- Email: [email protected]
SIRT1 as a potential biomarker of response to treatment with glatiramer acetate in multiple sclerosis
SIRT1 as a potential biomarker of response to treatment with glatiramer acetate in multiple sclerosis Daniel Hewes, Alexandru Tatomir, Adam M. Kruszewski, Gautam Rao, Cosmin A. Tegla, Jonathan Ciriello, Vingh Nguyen, Walter Royal III, Christopher Bever, Violeta Rus, Horea Rus PII: DOI: Reference:
S0014-4800(17)30031-X doi:10.1016/j.yexmp.2017.01.014 YEXMP 4007
To appear in:
Experimental and Molecular Pathology
Received date: Accepted date:
17 January 2017 17 January 2017
Please cite this article as: Hewes, Daniel, Tatomir, Alexandru, Kruszewski, Adam M., Rao, Gautam, Tegla, Cosmin A., Ciriello, Jonathan, Nguyen, Vingh, Royal III, Walter, Bever, Christopher, Rus, Violeta, Rus, Horea, SIRT1 as a potential biomarker of response to treatment with glatiramer acetate in multiple sclerosis, Experimental and Molecular Pathology (2017), doi:10.1016/j.yexmp.2017.01.014
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT SIRT1 as a potential biomarker of response to treatment with Glatiramer Acetate in multiple sclerosis
PT
Daniel Hewes, Alexandru Tatomir, Adam M. Kruszewski, Gautam Rao, Cosmin A.
RI
Tegla§, Jonathan Ciriello, Vingh Nguyen‡, Walter Royal III*†, Christopher Bever*†,
SC
Violeta Rus‡, Horea Rus§†.
NU
Department of Neurology, University of Maryland, School of Medicine, §Research
MA
Service, Veterans Administration Maryland Health Care System, ‡ Department of Medicine, Division of Rheumatology and Clinical Immunology, University of Maryland,
D
School of Medicine, †Veterans Administration Multiple Sclerosis Center of Excellence,
TE
Baltimore, MD, USA
AC CE P
Running title: SIRT1 and multiple sclerosis Address correspondence to: Horea Rus M.D., Ph.D., University of Maryland School of Medicine, Department of Neurology, 655 W Baltimore St, BRB 12-033, Baltimore, MD 21201. Tel: (410) 706-3170; FAX: (410) 706-0186; E-mail: [email protected] D. Hewes, A. Tatomir and A. M. Kruszewski contributed equally to this manuscript.
Key words: SIRT1, Multiple Sclerosis, Glatiramer Acetate, Biomarker, Histone H3, Methylation
1
ACCEPTED MANUSCRIPT Abstract SIRT1, a NAD dependent histone and protein deacetylase, is a member of the histone
PT
deacetylase class III family. We previously showed that SIRT1 mRNA expression is
RI
significantly lower in peripheral blood mononuclear cells (PBMCs) of multiple sclerosis
SC
(MS) patients during relapses than in stable patients. We have now investigated SIRT1 as a possible biomarker to predict relapse as well as responsiveness to glatiramer
NU
acetate (GA) treatment in relapsing-remitting MS (RRMS) patients. Over the course of
MA
2 years, a cohort of 15 GA-treated RRMS patients were clinically monitored using the Expanded Disability Status Scale and assessed for MS relapses. Blood samples
D
collected from MS patients were analyzed for levels of SIRT1 and histone H3 lysine 9
TE
(H3K9) acetylation and dimethylation. During relapses, MS patients had a lower expression of SIRT1 mRNA than did stable MS patients. In addition, there was a
AC CE P
significant decrease in H3K9 dimethylation (H3K9me2) during relapses in MS patients when compared to stable patients (p=0.01). Responders to GA treatment had significantly higher SIRT1 mRNA (p=0.01) and H3K9me2 levels than did nonresponders (p=0.018). Receiver operating characteristic analysis was used to assess the predictive power of SIRT1 and H3K9me2 as putative biomarkers: for SIRT1 mRNA, the predictive value for responsiveness to GA treatment was 70% (p=0.04) and for H3K9me2 was 71% (p=0.03). Our data suggest that SIRT1 and H3K9me2 could serve as potential biomarkers for evaluating patients’ responsiveness to GA therapy in order to help guide treatment decisions in MS.
2
ACCEPTED MANUSCRIPT Introduction Multiple sclerosis (MS) is a demyelinating disease characterized by chronic
PT
inflammation of the central nervous system (CNS), in which many factors (genetic and
RI
environmental) act in concert to influence disease susceptibility and progression (Yadav
SC
et al.,2015). MS patients often present with the relapsing-remitting form of the disease (RRMS), which consists of recurring episodes of unpredictable and sudden neurologic
NU
worsening (relapse) that may or may not leave permanent damage followed by periods
MA
of months to years of remission. RRMS is commonly managed with first-line immunomodulatory drug treatments, such as glatiramer acetate (GA) or interferon-beta,
D
which have been shown to reduce relapses by about 30% (Ford et al., 2010;Johnson,
TE
2012). However, given the unpredictable and heterogeneous nature of RRMS, it is difficult to anticipate patient prognosis or response to treatment (Graber and Dhib-
AC CE P
Jalbut, 2011). Currently there is a significant need for the development and validation of reliable biomarkers for the prediction of disease activity and treatment response. Epigenetic regulators such as histone deacetylases (HDACs) and histone acetyltransferases are increasingly being implicated as direct or indirect components in the regulation of expression of neuronal, immune, and other tissue-specific genes (Houtkooper et al., 2012). Histone protein post-translational modifications have the ability to affect chromatin structure and regulate gene expression (Guarente, 2011; Martin et al., 2015). Sirtuin 1 (SIRT1) is a member of the HDAC class III family of proteins (Feige and Auwerx, 2008). It is a NAD-dependent histone and protein deacetylase that catalyzes the removal of acetyl groups from a variety of protein substrates, including histones H1, 3
ACCEPTED MANUSCRIPT H3, and H4 (Vaquero et al., 2004). In addition, SIRT1 has been found to promote histone H3 lysine 9 (H3K9) methylation, resulting in epigenetic gene silencing (Feige
PT
and Auwerx, 2008). SIRT1 is involved in the regulation of a number of cellular
RI
processes, including transcription, metabolism, DNA repair, and aging (Guarente, 2011). SIRT1 can induce chromatin silencing through the deacetylation of histones
SC
(Vaquero et al., 2004) and can modulate cell survival by regulating the transcriptional
NU
activities of p53 (Luo et al., 2000), NF-κB (Yeung et al., 2004), FOXO proteins (Brunet
MA
et al., 2004; Motta et al., 2004), and p300 (Bouras et al., 2005). Recently, SIRT1 was found to positively regulate the function of Th17 cells by
D
modulating the activity of RORγt. SIRT1 deficiency results in impaired production of
TE
proinflammatory Th17 cells and reduced susceptibility to Th17 cell–mediated autoimmune disease (Lim et al., 2015). Consistent with these findings, treatment of
AC CE P
mice with Ex-527, a SIRT1-specific inhibitor, strongly suppressed the development of experimental autoimmune encephalomyelitis (EAE), a mouse model of MS (Lim et al., 2015). On the other hand, resveratrol, a SIRT1 activator, was shown in a separate study to ameliorate the disease course of EAE (Singh et al., 2007). Several studies have shown the ability of resveratrol to trigger apoptosis in activated T cells and also to induce a decrease in spinal cord inflammation during EAE (Singh et al., 2007). Other studies have demonstrated that resveratrol has immunomodulatory effects, altering the percentage of IL-17-positive T cells in the periphery and the CNS following long-term treatment in a relapsing-remitting EAE model (Imler and Petro, 2009) and shown that it is neuroprotective (Shindler et al., 2010). When EAE was induced in SIRT1overexpressing mice, a significant improvement in the clinical score, axon preservation,
4
ACCEPTED MANUSCRIPT and neuronal survival, as well as a decrease in inflammation and myelin loss were seen when compared to wild type mice. This protective phenotype appears to be associated
PT
with increased NAD+ and brain-derived neurotropic factor levels (Nimmagadda et al.,
RI
2013).
SC
Our studies have shown that SIRT1 is expressed by peripheral blood mononuclear cells (PBMCs), and in MS patients SIRT1 expression is significantly
NU
decreased during relapses (Tegla et al., 2014). SIRT1 was found to be expressed by
MA
perivascular CD3+ and CD68+ inflammatory cells in both acute and chronic active lesions in MS brains (Tegla et al., 2014), and it was also found to be co-localized with
D
oligodendrocytes and astrocytes in MS brains. A recent study (Miyazaki et al., 2014)
TE
found a decrease in SIRT1 mRNA expression in B cells from MS patients when
AC CE P
compared to healthy controls. This work suggested that an increase in miR-132, an endogenous suppressor of SIRT1, might be partially responsible for the decreased SIRT1 expression in B cells from MS patients. These results outline a novel miR-132SIRT1 axis that controls pro-inflammatory cytokine secretion by human B cells and demonstrate that dysregulation of this axis may underlie abnormal pro-inflammatory B cell cytokine responses in patients with MS (Miyazaki et al., 2014). SIRT1 has been identified as a potential biomarker of disease activity in MS (Tegla et al., 2014) but it is not known whether SIRT1 could serve as a biomarker of response to treatment. In the present study we longitudinally investigated expression of SIRT1 and its downstream effects on H3K9 to assess their effectiveness in detecting response to GA therapy. We found that responders to GA treatment had significantly higher SIRT1 mRNA levels when compared to non-responders. In addition, responders 5
ACCEPTED MANUSCRIPT to GA treatment had significantly higher H3K9me2 levels than did non-responders. Our data suggest that SIRT1 mRNA and H3K9me2 are potential biomarkers for evaluating
PT
MS patients’ responsiveness to GA therapy.
RI
Materials and Methods
SC
Patients and controls. A cohort of 15 patients with RRMS was enrolled in the study. The patients were recruited from the University of Maryland Multiple Sclerosis Center.
NU
The mean age was 40 (range 22-60) and consisted of 60% females (n=9) and 40%
MA
males (n=5). The criteria for inclusion of MS patients in the study were: (i) age 18 to 65 years; (ii) fulfillment of McDonald criteria for definite MS (Polman et al., 2005; Tegla et
D
al., 2013); (iii) relapsing-remitting course; (iv) having newly diagnosed MS, or MS not
TE
treated with currently used immunomodulatory drugs (interferon-β or GA) for 3 months
AC CE P
prior to study entry; (v) no exacerbations in the 4 weeks before the study; (vi) no i.v. or p.o. steroids for 4 weeks prior to study enrollment; (vii) no treatment with natalizumab, fingolimod , mitoxantrone, cyclophosphamide, or any investigational drugs during the past year; and (viii) a disability score of 0-5.5, as defined by the expanded disability status scale (EDSS) (Kruszewski et al., 2015). Exclusion criteria for MS patients were: (i) a history of autoimmune disorders, vascular disease, or active acute or chronic infections; (ii) use of antibiotics in the last 30 days; (iii) a history of intracranial or intraspinal tumor or metabolic myelopathy; or (iv) a history of alcohol or drug abuse. Study design. All MS patients received 20 mg of GA injected subcutaneously every day for 2 years. During this 2-year period, patients were clinically evaluated and peripheral blood samples were collected at baseline (month 0), and subsequently at 3, 6, and 12 months following the initiation of GA. Patients with symptoms suggestive of a 6
ACCEPTED MANUSCRIPT clinical relapse were requested to call the University of Maryland Multiple Sclerosis Center. Clinical relapse was defined as substantial worsening of pre-existing symptoms
PT
or appearance of new neurological deficits in the absence of fever or infections lasting
RI
more than 24 h. An EDSS evaluation was completed at each visit. Clinical records, consultation reports, and inpatient records were reviewed by a neurologist (H.R., C.B.,
SC
or W.R.) to ensure that the data obtained were complete. In the case of patients with
NU
relapse, the administration of 1g of methylprednisolone (Solu-Medrol) i.v. for 3 days was used to treat the disease exacerbation. A prednisone taper was also used after i.v.
MA
Solu-Medrol administration in certain cases. In such cases, blood samples were obtained prior to Solu-Medrol treatment. Responders to GA treatment were defined as
TE
D
patients who exhibited no relapse or no more than one relapse event during the 2-year span following the initiation of GA. Non-responders were defined as patients who
AC CE P
exhibited two or more relapse events during the 2-year span following the initiation of GA. According to these criteria, the present cohort consisted of 11 responders and 4 non-responders to treatment.
Collection of PBMCs, total RNA purification, and cDNA synthesis. PBMCs were isolated from each patient’s fresh blood sample using BD Vacutainer CPT tubes (Becton Dickinson, Franklin Lakes, NJ) at the time of their outpatient visits, and RNA isolation was performed the same day. Total RNA (0.5 µg per sample) was purified (Qiagen, Santa Clarita, CA), denatured, and reverse-transcribed using random primers to synthesize cDNA for each sample as previously described (Kruszewski et al., 2015). Real-time quantitative PCR. Real-time quantitative PCR was performed using a StepOne real-time PCR system (Applied Biosystems, Foster City, CA). The SIRT1 (For: 7
ACCEPTED MANUSCRIPT 5′-TGGCAAAGGAGCAGATTAGTAG-3′, Rev: 5′-GGCATGTCCCACTATCACTGT-3′) and FasL (For: 5′-GCCCATTTAACAGGCAAGTC-3′, Rev: 5′-
PT
ATCACAAGGCCACCCTTCTT-3′) primers were designed and synthesized by IDT
RI
(Coralville, IA) and used in conjunction with LightCycler FastStart DNA Master SYBR Green I (Roche, Indianapolis, IN) along with sample cDNA according to the
SC
manufacturer’s protocol. The same reaction was performed in the absence of sample
NU
cDNA as a negative control for each real-time PCR assay. For each gene, the cycle threshold (CT) values were determined in the exponential phase of the amplification plot
MA
and normalized to the mRNA expression of L13 ribosomal protein, a housekeeping gene. A standard curve was generated using serial dilutions of qPCR human reference
TE
D
total cDNA (Clontech, Mountain View, CA), and the normalized mRNA value (NRV) was calculated according to the following formula for relative expression of target mRNA:
AC CE P
NRV = (TarS/L13), where TarS represents the level of mRNA expression of the target gene, and L13 corresponds to that of the amplified L13 mRNA (Cudrici et al., 2008). Western Blotting. Western blot protein analysis was performed using patient PBMC samples that were lysed in RIPA buffer and processed as previously described (Niculescu et al., 1997). Whole-cell lysates (total protein = 20-30 g) were analyzed by 12% SDS-PAGE, followed by western blotting. Each membrane was analyzed for the expression of SIRT1 (Active Motif, Carlsbad, CA), histone H3 (CST, Danvers, MA), H3K9me2 (Active Motif), and H3K9ac (Active Motif) using specific antibodies. β-actin (Rockland Immunochemicals, Limerick, PA) was used as a loading control for normalization. Anti-rabbit or anti-mouse HRP-conjugated antibody (Santa Cruz Biotech, Dallas, TX) was used as the secondary antibody, and signals were visualized by
8
ACCEPTED MANUSCRIPT enhanced chemiluminescence (Denville, Holliston, MA) and autoradiography. Blots were then stripped and re-probed. The radiographic band density was measured using
PT
UN-SCAN-IT software (Silk Scientific, Orem, UT), and results were expressed as a ratio
RI
to β-actin (Rus et al., 1996).
SC
Statistical analysis. Comparisons between groups were performed using a two-tailed t-test, assuming unequal variances; p values <0.05 were considered significant.
NU
Statistical analysis was performed using GraphPad Prism software, version 6. All values
MA
are shown as means ± SEM and are representative of three experiments unless otherwise noted. Receiver operating characteristic (ROC) curve analysis was used to
D
assess the predictive accuracy of each potential biomarker. The predictive probabilities
TE
of binary outcomes regarding clinical state and response to GA treatment were reported as a C-statistic, or area under the curve (AUC, represented as a percentage, with a
Results
AC CE P
perfect score being 100% predictability).
Expression of SIRT1 mRNA in PBMCs of RRMS patients with acute relapse We first examined the expression of SIRT1 mRNA in unstimulated PBMCs to determine the differences in expression levels between patients who were having relapse and those who were clinically stable. The PBMC samples were obtained at baseline and subsequently collected at 3, 6, and 12 months following the initiation of GA treatment. A statistically significant decrease in SIRT1 mRNA expression level was found in patients during relapse when compared to those in remission (p<0.003) (Figure
9
ACCEPTED MANUSCRIPT 1A). These data confirm and extend our previous work suggesting that SIRT1 mRNA expression is a potential biomarker of disease activity in MS.
PT
Expression of SIRT1 mRNA in PBMCs in Responders vs. Non-Responders
RI
Since GA is an effective treatment for RRMS, we investigated the relationship
SC
between SIRT1 mRNA expression and responsiveness to treatment with GA. Responders to GA treatment were defined as patients who exhibited no relapse or no
NU
more than one relapse event during the 2-year span following the initiation of GA,
MA
whereas non-responders exhibited two or more relapse events. Overall, we found a significant decrease in SIRT1 mRNA expression in GA non-responders when compared
D
to responders (p=0.01) (Figure 1B). These data suggest that SIRT1 mRNA expression
TE
is a possible biomarker of response to therapy with GA. We also analyzed the time
AC CE P
course of SIRT1 mRNA expression levels over 0, 3, 6, and 12 months in responders versus non-responders. Over time, responders to GA showed persistently higher levels of SIRT1 mRNA, whereas non-responders showed persistently lower levels (Figure 2). The difference seen in mRNA expression at all the time points investigated did not reach statistical significance. We also examined the relationship between the expression of SIRT1 protein and responsiveness to treatment with GA. We did not see any statistically significant changes in SIRT1 protein levels in GA responders vs nonresponders (data not shown). Expression profile of histone H3K9me2 and H3K9ac in MS patients In addition to SIRT1 mRNA expression, we also measured the acetylation and methylation of histone H3 at lysine 9 by western blotting using specific antibodies. We
10
ACCEPTED MANUSCRIPT found that patients that were in a relapse had a significantly decreased (p=0.01) level of H3K9me2 when compared to patients that were stable (Figure 3A). In addition, we
PT
found that responders to GA treatment had significantly higher levels (p=0.018) of
RI
H3K9me2 than did non-responders (Figure 3B). However, for H3K9ac we found no
responders and non-responders (Figure 4).
SC
significant differences between stable patients and relapse patients, nor between
NU
We also examined the correlation between the levels of H3K9me2 and H3K9ac
MA
and those of SIRT1. In patients that were non-responders, there was a significant positive correlation (R=0.495, p=0.0304) between SIRT1 levels and H3K9me2 (Figure
D
5A) as well as a significantly positive correlation (R=0.504, p=0.0278) between SIRT1
TE
and H3K9ac (Figure 5B).
AC CE P
Correlation of H3K9me2 with FasL mRNA in RRMS Both FasL and RGC-32 mRNA levels have been previously shown to be significantly decreased in the PBMCs of MS patients during relapse and increased in responders to GA therapy (Kruszewski et al., 2015)). Similar trends were observed in our present investigations with H3K9me2 protein expression. Thus, H3K9me2 protein levels were correlated with those of FasL mRNA in MS patients and assessed using Pearson correlation analysis. H3K9me2 protein and FasL mRNA were positively correlated in patients during relapse (r=0.644, p=0.0129, Figure 6A) and in GA nonresponders (r=0.6953, p=0.0083, Figure 6B).
11
ACCEPTED MANUSCRIPT ROC analysis ROC analysis was used to assess the predictive accuracy of using SIRT1 mRNA
PT
levels or H3K9me2 as putative biomarkers to detect patient relapse and to assess
RI
responsiveness to GA treatment. We used ROC curve analysis in order to assess the
SC
probability (C-statistic or AUC) of SIRT1 or H3K9me2 in detecting a relapse and a response to treatment. The probability (C-statistic or AUC) of accurately detecting
NU
relapse was 72% using SIRT1 (p=0.02) (Figure 7A). The probability (C-statistic or AUC)
MA
of accurately detecting the response to GA was 70% (p=0.04) (Figure 7B). The probability (C-statistic or AUC) of accurately detecting relapse was 73%
D
using H3K9me2 (p=0.02) (Figure 8A). The probability (C-statistic or AUC) of accurately
AC CE P
Discussion
TE
detecting the response to GA using H3K9me2 was 71% (p=0.03) (Figure 8B).
The aim of the present study was to evaluate the expression of SIRT1 as a potential biomarker for GA responsiveness. We have previously shown that SIRT1 mRNA expression is decreased in unstimulated PBMCs during relapses in MS patients (Tegla et al., 2014). Our current data extend and confirm these prior results (Tegla et al., 2014) in a larger number of patients, also suggesting that SIRT1 is a possible biomarker of disease activity in MS. This paper is the first to investigate the role of SIRT1 and H3K9me2 as potential predictors of response to GA treatment in RRMS patients. We measured SIRT1 mRNA expression in RRMS patients who had started on GA therapy and followed their mRNA expression in relation to their clinical response to therapy. Our current data have demonstrated that the levels of SIRT1 mRNA are
12
ACCEPTED MANUSCRIPT significantly decreased in RRMS patients that are non-responders to GA treatment when compared to responders over a period of 12 months, suggesting that SIRT1 could
PT
be used as a possible biomarker for response to GA therapy. The results of our ROC
RI
analysis to calculate the diagnostic power of SIRT1 in assessing clinical activity and GA responsiveness also suggested that SIRT1 mRNA may be an effective biomarker to
SC
assess disease activity and response to GA therapy in MS patients. It is important to
NU
mention that we performed this analysis in unstimulated PBMCs in order to mimic the in vivo parameters as much as possible, since alterations in mRNA expression in
MA
unstimulated cells are more likely to have a predictive value for clinical exacerbations
D
and subsequently for response to therapy (Tegla et al., 2014).
TE
We have previously shown that the changes in SIRT1 expression levels during relapses are positively correlated with those of histone H3K9ac and histone H3K9me2
AC CE P
(Tegla et al., 2014). To further investigate the downstream effects of SIRT1 expression, we measured the effects of acetylation and methylation on histone H3 lysine 9 and compared the levels in responders to those in non-responders. Our results showed that non-responders had statistically significant lower levels of H3K9me2 and also suggested that H3K9me2 may be an effective biomarker to assess disease activity and response to GA therapy in MS patients. In addition, we found that SIRT1 mRNA expression correlated well with the expression of H3K9me2, implying that histone methyltransferase G9a and its partner protein GLP also play a role in the regulation of response to GA therapy. It is well known that histone lysine methylation participates in various chromatin-associated functions, including transcriptional regulation, heterochromatin formation, DNA repair, and recombination (Martin and Zhang, 2005).
13
ACCEPTED MANUSCRIPT G9a is the major mammalian methyltransferase responsible for mono- and dimethylation of histone H3K9 in euchromatin regions (Shinkai and Tachibana, 2011).
PT
Dimethylated H3K9 contributes to the binding of heterochromatin protein 1 (HP1). The
RI
epigenetic transcriptional silencing of proinflammatory genes requires G9a-dependent dimethylation of histone H3 lysine 9, recruitment of HP-1, CpG methylation of DNA, and
SC
assembly of H1 linker histone and high mobility group box 1 (Shinkai and Tachibana,
NU
2011) . We can speculate that the reduction in the dimethylation of H3K9 in patients with relapses and in non-responders might indicate transcriptional activation of
MA
proinflammatory genes. In addition, other studies have implicated H3K9me2 as a suppressor of IFN type I and of IFN response genes which are key regulators of innate
TE
D
antiviral immunity (Fang et al., 2012).
Since H3K9Ac and transcriptionally repressive H3K9 methylation (Turner, 2005)
AC CE P
are mutually exclusive, we expected to see an increase in H3K9Ac in non-responders to GA. Even though global H3K9Ac levels were unchanged in these patients, is still possible that changes in H3K9Ac can be seen at various proinflammatory gene promoters. H3K9 methylation and acetylation also play a significant role in establishing a transcriptionally active state of the FasL promoter in CD4+ cells in response to TCR stimulation (Ghare et al., 2014). The essential role of the FasL promoter H3K9Ac in transcriptional activation is supported by earlier work in which treatment of CD4+ T cells with a histone deacetylase inhibitor, trichostatin A (known to increase histone acetylation), led to enhanced activation-induced FasL expression and apoptosis (Turner, 2005). These findings are in agreement with our data showing a significant
14
ACCEPTED MANUSCRIPT correlation between FasL mRNA expression and histone H3K9 methylation and acetylation.
PT
In conclusion, our data suggest that SIRT1 mRNA and H3K9me2 can potentially
RI
serve as markers for predicting MS relapse and evaluating a patients’ responsiveness to GA therapy. Along with the other previously delineated biomarkers of response to GA
SC
(RGC-32, FasL and IL-21 (Kruszewski et al., 2015)), SIRT1 and H3k9me2 could
MA
NU
potentially be used in the future to aid physicians in predicting response to therapy.
ACKNOWLEDGMENTS
D
We thank Dr. Deborah McClellan for editing this manuscript. This work was supported in
TE
part by a Veterans Administration Merit Award BX001458 (to H.R.). Daniel Hewes was
AC CE P
supported in part by a medical student research grant from the Foundation of the Consortium of Multiple Sclerosis Centers’ MS Workforce of the Future program and the University of Maryland School of Medicine’s Office of Student Research. Adam Kruszewski was supported in part by a medical student research grant from the Foundation of the Consortium of Multiple Sclerosis Centers’ MS Workforce of the Future program and the Leslie B. Barnett Memorial Student Fellowship. Jonathan Ciriello was supported in part by the University of Maryland School of Medicine’s Office of Student Research.
15
ACCEPTED MANUSCRIPT References Bouras, T., Fu, M., Sauve, A.A., Wang, F., Quong, A.A., Perkins, N.D., Hay, R.T., Gu,
PT
W., and Pestell, R.G. (2005). SIRT1 deacetylation and repression of p300 involves
RI
lysine residues 1020/1024 within the cell cycle regulatory domain 1. J Biol Chem 280, 10264-10276.
SC
Brunet, A., Sweeney, L.B., Sturgill, J.F., Chua, K.F., Greer, P.L., Lin, Y., Tran, H., Ross,
NU
S.E., Mostoslavsky, R., Cohen, H.Y., et al. (2004). Stress-dependent regulation of FOXO transcription factors by the SIRT1 deacetylase. Science 303, 2011-2015.
MA
Cudrici, C., Ito, T., Zafranskaia, E., Weerth, S., Rus, V., Chen, H., Niculescu, F., Soloviova, K., Tegla, C., Gherman, A., et al. (2008). Complement C5 regulates the
TE
D
expression of insulin-like growth factor binding proteins in chronic experimental allergic encephalomyelitis. J Neuroimmunol 203, 94-103.
AC CE P
Fang, T.C., Schaefer, U., Mecklenbrauker, I., Stienen, A., Dewell, S., Chen, M.S., Rioja, I., Parravicini, V., Prinjha, R.K., Chandwani, R., et al. (2012). Histone H3 lysine 9 dimethylation as an epigenetic signature of the interferon response. J Exp Med 209, 661669.
Feige, J.N., and Auwerx, J. (2008). Transcriptional targets of sirtuins in the coordination of mammalian physiology. Curr Opin Cell Biol 20, 303-309. Ford, C., Goodman, A.D., Johnson, K., Kachuck, N., Lindsey, J.W., Lisak, R., Luzzio, C., Myers, L., Panitch, H., Preiningerova, J., et al. (2010). Continuous long-term immunomodulatory therapy in relapsing multiple sclerosis: results from the 15-year analysis of the US prospective open-label study of glatiramer acetate. Mult Scler J 16, 342-350.
16
ACCEPTED MANUSCRIPT Ghare, S.S., Joshi-Barve, S., Moghe, A., Patil, M., Barker, D.F., Gobejishvili, L., Brock, G.N., Cave, M., McClain, C.J., and Barve, S.S. (2014). Coordinated Histone H3
PT
Methylation and Acetylation Regulate Physiologic and Pathologic Fas Ligand Gene
RI
Expression in Human CD4+ T Cells. J Immunol 193, 412-421.
Graber, J.J., and Dhib-Jalbut, S. (2011). Biomarkers of disease activity in multiple
SC
sclerosis. J Neurol Sci 305, 1-10.
NU
Guarente, L. (2011). Franklin H. Epstein Lecture: Sirtuins, aging, and medicine. N Engl J Med 364, 2235-2244.
MA
Houtkooper, R.H., Pirinen, E., and Auwerx, J. (2012). Sirtuins as regulators of metabolism and healthspan. Nature Reviews Molecular Cell Biology.
TE
D
Imler, T.J., and Petro, T.M. (2009). Decreased severity of experimental autoimmune encephalomyelitis during resveratrol administration is associated with increased IL-
AC CE P
17+IL-10+ T cells, CD4− IFN-γ+ cells, and decreased macrophage IL-6 expression. Int Immunopharmacol 9, 134-143.
Johnson, K.P. (2012). Glatiramer acetate for treatment of relapsing–remitting multiple sclerosis. Expert Rev Neurother 12, 371-384. Kruszewski, A.M., Rao, G., Tatomir, A., Hewes, D., Tegla, C.A., Cudrici, C.D., Nguyen, V., Royal Iii, W., Bever Jr, C.T., Rus, V., and Rus, H. (2015). RGC-32 as a potential biomarker of relapse and response to treatment with glatiramer acetate in multiple sclerosis. Exp Mol Pathol 99, 498-505. Lim, H.W., Kang, S.G., Ryu, J.K., Schilling, B., Fei, M., Lee, I.S., Kehasse, A., Shirakawa, K., Yokoyama, M., Schnölzer, M., et al. (2015). SIRT1 deacetylates RORγt and enhances Th17 cell generation. J Exp Med 212, 607-617.
17
ACCEPTED MANUSCRIPT Luo, J., Su, F., Chen, D., Shiloh, A., and Gu, W. (2000). Deacetylation of p53 modulates its effect on cell growth and apoptosis. Nature 408, 377-381.
PT
Martin, A., Tegla, C.A., Cudrici, C.D., Kruszewski, A.M., Azimzadeh, P., Boodhoo, D.,
and neurodegeneration. Immunol Res 61, 187-197.
RI
Mekala, A.P., Rus, V., and Rus, H. (2015). Role of SIRT1 in autoimmune demyelination
SC
Martin, C., and Zhang, Y. (2005). The diverse functions of histone lysine methylation.
NU
Nat Rev Mol Cell Biol 6, 838-849.
Miyazaki, Y., Li, R., Rezk, A., Misirliyan, H., Moore, C., Farooqi, N., Solis, M., Goiry,
MA
L.G., de Faria Junior, O., Dang, V.D., et al. (2014). A Novel MicroRNA-132-Surtuin-1 Axis Underlies Aberrant B-cell Cytokine Regulation in Patients with Relapsing-Remitting
TE
D
Multiple Sclerosis. PLOS ONE 9, e105421. Motta, M.C., Divecha, N., Lemieux, M., Kamel, C., Chen, D., Gu, W., Bultsma, Y.,
AC CE P
McBurney, M., and Guarente, L. (2004). Mammalian SIRT1 represses forkhead transcription factors. Cell 116, 551-563. Niculescu, F., Rus, H., van Biesen, T., and Shin, M.L. (1997). Activation of Ras and mitogen-activated protein kinase pathway by terminal complement complexes is G protein dependent. J Immunol 158, 4405-4412. Nimmagadda, V.K., Bever, C.T., Vattikunta, N.R., Talat, S., Ahmad, V., Nagalla, N.K., Trisler, D., Judge, S.I.V., Royal, W., Chandrasekaran, K., et al. (2013). Overexpression of SIRT1 Protein in Neurons Protects against Experimental Autoimmune Encephalomyelitis through Activation of Multiple SIRT1 Targets. J Immunol 190, 45954607.
18
ACCEPTED MANUSCRIPT Polman, C.H., Reingold, S.C., Edan, G., Filippi, M., Hartung, H.-P., Kappos, L., Lublin, F.D., Metz, L.M., McFarland, H.F., O'Connor, P.W., et al. (2005). Diagnostic criteria for
PT
multiple sclerosis: 2005 revisions to the “McDonald Criteria”. Ann Neurol 58, 840-846.
RI
Rus, H.G., Niculescu, F., and Shin, M.L. (1996). Sublytic complement attack induces cell cycle in oligodendrocytes. J Immunol 156, 4892-4900.
SC
Shindler, K.S., Ventura, E., Dutt, M., Elliott, P., Fitzgerald, D.C., and Rostami, A. (2010).
NU
Oral Resveratrol Reduces Neuronal Damage in a Model of Multiple Sclerosis. J Neuroophthalmol 30, 328-339.
MA
Shinkai, Y., and Tachibana, M. (2011). H3K9 methyltransferase G9a and the related molecule GLP. Genes & Development 25, 781-788.
TE
D
Singh, N.P., Hegde, V.L., Hofseth, L.J., Nagarkatti, M., and Nagarkatti, P. (2007). Resveratrol (trans-3,5,4'-Trihydroxystilbene) Ameliorates Experimental Allergic
AC CE P
Encephalomyelitis, Primarily via Induction of Apoptosis in T Cells Involving Activation of Aryl Hydrocarbon Receptor and Estrogen Receptor. Mol Pharmacol 72, 1508-1521. Tegla, C.A., Azimzadeh, P., Andrian-Albescu, M., Martin, A., Cudrici, C.D., Trippe Iii, R., Sugarman, A., Chen, H., Boodhoo, D., Vlaicu, S.I., et al. (2014). SIRT1 is decreased during relapses in patients with multiple sclerosis. Exp Mol Pathol 96, 139-148. Tegla, C.A., Cudrici, C.D., Azimzadeh, P., Singh, A.K., Trippe, R., 3rd, Khan, A., Chen, H., Andrian-Albescu, M., Royal, W., 3rd, Bever, C., et al. (2013). Dual role of Response gene to complement-32 in multiple sclerosis. Exp Mol Pathol 94, 17-28. Turner, B.M. (2005). Reading signals on the nucleosome with a new nomenclature for modified histones. Nat Struct Mol Biol 12, 110-112.
19
ACCEPTED MANUSCRIPT Vaquero, A., Scher, M., Lee, D., Erdjument-Bromage, H., Tempst, P., and Reinberg, D. (2004). Human SirT1 interacts with histone H1 and promotes formation of facultative
PT
heterochromatin. Mol Cell 16, 93-105.
RI
Yadav, S.K., Mindur, J.E., Ito, K., and Dhib-Jalbut, S. (2015). Advances in the immunopathogenesis of multiple sclerosis. Curr Opin Neurol 28, 206-219.
SC
Yeung, F., Hoberg, J.E., Ramsey, C.S., Keller, M.D., Jones, D.R., Frye, R.A., and
NU
Mayo, M.W. (2004). Modulation of NF-kappaB-dependent transcription and cell survival
AC CE P
TE
D
MA
by the SIRT1 deacetylase. EMBO J 23, 2369-2380.
20
ACCEPTED MANUSCRIPT Figure Legends: Figure 1. SIRT1 mRNA expression in stable and relapsing RRMS patients and in
PT
responders and non-responders to GA treatment. PBMCs from RRMS patients were
RI
obtained at baseline (month 0) and then collected at 3, 6, and 12 months following the
SC
initiation of GA treatment. Target gene mRNA expression was measured in patients’ PBMCs by real-time qRT-PCR and expressed as a ratio to L13 mRNA expression.
NU
Overall, significantly lower levels of SIRT1/L13 (p<0.003) were found in patients during
MA
relapse when compared to clinically stable patients (A) and in patients who were nonresponders to GA treatment when compared with responders (B, p=0.01).
D
Figure 2. Time course of SIRT1 mRNA expression in RRMS patients treated with
TE
GA. PBMCs from RRMS patients were obtained at baseline (month 0) and then
AC CE P
collected at 3, 6, and 12 months following the initiation of GA treatment. Over time, responders to GA showed persistently higher levels of SIRT1 mRNA expression, while non-responders showed persistently lower levels. Figure 3. Expression of H3K9me2 expression in stable and relapsing RRMS patients and in RRMS patients classified as responders and non-responders to GA treatment. Dimethylation of H3K9 was measured in patients’ PBMCs by western blotting and expressed as a ratio to β-actin expression. Overall, significantly lower levels of H3K9me2 were found in patients during relapse when compared to the levels in stable patients (p=0.01) (A) and in patients who were non-responders to GA treatment when compared with responders (p=0.018) (B).
21
ACCEPTED MANUSCRIPT C. An example of a western blot from a non-responder to GA patient. PBMCs from RRMS patients were obtained at baseline (month 0) and then collected at 3, 6, and 12
PT
months following the initiation of GA treatment.
RI
D. An example of a western blot from a responder to GA. PBMCs from RRMS patients
SC
were obtained at baseline (month 0) and then collected at 3, 6 and 9 months following the initiation of GA treatment.
NU
Figure 4. Expression of H3K9ac in stable and relapsing RRMS patients and in
MA
RRMS patients classified as responders and non-responders to GA treatment. H3K9ac expression was measured in patients’ PBMCs by western blotting and
D
expressed as a ratio to β-actin protein expression. Overall, there were no significant
AC CE P
responders (B).
TE
differences in stable vs. relapse MS patients (A) or in responders to GA vs non-
Figure 5. SIRT1 protein expression is correlated with H3K9me2 and H3K9ac in non-responders to GA. SIRT1, H3K9me2, and H3K9ac protein was measured in MS patients’ isolated PBMCs by western blot analysis and expressed as a ratio to β-actin protein expression. Correlation of SIRT1 with H3K9me2 and H3K9ac expression was assessed using Pearson correlation coefficients. SIRT1 protein levels were positively correlated with levels of H3K9me2 (A; R=0.495, p=0.0304) and H3K9ac (B; R=0.504, p=0.0278). Figure 6. Correlation of H3K9me2 with FasL mRNA expression. (A) FasL mRNA was measured in patients’ PBMCs by qRT-PCR and expressed as a ratio to L13 mRNA expression. FasL mRNA expression was significantly decreased in 22
ACCEPTED MANUSCRIPT non-responders when compared to responders (p=0.0016). (B). H3K9me2 expression levels were positively correlated with levels of FasL mRNA (R=0.644, p=0.0129) in
PT
relapse patients. (C) H3K9me2 expression levels were positively correlated with levels
RI
of FasL mRNA (R=0.6953, p=0.0083) in patients who were non-responders.
SC
Figure 7. ROC curve analysis to assess the predictive accuracy of SIRT1 mRNA in detecting patient relapse and response to GA.
NU
We used ROC curve analysis to assess the probability (C-statistic, or AUC) of SIRT1
MA
mRNA of detecting a relapse and a response to treatment. The probability of accurately detecting relapse was 72% using SIRT1 (p=0.02) (A). The probability of accurately
D
detecting the response to GA was 70% (p=0.04) (B).
TE
Figure 8. ROC curve analysis to assess the predictive accuracy of H3K9me2 in
AC CE P
detecting patient relapse and response to GA. We used ROC curve analysis to assess the probability (C-statistic, or AUC) of H3K9me2 in detecting a relapse and response to treatment with GA. The probability of accurately detecting relapse was 73% using H3K9me2 (p=0.02) (A). The probability of accurately detecting the response to GA was 71% (p=0.03) (B).
23
ACCEPTED MANUSCRIPT Abbreviations
AC CE P
TE
D
MA
NU
SC
RI
PT
AUC - area under the curve CNS - central nervous system CPT - cell preparation tubes EAE - experimental autoimmune encephalomyelitis EDSS - expanded disability status scale FasL - fas ligand FOXO - forkhead box protein O GA - glatiramer acetate H3K9 - histone H3 lysine 9 HDAC - histone deacetylase MS - multiple sclerosis NAD - nicotinamide adenine dinucleotide NF-kB - nuclear factor kappa light chain enhancer of activated B cells NRV - normalized mRNA value PBMCs - peripheral blood mononuclear cells RGC-32 - response gene to complement 32 ROC - receiver operating characteristic RORγt - RAR-related orphan receptor gamma t RRMS - relapsing-remitting multiple sclerosis SIRT1 - sirtuin 1
24
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Fig. 1
25
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Fig. 2
26
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Fig. 3
27
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Fig. 4
28
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Fig. 5
29
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Fig. 6
30
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Fig. 7
31
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Fig. 8
32
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Fig. 9
33
ACCEPTED MANUSCRIPT Highlights The role of SIRT1 as a biomarker of response to treatment was investigated.
SIRT1 was found to be a biomarker of response to glatiramer acetate treatment.
SIRT1 was found to be a biomarker of clinical activity in multiple sclerosis.
Histone H3K9me2 was also found to be a biomarker of response to GA treatment.
AC CE P
TE
D
MA
NU
SC
RI
PT
34
S0014-4800(17)30031-X doi:10.1016/j.yexmp.2017.01.014 YEXMP 4007
To appear in:
Experimental and Molecular Pathology
Received date: Accepted date:
17 January 2017 17 January 2017
Please cite this article as: Hewes, Daniel, Tatomir, Alexandru, Kruszewski, Adam M., Rao, Gautam, Tegla, Cosmin A., Ciriello, Jonathan, Nguyen, Vingh, Royal III, Walter, Bever, Christopher, Rus, Violeta, Rus, Horea, SIRT1 as a potential biomarker of response to treatment with glatiramer acetate in multiple sclerosis, Experimental and Molecular Pathology (2017), doi:10.1016/j.yexmp.2017.01.014
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT SIRT1 as a potential biomarker of response to treatment with Glatiramer Acetate in multiple sclerosis
PT
Daniel Hewes, Alexandru Tatomir, Adam M. Kruszewski, Gautam Rao, Cosmin A.
RI
Tegla§, Jonathan Ciriello, Vingh Nguyen‡, Walter Royal III*†, Christopher Bever*†,
SC
Violeta Rus‡, Horea Rus§†.
NU
Department of Neurology, University of Maryland, School of Medicine, §Research
MA
Service, Veterans Administration Maryland Health Care System, ‡ Department of Medicine, Division of Rheumatology and Clinical Immunology, University of Maryland,
D
School of Medicine, †Veterans Administration Multiple Sclerosis Center of Excellence,
TE
Baltimore, MD, USA
AC CE P
Running title: SIRT1 and multiple sclerosis Address correspondence to: Horea Rus M.D., Ph.D., University of Maryland School of Medicine, Department of Neurology, 655 W Baltimore St, BRB 12-033, Baltimore, MD 21201. Tel: (410) 706-3170; FAX: (410) 706-0186; E-mail: [email protected] D. Hewes, A. Tatomir and A. M. Kruszewski contributed equally to this manuscript.
Key words: SIRT1, Multiple Sclerosis, Glatiramer Acetate, Biomarker, Histone H3, Methylation
1
ACCEPTED MANUSCRIPT Abstract SIRT1, a NAD dependent histone and protein deacetylase, is a member of the histone
PT
deacetylase class III family. We previously showed that SIRT1 mRNA expression is
RI
significantly lower in peripheral blood mononuclear cells (PBMCs) of multiple sclerosis
SC
(MS) patients during relapses than in stable patients. We have now investigated SIRT1 as a possible biomarker to predict relapse as well as responsiveness to glatiramer
NU
acetate (GA) treatment in relapsing-remitting MS (RRMS) patients. Over the course of
MA
2 years, a cohort of 15 GA-treated RRMS patients were clinically monitored using the Expanded Disability Status Scale and assessed for MS relapses. Blood samples
D
collected from MS patients were analyzed for levels of SIRT1 and histone H3 lysine 9
TE
(H3K9) acetylation and dimethylation. During relapses, MS patients had a lower expression of SIRT1 mRNA than did stable MS patients. In addition, there was a
AC CE P
significant decrease in H3K9 dimethylation (H3K9me2) during relapses in MS patients when compared to stable patients (p=0.01). Responders to GA treatment had significantly higher SIRT1 mRNA (p=0.01) and H3K9me2 levels than did nonresponders (p=0.018). Receiver operating characteristic analysis was used to assess the predictive power of SIRT1 and H3K9me2 as putative biomarkers: for SIRT1 mRNA, the predictive value for responsiveness to GA treatment was 70% (p=0.04) and for H3K9me2 was 71% (p=0.03). Our data suggest that SIRT1 and H3K9me2 could serve as potential biomarkers for evaluating patients’ responsiveness to GA therapy in order to help guide treatment decisions in MS.
2
ACCEPTED MANUSCRIPT Introduction Multiple sclerosis (MS) is a demyelinating disease characterized by chronic
PT
inflammation of the central nervous system (CNS), in which many factors (genetic and
RI
environmental) act in concert to influence disease susceptibility and progression (Yadav
SC
et al.,2015). MS patients often present with the relapsing-remitting form of the disease (RRMS), which consists of recurring episodes of unpredictable and sudden neurologic
NU
worsening (relapse) that may or may not leave permanent damage followed by periods
MA
of months to years of remission. RRMS is commonly managed with first-line immunomodulatory drug treatments, such as glatiramer acetate (GA) or interferon-beta,
D
which have been shown to reduce relapses by about 30% (Ford et al., 2010;Johnson,
TE
2012). However, given the unpredictable and heterogeneous nature of RRMS, it is difficult to anticipate patient prognosis or response to treatment (Graber and Dhib-
AC CE P
Jalbut, 2011). Currently there is a significant need for the development and validation of reliable biomarkers for the prediction of disease activity and treatment response. Epigenetic regulators such as histone deacetylases (HDACs) and histone acetyltransferases are increasingly being implicated as direct or indirect components in the regulation of expression of neuronal, immune, and other tissue-specific genes (Houtkooper et al., 2012). Histone protein post-translational modifications have the ability to affect chromatin structure and regulate gene expression (Guarente, 2011; Martin et al., 2015). Sirtuin 1 (SIRT1) is a member of the HDAC class III family of proteins (Feige and Auwerx, 2008). It is a NAD-dependent histone and protein deacetylase that catalyzes the removal of acetyl groups from a variety of protein substrates, including histones H1, 3
ACCEPTED MANUSCRIPT H3, and H4 (Vaquero et al., 2004). In addition, SIRT1 has been found to promote histone H3 lysine 9 (H3K9) methylation, resulting in epigenetic gene silencing (Feige
PT
and Auwerx, 2008). SIRT1 is involved in the regulation of a number of cellular
RI
processes, including transcription, metabolism, DNA repair, and aging (Guarente, 2011). SIRT1 can induce chromatin silencing through the deacetylation of histones
SC
(Vaquero et al., 2004) and can modulate cell survival by regulating the transcriptional
NU
activities of p53 (Luo et al., 2000), NF-κB (Yeung et al., 2004), FOXO proteins (Brunet
MA
et al., 2004; Motta et al., 2004), and p300 (Bouras et al., 2005). Recently, SIRT1 was found to positively regulate the function of Th17 cells by
D
modulating the activity of RORγt. SIRT1 deficiency results in impaired production of
TE
proinflammatory Th17 cells and reduced susceptibility to Th17 cell–mediated autoimmune disease (Lim et al., 2015). Consistent with these findings, treatment of
AC CE P
mice with Ex-527, a SIRT1-specific inhibitor, strongly suppressed the development of experimental autoimmune encephalomyelitis (EAE), a mouse model of MS (Lim et al., 2015). On the other hand, resveratrol, a SIRT1 activator, was shown in a separate study to ameliorate the disease course of EAE (Singh et al., 2007). Several studies have shown the ability of resveratrol to trigger apoptosis in activated T cells and also to induce a decrease in spinal cord inflammation during EAE (Singh et al., 2007). Other studies have demonstrated that resveratrol has immunomodulatory effects, altering the percentage of IL-17-positive T cells in the periphery and the CNS following long-term treatment in a relapsing-remitting EAE model (Imler and Petro, 2009) and shown that it is neuroprotective (Shindler et al., 2010). When EAE was induced in SIRT1overexpressing mice, a significant improvement in the clinical score, axon preservation,
4
ACCEPTED MANUSCRIPT and neuronal survival, as well as a decrease in inflammation and myelin loss were seen when compared to wild type mice. This protective phenotype appears to be associated
PT
with increased NAD+ and brain-derived neurotropic factor levels (Nimmagadda et al.,
RI
2013).
SC
Our studies have shown that SIRT1 is expressed by peripheral blood mononuclear cells (PBMCs), and in MS patients SIRT1 expression is significantly
NU
decreased during relapses (Tegla et al., 2014). SIRT1 was found to be expressed by
MA
perivascular CD3+ and CD68+ inflammatory cells in both acute and chronic active lesions in MS brains (Tegla et al., 2014), and it was also found to be co-localized with
D
oligodendrocytes and astrocytes in MS brains. A recent study (Miyazaki et al., 2014)
TE
found a decrease in SIRT1 mRNA expression in B cells from MS patients when
AC CE P
compared to healthy controls. This work suggested that an increase in miR-132, an endogenous suppressor of SIRT1, might be partially responsible for the decreased SIRT1 expression in B cells from MS patients. These results outline a novel miR-132SIRT1 axis that controls pro-inflammatory cytokine secretion by human B cells and demonstrate that dysregulation of this axis may underlie abnormal pro-inflammatory B cell cytokine responses in patients with MS (Miyazaki et al., 2014). SIRT1 has been identified as a potential biomarker of disease activity in MS (Tegla et al., 2014) but it is not known whether SIRT1 could serve as a biomarker of response to treatment. In the present study we longitudinally investigated expression of SIRT1 and its downstream effects on H3K9 to assess their effectiveness in detecting response to GA therapy. We found that responders to GA treatment had significantly higher SIRT1 mRNA levels when compared to non-responders. In addition, responders 5
ACCEPTED MANUSCRIPT to GA treatment had significantly higher H3K9me2 levels than did non-responders. Our data suggest that SIRT1 mRNA and H3K9me2 are potential biomarkers for evaluating
PT
MS patients’ responsiveness to GA therapy.
RI
Materials and Methods
SC
Patients and controls. A cohort of 15 patients with RRMS was enrolled in the study. The patients were recruited from the University of Maryland Multiple Sclerosis Center.
NU
The mean age was 40 (range 22-60) and consisted of 60% females (n=9) and 40%
MA
males (n=5). The criteria for inclusion of MS patients in the study were: (i) age 18 to 65 years; (ii) fulfillment of McDonald criteria for definite MS (Polman et al., 2005; Tegla et
D
al., 2013); (iii) relapsing-remitting course; (iv) having newly diagnosed MS, or MS not
TE
treated with currently used immunomodulatory drugs (interferon-β or GA) for 3 months
AC CE P
prior to study entry; (v) no exacerbations in the 4 weeks before the study; (vi) no i.v. or p.o. steroids for 4 weeks prior to study enrollment; (vii) no treatment with natalizumab, fingolimod , mitoxantrone, cyclophosphamide, or any investigational drugs during the past year; and (viii) a disability score of 0-5.5, as defined by the expanded disability status scale (EDSS) (Kruszewski et al., 2015). Exclusion criteria for MS patients were: (i) a history of autoimmune disorders, vascular disease, or active acute or chronic infections; (ii) use of antibiotics in the last 30 days; (iii) a history of intracranial or intraspinal tumor or metabolic myelopathy; or (iv) a history of alcohol or drug abuse. Study design. All MS patients received 20 mg of GA injected subcutaneously every day for 2 years. During this 2-year period, patients were clinically evaluated and peripheral blood samples were collected at baseline (month 0), and subsequently at 3, 6, and 12 months following the initiation of GA. Patients with symptoms suggestive of a 6
ACCEPTED MANUSCRIPT clinical relapse were requested to call the University of Maryland Multiple Sclerosis Center. Clinical relapse was defined as substantial worsening of pre-existing symptoms
PT
or appearance of new neurological deficits in the absence of fever or infections lasting
RI
more than 24 h. An EDSS evaluation was completed at each visit. Clinical records, consultation reports, and inpatient records were reviewed by a neurologist (H.R., C.B.,
SC
or W.R.) to ensure that the data obtained were complete. In the case of patients with
NU
relapse, the administration of 1g of methylprednisolone (Solu-Medrol) i.v. for 3 days was used to treat the disease exacerbation. A prednisone taper was also used after i.v.
MA
Solu-Medrol administration in certain cases. In such cases, blood samples were obtained prior to Solu-Medrol treatment. Responders to GA treatment were defined as
TE
D
patients who exhibited no relapse or no more than one relapse event during the 2-year span following the initiation of GA. Non-responders were defined as patients who
AC CE P
exhibited two or more relapse events during the 2-year span following the initiation of GA. According to these criteria, the present cohort consisted of 11 responders and 4 non-responders to treatment.
Collection of PBMCs, total RNA purification, and cDNA synthesis. PBMCs were isolated from each patient’s fresh blood sample using BD Vacutainer CPT tubes (Becton Dickinson, Franklin Lakes, NJ) at the time of their outpatient visits, and RNA isolation was performed the same day. Total RNA (0.5 µg per sample) was purified (Qiagen, Santa Clarita, CA), denatured, and reverse-transcribed using random primers to synthesize cDNA for each sample as previously described (Kruszewski et al., 2015). Real-time quantitative PCR. Real-time quantitative PCR was performed using a StepOne real-time PCR system (Applied Biosystems, Foster City, CA). The SIRT1 (For: 7
ACCEPTED MANUSCRIPT 5′-TGGCAAAGGAGCAGATTAGTAG-3′, Rev: 5′-GGCATGTCCCACTATCACTGT-3′) and FasL (For: 5′-GCCCATTTAACAGGCAAGTC-3′, Rev: 5′-
PT
ATCACAAGGCCACCCTTCTT-3′) primers were designed and synthesized by IDT
RI
(Coralville, IA) and used in conjunction with LightCycler FastStart DNA Master SYBR Green I (Roche, Indianapolis, IN) along with sample cDNA according to the
SC
manufacturer’s protocol. The same reaction was performed in the absence of sample
NU
cDNA as a negative control for each real-time PCR assay. For each gene, the cycle threshold (CT) values were determined in the exponential phase of the amplification plot
MA
and normalized to the mRNA expression of L13 ribosomal protein, a housekeeping gene. A standard curve was generated using serial dilutions of qPCR human reference
TE
D
total cDNA (Clontech, Mountain View, CA), and the normalized mRNA value (NRV) was calculated according to the following formula for relative expression of target mRNA:
AC CE P
NRV = (TarS/L13), where TarS represents the level of mRNA expression of the target gene, and L13 corresponds to that of the amplified L13 mRNA (Cudrici et al., 2008). Western Blotting. Western blot protein analysis was performed using patient PBMC samples that were lysed in RIPA buffer and processed as previously described (Niculescu et al., 1997). Whole-cell lysates (total protein = 20-30 g) were analyzed by 12% SDS-PAGE, followed by western blotting. Each membrane was analyzed for the expression of SIRT1 (Active Motif, Carlsbad, CA), histone H3 (CST, Danvers, MA), H3K9me2 (Active Motif), and H3K9ac (Active Motif) using specific antibodies. β-actin (Rockland Immunochemicals, Limerick, PA) was used as a loading control for normalization. Anti-rabbit or anti-mouse HRP-conjugated antibody (Santa Cruz Biotech, Dallas, TX) was used as the secondary antibody, and signals were visualized by
8
ACCEPTED MANUSCRIPT enhanced chemiluminescence (Denville, Holliston, MA) and autoradiography. Blots were then stripped and re-probed. The radiographic band density was measured using
PT
UN-SCAN-IT software (Silk Scientific, Orem, UT), and results were expressed as a ratio
RI
to β-actin (Rus et al., 1996).
SC
Statistical analysis. Comparisons between groups were performed using a two-tailed t-test, assuming unequal variances; p values <0.05 were considered significant.
NU
Statistical analysis was performed using GraphPad Prism software, version 6. All values
MA
are shown as means ± SEM and are representative of three experiments unless otherwise noted. Receiver operating characteristic (ROC) curve analysis was used to
D
assess the predictive accuracy of each potential biomarker. The predictive probabilities
TE
of binary outcomes regarding clinical state and response to GA treatment were reported as a C-statistic, or area under the curve (AUC, represented as a percentage, with a
Results
AC CE P
perfect score being 100% predictability).
Expression of SIRT1 mRNA in PBMCs of RRMS patients with acute relapse We first examined the expression of SIRT1 mRNA in unstimulated PBMCs to determine the differences in expression levels between patients who were having relapse and those who were clinically stable. The PBMC samples were obtained at baseline and subsequently collected at 3, 6, and 12 months following the initiation of GA treatment. A statistically significant decrease in SIRT1 mRNA expression level was found in patients during relapse when compared to those in remission (p<0.003) (Figure
9
ACCEPTED MANUSCRIPT 1A). These data confirm and extend our previous work suggesting that SIRT1 mRNA expression is a potential biomarker of disease activity in MS.
PT
Expression of SIRT1 mRNA in PBMCs in Responders vs. Non-Responders
RI
Since GA is an effective treatment for RRMS, we investigated the relationship
SC
between SIRT1 mRNA expression and responsiveness to treatment with GA. Responders to GA treatment were defined as patients who exhibited no relapse or no
NU
more than one relapse event during the 2-year span following the initiation of GA,
MA
whereas non-responders exhibited two or more relapse events. Overall, we found a significant decrease in SIRT1 mRNA expression in GA non-responders when compared
D
to responders (p=0.01) (Figure 1B). These data suggest that SIRT1 mRNA expression
TE
is a possible biomarker of response to therapy with GA. We also analyzed the time
AC CE P
course of SIRT1 mRNA expression levels over 0, 3, 6, and 12 months in responders versus non-responders. Over time, responders to GA showed persistently higher levels of SIRT1 mRNA, whereas non-responders showed persistently lower levels (Figure 2). The difference seen in mRNA expression at all the time points investigated did not reach statistical significance. We also examined the relationship between the expression of SIRT1 protein and responsiveness to treatment with GA. We did not see any statistically significant changes in SIRT1 protein levels in GA responders vs nonresponders (data not shown). Expression profile of histone H3K9me2 and H3K9ac in MS patients In addition to SIRT1 mRNA expression, we also measured the acetylation and methylation of histone H3 at lysine 9 by western blotting using specific antibodies. We
10
ACCEPTED MANUSCRIPT found that patients that were in a relapse had a significantly decreased (p=0.01) level of H3K9me2 when compared to patients that were stable (Figure 3A). In addition, we
PT
found that responders to GA treatment had significantly higher levels (p=0.018) of
RI
H3K9me2 than did non-responders (Figure 3B). However, for H3K9ac we found no
responders and non-responders (Figure 4).
SC
significant differences between stable patients and relapse patients, nor between
NU
We also examined the correlation between the levels of H3K9me2 and H3K9ac
MA
and those of SIRT1. In patients that were non-responders, there was a significant positive correlation (R=0.495, p=0.0304) between SIRT1 levels and H3K9me2 (Figure
D
5A) as well as a significantly positive correlation (R=0.504, p=0.0278) between SIRT1
TE
and H3K9ac (Figure 5B).
AC CE P
Correlation of H3K9me2 with FasL mRNA in RRMS Both FasL and RGC-32 mRNA levels have been previously shown to be significantly decreased in the PBMCs of MS patients during relapse and increased in responders to GA therapy (Kruszewski et al., 2015)). Similar trends were observed in our present investigations with H3K9me2 protein expression. Thus, H3K9me2 protein levels were correlated with those of FasL mRNA in MS patients and assessed using Pearson correlation analysis. H3K9me2 protein and FasL mRNA were positively correlated in patients during relapse (r=0.644, p=0.0129, Figure 6A) and in GA nonresponders (r=0.6953, p=0.0083, Figure 6B).
11
ACCEPTED MANUSCRIPT ROC analysis ROC analysis was used to assess the predictive accuracy of using SIRT1 mRNA
PT
levels or H3K9me2 as putative biomarkers to detect patient relapse and to assess
RI
responsiveness to GA treatment. We used ROC curve analysis in order to assess the
SC
probability (C-statistic or AUC) of SIRT1 or H3K9me2 in detecting a relapse and a response to treatment. The probability (C-statistic or AUC) of accurately detecting
NU
relapse was 72% using SIRT1 (p=0.02) (Figure 7A). The probability (C-statistic or AUC)
MA
of accurately detecting the response to GA was 70% (p=0.04) (Figure 7B). The probability (C-statistic or AUC) of accurately detecting relapse was 73%
D
using H3K9me2 (p=0.02) (Figure 8A). The probability (C-statistic or AUC) of accurately
AC CE P
Discussion
TE
detecting the response to GA using H3K9me2 was 71% (p=0.03) (Figure 8B).
The aim of the present study was to evaluate the expression of SIRT1 as a potential biomarker for GA responsiveness. We have previously shown that SIRT1 mRNA expression is decreased in unstimulated PBMCs during relapses in MS patients (Tegla et al., 2014). Our current data extend and confirm these prior results (Tegla et al., 2014) in a larger number of patients, also suggesting that SIRT1 is a possible biomarker of disease activity in MS. This paper is the first to investigate the role of SIRT1 and H3K9me2 as potential predictors of response to GA treatment in RRMS patients. We measured SIRT1 mRNA expression in RRMS patients who had started on GA therapy and followed their mRNA expression in relation to their clinical response to therapy. Our current data have demonstrated that the levels of SIRT1 mRNA are
12
ACCEPTED MANUSCRIPT significantly decreased in RRMS patients that are non-responders to GA treatment when compared to responders over a period of 12 months, suggesting that SIRT1 could
PT
be used as a possible biomarker for response to GA therapy. The results of our ROC
RI
analysis to calculate the diagnostic power of SIRT1 in assessing clinical activity and GA responsiveness also suggested that SIRT1 mRNA may be an effective biomarker to
SC
assess disease activity and response to GA therapy in MS patients. It is important to
NU
mention that we performed this analysis in unstimulated PBMCs in order to mimic the in vivo parameters as much as possible, since alterations in mRNA expression in
MA
unstimulated cells are more likely to have a predictive value for clinical exacerbations
D
and subsequently for response to therapy (Tegla et al., 2014).
TE
We have previously shown that the changes in SIRT1 expression levels during relapses are positively correlated with those of histone H3K9ac and histone H3K9me2
AC CE P
(Tegla et al., 2014). To further investigate the downstream effects of SIRT1 expression, we measured the effects of acetylation and methylation on histone H3 lysine 9 and compared the levels in responders to those in non-responders. Our results showed that non-responders had statistically significant lower levels of H3K9me2 and also suggested that H3K9me2 may be an effective biomarker to assess disease activity and response to GA therapy in MS patients. In addition, we found that SIRT1 mRNA expression correlated well with the expression of H3K9me2, implying that histone methyltransferase G9a and its partner protein GLP also play a role in the regulation of response to GA therapy. It is well known that histone lysine methylation participates in various chromatin-associated functions, including transcriptional regulation, heterochromatin formation, DNA repair, and recombination (Martin and Zhang, 2005).
13
ACCEPTED MANUSCRIPT G9a is the major mammalian methyltransferase responsible for mono- and dimethylation of histone H3K9 in euchromatin regions (Shinkai and Tachibana, 2011).
PT
Dimethylated H3K9 contributes to the binding of heterochromatin protein 1 (HP1). The
RI
epigenetic transcriptional silencing of proinflammatory genes requires G9a-dependent dimethylation of histone H3 lysine 9, recruitment of HP-1, CpG methylation of DNA, and
SC
assembly of H1 linker histone and high mobility group box 1 (Shinkai and Tachibana,
NU
2011) . We can speculate that the reduction in the dimethylation of H3K9 in patients with relapses and in non-responders might indicate transcriptional activation of
MA
proinflammatory genes. In addition, other studies have implicated H3K9me2 as a suppressor of IFN type I and of IFN response genes which are key regulators of innate
TE
D
antiviral immunity (Fang et al., 2012).
Since H3K9Ac and transcriptionally repressive H3K9 methylation (Turner, 2005)
AC CE P
are mutually exclusive, we expected to see an increase in H3K9Ac in non-responders to GA. Even though global H3K9Ac levels were unchanged in these patients, is still possible that changes in H3K9Ac can be seen at various proinflammatory gene promoters. H3K9 methylation and acetylation also play a significant role in establishing a transcriptionally active state of the FasL promoter in CD4+ cells in response to TCR stimulation (Ghare et al., 2014). The essential role of the FasL promoter H3K9Ac in transcriptional activation is supported by earlier work in which treatment of CD4+ T cells with a histone deacetylase inhibitor, trichostatin A (known to increase histone acetylation), led to enhanced activation-induced FasL expression and apoptosis (Turner, 2005). These findings are in agreement with our data showing a significant
14
ACCEPTED MANUSCRIPT correlation between FasL mRNA expression and histone H3K9 methylation and acetylation.
PT
In conclusion, our data suggest that SIRT1 mRNA and H3K9me2 can potentially
RI
serve as markers for predicting MS relapse and evaluating a patients’ responsiveness to GA therapy. Along with the other previously delineated biomarkers of response to GA
SC
(RGC-32, FasL and IL-21 (Kruszewski et al., 2015)), SIRT1 and H3k9me2 could
MA
NU
potentially be used in the future to aid physicians in predicting response to therapy.
ACKNOWLEDGMENTS
D
We thank Dr. Deborah McClellan for editing this manuscript. This work was supported in
TE
part by a Veterans Administration Merit Award BX001458 (to H.R.). Daniel Hewes was
AC CE P
supported in part by a medical student research grant from the Foundation of the Consortium of Multiple Sclerosis Centers’ MS Workforce of the Future program and the University of Maryland School of Medicine’s Office of Student Research. Adam Kruszewski was supported in part by a medical student research grant from the Foundation of the Consortium of Multiple Sclerosis Centers’ MS Workforce of the Future program and the Leslie B. Barnett Memorial Student Fellowship. Jonathan Ciriello was supported in part by the University of Maryland School of Medicine’s Office of Student Research.
15
ACCEPTED MANUSCRIPT References Bouras, T., Fu, M., Sauve, A.A., Wang, F., Quong, A.A., Perkins, N.D., Hay, R.T., Gu,
PT
W., and Pestell, R.G. (2005). SIRT1 deacetylation and repression of p300 involves
RI
lysine residues 1020/1024 within the cell cycle regulatory domain 1. J Biol Chem 280, 10264-10276.
SC
Brunet, A., Sweeney, L.B., Sturgill, J.F., Chua, K.F., Greer, P.L., Lin, Y., Tran, H., Ross,
NU
S.E., Mostoslavsky, R., Cohen, H.Y., et al. (2004). Stress-dependent regulation of FOXO transcription factors by the SIRT1 deacetylase. Science 303, 2011-2015.
MA
Cudrici, C., Ito, T., Zafranskaia, E., Weerth, S., Rus, V., Chen, H., Niculescu, F., Soloviova, K., Tegla, C., Gherman, A., et al. (2008). Complement C5 regulates the
TE
D
expression of insulin-like growth factor binding proteins in chronic experimental allergic encephalomyelitis. J Neuroimmunol 203, 94-103.
AC CE P
Fang, T.C., Schaefer, U., Mecklenbrauker, I., Stienen, A., Dewell, S., Chen, M.S., Rioja, I., Parravicini, V., Prinjha, R.K., Chandwani, R., et al. (2012). Histone H3 lysine 9 dimethylation as an epigenetic signature of the interferon response. J Exp Med 209, 661669.
Feige, J.N., and Auwerx, J. (2008). Transcriptional targets of sirtuins in the coordination of mammalian physiology. Curr Opin Cell Biol 20, 303-309. Ford, C., Goodman, A.D., Johnson, K., Kachuck, N., Lindsey, J.W., Lisak, R., Luzzio, C., Myers, L., Panitch, H., Preiningerova, J., et al. (2010). Continuous long-term immunomodulatory therapy in relapsing multiple sclerosis: results from the 15-year analysis of the US prospective open-label study of glatiramer acetate. Mult Scler J 16, 342-350.
16
ACCEPTED MANUSCRIPT Ghare, S.S., Joshi-Barve, S., Moghe, A., Patil, M., Barker, D.F., Gobejishvili, L., Brock, G.N., Cave, M., McClain, C.J., and Barve, S.S. (2014). Coordinated Histone H3
PT
Methylation and Acetylation Regulate Physiologic and Pathologic Fas Ligand Gene
RI
Expression in Human CD4+ T Cells. J Immunol 193, 412-421.
Graber, J.J., and Dhib-Jalbut, S. (2011). Biomarkers of disease activity in multiple
SC
sclerosis. J Neurol Sci 305, 1-10.
NU
Guarente, L. (2011). Franklin H. Epstein Lecture: Sirtuins, aging, and medicine. N Engl J Med 364, 2235-2244.
MA
Houtkooper, R.H., Pirinen, E., and Auwerx, J. (2012). Sirtuins as regulators of metabolism and healthspan. Nature Reviews Molecular Cell Biology.
TE
D
Imler, T.J., and Petro, T.M. (2009). Decreased severity of experimental autoimmune encephalomyelitis during resveratrol administration is associated with increased IL-
AC CE P
17+IL-10+ T cells, CD4− IFN-γ+ cells, and decreased macrophage IL-6 expression. Int Immunopharmacol 9, 134-143.
Johnson, K.P. (2012). Glatiramer acetate for treatment of relapsing–remitting multiple sclerosis. Expert Rev Neurother 12, 371-384. Kruszewski, A.M., Rao, G., Tatomir, A., Hewes, D., Tegla, C.A., Cudrici, C.D., Nguyen, V., Royal Iii, W., Bever Jr, C.T., Rus, V., and Rus, H. (2015). RGC-32 as a potential biomarker of relapse and response to treatment with glatiramer acetate in multiple sclerosis. Exp Mol Pathol 99, 498-505. Lim, H.W., Kang, S.G., Ryu, J.K., Schilling, B., Fei, M., Lee, I.S., Kehasse, A., Shirakawa, K., Yokoyama, M., Schnölzer, M., et al. (2015). SIRT1 deacetylates RORγt and enhances Th17 cell generation. J Exp Med 212, 607-617.
17
ACCEPTED MANUSCRIPT Luo, J., Su, F., Chen, D., Shiloh, A., and Gu, W. (2000). Deacetylation of p53 modulates its effect on cell growth and apoptosis. Nature 408, 377-381.
PT
Martin, A., Tegla, C.A., Cudrici, C.D., Kruszewski, A.M., Azimzadeh, P., Boodhoo, D.,
and neurodegeneration. Immunol Res 61, 187-197.
RI
Mekala, A.P., Rus, V., and Rus, H. (2015). Role of SIRT1 in autoimmune demyelination
SC
Martin, C., and Zhang, Y. (2005). The diverse functions of histone lysine methylation.
NU
Nat Rev Mol Cell Biol 6, 838-849.
Miyazaki, Y., Li, R., Rezk, A., Misirliyan, H., Moore, C., Farooqi, N., Solis, M., Goiry,
MA
L.G., de Faria Junior, O., Dang, V.D., et al. (2014). A Novel MicroRNA-132-Surtuin-1 Axis Underlies Aberrant B-cell Cytokine Regulation in Patients with Relapsing-Remitting
TE
D
Multiple Sclerosis. PLOS ONE 9, e105421. Motta, M.C., Divecha, N., Lemieux, M., Kamel, C., Chen, D., Gu, W., Bultsma, Y.,
AC CE P
McBurney, M., and Guarente, L. (2004). Mammalian SIRT1 represses forkhead transcription factors. Cell 116, 551-563. Niculescu, F., Rus, H., van Biesen, T., and Shin, M.L. (1997). Activation of Ras and mitogen-activated protein kinase pathway by terminal complement complexes is G protein dependent. J Immunol 158, 4405-4412. Nimmagadda, V.K., Bever, C.T., Vattikunta, N.R., Talat, S., Ahmad, V., Nagalla, N.K., Trisler, D., Judge, S.I.V., Royal, W., Chandrasekaran, K., et al. (2013). Overexpression of SIRT1 Protein in Neurons Protects against Experimental Autoimmune Encephalomyelitis through Activation of Multiple SIRT1 Targets. J Immunol 190, 45954607.
18
ACCEPTED MANUSCRIPT Polman, C.H., Reingold, S.C., Edan, G., Filippi, M., Hartung, H.-P., Kappos, L., Lublin, F.D., Metz, L.M., McFarland, H.F., O'Connor, P.W., et al. (2005). Diagnostic criteria for
PT
multiple sclerosis: 2005 revisions to the “McDonald Criteria”. Ann Neurol 58, 840-846.
RI
Rus, H.G., Niculescu, F., and Shin, M.L. (1996). Sublytic complement attack induces cell cycle in oligodendrocytes. J Immunol 156, 4892-4900.
SC
Shindler, K.S., Ventura, E., Dutt, M., Elliott, P., Fitzgerald, D.C., and Rostami, A. (2010).
NU
Oral Resveratrol Reduces Neuronal Damage in a Model of Multiple Sclerosis. J Neuroophthalmol 30, 328-339.
MA
Shinkai, Y., and Tachibana, M. (2011). H3K9 methyltransferase G9a and the related molecule GLP. Genes & Development 25, 781-788.
TE
D
Singh, N.P., Hegde, V.L., Hofseth, L.J., Nagarkatti, M., and Nagarkatti, P. (2007). Resveratrol (trans-3,5,4'-Trihydroxystilbene) Ameliorates Experimental Allergic
AC CE P
Encephalomyelitis, Primarily via Induction of Apoptosis in T Cells Involving Activation of Aryl Hydrocarbon Receptor and Estrogen Receptor. Mol Pharmacol 72, 1508-1521. Tegla, C.A., Azimzadeh, P., Andrian-Albescu, M., Martin, A., Cudrici, C.D., Trippe Iii, R., Sugarman, A., Chen, H., Boodhoo, D., Vlaicu, S.I., et al. (2014). SIRT1 is decreased during relapses in patients with multiple sclerosis. Exp Mol Pathol 96, 139-148. Tegla, C.A., Cudrici, C.D., Azimzadeh, P., Singh, A.K., Trippe, R., 3rd, Khan, A., Chen, H., Andrian-Albescu, M., Royal, W., 3rd, Bever, C., et al. (2013). Dual role of Response gene to complement-32 in multiple sclerosis. Exp Mol Pathol 94, 17-28. Turner, B.M. (2005). Reading signals on the nucleosome with a new nomenclature for modified histones. Nat Struct Mol Biol 12, 110-112.
19
ACCEPTED MANUSCRIPT Vaquero, A., Scher, M., Lee, D., Erdjument-Bromage, H., Tempst, P., and Reinberg, D. (2004). Human SirT1 interacts with histone H1 and promotes formation of facultative
PT
heterochromatin. Mol Cell 16, 93-105.
RI
Yadav, S.K., Mindur, J.E., Ito, K., and Dhib-Jalbut, S. (2015). Advances in the immunopathogenesis of multiple sclerosis. Curr Opin Neurol 28, 206-219.
SC
Yeung, F., Hoberg, J.E., Ramsey, C.S., Keller, M.D., Jones, D.R., Frye, R.A., and
NU
Mayo, M.W. (2004). Modulation of NF-kappaB-dependent transcription and cell survival
AC CE P
TE
D
MA
by the SIRT1 deacetylase. EMBO J 23, 2369-2380.
20
ACCEPTED MANUSCRIPT Figure Legends: Figure 1. SIRT1 mRNA expression in stable and relapsing RRMS patients and in
PT
responders and non-responders to GA treatment. PBMCs from RRMS patients were
RI
obtained at baseline (month 0) and then collected at 3, 6, and 12 months following the
SC
initiation of GA treatment. Target gene mRNA expression was measured in patients’ PBMCs by real-time qRT-PCR and expressed as a ratio to L13 mRNA expression.
NU
Overall, significantly lower levels of SIRT1/L13 (p<0.003) were found in patients during
MA
relapse when compared to clinically stable patients (A) and in patients who were nonresponders to GA treatment when compared with responders (B, p=0.01).
D
Figure 2. Time course of SIRT1 mRNA expression in RRMS patients treated with
TE
GA. PBMCs from RRMS patients were obtained at baseline (month 0) and then
AC CE P
collected at 3, 6, and 12 months following the initiation of GA treatment. Over time, responders to GA showed persistently higher levels of SIRT1 mRNA expression, while non-responders showed persistently lower levels. Figure 3. Expression of H3K9me2 expression in stable and relapsing RRMS patients and in RRMS patients classified as responders and non-responders to GA treatment. Dimethylation of H3K9 was measured in patients’ PBMCs by western blotting and expressed as a ratio to β-actin expression. Overall, significantly lower levels of H3K9me2 were found in patients during relapse when compared to the levels in stable patients (p=0.01) (A) and in patients who were non-responders to GA treatment when compared with responders (p=0.018) (B).
21
ACCEPTED MANUSCRIPT C. An example of a western blot from a non-responder to GA patient. PBMCs from RRMS patients were obtained at baseline (month 0) and then collected at 3, 6, and 12
PT
months following the initiation of GA treatment.
RI
D. An example of a western blot from a responder to GA. PBMCs from RRMS patients
SC
were obtained at baseline (month 0) and then collected at 3, 6 and 9 months following the initiation of GA treatment.
NU
Figure 4. Expression of H3K9ac in stable and relapsing RRMS patients and in
MA
RRMS patients classified as responders and non-responders to GA treatment. H3K9ac expression was measured in patients’ PBMCs by western blotting and
D
expressed as a ratio to β-actin protein expression. Overall, there were no significant
AC CE P
responders (B).
TE
differences in stable vs. relapse MS patients (A) or in responders to GA vs non-
Figure 5. SIRT1 protein expression is correlated with H3K9me2 and H3K9ac in non-responders to GA. SIRT1, H3K9me2, and H3K9ac protein was measured in MS patients’ isolated PBMCs by western blot analysis and expressed as a ratio to β-actin protein expression. Correlation of SIRT1 with H3K9me2 and H3K9ac expression was assessed using Pearson correlation coefficients. SIRT1 protein levels were positively correlated with levels of H3K9me2 (A; R=0.495, p=0.0304) and H3K9ac (B; R=0.504, p=0.0278). Figure 6. Correlation of H3K9me2 with FasL mRNA expression. (A) FasL mRNA was measured in patients’ PBMCs by qRT-PCR and expressed as a ratio to L13 mRNA expression. FasL mRNA expression was significantly decreased in 22
ACCEPTED MANUSCRIPT non-responders when compared to responders (p=0.0016). (B). H3K9me2 expression levels were positively correlated with levels of FasL mRNA (R=0.644, p=0.0129) in
PT
relapse patients. (C) H3K9me2 expression levels were positively correlated with levels
RI
of FasL mRNA (R=0.6953, p=0.0083) in patients who were non-responders.
SC
Figure 7. ROC curve analysis to assess the predictive accuracy of SIRT1 mRNA in detecting patient relapse and response to GA.
NU
We used ROC curve analysis to assess the probability (C-statistic, or AUC) of SIRT1
MA
mRNA of detecting a relapse and a response to treatment. The probability of accurately detecting relapse was 72% using SIRT1 (p=0.02) (A). The probability of accurately
D
detecting the response to GA was 70% (p=0.04) (B).
TE
Figure 8. ROC curve analysis to assess the predictive accuracy of H3K9me2 in
AC CE P
detecting patient relapse and response to GA. We used ROC curve analysis to assess the probability (C-statistic, or AUC) of H3K9me2 in detecting a relapse and response to treatment with GA. The probability of accurately detecting relapse was 73% using H3K9me2 (p=0.02) (A). The probability of accurately detecting the response to GA was 71% (p=0.03) (B).
23
ACCEPTED MANUSCRIPT Abbreviations
AC CE P
TE
D
MA
NU
SC
RI
PT
AUC - area under the curve CNS - central nervous system CPT - cell preparation tubes EAE - experimental autoimmune encephalomyelitis EDSS - expanded disability status scale FasL - fas ligand FOXO - forkhead box protein O GA - glatiramer acetate H3K9 - histone H3 lysine 9 HDAC - histone deacetylase MS - multiple sclerosis NAD - nicotinamide adenine dinucleotide NF-kB - nuclear factor kappa light chain enhancer of activated B cells NRV - normalized mRNA value PBMCs - peripheral blood mononuclear cells RGC-32 - response gene to complement 32 ROC - receiver operating characteristic RORγt - RAR-related orphan receptor gamma t RRMS - relapsing-remitting multiple sclerosis SIRT1 - sirtuin 1
24
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Fig. 1
25
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Fig. 2
26
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Fig. 3
27
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Fig. 4
28
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Fig. 5
29
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Fig. 6
30
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Fig. 7
31
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Fig. 8
32
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
Fig. 9
33
ACCEPTED MANUSCRIPT Highlights The role of SIRT1 as a biomarker of response to treatment was investigated.
SIRT1 was found to be a biomarker of response to glatiramer acetate treatment.
SIRT1 was found to be a biomarker of clinical activity in multiple sclerosis.
Histone H3K9me2 was also found to be a biomarker of response to GA treatment.
AC CE P
TE
D
MA
NU
SC
RI
PT
34