PARP-1 and cytokine-mediated β-cell damage: a nick in the Okamoto model?


PARP-1 and cytokine-mediated �-cell damage: a nick in the Okamoto model?

Nadia Cobo-Vuilleumier and Benoit R. Gauthier
Pancreatic Islet Development and Regeneration Unit, Department of Stem Cells, CABIMER-Andalusian Center for Molecular
Biology and Regenerative Medicine, Seville, Spain

Submitted 7 May 2012; accepted in final form 7 May 2012

TYPE 1 DIABETES MELLITUS (T1DM), a complex genetic disease
characterized by the destruction of pancreatic islet �-cells,
leads to insulin deficiency and, ultimately, hyperglycemia in
inflicted individuals. The etiology stems from a progressive
autoimmune assault, in which macrophages and T cells secrete
proinflammatory cytokines, such as IL-1�, IFN-�, and TNF-�,
as well as nitric oxide (NO), which provoke �-cell death. The
latter situation is further aggravated, as remaining �-cells
produce chemokines, thereby attracting more immune cells
and, thus, escalating the inflammatory process. Cytokine-me-
diated cell death proceeds through necrosis or apoptosis and
implicates the activation of the Jak/Stat and NF-�B signaling
pathways (4). A common downstream target gene of these
pathways is inducible NO synthase (iNOS), which generates
NO, a major effector of �-cell death along with reactive
oxygen species.

Historically, T1DM has been recapitulated in rodents with
use of diabetogenic agents, such as streptozotocin (STZ) and
alloxan, to destroy the �-cell mass. In particular, STZ is
distinctly taken up by �-cells, inducing DNA alkylation and
leading to impaired �-cell function and death. In addition, STZ
was also found to generate NO species, further deteriorating
�-cell survival (7). More than 30 years ago, Okamoto and
colleagues (13) demonstrated that STZ and alloxan cause cell
death by inducing DNA strand breaks and the activation of
poly(ADP)-ribose synthase. This enzyme, also known as poly-
(ADP)-ribose polymerase-1 (PARP-1), is activated by DNA
breaks and uses NAD� to covalently link units of ADP ribose,
thereby forming long polymers on nuclear acceptor proteins
such as histones. This has been suggested to diminish their
affinity for DNA, opening the site of DNA damage and
allowing access to the base excision repair machinery (8).
Despite this important role in signaling and facilitating DNA
repair, the split functional personality of PARP-1 as Mr. Hyde
and Dr. Jeckyll is revealed when PARP-1 is hyperactivated.
Indeed, in an attempt to replenish the NAD� pool that is being
drained by the PARP-1, the cellular ATP stores are rapidly
exhausted, resulting in energy crisis-induced necrosis. The
latter forms the basis of the “Okamoto model,” which provides
a unifying hypothesis on how hyperactivation of PARP-1 leads
to impaired �-cell function (insulin biosynthesis and secretion)
and, ultimately, death in response to chemical insults (12, 13).
In agreement with this model, several independent studies have
shown that mice lacking PARP-1 were resistant to STZ-
mediated �-cell death and to development of hyperglycemia
(3, 9, 11). In parallel, PARP-1-deficient mice were found to be

extremely resistant to LPS-induced endotoxic shock as a con-
sequence of a blunted NF-�B-dependent transcription, leading
to inhibition of iNOS, as well as TNF-� and IFN-�, expression
in macrophages (10). Thus, PARP-1 not only mediates the
cytotoxic effect of NO, but it also appears to stimulate inflam-
mation by regulating the NF-�B signal transduction pathway.

These compelling data led to a large randomized double-
blind, placebo-controlled trial in which pre-T1DM participants
recruited from 18 European countries, as well as the United
States and Canada, were given the PARP-1 inhibitor nicotin-
amide or a placebo. After five years of treatment, the incidence
of developing T1DM was identical in both groups (5). These
studies highlight the complexity of T1DM and the pitfalls in
translating lessons learned in animal models of diabetes to
human physiology. Indeed, one critical question arises: How
well does the STZ-induced �-cell death model mimic T1DM?
In fact, hyperactivation of PARP-1 by STZ results in �-cell
death through necrosis, while �-cell destruction in T1DM
predominantly proceeds through a cytokine-mediated apopto-
sis process. Interestingly, PARP-1 is cleaved and inactivated
by caspase-3 and -7 in the early phase of apoptosis, suggesting
that PARP-1 does not participate in this mode of cell death (8).
Furthermore, Heller et al. (6), as early as 1995, demonstrated
that islets isolated from PARP-1-deficient mice were as sensi-
tive as control islets to cell lysis by high concentrations of NO
and reactive oxygen species. These data clearly raise serious
doubts about the importance of PARP-1 as a key actor in �-cell
destruction in a setting of T1DM. Thus, although the role of
PARP-1 in �-cell destruction via necrosis has been irrefutably
associated with STZ and in establishing the validity of the
Okamoto model, its role in cytokine-mediated cell death re-
mains to be fully explored.

In this issue of the Journal, Andreone et al. (1) address this
gap and challenge the Okamoto model by demonstrating that
PARP-1 does not participate in the deleterious effect of cyto-
kines on islet function and, potentially, survival. Indeed, An-
dreone et al. demonstrate that, similar to control islets, PARP-
1-deficient islet cells exposed to the proinflammatory cytokines
IL-1 and IFN-� exhibit normal activation of the NF-�B sig-
naling pathway, which leads to induction of iNOS expression
and production of NO. Furthermore, deletion of PARP-1 did
not alter IL-1/IFN-�-mediated phosphorylation of Stat-1, indi-
cating that the Jak/Stat signaling pathway implicated in apo-
ptosis is activated by cytokines in PARP-1-deficient islets (4).
Using a biochemical assay that measures uptake of a neutral
red dye within acidic lysosomes, Andreone et al. show that
islets lacking PARP-1 are resistant to cytokine-mediated cyto-
toxicity, whereas they remain sensitive to the general apoptotic
agent staurosporine. As lysosome acidification is an ATP-
dependent process, which may be preserved in PARP-1-defi-
cient islets due to sustained levels of ATP, Andreone et al.
astutely used the terminal deoxynucleotidyl transferase-medi-

Address for reprint requests and other correspondence: B. R. Gauthier,
Pancreatic Islet Development and Regeneration Unit, Dept. of Stem Cells,
CABIMER-Andalusian Center for Molecular Biology and Regenerative Med-
icine, Avda Américo Vespucio, Parque Científico y Tecnológico Cartuja 93,
41092 Seville, Spain (e-mail: benoit.gauthier@cabimer.es).

Am J Physiol Endocrinol Metab 303: E170–E171, 2012;
doi:10.1152/ajpendo.00226.2012.Editorial Focus

0193-1849/12 Copyright © 2012 the American Physiological Society http://www.ajpendo.orgE170

Downloaded from journals.physiology.org/journal/ajpendo at Carnegie Mellon Univ (128.182.081.034) on April 5, 2021.



ated dUTP nick-end labeling (TUNEL) assay as an alternative
approach to evaluate cell death. The TUNEL assay is particu-
larly useful, as it reveals DNA strand breaks, which are a
hallmark of NO-mediated cell death. Consistent with preserved
NO production, PARP-1-deficient islets exhibited increased
TUNEL-positive cells in the presence of cytokines to levels
similar to those of treated control islets. Consequently, glu-
cose-induced insulin secretion was also impaired in PARP-1-
lacking islets treated with cytokine. As development of T1DM
results from an intricate cross talk between immune cells and
islets, Andreone et al. also assessed whether inflammatory cell
signaling was impaired in macrophages derived from PARP-
deficient mice, as previously reported by others (10). Surpris-
ingly, the NF-�B signaling pathway, as well as downstream
iNOS activation and NO production, were unaltered in mac-
rophages derived from PARP-1 knockout mice that were ex-
posed to cytokines or LPS. A potential explanation for discrep-
ancies in results may stem from the fact that previous studies
were performed with macrophages exposed in vivo to LPS
(10), while Andreone et al. studied naïve macrophages treated
in vitro with the endotoxin.

The work by Andreone et al. (1) demonstrates that, in
contrast to STZ-mediated cell death and the Okamoto model,
PARP-1 does not contribute to cytokine-induced �-cell dam-
age. Furthermore, PARP-1 does not appear to be implicated in
the signaling of the inflammatory response by macrophages.
These results are consistent with the clinical trial demonstrat-
ing that nicotinamide does not improve onset of T1DM, caus-
ing a serious “nick” in the Okamoto model. Nonetheless, one
question remains to be clearly resolved from the study of
Andreone et al.: Does PARP-1 deficiency prevent, beyond
doubt, islet cell death? Indeed, DNA damage was increased in
islets from STZ-treated PARP-1-deficient mice, yet these ani-
mals did not develop hyperglycemia, and islet integrity and
function were preserved (11). It is tempting to speculate that
PARP-2, which accounts for �10% of all PARP activity in
mammalian cells (8), may provide sufficient residual activity to
repair STZ-mediated DNA damage in PARP-1 knockout mice,
thereby avoiding �-cell death and development of hyperglyce-
mia. In this context, it is interesting to note that PARP-2 was
recently shown to be essential for �-cell function and expan-
sion in response to a high-fat diet (2). Unfortunately, the
long-term effect of cytokines on �-cell DNA damage and
insulin secretion was not addressed by Andreone et al., as all
experiments were performed within 24 h of treatment. A key
study that could resolve whether PARP-1 deficiency is an asset
in protecting �-cells against a continuous exposure to proin-
flammatory cytokines would involve breeding of PARP-1-
deficient mice to animal models of experimental autoimmune
diabetes. This approach could also offer a valuable tool to
highlight and resolve potential pitfalls associated with the
failure of PARP-1 inhibitors to delay or prevent onset of
T1DM in human subjects. Nonetheless, recent findings dem-
onstrating the importance of PARP-2 in �-cell function, com-
bined with the apparent redundancy of PARP-1 in cytokine-
mediated �-cell dysfunction, strongly argue against the use of
new and more specific PARP inhibitors such as olaparib as
novel therapies for the treatment of T1DM.

ACKNOWLEDGMENTS

We thank Dr. Felipe Cortés Ledesma for critical reading of the manuscript.

GRANTS

Financial support for this study was provided by the Consejeria de Salud,
Junta de Andalucia, the Spanish Ministry of Science and Innovation, Instituto
de Salud Carlos III, and cofunded by Fondos FEDER and the Fundacion
Publica Andaluza Progreso y Salud.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

N.C.-V. and B.R.G. drafted the manuscript; N.C.-V. and B.R.G. edited and
revised the manuscript; N.C.-V. and B.R.G. approved the final version of the
manuscript.

REFERENCES

1. Andreone T, Meares GP, Hughes KJ, Hansen PA, Corbett JA.
Cytokine-mediated �-cell damage in PARP-1-deficient islets. Am J
Physiol Endocrinol Metab (First published 24 April 2012). doi:
10.1152/ajpendo.00055.2012.

2. Bai P, Canto C, Brunyanszki A, Huber A, Szanto M, Cen Y,
Yamamoto H, Houten SM, Kiss B, Oudart H, Gergely P, Menissier-de
Murcia J, Schreiber V, Sauve AA, Auwerx J. PARP-2 regulates SIRT1
expression and whole-body energy expenditure. Cell Metab 13: 450 –460,
2011.

3. Burkart V, Wang ZQ, Radons J, Heller B, Herceg Z, Stingl L, Wagner
EF, Kolb H. Mice lacking the poly(ADP-ribose) polymerase gene are
resistant to pancreatic beta-cell destruction and diabetes development
induced by streptozocin. Nat Med 5: 314 –319, 1999.

4. Eizirik DL, Mandrup-Poulsen T. A choice of death—the signal-trans-
duction of immune-mediated beta-cell apoptosis. Diabetologia 44: 2115–
2133, 2001.

5. Gale EA, Bingley PJ, Emmett CL, Collier T. European Nicotinamide
Diabetes Intervention Trial (ENDIT): a randomised controlled trial of
intervention before the onset of type 1 diabetes. Lancet 363: 925–931,
2004.

6. Heller B, Wang ZQ, Wagner EF, Radons J, Burkle A, Fehsel K,
Burkart V, Kolb H. Inactivation of the poly(ADP-ribose) polymerase
gene affects oxygen radical and nitric oxide toxicity in islet cells. J Biol
Chem 270: 11176 –11180, 1995.

7. Lenzen S. The mechanisms of alloxan- and streptozotocin-induced dia-
betes. Diabetologia 51: 216 –226, 2008.

8. Luo X, Kraus WL. On PAR with PARP: cellular stress signaling through
poly(ADP-ribose) and PARP-1. Genes Dev 26: 417–432, 2012.

9. Masutani M, Suzuki H, Kamada N, Watanabe M, Ueda O, Nozaki T,
Jishage K, Watanabe T, Sugimoto T, Nakagama H, Ochiya T, Sug-
imura T. Poly(ADP-ribose) polymerase gene disruption conferred mice
resistant to streptozotocin-induced diabetes. Proc Natl Acad Sci USA 96:
2301–2304, 1999.

10. Oliver FJ, Menissier-de Murcia J, Nacci C, Decker P, Andriantsito-
haina R, Muller S, de la Rubia G, Stoclet JC, de Murcia G. Resistance
to endotoxic shock as a consequence of defective NF-�B activation in
poly(ADP-ribose) polymerase-1 deficient mice. EMBO J 18: 4446 –4454,
1999.

11. Pieper AA, Brat DJ, Krug DK, Watkins CC, Gupta A, Blackshaw S,
Verma A, Wang ZQ, Snyder SH. Poly(ADP-ribose) polymerase-defi-
cient mice are protected from streptozotocin-induced diabetes. Proc Natl
Acad Sci USA 96: 3059 –3064, 1999.

12. Yamamoto H, Uchigata Y, Okamoto H. DNA strand breaks in pancre-
atic islets by in vivo administration of alloxan or streptozotocin. Biochem
Biophys Res Commun 103: 1014 –1020, 1981.

13. Yamamoto H, Uchigata Y, Okamoto H. Streptozotocin and alloxan
induce DNA strand breaks and poly(ADP-ribose) synthetase in pancreatic
islets. Nature 294: 284 –286, 1981.

Editorial Focus

E171PARP-1 AND CYTOKINE-MEDIATED �-CELL DEATH

AJP-Endocrinol Metab • doi:10.1152/ajpendo.00226.2012 • www.ajpendo.org
Downloaded from journals.physiology.org/journal/ajpendo at Carnegie Mellon Univ (128.182.081.034) on April 5, 2021.

http://dx.doi.org/10.1152/ajpendo.00055.2012