Gene expression profiling in MS

[Guest guest]

This is dense reading, most of it is technical. Towards the middle

they start talking about the proteins e.g. osteopontin etc that are

increased in MS plaques etc. Worth reading the original from the

library (Nature reviews, microarrays collection, March 2004). The

present article is based on work done in 2003.

Yash

June 2003 Vol 3 No 6 REVIEW

Nature Reviews Immunology 3, 483-492 (2003); doi:10.1038/nri1108

[956K]

TRANSCRIPTIONAL ANALYSIS OF TARGETS IN MULTIPLE SCLEROSIS

Lawrence Steinman & Zamvil about the authors

Beckman Center for Molecular Medicine B002, Stanford University,

Stanford, California 94305, USA.

correspondence to: Lawrence Steinman steinman@...

Large-scale analyses of messenger RNA transcripts and autoantibody

responses, taken from the actual sites of disease, provide us with an

unprecedented view of the complexity of autoimmunity. Despite an

appreciation of the large number of pathways and pathological

processes that are involved in these diseases, a few practical

targets and several new strategies have emerged from these studies.

This review focuses on multiple sclerosis and on the approaches that

are being used to identify new targets that might be manipulated to

control this disease.

Our understanding of pathological processes in autoimmunity has been

limited until recently by an inability to follow these complex

processes on a large scale. For example, it has been impractical to

measure the immune response to all the components of a complicated

target organ. Although it has been possible to measure a few gene

transcripts (messenger RNA) to see if these genes are differentially

activated at the site of disease, it has not been practical to

measure the mRNA levels of all the genes in the genome at the site of

disease, until recently. The advent of technologies for the large-

scale analysis of transcription1, 2 and of autoantibody responses3

has enabled an unprecedented appreciation of the complexity of

autoimmune disease. This review focuses on multiple sclerosis (MS),

an autoimmune disease of the central nervous system (CNS) that

culminates in neurodegeneration4-6.

MS affects approximately one million people worldwide, with 350,000

cases in North America alone3, 4. Women are affected two times more

frequently than men4-6. The disease often begins in young adulthood

with recurrent inflammatory attacks against the white matter of the

brain, producing a myriad of neurological impairments, including

blindness, loss of sensation, lack of coordination, bowel and bladder

incontinence, and difficulty walking5. About one third of patients

who start out with the relapsing–remitting form of MS progress to a

more chronic form with more widespread disability.

The hallmark pathology in MS is the plaque — an area of myelin that

has been denuded by inflammation and subsequent scarring by non-

neural cells in the brain, including bone-marrow-derived MICROGLIA

and brain-derived star-shaped astroglia. The cause of MS is

enigmatic, although most investigators believe that the immune attack

against white matter is paramount, with the resulting degeneration of

axons and myelin being secondary to this inflammatory process4, 6

(Fig. 1). However, the possibility that the immune response is itself

a reaction to some initiating neurodegenerative process must also be

considered, analogous to the sequence of events that occur during

wound healing. First there is the injury, then the process of repair;

the repair process itself, in response to the wound, evokes a cascade

of immunological activity. Therefore, by examining mRNA levels on a

large scale from specimens taken both early and late in the disease

process, it might be possible to assess whether neurodegeneration is

actually the primary response and neuroinflammation of the white

matter is a secondary response.

Figure 1 | The inflammatory phase of multiple sclerosis.

T cells, B cells and antigen-presenting cells (APCs), including

macrophages, enter the central nervous system (CNS), where they

secrete certain chemicals known as cytokines that damage the

oligodendroglial cells. These cells manufacture the myelin that

insulates the neuronal axon. The injured myelin cannot conduct

electrical impulses normally, just as a tear in the insulation of a

wire leads to a short circuit. Lymphocytes diapedese into the CNS

through use of a surface receptor known as 4-integrin. This step is

impeded by antibodies specific for 4-integrin or by interferon- (IFN-

). Once the blood–brain barrier is breached, other inflammatory cells

accumulate in the white matter. Inside the brain, T cells and

accompanying macrophages and microglial cells release osteopontin

(OPN), interleukin-23 (IL-23), IFN- and tumour-necrosis factor (TNF),

all of which damage the myelin sheath. Also, the presence of OPN

might lead to the attraction of T helper 1 (TH1) cells. T-cell

activation can be blocked by altered peptide ligands (APLs), such as

copaxone, or by statins. Concomitantly, B cells (plasma cells)

produce myelin-specific antibodies, which interact with the terminal

complex in the complement cascade to produce membrane-attack

complexes that further damage oligodendroglial cells. DNA vaccination

can be used to tolerize T- and B-cell responses to myelin.

Large-scale approaches

Large-scale sequencing of mRNA from complementary DNA libraries

derived from brain plaques of patients with MS1, and gene microarray

analysis of transcripts from MS lesions of various types2, have shown

both the complexity of the pathological response in MS and several

targets that might be manipulated to control this disease. The

general approach of using a broad-ranging analysis of the immune

response to search for unusual patterns has so far led to some

practical approaches for the containment of this disease. Several

strategies have been used to analyse mRNA transcripts that are over

or under expressed in tissue from patients with MS, enabling

transcripts that are unique to MS plaques or that are differentially

expressed in acute compared with chronic MS material to be

identified. As shown in Fig. 2, we have used two parallel approaches.

The first strategy involved large-scale sequencing of mRNA

transcripts from cDNA libraries prepared from brain tissue. The

second approach used OLIGONUCLEOTIDE MICROARRAYS1, 2, 7. These

strategies in no way diminish the importance of other 'genomic'

approaches that are being undertaken to understand the regulation of

the immune response. Recently, several groups have embarked on the

analysis of SINGLE-NUCLEOTIDE POLYMORPHISMS (SNPs) across the entire

genome for evidence of linkage to susceptibility to MS. Discussions

of the various approaches for the statistical analysis and

hierarchical clustering of data are beyond the scope of this review.

The reader is referred to our own papers, and reviews of them1, 2, 7,

and to the discussions by Schadt and colleagues8 on statistical

analysis, and by Chu and colleagues9 on hierarchical clustering.

Figure 2 | Large-scale analysis of gene transcription from MS

lesions.

a | Robotic sequencing of expressed sequence tags (ESTs) produced by

reverse transcription of messenger RNA from multiple sclerosis (MS)

lesions. b | Strategy for the analysis of complementary RNA tagged

with biotin on oligonucleotide microarrays. These microarrays either

contain oligonucleotides that are printed on the surface by a process

similar to photolithography, where certain chemical groups are

protected whereas others are printed, or are made by printing

complementary DNA on a glass surface. The output from these

microarrays is analysed by statistical methods that allow groupings

to become apparent. Cluster analysis or statistical analysis of

microarrays (SAM) looks at correlations between different variables.

This is similar to the methods used by pollsters, who might correlate

variables such as voting preference and socioeconomic status.

Statistical methods for handling data in hierarchical clusters are

reviewed in Refs 1,2,8,9.

High-throughput sequencing of cRNA from EXPRESSED SEQUENCE TAGS

(ESTs), using non-normalized cDNA brain libraries generated from

brain lesions of patients with MS and from control individuals, has

indicated the most prominent transcripts that are found in the brains

of patients with MS. Using this protocol, the mRNA populations that

are present in the brain specimens are accurately represented,

enabling the quantitative estimation of transcripts and comparisons

between specimens. We have sequenced more than 11,000 clones from

these libraries from patients with MS and from controls1, and have

concentrated our analysis on mRNA species that are present in MS

libraries, but absent in control libraries. This analysis yielded 423

genes, including 26 novel genes. Of those, 54 genes showed a mean

increase in their level of expression of 2.5-fold or greater in

libraries derived from patients with MS compared with controls.

Transcripts encoding B-crystallin, an inducible heat-shock protein

that is localized in the myelin sheath and targeted by T cells in MS,

were the most abundant transcripts found to be unique to MS plaques.

The next five most abundant transcripts included those that encode

prostaglandin D synthase (PGDS), prostatic binding protein, ribosomal

protein L17 and osteopontin (OPN; also known as ETA1). The potential

role of OPN in the progression of MS is discussed in detail later1.

Few studies have looked at transcriptional profiles in MS lesions. We

compared our results with those of Biddison and colleagues10, who

used cDNA microarrays to profile MS lesions. They studied two MS

lesions from one brain and identified 29 genes that had an increased

level of expression in acute MS plaques compared with control brain.

All of these 29 genes were represented on the HuGeneFL chip that we

used in our study2, with the exception of 2-chimerin, which was

replaced by chimerin. We found an increased level of expression of

eight of these 29 genes in at least two of the four MS samples2.

Another recent study by Selmaj and colleagues11 directly compared

different regions of MS lesions with different activity from the same

individuals. A comparison of raw data sets between the study by

Selmaj and colleagues11 and the other previously reported analyses1,

2, 10 has not been carried out yet. The studies by Chabas and

colleagues1, Lock and colleagues2, and Whitney and colleagues12

looked for similarities between MS lesions and lesions from the

animal model EXPERIMENTAL AUTOIMMUNE ENCEPHALOMYELITIS (EAE)1, 2, 12.

Other studies13, 14 have analysed transcriptional profiles of EAE

lesions. The investigation by Ibrahim and colleagues13 looked at

acute EAE induced by injection of myelin oligodendrocyte glycoprotein

(MOG) in complete Freund's adjuvant (CFA) plus Bordetella pertussis

toxin, which is an additional immune adjuvant. The CFA and B.

pertussis toxin might have contributed to the alterations in gene

transcription that were observed. Furthermore, only two time points

on day 16 and day 22 after immunization, at the onset and peak of

acute clinical disease, were examined. It would be useful in the

future to study not only a model of acute EAE, but also one of the

several other varieties of this disease15. It is also desirable to

compare such results from different models of EAE with those obtained

from studies of MS tissue, similar to the attempts that have been

made by groups from Stanford and the National Institutes of Health1,

2, 12.

The pros and cons. It is possible to look at the transcription level

of all genes in the genome at the site of disease using microarrays

or robotic sequencing. This allows an unprecedented view of gene

activity. However, the alternative splicing of genes might be missed

or mRNA transcripts that are unproductive might be detected.

Therefore, it is crucial to support transcriptional analysis with

identification of the translated protein at the site of disease,

preferably using immunohistochemistry. Then, of course, it is

necessary to understand the biology of the gene of interest, to

understand the role that it might have in the pathogenesis of

disease. This is a major challenge in terms of time and financial

resources, and given the large amount of data resulting from large-

scale transcriptional analysis, it is difficult to mount an in-depth

effort to understand the biological role of more than a few

interesting genes at any one time. Nevertheless, we describe here

some unexpected pathways involved in autoimmune disease, the

discovery of which has resulted from such studies.

Osteopontin: a role in MS progression

As mentioned earlier, molecular mining of two sequenced libraries

from brain plaques of patients with MS and their comparison with a

normal brain library (matched for size and tissue type, and

constructed using an identical protocol) showed that transcripts

encoding OPN are frequently detected in and exclusive to the mRNA

population of MS plaques. Given the known pro-inflammatory role of

OPN, we investigated the potential role of this protein in MS

progression. First, we examined the cellular expression pattern of

this protein in human MS plaques and in control tissue by

immunohistochemistry. In active MS plaques, OPN was found to be

expressed on microvascular endothelial cells and macrophages, and in

the white matter adjacent to plaques. Reactive astrocytes and

microglial cells also expressed OPN1.

Studies in EAE models. The role of OPN in inflammatory demyelinating

disease was next examined using two models of EAE in mice — one

relapsing–remitting model, and one model that is initially relapsing

but is followed by a progressive phase, often culminating in death.

The relapsing–remitting model of EAE was first used to compare the

cellular expression of OPN at different stages of the disease.

Disease was induced in SJL mice by immunization with a peptide of

proteolipid protein (PLP139–151) in CFA, and the animals were scored

daily for signs of disease. Histopathological analyses showed that

OPN was expressed broadly in microglia during both relapse and

remission from disease, and this expression was focused near

perivascular inflammatory lesions. In addition to the expression of

OPN by glial cells, expression by neurons was detectable during acute

disease and relapse, but not during remission. These results

indicated that the level of expression of OPN in lesions correlates

with the severity of disease.

The potential role of OPN in demyelinating disease was next tested

using OPN-deficient mice. EAE was induced using MOG35–55 in CFA in

OPN-deficient mice and wild-type controls. EAE was observed in all

OPN-deficient and -sufficient mice after immunization with MOG35–55.

However, the severity of disease was reduced in all animals in the

OPN-deficient group, and these mice were totally protected from EAE-

related death. So, OPN significantly influenced the course of

progressive EAE induced by MOG35–55 (Ref. 1).

The rate of relapses and remissions was tested. During the first 26

days, OPN-deficient mice had a distinct evolution of EAE, with a much

higher percentage of mice undergoing remissions compared with the

control mice. Although the clinical courses in the two groups were

different, there were similar numbers and appearances of inflammatory

foci in the CNS. Therefore, although OPN might not influence the

extent of the inflammatory response, this protein might influence

whether or not the course of disease is progressive, or whether

relapses and remissions develop.

To examine whether different immune responses were involved in OPN-

deficient and OPN-sufficient animals, we tested the profile of

cytokine expression in these mice. T cells from OPN-deficient mice,

compared with OPN-sufficient T cells, had a reduced proliferative

response to MOG35–55. In addition, the production of interleukin-10

(IL-10) was increased in T cells reactive to MOG35–55 in OPN-

deficient mice that had developed EAE, compared with T cells from OPN-

sufficient mice. At the same time, the production of interferon- (IFN-

) and IL-12 was reduced in cultures of spleen cells from OPN-

deficient mice stimulated with MOG. IFN- and IL-12 are important pro-

inflammatory cytokines in MS4, 6; the finding that there is reduced

production of these cytokines in OPN-deficient mice is consistent

with the hypothesis that OPN might have a crucial role in the

modulation of T helper 1 (TH1) immune responses in MS and EAE.

Sustained expression of IL-10 might, therefore, be an important

factor in the reversal of relapsing MS, and its absence might allow

the development of secondary progressive MS.

OPN might have pleiotropic functions in the pathogenesis of

demyelinating disease (Fig. 1). The production of OPN by glial cells

might attract TH1 cells, and its expression by glial and ependymal

cells might allow inflammatory T cells to penetrate the brain.

Finally, our data indicate that neurons might also secrete this pro-

inflammatory molecule and participate in the autoimmune process.

Potentially, neuronal secretion of OPN could modulate inflammation

and demyelination, and influence the clinical severity of the

disease. Consistent with this idea, a role for neurons in the

pathogenesis of MS and EAE has been described recently4, 6.

CD44 is a known ligand of OPN, mediating a decrease in the production

of IL-10 (Ref. 16). OPN-deficient mice produce increased levels of IL-

10 during the course of EAE. We showed recently that CD44-specific

antibodies prevented EAE17, indicating that the pro-inflammatory

effect of OPN in MS and EAE might be mediated by CD44. The binding of

OPN to its integrin fibronectin receptor V3 through the arginine–

glycine–aspartate tripeptide motif might also perpetuate TH1-cell-

mediated inflammation1, 16, 17. In active MS lesions, the V subunit

of this receptor is overexpressed by macrophages and endothelial

cells, and the 3 subunit is expressed on the luminal surface of

endothelial cells. Through its tripeptide-binding motif, OPN inhibits

the function of inducible nitric-oxide synthetase (iNOS), which is

known to participate in autoimmune demyelination1. So, in conclusion,

OPN is situated at several checkpoints that would allow diverse

activities in the course of autoimmune-mediated demyelination.

Recently, Cantor's group18 has described concordant results in

another model of EAE. They studied a model of relapsing EAE induced

by PLP172–183 in C57BL/6129 OPN-deficient mice that had been

backcrossed to C57BL/6 mice for six generations. Wild-type OPN-

sufficient mice on the C57BL/6129 background, which were littermates

of the OPN-deficient mice, were used as controls. The incidence of

clinical disease and the mean day of disease onset were similar in

OPN-deficient C57BL/6129 mice compared with controls. OPN-deficient

C57BL/6129 mice had lower maximum clinical disease scores and

recovered faster without spontaneous relapses in this model.

Decreased levels of inflammatory infiltration and demyelination were

seen in this model in the OPN-deficient C57BL/6129 mice. Therefore,

these results were even more marked than those seen by our group in

the MOG35–55-induced model of EAE1. In the study by Cantor's group18,

CD4+ T cells reactive to PLP172–183 produced less IFN- and tumour-

necrosis factor (TNF) after restimulation with the myelin peptide.

Furthermore, OPN-deficient C57BL/6129 mice produced increased levels

of IL-10 after co-stimulation with CD3-specific antibody, compared

with wild-type littermates, indicating that a shift to a TH2-cell

phenotype had occurred. Taken together, our results1 and those of

Cantor's group18 indicate that OPN is a potent modulator of

autoimmune demyelinating disease. The role of OPN in the pathogenesis

of MS is described in Fig. 1. Note the inclusion of IL-23, a recently

discovered cytokine that contains the p40 subunit of IL-12 (Ref. 19).

The potential interactions between IL-23 and OPN are the subject of

intense investigation. It will be important to know where IL-23, IL-

12 and OPN fit into the hierarchy of regulation of TH1-cell responses

in the brain in human autoimmune disease.

OPN polymorphisms and disease. Further studies have been undertaken

looking at OPN polymorphisms and disease course in patients with MS.

In 821 patients with MS that were analysed, a trend for association

with disease course was detected in patients carrying at least one

1284A allele of the OPN gene, indicating that this polymorphism has

an effect on disease course. Patients with this genotype were less

likely to have a mild disease course and were at increased risk for a

secondary-progressive clinical type of disease20. In patients with MS

in Japan, polymorphisms in OPN were shown to be crucial for

determining susceptibility to progressive or relapsing MS21. Levels

of OPN were increased in the spinal fluid of patients with MS during

relapses22. In addition, OPN mRNA levels were increased in patients

with Huntington's disease, and levels of OPN mRNA and protein were

increased in brain tissue from a transgenic mouse model of

Huntington's disease after successful treatment with an inhibitor of

a key enzyme, transglutaminase, that is involved in the pathogenesis

of this neurodegenerative disease23. Levels of OPN mRNA and protein

were also increased in an experimental model of epilepsy24. So, the

role of OPN in a wide variety of neuroinflammatory and

neurodegenerative conditions continues to be an area worthy of

intense interest. Contradictory results published by Blom and

colleagues25 might be explained at several levels26.

The results described by Blom and Holmdahl25 might be due to

differences between the EAE model that they studied and the models

used by our group1 and by Cantor's group18. Holmdahl's group induced

EAE using recombinant rat MOG1–125, whereas we used a peptide

fragment of MOG, MOG35–55, which is the immunodominant epitope in

C57BL/6 mice. Cross and colleagues27, 28 have shown that the form of

inducing antigen (protein compared with peptide) has a role in the

pathogenesis of EAE27, such that B-cell-deficient mice do not develop

EAE when immunized with full-length MOG, although they are

susceptible to MOG35–55-induced disease. So, there are likely to be

different mechanisms at work in EAE induced using recombinant

proteins compared with that induced using peptide fragments of the

same protein. Furthermore, the recombinant MOG1–125 of rat origin

used by Holmdahl and colleagues has many differences compared with

the mouse sequence. The importance of OPN in EAE is further shown by

the fact that DNA vaccination to OPN ameliorated EAE, by inducing OPN-

specific antibodies26. Finally, to implicate linked genes

definitively as the basis for the differences seen between the data

of Holmdahl and colleagues25 and those of our group1 and of Cantor's

group18, EAE induced in earlier backcrosses of their25 OPN-deficient

strain would have to be examined. Until such differences are

explicitly analysed, it is difficult to draw any definitive

conclusions. It is not at all unusual for divergent results to arise

from studies of autoimmunity in knockout mice, especially when

different animal models are examined26, 29.

Lipid and cholesterol metabolism

Transcriptional profiling of MS tissue has indicated many changes in

genes that are involved in lipid and cholesterol metabolism2.

Expression of 3-hydroxy-3-methylglutaryl coenzyme A reductase (HMG-

CoA reductase) was downregulated in brain tissue from MS patients, as

were the expression levels of other genes involved in crucial

pathways of lipid metabolism, such as those encoding stearoyl CoA

desaturase (SCD), acetoacetyl-CoA thiolase, propionyl-CoA carboxylase

and enoyl-CoA hydratase. Initially, we have focused our attention on

the potential role of HMG-CoA reductase, because of its pleiotropic

effects on the immune system, including reduction in the expression

of inducible MHC class II molecules and the ability to block

leukocyte function-associated antigen 1 (LFA1) and its interactions

with intercellular adhesion molecule 1 (ICAM)30-32. The class of

drugs known as statins reduce cholesterol synthesis by inhibiting HMG-

CoA reductase. We showed more than twenty years ago that inhibition

of expression of MHC class II molecules could reverse autoimmune

disease in several animal models, including EAE, experimental

autoimmune myasthenia gravis and experimental autoimmune

thyroiditis30, 33-35. Recently, promising results in pre-clinical

studies have ignited interest in the potential application of the

cholesterol-lowering HMG-CoA reductase inhibitors (statins) for the

therapy of MS36-38.

In contrast with currently approved treatments for MS, which are

administered parenterally, statins are given orally and are well

tolerated. Atorvastatin (Lipitor®), which is given orally and is

currently the most potent statin for cholesterol reduction, can

prevent or reverse chronic and relapsing EAE36. In vivo treatment

with atorvastatin induced the secretion of anti-inflammatory TH2

cytokines by T cells and suppressed the secretion of TH1 cytokines by

T cells. In vitro, atorvastatin promoted TH0 cells to differentiate

into TH2 cells, which could adoptively transfer protection against MS

to recipient mice. Atorvastatin also suppressed the expression of MHC

class II molecules by microglia and, when tested in vitro, it

inhibited IFN--inducible expression of co-stimulatory and MHC class

II molecules by antigen-presenting cells (APCs). So, atorvastatin has

immunomodulatory effects on both APCs and T cells. These findings

indicate that statins could be useful in the early inflammatory phase

of MS (Fig. 1), as well as in the neurodegenerative phase of MS (Fig.

3). Statins also inhibit the secretion of metalloproteases, which

indicates that they might prevent T cells from entering the CNS.

Lovastatin, which partially suppressed acute EAE in rats, was shown

to inhibit the production of iNOS and TNF — pro-inflammatory

molecules that are neurotoxic — indicating that statins might be

beneficial in the chronic phase of MS38. At the moment, there is one

open-label trial testing simvastatin in relapsing–remitting MS being

carried out, and other trials using atorvastatin are being planned.

Because of their anti-inflammatory and potential neuroprotective

effects, statins are also being evaluated for potential efficacy for

the treatment of other CNS disorders, including Alzheimer's disease36.

Figure 3 | The neurodegenerative phase of multiple sclerosis.

T cells and antigen-presenting cells (APCs) such as macrophages

produce glutamate, a toxic substance, which injures oligodendroglial

cells and underlying axons. The activation of T cells can be blocked

by various approaches, including statins and altered peptide ligands

(APLs). Statins also inhibit the secretion of metalloproteases, and

it has been suggested that they might, therefore, also act to block T

cells from entering the central nervous system (CNS). Antibodies

specific for 4-integrin also block the movement of T cells into the

brain. OPN, osteopontin.

Neuroendocrine mediators in autoimmunity

Large-scale analysis of gene transcripts in MS lesions showed that

levels of the neuroendocrine mediators leptin, melanocortin-4

receptor and adrenocorticotropic hormone receptor (ACTHR) are

increased at the site of inflammation in the brain1, 2.There are

indications that other neuroendocrine mediators, such as the

melanocortins2 and corticotrophin-releasing factor (CRF)2, 39-42,

have important roles in diseases such as MS and rheumatoid arthritis

that are thought to be autoimmune in nature. Leptin is thought to

produce its profound effects on appetite and body weight by altering

the balance between the anorectic (appetite suppressing)

neuropeptides -melanocyte-stimulating hormone (-MSH) and CRF, and the

orexigenic (appetite enhancing) neuropeptides Agouti-related protein

(AGRP) and neuropeptide Y (NPY). Shifts in the balance between these

neuropeptides also contribute to the cytokine-mediated loss of

appetite and body weight, and malaise that is induced by

lipopolysaccharide39. One important mechanism of this regulation

occurs at the level of the pro-opiomelanocortin (POMC)-expressing

neurons in the arcuate nucleus of the hypothalamus, which are

activated by leptin to release -MSH39. It seems that this interplay

between neuropeptides in relation to hypothalamic function might be

reproduced at sites of autoimmune pathology (Fig. 4). All three of

these neuropeptides — leptin, melanocortin 4 and ACTH — have

pleiotropic roles in neuroendocrine and immune physiology.

Figure 4 | The intricate interplay between leptin, corticotropin-

releasing factor and the melanocortins in the regulation of TH1

immunity.

Leptin, corticotropin-releasing factor (CRF) and the melanocortins

modulate T helper 1 (TH1)-cell autoimmunity. Decreases in the level

of body fat or starvation decrease leptin levels and increase CRF

levels. Increased levels of CRF downregulate TH1-cell-mediated

autoimmunity, as do decreased levels of leptin. Other molecules of

the -melanocyte-stimulating hormone (MSH) family modulate TH1-cell-

mediated autoimmunity through their melanocortin-4 receptor (MC4R).

Leptin R, leptin receptor; CRFR, corticotropin-releasing factor

receptor. Reproduced, with permission from The Journal of Clinical

Investigation, from Ref. 39.

The microsomal enzyme SCD1 is required for biosynthesis of the

monounsaturated fats palmitoleate and oleate from saturated fatty

acids39, 43. Scd1 mRNA levels are highly elevated in the livers of

ob/ob mice, which contain a mutation in the leptin receptor and

develop obesity. Indeed, SCD1 probably has a decisive role in the

metabolic effects of leptin. Interestingly, SCD1 was downregulated in

brain tissue from patients with MS2, and both mRNA levels and

activity of this enzyme are repressed by leptin39. ob/ob mice with

additional mutations in Scd1 are markedly less obese than ob/ob

controls39. The role of leptin in autoimmune brain disease, and in

the immune system in general, might be mediated by downregulation of

this enzyme (SCD1) involved in the biosynthesis of monounsaturated

fats. Once again, we witness the remarkable choreography of molecules

that are involved in the regulation of body weight and energy

metabolism, and the parallel roles of these same molecules in the

finely tuned immune response (Fig. 4 and Table 1).

Table 1 | Validation of the role of neuroendocrine genes found in

MS lesions

Earlier work had shown that CRF, which is the main regulator of the

stress response in the hypothalamus–pituitary–adrenal axis, or

urocortin, which is a naturally occurring paralogue of CRF, acting

directly on T cells in adrenalectomized mice ameliorated EAE.

Antagonists of CRF blocked these effects42. So, CRF, similar to

leptin, is produced by the brain and might act directly on the immune

system. Expression of CRF itself can be regulated by cytokines,

adding another layer of complexity and a further target for

intervention. Another neuropeptide, ACTH, which is an important

mediator of the stress response produced in the pituitary gland, has

been used for more than 40 years to treat patients with MS. ACTHR is

expressed in MS lesions2.

These results involving neuroendocrine mediators of autoimmunity

imply that stress might be beneficial in autoimmunity and that brief

interludes of starvation might help to reverse disease. In fighting

microbial infections, perhaps 'stress' is detrimental and 'eating' is

recommended, but in the case of autoimmunity, the opposite appears to

hold39.

Connections between allergy and autoimmunity

Previous work has shown that self-antigens can trigger allergic

responses. In our recent work44, 45, we have brought to attention

Ehrlich's century-old scenario for autoimmunity, 'horror

autotoxicus', in which the immune system attacks the body's own

tissues44. Large-scale transcriptional sequencing of MS lesions has

shown that there are a large number of allergy-related gene

transcripts in MS lesions (Table 2). These transcripts include those

encoding prostaglandin D1, platelet-activating factor receptor

(PTAFR), tryptase, Fc receptor (FcR) and eosinophilic cationic

protein2, 45.

Table 2 | Validation of the role of allergy-associated genes found

in MS lesions

Since Rivers' description of EAE more than 60 years ago46, our

concepts of allergy and autoimmunity have been highly dichotomous46-

48. However, this distinction has been increasingly blurred as drugs

that are commonly used for the treatment of allergic diseases, such

as antihistamines, have been shown to ameliorate EAE46-48.

Furthermore, an association between sensitivity to histamine and

susceptibility to EAE has been described46-48. B. pertussis toxin

(PTX), which increases vasoactive-amine sensitization, is required as

an adjuvant to induce EAE in those strains of mice that are not

physiologically sensitive to histamine. B. pertussis histamine

sensitization (Bphs) is the gene controlling PTX-induced vasoactive-

amine sensitization, and susceptibility to EAE and other autoimmune

diseases is linked to an allele of this gene49. Interestingly,

Teuscher and colleagues49 reported recently that Bphs encodes the

histamine 1 receptor (H1R). Finally, the use of large-scale analysis

of gene transcripts from MS lesions has identified several molecules

that can also have important roles in the allergic response. Taken

together, this evidence indicates that components of classical

allergic responses can also markedly influence the pathogenesis of

autoimmune disease in the EAE model.

Several molecules that can have important roles in allergic responses

were shown to participate in EAE1, 2, 46-49. In MS lesions, we have

shown that there is an increased level of transcription of the genes

encoding H1R, PTAFR, tryptase, FcRI and PGDS. Transcripts encoding

tryptase, PTAFR and PGDS were present at an increased level in the

CNS in EAE. Moreover, the expression of H1R was increased by TH1

cells reactive to myelin, and immunohistochemical staining showed

that H1R and H2R are present in inflammatory lesions. EAE was

ameliorated in mice with disruptions of the -chain that is common to

FcRIII and FcRI. Even if TH1 cells are the main contributors to the

pathogenesis of EAE and MS, molecules that are involved in allergic

responses can potently modulate the disease also. In accordance with

this conclusion, both H1R antagonists and PTAFR antagonists markedly

blunted EAE46-48.

A proteomic approach to autoimmune disease

Perhaps again with homage to Ehrlich's ideas44, we can use DNA as a

drug to manufacture a 'magic bullet' to reverse the course of

autoimmunity. We have recently described a technology that allows the

large-scale analysis of autoantibody responses50, 51. We have adapted

this technology to construct a myelin proteome, representing the main

myelin proteins and their peptide epitopes. Using this microarray, we

can analyse, on a large scale, the autoantibody response as it

spreads to various components of the myelin sheath and targets the

immune system for attack. Using an approach whereby DNA encoding

these myelin components can tolerize or even deviate autoaggressive

anti-myelin responses, we can now take the autoantibody data from the

myelin microarray and construct a vaccine consisting of DNA encoding

these myelin targets52, 53.

The use of DNA vaccines can even be applied on a wider scale. Using

DNA constructs that actually target pathogenic cytokines, rather than

promoting tolerance to self, we can reverse the effects of pathogenic

molecules such as OPN by using a DNA vaccine to this pathogenic

cytokine26. Such an approach can reverse ongoing paralytic disease.

This approach has been used to target various pathogenic chemokines

and cytokines26, 54-57. So, DNA, when used as a drug, can rapidly and

conveniently target molecules that have been discovered by the large-

scale proteomic and transcriptional analysis of diseased tissues. DNA

itself can be used as a 'magic bullet'. This 'reverse proteomic'

approach, using DNA as a drug, is shown in Fig. 5.

Figure 5 | The 'reverse proteomic' approach.

Autoantibody microarrays3 are constructed to identify the nature of

the immune response. The specific autoantigens that have triggered

immunity can then be targeted, and the immune response to them

suppressed by encoding the autoantigens in a DNA plasmid. The plasmid

contains specially engineered immunosuppressive DNA sequences50, 51,

54.

Conclusions

Large-scale transcriptional profiling of lesions from the brains of

patients with MS has determined a large number of new targets and new

pathways for potential interventions. The course of inflammation has

indicated molecules that most experts would have considered unlikely

to have a role in autoimmune disease. OPN, neuroendocrine mediators,

enzymes involved in the metabolism of cholesterol, and molecules

associated with allergy all have important roles in the pathogenesis

of MS. Large-scale approaches to understanding the immune response

can be extended to proteomics, and these new technologies reveal

further complexities of autoimmune inflammation.

Links

DATABASES

LocusLink: B-crystallin | acetoacetyl-CoA thiolase | ACTHR | AGRP |

CD44 | CRF | enoyl-CoA hydratase | H1R | HMG-CoA reductase | ICAM |

IFN- | IL-10 | IL-12 | IL-23 | iNOS | leptin | LFA1 | melanocortin-4

receptor | MOG | NPY | OPN | POMC | propionyl-CoA carboxylase | PGDS

| prostatic binding protein | proteolipid protein | PTAFR | ribosomal

protein L17 | Scd1 | stearoyl-CoA desaturase | TNF | transglutaminase

| urocortin

OMIM: multiple sclerosis

References

1.

Chabas, D. et al. The influence of the proinflammatory cytokine,

osteopontin, on autoimmune demyelinating disease. Science 294, 1731-

1735 (2001).

This study describes the sequencing of messenger RNA transcripts

found in the brain using robotics. The biological role of osteopontin

(OPN), one of the sequences that are found in the brain of patients

with multiple sclerosis (MS), is described. | Article | PubMed | ISI

| ChemPort |

2.

Lock, C. et al. Gene-microarray analysis of multiple sclerosis

lesions yields new targets validated in autoimmune encephalomyelitis.

Nature Med. 8, 500-508 (2002).

This paper shows how oligonucleotide microarrays are used to assess

mRNA transcripts in acute and chronic MS lesions. The biological

roles of some of these transcripts, including those encoding G-CSF

and Fc receptor, are shown. | Article | PubMed | ISI | ChemPort |

3.

, W. H. et al. Antigen arrays for multiplex characterization

of autoantibody responses. Nature Med. 8, 295-301 (2002). | Article

| PubMed | ISI | ChemPort |

4.

Steinman, L. Multiple sclerosis: a two stage disease. Nature

Immunol. 2, 762-765 (2001). | Article | PubMed | ISI | ChemPort |

5.

Steinman, L. Autoimmune disease. Sci. Am. 269, 106-114 (1993). |

PubMed | ISI | ChemPort |

6.

Steinman, L., , R., Bernard, C. C. A., Conlon, P. & Oksenberg,

J. R. Multiple sclerosis: deeper understanding of its pathogenesis

reveals new targets for therapy. Annu. Rev. Neurosci. 25, 491-505

(2002). | Article | PubMed | ISI | ChemPort |

7.

Dyment, D. & Ebers, G. An array of sunshine in multiple sclerosis.

N. Engl. J. Med. 247, 1445-1447 (2002).

This paper reviews the promise of microarray technologies to discover

new targets in MS. | Article |

8.

Schadt, E. E., Li, C., Su, C. & Wong, W. H. Analyzing high-density

oligonucleotide gene expression array data. J. Cell Biochem. 80, 192-

202 (2000). | Article | PubMed | ISI | ChemPort |

9.

Tusher, V. G., Tibshirani, R. & Chu, G. Significance analysis of

microarrays applied to the ionizing radiation response. Proc. Natl

Acad. Sci. USA 98, 5116-5121 (2001). | Article | PubMed | ChemPort |

10.

Whitney, L. W. et al. Analysis of gene expression in multiple

sclerosis lesions using cDNA microarrays. Ann. Neurol. 46, 425-428

(1999). | Article | PubMed | ISI | ChemPort |

11.

Mycko, M. P., Papoian, R., Boschert, U., Raine, C. S. & Selmaj, K.

W. cDNA microarray analysis in multiple sclerosis lesions: detection

of genes associated with disease activity. Brain 126, 1048-1057

(2003). | Article | PubMed | ISI |

12.

Whitney, L. W., Ludwin, S. K., McFarland, H. F. & Biddison, W. E.

Microarray analysis of gene expression in multiple sclerosis and EAE

identifies 5-lipoxygenase as a component of inflammatory lesions. J.

Neuroimmunol. 121, 40-48 (2001). | Article | PubMed | ISI | ChemPort |

13.

Ibrahim, S. M. et al. Gene expression profiling of the nervous

system in murine experimental autoimmune encephalomyelitis. Brain

124, 1927-1938 (2001). | Article | PubMed | ISI | ChemPort |

14.

Nicot, A., Ratnakar, P. V., Ron, Y., Chen, C. C. & Elkabes, S.

Regulation of gene expression in experimental autoimmune

encephalomyelitis indicates early neuronal dysfunction. Brain 126,

398-412 (2003). | Article | PubMed | ISI |

15.

Steinman, L. Gene microarrays and experimental demyelinating

disease: a tool to enhance serendipity. Brain 124, 1897-1899 (2001).

| Article | PubMed | ISI | ChemPort |

16.

Weber, G. F., Ashkar, S., Glimcher, M. J. & Cantor, H. Receptor-

ligand interaction between CD44 and osteopontin (Eta-1). Science 271,

509-512 (1996). | PubMed | ISI | ChemPort |

17.

Brocke, S., Piercy, C., Steinman, L., Weissman, I. L. & Veromaa, T.

Antibodies to CD44 and integrin 4, but not L-selectin, prevent CNS

inflammation and experimental encephalomyelitis by blocking secondary

leukocyte recruitment. Proc. Natl Acad. Sci. USA 96, 6896-6901

(1999). | Article | PubMed | ChemPort |

18.

Jansson, M., Panoutsakapoulou, V., Baker, J., Klein, L. & Cantor, H.

Attenuated EAE in Eta-1/osteopontin deficient mice. J. Immunol. 168,

2096-2099 (2002). | PubMed | ISI | ChemPort |

19.

Cua, D. et al. Interleukin-23 rather than interleukin-12 is the

critical cytokine for autoimmune inflammation of the brain. Nature

421, 744-748 (2003).

This paper contends that IL-23 is paramount in the cytokine cascade

in experimental autoimmune encephalomyelitis (EAE). | Article |

PubMed | ISI | ChemPort |

20.

Caillier, S. et al. Osteopontin polymorphisms and disease course in

MS. Genes Immun. 4, 312-316 (2003). | Article | PubMed | ISI |

ChemPort |

21.

Niino, M., Kikuchi, S., Fukazawa, T., Yabe, I. & Tashiro, K. Genetic

polymorphisms of osteopontin in association with multiple sclerosis

in Japanese patients. J. Neuroimmunol. 136, 125-129 (2003). | Article

| PubMed | ISI | ChemPort |

22.

Vogt, M. et al. Elevated osteopontin levels are associated with

disease activity in relapsing-remitting MS patients. Ann. Neurol. (in

the press).

23.

Karpuj, M. V. et al. Prolonged survival and decreased abnormal

movements in transgenic model of Huntington's disease, with

administration of cystamine, a transglutaminase inhibitor. Nature

Med. 8, 143-149 (2002). | Article | PubMed | ISI | ChemPort |

24.

Graber, K. D., Fontoura, P., Hermans, G., Steinman, L. & Prince, D.

A. mRNA changes in epileptogenic and nonepileptogenic undercut rat

neocortex. Epilepsia 43, 21 (2002).

25.

Blom, T., Franzen, A., Heinegard, D. & Holmdahl, R. Comment on " The

influence of the proinflammatory cytokine, osteopontin, on autoimmune

demyelinating disease " . Science 299, 1845 (2003). | Article | PubMed

| ChemPort |

26.

Steinman, L. et al. Response to comment on " The influence of the

proinflammatory cytokine, osteopontin, on autoimmune demyelinating

disease " . Science 299, 1845 (2003). | Article | PubMed | ChemPort |

27.

Lyons, J. A., Ramsbottom, M. J. & Cross, A. H. Critical role of

antigen-specific antibody in experimental autoimmune

encephalomyelitis induced by recombinant myelin oligodendrocyte

glycoprotein. Eur. J. Immunol. 32, 1905-1913 (2002). | Article |

PubMed | ISI | ChemPort |

28.

Lyons, J. A., San, M., Happ, M. P. & Cross, A. H. B cells are

critical to induction of experimental allergic encephalomyelitis by

protein but not by a short encephalitogenic peptide. Eur. J. Immunol.

29, 3432-3439 (1999). | Article | PubMed | ISI | ChemPort |

29.

Steinman, L. Some misconceptions about understanding autoimmunity

through experiments with knockouts. J. Exp. Med. 185, 2039-2041

(1997). | Article | PubMed | ISI | ChemPort |

30.

Steinman, L., Rosenbaum, J. T., Sriram, S. & McDevitt, H. O. In vivo

effects of antibodies to immune response gene products: prevention of

experimental allergic encephalomyelitis. Proc. Natl Acad. Sci. USA

78, 7111-7114 (1981). | PubMed | ChemPort |

31.

Kwak, B., Mulhaupt, F., Myit, S. & Mach, F. Statins as a newly

recognized type of immunomodulator. Nature Med. 6, 1399-1402 (2000).

| Article | PubMed | ISI | ChemPort |

32.

Bottazo, G. F., Pujol-Borrell, R., Hanufusa, T. & Feldmann, M. Role

of aberrant HLA-DR expression and antigen presentation in induction

of endocrine autoimmunity. Lancet 12, 1115-1119 (1983). | Article |

33.

Waldor, M., Sriram, S., McDevitt, H. O. & Steinman, L. In vivo

therapy with monoclonal anti-I-A antibody suppresses immune response

to acetylcholine receptor. Proc. Natl Acad. Sci. USA 80, 2713-2717

(1983). | PubMed | ChemPort |

34.

Sriram, S. & Steinman, L. Anti I-A antibody suppresses active

encephalomyelitis: treatment model for IR gene linked diseases. J.

Exp. Med. 158, 1362-1367 (1983). | PubMed | ISI | ChemPort |

35.

Vladutiu, A. & Steinman, L. Inhibition of experimental autoimmune

thyroiditis in mice by anti-I-A antibodies. Cell. Immunol. 109, 169-

180 (1987). | PubMed | ISI | ChemPort |

36.

Youssef, S. et al. The HMG-CoA reductase inhibitor, atorvastatin,

promotes a TH2 bias and reverses paralysis in CNS autoimmune disease.

Nature 420, 78-84 (2002).

This paper describes the surprising role of statins as

immunomodulators. Statins induce a T helper 2 (TH2) cytokine shift. |

Article | PubMed | ISI | ChemPort |

37.

Neuhaus, O. et al. Statins as immunomodulators: comparison with

interferon-1b in MS. Neurology 59, 990-997 (2002). | PubMed | ISI |

ChemPort |

38.

Stanislaus, R., Pahan, K., Singh, A. K. & Singh, I. Amelioration of

experimental allergic encephalomyelitis in rats by lovastatin.

Neurosci. Lett. 269, 71-74 (1999). | Article | PubMed | ISI |

ChemPort |

39.

Steinman, L., Conlon, P., Maki, R. & , A. The intricate

interplay among body weight, stress and the immune response to friend

or foe. J. Clin. Invest. 111, 183-185 (2003). | Article | PubMed |

ISI | ChemPort |

40.

Matarese, G. et al. Requirement for leptin in induction and

progression of experimental autoimmune encephalomyelitis. J. Immunol.

166, 5909-5916 (2001). | PubMed | ISI | ChemPort |

41.

Sanna, V. et al. Leptin surge precedes autoimmune encephalomyelitis

and correlates with disease susceptibility, inflammatory anorexia and

the development of pathogenic T cell responses. J. Clin. Invest. 111,

241-250 (2003).

A description of how leptins modulate TH1-cell-mediated autoimmunity.

Short periods of fasting can modulate TH1 immune responses. | Article

| PubMed | ISI | ChemPort |

42.

Poliak, S. et al. Stress and autoimmunity: the neuropeptides

corticotropin releasing factor and urocortin suppress

encephalomyelitis via effects on both the hypothalamic-pituitary-

adrenal axis and the immune system. J. Immunol. 158, 5751-5756

(1997). | PubMed | ISI | ChemPort |

43.

Cohen, P. et al. Role for stearoyl-CoA desaturase-1 in leptin-

mediated weight loss. Science 297, 240-243 (2002). | Article | PubMed

| ISI | ChemPort |

44.

Ehrlich, P. Die Schutzstoffe des Blutes. Dtsch. med. Wschr. 27, 913-

916 (1901) (in German).

45.

Pedotti, R. et al. An unexpected version of horror autotoxicus:

anaphylactic shock to a self-peptide. Nature Immunol. 2, 216-222

(2001). | Article | PubMed | ISI | ChemPort |

46.

Rivers, T. M., Sprunt, D. H. & Berry, G. P. Observations on attempts

to produce acute disseminated encephalomyelitis in monkeys. J. Exp.

Med. 58, 39-53 (1933).

47.

Pedotti, R. et al. Multiple elements of the allergic arm of the

immune response modulate autoimmune demyelination. Proc. Natl Acad.

Sci. USA 100, 1867-1872 (2003). | Article | PubMed | ChemPort |

48.

Steinman, L. Optic neuritis, a new variant of experimental

encephalomyelitis, a durable model for all seasons, now in its

seventieth year. J. Exp. Med. 197, 1065-1071 (2003). | Article |

PubMed | ISI | ChemPort |

49.

Ma, R. Z. et al. Identification of Bphs, an autoimmune disease

locus, as histamine receptor H1. Science 297, 620-623 (2002).

This paper shows that histamine receptors are crucial for

susceptibility to EAE. | Article | PubMed | ISI | ChemPort |

50.

, W., Garren, H., Utz, P. J. & Steinman, L. Millennium

Award. Proteomics for the development of DNA tolerizing vaccines to

treat autoimmune disease. Clin. Immunol. 103, 7-12 (2002). | Article

| PubMed | ISI | ChemPort |

51.

, W., Steinman, L. & Utz, P. J. Proteomics technologies for

the study of autoimmune disease. Arthritis Rheum. 46, 885-893 (2002).

| Article | PubMed | ISI | ChemPort |

52.

Garren, H. & Steinman, L. DNA vaccination in the treatment of

autoimmune disease. Curr. Direct. Autoimmun. 2, 203-216 (2000). |

ChemPort |

53.

Garren, H. et al. Combination of gene delivery and DNA vaccination

to protect from and reverse TH1 autoimmune disease via deviation to

the TH2 pathway. Immunity 15, 15-22 (2001). | Article | PubMed | ISI

| ChemPort |

54.

Urbanek-Ruiz, I. et al. Immunization with DNA encoding an

immunodominant peptide of insulin prevents diabetes in NOD mice.

Clin. Immunol. 100, 164-171 (2001). | Article | PubMed | ISI |

ChemPort |

55.

Youssef, S. et al. Long lasting protective immunity to experimental

autoimmune encephalomyelitis following vaccination with naked DNA

encoding C-C chemokines. J. Immunol. 161, 3870-3879 (1998). | PubMed

| ISI | ChemPort |

56.

Youssef, S., Wildbaum, G. & Karin, N. Prevention of experimental

autoimmune encephalomyelitis by MIP-1 and MCP-1 naked DNA vaccines.

J. Autoimmun. 13, 21-29 (1999). | Article | PubMed | ISI | ChemPort |

57.

Wildbaum, G., Youssef, S., Grabie, N. & Karin, N. Neutralizing

antibodies to IFN--inducing factor prevent experimental autoimmune

encephalomyelitis. J. Immunol. 161, 6368-6374 (1998). | PubMed | ISI

| ChemPort |

Acknowledgements

This work was funded, in part, by the Phil N. Trust, the

National Institutes of Health, the National Multiple Sclerosis

Society, the Foundation, the Maislin Foundation and the

Wadsworth Foundation. L.S. has founded two biotechnology companies,

Neurocrine Biosciences and Bayhill Therapeutics.

back to top

NATURE REVIEWS | IMMUNOLOGY

© 2003 Nature Publishing Group