WO2009127826A1

WO2009127826A1 - Recombinant fsh including alpha 2,3- and alpha 2,6-sialylation

Info

Publication number: WO2009127826A1
Application number: PCT/GB2009/000978
Authority: WO
Inventors: Ian Cottingham; Daniel Plaksin; Richard Boyd White
Original assignee: Ferring International Center Sa
Priority date: 2008-04-16
Filing date: 2009-04-16
Publication date: 2009-10-22
Also published as: HUE030652T2; RU2537268C2; JP2018021037A; KR20110005863A; KR101622944B1; JP2015120696A; ZA201007373B; AU2017204258A1; DK3144318T3; MX355457B; AU2009237479A1; JP2024038000A; HK1146284A1; FI3098234T3; EP3144318A1; DK2268666T3; EP3144318B1; HRP20161520T1; JP6762916B2; BRPI0910461A2

Abstract

Preparations including recombinant FSH (rFSH).

Description

RECOMBINANT FSH INCLUDING ALPHA 2,3- AND ALPHA 2,6-SIALYLATION

The present invention relates to gonadotrophins for use in the treatment of infertility. In particular it relates to follicle stimulating hormone (FSH).

The gonadotrophins are a group of heterodimeric glycoprotein hormones which regulate gonadal function in the male and female. They include follicle stimulating hormone (FSH), luteinising hormone (LH) and chorionic gonadotrophin (CG).

FSH is naturally secreted by the anterior pituitary gland and functions to support follicular development and ovulation. FSH comprises a 92 amino acid alpha sub-unit, also common to the other glycoprotein hormones LH and CG, and a 111 amino acid beta sub-unit unique to FSH that confers the biological specificity of the hormone (Pierce and Parsons, 1981). Each sub-unit is post translationally modified by the addition of complex carbohydrate residues. Both subunits carry 2 sites for N-linked glycan attachment, the alpha sub-unit at amino acids 52 and 78 and the beta sub-unit at amino acid residues 7 and 24 (Rathnam and Saxena, 1975, Saxena and Rathnam, 1976). FSH is thus glycosylated to about 30% by mass (Dias and Van Roey. 2001. Fox et a/. 2001).

FSH purified from post-menopausal human urine has been used for many years in infertility treatment; both to promote ovulation in natural reproduction and to provide oocytes for assisted reproduction technologies. Two recombinant versions of FSH, Gonal-F (Serono) and Puregon (Organon) became available in the mid-1990's. These are both expressed in Chinese hamster ovary (CHO) cells (Howies, 1996).

There is considerable heterogeneity associated with FSH preparations which relates to differences in the amounts of various isoforms present. Individual FSH isoforms exhibit identical amino acid sequences but differ in the extent to which they are post-translationally modified; particular isoforms are characterised by heterogeneity of the carbohydrate branch structures and differing amounts of sialic acid (a terminal sugar) incorporation, both of which appear to influence the specific isoform bioactivity.

Glycosylation of natural FSH is highly complex. The glycans in naturally derived pituitary FSH can contain a wide range of structures that can include combinations of bi-, tri- and tetra-antennary glycans (Pierce and Parsons, 1981. Ryan et al., 1987. Baenziger and Green, 1988). The glycans can carry further modifications: core fucosylation, bisecting glucosamine, chains extended with acetyl lactosamine, partial or complete sialylation, sialylation with α2,3 and α2,6 linkages, and sulphated galactosamine substituted for galactose (Dalpathado et al., 2006). Furthermore, there are differences between the distributions of glycan structures at the individual glycosylation sites. A comparable level of glycan complexity has been found in FSH derived from the serum of individuals and from the urine of post-menopausal women (Wide et al., 2007).

The glycosylation of recombinant FSH products reflects the range of glycosyl-transferases present in the host cell line. Existing rFSH products are derived from engineered Chinese hamster ovary cells (CHO cells). The range of glycan modifications in CHO derived rFSH are more limited than those found on the natural products, derived either from pituitary extracts or urine. Examples of the reduced glycan heterogeneity found in CHO derived rFSH include a lack of bisecting glucosamine and a reduced content of core fucosylation and acetyl lactosamine extensions (Hard et al., 1990). In addition, CHO cells are only able to add sialic acid using the α2,3 linkage (Kagawa et al, 1988, Takeuchi et al, 1988, Svensson et al., 1990). This is different from naturally produced FSH which contains glycans with a mixture of α2,3 and α2,6-linked sialic acid. It has been demonstrated that a recombinant FSH preparation (Organon) differs in the amounts of FSH with an isoelectric point (pi) of below 4 (considered the acidic isoforms) when compared to pituitary, serum or post-menopausal urine FSH (Ulloa-Aguirre et al. 1995). The amount of acidic isoforms in the urinary preparations was much higher as compared to the recombinant products, Gonal-f (Serono) and Puregon (Organon) (Andersen et al. 2004). This must reflect a lower molar content of sialic acid in the rFSH since the content of negatively-charged glycan modified with sulphate is low in FSH. The lower sialic acid content, compared to natural FSH, is a feature of both commercially available FSH products and therefore must reflect a limitation in the manufacturing process (Bassett and Driebergen, 2005).

There is a large body of scientific work which analyses and tries to explain the variations in FSH glycosylation between individuals and changes over the course of an ovulation cycle. One of the major discussions relates to the observation that FSH concentration and sialic acid content both decrease during the preovulatory phase of the cycle.

The decreased sialic acid content results in a more basic FSH which is both cleared more rapidly and, in vitro at least, is more potent at the target receptor (Zambrano et al. 1996). The question as to the biological relevance of these changes and how they may be involved in selecting the dominant follicle remains unresolved (reviewed by Ulloa-Aguirre, 2003).

The circulatory life-time of FSH has been documented for materials from a variety of sources. Some of these materials have been fractionated on the basis of overall molecular charge, as characterised by their pi, in which more acid equates to a higher negative charge. As previously stated the major contributor to overall molecular charge is the total sialic content of each FSH molecule. For instance, rFSH (Organon) has a sialic acid content of around 8 mol/mol, whereas urine-derived FSH has a higher sialic acid content (de Leeuw et al. 1996). The corresponding plasma clearance rates in the rat are 0.34 and 0.14 ml/min (Ulloa-Aguirre et al. 2003). In another example where a sample of recombinant FSH was split into high and low pi fractions, the in vivo potency of the high pi (lower sialic acid content) fraction was decreased and it had a shorter plasma half-life (D'Antonio et al. 1999). It has also been reported that the more basic FSH circulating during the later stages of the ovulation cycle is due to the down- regulation of α2,3 sialyl-transferase in the anterior pituitary which is caused by increasing levels of estradiol (Damian-Matsumara et al. 1999. Ulloa- Aguirre et al. 2001). Results for the α2,6 sialyl-transferase have not been reported.

The total sialic acid content of FSH and rFSH is not directly comparable since sialic acids are commonly linked in two ways. Pituitary/ serum/ urinary FSH contain both α2,3 and α2,6-linked sialic acid, with a predominance of the former. However, CHO cell derived recombinants only contain α2,3 (Kagawa et al, 1988, Takeuchi et al, 1988, Svensson et al., 1990). This is another difference between natural and current recombinant products in addition to the lower overall sialic acid content of the latter.

CHO cells are commonly used for the production of pharmaceutical human recombinant proteins. Structural analysis has identified that sialic acid is exclusively attached by a α2,3-linkage. (Kagawa et al, 1988, Takeuchi et al, 1988, Svensson et al., 1990). Many human glycoproteins contain a mixture of both α2,3- and α2,6-linkages. Therefore recombinant proteins expressed using the CHO system will differ from their natural counterparts in their type of terminal sialic acid linkages. This is an important consideration in the production of biologicals for pharmaceutical use since the carbohydrate moieties may contribute to the pharmacological attributes of the molecule.

It is desirable to have a rFSH product that more closely replicates or mimics the physiochemical and pharmacokinetic profile of the product produced from human urine. It is desirable to have a rFSH product that has improved pharmacokinetic property or properties compared to the known recombinant product.

According to the present invention there is provided recombinant FSH ("rFSH" or "recFSH") including α2, 3 sialylation and α2, 6 sialylation and, optionally,α2, 8 sialylation. The rFSH (or rFSH preparation) according to the invention may have 10% or more of the total sialylation being α2,3- sialylation, for example 65-85% of the total sialylation may be α2,3- sialylation. The rFSH (or rFSH preparation) of the invention may have 50% or less of the total sialylation being α2,6-sialylation, for example 15-35% of the total sialylation may be α2,6- sialylation. The rFSH (or rFSH preparation) of the invention may have 5% or less of the total sialylation being α2,8-sialylation, for example 0.1-4% of the total sialylation may be α2,8- sialylation. The rFSH (or rFSH preparation) according to the invention may have a sialic acid content [expressed in terms of a ratio of moles of sialic acid to moles of protein] of 6 mol/mol or greater, for example of between 6 mol/mol and 15 mol/mol.

The applicants have found that the type of sialic acid linkage, α2,3- or α2,6-, can have a dramatic influence on biological clearance of FSH. Human cell lines, as opposed to CHO cell lines, can express recombinant FSH with sialic acids attached by both α2,3 and α2,6 linkages. In Example 4 a recombinant FSH cell line was made which expressed FSH containing glycans with low levels of both α2,3- and α2,6-linked sialic acid (Figure 6). This basic material, with limited sialic acid content (Figure 4) was cleared very quickly from the circulation in rat as would be predicted (Figure 7). The cell line was then subjected to a second engineering step with the addition of the gene encoding for the α2,6-sialyl-transferase (Example 5). The resulting rFSH was highly sialylated showing sialic acid content and pi distribution comparable with urinary FSH (Figure 5). However, the material was cleared very rapidly from circulation of rats at a rate comparable to the original material which had low sialic acid content (Figure 8). This was an unexpected observation since it is known that a proportion of sialic acid on natural and biologically active FSH is α2,6-linked. The clearance of the α2,6-sialylated rFSH was found to be mediated by the asialoglycoprotein (ASGP) receptor found in the liver (Example 9). This was demonstrated by transient blockade of the ASGP receptors using an excess of another substrate for the receptor. With the receptor blocked by asialofetuin, the expected clearance for the highly-sialylated material was restored (Figure 9). This was maintained for several hours until the blockade was overcome and the α2,6 linked highly sialylated rFSH resumed its rapid clearance.

Recombinant FSH with a mixture of both α2,3 and α2,6-linked sialic acid was made by engineering a human cell line to express both rFSH and c-2,3 sialyltransferase (Example 4 and 5). The expressed product is highly acidic and carries a mix of both α2,3- and α2,6-linked sialic acids; the latter provided by the endogenous sialyl transferase activity (Figure 6). This has two advantages over rFSH expressed in conventional CHO cells: first the material is more highly sialylated due to the combined activities of the two sialyltransferases; and secondly the material more closely resembles the natural FSH. This is likely to be more biologically appropriate compared to CHO cell derived recombinant products that have produce only α2,3 linked sialic acid (Kagawa et al, 1988, Takeuchi et al, 1988, Svensson et al., 1990) and have decreased sialic acid content (Ulloa-Aguirre et al. 1995., Andersen et al. 2004).

The applicants have surprisingly found that rFSH of the invention may more closely replicate or mimic the physiochemical and pharmacokinetic profile of the natural human urinary product than other recombinant products. In other words, rFSH of the invention may be closer to the "natural" FSH. This may have significant advantages regarding dosing etc. Further, a more "natural" or more "human" product may be more desirable to the patient, who may desire therapy, although in a sense artificial, to be as "natural" as possible. There may be other advantages (e.g. pharmacokinetic advantages) in a recombinant product having carbohydrate (e.g. glycan) structure which is closer to natural (e.g. human urinary) FSH than other recombinant products.

The invention is thus a recombinant version of FSH which carries a mix of α2,3 and α2,6 sialic acid and therefore more closely resembles natural FSH. It is expected that the use of this compound for controlled ovarian stimulation, in IVF techniques, and ovulation induction will result in a more natural stimulation of the ovary compared to existing recombinant products.

According to the present invention there is provided recombinant

FSH ("rFSH" ot .l'recFSH") (and/or a recombinant FSH preparation) including α2, 3 sialylation and α2, 6 sialylation. The rFSH or rFSH preparation may optionally further include α2, 8 sialylation.

Herein term "recombinant FSH preparation" includes a preparation for e.g. pharmaceutical use which includes recombinant FSH. In embodiments of the invention, the rFSH may be present as a single isoform or as a mixture of isoforms.

The rFSH (or rFSH preparation) according to the invention may have a sialic acid content [expressed in terms of a ratio of moles of sialic acid to moles of protein] of 6 mol/mol or greater (Example 8), for example between 6 mol/mol and 15 mol/mol, e.g between 8 mol/mol and 14 mol/mol, for example between 10 mol/mol and 14 mol/mol, e.g between 11 mol/mol and 14 mol/mol, e.g between 12 mol/mol and 14 mol/mol, e.g. between 12 mol/mol and 13 mol/mol. The rFSH of the invention may be produced or expressed in a human cell line.

The rFSH (or rFSH preparation) according to the invention may have 10% or more of the total sialylation being α2,3-sialylation. For example, 20, 30, 40, 50, 60, 70, 80 or 90% or more of the total sialylation may be α2,3-sialylation. The rFSH (or rFSH preparation) may include α2,3- sialylation in an amount which is from 65 to 85% of the total sialylation, for example from 70 to 80% of the total sialylation, for example from 71 to 79% of the total sialylation. The rFSH (or rFSH preparation) of the invention may have 50% or less of the total sialylation being α2,6- sialylation. For example 40, 30, 20, 10, 5% or less of the total sialylation may be α2,6- sialylation. The rFSH (or rFSH preparation) may include α2,6-sialylation in an amount which is from 15 to 35% of the total sialylation, for example from 20 to 30% of the total sialylation, for example from 21 to 29% of the total sialylation. The rFSH (or rFSH preparation) of the invention may have 5% or less of the total sialylation being α2,8- sialylation. For example 2.5% or less of the total sialylation may be α2,8- sialylation. The rFSH (or rFSH preparation) may include α2,8-sialylation in an amount which is from 0.1 to 4% of the total sialylation, for example from 0.5 to 3% of the total sialylation, for example from 0.5 to 2.5% of the total sialylation, By sialylation it is meant the amount of sialic residues present on the FSH carbohydrate structures. α2,3-sialylation means sialylation at the 2,3 position (as is well known in the art) and α2,6 sialylation at the 2,6 position (also well known in the art). Thus "% of the total sialylation may be α 2,3 sialylation" refers to the % of the total number of sialic acid residues present in the FSH which are sialylated in the 2,3 position. The term "% of the total sialylation being α2,6-sialylation" refers to the % of the total number of sialic acid residues present in the FSH which are sialylated in the 2,6 position.

The rFSH (or rFSH preparation) according to the invention may have a sialic acid content (amount of sialylation per FSH molecule) of (based on the mass of protein, rather than the mass of protein plus carbohydrate) of 6% or greater (e.g. between 6% and 15%, e.g. between 7% and 13%, e.g. between 8% and 12%, e.g. between 11% and 15%, e.g. between 12% and 14%) by mass. Recombinant FSH expressed in Chinese hamster ovary (CHO) cells includes exclusively α 2, 3 sialylation (Kagawa et al, 1988, Takeuchi et al. 1988, Svensson et al. 1990).

The rFSH of the invention may be produced or expressed in a human cell line. This may simplify (and render more efficient) the production method because manipulation and control of e.g. the cell growth medium to retain sialylation may be less critical than with known processes. The method may also be more efficient because there is little basic rFSH produced than in production of known rFSH products; more acidic rFSH is produced and separation/removal of basic FSH is less problematic. The rFSH may be produced or expressed in a Per.Cβ cell line, a Per.CΘ derived cell line or a modified Per.C6 cell line. The cell line may be modified using α2,3-sialyltransferase. The cell line may be modified using α2,6-sialyltransferase. Alternatively or additionally, the rFSH may include α2,6-linked sialic acids (α2,6 sialylation) provided by endogenous sialyl transferase activity [of the cell line].

The rFSH may be produced using α2,3- and/or α2,6- sialyltransferase. The rFSH may be produced using α2,3- sialyltransferase. The rFSH may include α2,6-linked sialic acids (α2,6 sialylation) provided by endogenous sialyl transferase activity.

According to the present invention in a further aspect there is provided a method of production of rFSH and/or an rFSH preparation as described herein (according to aspects of the invention) comprising the step of producing or expressing the rFSH in a human cell line, for example a Per.C6 cell line, a Per.CΘ derived cell line or a modified Per.CΘ cell line, for example a cell line which has been modified using α2,3- sialyltransferase.

The rFSH structure contains glycan moieties. Branching can occur with the result that the glycan may have 1 , 2, 3, 4 or more terminal sugar residues or "antennae", as is well known in the art. The rFSH of the invention may have glycans with sialylation presence on mono-antennary and/or di-antennary and/or tri-antennary and/or tetra-antennary structures. The rFSH may preferably include mono-sialylated, di- sialylated, tri- sialylated and tetra- sialylated glycan structures with relative amounts as follows: 9-15% mono-sialylated; 27 - 30% di— sialylated; 30 - 36% tri- sialylated and 25 - 29 % tetra-sialylated (e.g. as shown by WAX analysis of charged glycans, as set out in Example 8 c).

According to the present invention in a further aspect there is provided rFSH produced (e.g. expressed) in a human cell line. The rFSH may include α2,3- and α2,6-sialylation. The rFSH may be produced or expressed in a Per.Cδ cell line, a Per.C6 derived cell line or a modified Per.Cβ cell line. The cell line may be modified using α2,3- sialyltransferase. The cell line may be modified using α2,6- sialyltransferase. Alternatively or additionally, the rFSH may include α2,6- linked sialic acids (α2,6 sialylation) provided by endogenous sialyl transferase activity [of the cell line]. The rFSH (or rFSH preparation) may have 10% or more of the total sialylation being α2,3-sialylation, for example 65-85% of the total sialylation may be α2,3-sialylation. The rFSH (or rFSH preparation) of the invention may have 50% or less of the total sialylation being α2,6-sialylation, for example 15-35% of the total sialylation may be α2,6- sialylation. The rFSH (or rFSH preparation) of the invention may have 5% or less of the total sialylation being α2,8- sialylation, for example 0.5-4% of the total sialylation may be α2,8- sialylation. The rFSH may have a sialic acid content [expressed in terms of a ratio of moles of sialic acid to moles of protein] of 6 mol/mol or greater, for example between 6mol/mol and 15 mol/mol.

According to the present invention in a further aspect there is provided a pharmaceutical composition comprising rFSH including α2,3- sialylation and α2,6-sialylation (e.g. as set out above). The pharmaceutical composition may further comprise hCG and/or LH.

hCG can be obtained by any means known in the art. hCG as used herein includes human-derived and recombinant hCG. Human-derived hCG can be purified from any appropriate source (e.g. urine, and placenta) by any method known in the art. Methods of expressing and purifying recombinant hCG are well known in the art.

LH can be obtained by any means known in the art. LH, as used herein, includes human-derived and recombinant LH. Human-derived LH can be purified from any appropriate source (e.g. urine) by any method known in the art. Methods of expressing and purifying recombinant LH are known in the art.

The pharmaceutical composition may be for the treatment of infertility, e.g. for use in e.g. assisted reproductive technologies (ART), ovulation induction or intrauterine insemination (IUI). The pharmaceutical composition may be used, for example, in medical indications where known FSH preparations are used. The present invention also provides the use of rFSH and/or an rFSH preparation described herein (according to aspects of the invention) for, or in the manufacture of a medicament for, the treatment of infertility. The pharmaceutical compositions of the present invention can be formulated into well-known compositions for any route of drug administration, e.g. oral, rectal, parenteral, transdermal (e.g. patch technology), intravenous, intramuscular, subcutaneous, intrasusternal, intravaginal, intraperitoneal, local (powders, ointments or drops) or as a buccal or nasal spray. A typical composition comprises a pharmaceutically acceptable carrier, such as aqueous solution, non toxic excipients, including salts and preservatives, buffers and the like, as described in Remington's Pharmaceutical Sciences fifteenth edition (Matt Publishing Company, 1975), at pages 1405 to 1412 and 1461 - 87, and the national formulary XIV fourteenth edition (American Pharmaceutical Association, 1975), among others.

Examples of suitable aqueous and non-aqueous pharmaceutical carriers, diluents, solvents or vehicles include water, ethanol, polyols (such as glycerol, propylene glycol, polyethylene glycol, and the like), carboxymethylcellulose and suitable mixtures thereof, vegetable oils (such as olive oil), and injectible organic esters such as ethyl oleate.

The compositions of the present invention also can contain additives such as but not limited to preservatives, wetting agents, emulsifying agents, and dispersing agents. Antibacterial and antifungal agents can be included to prevent growth of microbes and includes, for example, paraben, chlorobutanol, phenol, sorbic acid, and the like. Furthermore, it may be desirable to include isotonic agents such as sugars, sodium chloride, and the like.

In some cases, to effect prolonged action it is desirable to slow the absorption of FSH (and other active ingredients, if present) from subcutaneous or intramuscular injection. This can be accomplished by the use of a liquid suspension of crystalline or amorphous material with poor water solubility. The rate of absorption of FSH then depends upon its rate of dissolution which, in turn, can depend upon crystal size and crystalline form. Alternatively, delayed absorption of a parenterally administered FSH combination form is accomplished by dissolving or suspending the FSH combination in an oil vehicle.

Injectable depot forms can be made by forming microencapsule matrices of the FSH (and other agents, if present) in biodegradable polymers such as polylactide-polyglycolide. Depending upon the ratio of FSH to polymer and the nature of the particular polymer employed, the rate of FSH release can be controlled. Examples of other biodegradable polymers include polyvinylpyrrolidone, poly(orthoesters), poly(anhydrides) etc. Depot injectable formulations are also prepared by entrapping the FSH in liposomes or microemulsions which are compatible with body tissues.

Injectable formulations can be sterilized, for example, by filtration through a bacterial-retaining filter, or by incorporating sterilizing agents in the form of sterile solid compositions which can be dissolved or dispersed in sterile water or other sterile injectable medium just prior to use. Injectable formulations can be supplied in any suitable container, e.g. vial, pre-filled syringe, injection cartridges, and the like.

Injectable formulations can be supplied as a product having pharmaceutical compositions containing FSH (optionally with hCG, LH etc.) If there is more than one active ingredient (i.e. FSH and e.g. hCG or LH) these may be suitable for administration separately or together. If administered separately, administration can be sequential. The product can be supplied in any appropriate package. For example, a product can contain a number of pre-filled syringes containing either FSH, hCG, or a combination of both FSH and hCG, the syringes packaged in a blister package or other means to maintain sterility. A product can optionally contain instructions for using the FSH and hCG formulations.

The pH and exact concentration of the various components of the pharmaceutical composition are adjusted in accordance with routine practice in this field. See GOODMAN and GILMAN's THE PHARMACOLOGICAL BASIS FOR THERAPEUTICES, 7^th ed. In a preferred embodiment, the compositions of the invention are supplied as compositions for parenteral administration. General methods for the preparation of the parenteral formulations are known in the art and are described in REMINGTON; THE SCIENCE AND PRACTICE OF PHARMACY, supra, at pages 780-820. The parenteral compositions can be supplied in liquid formulation or as a solid which will be mixed with a sterile injectable medium just prior to administration. In an especially preferred embodiment, the parenteral compositions are supplied in dosage unit form for ease of administration and uniformity of dosage.

Detailed description of the invention

The present invention will now be described in more detail with reference to the following Examples and to the attached drawings in which:

Figure 1 shows a plasmid map of the pFSHalpha/beta expression vector; Figure 2 shows the α2,3-sialyltransferase (ST3GAL4) expression vector; Figure 3 shows the α2,6-sialyltransferase (ST6GAL1) expression vector;

Figure 4 shows Isoelectric focussing of recombinant FSH produced by

Per.C6 cells stably expressing FSH;

Figure 5 shows example of clones analysed by isoelectric focussing of recombinant FSH produced by Per.Cδ cells stably expressing FSH after engineering with α2,3- or α2,6-sialyltransferase;

Figure 6 shows analysis of the Sialic acid linkages of Per.C6 FSH;

Figure 7 shows metabolic clearance rates (MCRs) of Per.Cδ FSH samples;

Figure 8 shows MCRs of α2,6-sialytransferase engineered Per.Cβ FSH samples;

Figure 9 shows MCRs of α2,6-sialytransferase engineered Per.Cδ FSH samples;

Figure 10 shows MCRs of α2,3-sialytransferase engineered Per.Cδ FSH samples; Figure 11 shows ovarian weight augmentation by Per.Cδ rFSH clones of parental Per.Cδ rFSH, according to the method of Steelman and Pohley

(1953);

Figure 12 shows ovarian weight augmentation by Per.Cδ rFSH clones of engineered (α2,6-sialyltransferase) Per.Cδ rFSH; and Figure 13 shows ovarian weight augmentation by Per.Cδ rFSH clones of engineered (α2,3-sialyltransferase) Per.Cδ rFSH. Sequence Selection

Human FSH The coding region of the gene for the FSH alpha polypeptide was used to according to Fiddes and Goodman. (1981). The sequence is banked as AH007338 and at the time of construction there were no other variants of this protein sequence. The sequence is referred herein as SEQ ID 1.

The coding region of the gene for FSH beta polypeptide was used according to Keene et al (1989). The sequence is banked as NM_000510 and at the time of construction there were no other variants of this protein sequence. The sequence is referred herein as SEQ ID 2

Sialyltransferase

α2,3-Sialyltransferase - The coding region of the gene for beta-galactoside alpha-2,3-sialyltransferase 4 (α2,3-sialyltransferase, ST3GAL4) was used according to Kitagawa and Paulson (1994). The sequence is banked as L23767 and referred herein as SEQ ID 3.

α2,6-Sialyltransferase - The coding region of the gene for beta- galactosamide alpha-2,6-sialyltransferase 1 (α2,6-sialyltransferase, ST6GAL1) was used according to Grundmann et al. (1990). The sequence is banked as NM_003032 and referred herein as SEQ ID 4.

EXAMPLES

Example 1 Construction of the FSH expression vector

The coding sequence of FSH alpha polypeptide (AH007338, SEQ ID 1) and FSH beta polypeptide (NM_003032, SEQ ID 2) were amplified by PCR using the primer combinations FSHa-fw and FSHa-rev and FSHb-fw and FSHb-rec respectively.

FSHa-fw 5'-CCAGGATCCGCCACCATGGATTACTACAGAAAAATATGC-S' FSHa-rev 5'-GGATGGCTAGCTTAAGATTTGTGATAATAAC-3^I

FSHb-fw 5'-CCAGGCGCGCCACCATGAAGACACTCCAGTmTC-3' FSHb-rev 5'-CCGGGTTAACTTATTATTCTTTCATTTCACCAAAGG-S'

The resulting amplified FSH beta DNA was digested with the restriction enzymes Asc\ and Hpa\ and inserted into the Asc\ and Hpa\ sites on the CMV driven mammalian expression vector carrying a neomycin selection marker. Similarly the FSH alpha DNA was digested with SamHI and Nhe\ and inserted into the sites BamHI and Nhe\ on the expression vector already containing the FSH beta polypeptide DNA.

The vector DNA was used to transform the DH5α strain of E.coli. Sixty colonies were picked for amplification and fifty seven contained the vector containing both FSH alpha and beta. Twenty of these were selected for sequencing and all contained the correct sequences according to SEQ ID 1 and SEQ ID 2. Plasmid pFSH A+B#17 was selected for transfection (Figure 1).

Example 2 Construction of the ST3 expression vector

The coding sequence of beta-galactoside alpha-2,3-sialyltransferase 4 (ST3, L23767, SEQ ID 3) was amplified by PCR using the primer combination 2,3STfw and 2,3STrev.

2,3STfw 5¹-CCAGGATCCGCCACCATGTGTCCTGCAGGCTGGAAGC-3^I 2,3STrev 5'- 1 I I I I I I CTTAAGTCAGAAGGACGTGAGGTTCTTG-3' The resulting amplified ST3 DNA was digested with the restriction enzymes BamHI and AfIW and inserted into the SamHI and AfIW sites on the CMV driven mammalian expression vector carrying a hygromycin resistance marker. The vector was amplified as previously described and sequenced. Clone pST3#1 (Figure 2) contained the correct sequence according SEQ ID 3 and was selected for transfection.

Example 3 Construction of the ST6 expression vector

The coding sequence of beta-galactosamide alpha-2,6-sialyltransferase 1 (ST6, NM_003032, SEQ ID 4) was amplified by PCR using the primer combination 2,6STfw and 2,6STrev.

2,6STfw δ'-CCAGGATCCGCCACCATGATTCACACCAACCTGAAG-S' 2,6STrev 5'- l I I I I I I CTTAAGTTAGCAGTGAATGGTCCGG-3'

The resulting amplified ST6 DNA was digested with the restriction enzymes SamHI and AfIW and inserted into the SamHI and AfIW sites on the CMV driven mammalian expression vector carrying a hygromycin resistance marker. The vector was amplified as previously described and sequenced. Clone pST6#11 (Figure 3) contained the correct sequence according SEQ ID 4 and was selected for transfection.

Example 4 Stable expression of pFSH A+B in PER.C6 cells. Transfection isolation and screening of clones.

Per.C6 clones producing FSH were generated by expressing both polypeptide chains of FSH from a single plasmid (see Example 1). To obtain stable clones a liposome based transfection agent with the pFSH A+B construct. Stable clones were selected in VPRO supplemented with 10% FCS and containing G418. Three weeks after transfection G418 resistant clones grew out. A total of 250 clones were selected for isolation. The isolated clones were cultured in selection medium until 70-80% confluent. Supernatants were assayed for FSH protein content using an FSH selective ELISA and pharmacological activity at the FSH receptor in cloned cell line, using a cAMP accumulation assay. Clones (98) expressing functional protein were progressed for culture expansion to 24 well, 6 well and T80 flasks.

Studies to determine productivity and quality of the material from seven clones were initiated in T80 flasks to generate sufficient material. Cells were cultured in supplemented media as previously described for 7 days and the supernatant harvested. Productivity was determined using the FSH selective ELISA. The isoelectric profile of the material was determined (Example 6). Representative samples are shown in Figure 4. The information from the IEF was used to select clones for metabolic clearance rate analysis (Example 9). Clones (005, 104, 179, 223, 144) with sufficient productivity and quality were selected for sialyltransferase engineering.

Example 5 Level of sialylation is increased in cells that over express α2,3- or α2,6-sialyltransferase. Stable expression of pST3 or pST6 in FSH expressing PER.C6 cells; Transfection isolation and screening of clones.

Per.C6 clones producing highly sialylated FSH were generated by expressing α2,3 sialyltransferase or α2,6 sialyltransferase from separate plasmids (see Examples 2 and 3) in Per.C6 cells already expressing both polypeptide chains of FSH (see Example 4). Four clones produced from PER.C6® cells as set out in Example 4 were selected for their characteristics including productivity, good growth profile, production of functional protein, and produced FSH which included some sialylation.

Stable clones were generated as previously described in Example 4. A total of 202 clones from the α2,3-sialyltransferase program and 210 clones from the α2,6-sialyltransferase program were isolated, expanded and assayed. The final clone number for the α2,3- study was 12 and 30 for the α2,6- study.

The α2,3-sialyltransferase clones were adapted to serum free media and suspension conditions.

As before clones were assayed using a FSH selective ELISA, functional response in an FSH receptor cell line, IEF (Example 6), metabolic clearance rate (Example 9) and Steelman Pohley analysis (Example 10). Results were compared to a commercially available recombinant FSH (Gonal-f, Serono) and the parental FSH Per.C6 cell lines. Representative samples are shown in Figure 5. Some clones did not demonstrate an increase in sialylation but it can be seen that FSH produced by most of the clones has significantly improved sialylation (i.e. on average more FSH isoforms with high numbers of sialic acids) compared to FSH expressed without α2,3- or α2,6- sialyltransferase.

In conclusion expression of FSH together with sialyltransferase in Per.Cβ cells results in increased levels of sialylated FSH compared to cells expressing FSH only.

Example 6 Analysis of the pi of Per.Cβ produced FSH isoforms by isoelectric focussing.

Electrophoresis is defined as the transport of charged molecules through a solvent by an electrical field. The mobility of a biological molecule through an electric field will depend on the field strength, net charge on the molecule, size and shape of the molecule, ionic strength and properties of the medium through which the molecules migrate.

Isoelectric focusing (IEF) is an electrophoretic technique for the separation of proteins based on their pi. The pi is the pH at which a protein has no net charge and will not migrate in an electric field. The sialic acid content of the FSH isoforms subtly alters the pi point for each isoform, which can be exploited using this technique to visualise the Per.C6 FSH isoforms from each clone.

The isoelectric points of the Per.Cβ produced FSH isoforms in cell culture supernatants were analyzed using isoelectric focussing. Cell culture media from Per.C6 FSH clones was produced as described in Example 4 and 5.

Per.C6 FSH samples were separated on Novex® IEF Gels containing 5% polyacrylamide under native conditions on a pH 3.0 -7.0 gradient in an ampholyte solution pH 3.0 - 7.0.

Proteins were transferred onto supported nitrocellulose and visualised using a primary anti-FSHα monoclonal antibody, secondary anti-mouse IgG alkaline phosphatase conjugated antibody and 5-bromo-4-chloro-3- indolyl-phosphate (BCIP) and nitro blue tetrazolium (NBT) reagent to visualise the bands.

As indicated in Figures 4 and 5, the bands represent isoforms of FSH containing different numbers of sialic acid molecules.

Using this method clones producing FSH isoforms with a higher number of sialic acid molecules were identified. Engineering with α2,3- or α2,6- sialyltransferase resulted in clones with more sialic acid and a lower pi. Example 7 Analysis of the Sialic acid linkages of Per.C6 FSH

Glycoconjugates were analyzed using a lectin based glycan differentiation method. With this method glycoproteins and glycoconjuagtes bound to nitrocellulose can be characterized. Lectins selectively recognize a particular moiety, for example α2,3 linked sialic acid. The lectins applied are conjugated with the steroid hapten digoxigenin which enables immunological detection of the bound lectins.

Purified Per.C6 FSH from a parental clone (no additional sialyltransferase), a α2,3-sialyltransferase engineered clone and a α2,6- sialyltransferase engineered clone were separated using standard SDS- PAGE techniques. A commercially available recombinant FSH (Gonal-f, Serono) was used as a standard.

Sialic acid was analyzed using the DIG Glycan Differentiation Kit (Cat. No. 11 210 238 001, Roche) according to the manufacturers instructions. Positive reactions with Sambucus nigra agglutinin (SNA) indicated terminally linked (2-6) sialic acid. Positive reactions with Maackia amurensis agglutinin Il (MAA): indicated terminally linked (α2-3) sialic acid

In summary the parental clone 005 contained low levels of both α2,3- and α2,6- sialic acid. The clones engineered with α2,3-sialyltransferase contained high levels of α2,3- sialic acid linkages and low levels of α2,6- sialic acid linkages. Clones engineered with α2,6-sialyltransferase contained high levels of α2,6- sialic acid linkages and low levels of α2,3- sialic acid linkages. The standard control Gonal-f only contains α2,3- sialic acid linkages. This is consistent with what is known about recombinant proteins produced in Chinese Hamster ovary (CHO) cells (Kagawa ef al, 1988, Takeuchi ef al, 1988, Svensson ef al., 1990). In conclusion, engineering of Per.Cβ FSH cells with α2,3- or α2,6- sialyltransferase successfully increased the number of sialic acid molecules conjugated to the FSH in the sample.

Example 8a Quantification of total Sialic acid

Sialic acid is a protein-bound carbohydrate considered to be a monosaccharide and occurs in combination with other mono- saccharides like galactose, mannose, glucosamine, galactosamine and fucose.

The total sialic acid on purified rFSH (Example 11) was measured using an enzymatic sialic acid quantification kit according to the manufacturers protocol (Sigma, Sialic-Q). In short N-acetylneuraminic acid aldolase catalyses sialic acid to N-acetylmannoasine and pyruvic acid. The pyruvic acid can be reduced to lactic acid by β-NADH and lactic dehydrogenase. B-NADH oxidation can be accurately measured spectrophotometrically.

Protein concentration was measured in microtiter plates using a commercial bicinchoninic acid (BCA) assay kit (Sigma, B 9643) based on the Lowry method (Lowry et al, 1951).

The total sialic acid content of Per.Cβ FSH was measured and found to be greater than 6 mol/mol.

Example 8b Quantification of relative amounts of α2,3, α2,6 and α2,8 sialic acid

The relative percentage amounts of α2,3, α2,6 and α2,8 sialic acid on purified rFSH (Example 11) were measured using known techniques.

Each sample of rFSH was immobilized (gel block), washed, reduced, alkylated and digested with PNGase F overnight. The N-glycans were then extracted and processed. N-glycans for NP-HPLC and WAX-HPLC analysis were labelled with the fluorophore 2AB as detailed in Royle et al. The N-glycans were run on normal phase (NP) HPLC on a TSK amide column (as detailed in Royle et al) with retention times expressed in glucose units (GU).

Samples of the extracted, pooled, glycans (extracted as above) were digested with different sialidases to determine the linkages. NAN 1 (recombinant sialidase) releases α2,3 linked non-reducing terminal sialic acids (NeuNAc and NeuNGc), ABS (Arthrobacter ureafaciens sialidase) releases α2,3, α2,6 and α2,8 linked non-reducing terminal sialic acids (NeuNAc and NeuNGc). Samples were analysed by NP-HPLC, to allow comparison of the undigested sample with that digested with NAN1 and that digested with ABS. Comparison of the three NP-HPLC traces (undigested, NAN1 digested, ABS digested) shows that digestion with ABS and NAN1 give different results. This indicates that the samples have sialic acids with α2,3, α2,6 and α2,8 linkages. The relative percentages were calculated from structures present in the undigested glycan pools and were found to be in the ranges 65% - 85% (e.g. 77.75%) for α2,3 sialylation; 15 to 35% (e.g. 21.46%) for α2,6 sialylation; and 0.1 to 3% for α2,8 sialylation.

Example 8c Quantification of relative amounts mono, di, tri and tetra antennary sialylated structures

The relative percentage amounts of mono, di, tri and tetra sialylated structures on glycans extracted from purified rFSH (Example 11) were measured using known techniques.

Each sample of rFSH was immobilized (gel block), washed, reduced, alkylated and digested with PNGase F overnight. The N-glycans were then extracted and processed. N-glycans for NP-HPLC and WAX-HPLC analysis were labelled with the fluorophore 2AB as detailed in Royle et al. Weak anion exchange (WAX) HPLC to separate the N-glycans by charge (Example 8b) was carried out as set out in Royle et al, with a Fetuin N- glycan standard as reference. Glycans were eluted according to the number of sialic acids they contained. All samples included mono (1S)₁ di(2S), tri(3S) and tetra(4S) sialylated structures. The relative amounts of sialylated structures were found to be in the following ratios (1S:2S:4S:4S): 9-15%: 27-30%: 30-36%: 25-29 % (for example 10.24:28.65:35.49:25.62).

Example 9 Determination of the metabolic clearance rates of rFSH

To determine the metabolic clearance rate (MCR) of Per.C6 FSH samples conscious female rats (3 animals per clone) were injected into the tail vein at time zero with a bolus of rFSH (1 - 10 μg/rat, based on ELISA quantification of samples, DRG EIA 1288). Blood samples (400 μl) were taken from the tip of the tail at 1 , 2, 4, 8, 12, 24 and 32 hours after test sample injection. Serum was collected by centrifugation and assayed for FSH content by ELISA (DRG EIA 1288).

The asialoglycoprotein receptor (ASGP-R) recognizes desialyated (galactose-terminated) glycoproteins such as asialofetuin (ASF). (Pricer and Ashwell, 1971. Van Lenten and Ashwell, 1972). The ASGP receptor and the bound desialyated glycoprotein are internalized into the cell where the receptor is recycled and the ligand is degraded (Regoeczi et al, 1978, Steer and Ashwell, 1980).

To investigate if Per.C6 FSH material was cleared via this mechanism, the ASGP-R was saturated with asialofetuin. The metabolic clearance rate of parental, α2,6 or α2,3-sialyltransferase engineered material was determined as described with co administration of a minimum 7500-fold molar excess of asialofetuin to saturate the ASGP-R for 1-2 h. The material produced by the parental Per.Cβ FSH clones contained some longer MCR material but a high percentage was cleared quickly (Figure 7). The lead clone 005 which contained the most sialylated material was engineered using α2,6- or α2,3- sialyltransferase (Example 5). Although the clones engineered with α2,6-sialyltransferase demonstrated increased sialylation (Figure 5) there was no improvement in the MCR (Figure 7). Blockade of the ASGR restored the MCR of the α2,6 material to that of the standard demonstrating that even with increased α2,6 linkages the material is cleared quickly (Figure 8). Engineering with α2,3- sialyltransferase resulted in clones with comparable MCR to the standard (Figure 9) and varying sialic content was consistent with what is known for the isoforms of FSH (Figure 10).

Example 10 Steelman-Pohley in vivo assay

To demonstrate increasing sialic acid content on FSH results in an increased biological effect, the increase in ovarian weights in rats by highly sialylated FSH such as produced in Example 5 was examined.

The increase in ovarian weights due to the Per.CΘ rFSH clones were analysed according to the method of Steelman and Pohley (1953). Per.C6 rFSH from filtered cell media samples was quantified by ELISA (DRG₁EIA- 1288). The samples (Per.C6 rFSH) and standards (Gonal-f rFSH) were tested at five different doses (3 animals/dose). Gonal-f was dosed at 50, 100, 200, 400, and 800 ng/rat, The sample doses were calculated using their AUC values relative to Gonal-f, typically 0.05 - 10 μg/rat.

In conclusion, the undersialylated material produced by the parental Per.CΘ FSH clones (Figure 11) was not as potent in the ovarian weight augmentation assay as the commercially available rFSH. Sialyltransferase engineering to add additional α2,6- linkages increased the sialic acid content but did not improve potency in the in vivo assay (Figure 12). However, additional α2,3- linkages significantly improved potency (Figure 13) and the two recombinant FSH preparations (Per.Cβ and CHO-derived) display very similar profiles in this assay.

Example 11 Production and purification overview

A procedure was developed to produce FSH in PER.C6 cells that were cultured in suspension in serum free medium. The procedure is described below and was applied to several FSH-producing PER.C6 cell lines.

FSH from the parental clone 005, α2,3- clone 007 and α2,6 clone 059 was prepared using a using a modification of the method described by Lowry et al. (1976).

For the production of PER.C6-FSH, the cell lines were adapted to a serum- free medium, i.e., Excell 525 (JRH Biosciences). The cells were first cultured to form a 70%-90% confluent monolayer in a T80 culture flask. On passage the cells were re-suspended in the serum free medium, Excell 525 + 4 mM L-Glutamine, to a cell density of 0.3x10⁶ cells/ml. A 25 ml cell suspension was put in a 250 ml shaker flask and shaken at 100 rpm at 37⁰C at 5% CO₂. After reaching a cell density of > IxIO⁶ cells/ml, the cells were sub-cultured to a cell density of 0.2 or 0.3x10⁶ cells/ml and further cultured in shaker flasks at 37°C, 5% CO₂ and 100 rpm.

For the production of FSH, the cells were transferred to a serum- free production medium, i.e., VPRO (JRH Biosciences), which supports the growth of PER.C6 cells to very high cell densities (usually > 10⁷ cells/ml in a batch culture). The cells were first cultured to > IxIO⁶ cells/ml in Excell 525, then spun down for 5 min at 1000 rpm and subsequently suspended in VPRO medium + 6 mM L-glutamine to a density of 1x10⁶ cells/ml. The cells were then cultured in a shaker flask for 7-10 days at 37°C, 5% CO2 and 100 rpm. During this period, the cells grew to a density of > 10⁷ cells/ml. The culture medium was harvested after the cell viability started to decline. The cells were spun down for 5 min at 1000 rpm and the supernatant was used for the quantification and purification of FSH. The concentration of FSH was determined using ELISA (DRG EIA 1288).

Thereafter, purification of FSH was carried out using a modification of the method described by Lowry et a/. (1976). This was achieved by chromatography on DEAE cellulose, gel filtration on Sephadex G100, adsorption chromatography on hydroxyapatite, and preparative polyacrylamide electrophoresis.

During all chromatographic procedures, the presence of immunoreactive FSH was confirmed by RIA (DRG EIA 1288) and IEF (Example 6).

References

Andersen CY, Westergaard LG, and van WeIy M. (2004). FSH isoform composition of commercial gonadotrophin preparations: a neglected aspect? Reprod Biomed Online. 912), 231-236.

Arey BJ, Stevis PE, Deecher DC, Shen ES, Frail DE, Negro-Vilar A, and Lopez FJ. (1997) Induction of promiscuous G protein coupling of the follicle-stimulating hormone (FSH) receptor: a novel mechanism for transducing pleiotropic actions of FSH isoforms. MoI Endocrinol. 11(5), 517-526.

Baenziger JU and Green ED. (1988). Pituitary glycoprotein hormone oligosaccharides: structure, synthesis and function of the asparagine- linked oligosaccharides on lutropin, follitropin and thyrotropin. Biochim Biophys Acta. 947(2). 287-306.

Bassett RM, and Driebergen R. (2005). Continued improvements in the quality and consistency of follitropin alfa, recombinant human FSH. Reprod Biomed Online. 10(2). 169-177.

Damian-Matsumura P, Zaga V, Maldonado A, Sanchez-Hernandez C, Timossi C, and Ulloa-Aguirre A. (1999). Oestrogens regulate pituitary alpha2,3-sialyltransferase messenger ribonucleic acid levels in the female rat. J MoI Endocrinol. 23(2), 153-165.

D'Antonio M., Borrelli F. , Datola A., Bucci R. , Mascia M. , Polletta P., Piscitelli D., and Papoian R. (1999) Biological characterization of recombinant human follicle stimulating hormone isoforms. Human Reproduction 14, 1160-1167

Dalpathado DS, lrungu J, Go EP₁ Butnev VY, Norton K, Bousfield GR, and Desaire H. (2006). Comparative glycomics of the glycoprotein follicle stimulating hormone: glycopeptide analysis of isolates from two mammalian species. Biochemistry. 45(28). 8665-8673. No copy

Dias JA, Van Roey P. (2001). Structural biology of human follitropin and its receptor. Arch Med Res. 32(6). 510-519

Fiddes, J. C. and Goodman, H. M. (1979) Isolation, cloning and sequence analysis of the cDNA for the alpha-subunit of human chorionic gonadotropin. Nature, 281 , 351-356.

Flack, M.R., Bennet, A.P., Froehlich, J. Anasti, JN and Nisula, B. (1994). Increased biological activity due to basic isoforms in recombinant human follicle-stimulating hormone produced in a human cell line. J. Clin. Endocrinol. Metab., 79, 756-760

Fox KM, Dias JA, and Van Roey P. (2001). Three-dimensional structure of human follicle-stimulating hormone. MoI Endocrinol. 15(3), 378-89

Grabenhorst E, Hoffmann A, Nimtz M, Zettlmeissl G, and Conradt HS. (1995). Construction of stable BHK-21 cells coexpressing human secretory glycoproteins and human Gal(beta 1-4)GlcNAc-R alpha 2,6- sialyltransferase alpha 2,6-linked NeuAc is preferentially attached to the Gal(beta 1-4)GlcNAc(beta 1-2)Man(alpha 1-3)-branch of diantennary oligosaccharides from secreted recombinant beta-trace protein. Eur J Biochem. 232(3), 718-25.

Green ED and Baenziger JU. (1988). Asparagine-linked oligosaccharides on lutropin, follitropin, and thyrotropin. II. Distributions of sulfated and sialylated oligosaccharides on bovine, ovine, and human pituitary glycoprotein hormones. J Biol Chem. 263(1). 36-44. Grundmann.U., Nerlich.C, Rein.T. and Zettlmeissl, G. (1990). Complete cDNA sequence encoding human beta-galactoside alpha-2,6- sialyltransferase. G Nucleic Acids Res. 18 (3), 667

Howies, CM. (1996). Genetic engineering of human FSH (Gonal-F). Hum Reprod. Update, 2,172-191.

Kagawa Y₁ Takasaki S, Utsumi J, Hosoi K, Shimizu H, Kochibe N, and Kobata A. (1988). Comparative study of the asparagine-linked sugar chains of natural human interferon-beta 1 and recombinant human interferon-beta 1 produced by three different mammalian cells. J Biol Chem. 263(33), 17508-17515.

Keene, J. L., Matzuk, M. M., Otani, T., Fauser, B₁C₁J₁M., Galway, A.B., Hsueh, A.J.W. and Boime, I. (1989). Expression of Biologically active Human Follitropin in Chinese Hamster Ovary Cells. The Journal of Biological Chemistry, 264(9). 4769-4775.

Kitagawa.H. and Paulson.J.C (1994) Cloning of a novel alpha 2,3- sialyltransferase that sialylates glycoprotein and glycolipid carbohydrate groups. J. Biol. Chem. 269(2). 1394-1401.

Lee EU, Roth J, and Paulson JC (1989) Alteration of terminal glycosylation sequences on N-linked oligosaccharides of Chinese hamster ovary cells by expression of beta-galactoside alpha 2,6-sialyltransferase. J Biol Chem. 264(23). 13848-13855.

de Leeuw, R., Mulders, J., Voortman, G. Rombout, F. Damm, J. and Kloosterboer, L. (1996) Structure-function relationship of recombinant follicle stimulating hormone (Puregon). MoI. Hum. Reprod., 2, 361-369. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. (1951) Protein measurement with the Folin phenol reagent. J Biol Chem. 193(1). 265-75.

Lowry, PJ, McLean, C, Jones RL and Satgunasingam N. (1976) Purification of anterior pituitary and hypothalamic hormones Clin Pathol Suppl (Assoc Clin Pathol). 7, 16-21.

Pierce JG, and Parsons TF (1981) Glycoprotein hormones: structure and function Annu Rev Biochem. 5O₁465-495.

Pricer WE Jr, and Ashwell G. (1971). The binding of desialylated glycoproteins by plasma membranes of rat liver. J Biol Chem. 246(15), 4825-33.

Rathnam P, and Saxena BB. (1975). Primary amino acid sequence of follicle-stimulating hormone from human pituitary glands. I. alpha subunit. J Biol Chem.;250(17):6735-6746.

Regoeczi E, Debanne MT, Hatton MC, and Koj A. (1978) Elimination of asialofetuin and asialoorosomucoid by the intact rat. Quantitative aspects of the hepatic clearance mechanism. Biochim Biophys Acta. 541(3), 372- 84.

Royle L, Radcliffe CM, Dwek RA and Rudd PM (2006) Methods in Molecular Biology, ed I Brockhausen-Schutzbach (Humana Press), 347: Glycobiology protocols, 125-144.

Ryan RJ, Keutmann HT, Charlesworth MC, McCormick DJ, Milius RP, Calvo FO and Vutyavanich T. (1987). Structure-function relationships of gonadotropins. Recent Prog Horm Res.;43,:383-429. Saxena BB and Rathnam P. (1976) Amino acid sequence of the beta subunit of follicle-stimulating hormone from human pituitary glands. J Biol Chem. 251 (4). 993-1005

Steelman SL, and Pohley FM. (1953) Assay of the follicle stimulating hormone based on the augmentation with human chorionic gonadotropin. Endocrinology. 53(6), 604-616.

Steer CJ, and Ashwell G. (1980) Studies on a mammalian hepatic binding protein specific for asialoglycoproteins. Evidence for receptor recycling in isolated rat hepatocytes. J Biol Chem. 255(7), 3008-13.

Svensson EC, Soreghan B, and Paulson JC. (1990) Organization of the beta-galactoside alpha 2,6-sialyltransferase gene. Evidence for the transcriptional regulation of terminal glycosylation. J Biol Chem. 265(34):20863-20868.

Takeuchi M, Takasaki S, Miyazaki H, Kato T, Hoshi S, Kochibe N, and Kobata A (1988). Comparative study of the asparagine-linked sugar chains of human erythropoietins purified from urine and the culture medium of recombinant Chinese hamster ovary cells. J Biol Chem. 263(8), 3657- 3663.

Timossi CM, Barrios de Tomasi J, Zambrano E, Gonzalez R, Ulloa-Aguirre A. (1998). A naturally occurring basically charged human follicle- stimulating hormone (FSH) variant inhibits FSH-induced androgen aromatization and tissue-type plasminogen activator enzyme activity in vitro. Neuroendocrinology. 67(3), 153-163.

Timossi CM, Barrios-de-Tomasi J, Gonzalez-Suarez R, Arranz MC, Padmanabhan V, Conn PM, and Ulloa-Aguirre A. (2000). Differential effects of the charge variants of human follicle-stimulating hormone. J Endocrinol. 165(2), 193-205.

Ulloa-Aguirre, A., Espinoza, R., Damian-Matsumura, P. and Chappel, S. C. (1988) Immunological and biological potencies of the different molecular species of gonadotrophins. Hum. Reprod. 3, 491-501.

Ulloa-Aguirre, A., Cravioto, A., Damian-Matsumura, P. Jimenez, M, Zambrano, E and Diaz-Sanchez, V. (1992) Biological characterization of the naturally occurring analogues of intrapituitary human follicle stimulating hormone. Hum. Reprod. 7, 23-30.

Ulloa-Aguirre A, Midgley AR Jr, Beitins IZ, and Padmanabhan V. (1995). Follicle-stimulating isohormones: characterization and physiological relevance. Endocr Rev.16(6), 765-787.

Ulloa-Aguirre A, Maldonado A, Damian-Matsumura P, and Timossi C (2001). Endocrine regulation of gonadotropin glycosylation. Arch Med Res. 32(6), 520-532.

Ulloa-Aguirre A, Timossi C, Barrios-de-Tomasi J, Maldonado A, and Nayudu P. (2003). Impact of carbohydrate heterogeneity in function of follicle-stimulating hormone: studies derived from in vitro and in vivo models. Biol Reprod. 69(2), 379-389.

Van Lenten L, and Ashwell G. (1972) The binding of desialylated glycoproteins by plasma membranes of rat liver. Development of a quantitative inhibition assay. J Biol Chem. 247(14), 4633-40.

Wide, L. and Albertsson-Wikland, K. (1990) Change in electrophoretic mobility of human follicle-stimulating hormone in serum after administration of gonadotropin-releasing hormone. J. Clin. Endocrinol. Metab. 7O₁ 271-276. Wide, L. and Bakos, O. (1993). More basic forms of both human follicle- stimulating hormone and luteinizing hormone in serum at midcycle compared with the follicular or luteal phase. J. Clin. Endocrinol. Metab., 76, 885-889.

Wide L, Naessen T, Sundstrom-Poromaa I, Eriksson K. (2007) Sulfonation and sialylation of gonadotropins in women during the menstrual cycle, after menopause, and with polycystic ovarian syndrome and in men. J Clin Endocrinol Metab.;92(11). 4410-4417.

Zambrano E, Zariήan T, Olivares A, Barrios-de-Tomasi J₁ and Ulloa- Aguirre A. (1999). Receptor binding activity and in vitro biological activity of the human FSH charge isoforms as disclosed by heterologous and homologous assay systems: implications for the structure-function relationship of the FSH variants. Endocrine. 10(2). 113-121.

Zhang X, Lok SH, and Kon OL (1998) Stable expression of human alpha- 2,6-sialyltransferase in Chinese hamster ovary cells: functional consequences for human erythropoietin expression and bioactivity. Biochim Biophys Acta. 1425(3). 441-452.

SEQ ID 1 Follicle stimulating hormone alpha polypeptide Accession number AH007338

Nucleotide sequence of FSH alpha

1 ATGGATTACT ACAGAAAATA TGCAGCTATC TTTCTGGTCA CATTGTCGGT GTTTCTGCAT

61 GTTCTCCATT CCGCTCCTGA TGTGCAGGAT TGCCCAGAAT GCACGCTACA GGAAAACCCA 121 TTCTTCTCCC AGCCGGGTGC CCCAATACTT CAGTGCATGG GCTGCTGCTT CTCTAGAGCA

181 TATCCCACTC CACTAAGGTC CAAGAAGACG ATGTTGGTCC AAAAGAACGT CACCTCAGAG 241 TCCACTTGCT GTGTAGCTAA ATCATATAAC AGGGTCACAG TAATGGGGGG TTTCAAAGTG

301 GAGAACCACA CGGCGTGCCA CTGCAGTACT TGTTATTATC ACAAATCTTA A

Protein sequence of FSH alpha i MKTLQFFFLF CCWKAICCNS CELTNITIAI EKEECRFCIS INTTWCAGYC

YTRDLVYKDP

61 ARPKIQKTCT FKELVYETVR VPGCAHHADS LYTYPVATQC HCGKCDSDST DCTVRGLGPS 121 YCSFGEMKE

SEQ ID 2

Follicle stimulating hormone beta polypeptide Accession number NM_000510 Nucleotide sequence of FSH beta

1 ATGAAGACAC TCCAGTTTTT CTTCCTTTTC TGTTGCTGGA AAGCAATCTG CTGCAATAGC

61 TGTGAGCTGA CCAACATCAC CATTGCAATA GAGAAAGAAG AATGTCGTTT CTGCATAAGC 121 ATCAACACCA CTTGGTGTGC TGGCTACTGC TACACCAGGG ATCTGGTGTA TAAGGACCCA

181 GCCAGGCCCA AAATCCAGAA AACATGTACC TTCAAGGAAC TGGTATATGA AACAGTGAGA 241 GTGCCCGGCT GTGCTCACCA TGCAGATTCC TTGTATACAT ACCCAGTGGC CACCCAGTGT

301 CACTGTGGCA AGTGTGACAG CGACAGCACT GATTGTACTG TGCGAGGCCT GGGGCCCAGC 361 TACTGCTCCT TTGGTGAAAT GAAAGAATAA

Protein sequence of FSH beta

1 MKTLQFFFLF CCWKAICCNS CELTNITIAI EKEECRFCIS INTTWCAGYC YTRDLVYKDP 61 ARPKIQKTCT FKELVYETVR VPGCAHHADS LYTYPVATQC HCGKCDSDST DCTVRGLGPS 121 YCSFGEMKE

SEQ ID 3

Beta-galactoside alpha-2,3-sialyltransferase 4

Accession Number L23767 Nucleotide sequence of ST3GAL4

1 ATGTGTCCTG CAGGCTGGAA GCTCCTGGCC ATGTTGGCTC TGGTCCTGGT CGTCATGGTG 61 TGGTATTCCA TCTCCCGGGA AGACAGGTAC ATCGAGCTTT TTTATTTTCC CATcccAGAG

121 AAGAAGGAGC CGTGCCTCCA GGGTGAGGCA GAGAGCAAGG CCTCTAAGCT CTTTGGCAAC

181 TACTCCCGGG ATCAGCCCAT CTTCCTGCGG CTTGAGGATT ATTTCTGGGT CAAGACGCCA 241 TCTGCTTACG AGCTGCCCTA TGGGACCAAG GGGAGTGAGG ATCTGCTCCT CCGGGTGCTA

301 GCCATCACCA GCTCCTCCAT CCCCAAGAAC ATCCAGAGCC TCAGGTGCCG CCGCTGTGTG 361 GTCGTGGGGA ACGGGCACCG GCTGCGGAAC AGCTCACTGG GAGATGCCAT CAACAAGTAC 421 GATGTGGTCA TCAGATTGAA CAATGCCCCA GTGGCTGGCT ATGAGGGTGA

CGTGGGCTCC

481 AAGACCACCA TGCGTCTCTT CTACCCTGAA TCTGCCCACT TCGACCCCAA

AGTAGAAAAC 541 AACCCAGACA CACTCCTCGT CCTGGTAGCT TTCAAGGCAA TGGACTTCCA

CTGGATTGAG

601 ACCATCCTGA GTGATAAGAA GCGGGTGCGA AAGGGTTTCT GGAAACAGCC

TCCCCTCATC

661 TGGGATGTCA ATCCTAAACA GATTCGGATT CTCAACCCCT TCTTCATGGA GATTGCAGCT

721 GACAAACTGC TGAGCCTGCC AATGCAACAG CCACGGAAGA TTAAGCAGAA

GCCCACCACG

781 GGCCTGTTGG CCATCACGCT GGCCCTCCAC CTCTGTGACT TGGTGCACAT

TGCCGGCTTT 841 GGCTACCCAG ACGCCTACAA CAAGAAGCAG ACCATTCACT ACTATGAGCA

GATCACGCTC

901 AAGTCCATGG CGGGGTCAGG CCATAATGTC TCCCAAGAGG CCCTGGCCAT

TAAGCGGATG

961 CTGGAGATGG GAGCTATCAA GAACCTCACG TCCTTCTGA

Protein Sequence of ST3GAL4

1 MCPAGWKLLA MLALVLVVMV WYSISREDRY IELFYFPIPE KKEPCLQGEA ESKASKLFGN 61 YSRDQPIFLR LEDYFWVKTP SAYELPYGTK GSEDLLLRVL AITSSSIPKN IQSLRCRRCV

121 VVGNGHRLRN SSLGDAINKY DWIRLNNAP VAGYEGDVGS KTTMRLFYPE SAHFDPKVEN 181 NPDTLLVLVA FKAMDFHWIE TILSDKKRVR KGFWKQPPLI WDVNPKQIRI LNPFFMEiAA

241 DKLLSLPMQQ PRKIKQKPTT GLLAITLALH LCDLVHIAGF GYPDAYNKKQ

TIHYYEQITL

301 KSMAGSGHNV SQEALAIKRM LEMGAIKNLT SF

SEQ ID 4

Beta-galactosamide alpha-2,6-sialyltransferase 1 Accession number NM_003032 Nucleotide sequence of ST6GAL1

1 ATGATTCACA CCAACCTGAA GAAAAAGTTC AGCTGCTGCG TCCTGGTCTT TCTTCTGTTT

61 GCAGTCATCT GTGTGTGGAA GGAAAAGAAG AAAGGGAGTT ACTATGATTC CTTTAAATTG 121 CAAACCAAGG AATTCCAGGT GTTAAAGAGT CTGGGGAAAT TGGCCATGGG GTCTGATTCC

181 CAGTCTGTAT CCTCAAGCAG CACCCAGGAC CCCCACAGGG GCCGCCAGAC CCTCGGCAGT 241 CTCAGAGGCC TAGCCAAGGC CAAACCAGAG GCCTCCTTCC AGGTGTGGAA CAAGGACAGC

301 TCTTCCAAAA ACCTTATCCC TAGGCTGCAA AAGATCTGGA AGAATTACCT AAGCATGAAC

361 AAGTACAAAG TGTCCTACAA GGGGCCAGGA CCAGGCATCA AGTTCAGTGC AGAGGCCCTG 421 CGCTGCCACC TCCGGGACCA TGTGAATGTA TCCATGGTAG AGGTCACAGA TTTTCCCTTC

481 AATACCTCTG AATGGGAGGG TTATCTGCCC AAGGAGAGCA TTAGGACCAA GGCTGGGCCT

541 TGGGGCAGGT GTGCTGTTGT GTCGTCAGCG GGATCTCTGA AGTCCTCCCA ACTAGGCAGA

601 GAAATCGATG ATCATGACGC AGTCCTGAGG TTTAATGGGG CACCCACAGC CAACTTCCAA

661 CAAGATGTGG GCACAAAAAC TACCATTCGC CTGATGAACT CTCAGTTGGT TACCACAGAG 721 AAGCGCTTCC TCAAAGACAG TTTGTACAAT GAAGGAATCC TAATTGTATG GGACCCATCT

781 GTATACCACT CAGATATCCC AAAGTGGTAC CAGAATCCGG ATTATAATTT CTTTAACAAC

841 TACAAGACTT ATCGTAAGCT GCACCCCAAT CAGCCCTTTT ACATCCTCAA GCcccAGATG

901 CCTTGGGAGC TATGGGACAT TCTTCAAGAA ATCTCCCCAG AAGAGATTCA GCCAAACCCC

961 CCATCCTCTG GGATGCTTGG TATCATCATC ATGATGACGC TGTGTGACCA GGTGGATATT 1021 TATGAGTTCC TCCCATCCAA GCGCAAGACT GACGTGTGCT ACTACTACCA GAAGTTCTTC

1081 GATAGTGCCT GCACGATGGG TGCCTACCAC CCGCTGCTCT ATGAGAAGAA TTTGGTGAAG 1141 CATCTCAACC AGGGCACAGA TGAGGACATC TACCTGCTTG GAAAAGCCAC ACTGCCTGGC 1201 TTCCGGACCA TTCACTGCTA A

Op- Protein Sequence of ST6GAL1

1 MIHTNLKKKF SCCVLVFLLF AVICVWKEKK KGSYYDSFKL QTKEFQVLKS LGKLAMGSDS

61 QSVSSSSTQD PHRGRQTLGS LRGLAKAKPE ASFQVWNKDS SSKNLIPRLQ KIWKNYLSMN 121 KYKVSYKGPG PGIKFSAEAL RCHLRDHVNV SMVEVTDFPF NTSEWEGYLP KESIRTKAGP 181 WGRCAVVSSA GSLKSSQLGR EIDDHDAVLR FNGAPTANFQ QDVGTKTTIR

LMNSQLVTTE

241 KRFLKDSLYN EGILIVWDPS VYHSDIPKWY QNPDYNFFNN YKTYRKLHPN QPFYiLKPQM

301 PWELWDILQE ISPEEIQPNP PSSGMLGIII MMTLCDQVDI YEFLPSKRKT

DVCYYYQKFF

361 DSACTMGAYH PLLYEKNLVK HLNQGTDEDI YLLGKATLPG FRTIHC

Claims

1. Recombinant FSH (rFSH) including α2,3- and α2,6-sialylation.

2. Recombinant FSH according to claim 1 having a sialic acid content [expressed in terms of a ratio of moles of sialic acid to moles of protein] of 6 mol/mol or greater.

3. Recombinant FSH according to claim 1 or claim 2 having a sialic acid content of between 6 mol/mol and 15 mol/mol.

4. Recombinant FSH according to any preceding claim wherein 10% or more of the total sialylation is α2,3-sialylation.

5. Recombinant FSH according to any preceding claim which includes α2,3-sialylation in an amount which is from 65 to 85% of the total sialylation.

6. Recombinant FSH according to any preceding claim which includes α2,3-sialylation in an amount which is from 70 to 80% of the total sialylation.

7. Recombinant FSH according to any preceding claim wherein 50% or less of the total sialylation is α2,6-sialylation.

8. Recombinant FSH according to any preceding claim which includes α2,6-sialylation in an amount which is from 15 to 35% of the total sialylation.

9. Recombinant FSH according to any preceding claim which includes α2,6-sialylation in an amount which is from 20 to 30% of the total sialylation,

10. Recombinant FSH according to any preceding claim which further includes α2,8-sialylation.

11. Recombinant FSH according to any preceding claim wherein the sialic acid content is 6% or greater by mass.

12. Recombinant FSH according to any preceding claim produced or expressed in a human cell line.

13. Recombinant FSH according to any preceding claim produced or expressed in a Per.C6 cell line, a Per.Cδ derived cell line or a modified Per.Cβ cell line.

14. Recombinant FSH according to claim 12 or 13 wherein the cell line has been modified using α2,3-sialyltransferase.

15. Recombinant FSH according to any of claims 12 to 14 which includes α2,6-linked sialic acids (α2,6 sialylation) provided by endogenous sialyl transferase activity.

16. Recombinant FSH expressed in a human cell line.

17. Recombinant FSH according to claim 16 wherein 10% or more of the total sialylation is α2,3-sialylation.

18. Recombinant FSH according to claim 16 or 17 which includes α2,3- sialylation in an amount which is from 65 to 85% of the total sialylation.

19. Recombinant FSH according to any of claims 16 to 18 wherein 50% or less of the total sialylation is α2,6-sialylation.

20. Recombinant FSH according to any of claims 16 to 19 which includes α2,6-sialylation in an amount which is from 15 to 35% of the total sialylation.

21. Recombinant FSH according to any of claims 16 to 20 which includes a sialic acid content [expressed in terms of a ratio of moles of sialic acid to moles of protein] of 6 mol/mol or greater.

22. Recombinant FSH according to any of claim 16 or 21 including α2,3- sialylation and α2,6-sialylation.

23. A recombinant FSH preparation including α2,3- and α2,6-sialylation.

24. A preparation according to claim 23 which is a pharmaceutical preparation.

25. A preparation according to claim 23 or 24 having a sialic acid content [expressed in terms of a ratio of moles of sialic acid to moles of protein] of 6 mol/mol or greater.

26. A preparation according to any of claims 23 to 25 wherein 10% or more of the total sialylation is α2,3-sialylation.

27. A preparation according to any of claims 23 to 26 which includes α2,3- sialylation in an amount which is from 65 to 85% of the total sialylation.

28. A preparation according to any of claims 23 to 27 wherein 50% or less of the total sialylation is α2,6-sialylation.

29. A preparation according to any of claims 23 to 28 which includes α2,6- sialylation in an amount which is from 15 to 35% of the total sialylation.

30. A preparation according to any of claims 23 to 29 produced or expressed in a human cell line.

31. A pharmaceutical composition comprising rFSH including α2,3- sialylation and α2,6-sialylation.

32. A pharmaceutical composition according to claim 31 wherein 10% or more of the total sialylation is α2,3-sialylation.

33. A pharmaceutical composition according to claims 31 or 32 which includes α2,3-sialylation in an amount which is from 65 to 85% of the total sialylation.

34. A pharmaceutical composition according to any of claims 31 to 33 wherein 50% or less of the total sialylation is α2,6-sialylation.

35. A pharmaceutical composition according to any of claims 31 to 33 which includes α2,6-sialylation in an amount which is from 15 to 35% of the total sialylation.

36. A pharmaceutical composition comprising rFSH according to any of claims 1 to 22 and/or a preparation according to any of claims 23 to 30.

37. A pharmaceutical composition according to any of claims 31 to 36 further comprising hCG and/or LH.

38. A pharmaceuticalcomposition according to any of claims 31 to 37 for use in the treatment of infertility.

39 . A method of treatment of infertility comprising a step of administering to a subject a composition comprising rFSH according to any of claims 1 to 22 and/or a preparation according to any of claims 23 to 30 and/or a pharmaceutical composition according to any of claims 31 to 37.

40. Use of rFSH according to any of claims 1 to 22 and/or an rFSH preparation according to any of claims 23 to 30 in the manufacture of a medicament for the treatment of infertility.

41. A method of production of rFSH according to any of claims 1 to 22 and/or an rFSH preparation according to any of claims 23 to 30 comprising the step of producing or expressing the rFSH in a human cell line.