The present invention relates to human monoclonal autoantibodies (MAbs) reactive with the thyroid stimulating hormone (TSH) receptor (TSHR). One of the human MAbs (K1-18) has the ability to bind to the TSHR and to stimulate TSHR cyclic AMP activity. The other human MAb (K1-70) has the ability to bind to the TSHR and to block stimulation of cyclic AMP mediated by TSH and TSHR stimulating antibodies. Both human MAbs were isolated from the peripheral lymphocytes of a patient who presented with clinical symptoms of hypothyroidism.
Thyroid function is regulated by TSH secreted by the pituitary (Szkudlinski M W, et al 2002. Physiological Reviews 82: 473-502). TSH binds to the TSHR on the surface of thyrocytes and this is the first step in initiating the TSHR signalling cascade. Binding of TSH to the TSHR leads to stimulation of formation and release of thyroid hormones; thyroxine (T4) and tri-iodothyronine (T3). A feedback mechanism involving the levels of T4 and T3 in the circulation and thyrotropin releasing hormone (TRH) secreted by the hypothalamus controls the release of TSH that in turn controls thyroid stimulation and the levels of thyroid hormones in serum (Szkudlinski M W, et al, 2002 supra). The TSHR is a G-protein coupled receptor and is composed of three domains:—a leucine rich repeat domain (LRD), a cleavage domain (CD) and a transmembrane domain (TMD) (Núñez Miguel R, et al 2004. Thyroid 14: 991-1011).
It is well documented in the art that some patients with autoimmune thyroid disease (AITD) develop autoantibodies which are reactive with the TSHR (Rees Smith B, et al 1988. Endocrine Reviews 9: 106-121). There are two main types of TSHR autoantibodies (TRAbs); a stimulating type and a blocking type. Thyroid stimulating type autoantibodies bind to the TSHR and mimic the actions of TSH thereby stimulating the thyroid to produce high levels of T4 and T3; these autoantibodies are also described as TRAbs with stimulating activity or TSH agonist activity (Rees Smith B, et al 2007. Thyroid 17: 923-938). The feedback control mechanism of thyroid function is no longer effective in the presence of thyroid stimulating autoantibodies and patients present with the clinical symptoms of a hyperactive thyroid characterised by an excess of thyroid hormones in serum and its metabolic consequences. This condition is known as Graves' disease. TRAbs with stimulating activity may also interact with the TSHRs in retroorbital tissue and contribute to the development of the eye signs of Graves' disease. A human monoclonal autoantibody which acts as a powerful thyroid stimulator (hMAb TSHR1; also referred to as M22) has been described in detail in WO 2004/050708A2. The structure of the complex of M22 Fab bound to the TSHR LRD has been solved by x-ray crystallography at 2.55 Å resolution as described in WO2008/025991A1. Analysis of the structure of the TSHR-M22 complex provides detailed information about the receptor residues and the stimulating autoantibody residues involved in interactions with each other.
M22 has been used in ELISA for TSHR antibody measurement (Zöphel, K et al, Clinica Chimica Acta 2009 and Zöphel, K et al, Clinica Chimica Acta 2008.
Blocking type TRAbs occur less frequently in patients with AITD than stimulating autoantibodies. Blocking type autoantibodies bind to the TSHR, prevent TSH from binding to the receptor but have no ability to stimulate TSHR activity. Consequently formation and secretion of thyroid hormones (T4 and T3) is greatly reduced and the patients with this type of TRAb can present with clinical symptoms of an under-active thyroid (hypothyroidism). Blocking type autoantibodies are known as TRAbs with blocking activity or TSH antagonist activity (Rees Smith B, et al 1988 supra and Rees Smith B, 2007 et al supra). TRAbs with blocking activity when present in serum of pregnant women cross the placenta and may block the TSHRs in the foetal thyroid leading to neonatal hypothyroidism and serious consequences for development. Furthermore, TRAbs with blocking activity can be found in breast milk of affected mothers and may cause clinical hypothyroidism in the baby (Evans C, et al 2004 European Journal of Endocrinology 150: 265-268). A human autoantibody to the TSHR with TSH antagonist activity (5C9) has been described in detail in WO 2008/099185A1. Clinical symptoms in patients with AITD and circulating TRAbs are related to the effect of autoantibodies on TSHR activity i.e. whether the TRAbs cause stimulation or blocking. It has been proposed, however, that in some patients a mixture of stimulating and blocking TRAbs may be present simultaneously with the overall clinical presentation related to higher concentration and/or activity of one type of the TRAbs (Rees Smith B et al 1988 supra; Furmaniak J et al 1993 Springer Seminars in Immunopathology 14: 309-321 and Schott M et al 2005 Trends in Endocrinology and Metabolism 16: 243-248). Furthermore, the concentrations and/or activities of stimulating or blocking TRAb may vary in the same patient during the course of the disease and indeed fluctuation of symptoms from hypo- to hyperthyroidism in the same patient over time has been reported (Rees Smith B et al 1988 supra; Furmaniak J and Rees Smith B 1993 supra and Schott M et al 2005 supra). However, attempts to separate the TRAbs with different bioactivity or to differentiate between these TRAbs in serum samples using currently available bioassays is difficult. More recently, the invention described in WO2006/016121A1 provides a means to discriminate between stimulating and blocking types of TRAbs using bioassays that employ TSHR mutated at R255.
Human recombinant TSH (Thyrogen®) is a preparation of human TSH produced under cGMP regulations as a recombinant protein and approved by the US FDA as an aid in the diagnosis of residual or recurrent thyroid cancer (Duntas L H, Cooper D S 2008 Thyroid 18: 509-516). Monitoring of thyroid cancer patients after treatment includes stimulation of thyroid remnants or metastases with recombinant human TSH followed by a thyroid scan and/or measurement of serum thyroglobulin levels (Duntas L H and Cooper D S 2008 supra). Human chorionic gonadotropin is a hormone produced during pregnancy which has mild thyroid stimulating effects (Grossmann M et al 1997 Endocrine Reviews 18: 476-501). Characterisation of stimulating or blocking types of TRAbs and how they interact with the TSHR is of critical importance for development of improved methods to diagnose and manage different forms of AITD. In addition these studies are critical for developing new strategies for the management of diseases associated with an autoimmune response to the TSHR. The availability of potent thyroid stimulators other than recombinant human TSH provides new alternatives for monitoring and managing thyroid cancer patients.
The invention described in WO2004/050708A2 provides details of the properties of a human monoclonal autoantibody (MAb) with powerful stimulating activity and its interaction with the TSHR. The interactions between this autoantibody (M22) and the TSHR LRD have been solved at the molecular level from an X-ray diffraction analysis (2.55 Å resolution) of a complex between the two molecules as described in WO2008/025991A1. WO2006/016121A1 discloses a mutated TSHR preparation including at least one point mutation which can be used in the differential screening and identification of patient serum stimulating TSHR autoantibodies, patient serum blocking TSHR autoantibodies and TSH in a sample of body fluid from a patient being screened. Generation and characterisation of a mouse MAb (9D33) with TSHR blocking activity is also described in WO2004/050708A2. 9D33 binds to the TSHR with high affinity (2×1010 L/mol) and is an effective antagonist of TSH, hMAb TSHR1 (M22) and patient serum TRAbs with stimulating or blocking activities. WO2008/099185A1 discloses the isolation and characterisation of human MAb (5C9) to the TSHR that is an effective antagonist of TSH and of stimulating TRAbs in patient sera. 5C9 has been found unexpectedly to inhibit TSHR constitutive activity (also referred to as the TSHR basal activity), that is to say the production of cyclic AMP in a test system in the absence of TSH or M22. Furthermore, 5C9 has been found to inhibit TSHR cyclic AMP activity associated with TSHR activating mutations. WO2008/091981A2 describes a mouse MAb that has the ability to suppress the constitutive activity of TSHR and the methods of using the MAb to treat thyroid diseases including hyperthyroidism and thyroid cancer. The properties of the MAb described in WO2008/091981A2 are also disclosed in Chen C R et al 2007 Endocrinology 148: 2375-2382.
Antibodies K1-18 and K1-70 have been isolated from the peripheral blood lymphocytes of a 54 year old female patient with hypothyroidism and high levels of TSHR autoantibodies. The patient had an 8 year history of AITD and first presented with hyperthyroidism and responded to treatment with methimazole which continued for 3 years. However, approximately 10 months after reaching the euthyroid state (i.e. having normal function) the patient developed hypothyroidism and was treated with thyroxine. The patient had been hypothyroid for approximately 4.5 years at the time of blood collection. At the time of lymphocyte isolation serum TRAb levels were 160 Units/L measured by TSH binding inhibition assay. The serum also showed an ability to block TSH stimulation of the TSHR (cyclic AMP based assay). Serum autoantibodies to thyroid peroxidase were positive at >500 Units/mL (Units are of the reference preparation 66/387 from National Institute for Standards and Control (NIBSC) Potters Bar, UK).
The patient's lymphocytes were immortalised by infection with Epstein Barr virus (EBV) and supernatants of cultures of the infected cells screened for their ability to inhibit 125I-TSH binding to TSHR coated tubes. Cells from positive cell cultures were fused with a mouse/human cell line and screened as above. 2 stable clones secreting TSHR autoantibodies were obtained. IgGs were purified from supernatants of the clone cultures and the ability of the 2 MAbs (K1-18 and K1-70) IgGs to bind to the TSHR and influence TSHR activity assessed. In particular, the ability of K1-18 or K1-70 to inhibit TSH binding to the TSHR was studied. The ability of K1-18 to stimulate the TSHR was also studied and compared to the activity of various other thyroid stimulators. The ability of K1-70 to inhibit the ability of TSH to stimulate the TSHR was studied and compared to the activities of other TSH antagonists. Furthermore, the ability of stimulating or blocking patient serum TRAbs to inhibit TSHR binding and biological activity of K1-18 and K1-70 was assessed. In addition, the use of K1-18 and K1-70 in assays for TSHR antibodies, TSH and related compounds was investigated. Variable region (V region) genes of the heavy (HC) and light chains (LC) of K1-18 and K1-70 were sequenced and the complementarity determining regions (CDRs) assigned. Furthermore, purified preparations of K1-70 Fab were crystallised and analysed using X-ray diffraction methods. These analyses provided molecular level details about the overall structure of K1-70 Fab and the topography of the antigen binding site of K1-70.
According to one aspect of the invention there is provided an isolated human antibody molecule which binds to a TSHR and which reduces ligand induced stimulation of said TSHR but has no effect on constitutive activity of said TSHR.
Preferably, there is provided an isolated human antibody molecule or fragment thereof which binds to the TSHR and which reduces ligand-induced stimulation of the TSHR but has no effect on said TSHR constitutive activity wherein said human antibody or fragment thereof has the characteristics of patient serum TSH receptor autoantibodies of inhibiting TSH and M22 binding to the TSHR. More preferably the isolated human antibody molecule or fragment thereof has at least one further characteristic of patient serum TSH receptor autoantibodies selected from having a binding affinity for the TSHR of at least 108 L/mol and the ability to cause detectable blocking of ligand-induced TSHR stimulation at an antibody concentration of less than 10 μg/mL. Even more preferably the further characteristics of patient serum TSH receptor autoantibodies are selected from having a binding affinity for the TSHR of at least 109 L/mol and the ability to cause detectable blocking of ligand-induced TSHR stimulation at an antibody concentration of less than 1 μg/mL, preferably less than 0.1 μg/mL. The isolated human antibody may be an antagonist of TSH and/or thyroid stimulating autoantibodies, and/or thyroid stimulating animal antibodies and/or of human chorionic gonadotropin.
The isolated antibody molecule may be an inhibitor of TSH receptor binding by at least one of TSH, M22 or K1-18.
The isolated antibody molecule may comprise an antibody VH domain selected from the amino acid sequence of
Preferably, the isolated antibody molecule comprises CDR I, II and III of
The isolated antibody molecule may have a molecular structure as shown in
According to another aspect of the invention there is provided an isolated antibody molecule which binds to the TSHR so as to stimulate the TSHR, the antibody molecule comprising an antibody VL domain selected from the amino acid sequences of
Binding of the isolated antibody molecule to the TSHR may be inhibited by patient serum TSHR antibodies with thyroid stimulating or blocking activities.
Binding of the isolated antibody molecule to the TSHR may be inhibited by at least one of M22, K1-70, 5C9, 9D33 and thyroid stimulating mouse monoclonal antibodies.
The isolated antibody molecule may comprise an antibody VL domain selected from the amino acid sequences of
Preferably, the isolated antibody molecule comprises CDR I (SEQ ID No 6 and 16), II (SEQ ID No 7 and 17) and III (SEQ ID No 8 and 18) of
The isolated antibody molecule may comprise an antibody VL domain consisting of an amino acid sequence of
Additionally or alternatively, an antibody according to the invention may comprise one or more amino acid sequences having substantial homology to those CDRs. Preferably the antibody shows 70-99.9% amino acid homology to the CDRs (SEQ ID No 24-26 and 34-36) shown in
In most applications a VH domain in an antibody molecule according to the invention will be arranged with a VL domain to provide a TSHR binding site. In some applications a VH domain alone may be provided to bind a TSHR.
Methods of grafting antibody domains are well known in the art such that an antibody molecule in accordance with the invention can be constructed using VH and VL domains or portions thereof from different sources.
The term “antibody molecule” and cognate terms, such as “antibody molecules”, used herein in relation to antibody molecules of the invention embraces, according to context, immunoglobulin-based binding moieties such as monoclonal, recombinant, synthetic and polyclonal antibodies, single chain antibodies, multi-specific antibodies and also binding moieties, which may be substituted by the skilled addressee for such immunoglobulin-based binding moieties, such as domain antibodies, diabodies, as well as IgG[Delta]CH2, F(ab′)2), Fab, scFv, VL, VH, dsFv, Minibody, Triabody, Tetrabody, (scFv)2, scFv-Fc, F(ab′)3 moieties (Holliger P, et al 1993 Proc Natl Acad Sci USA 90: 6444-6448.), (Carter P J 2006 Nat Rev Immunol 6: 343-357). The term also embraces fragments of such entities, preferably fragments which bind TSHRs, and more preferably have the effects of K1-18 or K1-70.
The terms “thyroid stimulating hormone receptor” and “TSHR” refer to full length human TSHR having the amino acid sequence shown in
The isolated antibody of the invention may preferably be in the form of a monoclonal antibody, a recombinant antibody or a synthetic antibody. CDRs I, II or III from the K1-18 or K1-70 VH or VL domains may be incorporated into a suitable framework. Variants of the K1-18 and K1-70 VH and VL domains and their CDRs can be produced by modifications using methods well known to those skilled in the field.
Such variants may comprise one or more amino acid sequence variations, including the addition, deletion, substitution or insertion mutations. The framework of K1-18 or K1-70 may also be modified in antibody molecules according to the invention. The isolated antibody according to the invention may have a framework which is human or non-human.
According to another aspect of the invention there is provided an isolated nucleotide encoding an isolated antibody molecule or fragment thereof according to the invention, comprising an antibody VL domain comprising an amino acid sequence of
The isolated nucleotide may comprise a nucleotide sequence of
A plurality of such nucleotides may be provided, for example in a bacteriophage display library. Such bacteriophage display libraries may be used to express a variety of antibody molecules or fragments thereof such as isolated domains.
The invention also provides a vector including an isolated nucleotide according to the invention, or a host cell including such a vector or a nucleotide according to the invention. The vector may be a plasmid, virus or fragment thereof. Many different types of vectors are known to the skilled addressee. The isolated cell may express an antibody according to the invention. Preferably, the isolated cell secretes an antibody according to the invention. Preferably an isolated cell according to the invention is from a stable hetero-hybridoma cell line.
Another aspect of the invention provides a method of producing an isolated antibody molecule or a fragment thereof such as an isolated domain in accordance with the invention, the method comprising expressing a nucleotide encoding such an antibody molecule, or a fragment thereof.
According to a further aspect of the invention there is provided a method of producing an antibody according to the invention, the method comprising culturing one or more isolated host cells according to the invention whereby the antibody is expressed by the cell. Preferably, the antibody is secreted by the cell,
According to another aspect of the invention there is provided a pharmaceutical composition comprising an isolated antibody molecule according to the invention, and a carrier.
A pharmaceutical composition according to the invention may be suitable for human administration.
Preferably a pharmaceutical composition according to the invention has no significant adverse effect on the immune system of the subject.
Various formats are contemplated for pharmaceutical compositions according to the invention. A pharmaceutical composition according to the invention for use in the treatment of a thyroid-related condition may be in an injectable format. A pharmaceutical composition according to the invention for use in the treatment of ophthalmic Graves' disease is preferably in the form of eye drops. Pharmaceutical compositions of this invention comprise an isolated antibody in accordance with the invention, with a pharmaceutically acceptable carrier, adjuvant or vehicle. Pharmaceutically acceptable carriers, adjuvants and vehicles that may be used in the pharmaceutical compositions of this invention include, but are not limited to, ion exchangers, alumina, aluminium stearate, lecithin, serum proteins, such as human serum albumin, buffer substances such as phosphates, glycine, sorbic acid, potassium sorbate, partial glyceride mixtures of saturated vegetable fatty acids, water, salts or electrolytes, such as protamine sulphate, disodium hydrogen phosphate, potassium hydrogen phosphate, sodium chloride, zinc salts, colloidal silica, magnesium trisilicate, polyvinyl pyrrolidone, cellulose-based substances, polyethylene glycol, sodium carboxymethylcellulose, polyacrylates, waxes, polyethylene-polyoxypropylene-block polymers, polyethylene glycol and wool fat. The pharmaceutical compositions of the invention may be administered orally, parenterally, by inhalation spray, topically, by eyedrops, rectally, nasally, buccally, vaginally or via an implanted reservoir. We prefer oral administration or administration by injection. The term “parenteral” as used herein includes subcutaneous, intracutaneous, intravenous, intramuscular, intra-articular, intrasynovial, intrasternal, intrathecal, intralesional and intracranial injection or infusion techniques. The pharmaceutical compositions may be in the form of a sterile injectable preparation, for example, as a sterile injectable aqueous or oleaginous suspension. This suspension may be formulated according to techniques known in the art using suitable dispersing or wetting agents (such as, for example, Tween 80) and suspending agents. The sterile injectable preparation may also be a sterile injectable solution or suspension in a non-toxic parenterally-acceptable diluent or solvent, for example, as a solution in 1,3-butanediol. Among the acceptable vehicles and solvents that may be employed are mannitol, water, Ringer's solution and isotonic sodium chloride solution. In addition, sterile, fixed oils are conventionally employed as a solvent or suspending medium. For this purpose, any bland fixed oil may be employed including synthetic mono- or diglycerides. Fatty acids, such as oleic acid and its glyceride derivatives are useful in the preparation of injectables, as are natural pharmaceutically-acceptable oils, such as olive oil or castor oil, especially in their polyoxyethylated versions. These oil solutions or suspensions may also contain a long-chain alcohol diluent or dispersant such as Ph. Helv or a similar alcohol. The pharmaceutical compositions of this invention may be orally administered in any orally acceptable dosage form including, but not limited to, capsules, tablets, and aqueous suspensions and solutions. In the case of tablets for oral use, carriers which are commonly used include lactose and corn starch. Lubricating agents, such as magnesium stearate, are also typically added. For oral administration in a capsule form, useful diluents include lactose and dried corn starch. When aqueous suspensions are administered orally, the active ingredient is combined with emulsifying and suspending agents. If desired, certain sweetening and/or flavouring and/or colouring agents may be added. The pharmaceutical compositions of this invention may also be provided in the form of suppositories for rectal administration. These compositions can be prepared by mixing a compound of this invention with a suitable non-irritating excipient which is solid at room temperature but liquid at the rectal temperature and therefore will melt in the rectum to release the active components. Such materials include, but are not limited to, cocoa butter, beeswax and polyethylene glycols. Topical administration of the pharmaceutical compositions of this invention is especially useful when the desired treatment involves areas or organs readily accessible by topical application. For application topically to the skin, the pharmaceutical composition should be formulated with a suitable ointment containing the active components suspended or dissolved in a carrier. Carriers for topical administration of the compounds of this invention include, but are not limited to, mineral oil, liquid petroleum, white petroleum, propylene glycol, polyoxyethylene polyoxypropylene compound, emulsifying wax and water. Alternatively, the pharmaceutical composition can be formulated with a suitable lotion or cream containing the active compound suspended or dissolved in a carrier. Suitable carriers include, but are not limited to, mineral oil, sorbitan monostearate, polysorbate 60, cetyl esters wax, cetearyl alcohol, 2-octyldodecanol, benzyl alcohol and water. The pharmaceutical compositions of this invention may also be topically applied to the lower intestinal tract by rectal suppository formulation or in a suitable enema formulation. Topically-transdermal patches are also included in this invention. The pharmaceutical compositions of this invention may be administered by nasal aerosol or inhalation. Such compositions are prepared according to techniques well-known in the art of pharmaceutical formulation and may be prepared as solutions in saline, employing benzyl alcohol or other suitable preservatives, absorption promoters to enhance bioavailability, fluorocarbons, and/or other solubilising or dispersing agents known in the art.
Antibodies according to the first mentioned aspect of the invention such as K1-70 have potential applications for management and control of conditions associated with TSHR activation for example: Graves' disease, Graves' opthalmopathy or hyperthyroidism due to abnormal levels of TSH or hCG.
Antibodies according to the second mentioned aspect of the invention such as K1-18 have applications for stimulating the TSHR in different clinical conditions and treatment situations. These conditions include diagnosis and management of thyroid cancer and its metastases, multinodular goitre or congenital hypothyroidism.
According to another aspect of the invention there is provided the use of an isolated antibody molecule or a pharmaceutical composition according to the invention in therapy. The invention also provides an isolated antibody molecule or a pharmaceutical composition according to the invention for use in therapy.
According to another aspect of the invention there is provided a method of characterising the activity of TSHR antibodies, TSH or human chorionic gonadotropin, the method comprising a step including the use of an isolated antibody molecule according to the invention.
According to another aspect of the invention there is provided an in vitro method of stimulating the TSHR in mammalian cells, the method comprising contacting the cells with an isolated antibody molecule according to the invention.
According to another aspect of the invention there is provided an in vivo method of stimulating the TSHR in mammalian cells, the method comprising contacting the cells with an isolated antibody molecule according to the invention. Preferably, cells of a subject with thyroid cancer and its metastases, multinodular goitre and/or congenital hypothyroidism are contacted with an isolated antibody according to the invention.
According to another aspect of the invention there is provided an in vivo method of preventing ligand induced stimulation of TSHRs in mammalian cells, the method comprising contacting a TSHR with an isolated antibody molecule according to the invention. The ligand may be a thyroid stimulating autoantibody, TSH or human chorionic gonadotropin. The mammalian cells may be thyroid cells or extra-thyroidal cells. Mammalian extra-thyroidal cells may be in retro-orbital tissue or pre-tibial tissue.
In methods according to this aspect of the invention the isolated antibody molecule may be used in combination with another TSHR binding antibody such as 5C9 or 9D33 referred to above.
The thyroid-related condition may be selected from hyperthyroidism, Graves' disease, ophthalmic Graves' disease and neonatal hyperthyroidism. Alternatively, the thyroid-related condition may be hypothyroidism related to the presence of TRAbs with blocking activity in patients with AITD, neonatal hypothyroidism due to transfer of maternal TRAbs (via placenta or breast milk).
The subject treated in the various methods of the invention described above is preferably human.
According to another aspect of the invention there is provided a diagnostic method for detecting autoantibodies to TSHRs, the method comprising contacting a sample, which has been isolated from a subject believed to contain such autoantibodies, and an antibody molecule according to the invention with a TSHR.
According to another aspect of the invention there is provided a diagnostic method for detecting an antibody in accordance with the invention, preferably a human antibody, to the TSHR or antibodies to the TSHR in human serum comprising contacting any one of the antibodies to the TSHR with a TSHR fragment comprising amino acids 22-260 of the TSHR (TSHR260) (
A suitable detectable label that can be employed in a method according to the present invention can be selected from the group consisting of enzymic labels, isotopic labels, chemiluminescent labels, fluorescent, dyes and the like.
In the case where an isotopic label (such as 125I, 14C, 3H or 35S) is employed, monitoring may therefore comprise measuring radioactivity dependent on binding of an antibody molecule according to the present invention. Radioactivity is generally measured using a gamma counter, or liquid scintillation counter.
According to another aspect of the invention there is provided a method of identifying small molecules that bind to TSHR260 (SEQ ID No 75), the method comprising contacting a candidate small molecule with TSHR260 for example in an ELISA and selecting small molecules that bind to TSHR260. Further, there is provided a method of identifying small molecules that have the ability to prevent TSHR autoantibody binding to TSHR260, the method comprising determining inhibition of binding of TSHR autoantibody (stimulating or blocking) to TSHR260 in the presence of a candidate small molecule and selecting small molecules that inhibit TSHR autoantibody binding. Small molecules identified in this way may be developed to provide new drugs to control autoimmune thyroid disease caused by TSHR autoantibodies (stimulating or blocking).
The present invention provides new and/or improved means to:
Antibody molecules and methods in accordance with the invention will now be described, by way of example only, with reference to the accompanying drawings,
Lymphocyte Isolation and Cloning of Human Monoclonal TSHR Autoantibodies.
The monoclonal autoantibodies K1-18 and K1-70 were isolated using the procedure described in WO2004/050708A2. Lymphocytes were isolated from a blood sample collected from a patient with an 8 year clinical history of AITD and high levels of TRAbs. Patient consent and Local Ethical Committee approval were obtained. The patient was first diagnosed with hyperthyroidism, reached the euthyroid state after treatment with methimazole, however, approximately 4.5 years prior to blood collection she developed hypothyroidism. At the time of blood collection the patient was being treated with thyroxine (50 μg daily). The lymphocytes were infected with Epstein Barr Virus (EBV) (European Collection of Cell Cultures—ECACC; Porton Down, SP4 0JG, UK) and cultured on mouse macrophage feeder layers as described in WO2004/050708A2. Immortalised lymphocytes secreting TSHR autoantibodies were fused with a mouse/human hybrid cell line K6H6/B5 (ECACC) and cloned four times by limiting dilution to obtain a single colony. The presence of TSHR autoantibody in cell culture supernatants at different stages of cloning was detected by inhibition of labelled TSH binding to the TSHR (WO2004/050708A2). Two single clones producing the TSHR autoantibodies were expanded and supernatants from the cultures were harvested for autoantibody purification. One clone was designated as K1-18 and the other as K1-70.
TSHR human MAbs IgGs were purified from culture supernatants using protein A affinity chromatography on MabSelect™ (GE Healthcare, UK) as described in Sanders J et al 2004. Thyroid 2004 14: 560-570) and purity assessed by SDS-polyacrylamide gel electrophoresis (PAGE). The heavy chain isotype was determined using a radial diffusion assay (The Binding Site; Birmingham, B29 6AT, UK), and the light chain isotype was determined by Western blotting with anti-human kappa chain and anti-human lambda chain specific mouse monoclonal antibodies (Sigma-Aldrich Company Ltd, Poole, UK). Purified K1-18 IgG was treated with mercuripapain (Sigma Aldrich, Poole, UK) at a IgG/enzyme ratio of 100:1 in phosphate buffered saline (PBS; 137 mmol/L NaCl, 8.1 mmol/L Na2HPO4, 2.7 mmol/L KCL, 1.47 mmol/L KH2PO4, pH 7.4 containing cysteine at final concentration of 1 mmol/L and EDTA at final concentration of 2 mmol/L) for 4 hours at 37° C. The reaction was stopped by addition of iodoacetamide (final concentration of 50 mmol/L) for 30 minutes at room temperature. The reaction mixture was then passed through a MabSelect column to remove any intact IgG or Fc fragments from the Fab preparation. The Fab containing solution was dialysed into PBS containing 3.1 mmol/L NaN3 and concentrated using a Centriprep concentrator (Millipore, Watford, WD18 8YH, UK) when appropriate. K1-70 Fab were obtained using similar method except that an IgG/enzyme ratio of 200:1 was used and the digestion with enzyme was for 1 hour at 37° C. Analysis by SDS-PAGE indicated that intact IgG was undetectable in the Fab preparations. IgG preparations were labelled with 125I as described in Sanders J et al 1999. Journal of Clinical Endocrinology and Metabolism. 1999 84: 3797-3802) or with biotin hydrazide (Perbio Science, Cramlington, UK) (Rees Smith et al 2004. Thyroid 14: 830-835).
Inhibition of 125I-TSH or 125I-Labelled Human MAbs Binding to the TSHR
Binding inhibition assays were carried out using TSHR coated tubes as described in WO2004/050708A2. In the assay, 100 μL of test sample (MAb preparation, patient serum or unlabelled TSH) and 50 μL of start buffer (RSR Ltd) were incubated in TSHR coated tubes for 2 hours at room temperature with gentle shaking. After aspiration, the tubes were washed and 100 μL of 125I-labelled protein (5×104 cpm) added and incubated for 1 hour at room temperature with shaking. The tubes were then aspirated, washed and counted in a gamma counter. Inhibition of labelled protein binding was calculated as 100×[1−(cpm bound in the presence of test material/cpm bound in the presence of control material)]. MAb preparations used in these experiments were K1-18, K1-70, M22, 5C9, 9D33 described above. TSMAbs 1-7 are mouse thyroid stimulating MAbs (WO 03/01863 and Sanders J et al 2002 supra). Control material was a pool of healthy blood donor sera or individual healthy blood donor sera or other materials as indicated in the results of various experiments.
Scatchard Analysis of Human MAb IgG Binding to the TSHR
Unlabelled K1-18 or K1-70 IgG in 50 μL of assay buffer (50 mmol/L NaCl, 10 mmol/L Tris pH 7.8 and 0.1% Triton X-100) and 50 μL start buffer (RSR Ltd) and 50 μL of 125I-labelled K1-18 or K1-70 IgG respectively (30,000 cpm in assay buffer) were incubated in TSHR coated tubes for 2 hours at room temperature with shaking (maximum binding occurred under these conditions), aspirated, washed twice with 1 mL of assay buffer and counted in a gamma counter. The concentration of IgG bound vs bound/free was plotted (Scatchard G 1949. Annals of the New York Academy of Sciences 51: 660-672) to derive the association constant.
Inhibition of TSH Binding to the TSHR Measured by ELISA
A TRAb ELISA based on TSH-biotin binding to TSHR coated ELISA wells was used as described previously (Bolton J, et al 1999 Clinical Chemistry 45: 2285-2287). In the assay 75 μL of test sample was added to 75 μL of start buffer in the plate wells and incubated for 2 hours at room temperature with shaking at about 500 shakes/minute. After washing 100 μL of TSH-biotin was added and incubation continued for 25 minutes without shaking. The wells were washed again, the reaction developed using described standard procedures and the absorbance of each well read at 450 nm.
Inhibition of TSH-biotin binding was calculated as: 100×[1−(test sample absorbance at 450 nm/negative control sample absorbance at 450 nm)]. MAb preparations used in these experiments were K1-18, K1-70, M22, 5C9, 9D33 described above. TSMAbs 1-7 are mouse thyroid stimulating MAbs (WO03/01863 and Sanders J, et al 2002 supra). Control sample material was a pool of healthy blood donor sera or other materials as indicated in the results of various experiments.
Inhibition of M22 Binding to the TSHR in ELISA
A TRAb ELISA based on labelled M22 (M22 Fab-POD) binding to TSHR coated ELISA wells was used (Rees Smith B, et al 2004 supra). The assay was carried out as the TSH-biotin based ELISA except the first incubation was for 1 hour. Results were expressed as inhibition of M22 binding using the formula: 100×[1−(test sample absorbance at 450 nm/negative control sample absorbance at 450 nm)]. MAb preparations used in these experiments were K1-18, K1-70, M22, 5C9, 9D33 described above. TSMAbs 1-7 are mouse thyroid stimulating MAbs (Patent application number WO03/01863 and Sanders J, 2002 supra). Control material was a pool of healthy blood donor sera or other materials as indicated in the results of various experiments.
Analysis of TSHR Stimulation
The ability of K1-18 or K1-70 IgG and other preparations to stimulate production of cyclic AMP in Chinese hamster ovary (CHO) cells transfected with the human TSHR was tested as described in WO2004/050708A2. CHO cells expressing either approximately 5×104 or approximately 5×105 TSHR per cell were seeded into 96-well plates at 3×104 cells per well, adapted into DMEM (Invitrogen Ltd, Paisley, UK) without foetal calf serum and then test samples (TSH, IgG or patient serum) added (100 μL diluted in cyclic AMP assay buffer i.e. NaCl free Hank's Buffered Salts solution containing 1 g/L glucose, 20 mmol/L HEPES, 222 mmol/L sucrose, 15 g/L bovine serum albumin and 0.5 mmol/L 3 isobutyl-1-methylxanthine pH 7.4; cyclic AMP assay hypotonic buffer) and incubated for 1 hour at 37° C. After removal of test solutions, cells were lysed and cyclic AMP concentration in the lysates assayed using Direct Cyclic AMP Correlate—EIA kits from Assay Designs; Cambridge Bioscience, UK. Results are expressed as pmol/mL of cyclic AMP in the cell lysate (200 μL). Some experiments were carried out under isotonic buffer condition. In these experiments Krebs Ringer Hepes buffer (KRH buffer) was used (124 mmol/L NaCl, 5 mmol/L KCl, 1.25 mmol/L MgSO4, 1.45 mmol/L CaCl2, 1.25 mmol/L KH2PO4, 25 mmol/L HEPES, 8 mmol/glucose, 0.5 g/L bovine serum albumin, 0.5 mmol/L 3 isobutyl-1-methylxanthine, pH 7.4). Cells were allowed to reach the required density, the culture medium removed and the cells washed with 1 mL of KRH buffer. Fresh KRH buffer was then added and the cells incubated for 30 minutes at 37° C. The buffer was then removed and replaced with fresh KRH buffer containing test sample (TSH, MAb preparations, serum samples etc). The next steps were then carried out as described above for the experiments under the hypotonic conditions (i.e. in cyclic AMP assay buffer). In some experiments the effect of various MAbs on the TSHR stimulating activity of various preparations (for example, TSH, human MAbs, patient sera) measured as described above was assessed. This was carried out by comparing (a) the stimulating activity of the sample alone with (b) stimulating activity in the presence of various MAbs.
Measurement of Antagonist (Blocking) Activity
The ability of K1-70 IgG and other preparations to inhibit the stimulating activity of porcine (p) TSH, native human (h) TSH and recombinant human (rh) TSH, MAb M22, MAb K1-18 and patient serum TRAbs in CHO cells expressing TSHRs was assessed. This was carried out by comparing the stimulatory effect of TSH, M22, K1-18 or TRAbs in the absence and in the presence of K1-70 IgG (or other preparations being tested). The assay was carried out as described above except 50 μL of K1-70 (or other preparations being tested) diluted in cyclic AMP assay buffer was added to the cell wells followed by 50 μL of TSH or M22 or K1-18 or patient serum (diluted as appropriate in cyclic AMP assay buffer) and incubated and tested as for the stimulating assay described above. Other MAbs and sera from patients with blocking type TRAbs were tested in this assay in addition to K1-70.
Association and Dissociation of K1-18 and K1-70 Binding to the TSHR
The association and dissociation of K1-18 IgG, K1-18 Fab, K1-70 IgG and K1-70 Fab binding to the full length TSHR and the TSHR260 was studied using the method as described in: Nakatake N, et al Thyroid 2006, 16; 1077-1084. The full length TSHR or TSHR260 were coated onto plastic tubes which had been pre-coated with an appropriate mouse MAb to the TSHR. In association experiments 100 μL of 125I-labelled IgG or Fab were incubated in the TSHR coated tubes at room temperature for 5-180 min. The tubes were then aspirated, washed with assay buffer and counted in a gamma counter. In the dissociation experiments 100 μL of 125I-labelled IgG or Fab were incubated in TSHR coated tubes for 180 min at room temperature followed by the addition of 10 μL of 1 mg/mL of various MAb IgG or Fab preparations and incubation for 0-180 min at room temperature. At different time points the tubes were aspirated, washed and counted. In some experiments TSH or buffer was added instead of a MAb preparation.
Amino Acid Mutations in the TSHR
The methods used to introduce specific mutations into the TSHR sequence have been described in patent application WO2006/016121A. Furthermore, transfection of mutated TSHR constructs into CHO cells using the Flp-In system is also described in WO2006/016121A. Flp-In-CHO cells expressing either wild type or mutated TSHRs were seeded into 96 well plates and used to test the ability of various preparations to stimulate cyclic AMP activity in the CHO cells expressing the TSHR containing amino acid mutations. These experiments were compared to similar experiments carried out using CHO cells expressing wild type TSHR. Flp-In-CHO cells expressing either wild type or mutated TSHRs were also used in experiments to study the ability of various preparations to block the stimulating activity of TSH, stimulating antibodies or patient serum TRAbs as described above.
Production of TSHR260-Alkaline Phosphatase (TSHR260-AP) Construct
The TSHR 260 construct (coding amino acids 1-260 of the human TSHR; amino acids 1-21 being the leader sequence) was amplified using full length human TSHR as the template (Oda Y, et al 1998. Journal of Molecular Endocrinology 20: 233-244) and joined to the coding sequence of a secreted alkaline phosphatase (minus the 17 amino acid alkaline phosphatase leader sequence) using the cloning vector pSEAP2-basic (Clontech) as the template. Two PCR reactions were carried out, the first used the full length TSHR amplified with specific primers SEQ ID No 77 and SEQ ID No 78 primers (Sigma Genosys) which added an EcoRI restriction site at the N-terminus, and a 1 amino acid linker (Asparagine) and the first 8 amino acids (excluding the 17 amino acid leader sequence) of the secreted alkaline phosphatase at the C terminus. The second PCR was carried out using the cloning vector pSEAP2-basic amplified with the primersSEQ ID No 79 and SEQ ID No 80 which adds amino acids 254-260 of the TSHR and a 1 amino acid linker (Asparagine) to the N terminus of the secreted alkaline phosphatase and a 6 histidine tag, a stop codon and an XhoI restriction site at the C-terminus of the secreted alkaline phosphatase gene. The PCR reactions were carried out for 30 cycles of 1 minute at 94° C., 1 minute at 40° C. and 1 minute at 72° C. followed by 7 minutes at 72° C. The PCR products were run on 1% agarose gels and the DNA extracted using a geneclean II kit (Anachem Ltd, Luton) following the manufacturer's instructions. Purified PCR products 1 and 2 were then used to set up a third PCR to construct the whole TSHR 260-alkaline phosphatase gene. The PCR 3 reaction contained 200 ng of PCR 1 and 200 ng of PCR 2 product and PCR 3 was carried out for 7 cycles at 94° C. for 1.5 minutes, 65° C. for 1.5 minutes and 72° C. for 1.5 minutes. The temperature was then increased to 94° C. again for 2 minutes and primers SEQ ID No 77 and 80 added followed by 30 cycles of 94° C. for 1 minute, 52° C. for 1 minute and 72° C. for 2 minutes. The PCR 3 product was cloned into pFastBac1 using EcoRI and XhoI restriction sites and the presence of the mutation was verified using sequencing by the Sanger-Coulson method (Sanger F et al 1997. Proceedings of the National Academy of Sciences of the USA 74: 5463-5467). Recombinant DNA was made using the Bac to Bac Baculovirus expression system (Invitrogen, UK) and transfected into Sf-9 cells to obtain and amplify recombinant baculovirus stock as described in WO02008/025991A1. TSHR260-AP was expressed in insect cells as described in WO02008/025991A1.
ELISA Based on TSHR260-AP
An ELISA was established based on the ability of divalent TSHR antibodies to bind with one antigen binding site to TSHR coated onto an ELISA plate well and with the other antigen binding site to TSHR260-AP in a liquid phase i.e. forming a bridge. TSHR in the form of full length detergent-solubilised receptor expressed in CHO cells was coated onto ELISA plate wells via a C-terminal antibody as described previously (Bolton J et al 1999 supra). In the assay 75 μL of start buffer (as described for TRAb ELISA; Bolton J, et al 1999 supra) and 75 μL of test sample (patient sera or monoclonal antibodies) were added to the ELISA plate wells coated with the full length detergent solubilised TSHR and incubated for 2 hours at room temperature with shaking (500 rpm). Then the contents of the wells were removed, the wells washed 3 times with wash buffer (50 mmol/L NaCl, 20 mmol/L Tris pH 7.8, 1% Triton X-100) followed by addition of 100 μL of TSHR260-AP (diluted in wash buffer containing 0.2 g/L MgCl2-6H2O and 2 g/L BSA). After incubation for 1 hour at room temperature with shaking (500 rpm) the wells were emptied, washed (3 times) and 100 μL of p-nitrophenyl phosphate (pNpp) substrate (Europa Bioproducts Ltd, Ely, Cambridge UK) added and the plate incubated in the dark for 45 minutes. Thereafter 100 μL of stop solution (1 mol/L NaOH) was added and the absorbance read at 405 nm in an ELISA plate reader. The results were expressed as OD405 nm absorbance values, values higher than those observed with a panel of healthy blood donor (HBD) sera indicated the presence of TSHR autoantibodies in the sample. In some experiments solubilised preparations of recombinant TSHR containing mutation R255D expressed in CHO cells were used to coat the ELISA plate wells.
Production of TSHR LRD C-CAP Construct
The TSHR LRD C-CAP construct coding amino acids 1-409 of the human TSHR with amino acids 306-384 removed, was amplified using full length human TSHR as the template (Oda Y, et al 1998. Journal of Molecular Endocrinology 20: 233-244). Two PCR reactions were carried out, the first used the full length TSHR amplified with T7 primer (SEQ ID No 81) and specific primer SEQ ID No 82 (Sigma Genosys, Gillingham, Dorset, UK) which added amino acids 385-342 of the TSHR to the C terminus of amino acid 305 of the TSHR. The second PCR was carried out using the full length TSHR amplified with BGH reverse primer SEQ ID No 83 and the specific primer (SEQ ID No 84), which adds amino acids 298-305 of the TSHR to the N terminus of amino acid 385 of the TSHR. The PCR reactions were carried out for 30 cycles of 1 minute at 94° C., 1 minute at 40° C. and 2 minutes at 72° C. followed by 7 minutes at 72° C. The PCR products were run on 1% agarose gels and the DNA extracted using a Geneclean II kit (Anachem Ltd, Luton, UK) following the manufacturer's instructions. Purified PCR products 1 and 2 were then used to set up a third PCR to construct a continuous TSHR sequence joining Ser305 to Tyr385 with amino acids 306-384 removed. The PCR 3 reaction which contained 200 ng of PCR 1 and 200 ng of PCR 2 product and PCR 3 was carried out for 7 cycles at 94° C. for 1.5 minutes, 65° C. for 1.5 minutes and 72° C. for 1.5 minutes. The temperature was then increased to 94° C. again for 2 minutes and T7 primer (SEQ ID No 81) and BGHR primer (SEQ ID No 83) were added followed by 30 cycles of 94° C. for 1 minute, 52° C. for 1 minute and 72° C. for 2 minutes. The PCR 3 product containing TSHR sequence omitting amino acids 306-384 was then run on a 1% agarose gel and the DNA extracted using a Geneclean II kit (Anachem Ltd) following the manufacturer's instructions. Purified PCR3 product was used as the template for construction of the TSHR LRD C-CAP gene in PCR 4. The PCR 4 reaction contained 200 ng of PCR 3 as template DNA and was amplified with T7 primer (SEQ ID No 81) and the specific primer (SEQ ID No 85) which adds a 6 histidine tag, a stop codon and an XhoI restriction site to the C-terminus of amino acid 409 of the TSHR sequence (1-409 with amino acids 306-384 deleted). PCR 4 was carried out for 30 cycles of 1 minute at 94° C., 1 minute at 40° C. and 1 minute at 72° C., followed by 10 minutes at 72° C. The PCR 4 product was cloned into pFastBac1 using BamHI and XhoI restriction sites and the presence of the mutation was verified using sequencing by the Sanger-Coulson method (Sanger F. et al 1997. Proceedings of the National Academy of Sciences of the USA 74: 5463-5467). Recombinant DNA was made using the Bac to Bac Baculovirus expression system (Invitrogen, Paisley, UK) and transfected into Sf-9 cells to obtain and amplify recombinant baculovirus stock as described in WO2008/025991A1. TSHR LRD C-CAP (
Comparison of Stability of Different TSHR Preparations
Temperature stability of different preparations of recombinant TSHR was compared. Full length solubilised TSHR expressed in CHO cells, TSHR260 expressed in insect cells, TSHR260-AP expressed in insect cells and TSHR. LRD C-CAP expressed in insect cells were tested. An aliquot of each of the above listed preparations was removed from −80° C. storage, thawed on ice, a sample returned to −80° C. as a control while the bulk was stored at room temperature (20-25° C.) for 24 or 48 hours. After 24 or 48 hours at room temperature the TSHR preparations were stored at −80° C. and then tested as described below. ELISA plate wells were coated with a F(ab′)2 preparation of the mouse TSHR MAb 14C4 (Jeffreys J et al 2002, Thyroid 12: 1051-1061 and Sanders J et al 2007 Thyroid 17: 395-410) at 1 μg/mL in coating buffer (Bolton J et al 1999 supra). TSHR preparations under investigation were diluted in 20 mmol/L NaCl, 10 mmol/L Tris pH7.8, 1% v/v Triton X-100, 1 g/L BSA, 200 mg/L NaN3 and 150 μL added to ELISA plate wells (in quadruplicate). After incubation overnight at 4° C. to allow the TSHR preparations to bind to the antibody (14C4 F(ab′)2) coated wells, the wells were washed and incubated with 75 μL of assay buffer (50 mmol/L NaCl, 20 mmol/L Tris pH7.8, 1% v/v Triton X-100, 1 g/L BSA) and 75 μL of healthy blood donor serum for 1 hour at room temperature at 500 shakes per minute on an ELISA plate shaker. Thereafter the contents of the wells were emptied, the wells washed and 100 μL of M22 Fab-peroxidase conjugate (see above) added to each well. After 25 minutes incubation at room temp without shaking the plate wells were washed again followed by addition of 100 μL of tetramethylbenzidine and a further incubation of 25 minutes at room temperature without shaking. The reaction was stopped by addition of 50 μL of 0.5 mol/L H2SO4 and the absorbance of each well read at 450 nm on an ELISA plate reader.
Variable Region Gene Analysis
The variable (V) region genes of K1-18 or K1-70 heavy and light chains were determined as described in WO2004/050708A2, using total RNA prepared from 1×107 hetero-hybridoma cells (secreting K1-18 IgG or K1-70 IgG) to produce mRNA for RT-PCR (reverse transcriptase PCR) reactions. Specific IgG1 HC and kappa LC sense and antisense strand oligonucleotide primers designed using the Medical Research Council's V-base (http://vbase.mrc-cpe.cam.ac.uk/) and synthesised by Invitrogen (Paisley, PA4 9RF, UK) were used in RT-PCR reactions with K1-18 mRNA. Specific IgG1 HC and lambda LC primers prepared as described above were used in RT-PCR reactions with K1-70 mRNA. The RT reaction was carried out at 50° C. for 15 minutes followed by 40 cycles of PCR at 94° C. for 15 seconds, 50° C. for 30 seconds and 72° C. for 30 seconds. DNA products were cloned into pUC18 and sequenced by the Sanger-Coulson method (Sanger F, et al 1977 supra). V region sequences were compared with available sequences of human Ig genes using Ig blast (http://www.ncbi.nlm.nih.gov/igblast/). The CDRs were assigned by the method of Kabat (Kabat E et al 1991 Sequences of proteins of immunological interest (US Public Health service, Bethesda, Md.) Fifth edition) and Ig blast (http://www.ncbi.nlm.nih.gov/igblast/). A second round of mRNA isolation was carried out from both the K1-70 and K1-18 hybridoma cell lines that had undergone further recloning by limiting dilution. The V-region sequences (K1-18 HC, K1-18 LC, K1-70 HC and K1-70 LC) were obtained by RT-PCR from the mRNA followed by cloning and sequencing as described above. In addition the RT-PCR reactions were also carried out using specifically designed PCR primers corresponding to the 5′ end of the respective leader sequences for each of the V regions. This allowed the identification of the actual oligonucleotide sequences (and derived amino acid sequences) at the N-termini of the HC and LC V regions of K1-18 and K1-70. In addition, the N-terminal amino acid sequence of K1-70 LC protein was analysed by Edman degradation reaction by Alta Bioscience (Birmingham, UK). This was possible after the N-terminal “deblocking” of the K1-70 LC protein preparation with pyroglutamate aminopeptidase. Purified K1-70 Fab (10 μg) was treated with 2.5 mU of pyroglutamate aminopeptidase (in 50 mmol/L Na2HPO4 pH 7.0; 10 mmol/L dithiothreitol and 1 mmol/L EDTA) for 6 hours at 75° C. An equal volume of SDS-PAGE sample buffer was added and after heating at 100° C. for 5 minutes, K1-70 Fab was resolved into the HC (Fd part) and the LC on 15% SDS-PAGE. The LC band was carefully cut out of the gel and the N-terminal protein sequence determined. Repeat rounds of RT-PCR and sequencing of K1-18 HC, K1-18 LC and K1-70 HC confirmed the V region sequences were the same as obtained before while the K1-70 LC V region sequence differed. The K1-70 LC sequence obtained in the repeat round of experiments was consistent with the protein sequence of the 2-21 consecutive N-terminal amino acids obtained by Edman reaction (
X-Ray Diffraction Analysis of K1-70 Fab
K1-70 Fab solutions prepared as described above were concentrated to 15.5 mg/mL using iCON concentrators (ThermoFisher Scientific, Loughborough, UK) with a 9000 Da cut off and stored at −20° C. in aliquots. Crystals of K1-70 Fab were grown using the hanging drop method of vapour diffusion using the Structure Screen 1 sparse matrix screen from Molecular Dimensions (Newmarket, UK). Several crystals were obtained in a number of conditions and all were screened to identify the crystal most suitable for the X-ray diffraction analysis at Biofocus DPI (Saffron Walden, UK). A crystal grown in 30% PEG 400, 0.1 M sodium Hepes pH 7.5, 0.2 M magnesium chloride was chosen. It was washed in well solution and flash frozen by plunging into liquid nitrogen. The data set was collected on a Rigaku R-Axis IV image plate detector and was indexed, integrated and scaled using MOSFLM and SCALA (from CCP4 program suite (Collaborative computational project, number 4. 1994. “The CCP4 Suite: Programs for Protein Crystallography”. Acta Cryst. D50, 760-763). Three structures from the Protein Data Bank (http://www.rcsb.org/pdb/home/home.do) 1LIL (VL and CL domains), 2B0S (VH domain) and 2EH7 (VL domain) were chosen for use in molecular replacement, based on sequence alignment. There were two complete Fab K1-70 molecules in the asymmetric unit and the resulting model was given ten cycles of atomic refinement with tight geometric weights using REFMAC5 (CCP4). The electron density maps calculated after molecular replacement and initial refinement were examined in the model building program COOT (Emsley P, Cowtan K 2004. Nature 355: 472-475) and automated model rebuilding was performed using BUCCANEER (CCP4). The model was reexamined and any remaining missing features were built by hand and the model was refined using REFMAC5 (CCP4). Water molecules were then added using the water placement option in COOT and refined using REFMAC5 (CCP4). The structural geometry of Fab K1-70 was checked using PROCHECK (CCP4) and RAMPAGE (CCP4). Finally, the residues in the model were renumbered in accordance with the Kabat numbering system (Kabat E et al 1991 supra).
Cloning and Expression of Recombinant K1-70 Fab in E. coli
The K1-70 HC RT-PCR product was cut with XhoI and SpeI restriction endonucleases and the K1-70 LC PCR product was cut with SacI and XbaI restriction endonucleases and both HC and LC cDNAs cloned into the Immunozap H/L vector (Stratagene Europe; Amsterdam, Netherlands) (Matthews I, et al 2002 Laboratory Investigation 82: 1-11) under the control of the lacZ promoter. Plasmid DNA was prepared using the Qiagen midi plasmid purification kit (Qiagen Ltd, Crawley, UK) and the presence of K1-70 HC and LC cDNAs confirmed by sequencing using the Sanger-Coulson method (Sanger F, et al 1977 supra). Plasmid DNA was transformed into the E coli strain HB2151 (GE Life Sciences, Little Chalfont, UK) and grown overnight at 37° C. on LB ampicillin (Tryptone 10 g/L, Yeast Extract 5 g/L, NaCl 10 g/L, 100 μg/mL final concentration ampicillin) agar plates (15 g/L agar). Precultures (one colony in 3 mL LB ampicillin+1% glucose) were grown overnight at 30° C. with shaking. Production of the recombinant Fab is inhibited in the presence of glucose. Precultures after overnight incubation were diluted 1/100 (0.5 mL in 50 mL LB ampicillin) and grown at 30° C. until the OD600 was 1.2 followed by addition of sucrose (final concentration 0.3 mol/L) and culture grown at 30° C. until OD600 returned to 1.2. Thereafter isopropyl-β-D thiogalactoside (IPTG) was added to a final concentration of 1 mmol/L and cultures continued to be incubated for 24 hours at 23° C. with shaking. The cultures were then centrifuged at 3000 rpm for 30 minutes at 4° C. and the culture supernatants recovered. The culture supernatants were filtered through a 0.45 μm filter and dialysed overnight into PBS (8.1 mmol/L Na2HPO4, 1.5 mmol/L KH2PO4, 2.7 mmol/L KCl, 137 mmol/L NaCl pH 7.4). Culture supernatant from HB2151 cells transformed with K1-70 plasmid (HB2151/K1-70) grown with glucose without IPTG ie non-induced were used as negative controls. The culture supernatants were assayed for (a) their ability to inhibit TSH binding to the TSHR and (b) their ability to inhibit TSH mediated stimulation of cyclic AMP production in CHO cells expressing TSHR.
Isolation and Cloning of Stable Cell Lines Secreting K1-18 or K1-70
Lymphocytes (26×106) obtained from 20 mL of patient's blood were infected with EBV and plated out at 1×106 cells per well in a 48 well plate on feeder layers of mouse macrophages. On day 13 post EBV infection the plate well supernatants were monitored for inhibition of 125I-TSH binding. Positive clones were tested further for their effects (stimulating or blocking) on the TSHR. Cells from positive wells (positive in any of the assays used) were expanded and fused with the K6H6/B5 hybridoma cell line and plated out in 96 well plates. Two clones stably producing antibodies with 125I-TSH binding inhibiting activity were obtained and re-cloned 4 times. One of the clones secreted a human MAb designated as K1-18 that had TSHR stimulating activity. K1-18 antibody purified from the hetero-hybridoma culture supernatants was subclass IgG1 with kappa light chains. The other stable clone secreted a human MAb designated as K1-70 that had the ability to block TSH stimulation of cyclic AMP production in TSHR transfected CHO cells. K1-70 antibody purified from hetero-hybridoma culture supernatants was subclass IgG1 with lambda light chains.
Inhibition of 125I-TSH Binding to TSHR Coated Tubes
The ability of different concentrations of K1-18 or K1-70 IgGs to inhibit binding of labelled TSH to TSHR coated tubes is shown in Tables 1a and 1b. As shown in Table 1a, K1-18 IgG diluted in healthy blood donor (HBD) serum showed maximum inhibition of 125I-TSH binding of approx 95% at 1 μg/mL concentrations. The inhibiting effect of K1-18 IgG at concentrations between 1-0.001 μg/mL was dose dependent. The inhibiting effect of K1-18 was comparable to the effect of M22 IgG at the same concentrations. K1-18 IgG at 1 μg/mL is a more potent inhibitor of 125I-TSH binding than 5C9 IgG, TSMAb 1-7 IgGs or 9D33 IgG (Table 1a). K1-70 IgG or Fab inhibiting effects on 125I-TSH binding are shown in Table 1b. K1-70 IgG diluted in HBD serum showed dose dependent inhibition ranging from 13.5±2.3% at 0.03 μg/mL to 95.9±0.8% at 100 μg/mL. The inhibiting effects of K1-70 Fab was comparable to the effects of K1-70 IgG at the same concentrations (Table 1b). Tables 1a and 1b also show the effects on 125I-TSH binding to TSHR coated tubes by K1-18 and K1-70 and different MAbs IgGs diluted in the coated tube assay buffer; in the case of all MAbs except 5C9 these effects were comparable to the results observed when MAbs were diluted in HBD serum. Table 2a shows inhibition of 125I-TSH binding to TSHR coated tubes by different preparations of K1-18, K1 donor serum and K1-donor serum IgG. In this experiment, approximately 12% inhibition was observed with as little as 0.01 μg/mL of K1-18 IgG diluted in HBD sera and the inhibition increased in a dose dependent manner up to 95% inhibition at 10 μg/mL of K1-18 IgG. K1-18 Fab at 0.01 μg/mL in HBD sera showed 5.6±7.3% inhibition and the inhibition increased in a dose dependent manner to a maximum inhibition of 82.2±0.9% at 10 μg/mL. This can be compared to 125I-TSH binding inhibition by donor serum IgG diluted in HBD sera; 13.7±1.3% inhibition at 0.125 mg/mL increasing in a dose dependent manner to 76.5±1.5% inhibition at 1 mg/mL. Donor serum at different dilutions also showed dose dependent inhibition of 125I-TSH binding; 9.1±0.8% inhibition and 1/160 dilution to 81.1±0.4% inhibition at 1/10 dilution. The data in Table 2a showed that purified K1-18 IgG was 6600 times more active in terms of inhibition of TSH binding compared to K1 donor serum IgG. When K1-18 IgG and donor serum IgG were diluted in assay buffer the ability of K1-18 IgG to inhibit TSH binding was 4700 times greater than that of the donor serum IgG (Table 2a). Table 2b shows inhibition of 125I-TSH binding to the TSHR in coated tube assay by different preparations of K1-18 compared to the effect of Thyroid Stimulating Autoantibody reference preparation 90/672 from National Institute for Biological Standards and Control (NIBSC; Potters Bar, UK). K1-18 IgG diluted in HBD serum showed 125I-TSH binding inhibiting activity of 69 NIBSC 90/672 units/mg (mean of activity calculated at three concentrations of K1-18 IgG; 30 ng/mL, 100 ng/mL and 300 ng/mL) (Table 2b). 125I-TSH binding inhibiting activity of K1-18 Fab (diluted in serum) calculated in the same experiment was 46 NIBSC 90/672 units/mg (activity at 30 ng/mL, 100 ng/mL and 300 ng/mL of K1-18 Fab was used for the calculations) (Table 2b). This can be compared with M22 IgG 125I-TSH binding inhibiting activity of 131 NIBSC units/mg (Table 2b). 125I-TSH binding inhibiting activities of dilutions of donor serum and donor serum IgG compared to the activity of NIBSC 90/672 are shown in Table 2c. 125I-TSH binding inhibiting activity of the donor serum was 0.075 NIBSC 90/672 units/mL (mean of values at 40× and 20× dilutions) and of donor serum IgG diluted in HBD serum was 0.011 units/mg (mean of values at 0.1; 0.3 and 1.0 mg/mL) (Table 2c). This can be compared to the activity of K1-18 IgG (diluted in HBD serum) measured in the same experiment of 63.3 NIBSC 90/672 units/mg (mean of values at 30, 100 and 300 ng/mL) and the activity of K1-70 IgG (diluted in HBD serum) of 114 units/mg (mean of values at 10, 30 and 100 ng/mL) (Table 2c). Consequently in this assay system the specific activity of K1-18 IgG was 5755× that of the donor serum IgG. Similarly, the specific activity of K1-70 IgG was 10,364× that of the donor serum IgG.
Scatchard Analysis of K1-18 and K1-70 Binding to TSHR Coated Tubes
The binding affinity of K1-18 IgG for the TSHR (full length) was 6.7±1.0×109 L/mol (mean±SD; n=3) while binding affinity of K1-18 Fab was 1.8±1.0×109 L/mol (mean±SD; n=3). Binding affinity of K1-18 IgG for the TSHR260 was 5.9±1.0×109 L/mol (mean±SD; n=3). K1-70 IgG binding affinity for the TSHR (full length) was 3.9±0.8×1010 L/mol (mean±SD; n=3) while binding affinity of K1-70 Fab was 2.3±0.3×1010 L/mol (mean±SD; n=3). Binding affinity of K1-70 IgG for the TSHR260 was 3.1±0.4×1010 L/mol (mean±SD; n=3) and of K1-70 Fab it was 9.3±0.4×109 L/mol (mean±SD; n=3). This can be compared to binding affinity of porcine TSH to the TSHR (full length) of 6.0±0.9×109 L/mol (mean±SD, n=5) (Nakatake et al 2006 supra).
Inhibition of TSH-Biotin Binding to the TSHR Measured by ELISA
The effects of K1-18 IgG on TSH-biotin binding to TSHR coated ELISA plate wells was studied and compared to the effects of various other MAbs. As shown in Table 3a K1-18 IgG diluted in HBD serum had a dose dependent inhibiting effect on TSH-biotin binding with 10.0±0.8% inhibition at 0.01 μg/mL, essentially maximum inhibition of 96.2±0.2% at 1 μg/mL and a maximum inhibition plateau at concentrations of 3 μg/mL and above. This can be compared to M22 IgG (diluted in HBD serum) inhibiting effect of 17.5±2.0% at 0.01 μg/mL and 98.3±0.0% at 1 μg/mL (Table 3a). TSH-biotin binding inhibiting activity of K1-18 IgG at 1 μg/mL (diluted in HBD serum) was greater than 5C9 IgG, TSMAb 1-7 IgGs and 9D33 IgG as illustrated by the examples shown in Table 3a. When K1-18 IgG was tested diluted in ELISA assay buffer (50 mmol/L NaCl, 10 mmol/L Tris pH 7.8, 0.1% v/v Triton X-100, 1 mg/mL BSA) the inhibiting effects were essentially the same as when the dilutions were made in HBD serum (Table 3a).
As shown by the examples in Table 3b K1-18 Fab diluted in HBD serum or in ELISA assay buffer was also an effective inhibitor of TSH-biotin binding in the ELISA. The inhibiting effects of K1-18 IgG diluted in ELISA assay buffer with addition of control MAb IgG (5B3 which is a human MAb to glutamic acid decarboxylase) at 100 μg/mL is shown in Table 3c. When diluted in the buffer containing control MAb K1-18 IgG showed similar TSH-biotin binding inhibition activity as when diluted in buffer containing BSA or in HBD serum (Table 3c). Consequently, the presence of an unrelated human MAb IgG at high concentration (100 μg/mL) had no effect on the inhibiting activity of K1-18 IgG nor M22 IgG nor 5C9 IgG. Table 3d shows the effects of K1-70 on TSH-biotin binding to the TSHR and these are comparable to the effects of K1-18 or M22 (Tables 3a & 3b). K1-70 IgG diluted in HBD serum had a dose dependent inhibiting effect on TSH-biotin binding with 13.6±1.4%; 74.1±0.4% and 97.4±0.2% inhibition at 0.01 μg/mL; 0.1 μg/mL and 1 μg/mL, respectively. K1-70 Fab was similarly active with inhibitions of 18.2±0.6%; 88.3±0.3% and 96.9±0.1% at 0.01 μg/mL; 0.1 μg/mL and 1 μg/mL, respectively. When K1-70 IgG or Fab preparations were diluted in ELISA assay buffer the inhibiting activities were essentially the same compared to dilutions made in HBD serum (Table 3d). The ability of K1-18 IgG to inhibit binding of M22 Fab-POD to TSHR coated ELISA plate wells is shown in Table 4a. K1-18 IgG diluted in HBD serum inhibited M22 Fab-POD binding in a dose dependent manner; in particular 21.0±1.3%, 81.6±0.5% and 97.2±0.1% inhibitions were observed at 0.03 μg/mL, 0.3 μg/mL and 3 μg/mL, respectively. This effect was comparable to the inhibiting effect of M22 IgG (diluted in HBD serum) of 51.0±2.4%, 93.2±0.3% and 98.0±0.2% at 0.03 μg/mL, 0.3 μg/mL and 3 μg/mL, respectively. K1-18 Fab showed similar ability to inhibit M22 Fab-POD binding as K1-18 IgG (Table 4b). As shown in Table 4a, K1-18 and M22 ability to inhibit labelled M22 binding to the TSHR were greater than the inhibiting activities of 5C9, TSMAbs 1-7 and 9D33. Inhibiting effects of all MAbs studied when diluted in ELISA assay buffer were similar to those observed when diluted in HBD serum (Table 4a and b). The inhibiting effects of K1-18 on M22 Fab-POD binding to the TSHR can be compared to the effects of K1-70 (Table 4c). K1-70 IgG diluted in HBD serum showed 34.5±3.8%, 91.1±0.3% and 97.6±0.1% inhibition at 0.03 μg/mL, 0.3 μg/mL and 3 μg/mL, respectively. Similar percentage inhibitions were observed when K1-70 IgG was diluted in ELISA assay buffer (Table 4c). As illustrated in Table 4c K1-70 Fab diluted in HBD serum or in ELISA assay buffer showed similar M22-POD binding inhibiting activity as K1-70 IgG.
Inhibition of 125I-Labelled K1-18 IgG or Fab Binding to TSHR Coated Tubes
In the presence of a control human MAb 4B4 IgG at concentrations from 0.01-100 μg/mL (diluted in HBD serum) binding of 125I-K1-18 IgG was essentially not affected (Table 5a). This can be compared with the effects of different concentrations of unlabelled K1-18 IgG (diluted in HBD serum); increasing doses of 0.001; 0.01; 0.1 and 1.0 μg/mL caused inhibition of binding of 125I-labelled K1-18 of 11.1±4.4%, 22.9±2.4%, 69.0±0.5% and 91.7±0.8%, respectively. Unlabelled K1-18 Fab tested at concentrations from 0.001-100 μg/mL showed inhibitions ranged from 10.3±2.2% (at 0.03 μg/mL) to 84.8±0.9% (at 100 μg/mL) (Table 5b). K1-70 IgG and Fab (both tested in the range of concentrations from 0.001-100 μg/mL) also inhibited 125I-K1-18 IgG binding in a dose dependent manner to an essentially complete inhibition of 95.1±0.3% at 10 μg/mL of K1-70 IgG and 92.8±1.1% at 3 μg/mL of K1-70 Fab (Table 5b). In addition, binding of 125I-K1-18 IgG was inhibited in a dose dependent manner by M22 IgG, M22 Fab, 5C9 IgG, TSMAb 1-7 IgGs and 9D33 IgG (Table 5a and 5b). When the same experiments were carried out using various MAb preparations diluted in coated tube assay buffer the inhibiting effects of the respective preparations were comparable to the effect observed when diluted in HBD serum except in the case of 5C9 IgG (Table 5a). In the case of 5C9 diluted in assay buffer the maximum inhibition at 100 μg/mL was 91.3±0.4% compared to 57.7±2.4% when diluted in HBD serum and the inhibitions at 0.01 μg/mL were 11.7±1.8% and −1.8±2.7%, respectively (Table 5a). Binding of 125I-K1-18 IgG to TSHR coated tubes was inhibited by the lymphocyte donor serum resulting in 35.2% and 59.3% inhibition at serum dilutions of 1:20 and 1:10, respectively (Table 5c). Sera from 20 Graves' patients inhibited binding of 125I-K1-18 IgG and the inhibiting effect was comparable to the inhibiting effect on 125I-TSH binding (Table 5c). Table 5c also shows the effect of dilutions of sera from two patients with blocking TRAbs (B1 and B2) and sera from two patients with stimulating TRAbs (S1 and S2) on both 125I-K1-18 IgG and 125I-TSH binding. Effect of various MAbs on 125I-K1-18 Fab binding to the TSHR coated tubes is shown in Table 5d. Both, unlabelled K1-18 IgG and K1-18 Fab had dose dependent inhibiting effect on 125I-K1-18 Fab binding and these effects were comparable to the effects of M22 IgG, M22 Fab and K1-70 IgG (Table 5d). 5C9 IgG, TSMAbs 1-7 IgGs and 9D33 IgG also inhibited on 125I-K1-18 Fab binding, however, their effects were smaller compared to M22, K1-18 and K1-70 preparations (Table 5d).
Stimulation of Cyclic AMP Production in CHO Cells Expressing the TSHR
K1-18 IgG stimulated cyclic AMP production in CHO cells expressing the TSHR in a dose dependent manner as shown in Table 6a. In hypotonic buffer, the levels of cyclic AMP in the presence of 0.1 ng/mL K1-18 IgG were 1.56±0.32 pmol/L, at 1.0 ng/mL were 4.08±0.28 pmol/L, at 10 ng/mL were 31.66±5.06 pmol/L, at 100 ng/mL were 64.95±9.61 pmol/L and at 1000 ng/mL they were 67.90±10.44 pmol/L. The cyclic AMP levels at different concentrations of K1-18 Fab in hypotonic buffer were 1.72±0.82 pmol/L, 9.99±3.52 pmol/L, 53.22 pmol/L and 66.94±6.93 pmol/L at 1 ng/mL, 10 ng/mL, 100 ng/mL and 1000 ng/mL of K1-18 Fab, respectively. M22 Fab at 1 ng/mL in hypotonic buffer stimulated cyclic AMP production of 29.80±0.97 pmol/L and at 10 ng/mL of 57.41±5.05 pmol/L (Table 6a). Table 6a also shows the effect of K1-18 IgG or Fab on cyclic AMP stimulation in CHO cells expressing the TSHR tested under the isotonic conditions. As illustrated by examples in Table 6a both K1-18 and K1-70 caused increase of cyclic AMP production under isotonic conditions although the levels of cyclic AMP produced were lower compared to experiments using hypotonic conditions. Comparison of stimulating activity of M22 IgG and K1-18 IgG tested in hypotonic buffer is shown in Table 6b. At 3 ng/mL concentration M22 IgG stimulated 24.3±2.3 pmol/mL of cyclic AMP while K1-18 IgG 8.3±0.5 pmol/mL. At 10 ng/mL M22 IgG and K1-18 IgG caused stimulation of 50.3±1.6 and 25.0±1.0 pmol/mL of cyclic AMP respectively and at 100 ng/mL 64.6±1.9 and 62.6±2.7 pmol/mL respectively. The stimulating activity of K1-18 IgG and Fab was also assessed relative to the activity of NIBSC reference preparation 90/672 (Table 6c). Calculated cyclic AMP stimulating activity of K1-18 IgG was 155 NIBSC 90/672 units/mg (mean of activity calculated at three concentrations of K1-18 IgG; 1 ng/mL, 3 ng/mL and 10 ng/mL) (Table 6c). Cyclic AMP stimulating activity of K1-18 Fab calculated in the same experiment was 22 NIBSC 90/672 units/mg (activity at 10 ng/mL, 30 ng/mL and 100 ng/mL of K1-18 Fab were used for the calculations) (Table 6c). This can be compared with M22 IgG cyclic AMP stimulating activity of 286 NIBSC units/mg (Table 6c). For comparative purposes the stimulating activities of porcine TSH, native human TSH and recombinant human TSH in isotonic and in hypotonic buffers are shown in Table 6d.
Further examples shown in Table 6e concern the stimulating effects of K1-18 IgG, M22 IgG or pTSH when mixed together in different combinations. The stimulating effect of pTSH, M22 or K1-18 appeared to be enhanced when two stimulators were mixed together compared to the effect of the stimulator alone at the same concentrations. In particular, cyclic AMP production of 11.01±0.99 pmol/mL (mean±SD) at 0.1 ng/mL pTSH alone and 35.17±6.38 pmol/mL (mean±SD) at 1 ng/mL of M22 IgG increased to 47.22±3.89 pmol/mL (mean±SD) when 0.1 ng/mL pTSH and 1 ng/mL of M22 IgG were mixed together. Also a mixture of 0.1 ng/mL of pTSH and 10 ng/mL of K1-18 IgG had a greater stimulating effect than these stimulators alone (Table 6e). Furthermore, two stimulating antibodies mixed together were more potent than a single antibody at the same concentrations. For example, 29.95±1.18 pmol/mL (mean±SD) of cyclic AMP was produced in response to 5 ng/mL of K1-18 IgG, 20.20±2.48 pmol/mL (mean±SD) was produced in response to 0.5 ng/mL of M22 IgG while 44.01±7.19 pmol/mL (mean±SD) cyclic AMP was produced in response to 5 ng/mL of K1-18 and 0.5 ng/mL of M22 mixed together (Table 6e).
The results of two experiments in which the ability of the K1-18 and K1-70 donor serum and donor serum IgG to stimulate cyclic AMP were compared to the stimulating activity of NIBSC 90/672 are shown in Tables 6f and 6g. In experiment 1, the stimulating activity of donor serum was 4.7±0.1 pmol/mL of cyclic AMP at 30 times dilution compared to the effect of HBD serum at the same dilution of 1.7±0.4 pmol/mL while the stimulating activity of donor serum IgG was 7.7±1.0 pmol/mL at 30 μg/mL which represented activity relative to NIBSC 90/672 of 0.013 units/mg (Table 6f). In experiment 2, donor serum diluted 30 times caused stimulation of cyclic AMP to 9.5±0.7 pmol/mL while donor serum IgG at 30 μg/mL caused stimulation to 15.6±0.7 pmol/mL which represented activity relative to NIBSC 90/672 of 0.014 units/mg (Table 6g). K1-18 IgG TSHR stimulating activity was inhibited by human MAbs with TSH antagonist activity (K1-70 and 5C9) as illustrated by the examples shown in Table 6h. In particular, K1-18 IgG at 10 ng/mL caused stimulation of cyclic AMP to 50.0±3.3 pmol/mL and this was reduced to 3.8±1.0 pmol/mL in the presence of 0.1 μg/mL of K1-70 IgG (92% inhibition). In the presence of 10 ng/mL of K1-18 IgG and 0.1 μg/mL of 5C9 IgG cyclic AMP levels were 4.4±1.5 pmol/mL (91% inhibition). At higher concentrations of K1-70 IgG or 5C9 IgG the inhibiting effect was complete (100% inhibition) (Table 6h). In further experiments the effect of K1-70 IgG mixed together with 5C9 IgG on K1-18 IgG stimulating activity was studied (Table 6i). As shown in the Table 6i K1-18 IgG stimulation at 10 ng/mL was effectively inhibited by 0.1 μg/mL of 5C9 IgG or 0.1 μg/mL of K1-70 IgG. When K1-70 IgG and 5C9 IgG were mixed to give a final total concentration of 0.1 μg/mL the stimulating activity of K1-18 IgG was also effectively inhibited (97.3% inhibition). However at lower concentrations K1-70 IgG and 5C9 IgG when mixed together were more effective inhibitors of K1-18 IgG stimulating activity than one antibody alone. For example, at 0.001 μg/mL K1-70 IgG and 5C9 IgG individually caused no inhibition (0% and 1% respectively) while when mixed together to the same final concentration of total IgG (ie 0.001 μg/mL) the inhibition was 25.5%. Table 6i also shows that cyclic AMP concentration in the presence of K1-70 IgG (100 μg/mL) was similar to that observed in the presence of assay buffer while in the presence of 5C9 IgG (100 μg/mL) the concentration of cyclic AMP was lower (0.89±0.13; 0.89±0.15 and 0.55±0.14, respectively) (mean±SD of triplicate determinations). When K1-70 IgG and 5C9 IgG were mixed together (final concentration of 100 μg/mL of total IgG) cyclic AMP concentrations were not reduced below the levels observed in the presence of buffer (ie not lower than basal or constitutive activity levels). For comparison essentially complete inhibition of M22 IgG (3 ng/mL) stimulating activity was observed at 1 μg/mL of 5C9 IgG or 1 μg/mL of K1-70 IgG (97.1% and 96.6% inhibition respectively) and inhibitions at 0.1 μg//mL of 5C9 IgG or 0.1 μg/mL of K1-70 IgG were 92.8% and 75.5%, respectively (Table 6j). However, when 5C9 and K1-70 were mixed together to a final total IgG concentration of 0.1 μg/mL 91.9% inhibition was observed (Table 6j). The effects of mixtures of K1-70 IgG and 9D33 IgG on K1-18 IgG stimulating activity are shown in Table 6k. In the case of 9D33 IgG 95% inhibition was observed at 1 μg/mL while K1-70 IgG at 0.1 μg/mL showed the same inhibition (95% inhibition). When the two blocking MAbs (9D33 and K1-70) were mixed together to a final total IgG concentration of 0.1 μg/mL 95% inhibition was also observed (Table 6k). 9D33 IgG at 10 μg/mL was able to essentially completely inhibit M22 IgG cyclic AMP stimulation (94% inhibition) while lower concentrations of K1-70 IgG (1 μg/mL) had similar effect (96% inhibition) (Table 6l). Esentially complete inhibition of M22 activity (96% inhibition) was evident at 1 μg/mL of the 9D33 and K1-70 mixture (Table 6l). This is comparable to the inhibiting effect of a mixture of 9D33 and K1-70 (1 μg/mL) on TSH stimulating activity (97% inhibition) (Table 6m). However, it should be noted that TSH stimulating activity was inhibited more effectively by K1-70 IgG alone (98% inhibition at 1 μg/mL) than by 9D33 IgG alone (95% inhibition at 100 μg/mL) (Table 6m). Table 6n shows the effect of the lymphocyte donor serum and three patient sera containing TRAbs with blocking activity (B1-B3) on TSHR stimulating activity of TSH, M22 IgG and K1-18 IgG. The donor serum inhibited TSH, M22 IgG and K1-18 IgG stimulating activities (63.8%, 80.1% and 79.5% inhibitions, respectively). Three different sera with blocking TRAbs that had a strong inhibiting effect on TSH and M22 IgG stimulation also inhibited the stimulating activity of K1-18 IgG (Table 6n). The inhibiting effects of the different patient sera on TSH, M22 IgG or K1-18 IgG stimulating activities were comparable.
Measurement of Antagonist (Blocking) Activity
Incubation of CHO cells expressing the TSHR with porcine TSH at 3 ng/mL caused stimulation of cyclic AMP production to 62.6±3.9 pmol/mL (Table 7a). In the presence of increasing amounts of K1-70 IgG stimulating activity of porcine TSH was inhibited in a dose dependent manner. In particular in the presence of 0.01, 0.05, 0.1 and 1 μg/mL of K1-70 IgG the levels of cyclic AMP were 60.1±1.6, 31.4±1.9, 5.8±2.8 and 2.0±0.2 pmol/mL respectively which represent 4.0%, 49.8%, 90.7% and 96.8% inhibition respectively relative to the effect of control MAb IgG (5B3) (Table 7a). Table 7a also show the inhibiting effects of 5C9 IgG for comparison. The effects of K1-70 Fab on stimulating activity of porcine TSH under 2 different experimental conditions are shown in Table 7b. In the presence of K1-70 Fab at 1 μg/mL porcine TSH stimulating activity was essentially completely inhibited in both conditions (ie in isotonic and in hypotonic medium). The effect of K1-70 Fab was dose dependent in the range of concentrations studied (0.003 μg/mL to 3 μg/mL) with as little as 0.05 μg/mL of Fab showing an ability to reduce porcine TSH stimulation down to 28.9±1.1 pmol/mL cyclic AMP from 39.5±1.9 pmol/mL in the presence of 1 μg/mL of control MAb (under isotonic conditions) (Table 7b). The potency of K1-70 Fab under hypotonic conditions was similar to that observed under isotonic conditions (Table 7b).
Increasing concentrations of K1-70 IgG (range 0.001-100 μg/mL) did not show any ability to inhibit TSHR constitutive (basal) activity as illustrated by the examples shown in Table 7c. This contrasts with the effects of 5C9 IgG as shown in Table 7c for comparison. The blocking mouse antibody 9D33 tested in the same experiment had no ability to affect TSHR constitutive activity (Table 7c) and some weak stimulating activity (about 2× basal) was observed with high concentrations of 9D33.
Blocking activity of K1-70 IgG was compared to the blocking activity of the lymphocyte donor serum as shown in Table 7d. Cyclic AMP levels after incubation with porcine TSH at 1 ng/mL were 61.7±4.3 pmol/mL and the levels dropped in the presence of donor serum (10× dilution) to 14.9±1.2 pmol/mL (75.9% inhibition) and to 51.6±2.6 pmol/mL with serum diluted 20 times (16.4% inhibition). Donor serum at higher dilutions did not have a detectable effect on TSH stimulating activity. The effect of donor serum can be compared to the effect of K1-70 IgG that at 0.1 μg/mL had a similar effect as serum diluted 10 times (67.6% and 75.9% inhibition respectively) (Table 7d). K1-70 IgG had the ability to block cyclic AMP stimulating activity of porcine TSH, human TSH and human recombinant TSH as illustrated by the examples in Table 7e. K1-70 IgG at 0.1 μg/mL was an effective blocker of stimulating activity of all three TSH preparations tested under hypotonic medium conditions. Blocking activity of K1-70 IgG was less effective under isotonic medium conditions (Table 7e). Effects of K1-70 IgG on M22 IgG mediated stimulation of cyclic AMP production in CHO cells expressing the TSHR are shown in Table 7f. Cyclic AMP levels observed at 3 ng/mL of M22 IgG were 33.1±1.8 pmol/mL and these decreased in the presence of K1-70 IgG for example, 4.3±2.4 pmol/mL at 0.1 μg/mL (87% inhibition) (Table 7f). The effects of K1-70 IgG were comparable to the effects of 5C9 IgG tested in the same experiment (Table 7f). Furthermore, K1-70 IgG showed the ability to block cyclic AMP stimulating activity of TRAbs in sera from patients with Graves' disease and the examples in Tables 7g-7k illustrate that 100 μg/mL concentrations of K1-70 IgG caused complete inhibition of stimulating activity in all 15 sera studied (inhibition of T1-T15 ranged from 90.8% to 98.7%). The effects of K1-70 IgG on the stimulating activities of sera T1-T15 were comparable to the effects of 5C9 IgG and 9D33 IgG tested in the same experiment except for one serum ie T11 (Table 7j). Stimulating activity of serum T11 was only weakly inhibited by 100 μg/mL of 5C9 IgG (8.5% inhibition) while in the presence of 100 μg/mL of K1-70 the inhibition was essentially complete (95.1% inhibition) (Table 7j). Effective inhibition (87.1%) of T11 was also observed with 100 μg/mL 9D33 (Table 7j). The effect of different blocking MAbs at different concentrations (0.01-100 μg/mL) on the stimulating activities of three Graves' sera (including serum T11) is shown in more detail in Tables 7l-7n. These experiments showed that K1-70 IgG, 5C9 IgG and 9D33 IgG are effective inhibitors at concentrations as low as 0.1 μg/mL except in the case of serum T11 on which 5C9 IgG had little or no effect (Table 7n). Table 7o shows inhibition of porcine TSH stimulation by K1-70 IgG and 5C9 IgG when the two blocking MAbs were mixed together in one experiment. These experiments showed that the two blocking MAbs were effective in combination in their ability to inhibit TSH stimulation of cyclic AMP production. In addition the effect of K1-70 IgG and 5C9 IgG mixed together on the constitutive activity of the TSHR was tested. As shown in Tables 7c and 7p K1-70 IgG had no effect on TSHR basal activity in contrast to 5C9 IgG. When K1-70 and 5C9 IgGs were mixed together to give a final IgG concentration of 2 μg/mL cyclic AMP levels dropped slightly from 58.04±8.52 pmol/mL (mean±SD, n=3) in the presence of buffer only to 55.28±6.17 pmol/mL (mean±SD, n=3) ie 4.8% inhibition (Table 7p). However, when 5C9 IgG was mixed with 5B3 IgG (control antibody to glutamic acid decarboxylase) to give a final IgG concentration of 2 μg/mL constitutive activity of the TSHR was inhibited to 52.1% of basal values (cyclic AMP level 27.78±2.96 pmol/mL; mean±SD, n=3) (Table 7p). These experiments show that in the presence of K1-70 IgG, 5C9 IgG is unable to act as an effective inhibitor of TSHR constitutive activity.
Effect of the TSHR Mutations on K1-18 Stimulating Activity
The effect of K1-18 IgG on stimulation of cyclic AMP production was tested using CHO cells expressing TSHRs with the following amino acid mutations: Lys58Ala, Arg80Ala, Tyr82Ala, Glu107Ala, Arg109Ala, Lys129Ala, Phe130Ala, Phe134Ala, Lys183Ala, Asp203Ala, Arg255Asp (Table 8a-k and summarised in Table 10). Mutation of TSHR amino acids Lys58, Arg80, Tyr82, Glu107, Arg109, Lys129, Phe130, Phe 134 and Asp203 to alanine had no effect on K1-18 IgG's ability to stimulate cyclic AMP production. The ability of K-18 IgG to stimulate cyclic AMP production was lost completely with CHO cells expressing TSHR containing mutations Lys183Ala and Arg255Asp and cyclic AMP concentrations in response to K1-18 IgG were similar to the concentrations observed in the presence of cyclic AMP buffer only (Table 8i and 8k). However, responsiveness to TSH was retained with the Lys183Ala and Arg255Asp mutations. In an additional series of experiments the effects of mutations of various amino acids of the TSHR on K1-18 IgG cyclic AMP stimulating activity was tested further (Tables 14a-14v and summarised in Table 16). Mutations (to alanine) of TSHR residues Asp43, Ile60, Glu61, Thr104, His105, Lys250, Arg255, Thr257, Asp276 and Ser281 had no effect on K1-18 IgG's ability to stimulate cyclic AMP production. Mutations of TSHR Asp151, Glu178, Lys209, Gln235, Glu251 to alanine caused a small reduction of K1-18 IgG stimulating activity, however, these mutations also affected TSH stimulating activity therefore the interactions with these TSHR residues were not considered specific for K1-18. In contrast, mutations of TSHR Glu157Ala, Lys183Asp, Tyr185Ala and Asp232Ala resulted in loss of the ability of K1-18 IgG to stimulate cyclic AMP (less than 20% of the wild type activity; Tables 14g, 14i, 14j, 14m). Furthermore, the ability of K1-18 IgG to stimulate the TSHR mutated at Tyr206, Trp258 and Arg274 to alanine was reduced to approximately 40-60% of the wild type activity (Tables 14k, 14s, 14t).
Effect of the TSHR Mutations on K1-70 Blocking Activity
The effect of K1-70 IgG on TSH stimulation of cyclic AMP production was tested in CHO cells expressing TSHRs containing mutations of the following amino acids to alanine: Lys58, Arg80, Tyr82, Glu107, Arg109, Lys129, Phe130, Phe134, Lys183 and Asp203. In addition the effect of TSHR mutation Arg255Asp was tested. (Table 9 a-k and summarised in Table 10). Mutation of TSHR amino acids Arg80, Glu107, Lys129, Phe130, Phe134 and Asp203 to alanine had no effect on K1-70 IgG's ability to inhibit TSH stimulated cyclic AMP production. The ability of K1-70 to inhibit TSH stimulated cyclic AMP production was reduced by mutation of TSHR Lys58, Tyr82, Arg109 and Lys183 to alanine. The mutation Lys58Ala had the greatest effect (Table 9a) followed by Arg109Ala, Lys183Ala and Tyr82Ala that had the smallest effect (Tables 9e, 9i, 9c respectively and Table 10). However, none of the mutations studied caused a complete loss of K1-70 IgG ability to block TSH stimulated cyclic AMP production. In an additional series of experiments the effects of mutations of various amino acids of the TSHR on K1-70 IgG blocking activity was tested further (Tables 15a-15v and summarised in Table 16). The ability of K1-70 to inhibit TSH stimulated cyclic AMP production in TSHR transfected CHO cells was not affected by the TSHR mutations Asp43, Thr104, His105, Asp151, Tyr185, Tyr206, Lys209, Asp232, Gln235, Glu251, Arg255, Thr257, Trp258, and Arg274 to alanine nor mutations Asp160Lys and Lys183Asp. Mutations (to alanine) at TSHR Glu178 and Ser281 had a small effect on the ability of K1-70 IgG to block the stimulating activity of TSH (80-100% of the wild type activity; Tables 15h and 15v). TSHR mutations Glu61, Lys250 and Asp276 to alanine caused some effect (60-80% of the wild type) on K1-70 IgG blocking activity (Tables 15c, 15o, 15u) while TSHR mutation Ile60Ala caused reduction of K1-70 IgG blocking activity to 40-60% of the wild type activity (Table 15b).
Inhibition of 125I-labelled K1-70 IgG or Fab binding to the TSHR Binding of 125I-K1-70 IgG to TSHR coated tubes was inhibited in a dose dependent manner by unlabelled K1-70 IgG and at concentrations of 0.003 μg/mL, 0.03 μg/mL, 0.3 μg/mL and 3 μg/mL (diluted in HBD serum) the inhibitions were 10.2±2.4%; 36.5±1.9%; 84.4±0.8% and 92.0±0.5%, respectively (Table 11 a). As shown in Table 11a binding of 125I-K1-70 IgG was inhibited in a very similar way by M22 IgG (diluted in HBD serum); 6.2% (mean of duplicate determinations); 33.8±0.9%; 84.6±1.1% and 91.6±0.5% inhibitions at 0.003 μg/mL, 0.03 μg/mL, 0.3 μg/mL and 3 μg/mL concentrations. M22 Fab showed a greater potency to inhibit 125I-K1-70 IgG binding at lower concentrations with the inhibitions of 16.7±6.0%; 60.2±1.8%; 89.9±01% and 92.0±0.3% at 0.003 μg/mL, 0.03 μg/mL, 0.3 μg/mL and 3 μg/mL concentrations (in HBD serum) (Table 11a). Also, K1-18 IgG and Fab inhibited 125I-K1-70 IgG binding to the TSHR in a dose dependent manner (Table 11b). Dilutions of K1-18 IgG at 0.03 μg/mL, 0.3 μg/mL and 3 μg/mL (in HBD serum) showed 20.0±1.9%, 73.9±0.6% and 91.0±0.3% inhibition respectively and K1-18 Fab at 0.03 μg/mL, 0.3 μg/mL and 3 μg/mL (in HBD serum) showed 10.2±2.1%, 60.9±1.1% and 80.5±0.6% inhibition respectively (Table 11b). In contrast higher concentrations of 5C9 IgG were required to inhibit 125I-K1-70 IgG binding; 1 μg/mL was needed for 15.4±3.4% inhibition and 100 μg/mL showed 68.5±0.7% inhibition (Table 11a). Tables 11a and 11b also show the effects of various human MAbs on 125I-K1-70 binding in assay buffer. The effects are stronger when dilutions were made in assay buffer compared with HBD sera. The effects of different mouse monoclonal antibodies to the TSHR with TSHR stimulating activity (TSMAbs 1-7) and with TSHR blocking activity (9D33) on 125I-K1-70 IgG binding to the TSHR are shown in Table 11c. All TSMAbs tested had the ability to inhibit 125I-K1-70 IgG binding and 100 μg/mL concentrations (in HBD serum) caused inhibition ranging from 33.5±3.7% (TSMAb 3) to 59.6±0.6% (TSMAb 5) (Table 11c). 9D33 IgG at 100 μg/mL (in HBD serum) showed 51.1±1.7% inhibition (Table 11c). When these MAbs were diluted in assay buffer the % inhibitions were slightly higher compared with HBD sera in some cases and slightly lower in others (Table 11c). Further experiments showed that 125I-K1-70 Fab binding to the TSHR was also effectively inhibited by K1-18 IgG, K1-18 Fab, M22 IgG, M22 Fab, K1-70 IgG, K1-70 Fab, and mouse TSMAbs in a dose dependent manner (Table 11d). Graves' patient sera n=20) containing TSHR autoantibodies with 125I-TSH binding inhibition activity in the range from 15.9% to 80.0% (in the TSHR coated tube assay) also showed the ability to inhibit 125I-K1-70 IgG or Fab binding to the TSHR (range 19.2% to 77.6% and 15.9% to 72.8%, respectively) (Table 11e). The % inhibitions for the three labelled ligands tested were comparable in the case of each serum (Table 11e). None of the HBD sera (n=10) had an effect on 125I-labelled TSH, K1-70 IgG or Fab binding (Table 11e). Table 11e shows examples of experiments using two sera with TSHR stimulating activity (S1 and S2) and two sera with TSHR blocking activity (B1 and B2). Sera with either type of activity inhibited 125I-labelled TSH, K1-70 IgG or Fab binding in a dose dependent manner and the degree of inhibition of binding of the different ligands was comparable for each serum (Table 11e).
Kinetics of K1-18 and K1-70 Binding to the TSHR
125I-labelled K1-70 IgG and Fab binding to TSHR (full length) coated tubes at room temperature reached a maximum after 180 min (45.5% and 37.1% binding respectively). 50% maximal binding occurred after about 35 min (
Effects of K1-18 or K1-70 IgG in an ELISA Based on TSHR260-AP
The ability of K1-18 IgG to form a “bridge” between the full length TSHR immobilised on ELISA plate wells and TSHR260-AP in the liquid phase is illustrated by the examples shown in Table 12a. OD405 nm values increased in a dose dependent manner with increasing concentrations of K1-18 IgG (diluted in HBD sera). In particular, OD405 nm values were 0.013, 0.191, 0.511, 0.660 and 0.706 at 0.005, 0.05, 0.5, 10 and 100 μg/mL K1-18 IgG respectively compared to OD405 nm of −0.002 in the presence of HBD serum alone. K1-70 IgG (diluted in HBD sera) also bound well in the bridging ELISA and showed OD 405 nm values of 0.045, 0.290, 0.661, 0.738 and 0.794 at 0.005, 0.05, 0.5, 10 and 100 μg/mL K1-70 IgG concentrations respectively (Table 12a). The effects of K1-18 and K1-70 IgGs can be compared to the ability of M22 IgG to bind to TSHR260-AP preparations as shown in Table 12a. In the assay, increasing doses of M22 IgG (ranging from 0.005 μg/mL to 10 μg/mL diluted in HBD sera) bound increasing amounts of the TSHRs with the OD405 nm values ranging between 0.045 and 0.796. When dilutions of MAbs were made in ELISA assay buffer rather than HBD sera, absorbances at 450 nm were higher particularly in the case of 5C9 (Table 12a). The principle of the “bridge type” ELISA on which divalent IgG binds to two molecules of the TSHR has been validated further by the results of the experiments shown in Table 12b. Intact IgGs of human MAbs to the TSHR (M22, 5C9, K1-18 and K1-70) showed dose dependent binding in the ELISA while the monovalent Fab fragments of the same MAbs showed little or no response (Table 12b). Mouse TSMAbs 1-7 also bound well in the TSHR260-AP ELISA as illustrated by the examples in Table 12c. OD 405 nm signal ranged from 0.103 to 0.561 at 10 μg/mL concentrations of TSMAbs 1-7 (Table 12c). Mouse TSHR blocking MAb 9D33 also bound in this assay system with an OD 405 nm signal of 0.481 at 10 μg/mL (Table 12d). Patient sera containing TRAbs with stimulating activity i.e. sera that showed an ability to stimulate cyclic AMP activity in CHO cells expressing the TSHR reacted well in the TSHR260-AP ELISA. Table 12e shows examples of 6 different sera tested at different dilutions and the OD405 nm signal ranged from 0.407 to 0.924 at 1/5 dilutions in HBD sera. Furthermore sera from patients with blocking type TSHR autoantibodies bound well in the TSHR260-AP ELISA as illustrated by the examples in Table 12f with the OD 405 nm signal ranging from 0.323 to 0.896 at 1/10 dilutions in HBD sera. Table 12g shows more examples of binding of patient sera in the TSHR260-AP ELISA. TRAb concentrations in the TSHR260-AP ELISA were calculated from a calibration curve prepared from the NIBSC reference preparation 90/672 and compared to TRAb concentrations (expressed as NIBSC U/L) measured in the same sera using a TSHR coated tubes assay. There was good overall agreement in TRAb measurements made using the TSHR260-AP ELISA and by inhibition of TSH binding to full length TSHR (coated tube assay) (r=0.913, n=57) (
Temperature Stability of Different TSHR Preparations
In the temperature stability experiments OD450 nm values of binding of M22 Fab-peroxidase to the full length TSHR in the ELISA were 1.748, 0.268 and 0.126 respectively for (a) preparations stored at −80° C. (untreated), (b) incubated for 24 hours at room temperature followed by return to −80° C. and (c) incubated for 48 hours at room temperature followed by return to −80° C., respectively. Consequently full length TSHR preparations stored at room temperature for 48 and 24 hours showed respectively only 7% and 15% activity relative to untreated preparations. M22 Fab-peroxidase binding OD450 nm values were 2.293 for untreated TSHR260 and 1.836 and 1.676 for TSHR260 stored at room temperature for 24 and 48 hours respectively. The activity of TSHR260 stored at room temperature for 24 and 48 hours relative to untreated preparations was 80% and 73% respectively. Similar results were observed in the case of TSHR260-AP with OD450 nm of 2.106 and 1.983 for samples stored at room temperature for 24 and 48 hours respectively compared to 2.395 for untreated samples. This represented 88% and 83% binding activity after 24 and 48 hours room temperature storage relative to untreated TSHR260-AP. In the experiments with untreated TSHR LRD C-CAP the OD450 nm was 1.826 and after 24 and 48 hour room temp storeage 1.158 and 1.155, respectively. TSHR LRD C-CAP showed 63% activity relative to untreated preparations after 24 and 48 hour storage at room temperature. The above described experiments showed that the ability to bind M22 of TSHR260, TSHR260-AP and TSHR LRD C-CAP after room temp treatment was greater than the full length TSHR preparations. This indicates that TSHR260, TSHR260-AP and TSHR LRD C-CAP are more stable at room temperature compared to full length TSHR.
Variable Region Sequences
Sequence analysis of the genes coding for K1-18 indicated that the HC V region genes were from the VH5-51*01 family, the D genes from the D3-16*02 (or D3-16*01) family and the JH genes from the J3*02 family. In the case of the LC, V region genes were from the V3-20*01 family and J region genes from the JK-1*01 germline. The HC nucleotide and amino acid sequences are shown in
K1-70 Fab Structure
The structure of Fab K1-70 has been determined at 2.22 Å resolution (
The overall surface of the K1-70 antigen binding area is highly irregular with a cavity near to the centre.
The cavity is surrounded mostly by aromatic residues and by LC Asp50 (
Furthermore, the interior of the cavity is also populated by aromatic residues. This suggests that aromatic contacts may be important for the interaction between K1-70 and the TSHR with the prominent aromatic residue on the surface of the TSHR “fitting” into the cavity on the K1-70 surface.
Recombinant K1-70 Fab
Table 17a shows that recombinant K1-70 Fab in E. coli culture supernatant had the ability to inhibit 125I-TSH binding to the TSHR. The inhibiting effect was complete at lower dilutions of the culture supernatants (91.9% at 1:2 dilution) while increasing dilutions of the supernatant cause dose dependent inhibiting effect (27.9% inhibition at 1:256 dilution) (Table 17a). The effect of recombinant K1-70 Fab on TSH mediated stimulation of cyclic AMP production in CHO cells expressing the TSHR is shown in Table 17b. Different dilutions of culture supernatants showed dose dependent inhibition of cyclic AMP stimulation; from 89.3% inhibition at 1:5 dilution to 39.7% inhibition at 1:40 dilution (Table 17b). Control culture supernatants from non-induced E. coli cultures did not produced detectable inhibition of TSH binding or inhibition of TSH mediated cyclic AMP stimulation (Tables 17a &b).
The experiments described above show that two monoclonal autoantibodies to the TSHR with very different biological activities (K1-18 stimulating and K1-70 blocking) can be isolated from a single preparation of a patient's lymphocytes. Consequently, the patient's immune system was producing both types of TSHR autoantibody ie stimulating type and blocking type at the same time. Once isolated (as described above) in the form of monoclonal autoantibodies, the properties of the two types of TSHR autoantibody can be investigated without interference from each other. The characteristics of the new human MAb with TSHR stimulating activity (K1-18) have been described and compared to the characteristics of some other known TSHR MAbs. Specifically of a stimulating human MAb (M22), a blocking human MAb (5C9), a blocking human Mab (K1-70), a blocking mouse MAb (9D33) and mouse stimulating MAbs (TSMAbs 1-7). Also the characteristics of the new human MAb with TSH antagonist activity (K1-70) have been described and compared to the characteristics of known MAbs. Specifically of a blocking human MAb (5C9), a blocking mouse MAb (9D33), a stimulating human MAb (M22), a stimulating human MAb (K1-18) and mouse stimulating MAbs (TSMAbs 1-7). It has been shown that the new human stimulating TSHR MAb K1-18 has properties similar to M22 in terms of:—inhibition of binding of labelled TSH to the TSHR, inhibition of binding of each other to the TSHR, inhibition of binding of blocking human MAbs (5C9 and K1-70) to the TSHR, inhibition of binding of mouse blocking and stimulating MAbs (9D33 and TSMAbs 1-7). Also the patient serum TRAbs inhibited binding of K1-18 to the TSHR. Furthermore, both M22 and K1-18 bind to the TSHR with a high affinity and are able to bind to a TSHR fragment consisting of amino acids 22-260 linked to alkaline phosphatase. Antibodies such as M22 and K1-18 have the ability to stimulate TSHR cyclic AMP activity although the potency of the two antibodies differs by about 1.5 fold. The studies show that the properties of TSHR stimulating autoantibodies are similar in different patients and they are representative of the properties of TSHR stimulating autoantibodies in all patients with Graves' disease studied so far. A summary of K1-18 characteristics is shown in Table 13a. Our experiments also showed that the new blocking type human MAb K1-70 (obtained from the same sample of lymphocytes as the stimulating MAb K1-18) has the ability to:—inhibit binding of labelled TSH to the TSHR, inhibit binding of human MAbs (M22, K1-18 and 5C9) to the TSHR, inhibit binding of mouse blocking and stimulating MAbs (9D33 and TSMAbs 1-7) to the TSHR. Furthermore binding of K1-70 to the TSHR was inhibited by patient serum TRAbs. K1-70 showed potent TSH antagonist activity and the ability to block stimulation of the TSHR by all patient serum TRAbs tested. K1-70 was shown to be a more effective inhibitor of TSH binding to the TSHR than 5C9. K1-70 binds to the TSHR with a high affinity and is able to bind to the TSHR fragment of amino acids 22-260 linked to the alkaline phosphatase. Consequently, K1-70 has the characteristics of patient sera with blocking TRAbs including the high binding affinity for the TSHR, the ability to inhibit TSH and M22 binding to the TSHR and the ability to block ligand induced TSHR stimulation at low concentrations of antibody. However, K1-70 has no effect on TSHR constitutive activity while 5C9 does. A summary of K1-70 characteristics is shown in Table 13b. K1-18 ability to stimulate cyclic AMP activity in CHO cells expressing TSHRs was lost when TSHR was mutated at Glu157Ala, Lys183Ala, Tyr185Ala, Asp232Ala or Arg255Asp. K1-70 ability to block TSH mediated cyclic AMP activity in CHO cells expressing TSHRs was reduced in the case of TSHR mutations Lys58Ala, Ile60Ala, Arg109Ala, Lys183Ala, Lys250Ala and slightly reduced by the TSHR mutation Tyr82Ala. Both K1-18 and K1-70 as well as M22 reacted well with the TSHR fragment of 22-260 in an ELISA based on TSHR 260-AP. Furthermore a panel of patient serum TSHR autoantibodies reacted well with TSHR amino acids 22-260 in the same assay. Patient sera with either type of TRAb activities (stimulating and blocking) bound to TSHR260 in the ELISA. In addition, ELISA plate wells coated with the TSHR fragment of amino acids 22-260 bound M22-peroxidase (from RSR Ltd) well and this M22-peroxidase binding was inhibited by a panel of patient serum TSHR autoantibodies. This inhibition of M22-peroxidase binding by the patient serum TSHR autoantibodies was similar to inhibition of M22-peroxidase binding to full length TSHR. Surprisingly therefore the TSHR fragment of amino acids 22-260 (or perhaps a smaller fragment) appears to be sufficient for routine assays of TSHR autoantibodies. Furthermore, M22 also bound well to a longer fragment of the TSHR (TSHR LRD C-CAP). In stability studies, the ability of M22 to bind TSHR260, TSHR260-AP and TSHR LRD C-CAP after they had been pre-incubated at room temperature was greater than the full length TSHR preparations which had been pre-incubated under the same conditions. This indicates that TSHR260, TSHR260-AP and TSHR LRD C-CAP are more stable at room temperature compared to full length TSHR. The TSHR mutation Arg255Asp had no effect on binding of K1-70 IgG while K1-18 IgG (at lower concentrations ie 0.1 μg/mL and below) bound less effectively to the mutated receptor. The experiments with different patient serum TRAbs indicate that sera with TSHR stimulating activities can be differentiated from sera with TSHR blocking activities on the basis of differences in binding to the TSHR containing R255D mutation. Binding of patient sera with stimulating activities is affected by the mutation while binding of patient sera with blocking activities is affected less or not at all. The experiments provide nucleotide and amino acid sequences of K1-18 and K1-70. Although heavy chain V genes of K1-18, K1-70 are derived from the same germline which belongs to the same family as the other stimulating human MAb M22 heavy chain V genes they are all combined with D and J genes from different families; furthermore K1-18 uses the kappa light chain, whereas M22 and K1-70 use lambda light chains. 5C9 (the other blocking type human MAb) germline genes are different from M22, K1-18 and K1-70 except that 5C9 and K1-70 use J4 heavy chain genes. Amino acid sequences of the CDRs of stimulating MAbs (M22 and K1-18) and blocking MAbs (5C9 and K1-70) are essentially different in particular within the heavy and the light chains CDR3s. These observations indicate that each of the 4 human autoantibodies is derived from distinct germlines. Also different CDR sequences may show similar biological activities towards the TSHR. The X-ray diffraction data provide molecular details of K1-70 Fab structure including the topography of the antigen binding site. A recombinant K1-70 Fab produced by cloning and expression of the K1-70 HC (SEQ ID No 46) and K1-70 LC (SEQ ID No 63 with SEQ ID No 64) in E. coli showed the ability to inhibit 125I-labelled TSH binding to the TSHR and the ability to inhibit TSH mediated stimulation of TSHR cyclic AMP activity. Overall the results indicate that antibodies in accordance with the invention such as K1-18 and K1-70 show similar TSHR binding activity and similar biological effects on TSHR function as TSHR MAbs described previously (M22 and 5C9) and as TSHR autoantibodies found in different sera from patients with autoimmune thyroid disease.
125I-TSH binding in the presence of assay buffer was 11%. 125I-TSH binding in the presence of HBD pool was 11.5%. Assay buffer is 50 mmol/L NaCl, 10 mmol/L Tris pH 7.8, 0.1% Triton X-100.
125I-K1-18 IgG
125I-TSH
125I-K1-18 Fab binding in the presence of assay buffer was 10%.
125I-K1-18 Fab binding in the presence of HBD pool was 9.4%.
aThe total final concentration of IgG mixture is shown; ie in the case of 100 μg/mL (K1-70 IgG + 5C9 IgG) the mixture contains 50 μg/mL K1-70 IgG and 50 μg/mL 5C9 IgG. Consequently, the combined effect of two IgGs at 100 μg/mL can be compared to the effect of the single IgG at the same concentration (100 μg/mL).
aThe total final concentration of IgG mixture is shown; ie in the case of 100 μg/mL (K1-70 IgG + 9D33 IgG) the mixture contains 50 μg/mL K1-70 IgG and 50 μg/mL 9D33 IgG. Consequently, the combined effect of two IgGs at 100 μg/mL can be compared to the effect of the single IgG at the same concentration (100 μg/mL).
aThe total final concentration of IgG mixture is shown; ie in the case of 100 μg/mL (K1-70 IgG + 9D33 IgG) the mixture contains 50 μg/mL K1-70 IgG and 50 μg/mL 9D33 IgG. Consequently, the combined effect of two IgGs at 100 μg/mL can be compared to the effect of the single IgG at the same concentration (100 μg/mL).
aThe total final concentration of IgG mixture is shown; ie in the case of 100 μg/mL (K1-70 IgG + 9D33 IgG) the mixture contains 50 μg/mL K1-70 IgG and 50 μg/mL 9D33 IgG. Consequently, the combined effect of two IgGs at 100 μg/mL can be compared to the effect of the single IgG at the same concentration (100 μg/mL).
aIn some experiments the MAbs were mixed in equal proportions and the concentration shown for a mixture represents the total amount of IgG added i.e.: 100 μg/mL (K1-70 + 5C9) = 50 μg/mL of K1-70 + 50 μg/mL of 5C9 used in the experiment.
aThe total final concentration of IgG mixture is shown; ie in the case of 2 μg/mL K1-70 IgG + 5C9 IgG, the mixture contains 1 μg/mL K1-70 IgG and 1 μg/mL 5C9 IgG. Consequently, the combined effect of two IgGs at 2 μg/mL can be compared to the effect of the single IgG at the same concentration (2 μg/mL).
aTest samples in hypotonic cyclic AMP assay buffer.
bTSH final concentration = 3 ng/mL
c% inhibition = 100 × [1 − (cyclic AMP in the presence of test samples and TSH/cyclic AMP in the presence of cyclic AMP buffer and TSH)].
125I-TSH binding to TSHR coated tubes by patient sera
125I-K1-70 Fab
125I-K1-70 IgG
125I-TSH
1Read off M22 IgG calibration curve (0.005, 0.001, 0.05, 0.1, 0.5, 1.0 and 10 μg/mL was run in each assay).
2Units are NIBSC 90/672.
3Stimulation of cyclic AMP production was tested (serum diluted 1:10 in hypotonic cyclic AMP buffer) using CHO cells expressing the full length TSHR. S1 = TSHR stimulating patient serum 1. S2 = TSHR stimulating patient serum 2. S3 = TSHR stimulating patient serum 3. S4 = TSHR stimulating patient serum 4. S5 = TSHR stimulating patient serum 5. S6 = TSHR stimulating patient serum 6. Serum dilutions for both ELISAs were made in HBD.
athe inhibition of 125I-TSH binding was carried out using the PEG precipitation assay (Southgate K, Creagh F, Teece M, Kingwood C, Rees Smith B. A receptor assay for the measurement of TSH receptor antibodies in unextracted serum. Clin Endocrinol 1984; 20: 539-548).
aA calibration curve using M22 IgG (0.005, 0.01, 0.05, 0.1, 0.5, 1.0 and 10 μg/mL) was run in each assay.
bA calibration curve using K1-70 IgG (0.005, 0.01, 0.05, 0.1, 0.5, 1.0 and 10 μg/mL) was run in each assay.
aA calibration curve using M22 IgG (0.005, 0.01, 0.05, 0.1, 0.5, 1.0 and 10 μg/mL) was run in each assay.
bA calibration curve using K1-70 IgG (0.005, 0.01, 0.05, 0.1, 0.5, 1.0 and 10 μg/mL) was run in each assay.
125I-TSH binding
125I-TSH binding
bTSH (2)
bTSH (2)
aTest samples in hypotonic cyclic AMP assay buffer.
bTSH final concentration = 3 μg/mL.
c% inhibition = 100 × [1 − (cyclic AMP in the presence of test samples and TSH/cyclic AMP in the presence of cyclic AMP bufferand TSH)]. 5B3 is a human monoclonal antibody to GAD (negative control for K1-70).
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bM22 (2)
bM22 (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
bTSH (2)
125I-TSH
1Concentration of recombinant K1-70 Fab in culture supernatant measured using the Easy-Titer Human IgG (H = L) assay kit (Pierce Biotechnology) using different concentrations of hybridoma produced K1-70 Fab as a calibration curve
2inhibition of binding was calculated using the formula: % inhibition = 100 − [A/B × 100] where A = % of 125I-TSH binding in the presence of test sample; and B = % of 125I-TSH binding in the presence of assay buffer (50 mM NaCl, 10 mM Tris, pH 7.8, 1% Triton X-100 and 1 mg/mL BSA).
1Assay buffer: Hanks' buffered salt solution (NaCl free) containing 1 g/L glucose, 20 mmol/L HEPES, 222 mmol/L sucrose, 15 g/L BSA and 0.5 mmol/L 2-isobutyl-1-methylxanthine pH7.4
2Dilutions in assay buffer
3Concentration of recombinant K1-70 Fab in culture supernatant measured using the Easy-Titer Human IgG (H = L) assay kit (Pierce Biotechnology) using different concentrations of hybridoma produced K1-70 Fab as a calibration curve
4% inhibition of TSH stimulation of cyclic AMP: % inhibition = 100 − [A/B × 100] where A = stimulation of cyclic AMP by 3 ng/mL TSH in the presence of test sample; and B = stimulation of cyclic AMP by 3 ng/mL TSH.
Number | Date | Country | Kind |
---|---|---|---|
0823562.4 | Dec 2008 | GB | national |
0908945.9 | May 2009 | GB | national |
This application is a divisional application of National Stage application under 35 U.S.C. §371 and claims benefit under 35 U.S.C. §119(a) of International Application No. PCT/GB2009/002946 having an International Filing Date of Dec. 23, 2009, which claims the benefit of priority of GB 0908945.9 having a filing date of May 22, 2009 and GB 0823562.4 having a filing date of Dec. 24, 2008.
Number | Date | Country | |
---|---|---|---|
Parent | 13142217 | Aug 2011 | US |
Child | 14790489 | US |