DNA Barcoding and Phylogenetic Assessment of Family Lamiaceae from Pakistan based on Plastid and Nuclear Sequence Data.

By Nadia Batool Zahra

Department of Biotechnology Faculty of Biological Sciences Quaid-i-Azam University Islamabad 2017

DNA Barcoding and Phylogenetic Assessment of Family Lamiaceae from Pakistan based on Plastid and Nuclear Sequence Data.

A thesis submitted in the partial fulfillment of requirements for the degree of Doctor of Philosophy In Plant Biotechnology By Nadia Batool Zahra

Department of Biotechnology Faculty of Biological Sciences Quaid-i-Azam University Islamabad 2017

DEDICATED TO MY LOVING PARENTS For their support & prayers

CONTENTS

Acknowledgements i Index of Figures iii Index of Tables v List of Abbreviations vi Abstract vii

Chapter 1 Introduction 1-31

1.1 Family Lamiaceae 1 1.2 Distribution 1 1.3 Medicinal and Economic Importance 2 1.4 Systematics 3 1.4.1 Subfamily Symphorematoideae Briq. 4 1.4.2 Subfamily Viticoideae Briq. 8 1.4.3 Subfamily Ajugoideae Kostel. 10 1.4.4 Subfamily Prostantheroideae Luerss. 12 1.4.5 Subfamily Scutellarioideae (Dumort.) Caruel 13 1.4.6 Subfamily Lamioideae Harley 14 1.4.7 Subfamily Nepetoideae (Dumort.) Luerss. 18 1.5 Molecular phylogenetics 20 1.5.1 Plastid and nuclear regions in angiosperm phylogeny 21 1.6 DNA Barcoding: A tool for standardization of herbal 22 medicinal products (HMPs) 23 1.6.1 What is a DNA Barcode?

1.6.2 Plastid and nuclear markers for DNA barcoding in plants 24

Maturase K gene (matK)

1.6.2.1 Ribulose-1,5-bisphosphate carboxylase/oxygenase large 25

subunit gene (rbcL)

1.6.2.2 trnH-psbA intergenic spacer 26 1.6.2.3 Internal transcribed spacer (ITS) 26

1.6.2.4 Challenges in DNA barcoding of medicinal plants/herbal 27 products 1.7 Research Objectives 31

Chapter 2 Materials & Methods 33-55

2.1 Section I: DNA Barcoding of herbal medicinal products 33 (HMPs) of Lamiaceae from Pakistan. 2.1.1 Sample collection from herbal markets/industry 33 2.1.2 DNA extraction from HMPs 33 2.1.2.1 Standard 2 X CTAB method 33 2.1.2.2 DNA isolation kit 38 2.1.3 Polymerase chain reaction (PCR) 38 2.1.3.1 PCR conditions for rbcL complete gene amplification 39 2.1.3.2 PCR conditions for matK partial gene amplification 39 2.1.3.2 PCR conditions for trnH-psbA spacer amplification 39 2.1.4 Purification of amplified products 40 2.1.5 DNA sequencing of barcoding loci 40 2.1.6 Nucleotide sequence data analysis 40 2.1.6.1 BLAST analysis 40 2.2 Section II: Phylogenetic utility of cpDNA rbcL, matK and 43 trnH-psbA regions 2.2.1 Sequence alignment and super matrix assembly 43 2.2.2 Data set construction 43 2.2.3 Phylogenetic analysis by RAxML method 43 2.3 Section III: Molecular phylogenetics of Lamiaceae based 44 on plastid (trnL-trnF) and nuclear (ITS) markers 2.3.1 Taxon sampling from wild and herbarium 44 2.3.2 DNA extraction from wild and herbarium specimen’s 44 2.3.2.1 2 X CTAB protocol 44 2.3.2.2 DNA extraction from herbarium samples 50 2.3.3 PCR 51 2.3.3.1 PCR parameters for trnL-trnF region 51 2.3.3.2 PCR parameters for ITS region 51 2.3.4 Purification of amplified products and DNA sequencing 51 2.3.5 Nucleotide sequence data analysis 54 2.3.5.1 Sequence alignment and super matrix assembly 54 2.3.5.2 Data set construction 54 2.3.6 Phylogenetic analysis 54 2.3.6.1 RAxML 55 2.3.6.2 Bayesian Inference (BI) 55

Chapter 3 Results 56-102

3.1 Section I: DNA Barcoding of herbal medicinal products 56 (HMPs) of Lamiaceae from Pakistan. 3.1.1 DNA extraction from HMPs and PCR of rbcL, matK and 56 trnH-psbA regions 3.1.2 Purification, sequencing and data analysis 56 3.1.3 BLAST analysis 56 3.1.3.1 rbcL gene analysis 58 3.1.3.2 matK nucleotide sequence analysis 62 3.1.3.3 trnH-psbA spacer region analysis 62 3.2 Section II: Evaluation of phylogenetic utility of cpDNA 65 rbcL, matK and trnH-psbA regions 3.3 Section III: Molecular phylogenetics of Lamiaceae based 79 on plastid (trnL-trnF) and nuclear (ITS) markers 3.3.1 DNA extraction and PCR 79 3.3.2 Purification, sequencing and data analysis 79 3.3.3 Phylogenetic analysis based on plastid trnL-trnF region 79 3.3.4 Phylogenetic analysis based on nuclear ITS region 92

Chapter 4 Discussion 103

4.1 Section I: DNA Barcoding of herbal medicinal products 103 (HMPs) of Lamiaceae from Pakistan. 4.1.1 Difficulty in DNA extraction and amplification of HMPs 103 4.1.2 Insufficient reference sequence data 104 4.1.3 Pros and cons of rbcL, matK and trnH-psbA regions 105 4.1.4 Challenges, improvements and regional recommendations for 105 DNA barcoding of HMPs 4.2 Section II: Evaluation of phylogenetic utility of cpDNA 109 rbcL, matK and trnH-psbA regions 4.3 Section III: Molecular phylogenetics of Lamiaceae based 110 on plastid (trnL-trnF) and nuclear (ITS) markers 4.3.1 Viticoideae 111 4.3.2 Scutellarioideae 112 4.3.3 Ajugoideae 112 4.3.4 Prostantheroideae 113 4.3.5 Symphorematoideae 113 4.3.6 Nepetoideae 114 4.3.7 Lamioideae 116 4.4 Conclusion 117 4.5 Future Recommendations 119

Publications 121

Literature Cited 122

ACKNOWLEDGEMENTS

All praise to Allah Almighty, the most beneficent, the most merciful, who gave me strength and enabled me to undertake and execute this research task.

I feel highly privileged in taking opportunity to express my deep sense of gratitude to my supervisor Dr. Zabta Shinwari, Professor, Department of Biotechnology. He was my greatest strength and my supreme mentor who not only supervised my work but also polished every aspect of my personality. I am thankful to him for his inspiration, reassurance and counseling from time to time and for his scholastic guidance and valuable suggestions throughout the study.

I wish to express my heartfelt appreciation to Dr. Melanie Schori and Prof. Dr. Allan Showalter who were a source of guidance for completion of my research work at Environmental and Plant Biology Lab, Ohio University, Ohio, US. I would also like to pay my cordial thanks to Dr. Evgeny Mavrodiev, Prof. Dr. Douglas Soltis and Prof. Dr. Pamela Soltis for their cooperation, guidance, support and valuable advices they offered for completion of my research at Laboratory of Molecular Systematics & Evolutionary Genetics, University of Florida, Florida, US. It gives me great pleasure to express my gratitude to Higher Education Commission of Pakistan, for providing Indigenous and IRSIP scholarships.

I would like to extend my deepest appreciation to those people, who helped me in one way or other in planning and executing this research work and writing up this thesis manuscript. I am obliged to Prof. Dr Wasim Ahmad, Dean, Biological Sciences and Dr. Muhammad Naeem, Chairman, Department of Biotechnology, Faculty of Biological Sciences, Quaid-i- Azam University, Islamabad for extending the research facilities of the department to accomplish this work. I am grateful to Faculty Department of Biotechnology, Quaid-i-Azam University, Islamabad for their help and support. Many thanks to Dr Muhammad Iqbal, Dr Mushtaq Ahmad, Dr Zafar Mahmood, Dr Anjum Parveen, Dr Tariq Mahmood, Dr Muhammad Ali and Mr Abdul Majid for extending their valuable advices and providing research material support. Sincere thanks to all my senior and junior lab fellows who were part of this journey at Molecular Sysytematics & Applied Ethnobotany Laboratory, Quaid-i- Azam University, Islamabad.

My earnest gratitude to Dr Anwar Nasim for his continuous moral support and encouragement. Special thanks to Sohail Irshad and Muhammad Bilal Khan who happily extended their assistance at every instance.

Solemn gratitude to my parents and my siblings who deserve special mention for their inseparable support and prayers. They have always been my source of strength and love. It wouldn’t have been this bearable if I didn’t have them in my life. Thank you for your unconditional support with my studies. Thank you for giving me a chance to prove and improve myself through all walks of my life.

Nadia Zahra

INDEX OF FIGURES

Figure Title Page

Fig 1.1 Clasification of family Lamiaceae proposed by Harley & colleagues 5 (2004). Fig 1.2 Flow chart representation of general transportation chain involved in 31 medicinal plants business in Pakistan. Fig 1.3 A double helix to show the relationship between classical taxonomy 32 and molecular database. Fig 2.1 Local herb store, Abpara, Islamabad, Pakistan. 34

Fig 2.2 Visit to herbal pharmaceutical industry, Pakistan. 34

Fig 2.3 Photograph showing the usual packaging of HMPs in plastic bags 35 purchased from Pansar stores. Fig 2.4 Field visit to Northern regions of Pakistan for collecting Lamiaceae 45 species.

Fig 2.5 Showing locations of primers used to amplify different regions of 53 plastid (trnL-trnF) genome. Fig 2.6 Showing locations of primers used to amplify different regions of 53 nuclear (ITS) genome. Fig 3.1 Visualizations of extracted genomic DNA and amplified products of 57 DNA barcoding loci. Fig 3.2 Performance/Percentage success of DNA barcoding loci for 64 Lamiaceae. Fig 3.3 DNA barcodes of 32 HMPs belonging to Lamiaceae, collected from 64 local herbal stores and herbal pharmaceutical industry. Fig 3.4 The RAxML phylogenetic tree based on rbcL gene sequences of 66 Lamiaceae. Fig 3.5 The RAxML phylogenetic tree based on matK gene sequences of 70 Lamiaceae. Fig 3.6 The RAxML phylogenetic tree based on trnH-psbA spacer 73 sequences of Lamiaceae. Fig 3.7 Comparison of rbcL, matK and trnH-psbA region alignments. 78 Fig 3.8 Visualizations of extracted genomic DNA and amplified products of 80 plastid and nuclear regions. Fig 3.9 The RAxML phylogenetic tree based on sequences of trnL-trnF 84 region of Lamiaceae. Fig 3.10 Bayesian 50% majority rule consensus tree of Lamiaceae based on 88 plastid trnL-trnF sequence data. Fig 3.11 The RAxML phylogenetic tree based on ITS sequences of 95 Lamiaceae. Fig 3.12 Bayesian 50% majority rule consensus tree of Lamiaceae based on 99 nuclear ITS sequence data.

INDEX OF TABLES

Table Title Page

Table 1.1 Comparison of major classifications of family Lamiaceae. 6 Table 2.1 List of the samples collected/purchased from herbal stores and 36 herbal pharmaceutical industry in Islamabad, Pakistan with vouchers information. Table 2.2 Primers used in DNA barcoding of Lamiaceae HMPs. 42 Table 2.3 List of species collected from wild (MOSEL NZ), Herbarium of 46 Pakistan (ISL) Quaid-i-Azam University and Karachi University Herbarium (KUH). Table 2.4 Primers used for amplification of plastid trnL-trnF and nuclear ITS 52 regions. Table 3.1 List of the HMPs used including scientific names and taxonomic ID 59 based on MEGABLAST.

LIST OF ABBREVIATIONS

APG Angiosperm Phylogeny Group BI Bayesian Inference bp Base Pair BS Bootstrap CTAB Cetyl Trimethyl Ammonium Bromide DNA Deoxyribonucleic acid EDTA Ethylene Diamine Tetra Acetic Acid HMPs Herbal Medicinal Products ISL Islamabad Herbarium ITS Internal Transcribed Spacer KUH Karachi University Herbarium matK Maturase Kinase MCMC Monte Carlo Markov chain

MgCl2 Magnesium Chloride ML Maximum Likelihood MOSEL Molecular Systematics & Applied Ethnobotany Laboratory MP Maximum Parsimony NaCl Sodium Chloride NCBI National Centre for Biotechnology Information NJ Neighbour Joining PCR Polymerase Chain Reaction PP Posterior Probability RAxML Randomized Axelerated Maximum Likelihood rbcL Ribulose 1-5 biphosphate carboxylase/oxygenase large subunit TE Tris Ethylene Diamine Tetra Acetic Acid trnH-psbA tRNA-His & photosystem II protein D1 spacer UPGMA Unweighted Pair Group Method with Arithmetic Means UV Ultra Violet

Abstract

The plant family Lamiaceae (Mints) contains ca. 7173 species distributed among 273 genera worldwide. It has highly varied phenotypic characters. Many species are of horticultural and economic significance and most importantly the family represents a wealth of medicinaly important aromatic herbs e.g Lavandula (Lavender), Mentha (Mint), Origanum (Oregano, Marjoram), Salvia (Sage), Rosmarinus (Rosemary), Melissa (Lemon balm) and Thymus (Thyme). There are 60 genera and about 212 species of Lamiaceae reported from Pakistan. This is the first ever investigation from Pakistan based on DNA barcoding identification of herbal medicinal products (HMPs) of family Lamiaceae and its subsequent phylogenetic assessment based on DNA sequence analysis. The study is divided into three sections:

DNA barcoding of HMPs belonging to Lamiaceae is conducted for their correct identification and to fix the problem of adulteration in the first section of present work. It can help in shielding consumers from the health hazards connected with the potential contamination and substitution of herbal medicinal products. HMPs representing 32 Lamiaceae plant samples were collected from three Pansar stores (herb stores) located at Islamabad and a herbal pharmaceutical industry. The extraction of total genomic DNA was carried out from these HMPs. Three plastid barcoding loci (rbcL, matK and trnH-psbA) were selected for PCR amplification and nucleotide sequencing was carried out. The comparison of DNA sequences obtained from these loci was carried out to ascertain the taxonomic identification of the plant material. We identified four mislabeled samples including one industry sample (Salvia haematodes) and three pansar store samples (Leucas linifolia, Lycopus europaeus and Salvia moorcroftiana II). Additionaly two product substitutions are also found in which Hyssopus officinalis is replaced by Nepeta bracteata and Nepeta ruderalis is substituted by Salvia spp. The identification of three HMPs (Lallemantia royleana, Origanum vulgare and Salvia aegyptiaca) is highly ambiguous because of lack of reference sequences available in GenBank. In Lamiaceae the overall amplification success for rbcL is 87% and for matK it is 81% while trnH-psbA showed 69%. Post sequencing analysis showed that trnH-psbA and matK have been able to discriminate the species relatively better with 40% success rate than rbcL (16%). On the whole a total of 22 sequences are genus-level barcodes (78%) and 12 sequences are species-level barcodes (44%). The nucleotide sequence data produced from the current study has been published in GenBank under the accession numbers KP172036-KP172082, KP218929- KP218945. We performed a comparative analysis of rbcL, matK and trnH-psbA to evaluate their performance for Lamiaceae barcoding. Our findings suggest matK as the potential barcode for Lamiaceae HMPs. The species-level identification was considerably challanging due to insufficient reference data and selection of plastid markers. Therefore, it is recommended for herbal pharmaceutical industries to develop a local (regional) herbal barcode library for their species of interest. The method of DNA barcoding can greatly assist the regulatory authorities and herbal industries to devise a mechanism for quality control and customer care. It can largely support the herbal pharmaceutical industries to restore the eroded consumer confidence.

In second part of the study, phylogenetic utility of three barcoding regions (rbcL, matK & trnH-psbA) is estimated. The data sets were comprised of 245 rbcL ingroup taxa, 235 for matK and 259 for trnH-psbA. This sequence data was acquired from GenBank including the accessions produced in the first part of work. We found that among these three selected markers matK seems to be the best gene region which is able to resolve the subfamilies and provide strongly supported monophyletic genera. rbcL was not able to resolve the clades with a strong bootstrap (BS) support. It was difficult to align the spacer region trnH-psbA across the whole family; as a result it affected the tree topology and could not produce the well resolved clades of subfamilies.

The third section of present work is the phylogenetic analysis based on plastid trnL-trnF and nuclear ribosomal internal transcribed spacer (nrITS) regions. A sum of 89 taxa was collected by visiting different wild areas and two herbaria of Pakistan. Additional publicly available data was downloaded from GenBank and the data sets were constructed. The data sets constitute 398 trnL-trnF taxa and 413 ITS ingroup taxa. The phylogenetic analyses were performed on the trnL-trnF and ITS data matrices by utilizing methods of ML (Maximum Likelihood) and BI (Bayesian Inference). We observed the Bayesian consensus trees showed more resolved nodes in comparison to ML consensus trees. The subfamilies received strong bootstrap values in the BI as compared to the ML results. In Pakistan’s Lamiaceae species, hybridization was observed, particularly evident in the nuclear ITS analysis. The analysis of taxa collected from Pakistan revealed that these species are undergoing possible radiation in place instead of dispersal. The taxonomic position of some species from Pakistan which were originally based on morphological characters did not corroborate with the findings of current molecular analysis. Therefore, it would be interesting to explore more plastid and nuclear loci with increased number of species for each group. Such approach will provide improved insight into relationships of mints from Pakistan. The intense studies more focused on each group (each subfamily) may draw a better picture of Pakistan’s Lamiaceae.

Introduction

1.1 Family Lamiaceae

The plant family Lamiaceae Martinov is also called as Labiatae Adans. Another common name is Mint family and it comprises of herbs perennial or annual, shrubs, subshrubs, trees and vines. These may be aromatic or not, nearly cosmopolitan distribution contains about 7173 species distributed among 273 genera (Harley et al., 2004). Mint family is considered the 6th largest family of angiosperms (Drew and Sytsma, 2012). Lamiaceae has highly varied phenotypic characters, but found with often hairy leaves with epidermal glands that function in secreting volatile oils. These oils give characteristic scents to many members. Many members are of horticultural and economic significance. There are seven subfamilies of Lamiaceae classified by Harley et al. (2004) (1) Symphorematoideae Briq. (2) Viticoideae Briq. (3) Ajugoideae Kostel. (4) Prostantheroideae Luerss. (5) Scutellarioideae (Dumort) Caruel (6) Lamioideae Harley (7) Nepetoideae (Dumort.) Luerss.

1.2 Distribution

The members of Lamiaceae are largely found in the temperate regions of the world including Mediterranean region and tropical savannas but not inhabiting the coldest regions of high altitude or high latitude. Hedge (1992) recognized six regions of high species diversity in Lamiaceae

1- Mediterranean & SW Central Asia; 2- Africa South of the Sahel & Madagascar; 3- China; 4- Australia; 5- South America; 6- Northern America & Mexico. Another region was added by Harley et al. (2004) known as Indomalesian region (SE Asia).

The biogeographical importance of Pakistan is evident in South Asia as it is bordered by Iran on the West, Afghanistan on the Northwest, Northeast has China, East occupied by India and South has The Arabian Sea. Pakistan is situated across one of the prime disjunctions in the biota of Southern Asia, and the line of delineation marked along the Western side of The Indus Basin and the Indus valley of Kohistan (Frodin, 1984). Hedge (1990) reported 60 genera and about 212 species of Lamiaceae in Pakistan. The genera belonging to subfamily Symphorematoideae and Prostantheroideae has not been reported from Pakistan.

1.3 Medicinal and economic importance

Lamiaceae represents a wealth of medicinaly important aromatic herbs e.g Lavandula (Lavender), Mentha (Mint), Origanum (Oregano, Marjoram), Salvia (Sage), Rosmarinus (Rosemary), Melissa (Lemon balm) and Thymus (Thyme), several species species are commonly found in Mediterannean and numerous regions of Asia (Ali et al., 2000; Celiktas et al., Hussain et al., 2008; Wink, 2003). Since ancient times, these species have been largely used as teas, spices and traditional medicines. Oregano is an important seasoning mix used in large quantities worldwide in pizza seasonings. Others include Rosemary, Lemon balm, Sage and Thyme consumed in Mediterranean region as typical seasonings (Baser, 2002; Esen et al., 2007). O. basilicum, O. tenuiflorum (basil) and P. amboinicus are the seasoning herbs of tropics.

Some Lamiaceae members are described for their toxic components; however, for more than 75 years many other members of this family are being consumed in culinary and essential oil industries (Martins et al., 1999; Manosroia et al., 2006). Today, the essential oils of several Mint species which are contained in their leaves have become a major raw material for the pharma, cosmetics and food industry (Burt, 2004; Edris, 2007). The US FDA’s Generally Recommended as Safe (GRAS) list includes Sage, Rosemary, Thyme and Oregano. There are many diseases like bronchitis and intestinal disorders wherein the essential oils of Lamiaceae plants are useful for treatment (Burt, 2004; Baris et al., 2006).

Many viticoids (members of Viticoideae) are big forest trees including Gmelina & Vitex, which have considerable commercial value because of their utility as timber. T. grandis (Incertae sedis), called as Teak commercially, is also a foremost timber tree. Lavandula and Pogostemon are utilized in toilet scents, perfumes and the oil of Perilla frutescens is extracted to be used in paints and varnishes. Heinrich (1992) reported that the Mints are largely utilized in traditional medical systems of North American Indian cultures and holds the third most important position in ethnobotany. Teucrium and Ajuga spp., have antifeedant compounds which are effective against the crop pests. L. cardiaca and Salvia spp. are found effective against the cardiac diseases (Rodriguez-Hahn et al., 1992) and there are few members of subfamily Lamioideae whose consumption in large amounts has been considered lethal. European gardens harbor over 50 genera alone being horticultural significant, however, several other genera are cultivated in different parts of the world including subfamily Symphorematoideae (Congea), Viticoideae (Callicarpa, Vitex), Ajugoideae (Ajuga, Caryopteris, Clerodendrum), Prostantheroideae (Prostanthera), Scutellarioideae (Holmskioldia, Scutellaria), Lamioideae (Phlomis, Physostegia, Lamium, Stachys) and Nepetoideae (Agastache, Monarda, Nepeta, Rosmarinus, Salvia, Thymus) (Harley et al., 2004).

There is a significant cultivation of Lamiaceae species in Pakistan along with the variety of wild species growing in different regions of the (Hedge, 1990). The occurrence of many Lamiaceae species has been reported from different localities of Pakistan. The cultivated crops include O. sanctum, O. basilicum, M. arvensis, M. piperita, T. vulgaris and T. linearis (Hussain et al., 2010). Frequent growth of other specis in mountainous terrains is also observed at different altitudes.

1.4 Systematics

The oldest complete taxonomic classification of Lamiaceae was proposed by Bentham (1832-1836) which was modified in 1876. Briquet (1895) brought improvements to Bentham’s classification. Another modification of this system was proposed by Melchior in 1964. Erdtman (1945) split Lamiaceae into two subfamilies: Lamioideae having tricolpate pollen shed-off in a 2-celled phase and subfamily Nepetoideae having hexacolpate pollen shed-off in a 3-celled phase. Wunderlich (1967) put forth a new system of classification which was built on Briquet’s system with many important modifications. A close relationship between Lamiaceae and Verbenaceae has long been recognized by Cronquist (1981) due to sharing many common characters. The distinguishing character is presence of deeply 4-lobed ovary having a gynobacic style (Lamiaceae) whereas an un-lobed ovary having a terminal style (Verbenaceae). However Cronquist (1988) proposed that boundaries between the two families are somewhat arbitrary and taxa with intermediate morphology are found in both families. It was supported by studies of Cantino (1992a, 1992b) that Lamiaceae was polyphyletic, having many clades born independent from Verbenaceae. He proposed the transfer of cymose genera of subfamily Chloanthoideae, subfamily Viticoideae, subfamily and tribe Monochileae from Verbenaceae to Lamiaceae. Cantino et al. (1992) published the list of accepted genera with their subfamilies, tribes and subtribes belonging to family Lamiaceae which was largely adopted by Thorne (1992). The phlogenetic trees of Lamiaceae and Verbenaceae constructed from rbcL showed the similar findings (Wagstaff and Olmstead, 1997). The sister families of Verbenaceae s. st. are Bignoniaceae and Martyniaceae instead of Lamiaceae, is reported by Olmstead et al. (2001) with weak bootstrap support.

The most recent full taxonomic hierarchy of Lamiaceae by Harley et al. (2004) was strongly based on morphological studies by Cantino and Sanders, 1986 & Cantino et al. 1992 and more recent molecular findings of Wagstaff et al. 1995, Wagstaff and Olmstead, 1997 & Wagstaff et al. 1998. Synapomorphies for the family include opposite leaves, a quadrangular stem, indumentums and hypogynous flowers, although there are rare irregularities in the first three traits (Harley et al., 2004). Harley et al. (2004) classified the Lamiaceae in seven subfamilies (Fig 1.1). A comparison among these major classifications of Lamiaceae is summarized in Table 1.1. The work performed in this dissertation is primarily following the classification proposed by Harley et al. (2004), hence providing below a brief account of the subfamilies proposed by them.

1.4.1 Subfamily Symphorematoideae Briq.

It is characterized by inflorescence 3-7 flowered capitate cymes with an involucre of bracts, ovary imperfectly 2-locular, ovules apical pendulous and fruit dry or subdrupaceous. The subfamily Symphorematoideae consists of only three genera viz. Sphenodesme Jack,

Subf. Lamioideae

Subf. Viticoideae Tri. Subf. Chloantheae Prostantheroide ae Tri. Subf. Westringieae Scutellarioideae Subt. Salviinae Lamiaceae Tri. Subt. Mentheae Menthinae Tri. Subt. Elsholtzieae Nepetinae Subt. Subf. Nepetoideae Lavandulinae Subt. Subf. Ajugoideae Hanceolinae Subf. Subt. Symphorematoide Tri. Ocimeae Hyptidinae ae Subt. Ociminae Subt. Plectranthina e

Fig 1.1: Clasification of family Lamiaceae proposed by Harley & colleagues (2004).

Subf. = Subfamily, Tri. = Tribe, Subt. = Subtribe

Table 1.1: Comparison of major classifications of family Lamiaceae.

Bentham Briquet Erdtman Wunderlich Cantino et al. Harley et al. (1876) (1895-1897) (1945) (1967) (1992) (2004) Subfam. Symphorematoideae Subfam. Viticoideae Subfam. Viticoideae Subfam. Chloanthoideae Tribe Subfam. Prostantherioideae Subfam. Prostantherioideae Subfam. Prostantherioideae Subfam. Prostantherioideae Prostanthereae Tribe Chloantheae Tribe Westringieae Tribe Prasieae Subfam. Prasioidea Tribe Ajugeae Subfam. Ajugoideae Subfam. Ajugoideae Subfam. Ajugoideae Subfam. Ajugoideae Tribe Ajugeae Tribe Rosmarineae Subfam. Teucrioideae Tribe Lamieae Subfam. Scutellarioideae Subfam. Scutellarioideae Subfam. Scutellarioideae Subfam. Scutellarioideae Subtr. Scutellariinae Subfam. Lamioidea Subfam. Lamioideae Subfam. Lamioideae Tribe Perilomieae Tribe Lamieae Subfam. Subfam. Lamioideae Subtr. Prunellineae Lamioideae Tribe Lamieae Subtr. Melittidinae Subtr. Melittidinae Subtr. Melittidinae Subtr. Lamiinae Subtr. Lamiinae Subtr. Lamiinae Tribe Marrubieae Tribe Marrubieae Subtr. Marrubinae Tribe Prasieae Subfam. Pogostemonoideae Tribe Nepeteae Tribe Nepeteae Subfam. Nepetoideae Subfam. Nepetoideae Subfam. Nepetoideae Tribe Nepeteae Tribe Salvieae Tribe Salvieae Tribe Salvieae Tribe Meriandreae Tribe Meriandreae Tribe Monardeae Tribe Monardeae

Tribe Mentheae Tribe Pogostemoneae Tribe Elsholtzieae Tribe Elsholtzieae Tribe Elsholtzieae Subtr. Pogostemoninae Tribe Mentheae Tribe Mentheae Tribe Mentheae Tribe Mentheae Subtr. Hyssopinae Subtr. Hyssopinae Subtr. Origaninae Subtr. Origaninae Subtr. Origaninae Subtr. Menthinae Subtr. Menthinae Subtr. Menthinae Subtr. Collinsonlinae Subtr. Collinsonlinae Subtr. Melissinae Subfam. Subtr. Melissinae Nepetoideae Tribe Prunelieae Subtr. Melissinae Tribe Glechoneae Tribe Glechoneae Tribe Hormineae Tribe Hormineae Tribe Lepechinieae Tribe Lepechinieae Subtr. Hormininae Subtr. Salviinae Subtr. Nepetinae

Tribe Ocimeae Tribe Rosmarineae Subtr. Lavandulinae Subfam. Lavanduloideae Tribe Lavanduleae Tribe Lavanduleae Subtr. Plectranthinae Subfam. Ocimoideae Tribe Ocimeae Tribe Ocimeae Tribe Ocimeae Subtr. Hyptidinae Subtr. Hyptidinae Subtr. Hyptidinae Subtr. Plectranthinae Subtr. Plectranthinae Subtr. Plectranthinae Subtr. Ociminae Subtr. Ociminae Subtr. Ociminae Subtr. Lavandulinae Subtr. Hanceolinae

Subfam. Catoferioideae Subfam. Catoferioideae

Symphorema Roxb. and Congea Roxb. found in South East Asia, India, Sri Lanka and Malaysia (Harley et al., 2004). Recognition of Symphorematoideae by Harley et al. (2004) is one of the main modifications to the traditional treatment (earlier treated as a separate family: Symphoremataceae Wight). Bentham (1876) & Briquet (1895-1897) ranked it as tribe Symphoremeae and subfamily Symphoremeoideae under the family Verbenaceae respectively. Junell (1934) recognized that the gynoecial structure of Congea (Symphoremataceae) was distinct from other Labiatae and Verbenaceae, but suggested its viticoid ancestory. Throne (1992) and Cantino et al. (1992) placed it as a separate family Symphoremataceae under suborder Lamiineae. Wagstaff et al. (1998) carried out a study where they showed Congea tomentosa nested within Labiatae s. l. based on rbcL analysis whereas it appeared at basal position in ndhF analysis and as a sister group to subfamily Nepetoideae based on combined analysis of ndhF and rbcL but concluded that the addition of more members from Symphoremataceae is required to further establish the relationship. Bendiksby et al. (2011) showed Congea as sister to a clade of viticoid genera.

1.4.2 Subfamily Viticoideae Briq.

Recognition of Viticoideae by Harley et al. (2004) within Lamiaceae is another of the major modifications to the traditional treatment where it has been part of Verbanaceae. Bentham (1876) treated it as tribe Viticeae of Verbenaceae. Briquet (1895-1897) placed it as subfamily Viticoideae belonging to Verbenaceae and divided the group into four tribes (Callicarpeae, Tectoneae, Viticeae, Clerodendreae) having the characteristic features of presence of cymes; hemianatropous ovules; whole fruit or divided into 4-10 locules and seeds without endosperm. Briquet’s Viticoideae included Vitex L., Gmelina L., Premna L., Callicarpa L., Cornutia L., Petitia Jacq., Tectona, Clerodendrum and other allied genera. Pieper (1928) stated that it was not yet possible to establish the exact generic boundries between Vitex and Premna. Junell (1934) made some modifications to Briquet's four tribes within the Viticoideae. He remarked the affinity of Viticipremna J. Lam, Tsoongia Merr. and Pseudocarpidium Millsp. to Vitex which was reported by Pieper (1928). In addition, Junell moved Peronema Jack, Hymenopyramis Wall. ex Griff. and Petraeovitex Oliv. from the subfamily Caryopteridoideae into the Viticeae based on the characteristic feature of Caryopteridoideae where fruits split easily into four. These genera have fruit which does not split, therefore transferred to Viticeae. Teijsmanniodendron was placed into its own tribe Teijsmanniodendreae, based on it having fruit which is a one-celled, one-seeded indehiscent capsule (Koorders, 1904).

Throne (1992) and Cantino et al. (1992) recognized the subfamily Viticoideae as part of Labiatae for the first time which was later followed by Harley et al. (2004). Of the seven subfamilies proposed by Harley et al. 2004, the Viticoideae has been considered as the least satisfactory circumscribed, which is clearly paraphyletic or possibly polyphyletic as shown by morphological, phytochemical and molecular evidences. It became the part of Labiatae with the removal of several genera. Clerodendrum and Rotheca Raf. were transferred to form part of the Ajugoideae, while Tectona, Peronema, Hymenopyramis, Petraeovitex and Callicarpa became listed as incertae sedis. This Viticoideae composed of the following genera: Vitex, Premna, Teijsmanniodendron, Gmelina, Paravitex H.R. Fletcher, Tsoongia, Viticipremna, Petitia, Cornutia L. and Pseudocarpidium. Viticipremna, Tsoongia and Teijsmanniodendron are recommended to be closely aligned to Vitex in Cantino's (1992b) phylogenetica trees of morphological traits. Wagstaff et al. (1998) suggested through a molecular study that Gmelina, Premna form a clade while Vitex, Petitia form another, therefore, monophyly of Viticoideae was not supported. Olmstead’s unpublished data revealed Hymenopyramis sister relation to Petraeovitex and these genera are dissimilar from other Viticoid genera. Olmstead reported in this finding that Tectona and Callicarpa appear to be in basal position, wherein the later has a weak bootstrap support as sister clade to Prostantheroideae, while former received a weak support being sister clade to most of the family members. Therefore, exclusion of the said 4 genera and keeping them incertae sedis would help in making the Viticoideae more homogeneous. Wagstaff et al. (1998) showed the sister relationship of Callicarpa with Nepetoideae. In a latest investigation, Bramley et al. (2009) based upon the phylogenetic investigation of ITS & ndhF sequence data provided evidence that the Viticoideae is not monophyletic. According to this study the most well supported clade, the Vitex group, contains Vitex, Paravitex, Tsoongia, Viticipremna, Petitia and Teijsmanniodendron. The inclusion of Paravitex, Viticipremna and Tsoongia in a larger Vitex is supported by molecular and morphological evidences, therefore new combinations are being proposed. The generic status of Teijsmanniodendron and Petitia is not resolved and upheld yet in Bramley’s investigation. Bendiksby et al. (2011) also demonstrated Viticoideae as non-monophyletic. Currently, different studies are providing inconsistent results which prove that the whole group needs combination of approaches for further comprehensive studies.

1.4.3 Subfamily Ajugoideae Kostel.

Subfamily Ajugoideae (Teucrioideae) as proposed by Harley et al. (2004) has ca. 1000 species divided into 24 genera, cosmopolitan, but many temperate, and especially South East Asia to Australia. In earlier classifications it was known as tribe Ajugeae sensu Bentham and Briquet and subfamily Ajugoideae sensu Wunderlich and reported as polyphyletic by Cantino (1992b). The traditional family boundry between Lamiaceae and Verbenaceae was transcended into Teucrioideae recognized by Cantino et al. (1992). Wagstaff and Olmstead (1997); Wagstaff et al. (1998) reported Teucrioideae to be paraphyletic with Ajuga L. in Ajugoideae sensu Cantino et al. (1992) nested within it, proved by ndhF and rbcL sequences. The positioning of Ajuga and its allied genera in subfamily Teucrioideae was suggested based on these findings. In later study, together they appeared to be monophyletic in Lamiaceae s. I. (Cantino et al., 1999). The name Ajugoideae has priority over Teucrioideae under the International Code of Botanical Nomenclature, therefore, corrected by Judd et al. (1999). The genera included in Ajugoideae are Rotheca, Clerodendrum, Aegiphila Jacq., Teucridium Hook. f., Teucrium L., Ajuga, Pseudocaryopteris P.D. Cantino, Schnabelia Hand.-Mazz., Trichostema L., Caryopteris Bunge, Faradaya F. Muell., Oxera Labill. and relatives.

Throughout the taxonomic history of Clerodendrum s. l., it is grouped among twelve different genera which are sometime placed in different families (De Necker, 1790). Clerodendrum L. which is a diverse genus having about 580 species widely distributed in America, Australia, Asia and Africa has highly variable morphological and cytological characters among the species which suggests its paraphyletic or polyphyletic origin (El Mokni et al., 2013). Through the advanced molecular systematic approaches, the delimitation of Clerodendrum still is in the process of modification. Based on the cpDNA restriction site analysis performed by Steane et al. (1997) and ITS sequences of Steane et al. (1999) concluded in the transfer of a large number of species from Clerodendrum s. l. to genus Rotheca (Steane and Mabberley, 1998). ndhF gene provided preliminary evidence that Clerodendrum is polyphyletic (Steane et al., 1997). Based on morphological characters like size of leaves, length of corolla and inflorescence type, few reports have divided the genus into 2 main sub-genera, Cyclonema (Hochst.) Guirke & Clerodendrum (Steane et al., 1999). Steane et al. (2004) reported the morphological similarity of Aegiphila, Amasonia L. f., Huxleya Ewart and Rees, and Kalaharia Baillon to Clerodendrum. Based on the molecular analysis reported by Steane et al., (2004) Huxleya was placed into Clerodendrum to give birth to a new combination, Clerodendrum linifolium, already supported by De Kok et al. (2000) who brought out chemical and morphological similarities between Clerodendrum & Huxleya. Yuan et al. (2010b) segregated the genus Volkameria L. from a formerly polyphyletic Clerodendrum based on molecular analysis. Recently Barrabe et al. (2015) reassessed the relationships of Oxera and reported that Clerodendrum is sister to Oxera. They have placed polyphyletic Faradaya in synonymy with Oxera because Faradaya was found partly nested within Oxera.

Another interesting group belonging to the subfamily Ajugoideae s. l. is Caryopteris- Trichostema complex. The complex also includes monotypic or very small genera i.e Amethystea L., Discretitheca P. D. Cantino, Pseudocaryopteris, Tripora P. D. Cantino, Rubiteucris Kudo and Schnabelia Hand.-Mazz. Chen and Gilbert (1994); Li and Hedge (1994); Moldenke (1983) placed the Caryopteris and Schnabelia in Verbenaceae. All the genera belonging to this complex are Asiatic except Trichostema which is North American. Most of the Caryopteris is endemic to China (Pei and Chen, 1982; Abu- Asab et al., 1993; Chen and Gilbert, 1994). The genus is treated either in Verbenaceae (Clarke, 1885; Briquet, 1895; Moldenke, 1980; Jafri and Ghafoor, 1974; Long, 1999; Press et al., 2000; Rajendran and Daniel, 2002) or in Lamiaceae (Junell, 1934; Cantino et al., 1992; Thorne, 1992; Harley et al., 2004). Cantino (1992b) and Rimpler et al. (1992) suggested that the genus Caryopteris is para or polyphyletic. Cantino et al. (1999) based on non-molecular data as well as rbcL and ndhF sequences found the similar corroborating results. Amethystea is a monotypic genus while Rubiteucris and Schnabelia include two and five species respectively after their expansion by Cantino et al. (1999) who transferred species from Caryopteris. Similarly Discretitheca, Pseudocaryopteris and Tripora were disintegrated from Caryopteris s. l. in the same study to delimit the Caryopteris as monophyletic. The genera comprising Caryopteris- Trichostema complex are closely related based on shared characters like pollen morphology, androecial structure, corolla and fruit morphology (Abu- Asab and Cantino 1989; Cantino, 1992a; Abu- Asab et al., 1993). The close affinity of Caryopteris to Trichostema was reported by Rimpler et al. (1992) based on phytochemical and morphological characters. Molecular phylogenetic studies further proved the sister relationship between these two genera (Steane et al., 1997; Wagstaff and Olmstead, 1997; Wagstaff et al., 1998). The close ties of genera constituting the complex based on combined morphological and molecular analysis were supported by low bootstrap (Cantino et al., 1999), therefore, intriguing the further investigations. Caryopteris s. str., Pseudocaryopteris, Schnabelia and Trichostema appeared to be monophyletic while Caryopteris s. l. as polyphyletic in a molecular study of ndhF conducted by Huang (2008). Same study based on ndhF data proved that the sister group of Trichostema is Caryopteris, with Amethystea the next most closely related taxon but on the other hand the combined ITS and ndhF data with morphological data showed that the sister group of Trichostema is Amethystea. Therefore, suggesting the need of further probe into this Caryopteris-Trichostema complex. In a recent investigation by Shi et al. (2003) based on matK and ITS sequence data, Schnabelia is found to be close to some species of Caryopteris.

1.4.4 Subfamily Prostantheroideae Luerss.

Harley et al. (2004) divided the Australian subfamily Prostantheroideae into two tribes viz. tribe Chloantheae Benth. & Hook. f. comprising of 10 genera and tribe Westringieae Bartl. including 6 genera. Bentham (1876) classified it as tribe Prostanthereae but in the later classifications of Briquet (1895-1897) and Wunderlich (1967) it was ranked as subfamily Prostantheroideae. The Australian tribe Prostanthereae showed close relationships with Verbenaceae subfamily Chloanthoideae based on gynoecial similarities proposed by Junell (1934). Cantino et al. (1992) included the Verbenaceous subfamily Chloanthoideae to Labiatae s. l. which was adopted by Thorne (1992). Chloanthoideae, a primarily Australian group circumscribed by Cantino et al. (1992) included members of Prostanthereae (traditionally in Labiatae s. str.) and Chloanthoideae, plus Tectona (traditionally in Verbenaceae s. l.). Olmstead et al. (1998) also proposed the monophyly of Prostantheroideae. Recently Bendiksby et al. (2011); Li et al. (2012) showed the monophyly of Prostantheroideae, however, the analysis includes members of tribe Westringeae only. The most recent synopsis of Chloantheae was presented by Conn et al. (2011) based on morphological and molecular study. They presented a key to distinguish the genera of tribe Chloantheae.

Wrixonia F. Muell. is one of the six genera within the Australian endemic tribe Westringeae which includes only two species (Conn, 2004; Harley et al., 2004). The said genus has clear morphological resemblance with the genus Prostanthera Labill., the most populated genus of the tribe Westringieae. Phylogenetic analyses based on morphological characters showed that Wrixonia and Prostanthera (including Eichlerago) are sister taxa (Cantino, 1992b; Conn, 1992; Abu-Asab and Cantino, 1993). Cantino (1992b) also demonstrated that Wrixonia has closer affinity to Prostanthera section Prostanthera than to section Klanderia by one synapomorphy: the closed fruiting calyx. Wilson (2010) using nuclear (ETS) and plastid (trnT-F) DNA reported that Prostanthera appears to be paraphyletic in comparison to Wrixonia. Based on these findings, Wrixonia is reduced to the synonymy of Prostanthera in order to maintain a monophyletic Prostanthera (Wilson et al., 2012).

There are other two genera Hemigenia R. Br. and Microcorys R. Br. of tribe Westringieae reported to be polyphyletic by means of morphological and molecular analysis (Guerin, 2008). However further use of molecular markers and additional taxa are recommended to evaluate the complete implications on the taxonomy of these genera (Guerin, 2008).

1.4.5 Subfamily Scutellarioideae (Dumort.) Caruel

Addition of Holmskioldia Retz. in subfamily Scutellarioideae is one of the notable modifications done by Harley et al. (2004) in the traditional classification. The other four genera are Wenchengia C. Y. Wu & S. Chow, Renschia Vatke, Tinnea Kotschy ex Hook. f. and Scutellaria L. (Harley et al. 2004). Bentham placed Brazoria Engelm. ex A. Gray, Prunella L. and Cleonia L. in his subtribe Scutellariinae which were excluded by Briquet in his subfamily Scutellarioideae but he also excluded Perilomia which was correctly placed by Bentham. Therefore, Briquet’s Scutellarioideae was paraphyletic. Wunderlich’s Scutellarioideae was monophyletic as she included the Perilomia. Cantino (1992a) supported the monophyly of Scutellarioideae based on two synapomorphies, bilabiate calyx with entire rounded lips and fruits having a distinctive tuberculate surface. Wagstaff et al. (1998) reported Scutellarioideae as sister subfamily to Lamioideae & Pogostemonoideae according to their molecular systematic analysis.

The phylogenetic position of Wenchengia has long been controversial, though it is a monotypic genus. The characteristic features of Wenchengia are alternate leaves, racemose inflorescences, vascular funicles and slender stalks. Wu and Chow (1965) established a separate subfamily Wenchengioideae based on morphological uniqueness of Wenchengia and adopted by Wu and Li (1977); Li and Hedge (1994); Takhtajan (2009). Abu Asab and Cantino (1993) recommended the genus as incertae sedis based on their morphological cladistic analysis which showed Wenchengia belonging to or near to Ajugoideae but its position in Scutellarioideae appeared to be only one step less parsimonious. Ryding (1996) suggested to keep considering the Wenchengia incertae sedis based on his morphological observations. However, Wenchengia is placed in subfamily Scutellarioideae by Cantino in Harley et al. (2004). Li et al. (2012) conducted an investigation on phylogenetic position of Wenchengia within mint family and revealed that Wenchengia emerged as a sister to the Holmskioldia-Tinnea-Scutellaria clade based on ndhF and rbcL analysis. The placement of Wenchengia in subfamily Scutellarioideae is recommended by them with further support of morphological, anatomical and cytological features.

1.4.6 Subfamily Lamioideae Harley

Subfamily Lamioideae (including Pogostemonoideae) is the second largest species-rich subfamily among the seven subfamilies proposed by Harley et al. (2004) classification. It contains 63 genera and about 1260 species. It exhibits remarkable morphological and habit diversity and occur in several temperate and subtropics of the world. Nearly cosmopolitan distribution, however, concentrated in Eurasia and northern to tropical Africa. Few genera are also reported from cold temperate regions. Briquet (1895-1897) recognized the huge subfamily Lamioideae (Stachyoideae) by subsuming the Bentham’s tribe Nepeteae, Salvieae (Monardeae), Mentheae (Satureineae) and most of his Lamieae (Stachydeae). Erdtman (1945) divided Labiatae in 2 divisions – Lamioideae, which usually have tricolpate pollen shed at the two-celled stage, a character which they share with many of Verbenaceae and the other is Nepetoideae, having hexacolpate pollen shed at the three-celled stage. This division was further correlated with other characters including myxospermy, presence or absence of endosperm, embryo shape, as well as a number of phytochemical characters (Wunderlich, 1967; Cantino and Sanders, 1986). Cantino and Sanders (1986) could not find an evidence for monophyly of Lamioideae. Wunderlich (1967) recognized Lamioideae (Stachyoideae) comprising Bentham’s Prasieae, most of Bentham’s Lamieae and five other genera.

Cantino et al. (1992) could not draw clear distinction between Pogostemonoideae and Lamioideae, although they proposed Pogostemonoideae as a separate subfamily from Lamioideae. The characteristic feature of Pogostemonoideae is stamens of equal length, and Lamioid’s are marked by presence of laballenic acid and an unusual embryo sac. On the other hand, pericarp structure (Ryding, 1995) and pollen morphology (Abu-Asab and Cantino, 1994) provide no distinction between the two groups. The cpDNA molecular phylogeny provides a poor support for segregation of Pogostemonoideae and Lamioideae, whereas the monophyletic group consisting of both subfamilies is strongly supported (Wagstaff et al., 1998). In later classification of the Lamiaceae, pogostemonoid taxa have been subsumed into Lamioideae, but the suprageneric relationships among the Lamioideae remained poorly understood providing no tribal ranks (sensu Harley et al., 2004). Wink and Kuafmann (1996); Wagstaff & Olmstead (1997); Wagstaff et al. (1998) reported Scutellarioideae as the closest relatives of the Pogostemonoideae–Lamioideae clade based on molecular analysis.

Scheen et al. (2010) presented another phylogenetic investigation based on three plastid markers (trnL, trnL-trnF, rps16) analyzing 159 species belonging to 50 genera. They found strong support for monophyly of Lamioideae s. l. (i.e., including Pogostemonoideae) with Cymaria Benth. as its sister group. Subfamily Lamioideae has 9 tribes. 3 new tribes are added: Gomphostemmateae Scheen & Lindqvist, Phlomideae Mathiesen, and Leucadeae Scheen & Ryding. Rest of the 6 tribes are: Pogostemoneae Briq., Synandreae Raf., Stachydeae Dumort., Leonureae Dumort., Lamieae Coss. & Germ., and Marrubieae Vis. The genus Betonica L. is reestablished and confirmed by Dundar et al. (2012) . The findings recommended strongly that the genera Ballota L., Sideritis L., Leucas R. Br and Stachys L. are either paraphyletic or polyphyletic. Yet 16 genera remained unclassified at the tribal level due to formation of monogeneric groups (Betonica, Colquhounia Wall., Eriophyton Benth., Galeopsis L., Paraphlomis (Prain) Prain, Roylea Wall. ex Benth.) or unavailability of molecular evidence (Ajugoides Makino, Alajja Ikonn., Hypogomphia Bunge, Loxocalyx Hemsl., Matsumurella Makino, Metastachydium Airy Shaw ex C. Y. Wu & H. W. Li, Paralamium Dunn, Pseudomarrubium Popov, Stachyopsis Popov & Vved. and Sulaimania Hedge & Rech. f.).

Despite the reasonable progress in the Lamioideae phylogenetics which has been recently made, yet it is considered as one of the most poorly resolved subfamily of Lamiaceae. Only limited groups have undergone phylogenetic analysis e.g tribe Lamieae (Ryding, 2003), tribe Leucadeae (Ryding, 1998; Scheen and Albert, 2009), tribe Phlomoideae (Ryding, 2008; Pan et al., 2009), tribe Synandrea (Scheen et al., 2008), Sideritis (Barbar et al., 2000, 2002, 2007) and the indigenous Hawaiin Labiates (Lindqvist and Albert, 2002; Lindqvist et al., 2003).

Bendiksby et al. (2011) proposed a taxonomic update of subfamily Lamioideae based on four plastid markers whose main purpose was to focus the genera which were omitted in the phylogenetic investigation by Scheen et al. (2010). They made 13 new combinations at rank of species and one at subgenus, established a new tribe Paraphlomideae Bendiksby in which Paraphlomis, Matsumurella & Ajugoides were added. Only three genera (Metastachydium, Paralamium, Pseudomarrubium) remain unrepresented in this study, remaining 61 presently recognized genera of Lamioideae are investigated. The incertae sedis genera, Cymaria Benth. and Acrymia Prain forms a clade with Lamioideae which has a strong support for subfamily Scutellarioideae as its sister clade. Another incertae sedis genus, Garrettia H. R. Fletcher appears as the sister of this larger clade constituting these four groups. Cymaria, Acrymia and Garrettia have shown a close morphological relationship previously (Cantino, 1992a; Harley et al., 2004). However, due to obvious morphological differences, none of these genera fit into Lamioideae (Bendiksby et al., 2011). Bendiksby et al. (2013) amalgamated the Stachyopsis and Eriophyton and also transferred Stachys tibetica to this expanded Eriophyton now containing 11 species. The group is supported as monophyletic by molecular phylogenetic tree. The morphological characters featuring this expanded Eriophyton are presence of usually hairy anthers, prominent and apically rounded to slightly emarginate lateral lobes of the lower lip of the corolla and apically truncate or subtruncate nutlets.

Molecular phylogenetics of tribe Stachydeae has been recently investigated to confirm the monophyly and to better resolve the poorly understood relationships within the tribe (Salmaki et al., 2013). Tribe Stachydeae, or some of the genera constituting it, have been studied through molecular phylogenetics in the previous works (e.g. Lindqvist and Albert, 2002; Barber et al., 2002, 2007; Lindqvist et al., 2003; Scheen et al., 2010; Bendiksby et al., 2011; Roy et al., 2013). The complexity of Stachydeae is due to presence of paraphyletic genera, considerable morphological plasticity, a range of ploidy levels, and presumably frequent natural hybridization. Salmaki and colleagues (2013) carried out the analysis of nuclear and plastid DNA sequence data to recognize major evolutionary patterns and to assess taxonomic hypotheses within this largest tribe of Lamioideae. Both nuclear and plastid data corroborate monophyly of the tribe, with Melittis L. as sister to all remaining Stachydeae. Still the authors could not convert the taxonomy of Stachydeae into a more ‘natural’ treatment.

Tribe Gomphostemmateae comprises 46 species divided into three genera—Bostrychanthera Benth., Gomphostemma Wall. ex Benth. and Chelonopsis Miq., and have strong support for its monophyly (Scheen et al., 2010; Bendiksby et al., 2011; Xiang et al., 2013). Members of this clade tend to have relatively large, four-lobed corollas that are strongly dilated distally (Harley et al., 2004). Possible synapomorphies include similarities in fruit pericarp structure (Ryding, 1994a, b) and the apparent branching of the columellae in the pollen exine (Pozhidaev, 1989; Abu-Asab & Cantino, 1994), but the sample size in these studies was too limited to be conclusive. Xiang et al. (2013) recently proposed the transfer of Bostrychanthera to Chelonopsis based on molecular, morphological and cytological data. The first ever study of Lamioideae based on low-copy nuclear marker has been recently conducted by Roy and Lindqvist (2015) by using PPR locus. They found the results consistent with previously studied cpDNA data of Scheen et al. (2010) and Bendiksby et al. (2011b), however, observed some important discordance among the cpDNA and PPR data, suggesting increased taxon sampling and use of multiple independent nuclear loci for further studies.

Yao et al. (2016) proposed a new infrageneric classification of Pogostemon consisting of two subgenera.

1.4.7 Subfamily Nepetoideae (Dumort.) Luerss.

The subfamily Nepetoideae is made up of about 105 genera (Harley et al., 2004) and is the largest subfamily in the Lamiaceae (Wagstaff et al., 1995; Wagstaff et al., 1998; Paton et al., 2004). Hexacolpate pollen, gynobasic style, an investing embryo, presence of rosmarinic acid and exalbuminous seeds are the noteworthy synapomorphies through which it appeared to be monophyletic (Cantino and Sanders 1986, Harley et al., 2004). Other studies also reported it as monophyletic (Wagstaff et al., 1995; Wagstaff and Olmstead, 1997).

The tribal segregation of Nepetoideae varied primarily from classification to classification (Bentham, 1876; Briquet, 1895-1897; Wunderlich, 1967). Nepetoideae sensu Wunderlich corresponds closely to Erdtman’s Nepetoideae, the only difference between these two circumscriptions is the Wunderlich’s segregation of Catoferia Benth. to subfamily Catoferioideae. Cantino (1992) provided a detailed overview of these treatments. Cantino et al. (1992) provided a new hierarchy of subfamily based on morphological and molecular analysis. This study constituted 4 tribes Ocimeae, Elsholtzieae, Mentheae and Lavanduleae wherein the tribe Mentheae has the major modifications in contrast to the earlier taxonomic treatments. Harley et al. (2004) adopted these findings with slight modifications and recognized three tribes i.e. Mentheae, Ocimeae & Elsholtzieae with the Mentheae the largest, containing about 65 genera. Tribe Mentheae gives rise to 3 sub tribes: Nepetinae, Menthinae and Salviinae. They repositioned the Lavandula L. which was the only member of Lavandulinae, within tribe Ocimeae, along with 4 other sub-tribes: Plectranthinae, Ociminae, Hyptidinae and Hanceolinae. Subtribe Hanceolinae has been recently recognized and Isodon (Benth.) a primarily Asiatic genus is present in it. Schrader ex Spach, which often had been placed in Plectranthus L’Her. There are a number of molecular investigations which have been carried out within the Nepetoideae (Wagstaff et al., 1995; Prather et al., 2002; Paton et al., 2004; Trusty et al., 2004; Walker et al., 2004; Bräuchler et al., 2005; Edwards et al., 2006; Walker and Sytsma, 2007; Bräuchler et al., 2010).

Mentheae is not only the largest tribe of Lamiaceae regarding the number of genera and species but it exhibits diversity in occurence, habit, breeding system and floral form (Drew and Sytsma, 2012). Mentheae has undergone a number of molecular phylogenetic investigations (Wagstaff et al., 1995; Prather et al., 2002; Trusty et al., 2004; Paton et al., 2004; Walker et al., 2004; Bräuchler et al., 2005, 2010; Edwards et al., 2006; Walker and Sytsma, 2007; Drew and Sytsma, 2011), where Mentheae appeared to be monophyletic. Since the treatment proposed by Harley et al. (2004), several molecular (Trusty et al., 2004; Walker et al., 2004; Bräuchler et al., 2005, 2010; Edwards et al., 2006; Walker and Sytsma, 2007; Drew and Sytsma, 2011) and morphological (Moon et al., 2008, 2009, 2010; Ryding, 2010a, b) studies have focused on Mentheae and groups within it. These studies showed the non-monophyly of the three subtribes of Mentheae proposed by Harley et al. (2004) and reported that a number of genera remain unplaced/misplaced (Ryding, 2010a; Drew and Sytsma, 2011).

Generic boundaries in subtribe Menthinae have been under debate especially those taxa associated with the former Satureja s. l. complex (Satureja L., Micromeria Benth., Calamintha, Clinopodium L., Acinos). Many authors favored Briquet’s (1895–1897) broad concept of Satureja (Thonner, 1915; Brenan, 1954; Hedberg, 1957; Killick, 1961; Epling and Jativa, 1964, 1966; Greuter et al., 1986) while others (Chater and Guinea, 1972; Ball and Getliffe, 1972; Davis, 1982; Morales, 1993) were in favour of the narrow delimitation classified by Bentham (1848, 1876).

In the recent years, there have been an increasing number of molecular studies in Nepetoideae with focus on the tribes Ocimeae (Paton et al., 2004) and especially Mentheae. All of the latter were restricted to selected genera, e.g. Bystropogon L’Her. (Trusty et al., 2004, 2005), Conradina A. Gray (Edwards et al., 2006, 2008a, b), Mentha L. (Bunsawat et al., 2004), Micromeria (Bräuchler et al., 2005), Minthostachys (Schmidt-Lebuhn, 2007, 2008), Monarda L. (Prather et al., 2002) and Salvia L. (Walker et al., 2004; Walker and Sytsma, 2007) with some preliminary studies at the tribal level only.

Drew and Sytsma (2012) in their recent study based on cpDNA and nrDNA phylogenetics showed conflicts with the subtribal delimitation of Mentheae proposed by Harley et al. (2004). They showed the monophyly of Mentheae and proposed two new subtribes, Prunellinae and Lycopinae in addition to Harley’s.

Harley et al. (2004) treated ten genera as incertae sedis (Acrymia Prain, Callicarpa L., Cymaria Benth., Garrettia H.R.Fletcher, Holocheila (Kudô) S. Chow, Hymenopyramis Wall. ex Griff., Ombrocharis Hand.-Mazz., Peronema Jack, Petraeovitex Oliv., and Tectona L. f. These were not placed into any of the seven subfamilies.

Recently Chen et al. (2016) placed incertea sedis Ombrocharis in Nepetoideae, a placement that is also supported by its hexacolpate pollen grains. They demonstrated that Ombrocharis and another monotypic genus of Nepetoideae, Perillula, form a clade that is sister to the remaining genera of tribe Elsholtzieae. The monophyly of Elsholtzieae (including Ombrocharis) is well supported, there is weak support for Elsholtzieae as sister to the rest of Nepetoideae and Elsholtzia may be polyphyletic.

1.5 Molecular Phylogenetics

The obejective of molecular phylogenetics is to show the ability of DNA molecular markers for investigating a huge number of evolutionary and taxonomic problems. DNA sequences offer more reliable and robust analysis as there is no selection of multiple characters in cladistic analysis than morphological data which becomes biased due to character selection. Similarly, DNA sequences perform better in phylogenetic analysis due to their natural composition of discrete instead of continuous characters. There has been incredible increase Since 1981, in the studies based on molecular markers and phylogenies constructed through them since 1981 (Pagel, 1999).

The fundamentals of the modern phylogenetics were laid by Hennig (1950, 1966). The key phylogenetic concept is testing of monophyly, which has evolved as the basis of all modern classification systems, Angiosperms are being considered as the primary branch of the ‘Tree of Life’ which has been grouped fundamentally based upon the cladistic analyses of nucleotide (DNA) sequences (APG I, 1998; APG II, 2003; APG III, 2009; APG IV, 2016). There are different methods of phylogenetic tree reconstruction depending on the type of analysis required. Distance Methods transform aligned nucleotide sequences into a pairwise distance matrix for building phylogenies, whereas Discrete Methods take into account each nucleotide site individually. NJ and UPGMA methods are categorized in Distance methods. Max. Likelihood (ML), Max. Parsimony (MP) & Bayesian analysis are used for discrete methods. There are different algorithms which work behind these methods. However, different investigations have proved that the data which contains low noise and clear signals, significantly similar analysis results are obtained through most of these methods (Muellner et al., 2003).

1.5.1 Plastid and nuclear regions in angiosperm phylogeny

Over the past three decades there has seen a blossoming of molecular systematics to study angiosperm phylogeny. The two primary sources for phylogenetic purposes have been the chloroplast genome and the nuclear ribosomal DNA repeat region. The mitochondrial genome in plants has been of very less importance in phylogenetic studies. cpDNA regions are employed basically due to the ease of amplification and they provide enough variability for medium and high rank classification because of conservative rate of nucleotide substitution. Chloroplast genome is relatively abundant in plant cells and contains mostly single copy genes. Nuclear genome usually offer more variable molecular characters for evolutionary studies than chloroplast genome as some nuclear genes evolve more rapidly, providing the reason for better evaluation of the evolutionary connection rather among the more similar species (Sang, 2002; Small, 2004). Due to maternal inheritance of cpDNA markers, they are not fully capable of providing the true picture of hybridization and recombination. Secondly, because of their linked nature they behave as a single unit, which may turn the evolutionary patterns doubtful (Small et al., 2004; Steele et al., 2008). Therefore, in such instances single or low copy markers are recommended for being more informative because of their biparental inheritance and unlinked nature providing high sequence variability (Sang, 2002; Curto et al., 2012). Low-copy nuclear genes exhibit faster pace of evolution, although they have shown certain shortcomings as well (Yuan et al., 2009, 2010; Drew and Sytsma, 2013). They need specific design of primers and nucleotide amplification is frequently followed by nuclear cloning, making them time consuming and laborious.

To date there are numerous studies based on the phylogenetic analysis of family Lamiaceae. These studies include investigation of intrafamilial relationships which have helped to identify the membership of major monophyletic groups and provide good starting hypotheses for both intrafamilial relationships and improved generic circumscriptions. However, an overwhelming number of these studies cannot adequately resolve relationships within and among major clades. Consequently, phylogenetic relationships among most major lineages remain unresolved or poorly supported, presenting a significant roadblock to our common goals for understanding the evolutionary history of the family and revising classifications. Less well known are the relationships within each of the major lineages; phylogenetic analyses of closely related species regularly provides with the unresolved phylogenetic relationships and regular methods of resolving phylogenes carrying the significant addition of genes and species into the analysis have been proved highly ineffective.. Thus, progress in mint systematics remains stalled to resolve these difficult phylogenetic problems.

1.6 DNA Barcoding, A tool for standardization of herbal medicinal products (HMP’s)

The popularity of herbal medicine/products has increased worldwide during the past couple of decades. In many countries herbal medicine is not only regarded as a conventional treatment strategy but also acts as a health care system. This knowledge of traditional medicine based on utility of plants/herbs is playing a pivotal role in today’s drug development and biological research. The active compounds of medicinal herbs are used as a lead molecule for the discovery of new drug. Globally speaking, demand for medicinal plants, herbal medicinal products, food supplements, pharmaceuticals and health products are significantly growing (Sen and Chakraborty, 2015). By 2050, the estimated rise in global market for medicinal and aromatic plants is US $5 trillion which was US $62 billion in 2002, indicating a global paradigm shift from an allopathic to a traditional healthcare system (Shinwari, 2010). Despite the increasing demand and acceptance of medicinal plants, there are pitfalls in drug standardization, efficacy, safety, quality control, information & regulatory system. The very first step in the process of quality assurance is the authentic identification of the plant species.

The identification of herbs is traditionally carried out by morphological characters, but in case of cryptic species or phenotypically variable species the chances of misidentification are greater (Vijayan and Tsou, 2010). There can be serious consequences of use of a misidentified medicinal plant. For example, Foxglove (D. purpurea) could be substituted with Comfrey (S. officinale) due to morphological similarity in their leaves. Symphytum officinale is effective in the treatment of tendons injury and fractures while Digitalis purpurea can help treating the congestive heart failure due to the presence of cardiac glycosides. Ingesting the Symphytum officinale tea resulted the cardiotoxicity in nine individuals, some others even required temporary pacemakers for the regulation of heart beat (Lin et al., 2010). There are studies which proved the deliberate adulteration of herbal medicinal products with such additives which are not mentioned on the labels (Stoeckle et al., 2011; Newmaster et al., 2013; New York City Press Office, 2015). Such reports provided the significance of an effective and precise science integrated method for taxonomic identification of the medicinal plants and their HMPs. The molecular method which has made it possible to identify the herbs and to find the adulterants in HMPs is ‘DNA barcoding’.

1.6.1 What is a DNA Barcode?

DNA barcoding is defined as a method in which a standardized, short piece of DNA (< 1000 bp) having significant variablity is compared to a reference database through a sequence alignment algorithm for species identification (Shinwari et al., 2014). A perfect DNA barcode should meet the following 3 criterias; use of universal primers for amplification, bidirectional sequencing and high nucleotide variation among species. However, such criterion raised many questions regarding the purpose and nature of systematics and related subjects (DeSalle, 2006; Rubinoff et al., 2006). Within few years DNA barcoding took the shape of reality (Frezal and Leblois, 2008). Proving the universality and high resolution of DNA barcoding, several coding/non-coding DNA regions are being utilized as molecular markers. It is essential for a barcoding molecular marker to demonstrate high inter specific and low intra specific variation. This variation b/w inter & intra specific distances is termed as “DNA barcoding gap”. This method is proposed as an integrated wholesome method along with classical taxonomy for discrimination/identification and validation of species (Newmaster et al., 2009; Sahare and Srinivasu, 2012; Vohra and Khera, 2013).

1.6.2 Plastid and nuclear markers for DNA barcoding in plants

DNA barcoding is being considered as a simple potential way out to a complicated problem providing a method of quality assurance to authenticate the raw plant material. In animals the COX1 belonging to mitochondrial genome is regarded as a universal barcode marker, although in case of plants proposing a general barcode has remained indefineable (Kress and Erickson, 2008). In case of plants the quest for barcoding region diverted towards cpDNA (chloroplast) and nrDNA (nuclear) regions due to slow rate of evolution and limited divergence of mitochondrial DNA (Hollingsworth et al., 2011). The important candidate coding/non-coding regions for plant barcodes are ITS, rbcl, matK, trnH-psbA, trnL-F, 5S- rRNA & 8S-rRNA. However, because of variation in their performance, it was propsed that instead of single-locus plant barcode, multilocus combinations needs to be tested. CBOL (Consortium for the Barcode of Life) and the iBOL (International Barcode of Life) are the two imternational initiative working for the development of DNA barcodes. The standard gene markers that have been used for DNA barcoding are chloroplast rbcL (ribulose 1, 5- bisphosphate carboxylase/oxygenase large subunit) and matK (maturase K) as core barcodes (Hollingsworth et al., 2009). The spacer between photosystem II protein D1 and tRNA-His (psbA-trnH spacer) and the nuclear ribosomal internal transcribed spacer 2 (ITS2), however, are now suggested by several studies (Chen et al., 2010; Gao et al., 2010; Yao et al., 2010; Fu et al., 2011, Han et al., 2012; Newmaster et al., 2013; Michel et al., 2016).

1.6.3 Challenges in DNA barcoding of medicinal plants/herbal products

In the case of herbal plants/products, the universal protocols developed for the DNA barcoding can be adopted for barcoding of dried plants and the herbal supplements. There are challenges involved in the DNA barcoding of medicinal plants in terms of developing barcodes and the analysis of data to measure the distinguishing power (Cowan and Fay, 2012). Sometimes presence of secondary metabolites in plants hinders the successful PCR of barcoding regions. Although the solution to these problems is available in the form of modified isolation protocols, primer sequences and the adoption of engineered DNA polymerase. The combination of barcode sequences from multiple loci has also been used successfully.

In Pakistan about 70% of the population is dependent on the plant derived traditional medicine for primary health care system. In developed countries like Germany the 80% of population have used herbal medicine at least once. Therefore, to provide standardization and quality control of herbal medicines with established therapeutic value, resulting in an increase in demand, to revive a traditional healthcare system facing extinction is an important task that can be fulfilled by DNA barcoding of medicinal plants. In Pakistan, herbal plants are used largely often in the form of packaged medicine manufactured by herbal medicine industries and as raw herbs and their decoctions which are formulated by indigenous people by practicing their indigenous knowledge. The raw material is collected from the wild and transported to national and international markets. The long transportation chain (Fig 1.2), with many middlemen, results in increased events of misidentification and adulteration. Due to lack of proper monitoring and regulatory system there is a risk that many medicinal plants and their products available through the herbal stores/market are taxonomically either misidentified/mislabeled or substituted/contaminated. The aforesaid situation emphasizes need for an effective and efficient identification system through barcoding these medicinal plants. Hence, where the barcoding will allow the pharmaceutical industry and consumers in Pakistan to authenticate the raw material, it will also provide reference sequences to the scientific community. The current study was designed on DNA barcoding of medicinal plants of family Lamiaceae for their correct identification and to fix the problem of adulteration. Many species are used as traditional medicines, as culinary herbs and spices and as source of essential oils. However, it is often difficult to differentiate closely related groups due to morphological complexity of the Lamiaceae, which leads to many taxonomic problems. Many members of Lamiaceae are available in the domestic herbal markets of Pakistan as raw herbs and packaged herbal preparations manufactured by herbal pharmaceutical industries. DNA barcode identification of doubtful raw constituents can prove or invalidate the taxonomic identity of medicinal plants before they are sent for processing. It enables the pharma companies to build its consumer confidence (Schori & Showalter, 2011). Fig 1.3 demonstrates the relationship between classical taxonomy and DNA barcoding data.

1.7 Objectives

Following targets are set to be achieved in the current work:

i- To determine the utility of a regional approach of DNA barcoding for identification of medicinal plants of Lamiaceae in Pakistan. ii- To provide an evaluation of the performances of candidate barcoding loci (rbcL, matK, trnH-psbA) for Lamiaceae. iii- Assessing the phylogenetic utility of DNA barcoding loci. iv- To use the nuclear ribosomal DNA ITS and plastid DNA trnL-trnF sequence data to determine the monophyly of the Lamiaceae from Pakistan. v- To improve the intra-familial relationships within Lamiaceae based on molecular data.

Subf. Lamioideae

Subf. Viticoideae Tri. Subf. Chloantheae Prostantheroideae Tri. Subf. Westringieae Scutellarioideae Subt. Salviinae Lamiaceae Subt. Tri. Mentheae Menthinae Tri. Subt. Elsholtzieae Nepetinae Subt. Subf. Nepetoideae Lavandulinae Subt. Subf. Ajugoideae Hanceolinae Subf. Subt. Tri. Ocimeae Symphorematoideae Hyptidinae Subt. Ociminae Subt. Plectranthinae

Fig 1.1: Clasification of family Lamiaceae proposed by Harley & colleagues (2004).

Subf. = Subfamily, Tri. = Tribe, Subt. = Subtribe

Wild collector

Herbal/Pansar Middlemen stores

Consumers Traders

Herbal pharma Tabibs/Hakeems Export industries

Fig 1.2: Flow chart representation of general transportation chain involved in medicinal plants business in Pakistan.

Fig 1.3: A double helix to demonstrate the association b/w molecular database and classical taxonomy. The figure is divided into 3 parts having 3 different colors. The process of taxonomic revision is symbolized by pink color, workflow of barcode database construction is denoted by yellow color & blue color shows the use of Bioinformatics, classical taxonomy and DNA barcoding data by employing the utility of internet tools and management systems from large herbaria.

Materials & Methods

2.1 Section I: DNA Barcoding of Herbal Medicinal Products (HMP’s) of family Lamiaceae from Pakistan.

2.1.1 Sample collection from herbal markets/industry

HMP’s (Herbal medicinal products) representing 32 samples of plants belonging to family Lamiaceae were bought/collected from three Pansar stores (herbal stores) of Islamabad city and a herbal pharmaceutical industry (names withheld) (Fig 2.1 & Fig 2.2). These sampled HMPs came in plastic packs, already weighed and packed or weighed on-spot according to demand of customer (Fig 2.3). HMP’s samples included plant sections, twenty-three samples were leaves, two seeds, two inflorescence, four samples were a mix of different dried plant parts (shoots, flowers, leaves) and one sample was grounded plant material beyond recognition. The voucher specimen’s were made by mounting the plastic packets containing herbal product material on the herbarium sheets. These voucher specimens were deposited to the herbarium collection of MOSAEL Laboratory, Deptt. of Biotech ,QAU, Islamabad, Pakistan. Voucher ID’s & the related information is given in Table 2.1.

2.1.2 DNA extraction from HMPs

Plant DNA extraction was performed using modified methods as explained below:

2.1.2.1 Standard 2 X CTAB method

Total genomic DNA isolation was performed by a standard 2 X CTAB (Cety-trimethyl- ammonium bromide) method (Doyle, 1991): Liquid nitrogen was used for grounding 0.1 g

Fig 2.1: Local herb store, Abpara, Islamabad, Pakistan.

Fig 2.2: Visit to herbal pharmaceutical industry, Pakistan.

Fig 2.3: Photograph showing the usual packaging of HMPs in plastic bags purchased from Pansar stores.

Table 2.1: List of the samples collected/purchased from herbal stores and herbal pharmaceutical industry in Islamabad, Pakistan with vouchers information. Vouchers were deposited to MOSAEL Laboratory Deptt. of Biotech, QAU aIslamabad, Pakistan.

Serial # Species Voucher ID Source of collection (Islamabad)

1. Ajuga bracteosa MOSEL 265 Herbal industry

2. Ajuga parviflora MOSEL 285 Herbal industry

3. Hyssopus officinalis * Herbal store

4. Lallemantia royleana MOSEL 289 Herbal industry

5. Leucas cephalotes MOSEL 266 Herbal industry

6. Leucas linifolia * Herbal store

7. Lycopus europaeus * Herbal store

8. Melissa officinalis MOSEL 267 Herbal industry

9. Mentha aquatica MOSEL 269 Herbal industry

10. Mentha arvensis MOSEL 268 Herbal industry

11. Mentha x piperita MOSEL 271 Herbal industry

12. Mentha pulegium MOSEL 272 Herbal industry

13. Mentha longifolia MOSEL 270 Herbal industry

14. Mentha spicata MOSEL 274 Herbal industry

15. Mentha suaveolens MOSEL 273 Herbal industry

16. Nepeta cataria MOSEL 275 Herbal industry

17. Nepeta ruderalis * Herbal store

18. Ocimum basilicum I MOSEL 276 Herbal industry

19. Ocimum basilicum II MOSEL 277 Herbal store

20. Ocimum x africanum MOSEL 278 Herbal industry 21. Ocimum tenuiflorum MOSEL 279 Herbal industry

22. Otostegia limbata MOSEL 280 Herbal industry

23. Origanum vulgare * Herbal store

24. Plectranthus rugosus MOSEL 286 Herbal industry

25. Rosmarinus officinalis MOSEL 281 Herbal industry

26. Salvia aegyptiaca MOSEL 282 Herbal store

27. Salvia haematodes * Herbal industry

28. Salvia lanata MOSEL 287 Herbal industry

29. Salvia moorcroftiana I MOSEL 288 Herbal industry

30. Salvia moorcroftiana II * Herbal store

31. Salvia plebeia MOSEL 283 Herbal store

32. Stachys byzantina MOSEL 284 Herbal industry

* The samples which were found misidentified after the process of DNA barcoding were not allotted the Voucher IDs.

dried leaf tissue in a bleached pestle & mortar. To this grinded powdered material 1000 µl 2 X CTAB buffer was added & shifted to a 2 ml centrifuge tube with a locking cap. 100 mM Tris-HCl (pH 8. 0), 20 mM ethylene diamine tetra acetic acid (EDTA) with pH 8.0 and 1.4 M Sodium Chloride (NaCl) were used for preparing 2 X CTAB. 1 µl of 2-merceptoethanol was added to each tube in fume hood. The incubation of these tubes was carried out at 65 °C and 30 mins. 400 ul of chloroform isoamyl alcohol (24:1) was added to each tube and rocked the tubes for 30 minutes on an electrical rocker. The centrifugation of these tubes was carried out at 8000 revolutions per minute (rpm) for ten minutes after rocking. The aqueous phase was poured to another centrifuge tube in fume hood. 2/3 volume of chilled 2-propanol was added and allowed to precipitate for up to two weeks in the freezer. Once the precipitation was achieved, a 5 mins centrifugation at 3000 rpm resulted in the formation of DNA pellet towards the bottom of tubes. DNA pellets were washed with 70 percent ethanol (500 µl) after removing the supernatant. These tubes were again centrifuged at 4000 rpm for 3 minutes. Poured off supernatant and allowed the tubes to dry. 100 µl TE (Tris ethylenediamine tetra acetic acid) buffer was used to resuspend the DNA pellet. Dirty pellets were cleaned with MO-BIO’s UltraClean® 15 DNA purification kit by adopting the manufacturer’s directions. To assess the quality of extracted DNA, 5 µl DNA mixed with loading dye (5 µl) was loaded on 1% agarose gel.

2.1.2.2 DNA isolation kit

Few genera of Lamiaceae (Ocimum, Lycopus, Nepeta, Origanum) proved to be difficult while DNA isolation with standard 2 X cetyltrimethylammonium bromide (CTAB) method. Therefore, PowerPlant Pro DNA extraction kit by MO BIO Labs Inc. was utilized by following the manufacturer’s instructions (Appendix I).

2.1.3 PCR (Polymerase Chain Reaction)

For DNA barcoding of Lamiaceae HMPs, two cpDNA gene regions (rbcL and matK) and an intergenic spacer region (trnH-psbA) was selected. PCR protocols were optimized by trying different temperatures and cycling conditions for the amplification of target regions. Primer sequences of rbcL and matK gene region and trnH-psbA intergenic spacer region are given in Table 2.2.

PCR amplification of the rbcL, matK and trnH-psbA regions was performed in a Veriti Thermal Cycler (Applied Biosystems, Carlsbad, California, USA) using the KAPA3G Plant PCR Kit (Kapa Biosystems, Woburn, Massachusetts, USA) (Schori et al., 2013; Shinwari et al., 2014). Each reaction contained KAPA3G Plant PCR Buffer whwrein 1 X final concentration included dNTP’s at 0.2 mM each, MgCl2 (1.5 mM final concentration), one unit KAPA3G Plant DNA polymerase enzyme, forward & reverse primers having a final concentration of 0.3 μM each, template DNA and PCR-grade water to make the final volume of reaction 50 μL. For successful PCR of nine samples which failed to amplify at 1.5 mM

MgCl2, A higher conc. of MgCl2 was needed. For most of the species successful amplification of our selected markers was obtained by employing the KAPA3G polymerase from dirty DNA pellets which were not purified after CTAB isolation. However other species (Ocimum, Origanum, Nepeta, Lycopus) didn’t amplified unless the genomic DNA was extracted or purified by using PowerPlant Pro DNA isolation kit.

2.1.3.1 PCR conditions for rbcL complete gene amplification

The following cycling parameters were used for rbcL : 95 °C 10 min; 50 cycles: 95 °C 20 s, 58 °C 15 s, 72 °C 90 s and final extension 72 °C 90 s (Schori et al., 2013; Shinwari et al., 2014). The rbcL primers 1F (Fay et al., 1997) and 1460R (Fay et al., 1998; Cuénoud et al., 2002) were used in this experiment.

2.1.3.2 PCR conditions for matK partial gene amplification

The cycling parameters for matK were: 95 °C 10 min; 40 cycles: 95 °C 20 s, 50 °C 15 s, 72 °C 90 s and final extension 72 °C 90 s (Schori et al., 2013; Shinwari et al., 2014). The matK 390 F and 1360 R primers (Cuénoud et al., 2002) were used to amplify matK region.

2.1.3.2 PCR conditions for trnH-psbA spacer amplification

A touchdown program was carried out for trnH-psbA using the PsbAF/PsbHR primers (Sang et al., 1997; Tate & Simpson, 2003), where the annealing temp was 58 °C for initial 11 cycles followed by touchdown to 48 °C for 29 cycles (Schori et al., 2013; Shinwari et al., 2014). The temperatures for other steps remained same as in rbcL and matK.

PCR products were kept at 4 °C till further processing. To ensure the successful amplification of the desired sequences, the PCR products were run on 1% agarose gel.

2.1.4 Purification of amplified products

PCR amplified products were purified with the Wizard SV Gel and PCR Clean-Up System kit (Promega Corporation, Madison, Wisconsin, USA) following the manufacturer’s protocol (Appendix II).

2.1.5 DNA sequencing of barcoding loci

The purified PCR products were sequenced at Ohio University’s Genomics Facility and analyzed using an ABI 3130xl Genetic Analyzer (Applied Biosystems, Carlsbad, California, US). Each sequencing reaction included 2 μL 5× buffer (Applied Biosystems), 0.5 μL dimethyl sulfoxide (DMSO ; Sigma), 0.5 μL BigDye (Applied Biosystems), 0.1 μL ThermoFidelase (Fidelity Systems, Gaithersburg, Maryland, USA), 10–40 ng template DNA, and PCR-grade water for a total volume of 8 μL. Cycle sequencing products were cleaned with the BigDye XTerminator Purification Kit (Applied Biosystems) according to manufacturer’s protocol (Appendix III). In case of rbcL, external primers 1F and 1460R and internal primers 636F and 724R (Fay et al., 1997) were used for sequencing. For matK and trnH-psbA sequencing, the similar primer pairs were used as for amplification.

2.1.6 Nucleotide sequence data analysis

2.1.6.1 BLAST analysis

The electropherograms were assembled and edited in Geneious 6.1.8 (Kearse et al., 2012). Sequences were then compared to GenBank nucleotide database using NCBI’s MEGABLAST default parameters. Percent similarity was recorded for the closest matches. The highest BLAST percent identity of the query sequence should be from the expected species or the species included in the expected genera for right identification. Ambiguous identification means that the highest BLAST percent identity for a query happened to match with several genera of the expected plant family. Incorrect identification means that the BLAST percent identity o the query was none from the expected species, expected genera and expected plant family. The data (sequences) produced in this current study were deposited in GenBank.

Table 2.2 Primers used in DNA barcoding of Lamiaceae HMPs.

Primer Primer Primer sequence (5’-3’) Reference region name

rbcL 1F ATGTCACCACAAACAGAAAC Fay et al. 1997; Fay et al. 1998; 636F GCGTTGGAGAGATCGTTTCT Cuénoud et al. 724R TCGCATGTACCTGCAGTAGCATTCAAGT 2002

1460R TCCTTTTAGTAAAAGATTGGGCCGAG

matK 390F CGATCTATTCATTCAATATTTC Cuénoud et al. 2002 1326R TCTAGCACACGAAAGTCGAAGT

trnH-psbA psbAF GTTATGCATGAACGTAATGCTC Sang et al. 1997; Tate & trnHR CGCGCATGGTGGATTCACAAATC Simpson, 2003

2.2 Section II: Phylogenetic utility of cpDNA rbcL, matK and trnH-psbA regions

2.2.1 Sequence alignment and super matrix assembly

As a starting point of our alignment all available nucleotide sequences of rbcL, matK and trnH-psbA regions from GenBank (https://www.ncbi.nlm.nih.gov/) were downloaded. After building the initial alignments the shorter sequences were removed as much as possible and retained the longest sequences. Ultimately only one sequence per species was included (few exceptions). All GenBank sequences were checked for accuracy by making phylogenetic trees for each region and those sequences which appeared to be mislabeled/misidentified were excluded from analysis. Sequences were initially aligned using the L-INS-i algorithm in MAFFT (Katoh et al., 2002), and subsequently manually refined in Mesquite version 3.03 (Maddison and Maddison, 2011).

2.2.2 Data set construction

Three primary data sets were constructed by employing combination of taxa/data from GenBank and our nucleotide sequence data:

(1) rbcL sequence alignment composed of 245 ingroup taxa, one outgroup taxa and 1537 characters; (2) matK sequence alignment containing 235 ingroup taxa, one outgroup taxa and 1591 characters and (3) trnH-psbA sequence alignment containing 259 ingroup taxa, one outgroup taxa and 1134 characters.

2.2.3 Phylogenetic analysis by RAxML method

The phylogenetic analysis was performed on rbcL, matK, and trnH-psbA data sets separately using RAxML Blackbox webserver by method of ML (Maximum Likelihood) (Stamatakis, 2008). The Maximum Likelihood method was based on the default settings of the GTRGAMMA nucleotide substitution model. 2.3 Section III: Molecular phylogenetics of Lamiaceae based on plastid (trnL-trnF) and nuclear (ITS) markers

2.3.1 Taxon sampling from wild and herbarium

In this part of studies the collection of species was done from wild (Fig 2.4) and two national herbaria of Pakistan representing the five subfamilies of Lamiaceae which have occurrence in Pakistan. The taxa we sampled from wild and these herbaria represented the collections from tropical and subtropical regions of Pakistan. The total number of sampled taxa was 89 which included 20 taxa from previous section of studies (DNA barcoding), 21 taxa were collected from different regions of Pakistan (wild), 29 taxa were collected from Herbarium of Pakistan (ISL), QAU, Islamabad, Pakistan, and 19 taxa were provided by Karachi University Herbarium (KUH), Pakistan. The voucher specimens of samples collected from wild and those used in the DNA barcoding analysis were stored in MOSAEL Laboratory, Deptt. of Biotech, QAU Isb, Pakistan. Taxa name & voucher details are given in Table 2.3.

2.3.2 DNA extraction from wild and herbarium specimen’s

Lamiaceae is known for presence of secondary metabolites in majority of its members and it sometimes hamper the extraction of good quality DNA, particularly for very old degraded herbarium specimen. Therefore, we used modified DNA isolation protocols for different type of specimen’s:

2.3.2.1 2 X CTAB protocol

DNA was isolated from fresh plant material or silica-dried material according to the Doyle, (1991) 2 X CTAB protocol which has already been described in the section I of the methodology under 2.1.2.1.

Fig 2.4: Field visit to Northern regions of Pakistan for collecting Lamiaceae species.

Table 2.3: List of species collected from wild (MOSEL NZ), Herbarium of Pakistan (ISL) QAU and Karachi University Herbarium (KUH). The species (MOSEL) identified in the previous section of study (DNA barcoding of HMPs) are also included. All these collections are from Pakistan.

Serial Species Voucher ID Location of # collection

Subfamily Ajugoideae

1 Ajuga bracteosa Wall. ex Benth. MOSEL 265 Islamabad

2 Ajuga parviflora Benth. MOSEL NZ 01 Kumrat

3 Ajuga reptans L. MOSEL NZ 02 Kumrat

Subfamily Lamioideae

4 Anisomeles indica (L.) Kuntze KUH89 Mirpur

5 Colebrookea oppositifolia Sm. ISL 31 Rawalpindi

6 Craniotome furcata (Link) ISL 527 Swat Kuntze

7 Eremostachys speciosa Rupr. KUH3765 (38658) Hangu

8 Eremostachys spectabilis Popov ISL 637 Quetta

9 Eremostachys thyrsiflora Benth. KUH1518 Quetta

10 Lagochilus cabulicus Benth. KUH3255 (70396) Chitral

11 Lagochilus cuneatus Benth. KUH S. N. Landikotal (30044)

12 Lamium album L. MOSEL NZ 14 Swat

13 Lamium amplexicaule L. ISL 143 Rawalpindi

14 Leonurus cardiaca L. ISL 658 Gilgit

15 Leucas cephalotes (Roth) MOSEL 266 Islamabad Spreng. 16 Leucas mollissima Wall. ex ISL 777 Islamabad Benth.

17 Leucas lanata Benth. ISL 156 Rawalpindi

18 Leucas urticifolia (Vahl) R. Br. ISL 78 Thatta ex Sm.

19 Lycopus europaeus L. KUH115 Peshawar

20 Marrubium vulgare L. ISL 158 Sabzazar (Baluchistan)

21 Moluccella laevis L. KUH3997 (22715) Karachi

22 Otostegia aucheri Boiss. KUH4943 (30236) Khuzdar

23 Otostegia limbata (Benth.) MOSEL 280 Islamabad Boiss.

24 Stachys byzantina K. Koch. MOSEL 284 Islamabad

25 Stachys emodi Hedge KUH701 (77324) Astore

26 Stachys floccosa Benth. KUH1104 (55611) Parachinar

27 Stachys parviflora Benth. KUH6854 (55636) Swat

28 Stachys purpurea Poir. MOSEL NZ 15 Swat

29 Stachys palustris L. ISL 22 Sanjwani

30 Stachys sylvatica L. ISL 25 Swat

31 Thuspeinanta brahuica (Boiss.) KUH1965 (15633) Kalat Briq.

Subfamily Nepetoideae

32 Calamintha debilis (Bunge) MOSEL NZ 12 Swat Benth.

33 Calamintha hydaspidis (Falc. ex KUH S. N. Azad Jammu Benth.) Hedge (21749) Kashmir

34 Clinopodium umbrosum (M. MOSEL NZ 13 Kumrat Bieb.) Kuntze

35 Clinopodium vulgare L. ISL 1052 Muzaffarabad

36 Dracocephalum moldavica L. ISL 2038 Skardu

37 Elsholtzia ciliata (Thunb.) Hyl. ISL 981 Azad Jammu Kashmir

38 Elsholtzia densa Benth. ISL 135 Skardu

39 Gontscharovia popovii (B. ISL Zardullah Chitral Fedtsch. & Gontsch.) Boriss.

40 Glechoma hederacea L. ISL 4748 Islamabad

41 Hyptis suaveolens (L.) Poit. ISL 988 Azad Jammu Kashmir

42 Isodon rugosus (Wall. ex Benth.) KUH4055 (73037) Hunza Codd

43 Isodon coesta (Buch.-Ham. ex KUH3434 Hazara D. Don) Kudô

44 Isodon lophanthoides(Buch.- KUH7884 Rawalpindi Ham. ex D. Don) H. Hara

45 Lallemantia royleana (Benth.) ISL 828 Nushki Benth.

46 Lavandula angustifolia Mill. KUH3898 Quetta

47 Melissa officinalis L. MOSEL 267 Islamabad

48 Mentha arvensis L. MOSEL 268 Islamabad

49 Mentha longifolia (L.) L. MOSEL 270 Islamabad

50 Mentha longifolia ssp. (L.) MOSEL NZ 20 Kumrat Huds.

51 Mentha pulegium L. MOSEL 272 Islamabad

52 Mentha suaveolens L. MOSEL 273 Islamabad 53 Mentha spicata L. MOSEL 274 Islamabad

54 Mentha aquatica L. MOSEL 269 Islamabad

55 Mentha x piperita L. MOSEL 271 Islamabad

56 Micromeria biflora (Buch.-Ham. MOSEL NZ 11 Swat ex D. Don) Benth.

57 Monarda fistulosa L. MOSEL NZ 16 Islamabad

58 Nepeta cataria L. MOSEL 275 Islamabad

59 O. americanum L. MOSEL NZ 10 Swat

60 O. basilicum L. MOSEL 276 Islamabad

61 O. tenuiflorum L. MOSEL 279 Islamabad

62 O. x africanum MOSEL 278 Islamabad

63 Origanum majorana L. MOSEL NZ 17 Swat

64 Origanum vulgare L. MOSEL NZ 23 Hattar

65 Perovskia abrotanoides Kar. KUH458 (31824) Quetta

66 Perovskia atriplicifolia Benth. ISL 05 Chitral

67 Perilla frutescens (L.) Britton ISL 1072 Muzaffarabad

68 Plectranthus barbatus Andrews ISL 966 Muzaffarabad

69 Prunella vulgaris L. MOSEL NZ 03 Kumrat

70 Rosmarinus officinalis L. MOSEL 281 Islamabad

71 S. aegyptiaca L. MOSEL 282 Islamabad

72 S. alba J.R.I.Wood MOSEL NZ 08 Swat

73 S. cabulica Benth. MOSEL NZ 05 Kumrat

74 S. lanata Roxb. MOSEL NZ 09 Swat

75 S. nubicola Wall. ex Sweet MOSEL NZ 21 Swat

76 S. plebeia R. Br. MOSEL 283 Islamabad 77 Salvia asperata Falc. ex Benth. MOSEL NZ 07 Swat in A.P.de Candolle

78 Salvia cordata Benth. MOSEL NZ 04 Kumrat

79 Salvia elegans Vahl MOSEL NZ 22

80 Thymus serpyllum L. ISL 54 Gilgit

81 Thymus sp. ISL 1524 Chitral

82 Thymus vulgaris L. ISL 920 Kurram Valley

83 Zataria multiflora Boiss. KUH2998 (55987) Booni

84 Ziziphora clinopodioides Lam. ISL 221 Quetta

85 Ziziphora tenuior L. ISL 59 Sibi

Subfamily Scutellarioideae

86 Scutellaria edelbergii Rech. f. ISL A.R. Baig Chitral

87 Scutellaria linearis Benth. ISL 951 Poonch

88 Scutellaria sp. MOSEL NZ 18 Kumrat

Subfamily Viticoideae

89 Vitex negundo L. MOSEL NZ 19 Islamabad

2.3.2.2 DNA extraction from herbarium samples

To extract quality DNA from old specimens of herbarium, the protocol was adapted from Doyle and Doyle (1987) and Cullings (1992). CTAB buffer was prepared and stored in a capped container. Before starting DNA extraction, 5 ml of CTAB was taken out and 0.2 g Polyvinylpyrrolidone (PVP) (Fisher Cat#: BP431-500) plus 25 μl of b-mercaptoethanol (Fisher Cat#: BP176 100) were added in it. The solution was stirred to dissolve PVP. Weighed 10-20 mg of plant tissue and put it in the tube. The samples from herbaria were grinded with the help of metal beads added in the tubes and beated in a beater GenoGrinder® (SPEX CertiPrep, Inc.) for two minutes or until the tissue was reduced to a fine powder. 500 μl of CTAB buffer was added to grind the samples further. Incubated the samples at 55 °C for 30 minutes to overnight. 500 μl of 24:1 Chloroform: Isoamyl alcohol was added and mixed well by shaking the tubes. These tubes were centrifuged for 5-10 minutes at maximum speed. Following centrifugation, three layers appeared (i) top: aqueous phase (ii) middle: debris and proteins (iii) bottom: chloroform. To avoid mixing of these layers the next step was performed quickly. The aqueous phase was pipetted out carefully and transferred to a separate tube. The volume of the aqueous phase was estimate and 0.08 volumes of cold 7.5 M ammonium acetate was added in it. Using the combined volume of aqueous phase and added ammonium acetate, 0.54 volumes of cold isopropanol (=2-propanol) was added. The mixture was vortexed and allowed to sit in freezer for 15 minutes to overnight. Longer times yielded more DNA, but also more contaminants. The resulting mixture was centrifuged for 3 minutes at maximum speed. The liquid was poured off carefully to retain the pellet in the tube. 700 μl of cold 70% Ethanol was added and mixed. Again the centrifugation was performed for 1 min at maximum speed. Poured off the liquid with care and added 700 μl of cold 95% Ethanol and mixed. Repeated the previous centrifugation step. The liquid was removed and pellet was retained. Dried the pellet by inverting the tubes on a Kim-wipe and allowed to stand for 1 hr or until dried. 100 μl of TE buffer was added in the tubes. The tubes were kept for 1hr at 55 °C or overnight in refrigerator. To assess the quality of isolated DNA, 5 µl DNA with 5 µl loading dye was loaded on 1% agarose gel.

2.3.3 PCR

PCR amplifications for trnL-trnF and ITS regions were performed using primers (Table 2.4) and procedures described in Dong et al. (2013). The 50 µl amplification reactions contained 1 µl DNA solution (10 ng), 10 µl PCR reaction buffer (5 X), 4 µl dNPT mix (0.2 mM), 5 µl of each primer , and 2.5 U of Apex TaqDNA polymerase (Gene Choice - Life Sciences Company, USA). For stubborn samples KAPA3G plant pcr kit manufactured by Kapa Biosystems Woburn, Massachusetts, USA was employed as outlined in Shinwari et al. (2014). Each reaction contained the KAPA3G Plant PCR Buffer (1 X final concentration, includes dNTPs at 0.2 mM each), MgCl2 (1.5 mM final concentration), 1 unit KAPA3G Plant DNA polymerase, primers at a final concentration of 0.3 μM each, template DNA and PCR- grade water to bring the volume to 50 μL.

2.3.3.1 PCR parameters for trnL-trnF region

The following cycling parameters were used for trnL-trnF: 95 °C 10 min; 40 cycles: 95 °C 20 s, 55 °C 15 s, 72 °C 90 s and final extension 72 °C 7 mins. There were two primer pairs used for the amplification of trnL-trnF region including trnL exon, trnL intron, trnL-F spacer and trnF exon (Table 2.4 and Fig 2.5)

2.3.3.2 PCR parameters for ITS region

The following cycling parameters were performed for ITS: 95 °C 10 min; 40 cycles: 95 °C 20 s, 53 °C 15 s, 72 °C 90 s and final extension 72 °C 7 mins. The entire ITS region was amplified by using two primer pairs as described by Blattner (1999) (Table 2.4 and Fig 2.6).

2.3.4 Purification of amplified products and DNA sequencing

Purification and sequencing reactions were performed at the University of Florida Genetic Institute where they use dideoxy chain termination method and reactions were run on an ABI

3730 DNA Analyzer (Applied Biosystems, Inc.).

Table 2.4 Primers used for amplification of plastid trnL-trnF and nuclear ITS regions.

Primer Primer name Primer sequence (5’-3’) Reference region

trnL-trnF c (Forward) CGAAATCGGTAGACGCTACG Taberlet et al. (1991) d (Reverse) GGGGATAGAGGGACTTGAAC

e (Forward) GGTTCAAGTCCCTCTATCCC

f (Reverse) ATTTGAACTGGTGACACGAG

ITS ITS1 GGAAGGAGAAGTCGTAACAAGG Blattner , (Forward) 1999 GCAATTCACACCAAGTATCGC ITS2 (Reverse) CTCTCGGCAACGGATATCTCG ITS3 (Forward) CTTTTTCCTCCGCTTATTGATATG

ITS4 (Reverse)

Fig 2.5: Showing locations of primers used to amplify different regions of plastid (trnL-trnF) genome.

Fig 2.6 Showing locations of primers used to amplify different regions of nuclear (ITS) genome.

2.3.5 Nucleotide sequence data analysis

2.3.5.1 Sequence alignment and super matrix assembly

Sequences were assembled and edited manually in Geneious 6.1.8 (Kearse et al., 2012). These sequences are in the process of submission to GenBank. As a starting point of our alignment we downloaded all available nucleotide sequences of trnL-trnF and ITS regions from GenBank (https://www.ncbi.nlm.nih.gov/). We then pruned or merged taxa in GenBank with synonymous names. After building the initial alignments we removed the shorter sequences and retained the longest sequences. Ultimately, we included only one sequence per species. All GenBank sequences were checked for accuracy by making phylogenetic trees for each individual gene region and those sequences which appeared to be mislabeled/misidentified were excluded from analysis. Sequences were initially aligned using the L-INS-i algorithm in MAFFT (Katoh et al., 2002), and subsequently manually refined in Mesquite version 3.03 (Maddison and Maddison, 2011). Ambiguously aligned characters were excluded prior to analyses.

2.3.5.2 Data set construction

We constructed two primary data sets by employing combinations of taxa/data from GenBank and our nucleotide sequence data:

(1) trnL-trnF sequence alignment composed of 398 ingroup taxa and 1129 nucleotides

(2) ITS sequence alignment containing 413 ingroup taxa and 1011 nucleotides.

2.3.6 Phylogenetic analysis

Phylogenetic tree reconstruction of the above prepared data sets of trnL-trnF and ITS sequences was performed by two methods described below: 2.3.6.1 RAxML

The phylogenetic analyses were performed on the trnL-trnF and ITS data sets using Maximum Likelihood (ML) as implemented in RAxML Blackbox webserver (Stamatakis, 2008). The ML analyses employed the GTRCAT nucleotide substitution model, with the default settings for the optimization of individual per-site substitution rates.

2.3.6.2 Bayesian Inference (BI)

Due to the difficulties of bootstrapping data sets with large amounts of non-randomly distributed missing data, we used Bayesian Inference (BI) to assess phylogenetic relationships and nodal posterior probabilities (PP). Bayesian analyses were performed using MrBayes 3.2.2 (Huelsenbeck and Ronquist, 2001) through XSEDE on the CIPRES Portal (Miller et al., 2010). The default settings (GTR + G + I model) were used for analyses of the trnL-trnF and for ITS HKY+G model was selected. The analysis was run with four Marko Chain Monte Carlo (MCMC) runs for ten million generations each. Trees were sampled every 500 generations. Convergence and mixing were assessed using Tracer 1.4 (Rambaut and Drummond, 2007). To evaluate convergence, it was observed that the standard deviation of split frequencies fell below 0.01. The burn-in value was 25% for obtaining a 50% majority rule consensus tree. Clade support was determined by Bayesian posterior probabilities (PP; Rannala and Yang, 1996). The 50% majority rule consensus trees were viewed with the program FigTree 1.4.0 (Rambaut, 2008).

Results

3.1 Section I: DNA Barcoding of herbal medicinal products (HMPs) of Lamiaceae from Pakistan.

3.1.1 DNA extraction from HMPs and PCR of rbcL, matK and trnH-psbA regions

Total genomic DNA was extracted from HMPs by using CTAB protocol and a DNA isolation kit. Isolated DNA was run on 1 % agarose gel for determining the quality of DNA (Figure 3.1 A), as good quality template is desired for the amplification of barcoding regions. For most of the species successful amplification of our selected markers (rbcL, matK and trnH-psbA) was achieved with the KAPA3G enzyme (Kapa Biosystems) from dirty pellets (not purified after CTAB extraction), while other species (Ocimum, Lycopus, Nepeta, Origanum) failed to amplify until genomic DNA had been purified or extracted with the PowerPlant Pro DNA Isolation Kit (MO BIO Laboratories, Inc.). The amplified products of different sizes from different barcoding regions were confirmed on 1 % agarose gel (Figure 3.1 B-D).

3.1.2 Purification, sequencing and data analysis

PCR products were cleaned with the Wizard SV Gel and PCR Clean-Up System kit (Promega Corporation, Madison, Wisconsin, USA). The purified PCR products were sequenced at Ohio University’s Genomics Facility and analyzed using an ABI 3130xl Genetic Analyzer. Sequences of amplified fragments were edited and analysed using Geneious 6.1.8 software (Kearse et al., 2012). Sequences were then compared to GenBank nucleotide database using NCBI’s MEGABLAST default parameters. Percent similarity was recorded for the closest matches.

3.1.3 BLAST analysis

Here a detailed account of the findings based on the BLAST analysis is given:

A M

~1500 bp

B

M

~850 bp

C

M

~500 bp

D

Fig 3.1: Visualizations of extracted genomic DNA and amplified products of DNA barcoding loci. A: Isolated genomic DNA of different taxa of Lamiaceae. B: PCR product of rbcL gene region. C: Amlified product of matK gene region. D: PCR product of trnH-psbA intergenic spacer region. M: 100 bp DNA marker (Fermentas).

In our investigation, rbcL proved to be the most successfully amplified and sequenced in 28 of the 32 HMPs (87%). matK provided 26 barcode sequences while trnH-psbA yielded 22 sequences. The barcode recovered for each HMP and plant part from which it was recovered can be found in Table 3.1. Recovered rbcL barcodes ranged from 576-1445 bp. Barcoding success for matK and trnH-psbA were 26/32 (81%) and 22/32 (69%), respectively (Fig 3.2). matK sequences ranged from 713–826 bp, while trnH-psbA sequences ranged from 506–586 bp. All nucleotide sequences from the current study are deposited and available in the GenBank under the accession numbers KP172036-KP172082, KP218929-KP218945. These accessions do not include misidentified and substituted/contaminated samples.

Twenty-two (22) of 32 sequenced samples matched their expected genera based on MEGABLAST of the selected barcode regions. One industry sample (Salvia haematodes) and three samples Leucas linifolia, Lycopus europaeus and Salvia moorcroftiana II bought from the local stores came up as completely different species not belonging to Lamiaceae (Table 3.1). Sequence chromatograms of these samples showed clear peaks with no indication of admixture. With respect to their obtained barcodes, all, except Lycopus europaeus, showed 99% sequence identity to a different unrelated species in MEGABLAST. Lycopus europaeus showed Proboscidea and Martynia with 95% sequence identity as top- hits.

3.1.3.1 rbcL gene analysis

Excluding the four misidentified samples and three other samples for which rbcL could not be amplified/sequenced, 21 of 25 rbcL sequences (84%) matched the expected genera based on MEGABLAST. Of these 21 only four rbcL sequences matched the expected species (Melissa officinalis, Nepeta cataria, Rosmarinus officinalis, Stachys byzantina). Nine rbcL sequences (Ajuga bracteosa, Ajuga parviflora, Plectranthus rugosus, Leucas cephalotes, Nepeta ruderalis, Otostegia limbata, Salvia aegyptiaca, Salvia lanata, Salvia moorcroftiana I) could not be identified to species level due to limited reference sequence data available in GenBank. rbcL lacked sequence variability for resulting in multiple congeneric species hits having the similar % identity. For example, top hits for the rbcL sequence for Mentha spicata sample did not include the expected species, but included Mentha suaveolens. matK for Mentha spicata came up with multiple congeneric hits including Mentha spicata, Mentha suaveolens and Mentha x piperita with 100% identity. On the other hand trnH-psbA resulted in 99% identity with Mentha spicata but 97% identity with other congeneric species. The same was true for Mentha pulegium and Mentha aquatica rbcL, which showed Mentha suaveolens and Mentha x piperita as top hits respectively while their matK and trnH-psbA provided us with the expected closest matches. Based on our rbcL barcode, Mentha suaveolens, Mentha x piperita, Mentha longifolia, Mentha arvensis, Ocimum basilicum I, Ocimum basilicum II, Ocimum x africanum, Ocimum tenuiflorum and Salvia plebeia did not match their expected species (Table 3.1). We could only prove their identification to the expected genus level.

3.1.3.2 matK nucleotide sequence analysis

A total of 25 matK barcodes excluding the four misidentified and three non-amplified/non- sequenced samples were obtained. Of these 25 matK sequences, 19 matched the expected genera based on MEGABLAST. 10 of these 19 matK sequences matched their expected species level (Melissa officinalis, Mentha aquatica, Mentha pulegium, Mentha spicata, Nepeta cataria, Ocimum basilicum I, Ocimum x africanum, Rosmarinus officinalis, Salvia plebeia, Stachys byzantina). No reference data could be retrieved for Ajuga bracteosa, Hyssopus officinalis, Lallemantia royleana, Leucas cephalotes, Nepeta ruderalis, Otostegia limbata, Salvia aegyptiaca and Salvia moorcroftiana from GenBank, therefore, species level identification was not possible based on matK barcodes (Table 3.1).

3.1.3.3 trnH-psbA spacer region analysis trnH-psbA spacer region had a lower rate of amplification/sequencing success as compared to rbcL and matK, however, it proved to be more successful for distinguishing the samples at species level. Only 20 trnH-psbA barcodes were found excluding four mislabeled and eight unamplified), resulting in 14 sequences which provided identification to expected genera and eight sequences (Mentha pulegium, Mentha spicata, Mentha longifolia, Mentha aquatica, Ocimum basilicum I, Ocimum basilicum II, Rosmarinus officinalis and Salvia plebeia) came up as successful barcodes for expected species (Table 3.1). Ajuga bracteosa and Leucas cephalotes could be identified to the expected genera due to lack of their reference trnH-psbA sequences in GenBank. There are three HMPs from our collection which appeared to be highly ambiguous. There was not sufficient reference sequence data available in GenBank for these species with respect to our chosen DNA barcoding markers. Although the barcodes obtained for these species (Lallemantia royleana, Origanum vulgare and Salvia aegyptiaca) were not completely unrelated because they belonged to Lamiaceae and none of the sequences for either of the barcoding region showed up to any other plant family. Genus Lallemantia had no records for rbcL, matK and trnH-psbA. Otostegia had no reference sequence for trnH- psbA, four un-verified matK sequences and three rbcL sequences which did not matched to our query blast. matK and trnH-psbA sequences of Origanum vulgare did not correspond to reference sequences in top-hits, rather in case of matK it appeared to be 99-100% identical to several Mentha species but also 99% identity to Origanum vulgare in lower-hits. trnH-psbA behaved in the same fashion where top-hit was Thymbra while lower hits contained Origanum vulgare, both with 97% identity. The only reference data available for Salvia aegyptiaca was a trnH-psbA sequence which did not show up in our blast hits. In case of Hyssopus only two rbcL reference sequences were available for Hyssopus which did not came up as top-hits for our query sequence. No reference sequences were present in GenBank for matK and trnH-psbA for genus Hyssopus. Similarly, Nepeta ruderalis had no sequence data in GenBank. However, we have marked Hyssopus officinalis and Nepeta ruderalis as product substitutes based on the sequences we obtained. Our rbcL, matK and trnH-psbA sequences for Hyssopus officinalis and Nepeta ruderalis showed them as Nepeta bracteata and Salvia respectively. As all the three markers showed same identification, it indicated the product substitution (Table 3.1).

The overall amplification success for rbcL and matK was 87% and 81% while trnH-psbA showed 69% (Fig 3.2). On the contrary, matK and trnH-psbA were able to distinguish the species relatively better with 40% success rate than rbcL (16%) (Fig 3.3). On the whole we generated a total of 22 genus-level barcodes (78%) and 12 species-level barcodes (44%) (Fig 3.3).

Table 3.1: List of the HMPs used including scientific names based on the common names under which the products were sold, plant part used, and taxonomic ID based on MEGABLAST top hits (best max score). Empty cells suggest that barcode was not recovered due to DNA extraction and/or amplification problems. “spp.” following genus name indicate multiple species possible within that genus. † represents misidentified HMPs. * indicates HMPs potentially substituted/contaminated.

Common Expected Plant Part Identification based on barcodes name scientific name rbcL ID matK ID trnH-psbA ID

Kauri booti Ajuga bracteosa Leaf Ajuga reptans Ajuga orientalis Ajuga ciliate

Tarkha booti Ajuga parviflora Leaf Ajuga - - decumbens

Zufa Khushk Hyssopus Flower *Nepeta *Nepeta *Nepeta officinalis bracteata bracteata bracteata

Tukhm Lallemantia Leaf - Glechoma - malanga royleana hederaceae

Chatra, Gul Leucas Leaf Marrubium Leucas sp. Leucas sp. dode cephalotes peregrinum

Gomi Leucas linifolia Shoot, leaf, †Portulaca - †Portulaca flower oleraceae oleraceae

Gypsywort Lycopus Shoot, leaf, - - †Proboscidea europaeus flower triloba

Lemon balm Melissa Leaf Melissa Melissa - officinalis officinalis officinalis

Watermint Mentha aquatica Leaf Mentha x Mentha Mentha aquatia piperita aquatica

Wild mint Mentha arvensis Leaf Mentha Mentha spicata Mentha suaveolens Canadensis

Peppermint Mentha x Leaf Mentha Mentha Mentha piperita suaveolens aquatica Canadensis Pennyroyal Mentha Leaf Mentha Mentha Mentha pulegium suaveolens pulegium pulegium

Velanay,Hors Mentha Leaf Mentha Mentha Mentha emint longifolia suaveolens aquatica longifolia

Spearmint Mentha spicata Leaf Mentha Mentha spicata Mentha spicata suaveolens

Apple mint Mentha Leaf Mentha spicata Mentha spicata Mentha suaveolens Canadensis

Catnip Nepeta cataria Leaf Nepeta cataria Nepeta cataria -

Badrang Nepeta ruderalis Flower *Salvia spp. *Salvia plebeia *Salvia spp. Boya, Indian catnip

Nyazbo, Basil Ocimum Leaf Ocimum Ocimum Ocimum basilicum I tenuiflorum basilicum basilicum

Nyazbo, Basil Ocimum Seed Ocimum Ocimum Ocimum basilicum II tenuiflorum americanum basilicum

Lemon basil Ocimum x Leaf Ocimum Ocimum x Ocimum africanum tenuiflorum africanum basilicum

Tulsi Ocimum Leaf Ocimum Ocimum - tenuiflorum basilicum basilicum

Chiti Booti Otostegia Leaf Stachys, Leucas Lamium limbata Phlomis, cephalotes galeobdolon Lamium, subsp. Ballota Flavidum

Banjawain, Origanum Shoot, leaf - Mentha spp. Thymbra, Wild vulgare Mentha marjoram

Khwangere Plectranthus Leaf Plectranthus - - rugosus barbatus

Rosemary Rosmarinus Leaf Rosmarinus Rosmarinus Rosmarinus officinalis officinalis officinalis officinalis Tak malanga Salvia Seed Dracocephalum Lallementia Rosmarinus aegyptiaca grandiflorum royleana officinalis

Behman Salvia Roots †Biebersteinia - - Surkh, Red haematodes multifida sage

Kianr Salvia lanata Leaf Savia - - moorcroftiana

Kalli-jarri Salvia Leaf Salvia Salvia - moorcroftiana I nemorosa nemorosa

Kalli-jarri Salvia Powder - †Baccharoides - moorcroftiana II adoensis

Samundar Salvia plebeia Leaf Salvia Salvia plebeia Salvia plebeia sokh miltiorrhiza

Lamb’s ears Stachys Leaf Stachys Stachys Prasium majus byzantina byzantina byzantina

100 90 81 80 76

70 60 rbcL 50 40 matK 40 Percentage% psbA-trnH 30 20 10 0 Barcoding success Genus-level ID Species-level ID

Fig 3.2: Performance/Percentage success of DNA barcoding loci for Lamiaceae. rbcL exhibited the highest barcoding and genus-level identification success, but with the lowest species-level identification. matK and trnH-psbA showed the same percentage of species- level identification. Our results proposed matK as the potential barcode for Lamiaceae HMPs.

90 80

70

60 50 40

30 Percentage% 20 Series1 10 0

Fig 3.3: DNA barcodes of 32 HMPs belonging to Lamiaceae, collected from local herbal stores and herbal pharmaceutical industry. 3.2 Section II: Evaluation of phylogenetic utility of cpDNA rbcL, matK and trnH-psbA regions

To estimate the phylogenetic utility of DNA barcoding regions (rbcL, matK and trnH-psbA ), the three sequence data sets are analysed independently by using Maximum Likelihood (ML) method as presented in RAxML analysis (Fig 3.4-3.6). The findings suggested matK as the best gene region which is able to resolve the subfamilies and provided the monophyletic or non-monophyletic genera with strong support. rbcL is not able to resolve the clades with a strong bootstrap (BS) support. It is difficult to align the spacer region trnH-psbA across the whole family (Fig 3.7). As a result it affected the tree topology and did not produce the well resolved clades of subfamilies.

Viticoideae included incertae sedis genus Teijsmanniodendron in both rbcL and matK phylogenetic trees, with weak support in rbcL but 100% BS is found in matK (Fig 3.4 & Fig 3.5). The BS of Viticoid genera in rbcL are weak overall and Petraeovitex is found in Scutellarioideae clade. Congea (Symphorematoideae) appeared to be part of Viticoideae in rbcL (Fig 3.4).

Ajugoideae made a single major clade in matK only while in rbcL and trnH-psbA it was split in two distant clades. Genus Westringia (Prostantheroideae) was found embedded between Ajuga and Teucrium in rbcL with BS 75 (Fig 3.4). Teucrium along with Spartothamnella showed a separate distant clade from rest of the Ajugoideae in trnH-psbA. The only other member of Prostantheroideae, genus Prostanthera was embedded between incertea sedis clade Cornutia and Gmelina with a BS value of 58. matK represented a single clade of tribe Westringieae belonging to subfamily Prostantheroideae, however, trnH-psbA represented tribe Westringieae and tribe Chloantheae not in a single clade (Fig 3.5 & Fig 3.6).

Scutellarioideae was found to be monophyletic except in rbcL where Petraeovitex (Viticoideae) occupied a basal position. The support values for rbcL and trnH-psbA clades were 68 and 50 respectively while matK received < 50 BS.

Tribe Gomphostemmateae and Pogostemoneae belonging to subfamily Lamioideae were not well resolved in matK (Fig 3.4). Tribe Leucadeae was divided into three subclades. The Lamioideae incertae sedis genera Betonica, Colquhounia, Galeopsis and Roylea remained unplaced. While in rbcL and trnH-psbA all the ten tribes of Lamioideae were unable to group accordingly (Fig 3.5 & Fig 3.6).

Subfamily Nepetoideae is the largest subfamily of Lamiaceae. rbcL and trnH-psbA failed to produce the tribal grouping of Nepetoideae. matK placed tribe Lavanduleae at the basal position of tribe Ocimeae. Elsholtzieae received a strong support of BS 98 (Fig 3.5). The largest tribe Mentheae was divided into several subclades.

The incertea sedis genera of Lamiaceae (proposed by Harley et al., 2004) Callicarpa, Cornutia, Gmelina, Premna and Tectona remained unplaced. Cymaria appeared with Lamioideae in rbcL and matK having very weak support. Teijsmanniodendron was found in Viticoideae clade with Vitex in rbcL and matK.

rbcL

matK

psbA

- trnH

Fig 3.7: Comparison of rbcL, matK and trnH-psbA region alignments. rbcL region showed least variability, while reasonable variation indicated by black arrows found in matK. However trnH-psbA spacer region was highly variable making it difficult to align across the family.

3.3 Section III: Molecular phylogenetics of Lamiaceae based on plastid (trnL-trnF) and nuclear (ITS) markers

3.3.1 DNA extraction and PCR

Total genomic DNA was extracted from wild and herbaria samples by using CTAB protocol. Isolated DNA was run on 1 % agarose gel for determining the quality of DNA (Fig 3.8 A). For most of the species successful amplification of our selected markers (trnL-trnF and ITS) was achieved with the Apex TaqDNA polymerase whereas with stubborn samples KAPA3G enzyme was used. It was observed that DNA from herbaria samples was technically more difficult to amplify. To amplify trnL-trnF and ITS two sets of forward and reverse primers were used for each region (details given in Chapter Materials and methods: Table 2.4). The size of amplified products was confirmed on 1 % agarose gel (Fig 3.8 B-C).

3.3.2 Purification, sequencing and data analysis

The purification and sequencing of PCR products were performed at the University of Florida Genetic Institute where they use dideoxy chain termination method and reactions were run on an ABI 3730 DNA Analyzer. Sequences of amplified fragments were edited and analysed. The data sets composed of GenBank sequences plus our generated sequences were analysed using Maximum Likelihood (ML) and Bayesian analysis for reconstructing molecular phylogeny of Lamiaceae. Results are elaborated here separately based on plastid and nuclear regions.

3.3.3 Phylogenetic analysis based on plastid trnL-trnF region

To estimate the phylogenetic relationships among different groups of Lamiaceae 1011 characters of 392 taxa were used in the ML (RAxML) and Bayesian analysis of the trnL-trnF region. In comparison to Maximum Likelihood (ML) consensus tree (Fig. 3.9) the Bayesian Inference (BI) tree (Figure 3.10) presents nodes which are better resolved. The subfamilies received strong bootstrap values in the Bayesian tree in contrast to the ML tree. Scutellarioideae, Ajugoideae, Prostantheroideae clades were well resolved and monophyletic having strong support (PP 1.0) in Bayesian tree.

A

M

~850 bp

bpbpbp B bbp

M

~750 bp

C

Fig 3.8: Visualizations of extracted genomic DNA and amplified products of plastid and nuclear regions. A: Total genomic DNA extracted from different taxa of Lamiaceae. B: PCR product of plastid trnL-trnF region. C: Amlified product of nuclear ITS region. M: 1 Kb DNA marker (Fermentas).

Viticoideae was not well resolved and represented by Vitex, Gmelina, Premna and Petraeovitex only. However, it was evident that Viticoideae was not monophyletic. Members of Gmelina and Premna formed the sister clade with Vitex having a significant posterior probability (PP) 0.97 obtained in Bayesian consensus tree (Fig 3.10), however, the ML provided a very weak bootstrap (BS) value of 60 (Fig 3.9). Petraeovitex turned out to be close to subfamily Lamioideae (BS = 83, PP = 0.98).

The Scutellarioideae clade was retrieved as monophyletic having BS value 100% and PP 1.0. Howevere, clade Scutellarioideae only comprised of genus Scutellaria, Holmskioldia and Wenchengia.

The clade representing the Ajugoideae included Ajuga, Teucrium, Teucridium, Oxera, Faradaya, Aegiphila, Tetraclea, Kalaharia, Caryopteris and Clerodendrum. Ajuga and Teucrium were monophyletic having strong support BS 100 while their PP values were 1.0 and 0.99 respectively. Teucrium and Teucridium were well supported sister genera (BS = 100, PP = 1.0). Oxera and Faradaya were recovered as clustered together in the similar clade with very strong support (BS = 100, PP = 1.0). Tetraclea was grouped together with Aegiphila making Aegiphila non-monophyletic in both trees, however the support value was significant in Bayesian tree (PP 0.92) and lower in ML tree. Clerodendrum appeared to be polyphyletic and clustered with Tetraclea-Aegiphila clade with weak support. Genus Rotheca did not appear in the Ajugoideae clade, it was recovered as a sister group to Scutellarioideae having strong support in Bayesian tree (PP 0.98) (Fig 3.10) and weak BS support in ML tree (Fig 3.9).

Clade Prostantheroideae appeared to be grouped close to Ajugoideae but the relationship was not evident. This clade was well resolved into two sub-clades representing tribe Chloantheae (BS = 99, PP = 0.99) and tribe Westringieae (BS = 100, PP = 1.0). Brachysola, Cyanostegia and Dicrastylis comprised the Chloantheae clade and all these three genera were monophyletic with considerable strong support in both analysis. Cyanostegia and Dicrastylis were sister groups having stong values of PP 1.0 and BS 100. Clade Westringieae included Prostanthera, Wrixonia, Westringia, Hemiandra, Microcorys and Hemigenia. Prostanthera was non-monophyletic due to embedded Wrixonia in it. Hemiandra was sister to monophyletic Westringia having strong BS 100 and PP 0.99. Microcorys and Hemigenia were gouped together and this Microcorys-Hemigenia group was strongly supported in both type of trees (BS = 100, PP = 1.0). Incertea sedis Callicarpa made a sister group with Prostantheroideae in ML tree.

Subfamily Nepetoideae constituted the largest clade among Lamiaceae subfamilies. Monophyly of the Nepetoideae clade was well supported in both trees (BS = 100, PP = 1.0). Tribe Elsholtzieae received strong support for its monophyly in Bayesian tree (PP 1.0) (Fig 3.10) while the support was significantly high in ML tree (BS 97) (Fig 3.9). The sister relationship between Elsholtzieae and Ocimeae received BS 62 and PP 0.67. Sub-tribal delimitation was not well resolved in Ocimeae. Clade Plectranthinae showed significant support PP 0.95 whereas the BS value was as low as 80. Inter-generic delimitation was also not clear, only Tetradenia was confirmed as monophyletic having BS 98 and PP 1.0 support values. Plectranthus was non-monophyletic and Coleus was distant from Plectranthus, Coleus strongly grouped with Anisochilus in Bayesian tree (PP 98), however, the support for this relationship was weak in ML tree (BS 69). Subtribe Ociminae was fairly monophyletic with lower support in both consensus trees (BS = 64, PP = 0.87). Within Ociminae, Syncolostemon and Hemizygia grouped together and this group was strongly supported in Bayesian tree (PP 1.0) and BS 95 in ML analysis. Monophyly of Ocimum and Orthosiphon was confirmed in both trees. Subtribe Hanceolinae and Hyptidinae were not delimited into distinct clades rather Hanceolinae was not presented as monophyletic. Subtribe Lavandulinae was not positioned clearly due to delimited Hanceolinae and Hyptidinae.

Tribe Mentheae was represented by a well supported clade in both trees (BS = 100, PP = 1.0) (Fig 3.9 & Fig 3.10). However, within Mentheae inter-generic boundaries of many genera were not delimited and bootstrap support was relatively lower in ML tree. The clade comprising subtribe Salviinae received a strong support in Bayesian consensus tree (PP 0.98) but the support was weak in ML tree (BS 66). Salvia was confirmed as non-monophyletic. Rosmarinus and Perovskia were monophyletic in both trees but Lepichinia made a sister group with other Salviinae and Melissa occupied a basal position to Salviinae plus Lepichinia. Subtribe Nepetinae received a low support of BS 56. In Bayesian tree it appeared to be divided into two clades and both were strongly supported. In both analysis members of Salvia and Hyssopus were embedded within Nepetinae. Agastache, Meehania and Schizonepata were strongly supported as monophyletic genera. Subtribe Menthinae was retrieved as non-monophyletic because Lycopus, Prunella, Horminum and Hyssopus were closer to Nepetinae. Lycopus formed a distinct clade which was strongly supported in Bayesian tree. Similarly, Prunella and Horminum were also supported as a distinct clade closer to Nepetinae. Hyssopus was embedded in Nepetinae through Lallemantia and it was strongly supported in Bayesian analysis (PP 0.99) while support was lower in ML tree where BS was 70%. Many genera within Menthinae were non-monophyletic including Micromeria, Clinopodium, Satureja, Thymus, Monarda and Cunila. The bootstrap support for many genera was low and their relationships were not well resolved in both analysis.

Clade Lamioideae received strong support in both consensus trees (BS = 97, PP = 1.0) (Fig 3.9 & Fig 3.10). The inter-generic support was moderate. Tribal boundries were largely unresolved particularly Leonureae and Paraphlomideae. Tribe Pogostemoneae was non- monophyletic, Leucosceptrum grouped with Gomphostemma in tribe Gomphostemmateae with strong Bayesian support PP 1.0 and BS value 100. However, the clade which included the rest of Pogostemoneae genera was recovered with a considerable support in Bayesian tree (PP 0.97) and ML tree (BS 80). Pogostemoneae occupied a basal position within Lamioideae clade. Clade Gomphostemmateae (BS = 61, PP = 0.67) missed the genus Bostrychanthera which was found with Stachys in ML tree with a negligible support and its position was also not evident in Bayesian tree. Chelonopsis also appeared as a separate clade confirming the non-monophyly of Gomphostemmateae.

The monophyly of Phlomideae, Marrubieae, Lamieae and Leucadeae was weakly to moderately supported in both analysis. These four tribes have a more derived position in Lamioideae. Ballota integrifolia (Marrubieae) and Eriophyton wallichii (Lamieae) did not occupy an evident position. Stachydeae was not well supported in both trees and Stachys was confirmed as polyphyletic or paraphyletic. Synandreae was not represented in the analysis. Lamioideae incertea sedis genera Colquhounia, Galeopsis and Roylea remained unplaced.

3.3.4 Phylogenetic analysis based on nuclear ITS region

The number of taxa for ML (RAxML) and Bayesian analysis of the nuclear ITS region was 413 while number of characters was 1011. In comparison to Maximum Likelihood (ML) consensus tree (Fig. 3.11) the Bayesian Inference (BI) tree (Figure 3.22) presents nodes which are better resolved. The subfamilies received strong bootstrap values in the Bayesian tree in contrast to the ML tree. Scutellarioideae, Ajugoideae, Prostantheroideae clades were well resolved and monophyletic having strong support (PP 1.0) in Bayesian tree.

Clade Viticoideae received moderate to strong support, ML demonstrated BS 87% (Fig 3.11) while Bayesian support is PP 0.93 (Fig 3.12). However, it was evident that Viticoideae was not monophyletic because Petraeovitex wass embedded within incertea sedis clade of Lamiaceae in ML tree with a negligible support. Similar relationship was found in Bayesian consensus tree. Vitex was also not monophyletic based on nuclear region analysis. Lamiaceae incertea sedis genera Tsoongia and Teijsmanniodendron were grouped in Viticoideae. Tsoongia axillariflora and Vitex vestita were clustered together with a strong support in both ML and Bayesian analysis (BS 100, PP 1.0). Teijsmanniodendron formed a strongly supported sub-clade within Viticoideae wherein support values were PP 1.0 and BS 97. Viticipremna and incertea sedis Teijsmanniodendron are recovered as monophyletic genera.

The Scutellarioideae clade was recovered as a strongly supported monophyletic group (BS = 100, PP 1.0). However, Scutellarioideae only included genus Scutellaria in the nuclear region analysis.

The clade representing the Ajugoideae additionaly included Trichostema, Amethystea, Schnabelia, Huxleya, and Rubiteucris to Ajuga, Teucrium, Teucridium, Oxera, Faradaya, Aegiphila, Tetraclea, Kalaharia, Caryopteris and Clerodendrum which were analysed in plastid region analysis. Trichostema and Amethystea together received a strong support in Bayesian analysis (PP 1.0) whereas bootstrap support is low in ML analysis (BS 86). Similar trend was observed in Rubiteucris and Schnabelia with support values PP 1.0 and BS 88 (Fig 3.12 & Fig 3.11) . The relationship between Rubiteucris-Schnabelia sub-clade and Teucrium- Teucridium sub-clade was strongly supported in Bayesian tree (PP 1.0), although the support is significant in ML tree (BS 90). Huxleya linifolia was embedded in Clerodendrum in both consensus trees. Teucrium and Teucridium appear to be sister genera (BS = 70, PP = 0.98). Oxera and Faradaya were retrieved together in the similar clade with very strong support (BS = 100, PP = 1.0). Genus Rotheca did not appear in the Ajugoideae clade, it was recovered within Lamioideae however, the support is weak in both trees.

Clade Prostantheroideae appeared to be grouped close to Viticoideae but the relationship was not strongly supported. The monophyly of Prostantheroideae was confirmed in Bayesian analysis having strong support (PP 1.0) (Fig 3.12) however the support was lower in ML tree (BS 61). This clade was not well delimited into two distinct tribal groups (Chloantheae and Westringieae). Brachysola, Chloanthes, Cyanostegia, Hemiphora, Newcastelia, Physopsis, Pityrodia and Dicrastylis comprise the Chloantheae. Westringieae included only two genera, Prostanthera and Westringia. Incertae sedis Callicarpa along with Incertae sedis Tectona hamiltoniana made a negligibly supported sister group with Prostantheroideae in ML tree (Fig 3.11).

Monophyly of the Nepetoideae clade was very weakly supported in both trees (BS = 26, PP = 0.53). Tribal groups were often not recovered non-monophyltic in Nepetoideae based on nuclear region analysis. Tribe Elsholtzieae was retrieved as non-monophyletic. Elsholtzia and Perilla are monophyletic in both consensus trees but not present in the same clade therefore, making Elsholtzieae non-monophyletic. There is weak support for clade Ocimeae in Bayesian tree (PP 0.57) and negligible in ML tree. Sub-tribal delimitation was not obvious and poorly resolved in Ocimeae. Eriope, Hypenia, Hyptidendron, Plectranthus, Orthosiphon, Hanceola, Lavandula and Siphocranion were significantly to strongly retrieved as monophyletic groups. Coleus and Plectranthus were not grouped together in both trees.

Tribe Mentheae was only monophyletic in ML analysis having BS 50% (Fig 3.11). However, within Mentheae inter-generic boundaries of many genera were not delimited in both trees. Here sub-tribal delimitation of Mentheae was almost impossible to draw based on nuclear region sequences. Sub-tribe Salviinae constitutes two clades. Salvia was confirmed as non- monophyletic. Melissa and Neoeplingia were found among Salviinae and their presence was strongly supported in both trees. Subtribe Nepetinae received a moderate support of BS 70. In Bayesian tree it appeared to be divided into two clades. In both analysis members of Salvia and Hyssopus were embedded within Nepetinae. Agastache, Meehania, Glechoma and Marmoritis were strongly supported as monophyletic genera. Subtribe Menthinae was retrieved as non-monophyletic because Lycopus, Prunella, Horminum and Hyssopus were closer to Nepetinae. Lycopus formed a distinct clade which was strongly supported in Bayesian and ML trees. Similarly, Prunella and Horminum were also supported as a distinct clade. Hyssopus was embedded in Nepetinae through Dracocephalum moldavica and it was strongly supported in Bayesian analysis (PP 1.0) while support is lower in ML tree where BS was 69%. Many genera within Menthinae were non-monophyletic including Micromeria, Clinopodium, Satureja, Thymus, Mentha and Cunila. The bootstrap support for many genera was low and their relationships were not well resolved in both analysis.

Based on nuclear region the monophyly of subfamily Lamioideae received very weak support in both consensus trees (see Fig 3.11 & Fig 3.12). Tribal boundries were largely unresolved though inter-generic support was significant. Tribe Pogostemoneae received weak Bayesian support (PP 0.71) and negligible support in ML tree. Lamioideae incertae sedis Colquhounia was found in Pogostemoneae clade as a non-monophyletic genus. Chelonopsis and Bostrychanthera constituted the clade Gomphostemmateae with strong support in Bayesian analysis (PP 0.97) and BS 76% in ML analysis, however, tribe Gomphostemmateae was non-monophyletic due to presence of Gomphostemma in other tribes. Lamioideae incertae sedis Galeopsis was grouped close to Chelonopsis-Bostrychanthera clade in both trees.

The monophyly of Phlomideae received a moderate Bayesian support (PP 0.64) and negligible in ML analysis. Genus Phlomoides was clearly non-monophyletic in both analysis. Stachydeae was well supported in both trees (BS = 97, PP = 1.0) and Stachys was confirmed as polyphyletic or paraphyletic.The rest of Lamioideae tribes occupied a more derived position and their delimitation was not clear. Synandreae was not represented in the analysis. Lamioideae incertea sedis Roylea remained unplaced.

Nepetoideae

Fig 3.4: The RAxML phylogenetic tree based on rbcL gene sequences of Lamiaceae. Bootstrap values are indicated, BP > 90 are considered high support and < 80 showing weak support. Groups of Lamiaceae are defined here according to the classification of Harley et al. (2004). Species highlighted in Bold letters were collected from Pakistan.

Nepetoideae

Ajugoideae Ajugoideae

Incertea sedis sedis Incertea

Fig 3.4 (continued)

Lamioideae

Scutellarioideae

Fig 3.4 (continued)

Viticoideae

Incertea sedis sedis Incertea

Fig 3.4 (continued)

Lamioideae

Fig 3.5: The RAxML phylogenetic tree based on matK gene sequences of Lamiaceae. Bootstrap values are indicated, BP > 90 are considered high support and < 80 showing weak support. Groups of Lamiaceae are defined here according to the classification of Harley et al. (2004). Species highlighted in Bold letters were collected from Pakistan.

Lamioideae

Scutellarioideae Scutellarioideae

Ajugoideae

Nepetoideae

Fig 3.5 (continued)

Nepetoideae

Prostantheroideae

Fig 3.5 (continued)

Nepetoideae

Lamioideae

Prostantheroideae

Ajugoideae

Fig 3.6: The RAxML phylogenetic tree based on trnH-psbA spacer sequences of Lamiaceae. Bootstrap values are indicated, BP > 90 are considered high support and < 80 showing weak support. Groups of Lamiaceae are defined here according to the classification of Harley et al. (2004). Species highlighted in Bold letters were collected from Pakistan.

Lamioideae Lamioideae

Ajugoideae

Nepetoideae

Fig 3.6 (continued)

Nepetoideae

Fig 3.6 (continued)

Nepetoideae

Ajugoideae

Fig 3.6 (continued)

Nepetoideae

Fig 3.9: The RAxML phylogenetic tree based on sequences of trnL-trnF region of Lamiaceae. Bootstrap values are indicated, BP > 90 are considered high support and < 80 showing weak support. Groups of Lamiaceae are defined here according to the classification of Harley et al. (2004). Species highlighted in Bold letters were collected from Pakistan.

Nepetoideae

Fig 3.9 (continued)

Nepetoideae

Lamioideae

Fig 3.9 (continued)

Lamioideae

Scutellarioideae

Prostantheroideae

Ajugoideae

Viticoideae

Fig 3.9 (continued)

Nepetoideae

Fig 3.10: Bayesian 50% majority rule consensus tree of Lamiaceae based on plastid trnL-trnF sequence data. Numeric values indicate Bayesian posterior probability (PP) support values. Values > 0.95 are considered as strong support, > 0.90 showing moderate support and < 0.90 with weak support. Lamiaceae group names are indicated to the right of tree in colours. Species names in Bold indicate taxa collected from Pakistan.

Nepetoideae

Fig 3.10 (continued)

Nepetoideae

Lamioideae

Fig 3.10 (continued)

Lamioideae

Scutellarioideae

Viticoideae

Prostantheroideae

ideae Ajugo

Fig 3.10 (continued)

Nepetoideae

Fig 3.11: The RAxML phylogenetic tree based on ITS sequences of Lamiaceae. Bootstrap values are indicated, BP > 90 are considered high support and < 80 showing weak support. Groups of Lamiaceae are defined here according to the classification of Harley et al. (2004). Species highlighted in Bold letters were collected from Pakistan.

Nepetoideae

Fig 3.11 (continued)

Nepetoideae

Lamioideae

Fig 3.11 (continued)

Lamioideae

Ajugoideae

cutellarioideae

S

Prostantheroideae

Vitioideae Vitioideae

Fig 3.11 (continued)

Nepetoideae

Fig 3.12: Bayesian 50% majority rule consensus tree of Lamiaceae based on nuclear ITS sequence data. Numeric values indicate Bayesian posterior probability (PP) support values. Values > 0.95 are considered as strong support, > 0.90 showing moderate support and < 0.90 with weak support. Lamiaceae group names are indicated to the right of tree in colours. Species names in Bold indicate taxa collected from Pakistan.

Nepetoideae

Fig 3.12 (continued)

Nepetoideae

Lamioideae

Fig 3.12 (continued)

Lamioideae

Ajugoideae

cutellarioideae

S

Prostantheroideae

Vitioideae Vitioideae

Fig 3.12 (continued)

Discussion

4.1 Section I: DNA Barcoding of herbal medicinal products (HMPs) of Lamiaceae from Pakistan.

4.1.1 Difficulty in DNA extraction and amplification of HMPs

Wallace et al. (2012) reported a relatively low DNA barcode success rate of 48% in their study. There may be different reasons of this low success rate including different manufacturing protocols of HMPs and the type of plant material (for example, seeds, leaf, stem, roots, barks) used in the herbal formulations. The natural presence of several secondary metabolites, including polysaccharides, glycoproteins and phenolics in plants can result in interference with DNA isolation, amplification and sequencing (Schori et al., 2013). Therefore, inefficient laboratory protocols used for DNA isolation, amplification and sequencing of haplotypes from herbal products may impede the success. In addition, degradation at primer binding sites may also contribute to differential amplification success of selected genes in samples with potentially degraded DNA (Newmaster et al., 2013). We faced some failures in DNA extraction by using the standard CTAB method for Ocimum, Lycopus, Nepeta and Origanum which we overcame by replacing the CTAB protocol with the PowerPlant Pro DNA Isolation Kit. It suggests that DNA extraction from HMPs is not unachievable from common forms of herbal materials (leaf, seed, root, flowers, and powder). There are previous studies (Cimino, 2010; Stoeckle et al., 2011; Baker et al., 2012; Wallace et al., 2012; Newmaster et al., 2013; Michel et al., 2016) in support of DNA based methods for quality control of herbal products wherein DNA was successfully extracted. Majority of our samples amplified at 1.5 mM MgCl2, a higher concentration (2 mM) of MgCl2 was required for successful PCR of nine samples only. We tried plant enhancer which comes with the KAPA3G Plant PCR Kit according to manufacturer’s protocol, yet failed to get the desired results for few samples. Hence, it suggests that there are instances when PCR from herbal pant material become challenging in the presence of inhibitors and absence of primers which do not perfectly match the target sequences.

4.1.2 Insufficient reference sequence data Several Lamiaceae species used as medicinal plants do not have reference sequence data available in GenBank and it became a major problem for our species-level identification of Ajuga bracteosa, Ajuga parviflora, Lallemantia royleana, Plectranthus rugosus, Leucas cephalotes, Nepeta ruderalis, Otostegia limbata, Salvia lanata and Salvia moorcroftiana I. There was not a single reference sequence available for these species for either of our selected markers. Hence, MEGABLAST hits showed congeneric species but not the said species, resulting in the genus-level identification only and decreasing the success rate of species-level discrimination. Our approach is in accordance with Hollingsworth et al. (2009) who stated that species with a single sample are considered potentially distinguishable if the sequence is unique (i.e. there is a potential that successful species-level discrimination may be achieved, but further sampling is needed to verify this). For Hyssopus officinalis only two rbcL reference sequences were available while Salvia aegyptiaca had only one psbA-trnH sequence in GenBank. Lallemantia was another exception for which no reference data was found for any member of this genus. Our voucher specimen are the first ever sequences for these aforementioned species deposited in GenBank and publicly available. With such scant data the possibility of misidentification by the suppliers, or the chances of mislabeled sequences in GenBank always remain. It is not always possible to identify an unknown or suspect plant where there is no reference sequence generated. The family or genus level identification may take place but it is unlikely that species identification will be confirmed. Newmaster et al. (2013) established a reference database of barcodes termed as Sequence Reference Material (SRM) Library to which they compared sequences to prevent these problems. However GenBank is adequately informative when we are working with limited resources, or performing preliminary research. Another potential problem which we encountered was the deposited reference data in GenBank do not always include the complete gene region. A partial sequence from a less variable portion of a gene may lead to a high match percentage that does not reflect an accurate identification of the query sequence. rbcL and matK regions are relatively long (approximately 1430 and 1550 bp respectively) but the rate of evolution might not be same for complete gene regions. The trnH-psbA spacer may be much shorter (200-650 bp, Kress et al., 2005) than rbcL and matK, so a 99% match using it as a barcode may be more accurate than a 99% match using either of these gene regions (Schori and Showalter, 2011). 4.1.3 Pros and cons of rbcL, matK and trnH-psbA regions

CBOL (The Consortium for the Barcode of Life) recommended the two-locus rbcL–matK combination as the universal plant DNA barcode in 2009. The trnH-psbA spacer and the nuclear internal transcribed spacer 2 (ITS2), are also widely used (Chen et al., 2010; Gao et al., 2010; Yao et al., 2010; Fu et al., 2011, Han et al., 2012; Newmaster et al., 2013; Michel et al., 2016). Our findings showed that rbcL and matK reasonably amplified among Lamiaceae with 89% success for both but trnH-psbA demonstrated a relatively lower success (71%). However, we were able to get the good quality bidirectional reads for rbcL and matK but rbcL was not variable enough to discriminate species. rbcL was able to distinguish only four samples at species-level while matK provided species identification of ten samples. trnH-psbA performed better than rbcL by pinpointing the taxonomic identity of eight species, although its amplification/sequencing success remained lower. The concatenated phylogenetic analyses of rbcL and matK was unable to resolve species because of insufficient reference sequence data in GenBank which resulted in extensive missing data of either rbcL or matK sequences in GenBank for many accessions per species (results not shown here). Both matK and trnH-psbA carried out 40% success rate for species-level identification wherein matK had better amplification/sequencing success. Therefore, based on this investigation, matK proved to be the potential barcode for Lamiaceae HMPs used in Pakistan. The overall success rate (excluding the four misidentified samples) of barcoding the Lamiaceae HMPs in this study is 82% at genus-level, however, it reduces to 44% at species-level identification attributed to the inadequate availability of reference sequence data and related issues discussed earlier.

4.1.4 Challenges, improvements and regional recommendations for DNA barcoding of HMPs

To date, as far as we know this is the first study published from Pakistan on the quality assurance of HMPs generally and particularly on Lamiaceae where we targeted a big herbal pharmaceutical industry and local herbal stores of the country’s capital city Islamabad. Contamination and adulteration in herbal products offers considerable health risks for consumers. Previously, contamination and substitution in several products with plants that have known toxicity, side effects and/or negatively interact with other herbs, supplements, or medications has been reported by other studies (Lin et al., 2010; Stoeckle et al., 2011; Newmaster et al., 2013). Our results found two product substitutions but it was difficult to affirm whether it was an accidental misidentification of a bulk product or a fraudulent market substitution for a cheaper product. Unlabeled plant fillers may also be found in herbal products, which may pose a potential health risk for consumers. With the recent advances in DNA barcoding methods, it is potentially possible to meet the challenge of authentication of herbal products or their raw material for routine market analysis (Sucher & Carles, 2008).

The first major challenge which created the limitations in our study was the lack of herbal barcode library for reference sequence data. The similar problem has been reported in market studies from North America, where scientists tested the authenticity of herbal products without herbal SRM barcode library (Stoeckle et al., 2011; Baker et al., 2012; Wallace et al., 2012). Therefore, it is important to develop the regional herbal barcode library where we can find the reference sequence data of all the medicinal plants and their haplotypes being used in the region. Once such a regional herbal barcode library will be created, it would be possible to overcome the insufficiency of authentic reference sequence data for quality control by regulatory authorities as well as the herbal pharmaceutical industries. A similar approach was taken by Newmaster et al. (2013) where they developed a SRM herbal barcode library for their study. Moreover, it is important to Pakistan that a broader regional herbal barcode library is established which not only includes the reference sequence data from Pakistani voucher specimen’s but also from its neighboring countries i.e. China, Iran, India and Afghanistan. It is because we found in our study that the raw material for HMPs do not necessarily comes from within the country but has been imported from the border countries. The ‘GenBank’ and ‘Medicinal Materials DNA Barcode Database’ (Lou et al., 2010) provides barcodes for many species of medicinal plants but without the information about voucher specimens which is an essential component of any DNA Barcode (Hebert et al., 2004). Therefore, it provides another reason for construction of herbal barcode library.

The second challenge occurred due to the selection of plastid barcode regions only, which provided low species resolution for herbal products. Chen et al. (2010) selected ITS2 for use in identifying medicinal plants provided the reasons (i) it comes from the nuclear genome, which has a different rate of evolution than the plastid genome (ii) it provides high species resolution (iii) it is a much shorter sequence allowing higher recovery from processed plant materials found within herbal products. Han et al. (2012) suggested the use of ITS2 for identification of medicinal plants of Lamiaceae. Hence, ITS2 may present a more successful barcode for Pakistani Lamiaceae HMPs and should be tested in future investigations.

There has been a considerable interest worldwide in traditional and alternative medicine, particularly herbal products over the past few decades. The World Health Organization also emphasizes the crucial role of alternative and traditional medicines in preventive and curative health, especially in developing countries and encourages member states to support traditional medicines and plan for formulation of policies with appropriate regulations (WHO, 2002).The herbal medicine as an alternative system of medicine is largely prevailing in South Asian countries such as Pakistan, India, Srilanka and Bangladesh. About half of the industrialized world’s population and 75-90% of the developing countries population still depend on the alternative systems of medicine. In Pakistan, traditional medicines have been a strong part of our cultural heritage and playing a significant role in providing health care to a large part of the population. The family Lamiaceae is one of the largest family of angiosperms found in Pakistan. Hedge (1990) reported 60 genera and about 212 species of Lamiaceae present in Pakistan. Many species in the family Lamiaceae have been widely used for the treatment of cardiovascular diseases, stroke, and other conditions, and are largely used by indigenous people, herbal pharmaceutical industries and the local herb stores. Hence authenticating these species has become an important area of research. We need to have scientific evidence on quality, efficacy, purity and safety of our HMPs in addition to good manufacturing practice (GMP), compliant facilities and manufacturing processes. The quality assurance or quality control starts right away from the authentication of raw material used in the preparation of HMPs, therefore the suppliers must be placed in quality chain, more importantly bringing our herbal collectors and sellers in spot light to ensure quality. There has been lack of concerted efforts for proper utilization of traditional medicines in the health care system. Globally there is a serious focus on regional and traditional ways of treatment to cure modern diseases. The Ministry of Health Pakistan along with National Council of Tibb (NCT), a regulatory authority for traditional and herbal system of medicine is determined to preserve and promote primitive practices by old practitioners and have recently initiated efforts to take measurements for quality assurance of herbal medicines, as it has bountiful capacity to enhance Pakistan’s economy in traditional medicine. The demand for a product authentication service that utilizes molecular biotechnology can be met by DNA barcoding which will restore the consumer confidence. The regulatory authorities should not only adopt DNA barcoding as an improved scientific method for standardization of herbs but also for conservation of these resources.

4.2 Section II: Evaluation of phylogenetic utility of cpDNA rbcL, matK and trnH-psbA regions

This section is in continuity with the previous section (DNA barcoding) of our studies. We used the sequence data produced in the previous part of our study in addition to the sequence data downloaded from GenBank for the selected cpDNA barcoding regions. To estimate the phylogenetic utility of our selected DNA barcoding regions (rbcL, matK and trnH-psbA), the three sequence data sets were analysed independently by using Maximum Likelihood (ML) as implemented in RAxML (Fig 3.4-3.6). The bootstrap support and the resolution of clades and nodes were described for the seven subfamilies of Lamiaceae based on individual cpDNA barcoding regions.

The findings suggested matK as the best gene region which is able to resolve the subfamilies and provided the monophyletic or non-monophyletic genera with strong support. matK coding region is one of the rapidly evolving cpDNA regions among angiosperms and therefore, exhibits considerable variation to provide phylogenetic signals. rbcL is not able to resolve the clades with a strong bootstrap (BS) support. The low variation in rbcL region makes it a bad choice for studying closely related species.

It is difficult to align the spacer region psbA-trnH across the whole family. As a result it affected the tree topology and did not produce the well resolved clades of subfamilies. Whitlock et al. (2010) reported frequent inversions in the psbA-trnH region and stated that if these inversions are not recognized and aligned appropriately, it may result in large overestimates of the number of substitution events separating closely related lineages and clustering more distantly related taxa closer that share the same form of the inversion. Zhu et al. (2010) also indicated the difficulty in sequence alignment of trnH-psbA due to frequent occurrence of deletions and insertions in this region and preferred the manual editing of sequence traces. Additionally presence of a poly- A/T structure in trnH-psbA region decreases the success rate of sequencing. Sang et al. (1997) observed phylogenetic performance of trnH-psbA sequences lower than of matK gene sequences. Kim and Lee (2005) encountered some resolution but rather low statistical support of nodes. Scheen et al. (2004) found variability similar to trnL–trnF but slightly lower resolution and support in resulting topologies. Short sequence length of the trnH-psbA spacer will also limit the power of this marker. Due to its specific molecular architecture, the trnH-psbA spacer can only be used for phylogeny inference among closely related taxa (e.g., species within genera), after excluding more or less large proportions of its sequence due to mutational hotspots (Borsch & Quandt, 2009).

4.3 Section III: Molecular phylogenetics of Lamiaceae based on plastid (trnL-trnF) and nuclear (ITS) markers

This investigation is aimed to determine the monophyly of Pakistani Lamiaceae in particular and to improve the infra-familial relationships in general within Lamiaceae based on plastid and nuclear datasets. Previously there is no such study undertaken from Pakistan which evaluates the current status of species based on DNA sequence data. The Flora of Pakistan (Labiatae) which is based on morphological studies was published in 1990 and is not revised since then. Lamiaceae is taxonomically cumbersome due to highly varied phenotypic characters and hybridization. Therefore, molecular systematics which is more reliable and robust as compared to morphological analysis (biased character selection) can provide better information about the present status of Lamiaceae from Pakistan. Subfamily Symphorematoideae and Prostantheroideae do not occur in Pakistan but it has been included in the analysis through GenBank data to reconstruct a comprehensive phylogeny of Lamiaceae worldwide. Harley et al. (2004) recognized seven subfamilies of Lamiaceae: (1) Symphorematoideae (2) Viticoideae (3) Ajugoideae (4) Prostantheroideae (5) Scutellarioideae (6) Lamioideae (7) Nepetoideae. This is the most recent and comprehensive treatment of Lamiaceae to date and we have followed it primarily for taxonomic groupings in this study. We employed RAxML Blackbox webserver (Stamatakis, 2008) and MrBayes v.3.2 (Huelsenbeck and Ronquist, 2001) softwares for building phylogenetic trees of trnL- trnF and ITS datasets.

4.3.1 Viticoideae

Recognition of Viticoideae by Harley et al. (2004) within Lamiaceae is another of the major modifications to the traditional treatment where it has been part of Verbanaceae. During the process a number of genera were removed and the Viticoideae sensu Briquet was retained as a subfamily. Rotheca and Clerodendrum were transferred to Ajugoideae whereas Callicarpa and Tectona were listed incertea sedis. Harley et al. (2004) recognized Viticoideaea as paraphyletic or possibly polyphyletic based on morphological, phytochemical and molecular data. They included ten genera in Viticoideae Petitia, Cornutia, Premna, Viticipremna, Tsoongia, Paravitex, Vitex, Teijsmanniodendron, Gmelina, and Pseudocarpidium. Cornutia, Premna, Tsoongia, Teijsmanniodendron and Gmelina were considered incertea sedis in some studies. Wagstaff and Olmstead (1997) reported two distinct clades within Viticoideae, one centered on Vitex and another constituted Premna, Gemelina and Cornutia. Similar relationship is inferred in our ITS consensus trees wherein two distinct clades are identified; one includes Gmelina, Premna, Petraeovitex and Peronema while the other clade includes Vitex, Viticipremna, Tsoongia, Teijsmanniodendron. This confirms the non-monophyly of Viticoideae in the present study (Fig 3.11 & Fig 3.12). Bramley et al. (2009) sunk the Paravitex, Tsoongia and Viticipremna into Vitex, decreasing the number of Viticoid genera to seven. Our findings also corroborates to Bramley et al. (2009) for reducing Tsoongia and Viticipremna in synonymy to Vitex. The present ITS analysis provides the convincing support to include Teijsmanniodendron in Vitex (Fig 3.11 & Fig 3.12) whose generic status remained upheld in previous studies. The identification between Teijsmanniodendron and Vitex species is often confusing based on morphological characters. Traditionally Teijsmanniodendron was separated from others because of its one-seeded, capsule-like fruit rather than drupaceous (Koorders, 1904). Vitex negundo sampled from Pakistan established a sister relationship with Vitex trifolia (Fig 3.9-3.12).

4.3.2 Scutellarioideae Addition of Holmskioldia in subfamily Scutellarioideae is one of the notable modifications done by Harley et al. (2004) in the traditional classification. The other four genera are Wenchengia, Renschia, Tinnea and Scutellaria. Monotypic Wenchengia has long been controversial for its phylogenetic position. It has been suggested to place Wenchengia in Scutellarioideae based on having similar corollas, stamens and tuberculate nutlets. Li et al. (2012) provided the molecular, cytological and morphological evidence to include Wenchengia in Scutellarioideae. Our observations confirmed the Harley et al. (2004) and Li et al. (2012) findings that Holmskioldia and Wenchengia are part of Scutellarioideae, however, our analysis does not prove the sister relationship of Wenchengia alternifolia with Holmskioldia-Scutellaria. We recovered the Wenchengia and Holmskioldia embedded within Scutellaria clade (Fig. 3.9 & Fig 3.10). Scutellaria linearis and the other Scutellaria sp. sampled from Pakistan occupied a basal position in Scutellarioideae clade along with Scutellaria amoena.

4.3.3 Ajugoideae

Subfamily Ajugoideae as proposed by Harley et al. (2004) has 24 genera, cosmopolitan, but many are temperate, especially South East Asia to Australia. There are many previous studies which have resolved the generic delimitations and phylogenetic relationships among Ajugoideae (Steane et al., 1997; Cantino et al., 1999; Steane et al., 1999; Steane et al., 2004; Huang, 2002; Shi et al., 2003; Yuan et al., 2010; Barrabe et al., 2015). Our analysis included Ajuga bracteosa, Ajuga parviflora and Ajuga reptans collected from Pakistan confirming Ajuga as monophyletic. Li et al. (2016) demonstrated Ajugoideae comprising of two large clades based on cpDNA analysis. One primarily temperate and centered on Teucrium and the other centered on Clerodendrum. The latter clade comprised a primarily tropical clade including Clerodendrum and related genera, and a primarily temperate clade made up of Ajuga, Trichostema, Caryopteris and related genera. This scheme has also been identified in our plastid trnL-trnF analysis but nuclear ITS trees showed some deviation wherein both the temperate clades centered on Teucrium and Ajuga, Trichostema, Caryopteris made one major clade while tropical Celrodendrum and its allies made the other clade. Faradaya was included in Oxera by Barrabe et al. (2015) and our findings coincide with them. Sinking Huxleya into Clerodendrum as proposed by Steane et al. (2005) is evident from our study and segregation of Kalaharia and Tetraclea from Clerodendrum reported by Yuan et al. (2010) is also supported here. The striking incongruence to previous Ajugoid studies is unresolved position of Rotheca in the present study (Fig. 3.9-3.12).

4.3.4 Prostantheroideae

Harley et al. (2004) divided the endemic Australian subfamily Prostantheroideae into two tribes viz. tribe Chloantheae comprising of 10 genera and tribe Westringieae including 6 genera. Our analysis included both tribes and we confirm the monophyly of each tribe. It also becomes obvious that combined Prostantheroideae is monophyletic here (Fig. 3.9- 3.12).Olmstead et al. (1998) proposed the monophyly of Prostantheroideae. Recently Bendiksby et al. (2011) and Li et al. (2012) showed the monophyly of Prostantheroideae however, the analysis included members of tribe Westringeae only. We have identified sister relationship between incertea sedis Callicarpa and Prostantheroideae and also confirm Callicarpa monophyletic. Wrixonia includes only two species (Conn, 2004; Harley et al., 2004). This genus has clear morphological resemblance with Prostanthera which is the largest genus of the tribe Westringieae. Wilson (2010) reported that Prostanthera is paraphyletic with respect to Wrixonia. Based on these findings, Wrixonia is reduced to the synonymy of Prostanthera (Wilson et al., 2012). Our study also provides a strong support for including Wrixonia in Prostanthera to maintain a monophyletic Prostanthera (Fig. 3.10).

4.3.5 Symphorematoideae

Recognition of Symphorematoideae by Harley et al. (2004) within Lamiaceae is one of the major changes to the traditional classification (earlier considered as a distinct family: Symphoremataceae). Subfamily Symphorematoideae is under represented in our study. There are only three genera of this subfamily and we could only include one species Congea tomentosa in our ITS analysis. Wagstaff et al. (1998) carried out a study where they showed Congea tomentosa nested within Labiatae s. l. based on rbcL analysis whereas it appeared at basal position in ndhF analysis and as a sister group to subfamily Nepetoideae based on combined analysis of ndhF and rbcL but concluded that the addition of more members from Symphoremataceae is required to further establish the relationship. Bendiksby et al. (2011) showed Congea as sister to a clade of viticoid genera. Our ITS analysis revealed Congea tomentosa embedded within the viticoid clade but we have not concluded this relationship due to scant sampling.

4.3.6 Nepetoideae

The subfamily Nepetoideae contains 105 genera (Harley et al., 2004) and is the largest subfamily in the Lamiaceae (Wagstaff et al., 1995; Wagstaff et al., 1998; Paton et al., 2004). Hexacolpate pollen, gynobasic style, an investing embryo, presence of rosmarinic acid and exalbuminous seeds are the noteworthy synapomorphies through which it appeared to be monophyletic (Cantino and Sanders 1986, Harley et al., 2004). Other studies also reported it as monophyletic (Wagstaff et al., 1995; Wagstaff and Olmstead, 1997). Our investigation also reports the Nepetoideae monophyletic (Fig. 3.9-3.12). Harley et al. (2004) recognized three tribes within Nepetoideae: Elsholtzieae, Mentheae, and Ocimeae. The relationship among these three tribes remains conflicting in our analysis. Harley et al. (2004) further divided tribe Mentheae into three subtribes: Salviinae, Menthinae and Nepetinae and tribe Ocimeae into five subsequent subtribes: Lavandulinae, Hanceolinae, Hyptidinae, Ociminae and Plectranthinae. Nevertheless the sub-tribal division of Nepetoideae is very confusing in present study as the findings from plastid and nuclear region are not congruent. We observed that many Nepetoid taxa collected from Pakistan are clumping together particularly in nuclear region analysis which hints towards the probability of hybridization taking place among Pakistan’s Lamiaceae. This view point is further strengthened by the presence of Lamioid species (Stachys and Eremostachys) from Pakistan within Nepetoidea. These Lamioid taxa (described in Lamioideae section) have grouped themselves with Pakistan’s Nepetoid genera therefore, providing the clue of species radiation in place instead of dispersal. There are a number of molecular investigations which have been carried out within the Nepetoideae (Wagstaff et al., 1995; Prather et al., 2002; Paton et al., 2004; Trusty et al., 2004; Walker et al., 2004; Bräuchler et al., 2005; Edwards et al., 2006; Walker and Sytsma, 2007; Bräuchler et al., 2010; Harley and Pastore, 2012).

Since the treatment proposed by Harley et al. (2004), several molecular (Trusty et al., 2004; Walker et al., 2004; Bräuchler et al., 2005, 2010; Edwards et al., 2006; Walker and Sytsma, 2007; Drew and Sytsma, 2011) and morphological (Moon et al., 2008, 2009, 2010; Ryding, 2010a, b) studies have focused on Mentheae and groups within it. These studies showed the non-monophyly of the three subtribes of Mentheae proposed by Harley et al. (2004) and reported that a number of genera remain unplaced/misplaced (Ryding, 2010a; Drew and Sytsma, 2011). Therefore Drew and Sytsma (2012) proposed two new subtribes, Prunellinae and Lycopinae in addition to Harley’s. Generic boundaries in subtribe Menthinae have been under debate especially those taxa associated with the former Satureja s. l. complex (Satureja, Micromeria, Calamintha, Clinopodium, Acinos). The Calamintha samples from Pakistan (Calamintha debilis and Calamintha umbrosa) should be treated as part of Clinopodium. Similarly Micromeria biflora from Pakistan collection is found embedded in Thymus in both plastid and nuclear DNA analysis. Clinopodium vulgare, Ziziphora tenuior, Thymus sp., Mentha longifolia, Mentha spicata, Mentha arvensis, Mentha pulegium, Mentha suaveolens, Mentha x piperita, Mentha longifolia ssp., Mentha aquatica, Clinopodium hydaspidis, Thymus vulgaris, Thymus serpyllum, Origanum vulgare, Origanum majorana, Zataria multiflora, Gontscharovia popovii, Salvia cabulica, Salvia elegans, Salvia plebeia, Salvia officinalis, Salvia aegyptiaca, Salvia cordata, Salvia alba, Salvia nubicola, Salvia asperata, Perovskia artiplicifolia, Perovskia abrotanoides, Rosmarinus officinalis, Melissa officinalis, Lycopus europaeus, Prunella vulgaris, Ziziphora clinopodioides, Lallemantia royleana, Dracocephalum moldavica, Nepeta cataria, Perilla sp., Ocimum basilicum, Ocimum tenuiflorum, Ocimum x citriodorum, Ocimum x africanum, Plectranthus barbatus, Isodon lophanthoides, Isodon coetsa, Isodon rugosus, Lavandula angustifolia, Elsholtzia ciliata, Elsholtzia densa, Perilla frutescens, Monarda fistulosa, Lagochilus cabulicus, Lagochilus cuneatus, Glechoma hederacea and Hyptis suaveolens are the other taxa from Pakistan included in the present analysis.

4.3.7 Lamioideae

Subfamily Lamioideae (including Pogostemonoideae) is the second largest species-rich subfamily among the seven subfamilies proposed by Harley et al. (2004). However, the suprageneric relationships among the Lamioideae remained poorly understood providing no tribal ranks (sensu Harley et al., 2004). Scheen et al. (2010) presented another phylogenetic investigation dividing Lamioideae into nine tribes: Gomphostemmateae, Phlomideae, Leucadeae, Pogostemoneae, Synandreae, Stachydeae, Leonureae, Lamieae, and Marrubieae. Our findings are not congruent with Scheen et al. (2010) because tribal boundries are largely unresolved in our analysis. However, we agree with them that the genera Stachys, Sideritis, Ballota, and Leucas are polyphyletic or paraphyletic as they are also non-monophyletic in our analysis. Bendiksby et al. (2011) proposed a taxonomic update of subfamily Lamioideae and established a new tribe Paraphlomideae which includes Ajugoides, Matsumurella and Paraphlomis. Otostegia aucheri from Pakistan should be treated as Molucella aucheri as proposed by Scheen and Albert (2007) (Fig. 3.9 & Fig 3.10). The relationship between Lagochilus and Otostegia limbata remains unclear, both samples are from Pakistan. Similarly Stachys emodi from Pakistan is not appearing as part of tribe Stachydeae in trnL-trnF analysis. On the other hand Stachys parviflora, Stchys flocossa, Stachys purpurea, Stachys emodi, Eremostachys thyrsiflora and Eremostachys speciosa are found in clade Nepetoideae instead of Lamioideae during nuclear ITS analysis. This might be an indication towards possibility of hybridization taking place among Lamiaceae in Pakistan because these taxa are clustered with Nepetoideae species from Pakistan. The other taxa from Pakistan includes Eremostachys spectabilis, Leucas mollissima, Leucas cephalotes, Leucas urticifolia, Leucas lanata, Lamium amplexicaule, Lamium album, Leonurus cardiaca, Moluccella laevis, Marrubium vulgare, Thuspeinanta brahuica, Stachys byzantina, Stachys palustris, Stachys sylvatica, Anisomeles indica, Craniotome furcata and Colebrookea oppositifolia which are part of analysis here. All these taxa clustered with their respective genera. Further studies of relationships within some tribes are provided by: Synandreae (Scheen et al., 2008; Roy et al., 2016); Leucadeae (Scheen and Albert, 2009); Phlomideae (Pan et al., 2009; Mathiesen et al., 2011; Salmaki et al., 2012); Lamieae (Bendiksby et al., 2011); Stachydeae (Salmaki et al., 2013); Gomphostemmateae (Xiang et al., 2013).

4.4 Conclusion

This is the first ever investigation based on DNA barcoding identification of herbal medicinal products (HMPs) of family Lamiaceae (mints) from Pakistan and its subsequent phylogenetic assessment based on molecular analysis. The study aimed for the correct identification of Lamiaceae HMPs to fix the problem of adulteration. The mints of Pakistan have never been studied involving nucleotide sequence data and all the previous findings presented in Flora of Pakistan heavily relied on morphological characters only. Hence the present study provides an updated account of Lamiaceae by utilizing the DNA sequence data from Pakistan in support with the data available from the global data bank. Following objectives have been achieved:

1- HMPs representing 32 Lamiaceae plant samples purchased/collected from three herbal stores (Pansar stores) in Islamabad and a herbal pharmaceutical industry were subjected to identification through DNA barcoding. Three plastid loci rbcL, matK and psbA-trnH were selected to barcode these HMPs. MEGABLAST sequence comparison was performed to verify the taxonomic identity of the samples. Four mislabeled samples including one industry sample (Salvia haematodes) and three pansar store samples (Leucas linifolia, Lycopus europaeus and Salvia moorcroftiana II) are identified as completely different species not belonging to Lamiaceae. Two product substitutions (Hyssopus officinalis and Nepeta ruderalis) are found. The overall amplification success for rbcL and matK is 87% and 81% while psbA-trnH showed 69%. matK and psbA-trnH are able to distinguish the species relatively better with 40% success rate than rbcL (16%). On the whole a total of 22 genus- level barcodes (78%) and 12 species-level barcodes (44%) are generated. The species-level identification is considerably low due to insufficient reference data and selection of plastid markers. Our results propose matK as the potential barcode for Lamiaceae HMPs.

2- The performances of these candidate barcoding loci (rbcL, matK, trnH-psbA) are evaluated for Lamiaceae:

(i) rbcL: 21 of 25 rbcL sequences (84%) matched the expected genera based on MEGABLAST. Of these 21, only four rbcL sequences matched the expected species (Melissa officinalis, Nepeta cataria, Rosmarinus officinalis, Stachys byzantina). rbcL lacked sequence variability for resulting in multiple congeneric species hits having the similar % identity. We could only prove their identification to the expected genus level. Nine rbcL sequences (Ajuga bracteosa, Ajuga parviflora, Plectranthus rugosus, Leucas cephalotes, Nepeta ruderalis, Otostegia limbata, Salvia aegyptiaca, Salvia lanata, Salvia moorcroftiana I) could not be identified to species level due to limited reference sequence data available in GenBank. (ii) matK: Of 25 matK sequences, 19 matched the expected genera based on MEGABLAST. 10 of these 19 matK sequences matched their expected species level (Melissa officinalis, Mentha aquatica, Mentha pulegium, Mentha spicata, Nepeta cataria, Ocimum basilicum I, Ocimum x africanum, Rosmarinus officinalis, Salvia plebeia, Stachys byzantina). No reference data could be retrieved for eight species including Ajuga bracteosa, Hyssopus officinalis, Lallemantia royleana, Leucas cephalotes, Nepeta ruderalis, Otostegia limbata, Salvia aegyptiaca and Salvia moorcroftiana I from GenBank, therefore, species level identification was not possible based on matK barcodes.

(iii) trnH-psbA: Lower rate of amplification/sequencing success as compared to rbcL and matK, however, more successful for distinguishing the samples at species level. Only 20 psbA-trnH barcodes were found resulting in 14 sequences which provided identification to expected genera. Eight sequences (Mentha pulegium, Mentha spicata, Mentha longifolia, Mentha aquatica, Ocimum basilicum I, Ocimum basilicum II, Rosmarinus officinalis and Salvia plebeia) came up as successful barcodes for expected species.

(iv) Ambiguous ID: Three HMPs appeared to be highly ambiguous. There was not sufficient reference data available in GenBank for these species with respect to our chosen DNA barcoding markers. Although the barcodes obtained for these species (Lallemantia royleana, Origanum vulgare and Salvia aegyptiaca) were not completely unrelated because they belonged to Lamiaceae and none of the sequences for either of the barcoding region showed up to any other plant family.

3- The phylogenetic utility of DNA barcoding regions (rbcL, matK and trnH-psbA) is estimated by analyzing the three sequence data sets independently by using Maximum Likelihood (ML) method. The number of taxa for ingroup analysis of rbcL, matK and trnH- psbA were 245, 235 and 259 respectively. The findings suggest matK as the best gene region which is able to resolve the subfamilies and provided the monophyletic genera with strong support. rbcL was not able to resolve the clades with a strong bootstrap (BS) support. It was difficult to align the spacer region trnH-psbA across the whole family. As a result it affected the tree topology and did not produce the well resolved clades of subfamilies. 4- Phylogenetic analysis based on plastid trnL-trnF and nuclear ITS region included 398 and 413 ingroup taxa. In comparison to ML consensus trees the Bayesian consensus trees have more resolved nodes. The subfamilies received strong support in the Bayesian analysis in contrast to the ML results. Scutellarioideae, Ajugoideae, Prostantheroideae clades were well resolved and monophyletic having strong support (PP 1.0) in Bayesian tree. Viticoideae is not well resolved. Monophyly of the Nepetoideae clade is well supported. Clade Lamioideae received strong support in both consensus trees although inter-generic support is moderate and tribal boundries are largely unresolved.

5- In Pakistan’s Lamiaceae, hybridization is taking place, particularly evident in the nuclear ITS analysis. Species are radiating in place instead of dispersal. The taxonomic position of some species which was originally based on morphological characters do not corroborate with the molecular analysis.

It concludes that to overcome the gaps still present in this investigation, it is crucial to exploit more plastid and nuclear loci with increased number of taxa for each group. The intense studies more focused on each group (each subfamily) may draw a better picture of Pakistan’s Lamiaceae.

4.5 Future Recommendations

1- The herbal pharmaceutical industries are recommended that they should voluntarily establish the DNA barcoding for the testing of raw materials used in manufacturing their medicinal products.

2- The herbal pharmaceutical industry may build their own herbal barcode library for reference sequence data depending on which plant species do they use in their products. It will be cost effective and will facilitate the rapid authentication to increase sovereign business interests and provide considerable health safety to consumers.

3- DNA barcoding can help the regulatory authorities to devise a mechanism for quality control and may develop check-points during the long supply chain starting from local collectors to the market shelf.

4- Nuclear regions should be tested in future barcoding studies. 5- More comprehensive sampling should be done to test the radiation vs dispersal hypothesis with the addition of more molecular markers.

Literature Cited Abu-Asab MS and Cantino PD (1989). Pollen morphology of Trichostema (Labiatae) and its systematic implications. Systematic Botany; 14(3):359-369. Abu-Asab MS and Cantino PD (1993). Phylogenetic implications of pollen morphology in tribe Ajugeae (Labiatae). Systematic Botany; 18(1):100-122. Abu-Asab MS and Cantino PD (1994). Systematic implications of pollen morphology in subfamilies Lamioideae and Pogostemonoideae (Labiatae). Annals of the Missouri Botanical Garden; 81(4):653-686. Abu-Asab MS, Cantino PD, Nowicke JW and Sang T (1993). Systematic implications of pollen morphology in Caryopteris (Labiatae). Systematic Botany; 18(3):502-515. Alan PL, Monfils AK, Posto AL and Williams RA (2002). Monophyly and phylogeny of Monarda (Lamiaceae): evidence from the internal transcribed spacer (ITS) region of nuclear ribosomal DNA. Systematic Botany; 27(1):127-137. Ali MS, Saleem M, Ali Z and Ahmad VU (2000). Chemistry of Zataria multifora (Lamiaceae). Phytochemistry; 55:933-936. APG (= Angiosperm Phylogeny Group) I (1998). An Ordinal Classification for the Families of Flowering Plants: APG I. Annals of the Missouri Botanical Garden; 85(4):531- 553. APG (= Angiosperm Phylogeny Group) II (2003). An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG II. Botanical Journal of the Linnean Society; 141(4):399-436. APG (= Angiosperm Phylogeny Group) III (2009). An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG III. Botanical Journal of the Linnean Society; 161(2):105-121. APG (= Angiosperm Phylogeny Group) IV (2016). An update of the Angiosperm Phylogeny Group classification for the orders and families of flowering plants: APG IV. Botanical Journal of the Linnean Society; 181(1):1-20. Bailey CD, Carr TG, Harris SA and Hughes CE (2003). Characterization of angiosperm nrDNA polymorphism, paralogy and pseudogenes. Molecular Phylogenetics & Evolution; 29:435-455. Baker DA, Stevenson DW and Little DP (2012). DNA barcode identification of black cohosh herbal dietary supplements. Journal of AOAC International; 95(4):1023-1034. Ball PW and Getliffe FM (1972). Satureja, Acinos, Clinopodium, Calamintha. In: Flora Europaea (eds. Tutin TG, Heywood VH, Burges NA, Moore DM, Valentine DH, Walters SM and Webb DA). Cambridge University Press, Cambridge; 3:163-167. Barber JC, Finch CC, Francisco-Ortega J, Santos-Guerra A and Jansen RK (2007). Hybridization in Macaronesian Sideritis (Lamiaceae): Evidence from incongruence of multiple independent nuclear and chloroplast sequence datasets. Taxon; 56(1):74-88. Barber JC, Francisco-Ortega J, Santos-Guerra A, Turner KG and Jansen RK (2002). Origin of Macaronesian Sideritis L. (Lamioideae: Lamiaceae) inferred from nuclear and chloroplast sequence datasets. Molecular Phylogenetics and Evolution; 23(3):293- 306. Barber JC, Francisco-Ortega J, Santos-Guerra A, Marrero A and Jansen RK (2000). Evolution of endemic Sideritis (Lamiaceae) in Macaronesia: insights from a chloroplast DNA restriction site analysis. Systematic Botany; 25(4):633-647. Barrabe L, Abdelkader GK, Ounemoa J, De Kok RPJ, Robert N And Gateble G (2015). Recircumscription of Oxera (Lamiaceae: Ajugoideae) to include Faradaya based on molecular and anatomical data. Botanical Journal of Linnean Society; 179:693-711. Baris O, Golloce M, Sahin R, Ozer H, Killic H, Ozkan H, Sokmen M and Ozbek T (2006). Biological activities of essential oil and methanol extract of Achillea biekersteinii Afan. (Asteraceae). Tukish Journal of Biology; 30: 65-73. Barthet MM and Hilu KW (2007). Expression of matK: functional and evolutionary implications. American Journal of Botany; 94:1402-1412. Baser KHC (2002). The Turkish Origanum species. In: Oregano. The Genera Origanum and Lippia (ed. Kintzios SE). Taylor & Francis. London; pp 109. Bendiksby M, Salmaki Y, Bräuchler C and Ryding O (2013). The generic position of Stachys tibetica Vatke and amalgamation of the genera Eriophyton and Stachyopsis (Lamiaceae subfam. Lamioideae). Plant Systematics and Evolution; 300(5):961-971. Bendiksby M, Thorbek L, Scheen AC, Lindqvist C and Ryding O (2011). An updated phylogeny and classification of Lamiaceae subfamily Lamioideae. Taxon; 60(2):471- 84. Bentham G (1832–1836). Labiatarum Genera et Species. Ridgeway and Sons, London. pp783.

Bentham G (1836). Labiatae. In: Prodromus Systematis Universalis Regni Vegetabilis (ed. DeCandolle AP). Treuttel, Würtz, Paris; 12:212-226.

Bentham, G (1876). Verbenaceae and Labiatae. In: Genera plantarum (eds. Bentham G & Hooker JD). Reeve and Co., London; 2: 1131-1223.

Blattner FR (1999). Direct amplification of the entire ITS region from poorly preserved plant material using recombinant PCR. Biotechniques; 27(6):1180-1186.

Borsch T and Quandt D (2009). Mutational dynamics and phylogenetic utility of non-coding chloroplast DNA. Plant Systematics and Evolution; 282:169-199.

Bramley GLC, Forest F and DeKok RPJ (2009). Troublesome tropical mints: Re-examining generic limits of Vitex and relations (Lamiaceae) in South East Asia. Taxon; 58(2):500-510. Bräuchler C, Meimberg H, Abele T and Heubl G (2005). Polyphyly of the genus Micromeria (Lamiaceae) - Evidence from cpDNA sequence data. Taxon; 54(3):639-650. Brauchler C, Meimberg H and Heubl G (2010). Molecular phylogeny of Menthinae (Lamiaceae, Nepetoideae, Mentheae)-Taxonomy, biogeography and conflicts. Molecular Phylogenetics and Evolution; 55(2):501-523. Brenan JPM (1954). Plants collected by the Vernay Nyasaland Expedition of 1946. Memoirs of New York Botanical Garden; 9: 1-132.

Briquet J (1895). Verbenaceae. In: Die Naturlichen Planzenfamilien (eds. Engler HGA and Prantl KAE). Leipzig; 4(3a):132-182.

Briquet J (1895-1897). Labiatae. In: Die Naturlichen Planzenfamilien (eds. Engler HGA and Prantl KAE). Leipzig; 4(3a):183-375.

Bunsawat J, Elliott NE, Hertweck KL, Sproles E and Alice LA (2004). Phylogenetics of Mentha (Lamiaceae): evidence from chloroplast DNA sequences. Systematic Botany; 29(4):959-964. Burt S (2004). Essential oils: their antimicrobial properties and potential application in foods- A review. International Journal of Food Microbiology; 94:223-253. Cantino PD (1982). Affinities of the Lamiales: A cladistic analysis. Systematic Botany; 7(3): 237-248. Cantino PD (1992a). Evidence for a polyphyletic origin of the Labiatae. Annals of the Missouri Botanical Garden; 79(2):361-379. Cantino PD (1992b). Towards a phylogenetic classification of the Labiatae. In: Advances in Labiate Science (eds. Harley RM and Reynolds T). Royal Botanic Gardens, Kew; 27- 37. Cantino PD, Harley RM and Wagstaff SJ (1992). Genera of Labiatae status and classification. In: Advances in Labiate Science (eds. Harley RM and Reynolds T). Royal Botanic Gardens, Kew; 511-522. Cantino PD and Sanders RW (1986). Subfamilial classification of Labiatae. Systematic Botany; 11(1):163-185. Cantino PD and Wagstaff SJ (1998). A reexamination of North American Satureja s.l. (Lamiaceae) in light of molecular evidence. Brittonia; 50(1):63-70. Cantino PD, Wagstaff SJ and Olmstead RG (1998). Caryopteris (Lamiaceae) and the conflict between phylogenetic and pragmatic considerations in botanical nomenclature. Systematic Botany; 23(3):369-386. Casiraghi M, Labra M, Ferri E, Galimberti A and Mattia FD (2010). DNA barcoding: a six question tour to improve users’ awareness about the method. Briefings in Bioinformatics; 11:440-453. CBOL Plant Working Group (2009). A DNA barcode for land plants. Proceedings of National Academy of Sciences USA; 106(31):12794-12797. Celiktas OY, Kocabas EE, Bedir E, Sukan FV, Ozek Tand Baser KH (2006). Antimicrobial activities of methanol extracts and essential oils of Rosmarinus officinalis, depending on location and seasonal variations. Food Chemistry; 100:553-559. Charles DJ and Simon JE (1992). Comparison of extraction methods for the rapid determination of essential oil content and composition. Journal of the American Society for Horticultural Science; 115:458-462.

Chater AO and Guinea E (1972). Micromeria. In: Flora Europaea (eds. Tutin TG, Heywood VH, Burges NA, Moore DM, Valentine DH, Walters SM and Webb DA). Cambridge University Press, Cambridge; 3:167-170. Chen F, Chan HY, Wong KL, Wang J, Yu MT, But PPH and Shaw PC (2008). Authentication of Saussurea lappa, an endangered medicinal material, by ITS DNA and 5S rRNA sequencing. Planta Medica; 74:889-892.

Chen P, Drew BT, Li B, Soltis DE, Soltis PS and Xiang CL (2016). Resolving the phylogenetic position of Ombrocharis (Lamiaceae), with reference to the molecular phylogeny of tribe Elsholtzieae. Taxon; 65:123-136.

Chen S, Yao H, Han J, Liu C, Song J, Shi L, Zhu Y, Ma X, Gao T and Pang X (2010). Validation of the ITS2 region as a novel DNA barcode for identifying medicinal plant species. PloS One; 5(1):e8613.

Chen SL and Gilbert MG (1994). Verbenaceae. In: Flora of China, Verbenaceae through Solanaceae, (eds. Wu ZY and Raven PH). Science press Beijing and Missouri Botanical Garden St. Louis; 17:1-49

Cimino MT (2010). Successful isolation and PCR amplification of DNA from National Institute of Standards and Technology (NIST) herbal dietary supplement standard reference material powders and extracts. Planta Medica; 76(5):495.

Clarke CB (1885). Verbenaceae. In: Flora of British India (ed. Hooker JD). London; 4:560- 604.

Conn BJ (1992). Relationships within the tribe Prostanthereae (Labiatae). In: Advances in Labiate Science (eds Harley RM and Reynolds T). Royal Botanic Gardens, Kew; 55- 64.

Conn BJ (2004). Subfamily Prostantheroideae. In Flowering Plants-Dicotyledons, Lamiales (except Acanthaceae including Avicenniaceae) (ed. Kadereit JW). Springer-Verlag, Berlin; 7:167-275.

Conn BJ, Henwood MJ and Streiber N (2011). Synopsis of the tribe Chloantheae and new nomenclatural combinations in Pityrodia s.l. (Lamiaceae). Australian Systematic Botany; 24:1-9.

Cowan RS and Fay MF (2012). Challenges in the DNA barcoding of plant material. Methods in Molecular Biology; 862:23-33. Cronquist A (1981). An integrated system of classification of flowering plants. Columbia University Press, New York. pp. 1262. Cronquist A (1988). The evolution and classification of flowering plants, 2nd ed. New York Botanical Garden, New York. Cuenoud P, Savolainen V, Chatrou LW, Powell M, Grayer RJ and Chase MW (2002). Molecular phylogenetics of Caryophyllales based on nuclear 18S rDNA and plastid rbcL, atpB, and matK DNA sequences. American Journal of Botany; 89(1):132-144. Cullings KW (1992). Design and testing of a plant-specific PCR primer for ecological and evolutionary studies. Molecular Ecology; 1:233-240. Curto MA, Puppo P, Ferreira D, Nogueira M and Meimberg H (2012). Development of phylogenetic markers from single-copy nuclear genes for multi locus, species level analyses in the mint family (Lamiaceae). Molecular Phylogenetics and Evolution; 63: 758-767. Davis PH (1982). Flora of Turkey and the East Aegean Islands, vol. 7. Edinburgh University Press, Edinburgh. DeCandolle A (1836). Prodromus systematis universalis regni vegetabilis 5. Typis Crapelet, Dicta de Vourigard; 9:1883. De Kok RP, Grayer RJ and Kite GC (2000). Relationships of the endemic Australian genus Huxleya Ewart & Rees (Labiatae) based on of fruit and flavonoid characters. Australian Systematic Botany; 13(3):425-428. De Necker N (1790). Plasyrgophytorum. In: Elementa Botanica I. Neowedae; 359-389. DeSalle R (2006). Species discovery versus species identification in DNA barcoding efforts: response to Rubinoff. Conservation Biology; 20:1545–1547. Devey DS, Chase MW and Clarkson JJ (2009). A stuttering start to plant DNA barcoding: microsatellites present a previously overlooked problem in non-coding plastid regions. Taxon; 58:7-15. Doebley J, Durbin M, Golenberg EM, Clegg MT and Ma DP (1990). Evolutionary analysis of the large subunit of carboxylase (rbcL) nucleotide sequence among the grasses (Gramineae). Evolution; 44:1097-1108. Dong W, Xu C, Cheng T, Lin K, and Zhou S (2013). Sequencing angiosperm plastid genomes made easy: a complete set of universal primers and a case study on the phylogeny of Saxifragales. Genome Biology and Evolution; 5:989-997. Doyle JJ (1991). DNA protocols for plants. In: Molecular techniques in taxonomy (eds. Hewitt G, Johnson AWB and Young JPW). NATO ASI Series H, Cell Biology 57. Springer-Verlag, Berlin, Germany; 283-293. Doyle JJ and Doyle JL (1987). A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin; 19:11-15. Drew BT and Sytsma KJ (2011). Testing the monophyly and placement of Lepechinia in the tribe Mentheae (Lamiaceae). Systematic Botany; 36:1038-1049. Drew BT and Sytsma KJ (2012). Phylogenetics, biogeography, and staminal evolution in the tribe Mentheae (Lamiaceae). American Journal of Botany; 99(5):933-953. Drew BT and Sytsma KJ (2013). Phylogenetics, biogeography, and evolution of dioecy in South American Lepechinia (Lamiaceae). Botanical Journal of Linnaean Society; 171:171-190. DÜNDAR E, Akcicek E, Dirmenci T and AKGÜN Ş (2013). Phylogenetic analysis of the genus Stachys sect. Eriostomum (Lamiaceae) in Turkey based on nuclear ribosomal ITS sequences. Turkish Journal of Botany; 37(1):14-23. Edwards CE, Lefkowitz D, Soltis DE and Soltis PS (2008a). Phylogeny of Conradina and related southeastern scrub mints (Lamiaceae) based on GapC gene sequences. International Journal of Plant Sciences; 169(4):579-594. Edwards CE, Soltis DE and Soltis PS (2006). Molecular phylogeny of Conradina and other scrub mints (Lamiaceae) from the southeastern USA: evidence for hybridization in Pleistocene refugia? Systematic Botany; 31(1):193-207. Edwards CE, Soltis DE and Soltis PS (2008b). Using patterns of genetic structure based on microsatellite loci to test hypotheses of current hybridization, ancient hybridization and incomplete lineage sorting in Conradina (Lamiaceae). Molecular Ecology; 17(23):5157-5174. Edris AE (2007). Pharmaceutical and therapeutic potential of essential oils and their individual volatile constituents: A review. Phytotherapy Research; 21:308-323. El Mokni R, Kasri M and El Aouni MH (2013). Volkameria inermis (Lamiaceae) a new alien species naturalized to the Tunisian coast, first record for North-Africa. Flora Mediterranea; 23:117-122. Epling C and Játiva C (1964). Revisión del género Satureja en América del Sur. Brittonia; 16(4):393-416. Epling C and Játiva C (1966). Supplementary notes on American Labiatae. IX. Brittonia; 18(3):255-265. Erdtman G (1945). Pollen morphology and plant taxonomy. IV. Labiatae, Verbenaceae and Avicenniaceae. Svensk Botanisk Tidskrift; 39:279-285. Esen G, Azaz AD, Kurkcuoglu Baser KH and Tinmaz A (2007). Essential oil and antimicrobial activity of wild and cultivated Origanum vulgare L. subsp. hirtum (Link) Ietswaart from the Marmara region, Turkey. Flavour and Fragrance Journal; 22:371-376. Fay MF, Bayer C, Alverson WS, de Bruijn AY and Chase MW (1998). Plastid rbcL sequence data indicate a close affinity between Diegodendron and Bixa. Taxon; 47(1):43-50. Fay MF, Swensen SM and Chase MW (1997). Taxonomic affinities of Medusagyne oppositifolia (Medusagynaceae). Kew Bulletin; 52:111-120. Fazekas AJ, Kesanakurti PR, Burgess KS, Percy DM, Graham SW, Barrett SC, Newmaster SG, Hajibabaei M and Husband BC (2009). Are plant species inherently harder to discriminate than animal species using DNA barcoding markers? Molecular Ecology Resources; 9:130-139. Ferri G, Corradini B, Ferrari F, Santunione AL, Palazzoli F and Alu M (2015). Forensic botany II, DNA barcode for land plants: which markers after the international agreement? Forensic Science International: Genetics; 15:131-136. doi: 10.1016/j.fsigen.2014.10.005. Frezal L and Leblois R (2008). Four years of DNA barcoding: current advances and prospects. Infection, Genetics & Evolution; 8:727-736. Frodin DG (1984). Guide to standard floras of the world. Cambridge University Press, Cambridge. Fu YM, Jiang WM and Fu CX (2011). Identification of species within Tetrastigma (Miq.) Planch.(Vitaceae) based on DNA barcoding techniques. Journal of Systematics and Evolution; 49(3):237-245. Gao T, Yao H, Song J, Zhu Y, Liu C and Chen S (2010). Evaluating the feasibility of using candidate DNA barcodes in discriminating species of the large Asteraceae family. BMC Evolutionary Biology; 10(1):324. Gaut BS, Muse SV, Clark WD and Clegg MT (1992). Relative rates of nucleotide substitutions at the rbcL locus of monocotyledonous plants. Journal of Molecular Evolution; 35:292-303. Gonzalez MA, Baraloto C, Engel J, Mori SA, Petronelli P, Riera B, Roger A, Thebaud C and Chave J (2009). Identification of amazonian trees with DNA barcodes. PLoS ONE; 4: e7483. Greuter W, Burdet HM and Long D (1986). Satureja. In: Med-Checklist. Conservatoire et Jardin Botaniques de la ville de Genève, Geneva; 3:323-341. Guerin G (2008). Evidence for polyphyly in Hemigenia and Microcorys (Lamiaceae: Westringieae). Australian Systematic Botany; 21(5):313-325. Han JP, Shi LC, Chen XC and Lin YL (2012). Comparison of four DNA barcodes in identifying certain medicinal plants of Lamiaceae. Journal of Systematics and Evolution; 50(3):227-234.

Harley RM, Atkin S, Budanstev AL, Cantino PD, Conn BJ, Grayer R, Harley MM, de Tok R, Krestovskaja T, Morales R, Paton AJ, Ryding O and Upson T. 2004. Flowering plants, dicotyledons. In: The families and genera of vascular plants (ed. Kubitzki K). Springer Verlag, Berlin; 6:167-275.

Harley RM and Pastore JFB. A generic revision and new combinations in the Hyptidinae (Lamiaceae), based on molecular and morphological evidence. Phytotaxa; 58:1-55.

Hebert PD, Stoeckle MY, Zemlak TS and Francis CM (2004). Identification of birds through DNA barcodes. PLoS Biol; 2(10):e312.

Hedberg O (1957). Afroalpine vascular plants: a taxonomic revision. Symbolae Botanicae Upsalienses; 15:317-318.

Hedge IC (1990). Labiatae. In: Flora of Pakistan (eds. Ali SI and Nasir YJ). 92; 1-310. Hedge IC (1992). A global survey of the biogeography of the Labiatae. In: Advances in Labiate science (eds. Harley RM and Reynolds T). Royal Botanic Gardens, Kew; 7- 17. Heinrich M (1992). Economic botany of American Labiatae. In: Advances in Labiate science (eds. Harley RM and Reynolds T). Royal Botanic Gardens, Kew; 475–488. Hennig W (1950). Grundzüge einer Theorie der phylogeneti-schen Systematik. (Deutscher Zentralverlag). Berlin; pp. 370. Hennig W (1965). Phylogenetic systematics. Annual Review of Entomology; 10:97-116. Hidalgo O, Garnatje T, Susanna A and Mathez J (2004). Phylogeny of Valerianaceae based on matK and ITS markers, with reference to matK individual polymorphism. Annals of Botany; 93:283-293. Hilu KW, Borsch T, Muller K, Soltis DE, Soltis PS, Savolainen V, Chase MW, Powell MP, Alice LA, Evans R, Sauquet H, Neinhuis C, Slotta TA, Rohwer JG, Campbell CS and Chatrou LW (2003). Angiosperm phylogeny based on matK sequence information. American Journal of Botany; 90:1758-1776. Hollingsworth ML, Clark AA, Forrest LL, Richardson J, Pennington R, Long DG, Cowan R, Chase MW, Gaudeul M and Hollingsworth PM (2009). Selecting barcoding loci for plants: Evaluation of seven candidate loci with species-level sampling in three divergent groups of land plants. Molecular Ecology Resources; 9:439-457. Hollingsworth PM, Graham SW and Little DP (2011). Choosing and using a plant DNA barcode; PLoS ONE: 6, e19254. Huang M, Crawford DJ, Freudenstein JV and Cantino PD (2008). Systematics of Trichostema (Lamiaceae): Evidence from ITS, ndhF, and Morphology. Systematic Botany; 33(2):437-446. Huang M, Freudenstein J and Crawford D (2000). Phylogenetic relationships of the Caryopteris-Trichostema complex (Lamiaceae) based on ndhF sequence data. American Journal of Botany; 87(6):174-175. Huelsenbeck JP and Ronquist F (2001). MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics; 17:754-755. Hussain AI, Anwar F, Sherazi ST and Przybylski R (2008). Chemical composition. Antioxidant and antimicrobial activities of basil (Ocimum basilicum) essential oils depends on seasonal variations. Food Chemistry. 108:986-995. Hussain AI, Anwar F, Shahid M, Ashraf M and Przybylski R (2010). Chemical composition, antioxidant and antimicrobial activities of essential oil of spearmint (Mentha spicata L.) from Pakistan. Food Chemistry; 22:78-84. Jafri SMH and Ghafoor A (1974). Verbenaceae. In: Flora of West Pakistan (eds. Nasir E and S. Ali SI). Karachi; 7:71–40.

Judd WS, Cambell CS, Kellogg EA, Stevens PF and Donoghue MJ (2002). Plant systematics: A phylogenetic approach, 2nd ed. Sinaauer Associates, Inc, USA.

Junell S 1934. Zur Gynaceummorphologie und Systematik der Verbenaceen und Labiaten. Symbolae Botanicae Upsalienses; 4:1-219.

Kane NC and Cronk Q (2008). Botany without borders: barcoding in focus. Molecular Ecology; 17:5175-5176.

Katoh K, Misawa K, Kuma K and Miyata T (2002). MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleac Acids Research; 30(14):3059-66.

Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, Buxton S, Cooper A, Markowitz S, Duran C, Thierer T, Ashton B, Mentjies P and Drummond A (2012). Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics; 28(12):1647-1649.

Killick DJB (1961). South African species of Satureja. Bothalia; 7: 435-437.

Kim KJ and Lee HL (2005). Widespread occurrence of small inversions in the chloroplast genomes of land plants. Molecules and Cells; 19:104-113.

Koorders, SH (1904). Teijsmanniodendron, eine neue Gattung der Verbenaceae im Botanischen Garten von Buitenzorg. Annales du Jardin Botanique de Buitenzorg; 19: 19-31. Kress WJ and Erickson DL (2008). DNA barcodes: genes, genomics, and bioinformatics. Proceedings of National Academy of Sciences USA; 105:2761-2762. Kress WJ, Wurdack KJ, Zimmer EA, Weigt LA and Janzen DH (2005). Use of DNA barcodes to identify flowering plants. Proceedings of the National Academy of Sciences USA; 102(23):8369-8374. Kugita M, Kaneko A, Yamamoto Y, Takeya Y, Matsumoto T and Yoshinaga K (2003). The complete nucleotide sequence of the hornwort (Anthoceros formosae) chloroplast genome: insight into the earliest land plants. Nucleic Acids Research; 31:716-721. Lahaye R, van der Bank M, Bogarin D, Warner J, Pupulin F, Gigot G, Maurin O, Duthoit S, Barraclough TG and Savolainen V (2008). DNA barcoding the floras of biodiversity hotspots. Proceedings of the National Academy of Sciences USA; 105:2923-2928. Law SK, Simmons MP, Techen N, Khan IA, He MF, Shaw PC and But PPH (2010). Molecular analyses of the Chinese herb Leigongteng (Tripterygium wilfordii Hook.f.). Phytochemistry; 72:21-26. Li B, Xu W, Tu T, Wang Z, Olmstead RG, Peng H, Francisco-Ortega J, Cantino PD and Zhang DX (2012). Phylogenetic position of Wenchengia (Lamiaceae): A taxonomically enigmatic and critically endangered genus. Taxon; 61(2):392-401. Li B, Cantino PD, Olmstead RG, Bramley GLC, Xiang CL, Ma ZH, Tan YH and Zhang DX (2016). A large-scale chloroplast phylogeny of the Lamiaceae sheds new light on its subfamilial classification. Scientific Reports; 6:34343 DOI: 10.1038/srep34343. Li M, Cao H, But PPH and Shaw PC (2011). Identification of herbal medicinal materials using DNA barcodes. Journal of Systematics & Evolution; 49:271-283.

Li XW and Hedge IC (1994). Lamiaceae. In: Flora of China, Verbenaceae through Solanaceae, (eds. Wu ZY and Raven PH). Science press Beijing and Missouri Botanical Garden St. Louis; 17:50-299.

Lin CC, Yang CC, Phua DH, Deng JF and Lu LH (2010). An outbreak of foxglove leaf poisoning. Journal of Chinese Medical Association; 73(2):97-100. Lindqvist C and Albert VA (2002). Origin of the Hawaiian endemic mints within North American Stachys (Lamiaceae). American Journal of Botany; 89(10):1709-1724. Lindqvist C, Motley TJ, Jeffrey JJ and Albert VA (2003). Cladogenesis and reticulation in the Hawaiian endemic mints (Lamiaceae). Cladistics; 19(6):480-495. Lipscomb D, Platnick L and Wheeler Q (2003). The intellectual content of taxonomy: a comment on DNA taxonomy. Trends in Ecology & Evolution; 18:65-66. Lou SK, Wong KL, Li M, But PP, Tsui SK and Shaw PC (2010). An integrated web medicinal materials DNA database: MMDBD (Medicinal Materials DNA Barcode Database). BMC Genomics; 11(1):402.

Long DG (1999). Verbenaceae. In: Flora of Bhutan (eds. Grierson AJC and Long DG). Royal Botanic Gardens, Edinburgh; 2(2):911-937.

Maddison WP and Maddison DR (2011). Mesquite: a modular system for evolutionary analysis. Version 3.03. http://mesquiteproject.org.

Manosroia J, Dhumtanoma P and Manosroia A (2006). Anti-proliferative activity of essential oil extracted from Thai medicinal plants on KB and P388 cell lines. Cancer Letters; 235:114-120.

Martins AP, Salgueiro LR, Vila R, Tomi F, Canigueral S, Casanova J, da Cunha AP and Adzet T (1999). Composition of the essential oils of Ocimum canum, O. gratissimum and O. minimum. Planta Medica; 65:187-189.

Mathiesen C, Scheen AC and Lindqvist C. Phylogeny and biogeography of the lamioid genus Phlomis (Lamiaceae). Kew Bull; 66: 83-99.

Melchior H (1964). A. Engler’s Syllabus der Pflanzenfamilien.Borntraeger, Berlin; 2:666.

Michel CI, Meyer RS, Taveras Y and Molina J (2016). The nuclear internal transcribed spacer (ITS2) as a practical plant DNA barcode for herbal medicines. Journal of Applied Research on Medicinal and Aromatic Plants; 3(3):94-100.

Miller MA, Pfeiffer W and Schwartz T (2010). Creating the CIPRES Science Gateway for inference of large phylogenetic trees. In: Proceedings of the Gateway Computing Environments Workshop (GCE), 2010, New Orleans, Louisiana, USA; 1-8.

Mishra P, Kumar A, Nagireddy A, Mani DN, Shukla AK, Tiwari R and Sundaresan V (2016). DNA barcoding: an efficient tool to overcome authentication challenges in the herbal market. Plant Biotechnology Journal; 14:8-21.

Moldenke HN (1980). A sixth summary of the Verbenacaeae, Avicenniaceae, Stilbaceae, Dicrastylidaceae, Symphoremaceae, Nyctanthaceae and Eriocaulaceae of the world as to valid taxa, geographic distribution and synonomy. Phytologia Memoirs; 2:1-629. Moldenke HN (1983). Notes on the genus Caryopteris (Verbenaceae). Phytologia; 52:415- 437.

Moon HK, Hong SP, Smets E and Huysmans S (2009). Micromorphology and Character Evolution of Nutlets in Tribe Mentheae (Nepetoideae, Lamiaceae). Systematic Botany; 34(4):760-776. Moon HK, Smets E and Huysmans S (2010). Phylogeny of tribe Mentheae (Lamiaceae): The story of molecules and micromorphological characters. Taxon; 59(4):1065-1076. Moon HK, Vinckier S, Smets E and Huysmans S (2008). Palynological evolutionary trends within the tribe Mentheae with special emphasis on subtribe Menthinae (Nepetoideae: Lamiaceae). Plant Systematics and Evolution; 275(2):93-108.

Morales R (1987). El genero Thymbra L. (Labiatae). Anales del Jardín Botánico de Madrid; 44: 349–380.

Morgan DR and Soltis DE (1993). Phylogenetic relationships among members of Saxifragaceae sensu lato based on rbcL sequence data. Annals of Missouri Botanical Garden; 80:631-660.

Muellner AN, Samuel R, Johnson SA, Cheek M, Penning-Ton TD and Chase MW (2003). Molecular phylogenetics of Meliaceae based on nuclear and plastid DNA sequences. American Journal of Botany; 90:471-480.

Newmaster SG, Grguric M, Shanmughanandhan D, Ramalingam S and Ragupathy S (2013). DNA barcoding detects contamination and substitution in North American herbal products. BMC Medicine; 11(1):222. New York City Press Office (2015). Schneiderman AG asks major retailers to haltsales of certain herbal supplements as DNA tests fail to detect plant materialslisted on majority of products tested. 3 Feb, available from http://www.ag.ny.gov/press- release/ag-schneiderman-asks-major-retailers-halt-sales-certain-herbal-supplements- dna-tests (accessed 16.01.2016). Nithaniyal S, Newmaster SG, Ragupathy S, Krishnamoorthy D, Vassou SL and Parani M (2014). DNA barcode authentication of wood samples of threatened and commercial timber trees within the tropical dry evergreen forest of India. PLoS ONE; 9:e107669. Olmstead R, Reeves P and Lepschi B (1998). Confirmation of monophyletic Chloanthoideae (Lamiaceae) comprising tribes Chloantheae and Prostanthereae. Lamiales Newsletter; 6:7-10. Olmstead RG, Depamphilis CW, Wolfe AD, Young ND, Elisons WJ and Reeves PA (2001). Disintegration of the Scrophulariaceae. American Journal of Botany; 88(2):348-361. Pagel M (1999). Inferring the historical patterns of biological evolution. Nature; 401:877- 884. Pan YZ, Fang LQ, Hao G, Cai J and Gong X (2009). Systematic positions of Lamiophlomis and Paraphlomis (Lamiaceae) based on nuclear and chloroplast sequences. Journal of Systematics and Evolution; 47(6):535-542. Pang X, Liu C, Shi L, Liu R, Liang D, Li H, Cherny Stacey S and Chen S (2012). Utility of the trnH–psbA intergenic spacer region and its combinations as plant DNA barcodes: a meta-analysis. PLoS ONE; 7:e48833. Paton AJ, Springate D, Suddee S, Otieno D, Grayer RJ, Harley MM, Willis F, Simmonds MS, Powell MP and Savolainen V (2004). Phylogeny and evolution of basils and allies (Ocimeae, Labiatae) based on three plastid DNA regions. Molecular Phylogenetics and Evolution; 31(1):277-299.

Pei C and Chen SL (1982). Verbenaceae. In: Flora Republicae Popularis Sinicae, vol 65 (1). Beijing, Science Press.

Pieper W (1928). Voraibeiten zu einer Revision der afrikanischen Vitex-Arten mit Berücksichtigung der Übrigen Bot. Jb, 62 (3), Beibl. 142: 2-80.

Pillon Y, Johansen J, Sakishima T, Chamala S, Barbazuk WB, Roalson EH, Price DK and Stacy EA (2013). Potential use of low-copy nuclear genes in DNA barcoding: a comparison with plastid genes in two Hawaiian plant radiations. BMC Evolutionary Biology; 13:1-10.

Pozhidaev AE (1989). Exine structure in pollen grains of the Lamiaceae family. Botanicheskii Zhurnal; 74:1410-1422.

Prather AL, Monfils AK, Posto AL and Williams RA (2002). Monophyly and phylogeny of Monarda (Lamiaceae): evidence from the Internal Transcribed Spacer (ITS) region of nuclear ribosomal DNA. Systematic Botany; 27:127-137. Press JR, Shrestha KK and Sutton DA (2000). Annotated checklist of the flowering plants of Nepal. The Natural History Museum, London; pp. 430.

Rajendran A and Daniel P (2002). The Indian Verbenaceae (a taxonomic revision). Bishen Singh Mahendra Pal Singh, Dehradun; pp. 431.

Rambaut A and Drummond AJ (2007). Tracer v1.4. Online program available at http://beast.bio.ed.ac.uk/Tracer.

Rambaut A (2008). FigTree v1.1.1: Tree figure drawing tool. Available at http://tree.bio.ed.ac.uk/soft ware/figtree/.

Rannala, B and Yang ZH (1996). Probability distribution of molecular evolutionary trees: A new method of phylogenetic inference. Journal of Molecular Evolution; 43:304-311.

Rimpler HC, Winterhalter and Falk U (1992). Cladistic analysis of the subfamily Caryopteridoideae Briq. and related taxa of Verbenaceae and Lamiaceae using morphological and chemical characters. In: Advances in Labiate Science (eds. Harley RM and Reynolds T). Royal Botanic Gardens, Kew; 39-54.

Rodriguez-Hahn L, Esquivel B, Cardenas J and Ramamoorthy TP (1992). The distribution of diterpenoids in Salvia. In: Advances in Labiate science (eds. Harley RM and Reynolds T). Royal Botanic Gardens, Kew; 335–347.

Roy T, Chang TH, Lan T and Lindqvist C (2013). Phylogeny and biogeography of New World Stachydeae (Lamiaceae) with emphasis on the origin and diversification of Hawaiian and South American taxa. Molecular Phylogenetics and Evolution; 69(1):218-238. Roy T and Lindqvist C (2015). New insights into evolutionary relationships within the subfamily Lamioideae (Lamiaceae) based on pentatricopeptide repeat (PPR) nuclear DNA sequences. Americn Journal of Botany; 102(10):1721-1735. Roy, T. Nathan SC, Drake MG, Cantino PD, Scheen AC and Lindqvist C (2016). Evolutionary relationships within the lamioid tribe Synandreae (Lamiaceae) based on multiple low-copy nuclear loci. PeerJ; 4:e2220; DOI 10.7717/peerj.2220 Rubinoff D, Cameron S and Will K (2006). Are plant DNA barcodes a search for the Holy Grail? Trends in Ecology & Evolution; 21:1–2. Ryding O (1994a). Pericarp structure and phylogeny of Lamiaceae subfamily Pogostemonoideae. Nordic Journal of Botany; 14(1):59-63. Ryding O (1994b). Pericarp structure in the subtribe Melittidinae (Lamiaceae-Laminoideae) and its systematic implications. Botanische Jahrbücher fur Systematik; 115:547-555. Ryding O (1995). Pericarp structure and phylogeny of the Lamiaceae-Verbenaceae-complex. Plant Systematics and Evolution; 198(1-2):101-141. Ryding O (1996). Pericarp structure and phylogenetic position of the genus Wenchengia (Lamiaceae). Botanische Jahrbucher Fur Systematik; 118:153-158. Ryding O (1998). Phylogeny of the Leucas group (Lamiaceae). Systematic Botany; 23:235- 247.

Ryding O (2003). Phylogeny of the genus Lamium and reconsideration of the genus Wiedemannia (Lamiaceae). Botanische Jahrbücher fur Systematik; 124: 325–335.

Ryding O (2008). Pericarp structure and phylogeny of the Phlomis group (Lamiaceae subfam. Lamioideae). Botanische Jahrbücher fur Systematik; 127(3):299-316. Ryding O (2010a). Crystals in calyces of Lamiaceae and their phylogenetic and adaptive significance. Plant systematics and evolution; 290(1-4):201-215. Ryding O (2010b). Pericarp structure and phylogeny of tribe Mentheae (Lamiaceae). Plant systematics and evolution; 285(3-4):165-175. Sahare P and Srinivasu T (2012). Barcoding for authentic identification of medicinal plants. International Journal of Engineering Science; 1:33-36. Salmaki Y, Zarre S, Ryding O, Lindqvist C, Brauchler C, Heubl G, Barber J and Bendiksby M (2013). Molecular phylogeny of tribe Stachydeae (Lamiaceae subfamily Lamioideae). Molecular Phylogenetics and Evolution; 69(3):535-551. Sang T (2002). Utility of low-copy nuclear gene sequences in plant phylogenetics.Critical Reviews in Biochemistry and Molecular Biology; 37:121-147. Sang T, Crawford D and Stuessy T (1997). Chloroplast DNA phylogeny, reticulate evolution, and biogeography of Paeonia (Paeoniaceae). American Journal of Botany; 84(8):1120. Scheen AC and Albert VA (2009). Molecular phylogenetics of the Leucas group (Lamioideae; Lamiaceae). Systematic Botany; 34(1):173-181. Scheen AC, Bendiksby M, Ryding O, Mathiesen C, Albert VA and Lindqvist C (2010). Molecular Phylogenetics, Character Evolution, and Suprageneric Classification of Lamioideae (Lamiaceae). Annals of the Missouri Botanical Garden; 97(2):191-217. Scheen AC, Brochmann C, Brysting AK, Elven R, Morris A, Soltis DE, Soltis PS, Albert VA (2004). Northern hemisphere biogeography of Cerastium (Caryophyllaceae): insights from phylogenetic analysis of noncoding plastid nucleotide sequences. American Journal of Botany; 91:943-952. Scheen AC, Lindqvist C, Fossdal CG and Albert VA (2008). Molecular phylogenetics of tribe Synandreae, a North American lineage of Lamioid mints (Lamiaceae). Cladistics; 24(3):299-314. Schmidt-Lebuhn A (2008). Monophyly and phylogenetic relationships of Minthostachys (Labiatae, Nepetoideae) examined using morphological and nrITS data. Plant Systematics and evolution; 270(1-2):25-38. Schmidt‐Lebuhn An (2007). Using amplified fragment length polymorphism (AFLP) to unravel species relationships and delimitations in Minthostachys (Labiatae). Botanical Journal of the Linnean Society; 153(1):9-19. Schori M, Appel M, Kitko A and Showalter AM (2013). Engineered DNA polymerase improves PCR results for plastid DNA. Applications in Plant Sciences; 1(2):1-7 Schori M and Showalter AM (2011). DNA barcoding as a means for identifying medicinal plants of Pakistan. Pakistan Jornal of Botany; 43(SI):1-4. Sen S and Chakraborty R (2015). Toward the integration and advancement of herbal medicine: a focus on traditional Indian medicine. Botanics: Targets and Therapy; 5:33-44. Shi S, Du Y, Boufford D, Gong X, Huang Y, He H and Zhong Y (2003). Phylogenetic position of Schnabelia, a genus endemic to China: Evidence from sequences of cpDNA matK gene and nrDNA ITS regions. Chinese Science Bulletin; 48(15):1576- 1580. Shinwari ZK (2010). Medicinal plants research in Pakistan. Journal of Medicinal Plants Research; 4(3):161-176.

Shinwari ZK and Chaudheri MN (1992). A taxonomic revision of genus Mentha from Pakistan. Acta Phytotaxonomica et Geobotanica; 43: 97-110. Shinwari ZK, Jamil K and Zahra NB (2014). Molecular systematics of selected genera of subfamily mimosoidae-fabaceae. Pakistan Jornal of Botany; 46(2):591-598. Simmonds MSJ and Blaney WM (1992). Labiate-insect interactions: effects of labiate- derived substances on insect behaviour. In: Advances in Labiate science (eds. Harley RM and Reynolds T). Royal Botanic Gardens, Kew; 375–392. Simon JE, Quinn J and Murray RG (1990). Basil: a source of essential oils, Advances in New Crops. Timber, Portland, OR; pp. 484–489.

Sivadas D and Sreelekha (2011). Systematics of selected members of Verbenaceae and Lamiaceae inferred from morphological studies. Journal of Economic and Taxonomic Botany; 35:396-407.

Small RL, Cronn RC and Wendel JF (2004). Use of nuclear genes for phylogeny reconstruction in plants. Australian Systematic Botany; 17:145-170.

Stamatakis A, Hoover P and Rougemont J (2008). A rapid bootstrap algorithm for the RAxML web-servers. Systematic Biology; 75: 758-771.

Steane DA, de Kok RP and Olmstead RG (2004). Phylogenetic relationships between Clerodendrum (Lamiaceae) and other Ajugoid genera inferred from nuclear and chloroplast DNA sequence data. Molecular Phylogenetics and Evolution; 32(1):39- 45. Steane DA and Mabberley DJ (1998). Rotheca (Lamiaceae) revived. Novon; 8:204-206. Steane DA, Scotland RW, Mabberley DJ, Wagstaff S, Reeves P and Olmstead RG (1997). Phylogenetic relationships of Clerodendrum s. l. (Lamiaceae) inferred from chloroplast DNA. Systematic Botany; 22:229-243. Steane DA, Scotland RW, Mabberley DJ and Olmstead RG (1999). Molecular systematics of Clerodendrum (Lamiaceae): ITS sequences and total evidence. American Journal of Botany; 86(1):98-107. Steele PR, Guisinger-Bellian M, Linder CR and Jansen RK (2008). Phylogenetic utility of 141 low-copy nuclear regions in taxa at different taxonomic levels in two distantly related families of rosids. Molecular Phylogenetics and Evolution; 48:1013-1026. Stoeckle MY, Gamble CC, Kirpekar R, Young G, Ahmed S and Little DP (2011). Commercial teas highlight plant DNA barcode identification successes and obstacles. Scientific Reports; 1: 42, doi:10.1038/srep00042. Sucher NJ and Carles MC (2008). Genome-based approaches to the authentication of medicinal plants. Planta Medica; 74(6):603-623. Taberlet P, Gielly L, Pautou G and Bouvet J (1991). Universal primers for amplification of three non-coding regions of chloroplast DNA. Plant Molecular Biology; 17:1105- 1109.

Takhtajan A (2009). Flowering plants, 2nd ed. Berlin: Springer.

Tate JA and Simpson BB (2003). Paraphyly of Tarasa (Malvaceae) and diverse origins of the polyploid species. Systematic Botany; 28(4):723-737.

Thonner F (1915). The Flowering Plants of Africa. Dulau, Co. Ltd., London.

Thorne RF (1992). Classification and geography of the flowering plants. The Botanical Review; 58(3):225-327. Trusty JL, Olmstead RG, Bogler DJ, Santos-Guerra A and Francisco-Ortega J (2004). Using molecular data to test a biogeographic connection of the Macaronesian genus Bystropogon (Lamiaceae) to the New World: a case of conflicting phylogenies. Systematic Botany; 29(3):702-715. Trusty JL, Olmstead RG, Santos-Guerra A, Sa-Fontinha S and Francisco-Ortega J (2005). Molecular phylogenetics of the Macaronesian-endemic genus Bystropogon (Lamiaceae): palaeo-islands, ecological shifts and interisland colonizations. Molecular Ecology; 14(4):1177-1189. Verma SK and Goswami GK (2014). DNA evidence: current perspective and future challenges in India. Forensic Science International; 241:183-189. Vijayan K and Tsou C (2010). DNA barcoding in plants: taxonomy in a new perspective. Current Science (Bangalore); 99(11):1530-1541. Vohra P and Khera KS (2013). DNA barcoding: current advance and future prospects-a review. Asian Journal of Biological & Life Sciences; 3:185-189. Wagstaff SJ, Hickerson L, Spangler R, Reeves PA and Olmstead RG (1998). Phylogeny in Labiatae s. l., inferred from cpDNA sequences. Plant Systematics and Evolution; 209(3-4):265-274. Wagstaff SJ and Olmstead RG (1997). Phylogeny of Labiatae and Verbenaceae inferred from rbcL sequences. Systematic Botany; 22:165-179. Wagstaff SJ, Olmstead RG and Cantino PD (1995). Parsimony analysis of cpDNA restriction site variation in subfamily Nepetoideae (Labiatae). American Journal of Botany; 82:886-892. Walker JB and Sytsma KJ (2007). Staminal evolution in the genus Salvia (Lamiaceae): molecular phylogenetic evidence for multiple origins of the staminal lever. Annals of Botany; 100(2):375-391. Walker JB, Sytsma KJ, Treutlein J and Wink M (2004). Salvia (Lamiaceae) is not monophyletic: implications for the systematics, radiation, and ecological specializations of Salvia and tribe Mentheae. American Journal of Botany; 91(7):1115-1125. Wallace LJ, Boilard SM, Eagle SH, Spall JL, Shokralla S and Hajibabaei M (2012). DNA barcodes for everyday life: Routine authentication of Natural Health Products. Food Research International; 49(1):446-452. Wazir SM, Dasti AA and Shah J (2004). Common medicinal plants of Chapursan valley, Gojal II, Gilgit-Pakistan. Journal of Research (Science); 15(1):41-43.

Westman DJ (1744). Plantae Baccatae, In: Oratorio de Telluris Habitabilis Incremento (ed. Linnaeus C). Batavorum, Leiden; 59-61.

Whitlock BA, Hale AM and Groff PA (2010) Intraspecific Inversions Pose a Challenge for the trnH-psbA Plant DNA Barcode. PLoS ONE; 5(7):e11533. doi:10.1371/journal.pone.0011533

WHO (World Health Organization) (2002). WHO traditional medicine strategy 2002-2005. Wicke S and Quandt D (2009). Universal primers for amplification of the trnK/matK region in land plants. Anales del Jardín Botánico de Madrid; 66:285-288. Wilson TC (2010). Evolution of pollination in Prostanthera Labill. (Lamiaceae). University of Sydney. Wilson TC, Henwood MJ and Conn BJ (2012). Status of the genus Wrixonia F. Muell. (Lamiaceae). Telopea; 14:1-3. Wink M (2003). Evolution of secondary metabolites from an ecological and molecular phylogenetic perspective. Phytochemistry; 64:3-19. Wink M and Kaufmann M (1996). Phylogenetic Relationships between some Members of the Subfamily Lamioideae (Family Labiatae) Inferred from Nucleotide Sequences of the rbcL Gene. Botanica Acta; 109(2):139-148.

Wu CY and Chow S (1965). Duo taxa nova Labiatarum. Acta Phytotaxonomica Sinica; 10: 249-255.

Wu CY and Li HW (1977). Wenchengia. In: Flora Reipublicae Popularis Sinicae (eds. Wu CY and Li HW). Science pess, Beijing; 65(2):96-100.

Wunderlich R (1967). Ein Vorschlag zu einer natürlichen Gliederung der Labiaten auf Grund der Pollenkörner, der Samenentwicklung und des reifen Samens. Oesterreichische Botanische Zeitschrift; 114(4-5):383-483. Xiang CL, Zhang Q, Scheen A-C, Cantino PD, Funamoto T and Peng H (2013). Molecular phylogenetics of Chelonopsis (Lamiaceae: Gomphostemmateae) as inferred from nuclear and plastid DNA and morphology. Taxon; 62(2):375-386. Yao H, Song JY, Ma XY, Liu C, Li Y, Yu HX, Han JP, Duan LS and Chen SL (2009). Identification of Dendrobium species by a candidate DNA barcode sequence: the chloroplast psbA-trnH intergenic region. Planta Medica; 75:667-669. Yao H, Song J, Liu C, Luo K, Han J, Li Y, Pang X, Xu H, Zhu Y and Xiao P (2010). Use of ITS2 region as the universal DNA barcode for plants and animals. PloS One; 5(10):e13102. Yao G, Drew BT, Yi TS, Yan HF, Yuan YM and Ge XJ (2016). Phylogenetic relationships, character evolution and biogeographic diversification of Pogostemon s.l. (Lamiaceae). Molecular Phylogenetics and Evolution; 98:184-200. Yuan YW, Liu C, Marx HE and Olmstead RG (2009). An empirical demonstration of using pentatricopeptide repeat (PPR) genes as plant phylogenetic tools: Phylogeny of Verbenaceae and the Verbena complex. Molecular Phylogenetics and Evolution; 54(1):23-35. Yuan YW, Mabberley DJ, Steane DA and Olmstead RG (2010). Further disintegration and redefinition of Clerodendrum (Lamiaceae): implications for the understanding of the evolution of an intriguing breeding strategy. Taxon; 59(1):125-133. Yu MT, Wong KL, Zong YY, Shaw PC and Che CT (2008). Identification of Swertia mussotii and its adulterant Swertia species by 5S rRNA gene spacer. Zhongguo Zhong Yao Za Zhi; 33:502-504. Zhu YJ, Chen SL, Yao H, Tan R, Song JY, Luo K and Lu J (2010). DNA barcoding the medicinal plants of the genus Paris. Acta Pharmaceutica Sinica; 45:376-382. Zoz I and Litvinenko V (1979). Division of the family Lamiaceae Juss. into natural groups. Botanicheskii Zhurnal; 64: 989-997.