Oral and pulmonary delivery of FSH–Fc fusion proteins via neonatal Fc receptor-mediated transcytosis

S.C. Low1,3, S.L. Nunes2, A.J. Bitonti and J.A. Dumont

1 Syntonix Pharmaceuticals, Inc., 9 Fourth Avenue,Waltham, MA 02451, USA, 2 Current address: Protein Structure Group/Discovery Technologies, Novartis Institutes for Biomedical Research Inc., 250 Massachusetts Avenue, Cambridge, MA 02139, USA

3 To whom correspondence should be addressed: Email: slow{at}syntnx.com


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
BACKGROUND: The {alpha} and {beta} subunits of FSH were fused to the Fc domain of IgG1 either in a single chain or a heterodimer format. These molecules were absorbed through the epithelium in lung and intestine by neonatal Fc receptor (FcRn)-mediated transcytosis. METHODS AND RESULTS: Single chain and heterodimer FSH–Fc were made recombinantly in Chinese hamster ovary cells. Treatment of rats with a single s.c. dose of single chain or heterodimer FSH–Fc resulted in greater stimulation of ovarian weight (20.8±3.9 and 26.9±6.1 mg respectively) compared to those receiving vehicle (12.1±1.0 mg) or an equimolar dose of recombinant human FSH (14.3±1.7 mg). Both FSH–Fc fusion proteins were absorbed after oral dosing of newborn rats with long terminal half-lives of ~60 h, and pulmonary delivery in four cynomolgus monkeys produced maximum serum concentrations between 69 and 131 ng/ml with long terminal half-lives between 55 and 210 h. Serum inhibin levels increased after pulmonary dosing with single chain FSH–Fc (1.3- and 1.4-fold) and heterodimer FSH–Fc (5.9- and 7.1-fold) and remained elevated for >12 days after treatment with heterodimer FSH–Fc. CONCLUSIONS: We have shown that FSH–Fc fusion proteins have increased stability in blood and improved bioactivity in vivo, and that heterodimer FSH–Fc is more active in rats and monkeys than single chain FSH–Fc. These data suggest that Fc fusion proteins offer the potential for oral and pulmonary delivery of FSH.

Key words: FcRn/FSH/lung/pulmonary delivery


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Recombinant human (r)FSH is a common therapy in the treatment of both male and female infertility. Induction of folliculogenesis in women requires daily s.c. or i.m. administration of FSH for ~8–10 days, while daily injections are required for up to several months to induce spermatogenesis in hypogonadotrophic males (Schaison et al., 1993Go; Schoot et al., 1994Go). Longer-acting FSH preparations would therefore have obvious advantages in the tolerability of infertility treatments. A novel and less invasive means of delivering FSH without the use of needles would also be desirable but remains problematic for a variety of reasons including the relatively large size of the protein, its heterodimeric nature (i.e. non-covalently linked {alpha} and {beta} subunits; Pierce and Parsons, 1981Go) and the natural barrier represented by epithelial cells. An alternate approach to the delivery of FSH and other therapeutic proteins would therefore be the utilization of a natural transport system that exists within epithelial cells.

One candidate transport pathway involves the neonatal Fc receptor (FcRn) that was first described in newborn rodents as being responsible for the transport of IgG from maternal milk into the neonatal bloodstream (Simister et al., 1997Go). Human FcRn was first described in the placenta where it transports IgG from mother to fetus (Story et al., 1994Go). Although rodent FcRn levels decrease in epithelial tissues after weaning, human FcRn was recently detected in epithelial cells of adult human lung and intestine (Israel et al., 1997Go), and has been shown to transport IgG across human epithelial cells in vitro (Dickinson et al., 1999Go). Various proteins have been fused to the Fc domain of IgG1 and shown to retain biological activity (Ashkenazi et al., 1993Go; Chamow and Ashkenazi, 1996Go; Economides et al., 2003Go; Bitonti et al., 2004Go). Conjugation to Fc also increases the serum half-life of the fusion partner (Kim et al., 1998Go) due to the ability to bind FcRn which is the major regulator of IgG half-life (Kim et al., 1994Go; Junghans and Anderson, 1996Go; Simister et al., 1997Go). We therefore propose that FcRn may be used to transport therapeutic proteins, including FSH, across epithelial barriers and concomitantly increase the circulating half-life of the molecule.

FSH is a non-covalently linked dimeric protein consisting of {alpha} and {beta} subunits (Pierce and Parsons, 1981Go). Subunit assembly is essential for bioactivity of FSH (Jia and Hseuh, 1986Go) as well as for the stability of the {beta} subunit (Keene et al., 1989Go). However, it has been shown that a single chain fusion of the {alpha} and {beta} subunits of FSH is fully active (Sugahara et al., 1996Go) and has an increased serum half-life when fused with the carboxy-terminal peptide of hCG (Bouloux et al., 2001Go; Duijkers et al., 2002Go; Klein et al., 2003Go).

In this study we have investigated the effect of linking the Fc domain of immunoglobulin G1 (IgG1) to FSH, either as a single chain or with the {alpha} and {beta} subunits linked to separate arms of the Fc fragment as a heterodimer FSH-Fc molecule. These fusion proteins were used to examine carrier-medicated transport via FcRn across intestinal epithelia in neonatal rats and lung epithelia in cynomolgus monkeys.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Construction of single chain and heterodimer FSH–Fc fusion molecules
For single chain FSH–Fc constructs, FSH{beta} was isolated with its native signal sequence from a human pituitary mRNA library (Clontech, USA) using the following primers: 5'-CTAGCCTGCAGGCCACCATGAAGACACTCCAGTTTTTC-3' and 5'-TCCATTCCCGGGTTCTTTCATTTCACCAAAGGAG-3'.

FSH{alpha} was isolated from the same human pituitary mRNA library but without the signal sequence using the following primers: 5'-TACTTTCCCGGGGCTCCTGATGTGCAGGATT-3' and 5'-GCATCCGAATTCAGATTTGTGATAATAACAAGTAC-3'.

The area of each primer predicted to anneal to the corresponding template is underlined. The two FSH subunits were ligated to form a contiguous FSH{beta}–FSH{alpha} fusion without a 3' termination codon. The Fc fragment of human IgG1 (hinge, CH2 and CH3 domains; amino acids 221–447, EU numbering) was isolated from a human leukocyte cDNA library (Clontech) using the following primers: 5'-AAGAATTCGCCGGCGCCGCTGCGGTCGACAAAACTC-3' and 5'-TTCAATTGTCATTTACCCGGAGACAGGG-3'.

The area of each primer predicted to anneal to the Fc region of IgG1 is underlined. The 5' primer also contains sequence shown in bold that will create an eight amino acid linker sequence (EFAGAAAV) on the amino terminus of the Fc fragment. The Fc PCR product was cloned into the mammalian expression vector, pED.dC (Wyeth, USA) that contains an adenovirus major late promoter and a mouse dihydrofolate reductase gene as a selectable marker. The single chain FSH molecule was then cloned into pED.dC containing the human Fc sequence, thus creating a fusion molecule of FSH{beta}–FSH{alpha}–Fc with an eight amino acid linker between the FSH{alpha} and the Fc moieties (Figure 1a). Single chain FSH–Fc protein generated from this DNA construct will be ~128 kDa in size.



View larger version (17K):
[in this window]
[in a new window]
 
Figure 1. Schematic diagram of DNA constructs created to make single chain FSH–Fc (a), and heterodimer FSH–Fc proteins (b). Schematic diagram of single chain and heterodimer FSH–Fc fusion proteins (c). SDS–PAGE gel of single chain FSH–Fc and heterodimer FSH–Fc run under reducing and non-reducing conditions (d). Lane 1: single chain FSH–Fc reduced; lane 2: heterodimer FSH–Fc reduced; lane 3: single chain FSH–Fc non-reduced; and lane 4: heterodimer FSH–Fc non-reduced.

 
For heterodimer FSH–Fc, two DNA constructs were made encoding FSH{alpha}–Fc and FSH{beta}–Fc in unique expression plasmids. For FSH{alpha}–Fc, FSH{alpha} was isolated from a human pituitary mRNA library (Clontech) with its native signal sequence using the following primers: 5'-GGCTAGCCTGCAGGCCACCATGGATTACTACAGAAAATATGC-3', 5'-TCCACCGGATCCGCCTCCACCAGATTTGTGATAATAACAAGTACT-3'.

In the same way, the Fc fragment of IgG1 was isolated from a human leukocyte cDNA library (Clontech) using the following primers: 5'-GGAGGCGGATCCGGTGGAGGCGGGTCCGGCGGTGGAGGGAGCGACAAAACTCACACGTGCC-3' and 5'-GTTAGGCGGCCGCTCATTTACCCGGAGACAGGG-3'.

In each case, the sequence of the primers predicted to anneal to the corresponding template is underlined. A 15 amino acid linker sequence (GGGGSGGGGSGGGGS) was generated between the carboxy terminus of FSH{alpha} and the amino terminus of Fc. The sequence of each primer corresponding to this linker sequence is shown in bold. The adenovirus major late promoter in pED.dC was replaced by a cytomegalovirus (CMV) promoter using standard PCR techniques. The mammalian expression vector pcDNA6 (Invitrogen, USA) that contains a CMV promoter was the template for this PCR reaction. The FSH{alpha} and Fc PCR fragments were then cloned into this CMV–pED.dC vector (Figure 1b).

For FSH{beta}–Fc, FSH{beta} was isolated from a human pituitary mRNA library (Clontech) using the following primers: 5'-GGCTAGCCTGCAGGCCACCATGAAGACACTCCAGTTTTTCT-3' and 5'-TCCACCGGATCCGCCTCCACCTTCTTTCATTTCACCAAAGGAG-3'.

In the same way, the Fc fragment of IgG1 was isolated from a human leukocyte cDNA library (Clontech) using the following primers: 5'-GGAGGCGGATCCGGTGGAGGCGGGTCCGGCGGTGGAGGGAGCGACAAAACTCACACGTGCC-3' and 5'-GTTAGGCGGCCGCTCAGTGGTGATGGTGATGATGATGTTTACCCGGAGACAGGG-3'.

The sequence of each FSH{beta} and Fc primer predicted to anneal to the corresponding template is underlined. A 15 amino acid linker sequence (GGGGSGGGGSGGGGS) was generated between the carboxy terminus of FSH{beta} and the amino terminus of Fc. The sequence of each primer corresponding to this linker sequence is shown in bold. To facilitate protein purification, a carboxy terminal 6Histidine (6His) tag sequence was created on the Fc fragment. The sequence of the Fc primer corresponding to the 6His tag is shown in italics. The FSH{beta} and Fc–6His fragments were then cloned into the mammalian expression vector pcDNA6 (Invitrogen) that contains a CMV promoter and blasticidin gene for selection purposes (Figure 1b). Heterodimer FSH–Fc protein generated from co-transfection of FSH{alpha}–Fc and FSH{beta}–Fc constructs will be ~94.5 kDa in size.

Expression and purification of single chain and heterodimer FSH–Fc
Single chain FSH–Fc was transfected into CHO DG44 cells lacking a dihydrofolate reductase gene (obtained from Dr L.Chasin, Columbia University, NY, USA; Urlaub et al., 1986Go) using standard Superfect transfection protocols (Qiagen, USA). After 48 h, transfected cells were selected in minimum essential medium (MEM) {alpha} without ribonucleosides and deoxyribonucleosides containing 5% dialysed fetal bovine serum (FBS). To obtain higher protein expression levels, cells were treated with methotrexate at concentrations ≤200 nmol/l. For protein production, cells were seeded into roller bottles in Dulbecco's modified Eagle's medium (DMEM): F-12 (Invitrogen) +10% FBS and incubated for 3 days before changing the medium to DMEM: F-12+5 µg/ml human insulin (Invitrogen). Conditioned medium was collected daily for 10 days, then filtered through 0.2 µm filters, and stored at 4°C until purification. Single chain FSH–Fc was purified from culture medium using protein A affinity chromatography. After medium containing single chain FSH–Fc was loaded, protein A columns were washed with 5–10 column volumes of PBS (10 mmol/l phosphate pH 7.4, 2.7 mmol/l KCl and 137 mmol/l NaCl) and bound protein eluted with 0.1 mol/l glycine, pH 3.0. Eluted single chain FSH–Fc was then dialysed into PBS and stored at –80°C in aliquots containing 10% glycerol. Single chain FSH–Fc was run on Tris–glycine gels (Invitrogen) and stained with GelCode Blue (Pierce Chemical Company, USA) under reducing and non-reducing conditions to determine the purity of the protein. Single chain FSH–Fc was ~90% pure following a single protein A chromatography step (Figure 1d).

Heterodimer FSH–Fc was expressed by co-transfection of FSH{alpha}–Fc and FSH{beta}–Fc–6His expression vectors in CHO DG44 (Urlaub et al., 1986Go) cells using standard Superfect transfection methods. Forty-eight hours after transfection, cells were incubated in MEM{alpha} without ribonucleosides and deoxyribonucleosides containing 5% dialysed FBS and 10 µg/ml blasticidin (Invitrogen) to select only for cells containing both FSH{alpha}–Fc and FSH{beta}–Fc expression vectors. To obtain higher expression levels, cells were treated with methotrexate ≤50 nmol/l. For protein production, cells were seeded into roller bottles and culture medium containing secreted protein collected in the same way as for single chain FSH–Fc. Because FSH{alpha}–Fc and FSH{beta}–Fc are co-transfected, culture medium contains not only the desired heterodimer FSH–Fc, but also FSH{alpha}–Fc homodimer and FSH{beta}–Fc homodimer that require separation. An initial purification using protein A affinity chromatography was performed in the same way as described above for single chain FSH–Fc. Proteins were further purified using nickel affinity chromatography. FSH{beta}–Fc homodimers and heterodimer FSH–Fc bound to nickel affinity columns due to the presence of the 6His tag on FSH{beta}–Fc. Heterodimer FSH–Fc was separated from FSH{beta}–Fc homodimer by elution with an imidazole gradient (0–500 mmol/l). The imidazole concentration was then increased to 1 mol/l to remove any remaining protein. Heterodimer FSH–Fc eluted at ~30 to 90 mmol/l imidazole. Heterodimer FSH–Fc was run on Tris–glycine gels (Invitrogen) and stained with GelCode Blue (Pierce Chemical Company) under reducing and non-reducing conditions to determine the purity of the protein. Heterodimer FSH–Fc was ~90% pure following protein A and nickel affinity chromatography steps (Figure 1d).

In vivo bioactivity: ovarian weight gain assay
In vivo activity assays were carried out according to the protocol of Steelman and Pohley (1953)Go with minor modifications. Briefly, 21 day old female rats (10 rats per group) were given a single s.c. dose (1 nmol/kg) of recombinant (r)FSH (Follistim; Organon, The Netherlands), single chain FSH–Fc or heterodimer FSH–Fc in PBS. Seventy-two hours after dosing, ovarian weight was measured in each rat. Statistics were analysed using SigmaStat version 2.0 (RockWare Inc., USA).

Pharmacokinetic studies after oral dosing in neonatal rats
Ten day old neonatal rats (four rats/group) were dosed orally with 0.3 mg/kg single chain FSH–Fc or heterodimer FSH–Fc in PBS containing 5 mg/ml soybean trypsin inhibitor. At various times after dosing, blood was collected by cardiac puncture, allowed to clot, then serum prepared and stored at –20°C. A sandwich enzyme-linked immunosorbent assay (ELISA) was developed using an FSH coating antibody (Fitzgerald Industries, USA) and a horseradish peroxidase conjugated Fc detection antibody (Pierce Chemical Company, USA). The standard curve for the ELISA (200–0.78 ng/ml in 2-fold dilutions, limit of detection ~3 ng/ml) was created with the same lot of protein used to dose the rats. Samples were analysed in triplicate. Pharmacokinetic parameters were estimated using WinNonlin version 4.1 (Pharsight, USA).

Role of FcRn in oral delivery of single chain FSH–Fc and heterodimer FSH–Fc in neonatal rats
Single chain FSH–Fc and heterodimer FSH–Fc were iodinated with [125I]sodium iodide (Perkin Elmer, USA) using iodobeads (Pierce Chemical Company) according to manufacturer's protocols. Free iodine was separated from iodinated protein on a PD-10 desalting column. Ten day old rats were dosed orally with iodinated single chain or heterodimer FSH–Fc (25 pmol) or a mixture of iodinated single chain or heterodimer FSH–Fc (25 pmol) and a 300-fold molar excess of unlabelled human IgG (7.5 nmol; ICN, USA) in PBS with 5 mg/ml soybean trypsin inhibitor. Three hours after dosing, blood was collected by cardiac puncture and serum prepared. A 100 µl aliquot of serum was incubated with protein A Tris–acrylamide beads (Pierce Chemical Company) at 4°C for 1 h. Protein A beads were then washed twice with PBS and eluted with sodium dodecyl sulphate (SDS) sample buffer containing 20% {beta}-mercaptoethanol. Samples were boiled and analysed on 4–20% Tris–glycine gels, dried and quantification was performed on a Storm Phosphorimager (Molecular Dynamics, USA).

Oral bioactivity: testis weight gain assay
Oral bioactivity in male rats was determined according to a published protocol (Meachem et al., 1996Go) that demonstrated an increase in Sertoli cell number, resulting in an increase in testis weight, in neonatal rats treated with FSH daily for 10–20 days. Briefly, 2 day old male rats (10 rats per group) were orally dosed with 1 nmol/kg human rFSH (Follistim; Organon), single chain FSH–Fc or heterodimer FSH–Fc in PBS with 5 mg/ml soybean trypsin inhibitor. Rats were dosed daily for 14 days before measuring testis weight in each rat. Statistics were analysed using SigmaStat version 2.0 (RockWare, Inc.).

Pharmacokinetic/pharmacodynamic studies in cynomolgus monkeys
All studies with cynomolgus monkeys were conducted using approved protocols, following NIH guidelines for the care and use of research animals. Prior to pulmonary administration, animals were anesthetized with a combination of ketamine and valium and intubated with endotracheal tubes. Aerosols of single chain FSH–Fc (in PBS, pH7.4) or heterodimer FSH–Fc (in PBS, pH 7.4 with 0.1% HSA) were created with an Aeroneb ProTM nebulizer (Aerogen, USA) and administered to cynomolgus monkeys (deposited dose ~45 µg/kg) through the endotracheal tubes. A Bird Mark 7A respirator (Bird Products, USA) regulated the depth (20–40% vital capacity) and rate of respiration (28–30 breaths per minute) of each monkey such that the delivery of single chain FSH–Fc or heterodimer FSH–Fc was targeted to the central airways where epithelial expression of FcRn is predominantly located (Bitonti et al., 2004Go). Aerosol particle size was ~4–5 µm. Blood samples were collected at various times after pulmonary dosing and serum prepared. Single chain FSH–Fc and heterodimer FSH–Fc levels were quantified using commercially available FSH ELISA kits (DRG International, USA) according to manufacturer's directions. The standard curve for each assay (60 ng/ml to 2.5 ng/ml, limit of detection ~10 ng/ml) was created with the same lot of single chain FSH–Fc or heterodimer FSH–Fc used to dose the monkeys. Pharmacokinetic parameters were estimated using WinNonlin version 4.1 (Pharsight). Serum samples obtained after pulmonary dosing of single chain FSH–Fc or heterodimer FSH–Fc were also used to determine inhibin levels using commercially available ELISA kits (Diagnostic Systems Laboratories, USA) following manufacturer's instructions.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Protein characteristics
Figure 1c shows a schematic diagram of single chain FSH–Fc and heterodimer FSH–Fc fusion proteins. Single chain FSH–Fc is a fusion molecule of FSH{alpha}, FSH{beta} and the Fc portion of a human IgG1 molecule including the hinge, CH2 and CH3 domains. Thus an Fc dimer of single chain FSH–Fc contains two FSH{alpha} and two FSH{beta} subunits. In contrast, heterodimer FSH–Fc was made in such a way that a single FSH{alpha} subunit and a single FSH{beta} subunit were added to separate arms of the Fc dimer. Because of the extra FSH subunits in the single chain FSH–Fc molecule, this protein is larger than heterodimer FSH–Fc when the proteins are run under either reducing (~75 compared to 50 kDa respectively) or non-reducing (~150 compared to 100 kDa) conditions on SDS–polyacrylamide gel electrophoresis gels (Figure 1d). After purification, both single chain and heterodimer FSH–Fc are ~90% pure, and do not appear to contain Fc dimer (~26 or 52 kDa under reducing or non-reducing conditions respectively) or free FSH subunits (~30 kDa) on SDS–PAGE gels (Figure 1d).

In vivo bioactivity: ovarian weight gain assay
We compared the activity of single chain FSH–Fc and heterodimer FSH–Fc fusion proteins to that of recombinant human FSH in 21 day old female rats. Rats were given a single s.c. dose of 1 nmol/kg of either rFSH, single chain FSH–Fc or heterodimer FSH–Fc. Seventy-two hours after dosing, the right ovary from each animal was weighed (Figure 2). Ovarian weight was significantly increased in female rats treated with a single dose of rFSH compared to vehicle (14.3±1.7 mg compared to 12.1±1.0 mg, P=0.003). Single chain FSH–Fc and heterodimer FSH–Fc produced a greater increase in ovarian weight compared to vehicle- and FSH-treated groups (20.8±3.9 mg and 26.9±6.1 mg respectively; P<0.001). Heterodimer FSH–Fc was significantly more active than single chain FSH–Fc in this experiment (P=0.016).



View larger version (16K):
[in this window]
[in a new window]
 
Figure 2. Ovarian weight in 21 day old female rats treated with a single s.c. dose of 1 nmol/kg recombinant FSH, single chain FSH–Fc or heterodimer FSH–Fc. Ovarian weight was measured 72 h after dosing. Data are presented as average ovarian weight ± SD. n=10/group.

 
Pharmacokinetic studies after oral dosing in neonatal rats
Neonatal rats express high levels of FcRn in the small intestine during the first 3 weeks of life and therefore can be used to study oral uptake of FcRn binding molecules such as Fc-fusion proteins. Ten day old rats were dosed orally with single chain FSH–Fc or heterodimer FSH–Fc (0.3 mg/kg) and serum levels of protein determined with an FSH–Fc sandwich ELISA (Figure 3). Concentrations of single chain FSH–Fc and heterodimer FSH–Fc in neonatal rat serum reached maximum levels of 2.4 and 3.8 µg/ml respectively. Both proteins had long terminal half-lives of 60 h for single chain FSH–Fc and 69 h for heterodimer FSH–Fc.



View larger version (17K):
[in this window]
[in a new window]
 
Figure 3. Oral delivery of single chain and heterodimer FSH–Fc. Neonatal rats were given a single oral dose (0.3 mg/kg) of single chain or heterodimer FSH–Fc. Blood was collected at various times after administration, and serum levels of each protein measured with an FSH/Fc sandwich ELISA. n=4/time-point.

 
Role of FcRn in oral delivery of FSH–Fc single chain and FSH–Fc Heterodimer in neonatal rats
To show that oral delivery of single chain FSH–Fc and heterodimer FSH–Fc is due to FcRn binding and transcytosis, 10 day old rats were orally dosed with 125I-labelled single chain or heterodimer FSH–Fc, or a mixture of 125I-labelled single chain or heterodimer FSH–Fc with a 300-fold excess of unlabelled human IgG. Both single chain and heterodimer FSH–Fc were readily detected in the serum and were intact. Oral delivery of both single chain FSH–Fc and heterodimer FSH–Fc was greatly reduced in the presence of excess IgG (83% reduction and 53% reduction respectively as determined by phosphorimage analysis; Figure 4). Since IgG is a natural ligand for FcRn, this suggests that single chain and heterodimer FSH–Fc bind specifically to, and are transported by, FcRn.



View larger version (77K):
[in this window]
[in a new window]
 
Figure 4. Effect of excess IgG on oral uptake of single chain FSH–Fc and heterodimer FSH–Fc. Neonatal rats were dosed with 25 pmol [125I]single chain or heterodimer FSH–Fc in the presence or absence of 300-fold molar excess (7.5 nmol) of human IgG1. Protein A pulldowns from serum prepared were run under reducing conditions on a 4–20% Tris–glycine gel. Lane 1: 50 000 cpm single chain FSH–Fc; lane 2: serum [125I]single chain FSH–Fc; lane 3: serum [125I]single chain FSH–Fc in the presence of excess IgG; lane 4: 50 000 cpm heterodimer FSH–Fc; lane 5: serum [125I]heterodimer FSH–Fc; lane 6: serum [125I]heterodimer FSH–Fc in the presence of excess IgG.

 
Oral bioactivity: testis weight gain assay
Two day old male rats were orally dosed with either rFSH, single chain FSH–Fc or heterodimer FSH–Fc at a dose of 1 nmol/kg/day. After 14 days of treatment, the right testis of each animal was weighed. rFSH did not produce an increase in testis weight compared to the vehicle-treated group (49.1±8.1 compared to 55.4±8.1 mg respectively; Figure 5a). In contrast, single chain FSH–Fc (Figure 5a and b) and heterodimer FSH–Fc (Figure 5b) treatments resulted in significant increases in testis weight compared to vehicle-treated animals (113.0±19.8 and 139.6±11.9 mg compared to 58.6±10.4 mg respectively; P<0.001). Heterodimer FSH–Fc was significantly more active than single chain FSH–Fc in this experiment (P=0.003) which is consistent with the results of the s.c. dosing experiment in female rats.



View larger version (14K):
[in this window]
[in a new window]
 
Figure 5. Testis weight in 2 day old male rats treated daily for 14 days with single oral doses (1 nmol/kg) of (a) recombinant FSH or single chain FSH–Fc, and (b) single chain FSH–Fc or heterodimer FSH–Fc. Data presented as average testis weight ± SD. n=6–10/group.

 
Pharmacokinetic/pharmacodynamic studies in cynomolgus monkeys
We have previously shown expression of FcRn in cynomolgus monkey and human lung, and that an erythropoietin–Fc fusion protein is absorbed and retains activity after pulmonary administration (Spiekermann et al., 2002Go; Bitonti et al., 2004Go). We therefore tested whether FSH–Fc fusion proteins could be absorbed through the lung in a non-human primate and retain biological activity. Two cynomolgus monkeys were treated with a deposited dose of 45 µg/kg of either single chain FSH–Fc or heterodimer FSH–Fc, resulting in maximum serum concentrations of 92 and 94 ng/ml or 69 and 131 ng/ml respectively (Figure 6a and b). The terminal half-lives of single chain FSH–Fc and heterodimer FSH–Fc were long in both cases (55 and 210 h or 182 and 219 h respectively). To demonstrate biological activity of single chain FSH–Fc and heterodimer FSH–Fc after absorption through the lungs, inhibin levels were measured in the same serum samples analysed for protein levels (Figure 7a and b). In each of the two monkeys treated with single chain FSH–Fc, inhibin levels increased to a maximum concentration of 1 and 1.6 ng/ml corresponding to a 1.3- and 1.4-fold stimulation above baseline. In each of the two monkeys treated with heterodimer FSH–Fc, inhibin reached maximal serum concentrations of 2.7 and 7.4 ng/ml corresponding to a 5.9- and 7.1-fold stimulation over baseline levels. Similar to the results of the two rat models of FSH activity shown above, heterodimer FSH–Fc was more active in monkeys after pulmonary dosing compared to single chain FSH–Fc.



View larger version (13K):
[in this window]
[in a new window]
 
Figure 6. Pulmonary delivery of single chain FSH–Fc (a) or heterodimer FSH–Fc (b) in cynomolgus monkeys. FSH–Fc fusion proteins were aerosolized and a deposited dose of ~45 µg/kg delivered directly into monkey lungs through an endotracheal tube. Serum concentrations of single chain or heterodimer FSH–Fc were measured with an FSH ELISA. Each curve is representative of a single monkey.

 


View larger version (14K):
[in this window]
[in a new window]
 
Figure 7. Inhibin B levels in cynomolgus monkey serum following a pulmonary dose of single chain FSH–Fc (a) or heterodimer FSH–Fc (b). Bioactivity of single chain and heterodimer FSH–Fc were quantified using a specific inhibin B ELISA on serum samples following pulmonary administration of ~45 µg/kg of the fusion proteins.

 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
FcRn is an endogenous receptor expressed in neonatal rodent epithelial cells of the small intestine, and is responsible for the transport of antibodies from maternal milk into the neonatal bloodstream (Simister et al., 1997Go). We have previously used this model to demonstrate pulmonary transport of another Fc fusion protein, erythropoietin–Fc (Bitonti et al., 2004Go). The results of the studies reported here demonstrate that single chain and heterodimer FSH–Fc fusion proteins of large molecular weight (128 and 94.5 kDa respectively) can be delivered systemically at high concentrations after oral dosing in neonatal rats. The terminal half-life of single chain and heterodimer FSH–Fc in this model (60–69 h) is significantly longer than that of rFSH (11.4 h; de Leeuw et al., 1996Go). We have also shown that the transport of single chain and heterodimer FSH–Fc fusion proteins is mediated by FcRn transcytosis since an excess of the endogenous ligand for FcRn (IgG) reduced the transport of FSH–Fc fusion proteins after oral dosing in this model.

Single chain and heterodimer FSH–Fc are active in both ovarian and testis weight gain assays in rats. This is somewhat surprising since it has been shown that the carboxy terminal of the {alpha} subunit of FSH is critical for binding to the FSH receptor (Chen et al., 1992Go; Arnold et al., 1998Go). In addition, the recent publication of the crystal structure of human FSH complexed with its receptor clearly shows that the carboxy terminus of FSH{alpha} is in close contact with the FSH receptor (Fan and Hendrickson, 2005Go), and comparison of this crystal structure with that of FSH alone (Fox et al., 2001Go) indicates that the carboxy terminus of FSH{alpha} undergoes a conformational change when bound to the FSH receptor, causing the carboxy terminus of FSH{alpha} to become more rigid in structure. Since the carboxy terminus of FSH{alpha} is critical for receptor binding, it would seem unlikely that FSH would retain biological activity if the FSH{alpha} subunit is tethered at the carboxy terminus by Fc. In both single chain and heterodimer FSH–Fc fusion proteins, a linker sequence (of eight or 15 amino acids respectively) was used to connect the carboxy terminus of FSH{alpha} to the Fc moiety. It is possible that these linker sequences provide enough flexibility for FSH to bind to its receptor, thus retaining in vivo FSH activity for both fusion proteins.

Single chain and heterodimer FSH–Fc both retain bioactivity after oral dosing in male neonatal rats, leading to an increase in testis weight in treated animals compared to rFSH and vehicle controls. Since FSH binding in the testis is restricted to the Sertoli cell (Simoni et al., 1997Go), and since it has been shown that daily treatment of newborn male rats with FSH for 10–20 days results in an increase in Sertoli cell number resulting in an increase in testis weight (Meachem et al., 1996Go), it seems likely that single chain and heterodimer FSH–Fc treatment results in an increase in Sertoli cell number that is similarly reflected by an increase in testis weight. Although single chain and heterodimer FSH–Fc are active after oral dosing in neonatal rats, rFSH is not. Since we also show that transport of FSH–Fc fusion proteins is specifically mediated by FcRn binding and transport through the Fc moiety, it is likey that rFSH is not transported effectively after oral administration, thus explaining the lack of effect on testis weight.

Bioactivity of single chain and heterodimer FSH–Fc after s.c. dosing was also demonstrated in 21 day old female rats using the Steelman–Pohley assay (Steelman and Pohley, 1953Go). FSH binds to specific receptors on granulosa cells of the ovary and is responsible for the selection and growth of ovarian follicles that under specific conditions results in an increase in ovarian weight. Rats treated with rFSH or either FSH–Fc fusion protein had increased ovarian weights compared to vehicle-treated controls, likely due to an increased number and size of ovarian follicles. However, the stimulation in ovarian weight was greater for the FSH–Fc fusion proteins compared to rFSH. Since it is unlikely that the FSH–Fc fusion proteins will inherently have more activity than rFSH, the increased activity is probably due to the long terminal half-life (60–69 h) of these molecules in rodent circulation.

Ovarian and testis weight gain assays in rats indicated that both single chain and heterodimer FSH–Fc are significantly more active than rFSH. However, it is apparent that heterodimer FSH–Fc is also significantly more active than single chain FSH–Fc. Pulmonary delivery of single chain and heterodimer FSH–Fc in cynomolgus monkeys also supports this observation in that inhibin levels are increased to a higher level in response to heterodimer FSH–Fc compared to single chain FSH–Fc. Since single chain and heterodimer FSH–Fc have similar terminal half-lives in both rodents and monkeys, it is likely that the structure of heterodimer FSH–Fc may present the FSH{alpha} and FSH{beta} subunits in a more favourable conformation for bioactivity than if the FSH{alpha} and FSH{beta} are fused directly together with Fc to form single chain FSH–Fc.

rFSH is a common therapy in the treatment of both male and female infertility that requires daily injections for varying lengths of time. For example, treatment of hypogonadotrophic men currently requires daily injections of FSH for up to several months in order to induce spermatogenesis (Schaison et al., 1993Go; Schoot et al., 1994Go). The half-life of heterodimer FSH–Fc in cynomolgus monkeys in this study is 182–219 h. This is significantly longer than the half-life of rFSH of ~24 h in humans (le Cotonnec et al., 1994Go) and in non-human primates (Porchet et al., 1993Go; Weinbauer et al., 1994Go). Thus an obvious advantage of using heterodimer FSH–Fc in infertility treatments is the potential for a greatly reduced dosing frequency. In addition, pulmonary or oral delivery of FSH–Fc fusion proteins using endogenous FcRn expressed in epithelial cells of the lung and intestine could significantly improve tolerability of current infertility treatments.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Arnold CJ, Liu C, Lindau-Shepard B, Losavio ML, Patrascu MT and Dias JA (1998) The human follitropin {alpha}-subunit C terminus collaborates with a {beta}-subunit cystine noose and an {alpha}-subunit loop to assemble a receptor-binding domain competent for signal transduction. Biochemistry 37, 1762–1768.[CrossRef][ISI][Medline]

Ashkenazi A, Capon DJ and Ward RH (1993) Immunoadhesins. Int Rev Immunol 10, 219–227.[Medline]

Bitonti AJ, Dumont JA, Low SC, Peters RT, Kropp KE, Palombella VJ, Stattel JM, Lu Y, Tan CA, Song JJ et al. (2004) Pulmonary delivery of an erythropoietin Fc fusion protein in non-human primates through an immunoglobulin transport pathway. Proc Natl Acad Sci USA 101, 9763–9768.[Abstract/Free Full Text]

Bouloux PM, Handelsman DJ, Jockenhovel F, Nieschlag E, Rabinovici J, Frasa WL, de Bie JJ, Voortman G and Itskovitz-Eldor J, On behalf of the FSH–CTP study group. (2001) First human exposure to FSH–CTP in hypogonadotropic hypogonadal males. Hum Reprod 16, 1592–1597.[Abstract/Free Full Text]

Chamow SM and Ashkenazi A (1996) Immunoadhesins: principles and applications. Trends Biotechnol 14, 52–60.[CrossRef][ISI][Medline]

Chen F, Wang Y and Puett D (1992) The carboxy-terminal region of the glycoprotein hormone {alpha}-subunit: contributions to receptor binding and signaling in human chorionic gonadotropin. Mol Endocrinol 6, 914–919.[Abstract]

de Leeuw R, Mulders J, Voortman G, Rombout F, Damm J and Kloosterboer L (1996) Structure–function relationship of recombinant follicle stimulating hormone (Puregon). Mol Hum Reprod 2, 361–369.[Abstract]

Dickinson BL, Badizadegan K, Wu Z, Ahouse JC, Zhu X, Simister NE, Blumberg RS and Lencer WI (1999) Bidirectional FcRn-dependent IgG transport in a polarized human intestinal epithelial cell line. J Clin Invest 104, 903–911.[Abstract/Free Full Text]

Duijkers IJ, Klipping C, Boerrigter PJ, Machielsen CS, De Bie JJ and Voortman G (2002) Single dose pharmacokinetics and effects on follicular growth and serum hormones of a long-acting recombinant FSH preparation (FSH–CTP) in healthy pituitary-suppressed females. Hum Reprod 17, 1987–1993.[Abstract/Free Full Text]

Economides AN, Carpenter LR, Rudge JS, Wong V, Koehler-Stec EM, Hartnett C, Pyles EA, Xu X, Daly TJ, Young MR et al. (2003) Cytokine traps: multi-component, high-affinity blockers of cytokine action. Nat Med 9, 47–52.[CrossRef][ISI][Medline]

Fan QR and Hendrickson WA (2005) Structure of human follicle-stimulating hormone in complex with its receptor. Nature 433, 269–277.[CrossRef][ISI][Medline]

Fox KM, Dias JA and Van Roey P (2001) Three-dimensional structure of human follicle stimulating hormone. Mol Endocrinol 15, 378–389.[Abstract/Free Full Text]

Israel EJ, Taylor S, Wu Z, Mizoguchi E, Blumberg RS, Bhan A and Simister NE (1997) Expression of the neonatal Fc receptor, FcRn, on human intestinal epithelial cells. Immunology 92, 69–74.[CrossRef][ISI][Medline]

Jia XC and Hsueh AJW (1986) Granulosa cell aromatase bioassay for follicle-stimulating hormone: validation and application of the method. Endocrinology 119, 1570–1577.[Abstract]

Junghans RP and Anderson CL (1996) The protection receptor for IgG catabolism is the beta2-microglobulin-containing neonatal intestinal transport receptor. Proc Natl Acad Sci USA 93, 5512–5516.[Abstract/Free Full Text]

Keene JL, Matzuk MM, Otani T, Fauser BC, Galway AB, Hsueh AJ and Boime I (1989) Expression of biologically active human follitropin in Chinese hamster ovary cells. J Biol Chem 264, 4769–4775.[Abstract/Free Full Text]

Kim J-K, Tsen M-F, Ghetie V and Ward ES (1994) Localization of the site of the murine IgG1 molecule that is involved in binding to the murine intestinal Fc receptor. Eur J Immunol 24, 2429–2434.[ISI][Medline]

Kim YS, Maslinski W, Zheng XX, Schachter AD and Strom TB (1998) Immunoglobulin–cytokine fusion molecules: the new generation of immunomodulating agents. Transplant Proc 30, 4301–4306.

Klein J, Lobel L, Pollak S, Lustbader B, Ogden RT, Sauer MV and Lustbader JW (2003) Development and characterization of a long acting recombinant hFSH agonist. Hum Reprod 18, 50–56.[Abstract/Free Full Text]

le Cotonnec J-Y, Porchet HC, Beltrami V, Khan A, Toon S and Rowland M (1994) Clinical pharmacology of recombinant human follicle-stimulating hormone. II. Single doses and steady state pharmacokinetics. Fertil Steril 61, 679–686.[ISI][Medline]

Meachem SJ, McLachlan RI, de Kretser DM, Robertson DM and Wreford NG (1996) Neonatal exposure of rats to recombinant follicle stimulating hormone increases adult sertoli and spermatogenic cell numbers. Biol Reprod 54, 36–44.[Abstract]

Pierce JG and Parsons TF (1981) Glycoprotein hormones: structure and function. Annu Rev Biochem 50, 465–495.[CrossRef][ISI][Medline]

Porchet HC, le Cotonnec J-Y, Canali S and Zanolo G (1993) Pharmacokinetics of recombinant human follicle stimulating hormone after intravenous, intramuscular, and subcutaneous administration in monkeys, and comparison with intravenous administration of urinary follicle stimulating hormone. Drug Metab Dispos 21, 144–150.[Abstract/Free Full Text]

Schaison G, Young J, Pholsena M, Nahoul K and Couzinet B (1993) Failure of combined follicle-stimulating hormone–testosterone administration to initiate and/or maintain spermatogenesis in men with hypogonadotropic hypogonadism. J Clin Endocrinol Metab 77, 1545–1549.[Abstract]

Schoot DC, Harlin J, Shoham Z, Mannaerts BM, Lahlou N, Bouchard P, Coelingh Bennink HJ and Fauser BC (1994) Recombinant human follicle-stimulating hormone and ovarian response in gonadotropin-deficient women. Hum Reprod 9, 1237–1242.[Abstract]

Simister NE, Jacobowitz Israel E, Ahouse JC and Story CM (1997) New functions of the MHC class I-related Fc receptor, FcRn. Biochem Soc Trans 25, 481–486.[ISI][Medline]

Simoni M, Gromoll J and Nieschlag E (1997) The follicle-stimulating hormone receptor: biochemistry, molecular biology, physiology, and pathophysiology. Endocr Rev 18, 739–773.[Abstract/Free Full Text]

Spiekermann GM, Finn PW, Ward ES, Dumont J, Dickinson BL, Blumberg RS and Lencer WI (2002) Receptor-mediated immunoglobulin G transport across mucosal barriers in adult life: functional expression of FcRn in the mammalian lung. J Exp Med 196, 300–310.

Steelman SL and Pohley FM (1953) Assay of the follicle stimulating hormone based on the augmentation with human chorionic gonadotropin. Endocrinology 53, 604–616.[ISI][Medline]

Story CM, Mikulska JE and Simister NE (1994) A major histocompatibility complex class I-like Fc receptor cloned from human placenta: possible role in transfer of immunoglobulin G from mother to fetus. J Exp Med 180, 2377–2381.[Abstract/Free Full Text]

Sugahara T, Grootenhuis PD, Sato A, Kudo M, Ben-Menahem D, Pixley MR, Hsueh AJ 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, 71–77.[CrossRef][ISI][Medline]

Urlaub G, Mitchell PJ, Kas E, Chasin LA, Funanage VL, Myoda TT and Hamlin J (1986) Effect of gamma rays at the dihydrofolate reductase locus: deletions and inversions. Somat Cell Mol Genet 12, 555–566.[CrossRef][ISI][Medline]

Weinbauer GF, Simoni M, Hutchison JS and Nieschlag E (1994) Pharmacokinetics and pharmacodynamics of recombinant and urinary human FSH in the male monkey (Macaca fasicularis). J Endocrinol 141, 113–121.[Abstract]

Submitted on January 27, 2005; resubmitted on March 3, 2005; accepted on March 8, 2005.





This Article
Abstract
Full Text (PDF )
All Versions of this Article:
20/7/1805    most recent
deh896v1
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Request Permissions
Google Scholar
Articles by Low, S.C.
Articles by Dumont, J.A.
PubMed
PubMed Citation
Articles by Low, S.C.
Articles by Dumont, J.A.