Skip main navigation
×

Targeting Treatment Refractory NET by EZH2 Inhibition in Postural Tachycardia Syndrome

Originally publishedhttps://doi.org/10.1161/CIRCRESAHA.119.315654Circulation Research. 2020;126:1058–1060

Meet the First Author, see p 944

Postural tachycardia syndrome (POTS) is a form of chronic orthostatic intolerance for which the hallmark is an excessive increase in heart rate following upright posture in the absence of orthostatic hypotension.1,2 Faulty neuronal reuptake of the sympathetic nervous system3 by the transmitter norepinephrine is implicated by suppression of the norepinephrine transporter (NET) encoded by the SLC6A2 gene (abbreviated as NET). A challenge in POTS biology is determining how NET is silenced and why this state of default suppression is refractory to reactivation.

HDAC (histone deacetylase) inhibitors such a suberanilohydroxamic acid (SAHA) can only—partially attenuate transcriptional suppression in some—but not all cases of treatment refractory POTS.4 This has resulted in a paucity of new therapeutic approaches to reactivate NET and remains an unmet need in POTS treatment. The histone methyltransferase EZH2 (enhancer of zeste homolog 2) is responsible for trimethylation of lysine 27 of histone 3 (H3K27me3) and is a powerful epigenetic mark on NET.5 We examined cell viability in response to titrating concentrations of SAHA and GSK126, a highly selective pharmacological inhibitor of EZH2 methyltransferase activity. Increasing concentrations of GSK126 and SAHA did not significantly reduce cell viability (Online Figure I). GSK126 reactivated NET expression at low (1–5 µM) and higher doses (10–20 µmol/L) whereas SAHA failed to influence transcription (Online Figure II). LI-COR Odyssey quantitation show significant reductions in H3K27me3 by GSK126 (Online Figure III). Despite the massive increases in H3K9/14ac (Online Figure IV), NET cannot be reactivated by SAHA in cells that we have shown GSK126 can influence transcription. To determine if NET suppression is mediated by EZH2 and the H3K27me3 state, we immunoprecipitated chromatin (ChIP).5 Proteins were cross-linked to DNA and soluble chromatin fractionated by sonication and DNA immunopurified using an antibody that specifically recognizes H3K27me3. DNA was analyzed by qPCR using amplimers specifically designed to detect human NET. We observed significant reduction in H3K27me3 at the NET promoter in cells treated with GSK126 (Online Figure V) whereas SAHA did not influence H3K27me3. The presence of H3K27me3 makes local chromatin structure less accessible to transcriptional machinery and molecular probes such as endonucleases. Micrococcal nuclease cleaves DNA and is widely used to assess chromatin accessibility. To determine whether this inactive chromatin structure is brought about by H3K27me3, we assessed NET using micrococcal nuclease in cells treated with GSK126 and SAHA. This technique distinguishes chromatin hypersensitivity and our results show enhanced NET accessibility in cells treated with GSK126, whereas hypersensitivity was marginally affected by SAHA (Online Figure VI).

To determine if NET suppression in POTS is consistent with a model of default repression mediated by EZH2, we assessed H3K27me3 by ChIP. We observed significant H3K27me3 enrichment on NET in leukocytes derived from treatment-refractory POTS when compared with healthy individuals (Figure [A]). Next, we used GSK126 to competitively reduce H3K27me3 content in an attempt to reactivate NET mRNA levels. POTS leukocytes stimulated with increasing concentrations of GSK126 (Online Figure VII) resulted in pronounced reactivation (Online Figure VIII). Ex vivo stimulation was also compared with SAHA because NET silencing is characterized by tight epigenetic control.4 GSK126 has a 4-fold transcriptional reactivation index when compared with SAHA (Figure [B]). SAHA did not influence NET expression even at higher tested doses (Online Figure IX). These results suggest the failure to maximally reactivate NET transcription in POTS cells by SAHA is likely due to the inability to restore activation states despite massive H3K9/14 acetylation (Figure [C]) whereas the tight suppression by H3K27me3 is selectively targeted by GSK126. If EZH2 inhibition is responsible for NET reactivation, then GSK126 treatment should reduce H3K27me3 content. We treated healthy and POTS-derived leukocytes with titrating concentrations of GSK126 and fractionated soluble histones. LI-COR analyses show a quantitative reduction in H3K27me3 levels in leukocytes (Figure [D]). To investigate the role of H3K27me3 in NET silencing, leukocytes from healthy and POTS individuals were treated in the presence (GSK126) or absence (vehicle control, DMSO) of the EZH2 inhibitor. ChIP was used to specifically isolate highly enriched H3K27me3 fractions and the DNA was assessed by qPCR. In healthy leukocytes, we recovered NET DNA equally well from vehicle and GSK126-treated cells. In contrast to healthy leukocytes, H3K27me3 content was highly enriched on the NET promoter in POTS cells, and GSK126 treatment almost abolished this repressive epigenetic mark on the gene (Figure [E]).

Figure.

Figure. Refractory chromatinization characterized by H3K27me3 in cells derived from patients with postural tachycardia syndrome (POTS) whose NET gene is transcriptionally silent can be pharmacologically reactivated by GSK126.A, ChIP reveals the NET promoter is enriched with H3K27me3 in leukocytes derived from treatment refractory POTS. Participants included in this study were treatment refractory. They had ongoing symptoms despite nonmedication strategies including maintenance of salt and fluid intake, use of compression garments, and involvement in a graded exercise program. At the time of inclusion in this study, patients had a mean symptom duration of 7.45 y. Leukocytes isolated from healthy (n=4) and POTS (n=4) participants were assessed for EZH2-dependent H3K27me3. Quantitative polymerase chain reaction (PCR) analysis of DNA in ChIP with anti-H3K27me3 antibody. DNA was assessed using amplimers that specifically detect NET. Data are represented as the mean±SEM, *P=0.028 was calculated using 2-tailed Mann-Whitney test. B, Pharmacological NET reactivation by GSK126. Comparison of the transcriptional reactivation index in refractory POTS leukocytes by GSK126 and SAHA. Vehicle control is dimethyl sulfoxide (DMSO; n=4). Data are represented as the mean±SEM. Adjusted P value (*P=0.0001) was calculated using a 2-way ANOVA with Bonferroni multiple comparison test. The comparison is made against the vehicle for each drug. C, H3K9/14 acetylation content in leukocytes isolated from POTS participants and stimulated with suberanilohydroxamic acid (SAHA). Bar graph represents mean values of LI-COR Odyssey quantification of protein adjusted against histone H3. Vehicle control is DMSO (n=4). Data are represented as the mean±SEM. Adjusted P values *P=0.0002 (SAHA, 2 µM), **P=0.0001 (SAHA, 5 µM), and ***P=0.0003 (SAHA, 15 µM) were calculated using 1-way ANOVA with Bonferroni multiple comparison test where each drug concentration is compared with vehicle control. D, Pharmacological inhibition of EZH2 activity by GSK126 attenuates H3K27me3 protein content in healthy and treatment refractory POTS leukocytes. Bar graph represents mean values of LI-COR Odyssey quantification of protein when adjusted against histone H3. Vehicle control is DMSO (n=4). Data are represented as the mean±SEM, adjusted P values *P=0.004 (healthy) and **P=0.002 (POTS) were calculated using a 2-way ANOVA with Bonferroni multiple comparison test. The comparison is made against the vehicle for each drug. E, GSK126 abolishes H3K27me3 on NET chromatin in refractory POTS but not in healthy leukocytes. Leukocytes isolated from healthy (n=4) and POTS (n=4) participants were treated with GSK126 and chromatin fractionated for EZH2-dependent H3K27me3 content on the NET gene. Quantitative PCR analysis of DNA in ChIP with anti-H3K27me3 antibody. DNA was assessed using amplimers that specifically detect NET. Vehicle control is DMSO. Data are represented as the mean±SEM, adjusted *P=0.022 and **P=0.0001 were calculated using a 2-way ANOVA with correction performed for multiple comparison using Bonferroni multiple comparison test. F, Schematic representation of NET chromatin modification and conversion of default repression state to transcriptional reactivation by pharmacological inhibition of EZH2 by GSK126.

Although GSK126 is not licensed for the treatment of super-refractory POTS, the primary findings described here add to previous reports5 that chromatin modification might be a therapeutic option where other approved therapeutic protocols fail. Further studies are warranted to better understand potential drug interactions and unforeseen synergistic benefits, which we predict will be an important area of clinical epigenetics. This is the first description for gain of H3K27me3 content on NET that advance our understanding of default suppression in POTS (Figure [F]). We provide proof-of-concept and a therapeutic strategy that is seemingly amenable to effectively restore histone content. These results suggest exciting possibilities to reactivate refractory and transcriptionally silent genes implicated in POTS.

Nonstandard Abbreviations and Acronyms

EZH2

enhancer of zeste homolog 2

HDAC

histone deacetylase

POTS

postural tachycardia syndrome

SAHA

suberanilohydroxamic acid

Footnotes

*H.K. and A.W.K. contributed equally to this article.

For Sources of Funding and Disclosures, see page 1059.

Correspondence to: Assam El-Osta, Epigenetics in Human Health and Disease, Central Clinical School, Monash University, 99 Commercial Road, Melbourne 3004, Australia. Email

References

  • 1. Shannon JR, Flattem NL, Jordan J, Jacob G, Black BK, Biaggioni I, Blakely RD, Robertson D. Orthostatic intolerance and tachycardia associated with norepinephrine-transporter deficiency.N Engl J Med. 2000; 342:541–549. 10.1056/NEJM200002243420803CrossrefMedlineGoogle Scholar
  • 2. Raj SR. Postural tachycardia syndrome (POTS).Circulation. 2013; 127:2336–2342. 10.1161/CIRCULATIONAHA.112.144501LinkGoogle Scholar
  • 3. Goldstein DS, Brush JE, Eisenhofer G, Stull R, Esler M. In vivo measurement of neuronal uptake of norepinephrine in the human heart.Circulation. 1988; 78:41–48. 10.1161/01.cir.78.1.41LinkGoogle Scholar
  • 4. Khan AW, Ziemann M, Corcoran SJ, K N H, Okabe J, Rafehi H, Maxwell SS, Esler MD, El-Osta A. NET silencing by let-7i in postural tachycardia syndrome.JCI Insight. 2017; 2:e90183. doi: 10.1172/jci.insight.90183CrossrefMedlineGoogle Scholar
  • 5. Bayles R, Harikrishnan KN, Lambert E, Baker EK, Agrotis A, Guo L, Jowett JB, Esler M, Lambert G, El-Osta A. Epigenetic modification of the norepinephrine transporter gene in postural tachycardia syndrome.Arterioscler Thromb Vasc Biol. 2012; 32:1910–1916. 10.1161/ATVBAHA.111.244343LinkGoogle Scholar