|
|
|
|||||||
|
|||||||||
From the
Monoclonal human light chains, i.e. Bence Jones
proteins, and their recombinant variable fragments (V
Bence Jones proteins are monoclonal antibody light chains found
in the urine of approximately 60% of multiple myeloma
patients(1, 2) . The mechanisms underlying the
pathophysiology of these tumors are not understood fully. Light chain
aggregation has been advanced as one of the causes of tissue damage,
particularly involving the kidney(3) . The antibodies produced
by tumor cells in multiple myeloma are described to bind certain
antigens, including determinants found in carbohydrates (4) and
autoantigens expressed by neuronal cells(5, 6) . These
findings have raised the possibility that antigen recognition by the
tumor cell antibody products may contribute to the pathophysiology of
multiple myeloma.
Some antibodies possess catalytic activities
similar to those of
enzymes(7, 8, 9, 10) . Antibody light
chain subunits free of heavy chains can bind
antigen(11, 12) . The catalytic activity of one of these
antibodies has been localized to an active site in the light chain
subunit(13, 14) . Bence Jones proteins accumulate to
millimolar levels in multiple myeloma patients(15, 16) ,
at which even low level catalytic activity may be biologically
important. Here, we describe evidence for peptidase activity
attributable to the variable domain in a majority of light chains
isolated from myeloma patients. This observation indicates that
catalysis by antibody subunits is not a rare phenomenon and suggests a
possible link between antibody catalysis and the pathophysiology of
multiple myeloma.
Cleavage of 10
additional MCA-conjugates of varying lengths and containing uncharged,
acidic, or aromatic residues at the scissile bond by two light chains
was measured (). The activity of one of these light chains
(Bence Jones protein B6) was comparatively indiscriminate, in that
there was detectable cleavage at Ala-MCA, Phe-MCA, Met-MCA, Arg-MCA,
and Lys-MCA bonds. Even single amino acid-MCA conjugates were cleaved
by this light chain, but with increasing chain length, the activity
acquired preference for Arg- and Lys-containing bonds. The second light
chain (Bence Jones protein RHY) displayed essentially no hydrolysis of
substrates containing MCA linked to uncharged, acidic, or aromatic
amino acids. These observations indicate preferential cleavage at basic
residue-MCA bonds by the light chains and a variable influence of
N-terminal flanking residues in catalysis by the two light chains.
The results of our studies show that certain human light
chains have the capability to cleave synthetic protease substrates and
natural peptides. These light chains were judged to be pure on the
basis of SDS-polyacrylamide gel electrophoresis in the presence of a
reducing agent. The gel filtration experiments furnish direct proof
that the light chain itself is responsible for the observed activity
since noncovalent complexation of adventitious proteins with the light
chains would be precluded in the strong denaturant employed for
chromatography. The expression of VIP-hydrolyzing activity in the 12.5
kDa peak of a recombinant V
The presence of basic
residues in the peptide substrates appears to favor expression of light
chain catalytic activity. However, the sequence requirements are not
uniformly stringent, and one light chain even hydrolyzed single amino
acid-MCA conjugates. Recognition of the MCA conjugates by the light
chains does not appear analogous, therefore, to typical high affinity
binding interactions between antibodies and polypeptide antigens, which
can involve extensive contacts at >15 residues(24) . This
conclusion is supported by observations of high micromolar K
Catalysis has been assumed to be a rare property
of antibody subunits. Our finding that 17 of 24 light chains and
V
Light chains dimers serve as models for
the antibody combining site (30, 31). In the instance of one light
chain studied in detail (B6), the monomer form displayed substantially
greater hydrolytic activity than the dimer. Thus, formation of an
antibody-like site does not appear to be a prerequisite for
catalytically productive substrate contact. This conclusion is relevant
to potential intracellular expression of catalytic activity by light
chains, wherein a reducing environment is likely to favor the monomer
form. The extracellular concentrations of light chains in patients with
multiple myeloma can reach millimolar levels(15, 16) .
Many light chains are capable of cleavage of small protease substrates
containing basic residues, and some are capable of cleavage of the
larger polypeptide, VIP. These results provide a rationale for
examination of light chain-catalyzed peptide breakdown as a
pathophysiological mechanism in patients with multiple myeloma. The
example of VIP-hydrolysis may be directly relevant to tumor cell
activity in vivo because this peptide has been shown to
regulate antibody and interleukin synthesis by
lymphocytes(32, 33, 34) , and myeloma cells
express receptors for VIP(28) .
Values are
Values are
K
We thank Robert Dannenbring for technical support Drs.
T. T. Wu and M. Schiffer for discussion, H. Kolmar for the generous
gift of the REI plasmid, Y.-L. Deng and D. Hanson for the rREI
construct, and X. Jiang and R. Raffen for preparation of recombinant
proteins.
Volume 270,
Number 25,
Issue of June 23, pp. 15257-15261, 1995
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
Fragment (*)
ABSTRACT
INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
) were
screened for proteolytic activity using peptide-methylcoumarinamide
(peptide-MCA) conjugates and vasoactive intestinal polypeptide (VIP) as
substrates. Sixteen of 21 Bence Jones proteins and one of three V fragments were capable of detectable cleavage of one or more
substrates. The magnitude and kinetic characteristics of the activity
varied with different substrates. Among the peptide-MCA substrates, the
presence of tripeptide or tetrapeptide moieties with a basic residue at
the scissile bond generally favored expression of the activity. The
influence of N-terminal flanking residue recognition was evident from
differing values of K and k
(turnover number) observed using different
Arg-containing peptide-MCA substrates. Different light chains displayed
different kinetic parameters for the same substrate, suggesting unique
catalytic sites. Hydrolysis of VIP was characterized by nanomolar
Michaelis-Menten constants (K
),
suggesting comparatively high affinity recognition of this peptide. The
25-kDa monomer and the 50-kDa dimer forms of one light chain
preparation were resolved by gel filtration in 6 M guanidine
hydrochloride. Following renaturation, the monomer displayed 51-fold
greater peptide-MCA-hydrolyzing activity than the dimer. A renatured
V
domain prepared by gel filtration in 6 M guanidine hydrochloride displayed VIP-hydrolyzing activity in the
12.5-kDa peak fractions. These results provide evidence for the
proteolytic activity of certain human light chains and imply that this
phenomenon may have a pathophysiological significance.
Bence Jones Proteins and V
Monoclonal Fragments
and light chains were
purified from the urine of 21 myeloma patients with multiple myeloma as
described previously(17) . Recombinant V
domains
(rREC, rLEN, rREI) were expressed in Escherichia coli from
synthetic genes and purified as in Ref. 18. SDS-polyacrylamide gel
electrophoreses in the absence or presence of 20 mM 2-mercaptoethanol was on Phast gradient gels (8-25%;
Pharmacia Biotech Inc.). Each of the 21 light chain preparations
contained a 50-kDa dimer band and a 25-kDa monomer band (Fig. 1).
The disulfide bond in the dimer was reduced in mercaptoethanol,
producing a single 25-kDa monomer band. A single 12.5 kDa band was
observed in the two V preparations. Gel filtration of light
chain and V was in 6 M guanidine hydrochloride, pH
6.5 (Sigma) on a Superose-12 fast protein liquid chromatography column
(Pharmacia). The column fractions were renatured by dialysis against 50
mM Tris-HCl, 100 mM glycine, 0.025% Tween 20, 0.02%
sodium azide, pH 7.7, for 2 days with four buffer changes using a Life
Technologies, Inc. multiwell dialysis device (final guanidine
concentration < 1 nM, assuming equilibration across the
dialysis membrane).
Figure 1:
Silver-stained
SDS-polyacrylamide gels run in nonreducing (A) and reducing (B) conditions. Lanes1 and 2,
recombinant V fragments rRec and rLEN, respectively. Lanes3 and 4, Bence Jones proteins B6 and
RHY, respectively.
Catalysis Assays
Light chains were incubated with
peptide-MCA
()conjugates (Peptides International
or Sigma) in 50 µl of 50 mM Tris-HCl, 100 mM glycine, 0.025% Tween 20, pH 7.7, in 96-well plates (MicroFluor W,
Dynatech) at 37 °C in a humidified incubator. Hydrolysis of
peptide-MCA substrates was determined as the fluorescence of the
aminomethylcoumarin leaving group ( 460 nm,
370 nm) using a plate reader (Perkin Elmer LS50
fluorimeter). Product concentrations were computed by comparison of the
fluorescence yield with aminomethylcoumarin (Peptides International)
measured in identical volumes (18.3 fluorescence
units/µM/50 µl). Background fluorescence measured in
wells containing the substrate in diluent was generally less than 20
fluorescence units and was subtracted from values observed in the
presence of catalyst. [Tyr
-
I]VIP
prepared as in (19) was incubated with light chains at 37 °C
in 200 µl of 50 mM Tris-HCl, 100 mM glycine,
0.025% Tween 20, pH 7.7, for 6 h, and peptide hydrolysis was estimated
by measuring the radioactivity soluble in 10% trichloroacetic acid.
Estimates of peptide breakdown by this method are essentially identical
to those obtained by reversed-phase high performance liquid
chromatography separation of intact and degraded fragments(20) .
Trypsin was from bovine pancreas (3
crystallized, 3080 units/mg;
U. S. Biochemical Corp). Initial rate data at varying substrate
concentration were fitted using nonlinear regression analysis to the
Michaelis-Menten-Henri equation (v = (V
[S])/(K + [S])) (Enzfitter, Elsevier-Biosoft). Since
the catalyst concentration in some assays was in the range of the low
end of VIP substrate concentrations analyzed, kinetic constants were
also computed by a graphing routine as described in Ref. 21, which
eliminates errors due to substrate depletion effects. The K
and k
values
estimated by this method were essentially identical (<10%
difference) to the kinetic constants reported (see I).
VIP and Peptide-MCA Cleavage
Twenty-one myeloma
light chains and three V domains were initially screened
for proteolytic activity using radiodiodinated VIP as substrate, a
28-amino acid neuropeptide previously shown to serve as a target for
catalytic antibodies(7, 14) . Four light chains and one
recombinant V preparation hydrolyzed radiolabeled VIP (). Further screening was performed using a panel of five
peptide-MCA conjugates. Cleavage of the amide bond linking an amino
acid with the terminal coumarin moiety in these conjugates serves as a
surrogate for peptide bond hydrolysis, permitting rapid measurement of
protease activity(22) . Arg-MCA- and Lys-MCA-containing
substrates were employed in the screening assays because recombinant
anti-VIP light chains, thyroglobulin-specific autoantibodies, and
polyclonal IgG preparations have previously been shown to display
pronounced preference for cleavage at bonds containing basic amino
acids(10, 14) . A majority of the light chains (76%) and
one V preparation displayed detectable hydrolytic activity
with one or more peptide-MCA substrates (). Wide variations
were evident in the levels of hydrolysis of different substrates by
individual light chains and of individual substrates by different light
chains. The greatest frequency of hydrolysis was observed with
Boc-Glu-Ala-Arg-MCA as substrate. Four light chain preparations
displayed detectable hydrolysis of all five peptide-MCA substrates, and
two displayed hydrolysis of four substrates.
Kinetics
The initial rate data obtained at
increasing substrate concentrations were fitted to the Michaelis-Menten
equation (Fig. 2). The K for light
chain-catalyzed hydrolysis for the peptide-MCA substrates ranged from
15-290 µM (I), values in the same range
as those observed for conventional proteases. In contrast to the
peptide-MCA substrates, VIP hydrolysis by both catalyst preparations
examined was notable for low K
values,
suggesting comparatively high affinity recognition of this substrate.
In comparison, the K
for
trypsin-catalyzed VIP hydrolysis, estimated by methods identical to
those employed here, is approximately 3 orders of magnitude
greater(14) . Because of the low K
values, the kinetic efficiencies (k
/K
) of the
VIP-hydrolyzing light chains approach those of conventional proteases.
Apparent turnover numbers of the light chain (k
)
are in the same range or higher than of acyltransferase antibodies
raised by immunization with presumed transition state analogs (for
review, see Ref. 23). The hydrolysis of Boc-Ile-Glu-Gly-Arg-MCA and
Boc-Glu-Ala-Arg-MCA by the same light chain (B6) was characterized by
kinetic parameters that varied over 1 order of magnitude, confirming
the conclusion that residues flanking the scissile bond (P
,
P
, P
) play a role in substrate recognition.
Figure 2:
Initial rate data for VIP hydrolysis by
rREC V fragment (A) and Boc-Ile-Glu-Gly-Arg-MCA
hydrolysis by Bence Jones protein B6 (B) fitted to the
Michaelis-Menten equation. Catalyst concentrations: B6, 125
nM; rREC, 25 nM. Values are means of closely agreeing
duplicate determinations. Reaction times were 20 h and 6 h for
Boc-Ile-Glu-Gly-Arg-MCA and VIP,
respectively.
Mapping of Activity by Gel
Filtration
Chromatography experiments were performed in a strong
denaturant (6 M guanidine hydrochloride) to preclude
noncovalent association of adventitious protease contaminants with
light chains, and to dissociate noncovalent light chain dimers and
higher order aggregates formed at neutral pH in nondenaturing solvents
(3). The VIP-hydrolyzing specific activities of fractions spanning the
12.5 kDa optical density peak of a V preparation (Fig. 3A) were essentially identical (fractions
46-48, 3.6 10
, 3.8 10
,
and 3.8 10
cpm hydrolyzed/µg of protein/6 h),
suggesting catalyst homogeneity. Gel filtration of B6 light chain
yielded two optical density peaks corresponding to the disulfide-linked
dimer and monomer components (Fig. 3B). Most of the
peptide-MCA activity in the renatured column fractions was associated
with the monomer peak. Rechromatography of the monomer and dimer
yielded peptide-MCA-hydrolyzing activity peaks that tracked exactly
with the optical density peaks (Fig. 3C). The specific
activities of fractions spanning the width of the dimer and monomer
peaks were essentially constant (upperpanel,
fractions 21-23: 29.4, 31.8, and 27.5 units
F/µg of protein/20 h, respectively; lowerpanel, fractions 24-26: 1.4 10
,
1.5 10
and 1.5 10
units
F/µg of protein/20 h, respectively).
Figure 3:
Gel filtration of light chain and V fragment catalytic activity. Chromatography was on a Superose-12
gel filtration column in denaturant (6 M guanidine
hydrochloride). Effluent fractions were renatured by removal of the
guanidine hydrochloride by dialysis. Insets are silver-stained
SDS-polyacrylamide gels of the peak optical density fraction (lane1) and marker proteins (lane2). A, rREC V fragment (flow rate, 0.25
ml/min; fraction size, 0.25 ml); B, Bence Jones protein B6
(flow rate, 0.25 ml/min; fraction size, 0.25 ml); C,
rechromatography of B6 dimer (upperprofile;
retention time 37 min in B) and B6 monomer (lowerprofile; retention time 43 min in B),
flow rate 0.5 ml/min, fraction size, 0.5 ml.
fragment shows that the
activity is attributable to the variable domain. This conclusion is
supported by observations that many but not all light chains possess
the activity, and that different light chains show varying activity
levels and distinct specificity profiles.
values for the peptide-MCA hydrolysis
reaction. K
estimates, obtained as in
Ref. 21, are close to the K
values,
suggesting poor binding to the ground state of the substrate. Light
chain and V
fragment catalyzed hydrolyses of VIP, on the
other hand, was characterized by lower K values, indicating comparatively high affinity recognition
of the VIP ground state. VIP binding autoantibodies are found in
healthy individuals (25) and patients with asthma(26) .
The apparent specialization of certain catalytic light chains for
binding and hydrolysis of an autologous peptide like VIP is consistent
with recognition of other autoantigens by myeloma
proteins(4, 5, 6) . The second feature
distinguishing the peptide-MCA substrates and VIP is the frequency with
which light chain hydrolytic activity was encountered. Seventeen
proteins hydrolyzed the peptide-MCA substrates, whereas only five
displayed VIP-hydrolyzing activity. This may reflect differences in
structural requirements for cleavage of small versus large
peptide substrates, as has been described for trypsin(27) . For
instance, the greater flexibility of small peptides may permit
productive contact with the catalytic site, whereas comparatively rigid
conformations found in larger molecules may make potential scissile
bond inaccessible.
fragments tested possess detectable proteolytic activity
suggests that this may not be an uncommon phenomenon. There is no
association evident between the observed catalytic activity and light
chain isotype ( or ) or V
subgroup (). Taken together, the high frequency of catalysis and
apparent low affinity recognition of structurally different peptide-MCA
substrates suggest the presence of conserved structural components of
the catalytic site in different light chains. Alternatively, these
results may be explained by a tendency toward de novo development of catalytic activity during sequence diversification
of light chain variable regions. The identities of amino acids
responsible for the activity and the contributions of germ line
inheritance, V-J gene rearrangement, and somatic mutation mechanisms in
catalysis by different light chains are not yet known. A catalytic
triad of Ser, His
, and Asp
residues (Kabat numbering) in spatial arrangement similar to that
of the active site of subtilisin has been identified in an anti-VIP
light chain by molecular modeling(14) , and site-directed
mutagenesis experiments support the role of Ser and
His
as essential catalytic residues.
()Erhan and Greller (29) have previously noted a
significant sequence similarity between the first complementarity
determining region of certain Bence Jones proteins and the active site
region in serine proteases.
Table: Peptidase activity of Bence Jones
protein and recombinant Vfragment
F/20 h for peptide-MCA substrates and percent hydrolysis/6 h for I-VIP. ND denotes not detectable (
F for peptide-MCA
substrates < 11, percent VIP hydrolysis < 10). Light chain
isotype and subgroup is given in parentheses. Peptide-MCA and VIP
substrate concentrations, 0.2 mM and 0.5 nM,
respectively. Catalyst concentrations in peptide-MCA and VIP hydrolysis
assays, 2 µM and 0.25 µM, respectively. rREI,
rLEN, rREC are recombinant V fragments. The remaining
proteins are light chains purified from urine. Amino acids are
identified by one-letter codes.
Table: Cleavage of various peptide-methylcoumarinamide
substrates by two Bence Jones proteins, B6 and RHY
F/20 h from a representative experiment performed at 200
µM substrate and 2 µM light chain
concentrations.
Table: Kinetic constants for catalysis by Bence Jones
proteins
and k
values were computed by fitting initial rate data at increasing
substrate concentrations (peptide-MCA substrates, 10-3000
µM; VIP, 10-1000 nM) to the
Michaelis-Menten equation. Catalyst concentrations: B6, 125
nM; LAY, 125 nM and 21 nM using Boc-EAR-MCA
and VIP substrates, respectively; rREC, 25 nM. S.E. values
were <15% in all cases.
, variable domain of light chain; Boc, t-butyloxycarbonyl.
©1995 by The American Society for Biochemistry and Molecular Biology, Inc.
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  B. Wootla, S. Dasgupta, J. D. Dimitrov, J. Bayry, H. Levesque, J.-Y. Borg, A. Borel-Derlon, D. N. Rao, A. Friboulet, S. V. Kaveri, et al. Factor VIII Hydrolysis Mediated by Anti-Factor VIII Autoantibodies in Acquired Hemophilia J. Immunol., June 1, 2008; 180(11): 7714 - 7720. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y.-R. Kim, J.-S. Kim, S.-H. Lee, W.-R. Lee, J.-N. Sohn, Y.-C. Chung, H.-K. Shim, S.-C. Lee, M.-H. Kwon, and Y.-S. Kim Heavy and Light Chain Variable Single Domains of an Anti-DNA Binding Antibody Hydrolyze Both Double- and Single-stranded DNAs without Sequence Specificity J. Biol. Chem., June 2, 2006; 281(22): 15287 - 15295. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Lacroix-Desmazes, J. Bayry, S. V. Kaveri, D. Hayon-Sonsino, N. Thorenoor, J. Charpentier, C.-E. Luyt, J.-P. Mira, V. Nagaraja, M. D. Kazatchkine, et al. From the Cover: High levels of catalytic antibodies correlate with favorable outcome in sepsis PNAS, March 15, 2005; 102(11): 4109 - 4113. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Planque, Y. Bangale, X.-T. Song, S. Karle, H. Taguchi, B. Poindexter, R. Bick, A. Edmundson, Y. Nishiyama, and S. Paul Ontogeny of Proteolytic Immunity: IgM SERINE PROTEASES J. Biol. Chem., April 2, 2004; 279(14): 14024 - 14032. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. DEBAT, B. AVALLE, O. CHOSE, C.-O. SARDE, A. FRIBOULET, and D. THOMAS Overpassing an aberrant V{kappa} gene to sequence an anti-idiotypic abzyme with {beta}-lactamase-like activity that could have a linkage with autoimmune diseases FASEB J, March 1, 2001; 15(3): 815 - 822. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Paul, A. Tramontano, G. Gololobov, Y.-X. Zhou, H. Taguchi, S. Karle, Y. Nishiyama, S. Planque, and S. George Phosphonate Ester Probes for Proteolytic Antibodies J. Biol. Chem., July 20, 2001; 276(30): 28314 - 28320. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| All ASBMB Journals | Molecular and Cellular Proteomics |
| Journal of Lipid Research | ASBMB Today |