Characterization of Long Noncoding RNA and mRNA Profiles in Sepsis-Induced Myocardial Depression

Original Article

Characterization of Long Noncoding RNA and mRNA Profiles in Sepsis-Induced Myocardial Depression Tie-Ning Zhang,1 Julie E. Goodwin,2,3 Bing Liu,2,3 Da Li,4 Ri Wen,1 Ni Yang,1 Jing Xia,1 Han Zhou,2,3 Tao Zhang,1 Wen-Liang Song,1 and Chun-Feng Liu1 1Department

of Pediatrics, PICU, Shengjing Hospital of China Medical University, Shenyang, Liaoning Province, 110004, China; 2Department of Pediatrics, Yale University

School of Medicine, New Haven, CT, USA; 3Department of Vascular Biology and Therapeutics Program, Yale University School of Medicine, New Haven, CT, USA; 4Department

of Obstetrics and Gynecology, Shengjing Hospital of China Medical University, Shenyang, China

Septic shock with heart dysfunction is very common in intensive care units. However, whether long noncoding RNA (lncRNA) and mRNA profiles differ between patients with and without myocardial depression is unknown. We generated rat models of hypodynamic septic shock induced by lipopolysaccharide. A total of 12 rat models was constructed and heart tissue from each was collected. Whole genomic RNA sequencing was performed on left ventricular tissue; 6,508 novel lncRNAs and 432 annotated lncRNAs were identified in heart samples, and 74 lncRNAs were expressed differently in the sepsis and control groups. Gene ontology term enrichment indicated apoptosis and its related pathways showed obvious enrichment, which suggested cell apoptosis could play a critical role in the process of myocardial depression. Furthermore, we focused on one lncRNA from the Pvt1 gene. By silencing this lncRNA, we demonstrated knockdown of Pvt1 expression could induce cell apoptosis in lipopolysaccharide-induced heart cells, through increasing the expression of c-Myc, Bid, Bax, and caspase-3 and decreasing the expression of Myd88 and Bcl-2, thereby proving its functional role in myocardial depression. These results demonstrate that lncRNAs both participate in and mediate the pathological process of myocardial depression. Our study improves the understanding of the basic molecular mechanisms underlying myocardial depression.

INTRODUCTION Sepsis is a syndrome of physiological, pathological, and biochemical abnormalities caused by the altered systemic host response to infection.1,2 Septic shock is a subset of sepsis, in which underlying circulatory and cellular abnormalities are profound enough to greatly increase mortality.1 Notably, the myocardial dysfunction during sepsis is usually called sepsis-induced myocardial depression or sepsis-induced heart dysfunction,3 characterized by impaired myocardial contractility and reduced ejection fraction.3,4 This phenomenon can lead to higher mortality, especially in pediatric septic

852

shock patients.5 However, the underlying molecular mechanism of sepsis-induced myocardial depression remains unclear. Recent studies indicate that microRNAs (miRNAs) are involved in the process of sepsis-induced myocardial depression.6 Long noncoding RNAs (lncRNAs) are a type of noncoding RNA with a length exceeding 200 nt. These lncRNAs contribute to transcriptional and post-transcriptional functions, and they can be broadly classified either as signaling molecules, decoy molecules, guide molecules, or scaffold molecules.7 Unlike miRNAs, which have been extensively studied, the lncRNAs present in sepsis-induced myocardial depression have never been sequenced. As is well known, next-generation sequencing has become a powerful approach that can reveal different expression profiles underlying phenotypic differences, as well as decipher non-annotated transcriptional activity by identifying various novel transcripts (protein-coding and noncoding) and additional alternative splice variants of known annotated transcripts.8,9 The limited number of identified lncRNAs in sepsis-induced myocardial depression may impede further study of its complex molecular mechanism. Therefore, we hypothesized that lncRNAs are involved in the pathological process of sepsis-induced myocardial depression. The present study aimed to describe the features of lncRNAs and mRNAs in sepsis-induced myocardial depression. Furthermore, we focused on one particular lncRNA (transcript ID: ENSRNOT00000092896, gene ID: ENSRNOG00000062170, and gene name: Pvt1) for further research, which was the transcript from the rat Pvt1 gene. Our study provides a useful resource to study the functional roles of lncRNA and link the characteristic lncRNAs and mRNAs to myocardial depression phenotypes.

Received 10 April 2019; accepted 24 July 2019; https://doi.org/10.1016/j.omtn.2019.07.020. Correspondence: Chun-Feng Liu, Department of Pediatrics, PICU, Shengjing Hospital of China Medical University, No. 36, SanHao Street, Shenyang, Liaoning 110004, China. E-mail: [email protected]

Molecular Therapy: Nucleic Acids Vol. 17 September 2019 ª 2019 The Author(s). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

www.moleculartherapy.org

A MAP, mmHg

Control

B

150

Sepsis

Sepsis Control

120 90 60 30

h

h 12

8h

10

6h

4h

e lin Ba

se

2h

0

DAPI

TUNEL

C

Merged

D

20 Apoptosis rate (%)

Control

15

10

Sepsis

5 0

Control

Sepsis

E

0.15

0.10

0.1

0.05

0

Control

Sepsis

0

Control

Sepsis

30

1.0

Content of cTNI (pg/ml)

0.2

Content of CK-MB (ng/ml)

0.20 Content of IL-6 (ng/ml)

Content of TNF-α(ng/ml)

0.3

0.8

20

0.6 0.4

10

0.2

0

0

Control

Sepsis

Control

Sepsis

Figure 1. Basic Characteristics of Animal Models between Septic Shock Group and Control Group (A) In vivo mean arterial pressure (MAP) hemodynamic studies. n = 6 in each group. (B) Pathological comparation in heart tissue by H&E staining (10
). (C and D) Apoptosis detection in heart tissue by TUNEL staining (400
). n = 3 in each group. (E) Serum inflammatory biomarker (TNF-a and IL-6) and cardiac enzyme (cTnI and CK-MB) levels were detected by ELISA. n = 6 in each group; *p < 0.05 compared to the control; **p < 0.01 compared to the control.

RESULTS Model Identification of Myocardial Depression in Rats with Septic Shock

Our previous study showed that there is a positive relationship between mean arterial pressure (MAP) and heart function.10 Therefore, we measured MAP and found that MAP started to decline approximately 1.5 h after the administration of LPS. Septic shock occurred approximately 2 h after LPS administration, and it lasted for the rest of the observation period (Figure 1A). In the control group, MAP appeared unchanged throughout the entire observation period. Furthermore, H&E staining showed that there was infiltration of inflammatory cells and rupture of myocardial cells in hearts of the septic rats compared to those in the control group (Figure 1B). Terminal deoxynucleotidyl transferasemediated dUTP nick-end labeling (TUNEL) staining of heart tissue showed a statistically significant increase in apoptosis rates in septic shock animals (Figures 1C and 1D). Finally, we examined

serum inflammatory biomarkers (tumor necrosis factor-a [TNF-a] and interleukin-6 [IL-6]) and cardiac enzymes (cardiac troponin I [cTnI] and creatine kinase MB form [CK-MB]) (Figure 1E), and we found that all of these were significantly elevated in the septic animals. RNA Sequencing Identified the Features of lncRNA and mRNA in Myocardial Depression of Septic Shock

We analyzed RNA sequencing (RNA-seq) data from 12 rat heart samples, in which 82,000,000–115,000,000 raw data and 79,000,000–110,000,000 clean data were obtained (Table 1). The classification of mapped reads in each heart tissue is shown in Figure S1. After comparing the SNP, insertion or deletion (indel), and alternative splicing (AS) events in the sepsis and control groups, we found that the SNP and indel counts in sepsis were less than those in the control group, although there were no obvious differences in AS events (Figure S2).

Molecular Therapy: Nucleic Acids Vol. 17 September 2019

853

Molecular Therapy: Nucleic Acids

Table 1. The Detailed Information of RNA Sequencing Sample

Raw Reads

Clean Reads

Clean Bases (G)

Error Rate (%)

Q20 (%)

Q30 (%)

Total Mapped (%)

Control 1

104,887,116

101,842,730

15.28

0.01

98.09

95.02

96.42

Control 2

110,482,096

107,303,972

16.10

0.01

98.17

95.17

96.62

Control 3

109,101,010

105,976,634

15.90

0.01

98.20

95.27

96.41

Control 4

114,063,808

110,459,742

16.57

0.01

98.30

95.49

96.55

Control 5

104,412,260

101,257,002

15.19

0.01

98.11

95.06

96.22

Control 6

102,114,088

99,322,108

14.90

0.01

98.17

95.20

96.43

Sepsis 1

85,541,078

82,755,500

12.41

0.02

97.22

92.88

95.85

Sepsis 2

83,814,640

80,941,514

12.14

0.02

96.87

92.16

95.48

Sepsis 3

92,375,116

90,576,722

13.59

0.01

97.21

93.01

95.09

Sepsis 4

82,418,946

79,773,136

11.97

0.02

97.13

92.70

95.91

Sepsis 5

101,060,828

98,158,896

14.72

0.01

98.08

94.99

96.43

Sepsis 6

115,144,142

110,000,016

16.50

0.02

97.22

92.90

95.68

To minimize the false-positive rate in identifying lncRNAs from 287,301 assembled transcripts, we developed a filtering flowchart to discard transcripts without all characteristics of lncRNA (Figure 2A). By this method, we identified 6,508 novel lncRNAs from an intersection of the analysis results of CPC, CNCI, and PFAM (Figures 2B and 2C), which included 4,642 lncRNAs (71.3%), 979 antisense lncRNAs (15.0%), and 887 intronic lncRNAs (13.6%) (Figure 2D). We compared the number of overlapping lncRNAs in the sepsis and control groups, which suggested that the lncRNAs that appeared in the sepsis group also appeared in the control group (Figure 2E). There were 23,010 mRNAs included in our study, and we compared the number of overlapping mRNAs in the sepsis and control groups (Figure 2F). In addition, our study found that annotated and novel lncRNAs had a smaller size, fewer exons, and fewer open reading frames than mRNAs (Figures 2G–2I). However, there was no significant difference in transcript levels in heart tissue from the sepsis and control groups (Figures 2J–2L).

lncRNAs is summarized in Table S1, and the top 20 upregulated and 20 downregulated mRNAs are summarized in Table S2. The relative expression levels of several prominent DE mRNAs are shown in Figure 3G. These results reflect distinct lncRNA and mRNA expression profiles between sepsis-induced myocardial depression and the control group, implying different underlying pathophysiology in sepsisinduced heart dysfunction. Systematic Functional Analysis of Differentially Expressed lncRNAs and mRNAs

To check if lncRNAs were involved in the pathological process of sepsis-induced myocardial depression, the expression patterns of lncRNAs and mRNAs were analyzed. We analyzed differential expression (DE) lncRNAs and mRNAs using significance analysis following the criterion q < 0.05. As a result, there were 74 DE lncRNAs (33 upregulated and 41 downregulated) and 4,011 DE mRNAs (1,918 upregulated and 2,093 downregulated) (Figures 3A and 3B). The expression pattern of DE lncRNAs and mRNAs is shown using a cluster heatmap (Figures 3C and 3D).

To elucidate the possible functional significance of observed changes in lncRNA and mRNA levels between sepsis-induced myocardial depression and the control group, we performed a Gene Ontology (GO) term enrichment analysis. There were 20,121 background genes in total. We summarized the significantly enriched GO terms of lncRNAs (Figures 4A–4C) and mRNAs (Figures 4D–4F) regarding biological process, cellular component, and molecular function, respectively. Notably, the number of enriched genes at the lncRNA level was much less than that at the mRNA level regarding biological process. Interestingly, the differentially expressed lncRNAs and mRNAs were similar and significantly associated with regulation and metabolic processes in biological process term enrichment (Figures 4A and 4D). Importantly, the GO term RNA metabolic process was a significant enrichment on both the lncRNA and mRNA levels, which indicates that noncoding RNAs and mRNAs could make sense in the pathological process of sepsis. It is noteworthy that cell apoptosis and cell death were enriched in both lncRNAs and mRNAs, suggesting that cell apoptosis and cell death might play a critical role in sepsis-induced myocardial depression.

In addition, among all DE lncRNAs, there were 35 annotated lncRNAs and 39 novel lncRNAs, and their relative levels of expression are shown in Figures 3E and 3F. There was one novel downregulation lncRNA (LNC_006245) that was not detected in the sepsis group. The detailed information of the top 10 upregulated and 10 downregulated

In addition, the enrichment of cellular components and molecular function also showed a similar pattern. For example, molecularbounded organelle, intracellular part, and nucleus were all obviously enriched on the lncRNA and mRNA levels (Figures 4B and 4E). For molecular function enrichment, terms regarding binding such as

Differential Expression and Cluster Analysis of lncRNAs and mRNAs

854

Molecular Therapy: Nucleic Acids Vol. 17 September 2019

www.moleculartherapy.org

B

Transcript Assembled

Novel lncRNA

11,359

6,508

100,158

100,301

16,020

Assembled transcripts Step 1 Step 2 Step 3 Step 4 Step 5

6,508 1,851

0

St

ep

3

4

5

228 ep

Coding protential prediction

870

314

A tra s s e ns mb cr l e ip d ts St ep 1

Step 5

Annotated lncRNA

St

Step 4

ep

Expression level selection

282

10

St

Known transcripts selection

2

Step 2

Step 3

CNCI

777

ep

Transcripts length selection

20

St

Number of transcripts (log2-transformed)

Exon selection

Step 1

CPC

C 30

287,301

A

PFAM

D

E

Novel lncRNA classification

F

lncRNA overlap

mRNA overlap

15.0%

lincRNA

13.6%

776

antisense lncRNA intronic lncRNA

71.3%

5,768

0

Control

G

995

Sepsis

21,223

Control

H

I

Sepsis

ORF

Length

Exon Number

792

0.0100 8*10^-4

0.0075

Density

Density

Annotated lncRNA mRNA Novel lncRNA

Density

6*10^-4

0.4

Annotated lncRNA mRNA Novel lncRNA

4*10^-4

Annotated lncRNA mRNA Novel lncRNA

0.0050

0.2 0.0025

2*10^-4

0

0

0 0

10

20

30

0

40

FPKM distribution

J

5,000

0

10,000

FPKM distribution

K

5

5

4

4

500

1,000 1,500 2,000

FPKM distribution

L Control

2

is ps Se

tro on C

Sepsis Control

0

l

0

tro

0

l

1

is

0.50

0.25

1

ps

Density

Sepsis Control

on

2

3

0

1

2

3

4

5

log10 (FPKM+1)

C

Sepsis Control

log10 (FPKM+1)

3

Se

log10 (FPKM+1)

0.75

(legend on next page)

Molecular Therapy: Nucleic Acids Vol. 17 September 2019

855

Molecular Therapy: Nucleic Acids

protein binding and enzyme binding were enriched, which suggested that complex pathological processes are involved in sepsis-induced myocardial depression. To determine if there were some specific pathways changed in sepsis, we performed the Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis in lncRNA target genes and mRNAs. Interestingly, the apoptosis pathway was also enriched in both lncRNAs and mRNAs, which is consistent with the GO term enrichment results (Figures 4G and 4H). Therefore, we performed further chromosomal localization, and we analyzed the interactions and co-expression of genes in apoptosis-related pathways, at the lncRNA, lncRNA-mRNA, and mRNA levels, respectively (Figures 4I–4K). The TNF-signaling pathway,11,12 nuclear factor kB (NF-kB)-signaling pathway,13,14 and Jak-Stat pathway15,16 were enriched, which previous studies have implicated in sepsis-induced myocardial depression; our KEGG enrichment analysis also suggested their significance. Validation of DE lncRNAs by qPCR

According to the KEGG pathway enrichment analysis and previous sepsis studies, we concluded that the following pathways might be relevant in sepsis-induced myocardial depression: TNF-signaling pathway,11,12 NF-kB-signaling pathway,13,14 Jak-Stat pathway,15,16 and apoptosis pathway.17,18 To validate the reliability of the sequencing results and to provide the basis for further study, five annotated (ENSRNOT00000000787, ENSRNOT00000092896, ENSRNOT00000086178, ENSRNOT00000091713, and ENSRNOT 00000076955) and four novel (LNC_001626, LNC_004326, LNC_004590, and LNC_004777) DE lncRNAs were chosen to analyze by qPCR, which included at least one target gene that appeared in the aforementioned four common pathways (Figure 5). To detect widespread DE lncRNAs in sepsis-induced myocardial depression, we expanded the sample numbers (n = 15 in each group) for validation using our previous heart tissue samples. All lncRNAs selected for validation were statistically significant and consistent with the results obtained from next-generation sequencing, which indicated the high quality and validity of RNA-seq. The Role of lncRNA rPvt1 in Sepsis-Induced Myocardial Depression

By considering GO terms, KEGG pathway enrichment, and our findings using the septic shock model, we found that apoptosis could be a critical factor in myocardial depression. The Pvt1 gene and its related lncRNAs have been reported in research in humans, which suggested that this gene was related to cell apoptosis.19,20 We focused on an

lncRNA from Pvt1, which was named lncRNA rPvt1 for further study. Surprisingly, the fold changes from qPCR and RNA-seq were almost the same and showed significant upregulation in the sepsis group. In addition, we realized that its target genes included several classical apoptosis-related genes, such as Myc, p53, Bid, and Myd88 (Figure S3), which is highly suggestive of a potential role in the apoptosis process. We first located its expression area by RNA fluorescence in situ hybridization (FISH) in the H9c2 cell line after LPS administration, and we noticed that lncRNA rPvt1 could express in both the nucleus and cytoplasm (Figure 6A). In addition, we performed northern blot to evaluate its total sequence length, and we found that the expression bands were near the second shortest marker (representing the length of 1,821 bp) (Figure 6B). To validate the sequence accuracy and acquire its total length, we conducted rapid amplification of cDNA ends (RACE) clone experiments twice: the first time RACE clone was used to test the known sequence and acquire the 30 end and 50 end sequence, and the second RACE test aimed to demonstrate the acquired total sequence and show congruence. Similar to the results of the northern blot, we acquired an unknown 91-bp sequence from the 50 end RACE and an unknown 325-bp sequence from the 30 end RACE (Figure 6C). To elucidate the potential role of lncRNA rPvt1 in apoptosis, we first detected that lncRNA rPvt1 showed upregulation in the LPS-induced H9c2 cell line compared with the control group (Figure 6Di), which was consistent with the result from animal models. We then performed lentivirus transfection to silence this lncRNA, and we observed that its expression level was clearly decreased (Figure 6Dii). Furthermore, we performed the detection of cell apoptosis and cell cycle in four groups, including a control group, LPS-induced group, LPS-induced negative control (NC) group, and LPS-induced knockdown (KD) group. We found that the apoptosis rates increased significantly in the LPS-induced cells compared with the control group after 24 and 48 h. As for the LPS-induced NC and KD groups, the cell apoptosis rate increased in the silencing group compared with the NC group following LPS administration after 24 and 48 h, respectively (Figure 6E). Regarding cell cycle analyses, we noticed that there was a decreasing trend of S phase cells in the LPS-induced group compared with normal cells, and there was also an obvious decrease of S phase cells in the LPS-induced KD group compared with the NC group after 48 h (Figure 6F). We also performed cell counting kit-8 (CCK-8) to determine both the proliferation ability in the control and LPS

Figure 2. Characterization of lncRNAs and mRNAs in Rat Heart Tissue (A and B) The five steps of novel lncRNA screening. (C) Coding potential analysis via CPC, PFAM, and CNCI. Those sequences simultaneously shared by the above three tools were selected as candidate lncRNAs. (D) The classification of identified lncRNAs. (E) Venn diagram showing the number of overlapping lncRNAs in sepsis group compared with control group. (F) Venn diagram showing the number of overlapping mRNAs in sepsis group compared with control group. (G–I) Density distribution diagram showing the expression features of exon number (G), length (H), and opening reading frame (ORF) (I) of annotated lncRNAs, novel lncRNAs, and mRNAs in rat heart tissue. (J–L) Boxplot (J), violin plot (K), and density distribution diagram (L) showing the expression features of rat heart tissue from the sepsis group and the control group. FPKM, fragments per kilobase million.

856

Molecular Therapy: Nucleic Acids Vol. 17 September 2019

www.moleculartherapy.org

A

B

DE lncRNAs_Sepsis vs Control

C

DE mRNAs_Sepsis vs Control

Red: Up regulated (33)

D

DE lncRNAs cluster heatmap

Red: Up regulated (1,918) Green: down regulated (2,093)

Green: down regulated (41)

-log10 (qvalue)

2 1.3

2

2

1

1

0

0

-1

-1

-2

-2

2 1.3

0

ENSRNOT00000089619

ENSRNOT00000091080

ENSRNOT00000088340

ENSRNOT00000081637

C on t C rol 2 on tr C ol 4 on tro l C on 6 tro l C on 5 tro l C on 1 tro Se l 3 ps is Se 4 ps is Se 3 ps is Se 2 ps Se is 1 ps is Se 5 ps is 6

ENSRNOT00000084571

ENSRNOT00000076955

ENSRNOT00000086178

ENSRNOT00000092100

ENSRNOT00000079756

ENSRNOT00000092515

ENSRNOT00000007070

ENSRNOT00000080883

ENSRNOT00000076493

ENSRNOT00000087595

C

C

ENSRNOT00000093470

ENSRNOT00000080531

ENSRNOT00000077454

ENSRNOT00000089995

ENSRNOT00000090459

ENSRNOT00000090651 LNC_004363

ENSRNOT00000087021

ENSRNOT00000090331 LNC_003208

ENSRNOT00000000787

ENSRNOT00000091713

ENSRNOT00000085860

ENSRNOT00000081316

ENSRNOT00000086135

-4

ENSRNOT00000090485

-2

ENSRNOT00000092896

2

ENSRNOT00000092503

Relative level

DE Annotated lncRNAs_Sepsis vs Control 4

0

on tro l on 3 tro l on 5 tr C ol 2 on tro l C on 4 tr C ol 1 on tro Se l 6 ps is Se 4 ps is Se 3 ps is Se 2 ps is Se 1 ps is Se 5 ps is 6

20

C

0 log2 (fold change)

ENSRNOT00000086642

-20

20

ENSRNOT00000076620

0 log2 (fold change)

ENSRNOT00000074532

-20

ENSRNOT00000076276

0

LNC_001526

Rbp7

LNC_000096

LNC_004461

LNC_002550

Smim5

LNC_004460

LNC_002608

LNC_002610

Susd2

LNC_000813

LNC_002611

Nxpe1

Lhx6

LNC_003811

LNC_006347

Rab3a

LNC_005106

LNC_001626

Sult1a1

Aox3

LNC_002800

Pde4c

C1qtnf9

LNC_001260

LNC_002808

Perm1

LNC_004459

LNC_004458

Omd

LNC_001346

LNC_004197

LNC_005658

LNC_001151

Cxcl17

Ptx3

LNC_004311

LNC_004032

LNC_005883

LNC_006104

LNC_004777

LNC_002702

LNC_003456

LNC_002703

LNC_002898

LNC_006244

Acod1

G

LNC_004326

-5

LNC_000464

0

LNC_000590

Relative level

5

Cxcl9

DE Novel lncRNAs_Sepsis vs Control 10

F

LNC_004590

E

4

ENSRNOT00000076147

-log10 (qvalue)

4

DE mRNAs cluster heatmap

DE mRNAs_Sepsis vs Control 15

C2cd4b

Btnl9

Gpihbp1

Slc26a10

Klf12

Cyp2t1

Chp2

Akr1c19 Pop1

Ccl12

Ccl2

Ccl22

Ido1

Ptges

Ccl7

Cxcl1

Ubd

Cxcl2

Csf3

Cxcl10

-5

Selp

IL6

0

Ciita

5

Cxcl6

Relative level

10

-10

Figure 3. The Expression Profiling Changes of lncRNAs and mRNAs in Rat Heart Tissue (A and B) Volcano plot indicating up- and downregulated lncRNAs (A) and mRNAs (B) of rat heart tissue in the sepsis group compared with the control group; up- and downregulated genes are colored in red and blue, respectively. (C and D) Heatmap of lncRNAs (C) and mRNAs (D) showing hierarchical clustering of changed lncRNAs and mRNAs of rat heart tissue in the sepsis group compared with the control group; up- and downregulated genes are colored in red and blue, respectively. (E–G) Differences in the expressions of annotated lncRNAs (E), novel lncRNAs (F), and mRNAs (G) in rat heart tissue from the sepsis and normal groups. DE, differential expression.

administration groups and the influence of lncRNA rPvt1 on cell proliferation. This showed that there was a statistically significant difference between the control and LPS administration groups at 24, 36, and 48 h (Figure 6Gi). Additionally, proliferation in the knockout (KO) group was decreased compared with the NC group after LPS

administration at both 36 and 48 h (Figure 6Gii). However, the absolute difference of estimated values in these two groups was less than 20%, which suggests that our results provide limited evidence that lncRNA rPvt1 can influence cell proliferation in LPS-induced H9c2 cells.

Molecular Therapy: Nucleic Acids Vol. 17 September 2019

857

Molecular Therapy: Nucleic Acids

A

B

C

D

E

F

G

I

H

J

K

(legend on next page)

858

Molecular Therapy: Nucleic Acids Vol. 17 September 2019

www.moleculartherapy.org

1

4

2

0

4 2

l on C

ps

is

tro

l tro on C

The Potential Mechanism of lncRNA rPvt1 in the Regulation of Cell Apoptosis

We identified that major target genes of lncRNA rPvt1 relating to apoptosis included Myd88, c-Myc, Bid, and p53. As such, we investigated the changes in these four proteins and other apoptosisrelated proteins, such as Bax, Bcl-2, and caspase-3, in heart tissue from control and sepsis groups, using immunohistochemistry (IHC) and western blot. We found that the protein expressions of Myd88, c-Myc, cleaved-Bid, Bax, and cleaved-caspase-3 significantly increased and the expression of Bcl-2 obviously decreased in the sepsis group compared with the control group. However, there was no difference in the p53 protein between the control and sepsis groups (Figures 7A and 7B). Furthermore, to illustrate which genes were regulated and influenced by lncRNA rPvt1, we performed western blot of these proteins in a control cell line, an LPS-induced cell line, an LPS-induced NC group, and an LPS-induced KD group

is is

Se

8 6

0

is ps Se

tro

tro on C

6

l

0

10

ENSRNOT00000092896

2

I

8

ENSRNOT00000091713

3

on

l

is ps Se

l tro C

Se

on

ps

is

l tro on

H

C

0

Additionally, we performed RNA pull-down combined with mass spectrographic analysis to analyze whether the proteins were involved in the regulation process of lncRNA rPvt1 (Figure 7D). We found that the Irak-2 protein, which participates in the IL-1R/Myd88 pathway, could bind directly with lncRNA rPvt1 in the sepsis group, but not in the control group. Furthermore, we proved that the expressions of Irak-2 and IL-1R were different in the control and sepsis groups (Figure 7Ei) and that, after silencing lncRNA rPvt1, the expression of Irak-2 obviously decreased in the LPSinduced KD group compared with the LPSinduced NC group (Figure 7Eii). These results suggest that the IL-1R/Myd88/Irak-2 pathway could be involved in septic myocardial depression and that lncRNA rPvt1 might influence Myd88 through binding with the Irak-2 protein. ps

0.5

0

4

Se

C 1.0

is

2

1.5

ENSRNOT00000076955

ENSRNOT00000086178

0.5

ps

tro on

ps

on

Se

C

F

1.0

4

C

ENSRNOT00000000787

is

tro

is ps

tro on

E 1.5

8

0

G

0

(Figure 7C). We noticed that the Myd88 and Bcl-2 proteins decreased in the LPS-induced KD group but that the expressions of c-Myc, cleaved-Bid, Bax, and cleaved-caspase-3 increased compared with the LPS-induced NC group.

ps

0

l

0

0.5

l

10

6

LNC_004326

20

1

C

D

1.0

30

Se

2

40

LNC_001626

LNC_004590

3

l

LNC_004777

4

Figure 5. qPCR Validations (A–I) qPCR validations of nine regulated lncRNAs in rat heart tissue from the sepsis group and the control group. n = 15 in each group; **p < 0.01 compared to the control.

C 1.5

50

Se

B

5

Se

A

DISCUSSION Sepsis, as a clinical emergency, usually causes multi-organ dysfunction and can lead to high mortality.21 Despite improvements in intensive care treatments, septic shock, which is a subtype of sepsis with circulatory and cellular and/or metabolic dysfunction, is still associated with a higher risk of mortality. Unlike warm shock with highcardiac output and low peripheral vascular resistance occurring in adult septic shock, pediatric septic shock often occurs as cold shock and is characterized by low cardiac output and high peripheral vascular resistance. Therefore, it is necessary to understand the mechanisms of myocardial depression and aim to reduce the mortality in clinic. To reflect and simulate the pathological process of pediatric septic shock, which is usually modeled by cecal ligation and puncture,

Figure 4. GO and KEGG Analyses of Differentially Expressed lncRNA Target Genes and mRNAs in Rat Heart Tissue from the Sepsis Group and the Control Group (A–C) GO categories (biological process, A; cellular components, B; and molecular function, C) of differential lncRNA target genes in rat heart tissue from the sepsis group and the control group. (D–F) GO categories (biological process, D; cellular components, E; and molecular function, F) of differential mRNAs in rat heart tissue from the sepsis group and the control group. (G) KEGG analysis of differential lncRNA target genes in rat heart tissue from the sepsis group and the control group. (H) KEGG analysis of differential mRNAs in rat heart tissue from the sepsis group and the control group. (I–K) RCircos package analysis displaying chromosomal localization and the interactions of the co-expression of genes (STRING) regarding apoptosis-related pathways at the lncRNA level (I), lncRNA-mRNA level (J), and mRNA level (K).

Molecular Therapy: Nucleic Acids Vol. 17 September 2019

859

Molecular Therapy: Nucleic Acids

A

Probe

DAPI

B

Merged

ii

Ci M C123 4 5 6

18S

iii

3’-end RACE

5’-end RACE

1

M

2

M

Second RACE

1

1

2

M

6,948bp 4,742bp 2,661bp

U6

1,821bp 1,517bp

lncRNA

1.36%

5.28%

3.38%

1.05%

0.36%

**

9.39%

7.82%

2.79%

24h

4

**

0.6

**

+L

KD

C

+L

PS

PS N

KD

C

+L

l tro

N

PS +L KD

l tro on

PS

PS

10

PS +L KD

+L C N

l

S LP

tro

*

20

S

10

PS

LP S

on C

KD

20

LP

PS

PS +L

LP S

C

10

C

S cell rate (%)

+L

l tro on C

N

PS

PS

+L C PS +L

l tro on

ii 0.8

20

30

**

20

C

Gi

KD

N

40

+L

G2 19.39%

G2 24.67%

KD

S 12.30%

S 15.32%

G2 23.64%

S

S 17.08%

G2 23.37%

30

**

30

**

60

LP

S 27.06%

+L

LP S

tro l on C

48h

G1 68.31%

G1 60.01%

G1 61.59%

G1 cell rate (%)

80

G1 49.57%

10 5

G2 23.02%

G2 25.46%

G2 cell rate (%)

S cell rate (%)

20

40

15

on

S 17.71%

40

*

G2 cell rate (%)

20

C

G1 59.26% S 17.67%

G2 21.11%

G2 27.03%

25

C

S 17.08%

S 21.48%

80

60

KD+LPS

G1 58.76%

G1 61.82%

G1 51.22%

24h

NC+LPS

N

LPS

G1 cell rate (%)

Control

N

C

F

PS

5.90%

72.28%

+L

8.14%

79.88%

11.03%

l

78.62%

1.32%

on

KD

97.02%

tro

0

10

+L

48h

0.5

**

KD

1.0

** 20

S

16.98%

4.84%

S

9.24%

2.74%

LP

8.21%

2.14%

1.06%

+L P

C

on

LP

tro

48h Apoptosis rate (%)

l

S

30

0.59%

N

S

5.68%

C

77.11%

7.74%

88.11%

N

5.14%

LP

86.20%

tro

2.39%

1.5

C

ENSRNOT00000092896

5

on

96.20%

0

ii

10

l

2

*

**

15

C

6

KD+LPS

C

ENSRNOT00000092896

8

NC+LPS

PS

LPS

+L

Control

PS

E

24h Apoptosis rate (%)

20

Di

0.6

LPS

**

Control

**

KD+LPS NC+LPS

OD450

OD450

0.4

**

0.4

0.2 0.2

0 0h

12h

0 24h

Time

36h

48h

0h

12h

24h

36h

48h

Time

Figure 6. Functional Study of lncRNA rPvt1 (A) Expression location of lncRNA rPvt1 in LPS-induced H9c2 cell lines by fluorescence in situ hybridization (FISH) under confocal microscopy (600
). (B) The expression of lncRNA rPvt1 detected by northern blot (M, marker; C, positive control product; bands 1–3, heart tissue from the control group; bands 4–6, heart tissue from the sepsis group). (C) Rapid amplification of cDNA ends (RACE) to acquire the total sequence of lncRNA rPvt1 (i, 30 end RACE; M, marker; 1, 30 -RACE inner PCR product (+); 2, 30 -RACE inner PCR product (); ii, 50 end RACE; M, marker; 1, PCR product; iii, second RACE for validation; M, marker; 1, validation PCR product [+]; (legend continued on next page)

860

Molecular Therapy: Nucleic Acids Vol. 17 September 2019

www.moleculartherapy.org

we used a rat model subjected to a high dose of LPS, which showed an obvious and rapid shock and accompanying myocardial dysfunction. Our model is relatively stable and has a high reproducibility rate; therefore, it is suitable for studying the mechanism of septic shock in children. Recently, with the development of next-generation high-throughput sequencing, multiple lncRNAs have been identified and shown to play important roles in many diseases.22,23 Compared with microarray, next-generation sequencing has many strengths, including high sensitivity and specificity, and the ability to discover important novel lncRNAs. However, there is no relevant study using whole genomic sequencing that reports and compares changes in lncRNAs and mRNAs in sepsis-induced myocardial depression. To realize the basic characteristics of lncRNAs and mRNAs and evaluate the potential role of lncRNAs in septic myocardial depression, our study identified entire lncRNAs and mRNAs changing significantly in left ventricular tissues compared with the non-sepsis group. In addition, to improve the reliability and validity of sequencing, we developed six septic shock animals compared with six control animals to find the ubiquitous DE lncRNAs and mRNAs. According to quality control and qPCR validation, our sequencing results had high reliability and quality. Through bioinformatics analysis, we predicted potential functions of lncRNAs and mRNAs by GO term enrichment and KEGG pathway enrichment, and we found important changes that help us to understand the molecular mechanisms of sepsisinduced myocardial depression more clearly. In addition, we focused on one specific lncRNA, lncRNA rPvt1, and we conducted several experiments to thoroughly investigate its characteristics and evaluate its potential role in cell function. Our study indicates that lncRNAs, including lncRNA rPvt1, play important roles in the pathological process of septic myocardial depression and provide a new direction toward understanding this pathological process. We identified 74 DE lncRNAs between sepsis and control groups; the underlying mechanisms of these lncRNAs are poorly understood. We also performed GO enrichment and found that the terms of biological processes occupied most of the significantly enriched terms. According to our KEGG enrichment results, we found that TNF-signaling pathway, NF-kB-signaling pathway, Jak-Stat pathway, and apoptosis pathway were the common pathways that appeared at both the lncRNA and mRNA levels. Although some published studies11,14,16,17 stated the potential role of the aforementioned four common pathways in sepsis, the regulatory relationship between lncRNAs and these pathways is still unclear and should be studied in more depth in future basic studies.

Although a recent study conducted by Takasu et al.24 showed that myocardial tissue from patients with sepsis lacked significant necrosis, apoptosis, or autophagy compared with control donor hearts, we detected a significant difference in both apoptosis rates and the levels of several proteins related to cell apoptosis, including Myd88, c-Myc, Bid, Bax, Bcl-2, and caspase-3, in heart tissue from the septic shock and control groups. These results are also consistent with the results of GO term and KEGG pathway enrichment. To realize the coexpression relationships of genes relating to the apoptosis pathway, we also performed gene analyses of chromosomal localization and regulatory association between lncRNA and mRNA. We detected an obvious rate of apoptosis, which is partly due to the fact that we performed the septic shock model with a large dose of LPS, producing a more serious subtype of sepsis characterized by severe heart dysfunction and damage. In the previous clinical study,24 the included patients were septic patients rather than those in septic shock, which might explain why they did not develop serious myocardial damage. Another potential explanation is that we used a pediatric septic shock model, which has a different pathological process compared to adult sepsis. The low cardiac output in pediatric sepsis may reflect a severely impaired heart function, whereas there is usually normal or high cardiac output in adult sepsis. Therefore, apoptosis could be an important mechanism of heart dysfunction in pediatric septic shock. We also focused on one particular lncRNA and estimated its role in a septic model in cell culture. From previous studies, we knew that the lncRNA from human Pvt1 could regulate the apoptosis process. For example, a study conducted by He et al.20 showed that silencing lncRNA Pvt1 could increase cell apoptosis in nasopharyngeal carcinomas, and other studies support this notion in different malignancies.19,25 Although the sequence of lncRNA rPvt1 is different from that lncRNA from the human Pvt1 gene, we speculated that our lncRNA could influence and regulate cellular function in heart tissue through bioinformatics analysis. Therefore, to evaluate the potential role of lncRNA rPvt1 in sepsis, we created a lentiviral vector and silenced lncRNA, after which we observed that knockdown of Pvt1 expression could induce cell apoptosis in the LPS-induced heart cell line. This result suggests that Pvt1 could play a role in sepsisinduced myocardial depression. Compared with the control group, lncRNA rPvt1 in the sepsis group had a tendency toward high expression levels, and therefore this lncRNA could have a protective effect in sepsis. Furthermore, we evaluated the influence of lncRNA rPvt1 on its potential target genes, and we found that lncRNA rPvt1 could regulate the expression of several genes, including Myd88, c-Myc, and Bid, and then further influence the expression of genes related to cell apoptosis, such as Bax, Bcl-2, and caspase-3. Notably, in cancer-related research, several studies26,27 had reported that Pvt1 controls levels of Myc through regulation of

2, validation PCR product []). (D) Expression level of lncRNA rPvt1 in control and LPS-induced H9c2 cell line (i); detection of knock down rate by qPCR after lentiviral transfection for silencing lncRNA rPvt1 (ii). (E) Apoptosis detection in H9c2 cell line by flow cytometry. (F) Cell cycle detection in H9c2 cell line by flow cytometry. (G) Proliferation detection in control and LPS-induced cell lines (i), LPS-induced NC cell line, and LPS-induced KD cell line (ii) by CCK-8. NC, negative control group; KD, lncRNA rPvt1 knockdown group; *p < 0.05 compared to the control; **p < 0.01 compared to the control.

Molecular Therapy: Nucleic Acids Vol. 17 September 2019

861

862

Molecular Therapy: Nucleic Acids Vol. 17 September 2019

0.4

0.2

0

1.0

** PS

PS

+L

ii +L

+L

PS

PS

+L

**

KD

KD

C

S

S

+L P

S

+L P

C

KD

N

**

S

*

is

l

l

tro

is

ps

on

Se

C

ps

tro

Bcl-2 Protein Expression Level

cleaved caspase-3 Protein Expression Level

0.8

+L P

0.6

KD

β-actin C

Sepsis S

0.5 1.0

N

1.0 1.5

PS

**

N

0 LP

0.5

S

1.0

LP

l

** Se

on

0

+L

l

tro

1.5

cleaved-Bid Protein Expression Level

C

**

C

tro

on

S

1.0

LP

on

C

+L P

Bcl-2

N

Irak-2 C

l

1.5

cleaved Caspase-3 Protein Expression Level

KD

1.5

S

tro

0 on

**

C

S

is

Control

LP

Control +L P

ps

0

l

0 PS

**

0.5

tro

0.5 0.5

Irak-2 Protein Expression Level

1.5 +L

1.5

on

1.0 C

Bax

C

*

1.0

is

Ei KD

0

**

ps

0.5 N

Sepsis

Se

1.0

PS

0

+L

l

0

C

**

N

** tro

0.2

Se

0.4

l

0.4 on

0.6

Bax Protein Expression Level

0.8

tro

0.2 C

cleaved-Caspase 3

on

l

cleaved-Bid

S

tro

Bcl-2

LP

on

p53

C

C

is

l

c-Myc

S

l

ps

tro

p53 Protein Expression Level

Bax

LP

tro

Se

on

Myd88

IL-1R Protein Expression Level

Sepsis on

**

C

0.6

c-Myc Protein Expression Level

**

Bcl-2 Protein Expression Level

PS

+L

KD

0.8

is

PS

C

p53

ps

KD +L

1.5 PS

1.0

+L

cleaved-Bid

Se

l

C

0

tro

N

is

l

0.5

PS

ps

1.0

+L

S

Se

*

on

Control C

β-actin N

cleaved-Caspase 3 LP

l

cleaved-Bid Protein Expression Level

1.5

C

Irak-2 Protein Expression Level

Bax

S

tro

c-Myc

LP

on

cleaved-Bid C

tro

0

r

l

on

0.5

ke

tro

C

1.5

ar

on

Myd88

Myd88 Protein Expression Level

is

c-Myc

M

C

Bcl-2

Bax Protein Expression Level

PS

l

1.0

en se

+L

ps

tro

**

tis

e

KD

Se

Control

An

ns

PS

c-Myc Protein Expression Level

Myd88

Se

+L

on

is

0

C

C

ps

0.5

S

l

l

1.5

N

LP

tro

C on

tro

1.0

e

C

Se

on

**

ns

se

r

Myd88 Protein Expression Level

B

Se

en

ke

D

tis

ar

C

A

An

M

Molecular Therapy: Nucleic Acids

cleaved-Caspase 3

Control

Sepsis

Sepsis

β-actin 1.0

0.6

**

0.4

0.2 0

1.0

**

0.5

0

0.8

**

0.6

0.2 0.4

0

IL-1R Irak-2

β-actin

1.5

**

0.5

0

(legend on next page)

www.moleculartherapy.org

the protein’s stability. However, there was no prior study focusing on the role of lncRNA rPvt1 in heart tissue in sepsis. In our study, we found that silencing lncRNA rPvt1 could increase both the protein level of c-Myc and the expressions of Bax and caspase-3, therefore increasing cell apoptosis in rat heart tissue. Our results also showed that the expression of Myd88 decreased in the LPS-induced KD group compared with the LPS-induced NC group. After performing RNA pull-down, we hypothesized that this process might be mediated by Irak-2, which could bind with lncRNA rPvt1 directly, and that the IL-1R/Myd88/Irak-2 pathway might be involved in this process. Although silencing lncRNA rPvt1 could inhibit the expression of Myd88 and Irak-2, which would decrease the inflammatory response to a certain extent, in fact we found that silencing lncRNA rPvt1 led to an increase in cell apoptosis. This effect may be explained by a greater relative increase in other apoptosis-related proteins, such as Bax and Bid, and a decrease in the anti-apoptotic protein Bcl-2. However, the mechanism by which lncRNA rPvt1 influences c-Myc and regulates the expressions of Bax, Bid, and caspase-3 still requires further research to fully understand. Our study is an investigation into the role of lncRNAs in the pathological process of myocardial depression in models of septic shock. By applying whole genomic next-generation sequencing, we compared the DE lncRNAs and mRNAs between sepsis and control groups. These abundant data prompted the proposal that lncRNAs may interact and regulate their related protein-gene expression and, thus, play a key role in the pathogenesis of sepsis-induced myocardial depression. Additionally, we investigated lncRNA rPvt1 and detected its role in the myocardial cell model of sepsis. lncRNAs may provide new insights into the pathophysiology of sepsis-induced myocardial depression, as well as the potential for development of lncRNA mediators in the future.

MATERIALS AND METHODS Animal Model of Septic Shock

The study was approved by the Ethics Committee of Shengjing Hospital of China Medical University (2019PS073K). All experiments conformed to all relevant regulatory standards. The adolescent rat model of septic shock was generated by intraperitoneal injection of LPS, as in our previous description.10 Briefly, male pathogenfree Wistar rats from Changsheng Bio (Benxi, China) weighing from 170 to 190 g were anesthetized by 20% urethane (1 g/kg i.p.). We cannulated the left femoral artery to continue monitoring the MAP of the models (Biopac MP150 Biopac Systems, Goleta, CA, USA). After the MAP was stable, we challenged the rat with Escherichia coli 055:B5 (L-2880, Sigma-Aldrich, St. Louis, MO, USA;

20 mg/kg, 10 mg LPS dissolved in 1 mL 0.9% saline), and septic shock was established when the MAP decreased to 25%–30% of the baseline value. For the control group, 0.9% saline (2 mL/kg) was injected into the peritoneal cavity of rats. The left ventricle of the heart was excised after 12-h LPS or saline administration and immediately snap frozen in liquid nitrogen and stored at –80 C for further experiments. In addition, we collected blood from the abdominal aorta at 12 h after LPS or saline administration, which was centrifuged (3,000 rpm, 10 min) after being maintained at room temperature for 15 min. The blood supernatant was collected and stored at –80 C for ELISA. H&E Staining, TUNEL Assay for Apoptosis, and ELISA Determination of Rat Serum

The heart samples were fixed in 10% formaldehyde for 48 h. The specimens were subjected to dehydration and permeabilization and then embedded in paraffin. The paraffin tissues were sliced into serial sections (0.4 mm) and stained with H&E. Pathological changes were observed under a microscope. We used the TUNEL technique to detect apoptosis of heart tissues. A TUNEL apoptosis assay kit was purchased from WanleiBio (Shenyang, China), and we performed the assay in accordance with the manufacturer’s instructions. We detected TNF-a, IL-6, CK-MB, and cTnI in blood serum according to the instructions of the ELISA kit (CusaBio, Wuhan, China). The concentration of each sample was calculated according to the optical density value and the standard curve. RNA-Seq and Bioinformatics Analyses

Sequencing data were deposited with the NCBI (Sequence Read Archive [SRA]: PRJNA527717). The procedures regarding sequencing, data processing, and further bioinformatics analyses can be found in detail in the Supplemental Materials and Methods. Total RNAs from rats subjected to septic shock (n = 6) and controls (n = 6) were isolated and quality controlled. The preparation of whole-transcriptome libraries and next-generation sequencing were conducted by Novogene Bioinformatics Technology (Beijing, China). RNA-seq was performed on an Illumina Hiseq 4000 platform, and 150-bp-paired single-end reads were generated according to Illumina’s protocol. All the downstream bioinformatics analyses were based on the clean data with high quality. Validation of Gene Expression in RNA-Seq by qPCR Analysis

Total RNA was isolated from heart tissue or cultured cells, according to the manufacturer’s instructions supplied with the TRIzol Reagent

Figure 7. Mechanism Study of lncRNA rPvt1 (A) Immunohistochemical staining of Myd88, c-Myc, cleaved-Bid, p53, Bax, Bcl-2, and cleaved-caspase-3 in rat heart tissue from the control and sepsis groups (400
). n = 3 in each group. (B) Western blot analysis of Myd88, c-Myc, cleaved-Bid, p53, Bax, Bcl-2, and cleaved-caspase-3 in rat heart tissue from control and sepsis rat models. n = 3 in each group. (C) Western blot analysis of Myd88, c-Myc, cleaved-Bid, Bax, Bcl-2, and cleaved-caspase-3 in the normal H9c2 cell line, LPS-induced cell line, LPS-induced NC cell line, and LPS-induced KD cell line. n = 3 in each group. (D) Silver staining of rat heart tissue during RNA pull-down in control and sepsis groups. n = 1 in each group. (E) Western blot analysis of Irak-2 and IL-1R in rat heart tissue (i) and H9c2 cell line (ii). n = 3 in each group; *p < 0.05 compared to the control; **p < 0.01 compared to the control.

Molecular Therapy: Nucleic Acids Vol. 17 September 2019

863

Molecular Therapy: Nucleic Acids

(Invitrogen, Life Technologies, Carlsbad, CA, USA), and then reverse transcribed with a Thermo First Synthesis Kit. Primers of lncRNAs were designed and synthesized by GenechemBio (Shanghai, China). The specific quantitative primers for nine lncRNAs are listed in Table S1. Each targeted cDNA was amplified using SYBR Green via the StepOnePLUS Real-Time PCR system (Applied Biosystems, Waltham, MA, USA). In addition, the primers of the GAPDH gene were designed as an endogenous control. RACE Clone of lncRNA rPvt1

The PCR and RACE primers were designed and synthesized by Takara Bio (Dalian, China). Total isolated RNA was converted into cDNA using PrimeScript 1st Strand cDNA Synthesis Kit (Takara, 6110A, Japan) with oligo dT primer. All kits used in this section were provided with SMARTer RACE 50 /30 Kit component (Takara, 634858, Japan). We strictly conducted the 30 -RACE and 50 -RACE PCR reactions that generated the 30 and 50 cDNA fragments according to the manufacturer’s instructions. Then, purified PCR products were cloned into the pUC-19 vector and sequenced by Takara Bio (Dalian, China). Northern Blot Analysis

The heart tissues from the septic shock and control groups were used for RNA isolation and detection, run on a 1% agarose gel, and transferred to positively charged nylon membranes. The membranes were subjected to hybridization with 10.0 mL DIG Easy Hyb-labeled probes overnight at 50 C. Then, the band lncRNA rPvt1 was detected with X-ray film in a dark room. The probe sequences for lncRNA rPvt1 gene were 50 -AGATGTTACAAGTGGCCTAATG-30 (forward) and 50 -AGAGGTCTATCCTTGCTCTTTA-30 (300 bp) (reverse), which were designed and synthesized by Sangon Bio (Shanghai, China). Cell Culture and Treatment Protocol

The rat cardiomyoblast H9c2 cell line, obtained from the Shanghai Institute of Biochemistry and Cell Biology (Shanghai, China), was cultured at 37 C under 5% CO2 in DMEM containing 10% fetal bovine serum with 100 U/mL penicillin and 100 mg/mL streptomycin. Cells were passaged regularly and subcultured to 80% confluence prior to experimental procedures. For a septic cell model,10 we treated the cell line with LPS (10 mg/mL) for 12, 24, 36, and 48 h. FISH of lncRNA rPvt1

RNA FISH was performed using a Fluorescent in situ Hybridization kit from RiboBio (Guangzhou, China), in accordance with the manufacturer’s instructions. We used U6 and 18 s as our endogenous controls, which were obtained from RiboBio. The stained samples were imaged under confocal microscopy. Lentiviral Silencing of lncRNA rPvt1

The development of a lentiviral vector system to knock down lncRNA rPvt1 expression was designed to express small interfering RNA (siRNA) sequences for targeting Pvt1. A lentiviral vector that carried a nontargeting sequence was used as the NC of siPVT1. To track

864

Molecular Therapy: Nucleic Acids Vol. 17 September 2019

targeted cells, we designed a co-expressed GFP marker. The sequence of the lncRNA rPvt1 targeting siRNA was as follows: 50 CCTATGAGGTGATGATAAA-30 . The H9c2 cell line was seeded into 6-well plates for 24 h and subsequently transfected with lentivirus in both the KO and NC groups. The transfected cells were grown at 37 C with 5% CO2 for 96 h and then used for further experiments. Measurement of Cell Apoptosis and Cell Cycle

The H9c2 cells were divided into four groups: normal cells, LPSinduced H9c2 cells, and LPS-induced H9c2 cells in the KO and NC groups. All cells were trypsinized and subsequently harvested and incubated using an Annexin V-APC apoptosis detection kit (eBioscience, Waltham, MA, USA) after regular centrifugation and washing. Cellular apoptosis was evaluated via flow cytometry (Phoenix Flow Systems, San Diego, CA, USA) within 5 min and analyzed using a flow cytometer (Coulter EPICS XL-MCL FACScan, BD Biosciences, Franklin Lakes, NJ, USA). The data were analyzed using the guava InCyte Software to determine the cell apoptosis rates. As for the measurement of cell cycle, cells were stained with a fluorescent dye (propidium iodide) that directly binds to the DNA in the nucleus. After ethanol fixation and centrifugation, cells were then resuspended and 100 mL RNase A (Sigma, St. Louis, MO, USA) was added into each sample. After 30-min incubation at 37 C, 500 mL propidium iodide was added to each tube for at least 30 min to provide the nuclear signal for fluorescence-activated cell sorting (flow cytometry). Finally, cell cycle was evaluated with a flow cytometer. Cell Proliferation Assay

The cells transfected with siRNA were seeded into 96-well plates (2,000 cells per well), incubated for 24 h, and then treated with LPS. According to the manufacturer’s instructions, a CCK-8 (Sigma-Aldrich, St. Louis, MO, USA) was used to determine cell proliferation after LPS stimulation. The absorbance of each well was monitored using a spectrophotometer at a wavelength of 450 nm. Immunohistochemical Staining

Immunohistochemical staining for Myd88, c-Myc, cleaved-Bid, p53, Bax, Bcl-2, and cleaved-caspase-3 was conducted on the left ventricular tissue from control and sepsis group. The myocardial tissue was deparaffinized and rehydrated in a graded series of alcohol solutions. Sections were treated with 3% H2O2 for 30 min to block endogenous peroxidase activity and then with 1% BSA in PBS for 30 min. Slides were incubated overnight at 4 C with primary antibody (Myd88, c-Myc, p53, Bax, and Bcl-2; 1:200; Abcam, Cambridge, MA, USA; cleaved Bid and cleaved-caspase-3; 1:200; WanleiBio, Shenyang, Liaoning Province, China). The secondary antibody with horseradish peroxidase was used for incubation for 1 h. Images were viewed by a bright-field microscope (Nikon). Western Blot Analysis

The rat heart tissue and cultured H9c2 cell line were homogenized in RIPA buffer. Proteins from the homogenate (20 mg/well) were

www.moleculartherapy.org

separated on a one-dimensional 4%–10% SDS polyacrylamide gel and transferred to a polyvinylidene membrane (Bio-Rad, Hercules, CA, USA) with a semidry electrotransfer apparatus for 1.5 h at 300 mA. The expression levels of Myd88, c-Myc, cleaved-Bid, p53, Bax, Bcl2, cleaved-caspase-3, Irak-2, and IL-1R were determined (Myd88, cMyc, p53, Bax, Bcl-2, and IL-1R; 1:1,000; Abcam, Cambridge, MA, USA; Irak-2; 1:1,000; Cell Signaling Technology, Danvers, MA, USA; cleaved Bid and cleaved-caspase-3; 1:500; WanleiBio, Shenyang, Liaoning Province, China). The immunoblots were visualized by enhanced chemiluminescence (ECL detection kit from GE Healthcare, Chicago, IL, USA). RNA Pull-down Assay

Briefly, biotin-labeled lncRNA rPvt1 and its antisense RNAs from heart tissue in control and sepsis groups were transcribed in vitro with the Biotin RNA labeling mix and T7 RNA polymerase, then treated with RNase-free DNase I. All the pull-down experiment was used and followed the protocol from Magnetic RNA-protein Pull-Down Kit (Thermo Fisher, Waltham, MA, USA). Then, the proteins were measured using mass spectrographic analysis. Statistical Analysis

All experiments were performed at least three times. Data of a normal distribution were presented as the mean ± SEM. The Student’s t test for unpaired data was used to compare between the two groups. Twoway ANOVA of repeated measures was used to analyze the data of MAP changes with LPS administration. The statistical analysis was done using SPSS 23.0 software (IBM, Armonk, NY, USA). p < 0.05 was considered statistically significant.

SUPPLEMENTAL INFORMATION Supplemental Information can be found online at https://doi.org/10. 1016/j.omtn.2019.07.020.

AUTHOR CONTRIBUTIONS T.-N.Z. and C.-F.L. conceived and designed the study. T.-N.Z performed most assays. T.-N.Z. and R.W. performed animal models and heart tissue collection. T.-N.Z., T.Z., and W.-L.S. performed qPCR and western blot. T.-N.Z., J.E.G., B.L., D.L., N.Y., J.X., H.Z., and C.-F.L performed RNA-seq data analysis and interpretation. T.-N.Z., J.E.G., B.L., D.L., and C.-F.L. wrote the manuscript. All the authors read and approved the final manuscript.

CONFLICTS OF INTEREST The authors declare no competing interests.

ACKNOWLEDGMENTS This study was supported by the National Natural Science Foundation of China (81372039), the Natural Science Foundation of Liaoning Province (2017225003 and 2018108001), and the Science and Technology Foundation of Shenyang (F13-220-9-38).

REFERENCES 1. Singer, M., Deutschman, C.S., Seymour, C.W., Shankar-Hari, M., Annane, D., Bauer, M., Bellomo, R., Bernard, G.R., Chiche, J.D., Coopersmith, C.M., et al. (2016). The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA 315, 801–810. 2. Angus, D.C., and van der Poll, T. (2013). Severe sepsis and septic shock. N. Engl. J. Med. 369, 2063. 3. Raj, S., Killinger, J.S., Gonzalez, J.A., and Lopez, L. (2014). Myocardial dysfunction in pediatric septic shock. J. Pediatr. 164, 72–77.e2. 4. Brierley, J., and Peters, M.J. (2008). Distinct hemodynamic patterns of septic shock at presentation to pediatric intensive care. Pediatrics 122, 752–759. 5. Weiss, S.L., Fitzgerald, J.C., Pappachan, J., Wheeler, D., Jaramillo-Bustamante, J.C., Salloo, A., Singhi, S.C., Erickson, S., Roy, J.A., Bush, J.L., et al.; Sepsis Prevalence, Outcomes, and Therapies (SPROUT) Study Investigators and Pediatric Acute Lung Injury and Sepsis Investigators (PALISI) Network (2015). Global epidemiology of pediatric severe sepsis: the sepsis prevalence, outcomes, and therapies study. Am. J. Respir. Crit. Care Med. 191, 1147–1157. 6. Ho, J., Chan, H., Wong, S.H., Wang, M.H., Yu, J., Xiao, Z., Liu, X., Choi, G., Leung, C.C., Wong, W.T., et al. (2016). The involvement of regulatory non-coding RNAs in sepsis: a systematic review. Crit. Care 20, 383. 7. Boon, R.A., Jaé, N., Holdt, L., and Dimmeler, S. (2016). Long Noncoding RNAs: From Clinical Genetics to Therapeutic Targets? J. Am. Coll. Cardiol. 67, 1214–1226. 8. Trapnell, C., Williams, B.A., Pertea, G., Mortazavi, A., Kwan, G., van Baren, M.J., Salzberg, S.L., Wold, B.J., and Pachter, L. (2010). Transcript assembly and quantification by RNA-Seq reveals unannotated transcripts and isoform switching during cell differentiation. Nat. Biotechnol. 28, 511–515. 9. Griffith, M., Griffith, O.L., Mwenifumbo, J., Goya, R., Morrissy, A.S., Morin, R.D., Corbett, R., Tang, M.J., Hou, Y.C., Pugh, T.J., et al. (2010). Alternative expression analysis by RNA sequencing. Nat. Methods 7, 843–847. 10. Yang, N., Shi, X.L., Zhang, B.L., Rong, J., Zhang, T.N., Xu, W., and Liu, C.F. (2018). The Trend of b3-Adrenergic Receptor in the Development of Septic Myocardial Depression: A Lipopolysaccharide-Induced Rat Septic Shock Model. Cardiology 139, 234–244. 11. Winkler, M.S., Rissiek, A., Priefler, M., Schwedhelm, E., Robbe, L., Bauer, A., Zahrte, C., Zoellner, C., Kluge, S., and Nierhaus, A. (2017). Human leucocyte antigen (HLADR) gene expression is reduced in sepsis and correlates with impaired TNFa response: A diagnostic tool for immunosuppression? PLoS ONE 12, e0182427. 12. Wu, H., Liu, J., Li, W., Liu, G., and Li, Z. (2016). LncRNA-HOTAIR promotes TNF-a production in cardiomyocytes of LPS-induced sepsis mice by activating NF-kB pathway. Biochem. Biophys. Res. Commun. 471, 240–246. 13. Qi, J., Qiao, Y., Wang, P., Li, S., Zhao, W., and Gao, C. (2012). microRNA-210 negatively regulates LPS-induced production of proinflammatory cytokines by targeting NF-kB1 in murine macrophages. FEBS Lett. 586, 1201–1207. 14. Filgueiras, L.R., Jr., Martins, J.O., Serezani, C.H., Capelozzi, V.L., Montes, M.B., and Jancar, S. (2012). Sepsis-induced acute lung injury (ALI) is milder in diabetic rats and correlates with impaired NFkB activation. PLoS ONE 7, e44987. 15. Jin, H., Fujita, T., Jin, M., Kurotani, R., Namekata, I., Hamaguchi, S., Hidaka, Y., Cai, W., Suita, K., Ohnuki, Y., et al. (2017). Cardiac overexpression of Epac1 in transgenic mice rescues lipopolysaccharide-induced cardiac dysfunction and inhibits Jak-STAT pathway. J. Mol. Cell. Cardiol. 108, 170–180. 16. Lv, X., Zhang, Y., Cui, Y., Ren, Y., Li, R., and Rong, Q. (2015). Inhibition of microRNA-155 relieves sepsis-induced liver injury through inactivating the JAK/ STAT pathway. Mol. Med. Rep. 12, 6013–6018. 17. Li, Z., Jia, Y., Feng, Y., Cui, R., Miao, R., Zhang, X., Qu, K., Liu, C., and Zhang, J. (2019). Methane alleviates sepsis-induced injury by inhibiting pyroptosis and apoptosis: in vivo and in vitro experiments. Aging (Albany N.Y.) 11, 1226–1239. 18. Girardot, T., Rimmelé, T., Venet, F., and Monneret, G. (2017). Apoptosis-induced lymphopenia in sepsis and other severe injuries. Apoptosis 22, 295–305. 19. Cho, S.W., Xu, J., Sun, R., Mumbach, M.R., Carter, A.C., Chen, Y.G., Yost, K.E., Kim, J., He, J., Nevins, S.A., et al. (2018). Promoter of lncRNA Gene PVT1 Is a TumorSuppressor DNA Boundary Element. Cell 173, 1398–1412.e22.

Molecular Therapy: Nucleic Acids Vol. 17 September 2019

865

Molecular Therapy: Nucleic Acids

20. He, Y., Jing, Y., Wei, F., Tang, Y., Yang, L., Luo, J., Yang, P., Ni, Q., Pang, J., Liao, Q., et al. (2018). Long non-coding RNA PVT1 predicts poor prognosis and induces radioresistance by regulating DNA repair and cell apoptosis in nasopharyngeal carcinoma. Cell Death Dis. 9, 235.

24. Takasu, O., Gaut, J.P., Watanabe, E., To, K., Fagley, R.E., Sato, B., Jarman, S., Efimov, I.R., Janks, D.L., Srivastava, A., et al. (2013). Mechanisms of cardiac and renal dysfunction in patients dying of sepsis. Am. J. Respir. Crit. Care Med. 187, 509–517.

21. Rhodes, A., Evans, L.E., Alhazzani, W., Levy, M.M., Antonelli, M., Ferrer, R., Kumar, A., Sevransky, J.E., Sprung, C.L., Nunnally, M.E., et al. (2017). Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock: 2016. Intensive Care Med. 43, 304–377.

25. Li, W., Zheng, Z., Chen, H., Cai, Y., and Xie, W. (2018). Knockdown of long non-coding RNA PVT1 induces apoptosis and cell cycle arrest in clear cell renal cell carcinoma through the epidermal growth factor receptor pathway. Oncol. Lett. 15, 7855–7863.

22. Zhou, J., Xiong, Q., Chen, H., Yang, C., and Fan, Y. (2017). Identification of the Spinal Expression Profile of Non-coding RNAs Involved in Neuropathic Pain Following Spared Nerve Injury by Sequence Analysis. Front. Mol. Neurosci. 10, 91. 23. Wang, W., Gao, Z., Wang, H., Li, T., He, W., Lv, W., and Zhang, J. (2016). Transcriptome Analysis Reveals Distinct Gene Expression Profiles in Eosinophilic and Noneosinophilic Chronic Rhinosinusitis with Nasal Polyps. Sci. Rep. 6, 26604.

866

Molecular Therapy: Nucleic Acids Vol. 17 September 2019

26. Tseng, Y.Y., Moriarity, B.S., Gong, W., Akiyama, R., Tiwari, A., Kawakami, H., Ronning, P., Reuland, B., Guenther, K., Beadnell, T.C., et al. (2014). PVT1 dependence in cancer with MYC copy-number increase. Nature 512, 82–86. 27. Colombo, T., Farina, L., Macino, G., and Paci, P. (2015). PVT1: a rising star among oncogenic long noncoding RNAs. BioMed Res. Int. 2015, 304208.