вход по аккаунту


Anti-neuronal antibodies in patients with cancer.

код для вставкиСкачать
cAMP Response Element
Binding Protein Family
Transcription Factors: The
Holy Grail of Neurological
On a trip to Glastonbury, England nearly a decade
ago, I had the privilege of visiting the Chalice Well.
There I learned of some of the lore connecting the
Chalice (ie, the Cup of the Last Supper) and the water
from the Chalice Well with the Holy Grail. Grail legend has captured the imaginations of many with the
notion that the contents of the Holy Grail are a mysterious source of life, healing. and prosperity. The nature of the Grail contents is a secret to be sought and
a prize to be won, but only by those who are worthy.
The parallels between the search for the Holy Grail
and the search for one or more therapeutic interventions that can protect and repair the brain are striking.
In this issue of the Annals, Lee and colleagues1 present
new data that fortifies arguments that cAMP response
element binding protein (CREB) family transcription
factors are the long sought source of life and healing, if
not prosperity in the nervous system.
What makes CREB so special as a therapeutic target?
CREB is a 43kDa member of a functional class of proteins known as transcription factors.2 Transcription
factors are proteins that localize to the nucleus to bind
to noncoding, regulatory regions of DNA. Classically,
CREB and its family members (CREM, ATF-1) bind
to specific sites corresponding to the sequence, 5⬘TGACGTCA-3⬘ in the promoter regions of genes.
When modified posttranslationally by phosphorylation,
CREB constitutively bound to DNA can recruit other
proteins involved in activating new gene expression.
CREB and its associated transcriptional proteins
(known as coactivators) form a complex on specific
sites of DNA with RNA polymerase II (an enzyme that
makes mRNA from DNA). The complex of CREB, its
coactivators, and RNA polymerase II is able to induce
expression of a host of genes with established roles in
cell survival, energy metabolism, regeneration, and
memory.2,3 Given CREB’s ability to regulate genes in
arenas critical for the prevention or recovery from neurological injury, it is not surprising that CREB phosphorylation and CREB-dependent gene expression can
be activated by established regulators of neuronal survival or regeneration. These include peptide hormones,
growth factors, and synaptic activity. Specific CREBregulated genes include the prosurvival protein, Bcl-24;
the survival, differentiation, and plasticity growth factor, brain-derived neurotrophic factor5; the proregeneration gene, arginase I6; and the metabolic control
protein, cytochrome c.7 CREB thus has the capacity to
activate a “cassette” of genes that could act at cellular,
local, and systemic levels to optimize survival, metabolism, and plasticity in the nervous system. It is the putative ability of CREB to activate, with one “flip of a
switch,” a multivalent protective and regenerative program that makes it such a compelling therapeutic target in the central nervous system, particularly for diseases with complex pathophysiology. Indeed, activation
of CREB is a therapeutic approach that is actively being investigated for Alzheimer’s disease,8 Huntington’s
disease,9 and recovery from spinal cord injury.10
Lee and colleagues1 extend the consideration of
CREB as a therapeutic target to include hypoxicencephalopathy (HI) in neonates. Perinatal HI is a major cause of morbidity and mortality in infants for
whom there are no effective treatments. The approach
the authors use to solve this important clinical problem
is an elegant one; they take advantage of the paradigm
of ischemic preconditioning to understand molecular
mechanisms underlying “tolerance” in the neonatal
brain. In brief, Lee and colleagues demonstrate that delaying the onset of global hypoxia by 24 hours, not 1
hour, after permanent unilateral carotid ligation in
neonatal rats significantly diminishes neuronal injury
and enhances behavioral performance. The 24-hour
preconditioning period is associated with increased
phosphorylation of CREB. Phosphorylation of CREB
at serine 133 allows CREB to recruit coactivators such
as CREB binding protein (CBP) and RNA polymerase
II to activate transcription of CREB-dependent genes,
including brain-derived neurotrophic factor. The findings by Huang and colleagues correlate the protective
effects of ischemic preconditioning with activation of
CREB-dependent gene expression and are consistent
with prior observations that CREB phosphorylation is
associated with resistance to ischemic injury in adult
gerbils or rats.11 Moreover, the observation that 1 hour
is not sufficient to induce tolerance is consonant with
the notion that de novo gene expression dependent on
CREB takes more than 1 hour to activate and execute.
To establish a causal role for CREB in tolerance induced by ischemic preconditioning, the authors use antisense oligodeoxynucleotides (AS ODNs). AS ODNs
are synthetic polymer deoxynucleotides similar to those
deoxynucleotides found in DNA that are 15 to 20
bases in length and whose sequence is antisense (3⬘-5⬘),
that is, complementary to the sense sequence (5⬘-3⬘) of
a molecule of mRNA . An AS ODN has the capacity
to bind to the mRNA from which the CREB protein is
normally synthesized. Binding of the AS ODN to the
CREB mRNA inhibits CREB synthesis by physically
blocking the ability of ribosomes to move along mes-
© 2004 American Neurological Association
Published by Wiley-Liss, Inc., through Wiley Subscription Services
senger RNA or by hastening the rate of degradation of
CREB mRNA in the cytosol. Lee and colleagues confirm that the AS ODN targeted to CREB, but not a
scrambled AS ODN, is capable of reducing CREB protein levels. The reductions in CREB levels are associated with decreased ischemic tolerance in a 24-hour
preconditioning paradigm. The data support the Lee
hypothesis that CREB is a critical mediator of ischemic
tolerance in the neonatal brain. However, AS ODNs
are notorious for low to moderate stringency binding
to undesired DNA sequences.12 Indeed, evidence that
the CREB ODNs used by Lee and colleagues specifically affect CREB (CREB-1) and not other CREB
family members (CREM, ATF-1, ATF-4) or neuroprotective transcription factors from distinct families (eg,
NF-␬B, HIF-1, or Sp1) is noticeably absent. Some of
these controls are particularly important given the established ability of CREB family member CREM to be
upregulated and to compensate for CREB function in
animals in which CREB is genetically ablated.9 The
advent of small interfering RNAs (siRNA) as a powerful and potentially more specific tool to knock down
CREB mRNA offers an alternative that could be used
in future studies of ischemic tolerance in neonatal
brains to verify Lee and colleagues’ tantalizing observations that CREB is necessary for ischemic preconditioning. Moreover, because CREB has been found to
localize to the mitochondria as well as the nucleus, future studies should also evaluate relative reductions in
CREB in the nucleus as compared with mitochondria
in susceptible neuronal populations.13
One of the most clinically relevant strategies to test
the involvement of a pathway in a neurological disease
is to identify small molecule “drugs” that can modulate
the target of interest and determine the effects of these
small molecules on neuronal damage and behavior in
an appropriate animal model. Following this logic, Lee
and colleagues thoughtfully investigated the ability of
two established, pharmacological activators of the
CREB pathway, forskolin and rolipram, to reduce HI
when administered before hypoxia in neonatal rodents.
Forskolin, derived from an Ayurvedic herb, increases
cAMP by activating the enzyme involved in its production, adenylate cyclase. In contrast, rolipram increases
cAMP by inhibiting one of the enzymes involved in its
degradation, phosphodiesterase 4. Increases in cAMP
via forskolin or rolipram enhance phosphorylation of
CREB at serine 133 by activating protein kinase A. Increased CREB phosphorylation leads to increased
CREB-dependent gene expression. Consistent with the
involvement of this pathway in ischemic preconditioning in neonates, Lee and colleagues demonstrated that
both agents administered before hypoxia enhance
CREB phosphorylation and decrease the degree of
brain injury. As pointed out by the authors, future
studies will clarify whether posthypoxia treatment with
Annals of Neurology
Vol 56
No 5
November 2004
either of these agents will be effective in minimizing
injury and optimizing functional recovery. It will also
be of academic interest to determine whether CREB is
necessary for the protective effects of forskolin or rolipram, and to evaluate whether CREB acts by binding
to cAMP response elements or hypoxia response elements to activate the expression of genes involved in
hypoxic adaptation.13
The finding that rolipram (a phosphodiesterase 4 inhibitor) can stimulate neonatal ischemic tolerance is an
exciting finding that, for several reasons, could have an
almost immediate impact on therapeutic trials of neonates who are deemed to be at high risk for HI.14 First,
unlike forskolin or peptide growth factors, rolipram
can be delivered orally or subcutaneously and can penetrate the blood–brain barrier. Second, it is a Food and
Drug Administration–approved agent developed initially as an antidepressant that has a track record of use
in humans. Rolipram’s utility as an antidepressant in
humans has been limited by its ability to induce disabling emesis in some patients. However, dosing schedules are being developed that limit emesis and may circumvent this important side effect in neonates as well
as adults. Third, because the target of rolipram, PDE4,
has been implicated in therapies for memory loss (Alzheimer’s disease), neuronal survival (Huntington’s disease), inflammation (multiple sclerosis), and regeneration (spinal cord injury), intense efforts are afoot to
develop PDE4 inhibitors that are as effective as rolipram but do not induce emesis.15,16 Finally, recent
studies by scientists at deCODE Genetics in Iceland
have identified polymorphisms in the regulatory region
of the PDE4 gene that confer increased risk for ischemic stroke.17 Taken together with the data presented
by Lee and colleagues, these findings suggest that dysregulation of PDE4 may increase stroke susceptibility
in the Icelandic population by lowering cAMP, decreasing CREB activation, and altering the level of endogenous tolerance mechanisms rather than by altering
the course of atherosclerosis as has been postulated.
The observations also support the possibility that rolipram may be effective at primary or secondary prevention of neonatal or adult stroke.
The impressive study of Lee and colleagues adds to
the growing exuberance surrounding activators of the
CREB pathway, particularly phosphodiesterase inhibitors such as rolipram in the prevention and repair of
brain injury. Continued investigation of the transcriptional mechanisms of tolerance in ischemic preconditioning in neonates and adults is likely to reveal other,
possibly novel, transcription factors in addition to
CREB whose activation by small molecule drugs will
complement agents such as forskolin or rolipram in activating salutary gene programs. The strategy offers a
compelling code from which the sources of life, healing,
and prosperity can be deciphered in the nervous system.
Rajiv R. Ratan, MD, PhD
Anti-neuronal Antibodies in
Patients with Cancer
Burke/Cornell Medical Research Institute
White Plains, NY
1. Lee H.-T., Chang Y.-C., et al. cAMP response element-binding
protein activation in ligation preconditioning in neonatal brain.
Ann Neurol 2004;611– 624.
2. Lonze BE, Ginty DD. Function and regulation of CREB family
transcription factors in the nervous system. Neuron 2002;35:
605– 623.
3. Walton MR, Dragunow I. Is CREB a key to neuronal survival?
Trends Neurosci 2000;23:48 –53.
4. Riccio A, Ahn S, Davenport CM, et al. Mediation by a CREB
family transcription factor of NGF-dependent survival of sympathetic neurons. Science 1999;286:2358 –2361.
5. Shieh PB, Ghosh A. Molecular mechanisms underlying activitydependent regulation of BDNF expression. J Neurobiol 1999;
6. Lange P, Langley B, Lu P, Ratan RR. Novel roles for arginase
I in cell survival, regeneration, and translation in the nervous
system. J Nutrition (in press).
7. Scarpulla R. Transcriptional activators and coactivators in the
nuclear control of mitochondrial function in mammalian cells.
Gene 2002;286:81– 89.
8. Vitolo OV, Sant’Angelo A, Costanzo V, et al. Amyloid Beta
peptide inhibition of the PKA/CREB pathway and long term
potentation: reversibility by drugs that enhance cAMP signaling. Proc Natl Acad Sci USA 2002;99:13217–13221.
9. Mantamadiotis T, Lernberger T, Bleckman SC, et al. Disruption of CREB function in brain leads to neurodegneration. Nat
Genet 2002;31:47–54.
10. Nikulina E, Tidwell JL, Dai HN, et al. The phosphodiesterase
inhibitor rolipram delivered after a spinal cord lesion promotes
axonal regeneration and functional recovery. Proc Natl Acad Sci
USA 2004;101:8786 – 8790.
11. Hara T, Hamada J, Yano S, et al. CREB is required for the
acquisition of ischemic tolerance in gerbil hippocampal CA1
region. J Neurochem 2003;86:805– 814.
12. Stein CA, Narayanan R. Antisense oligodeoxynucleotides. Curr
Opin Oncol 1994;6:587–594.
13. Cammarota M, Paratcha G, Bevilaqua LR, et al. Cyclic AMPresponsive element binding protein in brain mitochondria.
J Neurochem 1999;72:2272–2277.
13. Zaman K, Ryu H, Hall D, et al. Protection from oxidative
stress-induced apoptosis in cortical neuronal cultures by iron
chelators is associated with enhanced DNA binding of hypoxiainducible factor-1 and ATF-1/CREB and increased expression
of glycolytic enzymes, p21 waf1/cip1 and erythropoietin.
J Neurosci 1999;19:9821–9830.
14. Zhu J, Mix E, Winblad B. The antidepressant and antiinflammatory effects of rolipram in the central nervous system.
CNS Drug Rev 2001;7:387–398.
15. Bielekova B, Lincoln A, McFarland H, Martin R. Therapeutic
potential of phosphodiesterase-4 and -3 inhibitors in Th1mediated autoimmune diseases. J Immunol 2000;164:
16. Dyke HJ, Montana JG. Update on the therapeutic potential of
PDE4 inhibitors. Expert Opin Invest Drugs 2002;11:1–13.
17. Gretarsdottir S, Thorleifsson G, Reynisdottir ST, et al. The
gene encoding phosphodiesterase 4D confers risk of ischemic
stroke. Nat Genet 2003;35:131–138.
DOI: 10.1002/ana.20308
In this issue of the Annals, Pittock and colleagues1 address several important clinical issues with respect to
paraneoplastic syndromes that are not widely recognized by neurologists encountering such patients. Taking advantage of a database that includes approximately
60,000 patients suspected of suffering from paraneoplastic disorders, the investigators searched for the presence of multiple antibodies against antigens shared by
the cancer and the nervous system. Their search was
for not only antibodies against the commonly recognized neuronal nuclei and cytoplasmic antigens, but
also antibodies against ion channels, acetylcholine receptors, and muscle. They found that with the exception
of the PCA-1 (anti-Yo) antibody, more than one autoantibody occurred in a significant number of patients.
For example, in the serum of 217 patients harboring the
ANNA-1 (anti-Hu) antibody, a paraneoplastic antibody
generally associated with encephalomyelitis,2 17 patients
also harbored CRMP-5, 2 harbored PCA-2, 3 harbored
ANA-2, and 1 harbored ANNA-3. Furthermore, ion
channel, acetylcholine receptor, and striational antibodies occurred with a cumulative frequency of 28 patients. Thus, 43 patients harboring the anti–Hu antibody had one or more coexisting paraneoplastic
antibodies in their serum, an overall frequency of approximately 20%. As the authors point out, multiple
antibodies in some patients with paraneoplastic syndromes have been reported. However, the phenomenon has not been widely recognized, and, in many series, including our own experience, only a single
antibody has been identified or reported.2– 4 The difference probably results both from the very high number of patients in this series and the high number of
antibodies screened. These findings should lead neurologists to rethink how they submit sera for paraneoplastic investigation.
Several important issues, both clinical and scientific,
are raised by this article. First, among the 60,000 sera
assayed, only 553 were positive for any of the welldefined paraneoplastic antibodies calling attention to
the rarity of antibody-positive paraneoplastic syndromes even in patients suspected of that diagnosis by
the physicians who referred the serum to Dr Lennon’s
laboratory. Second, because different antibodies can be
associated with the same clinical findings5 and the
same antibody can be associated with different clinical
syndromes,4 paraneoplastic antibodies should be
searched for by screening rather than by focusing on a
specific antibody. In this study, as in our laboratory,
© 2004 American Neurological Association
Published by Wiley-Liss, Inc., through Wiley Subscription Services
serum and cerebrospinal fluid are assayed by the combination of immunohistochemistry against neuronal
tissue and Western blotting against both neuronal tissue and cloned antigens. This approach not only identifies the presence of a neuronal autoantibody, but also
characterizes it definitively. It also allows identification
of new and novel paraneoplastic antibodies. Third, although paraneoplastic neuronal nuclear or cytoplasmic
autoantibodies do not definitively identify the neurological syndrome (eg, cerebellar syndromes can accompany anti–Hu, anti–Yo, or anti–Ri antibodies as well as
others), the presence of these antibodies strongly suggests that a careful search will identify that the patient
has cancer (see Table 2 of Pittock and colleagues1) and,
furthermore, point with good accuracy to where that
cancer will be located.
The article also raises scientific issues. The first, of
course, is why a patient with cancer develops antibodies at all? Most evidence suggests that the neuronal antigens found in the cancer are identical to those found
in the nervous system; that is, the antigen in the cancer
is “self” and thus would not be expected to be recognized as foreign by the host. The second question is
why more than one antigen? The authors have suggested intermolecular epitope spreading as one possibility. In this regard, the absence of multiple antibodies in
patients with PCA-1 (anti-Yo) is of interest. The
PCA-1 antibody usually occurs in patients with gynecological, especially ovarian, or breast cancer tumors
Annals of Neurology
Vol 56
No 5
November 2004
that rarely express neuronal antigen. The other antibodies measured in this study often are associated with
small cell lung cancer, a tumor that often expresses several neuronal antigens. This suggests that the epitope
spread may come from an immune reaction to the tumor. It would be interesting to know whether the presence of multiple antibodies correlates with better tumor outcome. These scientific questions await
Jerome B. Posner, MD
New York, NY
1. Pittock SJ, Kryzer TJ, Lennon VA. Paraneoplastic antibodies coexist and predict cancer, not neurological syndrome. Ann Neurol
2. Graus F, Keime-Guibert F, Reñe R, et al. Anti-Hu-associated
paraneoplastic encephalomyelitis: analysis of 200 patients. Brain
2001;124:1138 –1148.
3. Vianello M, Vitaliani R, Pezzani R, et al. The spectrum of antineuronal autoantibodies in a series of neurological patients.
J Neurol Sci 2004;220:29 –36.
4. Dalmau J, Graus F, Villarejo A, et al. Clinical analysis of antiMa2-associated encephalitis. Brain 2004;127:1831–1844.
5. Gultekin SH, Rosenfeld MR, Voltz R, et al. Paraneoplastic limbic encephalitis: neurological symptoms, immunological findings
and tumour association in 50 patients. Brain 2000;123:
DOI: 10.1002/ana.20311
Без категории
Размер файла
54 Кб
cancer, patients, antibodies, neuronal, anti
Пожаловаться на содержимое документа