Stress Reactivity, Susceptibility to Hypertension, and Differential Expression of Genes in Hypertensive Compared to Normotensive Patients

Although half of hypertensive patients have hypertensive parents, known hypertension-related human loci identified by genome-wide analysis explain only 3% of hypertension heredity. Therefore, mainstream transcriptome profiling of hypertensive subjects addresses differentially expressed genes (DEGs) specific to gender, age, and comorbidities in accordance with predictive preventive personalized participatory medicine treating patients according to their symptoms, individual lifestyle, and genetic background. Within this mainstream paradigm, here, we determined whether, among the known hypertension-related DEGs that we could find, there is any genome-wide hypertension theranostic molecular marker applicable to everyone, everywhere, anytime. Therefore, we sequenced the hippocampal transcriptome of tame and aggressive rats, corresponding to low and high stress reactivity, an increase of which raises hypertensive risk; we identified stress-reactivity-related rat DEGs and compared them with their known homologous hypertension-related animal DEGs. This yielded significant correlations between stress reactivity-related and hypertension-related fold changes (log2 values) of these DEG homologs. We found principal components, PC1 and PC2, corresponding to a half-difference and half-sum of these log2 values. Using the DEGs of hypertensive versus normotensive patients (as the control), we verified the correlations and principal components. This analysis highlighted downregulation of β-protocadherins and hemoglobin as whole-genome hypertension theranostic molecular markers associated with a wide vascular inner diameter and low blood viscosity, respectively.


Introduction
Hypertension is a fatal yet preventable risk factor of ischemic heart disease [1], the top cause of death worldwide [2]. Besides essential hypertension, there are many cases of hypertension that are not clinically classified as essential. In all these cases, there is an increase in intravascular pressure (only local sometimes) together with vascular shear stress, oxidative stress, inflammatory reactions, and remodeling of the vascular wall. These pathogenic mechanisms common to all hypertensive conditions share, at least in part, a molecular basis that we are trying to pinpoint here.
In our previous studies within this mainstream paradigm, we measured stress reactivity in rats [59] and created an inbred ISIAH rat strain (i.e., inherited stress-induced arterial hypertension) [60] and two outbred strains-tame and aggressive rats-corresponding to low and high stress reactivity [61][62][63][64]. On this basis, we sequenced transcriptomes in the brain stem [ [42] in OXYS rats (ICG SB RAS, Novosibirsk, Russia), which spontaneously develop the acceleratedsenescence phenotype against a background of moderately high blood pressure [65][66][67][68] with respect to normotensive Wistar rats. In the present work, we meta-analyzed our eight abovementioned RNA-Seq datasets to ensure out of caution that among them (together with those available in PubMed [69]), there are still no invariant molecular markers of hypertension. Accordingly, we sequenced the hippocampal transcriptome of tame compared to aggressive rats and identified the stress-reactivity-related rat DEGs and-using our bioinformatics model [70][71][72]-compared them by homology with all the available hypertension-related animal DEGs that we could find. The results were verified using the DEGs of hypertensive versus normotensive patients.

RNA-Seq and Mapping to the Reference Rat Genome
We sequenced the hippocampal transcriptome of three adult male tame gray rats (Rattus norvegicus)-in comparison with that of three aggressive ones-on an Illumina NextSeq 550 system (see Section 4.2). We chose the hippocampus because its functions contribute to learning under stress [73]. The rats were derived from two outbred tame and aggressive strains selectively bred at the ICG SB RAS [59,64] for over 90 generations using the glove test as described elsewhere [74]. The rats were not consanguineous (see Section 4.1). This procedure yielded 169,529,658 raw reads of 75 nt in length (Table 1); we deposited them in the NCBI SRA database [75] (ID PRJNA668014). In Table 1, the reader can see that 146,521,467 reads could be aligned with rat reference genome Rn6 and yielded 14,039 genes expressed within the hippocampus of the rats under study. Using Fisher's Z-test with Benjamini's correction for multiple comparisons, we found 42 DEGs that were not hypothetical, tentative, predicted, uncharacterized, or proteinnon-coding genes; this approach reduced the false-positive error rates (Tables 1 and 2).

Quantitative PCR (qPCR)-Based Selective Verification of the DEGs Identified in this Work in the Hippocampus of Tame versus Aggressive Rats
First, we used 16 additional unrelated rats, namely: eight aggressive and eight tame rats that scored "-3" and "3", respectively, on a scale from -4 (most aggressive rat) to 4 (tamest rat) in the glove test [74] conducted one month before the extraction of hippocampus samples (Table 3). Next, among the 42 DEGs listed in Table 2, we chose Ascl3 and Defb17; our qPCR data on them in the hippocampus of the tame and aggressive rats (see Section 4.4) are in Table 3 as the "mean ± standard error of the mean" (M 0 ± SEM) of their expression relative to four reference genes (B2m, Hprt1, Ppia, and Rpl30) [76] in triplicate. Arithmetic-mean estimates of the expression levels of each gene (Ascl3 and Defb17) in the hippocampus of these tame and aggressive rats in question are given in Table 3 and Figure 1a.
According to both the Mann-Whitney U test and Fisher's Z-test, both Ascl3 and Defb17 are significantly overexpressed in the hippocampus of the tame (white bars) versus aggressive (grey bars) rats according to the qPCR data obtained here (Figure 1a: p < 0.05, asterisks), consistently with the RNA-Seq data ( Table 2). Figure 1b depicts a significant Pearson's linear correlation (p < 0.00005), Spearman's rank correlation (p < 0.05), and Kendall's rank correlation (p < 0.05) between the log2 values (hereinafter, log2: the log2transformed ratio of an expression level of a given gene in tame rats to that in aggressive rats) for five genes-Ascl3, Defb17, B2m, Ppia, and Rpl30 (open circles)-within the RNA-Seq (X-axis) and qPCR (Y-axis) data obtained here. Note. Hereinafter, log2: the log2-transformed fold change (i.e., ratio of an expression level of a given gene in tame rats to that in aggressive rats); p and P ADJ : statistical significance according to Fisher's Z-test without and with the Benjamini correction for multiple comparisons, respectively.

Comparison of the Known DEGs (of Hypertensive versus Normotensive Animals) with Their Homologous Genes among the 42 Hippocampal DEGs (of Tame versus Aggressive Rats) Identified Here
In this study, using the PubMed database [69], we compiled all the transcriptomes (that we could find) of hypertensive versus normotensive animals, as presented in Table 4. The total number of DEGs was 4216 in 14 tissues of four animal species, as cited in the rightmost column of Table 4 [7,40-57].  Table S1 (hereinafter: see Supplementary Materials)), as shown in Figure S1 using a Venn diagram and in the table. Next, for the first time. we found that stress-reactivity-related and hypertension-related log2 values of the homologous animal DEGs statistically significantly correlate with each other according to Pearson's linear correlation (r = −0.29, p < 0.0005), the Goodman-Kruskal generalized correlation (γ = −0.20, p < 0.0005), and Spearman's (R = −0.29, p < 0.00025) and Kendall's (τ = −0.20, p < 0.0005) rank correlations. Finally, we processed Table S1 by principal component analysis in the Bootstrap mode of the PAST4.04 software [77] that yielded principal components PC1 and PC2, corresponding to a half-difference and half-sum of the stress reactivity-related and hypertension-related log2 values of the homologous animal DEGs ( Figure S1).

Verification of the Results Obtained on the Hypertensive versus Normotensive Animals Examined in this Work with respect to the DEGs-Of Hypertensive versus Normotensive Patients-That We Could Find
Using the PubMed database [69], we collected all the DEGs (of hypertensive compared with normotensive patients) that we could find ( Table 5). The total number of hypertensionrelated human DEGs found was 7865, as cited in the rightmost column of Table 5 [26-39].  Figure 2 shows exactly how we reproduced step-by-step the results obtained from the hypertension-related animal DEGs only by replacing them with the hypertension-related human DEGs (Table 5) as independent control clinical data that are documented in Table  S2. The lower half of this figure presents robust correlations between the stress-reactivityrelated and hypertension-related log2 values corresponding to animal and human DEG homologs as well as principal components PC1 and PC2 proportional to the half-difference and half-sum, respectively, of these log2 values; this was the essence of the verification.

Searching for the Hypertension-Related Molecular Markers among the Human Genes Orthologous to the 42 Hippocampal DEGs (of Tame versus Aggressive Rats) Identified in this Work
To this end, first of all, using the PubMed database [69], we characterized each of the 42 hippocampal DEGs (of tame versus aggressive rats) identified in this work (Table 2), in terms of how downregulation or upregulation of their orthologous human genes can manifest itself in hypertension, as presented  in Table S3 (hereinafter: see Supplementary Materials).

Figure 2.
A step-by-step diagram of verification of the obtained results on the hypertensive versus normotensive animals examined in this work with respect to all the transcriptomes (that we could find) of hypertensive versus normotensive patients (Table 5). Legend: see the footnote of Table 2; PC1 and PC2: principal components calculated using the PAST4.04 software [77].

Searching for the Hypertension-Related Molecular Markers among the Human Genes Orthologous to the 42 Hippocampal DEGs (of Tame versus Aggressive Rats) Identified in this Work
To this end, first of all, using the PubMed database [69], we characterized each of the 42 hippocampal DEGs (of tame versus aggressive rats) identified in this work (Table 2), in terms of how downregulation or upregulation of their orthologous human genes can manifest itself in hypertension, as presented  in Table S3 (hereinafter: see Supplementary Materials).
Next, for each hippocampal DEG (of tame versus aggressive rats) in question ( Table  2), we determined how many homologous DEGs of hypertensive versus normotensive subjects (i.e., patients and animals) have the opposite (NPC1) or the same (NPC2) sign of their log2 values related to hypertension, as compared with the sign of the log2 value of this  (Table 5). Legend: see the footnote of Table 2; PC1 and PC2: principal components calculated using the PAST4.04 software [77].
Next, for each hippocampal DEG (of tame versus aggressive rats) in question (Table 2), we determined how many homologous DEGs of hypertensive versus normotensive subjects (i.e., patients and animals) have the opposite (N PC1 ) or the same (N PC2 ) sign of their log2 values related to hypertension, as compared with the sign of the log2 value of this hippocampal DEG in tame versus aggressive rats, because principal components PC1 and PC2 correspond to a half-difference and half-sum, respectively, of these log2 values ( Figure S1 and Figure 2). Table 6 presents these determined quantities (N PC1 and N PC2 ) together with their statistical significance assessed via the binomial distribution both without (p-values) and with (P ADJ -values) Bonferroni's correction for multiple comparisons. Note. p and P ADJ : a significance estimate according to the binomial distribution without or with Bonferroni's correction for multiple comparisons, respectively; ND: not detected; underlining: statistically significant hypertension-related molecular markers identified in this work.
As shown in this table, only two of the 42 DEGs (in the hippocampus of the tame versus aggressive rats) found here are linked with PC1 (i.e., Hbb-b1 and Pcdhb9, as described in Table 7). Looking through Table 7, readers can see the statistically significant upregulation of both β-protocadherin and hemoglobin subunit DEGs in the tissues of the hypertensive versus normotensive subjects (patients and animals). This result allowed us to propose the statistically significant downregulation of their homologous DEGs (in the hippocampus of tame versus aggressive rats), identified here ( Table 2) as candidate hypertension theranostic molecular markers.

Verification of Downregulation of Human β-Hemoglobin and β-Protocadherins as HypertensionTtheranostic Molecular Markers using the DEGs (That We Could Find) of Domestic versus Wild Animals
For this purpose, using the PubMed database [69], we collected all the transcriptomes (that we could find) of domestic animals compared with their wild congeners, as shown in Table 8. The bottom row of this table indicates that we found 2393 DEGs in the tissues of domestic versus wild animals, as cited in the rightmost column of this table [72,[187][188][189][190][191][192][193].  Using the 42 DEGs (from the hippocampus of the tame versus aggressive rats) identified here (Table 2), together with these 2393 DEGs of domestic versus wild animals (Table 8), we revealed three β-protocadherin DEGs and seven hemoglobin subunit DEGs, which are compared in Table 9 with the human homologous genes (HBB, HBD, and PCDHB9), annotated with respect to hypertension in Table S3. Within columns viii and ix of this table, we transformed the log2 value characterizing the animal hemoglobin subunit and β-protocadherin DEGs into either underexpression or overexpression of the corresponding gene during divergence of domestic and wild animals from their most recent common ancestor, which is the most widely used phylogeny concept [194][195][196][197][198]. Downregulation of human genes HBB and HBD reduces blood viscosity [199] and corresponds to downregulation of the homologous genes Hbb-b1, Hbbl, Hba1, Hbad, Hbm, and Hbz1 in the tame rat, domestic chicken, or dog during their divergence from their most recent ancestors with respect to their wild congeners ( Table 9). As for human hemoglobin upregulation, a high-altitude environment provokes both hypertension and hyperhemoglobinemia [103]. This hemoglobin upregulation in humans corresponds to high hemoglobin subunit levels in aggressive rats [72], wolves [189], and wild chickens [193] during their microevolution (Table 9). Likewise, human gene PCDHB9 (protocadherin β9) downregulation leads to a wide vascular inner diameter [200] and corresponds to downregulation of β-protocadherins in tame rats [72] and domestic rabbits [188] during their microevolution (Table 9). Finally, PCDHB9 upregulation in humans elevates the risk of gastric cancer [201] (the surgical removal of which leads to hypertensive remission [12]) and corresponds to upregulation of β-protocadherins in aggressive rats [72] and wild rabbits [188] during their microevolution ( Table 9). As a standard Fisher's 2 × 2 table, Table 10 summarizes the observations detailed in Table 9.
As one can see in Table 10, downregulation of the genes of β-protocadherins and hemoglobin subunits, which were associated with a wide vascular inner diameter [200] and low blood viscosity [199], respectively, was observed only in domestic animals (not in their wild congeners). This difference is statistically significant according to the binomial distribution (p < 0.0001), Pearson's χ 2 test (p < 0.001), and Fisher's exact test (p < 0.001). Thus, downregulation of β-protocadherins and downregulation of hemoglobin subunits in animals are molecular markers of low stress reactivity [24], which is both a key physiological trait for domestic animals [61,62] and a clinically proven hypertension theranostic physiological marker in everyone, everywhere, anytime [23]. Table 9. Comparing the effects of changes to the expression of homologous genes (a) on hypertension development in humans and (b) during the divergence of domestic and wild animals from their most recent common ancestors.

Discussion
Here, we observed for the first time that downregulation of hemoglobin subunits or β-protocadherins corresponds to low blood viscosity or a wide vascular inner diameter, i.e., two universal genome-wide hypertension theranostic molecular markers applicable to everyone, everywhere, anytime, as readers can see in Table 7. Because of atherosclerosis comorbid with hypertension, this may support our previous finding that natural selection against underexpression of atheroprotective genes slows atherogenesis [202].
Nevertheless, it seems to be highly debatable how low expression levels of human genes HBB, HBD, and PCDHB9 would be adaptive under natural selection, favoring their downregulation that could cause their loss. For this reason, here, we analyzed these genes using our web service SNP_TATA_Comparator [203] applicable to research on hypertension, owing to its successful use in a clinical study on pulmonary tuberculosis [204] comorbid with hypertension [205]. Figure S2 exemplifies how we also used the UCSC Browser [206], Bioperl toolkit [207], and a package of R [208], together with both Ensembl [209] and dbSNP [210] databases in the case of the candidate SNP marker (rs34166473) reducing blood viscosity via HBD downregulation [199], as outlined here (Table S4). In total, we examined all 85 SNPs within the 70 bp proximal promoters of the genes HBD, HBD, and PCDHB9 within build #153 of the dbSNP database [210]. As a result of this work, we found 27 candidate SNP markers of hypertension, as indicated [12,103,[199][200][201]211] in Table S4 and described [212][213][214][215][216][217][218][219][220] in Section S1 "Supplementary methods for DNA sequence analysis" (see Supplementary Materials).
Finally, according to the semicentennial tradition, to assess the relative mutation rates (e.g., transitions versus transversions [224], synonymous versus non-synonymous substitutions [225], and insertions versus deletions [226]), we compared the genes HBB, HBD, and PCDHB9 in question with the human genome as a whole [227][228][229] (Table 11). Notes. Hypertension (HT): normotensive (←HT) and hypertensive (→HT). N GENE and N SNP : total numbers of the human genes and of their SNPs meeting the criteria for this study. N RES : the total number of the candidate SNP markers that can increase (N > ) or decrease (N < ) the affinity of TATA-binding protein (TBP) for these promoters and to respectively affect the expression of these genes. N ←HT and N →HT : total numbers of the candidate SNP markers that can prevent or provoke hypertension. p(H 0 ): the estimate of probability for the acceptance of this H 0 hypothesis, in accordance with the binomial distribution. TBP-site: TATA-binding-protein binding site.
At the top of this table is a genome-wide SNP pattern of TBP sites-where SNPs decreasing the TBP-DNA affinity dominate over SNPs, thus increasing this affinity within the human genome-as predicted by taking into account many mutagenesis molecular mechanisms (e.g., epistatic effects) [227] and as proven within the "1000 Genomes" project [228]. In accordance with Haldane's dilemma [229] and neutral evolution theory [230], this wholegenome trait reflects neutral mutation drift as a norm. At the bottom of Table 11 is the hypertension-related candidate SNP markers identified here, which often significantly reduce the affinity of TBP for promoters of the genes HBB, HBD, and PCDHB9, representing the genome-wide neutral mutational drift antagonizing hypertension.
Altogether, the hypertension-related candidate SNP markers discussed above fit the newest concept [231]: in addition to the accumulation of degenerative SNPs owing to their uncontrollability during neutral mutational drift, some adaptive SNPs can also accumulate in this way (Table 11).

Animals
The study was conducted on adult male gray rats (R. norvegicus) artificially bred for over 90 generations for either aggressive or tame behavior (as two outbred strains). The rats were kept under standard conditions of the Conventional Animal Facility at the ICG SB RAS (Novosibirsk, Russia), as described elsewhere [64, 74,232]. The total number of rats was 22 (11 aggressive and 11 tame ones), each four months old and weighing 250-270 g, all from different unrelated litters. All the rats were decapitated. Using a handbook technique [233], we excised samples of the hippocampus, which were then flash-frozen in liquid nitrogen and stored at −70 • C until use. Every effort was made to minimize the number of animals under study and to prevent their suffering. This work was conducted in accordance with the guidelines of the Declaration of Helsinki, Directive 2010/63/EU of the European Parliament, and of the European Council resolution of 22 September 2010.
The research protocol was approved by the Interinstitutional Commission on Bioethics at the ICG SB RAS, Novosibirsk, Russia (approval documentation no. 8 dated 19 March 2012).

RNA-Seq
Total RNA was isolated from~100 mg of the hippocampus tissue samples of tame (n = 3) and aggressive (n = 3) rats using the TRIzol™ reagent (Invitrogen, Carlsbad, CA, USA). The quality of the total-RNA samples was evaluated using a Bioanalyzer 2100 (Agilent, Santa-Clara, CA, USA). Samples with optimal RNA Integrity Numbers (RINs) were chosen for further analysis. Additionally, the total RNA was analyzed quantitatively on an Invitrogen Qubit™ 2.0 fluorometer (Invitrogen). Different RNA types were separated with the mirVana™ Kit (Thermo Fisher Scientific, Waltham, MA, USA). The Dynabeads mRNA Purification Kit (Invitrogen) was used to prepare highly purified mRNA from 5 µg of the RNA fraction depleted of small RNAs. Preparation of RNA-seq libraries from 15-30 ng of an mRNA fraction was carried out using the ScriptSeq™ v2 RNA-Seq Library Preparation Kit (epicenter ® , Madison, WI, USA). The quality of the obtained libraries was checked on a Bioanalyzer 2100. After normalization, barcoded libraries were pooled and handed over to the Multi-Access Center of Genomic Research (ICG SB RAS, Novosibirsk, Russia) for sequencing on an Illumina NextSeq 550 instrument in a NextSeq ® 500/550 High Output Kit v2 cassette (75 cycles) under the assumption of a direct read of 75 nucleotides, with at least 40 million reads.

Mapping of RNA Sequences to the R. norvegicus Reference Genome
First, the primary raw Fastq files were checked by means of a quality control tool FastQC (https://www.bioinformatics.babraham.ac.uk/projects/fastqc; accessed on 19 December 2018) for high-throughput sequencing data. After that, using the Trimmomatic tool [234], we improved the quality of the raw reads step-by-step as follows: (i) removing a base from either the start or end position if the quality was low, (ii) trimming bases by a sliding-window method, and (iii) removing any remaining reads that were less than 36 bases long. Next, with the help of the TopHat2 toolbox [235], we aligned the trimmed reads to the R. norvegicus reference genome (RGSC Rnor_6.0, UCSC version Rn6, July 2014 assembly). Then, in SAMTools version 1.4 [236], we reformatted these alignments into sorted BAM files. After that, using the htseq-count tool from preprocessing software HTSeq v.0.7.2 [237], along with gtf files carrying coordinates of the rat genes according to Rnor_6.0 and an indexed SAM file, we assigned the reads in question to these genes. Finally, in DESeq2 [238] via Web service IRIS (http://bmbl.sdstate.edu/IRIS/; accessed on 16 January 2020), we rated the differential expression of the abovementioned rat genes, and to minimize false-positive error rates, applied Fisher's Z-test [239] with Benjamini's correction for multiple comparisons, as well as discarded all the hypothetical, tentative, predicted, uncharacterized, and protein-non-coding genes.

qPCR
To selectively and independently verify the tame-versus-aggressive rat hippocampal DEGs found here (Table 2), in this work, we performed a qPCR control assay on the total RNA taken only from the remaining samples of the hypothalamus of tame (n = 8) and aggressive (n = 8) rats. First, with the help of TRIzol™, we isolated total RNA, purified it on Agencourt RNAClean XP Kit magnetic beads (Beckman, #A63987), and quantified it by means of a Qubit™ 2.0 fluorometer (Invitrogen/Life Technologies) along with an RNA High-Sensitivity Kit (Invitrogen, cat. # Q32852). After that, we synthesized cDNA using the Reverse Transcription Kit (Syntol, #OT-1). Next, using web service PrimerBLAST [240], we designed oligonucleotide primers for qPCR (Table 12).

Human Genes under Study
Here, we analyzed the 42 human genes that are orthologous to the 42 hippocampal DEGs of the tame versus aggressive rats (Table 2). Using the PubMed database [69], we characterized each of these 42 human genes in terms of what is already clinically known about how their underexpression or overexpression can manifest itself in hypertension (Table 9 and Tables S3 and S4).

DNA Sequences under Study
For in silico analysis of the human genes encoding candidate molecular markers for hypertension that were for the first time suggested in this work, we retrieved both DNA sequences and SNPs of their 70 bp proximal promoters from the Ensembl database [209] and from the dbSNP database [210], respectively, relative to reference human genome assembly GRCh38/hg38 using the UCSC Genome Browser [206] in the dialog mode and additionally by means of toolbox BioPerl [207] in the automated mode, as shown in Figure S2.

In Silico Analysis of DNA Sequences
We examined SNPs within DNA sequences using our previously developed public web service SNP_TATA_Comparator [203], which applies our bioinformatic model of threestep binding between TBP and a human gene promoter, as detailed in the Supplementary Materials (i.e., Section S1 "Supplementary methods for DNA sequence analysis") and additionally exemplified in Figure S2.

In Vitro Measurements
In this project, we in vitro measured K D values expressed in "moles per liter" units of the equilibrium dissociation constant of TBP promoter complexes by means of the EMSA, for each of the nine chosen candidate SNP markers for hypertension subjected to this experimental verification-i.e., rs35518301:g, rs34166473:c, rs34500389:t, rs33980857:a, rs34598529:g, rs33931746:g, rs33931746:c, rs281864525:c, and rs63750953:deletion-as described in-depth in the Supplementary Materials (i.e., Section S2 "Supplementary methods for in vitro measurement").

Knowledge Base on Domestic Animals' DEGs with Orthologous Human Genes that Can Affect Hypertension
In files with the flat Excel-compatible textual format, here, on the one hand, we first documented all the suggested associations between DEGs (of domestic versus wild animals) homologous to the 42 DEGs (in the hippocampus of tame and aggressive rats) identified in this study. On the other hand, we documented how underexpression or overexpression of the human genes homologous to these hippocampal rat DEGs can affect hypertension. Next, using the MariaDB 10.2.12 web environment (MariaDB Corp AB, Espoo, Finland), we added the current findings to our previously created PetDEGsDB knowledge base, which is publicly available at www.sysbio.ru/domestic-wild (accessed on 16 January 2020).

Statistical Analysis
Using the options in the standard toolbox of Statistica (Statsoft TM ), we applied the Mann-Whitney U test, Fisher's Z-test, Pearson's linear correlation test, the Goodman-Kruskal generalized correlation test, Spearman's and Kendall's rank correlation tests, Pearson's χ 2 test, Fisher's exact test, and binomial-distribution analysis.

Conclusions
First of all, in this work, we performed high-throughput sequencing of the hippocampus transcriptome for three tame adult male rats compared with three aggressive ones (all unrelated animals). The primary experimental data are publicly available for those who would like to use them (NCBI SRA database ID: PRJNA668014) [75].
With the help of this transcriptome, we found the 42 hippocampal DEGs-in the tame versus aggressive rats in question-with statistical significance (P ADJ < 0.05, Fisher's Z-test with Benjamini's correction for multiple comparisons) that was conventionally acceptable (Table 2). Moreover, we selectively validated these DEGs by independent experimental analyses (qPCR) of the other eight tame versus eight aggressive adult male rats from different unrelated litters of the same two outbred strains (Table 3 and Figure 1).
Besides this, using these 42 hippocampal tame-versus-aggressive rat DEGs, which reflect rat stress reactivity, we meta-analyzed (by homology) all the highly specific DEGsof hypertensive versus normotensive subjects (i.e., patients and animals)-that we could find within mainstream hypertension-related transcriptomic research articles. First, we found significant correlations between stress reactivity-related and hypertension-related conventional log2 values (fold changes) of the homologous DEGs analyzed. Next, we found principal components, PC1 and PC2, corresponding to a half-difference and halfsum of these log2 values. Finally, these data pointed to downregulation of hemoglobin or β-protocadherins, corresponding to low blood viscosity [199] or a wide vascular inner diameter [200], as two hypertension theranostic molecular markers applicable to everyone, everywhere, anytime.

Data Availability Statement:
The primary RNA-Seq data obtained in this work were deposited in the NCBI SRA database (ID = PRJNA668014).

Acknowledgments:
We are grateful to Nikolai Shevchuk (Shevchuk Editing Co., Brooklyn, NY, USA) for fruitful discussions of the work and assistance with adapting it to the requirements of the English language for scientific publications. We are also thankful to the Multi-Access Center "Bioinformatics" for the use of computational resources as supported by Russian government project FWNR-2022-0020. We are appreciative of the Center for Genomic Research at the ICG SB RAS, where RNA-Seq was carried out, as supported by the Russian Federal Science and Technology Program for the Development of Genetic Technologies.

Conflicts of Interest:
The authors declare no conflict of interest.