documents 1997 - rheumatoid factor autoantibodies - mageed

8/12/2019 Documents 1997 - Rheumatoid Factor Autoantibodies - Mageed

http://slidepdf.com/reader/full/documents-1997-rheumatoid-factor-autoantibodies-mageed 1/16

Rheumatoid Factor Autoantibodies

in Health and Diseasea

RIZGAR A. MAGEED,b MARIE B@RRETZEN,cSASHA P. MOYES,b KEITH M. THOMPSON,C

AND JACOB B. NATVIGC

bKennedy Institute of RheumatologyHammersmith, London W6 7c x United Kingdom

=Instituteof Immunology and RheumatologyThe National Hospital0172 Oslo, Norway

INTRODUCTION

Since their discovery, a vast amount of work has been done on the incidence,

specificity, and genetics of human rheumatoid factors (RF) in a great many laborato-

ries. In the main, these studies were inspired by the association of RFwith RA and the

hope that understanding mechanisms of their production may help define disease

pathogenesis. However, studies of RF production and incidence have shown that RF

are not restricted to patients with RA and that genes encoding RF are inherited in the

normal human population. Furthermore, the docum entation that RF specificity can bepart of the natural antibody reperto ire during early fetal life and that norm al adults can

produce RF following imm unization are important indications for their physiological

role in normals and in understanding mechanisms of B cell regulation in health and

disease.

Th e initial attempts to define RF structure and genetics benefited from serological

analysis with anti-idiotypic reagents and amino acid sequence analysis of RFparapro-

teins, or “M” components, from patients with mixed cryoglobulinemia, Walden-

strom’s macroglobulinemia, and chronic lymphocytic leukemia. More recently,

nucleotide sequence analysis of RF genes isolated from immortalized human B cell

hybridomas from normals and RA patients in our laboratories, and others, provided

significant insight into how RF are produced and likely to be r e g ~ l a t e d . ~ ~ere, we

show that RF produced in normal individuals are different from those found in the

synovium of RApatients. RF in normals re frequently encoded by a limited set of IgV

genes, do not appear to switch isotype, and have few replacement mutations in the

CDRs. These observations support the concept that RF are under strict control to

preven t the emergence of high-affinity RF in normals.

asupport for this work was provided by the Arthritis and Rheumatism Council of GreatBritain (to R. A. Mageed and S. P. Moyes) an d by the Norwegian Women’s Health Organization

(NKS) and the Norwegian Research Council (to M. Berretzen, K. M. Thompson, and J. B.

Natvig).

296



298 ANNALS NEW YORK ACADEMY OF SCIENCES

bind com plement, perpetuating the inflammatory processes and con tributing to tissue

damage.I6 The production of IgG RF is particularly important due to the ability of

IgG RF to self-associate and form large aggregates with high complement-fixing

potential.

The involvement of T cells in the synovial inflamm ation and the interaction with B

cells have been the focus of numerous studies.T cell involvement in RA is suggested

by the presence ofT cell infiltrates in the synovium and by the genetic association with

HLA -DRl and DR4 haplotypes.17J8 Synovial T cell infiltrates seem to be activated,

but lack specificity for any particular a~ to a n ti g en .' ~nly specificity for m ycobacterial

antigens has clearly been demonstrated.20 Furthermore, investigation of possible

clonal dominance w ithin T cell populations in rheumatoid synov ium has shown little

evidence for interclonal similarity or dom inance of T cell populations, suggesting that

the synovialT cell population may have been selected by their activation state rather

than by specificity.21

STRUCTURE-FUNCTIONALASPECTS OF HUMAN RF

The term RF is common for antibodies with specificities for determinan ts within

the Fc region of IgG. Antibod ies with these specificities are de tectable in a variety of

autoimmune and infectious diseases at varying levels and freq~encies?~-~~any

different specificities have been described for different RF. In general, IgG determi-nants recogn ized by RF can be categorized into the following groups: (1) subclass-

specific antigens, or isotypic antigens, found on all or som e of the four subclasses of

human IgG; (2) genetically defined alloantigens of the Gm type; (3) cross-reactive

antigens shared by human and animal IgG; (4) species-specific antigens found in

human IgG, but not animal IgG; 5 ) neoantigens expressed on aggregated, denatured,

or enzymatically digested IgG; and (6) determinants expressed on the Fab region of

IgG (antibodies with specificity for these determinants are often referred to as

antiglobulins, not RF). Extensive studies have shown that reactivity with an isotypic

antigen expressed on I gG l, 2, and 4, but not IgG3 (the Ga determinant) is the mainspecificity known for RF.25-28Despite some minor differences, the overall conclusion

has been that most FW,ncluding RF established from normal individuals, bind sites

related to the Ga deter1ninant.2~This site appears to be related to or overlaps with the

site of staphylococcal protein A (SPA) binding.30 Furthermore, pH titration and

chemical modification studies implicate histidine and tyrosine residues on the IgG Fc

in the binding of RF to the Ga dete~ min ant.~ 'he epitope is situated in the C y2 -C y3

interface region of IgG and involves three polypeptide loops, two from the Cy2 and

one from the C y3 domain, which come together in close proximity in the C y2 -C y3

interface region. The amino acids involved in the binding are likely to include

histidine at positions 310, 433, and 435 and tyrosine at 436. Despite the differing

causes of RF production, the specificity to this region appears to be generally

preserved.



MAGEED et aL: RF 299

IDIOTYPIC AND GENETIC BASIS OF HUMAN RF PRODUCTION

Idiotypic Analysis

The main questions addressed in studying the RF idiotype have been whethernatural RF found in patients with lymphoproliferative disorders and pathogenic RF n

RA patients are related and whether they are direct copies of germline genes or

products of somatic mutations. The first set of human RF to be extensively studied

were IgM paraproteins (or “M ” components) isolated from the serum of patients w ith

lymph oproliferative disorders (Waldenstrom’s macrog lobulinemia and cryoglobuline-

mia). Anti-idiotypic antisera to these monoclonal antibodies were obtained following

absorption of rabbit antisera with paraproteins from different individuals and with

polyclonal IgG. This approach allowed the d efinition of three cross-reactive idiotypic

families, or groups, designa ted Wa, Po, and Bla.’*32 decade later, practically similarprotocols were followed in the selection of mouse monoclonal antibodies identifying

cross-reactive idiotopes expressed by RF para protein^ ^^ A rapidly growing library of

antibody sequences (derived from mRNA or DNA) from hybridomas whose protein

products reacted with the antibodies has enab led the determ ination of the contribution

of germline genes to RF production and has established the range of genes whose

protein products were recognized by the monoclonal anti-idiotypic a n t i b o d i e ~ . ~ ~

We investigated the expression of V gene markers and cross-reactive idiotopes

(CRI) using various murine monoclonal anti-idiotypic antibodies. These included G 6

and G 8 , which recognize VH1-associated CRI;33+35 6 and D12, which recognizeVH3-associatedCRI;36and Lc 1 , which recognizes protein products of the vH4 family

of gene s3’ V, gene family products were determined by C 7 (specific for the VK 3

family) and C6 (specific for the V 3b ~ u b f a m i l y ) . ~ ~7.109 was used to identify a C RI

that was encoded by the VK3b-subfamily ariable region gene, K ~ 3 2 5 . ~ ~he m ajority

of the RF paraproteins and RF from imm unized donors expressed at least one of these

markers (TABLE ). Within individual donors, particularly the normal immunized

donor, MR , groups of RF were found that shared identical patterns of V gene markers,

suggesting a restricted r e ~ p o n s e .~

Analyses of the monoclonal RF paraproteins showed a high degree of CRIe x p r e s s i ~ n . ~ ~he m ost-striking feature was the predominant use of the VK 3-associated

light-chain subgroup and CRI, with more than 90% expressing the V,3 light-chain

s ~ b g r o u p . ~ . ~ ~urthermore, it transpired that at least 60% of all IgMK RF express light

chains encoded by two V,3 genes whose protein products are recognized by the

monoclonal an ti-idiotypic antibodies 17.109 and 6B6.6.40Similarly, albeit to a lesser

extent, the RF paraproteins were also shown to widely react with the G6 and G8

monoclonal anti-idiotypic antibodies. When the VL and VH genes were considered

together, three different combinations of VH-VL could be iden tified. The largest grou p

of these RF were those that expressed kappa light chains reactive with the m onoclonal

antibody 17.109 in association with VH genes w hose produc ts were reactive with the

monoclonal antibody G6, which was subsequently shown to be a serological marker

for the DP- 10 and h vl 26 3 genes.40A second group of these RF expressed a V K 3 ight



MAGEED et al : RF 301

V,3 in the R F from normals, together with a distribution of VH gene segments of the

v"1, vH3, and vH4 families, closely resembled the distribution seen among the RF

paraproteins. In contrast, RF from the synovial membrane of RA patients predomi-

nantly used heavy-chain segments encoded by genes from the vH3 family. These data

imply that RF paraproteins may be representative of R F used in normal physiological

responses, but that have undergone neoplastic transformation. Th e overall conclusion

from these studies is that RF produced in the synovial tissue of RA patients are

encoded by a wider range of IgV genes than the RF paraproteins and R F in normal

TABLE 2. Frequency of VHGen e Expression by RF Paraproteins and Monoclonal

IgM RF Produced by Hybridomas Established from Peripheral Blood of Immunized

Norm als and Synovial Membrane of RA Patients"

No. of No. of RF No. of RF

Family Individual Expressing Expressing ExpressingV H Paraproteins from Normals from RA SM

Used VHGene Used the Gene the Gene theG ene Total

DP-10DP-75HG31-13

total

3 DP-54DP-49DP-46b18DP-35DP-47DP-42v3-53DP-3 1

total

4 4.18DP-7 1

VH4.35DP-65DP-66

total

7 DP-2

Overall total

1

1

I

0

00

3

10

310

1

1

0

8

17

1 11

0 21 1

1

3 15

0 43 42 30 1

1 21 20

0

18 19

1 1

0 10 1

0

1 1

2 5

1 2

14 41

"Summary of the data on VH ene use in RF-producing B cell hybridomas generated in our

laboratories. For the sake of simplicity, data generated in other laboratories are not included.Only IgM RF paraproteins and IgM RF-producing lineshybridomas that have been well

characterized are included. IgG and I gA RF-producing lineshybridomas are not given. Normal

PBL and RA SM indicate the original sources of the RF-producing cells used to establish the

lineshybridomas.




human immunized donors (HID). Furthermore, there is an increase in the representa-

tion of vH3 genes in the synovial RF hybridomas.

Interestingly, when the values of RF expressing the CRI detected by the monoclo-

nal antibodies G6, G8, B6, D12, and LC1 were assessed in monozygotic twins,

concordant or discordant for RA, it appeared that the absolute levels of the CRImarkers were higher than in normals and that the values were very similar within

twins, irrespective of disease status, but different between unrelated twins.42 These

findings were taken to imply that there may be additional regulatory mechanisms,

possibly genetic or idiotypic, that control the level of natural R F and that some of

these regulatory mechanisms may be inherently abnormal in individuals with a

genetic predisposition to develop

NATURE OF SOMATIC MUTATION IN FtF

The increase in the aftinity of antibodies to antigen and isotype switching is often

referred to as maturation of the immune response. This maturation requires that the

Ag-binding s ite of the antibody molecule be structurally different from antibodies in

the primary response. This is typically achieved by som atic mutation and repertoire

shift. Sequence analysis of IgV gene mRNA from a large collection of B cell

hybridomas with specificity for a variety of different antigens has shown that such

structural changes are often achieved through mutations leading to amino acid

replacements in the CDRS ~ In such responses, hypermutation of IgV genes is

followed by bo th positive and negative selection pressures through w hich B cells withhigh-affinity receptors are selected for expansion. B cells with no mutations or with

limited mutated receptors that have low affinity for antigen die, presumably by

apoptosis. Besides an ncrease in affinity, positively selected B cells show a character-

istic elevation of the ratio of R:S in the C D R S . ~urprisingly, R F produced in the

healthy immunized normals also show evidence in their IgV genes of extensive

somatic mutations (9-16 mutations per VHgene). There was, however, evidence for

selection against replacement mutations in the CDRs, suggesting that high-affinity

variants were negatively selected. Without any selective pressures, random mutations

in CDRl+2 of the DP-10 gene segment would generate R:S ratios of 4.4 s Theoverall R:S ratio of the C D R l + 2 in the VH region of the RF rom the immunized

donor MR, who had 10R F encoded by the D P- 10 gene, was significantly lower at 0.4

p < 0.0027, Fisher’s exact test, two-sided) (TABLE ). In contrast, RF from RA

synovia, bone marrow, and peripheral blood did not show a similar counterselection of

replacement mutations in the CDRs as seen in RF from the normal immunized donors.

RF from RA atients had between 2 and 18 nucleotide mutations compared with 9-1 6

in immunized normals (TABLE and FIGURE).

Nucleotide sequencing of the light chains of 5 of the RF established from the

normal immunized donor, MR, confirmed the clonal relatedness and showed that all

used the Kv325 light-chain germline gene rearranged to the JK1gene segment with

identical VK-J, junctions (RF-MR2, R F-MR13, RF-MR16, RF-MR28, and RF-MR30). Hypermutation was also evident in the light chains, but to a lesser degree than

in the heavy chains. Four of the 5 related clones had 1 shared replacement mutation in



MAGEED et al.: RF 303

the light chain. These data confirmed that RF from immunized donors clearly undergo

somatic hypermutation.

ISOTYPE SWITCH IN RF-PRODUCING B CELLS

Following the initial interaction between virgin B cells and Ag, gennline gene-

encoded an tibodies of the IgM isotype are produced. The ability of the immune system

to adapt to combat antigenic challenge is manifested by further differentiation of

Ag-specific B cells to produce antibodies with high affinity and of different i ~ o t y p e s . ~ ~

Individual B cells with specificity for Ag can switch isotype while retaining the same

recombined IgV genes and an tibody specificity. These features of the immune system

have been widely used to assess the nature and mechanisms of induction of immune

responses to self-antigens in murine models of autoim mune disease^:^.^' Specifically,

TABLE 3. Total Number of Nucleotide Changes Observed in the VHGenes of RF

Derived from Peripheral Blood of Imm unized Normals, Synovial Membrane of RA

Patients, and IgM Anti-Red Blood Cell Rh(D) Antigen from Immunized Normals

Average Total No . of Total No. ofNo. Of Nucleotide Nucleotide

No. of Nucleotide Changes ChangesAntibodies Changes Seen in Seen in

CDR and 2ncluded from the 1.2 , and 3Source of in the Germline R:S R:SAntibody Analysis Gene R S Ratio R S Ratio

immunizednormals 10 12.0 17 16 1.1 7 16 0.4

RA SM 7 16.1 30 34 0.9 33 14 2.4

anti-Rh(D) 7 4.3 8 12 0.7 8 8.0

RF from

the ability to detect autoan tibodies of the IgG, and to a lesser extent IgA, isotypes withincreased affinity and similar specificity to autoantigens has been used as an indicator

of specific autoantigen-driven imm une responses in autoan tibody production.

One of the most-characteristic serological features of RA patients with severe

active disease and extra-articular symptom s is high levels of IgG W 8 High levels of

IgG RF are often associated with circulating immune complexes and complement

activation. The finding of large-size immune complexes is particularly important in

RA synovial inflammation because such complexes can contribute to pathogenesis

through com plement activation, cell recruitment, and cartilage and bone damage.I6 In

addition to the finding of high levels of IgG RF in the serum of a high proportion of

RA patients, IgG RF-producing B cell hybridomas were also isolated from three RA

patients in our stud ies.49 n con trast, none of the imm unized norm al donors produced

IgG RF or IgA RF, mplying that the switch of isotype to IgG and IgA is a more

common feature of R F-produc ing B cells in RA, but not norm al, individuals.




?Rl ..**..*'E~~'****.* .*ttt*ttt~~~.* * **..

4 13 24 26 27 28 29 3 0 3 1 32 33 34 35 31 ( 3 53 54 56 57 59 63

AGC T A T DCT A W AGC GTU ACA CCT ATC DOT ACA M C Au:P-10 CTD AAG CCT GG OOC ACC TK AGC

hsM ERp m1 4 __.._ _._ _._- - G

Flu69 7 74 I1 79 80 82 83 84 00 89 93 94 95

ACG ATC GAA GC GCC TAC GAQ CTD ACC TCT GAG 0 TAT TACP-I0

l L 4 s a FRp'ISI .._ .._ .._ __.a .__

~p m14 A -T- __._ A a --t

Rp-- A -0- __.-- - - _ _ - - - - - --- - - - --t

~ p - M R 3 3 _.__.A ... _._.. .a .. t

~p-w _._C A ...__. __.a . -t

Rp rnl - - - --- c --c

FIGURE 1 Nucleotide sequences of the rearranged DP-I0 VH genes of IgM RF-producinghybridomas established from the synovial membrane (SM) of an RA patient, from bone marrow(BM) and peripheral blood lymphocytes (PBL ) of other RA patients, and from PBL of a normalhuman immunized donor (HID), MR, in comparison with the DP-10 ge m lin e gene sequence.Nucleotide sequences of the reamanged genes from RF-producing hybridomas are comparedwith the sequence of the DP-10 germline gene to show nucleotide mutations. B ases similar tothe germline gene are given as dashes, mutations that result in amino acid replacements aregiven in uppercase letters, and silent mutations are given in lowercase letters. Only co dons that

are different from the germline in the HID RF are given in the figure. RF established from thebone marrow (BM) B cells and peripheral blood (PBL ) of RA patients are included in the figureto show that RF rom R A patients, irrespective of their original anatomical compartment, havedifferent patterns of mutations compared with HID RF. The data are based o n references 4543.



MAGEED ef al.: RF 305

CLONA L EXPANSION AND AFFINITY M ATURATION

It is now recognized that generation of mutations in the rearranged IgV genes

alone would not be adequate to account for the phenom enon of affinity maturation in

antibodies produced during the secondary phase of imm une responses.43A process (or

processes) capable of selectively activating and expanding B cells producing high-

affinity antibodies is also essential. In T cell-dependent immune responses, the

process of selective activation and expansion of higher affinity antibodies occurs

initially in the T cell areas of the spleen and lymph nodes.50 Subsequently, selected B

cells migrate to B cell follicles where they expand and initiate a germinal center

reaction, leading to clonal expansion of the higher affinity antibody-producing B

cells. This feature of the immune response has been used to study the kinetics of

immune responses to exogenous antigen^.^' Such studies have been achievable with

small hapten antigens, such as oxazolone and nitrophenyl, but are thought to bedifficult with complex antigens, which have numerous immunogenic epitopes. In our

studies, however, two IgM RF (RF-SJ1 and RF-SJ2) isolated from the same RA

patients were found to be clonally related.52The VHof RF-SJ2 had only 2 nucleotide

differences from its germline counterpart (GL-SJ), whereas RF-SJ1 had accumulated

18 nucleotide mutations. The light chain of RF-SJ1 also had acquired 18 nucleotide

mutations compared with the RF-SJ2. Functional affinity studies showed that the

extensively mutated antibody (RF-SJI) had a 100-fold higher affinity fo r human IgG

than the RF-SJ2. These observations provide further evidence that RF in RA do

undergo m aturation including clonal expansion.The derivation of the clonally related RF from the normal immunized donor MR

allowed a detailed examination of the relationship between somatic hypermutation

and affinity (TABLE).Among this group of RF, there is very little increase in affinity

with the accumulation of mutations in the FRs. On average, the RF had Kd values

of 2 X M). When compared with the affmity of RF from

RA patients, using the same gene segments, the mean K d was approximately 10-fold

lower, that is, 3 X M to 5 X lo-’ M) (TABLE).RF in RA

thus do not appear to be subject to regulatory mechanisms that control their affinity.

M (range, 0 63 X

M (range, 8 X

REPERTOIRE SHIFT IN THE IgV GENES ENCO DING RF

In addition to diversifying the antibody repertoire of an immune response by

somatic mutation, maturation of the antibody response is frequently accompanied by a

shift in the repertoire of the IgV genes used.53Studies of primary immune responses to

defined hapten antigens, such as oxazolone, in mice revealed that, although the

primary response was entirely dominated by a particular pair of VHand VL germline

genes, only 20 of the antibodies utilized the same com bination in the secondary and

tertiary responses.a This modification in the repertoire is called “repertoire shift.”

In humans, studies of immune responses to rabies vaccine have provided evidence

for repertoire shift.54For RF, there are limited data to suggest that repertoire shift may

contribute to the diversity of the RF repertoire in R A synovia (TABLE ).

Repertoire shift signifies that IgV genes utilized by low-affinity antibodies in the



3 6 ANNALS NEW YORK ACADEMY OF SCIENCES

TABLE . Functional ffi ni ti es of IgM RF rom Different Sourcesa

Nucleotide NucleotideMutations

in FR1+ 2+3ource

and CloneExamined R S

Mutationsin C D R l + 2

R S

HID PBLRF-MR 1RF-MR30RF-MR2RF-MR 16RF-MR37RF-MR33RF-MR25RF-MR14

RF-MR28

RA SMRF-SJlRF-SJ2RF-SJ3RF-SJ4RF-TS 1RF-TS2RF-TS5RF-KL1

Ki (MI

2 x 10-65 x 10-6

1 x 10-66 x2 x 10-69 x 10-72 x 10-6

2 x 10-6

1 x 10-6

2.4 x2.7 x4.6 x7.9 x 10-87.7 x 10-87.1 X 10-8

6.8 X lo-’2.1 x 10-7

EMCKAS ND ND ND ND 2.5 x 10-7

“The data are based on references 45 and 52. Only IgM R F paraprotein KAS and IgM

RF-producing lineshybridomas that have been characterized for specificity and gene use are

given. HID PEL and RA SM indicate the original RF-producing cells used to establish the

lineshybridomas. EM C indicates that the RF paraproteinKAS was from a patient with essential

mixed cryoglobulinem ia. All RF-producing hybridomas were sequenced at the mR NA level,

whereas paraprotein KAS was sequenced at the am ino acid level. ND = not determined. The

dissociation constant, Kd as deduced from the Scatchard plot using an ELISA assay for

binding to native human IgG.

primary response may not be capable of, or adequate for, encoding high-affinity

antibodies that dominate the secondary response. The reasons for this may lie in

intrinsic properties of the sequence or may reflect regulatory influences of B cell

selection in germinal centers.55 n responses to exogenous an tigens, the limiting factor

in determining repertoire is Ag concentration. Levels of exogenous antigens decline

gradually, reaching a minimum at the end of the acute response. This leads to

competition for the limited concentration of Ag, thus allowing only B cells with the

highest affinity to survive. In responses to self-antigens in normal individuals,however, the process of B cell recruitment in germinal centers is regulated by a

number of mechanisms, including high levels of soluble self-antigens that block

self-reactive B cells from reaching germinal centers and prevent maturation.’6

Some of these regulatory processes may be abnormal in the ectopic germinal



MAGEED ef uL: RF 307

TABLE . IgV Gene Use in RF Paraproteins and Monoclonal FW-producing Clones

Established by Hybridoma Technology from Peripheral Blood of Normal Immunized

Donors and the Synovial Membrane of RA Patients0

Closest ClosestVH Germline GermlineRF Family VHGene VLFamiIy VLGene Source

RF-lT9RF-MR514RF-TT5

RF-MRCRF-MRlRF-TrM7BOR

KASHULRF-TS 1RF-TS2RF-TS3RF-TS4RF-TS5RF-SJ1RF-SJ2RF-SJ3RF-SJ4RF-KLIB42c93D53H4H6HAFIOC6G4D5G9A2A2

333111

31

1

131

3333343334311

31

343

DP-54DP-54DP-54DP-10DP-10DP-10DP-54DP- I0

DP-10DP-I0DP-I0DP-494.16DP-49DP-35DP-46DP-46DP-49DP-66DP-47DP-478-1B4.41DP-47DP-75hvlflONDDP-46DP-46DP-79DP-49

K3aK3aK3a~ 3 b~ 3 b~ 3 bK3a~ 3 b

~ 3 b~ 3 b~ 3 bK3a

ND

A1

A l

~ 3 b~ 3 b

A1

h3A3A3

A

~ 3 b~ 3 bx3

K2

K1

K 1

K4

K 1

K1

Kv328Kv328Kv328Kv325Kv325Kv325Kv328Kv325

Kv325Kv325Kv325Kv328A23NDllL102lv117lv117Kv325Kv305Vdlv1042

15011150

11150

k18lv80 1HK102

v gv glv318OlUOZ

HID PBLHID PBLHID PBLHID PBLHID PBLHID PBLEMCEMC

EMCEMCRA SMRA SMRASMRA SMRA SMRA SMRA SMRA SMRA SMJRA SMRA SMRA SMRA SMRA SMRA SMRA SMRA SMRA SMRA SMRA SMRA SM

The data are based on references 4, 5 45, and 57-63. Only IgM R F paraproteins or IgM

RF-producing lineshybridomas that have been well characterized are given. HID PBL and RA

SM indicate the original RF-producing cells used to establish the linesfhybridomas. JRA refers

to juvenile rheumatoid arthritis. M7, BOR, KAS, and HUL are RF paraproteins from patients

with essential mixed cryoglobulinemia EMC)orWaldenstrom’s macroglobulinemia.These RF

were sequenced at the protein level. ND = not determined.

centers in synovial membranes of patients with RA, leading to immune response

maturation.

SUMMARY

Recent advances in molecular biological and human cell hybridization technology

have significantly advanced the knowledge of mechanisms that underlie human




rheumatoid factor (RF) production. These advances have provided insight into the

etiopathogenesis of synovial inflammation and lymphocyte recruitment in rheumatoid

arthritis (RA)joints.

We have examined the mechanisms that lead to RF production in RA patients and

those that regulate RF production in normals. The studies revealed structural featuresthat distinguish RF produced in normals from those produced in RA synovial tissue.

There are significant differences in the use of VL and V H genes between the two RF

populations. Furthermore, IgV genes encoding synovial RF in RA have extensive

evidence for nucleotide changes, leading to amino acid replacement in the complemen-

tarity determining regions (CDRs). In addition, RF produced in RA synovia show

evidence for affinity maturation, isotype switch to IgG RF, and repertoire shift

indicative of a continued recruitment of B cells. Together with computer modeling and

crystallographic studies, our data suggest that the mechanisms that operate on RF

selection in RA synovia are similar to immune responses to exogenous antigens. Incontrast, RF established from human immunized donors (HID) are characterized by a

very low ratio of replacement to silent R:S) ucleotide changes in the CDRl+2. In

addition, there is little increase in affinity with increasing num bers of mutations. There

is thus evidence for regulatory mechanisms that limit affinity maturation of RF n

normals.

ACKNOWLEDGMENTS

We would like to thank D. G. W illiams, of the Kennedy Institute of Rheumatology,for useful discussions and for the literature survey used in TA BLE .

REFERENCES

1. KUNKEL, . G., V. AGNELLO, . G. JOSLIN,R. J. WINCHESTER J. D. CAPRA.1973.Cross-idiotypic specificity among monoclonal IgM proteins with anti-gamma-globulin activ -ity. J Exp. Med. 137: 331-342.

KUNKEL, .G., R. J. WINCHESTER,. G. JOSLIN J. D. CAPRA. 974. Similarities in the lightchains of anti-gamma-globulins showing cross-id iotyp icspecificities. . Exp. Med. 139: 128-

136.3. RANDEN,., K. M. THOM PSON,. PASCUAL, . VICTOR, . BEALE, . COADWELL, ~R RE ,

J. D. CAPRA& J. B. NATVIG. 992. Rheumatoid factor V genes from patients withrheumatoid arthritis are diverse and show evidence of an antigen-driven response. Immunol.Rev. 128: 49-71.

4. BROWN, . M., K. J. FITZGERALD,.M o m , R. A. MAGEED, . G. WILLIAMS R. N.MAINI. 995. Sequence analysis of im munoglobu linheavy-chain va riable region genes fromthe synovium of a rheumatoid arthritis patient shows little evidence of mutation, but diverseCDR3. Immunology 84: 367-374.

5 . THOMPSON, . M., M . B ~ R R E ~ ~ , E N ,. RANDEN, ~R RE J. B. NATVIG. 995. V-generepertoire and hypermutation of rheumatoid fac tors produced in rheumatoid synov ial inflam-mation and imm unized healthy donors. Ann. N.Y. Acad. Sc i. 764: 440449 .

6. NOSSAL, . J. V. 1992. Cellular and m olecu lar mechanisms of B -lymphocyte tole rance . Adv.Immunol. 52: 283-33 1.

7. HARTLEY,. B., J. CROSBIE, . BRINK,. B. KANTOR,. BASTEN C. C. GOODNOW.991.Elimination from peripheral lymphoid tissues of self-reactive B lymph ocytes recognisingmembrane-bound antigens. Nature 353: 765-769.

2.




8.

9.

10.

1 1 .

12.

13.

14.

15.

16.

17.

18 .

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

CARSON, .A. P. P. CHEN& T. J. KIPPS. 1991.New roles for rheumatoid factor. J. Clin. Invest.

BIKOFF, . K. 1983. Tolerance for self Ig at the level of the Lyl+ T cell. I. Exp. Med.

ROOSNEK, . & A. LANZAVECCHIA.991. Efficient and selective presentation of antigen-

antibody comp lexes by rheum atoid factor B cells. J . Exp. Med. 173: 4 8 7 4 8 9 .TIGHE,., P. P.CHEN,R. TUCKER,. J. KIPPS J. ROUDIER,. R. JIRIK &D.A. CARSON. 993.

Function of B cells expressing a human immunoglobulin M rheumatoid factor autoantibody

in transgenic mice. J. Exp. Med. 177: 109-1 18.

region gene usage of human monoclonal rheumatoid factors derived from healthy donorsfollowing immunisation. Eur. J . Immunol. 24: 177 1-1778.

RANDEN,. 0 J. MELLEYE, ~RRE J . B. NAT VIG. 995. The identification of germ inalcentres and follicular dendritic cell networks in rheumatoid synovial tissue. Scand. J .

Immunol. 41: 4 8 1 4 8 6 .

MUNTHE,. & J . B. NATVIG.972. Immunoglobulin classes, subclasses, and com plexe s of IgG

rheumatoid factor in rheumatoid plasma cells. Clin. Exp. Imm unol. 12: 55-70.MOYNIER,., M. ABDERRAZIK,. DIDRY,. SANY& J . BROCHIER992.The B cell repertoire

in rheumatoid arthritis. III. Preferential homing of rheumatoid factor-producing B cell

precursors in the synovial fluid. Arthritis Rheum . 35: 49-54.

POPE,R. M., D. C. TELLER& M. MANNIK . 975. Intermediate complexes formed by

self-association of IgG rheumatoid factors. Ann. N.Y. Acad. Sci. 256: 82-87.

PANAYI, S. , J . S. LANCHBURY G. H. KNGSLEY.992. The importance of the T cell in

initiating and maintaining the chronic synovitis of rheumatoid arthritis. Arthritis Rheum.

ALARCON,. S. 1995. Epidemiology of rheumatoid arthritis. Rheum. Dis. Clin. North Am.

21: 5 8 9 4 0 4 .

ZIFF,M. 1993. Role of endothelium in the pathogenesis of rheumatoid synovitis. Int. J. TissueReact. 15: 135-137.

NATVIG,. B. A . J . QUAYLE. 995.V-genes of T-cell receptors in rheum atoid arthritis. Ann.

N.Y.Acad. Sci. 756: 138-146.

MELCHERS,. H. H. PETER& H. EIBEL.1995. The T and B cell repertoire o f patien ts with

rheumatoid arthritis. Scand . J. Rheumatol. Suppl. 101: 153-162.

SHMERLING,. H. & T. L. DELBANCO.991. The rheumatoid factor: an analysis of clinical

utility. Am. J . Med. 91: 528-534.

GRIPENBERG,., A. M. TEppo, P. KURKI, . GRIPENBERGT. HELVE. 1988. Autoantibodyactivity of cryoglobulins and sera in systemic lupus erythematosus: association of IgM c lass

rheumatoid factors with Rayn aud's syndrome. Scand . J . Rheumatol. 17: 249-254.

MULLEQ . , P. OXH OLM , . MIER-MADSEN A. WIIK.1989. IgA and IgM rheumatoid

factors in patients with primary Sjogren's syndrome: correlation to extrag landu lar manifesta-tions. Scand. J . Rheumatol. 18: 29-31.

SASSO, . H., C. V. BARBER, . A. NARDELLA, . J YOUNT& M. MA NNIK . 988. Antigen

specificities of human monoclonal and polyclonal IgM rheumatoid factors. J. Immunol.

RANDEN,., K. M. THOMPSON,. B. NATVIG, 0RRE & K. WAALEN.1989. Human

monoclonal rheumatoid factors derived from the polyclonal repertoire of rheumatoid synovialtissue: production and characterization. Clin. Exp. Immunol. 78: 13-18.

MAGEED, . A. , D . A. CARSON R. JEFFERIS. 1988. Immunogenic and antigenic epitopes ofimmunoglobulins. XXIII. Idiotype and molecular specificity of human rheumatoid factors:analysis of cross-reactive idiotype of rheumatoid factor paraproteins from the Wa idiotype

gro up in relation to their IgG subclass specificity. Scand. J . Immunol. 28: 233-240.NARDELLA,. A. I . R. OPPLIGER,.C. STONE, . H. SASSO,M. MANNIK,. SJOQUIST,. K.

SCHROEDER,. CHRISTENSEN,. J . JOHANSSON & L. BJORCK.988. Fc epitopes for human

rheumatoid factors and the relationships of rheumatoid factors to the Fc binding proteins ofmicroorganisms. Scand. J. Rheumatol. Suppl. 75: 190-198.

87: 379-383.

158: 1868-1880.

THOMPSON,.M., I. RANDEN,M. BBRRETZEN,0 0RRE & J . B. NATVIG.994. Variable

35: 729-735.

140: 3098-3107.




47.

48.

49.

50.

51.

52.

53.

54.

55 .

56.

57.

58.

59.

60.

61.

62.

63.

SHLOMCHIK,., M. M ASC ELLI, . SHAN,M. Z. RADIC,D. PISETSKY, . MARSHAK-ROTHSTEIN M. WEIGERT. 990. Anti-DNA antibodies from autoimm une mice arise byclonal expansion and somatic mutation. J. Exp. Med. 171: 265-292.

MA GEE D, . A., J. R. KIRWAN,. W. THOMPSON,. A. MCCARTHY E. J. HOLBOROW.1991. Characterization of the size and composition of c irculating immune com plexes in

patients with rheumatoid arthritis. Ann. Rheum . Dis. 5 0 231-236.RANDEN,., V. PASCUAL,. VICTOR, K. M. THOMPSON,0 ~ R R E, D. CAPRA& J. B.

NATVIG.993. Synovid IgG rheum atoid factors show ev idence of an antigen-driven imm uneresponse and a shift in the V-gene repertoire compared to IgM rheumatoid factors. Eur. J.Immunol. 23: 1220-1225.

LIU,Y-J., J. ZHANG,. L. LANE,E. Y. CHAN I. C. MACLENNAN.991. Sites of specific Bcell activation in primary and secondary responses to T cell-dependent and T cell-independent antigens. Eur. J. Immunol. 21: 295 1-2962.

JACOB,. & G.KELSOE.1992.In siru studies of the primary imm une response to (4-hydrox y-3-nitropheny1)-acetyl. 11.A common clonal origin fo r periarteriolar ymphoid sheath-associatedfoci and germinal cen ters. J. Exp. Med. 76: 679488 .

RANDEN,., D. BROWN, . M. THOMPSON,. C. HUGHES-JONES,. PASCUAL,. VICTOR,J. D. CAPRA, 0RRE & J. B. NATVIG.992. Clonally related IgM rheumatoid factorsundergo affinity maturation in the rheumatoid synov ial tissue. J. Lmmunol. 148: 32963301.

BEREK , ., A. BERCER& M. APEL.1991. Ma turation of the imm une response in germinalcenters. Cell 67: 1121-1129.

UEKI,Y., 1.S. GOLDFARB,. HARINDRANATH,. GORE,H. KOPROWSKI,A.. NOTKINS P.CASALI.990. Clonal analysis of a human antibody response: quantitation of precursors ofantibody-producingcells and generation and characterization of monoclonal IgM , IgG, andIgA to rab ies virus. J. Exp. Med. 171: 19-34.

S LEBECQUE.996. Sequen tial riggering of apoptosis, somatic mutation, and isotype switchduring germinal center developm ent. Semin . Immuno l.8: 169-177.

PULENDRAN,., M. KARVELAS G. J. V. NOSSAL. 994. A form of imm unolog ic tolerancethrough impairment of germinal cen ter development.Roc.Natl. Acad. Sci. U.S.A. 91: 2639-2643.

PASCUAL,., I. RANDEN, . M. THOMPSON,. SlOUD,0 0RRE. J. NATVIG J. D. CAPRA.1990. The com plete nucleotide sequences of the heavy chain variable regions of sixmonospecific rheumatoid factors derived from Epstein-Barr virus-transformed B ce llsisolated from the synovial tissue of patients with rheumatoid arthritis: furthe r evidence thatsome autoa ntibodies are unmutated cop ies of germline genes. J. Clin. Invest. 86: 1320-1328.

YOUNGBLOOD, ., L. FRUCHTER,. DING, . LOPEZ,V. BONAGURAA. DAVIDSON.994.Rheumatoid factors from the peripheral blood of two patients with rheumatoid arthritis aregenetically heterogeneous and somatically mutated. J. Clin. Invest.93: 852-861.

EZAKI,., H. KANDA, . SAKAI,N. FUK UI, . SHING U, . NOBUNAGA T. WATANA BE.

1991. Restricted diversity of the variable region nucleotid e sequences of the heavy and lightchain s of a human rheumatoid factor. Arthritis R heum. 34: 343-350.

NEWKIRK,. M., R. A. MAG EED, . JEFFER IS,. P, CHEN& J. D. CAPRA. 987. Com pleteamino acid sequences of variable regions of two human 1gM rheumatoid factors, BOR andKAS of the Wa idiotypic family, reveal restricted use of heavy and light chain variable andjoinin g region gene segments. J. Exp. Med. 166: 550-564.

ERMEL , . W., T. P. KENNY, . WONG, . SOLOMON,. P. CHEN& D. L. ROBBINS.994.Preferential utilization of a novel V lambda 3 gene in monoclonal rheumatoid factors derivedfrom the synovia l cells of rheumatoid arthritis patients. Arthritis Rheum. 37: 860-868.

SOTO-GIL, . W., T. OLE E,B. K. K LINK, . P. KE NN Y, . L. ROBBINS,. A . CARSON P. P.CHEN. 992.A systematic approach to defining the germline gene counte rparts of a mutatedautoantibody from a patient with rheum atoid arthritis. Arth ritis Rheum. 35: 356-363.

ERMEL, . W., T. P. KENNY,P. P. CHEN& D. L. ROBBINS.993. Mo lecular analysis ofrheumatoid factors derived from rheumatoid synovium suggests an antigen-driven responsei n inflamed joints. Arthritis Rheum. 6:380-388.

LIU, Y-J., C.ARPIN, E BOUTEKLER,.GURET, J. BA NC HE REAU ,.MARTINEZ-VALDEZ