LncRNA INPP5F ameliorates stress‐induced hypertension via the miR‐335/Cttn axis in rostral ventrolateral medulla

Abstract Aims The rostral ventrolateral medulla (RVLM) is an essential vasomotor center responsible for regulating the development of stress‐induced hypertension (SIH). Long non‐coding RNAs (lncRNAs) play critical roles in various physiopathology processes, but existing research on the functions of RVLM lncRNAs on SIH has been lacking. In this study, we investigated the roles of RVLM lncRNAs in SIH. Methods Genome‐wide lncRNA profiles in RVLM were determined by RNA sequencing in a SIH rat model established using electric foot shocks plus noises. The hypotensive effect of lncRNA INPP5F and the underlying mechanisms of lncRNA INPP5F on SIH were explored through in vivo and in vitro experiments, such as intra‐RVLM microinjection and immunofluorescence. Results We discovered 10,179 lncRNA transcripts, among which the lncRNA INPP5F expression level was significantly decreased in SIH rats. Overexpression of lncRNA INPP5F in RVLM dramatically reduced the blood pressure, sympathetic nerve activity, and neuronal excitability of SIH rats. LncRNA INPP5F overexpression markedly increased Cttn expression and reduced neural apoptosis by activating the PI3K‐AKT pathway, and its inhibition had opposite effects. Mechanistically, lncRNA INPP5F acted as a sponge of miR‐335, which further regulated the Cttn expression. Conclusion LncRNA INPP5F was a key factor that inhibited SIH progression, and the identified lncRNA INPP5F/miR‐335/Cttn/PI3K‐AKT/apoptosis axis represented one of the possible mechanisms. LncRNA INPP5F could serve as a therapeutic target for SIH.


| 1831
ZHANG et al. also cause hypertension. 5,6 This condition is often called stressinduced hypertension (SIH). The autonomic nervous system and its sympathetic arm have been confirmed to exert pivotal roles in the pathogenesis of hypertension. 7 Stress activating the sympathetic nervous system's response is believed to be involved in the development of SIH. 8,9 The rostral ventrolateral medulla (RVLM), caudal ventrolateral medulla (CVLM), nucleus tractus solitarius (NTS), and paraventricular nucleus (PVN) are important central sites that control the sympathetic outflow. 10,11 Among them, the RVLM oblongata region contains neurons that receive inputs from various sources, playing a predominant role in controlling sympathetic vasomotor tone and blood pressure. 12 Zhang et al. 13 concluded that PLIN2 knockdown in RVLM could block oxidative/nitrosative stress, alleviate sympathetic overdrive, and suppress SIH progression. Sigma-1 receptor activation inhibited RVLM neuroinflammation and subsequently ameliorated sympathetic hyperactivity and blood pressure in SIH rats. 14 Evidence revealed that NaV1.6 overexpression in RVLM mediated sympathetic activity and SIH development via glutamate regulation. 15 Other research also indicated that the increase in sympathetic outflow by activating RVLM sympathoexcitatory neurons in stress was responsible for generating SIH. 16,17 Thus, dysregulated gene expression in RVLM triggers the augmented sympathetic activity involved in the pathogenesis of SIH. Understanding regulatory networks of gene expression and the underlying molecular mechanisms in RVLM is necessary to slow down SIH development.
Non-coding RNAs (ncRNAs), which are RNA molecules transcribed from the genome DNA but not encoding proteins, serve as master regulators of gene expression in diverse manners. 18 Long non-coding RNAs (lncRNAs) are the best-known and largest class of regulatory ncRNAs that have a size of more than 200 bp. 19 LncRNAs perform a wide range of functions in many complex biological and pathological processes. Several peripheral dysregulated lncRNAs have been proven to be associated with hypertension, such as lncRNA-Ang362, lncRNA PAXIP1-AS1, and lncRNA HOXA-AS3. [20][21][22] However, studies involving the identification of lncRNAs in the cardiovascular center and their effects on sympathetic nervous excitement and blood pressure are still in their infancy. To date, no systematic research that investigated the lncRNA profiles of RVLM participating in the pathological process of SIH has been reported.
In this work, deep RNA sequencing was performed on RVLM tissue isolated from control and SIH rats to evaluate the changes in lncRNA transcriptome systemically upon SIH. A SIH rat model was induced using electric foot shocks plus noises, and it was proposed as a plausible model for exploring the complexity of SIH. [13][14][15][16][17] A total of 39 differentially expressed lncRNAs were uncovered based on in silico analysis upon SIH. In particular, a functionally important lncRNA, called lncRNA INPP5F, was identified. It functioned as a competing endogenous RNA (ceRNA) to target Cttn by competitively sponging miR-335, which is required to inhibit sympathetic discharges and lower blood pressure. To our knowledge, this was the first research to build a roadmap to facilitate the discovery of functional lncRNAs for central regulation of blood pressure involved in SIH. The modulation of lncRNA INPP5F may represent a novel approach for interventional treatment of SIH.

| Ethics statement
All animal experiments were approved by the Animal Care Ethics  Figure 1. The SIH rat model was established as previously described. 15,25 In brief, the rats were placed in a cage (22 cm × 22 cm × 28 cm) with a grid floor, and they received intermittent electric foot shocks (35-80 V for a duration of 50-100 ms), which were controlled by a computer, every 2-30 s. Meanwhile, noises with a level between 88 and 98 dB produced by a buzzer were given as the conditioned stimulus. The rats were subjected to stress stimulation for 2 h twice daily (9-11 a.m. and 3-5 p.m.) for 15 consecutive days. The control rats were placed in cages for the same period of time but not subjected to the stressful stimuli mentioned above.

| Measurement of blood pressure and heart rate (HR)
The systolic blood pressure (SBP), mean arterial pressure (MAP), and HR of rats were measured using ALC-NIBP non-invasive tailcuff system (Alcott Biotech, China) in accordance with the manufacturer's instructions. The detection time points are illustrated in Figure 1. The rats were placed into restraining chambers for 30 min prior to recording to adapt to the measurement procedure.
The whole recording process was kept in a proper environment (34 ± 0.5°C body temperature and noise-free atmosphere). An average value of three replicate measurements for each rat was obtained.

| Renal sympathetic nerve activity (RSNA) recording
As described previously, 25 the RSNA in rats was recorded. In brief, a left flank incision was performed to identify and isolate the renal sympathetic nerve under isoflurane anesthesia (the rats were placed in the induction chamber, the oxygen flowmeter was adjusted to 0.8-1.5 L/min, and the isoflurane vaporizer was adjusted to 3%-5%). A pair of platinum-iridium electrodes were placed on the nerve. Subsequently, Kwik-Sil gel (World Precision Instruments, USA) was used to cover the nerve-electrode complex, and a grass P55C preamplifier was used to amplify (×100) and filter (bandwidth: 100-3000 Hz) the nerve activity. The signal was recorded for 60 min using a Power Lab data acquisition system (RRID: SCR_001620, AD Instruments, Australia). The maximum nerve activity occurred 1-2 min after the rats were euthanized with pentobarbital sodium (≥150 mg/kg, i.p). The background noise level for the nerve activity was calculated 20-30 min after the rats were sacrificed. Baseline RSNA was taken as a percentage of maximum after the background noise was subtracted.

| Plasma norepinephrine (NE) examination
Rat blood samples were collected by cardiac puncture using EDTA as an anticoagulant under inhalational anesthesia with isoflurane as above. They were centrifuged at 1000 g for 15 min at 4°C within 30 min after sample collection, and the supernatant was taken for detection. The level of plasma NE was tested with an ELISA kit (Cat. No. EU2565, FineTest, China) following the manufacturer's specification.

| Total RNA extraction, lncRNA library preparation, and sequencing
The RVLM tissues were extracted by punching coronal sections according to the standard rat atlas. 26

| Sequencing data analysis and identification of differentially expressed lncRNAs
High-quality clean reads were obtained using Cutadapt 27 by removing the reads that contained adaptor contamination, low-quality bases, and undetermined bases from raw data. The clean reads were mapped to the genome of rat (ftp://ftp.ensem bl.org/pub/relea se-104/gtf/rattus_norve gicus/) by using Bowtie2 and Hisat2. 28,29 The mapped reads of each sample were assembled to transcripts by StringTie. 30 Transcripts shorter than 200 bp; less than three reads coverage; and less than one exon and transcripts that overlapped with known mRNAs and other classes of RNAs, such as snRNA, snoRNA, and pseudogenes, were first discarded. The remaining transcripts were then assessed by CPC and CNCI. 31,32 Transcripts with CPC score < −1 and CNCI score < 0 were removed. The qualifying transcripts were selected and considered as lncRNAs. The expression levels of lncRNAs were measured as fragments per kilobase of exon model per million mapped fragments by using StringTie. 30 Differential expression analysis was performed using edgeR. 33 The p value was adjusted using Benjamini-Hochberg method. LncRNAs with p-adjusted value <0.01 between the two groups were considered statistically significant.

| Quantitative reverse transcription polymerase chain reaction (qRT-PCR)
TRIzol Reagent (Cat. No. 15596026, Invitrogen, USA) was used to extract the total RNA from the RVLM tissues or B104 cells. The total RNA was reversely transcribed into cDNA by using Hifair II 1st The primer sequences are listed in Table 1. The measurements were repeated three times for each rat, and the average value was taken.

| Intra-RVLM microinjection
Intra-RVLM microinjection was performed as described in previous study. 15 In brief, the rats were placed in prone position, and the heads were mounted in a stereotaxic apparatus (Model 69,100, RWD Life Science, China) under inhalational anesthesia with isoflurane as described above. The skull was exposed via a midline in-  Table S1). They were microinjected into the bilateral RVLM (located 3.7-4.0 mm caudal to lambdoid suture, 2 mm lateral to the midline, and 8.0 mm ventral to the surface of the dura) at 1 μL/side through a glass micropipette ( Figure S1). Injection sites in the area of the RVLM were confirmed by reference to the standard rat atlas of Paxinos and Watson. 26 The time points of the microinjection are illustrated in Figure 1. After the microinjection was completed, the surgical incision was sutured and covered. No analgesic drugs were administrated after stereotaxic surgery to prevent skewing the results.  Table S1), ASO NC (RiboBio, China; the sequence was protected by a patent from RiboBio), miR-335 agomir (GenePharma, China, Table S1), agomir NC (GenePharma, China, Table S1), miR-335 antagomir (GenePharma, China, Table S1), and antagomir NC (GenePharma, China, Table S1)

| Immunofluorescence
The rats were perfused through the ascending aorta with cold saline, and 4% paraformaldehyde was freshly prepared in PBS after the rats were anesthetized with pentobarbital sodium ( The sections were incubated overnight at 4°C with the following pri-

| Western blot
The RVLM tissues or B104 cells were lysed in RIPA buffer (Cat.

| Flow cytometry
The apoptosis of B104 cells was examined using the Annexin V-Alexa

| Statistical analysis
Data were assessed with version 9.1 GraphPad Prism software and expressed as the mean ± standard error of the mean (SEM). Samples sizes used were similar to those used in our previous studies. 15,17,25 Shapiro-Wilk method was used to evaluate the normality of the data.
The data conformed to a normal distribution. Two-tailed unpaired Student's t-test was conducted to assess the significant differences between two groups, and one-way ANOVA with subsequent post hoc Bonferroni test was performed to compare multiple groups.  Figure 2A and Figure S2A-C). The weight of SIH rats was lower than that in the control rats but markedly increased after the 12th day of stress ( Figure S3). This result was observed to be consistent with our previous study. 17 These data revealed that the SIH rat model was established successfully. RNA sequencing was used to assess the lncRNA expression profiles in RVLM tissues of the control and SIH rats. A total of 524,098,064 raw reads (265,059,924 for SIH rats and 259,038,140 for control rats) were generated.
After the low-quality reads were discarded, 501,346,052 clean reads (254,254,380 for SIH rats and 247,091,672 for control rats) were detected. We aligned the clean reads to the rat reference genome, and the aligning rates were about 76.2% and 75.5% in the SIH and control groups, respectively. A total of 10,179 lncR-NAs were identified and used for subsequent analyses. With the p-adjusted value <0.01 as the significance threshold, 39 differentially expressed lncRNAs in the SIH and control rats were detected. Through expression intensity sorting within SIH and control groups, the five mostly increased and decreased lncR-NAs in SIH rats as compared to that in control rats are shown in Figure 2B. Next, the qRT-PCR was carried out to confirm the expression of these 10 mostly changed lncRNAs in RVLM between the two groups, 4 of which exhibited differential expression and agreed with the lncRNA sequencing results ( Figure 2C). Some data do not correspond to the lncRNA sequencing results ( Figure S4), which may be due to the biological differences between samples.
Among them, we found that lncRNA INPP5F expression was consistently and significantly decreased in SIH rats as compared to that in matched controls ( Figure 2C). Moreover, the tissue-specific expression of lncRNA INPP5F was determined using qRT-PCR.
Next, the results of RSNA recording and plasma NE ELISA test showed that the RSNA and plasma NE values in the SIH + pLV-lncRNA INPP5F rats were robustly decreased compared with those in the SIH and SIH + pLV-NC rats ( Figure 3B,C). Furthermore, c-Fos protein was determined by immunofluorescence assay. The findings suggested that lncRNA INPP5F upregulation significantly reduced the proportion of c-Fos-positive TH+ neurons in RVLM of SIH rats ( Figure 3D). All these data indicated that the increased expression of lncRNA INPP5F restored neuronal excitability, inhibited sympathetic nerve activity, and participated in blood pressure recovery.

| LncRNA INPP5F functioned as a positive regulator of Cttn and restrained apoptosis by activating the PI3K-AKT pathway
Neuronal apoptosis in RVLM contributes to sympathetic overactivity and increases the risk of hypertension. 25,37 Moreover, Cttn strongly affects apoptosis. 38,39 The PI3K-AKT signal pathway has been proven to play an important role in apoptosis. 40,41 Cttn is associated with the PI3K-AKT pathway, 42,43 which participates in apoptosis. 43,44 In the present study, the expression levels of Cttn, p-PI3K, and p-AKT in RVLM were detected by Western blot in SIH and control rats. We observed that compared with the control rats, Cttn, p-PI3K, and p-AKT protein expression levels were markedly decreased in the SIH rats ( Figure 4A). Western blot was also conducted to evaluate the expression of Cttn, p-PI3K, p-AKT, BCL2, BAX, and cleaved Caspase 3 in RVLM of SIH rats following pLV-lncRNA INPP5F plasmid microinjection. As shown in Figure 4B

| LncRNA INPP5F acted as a sponge of miR-335
The underneath mechanism by which lncRNA INPP5F regulated Cttn expression involved in SIH progression continued to be unveiled. Accumulating evidence has confirmed that the lncRNA-miRNA-mRNA network may play a key role in many physiological and pathological processes. 45  protective effects against neuronal apoptosis in SIH. 23 qRT-PCR indicated that the expression of miR-335 was significantly higher in RVLM of SIH rats than that in control rats ( Figure 6A). Figure 2C shows observably decreased compared with that in SIH and SIH + pLV-NC rats ( Figure 6G). The overexpression efficiency of lncRNA INPP5F in RVLM is illustrated in Figure S6A. Taken together, these data confirmed that lncRNA INPP5F served as a sponge for miR-335. , BAX, and cleaved Caspase 3 protein levels were detected by Western blot, cleaved Caspase 3 protein was determined by immunofluorescence, and apoptosis level was tested by TUNEL staining. Data were presented as mean ± SEM. Statistical significance was determined by two-tailed unpaired Student's t-test (A) and one-way ANOVA, followed by post hoc Bonferroni test (B-D). n = 3 rats per group (A, B). n = 12 slices from 6 rats, two slices per rat (C, D). *p < 0.05, **p < 0.01, and ***p < 0.001 versus SIH group. ns means non-significant versus SIH group. DAPI, 4′,6-diamidino-2-phenylindole; NC, negative control; RVLM, rostral ventrolateral medulla; SEM, standard error of the mean; SIH, stress-induced hypertension; TUNEL, TdT-mediated dUTP-biotin nick end labeling.

| DISCUSS ION
Accompanied by the rapid promotion of high-throughput sequencing, it became evident that eukaryotic genome transcribes up to 90% of the genomic DNA, and 98% of these transcripts are transcribed as ncRNAs. 46 LncRNAs are a relatively well-characterized class of  membrane dynamics to cortical actin assembly 48 and has an antiapoptotic effect in various diseases. 38,39,49 The PI3K-Akt signaling pathway modulates diverse cellular processes, including cell apoptosis. 40,41,50 Cttn is considered as an upstream of the PI3K-AKT pathway, 42,43 which is related to apoptosis. 43

CO N FLI C T O F I NTE R E S T S TATE M E NT
The authors have declared that no conflict of interest exists.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request. The lncRNAsequencing clean data reported in this study have been deposited in the NCBI Sequence Read Archive (SRA). The accession number is PRJNA930747.