|
|
 Previous Article | Table of Contents | Next Article 
Blood, 15 May 2001, Vol. 97, No. 10, pp. 3319-3321
CORRESPONDENCE
To the editor:
WA monoclonal rheumatoid factors and non-Hodgkin
lymphoma
De Re et al describe the sequences obtained from B-cell
monoclonal expansion analysis of patients with non-Hodgkin lymphoma (NHL).1 The characterization by the authors of 17 patients with NHL makes comparisons of the antibodies found to rheumatoid factor
genes and hepatitis C virus (HCV) anti-E2 antibodies. The study did not
include serologic or sequence determinations demonstrating identity
between the serum monoclonal rheumatoid factors (mRFs) and the
immunoglobulins (Igs) antigen receptors of the lymphoma cells in the
patients. Sequence data can be useful in attempting to understand the
origin of NHL in patients with chronic HCV infection, but there are
pitfalls in drawing conclusions on specificities of antibodies based on
sequence homologies to only part of the sequences that constitute the
combining site. The sequence homology analyses in this study were
flawed. Consequently, the conclusions that malignant cells in type II
cryoglobulinemia are derived from RF-producing cells may be only
partially correct, and the speculation that HCV E2 protein is involved
in the production of RF antibody is not warranted by the data presented. Because there are extensive sequence and other criteria for mRF bearing
the WA cross-idiotype (WA mRF), the predominant mRF in type II
cryoglobulinemia,3 in the absence of demonstrating identity between the serum mRF and the malignant cell Ig antigen receptor, these criteria may be used to provide indirect evidence that
the malignant cells arose from the major mRF-producing cells that
proliferate during the nonneoplastic course of type II
cryoglobulinemia. WA mRF bear a combining-site cross-idiotype (WA) that
requires both light- and heavy-chain V-region sequences. Typically, the WA cross-idiotype arises from coexpression of a VH1 gene
(DP-10), a D region (D21/9) that encodes 9-12 amino acids beginning with glutamic acid and ending in a proline,
and JH4 with a VK3 (Kv325) rearranged to JK1. Although VH1 is most common,
VH3 can be used (eg, M7-RF-WA; GenBank accession number
U03400); the latter group differs from the classic WA in that a VH3
gene (DP-54) rearranged to JH3 or
JH4 is coexpressed with Kv328 rearranged to
JK1. The heavy chain CDR3 amino acid sequence, 13 amino
acids long, differed significantly from the classic WA. Thus it appears
that there are 2 classes of WA mRF.2,3 The speculation
that De Re et al made regarding HCV E2 protein is untenable because
neither class of WA mRF has any significant homology with the anti-E2 HCV antibodies cited. Of the 17 immunoglobulin sequences studied by De Re et al, only 4 fit
criteria for the WA mRF. The homologies of the sequences deduced in
their Table 4 were arbitrary. The E value scores are listed for each
IgH CDR3-deduced amino acid sequence to give legitimacy to the
assignment, but this was used improperly. The inclusion of the
JH sequence shifts the score higher and muddles the
interpretation by making the similarities appear more significant than
they are. Moreover, as noted above, sequence homology is not the only
criterion for WA mRF. The translated D-region sequences of the CDR3
from classic WA often bears little homology. But the structural
conformation imposed by the size of the D region, the presence of
critical D-region amino acids at precise locations, and the restricted use of specific immunoglobulin genes generate the WA mRF. Thus among
the patients 1-4 that were considered by De Re et al to be similar to
WOL (a classic WA mRF), only the patient 1 sequence bears similarity to
WOL because the others lack critical D-region amino acids. Patients 5, 6, and 10 appear to be similar to M7-RF-WA (accession number U03400).
The IgH from patient 13 does not appear to be similar to BOR (a classic
WA mRF), based upon the CDR3 sequence and the presence of a VH4 gene
instead of VH1, which makes it more similar to the cold agglutinins.
The patient 14 IgH was reported to have a CDR3 similar to a mouse
rheumatoid factor, RF-MR20. Establishing lineage of malignant cells to mRFs requires the
demonstration of identity of the Ig antigen receptor and the preneoplastic mRFs. The data presented by De Re et al did not accomplish this, but the presence of 4 WA mRF NHLs did provide additional indirect evidence for the original hypothesis that HCV is
responsible for proliferation of a specific set of B cells in patients
with type II cryoglobulinemia and that transformation to malignant
cells occurs in some of these cells.4 Although HCV is not an oncogenic virus, HCV nonstructural protein NS3
and HCV core protein have been reported to transform cells in vitro. It
remains to be determined if the malignant transformation of the
HCV-driven proliferating B cells is a stochastic event, a result of a
subset of these B cells prone to malignant-transforming mutational
events, or a direct effect of the virus.
Glenn Knight and Vincent Agnello
Correspondence: Vincent Agnello, Lahey Hitchcock Clinic, 41 Mall
Rd, Burlington, MA 01805-0001
References
1.
De Re V, DeVita S, Marsotto A, et al.
Sequence analysis of the immunoglobulin antigen receptor of hepatitis C virus-associated non-Hodgkin lymphomas suggests that the malignant cells are derived from the rheumatoid factor-producing cells that occur mainly in type II cryoglobulinemia.
Blood.
2000;96:3578-3584[Abstract/Free Full Text].
2.
Knight GB, Agnello V, Bonagura V, Barnes JL, Panka DJ, Zhang QX.
Human rheumatoid factor cross-idiotypes, IV: studies on WA XID-positive IgM without rheumatoid factor activity provide evidence that the WA XId is not unique to rheumatoid factors and is distinct from the 17.109 and G6 XIds.
J Exp Med.
1993;178:1903-1911[Abstract/Free Full Text].
3.
Børretzen M, Randen I, Natvig JB, Thompson KM.
Structural restriction of the heavy chain CDR3 of human rheumatoid factors.
J Immunol.
1995;155:3630-3637[Abstract].
4.
Agnello V, Chung RT, Kaplan LM.
Evidence of a role for hepatitis C virus infection in pathogenesis of type II mixed cryoglobulinemia.
N Engl J Med.
1992;327:1490-1495[Abstract].
Response:
Relationship between IgR expressed by hepatitis C
virus-associated non-Hodgkin lymphomas and rheumatoid factors
The primary aim of our studyl was to demonstrate
that a large proportion of hepatitis C virus (HCV)-associated
non-Hodgkin lymphomas (NHLs) derive from B-cell clones chronically
stimulated by a common antigen. This is deduced on the basis of
peculiar properties that these NHLs express: intraclonal diversity, an R/S mutation ratio in the FR segments of IgR lower than expected by
chance, and a highly restricted use of gene segments in assembling IgR.
We believe that our data support the proposed pathogenetic model adequately. Concerning the derivation of the NHLs from B-cell clones that produce
rheumatoid factors (RFs), this is only supposed in our study
(the title reads "Sequence analysis of the immunoglobulin antigen
receptor of hepatitis C virus-associated non-Hodgkin lymphomas suggests that the malignant cells are derived from the
rheumatoid factor producing cells that occur mainly in type II
cryoglobulinemia") on the basis of the significant homologies found
between gene segments (both VH and VK) used by
the NHLs in assembling IgR and gene segments used by antibodies with RF
activity. It is worth noting that 6 of 9 patients with a previous
history of open mixed cryoglobulinemia (MC) syndrome, but none of the 8 patients without MC, specifically used the D21/9 segment gene, a region
also frequently used by the WA cross-idiotype of RF found in MC. This
finding supports a possible correlation between NHLs and B-cell clones producing RFs. Moreover, we previously demonstrated that premalignant and malignant lymphoproliferations in an HCV-infected type II MC
patient were sequential phases of a unique antigen-driven pathologic process.2 This finding again supports the possibility that the NHL originated from a B-cell clone that was already
present and antigenically stimulated at the time of MC onset. We agree that a formal demonstration of the identity of HCV-associated
NHLs with B-cell clone(s) that produce RFs could be derived only by a
direct comparison of the amino acid (AA)- deduced sequence of
the IgR expressed by the lymphomatous clone with the AA sequence of the
specific RFs present in the sera of each single patient. But this was
beyond the scope of our study. Nevertheless, it is our intent to prove
this point by comparing the AA-deduced Ig sequence obtained from serum
of some of these patients with their NHL IgR sequence. Concerning the suggested implication of HCV as the pathogenetic agent
of the HCV-associated NHLs, this was derived from the analysis of
sequence homologies between the IgR expressed by the NHLs and that of
antibodies specific for the E2 protein of HCV. In our opinion, such a
hypothesis is interesting in the light of the almost complete
association and supposed pathogenetic relationship between HCV
infection and MC syndrome, which precedes NHL onset, at least in a
large group of NHL patients. Although such a hypothesis needs a formal
demonstration, it is worth noting that Chan et al3 have
recently reported that HCV-associated NHLs and normal B cells
responding to E2 viral antigen preferentially use the VHI-69
gene, which is also used by some NHLs we analyzed
(VHI-69 is synonymous with VHl/DPIO, and
VHl/DP-IO and VHl/DP-88 genes differ in only 1 nucleotide) and is typically used by RF-WAs present in MC. Thus these
data indicate that some RF-WAs may have an anti-HCV specificity.
Moreover, in HCV infection the reactivity of IgM with the corresponding
IgG is inhibited by the addition of HCV antigens, suggesting that the
antigen-binding site of the IgM is cross-reactive with HCV
antigens.4 Furthermore, IgG-IgM WA immune complexes were
found in HCV-infected patients but not in acute and chronic hepatitis B
and acute hepatitis A infections. Thus IgG-IgM WA immune complexes
appear to be uniquely associated with HCV infection, supporting the
possibility that they derive from an antigen-driven response closely
related to the virus.4 Finally, since antibody specificity is primarily dependent on the
CDR3 region, which is the most variable part of the V region, we have
limited the search for sequence homologies to this part of the IgR
region.1 But the results of database research using the
entire AA-deduced V sequence again confirms significant homologies with
some RFs in most of the cases. Concerning the IgR sequence reported for
patient 13, we agree that it may not be similar to that of RF-Bor (the
E value was high). In contrast, the homology reported for patient 14 with RF is valid, RF-MR20 being a human rheumatoid factor.
Mauro Boiocchi, Valli De Re, Daniela Gasparotto, and Salvatore De
Vita
Correspondence: Mauro Boiocchi, Experimental Oncology 1, Cnetro
di Riferimento Oncologico, Via Pedemontana Occidentale 12, Aviano, PN
33081, Italy
References
1.
De Re V, De Vita S, Marzotto A, et al.
Sequence analysis of the immunoglobulin antigen receptor of hepatitis C virus-associated non-Hodgkin lymphomas suggests that the malignant cells are derived from the rheumatoid factor-producing cells that occur mainly in type II cryoglobulinemia.
Blood.
2000;96:3578-3584.
2.
De Re V, De Vita S, Marzotto A, et al.
Pre-malignant and malignant lymphoproliferations in an HCV-infected type II mixed cryoglobulinemic patient are sequential phases of an antigen-driven pathological process.
Int J Cancer.
2000;87:211-216[CrossRef][Medline]
[Order article via Infotrieve].
3.
Chan CH, Hadlock KG, Foung SK, et al.
VHI-69 gene is preferentially used by hepatitis C virus-associated B cell lymphomas and by normal B cells responding to the E2 viral antigen.
Blood.
2001;97:1023-1026[Abstract/Free Full Text].
4.
Sansonno D, Iacobelli AR, Comacchiulo V, et al.
Immunochemical and biomolecular studies of circulating immune complexes isolated from patients with acute and chronic hepatitis C virus infection.
Eur J Clin Invest.
1996;26:465-475[CrossRef][Medline]
[Order article via Infotrieve].
 CiteULike  Connotea  Del.icio.us  Digg  Reddit  Technorati What's this?
Related Article in Blood Online:
-
Sequence analysis of the immunoglobulin antigen receptor of hepatitis C virus-associated non-Hodgkin lymphomas suggests that the malignant cells are derived from the rheumatoid factor-producing cells that occur mainly in type II cryoglobulinemia
- Valli De Re, Salvatore De Vita, Alessandra Marzotto, Maurizio Rupolo, Annunziata Gloghini, Barbara Pivetta, Daniela Gasparotto, Antonino Carbone, and Mauro Boiocchi
Blood 2000 96: 3578-3584.
[Abstract]
[Full Text]
[PDF]
This article has been cited by other articles:
|
|
|
|
 
C. Mazzaro, D'A. Little, G. Pozzato, L. A. Emens, M. S. Sulkowski, V. Agnello, C. Mecucci, M. Casato, F. Lefrere, X. Troussard, et al.
Regression of Splenic Lymphoma after Treatment of Hepatitis C Virus Infection
N. Engl. J. Med.,
December 26, 2002;
347(26):
2168 - 2170.
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. Casato, C. Mecucci, V. Agnello, M. Fiorilli, G. B. Knight, C. Matteucci, L. Gao, and J. Kay
Regression of lymphoproliferative disorder after treatment for hepatitis C virus infection in a patient with partial trisomy 3, Bcl-2 overexpression, and type II cryoglobulinemia
Blood,
March 15, 2002;
99(6):
2259 - 2261.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
