The n e w e ng l a n d j o u r na l of m e dic i n e

Brief Report

Clonal Immunoglobulin against Lysolipids

in the Origin of Myeloma
Shiny Nair, Ph.D., Andrew R. Branagan, M.D., Jun Liu, Ph.D.,
Chandra Sekhar Boddupalli, Ph.D., Pramod K. Mistry, M.D.,
and Madhav V. Dhodapkar, M.B., B.S.​​

Sum m a r y

Antigen-driven selection has been implicated in the pathogenesis of monoclonal gam-

mopathies. Patients with Gaucher’s disease have an increased risk of monoclonal gam-
mopathies. Here we show that the clonal immunoglobulin in patients with Gaucher’s
disease and in mouse models of Gaucher’s disease–associated gammopathy is reactive
against lyso-glucosylceramide (LGL1), which is markedly elevated in these patients and
mice. Clonal immunoglobulin in 33% of sporadic human monoclonal gammopathies
is also specific for the lysolipids LGL1 and lysophosphatidylcholine (LPC). Substrate
reduction ameliorates Gaucher’s disease–associated gammopathy in mice. Thus, long-
term immune activation by lysolipids may underlie both Gaucher’s disease–associated
gammopathies and some sporadic monoclonal gammopathies.

M
ultiple myeloma and monoclonal gammopathy of undeter- From the Department of Medicine, Sec-
mined significance (MGUS) are characterized by clonal expansion of trans- tion of Hematology (S.N., A.R.B., C.S.B.,
M.V.D.), Section of Digestive Diseases
formed plasma cells.1 Analyses of immunoglobulin genes in tumor cells have (J.L., P.K.M.), Yale Liver Center (P.K.M.),
provided evidence of antigen-driven selection, with restricted heavy-chain variable- and Yale Cancer Center (M.V.D.), Yale
region use and highly hypermutated immunoglobulin heavy- and light-chain genes.2-6 University School of Medicine, New Ha-
ven, CT. Address reprint requests to Dr.
However, the antigens underlying the origins of most MGUS and myeloma clones Dhodapkar at the Department of Medi-
remain unknown. Hyperphosphorylated modification of stomatin (EPB72)-like 2 cine and Immunobiology, Yale Universi-
protein (STOML2, which is identical to paratarg-7) due to the inactivation of pro- ty, 333 Cedar St., New Haven, CT 06520,
or at ­madhav​.­dhodapkar@​­yale​.­edu.
tein phosphatase 2A was identified as a target of certain paraproteins and an in-
herited risk factor for the development of gammopathies.7-9 A recent study identi- Drs. Mistry and Dhodapkar contributed
equally to this article.
fied sumoylated heat-shock protein 90 as another inherited risk factor for plasma-cell
N Engl J Med 2016;374:555-61.
dyscrasia.10 However, there remains a need to identify the antigenic origins of
DOI: 10.1056/NEJMoa1508808
MGUS and myeloma that may be amenable to targeted prevention. Copyright © 2016 Massachusetts Medical Society.
Lipids (such as pristane) were implicated in the earliest models of murine plas-
macytoma,11 and lipid disorders such as Gaucher’s disease and obesity are associ-
ated with an increased risk of myeloma.12,13 The risk of myeloma is markedly higher
among patients with Gaucher’s disease, in whom myeloma is now emerging as a
leading cause of cancer-related death, than in the general population.13 Glucocer-
ebrosidase deficiency in Gaucher’s disease leads to increases in the level of LGL1.14
Recently, we identified a subset of human and murine LGL1-specific CD1d-restrict-
ed type 2 natural killer T cells that constitutively expressed markers of follicular
helper T cells and helped in the differentiation of lipid-reactive plasma cells.15 In
an earlier study, we found an elevation of type 2 natural killer T cells against an-
other bioactive lysolipid, LPC, in myeloma.16 These studies led us to test whether
the clonal immunoglobulin in Gaucher’s disease–associated myeloma and spo-
radic myeloma was reactive against LGL1 and LPC.

n engl j med 374;6 nejm.org  February 11, 2016 555

The New England Journal of Medicine
Downloaded from nejm.org on September 30, 2017. For personal use only. No other uses without permission.
Copyright © 2016 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

A SPEP Analysis with LGL1 Blots in Patients with B SPEP Analysis with LGL1 Blots in GBA1–/– Mice
Gaucher’s Disease and Controls
Patients with
Gaucher’s Disease Controls
With With
monoclonal polyclonal GBA1−/− Mice Control Mice
gammopathy gammopathy

SPEP SPEP
Immuno- Immuno-
globulin globulin
M spikes M spikes

LGL1 LGL1
Blot Blot

C Germinal-Center B Cells in Spleen

Wild-Type Mice GBA1−/− Mice P=0.04
2.5
PBS PBS LGL1
2.0
105 P=0.02
0.24% 0.68% 2.18%

Cells (%)
1.5
104
GL7

1.0
103

102 0.5

0.0
102 103 104 105 102 103 104 105 102 103 104 105 PBS PBS LGL1
Fas Fas Fas Wild GBA1−/−
Type

D CD38+CD138+ Plasma Cells in Bone Marrow Mononuclear Cells E Anti-LGL1 Antibodies

GBA1−/− Mice
PBS LGL1 P=0.01
0.4 P=0.02 1.0
105 0.10% 0.32%
Anti-LGL1 Antibody

0.8
0.3
104
Cells (%)

0.6
CD138

0.2
103 0.4
0.1
102 0.2

0.0 0.0
102 103 104 105 102 103 104 105 PBS LGL1 PBS LGL1
CD38 CD38 GBA1−/− GBA1−/−

Me thods pliance with the guidelines of the institutional

Patients and Mice
Peripheral-blood or bone marrow samples were
obtained from patients with MGUS or myeloma
and Gaucher’s disease and from healthy blood
donors. Written informed consent was obtained
from all the participants. The study was approved
by the institutional review board at Yale University.
The generation of glucocerebrosidase-deficient
(GBA1−/−) mice has been described previously.17
All the mice were bred and maintained in com-
animal care committee at Yale University.
Lipids
LGL1 (with purity assessed at >98% by thin-layer
chromatography) was purchased from Matreya and
was stored frozen (at −20°C) at a concentration of
500 μg per milliliter of 50% dimethyl sulfoxide in
distilled water as storage stock.15 LPC was pur-
chased from Avanti Polar Lipids, dissolved in chlo-
roform at a concentration of 10 mg per milliliter,
and stored at −20°C as storage stock.16 Diacylglyc-

556 n engl j med 374;6 nejm.org  February 11, 2016

The New England Journal of Medicine

Downloaded from nejm.org on September 30, 2017. For personal use only. No other uses without permission.
Copyright © 2016 Massachusetts Medical Society. All rights reserved.
 Brief Report
Figure 1 (facing page). Lipid Reactivity of Immuno­
globulin in Patients and Mice with Gaucher’s Disease
and in Controls.
Serum specimens obtained from patients with Gauch-
er’s disease–associated monoclonal gammopathy or
polyclonal gammopathy and from healthy controls
were analyzed by means of serum protein electropho-
resis (SPEP) (Panel A). The bracket shows the immu-
noglobulin component. In parallel, SPEP gel was blot-
ted onto polyvinylidene fluoride membrane that was
preincubated with lyso-glucosylceramide (LGL1) to
identify LGL1-reactive immunoglobulins. SPEP was
performed on serum specimens obtained from gluco-
cerebrosidase-deficient (GBA1−/−) mice and from con-
trol mice that were matched for age and sex (Panel B).
LGL1-specific blotting was performed simultaneously
on serum specimens obtained from the GBA1−/− mice
and the control mice. The M spikes in the serum spec-
imens obtained from the GBA1−/− mice were reactive
against LGL1. Representative contour plots show the
percentage of FAS+GL7+ germinal-center B cells
among total B cells (CD19+B220+) in splenocytes ob-
tained from wild-type mice or GBA1−/− mice 7 days af-
ter three weekly injections with phosphate-buffered
saline (PBS) or LGL1 (Panel C). The values above the
insets indicate the percentage of cells in that particu-
lar gate. The bar graph shows the summary of the per-
centage of germinal-center B cells in splenocytes from
wild-type mice and GBA1−/− mice 7 days after injection
with PBS or LGL1. Data are means (from three mice);
T bars indicate standard errors. A representative fluo-
rescence-activated cell sorting plot (Panel D) shows
the expression of CD38+CD138+ plasma cells in
CD19−CD45lo bone marrow mononuclear cells (the
gating strategy is shown in Fig. S1 in the Supplemen-
tary Appendix) obtained from GBA1−/− mice 7 days af-
ter PBS or LGL1 injection. Cumulative data show the
percentage of CD38+CD138+ plasma cells. Data are
means (from three mice); T bars indicate standard er-
rors. The bar graph (Panel E) shows the presence of
anti-LGL1 antibodies, with an optical density of 450 nm,
in serum specimens obtained from GBA1−/− mice that
had been injected with PBS or LGL1. Data are means;
T bars indicate standard errors.
erol (DAG; at a concentration of 1 mg per milliliter
of chloroform), cardiolipin (at a concentration of
25 mg per milliliter of chloroform), and lipid A (at
a concentration of 1 mg per milliliter of dimethyl
sulfoxide) were all purchased from Avanti Polar
Lipids and were stored at −20°C as storage stock.
Antigen-Specific Immunoblotting
Polyvinylidene fluoride membranes that were satu-
rated with LGL1 and LPC (BioRad) were prepared
as described previously.18 Briefly, filters were incu-
bated in 100 μg per milliliter of LGL1 and LPC
in 0.5 M sodium bicarbonate, rinsed in phosphate-
buffered saline (PBS) and 0.05% Tween 20 deter-
n engl j med 374;6 nejm.org  February 11, 2016
The New England Journal of Medicine
Downloaded from nejm.org on September 30, 2017. For personal use only. No other uses without permission.
Copyright © 2016 Massachusetts Medical Society. All rights reserved.
Table 1. Immunoglobulin Reactivity against Lysolipids.*
Mouse Model or Human Cohort Lysolipid Reactivity
no./total no. (%)
Mice
GBA1−/− mice 6/6 (100)
Control mice 0/10
Humans
Healthy donors 0/25†
Patients with polyclonal gammopathy without 0/10
Gaucher’s disease
Patients with Gaucher’s disease
Normal immunoglobulins 0/5
Monoclonal gammopathy 7/9 (78)
Polyclonal gammopathy 10/11 (91)
Patients with sporadic monoclonal gammopathy
MGUS or asymptomatic myeloma 5/16 (31)†
Myeloma 17/50 (34)†
* Data show the number of mice or patients with positive test results among
the total number of mice or patients tested. Control mice were C57BL/6 and
were matched for age and sex with glucocerebrosidase-deficient (GBA1−/−)
mice. MGUS denotes monoclonal gammopathy of undetermined significance.
† Lipid reactivity was determined against both lyso-glucosylceramide (LGL1)
and lysophosphatidylcholine (LPC).
gent, and blocked with 1% bovine serum albumin
(BSA). Gels for serum protein electrophoresis
were blotted onto lipid membranes with the use
of modified diffusion blotting.19 After blocking
with 1% BSA in PBS and Tween 20 detergent, the
membrane was incubated with appropriate horse-
radish peroxidase (HRP)–conjugated secondary
antibody and was washed and developed with
the use of SuperSignal West Pico chemilumines-
cent substrate (Thermo Scientific).
Enzyme-Linked Immunosorbent Assay
Diluted human plasma (1:250 dilution) was added
to plates coated with LGL1 (500 ng per well), and
the levels of kappa and lambda immunoglobulin
light chains were determined with the use of hu-
man kappa- and lambda-specific enzyme-linked
immunosorbent assay (ELISA) quantification kits
(Bethyl Laboratories).15 To measure hen-egg lyso-
zyme (HEL)–specific and lipid (LGL1, DAG, cardio-
lipin, and lipid A)–specific antibodies, plates (Nunc-
Immuno plate, Thermo Scientific) were coated with
an equimolar concentration of antigens overnight at
room temperature and then blocked with 1% BSA in
PBS and Tween 20 detergent for 2 hours at room
temperature. The test serum and purified immuno-
globulin sample were diluted in 1% BSA in PBS and
Tween 20 detergent and incubated overnight at
4°C. The antigen-specific antibody was detected
557
 The n e w e ng l a n d j o u r na l of m e dic i n e

A SPEP Analysis with LGL1 and Control Blots C Analyses of Purified Immunoglobulin
Lipid-Reactive Lipid-Nonreactive
Monoclonal Monoclonal
Gammopathy Gammopathy Control
Monoclonal Gammopathy Control
Sporadic
SPEP monoclonal Gaucher’s
Immuno- gammopathy Disease
globulin
SPEP
LGL1 Blot

LGL1
Control Blot

Cardiolipin
B LPC Blots
Lipid-Reactive Lipid-Nonreactive
Monoclonal Monoclonal Lipid A
Gammopathy Gammopathy Control

LPC Blot DAG

D Ligand-Dependent Enrichment of Clonal Immunoglobulin E Ligand-Dependent Depletion of Clonal Immunoglobulin

Patient 1 Patient 2 Patient 3 Patient 1 Patient 2 Patient 3
C S C S C S C S C S C S
Lipid-Reactive Lipid-Reactive
Monoclonal Gammopathy Monoclonal Gammopathy
1500 600
Units (×10−3)

Units (×10−3)

1000 400

500 200

0 0
1 2 3 4 5 6 1 2 3 4 5 6
Lane Lane

Patient 1 Patient 2 Patient 3 Patient 1 Patient 2 Patient 3

C S C S C S C S C S C S
Lipid-Nonreactive Lipid-Nonreactive
Monoclonal Gammopathy Monoclonal Gammopathy
1500 640
Units (×10−3)

Units (×10−3)

480
1000
320
500
160

0 0
1 2 3 4 5 6 1 2 3 4 5 6
Lane Lane

F Reduction in Anti-LGL1 Antibody, According to Diet, in Mice G Reduction in M-Spike Intensity, According to Diet, in Mice
P<0.001 P<0.001
100 100
Percent Reduction

Percent Reduction

80
60
50
40
20
0 0
Before After Before After Before After Before After
Control Diet Test Diet Control Diet Test Diet

558 n engl j med 374;6 nejm.org  February 11, 2016

The New England Journal of Medicine

Downloaded from nejm.org on September 30, 2017. For personal use only. No other uses without permission.
Copyright © 2016 Massachusetts Medical Society. All rights reserved.
 Brief Report

Figure 2 (facing page). Lipid Reactivity of Immuno­

was analyzed by means of serum protein electro-
globulin in Sporadic Gammopathy. phoresis. The bead pellet was washed four times
Panel A shows the results of SPEP analysis, LGL1-specific with the coimmunoprecipitation buffer. The anti-
blotting, and control blotting (with the use of human al- body bound to the beads was eluted by boiling the
bumin as the control antigen) on serum specimens ob- sample in sodium dodecyl sulfate–polyacrylamide
tained from patients with monoclonal gammopathy, gel electrophoresis sample buffer for 5 minutes.
showing lipid reactivity or no lipid reactivity, and from
healthy controls. Panel B shows the results of lysophos-
Samples were analyzed by means of Western
phatidylcholine (LPC)–specific blotting, which was per- blotting with antihuman IgG, IgA, and IgM heavy
formed on serum specimens obtained from the same chains and kappa and lambda light chains HRP
patients and healthy controls. Panel C shows the results antibodies. Signal intensities were densitometrically
of SPEP analysis, followed by LGL1, cardiolipin, lipid A, quantitated with the use of ImageJ software, ver-
and diacylglycerol (DAG)–specific blotting on purified
immunoglobulin obtained from patients with monoclo-
sion 1.41 (National Institutes of Health).
nal gammopathy without Gaucher’s disease (sporadic
monoclonal gammopathy), patients with Gaucher’s dis- Injection of GBA1−/− Mice
ease–associated monoclonal gammopathy, and healthy GBA1−/− mice were injected intraperitoneally with
controls. Panel D shows the ligand-dependent enrich- either PBS or LGL1 (at a dose of 200 μg per mouse
ment of clonal immunoglobulin. Serum specimens ob-
tained from three patients with lipid-reactive monoclonal
each week for 3 weeks) in a volume of 100 mm3.
gammopathy and three patients with lipid-nonreactive The mice were euthanized 7 days after the last in-
monoclonal gammopathy were incubated with control jection. The diluted serum specimen (1:250 dilu-
(C) or sphingosine (S)–coated sepharose beads; the tion) was used for the quantification of lipid-spe-
bead-binding fraction was analyzed for the presence of cific antibodies with the use of ELISA, as described
clonal immunoglobulin by Western blot. The bottom
panels show densitometric quantitation of bands per-
previously.15 HEL, which was purchased from
formed with the use of ImageJ software. Panel E shows Sigma, was mixed with an adjuvant to yield an
the ligand-dependent depletion of clonal immunoglobu- emulsion of 2 mg per milliliter, of which 50 mm3
lin. Serum specimens from three patients with lipid-reac- (100 μg per mouse) was injected intraperitone-
tive monoclonal gammopathy and three patients with ally in three GBA1−/− mice (8 to 12 weeks old); three
lipid-nonreactive monoclonal gammopathy were incubat-
ed with control or sphingosine-coated sepharose beads.
GBA1−/− mice received alum (Imject Alum Adjuvant,
The flow-through fraction was analyzed for M-spike de- Thermo Fisher Scientific) in PBS as a control.
pletion by means of SPEP. The bottom panels show den-
sitometric quantitation of bands performed with the use Substrate-Reduction Therapy
of ImageJ software. The bar graph in Panel F shows the
The drug eliglustat (Genz-112638, Genzyme), a
percent reduction in anti-LGL1 antibodies in serum spec-
glucocerebroside synthase inhibitor that prevents
imens obtained from GBA1−/− mice after treatment with
the overproduction of LGL1, was formulated in
eligulstat (test diet), as compared with mice that received
a standard rodent food (test diet) in a weight-to-
the control diet. Data are means; T bars indicate stan-
weight ratio of 0.075. Seven mice with Gaucher’s
dard errors. The bar graph in Panel G shows the percent
reduction in the M-spike intensity in serum specimens
disease received the test diet daily at a dose of
obtained from GBA1−/− mice after treatment with the test
150 mg per kilogram of body weight, and seven
diet, as compared with mice that received the control
mice with Gaucher’s disease that were matched
diet. Data are means; T bars indicate standard errors.
for age and sex were given a base diet (control
diet). The animals received the assigned diets for
by HRP-conjugated mouse and human immuno- 2 months and were then euthanized for tissue
globulins that had been developed with TMB chro- and blood collection.
mogen (3,3′,5,5′-tetramethylbenzidine; Invitrogen).
Statistical Analysis
Ligand-Mediated Enrichment and Depletion The cohorts of patients with Gaucher’s disease
of Clonal Immunoglobulin and sporadic monoclonal gammopathy (MGUS
Diluted human plasma (1:250 dilution) was incu- or myeloma) and of healthy donors were com-
bated with 50 mm3 of either control or sphingosine- pared with the use of nonparametric statistics,
coated beads (Echelon Biosciences). Plasma and and P values were calculated. Two-tailed P values
beads were mixed together by incubating the sus- at an alpha level of 0.05 were considered to indi-
pension under rotary agitation (on an orbital cate statistical significance. Fisher’s exact tests
shaker) at 4°C. After incubation, the tubes were and Mann–Whitney U tests were performed with
centrifuged, and the supernatant (flow-through) the use of Stata software (StataCorp).

n engl j med 374;6 nejm.org  February 11, 2016 559

The New England Journal of Medicine
Downloaded from nejm.org on September 30, 2017. For personal use only. No other uses without permission.
Copyright © 2016 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

R e sult s globulins did not react against cardiolipin, lipid A,

Clonal Immunoglobulin in Gaucher’s Disease
To evaluate whether the monoclonal immuno-
globulin in the context of Gaucher’s disease reacts
with LGL1, we analyzed the reactivity of clonal
immunoglobulin against LGL1 with an antigen-
specific immunoblot (Fig. 1A and 1B). Clonal im-
munoglobulin in 17 of 20 patients with Gaucher’s
disease and in 6 of 6 GBA1−/− mice (which faith-
fully recapitulate type 1 Gaucher’s disease in hu-
mans)17 with monoclonal gammopathy was spe-
cific for LGL1. LGL1 reactivity was also observed
for Gaucher’s disease–associated polyclonal gam-
mopathy (Fig. 1A). In contrast, only low-level or
background reactivity was detected in samples
obtained from healthy donors or control mice.
To show the capacity of LGL1 to mediate B-cell
and plasma-cell activation in Gaucher’s disease in
vivo, we hyperimmunized young GBA1−/− mice with
LGL1. These mice already have a mildly higher
level of Fas+GL7+ germinal-center B cells at base-
line than the level in control mice (Fig. 1C). How-
ever, the injection of LGL1 led to a further increase
in Fas+GL7+ splenic germinal-center B cells (Fig. 1C)
as well as an increase in bone marrow CD38+CD138+
plasma cells (Fig. 1D, and Fig. S1 in the Supplemen-
tary Appendix, available with the full text of this
article at NEJM.org). Injection of these mice with
LGL1 — but not injection with an unrelated antigen
(HEL) — led to the induction of anti-LGL1 antibod-
ies in vivo (Fig. 1E, and Fig. S2 in the Supplemen-
tary Appendix). These data show the capacity of
LGL1 to mediate B-cell activation in Gaucher’s dis-
ease and serve as an antigenic target in Gaucher’s
disease–associated monoclonal gammopathy.
Clonal Immunoglobulin in Sporadic
Gammopathies
The dysregulation of bioactive lipids and lipid-
reactive T cells has also been observed in patients
with sporadic myeloma.16,20 Therefore, we analyzed
whether clonal immunoglobulin in sporadic MGUS
and myeloma is also lipid-reactive. M spikes in
33% of the patients tested (22 of 66 patients)
were LGL1-reactive (Table 1 and Fig. 2A). LGL1-
reactive clonal immunoglobulins were cross-reac-
tive against LPC (Fig. 2B). In contrast, no reactivity
was observed in patients with polyclonal gam-
mopathy that was not associated with Gaucher’s
disease (Table 1, and Fig. S3 in the Supplemen-
tary Appendix).
Purified LGL1- and LPC-reactive clonal immuno- munoglobulin not only has direct implications for
or DAG in an antigen-specific immunoblot (Fig. 2C)
or an ELISA with similar levels of purified im-
munoglobulin (Fig. S4 in the Supplementary
Appendix). The lipid reactivity of clonal immu-
noglobulins was verified by means of several
approaches. Analysis of kappa and lambda light-
chain isotypes of anti-LGL1 antibodies revealed a
strong light-chain skew toward the involved clon-
al light chain only in patients with lipid-reactive
M spikes (Fig. S5 in the Supplementary Appendix).
LGL1-reactive immunoglobulins showed the
same heavy- and light-chain specificity as the
clonal immunoglobulin in immunoblotting (Fig.
S6 in the Supplementary Appendix). The incuba-
tion of serum specimens with sphingosine-coated
beads led to the specific enrichment of clonal
immunoglobulin in the bead-binding fraction in
patients with lipid-reactive M spikes, but not when
the clonal immunoglobulin was not lipid-reac-
tive (Fig. 2D). Similarly, the analysis of the bead
flow-through fraction showed a specific deple-
tion of M spikes only in patients with lipid-reactive
M spikes (Fig. 2E).
F(ab′)2 fragments that were isolated from
clonal immunoglobulins retained a capacity for
antigen recognition that was similar to that for
intact immunoglobulin in an ELISA and could
be enriched by ligand-conjugated beads (Fig. S7 in
the Supplementary Appendix), which indicated that
the binding of these clonal immunoglobulins to
ligands is F(ab′)2-mediated. The binding affinity
of LGL1-reactive clonal immunoglobulins to anti-
gen was estimated to be approximately 57×10−9 M,
which is comparable to that reported for other an-
tibodies (Fig. S8 in the Supplementary Appendix).
The cohort of patients with lipid-reactive gam-
mopathies had a higher proportion of kappa
light-chain, stage I myeloma and a lower propor-
tion of intermediate- or high-risk cytogenetic
factors than did patients who did not have lipid-
reactive clonal immunoglobulins (Table S1 in
the Supplementary Appendix). These data show
that the clonal immunoglobulin reacts with a
bioactive lysolipid in nearly one third of patients
with sporadic MGUS or myeloma, and the cohort
of patients with lipid-reactive gammopathies may
have a distinct clinical profile.
Discussion
Understanding the antigenic reactivity of clonal im-

560 n engl j med 374;6 nejm.org  February 11, 2016

The New England Journal of Medicine

Downloaded from nejm.org on September 30, 2017. For personal use only. No other uses without permission.
Copyright © 2016 Massachusetts Medical Society. All rights reserved.
 Brief Report

antigenic origins of myeloma but also may lead to
new strategies to prevent or treat clinical cancer
by targeting the underlying antigen. Long-term
antigenic stimulation may, in principle, also
promote genomic instability in myeloma by en-
gaging cytidine deaminases.21 In Gaucher’s dis-
ease–associated gammopathy, the reduction of
LGL1 may be achieved by substrate-reduction
therapies.22 Feeding eliglustat to GBA1−/− mice with
clonal immunoglobulins led to a reduction in
anti-LGL1 antibodies (Fig. 2F) as well as to a reduc-
tion of clonal immunoglobulin in vivo (Fig. 2G),
which indicates that Gaucher’s disease–associat-
ed gammopathy can be targeted by the reduction
of the underlying antigen. These data are also
consistent with recent findings regarding a re-
duced risk of B-cell cancers with substrate reduc-
tion in another model of Gaucher’s disease.23
The dysregulation of lysolipids has also been
described in the context of obesity.24 The risk of
myeloma has been shown to be higher among
obese persons than among persons of normal
weight,1 and diet-induced obesity was recently
shown to promote a myeloma-like condition in
mice.25 Further studies in larger cohorts are
needed to confirm clinical correlations and bet-
ter define the genetics of lipid-reactive myeloma.
Clinical studies are needed to assess whether
altering the levels of bioactive lipids can influ-
ence the natural history of gammopathies with
specificity for such lipids.
Supported in part by grants (CA135110, CA106802, and
CA156689) from the National Institutes of Health (NIH), by a
grant (65932) from the NIH and the National Institute of Ar-
thritis and Musculoskeletal and Skin Diseases, and by a Center
of Excellence Grant in Clinical Translational Research from
Genzyme.
Disclosure forms provided by the authors are available with
the full text of this article at NEJM.org.
We thank Lin Zhang for help with sample processing, and
Joan Keutzer, of Genzyme, for providing eliglustat–formulated
diets.

References
1. Palumbo A, Anderson K. Multiple my-
eloma. N Engl J Med 2011;​364:​1046-60.
2. Sahota SS, Leo R, Hamblin TJ, Steven-
son FK. Myeloma VL and VH gene se-
quences reveal a complementary imprint
of antigen selection in tumor cells. Blood
1997;​89:​219-26.
3. Sahota SS, Leo R, Hamblin TJ, Steven-
son FK. Ig VH gene mutational patterns
indicate different tumor cell status in hu-
man myeloma and monoclonal gammop-
athy of undetermined significance. Blood
1996;​87:​746-55.
4. Zojer N, Ludwig H, Fiegl M, Steven-
son FK, Sahota SS. Patterns of somatic
mutations in VH genes reveal pathways of
clonal transformation from MGUS to mul-
tiple myeloma. Blood 2003;​101:​4137-9.
5. Bakkus MH, Heirman C, Van Riet I,
Van Camp B, Thielemans K. Evidence that
multiple myeloma Ig heavy chain VDJ
genes contain somatic mutations but
show no intraclonal variation. Blood
1992;​80:​2326-35.
6. Vescio RA, Cao J, Hong CH, et al. My-
eloma Ig heavy chain V region sequences
reveal prior antigenic selection and marked
somatic mutation but no intraclonal diver-
sity. J Immunol 1995;​155:​2487-97.
7. Preuss KD, Pfreundschuh M, Ahl-
grimm M, et al. A frequent target of para-
proteins in the sera of patients with mul-
tiple myeloma and MGUS. Int J Cancer
2009;​125:​656-61.
8. Grass S, Preuss KD, Thome S, et al.
Paraproteins of familial MGUS/multiple
myeloma target family-typical antigens:
hyperphosphorylation of autoantigens is
a consistent finding in familial and spo-
radic MGUS/MM. Blood 2011;​118:​635-7.
9. Preuss KD, Pfreundschuh M, Fadle N,
et al. Hyperphosphorylation of autoanti-
genic targets of paraproteins is due to in-
activation of PP2A. Blood 2011;​118:​3340-6.
10. Preuss KD, Pfreundschuh M, Weigert
M, Fadle N, Regitz E, Kubuschok B. Su-
moylated HSP90 is a dominantly inherit-
ed plasma cell dyscrasias risk factor.
J Clin Invest 2015;​125:​316-23.
11. Anderson PN, Potter M. Induction of
plasma cell tumours in BALB-c mice with
2,6,10,14-tetramethylpentadecane (pris-
tane). Nature 1969;​222:​994-5.
12. Landgren O, Rajkumar SV, Pfeiffer
RM, et al. Obesity is associated with an
increased risk of monoclonal gammopathy
of undetermined significance among black
and white women. Blood 2010;​116:​1056-9.
13. Mistry PK, Taddei T, vom Dahl S,
Rosenbloom BE. Gaucher disease and
malignancy: a model for cancer patho-
genesis in an inborn error of metabolism.
Crit Rev Oncog 2013;​18:​235-46.
14. Dekker N, van Dussen L, Hollak CE, et
al. Elevated plasma glucosylsphingosine in
Gaucher disease: relation to phenotype,
storage cell markers, and therapeutic re-
sponse. Blood 2011;​118(16):​e118-27.
15. Nair S, Boddupalli CS, Verma R, et al.
Type II NKT-TFH cells against Gaucher
lipids regulate B-cell immunity and in-
flammation. Blood 2015;​125:​1256-71.
16. Chang DH, Deng H, Matthews P, et al.
Inflammation-associated lysophospholip-
ids as ligands for CD1d-restricted T cells in
human cancer. Blood 2008;​112:​1308-16.
17. Mistry PK, Liu J, Yang M, et al. Gluco-
cerebrosidase gene-deficient mouse reca-
pitulates Gaucher disease displaying cel-
lular and molecular dysregulation beyond
the macrophage. Proc Natl Acad Sci USA
2010;​107:​19473-8.
18. Nooij FJ, Van der Sluijs-Gelling AJ, Jol-
Van der Zijde CM, Van Tol MJ, Haas H,
Radl J. Immunoblotting techniques for
the detection of low level homogeneous
immunoglobulin components in serum.
J Immunol Methods 1990;​134:​273-81.
19. Braun W, Abraham R. Modified diffu-
sion blotting for rapid and efficient pro-
tein transfer with PhastSystem. Electro-
phoresis 1989;​10:​249-53.
20. Sasagawa T, Okita M, Murakami J,
Kato T, Watanabe A. Abnormal serum ly-
sophospholipids in multiple myeloma pa-
tients. Lipids 1999;​34:​17-21.
21. Koduru S, Wong E, Strowig T, et al.
Dendritic cell-mediated activation-in-
duced cytidine deaminase (AID)-depen-
dent induction of genomic instability in
human myeloma. Blood 2012;​119:​2302-9.
22. Mistry PK, Lukina E, Ben Turkia H, et
al. Effect of oral eliglustat on splenomeg-
aly in patients with Gaucher disease type
1: the ENGAGE randomized clinical trial.
JAMA 2015;​313:​695-706.
23. Pavlova EV, Archer J, Wang S, et al.
Inhibition of UDP-glucosylceramide syn-
thase in mice prevents Gaucher disease-
associated B-cell malignancy. J Pathol
2015;​235:​113-24.
24. Pietiläinen KH, Sysi-Aho M, Rissanen
A, et al. Acquired obesity is associated
with changes in the serum lipidomic pro-
file independent of genetic effects — a
monozygotic twin study. PLoS One 2007;​
2(2):​e218.
25. Lwin ST, Olechnowicz SW, Fowler JA,
Edwards CM. Diet-induced obesity pro-
motes a myeloma-like condition in vivo.
Leukemia 2015;​29:​507-10.
Copyright © 2016 Massachusetts Medical Society.

n engl j med 374;6 nejm.org  February 11, 2016 561

The New England Journal of Medicine
Downloaded from nejm.org on September 30, 2017. For personal use only. No other uses without permission.
Copyright © 2016 Massachusetts Medical Society. All rights reserved.