By Charles Gregory Lutz1, Alvaro M Armas-Rosales2 & ArnoldMSaxton3
Early attempts to combine growth and conformation traits from other species with the orange-red coloration originally observed as a recessive trait in O. mossambicus resulted in red hybrids and synthetic varieties derived from them, and over time interest in their salinity tolerance has increased. Commercial tilapia varieties that combine salinity tolerance and superior growth can allow for economically feasible production in many coastal and desert regions, but an understanding of the genetic effects involved is a prerequisite for their development.
The goal of this study was to examine the salinity tolerance of six varieties of Oreochromis exhibiting a broad range of salinity tolerance, and their reciprocal crosses, in a full diallel mating design and to develop appropriate statistical procedures to evaluate the influences of various genetic effects on this trait. These varieties included a line of blue tilapia originally from Lake Manzala Egypt (hereafter referred to as BL), the synthetic variety Florida red tilapia (FL), another synthetic variety descended in large part from the Rocky Mountain White Tilapia® and referred to as Mississippi Commercial strain (MC), a line of O. mossambicus resulting from random mating between two South African strains (MO), O. niloticus descended from the Auburn-Egypt strain (NI), and O. niloticus F1 crossbreds of NI females with males from the Stirling red Nile line (RE).
Over two consecutive summers, randomly selected fish from all parental varieties were used to produce diallel crosses, one each year, at the Louisiana State University Agricultural Center’s Aquaculture Research Station in Baton Rouge, Louisiana. Males of each variety were stocked with females of each variety, such that thirty-six outdoor fiberglass tanks were stocked with 2-3 females per male, with eight to 12 fish per pool, depending on individual weights. Fish were allowed to spawn of their own volition and females incubated eggs and fry within the tanks. Beginning at approximately 45 d post-stocking, samples of fingerlings from each cross were collected at 15- to 30-day intervals. A total of five trials were conducted over 13 months to estimate salinity tolerances of juveniles from parental varieties and their reciprocal crosses. A recirculating system was used for salinity tolerance trials (see illustration).
Mortality data were analyzed as interval sensitive. In the absence of significant year, trial or size (total length) effects, salinity tolerance data from the two diallels were pooled and each trial was considered a replicate. The genetic model for statistical analysis of offspring included line effects (li) defined as the average direct (transmitted) genetic effects of parental varieties, and maternal effects (mi) similarly defined as the average maternal genetic influence of each variety, each with an expected sum of zero. Reciprocal effects (rij) were defined as differences between reciprocal crosses, while specific reciprocal effects (r**ij) represented reciprocal effects less average maternal differences between parental varieties.
Heterosis in a cross between parental varieties (hij) was partitioned into overall heterosis contributed by the varieties included in the diallel ( ), plus the direct heterosis of each variety in the cross as a deviation from overall heterosis (hi or hj), and the specific heterosis [sij or specific combining ability (SCA)] of the variety combination in the cross. General combining ability (GCA) of any given variety, or the average performance of the variety in crosses with other varieties, was defined as the sum of one-half of the variety’s line effect and its direct heterosis, also with an expected sum of zero across all parental varieties.
A total of 2,205 F1 juveniles were evaluated for salinity tolerance. MST ranged from 25.0 ppt (NI) to 48.7 ppt (FL) among parental varieties, and cumulative survival curves of parental varieties indicated three distinct survival patterns, with MO and FL grouped together, BL alone, and MC, NI and RE together. Results indicated FL and MO dams would generally produce offspring with mean salinity tolerances of 45 ppt or greater. Line effect estimates indicated that offspring of BL, FL or MO were significantly more salinity tolerant than those of MC, NI and RE, and GCA estimates indicated that FL would contribute to the most salinity tolerant combinations among parental varieties.
Mean heterosis (h) for salinity tolerance was highly significant, and crosses between parental varieties were on average 4.46 ppt more tolerant than parental varieties. Ten of 15 variety combinations exhibited significant or highly significant hij estimates. Reciprocal effect estimates highlighted probable differences in gene frequencies among varieties and specific reciprocal effects were almost equally influential.
The FL variety used in this study tolerated up to 77 ppt, similar to previous reports for this synthetic line. Our MO variety tolerated up to 84 ppt, with an estimated MST of 46 ppt. In contrast with O. mossambicus and Florida Red tilapia, the poor salinity tolerance of O. niloticus is widely recognized. Among parental varieties, NI exhibited the lowest MST (25 ppt) in this study. Nonetheless, crossing NI males with FL females resulted in the highest MST estimate (52.5 + 3.14 ppt) for any cross in the study, highlighting the influence of dominance effects on this trait in certain crosses.
For those wishing to develop their own salt-tolerant lines of tilapia, varieties with the highest GCAs would also be expected to exhibit the highest net crossing effects. Results suggest that maternal effects can also be utilized in development of salt tolerant stocks. Although NI and RE varieties (both O. niloticus) were virtually indistinguishable in terms of salinity tolerance and cumulative mortality, offspring from RE females were on average 7.7 ppt more salinity tolerant than those of NI, presumably due to heterotic maternal effects.
Genetic sex ratios could also be an important consideration when evaluating tilapia stocks for salt water grow out. As a result of combined genetic effects the MO x BL cross was among the most salinity tolerant in these trials (MST of 51 ppt) and appeared to produce predominantly male juveniles, but this finding could not be considered conclusive. An aquaculture facility capable of conducting a simple 3 x 3 diallel design may obtain sufficient performance information to improve salinity tolerance based only on cross means. In the absence of software to statistically estimate genetic effects, cumulative mortality curves are also useful in identifying potential heterosis and maternal effects within variety combinations.
Since approximately 50% of heterosis is retained when individuals within a line mate at random in the F1 and subsequent generations, heterosis for salt tolerance exhibited in many of the crosses in this study emphasizes its potential importance in the formation of synthetic strains of salt tolerant tilapia. The persistence of a significant portion of positive heterosis in the original hybrid cross could partially explain the superior salt tolerance of the Florida Red synthetic line.
Evaluation of salinity tolerance in juveniles from diallel mating designs provided clear evidence of additive, dominance and maternal genetic effects influencing this trait in all crosses. All these influences must be taken into account in the development of breeding programs to combine salinity tolerance with superior production characteristics.
1 Aquaculture Research Station, Louisiana State University Agricultural Center, Baton Rouge, LA, USA
2 Coastal Fisheries Institute, Louisiana State University, Baton Rouge, LA, USA
3 Department of Animal Science, University of Tennessee, Knoxville,TN, USA