Assessment of Genetic Relationship among Male and Female Fig Genotypes Using Simple Sequence Repeat ( SSR ) Markers

Fig (Ficus carica L.) is a traditional crop in Turkey and widely cultivated around the Mediterranean areas. The gynodioecious fig species is present in two sexual forms, i.e. the domesticated fig (female tree) and the caprifig (male tree). Caprifigs are crucial for high quality fig production and breeding while, the studies on assessment of genetic relationship among caprifigs is limited. The aim of this study was to determine genetic diversity among 45 caprifigs and 2 female figs collected from four provinces in Marmara and Aegean Sea Regions of Turkey using simple sequence repeat (SSR) markers. In this work, 24 SSR markers were tested, one was monomorphic and the remaining markers amplified 82 alleles. The number of polymorphic alleles per SSR marker ranged from 2 to 7. The observed heterozygosity (Ho) differed from 0.18 to 0.76 and expected heterozygosity (He) ranged between 0.24 and 0.81. The polymorphism information content (PIC) varied from 0.42 to 0.98. A UPGMA analysis based on Dice similarity matrix clustered fig genotypes into two main groups and similarly, STRUCTURE analysis placed fig genotypes into two different gene pools (K=2). Fig genotypes collected from the same region were not clustered together in a group indicating that the fig genotypes did not cluster on the basis of their collection sites. Our results demonstrated that caprifigs and female figs are not genetically distinct and they clustered together in a group. All fig genotypes had distinct SSR marker profiles suggesting that there were no synonyms or homonyms. These results revealed a high genetic variation among fig genotypes and 23 SSR markers were enough to discriminate all fig genotypes analysed in this study demonstrating that SSR marker system is suitable for genetic analysis in figs.


Introduction
Fig (Ficus carica L.) has been cultivated since ancient times in the world (Janick, 2005) and Mediterranean region where Turkey is located is the origin of common figs (Caliskan and Polat, 2008).Traditionally, figs are an important ingredient in human diets in the region (Aksoy, 1998).Currently, they can be consumed as dried or fresh and excellent sources of vitamins, minerals, fiber and antioxidant compounds with health benefits (Caliskan et al., 2012;Mawa et al., 2013).
Due to its geographical and ecological conditions, Turkey is the major fig-producing country in the world.The world fig production in 2014 was 1.137.730tons and Turkey was the leading country with the production of 300.282 t (26.3%) (FAOSTAT, 2014).According to Turkish Statistical Institute in 2015, the fig production in Turkey was mainly spread in the Aegean (76.3%),Marmara (9.6%) and Mediterranean (7.8%) regions.In addition, Turkey has a remarkable dried fig export potential with 69.3% of worldwide dried fig export (FAOSTAT, 2013).
Figs are morphologically gynodioecious but functionally dioecious species having a specific reproduction system (Stover et al., 2007;Flaishman et al., 2008).Based on pollination mechanism, there are two sexual forms of fig, i.e. the caprifig (male tree), contain both male flowers and short-styled female flowers (abortive) together so that they can be described as morphologically hermaphrodite.Therefore, caprifigs can be used only as a pollen source for pollination and they don't environmental factors.To overcome this limitation, molecular marker based analysis have been carried out using isoenzyme markers (Cabrita et al., 2001); RAPDs (Khadari et al., 1995;Dalkilic et al., 2011); AFLPs (Baraket et al., 2009); ISSR (Guasmi et al., 2006;Ikegami et al., 2009).Microsattelites or SSRs have become the choice of marker due to their high polymorphism, co-dominance and reproducibility in recent genetic diversity studies for fig cultivars (Khadari et al., 2001;2004;Giraldo et al., 2005;2008;Achtak et al., 2009;Çalışkan et al., 2012;Perez-Jiménez et al., 2012;Ganopoulos et al., 2015;Ferrara et al., 2016) However, genetic characterization of caprifig germplasms using molecular markers such as SSR and RAPD is limited to few studies (Dalkilic et al., 2011;Essid et al., 2015).Thus, the purpose of this study was to investigate genetic relationship between 45 caprifig genotypes and 2 common Turkish female fig cultivars with SSR markers.The best caprifig genotypes as pollinators of important Turkish fig cultivars such as 'Sarılop' and 'Bursa Siyahı' are needed and in this respect, caprifig collection in Uludag University can serve as germplasm resources.Therefore, this study is important to assess and characterize the genetic relationship among caprifigs in this collection.

Plant material
In this study, the genetic relationship between 47 fig genotypes was analyzed (Table 1).'Bursa Siyahı' and 'Sarılop' are the common female fig cultivars while the remaining 45 genotypes were the caprifigs collected from Bursa, Yalova, Balıkesir and Aydın provinces of Turkey (Fig. 1).Caprifig genotypes were collected to determine best pollinator genotypes for fig cultivars such as'Bursa Siyahı' and 'Sarılop'.
produce any edible fruits.Male fig plants give three types of fruit annually; profichi (in summer), mammoni (in autumn) and mamme (in winter).Profichi (the main crop) is used for pollination of main summer crop of female trees (Ferguson et al., 1990).However, female fig plants, called edible figs, have long-styled female flowers and they can produce commercial fruits.Pollination process, also called caprification in ficus species, occurs by transferring pollen from caprifigs to female flowers of edible figs by a pollinator wasp, Blastophaga psenes L. The wasp enters, oviposits and develops in caprifigs, so there has been a symbiotic relationship between Blastophaga psenes L. and caprifigs (Kjelberg et al., 1987).
Although some parthenocarpic female fig genotypes require no pollination, caprification has become a common practice in commercial fig production in order to obtain sufficient fruit yield and high quality fruit set.Various studies reported that caprified fig fruits were larger and had more phytochemicals than uncaprified figs; and also caprification increased fruit weight and yield (Rahemi and Jafari, 2008;Trad et al., 2013).Pollination is required for many important Turkish female fig cultivars such as 'Bursa Siyahı' and 'Sarılop' which are well-known in Marmara and Aegean Regions, respectively.'Bursa Siyahı' is one of the best cultivars for fruit quality and fresh consumption and 'Sarılop' is the well-known cultivar for dried There have been various studies on the characterization of fig genotypes using of morphological traits (Caliskan and Polat, 2008;Giraldo et al., 2010;Pérez-Sánchez et al., 2016).However, morphological characters can be influenced by 173

DNA Extraction
DNA extraction from young leaf samples were collected and powdered with a paint shaker after lyophilization.DNA samples were extracted from 25 mg powdered leaf samples by using a modified CTAB method (Futterer et al., 1995).The concentration of each DNA sample was measured using a Quibit fluorometer (Invitrogen, USA) and adjusted to 50 ng/ µL and stored at -80 o C until use.

SSR analysis
Twenty four SSR primer pairs previously developed by Giraldo et al. (2005) were used for SSR analysis (Table 2).Forward primers were tailed with M13 sequence (GACGTTGTAAAACGACGGCC) at the 5´ end (Schuelke, 2000).Each 20 µL PCR solution contained 1.0 U Taq DNA polymerase (Thermo Scientific, USA) with 1 × reaction buffer, 0.10 µM M13 sequence tailed forward primer, 0.20 µM reverse primer, 0.20 µM M13 primer labelled with infrared dye either 700 nm or 800 nm (Licor, USA), 0.2 mM dNTPs, and 50 ng of DNA.SSR makers were amplified with a Veriti 96 well thermal cycler (Applied Biosystems, USA) using the following thermal cycling condition: 2 min at 94 °C; 6 cycles of 40 s at 94 °C, 1 min at 60 °C (annealing temperature was reduced 1 °C after every cycle for touchdown PCR protocol), 1 min and 30 s at 72 °C; 28 cycles of 40 s at 94 °C, 1 min at 55 °C and 1 min and 30 s at 72°C; 7 cycles of 40 s at 94 °C, 1 min at 54 °C and 1 min and 30 s at 72 °C and final extension step of 10 min at 72 °C.The PCR products were separated on 7.5% polyacrylamide gels at 30 W for 2 to 3 h using a Li-COR 4300 automated sequencer system (LI-COR).
Data analysis SSR markers were scored manually as present (1) or absent (0) and a genetic similarity matrix of fig genotypes was calculated using Dice coefficient (Dice, 1945) with the NTSYSpc v2.21 program (Exeter Software, New York, NY, USA).An unweight pair group method with arithmetic averages (UPGMA) dendrogram based on the Dice similarity matrix was constructed.
Expected heterozygosity (He) and observed heterozygosity (Ho) were calculated according to the method of Levene (1949) using POPGEN32 software v.1.32(Yeh et al., 1997).The polymorphism information content (PIC) was calculated using the formula: where Pi and Pj are the frequencies of the ith and jth alleles at a locus with l allele in a population, respectively (Botstein et al., 1980).
The population structure of the fig genotypes used in this study was determined using a model based Bayesian clustering implemented in STRUCTURE v.2.3.4 (Pritchard et al., 2000).Possible Ks (where K is an assumed fixed number of subpopulations in the entire population) from 1 to 12 were examined with 5 replicates.Each replication run was conducted with a burn-in period of 100,000 steps followed by 20,000 Monte Carlo Markov Chain (MCMC) replications using an admixture model and correlated allele frequencies options.The most likely number of subpopulations (K) was determined using the method described by Evanno et al. (2005).

Results and Discussion
A total of 24 SSR markers were tested to assess genetic variation between 45 caprifigs and two female Turkish fig cultivars (Table 2).While 23 of these SSR markers were polymorphic, LMFC12 was monomorphic with a single allele.The number of polymorphic alleles per SSR marker ranged from 2 (LMFC22-1, LMFC22-2, LMFC25, LMFC27, LMFC31 and LMFC37) to 7 (LMFC30) (Fig. 2).Besides, LMFC22 was considered to be multiple loci (Table 2), therefore, each locus was scored separately because the sizes of each loci were different.Aradhya et al. (2010) reported that the number of alleles ranged from 4 (LMFC22, LMFC31 and LMF35) to 9 (LMFC30) with a mean of 4.9.Similar results were also obtained by Essid et al. (2015) with 20 caprifig accessions that the number of alleles per locus ranged from 2 (LMFC32, LMFC15, LMFC21, LMFC31, LMFC18, LMFC27, LMFC23) to 6 (LMFC30).Thus, results from present study and previous demonstrated that LMFC30 produced the highest number of alleles per locus.
Among 47 genotypes, 82 polymorphic SSR alleles were amplified, and the average number of alleles per SSR marker was 3.56 (Table 2).Ho values of each SSR marker varied from 0.18 (LMFC30) to 0.76 (LMFC31) with an average of 0.45 while He values ranged from 0.24 (LMFC31) to 0.81 (LMFC30) with an average of 0.54.Ikegami et al. (2009) found that the average He and Ho values were 0.44 and 0.44, respectively in 19 European and Asian fig varieties.In another study assessing genetic variation among caprifig cultivars, it was reported that the average He and Ho values were 0.29, 0.33, respectively (Essid et al., 2015).On the other hand, Khadari et al. (2004) reported an average He of 0.60 and average of 6.3 alleles per locus with six SSRs and 72 fig cultivars.Our results are also comparable to the results of previous studies and also confirmed that genetic diversity among fig genotypes was large.
The Ho was higher than the He for LMFC20, LMFC21, LMFC22-1, LMFC25, LMFC26, LMFC27, LMFC31 and LMFC37 while Ho was lower than He for the remaining SSR markers in this study.A heterozygote excess (Ho>He) was observed for the LMFC15, LMFC31, LMFC18, LMFC27 loci whereas a heterozygote deficiency (Ho<He) was observed in LMFC24 and MFC2 loci in a study dealing with genetic variation of caprifigs (Essid et al., 2015).According to Giraldo et al. (2008), for all SSRs except LMFC15 and LMFC21, Ho values were higher than the expected.
The polymorphism information content (PIC) value ranged from 0.42 (LMFC18) to 0.98 (LMFC22-2) with an average of 0.73 in this study.Ikegami et al. (2009) found that the highest, the lowest and the average PIC values were 0.82 (LMFC30), 0.26 (LMFC15) and 0.39, respectively.Ferrara et al. (2016) reported that, the lowest PIC observed was 0.07 for LMFC23 and LMFC27 primers while the highest was 0.91 for Frub422 SSR marker.These results suggested that the genetic variation within fig genotypes in this study is high.
The number of gene pools indicated by the STRUCTURE analysis was estimated to be K=2 with the highest ΔK value (29.86) (Fig. 3), followed by K=6 and K=3 with ΔKs of 3.81 and 1.63, respectively.According to the STRUCTURE analysis, 47 fig genotypes were divided into   (2015) detected three undistinguished accessions due to synonyms and analyzed several homonyms within 20 Tunisian caprifig genotypes using SSR markers.However, our results showed that there were no homonyms, mislabeling or synonymies.Thus, all fig genotypes analyzed in this study were unique.
In addition, there was no clear distinction among fig genotypes based on their sites of collection in the present study.Similarly, Giraldo et al. (2008); Dalkilic et al. (2011) and Essid et al. (2015) also found that genotypes from the same regions were placed in different groups.On the other hand, Aradhya et al. (2010) reported that fig accessions from Turkmenistan are genetically different from the Mediterranean and the Caucasus figs.

Conclusions
The present study proved that SSR markers are reliable technique for identifying genetic diversity of both caprifig and female fig genotypes.Our results showed a high genetic diversity among 45 caprifig genotypes and 2 female cultivars commonly grown in Turkey and these results highlighted that all fig genotypes analyzed in this study were unique.In this respect, this genetic variation could be useful for constituting a caprifig gene bank for future breeding programs and fig production via caprification.
fig production.Due to the easy propagation of fig via cuttings, fig genotypes or cultivars can be transferred from one region to another region without any data.Therefore, it is crucial to characterize the local fig germplasm and discover genetic diversity among the fig accessions for intensive fig cultivation (Flaishman et al., 2008).

Table 1 .
List of genotypes and their sites of collection

Table 2 .
Number of polymorphic alleles, observed heterozygosity (Ho), expected heterozygosity (He) and polymorphism information content (PIC) of SSR markers