Content uploaded by Michele WL Teng
Author content
All content in this area was uploaded by Michele WL Teng on Aug 31, 2015
Content may be subject to copyright.
DOI: 10.1126/science.1249486
, 147 (2014);343 Science
Michele W. L. Teng and Mark J. Smyth
Can Cancer Trigger Autoimmunity?
This copy is for your personal, non-commercial use only.
clicking here.colleagues, clients, or customers by , you can order high-quality copies for yourIf you wish to distribute this article to others
here.following the guidelines can be obtained byPermission to republish or repurpose articles or portions of articles
): January 9, 2014 www.sciencemag.org (this information is current as of
The following resources related to this article are available online at
http://www.sciencemag.org/content/343/6167/147.full.html
version of this article at: including high-resolution figures, can be found in the onlineUpdated information and services,
http://www.sciencemag.org/content/343/6167/147.full.html#related
found at: can berelated to this article A list of selected additional articles on the Science Web sites
http://www.sciencemag.org/content/343/6167/147.full.html#ref-list-1
, 2 of which can be accessed free:cites 8 articlesThis article
registered trademark of AAAS. is aScience2014 by the American Association for the Advancement of Science; all rights reserved. The title CopyrightAmerican Association for the Advancement of Science, 1200 New York Avenue NW, Washington, DC 20005.
(print ISSN 0036-8075; online ISSN 1095-9203) is published weekly, except the last week in December, by theScience
on January 9, 2014www.sciencemag.orgDownloaded from on January 9, 2014www.sciencemag.orgDownloaded from on January 9, 2014www.sciencemag.orgDownloaded from
www.sciencemag.org SCIENCE VOL 343 10 JANUARY 2014 147
PERSPECTIVES
Autoimmune diseases occur when
immune B and T cells fail to distin-
guish the body’s own proteins as self
and attack them, ultimately damaging tissues
and organs. There are many possible causes
of autoimmune diseases, such as chemical
exposures, infections, and genetic factors.
On page 152 of this issue, Joseph et al. ( 1)
raise another possibility—nascent growing
cancers might express new proteins (neoanti-
gens) whose exposure to the immune system
creates the potential for an autoimmune dis-
ease to develop.
Systemic sclerosis (scleroderma) is a
rare chronic autoimmune rheumatic dis-
ease associated with fi brosis of the skin and
widespread destruction of blood vessels that
affects organs, with life-threatening conse-
quences. Some rheumatic diseases are also
paraneoplastic—they are present with can-
cers—and removal of the tumor or its medi-
cal treatment causes regression of the clini-
cal manifestations. Coincidental timing of
scleroderma and cancer had been observed
in patients with autoantibodies specifi c for
RNA polymerase III subunit (RPC1) ( 2).
Joseph et al. examined tumor tissue and
blood samples from 16 scleroderma patients
with various types of cancer. Among the
eight patients who had anti-RPC1 autoanti-
bodies and coincident cancer, six had genetic
mutations—either somatic mutations or
loss of heterozygosity (only one copy of an
allele) in POLR3A (polymerase III polypep-
tide A, the gene encoding RPC1) (see the
fi gure, below). The other eight patients had
autoantibodies to either topoisomerase 1
(anti-TOP1) or centromere protein B (anti-
CENPB) and developed delayed cancer.
Interestingly, all the anti-RPC1 anti-
bodies recognized wild-type and mutated
RPC1, indicating that the humoral immune
response does not directly target the area
of the mutation or discriminate between
mutant and wild-type versions of RPC1.
But some patients with scleroderma who
possessed defi ned POLR3A mutations had
T cells that reacted to RPC1 protein frag-
ments produced from the mutated gene. The
reactivity was specific to the patient and
peptide, but the frequencies of these T cells
were comparable to those observed in other
autoimmune diseases. Given that POLR3A
mutations are exceedingly rare in cancer
(0.7% overall), it is unlikely that the onset
Can Cancer Trigger Autoimmunity?
CANCER
Michele W. L. Teng
1, 2 and Mark J. Smyth
1 ,2
A chronic autoimmune disease sometimes
appears coincidently in patients with cancer.
1Cancer Immunoregulation and Immunotherapy and Immu-
nology in Cancer and Infection Laboratories, QIMR Berg-
hofer Medical Research Institute, 300 Herston Road, Her-
ston, 4006 Queensland, Australia. 2School of Medicine,
University of Queensland, Herston, 4006 Queensland, Aus-
tralia. E-mail: mark.smyth@qimrberghofer.edu.au
during axonal growth and/or migration ( 5).
As neurons of different types and origins
adopt different strategies to polarize ( 6), it
will be interesting to explore whether apical
shedding is a marginal or widespread mecha-
nism in neurogenesis.
During mitosis of apical progenitors in the
mouse cortex, a ciliary remnant associates
with the centrosome at one pole of the mitotic
spindle, resulting in asymmetric inheritance
of a ciliary structure ( 7). Thus, the cilium is
partly conserved rather than dismantled dur-
ing cell division. The cell inheriting the cili-
ary remnant rapidly reconstitutes an apical
cilium and is more likely to retain apical pro-
genitor identity. Its sister produces a basolat-
eral cilium, a phenomenon associated with
basal progenitor identity. Basal progenitors’
restricted capacity to proliferate may result
from the reduced exposure of the basolat-
eral cilium to luminal mitogens ( 8). Future
research should address whether apical shed-
ding is also present in the developing cortex,
where indirect neurogenesis, through basal
progenitors, is predominant, and how shed-
ding integrates with asymmetric ciliary recy-
cling and basolateral cilia formation.
The cortex and neural tube appear to have
a remarkable plasticity of ciliary structures
during neurogenesis. Although mechanisti-
cally different, asymmetric recycling of cili-
ary remnants ( 7), differential positioning of
cilia ( 8), and dismantling of the cilium and
shedding of the ciliary membrane are many
ways for a cell to modulate the timing and
dosage of its exposure to environmental sig-
nals. These new layers of control in signal
reception are independent of the source con-
centration and may therefore contribute to the
diversity of cell fate decisions, allowing pro-
gression of neural differentiation and mainte-
nance of a pool of apical progenitors to occur
simultaneously in a niche.
References
1. R. M. Das, K. G. Storey, Science 343, 200 (2014).
2. R. M. Das, K. G. Storey, EMBO Rep. 13, 448 (2012).
3. D. L. Rousso et al., Neuron 74, 314 (2012).
4. J. P. Thiery, H. Acloque, R. Y. Huang, M. A. Nieto, Cell
139, 871 (2009).
5. P. L. Cheng, M. M. Poo, Annu. Rev. Neurosci. 35, 181
(2012).
6. Y. Hatanaka, K. Yamauchi, F. Murakami, Dev. Growth
Differ. 54, 398 (2012).
7. J. T. Paridaen, M. Wilsch-Bräuninger, W. B. Huttner, Cell
155, 333 (2013).
8. M. Wilsch-Bräuninger, J. Peters, J. T. Paridaen, W. B.
Huttner, Development 139, 95 (2012).
10.1126/science.1249497
POLR3A mutation or loss
of heterozygosity
Autoantibodies
Severe
Coincident Delayed,
~14 years later Not detectable
(occult? cured? never?)
Less severe ?
?
Anti-RPC1
Ant i-TOP1
or anti-CENPB
Scleroderma
Cancer
Yes
Anti-RPC1 Anti-RPC1
NoNo
Severe
Coincident
Coincidence. Four subsets of scleroderma patients
are shown (as observed by Joseph et al. (1)) with dif-
ferent autoantibody profi les, mutations, and coin-
cident cancer.
CREDIT: V. ALTOUNIAN/SCIENCE
Published by AAAS
10 JANUARY 2014 VOL 343 SCIENCE www.sciencemag.org
148
PERSPECTIVES
of scleroderma and the cancer genomes of
these patients were unrelated, and that the
mutations, and T cell responses directed
against them, were coincidental. Rather,
POLR3A mutation in the occasional cancer
triggers the scleroderma.
According to cancer immunoediting ( 3),
the host can control tumor growth through
innate and adaptive immune mechanisms.
Genetic instability in the tumor may cre-
ate neoantigens that are recognized by the
immune system, leading to the selection of
tumor cells that can escape immune pres-
sure. Neoantigens have been demonstrated in
tumors by means of epitope prediction algo-
rithms ( 4). These principles have been dem-
onstrated in mouse models of carcinogenesis
and multistage cancer progression ( 3, 4), but
also recently, the importance of a patient’s
immune reaction with cancer in dictating
disease-free survival (immune contexture)
has been established ( 5).
The relatively low fraction of neoplas-
tic cells with genetic alterations in the can-
cers from some of the scleroderma patients
studied by Joseph et al. suggests that cancer
immunoediting had occurred. Cancer cells
targeted by T cells appear to selectively lose
the mutant allele, and loss of heterozygos-
ity occurred in most of the tumors that were
present synchronously with the scleroderma.
In some tumors, the loss of heterozygosity
was not yet complete (clonal heterogene-
ity was observed in which a small percent-
age of the tumor cells still carried the muta-
tion). Presumably, if those tumors had been
sampled a few months later, there might not
be any detectable mutant cells because loss
of heterozygosity would have been com-
plete. The loss of heterozygosity in the other
tumors without detectable mutations is quite
possibly a “historical” record of the tumors
once carrying mutations. Interestingly, no
mutant allele or loss of heterozygosity was
detected in some patients with anti-RPC1
autoantibodies, suggesting that other mech-
anisms can induce their production. The eti-
ology of scleroderma development in these
patients may resemble that in patients with
anti-TOP1 and anti-CENPB autoantibod-
ies. It is also possible that patients with
RPC1-specifi c autoantibodies and no can-
cer may once have had nascent cancer or
harbor cancer in a state of immune-medi-
ated dormancy. The loss of heterozygos-
ity of POLR3A in some patients raises the
possibility that immunoediting may involve
gene copy-number alterations in the cancer.
Alternatively, because RPC1 plays a role in
sensing and limiting infection by intracellu-
lar bacteria and viruses ( 6), cancer may also
Second-closest to Jupiter (after Io) of
the four large Galilean satellites,
icy Europa is one of the strangest
objects in the solar system ( 1). On page 171
of this issue, Roth et al. ( 2) present strong
evidence for ongoing eruptions of plumes
of water vapor from Europa’s surface. This
is a potentially major discovery, making
Europa only the fourth object in the solar
system known to exhibit ongoing internally
powered geological activity, after Earth,
Europa’s volcanic neighbor moon Io, and
Saturn’s icy moon Enceladus.
Europa is slightly smaller than our own
Moon, with a radius of 1561 km and a density
of 3010 kg m–3, implying an interior mostly
composed of silicates plus about 10% water
by mass. Gravity observations by the Gali-
leo Jupiter orbiter revealed that Europa is
strongly differentiated, with the lower-density
water component concentrated near the sur-
face ( 3). The surface is dominated by water
ice, along with other hydrated species, proba-
bly salts. Europa’s surface appearance is truly
bizarre, crisscrossed with fractures, some-
times global in extent, that on closer inspec-
tion turn out to be double ridges (see the fi g-
Glimpsing Eruptions on Europa
PLANETARY SCIENCE
John R. Spencer
Recent observations with the Hubble Space Telescope reveal what are probably eruptions of water
vapor from Jupiter’s moon Europa.
Southwest Research Institute, 1050 Walnut Street, Boulder,
CO 80302, USA. E-mail: spencer@boulder.swri.edu
mutate or lose POLR3A to evade activating
innate immunity.
The association between cancer and rheu-
matic diseases has been intriguing for many
years ( 7). Paraneoplastic syndromes may
be mediated by autoantibodies due either to
tumor antigens that are also expressed by
cells targeted by the autoimmune disease or
to the release of intracellular antigens from
apoptotic tumor cells ( 8). However, these
responses are directed to the normal pro-
tein, and there is yet no evidence that encod-
ing genes were mutated in the tumors. Joseph
et al. extend causation to include mutant
POLR3A in human tumors that elicits an
immune response against the mutant RPC1
but cross-reacts with the normal RPC1,
resulting in autoimmunity. However, the can-
cer mutation is likely not enough and addi-
tional factors (genetic, environmental, or
target tissue–specific) may be required to
generate damage to normal tissue.
From an epidemiological viewpoint, the
study by Joseph et al. is underpowered, but
these fi ndings may also prompt research into
whether antigens other than RPC1 might
trigger autoimmunity (e.g., in myositis and
lupus). In larger studies, mining exome
sequencing data can help identify a patient’s
specifi c tumor antigen profi le and potential
tumor immunity and autoimmunity. How-
ever, the fi ndings by Joseph et al. raise inter-
esting issues about tumors with high mutation
rates (lung, melanoma) or microsatellite DNA
instability (colorectal) and their predisposi-
tion to cause autoimmunity. Simple analysis
of large tissue collections from cancer patients
with accompanying “immunoscores” and
patient disease history might reveal correla-
tions with autoimmune syndromes.
Very little autoimmunity has been caus-
ally linked to cancer mutations and subse-
quent immune reactions, but these events
have only been examined in cancer patients
with clinically detectable disease. Do autoim-
mune patients in general have an undetectable
burden or history of cancer? New immuno-
therapeutic treatments for cancer that break
tolerance will help to clarify the underlying
natural human immune reaction to cancer and
its side effects on normal tissues.
References and Notes
1. C. G. Joseph et al., Science 343, 152 (2014); 10.1126/
science.1246886.
2. A. A. Shah, A. Rosen, L. Hummers, F. Wigley, L. Casciola-
Rosen, Arthritis Rheum. 62, 2787 (2010).
3. R. D. Schreiber, L. J. Old, M. J. Smyth, Science 331, 1565
(2011).
4. H. Matsushita et al., Nature 482, 400 (2012).
5. J. Galon et al., Science 313, 1960 (2006).
6. A. Ablasser et al., Nat. Immunol. 10, 1065 (2009).
7. J. E. Naschitz, Curr. Opin. Rheumatol. 13, 62 (2001).
8. M. L. Albert, R. B. Darnell, Nat. Rev. Cancer 4, 36 (2004)
.
Acknowledgments: M.W.L.T is supported by a CDF1 Fellow-
ship and Project Grant from the National Health and Medical
Research Council of Australia (NH&MRC) and Prostate Cancer
Foundation of Australia. M.J.S. is supported by a NH&MRC
Australia Fellowship and Program Grant, and Program Grant
from the Susan G. Komen for the Cure.
10.1126/science.1249486
Published by AAAS