K. Tessanne, H. C. Hines, and M. E. Davis1
The Ohio State University Department of Animal Sciences
The ob gene, which produces the 16-kilodalton protein leptin, is believed to be involved in the regulation and deposition of fat. In this study, two microsatellites and two restriction fragment length polymorphisms were used to identify polymorphisms in the bovine leptin genes of The Ohio State University beef cattle research herd, located at the Eastern Ohio Resource Development Center (EORDC). The microsatellites were those described by Wilkins and Davey (1997) and by Stone et al. (1996 [GenBank Accession No. G18586]), while the PCR-RFLPs were those characterized by Lien et al. (1997). Polymerase Chain Reaction (PCR) amplification of DNA from 160 beef bulls was performed using fluorescent primers designed for both of the microsatellites, as well as for the PCR-RFLPS. PCR products amplified with the PCR-RFLP primers were subjected to restriction enzyme digestion using Hinfl and BsaAI restriction endonucleases. Samples of the microsatellite PCR products, combined with the enzyme digestion products, were examined with an ABI 377 DNA Sequencer (Perkin-Elmer/Applied Biosystems), using GENESCAN 672 software. The genotypes were then determined with the aid of Perkin-Elmer/ABI Genotyper software. After all animals were genotyped, carcass data collected on the EORDC cattle were analyzed using the General Linear Models procedure of the SAS computer program. Significant differences in allelic frequencies between the high and low IGF-I selection lines were found at the Wilkins & Davey microsatellite (WD-MS) and Stone microsatellite (Stone-MS) sites, but not at the PCR-RFLP site. The WD-MS genotypes had a significant effect on ribeye area (P < 0.02). However, no other statistically significant relationships were found. Noteworthy relationships (P < 0.20) were found between both the Stone-MS and the PCR-RFLP genotypes and marbling score and between the PCR-RFLP genotypes and kidney, pelvic, and heart fat (KPH) percentage.
Leptin is a 16-kilodalton protein produced by the obesity (ob) gene. It was first discovered in mice by Jeffrey Friedmann’s research team at Rockefeller University in New York City in 1994 (Gura, 1997). Defects in both the mutant and wild-type copies caused the animals to act as if in starvation, depositing up to three times the body fat as mice without defective copies. Leptin was traced back to fat cells, called adipocytes, where it is produced and then released into the bloodstream. In addition, leptin receptor locations have been identified in the brain, also suggesting that leptin plays a role in appetite regulation and weight control. Specifically, leptin receptors are expressed in four main regions of the hypothalamus, including two that have been previously noted for involvement in feeding and metabolism (the articulate nucleus and the paraventricular nucleus) (Gura, 1997). Although it has been determined that leptin is related to fat deposition, its specific role is not yet known.
Even though the specific cause of fat deposition in animals is not yet known, the involvement of leptin in this process is thought to be very important in both humans and animals. In the bovine species, the leptin gene has been mapped to chromosome four (Stone et al., 1996; Pomp et al., 1997). The availability of highly polymorphic markers within and adjacent to the bovine leptin gene will facilitate genetic studies in cattle to determine the specific roles leptin plays. A study by Wilkins and Davey (1997) revealed a polymorphic microsatellite in the bovine leptin gene, this being microsatellite 5’ AAAAAAAAAAAAAAAATATATATATATATATATATA3’ in the 5' UTR region of the bovine leptin gene. This fragment corresponds to nucleotides 247 to 282 in the deposited sequence (GenBank Accession No. U50365). The microsatellite proved to be highly polymorphic, with 18 alleles revealed after genotyping 97 individuals (Wilkins and Davey, 1997).
Additionally, another microsatellite and several point mutations have also been detected in the bovine leptin gene. A microsatellite of the composition 5' GATA(CA)nCTAG 3' has been detected in the DNA flanking the 3' end of the obesity gene (Stone et al., 1996 [GenBank Accession No. G18586]). This microsatellite has at least four alleles (Fitzsimmons et al., 1998). The two most polymorphic point mutations are readily detected by restriction enzyme digestion of segments amplified by the polymerase chain reaction (PCR) (Lien et al., 1997). Polymorphisms such as these are referred to as PCR-amplified restriction fragment length polymorphisms (PCR-RFLPs).
The ob gene is believed to be involved in the regulation and deposition of fat tissue. Therefore, variations in it may make significant contributions to individual differences in cattle growth and carcass characteristics. In this project, the microsatellites described by Wilkins and Davey (1997) and Stone et al. (1996 [GenBank Accession No. G18586]), together with the PCR-RFLPs characterized by Lien et al. (1997), were used to identify polymorphisms in the bovine leptin genes of 139 bulls in the Ohio State University Angus beef cattle research herd located at the Eastern Ohio Resource Development Center (EORDC) in Belle Valley, Ohio. Extensive research is ongoing in this herd concerning selection for blood-serum insulin-like growth factor I (IGF-I) concentrations (Davis et al., 1995; Davis and Simmen, 1997). DNA samples collected for that project were analyzed in this study.
Justification for finding and using genetic markers in the bovine leptin gene and other fat/growth associated gene(s) lies in the economic importance of carcass traits desired for beef cattle. Public demands for leaner, higher quality meat govern the direction of the beef industry. If a polymorphic genetic marker for the leptin gene is identified and the role leptin plays in fat deposition is determined, research on genetically controlling the fat content of meat would progress greatly. Cattle with leaner body composition could then be selected using markers at the leptin locus, resulting in carcasses with either a higher or lower marbling score and a higher quality grade.
In this project, three objectives were identified: (1) to determine if the Angus bulls at EORDC were polymorphic for three segments of the bovine leptin gene, (2) to evaluate the allelic frequencies in the line selected for high IGF-I concentrations and in the line selected for low IGF-L, and (3) to evaluate the relationship of detected marker variation with differences in the carcass traits of these Angus bulls.
In this study, DNA previously collected on 139 Angus bulls from the EORDC beef cattle research herd in Belle Valley was analyzed to determine polymorphisms in their leptin genes. This DNA was on storage in the Animal Genetics Lab in the Department of Animal Sciences at Ohio State
University. Bulls were chosen from both the line selected for high IGF-I concentration and the line selected for low IGF-I concentration. The bulls used were born in the following seasons: Fall 1995, Spring 1996, Fall 1996, and Spring 1997.
The first step was to perform a polymerase chain reaction (PCR) amplification on each sample of DNA from these bulls. The components of this reaction were as follows:
100 ng/µl template DNA, 10X PCR standard reaction buffer, 125 m of dNTPs, 10 pmoles of each specific primer, and 1.0 µl of Taq polymerase.
Primers used for each marker included:
Wilkins & Davey Microsatellite (WD-MS):
Forward: 5' TTGTAATCCTGCAATATC TTGTCC 3'
Reverse: 5' TAAACAGGCCGTAGCA GTACAG 3'Stone Microsatellite (STONE-MS):
Forward: 5' GATGCAGCAGACCAA GTGG 3'
Reverse: 5' CCCAT17GCTAGAACCCAGG 3'PCR-RFLP:
Forward: 5' GGCTGGACGCAAAGGGC AGAGT 3'
Reverse: 5' CCCTGACGCCGCATTTCCCTA 3'
All primer sets were amplified in separate reactions. The PCRs were performed using a Perkin-Elmer thermocycler. The samples were subjected to 31 amplification cycles consisting of denaturation at 94°C for 30 seconds, annealing at 58°C for 30 seconds, and extension at 72°C for one minute. The final extension step was lengthened to seven minutes to ensure that polymerization was complete. The samples were run on an agarose gel to confirm amplification.
The PCR products amplified with the PCR-RFLP primers were then digested using Hinfl and BsaAl restriction endonucleases. The digestion included 4 µl of PCR sample, 0.3 µl of enzyme, 1 µl of specific buffer, and 4.7 µl of water. The samples were incubated for two hours at 37°C, after which the reaction was stopped by adding 4 µl of loading buffer.
Following completion of restriction enzyme digestion, alliquots of the different polymorphic products were combined as follows: 9 µl of WD-MS, 4 µl of Stone-MS, and 9 µl of PCR-RFLP. The mixed samples were transported to Dr. Christopher Weghorst’s laboratory in The James Cancer Hospital at Ohio State University. In this laboratory, 6 µl of each sample were added to 4 µl of formamide and 0.5 µl of ROX internal lane size standard (GS2500 Applied Biosystems). The samples were heat-denatured at 95°C for four minutes, chilled on ice, and loaded onto a standard 6% polyacrylamide denaturing sequencing gel in an ABI 377 DNA Sequencer (Applied Biosystems). The PCR products were detected by fluorescent dyes attached to the specific PCR primers. The dyes were fam (blue) for WD-MS, tet (green) for Stone-MS, and hex (yellow) for PCR-RFLPS. The PCR products were automatically sized by reference to the internal lane standard, results were transferred to the Animal Genetics Lab, and genotypes were determined there with the aid of Perkin-Elmer/ABI Genotyper software.
Once genotypes were determined, allelic frequencies were calculated. Allelic frequencies for each selection line were computed simply by dividing the counts for each allele in each selection line by the total number of alleles found in that line. Total allele frequencies were also calculated by dividing the count of a specific allele by the total number of alleles in each group (WD-MS, Stone-MS, or PCR-RFLP). Chi-square analysis was employed to evaluate differences in allelic frequencies between selection lines. A probability level of 0.05 was considered indicative of significance.
The genotypes were encoded numerically and their effects on carcass traits were determined using General Linear Models procedures (SAS, 1985). The carcass traits included backfat thickness; ribeye area; kidney, pelvic and heart fat percentage; hot carcass weight; curability; marbling; and yield grade. The statistical model included fixed effects of genotype, birth year, season of birth, IGF-I selection line, and age of dam. A covariate for on-test age of calf was also included in the model.
At the WD-MS mutation site, six alleles were found in 137 Angus bulls, these being detected as amplified segments that were 177, 184, 186, 190, 192, and 209 base pairs in length. Chi-square analysis of the frequencies of these alleles indicated significant differences between the high and low IGF-I selection lines (P < 0.0l; Table 1).
Table 1. Allele Frequencies and Chi-Square Analysis of Allele Frequencies. |
|||
|---|---|---|---|
| W-D Microsatellite*: | |||
| Allele | Frequency in High Line | Frequency in Low Line | Overall Frequency |
| 177 | 0.51 | 0.63 | 0.57 |
| 184 | 0.15 | 0.20 | 0.17 |
| 186 | 0.05 | 0.01 | 0.04 |
| 190 | 0.02 | 0.04 | 0.03 |
| 192 | 0.30 | 0.11 | 0.17 |
| 209 | 0.05 | 0.00 | 0.03 |
| Stone Microsatellite: | |||
| Allele | Frequency in High Line | Frequency in Low Line | Overall Frequency |
| 135 | 0.49 | 0.36 | 0.44 |
| 144 | 0.51 | 0.64 | 0.56 |
| PCR-RFLP: | |||
| Allele | Frequency in High Line | Frequency in Low Line | Overall Frequency |
| B | 0.44 | 0.47 | 0.45 |
| C | 0.56 | 0.53 | 0.55 |
| Chi-Square Analysis: | |||
| Locus | Observed X2 Value | Degrees of Freedom | P Value |
| WD-MS** | 16.37 | 5 | P < 0.0l |
| Stone-MS** | 4.51 | 1 | P < 0.05 |
| PCR-RFLP | 0.31 | 1 | P = 0.55 |
| *W-D for Wilkins & Davey, who first discovered
the microsatellite. **MS = microsatellite. |
|||
For the Stone microsatellite, 131 bulls were analyzed. Two alleles were found, and these were detected as amplified segments that were 135 and 144 base pairs in length. Neither of these alleles have been previously reported in the literature. Chi-square analysis detected significant differences in allele frequencies between the high and low IGF-I selection lines (P < 0.05; Table 1).
At the PCR-RFLP mutation site, two alleles were found in 139 bulls. The B allele was associated with an uncut fragment length of 170 base pairs, while the C allele was represented by two cut fragments, with lengths of 81 and 89 base pairs. Chi-square analysis of the frequencies of these alleles in the high and low IGF-I selection lines showed no significant difference between the lines (P = 0.55; Table 1).
Five previously unreported alleles were found during the course of this experiment. At the WD-MS site, new alleles consisting of 190, 192, and 209 base pairs were found, and at the Stone-MS site, new alleles consisting of 135 and 144 base pairs were found. These two new alleles were the only ones found at the Stone-MS locus in these Angus bulls.
Table 2. Least Squares Means and Standard Errors of Carcass Traits by Genotype. |
|||||||
|---|---|---|---|---|---|---|---|
| W-D Microsatellite**: | |||||||
| Dependent Variable | |||||||
| Independent Variable | Fat, mm | Ribeye, cm2 | KPH*, % | HCW*, kg | Cutability, % | Marbling | Yield Grade |
| Genotype (n = 137) | P = 0.69 | P < 0.02 | P = 0.90 | P= 0.25 | P = 0.74 | P = 0.51 | P = 0.72 |
|
177/177 |
8.9 ± 0.6 | 76.4 ± 1.3 | 2.2 ± 0.1 | 277.8 ± 5.00 | 51.3 ± 0.2 | 4.7 ± 0.1 | 2.4 ± 0.1 |
|
177/190 or 177/192 |
9.2 ± 0.8 | 73.8 ± 1.8 | 2.2 ± 0.2 | 264.2 ± 6.80 | 51.3 ± 0.3 | 4.9 ± 0.2 | 2.4 ± 0.1 |
|
177/184 |
9.3 ± 0.9 | 80.6 ± 1.9 | 2.2 ± 0.2 | 283.4 ± 7.10 | 51.6 ± 0.3 | 4.8 ± 0.2 | 2.2 ± 0.1 |
|
190/190 or 192/192 |
8.8 ± 1.0 | 71.9 ± 2.1 | 2.1 ± 0.2 | 261.1 ± 7.80 | 51.3 ± 0.4 | 4.7 ± 0.2 | 2.4 ± 0.2 |
|
184/194 or 194/192 |
9.3 ± 0.9 | 79.1 ± 2.0 | 2.1 ± 0.2 | 274.1 ± 7.50 | 51.7 ± 0.4 | 4.7 ± 0.2 | 2.2 ±0.2 |
|
192/209 |
4.8 ± 2.5 | 80.6 ± 5.5 | 1.7 ± 0.5 | 282.6 ± 20.6 | 53.0 ± 1.0 | 4.9 ± 0.5 | 1.7 ± 0.4 |
|
209/209 |
7.0 ± 2.3 | 84.2 ± 4.9 | 1.7 ± 0.4 | 301.1 ± 18.3 | 52.4 ± 0.9 | 5.4 ± 0.5 | 1.9 ± 0.4 |
|
177/186 |
8.8 ± 1.7 | 79.9 ± 3.7 | 2.1 ± 0.3 | 283.2 ± 13.8 | 51.8 ± 0.7 | 4.7 ± 0.4 | 2.2 ± 0.3 |
|
186/186 |
5.6 ± 2.0 | 73.5 ± 4.4 | 1.8 ± 0.4 | 272.4 ± 16.4 | 52.1 ± 0.8 | 4.0 ± 0.4 | 2.1 ± 0.3 |
| Stone Microsatellite: | |||||||
| Dependent Variable | |||||||
| Independent Variable | Fat, mm | Ribeye, cm2 | KPH*, % | HCW*, kg | Cutability, % | Marbling | Yield Grade |
|
Genotype (n = 131) |
P = 0.68 | P = 0.52 | P = 0.54 | P = 0.89 | P = 0.31 | P = 0.15 | P = 0.31 |
|
135/135 |
7.7 ± 0.9 | 77.8 ± 1.9 | 2.0 ± 0.2 | 271.9 ± 7.2 | 52.0 ± 0.3 | 4.7 ± 0.2 | 2.1 ± 0.1 |
|
135/144 |
8.2 ± 0.6 | 76.5 ±1.4 | 2.1 ± 0.1 | 269.6 ± 5.5 | 51.8 ± 0.3 | 4.8 ± 0.1 | 2.2 ± 0.1 |
|
144/144 |
8.4 ± 0.7 | 75.6 ± 1.7 | 2.2 ± 0.1 | 272.3 ± 6.3 | 51.5 ± 0.3 | 4.5 ± 0.0l | 2.3 ± 0.1 |
| PCR-RFLP: | |||||||
| Dependent Variable | |||||||
| Independent Variable | Fat, mm | Ribeye, cm2 | KPH*, % | HCW*, kg | Cutability, % | Marbling | Yield Grade |
|
Genotype (n = 139) |
P = 0.30 | P = 0.50 | P = 0.20 | P = 0.95 | P = 0.25 | P = 0.19 | P = 0.25 |
|
B/B |
9.2 ± 0.8 | 74.3 ± 1.8 | 2.3 ± 0.2 | 271.5 ± 7.0 | 51.1 ± 0.3 | 4.5 ± 0.2 | 2.5 ± 0.1 |
|
B/C |
8.2 ± 0.6 | 76.6 ± 1.3 | 2.0 ± 0.1 | 272.7 ± 4.9 | 51.7 ± 0.2 | 4.8 ± 0.1 | 2.2 ± 0.1 |
|
C/C |
9.1 ± 0.7 | 76.8 ± 1.5 | 2.2 ± 0.1 | 270.9 ± 5.8 | 51.5 ± 0.3 | 4.6 ± 0.1 | 2.3 ± 0.1 |
| * KPH = Kidney, Pelvic, and Heart Fat, HCW
= Hot Carcass Weight. ** W-D for Wilkins and Davey, who first discovered the microsatellite. *** B = 170 bp, C = 81 & 89 bp. |
|||||||
Analysis of variance showed that the WD microsatellite had a significant effect (P < 0.02) on ribeye area (Table 2). It is important to note also those traits for which P < 0.20 because of the trends that they may reveal. These traits were marbling score for both the Stone microsatellite and the PCR-RFLP (P= 0.15 and 0.19, respectively), and KPH for the PCR-RFLP (P = 0.20). No other important relationships between leptin genotypes and carcass traits were found.
It was surprising to find only one significant relationship between leptin genotype and carcass traits, because leptin has been shown to be involved in fat deposition (Gura, 1997). This lack of significance may be due to several causes.
First, the animals studied were all purebred Angus bulls, and the sample therefore lacked the genetic variation that would be expected of a group of crossbred animals or of a diverse group of breeds. In the experiment by Wilkins and Davey (1997) in which the polymorphic WD-MS locus was discovered, the sample consisted of different breeds of dairy cattle. Also, in another investigation by Pomp et al. (1997), the sample consisted of Limousin, Simmental, Holstein, Gelbvieh, Hereford, Angus, Brahman, and Brangus breeds. This sample was used to find a PCR-RFLP with Sau3AI restriction endonuclease (Pomp et al., 1997). It is recommended that future studies of the leptin gene include diverse breeds, thereby increasing the genetic variation within the sample.
Secondly, genotypes of only 139 Angus bulls were analyzed. The sample size was small due to the limited number of DNA samples available from bulls that had already been slaughtered and from which carcass data had been obtained and recorded. In future studies, a larger sample size may reveal more significant leptin genotype-carcass trait relationships.
Finally, the exact role leptin plays in fat deposition has not yet been determined. The carcass traits examined in this experiment may have been too superficial to elucidate the effects of the leptin genotypes on the biology of the animals studied. Comparison of leptin polymorphisms and actual serum leptin concentrations may be more revealing. This comparison is especially recommended because, although only one significant relationship was found between leptin genotypes and carcass traits, two of the sites revealed significant allelic frequency differences between the high and low IGF-I selection lines. This result may be an indicator of other effects of leptin genotypes in the bovine system.
Alberts, B., Bray, D., Lewis, J., Raff, M., Roberts, K., and Watson, J. D. Molecular Biology of the Cell. 3rd Ed. 1983.
Applied Biosystems. Best of the Breed Determined Automatically Using Microsatellite Markers. Biosystems Reporter. Issue No. 22, April 1994.
Applied Biosystems. GENESCAN 672 Software Performance Specifications.
Davis, M. E., Bishop, M. D., and Simmen, F-C. M. 1995. Divergent selection for blood serum insulinlike growth factor I concentration in beef cattle: 1. Non-genetic effects. J. Anim. Sci. Vol. 73, p. 1927—1932.
Davis, M. E. and R. C. M. Simmen. 1997. Genetic parameter estimates for serum insulin-like growth factor I concentration and performance in Angus beef cattle. J. Anim. Sci. Vol. 75, p. 317—324.
Fitzsimmons, C. J., Schmutz I, S. M., Bergen, R. D., and McKinnon, J. J. 1998. Effects of the obese gene in beef cattle on carcass fat characteristics. Plant and Animal Genome. VI, Jan 18-22, 1998, San Diego, Calif.
Glowatzki-Mullis, M-L, Gaillard, C., Wigger, G. and Fries, R. 1995. Microsatellite-based parentage control in cattle. Animal Genetics. Vol. 26, p. 7—12.
Gura, T. 1997. Obesity sheds its secrets. Science. Vol. 275, 7 February.
Lien, S., Sundvold, H., Klungland, H., and Vage, D. I. 1997. Two novel polymorphisms in the bovine obesity gene (OBS). Animal Genetics. Vol. 28, p. 245.
Pomp, D., Zou, T., Clutter, A. C., and Barendse, W. 1997. Rapid communication: Mapping of leptin to bovine chromosome 4 by linkage analysis of a PCR-based polymorphism. J. Anim. Sci. Vol. 75, p. 1427.
SAS. 1985. SAS User’s Guide: Statistics. Version 5 Edition. SAS Institute, Inc., Cary, NC. p. 433—506.
Stone, R. T., Kappes, S. M., and Beattie, C. W. 1996. The bovine homologue of the obese gene maps to chromosome 4. Mammalian Genome. Vol. 7, p. 399—400.
Wilkins R. J. and Davey, H. W. 1997. A polymorphic microsatellite in the bovine leptin gene. Animal Genetics. Vol. 28, p. 370—383.
1 For more information, contact at: The Ohio State University, 221 Plumb Hall, 2027 Coffey Road, Columbus, OH 43210; (614) 292-4984, Fax (614) 292-7116; email:davis.28@osu.edu