Biol Reprod Track the topics, authors and articles important to you
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow My Folders
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Bousfield, G. R.
Right arrow Articles by Butnev, V. Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bousfield, G. R.
Right arrow Articles by Butnev, V. Y.
Agricola
Right arrow Articles by Bousfield, G. R.
Right arrow Articles by Butnev, V. Y.
Biology of Reproduction 64, 136-147 (2001)
© 2001 Society for the Study of Reproduction, Inc.

Regular Article

Identification of Twelve O-Glycosylation Sites in Equine Chorionic Gonadotropin ß and Equine Luteinizing Hormone ß by Solid-Phase Edman Degradation1

George R. Bousfield2,a, Vladimir Y. Butneva, and Viktor Y. Butnev3,a

a Department of Biological Sciences, Wichita State University, Wichita, Kansas 67260-0026

ABSTRACT

The O-glycosylation sites for equine LHß (eLHß) and eCGß were identified by solid-phase Edman degradation of four glycopeptides derived from the C-terminal region. Both subunits were O-glycosylated at the same 12 positions, rather than the 4–6 sites anticipated. These sites were partially glycosylated, with carbohydrate attachment ranging from 20% to 100% for eCGß and from 10% to 100% for eLHß. When the C-terminal peptide containing all but one of the O-linked oligosaccharides was removed by mild acid hydrolysis of either eLHß or eCGß, hybrid hormones could be obtained by reassociating eLH{alpha},eFSH{alpha}, or eCG{alpha} with the truncated ß subunit derivatives. These hybrid hormones were identical in LH receptor-binding activity when des(121-149)eLHß or des(121-149)eCGß were combined with the same {alpha} subunit preparation. Thus, O-glycosylation appears to be responsible for the ß subunit contribution to the substantial difference in LH receptor-binding activity between eLH and eCG. Comparison of the equid LH/CGß sequences with those available for the primate CGß subunits indicated a greater conservation of glycosylation patterns in the former.

anterior pituitary, hormone action, LH, placenta

INTRODUCTION

A unique structural feature of primate chorionic gonadotropins and equid LH (eLH)/chorionic gonadotropins is an O-glycosylated C-terminal extension of the ß subunit that appears to extend survival of these hormones in the circulation but seems to have little to do with cellular activation [1]. Deletion of the C-terminus from eLHß and hCGß by mild acid hydrolysis [2, 3] or by mutation of recombinant hCGß [4, 5] produced no significant effect on LH receptor binding or in vitro biological activity. The potential role of the C-terminal extension in extending circulatory survival was demonstrated by adding it to recombinant bovine LHß and recombinant human FSH subunits and noting the decreased metabolic clearance rates of the resulting LH and FSH derivatives [6, 7]. Recently, however, a series of eCG glycosylation isoforms were described that exhibited the same patterns of N-glycosylation but differed in their ß subunit O-glycosylation [8]. The most heavily O-glycosylated eCG isoform was significantly less active than the other two eCG isoforms in terms of LH receptor-binding activity. Studies involving hybrid hormone preparations composed of all possible combinations of equine gonadotropin subunits have demonstrated that both {alpha} and ß subunits contributed to the difference in receptor-binding affinity between eLH and eCG [9]. The {alpha} subunit contribution was largely due to the addition of a Man({alpha}1-6)Man branch to asparagine (Asn)56 oligosaccharides and its extension by addition of lactosamine repeats. The ßAsn13 oligosaccharides have been reported to consist of largely biantennary structures that were terminated with GalNAc sulfate in eLHß and sialylated galactose in eCGß [10–12]. Because these structural differences were unlikely to be responsible for the reduced activity of eCG in vitro and its extended survival in vivo, we began to characterize the O-glycosylation of both eLH and eCG. In terms of O-glycosylation, the eCGß C-terminal extension possessed a greater carbohydrate content than that of eLHß [13]. It is not known whether this resulted from glycosylation of more of the five threonine (Thr) or 11 serine (Ser) residues located in the C-terminal extension, a greater percentage of carbohydrate present at the same sites, larger oligosaccharides attached to the same sites, or all three factors. Unlike N-glycosylation, which occurs at the consensus motif Asn-Xaa-Thr/Ser (Xaa is any amino acid other than proline), several O-glycosylation motifs have been suggested [14]. The existence of at least six mammalian GalNAc-transferases [15] makes it unlikely that O-glycosylation can be predicted with the same level of confidence as N-glycosylation. Accordingly, we identified the O-glycosylation sites for eLHß and eCGß using solid-phase Edman degradation.

MATERIALS AND METHODS

Hormone and Subunit Preparations

The equine gonadotropins eLH, eCG, and eFSH and their subunits were prepared by standard isolation procedures [8, 16, 17]. The laboratory eLH reference preparation potency was 2.0x NIDDK-oLH-26 in a rat testis LH receptor binding assay using 125I-hCG tracer. The eCG preparation employed for identification of the O-glycosylation sites consisted of a mixture of the two most abundant eCG isoforms, eCG-M and eCG-H [8]. These eCG isoforms possessed identical patterns of {alpha} subunit N-glycosylation but differed in the sizes of their ß subunits because of differential O-glycosylation. However, because the masses of the eCGß subunits were very broadly distributed, the imprecise terms low (L), medium (M), and high (H) have been employed to describe eCG preparations possessing ß subunits that migrated as broad bands of 71 000–217 000 Mr, 61 000–123 000 Mr, and 56 000–110 000 Mr, respectively. Truncated des(121-149)eLHß and des(121-149)eCGß derivatives (eLHßt and eCGßt, respectively) were produced by mild acid hydrolysis as described previously [3, 9]. These derivatives were combined with eLH{alpha}, eFSH{alpha}, and eCG{alpha} preparations as described below. The C-terminal peptide byproducts of these ß subunit derivatives, which corresponded to CTP-121-149, were used to identify O-glycosylation sites.

Preparation of eLHßt and eCGßt Hybrids

We combined 1 mg of eLH{alpha}, eFSH{alpha}, or eCG{alpha} with 1 mg eLHßt in 1 ml 0.5 M Tris-acetate buffer, pH 7.0, for 72 h at 37°C. The extended incubation period was necessary because eFSH subunits reassociated more slowly than eLH subunits (unpublished data) and eCG subunits reassociated less efficiently [9]. Reassociation buffer was replaced with 0.126 M ammonium bicarbonate by ultrafiltration in a Centricon P-10 (Amicon, Millipore, Bedford, MA), and protein was recovered by vacuum drying in a SpeedVac (Savant, Holbrook, NY). The protein fractions were dissolved in 200 µl of 0.126 M ammonium bicarbonate, and chromatography was performed on a Superdex 75 column (Pharmacia, Piscataway, NJ) equilibrated with the same buffer at a flow rate of 0.4 ml/min. Fractions were collected by hand, and protein was recovered by vacuum drying.

LH Receptor Binding and Steroidogenic Activities of Truncated ß Subunit Dervatives

The LH receptor binding activity was determined using 125I-hCG tracer and rat testis homogenate as previously described [8]. A 10-µg sample of hCG was iodinated by the chloramine-T method to a specific activity of 49 µCi/µg. The 125I-hCG was diluted to a final concentration of 25 ng/ml and stored at 4°C. Duplicate doses of hormone in 100 µl of RLA buffer (0.1 M Tris HCl, pH 7.4, 0.02% sodium azide, and 0.1% BSA) were added to pairs of 12- x 75-mm polypropylene tubes containing 100 µl of RLA buffer. A 2.5-ng sample of 125I-hCG in 100 µl of RLA buffer was added to each tube. Receptor preparation, consisting of 25 mg of rat testis homogenate suspended in 200 µl of RLA buffer, was added to each tube immediately before incubation. The assay tubes were vortexed and then placed in a shaking water bath and incubated for 2 h at 37°C. Following incubation, the tubes were centrifuged at 4600 x g for 30 min at 4°C in the H6000 rotor of an RC-3B centrifuge (Sorvall, Newtown, CT). The supernatants were aspirated, and the pellets were counted in a Cobra II gamma counter (Packard, Meriden, CT). A semilogarithmic plot of the percentage of specific binding versus hormone concentration was generated with the software package DeltaGraph (SPSS, Chicago, IL). The dose-response curves were simultaneously fitted to a four-parameter logistic equation using a Macintosh implementation of Allfit [18] called MacAllfit, running on a Macintosh IIci computer (Apple, Cupertino, CA). The 50% inhibitory concentrations, relative potencies, and their standard errors were calculated by MacAllfit for each curve. The 95% confidence limits were determined from these data. The within-assay variability was <5%, and the between-assay variability was <15%.

The in vitro biological activity of LH was determined in a rat testis Leydig cell bioassay [19, 20]. In a typical assay, two 50- to 80-day-old male rats were killed, and their testes were removed and decapsulated. The testes were placed in a 50-ml plastic culture tube containing 5 mg type I collagenase (Sigma Chemical Co., St. Louis, MO) dissolved in 20 ml of minimum essential medium (MEM; Gibco-BRL, Bethesda, MD). The tube was placed longitudinally in a shaking water bath and incubated for 10 min at 37°C with gentle agitation (75 cycles/min). Following incubation, 15 ml of cold MEM was added, and the tube inverted 50 times and then allowed to stand at 4°C for 5 min. The supernatant, which contained the Leydig cells, was transferred to a clean 50-ml culture tube. An additional 15-ml aliquot of MEM was added to the pellet, and the process was repeated. The second supernatant fraction was combined with the first, and the combined sample was centrifuged at 160 x g for 10 min at 4°C. The supernatant was discarded, the Leydig cell pellet was resuspended in 30 ml of cold MEM, and centrifugation was repeated. The cells were resuspended in cold MEM, and the open tube was placed in a 34°C shaker bath and incubated for 1 h with slow (75 cycles/min) shaking. Following the preincubation step, the cells were centrifuged at 160 x g for 10 min at 4°C. The cell pellet was resuspended in 25°C MEM and diluted to a final concentration of 60 000 cells/200 µl. Triplicate hormone samples that had been diluted in 100 µl MEM were combined with 200 µl of Leydig cell suspension in 12- x 75-mm polypropylene test tubes. The tubes were incubated at 34°C for 3 h with gentle shaking (75 cycles/min) and then frozen. On the following day, the tubes were thawed and centrifuged, and the testosterone content of each supernatant was determined by RIA [21]. Antiserum for the testosterone RIA was obtained from Dr. Gordon Niswender (Colorado State University, Ft. Collins, CO). RIA data were analyzed using the software package AssayZap (Biosoft, Ferguson, MO). The within-assay variability and between-assay variability for the testosterone RIA were <10% and <12%, respectively. The Leydig cell assay dose-response curves were plotted using DeltaGraph, and statistical analysis was carried out using MacAllfit [18]. Relative potencies and 95% confidence limits were determined from the results of this analysis.

Isolation of C-Terminal Peptides CTP-100-149, CTP-121-149, and CTP-140-149

A 9.4-mg sample of reduced, carboxymethylated (RCM)-eCGß was dissolved in 0.05 M sodium phosphate buffer, pH 7.8, and incubated for 24 h in the presence of 2% Staphylococcus aureus V8 protease [22]. An 11.8-mg sample of RCM-eLHß was digested with V8 protease using the same conditions. A fraction consisting of two O-glycosylated peptides was separated from the N-glycosylated amino terminal glycopeptide fraction by gel filtration over Sephacryl S-200, as previously described [22]. The O-linked glycopeptides were identified by automated Edman degradation using an updated 890B sequencer (Beckman, Fullerton, CA) [22]. The larger glycopeptide (residues 62-149) was separated from the shorter form (residues 100–149) by reverse-phase HPLC using a Vydac C-18 column (The Separations Group, Hesperia, CA) attached to an HPLC system (Waters, Milford, MA) as previously described [22]. The column was equilibrated with 0.01% trifluoroacetate (TFA) in water at a flow rate of 1 ml/min. Samples consisting of the C-terminal glycopeptide fractions derived from either eLH or eCG were dissolved in the 0.01% TFA and applied to the column. The column was washed with 0.01% TFA in water for 5 min. Following the wash, the eluant composition was changed to 5% acetonitrile for eLHß peptides and 9% acetonitrile for eCGß peptides in a stepwise manner. A linear gradient to 35% acetonitrile applied over the next 80 min developed the chromatogram. A stepwise increase to 80% acetonitrile for 10 min followed by a 10-min re-equilibration period with 0.01% TFA/water completed the chromatogram. The absorbance at 214 nm was monitored, and peptide fractions were collected by hand and dried under vacuum in a SpeedVac. The shorter peptide, designated CTP-100-149, was subjected to carbohydrate composition analysis and solid-phase N-terminal sequencing.

The longer peptide, CTP-62-149, was cleaved by mild acid hydrolysis to obtain CTP-121-149. CTP-62-149 was dissolved in 0.013 N HCl and incubated at 110°C for 30 min [13]. The sample was applied to a Vydac C-18 column coupled to a different HPLC system (Waters) that consisted of two model 501 HPLC pumps, an automated gradient controller, a U6K sample injector, and a model 484 tunable absorbance monitor. Data were acquired and processed using Maxima software (Waters) running on a PowerMate I computer (NEC, Boxborough, MA). The Vydac C-18 column was equilibrated with 0.01% TFA/water at a flow rate of 1 ml/min. The glycopeptides resulting from mild acid hydrolysis were separated from each other using the same 0–50% acetonitrile gradient employed to separate CTP-121-149 from the ß core fragment of eLHß or eCGß [13].

We obtained a third glycopeptide, designated CTP-140-149, by tryptic digestion of the intact ß subunit preparations following an approach previously reported for eLHß [22]. Samples consisting of either 1.1 mg intact eLHß or 1.1 mg intact eCGß were incubated in 0.1 M ammonium bicarbonate buffer, pH 7.8, containing 1% (w/w) trypsin at 37°C for 4 h. Glycopeptides were separated from the eLHß core fraction by reverse-phase HPLC using a Vydac C-18 column equilibrated with 0.01% TFA/water at 1 ml/min. The chromatogram was developed with a 0–30% acetonitrile gradient over 50 min as previously described [22]. Digestion of the eCGß fraction was repeated with 10% (w/w) trypsin at 37°C for 48 h, and the products were chromatographed under the same conditions. Peptide fractions were characterized by amino acid analysis and N-terminal sequencing. Based on results of these preliminary experiments, a 10-mg sample of eLHß was digested with trypsin, and the glycopeptides were isolated by reverse-phase HPLC. A fraction consisting of purified CTP-140-149 was coupled to Sequelon-aryl amine (AA) and sequenced. CTP-140-149 could not be obtained from eCGß.

Identification of O-Glycosylation Sites by Covalent Edman Degradation

O-glycosylation sites were identified by covalent sequencing [14]. Glycopeptides were coupled either to Sequelon-diisothiocyanate (DITC) or Sequelon-AA membranes (Millipore, Bedford, MA) and subjected to automated Edman degradation using a MilliGen model 6600 ProSequencer (Millipore). In initial experiments, a model 600 HPLC system (Waters), including a model 890 multiwavelength absorbance monitor, was employed to monitor the HPLC effluent simultaneously at two wavelengths. The absorbance at 269 nm was used to detect most phenylthiohydantoin (>PhNCS) derivatives, and absorbance at 313 nm was employed to detect the dehydro derivatives of >PhNCS-Thr and >PhNCS-Ser. In later experiments, a model 625 HPLC system (Waters) was employed, which included a model 486 tunable absorbance detector. A composite chromatogram was obtained in this manner. During most of the chromatogram, the absorbance at 269 nm was monitored, except between 10.8 and 11.35 min, when the wavelength was changed to 313 nm to monitor the dehydro derivatives of >PhNCS-Thr and >PhNCS-Ser.

Estimation of O-Glycosylation Site Occupancy by Carbohydrate Composition Analysis

Because the recoveries of >PhNCS-Thr and >PhNCS-Ser were very low during automated Edman degradation [23], it was necessary to measure the carbohydrate or the glycosylated >PhNCS-Thr/Ser derivatives to estimate the degree of occupancy at each O-glycosylation site. When we measured the carbohydrate released during automated Edman degradation, we found that it was released at each cycle regardless of whether it was a potential glycosylation site. Better results were obtained when we measured the carbohydrate associated with glycosylated >PhNCS-amino acid derivatives. Samples coupled to either Sequelon-DITC or Sequlelon-AA membranes were subjected to automated Edman degradation as described above. We collected portions of the >PhNCS-amino acid chromatogram corresponding to the breakthrough fraction (3.5–4.5 min) and the interval (6–9 min) where glycosylated >PhNCS-Thr and >PhNCS-Ser derivatives were eluted in 10- x 75-mm hydrolysis tubes. The volatile HPLC solvent consisting of 0.035 M ammonium acetate, pH 4.9, and 31–48% acetonitrile was removed under vacuum. Aliquots (200 µl) of 4 N TFA were added to each tube, the tubes were sealed, and the contents were hydrolyzed for 4 h at 100°C. Following hydrolysis, the tubes were opened and dried under vacuum. Hydrolysates were dissolved in 25 µl of water, and 20-µl samples were subjected to neutral and amino sugar analysis using a carbohydrate analyzer/ion chromatograph (Dionex, Sunnyvale, CA) [24]. The total galactosamine value was divided by the expected yield of Ser or Thr, based on the repetitive yield determined for the sequencer experiment. We also calculated the extent of glycosylation using the >PhNCS-amino acid chromatography data alone according to the method of Pisano et al. [25]. The combined areas of the two >PhNCS-Ser peaks or the four >PhNCS-Thr peaks were multiplied by 1.25 and then normalized to the >PhNCS-valine peak in the standard chromatogram. The amount of glycosylated amino acid derivative was divided by the expected amount of Ser or Thr. The repetitive yield calculated for the sequencer experiment was used to estimate the expected yield at each cycle. We employed the published conversion factor because partial glycosylation was the norm for the equine gonadotropins; therefore, our samples did not provide 100% glycosylated Thr and Ser residues that could be employed to calculate our own conversion factors.

RESULTS

Preparation of eLH/CGßt Hybrids

Equine {alpha} subunit preparations were combined with truncated ß subunit derivatives, eLHßt and eCGßt, and the heterodimer fractions were separated from unassociated subunit fractions by Superdex 75 chromatography. Purification of eLHßt hybrids is illustrated in Figure 1. Based on relative peak heights, the proportion of the unnassociated subunit fraction in the reassociation mixture increased from 14% to 24% as the carbohydrate content of the {alpha} subunit increased from 28% for eLH{alpha} to 41% for eCG{alpha}. The heterodimer peak constituted 60–72% of the total material in the chromatogram, which was comparable to results obtained following association of eLH{alpha} with intact eLHß and was more efficient than associating eFSH{alpha} or eCG{alpha} with intact eLHß in a previous study [9]. Because the other components were removed by the chromatographic procedure, we were able to eliminate the effect of subunit association efficiency on biological activity determinations.



View larger version (19K):
[in this window]
[in a new window]
  		FIG. 1. Isolation of truncated eLHß hybrids by Superdex 75 gel filtration. The bar indicates the portion of the chromatogram pooled and lyophilized to obtain the equine {alpha}:eLHßt dimer fraction. A) eLH{alpha}:eLHßt. B) eFSH{alpha}:eLHßt. C) eCG{alpha}:eLHßt

LH Radioligand and Leydig Cell Bioassay

Intact eCG-L, the most active eCG preparation available for these studies [8], exhibited only 13% as much rat LH receptor-binding activity as eLH (Fig. 2A and Table 1). The intact eLHß hybrids were all less active than intact eLH because of the inactive, unassociated eLHß present in each hybrid preparation [26]. The highest LH receptor-binding activity was exhibited by eLH{alpha}:eLHß (0.6 x eLH). Although eFSH{alpha}:eLHß was less active than eLH{alpha}:eLHß, it was more active than eCG{alpha}:eLHß, which had lower activity than eCG-L. Elimination of the C-terminal extension permitting separation of the heterodimer fraction from unassociated subunits revealed the impact of the ß subunit. The intact eCGß subunit produced a 50% reduction in the LH receptor-binding activity of eCG-L as compared with eCG{alpha}:eCGßt. The fact that the eCG{alpha}:eLHßt preparation was more active than eCG-L indicated that the eCGßC-terminal extension also contributed to the reduced LH receptor binding of eCG.



View larger version (27K):
[in this window]
[in a new window]
  		FIG. 2. LH receptor binding and steroidogenesis assay results. A) Rat testis LH radioligand assay employing 125I-hCG and rat testis homogenate. B) Leydig cell assay of intact eLHß hybrids. C) Leydig cell assay of truncated eCGß hybrids


View this table:
[in this window]
[in a new window]
  		TABLE 1. LH biological activities of truncated eLH{ß} and eCG{ß} hybrids. The program Allfit was employed to determine the relative potency and 95% confidence limits for a representative LH RLA and two Leydig cell LH bioassays (Fig. 2)

A progressive decrease in LH receptor-binding activity associated with increasing size of {alpha} subunit oligosaccharide was observed for both sets of truncated ß subunit hybrids. The eLH{alpha}:eLHßt preparation was 0.8 x that of eLH and was more active than eFSH{alpha}:eLHßt, which was more active than eCG{alpha}:eLHßt. The same pattern of {alpha}-specific reduced LH receptor-binding activity was observed with truncated eCGß hybrids. The most active eCGßt preparation was eLH{alpha}:eCGßt, which was more active than intact eLH itself, followed by eFSH{alpha}:eCGßt and eCG{alpha}:eCGßt. Whenever the same {alpha} subunit preparation was combined with either eLHßt or eCGßt, the resulting hybrids exhibited very similar LH receptor-binding activities. The most potent hybrid hormone preparations were eLH{alpha}:eLHßt and eLH{alpha}:eCGßt; their inhibition curves bracketed that of eLH. The eFSH{alpha}:eLHßt and eFSH{alpha}:eCGßt were the next most active preparations, and the eCG{alpha}:eLHßt and eCG{alpha}:eCGßt hybrids were the least active. Thus, removal of the O-glycosylated C-terminal extension from both eLHß and eCGß virtually eliminated the receptor-binding differences between hormone preparations possessing either of the truncated ß subunit preparations and the same {alpha} subunit preparation.

The LH activities of intact eLHß hybrids (Fig. 2B) and truncated eCGßt hybrids (Fig. 2C) were compared with that of intact eLH in the rat testis Leydig cell assay. Insufficient amounts of truncated eLHßt hybrids remained due to extensive testing in rat, human, and chicken FSH receptor-binding assays [9]. Comparing the LH receptor binding ID50 values with the steroidogenic EC50 values, the hybrids showed >100-fold amplification of the steroidogenic response, indicating that fully functional LH hybrids had been prepared in each case. There was no difference in the steroidogenic potencies of the intact eLHß hybrids. For the truncated eCGßt hybrids, both eLH{alpha}:eCGßt and eFSH{alpha}:eCGßt were as active as eLH (Table 1). The third hybrid, eCG{alpha}:eCGßt, exhibited a little more than half the activity of these preparations, consistent with its reduced LH receptor-binding activity.

Isolation and Characterization of O-Glycosylated Peptides

Identification of the O-glycosylation sites in eLHß and eCGß involved isolation and solid-phase Edman degradation of a series of progressively shorter C-terminal glycopeptides, beginning with a 50-amino acid residue peptide that possessed all of the O-glycosylation sites. Results of peptide isolation from RCM-eLHß are illustrated in Figure 3, and similar results were obtained with RCM-eCGß. Sephacryl S-200 chromatography following V8 protease digestion of an intact RCM-eLHß preparation under conditions permitting cleavage at both aspartic acid and glutamic acid residues [27] yielded the expected broad peak containing the O-glycosylated peptide fraction [22]. However, N-terminal sequence determination revealed that this fraction contained two peptides, CTP-62-149 and CTP-100-149. These peptides were separated by reverse-phase HPLC (Fig. 3B). Purified CTP-100-149 was used to characterize glycosylation of the sequence Ser115-Ser116-Ser117-Ser118. CTP-62-149 was subjected to mild acid hydrolysis, and the resulting fragments were separated by reverse-phase HPLC (Fig. 3C). Fraction A consisted of carbohydrate released by the hydrolysis procedure, and CTP-121-149 was recovered from fraction B. We attempted to obtain a tryptic C-terminal decapeptide by digestion of intact eLHß and eCGß to assess glycosylation of Thr148-Ser149 (Fig. 4). This approach was successfully applied to eLHß; however, eCGß was resistant even to prolonged digestion with 10% (w/w) trypsin (Fig. 4C). The dipeptide Thr148-Ser149 was released from DITC-coupled eCGß CTP-121-149 by automated Edman degradation as described in Materials and Methods.



View larger version (24K):
[in this window]
[in a new window]
  		FIG. 3. Isolation of C-terminal glycopeptides from RCM-eLHß. A) Sephacryl S-200 chromatography of S. aureus V8 protease digest of RCM-eLHß. The first peak eluted at 424 ml was not protein, according to SDS-PAGE analysis. The open bar indicates the portion of the chromatograms pooled to obtain the C-terminal glycopeptides. B) Separation of glycopeptides CTP-100-149 and CTP-62-149 (fractions I and II, respectively) obtained from the chromatogram in A by reverse-phase HPLC. The Vydac C-18 column was equilibrated with 0.01% TFA/water. The chromatogram was developed with a linear gradient of 0.01% TFA/acetonitrile as indicated by the dashed lines, using a flow rate of 1 ml/min. The solid bars indicate the portions of the chromatogram pooled to obtain each glycopeptide. C) Isolation of CTP-121-149 by reverse-phase HPLC of the products obtained by mild acid hydrolysis of CTP-62-149. The same solvents and flow rates were employed as in B. The gradient is indicated by the dashed line. The open bars indicate portions of the chromatogram pooled to obtain carbohydrate and glycopeptide products of the hydrolytic procedure



View larger version (23K):
[in this window]
[in a new window]
  		FIG. 4. Isolation of CTP-140-149 from tryptic digests of intact eCG and eLH ß-subunit preparations. Samples of eCGß and eLHß were incubated with trypsin and chromatographed on a Vydac C-18 column equilibrated with 0.01% TFA/water. A) Because of C-terminal heterogeneity, three CTP-140-149 fractions were isolated from eLHß, as indicated. Inset: Chromatography of 10-mg eLHß tyrptic digest. The solid bars indicate portions of the chromatograms pooled to obtain CTP-140-149. B) Tryptic digest of eCGß showing a single CTP-140-147 fraction at an enzyme/subtrate ratio of 1% trypsin. The solid bar shows the portion of the chromatogram that was pooled and subjected to a second round of tryptic digestion. C) No additional peptides were released when digestion was repeated for 48 h with 10% trypsin.

Carbohydrate composition analysis of eLHß and eCGß CTP-100-149 glycopeptides suggested glycosylation at five sites based on 4.7 mol galactosamine/mol eCGß CTP-100-149, whereas the 2.4 mol galactosamine/mol glycopeptide composition of eLHß CTP-100-149 indicated only three glycosylation sites (Table 2). The very high galactose and galactosamine content of the eCGß CTP-100-149 preparation was consistent with the presence of lactosamine repeats, although the sialic acid content was lower than expected for branched disialylated structures that have been reported [28]. Mild acid hydrolysis of CTP-62-149 quantitatively removed sialic acid, as reported previously [3]. This removal appeared in the breakthrough fractions following reverse-phase HPLC along with other portions of the oligosaccharide. Fifty-four percent of the galactose and 66% of the glucosamine but only 5% of the galactosamine recovered from the eCGß CTP-121-149 chromatogram were found in the 4 N TFA hydrolysate of this fraction, suggesting that the lactosamine repeats present in eCGßO-linked oligosaccharides [28] were also acid labile. The carbohydrate compositions for CTP-121-149 obtained from eCGß revealed that half the galactose and glucosamine and all of the sialic acid were absent. The reduction of galactosamine occurred to a lesser extent but suggested the loss of one glycosylation site. Elimination of one glycosylation site in eLHß was also indicated by comparing the galactosamine composition of eLHß CTP-100-149 with that of CTP-121-149. The composition of eLHß CTP-140-149 reflected that of a single glycosylation site.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Carbohydrate compositions of eCG{ß} and eLH{ß} C-terminal glycopeptides. Neutral and amino sugar analysis as well as sialic acid analysis were performed with a carbohydrate analyzer (Dionex, Sunnyvale, CA). Peptide concentrations were determined by amino acid analysis using the Pico·Tag (Waters, Milford, MA) method. Values are mol monosaccharide/mol peptide

Glycosylation of individual Ser and Thr residues was evaluated by solid-phase Edman degradation and carbohydrate composition analysis as described in Materials and Methods. Glycosylated >PhNCS-Ser was indicated by two peaks that emerged before >PhNCS-Ser itself (Fig. 5). These peaks were absent from >PhNCS chromatograms associated with nonglycosylated Ser residues, as previously reported [29]. Glycosylated Thr residues yielded four peaks that emerged ahead of >PhNCS-Thr. The >PhNCS-Thr chromatograms for eLHß Thr residues 127 and 133 differed from the corresponding Thr residues in eCGß by the shoulders associated with the first two peaks (compare Fig. 5, A and B). It is not known what oligosaccharide structural feature was associated with these additional peaks; however, it was absent from eLHß Thr residues 129 and 131 (data not shown).



View larger version (31K):
[in this window]
[in a new window]
  		FIG. 5. Comparison of eCGß and eLHß O-glycosylation at Thr127 and Ser128. Representative >PhNCS chromatograms obtained during automated solid-phase Edman degradation of DITC-coupled CTP-121-149 derived from each hormone. Positions of the glycosylated >PhNCS-Ser and >PhNCS-Thr peaks are indicated by the arrows. Shoulders present on the two larger glycosylated >PhNCS-Thr peaks obtained from eLHß were absent in the eCGß Thr127 chromatogram. The solid bars indicate portions of the chromatograms that were collected and subjected to carbohydrate analysis. A) eCGß Thr127. B) eLHß Thr127. C) eCGß Ser128. D) eLHß Ser128

Analysis of CTP-100-149 revealed that only Ser118 in the conserved sequence Ser115-Ser116-Ser117-Ser118 was glycosylated. For technical reasons, it was not possible to estimate the extent of glycosylation of this Ser residue. However differences in carbohydrate composition between CTP-100-149 and CTP-121-149 suggested 100% carbohydrate occupancy of this site, and the complete resistance of Lys119 to tryptic digestion provided additional support for this conclusion. Partial glycosylation of most Thr and Ser residues was observed with CTP-121-149. Quantitative analysis of O-glycosylation using the glycosylated >PhNCS amino acid derivative peaks or the galactosamine obtained from hydrolysates of the same region of the >PhNCS chromatograms during extensive analysis of CTP-121-149 preparations provided comparable results in most cases (Fig. 6). Larger variations in galactosamine recoveries were associated with apparent differences in average glycosylation estimates. These values ranged from 10% to 77% for eLHß and from 20% to 100% for eCGß. The most heavily glycosylated segment of eLHß consisted of Thr127-Thr131, which was also heavily glycosylated in eCGß. eCGß was also heavily glycosylated at Ser140-Ser141, consistent with its resistance to prolonged tryptic digestion.



View larger version (32K):
[in this window]
[in a new window]
  		FIG. 6. Comparison of percentage of glycosylation at each O-glycosylation site estimated by measuring galactosamine recovered from the HPLC chromatogram or calculated from the size of the >PhNCS amino acid peaks corresponding to glycosylated Thr and Ser. Bars with standard deviations indicated represent the means of three determinations. Galactosamine composition data for eCGß were successfully obtained from only one experiment. A) Glycosylation of eLHß CTP-121-149. B) Glycosylation of eCGß CTP-121-149

Analysis of the glycosylation status at the C-terminal ends required different strategies for each protein. We were able to isolate a decapeptide released by tryptic digestion of an intact eLHß preparation. Sequence analysis of CTP-140-149 recovered from fraction 7 (Fig. 4A) indicated the absence of glycosylated Ser residues at the N-terminus, the absence of glycosylation at Thr148, and the presence of carbohydrate at Ser149, and composition analysis indicated 80% glycosylation of Ser149 (Table 2). This experiment was repeated on a larger scale, and although the CTP-140-149 preparation recovered from fraction 10 (Fig. 4A, inset) was glycosylated at both Ser140 and Ser148, the absence of glycosylated Thr148 was confirmed (Fig. 7). To complete the analysis of eCGß, we obtained the C-terminal dipeptide Thr148-Ser149 by Edman degradation of CTP-121-140. Manual Edman degradation of the dipeptide revealed no glycosylated >PhNCS-Thr following the first degradation cycle, but glycosylated >PhNCS-Ser was observed following the second cycle (data not shown).



View larger version (16K):
[in this window]
[in a new window]
  		FIG. 7. Examination of O-glycosylation sites on eLHß CTP-140-149. The peptide isolated from peak 10 in Figure 4A (inset) was coupled to a Sequelon-AA membrane and subjected to 10 cycles of solid-phase Edman degradation. Chromatograms for the four hydroxy-amino acids found in this peptide are shown. Glycosylated Ser149 was indicated by a single peak instead of the usual pair of peaks (arrow)

DISCUSSION

Increased LH receptor-binding activity of eCG hybrid hormone preparations possessing a truncated eCGß subunit differed from results reported for similar modifications to eLH and hCG. Deletion of the C-terminal extension from eLHß and hCGß by mild acid hydrolysis [2, 3] and by mutation of recombinant hCGß [4, 5] had little effect on the in vitro biological activities of the truncated eLH, hCG, and recombinant hCG derivatives. These results led to the conclusion that the C-terminal extension only affected the activities of these hormones in vivo. Prolonged circulatory survival and enhanced in vivo potency following introduction of the CTP into both LH and FSH have been experimentally confirmed [6, 7]. However, truncation of the hCGß C-terminus by site-directed mutagenesis affected N-linked oligosaccharide processing, thereby revealing an intracellular function associated with this structural element [30, 31]. The results reported herein were consistent with recent studies involving eCG isoforms, which revealed that the high degree of O-glycosylation associated with eCG-H reduced its LH and FSH receptor-binding activity [8]. In the present study, we demonstrated that differences inO-glycosylation between eLH and eCG accounted for the eCGß subunit contribution to the reduced LH receptor-binding activity of eCG. This result was not demonstrated in a straightforward manner. Progressively increased amounts of unassociated eLHß present in hybrid hormone preparations, eLH{alpha}:eLHß, eFSH{alpha}:eLHß, and eCG{alpha}:eLHß, combined with the inhibitory effects of increased {alpha}Asn56 oligosaccharide size to suggest that {alpha}Asn56 oligosaccharide size was the only factor responsible for the reduced LH receptor-binding activity of eCG, because eCG{alpha}:eLHß exhibited the same reduced level of receptor-binding activity. However, the eCG{alpha}:eLHß hybrid preparation also possessed the highest content of inactive eLHß. (For a complete discussion of the {alpha}Asn56 oligsoaccharide structures responsible for reduced receptor-binding activity, see Butnev et al. [9]. The present discussion is restricted to the effects of O-glycosylation.) Removal of the C-terminal extension from eLHß and eCGß permitted separation of the heterodimer from unassociated subunits. When this confounding factor was eliminated, all the truncated ß subunit hybrids were at least twofold more active than eCG-L, the most active of the eCG isoforms. Despite the fact that the mass of eLHßt was 1053 mass units less than that of eCGßt [9], the LH receptor-binding activities of hybrids prepared by combining the truncated ß subunit preparations with the same {alpha} subunit preparation were virtually identical. This result was consistently obtained with all three equine {alpha} subunit preparations. The influence of ßAsn13 oligosaccharide on receptor-binding activity was probably insignificant because of overall similarities in their overall structures. The most abundant N-linked oligosaccharides obtained from eLHß and eCGß were biantennary [10, 12]. In the case of eCGß, these oligosaccharides were quantitatively desialylated by the mild acid treatment [13], whereas the sulfate moiety in eLHß oligosaccharides may have been only partially released because of greater resistance to mild hydrolysis conditions [32]. Even if substantial hydrolysis of the Asn13 oligosaccharides had occurred, receptor binding should not have been affected. Chemically deglycosylated oLHß, which possessed only the GlcNAc2 remnant of the Asn13 oligosaccharide, combined with intact oLH{alpha} to produce an oLH derivative with the same LH receptor-binding activity as intact oLH [33]. The most likely explanation for the ß subunit contribution to the reduced LH receptor-binding activity of eCG as compared with eLH was differential O-glycosylation of C-terminal residues 121–149. This differential O-glycosylation could have been the result of utilization of more potential glycosyation sites in eCGß as compared with eLHß, a greater occupancy of glycosylation sites in eCGß than in eLHß, or larger oligosaccharides attached to eCGß than to eLHß.

Qualitative assessment of O-glycosylation revealed 12 O-glycosylation sites in both hormones, a surprisingly large number (Fig. 8). Carbohydrate attachment to Ser118, Ser123, Thr127, Ser128, Thr129, Ser130, Thr131, Thr133, Ser137, Ser140, Ser141, and Ser149 was observed in the >PhNCS chromatograms and confirmed by increased galactosamine recovered from hydrolysates of the region of the chromatogram where the diagnostic peaks emerged. Previous estimates of the number of O-glycosylation sites based on carbohydrate composition [13] or yield of O-linked oligosaccharide [10] greatly underestimated the actual number because of the partial glycosylation observed in the present study.



View larger version (32K):
[in this window]
[in a new window]
  		FIG. 8. Comparison of eLHß and eCGß O-glycosylation, summarizing the analysis of all four glycopeptides, CTP-100-149, CTP-121-149, CTP-140-149, and CTP-148-149, indicated by the bars at the bottom of the figure. The open bars indicate glycosylation estimates for peptides where repetitive yields were unsatisfactory. Tryptic digestion indicated quantitative glycosylation at Ser118. The results for residues 121–147 are the means of three experiments, and the standard deviations are indicated. Quantitative glycosylation at Ser149 was indicated by its resistance to carboxypeptidase digestion and the very low coupling efficiency to arylamine membranes. The 80% glycosylation indicated for eLHß Ser149 was derived from data in Table 2

Quantitative estimates of the extent of glycosylation at individual Ser or Thr residues was absolutely dependent on accurate determination of the repetitive yield obtained during automated Edman degradation. Accordingly, these estimates were restricted to 10 of the 12 glycosylation sites that could be assessed in the course of experiments involving CTP-121-149. Fortunately, estimates for Ser118 and Ser149 could be obtained by other means. In both eLHß and eCGß, Ser118 appeared to be 100% glycosylated based on the difference in galactosamine composition between CTP-100-149 and CTP-121-149 derived from both subunit preparations and the complete resistance of lysine (Lys)119 to cleavage by trypsin (Fig. 4). Ser149 was reportedly 100% glycosylated; carboxypeptidase digestion failed to release any amino acids from eCGß and released only isoleucine146 from 20% of the eLHß preparation that lacked the C-terminal tripeptide Lys-Thr-Ser [22, 34]. We encountered extremely low coupling efficiencies when eCGß CTP-121-149 was coupled to Sequelon-AA membranes. Because Ser149 provided the only carboxyl group in the entire peptide, steric hindrance due to carbohydrate attached to its side chain appeared to be responsible for the reduced coupling efficiency. The composition of eLHß CTP-140-149 indicated 80% glycosylation of Ser149, which was the only glycosylated residue in this peptide preparation.

Extensive analysis of CTP-121-149 revealed a higher percentage of glycosylation in eCGß than was observed at four positions: Thr133, Ser137, and especially Ser140 and Ser141 (Fig. 8). Similar levels of glycosylation in both proteins were observed at the remaining O-linked sites. Despite the dibasic residues separating Ser137 from Ser140, these oligosaccharides largely protected eCGß from digestion with trypsin or any other protease [34]. Thus, limited proteolysis is unlikely to provide eCGß derivatives that could be used to determine if the greater extent of glycosylation at Ser140 and Ser141 contributed to the reduced receptor-binding activity of eCG.

Oligosaccharide size was another major difference between eCGß O-glycosylation and eLHß O-glycosylation. Although eLHß appeared to possess only core type 2 oligosaccharides, consisting of GalNAc, GlcNAc, galactosamine, and sialic acid, eCGß O-linked oligosaccharides had a very high content of GlcNAc and galactosamine. This finding was consistent with the presence of lactosamine repeats in eCGß O-linked oligosaccharides [28]. Some of these lactosamine repeats have been found to be sensitive to endo-ß-galactosidase digestion, particularly those in eCG-L and eCG-M. These repeats affected both the appearance and the electrophoretic mobility of the eCGß band following SDS-PAGE under reducing conditions [8]. For example, the molecular weight of the major eCGß band decreased from 40 000–50 000 to 35 000. However, glycosidase-sensitive lactosamine repeats were not entirely responsible for the differences in O-glycosylation between eLH and eCG ß subunits. Although the more active eCG preparations, eCG-L and eCG-M, almost doubled their FSH receptor-binding activities and eCG-H tripled its activity, they all remained lower in activity than eLH [8]. Future studies will have to be directed at characterizing the structures of oligosaccharides found at individual glycosylation sites to identify the functionally significant differences between eLH and eCG. Their location on adjacent residues will make this task very difficult because conventional glycopeptide isolation techniques cannot be employed.

Comparison of the primate and equid CGß amino acid sequences indicated an overall similarity in glycosylation patterns within each group, with more species-specific differences in the primate sequences (Fig. 9). Human CGß is N-glycosylated at Asn13 and Asn30. Baboon CGß possesses both potential glycosylation sites, and the predicted macaque CGß amino acid sequence lacks the Asn13 glycosylation site but substitutes a potential N-glycosylation site at the location of O-glycosylated Ser127 in hCGß. The horse LH/CGß subunits are N-glycosylated only at Asn13 [22, 34]. The predicted sequences for the donkey and zebra LH/CGß subunits indicate the same pattern of N-glycosylation. O-glycosylation of hCGß Ser residues 121, 127, 132, and 138 has been experimentally defined by several laboratories [35–37]. The predicted amino acid sequences for baboon and macaque CGß subunits suggest the baboon CGß possesses three of the four O-glycosylation sites found in hCGß. The sequence for the macaque CGß differed at three sites, including substitution of an N-glycosylation consensus sequence at one position where hCGß is O-glycosylated. The zebra and horse ß subunit sequences were 95% identical with the zebra LH/CGß subunit possessing the same Ser and Thr residues as the horse. There was only one difference between the horse and donkey LH/CGß subunit sequences that could affect O-glycosylation. The predicted donkey sequence included cysteine (Cys) instead of Ser118. In view of the otherwise strict conservation of potential O-glycosylation sites and because of the potential for intermolecular cross-linking by disulfide bonds involving this single Cys residue, this possible substitution should be examined further. A single base change would convert the TGT sequence found in the donkey [38] to the TCT sequence found in the horse [39]. Therefore, this difference in amino acids may be a sequencing artifact. Sequence conservation beyond Cys110 is low in the glycoprotein hormone family. Typical of that, the primate CGß sequences showed only 50–71% sequence identity when residues 111–145 were compared. In contrast, the equid LH/CGß subunitC-terminal sequences were 92–97% identical, suggesting an important role for O-linked carbohydrate in these species. Chorionic gonadotropins have arisen twice during evolution. In the primates, gene duplication produced a CGß gene that incorporated a C-terminal extension as well as other mutations in the core ß subunit [40]. In the equids, a single LH/CGß subunit gene is expressed in both pituitary and placenta, producing LH and CG, respectively [39]. The human ligands for the LH receptor differ at both the protein (80% identity over residues 1–121) and carbohydrate levels. Their equine counterparts are distinguished only by their carbohydrate moieties; however, the horse LH receptor is at least fivefold more sensitive to glycosylation differences [41].



View larger version (15K):
[in this window]
[in a new window]
  		FIG. 9. Comparison of primate CGß and equid LH/CGß C-terminal amino acid sequences. The amino acid sequences were obtained from the Swiss-Prot database [42], and the mature sequences were aligned using the DNAStar module Megalign (Madison, WI). The alignment report was highlighted using Adobe Illustrator (Adobe Systems, San Jose, CA). The amino terminal 110 residues are indicated by solid bars and the N-linked oligosaccharides are indicated by branched figures. The experimentally determined O-glycosylation sites in hCGß and eLH/CGß are boxed. Potential glycosylation sites are indicated in the other sequences by shading. Accession numbers: eLH/CGß, P08751; donkey LH/CGß, P19794; zebra LH/CGß O46641; hCGß, P01233; babbon CGß, P07434; macaque CGß, P51500

ACKNOWLEDGMENTS

We are grateful to Ms. Vanda Baker and Ms. Diana Hogan for their excellent technical assistance. We thank Dr. Shafiq Khan, Texas Tech University, Lubbock, for the Leydig cell bioassay protocol, Dr. Gordon Niswender, Colorado State University, Ft. Collins, for providing the testosterone antibody, and NIDDK, NHPP, and Dr. A. F. Parlow for the oLH reference preparation.

FOOTNOTES

First decision: 6 June 2000.

1 This work was supported by NIH grants AG15428 and DK52383. Back

2 Correspondence: George R. Bousfield, Department of Biological Sciences, Box 26, Wichita State University, 1845 Fairmount, Wichita, KS 67260-0026. FAX: 316 978 3772; bousfiel{at}twsuvm.uc.twsu.edu Back

3 Current address: Department of Physiology and Biophysics, University of Iowa College of Medicine, Iowa City, IA 52242. Back

Accepted: August 15, 2000.

Received: May 11, 2000.

REFERENCES

  1. Bousfield GR, Perry WM, Ward DN. Gonadotropins: chemistry and biosynthesis. In: Knobil E, Neill JD (eds.), The Physiology of Reproduction, vol. 1, 2nd ed. New York: Raven Press; 1994: 1749–1792.
  2. Bousfield GR, Ward DN. Biologic activity of eLH and hCG following removal of the C-terminal glycopeptide. In: Program of the 68th annual meeting of the Endocrine Society; 1986; Anaheim, CA. Abstract 530.
  3. Bousfield GR, Liu WK, Ward DN. Effects of removal of carboxy-terminal extension from equine luteinizing hormone (LH) ß-subunit on LH and follicle-stimulating hormone receptor binding activities and LH steroidogenic activity in rat testicular Leydig cells. Endocrinology 1989; 124:379–387.[Abstract/Free Full Text]
  4. El-Deiry S, Kaetzel D, Kennedy G, Nilson J, Puett D. Site-directed mutagenesis of the human chorionic gonadotropin ß-subunit: bioactivity of a heterologous hormone, bovine {alpha}-human des-(122-145)ß. Mol Endocrinol 1989; 3:1523–1528.[Abstract/Free Full Text]
  5. Matzuk MM, Hsueh AJW, Lapolt P, Tsafriri A, Keene JL, Boime I. The biological role of the carboxyl-terminal extension of human chorionic gonadotropin ß-subunit. Endocrinology 1990; 126:376–383.[Abstract/Free Full Text]
  6. Fares FA, Suganuma N, Nishimori K, LaPolt PS, Hsueh AJW, Boime I. Design of a long-acting follitropin agonist by fusing the C-terminal sequence of the chorionic gonadotropin ß subunit to the follitropin ß subunit. Proc Natl Acad Sci U S A 1992; 89:4304–4308.[Abstract/Free Full Text]
  7. Risma K, Clay CM, Nett TM, Wagner T, Yun J, Nilson JH. Targeted overexpression of luteinizing hormone in transgenic mice leads to infertility, polycystic ovaries, and ovarian tumors. Proc Natl Acad Sci U S A 1995; 92:1322–1326.[Abstract/Free Full Text]
  8. Butnev VY, Gotschall RR, Baker VL, Moore WT, Bousfield GR. Negative influence of O-linked oligosaccharides of high molecular weight equine chorionic gonadotropin on its luteinizing hormone and follicle-stimulating hormone receptor-binding activities. Endocrinology 1996; 137:2530–2542.[Abstract]
  9. Butnev VY, Gotschall RR, Butnev VY, Baker VL, Moore WT, Bousfield GR. Hormone-specific inhibitory influence of {alpha}-subunit Asn56 oligosaccharide on in vitro subunit association and FSH receptor binding of equine gonadotropins. Biol Reprod 1998; 58:458–469.[Abstract/Free Full Text]
  10. Damm JBL, Hard K, Kammerling JP, van Dedem GWK, Vliegenthart FG. Structure determination of the major N- and O-linked carbohydrate chains of the ß subunit from equine chorionic gonadotropin. Eur J Biochem 1990; 189:175–183.[Medline]
  11. Matsui T, Sugino H, Miura M, Bousfield GR, Ward DN, Titani K, Mizuochi T. ß-Subunits of equine chorionic gonadotropin and luteinizing hormone with an identical amino acid sequence have different asparagine-linked oligosaccharide chains. Biochem Biophys Res Commun 1991; 174:940–945.[CrossRef][Medline]
  12. Matsui T, Mizouchi T, Titani K, Okinaga T, Hoshi M, Bousfield GR, Sugino H, Ward DN. Structural analysis of N-linked oligosaccharides from equine chorionic gonadotropin and lutropin ß-subunits. Biochemistry 1994; 33:14039–14048.[CrossRef][Medline]
  13. Bousfield GR, Sugino H, Ward DN. Demonstration of a COOH-terminal extension on equine lutropin by means of a common acid-labile bond in equine lutropin and equine chorionic gonadotropin. J Biol Chem 1985; 260:9531–9533.[Abstract/Free Full Text]
  14. Pisano A, Packer NH, Redmond JW, Williams KL, Gooley AA. Characterization of O-linked glycosylation motifs in the glycopeptide domain of bovine {kappa}-casein. Glycobiology 1994; 4:837–844.[Abstract/Free Full Text]
  15. Bennett EP, Hassan H, Mandel U, Hollingsworth MA, Akisawa N, Ikematsu Y, Merkx G, van Kessel AG, Olofsson S, Clausen H. Cloning and characterization of a close homologue of human UDP-N-acetyl-{alpha}-D-galactose:polypeptide N-acetylgalactosaminyltransferase-T3, designated GalNAc-T6: evidence for genetic but not functional redundancy. J Biol Chem 1999; 274:25362–25370.[Abstract/Free Full Text]
  16. Bousfield GR, Ward DN. Purification of lutropin and follitropin in high yield from horse pituitary glands. J Biol Chem 1984; 259:1911–1921.[Abstract/Free Full Text]
  17. Moore WT Jr, Ward DN. Pregnant mare serum gonadotropin: rapid isolation procedures for the purification of intact hormone and isolation of subunits. J Biol Chem 1980; 255:6923–6929.[Abstract/Free Full Text]
  18. DeLean A, Munson PJ, Rodbard D. Simultaneous analysis of families of sigmoidal curves: application to bioassay, radioligand assay, and physiological dose-response curves. Am J Physiol 1978; 235:E97–E102.
  19. Syed V, Khan SA, Ritzen EM. Stage-specific inhibition of interstitial cell testosterone secretion by rat seminiferous tubules in vitro. Mol Cell Endocrinol 1985; 40:257–264.[CrossRef][Medline]
  20. Khan SA, Keck C, Guderman T, Nieschlag E. Isolation of a protein from human ovarian follicular fluid which exerts major stimulatory effects on in vitro steroid production of testicular, ovarian, and adrenal cells. Endocrinology 1990; 126:3043–3052.[Abstract/Free Full Text]
  21. Gay VL, Kerlan JT. Serum LH and FSH following passive immunization against circulating testosterone in the intact male rat and in orchidectomized rats bearing subcutaneous silastic implants of testosterone. Arch Androl 1978; 1:257–266.[Medline]
  22. Bousfield GR, Liu WK, Sugino H, Ward DN. Structural studies on equine glycoprotein hormones: amino acid sequence of equine lutropin ß subunit. J Biol Chem 1987; 262:8610–8620.[Abstract/Free Full Text]
  23. Tarr GE. Improved manual sequencing methods. Methods Enzymol 1977; 47:335–357.[Medline]
  24. Gotschall RR, Bousfield GR. Oligosaccharide mapping reveals hormone-specific glycosylation patterns on equine gonadotropin {alpha}-subunit Asn56. Endocrinology 1996; 137:2543–2557.[Abstract]
  25. Pisano A, Jardine DR, Packer NH, Redmond JW, Williams KL, Gooley AA, Farnsworth V, Carlson W, Cartier III PK. Identifying sites of glycosylation in proteins. In: Townsend RR, Hotchkiss AT Jr (eds.), Techniques in Glycobiology. New York: Marcel Dekker; 1997: 299–320.
  26. Bousfield GR, Liu WK, Ward DN. Hybrids from equine LH: alpha enhances, beta diminishes activity. Mol Cell Endocrinol 1985; 40:69–77.[CrossRef][Medline]
  27. Houmard J, Drapeau GR. Staphylococcal protease: a proteolytic enzyme specific for glutamoyl bonds. Proc Natl Acad Sci U S A 1972; 69:3506–3509.[Abstract/Free Full Text]
  28. Hokke CH, Roosenboom MJH, Thomas-Oates JE, Kamerling JP, Vliegenthart JFG. Structure determination of the disialylated poly-(N-acetyllactosamine)-containing O-linked carbohydrate chains of equine chorionic gonadotropin. Glycoconj J 1994; 11:35–41.[CrossRef][Medline]
  29. Bousfield GR, Butnev VY, Gotschall RR, Baker VL, Moore WT. Structural features of mammalian gonadotropins. Mol Cell Endocrinol 1996; 125:3–19.[CrossRef][Medline]
  30. Muyan M, Furuhashi M, Sugahara T, Boime I. The carboxy-terminal region of the ß-subunits of luteinizing hormone and chorionic gonadotropin differentially influence secretion and assembly of the heterodimers. Mol Endocrinol 1996; 10:1678–1687.[Abstract/Free Full Text]
  31. Muyan M, Boime I. The carboxyl-terminal region is a determinant for the intracellular behavior of the chorionic gonadotropin ß subunit: effects on the processing of the Asn-linked oligosaccharides. Mol Endocrinol 1998; 12:766–772.[Abstract/Free Full Text]
  32. Ward DN, Wen T, Bousfield GR. Sulfate and phosphate analysis in glycoproteins and other biologic compounds using ion chromatography: application to glycoprotein hormone and sugar esters. J Chromatogr 1987; 398:255–264.[CrossRef][Medline]
  33. Liu WK, Young JD, Ward DN. Deglycosylated ovine lutropin: preparation and characterization by in vitro binding and steroidogenesis. Mol Cell Endocrinol 1984; 37:29–39.[CrossRef][Medline]
  34. Sugino H, Bousfield GR, Moore WT Jr, Ward DN. Structural studies on equine gonadotropins: amino acid sequence of equine chorionic gonadotropin ß-subunit. J Biol Chem 1987; 262:8603–8609.[Abstract/Free Full Text]
  35. Birken S, Canfield RE. Isolation and amino acid sequence of COOH-terminal fragments from the ß subunit of human choriogonadotropin. J Biol Chem 1977; 252:5386–5392.[Abstract/Free Full Text]
  36. Keutmann HT, Williams RM. Human chorionic gonadotropin: amino acid sequence of the hormone-specific COOH-terminal region. J Biol Chem 1977; 252:5393–5397.[Free Full Text]
  37. Kessler MJ, Mise T, Ghai RD, Bahl OP. Structure and location of the O-glycosidic carbohydrate units of human chorionic gonadotropin. J Biol Chem 1979; 254:7909–7914.[Free Full Text]
  38. Leigh SEA, Stewart F. Partial cDNA sequence for the donkey chorionic gonadotrophin-ß subunit suggests evolution from an ancestral LH-ß gene. J Mol Endocrinol 1990; 4:143–150.[Abstract/Free Full Text]
  39. Sherman GB, Wolfe MW, Farmerie TA, Clay CM, Threadgill DS, Sharp DC, Nilson JH. A single gene encodes the ß-subunits of equine luteinizing hormone and chorionic gonadotropin. Mol Endocrinol 1992; 6:951–959.[Abstract/Free Full Text]
  40. Talmadge K, Vamvakopolous NC, Fiddes JC. Evolution of the genes for the ß subunits of human chorionic gonadotropin and luteinizing hormone. Nature 1984; 307:37–40.[CrossRef][Medline]
  41. Stewart F, Allen WR. The binding of FSH, LH, and PMSG to equine gonadal tissues. J Reprod Fertil Suppl 1979; 27:431–440.
  42. Bairoch A, Apweiler R. The SWISS-PROT protein sequence database and its supplement TrEMBL in 2000. Nucleic Acids Res 2000; 28:45–48.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J Mol EndocrinolHome page
S. Legardinier, J.-C. Poirier, D. Klett, Y. Combarnous, and C. Cahoreau
Stability and biological activities of heterodimeric and single-chain equine LH/chorionic gonadotropin variants
J. Mol. Endocrinol., April 1, 2008; 40(4): 185 - 198.
[Abstract] [Full Text] [PDF]

Home page
 GlycobiologyHome page
S. Legardinier, D. Klett, J.-C. Poirier, Y. Combarnous, and C. Cahoreau
Mammalian-like nonsialyl complex-type N-glycosylation of equine gonadotropins in MimicTM insect cells
Glycobiology, August 1, 2005; 15(8): 776 - 790.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
S. Nakav, A. Jablonka-Shariff, S. Kaner, P. Chadna-Mohanty, H. E. Grotjan, and D. Ben-Menahem
The LH{beta} Gene of Several Mammals Embeds a Carboxyl-terminal Peptide-like Sequence Revealing a Critical Role for Mucin Oligosaccharides in the Evolution of Lutropin to Chorionic Gonadotropin in the Animal Phyla
J. Biol. Chem., April 29, 2005; 280(17): 16676 - 16684.
[Abstract] [Full Text] [PDF]

Home page
 J Mol EndocrinolHome page
S. Legardinier, M. Duonor-Cerutti, G. Devauchelle, Y. Combarnous, and C. Cahoreau
Biological activities of recombinant equine luteinizing hormone/chorionic gonadotropin (eLH/CG) expressed in Sf9 and Mimic insect cell lines
J. Mol. Endocrinol., February 1, 2005; 34(1): 47 - 60.
[Abstract] [Full Text] [PDF]

Home page
 J EndocrinolHome page
M. Saint-Dizier, F. Foulon-Gauze, F. Lecompte, Y. Combarnous, and M. Chopineau
Cloning and functional expression of the equine luteinizing hormone/chorionic gonadotrophin receptor
J. Endocrinol., December 1, 2004; 183(3): 551 - 559.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow My Folders
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Bousfield, G. R.
Right arrow Articles by Butnev, V. Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bousfield, G. R.
Right arrow Articles by Butnev, V. Y.
Agricola
Right arrow Articles by Bousfield, G. R.
Right arrow Articles by Butnev, V. Y.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS