Potential Therapeutic Uses of Bovine Somatotropin in Cattle

 

Potential Therapeutic Uses of Bovine Somatotropin in Cattle


Robert J. Collier and John L. Vicini

Protiva, a Unit of Monsanto Company,
700 Chesterfield Parkway North/BB3F,
St. Louis, MO, USA, 63198.
E-mail: jcoll@ccmail.monsanto.com

Take Home Messages

Somatotropin is best known for its role in regulating milk production in cattle. However, this hormone has beneficial effects on other tissues which may provide therapeutic and prophylactic uses for it in the future.

Introduction

Bovine somatotropin (bST) has now been marketed continuously for five years in Brazil and Mexico and for two years in the United States. A considerable body of data now exists on its effects under a wide variety of management and environmental conditions in all the major cattle breeds. Concerns about catastrophic health issues in cattle treated with somatotropin have not materialized. Furthermore, consumers in these countries have accepted use of this technology in the dairy industry. In addition, new concepts for use of somatotropin in dairy cattle have been proposed. The objective of this paper will be to describe these uses and to explain their biological basis. The majority of these uses will require additional work to obtain regulatory approval in each country where the use may be considered commercially viable.

Somatotropin and Stress

A common misconception regarding use of bST in dairy cows is that these animals are "stressed" by the higher levels of milk yield they achieve when being supplemented. Stress is normally measured by the degree of "strain" it exerts on a system. In the case of metabolic systems this typically is measured by a change in metabolic rate. Thus, degree of heat and cold stress can be characterized by the change in basal metabolic rate induced by the stress. In the case of somatotropin, careful bioenergetic studies have demonstrated that basal metabolic rate of somatotropin-treated cows is unaltered and that nutritional requirements of bST-treated cows are the same as those for untreated cows and are a function of the animal's maintenance requirement, body condition, and requirements for milk synthesis (17, 39). This is in clear contrast to the impact of thyroid compounds (e.g. thyroprotein) which do change metabolic rate. Milk yield increases to these compounds are associated with losses in body condition and eventually milk yield (3).

Another way that stress is measured in domestic animals is to evaluate production. Typically, animals that are sick or suffering discomfort demonstrate clear decreases in production and/or production efficiency. This is clearly not the case with somatotropin treatment which results in increases in production and biological efficiency of animals. Speculation that use of somatotropin would cause cows to burn out were based on incorrect assumptions regarding the mechanism of action. As pointed out by Bauman and McGuire (3), "Metabolic disorders would most likely occur the first few days of bST treatment when milk yield has increased but intake has not. Suffice to say, there is not a single mention of clinical ketosis or milk fever occurring during the first weeks of bST treatment in any of the hundreds of published studies". Even when cows are not adequately fed we do not see development of a disease state with onset of somatotropin treatment. Underfed cows which do not have available body stores to increase production when treated with bST demonstrate a negligible milk yield response (4, 12, 23). Thus, cows are not "forced" to produce milk if they are not nutritionally capable of responding to somatotropin treatment.

An additional method of assessing impact of potential stressors on domestic animals is to examine the function of the hypothalamic-pituitary axis. This is based on evidence that under both acute and chronic stress the central nervous system of mammals evokes physiological responses that culminate in changes in secretion rate of hormones of the sympatho-adrenal axis (31). Typically, one can measure increases in secretion of adrenal corticotrophic hormone and glucocorticoids of the adrenal. The general endocrine status of somatotropin-treated animals was evaluated in the acute and chronic toxicology studies as well as an endocrine challenge study during a fourth consecutive lactation of bST treatment (1, 2). In all of these studies there was no evidence of increased secretion of glucocorticoids in animals treated with somatotropin. In fact, the only consistent evidence was a slight decrease in glucocorticoid concentration in somatotropin-treated animals that was attributed to metabolic adaptation to increased gluconeogenesis in treated animals. In addition to the hypothalamic adrenal axis, the pituitary challenge study indicated that four consecutive lactations of somatotropin treatment did not alter ability of the pituitary to respond to endocrine challenge.

Finally, chronic stress is recognized to be immune suppressive (31). Animals that are immune suppressed are more susceptible to disease. The exact cause of the immune suppression still remains undefined, but is believed to be related in part to increased secretion of adrenal glucocorticoids which suppress many immune functions. Since glucocorticoids are not increased in somatotropin-treated animals, one prerequisite for immune suppression would appear to be missing. Immune function of somatotropin-treated animals has been measured using a variety of approaches. The immune system is sensitive to somatotropin stimulation with a number of studies demonstrating improved immune function in somatotropin-treated animals. The effects of somatotropin include improved thymus weight and thymosin concentration in plasma (21), improved primary lymphoid tissue in bone marrow, stimulated activation of peripheral lymphocytes and macrophages (18), improved cytokine responses to challenge (20), improved antibody synthesis (36), improved IgG level (29), and augmented production of superoxide anion by macrophages and neutrophils (33). Vandeputte-Van Messom et al. (37) demonstrated improved production recovery from E. coli mastitis in lactating dairy cows when cattle were treated with bovine somatotropin. Thus, all of the evidence gathered to date does not support a negative effect of somatotropin on immune function. Quite to the contrary, the accumulated evidence supports a stimulatory role of somatotropin in regulation of the immune system. Evaluation of the mastitis incidence and duration of somatotropin-treated cows indicates that duration of mastitis was not altered in cows treated with rbST. Thus, there is also no evidence that acute or chronic treatment of lactating dairy cows results in suppression of the immune system.

In conclusion, the biology of bST in dairy cows has been extensively studied and collectively these studies do not support the hypothesis that bST induces a stress in lactating dairy cattle. There is no change in basal metabolic rate, there is no activation of the hypothalamic-adrenal axis, productive efficiency does not decline, it increases and finally, the immune system is not suppressed, in fact there is evidence that it is enhanced. Therefore, there is no basis in fact to support a contention that cattle treated with bST are stressed.

Somatotropin and Reproduction

Somatotropin has both direct and indirect effects on reproductive performance of dairy cattle. These effects are further influenced by the energy status of the cow and the stage of the estrus cycle or pregnancy when somatotropin treatment is initiated. This has made it difficult to correctly assess the true impact of somatotropin under a given set of management conditions and physiological state of the animals. In most studies animals are at various stages of the estrus cycle or pregnancy and at different levels of milk yield when somatotropin treatment is initiated. Very few studies have controlled the stage of the estrus cycle or energy state of the animal in order to get good estimates on a parameter. The result is that we have only general observations at this point and we know that there are both negative and positive effects on reproductive performance in cattle treated with somatotropin.

Direct effects of somatotropin on reproductive performance of cattle include enhanced growth of the corpus luteum, increased secretion of progesterone during the luteal phase of the estrus cycle, and a slight extension of the estrus cycle by extending the lifespan of the corpus luteum (26). These direct effects of somatotropin on function of the corpus luteum are explained by the presence of abundant somatotropin receptors in corpora lutea and not follicles (27). Further, it appears that the majority of the somatotropin receptors in the corpus luteum are found in the giant luteal cells and not the small luteal cells (27). The significance of this is unknown. However, increased progesterone and improved lifespan of the corpus luteum have potential positive effects on reproduction by improving the survival rate of the developing embryo. Additionally, increased IGF-1 concentrations in body fluids including plasma and follicular fluid may have beneficial effects on the ovum and developing embryo (22).

A majority of embryos not making it to the fetal stage are believed to be lost due to failure of maternal recognition of pregnancy (28). In other words, the embryo fails to get a sufficiently strong signal to the dam to prevent demise of the corpus luteum and the pregnancy is subsequently terminated. Thus, higher progesterone levels and a slightly longer luteal phase may very likely result in improved embryo viability. This possibility and the potential beneficial effects of increased IGF-1 exposure is supported by reports of increased embryo viability in embryo transfer studies (24) and an increase in twinning in bST-treated cattle (13). This increase in twinning is not consistently supported by increase ovulation rate (32,44) and could be due to an increase in embryo survival although this hypothesis is not yet proven. The percentage of multiple ovulations that naturally occur in cattle is approximately 30 (6). An increase in embryo survival would therefore clearly lead to an increased incidence of twinning.

The improved embryo quality and survival may be beneficial to the embryo transfer industry. However, too little is presently known regarding the viability of embryos from somatotropin-treated cattle to state that this potential is a reality.

The paradox of potential positive effects of bST on reproduction is that increased milk yield and the associated shift in energy balance with onset of somatotropin use adversely impacts the same parameters resulting in reduced progesterone concentrations and even anestrus in severe cases (43). Thus, it is clear that one cannot pursue both objectives (increase milk yield and improve fertility). However, it may be possible to pursue one objective at a time. Low doses of bST which are insufficient to improve milk yield have been reported to improve fertility (35). This approach would remove the negative impact of altered energy balance while obtaining the desired positive effects. Presently, a patent on this approach has been filed (34), but the status of that patent application is unknown. However, this concept deserves further study to determine if there is sufficient improvement in fertility to justify development of a product.

Improved ovulatory response to a gonadotropin challenge is another potential use for somatotropin in embryo transfer cows. Somatotropin advances the second follicular wave when treatment is initiated at the start of the estrus cycle (26). However, it takes two estrus cycles for antral follicles to reach the ovulatory stage. Although there is no evidence to support increased ovulation in somatotropin-treated animals, we do not know what potential exists if additional gonadotropins are administered to somatotropin-primed animals. Additional work needs to be conducted examining the effects of chronic somatotropin treatment (over at least two cycles) on response to superovulatory treatments.

To summarize; sufficient data exist to justify testing the following concepts for somatotropin use in lactating cattle:

Use of low doses in open heifers and lactating cows in early lactation to enhance fertility. There may be some potential for use in anestrus beef animals.
Use in embryo transfer to improve number of viable embryos.
Use in embryo transfer to increase number of ovulations.

Somatotropin to Extend Lactation

At the present time low cull cow prices and high replacement heifer costs have created a significant cash flow squeeze on dairy operations. Presently, it takes four cull cows to generate sufficient income to buy one replacement. At typical culling rates of 30%, it is easy to understand why many dairy farmers are having trouble maintaining their herd size. Keeping cows in the dairy herd can be very profitable if their production can be maintained. Therefore, using somatotropin to extend lactation well beyond 305 days might be a viable commercial opportunity. However, to be significant, this strategy would also need to utilize delayed breeding to avoid negative effects of pregnancy on the lactation performance. Delayed breeding with somatotropin use might then allow the producer to extend the lactation to 610 days rather than the standard 305. Data on lactation performance of somatotropin-treated open cows is now being collected at Cornell University by Dr. Dave Galton to examine this hypothesis. He is evaluating somatotropin use to extend lactation in open cows in several New York dairies and plans to summarize their performance at the American Dairy Science Meetings in 1996.

Somatotropin Use in Mastitis

Although it has been demonstrated that there is a slight increase in mastitis in lactating cows treated with somatotropin (45), this increase is associated primarily with the increase in milk yield. Somatotropin has been shown to have positive effects on the immune system (14). Burvenich et al. (10,11) and Vandeputte-Van Messom and Burvenich (37) demonstrated that somatotropin treatment shortened the mammary gland's recovery time following experimentally induced E. coli infection. Recovery was measured primarily as the rate and extent of restoration of the blood-milk barrier in inflamed quarters. Somatotropin treatment reduced milk sodium and chloride concentrations toward normal levels. Thus, cellular polarity and junctional complexes between lactating mammary cells were restored sooner in glands of cows treated with somatotropin. Other studies have demonstrated improved lymphoblastogenesis, immunoglobulin production, and increased neutrophil killing activity in somatotropin-treated cattle which indicates that somatotropin may improve the immune system response to a bacterial challenge (7,8,9,14). Additional work is needed to determine if there are therapeutic uses for somatotropin in treatment and recovery from mastitis.

Somatotropin for Prevention and Treatment of Metabolic Disease

Peripartum disorders involving periparturient hypocalcemia, reproductive disorders, and ketosis generally occur as a complex. Dystocia, milk fever, mastitis, retained placenta, metritis, LDA, and ketosis have been shown to be inter-related in peripartum dairy cows (15,16). For example, Curtis et al. (16) found that cows which had milk fever were 7.2, 4.0, 5.4, and 23.6 times more likely to suffer dystocia, retained placenta, clinical mastitis, or complicated ketosis (ketosis with an occurrence of at least one other clinical disease present concurrently or previously). Similarly, retained placenta, left displaced abomasum, and milk fever directly increased the risk of complicated ketosis by 16.4, 53.5, and 23.6 times (16).

Ketosis itself is a metabolic disease affecting early lactation dairy cows. Symptoms include decreased milk production, nervousness or lethargy, and inappetance. Metabolic implications include decreased glucose concentration in blood, increased ketones in urine, and increased blood betahydroxybutyrate (25). Recent research suggests that increased fatty liver and decreased glycogen stores may have an effect on whether a cow suffers from clinical ketosis (30,38).

Peripartal treatment with bovine somatotropin has been demonstrated to increase bone calcium mobilization in peripartal dairy cows (19). Although milk fever was not improved in that study, somatotropin's effect on calcium metabolism could have ameliorated the related disorders of displaced abomasum and ketosis. Also of interest is the increase in gluconeogenesis, which is attributed to the direct effect on bST on the liver (5), whereas the increase in milk production is due to stimulation by IGF-1 at the mammary gland. Also, Vicini et al. (42) showed that while lactating dairy cows have relatively high bST levels in early lactation, IGF-1 concentrations are low.

Considerable research will be required to determine if somatotropin can be used as a prophylactic agent in prevention of metabolic disease or whether it may be a successful treatment for certain metabolic diseases such as ketosis and milk fever. However, there is sufficient positive data on somatotropin's mechanism of action to pursue these as real possibilities.

Somatotropin to Accelerate Growth and Age at First Calving

Traditionally, management of replacement heifers has received little attention in comparison to overall dairy farm operations. However, the value of bringing vigorous heifers into production as soon as is economically feasible has received greater attention. The value in this approach is apparent when one considers the large portion of the typical dairy cow's herd-life that is prior to her first lactation (Figure 1) (40).

Feeding high energy diets, by incorporating more concentrates, has improved efficiency of animal performance in most areas of animal agriculture. However, this approach has not been feasible in raising replacement heifers because of reduced milk yields in their first lactation. Lowered blood somatotropin and reduced parenchymal tissue, in mammary glands of animals fed high energy diets prior to puberty, have been observed. These changes may be involved in the etiology of the lowered production of heifers fed high energy dense diets (40).

In a trial previously reported, Vicini et al. (41) (Table 1) administered somatotropin to heifers fed for accelerated growth prior to puberty. Weight gain was increased and days to puberty were reduced by the combination of high energy diets and somatotropin administration. Subsequent milk production of these heifers is shown in Figure 3. In a second trial, Vicini et al. (41) tested to see whether somatotropin would affect weight gain of heifers fed restricted quantities of feed.

Trial 1. Throughout the treatment period mean rates of gain were higher when heifers were fed the accelerated diets and rates of gain were even higher when bST was administered (Table 2). Heifers reached puberty at a target body weight regardless of weight gain. Breeding was done on a pen basis when the average weight was 350 kg. Weight differences were maintained throughout the breeding period and gestation. Accelerated pre-pubertal rates of gain resulted in animals coming into production earlier (Figure 2) with no loss in milk production (Figure 3). Dystocia and other calving difficulties were not affected by treatments (P>.10).

Trial 2. Heifer calves were tested in a similar design with the exception that an additional group was added to determine if bST treatment would increase growth of calves when fed restricted amounts of concentrates at lower rates of body weight gain. The high energy diets increased all measurements of body weight gain (Figure 4, Table 3) and similarly, bST increased body weight, ADG, and girth. Age at breeding was decreased by diet and bST.

To summarize:

Pre-pubertal rates of body weight gain were increased with higher energy diets and were improved further by administration of bST.
Increased ADG was associated with heifers reaching puberty at a younger age.
Milk production was numerically lowest for heifers fed high-energy diets and intermediate when bST was administered; but, these differences were not significant for this number of animals.
Lowering the age of parturition results in greater economic returns due to:
increased feed efficiency,
lowered maintenance costs,
reduction in replacement heifer inventories.

Conclusions

A wide variety of therapeutic and prophylactic uses of somatotropin for dairy cattle are being tested. Final commercial value of these opportunities will be determined by the dairy industry and cost of regulatory approvals relative to their value. However, it is clear that opportunities exist to expand use of somatotropin beyond increasing milk yield. In some countries the potential for therapeutic use may be of more value to the dairy industry than use for increasing milk yield.

Additional Readings

Alberta Dairy Management Fact Sheet:
Growth Hormone - 1. Introduction and Production Response

References

1. Adriaens, F.A., D.L. Hard, M.A. Miller, R.H. Phipps, R.H. Sorbet, R.L. Hintz and R.J. Collier. 1994. Domestic Animal Endo. 12:301-316.
2. Adriaens, F.A., M.A. Miller, D.L. Hard, R.F. Weller, M.D. Hale and R.J. Collier. 1992. J. Dairy Sci. 75:472-480.
3. Bauman, D.E. and M.A. McGuire. 1994. Paradox of BST: Why Cows Don't Burnout. In: Proceedings Minnesota Dairy Health Conference. U. of Minn. St. Paul, MN pp. 27-40.
4. Bauman, D.E. and R.G. Vernon. 1993. Ann. Rev. Nutr. 13:437-461.
5. Bauman, D.E., C.J. Peel, W.D. Steinhour, P.J. Reynolds, H.F. Tyrrell, A.C.G. Brouwn and G.L. Haaland. 1988. J. Nutr. 118:1031-1040.
6. Britt, Jack. Personal communication.
7. Burton, J.L., B.W. McBride, B.W. Kennedy, J.H. Burton, T.H. Elsasser and B. Woodward. 1992. J. Dairy Sci. 74:1589-1598.
8. Burton, J.L., B.W. McBride, B.W. Kennedy, J.H. Burton, T.H. Elsasser and B. Woodward. 1992. J. Dairy Sci. 75:747-755.
9. Burton, J.L., B.W. McBride, B.W. Kennedy, J.H. Burton, T.H. Elsasser and B. Woodward. 1991. J. Dairy Sci. 74:916-928.
10. Burvenich, C., G. Vandeputte-Van Messom, E. Roets, J. Fabry and A.M. Massart-Leen. 1989. Effect of bovine somatotropin on milk yield and milk composition in periparturient cows experimentally infected with Escherichia coli. In: Use of Somatotropin in Livestock Production, Eds. K. Sejrsen, M. Vestergaard and A. Neimann-Sorensen. Elsevier Appl. Sci., New York, NY, pp. 277-284.
11. Burvenich, C., G. Vandeputte-Van Messom, E. Roets, A. Massart-Leen, R.M. Akers, R. Reynart-Van Sichem and L. Devrieze. 1988. Ann. Med. Vet. 132:601-606.
12. Chalupa, W. and D.T. Galligan. 1989. J. Dairy Sci. 72:2510-2524.
13. Cole, W.J., K.S. Madsen, R.L. Hintz and R.J. Collier. 1991. Theriogenology 36:573-595.
14. Comens-Keller, P.G., P.J. Eppard and R.J. Collier. 1995. Evaluation of somatotropin as a homeorhetic regulator of immunity. In: Animal Science Research and Development: Moving Toward a New Century, Ed. M. Ivan. Centre for Food and Animal Research, Agriculture and Agri-Food Canada, Ottawa, ON K1A 0C6, Canada, pp. 79-94.
15. Correa, M.T., C.R. Curtis, H.N. Erb, J.M. Scarlett and R.D. Smith. 1990. J. Dairy Sci. 73:1515-1524.
16. Curtis, C.R., H.N. Erb, C.J. Sniffen, R.D. Smith and D.S. Kronfeld. 1985. J. Dairy Sci. 68:2347-2360.
17. Effect of somatotropin on nutrient requirements of dairy cattle. 1994. In: Metabolic Modifiers: Effects on the Nutrient Requirements of Food Producing Animals. National Academy Press, Washington, D.C. pp. 23-29.
18. Elsasser, T.H. and N.C. Steele. 1992. Growth hormone directed nutrient partitioning: Immune system-pituitary gland communication. In: Pennington Cent. Nutr. Ser. 2, Science of Food Regulation. pp. 164-186.
19. Eppard, P.J., J.J. Veenhuizen, W.J. Cole, P.G. Comens-Keller, G.F. Hartnell, R.L. Hintz, P.K. Olsson, R.H. Sorbet, T.C. White, C.A. Baile and R.J. Collier. 1995. Effect of bovine somatotropin administered to periparturient dairy cowson incidence of metabolic disease. Submitted J. Dairy Sci.
20. Exon, J.H., J.L. Bussiere and J.R. Williams. 1990. Brain Behav. Immun. 4:118-128.
21. Gelato, M.C. 1993. Trends Endocrinol. Metab. 4:106-110.
22. Herrler, A., R. Einspanier, D. Schams and H. Niemann. 1994. Theriogenology 41:601-611.
23. Hoogendoon, C.J., S.N. McCutcheon, G.A. Lynch, B.W. Wickham and A.K.H. MacGibbon. 1990. Anim. Prod. 51:431-439.
24.Kuehner, L.F., D. Rieger, J.S. Walton, X. Zhao and W.H. Johnson. 1993. Theriogenology 40:1003-1013.
25. Littledike, E.T., J.W. Young and D.C. Beitz. 1981. J. Dairy Sci. 64:1465-1482.
26. Lucy, M.C., T.L. Curran, R.J. Collier and W.J. Cole. 1994.Theriogenology 41:561-572.
27. Lucy, M.C., R.J. Collier, M.L. Kitchell, J.J. Dibner, S.D. Hauser and G.G. Krivi. 1993. Biol. Reprod. 48:1219-1227.
28. Lucy, M.C., W.W. Thatcher, J.D. Savio, G. Danet-Desnoyers, M.T. Moser, L. Badinga, F.A. Simmen and R.J. Collier. 1992. J. Anim. Sci. 70:(Suppl.1):271.
29. Marsh, J.A., W.C. Gause, S. Sandhy and C.G. Scanes. 1984. Proc. Soc. Exp. Biol. Med. 175:351-360.
30.Mills, S.E., D.C. Beitz and J.W. Young. 1986. J. Dairy Sci. 69:352-361.
31. Minton, J. E. 1994. J. Anim. Sci. 72:1891-1898.
32.Rieger, D., J.S. Walton, M.L. Goodwin and W.H. Johnson. 1991. Theriogenology 35:863-868.
33. Snow, C.E. 1985. J. Immunol. (Suppl.2):776S-778S.
34. Stanisiewski, E.P. 1993. Somatotropin for increasing reproductivity performance in cattle. Patent Application. WO 9303756 A1.
35. Stanisiewski, E.P., L.F. Krabill and J.W. Lauderdale. 1992. J. Dairy Sci. 75:2149-2164.
36. Takada, Y., H. Bando, Y. Miyamoto, M. Kosaka and T. Sano. 1991. Nippon Naibunpi Gakkai Zasshi. 67:1162 -1177.
37. Vandeputte-Van Messom, G. and C. Burvenich. 1993. J. Dairy Sci. 76:3727-3741.
38. Veenhuizen, J.J., J.K. Drackley, M.J. Richard, T.P. Sanderson, L.D. Miller and J.W. Young. 1991. J. Dairy Sci. 74:4238-4253.
39. Vernon, R.G. 1989. Influence of somatotropin on metabolism. In: Use of Somatotropin in Livestock Production, Eds. K. Sejresen, M. Vestergaard and A. Neimann-Sorensen.Elsevier Applied Science, New York, NY, pp. 31 - 50.
40. Vicini, J.L., P.K. Olsson, B. Barnes and R.J. Collier. 1995. J. Dairy Sci. 78 (Suppl. 1):180.
41. Vicini, J.L., C.A. Baile, G.F. Hartnell and J.J. Veenhuizen. 1993. J. Dairy Sci. 76(Suppl. 1):276.
42.Vicini, J.L., F.C. Buonomo, J.J. Veenhuizen, M.A. Miller, D.R. Clemmons and R.J. Collier. 1991. J. Nutr. 121:1656-1664.
43. Waterman, D.F., W.J. Silvia, R.W. Hemken, G. Heersche, Jr., T.S. Swenson and R.G. Eggert. 1993. Theriogenology 40:1015-1028.
44. Webb, R., J.G. Gong and T.A. Bramley. 1994. Theriogenology 41:25-30.
45. White, T.C., K.S. Madsen, R.L. Hintz, R.H. Sorbet, R.J. Collier, D.L. Hard, G.F. Hartnell, W.A. Samuels, G. de Kerchove, F. Adriaens, N. Craven, D.E. Bauman, G. Bertrand, P. Bruneau, G.O. Gravert, H.H. Head, J.T. Huber, R.C. Lamb, A.N. Pell, R. Phipps, R. Weller, G. Piva, Y. Rijkema, J. Skarda, F. Vedeau and C. Wollny. 1994. J. Dairy Sci. 77:2249-2260.