Background
Tilapias are members of the Cichlids, a large, diverse group of fishes including the perches and wrasses. Although tilapias are native to Africa, they have been introduced throughout the world for aquaculture purposes. Tilapias are tolerant of stressful water conditions and are able to flourish in a variety of freshwater and brackish water environments. They are opportunistic feeders and can feed on a wide range of both plant and animal natural food sources. Besides they can be cultivated in high density in ponds, cages, or tanks. The Nile tilapia (O. niloticus) is the most extensively cultured species among all tilapia (FAO, 2013). However due to the short life cycle and prolific breeding that result in uneven sized overcrowded population, most hatchery operators and farmers intend to culture monosex (male) Nile tilapia in commercial aquaculture practice because this system benefit the stakeholders to control unwanted reproduction by females of Nile tilapia, to obtain larger and even sized males and to gain profit.
 
All male population of Nile tilapia can be produced by several ways. Today the most practiced method is hormone feeding in diets. Administrating androgen 17-α-methyltestosterone (MT) is considered to be the most effective method (Guerrero and Guerrero, 1988). Use of hormones in direct food chain is said to have negative impacts on public health and environment. In many countries, hormonal sex reversal is therefore unacceptable for this reason. Direct observation of external morphology (particularly male and female genital papillae) is cumbersome and time consuming. An alternative way could be production of supermales (YY males) in indirect method and rear them up to the sexual maturity to cross with the normal females (XX) so that the offspring is normal males (XY). This system is also known as the production of genetically male tilapia (GMT). Production of GMT is environment friendly and acceptable method of producing monosex Nile tilapia.

The sex-determination system of Nile tilapia is complex. Interspecies hybridization, gynogenesis and progeny testing following sex inversions by hormone treatments, demonstrated that O. niloticus has an XX/XY chromosome sex determination similar to mammals (Jalabert et al., 1974; Müller-Belecke and Hörstgen- Schwark, 1995). A variety of evidence suggests that sex determination is principally monofactorial in tilapias (Wohlfarth and Wedekind 1991, Mair et al., 1995, 1997) but other autosomal influences (Hussain et al., 1998; Shirak et al., 2006; Cnanni et al., 2008), environmental effects particularly temperature (Baroiller et al., 2009) and polyfactorial mechanism including multiple allelism (Penman and Piferrer, 2008, Khan, 2011; Palaiokostas et al., 2013) and also override the simple chromosomal system to establish the sex. Sex-linked markers could play potential role to locate the QTL for sex as the genetic map of this species is available (Lee et al., 2005). One such marker, UNH995, a microsatellite, was used in the present study to analyze the suitability for the identification of polymorphocity and allelic discrimination with polyacrylamide gel electrophoresis (PAGE) in three different families (HRT23, HRT26 and HTZF824) of Nile tilapia.

1. Results
The present study was performed to observe any sex linkage and ability to identify variation in the Nile tilapia using microsatellite DNA marker UNH995. The results of the present study are given below in terms of ‘polymorphism’ and/or ‘monomorphism’ and patterns of allelic inheritance, if polymorphism existed.

1.1 Detection of polymorphism or monomorphism following PAGE
UNH995 was found to be polymorphic in the present research following PAGE in progeny (N=44) of family HRT23, HRT26 and HTZF824 (Fig. 1, Fig. 2 and Fig. 3 respectively).

1.2 Observation of allelic inheritance in both sexes of Nile tilapia in different families
PAGE resulted in marker homozygosity in the Dam of Family HRT23 (176/176) and in 4 out of 18 female progeny in locus 1. The rest female individuals were heterozygous for 170/176. Among male progeny 5 and 15 individuals expressed similar nature of allele inheritance respectively. Interestingly no putative YY males showed heterozygosity (Fig. 1). Null alleles were hypothesized in locus 2 for all progeny with alternative occurrence of 209 bp and 231 bp. A summary of inheritance of different alleles with UNH995 in family HRT23 is given in Table1. It depicts equal frequency of Null/209 in both female and male progeny (5/18) but unequal distribution of Null/231 in both sexes (females 5/18, whereas males 3/18) with none in the putative YY males (χ2: P<0.05).

In family HRT26, marker homozygosity was observed in the Dam, 7 female progeny (out of 18) and 12 male progeny (out of 18) along with 6 (out of 8) putative YY males in locus 1. The gradual decline of heterozygosity in locus 1 was evident from female progeny (11/18) through the male progeny (6/18) to the putative YY males (2/8). Null alleles were presumed in locus 2 as in family HRT23. The distribution of the four types of genotypes in locus 2 was heterogeneous and the highest frequencies of Null/209 bp were observed in male progeny (13/18) and putative YY males (5/8). A summary of inheritance of different alleles with UNH995 in family HRT26 is given in Table 2.

In family HTZF824, two genotypes (i.e., Null/136 and 136/142) were equally distributed in female progeny (7 for each genotype) and in male progeny (8 for each genotype) in locus 1, but individuals of former genotype was found to be absent in putative YY males with highest frequency (P<0.05) of the latter type. In locus 2, double null and single null alleles were assumed for the dam and the sire respectively resulting in almost equal distribution of both types of genotypes in female and male progeny but with higher frequency of Null/209 in putative YY males. A summary of inheritance of different alleles with UNH995 in family HTZF824 is given in Table 3.

2. Discussion
Due to the advancement of molecular techniques, large numbers of highly informative DNA markers have been developed for the identification of genetic polymorphism. In the last decade, the microsatellite DNA marker technique based on the polymerase chain reaction (PCR) has been one of the most commonly used molecular techniques for DNA profiling. Microsatellites, having many tandem repeats have great potential utility as genetic tags for use in aquaculture and fisheries biology. Recently microsatellite DNA marker has been extensively used as a tool in sex linkage study of many species. Such studies are helpful to find out informative markers that can increase the production of genetically improved male tilapia. In the present study several families of Nile tilapia population were studied to find sex linkage with one sex-linked marker, UNH995.
 

It has been observed recently that the sex determining mechanism of Nile tilapia is not a simple XX-XY system (Khan, 2011). Many approaches such as sex reversal and progeny testing, interspecific hybridization and molecular studies have been performed for clarifying the basic sex determining system in tilapia and to understand what factors affect sex determination mechanism. Apart from genetic factors many authors suggested low level of environmental or other autosomal or polyfactorial mechanism behind sex ratios in the progeny (Penman and Piferrer, 2008).

The current study was done to identify suitability of a sex linked marker to understand inheritance pattern which could play significant role in differentiating normal males and normal females. Inheritances of various markers from different linkage groups were studied by different scientists. LG3 markers were found to be useful in study of Oreochromis aurius and O. karongae (Lee et al., 2004, 2005; Cnaani et al., 2008). LG23 markers were important in segregation study of Oreochromis aurius x O. mossambicus (Shirak et al., 2006). Khan et al., (2014) found that LG1 markers were suitable in sex linkage study of O. niloticus. Lee et al., (2003) demonstrated the existence of a sex-determining region on LG1, using microsatellite markers, which were mapped to an interval of 10 cM. Eshel et al. (2010, 2012) found that markers in LG23 showed the highest association with phenotypic sex in a cross in a population of O. niloticus derived from Lake Manzala in Egypt. Some authors suggest multiple allelism in a single locus could play role in sex determination of Nile tilapia.

In family HRT23 of the present study PAGE resulted in marker homozygosis in the Dam along with 4 (out of 18) female progeny and 5 (out of 20) male progeny in locus 1 expressing approximately one-fourth of the genotypes. The rest (75%) individuals were heterozygous for 170/176. Interestingly no putative YY males showed heterozygosis. In locus 2, null alleles were hypothesized to explain the heritable genotypes for the progeny. The equal frequency of Null/209 in both female and male progeny (5/18) shows ‘ambivalent nature’ that may have skewness towards males or females equally, an explanation previously obtained in some studies (Khan, 2011). The unequal distribution of Null/231 in both sexes (females 5/18, whereas males 3/18) with none in the putative YY males (χ2: P<0.05) depicts that 231 bp is inclined to be inherited for ‘femaleness’.

In family HRT26, the gradual decline of heterozygosity in locus 1 was evident from female progeny (11/18) through the male progeny (6/18) to the putative YY males (2/8) and seems males were skewed for 176 bp. The highest frequencies (>60% of all possible genotypes expressed) of Null/209 bp were observed in male progeny (13/18) and putative YY males (5/8) compared to less than 30% occurrence in females (5/18). Allele 209 could be responsible for skewing the population towards maleness in contrast to allele 231 which has greater distribution in females (χ2: P<0.05).

In family HTZF824, equally distributed two genotypes (i.e., Null/136 and 136/142) in females and in males in locus 1 can be compared with those in family HRT23. Individuals of Null/136 was found to be absent in putative YY males with highest frequency (P<0.05) of the latter type. However it does not support the hypothesis that 142 bp was responsible for the maleness since in female progeny it was expressed in sufficient quantity.

It is desired to identify the tightly linked markers to the expected QTL (in this case, sex), and two tightly linked markers in the flanking region (one on each side of the QTL) will tremendously help to go for marker-assisted selection. The present study dealt with only one marker (in absence of other such markers) which is a limitation of the study. Two tightly linked markers can deliberately be used to separate ‘true’ supermales (YY) in the first generation offspring as segregation of ‘strong’ dominant homozygous alleles are hypothesized to form YY individuals.

Future work can be performed based on the observation on the departures from the sex ratios predicted by using a “simple” XX/XY model. The allelic variation in this XX/XY model demonstrates some alleles could be stronger in effect (producing close to all male) while some others are weaker giving intermediate sex ratios in the progeny. The allele giving intermediate sex ratio can also be of scientific interest as it may explain, the failure of an association of phenotypic sex with any region of the genome in one out of three families in the study of Lee et al., (2003) (e.g. if the female parent was XX and the male AA, progeny would all be XA and expected to have a sex ratio of approximately 1:1). Likewise, this could explain the single fully inbred clonal line with a sex ratio close to 1:1 in the study of Sarder et al. (1999) (if the clonal line was AA). Future research can be conducted based on the observation on allelic variations (stronger and weaker) which could be a stronger tool in producing all male population. In addition, the interaction between these alleles and interactions with other QTL in the genome should be fine mapped with more markers to discover any other factors (other than multiple alleles) for sex determination in this species.

3. Materials and methods
3.1 Study period and sites
The sites of the experiment were Fish Field Laboratory Complex, Bangladesh Agricultural University (BAU), Aquarium facilities of the Faculty of Fisheries, BAU and Fish Genetics and Biotechnology Laboratory, Department of Fisheries Biology and Genetics, BAU, Mymensingh, Bangladesh. The study was conducted for one year since May 2014 till April 2015.

3.2 Sample collection
The samples used in this study were previously collected and experiments can briefly be mentioned as follows: One day old spawns of first feeding stage of Nile tilapia (O. niloticus) were collected from Agro-3 fish farm and hatchery, situated at Boilor of Trishal upazilla, in the district of Mymensingh. Initially a mixture of fingerlings (N=500) of Nile tilapia containing normal males (XY) and normal females (XX) were collected. They were stocked in three different ponds of Field Laboratory Complex, BAU having an area of three decimals each.

3.3 Feeding and rearing 
The fingerlings were fed daily at the rate of 4% body weight four times a day up to thirty days. Crumble feed (Nursery, Saudi Bangla Fish Feed Ltd) was used to feed that contained 35% protein, 5% lipid, 5% carbohydrate, 11% ash and 1% minerals. For another ninety days they were reared up to sex determination stage using pellet feed (Starter 2, Saudi Bangla Fish Feed Ltd) containing, 26% protein, 6% lipid, 5% carbohydrate, 20% ash, 11% moisture and 1.8% calcium.

3.4 Identification of sex after sexual maturity
The sex of the experimental species O. niloticus was identified on the basis of visual observation of the genital papilla. Males were identified by observing elongated genital papilla and/or presence of milt while females were identified by round genital papilla and/or presence of egg by applyiny gentle pressure on the abdomen.

3.5 Breeding and sex-reversal
For the purpose of breeding 15 pairs of male and female were selected after six months and kept in individual hapas with 1:1 sex ratio. Among them offspring from nine pairs of parents were collected after accomplishment of breeding. The hatchlings were then fed DES hormone with the feed (100 mg DES hormone /kg feed). The hormone treatment was conducted for a period of one month. After the completion of the hormonal trial the fry were released in cisterns. Normal feed was provided to grow them up to sexual maturity. After sexual maturity, 3 females were identified as ‘XY’ females. These females along with their progeny were the experimental item of the present study.

3.6 Collection of DNA sample
Fin clips were collected from family HRT23, HRT26 and HTZF824 females and the sires. PIT tagging was done in putative YY supermales (in a Philippine family) and fry of family HRT23, HRT 26 and HTZF 824 that included both types of males (including XY and putative YY). The samples were preserved in separate eppendorfs containing 95% ethanol. Before taking each sample, all scientific procedures were rigorously maintained.
 
3.7 DNA extraction and quantification
Genomic DNA was extracted from fin clip tissues according to the method described by Islam and Alam (2004). Phenol-chloroform protocol is used for this extraction. Quantification was done by Eppendorf Biophotometer.
 
3.8 PCR amplification
PCR was performed in a 12 μl reaction volume containing 50 ng template DNA, 2.0 μM of each primer, 0.25 mM each of the dNTPs, 1 unit of Taq DNA polymerase, 1.5 mM MgCl2 and1μl 10X reaction buffer. The thermocycler condition consisted of 3 min initial denaturation at 94ºC followed by 35 cycles of 30 sec at 94ºC (denaturation), 30 sec at the respective annealing temperature, 1 min at 72ºC (extension) and ending with 5 min at 72ºC (final extension). When the PCR was completed, the PCR products were kept in a freeze (4ºC) for electrophoresis.

3.9 Polyacrylamide gel electrophoresis (PAGE) for microsatellite marker analysis
3.9.1 Glass plate preparation
The glass plate was washed with detergent followed by methanol. The plate was air-dried and then smeared with a solution containing 950 μl 95% ethanol, 5 μl 0.5% acetic acid and 3 μl silane (γ Methacryloxypropyl Trimethoxy Silane) and wiped out with tissue paper and kept in air. After 3 min, the glass plate was sprayed with 95% ethanol and wiped carefully with tissue paper to remove excess silane. The same procedure was repeated three times. At the same time the vertical gel apparatus was washed with deionized water and wiped out with tissue paper and air-dried. Then the vertical gel apparatus was smeared with water repellent (Clear View, USA) and wiped after 2 min.

3.9.2 Polyacrylamide (6%) gel preparation
For preparation of 6% polyacrylamide, 25.24 g urea (5M), 9 ml of 40% acrylamide: bisacrylamide (19:1) and 12 ml 5×TBE buffer were taken in 100 ml beaker and deionized water was added to make the solution approximately 60 ml. The solution was stirred for few min with magnetic stirrer until the urea was no longer visible. Finally, the total volume of the solution was made to 60 ml by adding deionized water. The gel chamber (38×30-cm Sequi-Gen GT sequencing gel electrophoresis system BIORAD) was set horizontally and leveled properly with a magnetic leveler. The acrylamide gel solution was taken in 60ml injection syringe just after addition of 420 μl of 10% APS (Ammonium persulphate) and 84 μl of TEMED and poured into the gel chamber. A clear level of gel at the sample loading edge was maintained by placing the comb directing opposite to the teeth to the gel. A gel with uniform 0.4 mm thickness was used for separation of single microsatellite loci. The gel was kept for 30 min for solidification. Then the upper side of the glass was cleaned with scissors and deionized water to have a clear level for sample loading.

3.9.3 Electrophoresis of PCR products in polyacrylamide gel
The gel was pre-run for 30 min at 120 W to raise the temperature up to 50oC before sample loading. After preheating, the air bubbles were removed carefully with micropipette. Meanwhile, PCR products and 4 μl 100 bp DNA ladder were preheated at 95oC for 5 min. The preheated PCR-products were immediately kept on ice and spin for few seconds in a microcentrifuge machine. The preheated PCR-products were loaded immediately between the teeth of the comb and the gel was run with the power set at 60 W and temperature set at 50oC for required length of time (1 h 30 min) according to the size of the DNA fragment. After completion of electrophoresis, the gel was stained with silver nitrate following Promega silver staining protocols.

3.10 Scoring and statistical analysis of Microsatellite data
The software Alpha View version 3.2.4.0 was used to estimate marker and allelic length for the male and female haplotypes. Segregation of alleles based on actual length (in bp) resulted in monomorphism or polymorphism. The genotypic scores were calculated under the Mendelian law of inheritance and any unknown genotype (for example, in sire of all the families) was discovered or calculated. A chi-square test was performed to determine the departure of allele frequencies from 1:1 ratio in males and in females.

Authors’ contributions
Conceived and designed the experiments: MGQK, NS, MSA. Performed the experiments: MGQK, NS. Analyzed the data: MGQK, NS. Contributed reagents/materials: MSA. Wrote the paper: MGQK. Contributed to editing: MSA, NS. All authors read and approved the final manuscript.

Acknowledgements
We thank Bangladesh Agricultural Research Council to provide financial support from the core research grant (BARC Core 116/2014).

References
Baroiller J.F., D'Cotta H., Bezault E., Wessels S., and Hoerstgen-Schwark G., 2009, Tilapia sex determination: Where temperature and genetics meet. Comparative Biochemistry and Physiology, Part A, 153: 30-38

Cnaani A., Lee B.-Y., Zilberman N., Ozouf-Costaz C., Hulata G., Ron M., D’Hont A., Baroiller J.-F., D’Cotta H., Penman D.J., Tomasino E., Coutanceau J.-P., Pepey E., Shirak A., and Kocher T.D., 2008, Genetics of sex determination in tilapiine species, Sexual Development, 2: 43-54
http://dx.doi.org/10.1159/000117718

Eshel O., Shirak A., Weller J.I., Slossman T., Hulata G., Cnaani A., and Ron M., 2010, Fine mapping of a locus on linkage group 23 for sex determination in Nile tilapia (Oreochromis niloticus), Animal Genetics, 42: 222-224
http://dx.doi.org/10.1111/j.1365-2052.2010.02128.x

Eshel O., Shirak A., Weller J.I, Hulata G., and Ron M., 2012, Linkage and physical mapping of sex region on LG23 of Nile tilapia (Oreochromis niloticus), G3, 2: 35-42
http://dx.doi.org/10.1534/g3.111.001545

FAO., 2013, The state of world Fisheries and Aquaculture. Food and Agriculture Organization for the United Nations, Rome

Guerrero R.D., and Guerrero L.A., 1988, Feasibility of commercial productions of sex-reversed Nile tilapia fingerlings in the Philippines, In: Pullin RSV, Bhukaswan T, Tonguthai K, Maclean JL (Eds.), Proceedings of the Second International Symposium on Tilapia in Aquaculture, Bangkok, Thailand, Department of Fisheries, Bangkok, Thailand, and ICLARM, Manila, Philippines, Conference Proceedings, 15, pp.183-186

Hussain M.G., Penman D.J., and McAndrew, B.J., 1998, Production of heterozygous and homozygous clones in Nile tilapia. Aquaculture International, 6: 197-205
http://dx.doi.org/10.1023/A:1009250626496

Jalabert B., Billard R., and Chevassus B., 1975, Preliminary experiments on sex control in trout: production of sterile fishes and self fertilization of hermaphrodites. Annales de Biologie Animale, Biochimie, Biophysique, 15: 19-28
http://dx.doi.org/10.1051/rnd:19750103

Khan M.G.Q., 2011, Marker-assisted selection in genetically male Nile tilapia (Oreochromis niloticus L) production, Dissertation for Ph.D., University of Stirling, Stirling, UK, Supervisor: Penman D.J., and McAndrew B.J., pp. 204

Khan M.G.Q., McAndrew B.J., and Penman D.J., 2014, Inheritance of sex and sex-linked markers in Nile tilapia (Oreochromis niloticus L), In: Wahab M.A., Shah M.S., Hossain M.A.R., Barman B.K., and  Hoq M.E. (eds.), Advances in Fisheries Research in Bangladesh: I. Proceedings of the 5th Fisheries Conference and Research Fair, 8-10 January 2012, Bangladesh Agricultural Research Council, Dhaka, Bangladesh, pp. 59-74

Lee B.-Y., Penman D.J., and Kocher T.D., 2003, Identification of a sex determining region in Nile tilapia (Oreochromis niloticus) using bulked segregant analysis, Animal Genetics, 34: 379-383
http://dx.doi.org/10.1046/j.1365-2052.2003.01035.x

Lee B.-Y., Hulata G., and Kocher T., 2004, Two unlinked loci controlling the sex of blue tilapia (Oreochromis aureus), Heredity, 92: 543-549
http://dx.doi.org/10.1038/sj.hdy.6800453

Lee B.-Y., Lee W.J., Streelman J.T., Carleton K.L., Howe A.E., Hulata G., Slettan A., Stern J.E., Terai Y., and Kocher T.D., 2005, A second-generation genetic linkage map of tilapia (Oreochromis spp.), Genetics, 170: 237-244
http://dx.doi.org/10.1534/genetics.104.035022

Mair G.C., Abucay J.S., Beardmore J.A., Skibinski D.O.F., 1995, Growth performance trials of genetically male tilapia (GMT) derived from YY-males in Oreochromis niloticus L.: On station comparisons with mixed sex and sex reversed male populations. Aquaculture, 137: 313-322
http://dx.doi.org/10.1016/0044-8486(95)01110-2
 
Mair G.C., Abucay J.S., Skibinski D.O.F., Abella T.A., and Beardmore J.A., 1997, Genetic manipulation of sex ratio for the large-scale production of all-male tilapia, Oreochromis niloticus, Canadian Journal of Fisheries and Aquatic Science, 54: 396-404
http://dx.doi.org/10.1139/f96-282

Müller-Belecke A., and Hörstgen-Schwark G., 1995, Sex determination in tilapia (Oreochromis niloticus) sex ratios in homozygous gynogenetic progeny and their offspring, Aquaculture, 137: 57-65
http://dx.doi.org/10.1016/0044-8486(95)01114-5

Palaiokostas C., Bekaert M., Khan M.G.Q., Taggart J.B., Gharbi K., McAndrew B.J., and Penman D.J., 2013, Mapping and validation of the major sex determining QTL in Nile tilapia (Oreochromis niloticus L.) using RAD sequencing. PLoS ONE 8(7): e68389, doi:10.1371/journal.pone.0068389
http://dx.doi.org/10.1371/journal.pone.0068389

Palaiokostas C., Bekaert M., Khan M.G.Q., Taggart J.B., Gharbi K., McAndrew B.J., and Penman D.J., 2015, A novel sex-determining QTL in Nile tilapia (Oreochromis niloticus), BMC Genomics, 16:171: doi 10.1186/s12864-015-1383-x
http://dx.doi.org/10.1186/s12864-015-1383-x

Penman D.J., and Piferrer F., 2008, Fish Gonadogenesis. Part I: Genetic and environmental mechanisms of sex determination. Reviews in Fisheries Science, 16: 16-34
http://dx.doi.org/10.1080/10641260802324610

Shirak A., Seroussi E., Cnaani A., Howe A.E., Domokhovsky R., Zilberman N., Kocher T.D., Hulata G., and Ron M., 2006, Amh and Dmrta2 genes map to tilapia (Oreochromis spp.) linkage group 23 within quantitative trait locus regions for sex determination. Genetics, 174: 1573-1581
http://dx.doi.org/10.1534/genetics.106.059030

Wohlfarth G.W., and Wedekind H., 1991, The heredity of sex determination in tilapias. Aquaculture, 92: 143-156
http://dx.doi.org/10.1016/0044-8486(91)90016-Z