Commensal Gut Derived Anaerobes As Novel Therapy For ...

Copy and paste this link to your website, so they can see this document directly without any plugins.



Keywords

mice, with, that, transgenic, were, disease, histicola, from, mice., treated, compared, showed, *0401, have, immune, cells, bacteria, *0402, levels, response, specific, class, data, role, cell, using, between, autoimmune, also, microbiome

Transcript

AD_________________
Award Number: W81XWH-10-1-0254
TITLE: Commensal gut derived anaerobes as novel therapy for inflammatory
autoimmune diseases
PRINCIPAL INVESTIGATOR: Dr. Ashutosh Mangalam
Dr. Veena Taneja
CONTRACTING ORGANIZATION: Mayo Clinic
Rochester, MN 55905
REPORT DATE: May 2012
TYPE OF REPORT: Annual
PREPARED FOR: U.S. Army Medical Research and Materiel Command
Fort Detrick, Maryland 21702-5012
DISTRIBUTION STATEMENT: Approved for Public Release;
Distribution Unlimited
The views, opinions and/or findings contained in this report are those of the author(s) and
should not be construed as an official Department of the Army position, policy or decision
unless so designated by other documentation.
REPORT DOCUMENTATION PAGE
Form Approved
OMB No. 0704-0188
Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the
data needed, and completing and reviewing this collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing
this burden to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 222024302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently
valid OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS.
1. REPORT DATE
May 2012
2. REPORT TYPE
Annual
3. DATES COVERED
15 April 2011 – 14 April 2012
4. TITLE AND SUBTITLE
5a. CONTRACT NUMBER
Commensal gut derived anaerobes as novel therapy for inflammatory autoimmune
diseases
5b. GRANT NUMBER
W81XWH-10-1-0254
5c. PROGRAM ELEMENT NUMBER
6. AUTHOR(S)
5d. PROJECT NUMBER
Ashutosh Mangalam
5e. TASK NUMBER
E-Mail: mangalam.ashutosh@mayo.edu
5f. WORK UNIT NUMBER
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES)
8. PERFORMING ORGANIZATION REPORT
NUMBER
Mayo Clinic
Rochester, MN 55905
9. SPONSORING / MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)
U.S. Army Medical Research and Materiel Command
Fort Detrick, Maryland 21702-5012
11. SPONSOR/MONITOR’S REPORT
NUMBER(S)
12. DISTRIBUTION / AVAILABILITY STATEMENT
Approved for Public Release; Distribution Unlimited
13. SUPPLEMENTARY NOTES
14. ABSTRACT
Rheumatoid arthritis (RA) and multiple sclerosis (MS) are chronic inflammatory autoimmune diseases affecting millions of
people. Here we are proposing a novel approach to cure MS, by administration of a specific strain of human commensalbacteria. Recent studies have shown that intestinal microflora plays an important role in the health of the host and posses
probiotics like qualities. We hypothesize that Gram-negative commensal bacteria from human gut have the potential to be
used as a therapeutic agent. We have used collagen induced arthritis (CIA) in HLA-DR4DQ8 mice and PLP91-110 induced
experimental autoimmune encephalomyelitis (EAE) HLA-DR3DQ8 mice to test our hypothesis that treatment with commensal
bacteria Prevotella histicola can modulate disease. First using various doses of bacteria, we have identified the optimal dose to
be used for treatment of CIA as well as EAE. Our in vitro and in vivo data showing suppression of antigen-specific immune
response in EAE and arthritis in P. histicola treated mice suggest generation of peripheral tolerance via gut. Our studies show
that treatment with P histicola suppresses arthritis/EAE via gut and generation of regulatory DCs and T regulatory cells. Our
data indicates that P histicola induced immune responses in the gut cause induction of immune tolerance in periphery leading
to suppression of antigen specific response.
15. SUBJECT TERMS
Commensal, inflammatory disease, therapeutic agent, multiple sclerosis, rheumatoid arthritis, HLA transgenic mice,
tolerance, P histicola.
16. SECURITY CLASSIFICATION OF:
17. LIMITATION
OF ABSTRACT
18. NUMBER
OF PAGES
19a. NAME OF RESPONSIBLE PERSON
USAMRMC
a. REPORT
U
b. ABSTRACT
U
c. THIS PAGE
U
UU
39
19b. TELEPHONE NUMBER (include area
code)
3
Table of Contents
Page
Introduction…………………………………………………………….………..….. 4
Body………………………………………………………………………………….. 5-14
Key Research Accomplishments………………………………………….…….. 15- 16
Reportable Outcomes……………………………………………………………… 17
Conclusion…………………………………………………………………………… 18
References……………………………………………………………………………. 19
Appendices…………………………………………………………………………… 20-21
4
This progress report is from Feb ‘2011 to April ’2012
Introduction
Rheumatoid arthritis (RA) and multiple sclerosis (MS) are chronic inflammatory autoimmune diseases
affecting millions of people. Since these diseases occur more often among young and middle-aged
adults, they cause significant loss of productive years in this population. Beside these autoimmune
diseases also cause significant economic burden (hundreds of Billions of USD) on society. Although
several therapies are in use, none of them cure disease. In this study we are investigating a unique
approach to ameliorate RA and MS, by administration of a specific strain of human commensal bacteria
Prevotella histicola (P histicola), which was recently isolated from human gut. Among all the genetic
factors linked with RA and MS, the strongest association has been with the MHC class II region on
chromosome 6 (1) and we have generated novel humanized HLA class II transgenic animal models of
RA and MS. We are utilizing these animal models to test the therapeutic efficacy of P histicola. Using an
experimental autoimmune encephalitomyelitis (EAE), which is an animal model of MS, we showed that
HLA-DR3DQ8 transgenic mice develop MS like disease characterized by brain plaques (2). HLADR3DQ8 mice were immunized with CNS antigen PLP91-110 and received either Prevotella histicola or
medium staring day 7 post-immunization and every other day for a total of 7 doses. Previously we have
shown that treatment of mice with 3-4 doses of P. histicola in PLP-immunized mice i) led to suppression
of antigen-specific immune response in vitro; ii) suppressed inflammatory cytokine IL-17 and increase in
levels of anti-inflammatory cytokine IL-10; and iii) lower clinical disease incidence as well as severity
indicating immunosuppressive properties of P histicola.
In 2nd year we have observed that P histocola treatment can suppress ongoing EAE in DR3DQ8.AEo
transgenic mice. Treatment of DR3DQ8 mice with EAE on day 11 with P histicola suppressed or down
regulated disease severity in 50% of mice. Next we investigated requirement of live vs. dead bacteria for
therapeutic effect and observed that live bacteria is needed for suppressing EAE as heat killed bacteria
failed to suppress EAE in DR3DQ8 transgenic mice significantly. We also performed histopathology of
brain as well as spinal cord and have observe that P histocola treated group have reduced brain and
spinal cord pathology compared to medium treated group. Treated group also showed reduced number
of inflammatory cells in CNS as well as reduced level of inflammatory cytokine such as IFNγ and IL-17.
Further we have standardized method to isolate CD4 regulatory T cells and suppressive macrophages.
Our preliminary studies show that P histicola mediate its immunosuppressive activity through increase in
regulatory CD4 T cells and tolerogenic CD103+CD11c+ dendritic cells (DCs).
Similarly, we studied therapeutic efficacy of P histicola to modulate arthritis in murine model of RA known
as collagen induced arthritis (CIA) (3). Our in vitro studies showed that treatment of mice with P. histicola
in collagen-immunized mice led to suppression of antigen-specific immune response and reduction in
production of inflammatory cytokines. In the second year we show that P. histicola can protect DQ8 and
DR4/DQ8 mice from arthritis in protective and therapeutic protocol. On the other hand, treatment with a
control bacteria Prevotella melanogenics did not result in protection from arthritis. We further show that
arthritis is regulated via generation of T regulatory cells and regulatory DCs. The treatment led to
increased production of regulatory cytokines like IL-10 with lower levels of IL-17. Further, rtPCR for
various cytokines and chemokines showed that treatment led to lower expression of pro-inflammatory IL23. To determine if gut immune system is different in naïve mice we used DRB1*0401 mice that show
sex-biased arthritis and also *0402 mice that are resistant to arthritis (6). Our data shows that gut
immune system is different between male and female *0401 mice and also between *0401 and *0402
mice suggesting a crucial role of gut in pathogenesis (7, 8). This also confirms our hypothesis that
arthritis can be modulated via gut and such a therapy may be possible in humans. Thus in summary, we
have made good progress towards completing our aims propsed in the grant and our data suggests that
P histicola induced immune responses in the gut might cause induction of immune tolerance in periphery
leading to suppression of antigen specific response.
5
Fig.2. P histicola treated mice show
decreased cumulative clinical score as
compared to sham treated mice in an
ongoing EAE model
Fig 1. Treatment with P. histicola of mice induced for collageninduced arthritis suppresses disease incidence while a control
bacteria P. melanogenica did not show suppressive activity. P.
histicola alone did not lead to arthritis.
Progress report (arranged according to approved SOW)
SOW 1-e) Immunization of mice with relevant antigen and feeding bacteria in ongoing disease in vivo
(11-12th months)
CIA 20 mice Controls 20 Dr. Taneja Completed
EAE 20 mice Controls 20 Dr. Mangalam Completed
SOW 1-f ) Score immunized mice for arthritis in CIA model. Bleed mice via tail technique (11-16th
months) Dr. Taneja.
Mice were immunized with type II
collagen (CII) to induce arthritis and
the fed them P histicola on alternate
days 2 weeks following
immunization. Next we tested mice
in preventive protocol. Mice were fed
P histicola 12 days prior to
immunization with type II collagen.
Mice immunized with type II collagen
and fed media without bacteria as
well as mice fed bacteria without CII
immunization were used as controls.
Disease phenotype in CIA model is
characterized by paw swelling,
scored on scale of 0-3. Our data
showed that both preventive and
therapeutic protocol suppressed
disease incidence and antibodies to
Collagen II. We used a control bacteria P. melanogenica; it did not show any suppressive activity (Fig.1).
Score immunized mice for paralysis in EAE model. Bleed mice via tail technique (11-16th months)
Dr. Mangalam
In these set of experiments, we wanted to investigate if P histicola can suppress or modulate an ongoing
disease in EAE or CIA model. As shown in Tab 1, P histicola treatment of DR3DQ8.AEo mice with EAE
lead to 50% decrease in disease incidence and severity as compared to medium treated control group,
which showed 100% disease incidence. P histicola treated group also showed less severe disease as
compared to control group as shown be cumulative clinical score in Fig. Medium treated control group
with EAE showed a cumulative score of 83 [9 mice with score of 5 (45), 5 mice with score of 4 (20), and
6 mice with a score of 3 (9)], while P histicola treated mice had a cumulative score of 28 [1 mice with
score of 5 (5), 2 mice with score of 4 (4), 3 mice with a score of 3 (6), 4 mice with a score of 2 (2) and 10
mice with no disease (0)]. Thus our data indicate that P histicola can modulate an ongoing disease.
Table 1: Effect of P histocola on PLP91-110 induced EAE in
DR3DQ8.AEo transgenic mice
Mouse strain Disease
incidence
(%)
Mean
onset of
disease ±
SD
Number of mice
with maximum
severity score
1 2 3 4 5
DR3.DQ8.AEº
(Medium fed)
20/20
(100%)
10.5±1 - - 4 10 6
DR3.DQ8.AEº (P 10/20 15±2 - 4 3 2 1
6
histicola) (50%)
Feeding bacteria in ongoing disease modulate disease severity and treated group showed
decrease in disease severity compared to control.
SOW 1-g) Sacrifice CIA group mice and harvest various organs and snap freeze a part of organs and
one part for paraffin blocks (16-17th months) Dr. Taneja
Completed
Sacrifice EAE mice and harvest various organs and snap freeze a part of organs and one part for
paraffin blocks (16-17th months) Dr. Mangalam
Completed
SOW 1-h) Harvest spleen and lymph nodes from the treated and controls mice in CIA (arthritis) model
and do in vitro assay for T cell response to antigen and measure cytokines (17 month) Dr. Taneja
CIA (arthritis) model (20th month). Dr. Taneja (completed last year)
Prevotella histicola modulates antigens-specific responses: Since collagen specific T-cell responses
play an important role in disease pathogenesis of CIA, we investigated effect of Prevotella on antigen
specific immune response and production of pro-inflammatory cytokines by antigen specific T-cells. Mice
were fed bacteria before or after immunization with CII. Mice immunized with but no bacteria and mice
fed bacteria without CII-immunization were used as controls. As shown below in Fig 3, antigen-specific T
cell response was suppressed in mice fed P. histicola before and after immunization with CII as
compared to mice immunized with CII only. As expected, mice fed bacteria in the absence of CIIimmunization did not show any antigen specific response. We further tested and compared production of
pro-inflammatory cytokines in mice immunized with CII and fed medium and mice immunized with CII
and fed P histicola (Fig 4). Mice treated with bacteria after CII-immunization showed a much lower
production of proinflammatory Th1 (IL-1, TNF and IFN) as well as Th17 (IL-12(p40), IL-17, IL-6)
cytokines compared to mice immunized with CII and fed media without bacteria. These in vitro studies
clearly show an immunomodulatory role of commensal bacteria like P histicola. Our studies suggest that
P. histicola may be able to generate systemic suppression via mucosal immune regulation.SOW 1-h)
Harvest spleen and lymph nodes from the treated and controls mice in EAE (multiple sclerosis) model
and do in vitro assay for T cell response to antigen and measure cytokines (17 month) Dr. Mangalam
0
1000
2000
3000
4000
5000
Bact+CII CII+Bact Bact CII
CII

cp m Fig 3. Lymph node T cell response to type II
collagen in DR4/DQ8 transgenic mice fed P
histicola before and after CII-immunization and
control CII-immunized without bacteria and only
bacteria fed mice.
0
200
400
600
800
1000
IL-1a IL-1b IL-12P40 IL-6 IL-17 TNF IFN
Med Control P histicola
Pg /m l Fig 4. Cytokine production analyzed by ELISA in
supernatants of mice immunized with CII and fed media
(Med Control) or P histicola. Transgenic mice fed P
histicola show a much lower production of
proinflammatory cytokines.
7
Fig.6 - P. histicola treated DR3DQ8 mice
exhibited reduced level of IL-17 and increased
levels of IL-10 as compared to sham treated
mice. Levels of IFN-γ were not different
Fig. 5. P. histicola treated DR3DQ8 mice exhibited reduced
PLP91-110 specific T cell proliferation as compared to sham
treated mice. Splenocytes were collected from mice immunized
with PLP and treated with P. histicola or medium (sham) and
were stimulated in vitro with the PLP91-110 polypeptide.
Fig.7. H&E stain of gut of mice treated either with P histicola or
left untreated in CIA model.
EAE (MS) model (20th month). Dr. Mangalam (completed last year)
Modulation of antigen specific T cell and cytokine response by P histicola
To determine if this protective effect of P. histicola is due to own-regulation of antigen specific T cell
responses, we isolated splenocytes from mice treated with bacteria or medium and stimulated with
PLP91-110 peptide. As shown in Fig. 5, antigen specific T cell response was suppressed in DR3DQ8
mice treated P. histicola as compared to sham treated mice. Splenocytes from bacteria fed mice also
produce less pro-inflammatory cytokine IL-17 on stimulation with PLP, while levels of anti-inflammatory
cytokine IL-10 was increased (Fig. 6). Levels of IFN-γ were not significantly different between two groups
(Fig. 6).
Milestone#7 immunomodulatory effect of bacteria in disease.
1-i) Sections of paraffin blocks and frozen blocks from CIA (arthritis) group and staining with
heamatoxylin and eosin. ELISA for antibodies(18-20 months) Dr. Taneja
To ensure that treatment of mice
did not result in any pathology of
the gut, we did histopathology of
mice immunized with CII and
treated with either medium or with P
histicola. We observed that P.
histicola treatment did not cause
any pathology in the gut (Fig.7).
8
Fig. 8. P histicola treated group showed decreased
inflammation and demylination in brain as well as spinal cord
compared to control untreated group.
Fig 10. DQ8 mice were immunized with type II collagen (CII) and treated with P. histicola
showed suppression of T cell response when treatment was before immunization. Autoantibodies
to CII were significantly reduced in P. histicola treated mice. Sera collected at day 20 after
immunization with CII, and day 40 after treatment started at day 21 were tested by ELISA.
Sections of paraffin blocks and frozen blocks from EAE (multiple sclerosis) group and staining with
heamatoxylin and eosin. ELISA for antibodies(18-20 months) Dr. Mangalam
Immunization of DR3DQ8 transgenic mice with
PLP antigen lead to development of EAE. The
disease is characterized by inflammation and
demyelination in spinal cord as well as brain tissue,
resulting in neurological deficit. Although, treatment
with P histicola suppressed clinical disease, it was
still possible that treated mice have brain and
spinal cord pathology. Therefore we analyzed
these tissues and observed that P histocola treated
group showed decreased inflammation and
demyelination compared to sham treated group
(Fig 8 ). Bacteria treated group also showed
decreased levels of inflammatory cells ( Fig 9A),
as well as inflammatory cytokines (IL-17 and IFNγ)
( Fig 9 B and C) compared to medium treated
group. Treated group also showed decreased
levels of anti-myelin antibodies (data not shown).
Milestone# 8. Histopathology of various organs and antibodies for diagnosis of disease.
1-j) compilation of in
vivo data and T cell
response and
cytokine data in
CIA (arthritis)
model (20th month).
Dr. Taneja
Fig. A) P histicola treatment reduces brain infiltrating cells in PLP91-110 immunized mice compared to medium treated
mice. P histocola treated mice also show reduced levels of inflammatory cytokine IFNγ and IL-17 confirming its
immunomodulatory role in EAE
9
Fig 11. Cytokines in medium treated and P. histicola treated mice showed suppression of
proinflammatory cytokines.
In vivo
data is
compiled in Fig 11. T cell response was significantly decreased when mice were first treated and then
immunized with CII compared to untreated mice. However, treatment after CII immunization did not
significantly reduce T cell response suggesting regulation of disease could be via other regulatory
mechanism.
Compilation of in vivo data and T cell response and cytokine data in EAE (multiple sclerosis)
model (20th month). Dr. Mangalam
Analysis of T cell response and cytokine in in vivo study, showed that treatment with P histicola led to
suppression of antigen response only in lamina propia, while T cell response in mesenteric lymph node
and spleen were not affected (12A). However, P histicola treated and protected mice showed decreased
levels of pro-inflammatory cytokines (12B).
Fig 12.A) P histicola treatment suppressed antigen specific response only in Lamina propia (LP)
cells but not in Mesenteric Lymph node (MLN) cells or spleen. B) Cytokines in medium treated and
P. histicola treated mice showed suppression of proinflammatory cytokines.
10
Mouse strain Disease incidence (%) Mean onset of
disease ± SD
Number of mice with maximum
severity score
1 2 3 4 5
DR3.DQ8.AEº (Medium fed) 10/20 (100%) 10.5±1 - - 4 2 4
DR3.DQ8.AEº
(P histocola live)
2/10 (20%)
15±2 - 1 - - 1
DR3.DQ8.AEº
(P histocola heat killed)
8/10 (80%)
14±2 - 1 1 2 4
Milestone#9 Analysis of data and publication (2 publications expected one for each disease).
Four abstracts, two in April’2011, one in Nov’2011 and Mar 2012, one manuscript 2012.
Aim#2 Mechanism of anti-inflammatory action of commensal bacteria P histicola.
2a) Heat inactivate bacteria and feed in therapeutic protocol and monitore for disease (18-24 months)
Mice sacrificed CIA 20 Controls 20 Dr. Taneja
Experiments with EAE model did not show any difference in heat killed and live bacteria. Considering
this, this experiment was not carried out for CIA studies as while EAE model can be done in one month,
CIA takes 4-6 month to complete. It was decided not to waste valuable mice and manpower.
EAE 20 Controls 20 Dr. Mangalam Ongoing (done once being repeated
Table 2: Requirement of Live vs. heat killed P histocola on PLP91-110 induced EAE in DR3DQ8.AEo
transgenic mice
Milestone# 10 Our preliminary data indicate that heat killed bacteria do not modulate disease suggesting
that live bacteria is required for disease suppression.
2b) Feed mice bacteria for a week and then immunize with Type II collagen for DR4/DQ8 and PLP for
DR3/DQ8 mice. Harvest LNCs and spleen cells duodenum, ileum, jejunum and colon after 10 days of
immunization. (21-22 months)
DR4/DQ8 mice -2 Dr. Taneja
DR3/DQ8 mice -2 Dr. Mangalam
We have stardaized this method and observed that we need 5 mice to get enough cells to perform
mechanistic experiments.
2c) Repeat experiment in 2a and
isolate cells from mice and pool cells from mice necessary for T cell response. Characterize cells, T reg,
memory T, DCs, Mac, B cells and TLR expression in various parts of intestines, T cell responses, freeze
supernatants from culture (22-24 months)
DR4/DQ8 mice- 10 mice Controls 10 Dr. Taneja
2b, 2f, 2g) Characterization of T regulatory cels from intestine and periphery and cytokines.
Prevotella histicola modulates antigen-specific responses via generation of T regulatory and
regulatory DCs. These studies suggested P. histicola may be able to generate systemic suppression
via mucosal immune regulation. We tested various cells in CII-immunized and mice immunized and
treated with bacteria in spleen and lamina propria. As shown in Fig 13, both spleen and lamina propria
showed an increase in T regulatory, CD4+CD25+FoXP3 and CD4+GITR+, cells as well as suppressive
DCs that express CD103. These studies suggest that treatment with P histicola leads to generation of
regulatory DCs that activates more T regulatory cells, these T cells and DCs can produce regulatory and
anti-inflammatory cytokines. Both T reg and reg DCs can migrate into the systemic immune system.
11
Prevotella histicola regulates systemic immune response via DCs and production of IL-10.
Mice treated with P. histicola and immunized with CII had DCs that when cultured with CD4 cells from
treated or Sham mice suppressed immune response when challenged in vitro with CII. Supernatants
from these cultures showed that treated DCs and CD4 led to production of IL-10 while IL-17,
proinflammatory cytokine, was suppressed in comparison to sham treated mice that produced more IL17 and lower IL-10. These studies suggest that P. histicola treatment has led to generation of regulatory
DCs phenotypically and functionally (Fig 13 and 14).
DR3/DQ8 mice-10 mice Controls 10 Dr. Mangalam
Spleen
Lamina
Propria
50 65
6.8 9.8
Sham P histicola
CD4+CD25+FoXP3 CD11c/CD103
CD4/GITR
CII+P.Hist CII
22
11
Fig 13. Treatment of DQ8 mice with P.histicola resulted in generation of regulatory DCs
and T regulatory cells as observed in splenic and lamina propria isolated cells. Various cells
were enumerated by FACS analysis.
14
21
0
50
100
150
200
S_CD4XDC S_CD4X P.HistDC P Hist_CD4X DC P Hist_CD4XS_DC
IL-17
IL-10
P
g/ m l
C
PM
0
200
400
600
800
1000
1200
Sham DC P Hist DC
Sham CD4
P Hist CD4
Fig 14. P. histicola suppresses systemic immune response via dendritic cells. Mice treated with P.
histicola or medium (Sham) were immunized with CII. Dendritic cells (DCs) and CD4 T cells
were isolated from spleen and in vitro challenged with CII in a criss-cross culture. As shown in
right panel, DCs from treated mice led to suppression of CD4+ T cell response in vitro. Also,
when DCs and CD4+ cells of treated mice were cultured, there was increased production of IL-10
(right panel).
12
Fig.16. A) P histcola treated mice show increase in CD11c+CD103+ tolerogenic
dendritic cells compared to o medium treated group. B) histogramplot
CD11c+CD103+ tolerogenic dendritic cells in treated vs untreated group. Dendritic
cells from P histicola treated mice increased levels of IL-10 (C) and decreased levels
P histicola can modulate disease by inducing anti-inflammatory immune response mediated by
regulatory cells such as
CD4+FoxP3+ Tregs, suppressive
macrophage and or tolerogenci
DCs. To determine if these cell
subsets are involved in the
suppressive effect of Phisticola, We
isolated Tregs, macrophage and
DCs from P histicola treated and
PBS treated mice. Our data
indicate that P histicola modulate
disease through increase in
Regulatory T cells number ( Fig. 15
A and B) as well as their function.
CD4+CD25+ Tregs isolated from P
histicola treated mice showed
increased ability to suppress
antigen specific immune response
as compared to medium treated
control group ( Fig 15 C).
Treatment with P histicola also
resulted in increase in number of
tolerogenic DCs (Fig. 16 A), which
have reduced antigen presentaion
capacity as compared to DCs
isolated from naive mice or
medium treated control group
(Fig 16B). Tolerogenic Dcs are
characterized by high IL-10 to
IL23 ratio and we also
observed a similar phenotype
as DCs isolated from bacteria
treated group produce more IL10 than IL23 ( Fig16 C and D).
Thus P histicola suppress
disease throgh modulation of
Tregs, and tolerogenic DCs,
which lead to decreased
antigen presentation and
inflammation.
Milestone#11
Immunomodulatory effects of
bacteria in immune response.
2d) Analysis of the data
generated by heat killed
bacteria and cytoplasmic
component and in vitro data
DR4/DQ8 mice Dr. Taneja
Ongoing
DR3/DQ8 mice Dr. Mangalam Ongoing
Fig. 15 A) P histcola treated mice show increase in CD4+FoxP3+ regulatory T
cells compared to o medium treated group. B) histogramplot for Tregs in treated
vs untreated group. C) CD4+CD25+ Tregs from P histocola treated mice have
increased suppressive activity as compared to medium treated group.
13
2g, h and i) Cytokines, primer design and RT-PCR in gut
Expression of cytokines/chemokines in CIA (arthritis) model Dr. Taneja
Next we determined the effect of treatment in jejenum of DQ8 mice and compared to control by doing
rtPCR for various cytokines. In addition, we also used naïve *0401 and *0402 mice to determine if gut
immune system can impact susceptibility in association with genotype.
rtPCR for cytokines in jejenum: We did rtPCR for various cytokines in jejenum of mice treated and
immunized with CII and those only immunized with CII using published primers. As shown in Fig 17,
mice treated with P.histicola showed a much higher expression of anti-inflammatory cytokines, IL-4 and
IL-10 and lower expression pro-inflammatory cytokine IL-23 further confirming that treatment modulates
arthritis via gut As shown above, P.histicola alone did not cause arthritis or any pathology suggesting,
this might be a good treatment for clinical trials in RA patients.
*0401 female mice show different jejenum profile than males.
DQ8 CII
DQ8 CII +P.Histicola
Fig 17 rtPCR for cytokines in jejenum of DQ8 mice immunized with CII and treated with P.
histicola compared to CII immunized mice showed differences in pro and anti-inflammatory
cytokines.
Fo ld c ha ng e re la tiv
e ex pr es si on -25
-20
-15
-10
-5
0
5
10
15
20
IL-
4
IL-
5
IL-
6
Sta
t4 Fo xP 3
IFN
g IL1
2 r
b1 IL-
17
a IL
23
a IL3 IL21
IL-
22
CC
L2
0
Fig 18 Comparison of fold change in gene transcript levels between *0401 females and males.
14
Fig.18. Treatment with P histicola cause increase in levels of
anti-inflammatory cytokine and transcription factors associated
with suppressive immune response.
We tested the jejuna of naïve mice for expression of cytokine and chemokine transcripts involved in the
Th17 regulatory network by rtPCR (Figure 18). Susceptible *0401 females showed a distinct cytokine
and chemokine profile as compared to males that was characterized with a significant increase in IL-23α
and IFNγ along with a decrease in the regulatory cytokines IL-4, IL-22 and CCL20. Similarly, *0401
females showed more than 3 fold increased gene transcripts for Th17 cytokines IL-17, IL-23, IL-6 and
Th1 cytokines IFNγ, Stat 4 and TBX21 while *0402 females had several fold increase in genes regulating
Th2 cytokines and regulatory networks like ICOS, GATA3 and IL-4. *0401 male mice did not show an
increase in transcripts for TH17 encoding genes compared to *0402 mice.
Our data showed a bias towards TH1/TH17 cytokine expression with significant decrease in cytokine
gene transcripts required for negative regulation of Th17 profile, like IL-4, IL-21 and IL-22, in *0401
females as compared to *0401 males and *0402 females. Interestingly CCL20 and CCL22 which are
required for the generation of regulatory CD4 T cells and DCs, are reduced several fold in *0401 females
as compared to *0401 males and *0402 females. These data suggest that events in gut may be involved
in pathogenesis.
Expression of
cytokines/chemokines in EAE
(multiple sclerosis) model Dr.
Mangalam
To determine the effect of P histicola
on mucosal immune system,
duodenum, jejunum and ileum were
isolated from P histicola treated and
medium treated mice immunized for
EAE.
RNA was extracted from tissue,
reverse transcribed into cDNA and
expression of cytokines and
transcription associated with antiinflammatory response were analyzed
by Real time PCR using specific
primers. As shown in figure 18, P
histicola treated mice showed
increased levels of anti-inflammatory
cytokine such as TGFβ, IL-10, IL-25
and TSLP compared to control group
with EAE. The levels differed between
tissue but maximum expression was
observed in ileum. Bacteria treated
group also showed increase in levels
of FoxP3, a transcription factors
marker for the regulatory CD4 T cells.
These data suggests that treatment with P histicola cause increase in levels of Th2 and antiinflammatory cytokines. These cytokine might be responsible for modulation of inflammatory immune
response in periphery leading to down-regulation of pro-inflammatory response and disease.
Milestone#13-15 cytokines, design of primers for mRNA expression for cytokine/chemokine expression
in intestines.
15
The Key Research Accomplishments
1st Year
• Culture of P Histicola for use in both CIA and EAE model
• Generation of DR4/DQ8 transgenic mice for in vivo use. HLA-DR4 and HLA-DQ8 transgenic mice
are mated to generate double transgenic mice. Double transgenic mice are characterized for the
presence of HLA transgenes by flow cytometry using specific conjugated antibodies. Mice
positive for both genes are identified and mated. DR4 and DQ8 transgenes can segregate which
necessitates typing for the transgene positivity.
• Generation of DR3/DQ8 transgenic mice for in vivo use. HLA-DR3 and HLA-DQ8 transgenic mice
were mated to generate double transgenic mice. Double transgenic mice are characterized for
the presence of HLA- DR3 and –DQ8 transgenes by flow cytometry using specific conjugated
antibodies. Mice positive for both genes are identified and mated. DR3 and DQ8 transgenes can
segregate which necessitates typing for the transgene positivity.
• Mice were gavaged with P histicola for 2 weeks and then immunized with either type II collagen
(DR4DQ8 mice) PLP91-110 peptide (DR3DQ8 mice). In addition, control mice were gavaged with
media in which P histicola were cultured and immunized with type II collagen or PLP91-110 peptide.
Mice are being monitored for disease (CIA in DR4DQ8 and EAE in DR3DQ8).
• Sera from all test and control mice is being collected and will be used to study antibodies at the
termination of the experiment.
• In vitro experiments show that feeding bacteria suppressed antigen-specific T cell response and
reduced production of inflammatory cytokines in both CIA and EAE models.
2nd Year
• Mice were gavaged with P histicola for 2 weeks and then immunized with either type II collagen
(DR4DQ8 mice) PLP91-110 peptide (DR3DQ8 mice). In addition, control mice were gavaged with
media in which P histicola were cultured and immunized with type II collagen or PLP91-110 peptide.
Treatment with P histocola suppressed disease incidence and severity in both CIA as well
as EAE model.
• In vitro experiments show that feeding bacteria suppressed antigen-specific T cell response and
reduced production of inflammatory cytokines in both CIA and EAE models.
Milestone#8. P histocola treated group showed decreased inflammation and demyelination compared
to sham treated group. Treated mice also showed decrease levels of inflammatory cytokine IL-17 and
IFNγ. Histopathology of in vivo studies and antibodies in mice treated with P. histicola have been
completed. • Treatment with P histocola suppressed RF and anticitrullin antibody in CIA model.
Treatment with P histocola suppressed anti-myelin antibody in EAE model.
Milestone #9 Data has been presentation at National meetings and manuscript is being analyzed for
publication.
Milestone# 10. Our preliminary data indicate that heat killed bacteria do not modulate disease
suggesting that live bacteria is required for disease suppression. We have performed this experiment
once and repeating it one more time to confirm the finding.
We have also standardized isolation of cells from various organs
16
Milestone#11 Immunomodulatory effects of bacteria in immune response. We have isolated Tregs,
macrophage and DCs from P histicola treated and PBS treated mice. Our data indicate that P histicola
modulate disease through increase in Regulatory cells number as well as suppressive function of Tregs.
Treatment with P histicola also leads to generation of tolerogenic DCs and macrophages which have
reduced antigen presentation capacity. Thus P histicola suppress disease through modulation of Tregs,
suppressive macrophages and tolerogenic DCs, which lead to decreased antigen presentation and
inflammation.
Milestone #11 Immunomodulatory effect of bacterial treatment was studied by comparison of mRNA
expression levels of various cytokines by rtPCR of cytokines in jejenum of treated and untreated mice
We have designed the primers for mRNA expression of various cytokine/chemokine expressions in
intestines.
Milestone #12 Publication of in vivo data is in progress.
Milestone #13 DR4.TLR4-/- and DR3DQ8.TLR4-/- mice are being characterized.
Milestone #14 and 15 Modulation of cytokines by bacteria in spleen has been studied by using bioplex
array system. We have designed of primers for mRNA expression for cytokine/chemokine expression in
intestines and analyzed expression levels using RT-PCR.
17
Reportable Outcome
We have presented our work based on above findings at various international meetings
Abstracts Published at International Meetings:
• David Luckey, Melissa Karau, Robin Patel, Moses Rodriguez, Joseph Murray, Chella David,
Veena Taneja and Ashutosh Mangalam (2011). Human commensal bacteria as a novel
therapeutic agent for Multiple Sclerosis. Microbial and Mucosal Immunology: the interface in
health and disease, San Francisco, CA, USA.
• David Luckey, Marshall Behrens, Melissa Karau, Robin Patel, Ashutosh Mangalam, and Veena
Taneja (2011). Microbial Mucosal Modulation of Arthritis. Microbial and Mucosal Immunology: the
interface in health and disease, San Francisco, CA, USA.
• Premila Samuel, Arika Wussow, Ashutosh K. Mangalam (2011). The Mechanism of action of
Prevotella histocola in the immunomodulation of Experimental autoimmune encephalomyelitis
(EAE). 96th Annual Meeting of KAS (Kentucky Academy of Science), Murray State University,
Murray, KY, USA.
• Taneja V. Gut and Autoimmunity. “ Invited talk” 5th Federation of Immunological Societies
Association. New Delhi, India, March 2012.
Publication in scientific Journals
1. Luckey D, BastaKoty D, and Mangalam A (2011). Role of HLA class II genes in susceptibility and
resistance to multiple sclerosis: studies using HLA transgenic mice. J Autoimmunity 37: 122-128
PMID: 21632210
2. Gomez A, Yoeman C, Luckey D, Marietta EV, Miller ME, Murray JA, White BA and Taneja V.
2012. HLA-DR polymorphism, gut microbiome and sex may predict susceptibility or resistance to
arthritis in humanized mice. PloS ONE 7(4):e36095. doi:10.1371 / journal .pone.0036095.
18
Conclusions. Our data show that P histicola can suppressed ongoing EAE. P histicola treated group
showed decreased organ specific pathology in both CIA as well as EAE model. Treatment with P
histicola cause decrease infiltration of inflammatory cells in to tissue leading to decreased pathology. We
further show that suppressive effect of P histicola is through modulation of regulatory CD4 T cells and
tolerogenic DCs as both of these population is increased in treated group. Both Tregs as well as
tolerogenic dendritic cells had been shown to suppress inflammatory response. Our experiment on
requirement of live vs. dead bacteria for its therapeutic effect is ongoing but our preliminary data
suggests that live bacteria is required for therapeutic effect of bacteria as heat killed bacteria did not
significantly suppressed disease in one experiment. We are repeating these experiments to confirm
these findings. Our ongoing studies on the role of TLR4 will delineate if innate immunity is involved in
antigen-specific tolerance in humanized model of arthritis.The results from our ongoing in vivo
experiments will help us in understanding if Prevotella histicola can be used as a treatment of EAE in
HLA-DR3DQ8 transgenic mice and CIA in DR4DQ8 mice.
19
References
1. Baranzini, S. E., J. Wang, R. A. Gibson, N. Galwey, Y. Naegelin, F. Barkhof, E. W. Radue, R. L.
Lindberg, B. M. Uitdehaag, M. R. Johnson, A. Angelakopoulou, L. Hall, J. C. Richardson, R. K.
Prinjha, A. Gass, J. J. Geurts, J. Kragt, M. Sombekke, H. Vrenken, P. Qualley, R. R. Lincoln, R.
Gomez, S. J. Caillier, M. F. George, H. Mousavi, R. Guerrero, D. T. Okuda, B. A. Cree, A. J.
Green, E. Waubant, D. S. Goodin, D. Pelletier, P. M. Matthews, S. L. Hauser, L. Kappos, C. H.
Polman, and J. R. Oksenberg. 2009. Genome-wide association analysis of susceptibility and
clinical phenotype in multiple sclerosis. Human molecular genetics 18:767-778.
2. Mangalam, A., D. Luckey, E. Basal, M. Jackson, M. Smart, M. Rodriguez, and C. David. 2009.
HLA-DQ8 (DQB1*0302)-restricted Th17 cells exacerbate experimental autoimmune
encephalomyelitis in HLA-DR3-transgenic mice. J Immunol 182:5131-5139.
3. Taneja, V., and C. S. David. Role of HLA class II genes in susceptibility/resistance to
inflammatory arthritis: studies with humanized mice. Immunological reviews 233:62-78.
4. David Luckey, Melissa Karau Robin Patel, Moses Rodriguez, Joseph Murray, Chella David,
Veena Taneja and Ashutosh Mangalam. “Human commensal bacteria as a novel therapeutic
agent for Multiple Sclerosis” for meeting titled Microbiota and mucosal immunology: the interface
in health and disease, April 14-16, 2011, San Francisco, CA, USA
5. David Luckey, Marshall Behrens, Melissa Karau, Robin Patel, Ashutosh Mangalam, and Veena
Taneja (2011). Microbial Mucosal Modulation of Arthritis. Microbial and Mucosal Immunology: the
interface in health and disease, San Francisco, CA, USA.
20
Appendices (3 abstracts and one published manuscript )
1. Human commensal bacteria as a novel therapeutic agent for Multiple Sclerosis
Microbiota and mucosal immunology: the interface in health and disease, April 14-16, 2011, San
Francisco, CA, USA
David Luckey1, Melissa Karau2, Robin Patel2, Moses Rodriguez1,3, Joseph Murray4, Chella David1, Veena
Taneja1 and Ashutosh Mangalam1.
Department of Immunology, Clinical Microbiology, Neurology, and Gastroenterology, Mayo Clinic,
Rochester, MN -55905 USA.
Multiple sclerosis (MS), a chronic inflammatory disease of the CNS, is strongly associated with the MHC
class-II genes HLA-DR2, DR3, DR4, DQ8. Here we are proposing a novel approach to cure MS, by
administration of a s pecific strain of human commensal-bacteria. R ecent studies have shown that
intestinal microflora plays an important role in the health of the host and posses probiotics like qualities.
We hypothesize that Gram-negative commensal-bacteria Prevotella histicola from human gut have the
potential to be used as a therapeutic agent. We have used HLA-DR3DQ8 transgenic mice to test our
hypothesis that treatment with commensal-bacteria P histicola can modulate experimental autoimmune
encephalomyelitis (EAE), an animal model of MS. Previously we showed that PLP91-110 can induce EAE
in HLA-DR3DQ8 transgenic mice. First using various doses of bacteria, we have identified the optimal
dose to be used for treatment of EAE. Our study showed that treatment of mice with 3-4 doses of P.
histicola in PLP91-110-immunized mice led to suppression of antigen-specific immune response in-vitro.
Treatment of mice with P histicola as probiotics is ongoing. Our data indicates that P histicola induced
immune responses in the gut cause induction of immune tolerance in periphery leading to suppression of
antigen-specific response.
2. Microbial mucosal modulation of arthritis
David Luckey, Marshall Behrens, Melissa Karou, Robin Patel, Joseph Murray, Ashutosh Mangalam and
Veena Taneja
Microbiota and mucosal immunology: the interface in health and disease, April 14-16, 2011, San
Francisco, CA, USA
Department of Immunology, Gastroenterology and Microbiology, Mayo Clinic Rochester, MN -55901
USA.
Predisposition to rheumatoid arthritis (RA) is associated with the presence of genetic factors, HLA class
II molecules, DR4 and DQ8, being the strongest. Recent reports that patients with RA have decreased
fecal levels of certain commensal bacteria suggested that intestinal microbes might be critical in
regulation of disease. We isolated Prevotella histicola, anaerobic commensal bacteria of Human gut,
from bowel of a patient and have shown that it possesses anti-inflammatory activity. We propose that gut
microbiota can influence peripheral immune response and may modulate arthritis in a murine model. We
have established a murine model of rheumatoid arthritis using mice expressing RA-associated HLA
genes, DRB1*0401 and DQ8. DR4 and DQ8 mice develop collagen-induced arthritis (CIA) following
immunization with type II collagen (CII). We have used HLA-DR4/ DQ8 mice to test our hypothesis that
treatment with commensal bacteria like Prevotella histicola can modulate CIA. In vitro data showed that
treatment of mice with P. histicola in CII-immunized mice led to suppression of antigen-specific immune
response and reduction in production of inflammatory cytokines suggesting P histicola has antiinflammatory properties in this model. Treatment of CIA in transgenic mice in therapeutic protocol is
21
ongoing. Our data suggests that P histicola induced immune responses in the gut causes systemic
immune suppression and may be able to regulate autoimmunity.
3. The Mechanism of action of Prevotella histocola in the immunomodulation of Experimental
autoimmune encephalomyelitis (EAE).
96th Annual Meeting of KAS (Kentucky Academy of Science), Murray State University, Murray, KY, USA
PREMILA SAMUEL*¹, Arika Wussow, ASHUTOSH K. MANGALAM². 1Department of Chemistry, Berea
College, Berea, KY 40404. ²Immunology Department, Mayo Clinic, Rochester, MN 55906. ).
Multiple sclerosis (MS) is an inflammatory autoimmune disease affecting the central nervous system.
The etiology of MS is extremely complex as both genetic as well as environmental factors may interact in
different ways to influence the outcome of disease. In recent year there had been lots of interest in
studying the role of gut microbiota as an environmental factors in the immunopathogenesis of
inflammatory diseases such as MS. This hypothesis is supported by recent reports that patients with
inflammatory diseases have reduced fecal levels of certain commensal bacteria, suggesting possible
immunomodulatory role of commensal bacteria in these diseases. Previously, we have shown that
human gut specific commensal bacteria, Prevotella histocola can suppress Experimental autoimmune
encephalomyelitis (EAE), an animal model of MS, in HLA-DR3DQ8 transgenic mice. In this study , we
investigated the mechanism of action of immunomodulatory activity of P histicola by investigating the
ability of P histicola to modulate pro and anti-inflammatory cytokine production from LPS stimulated
epithelial and dendritic cells. Our preliminary data show that the in vitro treatment of Caco-2, a human
intestinal epithelial cell line, and THP-1, a human monocytic cell line with P.histocola supernatant
resulted in the down regulation of pro-inflammatory cytokines TNFα and IL-8 and the up regulation of
anti-inflammatory cytokine IL-10. Likewise, the in vitro treatment of dendritic cells and macrophages
(derived from HLA-DR3.DQ8 double transgenic mice) with P. histocola supernatant suppressed the
production of TNFα, while inducing the production of IL-10. Thus our preliminary data indicate that P.
histocola might suppress EAE in HLA class II transgenic mice through modulation of pro and antiinflammatory cytokines. More detailed studies are underway to understand the detailed mechanism of
immunomodulatory action of the commensal bacteria.
Publication in scientific Journals
4. Luckey D, BastaKoty D, and Mangalam A (2011). Role of HLA class II genes in susceptibility and
resistance to multiple sclerosis: studies using HLA transgenic mice. J Autoimmunity 37: 122-128
PMID: 21632210
5. Gomez A, Yoeman C, Luckey D, Marietta EV, Miller ME, Murray JA, White BA and Taneja V.
2012. HLA-DR polymorphism, gut microbiome and sex may predict susceptibility or resistance to
arthritis in humanized mice. PloS ONE 7(4):e36095. doi:10.1371 / journal .pone.0036095.
lable at ScienceDirect
Journal of Autoimmunity 37 (2011) 122e128Contents lists avaiJournal of Autoimmunity
journal homepage: www.elsevier .com/locate/ jaut immRole of HLA class II genes in susceptibility and resistance to multiple sclerosis:
Studies using HLA transgenic mice
David Luckey, Dikshya Bastakoty 1, Ashutosh K. Mangalam*
Immunology Department, Mayo Clinic, 200 Ist ST SW, Rochester, MN 55905, United Statesa r t i c l e i n f o
Article history:
Received 27 April 2011
Accepted 2 May 2011
Keywords:
EAE/MS
HLA transgenic mice
Cytokine
Anti-myelin antibody
Complement* Corresponding author. Tel.: þ1 507 284 4562; fax
E-mail address: mangalam.ashutosh@mayo.edu (A
1 Current address: Interdisciplinary Graduate Prog
Nashville, TN 37240, United States.
0896-8411/$ e see front matter  2011 Elsevier Ltd.
doi:10.1016/j.jaut.2011.05.001a b s t r a c t
Multiple sclerosis (MS), an inflammatory and demyelinating autoimmune disease of CNS has both,
a genetic and an environmental predisposition. Among all the genetic factors associated with MS
susceptibility, HLA class II haplotypes such as DR2/DQ6, DR3/DQ2, and DR4/DQ8 show the strongest
association. Although a direct role of HLA-DR alleles in MS have been confirmed, it has been difficult to
understand the contribution of HLA-DQ alleles in disease pathogenesis, due to strong linkage disequilibrium. Population studies have indicated that DQ alleles may play amodulatory role in the progression of
MS. To better understand themechanism bywhich HLA-DR and -DQ genes contribute to susceptibility and
resistance to MS, we utilized single and double transgenic mice expressing HLA class II gene(s) lacking
endogenous mouse class II genes. HLA class II transgenic mice have helped us in identifying immunodominant epitopes of PLP in context of various HLA-DR and -DQ molecules. We have shown that HLA-DR3
transgenic mice were susceptible to PLP91e110 induced experimental autoimmune encephalomyelitis
(EAE), while DQ6 (DQB1*0601) and DQ8 (DQB1*0302) transgenic mice were resistant. Surprisingly
DQ6/DR3 double transgenic mice were resistant while DQ8/DR3 mice showed higher disease incidence
and severity than DR3mice. The protective effect of DQ6 inDQ6/DR3micewasmediated by IFNg, while the
disease exacerbating effect of DQ8moleculewasmediated by IL-17. Further, we have observed thatmyelinspecific antibodies play an important role in PLP91e110 induced EAE in HLA-DR3DQ8 transgenicmice. Based
on these observations, we hypothesize that epistatic interaction between HLA-DR and -DQ genes play an
important role in predisposition toMS and our HLA transgenicmousemodel provides a novel tool to study
the effect of linkage disequilibrium in MS.
 2011 Elsevier Ltd. All rights reserved.1. Introduction
Multiple sclerosis (MS) is presumed to be an autoimmune disease
of the central nervous system (CNS) leading to demyelination, axonal
damage, and progressive neurologic disability. Collective evidence
suggests that the onset of the disease might result from an aberrant
immune response to a number of myelin antigens that is T-cell
mediated. The first process of autoimmunity is the peripheral
activation of autoreactive CD4þ T cells via the presentation of autoantigens by susceptible MHC class II molecule(s). Therefore it is not
surprising that autoimmune diseases such as MS show a strong
association with certain HLA class II genes [1e8].
The HLA class II region of the MHC on chromosome 6p21
accounts for the majority of familial clustering in MS and is by far: þ1 507 266 0981.
.K. Mangalam).
ram, Vanderbildt Universirty,
All rights reserved.the major susceptibility locus. The class II linkage in MS differs
in various populations with the highest association with HLA-DR2
(DRB1*1501)/DQ6 (DQB1*0602) [9e12], Elegant studies by
Dyment et al. [4] have shown that the DRB1*17 (DR3) allele is
also associated with MS susceptibility. A similar finding on the
association of DR3 with MS has been shown in Southern European,
Canadian, Mexican and Sardinian MS patients [1,13e15]. Beside
DR2/DQ6, DR3/DQ2 and DR4/DQ8 genes are also linked with
predisposition to MS [1,12,14,16e18]. Recent studies have shown
that disease outcome might be decided by a complex interaction
among different class II genes present in a ‘haplotype’, suggesting
that the ‘haplotype’ might be the basic immunogenetic unit of
susceptibility or resistance [3,4,7,8,19].
Although no animal model can mimic all the facets of human
MS, the experimental autoimmune encephalomyelitis (EAE) model
in rodents has helped immensely in improving our understanding
of the immunopathogenesis of MS [20e22]. EAE can be induced in
various inbred animal strains by inoculation of whole myelin or
defined myelin proteins such as myelin basic protein (MBP), myelin
D. Luckey et al. / Journal of Autoimmunity 37 (2011) 122e128 123oligodendrocytes glycoprotein (MOG), and proteolipid protein
(PLP) in complete Freund’s adjuvant [20e22]. Elegant studies in
murine/rodent EAE have documented that encephalitogenic T cells
are CD4þ, T helper (Th1)-type cells secreting TNF-a/b and IFNg
[23e25]. However recent studies have indicated that a new T cell
phenotype Th17 secreting IL-17, IL-17F, IL-21, IL-22 and IL-23 might
also play an important role in the immunopathogenesis of EAE [26].
Thus current hypothesis of EAE indicates that both Th1 and Th17
cytokines play important roles in the immunopathogenesis of EAE.
2. HLA class II transgenic mice expressing HLA-DR or -DQ
molecule as an animal model of MS
Despite the fact that MHC genes show the strongest association
with MS, the exact role of HLA-DQ and -DR genes in disease
pathogenesis is not well understood due to the high polymorphism
and heterogeneity of human populations. The strong linkage
disequilibrium among HLA-DR, -DQ and other genes within the
HLA region makes it difficult to identify the role of individual genes
in the immunopathogenesis of MS. In order to understand the role
of class II molecules in MS, transgenic mice were generated that
express human HLA-DR or -DQ genes lacking endogenous mouse
class II genes [27]. A EAEmousemodel where the autoreactive Tcell
repertoire is selected and shaped by human MHC class II molecules
has helped us in understanding the immunopathogenesis of
inflammatory and demyelinating diseases such as human MS.
Using these class II transgenic mice, first we tested whether these
human HLA class II molecules are functional by analyzing the T cell
proliferative response against various CNS antigens.
2.1. Identification of T cell epitopes
We carried out experiments to determine whether HLA class II
molecules in the transgenic mice can efficiently present myelin
antigen proteolipid protein (PLP). PLP is the most abundant myelin
antigen inCNS and Tcells reactive to PLP peptides had been identified
in bothMS patients and normal controls [28e30]. Using overlapping
PLP peptides encompassing the entire sequence of the human PLP
molecule (human andmouse PLP are 100% conserved), we identified
a number of epitopes restricted to various HLA-DR or -DQ molecules
(Table 1). T cell epitopes were spread throughout the entire sequence
of PLP molecule [31] and major immunodominant regions were
31e70, 81e120, 140e160, 178e227 and 264e277. Both HLA-DR2
(*1502) and HLA-DR4 (*0401) molecules recognized similar residues on the PLP protein encompassing residues 31e60, 81e120,
178e197, 208e227 and 264e277. The exceptions were PLP51e70,
recognized only by -DR4 molecule, and residue 198e207 recognized
only by -DR2 molecule. HLA-DR3 molecules recognized residues
41e60, 91e110, and 178e227. In summary, all -DR (DR2, DR3 and
DR4) and -DQ (DQ6 and DQ8) molecules, recognized regions 41e60,
91e110, 178e197 and 208e227 of PLP. The majority of epitopes
identified largely encompassed regions previously reported to
be immunogenic in humans [28e30,32]. More importantly not allTable 1
Human myelin proteolipid protein specific T cell epitopes recognized by HLA class II mo
PLP epitope
1e20 31e50 41e60 51e70 81e100 91e110
DR2 (*1502)
DR3 (*0301)
DR4 (*0401)
DQ6 (*0601)
DQ8 (*0302)
Each black box represent a positive response to the peptide while white box denote no
a Only immunogenic region are shown.epitopes were restricted to various class II molecules tested. While,
some peptides elicited a response specific to a particular HLA class II
allele, others were promiscuous. PLP139e154 was immunogenic in
DQ6 and DQ8 transgenic mice but not in DR2, DR3 and DR4
transgenic mice. PLP peptide 1e20 was immunogenic only in DQ8
mice. In addition,we observed both DQandDR restricted response to
PLP91e110 whereas responses to 264e277 residues of PLP were
restricted only to DRmolecule. A similar observations have also been
reported in a study of MS patients from Japan [32]. Thus HLA class II
transgenic mice authenticate restriction of the proteolipid protein
(PLP) specific immune response implicated in MS pathogenesis.
2.2. PLP induced EAE in HLA class II transgenic mice
Our T cell epitope mapping data identified four immunodominant PLP epitopes (PLP 41e60, 91e110, 178e197, and 208e227)
in all DR and DQ transgenic mice. Previously, PLP95e116 had been
shown to be restricted by HLA-DR and -DQ using T cell lines from
MS patients while PLP95e116 specific T cell clones from HLA-DR2
transgenic mice can induce EAE in Rag2/ mice. These results
support a pathogenic role for PLP95e116-specific Tcells in HLA-DR2þ
MS patients, and shed light on the possible correlation between
autoimmune target epitope and disease phenotype in human CNS
autoimmune diseases [33].
Based on these studies, we selected PLP91e110 peptide for
induction of EAE in HLA class II transgenic mice. PLP91e110 peptide
induced a progressive, chronic EAE in 67% of the HLA-DR3.Ab mice
[31] and disease was characterized by a typical course of ascending
paralysis. The mean onset of the disease was 16  3 days, and
maximum disease severity score ranged from 1e3. No clinical sign
of disease was seen in transgene negative controls. The majority of
affected DR3 mice never went into remission for the length of
the test period (10 weeks), and developed a chronic form of EAE. No
clinical symptoms were observed in HLA-DR2, -DR4 or HLA-DQ6 or
-DQ8 transgenic mice [31]. DR3 transgenic mice with clinical signs
of EAE had diffuse meningeal infiltrates in both the spinal cord and
the brain. In addition, occasional sections of the spinal cord showed
paragonal mononuclear cell infiltrates that were closely associated
with the meningeal infiltrates. In the brain, mononuclear cell
infiltrates were seen primarily in themeningeal surfaces of the brain
stem, cerebellum, and surrounding the ventricles. Small areas of
demyelination were observed in spinal cord of DR3 mice with EAE.
Next we examined whether induction of EAE by PLP91e110
could lead to intra-molecular or inter-molecular spread of T cell
responses to other PLP regions or CNS antigens. Beside PLP91e110
induced proliferation responses, T cell proliferation was detected
against PLP peptides 141e160, 170e197, 188e207, 208e227, and
recombinant MOG, but not MBP protein. Using single amino acid
truncation and alanine substitution of encephalitogenic PLP91e110,
we identified the minimal epitope necessary for binding to the DR3
molecule and to induce EAE [34]. Residues necessary for binding
to HLA-DR3 molecule were identified as amino acid 97e108 of PLP.
Immunization of DR3 transgenic mice with the minimal epitopelecules.a
101e120 139e154 178e197 188e207 208e227 264e277
response to peptide.
D. Luckey et al. / Journal of Autoimmunity 37 (2011) 122e128124PLP97e108 led to induction of EAE and these mice showed classical
pathology associated with EAE. The alanine substitutions study
showed that residues 99, 102, and 103 are critical for immune
recognition of HLA-DR3 molecule [34].
Beside PLP91e110 peptide, Ito et al. [35] showed that PLP175e192
can induce a strong proliferative response and EAE in HLA-DR4
transgenic mice. Recently using the MBPePLP fusion protein
(MP4) [36], we have identified that PLP178e197 peptide can induce
EAE in HLA-DR2 (*1502) transgenic mice (Mangalam et al. unpublished observation). Our observation along with previous studies
indicate that presence of HLA-DR molecule is required for susceptibility to EAE, as transgenic mice expressing either the human DQ6
or DQ8 genes do not develop disease [27]. Based on these observation, we propose that HLA-DR genes such as HLA-DR2, -DR3 and
-DR4 are responsible for predisposition and susceptibility to
demyelinating disease, while polymorphism in DQ gene(s) might
play a modulating role. This hypothesis is supported by population
studies showing that the epistatic interaction between HLA molecules of the disease susceptible haplotypes plays an important role
in the final disease outcome in MS. While HLA-DQB1*0601 and
-DQB1*0603 protect against MS [3,4,19,37,38], DQB1*0602 and
DQB1*0302 alleles can increase disease susceptibility [5,11,37,39,40].
To understand the role of HLA-DQ molecules in the disease process,
we generated double transgenic mice expressing HLA-DQ6 or HLADQ8 gene on a disease susceptible HLA-DR3 background.
3. HLA-DQ6 (DQB1*0601) suppress EAE in HLA-DR3
transgenic mice by generating anti-inflammatory IFNg
Human population studies have suggested that HLA-DQ6
(DQB1*0601), found mostly in Asian populations protects from MS
[37,41]. To test this protective effect of DQ6 in an experimentalmodel,
we generated double transgenic mice expressing both HLA-DR3
and DQ6 on mouse class II negative background. Administration of
PLP91e110 to DR3.DQ6.Ab double transgenic mice with PLP91e110 led
to disease development only in 40% of double transgenic mice as
compared to 70% disease incidence in parental DR3.Ab transgenic
mice indicating, a protective role of the DQ6 gene [42]. The onset of
disease between these two groups was similar. Transgene negative
littermates or control Ab mice and DQ6 transgenic mice did not
develop clinical disease. This clinical disease data suggested that DQ6
plays a protective role by inhibiting development of EAE in disease
susceptible DR3 transgenic mice.
Next we analyzed if the protective effect of DQ6 is due to defect
in ability of DQ6 molecule to recognize PLP91e110 antigen. Interestingly, DR3.DQ6.Ab mice showed a very strong, dose dependent T
cell response to PLP91e110 antigen, which were at least three to four
folds higher in magnitude as compared to disease susceptible
DR3.Ab mice. Similar to double transgenic mice, HLA-DQ6 transgenic mice also showed a very strong Tcell proliferative response to
PLP91e110 as compared to DR3.Ab transgenic mice indicating that
the DQ6 molecule can present PLP91e110 antigen better than
DR3.Ab transgenic mice. Using the antibody blocking experiment,
we confirmed that the strong T cell response observed in
DR3.DQ6.Ab transgenic mice was restricted to DQ molecule.
As EAE was considered to be Th1 mediated disease, we argued
that resistance to EAE in DQ6 mice might be due to expression of
Th2 cytokines, which had been shown to be protective in EAE.
However, we observed that both the disease resistant DQ6 and
protected DR3.DQ6.Ab transgenic mice produced high levels of
IFNg, a cytokine normally associated with development of EAE
[43,44]. Beside high levels of IFNg, DR3.DQ6.Ab double transgenic
mice also produced a moderate level of IL-10 and high levels of IL-2
and IL-27. IL-4 levels were below detection limits in all samples
from single and double transgenic mice. In contrast, T cells fromdisease susceptible DR3.Ab transgenic mice produced higher
levels of IL-17, IL-22, and IL-23 as compared to DQ6 and DR3.DQ6
mice. We performed an IFNg-Elispot in the presence or absence of
blocking antibodies to confirm that HLA-DQ restricted CD4þ T cells
were the source of IFNg and not CD8 T cells or NK cells.
Our cytokine data indicated that DQ6 restricted IFNg might be
responsible for the protective effect observed in DR3.DQ6.Ab
transgenic mice. To confirm the role of IFNg in disease protection,
we performed in-vivo studies using neutralizating IFNg antibody.
DR3DQ6 transgenic mice treated with anti-IFNg but not with
isotype control showed increased disease incidence and severity,
similar to DR3.Ab transgenic mice, confirming a protective role of
IFNg in this model of EAE. Neutralizing antibody treatment in
DQ6.Abo mice had no effect. Thus high level of IFNg plays an
anti-inflammatory role and can suppress disease.
IFNg shows its anti-inflammatory effect through various
pathways such as induction of nitric oxide, generation of induced
Tregs and apoptosis of antigen specific T cells. PLP91e110 specific CD4
Tcells fromDQ6mice andDR3DQ6 transgenicmice produced higher
level of nitric oxide and had an increased frequency of CD4þFoxP3þ
T cells as compared to disease susceptible HLA-DR3 restricted CD4 T
cells. We also observed that T cells fromDQ6 aswell as DR3.DQ6.Ab
mice undergo increased proliferation and apoptosis as compared to
DR3 specific Tcells. Thus the protective effect of DQ6 in DR3.DQ6.Ab
double transgenic mice, was due to high levels of IFNg produced by
DQ6 restricted T cells, which suppressed proliferation of encephalitogenic DR3-restricted T cells by inducing apoptosis. Our study
suggests that DQ6 modifies the PLP91e110 specific T cell response in
DR3 through the anti-inflammatory effects of IFNg [42].
4. HLA-DQ8 (DQB1*0302) exacerbate disease in HLA-DR3
mice by generating pro-inflammatory IL-17
Presence of the HLA-DQ8/DR4 haplotype has been associated
with susceptibility to MS [16e18,45]. To test the role of DQ8
(DQB1*0302) in the immunopathogenesis of MS, we generated HLA
class II transgenic mice that express HLA-DQ8 and EAE susceptible
DR3 on an MHC II deficient background. Administration of PLP91e110
to DR3.DQ8.Ab double transgenic mice led to development of
disease in 100% of animals as compared to 70% disease incidence in
parental DR3.Ab transgenic mice, indicating that DQ8 synergizes
with DR3 for increased disease penetration. DR3DQ8 double transgenic mice showed an earlier disease onset with increased severity
as compared toDR3 transgenicmice (mean clinical score 3.4 0.2 vs.
2.3 0.3, p< 0.5). Thus DQ8plays amodulatory role in DR3.DQ8.Ab
double transgenic mice by inducing more severe EAE in disease
susceptible DR3 transgenic mice [46].
Disease susceptible DR3.Ab transgenic mice producedmoderate
to high levels of IFNg, TNFa, IL-2, IL-6, and IL-12 cytokines, showing
classical Th1 phenotype. Although CD4 Tcells fromDQ8mice did not
produce IFNg, they produced significantly higher levels of IL-17 and
GM-CSF (p < 0.01). Double transgenic DR3.DQ6.Ab mice also
produced higher levels of IL-17 and GM-CSF as well as IFNg, besides
producing moderate to high levels of TNFa, IL-1, IL-6, and IL-12
cytokines. IL-4 levels were below detection limits in all samples
from single and double transgenic mice. DR3.Ab transgenic mice
also producedmoderate amounts of IL-17, IL-21, and IL-23, however,
levels were significantly less (p < 0.01) as compared to DQ8.Ab or
DR3.DQ8.Ab mice. Both DR3 as well as DR3.DQ8.Ab transgenic
mice produced comparable levels of IL-27. Thus, disease susceptible
DR3.DQ8.Ab mice produced higher levels of IL-17, IL-21, IL-23, and
IFNg as compared to DR3 mice.
Using antibody blocking and cytokine Elispot assay, we
confirmed that increased levels of IFNg was produced by DR3
specific T cells, while IL-17 was produced by DQ8 specific T cells. To
Fig. 1. Antibody response against myelin antigens in HLA class II transgenic mice immunized with PLP91e110 peptide. DR3DQ8 mice with EAE showed higher level of antibody
against PLP91e110 peptide (A) as well as whole PLPeMBP fusion protein (B) compared to DR3 mice with EAE at day 7, 20 and 40 post-immunization. Sera were collected at indicated
time points and assayed using PLP91e110 or PLPeMBP coated plates using alkaline phosphatase-conjugated (AP) goat anti-mouse IgG (Jackson ImmunoResearch) and p-Nitrophenyl
phosphate (PNPP; Southern Biotechnology Associates Inc., Birmingham, AL) as substrate [45].
D. Luckey et al. / Journal of Autoimmunity 37 (2011) 122e128 125confirm the respective role of each cytokine in the EAE model, we
neutralized either IL-17 or IFNg in DR3.DQ8.Ab mice using specific
blocking antibodies and their respective isotype controls. While
neutralization of in-vivo levels of IFNg showed no effect on disease
incidence or severity, treatment with neutralizing IL-17 by its
specific antibody led to decrease in disease incidence and severity in
double transgenic DR3.DQ8.Ab mice as compared to mice treated
with isotype control antibody [46]. These set of data clearly indicates
that the increased disease severity observed in double transgenic
DR3.DQ8.Ab mice was due to high levels of IL-17 produced by DQ8
specific T cells. Blocking of IFNg or IL-17 in DQ8.Ab mice by
neutralizing antibody did not lead to development of disease.
Pathological analysis of brain and spinal cord tissues of mice with
EAE showed that DR3.DQ8.Ab double transgenic mice developed
severe CNS pathology as compared to DR3.Ab transgenic mice.
DR3.DQ8.Ab mice showed more widespread brain pathology with
severe inflammation anddemyelination in all parts of the brain tissue,
including cerebellum, brain stem, cortex, corpus callosum, stratium,
and meninges. In contrast, DR3.Ab transgenic mice showed inflammationprimarily localized to themeninges of the brain. DR3.DQ8.Ab
transgenic mice with EAE also showed typical parenchymal white
matter loss in brain, the classical pathology observed in MS. A similarFig. 2. Immunoflourescence detection of IgG and complement C3 deposition in brain of DR3
(red) and C3 deposition (green) in brain section as compared to DR3 mice with EAE. Expressi
IgG, anti-C3 antibody and fluorescence conjugate secondary antibody. Sections were fixed
(For interpretation of the references to color in this figure legend, the reader is referred topattern of pathology was also observed in the spinal cord with
increased inflammation and demyelination in DR3.DQ8.Ab mice as
compared to DR3.Ab mice. Thus, DR3.DQ8.Ab double transgenic
mice develop severe brain and spinal card pathology similar to brain
pathology observed in human MS.
5. Myelin-specific antibodies play an important role in
PLP91e110 induced EAE in HLA-DR3DQ8 transgenic mice
IL-17 can promote autoimmune disease through a mechanism
distinct from its pro-inflammatory effects [47]. It has been linked to
the induction of autoreactive humoral immune responses because
a deficiency in the blockade of IL-17 results in the decline of the
autoantibody response [48]. IL-17 can also induce the formation
of germinal centers, leading to the activation of B cells and an
increased level of antigen presentation and antibody production.
It is possible that the IL-17 induced B cells play an important role
in the disease exacerbation of DR3.DQ8.Ab mice. Further, administration of anti-myelin antibodies has been shown to enhance
demyelination in animal models of MS [49e52]. These pathogenic
antibodies have been shown to mediate tissue damage by recruitment of classical complement cascade [52,53]. The involvement ofand DR3DQ8 mouse with EAE. DR3DQ8 mice with EAE showed higher expression of IgG
on of IgG (red) and C3 (green) were detected by staining brain section with anti-mouse
, counterstained with nuclear DAPI stain and analyzed using fluorescent microscope.
the web version of this article.)
Fig. 3. The relative mRNA levels of C5 in CNS of DR3.Ab single and DR3.DQ8.Ab
double transgenic mice. DR3.DQ8.Ab showed higher expression of C5 mRNA as
compared to DR3.Ab mice. Expressions of C5 mRNA in different transgenic mice were
quantified by real-time PCR. Expression of b-actin was used as an internal control.
The expression of C5 mRNA in CNS of mice immunized with PLP91-110 relative to mice
immunized with control PLP peptide was calculated by the DDCt method.
D. Luckey et al. / Journal of Autoimmunity 37 (2011) 122e128126complement-dependent mechanisms in antibody mediated
demyelination and pathogenesis in EAE and MS had been noted in
several studies [54,55]. Therefore, we investigated the role of antimyelin antibodies and complement in disease exacerbation and the
increased CNS pathology observed in DR3DQ8 mice.
We first analyzed levels of anti-myelin specific antibodies in
single and double transgenic mice with EAE. DR3.DQ8.Ab double
transgenic mice and DR3.Ab single transgenic mice were immunized with PLP91e110 plus adjuvant as described previously [46] and
sera was collected at different time points. Levels of anti-PLP91e110
antibodies were detected using ELISA with PLP91e110 as the capture
antigen. As shown in Fig. 1A, DR3.DQ8.Ab double transgenic mice
with EAE showed higher levels of anti-PLP91e110 antibodies at all
time points tested from day 7 to 40 days post-immunization. Levels
of anti-PLP91e110 IgG increased overtime and maximum levels wereFig. 4. Schematic diagram of disease susceptibility and resistance in HLA class II transg
recognizing PLP91e110 peptide produce moderate levels of both Th1 (IFNg) as well as Th17
peptide produce high levels of IFNg and are resistant to EAE. IFNg produced by DQ6 restr
contrast, DQ8 mice recognizing PLP91e110 peptide produce high levels of IL-17 and GM-CSF
leads to severe disease indicating that DQ8 restricted IL-17 synergizes with DR3 to cause mo
in DR3DQ8 double transgenic mice may be due to induction of anti-myelin antibodies by ILthrough complement mediated cytotoxicity or through antibody dependent cytotoxicity.observed at day 40 post-immunization. Mice immunized with
PLP control peptide or CFA alone showed no reactivity (data not
shown). We also tested these sera against the whole PLPeMBP
molecules (MP4) using a fusion protein expressing both PLP and
MBP protein [36]. Similar to anti-PLP antibodies, we observed that
DR3.DQ8.Ab double transgenic mice with EAE produced higher
levels of anti-MP4 antibodies as compared to DR3.Ab single
transgenic mice with EAE at all time points (Fig. 1B).
Since, DR3.DQ8.Ab double transgenicmice showed severe disease
and a higher anti-PLP antibodies level, we hypothesized that increase
in clinical and pathological disease might be due to IgG
and complement deposition in the CNS leading to tissue destruction.
Fresh frozen section were stained with mouse anti-IgG or anti-C3
specific antibody and visualized using fluorescence microscope. As
shown in Fig. 2, we observed higher IgG and complement C3 deposition measured in brain sections from DR3.DQ8.Ab double transgenic mice with EAE as compared to DR3.Ab single transgenic mice
with EAE. The deposition was observed in area with inflammatory
lesions.We next analyzed expression of C5mRNA levels in CNS of and
observed that themean C5mRNA levels in single transgenic DR3mice
were 2 fold higher over control mice(Fig. 3). Whereas in DR3.DQ8.Ab
mice with EAE, C5 mRNA levels were 15 fold higher than control (7.5
fold increased over single transgenic DR3.Ab mice with EAE)(Fig. 3).
These data indicate that antibody and compliment play a role in
inducing the severe pathology observed in double transgenic mice.
These findings are in agreement with earlier reports that anti-myelin
antibodies and complement might play an important role in inducing
CNS demyelination in MS and EAE [49,52,56e60]. Antibodies against
myelin oligodendrocytes glycoprotein (MOG) had been shown to be
presentwithin demyelination lesions of caseswith acuteMS aswell as
in the marmoset model of EAE [57]. In addition, adoptive transfer of
MBP specific CD4 T cells in combination with demyelination monoclonal antibody specific for MOG in Lewis rats has been shown to
induce severe disease associated with large plaques of demyelination
[58]. Further, pathogenic anti-myelin antibodies had been shown toenic mice and disease modulation by HLA-DQ molecules. HLA-DR3 transgenic mice
(IL-17) cytokines and are susceptible to EAE. Whereas DQ6 mice recognizing the same
icted T cells is anti-inflammatory as it protects DR3DQ6 from development of EAE. In
but are also resistant to development of EAE. Interestingly, presence of DQ8 with DR3
re severe disease. Our recent data show that increased disease and pathology observed
17, which might lead to complement deposition and subsequent neuropathology either
D. Luckey et al. / Journal of Autoimmunity 37 (2011) 122e128 127induce demyelination through complement activation as mice deficient in C3 develop a milder form of EAE compared to C3 sufficient
mice [52,60].6. Concluding remarks
The main advantage of mouse EAE is that genetically engineered
mutants can be generated and bred. Thus, the influence of genetics
on susceptibility, disease course, inflammation and demyelination
can be studied. An EAE mouse model where the autoreactive T cells
repertoire is selected and shaped by human MHC class II molecules
will provide new information on immunopathogenesis of inflammatory and demyelinating diseases such as human MS. Thus HLA
transgenic mice had been quite useful in understanding the role
of HLA class II molecule in immunopathogenesis of EAE/MS. The
data from our single and double transgenic mice indicate the final
outcome of disease might be dependent on interaction between
HLA-DR and -DQ molecules. Utilizing these mice we were able to
identifying immunodominant as well as encephalitogenic epitopes
ofmyelin antigen such as PLP,MBPandMOG. Further, data generated
from EAE in HLA class II transgenic mice suggest that HLA-DR
molecules is required for susceptibility to disease while the
HLA-DQmoleculemight play amodulatory role through the pro and
anti-inflammatory cytokine network(Fig. 4). We have shown that
while HLA-DQ6 (DQb1*0601) can protect DR3 mice from EAE by
producing anti-inflammatory IFNg; HLA-DQ8 (DQb1*0302) synergize with DR3 to induce a severe disease in DR3DQ8 double transgenic mice by producing pro-inflammatory IL-17 and GM-CSF.
Further we have presented data indicating that IL-17 directly or
indirectly helps in induction of B cells producing anti-myelin antibodies. We have also shown increased deposition of IgG and
complement together with an increased expression of complement
in the brain of double transgenic mice with severe EAE. It will be
interesting to investigate the mechanism responsible for the
production of protective IFNg from HLA-DQ6 and inflammatory IL17 from HLA-DQ8 restricted T cells. We join other contributors of
this special issue in recognizing the immense contribution of Chella
David for his many achievements in the field of autoimmunity
including generation of HLA transgenic mice. Advent of these HLA
transgenic mice have helped immensely in understanding the
immunopathogenesis of various inflammatory, autoimmune,
allergic and infectious diseases. Finally, we note that this issue is part
of the Journal of Autoimmunity’s commitment in the recognition of
outstanding scientists in thefield of autoimmunity.We are delighted
that Chella Davis is being so honored and that he is part of the
continuum of distinguished autoimmunologists and dedicated
themes in the journal [61e71].Acknowledgements
This work was supported by NS52173 from NINDS. AKM was
supported by DOD grant 10-1-0254 and NS52173. We thank Julie
Hanson and her staff for mouse husbandry and Michele Smart for
tissue typing of transgenic mice.References
[1] Alvarado-de la Barrera C, Zuniga-Ramos J, Ruiz-Morales JA, Estanol B,
Granados J, Llorente L. HLA class II genotypes in Mexican Mestizos with
familial and nonfamilial multiple sclerosis. Neurology 2000;55:1897e900.
[2] Barcellos LF, Oksenberg JR, Begovich AB, Martin ER, Schmidt S, Vittinghoff E,
et al. HLA-DR2 dose effect on susceptibility to multiple sclerosis and influence
on disease course. American Journal of Human Genetics 2003;72:710e6.
[3] DeLuca GC, Ramagopalan SV, Herrera BM, Dyment DA, Lincoln MR,
Montpetit A, et al. An extremes of outcome strategy provides evidence that
multiple sclerosis severity is determined by alleles at the HLA-DRB1 locus.Proceedings of the National Academy of Sciences of the United States of
America 2007;104:20896e901.
[4] Dyment DA, Herrera BM, Cader MZ, Willer CJ, Lincoln MR, Sadovnick AD, et al.
Complex interactions among MHC haplotypes in multiple sclerosis: susceptibility and resistance. Human Molecular Genetics 2005;14:2019e26.
[5] Oksenberg JR, Barcellos LF, Cree BA, Baranzini SE, Bugawan TL, Khan O, et al.
Mapping multiple sclerosis susceptibility to the HLA-DR locus in African
Americans. American Journal of Human Genetics 2004;74:160e7.
[6] Olerup O, Hillert J. HLA class II-associated genetic susceptibility in multiple
sclerosis: a critical evaluation. Tissue Antigens 1991;38:1e15.
[7] Ramagopalan SV, Deluca GC, Morrison KM, Herrera BM, Dyment DA,
Lincoln MR, et al. Analysis of 45 candidate genes for disease modifying activity
in multiple sclerosis. Journal of Neurology 2008;255:1215e9.
[8] Ramagopalan SV, Morris AP, Dyment DA, Herrera BM, DeLuca GC, Lincoln MR,
et al. The inheritance of resistance alleles in multiple sclerosis. PLoS Genetics
2007;3:1607e13.
[9] McDonald WI. Multiple sclerosis: epidemiology and HLA associations. Annals
of the New York Academy of Sciences 1984;436:109e17.
[10] Oksenberg JR, Begovich AB, Erlich HA, Steinman L. Genetic factors in multiple
sclerosis. Jama 1993;270:2362e9.
[11] Barcellos LF, Sawcer S, Ramsay PP, Baranzini SE, Thomson G, Briggs F, et al.
Heterogeneity at the HLA-DRB1 locus and risk for multiple sclerosis. Human
Molecular Genetics 2006;15:2813e24.
[12] Dyment DA, Sadovnick AD, Ebers GC. Genetics of multiple sclerosis. Human
Molecular Genetics 1997;6:1693e8.
[13] CullenCG,MiddletonD, SavageDA,Hawkins S.HLA-DRandDQDNAgenotyping in
multiple sclerosis patients inNorthern Ireland. Human Immunology 1991;30:1e6.
[14] Marrosu MG, Murru MR, Costa G, Cucca F, Sotgiu S, Rosati G, et al. Multiple
sclerosis in Sardinia is associated and in linkage disequilibrium with HLA-DR3
and -DR4 alleles. American Journal of Human Genetics 1997;61:454e7.
[15] Masterman T, Ligers A, Olsson T, Andersson M, Olerup O, Hillert J. HLA-DR15 is
associated with lower age at onset in multiple sclerosis. Annals of Neurology
2000;48:211e9.
[16] Weinshenker BG, Santrach P, Bissonet AS, McDonnell SK, Schaid D, Moore SB,
et al. Major histocompatibility complex class II alleles and the course and
outcome of MS: a population-based study. Neurology 1998;51:742e7.
[17] Marrosu MG, Muntoni F, Murru MR, Spinicci G, Pischedda MP, Goddi F, et al.
Sardinian multiple sclerosis is associated with HLA-DR4: a serologic and
molecular analysis. Neurology 1988;38:1749e53.
[18] Zivadinov R, Uxa L, Bratina A, Bosco A, Srinivasaraghavan B, Minagar A, et al.
HLA-DRB1*1501, -DQB1*0301, -DQB1*0302, -DQB1*0602, and -DQB1*0603
alleles are associatedwithmore severe disease outcome onMRI in patients with
multiple sclerosis. International Review of Neurobiology 2007;79:521e35.
[19] Chao MJ, Barnardo MC, Lincoln MR, Ramagopalan SV, Herrera BM, Dyment DA,
et al. HLA class I alleles tag HLA-DRB1*1501 haplotypes for differential risk in
multiple sclerosis susceptibility. Proceedings of the National Academy of
Sciences of the United States of America 2008;105:13069e74.
[20] Gold R, Hartung HP, Toyka KV. Animal models for autoimmune demyelinating
disorders of the nervous system. Molecular Medicine Today 2000;6:88e91.
[21] Martin R, McFarland HF, McFarlin DE. Immunological aspects of demyelinating diseases. Annual Review of Immunology 1992;10:153e87.
[22] Swanborg RH. Experimental autoimmune encephalomyelitis in the rat: lessons in
T-cell immunology and autoreactivity. Immunological Reviews 2001;184:129e35.
[23] Ando DG, Clayton J, Kono D, Urban JL, Sercarz EE. Encephalitogenic T cells in
the B10.PL model of experimental allergic encephalomyelitis (EAE) are of the
Th-1 lymphokine subtype. Cell Immunology 1989;124:132e43.
[24] Kuchroo VK, Anderson AC, Waldner H, Munder M, Bettelli E, Nicholson LB.
T cell response in experimental autoimmune encephalomyelitis (EAE): role of
self and cross-reactive antigens in shaping, tuning, and regulating the autopathogenic T cell repertoire. Annual Review of Immunology 2002;20:101e23.
[25] Zamvil SS, Steinman L. The T lymphocyte in experimental allergic encephalomyelitis. Annual Review of Immunology 1990;8:579e621.
[26] Korn T, Bettelli E, Oukka M, Kuchroo VK. IL-17 and Th17 cells. Annual Review
of Immunology 2009;27:485e517.
[27] Mangalam AK, Rajagopalan G, Taneja V, David CS. HLA class II transgenic mice
mimic human inflammatory diseases. Advances in Immunology2008;97:65e147.
[28] Greer JM, Csurhes PA, Cameron KD, McCombe PA, Good MF, Pender MP.
Increased immunoreactivity to two overlapping peptides of myelin proteolipid protein in multiple sclerosis. Brain 1997;120(Pt 8):1447e60.
[29] Martino G, Hartung HP. Immunopathogenesis of multiple sclerosis: the role of
T cells. Current Opinion in Neurology 1999;12:309e21.
[30] Schmidt S. Candidate autoantigens in multiple sclerosis. Multiple Sclerosis
(Houndmills, Basingstoke, England) 1999;5:147e60.
[31] Mangalam AK, Khare M, Krco C, Rodriguez M, David C. Identification of T cell
epitopes on human proteolipid protein and induction of experimental
autoimmune encephalomyelitis in HLA class II-transgenic mice. European
Journal of Immunology 2004;34:280e90.
[32] Minohara M, Ochi H, Matsushita S, Irie A, Nishimura Y, Kira J. Differences
between T-cell reactivities to major myelin protein-derived peptides in
opticospinal and conventional forms of multiple sclerosis and healthy
controls. Tissue Antigens 2001;57:447e56.
[33] Kawamura K, Yamamura T, Yokoyama K, Chui DH, Fukui Y, Sasazuki T, et al.
HLA-DR2-restricted responses to proteolipid protein 95-116 peptide cause
autoimmune encephalitis in transgenic mice. The Journal of Clinical Investigation 2000;105:977e84.
D. Luckey et al. / Journal of Autoimmunity 37 (2011) 122e128128[34] Mangalam AK, Khare M, Krco CJ, Rodriguez M, David CS. Delineation of the
minimal encephalitogenic epitope of proteolipid protein peptide(91-110) and
critical residues required for induction of EAE in HLA-DR3 transgenic mice.
Journal of Neuroimmunology 2005;161:40e8.
[35] Ito K, Bian HJ, Molina M, Han J, Magram J, Saar E, et al. HLA-DR4-IE chimeric
class II transgenic, murine class II-deficient mice are susceptible to experimental allergic encephalomyelitis. The Journal of Experimental Medicine
1996;183:2635e44.
[36] Elliott EA, McFarland HI, Nye SH, Cofiell R, Wilson TM, Wilkins JA, et al.
Treatment of experimental encephalomyelitis with a novel chimeric fusion
protein of myelin basic protein and proteolipid protein. The Journal of Clinical
Investigation 1996;98:1602e12.
[37] Amirzargar A, Mytilineos J, Yousefipour A, Farjadian S, Scherer S, Opelz G,
et al. HLA class II (DRB1, DQA1 and DQB1) associated genetic susceptibility in
Iranian multiple sclerosis (MS) patients. European Journal of Immunogenetics
1998;25:297e301.
[38] Herrera BM, Ebers GC. Progress in deciphering the genetics of multiple
sclerosis. Current Opinion in Neurology 2003;16:253e8.
[39] Das P, Drescher KM, Geluk A, Bradley DS, Rodriguez M, David CS. Complementation between specific HLA-DR and HLA-DQ genes in transgenic mice
determines susceptibility to experimental autoimmune encephalomyelitis.
Human Immunology 2000;61:279e89.
[40] Serjeantson SW, Gao X, Hawkins BR, Higgins DA, Yu YL. Novel HLA-DR2-related
haplotypes in Hong Kong Chinese implicate the DQB1*0602 allele in susceptibility tomultiple sclerosis. European Journal of Immunogenetics 1992;19:11e9.
[41] Marrosu MG, Muntoni F, Murru MR, Costa G, Pischedda MP, Pirastu M, et al.
HLA-DQB1 genotype in Sardinian multiple sclerosis: evidence for a key role of
DQB1 *0201 and *0302 alleles. Neurology 1992;42:883e6.
[42] Mangalam A, Luckey D, Basal E, Behrens M, Rodriguez M, David C. HLA-DQ6
(DQB1*0601)-restricted T cells protect against experimental autoimmune
encephalomyelitis in HLA-DR3.DQ6 double-transgenic mice by generating
anti-inflammatory IFN-gamma. Journal of Immunology 2008;180:7747e56.
[43] Steinman L. A rush to judgment on Th17. The Journal of Experimental
Medicine 2008;205:1517e22.
[44] Stromnes IM, Cerretti LM, Liggitt D, Harris RA, Goverman JM. Differential
regulation of central nervous system autoimmunity by T(H)1 and T(H)17 cells.
Nature Medicine 2008;14:337e42.
[45] Khare M, Mangalam A, Rodriguez M, David CS. HLA DR and DQ interaction in
myelin oligodendrocyte glycoprotein-induced experimental autoimmune
encephalomyelitis in HLA class II transgenic mice. Journal of Neuroimmunology 2005;169:1e12.
[46] Mangalam A, Luckey D, Basal E, Jackson M, Smart M, Rodriguez M, et al. HLADQ8 (DQB1*0302)-restricted Th17 cells exacerbate experimental autoimmune
encephalomyelitis in HLA-DR3-transgenic mice. Journal of Immunology 2009;
182:5131e9.
[47] Hsu HC, Yang P, Wang J, Wu Q, Myers R, Chen J, et al. Interleukin 17-producing
T helper cells and interleukin 17 orchestrate autoreactive germinal center
development in autoimmune BXD2 mice. Nature Immunology 2008;9:166e75.
[48] Komiyama Y, Nakae S, Matsuki T, Nambu A, Ishigame H, Kakuta S, et al. IL-17
plays an important role in the development of experimental autoimmune
encephalomyelitis. Journal of Immunology 2006;177:566e73.
[49] Piddlesden SJ, Lassmann H, Zimprich F, Morgan BP, Linington C. The demyelinating potential of antibodies to myelin oligodendrocyte glycoprotein is
related to their ability to fix complement. The American Journal of Pathology
1993;143:555e64.
[50] Raine CS, Cannella B, Hauser SL, Genain CP. Demyelination in primate autoimmune encephalomyelitis and acute multiple sclerosis lesions: a case for
antigen-specific antibody mediation. Annals of Neurology 1999;46:144e60.
[51] Schluesener HJ, Sobel RA, Linington C,Weiner HL. Amonoclonal antibody against
a myelin oligodendrocyte glycoprotein induces relapses and demyelination incentral nervous system autoimmune disease. Journal of Immunology 1987;139:
4016e21.
[52] Urich E, Gutcher I, Prinz M, Becher B. Autoantibody-mediated demyelination
depends on complement activation but not activatory Fc-receptors.
Proceedings of the National Academy of Sciences of the United States of
America 2006;103:18697e702.
[53] Pollinger B, Krishnamoorthy G, Berer K, Lassmann H, Bosl MR, Dunn R, et al.
Spontaneous relapsing-remitting EAE in the SJL/J mouse: MOG-reactive
transgenic T cells recruit endogenous MOG-specific B cells. The Journal of
Experimental Medicine 2009;206:1303e16.
[54] Du C, Sriram S. Increased severity of experimental allergic encephalomyelitis in lyn-/- mice in the absence of elevated proinflammatory cytokine
response in the central nervous system. Journal of Immunology 2002;168:
3105e12.
[55] Rus H, Cudrici C, Niculescu F, Shin ML. Complement activation in autoimmune
demyelination: dual role in neuroinflammation and neuroprotection. Journal
of Neuroimmunology 2006;180:9e16.
[56] Cross AH, Stark JL. Humoral immunity in multiple sclerosis and its animal
model, experimental autoimmune encephalomyelitis. Immunologic Research
2005;32:85e97.
[57] Genain CP, Cannella B, Hauser SL, Raine CS. Identification of autoantibodies
associated with myelin damage in multiple sclerosis. Nature Medicine 1999;5:
170e5.
[58] Linington C, Engelhardt B, Kapocs G, Lassman H. Induction of persistently
demyelinated lesions in the rat following the repeated adoptive transfer of
encephalitogenic T cells and demyelinating antibody. Journal of Neuroimmunology 1992;40:219e24.
[59] Martin Mdel P, Monson NL. Potential role of humoral immunity in the pathogenesis of multiple sclerosis (MS) and experimental autoimmune encephalomyelitis (EAE). Frontiers in Bioscience 2007;12:2735e49.
[60] Szalai AJ, Hu X, Adams JE, Barnum SR. Complement in experimental autoimmune encephalomyelitis revisited: C3 is required for development of maximal
disease. Molecular Immunology 2007;44:3132e6.
[61] Youinou P. Haralampos M. Moutsopoulos: a lifetime in autoimmunity. Journal
of Autoimmunity 2010;35:171e5.
[62] Miyagawa F, Gutermuth J, Zhang H, Katz SI. The use of mouse models to better
understand mechanisms of autoimmunity and tolerance. Journal of Autoimmunity 2010;35:192e8.
[63] Scheinecker C, Bonelli M, Smolen JS. Pathogenetic aspects of systemic lupus
erythematosus with an emphasis on regulatory T cells. Journal of Autoimmunity 2010;35:269e75.
[64] Youinou P, Pers J-O. The international symposium on Sjogren’s syndrome in
Brest: the "top of the tops" at the "tip of the tips". Autoimmunity Reviews
2010;9:589e90.
[65] Lu Q, Renaudineau Y, Cha S, Ilei G, Brooks WH, Selmi C, et al. Epigenetics in
autoimmune disorders: highlights of the 10th Sjogren’s syndrome symposium. Autoimmunity Reviews 2010;9:627e30.
[66] Youinou P, Pers J-O, Gershwin ME, Shoenfeld Y. Geo-epidemiology and
autoimmunity. Journal of Autoimmunity 2010;34:J163e7.
[67] Chang C, Gershwin ME. Drugs and autoimmunity - A contemporary review
and mechanistic approach. Journal of Autoimmunity 2010;34:J266e75.
[68] Borchers AT, Naguwa SM, Keen CL, Gershwin ME. The implications of
autoimmunity and pregnancy. Journal of Autoimmunity 2010;34:J287e99.
[69] Schilder AM. Wegener’s granulomatosis vasculitis and granuloma. Autoimmunity Reviews 2010;9:483e7.
[70] Ansari AA, Gershwin ME. Navigating the passage between Charybdis and
Scylla: recognizing the achievements of Noel Rose. Journal of Autoimmunity
2009;33:165e9.
[71] Mackay IR. Clustering and commonalities among autoimmune diseases.
Journal of Autoimmunity 2009;33:170e7.
Loss of Sex and Age Driven Differences in the Gut
Microbiome Characterize Arthritis-Susceptible *0401
Mice but Not Arthritis-Resistant *0402 Mice
Andres Gomez1, David Luckey2, Carl J. Yeoman1, Eric V. Marietta3, Margret E. Berg Miller1,
Joseph A. Murray2,3, Bryan A. White1,5*, Veena Taneja2,4*
1 Institute for Genomic Biology, University of Illinois, Urbana, Illinois, United States of America, 2 Department of Immunology, Mayo Clinic, Rochester, Minnesota, United
States of America, 3 Department of Gasteroenterology, , United States of America, 4 Department of Rheumatology, Mayo Clinic,
Rochester, Minnesota, United States of America, 5 Department of Animal Sciences, University of Illinois, Urbana, Illinois, United States of America
Abstract
Background: HLA-DRB1*0401 is associated with susceptibility, while HLA-DRB1*0402 is associated with resistance to
developing rheumatoid arthritis (RA) and collagen-induced arthritis in humans and transgenic mice respectively. The
influence of gut-joint axis has been suggested in RA, though not yet proven.
Methodology/Principal Findings: We have used HLA transgenic mice carrying arthritis susceptible and -resistant HLA-DR
genes to explore if genetic factors and their interaction with gut flora gut can be used to predict susceptibility to develop
arthritis. Pyrosequencing of the 16S rRNA gene from the fecal microbiomes of DRB1*0401 and DRB1*0402 transgenic mice
revealed that the guts of *0401 mice is dominated by a Clostridium-like bacterium, whereas the guts of *0402 mice are
enriched for members of the Porphyromonadaceae family and Bifidobacteria. DRB1*0402 mice harbor a dynamic sex and
age-influenced gut microbiome while DRB1*0401 mice did not show age and sex differences in gut microbiome even
though they had altered gut permeability. Cytokine transcripts, measured by rtPCR, in jejuna showed differential TH17
regulatory network gene transcripts in *0401 and *0402 mice.
Conclusions/Significance: We have demonstrated for the first time that HLA genes in association with the gut microbiome
may determine the immune environment and that the gut microbiome might be a potential biomarker as well as
contributor for susceptibility to arthritis. Identification of pathogenic commensal bacteria would provide new
understanding of disease pathogenesis, thereby leading to novel approaches for therapy.
Citation: Gomez A, Luckey D, Yeoman CJ, Marietta EV, Berg Miller ME, et al. (2012) Loss of Sex and Age Driven Differences in the Gut Microbiome Characterize
Arthritis-Susceptible *0401 Mice but Not Arthritis-Resistant *0402 Mice. PLoS ONE 7(4): e36095. doi:10.1371/journal.pone.0036095
Editor: Sarah K. Highlander, Baylor College of Medicine, United States of America
Received December 14, 2011; Accepted March 27, 2012; Published April 24, 2012
Copyright:  2012 Gomez et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: The study was supported by seed funds from the Mayo Clinic/Illinois Alliance for Technology based Healthcare and a grant from Department of
Defense to VT. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: taneja.veena@mayo.edu (VT); bwhitw44@illinois.edu (BAW)
Introduction
Rheumatoid arthritis (RA) is a chronic inflammatory disease
that is characterized by synovial inflammation and erosion of bone
and cartilage leading to the destruction of joints. Although the
etiology of RA is unknown, both genetic and environmental
factors contribute to the susceptibility to developing arthritis [1].
Among the known genetic factors, strong associations are observed
between RA and the presence of certain HLA-DR alleles that
share the 3rd hypervariable region with DRB1*0401 gene, known
as the ‘shared epitope’ hypothesis. In contrast, DRB1*0402
confers resistance to the development of arthritis. Some evidence
points to an infectious etiology for RA, such the presence of certain
oral and gut commensal bacterial antigens in synovial fluids of
patients [1,2,3]. Migration of gut commensals or their products to
peripheral organs may be facilitated by loss of intestinal integrity,
resulting in mucosal or systemic immune stimulation. Recent
studies have shown that specific intestinal commensals or their
specific molecular patterns may modulate the integrity of the
intestinal mucosal barrier by inducing the expression of pro or
anti-inflammatory cytokines [4,5]. Thus, alterations of a normal
gut microbiome can affect mucosal immunity and have an
extended effect on non-intestinal diseases like diabetes and RA [6].
For instance, previous analysis of the fecal microbiome of patients
with RA revealed significantly fewer Bifidobacterium and bacteria of
the Bacteroides-Porphyromonas-Prevotella group, B. fragilis subgroup,
and the Eubacterium rectale–Clostridium coccoides group than the fecal
microbiota of patients with non-inflammatory fibromyalgia [7].
Because these bacterial species are known to belong to common
taxa in the human fecal microbiome [8,9,10], their low levels in
RA patients might suggest an altered gut microbiome. Further,
specific gut commensals such as Bifidobacterium infantis, can induce
an anti-inflammatory response in the intestinal mucosal and
peripheral immune systems by suppressing T-cell proliferation and
production of IL-10 and Th2 cytokines, and by inhibiting nuclear
factor kappa B (NF-kB) activation [11,12,13]. Although dendritic
PLoS ONE | www.plosone.org 1 April 2012 | Volume 7 | Issue 4 | e36095
Mayo C inic, Rochester, Minnesotal
cells (DCs), directly in contact with intestinal lumen contents, can
instruct naive CD4+ T-cells to differentiate into Th1, Th17, Th2
or T-regulatory cells, a unique gastro-intestinal environment may
favor the proliferation of the latter, a process possibly dependent
on the presence of specific gut commensal bacteria thus setting up
the basis for immunotolerance [14,15,16].
We have generated two lines of HLA transgenic mice carrying
the RA-susceptible DRB1*0401 and RA-resistant DRB1*0402
genes that lacked all four classical murine chains, Aa, Ab, Ea, Eb.
Because human class II molecules shape the T-cell repertoire in
these humanized mice, they show the same HLA restrictions in an
immune response as humans [17,18,19]. Upon immunization with
type II collagen (CII), *0401 mice develop collagen-induced
arthritis (CIA), while *0402 mice do not. The most remarkable
features of CIA in *0401 mice that is not observed in any other
model is a sex-bias in the onset of arthritis with a ratio of 3
Females: 1 Male, production of rheumatoid factor (RF) and anticyclic citrullinated peptide antibodies (ACPAs), diagnostic markers
for RA patients [18,20]. Host MHC genes affect the microbial
composition of the gut [21,22]. However, the interactions between
host genetic factors like MHC and their gut microbiota, and their
impact on the development of RA are difficult to study in humans
due to several factors that include: high HLA polymorphism, diet
and the fact that the disease is well established at the time of
diagnosis. Few studies describing microbiome of RA patients have
not done tag sequencing and also not analyzed the data according
to sex and age. Thus, HLA transgenic mice described here provide
a useful tool to understand the role of gut microbiota in the
pathogenesis of RA. Herein, we show that mice with the RAsusceptible DRB1*0401 gene harbor altered patterns of gut
microbiome characterized by an abundance and/or lack of
specific commensals as compared to mice with the RA-resistant
DRB1*0402 gene whose gut microbiomes are shaped by age and
sex. A differential expression of Th17 regulating gene transcripts, a
compromised gut permeability in *0401 mice and observed
dysbiosis in *0401 mice may in combination or independently
contribute to susceptibility to arthritis.
Results
Arthritis-susceptible DRB1*0401 and –resistant
DRB1*0402 mice differ in their gut microbiomes
The gut microbiome plays a crucial role in the homeostasis of
the immune system and is also linked to gut permeability. We
tested if an arthritis-susceptible genotype may be associated with
the presence or absence of specific gut bacteria by sequencing the
microbiome community structure in fecal samples of 87 mice
(n = 41 for *0401 and n = 45 for *0402 mice) using Roche 454 GSFLX Titanium Pyrosequencing technology. This included both
male and female mice of various ages for both strains. After
processing, 568,571 high quality sequences were used for further
analysis (sequence lengths ranged from 417 to 534 bp with a 506
median sequence length). A total of 5,267 operational taxonomic
units (OTUs) clustered at 97% sequence similarity were used for
microbiome analysis (1,953 to 60,915 reads per sample).
Non-metric multidimensional scaling (NMDS) and analysis of
similarities (ANOSIM) suggest that *0401(n = 41) and *0402
(n = 45) mice display only minor differences in their fecal
microbiome profiles (ANOSIM R-statistic = 0.14, P = 0.001)
(Figure 1 a). There were no significant differences in bacterial
richness between the 2 strains (P.0.1). However, the fecal
bacterial species in *0401 mice were slightly more evenly
distributed than those in *0402 mice (Shannon evenness index,
P = 0.04). Both NMDS and ANOSIM analysis of males and
females from each strain showed that sex was a confounding factor
and that males were masking the differences between the fecal
bacterial profiles of resistant and susceptible mice (Figure 1 b); the
ANOSIM R value between *0401 (n = 22) and *0402 (n = 21)
males was only 0.145 (P,0.001). In contrast, differences in fecal
microbiome structure were more evident between *0401 (n = 19)
and *0402 (n = 24) females (ANOSIM R statistic = 0.436,
P,0.001), (Figure 1 c and d). DRB1*0402 mice showed
dynamically different fecal microbiomes based on sex (males and
females, n = 21 and 24 respectively and age (,4 and .4 months
old, n = 27 and n = 18 respectively) factors (ANOSIM Rstatistic = 0.302 and 0.423 for sex and age differences respectively,
P,0.001) (Figure 2a and 2b). Unlike arthritis-resistant *0402 mice
however, the structure of the fecal microbiomes of *0401 mice lost
these sex (males and females, n = 22 and n = 19 respectively) and
age (,4 and .4 months old, n = 30 and n = 11 respectively) driven differences (ANOSIM R values of 0.052 and 0.043 for
gender and age differences respectively, P.0.1) (Figures 3 c and
d).
Specific gut commensals contribute to strain, sex and
age differences in mice fecal microbiomes
A percentages-species contribution analysis (SIMPER) and
Taxonomic search using RDP and NCBI databases allowed us
to identify and characterize the relative abundance distributions of
the five Operational Taxonomic Units (OTUs) that contributed
more than 2.5% to the observed differences of the fecal
microbiomes between resistant and susceptible transgenic mice
(Table 1, Figure 3a). The phyla distributions of OTUs followed
similar patterns when taking into account all of the 5,267 OTUs
detected (Figure 3 b); at this taxonomic level, *0401 mice
presented a more even Bacteroidetes/Firmicutes ratio (,1:1) than
*0402 mice (,2:1).
An OTU related to Allobaculum sp. (84% identity to A.
stercoricanis) or an unclassified member of the Clostridiales (87%
identity) was more abundant in disease-susceptible (*0401) mice
compared to *0402 mice (P,0.00001). On the other hand, OTUs
related to Bifidobacterium, Barnesiella and Parabacteroides spp., were
more abundant in disease-resistant mice (P = 0.0029) (Figure 3 a).
Data on the relative abundance of the OTU’s driving microbiome
differences between the two strains were used to construct a simple
correspondence analysis plot (CA) showing the level of correlation
between each OTU and members of either strain (Figure 4 c).
Dimension 1 (Axis CA1) of the CA plot explained 42.94% of the
total variation in the data and distinguished between susceptible
mice, more correlated to the abundance of Allobaculum sp. and
resistant mice, associated to greater proportions of the Bifidobacteria
and the Parabacteroides-Barnesiella group.
Sex based differences in the fecal microbiomes of *0402 mice
were driven mainly by Bifidobacterium pseudolongum subsp. Globosum
and Parabacteroides distasonis, each being more prevalent in females
(P = 0.018. and 0.00017 for Bifidobacterium and Parabacteroides
respectively) (Figure 4a) while, Barnesiella viscericola was more
abundant in males (P = 0.00018. n = 21 and 24 for resistant male
and female mice respectively). Dimension 1 of a CA plot
describing the level of association between either sex and specific
OTUs explained 57.29% of the total variation in the data and
showed high correlation between the relative abundance of
Barnesiella viscericola and *0402 males, and of Bifidobacterium
pseudolongum and Parabacteroides distasonis and *0402 females
(Figure 4 b). Even though fecal microbiome of *0401 mice were
dysbiotic and less dynamic (Figure 4 d), susceptible males did show
significantly higher abundances of B. pseudolongum than susceptible
females (P = 0.02, n = 22 and 19 for *0401 males and females
Gut Microbiome Identifies Arthritis Susceptibility
PLoS ONE | www.plosone.org 2 April 2012 | Volume 7 | Issue 4 | e36095
Figure 1. Multidimensional Scalinge analysis of the fecal microbiomes of arthritis-susceptible *0401 and –resistant *0402 mice. 16SrDNA bacterial community structure differences can be visualized with each symbol representing data from a single mouse fecal sample. (a) 16SrDNA bacterial community structures between *0401 (n = 41) and *0402 (n = 45) mice do not differ significantly (ANOSIM R = 0.14). (b) *0401 (n = 22)
and *0402 (n = 21) males do not show significant differences in fecal microbiome structure (ANOSIM R = 0.14), while (c) fecal microbiomes of *0401
(n = 19) and *0402 (n = 24) females differ significantly (ANOSIM R = 0.436). (d) Shaded area shows that the fecal microbiomes of *0402 females are
compact and may be driving differences between both genotypes.
doi:10.1371/journal.pone.0036095.g001
Figure 2. Sex and Age based Multidimensional Scalinge analysis of fecal microbiomes. (a) *0402 mice show significantly different fecal
microbiome structure according to sex (ANOSIM R = 0.302) and (b) age (ANOSIM R = 0.423). (c) *0401 mice lost sex and (d) age-based differences in
fecal microbiome (ANOSIM R = 0.052 and R = 0.043 respectively). Shaded areas show compact microbiome structures.
doi:10.1371/journal.pone.0036095.g002
Gut Microbiome Identifies Arthritis Susceptibility
PLoS ONE | www.plosone.org 3 April 2012 | Volume 7 | Issue 4 | e36095
respectively) (Figure 4 c). Age dependent fecal microbiota
differences in resistant mice were driven mainly by B. viscericola
that more abundant in older mice (,4 compared to .4 months,
n = 18 and 27 respectively, P = 0.0001). However, the relative
abundances of Bifidobacterium and Parabacteroides were not significantly different between older and younger *0402 mice (P = 0.766
and P = 0.0567 for Bifidobacterium and Parabacteroides respectively.
There were no significant age-driven differences in the relative
abundance of specific OTUs between susceptible mice.
Arthritis susceptible DRB1*0401 mice show altered
mucosal immune function and increased gut
permeability compared to resistant DRB1*0402 mice
We tested the hypothesis that dysbiosis in gut flora of *0401
mice may be associated with an altered intestinal permeability as
well as a distinct expression of pro and anti-inflammatory
cytokines in the gut as compared to arthritis-resistant mice, and
that this dysbiosis may play a role in the pathogenesis of arthritis.
A comparison of gut permeability between arthritic and naı̈ve
Figure 3. Relative abundance of OTUs in the fecal microbiomes of *0401 and *0402 mice. (a) Allobaculum sp. is is the most abundant OTU
in the *0401 mice, while Barnesiella sp. occurs with highest frequency in *0402 mice (P,0.05). (b) Similar taxa distributions are observed at phyla level
with Bacteroidetes: Firmicutes ratios more even in *0401 (n = 41) compared to *0402 (n = 45) mice. Data are presented as mean 6 S.E., *P,0.05,
**P,0.01, ***P,0.001. (c) Correspondence analysis plot shows the degree of correlation between specific OTUs and mice genotype. The red vectors
point to the center of gravity of the samples where each OTU mostly occurs. The distance between the tip of the vector and the samples (dots) give
an indication of the probability of OTU content in each sample. Green and black dots represent *0401 and *0402 fecal samples respectively.
Percentages in parentheses in the CA plots describe the amount of variation explained by each axis.
doi:10.1371/journal.pone.0036095.g003
Gut Microbiome Identifies Arthritis Susceptibility
PLoS ONE | www.plosone.org 4 April 2012 | Volume 7 | Issue 4 | e36095
*0401 mice showed a significant increase in gut permeability in
arthritic (n = 5) mice compared to naı̈ve (n = 5) mice (P,0.0001,
Figure 5a). To determine if the host genotype and gut flora may
determine gut permeability, naı̈ve male and female *0401 and
*0402 mice were kept on a similar diet, cage bedding and room.
Our data showed that there is a basal level of intestinal
permeability which is significantly higher in *0401 mice as
compared to *0402 mice and that it is age and sex-dependent in
susceptible mice (Figure 5b). There was no difference in gut
permeability between sexes at a young age (,4 months); however,
as the *0401 mice aged, females (.4 months age) showed an
increase in gut permeability as compared to the younger group,
P,0.04 and older *0402 females (P,0.03). Resistant mice did not
show any significant changes in gut permeability with age or sex
(n = 5–8 in each group).
To determine if gut microbial composition is also associated
with a different gut immune profile, we tested the jejuna of naı̈ve
mice for expression of cytokine and chemokine transcripts
involved in the Th17 regulatory network by rtPCR (Figure 6 a–
e, Figure S1, File S1). Susceptible *0401 females showed a distinct
cytokine and chemokine profile as compared to males that was
characterized with a significant increase in IL-23a and IFNc along
with a decrease in the regulatory cytokines IL-4, IL-22 and
CCL20. Similarly, *0401 females showed more than 3 fold
increased gene transcripts for Th17 cytokines IL-17, IL-23, IL-6
and Th1 cytokines IFNc, Stat 4 and TBX21 while *0402 females
had several fold increase in genes regulating Th2 cytokines and
regulatory networks like ICOS, GATA3 and IL-4. *0401 male
mice did not show an increase in transcripts for TH17 encoding
genes compared to *0402 mice.
Next we determined if the relative abundance of the OTUs
showing sex, age and strain differences in the transgenic mice were
associated with specific cytokine/chemokine transcripts in jejuna
(n = 12, 3 mice from each group, *0401 and *0402 males and
females). Spearman correlation tests showed that Bifidobacterium
species were negatively correlated to IL-17a (P = 0.06) and TBx21
(P = 0.004) transcript levels, while Parabacteroides species were
negatively correlated to TBx21 (P = 0.012) (Table S1). The relative
abundance of Allobaculum species were negatively correlated to
CCL22 (P = 0.017) and IL-21 (P = 0.06). Since cytokines transcripts were studied in only six mice/strain, these results need to be
interpreted with caution.
Discussion
Interaction between genetic and environmental factors is
required for predisposition to develop RA. The presence of
bacterial DNA of gut-residing commensals in synovial fluid [3] has
led to the hypothesis that certain mucosal bacteria may have a role
in the susceptibility to develop arthritis. By taking advantage of
transgenic mice that express the RA-susceptible *0401 transgene
or the RA-resistant *0402 transgene, we have shown that genetic
factors, along with sex background and disruption of gut
microbiome may influence susceptibility and/or resistance toward
developing arthritis in humanized mice. Our data on gut
microbiome in genetically resistant mice is consistent with previous
reports of age and sex based differences in the fecal microbiomes
of healthy individuals [23,24]. Resistant transgenic female mice,
whose microbiomes were more similar, drove differences in
microbiome structure between the two strains. This suggests that
host genotype, rather than sex background, is a major regulator of
gut microbial composition, an observation consistent with a recent
report in inbred lines of mice [25]. However, as different human
and murine models have shown, it is under debate whether the
T
a b le 1
.
Id e n ti ty o f th e m ai n O
T
U
S’
s d ri vi n g st ra in ,
ag e an d g e n d e r d if fe re n ce s am o n g m ic e fe ca l m ic ro b io m e s. P
h y lo g e n e ti c A
ff il ia ti o n C
lo se st m a tc h in th e R
D
P
e n v ir o n m e n ta l se q u e n ce d a ta b a se a n d re fs e q _
rn a %
ID
C
lo se st m a tc h in th e n r/ n t N
C
B
I
d a ta b a se %
ID
C
lo se st u n cu lt u re d m a tc h in th e n r/ n t N
C
B
I
d a ta b a se %
ID
N
o .
o f ti m e s d e te ct e d in th e d a ta se t A
ct in o b ac te ri a B
if id o b a ct er iu m p se u d o lo n g u m su b sp .
g lo b o su m st ra in JC
M
5
8
2
0
9
9
B
if id o b a ct er iu m p se u d o lo n g u m su b sp .
g lo b o su m st ra in 0
2
-2
1
0
0
U
n cu lt u re d B
if id o b ac te ri u m sp .
cl o n e P
P
1
8
7
-b 1
5
1
0
0
3
6
,2
6
1
N
R
_
0
4
3
4
4
1
.1
A
Y
1
6
6
5
1
5
.1
G
U
9
0
2
7
5
4
.1
B
ac te ro id e te s B
a rn es ie lla
in te st in ih o m in is Y
IT
1
1
8
6
0
st ra in Y
IT
8
4
G
ra m -n e g at iv e b ac te ri u m cL 1
0
-2
b -4
8
8
U
n cu lt u re d b ac te ri u m cl o n e R
M
A
M
0
3
9
1
9
9
3
0
,8
9
1
N
R
_
0
4
1
6
6
8
.1
A
Y
2
3
9
4
6
9
.1
H
Q
3
1
9
3
2
1
.1
B
ac te ro id e te s B
a rn es ie lla
vi sc er ic o la D
SM
1
8
1
7
7
st ra in JC
M
1
3
6
6
0
8
2
G
ra m -n e g at iv e b ac te ri u m cL 1
0
-2
b -4
8
7
U
n cu lt u re d b ac te ri u m g e n e 9
5
3
5
,5
8
5
N
R
_
0
4
1
5
0
8
.1
A
Y
2
3
9
4
6
9
.1
A
B
6
2
6
9
2
7
.1
B
ac te ro id e te s P
a ra b a ct er o id es d is ta so n is st ra in JC
M
5
8
2
5
8
3
G
ra m -n e g at iv e b ac te ri u m cT P
Y
-1
3
8
9
U
n cu lt u re d b ac te ri u m cl o n e 1
6
sa w 3
8
-
2
d 0
6
.q 1
k 9
9
3
5
,2
3
4
N
R
_
0
4
1
3
4
2
.1
A
Y
2
3
9
4
6
1
.1
EF
6
0
4
5
9
3
.1
Fi rm ic u te s A
llo
b a cu lu m st er co ri ca n is D
SM
1
3
6
3
3
8
4
B
ac te ri u m H
B
N
D
8
7
U
n cu lt u re d b ac te ri u m cl o n e R
-9
0
8
5
8
9
3
4
,0
4
2
N
R
_
0
4
2
1
1
0
.1
A
Y
4
9
8
7
4
8
.1
FJ
8
8
1
0
7
7
.1
P
h yl o ty p e id e n ti ty w as as si g n e d u si n g th e R
D
P
(R
ib o so m al D
at ab as e P
ro je ct )
cl as si fi e r at 8
0
%
B
ay e si an b o o ts tr ap cu to ff fr o m co m p ar is o n s to th e e n vi ro n m e n ta l su rv e y se q u e n ce d at ab as e .
R
N
A
e n tr ie s fr o m N
C
B
I’s
R
e fe re n ce Se q u e n ce p ro je ct (r e fs e q _
rn a) an d n o lo n g e r ‘‘n
o n -r e d u n d an t’
n u cl e o ti d e co lle
ct io n o f th e N
C
B
I
w e re al so u se d to as si g n p h yl o ty p e .
d o i:1
0
.1
3
7
1
/j o u rn al .p o n e .0
0
3
6
0
9
5
.t 0
0
1
Gut Microbiome Identifies Arthritis Susceptibility
PLoS ONE | www.plosone.org 5 April 2012 | Volume 7 | Issue 4 | e36095
maintenance of adaptive immune mechanisms are mainly applied
top-bottom (host-controlled) or bottom-up (driven by the gut
microbiome), with gut bacterial communities acting as puppets or
masters of the immune system [4,26,27,28]. In this respect, the
findings presented herein imply that the selection of a different T
cell repertoire by two distinct HLA transgenes [29] modulates gut
bacterial communities. This is consistent with studies that show
increases in the incidence of autoimmune disorders driven by
genotype, in such studies, interactions with specific commensals,
harmless under immunocompetent conditions, could trigger
disease [30,31]. Conversely, studies with germ-free and specific
pathogen-free mice have shown that disruptions to gut bacteria
can promote increased levels of pro-inflammatory cytokine and
interlukin-17 producing Th-17 cells, even in tissues distant to the
gut [27,31,32], suggesting that global adaptive immune responses
are also controlled by gut bacteria. Our data showed a bias
towards TH1/TH17 cytokine expression with significant decrease
in cytokine gene transcripts required for negative regulation of
Th17 profile, like IL-4, IL-21 and IL-22, in *0401 females as
compared to *0401 males and *0402 females. Interestingly CCL20
and CCL22 which are required for the generation of regulatory
CD4 T cells and DCs [33,34,35], are reduced several fold in *0401
females as compared to *0401 males and *0402 females. A recent
study showed that decreased levels of CCL20 during aging, are
Figure 4. Sex based relative abundance of OTUs in the fecal microbiomes of *0401 and *0402 mice. (a) *0402 females (n = 24) show
significantly higher relative abundances of Bifidobacterium-Parabacteroides OTUs compared to males (n = 21), whose microbiomes present
significantly higher levels of Barnesiella viscericola. (b) Correspondence analysis plot displays sex-based correlation between OTUs in *0402 mice. (c)
Significantly higher relative abundances of Bifidobacterium sp. were observed in *0401 males (n = 22) compared to females (n = 19), (d) despite loss of
dynamic sex based differences in the fecal microbiomes of *0401 mice. Percentages in parentheses in the CA plots describe the amount of variation
explained by each axis. Data are presented as mean 6 S.E. *P,0.05, **P,0.01, ***P,0.001.
doi:10.1371/journal.pone.0036095.g004
Figure 5. Gut permeability. (a) Transgenic arthritic mice showed significantly higher gut permeability compared to naı̈ve mice (n = 5 each group).
(b) Intestinal permeability in naı̈ve *0401 and *0402 transgenic male and female mice at .4 and ,4 months of age. *0401F,4 mo vs .4 mo,
P,0.04; *0401F vs *0402F .4 mo, P,0.03 (n = 5–8 in groups).
doi:10.1371/journal.pone.0036095.g005
Gut Microbiome Identifies Arthritis Susceptibility
PLoS ONE | www.plosone.org 6 April 2012 | Volume 7 | Issue 4 | e36095
associated with isolated lymphoid dysfunction and mucosal
immunosenescence [35]. These studies along with the present
results that show a decreased CCL20 and loss of age differences in
gut microbiome of *0401 females may confer mucosal dysfunction
and immunosenescence. Arthritis-susceptible males however,
showed a significant increase in Th1/Th2 but not in Th17
cytokines compared to resistant males, along with significant
increases in genes for cd T cells, suggesting a role for these cells in
Th1/Th2 profile. Differences in chemokine gene transcripts
observed between *0401 and *0402 mice and their correlation
with microbiome profiles further supported our contention of
dysbiosis leading to altered mucosal immune function in
susceptible mice. However, these events could be contributing to
pathogenesis independtly or in combination.
The significantly higher evenness index and even Bacteroidetes/
Firmicutes ratios in *0401 as compared to *0402 mice, and their
specific associations with either genotype, may be a reflection of an
apparent dysbiosis phenomenon similar to that observed in other
disease conditions [36]. Our observations suggest that *0402 mice
maintain a homeostatic gut bacterial environment characterized
by the overrepresentation and/or absence of specific microbiome
structure. Specifically, members of Bacteroidetes and Actinobacteria occur twice as often as Firmicutes in *0402 mice as opposed
to the stable ratios observed in *0401 mice. This observation
implies that host genotype and environmental stimuli can cause
expansion and/or contraction of certain members of a core or
signature gut microbiome to modulate immunity. In the present
study, microbiome differences between arthritis-susceptible and
resistant mice are higher relative abundance of Bifidobacterium
pseudolongum and members of the Porphyromonadaceae family in the
latter that are positively correlated with regulatory cytokines.
These observations support the importance of these taxa in the
maintenance of a homeostatic gut microbiome. Bifidobacteria sp.
have been recognized for their probiotic and immuno-modulating
properties including down-regulating the expression of inflammatory pathways and enhancing gut barrier function [4,36,37,38].
Parabacteroides distasonis, a commensal detected in higher proportions in arthritis-resistant mice, has been reported to reduce
intestinal inflammation in murine models upon oral administration of its antigens [39] and is also involved in ‘‘educating’’ the
immune system towards the tolerance of commensal antigens by
enhancing Treg cell recognition mechanisms [40]. Thus in *0401
mice, particularly in females, reduced relative abundance of
Bifidobacterium sp. and commensals from the Porphyromonadaceae
family may lead to dysbiosis, enhanced pro-inflammatory
responses, and a subsequent skewed immune response. Interestingly, the proportions of Bifidobacterium are inversely proportional
or negatively correlated to the presence of Allobaculum sp.
(Clostridiales order). Segmented filamentous bacteria (SFB), also
from the Clostridiales, have been linked to immunosuppression
Figure 6. A)) Heat map showing expression levels of cytokines and chemokine transcripts in jejenum of *0401 and *0402 male and
female mice (n = 3 in each group). (b) Comparison of fold change in gene transcript levels between *0401 females and males, (c) *0402 females
and males, (d) females of each genotype and (e) males of each genotype. Results are given as fold-changes of mean copy-numbers relative to the
mean copy-numbers of the comparative group. *P,0.05 and **P,0.01 and more. Data points with 3 or more fold differences and significance of
more than P,0.05 are shown.
doi:10.1371/journal.pone.0036095.g006
Gut Microbiome Identifies Arthritis Susceptibility
PLoS ONE | www.plosone.org 7 April 2012 | Volume 7 | Issue 4 | e36095
and an increase of pro-inflammatory responses in arthritis driven
by Th-17 cell proliferation [41,42]. Although SFB were not
detected in this study, the gut microbiomes of *0401 mice were
characterized by a 7 fold increase in the relative abundance of
Allobaculum sp. compared to *0402 mice. Based on 16S rRNA gene
phylogenetic analysis, A. stercoricanis forms a branch closer to the
XVI Clostridial cluster constituted by Clostriudium innocum,
Streptococcus pleomorphus and several Eubacterium spp. [43]. In fact
C. innocum was the closest hit in RDP database (80% ID), which
could imply high phylogenetic concordance of our sequence to
members of this Clostridiales group. Consistent with our results,
Clostridium spp. have been reported to be enriched in immunecompromised subjects [44] and reductions in C. innocuum levels are
reported upon oral administration of Bifidobacterium spp. [45]. A
broader search in the non-redundant nucleotide NCBI database,
also related this Clostridium-like sequence to a bacterium isolated
from mice deficient in secretory antibodies (87% identity), in
which there was increased recognition of gastrointestinal tract flora
antigens by systemic antibodies and increased bacterial translocation [46]. Thus, these findings may indicate an association
between inflammation and the high abundance of this OTU in
*0401 mice.
Our data in humanized mice is also supported by a study in
early RA patients, in which lower levels of Bifidobacteria and
bacteria of the Bacteroides-Porphyromonas-Prevotella group were
observed in RA patients compared to non-arthritic patients [7].
Members of the Porphyromonadaceae family are common dwellers of
intestinal, oral and urogenital human and murine flora and have
been identified as opportunistic commensals potentially pathogenic after immune disruption [47]. Herein, the relative abundance of
members of the Porphyromonadaceae family (Barnesiella and Parabacteroides spp.) were significantly reduced in susceptible mice whenever
the Clostridia-like bacterium became abundant. This observation
implies that the Clostridia-like bacterium may induce disruption of
normal commensals that are non-pathogenic under immunocompetent conditions. An increase in the gut permeability has been
suggested to play a role in pathogenesis of arthritis [48]. Thus, the
differences in microbiome and gut permeability seen in the *0401
and *0402 raises the possibility that, under certain conditions,
disease causing bacteria like Clostridia, and other pathobionts or
symbionts could produce translocation and a systemic immune
response resulting in arthritis in those with a genetic susceptibility.
An interesting observation is the dynamic different microbiome
structures based on age and/or sex background, driven by specific
bacterial groups in *0402 mice. In contrast, this kind of microbial
axis dynamism is completely lost in susceptible mice. This suggests
the execution of specific, age/sex-based, immune regulation
mechanisms in resistant mice. In this case, only females may have
benefited from relative high abundance of Bifidobacteria and
Parabacteroides. However, both *0402 males and females, regardless
of age, seem to benefit from an absence of the Clostridia-like
bacterium. This observation, along with the fact that in *0401
mice males may be taking advantage of significantly higher
abundance of Bifidobacteria compared to females, raises the
hypothesis that it is the loss of bacterial dynamism that is
associated with disease susceptibility, particularly in females. A
similar situation can be envisaged in human models in which
beneficial microbiome may act as a modulator of proper immune
response in the absence of host-genetic immune-regulators, and
presence of pro-inflammatory bacteria potentially triggering
disease. Although, specific molecular mechanisms remain largely
unexplored, together, these results suggest that susceptibility to RA
could be triggered by gut dysbiosis in genetically susceptible
individuals. In turn, the onset of resistance may be characterized
by a more dynamic microbiome, whose members expand and/or
contract to provide sex and/or age based competent immune
function, especially, in individuals that are not prone to develop
RA. The hypotheses proposed herein (Figure 7), could be tested in
future studies through experimentation with germ-free and SPF
mice using various ways to manipulate gut microbiome and
measure its impact in triggering disease. These approaches may
include administration of probiotics which have been shown to
alter intestinal microbiota and immune response and suppress CIA
[49,50]. Additionally, this model points a way forward to further
probe gut microbial communities in various disease conditions,
which would allow us to identify novel biomarkers and develop
preclinical models to manipulate them for therapeutics.
Materials and Methods
Transgenic mice
The generation of DRB1*0401and DRB1*0402 transgenic (Tg)
mice has been described previously [20,29]. Abu.DRB1*0401 and
Abu.DRB1*0402 mice were mated with MHCIID/D (AEu) mice
[51] to generate AEu.DRB1*0401 and AEu.DRB1*0402 mice.
Mice of both sexes (8–12 weeks of age) used in this study were bred
and maintained in the pathogen-free Immunogenetics Mouse
Colony at the Mayo Clinic, Rochester, MN in accordance with the
Institutional Animal care and use Committee (IACUC). For
convenience, DRB1*0402 mice will be referred to as *0402, and
DRB1*0401 mice as *0401.
The expression of DR on PBLs of transgenic mice was analyzed
by flow cytometry using mAbs L227 (anti-DR) conjugated
antibodies to characterize transgene positive mice.
16 s rDNA analysis of Mice Fecal microbiome
Microbial DNA was extracted using the MoBio UltraClean Soil
Kit (Mo Bio Laboratories Inc., Carlsbad, CA, USA) with a beadbeating step from fecal material of a single mouse. The V1–V3
region of the 16S ribosomal RNA gene was amplified by
Polymerase chain reaction (22 cycles of 94uC (30 s), 48uC (30 s),
72uC (2 min)) using primers 27f (CGTATCGCCTCCCTCGCGCCATCAG-AGAGTTTGATYMTGGCTCAG; corresponding to nucleotides 8–27 of the Escherichia coli 16 s rRNA
gene) and 534r (CTATGCGCCTTGCCAGCCCGCTCAG[MID tag 1–15]-ATTACCGCGGCTGCTGGCA; corresponding to nucleotides 514–534 of the E. coli 16 s rRNA gene). The
amplicons were subjected to pyrosequencing using 454 FLXTitanium technologies at the UIUC KECK. The resulting
sequences were processed using a combination of tools from
Mothur [52] and custom Perl scripts. Preliminary quality control
steps included the removal of sequences shorter than 400 nt with
homopolymers longer than 6 nucleotides and all reads containing
ambiguous base calls or incorrect primer sequences. Sequences
were aligned against the silva database and then trimmed so
subsequent analyses were constrained to the same portion of the
16S rDNA. Potentially chimeric sequences were detected using
chimera slayer (http://www.mothur.org/wiki/chimera.slayer/)
and removed. The remaining reads were pre-clustered to remove
sequences that are likely to have derived from sequencing errors
(http://www.mothur.org/wiki/Pre.cluster) and then clustered
using Mothur’s average algorithm. Taxonomic classification of
each OTU (clustered at 97% sequence similarity) was obtained by
Blastn alignments to NCBI RNA reference sequence and nonredundant nucleotide databases and with the Ribosomal Database
project (RDP) multiclassifier at 80% Bayesian bootstrap cutoff
from comparisons to the environmental survey sequence database.
All new data were deposited in the sequence read archive (http://
Gut Microbiome Identifies Arthritis Susceptibility
PLoS ONE | www.plosone.org 8 April 2012 | Volume 7 | Issue 4 | e36095
www.ncbi.nlm.nih.gov/sra/), accession number SRA043819.
NMDS plots and SIMPER analyses were constructed based on
Bray-Curtis distance metrics using Primer-E from normalized
OTU-abundance data. Correspondence analysis was plotted using
the ca package and bipartite network analysis from the R project
statistical software [53].
Induction and evaluation of CIA
To induce CIA, 8–12 weeks old *0401 transgenic mice and
negative littermates were immunized with 100 mg of type II
collagen (CII) (Chondrex Inc.) emulsified 1:1 with complete
Freunds’ adjuvant H37Ra (CFA, Difco Laboratories, Detroit, MI)
intradermally at the base of the tail as previously described for CIA
protocol [18]. Mice were monitored for the onset and progression
of CIA from 3–12 weeks postimmunization. The arthritic severity
of mice was evaluated as described previously with a grading
system for each paw from 0–3 as described [20]. Mice with a score
of 2 or more were used as arthritic mice.
Intestinal permeability
As gut permeability may be diet dependent, all transgenic mice
were kept on standard diet. Changes in intestinal permeability
were determined using 4-KDa FITC-labeled dextran. Mice were
deprived of food for 3 hours, then gavaged with FITC–labeled
dextran (0.6 mg/g body weight). Three hours later, mice were
bled and serum collected. FITC-dextran content of the sera was
determined by using a microplate reader with an excitation of
490 nm and emission detection at 525 nm. Gut permeability was
tested in age and sex matched arthritic (7 weeks post-immunization with CII) and naı̈ve (non-immunized) mice.
RNA isolation and Real- time Polymerase chain reaction
Jejenum of naı̈ve transgenic mice were isolated from 4 months
old mice. After a midline celiotomy, the intestine was flushed with
cold (4uC) phosphate buffered saline (PBS) to remove intraluminal
content and jejunal segments were placed in RNA stabilization
buffer (Qiagen). Total RNA from the isolated tissue was extracted
using the RNeasy kit and protocol (Qiagen). cDNA was prepared
using RT2 First Strand Kit cDNA Synthesis Kit and Primer Mixes
(SABiosciences). The quantification of gene expression related to
the Th17 Regulatory Network was performed using the RT2
Profiler PCR Array PAMM-0773 (SABiosciences) and the
HT7900 Fast Real-Time PCR System (ABI). Product amplification was measured and analyzed according to the manufacturer’s
instructions.
Statistical analyses
NMMDS (Non-metric Multi-dimensional scale) plots, SIMPER
(Percentages-species contribution) analyses and ANOSIM (Analysis of similarities) were constructed based on Bray-Curtis distance
metrics using Primer-E (PRIMER 5, version 5.2.7 (Primer-E Ltd.,
Plymouth, United Kingdom. Clark, 2005) from square root
transformed OTU-abundance data. The ANOSIM procedure
generates a test statistic, R, calculated as: R = (rB2rW)/[1/
4n(n21)], where n is the total number of samples, rB is the
average of rank similarities arising from all pairs of replicates
between different mice fecal samples groups (Strain, sex or age),
and rW is defined as the average of all rank similarities among
replicates within mice fecal samples groups. An R value of 1
indicates complete dissimilarity between groups; an R of 0
indicates a high degree of community similarity among groups.
Relative abundance of each OTU calculated as number of
reads of a taxon/total number of reads in a sample was used to
Figure 7. Role of the gut microbiome in susceptibility to arthritis. HLA genotype may shape the gut microbiome in an individual. The
DRB1*0401 gene associated with predisposition to Rheumatoid arthritis, may induce a lower gut Bacteroidetes: Firmicutes ratio compared to that
shaped by the DRB1*0402 gene, known to be associated with resistance to arthritis. This model suggests that a dysbiotic or arthritogenic gut
microbiome may be dominated by a Clostridia-like bacterium (Firmicutes phylum) in susceptible individuals, while competent/tolerant immune
responses are enhanced by increased abundances of Bifidobacterium spp. in resistance to RA. The gut microbiome has a crucial influence on
maintaining homeostasis of the gut immune system by predicting pro-inflammatory (TH1/Th17) or anti-inflammatory (TH1/TH2) responses.
Environmental triggers like smoking, diet and infectious agents along with sex hormones and age-dependent changes, may further modulate the gut
immune system and enhance pro-inflammatory conditions in genetically susceptible individuals. In synthesis, an overall/systemic immune response
generated by innate immune cells, may be originated at gut level and this response may be regulated by the gut microbiome via HLA genotype. This
chain of events may determine the onset of autoimmune diseases like rheumatoid arthritis.
doi:10.1371/journal.pone.0036095.g007
Gut Microbiome Identifies Arthritis Susceptibility
PLoS ONE | www.plosone.org 9 April 2012 | Volume 7 | Issue 4 | e36095
construct correspondence analysis plots using the ca package (50)
from the R project statistical software. To assess significant
differences in relative abundances of specific OTUs between
strain, sex or age groups non-parametric Mann-Whitney U/
Wilcoxon rank sum tests were conducted using the R project
statistical tool. All data for the non-parametric tests were checked
for homogeneity of error variances using the Brown-Forsythe test.
Statistical significance was set to P,0.05. Non-parametric
Spearman correlation coefficients were determined using the
PROC COR procedure from the SAS software platform (SAS
version 9.1.3; SAS Institute, Cary NC), appropriate significance
level, error correction and power for all tests were determined
using the pwr package from the R project statistical software.
Significance difference in expression of gene transcripts for Th17
cytokines and regulating genes were analyzed by online tool
available from the manufacturer for the PAMM0733 array
(SABiosciences) and is reported as significant fold change
difference of p,0.05 between groups. Difference in gut permeability between ages and sex was calculated by student’s T test with
significance set at p,0.05.
Supporting Information
Figure S1 Heat map showing expression levels of
cytokines and chemokine transcripts in jejenum of
*0401 and *0402 male and female mice.
(TIF)
Table S1 Correlation coefficients between OUT’s and
cytokine/chemokine transcript levels.
(DOCX)
File S1 Differences in Th17 regulatory network transcripts in *0401 and *0402 male and female mice.
(XML)
Acknowledgments
We thank Julie Hanson and her staff in the Mayo Immunogenetics mouse
colony for breeding and taking care of the mice. We thank Michele Smart
for tissue typing of transgenic mice.
Author Contributions
Conceived and designed the experiments: VT JAM BAW. Performed the
experiments: AG DL EVM CJY MEB. Analyzed the data: AG CJY VT
BAW. Contributed reagents/materials/analysis tools: VT BAW. Wrote the
paper: AG VT CJY JAM BAW.
References
1. Klareskog L, Padyukov L, Lorentzen J, Alfredsson L (2006) Mechanisms of
disease: Genetic susceptibility and environmental triggers in the development of
rheumatoid arthritis. Nat Clin Pract Rheumatol 2: 425–433.
2. Moen K, Brun JG, Valen M, Skartveit L, Eribe EK, et al. (2006) Synovial
inflammation in active rheumatoid arthritis and psoriatic arthritis facilitates
trapping of a variety of oral bacterial DNAs. Clin Exp Rheumatol 24: 656–663.
3. Kempsell KE, Cox CJ, Hurle M, Wong A, Wilkie S, et al. (2000) Reverse
transcriptase-PCR analysis of bacterial rRNA for detection and characterization
of bacterial species in arthritis synovial tissue. Infect Immun 68: 6012–6026.
4. Round JL, O’Connell RM, Mazmanian SK (2010) Coordination of tolerogenic
immune responses by the commensal microbiota. J Autoimmun 34: J220–225.
5. Chow J, Lee SM, Shen Y, Khosravi A, Mazmanian SK. Host-bacterial symbiosis in health and disease. Adv Immunol 107: 243–274.
6. Amirzargar A, Mytilineos J, Yousefipour A, Farjadian S, Scherer S, et al. (1998)
HLA class II (DRB1, DQA1 and DQB1) associated genetic susceptibility in
Iranian multiple sclerosis (MS) patients. Eur J Immunogenet 25: 297–301.
7. Vaahtovuo J, Munukka E, Korkeamaki M, Luukkainen R, Toivanen P (2008)
Fecal microbiota in early rheumatoid arthritis. J Rheumatol 35: 1500–1505.
8. Finegold SMSV, Mathisen GE, In: editor. New York: Academic Press; 1983:3–
31 (1983) Normal indigenous intestinal flora. In: Hentges DJ, ed. Human
intestinal microflora in health and disease. New York: Academic Press. pp 3–31.
9. Langendijk PS, Schut F, Jansen GJ, Raangs GC, Kamphuis GR, et al. (1995)
Quantitative fluorescence in situ hybridization of Bifidobacterium spp. with
genus-specific 16S rRNA-targeted probes and its application in fecal samples.
Appl Environ Microbiol 61: 3069–3075.
10. Ventura M, Turroni F, Canchaya C, Vaughan EE, O’Toole PW, et al. (2009)
Microbial diversity in the human intestine and novel insights from metagenomics. Front Biosci 14: 3214–3221.
11. Liu YJ, Soumelis V, Watanabe N, Ito T, Wang YH, et al. (2007) TSLP: an
epithelial cell cytokine that regulates T cell differentiation by conditioning
dendritic cell maturation. Annu Rev Immunol 25: 193–219.
12. Moore KW, de Waal Malefyt R, Coffman RL, O’Garra A (2001) Interleukin-10
and the interleukin-10 receptor. Annu Rev Immunol 19: 683–765.
13. Saenz SA, Taylor BC, Artis D (2008) Welcome to the neighborhood: epithelial
cell-derived cytokines license innate and adaptive immune responses at mucosal
sites. Immunol Rev 226: 172–190.
14. Baba N, Samson S, Bourdet-Sicard R, Rubio M, Sarfati M (2008) Commensal
bacteria trigger a full dendritic cell maturation program that promotes the
expansion of non-Tr1 suppressor T cells. J Leukoc Biol 84: 468–476.
15. Christensen HR, Frokiaer H, Pestka JJ (2002) Lactobacilli differentially
modulate expression of cytokines and maturation surface markers in murine
dendritic cells. J Immunol 168: 171–178.
16. Kelsall BL, Leon F (2005) Involvement of intestinal dendritic cells in oral
tolerance, immunity to pathogens, and inflammatory bowel disease. Immunol
Rev 206: 132–148.
17. Taneja V, Behrens M, Basal E, Sparks J, Griffiths MM, et al. (2008) Delineating
the role of the HLA-DR4 ‘‘shared epitope’’ in susceptibility versus resistance to
develop arthritis. J Immunol 181: 2869–2877.
18. Taneja V, Behrens M, Mangalam A, Griffiths MM, Luthra HS, et al. (2007)
New humanized HLA-DR4-transgenic mice that mimic the sex bias of
rheumatoid arthritis. Arthritis Rheum 56: 69–78.
19. Geluk A, Taneja V, van Meijgaarden KE, Zanelli E, Abou-Zeid C, et al. (1998)
Identification of HLA class II-restricted determinants of Mycobacterium
tuberculosis-derived proteins by using HLA-transgenic, class II-deficient mice.
Proc Natl Acad Sci U S A 95: 10797–10802.
20. Behrens M, Trejo T, Luthra H, Griffiths M, David CS, et al. (2010) Mechanism
by which HLA-DR4 regulates sex-bias of arthritis in humanized mice.
J Autoimmun.
21. Vaahtovuo J, Toivanen P, Eerola E (2003) Bacterial composition of murine fecal
microflora is indigenous and genetically guided. FEMS Microbiol Ecol 44:
131–136.
22. De Palma G, Capilla A, Nadal I, Nova E, Pozo T, et al. (2009) Interplay
Between Human Leukocyte Antigen Genes and the Microbial Colonization
Process of the Newborn Intestine. Curr Issues Mol Biol 12: 1–10.
23. Mariat D, Firmesse O, Levenez F, Guimaraes V, Sokol H, et al. (2009) The
Firmicutes/Bacteroidetes ratio of the human microbiota changes with age. BMC
Microbiol 9: 123.
24. Mueller S, Saunier K, Hanisch C, Norin E, Alm L, et al. (2006) Differences in
fecal microbiota in different European study populations in relation to age,
gender, and country: a cross-sectional study. Appl Environ Microbiol 72:
1027–1033.
25. Kovacs A, Ben-Jacob N, Tayem H, Halperin E, Iraqi FA, et al. (2011) Genotype
is a stronger determinant than sex of the mouse gut microbiota. Microb Ecol 61:
423–428.
26. Chervonsky AV (2010) Influence of microbial environment on autoimmunity.
Nat Immunol 11: 28–35.
27. Lee YK, Mazmanian SK (2010) Has the microbiota played a critical role in the
evolution of the adaptive immune system? Science 330: 1768–1773.
28. Ley RE, Peterson DA, Gordon JI (2006) Ecological and evolutionary forces
shaping microbial diversity in the human intestine. Cell 124: 837–848.
29. Taneja V, Taneja N, Behrens M, Pan S, Trejo T, et al. (2003) HLADRB1*0402 (DW10) transgene protects collagen-induced arthritis-susceptible
H2Aq and DRB1*0401 (DW4) transgenic mice from arthritis. J Immunol 171:
4431–4438.
30. Wen L, Ley RE, Volchkov PY, Stranges PB, Avanesyan L, et al. (2008) Innate
immunity and intestinal microbiota in the development of Type 1 diabetes.
Nature 455: 1109–1113.
31. Ichinohe T, Pang IK, Kumamoto Y, Peaper DR, Ho JH, et al. (2011)
Microbiota regulates immune defense against respiratory tract influenza A virus
infection. Proc Natl Acad Sci U S A 108: 5354–5359.
32. Zaph C (2010) Which species are in your feces? J Clin Invest 120: 4182–4185.
33. Ito T, Carson WFt, Cavassani KA, Connett JM, Kunkel SL. CCR6 as a mediator of immunity in the lung and gut. Exp Cell Res 317: 613–619.
34. Lugering A, Floer M, Westphal S, Maaser C, Spahn TW, et al. (2005) Absence
of CCR6 inhibits CD4+ regulatory T-cell development and M-cell formation
inside Peyer’s patches. Am J Pathol 166: 1647–1654.
Gut Microbiome Identifies Arthritis Susceptibility
PLoS ONE | www.plosone.org 10 April 2012 | Volume 7 | Issue 4 | e36095
35. McDonald KG, Leach MR, Huang C, Wang C, Newberry RD. Aging impacts
isolated lymphoid follicle development and function. Immun Ageing 8: 1.
36. Mazmanian SK, Round JL, Kasper DL (2008) A microbial symbiosis factor
prevents intestinal inflammatory disease. Nature 453: 620–625.
37. Matsumoto M, Kurihara S, Kibe R, Ashida H, Benno Y (2011) Longevity in
mice is promoted by probiotic-induced suppression of colonic senescence
dependent on upregulation of gut bacterial polyamine production. PLoS One 6:
e23652.
38. Cerf-Bensussan N, Gaboriau-Routhiau V (2010) The immune system and the
gut microbiota: friends or foes? Nat Rev Immunol 10: 735–744.
39. Kverka M, Zakostelska Z, Klimesova K, Sokol D, Hudcovic T, et al. (2011) Oral
administration of Parabacteroides distasonis antigens attenuates experimental
murine colitis through modulation of immunity and microbiota composition.
Clin Exp Immunol 163: 250–259.
40. Lathrop SK, Bloom SM, Rao SM, Nutsch K, Lio CW, et al. (2011) Peripheral
education of the immune system by colonic commensal microbiota. Nature 478:
250–254.
41. Caselli M, Holton J, Boldrini P, Vaira D, Calo G (2010) Morphology of
segmented filamentous bacteria and their patterns of contact with the follicleassociated epithelium of the mouse terminal ileum: implications for the
relationship with the immune system. Gut Microbes 1: 367–372.
42. Fuentes S, Egert M, Jimenez-Valera M, Monteoliva-Sanchez M, Ruiz-Bravo A,
et al. (2008) A strain of Lactobacillus plantarum affects segmented filamentous
bacteria in the intestine of immunosuppressed mice. FEMS Microbiol Ecol 63:
65–72.
43. Greetham HL, Gibson GR, Giffard C, Hippe H, Merkhoffer B, et al. (2004)
Allobaculum stercoricanis gen. nov., sp. nov., isolated from canine feces.
Anaerobe 10: 301–307.
44. Mangin I, Bonnet R, Seksik P, Rigottier-Gois L, Sutren M, et al. (2004)
Molecular inventory of faecal microflora in patients with Crohn’s disease. FEMS
Microbiol Ecol 50: 25–36.
45. Benno Y, Mitsuoka T (1992) Impact of Bifidobacterium longum on human fecal
microflora. Microbiol Immunol 36: 683–694.
46. Sait LC, Galic M, Price JD, Simpfendorfer KR, Diavatopoulos DA, et al. (2007)
Secretory antibodies reduce systemic antibody responses against the gastrointestinal commensal flora. Int Immunol 19: 257–265.
47. Alauzet C, Mory F, Carlier JP, Marchandin H, Jumas-Bilak E, et al. (2007)
Prevotella nanceiensis sp. nov., isolated from human clinical samples. Int J Syst
Evol Microbiol 57: 2216–2220.
48. Picco P, Gattorno M, Marchese N, Vignola S, Sormani MP, et al. (2000)
Increased gut permeability in juvenile chronic arthritides. A multivariate analysis
of the diagnostic parameters. Clin Exp Rheumatol 18: 773–778.
49. Ng SC, Hart AL, Kamm MA, Stagg AJ, Knight SC (2009) Mechanisms of
action of probiotics: recent advances. Inflamm Bowel Dis 15: 300–310.
50. So JS, Kwon HK, Lee CG, Yi HJ, Park JA, et al. (2008) Lactobacillus casei
suppresses experimental arthritis by down-regulating T helper 1 effector
functions. Mol Immunol 45: 2690–2699.
51. Kouskoff V, Fehling HJ, Lemeur M, Benoist C, Mathis D (1993) A vector
driving the expression of foreign cDNAs in the MHC class II-positive cells of
transgenic mice. J Immunol Methods 166: 287–291.
52. Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann M, et al. (2009)
Introducing mothur: open-source, platform-independent, community-supported
software for describing and comparing microbial communities. Appl Environ
Microbiol 75: 7537–7541.
53. Nenadic O, Greenacre M (2007) Correspondence analysis in R, with two and
three dimensional Graphics: the ca package. J Statis Soft 20: 1–13.
Gut Microbiome Identifies Arthritis Susceptibility
PLoS ONE | www.plosone.org 11 April 2012 | Volume 7 | Issue 4 | e36095

PDF Document reader online

This website is focused on providing document in readable format, online without need to install any type of software on your computer. If you are using thin client, or are not allowed to install document reader of particular type, this application may come in hand for you. Simply upload your document, and Docureader.top will transform it into readable format in a few seconds. Why choose Docureader.top?

  1. Unlimited sharing - you can upload document of any size. If we are able to convert it into readable format, you have it here - saved for later or immediate reading
  2. Cross-platform - no compromised when reading your document. We support most of modern browers without the need of installing any of external plugins. If your device can oper a browser - then you can read any document on it
  3. Simple uploading - no need to register. Just enter your email, title of document and select the file, we do the rest. Once the document is ready for you, you will receive automatic email from us.

Previous 10

Next 10