Review

Innate and adaptive immune responses in the CNS

Ari Waisman, Roland S Liblau, Burkhard Becher

Almost every disorder of the CNS is said to have an inflammatory component, but the precise nature of inflammation Lancet Neurol 2015; 14: 945–55
in the CNS is often imprecisely defined, and the role of CNS-resident cells is uncertain compared with that of cells Institute for Molecular
that invade the tissue from the systemic immune compartment. To understand inflammation in the CNS, the term Medicine, University Medical
Center of the Johannes
must be better defined, and the response of tissue to disturbances in homoeostasis (eg, neurodegenerative processes)
Gutenberg University, Mainz,
should be distinguished from disorders in which aberrant immune responses lead to CNS dysfunction and tissue Germany (Prof A Waisman PhD);
destruction (eg, autoimmunity). Whether the inflammatory tissue response to injury is reparative or degenerative Centre de Physiopathologie
seems to be dependent on context and timing, as are the windows of opportunity for therapeutic intervention in Toulouse-Purpan, Université
Toulouse 3, Toulouse, France
inflammatory CNS diseases.
(Prof R S Liblau MD); and
Institute of Experimental
Introduction We classify CNS disorders as innate or adaptive immune- Immunology, University of
The role of inflammation in various CNS diseases is mediated disorders, with specific examples of each Zurich, Zurich, Switzerland
(Prof B Becher PhD)
debated. Although the role of the immune system in mechanism. We will also discuss innate inflammatory
Correspondence to:
disturbed tissue homoeostasis is becoming better responses to pathological CNS disorders and compare
Prof Ari Waisman, Institute for
understood, inflammation in the CNS is thought to be them with antigen-driven autoimmune disorders, in Molecular Medicine,
different from that in the periphery. The CNS is often which T cells and B cells (panel 1) orchestrate an attack Universitätsmedizin der
described as an immune-privileged site, and evidence against CNS autoantigens. Johannes Gutenberg-Universität,
Obere Zahlbacher Strasse,
supports the notion that the CNS receives limited
55131 Mainz, Germany
immune surveillance by peripheral lymphocytes (panel 1) Inflammation and innate versus adaptive waisman@uni-mainz.de
under physiological conditions:1 the blood–brain barrier immunity
limits the movement of cells and macromolecules The term inflammation, from the Latin verb inflammare
between the blood and CNS tissue; there was thought to (to burn), is not actually synonymous with infection,
be no professional antigen-presenting cells in the CNS, although infection is often the cause of inflammation. To
and little expression of MHC molecules, limiting antigen understand inflammation, it is important to distinguish
recognition by invading T lymphocytes; and the immune innate and adaptive immune responses. There are many
response to allografts implanted into the CNS is delayed definitions of the innate and adaptive aspects of
compared with those in the periphery. immunity involved in the formation of an immune
The notion that the CNS must be immune privileged, response, and these components are not separate but are
hiding behind a selectively permeable barrier, has now functionally intertwined. Innate immunity is a feature of
been revised. Although there are no antigen- most life forms, from plants and fungi to invertebrate
presenting cells within the CNS parenchyma (panel 1),2 and vertebrate animals.8 The innate immune system
these cells exist in association with blood vessels in the components, such as tissue-resident macrophages
CNS.2–4 Neural-derived antigens are reportedly released (panel 1), dendritic cells, mast cells, circulating
from the CNS and are subsequently detected in CNS- phagocytes, and complement molecules (panel 1),
draining cervical lymph nodes. Hence, immune represent the first line of defence against invading
responses to CNS antigens or pathogens in the CNS can pathogens or malignancy. Adaptive immune responses
be mounted in these dedicated secondary lymphoid exist only in more complex vertebrates and involve a
structures.5,6 More importantly, if immune privilege were specialised leucocyte (panel 1) population that uses
absolute, why would immunosuppressed patients variable cell-surface antigen receptors generated by
develop disorders such as primary CNS lymphoma or somatic recombination or gene-conversion events.9 In
progressive multifocal leukoencephalopathy, a usually mammals, only B cells and T cells are capable of
fatal viral disease characterised by progressive damage of generating specialised antigen receptors to interact with
white matter, when the virus can multiply in the absence microbial pathogens or eliminate tumours. This highly
of control by CNS lymphocytes? These occurrences evolved immune system is, however, also the cause of
suggest that under physiological conditions, the CNS is several autoimmune diseases.
under close surveillance by the immune system, and The innate and adaptive aspects of inflammation in
pathogens or aberrant cells in the CNS are well controlled.7 CNS diseases are easily distinguished. Part of the innate
In this Review, we show that the CNS is not only an immune defence against pathogens entering the CNS is
immune competent organ, closely interacting with the the blood–brain barrier. This selectively permeable
systemic immune compartment under physiological barrier is formed by capillary endothelial cells connected
conditions, but also that almost all pathological changes by tight junctions along most CNS capillaries; some
within the CNS elicit a prominent inflammatory reaction. regions of the brain, such as the circumventricular
We aim to define and classify the most prominent organs, have widespread vasculature but no healthy
features of inflammation in the context of CNS diseases. blood–brain barrier.10–12

www.thelancet.com/neurology Vol 14 September 2015 945

 Review
Panel 1: Glossary
Amoeboid microglia
Microglia that have retracted their long and branched cellular
processes (present during normal function), thus developing an
amoeboid appearance, in response to pathological changes and
tissue damage.
Astrocytes
Glial cells of neuroectodermal origin that are dispersed
throughout the CNS. They respond to pathological changes
with hypertrophy and hyperplasia (called reactive astrogliosis).
B cells
Immune cells that arise from the bone marrow and mature in
the spleen. B-cell receptors can be secreted in the form of
soluble immunoglobulins.
Complement system
Part of the immune system that helps or complements the
ability of antibodies and phagocytic cells to clear pathogens
from an organism.
Gliosis
Reactive change of glial cells in response to damage or change
in tissue homoeostasis, usually involving hypertrophy and
proliferation.
HLA
The HLA locus encompasses genes that encode HLA class I
molecules, which present antigenic peptides to CD8 T cells, and
HLA class II molecules, which present antigenic peptides to CD4
T cells.
Leucocytes
All white blood cells.
Lymphocytes
T cells, B cells, and innate lymphoid cells such as natural killer cells.
Macrophages
Tissue-resident myeloid cells. In the CNS, most macrophages
are called microglia and differ from perivascular macrophages in
morphology, function, and ontogeny. By contrast with
microglia, perivascular macrophages derive from monocytes.
Microglia
Glial cells derived from primitive yolk-sac macrophages that
seed the CNS early during embryogenesis. They bring about
synaptic pruning and respond to damage in the CNS.
In the CNS parenchyma, two additional cell types are
actively involved in immune defence: astrocytes and
microglia, the CNS-resident macrophages (panel 1).
Astrocytes are large, ramified cells of neuroectodermal
origin that share a developmental pathway with neurons
and oligodendrocytes (panel 1)13 and are thought to have
several functions in the CNS, including structural
support, metabolism, synaptic plasticity, and immune
defence.14 Astrocytes are also thought to be crucial for the
946
Monocytes
Bone-marrow-derived myeloid cells. Monocytes carry only
germ-line-encoded antigen receptors and invade the
inflamed tissues. They are a potent source of free radicals and
cytokines. Under pathological conditions, most CNS-
infiltrating immune cells belong to this category and are
usually recruited to sites of inflammation to become
inflammatory phagocytes.
Myeloid cells
All non-lymphocytic leucocytes including monocytes,
macrophages, dendritic cells, and polymorphonucleated
granulocytes.
Natural killer cells
Cells from the innate immune system that are endowed with
cytotoxic properties. Upon recognition of stressed cells, they
secrete cytokines and release cytolytic granules.
Oligodendrocytes
Glial cells of neuroectodermal origin that are restricted to the
CNS. Oligodendrocytes support axonal function by producing
the myelin and supplying energy and nutrients.
Parenchyma
The functional tissue in the brain, protected by the blood–brain
barrier.
Perivascular cuffs
Perivascular accumulation of leucocytes recorded in infectious,
inflammatory, or autoimmune CNS diseases.
Perivascular spaces
The spaces between the arteries and veins (not capillaries) and
pia mater that can be populated by leucocytes. When inflamed,
they often appear as perivascular cuffs.
T cells
Immune cells that arise from the bone marrow and mature in
the thymus. They are characterised by surface expression of
highly variable T-cell receptors (TCRs) associated with antigen
recognition. T cells function as both effector and regulatory
cells in immune responses.
integrity of the blood–brain barrier.12 However, very few
in-vivo studies directly address the function of astrocytes
under physiological or pathological conditions, and
much of what we know about the biology of astrocytes
has been learned from in-vitro studies.
Microglia are haemopoietic cells with self-renewing
capacity, which are derived early during embryonic
development from yolk-sac myeloid cells (panel 1).15
Therefore, there is a phenotypically and functionally
www.thelancet.com/neurology Vol 14 September 2015

homogeneous population of myeloid cells within the
CNS parenchyma. Under pathological conditions,
monocytes (panel 1) from the blood cross the blood–brain
barrier, enter the CNS, and invade the microglial niche.
The observation of this process led to the idea that some
microglia might be of bone-marrow or monocytic origin,
or even constantly replaced by fresh bone-marrow
emigrants.16 However, a series of experiments showed
that the previously recorded microglia turnover is the
result either of pathological changes, inflammation, or
an experimental artifact induced by CNS irradiation.17
Although we have come a long way in understanding the
developmental origin of microglia, we know little about
their contribution to immune responses within the CNS.
Microglia have a defined role in CNS development, in
which they are needed as phagocytic cells, clearing debris
of apoptotic cells, and for shaping the synaptic
architecture.18 Furthermore, microglia maintain trophic
support and synapse remodelling throughout life.19
Evidence for the association of microglia in autoimmune
diseases is mostly circumstantial, as good models for
testing their direct association with autoimmunity have
Healthy
Astrocyte
end foot
Tight junctions
Monocytes Resting microglia
T lymphocytes Reactive microglia
B lymphocytes NG2 cells
Figure: Involvement of various cell types in CNS inflammation
In almost every non-physiological disorder, the CNS-resident astroglia and microglia become activated, proliferate, and change to have an inflammatory expression
signature. In some inflammatory diseases of the CNS, myeloid cells cross the neurovascular unit and invade the CNS. Lymphocytes with recombined antigen receptors
(T cells and B cells) can also penetrate the CNS and target CNS-resident antigens in many inflammatory diseases (panel 2).
www.thelancet.com/neurology Vol 14 September 2015
Review
been developed only recently.20,21 Microglia are phagocytic,
express MHC molecules, and release proinflammatory
cytokines, and therefore have the capacity to interact with
other immune cells. It is also possible that microglial cells
have only a minor role under pathological conditions and
that their abundance in the CNS is because of their
function in CNS homoeostasis and neuronal
development.22 Under physiological conditions, microglia
seem to monitor the CNS constantly. Whichever is true,
microglia are among the earliest responders to changes
in the CNS and therefore are the first cell types to
respond to any pathological insult.23 Whether the
morphological and functional changes of microglia
actually contribute to pathogenesis is a matter of some
debate. We have illustrated the part that microglia and
astrocyte activation might play under physiological and
pathological conditions in the figure.
Few patients have true autoimmune diseases driven by
adaptive immune responses involving self-reactive
B cells, T cells, or both. This does not mean that these
innate responses are not compounding or restricting the
underlying pathogenesis, but it does mean that this sort
Neuroinflammation
Neurodegeneration or reactive glia Autoimmunity or leucocyte invasion
Astrogliosis Astrogliosis
Compromised blood–brain barrier
Astrocytes Inflammatory mediators
Oligodendrocytes Endothelial cells
Pericytes
947
 Review
of inflammation is rarely the primary driver of tissue
damage and neurological deficits. By contrast, there are
some chronic inflammatory CNS diseases in which
adaptive, rather than innate, immunity dominates the
pathogenesis.
Innate immunity in the CNS
In diseases such as Alzheimer’s disease, Parkinson’s
disease, and amyotrophic lateral sclerosis (ALS), the
microglia show many morphological and molecular
changes (panel 2). Microglia have the ability to
phagocytose and to produce trophic factors,
prostaglandins, chemokines, cytokines, complement
proteins, proteases, reactive oxygen species, and nitric
oxide. Because some of these mediators might have an
adverse effect on CNS-resident cells, or attract other cells
that can cause damage, reactive microglia are thought to
contribute to tissue damage. However, although
microglia display a more amoeboid appearance (panel 1)
and retract their processes during phagocytosis, this
might not signify activated microglia, but rather
exhausted or dysfunctional microglial cells.57
Innate immunity in amyotrophic lateral sclerosis
ALS is a progressive and fatal neurodegenerative disease
that affects motor neurons in the spinal cord, motor
cortex, and brainstem.58 The exact pathophysiological
mechanisms underlying neurodegeneration in ALS
remain largely unclear, but a common pathological
hallmark is the presence of cytoplasmic inclusions in
degenerating neurons, followed by a local inflammatory
reaction.26 ALS is not thought to be mainly an
inflammatory or immune-mediated disease, but immune
mechanisms seem to play a part in its pathogenesis.59
Typically, inflammation in ALS is characterised by
astrocyte activation (gliosis; panel 1) and the accumulation
of large numbers of amoeboid (phagocytic) microglia in
the brain. This altered activation state of glial cells in ALS
is marked by increased production of potentially cytotoxic
molecules such as reactive oxygen species, inflammatory
mediators such as prostaglandins produced by the enzyme
COX-2, and proinflammatory cytokines including
interleukin 1b, tumour necrosis factor α (TNFα), and
interleukin 6.26 Whether this local immune response
limits or accelerates the pathogenesis of ALS is unclear.
ALS is modelled in mice with different mutations in
the Cu–Zn superoxide dismutase (SOD1) gene.60,61
Mutations of the SOD1 gene occur in about 20% of
patients with ALS,62 and reactive microglia and some
tissue-invading myeloid cells can also be found in the
CNS of SOD1 mutant mice.63 In human brain material,
neuropathologists cannot distinguish between invading
monocytes and reactive microglia.64 Using the SOD1
mutant murine model of ALS, it was shown that
monocytes are recruited to the CNS during disease
development, and the inhibition of this process by
neutralising antibodies against the chemokine receptor
948
CCR2—needed for monocyte recruitment into tissues—
resulted in slight disease amelioration.65 Monocytes with
a similar transcriptional signature were identified in the
blood of patients with progressive ALS, suggesting that
these cells contribute to neuronal damage.65
Although soluble mediators such as interleukin 1β and
TNFα were shown to have neurotoxic properties in vitro,
disease progression is not affected by deletion of of the
genes coding for TNFα or interleukin 1β alone in SOD1
mutant mice, only by deficiency of both together.66 Motor
neuron death in ALS is therefore probably the result
of many factors acting in a redundant and
complementary manner.
Even though motor neurons are the main cells affected
in ALS, evidence points to the involvement of
neighbouring astrocytes when these cells are directly
affected by mutations.67 Disease progression is slower
when mutant SOD1 is specifically deleted from
astrocytes.58,67 Similarly, SOD1 mutant mice in which the
mutant gene was deleted exclusively in microglia showed
slower progression of the disease.58,67 However, this could
be because mutant SOD1 might also be deleted in other
myeloid cells. Similar results were seen in bone-marrow
chimeric mice in which only haemopoietic cells were
deficient for SOD1, and it was shown that the presence of
host, wild-type SOD1 astrocytes substantially delayed
onset of motor neuron degeneration.67 Microglia are
hypersensitive to any trigger, including those induced by
activated astrocytes. Primary activation of astrocytes
would probably result in the activation of microglia,
which can propagate the damage within the CNS.
Damage caused first in astrocytes, as probably occurs
with the SOD1 mutation, leads to local, sterile
inflammation in the CNS, which in turn can contribute
to neuronal damage.
Innate immunity in Parkinson’s disease
Parkinson’s disease is the second most common
progressive neurodegenerative disorder. The motor
symptoms of Parkinson’s disease result from the death
of dopaminergic neurons in the substantia nigra, located
in the midbrain. Like ALS, Parkinson’s disease is
associated with neuroinflammation, the hallmarks of
which are the presence of reactive microglia and
astrocytes in the substantia nigra.68 Parkinson’s disease is
associated with increased production of cytokines,
chemokines, reactive oxygen species, and nitric oxide in
the substantia nigra.69,70 Cumulative oxidative damage in
the CNS is deemed an important mechanism because
age is the single most important factor contributing to
induction of sporadic forms of Parkinson’s disease.
Neuronal death itself, including release of protein
aggregates, is proposed to induce activation of microglia
during Parkinson’s disease. Additional activation of
microglia might be caused by the release of aggregated
α-synuclein from neurons into the extracellular space.
Extracellular α-synuclein can be phagocytosed by
www.thelancet.com/neurology Vol 14 September 2015

microglia,71 and oxidised forms of α-synuclein have been
seen to induce microglial activation.71,72 The internalisation
of α-synuclein by microglia is followed by activation of
NADPH oxidase, resulting in production of reactive
oxygen species.71 Local immune cells—namely
microglia—together with reactive astrocytes, are thought
to cause the neuronal damage in a non-antigen-specific
manner, although this is under debate.73 It is possible
that interrupting microglia activation in patients with
Parkinson’s disease might be beneficial in reducing
further neuronal damage.
Adaptive immunity in the CNS
Inflammatory CNS diseases that are very probably
caused by pathogenic T cells, B cells, or both, are varied
(panel 2), and include infectious meningoencephalitis, in
which the damage caused by the immune reaction
against the pathogen is sometimes more harmful than
the direct damage caused by the infectious agent. In
patients with progressive multifocal leukoencephalopathy
and immune reconstitution inflammatory syndrome,
who are recovering from an immunodeficient state,74 a
T-cell response specific to John Cunningham virus in the
brain, largely involving CD8 T cells, probably brings
about virus elimination and killing of CNS-resident cells
with concomitant inflammation.75 Multiple sclerosis is
another example, but because reviews have discussed the
potential contribution of T cells and B cells and have
speculated about their antigenic targets in multiple
sclerosis,76,77 we will concentrate on rarer autoimmune
diseases of the CNS in which the immunopathogenesis
is better understood and in which the target antigens
have been identified, such as anti-Hu-antibody-associated
syndrome, a paraneoplastic syndrome in which auto-
reactive T cells are probably the major immune effectors,
and neuromyelitis optica, a disorder in which
autoantibodies have proven pathogenic properties.
T-cell-mediated autoimmunity in CNS paraneoplastic
syndromes
T-cell-mediated autoimmunity is usually studied with
laboratory models of autoimmune encephalomyelitis
induced by immunisation of rodents with specific myelin
components or adoptive transfer of myelin-reactive
T cells.78 In human beings, T-cell-mediated CNS
inflammatory diseases include acute disseminated
encephalomyelitis,37 Rasmussen’s encephalitis,39 and
some paraneoplastic neurological disorders, such as the
anti-Hu-antibody-associated syndrome40 and paraneo-
plastic cerebellar degeneration (panel 2). Paraneoplastic
encephalomyelopathy and paraneoplastic sensory
neuronopathy (anti-Hu syndrome) are associated with a
partly effective antitumour immune response, in which
tumours are reduced but not eliminated, and usually
follow a sub-acute course, characterised by autoantibodies
targeting intracellular neuronal antigens. These neuronal
self-antigens are ectopically expressed in tumour cells.
www.thelancet.com/neurology Vol 14 September 2015
Review
Panel 2: Pathogenic association of innate and adaptive immune components with
CNS disease pathogenesis
CNS inflammatory diseases with a suspected prominent contribution from
(CNS-resident) innate immune cells
• Alzheimer’s disease24
• Parkinson’s disease25
• Amyotrophic lateral sclerosis26
• Traumatic brain injury27
• Prion disease28
• HIV-associated neurocognitive disorders29
• X-linked adrenoleukodystrophy30
• Neurosarcoidosis31
• Rett’s syndrome32
• Acardi-Goutières syndrome and related nuclease deficiency-associated syndromes33
• Gain-of-function mutations in IFIH134
CNS inflammatory diseases with a suspected prominent contribution from T cells
• Infectious encephalitis and meningoencephalitis35
• Multiple sclerosis36
• Acute disseminated encephalomyelitis37
• Narcolepsy38
• Rasmussen’s disease39
• Paraneoplastic encephalomyelopathy/paraneoplastic sensory neuronopathy (anti-Hu
autoimmunity)40
• Paraneoplastic cerebellar degeneration (anti-Yo autoimmunity)41
• Encephalitis associated with anti-Ma1/2 antibody42
• Stiff-person syndrome43
• Limbic encephalitis associated with anti-glutamic acid decarboxylase (anti-GAD)
antibody 44
• Chronic lymphocytic inflammation with pontine perivascular enhancement
responsive to steroids (CLIPPERS)45
• Idiopathic central diabetes insipidus46
CNS inflammatory diseases with a suspected prominent contribution from B cells
and antibodies
• Neuromyelitis optica (Devic’s disease)47
• Neuro-lupus48
• Encephalitis associated with anti-NMDA receptor antibodies49
• Limbic encephalitis associated with anti-AMPA receptor antibodies50
• Limbic encephalitis associated with anti-leucine-rich glioma-inactivated 1 (anti-LG1)
antibodies51
• Limbic encephalitis associated with anti-GABAB receptor antibodies52
• Epilepsy associated with anti-GABAA receptor antibodies53
• Anti-glycine receptor (anti-GlyR) antibody-associated disorders54
• Morvan’s syndrome with anti-contactin-associated protein-related 2 (anti-Caspr2)
antibodies55
• Parasomnia associated with anti-IgLON family member 5 (anti-IgLON5) antibodies56
For example, small cell lung cancers consistently express
the HuD protein, and low concentrations of anti-HuD
antibody can be detected in 16–22·5% of patients with
small cell lung cancers.79 However, about 1.4% of patients
with small cell lung cancers develop high serum antibody
titres, associated with the anti-Hu syndrome. Serum IgG
in these patients shows cross-reactivity for HuB, HuC,
and HuD, which are physiologically expressed neuron-
949
 Review
specific RNA-binding intracellular proteins. Detection of
high anti-Hu IgG serum titres in patients with
neurological manifestations represents a highly specific
diagnostic test and should lead to a thorough search for
the underlying cancer.
Pathophysiologically, the anti-Hu syndrome is
interpreted as an autoimmune disease triggered by
aberrant expression of neuronal self-antigens by
malignant cells. One key question with major therapeutic
implications with regard to disease pathogenesis is which
components of the immune system mediate nervous
tissue damage. Intrathecal synthesis of HuD-specific IgG
was strongly suggested by isoelectric focusing followed by
immunoblotting on HuD-loaded membranes and
calculation of a CSF–serum anti-Hu antibody index.80
Anti-Hu IgG is also deposited in the nervous tissue, but
whether this shows a post-mortem technical artifact or an
in-vivo occurrence of potential pathogenic relevance
remains uncertain. Because the anti-Hu antibodies are
specific for intracellular and predominantly nuclear
antigens, they might not access their target antigens in
vivo. Their pathogenic potential is therefore questionable.
Infusion of mice with purified IgG from patients with
high titres of anti-Hu antibody did not result in any CNS
lesion despite strong cross-reactivity between the two
species, and the effect of anti-Hu antibodies on in-vitro
neuronal cell cultures is difficult to interpret.40
Inflammatory infiltrates are commonly identified in the
CNS perivascular space (panel 1) and parenchyma of
patients with anti-Hu syndrome. The perivascular cuffs
(panel 1) are mainly composed of T cells, macrophages,
and occasionally B cells. Inflammatory cells within the
parenchyma are mostly T cells and microglial cells. CD8
T cells are the prominent lymphocyte population within
the inflamed CNS parenchyma, and usually form clusters
around neurons.44 Immunohistochemical analyses
revealed intimate contact between the targeted neurons
and cytotoxic granule-containing T cells.81 HLA class I
molecule expression (panel 1) has been described on
neurons in inflamed nervous tissue from patients with
anti-Hu syndrome;44 therefore, the close apposition of
CD8 T cells with neurons is probably indicative of in-vivo
antigen presentation by neurons to CD8 T cells, leading to
neuronal cell death. Microglia are located in the clusters of
inflammatory cells surrounding neurons, indirectly
suggesting their association with neuronophagy, but this
finding should be interpreted with caution.82
The exact antigen recognised by the CNS-infiltrating
T cells remains elusive. However, blood mononuclear cells
from patients with high anti-Hu antibody titres secrete
more interferon γ in response to HuD protein than blood
mononuclear cells from cancer patients without anti-Hu
antibodies. These data suggest that autoreactive CD4
T cells could be implicated in the anti-Hu syndrome.83 It
is unclear whether HuD-specific CD8 T cells are present
more frequently or with greater functional properties in
the blood of patients with anti-Hu syndrome. However, a
950
report suggested that an unconventional non-cytotoxic
interleukin-5-producing and interleukin-13-producing
anti-HuD CD8 T-cell population could be detected in
some patients.84 By contrast, a classic cytotoxic Yo antigen-
specific CD8 T-cell response is detected in paraneoplastic
cerebellar degeneration.36 Further studies on CNS-
infiltrating T cells and blood mononuclear cells of
patients and controls should better define the antigen-
specificity and molecular characteristics of the pathogenic
T cells, thereby offering much needed therapeutic targets.
Antibody-mediated autoimmunity in neuromyelitis
optica
The pathogenesis of autoantibody-mediated neurological
diseases was first studied at the neuromuscular junction
in myasthenia gravis and Lambert-Eaton myasthenic
syndrome. Newly described neurological or neuro-
psychiatric disorders have been associated with
autoantibodies targeting neuronal or axonal surface
antigens, most notably neurotransmitter receptors or
channel proteins (panel 2). The autoantibody-related
mechanisms of tissue injury are best defined in
neuromyelitis optica. However, these mechanisms might
not be applicable to CNS diseases associated with
autoantibodies against neuronal synaptic proteins, which
have been shown to induce a reversible decreased density
of the target protein through internalisation but without
neuronal death.
Neuromyelitis optica is typically characterised by
inflammatory demyelinating lesions in the optic nerves
and spinal cord. Pathologically, neuromyelitis optica
lesions are generally more destructive than multiple
sclerosis lesions, show loss of astrocytes consistent with
demyelination and axonal damage, and contain
eosinophils and neutrophils.45,85,86 The implication of
humoral immunity in neuromyelitis optica lesions
became clear on the basis of the prominent deposition of
immunoglobulins and activated complement
components, most notably at the outer rim of vessel
walls.87 IgG autoantibodies specific for aquaporin 4
(AQP4), a water channel localised on astrocyte foot
processes and pial surface, were identified as a highly
specific serum biomarker for neuromyelitis optica.88,89
AQP4 is particularly expressed in the spinal cord grey
matter and in the optic nerve. AQP4 aggregates into
larger structures forming orthogonal arrays of particles
on the astrocyte membrane.89
Although some plasma cells are able to synthesise anti-
AQP4 IgG intrathecally,90 the serum:CSF anti-AQP4 IgG
ratio in patients with neuromyelitis optica suggets there
is substantial production of IgG outside the CNS.
Colonies of blood plasmablasts are expanded during
neuromyelitis optica relapse and are a source of anti-
AQP4 IgG, which is detected at high concentrations in
the serum and CSF of patients with neuromyelitis optica
and sustains the survival of these cells, and therefore
represents a promising therapeutic target.91
www.thelancet.com/neurology Vol 14 September 2015

Patient IgG has shown pathogenic potential in vivo
after injection into laboratory animals.92,93 This pathogenic
property is lost when IgG preparations are depleted of
anti-AQP4 antibodies, suggesting that anti-AQP4 IgG is
the probable cause of the damage.92,94 In primary astrocyte
cultures, anti-AQP4 IgG bind the extracellular domains
of AQP4 resulting in internalisation of the AQP4–
autoantibody complexes with subsequent endolysosomal
proteolysis of AQP4 and activation of the complement
pathway or of cell-surface-bound Fc receptors by the anti-
AQP4 IgG.95
The modulation of AQP4 on the astrocyte surface has
consequences beyond water movement. The excitatory
aminoacid transporter-2 (EAAT2), which plays a key part
in glutamate homoeostasis in the CNS, forms a
macromolecular complex with AQP4 on astrocytes. Loss
of both AQP4 and EAAT2 has been documented in
neuromyelitis optica lesions.95 When primary astrocytes
are incubated with anti-AQP4 IgG, AQP4 and EAAT2 are
downregulated and glutamate uptake is substantially
reduced. Activation of the ionotropic glutamate receptors
might result in axonal injury and oligodendrocyte
changes and death.96 This could underlie the clinical
benefit of plasma exchange in some patients.97
On IgG binding to AQP4, the Fc portion of the IgG can
trigger complement activation and antibody-
dependent cell-mediated cytotoxicity. The large and
consistent deposition of immunoglobulin and of
activated complement fractions in lesions suggests a
crucial role of complement in tissue injury.87,98
Complement activation also generates strongly
chemoattractive molecules, notably C5a, which is
recorded at high concentrations in the CSF of patients
with neuromyelitis optica.99 Eculizumab, a monoclonal
antibody against the C5 protein, showed efficacy in
seropositive neuromyelitis optica patients in an open-
label trial,100 pointing to the pathological relevance of this
pathway in patients.
Tissue-invading myeloid cells such as monocytes,
eosinophils, and neutrophils probably have an amplifying
role in neuromyelitis optica pathophysiology. Fc receptors
are present on the surface of all these cell types, and their
interaction with anti-AQP4 IgG bound to tissue can
stimulate release of cytokines, chemokines, reactive
oxygen species and cytotoxic degranulation, resulting in
antibody-dependent cell-mediated cytotoxicity.86,94,101
Activation of adaptive immune responses in the
periphery
CNS inflammatory diseases that are attributable to
adaptive immune responses probably start within
lymphoid structures outside the nervous tissue. Specific
antigen receptors of T cells or B cells are either activated
by CNS-derived soluble antigens presented by peripheral
antigen-presenting cells, by autoantigens also expressed
in peripheral tissues,102 by cross-reactive foreign
antigens,103,104 or by neo-autoantigens of tumours.40 The
www.thelancet.com/neurology Vol 14 September 2015
Review
adaptive immune system is associated with active
immune surveillance of the CNS, so aberrantly activated
T cells and B cells can—in principle—access the CNS
and cause antigen-driven inflammatory CNS diseases
(panel 2). Tissue damage then allows CNS antigens to
access the cervical lymph nodes, thereby potentially
sustaining autoimmunity by introducing the antigen to
the peripheral immune system.5
Therapeutic strategies targeting CNS
autoimmune mechanisms
The treatment of CNS autoimmune diseases relies on
various approaches depending on the lymphocyte
subset(s) implicated. Immune cell-depleting strategies,
such as those using anti-CD52 (alemtuzumab) or anti-
CD20 (rituximab, ocrelizumab, ofatumumab) mono-
clonal antibodies, are probably the more radical
approaches. These strategies efficiently inhibit CNS
inflammation in multiple sclerosis but, with their
indiscriminate targeting of both pathogenic and
protective immune cells, might have side-effects.105
A more selective option is the pharmacological
interference of immune cell migration to the CNS.
Inhibition of immune cell trafficking through blockade
of the α4 integrin with natalizumab or small molecule
antagonist against S1P and S1P-receptor functional
antagonists (fingolimod, siponimod) has proven effective
in multiple sclerosis. Such approaches could be
applicable in other CNS inflammatory or autoimmune
diseases. Diseases in which CD8 T cells or autoreactive
helper T cells are the main immune effectors might also
be controlled by strategies acting on adhesion molecules
other than α4 integrin106 or on specific chemokine
receptors such as CCR5.107,108 The major drawback of this
approach is low selectivity, thereby resulting in reduced
overall immune surveillance of the CNS.
Neutralising of the harmful immune effectors without
affecting the rest of the immune response could include
selective neutralisation of effector molecules implicated
in the disease process (either in the initiation or the
effector phase). For some antibody-mediated diseases,
such as neuromyelitis optica, plasmapheresis and
neutralisation of complement factors can be efficient
strategies. However, in patients with other antibody-
mediated diseases, such as encephalitis associated with
anti-NMDA receptor antibodies, B-cell clonal expansion
and affinity maturation can take place within CNS
niches, blunting the efficiency of strategies targeting
humoral immune components in the periphery, because
these cannot cross the blood–brain barrier. It seems that
inflamed tissues have the capacity to host B cell
maturation, albeit not as effectively as dedicated
secondary lymphoid organs such as lymph nodes.
Studies of multiple sclerosis reveal that disease-
associated B-cell clones can mature either in the
periphery or within the CNS, with an active connection
between the two compartments.109
951
 Review
Neutralisation of proinflammatory cytokines such as
interleukin 23, interleukin 17A, granulocyte-macrophage
colony-stimulating factor, interleukin 6, or their receptors
is under investigation in multiple sclerosis and
neuromyelitis optica. This approach has been successful
in chronic inflammatory diseases outside the CNS but
whether it will also be useful to treat CNS autoimmunity
remains unclear. Surprisingly, blocking the
interleukin 2Rα chain with daclizumab seems to have
indirect mechanisms of action involving inhibition of
T-cell activation by dendritic cells110 and expansion of
regulatory CD56bright natural killer cells (panel 1).
Inhibition of T-cell or B-cell activation, proliferation
signals, or promotion of regulatory pathways—including
boosting the numbers or functions of CD4 regulatory
T cells—are plausible future avenues for immune
intervention in neuroimmunological diseases.
Local activation of innate immune cells, namely
microglia and astrocytes, can be the result but also the
cause of CNS tissue damage. Therefore, strategies that
target innate immune cells or their mediators for the
treatment of inflammatory diseases of the CNS are likely
to gain momentum in the coming years. These could
include inhibition of tyrosine kinases, inhibition of NF-
κB, and scavengers for reactive oxygen species and nitric
oxide, or pharmacological interference with their
production.
Inflammasomes are key multiprotein platforms
associated with the inflammatory cascade, notably
production of active interleukin 1 and interleukin 18 as a
result of recognition of foreign or endogenous danger
signals. Accumulating evidence supports the rationale of
targeting inflammasome activation in the CNS in future
trials.111 Interfering with cytokine networks can also
dampen adverse microglial activation.112
These methods are selected examples of strategies to
control the destructive activation of local immune cells in
the CNS. We need to keep in mind that in some groups of
diseases, such as ALS and Parkinson’s disease, it is
probably not sufficient to block the influx of immune cells
from the peripheral immune system, as described above,
but the use of new tactics to restrain microglia and
astrocyte activation should be considered. Repurposing
existing therapies to this end is also an actively investigated
avenue. The use of statins in multiple sclerosis and the
Search strategy and selection criteria
We searched PubMed for articles published in English
between January, 1980, and July, 2015, and references from
relevant articles. The search terms “CNS inflammation”, “CNS
autoimmunity”, “Alzheimer’s disease AND inflammation”,
“Parkinson’s disease AND inflammation”, “amyotrophic
lateral sclerosis AND inflammation”, ”neuromyelitis optica”,
and “SOD1” were used. The final reference list was generated
on the basis of relevance to the topics covered in this Review.
952
use of monoclonal antibodies or antagonists blocking
cytokines binding to their receptors (interleukin 1,
interleukin 6, interleukin 12) might prove successful.29,99,113
In autoimmune diseases of the CNS, rather than using
broad strategies that target useful immune components
indiscriminately to neutralise a few harmful
immune cells, the ultimate goal is to selectively modulate
the autoreactive lymphocytes associated with patho-
genesis. Antigen-specific immunotherapy shows effects
in preclinical animal models, but has not yet met clinical
expectations.114 Recent efforts aim either to use myelin-
derived peptides or proteins to inactivate autoreactive T
and B cells in multiple sclerosis,115 or selectively block the
interaction between the pathogenic anti-AQP4 antibodies
and their target.116
Conclusions and future directions
In this Review, we have attempted to explain
inflammation in the context of CNS diseases. There is
ample evidence that inflammation occurs in almost all
CNS disorders, irrespective of whether immunity is the
primary driver of disease (as in autoimmune
encephalitides) or represents the tissue response to
degenerative processes (eg, Parkinson’s disease, ALS,
Alzheimer’s disease). CNS tissue destruction can be
initiated by neural antigen-specific T or B cells, as seen in
neuromyelitis optica and in paraneoplastic neurological
diseases. Conversely, CNS inflammation and ultimately
damage can also be initiated by malfunction or aberrant
activation of the CNS-resident cells. We have an
incomplete understanding of the molecular pathways
associated with the pathogenic, and potentially protective,
contribution of both CNS-resident and CNS-infiltrating
immune cells to CNS diseases. Several issues remain
unsettled, including understanding the roles of blood-
borne myeloid cells and microglia in the diseased CNS.
Additionally, there is evidence that immune responses
favour repair after CNS damage.117 If CNS inflammation
can actually be either damaging or disease limiting, we
need to understand the mechanisms of each aspect to
develop new therapeutic strategies that exploit the
protective elements of inflammation.
Most neurological diseases show clear involvement of
the immune system; therefore, therapies aimed at
blocking immune cells is an obvious strategy. Such
strategies are already successful in the treatment of
people with multiple sclerosis, and could prove beneficial
for other inflammatory diseases of the CNS. Beyond
these proven therapies, the question of whether we can
harness the innate and adaptive immune system to
eliminate toxic components such as aggregated or
misfolded proteins within the CNS remains unanswered.
One strategy would be to engineer CNS-infiltrating
immune cells to express factors important in
neuroprotection or neuroregeneration, and in such a way
to not only reduce inflammation, but also protect the
tissue and promote the repair process.
www.thelancet.com/neurology Vol 14 September 2015

CNS diseases involving the influx of systemic
leucocytes, including B and T cells, are amenable to
treatment with modern biological drugs. Diseases such
as ALS, Parkinson’s disease, and Alzheimer’s disease, in
which lymphocytes probably have at most a minor role,
all share a clear inflammatory tissue response to the
degenerative processes. This tissue inflammation is only
now recognised as having detrimental effects and to
possibly accelerate disease progression. In view of the
speed at which fundamental discoveries are now being
translated to treat patients, we predict that controlling
innate inflammation will hold great promise for the
treatment of neurodegenerative diseases. Moreover,
knowledge about the genetic bases and the immune
mechanisms underlying CNS inflammatory diseases
should offer unique opportunities for more targeted
therapies.
Contributors
All authors contributed equally.
Declaration of interests
AW declares a grant and consultancy fees from Novartis. RSL declares
grants from Pierre Fabre and consultancy fees from Genzyme, Servier,
and GSK. BB holds a patent application by the University of Zurich and
the Charité–Universitätsmedizin Berlin entitled “Modulators of IL-12
and/or IL-23 for the Prevention or Treatment of Alzheimer’s Disease”
(PCT/EP2012/050066).
Acknowledgments
We thank Khalad Karram for the preparation of the figure. The authors’
work is supported by Inserm, CNRS, Toulouse III University, and ANR
(RSL); the Swiss National Science Foundation, the European Union, FP7
TargetBrain (279017), and the Swiss Multiple Sclerosis Society (BB); the
German Research Foundation SFB/TR 128, the Research Centre for
Immunotherapy, and the Focus Program in Neuroscience (AW), and a
grant from the European Union, FP7-PEOPLE-2012-ITN NeuroKine
(AW, RSL, and BB). Funding sources had no roles in the writing of the
manuscript or the decision to submit it for publication.
References
1 Murphy JB, Sturm E. Conditions determining the transplantability
of tissues in the brain. J Exp Med 1923; 38: 183–97.
2 Greter M, Heppner FL, Lemos MP, et al. Dendritic cells permit
immune invasion of the CNS in an animal model of multiple
sclerosis. Nat Med 2005; 11: 328–34.
3 McMenamin PG. Distribution and phenotype of dendritic cells and
resident tissue macrophages in the dura mater, leptomeninges, and
choroid plexus of the rat brain as demonstrated in wholemount
preparations. J Comp Neurol 1999; 405: 553–62.
4 Bartholomäus I, Kawakami N, Odoardi F, et al. Effector T cell
interactions with meningeal vascular structures in nascent
autoimmune CNS lesions. Nature 2009; 462: 94–98.
5 de Vos AF, van Meurs M, Brok HP, et al. Transfer of central
nervous system autoantigens and presentation in secondary
lymphoid organs. J Immunol 2002; 169: 5415–23.
6 Locatelli G, Wortge S, Buch T, et al. Primary oligodendrocyte death
does not elicit anti-CNS immunity. Nat Neurosci 2012; 15: 543–50.
7 Becher B. Central nervous system immune surveillance: on
natalizumab, dendritic cells, and dangerous immune privilege.
Arch Neurol 2008; 65: 1566–67.
8 Lemaitre B. The road to Toll. Nat Rev Immunol 2004; 4: 521–27.
9 Hirano M, Das S, Guo P, Cooper MD. The evolution of adaptive
immunity in vertebrates. Adv Immunol 2011; 109: 125–57.
10 Obermeier B, Daneman R, Ransohoff RM. Development,
maintenance and disruption of the blood-brain barrier. Nat Med
2013; 19: 1584–96.
11 Bentivoglio M, Kristensson K. Tryps and trips: cell trafficking across
the 100-year-old blood-brain barrier. Trends Neurosci 2014;
37: 325–33.
www.thelancet.com/neurology Vol 14 September 2015
Review
12 Alvarez JI, Katayama T, Prat A. Glial influence on the blood brain
barrier. Glia 2013; 61: 1939–58.
13 Chaboub LS, Deneen B. Developmental origins of astrocyte
heterogeneity: the final frontier of CNS development. Dev Neurosci
2012; 34: 379–88.
14 Finsen B, Owens T. Innate immune responses in central nervous
system inflammation. FEBS Lett 2011; 585: 3806–12.
15 Ginhoux F, Greter M, Leboeuf M, et al. Fate mapping analysis
reveals that adult microglia derive from primitive macrophages.
Science 2010; 330: 841–45.
16 Simard AR, Soulet D, Gowing G, Julien J-P, Rivest S. Bone
marrow-derived microglia play a critical role in restricting senile
plaque formation in Alzheimer’s disease. Neuron 2006;
49: 489–502.
17 Mildner A, Schmidt H, Nitsche M, et al. Microglia in the adult brain
arise from Ly-6ChiCCR2+ monocytes only under defined host
conditions. Nat Neurosci 2007; 10: 1544–53.
18 Paolicelli RC, Bolasco G, Pagani F, et al. Synaptic pruning by
microglia is necessary for normal brain development. Science 2011;
333: 1456–58.
19 Sierra A, Abiega O, Shahraz A, Neumann H. Janus-faced microglia:
beneficial and detrimental consequences of microglial phagocytosis.
Front Cell Neurosci 2013; 7: 6.
20 Yona S, Kim KW, Wolf Y, et al. Fate mapping reveals origins and
dynamics of monocytes and tissue macrophages under
homeostasis. Immunity 2013; 38: 79–91.
21 Heppner FL, Greter M, Marino D, et al. Experimental autoimmune
encephalomyelitis repressed by microglial paralysis. Nat Med 2005;
11: 146–52.
22 Kettenmann H, Kirchhoff F, Verkhratsky A. Microglia: new roles
for the synaptic stripper. Neuron 2013; 77: 10–18.
23 Becher B, Prat A, Antel JP. Brain-immune connection: immuno-
regulatory properties of CNS-resident cells. Glia 2000; 29: 293–304.
24 Heneka MT, Carson MJ, El Khoury J, et al. Neuroinflammation in
Alzheimer’s disease. Lancet Neurol 2015; 14: 388–405.
25 Gratwicke J, Jahanshahi M, Foltynie T. Parkinson’s disease
dementia: a neural networks perspective. Brain 2015; 138: 1454–76.
26 McGeer PL, McGeer EG. Inflammatory processes in amyotrophic
lateral sclerosis. Muscle Nerve 2002; 26: 459–70.
27 Bergold PJ. Treatment of traumatic brain injury with anti-
inflammatory drugs. Exp Neurol 2015; published online June 23,
2015. DOI: 10.1016/j.expneurol.2015.05.024.
28 Gomez-Nicola D, Fransen NL, Suzzi S, Perry VH. Regulation of
microglial proliferation during chronic neurodegeneration.
J Neurosci 2013; 33: 2481–93.
29 Clifford DB, Ances BM. HIV-associated neurocognitive disorder.
Lancet Infect Dis 2013; 13: 976–86.
30 Ferrer I, Aubourg P, Pujol A. General aspects and neuropathology
of X-linked adrenoleukodystrophy. Brain Pathol 2010; 20: 817–30.
31 Krumholz A, Stern BJ. Neurologic manifestations of sarcoidosis.
Handb Clin Neurol 2014; 119: 305–33.
32 Derecki NC, Cronk JC, Lu Z, et al. Wild-type microglia arrest
pathology in a mouse model of Rett syndrome. Nature 2012;
484: 105–09.
33 Lee-Kirsch MA, Wolf C, Gunther C. Aicardi-Goutieres syndrome:
a model disease for systemic autoimmunity. Clin Exp Immunol
2014; 175: 17–24.
34 Rice GI, del Toro Duany Y, Jenkinson EM, et al. Gain-of-function
mutations in IFIH1 cause a spectrum of human disease phenotypes
associated with upregulated type I interferon signaling. Nat Genet
2014; 46: 503–09.
35 Joubert B, Honnorat J. Autoimmune channelopathies in
paraneoplastic neurological syndromes. Biochim Biophys Acta 2015;
published online April 14, 2015. DOI:10.1016/j.bbamem.2015.04.003.
36 Hemmer B, Kerschensteiner M, Korn T. Role of the innate and
adaptive immune responses in the course of multiple sclerosis.
Lancet Neurol 2015; 14: 406–19.
37 Tenembaum SN. Acute disseminated encephalomyelitis.
Handb Clin Neurol 2013; 112: 1253–62.
38 Liblau RS, Vassalli A, Seifinejad A, Tafti M. Hypocretin (orexin)
biology and the pathophysiology of narcolepsy with cataplexy.
Lancet Neurol 2015; 14: 318–28.
953
 Review
39 Varadkar S, Bien CG, Kruse CA, et al. Rasmussen’s encephalitis:
clinical features, pathobiology, and treatment advances.
Lancet Neurol 2014; 13: 195–205.
40 Pignolet BS, Gebauer CM, Liblau RS. Immunopathogenesis of
paraneoplastic neurological syndromes associated with anti-Hu
antibodies: A beneficial antitumor immune response going awry.
Oncoimmunology 2013; 2: e27384.
41 Albert ML, Darnell JC, Bender A, Francisco LM, Bhardwaj N,
Darnell RB. Tumor-specific killer cells in paraneoplastic cerebellar
degeneration. Nat Med 1998; 4: 1321–24.
42 Leyhe T, Schüle R, Schwärzler F, Gasser T, Haarmeier T. Second
primary tumour in anti-Ma1/2-positive paraneoplastic limbic
encephalitis. J Neurooncology 2006; 17: 49–51.
43 Baizabal-Carvallo JK, Jankovic J. Stiff-person syndrome: insights
into a complex autoimmune disorder. J Neurol Neurosurg Psychiatry
2015; 86: 840–48.
44 Bien CG, Vincent A, Barnett MH, et al. Immunopathology of
autoantibody-associated encephalitides: clues for pathogenesis.
Brain 2012; 135: 1622–38.
45 Simon NG, Parratt JD, Barnett MH, et al. Expanding the clinical,
radiological and neuropathological phenotype of chronic
lymphocytic inflammation with pontine perivascular enhancement
responsive to steroids (CLIPPERS). J Neurol Neurosurg Psychiatry
2012; 83: 15–22.
46 Scheikl T, Pignolet B, Dalard C, et al. Cutting edge: neuronal
recognition by CD8 T cells elicits central diabetes insipidus.
J Immunol 2012; 188: 4731–35.
47 Bruck W, Popescu B, Lucchinetti CF, et al. Neuromyelitis optica
lesions may inform multiple sclerosis heterogeneity debate.
Ann Neurol 2012; 72: 385–94.
48 Jeltsch-David H, Muller S. Neuropsychiatric systemic lupus
erythematosus: pathogenesis and biomarkers. Nat Rev Neurol 2014;
10: 579–96.
49 Hughes EG, Peng X, Gleichman AJ, et al. Cellular and synaptic
mechanisms of anti-NMDA receptor encephalitis. J Neurosci 2010;
30: 5866–75.
50 Lai M, Hughes EG, Peng X, et al. AMPA receptor antibodies in
limbic encephalitis alter synaptic receptor location. Ann Neurol
2009; 65: 424–34.
51 Lai M, Huijbers MG, Lancaster E, et al. Investigation of LGI1 as the
antigen in limbic encephalitis previously attributed to potassium
channels: a case series. Lancet Neurol 2010; 9: 776–85.
52 Hoftberger R, Titulaer MJ, Sabater L, et al. Encephalitis and GABAB
receptor antibodies: novel findings in a new case series of
20 patients. Neurology 2013; 81: 1500–06.
53 Petit-Pedrol M, Armangue T, Peng X, et al. Encephalitis with
refractory seizures, status epilepticus, and antibodies to the GABAA
receptor: a case series, characterisation of the antigen, and analysis
of the effects of antibodies. Lancet Neurol 2014; 13: 276–86.
54 Hutchinson M, Waters P, McHugh J, et al. Progressive
encephalomyelitis, rigidity, and myoclonus: a novel glycine receptor
antibody. Neurology 2008; 71: 1291–92.
55 Irani SR, Alexander S, Waters P, et al. Antibodies to Kv1 potassium
channel-complex proteins leucine-rich, glioma inactivated 1 protein
and contactin-associated protein-2 in limbic encephalitis, Morvan’s
syndrome and acquired neuromyotonia. Brain 2010; 133: 2734–48.
56 Sabater L, Gaig C, Gelpi E, et al. A novel non-rapid-eye movement
and rapid-eye-movement parasomnia with sleep breathing disorder
associated with antibodies to IgLON5: a case series, characterisation
of the antigen, and post-mortem study. Lancet Neurol 2014;
13: 575–86.
57 Prokop S, Miller KR, Heppner FL. Microglia actions in Alzheimer’s
disease. Acta Neuropathol 2013; 126: 461–77.
58 Boillee S, Vande Velde C, Cleveland DW. ALS: a disease of motor
neurons and their nonneuronal neighbors. Neuron 2006; 52: 39–59.
59 Beers DR, Henkel JS, Zhao W, et al. Endogenous regulatory
T lymphocytes ameliorate amyotrophic lateral sclerosis in mice and
correlate with disease progression in patients with amyotrophic
lateral sclerosis. Brain 2011; 134: 1293–314.
60 Vehvilainen P, Koistinaho J, Gundars G. Mechanisms of mutant
SOD1 induced mitochondrial toxicity in amyotrophic lateral
sclerosis. Front Cell Neurosci 2014; 8: 126.
61 Garbuzova-Davis S, Sanberg PR. Blood-CNS Barrier Impairment in
ALS patients versus an animal model. Front Cell Neurosci 2014; 8: 21.
954
62 Wijesekera LC, Leigh PN. Amyotrophic lateral sclerosis.
Orphanet J Rare Dis 2009; 4: 3.
63 Sreedharan J, Brown RH Jr. Amyotrophic lateral sclerosis:
Problems and prospects. Ann Neurol 2013; 74: 309–16.
64 Varvel NH, Grathwohl SA, Baumann F, et al. Microglial repopulation
model reveals a robust homeostatic process for replacing CNS
myeloid cells. Proc Natl Acad Sci USA 2012; 109: 18150–55.
65 Butovsky O, Siddiqui S, Gabriely G, et al. Modulating inflammatory
monocytes with a unique microRNA gene signature ameliorates
murine ALS. J Clin Invest 2012; 122: 3063–87.
66 Gowing G, Dequen F, Soucy G, Julien JP. Absence of tumor necrosis
factor-alpha does not affect motor neuron disease caused by
superoxide dismutase 1 mutations. J Neurosci 2006; 26: 11397–402.
67 Yamanaka K, Boillee S, Roberts EA, et al. Mutant SOD1 in cell
types other than motor neurons and oligodendrocytes accelerates
onset of disease in ALS mice. Proc Natl Acad Sci USA 2008;
105: 7594–99.
68 Lee JK, Tran T, Tansey MG. Neuroinflammation in Parkinson’s
disease. J Neuroimmune Pharmacol 2009; 4: 419–29.
69 Ransohoff RM, Perry VH. Microglial physiology: unique stimuli,
specialized responses. Ann Rev Immunol 2009; 27: 119–45.
70 Hirsch EC, Hunot S. Neuroinflammation in Parkinson’s disease: a
target for neuroprotection? Lancet Neurol 2009; 8: 382–97.
71 Zhang W, Wang T, Pei Z, et al. Aggregated alpha-synuclein activates
microglia: a process leading to disease progression in Parkinson’s
disease. FASEB J 2005; 19: 533–42.
72 Roodveldt C, Labrador-Garrido A, Gonzalez-Rey E, et al. Glial innate
immunity generated by non-aggregated alpha-synuclein in mouse:
differences between wild-type and Parkinson’s disease-linked
mutants. PLoS One 2010; 5: e13481.
73 Biber K, Owens T, Boddeke E. What is microglia neurotoxicity
(Not)? Glia 2014; 62: 841–54.
74 Martin-Blondel G, Delobel P, Blancher A, et al. Pathogenesis of the
immune reconstitution inflammatory syndrome affecting the
central nervous system in patients infected with HIV. Brain 2011;
134: 928–46.
75 Martin-Blondel G, Bauer J, Cuvinciuc V, et al. In situ evidence of JC
virus control by CD8+ T cells in PML-IRIS during HIV infection.
Neurology 2013; 81: 964–70.
76 Nylander A, Hafler DA. Multiple sclerosis. J Clin Invest 2012;
122: 1180–88.
77 Steinman L. Immunology of relapse and remission in multiple
sclerosis. Ann Rev Immunol 2014; 32: 257–81.
78 Liblau RS, Gonzalez-Dunia D, Wiendl H, Zipp F. Neurons as
targets for T cells in the nervous system. Trends Neurosci 2013;
36: 315–24.
79 Monstad SE, Knudsen A, Salvesen HB, Aarseth JH, Vedeler CA.
Onconeural antibodies in sera from patients with various types of
tumours. Cancer Immunol Immunother 2009; 58: 1795–800.
80 Rauer S, Kaiser R. Demonstration of anti-HuD specific oligoclonal
bands in the cerebrospinal fluid from patients with paraneoplastic
neurological syndromes. Qualitative evidence of anti-HuD specific
IgG-synthesis in the central nervous system. J Neuroimmunol 2000;
111: 241–44.
81 Wanschitz J, Hainfellner JA, Kristoferitsch W, Drlicek M, Budka H.
Ganglionitis in paraneoplastic subacute sensory neuronopathy:
a morphologic study. Neurology 1997; 49: 1156–59.
82 Yamasaki R, Lu H, Butovsky O, et al. Differential roles of microglia
and monocytes in the inflamed central nervous system. J Exp Med
2014; 211: 1533–49.
83 Benyahia B, Liblau R, Merle-Beral H, Tourani JM, Dalmau J,
Delattre JY. Cell-mediated autoimmunity in paraneoplastic
neurological syndromes with anti-Hu antibodies. Ann Neurol 1999;
45: 162–67.
84 Roberts WK, Deluca IJ, Thomas A, et al. Patients with lung cancer
and paraneoplastic Hu syndrome harbor HuD-specific type 2 CD8+
T cells. J Clin Invest 2009; 119: 2042–51.
85 Saadoun S, Waters P, MacDonald C, et al. Neutrophil protease
inhibition reduces neuromyelitis optica-immunoglobulin
G-induced damage in mouse brain. Ann Neurol 2012; 71: 323–33.
86 Misu T, Hoftberger R, Fujihara K, et al. Presence of six different
lesion types suggests diverse mechanisms of tissue injury in
neuromyelitis optica. Acta Neuropathol 2013; 125: 815–27.
www.thelancet.com/neurology Vol 14 September 2015

87 Lucchinetti CF, Mandler RN, McGavern D, et al. A role for humoral
mechanisms in the pathogenesis of Devic’s neuromyelitis optica.
Brain 2002; 125: 1450–61.
88 Lennon VA, Kryzer TJ, Pittock SJ, Verkman AS, Hinson SR. IgG
marker of optic-spinal multiple sclerosis binds to the aquaporin-4
water channel. J Exp Med 2005; 202: 473–77.
89 Papadopoulos MC, Verkman AS. Aquaporin water channels in the
nervous system. Nat Rev Neurosci 2013; 14: 265–77.
90 Bennett JL, Lam C, Kalluri SR, et al. Intrathecal pathogenic anti-
aquaporin-4 antibodies in early neuromyelitis optica. Ann Neurol
2009; 66: 617–29.
91 Chihara N, Aranami T, Sato W, et al. Interleukin 6 signaling promotes
anti-aquaporin 4 autoantibody production from plasmablasts in
neuromyelitis optica. Proc Natl Acad Sci USA 2011; 108: 3701–06.
92 Bradl M, Misu T, Takahashi T, et al. Neuromyelitis optica:
pathogenicity of patient immunoglobulin in vivo. Ann Neurol 2009;
66: 630–43.
93 Saadoun S, Waters P, Bell BA, Vincent A, Verkman AS,
Papadopoulos MC. Intra-cerebral injection of neuromyelitis optica
immunoglobulin G and human complement produces
neuromyelitis optica lesions in mice. Brain 2010; 133: 349–61.
94 Zhang H, Verkman AS. Eosinophil pathogenicity mechanisms and
therapeutics in neuromyelitis optica. J Clin Invest 2013; 123: 2306–16.
95 Hinson SR, Romero MF, Popescu BF, et al. Molecular outcomes of
neuromyelitis optica (NMO)-IgG binding to aquaporin-4 in
astrocytes. Proc Natl Acad Sci USA 2012; 109: 1245–50.
96 Marignier R, Nicolle A, Watrin C, et al. Oligodendrocytes are
damaged by neuromyelitis optica immunoglobulin G via astrocyte
injury. Brain 2010; 133: 2578–91.
97 Keegan M, Pineda AA, McClelland RL, Darby CH, Rodriguez M,
Weinshenker BG. Plasma exchange for severe attacks of CNS
demyelination: predictors of response. Neurology 2002; 58: 143–46.
98 Sabater L, Giralt A, Boronat A, et al. Cytotoxic effect of
neuromyelitis optica antibody (NMO-IgG) to astrocytes: an in vitro
study. J Neuroimmunol 2009; 215: 31–35.
99 Kuroda H, Fujihara K, Takano R, et al. Increase of complement
fragment C5a in cerebrospinal fluid during exacerbation of
neuromyelitis optica. J Neuroimmunol 2013; 254: 178–82.
100 Pittock SJ, Lennon VA, McKeon A, et al. Eculizumab in
AQP4-IgG-positive relapsing neuromyelitis optica spectrum
disorders: an open-label pilot study. Lancet Neurol 2013; 12: 554–62.
101 Ratelade J, Zhang H, Saadoun S, Bennett JL, Papadopoulos MC,
Verkman AS. Neuromyelitis optica IgG and natural killer cells
produce NMO lesions in mice without myelin loss.
Acta Neuropathol 2012; 123: 861–72.
102 Krishnamoorthy G, Saxena A, Mars LT, et al. Myelin-specific T cells
also recognize neuronal autoantigen in a transgenic mouse model
of multiple sclerosis. Nat Med 2009; 15: 626–32.
www.thelancet.com/neurology Vol 14 September 2015
Review
103 Birnbaum ME, Mendoza JL, Sethi DK, et al. Deconstructing the
peptide-MHC specificity of T cell recognition. Cell 2014;
157: 1073–87.
104 Varrin-Doyer M, Spencer CM, Schulze-Topphoff U, et al. Aquaporin
4-specific T cells in neuromyelitis optica exhibit a Th17 bias and
recognize Clostridium ABC transporter. Ann Neurol 2012; 72: 53–64.
105 Haghikia A, Hohlfield R, Gold R, Fugger L. Therapies for multiple
sclerosis: translational achievements and outstanding needs.Trends
Mol Med 2013; 19: 309–19.
106 Rothhammer V, Muschaweckh A, Gasteiger G, et al. alpha4-
integrins control viral meningoencephalitis through differential
recruitment of T helper cell subsets. Acta Neuropathol Commun
2014; 2: 27.
107 Giacomini PS, Rozenberg A, Metz I, et al. Maraviroc and JC
virus-associated immune reconstitution inflammatory syndrome.
N Engl J Med 2014; 370: 486–88.
108 Martin-Blondel G, Bauer J, Uro-Coste E, et al. Therapeutic use of
CCR5 antagonists is supported by strong expression of CCR5 on
CD8 T cells in progressive multifocal leukoencephalopathy-
associated immune reconstitution inflammatory syndrome.
Acta Neuropathol 2015; 129: 463–65.
109 Lu DR, Robinson WH. Street-experienced peripheral B cells traffic
to the brain. Sci Transl Med 2014; 6: 248fs31.
110 Bielekova B. Daclizumab therapy for multiple sclerosis.
Neurotherapeutics 2013; 10: 55–67.
111 Walsh JG, Muruve DA, Power C. Inflammasomes in the CNS.
Nat Rev Neurosci 2014; 15: 84–97.
112 Vom Berg J, Prokop S, Miller KR, et al. Inhibition of IL-12/IL-23
signaling reduces Alzheimer’s disease-like pathology and cognitive
decline. Nat Med 2012; 18: 1812–19.
113 Chataway J, Schuerer N, Alsanousi A, et al. Effect of high-dose
simvastatin on brain atrophy and disability in secondary progressive
multiple sclerosis (MS-STAT): a randomised, placebo-controlled,
phase 2 trial. Lancet 2014; 383: 2213–21.
114 Dolgin E. The inverse of immunity. Nat Med 2010; 16: 740–43.
115 Lutterotti A, Yousef S, Sputtek A, et al. Antigen-specific tolerance by
autologous myelin peptide-coupled cells: a phase 1 trial in multiple
sclerosis. Sci Transl Med 2013; 5: 188ra75.
116 Tradtrantip L, Zhang H, Saadoun S, et al. Anti-aquaporin-4
monoclonal antibody blocker therapy for neuromyelitis optica.
Ann Neurol 2012; 71: 314–22.
117 Schwartz M, Raposo C. Protective autoimmunity: a unifying model
for the immune network involved in CNS repair. Neuroscientist 2014;
20: 343–58.
955