Embryonic development of diploid and triploid eggs of Clarias gariepinus (Burchell, 1822)

ABSTRACT

This study was conducted to evaluate the spawning performance and embryonic development of diploid and triploid eggs of the African Catfish, Clarias gariepinus. Triploidization was induced by heat shock at 41°C (for 20 min) at approximately 3 min after fertilization. Result obtained shows that the egg characteristics, fertilization percentage and the biometric parameters of hatched fry were similar in both treatments. However, more of the diploid eggs hatched (71.98%) than their triploid counterparts (56.31%). In addition, there were more observed abnormalities in the latter (35.43%) than the formal (2.71%). During the early embryogenetic cleavages stages (i.e. between two-cell and 64-cell stages), the diploid eggs advanced 5–23 min earlier than the triploid eggs. However, beyond the blastula stage, embryonic development was faster in the triploid; hence, hatching occurred 194 min earlier than the diploid eggs. The abnormal cleavages (unequal and asymmetric in nature) observed in the early stages of development in the triploids were thought to have resulted in the development of the somite at 50% epiboly. Consequently, abnormal hatched triploid fry from such eggs were short trunked. This study also affirmed the suitability of erythrocyte morphological as a simple and reliable index of discrimination between diploid and triploid African catfish.

KEYWORDS:

• African catfish
• heat shock
• abnormality
• embryogenesis

Introduction

Triploidization is a chromosome set manipulation method, used to improve the performance of fish by the addition of an extra set of chromosomes to the two-chromosome set possessed by normal organisms. Triploidy induction in many fish species has been initiated to take advantage of certain performance characters such as increased survival rate, growth performance, and meat production (Piferrer et al. 2009; Berrill et al. 2012; Fraser et al. 2012). Theoretically, triploid organisms are either reproductively sterile or have reduced reproductive functionality. This is largely due to the inability of the three homologous set of chromosomes to progress through recombination during the first meiotic division (Benfey and Bennett 2009). Hence, the rationale of environmental protection (i.e. escape of domesticated animals to the wild) and the quest for an efficient management tool in aquaculture has largely encouraged the production of triploid food fish (Benfey 1999; Felip et al. 2001; Maxime 2008).

Natural triploidy does exist in the wild (Zhou and Jianfang 2017). This is largely as a result of problems in the egg-ripening mechanisms coupled with the failure of the second polar body extrusion from the fertilized egg (Thorgaard and Gall 1979a; Flajsˇhans et al. 1993). It could also occur as a result of unreduced gametes during spawning or hybridization between different fish family due to reduce habitat complexity (Xiao et al. 2011; Wang et al. 2016; Zhou and Gui 2017). However, in the laboratory, triploidization induction is mainly done by preventing the release of the second polar body during the second stage of meiosis, which occurs 3–7 min after fertilization in many fishes (Carman et al. 1991). This is achieved by physical (heat, cold and pressure shock) or chemical (anti cleavage substances such as colchicine, vincristine, and cytochalasin) treatments (Piferrer et al. 2009).

However, shock-induced triploidization leads to embryonic mass mortality (Piferrer et al. 2000, 2003). The underlining principle and mechanism resulting in this are still not well understood. In previous studies, the morphology and division pattern of the animal pole or blastomere of fish eggs have been widely used not only as an indication of egg quality (Shields et al. 1997; Kjørsvik et al. 2003) but for larval characteristics in pure and hybrid fishes (Okomoda et al. 2017a). These include hatchability rates (Devauchelle et al. 1988; Kjørsvik et al. 1990, 2003; Pickova et al. 1997), abnormality percentage (Okomoda et al. 2017a, 2017b, 2018a), as well as susceptibility to infection (Hansen and Olafsen 1989; Pavlov and Moksness 1993), pollution (Longwell et al. 1992) or water quality (Okomoda et al. 2017b, 2018b). This line of research has not been given much attention in many previous triploidization trials.

Clarias gariepinus is one of the species of importance widely cultured in West Africa, Southeast Asia and many parts of the world (Verreth et al. 1992; Adamek and Sukop 1995; Hecht et al. 1996). This is largely due to its high fecundity, growth performance, as well as hardiness to poor water quality (Solomon et al. 2015). It is also considered one of the most excellent animal models for developmental and embryological studies (Sule and Adikwu 2004). This is largely because of its non-pigmented, oval-shaped and transparent eggs which facilitate easy observation of the embryonic changes under a simple light microscope (Sule 2002; Sule and Adikwu 2004). Triploidization of the African catfish is not a new concept as many previous studies have demonstrated its possibility and potential for aquaculture (Richter et al. 1986; Henken et al. 1987; Karami et al. 2010; Normala et al. 2016, 2017). However, the culture potential and propagation still needs to be improved. Olufeagba et al. (2015) stated that understanding the embryogenesis and early life development is the first step to improving artificial propagation of cultured fish species. Hence, since previous studies had focused on shock optimization (intensity and timing) and triploid progeny identification (erythrocyte and chromosome characterization), this study was designed to investigate for the first time the events and timing of the embryonic development of triploid induced eggs of African catfish, C. gariepinus.

Materials and methods

Sexually mature broodstocks C. gariepinus (above 1 kg) maintained in fiber glass tanks at the School of Fisheries and Aquaculture Science hatchery of the Universiti Malaysia Terengganu, in Malaysia were used for this study. Breeding and embryogenetic development reported in this study were the combinations of results from three different breeding trials using two pairs of male and female brood fish per trial. Females were injected with Ovaprim® hormone at a dose of 0.5 ml kg⁻¹ of the fish. They were separated into different rearing tanks (80 × 60 × 40cm³) and allowed to swim freely for a latency period of 11 h. Eggs were then collected in a clean bowl by stripping the female softly along its abdomen. The male broodstocks were euthanized, the abdominal cavity opened with the aid of a scissor to obtain the testis and then lacerated to obtain the milt. Fertilization was achieved by mixing the eggs and milt while the gametes were activated by the addition of water.

The fertilized eggs were quickly divided into two bowls for the two treatments designed for this study. A portion of the eggs was used for triploidization by “heat shock” at 41°C, while the other portion was not, hence regarded as the control group. The protocol for the heat shock had earlier been optimized (but not reported in this study) using established baseline time parameters set for this species by previous authors who used cold shocks (Wolters et al. 1981; Richter et al. 1987; Manickam 1991). In brief, fertilized eggs for triploidization were exposed to a 41°C water bath for 20 min, at approximately 3 min after fertilization. The control and triploid portions of eggs were immediately incubated in already prepared triplicates batches of 100 l fiberglass tank with continuous aeration. Water quality before and after heat shock (i.e. during the incubation) was monitored and kept optimum (temperature = 27.3 ± 0.9°C; pH = 7.50 ± 0.17; conductivity = 525 ± 5.05 µS cm⁻¹; total dissolved solid = 270.0 ± 1.23 mg l⁻¹; dissolved oxygen = 5.22 ± 0.98 mg l⁻¹). It is important to note that, due to the sensitivity of temperature to the process of embryogenesis, digitally regulated heaters set at 28°C were used to maintain the temperature of the triplicated treatment baths in which the eggs were incubated.

Approximately 50 fertilized eggs were collected (in Petri dishes) from each of the treatments and observed under a Nikon dissecting microscope (model number C-DSLS, Japan). The embryonic observations started 15 min before the attainment of each stage as reported by Olufeagba (1999, 2016) for the same species. The observed developmental stages, as well as abnormalities, were captured using a Sony camera (Cyber-shot 16.2MP Model number: DSC-TX10 50i, Japan) fitted to the microscope. Monitoring was maintained from the point of fertilization up to when fry was hatched. It is also important to note that incubation in the fiberglass tanks and embryonic observation under the microscope was done in an enclosed room in the Pusat Pengajian Sains Perikanan dan Akuakultur (PPSPA) hatchery with the air conditioning system set at 28°C throughout the period of observation. Hence, it is believed that the environmental and water temperature during the study was substantially controlled in all treatment as well as in all the trials of the study. Also, the egg size (n = 20) before and after fertilization were obtained using a Nikon profile projector (Model number V-12BD/JA) attached to a Nikon digital counter (Model number SC-212).

Fertilization and hatchability rate in this study was determined using the equations below:

Upon hatching, the biometric characteristics of the larvae (Figure 1) were recorded using the Nikon profile projector. Abnormality percentage was determined.

Descriptive statistics for breeding and larval characteristics as well as the embryonic development timing were performed using Minitab 14® computer software (Minitab, Inc, Pennsylvania State College, USA) followed by the Student’s t-test. Also, the status of triploidization and efficiency of the optimized induction process used in this study was confirmed by identifying the exclusive triploid range of the cell major axis (11.9–14.9 μm) already established for the species by Normala et al. (2016).

Results and discussion

Ola-Oladimeji (2015) had earlier stated that relative equality in fertilization rates is a strong indication of similarities in egg and milt qualities of broodstocks used. Hence, the observed similarity in fertilization rate in both diploid (82.10%) and triploid (80.09%) eggs was expected (P ≥ 0.05) since they were pooled from similar broodstock sources (Table 1). More so, triploidization was induced three minutes after fertilization; thus, the consequential effect of the heat shock process could only be evident after the treatment application and not before. Egg size variation had previously been linked to broodstock quality (Bromage and Roberts 1995) and size (Ataguba et al. 2013). However, the observation of an increased egg size after fertilization in this study resonates perfectly with the assumptions of Olufeagba et al. (2016). These researchers had linked increased egg size to hydration of the eggs post-fertilization. Egg sizes reported in this study were within the range reported by de Graaf and Janssen (1996), Olaniyi and Omitogun (2013), Olaniyi and Omitogun (2014) for the same species. Generally, the effect of egg size transcends beyond the developmental ability of the embryo to the characteristics of hatched fry (Ataguba et al. 2013). Okomoda et al. (2017a) had also demonstrated the preponderant influence of maternal origin on the early morphometry characteristic of hatched larvae in their study. Hence, the similarities in egg size (before and after fertilization) and the characteristics of normal hatched larvae (head length, body width, total length, yolk length and width) in both treatments (P ≥ 0.05) could be explained by common parents used in this study.

Table 1. Egg, breeding and larval characteristics of diploid and Triploid of C. gariepinus. Numbers are means ± standard errors.

	Diploids	Triploids	P-value
Egg size pre-fertilization (mm)	0.87 ± 0.12	0.90 ± 0.10	0.435
Egg size post fertilization (mm)	1.01 ± 0.21	0.98 ± 0.04	0.350
%Fertilization	82.10 ± 0.92	80.09 ± 4.09	0.551
%Hatchability	71.78 ± 0.76^a	56.31 ± 2.06^b	0.001
%Abnormality	2.71 ± 0.60^a	35.43 ± 4.20^b	0.001
Head length (mm)	1.14 ± 0.06	1.09 ± 0.03	0.442
Body width (mm)	0.45 ± 0.13	0.46 ± 0.22	0.112
Total length (mm)	4.20 ± 0.05	4.15 ± 0.16	0.421
Yolk length (mm)	1.37 ± 0.05	1.42 ± 0.05	0.098
Yolk width (mm)	0.90 ± 0.14	0.93 ± 0.19	0.150

Mean in the same row with different superscripts differ significantly (ANOVA, P ≤ 0.05).

Despite these similarities, hatchability was higher in the diploids (71.78%) and lower in the triploid fish (56.31%). In contrast, the percentage of abnormality was higher with triploidization (35.43%) compared to the control diploid group (2.71%) (P ≤ 0.05). These observations are probably linked to the developmental pattern of the different treatment during embryogenesis. Despite similarities in the embryonic stages of the different treatment (Table 2 and Figure 2(A–Q)) with many previously reported catfishes (Aluko 1995; Olufeagba et al. 1999; Arockiaraj et al. 2003; Da Rocha et al. 2009), abnormal developmental pattern characterized the early mitotic division of some of the triploid progenies (Figure 3(A–C)). The abnormal developing embryos were characterized by unequal cleavage (Figure 3(A)) as well as asymmetric-cleavage (Figure 3(B), 3(C)). However, in the normal developing embryos of both diploids and triploids, the early mitotic divisions were mainly discoidal meroblastic in nature, resulting in blastomeres of relatively equal sizes (similar to reported studies of Kimmel et al. 1995; Ninhaus-Silveira et al. 2006).

Table 2. Description of the embryogenetic development of diploid and triploid C. gariepinus under laboratory conditions (temperature = 27 ± 2.1°C). Numbers are means ± standard errors.

S/No	Stages	Diploid	Triploid	Description/observation
1	Fertilized egg	0 ± 0.0	0 ± 0.0	The unfertilized egg of C. gariepinus was greenish and oval, possessing a thin perivitelline membrane whose space was filled with protoplasmic layer. Upon fertilization, the eggs expand resulting in the separation of the inner yolk and the outer membrane. The adhesive nature of the eggs at this stage facilitated their attachment to the hatching substrate (Figure 2(A)).
2	One-cell stage	30 ± 2.1^b	36 ± 2.6^a	Granulated cytoplasmic accumulation at the anterior end of the yolk, hence, segregating the animal pole from the vegetal pole (Figure 2(B)).
3	Two-cell stage	40 ± 0.91^b	45 ± 3.20^a	The first mitotic division involving meroblastic cleavage results in the separation of the animal pole into two, producing two equal cells (Figure 2(C)).
4	Four-cell stage	52 ± 0.41^b	58 ± 3.03^a	The second cleavage just at right angle to the first division producing leads to the formation of the four-cell stage (Figure 2(D)).
5	Eight-cell stage	63 ± 3.0^b	71 ± 4.11^a	The third division results in eight cells (Figure 2(E)).
6	16-cell stage	71 ± 0.55^b	81 ± 1.40^a	The fourth division produced 16 cells either in regular four by four layers (Figure 2(F)) or asymmetric (Figure 3(S)) and irregular scattered layers (Figure 3(B), 3(C)). The latter eggs were mainly triploid and characterized by unequal shaped cells.
7	32-cell stage	86 ± 3.12^b	99 ± 2.4^a	Further division produced 32 cells (Figure 2(G)).
8	64-cell stage	118 ± 0.72^b	131 ± 1.52^a	Cell cleavages producing the 64 daughter cells were mainly latitudinal. Hence, cells tend to lie on one another making it extremely difficult to count (Figure 2(H)).
9	Morula stage	210 ± 4.30	215 ± 3.56	Further latitudinal cell divisions produced tiers of numerous cells with much more reduced sizes (Figure 2(I)).
10	Blastula stage	288 ± 2.63^a	272 ± 3.05^b	A domed shape cell heap on the yolk results from further division (Figure 2(J)).
11	Gastrulation stage	471 ± 1.33^a	413 ± 3.15^b	Germ cell rings characterized by random quasiperistaltic transitional wave movement is observed at this stage (Figure 2(K)). The germ rings increased until it covers 95% of the eggs (Figure 2(M), 2(N)). It is important to note at this point that some triploid embryos did not elongate beyond 50% of the egg yolk (Figure 3(U)).
12	Somite Stage	573 ± 1.20^a	509 ± 4.51^b	Somite formation begins with noticeable pairs of cells developing along the back of the embryo (Figure 2(O)). This arranged cells formed the vertebral column of the hybrid (Figure 2(P)).
13	Hatching	1419 ± 1.30^a	1225 ± 2.08^b	Movement of the tail at the prime of embryogenetic development (Figure 2(Q)) against the chorion wall forces it to rupture (Figure 2(R)).

Figure 1. Biometric parameters of the hatchling (adapted from Okomoda et al. 2017a).

Figure 2. Normal embryogenesis stages as observed in the diploid and triploid C. gariepinus under laboratory conditions. (A) Fertilized egg; (B) one-cell stage; (C) two-cell stage; (D) four-cell stage; (E) eight-cell stage; (F) 16-cell stage; (G) 32-cell stage; (H) 64-cell stage; (I) morula stage; (J) blastula stage; (K) gastrula stage; (L) 75% epiboly; (M) 90% epiboly; (N) 95% epiboly; (O) somite begins; (P) advance somite stage; (Q) prime; (R) hatchlings. Bar = 0.5mm.

Figure 3. Abnormalities in egg and hatchlings in the triploid of C. gariepinus under laboratory conditions (A) unequal cleavage; (B, C) different types of asymmetric cleavage; (D) abnormal somite development at 50% epiboly; (E, F) abnormal hatchlings with short/truncated tail. Bar = 0.5mm. <<ts: change part labels in Figure 3 to A to F, ie: change S to A, T₁ to B, T₂ to C, U to D, V₁ to E, V₂ to F>>

The abnormal embryos in this study are similar in pattern to those previously reported in triploids of Black Sea turbot Psetta maxima eggs (Aydın and Okumus 2017) and in reciprocal crosses of Asian and African catfishes (Okomoda et al. 2017a). While the observation in the latter study was explained by consequential postzygotic isolation mechanisms (Amini et al. 2007; Okomoda et al. 2017a), the observation in the present study and those of Aydın and Okumus (2017) could be a clear indication of the deleterious “triploidization effect” of heat or cold shocking (respectively). This probably resulted in unsynchronized cell cleavage, thereby making some cells divide much later than the sister does. The “triploidization effect” hypothesized in this study is also likely responsible for much earlier development of somite at 50% epiboly (Figure 3(D)), faster hatching time (Table 2), lowered hatchability rate and high percentage of abnormal fry (Table 1). It is no surprise therefore that the abnormal hatched triploid fry were characterized by short trunks, since somite development was much earlier (at 50% epiboly) than what was observed in the normal egg (at 95% epiboly). This is similar to the findings of Okomoda et al. (2017a) for the progenies of ♀P. hypophthalmus × ♂C. gariepinus.

The early embryogenetic stages (i.e. two-cell and 64-cell stages) were slower in the triploid (5–23min late) when compared to the diploid eggs. However, beyond the blastula stage, the observed sequence of timing was upturned in favor of triploid eggs. It is hypothesized that the 20 min heat shock applied 3 min after fertilization for the triploidization process may have halted some metabolic activities. Hence, beyond retention of the second polar body, it could have led to delayed development time during the early embryonic stages. The underlying principles of faster embryogenetic process in the blastula stage and through the organogenesis stages are not well understood. Similar earlier hatching tendency was previously reported in triploid rainbow trout, Saho gairdneri, by Happe et al. (1988). Unravelling the necessary scientific facts and explanation for this could be the focus of future studies; however, this could be connected to the extra chromosome set in the triploid eggs.

Normala et al. (2017) had earlier reiterated the ease of ascertaining the ploidy levels of the African catfish C. gariepinus using erythrocyte shape observation. Ellipsoidal and rounded erythrocytes have been previously reported as specific characteristics of triploid and diploid African catfish, respectively (Karami et al. 2010; Normala et al. 2017). This oval shape or increased erythrocyte size has been explained as a biological response to the increment of one chromosome set in the triploid fish (Benfey 1999; Normala et al. 2016). This phenomenon has also been demonstrated in the triploids of Wels catfish, Silurus glanis (Flajšhans 1997), pondloach, Misgurnus anguillicaudatus (Gao et al. 2007), Caspian trout, Salmo caspius (Dorafshan et al. 2008) and red hybrid tilapia, Oreochromissp. (Pradeep et al. 2011). The observation of erythrocytes within the exclusive triploid ranges of the cell major axis (11.9–14.9 μm) as recommended by Normala et al. (2016) justifies the suitability and reliability of this method as a simpler index of discrimination.

In conclusion, this study highlights the differences in the timing and sequence of embryogenetic development between diploid and triploid African catfish. This advances possible scientific explanations for the observed breeding performance of both groups. Future studies can focus on comparative assessment of different shock protocols (physical or chemical) so as to elucidate the extent of embryonic alterations, and hence identify the most suitable shock process for optimum production of triploid African catfish C. gariepinus.

Acknowledgments

The authors are indebted to the School of Fisheries and Aquaculture Science, Universiti Malaysia Terengganu, Malaysia for providing broodstocks of C. gariepinus used in this study. We also acknowledge the help of some technical staffs of the PPSPA hatchery department during experimental trials of this study.

Disclosure statement

No potential conflict of interest was reported by the authors.