вход по аккаунту


Monoclonal antibody 6b6.6 defines a cross-reactive kappa light chain idiotope on human monoclonal and polyclonal rheumatoid factors

код для вставкиСкачать
Mouse monoclonal antibody (MAb) 6B6.6 was
raised against a cross-reactive idiotope (CRI) present on
the light chains of 2 human IgM paraproteins with
rheumatoid factor (RF) activity. The MAb inhibited the
IgG-binding activity OF these proteins, and thus appears
to react with an epitope located at or near the RFbinding site. Enzyme-linked immunosorbent assay
(ELISA) and Western immunoblotting studies indicate
that the 6B6.6 CRI is associated with d I I a subsubgroup light chains, is not related to the Wa, Po, and
Bla RF cross-idiotypic specificities, and is clearly distinct from the dIIb-associated CRI detected by MAb
17.109. Using an ELISA, we detected 6B6.6 CRI in 59%
of 107 sera and 48% of 50 synovial fluids From patients
with seropositive rheumatoid arthritis (RA). However,
the quantities OF CRI-positive RF were small, and the
amount of CRI-positive RF did not correlate with the
amount of IgM-RF. The 6B6.6 CRI was shown to occur
primarily id the IgM fraction of RA sera by both
chromatographic studies and isotype-specific ELISA,
although small quantities appeared to be associated with
IgA and IgG in some sera. The presence of 6B6.6 CRI
on both monoclonal and polyclonal RF is consistent with
- ..
From the Division of Clinical lmmunology and Rheumatology. Department of Medicine. University of Alabama School of
Medicine, The University of Alabama at Birmingham, and the
Birmingham Veterans Administration Medical Center, Birmingham,
Supported by NIH grants AR-03555 and AI-18745 and by
the Veterans Administration Research Program.
Ralph E. Schrohenloher, PhD; Mary Ann Accavitti, PhD;
Ajit S. Bhown. PhD; William J. Koopman. MD.
Address reprint requests to Ralph E. Schrohenloher, PhD.
Division of Clinical Immunology and Rheumatology The University
of Alabama at Birmingham, UAB Station. Birmingham. AL 35294.
Submitted for publication March 31, 1989; accepted in
revised form September 15. 1989.
Arthrltls and Rheumatism, Vol. 33, No. 2 (February 1990)
the view that both are derived, at least in part, from a
common gene pool. However, its occurrence in relatively low levels suggests that the number of germline
genes encoding For R F is large or that extensive mutation occurs in the course of RF expression in RA.
Rheumatoid factors (RF) are a heterogeneous
group of autoantibodies that bind the Fc portion of
IgG. Although found with greatest frequency and at
highest titers in patients with rheumatoid arthritis (RA)
and other connective tissue diseases, RF also occur in
a variety of other conditions including certain liver,
pulmonary, and infectious diseases. Moreover, transient R F production may accompany secondary immune responses in healthy individuals (for review, see
ref. 1).
It is now clear that RF activity can be associated with IgG and IgA, as well as IgM, immunoglobulins (2-6); however, the proportions of RF isotypes
expressed in different diseases or among individuals
with the same disease can vary widely (6-9). Similarly,
RF exhibit considerable heterogeneity with respect to
their interaction with IgG. Several antigenic determinants distributed among the Fc fragments of the 4
subclasses of IgG have been shown to interact with RF
in humans (10). In addition, RF show variable degrees
of cross-reactivity with IgG from other species ( 1 1)
and with certain nonimmunoglobulin antigens, including deoxyribonucleic acid and nucleoprotein (12,13).
Since idiotypic determinants are frequently associated with the binding site, idiotypic analysis can
serve as a sensitive indicator of differences among
antibodies that have the same or similar specificities,
such as RF. Indeed, several RF idiotypic groups have
been identified using rabbit aniisera or mouse mono-
clonal antibodies (MAb) raised against human monoclonal paraproteins with RF activity, which occur in
some patients with certain lymphoproliferative diseases and other conditions (14-19). When investigated, these antisera or antibodies have appeared to
identify RF subsets that occur with limited frequency
among patients with polyclonal RF (16-21).
In seeking additional shared RF idiotypes that
may be expressed in the polyclonal repertoire, we
recently compared the idiotypic properties of 6 human
monoclonal I ~ M KRF paraproteins, using rabbit antisera prepared against each protein (22). Two of the
proteins, COR and LEW, demonstrated a prominent
V, shared idiotype that appeared to differ from previously described RF idiotypic groups. Evidence was
also obtained for the occurrence of this cross-reactive
idiotype (CRI) among polyclonal RF present in RA
Because of the limitations in using heterologous
antisera, in terms of specificity and reproducibility by
different groups of investigators, we undertook the
preparation of mouse MAb to the shared idiotypic
determinants present on COR and LEW IgM-RF (23).
Herein we describe the production of a MAb to a
major idiotypic determinant on the light (L) polypeptide chains of these paraproteins and its reaction with
polyclonal RF from patients with RA.
Monoclonal RF and other proteins. DAU and GRA
IgM and COR, LEW, POM, LAY, and GLO IgM-RF
paraproteins were isolated from plasma by precipitation 3
times as euglobulins and were further purified by gel chromatography through Sephacryl S-300 (Pharmacia, Piscataway, NJ) in O.1M sodium acetate buffer containing O.5M
NaCl (pH 3.8). MIL IgM-RF and SCH polymeric IgA-RF
were isolated from sera or plasma as described elsewhere
Patients and controls. Serum and synovial fluid samples were obtained from patients attending the outpatient
rheumatology clinics of the Department of Medicine, the
University of Alabama at Birmingham, who fulfilled the
American Rheumatism Association criteria for definite or
classic RA (26). Control sera were from healthy adult
volunteers with no history of rheumatic disease.
Heavy and light immunoglobulin chains. Immunoglobulin heavy (H) and L chains of the IgM used for
immunization and enzyme-linked immunosorbent assay
(ELISA) were prepared under nondenaturing conditions.
Interchain disulfide bonds of the proteins were cleaved by
reaction with 10 mM dithiothreitol (DTT) in O.ISM NaCl
buffered at pH 7.4 with 20 mM phosphate buffered saline
(PBS) for 3 hours at room temperature. Free sultlydryl
groups were then alkylated by I-hour reaction with a small
excess of iodoacetamide. The reduced and alkylated proteins were dissociated into the component H and L chains by
dialysis into 1M acetic acid and fractionated by gel chromatography through Sephadex G-100 (Pharmacia) in 1M acetic
acid (27).
COR L chain used in ELISA was passed through a
Sepharose 4B immunosorbent column (Pharmacia) containing covalently bound F(ab'), fragments of affinity-purified
rabbit antibody specific for human p chain (Jackson ImmunoResearch, Avondale, PA) (28), in order to remove small
quantities of H chain contaminants. Similarly, the COR H
chain was passed through an immunosorbent column prepared with the F(ab'), fragments of affinity-pudied goat
antibody specific for human K chain, to remove small quantities of L chains in the preparations.
COR, LEW, and DAU IgM L chains used for aminoterminal amino acid sequence analysis were prepared from
IgM that was reduced in 6M guanidine hydrochloride (GuHCI), 0.275M Tris HCI, and 0.2mM EDTA (pH 8.2) with 20
mM DTT for 1 hour at 37°C (29) and alkylated with 44 mM
recrystallized iodoacetamide for 1 hour at 37°C. L chains
were isolated by gel chromatography through a Sephadex
G-100 column (2.5 x 95 cm) in SM GuHCI, 1M acetic acid
(29). Because the supply of MIL IgM was limited, aminoterminal sequence analysis was performed on MIL L chain
obtained under nondenaturing conditions, after further reduction and alkylation in 6M GuHCl using the same conditions as those used for COR, LEW, and DAU IgM.
Monoclonal antibody. (DBA x BALB/c)F, mice were
immunized with 100 pg of COR L chain in Freund's complete adjuvant, injected subcutaneously, as described by
Lieberman et a1 (30). Three weeks after the primary injection, the mice were given a booster of 50 pg of LEW L chain
in O.1SM NaCI. Three days after the challenge, lymphocytes
from the popliteal and inguinal lymph nodes were fused with
X63-Ag8.653 myeloma cells (31). according to the method of
Kearney (32). Approximately 14 days after fusion, culture
supernatants obtained from Ig-secreting cells were screened
by ELISA for the presence of Ig that was reactive with COR,
LEW, and DAU L chains. Hybridomas producing antibody
that reacted with the L chains from COR and LEW, but not
with L chains from DAU (a human monoclonal IgMK protein
without RF activity) were subcloned, expanded in cell
culture, and grown in ascites by standard procedures. Monoclonal antibody 6B6.6 was selected for further characterization. Its isotype was determined to be IgGlK by ELISA,
using alkaline phosphatase-conjugated goat anti-mouse Ig
isotype-specific antibodies (Southern Biotechnology Associates, Birmingham, AL).
The MAb was isolated from cell-free ascites fluid by
affinity chromatography on a protein A-Sepharose CL-4B
column (1 x 5 cm; Pharmacia), according to the manufacturer's instructions. One milliliter of ascites fluid diluted in 1
ml of 0.15M glycine, 3M NaCl buffer (pH 8.9) was applied to
the column, which had been previously equilibrated with the
same buffer. The column was then washed free of unbound
proteins with the starting buffer, and the MAb was recovered
by elution with 0.1M sodium citrate buffer (pH 6.0).
ELISA. The activity of MAb 6B6.6 was investigated
using both a direct-binding ELISA and an RF-inhibition
ELISA. Flat-bottom polystyrene microtiter plates (Immulon
2; Dynatech, Alexandria, VA) were used for all assays. All
steps were performed at room temperature. PBS containing
10 mg/ml bovine serum albumin (BSA) was used as the
diluent for samples and for all reagents except coating
solutions, which were prepared in PBS. Wells were aspirated and then washed 3 times with 0.05% Tween 20 in PBS
between steps of the assay.
The direct-binding ELISA was performed by coating
the wells with 0.25 ml of purified protein or isolated Ig
polypeptide chain (0.01 mg/ml) for 2 hours at room temperature. Remaining protein-binding sites were blocked by
incubation with 0.4 ml of 1 mg/ml BSA in PBS for 1 hour.
Duplicate 0.25-ml samples of dilutions of ascites fluid containing the MAb were placed in the wells and incubated for
16 hours. Experiments with isolated MAb were performed
using 0.2-ml samples (100 ng/ml) and an incubation time of 3
hours. Wells for the latter experiments were coated with 0.2
ml of protein solution (0.01 mg/ml).
Bound antibody was detected by reaction for 1 hour
with 0.25 ml (or 0.2 ml) of alkaline phosphataseconjugated
affinity-purified F(ab’), fragments of goat anti-mouse IgG
antibody diluted 1 :5,000 (H chain- and L chain-specific;
Jackson ImmunoResearch), followed by reaction with 0.25
ml (or 0.2 ml) of 0.05M sodium carbonate buffer (PH 9.8)
containing 1 mM magnesium chloride and 1 mg/ml of disodium p-nitrophenyl phosphate for 30 minutes. Color development was stopped by adding 0.05 ml of 1M sodium
hydroxide, and absorbance was measured at 405 nm using a
Dynatech model MR580 or a Bio-Rad (Richmond, CA)
model 2550 plate reader. Values were corrected for background binding of reagents to the coated wells.
The ability of MAb 6B6.6 to inhibit RF activity was
determined by a modification of the method described by
Carson and Fong (17). Wells were coated with 0.25 ml of
0.05 mg/ml human IgG in PBS for 1 hour, and remaining
protein-binding sites were blocked with BSA as described
above. Equal volumes of IgM-RF (50 ng/ml) and ascites fluid
containing the MAb (serially diluted) were mixed and incubated for 1 hour at room temperature. Total activity controls, prepared by substituting diluent for the antibody
dilution, were included in each assay. Duplicate 0.25-ml
aliquots of the mixtures and controls were then reacted with
the IgG-coated wells for 16 hours. Bound IgM-RF was
detected by reaction for 1 hour with 0.25 ml of alkaline
phosphataseconjugated affinity-purified F(ab‘), fragments
of goat anti-human IgM ( p chain-specific; Pel-Freez Biologicals, Rogers, AR) diluted 1 :5,000. Color was developed and
absorbances determined as described above for the directbinding assay.
The presence of RF-associated CRI in sera and
synovial fluids was determined by a direct-binding ELISA
specific for RF. Wells were coated with human IgG in PBS
and blocked with BSA as described for the =-inhibition
ELISA. Duplicate 0 . 2 5 4 samples of the sera or synovial
fluids (diluted 1: loo), COR IgM-RF standards (2-250 ng/ml),
6B6.6 CRI-positive and -negative control sera (also diluted
1:100), and diluent controls were reacted with the IgGcoated wells for 16 hours. CRI-positive RF on the wells was
then detected by reaction with 0.25 ml of 6B6.6 ascites fluid
(diluted 1 : 1,000-1 :5,OOO) for 4 hours, followed by alkaline
phosphatase-conjugated F(ab’), fragments of affinity-
purified goat anti-mouse IgG and disodium p-nitrophenyl
phosphate, as described for the direct-binding ELISA. U p
take of RF was verified by reaction of a duplicate plate with
alkaline phosphataseconjugated affinity-purified F(ab’),
fragments of goat anti-human IgM (1 :5,000) in place of the
anti-mouse IgG reagent. Later assays were performed using
0.2 ml of the coating solution, samples, standards, controls,
and reagents, with no loss of sensitivity.
MAb 6B6.6 CRI activity of serum IgM, IgA, and IgG
was examined using an Ig-capture ELISA. Microtiter wells
were coated with 0.2 ml of F(ab’), fragments of affinitypurified rabbit antibody specific for human IgM, IgA, or IgG
(Jackson ImmunoResearch), 0.01 mg/ml in PBS for 3 hours,
and the remaining protein-binding sites were blocked with
BSA as indicated for the other ELISA. Duplicate 0.2-ml
samples of sera, diluted 1 :100, were placed in the antibodycoated wells and incubated for 16 hours. CRI-positive Ig on
the wells was then detected by reaction with 0.2 ml of 6B6.6
ascites fluid (1:5,000) for 4 hours, followed by 0.2 ml of
alkaline phosphatasecoqjugated F(ab’), fragments of aBnity-purified anti-mouse IgG (1 :5,000) for 1 hour.
Wells were reacted with substrate, and absorbance
was measured as described for the direct-binding ELISA.
Results were corrected for background binding of the reagents as determined from duplicate control wells for each
sample, developed in the same manner except that BSA-PBS
diluent was substituted for the ascites fluid. In addition,
uptake of Ig was verified for each sample by reaction of a
second set of control wells with alkaline phosphatase
conjugated F(ab’), fragments of antibody specific for IgM,
IgA, or IgG (Jackson ImmunoResearch) diluted 1 :5,000,
1 :5,OOO, and 1:2,500, respectively.
Western immunoblot analysis (33). Protein electrophoresis was performed in 12% polyacrylamide gels prepared in 0.375M Tris HC1 buffer (pH 8.8), containing 0.1%
sodium dodecyl sulfate (SDS), using a Bio-Rad Mini-Protean
I1 dual slab cell, according to the general procedure of
Laemmli (34). Samples of the IgM (-0.25 mg/ml) were
reduced by heating at 95°C for 4 minutes in 0.063M Tris HCl
buffer (pH 6.8) containing 0.5M 2-mercaptoethanol, 2%
SDS, 10% glycerol, and 0.001% bromphenol blue, and 0.01
ml was applied to the running gel in a stacking gel that
contained 4% polyacrylamide in 0.05M Tris HCl buffer (pH
6.8). Electrophoresis was performed in 0.19M glycine,
0.025M Tris buffer, (PH 8.3), for 30 minutes at 175V.
Transfer to nitrocellulose membranes (Bio-Rad) was carried
out in 0.19M glycine, 0.025M Tris buffer (pH 8.3) containing
20% methanol, by the application of 125 mA for 30 minutes,
using a Bio-Rad Mini-Transblot electrophoretic transfer cell.
Nonspecific protein-binding sites were blocked by
incubation in O.15M NaCl buffered to pH 7.5 with 20 mM
Tris base (Tris buffered saline; TBS) containing 3% gelatin.
Membranes were then incubated with MAb 6B6.6 (100 ng/ml
or 15,000 ascites), control 1 g G l ~MAb (6A6.7) specilic for
an unrelated antigen (100 ng/ml), or a polyvalent affinitypurified rabbit anti-human Ig preparation (1 :5,000; Jackson
ImmunoResearch) in TBS containing 1% gelatin and 0.2%
Tween 20, for 16 hours. Membranes reacted with MAb were
then incubated with alkaline phosphatase-conjugated affinity-purified F(ab’), goat anti-mouse I& antibody (1 :5,000) in
the same buffer for 1 hour. Alkaline phosphataseconjugated
F(ab’), goat anti-rabbit IgG (1 :5,000; Jackson ImmunoResearch) was used for membranes that were reacted with the
rabbit anti-human Ig antibodies.
Reactions were visualized by addition of a mixture of
p-nitro blue tetrazolium chloride (0.3 mg/ml) and 5bromo-4-chloro-3-indolyl phosphate toluidine (0.15 mg/ml)
in 0.1M sodium carbonate, 1.0 mM magnesium chloride
buffer (pH 9.8) for 5-30 minutes as required. Membranes
were subjected to two or three 5-minute washes with 0.2%
Tween 20 in TBS between steps, and to two additional
5-minute washes with TBS alone prior to color development.
Color development was stopped by washing thoroughly with
Gel chromatography of sera. Sera were fractionated
by gel chromatography through Sephacryl S-300 as previously described (35). Column fractions were analyzed for
IgM-RF, IgM, IgA, and IgG by solid-phase ELISA (35,36).
IgA-RF was determined by the procedure used for IgM-RF,
modified by substitution of alkaline phosphatasexonjugated
F(ab’), fragments of affinity-purifiedgoat anti-human IgA (a
chain-specific; Pel-Freez) for the corresponding anti-human
IgM antibody reagent. MAb 6B6.6 CRI activity was measured by the RF-binding ELISA.
Amino acid sequence analysis. Amino acid sequence
analysis was performed by stepwise Edman degradation of the
sample in a Beckman model 890M liquid-phase sequenator
(Beckman Instruments, Fullerton, CA), using a doublecouple/double-cleavageprogram. The sequencing program was
always initiated from the acid delivery step, and the
first cycle so obtained was discarded. The phenylthiohydantoin
amino acid derivatives were identified by high pressure liquid
chromatography, as described by Bhown and Bennett (37).
Statistical analysis. Comparisons between means
were performed by Student’s t-test. Simple linear regression
was used to compare paired data. P values less than 0.05
were considered significant. Variations within groups are
expressed as standard deviations.
Reaction of MAb 6B6.6 with monoclonal RF
paraproteins. Monoclonal antibody 6B6.6 strongly
bound to microtiter wells coated with COR and LEW
IgM-RF (Figure 1). Essentially identical titration
curves were obtained with these 2 proteins. The
antibody also bound weakly to wells coated with POM
and MIL IgM-RF. In contrast, the antibody did not
show significant levels of binding to wells coated with
LAY IgM-RF or SCH IgA-RF. Similarly, no reaction
was o h e r v e d using microtiter wells coated with GLO
IgM-RF and 2 non-RF IgM paraproteins, DAU and
GRA (results not shown).
The results shown in Figure 1 were obtained
using cell-free ascites fluid containing the MAb. Isolated 6B6.6 antibody gave comparable results (Table
1). An isolated control MAb (6A6.7) of the same
isotype (IgG 1K ) exhibited negligible reactivity with
686.6 Ascites (1IDilution x lo-‘)
Figure 1. Binding of monoclonal antibody 6B6.6 with human monoclonal rheumatoid factors, as determined by enzyme-linked immunosorbent assay. Microtiter wells coated with COR IgM, LEW
IgM, POM IgM, MIL IgM, LAY IgM, or SCH IgA (all at 0.01
mg/ml) were reacted with 686.6 ascites diluted in phosphate buffered saline-bovine serum albumin as indicated on the abscissa and
developed with alkaline phosphatase-labeled F(ab’), fragments of
goat anti-mouse IgG antibody followed by substrate.
RF-coated wells. Thus, uptake of MAb 6B6.6 on wells
coated with COR, LEW, POM, and MIL was not due
to the R F activity of these proteins.
Reaction of both COR and LEW R F with MAb
6B6.6 in solution inhibited their interaction with IgGcoated microtiter wells in the inhibition assay (Figure
2). However, the reactions of POM and MIL R F with
the IgG-coated wells were not significantly affected
under the same conditions.
When microtiter wells coated with the component polypeptide chains of COR IgM-RF were used,
binding of MAb 6B6.6 was limited to the L chains
(Figure 3). Failure to bind to H chains was not due to
insufficient adsorption, since the H chain-coated wells
gave a strong reaction with alkaline phosphatase
labeled antibody specific for human p chain. Specificity of MAb 6B6.6 for structures in the L chains of
COR IgM-RF, as well as MIL and POM IgM-RF, was
verified by Western immunoblot analysis (results not
shown). Although LEW IgM-RF reacted only weakly
with the MAb by this technique, the activity that was
demonstrated was also confined to the L chains.
As indicated in Table 1, all four 6B6.6 CRIpositive IgM-RF possessed L chains of the KIIIa
sub-subgroup. Classification of the COR IgM K chains
was determined by amino-terminal sequence analysis
(see below). In addition, the K chain from this protein
failed to react with a mouse MAb (7C2.1) that strongly
reacts with d I I b chains, but not with KIIIa chains, by
Table 1. Reaction of human monoclonal immunoglobulins with
monoclonal antibody (MAb) 6B6.6 anti-COR idiotope and a control
MAb (6A6.7) of the same isotype, but specific for an unrelated
antigen, by ELISA*
Absorbance at 405 nm
L chaint
K m a
MAb 6B6.6
MAb 6A6.7
28 1
Western immunoblotting (Schrohenloher RE, Accavitti MA: unpublished observations). Using Western
immunoblotting with K chain subgroupspecific rabbit
antipeptide antibodies (38), the L chains of LEW,
MIL, and DAU IgM have been identified as belonging
to the KIII subclass (Silverman GJ, Carson DA: personal communication). Assignment to the KIIIa subsubgroup was based on weak or negative reactions
with MAb 7C2.1 by Western immunoblotting (Schrohenloher RE, Accavitti MA: unpublished observations). Classification of the L chains of these proteins
as KIIIa was also consistent with the results of amino
acid sequence analysis of the amino-terminal portions
of the chains (see below).
The amino acid sequence of the amino-terminal
regions of COR, LEW, and MIL IgM-RF K chains, as
determined by automated amino acid sequence analysis, and the first 30 residues of POM IgM K chain (39)
686.6 Ascites( IlDilution x
* Microtiter wells coated with the IgM or IgA proteins were reacted
with 1Wnglml dilutions of the respective MAb for 3 hours at room
temperature, and bound antibody was subsequently detected by
reaction with alkaline phosphatase-labeled goat F(ab‘), anti-mouse
antibody. See Patients and Methods for details. ELISA = enzymelinked immunosorbent assay; R F = rheumatoid factor.
t Assignment of COR IgM L chains to the KIIIa sub-subgroup was
based on amino-terminal sequence analysis (see text) and a negative
reaction with a mouse MAb (7C2.I) specific for KIIIb chains by
Western immunoblotting (Schrohenloher RE, Accavitti MA: unpublished observations). LEW, MIL, and DAU IgM L chains were
shown to belong to the KIII subclass by Western immunoblotting
with L chain subgroupspecific rabbit antipeptide antibodies (ref. 38
and Silverman GJ, Carson DA: personal communication). Further
characterization as d I I a sub-subgroup was determined by aminoterminal sequence analysis (see text) and weak or negative reaction
with MAb 7C2.1. POM, LAY, and GLO IgM L chains were
classified by amino acid sequence and antigenic analysis (42). SCH
IgA L chains were assigned to the KIIIb sub-subgroup based on the
presence of the KIIIb-associated 17.109 cross-reactive idiotype (21)
and positive reaction with MAb 7C2.1.
$ DAU is a monoclonal IgM paraprotein of unknown specificity,
obtained from a patient with Waldenstrom’s macroglobulinemia.
Figure 2. Inhibition of rheumatoid factor (RF) binding to IgGcoated microtiter wells by monoclonal antibody 6B6.6. Microtiter
wells coated with human IgG (0.05 mglml) were reacted with
preincubated mixtures containing equal volumes of R F (SO ng/ml)
and 6B6.6 ascites diluted in phosphate buffered salinebovine serum
albumin as indicated on the abscissa and developed with alkaline
phosphatase-labeled F(ab’), fragments of rabbit anti-human IgM
antibody followed by substrate.
are shown in Figure 4. While the four 6B6.6 CRIpositive K chains exhibited a high degree of homology
among the residues identified, differences were found
in framework residues at positions 9, 15, and 17 and in
a complementarity-determining region residue at position 29. A high degree of homology (95% compared
with COR K chain) was also evident for the first 20
686.6 Ascites (IIDilution x W 3 )
Figure 3. Reaction of monoclonal antibody 6B6.6 with the heavy
(H) and light (L) polypeptide chains of COR IgM, as determined by
enzyme-linked immunosorbent assay. Microtiter wells coated with
isolated COR H and L chains (0.01 mglml) were reacted with 6B6.6
ascites diluted in phosphate buffered salinebovine serum albumin
as indicated on the abscissa and developed with alkaline phosphatase-labeled F(ab’), fragments of goat anti-mouse IgG antibody
followed by substrate.
L Chain
E I V M T Q S P A T L S V S P C K R A T L S ( C ) R A S Q S V S
_ ~ _ _ _ _ _ _ _ _ _ _ _ L_
- - - - - - - - - - - - - - - - E - - - - - ( - ) ? - - - - - -
- ~ - - - - - - V - - - - - - - E - - - - -
- - - - - - - - - - - - - - - - E - - . .
- E
- - - - - I -
Figure 4. Light (L) chain amino-terminal amino acid sequences of monoclonal antibody 6B6.6
cross-reactive idiotype (CRI)-positive IgMK rheumatoid factors (RF) (COR, LEW, MIL, and POM)
and a 6B6.6 CRI-negative IgMK of unknown (Unk.) antibody activity (DAU). Data for the POM
IgM-RF L chain are from ref. 39. Sequences of the other L chains were determined by automated
stepwise Edman degradation (see Patients and Methods). Amino acid residues are designated by the
I-letter code; residues identical to the corresponding residue in the COR L chain are indicated by
hyphens. The question mark in the MIL L chain indicates a residue that was not identified. Residues
shown in parentheses at position 23 of the COR and MIL L chains are presumed to be cysteine by
analogy to other human K chains (40); cysteinyl residues could not be identified by the method used.
amino-terminal residues of the K chain from DAU
IgM, a 6B6.6 CRI-negative protein of unknown antibody specificity (Figure 4). Nonetheless, comparison
of the available sequences with the prototype chains of
the 4 K chain subgroups supports classification of each
as a member of the KIII subgroup (40). The presence of
Ala at position 9, in place of the glycyl residue
characteristic of the d I I b sub-subgroup (41), further
indicates that COR, LEW, MIL, and DAU belong to
the KIIIa sub-subgroup.
Occurrence of 6B6.6 CRI on RF from RA patients. Sera from 107 patients with RA, selected for the
presence of IgM-RF by ELISA, were examined by the
RF-binding assay for the occurrence of R F that was
cross-reactive with MAb 6B6.6. The CRI was detected
in 63 (5%) of the sera, at levels that ranged from a
minimally detectable amount (equivalent to 0.2 pg/ml)
to 6 pg/ml, as determined by comparison with a
reference curve prepared- using COR IgM-RF. As
illustrated in Figure 5, more than half of the reactive
sera (i.e., 35) contained <0.5 pg/ml of the CRI. The
quantity of CRI did not correlate with serum IgM-RF
levels (r = 0.039, P = 0.69). Similarly, although the
levels of IgM-RF in sera without detectable 6B6.6 CRI
(294 ? 608 pg/ml, mean ? SD) were somewhat less
than in those t h a t demonstrated the CRI (456 552
pg/d), the difference was not, statistically significant
(' = 0'138)*The ranges Of lgM-RF levels were
parable in the 2 groups (2-3,721 Pg/ml in the CRInegative group and 6 2 , 7 4 2 pg/ml in the CRI-positive
group). The quantity of 6B6.6 CRI in the CRI-positive
group represented 0.01-6.5% (mean 2 SD 0.6 ? 1.0%)
of the serum IgM-RF.
Mouse MAb of the Same k G Subclass, but
specific for an unrelated antigen (MAb 6A6.7), did not
react with a panel of the 6B6.6 CRI-positive sera (n =
16) under the same conditions. Not unexpectedly,
normal sera (n = 30) were negative for 6B6.6 CRI by
the RF-binding ELISA.
Fifty RA patient synovial fluid samples that
contained IgM-RF as shown by ELISA were also
examined for the presence of 6B6.6 cross-reactive RF.
5 1
R A Sera
RA Synovial Fluids
Figure 5. Levels of rheumatoid factor (RF)-associated monoclonal
antibody 6B6.6 cross-reactive idiotype (CRI) in sera and synovial
fluids from patients with rheumatoid arthritis (RA). Microtiter wells
coated with human IgG (0.05 mg/ml) were reacted with the serum or
synovial fluid diluted 1 : 100 in phosphate buffered salinebovine
serum albumin (PBS-BSA). After washing, the wells were reacted
with 6B6.6 ascites (diluted 1 :5,000 in PBS-BSA) and developed with
alkaline phosphatase-labeled F(ab'), fragments of goat anti-mouse
I& antibody followed by substrate. Quantities of CRI were estimated from reference curves prepared from isolated COR IgM-RF
included in each run. Sera from 107 patients and synovial fluids from
50 patients were tested; only the results obtained in the 63 sera and
24 synovial fluids with detectable amounts of 6B6.6 CRI (20.2
&ml) are shown.
The CRI was found in 24 (48%) of the fluids, at levels
that ranged from 0.3 pg/ml to 13 pglml (Figure 5 ) . As
was found with the RA sera, the quantity of CRI did
not correlate with the amount of IgM-RF present in the
fluids (r = -0.092, P = 0.650). Also. as observed with
the KA sera, the level of IgM-KF in synovial fluids
without detectable 6B6.6 CRI (236 f 358 gglml) was
somewhat less than that in samples demonstrating the
CRI (309 t 522 CLg/ml), but the difference was not
statistically significant ( P = 0.559). The 2 groups
demonstrated similar ranges of IgM-RF values (21,707 pg/ml and 9-2,236 pg/ml for the CRI-negative
and CRI-positive synovial fluids, respectively). When
CRI was present, it accounted for 0.02-1 1% (mean t
SD 2.3 t 2.9%) of the IgM-RF in the synovial fluids.
Isotype distribution of 6B6.6 CRI in RA sera.
The isotype distribution of RF reactive with MAb
686.6 was examined by gel chromatography, in 2
CRI-positive RA sera that contained both IgM-RF and
IgA-RF (Figure 6). Essentially identical results were
obtained with both sera. Elution profiles for 686.6 CRI
corresponded to that of the IgM-RF. but not IgA-RF.
Much of the latter was recovered in lower molecular
weight fractions, intermediate between IgM-RF and
monomeric IgA.
The distribution of the 6B6.6 CRI among the
serum immunoglobulins in R A was also examined by
ELISA. using microtiter wells coated with F(ab’),
fragments of rabbit anti-human IgM, IgA. and IgG
antibodies. Results obtained with I5 seropositive RA
sera are shown in Figure 7. Reaction of MAb 686.6
with the sera in the assay for IgM-associated CRI was
variable and correlated strongly with that observed in
the RF-binding assay (r = 0.966, P < 0.001). Control
MAb 6A6.7 failed to react with the R A sera in the
ELISA for IgM-associated CRI. further indicating the
specificity of the reaction of MAb 6B6.6 in this assay.
I n contrast to the results obtained for IgMassociated CRI. MAb 6B6.6 reacted weakly or not at
all with the RA sera in the assays for IgA- and
IgG-associated CRI. except in sample 3, which gave a
moderately strong reaction for IgG-associated CRI.
However, it is likely that the latter activity was due to
the uptake of complexes containing CRI-positive IgMRF by the antibody-coated wells or reaction of CRIpositive IgM-RF with the antibody-coated wells sccondary to the binding of IgG. since the IgG fraction
recovered from this senim by gel chromatography in
acidic buffer did not react with MAb 6B6.6 in the assay
for IgG-associated CRI. The corresponding IgM frac-
mlgA IgG
v v
Fraction Number
Figure 6. Distribution of IgM rheumatoid factor (IgM RF). IgA RF,
and RF-associated monoclonal antibody 6B6.6 cross-reactive idiotype (CKI) in sera from 2 rheumatoid arthritis patients (A and B).
following gel chromatography through Sephacryl S-300 in sodium
acetate-NaCI buffer. at pH 3.8 (35). Serum samples (0.2 ml) diluted
in 0.8 ml of the pH 3.8 buffer were applied to a 1.5 x 9s-cm column
of the gel. Fractions (2 ml per tube) were neutralized and analyzed
at I : 100 for IgM RF. I : 20 for IgA RF. and 1 :5 for RF-associated
6B6.6 CRI. by enzyme-linked immunosorbent assay. Fraction tubes
containing peak concentrations of monomeric IgA (rnIgA) and I&
are also indicated.
tion of the serum was, however, strongly positive in
the assay for IgM-associated CRI.
Immunoglobulin uptake control studies for each
serum demorlstrated that large amounts of IgM, IgA,
or IgG (i.e., optical density >2.0) were bound by the
antibody-coated wells in the respective assays. Thus,
weak or negative reactions did not appear to be a
function of the Ig-binding capacity of the antibodycoated wells or the Ig content of individual sera.
Reaction of MAb 686.6 with normal serum
immunoglobulins. The occurrence of the 686.6 CRI on
serum immunoglobulins from 15 healthy controls was
RA Sera
Control Sera
1.0 r
1 0 1 1 1 2 1 3 1 4 1 5
Serum Number
Figure 7. Occurrence of IgM-, IgA-, and IgG-associated monoclonal antibody 686.6
cross-reactive idiotype in sera from I5 patients with rheumatoid arthritis (RA)and 15
healthy controls. Microtiter wells coated with F(ab'), fragments of rabbit anti-lgM,
anti-lgA. and anti-lgG antibodies were reacted sequentially with the sera diluted 1 : 100
and 6B6.6 ascites diluted I : S.Oo0 in phosphate buffered s a l i n e h v i n e serum albumin and
developed with alkaline phosphatase-labeled F(ab'), fragments of goat anti-mouse IgG
antibody followed by substrate. Results were corrected for background binding of the
enzyme-labeled anti-mouse 1gG reagent by the individual sera in each assay. as
determined using antibody-coated wells that were reacted with the diluted sera but not
the ascites fluid.
also determined using the ELISA procedure for IgM-,
I@-, and IgG-associated CRI (Figure 7). Only low
levels of IgM-, IgA-, or IgG-associated 6B6.6 CRI
were observed in occasional sera. IgM-RF was not
detected by ELISA in the serum that gave a weak
reaction of IgM-associated 6B6.6 CRI (sample 4). As
noted for the RA sera, Ig uptake controls confirmed
that large amounts of Ig were bound to the antibodycoated wells in all 3 assays for each serum.
Monoclonal antibody 686.6 was shown to react
with an idiotope present on the K-type immunoglobulin
L chains of several monoclonal IgM-RF proteins from
unrelated individuals. In addition, the idiotope was
detected, although at lower levels, on polyclonal RF in
approximately half of the sera and synovial fluids from
patients with seropositive RA. The idiotope was found
prim&ly in the IgM fraction of RA sera. In contrast, it
was absent or present in only low levels in serum IgA
and IgG from RA patients and serum IgM, IgA, and
IgG from healthy controls. Inhibition of the IgGbinding activity of COR and LEW IgM-RF by the
MAb suggests that the CRI is located at or near the
RF-active site of these proteins. Failure to inhibit the
IgG-binding activity of MIL and POM IgM-RF may
reflect the partial expression of the CRI in these
paraproteins, evident from the lower levels of reaction
with the MAb.
The CRI identified by MAb 686.6 does not
appear to be related to several previously described
RF CRI. All 4 of the 6B6.6 CRI-positive monoclonal
RF in the panel of proteins used to characterize this
determinant possessed KIII L chains of the KIIIa
sub-subgroup. RF bearing the Wa CRI. the major RF
idiotypic group described by Kunkel et al (14), characteristically have KIII L chains carrying the KIIIb
sub-subgroup determinant (42). It is therefore unlikely
that 6B6.6 CRI is related to the Wa group. Furthermore, 6B6.6 CRI was not detected on GLO IgM-RF, a
Wa CRI-positive protein in the panel (42). Similarly,
the 6B6.6 CRI does not appear to be related to t h e
dIb-associated KF CRI identified by MAb 17.109
(prepared by Carson and Fong [17]). Monoclonal
antibody 6B6.6 did not react with two 17.109 CRIpositive monoclonal RF included in the panel (GLO
and SCH) (17,21), further demonstrating that the
6B6.6 CRI is distinct from the latter. In addition, all of
the remaining IgM used in this study have been
examined for reaction with MAb 17.109 (kindly provided by Dr. D. A. Carson). COR, LEW, POM, MIL,
LAY, and DAU IgM, when coated on polystyrene
micropkite wells, failed to bind the hybridoma antibody. In contrast, SCH IgA and GLO IgM, which
have been previously shown to react with MAb
17.109, ,strongly bound the antibody under the same
The panel of human monoclonal proteins used
to characterize MAb 6B6.6 also included POM and
LAY IgM-RF, 2 of the original members of the less
frequently observed Po CRI group (also described by
Kunkel et al [14]). The finding that MAb 6B6.6 reacted
with only 1 of these 2 RF indicates that the 6B6.6 CRI
is not related to the Po CRI group. It should be further
noted that 3 of the 686.6 CRI-positive RF (COR,
POM. and MIL) have been tested and found not to
cross-react with nuclear antigens by immunofluorescence using mouse kidney substrate (unpublished
data). It is therefore unlikely that the RF CRI detected
by MAb 6B6.6 is related to the Bla group reported by
Agnello et al (16). since the latter is associated with a
subset of RF that cross-reacts with DNA nucleoprotein. Failure to react with DAU IgM indicates that the
antibody does not bind to all immunoglobulins with
KIIIa L chains, but may be selective for a subset
associated with RF activity.
It is clear that MAb 6B6.6 and MAb 17.109
identify different RF CRI associated with K-type L
chains. The relationship of these CRI has been further
characterized (43). Examination of a panel of 163
human monoclonal IgM by Western immunoblotting
demonstrated that both 686.6 and 17.109 CRI occurred with greatest frequency among the 26 monoclonal IgM-RF in the panel and that each was found
exclusively on K chains. The 2 CRI were mutually
exclusive, not being found on the same protein in any
instance. Each occurred with the same frequency
(-30%) among the monoclonal RF with K chains. The
17.109 CRI was found only on K chains expressing the
KIIIb sub-subgroup determinant. In contrast, none of
the 6B6.6 CRI-positive proteins possessed KIIIb
Although it was detected in a signhcant proportion of RA sera and synovial fluids, the quantity of
6B6.6 CRI-positive Ig, as estimated by comparison
with standard curves prepared from COR IgM-RF,
was very small and represented only a small part of the
IgM-RF (<1% in most sera). Furthermore, the amount
of CRI was not related to the amount of IgM-RF
present in the sera or synovial fluids. Interpretation of
the results obtained with polyclonal R F is complicated
by the inability to distinguish between partial and
complete cross-reactivity of the MAb with the RF.
This limitation notwithstanding, the results indicate
the occurrence of the same or similar idiotypic determinants in a portion of the R F molecules in RA and
support the concept that monoclonal and polyclonal
RF are derived, at least in part, from a common gene
pool. The fractional contribution of the gene encoding
686.6 CRIt-positive KIII chains to this pool is, of
course, unknown. Recently reported evidence for extensive polymorphism in a human immunoglobulin
gene locus (44) raises the possibility that the KIII gene
pool may be large. The apparent selective association
of 686.6 CRI with IgM-RF in RA sera that contain
both IgM-RF and IgA-RF suggests that precursors of
the latter may utilize a different gene pool. Thus,
differential expression of this idiotope may be a function of regulation. Alternately, absence of the 6B6.6
CRI among IgA-RF (and, presumably, IgG-RF) may
reflect a marked susceptibility to somatic mutation in
the R F gene encoding this determinant.
A germline gene (Humkv328) that appears to
encode the variable region of POM K chain and possibly other 686.6 CRI-positive KIIIa chains has been
cloned and sequenced (45). The sequence of the 30
amino-terminal residues of COR K chain differs from
that encoded b y Humkv328 by only 1 residue (Lys for
Glu at position 17). Similarly, the amino-terminal 17
residues of LEW K chain differ from the Humkv328
amino acid sequence only by the presence of Leu
instead of Pro at position 15, while the first 30 residues
of MIL K chain are identical to the Humkv328 amino
acid sequence, with the possible exception of the
unidentified residue at position 24. Comparison of the
first 30 residues of POM K chain with Humkv328
indicates substitutions at position 9 (Val for Ala) and
position 29 (Ile for Val).
The significance of the differences among the
6B6.6 CRI-positive K chains is unknown, but they may
represent polymorphism among genes encoding for
d I I a chains o r mutations occurring in the course of
differentiation of the respective B cell clones. It should
also be noted that the amino-terminal sequences of
DAU K chain and Humkv328 are identical, illustrating
that a high degree of homology can occur among first
framework region residues of KIIIa chains, irrespective of association with R F activity and the 6B6.6 CRI.
The minimal reaction of MAb 6B6.6 with serum
IgM, IgA, and IgG from normal subjects is consistent
with our previous observations, using a different assay, indicating that only a small fraction (-0.1%) of
normal IgM expressed this CRI, even though it was
detected in more than half the sera tested (43). Because of uncertainty regarding the significance of low
ELISA absorbance values, no attempt was made in
the present study to estimate the frequency of detectable 6B6.6 CRI in the normal immunoglobulins. The
virtual absence of the 6B6.6 CRI among the IgM, as
well as the other immunoglobulins in normal sera,
indicates preferential utilization by R F of L chains
carrying this determinant. Nonetheless, the apparent
Occurrence in occasional sera of the RF-associated CRI
in the absence of RF activity suggests that 686.6 CRIpositive L chains are not restricted to RF, but may also
be utilized by a limited number of immunoglobulins
having other specificities. The occurrence of RFassociated CRI among non-RF paraproteins has been
observed by Chen et al (46) and Crowley et al(43).
In summary, we have prepared a mouse MAb
(6B6.6) specific for a prominent CRI present on the L
chains of several human monoclonal IgMK R F paraproteins. The CRI was found only on monoclonal RF
with L chains belonging to the KIIIa sub-subgroup, and
it appeared to be associated with o r located near the
RF active site. In addition, the 6B6.6 CRI was detected on the polyclonal R F in roughly 50% of RA sera
and synovial fluids from seropositive patients; however, the quantities of polyclonal RF-associated CRI
detected in individual sera or synovial fluids represented only a small fraction of the R F present. Examination of the distribution of 6B6.6 CRI among the
immunoglobulins in RA sera indicated that it was
present primarily on IgM-RF, although small quantities of IgA- and IgG-associated CRI were indicated in
some sera. In contrast, only minimal quantities of
6B6.6 CRI were detected on normal serum IgM, IgA,
and IgG. Our findings suggest that the number of
germline genes encoding for R F in patients with rheumatoid arthritis is large, or that extensive mutation
occurs in the course of R F expression in this disease.
We thank Fran AUen. Rebccca L. Bowman, Mary E.
Kallman, Merry Y. Liu, Vincent J. Mug, James I.. Wayland, David H. Williams. and Joseph T. Wooten for superb
technical assistance, and Drs. Dennis A. Carson and Gregg
J. Silverman for K chain subgroup analysis and for generously providing GLO plasma and MAb 17.109. We are also
grateful to Dr. Fiona Hunter for the control MAb (6B6.71,
Dr. Henry Metzger for the LAY plasma, and Dr. Alan
Solomon for the SCH plasma used in this study.
I . Koopman WJ, Schrohenloher RE: Rheumatoid factor,
Rheumatoid Arthritis. Edited by PD Utsinger, NJ Zvaifler, GE Ehrlich. Philadelphia, JB Lippincott, 1985
2. Heimer R. Levin FM: On the distribution of rheumatoid
factors among the immunoglobulins. Immunochemistry
3:l-10, 1966
3. Carson DA, Lawrance S, Catalan0 MA, Vaughan JH,
Abraham G: Radioimmunoassay of IgG and IgM rheumatoid factors reacting with human IgG. J Immunol
119:295-300, 1977
4. Dunne JV,Carson DA, Spiegelberg HL, Alspaugh MA,
Vaughan JH: IgA rheumatoid factor in the sera and
MAb 6B6.6 CRI ON R F
saliva of patients with rheumatoid arthritis and Sjogren’s
syndrome. Ann Rheum Dis 38: 161-165, 1979
5 . Elkon KB, Delacroix DL, Gharavi AE, Vaerman JP,
Hughes GRV: Immunoglobulin A and polymeric IgA
rheumatoid factors in systemic sicca syndrome: partial
characterization. J Immunol 129576581, 1982
6. Koopman WJ, Schrohenloher RE, Solomon A: A quantitative assay for IgA rheumatoid factor. J Immunol
Methods 50:89-98, 1982
7. Wernick R, LoSpalluto JJ, Fink CW, Ziff M: Serum IgG
and IgM rheumatoid factors by solid phase radioimmunoassay: a comparison between adult and juvenile rheumatoid arthritis. Arthritis Rheum 24:1501-1511, 1981
8. Czerkinsky C, Koopman WJ, Jackson S, Collins JE,
Crago SS, Schrohenloher RE, Julian BA, Galla JH,
Mestecky J: Circulating immune complexes and immunoglobulin A rheumatoid factor in patients with mesangial immunoglobulin A nephropathies. J Clin Invest
77: 1931-1938, 1986
9. Sinico RA, Fornasieri A, Oreni N , Benuzzi S, d’Amico
G: Polymeric IgA rheumatoid factor in idiopathic IgA
mesangial nephropathy (Berger’s disease). J Immunol
137536541, 1986
10. Natvig JB, Gaarder PI, Turner MW: IgG antigens of the
C y 2 and Cy3 homology regions interacting with rheumatoid factors. Clin Exp Immunol 12:177-184, 1972
1 1 . Butler VP Jr, Vaughan JH: Hemagglutination by rheumatoid factor of cells coated with animal gamma globulins. Proc SOCExp Biol Med 116585-593, 1964
12. Hannestad K, Johannessen A: Polyclonal human antibodies to IgG (rheumatoid factors) which cross-react
with cell nuclei. Scand J Immunol 5541-547, 1976
13. Aitcheson CT, Peebles C, J o s h F, Tan EM: Characteristics of antinuclear antibodies in rheumatoid arthritis:
reactivity of rheumatoid factor with a histone-dependent
nuclear antigen. Arthritis Rheum 23528-538, 1980
14. Kunkel HG, Agnello V, J o s h FG, Winchester RJ,
Capra JD: Cross-idiotypic specificity among monoclonal
IgM proteins with anti-yglobulin activity. J Exp Med
137:331-342, 1973
15. Fdrre 8 , Dobloug JH, Michaelsen TE, Natvig JB:
Evidence of similar idiotypic determinants on different
rheumatoid factor populations. Scand J Immunol 9:
281-289, 1979
16. Agnello V, Arbetter A, Ibanez de Kasep G, Powell R,
Tan EM, J o s h F: Evidence for a subset of rheumatoid
factors that cross-react with DNA-histone and have a
distinct cross-idiotype. J Exp Med 151:1514-1527, 1980
17. Carson DA, Fong S: A common idiotope on human
rheumatoid factors identified by a hybridoma antibody.
Mol Immunol 20:1081-1087, 1983
18. Pasquali J-L, Storck D: Monoclonal antibodies defining
a minor and a private idiotope on human rheumatoid
factors. J Clin Immunol6:349-354, 1986
19. Posnett DN, Wisniewolski R, Pernis B, Kunkel HG:
Dissection of the human antigammaglobulin idiotype
system with monoclonal antibodies. Scand J Immunol
23:16%181, 1986
Fong S, Chen PP, Gilbertson TA, Weber JR, Fox RI,
Carson DA: Expression of three cross-reactive idiotypes
on rheumatoid factor autoantibodies from patients with
autoimmune diseases and seropositive adults. J Immuno1 137:122-128, 1986
Fox RI, Chen P, Carson DA, Fong S: Expression of a
cross-reactive idiotype on rheumatoid factor in patients
with Sjogren’s syndrome. J Immunol 136:477483, 1986
Schrohenloher RE, Koopman WJ: Expression of crossreactive idiotypes in a panel of monoclonal rheumatoid
factors (abstract). Clin Res 33:793A, 1985
Schrohenloher RE, Koopman WJ: An idiotope common
to rheumatoid factors from patients with rheumatoid
arthritis identified by a monoclonal antibody (abstract).
Arthritis Rheum 29 (suppl 4):S28, 1986
Koopman WJ, Schrohenloher RE, Barton JC, Greenleaf
EC: Suppression of in vitro monoclonal human rheumatoid factor synthesis by antiidiotypic antibody: target
cells and molecular requirements. J Clin Invest 72:
1410-1419, 1983
Schrohenloher RE, Koopman WJ, Moldoveanu Z, Solomon A: Activity of rheumatoid factors of different
molecular sizes: comparison of autologous monomeric
and polymeric monoclonal IgA rheumatoid factors. J
Immunol 134:1469-1474, 1985
Ropes MW, Bennett GA, Cobb S, Jacox R, Jessar RA:
1958 revision of diagnostic criteria for rheumatoid arthritis. Bull Rheum Dis 9:175-176, 1958
Fleischman JB, Pain RH, Porter RR: Reduction of
yglobulins. Arch Biochem Biophys (suppl 1):174-180,
Porath J, Kristiansen T: Biospecific affinity chromatography and related methods, The Proteins. Third edition.
Vol. I. Edited by H Neurath, RL Hill. New York,
Academic Press, 1975
Goiii F, Chen PP, Pons-Estel B, Carson DA, Frangione
B: Sequence similarities and cross-idiotypic specificity
of L chains among human monoclonal IgMK with antiyglobulin activity. J Immunol 135:40734079, 1985
Lieberman R, Potter M, Humphrey W Jr, Mushinski
EB, Vrana M: Multiple individual and cross-specific
idiotypes on 13 levan-binding myeloma proteins of
BALB/c mice. J Exp Med 142:106-1 19, 1975
Kearney JF, Radbruch A, Liesegang B, Rajewsky K: A
new mouse myeloma cell line that has lost immunoglobulin expression but permits the construction of antibodysecreting hybrid cell lines. J Immunol 123: 1548-1550,
Keainey JE: Hybridomas and monoclonal antibodies,
Fundamental Immunology. Edited by WE Paul. New
York, Raven Press, 1984
Towbin H, Staehelin T, Gordon J: Electrophoretic
transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc
Natl Acad Sci USA 7k4350-4354, 1979
Laemmli UK: Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature 227:
680-685, 1970
Schrohenloher RE, Koopman WJ, Alarcon GS: Molecular forms of IgA rheumatoid factor in serum and
synovial fluid of patients with rheumatoid arthritis.
Arthritis Rheum 29:1194-1202, 1986
Tomana M, Schrohenloher RE, Koopman WJ, Alarc6n
GS, Paul WA: Abnormal glycosylation of serum IgG
from patients with chronic inflammatory diseases. Arthritis Rheum 31:333-338, 1988
Bhown AS, Bennett JC: A modified system for thiazolinone conversion to thiohydantoin derivatives and their
separation by high pressure liquid chromatography.
Anal Biochem 150:457462, 1985
Silverman GJ, Carson DA, Solomon A, Fong S: Human
kappa light chain subgroup analysis with synthetic peptide-induced antisera. J Immunol Methods 95:24%257,
Klapper DG, Capra JD: The amino acid sequence of the
variable regions of the light chains from two idiotypically cross reactive IgM anti-gamma globulins. Ann
Immunol (Paris) 127:261-271, 1976
Kabat EA, Wu ‘IT, Reid-Miller M, Perry HM, Gottesman KS: Sequences of Proteins of Immunological Inter-
est. Fourth edition. Washington, DC, US Dept. of
Health and Human Services, 1987
41. Ledford DK, Goiii F, Pizzolato M, Franklin EC, Solomon A, Frangione B: Preferential association of KIIIb
light chains with monoclonal human IgMK autoantibodies. J Immunol 131:1322-1325, 1983
42. Kunkel HG, Winchester RJ, J o s h FG, Capra JD:
Similarities in the light chains of anti-yglobulins showing cross-idiotypic specificities. J Exp Med 139:12g.136,
43. Crowley JJ, Goldfien RD, Schrohenloher RE, Spiegelberg HL, Silverman GJ, Mageed RA, Jefferis R,
Koopman WJ, Carson DA, Fong S: Incidence of three
cross-reactive idiotypes on human rheumatoid factor
paraproteins. J Immunol 140:3411-3418, 1988
44. Van Dijk KW, Schroeder HW Jr, Perlmutter RM, Milner ECB: Heterogeneity in the human Ig V, locus. J
Immunol 142:2547-2554, 1989
45. Chen PP, Robbins DL, Jirik FR, Kipps TJ, Carson DA:
Isolation and characterization of a light chain variable
region gene for human rheumatoid factors. J Exp Med
166:1900-1905, 1987
46. Chen PP, Got% F, Fong S, Jirik F, Vaughan JH, Frangione B, Carson DA: The majority of human monoclonal
IgM rheumatoid factors express a “primary structuredependent” cross-reactive idiotype. J Immunol 134:
3281-3285, 1985
Без категории
Размер файла
1 075 Кб
kappa, chains, light, polyclonal, cross, antibody, reactive, human, defined, 6b6, factors, monoclonal, idiotope, rheumatoid
Пожаловаться на содержимое документа