1 Department of Obstetrics and Gynecology and 2 Center for Reproductive Sciences and 3 Department of Biostatistics, School of Public Health, Columbia University, New York, NY 10032, USA 4 To whom correspondence should be addressed. e-mail: jwl2{at}columbia.edu
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: carboxyterminal peptide/FSH/hormone analogue/oligosaccharide/pharmacokinetics
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Two long-acting gonadotrophin agonists developed by fusing the carboxyterminal peptide of hCG (CTP) to native recombinant human FSH (rhFSH) either as a non-covalently bound heterodimer containing the common - and ß-FSH subunits (Fares et al., 1992
) or as a contiguous peptide (Klein et al., 2002
) have previously been reported. The increased half-life and biopotency of this analogue (rhFSH-CTP) is speculated to be due to the four O-linked glycosylation sites present on the CTP moiety (Matzuk et al., 1990
). Alternatively, the biology of an FSH analogue containing additional N-linked oligosaccharides has not been reported.
Addition of carbohydrate to the peptide backbone is a post-translational event that may occur through O-linkage with serine or threonine residues, or via N-linkage to asparagine. Whereas the glycosylation signal sequence for O-linked sugars is poorly defined, addition of N-linked carbohydrates to asparagine occurs whenever the primary amino acid sequence contains the trimer Asn-X-Ser/Thr, where X denotes any amino acid except proline (Imperiali and OConnor, 1999). The number of N-linked carbohydrates that are introduced in a synthetic polypeptide chain can thus be precisely defined. Consequently, this represents a potentially more powerful tool to refine the pharmacokinetic properties of rhFSH and perhaps other recombinant species in a clinically meaningful way, compared with adding CTP or other sequences containing clusters of O-linked sugars. To determine the feasibility of this approach, we designed a synthetic oligopeptide containing two copies of the N-linked glycosylation signal sequence, and tethered this sequence between the ß- and
-subunits of hFSH, creating a single-chain fusion hormone analogue (rhFSH-N2). The ability of this novel construct to bind and activate the hFSH receptor in vitro, and the in-vivo activity and half-life of this analogue compared to native recombinant hFSH and hFSH-CTP, were examined.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Controls
Native rhFSH (follitropin-ß) was generously provided by Serono, Inc. (Rockland, MD, USA). Single-chain rhFSH-CTP was produced in a CHO-K1 cell line as described previously (Klein et al., 2002).
Construction of a rhFSHß-N2- (rhFSH-N2) single-chain fusion clone
Two complementary DNA strands encoding the following polypeptide sequence were synthesized by Columbias protein core: Ser-Gly-Ser-Asn-Ala-Thr-Gly-Ser-Gly-Ser-Asn-Ala-Thr-Ser-Gly-Ser, such that following annealing of the strands 5' BamHI and 3' SpeI sticky ends were formed (DNA sequences available upon request). The synthetic DNA duplex was then ligated into a vector along with the rhFSHß- and -subunit cDNA. The in-frame ligation of these three DNA was accomplished by placing a XhoI site immediately preceding the start codon and replacing the terminator codon of the hFSH ß-subunit with a BamHI site. In addition a SpeI site was placed at the 5' end and a SacI site immediately following the terminator codon of the
-subunit. The three fragments were then inserted into an SV40-based expression vector at XhoI/SacI sites to form the rhFSH-N2 expression construct (Figure 1).
|
Affinity purification of rhFSH-N2 was accomplished using an -specific monoclonal antibody column [A103 (OConnor et al., 1994
)]. The column was prepared by coupling purified A103 immunoglobulins to cyanogen bromide (CNBr)-Sepharose-4B according to the manufacturers instructions (Amersham Pharmacia Biotech, Piscataway, NJ, USA) at a concentration of 5 mg antibody/ml Sepharose. After applying the cell supernatant, the column was washed with 50 bed volumes of PBS followed by 2 bed volumes of distilled water. rhFSH-N2 was eluted with 3 or 4 bed volumes of 1 mol/l acetic acid and immediately dried on a Speed-Vac concentrator (Savant Instruments, Holbrook, NY, USA).
Electrophoresis and Western blotting
Sodium dodecyl sulphatepolycrylamide gel electrophoresis was performed and proteins were transferred to nitrocellulose using standard techniques (Laemmli, 1970; Towbin et al., 1979
; Burnette, 1981
). After blocking, the nitrocellulose was incubated overnight in a 1:10 000 dilution of a monoclonal antibody to FSH-ß from Biodesign International (Saco, ME, USA). The membrane was then washed and incubated for 1 h in a 1:10 000 dilution of a peroxidase-conjugated polyclonal antibody to mouse immunoglobulins (Amersham Pharmacia Biotech). After washing, the nitrocellulose was incubated in a chemiluminescent detection reagent according to the manufacturers directions (Amersham Pharmacia). Bands were visualized by exposure to X-ray film.
Isoelectric focusing gel electrophoresis
Samples were electrophoresed on a Novex pre-cast IEF gel, with a pI range of pH 37 according to the manufacturers instructions (Invitrogen, Carlsbad, CA, USA) and then visualized by silver staining.
In-vitro bioactivity
The rhFSH-N2 was quantified using an hFSH-ß specific antibody (Biodesign International, Saco, ME, USA) radioimmunoassay. Biological activity was assessed using CHO cells transfected with the FSH receptor (CHO FSHr) as previously described (Kelton et al., 1992). A total of 2x104 cells was mixed with rhFSH-N2, rhFSH-CTP or rhFSH at varying concentrations in a total volume of 200 µl. Mixtures were incubated at 37°C for 15 min, and cAMP levels were measured using a radioimmunoassay kit (Perkin Elmer Life Sciences, Boston, MA, USA).
In-vivo bioactivity and half-life determinations
Approval was obtained from the Institutional Animal Care and Use Committee at Columbia University. Immature (21 day old) female SpragueDawley rats were obtained from Charles River Laboratories (Wilmington, MA, USA). Rats were housed three to a cage and given standard food and tap water ad libitum. Animals were randomly assigned to one of four treatment groups: rhFSH-N2, single-chain recombinant hFSH tethered to CTP (rhFSH-CTP), rhFSH, or saline. All hormones were diluted to 11 µg/ml using injection buffer containing bovine serum albumin (1 mg/ml). Hormone was administered as a single i.v. dose of 2800 ng/rat in 0.25 ml of solution via a dorsal tail vein. Serum was obtained by periorbital venipuncture at the following intervals post-injection: 0.5, 1.0, 3.0, 6.0 and 12 h. Anaesthesia with isoflurane was administered prior to injection and each bleed. Serum was assayed for hFSH by enzyme-linked immunosorbent assay (Diagnostic Systems Laboratory, Webster, TX, USA). Rats were killed 72 h post-injection by carbon dioxide inhalation followed by exsanguination via cardiac puncture. Ovaries were then extirpated and weighed.
Pharmacokinetic analysis was performed using Winnonlin 1.0 software (Pharsight Corporation, Mountain View, CA, USA). Mean values for each time-point within a group were used to estimate pharmacokinetic parameters for terminal (elimination) half-life (t1/2) and area under the serum concentrationtime curve (AUC). Clearance (Cl) was calculated using the relationship Cl = dose/AUC.
Statistical analysis
Mean ovarian plasma concentrations and ovarian weights were compared between groups in a one-way analysis of variance (ANOVA) model. Since groups with higher means tended to also have higher variances, the data were first log-transformed to stabilize the variance between groups and satisfy the model assumptions. The primary comparison of interest was between rhFSH-N2 and rhFSH, and this difference was tested using a single contrast within the one-way structure. In addition, mean serum plasma concentrations at all time-points were compared using a one-way ANOVA model. At each time-point, the comparisons of interest were rhFSH-N2 versus rhFSH and rhFSH-N2 versus rhFSH-CTP. These hypotheses were tested with simultaneous contrasts at each time-point using a Bonferroni adjustment to control for multiple comparisons, and the analysis for each time-point was considered separately.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
|
|
|
|
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Proteins are glycosylated by one of two distinct mechanisms: N-linkage via an amide group on asparagine, or O-linkage, in which the oligosaccharide is attached to the hydroxyl group on a serine or threonine residue. N-Oligosaccharides share a common trimannosyl-chitobiose core domain distinct from that of O-linked sugars, although overall size and composition is variable (Dell and Morris, 2001). In addition to structural disparities, differences in the function of native O-linked and N-linked oligosaccharides on glycoprotein biology exist. For example, removal or deglycosylation of the CTP of hCG, containing naturally occurring O-oligosaccharides, increases metabolism of the hormone in vivo (el-Deiry et al., 1989
) but has no effect on proper folding of the molecule and receptor activation. (el-Deiry et al., 1989
; Matzuk et al., 1990
). In contrast, the native N-linked oligosaccharides of hCG and the other glycoprotein hormones play critical roles in both signal transduction and hormone metabolism (Matzuk and Boime, 1988
; Van Zuylen et al., 1997
; Imperiali and OConnor, 1999
).
Addition of the CTP, containing four O-linked oligosaccharides, to rhFSH has been shown to increase the half-life of the molecule without altering in-vitro bioactivity (and therefore proper folding of the protein) (Klein et al., 2002). Lack of an identifiable O-oligosaccharide consensus signal sequence, however, limits precise adjustments in carbohydrate content when designing hormone analogues. In an attempt to provide a more flexible, well-defined means of delaying gonadotrophin metabolism in vivo, we developed a recombinant follitropin agonist containing additional N-linked carbohydrates on the FSHß-subunit. Unlike O-linked sugars, the number of N-linked moieties is easily modifiable in a synthetic oligopeptide sequence by inserting a specified number of consensus N-oligosaccharide signal sequences. Adjusting the number of added carbohydrates could theoretically result in FSH agonists with a spectrum of pharmacokinetic characteristics and bioactivity in vivo. These designer gonadotrophin formulations may be tailored to meet the needs of specific patient populations.
In this report, we examined the pharmacokinetics and pharmacodynamics of a recombinant FSH analogue containing two additional N-oligosaccharides to test the utility of this approach. Stable, high-secreting CHO cell clones transfected with our FSH analogue were isolated, and proper folding of the fusion hormone was confirmed using an in-vitro FSH bioassay. Serum levels of rhFSH-N2 following i.v. administration were comparable with that of rhFSH-CTP and significantly increased relative to native recombinant FSH, confirming the ability of added N-oligosaccharides to decrease clearance of FSH in a manner analogous to a CTP moiety. Pharmacokinetic analysis of the data indicates rhFSH-N2 has a half-life 2-fold that of native rhFSH and increased in-vivo bioactivity relative to native rhFSH following an i.v. bolus dose of hormone.
The altered in-vivo characteristics of rhFSH-N2 are most likely related to the increased carbohydrate load present on the peptide linker sequence used to covalently bind the two subunits. N-Linked oligosaccharides on native hFSH are known to confer protein stability, and FSH analogues containing added O-linked carbohydrates attached exclusively to the ß-subunit (without covalent binding to the common -subunit) have prolonged half-lives compared with native hFSH (Bouloux et al., 2001
). Furthermore, while both single-chain rhFSH-CTP and rhFSH-CTP lacking a covalent linkage between the subunits have never been compared, the prolongation in serum half-life relative to control rhFSH appears comparable (23-fold increase) (Bouloux et al., 2001
; Klein et al., 2002
). Nevertheless, the covalent bond between the two subunits of rhFSH-N2 may play a contributory role in conferring increased biopotency to our hormone analogue, independent of any alteration in carbohydrate load. By stabilizing subunit interaction through covalent linkage, deactivation of the intact heterodimer due to subunit dissociation may theoretically be inhibited. This theory has yet to be examined. Most single-chain glycoprotein hormone analogues have typically employed a carbohydrate-rich sequence as a tether between subunits, confounding the effect of covalent binding per se on bioactivity. A single-chain rhFSH construct devoid of any linker sequence was able to bind and activate the hFSH receptor in vitro (Sugahara et al., 1996
). A pharmacokinetic comparison between this analogue and native hFSH would help to clarify the role of single-chain assembly on hormone metabolism. The role of the peptide backbone of the linker sequence must also be resolved. We plan to perform a pharmacokinetic and pharmacodynamic comparison employing a single-chain FSH analogue containing a comparable tether but devoid of N-linked consensus sequences to clarify this issue.
rhFSH-N2, with two additional N-oligosaccharides, displayed similar in-vivo characteristics (pharmacokinetics and pharmacodynamics) compared with rhFSH-CTP, which contains four O-linked carbohydrates in the CTP tether. This is not unexpected, as N-oligosaccharides tend to be more complex and highly branched than O-linked moieties. However, mass spectrometric analysis is required to characterize the carbohydrates of these modified FSH species, as size (particularly of O-linked oligosaccharides) can be highly variable. The effect of carbohydrate mass, volume and/or spatial constitution on half-life and bioactivity of FSH analogues remains speculative, pending the outcome of comparative analyses of species with well-characterized carbohydrate constituents.
Others have shown that carbohydrate variability within native preparations of FSH can have an effect on half-life. More acidic, sialic acid-rich isoforms have longer half-lives (Blum and Gupta, 1985). In-vivo bioactivity of standard recombinant preparations is correlated with a more acidic pI (Mulders et al.,1997
). Increased sialic acid content of rhFSH-N2, as suggested by a lower pI on IEF gel electrophoresis may be a factor in its increased half-life and in-vivo bioactivity. A complex carbohydrate characterization and further studies using desialylated hormone may provide a more definitive answer.
In conclusion, we report a novel long-acting recombinant follitropin agonist containing an N-oligosaccharide-rich tether between the ß- and -subunits of FSH. Further studies are required to ascertain the clinical potential of this analogue in patients in whom exogenous gonadotrophin therapy is indicated. The immunogenicity of our analogue, and its effects on clinical parameters such as folliculogenesis, oocyte quality, embryo development, implantation and pregnancy, and spermatogenesis remain to be defined. The utility of this approach in the development of additional FSH and other recombinant protein analogues is promising, and merits further investigation.
![]() |
Acknowledgements |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Bouloux, P.M.G., Handelsman, D.J., Jockenhovel, F., Nieschlag, E., Rabinovici, J., Frasa, W.L.H., de Bie, J.J., Voortman, G., Itskovitz-Eldor, J., FSH-CTP study group (2001) First human exposure to FSH-CTP in hypogonadotropic hypogonadal males. Hum. Reprod., 16, 15921597.
Burnette, W.N. (1981) Western Blotting: electrophoretic transfer of proteins from sodium dodecyl sulfatepolyacrylamide gels to unmolded nitrocellulose and radiologic detection with antibody and radioiodinated protein A. Anal. Biochem., 112, 195203.[ISI][Medline]
Dell, A. and Morris, H.R. (2001) Glycoprotein structure determination by mass spectrometry. Science, 291, 23512356.
el-Deiry, S., Kaetzel, D., Kennedy, G., Nilson, J. and Puett, D. (1989) Site-directed mutagenesis of the human chorionic gonadotropin beta-subunit: bioactivity of a heterologous hormone, bovine alpha-human des-(122145)beta. Mol. Endocrinol., 3, 15231528.[Abstract]
Fares, F.A., Suganuma, N., Nishimori, K., Lapolt, P.S., Hsueh, A.J.W. and Boime, I. (1992) Design of a long-acting follitropin agonist by fusing the C-terminal sequence of the chorionic gonadotropin beta subunit to the follitropin beta subunit. Proc. Natl Acad. Sci. USA, 89, 43044308.[Abstract]
Imperiali, B. and OConnor, S.E. (1999) Effect of N-linked glycosylation on glycopeptide and glycoprotein structure. Curr. Opin. Chem. Biol., 3, 643649.[CrossRef][ISI][Medline]
Kelton, C.A., Cheng, S.V., Nugent, N.P., Schweickhardt, R.L., Rosenthal, J.L., Overton, S.A., Wands, G.D., Kuzeja, J.B., Luchette, C.A. and Chappel, S.C. (1992) The cloning of the human follicle stimulating hormone receptor and its expression in COS-7, CHO, and Y-1 cells. Mol. Cell. Endocrinol., 89, 141151.[CrossRef][ISI][Medline]
Klein, J., Lobel, L., Pollak, S., Ferin, M., Xiao, E., Sauer, M.V. and Lustbader, J.W. (2002) Pharmacokinetics and pharmacodynamics of single-chain recombinant human follicle-stimulating hormone containing the human chorionic gonadotropin carboxyterminal peptide in the rhesus monkey. Fertil. Steril., 77, 12481255.[CrossRef][ISI][Medline]
Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227, 680685.[ISI][Medline]
Matzuk, M.M. and Boime, I. (1988) The role of the asparagine-linked oligosaccharides of the alpha subunit in the secretion and assembly of human chorionic gonadotrophin. J. Cell Biol., 106, 10491059.[Abstract]
Matzuk, M.M., Hsueh, A.J., Lapolt, P., Tsafriri, A., Keene, J.L. and Boime, I. (1990) The biological role of the carboxyl-terminal extension of human chorionic gonadotropin beta-subunit. Endocrinology, 126, 376383.[Abstract]
Mulders, J.W., Derksen, M., Swolfs, A. and Maris, F. (1997) Prediction of the in vivo biological activity of human recombinant follicle stimulating hormone using quantitative isoelectric focusing. Biologicals, 25, 269281.[CrossRef][ISI][Medline]
OConnor, J.F., Birken, S., Lustbader, J.W., Krichevsky, A., Chen, Y. and Canfield, R.E. (1994) Recent advances in the chemistry and immunochemistry of human chorionic gonadotropin: impact on clinical measurements. Endocr. Rev., 15, 650683.[ISI][Medline]
Sugahara, T., Grootenhuis, P.D., Sato, A., Kudo, M., Ben-Menahem, D., Pixley, M.R., Hsueh, A.J. and Boime, I. (1996) Expression of biologically active fusion genes encoding the common alpha subunit and either the CG beta or FSH beta subunits: role of a linker sequence. Mol. Cell. Endocrinol., 125, 7177.[CrossRef][ISI][Medline]
Towbin, H., Staehelin, T. and Gordon, J. (1979) Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Natl Acad. Sci. USA, 76, 43504354.[Abstract]
Van Zuylen, C.W., Kamerling, J.P. and Vliegenthart, J.F. (1997) Glycosylation beyond the Asn78-linked GlcNAc residue has a significant enhancing effect on the stability of the alpha subunit of human chorionic gonadotropin. Biochem. Biophys. Res. Commun., 232, 117120.[CrossRef][ISI][Medline]
Submitted on July 15, 2002; accepted on October 4, 2002