Vol. 9(1), 2016 ISSN 1675 – 5650

MALAYSIAN COCOA JOURNAL ------

Advisor Datin Norhaini Udin

Vice Advisor Dr. Ramle Kasin

Editor Dr. Rosmin Kasran

Editorial Committee Dr. Alias Awang Dr. Ahmad Kamil Mohd Jaaffar Dr. Douglas Furtek Dr. Fisal Ahmad Dr. Hii Ching Lik Dr. Rozita Osman Dr. Samuel Yap Kian Chee Dr. Tan Chia Lock Dr. Zainal Baharum En. Mohamed Yusof Ishak Hjh. Winoryantie Sulaiman Pn. Harnie Harun Pn. Suzannah Sharif

Published by MALAYSIAN COCOA BOARD 5-7th Floor, Wisma SEDCO Lorong Plaza Wawasan, Off Coastal Highway 88999 Kota Kinabalu, Sabah, Malaysia

CONTENTS

ASSESSMENT OF SOMATIC EMBRYOGENESIS POTENTIAL OF ELITE MALAYSIAN Theobroma cacao CLONES 1 Siti Norhana, M. A., Haikal Eman, A., Ahmad Kamil, M. J. and Kasran, R.

THE EFFECT OF ASCORBIC ACID IN THE MATURATION OF COCOA SOMATIC EMBRYOS 5 Norasekin, T. and Ducos, J.P.

IMPROVEMENTS OF Theobroma cacao ACCLIMATIZATION TO EX VITRO ENVIRONMENT 11 Nik Aziz, N. I., Jasli, M., Misak, E., Abdul Asui, N. and Kasran, R.

DEVELOPMENT OF SNP GENOTYPING ASSAYS IN COCOA POD BORER USING REAL-TIME PCR Kasran, R., Roslina, M.S., David, A., Nuraziawati, M.Y., Navies, M.., Mellisa, G., Fahmi, W. and 19 Larry, C.

IDENTIFICATION OF THE POTENTIAL RESISTANCE GENES IN Theobroma cacao TO COCOA POD BORER INSECT (Conopomorpha cramerella) IN RELATION TO HOST- PATHOGEN INTERACTION USING RNA-SEQ TECHNOLOGY 40 Roslina, M.S., David, A., Anisah, S., Navies, M., Kasran, R., K., Mellisa, G.,Fahmi, W. and Larry, C.

DNA FINGERPRINTING OF THE MALAYSIAN COCOA GERMPLASM COLLECTION USING MICROSATELLITE MARKERS 50 Lea, J. and Haya, R.

EVALUATION OF CLONAL MATERIALS IN MALAYSIAN COCOA BOARD, CRDC MADAI SABAH 59 Aizat, J., Nuraziawati, M.Y., Azlan, S. and Haya, R.

ANALYSIS OF COCOA YIELD COMPONENTS IN RELATION TO ITS YIELD Tan, C.L. and Ling, A.S.C. 64

PHYSIOLOGICAL RESPONSES OF CACAO SEEDLINGS SUPPLEMENTED WITH MINERALS USING FOLIAR APPLICATION 71 Tee, Y.K., Bailey, B.A., Ali, S.S., Strem, M.D., Elson, M. and Baligar, V.C.

THE EFFICIENCY OF CACAO SPECIFIC COMPOUND FERTILIZER ON CACAO PRODUCTIONS 79 Helmi, S., Rozita.O.., Haya. R., Ling, A.S.C.and Halim, H.

UNRAVELLING THE INTERACTION BETWEEN LEPIDOPTERAN SODIUM CHANNEL AND NEUROTOXIN PEPTIDE: A STRUCTURAL INSIGHT 86 Rahman, M. Z. A. , Jainuddin, H. and Kasran, R.

TOXICITY STUDIES OF Bacillus thuringiensis TO COCOA POD BORER LARVA AND ADULT MOTH 97 Tan, C.L., Navies, M. and Anisah, S. CAFFEINE TOXICITY TEST ON COCOA POD BORER ADULT MOTH Anisah, S. 103

EVALUATION OF BEST MANAGEMENT PRACTICES FOR COCOA POD BORER IN COCOA ECOSYSTEM 108 Saripah, B. and Alias, A.

ESTIMATING BLACK POD DISEASE PROGRESS RATE USING THE AREA UNDER DISEASE PROGRESS CURVE (AUDPC) FOR DIFFERENT COCOA CLONES IN LABORATORY CONDITION 121 Ling, A.S.C., Ahmad Kamil, M.J., Chong, K.P. and Ho, C.M.

PLANT GROWTH HORMONES PRODUCED BY ENDOPHYTIC Bacillus subtilis STRAIN LKM-BK ISOLATED FROM COCOA Ishak, Z., Mohd Iswadi, M.K., Russman Nizam, A.H., Ahmad Kamil, M.J., Ernie Eileen, R.R., Wan 127 Syaidatul, A. and Ainon, H.

EFFECTIVENESS OF SYSTEMIC FUNGICIDE (TEBUCONAZOLE) IN CONTROLLING VASCULAR STREAK DIEBACK (VSD) ON COCOA SEEDLINGS 134 Suhaida, S., Shari Fuddin, S. and Alias, A.

INHIBITION ABILITY OF COCOA POD EXTRACT ON TYROSINASE ACTIVITY Azila, A.K., Nur Azilah, A. and Azrina, A. 139

THE EFFECTS OF PHYSIOCHEMICAL ACTIVATION ON ADSORPTION CAPACITY, POROSITY AND SURFACE FUNCTIONAL GROUPS OF COCOA (Theobroma cacao) NIB – BASED ACTIVATED CARBON 151 Khairul Adli, N., Fisal, A., Mohd Azmier, A.and Mohd Sukri, H.

HUMAN CHEEK SKIN TEXTURE IMPROVEMENT BY COCOA BEAN EXTRACT INCORPORATED IN A COSMETIC FORMULATION 160 Norliza, A.W., Russly, A.R., Puziah, H., Amin, I. and Shuhaimi, M.

BIOACTIVITY-GUIDED FRACTIONATION OF POTENT ANTI-CANCER PROPERTIES FROM NON-EDIBLE TISSUES OF Theobroma cacao 170 Zainal, B., Abdah M. A., Taufiq Yap Y. H., Roslida A. H., Mohd Redzuan, S. and Kasran, R.

DISINTEGRATION OF COCOA POWDER TABLET Samuel, Y.K.C., Chin, H.H., Sarini, H. and Arief Huzaimi, M.Y. 181

SHELF-LIFE STUDY OF Lactobacillus fermentum IN PROBIOTIC CHOCOLATE Rosmawati, M.S., Abdul Rahim, A.M. and Kasran, R. 187

MALAYSIAN RENAISSANCE COCOA: QUANTIFICATION OF FLAVANOLS AND METHYLXANTINES IN SELECTED MALAYSIAN COCOA CLONES 193 Azhar, M. and Kasran, R.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

ASSESSMENT OF SOMATIC EMBRYOGENESIS POTENTIAL OF ELITE MALAYSIAN Theobroma cacao L. CLONES Siti Norhana, M. A.1, Haikal Eman, A.1, Ahmad Kamil, M. J.2 and Kasran, R.1 1Malaysian Cocoa Board, Cocoa Biotechnology Research Centre, Kota Kinabalu Industrial Park, 88460 Kota Kinabalu, Sabah, Malaysia 2 Cocoa Research & Development Centre Kota Samarahan, Malaysian Cocoa Board, Lot 248, Block 14, Daerah Muara Tuang, Bahagian Samarahan, Locked Bag 3131, 93450, Kuching, Sarawak.

Malaysian Cocoa Journal 9(1): 1-4 (2016) ABSTRACT - Somatic embryogenesis provides an efficient in vitro clonal propagation method to propagate elite clones rapidly for research and production. Due to the high heterozygosity nature of most cocoa clones, adjustment on the concentration of nutrients in the culture medium is necessary to increase the efficiency of embryogenic response of the local cocoa clones in Malaysia. Screening of 16 elite Malaysian cocoa clones from Class I and II using both patented and modified Nestle induction medium was carried out. Induction of somatic embryogenesis from floral staminodes of cocoa was studied with respect to primary somatic embryogenesis potential. The result showed five clones that were previously less responsive started to produce primary somatic embryos consistently between four to eight weeks after culture initiation on modified induction medium supplemented with 1mg/L Myo-inositol and 6.25mg/L L- Glutamine. Compared to the patented medium, the use of much lower concentration of Myo-inositol and L- Glutamine (100X and 40X lower, respectively) in the induction medium could reduce the time for somatic embryo production by half and able to stimulate higher frequency of embryogenesis in selected clones. The percentage of primary somatic embryos observed after eight weeks on the modified medium range from 5% to 43.3%. The study is currently concentrating on establishing highly embryonic calli stock and mass propagating secondary somatic embryos for embryo conversion trials.

Keywords: Somatic embryogenesis, Primary somatic embryos, Theobroma cacao, Myo-inositol, Glutamine

INTRODUCTION of two compounds in the induction medium, myo-inositol and L-glutamine, with respect to Somatic embryogenesis as an alternative for mass primary somatic embryogenesis potential were propagation of elite Theobroma cacao clones has investigated. been well studied for years. It provides an efficient in vitro clonal propagation method to This study involved 16 Malaysian cocoa propagate elite clones rapidly for research and clones from Class I and Class II. These cocoa production. The main advantages of this method clones were previously classified into four classes include the possibility of rapidly generating - Class I, II, II and IV - based on their asexually propagated uniform plants of high adaptability to a wide range of Malaysian agro- genetic value, and the clonal production of climatic conditions, yield potential, pod and bean orthotropic plants with normal dimorphic characteristics, tolerant level to major pest and architecture and taproot formation (Maximova et diseases and possession of high butter content al., 2002). and good flavor (MCB, 2012). Class I clones are generally suitable for planting throughout Due to the high heterozygosity nature of Malaysia with good yield potential (>2.5 most cocoa clones, their degrees of response ton/ha/year) and tolerant to major pest and towards culture media treatments varied. In order diseases, while Class II clones are suitable for to increase the efficiency of embryogenic planting in selective areas (such as inland soil) response of each clone, it is necessary to modify with good yield potential, moderately tolerant to the concentration of the nutrients in the culture. major diseases and possess good pod and bean In this study, the effects of various concentrations quality (MCB, 2012).

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

This study aimed to assess the somatic medium contained macro-nutrients DKW, micro- embryogenesis potential of elite Malaysian cocoa nutrients DKW, vitamins DKW, glucose 1 g/l, clones from Class I and II, using both Nestle sucrose 30 g/l, Phytagel (Sigma), pH5.7. patented induction medium as well as modified version of the same medium. Induction of The experiment was carried out in somatic embryogenesis was studied with respect triplicate. 20 explants were cultured per clone per to primary somatic embryogenesis potential. replicate. The average percentage of somatic embryos produced, representing the total number of somatic embryos per total number of explant, MATERIALS AND METHODS was determined eight weeks after the culture initiation. The earliest time taken to produce Staminodes from unopened cocoa flower buds primary somatic embryo from initial culture for (age 2-4 weeks) were used as explants. Samples each clone was also recorded. However, due to were obtained from bud grafted, field and seasonal availability of flowers, the experiments glasshouse-grown clonal plants at Centre for carried out on the patented and modified medium Cocoa Biotechnology Research, Kota Kinabalu were not able to be carried out side by side, hence Industrial Park, Sabah, Malaysia. done on different dates.

Unopened cocoa flower buds were washed with tap water and surface-sterilized RESULTS AND DISCUSSIONS using 10% commercial bleach (5.25% (v/v) sodium hypochlorite) for 20 min followed by Using the patented Nestle induction media, the three rinses for five min each in sterile reverse mean percentage of somatic embryos produced osmosis water. Following dissection, staminodes ranged from 8.33 to 18.33 (Table 1). Clones were extracted and placed on the culture medium. MCBC1, PBC123 and KKM22 from Class I gave the highest percentage of somatic embryos For callus initiation, staminodes were produced – 18.33%, 16.67% and 15%, cultured on Nestle patented induction medium respectively. Time taken to produce primary which comprised of macro-nutrients DKW somatic embryos were consistent and fast for (Driver and Kuniyaki, 1984), micro-nutrients both MCBC1 and PBC123 (around 6-7 weeks), DKW, vitamins DKW, Glucose 20 g/l, glutamine however, it was inconsistent and much slower for 250 mg/l, myo-inositol 100 mg/l, 2,4-D 2 mg/l, KKM22 (range from 7 to 13 weeks). Using the TDZ 5µg/l, Gelrite 3g/l, pH5.8. For modified modified induction medium, preliminary results medium, the modifications in the medium include showed that the mean percentages for primary adjusting the concentrations of L-glutamine and somatic embryos production were much higher myo-inositol to (6.25 mg/l, 0.156 mg/l) and (1 for PBC123 and PBC112 with 43.33% and mg/l, 0.01 mg/l) respectively. 38.33%, respectively (Table 2). The time taken for somatic embryos production was also fast The cultures were incubated in the dark (around 5 to 7 weeks) and consistent. at 25±2oC for two weeks, then transferred for another two weeks in a modified secondary callus Table 1. Embryogenic potential of cocoa clones growth medium. The modifications in this on Nestle medium (L-glutamine 250 mg/l, myo- medium include substitution of Woody Plant inositol 100 mg/l) Media (WPM) with DKW basal salt and Time To Produce Mean ± Std Clones Class Somatic Embryo (From solidifying agent Gelrite to Phytagel. It contained Error Initial Culture) DKW Basal Salt (Duchefa), glucose 20 g/l, 2.2g/L Phytagel (Sigma), B5 Vitamins, 2,4- KKM22 I 15±2.887 7 to 13 weeks dichlorophenoacetic acid 2mg/l, Kinetin MCBC1 I 18.333±1.667 6 to 7 weeks

0.25mg/l, pH5.7. PBC123 I 16.667±1.667 6 to 7 weeks

Somatic embryos were induced when KKM19 II 10±0 6 weeks the cultures were transferred to a PGR-free MCBC9 II 13.333±1.667 12 to 13 weeks differentiation medium, under the same PBC112 II 8.333±1.667 15 to16 weeks incubation conditions as described above. This

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 2. Embryogenic potential of cocoa clones inositol starvation. The effects of different myo- on modified Nestle induction medium (L- inositol level in the growth medium were also glutamine 6.25 mg/l, myo-inositol 1 mg/l) reported in somatic embryogenesis of other plants Time To Produce such as Algerian fir (Abies numidica) (Vooková Clones Class Mean ± Std Error Somatic Embryo (From Initial Culture) et al., 2001) and Egyptian clover (Trifolium

KKM22 I 15±2.887 6 weeks Alexandrinum) cultivars (Aly and Rashad, 2001).

PBC123 I 43.333±1.667 5 to 6 weeks Glutamine is a common, readily BR25 II 3.333±1.667 7 to 8 weeks assailable organic nitrogen source used in plant KKM19 II 5±2.887 4 weeks tissue culture media. The addition of amino acids to the culture medium has a dual effect they MCBC9 II 10±0 6 to 7 weeks either promote somatic embryogenesis PBC112 II 38.333±1.667 5 to 7 weeks development or show inhibitory effect. The response to these amino acids also varies with the species and type of explant. In plants such as date palms, Ageel and Elmeer (2011) reported that addition of amino acid specifically glutamine stimulated growth of date palm callus tissue significantly, which suggested that organic nitrogen was a growth-limiting factor in date palm cultures. Similarly, Ogita et al. (2001) reported that when embryogenic tissues of sugi (Cryptomeria japonica) were cultured on a high glutamine-containing (2,400 mg/l) medium, a large number of embryogenic aggregates was generated. In contrast, the usage of glutamine at higher concentrations (150 mg/l and 200 mg/l) Figure 1. Clone PBC123 on expression medium had a negative effect in the formation of eight weeks after culture initiation embryogenic calli of caroá, a native fiber plant in Brazil (Silveira et al., 2013). Myo-inositol or inositol is a sugar-like carbohydrate produced by most plants, and is The use of significantly lower usually added to stimulate cell growth and concentration of L-glutamine and myo-inositol in development. It is generally used in plant cell and the induction medium could potentially increase tissue culture media at concentrations of 50-5000 the efficiency of somatic embryogenesis of mg/l (Khadimi et al., 2014). The essential roles of selected cocoa clones from Class I and II in inositol in many cellular processes include Malaysia. The preliminary results showed that it membrane formation, plant cell wall biosynthesis, could reduce the time for somatic embryo stress response and signal transduction (Lackey et production by half and able to stimulate higher al., 2003). frequency of embryogenesis in selected cocoa clones. Nevertheless, no definitive conclusion In other plants, manipulation of myo- could be derived from these results as this study inositol level in in vitro cultures has been is currently still on-going. reported to influence callus formation and somatic embryos production. Kumar and Tuli CONCLUSIONS (2004) reported an increase in somatic embryogenesis efficiency and developmental The result showed five clones that were synchronization in embryogenic callus cultures of previously less responsive started to produce cotton by a single cycle of myo-inositol depletion primary somatic embryos consistently between in liquid culture. They reported that the number four to eight weeks after culture initiation on of mature somatic embryos obtained per explant modified induction medium supplemented with was 4 to 5-fold higher than that obtained without 1mg/L Myo-inositol and 6.25mg/L L-Glutamine.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Compared to the patented medium, the use of embryogenesis. National Botanical much lower concentration of Myo-inositol and L- Research Institute, Rana Pratap Marg, Glutamine (100X and 40X lower, respectively) in Lucknow-226001 (Uttar Pradesh), India. the induction medium could reduce the time for somatic embryo production by half and able to Lackey, K.H., Pope, P.M. and Johnson, M.D. stimulate higher frequency of embryogenesis in (2003). Expression of 1L-myo-inositol- selected clones. The percentage of primary 1-phosphate synthase in organelles. somatic embryos observed after eight weeks on Plant Physiol. 132: 2240-2247. the modified medium range from 5% to 43.3%.

Malaysia Cocoa Clones (2012). Malaysian Cocoa

Board. ACKNOWLEDGEMENTS

The author would like to thank Director General Maximova, S.N., Alemanno, L., Young, A., of Malaysian Cocoa Board for permission to Ferrière, N., Traore, A. and Guiltinan, publish these results. The technical assistance of M.J. (2002). Efficiency, genotypic Siti Zakiyyah Md. Soib, Norasekin Tamchek and variability, and cellular origin of primary Nik Iryani Nik Aziz are gratefully acknowledged. and secondary somatic embryogenesis of Theobroma cacao L. In Vitro Cellular and Developmental Biology – Plant 38: REFERENCES 252-259.

Ageel, S. and Elmeer, K. (2011). Effects of Ogita, S., Sasamoto, H., Yeung, E.C. and Thorpe, casein hydrolysates and glutamine on T.A. (2001). The Effects of glutamine callus and somatic embryogenesis of on the maintenance of embryogenic date palm (Phoenix dactylifera L.). New cultures of Cryptomeria japonica. In York Science Journal 4(7). Vitro Cell. Dev. Biol.–Plant 37: 268- 273. Aly, M.A.M. and Rashad, M.H. (2001). The effect of growth regulators, vitamins and Silveira, D.G., Lino, L.S.M., Souza, A. Da S. and Myo-inositol on callus and somatic Souza, F.V.D. (2013). Somatic embryogenesis induction in different embryogenesis of Neoglaziovia genetic backgrounds of Egyptian clover variegata (Arruda) Mez, an important (Trifolium Alexandrinum L.). source of fiber from native Brazilian Conference Faculty of Agriculture, bromeliads. Brazilian Archives of Cairo University, Giza, Egypt. Biology and Technology 56(4): 547-555.

Driver, J.A. and Kuniyuki, A.H. (1984). In vitro Vooková, B., Kormut'ák, A. and Hřib. (2001). propagation of Paradox walnut Effect of myo-inositol on somatic rootstock. HortScience 19: 507-509. embryogenesis of Abies numidica. J. Appl. Bot. 70: 46–49. Khadimi, A.A., Alhasnawi, A.N., Mohamad, A., Yusoff, W.M.W. and Zain, C.R.C.M. (2014). Tissue culture and some of the factors affecting them and the micropropagation of strawberry. Life Science Journal 11(8).

Kumar, M. and Tuli, R. (2004). Plant regeneration in cotton: A short-term inositol starvation promotes developmental synchrony in somatic

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

THE EFFECT OF ASCORBIC ACID IN THE MATURATION OF COCOA SOMATIC EMBRYOS

Norasekin, T.1 and Ducos, J.P.2 1Malaysian Cocoa Board, Cocoa Biotechnology Research Centre, Kota Kinabalu Industrial Park, 88460 Kota Kinabalu, Sabah, Malaysia 2Nestlé R&D Centre, 101 Avenue Gustave Eiffel, 37097 Tours Cedex 2, France

Malaysian Cocoa Journal 9(1): 5-10 (2016) ABSTRACT - This study was conducted to determine the effect of ascorbic acid (0mM, 0.05mM, 0.1mM) in the maturation of cocoa somatic embryos. Maturation in this study was defined by the ability of somatic embryos to convert to a plantlet through a germination procedure. Somatic embryos of Theobroma cacao were harvested from temporary immersion bioreactor and placed onto maturation media supplemented with given concentration of ascorbic acid. Data obtained suggest that addition of ascorbic acid at 0.1mM results in longer stem length compared to the control. However, supplementation of media with this antioxidant at 0.05mM and 0.1mM gave no effect to the plantlet conversion number.

Key words: Ascorbic acid, Antioxidant, Germination, Maturation, Plantlet, Somatic embryos, Theobroma cacao

INTRODUCTION different biosynthetic pathways (Arrigoni and de Tullio, 2000), regulates cell division, elongation Theobroma cacao L. also known as cocoa plant is and cell multiplication (Gonzalez-Reyes et al., exclusively cultivated in the inter-tropical area of 1998), growth and development including the world and a major source of income for flowering, senescence and root development. developing countries (Alemanno et al., 2007). It According to Kerk and Feldman (1995), ascorbic is mainly propagated via seeds, grafting, or acid involved in the initiation and maintenance of rooted cuttings. However, high degree of the quiescent center in Zea mays root. Ascorbate segregation for many traits was demonstrated has also been extensively used in vitro systems when propagated by seeds (Maximova et al., where it has been found to enhance shoot forming 2008). Because of these, somatic embryogenesis processes (Joy et al., 1988). Other than that, (SE) continues to be the primary method for in ascorbic acid also regulates defense response and vitro clonal propagation method of cocoa survival of plants under abiotic and biotic stress (Guiltinan, 2007) as it has the potential for mass (Gallie, 2013). propagation at reduced time and at lower cost. Plant regeneration via SE also provides an The main objective of this short study is alternative method for large scale production of to improve conversion rate of the embryos plants. However some limitation still occur such through manipulation of the culture conditions by as low percentage of primary somatic embryo adding ascorbic acid. production and low percentage of embryo conversion during the early stages of germination. This low conversion frequency is MATERIALS AND METHODS probably because of lack of mitotic activity within the meristematic regions (Stasolla and Induction of somatic embryos and culture Yeung, 1999). For complete maturation to occur, conditions embryos must achieve both morphological and This study was conducted at Nestlé R&D physiological maturity which is initiated from the Research Centre, Tours, France. Plant materials development of the somatic embryos. used for maturation and germination study is Theobroma cacao L. (cocoa) genotype EET103. Ascorbic acid is a major antioxidants Embryogenic tissues were firstly prepared from that has various functions in physiological and cocoa staminodes and petals by cocoa team in biochemical processes in both animal and plant Nestlé R&D Centre, Tours through induction and systems. This metabolite acts as electron donor in expression media according to available

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

protocols. Somatic embryos were mass means were compared using Tukey’s test at p = propagated via liquid and temporary immersion 0.05 using the software MINITAB version 14. bioreactor techniques. RESULTS Maturation The effect of maturation media supplemented Maturation and germination of somatic embryos with various concentration ascorbic acid was To test the effectiveness of different evaluated in terms of subsequent embryo concentration of ascorbic acid on the cocoa germination and plantlet conversion ability. maturation and germination stages, somatic embryos were cultured on media containing White-opaque cotyledonary somatic 0mM, 0.05mM and 0.1mM of this antioxidant. embryos, 5-10 mm in length, were isolated from Four weeks after culture initiation, somatic embryogenic culture established in temporary embryos showed size and length increment for all immersion bioreactor and cultured on maturation treatment. All embryos were transferred onto media supplemented with different concentration germination media. of ascorbic acid. The ascorbate solutions were filter-sterilized before added to autoclaved media After 6 weeks in germination media, a and poured into the plates. The final difference was observed when the morphology of concentration of ascorbic acid used were 0mM embryos converted to a plantlet was examined (control), 0.05mM and 0.1mM. The petri dish (Figure 1). The size of plantlets’ stem slightly were sealed and placed in the dark for increased affected by the previous maturation approximately 4 weeks at a constant temperature treatment, which is the greatest stem length was of 26 ± 2OC. exhibited by somatic embryos mature on media supplemented with 0.1mM ascorbic acid. The Each treatment was applied to 25 highest is up to 7cm (Figure 1c). However, somatic embryos with 5 replicate petri dishes. plantlet from media supplemented with 0.05mM ascorbic acid showed no different in size if Germination compared to control media. After 4 weeks of maturation in the presence of ascorbic acid, somatic embryos were transferred As recorded, out of 125 embryos from onto germination media. There is no ascorbic acid genotype EET103 tested in each treatment, only was added to this media. The petri dish were 20% embryos from control, 9.6% from 0.05mM placed under light, at a constant temperature of and 24.8% from 0.1mM media supplemented 26 ± 2OC for at least 6 weeks. After 6 weeks in with ascorbic acid converted to a normal plantlet germination media, performance of somatic (with at least 2 leaves) (Figure 2). Addition of embryos was evaluated in terms of percentage of 0.05mM ascorbic acid significantly (p < 0.05) embryos converted to a normal plantlet. The decrease the number of converted embryos. normal plantlets in this study was defined as However, the two media of control and somatic embryos that produce at least two leaves supplemented with 0.1mM did not show any and root system. Concurrently, the external significant differences (p > 0.05) in the number of morphology of the embryos was examined in the converted embryos (Figure 3). Embryos term of stem length. The plant height was which failed to convert to normal plantlet were measured using thread and a standard 12 inch categorised as semi normal and abnormal (Figure ruler. 4) (data not shown).

All converted embryos were later In summary, addition of ascorbic acid at acclimated to ex vitro conditions. Acclimated both concentration (0.05mM and 0.1mM) in plants were allowed to develop further in the maturation media did not gave major effect to the greenhouse under natural conditions. somatic embryos maturation, germination and conversion compared to standard media. The data collected were statistically analyzed by one-way ANOVA and presented as All converted embryos were allowed to the percentage means ± standard error. The develop further in the greenhouse and plantlets developed normally.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

a b c

Figure 1. Difference size of embryos converted to a plantlet in maturation and germination media for EET103; (a) control media, (b) 0.05mM ascorbic acid (c) 0.1mM ascorbic acid.

Figure 2. Percentage and mean of embryos EET103 converted to a normal plantlet for five replicate from three different treatment.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Figure 3. Mean of the percentage (mean ± standard error) for embryo EET103 converted to a normal plantlet. Different letters (a,b) shows significant differences responsiveness according to ascorbic acid concentration by Tukey’s test (p=0.05).

a b c

Figure 4: Type of embryos after maturation; (a) normal plantlet, (b) semi normal plantlet and (c) abnormal plantlet with abnormal shoot with cotyledon like leaves.

DISCUSSIONS leaf primordia in spruce (Kong and Yeung, 1992) and increase germination on Medicago sativa L. Complete maturation will only occur if embryos (Arab and Ehsanpour, 2006). However, addition achieve both morphological and physiological of ascorbic acid into Theobroma cacao genotype maturity (Stasolla and Yeung, 2003). In this short EET103 maturation media were not really study, we investigated the effect of ascorbic acid effective to increase the conversion rate as the (0 mM, 0.05 mM, 0.1 mM) in the maturation percentage is very low and almost no difference media to optimize conversion rate of somatic with control. This result is significant with embryos to plantlet. It is well established that another trial on other genotype done in Nestlé ascorbic acid is required in many processes of R&D Centre, Tours (data not shown). According plant growth and development (Smirnoff, 1996), to Ko et al. (2009), addition of ascorbic acid to continuous production of shoot growth and new the surface of culture media not only prevented

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

the development of lethal browning but also CONCLUSIONS greatly increased the number of shoot produced, hence the effectiveness of ascorbic acid to Results of this study suggest that, Theobroma increase conversion rate using this technique cacao genotype EET103 does not require the should be tried. presence of ascorbic acid in maturation media to produce normal converted plantlet. Omitting This study also showed that ascorbic ascorbic acid during maturation could reduce acid treatment during maturation produced production cost. However, further research may plantlet with a little bit higher rather than be needed to other genotype especially standard media and they did develop to whole recommended by Malaysian Cocoa Board as the plantlets. Increasing in growth might be from real function of ascorbic acid during plant increasing in mitotic division in meristematic maturation remain unclear. zones and cell growth in length due to ascorbic acid influences (Smirnoff et al., 2000) and it was confirmed that cellular expansion and ACKNOWLEDGEMENTS accumulation of storage products occur during embryo maturation, generally leading to an I would like to express my very great increase in embryo size (Merkle et al., 1995). appreciation to Dr Pierre Broun for accepting me However, from the result, this technique neither and my sincere appreciation to Malaysian Cocoa improve the percentage of normal plantlets nor Board for providing me an opportunity to do my decrease the abnormality of plantlets formation. one year training in Nestlé R&D Research Centre, Tours, France. Special thanks to Cocoa Influence of genotype on the rate of embryo Team in Nestlé for their guidance and valuable maturation technical support on this project. Any opinions, Results shown that a lot of embryos failed to conclusions and recommendations in this material convert to a normal plantlet, thus the percentage are those of the author and not necessarily to be of converted plantlet is low even though in attributed to the Nestlé R&D Research Centre, control. It might be due to the fact that cocoa Tours, France. genotype will respond differently to every stages in tissue culture process that may be attributed to their genetic makeup (da Silva et al., 2008; REFERENCES Maximova et al., 2002; Maximova et al., 2014; Quainoo and Dwomon, 2012). However, Alemanno, L., Devic, M., Niemenak, N., Sanier, currently none of the existing reports in C., Guilleminot, J., Rio, M., Verdeil, J.L. Theobroma cacao describe the genotype-specific and Montoro, P. (2007). Characterization response to ascorbic acid during maturation stage. of leafy cotyledon1-like during embryogenesis in Theobroma cacao L. Furthermore, embryo selection based on Planta An International Journal of Plant morphology of the somatic embryos is important Biology. because structural abnormality of somatic embryos is linked to low conversion rate into Arab, L. and Ehsanpour, A.A. (2006). The effects plant (Rodriguez and Wetzstein, 1994) and of ascorbic acid on salt induced alfalfa according to Quainoo and Dwomon (2012), (Medicago sativa L.) in vitro culture. torpedo stage embryos were more efficient than Biokemistri 18(2): 63-69. globular and heart stage embryos in converting into plantlets. Structural abnormality were Arrigoni, O. and Detullio, M. C. (2000). The roll frequently viewed as consequences of the recent of ascorbic acid in cell metabolism: exposure to exogenous condition during tissue between gene-directed functions and culture process such as phytohormones and unpredictable chemical reactions. Plant culture conditions. Physiology, 157: 481-488.

da Silva, T.É.R., Cidade, L.C., Alvim, F.C., de Mattos Cascardo, J.C. and Costa, M.G.C. (2008). Somatic embryogenesis and plant

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

regeneration in elite clones of Theobroma of Theobroma cacao L plants propagated cacao. Pesquisa Agropecuária Brasileira. via somatic embryogenesis. In vitro 43(10): 1433-1436. Cellular and Developmental Biology- Plant 44: 487-493. Gallie, D.R. (2013). L-Ascorbic Acid: A multifunctional molecule supporting plant Maximova, N.S., Florez, S., Shen, X., Niemenak, growth and development. Scientifica,1-24. N., Zhang, Y., Curtis, W. and Guiltinan, M.J. (2014). Genome-wide analysis Gonzalez-Reyes, J.A., Cordoba, F. and Navas, P. reveals divergent patterns of gene (1998). Involvement of plasma membrane expression during zygotic and somatic redox systems in growth control of animal embryo maturation of Theobroma cacao and plant cells. In: Plasma membrane L., the chocolate tree. BioMed Central redox systems and their role in biological Plant Biology 14: 185. stress and disease, Asard, H., Berczi, H. and Caubers, R.J. (eds). Kluwer Academic Merkle, S.A., Parrot, W.A. and Flinn, B.S. Publishers, Dordrecht, Netherlands, 193- (1995). Morphogenic aspects of somatic 213. embryogenesis. In: Thorpe T.A. (ed). In vitro embryogenesis in plants. Dordrecht: Guiltinan, M. J. (2007). Cacao. In Pua EC, Kluwer Academic Publishers, 155-203. Davey MR (eds) Biotechnology in Agriculture and Forestry - Transgenic Quainoo, A.K. and Dwomon, I.B. (2012). The Crops VI. Springer-Verlag, Berlin effect of abscisic acid in the conversion of Heidelbelg (in press). cocoa somatic embryos into plantlets. Frontiers in Science 2(2): 6-10. Joy, R.W.IV., Patel, K.R. and Thorpe, T.A. (1988). Ascorbic acid enhancement of Rodriguez, A.P.M. and Wetzstein, H.Y. (1994). organogenesis in tobacco. Plant Cell The effect of auxin type and concentration Tissue Organ Culture, 13: 219-228. on pecan (Carya illinoinensis) somatic embryo morphology and subsequent Kerk, N.M. and Feldman, L.J. (1995). A conversion into plants. Plant Cell Reports biochemical model for initiation and 13: 607-611. maintenance of the quiescent center: Implications for organization of root Smirnoff, N. (1996). The function and meristems. Development 121: 2825– 2833. metabolism of ascorbic acid in plants. Annals of Botany 78: 661–669. Kong, L. and Yeung, E.C. (1992). Development of white spruce somatic embryos: II. Smirnoff, N. and Wheeler, G.L. (2000). Ascorbic Continual shoot meristem development acid in plants: Biosynthesis and function. during germination. In vitro Cellular and Critical Reviews in Plant Sciences 19: Developmental Biology-Plant 28p: 125- 267-290. 131. Stasolla, C. and Yeung, E.C. (1999) Ascorbic Maximova, N.S., Alemanno, L., Young, A., acid embryos. In vitro Cellular and Ferriere, N., Traore, A. and Guiltinan, Developmental Biology-Plant 35: 316- M.J. (2002). Efficiency, genotypic 319. variability, and cellular origin of primary and secondary somatic embryogenesis of Stasolla, C. and Yeung, E.C. (2003). Recent Theobroma cacao L. In vitro Cellular and advances in conifer somatic Developmental Biology-Plant 38: 252- embryogenesis: Improving somatic 259. embryos quality. Plant Cell Tissue and Organ Culture 74: 15-35. Maximova, N.S., Young, A., Pishak, S. and Guiltinan, M.J. (2008). Field performance

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

IMPROVEMENTS OF Theobroma cacao ACCLIMATIZATION TO EX VITRO ENVIRONMENT

Nik Aziz, N. I1., Jasli, M1., Misak, E2., Abdul Asui, N3. and Kasran, R1. 1Malaysian Cocoa Board, Cocoa Biotechnology Research Centre, Kota Kinabalu Industrial Park, 88460 Kota Kinabalu, Sabah, Malaysia 2Politeknik Malaysia Sandakan, 90000 Sandakan, Sabah 3Universiti Malaysia Perlis, 02600 Arau, Perlis

Malaysian Cocoa Journal 9(1): 11-18 (2016) ABSTRACT- Successful acclimatization of tissue culture plantlets from in vitro to ex vitro environment is a prerequisite for any tissue culture strategy. On a commercial scale, mass multiplication of tissue culture plantlets are deemed successful when substantial amount of plantlets transplanted to ex vitro environment at a low cost demonstrated low mortality. In this study, we aim to increase the success rate of cocoa tissue culture plantlets acclimatization by improving its influencing factors through preparatory test cultivations in growth chambers. The factors monitored were: temperature, relative humidity, types of potting medium and acclimatization chamber improvements. Two acclimatization chambers were built for the purpose. The tissue culture plantlets involved was two months old QH1287 genotype. Temperature and relative humidity were recorded daily. Misting was done thrice daily. Compared to previous acclimatization trials, the use of acclimatization chambers improved plantlet survival to 39.1%. Average temperature and relative humidity recorded from October 2015 to April 2016 was 30.4⁰C and 77.5% respectively. Plantlets with at least two normal leaves, green stem and complete root system gave the best survival rate. By day seven in the chamber, plantlets started to produce new flush. The use of acclimatization chambers greatly improved cocoa tissue culture plantlet acclimatization and successfully decrease transplanting shock.

Key words: Cocoa, Tissue culture plantlet, Acclimatization, Transplanting shock, Preparatory test cultivations

INTRODUCTION a low cost demonstrated low mortality. Most tissue culture plantlets could not overcome the The seeds of cocoa (Theobroma cacao L.), dramatic environmental changes (Hayashi and commonly referred to as beans, are the defining Kozai, 1987) that consequently leads to ingredients of chocolate. Naturally farmers transplanting shock when moved from culture to would prefer to sell their beans instead of outside environment. To this day, reserving it for planting material. Thus, acclimatization of tissue culture plantlets establishing a rapid cocoa mass multiplication remains a bottleneck that often restrict this protocol to address our farmers’ planting technology’s extensive use in mass material need is imperative. Tissue culture multiplication for most plants. It is also a provides an alternative to traditional cocoa challenge to cocoa tissue culture, especially to propagation which usually involves cultivation Malaysia’s own genotypes done in local of cross-pollinated seeds, rooted cuttings or environment. grafting (Guillou et al., 2014). The slow production via rooted cuttings or grafting and Acclimatization is a process where in non-uniform agronomic performance of seed vitro plantlets are prepared for ex vitro derived plants can be resolved by this environment conditions adaptations. In contrast technology. to high relative humidity (RH) and low light surroundings when in culture, plantlets are Successful acclimatization of tissue gradually exposed to lower RH and greater light culture plantlets from in vitro to ex vitro intensities before transplanting to the field. This environment is a prerequisite for any tissue is critical since most tissue culture derived culture strategy. On commercial scale, mass plantlets are morphologically and multiplication of tissue culture plantlets are physiologically divergent from normal seedlings. deemed successful when substantial amount of The stress imposed on plant cells during tissue plantlets transplanted to ex vitro environment at culture process more often than not induced the

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

aberrant characteristics of tissue culture plantlets a hydroponic setting. The trial run in their (Wang et al., 2013). Numerous studies have improved acclimatization unit prototype resulted discussed the aberrant characteristics of tissue in decreased death rate and increased plantlet culture plantlets (Pospisilova et al., 1999; growth rate. Hazarika, 2006; Chandra et al., 2010; Bairu and Kane, 2011; Kumar and Rao, 2012; Ruffoni and Cocoa plantlet acclimatization have Savona, 2013; Isah, 2015). been a challenge to the Centre’s tissue culture programme as well. The common practice for Their studies reported that tissue culture plantlet acclimatization was to cover individual plantlets exhibit poor development of plantlets in pots with plastic covering. The epicuticular wax, malformed stomata and/or plastic coverings served as a shade for gradual impaired stomatal functioning, decreased plantlet familiarization to much intense outside photosynthetic efficiency and leaves that have light. Throughout the day, water was sprayed to poor mesophyll differentiation and weak prevent the plantlets from wilting (a minimum vasculature. These morphological and frequency of three times). Nutrients were physiological abnormalities may render them supplied in the form of half strength McCown’s incapable to withstand the sudden challenges of salt solution, which was administered every two higher light intensities, fluctuating temperature, week. The longest plantlet survival achieved was water stress, considerably lower relative ten weeks in acclimatization although they did humidity, lack of exogenous continuous nutrient not show any growth (no stem elongation and no supply and the septic condition of ex vitro new flush production) and remains the same as environment. in culture before gradually showing signs of hiperhydricity and consequently die out (data not It is not uncommon for tissue culture shown). Acclimatization in a closed hydroponic plantlets to rapidly wilt after transplanting from setting was also done but to no avail (data not in vitro. In addition to that, the plantlets also face shown). The limiting factor to continuous the need to quickly develop autotrophic mode of acclimatization trial is lack of plantlets. This nutrition in order to avoid death, since high study reports the acclimatization improvements sucrose content in culture media negated the by constructing cost effective acclimatization necessity of photosynthetic apparatus chambers and monitoring the factors influencing development (Kumar and Rao, 2012). Isah plantlet survival such as temperature, relative (2015) and Hazarika (2006) also noted that humidity, and types of potting medium. tissue culture plantlets have insufficient chlorophyll content coupled with inactive or absent photosynthesis enzymes. All of these MATERIALS AND METHODS circumstances explained why tissue culture derived plantlets are more prone to transplanting This study was initiated in September 2015 and shock compared to greenhouse or field raised is still on going. It is being conducted in the seedlings. Centre for Cocoa Biotechnology Research’s nursery. The utilization of acclimatization chambers has long been a part of tissue culture Plantlet regeneration strategy. Kozai et al. (1987) proposed the Plantlets were generated from QH1287 cocoa construction of computerized acclimatization genotype via tissue culture protocol as described unit for accelerated strawberry plantlets growth. by Maximova et al. (2005). Flower samples The unit was then improved to achieve better were obtained from QH1287 trees in the environmental control, better survival rates and germplasm collection plot of the Centre for more rapid growth of plantlets (Hayashi and Cocoa Biotechnology Research, Kota Kinabalu, Kozai, 1987). Factors crucial to a successful Sabah. Somatic embryogenesis and plantlet acclimatization like temperature, air humidity, regenerations were done at the Centre’s Tissue light intensity, carbon dioxide concentration, air Culture Laboratory. Two months old QH1287 flow rate and temperature of the nutrient solution plantlets with at least 2 normal leaves and were controlled by a microcomputer. Their developed root systems were used for the strawberry plantlets test cultivation were done in acclimatization trials.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

sprinkler are switched on manually thrice daily Acclimatization chamber (7.30 am, 12.30 pm and 4.30 pm). The reason Two acclimatization chambers were constructed. for this is to monitor the plantlets progression The chambers are placed inside a nursery with and the chamber performance before finalizing 70% shade black netting. The diagram of the the parameters later on. Daily temperature, RH acclimatization chamber is depicted in Figure 1 and observations were recorded. & 2. Each chamber was installed with three misting sprinklers with four nozzles each, walls The duration of misting depends on the made of semi-transparent plastic sheets and is RH recorded by the thermometer hygrometer connected to a water source equipped with a (HTC-1 LCD Digital Temperature Humidity filter (1” 120 mesh 10 bar). This is to ensure the Meter Thermometer). Table 1 shows the cleanliness of water source and the absence of duration of misting according to RH inside the obstructing particles that can interfere with the chamber. The main pipe for the water source is misting system. The main part of the chamber is also connected to a water pump to ensure 3m long, 1.82m high (chamber and table height) adequate water pressure for the misting system. and 1 m wide. For the time being, the misting

Misting sprinkler

1.82 m

1 m

3 m

To water source with

filter

Figure 1. A diagram of the acclimatization chamber

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Figure 2. The acclimatization chambers

Table 1. Duration of misting according to relative humidity inside the chamber Relative humidity 40-50% 51-60% 61-70% 71-80% 81-90% 91-100% (RH) range Misting duration 60 seconds 30 seconds 15 seconds 10 seconds No misting No misting

Experimental design culture plantlets from total mortality to 39.1%. A randomized complete block design (RCBD) is The main probable reason for plantlet mortality used for the trials, with the chambers serving as is extreme weather caused by El Nino. experimental blocks. At the beginning of the According to the Malaysian Meteorological trials, QH1287 plantlets are grown in 10 x 18 cm Department (2016), the nation experienced El polybags filled with black soil and placed Nino from December 2015 to May 2016. Most randomly inside the chamber. After month three plantlet mortality occurred in December 2015 in the chamber, they are transferred to 12 x 23 and January 2016. Although the average cm polybags with the same planting medium. temperature and average RH for these two Two types of planting medium are used, black months (average temperature and RH for soil and Jiffy 7® peat pellet. For each planting December were 29.8⁰C and 77.2% while for medium, three replications are planned with 46 January 30.2⁰C and 76.8% respectively) were plantlets per replicate. Due to plantlet not as severe as the readings recorded in March availability limitation (seasonal availability of 2016 and April 2016 (32.7⁰C and 74% in March QH1287 flower source and efficiency of plantlet and 32.1⁰C and 73.4% in April), there are regeneration), only the preliminary result of occasional spikes of extreme daily temperature replicate 1 are presented here. Replicate 2 and 3 up to 40⁰C that resulted in too low RH (as low as will be completed in concert with future plantlet 47%) which can be challenging to the plantlets. availability. The plantlets are fertilized with 2 gram NPK green fertilizer monthly. For the first The extreme weather during this three months, misting was done solely. It was preparatory test cultivation posed much intense then combined with watering starting from water and heat stress than normally faced by month four onwards. Only plantlets with dry tissue culture plantlets during acclimatization. planting medium are watered. Pruning and As discussed above, most tissue culture leaves weeding were done once weekly. sport physiological and morphological abnormalities. Excessive water loss through inadequate stomatal structure and functioning RESULTS AND DISCUSSIONS and insufficient epicuticular wax coverage on leaf surface was exacerbated by the weather The use of acclimatization chamber increased stress factor. Normally plant photosynthesis and the survival rates of genotype QH1287 tissue

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

transpiration rate are reduced when facing fertilizer and lime to help in early plant growth. temperature increase in the environment (Zhang Figure 3 showed that with the increase of et al., 2010 and Lipiec et al., 2013). However average temperature and consequent drop in RH, due to culture induced phenotype, plantlets the death frequency of plantlet planted in Jiffy undergo water loss much rapidly than the rate of 7® also increases. Plantlets sown in black soil its replenishment (through means like absorbing fared better. It is too early to conclude any water from soil or planting medium to replete effects by this two planting medium on the the loss) which eventually lead to death. plantlets since the result presented here is preliminary. However, starting from February We observed that plantlets planted on the death rate gradually decreases and fully stop black soil resulted in 50.0% survival while only in April. This might due to the plantlets maturing 28.3.0% survived when Jiffy 7® peat pellets and have overcome the morphological and were used. The Jiffy 7® peat pellets are physiological abnormalities developed during Canadian sphagnum peat moss added with culture.

Figure 3. Plantlet mortality on different types of potting medium (black soil and Jiffy 7®) from October 2015 to April 2016 with the monthly average relative humidity (RH) and temperature

Figure 4. Scorched plantlet due to El Nino

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

There are two types of in vitro leaves, reported by Palee et al. (2012) from their study described by Hazarika (2006) as the on the influence of different plantlet age on photosynthetic competent group and non- acclimatization success of Stemona curtisii. Prior competent group. According to our observation, acclimatization trials used plantlets that were kept cocoa plantlet possessed the former type of in up to six months in rooting media, due to the vitro leaves. This is because cocoa plantlet leaves observation that plantlets with substantial root did not deteriorate after transplantation as the development will survive the acclimatization non-competent group would. In the chamber, the longer. However, the utilization of plantlets were observed to produce new flush acclimatization chamber showed that within day seven of acclimatization. This is a unfavourable environmental conditions might be great improvement since plantlets in previous the major contribution to plantlet inability to acclimatization trials never develop new flush survive. This is because plantlets undergoing ex and remained morphologically the same as in vitro rooting showed favourable outcome in the culture. However, the role of persistent in vitro chamber (data not shown). Apart from that, 2 leaf in still debatable (Hazarika, 2006). He months old plantlets showed more vigour and are suggested that in the beginning of less prone to develop hiperhydricity compared to acclimatization, in vitro leaves did not conduct plantlets that are kept in culture longer. The use photosynthesis. Only after a while in of chamber for acclimatization has enabled ex acclimatization that the leaves will attain a semi- vitro rooting and this will cut down on the time photosynthetic competent form. This suggestion needed in the production of tissue culture is in line with the findings of Lee et al. (1985). plantlets. Lee reported that cultured Liquidambar styracifua L. leaves have large vacuoles, limited cytoplasmic content and flattened chloroplast CONCLUSIONS with an irregularly arranged internal system. This lack of internal chloroplast membrane The use of acclimatization chamber greatly development and lack of palisade parenchyma improved the cocoa tissue culture seedlings differentiation might lower the photosynthetic survival compared to total mortality in previous capacity of plantlets compared to normal acclimatization trials. The chamber provided a seedlings. Their study also discovered that more ambient environment that prepares the plantlet leaves are devoid of starch granules, the plantlets against the dramatic change of product of photosynthesis. Hence the question, environment. Further trials with more parameters did cocoa plantlet persistent in vitro leaves to investigate such as growth, morphological and conduct photosynthesis to provide energy for the physiological attributes of the plantlets will be development of critical new flush or it is a semi- carried out in the future in order to guarantee a photosynthetically competent leaves to begin truly successful tissue culture strategy suited to with? This question arises since cocoa persistent the nation’s climate and our local cocoa in vitro leaves did not show any growth genotypes. increment and remained the same. However, we observe that the pruning of persistent in vitro leaves encouraged more vigorous new flush ACKNOWLEGDEMENT formation. This supports the theory proposed by Machado and Hardwick (1987) as reported by The authors wish to thank the Ministry of Almeida and Valle (2007), where new flush Science, Technology and Innovation, Malaysia, occurs when the carbohydrate supply is for the financial support (Project No. 02-03-13- replenished within the plant, usually after an SF0069) and the Director General of Malaysian interflush period. In our case, since the plantlets Cocoa Board for the permission to publish these did not have to cater to the nourishment of pruned results. Constructive inputs from Dr. Ahmad persistent in vitro leaves, its carbohydrate supply Kamil Mohd. Jaaffar (CRDC Tawau), Dr. can be dedicated to new flush formation. Rebicca Edwards (UNIMAS) and fellow colleagues from Center for Cocoa Biotechnology For plantlet age, we observe that 2 Research are deeply appreciated, not forgetting months old plantlet perform well during technical assistance from Mahani Jasli, Mohd. acclimatization. This is similar to the result

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Shah David, Haikal Eman Abdullah, Emelda Environmental control for Misak, Nabilah Abdul Asui and Sahrizal Saipin. acclimatization of in vitro cultured plantlets. (1) Development of the acclimatization unit for accelerating the REFERENCES plantlet growth and the test cultivation. J. Agr. Met., 42(4): 349-358 (In Almeida, -A. A. F and Valle, R. R. (2007). Japanese with English summary) Ecophysiology of the cacao tree. Braz. J. Plant Physiol. 19(4): 425-448. Kumar, K. and Rao, I. U. (2012). Morphophysiologicals problems in Bairu, M. W. and Kane, M.E. (2011). acclimatization of micropropagated Physiological and developmental plants in - Ex vitro conditions - A problems encountered by in vitro review. Journal of Ornamental and cultured plants. Plant Growth Horticultural Plants, 2(4): 271-283 Regulation. 63: 101–103. Lee, N., Wetzstein, H. Y. and Sommer, H. E. Chandra, S., Bandopadhyay, R., Kumar, V. and (1985). Effects of quantum flux density Chandra, R. (2010). Acclimatization of on photosynthesis and chloroplast tissue cultured plantlets: from laboratory ultrastructure in tissue-cultured plantlets to land. Biotechnol Lett. 32: 1199-1205. and seedlings of Liquidambar styraciflua L. towards improved Guillou, C., Fillodeau, A., Brulard, E., Verdier, acclimatization and field survival. Plant D., Simon, M., Landmann, A., Physiol. 78: 637-641 Lausanne, F., Fontanel, A., Ducos J. P., Buchwalder, A. and Broun, P. (2014). Lipiec, J., Doussan, C., Nosalewicz, A. and Nestlé Cocoa plan: Cocoa propagation Kondracka, K. (2013). Effect of drought by somatic embryogenesis. Proceedings and heat stresses on plant growth and of the Third International Conference of yield: a review**. Int. Agrophys. 27: the IUFRO unit 2.09.02 on “Woody 463-477 Plant Production Integrating Genetic and Vegetative Propagation Machado, R. C. R. and Hardwick, K. (1987). Technologies”. September 8-12, 2014. Dynamics and histology of individual Vitoria-Gasteiz, Spain leaf and whole flush development yield clues on the control of the cocoa flush Hayashi, M. and Kozai, T. (1987). Development cycle. In: Proc. 10th Int. Cocoa Res. of a facility for accelerating the Conf., Santo Domingo, Dominican acclimatization of tissue-cultured Republic, pp.143-149. plantlets and the performance of test cultivations. Plant micropropagation in Malaysian Meteorological Department. (2016). horticultural industries. Symposium http://www.met.gov.my/documents/106 Florizel 87 Arlon – Belgium. 61/101803/status_elnino

Hazarika, B. N. (2006). Morpho-physiological Maximova, S. N., Young, A., Pishak, S., Miller, disorders in in vitro cultured plants. C., Traore, A., and Guiltinan, M. J. Scientia Horticulturae 108: 105–120 (2005). Integrated system for propagation of Theobroma cacao L., In: Isah, T. (2015). Adjustments to in vitro culture Protocol for Somatic Embryogenesis in conditions and associated anomalies in Woody Plants, Series: Forestry Sciences, plants. ACTA BIOLOGICA Vol. 77, Jain, S. Mohan; Gupta, Pramod CRACOVIENSIA Series Botanica. 57/2: K. (Eds.) Springer, Dordrecht, The 9–28. Netherlands, ISBN: 1-4020-2984-5.

Kozai, T., Hayashi, M., Hirosawa, Y., Kodama, Palee, J., Dheeranupattana, S., Jatisatienr, A., T. and Watanabe, I. (1987). Wangkarn, S., Mungkornasawakul, P.,

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Pyne, S., Ung, A. and Sastraruji, T. (2012). Influence of plantlet age and different soilless culture on acclimatization of Stemona curtisii Hook.f. Asian Journal of Plant Sciences, 11(6): 294-299

Pospisilova, J., Ticha, I., Kadlecek, S., Haisel, D. and Pizakova, S. (1999). Acclimatization of micropropagated plants in ex vitro conditions. Biologia Plantarum. 42: 481-497

Ruffoni, B. and Savona, M. (2013). Physiological and biochemical analysis of growth abnormalities associated with plant tissue culture. Horticulture, Environment and Biotechnology. 54 (3): 191-205

Wang, X., Wu, R., Lin, X., Bai, Y., Song, C., Yu, X., Xu, C., Zhao, Na., Dong, Y. and Liu, B. (2013). Tissue culture-induced genetic and epigenetic alterations in rice pure-lines, F1 hybrids and polyploids. BMC Plant Biology. 13:77

Zhang B., Liu W., Chang S.X., and Anyia A.O. (2010).Water-deficit and high temperature affected water use efficiency and arabinoxylan concentration in spring wheat. J. Cereal Sci. 52: 263-269.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

DEVELOPMENT OF SNP GENOTYPING ASSAYS IN COCOA POD BORER USING REAL-TIME PCR

Kasran, R.1, Roslina, M.S.1, David, A.1, Nuraziawati, M.Y.2, Navies, M.3, Mellisa, G.4, Fahmi, W.4 and Larry, C.4 1Malaysian Cocoa Board, Cocoa Biotechnology Research Centre, Kota Kinabalu Industrial Park, 88460 Kota Kinabalu, Sabah, Malaysia 2Cocoa Research & Development Centre Hilir Perak, Malaysian Cocoa Board, Jalan Sg. Dulang, 36307 Perak. 3Cocoa Research & Development Centre, Malaysian Cocoa Board, Batu 10, Jalan Apas, Peti Surat 60237, 91012 Tawau, Sabah, Malaysia 4Malaysian Genomics Resource Centre Berhad (652790-V), 27-9, Level 9 Boulevard Signature Office, Mid Valley City, 59200 Kuala Lumpur, Malaysia

Malaysian Cocoa Journal 9(1): 19-39 (2016) ABSTRACT – This project is to validate three TaqMan SNP genotyping assays as breeding markers for cocoa pod borer resistance, using QuantStudio 12K Flex Real-time Polymerase Chain Reaction (PCR) system. Fourteen known cocoa clones which comprised of seven resistant and seven susceptible plants were tested. Successful validation of the genotyping assay sets of 14 clones further led to screening of additional 137 cocoa clones of Malaysian Cocoa Board germplasm collections to identify the allelic discrimination of resistant and susceptible plants to pod borer disease. A total of 151 cocoa DNA samples were used in this project: seven known-resistant clones (QH441, KKM22, NA33, QH22, DESA104, TG149, and KKM2); seven known- susceptible clones (PBC123, PBC221, DESA102, PBC112, DESA105, KKM21, and TG146); and 137 unknown clones. The cocoa leaf samples were disrupted into fine powder using a Tissue Lyser (Qiagen, Germany) or a plastic pestle (Axygen, USA). Cocoa DNA extraction was performed using either the NucleoSpin Plant II kit (Macherey-Nagel, Germany) or the DNeasy Plant Mini Kit (Qiagen, Germany), both according to the manufacturer’s protocol. The cocoa DNA samples were quantified using Nanodrop 2000 (Thermo Fisher Scientific Inc., USA). The samples were also run on a 1% agarose gel to determine the integrity of genomic DNA. Typically, at least 10 ng of DNA with intact DNA is recommended to proceed to real-time PCR. The extracted cocoa DNA samples were further diluted to 10 ng/µL using elution buffer. The final concentration of diluted DNA was re-quantified using Nanodrop to confirm the concentration of DNA. The normalized DNA were run as triplicates and tested with three SNP genotyping assays: AH0JEOH, AHX1IB1 and AHZAGH9. TaqMan SNP Genotyping Assay data was analyzed with TaqMan Genotyper Software using end-point experiment type with auto-calling. The three SNP assays tested on the 14 control samples in this study showed that they are able to distinguish phenotypes based on genotype, and were assumed to show similar efficiency when applied to other unknown cocoa clones. However, when these SNPs were tested on a larger group of 137 samples, only 37.5% of clones had matching phenotype and genotype. To ensure that the SNPs found from the 14 samples were not caused by being closely genetically related to each other, and that the samples selected were indeed from a random group of samples, a neighbor-joining tree was generated by using all SNPs identified. The resistant and susceptible clones are dispersed about the tree indicating that these 14 known clones are from random sampling and the three SNPs that were selected were not due to population effect. Moreover, the SNPs are located within a few kb of each other, further strengthening the likelihood of them being real markers of resistance. The positioning of these three SNPs in the intron of a potential hexose symporter is intriguing. It is possible this Major Facilitator Superfamily (MFS) symporter transports different substrates than hexose. The inconsistency of the phenotype determined by genotyping and Average Damage

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Severity Index (ADSI) could be due to copy number variation of the SNPs in the unknown clones. Other possibilities include the growth condition of each clone as they were sourced from different locations; clone origin of each clone; and the nature of how ADSI is scored, that is, it being a qualitative (observation of damage on the cocoa pod) rather than a quantitative measure could all be potential reasons for the observation. The SNP assays AH0JEOH, AHX1IB1 and AHZAGH9 have been designed based on the 14 previously sequenced clones and were able to clearly distinguish allelic differences among these samples to categorize the cultivars correctly as either resistant or susceptible plants. However, the same efficiency was not observed in the larger sample of 137 clones, highlighting a possibility that the resistance to cocoa pod borer is conferred by multiple genes. Increasing the number of known clones to be used in the SNP calling process will also increase the possibility of correctly identifying the gene(s) involved in cocoa pod borer resistance.

Key words: Theobroma cacao, Cocoa pod borer, SNP genotyping

INTRODUCTION flavour and high cocoa butter. The focus of the Malaysian Cocoa Genome Programme is to carry The immediate priority of the Cocoa Genome out genome wide association studies on sets of Program is to assist cocoa breeders in identifying genetically diverse clones with contrasting traits in seedlings that are most likely to mature into high order to identify biomarkers for superior traits. The yielders, possess large bean sizes, good cocoa biomarkers will serve as diagnostic tools for flavour and resistance to common pest and screening seedlings with the desirable traits that diseases using Single Nucleotide Polymorphism are more likely to mature into superior trees than (SNP) markers. the initial breeding population.

In the case of Malaysia, cocoa genomics In recent study, we have sequenced and research includes screening for desirable traits in de-novo assembled 14 cocoa genomes, seven the germplasm collection and development of susceptible to pod borers and seven resistant molecular markers. Breeding populations are varieties (Roslina et al., 2014). Six million SNPs currently being established to select clones for were identified across the 14 samples. Six of these desirable traits using conventional breeding with SNPs showed enrichment within the resistant group. Four SNPs were found solely in all seven molecular tools. Such advanced breeding resistant samples and none of the seven susceptible techniques have been successful in low high value samples, and one SNP was homozygous in all crops, and it is clear the economics of this seven resistant samples. A study testing these approach for high value crops such as cocoa are SNPs in a larger number of samples is needed to exceptional. The availability of sequenced cocoa demonstrate that the SNPs can be confidently used genomes across many breeding varieties will as high quality breeding markers for pod borer create millions of molecular markers directly resistance. useful for future breeding improvements. The main objective of this project is to The challenge is to establish superior validate three TaqMan SNP genotyping assays: planting materials that are resistant to pest and AH0JEOH, AHX1IB1 and AHZAGH9 as disease, and also exhibit high yields, good cocoa breeding markers for cocoa pod borer resistance,

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

using QuantStudio 12K Flex Real-time PCR genotyping assay sets of 14 clones further led to system. Fourteen known cocoa clones which screening of additional 137 cocoa clones that is to comprised of seven resistant and seven susceptible identify the allelic discrimination of resistant and plants for testing. Successful validation of the susceptible plants to cocoa pod borer.

MATERIALS AND METHODS

SNP Genotyping Data Workflow is as follows:-

Receipt of leaf samples

Fail DNA extraction

Fail

Quality assessment of DNA: purity, quantity and integrity

Gel electrophoresis

Fail

Set up PCR reactions (TaqMan Master Mix)

Perform real-time PCR (384-well plate format)

Post PCR plate read & analysis of data and findings

Figure 1. Work process towards genotyping as well as analysis and interpretation of results.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Laboratory Procedures RESULTS A total of 151 cocoa DNA samples were used in this project: seven known-resistant clones (QH441, Analysis of Genotyped samples KKM22, NA33, QH22, DESA104, TG149, and

KKM2); seven known-susceptible clones (PBC123, PBC221, DESA102, PBC112, Analysis of 14 control cocoa samples DESA105, KKM21, and TG146); and 137 The SNP validation conducted on the known- unknown clones. resistant and susceptible cocoa DNA clones showed that the allelic discrimination for all three Upon receipt of sample, the cocoa leaf assays worked. Furthermore, clustering and signal samples were disrupted into fine powder using a intensities were observed based on three different Tissue Lyser (Qiagen, Germany) or a plastic pestle SNP assays which showed that there is (Axygen, USA). Cocoa DNA extraction was amplification of the DNA template in all samples. performed using either the NucleoSpin Plant II kit Values were assigned to each SNP, whereby a: (Macherey-Nagel, Germany) or the DNeasy Plant Mini Kit (Qiagen, Germany), both according to the 1) Homozygous major allele is 0 manufacturer’s protocol. The cocoa DNA samples 2) Heterozygous is 0.5 and were quantified using Nanodrop 2000 (Thermo 3) Homozygous minor allele is 1 Fisher Scientific Inc., USA). The samples were The values for all three SNPs were also run on a 1% agarose gel to determine the accumulated to give one final value which was integrity of genomic DNA. Typically, at least 10 used to determine the phenotype (based on ng of DNA with intact DNA is recommended to genotype). A cut-off was determined whereby proceed to real-time PCR. samples having an assigned value of:

The sample quality results of the final 1) > 2.5 is deemed susceptible 2) < 0.5 are assumed to be resistant samples selected for real-time polymerase chain 3) 0.5- 2.5 were assumed to have a moderate reaction (PCR) are shown in Table A1 and Figure trait. A1 in the appendix. The extracted cocoa DNA samples were further diluted to 10 ng/µL using For the ADSI, values > 3 were treated as elution buffer. The final concentration of diluted susceptible, < 2 as resistant and anything in DNA was re-quantified using Nanodrop to confirm between as moderate, as determined by Malaysian the concentration of DNA. The normalized DNA Cocoa Board. were run as triplicates and tested with three SNP All 14 samples had matching phenotypes genotyping assays: AH0JEOH, AHX1IB1 and based on real-time PCR, sequencing results and AHZAGH9. ADSI except for QH441, as summarized in Table 1. The discrepancy observed could have something TaqMan SNP Genotyping Assay data was to with the fact that different starting leaf materials analyzed with TaqMan Genotyper Software using were used in the sequencing in Phase I and end-point experiment type with auto-calling. genotyping in the current Phase II. As for the Allele discrimination plots were reviewed and remaining 13 samples, both phenotypes determined by genotyping and ADSI matches, manually called when required. confirming that the three SNP assays selected are able to distinguish allelic differences and can be used to accurately determine the phenotype of the clones.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 1. Summary of genotype and phenotype of 14 known cocoa clones based on real-time PCR, sequencing results and ADSI values.

Real-time PCR Results Sequencing Results ADS I Clone Sample EOH (A>G) IB1 (G>A) GH9 (A>T) EOH (A>G) IB1 (G>A) GH9 (A>T) Phenotype from Name ID Total Total from MCB Phenotype Phenotype MC Value Genoty Genoty Value Genotype Value Genotype Value Genotype Value Value Value Genotype Value B pe pe

QH441 C0923 G/G 1 A/A 1 T/T 1 3 Susceptible A/A 0 G/G 0 A/A 0 0 Resistance 0.91 Resistance

KKM22 C0924 A/G 0.5 G/G 0 A/A 0 0.5 Resistance A/A 0 G/A 0.5 A/A 0 0.5 Resistance 1.26 Resistance

NA33 C0925 A/G 0.5 G/G 0 A/A 0 0.5 Resistance A/A 0 G/G 0 A/A 0 0 Resistance 1.44 Resistance

QH22 C0926 A/A 0 G/G 0 A/A 0 0 Resistance A/A 0 G/G 0 A/A 0 0 Resistance 1.45 Resistance

DESA104 C0927 A/A 0 G/G 0 A/A 0 0 Resistance A/A 0 G/G 0 A/A 0 0 Resistance 1.5 Resistance

TG149 C0928 A/G 0.5 G/G 0 A/A 0 0.5 Resistance A/A 0 G/G 0 A/A 0 0 Resistance 1.67 Resistance

KKM2 C0929 A/A 0 G/G 0 A/A 0 0 Resistance A/A 0 G/G 0 A/A 0 0 Resistance 1.89 Resistance

PBC123 C0930 G/G 1 A/A 1 T/T 1 3 Susceptible G/G 1 A/A 1 T/T 1 3 Susceptible 3.02 Susceptible

PBC221 C0931 G/G 1 A/A 1 T/T 1 3 Susceptible G/G 1 A/A 1 T/T 1 3 Susceptible 3.06 Susceptible

DESA102 C0932 G/G 1 A/A 1 T/T 1 3 Susceptible G/G 1 A/A 1 T/T 1 3 Susceptible 3.33 Susceptible

PBC112 C0933 G/G 1 A/A 1 T/T 1 3 Susceptible G/G 1 A/A 1 T/T 1 3 Susceptible 3.33 Susceptible

DESA105 C0934 G/G 1 A/A 1 T/T 1 3 Susceptible G/G 1 A/A 1 A/T 0.5 2.5 Susceptible 3.37 Susceptible

KKM21 C0935 G/G 1 A/A 1 T/T 1 3 Susceptible G/G 1 A/A 1 T/T 1 3 Susceptible 3.45 Susceptible

TG146 C0936 G/G 1 A/A 1 T/T 1 3 Susceptible G/G 1 A/A 1 T/T 1 3 Susceptible 3.56 Susceptible

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

F-test revealed that the two sample unknown clones to categorize the phenotypes groups (susceptible and resistant) have unequal based on genotypes. The phenotype based on variances, so a t-test assuming unequal variances ADSI was also determined using the same cut-off was carried out which showed that the null values described in the previous section. There hypothesis is rejected, hence indicating that there were several samples from MCB without assigned is a correlation between phenotype (as determined ADSI values, treated as blind samples for which a through genotyping) and the ADSI provided by comparison could not be carried out, omitting Malaysian Cocoa Board. A scatter plot showing another 66 samples from further analysis. The final similar output is also shown in Figure 2. QH441 is number of samples that were analysed was 66 shown as an outlier in this plot. samples.

Analysis of 137 other cocoa samples A comparison of both phenotypes Further SNP validation of 137 unknown cocoa determined through genotyping and ADSI revealed clones were also carried out in triplicates and the that only 37.5% of the unknown samples were results are summarized in Table 2. Several samples correctly phenotyped based on the genotype. did not show any amplification (“NO AMP”) and this could be due to the occurrence of proximal F-test revealed that the two sample SNPs within the primer binding site in these groups (susceptible and resistant) have unequal samples. This would cause primers to not variances, so a t-test assuming unequal variances hybridise to the DNA template and halt was carried out and showed that the null amplification. These samples were not assigned hypothesis is accepted, which indicated that there with values or phenotypes based on genotype, is no correlation between phenotype (as omitting nine samples from further analysis. determined through genotyping) and the ADSI provided by Malaysian Cocoa Board for these The same cut-off values used with the unknown cocoa clones. A scatter plot showing known cultivars were applied to these 137 similar output is also shown in Figure 3.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Figure 2. A scatter plot for the 14 known clones with the ADSI value on the vertical axis and the susceptible / resistant trait on the horizontal axis .

Susceptible Moderate Resistant

Figure 3. A scatter plot for the 66 unknown cultivars with the ADSI value on the vertical axis and the susceptible/ moderate/ resistant trait on the horizontal axis

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 2. Summary of genotype and phenotype of 137 unknown cocoa clones based on real-time PCR and ADSI values. NO AMP refers to samples that failed to amplify during real-time PCR

Real-time PCR Results Sample ADSI from Phenotype Clone Name EOH (A>G) IB1 (G>A) GH9 (A>T) Total ID Phenotype MCB from MCB Genotype Value Genotype Value Genotype Value Value

ICS 84 C0940 A/A 0 G/G 0 A/A 0 0 Resistant 2.28 Moderate PBC 137 C0938 A/G 0.5 G/A 0.5 A/A 0 1 Moderate 2.23 Moderate PBC 132 C0941 G/G 1 A/A 1 T/T 1 3 Susceptible 2.29 Moderate PBC 154 C0937 A/G 0.5 G/A 0.5 A/A 0 1 Moderate 2.22 Moderate DESA 2 C0939 G/G 1 A/A 1 T/T 1 3 Susceptible 2.26 Moderate DESA CT100 C0942 A/G 0.5 G/A 0.5 A/A 0 1 Moderate 2.81 Moderate PA 150 C0944 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 2.91 Moderate SCA 6 C0943 G/G 1 A/A 1 T/T 1 3 Susceptible 2.83 Moderate PA 30 C0945 A/A 0 G/A 0.5 A/A 0 0.5 Resistant 2.91 Moderate QH 1003 C0946 A/A 0 G/A 0.5 A/A 0 0.5 Resistant 2.97 Moderate DESA E93 C0947 G/G 1 A/A 1 NO AMP - - - 2.55 Moderate QH 1560 C0948 G/G 1 A/A 1 T/T 1 3 Susceptible 2.69 Moderate TG 148 C0949 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 2.48 Moderate DESA 101 C0950 G/G 1 A/A 1 T/T 1 3 Susceptible 2.79 Moderate KKM 25* C0951 A/A 0 G/G 0 A/A 0 0 Resistant 2.57 Moderate PBC 179 C0952 G/G 1 A/A 1 T/T 1 3 Susceptible 2.41 Moderate DESA 103 C0953 G/G 1 A/A 1 T/T 1 3 Susceptible 2.73 Moderate DESA 1 C0954 G/G 1 A/A 1 T/T 1 3 Susceptible 2.43 Moderate KKM 5 C0955 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 2.55 Moderate QH 37 C0956 G/G 1 A/A 1 T/T 1 3 Susceptible 2.32 Moderate QH 794* C0957 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 2.44 Moderate DESA SP7 C0958 G/G 1 A/A 1 T/T 1 3 Susceptible 2.56 Moderate PBC 230* C0959 A/A 0 G/G 0 A/A 0 0 Resistant 2.56 Moderate

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Real-time PCR Results Sample ADSI from Phenotype Clone Name EOH (A>G) IB1 (G>A) GH9 (A>T) Total ID Phenotype MCB from MCB Genotype Value Genotype Value Genotype Value Value

PA 107* C0960 G/G 1 A/A 1 T/T 1 3 Susceptible 2.63 Moderate KKM 28 C0961 A/G 0.5 G/G 0 A/A 0 0.5 Resistant N/A - QH 1176 C0962 G/G 1 A/A 1 T/T 1 3 Susceptible N/A - PBC 159 C0963 A/G 0.5 G/G 0 A/A 0 0.5 Resistant N/A - QH 670 C0964 A/G 0.5 G/G 0 A/A 0 0.5 Resistant N/A - KKM 4 C0965 A/G 0.5 G/G 0 A/A 0 0.5 Resistant N/A - QH 326 C0966 G/G 1 A/A 1 T/T 1 3 Susceptible N/A - KKM 1 C0967 A/G 0.5 G/G 0 A/A 0 0.5 Resistant N/A - TG 266 C0968 A/G 0.5 G/G 0 T/T 1 1.5 Moderate N/A - KKM 24 C0969 A/G 0.5 G/G 0 A/A 0 0.5 Resistant N/A - KKM 15 C0970 A/A 0 G/G 0 A/A 0 0 Resistant N/A - KKM 19 C0971 A/A 0 G/G 0 A/A 0 0 Resistant N/A - LKM J 25 C0972 A/G 0.5 G/G 0 A/A 0 0.5 Resistant N/A - EEN 202-S1 C0973 A/G 0.5 G/G 0 A/A 0 0.5 Resistant N/A - ICS 32 C0974 A/A 0 G/G 0 A/A 0 0 Resistant N/A - MCB C11 C0975 G/G 1 A/A 1 T/T 1 3 Susceptible N/A - MCB C14 C0976 A/G 0.5 G/G 0 A/A 0 0.5 Resistant N/A - MCB C10 C0977 A/G 0.5 G/G 0 A/A 0 0.5 Resistant N/A - AMH 2 C0978 A/G 0.5 G/G 0 A/A 0 0.5 Resistant N/A - EEN 203_S1 C0979 A/A 0 G/G 0 A/A 0 0 Resistant N/A - MHP 437 C0980 A/G 0.5 G/G 0 A/A 0 0.5 Resistant N/A - MCB C13 C0981 A/G 0.5 G/A 0.5 A/A 0 1 Moderate N/A - Cas 1 C0982 G/G 1 A/A 1 T/T 1 3 Susceptible N/A - CL 19/51 C0983 G/G 1 A/A 1 T/T 1 3 Susceptible N/A - ACT 2/8 C0984 A/A 0 G/A 0.5 A/A 0 0.5 Resistant N/A -

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Real-time PCR Results Sample ADSI from Phenotype Clone Name EOH (A>G) IB1 (G>A) GH9 (A>T) Total ID Phenotype MCB from MCB Genotype Value Genotype Value Genotype Value Value

22 P C0985 A/A 0 G/A 0.5 A/A 0 0.5 Resistant N/A - BE 2 C0986 G/G 1 A/A 1 T/T 1 3 Susceptible N/A - 23 P C0987 A/A 0 G/A 0.5 A/A 0 0.5 Resistant N/A - EEN 302/S201 C0988 A/A 0 G/A 0.5 A/A 0 0.5 Resistant N/A - EEN 158 C C0989 G/G 1 A/A 1 T/T 1 3 Susceptible N/A - EEN 127 C0990 G/G 1 A/A 1 T/T 1 3 Susceptible N/A - EEN 162/1010 C0991 G/G 1 A/A 1 T/T 1 3 Susceptible N/A - SC 1 C0992 A/A 0 G/A 0.5 A/A 0 0.5 Resistant N/A - PJ 21 C0993 A/A 0 G/A 0.5 A/A 0 0.5 Resistant N/A - ICS 46 C0994 NO AMP - A/A 1 T/T 1 - - N/A - MX 75/3 C0995 A/G 0.5 G/G 0 A/A 0 0.5 Resistant N/A - Cas 3 C0996 A/A 0 G/G 0 A/A 0 0 Resistant N/A - IMC 55 C0997 A/A 0 G/A 0.5 A/A 0 0.5 Resistant N/A - EEN 258 H C0998 G/G 1 A/A 1 T/T 1 3 Susceptible N/A - ICS 1 C0999 G/G 1 A/A 1 T/T 1 3 Susceptible N/A - EET 156 C1000 A/G 0.5 G/A 0.5 A/A 0 1 Moderate N/A - HY 271419 C1001 A/G 0.5 G/A 0.5 A/A 0 1 Moderate N/A - EQX 69 C1002 A/A 0 G/A 0.5 A/A 0 0.5 Resistant N/A - CC 49 C1003 A/G 0.5 G/A 0.5 A/A 0 1 Moderate N/A - EET 040 C1004 A/G 0.5 G/A 0.5 A/A 0 1 Moderate N/A - EET 064 C1005 A/A 0 G/G 0 A/A 0 0 Resistant N/A - EQX 3360/3 C1006 G/G 1 A/A 1 T/T 1 3 Susceptible N/A - ACT 2/5 C1007 NO AMP - A/A 1 T/T 1 - - N/A - ICS 42 C1008 A/A 0 G/G 0 A/A 0 0 Resistant N/A - GA 57 C1009 G/G 1 A/A 1 T/T 1 3 Susceptible N/A -

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Real-time PCR Results Sample ADSI from Phenotype Clone Name EOH (A>G) IB1 (G>A) GH9 (A>T) Total ID Phenotype MCB from MCB Genotype Value Genotype Value Genotype Value Value

GS 17 C1010 A/A 0 G/A 0.5 A/A 0 0.5 Resistant N/A - HY 2714 18-6 C1011 A/A 0 G/A 0.5 A/A 0 0.5 Resistant N/A - IML 44 DM C1012 A/A 0 G/A 0.5 A/A 0 0.5 Resistant N/A - ICS 39 DM C1013 A/A 0 G/A 0.5 A/A 0 0.5 Resistant N/A - ACT 11 C1014 A/A 0 G/A 0.5 A/A 0 0.5 Resistant N/A - C12 18 C1015 G/G 1 NO AMP - T/T 1 - - N/A - UF 1/22 C1016 A/A 0 G/A 0.5 A/A 0 0.5 Resistant N/A - AMA 2 12-4 C1017 G/G 1 A/A 1 T/T 1 3 Susceptible N/A - ICS 129 C1018 A/A 0 G/A 0.5 A/A 0 0.5 Resistant N/A - EEN 241 or 163 C1019 A/A 0 G/G 0 A/A 0 0 Resistant N/A - B BR25 C1020 G/G 1 A/A 1 T/T 1 3 Susceptible N/A - BR25 x PBC C1021 G/G 1 A/A 1 T/T 1 3 Susceptible N/A - 123 BR25 x PBC C1022 G/G 1 A/A 1 T/T 1 3 Susceptible N/A - 123 BR25 x NA3 C1023 G/G 1 A/A 1 T/T 1 3 Susceptible N/A - UD 22 x ICS 95 C1024 G/G 1 A/A 1 T/T 1 3 Susceptible N/A - UD 22 x ICS 95 C1025 A/G 0.5 G/G 0 A/A 0 0.5 Resistant N/A -

BR25 x NA33 C1026 NO AMP - A/A 1 T/T 1 - - N/A - ICS 40 C1027 G/G 1 A/A 1 T/T 1 3 Susceptible 3.09 Susceptible ICS 45 C1028 A/A 0 G/G 0 A/A 0 0 Resistant 2.66 Moderate ICS 55 C1029 A/A 0 G/G 0 A/A 0 0 Resistant 1.74 Resistant ICS 75 C1030 A/A 0 G/G 0 A/A 0 0 Resistant 3.66 Susceptible ICS 84 C1031 A/A 0 G/G 0 A/A 0 0 Resistant 2.43 Moderate IMC 14 C1032 A/A 0 G/G 0 A/A 0 0 Resistant 3.35 Susceptible

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Real-time PCR Results Sample ADSI from Phenotype Clone Name EOH (A>G) IB1 (G>A) GH9 (A>T) Total ID Phenotype MCB from MCB Genotype Value Genotype Value Genotype Value Value

IMC 23 C1033 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 2.47 Moderate IMC 60 C1034 A/A 0 NO AMP - A/A 0 - - 3.18 Susceptible KKM 1 C1035 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 2.28 Moderate KKM 22 C1036 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 0.75 Resistant KKM 25 C1037 A/A 0 G/G 0 A/A 0 0 Resistant 2 Resistant KKM 4 C1038 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 1.03 Resistant NA 162 C1039 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 2.1 Moderate NA 214 C1040 G/G 1 A/A 1 T/T 1 3 Susceptible 2.1 Moderate NA 226 C1041 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 3.09 Susceptible NA 235 C1042 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 0.87 Resistant NA 250 C1043 G/G 1 A/A 1 T/T 1 3 Susceptible 2.7 Moderate NA 33 C1044 A/A 0 G/G 0 A/A 0 0 Resistant 2.15 Moderate PA 107 C1045 A/A 0 G/G 0 A/A 0 0 Resistant 2.02 Moderate PA 150 C1046 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 3 Susceptible PA 156 C1047 G/G 1 NO AMP - T/T 1 - - 2.98 Moderate PA 171 C1048 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 1.98 Resistant PA 181 C1049 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 3.48 Susceptible PA 48 C1050 G/G 1 A/A 1 T/T 1 3 Susceptible 3.16 Susceptible PA 7 C1051 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 1.5 Resistant PBC 112 C1052 G/G 1 A/A 1 T/T 1 3 Susceptible 1.59 Resistant PBC 123 C1053 G/G 1 A/A 1 T/T 1 3 Susceptible 1.56 Resistant PBC 137 C1054 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 2.46 Moderate PBC 154 C1055 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 1.57 Resistant PBC 159 C1056 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 1.88 Resistant PBC 221 C1057 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 1.09 Resistant

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Real-time PCR Results Sample ADSI from Phenotype Clone Name EOH (A>G) IB1 (G>A) GH9 (A>T) Total ID Phenotype MCB from MCB Genotype Value Genotype Value Genotype Value Value

PBC 230 C1058 A/A 0 G/G 0 A/A 0 0 Resistant 1.8 Resistant QH 1003 C1059 A/A 0 G/G 0 A/A 0 0 Resistant 1.12 Resistant QH 1176 C1060 G/G 1 A/A 1 T/T 1 3 Susceptible 0.64 Resistant QH 1213 C1061 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 2.55 Moderate QH 22 C1062 A/A 0 NO AMP - A/A 0 - - 1.11 Resistant QH 326 C1063 G/G 1 NO AMP - T/T 1 - - 1.8 Resistant QH 794 C1064 A/G 0.5 G/G 0 A/A 0 0.5 Resistant 2.78 Moderate R 106 C1065 A/A 0 G/G 0 A/A 0 0 Resistant 3.49 Susceptible R 13 C1066 A/A 0 G/G 0 A/A 0 0 Resistant 3.54 Susceptible R 21 C1067 A/A 0 G/G 0 A/A 0 0 Resistant 3.55 Susceptible R 30 C1068 A/A 0 G/G 0 A/A 0 0 Resistant 3.76 Susceptible R 41 C1069 A/A 0 G/G 0 A/A 0 0 Resistant 3.33 Susceptible R 78 C1070 A/A 0 G/G 0 A/A 0 0 Resistant 2.37 Moderate SCA 19 C1071 G/G 1 A/A 1 T/T 1 3 Susceptible 2.86 Moderate SCA 20 C1072 G/G 1 A/A 1 T/T 1 3 Susceptible 2.5 Moderate TG 266 C1073 G/G 1 A/A 1 T/T 1 3 Susceptible 3.2 Susceptible

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

DISCUSSION generated by using all SNPs identified. As shown in Figure 4, the resistant and susceptible clones are The three SNP assays tested on the 14 control dispersed about the tree indicating that these 14 samples in this study showed that they are able to known clones are from random sampling and the distinguish phenotypes based on genotype, and three SNPs that were selected were not due to were assumed to show similar efficiency when population effect. Moreover, the SNPs are located applied to other unknown cocoa clones. However, within a few kb of each other, further when these SNPs were tested on a larger group of strengthening the likelihood of them being real 137 samples, only 37.5% of clones had matching markers of resistance. The positioning of these phenotype and genotype. three SNPs in the intron of a potential hexose symporter is intriguing. The closest matching gene To ensure that the SNPs found from the in Arabidopsis is a sugar transporter expressed 14 samples were not caused by being closely only in pollen. It is possible this major facilitator genetically related to each other, and that the superfamily (MFS) symporter transports different samples selected were indeed from a random substrates than hexose (Remy and Duque, 2014). group of samples, a neighbor-joining tree was

Figure 4. Cocoa sample relatedness phylogram showing resistant and susceptible clones are randomly dispersed.

The inconsistency of the phenotype al., 2012). Methylation of the genome have also determined by genotyping and ADSI could be due been reported to affect resistance (Peng and to copy number variation of the SNPs in the Zhang, 2008; Dowen et. al., 2012; Sha et al., unknown clones. In soybean, different copy 2005; Stokes et al., 2002). Other possibilities numbers of the resistance genes conferred different include the growth condition of each clone as they degrees of resistance towards nematodes (Cook et were sourced from different locations; clone origin

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

of each clone; and the nature of how ADSI is 10.1126/science.1228746. Epub 2012 Oct scored, that is, it being a qualitative (observation 11. of damage on the cocoa pod) rather than a quantitative measure could all be potential reasons Dowen, R.H., Pelizzola, M., Schmitz, R.J., Lister, for the observation. R., Dowen, J.M., Nery, J.R., Dixon, J.E. and Ecker, J.R. (2012). Widespread dynamic DNA methylation in response to CONCLUSIONS biotic stress. Proc Natl Acad Sci U S A. 109(32): E2183-91.doi:10.1073/pnas. The SNP assays AH0JEOH, AHX1IB1 and 1209329109. Epub 2012 Jun 25. AHZAGH9 have been designed based on the 14 previously sequenced cultivars and were able to Peng, H. and Zhang, J. (2008). Plant genomic clearly distinguish allelic differences among these DNA methylation in response to stresses: samples to categorize the cultivars correctly as Potential applications and challenges in either resistant or susceptible plants. However, the plant breeding. Progress in Natural same efficiency was not observed in the larger Science, 19 (2009): 1037–1045 sample of 137 clones, highlighting a possibility that the resistance to cocoa pod borer is conferred Remy, E. and Duque, P. (2014). Beyond cellular by multiple genes. Increasing the number of detoxification: a plethora of physiological known clones to be used in the SNP calling roles for MDR transporter homologs in process will also increase the possibility of plants. Front Physiol. 5: 201. doi: correctly identifying the gene(s) involved in cocoa 10.3389/fphys.2014.00201. eCollection pod borer resistance. The inconsistency of the 2014. phenotype determined by genotyping and ADSI could be due to copy number variation of the SNPs Roslina, M.S., Rosmin, K., Lea, J., Navies, M., in the unknown clones. Other possibilities include Nor Aisyah, Sumayyah, A.K, Zafirah, Z., the growth condition of each clone as they were Fahmie, W. and Larry, C. (2014). sourced from different locations; clone origin of Identification of genetic markers for pod each clone; and the nature of how ADSI is scored, resistance by sequencing 14 cocoa that is, it being a qualitative (observation of genomes. Malaysian Cocoa J. 8: 40–48. damage on the cocoa pod) rather than a quantitative measure could all be potential reasons Sha, A.H., Lin, X.H., Huang, J.B. and Zhang, D.P. for the observation. (2005). Analysis of DNA methylation related to rice adult plant resistance to bacterial blight based on methylation- REFERENCES sensitive AFLP (MSAP) analysis. Mol Genet Genomics. 273(6): 484-90. Epub Cook, D.E., Lee, T.G., Guo, X., Melito, S., Wang, 2005 Jun 21. K., Bayless, A.M., Wang, .J, Hughes, T.J., Willis, D.K, Clemente, T.E., Diers, Stokes, T.L., Kunkel, B.N. and Richards, E.J. B.W., Jiang, J., Hudson, M.E. and Bent, (2002). Epigenetic variation in A.F. (2012). Copy number variation of Arabidopsis disease resistance. Genes multiple genes at Rhg1 mediates Dev. 16(2): 171-82. nematode resistance in soybean. Science. 2012 Nov 30: 338(6111):1206-9. doi:

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Appendix Table A1: QC results of DNA extracted from all 151 cocoa clones samples used in this study.

Nanodrop Clone Sample MGRC Intact Vol (µL) Conc Name ID Barcode A260/280 A260/230 DNA (ng/µL) QH441 C0923 C0923 30 15.5 1.73 1.59 Yes KKM22 C0924 C0924 30 29.9 1.82 1.01 Yes NA33 C0925 C0925 30 14.4 1.64 1.38 Yes QH22 C0926 C0926 30 33.3 1.84 2.08 Yes DESA104 C0927 C0927 30 98.7 1.84 2.22 Yes TG149 C0928 C0928 30 43.9 1.85 2.14 Yes KKM2 C0929 C0929 30 109.6 1.87 2.19 Yes PBC123 C0930 C0930 30 26.2 1.79 1.48 Yes PBC221 C0931 C0931 30 55.7 1.85 2.24 Yes DESA102 C0932 C0932 30 21 1.95 1.8 Yes PBC112 C0933 C0933 30 55.7 1.81 1.95 Yes DESA105 C0934 C0934 30 49.5 1.8 1.73 Yes KKM21 C0935 C0935 30 174.1 1.83 2.31 Yes TG146 C0936 C0936 30 83.3 1.68 1.15 Yes ICS 84 C0940 C0940 30 31.7 1.81 1.75 Yes PBC 137 C0938 C0938 30 40.7 1.89 2.11 Yes PBC 132 C0941 C0941 30 118.8 1.81 1.77 Yes PBC 154 C0937 C0937 30 37.7 1.83 1.65 Yes DESA 2 C0939 C0939 30 30 1.79 1.54 Yes DESA C0942 C0942 30 30.6 1.8 1.47 Yes CT100 PA 150 C0944 C0944 30 45 1.85 1.45 Yes SCA 6 C0943 C0943 30 70.9 1.81 1.73 Yes PA 30 C0945 C0945 30 14.3 1.79 1.28 Yes QH 1003 C0946 C0946 30 47.4 1.83 1.92 Yes DESA E93 C0947 GSA00023 50 31 1.87 1.69 Yes QH 1560 C0948 GSA00024 50 17 1.86 2 Yes TG 148 C0949 GSA00025 50 17.2 1.86 1.58 Yes DESA 101 C0950 GSA00026 50 19.6 1.82 1.62 Yes KKM 25* C0951 GSA00027 50 34.7 1.87 1.93 Yes PBC 179 C0952 GSA00028 50 49.5 1.88 1.86 Yes DESA 103 C0953 GSA00029 50 18.6 2.01 1.6 Yes DESA 1 C0954 GSA00030 50 18.3 1.86 1.44 Yes KKM 5 C0955 GSA00031 50 25 1.82 1.07 Yes QH 37 C0956 GSA00032 50 25.7 1.86 1.88 Yes QH 794* C0957 GSA00033 50 14.8 1.73 1.98 Yes DESA SP7 C0958 GSA00034 50 18.2 1.84 1.57 Yes

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Nanodrop Clone Sample MGRC Intact Vol (µL) Conc Name ID Barcode A260/280 A260/230 DNA (ng/µL) PBC 230* C0959 GSA00035 50 20 1.89 1.61 Yes PA 107* C0960 GSA00036 50 17.7 1.92 1.69 Yes KKM 28 C0961 GSA00037 50 32.1 1.84 1.62 Yes QH 1176 C0962 GSA00038 50 20.8 1.81 1.79 Yes PBC 159 C0963 GSA00039 50 24.8 1.88 1.86 Yes QH 670 C0964 GSA00040 50 20.1 1.79 1.3 Yes KKM 4 C0965 GSA00041 50 14.2 1.86 1.96 Yes QH 326 C0966 GSA00042 50 21.5 1.86 1.94 Yes KKM 1 C0967 GSA00043 50 47 1.86 1.77 Yes TG 266 C0968 GSA00044 50 18.1 1.81 1.58 Yes KKM 24 C0969 GSA00045 50 13 1.95 1.63 Yes KKM 15 C0970 GSA00046 50 21.9 1.94 2.01 Yes KKM 19 C0971 GSA00047 50 15.1 1.93 1.92 Yes LKM J 25 C0972 GSA00048 50 68.1 1.89 2.34 Yes EEN 202- C0973 GSA00049 50 25.6 1.87 2.14 Yes S1 ICS 32 C0974 GSA00050 50 253.7 1.85 2.36 Yes MCB C11 C0975 GSA00051 50 28.2 1.89 2.05 Yes MCB C14 C0976 GSA00052 50 125.9 1.86 2.29 Yes MCB C10 C0977 GSA00053 50 91.3 1.85 2.24 Yes AMH 2 C0978 GSA00054 50 77.4 1.87 2.3 Yes EEN C0979 GSA00055 50 73.6 1.85 2 Yes 203_S1 MHP 437 C0980 GSA00056 50 40.8 1.86 1.96 Yes MCB C13 C0981 GSA00057 50 83.3 1.85 2.18 Yes Cas 1 C0982 GSA00058 50 26.3 1.82 1.65 Yes CL 19/51 C0983 GSA00059 50 75.8 1.86 2.22 Yes ACT 2/8 C0984 GSA00060 50 172.1 1.83 2.36 Yes 22 P C0985 GSA00061 50 50.8 1.85 2.06 Yes BE 2 C0986 GSA00062 50 55.5 1.88 2.18 Yes 23 P C0987 GSA00063 50 32.8 1.83 2.05 Yes EEN C0988 GSA00064 50 72.7 1.87 2.17 Yes 302/S201 EEN 158 C C0989 GSA00065 50 135 1.84 2.3 Yes EEN 127 C0990 GSA00066 50 82.2 1.86 2.11 Yes EEN C0991 GSA00067 50 74.7 1.85 2.01 Yes 162/1010 SC 1 C0992 GSA00068 50 224.2 1.84 2.35 Yes PJ 21 C0993 GSA00069 50 78.7 1.87 2.18 Yes ICS 46 C0994 GSA00070 50 97.4 1.87 2.4 Yes

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Nanodrop Clone Sample MGRC Intact Vol (µL) Conc Name ID Barcode A260/280 A260/230 DNA (ng/µL) MX 75/3 C0995 GSA00071 50 48.6 1.89 2.32 Yes Cas 3 C0996 GSA00072 50 29.5 1.93 2.05 Yes IMC 55 C0997 GSA00073 50 89.8 1.89 2.36 Yes EEN 258 C0998 GSA00074 50 116 1.86 2.26 Yes H ICS 1 C0999 GSA00075 50 42.5 1.88 2.21 Yes EET 156 C1000 GSA00076 50 162.1 1.85 2.39 Yes HY C1001 GSA00077 50 52.2 1.88 2.27 Yes 271419 EQX 69 C1002 GSA00078 50 57.2 1.88 2.17 Yes CC 49 C1003 GSA00079 50 39.9 1.91 2.31 Yes EET 040 C1004 GSA00080 50 119.4 1.85 2.38 Yes EET 064 C1005 GSA00081 50 141.5 1.85 2.35 Yes EQX C1006 GSA00082 50 77.3 1.88 2.27 Yes 3360/3 ACT 2/5 C1007 GSA00083 50 83.7 1.86 2.2 Yes ICS 42 C1008 GSA00084 50 169.9 1.85 2.36 Yes GA 57 C1009 GSA00085 50 129.7 1.86 2.32 Yes GS 17 C1010 GSA00086 50 97.1 1.87 2.35 Yes HY 2714 C1011 GSA00087 50 71.5 1.89 2.33 Yes 18-6 IML 44 C1012 GSA00088 50 52.2 1.87 2.12 Yes DM ICS 39 C1013 GSA00089 50 97.8 1.87 2.27 Yes DM ACT 11 C1014 GSA00090 50 85.5 1.88 2.2 Yes C12 18 C1015 GSA00091 50 42.3 1.85 2.18 Yes UF 1/22 C1016 GSA00092 50 177.8 1.84 2.39 Yes AMA 2 C1017 GSA00093 50 149.7 1.86 2.32 Yes 12-4 ICS 129 C1018 GSA00094 50 62.4 1.86 2.34 Yes EEN 241 C1019 GSA00095 50 59.5 1.88 2.23 Yes or 163 B BR25 C1020 GSA00096 50 79.6 1.87 2.27 Yes BR25 x C1021 GSA00097 50 52.7 1.86 2.27 Yes PBC 123 BR25 x C1022 GSA00098 50 116.9 1.86 2.33 Yes PBC 123 BR25 x C1023 GSA00099 50 44.5 1.89 2.17 Yes NA3 UD 22 x C1024 GSA00100 50 33 1.94 2.12 Yes ICS 95 UD 22 x C1025 GSA00101 50 55.3 1.86 2.14 Yes ICS 95

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Nanodrop Clone Sample MGRC Intact Vol (µL) Conc Name ID Barcode A260/280 A260/230 DNA (ng/µL) BR25 x C1026 GSA00102 50 33.4 1.88 2.22 Yes NA33 ICS 40 C1027 GSA00108 50 32.9 1.81 2.2 Yes ICS 45 C1028 GSA00109 50 17.4 1.76 1.96 Yes ICS 55 C1029 GSA00110 50 51.8 1.87 2.31 Yes ICS 75 C1030 GSA00111 50 114.8 1.84 2.29 Yes ICS 84 C1031 GSA00112 50 42.4 1.84 1.88 Yes IMC 14 C1032 GSA00113 50 53 1.83 1.91 Yes IMC 23 C1033 GSA00114 50 24.2 1.83 2.18 Yes IMC 60 C1034 GSA00115 50 27.9 1.85 2.14 Yes KKM 1 C1035 GSA00116 50 116.5 1.84 2.13 Yes KKM 22 C1036 GSA00117 50 97.4 1.86 2.37 Yes KKM 25 C1037 GSA00118 50 51.1 1.82 1.93 Yes KKM 4 C1038 GSA00119 50 17.4 1.83 1.41 Yes NA 162 C1039 GSA00120 50 28 1.84 1.89 Yes NA 214 C1040 GSA00121 50 32.6 1.82 1.62 Yes NA 226 C1041 GSA00122 50 103.2 1.85 2.22 Yes NA 235 C1042 GSA00123 50 48.5 1.85 1.94 Yes NA 250 C1043 GSA00124 50 16.3 1.93 1.83 Yes NA 33 C1044 GSA00125 50 76.3 1.85 2.19 Yes PA 107 C1045 GSA00126 50 113.8 1.86 2.33 Yes PA 150 C1046 GSA00127 50 103 1.84 2.24 Yes PA 156 C1047 GSA00128 50 19.1 1.9 2.28 Yes PA 171 C1048 GSA00129 50 14.9 1.77 1.74 Yes PA 181 C1049 GSA00130 50 83.1 1.88 2.36 Yes PA 48 C1050 GSA00131 50 92.7 1.82 1.98 Yes PA 7 C1051 GSA00132 50 97.3 1.83 1.97 Yes PBC 112 C1052 GSA00133 50 65 1.83 1.99 Yes PBC 123 C1053 GSA00134 50 131.7 1.83 2.24 Yes PBC 137 C1054 GSA00135 50 152.3 1.85 2.3 Yes PBC 154 C1055 GSA00136 50 72.9 1.82 1.82 Yes PBC 159 C1056 GSA00137 50 159.2 1.83 2.2 Yes PBC 221 C1057 GSA00138 50 40.9 1.87 2.15 Yes PBC 230 C1058 GSA00139 50 82.8 1.77 1.48 Yes QH 1003 C1059 GSA00140 50 121.9 1.84 2.21 Yes QH 1176 C1060 GSA00141 50 168.1 1.83 2.25 Yes QH 1213 C1061 GSA00142 50 65.6 1.86 2.25 Yes QH 22 C1062 GSA00143 50 76.1 1.84 1.97 Yes QH 326 C1063 GSA00144 50 27.5 1.95 2.4 Yes

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Nanodrop Clone Sample MGRC Intact Vol (µL) Conc Name ID Barcode A260/280 A260/230 DNA (ng/µL) QH 794 C1064 GSA00145 50 58.2 1.86 2.1 Yes R 106 C1065 GSA00146 50 36.1 1.89 1.98 Yes R 13 C1066 GSA00147 50 77.7 1.86 2.2 Yes R 21 C1067 GSA00148 50 122.8 1.84 2.2 Yes R 30 C1068 GSA00149 50 205.6 1.84 2.3 Yes R 41 C1069 GSA00150 50 229 1.84 2.33 Yes R 78 C1070 GSA00151 50 116.1 1.84 2.28 Yes SCA 19 C1071 GSA00152 50 110.9 1.85 2.08 Yes SCA 20 C1072 GSA00153 50 53.9 1.9 2.29 Yes TG 266 C1073 GSA00154 50 72.7 1.86 2.26 Yes

Figure A1. Gel electrophoresis images of 151 known and unknown cocoa clones as run 1% agarose gel. Ladders used were 1kb (NEB, USA) and HindIII (NEB, USA).. continued

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Figure A1. Gel electrophoresis images of 151 known and unknown cocoa clones as run 1% agarose gel. Ladders used were 1kb (NEB, USA) and HindIII (NEB, USA).

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

IDENTIFICATION OF THE POTENTIAL RESISTANCE GENES IN Theobroma cacao TO COCOA POD BORER INSECT (Conopomorpha cramerella) IN RELATION TO HOST- PATHOGEN INTERACTION USING RNA-SEQ TECHNOLOGY

Roslina, M.S1., David, A.1, Anisah, S.1, Navies, M.2, Kasran, R.1, Mellisa, G.3, Fahmi, W.3, Larry, C.3 1Malaysian Cocoa Board, Cocoa Biotechnology Research Centre, Kota Kinabalu Industrial Park, 88460 Kota Kinabalu, Sabah, Malaysia 2Malaysian Cocoa Board, Centre of Cocoa Research and Development, Mile 10, Apas Road P.O.Box 60237, 91012 Tawau, Sabah 3Malaysian Genomics Resource Centre Berhad (652790-V), 27-9, Level 9 Boulevard Signature Office, Mid Valley City, 59200 Kuala Lumpur, Malaysia

Malaysian Cocoa Journal 9(1): 40-49 (2016) ABSTRACT - Plants developed different mechanisms to reduce insect attack, including specific responses that activate different metabolic pathways which considerably alter their chemical and physical aspects. On the other hand, insects developed several strategies to overcome plant defense barriers, allowing them to feed, grow and reproduce on their host plants. The Cocoa Pod Borer (CPB), also known as cocoa moth, is caused by the insect Conopomorpha cramerella, causing vast losses in the cocoa industry during the 1890s and 1900s. CPB attacks both young and mature cocoa pods. A common symptom of infested pods is unevenness and premature ripening. The transcriptome was sequenced from Theobroma cacao tissues (pod husk and placenta) of resistant and susceptible clones, before and after infection. The unique characterisation of the differences observed; if any; in pod husk, placenta and other relevant tissues may be used to understand the molecular variations underlying the phenotypic and disease resistance characteristics, and relevant molecular markers identified can be used in downstream breeding research. P450 cytochrome gene, an aluminium detoxification genes, sugar transporter genes, defensin genes are among those found to be upregulated or downregulated in resistance or susceptible clone. The resistant clone also synthesises more enzymes for phytoalexins that was known to inhibit insect feeding. The cocoa defense is comprised of a set of chitinases for digesting the CPB larva’s cuticle and possibly phytoalexins which may deter the borers with unpleasant flavours. The whole defense system for any threat appears to be more active in the resistant clones.

Key words: Cocoa, Cocoa pod borer, Resistance, Susceptible, Genes

INTRODUCTION: reasons behind the phenotypic and disease resistance characteristics, and relevant molecular The Cocoa Pod Borer (CPB), also known as markers may be identified that can be used in cocoa moth, is caused by the insect downstream breeding research (Wang et al., Conopomorpha cramerella, (Ooi, 2005) causing 2009; Salgotra et al., 2014) . The larger the vast losses in the cocoa industry during the number of samples for resistant and susceptible 1890s and 1900s (Valenzuela et al., 2014). CPB plant transcriptomes sequenced, the higher the attacks both young and mature Theobroma probability there is to identify biologically cacao pods (Ooi., 2005). To gain insights into relevant gene expression differences between the genetic basis of the natural resistance of resistant and susceptible plants. These gene cocoa clones to CPB attacks, comparison expression differences then can be correlated between transcriptional response to the with phenotypic characters. controlled CPB attacks of a susceptible and resistant clones by RNAseq analysis fifteen days after infection were performed (Chen et al., 2013; Gupta et al., 2015). The unique characterisation of the differences observed; if any; in pod husk, placenta and other relevant tissues may be used to understand the molecular

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

MATERIALS AND METHODS Gel Electrophoresis Gel Electrophoresis

Rearing of CPB moths and selection of suitable pods for caging and controlled infection were performed before any laboratory works to be done. Resistant and susceptible clones, KKM22 and PBC123 (Malaysian Cocoa Clones, 2012; Haya et al., 2007) were selected and the controlled infection were done in one cocoa farmers plot in Kota Marudu. After fifteen days of infection with CPB moths, pods were collected and divided into tissues of pod husk Result: PBC 123 Result: KKM 22 and placenta of resistant and susceptible clones, before and after infection (Tan et al., 2009). L1: Pod Husk (control), Transcriptomes were performed for samples L1: Pod Husk (control), L2: Pulp (control), L4: listed as below; L2: Pulp (control), L4: Pod Husk (CPB Pod Husk (CPB infected), L5: Pulp (CPB a) Resistant, pod husk, before infection. infected), L5: Pulp (CPB b) Resistant, pod placenta, before infected) infected) infection. c) Resistant, pod husk, after infection. Figure 1. RNA extraction of different tissues d) Resistant, pod placenta, after infection. from PBC 123 and KKM 22 e) Resistant, larvae + placenta, after infection. It is important to maintain high RNA f) Susceptible, pod husk, before infection integrity number (RIN) for each samples before g) Susceptible, pod placenta, before any deep sequencing to be performed. The infection. selected samples quality assessment results are h) Susceptible, pod husk, after infection. shown in Table 1. Next, the sample is processed i) Susceptible, pod placenta, after through messenger RNA isolation and cDNA infection. synthesis using NEBNext® Ultra™ RNA j) Susceptible, larvae + placenta, after Library Prep Kit for Illumina® according to infection. manufacturer’s protocol. Quantification of the k) Larvae from artificial diet. synthesized cDNA was made using Qubit 2.0 l) Larvae treated with caffeine from DNA Broad Range Assay (Invitrogen, USA). artificial diet. Sequencing A minimum of 10ng cDNA was fragmented RESULTS AND DISCUSSION using a Covaris S220 (Covaris Inc, USA) to a targeted size of 200 – 300 bp. The fragmented RNA extraction was performed for eight plant cDNA was then end-repaired, ligated to tissues samples and four RNA larvae were also NEBNext adapters, and was further PCR- extracted (Goh et al., 2014; Tan et al., 2009). All enriched using NEBNext® Ultra™ RNA Library extraction samples were subjected to quality Prep Kit according to manufacturer’s protocol. assessment process measured using QiaXpert Quantification of the final sequencing libraries (Qiagen) to verify the quality and quantity of the was performed using KAPA kit (KAPA RNA (Figure 1). Biosystem, USA) on Agilent Stratagene Mx- 3005p quantitative PCR (Agilent, USA). Sizes were confirmed using Agilent Bioanalyzer High Sensitivity DNA Chip (Agilent, USA). Finally, the libraries was sequenced using an Illumina flow cell with 209 cycles on the Illumina HiSeq 2000 platform (Illumina, USA). For this project, the sequencing run has generated a total number of 96.63 Gb raw data.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 1. Sample quality assessment results

Tube Sample Sample MGRC QiaXpert RIN No. Category Description barcode Conc A260/ A260/ Vol. Total (ng/ul) 280 230 (ug) 1 Cocoa Larvae from RNA KKM22 in pulp Cocoa_K2 680.700 2.030 1.530 50 34.040 9.6 Larvae infected with CPB 2 Cocoa Artificial Diet 1 Cocoa_A RNA RNA Larvae 708.100 1.790 2.130 50 39.910 9.1 D1 Larvae 3 Cocoa Caffeeine 1 Cocoa_CF RNA RNA Larvae 228.800 1.830 1.880 50 11.440 8.6 1 Larvae 4 Cocoa Larvae from RNA pbc123 in pulp Cocoa_5E 304.700 2.090 1.400 30 9.141 9.5 Larvae infected with CPB 5 Plant KKM22 KKM RNA Control POD Husk 84.300 1.440 0.840 30 1.260 8.5 HUSK 1 XCPB 6 Plant KKM22 CPB KKM RNA Infected POD 24.000 1.610 0.470 40 0.960 8.7 Husk CPB HUSK 1 7 Plant KKM22 KKM RNA Control POD Pulp 24.500 1.830 1.340 30 0.740 9.3 PULP 1 XCPB 8 Plant KKM22 CPB KKM RNA Infection POD 152.400 0.880 2.250 40 6.096 7.9 Pulp CPB PULP 9 Plant PBC 123 PBC Husk RNA Control POD 33.100 1.990 1.910 30 0.993 9.5 XCPB HUSK 1 10 Plant PBC 123 CPB PBC Husk RNA Infected POD 75.900 1.990 1.830 30 2.277 7.0 CPB HUSK 1 11 Plant PBC 123 PBC Pulp 115.000 1.740 1.080 30 3.450 8.7 RNA Control PULP 1 XCPB 12 Plant PBC 123 CPB PBC Pulp RNA Infected PULP 36.600 1.820 1.790 30 1.098 7.0 1 CPB

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Bioinformatic Stages Cocoa Transcriptome Assembly Statistics Table 2 showed stages of bioinformatics After the pre-processing stage, the clean reads analysis. from the data sets were assembled using the MGRC Transcriptome Pipeline. All reference Table 2. Stages of bioinformatics analysis. transcripts were generated by combining all clean reads of the Illumina sequencing data sets Process Description (Table 3). A simpler representation of transcripts length distribution identified in Cocoa Preprocessing of An assessment of transcriptome data sets were shown in Figure 2. Illumina pair-end sequence data data characteristics was conducted. This Table 3. Statistics of assembled transcripts assessment was used to identified in Cocoa transcriptome data sets. make informed decisions on how to best process the sequence library. In addition to removing lower base quality data, sequence reads with adaptor sequence were clipped and filtered based on quality. Sequences containing excessive numbers of ambiguous bases (N) were also removed.

De novo assembly of The sequence reads were sequenced reads assembled to form transcripts for a reference cocoa transcriptome.

Downstream analysis Assembled sequences were annotated using public protein databases. Gene expression of transcripts were analyzed to identify up-regulated and down-regulated differential expressed genes and to

characterisation and identification of the Figure 2. Grapical length distribution summary potential resistance genes of transcripts identified in cocoa transcriptome in Theobroma cacao to data sets. cocoa pod borer insect (Conopomorpha cramerella).

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Cocoa Transcriptome Alignment Statistics Gene Ontology (GO) Alignment statistics were reported from the Annotated transcripts were further categorized SAM format alignment files using custom Perl according to the three main GO domains: scripts (Table 4). A large number of the reads biological processes, cellular components, and (96.16%) aligned back to the transcripts as molecular function as shown in Table 6 and expected. The reads that did not map back to the Table 7. assembled transcripts corresponded to either low quality reads or lowly-expressed transcripts that Table 6. Characterization of annotated could not be assembled due to the minimum transcripts according to gene ontology domains. length requirement for transcripts (≥ 300nt). In this case, only a small portion of the reads (less than 3% of the reads) did not align back to the reference cocoa transcriptome.

Table 4. Reads alignment statistics of transcripts identified in cocoa transcriptome data sets

Differential Expression Analysis Paired-end reads were aligned back to the assembled transcripts (length ≥ 300bp) through the MGRC Transcriptome Pipeline. For expression analysis the MGRC Transcriptome Pipeline was used to estimate abundance of transcripts. The results showed that around 86% of the paired-end reads from the cocoa Downstream Analysis transcriptome datasets were mapped back to the In the final step, annotation and gene expression reference transcripts. The lower mapping analysis were performed to characterize and libraries around 73% are larvae samples. This is identify potential resistance genes in T. cacao likely due to the lower depth of transcriptome towards Cocoa Pod Borer insect. sequencing coupled with the many orders of magnitude increased complexity of the insect Annotation Analysis of Transcripts transcriptome. SynaSearch was used to map assembled transcripts against the UniProt database with a cutoff e-value of 1e-5 (Table 5) The best hit from each of these assembled transcript comparisons was used as the annotation reference for the respective transcripts.

Table 5. SynaSearch result summary of transcripts in the cocoa transcriptome.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 7. Top ten GO terms in each GO domain for Cocoa/Pod Borer transcriptome data sets.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

A Volcano plot of –Log10 (p-value) versus Log2FoldChange was plotted for each analyzed pair of transcriptomes (Figure 3). The Volcano plot assists in the identification of genes or transcripts with a strong statistically significant expression difference.

Figure 4. Heatmap of gene expression values show clustering (LHS dendrogram) of genes (RHS) between samples (kkm_husk_cpb vs pbc_HuSK_CPB) based on expression of mRNAs Figure 3. Differential expression analysis of (color differences) for a set of significant genes kkm_husk_cpb vs pbc_HuSK_CPB at the gene (padj < 1e-10). level. Differential Expression Analysis

The heatmap of Figure 4 is a scale Pod borer component of the transcriptome colour image for representing the observed values A P450 cytochrome was identified as of two different samples (kkm_husk_cpb vs differentially expressed between normally fed pbc_HuSK_CPB). The dendrogram at the side and caffeine fed (caffeine is a close analog of the shows a hierarchical clustering of the significant drug theobromine found in cocoa). This is likely genes (padj < 1e-10). In the main panel, the to be the P450 responsible for detoxification of sample names are represented in columns and the these purinergic xenobiotics from the insect significant genes are represented in rows. The (Magnanou et al., 2013). No differential clustering algorithm groups related rows together expression of this gene was found between the by expression similarity. Low expression values insects grazing on tolerant and susceptible clones. of a sample tend towards red in color while This suggests there is no chemical defense higher expression values tend to green in color. difference between tolerant and susceptible cultivars. An endogenous chemical insecticide is less likely to be the cause of cocoa clone resistance.

Other differentially expressed genes between borers fed caffeine and not are likely to be due to retarded lifestage (fibroin), irritation of the insect digestive system (intestinal mucin, peritrophin). The LigB-like gene may also be involved in caffeine metabolism, but it is

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

paradoxically down expressed in the caffeine fed A second cluster of three SNPs, which borers. didn’t quite pass statistical significance in the previous genome sequencing project, shows Cocoa component of the transcriptome differential expression between the clones. The Aluminium detoxification genes (Ryan et al., genomic region containing these SNPs is not 2011) are active in infected plants, but not in particularly well conserved, contains expressed uninfected plants. Low soil pH (under 4) regions which do not form long ORFs. Defensin mobilizes aluminium. The low pH or the uptake genes in plants and animals fit these criteria and of aluminium due to low pH may stress the cocoa in some cases there is 100 fold difference in plant and make both resistant and susceptible expression in this region between the susceptible plants more open to attack by borers. and tolerant clones. This locus, identified by genome sequencing, and verified by differential There are clear plant defence responses expression, may be potential defensin genes to borers in husk and pod in both resistant and which produce short peptides that inhibit insect susceptible clones. The susceptible clone has a attack. Much more works needs to be done to greater response than the resistant clone. The verify this finding. most likely explanation is that the quantity of larvae in the pod determines the strength of the General transcriptome differences between the response. The fecundity of each insects is susceptible and tolerant clones different within each other. The resistant clone Expression differences between the two clones has less borers and so the defense response is are small in the samples uninfected by borers. Big smaller. differences appear when the plant defense system becomes active after infection by borers. In This is suggestive of the resistant plants uninfected pulp, the resistant clone seems to have being physically more protected from borer attack higher oil production genes, possibly clone than the susceptible plants. There is a bigger differences in oil content unrelated to insect response in the susceptible plants as there are tolerance. more borers getting through. This fits current models that the husk is thicker and harder for the The susceptible clone has more terpene borers to penetrate in the resistant clones (Haya et biosynthesis activity in the husk than the resistant al., 2007). It may also be resistant plants are less clone. Terpenes are often used as plant defenses attractive to borers, or outer husk biochemical against insects and bacteria. Increased terpene defense is stronger, which would both limit borer production in the susceptible clone does not attack and hence plant defense activation. discourage the borers, so if the terpene biosynthesis is a defense against borers it is The cluster of three SNPs identified ineffectual. Increased adaptability of CPB in from previous genome sequencing of seven nature against terpenes as defence mechanism tolerant and seven susceptible clones is in a sugar might be one of the reason why the effect of transporter (Roslina et al., 2014). There were no terpenes were not as expected it to be. big differential expression changes between the tolerant and susceptible clones. These SNPs are Looking at a comparison of the borer in the intron of the sugar transporter and the sugar infected but tolerant clone and the borer infected transporter is lowly expressed in the susceptible susceptible clone there are many expression cultivar. It is possible increased sugar transport differences. Quite a number are clearly insect into the cell wall, means that the pods in the defence related. In pulp there are a set of resistant clone are less sweet owing to faster chitinases (for digesting insect cuticle or fungal metabolism of sugar. The lower free sugar in the hyphae) which are strongly upregulated in the pods may decrease the borers’ interest in resistant clone. These are upregulated along with attacking the pods. This hypothesis can be osmotin which is a plant defense gene for followed by a brix test of the pulp of tolerant and bacteria, suggesting a plant defense response susceptible clones. The small expression which is much more active in the resistant clone. difference may still be mechanistically relevant, This defense appears not to distinguish between or a SNP isoform of the gene is in linkage with bacteria and insects, or insect ingress brings in the SNPs. bacteria which the plant reacts to. In husk tissue

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

similar large differences are observable between mechanistically significant. The second locus has the clones. a large expression difference between the two cultivars but the function of the genes in this Potential pigment genes are upregulated region are unknown. The transcriptome of the in the resistant clone along with again, a set of cocoa pod borer has also been assembled from chitinases for defense against insect/fungal this dataset and a high confidence gene for predation. These pigment genes (anthocyanin) are xenobiotic resistance identified. usually related to fruit coloration typically red- skinned fruit. The chitinases appear to be effective insect inhibitors as an insect fat storage REFERENCES protein is strongly upregulated in the resistant sample, suggesting the borers are struggling in Chen, T., Lv, Y., Zhao, T., Li, N. and Yang, Y., the resistant clone, and fat and healthy in the (2013). Comparative transcriptome susceptible clone. profiling of a resistant vs. susceptible tomato (Solanum lycopersicum) cultivar The resistant clone also synthesises in response to infection by tomato more enzymes for phytoalexins (known to inhibit yellow leaf curl virus. PLoS ONE 8(11): insect feeding PMID: 24302045). These e80816. compounds are mostly antimicrobial and the lack of P450 response in the larvae fed with cocoa Goh, L.P.W., Chia, Y.C. and Tan, C.L., (2014). pulp suggests these phytoalexin-like compounds Isolation of DNA and RNA from Cocoa are not effective poisons against borers, but the Pod Borer, Conopomorpha cramerella phytoalexin response is are of the general (Snellen) and cloning its genes. response of the resistant clone to attack (whether Malaysian Cocoa Journal 8: 31-39 insect, fungi, or bacterial). Gupta, V., Estrada, A.D., Blakley, I., Reid, R., Components of the salicylic acid Patel, K., Meyer, M.D., Andersen, S.U., biosynthesis pathway are also upregulated in the Brown, A.F., Lila, M.A. and Loraine, resistant clone. This plant defense signalling A.E., (2015). RNA-Seq analysis and hormone and the enhancecd phytoalexin annotation of a draft blueberry genome biosynthesis suggests a more active defense assembly identifies candidate genes system is present in the resistant clone. The cocoa involved in fruit ripening, biosynthesis defense is comprised of a set of chitinases for of bioactive compounds, and stage- digesting the borer cuticle and possibly specific alternative splicing. phytoalexins which may deter the borers with GigaScience 4: 5 unpleasant flavours. A possible defensin locus is also upregulated. The whole defense system for Haya, R., Navies, M., Kelvin, L., Albert, L., any threat appears to be more active in the Alias, A., and Nuraziawati, M.Y., resistant clone. (2007). Screening for cocoa pod borer resistance based morphological descriptors. Presentation at the 2007 CONCLUSIONS Conference on Plantation Commodities, Putra World Trade Centre, Kuala This transcriptomic study of two clones of cocoa, Lumpur, Malaysia, 3-4 July 2007. one tolerant of cocoa pod borer attack and the other susceptible, has thrown a great deal of light Magnanou, E., Malenke, J.R. and Dearing, M.D., on the plant defense mechanisms involved in (2013). Hepatic gene expression in borer tolerance. The resistant clone appears to herbivores on diets with natural and have a heightened response to attack, but also novel plant secondary compounds. there is less activity of defensive genes, possibly Physiol Genomics 45: 774–785. due to lower burden of borers. Two loci identified by previous genetic methods show differential Ooi, P.A.C. (2005). Report of the APO Seminar expression. The first locus has a small expression on Nonpesticide Methods for difference which may or may not be

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Controlling Diseases and Insect Pests, Tan, C.L., Verica, J.A., Young, A., Pishak, S., APO 2005, ISBN: 92-833-7037-6. Maximova, S.N. and Guiltinan, M.J., (2009). Isolation of pod-specific genes Malaysian Cocoa Board (2012). Malaysian Cocoa in Theobroma cacao. Malaysian Cocoa Clones. Journal 5: 37-42.

Poczai1, P., Varga, I., Laos, M., Cseh, A., Bell, Teixeira, P.J.L., Lima, P.J.P., Thomazella, N., Valkonen, J.P.T. and Hyvönen, J., D.P.T., Prado, O.R.P.F., Rio, M.C.S., (2013). Advances in plant gene-targeted Mondego, J.M.C., Mieczkowski, P. and and functional markers: a review. Plant Pereiraa, G.A.G. (2014). High-resolution Methods 9: 6 transcript profiling of the atypical biotrophic interaction between Roslina, M.S., Rosmin, K., Lea, J., Navies, M., Theobroma cacao and the fungal Nor Aisyah., Sumayyah, A.K., Zafirah, pathogen Moniliophthora perniciosa. Z., Fahmie, W. and Croft, L. (2014). Identification of Genetic Markers for The Plant Cell, 26: 4245–4269. Pod Borer Resistance by Sequencing 14 Valenzuela, I., Purung, H., Roush, R.T. and Cocoa Genomes. Malaysian Cocoa Hamilton, A.J. (2014). Practical yield Journal 8: 40-48 loss models for infestation of cocoa with cocoa pod borer moth, Conopomorpha Ryan, P.R., Tyerman, S.D., Sasaki, T., Furuichi, cramerella (Snellen), Crop Protection T., Yamamoto, Y., Zhang, W.H and 66: 19-28. Delhaize, E. (2011). The identification of aluminium-resistance genes provides Wang, Z., Gerstein, M. and Snyder, M. (2009). opportunities for enhancing crop RNA-Seq: a revolutionary tool for production on acid soils. Journal of transcriptomics. Nat Rev Genet. 2009 Experimental Botany 62(1): 9–20 January; 10(1): 57–63

Salgotra, R.K., Gupta, B.B. and Stewart, C.N. Xu, Q., Liu, C., Biswas, M.K, Pan, Z. and Deng, (2014). From genomics to functional X., (2014). Recent advances in fruit crop markers in the era of next-generation genomics. Front. Agric. Sci. Eng. DOI sequencing. Biotechnol Lett (36): 417– 10.15302/J-FASE-2014002 426.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

DNA FINGERPRINTING OF THE MALAYSIAN COCOA GERMPLASM COLLECTION USING MICROSATELLITE MARKERS.

Lea, J.1 and Haya, R.2 1Malaysian Cocoa Board, Cocoa Biotechnology Research Centre, Kota Kinabalu Industrial Park, 88460 Kota Kinabalu, Sabah, Malaysia 2 Malaysian Cocoa Board, 5th – 7th Floor, Wisma SEDCO, Locked Bag 211, 88999 Kota Kinabalu, Sabah.

Malaysian Cocoa Journal 9(1): 50-58 (2016) ABSTRACT - Cocoa (Theobroma cacao L.), the main source of cocoa-based products and chocolate, must be conserved live in situ or ex situ as its seeds do not remain viable for more than a couple of weeks once the pod has been harvested. The Malaysian Cocoa Germplasm Collection (MCGC) is one of the ex situ collection of cocoa clones, the fourth largest in the world and regularly importing new clones from other cocoa genebank as well as exporting clones to other countries. The MCGC was established in 1992 and currently holding more than two thousands imported and local cocoa clones. As in many germplasm collection centres, mislabelling is a critical known problem and correction of the problem is crucial to improve the information reliability and efficient management of germplasm. Using microsatellite DNA markers, the germplasm collection was assessed in its amount of synonymies and homonymies and also its population membership based on the Bayesian clustering analysis. Comparison of homonymous plants across the collection revealed a significant misidentification rate estimated to be 37.5% and 10.77% synonymous errors. The microsatellite DNA markers amplified a total of 270 alleles with mean allelic richness of 18.2 alleles per locus and average polymorphism information content (PIC) value of 0.9948. The observed heterozygosity (Hobs) is 0.6855, indicate a high allelic diversity in this collection. Based on population structure analysis and a Bayesian assignment test showed that the germplasm collection can be classified into ten distinct clusters indicates that it is well represented and there is not much genetic gaps problem in the collection even compared with the USDA and CATIE cocoa germplasm collection.

Key words: Cocoa, Germplasm, Microsatellite markers, DNA Fingerprinting, Mislabeled clones

INTRODUCTION around the world and are located at four separate areas in Malaysia; Lower Perak, Perak, Jengka, Theobroma cacao L. or commonly known as Pahang, Madai Baturong and Tawau in Sabah. cocoa is an important tropical crop native to Maintained in field collection of living plants South America (Dias et al,. 2003) that provides due to difficulty working with its seeds, it raw ingredients for the chocolate confectionary occupies a large and extensive areas, industries. It is a global commodity that has an burdensome, difficult to manage appropriately annual production exceeded 4 million tons in and demanding a lot of works involving various 2013 of which 75% was produced in West cultivation practices. Africa (FAOSTAT, http://faostat3.fao.org/home/ index.html). It contributes significantly to the Cocoa germplasm collection have been economy of many regions of the world with its shown to contain a variety of mislabeled beans as the source of a multibillion dollar individuals and mislabeling is estimated at 15 to industries for the production of cocoa butter, 44% in global cocoa collections (Motilal and chocolates as well as cosmetics and antioxidants Butler, 2003; Motilal, 2004; Sounigo et al. 2006; source. Boza et al. 2012; Irish et al. 2010). As the case with many other cocoa germplasm collections, Cocoa is an outcrossing species (Wood MCGC invariably contains duplicate accessions, and Lass, 2001) and germplasm is conserved as synonymy problems, homonymous and clonally propagated in field collection. The identification mistakes, which make the transfer germplasm collections of Malaysian Cocoa of results and recommendations among different Germplasm Collection (MCGC) possesses breeding programs quite difficult and increase around 2263 accessions of T. cacao (CacaoNet the level of time and resources needed for Surveys 2008-2012) from several countries evaluating genetic potential of the accessions.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

The use of germplasm for crop improvement is 1 - 2 days) and kept at -20oC upon arrival until also often interfered by the unavailability of DNA extractions were conducted. information on the clones’ origin, genetic identity and genetic diversity. DNA Isolation, SSR Markers and Polymerase As part of the cocoa germplasm Chain Reaction (PCR) Amplification collection centres in the world, it is important to Genomic DNA was extracted from 100mg of know the passport of each cocoa clone in the leaf sample using DNeasy Plant Mini Kit collection in order to ensure conservation and (Qiagen, Hilden, DE) according to selection of clones for interplanting and breeding manufacturer’s instruction. The study used 15 are correct and would produce expected SSR loci (Table 1) identified as the international desirable agronomic traits especially. These standard set for cocoa germplasm issues have brought up the awareness on the characterization as reported by Saunders et al. need to genotype cocoa germplasm all over the (2004). Primers were synthesized by First Base world to address the issues and in ensuring (Malaysia) and forward primers were 5’ labelled efficient utilization of cocoa germplasm held in using HEX or 6-FAM dyes. PCR reaction many germplasm collections around the world mixtures consisted of 1µl containing 10-200ng by using a standard set of molecular markers of genomic DNA template, 1.6µl of a stock (Lanaud et al., 1999; Saunders et al., 2004). solution containing both forward and reverse primers (10µM each), 2µl of 2.5mM dNTP mix, The study was implemented in order to 1x PCR buffer, 1.6µL of 2.5mM MgCl2, 0.5µL identify and access the mislabelling errors in the of 0.5unit/µl Taq DNA Polymerase and sterile MCGC. The correct identification of the genetic distilled water for a total volume of 20µL. resources used in the cocoa plant breeding programme is a critical in ensuring the success PCR amplification were carried out in a of the programme in producing desirable GeneAmp PCR System 9700 (PE Applied planting materials. Biosystems) and MyIQ Single Color Real-Time PCR Detection System iCycler (BioRad) with At the end of the study, the duplicates following profile: 94oC for 4 minutes, followed and mislabelling identification using the by 35 cycles of 94oC 30 seconds, 46oC to 60oC pairwise comparison based on the fifteen (depending upon the annealing temperature of international cocoa microsatellite standard the primer pair) for 1 minute; followed by a hold (Saunders et al., 2004) were implemented. The at 72oC for 5 minutes. rate of three type of mislabeling errors were identified and presented. The amplified microsatellite loci were separated by capillary electrophoresis and analysed on an ABI Prism 3730 Genetic MATERIALS AND METHODS Analyzer (fragment analysis service outsourced to First Base Sdn Bhd, Malaysia). A dataset of Germplasm Collection multi-locus genotype in fsa file format was A total of 1504 cocoa leaves were collected from generated as a result. the MCGC in Tawau and Madai Baturong Kunak in Sabah and Hilir Perak and Jengka in GeneMapper® (Applied Biosystems) Peninsular Malaysia. The germplasm samples were used to score the multi-locus dataset in fsa included 500 clones, each represented by 2 files format Experimental data obtained through replicate trees. The samples used for DNA the fragment analysis were then scored and fingerprinting profiles included fully expanded arranged in the Microsoft Excel sheet as a data cocoa leaves from various ages collected from source in the next data analysis. CONVERT individual cocoa clones leaves. Fully-expanded v1.31 (Glaubitz, 2004) was used to convert cocoa leaves were harvested from each tree and dataset into GENEPOP format before the data shipped to the MCB Centre for Cocoa were used in Microchecker software v2.2.3 to Biotechnology Researches in Kota Kinabalu, check on genotyping errors due to non-amplified Sabah for analysis. Leaf materials were kept at alleles (null alleles) dropouts and stutter peaks. surrounding temperature during shipment (took

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 1. Description for the 15 SSR loci genotyped in the cocoa germplasm collection.

Locus Name Linkage group Tm (oC) Estimated size range (bp) mTcCIR1 8 56 120-140 mTcCIR6 6 52 220-260 mTcCIR7 7 56 150-170 mTcCIR8 9 52 300-320 mTcCIR11 2 50 280-320 mTcCIR12 4 56 170-260 mTcCIR15 1 56 230-270 mTcCIR18 4 56 330-350 mTcCIR22 1 50 280-300 mTcCIR24 9 60 180-210 mTcCIR26 8 46 280-310 mTcCIR33 4 52 270-350 mTcCIR37 10 50 140-190 mTcCIR40 3 52 270-300 mTcCIR60 2 52 190-220

Identification of Duplicates and Mislabeling level of polymorphism associated with each of For the purpose of this study, two types of the loci used in this study. High PIC value loci mislabelling in the collection were identified. discriminate better than less informative loci. The first mislabelling error type was homonymous or intraplant error, where the trees Genetic Relationships of Accessions that had same name in the collection but Population structure analysis was carried out difference multilocus profiles. The second was using the model-based Bayesian cluster analysis synonymous mislabelling or duplicate error, software STRUCTURE v2.1 (Pritchard et al., whereby the clone had different names but 2000). All the dataset were used to infer the shared the same microsatellite profiles. genetic structure of the germplasm collection. The Structure software uses allele frequencies Duplicates were identified by using from what is assumed to be potentially unlinked pair-wise comparisons among all the 1504 loci. It uses these figures to identify which set of individuals based on their multilocus SSR genes belonged to which population. Based on profile. The program GenAlEX 6.5 (Peakall and the allele frequencies, the population was set to Smouse, 2006, 2012) was used for genotype K, which would be determined by the software. matching. Accessions with different names that The programme would theoretically indicate the were fully matched at all 15 loci were declared total of populations that data represents. as duplicates or synonymously mislabelled Thereafter, the individuals were assigned to their accessions. significant K population. The population was analysed using the twenty reference samples Analysis of Genetic Diversity from each of the ten previously described cocoa Allele frequencies were estimated for each of the genetic group by Motamayor et al. (2008). locus of the entire population using PopGene v1.31. Statistical analyses were carried on all the alleles on alleles per locus, as well as the RESULTS observed (Hobs) and expected (Hexp) heterozygosity values. Additionally, the Identification of Duplicates and Mislabeling polymorphic information content (PIC) which is Fingerprint profiles of all the 1504 trees were 1 -  , where is equal to the frequency of the generated with all the 15 microsatellite primers. ith allele at the locus, was then calculated Microchecker has showed that the population is (Powell et al. 1996). PIC values indicate the possibly in Hardy Weinberg equilibrium in all

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

loci with none showing null allele. The 6.0 programme among the samples revealed a estimated null allele frequency for each locus is significant misidentification rate estimated to be compared to the null allele frequencies obtained 37.5% homonymous and 10.77% synonymous using method Chakraborty (Chakraborty 1992) errors. A total of 62.5% of the collection studied and Brookfield (Brookfield 1996). matched with the established reference genotypes in the ICGD (Figure 1). Genotype data was used to identify homonymous and synonymous errors in the collection. Pairwise comparison using GenAlEX

Figure 1. Pairwise comparison using GenAlEX 6.0 programme among the samples revealed a significant misidentification rate estimated to be 37.5% homonymous and 10.77% synonymous errors

Analysis of Genetic Diversity Genetic Relationships of Accessions Statistical analyses carried using PopGene v1.31 When the germplasm population was evaluated showed high genetic diversity as the numbers of in their relatedness to the ten representative alleles were found tin each primer is between 15 cocoa genetic groups as defined by Motamayor – 23 alleles. The microsatellite DNA markers et al (2008). An admixture model with 200,000 amplified a total of 270 alleles with mean allelic iterations after a burn-in period of 100,000 was richness of 18.2 alleles per locus and average used. The number of clusters (K-value) was set from 1 to 10. The run with the highest Ln Pr polymorphism information content (PIC) value (X|K) value of 10 was chosen and presented as of 0.9948 (Table 2). It shows that the H is obs bar plots per genotype or group. slightly lower than the Hexp indicating probable inbreeding system of mating (redundancy of The result of the Bayesian clustering alleles) in the sample. The observed analysis largely agreed with the distance-based heterozygosity (Hobs) of 0.6855 also indicate a cluster analysis. The 1504 individuals could be high allelic diversity in this collection. grouped into ten most probable clusters representing the 10 main groups as mentioned by Motamayor et al. (2008) (Figure 2).

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 2. Description and summary statistics for 15 loci genotyped in the MCB germplasm collection

Linkage Size range No of Locus name H H PIC Value group (bp) alleles obs exp mTcCIR26 10 282-307 15 0.7311 0.7344 0.999983541 mTcCIR37 4 133-185 18 0.7133 0.838 0.999999853 mTcCIR6 4 222-247 16 0.674 0.7699 0.999997742 mTcCIR8 1 388-304 17 0.5869 0.691 0.999477681 mTcCIR11 8 288-317 20 0.6974 0.7626 0.999956055 mTcCIR24 2 185-203 17 0.5073 0.4685 0.926325058 mTcCIR40 2 259-284 15 0.5957 0.7684 0.999962893 mTcCIR7 4 155-296 16 0.6613 0.7284 0.997449701 mTcCIR12 6 188-251 23 0.7634 0.769 0.999999905 mTcCIR18 3 331-355 17 0.7622 0.7703 0.999996335 mTcCIR22 9 279-290 23 0.6314 0.6046 0.99952181 mTcCIR1 1 127-144 18 0.5934 0.5487 0.999781305 mTcCIR15 9 232-256 22 0.8418 0.8389 0.99999998 mTcCIR33 8 264-346 15 0.7886 0.8322 0.999999987 mTcCIR60 2 187-223 18 0.7344 0.7944 0.99999998 Mean 18.2 0.6855 0.728 0.994830124

Figure 2. Inferred clusters in the MCGC using STRUCTURE software. The most probable clusters were obtained at K=10. Each vertical line represent one individual genotype. Individuals with multiple colours have admixed genotypes from multiple clusters.

DISCUSSIONS markers such as AFLP (Amplified Fragment Length Polymorphism) have sufficient Mislabeled plants have been identified as a discriminatory power to distinguish cocoa serious problem in germplasm collection (Hurka accessions but most often failed to reach clear et al., 2004). Significant efforts have been conclusion that two genotypes are identical identified to solve the problem in some (Sounigo et al., 2001). international cocoa collections (Motilal et al., 2013; Zhang et al., 2009); however, the For years, the MCB has been importing mislabeling problem in most of the various and collecting clones from germplasm national collections has not been addressed collections all over the world. Yet, to our systematically. Until recently, molecular knowledge, this is the first time the MCGC was

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

characterized using microsatellite markers. clones have been deposited in the International Mislabeled clones have been a concern for cocoa Cocoa Germplasm Database at the University of germplasm management and cocoa breeding Reading, UK (http://www.icgd.rdg.ac.uk/index. programmes both in MCB and international php) cocoa germplasm collection. Genotypes can be difficult to distinguish morphologically and The comparison of multilocus identification relies mostly on plant labels and microsatellite profiles had led to the discovery of field maps. Over the years of cocoa germplasm clones that were intraplant errors. They were maintenance, a significant numbers of the clones defined as intraplant errors because they did not may have been mislabeled or labeled incorrectly matched at all the 15 loci but labeled with the due to uncertain phenotypic assessment. same names. Some clones were identified as Saunders et al. (2004) has demonstrated that a synonymously mislabeled as they shared the set of fifteen standardized microsatellite primers same exact alleles across all the 15 microsatellite can characterize all the T. cacao germplasm loci but were labeled with different names collection and was effective for the assessment (Figure 1). of genetic identity of cocoa germplasm. The difference in total number of alleles In the recent years, microsatellite found between the MCGC, USDA-ARS and the markers have been widely used in cocoa CATIE collection also indicated that the genetic genotyping and individual identification. This diversity of cocoa in MCGC is fully represented enable a systematic assessment of genetic with the number of major alleles comparable identity in national and international cocoa with the other two international germplasm germplasm (Zhang et al., 2009; Motilal et al., collections (Figure 3). 2010). Reference microsatellite profiles of cocoa

Figure 3. Comparison of genetic diversity between the USDA-ARS Mayaguez collection, the CATIE Costa Rica and the MCB collection.

The allelic diversity present in the in the CATIE cocoa collection (14.2 alleles per MCGC (18.2 alleles per locus; 270 total alleles) locus; 231 total alleles) which is considered to can be considered very high compared to that contain a larger number of accessions from much identified by Irish et al (2010) in the USDA-ARS broader geographic area (Zhang et al., 2009). cocoa collection (8.80 alleles per locus; 132 total alleles) and even greater than the allelic diversity

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

High levels of heterozygosity plant material development programmes, (Hexp=0.728) in the MCGC was detected (Table breeders should not rely merely on the genetic 2). The high gene diversity indicates substantial diversity data but also on the economic and levels of admixture present in the germplasm agronomic traits. These combined data should be collection. The Hexp value also higher than that considered as indicators for the cocoa germplasm described for the USDA-ARS cocoa collection management and development of new cocoa (Hexp=0.66; Irish et al 2010) and CATIE planting materials (Irish et al., 2010) (Hexp=0.51; Zhang et al., 2009)

The Structure software (Pritchard et al., ACKNOWLEDGEMENTS 2000) has been used widely for Bayesian clustering analysis which assigns groups of The authors thank the Ministry of Science, individuals based on coefficients of population Technology and Innovation (MOSTI) for membership and determine the degree of supporting this project (ScienceFund grant no. admixture of allelic contributions within the 02-03-13-SF0081) and the Director General of population (Kalinowski, 2010). When the MCGC the Malaysian Cocoa Board , Datin Norhaini analyzed together with the 10 reference genetic Udin, for permission to publish these results. We groups previously reported by Motamayor et al. also thank Ms. Sairan Asim, Ms. Rafiah Abdul (2008), all the 10 genetic groups were represented Karim, Ms. Nuraziawati Mohd Yazik and Hj in the MCGC (Figure 2). Zailaini Md. Jamil and Ms Sarina Abu Samah for their excellent experience in cocoa breeding, Based on this initial finding, we cocoa germplasm, technical and samples recognize the urgent need to genotype the all the collection assistance. MCGC to ensure that any misidentification/mislabeling can be identified to prevent hindrances in conserving and utilizing the REFERENCES germplasm efficiently and reduce waste of land/space due to genetic redundancy in the Boza, E.J., Irish, B.M., Meerow, A.W., germplasm collection. These mislabeled trees Rodriguez, O.A., Ventura-Lopez, M., would significantly affect breeding program by Gomez, J.A., Moore, J.M., Zhang, D., reducing possible genetic benefits in the progeny Motamayor, J.C. and Schnell II, R.J. when misidentified parents were used in breeding programmes. (2012). Genetic diversity, population structure, conservation and utilization of With the availability of microsatellite Theobroma cacao L. genetic resources in fingerprinting profiles for each cocoa accessions, the Dominican Republic. Tree Genetics information of each accession’s passport data and Genomes. (origin, history, genetic identity etc.) and genetic diversity between individuals and the whole Brookfield, J. F. Y. (1996). A simple new method collection can be identified and used in the for estimating null allele frequency from potential exploitation of germplasm in cocoa heterozygote deficiency. Molecular improvement such as in choosing the correct Ecology 5(3): 453–455 clones for breeding and plant materials development. Chakraborty, R. (1992). Sample size However, it needs to be pointed out that requirements for addressing the population the estimation of genetic diversity and simulation genetic issues of forensic use of DNA of genetic redundancy were based on typing. Hum Biology 64: 141-159. microsatellite markers-defined parameters alone, without taking into consideration economic and Dias, L. A. D. S., Marita, J., Cruz, C. D., Barros, agronomic traits. There are many clones that may E. G. and Salomão, T. M. F. (2003). not have high allelic richness, but they may Genetic distance and its association with possess important and valuable agronomic and heterosis in cacao. Brazilian Archives of economic traits. Thus, in the breeding and new Biology and Technology 46(3): 507-514.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Glaubitz J.C. (2004). Convert: A user-friendly relationship with Trinitario and other program to reformat diploid genotypic data clones held in The International Cocoa for commonly used population genetic Genebank, Trinidad Plant. Genet Resource software packages. Molecular Ecology 8: 106-110. Notes 4: 309-310. Motilal, L. A., Zhang, D., Mischke, S., Hurka H., Neuffer B. and Friessen, N. (2004). Meinhardt, L. W., Umaharan, P. (2013). Plant genetic resources in botanical Microsatellite-aided detection of genetic gardens. In: Forkmann, G. and Michaelis, redundancy improves management of the S. (Eds), Proceedings of the 21st International Cocoa Genebank Trinidad. International Symposium on Breeding Tree Genet. Genomes 9: 1395-1411. Ornamentals. Part Ii. Acta Hortic 651: 35- 44. Motilal L. A. and Butler, D. (2003). Verification of identities in global cacao germplasm Irish, B.M., Goenaga Portela, R.J., Zhang, D., collections. Genet. Resources Crop Ev. 50: Schnell II, R.J., Brown, J.S. and Motamayor, J. (2010). Microsatellite 799-807. fingerprinting of the USDA-ARS Tropical Agriculture Research Station Cacao Peakall, R. and Smouse P. E. (2006). Genalex 6: (Theobroma Cacao L.) germplasm genetic analysis in excel. Population collection. Crop Science. 50: 656-667. genetic software for teaching and research. Molecular Ecology Notes 6: 288-295. Kalinowski, S. T. (2010). The computer program structure does not reliably identify the Peakall, R. and Smouse P. E. (2012). Genalex main genetic clusters within species: 6.5: Genetic analysis in excel. Population Simulations and implications for human genetic software for teaching and research population structure. Heredity 2010: 1-8. - An update. Bioinformatics 28: 2537- 2539. Lanaud, C., Risterucci, A. M., Pieretti, I., Falque, M., Bouet, M. and Lagoda P. J. L. (1999). Powell, W., Morgante, M., Andre, C., Hanafey, Isolation and characterization of M., Vogel, J., Tingey, S. and Rafalski, A. microsatellites in Theobroma Cacao L. (1996). The comparison of RFLP, RAPD, Molecular Ecology 8: 2141-2143. AFLP and SSR (microsatellite) markers for germplasm analysis. Mol Breeding 2: Motamayor, J. C., Lachenaud, P., Da Silva e 225-238 Mota, J. W., Loor, R., Kuhn, D. N., Brown, J. S. and Schnell, R. J. (2008). Pritchard, J. K., Stephens M. and Donelly, P. Geographic and genetic population (2000). Inference of population structure differentiation of the Amazonian chocolate from multilocus genotype data. Genetics tree (Theobroma Cacao L). PLoS One 155: 945–959. 3(10):E3311, DOI: 10.1371/Journal.Pone. 0003311. Saunders, J. A., Mischke, S., Leamy, E. A. and Hemeida, A.A. (2004). Selection of Motilal, L. A. (2004). The potential of cacao international molecular standards for DNA microsatellites amplification across diverse fingerprinting of Theobroma Cacao. plant taxa. In: Thangadurai D, Pullaiah T. Theoretical and Applied Genetics 110(1): Balatti Pa, (Eds.) Genetic Resources and 41-47. Biotechnology, Vol. 2. Regency Publications, New Delhi, India pp: 24-49. Sounigo, O., Christopher, Y., Bekele, F., Motilal, L. A., Zhang, D., Umaharan, P., Mooleedhar, V. and Hosein, F. (2001). The Mischke, S., Moooleedhar, V. and detection of mislabeled trees in the Meinhardt, L. W. (2010). The relic: Criollo International Cocoa Genebank, Trinidad Cacao In Blize – Genetic diversity and (ICG, T). In: Proceedings of the

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

International Workshop on the New on Cocoa Germplasm Utilization and Technologies and Cocoa Breeding. 16th – Conservation: A Global Approach (1998- 17th October 2000, Kota Kinabalu, Sabah, 2004). CFC, Amsterdam, The Malaysia, INGENIC(International Group Netherlands/ICCO, London, UK/IPGRI, for Genetic Improvement of Cocoa). pp. Rome, Italy, pp. 82-86. 34-39. Wood G.A.R and Lass, R.A. (2001). Cocoa. 4th Sounigo, O., Risterucci, A. M., Clement, D., Edition 620 Pages, Wiley-Blackwell Fouet, O., and Lanaud, C. (2006). Identification of off-types of clones used in Zhang, D., Mischke, S., Johnson, E. S., Mora, A., the international clone trial using DNA Phillips-Mora, W. and Meinhardt, L. W. analyses. In: Eskes, Ab., Efron, Y (Eds) (2009). Molecular characterization of an Global Approaches to Cocoa Germplasm international cacao collection using Utilization and Conservation. Final microsatellite markers. Tree Genet. Report of the CFC/ICCO/IPGRI Project Genomes 5: 1-10.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

EVALUATION OF CLONAL MATERIALS IN MALAYSIAN COCOA BOARD, CRDC MADAI SABAH

Aizat, J.1, Nuraziawati, M.Y.2, Azlan, S.3, Haya, R.4 1Malaysian Cocoa Board, Cocoa Research & Development Centre Kota Samarahan, Lot 248, Block 14, Daerah Muara Tuang, Bahagian Samarahan, Locked Bag 3131, 93450, Kuching, Sarawak. 2Malaysian Cocoa Board, Cocoa Research & Development Centre Hilir Perak, Jalan Sg. Dulang, 36307 Hutan Melintang, Perak. 3Malaysian Cocoa Board, Cocoa Research & Development Centre, Batu 10, Jalan Apas, Peti Surat 60237 91012 Tawau, Sabah, Malaysia. 4Malaysian Cocoa Board, Tingkat 5, 6 & 7, Wisma SEDCO, Lorong Plaza Wawasan, Off Coastal Highway, Beg Berkunci 211, 88999 Kota Kinabalu, Sabah, Malaysia. e-mail: [email protected]

Malaysian Cocoa Journal 9(1): 59-63 (2016) ABSTRACT - A clonal trial at Cocoa Research and Development Centre Madai, Sabah was established in 2008. The objective of the trial was to evaluate the potential yield of the selected cocoa accessions based on vascular streak dieback (VSD) disease tolerance and suitable to local agro-climatic condition. Twenty two cocoa accessions were evaluated against three control clones; PBC 123, MCBC 1 and KKM 22.The analyses performed indicated that PT3A44 was the best clone to have the highest potential yield mean while MHO430 appeared to have the least potential yield. Most of the accessions showed moderately resistant to VSD disease.

Key words: Cocoa, Clonal trial, Vascular streak dieback, Potential yield.

INTRODUCTION been established to evaluate new cocoa clones and selection of superior clonal varieties. The Since the discovery of vascular streak dieback objective of the trial was to evaluate the potential (VSD) disease in Papua New Guinea in the yield of the selected cocoa accessions based on 1960s, the disease has been found spreading to VSD disease tolerance and suitable to local agro- cocoa growing areas in Southeast Asia, parts of climatic condition. Melanesia, and cause serious losses of mature trees and seedlings (Guest and Keane, 2006). The cause of this disease is fungus MATERIALS AND METHODS basidiomycete, Oncobasidium theobromae. The occurrence of VSD is most common in wetter Genetic Material regions where annual rainfall exceeds 2,500 mm Genetic materials of this clonal trial consist of especially in West Malaysia. The use of resistant selected genotypes from hybrid population planting materials has been identified as one of established at Cocoa Research and Development the effective solutions. This technology can be Centre (CRDC) Madai and CRDC Hilir Perak. A easily transferred to cocoa farmers without total of 22 clones were evaluated against three expensive inputs. commercial clones; PBC 123, MCBC 1 and KKM 22 (Table 1). The main objective of cocoa breeding program in Malaysia is to develop new cocoa Trial Design planting materials that are high yield and tolerant The 25 selected clones were planted in a field to major pests and diseases. In order to achieve experiment at Block 36A, CRDC Madai in the objective, Malaysia Cocoa Board (MCB) has September 2006 using a randomized complete apply strategies such as the introduction and block design with three replications. The site has conservation of cocoa germplasm, breeding and an average monthly temperature of 25oC, with a selection for high yield, pest and disease minimum of 20oC and a maximum of 35oC, resistance, bean quality and good flavour; and monthly precipitation of 150.1 mm and average inter-specific hybridization. Field clonal trial has relative humidity of 75%. Each plot consisted of

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

twenty trees planted in double rows with 3 m of year 12N-12P-17K-TE fertilizer; canopy pruning spacing between both trees and plots. Forage for height; regular pruning for disease control, tree legume (Gliricidia sepium) were planted as black pod control using metalaxyl fungicides permanent shade at a spacing of 6 m x 6 m. during wet seasons, and weed control by a Maintenance treatments were applied uniformly combination of herbicides and bush cutter. to each tree as follows: 4 x 270-300 g/tee per

Table 1. The list of selected clones

No Trial Code Selection Code/ Clone No Trial Code Selection Code/ Clone 1. A C744 14. N C1038 2. B PT3A44 15. O MHP 435 3. C MHP 428 16. P PT4D6 4. D PBC 123 17. Q MHP 438 5. E C119 18. R C712 6. F MCB C1 19. S MHP 436 7. G N/4/21 20. T PT3D41 8. H MHP 430 21. U MHP 441 9. I PT4K66 22. V C276 10. J PT4K66 23. W 13B/16/1 11. K C1132 24. X DESA CT100 12. L MHP 431 25. Y MHP 442 13. M PT4N99

Phenotypic data was recorded on a VSD resistance among the selected clones single tree basis and the number of mature pods (Table 2). The most resistant among them is was taken monthly over a 5-year period. The clone N (C1038) which had the lowest score total number of pods harvested per year was (2.33) and comparable to established clone, PBC evaluated as one of the main yield components. 123 which has mean score of 2.35. Contrary, The dry cocoa beans were prepared according to there are two clones that are susceptible in the normal procedure; fermentation of the raw beans group, clone U (MHP 441) and W (13B/16/1) for 4-5 days and sun-drying 5-7 days. The with score of 2.75. characteristics measures included average bean weight, pod weight, husk content, and number of Pod production and pod value are the beans per pod for each hybrid. Pod value was two major components of cacao yield. Results in expressed as the number of pods required to Table 3 showed that clone B (C744) had high produce 1 kg of dried beans. Pod value = 1000/ potential of yield with the highest mean number (bean number x mean bean weight). Potential of pods per tree (19.08) exceeding the control yield (Kg.ha-1.year-1) for each hybrid was clone D (PBC 123) with score 18.48. The yield estimated from pod production divided by pod performance can be correlated to mechanism of value multiplied by 625 trees/ ha and a resistance to VSD disease. Clones with have correction factor 0.83. The potential yield was score of mean pod number per tree more than 10 recorded at maturity on 6-year old trees. also have VSD score lower than 2.50. Most of the clones with mean score of VSD more than RESULTS AND DISCUSSIONS 2.50 have lower mean pod number per tree which is below than 10. The results of VSD evaluation for two years indicate that there was varying performance of

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 2. Mean score of VSD symptom damage of the tested clones for 2015-2016.

Year No Trial Code Selection Code/ Clones Mean 2015 2016 1. A C744 2.61 2.53 2.57 2. B PT3A44 2.67 2.28 2.48 3. C MHP 428 2.57 2.51 2.54 4. D PBC 123 2.57 2.13 2.35 5. E C119 2.56 2.34 2.45 6. F MCB C1 2.47 2.31 2.39 7. G N/4/21 2.65 2.19 2.42 8. H MHP 430 2.74 2.25 2.50 9. I PT4K66 2.83 2.53 2.68 10. J PT4K66 2.56 2.20 2.38 11. K C1132 2.51 2.48 2.50 12. L MHP 431 2.75 2.69 2.72 13. M PT4N99 2.65 2.40 2.53 14. N C1038 2.49 2.16 2.33 15. O MHP 435 2.77 2.46 2.62 16. P PT4D6 2.56 2.40 2.48 17. Q MHP 438 2.60 2.47 2.54 18. R C712 2.59 2.49 2.54 19. S MHP 436 2.47 2.83 2.65 20. T PT3D41 2.65 2.23 2.44 21. U MHP 441 2.83 2.67 2.75 22. V C276 2.36 2.50 2.43 23. W 13B/16/1 2.71 2.78 2.75 24. X DESA CT100 2.57 2.31 2.44 25. Y MHP 442 2.60 2.43 2.52

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 3. Mean of pod number per tree from 2012 until Mac 2016.

Pod Yield Per Tree No Code 2012 2013 2014 2015 Mac 2016 Mean 1. B 8.18 15.26 35.84 31.48 4.62 19.08 2. D 9.27 20.32 30.25 28.59 3.97 18.48 3. G 6.3 22.67 27.13 25.00 3.75 16.97 4. L 5.75 12 8.92 0.33 - 6.75 5. V 6.63 12.75 16.57 11.76 3.68 10.28 6. E 6.35 14.64 21.2 16.66 4.93 12.76 7. Y 6.65 5.07 10.14 1.33 4.67 5.57 8. N 7.35 12.34 14.9 20.67 3.33 11.72 9. K 7.15 8.95 18.31 12.93 3.88 10.24 10. F 7.51 14.62 19.66 19.17 2.59 12.71 11. J 5.26 13.46 15.77 15.66 4.41 10.91 12. P 7.82 14.05 13.58 15.98 3.55 11.00 13. R 5.11 8.97 9.59 7.37 2.36 6.68 14. A 5.13 8.61 12.74 9.44 3.23 7.83 15. Q 5.88 7.78 12.19 11.15 3.47 8.09 16. O 1.33 9.00 5.5 3.70 1.00 4.11 17. X 6.24 7.46 12.54 16.38 4.38 9.40 18. W 3.44 6.64 19.48 9.20 2.60 8.27 19. C 5.89 7.54 6.92 2.38 1.00 4.75 20. M NA 4.67 7.79 8.52 4.57 6.39 21. I 7 4.84 1.33 3.17 3.00 3.87 22. T 3.25 4.39 6.5 20.90 3.96 7.80 23. S 3.17 4.85 4.6 2.85 4.84 4.06 24. U NA 2 1.84 2.14 1.50 1.87 25. H NA NA NA 1.90 2.80 2.35 Mean 5.94 10.12 13.89 11.95 3.42 8.88

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

The evaluation on pod and bean quality SMC 1 according to Malaysia Standard was performed to several of the tested clones 230:2007). The pod value for clone Q and A (Table 4). The selected clone Q, D and A were up to 22.1 and 23.6, which are superior to produce high quality dried cocoa beans with that of clone D as the control. ADBW more than 1.0 g respectively (Grade

Table 4. Yield components of several clones tested.

No Trial Code Selection Code/ Clone ADBW BNP BCR PV SD SC

1. Q MHP 438 1.12 47 98.3 22.1 1.01 15.68

2. E C119 0.71 51 94.8 29.8 0.13 18.63

3. D PBC 123 1.21 39 96.3 29 0.97 17.59

4. J PT4K66 0.92 32 98.8 34.6 0.16 17.95

5. A C744 1.13 44 96.2 23.6 0.96 18.26

6. N C1038 0.95 43 91.6 28.5 0.24 18.58

7. F MCBC 1 0.92 36 90.9 37.4 0.15 19.08 ADBW-Average dried bean weight; BNP- Bean number per pod; Bean conversion rate; PV-Pod value; SD- Standard deviation; SC- Shell content %.

CONCLUSION extended to the Deputy Director-General (Research & Development) and Director of Most of the tested clones demonstrate moderate Cocoa Upstream Technology Division for his resistance to vascular streak dieback disease and guidance. Acknowledgement is also conveyed to the most resistant is clone N (C1038). Among CRDC Madai center manager for allowing the the selected VSD resistant clones, clone B trial to be done. Finally, special thanks to all (PT3A44) showed potency on in terms of yield cocoa breeding members for their assistance in performance. the field work.

ACKNOWLEDGEMENTS REREFENCE

This research was co-funded by the Ministry of Guest, D. & Keane, P. 2007. Vascular-Streak Finance, 11th Malaysia Plan, Projek Bahan Dieback: A New Encounter Disease of Tanaman Koko Unggul and Malaysia Cocoa Cacao in Papua New Guinea and Southeast Board. The author acknowledges the Director Asia Caused by the Obligat Basidiomycete General of Malaysian Cocoa Board for the Oncobasidium theobromae. The American permission to publish this paper. Appreciation is Phytopathological Society. 97-12-1654.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

ANALYSIS OF COCOA YIELD COMPONENTS IN RELATION TO ITS YIELD

Tan, C.L.1 and Ling, A.S.C.2 1Malaysian Cocoa Board, Cocoa Biotechnology Research Centre, Kota Kinabalu Industrial Park, 88460 Kota Kinabalu, Sabah, Malaysia 2Malaysian Cocoa Board, Tingkat 5, 6 & 7, Wisma SEDCO, Lorong Plaza Wawasan, Off Coastal Highway, Beg Berkunci 211, 88999 Kota Kinabalu, Sabah, Malaysia.

Malaysian Cocoa Journal 9(1): 64-70 (2016) ABSTRACT - Yield is the single most important agronomic character in many crop plants. In cocoa, dry bean yield per hectare is influenced by dry bean yield per tree. Dry bean yield per tree (DBY), in turn, is determined by pod yield per tree (PYT), pod value (PV), average bean weight (ADBW) and bean number per pod (BNP). The aim of this study is to elucidate how much each of these yield components contributed to the total yield in cocoa. Hence, this will assist in better design of selection to enhance cocoa yield. The yield components values were based on 42 clones obtained from Malaysian Cocoa Board (MCB). Data were analysed using correlation, regression and principal components analysis. From the analysis, R- square of the regression analysis on dry bean yield is 96% which means the selected yield components able to explain 96% of the total variation in the dry bean yield while 4% maybe due to residual effects. The t-test showed that PYT and PV are highly significant (P<0.01) towards DBY whereas ADBW and BNP are not significant in the analysis. PYT is directly correlated with DBY while PV is inversely correlated with DBY. The overall results reflect that the importance of the PYT and PV in cocoa clone selection for breeding programs.

Key words: Average dried bean weight, Pod value, Bean number, Pod yield, Dry bean yield.

INTRODUCTION (ADBW) and Bean Number per Pod (BNP). The objective of the study is to examine the Yield is the single most important trait for many contribution of each of these components to the commercially grown crop plants. The challenge overall yield of cocoa. in breeding and genetic analysis of yield is posed by the biological complexity of this trait, since yield reflects the cumulative effects of multiple MATERIALS AND METHOD factors over time and across plant organs and field environments (Amit and Zamir, 2015). Data for the studies were obtained from clonal studies by cocoa breeders from various Yield is highly influenced by plantations in Malaysia (Appendix I). The environments. In contrast, environments have following methods were how the pod and bean much less effects on yield components. parameters obtained; Consequently, it is more feasible to improve yield by selecting these yield-related traits in a. Average Dried Bean Weight (ADBW): breeding programs because of the more accurate ADBW (g) = Weight of dry good beans measurement and repeatability across / No. of good beans. environments in comparison with yield (Gao et b. Pod Value (PV): al., 2015). Different crop varies in yield PV = [No. of pod sampled / Dry bean determinants to the actual crop yield. For example in mungbean; harvest index, biomass, weight (g)] x 1000 seed index, pods per cluster and branching c. Bean Number per Pod (BNP): behaviour were identified as yield determinants BNP = No. of wet beans / No. of pods for yield improvement (Singh et al., 2016). In sampled. cocoa, yield is influenced by various yield d. Pod Yield per Tree (PYT): components; namely Pod Yield per Tree (PYT), Total number of monthly pod yield per Pod Value (PV), Average Dry Bean Weight tree which include healthy, diseased

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

and rodent-damaged pods throughout a contribution of these yield components (PYT, calendar year. PV, ADBW and BNP) to the yield (DBY) of e. Dry Bean Yield (DBY) kg/tree/year: cocoa tree. Yield is determined by a highly DBY = Total weight of dry bean yield complex process; therefore, a better understanding of the factors that improve yield per tree in a calendar year. will benefit the selection of parents for crosses (Muhammad et al., 2015). Components of yield were statistically analysed for Pearson correlation, regression and a. Correlation analysis principal components using statistical software ® MINITAB Release 14. Graphs were plotted Correlation analysis of yield and its components using Microsoft Excel software. indicated that Pod Yield per Tree (PYT) has the largest correlation coefficient (0.945) to Dry

Bean Yield (DBY) (Table 1). It is significant at RESULTS AND DISCUSSION 1% level of probability. For PV, it is inversely

Dry Bean Yield (DBY) per tree is the dependent correlated to DBY and it is significant at 5% variable and Pod Yield per Tree (PYT), Pod level of probability. ADBW is 37.2% correlated Value (PV), Average Dry Bean Weight to DBY and it is significant at 5% level of (ADBW) and Bean Number per Pod (BNP) are probability. ADBW and BNP are inversely the independent variables in this study. The correlated to PV and that are highly significant study is to determine what percentage of (1% level of probability).

Table 1. The Pearson Correlation analysis of yield and yield components of cocoa

DBY PYT PV ADBW BNP

DBY

PYT 0.945**

PV -0.340* -0.046ns

ADBW 0.372* 0.187ns -0.630**

BNP 0.085ns -0.060ns -0.505** -0.247ns

Note: The SQRT transformation is used to normalise the data prior to correlation analysis. * and **: means that it is significant at 5%, 1% level of probability. ns: Not significant.

b. Regression analysis components able to explain 96% of the total variation in the dry bean yield while 4% maybe The regression equation for yield components in due to residual effects. The t-test showed that relation to Dry Bean Yield (DBY) is, PYT and PV are highly significant (P<0.01) towards DBY (Table 2) whereas ADBW and DBY = 2.31 + 0.0447 PYT - 0.110 PV + 0.307 BNP are not significant in the analysis. PYT is ADBW - 0.0007 BNP directly correlated with DBY while PV is inversely correlated with DBY. R-square of the regression analysis on dry bean yield is 96% which means the selected yield

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 2 Regression analysis of yield components in relation to DBY Predictor Coef SE Coef t value P value

Constant 2.312 2.159 1.07 0.291

PYT 0.045 0.002 27.39 0.000

PV -0.110 0.033 -3.30 0.002

ADBW 0.307 0.661 0.46 0.645

BNP -0.001 0.019 -0.03 0.973

Regression Analysis: DBY versus PYT DBY = 0.069 + 0.0454 PYT

In the single factor analysis of PYT in relation to Regression analysis was carried out for PYT in DBY, the regression equation is, relation to DBY. T-test was found to be highly significant (P<0.01) towards DBY (Table 3).

Table 3 Regression analysis of PYT in relation to DBY Predictor Coef SE Coef T P

Constant 0.0689 0.1732 0.40 0.693

PYT 0.045360 0.002838 15.98 0.0000

R-square of the regression analysis is 86.5%. A plot of PYT against DBY and the best-fit line is shown in Figure 1.

Figure 1. Plot of PYT against DBY

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Regression Analysis: DBY versus PV Regression analysis was carried out for PV in In the single factor analysis of PV in relation to relation to DBY. T-test was found to be DBY, the regression equation is, significant (P<0.05) towards DBY (Table 4). PV is also inversely correlated to DBY. DBY = 5.48 - 0.128 PV

Table 4 Regression analysis of PV in relation to DBY

Predictor Coef SE Coef T P

Constant 5.478 1.328 4.13 0.000

PV -0.12795 0.05702 -2.24 0.030

R-square of the regression analysis is 11.2%. PV is inversely correlated to DBY. A plot of PV against DBY with the best-fit line is shown in Figure 2.

Figure 2. Plot of PV against DBY

c. Principal components analysis values greater than 1 should be retained. These three principal components accounted for 98.4% Principal Component Analysis: DBY, PYT, PV, variability in the cocoa yield (Table 6). The first ADBW, BNP principal component, which accounted for about 45.60% of the variation (Table 6), was strongly Eigen analysis of the Correlation Matrix were associated with DBY and PYT. In this divided into five Principal Components (Table component (PC1), PYT contributed 50.4% of the 5). These were arranged in the order of variability in yield (Table 7). The sign of the decreasing Eigen value. The cumulative loading indicates the direction of the relationship variance of these five principal components have between the component and the variable. The a total value of 1. The first three components second principal component (PC2), which has eigen value of more than 1. According to the accounted for about 29.60% of the total variation Kaiser (1960) criterion, only PCs with eigen was identified as secondary component. In this

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

component, PV is the major component that named minor component as it only consisted of accounted 60.8% of the variability in yield. The ADBW where the component accounted for third principal component (PC3) (Table 7) was 23.20% of the variation.

Table 5. Eigen analysis of the Correlation Matrix

PC1 PC2 PC3 PC4 PC5

Eigenvalue 2.2794 1.4775 1.1610 0.0625 0.0197

Proportion 0.456 0.296 0.232 0.012 0.004

Cumulative 0.456 0.751 0.984 0.996 1.000

Variable

DBY 0.609 0.247 -0.212 0.164 0.704

PYT 0.504 0.458 -0.295 -0.101 -0.662

PV -0.423 0.608 -0.136 -0.607 0.253

ADBW 0.428 -0.139 0.677 -0.582 0.017

BNP 0.112 -0.583 -0.625 -0.506 0.037

Table 6. Principal component analysis for the yield components in cocoa

Principal component Variance (%) Cumulative variance (%)

1 45.60 45.60

2 29.60 75.20

3 23.20 98.40

Table 7. Loadings of PCA for the estimated yield components in cocoa

Variables PC1 PC2 PC3

DBY 0.609 0.247 -0.212

PYT 0.504 0.458 -0.295

PV -0.423 0.608 -0.136

ADBW 0.428 -0.139 0.677

BNP 0.112 -0.583 -0.625

Eigen values 2.279 1.478 1.161

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

CONCLUSION Gao, F., Wen, W., Liu, J., Rasheed, A., Yin, G., Xia, X., Wu, X. and He, Z. (2015). From this study, we can conclude that yield Genome-wide linkage mapping of QTL component PYT is the major contributor (R2 = for yield components, plant height and 86.5). This is followed by PV (R2 = 11.2) where yield-related physiological traits in the it is inversely correlated to DBY. The overall Chinese wheat cross results reflect that the importance of the PYT Zhou8425b/Chinese Spring. and PV in cocoa clonal selection for breeding Frontiers in Plant Science doi: programs. 10.3389/fpls.2015.01099

Kaiser, H.F. (1960). The application of ACKNOWLEDGEMENT electronic computers to factor analysis. Educ. Psychol. Meas. 20: 141-151. We wish to thank the Director-General of Malaysian Cocoa Board and the Director of Muhammad, R.W., Qayyum, A., Hamza, A., Biotechnology for permission to publish the Ahmad, M.Q., Naseer, N.S., Liaqat, S., above study. We also like to thank the cocoa Ahmad, B., Malik, W. and Noor, E. breeders for availing the yield and yield (2015). Analysis of genetic traits for components data for our present study. drought tolerance in maize. Genet. Mol. Res. 14 (4): 13545-13565.

REFERENCES Singh, C.M., Mishra, S. B. and Pandey, A. (2016). A study on correlation and Amit, G. and Zamir, D. (2015) Mendelizing all regression analysis in mungbean [Vigna components of a pyramid of three yield radiata (L.) Wilczek]. Legume Res. 39 QTL in tomato. Frontiers in Plant (1): 20-25. Science doi: 10.3389/fpls.2015.01096.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Appendix 1 PODVALUE CLONE DBY PYT ADBW BNP (PV) 1 BAL244 1.59 30.7 19.3 1.35 40 2 DESA1 1.74 45.2 26 1.12 27 3 FP10 0.93 22.4 26.7 1 41.6 4 FP7 0.71 17.8 27.7 1.05 36.7 5 FP9 0.99 21.4 24.1 1.18 34.7 6 KKM1 3.93 68.4 17.4 1.06 51 7 KKM17 2.26 42.7 18.9 1.07 49 8 KKM25 2.44 72.5 29.7 1.3 24 9 KKM26 1.76 47.9 27.2 1.15 33 10 KKM27 1.86 37.4 20.1 1.18 37 11 MCBC10 4.63 92 22.1 1.18 35 12 MCBC11 4.62 76 18.3 1.38 43 13 MCBC5 5.74 109.1 19 1.41 37 14 MCBC6 3.46 75 21.7 1.38 37 15 MCBC9 3.76 90.7 24.13 1.29 33 16 PBC112 1.53 38.3 25 1.01 38 17 PBC113 1.5 26.5 19.6 1.15 44 18 PBC130 1.91 42 22 1.28 36 19 PBC131 1.44 26.8 18.6 1.28 42 20 PBC140 1.68 42 25 1.26 33 21 PBC159 1.56 29.6 19 1.22 47 22 PBC179 1.59 33.1 20.8 1.26 38 23 PBC221 2.5 41.8 16.7 1.47 41 24 QH1176 3.06 81.4 26.6 1.05 37 25 QH1213 5.5 104 21.1 1.14 42.5 26 QH1287 3.66 90 24.6 1.05 40 27 QH186 3.33 84.2 25.3 0.94 45 28 QH22 3.51 75.1 21.4 1.15 42 29 QH240 3.1 80.9 26.1 0.97 41 30 QH441 3.88 108.6 28 0.96 39 31 QH670 3.89 77.8 22.2 1.17 40.4 32 SDS20 1.12 23 22.8 1.06 41 33 SDS21 1.32 35.2 29.6 0.94 36 34 SDS41 1.25 28.9 25.8 1.02 38 35 SDS46 1.36 31.3 25.6 1 39 36 SDS58 1.6 34.2 23.7 1.08 39 37 SDS6 1.07 25.6 26.5 1.02 37 38 SDS74 1.56 32.9 23.4 1.07 40 39 SDS84 0.81 18.1 24.8 1.12 36 40 TG137/B 4.18 78.9 21 1.39 38.8 41 TG148 3.93 53 15 1.62 36 42 TG446 4.15 89.7 24 1.43 30.8

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

PHYSIOLOGICAL RESPONSES OF CACAO SEEDLINGS SUPPLEMENTED WITH MINERALS USING FOLIAR APPLICATION

Tee, Y.K.1, Bailey, B.A.2, Ali, S.S.2, Strem, M.D.2, Elson, M.2 and Baligar, V.C.2 1Malaysian Cocoa Board, Locked Bag 211, 88999 Kota Kinabalu, Sabah, Malaysia 2US Department of Agriculture/Agricultural Research Service, Beltsville Agricultural Research Center, Beltsville, MD 20705, USA

Malaysian Cocoa Journal 9(1): 71-78 (2016) ABSTRACT - Cacao seedlings of IMC47 were grown in a glasshouse pot experiment to investigate the effects of mineral foliar applications on the physiological responses of cacao seedlings. The plants were subjected to the following eight treatments: (1) 4000 mg/L Ca(NO3)2 + 2 mg/L CuSO4 + 2 mg/L borax; (2) 2 mg/L CuSO4 + 2 mg/L borax; (3) 4000 mg/L Ca(NO3)2 + 2 mg/L borax; (4) 4000 mg/L Ca(NO3)2 + 2 mg/L CuSO4; (5) 2 mg/L borax; (6) 2 mg/L CuSO4; (7) 4000 mg/L Ca(NO3)2 and (8) control (water) as foliar application. Minerals were supplemented when seedlings were one month after sowing. The results showed that new flush leaf of the seedlings after mineral treatment had higher net photosynthesis (PN), stomatal conductance (gs), transpiration rate (E), photosynthetic efficiency photosystem II (FV/ Fm) and intercellular CO2 concentration (Ci). Among all the treatments, treated seedlings with boron and copper, alone or in combination, showed better growth performances in the new flush leaves compared to other treatments, suggesting that boron and copper play some effects on the growth by enhancing the photosynthesis mechanism, increase the rate of passage of CO2 and water vapor through the stomata of a leaf and improve the water evaporation from the leaves. Furthermore, seedlings treated with calcium, alone or combination with other minerals had the same growth or lower, with the control in net photosynthesis, stomatal conductance and transpiration rate. This might be due to the high concentration of calcium which could ameliorate the inhibitory effects of seedlings growth in cacao. These findings suggested that boron and copper supplementation could be applied to enhance the physiological aspects in cacao seedlings.

Keywords: Minerals, Foliar, Physiology, Photosynthesis, Stomatal conductance, Transpiration

INTRODUCTION Minerals such as copper and boron are physiologically important in plants as they are Nutrition management for a plant is important required for plant activities, including especially at the beginning of growth and meristematic development, chlorophyll development until the plants produce yield at the formation, photosynthesis and phenolic field through increasing photosynthetic compounds development. Thus, for better efficiency. Generally, micronutrients deficiency performance in plants, supplementation of in plants is a worldwide nutritional problem and minerals, especially the micronutrients is plants vary in their demand for micronutrients, as essential. Applications of micronutrients such as these are involved in almost of the physiological zinc and boron in tomato have been recorded in functions of plants. Micronutrients are required increasing seed yield. Studies by Fatemi (2001) by the plants in small quantities but are essential showed that foliar application of boron (B) had a for optimum plant biochemistry and significant effect on seed performance. In physiological processes. Most of the addition, Naz et al. (2012) observed that B micronutrients are cofactors in many enzymes increases growth and flowering of tomatoes. The and thus have enzyme-activating functions and combination of zinc (Zn) and B has significantly important in stabilizing proteins (Hänsch and enhanced fruit yield, total soluble solids and non- Mendel, 2009). Therefore, deficiency of these reducing sugar when applies through foliar (Asad nutrients can markedly reduce crop yield and et al., 2003). causing slow plant growth. Besides, element of copper (Cu) plays an important role as an activator of enzymes and

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

cooperates in electron transfers chain of reaction. CuSO4∙7H2O, MnSO4∙H2O and ZnSO4∙7H2O. This might effectively increased photosynthesis Media with nutrient solution were left in the and translocation of assimilates to the plants greenhouse for a week before transplanting the when plants are at a sufficient level of Cu. As for germinated seeds. calcium (Ca), it plays a role in determining the structural rigidity of the cell wall. During cell Seedlings were grown in ambient light wall development in a plant, low Ca and temperature conditions in a greenhouse concentration has the possibility to make the cell (USDA, Beltsville Agricultural Research Center) wall more pliable and easily ruptured, whereas with humidity was maintained above 75% using a high Ca concentration should rigidify the wall misting system (Atomizing Systems, Inc., Ho- and thus provides strength to the cell wall Ho-Kus, NJ, USA 07423-1433). In addition, the (Hepler, 2005). greenhouse was maintained at optimum growth conditions for the seedlings with heat was However, no information is available as provided when temperatures dropped below 20 regards to the effects of mineral elements such as °C and water-cooled air was provided when B, Cu and Ca on cacao seedlings. The main temperatures exceeded 29 °C. objective was to study the physiological responses of single or combination of minerals Spray application and sampling (boron, copper and calcium) through foliar Eight solutions of minerals were prepared application at cacao seedlings. according to treatments: (1) 4000 mg/L Ca(NO3)2 + 2 mg/L CuSO4 + 2 mg/L borax; (2) 2 mg/L CuSO4 + 2 mg/L borax; (3) 4000 mg/L Ca(NO3)2 MATERIALS AND METHODS + 2 mg/L borax; (4) 4000 mg/L Ca(NO3)2 + 2 mg/L CuSO4; (5) 2 mg/L borax; (6) 2 mg/L Preparation of seedlings CuSO4; (7) 4000 mg/L Ca(NO3)2 and (8) De- Well-developed cacao pods of genotype, IMC 47, ionized water. Foliar sprays were applied to the were received from USDA-ARS-Tropical seedlings at two-month old with a Binks model Agriculture Research station, Mayaguez, Puerto 15 sprayer at 15 psi at a rate of 50 ml m-2. Rico. Cacao seeds were extracted from the pods with the seed coats being removed, washed and Before spraying, the youngest or new cleaned with sterile deionized water. Seeds were emerged leaves (≥ 1 cm) were tied and tagged then allowed to germinate in water agar under using thread for each seedling to differentiate the laboratory conditions at 25 °C (± 2 °C) until the sprayed leaves (treated) and the new flush leaves emergence of radical from the seeds. (untreated) from the same plants. Physiological measurements were carried out on the treated and Nutrition solution, growth media and growth untreated leaves. conditions Seeds with at least one mm of radical were Physiological characteristics transplanted to 2.5 L of plastic pots (Classic 300 A month after sprayed, the net photosynthetic rate pots) with adequate bottom drainage. Soilless by unit of leaf area (PN), stomatal conductance to media (2:2:1, sand/perlite/promix) was prepared water vapor (gs), intercellular air space CO2 (Ci), using a cement mixer and nutrient solution with leaf transpiration rate (E) and the light adapted certain concentrations was mixed into the media photochemical efficiency of photosystem II (Fv/ to provide essential and same level of nutrients Fm) were measured by the infrared gas analyzer to each of the seedlings. The nutrient solution of a portable photosynthetic system (LI-6400; LI- contained macro- and micronutrients at COR, Inc., Lincoln, NE, USA) with an attached concentrations: 600 mg N/kg media, 331 mg P/kg leaf chamber fluorometer (LI6400-40). media, 240 mg K/kg media, 860 mg Ca/kg media, Measurements were measured from 0900 to 1400 288 mg Mg/kg media, 328 mg S/kg media, 100 h under a steady-state with the following mg Fe/kg media, 0.1 mg Mo/kg media, 1.0 mg conditions: leaf temperature = 25°C; [CO2] = 400 B/kg media, 1.0 mg Cu/kg media, 1.0 mg Mn/kg µmol mol-1; photosynthetic photon flux density media and 2.0 mg Zn/kg media. Nutrient solution (PPFD) = 500 µmol photons m-2s-1 and air flow = was prepared using salts of Fe DTPA (Sprint 100 µmol s-1. The tagged leaves was placed in the –2 –1 330), CaSO4∙2H2O, (NH4)6Mo7O24∙4H20, H3BO3, cuvette at PPFD of 600 µmol m s , temperature

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

of 28 0C, carbon dioxide concentration of 400 RESULTS AND DISCUSSION cm3m–3 and vapor pressure deficit (VPD) of 2.0 - 2.1 kPa. The leaf was allowed to equilibrate for Effects of minerals on plant photosynthesis 15 min and the rate of PN was recorded. Despite, All the treatments had higher net photosynthesis measurements on growth such as height of plant rate (PN) in treated leaves sprayed with minerals and chlorophyll were also recorded. except B+Cu+Ca as compared to control (water). Plants sprayed with treatments B+Cu+Ca, B+Cu, Experimental design B and Cu produced new flush leaves with higher A randomized complete block design was used efficiency in photosynthesis with the rate ≥ 3.50 and the experimental units were replicated three µmol m-2s-1 compared to other treatments (Figure times. Result was performed using Microsoft 1). Interestingly, Ca combination with B and Cu Excel (2007) as mean ± standard deviation. did not photosynthesize actively after sprayed, especially plants sprayed with Ca+Cu produced - new flush leaf with even lower PN (2.49 µmol m 2s-1 ) compared to sprayed leaf (2.63 µmol m-2s-1).

Figure 1. Effects of minerals on photosynthesis of treated leaves and untreated new flush leaves.

The plants might show symptom of Furthermore, it might disrupt the activity of stunted growth (data not shown) and the spraying Calvin cycle enzymes such as the Rubisco of Ca+Cu showed adverse effects on the plants (Panou-Filotheou et al., 2001; Pätsikkä et al., due to photosynthesis limitation promoted by 2002; Shaul, 2002). A reduction in PN was Ca+Cu stress in cacao seedlings. This might be observed when the plants were exposed to excess due to the excess Cu or Ca inhibits the activity of of Cu (Moustakas et al., 1997; Vassilev et al., several enzymes and interferes with plant 2002; Shi-Sheng, 2007; Cambrollé et al., 2011). biochemistry such as photosynthesis, respiration Overall, emergence of new flush leaves without and integrity of membranes (Almeida et al., 2007 minerals spray showed higher PN compared to not in the references). Consequently, excess Cu leaves treated with minerals spray except or Ca might leads to changes in chloroplasts, treatment Ca+Cu (Figure 1). New flush leaves causing reduction of electron transport in with active photosynthetic rate, particularly in Photosystems I and II (PS I and PS II). cacao seedlings sprayed with B alone or

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

combinations, this might due to the reason that stresses on photosynthesis (Baker, 2008). Krause boron is considered as one of the important and Weis (1991) reported that Fv/Fm values can mineral in the process of germination, fruit be related to changes in the quantum efficiency of formation and transfer of photosynthetic photosynthesis. This can be explained that the components (Malakoti and Keshavarz, 2003). efficiency where the light absorbed by the leaf’s Recent study in cashew recorded that significant pigment during PSII is used to drive a stable effects were observed in PN sprayed with ZnSO4 photochemistry in photosynthesis. Thus, (0.5%) + B (0.1%) (Lakshmipathi et al., 2015). untreated new flush leaves with lower Fv/Fm compared to treated leaves in B+Ca and control Overall, treated leaves sprayed with the were assumed that the cacao seedlings might treatments produced new flush leaves with better suffered from abiotic stress, probably toxicity in photosynthetic efficiency at about 20% increase B+Ca or nutrient insufficient level in control. in Fv/Fm values except B+Ca and control (Figure According to Denamay (1995), cocoa leaf with 2). The Fv/Fm ratio is considered as a good Ca nutrient of more than 0.4% consider normal indicator to measure the effects of environmental and only deficient when Ca is less than 0.3%.

Figure 2. Photosynthetic efficiency in treated and untreated cacao seedlings.

Effects of minerals on stomatal conductance seedlings sprayed with B, 0.051 mmol m⁻² s⁻1. In The result was coincided with the net addition, leaves sprayed with B+Cu+Ca had photosynthetic rate where cacao seedlings impressively increased 78.3% of stomatal sprayed with B+Cu+Ca, B+Cu and B conductance in the untreated new flush leaf successfully produced new flush of leaves with followed by B+Cu (48%). Seedlings sprayed with had higher stomatal conductance (Figure 3). Ca+Cu had lowered the stomatal conductance in Cacao seedlings sprayed with B+Cu produced the new flush leaves as compared to the sprayed new flush leaf with the highest stomatal leaves (Figure 3). conductance, 0.055 mmol m⁻² s⁻1, followed by

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Figure 3. Analysis of stomatal conductance on cacao leaves supplemented with minerals and untreated new flush leaves with no mineral supplementation.

Stomatal conductance measures the to control except B+Ca. Seedlings sprayed with B efficiency of water and gaseous exchange and B+Cu produced new flush leaves with the between the plant and the atmosphere. highest transpiration rate of 0.66 mmol m-2 s-1 for Theoretically, leaves with higher stomatal both the treatments. Cacao seedlings sprayed with conductance shows better photosynthetic B+Ca had the lowest transpiration rate at 0.40 efficiency and this leads to a better food mmol m-2 s-1 in new flush leaves (Figure 4). production and nutrient uptake by the plants. As a Normally, plants with higher transpiration rate result, new flush leaves of seedlings sprayed with have higher photosynthetic efficiency and this B+Cu+Ca, B+Cu and B which had higher result was coincided with higher stomatal stomatal conductance can allow seedlings to conductance. increase their CO2 uptake and subsequently enhance photosynthesis. Lakshmipathi et al. Transpiration acts as a main force in the (2015) reported that cashew sprayed with B alone xylem stream (White and Broadley, 2003) and or in combination of ZnSO4 had the maximum seedlings with high transpiration rate might stomatal conductance. changes the osmotic pressure of cells, thus enables mass greater flow of nutrients and water Effects of minerals on plant transpiration rate from roots to shoots (Taiz et al., 2015). Figure 4 showed that seedlings sprayed with minerals had higher transpiration rate compared

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Figure 4: Transpiration rate of cacao seedlings for treated and new flush untreated leaves.

Effects of minerals on plant intercellular CO2 may be an indicative to increased limitations of concentration photosynthesis (Tardieu and Simonneau, 1998). There were not much differences among the In this study, seedlings sprayed with minerals did -1 treatments (200 – 260 µmol mol ) except not show impact on the plant intercellular CO2 treatment with the combination of all the concentration as compared to control. Study micronutrients, B+Cu+Ca with intercellular CO2 showed a gradual decrease in the leaf intercellular -1 concentration of less than 200 µmol mol (Figure CO2 concentration of dryland maize after exposed 5) in treated leaves. Lower intercellular CO2 to increasing rate of organic fertilizer (Wang et concentration in the leaf sprayed with B+Cu+Ca al., 2012).

Figure 5: Intercellular carbon dioxide concentration on treated and untreated cacao seedlings.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

CONCLUSIONS Denamay, G. (1995). Cocoa leaf nutrient standards-Reviewed. Seminar on Crop Overall, cacao seedlings supplemented with Nutrition of Field Crops. MARDI. different micronutrients and combinations 1995. showed different physiological responses. Seedlings treated with combination of B+Cu and Fatemi, N. (2001). Determination effects of B showed positive physiological performances planting date and nitrogen and boron by producing new flush leaves with active foliar application on quality and photosynthetic efficiency, gaseous exchange quantity of soybean seeds. M.Sc, through higher stomatal conductance and Thesis, Univ. of Tarbiat Modars, Iran. transpiration rate compared with other treatments. Seedlings sprayed with Ca+Cu Hänsch, R. and Mendel, R. R. (2009). performed the least on physiology parameters. Physiological functions of mineral In conclusion, minerals of B+Cu and B are micronutrients (Cu, Zn, Mn, Fe, Ni, encouraged to be sprayed on cacao seedlings as Mo, B, Cl). Curr. Opin. Plant Biol., 12: they produced new flush leaves with better 259-266. physiological responses while the combination of Ca+Cu is not suitable to be sprayed on cacao Hepler, P. (2005). Calcium: a central regulator seedlings at current concentration. of plant growth and development. The Plant Cell, 17(8): 2142-2155.

ACKNOWLEDGEMENT Krause, G. H. and Weis, E. (1991). Chlorophyll fluorescence and photosynthesis: The The study was financially supported by World basics. Ann. Rev. Plant Physio. and Cocoa Foundation and US Department of Mol. Bio. 42: 313–349. Agriculture under the Program of Cocoa Borlaugs Fellowship. Special thanks to the Lakshmipathi, Dinakara Adiga, J., Kalaivanan, Director of Cocoa Upstream Technology D., Mohana, G. S. and Ramkesh Division of Malaysian Cocoa Board for Meena. (2015). Effect of foliar supporting the fellowship program to USDA, application of certain micronutrients on Beltsville. photosynthesis and yield of Cashew (Anacardium occidentale L.) var. Bhaskara under South West Coast REFERENCES region of Karnataka. Eco. Env. and Cons. 21 (1) : 517-520. Asad, A., Blamey, F.P.C. and Edwards, D.G. (2003). Effects of boron foliar Malakoti, M. and Keshavarz, P. (2003). Boron applications on vegetative and place in plant efficient nutrition. 1st ed. reproductive growth of sunflower. Ann. Iran. Sana Publication. Bot., 92: 565-570. Moustakas, M., Ouzounidou, G., Symeonidis, L. Baker, N. R. (2008). Chlorophyll fluorescence: and Karataglis, S. (1997). Field study of A probe of photosynthesis in vivo. Ann. the effects of excess copper on wheat Review of Plant Bio. 59: 89–113. photosynthesis and productivity. Soil Sci. Plant Nutr. 43:531–539 Cambrollé, J., Mateos-Naranjo, E., Redondo- Gómez, S., Luque, T. and Figueroa, Naz, R. M. M., Muhammad, S. M., Hamid, A. M.E. (2011). Growth, reproductive and and Bibi, F. (2012). Effect of boron on photosynthetic responses to copper in the flowering and fruiting of tomato. the yellow-horned poppy, Glaucium Sarhad J. Agric., 28: 37-40. flavum Crantz. Environ. Exp. Bot. 71:57–64 Panou-Filotheou, H., Bosabalidis, A. M. and Karataglis, S. (2001). Effects of copper toxicity on leaves of oregano

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

(Origanum vulgare subsp. hirtum). Ann. Bot. 88:207–214. Pätsikkä, E., Kairavuo, M., Šeršen, F., Aro, E.M. and Tyystjärvi, E. (2002). Excess copper predisposes photosystem II to photoinhibition in vivo by outcompeting iron and causing decrease in leaf chlorophyll. Plant Physiol. 129:1359–1367.

Shaul, O. (2002). Magnesium transport and function in plants: the tip of the iceberg. BioMetals 15:309–323

Shi-Sheng, K. E. (2007). Effects of copper on the photosynthesis and oxidative metabolism of Amaranthus tricolor seedlings. Agr Sci in China 6:1182– 1192

Taiz, L., Zeiger, E., Moller, I. and Murphy, A. (2015). Plant Physiology and Development. Sunderland, MA: Sinauer Associates. Inc.

Tardieu F. & Simonneau T. (1998) Variability among species of stomatal control under fluctuating soil water status and evaporative demand: modelling isohydric and anisohydric behaviours. J Expt. Botany. 49: 419–432.

Vassilev, A, Lidon, F. C., Matos, M. C., Ramalho, J. C. and Yordanov, I. (2002). Photosynthetic performance and content of some nutrients in cadmium- and copper-treated barley plants. J Plant Nutr. 25:2343– 2360.

Wang, X. J., Jia, Z. K., Liang, L. Y., Ding, R. X., Wang, M. and Li, H. (2012). Effects of organic fertilizer application rate on leaf photosynthetic characteristics and grain yield of dryland maize. J. Appl. Ecol. 23(2): 419-425.

White, P. and Broadley, M. R. (2003). Calcium in plants. Ann. Bot. 92: 487-511.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

THE EFFICIENCY OF CACAO SPECIFIC COMPOUND FERTILIZER ON CACAO PRODUCTIONS

Helmi, S.1, Rozita.O.2., Haya. R.3, Ling, A.S.C.3 and Halim, H.4 1Malaysian Cocoa Board, Cocoa Research and Development Center Hilir Perak, Peti Surat 30, Jalan Sg.Dulang, 36307Sg. Sumun,Perak Darul Ridzuan 2Malaysian Cocoa Board, Cocoa Research and Development Center Jengka, Jalan Jengka 23, Peti Surat 34, 28000 Temerloh,Pahang Darul Makmur. 3 Malaysian Cocoa Board, 5th,6th & 7th Floor,Wisma SEDCO, Locked Bag 211 88999 Kota Kinabalu, Sabah. 4Malaysian NPK Fertilizer Sdn. Bhd.P.O.Box 24, Batu 2, Jalan Jeniang, 08300 Gurun Kedah Darul Aman. Email: [email protected]

Malaysian Cocoa Journal 9(1): 79-85 (2016) ABSTRACT - The trial aims to evaluate the efficacy of two fertilizer formulation compared to current compound fertilizer available in the market to increase cocoa yield productions. The fertilizer formulations were derived from nutrient trial done by Malaysian Cocoa Board previously. Granular semi-true compound fertilizers were then produced using method similar to mass production of compound fertilizer in the factory. Three fertilizer formulations were applied in four different locations since April 2014. The two formulations are 10N : 7P : 19K : 2 + 17% Zeolite (T1) and 15N : 7P :20K : 2 + 7% Zeolite (T2). Compound fertilizer NPK Blue or 12N : 12P : 17K : 2 were use as a control (C). One kg of fertilizer per tree were applied annually on all four different locations ie. Cocoa Research and Development Centre (CRDC) Hilir Perak, Sungai Air Tawar, CRDC Jengka and Kuala Medang. Yield, soil, leaf nutrient analysis pest and diseases were recorded. The result showed that treatment 2 (NPK 15:7:20:2 with 7% Zeolite + TE) produces significantly higher yield compared to other treatments in three of four locations, while treatment 1 (NPK 10:7:19:2 with 17% Zeolite + TE) produces the highest yield in Kuala Medang. It is found that treatment 2 formulations showed better than current fertilizer practices.

Key words: Compound fertilizer, Slow release, Cacao, Alluvial soil, Zeolite.

INTRODUCTION in cocoa cultivation, mainly due to increase cost of fertilizers, and it is a merit that the efficiency of Each type of plant has its unique optimum nutrient added fertilizers to be increased. range. These ranges consist of minimum and maximum critical level (Uchida and Silva, 2000). Cocoa is considered one of the important When insufficient nutrient is provided to the plant commodity crops. However, cocoa is a less below the minimum level, plants will show nutrient favourable crop among farmers. Thus cocoa is deficiency symptoms. Meanwhile, excessive normally cultivated in unfavourable area such as nutrient uptake will also cause poor plant costal area with sandy costal alluvial soil. Sandy development due to toxicity. Hence, a proper and sedimentary/alluvial soils are very poor in retaining balance amount of fertilizer applied is important. soil nutrient. This is because negatively charged This is magnified much more on commodity crops nutrient normally does not have much affinity for such as cocoa. The total input cost on fertilizer for soil (sand) particle surfaces, and thus does not mature cocoa plantings is more than 60% (Harnie, absorbs on soils (Li, 2003). Huge amount of 2007) while 30-35% of cost is labour, inclusive of chemical fertilizer are applied to fields, of which activities related to fertilizer applications (usually around 50% remain unused and prone to getting 3-4 times per year). The escalating productions cost lost from the system (Smil, 1999; Liu et al., 2010).

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Loss of cationic plant nutrients can be reduced Bernam with costal alluvial costal soil, CRDC using cation exchangers as additives in fertilizers to Jengka and Kuala Medang with inland soil. The + control NH4 release. two new fertilizers formulations 10N : 7P : 19K : 2Mg with 17% Zeolite and 15N :7P : 20K : 2Mg Calinoptilolite zeolite has often been used with 7% Zeolite were evaluated. A standard NPK + as an inexpensive cation exchanger to preload NH4 Blue fertilizer formulation 12N :12P : 17K : 2Mg at + (Miller, 1998; Allen, 1996). Mixing NH4 saturated the rate of 1,066 kg/ha were used as a control zeolite with phosphate rock provide slow release of treatment. The amount fertilizers applied of each both P and N .Zeolite increases the CEC of soil treatment are similar. All the treatments (fertilizer (Suwardi et al., 1994) and reduces leaching of applications) were split into three applications per cationic plant nutrients (Huang and Petrovic, 1994; year. The plot size is 45 trees including buffer zone. Malekian et al., 2011). Zeolite may also improve Each treatment consists of three replicates, thus the physical properties of soils including water makes the trial at 405 trees for each location. The retention and resistance to water repellence trials were laid out in Randomize Complete Block (Petrovic, 1990; Xiubin and Zhanbin, 2001). Design (RCBD) for each location. Blocking was Natural zeolite has been used as a slow release done due to heterogeneity of soil condition. fertilizer providing either natural or added elements Regardless of treatment, the whole fields were (Carlino et al., 1998; Perrin et al., 1998a,b; provided with 440kg/ha/year of Lime Stone Dust. A Williams and Nelson, 1997). standard field management practices were also implemented to maintain the field. Our previous study found that mixing of straight fertilizer with zeolite and applied on mature Treatments cocoa trees had shown higher yield compared to no zeolite treatment (Helmi, 2014). Balanced fertilizer Treatment 1: NPK 10:7:19:2 with 17% Zeolite formulations with precise nutrient ratio for specific + TE plant such as cocoa are long overdue. Treatment 2: NPK 15:7:20:2 with 7% Zeolite + TE Compound or complex fertilizer contains Treatment 3: Compound fertilizer NPK Blue multiple nutrients in each individual granule. This 12:12:17:2 +TE differs from a blend of fertilizers mixed together to Each treatment was applied on to their respective achieve a desired average nutrient composition. plot with the amount of 323g/tree three times a This allows compound fertilizer to be spread so that year. each granule delivers a mixture of nutrients as it dissolves in the soil and eliminates the potential for Parameters segregation of nutrient sources (Piwowar, 2011). a. Yield component: actual harvested pods, This study is an assessment on the efficacy Pod Values and Dry Bean weight of compound fertilizer with specific nutrient ratio b. Pest assessment: Cocoa pod borer (CPB) towards mature cocoa productivity. The damage based on average damage severity formulations were derived from cocoa nutrient index (ADSI). requirement. The study compared two formulations c. Disease assessment: Vascular streak which are different version of a formulation for dieback assessment based on average cocoa compromising the amount of zeolite used. disease severity index. . d. Soil analysis: Two times per annum. (Exchangeable cations, pH, total nitrogen MATERIALS AND METHODS content, organic carbon and available phosphorus). The trial was done on four different locations, Cocoa Research and Development Centre (CRDC) Hilir Perak and Kg. Sungai Air Tawar, Sabak

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

RESULTS AND DISCUSSIONS 2014. The yield trend was somewhat distroted after the flood. Potential Yield Figure 1 and 2 showed the cumulative yield In CRDC Hilir Perak trial plot, treatment obtained from all locations. For CRDC Jengka trial 2 produces 1,916.37 kg of dry cocoa bean per plot, treatment 2 produces 2,463.34 kg of dry cocoa hectare for 26 months. The yield obtained is bean per hactare for 26 months. The yield obtained significantly higher compared to treatment 1 and is significantly higher compared to treatment 1 and control 1,769.78 kg/ha and 1,595.33 kg/ha, control 2,230.69 kg/ha and 2128.16 kg/ha, repectively. The data shows similar trend in Kg. repectively. Meanwhile, the yield data obtained Sungai Air Tawar trial plot; treatment 2 produces from Kuala Medang did not showed any significant 3,382.89 kg of dry cocoa bean per hactare for 26 difference among the treatments. As a disclaimer, months. The yield obtained is significantly higher Kuala Medang trial plot was flooded in December compared to treatment 1 and control 2,925.53 kg/ha and 2,909.57 kg/ha, repectively.

(A)

(B) Figure 1A and B. Dry bean data data collected from trial plots (inland plots).

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

(A)

(B)

Figure 2A and B. Dry bean data data collected from trial plots (Costal land plots)

Vascular Streak Dieback Assesment four different location. But there are significant Table 1 shows the VSD assessment in January differences among the locations. (P = 3.0x10-8). 2016, at the end of the project. Data were colected This indicatese that VSD infection are varries every four month. The results showed no among trial plots. significant difference among the treatments at all

Table 1. VSD assessment for 2016 Treatment CRDC Hilir Farmer's plot CRDC Jengka Farmer's plot Perak Ayer Tawar Kuala Medang T1 2.42a 0.48a 2.59a 0.21a T2 2.08a 0.43a 2.85a 0.05a C 2.52a 0.47a 2.62a 0.11a Means followed by same letters at the same rows are not significantly different (p ≥ 0.05)

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Cocoa Pod Borer (CPB) Assesment treatment in each location. This indicates that the Table 2 shows the CPB assessment in April 2016 treatment does not have any effect on CPB at the end of the project. The Average Damage infestations. Although there are differences in Severity Index (ADSI) assesment indicates that ADSI when compared locations to locations, this there were no significant difference among is merely the effect of different CPB exposure in.

Table 2. Average Damage Severity Index 2016 Treatment CRDC Hilir Farmer's plot CRDC Jengka Farmer's plot Perak Ayer Tawar Kuala Medang T1 1.37a 0.33a 0.91a 0a T2 1.5a 0.4a 1.18a 0a C 1.6a 0.5a 0.92a 0a Means followed by same letters at the same rows are not significantly different (p ≥ 0.05)

Soil Nutrient Analysis soil more acidic and thus the availability of Table 3 shows the results of soil analysis at the end nutrient is lower when soil pH is not maintained. of the project. Higher Cation Exchangeable Table 3 also includes the critical level of nutrient Capacity in coastal area is due to high amount of required for cocoa productions. clay in coastal alluvial soil. But this also makes the

Table 3.1: Soil analysis for plots in CRDC Jengka pH Org Total Avail P CEC Exch Exch Exch C N (mg/kg) (cmol K Ca Mg (%) (%) (+)/kg) Treatment 1: NPK 10:7:19:2 + TE 4.48 1.42 0.1 32.67 5.10 0.13 2.26 0.57 with 17% Zeolite Treatment 2: NPK 15:7:20:2 + TE 4.50 1.43 0.11 22.33 5.00 0.12 2.4 0.55 with 7% Zeolite Treatment 3: Compound fertilizer 4.49 1.42 0.11 27.50 5.05 0.13 2.33 0.56 NPK Blue 12:12:17:2 +TE (control)

Critical Level 5.5-6.5 2 0.16 15 12 0.24 2.5 2

Table 3.2: Soil analysis for plots in Kuala Medang pH Org Total Avail P CEC Exch Exch Exch C N (mg/kg) (cmol K Ca Mg (%) (%) (+)/kg) Treatment 1: NPK 10:7:19:2 + TE 4.25 2.69 0.15 6.07 8.56 0.23 1.32 0.51 with 17% Zeolite Treatment 2: NPK 15:7:20:2 + TE 4.26 2.67 0.13 10.58 11.53 0.24 0.81 0.61 with 7% Zeolite Treatment 3: Compound fertilizer 4.32 2.94 0.14 10.79 12.63 0.29 0.76 0.68 NPK Blue 12:12:17:2 +TE (control) Critical Level 5.5-6.5 2 0.16 15 12 0.24 2.5 2

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 3.3: Soil analysis for plots in CRDC Hilir Perak pH Total Avail P CEC Exch Exch Exch N (mg/kg) (cmol K Ca Mg (%) (+)/kg) Treatment 1: NPK 10:7:19:2 + TE with 4.62 0.24 38.67 23.06 0.97 2.02 6.91 17% Zeolite Treatment 2: NPK 15:7:20:2 + TE with 4.47 0.19 39 23.63 0.57 1.36 8.15 7% Zeolite Treatment 3: Compound fertilizer NPK 4.45 0.2 47 22.38 0.62 1.27 6.2 Blue 12:12:17:2 +TE (control) Critical Level 5.5-6.5 0.16 15 12 0.24 2.5 2

Table 3.4: Soil analysis for plots in Air Tawar pH Total Avail P CEC Exch Exch Exch N (%) (mg/kg) (cmol K Ca Mg (+)/kg) Treatment 1: NPK 10:7:19:2 + TE with 4.71 0.19 79.67 21.87 0.7 9.44 1.19 17% Zeolite Treatment 2: NPK 15:7:20:2 + TE with 7% 4.98 0.15 48 21.52 0.79 8.59 1.29 Zeolite Treatment 3: Compound fertilizer NPK 4.9 0.18 118.67 24.2 1.24 9.29 1.78 Blue 12:12:17:2 +TE (control) Critical Level 5.5-6.5 0.16 15 12 0.24 2.5 2

CONCLUSIONS General of Malaysian Cocoa Board for the permission to publish this paper. Appreciation is The duration of the trial has been more than 2 extended to the Deputy Director-General years and the results found that monthly dry bean (Research & Development) and Director of Cocoa yield data showed significant differences among Upstream Technology Division for his guidance. the treatments. Treatment 2 showed higher yield Acknowledgement is also conveyed to CRDC compared to other treatments in three out of four Hilir Perak center manager for allowing the trial to locations. Increase in yield can be detected as soon be done in its location. Gratitude is also shown to as 8 month after application. The treatments do not the late Mr. Denamany for his previous findings have any effects on VSD and CPB assessment. and notes, and guidance from fellow colleagues Dr. Rozita Osman, Dr. Mohd. Yusof Abdul Samad and Mr. Albert Ling . Finally, thanks to Mr. ACKNOWLEDGEMENT Shaifullah b. Sanusi, Mr. Sabar Saihone, En. Saful Mujahid A. Rahman, Mohd Zam Zam Bin Kasmin This research was co-funded by National Farmer and Miss Salwa Nazri for their assistance in the Organizations (NAFAS) and Malaysian Cocoa field work. Board. The author acknowledges the Director

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

REFREENCES Petrovic, M.A. (1990). The potential of natural zeolite as a soil amendment. Golf Allen E.R., Ming D.W., Hossner L.R., Course Management (November Henninger D.L., (1996). Soil Sci. Soc. issue), pp. 92-94. Am. J. 60: 1467. Perrin, T.S., Boettinger, J.L., Drost, D.T. and Carlino, J., Williams, K.A.,and Allen, E.R., Norton, J.M. (1998a). Decreasing (1998). Evaluation of zeolite-based nitrogen leachingfrom sandy soil with soilless root media for potted ammonium-loaded clinoptilolite. J. of chrysanthemum production. Environmental Quality 27: 656–663. HortTechnology 8: 373–378. Perrin, T.S., Drost, D.T., Boettinger, J.L. and Harnie, H. (2007). Cost Analysis for Cocoa Norton, J.M. (1998b). Ammonium- Production. Malaysian Cocoa Board. loaded clinoptilolite: a slow-release nitrogen fertilizer for sweet corn. J. of Helmi, S., Albert, L.S.C. and Rozita O. (2014). Plant Nutrition 21:515–530. Efficacy of slow release zeolite base compound fertilizer on Theobroma Piwowar, A., (2011) Innovations in the sphere of cacao. In Proceedings Malaysian mineral fertilization and their practical International Cocoa Conference (MICC application. Postępy Nauk Rolniczych 2013), Kuala Lumpur, Malaysia, 7- 3: 47-56 8th October 2013. Pp. 157 - 160. Uchida, R. and Silva, J. A. (2000). Plant Huang, Z.T. and Petrovic, A.M. (1994). Soil Nutrient Management in Hawaii’s process and chemical transport - Soils, Approaches for Tropical and clinoptilolite zeolite inﬂuence on Subtropical Agriculture, eds. nitrate leaching and nitrogen use College of Tropical Agriculture efficiency in simulated sand based golf and Human Resources, University of greens. Journal of Environmental Hawaii at Manoa. Quality 23: 1190-1194. Smil, V. (1999). Nitrogen in crop production: Li, Z . (2003). Use of surfactant-modified zeolite an account of global flows. Global as fertilizer carriers to control nitrate Biogeochemical Cycles 13: 647-662. release. Microporous and Mesoporus Materials 61: 181-188 Suwardi, Goto, I. and Ninaki, M. (1994). The quality of natural zeolites from Japan Liu, J., You, L., Amini, M., Obersteiner, M., and Indonesia and their application Herrero, M., Zehnder, A. J. B., and effects for soil amendment. Journal of Yang, H. (2010). A high-resolution Agricultural Science Tokyo Nogyo assessment on global nitrogen flows in Daigaku 39: 133-148. cropland. Proc. Natl. Acad. Sci., 107: 8035–8040, Williams, K.A., Nelson, P.V., (1997). Using precharged zeolite as a source of Malekian, R., Abedi-Koupai, J. and Saeid potassium and phosphate in a soilless Eslamian, S. (2011). Inﬂuences of container medium during potted clinoptilolite and surfactant-modified Chrysanthemum production. Journal of clinoptilolite zeolite on nitrate the American Society for Horticultural leaching and plant growth. Journal of Science 122:703–708. Hazardous Materials 185: 970-976. Xiubin, H. and Zhanbin, H. (2001). Zeolite Miller R.W., Gardiner D.T., Soils in Our application for enhancing water Environment, (1998). Eigth ed., infiltration and retention in loss soil. Prentice Hall, Upper Saddle River, NJ. Resources, Conservation and Recycling 34: 45–52.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

UNRAVELLING THE INTERACTION BETWEEN LEPIDOPTERAN SODIUM CHANNEL AND NEUROTOXIN PEPTIDE: A STRUCTURAL INSIGHT

Rahman, M. Z. A., Jainuddin, H. and Kasran, R. Malaysian Cocoa Board, Cocoa Biotechnology Research Centre, Kota Kinabalu Industrial Park, 88460 Kota Kinabalu, Sabah, Malaysia

Malaysian Cocoa Journal 9(1): 86-96 (2016) ABSTRACT - Insect pest especially Lepidoptera are burden to humans as they destroy economical important crops around the world. Moreover, a Lepidopteran species known as Conopomorpha cramerella was one of the major treat to the cocoa industry in Southeast Asia. Widespread of heavily used of agrochemicals have resulted in genetic selection pressure that led to the development of insecticide- resistant arthropods, as well as concerns over human health and the environment. Seeking and exploring for the new insecticide targets can be very challenging and daunting task. Most of the commercial insecticides target channel proteins as their main receptor, however, the structure of the protein was still absence for Lepidoptera. In order to understand the structure and function of Lepidopteran sodium channel, homology modeling was conducted to model the pore-forming tetrameric protein complex from Bombyx mori protein sequence. Molecular docking experiments were conducted to study the interaction of protein complex between the Lepidopteran sodium channel and targeted insecticides. Detail protein-ligand interaction was examined at molecular level and key interacting residues were identified in order to depict the mechanism underlying the inhibitory action of the insecticidal lead targets.

Key words: Sodium channel protein, Lepidoptera, Neurotoxin, Protein modeling, Molecular docking

INTRODUCTION an urgent need to develop novel classes of insecticides or alternative methods of insect pest Lepidoptera are the second most diverse pest control. insect order outnumbered only by the beetles. There is hardly any cultivated plant that is not A promising approach in the attacked by at least one lepidopteran pest. One of agricultural sector is to engineer crops to the notable pest that greatly impact cocoa produce insecticidal toxins. The introduction of production especially in Southeast Asia is cocoa GM crops that express Bacillus thuringiensis pod borer. Chemical insecticides remain the (Bt) toxins has provided an alternative and dominant solution for combating insect pests potentially safer method of insect control than despite the introduction of transgenic crops and chemical insecticides. Bt toxins kill certain other biological control methods. The major insect pests but are harmless to most other classes of chemical insecticides act on only six creatures including people. These molecular targets in the insect nervous system, environmentally friendly toxins have been used namely acetylcholinesterase, voltage-gated for decades in sprays by organic growers and sodium (NaV) channels, nicotinic acetylcholine since 1996 in engineered Bt crops by mainstream receptors, GABA- and glutamate-gated chloride farmers. However, over the past few years, some channels, and ryanodine receptors (Lai and Su, insect pest species have evolved to confer the 2011). The widespread use of chemical resistance towards Bt toxin by rare genetic insecticide over several decades has promoted mutations (Bates et al., 2005). Therefore, the evolution of resistant insect populations, with alternative insect-toxin transgenes are urgently more than 600 insect species developed the needed. There are very few well characterised resistance towards one or more classes of toxins that could be considered as alternatives or chemical insecticides (Bass and Field, 2011). In adjuncts to Bt. However, some of the most addition, key classes of insecticides have been promising candidates are novel insecticidal withdrawn from sale or their use has been peptides that have been isolated from the venom restricted by regulatory authorities due to of spiders (Maggio et al., 2010). Most of these growing environmental and human health peptides now are highly stable because they concerns (King and Hardy, 2013). Thus, there is contain an inhibitor cystine knot (ICK) motif

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

(Pallaghy et al., 1994) that provides them with The protein sequence of Helicoverpa armigera resistance to extremes of pH, high temperatures, (cotton bollworm), Bombyx mori (silkworm) and and proteolytic enzymes (Saez et al., 2010). One Heliothis virescens (tobacco budworm) were of the first insecticidal spider-venom peptides to retrieved form UniProt database be reported was Aps III from the venom of the (www.uniprot.org). Multiple Sequence trapdoor spider Apomastus schlingeri (Skinner et Alignment (MSA) was performed by using web- al., 1992). With a reported LD50 of 133 pmol/g based tool (http://www.ebi.ac.uk/Tools/msa) to against the tobacco hornworm Manduca sexta, align the selected protein sequences in order to this peptide is one of the most potent insect reveal the evolutionary relationships and toxins described to date according to sequence homology between those Lepidoptera ArachnoServer (Wood et al., 2009). Most toxins species. from venom target Nav channels for the purposes of prey subjuga¬tion and self-defense Protein modelling of Lepidopteran sodium (Billen et al., 2008). While the overall domain channel architec¬ture of insect and vertebrate Nav Sodium channel from B. mori was modelled by channels are very similar, there are considerable using homology modelling approach and the sequence differences; insect Nav channels are protein sequence was retrieved from UniProt only ~60 % homologous to their human Database (www.uniprot.org). The automated counterparts (King et al., 2008), which provides homology mode was conducted and YASARA ample opportunity for producing Nav channel software version 12.5.7 (YASARA Biosciences insecticides that are insect-specific. GmbH, Austria) was used to visualize and model the protein structures. The Structure Quality Advancement in the field of Assessment of the predicted models were computational biology paved a new route for evaluated using WHAT_CHECK tools. scientist to develop an effective and environmentally friendly insecticides. By Molecular docking of commercial insecticides employing molecular modelling and structure- The predicted protein model of B. mori sodium based insecticide design an ideal insecticide channel was selected as the receptor and six leads can be generated. This emerging approach commercial insecticides (permethrin, however need a detail understanding of the cypermethrin, deltamethrin, fenvalerate, mechanism of the molecular target such as the fluvalinate, cycloprothrin) were selected as the three dimensional structure. Voltage-gated ligands for the molecular docking experiment. sodium (NaV) channels is one of the common The three dimensional structure of the chemical molecular target for the commercial toxicants insecticides were retrieved from PubChem available in the market such as DDT, database (https://pubchem.ncbi.nlm.nih.gov). pyrethroids, N-alkylamides, and The global docking experiments were performed dihydropyrazoles. However, to date there was no using AutoDock 4.2 (Morris et al., 1998) using sodium channel protein structure reported yet for the default docking parameters supplied with the Lepidoptera family and without the structure, AutoDock and the point charges assigned progress on the insecticide lead design and according to the AMBER03 force field (Duan et discovery will be hindered. Therefore, in this al., 2003). The setup was done with the study, protein modelling technique will be used YASARA molecular modeling program version to unravel the structural insight of lepidopteran 12.5.7 (YASARA Biosciences GmbH, Austria). sodium channel. Molecular docking will be The docking grid was generated 5.0 Å around all performed to depict the interaction between the atoms with periodic boundaries and a total of 25 channel protein with the neurotoxin peptides as docking runs were set up for each docking well as the commercial chemical insecticides. experiments. The docked conformations from 25 runs were clustered and significant contacting amino acid residues were determined and ranked MATERIALS AND METHODS based on the binding energy. The top ranked conformations were visually analysed and detail Sequence analysis of Lepidopteran sodium residues involved in protein-ligand interaction channel protein were determined.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Data mining of potent insect-selective nervous system for pesticide development. Most neurotoxin peptides of the commercial chemical insecticides in the Data mining against ArachnoServer market right now, utilized this receptor for their (http://www.arachnoserver.org) were conducted insecticide mode of action. However, till now, in order to search for the ideal reference there is no sodium channel protein structure neurotoxin peptides originated from spiders. reported yet for lepidopteran. In order to study Several criteria were set such as exhibiting high the mode of binding and interaction between the toxicity, high selectivity and having solved active chemical compound of insecticide and protein structure deposited in the Protein Data sodium channel receptor protein at molecular Bank (http://www.rcsb.org). level, the protein structure of the receptor protein need to be modelled. Therefore, homology Molecular docking of neurotoxin peptides modelling was conducted to model the three The predicted protein model of B. mori sodium dimensional structure of Lepidoptera’s sodium channel was selected as the receptor and ten channel protein. By assuming that Lepidoptera neurotoxin peptides from eight different spider family would share similar protein folds species (ApsIII, OAIP-1, U3-cyrtautoxin-As1a, regarding their amino acid sequences, the ω-hexatoxin-Hv1a, ω-hexatoxin-Hv1f, ω- sodium channel protein sequences from H. hexatoxin-Hv2a, κ-theraphotoxin-Ps1a, π- armigera, B. mori and H. virescens were theraphotoxin-Pc1a, μ-theraphotoxin-Hhn1a, μ- retrieved form UniProt database theraphotoxin-Hhn2a) were selected as the (www.uniprot.org). Multiple sequence alignment ligands for the molecular docking experiment. was performed for the three lepidopterans and The three dimensional structure of the the similarities and the conserved regions were neurotoxin peptides were retrieved from Protein observed as shown in Figure 1. Because the Data Bank (http://www.rcsb.org). The global protein sequence among the lepidopterans are docking experiments were performed using highly conserved, all of them can be selected to AutoDock 4.2 (Morris et al., 1998) using the model Lepidopteran sodium channel, however default docking parameters supplied with for this study we chose sodium channel from B. AutoDock and the point charges assigned mori for the subsequent experiment. according to the AMBER03 force field (Duan et al., 2003). The setup was done with the In order to search for the suitable YASARA molecular modeling program version protein template, PSI-BLAST search was 12.5.7 (YASARA Biosciences GmbH, Austria). performed over the Protein Data Bank The docking grid was generated 5.0 Å around all (www.rcsb.org). From the PSI-BLAST search, atoms with periodic boundaries and a total of 25 sodium channel from Caldalkalibacillus docking runs were set up for each docking thermarum was ranked first with total score of experiments. The docked conformations from 25 24.30 and BLAST E-value of 2x10-16 (Table 1). runs were clustered and significant contacting The sodium channel protein from C. thermarum amino acid residues were determined and ranked shares 25% sequence identity with B. mori. The based on the binding energy. The top ranked alignment quality of all the alignments were conformations were visually analysed and detail assessed by using WHAT_CHECK (Hooft et al., residues involved in protein-ligand interaction 1996). By using the sodium channel from C. were determined. thermarum (PDB ID: 4BGN) as the template, the structure of B. mori’s sodium channel was successfully modelled by using YASARA RESULTS AND DISCUSSION Structure Suite 12.10.3 via automated mode. Model quality evaluation was conducted to Modelling of Lepidopteran sodium channel check for the dihedrals angle and protein structure packing as shown in Table 2. The overall Z- Voltage-gated sodium (NaV) channels is one of score of -0.899 was obtained from the model the potential molecular targets in the insect evaluation which fall in the good range category.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Figure 1. Multiple Sequence Alignment (ClustalW) of sodium channel protein between H. armigera, B. mori and H. virescens.

Table 1: PSI-BLAST Result. Sodium channel protein sequence from B. mori was used as the query. Template Total BLAST Alignment Coverage Template ID Resolution Score E-value Score

-16 1 24.30 2x10 103.0 46% 4BGN-A 9.00 Å

-30 2 23.81 2x10 122.0 44% 4MVM-D 3.20 Å

-30 3 23.14 1x10 117.0 44% 4MVQ-D 3.40 Å

-30 4 22.72 2x10 113.0 44% 4MVZ-D 3.30 Å

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 2. Model quality evaluation result from WHAT_CHECK. Check Type Quality Z-score Comment

Dihedral 0.408 Optimal Packing 1-D -0.749 Good Packing 3-D -1.431 Satisfactory Overall -0.899 Good

The predicted model of the sodium (Figure 3) as the reference structure, the B. mori channel was in monomeric form (Figure 2A), tetrameric complex structure was assembled however, in nature, the sodium channel is a (Figure 2B) from the predicted monomeric complex protein which the pore region was model. The newly assembled tetrameric complex formed when the protein assembled in the form of the B. mori’s channel protein was only tetrameric form. Since our purpose of modelling consisted pore-forming domain and all the the channel protein is to study the interaction voltage gating sensor domains were removed. As between the neurotoxin and the channel protein, shown in Figure 2A, the pore-forming domain functional complex structure of the channel consist of helical S5 & S6 and together with all protein must be obtained. By using the pore the tetramer subunit, they formed a funnel-like complex structure of the sodium channel from shaped pore inside the cytoplasmic membrane Magnetococcus marinus MC-1 (PDB ID: 4F4L) (Figure 2B).

Figure 2. Predicted model of B. mori sodium channel. (A) Sodium channel in monomeric form. (B) Sodium channel in tetrameric form.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Figure 3. Sodium channel from M. marinus MC-1 (PDB ID: 4F4L).

Molecular docking of commercial insecticides -6.40 kcal/mol and followed by deltamethrin with binding energy of -6.29 kcal/mol. In order to understand the interaction between Examining the detail interaction between the channel protein and the inhibitors such as cycloprothrin and the pore domain revealed that chemical insecticides at molecular level, global Trp318 of the first monomer and Trp328 of the docking experiments were conducted. Flexible second monomer that formed the pore interact docking for all the ligands were carried out on with the two benzene rings of the non-polar the active site of the open pore of the sodium region of cycloprothrin via hydrophobic channel (Figure 4). Based on the procedure interaction (Figure 5). On the contrary, the explained in the experimental section, the hydroxyl group of Tyr335 formed interaction predicted binding energy of these insecticides with the carbonyl group of cycloprothrin and into the pore domain are listed in Table 3. The Tyr329 formed hydrogen bonding with the binding energy is obtained by calculating the oxygen atom from the polar region of energy at infinite distance (the unbound state) cycloprothrin (Figure 5). Our docking studies and subtracting the energy of the entire reveals that all chemical insecticides adapt simulation cell at the bound state. Based on our similar interaction between the pore domain of docking results (Table 3), cycloprothrin and the sodium channel and acts as pore blocker cypermethrin showed highest binding energy of rather than gating modifier.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Figure 4. Docked structure of commercial chemical insecticides in open conformation of B. mori sodium channel. The pore forming domain (helical S5 & S6) is shown in grey, whereas, the ligands (insecticides) are shown in cyan.

Table 3: Binding energy between commercial insecticides (ligand) and sodium channel protein (receptor).

Compound Bind.energy[kcal/mol]

Cycloprothrin -6.4000

Cypermethrin -6.4000

Deltamethrin -6.2900 Fenvalerate -6.0600 Permethrin -5.8700 Fluvalinate -5.0300

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Figure 5. Cycloprothrin interaction with the sodium channel pore. Deltamethrin interact with the hydrophobic residues of helical S5 and S6 which consist of tyrosine (yellow) and tryptophan (red).

Molecular docking of neurotoxin peptides mining against ArachnoServer. ArachnoServer Insect-selective neurotoxin peptides have huge (http://www.arachnoserver.org) is a manually potential to be developed as bio-pesticide curated database containing information on the especially from spiders. Till now, more than 800 sequence, three-dimensional structure, and neurotoxin peptides were identified and curated biological activity of protein toxins derived from all over the world. However not all the spider venom developed by scientist from the neurotoxin derived from spiders are insect- University of Queensland. Based on few criteria selective, orally active and exhibit high toxicity (high toxicity, insect-selective and solved toward lepidopterans. In order to search the ideal protein structure) we have listed down 10 reference neurotoxin peptides for our protein neurotoxin peptides from various spider species design experiment, we had conducted data in Table 4.

Table 4. General characteristics of neurotoxin peptides from various spider species. Toxin Name Origin Toxicity Length Size PDB ID

(LD50) (a.a) (kDa) ApsIII Apomastus schlingeri 133 pmol/g 38 3.8 2m36

OAIP-1 Selenotypus plumipes 104 pmol/g 33 3.7 2ll1

U3-cyrtautoxin-As1a Apomastus schlingeri 396.0 pmol/g 32 3.5 1wqb

ω-hexatoxin-Hv1a Hadronyche versuta 279.0 nM 37 4.0 1axh

ω-hexatoxin-Hv1f Hadronyche versuta 1384.0 pmol/g 37 3.9 1zjv

ω-hexatoxin-Hv2a Hadronyche versuta 160.0 pmol/g 45 4.4 1hp3

κ-theraphotoxin-Ps1a Paraphysa scrofa 28.0 nM 29 3.5 1v7f

π-theraphotoxin-Pc1a Psalmopoeus cambridgei 0.7 nM 40 4.6 1lmm

μ-theraphotoxin-Hhn1a Haplopelma hainanum 42.3 nM 35 3.9 1zju

μ-theraphotoxin-Hhn2a Haplopelma hainanum 1.1 nM 33 3.6 2jtb

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Among the listed peptides, OAIP-1 blocked the inner groove of the channel. By neurotoxin from Selenotypus plumipes showed examining the molecular surface of the protein- highest toxicity with LD50 of 104 pmol/g and protein complex (Figure 7), the interaction that followed by APSIII neurotoxin from Apomastus majorly involve between the pore and the schlingeri with LD50 of 133 pmol/g. Most of the neurotoxin peptides are hydrophobic interaction neurotoxin peptides are in the range 29-45 amino since the inner pore predominantly comprised acids in length and approximately ~4 kDa in hydrophobic amino acid residues. The red size. molecular surface as shown in Figure 7 indicates negatively charged electrostatic surface potential Protein-protein docking experiments which also associated with hydrophobicity of the against sodium channel were conducted by protein domain. In contrast, the blue molecular selecting 10 potent neurotoxin peptides (Table 4) surface associated with hydrophilicity and as the ligand by retrieving their structures from indicates positively charged electrostatic surface the Protein Data Bank (http://www.rcsb.org). potential. The electrostatic surface potential was Flexible docking experiments were performed in calculated by using Particle Mesh Ewald (PME) a total of 25 runs per peptide target and the algorithm along with AMBER96 force field. docked peptide-protein complexes were Since, sodium channel is a member of summarized in Figure 6. The binding energies membrane bound protein, only the exposed for each protein-protein docking were tabulated surface is hydrophilic, whereas, the majority of (Table 5) and the complex structure were the surface are hydrophobic as they are enclosed visualized. Analyzing each of the docked inside lipid membrane. Therefore, hydrophobic structures, we found that all of the neurotoxin interaction account most of the force involve in peptides act as the pore blocker in which they binding interaction between the neurotoxin are blocking the opening of the pore (Figure 7) peptides and the channel’s pore. as opposed with chemical insecticides that

Figure 6. Docked structure of neurotoxin peptides from spiders in open conformation of B. mori sodium channel. The pore forming domain (helical S5 & S6) is shown in grey, whereas, the ligands (neurotoxin peptides) are shown in yellow surrounded with molecular surface.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 5: Binding energy between neurotoxin peptides (ligand) and sodium channel protein (receptor). Toxin Name Bind.energy[kcal/mol]

ApsIII -0.8600 OAIP-1 -4.6200 U3-cyrtautoxin-As1a -4.2200 ω-hexatoxin-Hv1a -7.6100 ω-hexatoxin-Hv1f -4.2800 ω-hexatoxin-Hv2a -3.7600 κ-theraphotoxin-Ps1a -3.6700 π-theraphotoxin-Pc1a -7.8500 μ-theraphotoxin-Hhn1a -6.1700 μ-theraphotoxin-Hhn2a -2.2200

Figure 7. Structural representation of protein-protein complex conformation using electrostatic surface potential. OAIP-1 neurotoxin (yellow) was docked against the pore domain of sodium channel. (A) Side- view of the protein complex. (B) Top-view of the protein complex. Positively charged electrostatic surface potential is colored in blue, whereas, the negatively charged surface is colored in red.

CONCLUSIONS enhance effectiveness can be design and could solve the insecticide resistance issue. The rapid In general, modelling Lepidopteran sodium increase in the demand of organic product channel protein via protein structure prediction especially in the western market had accelerated provide us structural insight that give us some the development of nature-based bio-pesticides. glance entailing the machinery of this transport Based on our findings, neurotoxin peptides from protein. As an important insecticide target, spider have a huge potential to be developed as understanding the mode of action or inhibitory an alternative for the current chemical pesticide mechanism of the commercial chemical in the market as concerns over human health and insecticides at molecular level would be useful the environment are given more attention over for the industry since new chemical leads with these days.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

ACKNOWLEDGEMENT Maggio, F., Sollod, B.L., Tedford, H.W., Herzig, V. and King, G.F. (2010) Spider toxins The authors gratefully acknowledge the financial and their potential for insect support received from Malaysian Cocoa Board control, in Insect Pharmacology: in conducting the research. Channels, Receptors, Toxins and Enzymes (Gilbert LI and Gill SS eds) pp 101–123, Academic Press, London. REFERENCES Morris, G. M., Goodsell, D. S., Halliday, R. S., Bass, C. and Field, L. M. (2011). Gene Huey, R., Hart, W. E., Belew, R. K. and amplification and insecticide resistance. Olson, A. J. (1998). Automated docking Pest Manag Sci. 67: 886-890. using a Lamarckian genetic algorithm and and empirical binding free energy Bates, S. L, Zhao, J. Z., Roush, R. T. and function. J. Comput. Chem.19: 1639- Shelton, A. M. (2005). Insect resistance 1662. management in GM crops: past, present and future. Nat. Biotechnol. 23: Pallaghy, P. K., Nielsen, K. J., Craik, D. J. and 57–62. Norton, R. S. (1994). A common structural motif incorporating a Billen, B., Bosmans, F. and Tytgat, J. (2008). cystine knot and a triple-stranded b- Animal peptides targeting voltage- sheet in toxic and inhibitory activated sodium channels. Curr polypeptides. Protein Sci. 3: 1833– Pharm. Des. 14: 2492–2502. 1839.

Duan, Y., Wu, C., Chowdhury, S., Lee, M. C., Saez, N. J., Senff, S., Jensen, J. E., Er, S. Y., Xiong, G., Zhang, W., Yang, R., Herzig, V., Rash, L. D. and King, G. F. Cieplak, P., Luo, R. and Lee, T. (2010). Spider-venom peptides as (2003). A point-charge force field for therapeutics. Toxins. 2: 2851–2871. molecular Mechanics Simulations of Proteins. J. Comput. Chem., 24:1999- Skinner, W. S., Dennis, P. A., Li, J. P. and 2012. Quistad, G. B. (1992). Identification of insecticidal peptides from venom of the Hooft, R.W.W., Vriend, G., Sander, C. and trap-door spider, Aptostichus schlingeri Abola, E.E. (1996). Errors in protein (Ctenizidae). Toxicon. 30: 1043–1050. structures. Nature. 381: 272-272. Wood, D. L., Miljenovic, T., Cai, S., Raven, R. King, G. F. and Hardy, M. C. (2013). Spider- J., Kaas, Q., Escoubas, P., et al. (2009). venom peptides: structure, Arachno- Server: a database of pharmacology, and potential for control protein toxins from spiders. BMC of insect pests. Annu. Rev. Entomol. 58: Genomics. 10: 375. 475-496.

King, G. F., Escoubas, P. and Nicholson, G. M. (2008). Peptide toxins that selectively target insect NaV and CaV channels. Channels. 2: 100–116.

Lai, T. C. and Su, J. Y. (2011). Assessment of resistance risk in Spodoptera exigua (Hubner) (Lepidoptera: Noctuidae) to chlorantraniliprole. Pest. Manag. Sci. 67: 1468-1472.

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

TOXICITY STUDIES OF Bacillus thuringiensis TO COCOA POD BORER LARVA AND ADULT MOTH

Tan, C.L.1, Navies, M.2 and Anisah, S.1 1Malaysian Cocoa Board, Cocoa Biotechnology Research Centre, Kota Kinabalu Industrial Park, 88460 Kota Kinabalu, Sabah, Malaysia 2Malaysian Cocoa Board, Cocoa Research and Development Centre, Miles 10, Apas Road, P.O. Box 60237, 91012 Tawau, Sabah, Malaysia

Malaysian Cocoa Journal 9(1): 97-102 (2016) ABSTRACT - Bacillus thuringiensis (Bt) is a soil bacterium that forms spores during the stationary phase of its growth cycle. The spores contain crystals, predominantly comprising one or more Crystal proteins (Cry) (also known as delta-endotoxins) that have potent and specific insecticidal activity. Different strains of Bt produce different types of toxin, each of which affects a narrow taxonomic group of insects. Therefore, Bt toxins have been used as topical pesticides to protect crops, and more recently the proteins have been expressed in transgenic plants to confer inherent pest resistance. In this study, two Bt biopesticides available (Florbac® and Vigor®) were evaluated against cocoa pod borer (CPB) larva and adult moth. For biopesticide Florbac®, after 3 days of Bt toxin treatment, all the CPB moths in treatments of 3 ppt and above were killed whereas all the CPB moths survive in the Control treatment, 50% of the CPB moths survive in the 1 ppt Bt toxin treatment and 25% of the CPB moths survive in the 2 ppt Bt toxin treatment. This indicates that the Florbac Bt endotoxin is effective in total killing CPB moths at 3 ppt and above. As for the larva, approximately 4 ppt concentration of the biopesticide is needed for effective killing of 50% of the population. For Florbac, the LC50 for larva is 4.2 ppt whereas for moth is 0.80 ppt. For Vigor, the LC50 for moth is 1.3.

Key words: Bacillus thuringiensis, Cocoa pod borer, Biopesticide Florbac®

INTRODUCTION there is no record on how CPB larva and moths respond toward B.t.a. exposure. Bacillus thuringiensis (Bt) is a soil bacterium that forms spores during the stationary phase of The mode of action of Cry proteins has its growth cycle. The spores contain crystals, been partially elucidated. After oral uptake, the predominantly comprising one or more Crystal crystals are dissolved in alkaline environment of proteins (Cry) (also known as d-endotoxins) that the larval midgut and the protoxin are processed have potent and specific insecticidal activity. by midgut proteases to 65 kD protease-resistant Different strains of Bt produce different types of toxic fragment. The toxic fragment binds to the toxin, each of which affects a narrow taxonomic receptor on the epithelial cells of the midgut and group of insects. Therefore, Bt toxins have been penentrates the cell membranes (Bosch et al., used as topical pesticides to protect crops, and 1994). This eventually leads to lysis of these more recently the proteins have been expressed cell and death of the larvae / insects. in transgenic plants to confer inherent pest resistance. MATERIALS AND METHODS Florbac (tradename) with active ingredient of Bacillus thuringiensis subsp. Two Bt biopesticides available (Florbac® and aizawai strain NB200 (B.t.a) is a biopesticide Vigor®) were evaluated against cocoa pod borer ® specific for controlling insect pest from (CPB) larva and adult moth. Florbac lepidopteran group which infesting vegetable. biopesticide were obtained from National ® Larvae of lepidopteran pest need to consume the Pingtung University, Taiwan whereas Vigor B.t.a., in order to get poisoned with bt toxin biopesticide was a gift from Vigor AgriBiotech released by the B.t.a. This product is effective to Solutions Sdn. Bhd., Johor, Malaysia. Florbac® control lepidopteran pest on vegetable. However, biopesticide consist of Bacillus thuringiensis subsp. aizawai strain NB200 Lepidopteran

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

active toxin (10%) and other ingredients (90%). (concentration + 1). A graph is plotted for the Vigor® biopesticide contained mixed strains of probit units against log of the concentrations. A Bacillus thuringiensis (Bt). The actual Bt strains regression line is drawn and equivalent equation used and ingredients is not known to the authors is formulated. The lethal concentration required as it is a trade secret. to kill 50% of the population (LC50) is determined by calculating for a probit value of The larva and adult moth were collected 5.00 and then taking the inverse log of the from cocoa fields in Tawau, Madai Baturong concentration it is associated with. and Tuaran, Sabah. The experiments were carried out over a period of one year. Five CPB moth different concentrations of the Bt biopesticides Five concentrations of Florbac B.t.a. NB200 by were used and the experiments were repeated Valent Bio Science Corporation USA were trice (for CPB moth). prepared, 1 ppt, 2 ppt, 3 ppt, 4 ppt and 5 ppt. Control was tested with distilled water. The CPB larva B.t.a toxin was soaked in cotton wool for Five concentrations of Florbac B.t.a. NB200 by feeding by the CPB moths kept in aerated Valent Bio Science Corporation USA were containers. prepared, 0.25 ppt, 0.50 ppt, 1.00 ppt, 1.50 ppt and 2.00 ppt. Control was distilled water. Five Four CPB moths were exposed to each CPB larva were exposed to each treatments and treatments and control (distilled water). Daily control (distilled water). Acute CPB larva observations of the survivorship of the CPB response was observed after 24 hours of moth is recorded. Any moth without any exposure. The experiment consisted of three response when disturbed with a hair brush was replicates per treatment. considered dead due to B.t.a. toxin. Percent mortality was calculated as above based on Any larvae without response when number of surviving moths after Day 4 (for disturbed with a hair brush will be considered Florbac) and Day 6 (for Vigor). Probit analysis dead due to B.t.a. toxin. Percent mortality was was used to define the CPB larva LC50 due to calculated using Abbott’s formula as follow; B.t.a toxin.

% mortality = RESULT where; x = % survivorship in the control group CPB larva Table 1 showed the effect of various y = % survivorship in the experiment group concentrations of Florbac B.t.a. NB200 on the

Probit analysis (Finney, 1952) was mortality of CPB larva. The corresponding carried out manually by converting percentage of percentage of mortality were also converted into Probits units using Finney’s Table (Finney and mortality into probit units using Finney’s Table Stevens, 1948). (Finney and Stevens, 1948). Subsequently, the concentrations used were changed to Log10

Table 1. CPB larva percent mortality and the probits value of various concentration Concentration % Log10 (X+1) Probits of Mortality (X) ppt Mortality

0.25 0.0969 33 4.56 0.50 0.1761 27 4.39 1.00 0.3010 33 4.56 1.50 0.3979 34 4.59 2.00 0.4771 47 4.92

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Figure 1 is a graph of Probits of with probit of mortality value Y at 5; LC50 = mortality against Log10 (concentration + 1) of 4.2095 ppt. Florbac B.t.a. NB200. A best fit line is obtained using Excel software and an equivalent equation Recommended concentration by the is formulated. Lethal concentration (LC50) is manufacturer is 1 ppt. However in our studies, derived by solving the equation whereby the in order to kill 50% of CPB larva, we need Probits of mortality value is 5. Linear regression approximately 4 ppt concentration of the function is Y = 0.9275X + 4.3352 (Figure 1); biopesticide. LC50 value is obtained by solving the function

Figure 1. Probits of mortality over concentration (transformed to log10 [concentration + 1]) with linear regression.

CPB moth mortality of CPB moth. The corresponding percentage of mortality were also converted into Florbac biopesticide Probits units using Finney’s Table (Finney and Table 2 showed the effect of various Stevens, 1948). concentrations of Florbac B.t.a. NB200 on the

Table 2. CPB moth percent mortality and the probits value on various concentrations of Florbac. Concentration Probits of (X) ppt of Log10 (X+1) % Mortality Mortality Florbac 0 0 6.7 3.52 1 0.301 70 5.52 2 0.4771 85 6.04 3 0.6021 93.3 6.48 4 0.699 100 8.09 5 0.7782 100 8.09

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Figure 2 is a graph of Probits of After 3 days of Bt toxin treatment, all the mortality against Log10 (concentration + 1) of CPB moths in Bt toxin treatments 3 ppt and Florbac B.t.a. NB200. A best fit line is obtained above were killed whereas all the CPB moths using Excel software and an equivalent equation survive in the Control treatment, 50% of the is formulated. Lethal concentration (LC50) is CPB moths survive in the 1 ppt Bt toxin derived by solving the equation whereby the treatment and 25% of the CPB moths survive in Probits of mortality value is 5. Linear regression the 2 ppt Bt toxin treatment (Table 2). This function is Y = 5.8488x + 3.5046 (Figure 2; indicates that the Florbac B.t.a. NB200 by LC50 value is obtained by solving the function Valent Bio Science Corporation USA is with probit of mortality value Y at 5; LC50 = effective in killing CPB moths at 3 ppt and 0.8017 ppt. above.

Figure 2. Probits of mortality over concentration (transformed to log10 [concentration + 1]) with linear regression.

Vigor biopesticide percentage of mortality were also converted into Table 3 showed the effect of various Probits units using Finney’s Table (Finney and concentrations of Vigor® biopesticide on the Stevens, 1948). mortality of CPB moth. The corresponding

Table 3. CPB moth percent mortality and the probits value on various concentrations of Vigor. Concentration (X) Probits of Log10 (X+1) % Mortality ppt of Vigor® Mortality 0 0 13.3 3.87 1 0.301 53.3 5.08 2 0.4771 60 5.25 3 0.6021 66.7 5.44 4 0.699 86.7 6.13 5 0.7782 86.7 6.13

100

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Figure 3 is a graph of Probits of Probits of mortality value is 5. Linear regression mortality against Log10 (concentration + 1) of function is Y = 2.8315x + 3.9682 (Figure Vigor® biopesticide. A best fit line is obtained 3);LC50 value is obtained by solving the using Excel software and an equivalent equation function with probit of mortality value Y at 5; is formulated. Lethal concentration (LC50) is LC50 = 1.314 ppt. derived by solving the equation whereby the

Figure 3. Probits of mortality over concentration (transformed to log10 [concentration + 1]) with linear regression.

DISCUSSIONS other trade name basing on the same Bt subsp. aizawai strain are Agree®, Design® and Laboratory testing of the Bt biopesticide Xentari®. The target insect is manly Lepidoptera ‘Florbac’ indicates that it is effective in killing (Sanahuja et al., 2011). Topical Bt sprays are CPB larva and moths. Although the advantageous in terms of their safety, specificity manufacturer’s recommended dosage is 1 ppt, and potency compared to chemical sprays, and we found that a higher dosage is required are also biodegradable. Bt biopesticide is also (approximate 4 ppt to kill 50% of CPB larva and one of the safest microbial products known. It at least 3 ppt or above for total killing of CPB has no known hazard to human health moth after 3 days). Santosa et al. (2004) in their (Mendelsohn et al., 2003) and environmentally work found that all the Bt toxins tested showed safe. However, Bt is only effective when present at least 40% mortality rate on CPB larva. The on the plant organs on which insects feed. increased dosage from the recommended could Usually Bt is applied when early instar larvae are be due to the product shelf life has expired present, because older larvae are more tolerant. (expiration date: 991011). Stage of growth of Bt sprays persist for only a few days because UV CPB larva used for testing also affect the results light, weather, the chemical environment of the obtained. Bioassays of neonate or 1st-instar leaf surface and the presence of proteinases larvae would have been preferable, because contribute to the degradation of Cry proteins. younger instars are generally more sensitive to Cry proteins (Santoso et al., 2004). CPB larva spend most of their life cycle inside the cocoa pods, making topical spray of Bt Florbac® is one of the many Bt difficult to reach the insect. Alternatively, biopesticide available in the market. Several expression of the Bt protein can be done in

101

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

planta, that is by inserting the Bt gene into the management. Nature Biotechnology genome of cocoa plant such as in rice 12: 915 – 918. (Chakraborty et al., 2016; Jiao et al., 2016). The Bt protein can also be expressed in targeted Chakraborty, M., Reddy, P.S., Mustafa, G., organs such as in cocoa pod shells and pulp to Rajesh, G., Narasu, V.M., efficient killing of the CPB larva. This can be Udayasuriyan, V. and Rana, D. (2016). done by fusion of a tissue-specific promoter to Transgenic rice expressing the the gene as in other crop plants (Pradhan et al., cry2AX1 gene confers resistance to 2016; Albuquerque et al., 2015). This would multiple lepidopteran pests. Transgenic also result in less stress for the cocoa plant as the Res. DOI 10.1007/s11248-016-9954-4 Bt protein is only localised in specific organs and not on the whole plant. Finney, D. J., Ed. (1952). Probit Analysis. Cambridge, England, Cambridge University Press. CONCLUSION Finney, D. J. and Stevens, W. L. (1948). A table In this study, we have found that the acute LC50 for the calculation of working probits Florbac Bt toxin against CPB larva is 4.2 ppt and weights in probit analysis. whereas for moth the chronic LC50 is 0.8 ppt. Biometrika 35(1-2): 191-201. For Vigor® Bt, the LC50 against CPB moth is 1.3 ppt. Even though both are of Bt origin, the Jiao, Y., Yang, Y., Meissle, M., Peng, Y. and Li, killing ability varies with different strains of Bt. Y. (2016). Comparison of susceptibility of Chilo suppressalis and Bombyx mori to five Bacillus thuringiensis proteins. ACKNOWLEDGEMENT J. Invertebr Pathol. 136: 95-99.

We would like to thank the Chairman of Mendelsohn, M., Kough, J., Vaituzis, Z., Malaysian Cocoa Board, Datuk Dr. Marcus Matthews, K. (2003). Are Bt crops Mojingoh for linking us with National Pingtung safe? Nature Biotechnology 21(9): University, Taiwan and thererby obtaining the 1003 – 1009. Florbac Bt sample and also Vigor AgriBiotech Solutions Sdn. Bhd. for providing the Vigor® Bt Pradhan, S., Chakraborty, A., Sikdar, N., sampel for this study. Chakraborty, S., Bhattacharyya, J., Mitra, J., Manna, A., Dutta Gupta, S. and Sen, S.K. (2016). Marker-free REFERENCES transgenic rice expressing the vegetative insecticidal protein (Vip) of Albuquerque, É.V., Bezerra, C.A., Romero, J.V., Bacillus thuringiensis shows broad Valencia, J.W., Valencia-Jiménez, A., insecticidal properties. Planta. Pimenta, L.M., Barbosa, A.E., Silva, doi:10.1007/s00425-016-2535-1. M.C., Meneguim, A.M., Sá M.E., Engler, G., de Almeida-Engler, J., Santoso, D., Chaidamsari, T., Wiryadiputra, S. Fernandez, D. and Grossi-de-Sá, M.F. and Ruud A de Maagd (2004). Activity (2015). Seed-specific stable expression of Bacillus thuringiensis toxins against of the α-AI1 inhibitor in coffee grains cocoa pod borer larvae. Pest Manag. and the in vivo implications for the Sci. 60: 735–738. development of the coffee berry borer. Trop Plant Biol. 8: 98-107. Sanahuja, G., Banakar, R., Twyman, R.M., Capell, T. and Christou, P. (2011). Bosch, D., Schipper, B., Hilde, van der Kleij, Bacillus thuringiensis: a century of Ruud, A. de Maagd and Willem, J. research, development and commercial Stiekema (1994). Recombinant Bacillus applications. Plant Biotechnology J. 9: thuringiensis crystal proteins with new 283–300. properties: possibilities for resistance

102

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

CAFFEINE TOXICITY TEST ON COCOA POD BORER ADULT MOTH

Anisah, S. Malaysian Cocoa Board, Cocoa Biotechnology Research Centre, Kota Kinabalu Industrial Park, 88460 Kota Kinabalu, Sabah, Malaysia

Malaysian Cocoa Journal 9(1): 103-107 (2016) ABSTRACT – This study was performed to determine the toxicity effect of caffeine against cocoa pod borer (CPB) Conopomorpha cramerella (Snellen), the most devastating insect pest to cocoa. Five concentrations of caffeine were prepared, 0.2 mg ml-1, 0.4 mg ml-1, 0.8 mg ml-1, 1.2 mg ml-1. Control was reverse osmosis water. Two days old CPB moths were collected, separated according to gender then introduced into cages, and exposed to each treatments and control. Each cage contained five CPB moths and the experiments were consisted of three replicates per treatment. Observations were made every 24 hours for five days. Probit analysis was used to define the CPB moth LC50 due to caffeine. In order to kill 50% of the CPB moth, 0.7245 mg ml-1 and 0.7288 mg ml-1 of caffeine were needed for females and males CPB moth, respectively. These data revealed that caffeine has detrimental effects at particular concentrations on CPB insects.

Key words: Cocoa Pod Borer, Caffeine, Alkaloid, Toxicity test, Moth, Probit analysis.

INTRODUCTION (Senanayake and Wijesekera, 1971). The content of caffeine is negligible until the cocoa pods Caffeine (1, 3, 7-trimethylxanthine) is a purine reached three-month when the kernel is alkaloids that is produced by plants including mucilaginous. However, the caffeine content coffee, tea, and cocoa as secondary metabolites increased markedly after four-month. (Zulak et al., 2006). Caffeine has been reported to play roles in plant chemical defense, acting as a Although various effects of caffeine have repellent (Hollingsworth et al., 2002), natural been reported in other insects, yet in cocoa pod pesticide (Nathanson, 1984), and allelopathic borer (CPB) insect (Conopomorpha cramerella) agent (Waller, 1989). Despite leaves cardiac and no study has been reported so far. This is a diuretic effects, caffeine also reported to serve as preliminary study on the effect of caffeine on CPB stimulant, and one that used widely as food insects. additive and added in drugs ingredients (Chou and Benowitz, 1994). MATERIALS AND METHODS In insect particularly, various effects of caffeine have been reported including inhibition of Insect feeding and sleep (Wu et al., 2009), increase heart The adult moths of cocoa pod borer (CPB) insect rate (Zornik et.al., 1999), mutagenic and mitotic were obtained from Pusat Penyelidikan dan action, olafactory and visual associative learning Pembangunan Koko, Madai. Harvested cocoa pods (Si et al., 2005), and mortality and oviposition heaps were covered with dry cocoa leaves to trap (Laranja et al., 2006). the emerging pre-pupae. The pupae on leaves were cut out and placed in a container or box before Caffeine is found to be present in young they were sent out via BUS to the laboratory at leaves, flower buds, and fruits. It is believed that Centre for Cocoa Biotechnology Research located the presence of caffeine in young leaves, fruits and at Industrial Park Kota Kinabalu. The newly flower buds acts to protect soft tissues from emerged moths were separated accordingly based predators such as insect larvae and beetles on their gender and fed with 5% honey solutions. (Hwavitharanage, 2000). In cocoa (Theobroma cacao), caffeine is detected in cotyledon and shell

103

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Insecticidal activity After 48 hours of exposure, the percent In order to test the toxicity of caffeine on the adult mortality of the female control group was still zero moth CPB, five concentrations of caffeine were while in the male control, the mortality already prepared, 0.2 mg ml-1, 0.4 mg ml-1, 0.8 mg ml-1, reached 21 percent. In the female group, mortality 1.2 mg ml-1. Control was reverse osmosis water. was less than 10 percent even after 72 hours of Two-day-old adult males and females CPB were exposure. This result suggested that the females used in this study. Experiments of insecticidal CPB moth is long living compared to the male activity against adult moth CPB were conducted CPB moth. according to gender but performed simultaneously. Each treatments and control cage are introduced This result too corresponded with the with five CPB moths and the experiments results obtained from the experimental moths consisted of three replicates per treatment. group. The male CPB moths are found to be more sensitive to caffeine than the female CPB moth. Each of the treatments and control Percent mortality of the males CPB moth (Table 2) solutions were prepared in 25 ml conical flask was always higher than the percent mortality of the where the total volume of each was 20 ml. A 12 ml females CPB moth (Table 1) at all concentrations test tube wrapped with lint cloth inserted into the of caffeine tested. However, the results were not conical flask to allow the solution to migrate statistically significant (F=0.421, df=1, p=0.563) . across the lint cloth. The moth feeds through the Furthermore, Probit analysis showed that LC50 lint cloth. All the insecticidal activity experiments value of caffeine against CPB moth males and were conducted at room conditions. Observations females are very close – considered the same. The -1 were made every 24 hours for five days. Any moth LC50 for males was 0.7288 mg ml and 0.7245 for laid on the surface of the cage’s bottom will be females. The LC50 values are obtained by solving considered dead due to caffeine toxicity. the Linear regression function with probit of mortality value Y at 5 (Figure 1 and Figure 2). Statistical analysis The percent (%) mortality was calculated using Caffeine has been reported to be toxic in Abbott’s formula while the lethal concentrations many other studies on insect larvae. We assumed (LC50) value after 24 hours exposure was that caffeine will be toxic to CPB larvae as well as determined by Probit analysis using SPSS the study presented here provides clear software. information that caffeine is toxic to the CPB moth at particular concentrations.

RESULTS AND DISCUSSIONS In the caffeine toxicity study of mosquito larvae (Laranja et al., 2003), caffeine blocks the Significant difference (p<0.05) was observed in development of Aedes aegypti (Diptera, Culicidae) percent mortality of CPB moths with the in the larval stage, consequently inhibiting the increasing concentrations of caffeine (F=11.918, production of adults. Laranja et al. (2003) reported df=3, P=0.0357) (Table 1 and Table 2). The higher that the concentrations of caffeine at 2.0 mg ml-1 the concentration used, the shorter the time to produced the strongest effect, killing 100% of the reach 100 percent mortality. At 1.2 mg ml-1 of larvae in the L1-sub-phase. At 1.0 mg ml-1, the caffeine, it takes 48 hours and 72 hours to resulted development of larva was blocked at instar two- 100 percent mortality for males and females, stage, while at 0.5 mg ml-1, the larvae died at instar respectively. At lower concentrations (0.4 mg ml-1 three or four-stage. and 0.8 mg ml-1), it did not resulted in 100 percent mortality for females even until 96 hours of In CPB insect, 1.2 mg ml-1 of caffeine exposure. The same result for males at 0.4 mg ml- caused 100 percent mortality to both CPB males 1, except the concentrations at 0.8 mg ml-1, where and females moth after 48 hours and 72 hours of the percent mortality was 100 percent after 72 exposure respectively. Although the 100 percent hours of exposure. killing effect needs more than 1 mg ml-1 of

104

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

caffeine, but to make half of the population mortal concentrations of caffeine and LC50 value is it needs lower than 1mg ml-1 of caffeine where the needed for CPB larvae as the larval stages are -1 LC50 for males was 0.7288 mg ml and 0.7245 for more fragile and sensitive compared to adult moth females. This result presumes that lower stage.

Table 1. Mean percent (%) mortality of the females CPB moths after exposure to caffeine. Treatment* Mean % mortality ( Mean ± SEM) Caffeine N** (mg ml-1) 0 hour 24 hours 48 hours 72 hours 96 hours Control 5 0.00 ± 0.00 0.00 ± 0.00 0.00 ± 0.00 6.67 ± 6.67 6.67 ± 6.67 (RO water) 0.4 mg/ml 5 0.00 ± 0.00 26.70 ± 17.64 66.67 ± 13.33 64.29 ± 13.33 71.43 ± 17.64 0.8 mg/ml 5 0.00 ± 0.00 46.70 ± 17.64 60.00 ± 11.55 71.43 ± 6.67 80.00 ± 0.00 1.2 mg/ml 5 0.00 ± 0.00 73.33 ± 13.33 93.33 ± 6.67 100.00 ± 0.00 100.00 ± 0.00 *Each treatment had three replications **Total numbers of females CPB moths per treatment

Table 2: Mean percent (%) mortality (mean ± SEM) of males CPB moths after exposure to caffeine. Treatment* Mean % mortality ( Mean ± SEM) Caffeine N** (mg ml-1) 0 hour 24 hours 48 hours 72 hours 96 hours Control 4 0.00 ± 0.00 0.00 ± 0.00 25.00 ± 25.00 6.67 ± 6.67 33.00 ± 33.33 (RO water) 0.4 mg/ml 5 0.00 ± 0.00 8.30 ± 11.55 65.00 ± 12.58 68.33 ± 22.42 83.33 ± 16.67 0.8 mg/ml 5 0.00 ± 0.00 80.00 ± 11.55 93.33 ± 6.67 100.00 ± 0.00 100.00 ± 0.00 1.2 mg/ml 5 0.00 ± 0.00 86.70 ± 13.33 100.00 ± 0.00 100.00 ± 0.00 100.00 ± 0.00 *Each treatment had three replications **Total numbers of males CPB moths per treatment

Probit of mortality

Log10 [concentration + 1]

Figure 1 Probits of mortality for females CPB over concentration (transformed to log10 [concentration + 1] with liner regression).

105

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Probit of mortality

Log10 [concentration + 1]

Figure 2. Probits of mortality for males CPB over concentration (transformed to log10 [concentration + 1] with liner regression).

Cocoa plant is one of among plants that CPB pupae collections, and Navies Maisin for the naturally produce alkaloids metabolite and cocoa statistical analysis assistance. has been reported to produce theobromine and caffeine. Many studies revealed the negative effects of alkaloids towards insect-pest. However REFERENCES to date, no study has been reported on the theobromine and caffeine that associated with Chou, T.M. and Benowitz, N.L. (1994). Caffeine plant chemicals defence functions in cocoa plant. and coffee: effects on health and Transgenic tobacco plants producing caffeine as a cardiovascular disease. Com. Biochem. potential new strategy for insect control has been Physiol. C. Pharmacol. Toxicol, created (Kim et al., 2006). This strategy may be Endocrinol., 109: 173-189. applicable to apply in cocoa insect-pest management in the future. However, further Hewavitharanage, H. et al. (2000) Effect of studies on the potential effects of caffeine against caffeine on shot-hole borer beetle CPB insects and the mechanism on how the (Xyleborus fornicatus) of tea (Camellia caffeine affect the CPB insects must be done to sinensis). Phytochemistry, 51: 35–41. suit how this caffeine can be used to control CPB insects in future. Hollingsworth, R.G., Armstrong, J.W. and Campbell, E. (2002). Caffeine as repellent for slugs and snails. USDA National CONCLUSIONS Wildlife Research Center – Staff Publications, 470: 914-916. This study provides preliminary information on toxicity effect of caffeine against CPB moth. The Laranja, A.T., Manzatto, A.J. and Bicudo, result obtained here may suggest that caffeine can H.E.M.C. (2003). Effects of caffeine and be used as a natural insecticide for CPB control. used coffee grounds on biological features of Aedes aegypti (Diptera, Culicidae) and their possible use in ACKNOWLEDGEMENT alternative control. Genet Mol Biol., 26: 419-429. The authors gratefully acknowledge the financial support by Malaysian Cocoa Board, Mavis Peter Laranja, A.T., Manzato, A.J. and Bicudo, Jaus (Research assistant), Paulus and team for the H.E.M.C. (2006). Caffeine effect on

106

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

mortality and oviposition in successive Pharmacol. Biochem. Behav., 82: 664- generations of Aedes aegypti. Rev Sauda 672. Publica., 40(6): 1112-1117. Waller, G.R. (1989). Biochemical frontiers of Kim, Y.S., Uefuji, H., Ogita, S. and Sano, H. allelopathy. Biol. Plant., 31: 418–447. (2006). Transgenic tobacco plants producing caffeine: a potential new Wu, M.N., Ho, K., Crocker, A., Yue, Z. and Koh, strategy for insect pest control. K. (2009). The effects of caffeine on Transgenic Res., 15: 667-672. sleep in Drosophila require PKA activity, but not the adenosine receptor. J. Nathanson, J.A. (1984). Caffeine and related Neurosci, 29: 11029-11037. methylxanthines: possible naturally occurring pesticides. Science, 226: 184- Zornik, E., Paisley, K. and Nichols, R. (1999). 187. PMID: 6207592. Neural transmitters and a peptide modulate Drosophila heart rate. Peptides, Senanayake, U.M and Wijesekera, O.B. (1971). 20: 45-51. Theobromine and caffeine content of the cocoa bean during its growth. J. Sci.Fd Zulak, K.G., Liscom, D.K., Ashihara, H., Facchini, Agric., 22: 262-263. P.J., 2006. Alkaloids. In: crozier, A., Clifford, M.N., Ashihara, H (Eds.), Plant Si, A., Zhang, S.W. and Maleszka, R. (2005). Secondary Metabolites: Occurrence Effects of caffeine on mitotic index in Structure, and Role in the Human Diet. drosophila prosaltans (Diptera). Blackwell, Oxford, pp. 102-136.

107

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

EVALUATION OF BEST MANAGEMENT PRACTICES FOR COCOA POD BORER IN COCOA ECOSYSTEM

Saripah, B. and Alias, A. Malaysian Cocoa Board, 5-7th floor, Wisma SEDCO, Lorong Plaza Wawasan, Off-Coastal Highway, Locked Bag 211, 88999 Kota Kinabalu, Sabah, Malaysia Email: [email protected], [email protected]

Malaysian Cocoa Journal 9(1): 108-120 (2016) ABSTRACT – Cocoa is subjected to be infested by various numbers of insect pests, and cocoa pod borer is the most serious pest of cocoa in Malaysia. Several control measures have been implemented, and the chemicals have been a major approach if not unilateral, in the management of CPB. Despite extensive use of insecticides, CPB continues to cause an unacceptable level of damage, thus the combination of several control approaches should be sought. The study was commenced at three blocks; Block 18C with complete management practices which includes insecticide application, pruning, fertilization and frequent harvesting; Block 17C was treated with frequent harvesting at intervals of 7-8 days; and Blok 19C was served as control block. The results showed that the mean numbers of CPB eggs were recorded higher in Block 17C compared with Block 18C in all sampling occasions. Block 18C shows the lowest mean number of CPB eggs in both sampling plots, outside and core plots and it was found significantly different (p≤0.05) compared to the other blocks. Frequent spraying with insecticides at the adjacent block (18C) helps in reducing CPB eggs in the control block (Block 19C), although there was no spraying was implemented Block 19C. Combination of complete management practices at Block 18C seems to have some effect on the CPB population at Block 17 and 19C, because all blocks are adjacent to each other. The second phase of the study was concentrated on the effect of heap pods to CPB eggs. The mean number of CPB eggs was significantly different at week 4, as compared to weeks 6, 10 and 12 after heaps of cocoa pods (n=100) were placed in the field. Observation on CPB egg at different range of heap pods shows that no significant difference between range 2.5 m, 5.0 m and 9.0 m. Field observation shows that the number of pupa collected was high at 50 and 75 pods and significantly different compared to heaps of 12 and 25 pods. CPB was managed to bore out until Day-10, with the highest number of pupa recorded at Day-4. Smallholders must be advised to cut infested cocoa pods into small pieces (cut more than 5 pieces) due to lesser number of pupa emergence. Implementation of insecticides with cultural control must be carried out continuously, for suppression of CPB population in the cocoa field.

Key words: Cocoa, Cocoa pod borer, Complete management practices, Conopomorpha cramerella, Insecticide, Theobroma cacao

INTRODUCTION region (Bateman, 2015). The ‘top dozen producers’ over the last five years are Ivory Coast, In the developing countries, small-scale farmers Ghana, Indonesia, Nigeria, Cameroon, Brazil, that dominated 72% of family farms (FAO, 2014) Equador, Dominican Republic, Peru, Papua New are engaged to cash cropping commodities of Guinea, Colombia and Mexico that represent 95% coffee, cocoa and oil palms (Curry et al., 2015). of global cocoa production. Estimated about 80- Cocoa, Theobroma cacao (Linnaeus) (Malvales: 90% of cocoa plantation all over the world was Sterculiaceae) is planted all over the world, and planted in small scale size (World Cocoa world cocoa production in 2013/2014 is recorded Foundation, 2014). For example, in Ghana, 90% of at 4,272 million tonnes. The production was the plantation was contributed by small scale dominated by African region at 72% (3,101 plantation, with a farm size 1 to 5 hectares million tonnes), followed by the Latin America (Duguma et al., 2001). As in African countries, the region at 16% (666,000 tonnes) and 12% or main actors in Malaysian cocoa commodity system 505,000 tonnes contributed by Asia and Oceania were smallholders that represent 94.85% of total

108

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

cocoa planting area in 2014 (MPIC, 2014). over 90% was observed in 2008, where every Indonesia, which currently dominated Australasian single tree only manages to bear an average of less production of cocoa increased their production than one healthy pod (Curry, 2009). Cocoa pod tremendously from 5% to 19% over the 20th borer continues spreading after that, and accounted century (Bateman, 2015). Smallholders play a vital 80% of crop loss in the year 2012. Cocoa pod role where they contribute to boom production of borer was also detected in 2011 in a commercial Indonesian cocoa industry. Smallholders are major cocoa plantation in far north Queensland (RIRDC, players in Sulawesi, that led this island, 2013). However, it was only detected in one contributing to over 60% of Indonesia’s total property and not established more broadly. The production (McMahon et al., 2015). pest was successfully eradicated using an eradication program that included pheromone Due to small scale plantation, cocoa trapping, property quarantine, pod free period, growers are exposed to limited resources, canopy thinning, crop hygiene, destruction of host technology and financial in managing their material and insecticide treatments. Malaysia also plantation (Curry et al., 2015). Crop loss suffered with CPB incursions. One of the key contributed by pests and diseases was estimated up points for vast declining trends of planting areas to 40% in cocoa growing regions of Africa, Asia and cocoa production was a heavy infestation of and the America (World Cocoa Foundation, 2014). CPB all over the country. Cocoa pod borer Planting low-yielding varieties, poor farm contributed to the loss of interest in cocoa, in maintenance practices and the incidence of pest addition to the labor constraints and instability of and diseases contributed to low productivity of the cocoa price (Azhar, 2007). Lower than cocoa per hectare basis in Ghana (Dormon et al., expected productivity (Azhar and Lee, 2004) is a 2004) and Cameroon (Coulibaly, 2002). Even major factor responsible for the declining of the though most of the growers are familiar with these cocoa growing areas, where the production was far problems, they may not do implement adequate from estimated potential yield which is at 2 to 4 approach, attributed by lack of knowledge, labor tonnes/ha. constraints, high cost of inputs and availability of proper control equipment. The infestation of CPB can end up at a very serious level, where this moth is capable to Unfortunately, as faced by most of inflict significant financial loss that may up to 31% producing countries, Indonesia is now facing a of the crop value (Teh et al., 2006). CPB will lay decline in productivity and in bean quality. eggs on the pod which are 10 weeks old, the Indonesian cocoa production was affected by the proportion of first instar larvae penetrating the pod cocoa pod borer (CPB), Conopomorpha surface was up to 80%. Infestation of CPB cramerella (Lepidoptera: Gracillariidae) with the occurred when the larvae feeding on the mucilage, proportion of cocoa beans. placenta and testa of the developing beans that reduced by 15% in this recent year (Bateman, lead to clump and inextricable beans. Heavy 2015). Back to the history, CPB was recorded as infestation which usually occurs with more than 3- the pest of cocoa in the South East Asia since 4 larva feeding inside may stop the pod from 1860’s when firstly reported at Sulawesi, Java developing, hence induce early ripening or ripen (1901-1966), followed by the Philippines (1936), prematurely. Larval entry can be recognized by the Sumatra (1971), Malaysia (1980), Papua New sign of entry holes on the pod surface. Guinea (2006) and Australia (2011). In the year of Approximately only 15% of larva were 2006, CPB was first detected in Papua New unsuccessful in penetrating pre-sclerotic layer of Guinea near Keravat, East New Britain Province cocoa pod. Meanwhile, 75% of larval was (ENBP) and West Sepik Province. Eradication survived at 13 weeks old pods and very few larva program was implemented for several months, survival recorded at the ripe pods (Day, 1989). however the pest re-emerged within the Infested pods will produce clumped beans, and eradication zone just a month after the eradication become unusable, hence, the whole pod must be program ended (Anon, 2010). Crop loss accounted

109

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

discarded when the infestation is severe Spraying of insecticide for pest control is (McMahon et al., 2015). not the only solution for yield increment, but also rely on husbandry practice that influence cocoa As CPB persisted, smallholders have no productivity (Kumi and Daymond, 2015). Low- choice either to adapt strategies in combating this input management should not be practiced to any pest or to suffer with a dramatic loss of cocoa further extent, where adopting a modern control income. The financial cost at heavily infested area approach is an obligatory. Curry et al. (2015) was critical due to yield loss and an increase in the suggested that low-input management methods production cost in order to minimize the were ineffective for managing CPB infestation, as infestation. Most smallholders with their limited in their observation in Papua New Guinea. This resources prone to partially abandon or in a worst pest can be effectively kept under control if the case fully abandon their cocoa plantation, as an smallholders are ready to adopt the best initial response towards severe pest and disease agricultural practices, hence increase the yield to a incidence. Where there is no unilateral technique more desirable level and economically viable was proven successful in managing this pest, investment (Azhar and Lee, 2004). Field combination of recommended agronomic management must emphasis on a complete set of practices, resistant planting materials, fertilizer practices, from cultural practices such as weekly programming and Integrated Pest Management is a pod harvesting, removal and burial of a diseased key to cocoa sustainability (Azhar, 2007). and CPB affected pods, regular pruning to Integrated Pest Management was defines as chemical practices, if needed (Curry et al., 2015). “careful consideration of all available pest control Adoption of best practices by the smallholders will methods and subsequent integration of appropriate incur minimum impact of CPB, and this package measures that discourage the development of pest should be adopted by cocoa growers in order to populations and keep pesticides and other reduce the use of insecticide (Lee, 2006). interventions to levels that are economically justified and reduce or minimize risks to human Owing to the interest of smallholders health and the environment” by the United Nations applying insecticides as a mean for controlling Food and Agricultural Organization (FAO). CPB infestation, the first phase of the study attempted to apply complete management practices In most cases, chemical control is the which include insecticide application, pruning, most preferable technique by the cocoa growers. fertilization and frequent harvesting as a one Insecticide spraying was commenced every two complete package. The second phase of study weeks, even control should be based on cocoa includes observations on the effects of heap pods cropping calendars. Even chemical has been to CPB eggs and the emergence of pupa. The aim considered as environmentally unsound in the of the study was to investigate the infestation of complex cocoa agro-ecosystem as well as CPB at adjacent blocks with different control expensive and limited cost efficacy, farmers approach. The study was undertaken to investigate always believe that the need of pesticides for CPB infestation when weekly pod harvesting was sustaining healthy crop (Bateman, 2015). implemented, and to compare with complete Beginning of 2014 in EU countries, under the management as well as a block with no control Sustainable Use Directive, pesticide should only approach. Assessment was also conducted in take place within the general principles of IPM, buffer areas and compared with trees located in with integration is a must. Establishing an optimal core plot, with regards to determining either mix of control techniques must include cultural control at one’s block will affect CPB infestation methods, genetic methods and clonal selection that in neighboring areas. Effects of heap pods were confers resistance to pests, conservation and/or evaluated at different range, number of pods per manipulation of biological agents (e.g. heap and number of CPB emerged from cut pods. biopesticides and insect predators) and application of chemical pesticides but only on the basis of rational and responsible use.

110

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

MATERIALS AND METHODS approach was implemented at Block 19C where pruning was carried every month, and weekly ripe The experiment was conducted at three mature pod harvesting. The study was commenced from cocoa blocks, at the Cocoa Research and December 2012 to May 2014. Data collection Development Center (CRDC), Malaysian Cocoa consists of CPB eggs, entry and exit holes. Board Hilir Perak, Malaysia (Longitude E.100 M, Numbers of CPB eggs were recorded at 200 52’ 0’, Latitude N3 53’ 42). Complete randomly selected pods per block, at the size of management practices which include insecticide more than 9cm. The number of exit holes was application, pruning, fertilization and frequent counted based on the distinct exit holes harvesting were adopted at Block 18C (Figure 1). (approximately 0.1 to 0.2cm) on the pod surface. Frequent insecticides were applied with an interval Meanwhile the number of entry hole was counted of 7-8 days, shorter interval compared with normal after thin layer of pod surface were shaved, and practice by cocoa farmers, which is 14 days. small penetration holes (<0.05cm) visibly seen. Preliminary data were obtained in December 2012, There were two distinct cropping seasons, which before first insecticide spraying was implemented the former was peak cropping season (December in January 2013. Insecticides were applied every 2012 through February 2013; October 2013 week with knapsack sprayer, and active ingredient through January 2015) and the latter was drought used was deltamethrin, at recommended dosage. cropping season (March through September 2013). Pruning and fertilization were conducted three Collected data were arranged by monthly basis, times per year, approximately at four month according to different treatments and pooled in interval. Harvesting of ripe pods was carried out Excel® program. Data were then analyzed using every week or 7-8 days interval. The same control Analysis of Variance (ANOVA) provided by the was implemented at Block 17C, excluding SAS statistical software (a SAS® system for insecticide spraying. Meanwhile, no control Windows® V8).

Figure 1. Control and management approach at Blocks 17C, 18C and 19C

111

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

The second phase of study was divided selected and divided into four treatments. Each into two sections, which the first was to evaluate treatment consists of 25 pods that cut into two, the effect of heaps of CPB infested pods on CPB. four, six and eight pieces per pod. Number of pupa The study was conducted for the period of four emergence was recorded for 7 days. Collected data months. Infested ripe pods from all treatments were arranged according to different treatments were collected and heaps were performed at three and pooled in Excel® program. Data were then different quantities of cocoa pods. The number of analyzed using SAS statistical software (a SAS® CPB eggs laid in the treated plots was monitored system for Windows® V8). at three different distances from the heaps, 2.5 m, 5.5 m and 9.0 m. Fifteen matures pods from three cocoa trees were examined for the period of 12 RESULTS weeks, and number of eggs deposited on cocoa pods were recorded at bimonthly interval. This For the first phase of data, the mean of CPB eggs study were conducted in five replicates and recorded the highest at block 19C, where no treatments were 10 pods, 25 pods and 50 pods per control approach was undertaken in 18-months of heap. observation (Figure 2). Means of CPB eggs were falling between 0.4 to 1.5 eggs/pod/month. The A second section of the second study was mean of CPB eggs was recorded the lowest at implemented in January 2016. The number of 12, block 18C, and less fluctuated where month of 25, 50 and 75 harvested pods was placed as a heap, November 2013 through January 2014 was among and covered with dry cocoa leaves. Number of the lowest months, which less than 0.2 pupa emerges from each heap was recorded for 7 eggs/pod/month. Block17C recorded dwindling days. Bigger heaps, consists of 150-200 pods were trends, where April 2013 recorded the highest placed in a large container and the number of pupa mean of CPB eggs. Regardless to different emerged from this big heaps were recorded. This treatment, all blocks harbored the lowest mean of study was replicated 10 times in the laboratory. In CPB eggs during peak crop season from other observations, 100 mature cocoa pods were November 2013 through January 2014.

Means followed by the same letters are not significantly different (p ≥ 0.05).

Figure 2. Mean of CPB eggs at Blocks 17C, 18C and 19C

112

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Both mean of CPB entry and exit holes The mean of CPB eggs in outside and shows similar trends where Block 18C recorded core plots, cumulated for three months after the lower entry and exit holes compared with Blocks first spraying date were shown in Table 1. Block 17C and 19C. However, both entry and exit holes 18C shows the lowest mean of CPB eggs in both at Block 18C increased significantly from June outside and core plots and were significantly onwards. This might be due to drought pod period different at p≤0.05 from other two blocks. The and only a few pods were available during lowest CPB eggs was recorded at core plot at sampling occasions. The mean gradually decreased Block 18C. Outside plot shows better results during peak crop season (October to December compared with the core plot in Block 17C. This 2013) for both 17C and 18C. The Highest mean of might due to frequent insecticides spraying at the entry holes recorded at Block 17C, 18C and 19C adjacent block (18C) helps in reducing CPB eggs was 27, 25 and 38 holes/pod/month, respectively in neighboring rows, although there was no (Figure 3). The highest mean of exit holes was spraying was implemented in this block. Overall, recorded in July or September 2013 (Figure 4), continuous insecticides spraying at Block 18C with 13 holes (Block 19C), 7.5 (Block 17C) and slightly affected CPB population at Blocks 17C 7.0 (Block 18C). and 19C.

Means followed by the same letters are not significantly different (p ≥ 0.05).

Figure 3. Mean of CPB entry holes at Blocks 17C, 18C and 19C

Means followed by same letters are not significantly different (p ≥ 0.05).

Figure 4. Mean of CPB exit holes at Blocks 17C, 18C and 19C.

113

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 1. Mean of CPB eggs in outside and core plots

Area Mean of CPB eggs 17 C outside 0.390 a 17 C Core 0.409 a 18 C outside 0.246 b 18 C Core 0.217 b 19 C 0.352 a Means followed by same letters are not significantly different (p ≥ 0.05). The second phase of the study was pupation period, and deposit their eggs on suitable concentrated on the effect of heap pods on CPB cocoa pods. Observation on CPB egg at different eggs. The mean number of CPB eggs observed on range of heap pods (Figure 6), shows that no cocoa pods was significantly different (p ≥ 0.05) at significant difference between range 2.5 m (0.400), week 4 (0.609a) as compared to week 2 (0.341b), 5.0 m (0.494) and 9.0 m (0.412). Different ranges 6 (0.429b), 10 (0.347b) and 12 (0.368 b) after of heaps to cocoa trees did not have a significant heaps of cocoa pods (n=100) were placed in the effect on the mean number of CPB eggs. The trend field (Figure 5). This might due to life stages of was similar to the density of pods per heaps CPB itself, where it took approximately six to (Figure 7). However, number of CPB eggs was eight days from pupa to emerge as an adult. An considerably high (more than 0.4 eggs/pod) for all adult took the next three to five days to oviposit treatments. This might due to the low number of their eggs. In addition, based on observation (see pods available during sampling occasions, which Figure 9), CPB is capable to emerge more than 10 may influence oviposition preference of CPB. days from heap date. Therefore, they successful Study may be repeated during peak crop seasons in continued their generation, and four weeks might the future. be the perfect time for all CPB to emerge from

Means followed by the same letters at vertical rows were not significantly different (p ≥ 0.05).

Figure 5. Mean of CPB eggs after augmentation of heaps of cocoa pod.

114

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Means followed by the same letters at vertical rows were not significantly different (p ≥ 0.05).

Figure 6. Mean of CPB eggs at different ranges from heaps of cocoa pod.

Means followed by the same letters at vertical rows were not significantly different (p ≥ 0.05). Figure 7. Mean of CPB eggs obtained from different numbers of heaped cocoa pod.

Some smallholders usually left heaps of the laboratory with the heaps of approximately harvested cocoa pods several days in the block 150-200 pods and number of pupa emergence was before pod breaking. In block with the existence of monitored for 10 days. Cocoa pod borer (CPB) CPB, this practice strongly should be avoided due was managed to bore out until Day-10 (Figure 9), to insidious and cryptic behavior of the pest. The with the highest number of pupa recorded at Day- larvae inside the pod are still capable to bore out 4. In another observation, smallholders must be and produce a silken cocoon once they emerge as a advised to cut infested cocoa pods into small pupa. In different observation, field observation pieces (cut more than 5 pieces per pod) due to shows that the number of pupa collected was the lesser number of pupa emergence (Figure 10). It highest at 50 and 75 pods and significantly will be better practice rather than just left infested different (p≤0.05) compared to heaps of 12 and 25 cocoa pods on the floor of the block. pods (Figure 8). The observation was continued in

115

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Figure 8. Number of pupa emerges from heap pods (12, 25, 50 and 75 pods) during seven day observed in the cocoa field

Figure 9. Number of pupa emerges from approximately 150-200 pods during10- days of observation in the laboratory

Figure 10. Number of pupa emerges from the cocoa pod that cut into different pieces

116

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

DISCUSSION innovations including pests and diseases control measures (Kumi and Daymond, 2015). Therefore, With the arrival of CPB in many countries, combination of control techniques that are massive effort to control the pest infestation have feasible, easy to understand, less technical yet to be undertaken and smallholders requires to efficient must be promoted to smallholders. spend more time in the cocoa field. Curry et al. (2015) stated that before the emergence of CPB in Combination of complete management Papua New Guinea, cocoa only required relatively practices that includes schedule insecticide little labor and provided a good return in yield. spraying, pruning, fertilization and weekly pod Heavy infestation of CPB is not negligible until harvesting was proven effective in reducing CPB smallholders can no longer maintain a low input eggs, entry as well as exit holes (Block 18C). system production for survival of their cocoa Where no insecticide was implemented, CPB production. To date, traditional farming with low- successfully deposited more eggs, even the weekly input cropping system at the very serious harvesting of ripe pods were implemented (Block infestation of pests, meant virtually no healthy 17C). This was in agreement with McMahon et al. mature cocoa pods were available to harvest. (2015) that denoted cocoa trees were exposed to pests and diseases incidence with the absence of In a country where cocoa growers cannot chemical control as in observation in North afford much pesticide use as in Ghana, farmers Kolaka, Sulawesi from 2010-2012. However, in have reverted to cultural approach in managing the same study, contradict results were obtained at pests and diseases. This traditional method two other locations, Pinrang and Polman where involved waste disposal, primarily careful weeding both locations produced better results in term of and pruning (Leiter and Harding, 2004). yield estimation and percentage of healthy pods Therefore, Ghanaian cocoa has the best flavor even no insecticide spraying was commenced. quality with lower pesticide use. Even though a Meanwhile observation by Kumi and Daymond package of management practices was (2015) concludes that more than 50% of farmers implemented, additional chemical application is a (n=150) claimed under the proper spraying must and smallholders usually do not practice program, the incidence of pest and disease can be insect control due to the high costs of equipment reduced and resulted in a yield increment (56.7%), and insecticides. Even mass spraying of chemical and only 14% and 7.3% disagree with the former insecticides for long period may affect human and latter statements. Spraying conducted in small health and the environment, a combination of good blocks may also affect other adjacent areas, even maintenance and chemical may increase the only with little advantage (Table 1). This can be productivity up to 50 to 100% in farm trials seen where outside or buffer area of Block 17C (Wessel and Quilt-Wessel, 2015). harbored less CPB eggs compared to the core plot, even with no significant difference between these In order to deal with a serious infestation two areas. However, in most cases, isolated of CPB, researchers and farmers can participate, applications of insecticide generally less effective collaborate, actively seeking, testing new since insects from untreated areas are capable to techniques and ideas. This can be achieved by the re-infest previously treated neighboring or potential synergy through interaction of formal adjacent areas over a period of time. agricultural research and farmer’s own research (Okali et al., 1994). Successful adoption of any Leaving cocoa blocks without any control techniques requires farmers not only stand as approach, plus with longer periods of harvesting passive recipients of technology, but managed to make the problem worse (Block 19C). This is identify, analyze and most important things is able because two weeks was enough for pupa survival, to implement research activities (Conway, 2001). as in observation using mass heap pods (Figure 9). However, with differences in education standard The returns for harvesting a number of healthy and experience of each smallholder might have pods are likely to be so low that in realistic implications on the efficient and effective use of scenarios, cocoa blocks will eventually be

117

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

abandoned, or smallholders only manage to adopt experience. Adaptation decisions and practices minimal management approaches (Anon, 2010). must be responsive to one specific risk or a single Abandoning the cocoa field with the CPB event (Curry et al. 2015). Therefore, the decisions environment clearly is not the best solution. It to adopt or modify existing agriculture practiced would never success in suppressing the infestation, must be practiced “not in a ‘once-off’ manner, but conversely this might increase the pest population requires a dynamic on-going ‘trial-by-error’ later on, and greater loss in term of financial profit. process” (Smith and Skinner, 2002). Therefore, smallholders must adopt any techniques that may reduce CPB incursion in their field, for the sustainability of their cocoa field. CONCLUSIONS

The results also show that insecticide Productivity in cocoa growing countries all over spraying was not the only technique for depressing the world was not viable with heavy pest the population of CPB in the cocoa field. Cultural infestation and disease incidence. Pests and technique such as field sanitation, bury infested diseases problems will lead to low yields and pods or breaking pods into small pieces may help severe repercussions on the income per hectare in suppressing CPB life cycle, especially the pupae basis. It is clear that plot management up to (Figure 10). This is because the entire life cycle of standard for effective control of CPB must be CPB is approximately 27-33 days, and larvae prioritized, before the other control approach takes spend at least 15 days in the pod. Pupa even can place. Smallholders have to understand the pest emerge after 10 days of pod harvesting, and status, occurrence, infestation, and later on they cumulative number of emergence was the greatest are able to experiment with a variety of at Day 4 (Figure 9). Therefore, cutting infested technological options based on their ideas and into small pieces (six or eight was recommended) personal experiences. Extensive training or any were reduced number of pupa emergence within a other transfer of technology program must week. emphasis on the use of new technique, plot sanitation or on how to modify their field practices Smallholders must be informed that one that appropriate to CPB environment. Quality of the best practices is to break the cocoa pods extension training and support programs to control soonest after pod harvesting. Leaving the infested CPB must be deliberate, until most of cocoa pods in longer period may induce a number of smallholders have willing and understood the CPB in the cocoa field. Leaving infested pods in principles of each control approach. The best mass quantity which more than 50 pods per heap management techniques must consist of non- (Figure 8), or close to the cocoa trees (Figure 6) unilateral, straightforward and undemanding will make things worse. Therefore, immature technique. Reluctance to adopt an appropriate infested pods must be removed, and all infested control approach and management technique will cocoa pod husks must be buried as it is the part of forever place CPB as an interminable problem. In CPB control approaches (McMahon et al. 2015). the worst scenario of CPB infestation, cocoa Never leave infested pods in the field must be plantation may become only a piece of history in practiced, to avoid they serve as a source of CPB Malaysia, in the future. to infest remaining healthy pods (Figure 7). To achieve satisfactory yield from cocoa, smallholders must be updated with any new ACKNOWLEDGEMENTS information, current suggestion management technique as well as recent technology. It can be We thank the Director General of Malaysian proven by the study conducted by Kumi and Cocoa Board, Datin Norhaini Udin and the Daymond (2015) in Ghana, where they found that Director of Upstream Division Technology, Mr. among 150 respondents, farmers with more years Haya Ramba for permission to publish this paper. of experience in growing cocoa had higher yield Data collection for almost three years and research per hectare compared to farmers with less technical assistance were provided by Entomology

118

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Staffs at the CRDC Hilir Perak; Mr. Roslan Sa’adi, Curry, G. N., Koczberski, G., Lummani, J., Mr. Ahmad Zaki Yusoff, Mr. Abdul Mutalib Abd. Nailina, R., Peter, E., McNally, G. and Kadir and Mr. Mohamad Faiz Yahya. Special Kuaimba, O. (2015). A bridge too far? thanks to practical internship students from The influence of socio-cultural values on UNISZA, Norhaniza Roselee and Noor Naimah the adaptation responses of smallholders Syafiqah Bt. Nokman that contributing in some to a devastating pest outbreak in cocoa. data collections. Global Environmental Change 35: 1-11.

Curry, G. N. and Koczberski, G. (2009). Finding REFERENCES common ground: relational concepts of land tenure and economy in the oil palm Anon. (2010). Final Report – Socioeconomic frontier of Papua New Guinea. impact assessment of cocoa pod borer in Geographical Journal 175(2): 98-111. East New Britain Province, Papua New Guinea FR 2010-25. Australian Centre Day, R. K. (1989). Effect of cocoa pod borer, for International Agricultural Research Conopomorpha cramerella, on cocoa (ACIAR). yield and quality in Sabah, Malaysia. Crop Protection 8: 332-339. Azhar, I. and Lee, M. T. (2004). Perspective for cocoa cultivation in Malaysia: Relook at Dormon, E. N. A., Van Huis, A., Leeuwis, C., the economic indicators. Malaysian Obeng-Afori, D. and Sakyi-Dawson, O. Cocoa Journal 1: 1-18. (2004). Causes of low productivity of cocoa in Ghana: farmers' perspectives and Azhar, I. (2007). The ways towards sustainability insights from research and the of cocoa industry in Malaysia. ICCO sociopolitical establishment. NJAS - Round Table on a Sustainable World Wageningen Journal of Life Sciences 52: Cocoa Economy, 3-6th October 2007, 237-259. Accra, Ghana. Duguma, B., Gockowski, J. and Bakala, J. (2001). Bateman, R. (2015). Pesticide use in cocoa. A Smallholder cacao (Theobroma cacao guide for training administrative and Linn.) cultivation in agroforestry systems research staff. Third Edition. of West and Central Africa: Challenges International Cocoa Organization and opportunities. Agroforestry systems, (ICCO), London, United Kingdom. P 51(3): 177-188. 109. FAO. (2014). The State of Food and Agriculture. Conway, G. R. (2001). The Doubly Green Innovation in Family Farming. FAO. Revolution: A context for farming Rome. systems research and extension in the 21st century. Journal for Farming Systems Lee, M. T. (2006). Agricultural practices towards a Research-Extension. Special Int. sustainable cocoa economy. Proceedings Symposium Edition PP 1-16 15th International Cocoa Research Conference, 9-14th October 2006, San Coulibaly, O., Mbila, D., Sonwa, D. J., Adesina, Jose, Costa Rica. A. and Bakala, J. (2002). Responding to economic crisis in sub-Saharan Africa: Leiter, J. and Harding, S. (2004). Trinidad, Brazil New farmer-developed pest management and Ghana: Tree melting moments in the strategies in cocoa based-plantations in history of cocoa. Journal of Rural Studies Southern Cameroon. Integrated Pest 20: 113-130. Management Reviews 7: 165-172.

119

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Kumi, E. and Daymond, A. J. (2015). Farmers’ RIRDC. (2013). Commercializing cocoa growing perceptions of the effectiveness of the in North Queensland. RIRDC Cocoa Disease and Pest Control Publication, No. 13/114. Rural Industries Programme (CODAPEC) in Ghana and Research and Development Corporation, its effects on poverty reduction. American Australia. Journal of Experimental Agriculture 7(5): 257-274. Smith, B. and Skinner, M. W. (2002). Adaptation options in agriculture to climate change: a McMahon, P., Hussin, P., Lambert, S., Mulia, S., typology. Mitigation and Adaptation Nurlaila., Susilo, A. W., Sulistyowati, E., Strategies for Global Change 7: 85-114. Sukamto, S., Israel, M., Saftar, A., Amir, A., Purwantara, A., Iswanto, A., Guest, Teh, C. L., Pang, J. T. and Ho, C. T. (2006). D. and Keane, P. (2015). Testing local Variation of the response of clonal cocoa cocoa selections in three provinces in to attack by cocoa pod borer Sulawesi: (i) Productivity and resistance Conopomorpha cramerella (Lepidoptera: to cocoa pod borer and Phytophthora pod Gracillariidae) in Sabah. Crop Protection rot (black pod). Crop Protection 70: 28- 25: 712-717. 39. Wessel, M. and Quist-Wessel, P. M. F. (2015). MPIC (2014). Laporan Tahunan 2014. Cocoa production in West Africa, a Kementerian Perusahaan Perladangan dan review and analysis of recent Komoditi Malaysia, Putrajaya, Malaysia, developments. NJAS-Wageningen Pp 205. Journal of Life Sciences 74-75: 1-7.

Okali, C., Sumberg, J. E and Farrington, J. (1994). World Cocoa Foundation (2014). Cocoa Market Farmer Participatory Research: Rhetoric Update. http://worldcocoafoundation.org/ and Reality. Intermediate Technology wp-content/uploads/Cocoa-Market- Publications, London, UK. P 159. Update-as-of-4-1-2014.pdf (sourced: 19.06.2016).

120

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

ESTIMATING BLACK POD DISEASE PROGRESS RATE USING THE AREA UNDER DISEASE PROGRESS CURVE (AUDPC) FOR DIFFERENT COCOA CLONES IN LABORATORY CONDITION

Ling, A.S.C.1, Ahmad Kamil, M.J.2, Chong, K.P.3 and Ho, C.M.3 1Malaysian Cocoa Board, Tingkat 5, 6 & 7, Wisma SEDCO, Lorong Plaza Wawasan, Off Coastal Highway, Beg Berkunci 211, 88999 Kota Kinabalu, Sabah, Malaysia. 2 Cocoa Research & Development Centre Kota Samarahan, Malaysian Cocoa Board, Lot 248, Block 14, Daerah Muara Tuang, Bahagian Samarahan, Locked Bag 3131, 93450, Kuching, Sarawak. 3Faculty of Science and Natural Resources, Universiti Malaysia Sabah, Jalan UMS, 88400 Kota Kinabalu Sabah, Malaysia.

Malaysian Cocoa Journal 9(1): 121-126 (2016) ABSTRACT – The black pod disease is still a major threat in cocoa planting in Malaysia. The losses due to the black pod disease which caused by Phytophthora exceed $400 million worldwide. Breeding work on identifying the resistance clone to black pod disease is important as the most economical, environmentally friendly and effective control method. The detached pod test is used in breeding program to identify clone with resistant to pests and diseases including cocoa black pod disease. This paper is to model the black pod disease progress rate in different category of resistant clones (KKM 4, KKM 5, QH 1003 and BR 25) using nonlinear models such as exponential model, monomolecular model, logistic model and gompertz model, then quantified the disease intensity over time using the area under disease progress curve (AUDPC). The goodness of fit test showed that lesion growth rate of clones BR 25 and KKM 5 are fitted well with Logistics model and Gompertz model, respectively. Meanwhile, Exponential model is performed better in clone KKM 4 and QH 1003. Comparison of AUDPC among the four clones showed that BR 25 has the lowest AUDPC value (35.76) followed by QH 1003 (40.97) then KKM 4 (55.60) and KKM 5 has the largest value of AUDPC (130.33). These results indicated that the resistance level in clones BR 25 and QH 1003 towards black pod disease is better than clones KKM 4 and KKM 5 that have less resistance to black pod disease. As this suggested approach in estimating black pod disease progress rate using AUDPC on fitted nonlinear models is useful in quantifying the resistance clone to black pod disease in the detached pod test.

Key words: Nonlinear model, Phytophthora, Cocoa clones, Area under disease progress curve.

INTRODUCTION to cocoa black pod disease widely described in the papers of Nyadanu et al. (2013), Thevenin et al. The black pod disease is still a major threat in (2012) and Barreto et al. (2015) meanwhile, cocoa planting which could caused major losses to Malaysia has successfully categorized cocoa cocoa farmers in Malaysia. The losses are clones planted in Malaysia into four categories estimated to be exceeded $400 million worldwide such as susceptible, moderate susceptible, (ICCO, 2013). There are several control measures moderate resistant and resistant to cocoa black pod have been used to overcome the cocoa black pod disease (Ramba et al., 2012). disease problem which caused by Phytophthora such as using copper–based fungicides, cultural The detached pod test has been used in practices and screening resistant clones to cocoa breeding program to screen the clones that black pod disease. Breeding program in resistant to pests and diseases including cocoa identifying resistant clones to black pod disease black pod disease. There is a study showed that has been the most economical, environmentally detached pod test is an effective method to assess friendly and effective control method (Iwaro and the clonal resistance to cocoa pod rot in laboratory Singh, 2004). Work on developing resistant clones due to its highly correlated with symptom of cocoa

121

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

pod rot observed in field over period of 3 years Mathematical models used in fitting the black (Iwaro et al., 2005). Currently, the measurement pod lesion growth used in determine the resistant level of cocoa Four nonlinear models are used to describe the clones to cocoa black pod disease in detached pod cocoa black pod disease progress or the lesion test is based on black pod lesion growth rate at day growth rate, namely exponential model, 6 after inoculation in the laboratory. However, this monomolecular model, logistic model and measurement didn’t utilize the information of the gompertz model (Madden et al., 2007). lesion growth rate from day 1 to day 5 which might different from day 6 in determine the Exponential Model: y = y0 exp(rEt) resistant level. Therefore, it is important to understand and estimate the black pod disease (1) progress rate by taken into account the information of lesion growth rate from day 1 to day 6 in the where y0 is representing the start of epidemic or laboratory to increase the accuracy of the initial disease intensity and rE is a rate parameter screening method using detached pod test in terms (constant). of resistant clones to cocoa black pod. Monomolecular Model: y = 1 – (1 – y0) exp(–rM t)

MATERIALS AND METHOD (2)

Experimental design where y0 is a constant of integration or value at Four cocoa clones of KKM 4, KKM 5, QH 1003 time, t = 0 and rM is a rate parameter (constant). and BR 25 that have different resistant level on Logistic Model: cocoa black pod disease are collected from the Cocoa Research and Development Center Madai, (3) Sabah, to be assessed for black pod disease progress rate using detached pod method. Five where y0 is a constant of integration and represents healthy ripe pods from each clone were brought to disease intensity at t = 0 and rL is a rate parameter the laboratory and wiped with a piece of cotton (constant). wool soaked with 70% ethanol and rinse within sterile distilled water and allowed to dry. Gompertz Model:

Detached pod test described by Iwaro et (4) al. (1997) was carried out. An isolate of Phytophthora sp. used in the study was obtained where B is a constant of integration equal to – from a naturally infected cocoa pod from a field at ln(y0) and rG is a rate parameter. the Cocoa Research and Development Center Madai, Sabah. Inoculated pods with mycelial plugs from a 3 day old Phytophthora palmivora Statistical analysis culture grown on 20% V8 juice calcium carbonate Curve fitting on the average diameter of lesion agar medium were arranged in a randomized growth is done using PROC NLIN (SAS Institute complete design with five pods per clone and 2011). PROC NLIN fits nonlinear models by incubated at room temperature in the laboratory. st th minimizing the error sum of squares. On the 1 day till 6 day of incubation, diameter of the established lesions were measured with a Calculation of measures for goodness-of-fit caliper meter. In order to identify the best fitted model among the four nonlinear models, two Goodness of Fit tests are used, namely the Akaike Information Criterion (AIC) and the Bayesian Information Criterion

122

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

(BIC). Those tests were chosen based on the study using the method of Yeh (2002) is used to estimate by Spiess and Neumeyer (2010) proved that AIC the black pod disease progress rate. Disease and BIC are performance better than R-squared for progress data is summarized into one value by nonlinear model. AUDPC which are proportioned between diameter lesion of cocoa black pod disease and duration of AIC = 2p – 2ln(L) the lesion developed. The AUDPC is calculated by (5) the trapezoidal rule that can be presented as integral function given as follows: with p = number of parameters and ln(L) = maximum log-likelihood of the estimated model ≈ T(a, b, n) = ((b – a) / n) × (((f(a) + and calculated as follows: f(b)) / 2) + Σ f(a + i (b – a) / n)) (8)

where the domain [a, b] of the integration function (6) are subdivided into n strips with the points of x0, x1, . . ., xn . x0 = a, xn = b and xr = x0 + r (b – a) / n. with x1, ..., xn = the residuals from the nonlinear least squares fit and N = their number. In our case of study, a = 1, b = 6 and n is set to 50 subdivisions or strips while the f(x) is BIC = p ln(n) – 2 ln(L) given as nonlinear function model of each clone. (7) with p = number of parameters, n = sample size RESULTS AND DISCUSSIONS and L = maximum likelihood of the estimated model. Table 1 gave the summary of the length and perimeter of pod used in the study prior to the Estimating on Area Under Disease Progress screening test. The ripe pods used in the study are Curve ranged from 159.00mm to 233.00mm in length The area under the disease progress curve and 235.00mm to 284.00mm in perimeter. (AUDPC) for each nonlinear model that calculated

Table 1. Descriptive statistics on pod length and perimeter for four different cocoa clones. Mean±Stderr Clone N Length (mm) Perimeter (mm) BR25 5 181.00±8.72 252.00±5.83 KKM4 5 170.00±15.41 284.00±9.48 KKM5 5 159.00±2.92 266.00±4.05 QH1003 5 233.00±3.39 235.00±3.16

Table 2 showed the mean lesion diameter indicator of clone’s resistance level towards black growth on four cocoa clones at different time pod disease in the detached pod method showed intervals after inoculation. Observation on the 6th BR 25 has the lowest value followed by QH 1003 day of lesion development that being used as an then KKM 4 and KKM 5 has the largest value.

123

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 2. Mean lesion diameter on cocoa pods of four different cocoa clones, measured on 6 consecutive days. Mean lesion diameter (mm) Clone 1st day 2nd day 3rd day 4th day 5th day 6th day KKM 4 2.0 9.5 16.0 21.2 32.7 50.7 KKM 5 6.0 19.0 36.7 57.5 82.0 106.5 QH 1003 0 3.5 9.6 15.5 26.4 47.3 BR 25 2.0 4.8 8.6 13.5 24.3 33.2

Table 3 showed the results of four developed models with small value in the standard different nonlinear models using PROC NLIN for errors of the estimate parameters. This suggests each clone. The overall F value is very large in all that the model’s fit was excellent.

Table 3. Comparison of four nonlinear models on lesion development of four cocoa varieties. Clone Model of black pod lesion development F value BR 25 Y(Exponential) = 2.224*EXP(0.456*DAY) 368.40** Y(Monomolecular) = 1 +1.812)*EXP(0.486*DAY) 284.24** Y(Logistics) = 55.524/(1+49.985*EXP(–0.721*DAY)) 854.40** Y(Gompertz) = 189.50*EXP(–5.557*EXP(–0.194*DAY)) 619.42** KKM 4 Y(Exponential) = 3.710*EXP(0.436*DAY) 452.12** Y(Monomolecular) = 1 + 3.245*EXP(0.456*DAY) 395.01** Y(Logistics) = 206.80/(1 + 64.616*EXP(–0.506*DAY)) 246.02** Y(Gompertz) = 3290*EXP(–7.334*EXP(–0.087*DAY)) 273.81** KKM 5 Y(Exponential) = 10.966*EXP(0.387*DAY) 200.01** Y(Monomolecular) = 1 – 10.449*EXP(0.393*DAY) 189.66** Y(Logistics) = 137.20/(1 + 29.720*EXP(–0.767*DAY)) 1225.68** Y(Gompertz) = 206.80*EXP(–4.61*EXP(–0.322*DAY)) 6626.88** QH 1003 Y(Exponential) = 1.430*EXP(0.584*DAY) 641.39** Y(Monomolecular) = 1– 1.160*EXP(0.616*DAY) 448.25** Y(Logistics) = 200.10/(1+180.30*EXP(–0.669*DAY)) 370.13** Y(Gompertz) = 9965.80*EXP(–9.510*EXP(–0.096*DAY)) 445.85**

Goodness of fit test showed that lesion models (Table 4). For KKM 5, the lesion growth growth rate of clone BR 25 is fitted well with rate is fitted well with Gompertz model. Logistics model as both AIC and BIC values are Meanwhile, Exponential model is performed better the smallest compared to three other nonlinear in clone KKM 4 and QH 1003.

Table 4. Comparison goodness of fit test for four different nonlinear models on lesion development of cocoa black pod disease. Exponential model Monomolecular model Logistics model Gompertz model Clone AIC BIC AIC BIC AIC BIC AIC BIC BR 25 24.48 24.06 26.03 25.61 17.30 16.68 19.23 18.60 KKM 4 28.12 27.70 28.92 28.51 29.61 28.98 34.85 34.23 KKM 5 42.88 42.46 43.19 42.77 29.89 29.27 19.77 19.15 QH 24.21 23.79 26.35 25.93 25.35 24.72 24.23 23.61 1003

124

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Figure 1 showed the value of AUDPC These results indicated that the resistance level for four different clones with the AUDPC values in clones BR 25 and QH 1003 towards black pod varied among the clones. BR 25 has the lowest disease is better than clones KKM 4 and KKM 5 AUDPC value followed by QH 1003 then KKM that have less resistance to black pod disease. 4 and KKM 5 has the largest value of AUDPC.

AUDPC = 35.76 AUDPC = 55.60

Logistics model for BR 25 Exponential model for KKM 4

AUDPC = 130.33 AUDPC = 40.97

Gompertz model for KKM 5 Exponential model for QH1003

Figure 1. Area under disease progress curve (AUDPC) for lesion development on four different cocoa clones for 6 days after inoculation with Phytopththora. palmivora.

CONCLUSION QH 1003 showed that BR 25 is most resistant to black pod disease compared to three other clones The fitted nonlinear models on the cocoa black as it has the lowest AUPDC value followed by pod progress rate or lesion growth rate from four QH 1003, KKM 4 and KKM 5. different clones enable us to estimate the pod susceptibility in laboratory condition using The estimation of black pod disease detached pod method. The estimate AUPDC progress rate using AUPDC on fitted nonlinear values from the fitted nonlinear models such as model can be used as indicator to compare the Logistics model for BR 25, Gompertz model for resistance level of clones tested against black KKM 5 and Exponential model for KKM 4 and pod disease using detached pod method in

125

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

future. As the AUDPC has considered all lesion Iwaro, A.D., Sreenivasan, T.N., and Umaharan, development from 1st day to 6th day of P. (1997). Foliar resistance to inoculation compared to using the mean lesion Phytophthora palmivora as an diameter of black pod disease development on indicator of pod resistance in 6th day of inoculation. Theobroma cacao. Plant Dis. 81: 619- 624.

ACKNOWLEDGEMENT Madden, L.V., Hughes, G. and van den Bosch, F. (2007). The study of Plant Disease The author also would like to thank the Director- Epidemics. APS Press, St. Paul, MN. General, Malaysian Cocoa Board (MCB) and Director of Upstream Technology Cocoa for Nyadanu, D., Akromah, R., Adomako, B., permission to publish and reviewing this paper. Kwoseh, C., Lowor, S. T., Dzahini- Thanks are also due to Mr. Yahya Mohd. Nor, Obiatey, H., Akrofu, A. Y., Ansah, F. Mr. Zamri Kamari, Pn. Sarinah Ambia and staffs O., Asiama, Y. O. and Assuah, M. K. at Plant Breeding Unit and Plant Pathology Unit, (2013). Biochemical mechanisms of CRDC Tawau, Sabah for their assistance in this resistance to black pod disease in cocoa study. (Theoborma cacao L.). American Journal of Biochemistry and Molecular Biology 3(1): 20–27. REFERENCES Barreto, M.A., Santos, J.C.S., Corrêa, R.X., Ramba, H., Yazik, N.M., Hussin, M.J. and Luz, E.D.M.N., Marelli, J. and Souza, A.P. Muzakkar, M.Q. (2012). Book of (2015). Detection of genetic resistance Malaysia Cocoa Clones. Malaysian to cocoa black pod disease caused by three Cocoa Board. Phytophthora species. Euphytica 206 (3): 677-687. SAS Institute. (2011). SAS for windows, version 9.3. SAS Institute, Cary NC. ICCO (2013). Pests & Diseases. Available at http://www.icco.org/about-cocoa/pest- Thevenin, J.M., Rossi, V., Ducamp, M,, Doare, a- diseases.html. Accessed on 4 July, F., Condina, V., Lachenaud, P. (2012). 2013. Numerous Clones Resistant to Phytophthora palmivora in the Iwaro, A.D. and Singh, V. (2004). Progress ‘‘Guiana’’ Genetic Group of report on the germplasm enhancement Theobroma cacao L. PLoS ONE 7(7): programme for resistance to e40915. black pod disease. In: Annual doi:10.1371/journal.pone.0040915. Report for 2003. Cocoa Research Unit, The University of the Spiess, A. N. and Neumeyer, N. (2010). An West Indies, St. Augustyine, evaluation of R2 as an inadequate Trinidad. pp. 43-45. measure for nonlinear models in pharmacological and biochemical Iwaro, A.D., Thévenin, J.M., Butler, D.R. and research: a Monte Carlo approach. Eskes, A.B. (2005). Useful of the BMC Pharmacology. 10: 6. detached pod test for assessment of cacao resistance to Yeh, Shi-Tao (2002). Using Trapezoidal Rule Phytophthora pod rot. European for the Area Under a Curve Calculation. Journal of Plant Pathology 113: SUGI 27 Proceedings, Cary, NC: SAS 173–182. Institute Inc.

126

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

PLANT GROWTH HORMONES PRODUCED BY ENDOPHYTIC Bacillus subtilis STRAIN LKM-BK ISOLATED FROM COCOA

Ishak, Z.1,2, Mohd Iswadi, M.K.2, Russman Nizam, A.H.2, Ahmad Kamil, M.J.3, Ernie Eileen, R.R.4 , Wan Syaidatul, A.1 and Ainon, H.1 1School of Bioscience and Biotechnology, Faculty of Science and Technology Universiti Kebangsaan Malaysia 43600, Bangi Selangor Malaysia. 2Division of Biotechnology, Cocoa Innovation & Technology Center, Malaysian Cocoa Board, Lot PT12621, Nilai Industrial Area, 71800 Nilai, Negeri Sembilan. 3Malaysia Cocoa Board, Cocoa Research and development Centre, Lot 248, Block 14 Daerah Muara Tuang, Bahagian Samarahan, Kota Samarahan, Sarawak, Malaysia. 4Faculty of Applied Sciences, Universiti Teknologi MARA, 40450 Shah Alam Selangor, Malaysia.

Malaysian Cocoa Journal 9(1): 127-133 (2016) ABSTRACT - Endophytic bacteria living within tissues of healthy Theobroma cacao plants are being assessed for their ability to produce plant growth hormones. Based on 16S ribosomal DNA sequence analysis selected endophytic bacteria was identified as Bacillus subtilis LKM-BK. Two different solvent, ethyl acetate and n-butanol were used to extract plant growth hormones. The extraction fractions were determined using thin-layer chromatography and identified by high pressure liquid chromatography. The results showed endophytic bacteria produced 1-naphthalene acetic acid (NAA), tryptamine, 3-indole propionic acid (IPA), indole-3-butyric acid (IBA), 3-indole acetic acid (IAA), gibberellic acid (GA) and trans-zeatin. The finding indicates that endophytic bacteria showed abilities to produce multiples auxin, gibberellins and cytokinin.

Key words: Endophytic, Bacteria, Plant growth hormones, Cocoa, 16S rDNA sequence

INTRODUCTION pneumonia, Burkhoderia (Pseudomonas) cepacia and Stenotrophomonas (Xanthomonas) Endophytic bacterial colonizes internal plant maltophilia (Dong et al., 2003). host without negative diseases effects and produce bioactive substances such as plant Successful trials of using endophytes as growth hormones to increase plant growth (Khan plant growth hormones has been shown in et al., 2014). Endophytic bacteria create a center profitable plants like corn (Szilagyi-Zecchin et of attention because of demand to bring down al., 2014), rice (Waqas et al., 2014), tomato the use of chemicals, especially when to be (Khan et al., 2014), soybean (Dalal and concerned about environmental protection (Vale Kulkarni, 2013), pepper (Jasim et al., 2013), et al., 2010). The approach is to use endophytes poplar trees (Taghavi et al., 2009), grapevine that may increase plants growth and health when (Barka et al., 2002), potatoes (Sturz et al., 1998), added as inoculants (Lucy et al., 2004). The and others. Barka et al., (2002) reported that commonly recognized classes of plant growth plant growth hormone, endophytic hormones are the auxins, gibberellins, rhizobacterium, Pseudomonas sp. strain PsJN, cytokinins, abscisic acid, and ethylene. demonstrated an induction of plant growth as Endophytes are well known mainly for their well as antagonistic effect on in vitro growth of ability to produce plant hormones in vitro, grapevine. Taghavi et al., (2009) through their among which are indoles, mainly IAA (Luz et greenhouse experiment revealed poplar plants al., 2008), gibberellins (Bottini et al., 2004) and cuttings inoculated with endophyte cytokinins (Tien et al., 1979). The hormones (Enterobacter sp. strain 638) repeatedly increase produced by endophytic bacteria are involved in in biomass production compared to cuttings of plant growth promoting and yield production for non-inoculated control plants. These endophyte examples endophytic Pseudomonas and Bacillus producing plant growth hormones were carried sp (Alam, et al., 2001; and Cakmakci et al., out on many crops under a variety of soil and 2001), Enterobacter (Pantoea), Klebsiella environmental condition, which frequently

127

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

resulted in significant increasing yield (95oC for 45 sec; 51oC for 15 sec; 72oC for 2 productivity (Okon and Labandera-Gonzalez, min) for denaturation, annealing and extension, 1994). and 1 cycle (72oC for 10 min) for final extension of the amplified DNA. The PCR products were The aim of this project is to isolate and purified by standard methods and directly identify selected endophytic bacteria from cocoa sequenced with primers 518F and 800R using plant. As the plant growth hormones produced BigDye® Terminator v3.1 cycle Sequencing Kit by endophytic bacteria can vary, it is important (Applied Biosystems). The sequences were to determine and identify the plant growth compared using BLAST search program hormones produced selected endophytic (http://www.ncbi.nlm.gov/BLAST). The bacteria. Since studies on endophytic bacteria analyzed sequences have been deposited in isolated from cocoa plant very limited, the study GenBank. is significant and will be developed as locally produced bio-fertilizer for cocoa plantation Extraction of plant growth hormones specifically in Malaysia. Selected endophytic bacterial was grown in 200 mL of liquid nutrient broth (Sigma, St. Louis, MO, USA) in 500 mL shake flask and incubated MATERIALS AND METHODS at 28 ± 2oC with agitation of 120 rpm for 24 hours. The extraction of plant growth hormone Isolation of endophytic bacteria produced by endophytic bacteria was performed Endophytic bacteria isolates were carried out by using methods from Tien et al., (1979). followed the protocol described by Shimizu et Endophytic bacteria culture (200 mL) was al. (2000) with modifications. Twigs along with centrifuged at 7,700 x g for 30 minutes to get leaves were collected from healthy cocoa tree cell-free supernatant. The cell-free supernatant grown in Sabah, Malaysia. The samples were was reduced to 50 mL by evaporation under washed under running water for 30 min prior to vacuum and adjusted to pH 2.8 using 1 N HCL. surface sterilization. After washing, the leaf The plant growth hormones from cell-free samples were cut into pieces (1 x 1 cm), and the supernatant, was extracted using ethyl acetate for twigs were cut into 1 cm long segments. Pieces 3 times to produce ethyl acetate fraction and of the leaves and twigs were soaked in 1% aqueous fraction. Ethyl acetate fraction was sodium hypochlorite for 1 min, followed by 70% evaporated under vacuum and dissolved in ethanol for 5 min. The sterilized plant pieces absolute methanol for Thin Layer were further rinsed with sterile water. All the Chromatography (TLC) and High Pressure pieces were placed on nutrient agar (NA) Liquid Chromatography (HPLC) bioassay. medium and incubated at 28oC for 7 days. The Meanwhile, aqueous fraction was adjusted to pH colonies that appeared on the culture plates 7.0 using 1 N NaOH and was extracted with around the plant tissues were streaked onto new water-saturated n-butanol for 3 times. Fraction NA plates, and the pure cultures of each isolate of n-butanol was separated and evaporated under were suspended in 20% glycerol and stored at - vacuum and dissolved in absolute methanol for 80oC for further study. Working culture of TLC and HPLC bioassay. selected endophytic bacteria was established by transferring from stock culture onto NA plate Thin Layer Chromatography (TLC) and incubated at 28oC for 24 hours. TLC chromatograms were run on 3 cm width x 10 cm length preparative silica gel plates (Gel Endophytic bacteria identification silica 60GF254, Merck). The solvent system used The selected endophytic bacterial 16S ribosomal were chloroform : ethyl acetate : formic acid DNA was amplified using the primer 1492R and (50:40:10 v/v) to separate indole compounds and primer 27F. The total reaction volume of 25 µl gibberellins in ethyl acetate fractions. Another contained DNA from crude bacteria lysate, 10 solvent n-butanol : acetic acid : water (12:3:5 pmol of each primer, deoxy nucleotides v/v) was used to separate cytokinin in n-butanol triphosphates (dNTPs, 400 µM each), 0.75 U of fraction. About 1 mL of fraction was spotted at 1 Taq DNA polymerase, and supplied buffer. The cm from the bottom of the TLC plate. The plate PCR was performed as follow: 1 cycle (95oC for was then transferred to glass jar containing 5 minutes) for initial denaturation; 30 cycles above mobile phase. The mobile phase was left

128

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

to rise until reached 1 cm at the end of plate. The RESULTS AND DISCUSSIONS separated spots on TLC chromatograms were visualized under UV254 nm (Oedjijono et al., Several endophytic bacteria that colonized inside 1993) and iodine vapor. The Rf value of each plant tissues were successfully isolated from separated spots were measured. surface-sterilized of healthy cocoa twigs along with leaves. The presence of endophytic bacteria High Pressure Liquid Chromatography from cocoa tree support earlier reports of various (HPLC) endophytic isolates can be found in almost all HPLC chromatograms were produced by plants (Yaojian et al., 2001). One of the isolates injecting 10 μl of the filtered extracts onto a was selected for further identification and reverse-phase column (Nucleosil EC 300/4 100- extraction of plant growth hormones. 10 C18) using JASCO ChromPass Identification of selected endophytic bacteria Chromatography System (JAPAN), equipped was performed by sequencing of the 16S rDNA with Photo Diode Array (MD 2010 plus) at 254 region of the bacteria. The 16S rDNA sequence nm. Two solvent systems were used to separate of the endophytic bacteria strain consisting of hormones compounds. Solvent A was 85% of 1361 bp was determined. The results of BLASTn wáter : acetonitrile : acetic acid (85:15:1 v/v) and search from the National Center of solvent B was 15% of 30% methanol in water, Biotechnology Information (NCBI) and with flow rate of 1.0 mL/min. Retention times comparison of sequences against the GenBank for peaks were compared to hormones standard; database indicated that the selected endophytic 3-indole acetic acid (IAA-Sigma), 1-naphthalene bacteria belong to Bacillus subtilis LKM-BK acetic acid (NAA-Sigma), indole-3-butyric acid (Figure 1). The 16S rDNA sequence was (IBA-Sigma), 3-indole propionic acid (IPA- submitted to GenBank, with the accession Aldrich), tryptamine (TRY-Aldrich), trans-zeatin number of KR 560045. (Sigma), and gibberellic acid (GA-Sigma).

Figure 1. Phylogenetic tree obtained of 16S rDNA sequences, showing the position of Bacillus subtilis LKM-BK among phylogenetic neighbors.

129

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Plant growth hormones extracted from et al., 1972). Martinez-Morales et al., (2003) selected endophytic bacteria Bacillus subtilis reported that Azospirillum brasilense produced LKM-BK using ethyl acetate fraction (Figure 2) substance associated with auxin activity in cell- showed four clear individual spots and several free supernatant can be spotted by TLC faded spots of indole (auxins) and gibberellins apparatus. Meanwhile, the n-butanol fraction hormones on TLC plate. The clear spots detected revealed cytokinin spot detected at Rf value of at Rf value of 0.3, 0.6, 0.7 and 0.9. Some of the 0.5 (Figure 2). Cytokinin has been observed in spot were faded due to the type of solvent several bacteria Rhizobium sp., Bacillus subtilis system used for TLC to separate indole BC1 and Escherichia coli K12 by TLC showing compounds, damaged the hormones (Schneider Rf value of 0.65 (Sabat et al., 2014).

Figure 2. TLC chromatogram of Bacillus subtilis LKM-BK. (A: Ethyl acetate fraction. B: n- butanol fraction)

Identification of indole (auxin) and gibberellic acid and cytokinin has been reported gibberellins hormones from ethyl acetic fractions to be produced by Bacillus sp., Micrococcus sp., using HPLC revealed that, Bacillus subtilis Pseudomonas sp. Flavobacterium sp. and LKM-BK produces NAA, TRY, IPA, IBA, IAA Serratia sp. (UmaMaheswari et al., 2013). and GA (Figure 3). Meanwhile, the cytokinin Interestingly, Bacillus subtilis LKM-BK isolated hormones from n-butanol fraction showed that from cocoa plant have multiple abilities to Bacillus subtilis LKM-BK only produced trans- produce several auxin (NAA, TRY, IPA, IBA zeatin (Figure 4). Khan et al., (2016) observed and IAA), GA and cytokinin (trans-zeatin). that Bacillus subtilis LK14 isolated from Moringa peregrine plant produced IAA (auxin) hormone to improve the growth of Solanum lycopersicum. Auxin hormones especially IAA was known to play important role in cell elongation, division and enlargement (Ishida et al., 2013). While GA stimulated shoot length, chlorophyll contents and biomass of rice plant growth (Waqas et al., 2014). IAA along with

130

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Figure 3. HPLC chromatograms of Bacillus subtilis LKM-BK using ethyl acetate fraction.

Figure 4. HPLC chromatograms of Bacillus subtilis LKM-BK using n-butanol fraction.

CONCLUSIONS cytokinin. Bacillus subtilis LKM-BK can be produced in the large scale and potential as Endophytic bacteria have been reported to live in naturally bio-fertilizer for cocoa growth. symbiosis within diverse plants but, their known function is still limited. This study showed one of the major contributions of endophytic bacteria ACKNOWLEDGEMENT towards plant growth is the production of growth hormones. Plant growth hormones produced Research project is supported by Intensive from selected endophytic bacteria Bacillus Research in Priority Area (IRPA) of Ministry of subtilis LKM-BK isolated from cocoa plant Sciences, Technology and Environment of showed the present of auxin, gibberellins and Malaysia.

131

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

REFERENCES Khan, A.L., Halo, B.A., Elyassi, A., Ali, S., Al- Hosni, J., Hussain, J., Al-Harrasi, A. Alam, M.S., Cui, Z.J., Yamagishi, T. and Ishii, and Lee, I.J. (2016). Indole acetic acid R. (2001). Grain yield and related and ACC deaminase from endophytic physiological characteristics of rice bacteria improves the growth of plants Oryza sativa L. inoculated with Solanum lycopersicum. Electronic free-living rhizobacteria. Plant Journal of Biotechnology, 21: 58-64. Production Science, 4: 125-130. Khan, A.L., Waqas, M., Kang, S.M., Al-Harrasi, Barka, E.A., Gognies, S., Nowak, J., Audran, A., Hussain, J., Al-Rawahi, A., Al- J.C. and Belarbi, A. (2002). Inhibitory Khiziri, S., Ullah, I., Ali, L., Jung, H.Y. effect of endophyte bacteria on Botrytis and Lee, I.J. (2014). Bacterial cinerea and its influence to promote the endophyte Sphingomonas sp. LK11 grapevine growth. Biological Control, produces gibberellins and IAA and 24: 135-142. promotes tomato plant growth. Journal of Microbiology, 52 (8): 689-695. Bottini, R., Cassan, F. and Piccoli, P. (2004). Gibberellin production by bacteria and Lucy, M., Reed, E. and Glick, B.R. (2004). its involvement in plant growth Application of free living plant growth- promotion and yield increase. Applied promoting rhizobacteria. Antonie van Microbiology Biotechnology, 65: 497- Leewenhoek, 86: 1-25 503. Luz, E.B., Hani, A. and Yoav, B. (2008). Cakmakci, R., Kantar F. and Sahin F. (2001). Involment of indole-3-acetic acid Effect of N2- fixing bacteria produced by the growth-promoting inoculations on yield of sugar beet and bacterium Azospirillum spp. In barley. Journal of Plant Nutrition and promoting growth of Chlorella Soil Science, 164: 527-531. vulgaris. Journal of Phycology, 44: 938-947. Dalal, J. and Kulkarni, N. (2013). Antagonistic and plant growth promoting potentials Martinez-Morales, L.J., Soto-Urzua, L., Beatriz, of indigenous endophytic bacteria of E.B. and Sanchez-Ahedo, J.A. (2003). soybean (Glycine max (L) Merril). Indole-3-butyric acid (IBA) production Current Research in Microbiology and in culture medium by wild strain Biotechnology 1(2): 62-69. Azospirillum brasilense. FEMS Microbiology Letters, 228:167-173. Dong, Y., Iniguez, A.L. and Triplett, E.W. (2003). Quantitative assessments of the Oedjijono, M.A.L., Line, M.A. and Dragar, C. host range and strain specificity of (1993). Isolation of bacteria endophytic colonization by Klebsiella antagonistic to a range of plant pneumonia 342. Plant Soil 257: 49-59. pathogenic fungi. Soil Biology Biochemistry 25(2): 247-250. Ishida, Y., Nakamura, A., Mitani, Y., Suzuki, M., Soeno, K., Asami T. and Shimada, Okon, Y. and Labandera-Gonzalez, C.A. (1994). Y. (2013). Comparison of indole Agronomic applications of derivatives as potential intermediates of Azospirillum: An evaluation of 20 year auxin biosynthesis in Arabidopsis. world-wide field inoculation. Soil Plant Biotech., 30: 185-190. Biology and Biochemistry, 26: 1591- 1601. Jasim, B., Jimtha, C.J., Jyothis, M. and Radhakrishnan, E.K. (2013). Plant Sabat, S., Murthy, V.K., Shantha, S.L., growth promoting potential of Kushnoor, D., Agarwal, G., Thomas, J. endophytic bacteria isolated from Piper and Devaraj, S. (2014). Comparative nigrum. Plant Growth Regul., 71: 1-11. study of cytokinin production isolated

132

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

from bacteria and shoot induction. UmaMaheswari, T., Anbukkarasi, K., Indian Journal of Biotechnology 13: Hemalatha, T. and Chendrayan, K. 544-546. (2013). Studies on phytohormone producing ability of indigenous Schneider, E.A., Gibson, R.A. and Wightman, F. endophytic bacteria isolated from (1972). Biosynthesis and metabolism of tropical legume crops. International indole-3yl-acetic acid. 1. The native Journal of Current Microbiology and indoles of barley and tomato shoots. Applied Sciences, 2(6): 127-136. Journal of Experimental Botany, 23: 152-170. Vale, M., Seldin, L., Araujo, F.F. and Lima, R. (2010). Plant growth promoting Shimizu, M., Nakagawa, Y., Sato, Y., Furumai, rhizobacteria: fundamentals and T., Igarashi, Y., Onaka, H., Yoshida, R. applications. In: Maheswari D.K. (ed). and Kunoh, H. (2000). Studies on Plant growth and health promoting endophytic actinomycetes (I) bacteria. Springer, Berlin, 21-43. Streptomycetes sp. isolated from rhododendron and its antifungal Waqas, M., Khan, A.L. and Lee, I.J. (2014). activity. Journal of General Plant Bioactive chemical constituents Pathology, 66: 360-366. produced by endophytes and effects on rice plant growth. Journal of Plant Sturz, A.V., Christie, B.R. and Matheson, B.G. Interactions, 9 (1): 478-487. (1998). Associations of bacterial endophyte population from red clover Yaojian, H., Jianfeng, W., Guiling L., Zhonghui, and potato crops with potential for Z. and Wenjin, S. (2001). Antitumor beneficial allelopathy. Canadian and antifungal activities in endophytic Journal of Microbiology, 44: 162-167. fungi isolated from pharmaceutical plants. FEMS Immunology and Medical Szilagyi-Zecchin, V.J., Ikeda, A.C., Hungria, M., Microbiology, 31: 163-167. Adamoski, D., Kava-Cordeiro, V., Glienke, C. and Galli-Terasawa, L.V. (2014). Identification and characterization of endophytic bacteria from corn (Zea mays L.) roots with biotechnological potential in agriculture. AMB Express, 4: 26.

Taghavi, S., Garafola, C., Monchy, S., Newman, L.A., Hoffman, A., Weyen, N., Barac, T., Vangronsveld, J. and van der Lelie, D. (2009). Genome survey and characterization of endophytic bacteria exhibiting a beneficial effect on growth and development of poplar. Applied and Environmental Microbiology, 75 (3): 748-757.

Tien, T.M., Gaskin, M.H. and Hubbell, D.H. (1979). Plant growth substances produced by Azopirillum brasilense and their effect on the growth of pearl millet (Pennisetum americanum L.). Applied and Environmental Microbiology, 37: 1016-1024.

133

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

EFFECTIVENESS OF SYSTEMIC FUNGICIDE (TEBUCONAZOLE) IN CONTROLLING VASCULAR STREAK DIEBACK (VSD) ON COCOA SEEDLINGS

Suhaida, S.1, Shari Fuddin, S.2 and Alias, A.1 1Malaysian Cocoa Board, Cocoa Research & Development Centre Hilir Perak, Jalan Sg. Dulang, 36307 Hutan Melintang, Perak. 2Malaysian Cocoa Board, Cocoa Research & Development Centre, Batu 10, Jalan Apas, Peti Surat 60237 91012 Tawau, Sabah, Malaysia. *Corresponding author email: [email protected]

Malaysian Cocoa Journal 9(1): 134-138 (2016) ABSTRACT - Vascular streak dieback (VSD) is considered one of the most important diseases of cocoa which reduces large amount of annual cocoa production. The most effective fungicide that previously applied to control this disease was triadimenol. However, it is no longer available in Malaysian market. Therefore, evaluation of alternative fungicide is important to control this disease. The study aims to evaluate the potential of tebuconazole in treating infected cocoa seedlings by VSD. This fungicide previously exhibited excellent performance in inhibiting the growth of VSD pathogen under in vitro conditions. Three months old of moderately infected cocoa seedlings were applied with different concentrations of tebuconazole (400, 200 and 1 ppm). The seedlings were treated at 7 and 14 day intervals followed by monthly scoring of disease severity for 8 months. Among the treatment tested, severity of VSD was significantly reduced on cocoa seedlings treated at 7 day intervals with T1 (400ppm) and T2 (200 ppm), providing disease reduction up to 79.63%. In contrast, all concentration applied at 14 day intervals showed a low disease reduction ranging from 18.52% to 29.63%. These results showed tebuconazole was effectively reducing the VSD disease but in a frequent application. Further field evaluations of this fungicide are required.

Key words: Cocoa, Tebuconazole, Vascular streak dieback

INTRODUCTION Vascular streak dieback is currently controlled by cultural practices, resistance Vascular streak dieback (VSD) of cocoa planting materials and chemicals. Since VSD is (Theobroma cacao), caused by the a systemic disease, the application of systemic basidiomycete fungus Ceratobasidium fungicides had been suggested. Guest and Keane theobromae (syn. Oncobasidium theobromae) is (2007) reported that systemic ergosterol one of the most threatening diseases on cocoa in biosynthesis inhibiting fungicides, such as Southeast Asia and Melanesia regions. In flutriafol, hexaconazole, propiconazole, Malaysia, VSD problem was first revealed by tebuconazole, and triadimenol, have a potential Keane and Turner in 1971, where they found in suppressing VSD in seedlings and mature particular form of VSD symptoms similar with cocoa trees. recorded in Papua New Guinea (Keane et al., 1972). Since then, it was reported to be In Malaysia, triadimenol has been potentially destructive disease on cocoa in extensively used in commercial cocoa Malaysia by several researchers (Zainal Abidin, plantations. However, all triadimenol-based 1982). Vascular streak dieback causes cocoa fungicide is no longer available in Malaysian branches dieback with infections capable of market. Therefore, evaluation of alternative killing seedlings and mature cocoa trees of fungicide is important to reduce the impact of susceptible clone (Samuels et al., 2012). The this disease on cocoa cultivation. Besides symptoms includes chlorosis and necrosis of triadimenol, tebuconazole also has a good leaves, enlarged lenticels causing roughening of potential in controlling VSD in cocoa plantations the bark, three blackened vascular traces on leaf (Choong and Ng, 1990; Holderness, 1990). abscission scars and dark streaks within the However, very little information is available on vascular tissue (Guest and Keane, 2007). the effectiveness of tebuconazole in suppressing VSD in Malaysia as well as in other countries in

134

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

the world. Hence, the aim of this study was to determine the efficacy of tebuconazole against DR = [1− DT/DC] × 100 VSD in infected cocoa seedlings under nursery condition and the results will be further observed where: DR = disease reduction under field conditions. DT = disease percentages in treatments DC = disease percentages in control

MATERIALS AND METHODS RESULTS AND DISCUSSIONS The study was conducted under open nursery at Cocoa Research and Development Centre Tebuconazole is a triazol, sterol biosynthesis- (CRDC) Hilir Perak. Prior to initiation on this inhibiting and systemic fungicide, widely used study, the seeds of susceptible cocoa clone (PBC against various economically important diseases 131) were grown in polyethylene bags (22.86 cm of crops grown not only in the temperate but also x 30.48 cm) and placed around readily infected in the tropical and sub-tropical regions (Choong seedlings to give natural infection of VSD. After and Ng, 1990). The assessment of tebuconazole three months, the moderate pre-infected at the nursery stage was conducted with aims to seedlings were treated with different study the potential of this fungicide in reducing concentrations of fungicide which are 1 ppm, the severity of VSD on seedlings, since it 200 ppm and 400 ppm (Table 1), prepared in previously exhibited complete inhibition on distilled water. Foliar spray, approximately pathogen growth under in vitro condition 20ml/seedling was applied at 7 and 14 day (Suhaida et al., 2015). Figure 1 illustrates the intervals. disease reduction achieved by application of six treatments of fungicide at different intervals and Table 1: The application fungicide at different concentrations. Foliar application at the concentrations and intervals concentration of 200 ppm at 7 day intervals (T2) was the most effective, resulting in a disease Concentration reduction of 79.63% during the 8 months Interval Experiment (ppm) observation. However, higher concentration of 400 (recommended fungicide (400 ppm), T1 exhibited lower T1 by the manufacturer) reduction (77.8%) compare with 200 ppm (T2). 7 days T2 200 This might be effected by the distribution of T3 1 spores from the source to the treated seedlings. C1 Distilled water The spores of C. theobromae were airborne and 400 (recommended their distribution was related to the strength of T4 by the manufacturer) wind blowing. They can be spread only about 10 14 days T5 200 m from the source during slow winds and up to T6 1 182 m during strong winds (Anita-Sari and C2 Distilled water Susilo, 2013). Since the study was carried out under open nursery, the blowing of wind around The seedlings were manually watered nursery might be slightly infected the using tap water daily and individually assessed distribution of spores. However, disease for VSD symptoms monthly for 8 months. The reductions of both treatments (T1 and T2) were symptoms were scored from 0 to 6 (Table 2), not significantly different (p ≤ 0.05) with each according to Ahmad Kamil et al. (2006) with other. This concludes the application of minor modification by assessing the terminal tebuconazole at concentration suggested by flushes of seedling, instead of each branches in manufacturer or half of it is effective for each quarter of tree canopy. The severity rating controlling infected cocoa seedlings at interval was then transformed into percentage disease of 7 days. reduction and calculated using the equation by Omar et al. (2006) as follows:

135

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 2. Severity scale used for disease assessment of VSD

Severity Primary symptom severity of a flush Associated symptom Scale 0 Apparently uninfected or healthy Leaves glossy, healthy 1 One or two infected leaves; infected leaves showing Smooth bark of twigs or stem no early signs of symptoms-loss of glossiness or shine; swollen lenticels discrete brown vascular bundles on petiole of leaf scar or midrib of the leaf lamina. 2 Few infected leave, one showing or more showing Lenticels on bark may or may not be chlorosis in progress swollen 3 Some leaves infected; one infected leaf abscised; Lenticels on bark may or may not be one or more of the leaves showing chlorosis and swollen necrosis in progress 4 Two infected leaves abscised, some or all of the Lenticels on bark may or may not remaining leaves showing chlorosis and/or partial swollen necrosis 5 Three or more infected leaves have abscised; Lenticels usually swollen; fruit bodies remaining leaves infected, chlorotic or necrotic; may or not be present apparent cessation of growth (of first flush) 6 Near complete or complete defoliation from Lenticels usually swollen; fruit bodies abscission of infected leaves; dieback (first flush) in may or may not be present; may or progress as indicated by the drying of the twig/stem. may not be any proliferation of auxiliary shoots. The data of percentage disease reduction were subjected to one-way analysis of variance (ANOVA) and the means were compared using fisher test at p ≤ 0.05.

T1 – 7 day intervals (400 ppm) T2 – 7 day intervals (200 ppm) T3 – 7 day intervals (1 ppm) T4 – 14 day intervals (400 ppm) T5 – 14 day intervals (200 ppm) T6 – 14 day intervals (1 ppm)

Means followed by same letters are not significantly different (p ≥ 0.05)

Figure 1. Disease reduction of fungicide at different intervals and concentrations.

136

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

The study also noted that all that application of fungicide applied at 14 day concentrations of applied fungicide at 14 day intervals is not effective to control VSD. Since intervals (T4, T5 and T6) exhibited small the experiment was conducted on pre-infected reduction of VSD severity. T4, applied at cocoa seedlings, frequent application of concentration of 400 ppm only provided disease fungicide is required in order to cure the reduction of 29.63%, followed by 22.22% at 1 seedlings. This is to ensure continuous supply of ppm (T6) and 18.52% at 200 ppm. Disease fungicide that will be systemically absorbed into severity exhibited by these treatments were also plant tissue and subsequently killed the pathogen high and not significantly different with control (Fairbanks et al., 2000). except T4 (Figure 2). These results suggested

T1 – 7 day intervals (400 ppm) T2 – 7 day intervals (200 ppm) T3 – 7 day intervals (1 ppm) T4 – 14 day intervals (400 ppm) T5 – 14 day intervals (200 ppm) T6 – 14 day intervals (1 ppm) C1 – 7 day intervals (distilled water) C1 – 14 day intervals (distilled water)

Means followed by same letters are not significantly different (p ≥ 0.05)

Figure 2. Disease severity of fungicide at different intervals and concentrations.

CONCLUSIONS ACKNOWLEDGEMENTS

The study concluded that tebuconazole was The authors would like to thank the Director markedly has a good potential in controlling General of Malaysian Cocoa Board, Datin infected cocoa seedlings caused by VSD. Norhaini Udin; Director of Cocoa Upstream However, the efficacy of tebuconazole for Technology, Mr. Haya Ramba and staffs of suppressing VSD disease reduced with Pathology Unit, Cocoa Research and prolonging the interval of foliar application. Development Centre (CRDC) Hilir Perak. Therefore, frequent application of fungicide is required in order to making sure complete disease suppression. The study will be further REREFNCES extended on mature cocoa trees in the near future. Ahmad Kamil, M.J,. Kelvin, L., Sapiyah, S., Lee, M.T., Shari Fuddin, S., Albert, L., Francis, A. and Bong, C.L. (2006). Evaluation of cocoa resistance to vascular streak dieback in Malaysia. In Global

137

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Approaches to Cocoa Germplasm Malaysia. In Proceeding of the Cocoa Utilization and Conservation, Final report and Coconuts Conference, Incorporated of the CFC/ICCO/IPGRI project on Society of Planters. Editors: Wastie, R.L. “Cocoa Germplasm Utilization and and Earp, D.A, Kuala Lumpur. pp 50-70. Conservation: a Global Approach” (1998- 2004). Editors: Eskes, A. B. and Efron, Y. Keane, P.J. Flentje, N.T and Lamb, K.P. (1972). CFC, Amsterdam, The Investigation of Vascular-Streak Netherlands/ICCO, London, UK/IPGRI, Dieback of Cocoa in Papua New Guinea. Rome, Italy. pp. 191–199. Australian Journal of Biological Sciences, 25: 553-564. Anita-Sari, I. and Susilo, A.W. (2013). Investigation of different characters of Omar, I., O’Neill, T.M. and Rossall, S. (2006). stomata on three cocoa clones with Biological control of fusarium crown resistance level difference to VSD and root rot of tomato with antagonistic (vascular-streak dieback) disease. Journal bacteria and integrated control when of Agricultural Science and Technology combined with the fungicide A, 3: 703–710. carbendazim. Plant Pathology, 55: 92– 99. Choong, C.T. and Ng, K.Y. (1990). A comparative evaluation of triadimenol Samuels, G.J., Ismaiel, A., Rosmana, A., Junaid, with tebuconazole, a new fungicide for M., Guest, D., McMahon, P., Keane, P., the control of VSD (vascular streak Purwantara, A., Lambert, S., Rodriguez- dieback) of cocoa in Malaysia. In Carres, M. and Cubeta, M.A. (2011). Proceedings of the 3rd International Vascular streak dieback of cacao in Conference on Plant Protection in the Southeast Asia and Melanesia: in planta Tropics: volume III, Malaysian Plant detection of the pathogen and a new Protection Society (Malaysia), Pahang, taxonomy. Fungal Biology, 116:11–23. Malaysia, 20-23rd Mar 1990, p. 50-53. Suhaida, S., Shari Fuddin, S. and Alias, A. Fairbanks, M.M., Hardy, G.E.St.J. and (2015). In vitro screening of selected McComb, J.A. (2000). Comparisons of fungicides for controlling vascular streak phosphite concentrations in Corymbia dieback disease of cocoa (Theobroma (Eucalyptus) calophylla tissues after cacao). In Proceeding International spray, mist or soil drench applications Congress of the Malaysian Society for with the fungicide phosphite. Microbiology (ICMSM 2015), Penang, Australasian Plant Pathology, 29: 96– Malaysia, 7-10th December 2015. 101. Zainal Abidin, M.A. (1982). Vascular streak Guest, D. and Keane, P. (2007). Vascular-Streak dieback of cocoa in West Malaysia. Dieback: A new encounter disease of Published master dissertation, Universiti cacao in Papua New Guinea and Pertanian Malaysia. Southeast Asia caused by the obligate basidiomycete Oncobasidium theobromae. Phytopathology, 97: 1654– 1657.

Holderness, M. (1990). Control of vascular- streak dieback of cocoa with triazole fungicides and the problem of phytotoxicity. Plant Pathology, 39: 286- 293.

Keane, P.J. and Turner P.D. (1971). Vascular Streak die-back of cocoa in West

138

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

INHIBITION ABILITY OF COCOA POD EXTRACT ON TYROSINASE ACTIVITY Azila, A.K.1, NurAzilah, A.1 and Azrina, A.2,3 1Malaysian Cocoa Board, Cocoa Innovation & Technology Centre, Lot PT12621, Nilai Industrial Area, 71800 Nilai, Negeri Sembilan. 2Halal Products Research Institute, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor 3Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor

Malaysian Cocoa Journal 9(1): 139-150 (2016) ABSTRACT – Cocoa pod is a waste from cocoa industry. About 75 percentage weight of cocoa fruit is pod. The pods have variety of colors which resemble the various active compounds with potential health benefits. Normally, the cocoa pods were utilized as fertilizers, source of potash and activated carbon. However, the application of cocoa pods as cosmeceutical ingredients has not been extensively explored and documented. The objective of this study was to investigate the effects of cocoa pod extract on tyrosinase activity where good suppression of tyrosinase activity may indicate its potential to be used as skin whitening agent. In tyrosinase assay, the effects of cocoa pod extract on the activity of tyrosinase from mushroom to oxidize L-DOPA substrate were measured. Results showed the cocoa pod extract had better performance of tyrosinase reduction activity at almost two-folds compared with kojic acid and ascorbic acid. Hence, cocoa pod extract has the potential to be used as skin whitening active ingredients.

Key words: Cocoa pod extract, Cosmeceutical, Skin whitening, Tyrosinase, L-DOPA.

INTRODUCTION

Various racial skin colors are influenced by the Tyrosinase activity rate of tyrosinase enzyme, in addition to the melanocytes cells size and distribution (Parvez et al., 2006). Lower activity rate of tyrosinase enzyme leads to lighter skin color. Tyrosinase On the other hand, the darker skin colors have melanocytes cells with longer, thicker and branched dendrites, as well as higher tyrosinase Cysteine, activity rate compared to the white skin. The Glutathione stratum corneum layer of the darker skin contains melanin while absence in the white skin (Rana et al., 1996). Excessive exposure to UV- light can caused skin pigmentation problems such as melasma while certain chemicals and TRP1, drugs can lead to hyperpigmentation (Lin et al., TRP2 2008).

The melanin synthesis pathway was elucidated in detail by Lin et al. (2008) and molecularly by Kim and Uyama (2005). In Figure 1: Melanin Pathway (Lin et al., 2008; brief, tyrosinase is the primary enzyme involved Parvez et al., 2006; Kim and Uyama, 2005) producing melanin in lower part of epidermis. Two steps of oxidation reaction involving The dopaquinone, was then tyrosinase enzyme in melanin production polymerized into eumelanin or pheomelanin (Figure 1). Initially, formation of melanin depending on the type of enzymes involved. started with oxidation of the L-tyrosine Eumelanin (darker melanin; brown and black (monophenol) to L-DOPA (o-diphenol; 3,4- skin) is produced when tyrosinase-related dihydroxyphenylalanine) and further to proteins (TRP-1 and TRP-1) metabolized the dopaquinone (Kim and Uyama, 2005). dopaquinone (Lin et al., 2008). On the other

139

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

hand, pheomelanin (lighter melanin; red and noncompetitive inhibitor also binds to the yellow) is formed with the presence of cysteine enzyme but not on the specific substrate-enzyme and glutathione enzymes (Rana et al., 1999). binding site. Another inhibition mode is Subsequently, the melanins were shifted to uncompetitive inhibitor, where the Km value keratinocytes layer or upper part of epidermis remains unchanged or decreases and the Vm (Parvez et al., 2006). However, the value decreases with the presence of inhibitor. pheomelanins degraded upon the keratinization An uncompetitive inhibitor only binds to the process while the eumelanin sloughed off from enzyme-substrate complex but not to the free the skin by natural desquamation (Rana et al., enzyme as exhibited by the competitive and 1996). noncompetitive inhibitor. The mixed type mode of inhibitor has the increasing value of Km with Kim and Uyama (2005) mentioned that reducing value of Vm, where the inhibitor can melanin synthesis can be inhibited by avoiding binds both the enzyme-substrate complex and UV exposure; inhibition of tyrosinase, free enzyme. melanocyte metabolism and proliferation; or removal of melanin. Plant extracts inhibit In order to be categorized as tyrosinase activity by various mechanisms, such competitive inhibitor, the plant extract as being a competitor, non-competitor, presumably contain compounds with similar uncompetitor or mixed mode to the substrate structure or side of reaction to tyrosine or L- (tyrosine or L-DOPA). As a competitor, plant DOPA, such as 3’,5’-di-C-β- extract binds to tyrosinase-active site and glucopyranosylphloretin in the extract of reduces the activity of the enzyme and immature calamondin peel (Lou, 2012). consequently lowers the production of melanin According to Kim and Uyama (2005), plant resulting fairer skin. According to Bowden polyphenols, especially flavonoids, possessed (1976), the classification of mode of inhibition inhibition ability against tyrosinase enzyme due was interpreted by the affinity strength (Km) and to the similarity between the dihydroxyphenyl maximum velocity (Vm) values (Table 1) group in L-DOPA (Figure 2) and the α-keto obtained from the double reciprocal plot. group in flavonoids. Other than structure similarity to the L-DOPA, flavonoids have the Table 1. Mode of enzyme inhibition based on ability to form complex metal ion chelating with double reciprocal plot (Bowden, 1976) the tyrosinase which is the copper-containing enzyme and disrupt its activity (Malešev and Mode Km Vm Kuntić, 2007; Morel et al., 1993). In this study, Competitive Increase Increase or the effect of cocoa pod extracts compared to remain other positive controls in inhibiting the Non- Reduce or Reduce tyrosinase enzyme from oxidizing L-DOPA competitive remain substrate was measured. Uncompetitive Reduce Increase

Mixed Increase Reduce

Km is the inverse measure of the affinity or strength of binding between the enzyme and its substrate, Vm is the maximum velocity

Pál (2013) explained in details the mode of enzyme inhibition. In brief, for Figure 2: L-DOPA Structure competitive inhibitor mode, the Km value increases with the presence of inhibitor (sample) (Parvez et al., 2005) whereby the Vm value remains unchanged or increase. Competitive inhibitor competes with Cocoa pod is an abundant waste of the substrate for the binding sites on the enzyme cocoa plantation. About 70-75% of the cocoa and therefore it usually has similar structure to fruit’s weight was amounted by the pod the substrate. On the other hand, for (Adzimah and Asiam, 2010; Owusu-Domfeh, noncompetitive inhibitor, the Km value reduces 1972). Cocoa pods are usually discarded in bulk or remains with reducing value of Vm. The to plantation floor to serve as fertilizer (Agbeniyi

140

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

et al., 2011). Other application of cocoa pods prepared by diluting 80 mL of denatured ethanol were as activated carbon (Ahmad and Wan (GmbH, Hamburg, Germany) with 20 mL of Zainal, 2009; Ahmad and Radzi, 2005), gum distilled water. (Samuel, 2006), food colorant (Figueira et al., 1993), anti-hypercholestromia supplement One gram of cocoa pod powder was (Sartini and Gemini, 2008), and source of soaked in 20 mL of aqueous ethanol in 50 mL potash. Cocoa pod in powder form has been conical flask with cap. The flask was placed in water bath shaker (BS-21, Lab Companion, used as composite material (Chun et al., 2014), o fibre in bread (Amir et al., 2013) and exfoliating Mapo-Gu, Seoul, Korea) for 30 min, 35 C at 120 agent (Karim and Abdullah, 2012). The cocoa rpm. Decanted portion was poured into tared pods composed of 32.3% carbohydrate, 21.4% pear-shaped flask after filtration with Whatmann lignin, 19.2% sugars, 8.6% protein and 27.7% filter No.1 to remove the remaining floating minerals (Vriesmann et al., 2011; Owusu- material. Solvent was removed using rotary Domfeh, 1972). The phenolic content of cocoa vacuum evaporator (IKA, Staufenim Breisgau, pod extract was reported by Vriesmann et al. Germany), until dry. The dried extract was (2011) at 45.6-46.4 mg gallic acid equivalent freeze-dried for at least 24 h (FDA8512, (GAE). Sartini and Gemini (2008) reported that Labconco, Kansas City, Missouri, USA). The cocoa pod contained 56.5+0.57 mg GAE/g of freeze-dried crude extract sample was re- phenolic content. In previous study, the cocoa dissolved in aqueous ethanol (80% v/v) to obtain pod extract was determined to contain 1000 µg/mL and was made into serial dilution of 49.54+3.69 mg GAE/g (Karim et al., 2014a). 500, 250, 125, 62.5, 31.1, 15.5 and 7.8 µg/mL, The aim of this study was to determine the effect before being subjected to the analysis. of cocoa pod extract on the assay of tyrosinase Ascorbic acid - Ascorbic acid (AA) was used as enzyme activity which associated to skin a positive control (Figure 3). Ascorbic acid has whitening. an effective role in maintaining cellular function (Padh, 1990; Murad et al., 1983) which affecting the skin condition. Ascorbic acid and its MATERIALS AND METHODS derivatives, including ascorbic 2-phosphate, induce collagen accumulation in fibroblasts Sample preparation (Yamamoto et al., 1992; Hata and Senoo, 1989; Geesin et al., 1988; Murad et al., 1983), increase Cocoa pod – The discarded cocoa pods were collagen synthesis with prolonged exposure to collected from the Research Station of the AA (Kishimoto et al., 2013) and promote topical Malaysian Cocoa Board in Jengka and labeled as treatment to the photo-damaged skin (Humbert CPE. In addition, another two samples of cocoa et al., 2003). L-(+)-ascorbic acid was acquired pod were collected by clones (PBC123 and from Sigma-Aldrich, St. Louis, Missouri,USA. MCB C1). PBC123 was selected because it is One milligram of ascorbic acid powder was the most planted clone, while MCB C1 is one of dissolved in ethanol to obtain concentration of the latest clones developed by the Malaysian 1000 µg/mL and serial dilution of 500, 250, 125, Cocoa Board. 62.5, 31.1, 15.5 and 7.8 µg/mL were prepared.

All cocoa pods were washed thoroughly with tap water. Mechanical fruit slicer (FC-312, Zhaoqing Fengxiang Food Machinery, Guangdong, China) were used to slice the pods into a smaller pieces to aid drying process using a high-performance dryer (FD-825, Protech, Selangor, Malaysia). The dried cocoa pods were ground into fine particle by using grinding machine to 1 mm mesh size (Automatic Hammer Figure 3: Ascorbic Acid Structure Mill Grinder, Guangdong, China). Extraction (www.pubchem.com) was carried out using 80% ethanol (Sanbogi et al., 1998). Aqueous ethanol (80% v/v) was

141

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Kojic acid - Kojic acid (MW 142.11g/mol, Sigma-Aldrich, Figure 4) in same serial dilution with ascorbic acid diluted in dimethylsulfoxide (DMSO, Merck Milipore, Darmstadt, Germany) was used as a positive control. Next, the inhibition percentage or reduction activity of the sample against tyrosinase enzyme is calculated by subtracting the remaining activity value from 100%. The inhibition concentration at 50% (IC50) was determined by linear regression of plotted data in the graph of reduction percentage versus concentration of the sample. The lower the IC50 value of the sample indicated the higher of its

Figure 4: Kojic Acid Structure inhibition capability against tyrosinase activity. (Bentley, 2006) To determine the mode of inhibition (competitive, non-competitive, uncompetitive inhibitor or mixed) of the samples, L-DOPA Tyrosinase assay substrate at various concentrations (0.49, 0.98, Tyrosinase reduction activity of the samples was 1.96, 3.92, 7.84, 15.68 and 31.36 µg/mL) were determined using method by Sabudak et al. mixed with each of the samples (at 490 µg/mL) (2013). Tyrosinase from mushroom was and the tyrosinase enzyme solution (1460 purchased from Sigma-Aldrich (1,715 U/mg µg/mL). The absorbance was monitored at 450 solid, 14.58 mg whitish powder). About 1.46 nm for 20 minutes. Mode or type of tyrosinase mg of this tyrosinase powder was dissolved in 1 inhibition by the samples or extract was mL of 50 mM phosphate buffer solution (PBS, analyzed using Lineweaver-Burk or double 10x, R&M Chemicals, Mississauga, Ontario, reciprocal plot (Eisenthal and Bowden, 1974) as Canada) to achieve concentration in liquid at in equation (2); 2,500 U/mL. In another vial, L-DOPA substrate solution at 2.5 mM was prepared by dissolving (2) 4.9 mg of L-3,4-dihydroxyphenylalanine (L- DOPA, MW 197.19 g/mol, Sigma Aldrich) in 10 Where 1/Vo is the reciprocal rate, where Vo was mL of PBS. obtained from the changes of the optical density A 20 µL of each prepared tested per minute (∆OD/min). Km is the inverse solution were pipetted into 96-microplate well measure of the affinity or strength of binding and 138 µL of PBS was added to each well. between the enzyme and its substrate, Vm is the Plate was incubated for 90 min at 37oC after maximum velocity and 1/S is the reciprocal of addition of 2 µL tyrosinase solution. Finally, 40 substrate concentration. µL of substrate solution was added to each well Statistical analysis before monitoring the absorbance at 450 nm for The results were presented as mean + standard 20 min. Negative control or labeled as no deviation and were determined for triplicates of inhibitor contained substrate solution, tyrosinase three independent samples. Linear regression to enzyme and PBS. determine the slope of tyrosinase remaining activity for the samples were accepted when the Slopes of the remaining tyrosinase 2 activity of the samples at the concentration range value of R-squared (R ) was higher than 0.8 and measured in this study (1000, 500, 250, 125, were plotted using Microsoft Excel 7.0. Minitab 62.5, 31.1, 15.5 and 7.8 µg/mL) were plotted Software version 14.12.0 (State College, based on the optical density recorded for 20 min. Pennsylvania, USA) was used to carry out the Then, the tyrosinase inhibition percentage was statistical analysis. The results of inhibition calculated as percentage of remaining activity as activity, IC50, maximum velocity (Vm) and in equation (1); affinity strength (Km) were compared using

142

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

ANOVA (Tukey test) and significantly different exhibited by the concentration range in this when p-value <0.05 (α=0.05). study.

Ranking for tyrosinase inhibition in this RESULTS study based on IC50 value was CPE > KA > PBC123 > AA (Table 4). The MCB C1 showed The slopes of remaining tyrosinase activity for saturated value of inhibition against tyrosinase the samples at the selected concentration range (Table 3), therefore, the IC50 could not be in this study (Table 2) were obtained from the determined. plotted optical density versus minutes (data not shown). Slope value equals almost to zero The mode of inhibition was interpreted indicated high reduction activity against by the Km and Vm value as in Table 4. The result tyrosinase enzyme. At concentration of 1000 shows that CPE was a mixed type inhibitor µg/mL, CPE (0.0023 + 0.0004) has the lowest where the Km (0.3322 + 0.0001 μg/mL) higher slope value compared to other tested solution and lower Vm (0.0176 + 0.0021 OD/min) including kojic acid at similar concentration compared to the initial stage with the presence of (KA; 0.0052 + 0.0005). Low concentration of CPE as in Figure 5 (a). Mode inhibition of AA CPE and KA (at 31.1-7.8 μg/mL) showed higher as in Figure 5 (b) against tyrosinase enzyme is slope values than the negative control, indicating uncompetitive as the Km value (0.1508 + 0.0109 low inhibition activity against tyrosinase μg/mL) is similar to the value of negative control enzyme. The AA obtained higher slopes values with lower of Vm (0.0499 + 0.0090 OD/min). than the negative control, despite of its KA (0.4503 + 0.1323 μg/mL) acted as concentration, indicating low tyrosinase competitive inhibitor based on the significantly inhibition activity. higher value of Km compared to the negative control (no inhibitor, 0.1037 + 0.0715 μg/mL). The slopes were interpreted into Meanwhile, the inhibition mode of PBC123 and percentage against the negative control (no MCB C1 extracts was uncertain due to the inhibitor) to obtain the reduction activity against negative slope obtained by Lineweaver-Burk tyrosinase (Table 3). Inhibition against plot. The probable explanation was that the tyrosinase enzyme activity at low concentration inhibition was irreversible mode (Chang, 2005) was almost similar for CPE and KA. However, where the affinity strength value was less than the pattern of inhibition activity was gradually 0.1. According to Chang (2005), the irreversible increased for CPE. At high concentration of mode of enzyme inhibition can also be observed CPE (1000 μg/mL), the percentage of inhibition by the saturated value of reduction tyrosinase against tyrosinase enzyme activity was 86.51 + activity as exhibited by MCB C1 and sigmoidal 2.12% which was better than KA (1000 μg/mL) curve as shown by PBC123 in Table 3. at 68.80 + 2.89%. For AA, smaller reduction . activity compared to other samples was

143

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 2. The slope of remaining tyrosinase activity due to presence of the samples compared to negative control

Concentration CPE PBC 123 MCB C1 AA KA No inhibitor µg/mL 7.8 0.0140 +0.0011 0.0138 +0.0012 0.0132 +0.0007 0.0153 +0.0011 0.0137 +0.0023 0.0146 +0.0004 15.5 0.0131 +0.0004 0.0135 +0.0005 0.0135 +0.0005 0.0155 +0.0007 0.0170 +0.0010 0.0138 +0.0019 31.1 0.0139 +0.0034 0.0141 +0.0005 0.0142 +0.0010 0.0143 +0.0025 0.0148 +0.0003 0.0133 +0.0002 62.3 0.0121 +0.0005 0.0139 +0.0001 0.0138 +0 0.0144 +0.0014 0.0147 +0.0014 0.0149 +0.0005 125 0.0097 +0.0012 0.0131 +0.0005 0.0133 +0.0001 0.0147 +0.0029 0.0127 +0.0008 0.0137 +0.0016 250 0.0063 +0.0014 0.0116 +0.0008 0.0123 +0.0015 0.0125 +0.0017 0.0080 +0.0013 0.0147 +0.0005 500 0.0057 +0.0007 0.0069 +0.0016 0.0128 +0.0014 0.0136 +0.0013 0.0087 +0.0007 0.0136 +0.0014 1000 0.0023 +0.0004 0.0059 +0 0.0126 +0.0006 0.0117 +0.0003 0.0052 +0.0005 0.0152 +0.0021 Values which are expressed in mean + standard deviation are from triplicates of three independent samples, using linear regression analysis, accepted when R2 value is at least 0.8 or higher. CPE is cocoa pod extract collected in bulk, PBC123 is Perang Besar clone, MCB C1 is Malaysian Cocoa Board clone, AA is ascorbic acid and KA is kojic acid. Kojic acid is positive control and negative control (no inhibitor) is the solution containing the substrate and tyrosinase enzyme.

Table 3. Reduction of tyrosinase activity in the presence of the samples

Concentration CPE PBC 123 MCB C1 AA KA µg/mL 7.8 16.17 +1.59aA 17.33 + 0.81acA 79.16 +3.88bAG 8.29 +0.48d A 18.07 + 0.25cA 15.5 21.52 +1.37abcB 18.86 + 3.06aAB 80.66 +3.19bcFG 7.25 +0.05d A 22.44 + 0.04cB 31.1 16.97 +2.08adA 15.75 + 2.76adAB 85.27 +5.82bEFG 14.15 +1.72d B 11.17 +1.71cdC 62.3 27.78 +1.90aC 16.65 +0.86bAB 82.43 +0.01cDFG 13.51 +0.84d B 11.71 +1.27eC 125 41.99 +4.13a D 21.42 +2.90bB 79.83 +0.38cCD 11.84 +0.28d C 24.12 +4.89bB 250 62.23 +0.28a E 30.76 + 4.75bC 73.98 +8.82cBC 25.26 +0.53d D 52.22 +7.46eD 500 65.87 +2.51a F 54.17 + 7.76bD 77.25 +8.35cΑCDF 18.61 +1.62d E 47.78 +4.07eE 1000 86.51 +1.27a G 64.39 + 1.00bE 75.66 +3.56cACDFG 29.62 +0.90d F 68.80 +2.89eF Values are expressed in mean + standard deviation, triplicates of three independent samples, a b c d e denotes significantly different between sample in row for respective concentration, using ANOVA (Tukey test) at α=0.05, A B C DE F denotes significantly different between concentration for each sample in respective column, using ANOVA (Tukey test) at α=0.05. CPE is cocoa pod extract collected in bulk, PBC123 is Perang Besar clone, MCB C1 is Malaysian Cocoa Board clone, AA is ascorbic acid and KA is kojic acid.

144

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 4. The IC50 and inhibition mode of CPE, AA and KA

Sample IC50 (μg/mL) Km (μg/mL) Vm (OD/min) Inhibition mode no inhibitor - 0.1037 +0.0715a 0.0406 +0.0051a - CPE 357.95 +1.89a 0.3322 +0.0001b 0.0176 +0.0021b Mixed PBC123 629.34 +2.99b -0.1627 +0.1047c 0.0393 +0.0016c irreversible1 MCB C1 - -0.1992 +0.0574c 0.0363 +0.0011c irreversible1 AA 670.82 +0.80c 0.1508 +0.0109a 0.0499 +0.0090d uncompetitive, Sakuma et al. (1999) KA 572.28 +1.68d 0.4503 +0.1323d 0.0478 +0.0040d competitive, Chang (2009)

Values are expressed in mean + standard deviation, triplicates of three independent samples, a b c d denotes significantly different between sample in column for respective parameter, using ANOVA (Tukey test) at α=0.05, IC50 is the concentration of sample that can inhibit the tyrosinase activity to 50%, Km is the inhibitory constant and Vm is the maximal velocity obtained using Lineweaver-Burk plot, 1 based on the saturated and sigmoidal of reduction activity value exhibited by MCB C1 and PBC123 (Table 3), respectively (Chang, 2005). CPE is cocoa pod extract collected in bulk, PBC123 is Perang Besar clone, MCB C1 is Malaysian Cocoa Board clone, AA is ascorbic acid and KA is kojic acid. No inhibitor is mixture of L-DOPA (substrate) and tyrosinase enzyme in the solution.

145

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

(a) (b)

Figure 5: Lineweaver-Burk Plots of Tyrosinase Reaction and its Inhibition by (a) CPE and KA showing the mixed and competitive mode to the negative control (no inhibitor), respectively; and (b) AA showing the uncompetitive mode where the AA exhibits almost parallel line to the negative control (no inhibitor).

146

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

DISCUSSIONS inhibition in melanin pathway. KA exhibited competitive mode with tyrosinase in the early CPE was reported earlier to contain malic acid, stage of melanin pathway and reacted by metal- rosmarinic acid, procyanidin, apigenin and chelating mechanism against tyrosinase enzyme dimethyl ellagic acid (Karim et al., 2016). In this (Aytemir and Karakaya, 2012). Consequently, study, the CPE was determined to have a mixed the rate of reaction became slower towards the type of tyrosinase inhibition probably due to end of melanin pathway due to the mixed type of these various compounds that have different inhibition (Cabanes et al., 1994). However, KA mechanism of inhibition. For example, malic could induce sensitization on skin (Draelos, acid was reported to inhibit the tyrosinase 2008) but at concentration of 1%, KA was the enzyme activity in non-competitive mode (Qiu best lightening agent (Aytemir and Karakaya, et al., 2015). Son et al. (2001) reported that 2012). malic acid, a hydroxycarboxylic compound, has medium tyrosinase inhibitory effect in comparison with tartaric acid and acetic acid. CONCLUSIONS On the other hand, rosmarinic acid induced the tyrosinase activity as studied by Lee Based on the above discussion and the results et al. (2007) on B16 mouse melanoma cells. obtained in this study, it was found out that the Further investigation by Oliveira et al. (2012) on CPE exhibited better performance to reduce the the effect of rosmarinic acid to tyrosinase tyrosinase activity compared to KA and AA, enzyme, at low concentration, the melanin considering the synergistic effect of various formation process induced while at high compound in CPE compared to single effect of concentration the tyrosinase activity reduced. kojic acid or ascorbic acid. Therefore, the CPE The inhibition of rosmarinic acid against has the potential to be used as skin whitening tyrosinase enzyme was found to be agent. noncompetitive mode (Lin et al., 2011). Procyanidin is trimer of catechin, a flavonol compound of polyphenolic. Flavonol has the ACKNOWLEDGEMENTS flavonoid structure with the hydroxyl group at the B-ring structure which exhibited the This project was funded by the Temporary inhibition against tyrosinase by metal chelating Research Fund (TRF 15272) of Malaysian mechanism similar to KA (Aytemir and Cocoa Board and supported by Universiti Putra Karakaya, 2012). Procyanidin trimer from apple Malaysia under grant GPIPS/2013/9399848 of inhibited the tyrosinase activity effectively Halal Products Research Institute. Thank you (Shoji et al., 2005) and found to be mixed mode for those who involved in data collection. type (Chen et al., 2014).

The AA acted as antioxidant agent on REFERENCES the later stage of melanin pathway, where the formation of dopachrome, an intermediate form Adzimah, S.K. and Asiam, E.K. (2010). Design of eumelanin, was converted back to of a cocoa pod splitting machine. Res. dopaquinone (Kim and Uyama, 2005). Sakuma J. Appl. Sci. Eng. Technol. 2(7): 622- et al. (1999) reported that AA has uncompetitive 634. inhibition mode against tyrosinase, where the AA tend to bind on the enzyme-substrate Agbeniyi, S.O., Oluyole, K.A. and Ogunlade, complex but not to the free tyrosinase enzyme in M.O. (2011). Impact of cocoa pod husk the assay. This explained the low activity of AA fertilizer on cocoa production in Nigeria. in the tyrosinase assay used in this study. World J. Agric. Sci.7(2): 113-116. KA bound to the active site of Ahmad, F. and Wan Zainal, W.N.H. (2009). tyrosinase enzyme that was supposed to bind the Adsorption isotherms and kinetics studies substrate, thus lowering the tyrosinase activity of methylene blue on cocoa pod husk (Draelos, 2008). Chang (2009) mentioned that based activated carbon. Malaysian KA can behave in two different modes of Cocoa J. 5: 55-62.

147

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Ali A, Akhtar N, Chowdhary F. (2014). Eisenthal, R. and Bowden, A.C. (1974). The Enhancement of humanskin facial direct linear plot: a new graphical revitalization by moringa leaf extract procedure for estimating enzyme kinetic cream. Postep. Derm. Alergol. 2: 71–6. parameters. Biochem. J. 139: 715-720. Amir, I.Z., Hanida, H.S. and Syafiq, A. (2013). Fischer, T.W., Alberti, W.E. and Elsner, P. Development and physical analysis of (1999). Direct and non-direct high fiber bread incorporated with cocoa measurement techniques for analysis of (Theobroma cacao sp.) pod husk powder. skin surfacetopography. Skin Pharmacol Int. Food Res. J. 20(3): 1301-1305. Appl Skin Physiol.12:1–11. Aytemir, M. D. and Karakaya, G. (2012). Geesin, J.C., Darr, D., Kaufman, R., Murad, S. Kojicacidderivatives. In Medicinal and Pinnel, S.R. (1988). Ascorbic acid Chemistry and Drug Design specifically increases type I and type III (DenizEkinci (Ed.), ISBN: 978-953-51- procollagen messenger RNA levels in 0513-8, InTech, Available human skin fibroblasts. J. Invest. from:http://www.intechopen.com/books/ Dermatol. 90: 420–424. medicinal-chemistry-and-drug- design/kojic-acid-derivatives. Hata, R.I. and Senoo, H. (1989). L-ascorbic acid 2-phosphate stimulates collagen Bentley, R. (2006). From miso, sake and shoyu accumulation, cell proliferation, and to cosmetics: a century of science for formation of a three-dimensional tissue kojic acid. Nat. Prod. Rep. 23: 1046- like substance by skin fibroblasts. J. Cell 1062. Physiol. 138(1): 8–16. Bowden, A.C. (1976). Chapter 4: Inhibitors and Humbert, P.G., Haftek, M., Creidi, P., Lapière, activators. In Principles of Enzyme C., Nusgens, B., Richard, A., Schmitt, D., Kinetics. pp.52-70. Rougier, A. and Zahouani, H. (2003). Topical ascorbic acid on photoaged skin. Cabanes, J., Chazarra, S. and Carmona, F.G. Clinical, topographical and ultrastructural (1994). Kojic Acid, a cosmetic skin evaluation: double-blind study vs. whitening agent, is a slow-binding placebo. Exp. Dermatol. 12(3): 237–244. inhibitor of catecholase activity of tyrosinase. J. Pharm. Pharmacol. 46: Karim, A.A.; Abdullah, N. A. (2012). Stability 982-985. study of cocoa butter body scrub formulated with ground cocoa pod husk. Chang, R. (2005). Chapter 10: Enzyme Malaysian Cocoa J. 7: 42-47. Kinetics. In Physical Chemistry for the Biosciences. U.S.A: University Science Karim, A.K., Azlan, A., Ismail, A., Hashim, P. Books, California. pp. 363- 400. and Abdullah, N.A. (2014a). Comparison of antioxidant properties of cocoa pods Chang, T. (2009). An updated review of and shells. Malaysian Cocoa J. 8: 49-56. tyrosinase inhibitors. Int. J. Mol. Sci. 10(6): 2440-2475. Karim, A.K., Azlan, A., Ismail, A., Hashim, P., Gani, S.S.A., Zainudin, B.H. and Chen, X., Shi, Y., Chai. W.M., Feng, H.L. Abdullah, N.A. (2014b). Phenolic Zhuang, J.X. and Chen, Q.X. (2014). composition, antioxidant, anti-wrinkles Condensed tannins from Ficus virens as and tyrosinase inhibitory activities of tyrosinase inhibitors: structure, inhibitory cocoa pod extract. BMC Complement. activity and molecular mechanism. PLoS Altern. Med. 14, 381. doi:10.1186/1472- ONE 9(3): e91809. doi:10.1371/ 6882-14-381. journal.pone.0091809 Karim, A.K., Azlan, A., Ismail, A., Hashim, P., Draelos, Z.D. (2008). The cosmeceutical realm. Gani, S.S.A., Zainudin, B.H. and Clin. Dermatol. 26: 627-632. Abdullah, N.A. (2016). Efficacy of cocoa pod extract as antiwrinkle gel on

148

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

human skin surface. J. Cosmet. OliveiraK.B.., Palú, E., Santos, A.M.W. and Dermatol.doi: 10.1111/jocd.12218. Santos, B.H. (2012). Influence of rosmarinic acid and Salvia officinalis Kim, Y.J. and Uyama, H. (2005). Tyrosinase extracts on melanogenesis of B16F10 inhibitors from natural and synthetic cells. Rev. Bras. Farmacogn. 23(2): 249- sources: structure, inhibitor mechanism 258. and perspective for the future. Cell Mol. Life Sci. 62: 1707-1723. Owusu-Domfeh, K. (1972). The future of cocoa and its by-products in the feeding of Kishimoto, Y., Saito, N., Kurita, K., Shimokado, livestock. Ghana J. Agric. Sci. 5: 57-64. K., Maruyama, N. and Ishigami, A. (2013). Ascorbic acid enhances the Padh, H. (1990). Cellular functions of ascorbic expression of type 1 and type 4 collagen acid. Biochem. Cell Biol. 68(10): 1166- and SVCT2 in cultured human skin 1173. fibroblasts. Biochem. Biophysic. Res. Comm. 430: 579-584. Pál, G. (2013). Chapter 9: Enzyme kinetics. In Introduction to Practical Biochemistry. Lee, J., Kim, Y.S. and Park, D. (2007). Eötvös Loránd University. Available Rosmarinic acid induces melanogenesis online. through protein kinase A activation signaling. Biochem. Pharmacol. 74(7): Parvez, S., Kang, M., Chung, H., Cho, C., Hong, 960-968. M., Shin, M. and Bae, H. (2006). Survey and mechanism of skin depigmenting and Lin, J., Chiang, H., Lin, Y. and Wen, K. (2008). lightening agents. Phytother. Res. 20: Natural products with skin-whitening 921-934. effects. J. Food Drug Anal. 16(2): 1-10. Qiu, A., Gou L. and Gou, S. (2015). Study of Lin, L., Dong, Y., Zhao. H., Wen, L., Yang, B. inhibition effect on L-malic acid for and Zhao, M. (2011). Comparative tyrosinase. Food Sci. Technol. 2010- evaluation of rosmarinic acid, methyl 2011. rosmarinate and pedalitin isolated from Rabdosia serra (MAXIM.) HARA as Rana, B. K., Hewett-Emmett, D., Li, J., Chang, inhibitors of tyrosinase and α- B.H., Sambuughin, N., Lin, M., Watkins, glucosidase. Food Chem. 129: 884-889. S., Bamshad, M., Jorde, L.B., Ramsay, M., Jenkins, T. and Li, W.H. (1999). Malešev, D. and Kuntić, V. (2007). High polymorphism at the human Investigation of metal-flavonoid chelates melanocortin 1 receptor locus. Genetics. and the determination of flavonoids via 151: 1547-1557. metal-flavonoid complexing reactions. J. Serb. Chem. Soc. 72 (10): 921–939. Rana, B. K., Hewett-Emmett, D. and Li, W.H. (1996). Variation in the human Morel, I., Lescoat, G., Cogrel, P., Sergent, O., melanocyte stimulating hormone receptor Pasdeloup, N., Brissot, P., Cillard, P. and locus (Abstr.). Pigm. Cell Res. 60: Cillard, J. (1993). Antioxidant and iron- (Suppl. 5). chelating activities of the flavonoids catechin, quercetin and diosmetin on Sanbogi, C., Osakebe, N., Natsume, M., iron-loaded rat hepatocyte cultures. Takizawa, T., Gomi, S. and Osawa, T. Biochem Pharmacol, 45: 13–19. (1998). Antioxidative polyphenols isolated from Theobroma cacao. J. Murad, S., Tajima, S., Johnson, G.R., Sivarajah, Agric. Food Chem. 46: 454-457. A. and Pinnell, S.H. (1983). Collagen synthesis in cultured human skin Shoji, T., Masumoto, S., Moriichi, N., Kobori, fibroblasts: Effect of ascorbic acid and its M., Kanda, T., Shinmoto, H. and analogs. J. Invest. Dermatol. 81: 158- Tsushida, T. (2005). Procyanidin trimers 162. to pentamers fractionated from apple inhibit melanogenesis in B16 mause

149

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

melanoma cells. J. Agric. Food Chem. www.pubchem.com. Accessed: 10/5/2016. 53(15): 6105-6111. Yamamoto, I., Muto, N., Murakami, K. and Son, S.M., Moon, K.D. and Lee, C.Y. (2001). Akiyama, J. (1992). Collagen synthesis in Inhibitory effects of various antibrowning human skin fibroblasts is stimulated by a agents on apple slices. Food Chem. stable form of ascorbate, 2-O-alpha-D- 73(1): 23-30. glucopyranosyl-L-ascorbic acid. J. Nutr. 122(4): 871-877. Visioscan, Visioscan VC98 – Unique description of the skin topography directly from the Skin (1998).

150

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

THE EFFECTS OF PHYSIOCHEMICAL ACTIVATION ON ADSORPTION CAPACITY, POROSITY AND SURFACE FUNCTIONAL GROUPS OF COCOA (Theobroma cacao) NIB – BASED ACTIVATED CARBON

Khairul Adli, N.1,4, Fisal, A.2, Mohd Azmier, A.3 and Mohd Sukri, H.1,5 1Faculty of Science and Technology, Universiti Sains Islam Malaysia, 71800 Nilai, Negeri Sembilan, Malaysia 2Malaysian Cocoa Board, Downstream Technology Cocoa Division, Lot 12621, 71800 Nilai, Negeri Sembilan, Malaysia 3School of Chemical Engineering, Engineering Campus, Universiti Sains Malaysia, Seri Ampangan, 14300 Nibong Tebal, SPS, Pulau Pinang, Malaysia 4Department of Forensic Medicine, Kuala Lumpur Hospital, Jalan Pahang, 50586 Kuala Lumpur, Malaysia 5Institute of Halal Research and Management, Universiti Sains Islam Malaysia, 71800 Nilai, Negeri Sembilan, Malaysia

Malaysian Cocoa Journal 9(1): 151- 159 (2016) ABSTRACT – Activated carbon with highly-porous structures was prepared from cocoa (Theobroma cacao) nibs pellet by chemical and physical (physicochemical) activation. Cocoa nib char was produced by pyrolysis at 700˚C under Nitrogen (N2) flow for an hour, impregnated with potassium carbonate (K2CO3) at the ratio of 1:1, 2:1 and 3:1 and placed in an oven at 105˚C for 24 hours (chemical activation). Physical activation was performed under N2 flow at 500˚C for one hour. Activated carbon impregnated with K2CO3 at the ratio of 3:1 exhibited the highest surface area by Brunauer, Emmett and Teller method and total pore volume at 1,729 m2g-1 and 1.57 cm3g-1, respectively. The Infrared spectroscopy (FTIR) analysis showed the presence of aliphatic and aromatic hydrocarbon together with nitrogen functional group.

Key words: Cocoa nib pellet, Activated carbon, Physicochemical activation, Ompregnation

INTRODUCTION precursor at certain temperature under inert atmosphere of nitrogen in the tubular furnace. Activated carbon is a highly porous material These two types of activation can be used to used to adsorb and remove substances. Activated control the characteristics of the prepared carbon is popular because it is easy to produce, (Alabadi et al., 2015). Activated carbon can be used in small amount and easy to obtain produced from agricultural by-products is called commercially. It has the desirable properties as a biosorbent (Kouakou et al., 2013). good adsorbent such as specific surface area, pore volume, pore size, and uniformity (Alabadi Functional groups can be found largely et al., 2015). The adsorption capability and on the edge of graphene layers (Barkauskas & selectivity is also depends on its surface Dervinyte, 2004). Well-developed porous chemistry (surface functional groups) and also structures and large internal surface consist of its pore size distribution (Barkauskas & hydrophobic graphene layers and hydrophilic Dervinyte, 2004). Production of activated carbon surface functional groups are essential to needs chemical activation (impregnation perform as adsorbents (Jia et al., 2002). process) or physical activation (temperature However, the surface of activated carbon can be setting) and sometimes the combination of both filled with uneven distribution of functional with the selected precursor (Fisal et al., 2013). groups and lots of heteroatom such as oxygen, nitrogen, sulphur and hydrogen which give Chemical activation involves variation in activated carbon’s characteristics carbonization process of precursor and chemical (Barkauskas & Dervinyte, 2004; Franz et al., which act as an activating agent under inert 2000). atmosphere of Nitrogen in the tubular furnace. Physical activation is by the pyrolysis of the

151

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

These surface groups containing lightweight nibs and discarded them as waste heteroatoms will have different chemical (Moore Jr, 2013). properties that will affect the function of the absorbents (Avecedo & Barriocanal, 2015). This study is aimed at preparing and Among these heteroatoms, the most common characterizing the activated carbon produced species and widely present on the surface of from discarded cocoa nibs. The objective of the activated carbons is oxygen-containing present study is to investigate the porosity, functional group (Shen et al., 2008). Oxygen- surface chemistry and adsorption capability of containing functional groups are well known for activated carbon obtained from pelletized cocoa heavy metals removal in aqueous solution nib. (Allwar, 2012, 2013). These complex surfaces also hold the carbon basal plane and inorganic ash (Franz et al., 2000). MATERIALS AND METHODS

Recently, researchers are using rice Materials and Reagents husks (Ghosh & Bhattacherjee, 2013; Raoul et Cocoa nib pellet was used as precursor in al., 2015); cocoa shells (Fisal et al., 2013; Ribas preparing activated carbon, which was obtained et al., 2014; Theivarasu & Mylsamy, 2011); from Malaysia Cocoa Board’s Pilot Plan located cocoa pod husk (Cruz et al., 2012; Chethana, in Bandar Baru Nilai, Negeri Sembilan. 2014); almond shells (Chayande et al., 2013); Potassium carbonate, K2CO3 was purchased palm kernel shells (Abechi et al., 2013); oil palm from Systerm and Hydrochloric acid, HCl was (Allwar, 2012); grass (Palanichamy et al., 2013); purchased from Merck. Deionized water was date (Hameed et al., 2009); date palm seed supplied by the Milli-Q® Integral Water (Halbus et al., 2013); olive stone (Larous & Purification System and distilled water was from Meniai, 2016; Bohli et al., 2013); and mango Favorit Water Still, Model W4L. nuts (Kwaghger & Ibrahim, 2013) as precursor for activated carbon. The performance of each of Precursor Characterization these precursors is varies to each other, but suit The moisture and ash content in cocoa nib pellet their objectives. were measured using muffle furnace method (Fisal et al., 2013). The moisture content As mentioned above, there were studies determination was performed by using crucible have been performed on cocoa-based materials which was pre-dried prior use. The crucible was as the suitable precursors for activated carbon. preheated in oven at 103ºC and then cooled for The most recognized precursors are cocoa pod 15 minutes in a desiccator. Approximately, 2 g husks and cocoa shells. Cocoa pod husk is the of sample was placed into the prepared crucible. outer part of the fruit, which protect the cocoa Both sample and crucible were heated in the beans from insects and the surrounding oven for three hours at 103⁰C. Then, the crucible environment. Cocoa shell is the thin coat at the was removed to lose heat for another 15 minutes outer layer of the cocoa bean, which contains the in the desiccator before the weight of the cocoa nib. Cocoa nib is the remaining embryonic crucible were recorded. Then the crucible was tissue of the cocoa seeds. Cocoa pod husk and heated again in the oven for two hours and was cocoa shell are abundant as agricultural waste cooled prior weighing. Weight obtained was and therefore, could easily be converted into recorded. activated carbon (Ismail et al., 2007). The sample preparation for the ash Grinding of cocoa nib will produce content determination was used the same cocoa liquor, cocoa butter and cocoa powder technique as in moisture content determination. which are the main ingredients to produce Prior analysis, the muffle furnace was set at chocolate (Ismail et al., 2007). About 7 % of 300⁰C for one hour. Approximately, 2 g of cocoa nibs especially nibs that was irregular in sample was added into the prepared crucible. shape, size and uneven was cast-off as waste The crucible was placed in the muffle furnace during the cracking and winnowing process and the temperature was raised to 600ºC. The (Akinnuli et al., 2014). The turbulent airflow sample was left in the furnace overnight. Prior during winnowing process took up the

152

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

weighing, the crucible was placed in a desiccator Depending on the IR, an amount of char to allow cooling. and K2CO3 were mixed together with hot deionized water in a 250 mL beaker. The The pH measurement of the precursor mixture was stirred on a hotplate with magnetic was determined by using pH multi parameter stirrer for one hour at 30ºC until the K2CO3 were (Mettler-Toledo Seven Excellence, Switzerland). completely dissolved. Then, the beaker was Approximately 2 g of cocoa nib pellet was stored for overnight inside an oven (Model crushed and dissolved in 100 mL boiled Memmert 600, Germany) at temperature 105ºC deionised water. Then the mixture was filtered to remove water and leaving only K2CO3 on through filter paper and the filtrate was used for char. pH measurement. The proximate analysis results are as listed in Table 1. The K2CO3-impregnated char was placed inside the vertical tubular furnace reactor Procedures for physical activation under nitrogen flow of Approximately, 60 g of precursor was weighed 1.5 mL/min. The temperature was ramped from and placed in a stainless steel vertical tubular ambient temperature to 500˚C. Once the required reactor that was placed in furnace (Fisal et al., temperature obtained, the process was held for 2013). Nitrogen gas was used as the purging gas one hour. Then, the prepared activated carbon through the furnace, to inhibit oxidization. The was cooled to room temperature. flow rate of nitrogen gas was set at 120 mL/min and the heating rate was held constant at Washing process was started initially 10ºC/min. The temperature was ramped from with boiled deionized water to recover the room temperature to 700ºC and held for one remaining K2CO3. A filter paper and filter funnel hour. After the one-hour holding time, the were used to hold the prepared activated carbon temperature was decreased to 30ºC to collect the while the hot deionized water passing through. produced char. The char was kept in desiccator The process was continued by using 0.1 M HCl, to cool down to room temperature and and a few more rounds with hot deionized water maintaining the moisture content. The char was until the pH of the washing solutions reached then stored in an air-tight container for further approximately 7. The pH measurement was treatment. Char yield was determined using made by using pH meter (Mettler-Toledo Seven following equation; Excellence, Switzerland). After the washing process finished, the activated carbon was dried in an oven (Memmert, Germany) at 105ºC. Then, the dried activated carbon was stored in air-tight container for further studies. where is the mass of char (g) after carbonization process and is the initial mass Surface Area and Carbon Porosity Analysis of cocoa nib pellet before carbonization. The surface area, pore size, micropore and mesopore volume were determined by The chars produced were impregnated Accelerated Surface Area and Porosimetry with potassium carbonate (K2CO3) at various System (Micromeritics ASAP 2020, US). There impregnation ratios (IR). Impregnation with were three steps taken to perform the analysis.

K2CO3 was a form of chemical activation. The impregnation ratio was calculated as: BET surface area and total pore volume were analysed by N2 as adsorptive gas at 77 K. The surface area and porosity were determined by N2 adsorption at 77 K. A relative pressure of -5 between 10 and 0.995 of N2 gas was used to

obtain the N2 adsorption isotherm. The BET where is the dry weight (g) of potassium surface area and total pore volume of the activated carbons were determined by the hydroxide pellets and is the dry weight (g) application of the Brunauer-Emmett-Teller of char. (BET), Langmuir and BJH (Barrett, Joyner, and Halenda) methods.

153

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Surface Morphology and Crystalline Structure Table 1. The ash and moisture contents of Surface morphology and crystalline structure precursor (cocoa nib pellet). was analysed by Scanning Electron Microscope (SEM). The surface morphologies were Proximate analysis (wt. %) examined by using a FEI scanning electron Ash content 11.71 microscope (FEI Quanta 400 FE-SEM). Moisture 3.46 pH 5.808

RESULTS AND DISCUSSIONS Figure 1 demonstrates the differences in Characteristic of Precursor morphology of original precursor, char and Table 1 exhibits the proximate analysis results, activated carbon of cocoa nib characterised by which show that the precursor contains low FE-SEM. Magnification of 2500X and 10,000X moisture content (3.46%), acidic (pH 5.808) and were used to observe the surface topology and ash content recorded at 11.71%. morphology. Figure 1.A1 and Figure 1.A2 demonstrate a rough surface that can be seen as stacks of cloud without any presence of pore,

cracks or voids which represent the cocoa nib pellet. This may be due to availability of cellulose, hemicellulose and fat in cocoa nib.

A1 A2

Figure1. FE-SEM micrographs (2,500X and 10,000X) of cocoa nib.

In Figure 2, the band at 3210 cm−1 is bands could be due to C=C vibration in a few assigned to O-H stretching vibration in hydroxyl structures and aromatic rings skeletal vibrations. groups. A strong band at 2918 cm−1 and 2850 The presence of the bands at 1400 cm−1 to 1300 cm−1 observed is attributed to C–H stretching cm−1 suggested the presence of phenol and vibration in the aliphatic group. A weak band at alcohol groups, while the bands at 1100 cm−1 to 2303 cm−1 is ascribed to C≡N in the nitrogen 900 cm−1 indicated the C–O. The weak bands at functional group. The band at 1735 cm−1 could 822 cm−1 to 746 cm−1 are assigned to C–H out- be attributed to C=O vibration in esters. In the of-plane bending in the aromatic rings. range 1700 cm−1 to 1400 cm−1, several observed

154

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Figure 2. FTIR spectra of cocoa nibs before treatments.

Morphology and Surface Chemistry of observed bands at 2841 cm−1, 1805 cm−1 and Activated Carbon 1397 cm−1 could be due to aliphatic and aromatic Figure 3.B1 and Figure 3.C1 show that the hydrocarbons which are assigned to C–H. physical activation process produced extensive external surfaces with irregular cavities and The FTIR spectrum obtained for the pores. The surface topology differed strongly prepared activated carbon displayed uniform between untreated cocoa nibs and activated one. bands at aliphatic hydrocarbon regions such as at 3100 cm−1 to 2850 cm−1 for C–H stretching and Porosity can be observed in Figure 1480 cm−1 to 1350 cm−1 for C–H bending. The 3.B1 and Figure 3.C1 however Figure 3.C1 bands at 2110 cm−1 to 2116 cm−1 in CNAC-IR- shows higher porosity than the other. 1, CNAC-IR-2 and CNAC-IR-3 could be Carbonization creates more pores as can be seen attributed to C≡C of the aliphatic hydrocarbon as on Figure 3.B2 however, the size of the pores found in char. A uniform band was also decreased in Figure 3.C2. Smaller and narrow observed in aromatic hydrocarbon region at pores were observed on the surface of the 1900 cm−1 to 1700 cm−1. The presence of the prepared activated carbon. This situation may be band at 3716 cm−1, 3100 cm−1 and 3044 cm−1 due to the evaporation of the chemical reagent may be subjected to hydroxyl groups, which was (K2CO3) during impregnation process. As a absenced in char but can be found in the result, micropores were developed more rather precursor. A weak band at 1600 observed in than mesopores. CNAC-IR-2 and CNAC-IR-3 were ascribed to C=C. Nitrile could be associated with the band at According to Table 2, char (CNAC-IR- 2400 cm−1 to 2300 cm−1 indicating the presence 0) loss bands at 2918 cm−1 and 2850 cm−1 during of nitrogen functional groups. Based on these carbonization. However, a band at 2345 cm−1 findings, it can be summarised that carbonization attributed to C≡N was still attached. The and IR were responsible in changing the surface presence of the band at 2113 cm−1 indicated the functional groups of the products. C≡C of the aliphatic hydrocarbon. A few

155

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

B1 C1

B2 C2

Figure 3. FE-SEM micrographs (2,500X and 10,000X) of char and activated carbon.

Table 2. FT-IR spectrum of cocoa nib pellet used as precursor for activated carbon.

Cocoa nib pellet AC Wave numbers Characteristics range (cm¯¹) IR = 0:1 IR = 1:1 IR = 2:1 IR = 3:1

C-H (stretching) - 2888 3100 2932, 3044 3100-2850 C-H (bending) 1397 1405 1396 1393 1480-1350 Aliphatic C=C 3716, 3806, - 3100 3044 hydrocarbon >3000 3999 C≡C 2113 2112 2110 2116 2260-2100 <900 827 821 827 826

Aromatic 2850-2820 2841 - - - hydrocarbon 1620-1400 - 1405 - - 1900-1700 1805 1802 1810 1856 O-H (stretching) Hydroxyl groups - 3716 3100 3044 3620-3000 C-O in ethers 1300-1000 - 1003 - - (stretching) Alkenes 1680-1600 - - 1619 1617 C≡N Nitrile 2345 2346 2343 2338 2400-2300

156

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Surface Area and Porosity Study of Activated isotherm as depicted at Table 3. The surface area Carbon of all the samples was exceeding 1,000 m2 g-1. The surface area and pore volumes of activated The surface area was increased from CNAC-IR- carbon are best developed in the carbonization 1 to CNAC-IR-3, exhibited the effect of temperature in the range of 500˚C to 800˚C increment of the IR. Observation on the pore when impregnated with K2CO3 (Cruz et al., volume and pore diameter against the IR, 2012). Selecting 500˚C as the activation indicated sample CNAC-1R-1 had smaller pore temperature was for the experiment purposes. A volume and average pore diameter (< 5 nm). different product of activated carbon is expected Activated carbon produced by chemical due to the impregnation ratio (IR) of K2CO3 to activation method has been shown with high char monolith. surface area and narrow micropore distribution (Adinata et al., 2007). Increment of the IR The BET surface area of the activated affects the surface area, the pore volume and the carbon was calculated from the N2 adsorption average pore diameter of the products.

Table 3. Coded samples of respective surface area of activated carbon from cocoa nibs at different activation times. S S S V D Sampel BET Langmuir BJH BJH BET (m2 g-1) (m2 g-1) (m2 g-1) (cm3 g-1) (nm) CNAC-1R-1 1,161.12 1,198.94 1,060.05 1.16 4.53 CNAC-1R-2 1,596.38 1,787.03 1,155.99 1.27 3.63 CNAC-1R-3 1,729.68 1,989.81 1,262.17 1.39 3.63

CONCLUSIONS Journal of Serbian Chemical Society 69 (5): 365-375. Wasted cocoa nibs from chocolate industry were [3] A. Fisal, W. D. Wan Mohd Ashri, A. Mohd used as precursor to prepare highly porous Azmier, R. Rosinah, A. A. Amirul activated carbon. The process involved chemical (2013). The effects of CO2 activation, activations by K2CO3 as activating agents and on porosity and surface functional physical activations at 500˚C as activating groups of cocoa (Theobroma cacao) – temperature. The resulting activated carbons shell based activated carbon. Journal of exhibited the maximum BET surface area was Environmental Chemical Engineering 1729 m2g-1 and mesopore volume was 1.16 1: 378–388. cm3g-1. The effects of impregnation ratio on the [4] U. Kouakou, A. S. Ello, J. A. Yapo, A. activated carbon demonstrated changes on the Trokourey (2013). Adsorption of iron presences of surface functional groups. The and zinc on commercial activated results indicated that K2CO3 and HCl treatments carbon. Journal of Environmental produced more nitrogen functional group on the Chemistry and Ecotoxicology 5 (6): activated carbon’s surfaces. 168-171. [5] Y. F. Jia, B. Xiao, K. M. Thomas (2002). Adsorption of metal ions on nitrogen REFERENCES surface functional groups in activated carbons. Langmuir 18: 470-478. [1] A. Alabadi, S. Razzaque, Y. Yang, S. Chen, [6] M. Franz, H. A. Arafat, N. G. Pinto (2000). B. Tan (2015). Highly porous activated Effect of chemical surface carbon materials from carbonized heterogeneity on the adsorption biomass with high CO2 capturing mechanism of dissolved aromatics on capacity. Chemical Engineering activated carbon. Carbon 38: 1807– Journal 281: 606-612. 1819. [2] J. Barkauskas, M. Dervinyte (2004). An [7] B. Avecedo, C. Barriocanal (2015). Texture investigation of the functional groups and surface chemistry of activated on the surface of activated carbons.

157

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

carbons obtained from tyre wastes. Fuel from cocoa (Theobroma cacao) pod Processing Technology 134: 275-283. husk. Civil & Environmental [8] W. Shen, Z. Li, Y. Liu (2008). Surface Engineering 2(2): 1-6. chemical functional groups [16] K. Chethana (2014). A research on cocoa modification of porous carbon. Recent pod husk activated carbon for textile Patents on Chemical Engineering 1: industrial wastewater colour removal. 27-40. International Journal of Research in [9] A. Allwar (2013). Porous structures of Engineering and Technology 3(3): 731- activated carbons derived from oil palm 737. empty fruit bunch by phosphoric acid [17] P. K. Chayande, S. P. Singh, and M. K. N. activation under nitrogen and carbon Yenkie (2013). Characterization of dioxide. International Journal of activated carbon prepared from almond Research in chemistry and Environment shells for scavenging phenolic 3(2): 62-68. pollutants. Chemical Science [10] Allwar (2012). Characteristics of micro- Transactions 2(3): 835-840. and mesoporous structure and surface [18] S. E. Abechi, C. E. Gimba, A. Uzairu, Y. A. chemistry of activated carbons Dallatu (2013). Preparation and produced by oil palm shell. characterization of activated carbon International Conference on Chemical, from palm kernel shell by chemical Ecology and Environmental Sciences. activation. Research Journal of Bangkok. 138-141. Chemical Sciences 3(7): 54-61. [11] R. Ghosh, S. Bhattacherjee (2013). A [19] K. Palanichamy, A. Ariharaputhiran, A. review study on precipitated silica and Darchen (2013). Activated carbon from activated carbon from rice husk. grass - a green alternative catalyst Chemical Engineering & Process support for water electrolysis. Technology 4(4): 1-7. International Journal of Hydrogen [12] T. T. D. Raoul, A. S. Gabche, N. G. Ndifor- Energy 38(25): 10364-10372. Angwafor, K. J. Mbadcam (2015). [20] B. H. Hameed, J. M. Salman, A. L. Ahmad Adsorption of salicylic and (2009). Adsorption isotherm and kinetic sulfosalicylic acid onto powdered modeling of 2,4-D pesticide on activated carbon prepared from rice and activated carbon derived from date coffee husks. International Journal of stones. Journal of Hazardous Materials Current Engineering and Technology 5 163: 121–126. (3): 1641-1652. [21] A. F. Halbus, Z. H. Athab, F. H. Hussein [13] M. C. Ribas, M. A. Adebayo, L. D. T. (2013). Adsorption of disperse blue dye Prola, E. C. Lima, R. Cataluna, L. A. on Iraqi date palm seeds activated Feris, M. J. Puchana-Rosero, F. M. carbon. International of Chemical Machado, F. A. Pavan, T. Calvete Science 11(3): 1219-1233. (2014). Comparison of a homemade [22] S. Larous, A-H. Meniai (2016). Adsorption cocoa shell activated carbon with of Diclofenac from aqueous solution commercial activated carbon for the using activated carbon prepared from removal of reactive violet 5 dye from olive stones. International Journal of aqueous solutions. Chemical Hydrogen Energy xxx, 1-11. Engineering Journal 248: 315–326. [23] T. Bohli, N. Fiol, I. Villaescusa, A. [14] C. Theivarasu, S. Mylsamy (2011). Ouederni (2013). Adsorption on Removal of malachite green from Activated Carbon from Olive Stones: aqueous solution by activated carbon Kinetics and Equilibrium of Phenol developed from cocoa (Theobroma Removal from Aqueous Solution. cacao) shell - a kinetic and equilibrium Chemical Engineering & Process studies. E-Journal of Chemistry 8 (S1): Technology 4(6): 1-5. 363-371. [24] A. Kwaghger, J. S. Ibrahim (2013). [15] G. Cruz, M. Pirilla, M. Huuhtanen, L. Optimization of conditions for the Carrion, E. Alvarenga, R. L. Keiski preparation of activated carbon from (2012). Production of activated carbon mango nuts using HCl. American

158

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Journal of Engineering Research 2(7): 28, 2013. [Cited: June 28, 2016.] 74-85. http://www.tejaschocolate.com/winnow [25] A. Ismail, K. Lamin, N. A. Nik Mat, M. ing-cocoa-beans/. Abdul, R. Kasran, O. Yusof, A. Awang, [28] A. Fisal, W. Daud, A. Mohd Azmier, R. D. K. P. Gangadharan, H. Hussein, F. Rosinah, A. Amirul Azri (2013). The L. Ening, H. Jaafar (2007). Malaysian effects of acid leaching on porosity and Cocoa - Naturally Nutritious, surface functional groups of cocoa Deliciously Healthy. Kota Kinabalu : (Theobroma cacao)– shell based Malaysian Cocoa Board. activated carbon. Chemical Engineering [26] B. O. Akinnuli, S. P. Ayodeji, A. J. Omeiza and Design 91: 1028–1038. (2015). Computer Aided Design for [29] D. Adinata, W. Daud, M. K. Aroua (2007). Cocoa Beans Processing Yield Preparation and characterization of Prediction. International Journal of activated carbon from palm shell by Applied Science and Technology 4(5): chemical activation with K2CO3. 82-91. Bioresour. Technol. 98: 145-149. [27] S. Moore Jr, Tejas Chocolate Craftory. winnowing cocoa beans. [Online] April

159

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

HUMAN CHEEK SKIN TEXTURE IMPROVEMENT BY COCOA BEAN EXTRACT INCORPORATED IN A COSMETIC FORMULATION

Norliza, A.W.1,2, Russly, A.R.2, Puziah, H.2, Amin, I.3 and Shuhaimi, M.2 1Malaysian Cocoa Board, Cocoa Innovation & Technology Centre, Lot 12621, Nilai Industrial Area, 71800 Nilai, Negeri Sembilan. 2Halal Products Research Institute, Universiti Putra Malaysia, 43400 Serdang, Selangor. 3Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, 43400 Serdang, Selangor.

Malaysian Cocoa Journal 9(1): 160-169 (2016) ABSTRACT – Evaluation of cosmetic formulation containing cocoa bean extract (CBE) in improving human cheek skin texture was carried out based on a non-invasive biophysical technique in a single-blinded study. The HPLC quantification on (–)-epicatechin and (+)-catechin from the CBE were 133.70±0.64 and 6.53±0.09 mg/g DW, respectively. A cosmetic formulation containing 0.1% concentrated extract of cocoa bean was developed by entrapping the extract in an internal aqueous phase. A similar formulation but without the extract was prepared as placebo. Each of the formulations was applied to the cheeks of 12 human volunteers for a period of two months. Results showed that the cosmetic formulation with CBE exhibited steady, significant improvement in skin parameters over the 2-months period of application. Images showed visible improvement of the overall skin appearance and reduction of fine lines by the cosmetic formulation with CBE. Furthermore, skin texture (variance and contrast) as well as skin evaluation of living skin (SELS) for skin roughness (SEr), skin scaliness (SEsc), skin smoothness (SEsm) and skin wrinkles (SEw), showed significant decline (p<0.05) and the texture parameter of energy showed significant increase (p<0.05). These findings indicated that the cosmetic formulation containing 0.1% CBE possesses anti-aging effect, thus improves skin surface appearance.

Key words: Human cheek skin, Cocoa bean extract, Cosmetic formulation, Anti-aging, Surface evaluation of living skin (SELS)

INTRODUCTION The research for a natural and healthy lifestyle has increased the interest for natural Constant exposure to ultraviolet (UV) irradiation bioactive compounds that could be introduced for may cause skin to age through a complex process. nutraceuticals as well as cosmeceuticals. In this UV generates reactive oxygen species (ROS) context, phenolic compounds are the most which has always been associated with collagen promising group of molecules due to their high deficiency and subsequent skin wrinkling (Fischer antioxidant capacity (Rice-Evans et al., 1997). et al., 1997, Varani et al., 2000). Aging of the skin There were subsequently increasing number of is visually characterized by irregular pigmentation, investigations on polyphenols due to their loss of elasticity, apparent wrinkling, dryness and encouraging effects on antibacterial, antiviral, anti- roughness (Soyun et al., 2009). Currently, inflammatory, anti-allergic, antithrombotic and cosmetic formulation containing natural bioactive vasodilatory actions (Cook and Samman, 1996). In has emerged as a promising vehicle for the drug vitro studies demonstrated that these compounds delivery to the body. Topical formulations have have several biological activities, such as the attracted significant interest as a vehicle for drug ability to scavenge superoxide radicals and delivery to the human skin due to easy compliance hydroxyl radicals, reduce lipid peroxyl radicals (Foldvari, 2000). Incorporation of antioxidant in a and inhibit lipid peroxidation (Salah et al., 1995). topical application has been widely investigated Polyphenols have also been incorporated into low especially for the decreasing of skin aging (Bisset density lipoprotein (LDL) particles and thereby et al., 1990). decrease their ability from oxidation. For

160

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

beneficial of human skin, cocoa polyphenols have air-dried to yield 45 g of fat-free material. been proven to exhibit a positive action on skin Ground cocoa cotyledons were further extracted elasticity (Gasser et al., 2008). with 70% aqueous ethanol for 2 hrs at 50oC using an orbital shaker (Unimax 1010, Heidolph, Malaysia used to be the fourth largest world Germany). The ratio between samples to extraction cocoa producer in 1989/1990, but unfortunately medium was 1:40. The extracts were combined, the ranking has dropped to twelfth position in and the organic solvent was removed by rotary 2006/2007 due to poor management input from the evaporator (IKA RV 10 control, Staufen, estate sector. Labor constraints continued to have a Germany) under partial vacuum at 40oC. The toll on the production and area cultivated despite aqueous extracts were freeze-dried and kept in air- of government’s effort to prop up production to tight container for further analyses. facilitate the industry’s grinding needs of around 300,000 tons over the last few years. The cocoa Cocoa bean extract formulation pod borer (CPB) infestation and labor issue which Water-based (aqueous phase) formulation was were often quoted as reasons for the losing of formulated from ion water S-100. The ethanolic interest shown in the crop, have been identified as CBE was dissolved in aqeous phase with the major factors responsible for its decline. To continuous stirring by propeller Silverson L5M-A. make things even worse following the declination The formula for the water-based cosmeceutical is of hectarage and dried cocoa beans, Azizah et al. given in Table 1. (2007) in their study have reported that Malaysian beans are traded at a low price due to inadequate Table 1: Formulation of water-based CBE- standard of cocoa aroma, astringency and bitter containing cosmetic. taste characteristics. One of the factors to such Part Materials Percentage (%) occurrence might be due to its high amount of A Cocoa bean extract 0.1 phenolic substances which is apparently not a B Ion water S-100 q.s. desirable property in chocolate and cocoa-based TOTAL 100% foods and beverages production. Ironically, high phenolic content of Malaysian cocoa bean High Performance Liquid Chromatography/ provides an efficient way to enrich the endogenous Diode Array Detector (HPLC/DAD) of cutaneous protection system and thus, may be a Ten (10) mg of the dried CBE was mixed with 5 successful strategy for diminishing UV radiation- mL 100% methanol before injected through a mediated oxidative damage in the skin. Therefore, reverse-phase C SEP-PAK cartridge (Waters it is necessary to diversify cocoa bean products 18 Associates, Milford, MA). (−)-Epicatechin and range in a cosmetic/cosmeceutical product (+)-catechin fractions retained in the cartridge, development in order to ensure cocoa existence were eluted with 10 mL of 40% aqueous methanol and its sustainability particularly in Malaysia. and subsequent 10 L of the solution was injected

into the HPLC/DAD. The HPLC/DAD included a

Waters Associates model 6000-A pump, U6K MATERIALS AND METHODS Universal injector, and Model 450 variable

wavelength detector. Separation of the compounds Preparation of the crude extract were accomplished on a reverse-phase analytical Fresh cocoa beans were manually deshelled before column with a mobile phase of water : methanol : grinding. The extract was prepared according to acetic acid (87:8:5), pumped at a flow rate of 1.5 the method of Hammerstone et al. (1999) with mL/min for 10 min. The eluting compounds were minor modification. The seeds were ground in a detected by monitoring absorbance at 280 nm and high-speed laboratory mill until the particle size quantified by comparing peak area of the sample was reduced to 1 mm. Lipids were removed from to those obtained from authentic standards (Sigma, 100 g of the ground seeds by extracting three times Co. Chemical, St. Louis, USA). Content of active with 450 mL of hexane. The lipid-free solids were compounds were expressed in mg/g DW.

161

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Ethical standards remained in the room for at least 15 min in order to This study was approved by the Jawatankuasa allow the skin to get acclimatized at 25oC. Etika Universiti Putra Malaysia (JKEUPM) for in vivo studies (Reference No UPM/TNCPI/RMC/ Visioscanning JKEUPM/1.4.18.1/F1) and was conducted Selection of MCB’s staffs (according to the criteria according to the guidelines of Helsinki declaration. of recruitment), was for the ease of monthly skin Prior to the study, each volunteer was provided measurement. For skin texture parameters and signed a volunteer protocol. This protocol was (volume, energy, contrast and variance) and a bilingual consent document indicating the terms surface evaluation of living skin (SELS) study, the and conditions such as purpose, duration, risks, Visioscan VC 98 system (Courage + Khazaka required procedures and potential benefits of the Electronic, Germany) was used. The protocol for testing. Volunteers were not informed regarding Visioscan VC 98 is as follows: the selected skin the content of formulations. area (human cheek) was monitored by a videocamera that transmitted images to the Study protocol computer screen. The camera was connected to a A total of 12 subjects, all women between 30 and computer via a digitalization unit (Video Digitizer 46 years of age at the Cocoa Innovation and VD 300) through FireWire port. This test was Technology Center, Malaysian Cocoa Board conducted before the treatment, as well as on the (MCB), Nilai, Negeri Sembilan, good health 4th and 8th week in the morning at 10 a.m. (2 condition and free from skin infections, were months duration of assessment). All measurements selected in the study. Informed consent was signed were made in a draught-free room at controlled by all volunteers before commencement of the temperature (18.021oC) and relative humidity study. Design of the study was single-blinded for (50-65%) (Ali et al., 2012). the comparisons of two formulations. All subjects were healthy females with no known Statistical analysis dermatological diseases or allergy to substances in The skin parameter values (volume, energy, the formulations. The exclusion criteria were contrast, variance and surface evaluation of living presence of any dermatitis and/or other skin or skin) of the right and left cheek of the volunteers allergic diseases, smokers and any previous were calculated at 0 h, 4th and 8th week of treatment of cheek skin with cosmetic assessment. Minitab ver. 14 was used for data formulations such as sunscreens, moisturizers or analysis on the computer by using the paired anti-aging cosmetics in the past two months prior sample t-test for the variation between the two to the commencement of the study. In vitro dermal formulations. The level of significance was set at and ocular irritection assays as well as in vivo p<0.05. patch test were previously performed to determine any possible irritation which might be caused by the product. Two weeks before the study and RESULTS AND DISCUSSION during the treatment period, subjects were permitted only to use cleansing product provided Identification and quantification using by investigator with no presence of any bioactive HPLC/DAD component. Each subject was then handed with Whilst total phenolic content may be a useful both the extract product and placebo. The subjects indicator of potential nutritional benefit, the actual were required to apply the extract product (A) and profile of phenolics within the bean is likely to be placebo (B) on the right and left cheek, more important. To know what is/are the respectively, focusing on the skin area below the responsible active ingredient(s) in the CBE, cheek bone, at a dose of 500 mg twice daily HPLC/DAD was used. The HPLC/DAD analysis (morning 7:00 – 9:00; evening, 22:00 – 24:00) at of CBE has revealed the presence of major home. The areas around the eyes were omitted. phenolics, namely (–)-epicatechin and (+)-catechin Prior to the parameters measurement, subjects identified by its chromatographic behavior and UV were required to wash their faces with water and spectra (280 nm). Figure 1 showed chromatograms

162

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

of the CBE as well as standard, indicating the Table 2 demonstrates the quantification of these existence of epicatechin and catechin whereas compounds.

(1I) (1)

(A ) (1I)

(1)

(B)

Figure 1. HPLC chromatograms of; A = mixture of standards, B = sample of CBE (Peaks: 1. Catechin; II. Epicatechin)

(−) -Epicatechin was detected to be the phenolics of cocoa beans from different countries major phenolic component in CBE sample, as compared to the present study of CBE. It is well contributing about 133.70±0.64 mg/g DW whereas established that epicatechin is the main polyphenol catechin content was 6.53±0.09 mg/g DW, thus found in cocoa beans (Kim and Keeny, 1984; lesser than the result obtained from previously Nelson and Sharpless 2003; Zhu et al., 2003). For measured, i.e. Folin-Ciocalteu method comparison, the content of epicatechin in (200.79±3.27 mg GAE/g DW). The outcome is Malaysian cocoa beans was at least 14-, 39-, 42- predictable due to the weak selectivity of the and 10-fold higher than the Malaysian, Ghanaian, Folin-Ciocalteu reagent, as it reacts positively with Cote d’Ivoire and Sulawesian of fermented cocoa different antioxidant compounds (phenolic and beans, respectively (Table 2). Epicatechin is a non-phenolic substances). The above mentioned major component of the polyphenol in cocoa bean HPLC/DAD results also indicated that such and it is a monomer of procyanidins. phenolic rich fractions from the cocoa bean may Approximately, 35% of the total phenolic content perform excellent antioxidant capacity in most in fresh/unfermented cocoa beans is constituted of assays. Table 2 shows literature data on the epicatechin. Fresh cocoa beans refers to the beans

163

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

that are sun-dried without fermentation or by Nazaruddin et al. (2006). Therefore, fresh fermented only for 1 to 2 days. Fermentation cocoa beans were rather chosen in this study due to process will lower the polyphenol content through its high epicatechin content compared to the oxidation and exudation, thus, reducing fermented beans. Furthermore, the fresh beans are astringency and bitterness (Misnawi et al., 2002). indeed involving less processing steps, i.e. Furthermore, 6–17% of epicatechin is degraded fermentation and roasting processes. during fermentation process in a study conducted

Table 2: Total phenolic content expressed as mg gallic acid equivalent (GAE)/g dry weight (DW), epicatechin and catechin content of cocoa beans from different countries

Sample Total Phenolic Epicatechin Catechin Content Content Content (mg GAE/g DW) (mg/g DW) (mg/g DW) *Fermented Malaysian 0.11±0.01b 8.2±0.12b ND Ghanaian 0.09±0.02b 3.0±0.16b ND Cote d’Ivoire 0.08±0.04b 2.8±0.05b ND Sulawesian 0.09±0.03b 12.2±0.07b ND **Fresh bean Malaysian CBE 200.79±3.27a 133.70±0.64a 6.53±0.09a

Values are expressed as mean ± standard deviation (n=3). Means with different letters in a same column were significantly different (p <0.05). * Study conducted by Azizah et al. (2007) ** Results of present study ND: Not detected

Skin texture parameters Highly hydrated elastic skin has a high energy Change in skin texture parameters (volume, value compared to older skin with many wrinkles energy, contrast and variance) after application of (Rasul and Akhtar, 2011). Furthermore, increase in placebo and extract product is shown in Figure 2. energy parameter also indicates moisturizing effect Increased in the energy of extract product were of the extract product and hence anti-aging statistically significant at all time intervals whilst properties of the CBE. Meanwhile, volume that of the placebo was not significant (p>0.05). measures the virtual amount of liquid (mm3) Furthermore, descending values for volume, required to fill the depths in image. In young and contrast, variance parameters were observed for smooth surface of the skin, less virtual liquid is the extract product and statistically significant at required to fill the depths. The extract product all time intervals, however, the values for the produced significant effects as showed in Figure 2 placebo were not significant (p>0.05). The where descending values were observed as well. formulation also showed significant effects when This shows skin smoothness as less rough skin paired sample t-test was applied for variation would require less amount of virtual liquid between the two preparations. Energy parameter (Khazaka, 2000). Variance is the average of a local describes the level of hydration of the skin. In this variance over an amount of pixels. The actual study, increased in the energy values for extract value of the pixel is compared to the average. High product was statistically significant at both reading variance indicates high roughness of skin surface intervals, however, placebo produced insignificant (Debowska et al., 2005). Luckily, our findings effects. Percentage of changes is represented in indicate reduced variance at both 4th and 8th weeks Figure 2. Energy increases with increase in the of evaluation (Figure 2). Finally, contrast hydration level of the skin and this in turn results parameter indicates the difference between gray in greater homogeneity of the output image. levels of two neighbored pixel. Higher contrast

164

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

always contributes to higher difference between reduction of roughness. Overall, the skin texture the values of two neighbors (Debowska et al., parameters describe the level of hydration of the 2005). Reduction in contrast indicates better skin. condition and smoothness of skin resulting in a

Figure 2. Change (%) in skin texture parameter (volume, energy, contrast and variance) of volunteers with time following application of extract product and placebo (Key: 0B=application of placebo at 0 hour, 0F=application of extract product at 0 hour, 1B=application of placebo after one month, 1F=application of extract product after one month, 2B=application of placebo after two months, 2F=application of extract product after two months)

Surface evaluation of living skin smoothness and wrinkles were identified for the The values of SELS parameters (SEr, SEsc, SEsm extract product. A decrease in the mean values of and SEw) before application of extract product and skin smoothness in contrast to skin roughness in at 1st and 2nd months of the study period are given the extract product suggested that it does possesses in Table 3 whereas Figure 4 represents the anti-aging characteristic. Smaller SEsc value has photographs of the right cheek of human always been associated with higher skin moisture. volunteers. The percent changes in the values for Moreover, lower values of SEw indicate that there 12 volunteers were calculated as in Figure 3. In were less wrinkles on the skin or the extract brief, SEr is the roughness parameter which product being administered, had successfully measures the proportion of dark pixels. SEsm is reduced fine wrinkles. An adverse reaction could the index of smoothness and is calculated from the be seen from the loss of collagen in relation with mean width and depth of wrinkles. SEsc is the transepidermal water loss (TEWL) (Williams et index of scaliness of skin which exhibits the al., 2009). Higher TEWL is equivalent to less intensity of skin dryness. SEw marks aging and is water retention by collagen in the dermis layer, measured from the proportion of horizontal and thus results in collagen deterioration. In addition, vertical wrinkles (Ishii et al., 2008). Gradual elastin is also essential due to its supports to the decrease in the values of roughness, scaliness, body’s natural elasticity. Elastin preventing skin

165

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

from sagging and enable body to naturally heal al., 2009). According to Alena and Psotova (2003) from wound. Therefore, both collagen and elastin as well as Nunu et al. (2008), phenolics inhibit could be easily degenerated due to greater TEWL, ROS and offer the best protection against enzymes which is why such circumstance should be strictly that cause the elimination of collagen and elastin avoided. Improvement in skin surface parameters in skin, thus prevent tissue injury. All of these can be attributed to the actives in CBE particularly findings collectively support our investigations epicatechin, and catechin as indicated in the that the CBE cosmetic possesses anti-aging HPLC/DAD chromatogram in Figure 1. properties. These results are similar to those Bioflavonoids and polyphenolic compounds in obtained by other authors who observed the use of botanicals have been shown to exhibit beneficial skin care products containing phenolic compounds antioxidant properties. Epicatechin is one of such that could lowered SELS values in skin roughness primary component and is effective against skin (De Paepe, 2000; Kampf and Ennen, 2006). inflammation and carcinogenesis (Udompataikul et

Figure 3: Change (%) in SELS parameters of volunteers’ skin following application of placebo and extract product (Key: 0B=application of placebo at 0 hour, 0F=application of extract product at 0 hour, 1B=application of placebo after one month, 1F =application of extract product after one month, 2B=application of placebo after two month, 2F=application of extract product after two month

Table 3: SELS parameters values (mean±SD)

Parameter Product 0 hour 1 Month 2 Months SEr Placebo 0.00±0.00 -14.29±0.02 -14.29±0.03 Extract formulation 0.00±0.00 -26.47±0.03 -65.38±0.02 SEsc Placebo 0.00±0.00 -13.33±0.01 -41.67±0.02 Extract formulation 0.00±0.00 -45.00±0.01 -81.25±0.01 SEsm Placebo 0.00±0.00 -3.88±1.37 -3.26±1.03 Extract formulation 0.00±0.00 -15.50±0.61 -22.77±0.78 SEw Placebo 0.00±0.00 0.36±0.00 -7.99±3.34 Extract formulation 0.00±0.00 -6.15±2.48 -30.88±5.46 SEr; skin roughness, SEsc; skin scaliness, SEsm; skin smoothness, SEw; skin wrinkles

166

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

(A) (B)

Figure 4. Photograph of the right cheek of a human volunteer at 0 hour (A) and after 2 months (B) of extract product application. The placebo produced statistically insignificant changes in all SELS values of the skin but the effect of the extract product was significant at all time intervals (p<0.05). This difference was also observed when the SELS parameters of the extract and placebo were directly compared (p<0.05)

CONCLUSION technique such as the one performed in this study is a valuable tool for assessment of anti-aging The results achieved in SELS investigation suggest effects of topical skin applications. Moreover, the that topical formulation of 0.1% CBE was capable formulation showed no harmful effects and it can in revitalizing the skin and reducing signs of skin be used as an effective topical anti-aging aging. The CBE formulation presents interesting treatment. effects in skin moisture as well as microrelief and the presence of CBE in the formulation possesses long term effects in the improvement of skin ACKNOWLEDGEMENTS roughness. Furthermore, in this study, it was found that the CBE formulation had the ability in This research was supported financially by the inducing skin tightness that enabling prevention of Ministry of Science, Technology and Innovation of the damage skin, thus delays aging process. The Malaysia (MOSTI) [Project No.: 02-03-13- formulation also increased the hydration of skin SF0074]. The authors would like to extend their and such hydration effect is crucial for normal thanks to Universiti Putra Malaysia (UPM) and cutaneous metabolism and may also prevent skin Cocoa Downstream Technology Division of alteration and early aging. This way, the CBE Malaysian Cocoa Board for the laboratory formulation can be a good choice when hydration facilities provided. effects are required. The non-invasive biophysical

167

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

REFERENCES Mass spectrometry. J. Agric. Food Chem., 47: 490496. Alena, S. and Jitka, P.D.W. (2003). Natural phenolics in the prevention of UV- Ishii, H., Kitamura, T., Todo, H., Sugibayashi, induced skin damage. A Review. Biomed. K., Sakurai, H. and Fukahori, M. (2008). Papers, 147: 137−145. Combined effect of sodium chondroitin sulfate and sodium hyaluronate on skin Bisset, D., Chaterjee, R. and Hannon, D. (1990). moisturization following single and Photoprotective effect of superoxide repeated application. Asian J. Pharm. Sci., scavenging antioxidants against 3: 94−101. ultraviolet radiation-induced chronic skin damage in the hairless mouse. Photoderm. Kampf, G. and Ennen, J. (2006). Regular use of Photoimmunol. Photomed., 7: 56−62. a hand cream can attenuate skin dryness and roughness caused by frequent hand Cook, N.C. and Samman, S. (1996). Flavonoids- washing. BMC Dermatol., 13: 15. Chemistry, metabolism, cardioprotective effects and dietary sources. J. Nutr. Khazaka, G. (2000). Information and operating Biochem., 7: 66–76. instructions for the Visioscan VC 98 and the software SELS (Surface evaluation of De Paepe, K., Lagarde, J.M., Gall, Y., Roseeuw, living skin). CK Electronic Cologne, D. and Rogiers, V. (2000). Microrelief of 34−36. the skin using a light transmission method. Arch. Dermatol. Res., 292: Kim, H. and Keeney, P.G. (1984). (−)- 500510. Epicatechin content in fermented and unfermented cocoa beans. J. Food Sci., Debowska, R., Rogiewicz, K., Iwanenko, T, 49: 10901092. Kruszewski, M. and Eris, I. (2005). Folic acid (Folacin) – New application of a Kubola, J. and Siriamornpun, S. (2008). cosmetic ingredient. Kosmetische Phenolic contents and antioxidant Medizin, 3: 16−22. activities of bitter gourd (Momordica charantia L.) leaf, stem and fruit fraction Fischer, G.H., Zeng, Q.W. and Subhashi, G. extracts in vitro. Food Chem., 110: (1997). Pathophysiology of premature 881890. skin aging induced by ultraviolet light. N. Engl. J. Med., 337: 1419−1428. Misnawi, Jinap, S., Nazamid, S. and Jamilah, B. (2002). Activation of remaining key Foldvari, M. (2000). Non-invasive enzymes in dried under-fermented cocoa administration of drugs through the skin: beans and its effect on aroma precursor challenges in delivery system design. formation. Food Chem., 78: 407417. Pharm. Sci. Technol. To., 3: 417−425. Nazaruddin, R., Seng, L.K., Hassan, O. and Gasser, P., Lati, E., Peno-Mazzarino, L., Said, M. (2006). Effect of pulp Bouzoud, D., Allegaert, L. and Bernaert, preconditioning on the content of H. (2008). Cocoa polyphenols and their polyphenols in cocoa beans (Theobroma influence on parameters involved in ex cacao) during fermentation. Ind. Crops vivo skin restructuring. Int. J. Cosmetic Prod., 24: 8794. Sci., 30: 339–345. Nelson, B.C. and Sharpless, K.E. (2003). Hammerstone, J.F., Sheryl, A.L., Alyson, E.M., Quantification of the predominant Robert, R. and Harold, H.S. (1999). monomeric catechins in baking chocolate Identification of procyanidins in cocoa standard reference material by LC/APCI- (Theobroma cacao) and chocolate using MS. J. Agric. Food Chem., 51: 531537. high-performance liquid chromatography/

168

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Nunu, M., Louisa, N., Ketevan, J., Marine, D. Udompataikul, M., Sripiroj, P. and and Eduard, C. (2008). Identification of Palungwachira, P. (2009). An oral free radicals induced by UV irradiation in nutraceutical containing antioxidants, collagen water solutions. J. Photochem. minerals and glycosaminoglycans Photobiol. B, 93: 61−65. improves skin roughness and fine wrinkles. Int. J. Cosmetic Sci., 31: Rasul, A. and Akhtar, N. (2011). Formulation 427−435. and in vivo evaluation for anti-aging effects of an emulsion containing basil Varani, J., Warner, R.L., Gharaee-Kermani, M., extract using non-invasive biophysical Phan, S.H., Kang, S., Chung, J.H., Wang, techniques. J. Pharm Sci., 19: 344−350. Z.Q., Datta, S.C., Fisher, G.J. and Voorhees, J.J. (2000). Vitamin A Rice-Evans, C.A., Miller, N.T. and Paganga, G. antagonizes decreased cell growth and (1997). Antioxidant properties of phenolic elevated collagen-degrading matrix compounds. Trends Plant Sci., 4: metalloproteinases and stimulates 304309. collagen accumulation in naturally aged human skin. J. Invest. Dermatol., 114: Salah, N., Miller, N.J., Paganga, G., Tijburg, L., 480−486. Bolwell, P. and Rice-Evans, C. (1995). Polyphenolic flavanoids as scavengers of Williams, S., Tamburic, S., Stensvik, H. and aqueous phase radicals and as chain Weber, M. (2009). Changes in skin breaking antioxidants. Arc. Biochem. physiology and clinical appearance after Biophy., 322: 339–346. microdroplet placement of hyaluronic acid in aging hands. J. Cosmetic Soyun, C., Serah, L., Min-Jung, L., Dong, H.L., Dermatol., 8: 216−225. Chong, H.W., Sang, M.K. and Jin, H.C. (2009). Dietary aloe vera supplementation Zhu, Q.Y., Hammerstone, J.F., Lazarus, S.A., improves facial wrinkles and elasticity Schmitz, H. and Keen, C.L. (2003). and it increases the type I procollagen Stabilizing effect of ascorbic acid on gene expression in human skin in vivo. flavan-3-ols and dimeric procyanidins Ann. Dermatol., 21: 6−11. from cocoa. J. Agric. Food Chem., 51: 828833.

169

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

BIOACTIVITY-GUIDED FRACTIONATION OF POTENT ANTI-CANCER PROPERTIES FROM NON-EDIBLE TISSUES OF Theobroma cacao

Zainal, B.1, Abdah, M.A.3, Taufiq Yap, Y.H.4, Roslida, A.H.3, Mohd Redzuan, S.1 and Kasran, R.2 1Division of Biotechnology, Cocoa Innovation & Technology Centre, Malaysian Cocoa Board, Lot Pt 12621, Nilai Industrial Park, 71800 Nilai, Negeri Sembilan, Malaysia 2Malaysian Cocoa Board, Cocoa Biotechnology Research Centre, Kota Kinabalu Industrial Park, 88460 Kota Kinabalu, Sabah, Malaysia 3Department of Biomedical Science, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, 43400 UPM, Serdang, Selangor, Malaysia 4Department of Chemistry, Faculty of Science, Universiti Putra Malaysia, 43400 UPM, Serdang, Selangor

Malaysian Cocoa Journal 9(1): 170-181 (2016) ABSTRACT - Theobroma cacao (T. cacao) is widely used especially in foods and beverages but only few studies on the use of the non-edible tissues of T. cacao as anti-cancer agents. This study investigated the cytotoxic activity of crude methanol extracts of T. cacao non-edible tissues against human cancer cell lines in vitro. This study was screened for cytotoxic activity of CL, CB, CH, UFCS, FCS, CR, CC and CP extracts against human cancer cell lines using MTT assay. The selected active extract against the most susceptible human cancer cell line was purified using bioassay-guided fractionation and the most active fractions were identified using GC-MS. The CL extract had strong cytotoxic activity against the breast cancer cell line (MCF-7) with its IC50 value was 41.43 ± 3.26 µg/ml. The extract was further purified and the hexane partitioned fraction had the highest cytotoxic activity (IC50 = 66.7 ± 7.95 µg/ml). Further purification was performed using flash column chromatography to produce a CL sub-fraction II/SF7, which was selected for GC-MS chemical characterization. The CL sub-fraction II/SF7 had the highest cytotoxic activity (IC50 = 6.36 ± 0.71 μg/ml). The CL sub-fraction II/SF7 from the T. cacao had potential cytotoxic properties against MCF-7 cells but not toxic to normal cell (WRL-68) and discovered that it contained about 9 major active compounds associated with cytotoxic activity. Eight compounds were known compounds and one compound was unidentified compound. T. cacao leaf contained potent bioactive compounds and should be investigated further, as it has the potential to decrease viability of human breast cancer cells.

Key words: Theobroma cacao, Leaf, Cancer, Cytotoxicity, Bioactivity-guided fractionation.

INTRODUCTION are associated with adverse side effects arising from the lack of tumour specificity (Aly, 2012). Cancer develops when there is uncontrolled cell proliferation due to oncogene gain-of-function Natural products have been recognized and tumour suppressor loss-of-function. In as a source for remedies since ancient times. general, cells in the body follow a regular path New bioactive natural products, especially of growth, division and death controlled by secondary metabolites from diverse sources such programmed cell death, or apoptosis; cancer as plants, bacteria, fungi, marine flora and fauna develops when this process is disrupted (Fearon are essential for the development of novel and Vogelstein, 1990). Typically, cancer is pharmaceuticals (Harvey, 2008; Ma and Wang, caused by DNA abnormalities in the affected 2009). Several plants have been identified as cells, which leads to the growth of an extra valuable sources of therapeutic agents for tissue mass, i.e. a tumour, which may be benign treating several types of cancer. For example, (noncancerous) or malignant (cancerous). The some effective anti-cancer drugs have natural possible causes of cancer are tobacco smoke, origins (Lee, 2010) such as the vinca alkaloids infection, chemicals, radiation and unhealthy vinblastine and vincristine (Catharanthus diet (Zaid et al., 2010). Cancer is a leading cause roseus), taxol (Taxus brevifolia), etoposide and of death in both developed and developing teniposide derived semisynthetically from countries. Current therapeutic approaches, which epipodophyllotoxin and an epimer of include surgery, chemotherapy and radiotherapy podophyllotoxin (Podophyllum species), and

170

Malaysian Cocoa Journal 2016, Vol. 9 No. 1 camptothecin (Camptotheca acuminate) (Fadeyi Nolvadex D) were purchased from AstraZeneca et al., 2013). UK (UK).

Theobroma cacao L. (Sterculiaceae), or Preparation of plant materials cocoa and its products are consumed worldwide Fresh non-edible cocoa plant tissues, i.e. cocoa and are studied mainly because of the in vitro leaf (CL), cocoa bark (CB), cocoa husk (CH), antioxidant, anti-tumour, anti-inflammatory and unfermented cocoa shell (UFCS), fermented antiradical properties of some of its polyphenolic cocoa shell (FCS), cocoa root (CR), cocoa constituents, specifically procyanidins and cherelle (CC) and cocoa pith (CP) were flavan-3-ol (Kim et al., 2010; Wollgast and collected from cocoa smallholder fields in Anklam, 2000). Wollgast and Anklam, (2000) Ranau, Sabah, Malaysia. Plant samples were identified three groups of polyphenols in cocoa collected during the peak cocoa fruiting season beans: catechins, anthocyanidins and in April 2011. A voucher specimen (SK proanthocyanidins, which constitute about 37%, 2434/14) was deposited at the Bioscience 4% and 58%, respectively, of the bean Institute of Universiti Putra Malaysia. The plant polyphenol content. Of the catechins, (−)- tissues were harvested, rinsed under tap water epicatechin is the most abundant (up to 35%), and air-dried in a covered space at room while (+)-catechin, (+)-gallocatechin, and (−)- temperature and then ground using a commercial epigallocatechin are present in smaller blender. Each powdered plant tissues (5 g) was quantities. Cocoa liquor procyanidins extracted by soaking in 200 ml methanol for 3 significantly reduced the incidence and days at room temperature. The mixture was then multiplicity of lung carcinoma and decreased filtered using a clean muslin cloth, followed by thyroid adenoma in male rats (Yamagishi et al., Whatman No. 1 filter paper. The filtrate was 2003). Certain cocoa procyanidins also inhibit evaporated to dryness using a rotary evaporator tumorigenesis, tumour growth and angiogenesis attached to a vacuum pump. Extracts were stored (Kenny et al., 2004). The main goal of this study at 2–8°C until used. was to investigate the cytotoxic activities of the extracts of 8 non-edible T. cacao tissues against Preparation of human cancer cell lines various human cancer cell lines also purified and Estrogen receptor–positive breast cancer (MCF- identified the bioactive compounds from the 7), estrogen receptor–negative breast cancer extract with the most potent cytotoxic activity (MDA-MB-231), liver cancer (HepG2), colon against the most susceptible human cancer cell carcinoma cancer (HT-29), lung cancer (A549), line using bioactivity-guided fractionation. cervical cancer (HeLa) and normal human liver (WRL-68) cell lines were purchased from MATERIALS AND METHODS American Type Culture Collection (USA). The HT-29, WRL-68 and MDA-MB-231 cells were Media, chemicals, reagents and drug cultured in DMEM with 10% FBS and 1% Silica gel 60 (particle size 0.06–0.2 mm, 70–230 penicillin/streptomycin. The MCF-7, HepG2, mesh ASTM), thiazolyl blue tetrazolium HeLa and A549 cells were cultured in RPMI bromide (MTT), dimethyl sulfoxide (DMSO), 1640 with 10% FBS and 1% trypan blue (0.4%) and phosphate-buffered penicillin/streptomycin. All cells were cultured saline (PBS) were purchased from Sigma at 37°C in 95% humidity and 5% CO2 for 3 Chemical Co. (USA). Roswell Park Media days. The cells were sub-cultured when they Institute (RPMI) 1640, Dulbecco’s modified reached 80–90% confluence. Eagle’s medium (DMEM), penicillin/streptomycin (100×), trypsin-EDTA Determination of target cell survival (1×) and foetal bovine serum mycoplex (FBS) The cytotoxic effect of the extracts against the were obtained from PAA, GmbH (Germany). cancer cell lines were determined using the The hexane, methanol (MeOH), microtitration colorimetric method of MTT dichloromethane (DCM), ethyl acetate and reduction based on Akim et al. (2011) with chloroform used were of the highest purity grade minor modifications. To test the biological and were obtained from Fisher (UK). activity, dried crude extracts were dissolved in Resublimed iodine was purchased from R&M DMSO to produce 10 mg/ml stock solution, Chemicals (UK). Tamoxifen tablets (20 mg, which was later mixed with culture media

171

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

(DMEM or RPMI 1640) to achieve the desired prepared by mixing 70, 27 and 10 g silica gel in concentration. Briefly, exponential-phase cells the fractionation (F), sub-fractionation (SF) and were harvested from maintenance cultures and sub-sub-fractionation (SSF), respectively, counted with the aid of a haemocytometer and followed by pouring 250, 200 and 150 ml trypan blue solution. Cell suspensions (100 μl) hexane for the F, SF and SSF phases, were dispensed in triplicate into 96-well culture respectively, into the column. The tap of the plates at optimized concentrations of ~1.0 × 105 glass column was left open to allow free flow of cells/ml per cell line. After 24-hour incubation at solvent into a conical flask below. The setup was 37°C, 100 μl culture medium was removed from deemed in order when the solvent drained freely the well and 100 μl fresh medium containing a without carrying silica gel or glass wool into the series of concentrations of the extracts was tap. At the end of the packing process, the tap added to each well and incubated for another 48 was locked and the column was allowed to hours. RPMI 1640 was used as the negative stabilize for 48 hours, after which the clear control. After 48 hours, the medium in each well solvent on top of the silica gel was allowed to was aspirated and replaced with 20 μl 5 mg drain down to the silica gel meniscus. The wet MTT working solution (MTT stock solution packing method was used for preparing the silica mixed with medium to attain a final gel column. The sample was mixed with 5, 3 and concentration of 0.5 mg/ml). MTT powder was 1 g silica gel in hexane for F, SF and SSF, dissolved in PBS to form a stock solution of respectively and dried at room temperature for MTT (5 mg/ml). The stock solution was stored 24 hours. The dried powder was shaken gently at -20°C. The cells were incubated at 37°C for 4 and lightly layered on top of the column. The hours and then the medium was aspirated and column tap was opened to allow the eluent to replaced with 100 μl DMSO to dissolve the flow at the rate of 40 drops per minute. Elution formazan crystals formed. The culture plates was performed using solvent systems of were shaken for 5 minutes and the absorbance of gradually increasing polarity in the following each well was measured using an microplate sequence: hexane (100% v/v), hexane:DCM (4:1 ELISA reader BioTek® EL808 (Winooski, VT, v/v), hexane:DCM (3:2 v/v), hexane:DCM (2:3 USA) at 570 nm, using 630 nm as the reference v/v), hexane:DCM (1:4 v/v), DCM (100%), wavelength. To quantify cell survival (S%), the DCM:MeOH (4:1 v/v), DCM:MeOH (3:2 v/v), absorbance of a sample with cells grown in the DCM:MeOH (2:3 v/v), DCM:MeOH (1:4 v/v) presence of different concentrations of the and MeOH (100% v/v). The eluted fractions investigated agents was divided by the were collected in aliquots of 50 ml (F), 10 ml absorbance of untreated cells grown only in (SF) and 5 ml (SSF). The chemical composition complete medium and multiplied by 100. The of each fraction was evaluated using thin-layer median inhibition concentration (IC50, µg/ml) chromatography (TLC) and visualized with was the concentration of the extract causing 50% ultraviolet (254 and 365 nm) to identify fractions inhibition of cell survival. containing similar constituents, which were then combined. The combined fractions were Flash column chromatography evaporated in vacuo at 40C. The yields of crude The selected methanolic crude extract was extracts, partitioned fractions and fractions were extracted by successive liquid-liquid partition in determined. hexane, followed by DCM and lastly MeOH to yield 3 partitioned fractions. The hexane, DCM Bioactivity-guided fractionation and MEOH partitioned fractions were obtained The crude extract with potent cytotoxic activity using a vacuum filter and evaporated under was tested against cancer cell lines using the reduced pressure. The partitioned fractions with MTT assay to confirm the cytotoxic activity. potent cytotoxic activity were subjected to flash The crude extract was partitioned in hexane, column chromatography based on the methods DCM and MeOH, The resultant fractions were of Lai et al. (2008) and Ode et al. (2011), with again tested for cytotoxic activity. Fractions that some modifications, to separate the extract into inhibited cell survival by 50% as determined by its component fractions using the silica gel 60. comparison of the IC50 were separated using In the column chromatography setup, the lower silica gel chromatography, yielding a series of part of the glass column was stocked with glass fractions. In the same manner, each fraction wool with the aid of a glass rod. The slurry obtained from the flash column chromatography

172

Malaysian Cocoa Journal 2016, Vol. 9 No. 1 was assayed for cytotoxic activity in the selected cells were plotted against the extract cancer cell line and was used to select fractions concentrations (µg/ml). Figure 1 illustrates the for further flash column chromatography. anti-cancer screening activities of the extracts Inactive extracts and fractions were not against various cancer cell lines. The CL extract investigated further. exhibited the most potent cytotoxic activity against MCF-7 cells (IC50 =41.43 ± 3.260 Gas chromatography–mass spectrometry (GC- µg/ml). Given the pronounced cytotoxic activity, MS) the in vitro cytotoxic study suggests that the CL The fraction with strong anti-cancer effect was extract might be a significant source of novel analyzed using GC-MS, which was carried out promising anti-cancer compounds. Cytotoxic using the method of Shunmugapriya and activity was classified into four groups based on Kalavathy (2012) with some modifications. GC- the IC50 values: active (≤20 µg/ml); moderately MS was performed using a QP2010 Plus system active (>20–100 µg/ml); weakly active (>100– (Shimadzu, Japan) and gas chromatograph 1000 µg/ml); inactive (>1000 µg/ml) interfaced to a mass spectrometer equipped with (Atjanasuppat et al., 2009). The CL, CB, CH, a SGE BPX5 fused silica capillary column (SGE UFCS, CR, and CC extracts were moderately incorporated, USA) with dimension of 30 m × active against MCF-7 cells; the respective IC50 0.25 mm ID × 0.25 µm composed of 100% were 41.43 ± 3.26 µg/ml, 71.97 ± 9.27 µg/ml, dimethyl polysiloxane. For GC-MS detection, an 62.23 ± 14.90 µg/ml, 65.03 ± 4.17 µg/ml, 76.40 electron ionisation system with ionising energy ± 13.81 µg/ml and 68.90 ± 11.41. The FCS of 70 eV was used. Helium gas (99.999%) was extract was also moderately active against HeLa used as the carrier gas at a constant flow rate 1.5 and HepG2 cells; the respective IC50 were 71.40 ml/min; an injection volume of 0.2 μl was ± 12.11 µg/ml and 68.90 ± 10.18 µg/ml. The CC employed with a split ratio of 1, injector extract was also moderately active against HeLa temperature of 250°C and ion source cells; the IC50 was 67.80 ± 9.36 µg/ml (Table 1). temperature of 200°C. The oven temperature The other extracts were weakly active (IC50 > was programmed from 50°C (isothermal for 2 100 μg/ml) against the tested cancer cell lines. minutes) with an increase of 50°C/min to 200°C, The differing cytotoxic activity values of the then 320°C, ending with a 9 minute isothermal extracts could have been due to the synergistic hold at 320°C. Mass spectra were taken at 70 effects of their components; a previous study eV; the scan interval was 0.5 seconds, and showed that crude plant extracts are more fragments were 30–750 m/z. The total GC pharmacologically active than their isolated running time was 21 minutes. The relative % active principles (Karagöz et al., 2007). CL amount of each component was calculated by extract was selected the for further analysis comparing its average peak area to the total because it had the most potent anti-cancer areas; software for handling mass spectra and activity against MCF-7 cells, as based on its chromatograms was a GC-MS solution. IC50, which was also significantly different from Interpretation of mass spectrum GC-MS was that of the positive control tamoxifen. The active conducted using the National Institute Standard ingredients in the methanolic CL extract may be and Technology (NIST) database, which valuable compounds that can kill cancer cells contains >62,000 patterns. The spectrum of the but are nontoxic in normal cells, given the low unknown component was compared with that of IC50 against the human normal liver (WRL-68) known components in the NIST database. The cells (Table 1). Generally, leaf extracts contain name, molecular weight, retention time, and the most anti-cancer constituents, e.g. Curcumin structure of the test material components were species (Prasanna et al., 2011), Alnus determined. sieboldiana (Ludwiczuk et al., 2011), Hibiscus sabdariffa (Lin et al., 2012), Azadirachta indica (Elumalai et al., 2012), and Carpinus betulus RESULTS AND DISCUSSIONS (Cieckiewicz et al., 2012). Furthermore, Kim et al. (Kim et al., 2011) reported that T. cacao Cytotoxic activity screening in vitro extracts, from the bean in particular, had anti- Methanolic extracts of non-edible cocoa plant cancer properties against various cancer cell tissues were evaluated using MTT assay and the lines due to polyphenolic contents such as percentage survival [S (%)] curves of treated procyanidin.

173

Figure 1. Survival [S (%)] of A549 (A), HeLa (B), HepG2 (C), HT-29 (D), MCF-7 (E), and MDA-MB-231 (F) cells grown for 48 hours in the presence of increasing concentrations of extracts of non-edible cocoa plant tissues. Each value represents the mean ± SEM of 3 independent experiments. CL, cocoa leaf; CB, cocoa bark; CH, cocoa husk; UFCS, unfermented cocoa shell; FCS, fermented cocoa shell; CP, cocoa pith; CR, cocoa root; CC, cocoa cherelle.

Extraction and partition yields of CL extract hexane, DCM and MeOH produced Solvent extraction is a critical part in analysis of approximately 20.65%, 6.84%, and 30.16% the bioactive compounds in plant materials for yields, respectively (Table 2). The percentage of extracting the desired chemical components for crude MeOH extract yield was based on the further separation and characterization. In this weight of dried and ground plant materials and study, the CL extract was prepared by extracting the percentage yield of the partitioned fraction 1000 g powdered CL in methanol, producing a was based on the weight of crude MeOH extract 6.73% yield. Further successive solvent-solvent used (Reddy et al., 2013). Compared to the other partitions were performed in hexane, DCM, partitioned fractions, the hexane fraction from followed by MeOH. Liquid-liquid partition for liquid-liquid partition was the most active

174

Malaysian Cocoa Journal 2016, Vol. 9 No. 1 against MCF-7 cells, while the MeOH fraction as the polar solvent may merely solubilize a had a higher yield compared to the other larger range of compounds, some of which may partitioned fractions (Table 2). These results have little or no anti-cancer activity (Anwar et suggest that a polar organic solvent is al., 2013). The yield of chemical extraction sufficiently strong for recovering a higher depends on the solvent type and varying extraction yield of anti-cancer components from polarities, extraction time and temperature, T. cacao leaf, which should be considered in sample-to-solvent ratio, as well as the chemical future extraction studies. However, it is composition and physical characteristics of the important to note that optimal extraction yields samples (Dai & Mumper, 2010). may not translate to higher anti-cancer activity,

Table 1. Cytotoxic activity (IC50) of the extracts of non-edible T. cacao tissues.

Extract IC50 (µg/ml) of cell line MDA-MB- A549 HeLa HepG2 HT-29 MCF-7 WRL-68 231 541.33 ± 430.67 ± 493.33 ± 559.00 ± 41.43 ± 504.33 ± 765.00 ± CL 69.82 82.16 17.15 16.01 3.26 30.22 34.0 574.67 ± 688.67 ± 828.33 ± 468.67 ± 71.97 ± 555.33 ± 842.67 ± CB 48.83 64.79 153.91 22.18 9.27 36.22 66.5 533.67 ± 372.67 ± 396.00 ± 443.33 ± 62.23 ± 550.67 ± 792.33 ± CH 63.51 16.48 40.43 23.36 14.90 6.57 75.5 613.67 ± 468.33 ± 464.33 ± 163.00 ± 65.03 ± 555.00 ± 97.90 ± UFCS 25.86 35.95 42.78 18.58 4.17 10.12 1.3 520.67 ± 71.40 ± 68.90 ± 197.00 ± 242.33 ± 486.33 ± 633.33 ± FCS 19.06 12.11 10.18 13.23 47.68 13.02 119.8 595.00 ± 641.33 ± 734.33 ± 530.33 ± 329.67 ± 360.00 ± 489.00 ± CP 82.61 61.00 81.61 41.07 52.66 34.77 50.6 636.33 ± 321.67 ± 237.33 ± 613.33 ± 76.40 ± 400.33 ± CR >1000.0 66.42 26.03 18.98 60.24 13.81 74.60 639.67 ± 67.80 ± 427.33 ± 629.33 ± 68.90 ± 614.67 ± 857.00 ± CC 45.74 9.36 31.35 37.75 11.41 11.29 44.0 3.36 ± 10.01 ± T - - - - - 0.89 0.84 Note: CL, cocoa leaf; CB, cocoa bark; CH, cocoa husk; UFCS, unfermented cocoa shell; FCS, fermented cocoa shell; CP, cocoa pith; CR, cocoa root; CC, cocoa cherelle; T, Tamoxifen. Bolded values indicated most potent cytotoxic activity.

Table 2. Extraction and partition yields of CL extract.

Samples Weight (g) Yield (%) MEOH extract 67.3 6.73 Hexane fraction 13.9 20.65 DCM fraction 4.62 6.84 MEOH fraction 20.34 30.16

Bioactivity-guided fractionation of CL extract fractions had significant cytotoxic bioactivity in Initial purification was performed on the crude MCF-7 cells (Figure 2). However, the non-polar CL extract to improve the selective cytotoxic solvent of the hexane partitioned fraction bioactive fractionation, using the cell line demonstrated the most effective activity against cytotoxic assay as a guide (Umthong et al., MCF-7 cells (IC50 = 66.67 ± 7.95 µg/ml) 2011). Sequential solvent partitioning in three compared to the DCM and MeOH partitioned different polarity solvents was used to fractions. Solvent partitioning is important for fractionate the extract. MTT assay revealed that obtaining active compounds in the active extract the hexane, DCM and MeOH partitioned distributed according to their polarity and is used

175

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

extensively for fractionation following antiproliferative activity (IC50 = 43.30 ± 2.83 bioactivity-guided fractionation (Duarte et al., µg/ml) (Table 3). Figure 3 shows that CL sub- 2012). The CL hexane partitioned fraction of the fraction II/SF7 was selected for further analysis T. cacao was further fractionated by flash due to its high cytotoxic activity. Figure 4 column chromatography with elution based on depicts the cytotoxic activity of CL sub-fraction solvents of increasing polarity. About 66 eluted II/SF7 and tamoxifen against MCF-7 cells and fractions (I/F1–F66) from the hexane partitioned the cytotoxic activity of CL sub-fraction II/SF7 fraction were collected and TLC- against WRL-68 cells. Seven concentrations of chromatographed, generating 9 combined 2-fold dilutions were used: 100, 50, 25, 12.5, fractions (II/F1–F9). The 9 combined fractions 6.3, 3.1 and 1.6 µg/ml. The IC50 of CL sub- were tested against MCF-7 cells, and combined fraction II/SF7 and tamoxifen against MCF-7 fraction II/F4 showed the most potent cells was 5.41 ± 0.54 and 3.55 ± 0.29 µg/ml, antiproliferative activity (IC50 = 7.58 ± 1.29 respectively, while the IC50 of CL sub-fraction µg/ml (Table 3). The fraction was thus selected II/SF7 against WRL-68 cells was 22.87 ± 0.09 for further purification by eluting again in µg/ml, indicating that CL sub-fraction II/SF7 solvent with increasing polarity, which had lower cytotoxic activity against WRL-68 generated about 45 sub-fractions (I/SF1–SF45), cells compared to MCF-7 cells. These results and TLC chromatography generated 7 combined were generally in agreement with that of the sub-fractions (II/SF1–SF7). The 7 combined screening test, confirming that MCF-7 cells were sub-fractions were tested against MCF-7 cells the most susceptible to CL sub-fraction II/SF7. and combined sub-fraction II/SF7 showed most As the IC50 of CL sub-fraction II/SF7 indicated potent antiproliferative activity (IC50 = 6.36 ± lower toxicity, therefore assumed that the CL 0.71 µg/ml) (Table 3). Sub-fraction II/SF7 was sub-fraction II/SF7 is non-toxic to normal liver selected for further purification using flash cells. It shown that CL sub-fraction II/SF7 has column chromatography and eluted in a series of anti-cancer properties without being toxic to solvents with increasing polarity to generate 10 normal cells (Abu et al., 2013; Al-Rashidi et al., combined sub-sub-fractions (II/SSF1–SSF10). 2011); for this reason, and given its effects on The 10 combined sub-sub-fractions were tested the MCF-7 cells, we selected it for GC-MS against MCF-7 cells and combined sub-sub- identification of its cytotoxic agents. fraction II/SSF8 showed the most potent

Figure 2. The IC50 of hexane partitioned fraction (HPF), DCM partitioned fraction (DPF) and MeOH partitioned fraction (MPF) in MCF-7 cells. Values are the means of triplicate samples (n = 3). Data are presented as the mean ± SEM.

176

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 3. Cytotoxic activity (IC50, μg/ml) of fractions, sub-fractions and sub-sub-fractions obtained from the CL extract. a a a Fractions IC50 (µg/ml) Sub- IC50 (µg/ml) Subsub- IC50 (µg/ml) fractions fractions II/F1 66.67 ± 6.44 II/SF1 42.07 ± 2.65 II/SSF1 >100 II/F2 70.30 ± 10.52 II/SF2 50.27 ± 1.48 II/SSF2 >100 II/F3 52.63 ± 6.35 II/SF3 43.90 ± 7.31 II/SSF3 >100 II/F4 7.58 ± 1.29 II/SF4 18.70 ± 1.04 II/SSF4 77.50 ± 2.68 II/F5 43.50 ± 3.48 II/SF5 40.67 ± 2.92 II/SSF5 71.87 ± 3.88 II/F6 60.60 ± 4.86 II/SF6 36.47 ± 3.52 II/SSF6 >100 II/F7 47.17 ± 5.20 II/SF7 6.36 ± 0.71 II/SSF7 >100 II/F8 52.60 ± 3.47 II/SSF8 43.30 ± 2.83 II/F9 45.70 ± 2.00 II/SSF9 >100 II/SSF10 >100 Note: aValues are the means of triplicate samples (n = 3). Data are presented as the mean ± SEM from 3 independent experiments. Bolded values indicate potent cytotoxic activity.

Figure 3. Mean IC50 of CL sub-fraction II/SF7 and tamoxifen against MCF-7 cells as compared with the mean IC50 of CL sub-fraction II/SF-7 against WRL-68 cells. Values are the means of triplicate samples (n = 3). Data are presented as the mean ± SEM.

GC-MS identification of anti-cancer agents in methyl ester (7%), octadecanoic acid methyl CL sub-fraction II/SF7 ester (7.34%), hexadecanoic acid-2- The bioactive compounds in CL sub-fraction hydroxyethyl ester (4.24%), octadecanoic acid- II/SF7, which had high cytotoxic activity, were 2-hydroxyethyl ester (2.39%), hexadecanoic identified using GC-MS and the data were acid-2-hydroxy-1-(hydroxymethyl)ethyl ester interpreted based on the NIST database. Figure (1.45%) and unidentified compound (33.08%). 4 depicts the GC-MS chromatogram, which The unidentified compound had the highest determined that about 39 compounds in CL sub- percentage of area and warrants further fraction II/SF7 contributed to its cytotoxic structural identification. The active compounds activity. Table 4 summarizes the compounds identified in CL sub-fraction II/SF7 are found in CL sub-fraction II/SF7 and shows that medicinally valuable and have various 9 major compounds had percentage of area > pharmaceutical applications (Table 4). The 1%: 4-(1-methyl)-2-cyclohexane-1-one (1.53%), antioxidant and antibacterial properties can be phenol,2,4-bis(1,1-dimethylethyl) (16.26%), 7- attributed to the presence of phenol,2,4-bis(1,1- oxabicyclo[4.1.0]heptane, 1-methyl-4-(2- dimethylethyl) (Manorenjitha et al., 2013); methyloxiranyl) (2.43%), hexadecanoic acid octadecanoic acid methyl ester has antioxidant,

177

Malaysian Cocoa Journal 2016, Vol. 9 No. 1 hypercholesterolemic and pesticide properties acid-2-hydroxyethyl ester has antiandrogenic, (Manorenjitha et al., 2013); and hexadecanoic hypocholesterolemic, nematicide, 5-alpha acid methyl ester has anti-inflammatory, reductase inhibitor, anti-acne, anti-inflammatory, antiandrogenic, cancer-preventive, cancer-preventive and dermatogenic properties dermatogenic, hypocholesterolemic, 5-alpha (Elezabeth & Arumugam, 2014). Hexadecanoic reductase inhibitor, anemiagenic, insectifuge and acid-2-hydroxy-1-(hydroxymethyl)ethyl ester flavouring properties (Kalaisezhiyen and has haemolytic, pesticide, flavouring and Sasikumar, 2012). Hexadecanoic acid-2- antioxidant activities. As stated earlier, the hydroxyethyl ester has antioxidant, unidentified compound requires further hypocholesterolemic, nematicide, antiandrogenic characterization. and 5-alpha reductase inhibitor properties (Sheela and Uthayakumari, 2013). Octadecanoic

Unidentified compound

Figure 4. GC-MS profile of CL sub-fraction II/SF7.

CONCLUSIONS characterization. Our current findings provide basic knowledge that will foster further studies We describe new findings for extracts of the in this field. non-edible parts of T. cacao, i.e. its cocoa leaf, which has strong cytotoxic activity against the ACKNOWLEDGEMENTS MCF-7 cells. The cytotoxic compounds present in the CL extract were purified using bioactivity- The authors would like to thank the Malaysian guided fractionation. The hexane-partitioned Cocoa Board (MCB), the cocoa smallholders of fraction had the highest cytotoxic activity Sabah for identifying and supplying plant against MCF-7 cells and was further purified specimens and the Universiti Putra Malaysia using column chromatography to produce (UPM) Department of Biomedical Sciences for fractions, sub-fractions and sub-sub-fractions. providing the human cancer cell lines and Based on the IC50, CL sub-fraction II/SF7 had laboratory facilities. This work was funded by the highest cytotoxic activity and identified 9 the MCB Temporary Research Fund and UPM major chemical constituents from the CL sub- Research University Grant Scheme (9302400). fraction II/SF7. We also detected an unidentified The authors also would like to thank Mohd compound in CL sub-fraction II/SF7, which Redzuan Shairi, Hazirah Abdul Radzak and Siti requires further identification and Sarah Sazali for their assistance.

178

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Table 4: Several active compounds were identified using GC-MS in CL sub-fraction II/SF7 of T. cacao.

Compound Formula Similarity Molecula RT index r weight (min) (%) (g) 2-Butanone C4H8O 93 72 4.017 2-Ethyl-2-butenal C6H10O 85 98 4.018 2-Ethyl-2-butenal C6H10O 85 98 4.325 6-Methyl-5-hepten-2-one C8H14O 86 126 5.125 2-Methyl-6-methylene-2-octane C10H18 91 138 5.667 (E)-3,7-Dimethyl-2,6-octadienal C10H16O 81 152 5.758 2-Methylbenzaldehyde C8H8O 95 120 6.042 1-Methyl-4-(2-methyloxiranyl)-7-oxabicyclo C10H16O2 83 168 6.542 4.1.0)heptanes 4-(1-Methyl)-2-cyclohexane-1-one C9H14O 83 138 6.750 2,4-Dimethyl-1-heptanol C9H20O 82 144 6.883 9-Methyl-1-decene C11H22 82 154 7.125 4-Propylbenzaldehyde C10H12O 92 148 7.292 Lilac aldehyde C10H18O2 81 170 7.358 6,10-Dimethyl-5,9-undecadien-2-one C13H22O 87 194 8.117 Oxalic acid C15H28O4 81 272 8.217 2,6-Bis(1,1-dimethylethyl)-2,5-cyclohexadiene- C14H20O2 84 220 8.258 1,4-dione Phenol,2,4-bis(1,1-dimethylethyl) C14H22O 96 206 8.450 6,11-Dimethyl-2,6,10-dodecatrien-1-ol C14H24O 83 208 8.525 Cyclohexanol, 2-methyl-3-(1-methylethenyl)-, C12H20O2 83 196 8.692 acetate, (1.alpha.,2.alpha.,3.alpha.) Methyl 12-methyltridecanoate C15H30O2 83 242 9.417 Cis-Z-alpha-bisabolene epoxide C15H24O 86 220 9.900 7-Oxabicyclo[4.1.0]heptane, 1-methyl-4-(2- C10H16O2 87 168 9.950 methyloxiranyl) Hexadecanoic acid methyl ester C17H34O2 97 270 10.292 1-Oxa-spiro[4.5]deca-6,9-diene-2,8-dione, 7,9-di- C17H24O3 83 276 10.342 tert-butyl- Benzenepropanoic acid, 3,5-bis(1,1-dimethyl)-4- C18H28O3 81 292 10.383 hydroxy-, methyl ester 4,8,12-Tetradecatrienal, 5,9,13-trimethyl C15H26O 89 222 10.617 1-Nonadecanol C19H40O 91 284 10.942 Phytol C20H40O 90 296 11.025 Octadecanoic acid methyl ester C19H38O2 97 298 11.075 3-Ethoxy-3,7-dimethyl-1,6-octadiene C12H22O 82 182 11.392 Hexadecanoic acid-2-hydroxyethyl ester C18H36O3 93 300 11.533 3-Chloropropionic acid C20H39ClO2 81 346 11.692 4,8,12,16-Tetramethylheptadecan-4-olide C21H40O2 94 324 11.95 Octadecanoic acid-2-hydroxyethyl ester C20H40O3 90 328 12.225 Glycidyl stearate C21H40O3 82 340 12.408 Mono(2-ethylhexyl) phthalate C16H22O4 91 278 12.525 Hexadecanoic acid-2-hydroxy-1- C19H38O4 92 330 12.492 (hydroxymethyl)ethyl ester Octadecanoic acid-2,3-dihydroxyprophyl ester C21H42O4 93 358 13.125 Unidentified compound - - - 17.216 Note: RT, retention time.

179

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

REFERENCES C., & Duarte, A. C. (2012). Analytical techniques for discovery of bioactive Abu, M. N., Ashraf, M., Salleh, M., Hani, N., & compounds from marine fungi. TrAC Radzman, M. (2013). Insulin sensitivity Trends in Analytical Chemistry, 34: 97– enhancement of the mixture of Tinospora 110. Crispa and Gelam (Melaleuca Cajuputi) Elumalai, P., Gunadharini, D. N., Senthilkumar, honey and its antiproliferative activity on K., Banudevi, S., Arunkumar, R., Benson, hepatocellular carcinoma, HepG2: A C. S., … Arunakaran, J. (2012). Ethanolic preliminary study. Journal of Medical neem (Azadirachta indica A. Juss) leaf Research and Development, 2(3): 48–54. extract induces apoptosis and inhibits the Akim, A., Ling, L. C., Rahmat, A., & Zakaria, Z. IGF signaling pathway in breast cancer A. (2011). Antioxidant and anti- cell lines. Biomedicine & Preventive proliferative activities of Roselle juice on Nutrition, 2(1): 59–68. Caov-3, MCF-7, MDA-MB-231 and HeLa Fadeyi, S. A., Fadeyi, O. O., Adejumo, A. A., cancer cell lines. African Journal of Okoro, C., & Myles, E. L. (2013). In vitro Pharmacy and Pharmacology, 5: 957–965. anticancer screening of 24 locally used Al-Rashidi, W., Mat Supri, N. N., & Manshoor, Nigerian medicinal plants. BMC N. (2011). Cytotoxic activity of crude Complementary and Alternative Medicine, extract from Costus malortieanus 13: 79. (Costaceae). American-Eurasian Journal Fearon, E. R., & Vogelstein, B. (1990). A of Toxicological Sciences, 3(2): 63–66. genetic model for colorectal Aly, H. A. A. (2012). Cancer therapy and tumorigenesis. Cell, 61(5): 759–767. vaccination. Journal of Immunological Harvey, A. L. (2008). Natural products in drug Methods, 382(1-2): 1–23. discovery. Drug Discovery Today, 13(19- Anwar, F., Kalsoom, U., Sultana, B., Mushtaq, 20): 894–901. M., Mehmood, T., & Arshad, H. A. Kalaisezhiyen, P., & Sasikumar, V. (2012). GC- (2013). Effect of drying method and MS evaluation of chemical constituents extraction solvent on the total phenolics from methanolic leaf extract of Kedrostis and antioxidant activity of cauliflower foetidissima (Jacq.) Cogn. Asian Journal (Brassica oleracea L.) extracts. of Pharmaceutical and Clinical Research, International Food Research Journal, 5(3): 0–4. 20(2): 653–659. Karagöz, A., Turgut-Kara, N., Çakir, Ö., Atjanasuppat, K., Wongkham, W., Meepowpan, Demirgan, R., & Ar, Ş. (2007). Cytotoxic P., Kittakoop, P., Sobhon, P., Bartlett, A., activity of crude extracts from Astragalus & Whitfield, P. J. (2009). In vitro chrysochlorus (Leguminosae). Biotechnol. screening for anthelmintic and antitumour & Biotechnol. EQ., 21(2): 2–4. activity of ethnomedicinal plants from Kenny, T. P., Keen, C. L., Jones, P., Kung, H.-J., Thailand. Journal of Ethnopharmacology, Schmitz, H. H., & Gershwin, M. E. (2004). 123(3): 475–82. Pentameric procyanidins isolated from Cieckiewicz, E., Angenot, L., Gras, T., Kiss, R., Theobroma cacao seeds selectively & Frédérich, M. (2012). Potential downregulate ErbB2 in human aortic anticancer activity of young Carpinus endothelial cells. Experimental Biology betulus leaves. Phytomedicine : and Medicine, 229: 255–263. International Journal of Phytotherapy and Kim, J., Lee, K. W., & Lee, H. J. (2011). Cocoa Phytopharmacology, 19(3-4): 278–83. (Theobroma cacao) seeds and Elezabeth, D. V., & Arumugam, S. (2014). GC- phytochemicals in human health. In Nuts MS analysis of ethanol extract of Cyperus and Seeds in Health and Disease rotundus Leaves. International Journal of Prevention (pp. 351–360). Elsevier Inc. Current Biotechnology, 2(1): 19–23. Kim, J.-E., Son, J. E., Jung, S. K., Kang, N. J., Dai, J., & Mumper, R. J. (2010). Plant phenolics: Lee, C. Y., Lee, K. W., & Lee, H. J. extraction, analysis and their antioxidant (2010). Cocoa polyphenols suppress TNF- and anticancer properties. Molecules α-induced vascular endothelial growth (Basel, Switzerland), 15(10): 7313–52. factor expression by inhibiting Duarte, K., Rocha-Santos, T. A. P., Freitas, A. phosphoinositide 3-kinase (PI3K) and

180

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

mitogen-activated protein kinase kinase-1 effect of Cassia auriculata leaf extract and (MEK1) activities in mouse epidermal curcumin through induction of apoptosis in cells. The British Journal of Nutrition, human breast and larynx cancer cell lines. 104: 957–964. Biomedicine & Preventive Nutrition, 1(2): Lai, C.-S., Mas, R. H. M. H., Nair, N. K., Majid, 153–160. M. I. a, Mansor, S. M., & Navaratnam, V. Reddy, A. S., Abd Malek, S. N., Ibrahim, H., & (2008). Typhonium flagelliforme inhibits Sim, K. S. (2013). Cytotoxic effect of cancer cell growth in vitro and induces Alpinia scabra (Blume) Náves extracts on apoptosis: an evaluation by the bioactivity human breast and ovarian cancer cells. guided approach. Journal of BMC Complementary and Alternative Ethnopharmacology, 118(1): 14–20. Medicine, 13: 314. Lee, K.-H. (2010). Discovery and development Sheela, D., & Uthayakumari, F. (2013). GC-MS of natural product-derived analysis of bioactive constituents from chemotherapeutic agents based on a Coastal Sand Dune Taxon-Sesuvium medicinal chemistry approach. Journal of portulacastrum (L.). Bioscience Natural Products, 73(3): 500–16. Discovery, 4(1): 47–53. Lin, H.-H., Chan, K.-C., Sheu, J.-Y., Hsuan, S.- Shunmugapriya, K., & Kalavathy, U. (2012). W., Wang, C.-J., & Chen, J.-H. (2012). GC-MS Analysis of bioactive constituents Hibiscus sabdariffa leaf induces apoptosis of Rauwolfia densiflora (Wall) Benth. ex of human prostate cancer cells in vitro and HK.f. International Journal of Applied in vivo. Food Chemistry, 132(2): 880–891. Biology and Pharmaceutical Technology, Ludwiczuk, a, Saha, a, Kuzuhara, T., & 3, 1–5. Asakawa, Y. (2011). Bioactivity guided Umthong, S., Phuwapraisirisan, P., Puthong, S., isolation of anticancer constituents from & Chanchao, C. (2011). In vitro leaves of Alnus sieboldiana (Betulaceae). antiproliferative activity of partially Phytomedicine, 18(6): 491–8. purified Trigona laeviceps propolis from Ma, X., & Wang, Z. (2009). Anticancer drug Thailand on human cancer cell lines. BMC discovery in the future: an evolutionary Complementary and Alternative Medicine, perspective. Drug Discovery Today, 11(1): 37. 14(23-24): 1136–42. Wollgast, J., & Anklam, E. (2000). Polyphenols Manorenjitha, M. S., Norita, A. K., Norhisham, in chocolate: Is there a contribution to S., & Asmawi, M. Z. (2013). GC-MS human health? Food Research analysis of bioactive components of Ficus International, 33: 449–459. religiosa (Linn.) stem. International Yamagishi, M., Natsume, M., Osakabe, N., Journal of Pharma and Bio Sciences ISSN, Okazaki, K., Furukawa, F., Imazawa, T., 4(2): 99–103. … Hirose, M. (2003). Chemoprevention of Ode, O. J., Asuzu, I. U., & Ajayi, I. E. (2011). lung carcinogenesis by cacao liquor Bioassay-guided fractionation of the crude proanthocyanidins in a male rat multi- methanol extract of Cassia singueana organ carcinogenesis model. Cancer leaves. Journal of Advanced Scientific Letters, 191: 49–57. Research, 2(4): 81–86. Zaid, H., Rayan, A., Said, O., & Saad, B. (2010). Prasanna, R., Chinnakonda Chandramoorthy, H., Cancer treatment by Greco-Arab and Ramaiyapillai, P., & Sakthisekaran, D. Islamic herbal medicine. The Open (2011). In vitro evaluation of anticancer Nutraceuticals Journal, 3(1): 203–212.

181

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

DISINTEGRATION OF COCOA POWDER TABLET

Samuel, Y.K.C., Chin, H.H., Sarini, H. and Arief Huzaimi, M.Y. Malaysian Cocoa Board, Cocoa Innovation & Technology Centre, Lot PT12621, Nilai Industrial Area, 71800 Nilai, Negeri Sembilan

Malaysian Cocoa Journal 9(1): 182-186 (2016) ABSTRACT - By its nature, cocoa powder is not inclined to disperse, but rather to float on the surface of fluid at room temperature or below. This is primarily result of the cocoa butter contained in the powder, which repels water and prevents wetting of the particles. The objective of this paper is to report factors that were contributed towards disintegrate processes of cocoa powder. Cocoa powder was encapsulated with lecithin, sieved and pressed into tablet form by direct compression method. Results showed that amount of lecithin used, which had direct correlation with cocoa powder particle size, and compression force affects the tablet disintegration time and its tensile strength.

Key words: Cocoa powder, Disintegration, Lecithin

INTRODUCTION surface and inhibit the dissolution or dispersion process (Phuong and Joachim, 2014). Cocoa powder is prepared from cocoa nibs, which have been adequately fermented, dried, Fast dissolving or dispersing products roasted and ground after the extraction of cocoa are of high interest in food industry as well as in butter (Beckett, 1994). Generally, up to about pharmaceutical industry. Hence, for easy 80-85% of the fat content can be removed from handling and preparation of cocoa powder drink, the cocoa mass by mechanical pressing. This either in hot or cold medium, development of typically leaves 10-20% of the fat residuals in effervescent cocoa tablet was the ultimate goal cocoa powder (Ian et al., 2010). of this project. An effervescent tablet refers to a tablet that is able to disperse or quickly broken Cocoa powder is used in the preparation apart by internal liberation of carbon dioxide of food stuffs, such as baked products, gases in water (Srinath et al., 2011). beverages, and confectionery. Beverages, such as drinking chocolate is one of the major usage of cocoa powder. In these applications, the cocoa MATERIALS AND METHODS powder is ideally able to disperse when mixed with cold aqueous medium such as milk or Lecithin used in this project was a gift from water. However, cocoa powder is not easily Lasenor, cocoa powder was purchased from wetted which can impede dispersion and results Barry Callebaut (M) Sdn Bhd., and all other in lumping due to the existence of cocoa butter chemicals were pharmaceutical grade from residual in the powder (Bernaert et al., 2013). Merck. Due to the fact that cocoa powder is covered by free fat (cocoa butter), it is not possible to obtain Various concentrations of lecithin the dispersion of powder in water or milk at (0.1% - 0.9%) were sprayed at the rate of 10 temperature below the melting point of the fat ml/min into cocoa powder in a fluid bed dryer (Van Neeuwanhuyzen, 1981). In addition, fine system set at 90oC of the chamber temperature. powder, such as cocoa powder, face Nozzle pressure was set at 2 bar. The resulting agglomeration phenomena when it is to be powder was tested for its wettability by placing dissolved or dispersed in a liquid. When fine dry 1 g powder sprinkle into 100 ml water at 25oC cocoa powder is poured into a liquid, the surface until all powder particle were completely wetted. boundary of the solid is in contact with liquid Particle size distribution of the powder was phases and form a cover layer, trapping other yet tested by Malvern Particle Size Analyser with to be wetted dry powder and air inside. As a alcohol as dispersion medium. result, agglomerated powder floats on the

182

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Powder with the best wettability friabilator and rotated at the speed of 25rpm for characteristic was chosen and mixed with other 100 rotations. Tablets were removed from the excipients of sugar, sodium bicarbonate and friabilator, dusted off the fines and weighted. citric acid at the ratio of 21:20:31:28, and passed Result was calculated in term of percent weight through 750µm sieve, pressed at 6 to 9 tonne of loss (Lakshmi and Sagar, 2011). compression force by direct compression method using Chamunda rotary tablet press machine. The die of the tableting machine was fixed at 16 RESULTS AND DISCUSSIONS thread counts. Test on the resulting tablets disintegration time was carried out by placing Particle size analysis reviewed that all treatments one tablet into 50ml drinking water at 25oC until gave similar particle size distribution profile the said tablet was fully disintegrated. Tablet with a small peak was observed at the region of hardness was measured using texture analyser 0.7 to 2µm and a major peak at 5 to 40µm with needle probe punched at the middle of the (Figure 1). More than 70% of the volume tablet samples and the highest reading just density falls in the region of 5 to 40µm with before the tablet break was recorded. The peaks at 10-20µm. Hence, the values of Dv(50) thickness of the tablets was measured using were taken to represent the average population Vernier calliper. particle size for each of the treatments respectively. Tablet friability was carried out by placing five pre-weighted tablets into Roche

Lecithin concentration

Figure 1. Particle size distribution.

Lecithin contains a lipophilic as well as system (Hamburg, 1982). Addition of lecithin a lipophobic moiety, and therefore, it can react into cocoa powder in this study was to improve as a surface-active agent. It is able to reduce the its wettability and absorption by water when interfacial or surface-tension in a colloidal cocoa powder is dispersed in water.

183

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Figure 2 shows that the wettability of particle was encapsulated with single layer of cocoa powder was strongly affected by the lecithin, thus giving bigger particle size on amount of lecithin in the formulation. Results average. Increasing lecithin concentration above showed that cocoa powder treated with 0.5% 0.5%, however, gave antagonistic effect due to lecithin gave fastest time (1.76 minutes) to the increase of free lecithin contents, and disperse completely in water at 25oC. At this consequently brought down the average particle point, its particle size was the largest (Figure 1). size value. At 0.5% lecithin, each of the cocoa powder

Figure 2. Effect of lecithin treatment on cocoa powder wettability and its particle size.

Tableting machine used in this project Figure 3 showed that higher is a “force control principle” variant, whereby, it compression force produced harder tablet and is based on the measurement of final improved solubility. Each excipients in the tablet compression force set by user. The die of the formulation were in closer proximity with higher tableting machine set at 16 thread counts gave compression force, and thus, giving faster 4.6g weight of cocoa powder as fill volumes, reaction rate. Nevertheless, compression force at and statistically, a slight variation in tablets 8.5 tonne would produce tablets with highest thickness were observed as the pressing force dissolution time and indicated the optimum increased (Table 1). compression force applied was 8.5 tonne. At this point, the tablets produced had 1,088.89 N in Disintegration, as a precursor to hardness and 2.50 minutes dissolution time in dissolution (Alderborn, 2005), was affected by average. Friability test gave 0.3% in average the tensile strength of a tablet. The energy of (Table 1), which was within the nominal failure, which is reflecting the inter- and intra- friability of 1% maximum of start weight. Above particle cohesion within a tablet is another this point, the tablets produced were too compact determining factor affecting the disintegrative and consequently yielding high tensile strengths and dissolution properties. Both the tensile and energy of failure, and hence, gave poor strength of the tablet and the energy of failure disintegration time. are predominantly affected by the applied compression force (Heng et al., 2010).

184

Malaysian Cocoa Journal 2016, Vol. 9 No. 1

Figure 3. Tablet hardness and its solubility in different compression force.

Table 1. Evaluation of physical parameters of cocoa tablet pressed in different compression force

Compressi Tablet Weight Thickness Disintegration/

on Force (mg)* (mm)** solubility time (tonne) Friability (%)* Hardness (N)** (min)**

6.00 4,669.82 ± 112.65 0.31 ± 0.01 636.19 ± 8.30 8.73 ± 0.04 4.40 ± 0.04

6.50 4,651.36 ± 123.05 0.32 ± 0.02 668.70 ± 6.43 8.70 ± 0.02 4.25 ± 0.03

7.00 4,578.04 ± 73.04 0.29 ± 0.02 807.43 ± 8.56 8.68 ± 0.02 4.08 ± 0.02

7.50 4,543.98 ± 110.93 0.33 ± 0.01 967.55 ± 7.99 8.67 ± 0.02 3.38 ± 0.03

8.00 4,449.17 ± 115.74 0.31 ± 0.02 976.60 ± 7.26 8.65 ± 0.03 3.08 ± 0.04

8.50 4,598.20 ± 107.19 0.30 ± 0.02 1088.89 ± 6.88 8.64 ± 0.03 2.50 ± 0.05

9.00 4,490.04 ± 158.86 0.30 ± 0.02 1710.32 ± 6.72 8.63 ± 0.03 4.43 ± 0.04

All values were the average of five determination ± Standard Deviation. *Indicate no significant differences between treatment at p>0.05 **Indicate significant differences between treatment at p<0.05

185

CONCLUSIONS Bernaert, H., Blondeel, I. and DeClercq, D. (2013). Method for Producing a Soluble The study showed that 0.5% lecithin was able to Cocoa Product from Cocoa Powder. improve wettability of cocoa powder, whereby it European Patent EP2094100B1. took 1.76 minutes in average to disperse completely in water at 25oC. Dry mixing of cocoa Hamburg, R.Z. (1982). Lecithin. Confectionery powder treated with 0.5% lecithin, sugar, sodium Production. July. 284-285, 290. bicarbonate and citric acid at the ratio of 21:20:31:28 and passed through 750µm sieve; Heng, D., Ogawa, K., J. Cutler, D., Chan, H.K., pressed at 8.5 tonne of compression force gave A. Raper, J., Ye, L., and Yun, J. (2010). acceptable effervescent effect on the produced Pure Drug Nanoparticles in Tablets: What cocoa tablets with enough hardness in normal Are the Dissolution Limitations? J. tablet handling based on its friability value. Nanopart Res.12: 1743-1754.

Ian, C.P., Todd, W.G., Bassam, F.J., and Frans, ACKNOWLEDGEMENTS W.B. (2010). Low Fat Cocoa Powder. US7709041B2. The authors would like to acknowledge Director General of the Malaysian Cocoa Board to Lakshmi, C.S.R. and Sagar, D.A. (2011). approving the funding of this project through Development and Characterization of Temporary Research Fund (TRF) of MCB and Melt-in-Mouth Tablets of Atenolol by Economy Planning Unit (EPU) of the Malaysia Sublimation Technique. IJPRD. 3(3); 5 Prime Minister Department (JPM) for the (May 2011): 27-36. financial assistance in research facilities procurement through RMK 10 Development Phuong, T.N.N. and Joachim, U. (2014). Fast Project Funds. Dispersible Cocoa Tablets: A Case Study of Freeze-Casting Applied to Foods. Chem. Eng. Technol. 37(8): 1376-1382. REFERENCES Srinath, K.R., Chowdary, C.P., Palanisamy, P., Alderborn, G. (2005). Tablets and Compaction. Vamsy Krishna, A., Aparna, S., Ali, S.S., In: Aulton ME (ed) Pharmaceutics. The Rakesh, P., and Swetha, K. (2011). Science of Dosage Form Design, 2nd Formulation and Evaluation of edition. Elsevier. London. Pp. 397-440. Effervescent tablets of Paracetamol. IJPRD. 3(3);12(May 2011): 76-104. Beckett, S.T. (1994). Industrial Chocolate Manufacture and Use, 2nd edition. Blackie Van Neeuwanhuyzen, W. (1981). The Industrial Academic & Professional. London. Used of Special Lecithin: A Review. J. Amer. Oil Chem. And Soc. 58(10): 886- 888.

186

SHELF-LIFE STUDY OF Lactobacillus fermentum IN PROBIOTIC CHOCOLATE

Rosmawati, M.S. 1, Abdul Rahim, A.M.2 and Kasran, R.3 1Division of Biotechnology, Malaysian Cocoa Board, Cocoa Innovation & Technology Center, Lot PT12621, Nilai Industrial Area, 71800 Nilai, Negeri Sembilan. 2Well-Being LS (International) Sdn Bhd. 43200 Balakong Selangor, Malaysia. 31Malaysian Cocoa Board, Cocoa Biotechnology Research Centre, Kota Kinabalu Industrial Park, 88460 Kota Kinabalu, Sabah, Malaysia. [email protected]

Malaysian Cocoa Journal 9(1): 187-192 (2016) ABSTRACT - Food which contains probiotic bacterial culture will assist the body to re-establishing the proper balance necessary for optimal intestinal function and good health. Probiotic bacteria in the intestinal tract also can generate intense metabolic activity and are necessary for overall human health because they promote normal gastrointestinal function and protect against infection. Chocolate is a superior carrier for the intestinal delivery of probiotic bacteria. Chocolate embedded probiotics shown the survival rate of probiotic bacteria was three times higher than in the milk and yogurt drink (Rich Benson, Barry Callebaut). Our aims was to develop, analyses nutritional content of probiotic chocolate and to identify the amount of probiotic bacteria (Lactobacillus fermentum) in probiotic chocolate. Milk and dark probiotic chocolates were developed using 1 % to 10 % of lactic acid bacteria powder, cocoa liqour, cocoa butter, sugar, skimmed milk, vanilin and soya lecithin. Results showed that there was no significant difference observed in nutrition content of probiotic milk and dark chocolate compared to control chocolate. Microbiological studies of probiotic milk and dark chocolates showed the presence of Lactobacillus spp. to the extent of 107 to 108 cfu/g during the 12 month storage period. In conclusion, 10 g of probiotic chocolate per day is more than enough to ensure the optimal activity on the intestinal flora in parallel to its higher survival rates of chocolate-embedded probiotics in the human body.

Key words: Probiotic chocolate, Lactobacillus, Lactic acid bacteria, Sensory analysis, Shelf-life

INTRODUCTION Recent study shown that probiotics provide a number of health benefits mainly Probiotic foods are well known to provide several through maintenance of normal intestinal health benefits when administered in adequate microflora, protection against gastrointestinal amounts on the host (FAO/WHO, 2001). Most of pathogens (D’Aimmo et al., 2007; Lourens- the probiotic food products are categorized as Hattingh and Viljoen, 2001), enhancement of the functional foods as they help in maintaining a immune system (Gilliland, 1990), reduction of good balance and composition of intestinal flora serum cholesterol level and blood pressure and increase the resistance against invasion (Rasic, 2003), anti-carcinogenic activity (Rasic, pathogens. The demand of probiotic functional 2003), improved utilization of nutrients and foods is growing rapidly due to increased improved nutritional value of food (Lourens- awareness of consumers. The global market for Hattingh and Viljoen, 2001). Therapeutic functional foods and beverages has grown from applications of probiotics include prevention of $33 billion in 2000 to $176.7 billion in 2013 that infantile diarrhea, urinogenital diseases, accounts for 5 % of the overall food market, and osteoporosis, food allergy and atopic diseases; is the driving growth for the food industry as a reduction of antibody-induced diarrhea; whole (Granato et al., 2010; Hennessy, 2013). It alleviation of constipation and has been estimated that probiotic foods comprise hypercholesterolemia; control of inflammatory between 60 % and 70 % of the total functional bowel diseases; and protection against colon and food market (Holzapfel, 2006; Kołozyn- bladder cancer (Lourens-Hattingh and Viljoen, Krajewskaa and Dolatowski, 2012; Stanton et al., 2001; Mattila-Sandholm et al., 2002; Salminen, 2001). 1996; Venturi et al., 1999).

187

Chocolate is a superior carrier for the Moisture (%) = Weigh before drying – Weight intestinal delivery of probiotic bacteria. after drying/ Weight before drying x 100 Chocolate embedded probiotics shown the survival rate of probiotic bacteria was three times Ash content higher than in the milk and yogurt drink (Rich A dry ashing method according to AOAC 2000 Benson, Barry Callebaut). Development of was used to determine the ash content. Five (5) g chocolate with adequate doses of probiotics at the samples were placed in the crucible and time of consumption is a challenge, because incinerated in the furnace at 550±10°C for 12 hrs. several factors during processing and storage The remaining inorganic material was cooled in affect the viability of probiotic organisms. The the dessicator and weighed. Percentage of ash presence of probiotics in chocolate may also was obtained following equation below: adversely affect their quality and sensory properties. In this study, Lactobacillus spp. Total ash (%) = Weight of ash/Weight of sample bacteria was incorporated into milk and dark x 100 chocolate as probiotics and the nutritional information and storage affect to a minimum Fat content level [109 and 1011 CFU/day (Mombelli, B.; Determination of fat content was performed Gismondo, 2000)] of microorganisms required to following AOAC (2000), using a soxhlet provide positive health benefit from consumption extraction apparatus. 3-5 g of chocolate was will be identified. weighted and transfered to the thimbles in the soxhlet unit. The petroleum ether was used to extract fat in the sample and was heated for 14 MATERIALS AND METHODS hrs. The solvent was evaporated and the residue was weighed. Percentage of fat was obtained Development of probiotic milk and dark following equation below: chocolate Melted fat components and dry powders were Fat (%) = Weight of fat/Weight of sample x 100 mixed homogeneously in a kneader and the chocolate mass (fats) was pre-refined on a lab Crude protein content scale three roller refiner. Then, the mass was Protein content was determined by Kjeldahl transferred to a lab scale conch and the remaining method (AOAC 2000). Sample (0.5-1.0 g) was cocoa butter (refined separately) was added and digested for 2 hrs at 420C in Kjeldahl tube in the mixed for a total of 8 hrs. Powder containing presence of one (5 g) Kjeltec catalyst tablet and Lactobacillus fermentum was only added 2 hrs 200 ml of sulfuric acid (H2SO4). The mixture was before final stage of conching, followed by then cooled and mixed with 60 ml of distilled addition of emulsifier. Then, tempering of water coutiously. The mixture was immersed in finished chocolate mass took place manually on a 0.2 N asid hydrochloric (HCl) and mixed until all marble slab and the chocolate was moulded to a NH3 is distilled. After receiver was removed, 7-10 g shape. Finally, the demoulded chocolate excess HCl distilled was titrated with 40 % was stored at 16C. All samples produced met the natrium hydroxide (NaOH) solution. Percentage legal standard of ‘quality chocolate’. of nitrogen was calculated following equation below: Preparation of sample Probiotic milk and dark chocolate was melted at Protein (%) = (vol of HCl used in sample – vol of 45-50C and ready for analysis. HCl used in blank) x Normality of HCl x 14.007 x 6.25 / weight of sample Moisture content Moisture content of milk and dark probiotic Crude fibre content chocolate was measured using oven method Crude fibre content was determined according to (AOAC 2000). Three (3) g sample was weighted Hand book of Food Analysis, Part IX (1984). Two (2) g of sample was dried in an oven at 100 and dried in oven set at 105C and left for 3 hrs to dry. The percentage of moisture content was ± 2C for 4 hrs and transferred to a digestion calculated as: flask. 200 ml of boiling 1.25 % sulphuric acid (H2SO4) was added until all the sample was

188

thoroughly wetted and finally washed with RESULTS AND DISCUSSION boiling water until washings are acid free. Residue was washed back with 200 ml of boiling Composition of chocolate and storage 1.25 % sodium hydroxide (NaOH) solution using environment have significant influence on the wash bottle marked to deliver 200 ml. The flask survival of probiotics (Mattila-Sandholm et al., was then connected to a reflux condenser and 2002). Chocolate compositions such as sugar, boiled briskly for exactly 30 minutes. The flask milk, protein, fat, vitamin and antioxidant either was immediately removed and the content was direct or indirectly will promote growth of filterred through gooch prepared with asbestos probiotic (Korbekandi et al.,2011). These mat and carefully transferred all the residue into supplements have significant positive effects on the gooch with hot water. Residue was washed the survival of probiotic microorganisms during thoroughly with hot water until the filtrate was storage (Mohammadi et al., 2011). However, alkali free. Then, washed with about 10 ml different percentage of probiotic chocolate alcohol. Gooch crucible was dried at 110ºC, compositions such as cocoa liquor, cocoa powder, cooled and weighed (W1). The gooch was cocoa butter, sugar and milk powder used in transfered to a muffle furnace controlled at 525 - probiotic milk and dark chocolate did not shown 550ºC (W2). The weight loss represented crude significant different on the effect to the survival fibre and was calculated as below: of probiotic bacteria.

Crude fibre (%) = [(W1 – W2 / Weight of The nutritional composition of probiotic sample) x (100 x 100 / 100 – Moisture)] milk and dark chocolate was summarized in Table 1. Fat was the major constituent in the Total carbohydrate content probiotic milk and dark chocolate which was 47.2 The total carbohydrate content (%) in the samples % and 53.0 % respectively. It supported probiotic was calculated by substraction of the sums of the cells by reducing their exposure to detrimental weights of crude protein, total fat, moisture and factors (Karimi et al., 2011). The high fat content, ash (Anonymous, 2009). anaerobic environment and buffering capacity of the matrix in chocolate help to protect the Total carbohydrate (%) = 100 – (% moisture + % probiotic cells, both in the product and during ash + % fat + % protein) intestinal transit (Lee and Salminen, 2009). High fat content used in probiotic milk and dark Energy value chocolate formulation were contributed to higher The energy value was calculated based on energy value which was 609.3 % and 648.6 % Pearson (1970) suggestion by sum of the respectively. percentages of proteins and carbohydrates multiplied by a factor of 4 (kcal/g), total fats Table 1: Nutritional composition of probiotic multiplied by a factor of 9 (kcal/g) and total fibre milk and dark chocolate multiplied by a factor of 2 (kcal/g). TEST Probiotic Probiotic Total Energy = (% protein x 4) + (% (% dry basis) milk dark carbohydrate x 4) + (% fat x 9) + (% fibre x 2) chocolate chocolate Moisture 2.55 1.29 Lactobacillus fermentum content Ash 2.43 1.64 Content of Lactobacillus spp. was determined Fat 47.20 53.00 according to Compendium of Methods for the Crude protein 9.21 6.51 Microbiological Examination, Chapter 19. The Crude fibre 3.40 2.33 sample was prepared into several decimal Total 35.21 35.23 dilutions and incubated in MRS agar at 35 C for carbohydrate 72 hours. Number of colony forming units of Energy value 609.3 648.6 microorganisms per gram sample was calculated (kCal/100g) using following formula: The second major nutritional constituent N = C / (n1 + 0.1n2) x d x V in milk and dark probiotic chocolate was total carbohydrate which was 35.21 % and 35.23 %

189

respectively. Studies have shown that the Table 2: Probiotic (lactobacillus spp) content of presence of disaccharides can stabilize the cell probiotic milk and dark chocolate. membrane during storage (Carvalho et al., 2002; Önneby et al., 2013). It was supported research Product Probiotic Content cfu/g by Yoo and Lee (1993) that showed certain Formulation (35 C ± 1 C for 72 hrs) glucose can prevent membrane damage by 0 mth 6 mth 12 mth interaction with it, and stabilizes protein Probiotic Milk 1.8 x 1.6 x 1.8 x functionality and structure. Chocolate 108 107 108

Different growth promoters such as Probiotic Dark 1.9 x 2.2 x 2.1 x vitamins, minerals, protein and antioxidant in Chocolate 108 107 108 probiotic chocolate may also increase the growth rate of probiotic species. These supplements have significant positive effects on the survival of CONCLUSIONS probiotic microorganisms during storage (Mohammadi et al., 2011). Although protein The foremost challenge in probiotic chocolate is content in probiotic milk and dark chocolate was to maintain the appropriate numbers of probiotic relatively low (9.21 % and 6.51 %, respectively), bacteria in chocolate during processing and certain protein derivatives were also found to storage, as insufficient doses (107 to 109 cfu/g) at promote growth of the probiotic by providing the time of consumption will not provide the nutrition for the cells, by reducing redox potential intended health benefit. However, this study of the medium as well as increasing buffering showed that chocolate is the superior carrier for capacity of the medium that results in a smaller probiotic bacteria and provide longer survival decrease in pH (Dave and Shah, 1998; period. As a conclusion, 10 g of probiotic milk or Mortazavian et al., 2010). dark chocolate per day is more than enough to ensure the optimal activity on the intestinal flora The moisture content of probiotic to provide positive health benefits. products was another factor influencing shelf-life stability of live bacteria. Low moisture content in both probiotic milk and dark chocolate (2.55 % ACKNOWLEDGEMENT and 1.29 %, respectively) may prevented detrimental effect for bacterial survival (Önneby The authors would like to thank Malaysia Cocoa et al., 2013). Board (MCB) for providing facility and funding to this research. The authors also would like to The foremost challenge is maintaining thank Ms. Siti Azriena, Pn. Hannim and En the appropriate numbers of these probiotic Rahmat for their technical support. bacteria in milk and dark chocolate during processing and storage to provide optimal health benefits at the time of consumption. An REFERENCES appropriate mixture of chocolate ingredients provides the perfect medium for probiotics Anonymous (2009). Code of Federal Regulations, bacteria to growth and survive during storage Title 21, Part 101.9 – Food Nutrition study. Table 2 shown the content of probiotics Labeling of Food, United State bacteria in milk and dark chocolate during 12 mth Government Printing Office, study period. Washington, USA.

Probiotic milk and dark chocolates AOAC International (2000). Official Methods of showed the presence of Lactobacillus spp. to the Analysis of The Association of Official 7 8 extent of 10 to 10 cfu/g during 12 month storage Analytical Chemistry, 16th Edition, period. It showed that chocolate composition is Washington, USA. the superior carrier for probiotics bacteria over the product shelf life in order to maintain Carvalho, A. S., Silva, J., Ho, P., Teixeira, P., consumer confidence in probiotic chocolates. Malcata, F. X. and Gibbs, P. (2002). Survival of freeze-dried Lactobacillus

190

plantarum and Lactobacillus rhamnosus microorganisms in cheese during during storage in the presence of production and storage: A review. Dairy protectants. Biotechnology Letters 24: Science & Technology, 91: 283–308. 1587–1591. Kołozyn-Krajewskaa, D. and Dolatowski, Z. J. Compendium of Methods for the Microbiological (2012). Probiotic meat products and Examination (2015). Chapter 19, Apha human nutrition. Process Biochemistry Press, Washington DC. (Barking,London, England) 47: 1761– 1772. D’Aimmo, M. R., Modesto, M. and Biavati, B. (2007). Antibiotic resistance of lactic Lee, Y. K. and Salminen, S. (2009). Handbook of acid bacteria and Bifidobacterium spp. probiotics and prebiotics (2nd ed.). Isolated from dairy and pharmaceutical Hoboken, NJ: JohnWiley and Sons, Inc. products. International Journal of Food Microbiology 115(1): 35–42. Lourens-Hattingh, A. and Viljoen, B. C. (2001). Yogurt as probiotic carrier food. Dave, R. I. and Shah, N. P. (1998). Ingredient International Dairy Journal 11(1–2): 1– supplementation effects on viability of 17. probiotic bacteria in yogurt. Journal of Dairy Science 81: 2804–2816. Mattila-Sandholm, T., Myllarinen, P. M., Crittenden, R., Mogensen, G., Fonden, Gilliland, S. E. (1990). Health and nutritional R. and Saarela, M. (2002). benefits from lactic acid bacteria. FEMS Technological challenges for future Microbiology Letters 87(1–2): 175–188. probiotic foods. International Dairy Journal 12: 173–182. Granato, D., Branco, G. F., Cruz, A. G., Faria, J. A. F. And Nazzaro, F. (2010). Mombelli, B. and Gismondo, M.R. (2000). The Functional foods and nondairy probiotic use of probiotics in medical practice. Int. food development: Trends, concepts and J. Antimicrob. Agents 16: 531–536. products. Comprehensive Reviews in Food Science and Food Safety 9: 292– Mortazavian, A. M., Khosrokhvar, R., Rastegar, 302. H. and Mortazaei, G. R. (2010). Effects of dry matter standardization order on Hand book of Food Analysis (1984). Part IX, p. biochemical and microbiological 44. characteristics of freshly made probiotic Doogh (Iranian fermented milk drink). Hennessy, M., (2013). What’s driving growth in Italian Journal of Food Science 22: 98– functional food and beverages? A 102. convergence of nutrition, convenience and taste. Accessed 14 Feb 2014. survival of freezedried Sphingobium sp.cWorld Journal of Microbiology and Holzapfel,W. H. (2006). Introduction to Biotechnology 29(8): 1399–1408. prebiotics and probiotics. In I. Goktepe, V. K. Juneja, & M. Ahmedna (Eds.), Pearson, D. (1970). The Chemical Analysis of Probiotics in food safety and human Foods. 6th Edition, Churchill, London. health (pp. 1–35). NewYork: CRC Press. Probiotics Market - Global Industry Analysis, Market Size, Share, Trends, Analysis, Karimi, R., Mortazavian, A. M. and Cruz, A. G. Growth and Forecast, 2012—2018. (2011). Viability of probiotic Available online:

191

http://www.transparencymarketresearch. probiotics. The American Journal of com/ Probioticsmarket.html2013 Clinical Nutrition 73: 476S–483S. (accessed on 24 October 2014). Venturi, A., Gionchetti, P., Rizzello, F., Rasic, J. L. (2003). Microflora of the intestine Johansson, R., Zucconi, E. and Brigidi, probiotics. In B.Caballero, L. Trugo, and P. (1999). Impact on the composition of P. Finglas (eds.), Encyclopedia of food the faecal flora by a new probiotic sciences and nutrition (pp. 3911–3916). preparation: Preliminary data on Oxford: Academic Press. maintenance treatment of patients with ulcerative colitis. Alimentary Salminen, S. (1996). Uniqueness of probiotic Pharmacology & Therapeutics 13(8): strains. IDF Nutrition Newsletter 5: 16– 1103–1108. 18. Yoo, B. and Lee, C. M. (1993). Thermoprotective Stanton, C., Gardiner, G., Meehan, H., Collins, effect of sorbitol on proteins during K., Fitzgerald, G. F., Lynch, P. B. and dehydration. Journal of Agricultural and Ross, R. P. (2001). Market potential for Food Chemistry 41: 190–192.

192

Short Communication

MALAYSIAN RENAISSANCE COCOA: QUANTIFICATION OF FLAVANOLS AND METHYLXANTINES IN SELECTED MALAYSIAN COCOA CLONES

Azhar, M. and Kasran, R. Malaysian Cocoa Board, Centre for Cocoa Biotechnology Research, Commercial Zone 1, South KKIP, Norowot Road, 88460 Kota Kinabalu, Sabah

Malaysian Cocoa Journal 9(1): 193-194 (2016) ABSTRACT - Cocoa beans are considered to be a rich source of pharmacologically active compounds such as flavanols and methylxanthines. The determination of the levels of these compounds in cocoa is becoming increasingly important due to the awareness of the health benefit of these compounds. The aim of this study is to quantify flavanol (catechin and epicatechin) and methylxanthines (caffeine and theobromine) in selected Malaysian cocoa clones. 20 cocoa clones will be tested. Currently, 32 cocoa pods (4 pods per clone) were collected from 8 clones (KKM1, KKM4, KKM25, MCBC1, MCBC4, PBC123, QH1003 and QH1213). Sample preparation includes cleaning the beans, freeze drying, grinding, eliminating fat content and metabolite isolation via solvent extraction method. Samples will be analyzed using liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS). This method permits the simultaneous determination of catechin, epicatechin, caffeine and theobromine. These monomers are expected to be present in all samples and the concentration of these monomers are varied for each sample.

Key words: Caffeine, Catechin, Cocoa, Epicatechin, Theobromine

Raw cocoa bean is nutritionally the richest (Radojcic Redovnikovic et al., 2009). Many fraction of cocoa plant. It is estimated to contain studies have described these groups mainly about 380 known chemicals (Andújar et al., because of their pharmacological effects and are 2012). Scientists have started to identify and used as medications. study the specific chemicals and chemical combinations to better understand its effects. Since the seventeenth century, cocoa and Large numbers of chemical in cocoa are chocolate have been described as potential constantly being screened for their possible medicines (Andújar et. al., 2012). Many studies pharmacological value. In general terms, have described cocoa phenolics as being pharmacology is the science of how chemical bioactive compounds, especially prominent for agents interact with biological systems. their positive health effects for humans. Pharmacology studies have been performed worldwide for many years as part of the Polyphenols have gained much interest nonclinical evaluation of pharmaceuticals for due to its antioxidant capacity and possible human use. The objective ofthis study is to benefits to human health such as anti- quantify and compare the concentration of carcinogenic, anti-atherogenic, anti-ulcer, anti- catechin, epicatechin, caffeine and theobromine thrombotic, anti-inflammatory, immune in selected cocoa clones. modulating, anti-microbial, vasodilatory and analgesic effects (Hii et al., 2009). Studies have It is well known that cocoa bean is a rich demonstrated that cocoa has more polyphenols source of many structurally different biologically and flavonoids than black tea, green tea and red active compounds, among them polyphenolics wine (Lee et al., 2003). and xanthine alkaloids being predominant

Catechin Epicatechin

193

Other compounds present in cocoa are relaxation of smooth muscle (especially bronchial methylxanthines. Caffeine and theobromine are a muscle), and diuretic effects (Apgar and Tarka, class of methylxanthines that have been widely 1999; Gilbert, 2004). Furthermore, discussed in the literature. Methylxanthines have methylxanthines (caffeine, theophylline and known physiological and pharmacological effects aminophylline) are also commonly used as a in the body such as central nervous system medication for apnea of prematurity (Adén, stimulation, cardiac muscle stimulation, 2011).

Theobromine Caffeine

REFERENCES Kenny, T.P., Keen, C.L., Jones, P., Kung, H., Adén U. (2011). Methylxanthines during Schmitz, H.H. and Gershwin, M.E. pregnancy and early postnatal life. (2004). Cocoa procyanidins inhibit Handb Exp Pharmacol. 200: 373-89. proliferation and angiogenic signals in human dermal microvascular endothelial Andújar, I., Recio, M.C., Giner, R.M. and Ríos, cells following stimulation by low-level J.L. (2012). Cocoa Polyphenols and H202. Experimental Biology and Their Potential Benefits for Human Medicine 229: 765-771. Health. Article ID 906252, Oxidative Medicine and Cellular Longevity 2012: Lee, K.W., Kim, J.Y., Lee, H.J. and Lee, C.Y. 23 (2003). Cocoa has more phenolic phytochemicals and a higher antioxidant Apgar, J.L. and Tarka, S.M. (1999). capacity than teas and red wine. Journal Methylxanthines. In I. Knight (Ed.), of Agricultural and Food Chemistry, 51: Chocolate and cocoa: Health and 7292-7295. nutrition 153-173. Radojcic Redovnikovic, I., Delonga, K., Mazor, C.L. Hii, C.L. Law, S. Suzannah, Misnawi and S., Dragovic-Uzelac, V., Caric, M. and M. Cloke. (2009). Polyphenols in cocoa Vorkapic-Furac, J.( 2009). Polyphenolic (Theobroma cacao L.). Asian Journal of Content and Composition and Food and Agro-Industry. 2(04): 702- Antioxidative Activity of Different 722. Cocoa Liquors. Czech Journal of Food Sciences 01/2009; 27(5): 330-337. Gilbert, S.G. (2004). A small dose of toxicology: The health effects of common chemicals. Boca Raton, FL: CRC Press.

194