- Email: [email protected]
Two Avian Luteinizing Hormone Radioimmunoassay Procedures Compared by Measurement of Changes During the Ovulatory Cycle of Turkey and Broiler hens
Two Avian Luteinizing Hormone Radioimmunoassay Procedures Compared by Measurement of Changes During the Ovulatory Cycle of Turkey and Broiler hens J. A. PROUDMAN US Department of Agriculture, ARS, Avian Physiology Laboratory, Beltsville, Maryland 20705 C. G. SCANES
H. OPEL US Department of Agriculture, ARS, Avian Physiology Laboratory, Beltsville, Maryland 20705 M. A. OTTINGER Department of Poultry Science, University of Maryland, College Park, Maryland 20742 (Received for publication October 14, 1983) ABSTRACT Two radioimmunoassay (RIA) procedures for measuring avian luteinizing hormone (LH) were compared using divided aliquots of plasma samples collected during the ovulatory cycles of turkey and broiler breeder hens in three separate experiments. The RIA systems compared were the homologous turkey LH assay of Wentworth et al. (1976) and the homologous chicken LH assay of Follett et al. (1972). Both assays provided the same assessment of circulating LH changes. The preovulatory surge of LH was observed 6 hr before ovulation in turkeys and 4 hr before ovulation in chickens. In addition, both assays detected an increase in chicken LH after ovulation, followed by a gradual decline prior to the next preovulatory surge. The correlation between RIA were highly significant (P<.001) in each experiment. The crossreaction of purified ostrich LH, FSH, and TSH and of the a and /3 subunits of turkey LH was assessed in each assay system. Ostrich LH, FSH, and TSH crossreacted in the chicken LH assay in a dose-related manner: ostrich LH showed the highest activity. Only ostrich TSH produced a response in the turkey assay that was dose related. The a subunit, but not the @ subunit, of turkey LH was quite potent in the chicken LH RIA, but neither subunit competed as well as the native hormone in the turkey RIA. Thus, the two LH antisera recognize different portions of the LH molecule. The chicken antiserum apparently recognizes a site on the a subunit that may be at least partially masked in the FSH and TSH molecules. The turkey antiserum recognizes a site on the native LH molecule that is altered when the subunits are separated. (Key words: luteinizing hormone, radioimmunoassay, chicken, turkey) 1984 Poultry Science 63:1269-1275
INTRODUCTION Several radioimmunoassay (RIA) procedures have been developed t h a t p e r m i t m e a s u r e m e n t of circulating levels of luteinizing h o r m o n e (LH) in avian species. One assay (Follett et al., 1972) was developed using chicken LH, while those of Wentworth et al. ( 1 9 7 6 ) and Burke et al. ( 1 9 7 9 a ) are h o m o l o g o u s for the turkey h o r m o n e . Each assay has also been used t o measure LH in avian species o t h e r t h a n the one for which it was developed ( " h e t e r o l o g o u s " assay). The physiological d a t a obtained with these different assay systems have been in general agreement. However, a direct comparison of LH m e a s u r e m e n t s of the same
samples is w a r r a n t e d t o assure t h e comparability of results o b t a i n e d by different p r o cedures in different laboratories. A recent publication (Goldsmith and Follett, 1 9 8 3 ) has provided a comparison of the assays of F o l l e t t et al. ( 1 9 7 2 ) and Burke et al. ( 1 9 7 9 a ) b y measuring t h e same plasma samples and various purified h o r m o n e preparations in a single l a b o r a t o r y . In t h e p r e s e n t paper, t h e results from two assay procedures for measuring avian LH (Follett et al, 1 9 7 2 ; Wentworth et al, 1976) were c o m p a r e d using divided aliquots of plasma from h o m o l o g o u s and h e t e r o l o g o u s species and using some purified h o r m o n e preparations.
1269
Downloaded from http://ps.oxfordjournals.org/ at University of South Dakota and School of Medicine on April 12, 2015
Department of Animal Sciences, Rutgers — The State University, New Brunswick, New Jersey 08903
1270
PROUDMAN ET AL MATERIALS AND METHODS
Luteinizing Hormone Assays
The chicken LH assay was performed at Rutgers University by the method of Follett et al. (1972). Briefly, this assay entailed iodination of LH fraction AE1 by chloramine T (100 fig) and purification of labeled chicken LH on CF11 cellulose. Antiserum #16/8, raised using LH fraction CM2 as antigen, was used at a final
Physiological Studies Experiment 1. The changes in LH secretion during the ovulatory cycle of the turkey were studied using mature (49 to 51 weeks) Diamond Hybrid Small White turkey hens housed in wire laying cages. The hens were maintained on a 14L:10D photoperiod (lights on at 0300 hr) and had laid at least two consecutive 2-egg clutches immediately prior to the experiment. Blood samples were drawn by venipuncture into heparinized syringes at 2-hr intervals for 24 hr, starting at 1100 hr on the day of lay of the terminal egg (Cj) of a sequence. The expected time of the subsequent C\ ovulation (0200 to 0400 hr) was verified by the position of the egg in the oviduct at autopsy approximately 12 hr after ovulation. Those birds that did not ovulate or that had an egg at a position in the oviduct inconsistent with die expected time of ovulation were excluded from the experiment. Blood samples from all experiments were centrifuged immediately after collection and each plasma sample was divided into two aliquots and frozen until assay. In all experiments, the LH concentrations were grouped to the nearest 2-hr period in relation to the estimated time of ovulation. Experiment 2. Changes of LH during the ovulatory cycles of chickens were studied using mature (42 week) Hubbard broiler breeder hens housed in individual laying cages under a 15.5L:8.5D photoperiod (lights on at 0600 hr). Hens selected for this study had a history of laying long clutches and had laid the Ci or C2 egg between 0730 and 1100 hr on the day of the experiment. Blood samples were drawn at 2- or 4-hr intervals over a period of 20 hr. The 2-hr intervals were chosen to coincide with the expected time of the preovulatory surge of LH. The time each hen ovulated was estimated to be
Downloaded from http://ps.oxfordjournals.org/ at University of South Dakota and School of Medicine on April 12, 2015
The turkey LH assay was performed at Beltsville using a modification of the procedure of Wentworth et al, (1976). The production and characteristics of the turkey LH antiserum (#753-10-22) have been described (Wentworth et al, 1976). Turkey LH B110B (Burke et al, 1979a) was used for the standard and for iodination. Iodination was accomplished using a stoichiometric chloramine T procedure as described previously (Proudman and Opel, 1981) but omitting the second gel chromatography step (Sephadex G-75). The modified assay showed no response to sera from hypophysectomized turkeys and chickens, but did respond to plasma and pituitary extracts in a dose-related manner parallel to the purified turkey LH standard. In Experiments 1 and 2, plasma samples were assayed in duplicate at a dose of 200 /il/tube. Standard and unknowns were diluted to a volume of 500 /il with assay buffer and the assay was initiated by. adding 200 |l(l of a 1:20,000 dilution of first antibody. After 24 hr at 4 C, radiolabeled turkey LH (100 fil; about 30,000 cpm) was added, and the incubation was continued for 48 hr. Second antibody (200 fil) was then added for a final overnight incubation. In Experiment 3, plasma samples were assayed at a dose of 100 fil, and the total assay volume was reduced by 50%. The final concentrations of all other components remained unchanged, and comparison of interassay plasma standards revealed no difference in the potency estimates obtained with the normal or reduced volume assays. Potency estimates were calculated using the spline-function program employed by the LKB Model 1270 gamma counter. All samples from one experiment were assayed in a single RIA to eliminate interassay variability within an experiment. The intraassay coefficient of variation (CV) was 7.4%, and the interassay CV (between experiments) was 14.2%. Assay sensitivity was .10 ng for the standard procedure and .05 ng for the reduced volume procedure.
dilution of 1:80,000 and LH fraction AE1 was used as standard. The assay, performed at 4 C, employed a 24-hr incubation with first antibody, an additional 24 hr incubation after adding labeled chicken LH, and a final 24 hr precipitation phase with second antibody (Antibodies Inc., Davis, CA). Plasma samples were assayed in duplicate at a single concentration (40 fil in a total assay volume of 200 fil). Samples from each experiment were assayed in a single RIA to exclude interassay variance. The intraassay CV was 9.5%, and the interassay CV (between experiments) was 13.5%.
RADIOIMMUNOASSAY FOR LUTEINIZING HORMONE
Avian Glycoprotein
Hormones
In each assay, the crossreaction of the following purified avian hormones was assessed: ostrich LH SC6BRB, TSH SC8RC, and FSH SC20B (Papkoff et al., 1982); turkey LH a subunit B124B and j3 subunit B129 (Burke et al., 1979b). Statistical Analysis Data from each blood collection series were analyzed separately by analysis of variance with assay, bird, assay X bird, time, and assay X time as factors. A significant F-test for assay X time interaction indicated nonparallelism between the LH profiles provided by the two assay procedures. We used regression analysis to determine die correlation between assays within each series. The data used for regression analysis were log transformed to equalize variance. The procedures of Barr et al. (1982) were used for all statistical analyses. RESULTS Regardless of the LH assay used in Experiment 1, LH levels measured in samples from nine turkey hens provided nearly identical profiles of LH change during the ovulatory cycle (Fig. 1). The absolute LH values were lower when determined relative to the turkey LH standard than when determined relative to the chicken standard because of differences in die potencies of the two standards. The correlation between potency estimates from the
two assay procedures was .75, which was highly significant (P<.001). Analysis of variance confirmed that the two profiles were indeed parallel (P = .94 for nonparallelism). The changes in LH levels in samples from the six broiler breeder chickens sampled in Experiment 2 are shown in Figure 2. Because this experiment provided fewer sampling periods than did Experiment 1 and fewer hens ovulated when expected, these data were characterized by larger standard errors and poorer definition of the preovulatory LH surge than was seen in the turkey experiment. The correlation between assays was .76 (P<.001). The test for nonparallelism was not significant (P = .15), indicating parallelism between assays. Experiment 3 provided more sampling periods, and hence, both a more precise estimate of the time of ovulation and a better definition of LH changes before the preovulatory surge than did Experiment 2. Figure 3A illustrates the LH profile determined from samples collected from 0900 to 0900 hr, and Figure 3B illustrates the profile determined from samples collected from 1500 to 1500 hr. The absolute LH values measured by the chicken assay procedure were lower than in previous experiments due to the use of a different (more potent) chicken LH standard. Under both collection schedules, there was a
5.0 4.0-
3.0
2.0
1.0
0L
-16 -14 -12 -10 -8 -6 -4 -2 0 »2 »4 *6 *8 TIME
FROM
OVULATION
(hr)
FIG. 1. Profile of luteinizing hormone (LH) secretion (mean ± SEM) during the ovulatory cycle of the turkey (n = 9) as measured by the chicken LH radioimmunoassay (RIA) (•- - -•) and the turkey LH RIA ( • — • ) . (The black horizontal bar indicates the hours of darkness.)
Downloaded from http://ps.oxfordjournals.org/ at University of South Dakota and School of Medicine on April 12, 2015
30 min after the time of oviposition of the C2 or C 3 egg. This estimate was checked by digital palpation, and any hen that did not ovulate or that appeared to have ovulated at an unexpected time was excluded from the experiment. Experiment 3. Two separate groups of Shaver minibroiler breeder hens were sampled at 2-hr intervals for 24 hr commencing at either 0900 or 1500 hr on different days. These birds, also housed in individual cages, were on a 16L:8D photoperiod (lights on at 0200 hr) and were sampled relative to the C 3 or C4 ovulation of a long sequence. The time each hen ovulated was estimated by the time of oviposition. Hens were autopsied at the end of the experiment to inspect the ovary for signs of atresia and to verify the time of ovulation by die position of the egg in the oviduct. Hens mat did not ovulate as expected were excluded from the experiment.
1271
PROUDMAN ET AL
1272
~ E
4.0
o>
-S
3.0
x
<
2.0
in
1 '°
DISCUSSION -20 -18 -16 -14 -12 -10 - 8 - 6 - 4 -2 TIME
FROM
OVULATION
0 *2
(hr)
FIG. 2. Profile of luternizing hormone (LH) secretion (mean ± SEM) during the ovulatory cycle of Hubbard broiler breeder hens as measured by the chicken radioimmunoassay (RIA) (• •) and turkey LH RIA (• •). (Numbers in parentheses represent the number of observations at each point.)
well-defined peak in LH 4 hr before ovulation. The changes before and after this preovulatory surge, as shown in Figure 3B, are in general agreement with those shown in Figure 2; i.e., declining LH levels before the surge and increasing LH levels after ovulation. As shown in Figure 3A, LH levels generally declined during the period that followed one ovulation and preceeded the surge of LH for the next ovulation, but no evidence was found of a welldefined LH peak during this midcycle period. The correlation between assay procedures was highly significant (P<.001) under both sampling schedules (r = .81 for samples taken beginning at 0900 hr; r = .61 for samples taken beginning at 1500 hr). However, analysis of variance indicated that the LH profiles in Figure 3A were parallel (P = .22 for nonparallellism), whereas those in Figure 3B were not (P= .005). The degree to which various purified avian glycoprotein hormones were able to compete for LH binding sites in the two assay systems is shown in Figure 4. Turkey LH and its a subunit were active in the chicken LH assay in a doserelated manner approximately parallel to that of the chicken LH standard (Fig. 4A). All ostrich hormones exhibited crossreaction in this assay, with LH exhibiting the highest crossreaction (19.4%) and ostrich TSH the least (5.4%). The (3 subunit of turkey LH exhibited little activity in the chicken assay. In the turkey
The chicken LH RIA of Follett et al. (1972) and the turkey LH RIA of Wentworth et al. (1976) gave virtually identical assessments of LH changes during the ovulatory cycles of turkeys and chickens. Both assay procedures revealed that LH peaked 6 hr before ovulation in the turkey and 4 hr before ovulation in the chicken, which was in agreement with the results of many previous studies of both species. In addition, the results of both assays suggested
< -24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 -2
E S»
3.0
-18 -16 -14 -12 -10 -8 - 6 - 4 -2 TIME
FROM
0
OVULATION
*2 - 4 -.6 »8 (hr)
FIG. 3. Profile of luteinizing hormone (LH) secretion (mean ± SEM) during the ovulatory cycle of Shaver minibroiler breeder hens as measured by the chicken LH radioimmunoassay (RIA) (•- - -•) and the turkey LH RIA ( • — • ) ; A) sampling beginning at 0900 hr and B) sampling beginning at 1500 hr. n = 9 different birds in each series, but because of differences in the time of ovulation, n was reduced to 4 or 5 at the extreme points in each panel.
Downloaded from http://ps.oxfordjournals.org/ at University of South Dakota and School of Medicine on April 12, 2015
LH assay, only chicken LH, ostrich TSH, and the /S subunit of turkey LH produced dose responses that were approximately parallel with the dose response of the turkey LH standard (Fig. 4B). The purified chicken LH was only 8% as active as the turkey standard in this assay system. Ostrich TSH showed 10% crossreaction, and ostrich LH and FSH (as well as turkey LH a subunit) showed varying degrees of crossreaction that were not dose-related or parallel to the dose response of the turkey LH standard.
5.0
RADIOIMMUNOASSAY FOR LUTEINIZING HORMONE
1273
100
80
60 40
20
o
100
UJ 0.
100 DOSE
(ng)
FIG. 4. Dose response curves for various purified avian glycoprotein hormones in A) the homologous chicken luteinizing hormone (LH) radioimmunoassay (RIA) and B) the homologous turkey LH RIA. (Specific identifications of the chicken (c), turkey (t), and ostrich (os) preparations used is given in the Materials and Methods section.)
that chicken LH levels may increase after ovulation and then gradually decline before die next preovulatory LH surge. Although die LH changes observed during mis period in our broiler breeders did not represent a consistent and well-defined peak, die levels found were clearly not as static as during the same period in the turkey. Reports of several previous studies with bioassay and RIA indicated increased LH secretion in some chickens at 23 to 20 hr before ovulation, at 14 to 12 hr before ovulation hr, or at both times before ovulation (Nelson et al., 1965; Bullock and Nalbandov, 1967; Cunningham and Furr, 1972; Shonodo et al., 1975; Etches and Cunningham, 1977; Ax, 1978). A significant midcycle peak of LH at 15 hr before ovulation was detected by Etches et al. (1981) with the RIA of Follett et al. (1972). Wilson et al. (198 3) observed an LH peak at 8 hr after ovulation, a time that would cor-
respond to 18 hr before ovulation in a midsequence hen. However, they found that this peak also occurred after the terminal egg of a sequence was laid, suggesting tiiat the midcycle LH peak noted by others may be related to the preceding ovulation rather than to a subsequent ovulation. Substantial differences were noted in die crossreaction of purified hormones in die two assay systems. Purified turkey LH crossreacted well in the chicken assay, but the purified chicken LH crossreacted poorly in the turkey assay. Yet, both assays performed equally well in measuring LH levels in plasma samples from both chickens and turkeys. These results suggested that the chicken LH standard that was used differed immunologically from circulating chicken LH in that the immunological binding site recognized by the turkey antiserum was altered, but the binding site for
Downloaded from http://ps.oxfordjournals.org/ at University of South Dakota and School of Medicine on April 12, 2015
I
1274
PROUDMAN ET AL
Ostrich hormones crossreacted in the turkey LH RIA, but LH and FSH had extremely shallow dose-response curves, and TSH had a crossreaction of about 10%. Papkoff et al. (1982) found that ostrich LH had a low crossreaction and a shallow dose-response curve in the homologous turkey LH RIA of Burke et al. (1979a). The chicken LH assay thus appears to be less species specific than either the turkey LH assay used here or that of Burke et al. (1979a), as all three ostrich hormones crossreacted in the chicken assay in a dose-related manner with ostrich LH exhibiting the highest activity. Comparisons between assays on the basis of crossreactivity with purified hormones are difficult to interpret, however, because of the differences in the species specificity of antisera and because of the uncertainty about the immunological purity and integrity of purified preparations. The current study focused mainly on the measurement of LH levels in plasma samples, because such measurement represents
the primary use of RIA. Both assays, used independently in different laboratories, provided the same assessment of circulating LH changes in both species studied. Goldsmith and Follett (1983) reached a similar conclusion in comparing the chicken LH assay used here witfi the turkey LH assay of Burke et al. (1979a). Because each assay was developed independently, with antisera to different purified LH preparations, their agreement in assessing physiological changes in circulating hormone levels increases our confidence in both the accuracy and specificity of the avian LH RIA's, and the comparability of data from different laboratories. ACKNOWLEDGMENTS
The authors express appreciation to G. Campbell, B. Russell, S. McGuire, and N. Zimmermann for their assistance; to D. Gavelek for preparation of figures; to E. Russek, Department of Animal Sciences, University of Maryland, for statistical analyses; and to W. H. Burke, University of Georgia, H. Papkoff, University of California-San Francisco, and B. C. Wentworth, University of Wisconsin, for their gifts of purified hormones and LH antiserum. This is Scientific Article #A-3729, Contribution #6705, of the Maryland Experiment Station (Department of Poultry Science).
REFERENCES Ax, R. L., Jr., 1978. Development and validation of a bioassay for chicken LH and application to chicken serum. Ph.D. thesis, Univ. Illinois, Urban a, IL. Barr, A. J., J. H. Goodnight, J. P. Sail, and T. J. Helwig, 1982. Statistical Analysis Systems. Statistical Analysis System Inst., Inc., Raleigh, NC. Bullock, D. W., and A. V. Nalbandov, 1967. Hormonal control of die hen's ovulatory cycle. J. Endocrinol. 38:407-415. Burke, W. H., P. Licht, H. Papkoff, and A. Bona Gallo, 1979a. Isolation and characterization of luteinizing hormone and follicle-stimulating hormone from pituitary glands of the turkey (Meleagris gallopavo). Gen. Comp. Endocrinol. 37:508— 520. Burke, W. H., H. Papkoff, P. Licht, and A. Bona Gallo, 1979b. Preparation and properties of luteinizing hormone (LH) subunits from the turkey (Meleagris gallopavo) and their recombination with subunits of ovine LH. Gen. Comp. Endocrinol. 37:501-507. Cunningham, F. J., and B.J.A. Furr, 1972. Plasma levels of luteinizing hormone and progesterone
Downloaded from http://ps.oxfordjournals.org/ at University of South Dakota and School of Medicine on April 12, 2015
the chicken antiserum remained intact. A relatively minor change in the structure or conformation of the hormone molecule during purification, or denaturation during storage, could have affected these binding sites. The crossreaction of turkey LH a and (3 subunits in each of the assays further suggested that the antisera recognize different portions of the LH molecule. The a subunit was as potent as the native turkey hormone in inhibiting the binding of iodinated chicken LH in the chicken assay when potency was calculated on a weight basis or 56% as potent as native turkey LH when calculated on a molar basis. The (3 subunit crossreacted very poorly in the chicken assay. Neither subunit competed as well as did the native hormone in the turkey assay, although the fi subunit, when assayed at relatively high doses, appeared to inhibit binding in a doserelated manner that was similar to the binding inhibition of native turkey LH. Thus, the immunological site recognized by the chicken antiserum apparently is predominately on the a subunit, whereas the turkey antiserum probably recognizes a site on the native molecule that is altered when the subunits are separated. However, in view of the greater potencies of chicken, turkey, and ostrich LH than of ostrich FSH or TSH, the immunoreactive site on the a subunit recognized by the chicken LH assay may be at least partially masked in the FSH and TSH molecules.
RADIOIMMUNOASSAY FOR LUTEINIZING HORMONE
1965. Changes in hypophysial and plasma LH levels during the laying cycle of the hen. Endocrinology 77:889-896. Papkoff, H., P. Licht, A. Bona-Gallo, D. S. MacKenzie, W. Oelofsen, and M.M.J. Oosthuizen, 1982. Biochemical and immunological characterization of pituitary hormones from the Ostrich (Struthio camelus). Gen. Comp. Endocrinol. 48:181—195. Proudman, J. A., and H. Opel, 1981. Turkey prolactin: Validation of a radioimmunoassay and measurement of changes associated with broodiness. Biol. Reprod. 25:573-580. Shodono, M., T. Nakamura, Y. Tanabe, and K. Wakabayashi, 1975. Simultaneous determinations of oestradiol-17-beta, progesterone and luteinizing hormone in the plasma during the ovulatory cycle of the hen. Acta Endocrinol. 78: 565—573. Wentworth, B. C , W. H. Burke, and G. P. Birrenkott, 1976. A radioimmunoassay for turkey luteinizing hormone. Gen. Comp. Endocrinol. 29:119-127. Wilson, S. C, R. C. Jennings, and F. J. Cunningham, 1983. An investigation of diural and cyclic changes in die secretion of luteinizing hormone in the domestic hen. J. Endocrinol. 98:137-145.
Downloaded from http://ps.oxfordjournals.org/ at University of South Dakota and School of Medicine on April 12, 2015
during the ovulatory cycle of the hen. Pages 51—64 in Egg Formation and Production. B. M. Freeman and P. E. Lake, ed. Br. Poult. Sci., Ltd., Edinburgh. Etches, R. J., F. Croze, and C. E. Duke, 1981. Plasma concentrations of luteinizing hormone, progesterone, testosterone and estradiol in follicular and peripheral venous plasma during the ovulation cycle of the hen. Page 89—98 in Advances in Physiological Science. Vol. 33. Recent Advances in Avian Endocrinology. G. Pethes, P. Peczely and P. Rudas, ed. Pergamon Press, New York, NY. Etches, R. J., and F. J. Cunningham, 1977. The plasma concentrations of testosterone and LH during the ovulation cycle of the hen (Gallus domesticus). Acta Endocrinol. 84:357—366. Follett, B. K., C. G. Scanes, and F. J. Cunningham, 1972. A radioimmunoassay for avian luteinizing hormone. J. Endocrinol. 64:87—101. Goldsmith, A. R., and B. K. Follett, 1983. Avian LH and FSH: Comparison of several radioimmunoassays. Gen. Comp. Endocrinol. 50:25—35. Nelson, D. M., H. W. Norton, and A. V. Nalbandov,
1275
H. OPEL US Department of Agriculture, ARS, Avian Physiology Laboratory, Beltsville, Maryland 20705 M. A. OTTINGER Department of Poultry Science, University of Maryland, College Park, Maryland 20742 (Received for publication October 14, 1983) ABSTRACT Two radioimmunoassay (RIA) procedures for measuring avian luteinizing hormone (LH) were compared using divided aliquots of plasma samples collected during the ovulatory cycles of turkey and broiler breeder hens in three separate experiments. The RIA systems compared were the homologous turkey LH assay of Wentworth et al. (1976) and the homologous chicken LH assay of Follett et al. (1972). Both assays provided the same assessment of circulating LH changes. The preovulatory surge of LH was observed 6 hr before ovulation in turkeys and 4 hr before ovulation in chickens. In addition, both assays detected an increase in chicken LH after ovulation, followed by a gradual decline prior to the next preovulatory surge. The correlation between RIA were highly significant (P<.001) in each experiment. The crossreaction of purified ostrich LH, FSH, and TSH and of the a and /3 subunits of turkey LH was assessed in each assay system. Ostrich LH, FSH, and TSH crossreacted in the chicken LH assay in a dose-related manner: ostrich LH showed the highest activity. Only ostrich TSH produced a response in the turkey assay that was dose related. The a subunit, but not the @ subunit, of turkey LH was quite potent in the chicken LH RIA, but neither subunit competed as well as the native hormone in the turkey RIA. Thus, the two LH antisera recognize different portions of the LH molecule. The chicken antiserum apparently recognizes a site on the a subunit that may be at least partially masked in the FSH and TSH molecules. The turkey antiserum recognizes a site on the native LH molecule that is altered when the subunits are separated. (Key words: luteinizing hormone, radioimmunoassay, chicken, turkey) 1984 Poultry Science 63:1269-1275
INTRODUCTION Several radioimmunoassay (RIA) procedures have been developed t h a t p e r m i t m e a s u r e m e n t of circulating levels of luteinizing h o r m o n e (LH) in avian species. One assay (Follett et al., 1972) was developed using chicken LH, while those of Wentworth et al. ( 1 9 7 6 ) and Burke et al. ( 1 9 7 9 a ) are h o m o l o g o u s for the turkey h o r m o n e . Each assay has also been used t o measure LH in avian species o t h e r t h a n the one for which it was developed ( " h e t e r o l o g o u s " assay). The physiological d a t a obtained with these different assay systems have been in general agreement. However, a direct comparison of LH m e a s u r e m e n t s of the same
samples is w a r r a n t e d t o assure t h e comparability of results o b t a i n e d by different p r o cedures in different laboratories. A recent publication (Goldsmith and Follett, 1 9 8 3 ) has provided a comparison of the assays of F o l l e t t et al. ( 1 9 7 2 ) and Burke et al. ( 1 9 7 9 a ) b y measuring t h e same plasma samples and various purified h o r m o n e preparations in a single l a b o r a t o r y . In t h e p r e s e n t paper, t h e results from two assay procedures for measuring avian LH (Follett et al, 1 9 7 2 ; Wentworth et al, 1976) were c o m p a r e d using divided aliquots of plasma from h o m o l o g o u s and h e t e r o l o g o u s species and using some purified h o r m o n e preparations.
1269
Downloaded from http://ps.oxfordjournals.org/ at University of South Dakota and School of Medicine on April 12, 2015
Department of Animal Sciences, Rutgers — The State University, New Brunswick, New Jersey 08903
1270
PROUDMAN ET AL MATERIALS AND METHODS
Luteinizing Hormone Assays
The chicken LH assay was performed at Rutgers University by the method of Follett et al. (1972). Briefly, this assay entailed iodination of LH fraction AE1 by chloramine T (100 fig) and purification of labeled chicken LH on CF11 cellulose. Antiserum #16/8, raised using LH fraction CM2 as antigen, was used at a final
Physiological Studies Experiment 1. The changes in LH secretion during the ovulatory cycle of the turkey were studied using mature (49 to 51 weeks) Diamond Hybrid Small White turkey hens housed in wire laying cages. The hens were maintained on a 14L:10D photoperiod (lights on at 0300 hr) and had laid at least two consecutive 2-egg clutches immediately prior to the experiment. Blood samples were drawn by venipuncture into heparinized syringes at 2-hr intervals for 24 hr, starting at 1100 hr on the day of lay of the terminal egg (Cj) of a sequence. The expected time of the subsequent C\ ovulation (0200 to 0400 hr) was verified by the position of the egg in the oviduct at autopsy approximately 12 hr after ovulation. Those birds that did not ovulate or that had an egg at a position in the oviduct inconsistent with die expected time of ovulation were excluded from the experiment. Blood samples from all experiments were centrifuged immediately after collection and each plasma sample was divided into two aliquots and frozen until assay. In all experiments, the LH concentrations were grouped to the nearest 2-hr period in relation to the estimated time of ovulation. Experiment 2. Changes of LH during the ovulatory cycles of chickens were studied using mature (42 week) Hubbard broiler breeder hens housed in individual laying cages under a 15.5L:8.5D photoperiod (lights on at 0600 hr). Hens selected for this study had a history of laying long clutches and had laid the Ci or C2 egg between 0730 and 1100 hr on the day of the experiment. Blood samples were drawn at 2- or 4-hr intervals over a period of 20 hr. The 2-hr intervals were chosen to coincide with the expected time of the preovulatory surge of LH. The time each hen ovulated was estimated to be
Downloaded from http://ps.oxfordjournals.org/ at University of South Dakota and School of Medicine on April 12, 2015
The turkey LH assay was performed at Beltsville using a modification of the procedure of Wentworth et al, (1976). The production and characteristics of the turkey LH antiserum (#753-10-22) have been described (Wentworth et al, 1976). Turkey LH B110B (Burke et al, 1979a) was used for the standard and for iodination. Iodination was accomplished using a stoichiometric chloramine T procedure as described previously (Proudman and Opel, 1981) but omitting the second gel chromatography step (Sephadex G-75). The modified assay showed no response to sera from hypophysectomized turkeys and chickens, but did respond to plasma and pituitary extracts in a dose-related manner parallel to the purified turkey LH standard. In Experiments 1 and 2, plasma samples were assayed in duplicate at a dose of 200 /il/tube. Standard and unknowns were diluted to a volume of 500 /il with assay buffer and the assay was initiated by. adding 200 |l(l of a 1:20,000 dilution of first antibody. After 24 hr at 4 C, radiolabeled turkey LH (100 fil; about 30,000 cpm) was added, and the incubation was continued for 48 hr. Second antibody (200 fil) was then added for a final overnight incubation. In Experiment 3, plasma samples were assayed at a dose of 100 fil, and the total assay volume was reduced by 50%. The final concentrations of all other components remained unchanged, and comparison of interassay plasma standards revealed no difference in the potency estimates obtained with the normal or reduced volume assays. Potency estimates were calculated using the spline-function program employed by the LKB Model 1270 gamma counter. All samples from one experiment were assayed in a single RIA to eliminate interassay variability within an experiment. The intraassay coefficient of variation (CV) was 7.4%, and the interassay CV (between experiments) was 14.2%. Assay sensitivity was .10 ng for the standard procedure and .05 ng for the reduced volume procedure.
dilution of 1:80,000 and LH fraction AE1 was used as standard. The assay, performed at 4 C, employed a 24-hr incubation with first antibody, an additional 24 hr incubation after adding labeled chicken LH, and a final 24 hr precipitation phase with second antibody (Antibodies Inc., Davis, CA). Plasma samples were assayed in duplicate at a single concentration (40 fil in a total assay volume of 200 fil). Samples from each experiment were assayed in a single RIA to exclude interassay variance. The intraassay CV was 9.5%, and the interassay CV (between experiments) was 13.5%.
RADIOIMMUNOASSAY FOR LUTEINIZING HORMONE
Avian Glycoprotein
Hormones
In each assay, the crossreaction of the following purified avian hormones was assessed: ostrich LH SC6BRB, TSH SC8RC, and FSH SC20B (Papkoff et al., 1982); turkey LH a subunit B124B and j3 subunit B129 (Burke et al., 1979b). Statistical Analysis Data from each blood collection series were analyzed separately by analysis of variance with assay, bird, assay X bird, time, and assay X time as factors. A significant F-test for assay X time interaction indicated nonparallelism between the LH profiles provided by the two assay procedures. We used regression analysis to determine die correlation between assays within each series. The data used for regression analysis were log transformed to equalize variance. The procedures of Barr et al. (1982) were used for all statistical analyses. RESULTS Regardless of the LH assay used in Experiment 1, LH levels measured in samples from nine turkey hens provided nearly identical profiles of LH change during the ovulatory cycle (Fig. 1). The absolute LH values were lower when determined relative to the turkey LH standard than when determined relative to the chicken standard because of differences in die potencies of the two standards. The correlation between potency estimates from the
two assay procedures was .75, which was highly significant (P<.001). Analysis of variance confirmed that the two profiles were indeed parallel (P = .94 for nonparallelism). The changes in LH levels in samples from the six broiler breeder chickens sampled in Experiment 2 are shown in Figure 2. Because this experiment provided fewer sampling periods than did Experiment 1 and fewer hens ovulated when expected, these data were characterized by larger standard errors and poorer definition of the preovulatory LH surge than was seen in the turkey experiment. The correlation between assays was .76 (P<.001). The test for nonparallelism was not significant (P = .15), indicating parallelism between assays. Experiment 3 provided more sampling periods, and hence, both a more precise estimate of the time of ovulation and a better definition of LH changes before the preovulatory surge than did Experiment 2. Figure 3A illustrates the LH profile determined from samples collected from 0900 to 0900 hr, and Figure 3B illustrates the profile determined from samples collected from 1500 to 1500 hr. The absolute LH values measured by the chicken assay procedure were lower than in previous experiments due to the use of a different (more potent) chicken LH standard. Under both collection schedules, there was a
5.0 4.0-
3.0
2.0
1.0
0L
-16 -14 -12 -10 -8 -6 -4 -2 0 »2 »4 *6 *8 TIME
FROM
OVULATION
(hr)
FIG. 1. Profile of luteinizing hormone (LH) secretion (mean ± SEM) during the ovulatory cycle of the turkey (n = 9) as measured by the chicken LH radioimmunoassay (RIA) (•- - -•) and the turkey LH RIA ( • — • ) . (The black horizontal bar indicates the hours of darkness.)
Downloaded from http://ps.oxfordjournals.org/ at University of South Dakota and School of Medicine on April 12, 2015
30 min after the time of oviposition of the C2 or C 3 egg. This estimate was checked by digital palpation, and any hen that did not ovulate or that appeared to have ovulated at an unexpected time was excluded from the experiment. Experiment 3. Two separate groups of Shaver minibroiler breeder hens were sampled at 2-hr intervals for 24 hr commencing at either 0900 or 1500 hr on different days. These birds, also housed in individual cages, were on a 16L:8D photoperiod (lights on at 0200 hr) and were sampled relative to the C 3 or C4 ovulation of a long sequence. The time each hen ovulated was estimated by the time of oviposition. Hens were autopsied at the end of the experiment to inspect the ovary for signs of atresia and to verify the time of ovulation by die position of the egg in the oviduct. Hens mat did not ovulate as expected were excluded from the experiment.
1271
PROUDMAN ET AL
1272
~ E
4.0
o>
-S
3.0
x
<
2.0
in
1 '°
DISCUSSION -20 -18 -16 -14 -12 -10 - 8 - 6 - 4 -2 TIME
FROM
OVULATION
0 *2
(hr)
FIG. 2. Profile of luternizing hormone (LH) secretion (mean ± SEM) during the ovulatory cycle of Hubbard broiler breeder hens as measured by the chicken radioimmunoassay (RIA) (• •) and turkey LH RIA (• •). (Numbers in parentheses represent the number of observations at each point.)
well-defined peak in LH 4 hr before ovulation. The changes before and after this preovulatory surge, as shown in Figure 3B, are in general agreement with those shown in Figure 2; i.e., declining LH levels before the surge and increasing LH levels after ovulation. As shown in Figure 3A, LH levels generally declined during the period that followed one ovulation and preceeded the surge of LH for the next ovulation, but no evidence was found of a welldefined LH peak during this midcycle period. The correlation between assay procedures was highly significant (P<.001) under both sampling schedules (r = .81 for samples taken beginning at 0900 hr; r = .61 for samples taken beginning at 1500 hr). However, analysis of variance indicated that the LH profiles in Figure 3A were parallel (P = .22 for nonparallellism), whereas those in Figure 3B were not (P= .005). The degree to which various purified avian glycoprotein hormones were able to compete for LH binding sites in the two assay systems is shown in Figure 4. Turkey LH and its a subunit were active in the chicken LH assay in a doserelated manner approximately parallel to that of the chicken LH standard (Fig. 4A). All ostrich hormones exhibited crossreaction in this assay, with LH exhibiting the highest crossreaction (19.4%) and ostrich TSH the least (5.4%). The (3 subunit of turkey LH exhibited little activity in the chicken assay. In the turkey
The chicken LH RIA of Follett et al. (1972) and the turkey LH RIA of Wentworth et al. (1976) gave virtually identical assessments of LH changes during the ovulatory cycles of turkeys and chickens. Both assay procedures revealed that LH peaked 6 hr before ovulation in the turkey and 4 hr before ovulation in the chicken, which was in agreement with the results of many previous studies of both species. In addition, the results of both assays suggested
< -24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2 0 -2
E S»
3.0
-18 -16 -14 -12 -10 -8 - 6 - 4 -2 TIME
FROM
0
OVULATION
*2 - 4 -.6 »8 (hr)
FIG. 3. Profile of luteinizing hormone (LH) secretion (mean ± SEM) during the ovulatory cycle of Shaver minibroiler breeder hens as measured by the chicken LH radioimmunoassay (RIA) (•- - -•) and the turkey LH RIA ( • — • ) ; A) sampling beginning at 0900 hr and B) sampling beginning at 1500 hr. n = 9 different birds in each series, but because of differences in the time of ovulation, n was reduced to 4 or 5 at the extreme points in each panel.
Downloaded from http://ps.oxfordjournals.org/ at University of South Dakota and School of Medicine on April 12, 2015
LH assay, only chicken LH, ostrich TSH, and the /S subunit of turkey LH produced dose responses that were approximately parallel with the dose response of the turkey LH standard (Fig. 4B). The purified chicken LH was only 8% as active as the turkey standard in this assay system. Ostrich TSH showed 10% crossreaction, and ostrich LH and FSH (as well as turkey LH a subunit) showed varying degrees of crossreaction that were not dose-related or parallel to the dose response of the turkey LH standard.
5.0
RADIOIMMUNOASSAY FOR LUTEINIZING HORMONE
1273
100
80
60 40
20
o
100
UJ 0.
100 DOSE
(ng)
FIG. 4. Dose response curves for various purified avian glycoprotein hormones in A) the homologous chicken luteinizing hormone (LH) radioimmunoassay (RIA) and B) the homologous turkey LH RIA. (Specific identifications of the chicken (c), turkey (t), and ostrich (os) preparations used is given in the Materials and Methods section.)
that chicken LH levels may increase after ovulation and then gradually decline before die next preovulatory LH surge. Although die LH changes observed during mis period in our broiler breeders did not represent a consistent and well-defined peak, die levels found were clearly not as static as during the same period in the turkey. Reports of several previous studies with bioassay and RIA indicated increased LH secretion in some chickens at 23 to 20 hr before ovulation, at 14 to 12 hr before ovulation hr, or at both times before ovulation (Nelson et al., 1965; Bullock and Nalbandov, 1967; Cunningham and Furr, 1972; Shonodo et al., 1975; Etches and Cunningham, 1977; Ax, 1978). A significant midcycle peak of LH at 15 hr before ovulation was detected by Etches et al. (1981) with the RIA of Follett et al. (1972). Wilson et al. (198 3) observed an LH peak at 8 hr after ovulation, a time that would cor-
respond to 18 hr before ovulation in a midsequence hen. However, they found that this peak also occurred after the terminal egg of a sequence was laid, suggesting tiiat the midcycle LH peak noted by others may be related to the preceding ovulation rather than to a subsequent ovulation. Substantial differences were noted in die crossreaction of purified hormones in die two assay systems. Purified turkey LH crossreacted well in the chicken assay, but the purified chicken LH crossreacted poorly in the turkey assay. Yet, both assays performed equally well in measuring LH levels in plasma samples from both chickens and turkeys. These results suggested that the chicken LH standard that was used differed immunologically from circulating chicken LH in that the immunological binding site recognized by the turkey antiserum was altered, but the binding site for
Downloaded from http://ps.oxfordjournals.org/ at University of South Dakota and School of Medicine on April 12, 2015
I
1274
PROUDMAN ET AL
Ostrich hormones crossreacted in the turkey LH RIA, but LH and FSH had extremely shallow dose-response curves, and TSH had a crossreaction of about 10%. Papkoff et al. (1982) found that ostrich LH had a low crossreaction and a shallow dose-response curve in the homologous turkey LH RIA of Burke et al. (1979a). The chicken LH assay thus appears to be less species specific than either the turkey LH assay used here or that of Burke et al. (1979a), as all three ostrich hormones crossreacted in the chicken assay in a dose-related manner with ostrich LH exhibiting the highest activity. Comparisons between assays on the basis of crossreactivity with purified hormones are difficult to interpret, however, because of the differences in the species specificity of antisera and because of the uncertainty about the immunological purity and integrity of purified preparations. The current study focused mainly on the measurement of LH levels in plasma samples, because such measurement represents
the primary use of RIA. Both assays, used independently in different laboratories, provided the same assessment of circulating LH changes in both species studied. Goldsmith and Follett (1983) reached a similar conclusion in comparing the chicken LH assay used here witfi the turkey LH assay of Burke et al. (1979a). Because each assay was developed independently, with antisera to different purified LH preparations, their agreement in assessing physiological changes in circulating hormone levels increases our confidence in both the accuracy and specificity of the avian LH RIA's, and the comparability of data from different laboratories. ACKNOWLEDGMENTS
The authors express appreciation to G. Campbell, B. Russell, S. McGuire, and N. Zimmermann for their assistance; to D. Gavelek for preparation of figures; to E. Russek, Department of Animal Sciences, University of Maryland, for statistical analyses; and to W. H. Burke, University of Georgia, H. Papkoff, University of California-San Francisco, and B. C. Wentworth, University of Wisconsin, for their gifts of purified hormones and LH antiserum. This is Scientific Article #A-3729, Contribution #6705, of the Maryland Experiment Station (Department of Poultry Science).
REFERENCES Ax, R. L., Jr., 1978. Development and validation of a bioassay for chicken LH and application to chicken serum. Ph.D. thesis, Univ. Illinois, Urban a, IL. Barr, A. J., J. H. Goodnight, J. P. Sail, and T. J. Helwig, 1982. Statistical Analysis Systems. Statistical Analysis System Inst., Inc., Raleigh, NC. Bullock, D. W., and A. V. Nalbandov, 1967. Hormonal control of die hen's ovulatory cycle. J. Endocrinol. 38:407-415. Burke, W. H., P. Licht, H. Papkoff, and A. Bona Gallo, 1979a. Isolation and characterization of luteinizing hormone and follicle-stimulating hormone from pituitary glands of the turkey (Meleagris gallopavo). Gen. Comp. Endocrinol. 37:508— 520. Burke, W. H., H. Papkoff, P. Licht, and A. Bona Gallo, 1979b. Preparation and properties of luteinizing hormone (LH) subunits from the turkey (Meleagris gallopavo) and their recombination with subunits of ovine LH. Gen. Comp. Endocrinol. 37:501-507. Cunningham, F. J., and B.J.A. Furr, 1972. Plasma levels of luteinizing hormone and progesterone
Downloaded from http://ps.oxfordjournals.org/ at University of South Dakota and School of Medicine on April 12, 2015
the chicken antiserum remained intact. A relatively minor change in the structure or conformation of the hormone molecule during purification, or denaturation during storage, could have affected these binding sites. The crossreaction of turkey LH a and (3 subunits in each of the assays further suggested that the antisera recognize different portions of the LH molecule. The a subunit was as potent as the native turkey hormone in inhibiting the binding of iodinated chicken LH in the chicken assay when potency was calculated on a weight basis or 56% as potent as native turkey LH when calculated on a molar basis. The (3 subunit crossreacted very poorly in the chicken assay. Neither subunit competed as well as did the native hormone in the turkey assay, although the fi subunit, when assayed at relatively high doses, appeared to inhibit binding in a doserelated manner that was similar to the binding inhibition of native turkey LH. Thus, the immunological site recognized by the chicken antiserum apparently is predominately on the a subunit, whereas the turkey antiserum probably recognizes a site on the native molecule that is altered when the subunits are separated. However, in view of the greater potencies of chicken, turkey, and ostrich LH than of ostrich FSH or TSH, the immunoreactive site on the a subunit recognized by the chicken LH assay may be at least partially masked in the FSH and TSH molecules.
RADIOIMMUNOASSAY FOR LUTEINIZING HORMONE
1965. Changes in hypophysial and plasma LH levels during the laying cycle of the hen. Endocrinology 77:889-896. Papkoff, H., P. Licht, A. Bona-Gallo, D. S. MacKenzie, W. Oelofsen, and M.M.J. Oosthuizen, 1982. Biochemical and immunological characterization of pituitary hormones from the Ostrich (Struthio camelus). Gen. Comp. Endocrinol. 48:181—195. Proudman, J. A., and H. Opel, 1981. Turkey prolactin: Validation of a radioimmunoassay and measurement of changes associated with broodiness. Biol. Reprod. 25:573-580. Shodono, M., T. Nakamura, Y. Tanabe, and K. Wakabayashi, 1975. Simultaneous determinations of oestradiol-17-beta, progesterone and luteinizing hormone in the plasma during the ovulatory cycle of the hen. Acta Endocrinol. 78: 565—573. Wentworth, B. C , W. H. Burke, and G. P. Birrenkott, 1976. A radioimmunoassay for turkey luteinizing hormone. Gen. Comp. Endocrinol. 29:119-127. Wilson, S. C, R. C. Jennings, and F. J. Cunningham, 1983. An investigation of diural and cyclic changes in die secretion of luteinizing hormone in the domestic hen. J. Endocrinol. 98:137-145.
Downloaded from http://ps.oxfordjournals.org/ at University of South Dakota and School of Medicine on April 12, 2015
during the ovulatory cycle of the hen. Pages 51—64 in Egg Formation and Production. B. M. Freeman and P. E. Lake, ed. Br. Poult. Sci., Ltd., Edinburgh. Etches, R. J., F. Croze, and C. E. Duke, 1981. Plasma concentrations of luteinizing hormone, progesterone, testosterone and estradiol in follicular and peripheral venous plasma during the ovulation cycle of the hen. Page 89—98 in Advances in Physiological Science. Vol. 33. Recent Advances in Avian Endocrinology. G. Pethes, P. Peczely and P. Rudas, ed. Pergamon Press, New York, NY. Etches, R. J., and F. J. Cunningham, 1977. The plasma concentrations of testosterone and LH during the ovulation cycle of the hen (Gallus domesticus). Acta Endocrinol. 84:357—366. Follett, B. K., C. G. Scanes, and F. J. Cunningham, 1972. A radioimmunoassay for avian luteinizing hormone. J. Endocrinol. 64:87—101. Goldsmith, A. R., and B. K. Follett, 1983. Avian LH and FSH: Comparison of several radioimmunoassays. Gen. Comp. Endocrinol. 50:25—35. Nelson, D. M., H. W. Norton, and A. V. Nalbandov,
1275