EP0535032A1 - Therapy and prophylaxis of autoimmune diseases - Google Patents

Abstract

The invention relates to a novel method for the treatment or prophylaxis of an autoimmune disease. The method comprises administering to a patient who requires it, an effective amount of at least one mammalian antibody with an idiotypic connection or one of its fragments with specific binding. An autoimmune disease of particular interest is insulin dependent diabetes mellitus (IDDM).

Description

THERAPY AND PROPHYLAXIS OF AUTOIMMUNE DISEASES

The present invention relates a method for the treatment or prophylaxis of autoimmune disease.

BACKGROUND OF THE INVENTION A number of investigators have demonstrated that

"physiological" autoantibodies exist in measurable

quantities in normal individuals (e.g., Dighiero et al ., 1985). Self-reactive B cells are, thus, continuously activated to effector functions. The most straightforward interpretation of these

observations is that pathology is not caused by the

presence or concentration of a given antibody or clone.

Rather, health or disease are systemic properties,

determined in part by components of the immune apparatus other than the antigen-specific clones (Holmberg and

Coutinho, 1985). A number of recent findings lend support to these interpretations.

The antibody population which includes auto-antibodies - that is, the "natural antibodies" or actual repertoire - is also characterized by a very high degree of idiotypic connectivity. In other words, a large proportion of these antibodies react with other natural antibodies produced in the same individual through V-region interactions (Holmberg et al, 1984a). Thus, one part of the system, consisting of resting

immunocompetent lymphocytes, obey the postulates of the clonal selection theory and constitutes the pool responding to antigenic challenge from the external environment. The other part of the immune system, constituting 5-10% of the total lymphocyte population, is organized as a highly connected network based on interactions amongst the

variable regions of free antibodies and activated lymphocytes, together with interactions between variable regions and other self-antigens (Couthinho, 1989, Holmberg et al., 1989). Structural and serological studies of the development of B-cell repertoires have led to the

conclusion that immune systems begin as a developmental controlled, germline-encoded high-connectivity network (Holmberg et al., 1989, Lundkvist et al. 1989). Such developmental constrains have been suggested to reflect a functional role of the early B-cell repertoires, with implications for the establishment of both B-cell and post-thymic T-cell repertoires.

There is increasing clinical and experimental evidence that many autoimmune diseases develop as a result of

abnormalities in the immune system, especially in T

lymphocyte-mediated immunity. Many signs and symptoms of infectious, inflammatory and neoplastic diseases evolve as a result of stimulation of cellular immunity. In addition, although immunocompetent cells may not necessarily be involved at the initial stage, abnormal regulation of otherwise appropriate cellular immune reactions may lead to acute and chronic diseases. These diseases are generally of unknown etiology and include systemic rheumatic diseases (e.g. rheumatoid arthritis (RA)), organ-specific endocrine diseases (e.g. insulin-dependent diabetes mellitus (IDDM)) and inflammatory diseases of the gut (e.g. Crohn's disease).

The treatments available in relation to said diseases are usually symptomatic or palliative treatments, i.e. most of the drugs prescribed in connection with said diseases are directed at the allaying of symptoms and usually have no curative effect. Other treatments are so-called

substitution therapies which involve life-long supplying to the patient of substances, e.g. hormones, needed due to a reduced/insufficient internal production of said

substance. Said treatments are often unsatisfactory and imply unwanted and often serious side-effects. Thus, improved methods of treatments and improved pharmaceutical compositions are needed.

BRIEF DESCRIPTION OF THE INVENTION

It is the object of the present invention to remedy the deficiencies of the prior art. More particularly, the present invention relates to treatment and prophylaxis of an autoimmune disease in accordance with a new strategy based upon surprising findings as described in the

examples. The strategy involves the treatment or

prophylaxis of autoimmune disease by administering, to a patient in need thereof, an effective amount of at least one mammalian idiotypically connected antibody or a

specifically binding fragment thereof.

It was surprisingly found that single injections of

moderate doses of isogenic, natural monoclonal antibodies can efficiently prevent the development of T cell mediated type I diabetes. The effect using monoclonal antibodies was found to be related to V-region specificity, as some, but not other, monoclonal antibodies of identical isotype and origin, had the observed effect.

Similar it was found that one of the monoclonal antibodies also showed an effect on the development of disease in another mouse model for a spontaneous autoimmune disease, the B6-lpr/lpr mouse, as measured as titers of RF and autoantibodies against DNA.

It was also found that repeated administration of high doses of polyclonal Ig given at birth, can to some extent prevent the development of T cell-mediated type I diabetes. A decrease in the development of insulitis was found to be more pronounced in mice treated with rabbit anti-mouse IgM as compared to normal polyclonal rabbit immunoglobulin. Although the present invention is not to be limited to any theory, it seems justified to assume that the surprising effect obtained according to the strategy of the invention is at least partially due to idiotypically connected antibodies playing an important role in the establishment or "normalization" of the immune system. This

establishment or normalization has been demonstrated in newborn mammals, and one aspect of the invention thus opens up the possibility of a treatment of newborns, in

particular newborns having a high risk of autoimmune disease. However, provided, as explained herein, idiotypic connectivity is a property related to normalization and control of the immune system, it appears conceivable that treatment with antibodies showing these characteristics will also be effective when administered to adult

individuals displaying the corresponding autoimmune

diseases. Also, it is conceivable that the antibodies in question have a general property not limited to species, that is, it is contemplated that idiotypically connected murine antibodies will be useful across species barriers, including being useful in humans.

DETAILED DISCLOSURE OF THE INVENTION

As mentioned above, the present invention thus relates to the treatment or prophylaxis of an autoimmune disease in mammals, including humans, by administrering an effective amount of at least one mammalian idiotypically connected antibody or a specifically binding fragment thereof to a patient in need thereof, or, expressed in another manner, to the use of at least one mammalian idiotypically

connected antibody or a specifically binding fragment thereof for the preparation of a pharmaceutical composition for the treatment or prevention of autoimmune disease in mammals, including humans.

Sundblad et al. (1989) disclose an experiment in which a murine anti-idiotypic antibody directed against a murine antibody with anti-acetylcholine receptor activity was administered to newborn BALB mice which were then immunized by injection of acetylcholine receptor from another species (Torpedo marmorata [electric eel]) in order to

experimentally induce myasthenia gravis. The group of mice which received this treatment showed a decreased antiacetylcholine receptor antibody response after the

immunization compared to mice which had not received the anti-idiotypic antibody. The only thing which this

experiment proves is that it is possible to alter the immune response to an experimentally induced heterologous acetylcholine receptor by administrering an anti-idiotypic antibody against an antibody with anti-acetylcholine receptor activity. In the present specification and claims, the term

"idiotypically connected" designates an antibody which is capable of reacting, by V-V interaction, with at least one naturally occurring antibody from the same individual mammal or the same mammal species. The term "V-V- interaction" (also known as "idiotypical interaction") indicates that the reaction is between the variable regions of the antibodies interacting. There are several way to asses that an interaction between antibodies is a V-V- interaction, such as by enzymatically cleaving the F_ab fragment from the antibody in question and assessing whether they interact; if they do, the interaction is a V-V interaction. A quick method for assessing the same is to show that the antibodies in question do not interact with another antibody of the same origin and with the same C region isotype.

In the present context the term "specifically binding fragment" is used to define a fragment of the idiotypically connected antibody, which is capable of exerting

substantially the same effect in the treatment or

prophylaxis of an autoimmune disease as the whole antibody. The specifically binding fragment may be obtained by well- known techniques, and e.g. derived directly from an

idiotypically connected antibody, e.g. by enzymatically cleavage. In the following the terms "fragment" and

"specifically binding fragment" are used interchangeably. Normally, idiotypically connected antibodies will be capable of idiotypically interacting with a plurality of naturally occurring antibodies from the same individual or species. A suitable test which will in most cases reveal idiotypical connection is to establish a panel of different natural antibodies from the same individual or another individual of the same species, such as a panel of

antibodies produced by B cell hybridomas, and assessing whether the antibody in question reacts with one or several of the antibodies in the panel, e.g. by means of a suitable test such as the ELISA test described in Example 2. As a guideline, the panel should preferably comprise at least 50 antibodies, but it is preferred that it comprises as many different antibodies as possible, so as to minimize the possibility that an idiotypically connected antibody is not recognized as such. In the experiments summarized in Table 1 below, the panel comprised 70 antibodies.

The initial panel may be randomly selected, or the

antibodies used in the panel may be deliberately selected for diversity. Once a candidate has been tested, it may be added to the panel to obtain a larger panel. As is

demonstrated in Table 1, idiotypical interaction will normally take place with several of the antibodies in a panel of sufficient size.

As used herein the term "pharmaceutical composition" comprises any composition suitable for human use, such as compositions which are administered parenterally. The administration will be dependent on the patient's age and weight and the particular condition being treated as well as the severity of the disease. The active ingredient i.e. the antibody or antibodies or fragments thereof may be contained in any appropriate amount in the composition, and are generally contained in an amount of about 0.1-10 mg per ml, in particular about 1 mg per ml. The composition may be in any appropriate unit dosage form.

Parenteral administration may comprise suitable injection, e.g. intravenous, intraperitoneal, intramuscular,

intraarticular, intraocular or retroocular injection, infusion or implantation of e.g. suitable delivery devices.

Formulations for parenteral use may be presented in unit dose forms, e.g. ampoules, or in multi-dose containers with an added suitable preservative. The composition may be in form of a solution, a suspension, an emulsion or a delivery device for implantation or may be presented as a dry powder to be reconstituted with water or another suitable solvent before use. Apart from the active antibody substance the compositions may comprise suitable pharmaceutically

acceptable carriers and/or excipients. Furthermore the composition may, in addition, conveniently comprise

suspending, stabilizing, pH-adjusting agents and/or

dispersing agents.

The dosage of the composition according to the invention depends on the administration method, the disease, the severity of the disease and of the weight and age of the patient. For parenteral administration, a dosis of about 0.01-μg to about 1 mg pr kg body weight, in particular from about 0.1 μg to about 10 μg, should be appropriate for a single parenteral administration. The administration can be performed daily or at intervals ranging from a couple of days to a couple of weeks, or even intervals of months or years in certain cases, depending upon the individual responsiveness. It is contemplated that for some autoimmune diseases, one single or just a few administrations could be sufficient to permanently cure the patient. Other disorders may require life-long treatment.

In every case, the dosage should be carefully adapted so as to imply the optimal dosis for obtaining the specific action in the lymphocyte interaction system, i.e.

attainment of optimal total dosages, dosage forms and dosage frequency. A conventional and convenient manner for adapting the dosage is to begin with a very low dosage and then increase the dosage if necessary to obtain the

desired effect. Dosage frequency and dosage form may likewise be optimised in a manner known in the art.

The strategy made possible according to the present

invention could also be utilized with more advanced

delivery systems, such as an implantation device for continuous administration of the antibodies, e.g. a

degradable polymer suitable for prolonged administration of proteins.

In particular, the invention relates to the use of at least one mammalian idiotypically connected antibody or a

specifically binding fragment thereof for the preparation of a pharmaceutical composition for the treatment or prevention of diabetes mellitus, in particular insulin- dependent diabetes mellitus (type 1), in particular for the prevention of progression of diabetes mellitus (type 1), especially substantially immediately after the first diagnostic establishment of diabetes mellitus, or for prophylaxis after establishment of being in a high risk group of developing diabetes mellitus in mammals, including humans. These uses as well as other uses and methods of the

invention will preferably be noted specifically and in detail as explained herein in instructions for use provided to the physician and/or to the patient together with the compositions to be used. As used in the present specification and claims, the term "substantially immediately after" comprises a period after the first diagnostic establishment of the disease or a high risk of developing the disease within a limited number of months or years which period should be as short as

possible, so as to enable intervention with the progression of irreversible changes in the pathological body tissues.

A pharmaceutical composition comprising at least one mammalian idiotypically connected antibody or a

specifically binding fragment thereof according to the present invention may be used for the prevention of

progression of irreversible pathological processes in the pancreatic tissues which otherwise lead to diabetes

mellitus (type 1). The therapy of the invention is

preferably instituted as soon as possible, i.e. very shortly after the onset of diabetes or, preferably,

immediately after diagnosis of pathological processes in the islets of pancreas (e.g. before clinical manifestation of diabetes mellitus (type 1)). The therapy of the

invention may also apply to certain identified high risk groups the identification being e.g. obtained by means of a screening program employing markers relatively specific for pathological changes seen in diabetes mellitus and/or preliminary stages thereof, especially markers of ongoing pathological changes in predisposed individuals (e.g. in HLA-DR 3/4-positive individuals). A person is contemplated to be in a high risk group of developing diabetes mellitus (type 1) if his risk is about 10-100%, such as about 50- 100%. Moreover, the present invention relates to the use of at least one mammalian idiotypically connected antibody or a specifically binding fragment thereof for the preparation of a pharmaceutical composition for the treatment or prevention of conditions related to the function of the thyroid gland, in particular hyper- or hypofunctioning of said gland, e.g. thyroiditis (acute, subacute or chronic), including Hashimoto's disease (lymphocytic thyroiditis;

lymphoid thyroiditis), Riedel's thyroiditis (chronic fibrous thyroiditis), de Quervain's thyroiditis (subacute granulomatous thyroiditis), subacute lymphocytic

thyroiditis, Graves' disease, Graves' subacute thyroiditis and Graves' ophthalmopathy in mammals, including humans.

In another aspect, the invention relates to the use of at least one mammalian idiotypically connected antibody or a specifically binding fragment thereof for the preparation of a pharmaceutical composition for the treatment or prevention of a condition related to hypofunction of the adrenal glands, in particular Addison's disease or

furthermore Simmonds' panhypopituitarism in mammals, including humans.

In another aspect, the invention relates to the use of at least one mammalian idiotypically connected antibody or a specifically binding fragment thereof for the preparation of a pharmaceutical composition for the treatment of progression of arthritis such as rheumatoid arthritis including juvenile rheumatoid arthritis, psoriatic arthritis or Reiter arthritis substantially immediately after the first diagnostic establishment of arthritis in mammals, including humans. The therapy of the invention is

preferably instituted as soon as possible, i.e. very shortly after the onset of arthritis or, preferably, immediately after diagnosis of pathological processes (e.g. before clinical manifestations) . The therapy of the

invention may also apply to certain identified high risk groups the identification being e.g. obtained by means of a screening program employing markers relatively specific for arthritic changes and/or preliminary stages of arthritis, especially markers of ongoing pathological changes in predisposed individuals, in particular individuals having specific HLA tissue types. In yet another aspect, the invention relates to the use of at least one mammalian idiotypically connected antibody or a specifically binding fragment thereof for the preparation of a pharmaceutical composition for the treatment of

Crohn's disease or ulcerative colitis, especially for the prevention of relapse or progression of Crohn's disease or ulcerative colitis, or for the treatment of pernicious anemia or celiac disease in mammals, including humans.

In yet another aspect, the invention relates to the use of at least one mammalian idiotypically connected antibody or a specifically binding fragment thereof for the preparation of a pharmaceutical composition for treatment or prevention of relapse of systemic lupus erythematosus, multiple sclerosis, sarcoidosis, primary biliary cirrhosis,

autoimmune hemolytic anemia, immune thrombocytopenia purpura and myasthenia gravis in mammals, including humans.

In yet another aspect, the invention relates to the use of at least one mammalian idiotypically connected antibody or a specifically binding fragment thereof for the preparation of a pharmaceutical composition for treatment or prevention of autoimmune glomerulonephritis and Goodpasture's

syndrome.

While the composition used according to the invention will in many cases comprise one single or a few monoclonal antibodies, it is also possible to use a larger number of antibodies, and even to use polyclonal antibodies. In such case, however, the predominant proportion of the antibodies should, of course, be the idiotypically connected

antibodies. Thus, the idiotypically connected antibody or antibodies should preferably constitute at least 80% by weight of the antibodies used for the preparation, more preferably at least 90% by weight of the antibodies used for the preparation, still more preferably at least 95% by weight of the antibodies used for the preparation, and most preferably substantially 100% by weight of the antibodies used for the preparation. Of course, the same

considerations apply when the composition contains one or more specifically binding fragments of the antibody or antibodies or a mixture of antibodies and fragments.

In cases where several antibodies are used, in particular where the antibodies are polyclonal antibodies, the

idiotypically connected antibodies may be isolated or concentrated from a mixture in which not idiotypically connected antibodies are present by, e.g., absorption to a panel of natural idiotypically connected antibodies

immobilized on a suitable matrix, e.g. Sephadex, such as a panel of antibodies as described above immobilized to a matrix, and subsequent release of the absorbed antibodies, e.g. by elution at low pH and/or high salt concentration.

It is contemplated that for many uses according to the invention, it will be preferred that the antibody is a monoclonal antibody or a mixture of monoclonal antibodies.

Particularly preferred antibodies are those in which the V_H gene segment is encoded by a member of the group

consisting of the V_H gene segments belonging to the murine V_H gene families 7183, 36-60 or Q52 and the human V_H gene family V_H3 and the corresponding gene families of other mammals. (Schroeder et al., 1990). The aforementioned murine gene families are known to be positioned D-proximal on the chromosome, to be homologous between human and mouse and to be associated with high connectivity. For the purpose of the present specification and claims, and in accordance with the definition by Brodeur et al., 1984, and Dildrop, 1984, the gene "families" are defined as

comprising genes which are at least 80% homologous with the specific genes disclosed by Dildrop et al.

Also preferred are antibodies which are encoded by genes which are devoid of template-independent insertions (N sequences) or in which the template-independent insertions comprise less than 5 such nucleotides. N-region diversity is believed to be random and therefore antibody

specificity, when derived from N-region addition, cannot be evolutionarily selected. It is argued in the present context that the property of high connectivity is

evolutionarily selected to be kept in germline genes expressed only during ontogeny, it follows that the lack of N-sequences corresponds to requirements of germline encoded specificities in the idiotypically connected antibodies. In a particular aspect, these antibodies also comprise the V_H gene segments described immediately above.

In practice, the antibodies used according to the invention will often be produced by hybridoma cell lines derived from B-lymphocytes producing naturally expressed idiotypically connected antibodies, such as the hybridoma cell lines disclosed in the Examples, e.g. the ones deposited at the European Collection of Animal Cell Cultures on 8 June 1990 under the accession numbers 90060809,

90060810, and

90060811, or they may be antibodies produced by cultivation of a suitable organism harboring a DNA fragment capable of expressing the antibody in question in the organism, such as DNA fragments comprising the fragments shown in Figs. 2 and 3 encoding the variable part of the antibody, and obtaining the antibodies from the organism. Suitable organisms for this purpose are bacteria, such as E. coli, harboring e.g. a plasmid containing the DNA fragment in question. In cases where glycosylation of the antibodies is decisive, the microorganism may be a cell line, such as a hybridoma cell line, carrying such a plasmid. The DNA fragment may, in such cases, be a fragment isolated from a naturally activated B cell, or it may be one which is wholly or partially produced by oligounucloetide synthesis and/or recombinant techniques and/or combinations thereof. Such antibodies may be antibodies which are identical or substantially identical to naturally occurring antibodies, or they may be chimeric antibodies, such as a chimeric antibody which has two pairs of light chain/heavy chain complexes, wherein at least one pair has chains comprising mouse variable regions such as the regions encoded by the sequences described herein, joined with human constant region segment, with or without naturally associated J and D segments.

The antibodies used may be antibodies characteristic to the same species as the species treated, or they may be

antibodies characteristic to another species, or they may be species-chimeric antibodies as described above. It is also contemplated that the variable fragments per se of the antibodies, such as the F_ab, F_ab(2) or other suitable fragment, may be used. These fragments may be obtained e.g. by enzymatically cleavage of the antibody in question. Specific monoclonal antibodies used according to the invention are selected from monoclonal antibodies BA.N 3:2.2, BA.N 4:3.40, BA.N 4:1.13, BA.N 4:4,57 (90060809), BA.N 1:5.24 (90060811), and BA.N 1:1.18 (90060810), among which the three last-mentioned antibodies have been found to be particularly effective, as appears from the

examples. The reactivity patterns of these antibodies are given in Fig. 1. Antibodies BA.N 4:4.57, BA.N 1:5.24 and BA.N 1:1.18 are produced by deposited hybridomas. (The abbreviation "BA.N" signifies BALB/c Neonatal. In one aspect of the invention, the antibodies used are antibodies the variable part of which is encoded by any of the DNA sequences shown in Figs 2 and 3 or by sequences which are at least 80% homologous therewith and capable of encoding the variable part of an idiotypically connected antibody. Recently presented data from studies of human autoimmune disorders have indicated a linkage to particular v_H-genes, known to be overrepresented in early B cell repertoires (Yang et al., 1990, Walter et al., 1991). The present invention is further described in figures without being limited thereto.

LEGENDS TO FIGURES

Legend to figure 1

Natural high connectivity antibodies preferentially use 7183 and Q52 V_H genes. Reactivity patterns of 27 "natural" B-cell hybridomas as tested on a large panel of ligands. TNP: 2,4,6-trinitrophenyl-BSA; NIP:

(4-hydroxy-5-iodo-3-nitrophenyl) acetyl-BSA; PC:

phosphorylcholine; Dx B512: α1-6 dextran; Dx 1355: α1-3d- extran; AC38 and AC146 (both monoclonal antibodies specific for the Bl-8 hybridoma molecule, and obtained from Dr K. Rajewsky, Cologne); 10/13-15 (monoclonal antibody specific for the TEPC 15 idiotype, obtained from Dr. C. Berek, Cambridge); 2-4 and 20-5 (both monoclonal antibodies specific for the SP603 idiotype, obtained from Dr. P.-A. Cazenave, Paris); IIIA 11C7 (monoclonal antibody prepared against the hybridoma molecule BA.N 1:1.8); VIB3C20

(monoclonal antibody prepared against Dx B512-specific monoclonal antibody 17-9); anti-J558 (monoclonal antibody specific for the J558 idiotype, obtained from Dr. J.

Kearney); M460, J558 and T15 (monoclonal antibodies derived from the myelomas MOPC460, J558 and TEPC15, respectively); F6(51) and 41-27 (monoclonal antibody specific for the MOPC460 idiotype, obtained from Dr. P.-A. Cazenave, Paris); G5Bb (monoclonal antibody specific for GAT, obtained from Dr. J. Theze, Paris); G6-3 (monoclonal antibody prepared against the BA.N 4:4.57 idiotype); H9-14.8, H40-315.7

(monoclonal antibodies both specific for I-Ek, obtained from Dr. M. Pierres, Marseille); H100-27/55 (monoclonal antibody specific for Kk, obtained from Dr. H. Lemke,

Kiel); 14-4-4 (monoclonal antibody specific for the la.7 epitope); KJ1.26.1 and KJ12.98.2 (monoclonal antibodies specific for idiotypic determinants on the antigen-specific T cell receptor, obtained from Dr. P. Marrack, Denver);

actine; tubuline; myosin; PO: peroxidase; renin; Neurof: neurofilaments, and DNA); AchR: acetylcholine receptor isolated from Torpedo California; TG: mouse thyroglobulin. Reactions scored as positive are indicated by filled squares. The cell lines are ordered according to the number of positive reactions against this panel of ligands. The VH-gene family expressed by each of the cell lines was determined by blotting poly(A) +RNA to nitrocellulose filters and hybridizing with 32P labeled DNA probes

specific for each of the VH gene families. NT=Not tested.

Legend to figure 2

V_H-region sequences from 4 hybridomas of neonatal origin, 1:5.24, 3:3.39, 4:1.13 and 4:3.40. The sequences are compared to the most homologous sequences found in the literature. The 1:5.24 and 3:3.39 clones are compared to the V_H 14.29 sequence, derived from a nonproductively rearranged allele of an Abelson virus transformed pre-B cell line of fetal liver origin and belonging to the 7183 V_H-gene family (Lawler et al., 1987). The 4:1.13 is

compared to the A8.2 sequence derived from an Abelson virus induced pre-B cell line, and typed as a member of the Q52 V_H-gene family (Gerondakis, 1984). The 4:3.40 is compared to the 13G3S sequence derived from a Ly-IB cell related hybridoma, and typed as a member of the J558 V_H-gene family (Förster et al., 1988). Dashes indicate sequence identity and ambiguities are indicated by a "x". Numbering is according to Rabat et al., 1987. The abbreviation "CLR" stands for complementary determining region, i.e., the actual antigen binding region. Legend to figure 3

V_K region sequences from the same hybridomas shown in Fig. 2 (indicated by boxes), compared to their most homologous sequences found in the literature. The 1:5.24 is compared to the L7 germline gene of the V_K23 subgroup (Pech et al., 1981). The 3:3.39 is compared to the VK 1.60 sequence derived from a fetal liver hybridoma and belonging to the V_Kl subgroup (Lawler et al., 1989). The 4:1.13 is compared to the NQ 22.87.L sequence derived from an oxazolone specific hybridoma and identified as a member of the V_KOx subgroup (Berek et al., 1987). The 4:3.40 is compared to the 3B9PC sequence derived from a Ly-1 B cell related hybridoma and typed as belonging to the V_K21 subgroup

(Förster et al., 1988). All other designations are as in Fig. 1.

Legend to figure 4

Part of the V_H region sequences of eight hybridomas of newborn origin and two of adult origin, covering the CDR3 region, the 3' end of the V-segment and the 5'end of the J- segment. Indicated in the figure are also the germline sequences of the D-regions utilized in the set of

hybridomas analyzed. The D-gene and J_H-gene segments expressed by each individual hybridoma is indicated within brackets. "N" indicates nucleotides which can not be assigned to any germline sequence. For other designations see Fig. 1.

Legend to Figure 5

Cumulative incidence of diabetes in female NOD-mice treated with rabbit-anti-mouse IgM (RalgM), normal rabbit Ig

(NRIg), BSA (1 mg/mouse, 3 times/week), or left untreated. Treatment was initiated within 24 hours after birth and continued for 4 weeks. The mice were thereafter left untreated. The incidence of diabetes was determined by testing for glucosuria.

Legend to Figure 6

Cumulative incidence of insulitis of male (o) or female (●) NOD-mice. The mice were sacrificed at the indicated time points and pancreas was histological examined for evidence of insulitis. 20 independent islets of Langerhans were examined and any individual showing at least one

infiltrated islet was scored as positive.

Legend to Figure 7

Incidence of insulitis in female NOD-mice at 10 and 15 weeks of age. The mice were treated with rabbit-anti-mouse IgM (RalgM) (closed bars) or normal rabbit Ig (NRIg)

(hatched bars) for the first 4 weeks of life. They were thereafter left untreated. At the age of 10 or 15 weeks the mice were sacrificed together with age-matched, non-treated controls (open bars) and the incidence of insulitis was determined by histological examination of pancreases.

Legend to Figure 8

Relative degree of insulitis in female NOD-mice at 15 weeks of age. Mice were treated with rabbit-anti-mouse IgM

(RalgM) or normal rabbit Ig (NRIg) for the time periods indicated on the abscissa. At 15 weeks mice were sacrificed and the frequency of Langerhans islets with mononuclear infiltrates was determined by histological examination of pancreas. Legend to Figure 9

Cumulative incidence of diabetes in female NOD-mice after treatment with BSA (●) or a mixture of monoclonal

antibodies (♦). Doses of 1 mg/mouse, 3 times/week was injected, treatment was started within 24 h after birth and continued for 4 weeks. The mice were than left untreated. Incidence of diabetes was scored by testing for

glucosuria. The figure also include a non-treated control group (Δ).

The mixture of monoclonal antibodies consisted of BA.N 1:5.24, BA.N 1:1.8, BA.N 4:5.57, BA.N 1:6.30, BA.N 4:1.13, BA.N 4:3.40, BA.N 3:2.2 in equal proportions.

Legend to Figure 10

Cumulative incidence of diabetes in female NOD-mice after treatment with 100 μg of the indicated monoclonal

antibodies given as one single injection within 24 h after birth.

The antibodies used were as follows:

Panel A: BA.N 1:6.30 (x), TNP.11 (●),

Panel B: BA.N 1:1.8 (Δ), BA.N 1:5.24 (●), BA.N 4:4.57 (♦),

Panel C: BA.N 4:3.40 (▲) , BA.N 3:2.2

(●), BA.N 4:1.13 (♦).

Legend to figure 11

Cumulative incidence of diabetes in female NOD-mice after treatment with one single injection of either 50 μg of F_ab- fragments prepared from the monoclonal antibody BA.N 1:1.8 (Δ) or 100 μg of the monoclonal antibody BA.N 1:5.24 (■) within 24 h after birth. The figure also include a nontreated control group (♦). Legend to figure 12

Cumulative incidence of diabetes in female NOD-mice after treatment a single injection of either 1 mg (■) or 100 μg (Δ) of the monoclonal antibody BA.N 1:1.8 at 9 weeks of age. The figure also include a non-treated control group

(☐).

Legend to figure 13

Titers of rheumatoid factor (RF), antibodies against single-stranded DNA (Anti-DNA ss) and antibodies against double-stranded DNA (Anti-DNA ds) in serum samples of lpr- mice at 20 weeks of age after treatment with 100 μg of the indicated monoclonal antibody given as one single injection within 24 h after birth. The antibodies used were BA.N 1:1.8 (▧ ) and BA.N 1:5.24 (▧ ). The figure also include a non-treated control group (☐). Serum samples were diluted 1/400 before assay of anti-DNA antibodies and 1/32 before assay of RF. (I) denotes standard error.

Legend to figure 14

Frequency of clones lacking N-sequence additions in the DJ_Hjunctional region. 7183 or J558 utilizing clones isolated from PCR libraries of BALB/c, C57B1/6 or NOD mice were sequenced, and the percentage of clones lacking N-sequence additions in the D-J_H junction was scored. The origin of the clones and the V_H gene family utilized is given below the respective bars. Filled bars represent neonatal clones and the hatched bars adult clones. N indicates the total number of clones isolated of each category. Legend to figure 15

Frequency of functional or non-functional 7183 clones lacking N-sequence additions in the D-J_H junctional region. 7183 clones of BALB/c (A) or C57B1/6 (B) mice, were

isolated and analyzed as described in legend to Figure 14. Closed bars represent functional rearrangements and hatched bars represent non-functional rearrangements.

Rearrangements displaying the V_Hgene segment and the J_H gene segments in different reading frames or that contained a termination codon in the CDR3 region were considered as non-functional. All other designations are as in legend to Figure 14.

Legend to figure 16

Frequency of functional or non-functional 7183 clones of NOD origin, lacking N-sequence additions in the D-J_H junctional region. Clones were isolated and analyzed as described in legend to Figure 14. All other designations as in legend to Figure 15.

EXAMPLE 1

Hybridomas and monoclonal antibodies

The hybridomas and monoclonal antibodies used in this study have been described before (Dighiero et al., 1985;

Holmberg et al., 1984a; Holmberg et al., 1984b; Holmberg et al., 1986). In brief, spleen cells from non-immunized 6- day-old BALB/c mice were hybridized to the non-secreting Sp2/o cell line using conventional methods (Köhler and Milstein, 1975) and directly cloned. All growing clones were kept and primary supernatants were used for screenings.

Most of the hybridomas were subsequently recloned under limiting dilution conditions and the serological reactions were confirmed using culture supernatant or partially purified antibodies from these clones.

EXAMPLE 2

Characterization of specific monoclonal antibodies

Enzyme-linked immunosorbent assays (ELISA)

The ELISA tests were performed as previously described (Holmberg et al., 1986a). 96-well microtiter plates were coated for 4 hours with the respective ligand at the indicated concentrations. After saturating the plates with 0.5% bovine serum albumin (BSA) in phosphate-buffered saline (PBS) for 30 minutes, culture supernatants or purified monoclonal antibodies were titrated and left for 2 hours. Bound material was revealed by adding rabbit antimouse IgM antibodies followed by peroxidase-labelled goat anti-rabbit Ig antibodies and substrate. The absorbance (A₄₅₀ nm) was then registered using a Multiscan (Flow

Labs., Irvine, Scotland). All reactions were carried out at room temperature and each step was followed by washing the plates three times with PBS. Only reactions giving an A₄₅₀ value above background + 3 SD were considered as positive (see Fig. 1A and 1B).

EXAMPLE 3

DNA sequencing of hybridomas INTRODUCTION

Distinctly higher degrees of V-region connectivity has been observed within the B cell population of neonatal as compared to adult individuals (Holmberg et al., 1984a;

Kearney et al., 1986; Holmberg et al., 1986a), indicating that an idiotypic network may play an important role for the establishment of the immune system. Studies of preimmune repertoires based on the identification of homology families of V_H-genes have demonstrated that an

overrepresentation of D-proximal V_H-genes is characteristic of the developing immune system (Yancopoulos et al., 1984; Perlmutter et al., 1985; Freitas et al., 1989), implying that early expressed genes encode V-regions with unique functional properties. In support of this suggestion, a correlation was found between utilization of D-proximal V_H genes (7183 and Q52 families) and high degree of idiotypic connectivity within a set of neonatally-derived natural B cell hybridomas (Holmberg, 1987). No such correlation, however, is observed in hybridomas derived from adult individuals (Holmberg et al., 1986a), indicating that properties of newborn antibodies cannot be explained on the basis of V_H-gene utilization only.

In order to analyze the structure-function relationship within the population of natural antibodies, the nucleotide sequence of eight V_H- and V_L- regions expressed by

hybridomas representing this B cell population has been determined. In addition to confirming the germline origin of these clones, the analysis also reveals a limited junctional diversity of neonatally derived hybridomas, and suggests unique structural properties of their H-chain CDR3 regions.

METHODS

Cells The hybridomas used in this study have been described before (Holmberg et al., 1984a; Holmberg et al., 1986a; Holmberg, 1986b; Holmberg et al., 1987). Briefly, BA.N hybridomas are spleen cells from non-immunized, 6 day old BALB/c mice hybridized to the non-secreting SP2/0 cell line using established methods (Kohler and Milstein, 1975) and directly cloned under limiting dilution conditions. Adult hybridomas are derived from spleen cells of 12 week old BALB/c mice stimulated with LPS for 48 hours prior to fusion. Cell lines were grown in RPMI supplemented with 10% fetal calf serum, 5x10-5 mol 2β-mercaptoethanol and 100 μg/ml streptomycin, 100 U/ml penicillin, 10 mM pyruvate and 2 mM L-glutamine. Each cell line was tested for Ig production (and thereby also, indirectly, for Ig transcription) in hemolytic plaque assays and ELISAs (see below) and if necessary the cell lines were subcloned under limiting dilution conditions.

Enzyme linked immunosorbent assays (ELISA)

The ELISA tests were performed as previously described (Holmberg et al., 1986a). In brief, 96-well

microtiterplates were coated for 4 hours with sheep antimouse Ig antibodies. After saturating the plates with 0.5% gelatin in PBS for 30 minutes, culture supernatants were titrated and incubated for 2 hours. Bound material was revealed by adding peroxidase-labelled goat anti-mouse IgM antibodies followed by substrate. The absorbance (A₄₅₀nm) was registered using a Multiscan (Flow Labs.). All reactions were carried out at room temperature and each step was followed by washing the plate in PBS, supplemented with 0.1% Tween 20.

Preparation of RNA and nucleotide sequencing Cells were grown to a density of 10⁶/ml and a total of 5 x 10⁸ cells were harvested and total cellular RNA was

extracted according to Brawerman et al., 1972. Poly(A)⁺-RNA was subsequently selected on an oligo d(T) cellulose column (Pharmacia, Uppsala). Poly(A)+-RNA was then used directly for sequencing using a modified version of the dideoxy sequencing method (Hamlyn et al., 1978) with synthetic primers complementary to the Cμ region (5'- GCTCGCTCTCGCAGGAGAC-3 ' ) or the C_K region

(5'-TGGATGGTGGGAAGATG-3'). cDNA synthesis was terminated by addition of dideoxynucleotide (Sanger et al., 1977) and reaction mixtures were separated on a polyacrylamide gel. The same sequence was determined at least twice with two different radioactive nucleotides (³⁵SdATP, ³⁵SdCTP) to overcome ambiguities. The numbering system of Kabat et al., 1987, was used to number the nucleotide positions.

RESULTS

The nucleotide sequence of the heavy and light chain V- regions from several B cell hybridomas originating from newborns was determined, representing different degrees of relative connectivity and expressing different V_H-gene families. Detailed reactivity patterns of each of these antibodies have been previously reported ( (Holmberg et al., 1984a; Holmberg et al., 1986a; Holmberg, 1986b; Holmberg et al., 1987) and are summarized in Table I. The ligands tested are indicated in the table, and include different self- and nonself-antigens, conventional idiotypes and anti-idiotypes as well as a panel of 70 IgM antibodies derived from the same hybridoma collection as those

investigated here. As can be seen in the table, two of the clones, BA.N 4:4.57 and 4:3.40, show a very high degree of connectivity when tested against other members of the hybridoma collection, while failing to react with most of the heterologous ligands. This is in agreement with the previously reported bias towards idiotypic interactions, rather than general properties of degeneracy (increased nonspecific binding), constituting the basis for the observed connectivity in neonatal B cell repertoires.

a) Fraction of positive reactions in ELISA to the following different antigenic ligands: TNP-BSA, NIP-BSA, PC, Dx B512, Dx 1355, Ac38 and AC146 (both mAb specific for the B1-8 idiotype and obtained from Dr. K. Rajewsky, Cologne),

10/13-15 (mAb specific for the TEPC15 idiotype, obtained from Dr. C. Berek, Cologne), 2-4 and 20-5 (both mAb

specific for the SP603 idiotype, obtained from Dr. P. -A. Cazenave, Paris), IIIA 11C7 (mAb prepared against the hybridoma molecule BA.N 1:1.8), VIB3C20 (mAb prepared against the Dx B512 specific mAb 17-9), anti-J558 (mAb specific for the J558 idiotype, obtained from Dr. J.

Kearney, Birmingham), MOPC 460, J558 and TEPC 15 (mAb derived from myelomas specific for TNP, Dxl355 and PC respectively), F6(51) and 41-27 (both mAb specific for the MOPC460 idiotype, obtained from Dr. P.-A. Cazenave), G5Bb (mAb specific for GAT, obtained from Dr. J. Theze, Paris), G6-3 (mAb prepared against the BA.N 4:4.57 idiotype), H9- 14.8, H40-315.7 (both mAb specific for I-E^k, obtained from Dr. M. Pierres, Marseille), H100-27/55 (mAb specific for K^k, obtained from Dr. H. Lemke, Kiel), 14-4-4 (mAb specific for the la.7 epitope), KJ1.26.1 and KJ12.98.2 (both mAb specific for idiotypic determinants on the antigen-specific T cell receptor, obtained from Dr. P. Marrack, Denver), actin; tubulin, myosin, peroxidase, renin, neurofilament, DNA, acetylcholine receptor isolated from Torpedo californica, mouse thyroglobulin (see Holmberg, 1987, for a detailed description of ligands). b) Fraction of positive reactions in ELISA or RIA with 70 other IgM antibodies within the BA.N collection. c) Nb = Newborn.

d) ND = Not determined.

The nucleic acids sequences of the heavy chain V-regions are presented in Figure 2 and of light chain V-regions in Figure 3. Included in the figures are also the most

homologous V regions found amongst previously reported sequences. The analysis of these nucleotide sequences confirmed their predicted germline origin. Thus, the clone BA.N 1:5.24 expresses a V_K-gene identical to the L7 (Pech et al., 1981) germline gene (a member of the V_K23 subgroup) as far as the sequence could be determined, joined to the J_K2 element. The V_H region of this clone is composed of a V gene identical to the V_H14.29 (Lawler et al., 1987)

sequence (a member of the 7183 V_H-family derived from an Abelson virus transformed pre-B cell line of fetal liver origin), and the same V_H-sequence was also found to be used by the clone BA.N 3:3.39. Although a corresponding germline gene has so far not been isolated, the fact that three clones, independently isolated from different mice,

express this V_H gene, strongly suggests their germline origin. A similar argument could be made for the V_K of the 3:3.39 clone which is identical to the V_K 1.60 (Lawler et al., 1989) sequence (obtained from a fetal liver hybridoma and belonging to the V_K1 subgroup) and for the 4:3.40 clone which is identical to two previously reported sequences of LyIB-related B-cell hybridomas 3B9PC and 5D6S (V_K21

subgroup). Interestingly in this case the sequence identity extends through the complete V_J-region (Förster et al., 1988). For the V_H region of the clone BA.N 4:3.40, as well as the V_L and V_H regions of BA.N 4:1.13, no corresponding germline V gene counterparts could be assigned. There is, however, no evidence for the presence of somatic mutations in these or any of the other hybridomas sequenced here. In every case, the most homologous sequences found differ in at least 10 nucleotide positions, making it more likely that they are derived from separate germline genes rather than somatically derived variants of already known germline genes. In addition, all hybridomas studied here were found to use D- and J-segments of germline origin.

The nucleotide sequence data of these clones, together with the partial sequence analysis of another set of V_H genes from hybridomas of neonatal or adult origin, also revealed a limited N-region diversity. Figure 4 shows the nucleotide sequences of these clones covering the CDR3 region. As can be seen in the figure, five out of eight hybridomas of neonatal origin totally lacked N-sequence additions. Thus, the clones BA.N 3:3.39, BA.N 1:1.8, BA.N 1:3.80, BA.N

4:4.57 and BA.N 4:3.40 totally lacked N-sequences while the BA.N 4:1.13, the BA.N 2:6.80 and the BA.N 1:5.24 contained a limited number of nucleotides 5' and/or 3' of the D segment that could not be assigned to any corresponding, known germline segment.

No obvious correlation of the degree of connectivity to these features exists in this limited number of clones. Rather they appear to reflect a difference between the neonatal and the adult B cell repertoires. Thus both of the hybridomas derived from normal, adult mice analyzed here show extensive N-region diversity (Fig. 4). Furthermore, out of 158 previously reported V_H-sequences (Kabat et al., 1987) derived from hybridomas and myelomas, 148 contains N- regions.

DISCUSSION

Analysis of B-cell repertoires at different points of ontogeny have revealed a developmentally controlled

expression of V_H-genes (Yancopoulos et al., 1984;

Perlmutter et al., 1985; Freitas et al., 1989). Thus, a bias towards utilization of V_H-genes positioned D- proximal on the chromosome is evident in early stages of ontogeny while in adults the representation of different V_H-gene families appear to correlate with their respective complexity (Dildrop et al., 1985). These observations have posed the question whether the observed bias reflects a functional role of D-proximal V_H-genes in the early immune system, or if this phenomenon is solely a fortuitous result of higher accessibility of these genes to rearrangement.

The correlation between high idiotypic connectivity and the utilization of V_H-genes of the most D-proximal families observed in hybridomas of newborn but not of adult origin have previously been described. This association between functional properties of high connectivity and the

utilization of D-proximal V_H-genes suggests that these genes may have been evolutionarily selected for

chromosomal position and structural properties in order to ensure the early appearance of clones constituting a highly connected idiotypic network. The data presented herein further support this view by confirming the germline origin of these antibodies, and by showing the paucity of somatic processes increasing diversity in CDR3. The most interesting observation reported here is the limited occurrence of N-sequences within the pool of natural, neonatal B cell hybridomas. Thus, while less than half of the neonatally derived V_H-sequences studied here contain N-sequences, the overall occurrence in previously reported V_H-genes is more than 90%. This low frequency of N-sequence addition within the neonatal B cell repertoire could indicate that a selective pressure exists to

minimize somatically derived diversity at this stage in development. Consequently, any properties encoded by the germline segments making up the CDR3 region, in particular by the D-region, would be kept. It could be speculated that some molecular mechanism might be specifically acting or lacking in either of the two populations. A possible candidate would be the enzyme terminal deoxynucleotidyl transferase (TdT) the expression of which has been

correlated to the addition of N-sequences (Alt and

Baltimore, 1982). It is interesting to note that similar observations have been made for the TCR,δ cells, in which the variable region of the TCRδ-genes expressed by fetal thymocytes contains comparatively less N-region insertions than those of adult origin (Chien et al., 1987).

The structural characteristics of neonatal antibodies described here may explain their general properties, such as multireactivity and V-region connectivity. In turn, these might be necessary for the early operation of the IS and to provide conditions for appropriate repertoire selection. EXAMPLE 4

Treatment of NOD mice with immunoglobulins

The NOD mouse as a model for diabetes

The non-obese diabetic (NOD) mouse is a well established animal model of human insulin dependent diabetes mellitus (Makino et al., 1980). The onset of overt diabetes in these animals from about three months onwards is preceded by a mononuclear cell infiltration of the islets of Langerhans leading to the selective destruction of insulin-producing beta-cells (Miyazaki et al., 1985). The infiltrating cells include T cells of both CD4+ and CD8+ phenotype,

macrophages and B cells. The disease is prevented by T cell depletion and can be transferred by a combination of CD4⁺ and CD8⁺ splenic T cells from diabetic donors,

demonstrating that the effector stage of the disease is T cell mediated.

Although B cells have also been suggested to take part in the development of disease (Miyazaki et al., 1985; Fujita et al., 1982), it was recently demonstrated that they are not required during the effector stage (Pontesilli et al., 1987). In view of recently presented evidence demonstrating B cell mediated influences on the formation of the

peripheral T cell repertoire in normal mice (Bendelac et al., 1988; Martinez et al., 1984), it was investigated by the present inventor whether B cells/immunoglobulins may influence the development of an auto-aggressive T cell repertoire in the NOD mouse.

The underlying mechanisms for the development of

autoagressive lymphocyte repertoires are, however, still not well understood. In view of recently presented evidence demonstrating B cell-mediated influences on the formation of the peripheral T cell repertoire in normal mice

(Pontesilli et al., 1987; Bendelac et al., 1988), it is investigated by the present inventor that B

cells/immunoglobulins may influence the development of an autoagressive T cell repertoire.

This point has been investigated by suppressing the

development of early B cell repertoire, by repeatedly administering high doses of rabbit anti-IgM antibodies for the first 4 weeks of life. In parallel, similar regimes for injecting normal rabbit immunoglobulins have been used.

MATERIALS AND METHODS Mice

NOD mice were bred in animal facilities at the University of Umea. All mice were tested for glucosuria once a week by using test strips and a colorimetric assay (Glukotest, Boehringer Mannheim). By postnatal week 35, the spontaneous incidence of diabetes (glycosuria) in our colony reached 62% in females and 10% in males.

Anti-IgM suppression

NOD mice were injected intraperitoneally with rabbit antimouse IgM (1 mg/mouse, 3 times/week). Treatment was

initiated within 24 hours after birth and continued for 4 weeks unless otherwise indicated. Normal rabbit Ig or BSA was injected according to the same protocol. The mice were then left untreated until sacrificed.

FACS analysis Spleen cells were tested for expression of Ig by incubating 10⁶ cells with fluorescein-conjugated goat anti-mouse Ig (Becton-Dickinson) (0.08 μg) in 50 μl of PBS-FCS for 30 minutes at 4°C, washed, and resuspended in 1 ml of PBS-FCS containing propidium iodide, and cell surface fluorescence was analyzed on a FACSCAN (Becton-Dickinson). Non-viable cells were excluded by virtue of high propidium iodide fluorescence. 10⁴ viable cells were accumulated for histograms using logarithmic amplification of fluorescence intensity.

Histopathology

Insulitis was determined by histological examination of pancreas. Paraffin sections (4 lm) of formalin-fixed pancreases were stained with hematoxylin and eosin. A minimum of 20 islets per individual were counted and only mice in which no affected islets could be detected were scored as negative.

Polyclonal Ig preparations

Partially purified normal rabbit Ig (normal polyclonal rabbit immunoglobulin) was obtained by precipitating normal rabbit serum twice with ammonium sulphate followed by extensive dialysis against PBS. Rabbit anti-mouse IgM

(rabbit anti-mouse IgM) was obtained in a similar fashion but using immune serum from rabbits hyperimmunized with mouse IgM antibodies.

RESULTS

Lower incidence of overt diabetes after normal polyclonal rabbit immunoglobulin or rabbit anti-mouse IgM treatment

Aiming at establishing the possible influence that B cells/Ig may have on the development of autoreactive T cell repertoires in the newborn period of NOD mice, high doses of normal polyclonal rabbit immunoglobulin or rabbit anti- mouse IgM antibodies were injected repeatedly for 4 weeks, starting treatment within the first 24 hours of life. At 4 weeks of age after this treatment, less than 0.5% Ig bearing cells could be detected in the spleen from mice treated with rabbit anti-mouse IgM antibodies, as monitored by cytofluorographic analysis with a fluorescein-conjugated goat anti-mouse Ig antibody (data not shown). The effect of such treatment on the development of overt diabetes in female NOD mice is shown in Figure 5. Both rabbit anti-mouse IgM and normal polyclonal rabbit

immunoglobulin clearly reduced the incidence of diabetes compared to untreated animals. In contrast, treatment with BSA injected according to the same protocol did not alter the normal incidence of disease, excluding the possibility that any protein would mediate this effect.

Development of insulitis

During the development of diabetes in the NOD mouse, mononuclear infiltration of the pancreatic islets is observed as early as 5 weeks after birth (Bottomly et al., 1980; and Figure 6). The incidence of insulitis is then gradually increasing with age to reach 100% in both males and females at around 10 to 12 weeks. As indicated in the figure, the development of insulitis precedes overt diabetes by 10 to 15 weeks. In contrast to overt diabetes, insulitis is observed in 100% of both males and females. Also the severity of insulitis among the affected

individuals progresses with age. At 15 weeks, the majority of mice have more than 50% of the islets infiltrated with mononuclear cells, and at 25 weeks all individuals tested showed severe insulitis with more than 70% of islets affected. Histological examination to determine incidence and degree of insulitis allows for a more rapid evaluation of possible effects on the development of pathology, and this was used throughout this study in addition to scoring incidence of overt diabetes. Reduction of insulitis by normal polyclonal rabbit

immunoglobulin and rabbit anti-mouse IgM

Mice treated with normal polyclonal rabbit immunoglobulin and rabbit anti-mouse IgM as described above were then sacrificed at different ages and their pancreases examined for lymphocyte infiltration. In each age group, the mean frequency of infiltrated islets was calculated and the results compared to untreated, age-matched controls. As can be seen in Figure 7, both types of treatment reduces the severity of insulitis in accordance with the effect seen on overt diabetes, and although treatment is interrupted at 4 weeks of age, the effect is persistent up to 30 weeks or more. However, treatment with rabbit anti-mouse IgM

results in a more pronounced reduction of insulitis and a more prolonged effect than observed after normal polyclonal rabbit immunoglobulin treatment. These data suggest that different mechanisms may underly the two types of

treatment.

Different requirements for duration of treatment with normal polyclonal rabbit immunoglobulin and rabbit anti- mouse IgM

In order to further characterize the discrepancy observed with normal polyclonal rabbit immunoglobulin and rabbit anti-mouse IgM, the effect of different duration of the treatment was analyzed. Female NOD mice were treated for different time periods ranging from one single injection at birth up to continuous treatment for the first 4 weeks of life. At the age of 15 weeks, the mice were sacrificed and the degree of insulitis was measured. The results shown in Figure 8 reveal two distinct effects obtained after

different durations of treatment with normal polyclonal rabbit immunoglobulin and rabbit anti-mouse IgM,

respectively. While the effect of one single injection is the same both for normal polyclonal rabbit immunoglobulin and rabbit anti-mouse IgM and also reflects the optimum effect of normal polyclonal rabbit immunoglobulin, the effect of rabbit anti-mouse IgM increases with the duration of treatment up to at least 4 weeks.

DISCUSSION It was found that both of these experimental protocols lead to a reduction in the incidence of overt diabetes as well as in the incidence and severity of insulitis. A

significantly better effect on the development of insulitis is, however, observed after anti-IgM suppression,

indicating that the neonatal B cell repertoire is involved in the process resulting in auto-aggression in the NOD mouse.

It is well established that the neonatal period plays a decisive role in the establishment of the mature repertoire of T and B cells. During this period, processes involving interactions between B and T cells based on idiotypic interactions have been suggested to contribute

significantly to repertoire selection. The demonstration that the neonatal B cell compartment contains unique specificity repertoires (Martinez et al., 1985; Dighiero et al., 1985) including a strong bias for V-region

connectivity (Dighiero et al., 1985; Holmberg et al.,

1984a; Holmberg et al., 1986b) has also been proposed to constitute the basis for network control of the development of the normal immune system (IS). In view of this it appears plausible that T cell mediated autoimmune processes may depend on such mechanisms influencing the selection of T cell receptor specificities.

At the effector phase, this disease has been shown to depend exclusively on T cells. However, in the light of the recent demonstration of Ig-dependent selection of T cell repertoires limited to the first three weeks of life, it could be argued that B cells may play a critical role in the early establishment of the autoimmune T cell

compartment.

To approach the possible impact of the neonatal repertoire on the development of autoimmunity and the possible role of B cells in T cell mediated autoimmunity, the B cell

development in NOD mice was suppressed with rabbit anti- mouse IgM antibodies from birth until 4 weeks of age, thereby not allowing Ig influence on T cell development during this period. When sacrificed 11 weeks later, these mice show a significant reduction of insulitis and, among positive individuals, a lower degree of insulitis compared to untreated, age-matched controls.

Suppressed mice rapidly develop normal numbers of B cells upon interruption of treatment. Together with the

observation that rabbit anti-mouse IgM suppression during the first 4 weeks of life is sufficient to mediate a significant reduction in the T cell mediated autoimmune process in NOD mice, this implies that the observed effect is due to properties of the neonatal system, suggesting a direct involvement of the neonatal B cell repertoire in the development of T cell mediated autoimmunity.

The present observation, if interpreted in this framework, could also be relevant for the understanding of the

mechanisms mediating the well documented effects of

intravenously administered polyclonal natural IgG on different forms of human autoimmune disorders (Imbach et al., 1981, Sultan et al., 1984).

EXAMPLE 5

Treatment of NOD mice with monoclonal antibodies Materials and methods were as in Example 4 except for: Monoclonal antibody preparations

The monoclonal antibodies used here were all of the IgM isotype and have previously been described. BA.N 1:1.8, BA.N 1:5.24, BA.N 3:2.2, BA.N 4:1.13, BA.N 4:3.40 and BA.N 4:4.57 were serologically characterized as idiotypically connected and polyreactive, while the BA.N 1:6.30 was found not to react with any ligand tested so far. All these antibodies were derived from B cell hybridomas isolated from spleen cells of normal, 6-day-old BALB/c mice. TNP. II is a TNP specific antibody isolated after a fusion of a TNP-KLH immunized BALB/c mouse.

All IgM antibodies were precipitated from cell supernatant twice, using PEG 6000 (Polyetylen glycol, approx Mw 6 kD, Riedel de Haen). Further purification was performed by gelfiltration on a TSK 3000 SWG column, 21.5 x 600 mm

(Pharmacia,LKB) using phosphate buffered saline as mobile phase. The antibody containing fraction was concentrated in an Omega concentration cell, Mw cutoff 30 kD (Filtron). The concentrated antibody solution was filtrated through a 0.2 μm Spin-x centrifuge filter (Costar) and stored at -80°C until use.

Fab-fragment preparations

Purified monoclonal antibodies were diluted in PBS to a concentration of 1 mg/ml and the following were added: L- cysteine (sigma Chemical Co, St Louis) to a final

concentration of 10 mM and EDTA (Riedel de Haen) to a final concentration of 5 mM and 0.01 % v/v Tween 20 (Sigma

Chemical Co). The pH was adjusted to 7.0 with 1 M NaOH and Papain (Sigma Chemical Co) was added to a w/w ratio of 1:100 papain/ antibody. The digestion was conducted in room temperature for 8 h and terminated with the addition of 5 mM iodoacetamide. The fragments obtained by the papain digestion were separated by gelfiltration on a TSK 300 SWG column (Pharmacia,LKB, Uppsala, Sweden). The peak with retention time corresponding to the expected molecular weight of Fab fragments (63kD) was collected and

concentrated in an Omega Cell (Filtron). Commassie stained SDS-polyacrylamide gels run under non-reducing conditions revealed one band at 63 kD. The concentrated F_ab fragments were filtered through a 0.2 μm Spin-x centrifuge filter (Costar) prior to injection.

Treatments

NOD mice were injected intraperitoneally with a mixture of monoclonal antibodies 1 mg/mouse, 3 times/week). Treatment was initiated within 24 hours after birth and continued for 4 weeks. As a control corresponding doses of BSA were given for the first 4 weeks of life. The mice were thereafter left untreated until sacrificed. NOD mice were also given a single intraperitoneal injection of either 100 μg of a single monoclonal antibody or 50 μg of F_ab-fragments within 24 hours after birth.

Adult (9 weeks old) NOD mice were injected intravenously with one dose of either 100 μg or 1 mg monoclonal antibody. RESULTS

Effect of high dose treatment of NOD mice using monoclonal natural antibodies

The observed result from treatment with polyclonal,

isogenic Ig mimics the beneficial effects of pooled, normal IgG on certain human autoimmune disorders. It could be hypothesized that both these effects are mediated by properties characteristic of the natural antibody

population. In order to test this hypothesis, a set of monoclonal, natural antibodies isolated from normal, neonatal BALB/c mice were used for treatment. 7 monoclonal antibodies were selected, 6 of which had previously been characterized as polyreactive and highly "idiotypically connected" (BA.N 1:1.8, BA.N 1:5.24, BA.N 3:2.2, BA.N

4:1.13, BA.N 4:3.40 and BA.N 4:4.57) and one that lacked these properties (BA.N 1:6.30). Figure 9 shows the results of the repeated administration, during the first 4 weeks of life, of high doses of a mixture of all 7 monoclonal antibodies. A similar reduction in the incidence of diabetes was observed as previously found after treatment with polyclonal Ig. To rule out the possibility that the observed results were due to

unspecific general effect exerted by any protein antigen administered in this fashion, a control group was given BSA according to the same protocol. As shown in the figure, this had no effect on the incidence of the disease.

Low doses of some monoclonal, natural antibodies

effectively inhibit development of diabetes in the NOD mouse

The fact that pooled monoclonal antibody species could mediate inhibition of development of diabetes in the female NOD mouse suggested that the effect could be mediated by one or several of the V-region specificities represented in this collection. Alternatively, as all the monoclonal antibodies tested were of the same isotype, IgM, properties of the IgM constant region could be responsible. In order to address this issue, the different monoclonal antibodies were injected individually, including an additional antibody of the same isotype but originating from a

conventional, TNP-specific B cell hybridoma (TNP 11.).

On the basis of their effects, the monoclonal antibodies could be divided into three groups: i) TNP.11 and BA.N

1:6.30 had no significant effect on the incidence of disease (Fig. 10 A). In contrast, the BA.N 1:1.8, BA.N 1:5.24 and BA.N 4:4.57 antibodies showed a dramatic effect (Fig. 10 B). The three remaining antibodies tested were found to have an intermediary effect (Fig. 10 C).

Treatment of NOD mice with Fab-fragments

As shown in Figure 11 treatment of NOD mice with one single injection of Fab-fragments prepared from the monoclonal antibody BA.N 1:1.8 had no effect on the development of diabetes. This clearly shows that the effect obtained with treatment with the complete antibody BA.N 1:1.8 (Figure 10 B) can not be attributed to a direct inhibitory effect on any of the steps in the interaction between T-cells and antigen presenting cells by the antibodies. The effect seen with the antibodies BA.N 1:1.8, BA.N 1:5.24 and BA.N

4:4.57 must instead be interpreted as an effect on the immune system as a whole and its ability to control self reactivity. These results does not exclude the possibility that other fragments prepared or derived from these

antibodies or any other idiotypically connected antibody may be used in the treatment of autoimmune diseases.

Treatment of adult NOD mice Figure 12 shows the result of administration, at 9 weeks of age, of two different doses of the monoclonal antibody BA.N 1:5.24. The lower dose, 100 μg, shows no effect on the development of disease, the higher dose, 1 mg, however, shows a reduction in the incidence of disease.

EXAMPLE 6

Treatment of lpr mice with monoclonal antibodies

The Lpr-mouse as a model for SLE

The autosomal recessive gene lpr (lymphoproliteration) was first observed in the 12th generation of brother x sister matings during the development of the MRL strain which originated with a series of crosses involving strains

AKR/J, C57BL/6J, C3H/Di, and LG/J (Murphy and Roths, 1978). The expression of the lpr gene in MRL/MpJ-lpr/lpr (MRL- lpr/lpr)² mice leads to massive generalized lymph node enlargement due to the proliferation of T cells. Such lymphadenopathy is notably associated with the early onset of fatal systemic lupus erythematosus (SLE), with a 50% cumulative mortality rate at 5 to 6 months of age, whereas MRL/MpJ-+/+ (MRL-+/+) mice lacking the lpr mutation have no lymphoproliferation and develop SLE in their second year of life (Murphy and Roths, 1978; Andrews et al., 1978).

Mice

B6-lpr/lpr mice were bred in animal facilities at the

Swedish Defence Research Establishment, Umea.

Serological assays

Antibodies to single-stranded DNA (ssDNA) and double- stranded DNA (dsDNA) were measured with an ELISA assay. Plates were coated with 100 μl of 10 μg/ml of ssDNA or dsDNA in citrate buffer, pH 6, and bound IgM was revealed by biotinylated goat anti-mouse IgM, avidin-alkaline phosphatase and substrate. The absorbance at 405 nm was the registered using a Multiscan (Flow Labs., Irvine, Scotland). For determinations of rheumatoid factor

activity, ELISA assays similar to those described above were used, coating plates with 100 μl of a solution

containing 12 μg/ml of F_c portions of purified mouse IgG₁. Serum samples were added to the wells in dilution from 1/32 to 1/512. RESULTS AND DISCUSSION

As can be seen in Figure 13 treatment of lpr mice with one injection of monoclonal antibody BA.N 1:1.8 interfered with the development of autoimmune disease as measured as titers of rhematoid factor (RF) and antibodies against single- stranded DNA (DNA ss) and double-stranded DNA (DNA ds), treatment with BA.N 1:5.24 on the other hand had no effect in this experiment.

EXAMPLE 7 Analysis of immunoglobulin V_H genes of neonatal BALB/c mice

INTRODUCTION

The utilization of members of the different V_H gene

families appears to be developmentally controlled. Thus, V_H gene segments from the most D-proximal V_H gene families have been shown to be preferentially used in Abelson virus-transformed pre-B cell lines (Yancopoulos et al

1984), and in hybridomas (Perlmutter et al 1985) as well as normal B cells of perinatal origin (Freitas et al., 1989). This seems to result from differentiated V_H gene

accessability to rearrangement. In contrast to the

perinatal repertoire, the representation of each VH gene family in adult splenic B cells corresponds approximately to it's relative size. In turn, such "normalization" appears to result from cellular selection.

The perinatal skewing of V_H gene repertoires could be a result of a fortuitous positioning of certain V_H gene segments on the chromosome, but it could as well be argued that D-proximity of certain V_H genes has been

evolutionarily selected to ensure their expression in early development. Analyses of neonatal B cell hybridomas have indeed demonstrated that functional properties of

multireactivity and idiotypic connectivity are correlated to utilization of D-proximal V_H genes. Moreover, N-region diversity in combination with germline gene expression within these clones suggested that random somatic

diversification would be minimized at this stage of

development.

We have here attempted to ascertain this hypothesis

analyzing the V-region diversity of newborn (Nb) and adult (Ad) B cell populations using the polymerase chain reaction (PCR) method to determine the nucleotide sequence of a large number of V_HD_HJ_H regions. The analysis was based on amplification of genomic DNA and restricted to members of the 7183 and J558 VH gene families by using 5' primers specific for either of these gene families.

MATERIALS AND METHODS

Mice

BALB/c mice were bred in animal facilities at the

University of Umea. Isolation and amplification of genomic DNA

Spleens of newborn (<24 h after birth) or 14 weeks old BALB/c mice were minced and washed in ice-cold PBS and then lysed in 1 ml of digestion buffer (100 mM NaCl, 10mM

TrisHCl, 25 mM EDTA (pH 8.0), 0.5% SDS and 0.1 mg/ml proteinase K) for 12 h at 50 oC. After phenol extraction and ethanol precipitation the DNA was digested with BamHl and was then used as template for PCR amplification. The DNA was dissolved in 50 μl 35 mM TrisHCl (pH 8.3), 50 mM KCl, 2.5 mM MgCl2, 200 μM dNTP's, 1 μM both primers and 4.0 units Tag polymerase. PCR amplification was performed on a thermal cycler (Hybaid Ltd) with 25 cycles of 45 sec at 94 °C, 1 min at 55 °C for 7183 V_H genes and 60 °C for J558 V_H genes and 45 sec at 72 °C per cycle. After ethanol precipitation, the reaction product was digested with NotI and Xhol restriction endonucleases and thereafter applied to a 1% Agarose gel. A band with the correct size was cut out, the DNA electroeluted and ligated into pBluescript (KS-) vector (Stratagene) and transformed into E. coli. The oligonucleotide primers were :

V_H 7183, 5'CGCGCGGCCGCGTGGAGTCTGGGGGAGGCTTA 3',

V_H J558, 5'CGCGCGGCCGCTGCAGCAGCCTGGGGCTGAG 3' and

J_H4, 5'GCGCTCGAGGAGACGGTGACTGAGGTT 3'. Screening of PCR libraries and DNA seguencing

About 500 ampicillin resistant colonies from each library was screened by colony hybridization using radioactively labelled cDNA probes, pVH81X and pVHB4 specific for the 7183- and J558- V_H gene families respectively. Plasmid DNA from positive colonies, was sequenced by the dideoxymethod using Sequenase (U.S. Biochemicals) or T7-polymerase

(Pharmacia), 7-deaza-2'-dGTP (Pharmacia) and T3 and T7 sequencing primers.

RESULTS

Diversity in V_H7183-D_H-J_H junctions of neonatal and adult origin

The comparison of junctional regions of the adult versus newborn sequences revealed that both the contribution of exonuclease nibbling of the terminals of the V_H and J_H segments as well as presumable random addition of

N-sequences was found to be less pronounced in sequences of neonatal as compared to adult origin. We also noted that while in the isolated clones of neonatal origin all sequences utilized the J_H4 gene segment, 4 of 56 adult sequences contained a J_H1 segment. As the J_H primer used for PCR amplification was designed to be homologous to the J_H4 segment, it is not surprising that the majority of the clones obtained utilized this gene segment. Evidently, however, the sequence homology to the J_H1 gene segment was enough to amplify also sequences containing the J_H1 segment under the stringency conditions used here. Although it can not be ruled out that the observed discrepancy between the neonatal and adult collections of sequences is due to the limited number of neonatal sequences analyzed, it remains a possibility that J_H1 containing sequences may be less frequent in the neonatal 7183 V_H repertoire than in the adult.

Utilization of D-segments in unusual codon reading frames

The analysis of the junctional regions also revealed a high frequency of neonatal V_H-D_H-J_H joinings that had placed the D-segment in reading frame 2 or 3. Thus wile 78% of the adult sequences analyzed here, and more than 80% of

sequences found in the literature use D-segments in the first codon reading frame (Ichihara et al 1989), this was found in less than 50% of the neonatal sequences. It has been suggested that the more frequent utilization of reading frame 2 and 3 in incomplete D_H-J_H joinings may indicate that the observed bias in functional joinings is the result of cellular selection (Ichihara et al. 1989). The uneven distribution of codons, particularly those encoding hydrophobic amino acids (a.a.), red in the three different reading frames (Figure 4) may be interpreted to support this argument. Thus, while hydrophobic a.a. are extremely rare in D-segments red in the first reading frame ( 7 of 95 codons) they are clearly overrepresented in the second and third reading frame ( 94 of 98 and 46 of

respectively). DISCUSSION

Nucleotide sequence analysis of a large number of

rearranged immunoglobulin heavy chain V-regions derived from 1 day old BALB/c mice demonstrate a very restricted diversity as compared to 14 weeks old mice. This is the result of usage of a limited number of individual V_H gene segments, together with a lower degree of junctional diversity in the neonatal as compared to the adult

repertoire. These data support a model of selection in the early development of the IS of distinct germline encoded specificities with minimal contribution of mechanisms generating random diversity. EXAMPLE 8

Analysis of N-region diversity in normal and NOD mice

INTRODUCTION

To create a functional Ig heavy chain an ordered

rearrangement process is performed involving the

juxtapositioning of a diversity (D) gene segment to a junctional (J_H) gene segment followed by the joining of a variable (V_H) gene segment to this DJ_H complex (Tonegawa, 1983). During this rearrangement process, exonuclease nibbling of the terminals of the joining gene segments, together with template-independent addition of nucleotides (N-nucleotide), contribute to the generation of V-region diversity. The addition of N-nucleotides in the V_H-D gene and the D-J_H junctions is thought to be mediated by a template independent DNA polymerase, terminal

deoxynucleotidyl transferase (TdT) (Desiderio et al., 1984), shown to be specifally expressed in lymphoid precursors (Bollum, 1979). N-region diversity is more pronounced in adult as with compared neonatal B cells, (Holmberg et al., 1989, Carlsson and Holmberg, 1990, Gu et al., 1990, Feeney, 1990, Meek, 1990). We have here compared the junctional regions from functional and non-functional V_HDJ_H rearrangements derived from splenic B cells of neonatal or adult mice. These analyses reveal that a selection against N-sequence

additions occurs both in adult and neonatal individuals. This selection is evident amongst 7183 but not J558 V_H rearrangements, suggesting that this process is dependent on V_H rearrangements, suggesting that this process is dependent on V_H gene specificity.

MATERIALS AND METHODS

Mice BALB/c, C57BL/6 and NOD mice were bred in animal facilities at the University of Umea.

Isolation and amplification of genomic DNA, screening of PCR libraries and DNA sequencing were performed as

described in Example 7.

RESULTS AND DISCUSSION

Analysis of V_HDJ_H junctions of B cell hybridomas, or primary B cells, of neonatal and adult origin have revealed that a limited N-region diversity is associated with B cells of fetal and neonatal origin (Holmberg et al., 1989, Carlsson and Holmberg, 1990, Gu et al., 1990, Feeney, 1990, Meek, 1990). The detailed analysis of the nucleotide sequences of a large number of rearranged Ig V_H-regions derived from adult or neonatal mice confirmed these

observations (Figure 14). Genomic DNA extracted from spleen cells of BALB/c, C57B1/6 or NOD mice, was amplified using 77183 V_H- or J558

V_H-specific primers together with a J_H specific primer by polymerase chain reaction (PCR). The vast majority (80-87%) of functional, neonatal rearrangements utilizing 7183 or J558 V_H genes were found to lack N- sequences in the D-J_H junction. In contrast, this was the case only in a minority (21-40%) of the adult sequences.

As the 3'end of the majority of the V_H gene segments utilized were unknown, the analyses were restricted to the D-J_H junction. However, in the cases where N-seguences could be identified in the V_H-D junction, a similar

discrepancy between neonatal and adult clones was observed (data not shown). As non-functional rearrangements should be representative of a non-selected pool of V-regions while functional rearrangements could be subjected to V-region specific selection, the frequency of N-sequence additions in these two populations was subsequently compared. As illustrated in Figures 15A and 15B, the lack of N-sequence additions was found to be more pronounced among the functional as compared to the non-function rearrangements of 7183 V_H genes. Thus, in adult BALB/c the frequency of clones lacking N-sequence additions in functional and

non-functional rearrangements were 40% and 14%

respectively. Similarly, in C57B1/6 25% and 8% of the functional and non-functional rearrangements,

respectively, were found to lack N-sequences.

A similar discrepancy was observed between functional and non-functional rearrangements of neonatal origin. The respective frequencies of the neonatal rearrangements were 87% and 80% for the functional and 25% and 14% for the non-functional rearrangements. These observations suggest that the lack of N-sequence additions observed in distinct B cell populations may in fact be due to a selective advantage for clones with these properties.

A more detailed analysis of the data revealed that the observed selection against N-region diversity was most pronounced amongst the rearrangements utilizing the 7183.1 V_H gene (Table 2). Thus, 10 of 10 functional rearrangements of BALB/c origin utilizing this VH gene were found to lack N-sequences in the D-J_H junction. Similarly, 5 of 5 rearrangements isolated from C57B1/6 mice, and utilizing the corresponding V_H gene segment (7183.1^b), lacked

N-sequence additions at this site.

In contrast to the observed selection against N-sequence additions amongst 7183 V_H utilizing rearrangements of

BALB/c and C57B1/6 mice, no such discrepancy was observed between functional and non-functional 7183 rearrangements of NOD mice (Figure 16). Thus, 89% and 100% of functional and non-functional rearrangements respectively, isolated from neonates, and 50% and 38% of functional and

non-functional rearrangements isolated from adults lacked N-sequences. The 7183.1 has been shown to preferentially rearranged in early B cells of fetal liver as well as adult bone marrow origin (Yancopoulos et al., 1984, Lawler et al., 1987). In Example 7 it was shown that although this preference for rearrangements of the 7183.1 gene remains in the adult individual, a strong selection against functional

rearrangements of this gene exists in the developmental process of a primary B cell repertoire. The apparently programmed utilization of D-proximal V_H genes thus,

appears to be dependent on the combination of a mechanistic preference for rearrangement of D-proximal V_H genes

together with cellular selection mechanisms favouring certain V_H gene specificities at different developmental stages. The observations made here suggest that this process is paralleled by a selection against N-region diversity within the population of specificities preferred in early ontogeny.

These observations could be interpreted as to reflect the assurance of the expression of a set of specificities making up the earliest B cell repertoire. This could appear to be the contribution of, on the one hand, an evolutionary selected preference for the rearrangement of certain V_H genes, e.g. the 7183.1 V_H gene, together with a relatively low TdT activity in early ontogeny as compared to the adult individual. On the other hand, a developmentally controlled selection mechanism acting on the B cells through specific interactions with their respective V-regions. While it remains an open guestion what would be the basis for such selection mechanisms, we note that unique

functional properties of multireactivity and connectivity have been ascribed to B cell clones utilizing D-proximal V_H genes, particularly of the 7183 family (Holmberg, 1987). It is tempting therefore to speculate that such properties are vital for the developing immune system.

In this context, it is interesting to note that the 7183 V_H family of the mouse on the basis of sequence homology has been categorized in a subset of V_H gene sequences derived from different species, shown to constitute the most evolutionary conserved group of V_H genes (Tutter and Riblet, 1989, Schroeder et al., 1990), and to be

preferentially expressed in early ontogeny of both mice and humans (Yancopouloss et al. 1984, Schroeder and Wang,

1990).

The fact that the selection against N-sequence additions is not observed amongst 7183 rearrangements of NOD origin, could be hypothesized to reflect a disturbance in the process of normal primary B cell repertoire establishment. It is tempting to speculate that such disturbances in the establishment of B cell repertoires in the NOD mouse could have implications on the development of an autoaggressive lymphocyte repertoire.

REFERENCES

- Alt, F. and Baltimore, D. 1982. Joining of

immunoglobulin heavy chain gene segments:

Implications from a chromosome with evidence of three D-JH fusions. Proc. Natl. Acad. Sci. USA 79:4118.

- Andrews, B.S., R.A. Eisenberg, A.N. Theofilopoulos,

S. Izui, C.B. Wilson, P.J. McConahey, E.D. Murphy, J.B. Roths, and F.J. Dixon. 1978. Spontaneous murine lupus-like syndrome. Clinical and immunopathological manifestations in several strains. J. Exp. Med.

148:1198. - Bendelac, A., C. Boitard, P. Bedossa, H. Bazin, J.-F.

Bach, and C. Carnaud. 1988. Adoptive T cell transfer of autoimmune nonobese diabetic mouse diabetes does not require recruitment of host B lymphocytes. J. Immunol. 141:2625. - Berek, C., Jarvis, J.M. and Milstein, C. 1987.

Activation of memory and virgin B cell clones in hyperimmune animals. Eur. J. Immunol. 17:1121. - Bollum, F. 1979. Terminal deoxynucleotidityl

transferase as a hemapoetic cell marker. Blood.

54:1203.

- Bottomly, K., CA. Janeway Jr., B.J. Mathieson, and

D.E. Mosier. 1980. Absence of an antigen-specific helper T cell required for the expression of the T15 idiotype in mice treated with anti-μ antibody. Eur. J. Immunol. 10:159. - Brawerman, G., Mendecki, J. and Lee, S.Y. 1972. A procedure for the isolation of mammalian messenger ribonucleic acid. Biochemistry 11:637. - Brodeur, P. and Riblet, R. 1984. Eur. J. Immunol. 14:922. - Carlsson, L. and Holmberg, D. 1990. Genetic basis of the neonatal antibody network: high connectivity correlates with limited N-region diversity and germline V-gene expression. Int. Immunol. 2:639 - Chien, Y., Iwashima, M., Wettstein, D.A., Kaplan, K.B., Elliott, J.F., Born, W. and Davies, M.M. 1987. T cell receptor δ gene rearrangements in early thymocytes. Nature. 330:722. - Coutinho, A. 1989. Beyond clonal selection and

network. Immunol. Rev. 110:6. - Dighiero, G., Lyberi, P., Holmberg, D., Lundkvist,

I., Coutinho, A. and Avrameas, S. 1985. High

frequency of natural autoantibodies in normal newborn mice. J. Immunol. 134:765. - Desiderio, S.V., Yancopoulos, G.D., Paskind, M.,

Thomas, E., Boss, M.A., Landau, N., Alt, F.W. and Baltimore, D. 1984. Insertion of N-regions into heavy-chain genes is correlated with expression of terminal deoxytransferase in B cells. Nature.

311:752. - Dildrop, R. 1984. A new classification of mouse V_H sequences. Immunology Today. 5:85. - Dildrop, R., Krawinkel. U., Winter, E. and Rajewsky, K. 1985. V_H-gene expression in murine

lipopolysaccharide blasts distributes over the nine known V_H-gene groups and may be random. Eur. J.

Immunol. 15:1154. - Förster, I., Gu, H. and Rajewsky, K. 1988. Germline antibody V region as determinants of clonal

persistence and malignant growth in the B cell compartment. EMBO J. 7:3693. - Freitas, A.A., Lembezat, M.-P. and Coutinho, A. 1989.

Expression of antibody V-regions is genetically and developmentally controlled and modulated by the B lymphocyte environment. Int. Immunol. 1:342.

- Fujita, T., R. Yui, Y. Kusimoto, Y. Serizawa, S.

Makino, and Y. Tochino. 1982. Lymphocyte insulitis in a "non-obese diabetic (NOD)" strain of mice: an immunohistochemical and electron microscope

investigation. Biomed. Res. 3:429. - Gerondakis, S., Boyd, A., Bernard, 0., Webb, E. and Adams, J.M. 1984. Activation of immunoglobulin y gene expression involves stepwise demethylation. EMBO J. 3:3013. - Gu, H., Förster, I. and Rajewsky, K. 1990. Sequence homologies, N sequenceinsertion and J_Hgene utilization in

V_HDJ_H joining: implications for the joining mechanism and the ontogenetic timing of Lyl B cell and B-CLL progenitor generation. EMBO J. 9:2133 - Hamlyn, P.M., Brownlee, G.G., Cheng, C.C., Gait, M.J.

and Milstein, C. 1978. Complete sequence of constant and 3' noncoding regions of an immunoglobulin mRNA using the dideoxynucleotide method of RNA sequencing. Cell 15:1067. - Holmberg, D. et Coutinho, A. 1985. Natural antibodies and autoimmunity. Immunology Today. 12: 356. - Holmberg, D., Forsgren, S., Ivars, F. and Coutinho, A. 1984a. Reactions amongst IgM antibodies derived from normal, neonatal mice. Eur. J. Immunol., 14:435.

- Holmberg, D., Forsgren, S., Forni, L., Ivars, F. and Coutinho, A. 1984b. Proc. Natl. Acad. Sci. USA.

81:3175. - Holmberg, D., Freitas, A., Portnoi, D., Jacquemart, F., Avrameas, S. and Coutinho, A. 1986a. Antibody repertoires of normal BALB/c mice: B lymphocyte populations defined by state of activation. Immunol. Rev. 23:147. - Holmberg, D., Wennerström, G., Andrade, L. and

Coutinho, A. 1986b. The high idiotypic connectivity of "natural" newborn antibodies is not found in adult, mitogen-reactive B cell repertoires. Eur. J. Immunol. 16:82. - Holmberg, D. 1987. High connectivity, natural

antibodies preferentially use 7183 and QUPC 52 V_H families. Eur. J. Immunol. 17:399. - Holmberg, D., Andersson, A., Carlsson, L. and

Forsgren, S. 1989. Establishment and functional

implications of B-cell

connectivity. Immunol. Rev. 110:89

Ichihara, Y. Hayashida, S. Myazawa, S. and Kurosawa, Y. 1989. Only DFL16, DS72 and DQ52 gene families exists in mouse immunoglobulin heavy chain diversity gene loci, of which DFL16 and DSP2 originate from the same primordiar P_H gene. Eur. J. Immunol. 19:1849. - Imbach, P., Barandun, S., d'Apuzzo, V., Baumgartner,C., Hirt, A., Morell, A., Rossi, E., Schöni, M., Vest,

M. and Wagner, H.P. 1981. High dose intravenous gammaglobulin, for idiopathic thrombocytopenic purpura in childhood. Lancet.

1:1228.

- Kabat, E., Wu, T.T., Reid-Miller, M., Perry, H.M. and Gottesman, K.S. 1987. In Sequences of Proteins of Immunological Interest. US Government Printing

Office, Bethesda, MD. - Kearney, J.F., Vakil, M. 1986. Non-random V_H gene

expression and idiotype-anti-idiotype expression on early B cells. In Kelsoe, G., Schulze, D. eds.,

Evaluation and Verteral immunity: The Antigen

Receptor and MHC Gene Families. Texas University Press, Austin. - Köhler, G. and Milstein, C. 1975. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature. 256:495. - Lawler, A.M., Lin, P.S. and Gearhart, P.J. 1987.

Adult B-cell repertoire is biased toward two heavy- chain variable-region genes that rearrange frequently in fetal pre-B cells. Proc. Natl. Acad. Sci. USA. 84:2454. - Lawler, A.M., Kearney, J.F., Kuehl, M. and Gearhart, P.J. 1989. Early rearrangements of genes encoding murine immunoglobulin K chains, unlike genes encoding heavy chains, use variable gene segments dispersed throughout the locus. Proc. Natl. Acad. Sci. USA. 86:6744. - Lundkvist, I., Coutinho, A., Varela, F. and Holmberg, D. 1989. Evidence for a functional idiotypic network among natural antibodies in normal mice. Proc. Natl. Acad. Sci. USA. 86:5074. - Makino, S., R. Kunimoto, Y. Muraoka, Y. Mitsuschima, K. Katagiri, and Y. Tochino. 1980. Breeding of a non- obese, diabetic strain of mice. Exp. Anim. (Tokyo) 29:1. - Martinez-A., C., P. Pereira, R.R. Bernabe, A.

Bandeira, E.-L. Larsson, P.-A. Cazenave, and A.

Coutinho. 1984. Internal complementarities in the immune system: Regulation of the expression of helper T-cell idiotypes. Proc. Natl. Acad. Sci. USA.

81:4520. - Martinez-A., C, R.R. Bernabe, A. de la Hera, P.

Pereira, P.-A. Cazenave, and A. Coutinho. 1985.

Establishment of idiotypic helper T-cell repertoires early in life. Nature. 317:721. - Miyazaki, A., T. Hanafusa, K. Yamada, J. Miyagawa, H.

Fujino-Kurihara, H. Nakajiama, K. Nonaka, and S.

Tarul. 1985. Predominance of T lymphocytes in

pancreatic islets and spleen of prediabetic non-obese diabetic (NOD) mice: a longitudinal study. Clin. Exp. Immunol. 70:84. - Murphy, E.D., and J.B. Roths. 1978. Autoimmunity and lymphoproliferation: induction by mutant gene lpr, and acceleration by a male-associated factor in strain BXSB mice. In Genetic Control of Autoimmune Disease. N.R. Rose, P.E. Bigazzi, and N.L. Warner, eds. Elsevier/North Holland, New York, p. 207. - Pech, M., Hochtl, J., Schnell, H. and Zachau, H.G.

1981. Differences between germ-line and rearranged immunoglobulin V_K coding sequences suggest a

localized mutation mechanism. Nature. 291:668

- Perlmutter, R., Kearny, J., Chang, S. and Hood, L.

1985. Developmentally controlled expression of Ig V_H genes. Science 227:1597.

- Pontesilli, O., P. Carotenuto, L.S. Gazda, P.F.

Pratt, and S.J. Prowse. 1987. Circulating lymphocyte populations and autoantibodies in non-obese diabetic (NOD) mice: a longitudinal study. Clin. Exp. Immunol. 70:84. - Sanger, F., Nicklen, S. and Coulson, A.R. 1977. DNA sequencing with chain termination inhibitors. Proc. Natl. Acad. Sci. USA 74:5463. - Schroeder, H.N., Hillson, J. L., and Perlmutter, R.

M. 1990. Structure and evolution of mammalian V_H families. Int. Immunol. 2:41. - Schroeder, H. and Wang, J.Y. 1990. Preferential utilization of conserved immunoglobulin heavy chain

variable gene segments during human fetal life. Proc. Natl. Acad. Sci. USA. 87:6146

- Sultan, Y., Kazatchkie, M.D., Maisonneuve, P. and Nydegger, U.E. 1984. Anti-idiotypic suppression of

autoantibodies to Factor Vlllby high-dose intravenous gammaglobulin. Lancet. 2:765.

- Sundblad, A., Heusser, S., Holmberg, D., Cazenave,

P.-A. and Coutinho, A., 1989. Suppression of antibody responses to the acetylcholine receptor by natural antibodies. Eur. J. Immunol. 19:1425. - Tonegawa, S. 1983. Somatic generation of antibody diversity. Nature. 302:575. - Tutter, A. and Riblet, R. 1989. Conservation of an immunoglobulin variable-region gene family indicates a specific, non-coding function. Proc. Natl. Acad. Sci. USA. 86:7460. - Walter, M.A., gibson, W.T., Ebers, G.C. and Cox, D.W. 1991. Susceptibility to multiple sclerosis is associated with the proximal immunoglobulin heavy chain variable region. J. Clin. Invest. 87:1266

- Yancopoulos, G.D., Desiderio, S.V., Paskind, M.,

Kearney, J.F., Baltimore, D. and Alt, F.W. 1984.

Preferential utilization of the most J_H proximal V_H gene segments in pre-B cell lines. Nature 311:727. - Yang, P.-M., Olsen, N.J., Siminovitch, K.A., Olee, T., Kozin, F., Carson, D. and Chen, P.J. 1990. Possible deletion of a developmentally regulated heavy- chain variable region in autoimmun diseases. Proc. Natl. Sci. USA. 87:7907.

Claims

1. The use of at least one mammalian idiotypically

connected antibody or a specifically binding fragment thereof for the preparation of a pharmaceutical composition for the treatment or prevention of an autoimmune disease in mammals, including humans, with the proviso that when the mammal is a mouse, the antibody is not an auto-idiotypic antibody directed against a murine antibody against the murine acetyl choline esterase receptor.

2. The use according to claim 1 wherein the idiotypically connected antibody or antibodies constitute at least 80% by weight of the antibodies used for the preparation.

3. The use according to claim 2 wherein the idiotypically connected antibody or antibodies constitute at least 90% by weight of the antibodies used for the preparation.

4. The use according to claim 3 wherein the idiotypically connected antibody or antibodies constitute at least 95% by weight of the antibodies used for the preparation.

5. The use according to claim 4 wherein the idiotypically connected antibody or antibodies constitute substantially

100% by weight of the antibodies used for the preparation.

6. The use according to any one of the preceding claims wherein the antibodies are polyclonal antibodies.

7. The use according to any one of the preceding claims wherein the antibody is a monoclonal antibody.

8. The use according to any one of claims 1-6 wherein the antibodies are a a mixture of monoclonal antibodies.

9. The use according to any one of the preceding claims wherein the antibody or antibodies is/are antibodies the V_H gene segment of which is encoded by a member of the group consisting of the V_H gene segments belonging to the murine V_H gene families 71.83, 36-60 or Q52 and the human V_H gene family V_H3 and the corresponding gene families of other mammals.

10. The use according to any one of the preceding claims wherein the antibody or antibodies is/are antibodies which are encoded by genes which are devoid of template- indepedent insertions (N-sequences) or in which the

template-independent insertions comprise less than 5 such nucleotides.

11. The use according to any one of the preceding claims in which the antibody/antibodies is/are derived from B- lymphocytes producing naturally expressed idiotypically connected antibodies, or are antibodies which are identical or substantially similar to said antibodies .

12. The use according to any one of the preceding claims in which the antibody or antibodies is/are selected from monoclonal antibodies BA.N 3:2.2, BA.N 4:3.40, BA.N 4:1.13, BA.N 4:4,57, BA.N 1:5.24, and BA.N 1:1.18.

13. The use according to any of the preceding claims in which the antibodies are antibodies the variable part of which is encoded by any of the DNA sequences shown in Figs 2 and 3 or by sequences which are at least 80% homologous therewith and capable of encoding the variable part of an idiotypically connected antibody.

14. The use according to any of the preceding claims in which the disease is a condition related to the function of the thyroid gland, in particular hyper- or hypofunctioning of said gland, e.g. thyroiditis (acute, subacute or

chronic), including Hashimoto's disease (lymphocytic thyroiditis; lymphoid thyroiditis), Riedel^,s thyroiditis (chronic fibrous thyroiditis), de Quervain's thyroiditis (subacute granulomatous thyroiditis), subacute lymphocytic thyroiditis, Graves' disease, Graves' subacute thyroiditis and Graves' ophthalmopathy or a condition related to hypofunction of the adrenal glands, in particular Addison's disease or furthermore Simmonds' panhypopituitarism, arthritis such as rheumatoid arthritis including juvenile rheumatoid arthritis, psoriatic arthritis or Reiter

arthritis, ankyloses spondylitis, Crohn's disease or ulcerative colitis, especially for the prevention of relapse or progression of Crohn's disease or ulcerative colitis, or for the treatment of pernicious anemia or celiac disease, systemic lupus erythematosus, multiple sclerosis, sarcoidosis, primary biliary cirrhosis,

autoimmune hemolytic anemia, immune thrombocytopenia purpura, myasthenia gravis, glomerulonephritis and

Goodpasture's syndrome.

15. The use according to any of claims 1-13 wherein the disease is diabetes mellitus type 1.

16. A method of treating or preventing an autoimmune disease comprising administering to a patient in need of such treatment a therapeutically or prophylactically amount of a monoclonal antibody according as characterized in any one of the claims 1-15.