Journal of Genetics (2019)98:96 Ó Indian Academy of Sciences

https://doi.org/10.1007/s12041-019-1145-7 (0123456789().,-volV)(0123456789().,-volV)

RESEARCH ARTICLE

Utilizing bioinformatics to detect genetic similarities between African bee subspecies

HOSSAM F. ABOU-SHAARA*

Department of Plant Protection, Faculty of Agriculture, Damanhour University, Damanhour 22516, Egypt *E-mail: [email protected].

Received 27 December 2018; revised 23 June 2019; accepted 30 July 2019

Abstract. Various honey bees, especially subspecies Apis mellifera, occur in Africa and are distribute across the continent. The genetic relationships and identical genetic characteristics between subspecies in Africa (African bee subspecies) have not been widely investigated using sequence analysis. On the other hand, bioinformatics are developed rapidly and have diverse applications. It is anticipated that bioinformatics can show the genetic relationships and similarities among subspecies. These points have high importance, especially subspecies with identical genetic characteristics can be infected with the same group of pathogens, which have implications on honey bee health. In this study, the mitochondrial DNA sequences of four African subspecies and Africanized bees were subjected to the analyses of base composition, phylogeny, shared gene clusters, enzymatic digestion, and open reading frames. High identical base composition was detected in the studied subspecies, and high identical results from all tests were found between A. m. scutellata and A. m. capensis followed by A. m. intermissa and A. m. monticola. The least genetic relationships were found between A. m. lamarckii and the other subspecies. This study presents insights into the genetic aspects of African bee subspecies and highlights similarity and dissimilarity aspects. Also, Africanized honey bees derived from A. m. scutellata showed high genetic similarities to other African bees, especially A. m. capensis. Additionally, specific primers to identify these subspecies were designed and tested.

Keywords. primers; mitochondrial DNA sequences; African honey bee; phylogeny; Apis mellifera L.

Introduction hybridization between A. m. lamarckii and Carniolan bees beside other subspecies from Europe (Page et al. 1981; Honey bees, Apis mellifera L., have many subspecies of Sheppard et al. 2001; Kamel et al. 2003). which some are distributed in Europe and Asia, and others in The phylogenetic analyses from previous studies support the Africa (Garnery et al. 1992; Arias and Sheppard 1996; presence of close relationships between African bees in one Oleksa and Tofilski 2015). Some of the studies support that branch (Garnery et al. 1992), and there are differences between Africa is the origin of honey bees while other studies do not African and European bees in random amplified polymorphic suppot this fact (Franck et al. 1998; Whitfield et al. 2006; DNA (Suazo et al. 1998). A few studies have been conducted in Han et al. 2012; Wallberg et al. 2014). The African honey Africa on the genetic characterization of the honey bees, e.g. bee subspecies include Apis mellifera lamarckii Cockerell, studying mtDNA polymorphisms in Moroccan and Spanish bee 1906 or commonly known as Egyptian honey bees in Egypt, populations (Garnery et al. 1995) and in subspecies from Kenya A. m. intermissa Maa, 1953 in other north African countries, (Meixner et al. 2000). Additionally, A. m. scutellata were A. m. monticola Smith, 1961 in east Africa, and A. m. introduced to South America giving rise to the problem of scutellata Lepeletier, 1836 and A. m. capensis Eschscholtz, Africanized honey bees or commonly known as killer bees 1822 in South Africa. Other subspecies exist in limited (Taylor 1977;Winston1992). Many genetic studies have con- regions, such as A. m. sudanese in Sudan (Woyke 1993) and centrated on Africanized bees (e.g. Sheppard et al. 1991, 1999). A. m. sahariensis in Morocco and south Algeria (Smith et al. Detecting the identical genetic similarities between honey bee 1991; Adjlane et al. 2016). Some of these subspecies have subspecies has high importance especially subspecies with exposed to extensive hybridization over years, e.g. identical genetic characteristics can share the same pathogens or bees are at present common in Egypt due to the be infected with the same group of pathogens, which have 96 Page 2 of 7 Hossam F. Abou-Shaara implications on honey bee health. In fact, comprehensive Biotechnology Information (https://www.ncbi.nlm.nih.gov/) studies to compare genetic characteristics and relationships (table 1). These subspecies occur in different parts of Africa, based on sequences between African honey bee subspecies are but the sequence of A. m. scutellata is to Africanized honey not common. Therefore, it is hypothesized that using bioin- bees occur in America (Gibson and Hunt 2016). formatics can greatly detect variations and similarity aspects in DNA sequence between African bee subspecies, and between Africanized bees and African bees. Sequence analysis Fortunately, the availability of many free genetic resources encourage researchers to extract more genetic information from The percentage of A, T, C and G, beside the AT-skew and target organisms. The full sequences of mitochondrial DNA of GC-skew (Perna and Kocher 1995) were calculated for each some subspecies in Africa are freely available (Hu et al. 2016; subspecies and 13 genes that was identified (table 2). Eimanifar et al. 2016, 2017a, 2017b). Therefore, this study aimed to utilize bioinformatics using various analytical tools to highlight the genetic relationships between some African honey bee subspecies and the Africanized honey bees derived from A. Phylogenetic tree m. scutellata. Also, specific primers to identify African and subspecies were designed. Two trees were constructed for the studied subspecies, the first one was based on the whole sequences while the second one was based on the sequences of the 13 genes. Sequences Methods were aligned using ClustalW and IUB as DNA weight matrix, then the maximum likelihood method as statistical Sequences of African honey bee subspecies method, the Jukes–Cantor model (Jukes and Cantor 1969), and bootstrap method to test of phylogeny were used to The mitochondrial DNA sequences of some African honey construct the two phylogenetic trees using MEGA7 (Kumar bee subspecies were downloaded from National Centre for et al. 2016).

Table 1. The information of A. mellifera subspecies used in this study.

Gene Occurrence Subspecies Size (bp) number Accession Reference

North Africa (Libya, Tunisia Morocco and Algeria) A. m. intermissa 16336 13 KM458618 Hu et al. (2016) North Africa (Egypt) A. m. lamarckii 16589 13 KY464958 Eimanifar et al. (2017a) East Africa A. m. monticola 16343 13 MF678581 Eimanifar et al. (2017b) Africanized honey bees in America A. m. scutellata 16411 13 KJ601784 Gibson and Hunt (2016) South Africa A. m. capensis 16470 13 KX870183 Eimanifar et al. (2016)

Table 2. Thirteen genes identified in sequences of the African honey bee subspecies.

Sequences (from - to)

Genes A. m. intermissa A. m. lamarckii A. m. monticola A. m. scutellata A. m. capensis

1 498–1499 513–1514 502–1503 507–1508 508–1509 2 1798–3358 1802–3362 1794–3354 1803–3368 1803–3363 3 3620–4297 3887–4564 3618–4295 3645–4320 3693–4368 4 4446–4604 4713–4871 4444–4602 4471–4629 4520–4678 5 4586–5266 4853–5533 4584–5264 4611–5291 4660–5340 6 5287–6066 5554–6333 5285–6064 5312–6091 5362–6141 7 6187–6540 6481–6834 6185–6538 6253–6606 6302–6655 8 Complement Complement Complement Complement Complement (6894–8558) (7196–8860) (6892–556) (6960–8624) (7009–8673) 9 Complement Complement Complement Complement Complement (8646–9983) (8949–10259) (8644–9987) (8712–10019) (8761–10071) 10 Complement Complement Complement Complement Complement (9989–10252) (10292–10555) (9991–10254) (10048–10311) (10104–10367) 11 10439–10942 10744–11247 10441–10944 10499–11002 10555–11058 12 11002–12153 11306–12457 11004–12155 11062–12213 11118–12269 13 Complement Complement Complement Complement Complement (12300–13217) (12606–13523) (12302–13219) (12364–13281) (12419–13336) Genetic similarities between African bee subspecies Page 3 of 7 96

Shared gene cluster Table 3. Components of sequences of the studied African bees.

The OrthoVenn as a platform for comparison of orthologous AT- GC- gene clusters (http://www.bioinfogenome.net/OrthoVenn/) Subspecies A% T% G% C% skew skew utilizing proteins downloaded from Uniprot was used to A. m. 43.17 41.41 5.65 9.75 0.02 -0.26 identify the shared gene clusters between studied subspecies. intermissa A. m. lamarckii 43.33 41.55 5.56 9.53 0.02 -0.26 A. m. monticola 43.11 41.60 5.58 9.60 0.01 -0.26 A. m. scutellata 43.24 41.45 5.61 9.68 0.02 -0.26 Enzymatic digestion A. m. capensis 43.24 41.51 5.61 9.61 0.02 -0.26

A method by Abou-Shaara (2019) was used to compare the fragments resulted from the enzymatic digestion. The Table 4. Components of sequences of the 13 genes of studied sequences of studied subspecies were digested using African bees. available restriction enzymes, namely BamHi, BclI, BglI, AT- GC- BglII, EagI, EcoRI, EcoRV, HindIII, HpaI, NdeI, NheI, Subspecies A% T% G% C% skew skew NotI, PstI, SacI, SalI,SamI, SphI, XbaI and XhoI at Gen- ome Compiler 2.2.88 (http://www.genomecompiler.com) A. m. 42.04 40.95 6.40 10.59 0.01 -0.24 and then fragments were simulated on gel using NEB intermissa 100-bp ladder. A. m. lamarckii 42.07 41.08 6.41 10.42 0.01 -0.23 A. m. 42.05 41.24 6.31 10.38 0.009 -0.24 monticola A. m. 41.99 41.01 6.41 10.56 0.01 -0.24 Open reading frames (ORFs) scutellata A. m. capensis 41.98 41.07 6.42 10.51 0.01 -0.24 The default options of Genome Compiler 2.2.88 were used to identify ORFs for the studied African bees with start codon of ATG, minimum length of amino acids of 60, and (table 4). Also for these genes, base A had the highest per- considering all frame types. The similarity between identi- centage followed by T, C and finally G. The difference fied ORFs of studied subspecies was then highlighted. between A and T was 1.09, 0.99, 0.81 and 0.98%, and between C and G was 4.19, 4.01, 4.07, and 4.09% for A. m. intermissa, A. m. lamarckii, A. m. monticola, A. m. scutellata Primers to discriminate between subspecies and A. m. capensis, respectively. Moreover, AT-skew and GC-skew support the high similarity between the subspecies Specific primers were designed first using Genome Compiler without essential differences. 2.2.88, utilizing the option of auto primer design to dis- criminate between the subspecies. Further, the designed Phylogenetic tree primers were tested using Primer-BLAST of NCBI against all the studied subspecies to ensure specificity of the The phylogenetic tree (figure 1) highlights the close rela- designed primers to target subspecies (i.e. bands given only tionship between A. m. intermissa and A. m. monticola, and with target sequence). between A. m. scutellata and A. m. capensis. On the other

Results and discussion

Sequence analysis

The mitogenomes of the studied African bees have high percentage of nucleotide A followed by T, C and finally G (table 3). The difference between A and T was 1.76, 1.78, 1.51, 1.79, and 1.73%, and between C and G was 4.1, 3.97, 4.02, 4.07, and 4% for A. m. intermissa, A. m. lamarckii, A. m. monticola, A. m. scutellata and A. m. capensis, respectively. In general, high similarity was detected in the base composition of the mitogenomes, which is also sup- ported by AT-skew and GC-skew. Another high similarity was found between studied bees Figure 1. The phylogenetic tree based on the whole sequences of with regard to base composition of the identified 13 genes the studied African bees using the maximum likelihood method. 96 Page 4 of 7 Hossam F. Abou-Shaara

Figure 2. The phylogenic tree based on the 13 genes of studied African bees using the maximum likelihood method. side, A. m. lamarckii was placed away from the other sub- species. Also, the 13 genes (figure 2) support the relation- ship detected based on the whole sequences. The differentiation of the Egyptian honey bees from those of other African areas was suggested based samples collected from honey bee colonies (Franck et al. 2001), and based on 13 protein-coding genes (PCG) (Eimanifar et al. 2017a). This differentiation can support the present result, especially A. m. lamarckii which was located away from the other African bees. The trees presented in Han et al. (2012) sup- Figure 3. Shared gene clusters between the studied African bees. port the high relationships between African subspecies used in this study, without supporting any close relationships between any two subspecies over others. It is clear that phylogenetic relationships obtained using the maximum Africanized honey bees derived from A. m. scutellata have likelihood method. high genetic relationships with the other African bees, especially A. m. capensis although the hybridization with European bees in America. This close relationships between ORFs these two subspecies is also supported by the phylogenetic tree constructed based on 13 PCG (Eimanifar et al. The detected ORFs (A and D) were identical among A. m. 2016, 2017a). scutellata, A. m. capensis and A. m. monticola while A. m. intermissa had extra ORFs (E) (figure 5). On the other side, A. m. lamarckii shared only ORFs (A) with the other sub- Shared gene cluster species and possessed unique ORFs (B and C). The results of this part are supported by the phylogenetic analysis All the subspecies shared the same 866 gene cluster families especially with regard to the close relationship among A. m. (figure 3). This confirms the high similarity between African scutellata, A. m. capensis and the separation of A. m. subspecies based on the available database. This result is in lamarckii than the other subspecies. In general, the detected line with the analysis of base composition. number of ORFs using the default options of Genome Compiler 2.2.88 was low in all the subspecies (figure 5).

Enzymatic digestion Primers to discriminate between subspecies Fragments with different sizes to the studied subspecies are resulted from the enzymatic digestion (figure 4). Identical Primers to identify each subspecies were identified (table 5). fragments were obtained to A. m. intermissa and A. m. These primers can give bands only with the target subspecies monticola except in one fragment, and between A. m. according to Primer-BLAST of NCBI. The product length of scutellata and A. m. capensis except in five fragments while these primers were 191, 140, 144, 112 and 121 nt for A. m. A. m. lamarckii differed than the other subspecies in four intermissa, A. m. lamarckii, A. m. monticola, A. m. scutellata fragments. The results of the enzymatic digestion support the and A. m. capensis, respectively. Genetic similarities between African bee subspecies Page 5 of 7 96

Figure 4. Fragments of the studied African bees simulated on gel using NEB 100-bp ladder by Genome Compiler 2.2.88.

Figure 5. ORFs for the studied African bees using default options of Genome Compiler 2.2.88. 96 Page 6 of 7 Hossam F. Abou-Shaara

Table 5. Specific primers to identify the African honey bee subspecies.

Species Type Length Sequences

A. m. intermissa Forward 50 23 16146 to 16168 TTTAAAAAATAAAAACTTAAAAA Reverse 50 23 16336 to 16314 AATATTTTTAAATAAGGTATTGA A. m. monticola Forward 50 23 16200 to 16222 TTTTAAAAAAATAAAAACTTAAA Reverse 50 23 16343 to 16321 AATTAAGTAAAATATTTTTAAAT A. m. capensis Forward 50 22 16350 to 16371 TTAAAAAATAAATAAAATAAAT Reverse 50 22 16470 to 16449 AAATTAAGTAATATTTTTAAAT A. m. lamarckii Forward 50 21 16450 to 16470 ACTAAGAATAAAAGTTTTTTT Reverse 50 22 16589 to 16568 AATTAAGTAAATATTTTAAAAT A. m. scutellata Forward 50 23 16300 to 16322 AAAAATAAATAAAATAAAAAAAA Reverse 50 22 16411 to 16390 AAATTAAGTAATATTTTTAAAT

In conclusion, this study highlighted the similarity aspects Eimanifar A., Kimball R. T., Braun E. L. and Ellis J. D. 2016 The between some African honey bee subspecies utilizing complete mitochondrial genome of the , Apis bioinformatics. Africanized honey bees derived from Afri- mellifera capensis Esch. (Insecta: : ). Mito- chondrial DNA B. 1, 817–819. can subspecies A. m. scutellata showed high identical Eimanifar A., Kimball R. T., Braun E. L., Moustafa D. M., Haddad genetic aspects and close phylogenetic relationship to A. m. N., Fuchs S. et al. 2017a The complete mitochondrial genome of capensis while those of north and east Africa, A. m. inter- the Egyptian honey bee, Apis mellifera lamarckii (Insecta: missa and A. m. monticola showed less identical degree Hymenoptera: Apidae). Mitochondrial DNA B. 2, 270–272. based on the analytical tools used. Bees from north Africa, Eimanifar A., Kimball R. T., Braun E. L., Fuchs S., Grunewald B. and Ellis J. D. 2017b The complete mitochondrial genome of an A. m. intermissa and A. m. lamarckii showed very few east African honey bee, Apis mellifera monticola Smith (Insecta: similarity aspects and no close relationships. In general, A. Hymenoptera: Apidae). Mitochondrial DNA B. 2, 589–590. m. lamarckii showed no close relationships with all the Franck P., Garnery L., Solignac M. and Cornuet J. M. 1998 The studied subspecies. Some pathogens have been detected in origin of west European subspecies of honeybees (Apis mellif- African bees, for examples Nosema ceranae and some era): new insights from microsatellite and mitochondrial data. Evolution 52, 1119–1134. viruses in A. m. intermissa (Higes et al. 2009; Adjlane et al. Franck P., Garnery L., Loiseau A., Oldroyd B. P., Hepburn H. R., 2016) and black queen cell virus and virus Solignac M. et al. 2001 Genetic diversity of the honeybee in 1inA. m. scutellata (Strauss et al. 2013). It is expected that Africa: microsatellite and mitochondrial data. Heredity 86, A. m. scutellata, A. m. capensis as well as A. m. intermissa 420–430. and A. m. monticola can be infected with the same strains or Garnery L., Cornuet J. M. and Solignac M. 1992 Evolutionary history of the honey bee Apis mellifera inferred from mitochon- types of pathogens due to the presence of identical genetic drial DNA analysis. Mol. Ecol. 1, 145–154. characteristics unlike A. m. lamarckii, a point requires further Garnery L., Mosshine E. H., Oldroyd B. P. and Cornuet J. M. 1995 investigations. Additionally, specific primers to discriminate Mitochondrial DNA variation in Moroccan and Spanish honey between African subspecies were designed and tested. bee populations. Mol. Ecol. 4, 465–472. Gibson J. D. and Hunt G. J. 2016 The complete mitochondrial genome of the invasive Africanized Honey Bee, Apis mellifera scutellata (Insecta: Hymenoptera: Apidae) Mitochondrial DNA A. 27, 561–562. References Han F., Wallberg A. and Webster M. T. 2012 From where did the Western honeybee (Apis mellifera) originate?. Ecol. Evol. 2, Abou-Shaara H. 2019 Highlights on the genetic relationships 1949–1957. between some honey bee viruses using various techniques. J. Higes M., Martı´n-Herna´ndez R., Garrido-Bailo´n E., Botı´as C. and Appl. Biotechnol. Rep. 6, 15–19. Meana A. 2009 The presence of Nosema ceranae (Mi- Adjlane N., Dainat B., Gauthier L. and Dietemann V. 2016 Atypical crosporidia) in North African honey bees (Apis mellifera viral and parasitic pattern in Algerian honey bee subspecies Apis intermissa). J. Apic. Res. 48, 217–219. mellifera intermissa and A. m. sahariensis. Apidologie 47, 631–641. Hu P., Lu Z. X., Haddad N., Adjlane N., Loucif-Ayad W., Wang Y. Arias M. C. and Sheppard W. S. 1996 Molecular phylogenetics of Z. et al. 2016 Complete mitochondrial genome of the Algerian honey bee subspecies (Apis mellifera L.) inferred from mito- honeybee, Apis mellifera intermissa (Hymenoptera: Apidae). chondrial DNA sequence. Mol. Phylogenet. Evol. 5, 557–566. Mitochondrial DNA A. 27, 1791–1792. Genetic similarities between African bee subspecies Page 7 of 7 96

Jukes T. H. and Cantor C. R. 1969 Evolution of protein molecules. Sheppard W. S., Shoukry A. and Kamel S. 2001 The Nile honey bee- In Mammalian protein metabolism (ed. H. N. Munro), the bee of ancient Egypt in modern times. Am. Bee J. 141, 260–263. pp. 21–132, Academic Press, New York. Smith D. R., Palopoli M. F., Taylor B. R., Garnery L., Cornuet J. Kamel S. M., Strange J. P. and Sheppard W. S. 2003 A scientific M., Solignac M. et al. 1991 Geographical overlap of two note on hygienic behavior in Apis mellifera lamarckii and A. m. mitochondrial genomes in Spanish honeybees (Apis mellifera carnica in Egypt. Apidologie 34, 189–190. iberica). J. Hered. 82, 96–100. Kumar S., Stecher G. and Tamura K. 2016 MEGA7: molecular Strauss U., Human H., Gauthier L., Crewe R. M., Dietemann V. and evolutionary genetics analysis version 7.0 for bigger datasets. Pirk C. W. 2013 Seasonal prevalence of pathogens and parasites Mol. Biol. Evol. 33, 1870–1874. in the savannah honeybee (Apis mellifera scutellata). J. Inver- Meixner M. D., Arias M. C. and Sheppard W. S. 2000 Mitochon- tebr. Pathol. 114, 45–52. drial DNA polymorphisms in honey bee subspecies from Kenya. Suazo A., McTiernan R. and Hall H. G. 1998 Differences between Apidologie 31, 181–190. African and European honey bees (Apis mellifera L.) in random Oleksa A. and Tofilski A. 2015 Wing geometric morphometrics and amplified polymorphic DNA (RAPD). J. Hered. 89, 32–36. microsatellite analysis provide similar discrimination of honey Taylor Jr, O. R. 1977 The past and possible future spread of bee subspecies. Apidologie 46, 49–60. Africanized honeybees in the Americas. Bee World 58, 19–30. Page R. E., Ibrahim M. M. and Laidlaw H. H. 1981 The history Wallberg A., Han F., Wellhagen G., Dahle B., Kawata M., Haddad of modern in Egypt. Glean. Bee Cult. 109, N. et al. 2014 A worldwide survey of genome sequence variation 24–26. provides insight into the evolutionary history of the honeybee Perna N. T. and Kocher T. D. 1995 Patterns of nucleotide Apis mellifera. Nat. Genet. 46, 1081. composition at fourfold degenerate sites of mitochondrial Whitfield C. W., Behura S. K., Berlocher S. H., Clark A. G., genomes. J. Mol. Evol. 41, 353–358. Johnston J. S., Sheppard W. S. et al. 2006 Thrice out of Africa: Sheppard W. S., Soares A. E. E., DeJong D. and Shimanuki H. 1991 ancient and recent expansions of the honey bee, Apis mellifera. Hybrid status of honey bee populations near the historic origin of Science 314, 642–645. Africanization in Brazil. Apidologie 22, 643–652. Winston M. L. 1992 The biology and management of Africanized Sheppard W. S., Rinderer T. E., Garnery L. and Shimanuki H. 1999 honey bees. Annu. Rev. Entomol. 37, 173–193. Analysis of Africanized honey bee mitochondrial DNA reveals Woyke J. 1993 Some behavioural characteristics of the Sudanese further diversity of origin. Genet. Mol. Biol. 22, 73–75. honey bee. Bee World 74, 133–140.

Corresponding editor: N. G. PRASAD