Expression Levels of XIST RNA Predict PTSS and Chronic Pain Outcomes

Expression levels of XIST RNA predict PTSS and chronic pain outcomes in women experiencing motor vehicle collision Linnstaedt SD1,2 ,Yu S1,2, Chen C1,2, Kurz M3, Pearson C4, Hendry PL5, Lewandowski C6, Domeier R7, Damiron K8, S, McLean SA1,2,9 From 1the Institute of Trauma Recovery; 2Department of Anesthesiology, UNC-Chapel Hill; 3Department of Emergency Medicine, University of Alabama; 4Department of Emergency Medicine Detroit Receiving; 5Department of Emergency Medicine Shands Jacksonville Medical Center; 6Department of Emergency Medicine Henry Ford Hospital; 7Department of Emergency Medicine St. Joseph Mercy Health System; 8Department of Emergency Medicine Albert Einstein Medical Center; 9Department of Emergency Medicine, UNC-Chapel Hill

FIGURE 1 and TABLE 1. Study design and characteristics. Sixty-six African American women were enrolled in INTRODUCTION one of 13 Emergency Departments in the early aftermath of motor vehicle collision. Blood was collected at the Characteristic RESULTS time of enrollment and CPTP and PTSS was assessed 6 weeks, 6 months, and 1 year following MVC. Women Enrolled, n 66 Women experiencing motor vehicle collision (MVC) are at Age, years, mean (SD) 30 (10) • African American women age 18 to 65 (Table 1) presenting in Education, n (%) substantially increased risk of both chronic post-traumatic pain HS or less 11 (16.7) the Emergency Department (ED) following MVC were enrolled (CPTP) and posttraumatic stress symptom (PTSS) severity. Post-HS not college 3 (4.5) Some college 35 (53) in the study (n=66). Blood was collected in the ED at the time 10-40% College 14 (21.2) X chromosome inactivation (XCl) is one candidate mechanism Post-college 3 (4.5) of enrollment and CPTP and PTSS were assessed (Figure 1). contributing to sexual dimorphism in individuals with CPTP and Collision characteristics Characteristics of the study population are provided in Table 1. Driver, n (%) 49 (74.2) PTSS. The long non-coding RNA, X-inactive specific transcript Severe vehicle damage, n (%) 38 (57.6) • Higher XIST RNA expression levels were associated with 1 Seatbelt worn, n (%) 61 (92.4) (XIST), is known to be a major regulator of XCI. Altered ability of Airbag deployed, n (%) 24 (36.4) Distress* in the early aftermath of increased risk of developing CPTP and PTSS six months XIST to coat the X chromosome is one factor that facilitates gene 25 (11) CPTP PTSS trauma, mean (SD) following MVC (Figure 2). Overall pain in the ED (0-10 NRS), escape from XCI; sixty-two human X chromosome genes are known 6.9 (2.3) 2,3 mean (SD) escapees. recover IESR score at 6 months, mean (SD) 29.9 (28.6) • Previously defined escapee genes were mapped on the Overall pain at 6 months, mean (SD) 5.2 (3.4) human X chromosome (Figure 3). Most genes that escape Previous studies have shown that 1) XIST RNA is over-expressed in > *distress measured with the peritraumatic distress 2,3 females with major affective disorders4, and 2) escapee genes are inventory XCI are located on the p arm of the X chromosome. associated with depression, bipolar disorder, and mental Eighteen of these escape genes (defined in blue, Figure 3) are 4,5 FIGURE 2. XIST RNA expression levels are FIGURE 3. X Chromosome genes known to TABLE 2. Competitive gene set analyses show that positively correlated with XIST RNA expression levels (p<0.05) impairment . higher in women who develop CPTP and PTSS the set of eighteen X chromosome genes that are escape XCI. The eighteen genes that are in women following MVC. Example correlations are shown in HYPOTHESIS six months following MVC than in women who positively correlated with XIST in women correlated with XIST in MVC study participants, are recover. following MVC are shown in blue. more differentially expressed in women that develop Figure 4. As a control, we examined the relationship between CPTP and PTSS vs recover following MVC than any Expression levels of XIST RNA predict CPTP and PTSS outcomes in * *

other set of eighteen genes. XIST RNA expression levels and the expression of genes n

women experiencing MVC, and X chromosome gene transcripts o 1 0 0 0 0 0 i

s never known to escape XCI; the expression of these s

known to escape XCI are correlated with XIST RNA expression levels e r EIF2S3; ZFX; p 6 transcripts were negatively correlated (data not shown).

and these genes are associated with CPTP and PTSS outcomes in x 9 0 0 0 0 EIF1AX; FAM9C; Gene set enrichment analysis using Camera E ACE2; CA5BP1; A # of genes Direction* P value • Gene set enrichment analyses demonstrated that the eighteen

women. N ZRSR2; S100G; -4 R 8 0 0 0 0 SYAP1; TXLNG; CPTP 18 Up 3.8x10 T X chromosome genes correlated with XIST in women who

-6 S METHODS I TCEANC; RAB9A; PTSD 18 Up 5.4x10 X TRAPPC2; OFD1; *direction of expression of genes in women who develop CPTP and develop CPTP and PTSS following MVC were more 7 0 0 0 0 GPM6B; GEMIN8; VENTXP1 PTSD relative to expression of the same genes in women who recover African American women (n = 66) age 18 to 65 presenting to one of C P T P P T S S differentially expressed than any other set of eighteen genes * * CA5B; AP1S2; following MVC n l o w C P T P / P T S S s e v e r it y CTPS2; RBBP7; thirteen Emergency Departments (EDs) after MVCo who did not have a i 1 0 0 0 0 0

s h i g h C P T P / P T S S s e v e r it y KAL1; STS; PNPLA4; KDM5C; IQSEC2; (Table 2).

s e

fracture or require hospital admission were enrolledr . CPTP was assessed NLGN4X; ARSH; SMC1A; FUNDC1; p

x 9 0 0 0 0 PRKX; CXorf28; UBA1; INE1; • Twelve of the eighteen genes that were positively correlated E at 20 body regions on a 0-10 Numeric Rating Scale (NRS) and PTSS was FIGURE 4. Of known escapee genes, eighteen FIGURE 5. Twelve of eighteen X chromosome genes A GYG2; ARSD; MXRA5 KDM6A; CDK16; N are positively correlated with XIST in MVC study that are positively correlated with XIST in women with XIST were expressed at significantly higher levels in assessed using the Impact of Events Scale-RevisedR 6 weeks, 6 months, 8 0 0 0 0

CXorf38; T participants (example correlations shown S following MVC, are expressed at higher levels in and 1 year after MVC. I DDX3X; USP9X women that developed CPTP and PTSS than women who X below), suggesting potential escape from XCI. 7 0 0 0 0 women who develop CPTP and/or PTSS than in Blood was collected in RNA PAXgene tubes in the ED andC P TtotalP RNAP TwasS S those who recover. Three examples of these gene recovered (p<0.05) following MVC (Figure 5). 4 0 0 0

n transcripts are shown below.

isolated using PAXgene blood miRNA kits (PreAnalytix). Total RNA was r value: 0.404 o

i p-value: 0.001

s s prepared for sequencing using Ovation Human Blood RNA seq kits 4 0 0 0 4 0 0 0 e 3 0 0 0

r CONCLUSIONS

o x RPS4X; JPX i

(NuGen). Sample libraries were sequenced on a HiSeq2500 system i

s s

e 3 0 0 0 3 0 0 0

e e

r These data suggest that XIST RNA and related X-chromosome A

2 0 0 0 r

N x

(Illumina). Raw sequencing reads were aligned to hg19 using STAR, x e

e 2 0 0 0 2 0 0 0

A transcript levels predict CPTP and PTSS in women experiencing

N N

quantified using RSEM, and normalized to the overall upper quartile. X

R R

K 1 0 0 0

1 0 0 0 1 0 0 0

R X

X MVC. Further studies are needed to replicate these findings and

K R

Repeated measures logistic regression analyses adjusted for age, study R P 5 0 0 0 0 1 0 0 0 0 0 1 5 0 0 0 0 P 0 0 l o w C P T P H i g h C P T P L o w P T S S H i g h P T S S examine potential mechanisms. X IS T R N A e x p re s s io n site, and time following MVC were used to evaluate the relationship NAP1L3; COX7B; 3 0 0 0 0

n P2RY10; GPR174 o

between XIST RNA expression levels and CPTP or PTSS severity. Mean i

s REFERENCES r value: 0.481 s 2 5 0 0 0 4 0 0 0 0 4 0 0 0 0 e p-value: <0.001

levels of expression in individuals with these outcomes versus those who r TAF7L; NXF5 1. Penny GD, Kay GF, Sheardown SA, Rastan S, Brockdorff N. Requirement for Xist in X chromosome

i i

x inactivation. Nature 1996; 379(6561): 131-137.

s s

e 2 0 0 0 0 3 0 0 0 0 3 0 0 0 0

s s

recover were derived from marginal means from regression models.

r r

A 2. Balaton BP, Brown, CJ. Escape Artists of the X chromosome. Trends in Genetics 2016; Vol 32 Issue 6

Bivariate analyses were used to determine the Pearson correlation 2 0 0 0 0 2 0 0 0 0 pp348-359.

R 1 5 0 0 0 ZCCHC16;

N N

X 3. Carrel L, Willard HF. X-inactivation profile reveals extensive variability in X-linked gene expression in R HTR2C; R coefficients and p values corresponding to the relationship between XIST 4 1 0 0 0 0 1 0 0 0 0

S females. Nature 2005; 434(7031): 400-404. X

1 0 0 0 0 ALG13 X

S S

R 4. Ji B, Higa KK, Kelsoe JR, Zhou X. Over-expression of XIST, the master gene for X chromosome P

RNA expression and previously identified escapee genes. Genes that R

0 0 P 5 0 0 0 0 1 0 0 0 0 0 1 5 0 0 0 0 R inactivation, in females with major affective disorders. EBioMedicine 2015; 2(8): 909-918. were previously shown to escape XCI were mapped to a schematic of the L o w C P T P H i g h C P T P L o w P T S S H i g h P T S S X I S T R N A e x p r e s s i o n 5. Zhang Y, Morales AC, Jiang M, Zhu Y, Hu L, Urrutia AO et al. Genes that escape X-inactivation in humans SH2D1A

X chromosome using coordinates derived from hg19 genome build. have high intraspecific variability in expression, are associated with mental impairment but are not slow

n o

i evolving. Molecular biology and evolution 2013: mst148. 3 0 0 0 4 0 0 0 4 0 0 0 6 s r value: 0.331

s 6. Wu D, and Smyth, GK. Camera: a competitive gene set test accounting for inter-gene correlation. Nucleic n

Competitive gene set analyses were used to compare whether e

p-value: 0.007 n

s s

p GPR112; 3 0 0 0 3 0 0 0 Acids Research 2012. 10.1093/nar/gks461.

e r

expression levels of genes positively correlated with XIST in the MVC r

e p

VGLL1; p

2 0 0 0 x

A 2 0 0 0 2 0 0 0

CD40LG A

N N

study predict CPTP or PTSS better than any other set of genes, in terms N

Research reported in this publication was supported by NIAMS of the NIH under

X 1 0 0 0 1 0 0 0

A X

of differential expression. This test accounts for inter-gene correlation. All A 1

1 R01-AR060852, The Mayday Fund, and an American Pain Society Future Leaders

F A

1 0 0 0 F

E 1 E 0 statistical analyses were conducted using either SPSS (v24) or R F 0 I in Pain Grant. The content is solely the responsibility of the authors and does not L o w C P T P H ig h C P T P L o w P T S S H ig h P T S S E L1CAM statistical programs (camera {limma}). 5 0 0 0 0 1 0 0 0 0 0 1 5 0 0 0 0 necessarily represent the official views of these funding agencies. X IS T R N A e x p re s s io n