Download Production and Immunodiagnostic Applications of Antihuman Light

HEMATOPATHOLOGY
Original Article
Production and Immunodiagnostic
Applications of Antihuman Light Chain
Monoclonal Antibodies
MASAHIRO ABE, M D , * f TETSUYA GOTO, M D , * f STEPHEN J. KENNEL, P H D , *
DENNIS WOLFENBARGER, MT,* SALLIE D. MACY, MT,* DEBORAH T. WEISS, BS,*
AND ALAN SOLOMON, MD*,§
cell and surface Ig of B-lymphocyte populations, respectively. These
antibodies facilitated the immunohistochemical detection and characterization of light-chain-associatcd amyloid (AL amyloid) and other types
of light-chain-related tissue deposits. Furthermore, the unti-C, -specific
MoAbs were used to measure serum and urinary Igx and IgX concentrations. Quantification of Bence Jones protein excretion, even in the presence of other urinary proteins, was possible using the highly sensitive
anti-C„ and anti-Cx MoAbs reactive only with free light chains. The
ability to identify and characterize, through the use of these antihuman
light chain MoAbs, light-chain-related epitopes at the protein, cellular,
and tissue level has clinical importance in the diagnosis and treatment
of patients with monoclonal plasma cell and related B-cell immunoproliferative diseases.(Key words: Monoclonal antibodies; Light chains;
Bence Jones proteins; Multiple myeloma; AL amyloidosis; Plasma
cells; B cells) Am J Clin Pathol 1993; 100:67-74.
Hybridomas producing antihuman light chain monoclonal antibodies
(MoAbs) were derived from fusion of SP2/0 mouse myeloma cells with
splenic lymphocytes from mice repeatedly immunized with purified Kand X-type Bence Jones proteins representative of the major V, (V,„
V.M, V„m, V„IV) and Vx (VM, Vxll/V, Vxll„ Vxlv, Vxv,) subgroups or gene
families. Monoclonal antibodies were obtained that had specificity for
constant-region (C,) determinants common to all K or X light chains (C,
and Cx, respectively) as well as for variable-region (V,) epitopes unique
to each of the V, or Vx subgroups. The capability of these reagents to
recognize C, and V, determinants on monoclonal immunoglobulin (Ig)
molecules was demonstrated in fluid-phase antigen-capturing enzymelinked immunosorbent assay (ELISA), solid-phase ELISA, and immunoblotting. In addition, these antilight chain MoAbs were used to establish immunocytochemically the K or X type and V, -subgroup nature
of light chains expressed by the cytoplasmic Ig of monoclonal plasma
resentative of the major V,. and Vx gene families.15 The results
of these serologic analyses have been confirmed from those
obtained from amino acid and nucleotide sequence data, and
reveal that the classification of light chains has considerable
biologic and clinical importance. Our studies with these polyclonal antisera and those of other investigators who have prepared antilight chain monoclonal antibodies (MoAbs) have
demonstrated that certain disease processes are associated with
particular VL subgroups or epitopes.6"9 The usefulness of these
reagents for immunodiagnosis, however, has been precluded
by their limited availability.
To address this limitation and as part of a longstanding effort
to elucidate the immunochemical properties of human light
chains, we have prepared a large panel of MoAbs that recognize
structurally distinct CL- and VL-associated light chain epitopes
expressed by secreted forms of monoclonal Ig molecules and by
monoclonal plasma cells and B lymphocytes. These reagents
From the * Human Immunology & Cancer Program. Department of have been used in serologic, immunocytochemical, and immuMedicine. University of Tennessee Medical Center/Graduate School of nohistochemical analyses to identify and characterize monocloMedicine. Knoxville. Tennessee, the tFirst Department of Medicine. nal Igs, cell populations, and light chain-related tissue deposits.
University of Tokushima. Tokushima. Japan, and the ^Biology Division, Oak Ridge National Laboratory, Oak Ridge. Tennessee.
MATERIALS A N D METHODS
Received April 7, 1992; accepted for publication May 22, 1992.
Proteins
Supported by Public Health Research Grant CA 10056 from the
National Cancer Institute and an Ina M. Barger Memorial grant for
Monoclonal urinary light chains and serum Igs, ie, K and X
cancer research (IM-430) from the American Cancer Society.
Bence Jones proteins, IgG and IgA myeloma proteins, and IgM
§ American Cancer Society (Clinical) Research Professor.
macroglobulins were obtained from patients with multiple myAddress reprint requests to Dr. Solomon: University of Tennessee
eloma, Waldenstrom's macroglobulinemia, chronic lymphoMedical Center, 1924 Alcoa Highway, Knoxville, Tennessee 37920.
Monoclonal immunoglobulin (Ig) light chains represent a
unique biomarker of malignant plasma cell and other lymphocyte-related immunoproliferative disorders, including lightchain-associated amyloidosis (AL amyloidosis) and have long
been implicated in many of the pathophysiologic manifestations of these diseases.1,2 The detection and characterization of
these components therefore have important diagnostic, therapeutic, and prognostic implications. We have employed serologic methods to classify light chains at the protein and cellular
levels using polyclonal antihuman light chain antisera prepared by immunizing rabbits with selected human monoclonal
light chains, ie, Bence Jones proteins. 34 Through the use of
these reagents, it has been possible to identify readily not only
the two constant-region (CL) light chain isotypes, K and X, but
also the variable-region (VL) subgroups and subsubgroups rep-
67
68
HEMATOPATHOLOGY
Original Article
cytic leukemia (CLL), or AL amyloidosis. The proteins were
isolated by zone (block) electrophoresis and purified by gel filtration as previously described.3 The purity of the proteins was
determined by electrophoresis in agarose and polyacrylamide
gels; the light chain type and heavy chain class were established
by immunofixation electrophoresis (Paragon® Electrophoresis
System, Beckman, Brea, CA) or by immunoelectrophoresis.
The light chain VL-subgroup nature of K and X monoclonal Igs
was determined serologically using rabbit anti-V, and anti-Vx
subgroup-specific antisera, respectively.3 Normal (polyclonal)
human light chains were prepared from pooled serum IgG3
(Cohn FRII, Sigma Chemical, St. Louis, MO), normal (polyclonal) IgG, IgA, and IgM proteins were obtained from Sigma
Chemical, and pooled normal human serum was obtained
from Chemicon (Temecula, CA).
Antibody
Preparation
Monoclonal antibodies were prepared using heat-precipitated Bence Jones proteins representative of the four V, (V,,,
V«„, V,M1> V,1V) and five Vx (Vx„ Vx]l/V, Vxl„, V xlv , V xvl ) subgroups.8 Five mice were immunized with each protein: 6- to
8-week-old pathogen-free, female Balb/c mice were injected
intraperitoneally (IP) every 2 to 4 weeks with 100 n% of the
immunogen emulsified in 0.2 mL of either complete Freund's
adjuvant or a mixture containing 50 fig of monophosphoryl
lipid A and 50 /ig of trehalose dicorynomycolate (RIBI ImmunoChem Research, Hamilton, MT). Ten days after injection, serum antilight chain antibody titers were determined in a
fluid-phase antigen-capturing enzyme-linked immunosorbent assay (ELISA) on specimens of blood obtained by retroorbital
bleeding. Three days before fusion, mice were injected intrasplenically with 100 j*g of protein suspended in 0.1 mL of phosphate-buffered saline. Hybridomas were generated from the
fusion of 1.6 X 108 immune spleen cells with 2.4 X 10s SP2/0
mouse myeloma cells in the presence of polyethylene glycol
according to the method of de St. Groth and co-workers.10
Hybridoma culture supernatants were screened for MoAb
production in fluid-phase antigen-capturing ELISA using biotinylated Bence Jones proteins. Based on these results, colonies
were selected, expanded, and recloned at least twice by limited
dilution. Hybridomas were grown in pristane-primed Balb/c
mice and the MoAbs isolated from ascitic fluid by ammonium
sulfate precipitation and ion-exchange chromatography." The
MoAb heavy chain isotype was identified by ELISA with subclass-specific rabbit-antimouse Igs (Mono AB-ID/EIA kit,
Zymed Laboratories, San Francisco, CA). The association constants of the MoAbs were determined from Scatchard plots
using nonlinear regression analyses of data obtained by immunoprecipitation of radioiodinated Bence Jones proteins.12
Enzyme-Linked
Immunosorbent
Assay
Procedures
To determine the mouse polyclonal antihuman light chain
antibody response and the antilight chain reactivity and specificity of the MoAbs, a fluid-phase antigen-capturing ELISA
was employed in which 96-well microtiter plates (Corning
Glass, Corning, NY) were coated with a polyclonal goat antimouse IgG antiserum (Sigma Chemical). After blocking and
washing, appropriately diluted samples of mouse serum, culture fluid supernatant, ascitic fluid, or purified MoAbs were
added to each well. Subsequent steps included the addition of a
biotinylated13 Bence Jones protein or intact Ig, an avidin-biotin
complex (Vectastain, Vector Laboratories, Burlingame, CA),
and a 2.2'-azino-bis[3-ethylbenzthiazoline-6 sulfonic acid]
substrate solution (Kirkegaard and Perry Laboratories, Gaithersburg, MD). Color development was terminated by the addition of 2% oxalic acid.
To select hybridoma clones for further propagation and to
test the reactivity of the MoAbs against a large panel of Bence
Jones proteins, a solid-phase ELISA was employed. Wells were
coated with a solution of monoclonal light chains, and after
blocking and washing, hybridoma culture supernatant (or appropriate dilutions of ascitic fluid or purified MoAb) was added
to each well. Detection of bound MoAb was accomplished using a peroxidase-labeled goat antimouse IgG antiserum
(BioRad, Richmond, CA) and the 2.2'-azino-bis[3-ethylbenzthiazoline-6 sulfonic acid] substrate.
A quantitative sandwich ELISA was used to measure the
concentration of bound or free light chains. Microtiter plates
were coated with purified mouse antihuman light chain
MoAbs, and, after blocking and washing, the wells were filled
with appropriately diluted serum or urine specimens or with
1.0- to 1000-ng/mL solutions of a reference standard consisting
of either (1) a monoclonal IgG* or IgGX protein or (2) an equimolar mixture of 20 different K or X Bence Jones proteins representative of each of the VL subgroups. After incubation and
washing, the wells were filled with peroxidase-labeled goat
F(ab')2 antihuman K or X light chain-specific antiserum (Tago,
Burlingame, CA). 2.2'-azino-bis[3-ethylbenzthiazoline-6 sulfonic acid] was used as a substrate.
For all three methods, color development was measured
after 15 minutes at 415 nm using an ELISA plate reader (BioTek). The reactivity was graded as: 0: no difference from background or +, ++, +++: >0.5, >0.9, and >1.3 optical density
units above background, respectively. In the quantitative
ELISA, the data were analyzed using the Kineti-Calc® software
program (BioTek Instruments, Winoski, VT).
Immunoblotting
Sodium dodecyl sulfite/polyacrylamide gel electrophoresis
(SDS/PAGE) was performed on Bence Jones proteins and
monoclonal IgGs using the Phast system (LKB-Pharmacia, Piscataway, NJ). The proteins were transferred by electroblotting
onto 0.45-/*m Immobilon membranes (Millipore, Bedford,
MA) and, after blocking, were exposed first to culture fluidcontaining supernatant or dilutions of purified mouse antihuman light chain MoAbs and then to an alkaline-phosphataselabeled horse antimouse IgG antiserum (ABC kit, Vector).
Bound protein was detected using the Alkaline Phosphatase
Substrate II kit (Vector).
Immunocytochemistry
and
Immunohistochemistry
Monoclonal B-cell populations were isolated using FicollHypaque gradient centrifugation from peripheral blood, bone
marrow, or lymph nodes. They were then either cytocentrifuged onto poly-L-lysine-coated slides or prepared for flow cytometry. Additional material for study included 4-fim sections
of fresh-frozen lymph nodes or of deparaffinized, formalinfixed tissue or bone marrow. Immunoperoxidase or alkaline
phosphatase techniques used to examine the light chain nature
of cytoplasmic or surface Igs (clg or slg, respectively) of cells
contained in the cytocentrifuge or tissue specimens were similar to those described previously.4
For flow cytometry, cells were incubated with appropriately
diluted MoAbs, after which they were exposed first to a fluores-
A.J.C.P.-July 1993
69
ABE ET AL.
MoAbs
Antihuman L
cein isothiocynate- (FITC) conjugated F(ab')2 goat antimouse
IgG antiserum absorbed with human Ig (Tago) and then to a
PE-conjugated F(ab')2 goat antihuman K or X chain antiserum
(Tago). After washing, the stained cells were examined in an
Epics Profile flow cytometer (Coulter Laboratories, Hialeah,
FL). Backgroundfluorescencewas determined by staining cells
with (1) an unrelated mouse MoAb, (2) the FITC-conjugated
antiserum, and (3) PE-conjugated F(ab')2 prepared from nonimmune goat IgG.
RESULTS
Generation and Characterization of Antihuman Light
Chain Monoclonal Antibodies
Initial attempts to generate antihuman light chain MoAbs
using Freund's adjuvant met with limited success in that the
resultant antibodies were of low affinity (Ka, <106 M"') and
recognized primarily V-region antigenic determinants unique
to the protein used for immunization, ie, antiidiotype specificity. In contrast, the use of the RIBI adjuvant resulted in MoAbs
of broader specificity and higher affinity (Ka, >106 M_1) for
light-chain-associated determinants. The maximum antibody
response (as determined from measurement of the mouse immune response to Bence Jones proteins of the same VL subgroup as the injected human protein) was usually achieved
after a 4- to 8-month period of immunization. Splenic lymphocytes were harvested only from animals that exhibited the highest serum titer of polyclonal antihuman light chain antibody.
Five mice were injected with each protein; a significant immune response developed in at least two. Typically, each fusion resulted in ~200 antibody-producing hybridomas, and,
based on results obtained by ELISA, selected clones were recloned and expanded for in vivo passage.
The specificity of the antihuman light chain MoAbs was
tested influid-and solid-phase ELISA against a reference panel
of >70 K- and X-type Bence Jones proteins, and >20 monoclonal IgG, IgA, and IgM proteins as well as polyclonal light
chains and intact Igs. Two groups of MoAbs were identified:
the first recognized a CL-related epitope associated with K or X
light chains (C, or Q, respectively) and reacted with free light
chains as well as those associated with intact Ig proteins, ie,
bound light chains. Other anti-CL antibodies were obtained
that reacted only with free monoclonal or polyclonal light
chains but not with intact monoclonal or polyclonal Ig molecules (Table 1). These reagents could detect in quantitative
ELISA light chain concentrations as low as 1.0 ng/ml.
The second group of MoAbs recognized antigenic determinants present on the V region of K or X light chains (V, or Vx).
This group included those with antiidiotype specificity as well
as antibodies that detected V, or Vx antigenic determinants
unique to each of the major VL subgroups of K or X chains,
respectively. Based on the results of ELISA, we selected from
these reagents two panels of antibodies having optimum reactivity with each of the VL subgroups. The first consisted of
anti-V, MoAbs that were specific for either KI, KII, KIII, or KIV
light chains; the second included anti-Vx MoAbs that reacted
exclusively with XI, XII/V, XIV, or XVI proteins. In the case of
t n e vxm subgroup, no single anti-VXII1 MoAb could detect all
XIII proteins. However, antibodies specific for each of the three
VMII subsubgroups14—VXIIIa, VMIIb, and VMllc—were identified
and mixed to create an anti-VMU-specific reagent (Table 2).
The anti-VL subgroup-specific MoAbs were used to determine by ELISA the V,- or Vx-subgroup nature of Bence Jones
proteins and light chains associated with monoclonal IgG, IgA,
and IgM proteins. The anti-Vx subgroup-specific MoAbs
reacted well in both fluid- and solid-phase ELISA, as did the
anti-V„ reagents, with the exception of anti-V,,, which had optimum reactivity in solid-phase ELISA.
All of the mouse antihuman light chain MoAbs were of the
IgG class. Their heavy chain isotypes and association
constants, which ranged from 0.3 to 80 X 107 M"\ are given in
Table 3.
Application of Antihuman Light Chain MoAbs
The capability of the antihuman light chain MoAbs to recognize CL and VL determinants associated with serum or urinary
Ig molecules, clgs or slgs of plasma cell and B-cell populations,
and light-chain-associated tissue deposits was tested using
various immunologic methods.
TABLE 1. REACTIVITY OF ANTIHUMAN LIGHT CHAIN MONOCLONAL ANTIBODIES
Immunoglobulin
Antitotal K
(14-6E4)*
Antifree K
(LKC8)
+++t
+++
Antitotal X
(21-3F4)
Antifree X
(18-9G11)
Monoclonal
BJPK
BJPX
Ig&c
IgGX
IgA/c
IgAX
IgM/c
IgMX
Polyclonal
Light chains x/X
Ig&c/X
IgAx/X
Iglvk/X
0
0
0
+++
0
+++
0
+++
0
0
0
0
0
0
0
0
+++
+++
+++
+++
+++
+++
+++
+++
+++
+++
+++
0
0
0
* Clone designation.
t Reactivity in fluid-phase ELISA.
Vol. 100-No. I
0
+++
0
+++
0
+++
0
0
0
0
0
0
+++
0
0
0
70
HEMATOPATHOLOGY
Original Article
TABLE 2. SPECIFICITY OF MOUSE ANTIHUMAN V«- AND Vx-SUBGROUP-SPECIFIC MONOCLONAL ANTIBODIES
Anti-Vu
(5-8C)*
Subgroup
Anti-VK„
(20-15H2)
Anti-VK„,
(27-4A6)
Anti-VKlv
(11-1F4)
0
++
0
0
0
0
++
0
0
0
0
+++
+t
V.,(10)t
V.„ (8)
V.„, (9)
V>1V (6)
0
0
0
Anti-Vv
(28-20E12)
Anti-Vwv
(23-7F7)
Anti-Vm,
(22- I5D11, 29-3D12,
24-8Dl2)§
Anti-V)jV
(21-18F9)
Anti-VM,
(19-6D6)
+++
0
0
0
0
0
+++
0
0
0
0
0
+++
0
0
0
0
0
+++
0
0
0
0
0
+++
VM(10)
Vx,„v (6)
Vx... 0 4 ) 1
Vx,v (2)
Vxv, (6)
* Clone designations.
t Number of monoclonal lighl chains teslcd.
$ Reactivity influid-phaseEL1SA.
§ Mixture of anti-Xllla. -lllb. and -lllc monoclonal antibodies.
|| Includes 5. 3. and 6 proteins of the Xllla. All lb. and Mile subsubgroups. respectively.
For protein analysis, the K or X type and the VL subgroup of
Bence Jones proteins and light chains of monoclonal Igs subjected to SDS/PAGE could be established in immunoblotting
analyses using the antilight chain MoAbs. The results of one
such study are illustrated in Figure 1, in which the KIV or XII/V
nature of the K or X Bence Jones protein, respectively, was evidenced.
The anti-C L reagents were used to quantitate by sandwich
EL1SA serum or urinary concentrations of bound or free light
chains, respectively. Analyses using the antitotal-K or -X MoAbs
revealed that in pooled normal human serum the level of Ig*
was 10.0 mg/mL, and that of IgX was 3.9 mg/mL. The concentration of polyclonal free-K and -X chains in random urine specimens (obtained from six normal individuals) was measured
using the antifree-K or -X MoAbs, and averaged 1.7 Mg/mL and
TABLE 3. CHARACTERIZATION OF MOUSE ANTIHUMAN
LIGHT CHAIN MONOCLONAL ANTIBODIES
Specificity
Total K
Free K
Total X
Free X
v.,
v.„
v.„,
V.,v
Vx,
Vxu/v
V
v
XlUn
VxiIIb
Vxltlc
Vxtv
Vxv
Clone
Designation
14-6E4
LKC8
21-3F4
18-9011
5-8C
20-15H2
27-4A6
1I-1F4
28-20E12
23-7F7
22-15D11
29-3D12
24-8D12
21-18F9
19-6D6
ND = not determined.
Heavy Chain
Isotype
Association
Constant
(XIO1 mol/L)
IgGl
IgGl
IgG2a
IgGl
lgG3
IgGl
IgGl
IgGl
IgGl
IgG2a
IgG2b
IgG2a
IgGl
IgG2a
IgGl
2.2
3.1
33.0
0.6
1.0
ND
ND
ND
ND
80.0
7.0
ND
1.1
10.0
0.3
0.7 jug/mL, respectively. The antifree light chain reagents also
were used to quantitate Bence Jones protein excretion, and
were particularly helpful in cases where the presence of this
component was obscured by other urinary proteins, eg, in patients with AL amyloidosis and nephrotic syndrome. The results of analyses quantitating urinary K or X Bence Jones proteins before and after chemotherapy in two such patients are
illustrated in Figure 2.
At the cellular level, the antitotal-K and -X chain antibodies
were tested against cell suspensions prepared from the bone
marrow of patients with multiple myeloma or AL amyloidosis
and from the peripheral blood, bone marrow, or lymph nodes
of patients with CLL and related B-cell neoplasms. Among
more than 100 cases studied, the monoclonal Ig immunophenotype of the plasma cell or lymphocyte population was appar-
>*^£
.' I *
*
*> # v* -^
anti-x
anti-X
anti-«IV
anti-A.ll/V
FIG. 1. Immunoblotting (Western blotting) analyses, (left) The lanes
containing K\, KII, KIII, and KIV Bence Jones proteins are as indicated.
The upper and lower membranes were immunostained with an anti-C,
MoAb and an anti-V„IV subgroup-specific MoAb, respectively, (right)
The lanes containing XI, XII/V, XIII, XIV, and XVI Bence Jones proteins are as indicated. The upper and lower membranes were immunostained with an anti-Cx MoAb and an anti-VXI|/v subgroup-specific
MoAb, respectively.
A.J.C.P.-July 1993
ABE ET AL.
Antihuman
lations was more evident; eg, the cells from patient RW reacted
with the anti-X and anti-V MI/v MoAbs but not with the other
antibodies in the panel (Fig. 3). For more than 30 patients
whose serum or urine contained a monoclonal Ig protein, there
was complete concordance between the light chain classification determined immunocytochemically and that found
through our serologic and sequence analyses of the isolated
protein. In addition to studies of fresh-frozen cell suspensions
or tissue, these antibodies could be used to immunostain formalin-fixed, paraffin-embedded tissue (Fig. 4).
Analyses by flow cytometry involved usage of both the antic s and anti-V L subgroup-specific MoAbs as primary reagents
for examination of peripheral blood cells or isolated nodal lymphocytes derived from patients with CLL and other forms of
B-cell neoplasia. The immunocytochemical demonstration of
the slg/cl+ or slg/clll+ nature of cells from two such patients is
shown in Figure 5.
The antilight chain MoAbs also were used to demonstrate
immunohistochemically the deposition of light chains as renal
tubular casts, basement membrane precipitates, or amyloid fibrils. Due to the variation among patients with AL amyloidosis
in the composition of the light chains found in the amyloid
deposits (ie, the predominant occurrence of VL fragments versus the presence of components containing the VL plus a portion of the C L domain 15 ), the immunohistochemical detection
of AL amyloid often required the use of the anti-V L subgroupspecific antibodies (Fig. 6). That the AL amyloid deposits in
patient VT consisted only of a VMIIa fragment while those in
patient IR contained both VMIIa plus Cx structure was confirmed through our sequence analyses of the purified amyloid
extracts.
free K*
0.03
0.08
0.23
0.17
0.18
0.14
0.11
4-14-91
BJP
IgG
FB
e
©
££
hssJL
9-4-90
I «-Sp
10.3
0.01
11-5-90
9 '•^Wp
12.5
0.11
12.5
0.18
18.0
0.05
12.8
0.03
23.0
0.03
15.1
0.02
MM
1 iH
! it
1-3-91
3-5-91
6-4-91
till
1
7-18-91
8-18-91
71
Lightchain MoAbs
DISCUSSION
BJP
* nig/ml
FIG. 2. Immunoquantification of Bence Jones proteinuria in patients
with nephrotic syndrome. Agarose gel electropherograms of lyophilized, reconstituted (50 mg/mL) serial urine specimens from AL amyloidosis patients RM and FB, whose urine contained a K Bence Jones
protein plus a monoclonal IgG* protein and a \ Bence Jones protein,
respectively. In both cases, the first specimen was obtained before administration of chemotherapy. Protein concentrations were measured
by a sulfosalicylic acid (SA) turbidity method and by quantitative
ELISA employing MoAbs specific for free K or X light chains as indicated. The location of the IgG and Bence Jones proteins (BJP) were
determined by immunofixation electrophoresis.
ent by the predominant K or X reactivity of the clg or slg of these
cells. When the cells were immunostained with anti-V^ or antiVx subgroup-specific MoAbs, the monoclonality of such popu-
A variety of polyclonal antihuman light chain antibodies
with specificities for CL- and VL-associated epitopes have been
generated for clinical and research purposes; however, the relative scarcity of these reagents, especially those directed to VL
antigenic determinants, as well as the broader specificity and
technical problems associated with polyclonal antisera, have
restricted their use. The ability to produce antilight chain
MoAbs of the desired specificity in virtually unlimited
amounts would provide an invaluable source of material for
diagnostic, prognostic, and possible therapeutic purposes.
Previously, few MoAbs with light-chain specificity have been
obtained.16"20 As part of a systematic effort to generate a battery
of these reagents, we developed an immunization protocol in
which a series of mice were injected with well-characterized Kand X-type Bence Jones proteins representative of the major VL
subgroups, V„„ V,„, V, m , V,IV and VM, VM1/V, VMU, VXIV, VXVI.
The availability of large numbers of these serologically and
chemically classified light chains, and of monoclonal IgG, IgA,
and IgM proteins made it possible to determine by ELISA the
appropriate clones for selection and propagation and to establish the specificity of the resultant antibodies for light-chain-associated C L and VL epitopes.
The antihuman light chain MoAbs generated included those
reacting with C L determinants—anti-C, or anti-Q—found on
free or bound light chains, and others that recognized Q or Cx
determinants on free light chains exclusively. Additionally, a
second group of antilight chain MoAbs were produced that
reacted with VL-associated idiotype and subgroup-associated
epitopes. Through extensive screening and clonal selection,
Vol. 100-No. 1
72
HEMATOPATHOLOGY
Original Article
'
^U Iff"
IW
#*
'hMm+fmm
OE
V
RW
HB
BR
GR
m
antl-i
antl J.I
anti-xii/v
antlxiv
antkw
MYELOMA
CH
LYMPHOMA
JR
MW
anti-t
anti-A
A.J.C.P.-July 1993
AS
73
ABE ET AL.
Antihuman Lightchain MoAbs
FIG. 3. Immunocytochemical characterization of the Vx subgroups expressed by the light chains of clg and slg in monoclonal plasma cell and
B-lymphocyte populations, respectively. Plasma cells derived from the
bone marrow of patients with multiple myeloma (RW, HB) and AL
amyloidosis (GR) and peripheral blood lymphocytes from patients
with CLL (DE, BR). Cells were examined using the avidin-biotin complex alkaline phosphatase technique and the antihuman light chain
MoAbs as indicated. Based on the patterns of reactivity, the light chains
of cells from patients DE, RW, HB, BR, and GR were classified as XI,
XI1/V, Mil, XIV, and XVI, respectively (hematoxylin counterstain;
X 1,000).
1
CR
HF
L
F
"? L
L
F
1 *
3
V
40P
4
-D>Nfc£bia
LFLZ
Anti-n
FIG. 4. Immunophenotyping analyses of formalin-fixed, paraffin-embedded tissue. (Left) Bone marrow specimens from patients CH and JR
with clgx+ and dgX+ multiple myeloma, respectively, and (right)
lymph node specimens from patients MW and AS with slg*+ and sIgX+
low-grade lymphoma, respectively (immunoperoxidase method, primary antibodies—anti-x and anti-X MoAbs, as indicated; X 1,000).
panels of antibodies were obtained that had specificity for the
four V, and five Vx subgroups.
We have demonstrated, in several types of immunologic
analyses (ELISA, immunoblotting, immunocytochemistry,
and immunohistochemistry), the capability of the antilight
chain MoAbs to recognize the K or X type and the VL-subgroup
nature of monoclonal Ig proteins, plasma cells, and B lymphocytes and have demonstrated the clinical applicability of these
reagents. The anti-C, and anti-Q Moabs were used not only to
establish protein and cell monoclonality but also to quantitate
serum and urine Ig concentrations. Due to the obscuring presence of non-Ig proteins in nephrotic urine, it has been virtually
impossible to measure the amount of Bence Jones protein in
2
HF
BG
BG
A#
LFL2
LFL2
Anti-x
Anli>«
FIG. 5. Flow cytometric analyses. Mononuclear cells isolated from the
peripheral blood of two patients (HF and BG) with slgic* CLL were first
stained with anti-V„, or anti-V,,,, subgroup-specific MoAbs and then
with the FITC- or PE-conjugated anti-lg antibodies. Data are expressed.
as bivariate dot plots with logfluorescenceintensity of FITC (surface
V„) on the Y axis, and logfluorescenceintensity of PE (surface *•) on the
X axis. The vertical and horizontal lines define the boundaries of positive and negative cells based on control staining with FITC and PE.
antl-x
anti-Mlla
FIG. 6. Immunohistochemical characterization of AL amyloid deposits. Use of anti-CL (anti-X) and -V, (anti-Xllla) MoAbs to characterize the light
chain nature of the green birefringent Congo red (CR)-stained amyloid deposits in the spleen and heart of patients VT and IR, respectively. Based on
the patterns of reactivity, the amyloid in patient VT consisted of a Vx fragment, while that of IR contained Q- as well as Vx-related structure
(immunoperoxidase technique, primary antisera as indicated; patient VT. X400: patient IR, X250).
Vol. 100-No. I
74
HEMATOPATHOLOGY
Original Article
such specimens using electrophoretic or other serologic techniques. The availability of anti-CL antibodies with specificity
forfree K or X light chains has made such analyses feasible; these
data are especially important in documentation of response to
treatment or relapse of patients with multiple myeloma or AL
amyloidosis.' Because these reagents can also be used to quantitate free polyclonal (as well as monoclonal) urinary K and X
light chains, it is possible to assess polyclonal B-cell activation
in autoimmune and other diseases, such as multiple sclerosis,21"23 and to monitor kidney function in patients with other
disorders associated with renal tubular dysfunction and polyclonal light chain proteinuria, eg, systemic lupus erythematosus and diabetes mellitus.24"27
The anti-VL subgroup-specific MoAbs also have clinical relevance. The ability to determine immunocytochemically the V,
or Vx subgroup is of diagnostic and prognostic importance. Not
only can monoclonality be proven even in cases in which cell
numbers are minimal, but the ability to identify residual monoclonal plasma cell or B-cell populations provides an objective
means to evaluate response to treatment or to predict disease
relapse. In addition, these reagents (as well as those with CL
specificity) can be used to identify and characterize light chainrelated tissue deposits, such as those found in myeloma (cast)
nephropathy, light chain deposition disease, and especially AL
amyloidosis.2 Finally, because certain VL subgroups are associated with particular disease processes, including autoimmune
disorders, AL amyloidosis, and B-cell neoplasia,6"9 it is important to establish the VL nature of monoclonal proteins or cells.
In addition to their diagnostic and prognostic usefulness, the
antihuman light chain MoAbs may also have therapeutic potential.28,29 These antibodies could serve as immunoadsorbants
to remove extracorporeally nephrotoxic or amyloidogenic light
chains, or to purge neoplastic B cells from bone marrow specimens used for autologous transplantation.30 Such applications
would be of obvious clinical benefit, and are currently under
investigation.
Acknowledgments. The authors thank Mary deBram Hart for assistance in the technical aspects of these studies and Julie Ottinger for
assistance in manuscript preparation.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
REFERENCES
1. Solomon A. Monoclonal immunoglobulins as biomarkers of
cancer. In: Sell S, ed. Cancer Markers. Clifton, NJ: Humana
Press, 1980, pp 57-87.
2. Solomon A. Clinical implications of monoclonal light chains. Semin Oncol 1986;13:341-349.
3. Solomon A. Light chains of human immunoglobulins. Methods
Enzymol 1985;116:101-121.
4. Solomon A, Weiss DT, Macy SD, Antonucci RA. Immunocytochemical detection of kappa and lambda light chain V region
subgroups in human B-cell malignancies. Am J Pathol
1990;137:855-862.
5. Solomon A, Weiss DT. Serologically defined V region subgroups
of human Might chains. J Immunol 1987;139:824-830.
6. Solomon A, Frangione B, Franklin EC. Bence Jones proteins and
light chains of immunoglobulins. Preferential association of the
V xv , subgroup of human light chains with amyloidosis AL(A). J
Clin Invest 1982;70:453-460.
7. Ledford DK, Gofii F, Pizzolato M, et al. Preferential association of
/clllb light chains with monoclonal human IgMx autoantibodies. J Immunol 1983;131:1322-1325.
8. Solomon A. Light chains of immunoglobulins: Structural-genetic
correlates. Blood 1986;68:603-610.
9. KippsTJ. The CD5 B cell. Adv Immunol 1989;47:117-185.
10. Fazekas de St Groth S. Scheidegger D. Production of monoclonal
24.
25.
26.
27.
28.
29.
30.
antibodies: Strategy and tactics. J Immunol Methods
1980;35:1-21.
Kennel SJ, Foote U , Lankford PK. Analysis of surface proteins of
mouse lung carcinomas using monoclonal antibodies. Cancer
Res 1981;41:3465-3470.
Kennel SJ, Chen JP, Lankford RK, Foote LJ. Monoclonal antibodies from rats immunized with fragment D of human fibrinogen.
Thrombosis Res 1981;22:309-320.
Klein-Schneegans A-S, Kuntz L, Fonteneau P, Loor F. An indirect
asymmetrical sandwich ELISA using anti-allotype antibodies
for the specific and quantitative measurement of mouse IgG2a
of Igh-l b allotype. J Immunol Methods 1989;125:207-213.
Eulitz M, Murphy C, Weiss DT, Solomon A. Serological and chemical differentiation of human XIII light chain variable regions. J
Immunol 1991;146:3091-3096.
Glenner GG. Amyloid deposits and amyloidosis: The ^-fibrilloses.
N Engl J Med 1980;302:1283-1292, 1333-1343.
Lowe J, Hardie D, Jefferis R, et al. Properties of monoclonal antibodies to human immunoglobulin kappa and lambda chains.
Immunology 1981;42:649-659.
Ling NR, Lowe J, Hardie D, et al. Detection of free K chains in
human serum and urine using pairs of monoclonal antibodies
reacting with C. epitopes not available on whole immunoglobulins. Clin Exp Immunol 1983;52:234-240.
Greenstein JL, Solomon A, Abraham GN. Monoclonal antibodies
reactive with idiotypic and variable-region specific determinants on human immunoglobulins. Immunology 1984;51:17—
25.
Walker MR, Solomon A, Ling NR, et al. Immunogenic and antigenic epitopes of immunoglobulins. XVIII. Subpopulations of
human lambda chains defined with a panel of monoclonal antibodies. Immunology 1986;59:467-471.
Walker MR, Solomon A, Weiss DT, et al. Immunogenic and antigenic epitopes of IgE: XXV. Monoclonal antibodies that differentiate the Mcg+/Mcg" and Oz+/Oz" C region isotypes of human
XL chains. J Immunol 1988;140:1600-1604.
Hopper JE, O'Brien J, Papagiannes E. Restriction of blood and
marrow CLL-B cells to free L-chain Ig secretion: Implication for
normal B-cell function and control. Am J Hematol
1988;29:125-133.
Hopper JE, Sequeira W, Martellotto J, et al. Clinical relapse in
systemic lupus erythematosus: Correlation with antecedent elevation of urinary free light-chain immunoglobulin. J Clin Immunol 1989;9:338-350.
Mehta PD, Cook SD, Troiano RA, Coyle PK. Increased free light
chains in the urine from patients with multiple sclerosis. Neurology 1991;41:540-544.
Spriggs B, Epstein WV. Clinical and laboratory correlates of L
chain proteinuria in systemic lupus erythematosus. J Rheumatol 1974;1:287-292.
Soiling K. Polymeric forms of free light chains in serum from normal individuals and from patients with renal diseases. Scand J
Clin Lab Invest 1976;36:447-452.
Soiling K, Soiling J, RomerFK. Free light chains of immunoglobulins in serum from patients with rheumatoid arthritis, sarcoidosis, chronic infections and pulmonary cancer. Acta Med Scand
1981;209:473-477.
Groop L, Makipernaa A, Stenman S, et al. Urinary excretion of
kappa light chains in patients with diabetes mellitus. Kidney Int
1990;37:1120-1125.
Tutt AL, Stevenson FK, Smith JL, Stevenson GT. Antibodies
against urinary light chain idiotypes as agents for detection and
destruction of human neoplastic B lymphocytes. J Immunol
1983;131:3058-3063.
Waldmann TA. Monoclonal antibodies in diagnosis and therapy.
Science 1991;252:1657-1662.
Gribben JG, Freedman AS, Neuberg D, et al. Immunologic purging of marrow assessed by PCR before autologous bone marrow
transplantation for B-cell lymphoma. N Engl J Med
1991;325:1525-1533.
AJ.C.P.-July 1993