help button home button Endocrine Society Molecular Endocrinology ENDO 08 Sessions Library
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Molecular Endocrinology, doi:10.1210/me.2005-0537
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
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 Download to citation manager
Right arrow Reprints, Permissions and Rights
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Jablonka-Shariff, A.
Right arrow Articles by Boime, I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Jablonka-Shariff, A.
Right arrow Articles by Boime, I.
Molecular Endocrinology 20 (6): 1437-1446
Copyright © 2006 by The Endocrine Society

Single-Chain, Triple-Domain Gonadotropin Analogs with Disulfide Bond Mutations in the {alpha}-Subunit Elicit Dual Follitropin and Lutropin Activities in Vivo

Albina Jablonka-Shariff1, T. Rajendra Kumar1, Joshua Eklund, Anna Comstock and Irving Boime

Department of Molecular Biology and Pharmacology (A.J.-S., A.C., I.B.), Washington University School of Medicine, St. Louis, Missouri 63110; and Departments of Molecular and Integrative Physiology (T.R.K., J.E.), and Pathology and Laboratory Medicine (T.R.K.), University of Kansas Medical Center, Kansas City, Kansas 66160

Address all correspondence and requests for reprints to: Irving Boime, Department of Molecular Biology and Pharmacology, Washington University School of Medicine, St. Louis, Missouri 63110. E-mail: iboime{at}wustl.edu.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
The human glycoprotein hormones chorionic gonadotropin (CG), TSH, LH, and FSH are heterodimers composed of a common {alpha}-subunit and a hormone-specific ß-subunit. The subunits assemble noncovalently early in the secretory pathway. LH and FSH are synthesized in the same cell (pituitary gonadotrophs), and several of the {alpha}-subunit sequences required for association with either ß-subunit are different. Nevertheless, no ternary complexes are observed for LH and FSH in vivo, i.e. both ß-subunits assembled with a single {alpha}-subunit. To address whether the {alpha}-subunit can interact with more than one ß-subunit simultaneously, we genetically linked the FSHß- and CGß-subunit genes to the common {alpha}-subunit, resulting in a single-chain protein that exhibited both activities in vitro. These studies also indicated that the bifunctional triple-domain variant (FSHß-CGß-{alpha}), is secreted as two distinct bioactive populations each corresponding to a single activity, and each bearing the heterodimer-like contacts. Although the data are consistent with the known secretion events of gonadotropins from the pituitary, we could not exclude the possibility whether transient intermediates are generated in vivo in which the {alpha}-subunit shuttles between the two ß-subunits during early stages of accumulation in the endoplasmic reticulum. Therefore, constructs were engineered that would direct the synthesis of single-chain proteins completely devoid of heterodimer-like interactions but elicit both LH and FSH actions. These triple-domain, single-chain chimeras contain the FSHß- and CGß-subunits and an {alpha}-subunit with cystine bond mutations (cys10–60 or cys32–84), which are known to prevent heterodimer formation. Here we show that, despite disrupting the intersubunit interactions between the {alpha}- and both CGß- and FSHß-subunits, these mutated analogs exhibit both activities in vivo comparable to nonmutated triple-domain single chain. Such responses occurred despite the absence of quaternary contacts due to the disrupted bonds in the {alpha}-subunit. Thus, gonadotropin heterodimer assembly is critical for intracellular events, e.g. hormone-specific posttranslational modifications, but when heterodimers are present in the circulation, the {alpha}/ß-contacts are not a prerequisite for receptor recognition.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
A KEY FEATURE of the glycoprotein hormone family, which consists of chorionic gonadotropin (CG), FSH, LH, and TSH, is its heterodimeric structure. The hormones are composed of a common {alpha}-subunit associated noncovalently to a variable ß-subunit that confers the receptor binding specificity for each dimeric ligand, and the gonadal biological activities require the {alpha}/ß-heterodimer (1). The crystal structures of human (h)CG (2, 3) and FSH (4) showed that the two subunits are associated with each other along much of their surfaces, each subunit having remarkably similar folds with two hairpin loops at one end and a single loop at the other (2, 3, 4). This configuration, together with the heterodimeric requirement for receptor binding, implies that the complex quaternary structure of the gonadotropins is crucial for biological activity.

In contrast to these modeling studies, a number of recent in vitro observations using single-chain glycoprotein hormone variants, in which the DNA encoding the {alpha}- and ß-subunits are genetically linked, have suggested that the complete heterodimeric configuration is not essential for receptor binding (5, 6, 7, 8, 9). These single-chain variants of hFSH and hCG displayed normal receptor binding and signaling, despite molecular modeling, which demonstrates that some of the {alpha}-ß-subunit alignments seen in the crystal structure could not be maintained. Moreover, single chains containing mutations in the {alpha}- or ß-subunits that inhibit heterodimer formation result in biologically active analogs in vitro, suggesting that the glycoprotein hormone-receptor interactions are flexible (7, 8, 9).

Additional evidence indicating the extensive flexibility of the interactions between the ligand and the receptor was obtained from studies performed with a single chain bearing three domains (i.e., one {alpha}- and two ß-subunits in tandem), which show that the receptors recognize the {alpha}/ß-domains in different conformations (10). Thus, a triple-domain gene encoding FSHß-CGß-{alpha} protein displayed high-affinity binding to both of the corresponding human receptors. We also demonstrated that two distinct bioactive populations of the triple-domain chimera corresponding to a single hormone activity were secreted from transfected Chinese hamster ovary (CHO) cells (11). The ability of two ß-subunits to interact with one {alpha}-subunit in a single chain and display two different biologically active conformations has important physiological implications. Both FSH and LH are made in the same cell, i.e. in the gonadotroph, and no functional FSH/LH complexes have been identified in vivo. Moreover, FSH and LH exhibit differences in the rates of assembly of their corresponding subunits in the endoplasmic reticulum (12). It is unclear, however, how in the endoplasmic reticulum the ß-subunits sort out from each other and assemble with a common {alpha}-subunit. Once the {alpha}-ß contacts are formed, the resultant species are stable. On the other hand, in vivo it cannot be excluded that a dually active molecule is formed in which a single {alpha}-subunit shares its contact sites with both ß-subunits simultaneously. It is likely that such triple complex interactions would not be stable, but would be, rather, transitory heterodimeric complexes pro tem.

Thus, the question arose whether a triple-domain single molecule lacking heterodimeric contacts, presumably resembling transient biosynthetic intermediates generated in vivo, is dually active. To address this issue, we engineered triple-domain (FSHß-CGß-{alpha}) analogs in which neither ß-subunit domain will form heterodimeric contacts with the {alpha}-subunit. Our approach was to delete disulfide bonds in the {alpha}-subunit which, based an earlier results, demonstrated that such mutations prevent {alpha}/ß-subunit association, resulting in the absence of the heterodimer (13, 14). This strategy permits the design of one construct rather than the necessity for generating several different ß-subunit analogs.

Here we show that, despite the disruption of the intramolecular heterodimeric contacts in the analogs, the triple-domain, single-chain hormones elicit both CG/LH and FSH activities in vivo. The data imply that a single form of the analog demonstrates dual activity and that the presence of heterodimeric contact sites are not required for receptor recognition.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 
Previously we demonstrated that at least two biologically active species are generated during synthesis of the triple-domain, single-chain FSHß-CGß-{alpha} analog, each corresponding to a single activity (FSH or CG). To address the question of whether a single species can exhibit dual activity, we engineered a triple-domain chimera containing an {alpha}-subunit-bearing disulfide bond mutations, at either the cys10–60 or cys32–84 positions. These disulfide bonds have been shown previously to be required for {alpha}-subunit assembly with the ß-subunit (13, 14). Moreover, the cys residues that make up the 10–60 and 32–84 bonds in the {alpha}-subunit are not associated with specific epitopes recognized by FSH 54 and B 109 monoclonal antibodies (mAbs). The FSH 54 epitope is exclusively on loop 2 of the FSHß-subunit (Ebo Bos, personal communication), and B 109 epitope is not involved with the {alpha}-subunit (11). Thus, the immunoreactivity of the mutants to these mAbs will be conformationally sensitive (see below).

The synthesis of the analogs was examined in transfected cells labeled with [35S]cysteine. Equal aliquots of lysate/medium were precipitated with CGß antiserum, and the proteins were resolved on sodium dodecyl sulfate (SDS) gels (Fig. 1Go). It is evident that greater than 75% (±0.5) of the total synthesized unmutated analog is secreted (WT TR lanes 1 and 2). By contrast, the fraction of secreted cys10–60 (lanes 3 and 4) and cys32–84 (lanes 5 and 6) mutants was 22% (±2) and 18% (±0.6), respectively. The reduced secretion of the mutated chimeras is consistent with our previous results (13), which showed that the extent of secretion of the noncombined {alpha}-subunit containing the same disulfide mutants reported here is dramatically reduced.


Figure 1
View larger version (23K):
[in this window]
[in a new window]
  		Fig. 1. Synthesis of Triple-Domain Single Chain

Cells transfected with either the single-chain, wild-type FSHß-CGß-{alpha} (TR WT; lanes 1 and 2), or the analogs containing either the {alpha}-domain missing the {alpha}10–60 (lanes 3 and 4), or the {alpha}32–84 (lanes 5 and 6) disulfide bonds were incubated overnight with [35S]cysteine. Polyclonal CGß antiserum was used to precipitate lysate (L) and medium (M) fractions, and the reduced proteins were resolved by 12.5% SDS-PAGE. The apparent molecular mass of the secreted forms is 88,000 kDa.

 
Western Blot Analysis of the Chimeras
As discussed above, formation of the heterodimer is prevented with subunits bearing either the cys10–60 or cys32–84 mutations (13, 14). To probe whether the triple-domain analogs with these mutations contain heterodimeric-like contacts, Western blot analysis was performed using two dimer-specific mAbs: FSH 54 and B109 (Figs. 2Go and 3Go). These mAbs should distinguish the chimeric forms containing FSHß/CGß heterodimeric contacts from those in which the ß-subunit is in a nonassembled configuration. A Western blot containing heterodimer standard and the chimeras was analyzed with mAb FSH 54 (Fig. 2AGo). As expected, there was no significant binding of mAb to the noncombined FSHß-subunit (lane 2) or the chimeras containing the mutations in the {alpha}-subunit (lanes 3 and 4). No signals were seen despite prolonged exposure of the blot (data not shown). By contrast, there was extensive binding of the antibody to the native heterodimer (FSHdi, lane 1), recombinant FSH dimer (rFSH, lane 6), and nonmutated triple-domain single chain (TR WT, lane 5). That the lack of detection was not due to protein degradation is demonstrated by the signals observed with the FSHß 4B mAb (Fig. 2BGo), which recognizes epitopes in the noncombined and assembled forms of the FHß-subunit. It is clear that the mutants are now detected (lanes 8 and 9), and the noncombined form of the FHß-subunit in the dimer preparations is also observed (lanes 7 and 11).


Figure 2
View larger version (18K):
[in this window]
[in a new window]
  		Fig. 2. Immunoreactivity of the Triple-Domain and Native Hormones with FSH mAbs

Comparable amounts of media samples were subjected to Western blot using the antibody designated FSH 54 (panel A) which is FSH heterodimer specific, or FSHß 4B, which (panel B) recognizes unassembled and dimer forms of the FSHß-subunit. Lanes 6 and 11 contain purified rFSH. Lanes 1 and 7 show FSH dimer (FSHdi) in the conditioned medium from the transfected cells. TR WT denotes the triple-domain, wild-type chimera. The TR arrow indicated the migration of the triple-domain protein.

 

Figure 3
View larger version (27K):
[in this window]
[in a new window]
  		Fig. 3. Immunoreactivity of Triple-Domain and Native Hormones with hCG mAb

Comparable amounts of media samples from transfected cells were subjected to Western blot using the CG heterodimer-specific mAb B109 (panel A), and polyclonal CGß antiserum (CGß AS) (panel B). The CGß antiserum recognizes both the heterodimer and unassembled forms of the CGß-subunit. Lane 1 contains recombinant CG dimer (CGdi). Lane 2 shows only the recombinant CGß-subunit, and lanes 3–5 display the triple-domain single chains. The same sample order is shown for the screen with CGß polyclonal antiserum (panel B). Note that the antiserum detects the mutated chimeras and the free CGß-subunit. TR WT denotes the triple-domain, wild-type chimera.

 
In the case of the heterodimer-specific mAb B109 (Fig. 3AGo), the pattern of CG immunoreactivity was analogous to that seen for FSH screening. Only the CG heterodimer (lane 1) and the nonmutated triple domain (lane 5) are immunoreactive. The noncombined ß-subunit (lane 2) and the {alpha}-mutated triple-domain analogs (lanes 3 and 4) displayed no detectable immunoreactivity. Previously we demonstrated that single-chain chimeras bearing the CGß-subunit and the {alpha}-subunit mutants exhibited no significant immunoreactivity when probed with several CG dimer-specific mAbs (designated 26, 40, and 55; see Ref5). Similar results were observed here when the same CG heterodimer-specific mAbs were used (data not shown). The absence of signal of the mutant single chains was still evident after overexposure of the blots. When a replicate blot was probed with polyclonal CGß antiserum (Fig. 3BGo), both the unassembled CGß-subunit (lanes 7) and the mutant chimeras (lanes 8, 9), which did not react with the dimmer-specific mAb B109, are clearly observed. [There are some higher molecular weight species seen in lanes 5 and 8–10, which are presumably aggregates because they are not observed when the proteins are resolved under nonreduced conditions. Although their composition is unclear, such aggregation has been reported previously for native hCG and a variety of single-chain analogs (5, 7, 8, 9).]. Although it is clear that the mutants do not display detectable immunoreactivity to the heterodimer-specific mAbs, we could not exclude the possibility that the mutated {alpha}-domain formed heterodimer-like contacts but it was dissociated from the ß-subunit due to denaturation on immunoblots. To address this issue, we performed an immunoprecipitation before the blotting step in the absence of SDS and boiling (Fig. 4Go). The mAbs FSH54 (panel A) and B109 (panel B) were incubated with either the nonmutated or mutated triple-domain chimeras, and the precipitates were solubilized and then subjected to blotting with polyclonal {alpha}-antiserum that recognizes the wild-type and mutant chimeras (Fig. 4Go). It is evident that whereas the nonmutated chimeras were immunoreactive with either B109 or FSH54 (panels A and B, lane 1), no signals were seen for the mutants when precipitated with either mAbs (panel A and B; lanes 2 and 3). As expected, no signals were seen when nonimmune mouse IgG was used instead of the mAbs (lane 4). Taken together, the data with the chimeras bearing two distinct disulfide bond mutations at different regions in the {alpha}-subunit, with the same result, show that the intramolecular heterodimeric contacts in the triple-domain mutants have been disrupted. Thus, lack of the heterodimeric-like contacts were not the result of denaturation during immunoblotting.


Figure 4
View larger version (16K):
[in this window]
[in a new window]
  		Fig. 4. Immunoprecipitation and Western Blotting of Triple-Domain Chimeras

The triple-domain chimeras were immmunoprecipitated using protein A agarose according to a modified protocol described by Santa Cruz Biotechnology, Inc. (Santa Cruz, CA). The identical amounts of conditioned media containing chimeras were precleared with protein A-agarose and normal mouse IgG. The pellets were centrifuged at 2500 rpm at 4 C for 5 min, and the supernatant was treated with mAb54 (panel A) or mAbB109 (panel B) overnight at 4 C, and the precipitates were collected by centrifugation at 2500 rpm for 5 min. The pellets were washed with PBS and solubilized in SDS buffer in the absence of heat denaturation, and proteins were separated on a 12.5% SDS gel and blotted with polyclonal {alpha}-antiserum. Nonimmune IgG and WT-TR correspond to the nonimmue mouse IgG and the nonmutated triple-domain chimeras, respectively. The arrowhead indicates the position of the triple-domain chimera.

 
FSH Bioactivity
Ovarian Weight-Gain Response.
Having demonstrated the synthesis and secretion of the mutant chimeras, their biological activities were next examined in vivo. Media containing the analogs (2.5 I.U. or 5 I.U. immunoreactive FSH equivalent) were injected into immature mice, and 48 h later the ovarian weights were determined (Fig. 5Go). In this assay, pregnant mare serum gonadotropin (PMSG) is used as a positive control for FSH action. The two mutants elicited a dose-dependent increase in ovarian weight comparable with either PMSG or the nonmutated triple-domain analog (Fig. 5Go). As expected, no ovarian weight changes were seen after an injection of comparable volume of PBS, or control media containing only the human {alpha}-subunit. The latter control demonstrates that the biological effect seen is not related to the presence of CHO cell-secreted proteins in the media.


Figure 5
View larger version (18K):
[in this window]
[in a new window]
  		Fig. 5. Bioassay of FSH in Vivo

Immature female mice (four to five per group) were injected with PBS or control medium or different doses of PMSG or the analogs, and 48 h later their ovaries were collected and weighed. The data shown are the mean ± SEM from three independent assays. Medium containing a comparable amount of nonassembled {alpha}-subunit is a negative control (Ctrl. Medium). The activities of the analogs compared with the controls (PBS, CHO media from nontransfected cells), and the dose responses were statistically significant; P < 0.05 by one-way ANOVA. TR WT denotes the triple-domain, wild-type chimera.

 
Ovarian Aromatase Induction Assay.
To more directly assay FSH activity by the analogs, we performed an ovarian aromatase induction assay (Fig. 6Go, A and B). Ovarian aromatase converts testosterone to estrogen and is exquisitely sensitive to FSH stimulation (15, 16). We have previously standardized a semiquantitative RT-PCR assay that measures aromatase expression in mouse ovaries (15). Clearly, PMSG and all of the triple-domain analogs induced the expression of aromatase RNA compared with the response obtained with different volumes of control medium (Fig. 6AGo). The induction of aromatase in the ovaries of experimental groups of mice reflects a true bioactivity of PMSG or the analogs, because an amplified product was not obtained in the negative control, which does not contain any cDNA (Fig. 6AGo, no cDNA). Moreover, both the nonmutant and mutated chimeras increased the expression of aromatase in a dose-dependent manner comparable to that seen with PMSG (Fig. 6BGo). The expression in the ovaries of the control medium-injected group is the baseline expression of aromatase RNA, typically observed in immature mice at this age (15, 16).


Figure 6
View larger version (34K):
[in this window]
[in a new window]
  		Fig. 6. RT-PCR Analysis of Ovarian Aromatase Expression

Immature female mice were injected with different volumes of control medium, or different does of PMSG, triple-domain, wild-type (WT TR), or the mutant analogs. Total ovarian RNA was analyzed by RT-PCR (panel A) as described in Materials and Methods. Note the induction of aromatase (Cyp19a1) (upper wells) with PMSG and all the triple-domain analogs in a representative RT-PCR gel. Amplification of cyclophilinA (CypA) was used as an internal control for the amount of RNA input in the experiment (lower wells). B, Densitometric scan of the ethidium bromide-stained gels is represented as a ratio of Cyp19a1/CypA. The data shown are mean ± SEM of two independent sets of experiments with three mice per each group. Both PMSG and the analogs at each dose showed significant induction of aromatase mRNA compared with that by control medium (P < 0.05). When the dose responses were compared, induction by PMSG and the 10–60 disulfide mutant analogs was significant (P < 0.05). Although WT and the 32–84 analogs at a dose of 5 IU demonstrated a trend of induction compared with the expression by 2.5 IU, the responses were not statistically significant (P > 0.05).

 
hCG Bioactivity: Superovulation of Immature Female Mice
Next, we assayed the hCG bioactivity of the analogs (Fig. 7Go). A typical superovulation protocol includes priming by PMSG of immature female mice that support ovarian follicle growth followed by an ovulatory dose of hCG (LH activity) that results in release of multiple ova into the oviducts (15). PMSG-primed immature female mice were injected with either hCG or different doses of the analogs, and the number of ova in the oviduct was counted the next day. The mutant analogs induced multiple ovulations similar to that seen by the native hCG heterodimer or the nonmutated triple-domain analog and displayed a dose-dependent hCG/LH-like activity (Fig. 7Go). Identical to the FSH assay, the negative controls (PBS or CHO medium containing only the {alpha}-subunit) did not cause superovulation in mice. Taken together these data confirm that, despite the disruption of these disulfide bonds in the {alpha}-subunit, the mutant single-chain analogs exhibit both FSH and CG/LH activities, similar to the nonmutated wild-type analog.


Figure 7
View larger version (19K):
[in this window]
[in a new window]
  		Fig. 7. CG Biological Activity of the Single Chains in Vivo

Immature female mice (three to four per group) were injected with 5 IU of PMSG, and 46 h later, injected with either PBS, control medium, or different doses of hCG or the analogs. Ova were collected from the oviducts 16–18 h later and counted. The data are from three independent experiments (nd = no released ova detected). The dose responses and differences of the chimeras compared with the control media were significant: P < 0.05 by one-way ANOVA. TR WT denotes the triple-domain, wild-type chimera. Ctrl., Control.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
MATERIALS AND METHODS
 REFERENCES
 
Numerous bioactive proteins contain cystine knots, a ring of cys residues comprised of two disulfide bonds through which a third disulfide penetrates (2, 3, 17, 18). Proteins containing this structure belong to a superfamily of growth factors including TGFß, nerve- and platelet-derived growth factors, and the subunits of the glycoprotein hormone family. Disrupting any of these disulfide bonds in either the glycoprotein hormone-specific ß- or common {alpha}-subunits dramatically reduces assembly and recovery of the heterodimer (13, 14, 19). In the case of the {alpha}-subunit, deleting the cys10–60 or cys32–84 bonds prevented assembly with the wild-type CGß-subunit, which resulted in the recovery of less than 5% of the secreted CG heterodimer (8, 13, 14).

Previously we showed that single-chain chimeras of the glycoprotein hormone family bearing two tandemly arranged FSHß- and CGß-subunits genetically linked to a single {alpha}-subunit displayed both FSH and CG/LH biological actions (10). It was demonstrated that two unique functional entities bearing heterodimeric-like contacts were formed during the synthesis of this chimera (11, 12). However, these experiments did not address whether or not a single-chain triple-domain molecule lacking {alpha}/ß-heterodimeric contacts exhibited dual activity (Fig. 7Go). Our present data demonstrate that when the heterodimeric contacts between the mutated {alpha}-subunit and the two ß-subunits are impaired, a single form of the analog is generated that elicits dual FSH and CG activities. Despite disrupting the intrachain quaternary relationships, the mutants displayed nearly comparable FSH and CG actions to the control standards. That the dimer-specific mAbs discriminated the {alpha}/ß-heterodimeric contacts in the nonmutated and the mutated forms and yet, they were nevertheless active, shows an uncoupling of those epitopes required for assembly and receptor recognition. Although heterodimeric contacts were not recognized by the mAbs used in our studies, we cannot exclude the existence of other regions of {alpha}-ß association not detected here. More biophysical analyses will be required to confirm to what extent such minimal contacts exist in single-chain mutants.

The biological implications for such functional analogs are significant. Given the structural permissiveness in the signaling responses to these analogs, the question arises as to why the gonadotropins evolved as heterodimers. Although FSH and LH are synthesized in the same cell, the pituitary gonadotroph, they exhibit differences in the rates of assembly of their corresponding subunits in the endoplasmic reticulum. This implies a potential regulative step in the production of the hormones. Moreover, it is well established that the diversity of hormone-specific oligosaccharides is the result of assembly of the common {alpha}-subunit with hormone-specific ß-subunit. This combination results in unique oligosaccharide patterns even if the hormones are synthesized in the identical cell and have access to the same processing enzymes. The differences in carbohydrates have major physiological consequences: The very short in vivo biological half-life of LH compared with FSH is due to its carbohydrates. Thus, in this case, the quaternary relationships in the heterodimer are crucial because the longevity of the gonadotropins is linked to oligosaccharide structure, which, in turn, is dependent on the nature of the heterodimeric interactions. The ability to by-pass the assembly step with the single-chain analogs permits incorporation of mutated subunits that otherwise would inhibit heterodimer formation from the individual subunits. Thus, these single-chain models have revealed a key structure-function feature of the glycoprotein hormone family, i.e. the quaternary interactions are essential for the intracellular trafficking for the heterodimers and for hormone-specific posttranslational modifications, but not for receptor recognition and signaling. In the latter case, the role of the heterodimer structure is to ensure that the appropriate epitopes in each subunit are brought in contact with the receptor triggering the biological response. Presumably then, the receptor-ligand contact sites generated at the receptor interface are established by small clusters of amino acids from both ß- and {alpha}-subunits (20, 21, 22, 23, 24, 25). Such determinants for bioactivity are not disrupted by the conformational changes induced by the disulfide bond mutations. On the other hand, given the flexibility between the gonadotropin receptor and the ligand, the plasticity of this complex may configure the ß- and mutated {alpha}-domains into a heterodimeric-like species in the vicinity of the receptor leading to downstream signaling.

The biological activity of the analogs observed in vivo is critical because previous studies tested the activity of disulfide bond mutants using in vitro models. Because these mutants are misfolded, we could not expect that a priori they would be active in vivo (26). It is remarkable that, although the dually active gonadotropin analogs contain both FSH and LH activities, they elicit sequential individual responses to FSH and LH. If this were not the case, we would not have observed comparable bioactivities (i.e. ovarian weight gain, aromatase induction, and superovulation) of the analogs to those elicited by PMSG and hCG, because simultaneous actions of FSH and LH often result in premature luteinization and a failure to induce multiple ovulations (27, 28, 29). Thus, although the analogs reported here can exert dual gonadotropin activities, similar triple-domain proteins, consisting of both FSH and LH subunits, have not been observed in vivo. Presumably, this ensures endocrine homeostasis such that secretion of FSH and LH is independently regulated. Moreover it allows sequential action of the gonadotropic hormones only during distinct phases of gonad development resulting in optimal gametogenesis and steroidogenesis.

Analysis of how the three gonadotropin subunits are configured in a single-chain orientation will be feasible in the future by purification and crystallographic studies of the analogs. Such studies will also be helpful to identify the minimal heterodimeric contact points required for glycoprotein hormone receptor recognition and signal transduction. The permissiveness in the structures of these single-chain analogs can provide a convenient platform to study other multisubunit bioactive proteins and growth factors of the cystine knot superfamily where subunit assembly will be vulnerable to mutation(s). Finally, the nonmutated and mutated single-chain variants, similar to those reported here, will also permit mapping of functional epitopes on numerous intracellular intermediates generated during the biosynthesis and secretion of other multisubunit proteins.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
REFERENCES
 
Materials
PMSG and hCG were obtained from Calbiochem (San Diego, CA) and Sigma Chemical Co. (St. Louis, MO). Restriction enzymes were purchased from Promega Corp. (Madison, WI) and New England Biolabs, Inc. (Beverly, MA). Hyaluronidase from bovine testes (type IV-S) was obtained from Sigma Chemical Co. Oligonucleotides used for sequencing were prepared by Washington University Sequencing Facility (St. Louis, MO) and from SeqWright (Houston, TX). All other analytical grade chemicals were deoxyribonuclease, ribonuclease, and protease free and purchased from EM Science (Gibbstown, NJ). Media and reagents for cell culture were prepared by Washington University Center for Basic Research (St. Louis, MO), except Ham’s F12, which was purchased from Mediatech, Inc. (Herndon, VA), and M-2 medium was obtained from Sigma. mAbs, designated FSH 54, FSHß 4B, and recombinant FSH (rFSH) were obtained from Organon (Oss, The Netherlands). mAb B109 was provided by Dr. Steven Birken (Columbia University Medical School, New York, NY). The CGß- and {alpha}-subunit-specific antisera were prepared in this laboratory. [35S]Cysteine was purchased from MP BioMedicals, Inc. (Irvine, CA). Pansorbin was purchased from EMD BioSciences (Darmstadt, Germany). Fetal bovine serum (FBS) was purchased from Harlan Bioproducts for Science, Inc. (Indianapolis, IN). Immature female CF-1 mice were obtained from Charles River (Wilmington, MA). Dialyzed FBS, DH5{alpha}-competent cells, G418, lipofectamine 2000, trizol, and RT-PCR reagents were all purchased from Invitrogen (Carlsbad, CA). Agarose and ethidium bromide were obtained from Shelton Scientific (Shelton, CT). Microtainer serum separator tubes were obtained from Becton Dickinson and Co. (Franklin Lakes, NJ). Vivacell 70 preparative concentrators were purchased from Sartorius (Goettingen, Germany). RIA kits for hCG and hFSH were purchased from Diagnostic Products, Inc. (Los Angeles, CA). These determinations employ enzyme-linked immunosorbent sandwich assay methods. In studies of the triple-domain single chain we validated this particular assay against a polyclonal based assay and also by measurements using BIACORE detection (10).

Engineering Single-Chain Variants
Triple-domain, single-chain mutants were constructed by domain swapping from three templates previously constructed: the CGß{alpha} single chain (30, 31, 32) containing mutations in the cystine knot of the {alpha}-subunit ({alpha}cys10–60 or 32–84) (8, 13) and the triple-domain, single-chain FSHß-CTP-CGß-{alpha} (10). The cystine residues were changed to alanine (13). In the first step, the triple-domain single chain was digested with BamHI/SalI. The fragment containing BamHI-FSHß-CTP-CGß-exon2-SalI sequence was ligated into BamHI/SalI site of pM2HA (10, 15). This construct was then digested with SalI. In a separate reaction, pM2HA bearing the CGß{alpha} mutants ({alpha}cys10–60 or {alpha}32–84) were digested with SalI. The fragments containing CGß exon 3 and the entire coding sequence of the {alpha}-subunit with the cys10–60 or cys32–84 mutations were inserted into the pM2HASalI site containing the FSHß-CTP-CGß exon 2 sequence. The final products, FSHß-CTP-CGß-{alpha} bearing the mutations, were sequenced to ensure no mistakes were introduced in the cloning steps.

Transfection and Cell Culture
The triple-domain analogs were transfected into CHO cells using Lipofectamine 2000, and stable clones were selected using G418 (250 µg/ml) approximately 11 d later. The clones were maintained in Ham’s F12 medium [supplemented with penicillin (100 U/ml), streptomycin (100 µg/ml), and 2 mM L-glutamine] containing 5% FBS and G418 (125 µg/ml) in a humidified atmosphere of 5% CO2/95% air. Collection media for in vivo studies were concentrated using a Vivacell 70 concentrator and frozen in aliquots at –80 C until further use. Gonadotropin immunoreactivities in concentrated media of triple-domain analogs were measured at the Fertility Center, University of Kansas Medical Center (Kansas City, KS), using the ImmuLite detection system with LH- and FSH-specific sandwich immunoassay kits.

Metabolic Labeling
To examine the synthesis of single-chain chimeras, CHO cells expressing the wild-type, triple-domain analog or the {alpha} mutated chimeras were labeled for 16 h in cysteine-free Ham’s F-12 medium containing dialyzed FBS and [35S]cysteine (10, 15). Aliquots of cell lysate and medium were immunoprecipitated with polyclonal antiserum directed against CGß-subunit, which was raised in our laboratory. The proteins were resolved on 12.5% SDS-PAGE. Labeled proteins from autoradiography were quantified by densitometry.

Western Blot Analysis
Equal amounts of analogs were resolved on 12.5% SDS-PAGE in the absence of heat and reducing agent and transferred onto nitrocellulose. Proteins were probed with mAbs: FSH dimer-specific [FSH 54 (11, 12)], FSHß and dimer specific (FSHß 4B), CG dimer-specific (B109), and with CGß polyclonal antiserum (CGß AS). Immunodetection was performed with Tropix chemiluminescent system (Tropix, Inc., Bedford, MA).

Mice
Mice were supplied with food and water ad libitum and maintained on a 12-h light and 12-h dark cycle. All the animal protocols were approved by University of Kansas Medical Center Animal Care Committee as per National Institute of Health guidelines.

FSH Bioassays
Ovarian Weight Gain Response.
The FSH activity of the chimeras was determined using the Steelman-Pohley assay, which is based on the ability of FSH to increase ovarian weight in immature mice (33). Mice (n = 4 or 5) were injected ip with either 100 µl of PBS or control medium containing only the {alpha}-subunit, PMSG (2.5 IU or 5 IU per mouse) or the triple-domain analogs (2.5 or 5 IU equivalent of FSH immunoreactivity). After 48 h, mice were exsanguinated under isofluorane anesthesia, and ovaries were isolated and individually weighed. This FSH bioassay was performed three times.

Ovarian Aromatase Induction Assay.
Ovarian aromatase assay using immature CF-1 female mice was performed as previously described (15, 16). Mice were injected with control medium, or different doses of PMSG (5 IU or 10 IU) or the triple-domain analogs (three mice per dose) and 48 h later, the ovaries were collected into Trizol solution. Total RNA was isolated according to the manufacturer’s protocol, and the integrity was checked on 1% agarose gels and spectrophotometrically quantified. One microgram of total RNA was used in reverse transcription reactions, and the cDNAs were used in quantitative amplification reactions using aromatase (Cyp19a1)- and cyclophilin A (CypA)-specific primers as described elsewhere (15). The PCR products were separated on 0.7% agarose gels and stained with ethidium bromide, and the intensities of bands were quantified using an Epson expression 1680 scanner and software supplied by the manufacturer. The aromatase assay was performed twice.

hCG Bioassay.
The hCG bioassay was performed based on the standard superovulation protocol (15). Immature CF-1 mice (four to five) were injected ip with PMSG (5 IU per mouse), and 46 h later they were injected with either PBS, control medium, different doses of hCG, or the analogs (2.5 IU or 5 IU per mouse). The cumulus masses from oviducts were collected 16/18 h later and oocytes were counted as described (15). Superovulation assays were performed three times.

Statistical Analysis
All data were analyzed by one-way ANOVA using a Microsoft Excel software program (Microsoft Inc., Redmond, WA). P < 0.05 was considered statistically significant.


   ACKNOWLEDGMENTS
 
We thank Dr. Vicenta Garcia-Campayo for a critical reading of the manuscript and helpful suggestions. We also thank Amanda Fisher, Diane Redmond, and Linda Lobos for their excellent assistance in preparing the manuscript.


   FOOTNOTES
 
This work was supported in part by National Institutes of Health Grant DK065155 (to I.B.).

First Published Online April 6, 2006

1 A.J.-S. and T.R.K. contributed equally to this study. Back

Abbreviations: CG, Chorionic gonadotropin; CHO, Chinese hamster ovary; mAbs, monoclonal antibodies; PMSG, pregnant mare serum gonadotropin; rFSH, recombinant FSH; SDS, sodium dodecyl sulfate.

Received for publication December 30, 2005. Accepted for publication March 29, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 RESULTS
 DISCUSSION
 MATERIALS AND METHODS
 REFERENCES
 

  1. Pierce JG, Parsons TF 1981 Glycoprotein hormones: structure and function. Annu Rev Biochem 50:465–495[CrossRef][Medline]
  2. Lapthorn AJ, Harris DC, Littlejohn A, Lustbader JW, Canfield RE, Machin KJ, Morgan FS, Isaacs NW 1994 Crystal structure of human chorionic gonadotropin. Nature 369:455–461[CrossRef][Medline]
  3. Wu H, Lustbader JW, Liu Y, Canfield RE, Hendrickson WA 1994 Structure of human chorionic gonadotropin at 2.6 A resolution from MAD analysis of the selenomethionyl protein. Structure 2:545–558[Medline]
  4. Fox KM, Dias JA, Van Roey P 2001 Three-dimensional structure of human follicle-stimulating hormone. Mol Endocrinol 15:378–389[Abstract/Free Full Text]
  5. Jackson AM, Berger P, Pixley M, Klein C, Hsueh AJ, Boime I 1999 The biological action of choriogonadotropin is not dependent on the complete native quaternary interactions between the subunits. Mol Endocrinol 13:2175–2188[Abstract/Free Full Text]
  6. Hiro’oka T, Maassen D, Berger P, Boime I 2000 Disulfide bond mutations in follicle-stimulating hormone result in uncoupling of biological activity from intracellular behavior. Endocrinology 141:4751–4756[Abstract/Free Full Text]
  7. Sugahara T, Grootenhuis PD, Sato A, Kudo M, Ben-Menahem D, Pixley MR, Hsueh AJ, Boime I 1996 Expression of biologically active fusion genes encoding the common {alpha} subunit and either the CGß or FSHß subunits: role of a linker sequence. Mol Cell Endocrinol 125:71–77[CrossRef][Medline]
  8. Sato A, Perlas E, Ben-Menahem D, Kudo M, Pixley MR, Furuhashi M, Hsueh AJ, Boime I 1997 Cystine knot of the gonadotropin {alpha} subunit is critical for intracellular behavior but not for in vitro biological activity. J Biol Chem 272:18098–18103[Abstract/Free Full Text]
  9. Ben-Menahem D, Kudo M, Pixley MR, Sato A, Suganuma N, Perlas E, Hsueh AJ, Boime I 1997 The biologic action of single-chain choriogonadotropin is not dependent on the individual disulfide bonds of the ß subunit. J Biol Chem 272:6827–6830[Abstract/Free Full Text]
  10. Kanda M, Jablonka-Shariff A, Sato A, Pixley MR, Bos E, Hiro’oka T, Ben-Menahem D, Boime I 1999 Genetic fusion of an {alpha}-subunit gene to the follicle-stimulating hormone and chorionic gonadotropin-ß subunit genes: production of a bifunctional protein. Mol Endocrinol 13:1873–1881[Abstract/Free Full Text]
  11. Garcia-Campayo V, Jablonka-Shariff A, Boime I 2004 A single-chain bifunctional gonadotropin analog is secreted from Chinese hamster ovary cells as two distinct bioactive species. J Biol Chem 279:44286–44293[Abstract/Free Full Text]
  12. Keene JL, Matzuk MM, Boime I 1989 Expression of biological active human follitropin in Chinese hamster ovary cells. J Biol Chem 264:4769–4775[Abstract/Free Full Text]
  13. Furuhashi M, Ando H, Bielinska M, Pixley MR, Shikone T, Hsueh AJ, Boime I 1994 Mutagenesis of cysteine residues in the human gonadotropin {alpha} subunit. Roles of individual disulfide bonds in secretion, assembly, and biologic activity. J Biol Chem 269:25543–25548[Abstract/Free Full Text]
  14. Darling RJ, Ruddon RW, Perini F, Bedows E 2000 Cystine knot mutations affect the folding of the glycoprotein hormone {alpha}-subunit. Differential secretion and assembly of partially folded intermediates. J Biol Chem 275:15413–15421[Abstract/Free Full Text]
  15. Garcia-Campayo V, Boime I, Ma X, Daphna-Iken D, Kumar TR 2005 A single-chain tetradomain glycoprotein hormone analog elicits multiple hormone activities in vivo. Biol Reprod 72:301–308[Abstract/Free Full Text]
  16. Burns KH, Yan C, Kumar TR, Matzuk MM 2001 Analysis of ovarian gene expression in follicle-stimulating hormone ß knockout mice. Endocrinology 142:2742–2751[Abstract/Free Full Text]
  17. Isaacs NW 1995 Cystine knots. Curr Opin Struct Biol 5:391–395[CrossRef][Medline]
  18. Muller YA, Christinger HW, Keyt BA, do Vos AM 1997 The crystal structure of vascular endothelial growth factor (VEGF) refined to 1.93 A resolution: multiple copy flexibility and receptor binding. Structure 5:1325–1338[Medline]
  19. Suganuma N, Matzuk M, Boime I 1989 Elimination of disulfide bonds affects assembly and secretion of the human chorionic gonadotropin ß subunit. J Biol Chem 264:19302–19307[Abstract/Free Full Text]
  20. Wells JA 1996 Binding in the growth hormone receptor complex. Proc Natl Acad Sci USA 93:1–6[Abstract/Free Full Text]
  21. Davies DR, Cohen GH 1996 Interactions of protein antigens with antibodies. Proc Natl Acad Sci USA 93:7–12[Abstract/Free Full Text]
  22. Jones S, Thornton JM 1996 Principles of protein-protein interactions. Proc Natl Acad Sci USA 93:13–20[Abstract/Free Full Text]
  23. Lee, C, Ji IJ, Ji TH 2002 Use of defined/function mutants to access receptor-receptor interactions. Methods 27:318–323[Medline]
  24. Puett D, Bhowmick N, Fernandez LM, Huang J, Wu C, Narayan P 1996 hCG-receptor binding and transmembrane signaling. Mol Cell Endocrinol 125:55–64[CrossRef][Medline]
  25. Grossmann M, Szkudlinski MW, Zeng H, Ji I, Tropea JE, Ji TH, Weintraub BD 1995 Role of the carboxy-terminal residues of the {alpha}-subunit in the expression and bioactivity of human thyroid-stimulating hormone. Mol Endocrinol 9:945–958
  26. Bernier V, Lagacé M, Bichet DG, Bouvier M 2004 Pharmacological chaperones: potential treatment for conformational diseases. Trends Endocrinol Metab 15:222–228[CrossRef][Medline]
  27. Cooperman AB, Horowitz GM, Kaplan P, Scott RT, Navot D, Hofmann GE 1995 Relationship between circulating human chorionic gonadotropin levels and premature luteinization in cycles of controlled ovarian hyperstimulation. Fertil Steril 63:1267–1271[Medline]
  28. Filicori M 1997 Cause for premature luteinization? Fertil Steril 67:1179–1181[Medline]
  29. Filicori M, Cognigni GE, Pocognoli P, Tabarelli C, Ferlini F, Perri T, Parmegiani L 2003 Comparison of controlled ovarian stimulation with human menopausal gonadotropin or recombinant follicle-stimulating hormone. Fertil Steril 80:390–397[CrossRef][Medline]
  30. Sugahara T, Pixley MR, Minami S, Perlas E, Ben-Menahem D, Hsueh AJ, Boime I 1995 Biosynthesis of a biologically active single peptide chain containing the human common {alpha} and chorionic gonadotropin ß subunits in tandem. Proc Natl Acad Sci USA 92:2041–2045[Abstract/Free Full Text]
  31. Narayan P, Wu C, Puett D 1995 Functional expression of yoked human chorionic gonadotropin in baculovirus-infected insect cells. Mol Endocrinol 9:1720–1726[Abstract/Free Full Text]
  32. Sugahara T, Sato A, Kudo M, Ben-Menahem D, Pixley MR, Hsueh AJ, Boime I 1996 Expression of biologically active fusion genes encoding the common {alpha} subunit and the follicle-stimulating hormone ß subunit. Role of a linker sequence. J Biol Chem 271:10445–10448[Abstract/Free Full Text]
  33. Steelman SL, Pohley FM 1953 Assay of the follicle stimulating hormone based on the augmentation with human chorionic gonadotropin. Endocrinology 53:604–616[Medline]



This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
A. Jablonka-Shariff, C. A. Pearl, A. Comstock, and I. Boime
A Carboxyl-terminal Sequence in the Lutropin {beta} Subunit Contributes to the Sorting of Lutropin to the Regulated Pathway
J. Biol. Chem., April 25, 2008; 283(17): 11485 - 11492.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
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 Download to citation manager
Right arrow Reprints, Permissions and Rights
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Jablonka-Shariff, A.
Right arrow Articles by Boime, I.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Jablonka-Shariff, A.
Right arrow Articles by Boime, I.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology Endocrine Reviews J. Clin. End. & Metab.
Molecular Endocrinology Recent Prog. Horm. Res. All Endocrine Journals