Tertiary mechanisms of brain damage: a new hope for
treatment of cerebral palsy?
Bobbi Fleiss, Pierre Gressens
Lancet Neurol 2012; 11: 556-66
Cerebral palsy is caused by injury or developmental disturbances to the immature brain and leads to substantial motor,
Centre for the Developing
cognitive, and learning defi cits. In addition to developmental disruption associated with the initial insult to the immature
Brain, Institute of Reproductive
brain, injury processes can persist for many months or years. We suggest that these tertiary mechanisms of damage
and Developmental Biology,
might include persistent infl ammation and epigenetic changes. We propose that these processes are implicit in prevention
Department of Surgery and
of endogenous repair and regeneration and predispose patients to development of future cognitive dysfunction and
Cancer, Imperial College,
sensitisation to further injury. We suggest that treatment of tertiary mechanisms of damage might be possible by various
London, UK (B Fleiss PhD,
means, including preventing the repressive eff ects of microglia and astrocyte over-activation, recapitulating
Prof P Gressens MD); INSERM,
developmentally permissive epigenetic conditions, and using cell therapies to stimulate repair and regeneration
U676, Paris, France (B Fleiss,
Recognition of tertiary mechanisms of damage might be the fi rst step in a complex translational task to tailor safe and
P Gressens); Universite Paris
Diderot, Faculte de Medecine,
eff ective therapies that can be used to treat the already developmentally disrupted brain long after an insult.
Paris, France (B Fleiss,
P Gressens); and PremUP, Paris,
eff ect on development, and we cannot completely
France (B Fleiss, P Gressens)
Cerebral palsy is caused by non-progressive malfor mations
separate the eff ect of a disrupted developmental trajectory
or lesions to the immature brain that lead to develop-
from the eff ects of persisting active mechanisms of
Dr Bobbi Fleiss, INSERM U676,
Hopital Robert Debre,
mental disturbance (panel 1).1 Although cerebral palsy tertiary injury. However, with additional research, these
48 Boulevard Serurier,
characteristically presents with motor dysfunction, active ongoing processes might be targets for
F-75019 Paris, France
substantial impairments in sensation, behaviour, cog-
pharmacological treatment to improve long-term
nition, and communication are common comorbidities. neurological outcome in patients with cerebral palsy.
Extensive multicentre trials have revealed that only
Changes that persist in the brain after perinatal
about 10% of cases of cerebral palsy can be attributed injury might include myelin defi cits, reduced plasticity,
to hypoxia-ischaemia around the time of birth, as defi ned and altered cell number. The tertiary mechanisms that
by blood lactate, pH and base defi cit, and Apgar score.2,3
govern these eff ects, we speculate, include persistent
ammation, genetic factors, low birthweight, and infl ammation and epigenetic changes, which cause a
prematurity are among the strongest risk factors for blockade of oligodendrocyte maturation, impaired
cerebral palsy;4-6 however, in most cases the causal neurogenesis, impaired axonal growth, or altered
mechanism is unknown. A large proportion of, but not all,
synaptogenesis (fi gure 2).8,9 Also, we suggest that, long
cases are caused by events in the perinatal period and we after injury, tertiary brain damage might become
focus much of this Review on the eff ects of injury to the apparent only in conjunction with a so-called second
brain in this developmental period. Also, because the hit. In such a scenario, sensitisation to cognitive
specifi c details of the perinatal events that lead to cerebral dysfunction, cell loss, and seizures during subsequent
palsy are often unknown, we cannot derive a causal relation
infl ammation10-12 would occur as a result of the tertiary
to later neurological dysfunction in human beings. As processes of persistent glial priming leading to
such, where there are limited data on the human neonatal
accumulation of excitatory neurotransmitters and
disorder, we will discuss animal models and studies of aberrant proinfl am matory signalling.
adult injury, where cause and eff ect are more easily
The processes of delayed cell death after hypoxia-
identifi ed. However, we acknowledge that direct extrapo-
ischaemia were only recognised in the 1980s, and soon
lation of conclusions on the basis of fi ndings from adult after the notion of the window of therapeutic intervention
studies is not possible because of the substantial diff erences was developed.13-15 These were key steps in the
in damage and repair mechanisms between the neonatal development of therapeutic hypothermia--the only
and adult brains.
eff ective therapy for perinatal brain injury.16 In a similar
Findings from animal and clinical studies7 suggest that
way, we suggest a change in the conceptualisation of
in a subset of infants with cerebral palsy there are injury to incorporate tertiary damage processes. We hope
persistent active mechanisms that prevent regeneration this concept will provide a new outlook on treatment
or exacerbate brain damage, or both, and sensitise the options for perinatal brain damage, which will expand
patient's brain to further injury. We defi ne these processes
the treatment window and substantially increase the
and their eff ects as tertiary brain damage (panel 2). In number of treatable patients.
conjunction with and beyond the commonly outlined
phases of acute and secondary cellular responses, such Mechanisms of damage
as cell death and metabolic disturbances (fi gure 1), this Gliosis
tertiary damage phase might persist for months or years Aberrant gliosis might cause tertiary brain damage
after the initiating injury. Cerebral palsy is defi ned by its after brain injury, even though coordinated and regulated
www.thelancet.com/neurology Vol 11 June 2012
glial responses are crucial in repair and regeneration.17
Evidence that glial activity is increased long after infl am-
Panel 1: The defi nition of cerebral palsy from the executive
mation or injury has come from studies in several
classifi cation committee1
species. In mice, microglial gene expression for pro-
"Cerebral palsy describes a group of disorders of the
infl ammatory cytokines was still increased 7 months
development of movement and posture, causing activity
after irradiation-induced infl ammation.18 In non-human
limitation, that are attributed to non-progressive
primates, prolonged microglial reactivity and increased
disturbances that occurred in the developing fetal or infant
cytokine expression have been noted 12 months after
brain. The motor disorders of cerebral palsy are often
traumatic brain injury (TBI).19 In human beings, PET
accompanied by disturbances of sensation, cognition,
imaging with a ligand against translocator protein (TSPO;
communication, perception, and/or behaviour, and/or by
panel 3)20,21 revealed that activation of microglia persists
a seizure disorder."
in the human brain for more than 10 years after TBI and
is associated with poor cognitive function.22 Also, in a
perinatal animal model of cerebral palsy, PET imaging
Panel 2: Tertiary brain damage
for activated microglia showed that severity of motor
defi cit was associated with TSPO binding in the short
term.23 Advances in imaging have been crucial for
Tertiary forms of brain damage are those that worsen
identifi cation of the persistence of injury processes, and
outcome, predispose a patient to further injury, or prevent
imaging of activated microglia could be refi ned into a
repair or regeneration after an initial insult to the brain
crucial diagnostic and monitoring aid in the future.
No long-term studies of microglial activation in animal
Mechanisms that facilitate this phase of damage include an
models of neonatal hypoxia-ischaemia or infl ammation
altered epigenome and ongoing infl ammation
or in human neonatal encephalopathy have been done.
However, a study undertaken in 7-year-old children with
cerebral palsy24 who had been born prematurely revealed
We suggest that these mechanisms are active processes for
increased tumour necrosis factor (TNF) concen-
which treatments can be developed, including treatments
trations in the plasma and supernatants of peripheral
that target microglia, innate immune processes, or
blood mononuclear cells after lipopolysaccharide
astrocytes; recapitulate permissive epigenetic conditions; or
stimulation. Because increased peripheral cytokines are
a marker for cerebral infl
mation, this fi nding
suggests that an altered infl ammatory response persists occur in perivascular macrophages (PVM). Located
for at least 7 years after brain damage in patients with within the neurovascular unit, PVM might infl uence BBB
cerebral palsy. This response could have deleterious function, including lymphocyte transmigration or
eff ects on clinical symptoms or later disease progression,
activation profi le, or both.28,29 Although the importance of
or both, and recognising and blocking such persistent BBB integrity and lymphocyte infi ltration in acute-phase
infl ammation could be of therapeutic value.
injury is known, further investigation into the subtle
After focal injury, reactive astrocytes are the primary changes that persist in the long term is warranted.
constituent cells of a glial scar. In the acute phase this
Boys suff er more severe brain damage due to perinatal
scar is protective, in part by forming a barrier where the brain injury than girls30 and also have poorer cognitive
blood-brain barrier (BBB) has been damaged, and thereby
outcome 1 year after paediatric TBI, even if the severity
preventing the uncontrolled access of blood proteins and of their injury is controlled.31 In-vitro astrocytes from
red blood cells into the brain. This protective astrocyte male neonatal mice that were stimulated with
response is initiated by activation of the TGF-/SMAD lipopolysaccharide expressed higher con centrations of
pathway by albumin.25 Although astrocytes are involved in
interleukin-6, TNF, and interleukin-1B than matched
regeneration and repair in the CNS, astrogliosis can female microglia, suggesting that sex might infl uence
persist for many years after injury and products of the eff ects of persistent gliosis.32 These diff erences
activated astrocytes inhibit axonal re
growth and might be relevant in vivo for several reasons. First,
remyelination.17 Microglial signalling might exacerbate interleukin-6 and TNF stimulate astrocyte proliferation
astrocyte cytotoxicity,26 further suggesting the need for and reduce neural precursor cell proliferation in
c immunomodulatory treatment to rodents,33,34 which might perpetuate the inhibitory
enable long-term recovery and regeneration. Astrocytes eff ects of the gliotic scar and delay repair in male
also regulate the bioavailability of neuro transmitters, and
rodents. Second, interleukin-1B sensitises the brain to
glutamate transporter 1 expression is decreased at least neurotoxicity35 and inhibits oligodendrocyte mat uration
21 days after hypoxia-ischaemia in neonatal rats,27 which in mice,9 increasing the severity of the acute and long-
might sensitise the brain to seizures10 and suggests a term eff
ects of any second-hit insult during
further mechanism by which the eff ects of injury persist infl ammation. Together with the immunomodulatory
in the brain. Persistent increased reactivity might also eff ects of sex hormones on outcome after damage,36,37
www.thelancet.com/neurology Vol 11 June 2012
cell proliferation, survival, migration, neuritogenesis,
synaptogenesis, and axonal regeneration.41 Similarly, the
chondroitin sulfate proteoglycans are crucial in neural
development because they are involved in complex
signalling with other extracellular matrix components.42
Chondroitin sulfate proteoglycans and IgCAMs are glial
Gliosis IgCAM and
scar components that inhibit regeneration of the injured
mouse spinal cord. The glial scar can persist for years after
injury, and expression of these adhesion molecules also
remains raised for weeks43 or months after injury.44 As
such, they might represent an active signalling mechanism
that inhibits regeneration and repair processes long after
the acute and secondary phases of injury.
Toll-like receptors (TLRs) are a family of pattern
recognition receptors. TLRs initiate innate immune
responses, including cytokine release, after activation by
damage-associated or pathogen-associated molecular
pattern molecules. Activation of TLRs in the brain causes
ammation, and cytokines produced because of
peripheral TLR activation signal via endothelial cells of
the neurovascular unit to induce central infl ammation.
Aberrant TLR activation during the perinatal period
sensitises the brain to further experimental injury in the
xia and ischaemia
acute infl ammatory phase.45,46 Additionally, in rodent
models perinatal TLR activation reprogrammes the adult
neuroimmune response, sensitising it to adult brain
infl ammation and injury, accelerating age-related cog-
Figure 1: Outline of the acute, secondary, and tertiary damage phases in cerebral palsy
nitive decline, and increasing susceptibility to seizures in
Activation of an array of receptors and ion channels leads to NF- activation, intrinsic and extrinsic apoptotic
cascades, and an increase in intracellular calcium, which in combination aff ect cell death, perpetuate infl ammation,
adults.11,47-50 Also, well into adulthood one exposure to the
and induce gliosis. Selected tertiary-phase mechanisms are also shown, including persisting eff ects on glia, the
TLR-4 agonist lipopolysaccharide in the neonatal mouse
epigenome, and neurogenesis. NF-=nuclear factor-. TLR=toll-like receptor. IgCAM=immunoglobulin cell
disrupts fate determination in cells of the subgranular
adhesion molecule. OPC=oligodendrocyte precursor cell.
zone (Smith LP, Naylor AS, Gothenburg University,
personal communication). Because neurogenic potential
diff erences in the endogenous cell-based infl ammatory is linked to the capacity for memory and learning,51
response might infl uence tertiary brain damage.
altered neurogenic potential probably represents a
Microglia contribute directly to the development of a tertiary mechanism by which the ability of the brain to
mature level of spontaneous pathway activity and plasticity
adapt and respond to later challenge is impaired.
by eliminating synapses.38 Microglia also participate in
Furthermore, arrest of oligodendrocyte maturation is
proliferation, causing phagocytosis of hundreds of new caused by activation of TLR-2 by hyaluronan, a product of
cells in the subgranular zone as part of the normal the glial scar.52 As such, persistent glial reactivity after brain
proliferative processes.39 Infl
ammation increases this injury (see Gliosis section) might be a mechanism by
phagocytic capacity, and concentrations of activated which repair and regeneration are impaired. Additionally,
microglia correlate with a reduction in hippocampal long-
accumulation of immature oligodendrocytes after perinatal
term potentiation in adult rats.40 Microglial interleukin-1B injury might sensitise the brain to further injury.53
plays an important part in memory acquisition, and after
neonatal infection hippocampal learning is impaired in Epigenetic changes
adult mice exposed to a second immune challenge.12 Thus,
Epigenetic modifi cations include enzymatic regulation
we know that microglial activation can persist long after of transcription by the modifi cation of permissive tags
brain injury,18,19,22 but we can only speculate that in the long
(acetylation, methylation, ubiquitination, phosphor-
term this activation also alters synaptic pruning, ylation, and sumoylation) on histones or DNA or via
neurogenesis, pathway excitability, and memory--key microRNA-mediated regulation of translation. In
features that predispose the brain to altered seizure human beings, acetylation regulates transcription of up
susceptibility and learning impairments.
to 5% of the genome, and every human gene can be
regulated by at least one microRNA.54-56 As such,
Infl ammatory receptor activation
epigenetic changes are, not surprisingly, crucial for
Immunoglobulin cell adhesion molecules (IgCAMs) have
every aspect of normal development and brain function.
diverse cell-dependent and context-dependent roles in Epigenetic modifi cations are a major mechanism by
www.thelancet.com/neurology Vol 11 June 2012
Figure 2: Schematic representation of tertiary phase damage
Changes in oligodendrocytes, maturational blockade, and the production of glial scar products leading to decreased myelination, changes to glia, astrogliosis and
microgliosis, and possible changes in EAA bioavailability are shown. Also shown is the NVU, a possible target for novel therapeutics because of its ability to mediate
transmigration of infl ammatory mediators. EAA= excitatory aminoacid. NVU= neurovascular unit.
which injury and destructive prenatal environmental
factors can lead to long-term disturbances of brain
Panel 3: Imaging microglia in vivo
development, such as cognitive, motor, or behavioural
The mitochondrial network-located translocator protein
(TSPO; 18 kDa) was previously known as the mitochondrial
During the acute and secondary phases of brain injury
peripheral benzodiazepine receptor (PBR). TSPO is located at
there is substantial loss of acetylation and methylation
the contact sites between the outer and inner mitochondrial
tags58,59 and considerable variation in microRNA
network membranes and is involved in cholesterol transport
expression.60,61 Reduced acetylation is associated with
and steroid synthesis. Activated microglia have an increased
cognitive decline,62,63 which is accelerated after brain
expression of TSPO, and various PET ligands for TSPO have
injury.64,65 Whether epigenetic changes persist after
enabled visualisation of microglial activation during
immature brain injury or in infl ammatory models in the
neurodegeneration and after injury.20,21
long term is unknown. However, prenatal stress increases
methylation and acetylation well into adulthood in
rodents and human beings, leading to substantial failure in further oligodendrocyte precursor cell (OPC)
changes in stress reactivity.66,67 Also, for at least 1 year maturation. These changes are key features of brain
after low-level radiation exposure in mice there is an damage in neonatal models and in preterm infants9,53,69,70
increase in concentrations of microRNA-21,68 an onco-
and typically cause substantial lasting cognitive
miRNA that stimulates cell growth and proliferation. impairments. OPCs are sensitive to injury because they
Another microRNA, miRNA-107 might contribute to the have a lower antioxidant capacity than other cells71 and
pathogenesis of Alzheimer's disease and is reduced after
possibly also because of increased death receptor
TBI, ischaemia, and hypoxia.60 Together, these fi ndings expression in these cells.72 However, the persisting
suggest that changes in epigenetic processes, which can active process of delayed oligodendrocyte maturation
disrupt neuro development, might also persist and cause might be linked to altered microRNA expression73,74 and
tertiary brain damage.
acetyl ation.9,75 In particular, oligo
dendrocyte diff er-
Changes to epigenetic processes might be particularly
entiation and maturation depend on a reduction in
relevant for white matter injury and the persistent acetylation during the transition from embryonic stem
www.thelancet.com/neurology Vol 11 June 2012
cell to OPC75 and the transition from OPC to myelin-
clinical trials, the combination of hypothermia and drugs
producing oligodendrocytes.76 We have recently such as xenon79,80 or melatonin.81 With these combinations
established a model of white matter injury in which the window of effi
cacy for therapeutic hypothermia can
chronic perinatal infl ammation, caused by interleukin-
be extended. For example, neuroprotection was still noted
1B exposure for the fi rst 5 days after birth, led to OPC when the antiepileptic drug topiramate was given before
maturational arrest that persists into adulthood and delayed therapeutic hypothermia in a rat model of
results in altered adult behaviour.9 In this model, global
neonatal hypoxia-ischaemia.82 The same group have since
histone deacetylase (HDAC) gene expression was shown that early phenobarbital treatment also enhances
altered. Expression was higher, contrary to what we the effi
cacy of delayed hypothermia.83
expected given that a reduction in acetylation facilitates
An alternative strategy to combination treatment would
the processes of oligiodendrocyte maturation. However,
be to use early but short-term hypothermia to increase
c and region-specifi
c acetylation and the window of opportunity for a protective drug. For
microRNA concentrations have not yet been assessed in
example, fructose-1,6-biphosphate (FBP) was neuro-
this model and we cannot exclude epigenetic changes as
protective against neonatal excitotoxic cortical damage in
a cause of OPC arrest.
mice;84 however, the drug had to be given within the fi rst
8 h to be neuroprotective. When a moderate but transient
Treatment options for tertiary brain damage
cooling was done immediately after the insult, FBP
Clinically, acute and secondary pathological processes of administered 24 h after the excitotoxic insult was still
brain injury in neonates can only be treated with signifi
cantly neuroprotective; thus the therapeutic
therapeutic hypothermia (see below). Although many window for FBP was extended.
alternative and additional treatments are being tested,
they have so far shown only limited safety and effi
cacy. Recovery of OPC maturation
These drugs and strategies have been thoroughly Oligodendrocyte maturation is dependent on appropriate
reviewed elsewhere.77 Long-term treatments for cerebral acetylation and microRNA regulation, suggesting that
palsy address the functional consequences of the causal the epigenome could be targeted to reduce the long-term
brain injury (eg, management of muscle tone and negative consequences of perinatal infl ammation or
physical and language therapy), which provide important
injury on white matter development. Also, the dependence
improvements in quality of life. However, on the basis of
of CD44 (an extracellular matrix receptor) activation in
the fi ndings presented above, we believe that active hyaluronan-mediated TLR-2 activation and oligo-
pathological processes persist in the months and years dendrocyte maturational arrest could be exploited to
after brain injury. As such, long-term treatment after reverse the persisting eff ects of the gliotic scar.85 Because
injury might improve neurological function. However, CD44 agonists also favour immunosuppressive Th2 type
possible interactions of treatment with the eff ects of T-cell expansion,86 targeting CD44 might represent a
developmental disturbances that arise from the initial multifaceted therapeutic strategy to treat brain damage,
brain injury will need to be considered in the choice and
including tertiary myelination defi cits. The glial scar has
timing of any treatment. As mentioned above (see been targeted with a peptide mimetic of IgCAM L1,
Gliosis), advances in glial imaging might be refi ned into which led to successful remyelination in an adult rat
a crucial diagnostic and monitoring aid that could be model of TBI.87 Another possible mechanism to stimulate
used in the development of treatments.
recovery of myelination might be to increase Axin2
expression, which overcomes OPC maturational arrest,88
Cell death and metabolic disturbance
as occurs in infants diagnosed with periventricular white
The long-term eff ects of brain damage stem from the matter damage.89
eff ects of exposure to various cytoactive and cytotoxic
substances during primary and secondary phases of Targeting long-lasting infl ammation
damage. As such, direct inhibition of primary and Given the wealth of evidence implicating persistent
secondary phases of damage would be the most effi
cient infl ammation in the development of tertiary brain
mechanisms to prevent tertiary damage. However, from damage, attempting to reduce ongoing infl ammation
a clinical point of view, the window of opportunity for seems a reasonable therapeutic strategy. However, co-
acute-phase treatments after brain injury is short and ordinated infl ammation can be a benefi cial process, and
drugs used in the secondary phase have limited effi
cacy. there is a complex relation between cytokine levels and
For example, therapeutic hypothermia must be initiated neurogenesis. Because neurogenesis is a key process in
within the fi rst 6 h of life to be protective in term infants
regeneration and repair, care must be taken with any
with neonatal encephalopathy.78 Such a short time approach that reduces infl ammation. Infl ammation,
window does not allow use of this treatment for all and in particular interleukin-6 expression, increases
neonates who might benefi t from it, and this treatment proliferation in the subventricular zone in adult mice,34
is eff ective for only one in every nine infants treated. As a
in direct opposition to the eff ects of infl ammation in the
result, diff erent groups are testing, in animal models and
subgranular zone, where proliferation decreases during
www.thelancet.com/neurology Vol 11 June 2012
infl ammation in the neonate and adult.90,91 This disparate
prevention of chronic infl ammation.98 An important
response might be due to diff erences in the cell caveat regarding increasing M2 signalling is that care
populations within these two neurogenic niches. That is,
must be taken to allow the requisite development of an
the subgranular zone harbours neural progenitor cells M1 response. M1-activated microglia are generally
whose expansion is depressed by interleukin-6, and the considered to be deleterious because of the neurotoxic
subventricular zone contains true neural stem cells eff ects of proinfl ammatory products on CNS injury and
proliferating in response to interleukin-6. Covey and in vitro.45 However, in a rat model of photothrombotic
colleagues34 further discussed the importance of their stroke, reduction of the acute-phase microglial response,
fi ndings when considering treatment options for patients
with antagonism of nuclear factor (NF-) via poly-
with brain injury induced by infl ammation or hypoxia-
ADP-ribose polymerase-1, reduced functional recovery
ischaemia. They suggest that enhancing the capacity for at 30 days after injury.99 Furthermore, ablation of acute-
multipotent stem cell proliferation might off er greater phase microglia exacerbated brain damage in models of
long-term benefi ts for recovery and repair than reducing neonatal and adult hypoxia-ischaemia.100,101
interleukin-6 concentrations to facilitate progenitor cell
Another important consideration in attempts to
proliferation. We can expect that, because of the enhance the M2 or anti-infl ammatory profi le of micro-
diff erences in levels of neurogenesis, these results will glia is the hypothesis that prolonged enhancement of
not directly extrapolate to neonates, but these fi ndings an inherent anti-infl ammatory response after perinatal
are still an exciting addition to our understanding of the infection might predispose patients to the development
complexity of treating neuroinfl ammation.
of a tertiary neuropsychiatric disorder: schizophrenia.102
With this hypothesis in mind, engineering obsolescence,
Modulation of the microglial phenotype
apoptosis, or inactivation into cellular vehicles that are
Overactivation of microglia is detrimental for the pro-
used as therapeutics, such as macrophages or microglia,
duction of hippocampal neurons.90 However, activation is a necessity.
of microglia and macrophages can also be benefi cial,
supporting neurogenesis and progenitor proliferation, Modulation of lymphocyte activation
survival, and diff erentiation in other brain regions.34
Phenotypic plasticity, as described above for microglia and
There is evidence that the phenotypes of macrophage macrophages, was fi rst described in lymphocytes, where a
and microglia can be damaging or capable of facilitating multitude of markers for activator or repressor lymphocytes
repair and regeneration, depending on stimulation and indicate their roles in a multifaceted set of innate and
at diff erent times, which explains in part these disparate adaptive immune responses. Maximum transmigration of
eff ects.92,93 Three activation states of microglia in the CNS
lymphocytes occurs in the fi rst few days after brain damage
have been proposed: classical activation (tissue defence but their phenotypes and role in repair and regeneration
ammatory signalling [M1]), alternative over the long term are not well understood. Regulatory
activation (repair, anti-infl ammatory signals, fi brosis, and type T lymphocytes reduced microglial activation and
extracellular matrix reconstruction [M2a]), and acquired disease progression in an animal model of Parkinson's
deactivation (immunosuppressive signalling and disease.103 Thus, persistent microglial acti vation might be
phagocytosis of apoptotic cells [M2b]).94
in part a product of dysfunctional lymphocyte activity after
Microglia and macrophage phenotypes are easily brain damage. Increasing the numbers of regulatory
modifi able in vitro, and macrophages, microglia, and T cells or reinstating lymphocyte homeostasis might be a
PVM share key features with stem cells that make them new treatment option to prevent neurodegenerative
exciting candidates for repair and regenerative treat-
diseases after brain damage.
ments, because they also migrate to the injury site and
A novel strategy to target lymphocyte activity might
produce trophic factors. Using these characteristics, cell be via PVM, which are responsible, at least in part, for
therapy with M2-induced (alternative activation and mediating the transmigration of lymphocytes into the
acquired deactivation) macrophages reduced disease brain during infl ammation.104 Overexpression of C-X-C
severity in an animal model of multiple sclerosis95 and motif chemokine 10 and underexpression of matrix
after adult ischaemia in gerbils.96 These fi ndings suggest metalloproteinase 9 can cause lymphocytes to remain
that M2-activated microglia are a potential neurotherapy,
in the perivascular space after migration from the
but this will need to be validated in the developing brain.
bloodstream.105 The production and release of cytoactive
Alternatively, or in parallel, strategies aimed at substances by PVM can be modifi ed because of their
supporting the endogenous development of the constitutively phagocytotic characteristics: they readily
M2 phenotype would also be of major interest. Early absorb liposomes, plasmids, siRNA, or naked oligo-
increases in expression of the M2-inducing cytokine nucleotides. By modifying in-vivo cytoactive signals
interleukin-10 are associated with improved outcome in from PVM, lymphocyte transmigration or the
adult stroke,97 and increasing the concentrations of the phenotype of transmigrating cells, or both, could
M2 cytokines interleukin-13 and interleukin-4 causes be altered to favour regulatory and regenerative
apoptosis of activated microglia in a process crucial for lymphocyte activity.
www.thelancet.com/neurology Vol 11 June 2012
Epigenetic modifi cation
We suggest that persistent eff ects of injury on the
In adult mice, adverse behavioural and neuroendocrine immature brain, including infl ammation, will decrease
defi cits occur after early life stress. These defects can be this plasticity by inhibiting endogenous repair. Use of
abolished in the adult mouse by central administration of growth factors and cell therapy to support endogenous
the HDAC inhibitor trichostatin-A, which also promotes plasticity after a lesion (see below) is being trialled
demethylation.56 Additionally, HDAC inhibi tors modulate clinically and seems to be a promising delayed inter-
innate and adaptive immune responses to enhance vention for improvement of long-term neurological and
Th2 responses to TLR4 activation.54 HDAC inhibitors cognitive function.
rapidly restore the loss of acetylation associated with in-
fl ammation and hypoxia-ischaemia and improve the patho-
logical and behavioural outcomes in these injury models Persistent positive eff ects of neural or mesenchymal
when administered during the acute phase.59,106 Given the stem cell therapy on lesion size or cognitive or motor
cacy of HDAC inhibitors at reducing microglial outcome have been noted after adult and perinatal brain
activity107 and the ongoing role of microgliosis in tertiary injury.113-116 These studies have key exciting features.
brain damage, delayed treatment might also be benefi cial. Implanted cells can migrate to the lesion site when
However, when considering this type of therapeutic implanted contralateral to the lesion or when injected
intervention, the importance of HDAC activity for normal peripherally.114,116 Furthermore, a regenerative microglial
development, including proliferation, diff erentiation, and phenotype is stimulated by treatment,115 and recovery
myelination by oligodendrocytes, must not be forgotten.108
can occur independent of reductions in cell loss,115,116
That said, oligodendrocyte maturation and myelination suggesting that there is an extended therapeutic window.
were not altered when modest neuroprotection was One dose of intravenous stem cell therapy given to adult
provided by peripheral trichostatin A in a neonatal model rats 30 days after middle cerebral artery occlusion
of lipopolysaccharide-sensitised hypoxia-ischaemia.109
improved neurobehavioural outcome, in
MicroRNAs are involved in the pathogenesis of acute-
synaptogenesis, vasculogenesis, and prolif eration, and
phase injury and in endogenous repair and regeneration.
decreased apoptosis.116 In this so-called super-delayed
At 3 days after middle cerebral artery occlusion, the cell therapy regimen, even when one dose of stem cells
expression of microRNAs that are involved in regulation was administered 90 days after middle cerebral artery
of T-cell function (miRNA-17 and miRNA-20a) and occlusion, sensorimotor testing was improved 50 days
intracellular ATP concentrations (miRNA-15b) was later; although this improvement did not extend across
increased,110 suggesting a specifi c temporal role for the entire neurobehavioural battery. This fi nding
microRNAs in endogenous repair and regeneration. suggests that the development of delayed therapy,
Modifi cation of microRNA activity with siRNA reduces possibly via expansion of autologous stem cells for
liver and cardiac injury without aff ecting endogenous repetitive treatments, might be an exciting next step for
microRNA production,60 and miRNA-145 inhibition treatment of tertiary brain damage by inducing repair
improved neuropathological outcome after ischaemia in and regeneration.
a mouse model,110 thus showing the potential of
A further intriguing alternative to treatment with stem
microRNA regulation for therapy. miRNA-210 is a pro-
cells is to stimulate the production of endogenous
survival factor that stimulates cell growth and increases neuronal stem cells or to redirect new cell production
vascular endothelial growth factor (VEGF) signalling, from astroglia to oligodendrocytes and neurons117 after
which is key in long-term regeneration. miRNA-210 was neurodegenerative insult. Stem cells accumulate in the
reportedly increased in multiple models of hypoxia that subventricular zone after an acute brain lesion but we do
were tested and in placenta from patients with pre-
not yet understand how to ensure these cells participate
eclampsia;61 therefore, this might be a therapeutic target.
in repair and regeneration. However, because self-
miRNA-124 is expressed only by microglia in the brain renewal, proliferative capacity, and multipotency are
and is involved in the regulation of microglial phenotypes.
common to stem cells and cancer, the theoretical risk of
Specifi cally, miRNA-124 decreases expres sion of major tumour development must be considered.
histocompatibility complex 2,111 and loss of miRNA-124
Although clinical interventions directed at the
preceded the onset of disease in experimental underlying neurobiological process are limited, trials of
autoimmune encephalomyelitis.111 In the search for autologous umbilical-cord-derived stem cells are ongoing
strategies to improve brain function long after at Duke University (Durham, NC, USA; NCT00593242)
infl ammation and injury, that central administration of and the Medical College of Georgia (Augusta, GA, USA;
miRNA-124 can reverse progression of experimental NCT01072370). At Duke University, multiple stem cell
autoimmune encephalomyelitis is of interest.111
infusions are to be administered within the fi rst
14 postnatal days to infants diagnosed with hypoxic-
Promoting positive brain plasticity after a lesion
ischaemic encephalopathy. At the Medical College of
After brain injury in human beings, functional im-
Georgia, patients with cerebral palsy are to be
provements can accrue over many months or years.112
administered one treatment of stem cells in early
www.thelancet.com/neurology Vol 11 June 2012
childhood in a double-blind experimental design.
There are positive anecdotal reports from the Duke
Search strategy and selection criteria
University study, which is an open-label pilot study, but
References for this Review were found through a search of
no peer-reviewed data are yet available.
PubMed from 1951 to March, 2012, and from a search of
Google Scholar from 1992 to March, 2012, with the terms
"perinatal", "brain", "cerebral", "hypoxic/ischemic",
Promotion of positive plasticity after a lesion with
"long-term", "delayed", "sensitisation", and "neuroprotection"
pharmacological drugs has the clinical benefi ts of
in various combinations. Reports in English and French were
reproducibility and does not require the previous banking
considered. Additional searches were done as needed for
of stem cells or the use of immunomodulatory therapies
specifi c topics from the initial screen. With the large number
to inhibit allogeneic cell rejection. Growth factors reduce
of citations retrieved through these search strategies, the
delayed neuronal death,118-121 and treatment with growth
published work cited inevitably represents the authors'
factors also seems to be a promising strategy to prevent
personal selection and was also limited by restrictions on
tertiary damage related to aberrant infl am matory
processes, which decrease growth factor produc
and inhibit neurite outgrowth and remyelination.17
Furthermore, in fl ammation can decrease neurogenesis,
Erythropoietin, when given on average 24 h after
and this eff ect, which is associated with defi cits in birth, had a signifi cant neuroprotective eff ect in human
memory and learning,51 can be reversed with growth term infants with neonatal encephalopathy.129 The
timing of intervention argues in favour of an eff ect of
Treatment with inhibitory growth factor agonists, such
erythro poietin on plasticity after a lesion, although a
as neurite growth-inhibitory protein, beginning 1 week direct eff ect on delayed neuronal cell death cannot be
after stroke, improved cortical regeneration in adult excluded.
rats.123 Also, delayed transfer to an enriched environment,
beginning 15 days after middle cerebral artery occlusion,
which is associated with increased BDNF expression, A great deal of evidence from neonatal and adult injury
improved neurobehavioural scoring in adult rats,124 and studies suggests that an ongoing and theoretically
VEGF treatment delayed to 24 h after middle cerebral treatable tertiary damage phase exists after perinatal
artery occlusion still improved long-term infarct volume brain insult. Persistent neurological defi cits after
and neurological function.125
perinatal brain injury remain an enormous societal
Although growth factors such as BDNF are large mol-
burden despite substantial research eff orts. We suggest
ecules that are unlikely to cross the intact BBB easily, that advances in imaging and knowledge of glial
ampa kines--allosteric positive modulators of glutamat-
activation states and the mechanisms that underpin
ergic AMPA receptors--are small and diff usible white matter injury mean that we can now consider
molecules that can induce BDNF production in the brain
treatment of damage in the tertiary phase. Translation of
when administered systemically. Ampakines mimicked treatment options into clinically applicable therapies
the eff ects of BDNF on axonal sprouting in a mouse will be a complex task. However, we hope a greater
model of excitotoxic white matter injury.121
understanding of the persistence of active injury
Similarly, melatonin promoted plasticity in the same mechanisms will eventually lead to improved brain
model of neonatal excitotoxic white matter damage.126
health in infants with brain injury.
Although melatonin did not prevent the initial appearance
of white matter damage, it promoted repair with axonal BF did the search of the published work, prepared the fi gures, and wrote
regrowth, sprouting, or both. Recent data have shown and revised the Review. PG contributed to the search of the published
that the window for intervention is at least 24 h after work and the writing and revision of the Review.
the insult (Gressens P, personal communication). Confl icts of interest
We declare that we have no confl icts of interest.
Behavioural studies support the hypothesis that
melatonin-induced white matter repair is accompanied Acknowledgments
The authors' research is funded by the Wellcome Trust (Programme
by improved learning capabilities. Neuroprotective Grant WT094823MA), INSERM, Universite Paris 7, APHP (Contrat
properties of melatonin have been confi rmed in several Hospitalier de Recherche Translationnelle to PG), Fondation Leducq,
animal models of perinatal brain damage, including in Fondation Grace de Monaco, Fondation Roger de Spoelberch, PremUP,
fetal sheep after umbilical cord occlusion.127 Melatonin is and Seventh Framework Program of the European Union (grant
agreement number HEALTH-F2-2009-241778/NEUROBID).
a safe compound to administer to newborns,128 and it
crosses the BBB as well as the placenta. On the basis of References
Bax M, Goldstein M, Rosenbaum P, et al. Proposed defi nition and
these data, a clinical trial testing the neuroprotective
classifi cation of cerebral palsy, April 2005. Dev Med Child Neurol
eff ects of melatonin has been initiated in preterm infants
2005; 47: 571-76.
at high risk of developing brain damage and neurological
Blair E, Stanley FJ. Intrapartum asphyxia: a rare cause of cerebral
palsy. J Pediatr 1988; 112: 515-19.
www.thelancet.com/neurology Vol 11 June 2012
Graham EM, Ruis KA, Hartman AL, Northington FJ, Fox HE.
28 Hawkes CA, McLaurin J. Selective targeting of perivascular
A systematic review of the role of intrapartum hypoxia-ischemia in
macrophages for clearance of beta-amyloid in cerebral amyloid
the causation of neonatal encephalopathy. Am J Obstet Gynecol 2008;
angiopathy. Proc Natl Acad Sci USA 2009; 106: 1261-66.
29 Guillemin GJ, Brew BJ. Microglia, macrophages, perivascular
Moreno-De-Luca A, Ledbetter DH, Martin CL. Genomic insights
macrophages, and pericytes: a review of function and identifi cation.
into the causes and classifi cation of the cerebral palsies.
J Leukoc Biol 2004; 75: 388-97.
Lancet Neurol 2012; 11: 283-92.
30 Tioseco JA, Aly H, Essers J, Patel K, El-Mohandes AA. Male sex and
Keogh JM, Badawi N. The origins of cerebral palsy.
intraventricular hemorrhage. Pediatr Crit Care Med 2006; 7: 40-44.
Curr Opin Neurol 2006; 19: 129-34.
31 Donders J, Hoff man NM. Gender diff erences in learning and
Nelson KB. Causative factors in cerebral palsy. Clin Obstet Gynecol
memory after pediatric traumatic brain injury. Neuropsychology
2008; 51: 749-62.
2002; 16: 491-99.
Dammann O. Persistent neuro-infl ammation in cerebral palsy:
32 Mouton PR, Long JM, Lei DL, et al. Age and gender eff ects on
a therapeutic window of opportunity? Acta Paediatr 2007; 96: 6-7.
microglia and astrocyte numbers in brains of mice. Brain Res 2002;
Leviton A, Gressens P. Neuronal damage accompanies perinatal
white-matter damage. Trends Neurosci 2007; 30: 473-78.
33 Santos-Galindo M, Acaz-Fonseca E, Bellini MJ, Garcia-Segura LM.
Favrais G, van de Looij Y, Fleiss B, et al. Systemic infl ammation
Sex diff erences in the infl ammatory response of primary astrocytes
disrupts the developmental program of white matter. Ann Neurol
to lipopolysaccharide. Biol Sex Diff er 2011; 2: 7.
2011; 70: 550-65.
34 Covey MV, Loporchio D, Buono KD, Levison SW. Opposite eff ect of
10 Galic MA, Riazi K, Heida JG, et al. Postnatal infl ammation
infl ammation on subventricular zone versus hippocampal
increases seizure susceptibility in adult rats. J Neurosci 2008;
precursors in brain injury. Ann Neurol 2011; 70: 616-26.
35 Favrais G, Schwendimann L, Gressens P, Lelievre V.
11 Spencer SJ, Auer RN, Pittman Q J. Rat neonatal immune challenge
Cyclooxygenase-2 mediates the sensitizing eff ects of systemic
alters adult responses to cerebral ischaemia.
interleukin-1-beta on excitotoxic brain lesions in newborn mice.
J Cereb Blood Flow Metab 2006; 26: 456-67.
Neurobiol Dis 2007; 25: 496-505.
12 Williamson LL, Sholar PW, Mistry RS, Smith SH, Bilbo SD.
36 Brown CM, Suzuki S, Jelks KA, Wise PM. Estradiol is a potent
Microglia and memory: modulation by early-life infection.
protective, restorative, and trophic factor after brain injury.
J Neurosci 2011; 31: 15511-21.
Semin Reprod Med 2009; 27: 240-49.
13 Pulsinelli WA, Brierley JB, Plum F. Temporal profi le of neuronal
37 Gibson CL, Gray LJ, Bath PM, Murphy SP. Progesterone for the
damage in a model of transient forebrain ischemia. Ann Neurol
treatment of experimental brain injury: a systematic review.
1982; 11: 491-98.
Brain 2008; 131: 318-28.
14 Kirino T, Tamura A, Sano K. Delayed neuronal death in the rat
38 Chung WS, Barres BA. The role of glial cells in synapse
hippocampus following transient forebrain ischemia.
elimination. Curr Opin Neurobiol 2011; published online Oct 27.
Acta Neuropathol (Berl) 1984; 64: 139-47.
15 Pulsinelli WA, Jacewicz M, Levy DE, Petito CK, Plum F. Ischemic
39 Sierra A, Encinas JM, Deudero JJ, et al. Microglia shape adult
brain injury and the therapeutic window. Ann NY Acad Sci 1997;
hippocampal neurogenesis through apoptosis-coupled
phagocytosis. Cell Stem Cell 2010; 7: 483-95.
16 Gluckman PD, Wyatt JS, Azzopardi D, et al. Selective head cooling
n R, Nally R, Nolan Y, McCartney Y, Linden J, Lynch MA.
with mild systemic hypothermia after neonatal encephalopathy:
The age-related attenuation in long-term potentiation is associated
multicentre randomised trial. Lancet 2005; 365: 663-70.
with microglial activation. J Neurochem 2006; 99: 1263-72.
17 Sofroniew MV. Molecular dissection of reactive astrogliosis and glial
41 Lavdas AA, Papastefanaki F, Thomaidou D, Matsas R. Cell adhesion
scar formation. Trends Neurosci 2009; 32: 638-47.
molecules in gene and cell therapy approaches for nervous system
18 Mildner A, Schlevogt B, Kierdorf K, et al. Distinct and
repair. Curr Gene Ther 2011; 11: 90-100.
non-redundant roles of microglia and myeloid subsets in mouse
42 Maeda N, Ishii M, Nishimura K, Kamimura K. Functions of
models of Alzheimer's disease. J Neurosci 2011; 31: 11159-71.
chondroitin sulfate and heparan sulfate in the developing brain.
19 Nagamoto-Combs K, McNeal DW, Morecraft RJ, Combs CK.
Neurochem Res 2011; 36: 1228-40.
Prolonged microgliosis in the rhesus monkey central nervous
43 Jakovcevski I, Wu J, Karl N, et al. Glial scar expression of CHL1,
system after traumatic brain injury. J Neurotrauma 2007; 24: 1719-42.
the close homolog of the adhesion molecule L1, limits recovery after
20 Venneti S, Lopresti BJ, Wiley CA. The peripheral benzodiazepine
spinal cord injury. J Neurosci 2007; 27: 7222-33.
receptor (translocator protein 18kDa) in microglia: from pathology
44 Tang X, Davies JE, Davies SJ. Changes in distribution, cell
to imaging. Prog Neurobiol 2006; 80: 308-22.
associations, and protein expression levels of NG2, neurocan,
21 Rupprecht R, Papadopoulos V, Rammes G, et al. Translocator
phosphacan, brevican, versican V2, and tenascin-C during acute to
protein (18 kDa) (TSPO) as a therapeutic target for neurological and
chronic maturation of spinal cord scar tissue. J Neurosci Res 2003;
psychiatric disorders. Nat Rev Drug Discov 2010; 9: 971-88.
22 Ramlackhansingh AF, Brooks DJ, Greenwood RJ, et al.
45 Aden U, Favrais G, Plaisant F, et al. Systemic infl ammation
Infl ammation after trauma: microglial activation and traumatic
sensitizes the neonatal brain to excitotoxicity through a pro-/
brain injury. Ann Neurol 2011; 70: 374-83.
anti-infl ammatory imbalance: key role of TNFalpha pathway and
protection by etanercept. Brain Behav Immun 2010; 24: 747-58.
23 Kannan S, Saadani-Makki F, Balakrishnan B, et al. Magnitude of
[(11)C]PK11195 binding is related to severity of motor defi cits in a
46 Wang X, Stridh L, Li W, et al. Lipopolysaccharide sensitizes neonatal
rabbit model of cerebral palsy induced by intrauterine endotoxin
hypoxic-ischemic brain injury in a MyD88-dependent manner.
exposure. Dev Neurosci 2011; 33: 231-40.
J Immunol 2009; 183: 7471-77.
24 Lin CY, Chang YC, Wang ST, Lee TY, Lin CF, Huang CC. Altered
47 Boisse L, Mouihate A, Ellis S, Pittman Q J. Long-term alterations in
infl ammatory responses in preterm children with cerebral palsy.
neuroimmune responses after neonatal exposure to
Ann Neurol 2010; 68: 204-12.
lipopolysaccharide. J Neurosci 2004; 24: 4928-34.
25 Ralay Ranaivo H, Patel F, Wainwright MS. Albumin activates the
48 Ellis S, Mouihate A, Pittman Q J. Neonatal programming of the rat
canonical TGF receptor-SMAD signaling pathway but this is not
neuroimmune response: stimulus specifi c changes elicited by
required for activation of astrocytes. Exp Neurol 2010; 226: 310-19.
bacterial and viral mimetics. J Physiol 2006; 571: 695-701.
26 Bezzi P, Domercq M, Brambilla L, et al. CXCR4-activated astrocyte
49 Bilbo SD. Early-life infection is a vulnerability factor for
glutamate release via TNF alpha: amplifi cation by microglia triggers
aging-related glial alterations and cognitive decline.
neurotoxicity. Nat Neurosci 2001; 4: 702-10.
Neurobiol Learn Mem 2010; 94: 57-64.
27 Bruhn T, Levy LM, Nielsen M, Christensen T, Johansen FF,
50 Wang X, Hagberg H, Nie C, Zhu C, Ikeda T, Mallard C. Dual role of
Diemer NH. Ischemia induced changes in expression of the
intrauterine immune challenge on neonatal and adult brain
astrocyte glutamate transporter GLT1 in hippocampus of the rat.
vulnerability to hypoxia-ischemia. J Neuropathol Exp Neurol 2007;
Neurochem Int 2000; 37: 277-85.
www.thelancet.com/neurology Vol 11 June 2012
51 Dobrossy MD, Drapeau E, Aurousseau C, Le Moal M, Piazza PV,
73 Dugas JC, Cuellar TL, Scholze A, et al. Dicer1 and miR-219 are
Abrous DN. Diff erential eff ects of learning on neurogenesis:
required for normal oligodendrocyte diff erentiation and
learning increases or decreases the number of newly born cells
myelination. Neuron 2010; 65: 597-611.
depending on their birth date. Mol Psychiatry 2003; 8: 974-82.
74 Zhao X, He X, Han X, et al. MicroRNA-mediated control of
52 Sloane JA, Batt C, Ma Y, Harris ZM, Trapp B, Vartanian T.
oligodendrocyte diff erentiation. Neuron 2010; 65: 612-26.
Hyaluronan blocks oligodendrocyte progenitor maturation and
75 Liu A, Han YR, Li J, et al. The glial or neuronal fate choice of
remyelination through TLR2. Proc Natl Acad Sci USA 2010;
oligodendrocyte progenitors is modulated by their ability to acquire
an epigenetic memory. J Neurosci 2007; 27: 7339-43.
53 Segovia KN, McClure M, Moravec M, et al. Arrested
76 Ye F, Chen Y, Hoang T, et al. HDAC1 and HDAC2 regulate
oligodendrocyte lineage maturation in chronic perinatal white
oligodendrocyte diff erentiation by disrupting the beta-catenin-TCF
matter injury. Ann Neurol 2008; 63: 520-30.
interaction. Nat Neurosci 2009; 12: 829-38.
54 Brogdon JL, Xu Y, Szabo SJ, et al. Histone deacetylase activities are
77 Johnston MV, Fatemi A, Wilson MA, Northington F. Treatment
required for innate immune cell control of Th1 but not Th2 eff ector
advances in neonatal neuroprotection and neurointensive care.
cell function. Blood 2007; 109: 1123-30.
Lancet Neurol 2011; 10: 372-82.
55 Friedman RC, Farh KK, Burge CB, Bartel DP. Most mammalian
78 Thoresen M. Cooling the newborn after asphyxia--physiological
mRNAs are conserved targets of microRNAs. Genome Res 2009;
and experimental background and its clinical use. Semin Neonatol
2000; 5: 61-73.
56 Weaver IC, Champagne FA, Brown SE, et al. Reversal of maternal
79 Faulkner S, Bainbridge A, Kato T, et al. Xenon augmented
programming of stress responses in adult off spring through methyl
hypothermia reduces early lactate/N-acetylaspartate and cell death
supplementation: altering epigenetic marking later in life.
in perinatal asphyxia. Ann Neurol 2011; 70: 133-50.
J Neurosci 2005; 25: 11045-54.
80 Ma D, Hossain M, Chow A, et al. Xenon and hypothermia combine
57 Ozanne SE, Constancia M. Mechanisms of disease: the
to provide neuroprotection from neonatal asphyxia. Ann Neurol
developmental origins of disease and the role of the epigenotype.
2005; 58: 182-93.
Nat Clin Pract Endocrinol Metab 2007; 3: 539-46.
81 Kelen D, Robertson NJ. Experimental treatments for hypoxic
58 Sinn DI, Kim SJ, Chu K, et al. Valproic acid-mediated
ischaemic encephalopathy. Early Hum Dev 2010; 86: 369-77.
neuroprotection in intracerebral hemorrhage via histone
82 Liu Y, Barks JD, Xu G, Silverstein FS. Topiramate extends the
deacetylase inhibition and transcriptional activation. Neurobiol Dis
therapeutic window for hypothermia-mediated neuroprotection
2007; 26: 464-72.
after stroke in neonatal rats. Stroke 2004; 35: 1460-65.
59 Faraco G, Pancani T, Formentini L, et al. Pharmacological inhibition
83 Barks JD, Liu YQ, Shangguan Y, Silverstein FS. Phenobarbital
of histone deacetylases by suberoylanilide hydroxamic acid
augments hypothermic neuroprotection. Pediatr Res 2010; 67: 532-37.
specifi cally alters gene expression and reduces ischemic injury in
84 Rogido M, Husson I, Bonnier C, et al. Fructose-1,6-biphosphate
the mouse brain. Mol Pharmacol 2006; 70: 1876-84.
prevents excitotoxic neuronal cell death in the neonatal mouse
60 Madathil SK, Nelson PT, Saatman KE, Wilfred BR. MicroRNAs in
brain. Brain Res Dev Brain Res 2003; 140: 287-97.
CNS injury: potential roles and therapeutic implications. Bioessays
85 Back SA, Tuohy TM, Chen H, et al. Hyaluronan accumulates in
2011; 33: 21-26.
demyelinated lesions and inhibits oligodendrocyte progenitor
61 Fasanaro P, Greco S, Ivan M, Capogrossi MC, Martelli F.
maturation. Nat Med 2005; 11: 966-72.
MicroRNA: emerging therapeutic targets in acute ischemic
86 Baaten BJ, Li CR, Deiro MF, Lin MM, Linton PJ, Bradley LM. CD44
diseases. Pharmacol Ther 2010; 125: 92-104.
regulates survival and memory development in Th1 cells. Immunity
62 Peleg S, Sananbenesi F, Zovoilis A, et al. Altered histone acetylation
2010; 32: 104-15.
is associated with age-dependent memory impairment in mice.
87 Klementiev B, Novikova T, Korshunova I, Berezin V, Bock E.
Science 2010; 328: 753-56.
The NCAM-derived P2 peptide facilitates recovery of cognitive and
63 Shen S, Liu A, Li J, Wolubah C, Casaccia-Bonnefi l P. Epigenetic
motor function and ameliorates neuropathology following
memory loss in aging oligodendrocytes in the corpus callosum.
traumatic brain injury. Eur J Neurosci 2008; 27: 2885-96.
Neurobiol Aging 2008; 29: 452-63.
88 Fancy SP, Harrington EP, Yuen TJ, et al. Axin2 as regulatory and
64 Dash PK, Orsi SA, Moore AN. Histone deactylase inhibition
therapeutic target in newborn brain injury and remyelination.
combined with behavioral therapy enhances learning and memory
Nat Neurosci 2011; 14: 1009-16.
following traumatic brain injury. Neuroscience 2009; 163: 1-8.
89 Buser JR, Maire J, Riddle A, et al. Arrested preoligodendrocyte
65 Perry VH. Contribution of systemic infl ammation to chronic
maturation contributes to myelination failure in premature infants.