A preliminary study of the effect of a high-salt diet on transcriptome dynamics in rat hypothalamic forebrain and brainstem cardiovascular control centers

Animals and animal care

The male WKY rats and SHRs used in this study were bred at the University of Malaya Animal Experimental Unit from stock obtained from BioLASCO (Taiwan). After being weaned at five weeks of age, rats were housed in groups of five to six under controlled laboratory conditions (temperature 23 ± 5 °C, 12:12-hour light/ dark cycle and humidity 50% to 60%) with food and water ad libitum for at least a week prior to the onset of experimentation. All the experimental protocols involving animals and housing thereof were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) of the University of Malaya (Reference: 2014-01-07/Physio/R/HSZ) which maintains a full Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) accreditation.

Grouping and experimental design

Six-week-old WKY rats and SHRs were randomly assigned to receive food with either a regular-salt (RS) content (0.2% w/v NaCl) or high-salt (HS) content (4% w/v NaCl; Harlan Teklad, Germany) with free access of water. Four groups were thus studied:

Group 1: WKY rats receiving RS diet (WRS)

Group 2: WKY rats receiving HS diet (WHS)

Group 3: SHRs receiving RS diet (SRS)

Group 4: SHRs receiving HS diet (SHS)

The treatment period continued for six weeks.

Tissue collection and preparation

At the end of diet treatment, rats were euthanized (between the hours of 0800 to 1100) and tissue was isolated and processed as described below. Rats were humanely killed by stunning, followed immediately by decapitation with an animal guillotine (Harvard Apparatus, Holliston, MMA, USA). Brains were quickly removed from the cranium, washed in PBS, then cut into two parts to separate the forebrain and the brainstem which were then snap-frozen in dry ice (within 3 min after stunning) before being stored at −80 °C. The SFO, SON, PVN, NTS and RVLM were accurately localised based on rat brain atlas (George Paxinos & Watson, 2014). Frozen forebrain and brainstem were mounted on the cryostat stage set at −20 °C, equilibrated for 10 min and sectioned at 60 µm using a Thermo Scientific Shandon Cryotome FE and FSEE Cryostat. The sections were mounted on glass slides and stained with Toluidine blue (Sigma Aldrich; 1% w/v in 70% v/v ethanol) then visualised on a light microscope to identify the SFO, SON, PVN (at forebrain) and NTS and RVLM (at brainstem). Once localised, punches of SFO, SON, PVN, NTS and RVLM were then taken with one mm and 0.5 mm micro punches (Fine Scientific Tools) from the unstained tissue. The SFO was collected upon identification of anterior commissure followed by choroid plexus meets the third ventricle to form interventricular foremen. Six to eight consecutive punches were made to collect SFO region. Meanwhile, the supraoptic chiasm which was observance by ‘naked’ eye was used as reference point to collect SON and 12 to 14 serial punches were made to collect this region as defined by the rat brain atlas. Meanwhile, about eight consecutive punches were made upon opening of third ventricle to collect PVN. On the other hand, hypoglossal nucleus and facial nucleus were used as references for dissecting NTS (caudal, intermediate and rostral) and RVLM, respectively. As defined by the rat brain atlas, 20 to 23 consecutive sections were made for NTS, with the first six slices being central punches and the rest bilateral punches. For the RVLM, eight consecutive sections were punched after the disappearance of the facial nucleus. Each of the SFO, SON, PVN, NTS and RVLM punches then dispensed into 1.5ml tubes, suspended QIAzol reagent then stored at −80 °C prior to RNA extraction.

RNA extraction and quality assessment

Total RNA was extracted using Qiagen RNeasy kit protocols (Qiagen, Hilden, Germany). The frozen punched samples were allowed to thaw to ambient temperature and QIAzol phase later was separated with 350 µl chloroform (15 min, 12,000 g, 4 °C). The upper aqueous phase was removed then mixed with 70% (v/v) ethanol to precipitate the total RNA, which was resuspended and applied to RNeasy columns in accordance with the manufacturer’s instruction. For RNAseq experiments, punched samples at QIAzol phase were pooled (5 per group), whereas individual samples were used in qRT-PCR (n = 6). The RNA concentration was measured using a Nanodrop spectrophotometer (ND-2000, Thermo Scientific, Wilmington, DE) and Qubit Fluorometer 2.0 (Life Technologies). The RNA samples were also analysed using 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA 95051) to obtain RNA integrity numbers (RIN numbers) as a measure of their quality (Schroeder et al., 2006). All RNA samples met the RIN quality criterion of >8.5.

RNA-Sequencing analysis

Amplified cDNA libraries were prepared from isolated SFO, SON, PVN, NTS and RVLM RNA samples from SRS and SHS groups (n = 3) and sequenced using Illumina HiSeq 2500 sequencer (Illumina Inc., USA). Briefly, total RNA samples were enriched by hybridization to bead-bound rRNA probes using Ribo Zero kit to obtain rRNA-depleted samples. This was followed by the construction of Illumina libraries using ScriptSeq v2 (Illumina Inc., USA) that applied unique barcode adapters. The libraries were assessed for their quality using Qubit dsDNA High Sensitivity DNA kit and Agilent 2100 Bioanalyzer (Agilent Technologies, A, USA; Agilent High Sensitivity DNA kit). This was followed with further enrichment and amplification of the libraries by qRT-PCR using KAPA Biosystems Library Quantification kit, and normalisation to 2nM. Equal volumes of individual libraries were pooled and run on a MiSeq using MiSeq Reagent kit v2 (Illumina) to validate the library clustering efficiency. The libraries were then re-pooled based on the MiSeq demultiplexing results and sequenced on a HiSeq 2500 sequencing platform (Illumina, San Diego, California, USA) and cBot with v3 flow cells and sequencing reagents. The library reads of greater than 30 to 35 million were generated for each individual library. The data were then processed using RTA and CASAVA thus providing four sets of compressed FASTQ files per library. All raw reads were pre-processed for quality assessment, adaptor removal, quality trimming and size selection using the FASTQC toolkit to generate quality plots for all read libraries. The RNA-Seq alignment and data analysis were all performed in-house using a high-performance computer; “Hydra”. The pipeline made use of Bash and Python scripting to accept RNA-Seq post-trimmed data as input, before ultimately producing output tables of differentially expressed transcripts. Paired-end (2X100bp) raw input data are initially aligned with Tophat to the sixth iteration of the Rattus norvegicus reference genome (Rn6) (Trapnell, Pachter & Salzberg, 2009). HTseq was used to generate read counts, using the ENSEMBL GRCh37 annotation for reference (Anders, Pyl & Huber, 2015). In the present study the pipeline used EdgeR statistical method from the R Bioconductor package to call differential gene expression (DGE) (Risso et al., 2014; Robinson, McCarthy & Smyth, 2010). This allowed us to predict low p-values (p < 0.05) and rank from highest to lowest fold changes (FC) which were utilised in downstream validation. Raw FASTQ files can be found at https://www.ebi.ac.uk/ena/data/view/PRJEB35016. A gene catalogues that described putative changes in gene expression as a consequence of being fed with HS diet in SHRs SFO, SON, PVN, NTS and RVLM were generated. All significantly expressed gene data then filtered to obtain genes expressed at a level of at least 20 read counts in one of the SHS versus SRS comparisons. This followed with sorting the genes according to FC from highest to lowest which later allowed us to select 10 genes (5 in each high and low FC) in each brain regions for subsequent validation with qRT-PCR. As well as validating the SHRs RNA-Seq data, the effect of the HS diet in WKY rats was also included.

Quantitative reverse transcription polymerase chain reaction (qRT-PCR)

The cDNA synthesis was performed using QuantiTect Reverse Transcription kit (Qiagen) using total input RNA of 75ng for SFO, 200 ng for SON and PVN; 300 ng for NTS and 100 ng for RVLM, in accordance with manufacturer’s instructions. Primers for qRT-PCR were designed from NCBI official website (http://www.ncbi.nlm.nih.gov). All primers for target and endogenous control genes were obtained from Integrated DNA Technologies and the primer sequences are provided in Table S1. The qRT-PCR reactions were carried out in duplicate in 96-well plates and each PCR sample consisted of 6 µl 2X SYBR green master mix buffer (Roche), 0.024 µl of both forward and reverse primers 25nmole and 3.953 µl of RNase-free water. The reactions were performed using Applied Biosystems StepOne Plus Real-time PCR system and FCs were assessed by establishing delta-delta cycle threshold (C_T) between Rpl19, a 60-s ribosomal protein L19 as a calibrator gene and target genes. The following temperature profile was used: two minutes at 95 °C for reverse transcription according to the manufacturer’s instruction, followed by 40 cycles at 95 °C for five seconds and 60 °C for 10 s. The average C_T values of target and calibrator genes obtained from qRT-PCR instrumentation were imported into a Microsoft Excel spreadsheet and the ΔΔC_T was calculated using (C)_TTarget - C_TRpl19 as described by Livak & Schmittgen (2001).

Statistical analysis

Statistical analysis was performed using GraphPad Prism (GraphPad Software, La Jolla, CA, USA). All data are expressed as the mean ± standard error of means (SEM) of 6 rats. Comparisons between groups i.e., HS and RS of each individual strains were performed by independent unpaired Student’s t-test. The differences were considered statistically significant at p values <0.05.

RNA-Seq analyses of SFO, SON, PVN, NTS and RVLM of WKY rats and SHRs fed with RS and HS diets

The transcriptomic data composed of gene catalogues with a wide variety of genes of diverse functions for all brain regions that compared the effect of high-salt (HS, 4% NaCl) diet with regular-salt (RS, 0.2% NaCl) diet in SHRs of SFO (Table S2), SON (Table S3), PVN (Table S4), NTS (Table S5) and RVLM (Table S6). All the significantly expressed genes of all brain regions are summarized in a Venn diagram (Figs. 1A to 1E). The Venn diagram explains the change in brain regions of SHRs fed with HS compared with SHRs fed with RS diet and similarly for WKY rats. In addition, the diagram also demonstrates the intersection between WKY rats and SHRs indicating the presence of genes in both strains of rats as consequence of HS diet intake.

In SFO, there were 586 genes found to be significantly changed (edgeR; p < 0.05) in SHRs fed with HS (4% NaCl) diet when compared with SHRs on RS (0.2% NaCl) diet. Meanwhile, there were 614 genes significantly changed in WKY rats receiving HS diet as compared to WKY rats on RS diet. As shown in Fig. 1A, 90 genes observed at the intersection, indicating expression changes of identical genes in SFO in SHRs (SHS compared to SRS) and WKY rats (WHS compared to WRS).

On the other hand, the expression levels of 1,293 genes were significantly (edgeR; p < 0.05) changed in SHRs fed with HS (4% NaCl) diet when compared with SHRs on RS (0.2% NaCl) diet. Meanwhile, 410 genes were found to be significantly changed in WKY rats receiving HS diet as compared to WKY rats on RS diet. As shown in Fig. 1B, 50 genes were observed in the intersection, indicating expression changes of identical genes in SFO in SHRs (SHS compared to SRS) and WKY rats (WHS compared to WRS).

Meanwhile, there were 1,451 genes showed significant changes (edgeR; p < 0.05) in their expression levels in SHRs fed with HS (4% NaCl) diet when compared with SHRs on RS (0.2% NaCl) diet in PVN. A comparison of WKY rats receiving HS diet with WKY rats fed with RS diet showed that the expression levels of 1,727 genes were significantly changed, as displayed in Fig. 1C. The same figure shows an intersection of 844 genes, indicating expression changes of identical genes in PVN in SHRs (SHS compared to SRS) and WKY rats (WHS compared to WRS).

In the NTS, 361 genes were found to be significantly changed (edgeR; p < 0.05) in SHRs fed with HS (4% NaCl) diet when compared with SHRs on RS (0.2% NaCl) diet. Meanwhile, 1,251 genes were significantly changed in WKY rats receiving HS diet as compared to WKY rats on RS diet. As shown in Fig. 1D, 51 genes observed at the intersection, indicating expression changes of identical genes in NTS in SHRs (SHS compared to SRS) and WKY rats (WHS compared to WRS).

Meanwhile, there were 511 genes showed significant changes (edgeR; p < 0.05) in their expression levels in SHRs fed with HS (4% NaCl) diet when compared with SHRs on RS (0.2% NaCl) diet in RVLM. A comparison of the WKY rats receiving HS diet with WKY rats fed with RS diet, showed that the expression levels of 304 genes were significantly changed, as displayed in Fig. 1E. The same figure shows an intersection of 25 genes, indicating expression changes of identical genes in RVLM in SHRs (SHS compared to SRS) and WKY rats (WHS compared to WRS).

qRT-PCR validations of selected genes

The validation analyses for all target genes were compared relative to Rpl19 as in all brain regions, Rpl19 expression showed no significant differences in the expression when compared with that of another two well-known calibrator genes, Gapdh and β-actin, between the four experimental groups (WRS, WHS, SRS and SHS). For each tissue, based on the qRT-PCR analysis, validated genes were categorised as follows:

(A) Genes upregulated in both strains of rat.

(B) Genes downregulated in both strains of rat.

(C) Genes that showed significant upregulation only in SHRs.

(D) Genes that showed significant downregulation only in SHRs.

SFO: Among the 10 genes of the SFO chosen for validation (Table 1), 4 (Apopt1, Lin52, AVP, and OXT) were significantly upregulated only in SHRs when analyzed by qRT-PCR as evidenced in Figs. 2A to 2D.

SON: Among the 10 genes of the SON chosen for validation (Table 2), 2 (Caprin2 and Sctr) were significantly upregulated in both strains of rats (Figs. 3A and 3B); 2 (AVP and OXT) were significantly upregulated only in SHRs (Figs. 3C and 3D) and 2 (Kcnv1 and Ndufaf2) were significantly downregulated only in SHRs (Figs. 3E and 3F) when analyzed by qRT-PCR.

PVN. Among the 10 genes of the PVN chosen for validation (Table 3), 2 (Orc6 and Gkap) were significantly downregulated in both strains of rats (Figs. 4A and 4B); 2 (Caprin2 and Sclt1; Figs. 4C and 4D) were significantly upregulated only in SHRs and 1 (Pi4k2a) was significantly downregulated only in SHRs (Fig. 4E) when analyzed by qRT-PCR.

NTS. Among the 10 genes of the NTS chosen for validation (Table 4), 7 (Rprd1a, Csrnp3, Snrpd2l, Iglon5, Rnf157, St6galnac6 and Ankrd) were significantly downregulated only in SHRs when analyzed by qRT-PCR as evidenced in Figs. 5A to 5G.

RVLM: Among the 10 genes of the RVLM chosen for validation (Table 5), 3 (A4galt, Slc29a4 and Cmc1) were significantly upregulated only in SHRs (Figs. 6A–6C) whilst 2 (Ttr and Faim) were significantly downregulated only in SHRs (Figs. 6D to 6E) when analyzed by qRT-PCR.

General consideration

In the present study, RNA-Seq was used to document changes in gene expression profile in SFO, SON, PVN, the hypothalamic forebrain region for maintaining homeostasis; and NTS and RVLM brainstem CV control centres in SHRs following the consumption of an HS diet. The SHR is a well-documented animal experimental model in functional genetic and physiological studies as it has been attributed to the similarity of its pathophysiology with essential HPN in human (Doggrell & Brown, 1998; Korner, 2010; Leong, Ng & Jaarin, 2015). Several expert panels have reported that SHRs are an excellent model of experimental HPN that could serve as a counterpart for clinical essential HPN as well as model for complications of HPN (Badyal, Lata & Dadhich, 2003). Moreover, it has been proposed that the pathogenesis of HPN in the SHR is heterogeneous with the underpinning of CNS, neurohumoral, renal, and cellular abnormalities (DePasquale et al., 1992; Fortepiani et al., 2003; Sarikonda et al., 2009). Our previous findings showed significant augmentation in mean arterial pressure (MAP) of SHRs fed with HS diet when compared with SHRs on a regular-salt diet. However, there was no significant change in the MAP of WKY rats as a consequence of the consumption of HS-diet. Thus, further confirming that SHRs developed a salt-sensitive component to the established HPN and are a model for studying the central mechanisms of salt-sensitive HPN.

RNA-Seq analysis and qRT-PCR validation

Using the SHR model, catalogues of genes that are differentially expressed in SFO, SON, PVN, NTS and RVLM as a consequence of consuming the HS diet were generated using RNA-Seq analysis which was then validated with qRT-PCR. We also asked if the HS differentially expressed genes validated in the SHRs were also altered in expression by HS in normotensive WKY rats’ SFO, SON, PVN, NTS and RVLM. The comparisons were made between HS and RS diets in both the rat strains that revealed four categories of genes: genes up-regulated (A) or down-regulated (B), in both SHRs and WKY rats; genes up-regulated only in SHRs (C); genes down-regulated only in SHRs (D). Here, one might assume that it is the latter genes (categories C and D) that might have importance in the elevated BP seen in this strain as a consequence of HS. However, it may well be that genes commonly regulated in both strains have different effects due to WKY rats- or SHRs-specific interactions dependent upon genetic background. We also concede that we have revealed a relatively very small number of HS-responsive genes and many more remain to be mined from our RNAseq datasets.

Subfornical organ (SFO)

The SFO is known to be an important CV control region regulating fluid homeostasis by initiating volumetrically controlled thirst and drinking responses (Denton et al., 1999; Freece et al., 2005; Smith, Beninger & Ferguson, 1995; Stachenfeld, 2008). The osmotic information through osmoreceptors from OVLT is transmitted neurally to the hypothalamus and ultimately results in thirst sensation, drinking behaviour and release of AVP; hence, retain water in the body. Thus, the up-regulation of AVP in the present study (Fig. 2C) is in accordance to the claim that the rat’s SFO to have high-affinity binding sites for AVP as well as its degradation product of AVP (Anthes et al., 1997; Jurzak, Fahrenholz & Gerstberger, 1993). Furthermore, the mammalian SFO has been reported to contain vasopressinergic fibre endings from magnocellular or parvocellular portions, and the presence of AVP mRNA in SFO implies that AVP is formed and released there (Anthes et al., 1997). On the other hand, OXT was shown to be expressed in a similar pattern as AVP. Both AVP and OXT have been reported to be secreted simultaneously in response to hyperosmolality and hypovolemia (Haanwinckel et al., 1995; Ventura et al., 2002), and they have synergistic effect on the inner medullary collecting ducts (Ventura et al., 2002). The oxytocinergic fibres have also been reported to be found in SFO suggesting that OXT may affect salt and water intake in rats (Hosono et al., 1999). As such, the up-regulation of OXT level in SFO of SHRs in the present study is defensible.

Table 3:

Summary of validation of putative PVN differentially expressed genes.

Gene	Gene name	RNA-Seq analysis				qRT-PCR validation
		SHS. AvgCount	SRS. AvgCount	EdgeR.FC	EdgeR P-value	Category
Pi4k2a	Phosphatidylinositol 4-kinase Type 2 alpha	232.15	108.97	2.13	4.27E–06	D
Prrc2a	Proline rich coiled-coil 2A	960.03	509.31	1.89	1.38E–06
AVP	Vasopressin	9,331.38	5,415.72	1.73	8.45E–09
Caprin2	Caprin family member 2	2,337.86	1,551.92	1.51	4.57E–04	C
OXT	Oxytocin	37,526.70	28,121.00	1.23	1.86E–03
Sclt1	Sodium Channel and Clathrin Linker 1	1,663.35	1,707.38	0.97	8.46E–01	C
Gkap1	G Kinase Anchoring Protein 1	1,417.12	2,419.13	0.59	2.80E–05	B
Orc6	Origin Recognition Complex Subunit 6	97.98	173.50	0.57	3.29E–02	B
Isy1	SY1 Splicing Factor Homolog	498.23	915.38	0.55	1.67E–05
Nsmce1	NSE1 Homolog, SMC5-SMC6 Complex Component	125.59	235.92	0.54	2.77E–03

DOI: 10.7717/peerj.8528/table-3

Notes:

Category B: downregulated genes in SHRs and WKY rats; Category C: upregulated genes only in SHRs and Category D: downregulated genes only in SHRs when analyzed with qRT-PCR.

Abbreviation

Avg count

average count of the genes’ number

FC

fold change

p-value

the sum of significance

SHS

SHRs fed with HS diet

SRS

SHRs fed with RS diet

Table 4:

Summary of validation of putative NTS differentially expressed genes.

Gene	Gene name	RNA-Seq analysis				qRT-PCR validation
		SHS. AvgCount	SRS. AvgCount	EdgeR.FC	EdgeR P-value	Category
Nkain4	Sodium/Potassium Transporting ATPase Interacting 4	165.64	110.68	1.50	1.38E–02
St6galnac6	ST6 N-Acetylgalactosaminide Alpha-2,6-sialyltransferase 6	638.48	436.19	1.47	2.83E–03	D
Ankrd29	Ankyrin Repeat Domain 29	161.05	142.31	1.50	4.96E–02	D
Rnf157	Ring Finger Protein 157	912.69	673.41	1.36	3.67E–02	D
Rprd1a	Regulation of Nuclear Pre-mRNA Domain Containing 1A	695.73	899.43	0.77	3.70E–02	D
Csrnp3	Cysteine and Serine Rich Nuclear Protein 3	980.12	1,302.83	0.75	2.56E–02	D
Pdyn	Prodynorphin	488.66	753.66	0.65	5.09E–03
Snrpd2l	Small Nuclear Ribonucleoprotein D2-like	199.17	377.78	0.53	5.68E–03	D
Romo1	Reactive Oxygen Species Modulator 1	159.40	119.00	3.06	2.43E–06
Iglon5	IgLON Family Member 4	128.26	155.44	0.51	4.71E–03	D

DOI: 10.7717/peerj.8528/table-4

Notes:

Category D: downregulated genes only in SHRs when analyzed with qRT-PCR.

Abbreviation

Avg count

average count of the genes’ number

FC

fold change

p-value

the sum of significance

SHS

SHRs fed with HS diet

SRS

SHRs fed with RS diet

In addition, both RNA-Seq and qRT-PCR data also revealed significant up-regulation of Lin52 and Apopt1 (Table 1 and Figs. 2A and 2B) in SHRs being fed with HS diet. The Lin52 (protein Lin52) which was significantly up-regulated in SHS has been associated with clinical CV events among African Americans from atherosclerosis risk communities’ study. A genome-wide analysis of meta-analysis study by Shendre et al. (2017) showed the presence of Lin52 gene in African-American of CV events such as stroke, atherosclerosis and myocardial infarction incidents. However, the study did not further evaluate the functional role of this gene with the occurring of these diseases and suggested further studies. Hence, our finding may propose that HS intake could be the initial triggering factor leading to CV diseases. However, its certainty needs further exploration. Meanwhile, the Apopt1, an apoptogenic protein 1, has been shown to be localised within mitochondria and related with mitochondrial disorders characterised by cyclooxygenase (COX) deficiency (Melchionda et al., 2014). The COX deficiency is one of the most common biochemical abnormalities found in mitochondrial disorders with half of all cases remain genetically undefined. Moreover, a connection between apoptosis and increased reactive oxygen species (ROS) production has been suggested to play an important role in the pathophysiology of mitochondrial diseases associated with COX deficiency (Di Giovanni et al., 2001; Kadenbach et al., 2004; Melchionda et al., 2014). Therefore, the expression of Apopt1 in SFO needs further work to elucidate its functional significance in salt-induced HPN.

Supraoptic nucleus (SON)

The SON is known as a specialised nucleus in the hypothalamus that contains magnocellular neurons that secrete only AVP and OXT (Armstrong, 1995). A large amount of AVP and OXT are released into bloodstream to play a critical role in the regulation of salt and water balance as well as acting as vasoconstrictor, antidiuretic, CV regulator, lactation and affiliative behaviour when the normal physiological osmolality and volume are challanged (Hatton, 1990; Hatton, 1997; Jansen, Van der Zee & Gerkema, 2007; Mlynarik et al., 2007; Young, 1999). Thus, the high rates of neuropeptide synthesis, transport, and release in AVP and OXT have made SON as an important experimental model for the study peptidergic neuronal cell biology (Burbach et al., 2001; Hatton, 1990). The present RNA-Seq and qRT-PCR analyses of SON showed up-regulation of AVP and OXT (Table 2 and Figs. 3C and 3D) genes in SHRs as a result of consuming HS. These findings are in accordance with many other previous studies that have shown large increases in expression of the principal neuropeptide genes, AVP and OXT in magnocellular neurons (MCNs) in SON during salt-loading (Burbach et al., 2001; Greenwood et al., 2015b; Johnson et al., 2015c). It has been claimed that chronic salt loading produces large increase in volumes of MCNs which are recognized to be due to global increase in transcription and protein synthesis in the MCNs under hyperosmotic stimulations (Johnson et al., 2015c; Miyata, Nakashima & Kiyohara, 1994; Zhang et al., 2001). In addition, the increased AVP level in SHRs is in accordance with the hypothesis that SHRs have high AVP release i.e., high transcription of AVP as increasing levels of mineralocorticoid receptor and mineralocorticoids have been regarded to be involved in central regulation of BP, and they are known to drive the release of AVP to maintain hypertension (Carmichael & Wainford, 2015; Pietranera et al., 2012). Hence, it is not a surprise to observe a drastic increase of AVP in SHRs fed HS but there remains a need to elucidate the mechanism of how AVP related to salt-sensitive HPN. The increased expression of AVP is associated with an increased expression of Caprin2. The Caprin2 is a gene that encodes RNA binding protein that plays a role in central osmotic defense receptor (Konopacka et al., 2015; Loh et al., 2017). It was found to directly bind to mRNA that encodes AVP and responsible to increase the length of poly-A tail (structures added to the end of all newly-made mRNAs as more AVP mRNAs molecules needed to produce during dehydration) (Konopacka et al., 2015). The team also concluded that Caprin2 plays a critical role in mediating brain responses to osmotic stress as it is expressed in MCNs in SON and PVN, and that Caprin2 expression in these neurones increases during osmotic stress. The study further demonstrated that Caprin2 knockdown in SON and PVN disrupts physiological osmoregulatory mechanisms. Therefore, the up-regulation of Caprin2 in the present study (Fig. 3A) further strengthen the findings of studies by Konopacka et al. (2015). On the other hand, the OXT which is known to be co-secreted with AVP in response to osmotic and blood volume changes also been reported to induce a decrease in MAP and total peripheral resistance (Dutil et al., 2001; Petersson et al., 1996). This further suggest that OXT may play an important role in the control of body fluids homeostasis, causing natriuresis and vasodilatation, and reducing cardiac contractility and heart rate. Thus, the upregulation of OXT in SHRs fed with HS (Fig. 3D) suggesting that OXT along with AVP is transcribed more than normal levels induced by HS diet in SHRs in order to maintain BP.

Table 5:

Summary of validation of putative RVLM differentially expressed genes.

Gene	Gene name	RNA-Seq analysis				qRT-PCR validation
		SHS. AvgCount	SRS. AvgCount	EdgeR.FC	EdgeR P-value	Category
A4galt	Alpha 1,4-galactosyltransferase	150.17	122.08	2.32	2.62E–03	C
Pnmt	Phenylethanolamine-N-methyltransferase	159.04	71.17	2.27	2.51E–03
Snrpd2l	Small Nuclear Ribonucleoprotein D2-like	426.26	214.49	1.99	4.60E–02
Slc29a4	Solute Carrier Family 29 Member 4	425.29	242.05	1.77	2.75E–03	C
Gmfb	Glia Maturation Factor, Beta	1,271.81	735.76	1.75	7.62E–03
Ttr	Transthyretin	237.20	1,103.01	0.22	1.35E–02	D
Gmfg	Glia Maturation Factor, gamma	125.93	178.24	0.34	1.68E–03
Cmc1	C-x(9)-C Motif containing 1	82.97	206.42	0.41	1.65E–07	C
S100a6	S100 Calcium Binding Protein A6	130.77	262.70	0.50	3.29E–03
Faim	Fas Apoptotic Inhibitory Molecule	278.34	386.77	0.73	3.80E–02	D

DOI: 10.7717/peerj.8528/table-5

Notes:

Category C: upregulated genes only in SHRs and Category D: downregulated genes only in SHRs when analyzed with qRT-PCR.

Abbreviation

Avg count

average count of the genes’ number

FC

fold change

p-value

the sum of significance

SHS

SHRs fed with HS diet

SRS

SHRs fed with RS diet

In addition, the Sctr, a secretin receptor is known as a potent regulator of pancreatic bicarbonate, electrolyte and volume secretions (Baiocchi et al., 1999; Chow et al., 2004; Chu et al., 2007) was found to up-regulated in both SHRs and WKY rats fed with HS diet (Fig. 3Aii). The secretin though originally isolated from upper intestinal mucosal extract, has been associated with translocation of water channels such as translocation of aquaporin 2 triggered by AVP and OXT to and from plasma membrane in renal tubules, a critical role for renal water absorption (Chu et al., 2007; Jeon et al., 2003; Noda & Sasaki, 2005). Since AVP has been proved in regulating renal water reabsorption and some of these mechanisms have been associated with cyclic-AMP-protein kinase A-mediated phosphorylation of water channels, it has been hypothesized that secretin could modulate renal water permeability by inducing AVP-independent translocation of functional aquaporin 2 (Chu et al., 2007; Jeon et al., 2003; Li et al., 2006; Lorenz et al., 2003; Wilke et al., 2005) which the water homeostasis is very much related to salt-sensitive HPN.

On the other hand, Ndufaf2 and Kcnv1 (Figs. 3E and 3F) were found to be down-regulated in SHRs fed with HS diet. Mutation of Ndufaf2 has been reported in mitochondrial related diseases in humans (Anderson et al., 2013; Chinnery & Hudson, 2013; Koene et al., 2012); however, its exact role pertaining to HPN has not been documented. Meanwhile, the Kcnv1, voltage-gated channels have been reported as one of the gene expressed in preeclampsia, a hypertensive condition in pregnant women (Chang et al., 2011). To the best of our knowledge, the information of this gene associating with SSH is still lacking and not well documented in animal models.