EP0215775A1

EP0215775A1 - Immunosorbent for removal of rheumatoid factors from whole blood or blood plasma

Info

Publication number: EP0215775A1
Application number: EP85901390A
Authority: EP
Inventors: Ole Nordfang
Original assignee: Nordisk Gentofte AS
Current assignee: Nordisk Gentofte AS
Priority date: 1985-03-13
Filing date: 1985-03-13
Publication date: 1987-04-01
Also published as: WO1986005397A1

Abstract

Les facteurs rhumatoïdes sont retirés du sang ou du plasma sanguin par absorption sur un immunosorbant sous la forme d'une matrice à base d'un matériau poreux auquel est couplé de l'IgG monomère natif ou modifié. La matrice utilisée est de préférence un gel ouvert avec un poids moléculaire d'exclusion dépassant 106 kD, dépassant de préférence 107 kD, en particulier un gel d'agarose. Si l'on utilise un tel immunosorbant, la teneur en facteurs rhumatoïdes du sang de patients souffrant d'arthrite rhumatoïde ou d'autres infections auto-immunes peut être abaissée au niveau du plasma normal.Rheumatoid factors are removed from blood or blood plasma by absorption on an immunosorbent in the form of a matrix based on a porous material to which is coupled native or modified IgG monomer. The matrix used is preferably an open gel with an exclusion molecular weight exceeding 106 kD, preferably exceeding 107 kD, in particular an agarose gel. If such an immunosorbent is used, the level of rheumatoid factors in the blood of patients with rheumatoid arthritis or other autoimmune infections may be lowered to normal plasma.

Description

Immunosorbent for removal of rheumatoid factors from whole blood or blood plasma

The present invention concerns an immunosorbent useful for reducing or removing rheumatoid factors from whole blood or blood plasma.

Rheumatoid factors (RF) are autoantibodies having speci- ficity to immunoglobulin G (IgG). These antibodies may be of both IgM and IgG class and are present i.a. in the blood of patients with rheumatoid arthritis and other autoimmune connective tissue diseases. Rheumatoid factors are believed to have importance for the formation of immune complexes causing e.g. arthritis and vasculitis (H.E. Jasin et al., Clinical Immunobiology b ^~_, p. 365, Ed. F.H. Bach, 1975). Jasin et al. describes a plurality of analytical methods for RF, in particular IgM RF, based on the binding of RF to IgG, in particular for use in differential diagnosis of rheumatic diseases.

Many attempts have been made to reduce the concentration of RF in blood from patients by the so-called plasma- pheresis, where the patient's plasma is removed from the cells during extracorporeal- circulation, which are infused together with an isotonic saline solution or isooncotic liquids (J.V. Jones et al.: . Lab.Clin.Med. 97, 589, 1981). This unspecific method only enables removal of a few litres of plasma at a time.

Consequently, it is not possible to reduce the concentra¬ tion of RF significantly (L.H. Calabrese: Cleveland Clinic Quarterly _7, 53, 1980; R. Schlansky et al.: Arthritis and Rheumatism 7Λ_, 49, 1981) and particularly not to carry out a specific reduction. In haemophiliacs with anti Factor IX antibodies (IgG) it has been possible to reduce the anti Factor IX titre by absorption treatment of patients' plasma or blood. I.M. Nilsson et al. (Blood 5_8, 38, 1981) has treated plasma from a Factor IX inhibitor patient by extracor¬ poreal protein A Sepharose CL-4B absorption. This un- specific absorption removes 80?ό of the total IgG and thus also IgG anti Factor IX. As demonstrated below, also some RF absorption takes place. C. Freiburghaus et al. (Thrombosis and Haemostasis _5TJ, 208, 1983) has treated whole blood from a Factor IX inhibitor patient with Factor IX coupled to Sepharose 6MB. This reduced the Factor IX inhibitor titre specifically from 120 to 7.2 Bethesda U/ml plasma.

Purification of RF, isolated from serum samples, by immunoabsorption by affinity chromatography with a view to analysis has been described by several authors. Thus, N.E. Bianco et al. (Clin.Exp. Immunol. 11_. 91, 1974) has purified RF by coupling to heat-aggregated human IgG Sepharose 6B, and W.J. Koopman et al. (Arthritis and Rheum. 2 _^, 302, 1980) has developed a sensitive radioimmunoassay based on purification of RF by coupling to heat-aggregated human IgG Bio Gel A15m with 5 mg of coupled IgG/ml. The rheumatoid factors were released by elution with an acid buffer solution. Heat-aggregated

IgG was used for the mentioned immunoabsorptions of isolated RF because the binding constant of RF for heat- aggregated IgG is relatively high (10 1/mole), whillιe the binding constant for native IgG is very low (102 - - 11..11 xx 1100⁴ 1/mole) (D. Dissanayake: Immunology 32, 309, 1977).

It is proposed in the DE-A 27 25 608 to remove specific factors, including antibodies such as IgG, anti-DNA antibodies and rheumatoid factors, from blood plasma by absorption on an immunosorbent, which consists of an insoluble carrier to which specific antigens to the antibodies concerned are bound. Examples of suitable carriers are stated to be cross-linked dextrans and polyacrylic amides, cellulose derivatives and agarose. To remove rheumatoid factors it is proposed - in good accordance with the above account of binding constants - to use aggregated IgG. That such removal is feasible in practice is not shown in that specification, however. In contrast, it is shown in Table 3 below that aggregated IgG couples poorly to agarose gels, so that no effective removal of rheumatoid factors can be obtained.

Conclusively it may be said that the art comprises pro- cesses for unspecific reduction of the RF concentration in blood or plasma based on plasmapheresis and for un¬ specific IgG removal by immunoabsorption on protein A Sepharose. Finally, it is known to remove antibodies of high avidity, such as IgG anti Factor IX, by specific immunoabsorption. The possibility of removing RF by specific absorption on aggregated IgG is mentioned in the DE-A 27 25 608. However, as mentioned, it is not demonstrated in the specification that such removal is feasible in practice.

Likewise, a plurality of methods of purifying and ana¬ lysing isolated RF is known, some of which are based on immunoabsorption by affinity chromatography . These methods have all focussed on isolated RF and not on the absorbed serum.

Thus, no effective process is known for specific removal or reduction of the content of antibodies of low avidity such as RF from whole blood or blood plasma.

The object of the present invention is to provide an immunosorbent enabling such a specific removal or reduc¬ tion of RF. The invention is based on the surprising finding that coupling of substantially monomeric IgG (native IgG or chemically or enzymatically modified IgG) to a matrix of a porous material provides an immuno¬ sorbent which allows effective specific removal of both IgM RF and IgG RF from plasma and whole blood without simultaneous removal of normal IgG and IgM.

It is extremely surprising that monomeric IgG coupled to a porous matrix, such as an agarose gel (Sepharose), can be used for removing rheumatoid factors, since the binding constant of rheumatoid factors to monomeric IgG is low (up to 1.1 x 10 1/mole, D. Dissanayake: Immunology 3_2_.> 309, 1977): with this low binding constant the skilled person would expect that only about 25 % of RF from 1 ml of plasma can be removed with 1 ml of Sepharose gel on which 13 mg of IgG (corresponding to the normal plasma level of IgG) are coupled since

By incubation of 1 ml of plasma (about 13 mg of IgG/ml) with 1 ml of gel (jgG] is = 13 mg/ml = 8.7 x 10^"5 mole/1 (Mini: 150,000).

^_rβjq^ = 8,7 x 10^"5 x 1,1 x 10⁴ = 0,96 'Ml,0=

CRF-IgG3 d RF , i.e. about 50?o' of RF in plasma is bound to IgG. Thus, it should only be possible to bind 25?ό of RF from plasma to IgG on the gel, since the plasma has the same IgG-level as the gel.

With an RF content in plasma corresponding to a Rose- Waaler titre ( RW titre) of 1920 (S.G. Andersson: Bull, Wld. Hlth. Org. 42_, 311, 1970), one would thus expect a drop to -an RW titre of 1440. It is shown in Table 2, line 4 below that a drop to a Rose-Waaler titre of 60 can surprisingly be obtained.

The immunosorbent of the invention is accordingly charac¬ terized by comprising a matrix of a porous material to which substantially monomeric IgG is coupled.

The invention also concerns a process for reducing or removing RF from blood or blood plasma, said process being characterized by using as immunosorbent a matrix of a porous material to which substantially monomeric IgG is coupled.

As mentioned, the IgG used for the coupling with a view to removal of RF is monomeric, preferably native human IgG. This is surprising because if the skilled person would think of examining the possibility of effectively removing RF by using IgG as antigen by immunoabsorption, he would expect on the basis of the above-mentioned literature that heat-aggregated IgG would be the preferred possibility because of the considerably higher binding constant. As appears from the following tests and discus¬ sion the monomeric IgG is far superior.

In certain patients, intravenous injection of purified IgG can activate the complement system by binding comple- ment Clq. In such cases it may be an advantage to use chemically or enzymatically modified, monomeric human IgG to prevent complement binding. Chemical reduction of IgG, e.g. by dithiothreitol or mercaptoethanol and subsequent alkylation e.g. by means of iodoacetamide or enzymatic modification e.g. by means of pepsin or plasmin to change the IgG conformation, thus reducing the Clq binding, is well-known, cf. the DE-C2 23 11 333,

IgG from other animal species, such as horse, sheep, rabbit and monkey, also react with RF (R.M. Pope: J. Lab. Clin. Med. 9J_, 842, 1981) and such IgG, too, can therefore be used in the immunosorbent.

The matrix in the immunosorbent is a porous material, such as a gel, preferably an open gel with an exclusion molecular weight exceeding 10 kD, in particular exceeding 10 kD. The matrix must be compatible with an extra- corporal blood circulation.

Preferred gels are agarose gels, in particular the so- called Sepharose gels, preferably Sepharose CL-2B or

Sepharose CL-4B, but also gels of acrylic polymers can be used.

To obtain sufficient RF absorption it is usually necessary to couple IgG to the matrix with a density exceeding 5 mg/ml, preferably exceeding 10 mg/ml, in particular exceeding 20 mg/ml.

It is known that rheumatoid factors also per se can have antinuclear antibody (ANA) - activity (V. Agnello:

J.Exp.Med. L51_, 1514, 1980; K. Hannestad: Nature, 275,

671, 1978), and that RF can form part of immune complexes (F.C. Hay: Ann. Theu . Dis. 3j3_.> 1, 1979). Therefore, also ANA activity and immune complexes will be reduced by adsorption of RF on the described matrices.

A. RF absorption with IgG coupled to various matrices

To examine the absorption on various matrices, native human IgG in an amount of 15 mg of bound IgG/ml gel IgG to the gels listed in Table 1, where "Eupergit C" is an acrylic amide gel, while the various "Sepharose" types are all agarose gels, where the numerical value represents the dry matter content in per cent. The coupling was effected analogously with example 1 below,

For comparison purposes, use was made of partly a Sepharose CL-2B without coupled IgG and partly Protein A Sepharose CL-4B, which, as described in the foregoing, binds immunoglobulins unspecifically and is known as an agent for purification of IgG.

1 ml of plasma and 1 ml of gel were used, and the follow¬ ing results were obtained.

TABLE 1 RF absorption with IgG coupled to various matrices

Matrix Exclusion ab:sorbed serum molecular weight RW-titre mg IgG/ml mg IgM/ l

Eupergit C 10⁶kD 192 17 4.0

Sepharose 6MB 4 x 10⁶ 192 14 4.2

Sepharose 6B 4 x 10⁶ 192 16 4.0

Sepharose 4B 2 x 10⁷ 96 16 3.6

Sepharose CL-4B 2 x 10⁷ 48 12 4.2

Sepharose CL-2B 4 x 10⁷ 24 16 4.0 Protein A Seph.CL-4B -96 2 3.9 Seph.CL-2B without IgG 384 17 4.0

The present IgM RF is determined by the Rose-Waaler titre. A Rose-Waaler titre below 40 corresponds to normal plasma. It will thus be seen that all the gels result in a reduction with respect to the starting titre of 384, and that the best effect is obtained with "Sepharose CL-2B". It is remarkable that the absorption is specific, so that normal IgG and IgM are not also removed by the absorption. For purposes of comparison, the effect of the absorption with protein A Sepharose is shown. This gel reduces the IgG content in the plasma significantly, and the bound IgG is capable of binding some rheumatoid factor. The protein A Sepharose, however, does not absorb RF as effectively as IgG coupled directly to Sepharose CL-2B.

B. Absorption of RF as a function of coupled IgG amount

Variation of the amount of monomeric IgG coupled to Sepharose CL-2B (Table 2) allowed complete removal of Rose-Waaler activity by coupling of more than 20 mg of IgG/ml, and IgG RF was reduced to normal plasma level (0. Nordfang: J. Immunol. Methods 4 , 87, 1981). As stated before, effective removal of RF could not be expected on the basis of the binding constant of rheumatoid factors to monomeric IgG.

TABLE 2 Absorption of RF as a function of coupled IgG amount

mg bound IgG/ml gel After absorption with After absorption med IqG-CL-2B IqG-CL-4B

CL-2B CL-4B RW-titre IgG-RF U RW-titre IgG-RF U

0 0 1920 5.5 1920 6.0

4 4 480 4.5 960 5.2

7 7 120 4.0 480 4.5

13 13 60 3.5 240 4.0

20 21 < 30 3.5 120 9.6

30 34 < 30 3.2 60 3.5

48 41 < 30 3.2 < 30 3.1

Normal values < 40 3.3

In the literature mentioned previously, the authors have concentrated on IgM RF. IgG RF can be shown in a predominant part of patients with juvenile rheumatoid arthritis (Florin-Christensen et al.: Ann. Theum. Dis. (1974) 33, 32), and the efficiency of the present process with respect to removal of IgG RF is therefore of great practical importance.

C. Comparison between native and heat-aggregated IgG

It is shown in Table 3 below that a more effective removal of RF is not obtained by heat-aggregating IgG. On the contrary, the fact is that the RF-binding component in heat-aggregated IgG (polymeric IgG) can be coupled to Sepharose CL-2B, only very weakly. Heat-aggregated IgG was produced by placing a 10?ό IgG solution in a

60°C water bath for 2 hours. Polymeric IgG was isolated from heat-aggregated IgG by gel filtration on Sepharose CL-4B.

TABLE 3

Coupling IgG % polymer mg coupled RW-titre per ml gel * after abs. native IgG < 1 7.2 48 heat-aggregated IgG 4 5.6 48 polymeric IgG > 95 1.4 192 control without IgG - - 384

* determined by Kjeldahl's method. 10 mg of IgG/ml gel were used

The surprisingly effective removal of RF with monomeric IgG coupled to Sepharose, in spite of the low binding constant of rheumatoid factors to monomeric IgG may be due to several reasons. The coupling of the monomeric IgG may cause the individual IgG molecules to change their conformation, so that the binding constant to RF is increased. The coupling may also have caused the IgG molecules to be seated so closely that an RF immuno¬ globulin is bound to more than one IgG molecule.

EXAMPLE 1

a) Preparation of the immunosorbent

20 g of Sepharose CL-2B are washed with 100 ml of distilled H„0 on a glass filter. To the drained gel is added 10 ml of distilled H„0, and pH is adjusted to 11 with 6 M NaOH. Two g. of CNBr dissolved in 20 ml of distilled H₂0 are added. pH is kept at 10.5 to 11.0 with 6 M NaOH. The acid development stops after 5 min., and the gel is washed on a glass filter with 100 ml of distilled H₂0 and 200 ml of 0.1 M NaHC0₃, 0.5 M NaCl, pH = 8.3 (buffer A). Eight ml of native human IgG (100 mg/ml in buffer A) are immediately added and incubated with the gel for 2 hours. IgG Sepharose CL-2B is washed with 40 ml of buffer A and incubated overnight after washing with 40 ml 1 M glycine in buffer A. IgG Sepharose CL-2B is washed with 40 ml of buffer A, 40 ml of 0.1 M glycine/HCl, pH 2.5, 50 ml of PBS (50 mM Na phosphate, 0.15 M NaCl, pH 7.35) and 100 ml of 0.9?,. NaCl, 0.5 unit heparin/ml. The immunosorbent prepared contains 20 mg of bound IgG/ml.

b) Removal of RF

After this treatment 1 ml of gel is incubated for 1 hour with 1 ml of plasma in which the Rose-Waaler titre is 2560. The plasma is washed out of the gel with 0.9?ό NaCl, 0.5 unit of heparin/ml. The Rose-Waaler titre in the absorbed plasma is below 40. E X AMP L E 2

Use of chemically modified monomeric IgG

IgG in a concentration of 50 mg/ml was incubated with between 0 and 20 mM dithiothreitol (DTT) for 30 minutes at room temperature. Then iodoacetamide was added in a concent-ration corresponding to the double of the DTT concentration. After 30 minutes at room temperature DTT and iodoacetamide were removed from the IgG solu¬ tion by dialysis against buffer A and then coupled to Sepharose CL-2B, as described in example 1. 15 mg of IgG/ml og gel were coupled. Then 1 ml of gel was incubated with 2 ml of plasma in which the RW titre is 1280. With the procedure described in example lb the results listed in Table 4 were achieved. As appears from Table 4 it was possible to obtain an immunosorbent which absorbs RF and only at some degree complement Clq. When IgG was reduced with 5 mM DTT, an immunosorbent could be produced which bound more RF and less Clq than protein A Sepharose.

TABLE 4 Effect of weak DTT reduction of monomeric IgG on Clq binding and RF absorption

DTT amount RW-titre Clq mM after abs u/ml al Fter abs

0 160 <0, .2

1.25 320 0, .20

2.5 320 0 .40

5 320 0 .50 10 320 0 .50 20 640 0 .50

Starting plasma 1280 1.0

Prot.A Sepharose 640 0.34

Claims

P A T E N T C L A I M S

1. An immunosorbent useful for reduction or removal of rheumatoid factors from blood or blood plasma, c h a r a c t e r i z e d by comprising a matrix of porous material to which substantially monomeric IgG is coupled.

2. An immunosorbent according to claim 1, c h a r a c t e r i z e d in that the IgG coupled to the matrix is unmodified.

3. An immunosorbent according to claim 1, c h a r a c t e r i z e d in that the IgG coupled to the matrix is chemically or enzymatically modified.

4. An immunosorbent according to any of claims 1, 2 or 3, c h a r a c t e r i z e d in that the matrix has an exclusion molecular weight exceeding 10 kD, preferably exceeding 10 kD.

5. An immunosorbent according to claims 1-4, c h a r a c t e r i z e d in that the matrix is selected among agarose gels, preferably -Sepharose CL-2B or Sepharose CL-4B.

6. An immunosorbent according to claims 1-5, c h a r a c t e r i z e d in that the substantially monomeric IgG is coupled to the matrix with a density exceeding 5 mg/ml, preferably exceeding 10 mg/ml, in particular exceeding 20 mg/ml.

7. A process for reducing or removing rheumatoid factors from blood or blood plasma by absorption on an immuno¬ sorbent, c h a r a c t e r i z e d by using as immunosorbent a matrix of a porous material to which substantially monomeric IgG is coupled.

8. A process according to claim 7, c h a r a c t e r i z e d in that the IgG coupled to the matrix is unmodified.

9. A process according to claim 7, c h a r a c t e r i z e d in that the IgG coupled to the matrix is chemically or enzymatically modified.

10. A process according to claim 7, c h a r a c t e r i z e d in that the matrix employed has an exclusion molecular weight exceeding 10 kD, preferably exceeding 10 kD.

11. A process according to claims 7-9, c h a r a c t e r i z e d in that the substantially monomeric IgG is coupled to the matrix with a density exceeding 5 mg/ml, preferably exceeding 10 mg/ml, in particular exceeding 20 mg/ml.

12. A process according to claim 10, c h a r a c t e r i z e d in that the matrix employed is selected among agarose gels, preferably Sepharose CL-2B or Sepharose CL-4B.