Genetica80: 195-200, 1990. © t990KluwerAcademicPublishers.Printedin Belgium

195

Inheritance of the red color in tilapias L. Reich, J. Don & R. R.Avtalion Laboratory offish Immunology and Genetics, Department of Life Sciences, Bar-Ilan University, Ramat-Gan Israel Received31.5.1989

Acceptedin revisedform 27.1.1990

Abstract The inheritance of the red color was studied in two different varieties of tilapia which are both considered as hybrids of Oreochromis mossambicus. Crosses between red tilapia from the Philippines (PRT) and Sarotherodon galilaeus, or Oreochromis aureus gave a 1:1 ratio of red: normal and crosses between F~ black fish gave only black offspring. On the other hand crosses between the F~ red fish gave a 3:1 ratio of red:black and crosses between F~ red and black offspring gave a 1:1 ratio. These results lead to the conclusion that red color is dominant over the normal black color and controlled by a single autosomal gene (R). A unique phenotype named 'albino with black eyes' was observed among offspring of PRT and a presumed model of inheritance of this trait is proposed. Genetic analysis of a second variety of red tilapia (derived from an unknown origin) showed the following results: crosses between parents and between their Fl offspring consistently gave 100% red fish and crosses between this red tilapia and Oreochromis aureus gave 100% black offspring. The crosses between red and black F~ of these last two crosses gave a 1:1 ratio and crosses carried out between the black Ft offspring gave a 1:3 ratio of red:black. It may be concluded from these results that the black color is dominant in this strain and that this color is controlled by a single autosomal gene (B). The presumed mode of action of the dominant gene (R) as well as of the recessive gene (b) are discussed.

Introduction The culture of red tilapia is of some importance because its red color (a blend of red, pink, yellow and gold) is more appreciated in some markets than black tilapia. It has been reported that this fish has a fast growth rate and the ability to grow in both brackish and salt water (Sipe, 1979; Liao & Chang, 1983; Galman & Avtalion, 1983). The origin and the history of the red tilapia originating in Taiwan and the Philippines is not well documented. It is considered as a hybrid of O. mossambicus X O. hornorum with O. niloticus. In fact various morphological traits of these species (e.g. shape of snout, width to width of head)could be observed in individual different red tilapia (Galman & Avtalion, 1983). Electrophoretic analysis of transferrins and esterases also revealed high polymorphism, and no specific electrophoretic markers could be found.

The genetic determination of color and pigmentation is not yet well understood. A number of models for the inheritance of body color and pigmentation in red tilapia have been suggested (Mires, 1987; Scott et al., 1987; McAndrew et al., !988) and in the present work some of these models have been considered in order to explain the experimental results.

Materials and methods Fish The following tilapias were used in this study: (1) Sarotherodon galilaeus (Gal) originally from the sea of Galilee in Israel; (2) red tilapia originally from the Philippines (PRT) showing red (rPRT) and black (bPRT) phenotypes; (3) F~ hybrids of Gal )< rPRT; (4) a strain of red tilapia of unknown origin considered by the farmers as a red mutant of Oreochromis mossam-

196 bicus (MRT); (5) O. aureus (Au); (6) Ft hybrids of Au X MRT. The fish were divided into families of 4-5 females and 1-2 males each. The families were kept in aquaria of 280 liters, each equipped with a closed circulating system (Koiller & Avtalion, 1985). The water temperature was 26-28°C during the spawning season and 24-26°C in winter. The fish were fed with pellets containing 40% protein.

Artificial fertilization Artificial fertilization of the eggs was carried out in Petri dishes using modified Eagle's medium (MEM), as described by Yeheskel and Avtalion (1988). Immediately before spawning, the females were gently caught and the eggs were stripped out into a small Petri dish. Sperm was stripped out of the male in a similar way and diluted with MEM (pH 7.2-7.4), containing 5% carp serum. The eggs were washed with UV-sterilized water from Zuger-bottles before being fertilized with sperm. The eggs were gently agitated and then transferred after 10 min to Zuger-bottles for incubation (Don et al., 1987). The water was mechanically filtered through activated charcoal and sterilized by passage through a UV irradiator. The temperature of the water was kept constant at 26°C. Embryos were kept in the Zuger-bottles until complete yolk sac absorption. In order to avoid the influence of differential mortality at the early ages, the ratios between red and black phenotypes (Fig. 1) were determined at the embryonic age of 8 days by microscopic examination (X 20). The observed segregation ratios were tested by the X2 test for acceptance or rejection of the proposed hypothetical model.

Fig. 1. Colored phenotypes of 8-10 day old embryos: (A) black; (B) red; - (C) red with black eyes (X20).

and examined by elecron microscopy (TEM 100SK, Jeol, Japan). Electron microscopy Small pieces ofiridal tissue from normal, red or blackeyed, 10-20 days old embryos, were fixed in 2.5% glutaraldehyde in 0.1M Sorensen buffer (pH=7.2). After washing in buffer and fixing with 1% osmium tetroxide in the same buffer, the tissues were dehydrated in an alcohol series and then embedded in spures (Polaron, UK), thin sectioned using Ultratome III (LKB), stained with uranyl acetate and lead citrate

Results

Crosses carried out between 7 pairs of rPRT, consistently gave a ratio of 3:1, red:black (P=0.5-0.2). A ratio of 1:1 (red:normal) was obtained in 3 crosses of rPRT X bPRT or when an rPRT female was crossed with S. galilaeus or with O. aureus (P=0.9-0.7, 0.5-0.2 and 1, respectively). A bPRT X bPRT cross gave a

197 ratio of 0:1, red: normal (P= 1). These results suggest that rPRT is heterozygous (Rr) for a red controlling gene (R) and that bPRT, S. galilaeus and O. aureus, are homozygous (rr) for a recessive gene (r) (Table 1). Table 1. Results of crosses showing red and black phenotype segregations in PRT, and presumed parental genotypes.

PRT Tag No.

Q

Replicate N

(3

Genotype Color of parents R B

Ratio R/B P

Q

bPRT bPRT 2 G PRT 2 A PRT 2

84 rr 72 rr 148 rr

rr Rr Rr

0 84 0:l 39 33 1:1 74 74 1:1

1 0.5-0.2 1

rr rr Rr

30 34 1:1 105 112 1:1 25 18 1:1

0.7-0.5 0.7-0.5 0.5-0.2

160 164

1:1

0.9-0.7

682 248 156 98 190 28 221

3:1 3:1 3:1 3:1 3:1 3:1 3:1

0.5-0.2 0.5-0.2 0.7-0.5 0.1 0.2-0.1 0.9-0.7 0.5-0.2

1623 560 3:1

0.5-0.2

9

5

2

64 Rr

6 4

1 7

1 1

217 Rr 43 rr

Total:

A unique .phenotype with bright color and black eyes (BE) was observed, at different rates (5-19.5%) among offspring of 7 different rPRT × rPRT crosses (Rr × Rr). The segregation results showed 3 different ratios: 9:3:4 (in 2 pairs, P=0.5-0.2), 21:3:8 (in 4 pairs, P=0.2-0.1), and 45:3:16 (in 1 pair, P=0.7-0.5) for the following 3 phenotypes: red, BE and black, respectively (Table 2). The segregation results for body and eye Table 2. Results of rPRT × PRT crosses showing segregation for red, black eyes and black.

PRT RepliTag No. cate N

Q

~

6 23

16 16

7 16 7 16 16 7 16

6 2 2 2 2 1 1

921 338 212 121 266 38 287

Rr Rr Rr Rr Rr

Rr Rr Rr Rr Rr

Rr Rr

Rr Rr

Total:

239 90 56 23 76 10 66

R, red; B, black; N, offspring number; bPRT, black PRT; G, S.

8 6 9 32

16 7 7 16

2 1

2 2 1 2

Total: 8

7

Ratio %BE R/

BE/B

Total: 8 8 6 6 32 9 23

Color R BE B

6

121 287

80 18 165 56

23 66

15 9:3:4 0.1 -0.05 19.5 9:3:4 0.9-0.7

245 74

89

17

338 226 22 212 141 15 38 25 3 266 163 27

90 6.5 56 7 10 7.9 76 10

9:3:4 0.5-0.2 21:3:8 21:3:8 21:3:8 21:3:8

0.2 -0.1 0.5-0.2 0.99-0.95 0.5-0.2

555 67

232

8

21:3:8 0.2 -0.1

921 636 46

239

5

45:3:16 0.7-0.5

R, red; BE, red with black eyes; B, black; N, offspring number.

galilaeus; A, O. aureus; PRT, Red tilapia from the Philippines.

Fig. 2. Ultrastructure of iridophores of the golden tissue of normal eye. lr, Iridophores; N, nucleus; RP, reflecting platlets (X500); insert

(×100O0).

198

Fig. 3. Meridionai section of pupiliary border of the iris of tilapia embryo: (A) black; (B) red with black eyes - AC, anterior chamber; PC,

posterior chamber; ML, melanophore layer; IL, iridophore layer, IE, inner epithelium; C, cornea 0 < 1000).

color point to the possibility that this phenotype is controlled by two recessive homozygous genes (Is, 12) which seem to be associated to the R gene and can only be expressed in its presence (discussed below). ' Light and electron microscopy examinations showed that in normal fish an external tissue layer of the iris (golden tissue) contains structures characteristic of pigment cells called iridophores (Fig. 2). This layer was found to be completely absent in the iridal tissue of the black-eyed embryos which contain only the black melanophore tissue supported by the inner epithelium (Fig. 3). This phenotype seems to be associated with a lethal trait because most of the fish died during the embryonic period before the feeding stage was reached. A small proportion of these fish (5-10%) were able to reach a weight of 4-5 g before dying. Crosses carried out between 5 pairs of MRT gave a ratio of 1:0 red:black (P=I), and a ratio of 0:1 red:black (P~- 1) was shown in FI hybrids of 3 different crosses of a MRT female with an O. aureus male. Three crosses carried out between a black hybrid male and 3 different MRT females gave rise to a ratio of 1:1 (P=0.7-0.5). Seven crosses carried out between 2 pairs of hybrids yielded a ratio of 1:3 red:black (P----0.7-0.5) (Table 3). Genetic analysis clearly showed that in this fish the red color is recessive to the black color. Therefore MRT, which gave 100% red offspring, seems to be a recessive homozygote (bb). Hence, in the

Ft black hybrid crosses (Bb) and in the crosses of MRT with O. aureus (BB) 75% and 100%, respectively of black offspring was obtained.

Table 3. Results of crosses showing red and black phenotype segregation in MRT and its presumed genotypes.

MRT Tag No.

Replicate N

Q

~

A

41

3

45 20 42 0 0

41 4 3 3 4

4 1 1 1 1

Genotype of parents

Q

c3

492

BB

bb

1531 44 200 105 39

bb bb bb bb bb

bb bb bb bb bb

Color R B

0 492 0:1 1 1531 44 200 105 39

Total: 1919 42 20 0

2 2

39 39 39

39 32

1 1 1

5 2

38 62 39

1350 663

bb bb bb

Bb Bb

Ratio R/B

0 1:0 1 0 I:0 1 0 1:0 1 0 1:0 1 0 1:0 1 0 1:0

1

Bb Bb Bb

19 33 21

19 1:1 1 29 1:1 0.7-0.5 18 1:1 0.7-0.5

Total:

73

66 1:1

Bb Bb

Total:

0.7-0.5

348 1002 1:3 0.7-0.5 144 519 1:3 0.05 492 1521 1:3

A, O. aureus; R, red; B, black; N, offspring number.

0.7-0.5

199 Discussion

The results of this study show that red color of PRT is a dominant trait which seems to be controlled by a single autosomal gene. According to this model it is possible to divide the F 2 red fish into two genotypes: homozygous red (RR) and heterozygous red (Rr). Identification of the homozygote (RR) was attempted by performing testcrosses between homozygotes (RR) and normal black fish (rr) which were predicted to give 100% red offspring. Unfortunately, the few F2 red fish we examined in the frame of this work were not identified as homozygotes (not shown). However a pink phenotype yielding 100% pink offspring, was obtained in our group (Galman et al., 1987) in PRT following three generations of inbreeding of the pink phenotype and by others in Taiwanese red tilapia (Huang et ai., 1988). It was then reported that only the pink color is homozygous while the red is heterozygous. In contrast, red homozygous individuals, which gave 100% red offspring when crossed with heterozygotic red (Rr) or normal black fish (rr), could be selected from a stock of O. niloticus originating from Egypt (McAndrew et al., 1988). Red PRT are characterized by black spots of different sizes, dispersed over different parts of the skin, especially on the head. We observed that these spots develop at the post-embryonic stage. Therefore we have come to the conclusion that two types of melanophores which develop at two different ontogenetic stages, are present in adult fish: the first, susceptible to the R regulating gene, develops during early embryonic development of PRT, while the second, which develops a few months later, is not susceptible to the suppressive action of this gene. A similar phenomenon was reported in carp by Katasonov (1978). An unusual phenotype characterized by bright color and black eyes wgtsobserved by us. This has also been reported by Fitzgerald (1979) and Galman et al. (1987). The latter named it 'albino with black eyes'. Genetic results of the segregation ratios of body and eye color point to the possibility that this phenotype is controlled by two recessive homozygous genes (11,12). It also seems to us that this phenomenon is associated with the R gene and can only be expressed in its presence. We would suggest that the corresponding

genotypes are as follows: Black (rr), red with black eyes (R-il il i2 i2), Red (R_I 1__12_, R_I 1_i 2 i 2,R_il i112 -) for the three ratios of 9:3:4, 21:3:8, 45:3:16 (red:red with black eyes: black) observed (Table 2). By using electron microscopy, we observed that the golden tissue which is lacking in the black eyes contains structures characteristic of iridophores (Fig. 2, 3). Therefore it is possible that a genetic defect related to the formation of the iridophores is the cause of the black eye phenotype, as reported in the Mexican axolotl (Benjamin, 1970). In this animal, there exists a color mutation controlled by the melanoid (m) gene, with complete inhibition of iridophore differentiation including the eyes, and, to a lesser extent inhibition of the formation of xanthophores. However, this m gene was found to enhance the differentiation of melanophores. It is still not clear to us whether the lack of iridophore formation in the black-eyed tilapia exists all over the body or is "limited to the eyes only. This point should be clarified by further studies to be carried out by our group in the next spawning season. The color inheritance in MRT was found to be different from PRT, because a dominance of the black color over red color was shown in crosses of MRT (red) with O. aureus (black) and in crosses between two different black hybrids of this cross which gave a ratio of 1:3 red:black (Table 3). These results suggest that the black color is determined by a dominant gene B and its recessive allele b is responsible for the red color. This model is in agreement with the results obtained in a red O. niloticus strain originating in Uganda (Mires, 1987) which showed three different phenotypes (red, orange and albino), and similar to that described by Scott et al. (1987) and by McAndrew et al. (1988) for the inheritance of blond color in another strain of O. niloticus originating in Egypt. Because of the occurrence of skin melanophores as seen by electron microscopy, the latter authors suggested that in their blond phenotype there is a partial block of melanin granule formation. In contrast, no melanophores at all could be observed in our MRT either with light or electron microscopy (not shown). In contrast to PRT, where one type of melanophores was found to be dominated by the R genes whereas the other type was not, it could be suggested that in MRT the red color depends on a recessive gene (b) which might be involved in the inhibition of melanin synthesis.

200 Therefore we suggest the following genotypes for the different tilapias studied: Oreochromis aureus, rrBB; Sarotherodon galilaeus, rrBB; rPRT, R__BB; bPRT, rrBB; rMRT, rrbb; bMRT, rrB_, where Rr and Bb are the genes which we suggest control the red and black colours, respectively.

Acknowledgements This research was supported by a grant (FR.8526) from the Bar-Ilan research and development company. We wish to thank Mrs. Susan Weingarten for reviewing this manuscript.

References Benjamin, C. P., 1970. The biochemical effects of the d, m and a genes on pigment cell differentiation in the Axolotl. Devel. Biol. 23: 62-85. Don, J., Koiller, M., Yeheskel, O. & Avtalion, R.R. 1987. Increased Tilapia embryo viability using ultraviolet irradiation in closed recirculating Zuger-bottle system. Aquacult. Engin. 6: 69-74. Fitzgerald, W. J., 1979. The red-orange tilapia, a hybrid that could become a world favourite. Fish Farming 6: 26-27. Galman, O. R. & Avtalion, R. R. 1983. A preliminary investigation of the characteristics of red tilapia from the Philippines and Taiwan. pp. 291-301. In: L. Fishelson & Z. Yaron (eds),

International Symposium on Tilapiain Aquaculture. Tel-Aviv Univ. Press, Israel. Galman, O.R., Moreu, J. & Avtalion, R.R. 1988. Breeding characteristics, growth performance and implications for aquaculture of Philippine red tilapia, pp. 169-175. In: R. S. V. Pullin, T. Bhukaswan, K. Tonguthaiand J. L. Maclean (eds), ICLARM Edit. (Manila and Bangkok). Huang, C. M., Chang, S. L., Cheng, H. J. & Liao, I. C. 1988. Single gene inheritance of red body coloration in Thaiwanese red tilapia. Aquaculture 74: 227-232. Katasonov, V. Ya., 1978. Color in hybrids common and ornamental (Japanese) carp. 3. Inheritance of blue and orange color types. Genetika 14: 2185-2192. Koiller, M. & Avtalion, R.R. 1985. A. laboratory scale recycling water unit for Tilapia breeding. Aquacult. Engin. 4: 235-246. Liao, I. C. & Chang, S. L. 1983. Studies on the feasibility of the red tilapia culture in saline water, pp. 524-533. In: L. Fishelson and Z. Yaron (eds), International Symposium on Tilapia in Aquaculture. Tel-Aviv Univ. Press, Israel. McAndrew, B. J., Roubal, F. R., Roberts, R. J., Bullock, A. M. & Mc Ewen, I. M. 1988. The genetics and histology of red blond and associated color variants in O. niloticus. Genetica 76: 127-137. Mires, D., 1988. The inheritance of black pigmentation in O. niloticus from two African origins, pp. 237-241. In: R.S.V. Pullin, T. Bhukaswan, K. Tonguthai and J. L. Maclean (eds). ICLARM Edit. (Manila and Bangkok). Scott, A. G., Mair, G. C., Skibinski, D. O. F. & Beardmore, J. A. 1987. 'Blond' - a useful new genetic marker in tilapia O. niloticus (L) Aquacult. Fish Manag. 18: 159-165. Sipe, M., 1979. Golden perch. Commercial Fish Farmer. Aquaculture News, 5.56. Yeheskel, O. & Avtalion, R.R. 1988. Artificial fertilization in tilapia - A preliminary study. Colloq. INRA 44: 169-175.