Protein Arginine Methyltransferase 4 (PRMT4) mediates lymphopenia in experimental sepsis.

Yandong Lai1#, Xiuying Li2#, Tiao Li1, Yan Chen3, Chen Long4, Toru Nyunoya2, Kong Chen1, Georgios D.

Kitsios1, Seyed Mehdi Nouraie1, Yingze Zhang1, Bryan J. McVerry5, Janet S. Lee6, Rama K. Mallampalli7#,

and Chunbin Zou8#*.

1: Division of Pulmonary, Allergy, Critical Care Medicine, Department of Medicine, University of

Pittsburgh School of Medicine, Pittsburgh, PA, USA.

2: Division of Pulmonary, Allergy, Critical Care Medicine, Department of Medicine, University of

Pittsburgh School of Medicine, Pittsburgh, PA, USA; Veterans Affairs Pittsburgh Healthcare System,

Pittsburgh, PA, USA.

3: Division of Pulmonary, Asthma, Critical Care Medicine, Department of Medicine, University of

Pittsburgh School of Medicine, Pittsburgh, PA, USA; Current address: Department of Respiratory

diseases, The Second Xiangya Hospital, Central South University, Changsha, Hunan, China.

4. Division of Pulmonary, Asthma, Critical Care Medicine, Department of Medicine, University of

Pittsburgh School of Medicine, Pittsburgh, PA, USA; Current address: Department of Minimally Invasive

Surgery, The Second Xiangya Hospital, Central South University, Changsha, Hunan, China.

5: Division of Pulmonary, Allergy, Critical Care Medicine, Department of Medicine; Vascular Medicine

Institute, School of Medicine; Department of Environmental and Occupational Health, Graduate School

of Public Health, University of Pittsburgh; Pittsburgh, PA;

6: Division of Pulmonary, Allergy, Critical Care Medicine; Vascular Medicine Institute, University of

Pittsburgh, Pittsburgh, PA, USA

7: Division of Pulmonary, Allergy, Critical Care Medicine, Department of Medicine, University of

Pittsburgh School of Medicine, Pittsburgh, PA; Department of Medicine, Ohio State University College of

Medicine, Columbus, OH, USA.

8: Division of Pulmonary, Allergy, Critical Care Medicine, Department of Medicine, School of Medicine;

Department of Environmental and Occupational Health, Graduate School of Public Health, University of

Pittsburgh; Veterans Affairs Pittsburgh Healthcare System, Pittsburgh, PA, USA.

#: those authors equally contributed to the study.

*: To whom correspondence should be addressed:

Chunbin Zou, MD, PhD., Pulmonary, Allergy and Critical Care Medicine, Montefiore UH NW628, 3459

Fifth Avenue, The University of Pittsburgh, Pittsburgh, PA 15213 Tel.: (412)-624-4404; E-mail:

zouc@upmc.edu

mailto:zouc@upmc.edu

The authors declare no interest conflict with this manuscript.

Key words: T cell, cell death, PRMT4, FBXO9, ubiquitin proteasome degradation, immunosuppression,

sepsis.

Running title: Impaired PRMT4 stability causes T cell death.

Abstract

One hallmark of sepsis is a reduced number of lymphocytes, termed lymphopenia, that occurs from

decreased lymphocyte proliferation or increased cell death contributing to immune suppression.

Histone modification enzymes regulate immunity by epigenetically modulating chromatin architecture,

however, the role of these enzymes in lymphopenia remains elusive. In this study, we identified that a

chromatin modulator Protein Arginine N-methyltransferase 4/Coactivator-Associated Arginine

Methyltransferase 1 (PRMT4/ CARM1) that is elevated systemically in septic patients and experimental

sepsis, and is crucial for inducing T-lymphocyte apoptosis. An E3 ubiquitin ligase SCFFBXO9 docks on

PRMT4 via a phosphodegron to ubiquitinate the protein at K228 for ubiquitin proteasomal degradation.

High PRMT4 expression resulted from reduced levels of SCFFBXO9 that led to increased lymphocyte cell

death after Escherichia coli or lipopolysaccharide (LPS) exposure. Ectopic expression of PRMT4 protein

caused substantial lymphocyte death via caspase 3 mediated cell death signaling, and knockout of

PRMT4 abolished LPS mediated lymphocyte cell death. PRMT4 inhibition with a small molecule

compound attenuated lymphocyte death in complementary models of sepsis. These findings

demonstrate a previously uncharacterized role of a key chromatin modulator in lymphocyte survival that

may shed light on devising unique therapeutic modalities to lessen severity of septic

immunosuppression.

Introduction

Immunosuppression, i.e. “immune paralysis”, is one sepsis hallmark characterized by inhibited immune

responses to microbial infection that also occurs in chronic infection and cancer. Typical features of

immunosuppression include a reduced population of immune cells, exhaustion and dysfunction of cell

immunity, suppression of pro-inflammatory factor release, or an increase in anti-proinflammatory factor

release (1-5). Immunosuppression is clinically manifested in patients by chronic or recurrent bacterial

and viral infections, or acute infections with opportunistic pathogens leading to sepsis (6-10).

Lymphopenia has emerged as a prominent feature in septic patients that renders a poor prognosis. So

far, studies have revealed that multichannel signal pathways including programmed cell death 1 (PD1),

tumor necrosis factor receptor (TNFR), IL-7, and IL-10 may be important factors in septic

immunosuppression (1, 3, 6, 11-15). Gram-negative bacteria derived LPS can induce apoptosis in NK

cells, B cells, T cells, and dendritic progenitor cells (16). Further, LPS-tolerance remarkably contributes

to mortality in polymicrobial septic models (17, 18). In addition, selective reprogramming of gene

expression substantially affects survival rates in the LPS tolerance model (19-21). Yet, the underlying

molecular mechanism(s) in septic immunosuppression remains to be defined.

Epigenetics governs DNA accessibility and concomitantly coordinates with transcriptional factors to

control lymphocyte development, lineage differentiation, and maturation (22-27). Epigenetic alterations

occur in multi-organ failure, animal models of sepsis, and in critically ill patients (28-32). Bacterial

infection regulates behavior of epigenetic enzymes to reprogram host immune defenses (33-36).

Although epigenetics is crucial to the pathogenesis of immunosuppression as blockade of programmed

cell death protein 1 (PD1) does not reverse the epigenetic changes in exhausted T cells in chronic viral

infection animal models suggesting a role for other effectors that may be linked to the pathobiology of

immunosuppression (37). Interestingly, mRNA levels of several histone deacetylases are increased in

preclinical sepsis models (38). Histone H3K27 methylation has been previously implicated as an

epigenetic mark in septic immunosuppression (39). Nevertheless, our understanding of epigenetic

regulation in sepsis is limited as mechanisms are poorly described.

PRMT4, a type I methyltransferase enzyme belonging to at least eleven members in the protein arginine

methyltransferase family, regulates crucial life processes including gene transcription, proliferation, RNA

splicing, and development (40-44). PRMT4 catalyzes the addition of mono- or asymmetric arginine

residues to the guanidino nitrogens of arginyl residues in histones and non-histone protein substrates.

Coordination with the action of histone acetyltransferases p300 and p160, PRMT4 dependent

asymmetric dimethylation of histone H3 at Arg-17 (H3R17me2a) results in transcriptional activation

(45). PRMT4 proteins are aberrantly expressed in a number of cancer cells including breast, colorectal,

and prostate cancers, and its high level is correlated with a poor prognosis in breast cancer patients (46).

PRMT4 promotes cellular proliferation by binding to the promoter region to enhance cell cycle related

CDK2 gene transcription in breast and prostate cancer cells (47). PRMT4 catalyzes methylation of a

range of non-histone proteins including p300 to exert its diverse biological functions in NF-B mediated

inflammation, in p53 related signal transduction, and in the mRNA maturation process (41). A recent

study revealed that PRMT4 catalyzes methylation of more than 130 substrates in breast cancer cells

(48). Knockout of PRMT4 in mice leads to neonatal death and developmental defects in the respiratory

system, reduced percentage of CD4-CD8 double negative T cells, and a block of thymocyte development

in mice (49-51). However, how bacterial pathogens reprogram PRMT4 at the protein level to regulate

lymphocytes is not fully understood.

The ubiquitin proteasome system degrades most cellular proteins in living cells, and at times in

coordination with the gene transcriptional machinery. Ubiquitination proceeds in an enzymatic cascade,

in which the final step involves an E3 ubiquitin ligase that recognizes a protein substrate to catalyze

ligation of ubiquitin to the substrate. This process often results in substrate disposal within the

proteasome. Among E3 ubiquitin ligases, a family of SCF (Skp1-Cullin1-Fbox) E3 proteins have been

linked to bacterial infection and host defense (26, 34, 35, 52-54). Interestingly, the F-box protein FBXO9,

a component of SCF-FBXO9 E3 ligase complex, was identified in a pooled RNAi screen as LPS-responsive in

the context of IL-6 production to regulate innate immunity (55). FBXO9 may be associated with patient

survival in multiple myeloma, is linked to adipocyte development, and targets peroxisome proliferator-

activated receptor gamma for degradation (56-59). Aside from these studies, our understanding of this

E3 ligase subunit is limited, particularly as it relates to sepsis.

In this study, we identified that microbial factors (i.e. endotoxin) increase PRMT4 expression, by

extending its cellular protein lifespan thereby inducing lymphocyte death and modulate host survival in

experimental sepsis. Our data show that bacterial pathogens down-regulate expression of an E3

ubiquitin ligase, SCF-FBXO9 that targets PRMT4 for its elimination in cells. Further, we observe that PRMT4

exerts an undescribed function by triggering lymphocyte death and by screening putative small molecule

inhibitors of PRMT4, we identified that one compound, TP064, specifically attenuates lymphocyte cell

death and protects mice after LPS lung injury and in polymicrobial sepsis.

Results

Gram-negative bacterial derived endotoxin increases PRMT4 protein expression.

We identified that E. coli derived LPS increased PRMT4 protein levels in human lymphoma Jurkat T cells

(Fig. 1A). LPS-mediated PRMT4 accumulation appeared in a concentration-dependent manner in the

cells. PRMT4 mRNA levels as assayed by quantitative reverse transcription- polymerase chain reaction

(qRT-PCR) showed no remarkable differences in LPS treated and untreated Jurkat cells (Fig. 1B). Live E.

coli increased PRMT4 cellular concentrations in Jurkat cells (Fig. 1C). To assess clinical relevance, we

observed that PRMT4 was barely detected in the de-identified human lung tissue samples, but protein

levels were higher in infected lung tissue samples (Fig. 1D). Further, when we conducted PRMT4

immunoblotting in human peripheral blood leukocytes, PRMT4 protein levels from septic patients were

substantially higher compared to that of normal leukocytes (Fig. 1E). Prmt4 protein levels were

remarkably higher in polymicrobial infected mouse sera as compared to that of untreated control mice

(Fig. 1F). Finally, PRMT4 protein levels were substantially higher in plasma from septic patients as

compared to that without sepsis (Fig. 1G, Table 1). Hence, these data suggest that PRMT4 expression is

induced in human lymphoma Jurkat cells in response to bacterial pathogen and PRMT4 levels are

increased during sepsis in vivo. In addition, bacteria elevate PRMT4 protein levels possibly by impairing

its native pathway for protein degradation.

PRMT4 is a labile protein degraded by the SCFFBXO9 E3 ubiquitin ligase.

The availability of a specific protein in the cells is the result of dynamic homeostasis between its gene

transcription rate, translational efficiency, and rate of protein decay. Our published studies show that

PRMT4 is a labile protein (t½ ≈ 3.5 h) degraded via the ubiquitin proteasomal pathway in lung epithelial

cells (60). However, little is known about the molecular mechanism of PRMT4 protein degradation.

Results show that the half-life of PRMT4 protein is approximately 3.5 h in Jurkat T cells (Fig. 2A, B).

PRMT4 degradation was via the proteasome, but not the lysosome, as the proteasome inhibitor MG132

but not the lysosome inhibitor leupeptin accumulated PRMT4. Proteins degraded via the proteasome

are in general poly-ubiquitinated and ubiquitin could be a limiting factor for protein degradation in a

specified cellular compartment. Ectopic expression of ubiquitin accelerated PRMT4 turnover in a

ubiquitin-dependent manner (Fig.2C). As shown before in epithelia (60), results from

immunoprecipitation (IP) studies showed that PRMT4 was also poly-ubiquitinated in lymphocytes (Fig.

2D, upper panel). By screening a library of SCF-E3 ubiquitin ligases, we identified that SCFFBXO9 specifically

degrades PRMT4 (Fig. 2E). Further, PRMT4 protein was degraded in an FBXO9 plasmid concentration

dependent manner (Fig. 2F, upper panels). As a control, FBXO24, another F-box protein, did not degrade

PRMT4 (Fig. 2F, lower panels). IP studies showed that FBXO9 protein associates with PRMT4, as do the

SCF complex components Skp1 and Cullin1 (Fig. 2G). Moreover, knockdown (Kd) of FBXO9 with small

hairpin RNA (shRNA) stabilized PRMT4 as compared with that of a scramble shRNA control (Fig. 2H).

Overall, these data indicate that PRMT4 is specifically targeted for its cellular disposal by the SCFFBXO9

ubiquitin proteasomal machinery.

SCFFBXO9 ubiquitinates PRMT4 at K228.

We next fine mapped the ubiquitination site (s) within PRMT4. We generated and expressed truncated

PRMT4 constructs in cells to examine protein degradation of the truncations (Fig. 3A). Cellular

expression of constructs harboring carboxyl terminal deletions resulted in PRMT4 accumulation

suggesting that molecular signatures that mediate protein disposal are not localized within the

fragments (Fig. 3B, left panels). In contrast, deletion of the NH2-terminal 150 amino acids (aa) resulted in

lack of PRMT4 accumulation (Fig. 3B, right panels), suggesting that aa100-150 contains a degradation

element. We analyzed the sequence and found that there was no ubiquitin acceptor lysine residue in

the aa100-150, but a nearby lysine residue K228 resided in the truncation mutant containing aa100-608.

We tested if K228 is a ubiquitin acceptor site. As compared with that of wild type (WT) PRMT4 and other

mutants, a K228A mutant was stable from proteasomal degradation (Fig. 3C, D). The K228A mutant was less

ubiquitinated as compared with that of WT PRMT4 (Fig. 3E). In addition, mutation of K228 abrogated

FBXO9 mediated PRMT4 degradation (Fig. 3F). These data suggest that K228 may be one authentic

SCFFBXO9 catalyzed ubiquitin acceptor site.

SCFFBXO9 recognizes a PRMT4 phosphodegron

As SCF E3 ubiquitin ligases often interact with phosphorylated substrates, we next studied the FBXO9

docking site within PRMT4. We focused on the NH2-terminus because the above studies suggested that

aa100-150 contains a degradation element, possibly an E3 ligase docking site. Results from in vitro pull-

down assays showed that deletion of NH2-terminal 150 aa impairs PRMT4 and FBXO9 binding (Fig. 4A).

We further fine mapped the docking site for FBXO9 within PRMT4 by pull-down assays (Fig. 4B, C). Pull

down results showed that amino acids after aa131 were critical for FBXO9 binding. We analyzed the

sequence and identified a phosphorylation motif in this region: 132-TxxxS (Fig. 4D, upper panel, x

donates any amino acid). To test if this motif is a potential phosphodegron, we substituted T132 and S136

with a cysteine to mimic dephosphorylation or introduced an aspartic acid at T132 to mimic a

phosphorylation site. Pull down results suggest that a dephosphorylated T132 and phosphorylated S136

sites constituted an optimal FBXO9 binding motif constituting a phosphodegron for E3 SCF ligase

interaction (Fig. 4D, E). Consistent with these observations, our recent published study showed that

glycogen synthase kinase beta (GSK-3β)-mediated T132 phosphorylation stabilized PRMT4 (60). We

further tested the protein stability with these mutations. T132C and S136D mutants degraded faster than

that of WT PRMT4. A S136C mutant displayed greater protein stability. T132D and T132AS136A double mutants

completely stabilized the protein from degradation. However, a T132CS136D PRMT4 variant when

expressed in cells degraded most rapidly (Fig. 4F, G). In all, these data indicate that SCFFBXO9 engages a

phosphodegron 132-TdxxxSp to ubiquitinate PRMT4 at K228.

LPS induced PRMT4 promotes caspase 3 activation in lymphocytes.

We next assessed a potential pathophysiological role for elevated PRMT4 after bacterial infection.

PRMT4 appeared to be endotoxin-responsive, as LPS increased PRMT4 protein levels in T cell-like Jurkat

cells, in human infected lung tissues, and in human peripheral blood leukocytes and plasma from septic

patients (Fig. 1). PRMT family members regulate apoptosis as PRMT2 inhibits NF-B signaling and

promotes apoptosis in mouse embryo fibroblasts (61). PRMT4 expression induced by high-glucose

loading triggers apoptosis of human retinal pigment epithelial cells via H3R17 dimethylation (62) Thus,

we tested if actions of LPS on lymphocyte death are mediated by PRMT4 through its E3 ligase, SCFFBXO9.

We first studied cell death using activated caspase 3 as a marker in a variety of cell types including Jurkat

cells, mouse splenic T cells, and human peripheral blood pan-T cells after 6 h of LPS stimulation. As

predicted, caspase 3 was activated in all of these cells, and its activation occurred in an LPS

concentration dependent manner (Fig. 5A-C, second upper panels). Notably, LPS down-regulated FBXO9

protein variably in all the tested cells, suggesting a widespread (shared) mechanism whereby LPS

stabilizes PRMT4 in cells via depletion of SCFFBXO9 E3 ubiquitin ligase (Fig. 5A-C, second lower panels). To

understand the potential role(s) of PRMT4 in caspase 3 cleavage in T cells, we ectopically expressed

PRMT4 in Jurkat cells. Immunoblotting results showed that PRMT4 ectopic expression was sufficient to

activate caspase 3 in Jurkat cells (Fig. 5D). Caspase 3 activation occurred in a PRMT4 plasmid

concentration-dependent manner. We then generated PRMT4 knockout (KO) Jurkat cells using

CRISPR/Cas9 to test our observations. Immunoblotting results showed that PRMT4 was completely

depleted in stable KO cells (Fig. 5E). KO of PRMT4 blocked LPS-mediated caspase 3 activation to baseline

levels (Fig. 5F). In contrast, ectopic expression of PRMT4 promoted LPS-induced caspase 3 activation

(Fig. 5G). Results from immunoblotting analysis of the isolated mouse splenic T-cells showed that

expression of PRMT4 promotes LPS-induced caspase 3 activation, and depletion of PRMT4 by shRNA

substantially blocks LPS-induced caspase 3 activation (Fig. 5H). To confirm the role of FBXO9 in PRMT4

F

i

g

.

5

.

L

P

S

d

o

w

mediated caspase 3 activation, we depleted FBXO9 in Jurkat cells by shRNA. Immunoblotting results

showed that depletion of FBXO9 increased PRMT4 mediated caspase 3 activation and this activation is

ablated in PRMT4 KO cells (Fig. 5I). Thus, these data suggest that LPS increases PRMT4 expression

possibly via down-regulation of FBXO9, thereby facilitating the methyltransferase to trigger caspase 3

signaling in lymphocytes.

PRMT4 depletion or chemical inhibition reduces lymphocyte cell death.

To explore the function of LPS-mediated PRMT4 elevation and subsequent caspase 3 activation, we

measured lymphocyte apoptosis. We conducted Annexin V flow-cytometry to confirm the role of PRMT4

in LPS-mediated early stage T cell death. Flow-cytometry results showed that the baseline levels of cell

death were similar in PRMT4 KO Jurkat cells (8.19% and 8.03%) as compared to that of non-targeted

cells (7.86%). LPS treatment substantially promoted cell death in non-targeted cells (26.65%) but not in

PRMT4 KO cells (10.47% and 8. 31%). Interestingly, PRMT4 overexpression promoted Jurkat cell death

(19.75%) and augmented LPS-induced cell death (41.78%) (Fig. 6A, B). To confirm this observation, we

determined the cell viability in LPS treated mouse splenic lymphocytes. As predicted, LPS did not affect

cell survival in PRMT4 depleted mouse splenic T-cells but remarkably reduced cell survival in PRMT4

overexpressed mouse splenic T-cells (Fig. 6C). Since LPS increases PRMT4 protein levels that induces T

cell death, we assessed if a PRMT4 small molecule inhibitor(s) might suppress T cell death. We first

screened commercially available PRMT4 inhibitors in LPS treated Jurkat cells to test if the inhibitor(s)

might inhibit caspase 3 activation. Among the inhibitors screened, a selective and cell active PRMT4

inhibitor TP064 (MedKoo Biosciences, Inc. Cat#: 407418) efficiently and most consistently inhibited

caspase 3 activation at a concentration of ~40 nM (Fig. 6D); TP064 concentration of ~ 10 nM effectively

inhibited PRMT4 mediated histone H3 arginine methylation data not shown). Further, experimental

results suggest that TP064 can block LPS mediated caspase 3 activation at both 25 nM and 50 nM in

mouse splenic T cells (Fig. 6E). These studies link PRMT4 to endotoxin-induced lymphocyte death and

showcase the ability of a PRMT4 specific small molecule inhibitor to effectively antagonize endotoxin-

induced lymphocyte death signaling.

PRMT4 mediates splenic lymphocyte death in experimental acute lung injury.

To test our hypothesis in animal models, we assessed PRMT4 in an LPS lung injury mice model. We

overexpressed PRMT4 or knocked down PRMT4 using lentiviral constructs (1×107 cfu/mouse,

intratracheal [i.t.] administration) in mice (C57BL/6J) for 14 d. LPS (5 mg/kg, i.t. in PBS buffer) was

administrated (i.t.) into mice for 24 h. Prior to the lentivirus administration, we administered a droplet

of lysoPC (20 L, diluted 10 times in PBS, i.t.) to facilitate lentiviral particle migration through the

surfactant layer and cell membrane and promote efficiency of gene transfer. Results from histological

studies showed that administration of LPS caused lung inflammation with inflammatory cell infiltration.

Overexpression of PRMT4 enhanced lung inflammation and knockdown of PRMT4 or application of

PRMT4 inhibitor attenuated lung inflammation (Fig. 7A, upper panels). Results from TUNEL staining

showed that LPS caused cell death in lung tissues, in which, overexpression of PRMT4 enhanced cell

death and knockdown of PRMT4 or application of PRMT4 inhibitor attenuated cell death in the lung (Fig.

7A, lower panels, Fig. 7B). Immunoblotting results from lung indicated that i.t. administration of LPS or

lentiviral particles did modulate PRMT4 protein levels in lung tissue (Fig. 7C). Surprisingly, we observed

that LPS induced substantial cell death in the spleen as evidenced by TUNEL staining. Splenic cell death

occurred in the white pulp, a zone that T cells reside (Fig. 7D, E). Further, overexpression of PRMT4

markedly enhanced LPS-induced lymphocyte death whereas silencing of PRMT4 by shRNA significantly

diminished cell death. TP064 partially protected splenic lymphocytes from death. Lung infection can

serve as a reservoir of microbial products that can disseminate systemically. Thus, we speculated that i.t.

administration of LPS might access the systemic circulation to disseminate to major organs such as the

spleen. We analyzed splenic cell death with flow cytometry and identified that overexpressed PRMT4

enhanced cell death in CD4+ T cells (Fig. 7F, G). In addition, results from survival studies indicated that

PRMT4 overexpressed mice showed a significantly lower survival rate. Depletion of PRMT4 by shRNA or

application of TP064 remarkably improved mouse survival (Fig. 7H). These data suggest that LPS-

enhanced PRMT4 protein expression promotes lymphocyte death and targeting on PRMT4 improves

survival in a LPS mouse lung injury model.

PRMT4 mediates splenic lymphocyte death in a polymicrobial sepsis model.

We next tested our hypothesis in a well characterized polymicrobial septic mouse model, Cecal ligation

and puncture (CLP). Mice were subjected to lenti-PRMT4 overexpression and lenti-shRNA as above, CLP

procedure were followed to induce sepsis (63) in animals. One group of mice receiving lenti-vector only

were given TP064 (0.2 g/mouse) intravenously following CLP. Consistent with above observations,

results from TUNEL staining of the splenic tissues showed that CLP was sufficient to cause splenic T cell

death (Fig. 8A, B). Flow cytometric data confirmed these observations (Fig. 8C, D). Moreover, PRMT4

overexpression mediated splenic CD4+ T cell death in this model whereas silencing of PRMT4 by shRNA

or administration of TP064 attenuated splenic T cell death. Lenti-PRMT4 overexpression during CLP

increased IL-1ra, IL-6, IL-10, IL-17, CCL1-, CXCL1, CXCL9, and CXCL10 levels in mouse sera when

compared to the CLP group (p

stabilized PRMT4 that mediates lymphocyte apoptosis; and (iii) chemical inhibition or genetic depletion

of PRMT4 attenuates lymphocyte death and protects mice after endotoxin-induced lung injury and

polymicrobial sepsis (Fig. 9). Dysregulation of PRMT4 has been reported in many pathophysiological

settings. PRMT4 is aberrantly expressed in breast, prostate, and colorectal cancers that is associated

with a poor prognosis by promoting tumor progress and cancer metastasis. Consistent with these

observations, we identified that PRMT4 expression is increased in sepsis. PRMT4 protein was

upregulated in cellular and mouse septic models, in human infected lung tissue samples, and in

peripheral blood leukocytes from septic patients. Increased PRMT4 protein caused lymphocyte

apoptosis and this finding may be one mechanism underlying the pathogenesis of septic

immunosuppression, as apoptosis of T lymphocytes is a key feature of sepsis-induced

immunosuppression (64) and lymphopenia that develops during sepsis predicts early and late mortality

(65).

PRMT4 is crucial in lymphocyte lineage differentiation and maturation (49-51). Knockout of PRMT4 in

mice results in less CD4+/CD8+ cells and prevents thymocyte development (provide citation here).

However, these global defects in lymphocyte development prevent assessment of PRMT4 function

during pathophysiologic states when PRMT4 may achieve a high level beyond a physiologic threshold,

and where the gene product may exert functions other than uncovered in a knockout mouse model. It is

not uncommon to observe that some proteins display unexpected functions in the context of aberrant

expression, or low concentrations, or total cellular depletion. For instance, both knockdown and

overexpressed acyl-CoA:lysocardiolipin acyltransferase-1 harms mitochondrial function, with distinct

mechanisms (66). While the significance of the aberrantly expressed PRMT4 in specific disorders has

not been fully characterized, PRMT4 expression in breast cancer is associated with poor prognosis by

promoting tumor progression and cancer metastasis (67). Thus, reduced tumor free survival in breast

cancer may be linked to a similar function posited for the methyltransferase in immunosuppression.

Our studies uncover a unique role for PRMT4 in cell death linked to sepsis. Increased cell death has long

been noticed as a key pathway in sepsis-induced immunosuppression but the factors leading to this

increased cell death has not been fully defined. Immune checkpoint proteins including PD1/PDL1 are

believed to be one of the major pathways that contribute to septic immunosuppression (68). However,

efficacy of anti-PD-L1 application is limited to a portion of patients of whom PD1/PD-L1 signaling is

positive and the efficacy of anti-PDL1 application in septic models remains to be determined. Further,

PD-L1 antibody administration does not reverse the epigenetic changes of immunosuppression in animal

models. Our findings here add a novel mechanism of immunosuppression at the epigenetic level: the

potential for a pathogen-derived product to induce cellular concentrations of chromatin modulator,

PRMT4, that in turn mediates lymphocyte death via caspase 3 related apoptosis, thus serving as a

potential modulator in sepsis. Further studies are indicated to elucidate the molecular mechanisms

whereby PRMT4 regulates apoptosis, epigenetically or via its methylation function driving post-

translational modifications of key regulatory proteins that dictate cellular lifespan. Recent mass

spectrometric studies uncovered more than 130 PRMT4 modified protein substrates in breast cancer

cell lines (69). These targets might be suitable candidates for interrogation of post-translational

modifications by PRMT4 associated with T-cell immunosuppression.

Here we identified a new molecular target of FBXO9 in its control of PRMT4 protein stability that

regulates immune responses through control of T cell lifespan. Before PRMT4 is targeted for

ubiquitination and subsequent degradation, PRMT4 is recognized by the E3 ubiquitin ligase SCFFBXO9

through a phosphodegron. In this motif (132-TdxxxSp), the phosphorylation status of serine and

threonine residues appear to be differentially modified for optimal FBXO9 binding and subsequent

ubiquitination. Phosphodegrons have been widely cited in governing the stability of crucial molecules in

the pathogenesis of diseases. In this regard, we previously identified that oxidative stress reduces

PRMT4 levels via downregulation of a protein kinase GSK3. GSK3 phosphorylates T132 that is crucial

in regulating the affinity between the phosphodegron and the E3 ubiquitin ligase (60). Interestingly, in

the present study, we observed that LPS downregulates the E3 ubiquitin ligase component FBXO9 at the

protein level to promote PRMT4 protein stability.

We observed higher plasma PRMT4 levels in septic patients compared to non-septic patients. PRMT4 is

a protein arginine methyltransferase located in both the cytoplasmic and nuclear compartments (cite

references here). Our results suggest that PRMT4 is highly expressed in leukocytes isolated from septic

patients and PRMT4 is detectable in mouse sera and human plasma. Considering the fact that high levels

of PRMT4 are cytotoxic, we speculate that PRMT4 in plasma of septic patients may originate from dying

or dead cells. In addition, we identified that plasma PRMT4 protein levels were positively correlated

with IL-6 (r=0.37, p=0.006) in septic patients (Fig. 8G). This is consistent with our observation in the

polymicrobial infection mouse model with PRMT4 overexpression, in which IL-6 expression is increased.

Interestingly, FBXO9 is known to control IL-6 production to regulate innate immunity in a LPS cellular

model (55). PRMT4-mediated histone arginine methylation is a transcriptional activation epigenetic

marker (please cite reference). Our data invites the possibility that PRMT4 may enhance IL-6 secretion

possibly through loss of FBXO9 control during endotoxin-mediated cellular stress. Interestingly,

inhibition of PRMT4 by shRNA or with the small molecule inhibitor improved mice survival in both the

LPS lung injury and CLP mouse models, suggesting that lymphocyte death may be a critical factor

underlying the pathogenesis of endotoxin-mediated injury. Therefore, pharmaceutical targeting of

PRMT4 may be a potential alternative therapeutic approach in septic patients with significantly low T

cell counts and immunosuppression.

Materials and Methods

Cell lines and reagents. Human lymphoma Jurkat cells (ATCC) and human primary pan-T cells (StemCell)

were cultured with RPMI1640 (Gibco) containing 10% FBS. Cells were maintained in a 37 °C incubator in

the presence of 5% CO2. V5 antibody, pcDNA3.1D-His-V5-TOPO cloning kit (Cat#: K490001) and

Escherichia coli Top10 competent cells (Cat#: C404006) were from Invitrogen (St. Louis, MO). PRMT4

(Cat#:12495) and Hemagglutinin (HA) tag (Cat#: 3724S, lot: 5) antibody were from Cell Signaling

(Danvers, MA). Ubiquitin (Cat#: sc-166553, lot: A2710) antibody was from Santa Cruz Biotechnology

(Santa Cruz, CA). FBXO9 (Cat#: PA5-23474) antibody was from Thermo (Rockford, IL). Alexa Fluor 647 Rat

Anti-Mouse CD8a and BV421 Rat Anti-Mouse CD45R/B220 were from BD Biosciences (San Jose, CA). The

PRMT4 shRNA was from Origene (Rockville, MD). Cycloheximide (Cat#: ALX-380-269-G001, lot:

01061518), E64D (Cat#: BML-PI107-0001, lot: 08181537), and Ubiquitin aldehyde (Cat#: BML-UW8450-

0050, lot: 07021447) were from Enzo Life Sciences (Farmingdale, NY). β-actin (Cat#: A3853) antibody,

bacterial lipopolysaccharide (LPS) from E. coli O111:B4 (Cat#: L4391, lot: 115M4090V) and L-α-

Lysophosphatidylcholine (lysoPC, Cat#: L-5254) were from Sigma (Carlsbad, CA). MG132 (Cat#: F1101,

lot: F11052079) was from UBPBio (Aurora, CO). TP064 (Cat#: 6008) was from Tocris Bioscience (Ellisville,

MO). QuickChange II XL site-directed mutagenesis kits (Cat#: 200522) were from Agilent Technologies

(Santa Clara, CA). TnT Quick Coupled Transcription/Translation Systems (Cat#: L1170) were from

Promega (Madison, WI). Immobilized protein A/G Agarose beads (Cat#: 20421) were from Pierce

(Rockford, IL). All other reagents were of the highest grade available commercially.

Cloning and mutagenesis. V5-tagged PRMT4 truncations were cloned into pcDNA3.1D-His-V5-TOPO

plasmid using PCR-based approaches as previously described (70). Mutagenesis was introduced by using

a QuickChange II XL site-directed mutagenesis kits according to the manufacturer’s instructions. The

accuracy of the mutagenesis was confirmed by sequencing. The primers used in the construction of

PRMT4 truncations and site-directed mutagenesis were listed in Table 3.

Plasmid transfection. All plasmids were introduced into cells using electroporation executed with a

nuclear transfection apparatus (Amaxa Biosystems, Gaithersburg, MD) with a preset program (X-001 for

Jurkat cells), following the manufacturer’s instructions as previously described (66). Briefly, one million

cells in 100 μL of transfection buffer (20mM Hepes in PBS buffer) were mixed with 3 μg of plasmids

(including expression and shRNA constructs). After electroporation, the cells were cultured with 2 mL

conditional medium in 6 well plates for 24 h for further analyses.

Immunoblotting and co-Immunoprecipitation. Immunoblotting and co-immunoprecipitation were

conducted as previously described (66). Briefly, for immunoblotting, whole cell extracts (normalized to

total protein concentration) were resolved by SDS-PAGE and transferred to membranes by

electroblotting. The membranes were blocked with 5% (w/v) non-fat milk in Tris-buffered saline and

probed with primary antibodies as indicated. Membranes were developed by an enhanced

chemiluminescence (ECL) system. For immunoprecipitation, 1 mg of cell lysates (in PBS with 0.5% Tween

20 plus protease inhibitors) were incubated with specific primary antibodies for 2 h at room

temperature. The mixture was added with 35 µl of protein A/G-agarose beads for an additional 2 h at

room temperature. The precipitated complex was washed for three times with 0.5% Tween 20 in PBS

and analyzed by immunoblotting with ECL system.

Human study and PRMT4 ELISA assays. Subjects with or without sepsis were selected from a cohort of

mechanically-ventilated patients at the University of Pittsburgh Medical Center (UPMC). Eligible

patients were 18 years or older with acute respiratory failure requiring mechanical ventilation via

endotracheal intubation with sepsis as defined by the presence or suspicion of infection and two or

more systemic inflammatory response syndrome (SIRS) criteria. Control patients were intubated and

mechanically ventilated for airway protection without sepsis. Deidentified baseline clinical data

(including demographics, comorbidities, physiologic and laboratory variables), severity of illness

(modified sequential organ failure assessment (SOFA) scores), and clinical outcomes: 30- and 90-day

mortality, duration of ICU stay, incidence of acute kidney injury (defined as elevation of 0.3 mg/dL or

more from baseline creatinine) or shock (defined as need for vasopressors) within 7-days from

intubation (10) were extracted from the registry database (Table 2). Deidentified human infectious lung

samples and normal controls were as described previously (71).

Deidentified cytokine data were extracted from the registry database, and deidentified plasma provided

for assay of PRMT4 expression. PRMT4 levels were determined using PRMT4 ELISA kit (Cat#

MBS3244080 for human PRMT4, and Cat# MBS9339704 for mouse Prmt4, Mybiosource, San Diego, SC,

USA) as directed by the manufacturer. Briefly, the plates, reagents and samples were brought to room

temperature. Plasma samples with a volume of 50 μl and 100 μl HRP-Conjugate Reagent were added to

each well of the 96-well plate. The plate was incubated at 37°C for 60 min and washed with Wash

Solution for 4 times. The results were visualized with addition of 50 μl Chromogen Solution A and B and

incubation of the plate at 37°C for another 15 min. The color reaction was stopped with 50 μl of Stop

Solution and the optical density (OD) was read at 450 nm within 15 min.

Real‐Time Polymerase Chain Reaction (RT‐PCR). Total RNA from cells was isolated with RNeasy Mini

Kit (Qiagen, Valencia, CA, USA), and reverse‐transcribed by using High-Capacity RNA-to-cDNA Kit

(Applied Biosystems, Foster City, CA, USA) following the manufacturer's manual instructions. Single

stranded cDNA was then amplified by RT‐PCR with specific primers of PRMT4 and GAPDH. RT‐PCR

was performed on an ABI Prism 7000 thermocycler (Applied Biosystems, Thermo-Fisher Scientific,

Waltham, MA, USA) with SYBR Green PCR Master Mix (Roche, Basel, Switzerland). For each experiment,

samples (n = 3) were run in triplicate. Relative gene expression was calculated using the comparative CT

method.

in vitro Pull-down assay. We conducted in vitro binding assays to identify the FBXO9 binding domain

within PRMT4. V5-tagged PRMT4 truncated or site directed mutant proteins were in vitro expressed

using a TnT coupled reticulocyte system. Endogenous FBXO9 protein was obtained by FBXO9

immunoprecipitation from Jurkat cell lysate. FBXO9-precipitated protein A/G-agarose beads were

incubated with a variety of PRMT4 truncations or mutants for 2 h. The beads were washed extensively

with 0.5% Tween 20 in PBS and analyzed by V5 and FBXO9 immunoblotting.

Lentivirus production. Lentiviruses were generated by Lenti-X Packaging Single Shots VSV-G (Cat#:

631276, Clontech, CA, USA) according to the manufacturer’s procedures. Lentivirus-containing

supernatants were concentrated using Lenti-X Concentrator (Cat#: 631231). Concentrated lentiviruses

were resuspended in PBS and titrated by Lenti-X GoStix (Cat#: 631280). The samples were aliquoted and

stored at -80°C.

Mouse splenic T cell isolation, flowcytometry, and annexin V apoptosis assay. Mouse spleen were

disrupted in PBS containing 2% FBS. The aggregates and debris were removed by passing suspension

through Falcon 70-μm cell strainer (BD, NJ, USA). Mouse splenic T cells were isolated using EasySep

Mouse T Cell Isolation Kit (Cat#: P19851, Stemcell Technologies, Vancouver, Canada) with EasySep

Magnets (Cat#: 18000) according to the instructions. Above isolated mouse splenic T cells were also

stained with anti-CD4, CD8a or CD45R antibodies combined. The cells were incubated for 15 - 30 min at

room temperature in the dark. After adding 400 μl Annexin V binding buffer, the samples were acquired

on the LSRII flow cytometers (BD Biosciences, MI, USA). All cells were analyzed with Flowjo software

(Tree Star, OR, USA).

FITC Annexin V Apoptosis Detection Kit with Propidium Iodide (PI) (Cat#: 640914, BioLegend, CA, USA)

was employed to detect apoptotic cells. Jurkat cells were harvest and washed twice with cold BioLend

Cell Staining Buffer and resuspend in annexin V binding buffer with cell density about 2.5 ~ 10 × 106

cells/ml. After 100 μl of cells were transferred to a 5ml test tube, 5 μl of FITC annexin V and 10 μl of PI

were added. The cells were gently vortexed and incubated for 15 min at room temperature in the dark.

After adding 400 μl Annexin V Binding Buffer, the samples were analyzed through BD Accuri C6 flow

cytometry (BD Biosciences, MI, USA).

Cytokine proteome profiler array assay. Cytokine proteome profile of mouse sera were measured using

Mouse XL Cytokine Array Kit (Cat#: ARY028, R&D system, MN, USA) according to the manufacturer’s

instructions. Briefly, mouse serum samples were incubated overnight with the membrane. The

membrane was washed to remove unbound material followed by incubation with a cocktail of

biotinylated detection antibodies. Streptavidin-HRP and chemiluminescent detection reagents were

then applied, and a signal is produced at each capture spot corresponding to the amount of protein

bound.

Mouse LPS-induced lung Injury and Cecal ligation and puncture (CLP )procedures. LPS-

induced lung Injury model was conducted as previously described. Briefly, C57BL/6J mice at the age of

10 weeks were anesthetized by injecting intraperitoneally a solution of 1:1 ketamine (75mg/kg) and

xylazine (15mg/kg) throughout the experiment. LPS (7mg/kg) were intratreackally administrated and the

mice were observed for 48 h. Polymicrobial sepsis was induced in mice by cecal ligation and

puncture(72). Briefly, mice were anesthetized throughout the experiment. The abdomen was shaved,

and wiped with 70% alcohol. A 1 to 2 cm midline abdominal incision was performed. The cecum was

exposed and ligated with a sterile silk suture 1 cm from the tip and double punctured with a 19-gauge

needle. The cecum was gently squeezed to extrude a small amount of fecal material and was returned

to the peritoneal cavity. The incision was closed with silk sutures. Mice were resuscitated with i.p.

injection of 1 ml of pre-warmed 0.9% saline solution. Mice were then monitored every 12 hours for

survival or euthanized at different time points for analysis of different parameters.

CRISPR/Cas9 PRMT4 Knock-out cell line. Jurkat cells were co-transfected with PRMT4 CRISPR/Cas9 KO

Plasmid (Cat#: sc-404087-KO-2) and HDR Plasmid (Cat#: sc-404087-HDR-2) using UltraCruz Transfection

Reagent (Cat#: sc-395739). Stable KO cells were selected by puromycin following the instruction from

the CRSPR/Cas9 manufacturer. PRMT4 knockout was confirmed by immnoblotting.

Statistics. The data represent the Mean + standard deviation in the graphs depicting the error bars or as

specifically indicated. Prism 7 (GraphPad software, San Diego, CA) was used to determine statistical

significance. Comparisons between groups were made using unpaired, two-tailed Student’s t-test (two

groups) and one-way analysis of variance (ANOVA) with post-hoc Tukey HSD or Bonferroni and Holm

multiple comparisons. Kaplan Meier estimate was used for survival analysis in mouse septic models.

P 

4. T. S. Wang, J. C. Deng, Molecular and cellular aspects of sepsis-induced immunosuppression.

Journal of molecular medicine 86, 495-506 (2008).

5. J. S. Boomer, J. M. Green, R. S. Hotchkiss, The changing immune system in sepsis: is

individualized immuno-modulatory therapy the answer? Virulence 5, 45-56 (2014).

6. O. M. Peck-Palmer et al., Modulation of the Bcl-2 family blocks sepsis-induced depletion of

dendritic cells and macrophages. Shock 31, 359-366 (2009).

7. K. C. Chang et al., Blockade of the negative co-stimulatory molecules PD-1 and CTLA-4 improves

survival in primary and secondary fungal sepsis. Crit Care 17, R85 (2013).

8. D. S. Chen, I. Mellman, Elements of cancer immunity and the cancer-immune set point. Nature

541, 321-330 (2017).

9. S. J. Patel et al., Identification of essential genes for cancer immunotherapy. Nature, (2017).

10. E. Cacan, Epigenetic-mediated immune suppression of positive co-stimulatory molecules in

chemoresistant ovarian cancer cells. Cell Biol Int 41, 328-339 (2017).

11. C. L. Day et al., PD-1 expression on HIV-specific T cells is associated with T-cell exhaustion and

disease progression. Nature 443, 350-354 (2006).

12. S. J. Im et al., Defining CD8+ T cells that provide the proliferative burst after PD-1 therapy.

Nature 537, 417-421 (2016).

13. M. Pellegrini et al., IL-7 engages multiple mechanisms to overcome chronic viral infection and

limit organ pathology. Cell 144, 601-613 (2011).

14. Y. Shindo, J. Unsinger, C. A. Burnham, J. M. Green, R. S. Hotchkiss, Interleukin-7 and anti-

programmed cell death 1 antibody have differing effects to reverse sepsis-induced immunosuppression.

Shock 43, 334-343 (2015).

15. M. Hiramatsu, R. S. Hotchkiss, I. E. Karl, T. G. Buchman, Cecal ligation and puncture (CLP) induces

apoptosis in thymus, spleen, lung, and gut by an endotoxin and TNF-independent pathway. Shock 7,

247-253 (1997).

16. J. S. Nielsen et al., Rough-Form Lipopolysaccharide Increases Apoptosis in Human CD4(+) and

CD8(+) T Lymphocytes. Scand J Immunol 75, 193-202 (2012).

17. K. Kopanakis et al., Pre-treatment with low-dose endotoxin prolongs survival from experimental

lethal endotoxic shock: Benefit for lethal peritonitis by Escherichia coli. Cytokine 62, 382-388 (2013).

18. K. Chen et al., Super-low dose endotoxin pre-conditioning exacerbates sepsis mortality.

EBioMedicine 2, 324-333 (2015).

19. E. S. Melo et al., Gene expression reprogramming protects macrophage from septic-induced cell

death. Mol Immunol 47, 2587-2593 (2010).

20. E. S. Melo et al., Endotoxin tolerance: selective alterations in gene expression and protection

against lymphocyte death. Immunobiology 215, 435-442 (2010).

21. B. Novakovic et al., beta-Glucan Reverses the Epigenetic State of LPS-Induced Immunological

Tolerance. Cell 167, 1354-1368 e1314 (2016).

22. L. Wang et al., Histone acetyltransferase activity is conserved between yeast and human GCN5

and is required for complementation of growth and transcriptional activation. Molecular and cellular

biology 17, 519-527 (1997).

23. H. H. Qi et al., Histone H4K20/H3K9 demethylase PHF8 regulates zebrafish brain and craniofacial

development. Nature 466, 503-507 (2010).

24. R. Pavri et al., Histone H2B monoubiquitination functions cooperatively with FACT to regulate

elongation by RNA polymerase II. Cell 125, 703-717 (2006).

25. S. J. Nowak, V. G. Corces, Phosphorylation of histone H3: a balancing act between chromosome

condensation and transcriptional activation. Trends in genetics : TIG 20, 214-220 (2004).

26. B. B. Chen et al., F-box protein FBXL2 targets cyclin D2 for ubiquitination and degradation to

inhibit leukemic cell proliferation. Blood 119, 3132-3141 (2012).

27. M. Okada, S. Hibino, K. Someya, A. Yoshmura, Regulation of regulatory T cells: epigenetics and

plasticity. Adv Immunol 124, 249-273 (2014).

28. S. W. Kang et al., Renal kallikrein excretion and epigenetics in human acute kidney injury:

expression, mechanisms and consequences. BMC nephrology 12, 27 (2011).

29. H. F. Li, C. F. Cheng, W. J. Liao, H. Lin, R. B. Yang, ATF3-mediated epigenetic regulation protects

against acute kidney injury. Journal of the American Society of Nephrology : JASN 21, 1003-1013 (2010).

30. W. F. Carson, K. A. Cavassani, Y. Dou, S. L. Kunkel, Epigenetic regulation of immune cell functions

during post-septic immunosuppression. Epigenetics 6, 273-283 (2011).

31. Y. Liu, H. Li, T. Xiao, Q. Lu, Epigenetics in immune-mediated pulmonary diseases. Clinical reviews

in allergy & immunology 45, 314-330 (2013).

32. M. Takebe et al., Inhibition of histone deacetylases protects septic mice from lung and splenic

apoptosis. J Surg Res 187, 559-570 (2014).

33. B. Ellis, L. Kaercher, C. Snavely, Y. Zhao, C. Zou, Lipopolysaccharide triggers nuclear import of

Lpcat1 to regulate inducible gene expression in lung epithelia. World journal of biological chemistry 3,

159-166 (2012).

34. C. Zou et al., SCF(Fbxw15) mediates histone acetyltransferase binding to origin recognition

complex (HBO1) ubiquitin-proteasomal degradation to regulate cell proliferation. The Journal of

biological chemistry 288, 6306-6316 (2013).

35. C. Zou et al., LPS impairs phospholipid synthesis by triggering beta-transducin repeat-containing

protein (beta-TrCP)-mediated polyubiquitination and degradation of the surfactant enzyme acyl-

CoA:lysophosphatidylcholine acyltransferase I (LPCAT1). The Journal of biological chemistry 286, 2719-

2727 (2011).

36. C. Zou et al., Mortality factor 4 like 1 protein mediates epithelial cell death in a mouse model of

pneumonia. Science translational medicine 7, 311ra171 (2015).

37. R. S. Hotchkiss, G. Monneret, D. Payen, Sepsis-induced immunosuppression: from cellular

dysfunctions to immunotherapy. Nature Reviews Immunology 13, 862 (2013).

38. H. T. Aung et al., LPS regulates proinflammatory gene expression in macrophages by altering

histone deacetylase expression. FASEB J 20, 1315-1327 (2006).

39. N. D. Das, K. H. Jung, Y. G. Chai, The role of NF-kappaB and H3K27me3 demethylase, Jmjd3, on

the anthrax lethal toxin tolerance of RAW 264.7 cells. PLoS One 5, e9913 (2010).

40. D. Cheng, J. Cote, S. Shaaban, M. T. Bedford, The arginine methyltransferase CARM1 regulates

the coupling of transcription and mRNA processing. Molecular cell 25, 71-83 (2007).

41. M. Covic et al., Arginine methyltransferase CARM1 is a promoter-specific regulator of NF-

kappaB-dependent gene expression. The EMBO journal 24, 85-96 (2005).

42. N. Yadav et al., Specific protein methylation defects and gene expression perturbations in

coactivator-associated arginine methyltransferase 1-deficient mice. Proc Natl Acad Sci U S A 100, 6464-

6468 (2003).

43. M. A. Kleinschmidt, G. Streubel, B. Samans, M. Krause, U. M. Bauer, The protein arginine

methyltransferases CARM1 and PRMT1 cooperate in gene regulation. Nucleic acids research 36, 3202-

3213 (2008).

44. D. Chen et al., Regulation of transcription by a protein methyltransferase. Science 284, 2174-

2177 (1999).

45. Y. Wei, J. C. Horng, A. C. Vendel, D. P. Raleigh, K. J. Lumb, Contribution to stability and folding of

a buried polar residue at the CARM1 methylation site of the KIX domain of CBP. Biochemistry 42, 7044-

7049 (2003).

46. M. B. Davis et al., Expression and sub-cellular localization of an epigenetic regulator, co-activator

arginine methyltransferase 1 (CARM1), is associated with specific breast cancer subtypes and ethnicity.

Molecular cancer 12, 40 (2013).

47. S. El Messaoudi et al., Coactivator-associated arginine methyltransferase 1 (CARM1) is a positive

regulator of the Cyclin E1 gene. Proc Natl Acad Sci U S A 103, 13351-13356 (2006).

48. E. Shishkova et al., Global mapping of CARM1 substrates defines enzyme specificity and

substrate recognition. Nat Commun 8, 15571 (2017).

49. K. B. O'Brien et al., CARM1 is required for proper control of proliferation and differentiation of

pulmonary epithelial cells. Development 137, 2147-2156 (2010).

50. J. Kim et al., Loss of CARM1 results in hypomethylation of thymocyte cyclic AMP-regulated

phosphoprotein and deregulated early T cell development. The Journal of biological chemistry 279,

25339-25344 (2004).

51. D. Kim et al., Enzymatic activity is required for the in vivo functions of CARM1. The Journal of

biological chemistry 285, 1147-1152 (2010).

52. J. Zhao et al., F-box protein FBXL19-mediated ubiquitination and degradation of the receptor for

IL-33 limits pulmonary inflammation. Nature immunology 13, 651-658 (2012).

53. B. B. Chen et al., A combinatorial F box protein directed pathway controls TRAF adaptor stability

to regulate inflammation. Nature immunology 14, 470-479 (2013).

54. Q. Wang et al., Upregulation of miR-27a contributes to the malignant transformation of human

bronchial epithelial cells induced by SV40 small T antigen. Oncogene 30, 3875-3886 (2011).

55. I. V. Yang et al., Identification of novel genes that mediate innate immunity using inbred mice.

Genetics 183, 1535-1544 (2009).

56. V. Fernandez-Saiz et al., SCFFbxo9 and CK2 direct the cellular response to growth factor

withdrawal via Tel2/Tti1 degradation and promote survival in multiple myeloma. Nat Cell Biol 15, 72-81

(2013).

57. H. Cheng et al., cDNA cloning and expression analysis of a novel human F-box only protein. Mol

Cells 14, 56-59 (2002).

58. K. W. Lee et al., F-box only protein 9 is required for adipocyte differentiation. Biochem Biophys

Res Commun 435, 239-243 (2013).

59. K. W. Lee et al., F-box only protein 9 is an E3 ubiquitin ligase of PPARgamma. Exp Mol Med 48,

e234 (2016).

60. X. Li, Y. Lai, J. Li, M. Zou, C. Zou, Oxidative Stress destabilizes Protein Arginine Methyltransferase

4 Via Glycogen Synthase Kinase 3beta To Impede Lung Epithelial Cell Migration. Am J Physiol Cell Physiol,

ajpcell 00073 02017 (2017).

61. L. Ganesh et al., Protein methyltransferase 2 inhibits NF-κB function and promotes apoptosis.

Molecular and cellular biology 26, 3864-3874 (2006).

62. D.-i. Kim et al., High-glucose-induced CARM1 expression regulates apoptosis of human retinal

pigment epithelial cells via histone 3 arginine 17 dimethylation: role in diabetic retinopathy. Archives of

biochemistry and biophysics 560, 36-43 (2014).

63. D. Rittirsch, M. S. Huber-Lang, M. A. Flierl, P. A. Ward, Immunodesign of experimental sepsis by

cecal ligation and puncture. Nat Protoc 4, 31-36 (2009).

64. J. E. McDunn, R. S. Hotchkiss, Leukocyte phenotyping to stratify septic shock patients. Crit Care

13, 127 (2009).

65. A. M. Drewry et al., Persistent lymphopenia after diagnosis of sepsis predicts mortality. Shock

42, 383-391 (2014).

66. C. Zou et al., LPS impairs oxygen utilization in epithelia by triggering degradation of the

mitochondrial enzyme Alcat1. J Cell Sci 129, 51-64 (2016).

67. H. Cheng et al., Overexpression of CARM1 in breast cancer is correlated with poorly

characterized clinicopathologic parameters and molecular subtypes. Diagn Pathol 8, 129 (2013).

68. R. S. Hotchkiss, G. Monneret, D. Payen, Sepsis-induced immunosuppression: from cellular

dysfunctions to immunotherapy. Nature reviews. Immunology 13, 862-874 (2013).

69. E. Shishkova et al., Global mapping of CARM1 substrates defines enzyme specificity and

substrate recognition. Nature communications 8, 15571 (2017).

70. C. Zou et al., LPS impairs oxygen utilization in epithelia by triggering degradation of the

mitochondrial enzyme Alcat1. J Cell Sci 129, 51-64 (2016).

71. B. B. Chen et al., E3 ligase subunit Fbxo15 and PINK1 kinase regulate cardiolipin synthase 1

stability and mitochondrial function in pneumonia. Cell Rep 7, 476-487 (2014).

72. M. G. Toscano, D. Ganea, A. M. Gamero, Cecal ligation puncture procedure. Journal of visualized

experiments: JoVE, (2011).

Figure and legends

Fig. 1. LPS and bacteria increase PRMT4 protein in human lymphoma Jurkat cells. A. Jurkat cells were

treated with LPS as indicated, cell lysates were immunoblotted with PRMT4 and β-actin antibodies. B.

PRMT4 mRNA levels were determined by qRT-PCR in LPS treated lymphocytes, Student T-test were

conducted. C. Jurkat cells were treated with live E. coli (MOI=10) as indicated, cell lysates were

immunoblotted with PRMT4 and β-actin antibodies. D, E. Lung tissue lysates from deidentified human

lung samples (D) and peripheral blood leukocytes from septic patients (E) were analyzed for PRMT4 and

β-actin by immunoblotting. Experiments n=3. F. C57BL/6J mice were subjected to CLP procedures. At 48

h, mice sera were collected from untreated control group (n=5) and cecal ligation puncture (CLP) group

(n=10) for Prmt4 ELISA analysis. G. PRMT4 protein levels (log transformed) were determined by ELISA

from blood plasma from septic patients (n=53) and non-septic control patients (n=53). Lines indicate the

median and IQR, Mann-Whitney U test, p=0.0004.

Fig. 2. PRMT4 is a labile protein degraded via the SCFFBXO9 E3 ligase machinery in lymphocytes. A. Cells

were treated with cycloheximide (CHX), leupeptin or MG132 as indicated. Cell lysates were

immunoblotted with PRMT4 and β-actin antibodies. B. Densitometry results in A were plotted. C.

Ubiquitin overexpression reduces PRMT4 protein in a concentration dependent manner. D. PRMT4 is

ubiquitinated as shown by co-immunoprecipitation (Co-IP) of precipitates analyzed with ubiquitin and

PRMT4 antibodies. E. Screen of F-box family members revealed that overexpression of FBXO9 degrades

PRMT4. F. FBXO9 degrades PRMT4 in a concentration dependent manner. G. Co-IP studies showing that

PRMT4 interacts with SCFFBXO9. H. Knockdown of FBXO9 stabilizes PRMT4. Experiments n=3.

Fig. 3. SCFFBXO9 ubiquitinates PRMT4 at K228. A. schematic presentation of PRMT4 truncation constructs.

B. Fragment 150-608 does not accumulate after MG132 treatment. C. K228A point mutant does not

accumulate after MG132 treatment. D. K228 mutant is stable after CHX treatment. E. K228A mutant is

less ubiquitinated compared to WT PRMT4. F. K228A is not degraded after FBXO9 overexpression.

Experiments n=3.

Fig. 4. FBXO9 docks on a phosphodegron 132TxxxS within PRMT4. A. Pull-down assays suggest that

aa100-150 of PRMT4 is important for FBXO9 binding. B, C. Fine mapping the FBXO9 binding sites with

pull-down assays indicate that a phosphodegron 132TxxxS is critical for FBXO9 binding. D, E. Pull-down

assays with phosphorylation/ dephosphorylation mimics indicate that T132 dephosphorylation and S136

phosphorylation promote the degron binding to FBXO9. F. T132 dephosphorylation and S136

phosphorylation accelerate PRMT4 degradation using cycloheximide (CHX). G. Densitometry results of F

were plotted. Experiments n=1.

Fig. 5. LPS reduces FBXO9 levels leading to PRMT4 accumulation and caspase 3 activation in

lymphocytes cells. A-C. Jurkat cells, mouse splenic lymphocytes, and human peripheral blood T cells

were treated with LPS as indicated. Cell lysates were analyzed by PRMT4, cleaved caspase 3, FBXO9, and

β-actin immunoblotting. D. Over-expression of PRMT4 increases cleaved caspase 3 baseline levels in

Jurkat cells. E. Knockout of PRMT4 in Jurkat cells with the CRISPR-Cas9 technique. F, G. KO of PRMT4

represses LPS-induced caspase 3 activation (F) and overexpression of PRMT4 enhances LPS-induced

caspase 3 activation in Jurkat cells (G). H. Lentiviral expression of PRMT4 enhances LPS-mediated

caspase 3 activation and depletion of PRMT4 by lenti-shPRMT4 reduces cleaved caspase 3 in mouse

splenic lymphocytes. I. Depletion of FBXO9 by shRNA increases caspase 3 activation and this activation is

ablated in PRMT4 KO cells. Experiments n=3.

Fig. 6. PRMT4 causes lymphocyte death. A, B. FACS analysis of apoptosis in PRMT4 KO or overexpressed

Jurkat cells with or without LPS treatment. Data of A were quantitated and analyzed with One-way

ENOVA in B (n=3). C, Empty vector (Vec), Lenti-PRMT4 (OE), or shRNA (sh) particles were delivered

intratracheally into the mouse for 14 d. Mouse splenic T cells were isolated and treated with LPS for 18

h, viable cells were counted and analyzed with One-way ENOVA (n=3). D. Jurkat cells were treated with

LPS and a range of PRMT4 inhibitors as indicated for 3 h. Cell lysates were analyzed for cleaved caspase

3. E. Isolated mouse splenic T cells were treated with LPS and TP064, cleaved caspase 3 was

immunoblotting analyzed. Experiments n=3.

Fig. 7. Inhibition of PRMT4 suppresses splenic lymphocyte death in a LPS lung injury model. A-D.

PRMT4 was knocked down (kd) or overexpressed by i.t. administrated lentiviral constructs for 14 d. Mice

were given (i.t.) LPS or PRMT4 inhibitor as indicated for 24 h (n=8). Lung tissues were stained with

Hematoxylin Eosin (H&E) staining (A, upper panels) and TUNEL (A, lower panels). TUNEL positive cells in

lung tissues were quantitated and analyzed with One-way ENOVA (B) (n=3). C. PRMT4 overexpression

and knockdown in lung tissues were analyzed by immunoblotting. D. Splenic tissues from above PRMT4

knockdown or overexpression experiments (A) were stained with TUNEL. E. TUNEL positive cells in

splenic tissues were quantitated and analyzed with One-way ENOVA (n=3). F, G. Isolated splenic T cells

were analyzed with flow cytometry. CD4 was used as a T cell marker. The data from E were plotted in F

and analyzed with One-way ENOVA (n=3). H. Survival studies were conducted in the LPS lung injury

model, mice were observed for 48 h (n=10). “*” indicates p

observed for 5 days, (n=10). G. The correlation of plasma PRMT4 and IL-6 is analyzed and plotted in

septic patients with lung injury (n=53). Pearson correlation analysis was conducted with the natural log

transformation and Partial after adjusting for age. “*” indicates p

Table 1 – Patient Characteristics

Control Sepsis

N 53 53

PRMT4, mean (SD, ng/mL) 7.8 (11.0) 18.4 (28.6)

Demographics

Age, mean (SD) 55(15.0) 55 (17.5)

Male, n (%) 31 (58.5) 24 (45.3)

BMI, mean (SD) 28.8 (9.1) 30.9 (9.4)

Comorbidities, n (%)

Diabetes 16 (30.2) 22 (41.5)

COPD 6 (11.3) 11 (20.7)

Immunosuppression 9 (17) 9 (17)

Chronic Cardiac Failure 3 (5.7) 7 (13.2)

Chronic Kidney Disease 6 (11.3) 5 (9.4)

Pulmonary Fibrosis 0 (0) 2 (3.8)

Active Neoplasm 3 (5.7) 3 (5.7)

Chronic Liver Disease 9 (17) 6 (11.3)

Alcohol Use 13 (24.5) 8 (15.1)

Severity of Illness

SOFA score1, mean (SD) 5.2 (2.4) 7.4 (2.6)

PaO2:FiO2 ratio, mean (SD) 224 (152) 196 (88)

Shock, n (%) 10 (18.9) 34 (64.2)

Acute Kidney Injury, n (%) 22 (41.5) 40 (75.5)

ICU LOS, days, mean (SD) 8.5 (9.4) 13.4 (9.8)

30-day mortality, n (%) 6 (11.3) 12 (22.6)

90-day mortality, n (%) 7 (13.2) 13 (24.5)

BMI - body mass index, COPD - chronic obstructive lung disease, SOFA - sequential organ failure assessment, PaO2 - arterial partial pressure of oxygen, FiO2 - fraction of inspired oxygen, ICU - intensive care unit, LOS - length of stay 1 modified SOFA excludes neurologic score

Table 2. The study of association between PRMT4* and cytokines in septic and non-septic

patients

Group 1 (Sepsis) Crude Adjusted for

age

OR (95%CI)

30 day mortality 1.27 (0.72-2.24) 1.23 (0.67-

2.24)

90 day mortality 1.27 (0.73-2.21) 1.23 (0.69-

2.21)

Use of vasopressor 1.70 (0.97-2.99) 1.79 (1.00-

3.23)

Hyperinflammatory phenotype 1.43 (0.85-2.39) 1.42 (0.84-

2.39)

Correlation (P value)

Ang2* 0.37 (0.007) 0.37 (0.007)

IL8* 0.28 (0.048) 0.27 (0.052)

IL6* 0.37 (0.006) 0.37 (0.007)

Procalcitonin* 0.12 (0.40) 0.14 (0.34)

St2* 0.15 (0.28) 0.15 (0.31)

Fractalkine* 0.05 (0.76) 0.02 (0.92)

IL10* 0.17 (0.23) 0.18 (0.22)

Pentraxin3* 0.23 (0.16) 0.22 (0.19)

Rage* -0.03 (0.85) -0.02 (0.92)

TNFr1* 0.27 (0.052) 0.27 (0.059)

* Natural log transformation

Group 2 (Control) Crude Adjusted for

age

OR (95%CI)

30 day mortality 0.75 (0.37-1.53) 0.62 (0.31-1.24)

90 day mortality 0.89 (0.46-1.72) 0.72 (0.38-1.37)

Use of vasopressor 0.95 (0.51-1.77) 1.00 (0.51-1.98)

Hyperinflammatory phenotype 3.01 (1.29-7.02)1 2.29 (0.95-5.51)

Correlation (P value)

Ang2* 0.04 (0.80) -0.03 (0.83)

IL8* 0.10 (0.49) 0.03 (0.83)

IL6* 0.25 (0.07) 0.20 (0.16)

Procalcitonin* 0.24 (0.08) 0.22 (0.12)

St2* 0.03 (0.82) 0.02 (0.89)

Fractalkine* 0.31 (0.035) 0.31 (0.037)

IL10* -0.08 (0.56) -0.12 (0.41)

Pentraxin3* 0.22 (0.14) 0.18 (0.23)

Rage* 0.21 (0.13) 0.17 (0.23)

TNFr1* 0.12 (0.39) 0.06 (0.67)

* Natural log transformation 1 P = 0.011

Table 3. Primers used in the study

Truncation PRMT4F1 CACCATGGCAGCGGCGGCAGCG

PRMT4F2 CACCATGCACGCGGAGCAGCAG

PRMT4F3 CACCATGATCACCCTGGGCTGCAAC

PRMT4F4 CACCATGTTCCAGTTCTATGGC

PRMT4F5 CACCATGTTTTTTGCTGCTCAAGC

PRMT4F6 CACCATGCAAGTGGACATTATC

PRMT4F7 CACCATGGAACAGCTCTACATGG

PRMT4R1 ACTCCCATAGTGCATGGTGTTG

PRMT4R2 GGAGTGGGTGTGATTGACAATC

PRMT4R3 TGTTCCACATATTCTCCGAGG

PRMT4R4 GTCATAGCTCTGTCTTTTG

PRMT4R5 TATGGAGCCAATGAAAGCAAC

PRMT4R6 CATCAGGATCCGGATGTCAAATG

DN131f CACCATGACACTGGAGCGCTCTGTGTTC

DN132f CACCATGCTGGAGCGCTCTGTGTTCAG

DN133f CACCATGGAGCGCTCTGTGTTCAGTG

DN134f CACCATGCGCTCTGTGTTCAGTGAG

DN135f CACCATGTCTGTGTTCAGTGAGCGGAC

DN136f CACCATGGTGTTCAGTGAGCGGACAG

DN137f CACCATGTTCAGTGAGCGGACAGAGG

DN138f CACCATGAGTGAGCGGACAGAGGAATC

DN105f CACCATGAACAGCGTCCTCATCCAG

DN110f CACCATGCAGTTTGCCACACCCCACG

DN115f CACCATGCACGATTTCTGTTCTTTCTAC

DN120f CACCATGTTCTACAACATCCTGAAAAC

DN130f CACCATGCACACACTGGAGCGCTCTG

DN140f CACCATGCGGACAGAGGAATCCTCAG

Mutagenesis T132Cf CCTGTCGGGGCCACTGCCTGGAGCGCTCTGTG

T132Cr CACAGAGCGCTCCAGGCAGTGGCCCCGACAGG

T132Df CCTGTCGGGGCCACGACCTGGAGCGCTCTGTG

T132Dr CACAGAGCGCTCCAGGTCGTGGCCCCGACAGG

S136Cf CCACACACTGGAGCGCTGTGTGTTCAGTGAGCGG

S136Cr CCGCTCACTGAACACACAGCGCTCCAGTGTGTGG

S136Df CCACACACTGGAGCGCGATGTGTTCAGTGAGCGG

S136Dr CCGCTCACTGAACACATCGCGCTCCAGTGTGTGG

T132AS136Af CCACGCACTGGAGCGCGCTGTGTTCAGTGAGCGG

T132AS136Ar CCGCTCACTGAACACAGCGCGCTCCAGTGCGTGG

T132CS136Df CCACTGCCTGGAGCGCGATGTGTTCAGTGAGCGG

T132CS136Dr CCGCTCACTGAACACATCGCGCTCCAGGCAGTGG

K463Af GACCAGACAGGCTCCGCGTCCAGTAACCTGCTGG

K463Ar CCAGCAGGTTACTGGACGCGGAGCCTGTCTGGTC

K471Af GTAACCTGCTGGATCTAGCGAACCCCTTCTTCAGG

K471Ar CCTGAAGAAGGGGTTCGCTAGATCCAGCAGGTTAC

K310Af CATGGAGCAGTTCACCGCAGCCAACTTCTGGTACC

K310Ar GGTACCAGAAGTTGGCTGCGGTGAACTGCTCCATG

K228Af GCAGAGGTCCTGGTGGCGAGTAACAATCTGACAG

K228Ar CTGTCAGATTGTTACTCGCCACCAGGACCTCTGC

K242Af CGTGGTCATCCCTGGCGCAGTAGAGGAGGTCTC

K242Ar GAGACCTCCTCTACTGCGCCAGGGATGACCACG

The orientation of the all primers are in 5’ to 3’ direction.