Abstract Background Long noncoding RNAs (lncRNAs) have emerged as critical regulators of the expression of genes involved in cardiovascular diseases. This project aims to identify circulating lncRNAs associated with protein‐coding mRNAs differentially expressed between hypertensive and normotensive individuals and establish their link with hypertension. Methods and Results The analyses were conducted in 3 main steps: (1) an unbiased whole blood transcriptome‐wide analysis was conducted to identify and replicate protein‐coding genes differentially expressed by hypertension status in 497 and 179 Black individuals from the GENE‐FORECAST (Genomics, Environmental Factors and the Social Determinants of Cardiovascular Disease in African‐Americans Study) and MH‐GRID (Minority Health Genomics and Translational Research Bio‐Repository Database) studies, respectively. Subsequently, (2) proximal lncRNAs, termed cis lncRNA quantitative trait loci, associated with each mRNA were identified in the GENE‐FORECAST study and replicated in the MH‐GRID study. Finally, (3) the lncRNA quantitative trait loci were used as predictors in a random forest model to predict hypertension in both data sets. A total of 129 mRNAs were significantly differentially expressed between normotensive and hypertensive individuals in both data sets. The lncRNA‐mRNA association analysis revealed 249 cis lncRNA quantitative trait loci associated with 102 mRNAs, including VAMP2 (vesicle‐associated membrane protein 2), mitogen‐activated protein kinase kinase 3, CCAAT enhancer binding protein beta, and lymphocyte antigen 6 complex, locus E. The 249 lncRNA quantitative trait loci predicted hypertension with an area under the curve of 0.79 and 0.71 in GENE‐FORECAST and MH‐GRID studies, respectively. Conclusions This study leveraged a significant sample of Black individuals, a population facing a disproportionate burden of hypertension. The analyses unveiled a total of 271 lncRNA‐mRNA relationships involving mRNAs that play critical roles in vascular pathways relevant to blood pressure regulation. The compelling findings, consistent across 2 independent data sets, establish a reliable foundation for designing in vitro/in vivo experiments. Keywords: Black individuals, hypertension, long noncoding RNA, mRNA, transcriptome Subject Categories: Hypertension, High Blood Pressure __________________________________________________________________ Nonstandard Abbreviations and Acronyms DBP diastolic blood pressure GENE‐FORECAST Genomics, Environmental Factors and the Social Determinants of Cardiovascular Disease in African‐Americans Study lncrQTL long noncoding RNA quantitative trait loci lncRNA long noncoding RNA MDRD Modification of Diet in Renal Disease MH‐GRID Minority Health Genomics and Translational Research Bio‐Repository Database RF random forests SBP systolic blood pressure Research Perspective. What Is New? * Our study unveiled significant associations between long noncoding RNAs and mRNAs related to hypertension, shedding light on novel mechanisms underlying blood pressure regulation that implicate long noncoding RNAs. * The study sheds light on the relationship between vesicle trafficking, VAMP2 (vesicle‐associated membrane protein 2) functionality, and long noncoding RNAs within the renin‐angiotensin system, providing insights into the molecular pathways contributing to hypertension development. What Question Should Be Addressed Next? * Future research should explore the functional roles of identified long noncoding RNAs and mRNAs, to elucidate their impact on renin release dynamics and blood pressure regulation. According to the Centers for Disease Control and Prevention, ≈1 in every 3 adults in the United States has hypertension, amounting to nearly 75 million people.[30] ^1 The economic burden of hypertension in the United States is staggering. The American Heart Association estimated that hypertension‐related health care costs amounted to ≈$131 billion in 2019.[31] ^2 Black individuals have a higher prevalence of hypertension compared with other racial and ethnic groups in the United States. According to data from the National Health and Nutrition Examination Survey, the age‐adjusted prevalence of hypertension among Black individuals is ≈54% for men and 57% for women, compared with 46% and 48% for White men and women, respectively.[32] ^3 The consequences of high blood pressure are particularly severe for Black individuals; they are more likely to experience earlier onset, more rapid progression, and complications such as stroke, heart disease, and kidney disease.[33] ^4 Hypertension, a multifactorial disorder characterized by chronically elevated blood pressure, arises from the interplay of genetic and environmental factors. Transcriptome analysis, which investigates the complete set of RNA molecules expressed in a given cell or tissue, provides valuable insights into the underlying molecular mechanisms driving hypertension. Transcriptome analysis facilitates the construction of gene regulatory networks, unraveling the intricate interplay between regulatory elements, such as long non‐coding RNAs (lncRNAs), and their target genes in hypertension. lncRNAs are transcripts of >200 nucleotides in length that lack a protein‐coding sequence. They have been shown to play important roles in the pathophysiology of cardiovascular diseases.[34] ^5 Jeong et al reported a lncRNA involved in the regulation of vascular calcification,[35] ^6 and another study revealed a lncRNA that plays a role in endothelial cell functions linked to the development of coronary artery disease.[36] ^7 Several studies have indicated that lncRNAs are implicated in the mechanisms of hypertension, including regulation of the proliferation, migration, and apoptosis of vascular smooth muscle cells and the production of inducible nitric oxide synthase and nitric oxide; lncRNAs also modulate blood pressure and the angiogenic function through their action on endothelial cells.[37] ^8 , [38]^9 , [39]^10 , [40]^11 , [41]^12 Moreover, discernible variations in plasma concentrations of numerous lncRNAs have been observed in individuals afflicted with vascular ailments, implying their significant utility as biomarkers for coronary artery disease.[42] ^13 The objectives of this project are 2‐fold: first, to discern circulating lncRNAs linked to differentially expressed protein‐coding mRNAs in hypertensive versus normotensive individuals across 2 cohorts of Black individuals, comprising a total of 676 participants; and second, to establish associations between these identified lncRNAs and hypertension. METHODS The data included in the analyses are from 2 sources: the GENE‐FORECAST (Genomics, Environmental Factors and the Social Determinants of Cardiovascular Disease in African Americans Study) and the MH‐GRID (Minority Health Genomics and Translational Research Bio‐Repository Database) study. Both the GENE‐FORCAST and the MH‐GRID studies were approved by the National Institutes of Health Institutional Review Board. The studies were conducted in accordance with the local legislation and institutional requirements, and all the participants provided their written informed consent to participate in the studies. The data presented in this article cannot be publicly shared because of privacy restrictions. Requests to access the data sets should be directed to the corresponding author. Phenotype Data The GENE‐FORECAST study is a research platform that establishes a strategic, multi‐omics systems biology approach amenable to the deep, multidimensional characterization of minority health and disease in Black individuals. The GENE‐FORECAST study is designed to create a cohort based on a community‐based sampling frame of self‐identified US‐born, Black men and women (aged 21–65 years) recruited from the metropolitan Washington, DC, area. A description of the baseline characteristics of the 497 GENE‐FOREAST study samples is outlined in Table [43]1. Table 1. Baseline Characteristics Characteristics GENE‐FORECAST (n=497) MH‐GRID study (n=179) Mean or count SD or proportion Mean or count SD or proportion Age, y 48 12 45 7 Sex (female/male) 349/148 70%/30% 67/112 37%/63% Hypertension (control/case) 111/386 22%/78% 78/101 44%/56% Systolic blood pressure, mm Hg 163 37 121 16 Diastolic blood pressure, mm Hg 75 58 77 11 Mean arterial pressure, mm Hg 100 10 89 10 eGFR‐AFR, mL/min per 1.73 m^2 94 23 103 20 Body mass index, kg/m^2 32 8 31 9 [44]Open in a new tab eGFR‐AFR indicates estimated glomerular filtration rate adjusted for African ancestry; GENE‐FORECAST, Genomics, Environmental Factors and the Social Determinants of Cardiovascular Disease in African‐Americans Study; and MH‐GRID, Minority Health Genomics and Translational Research Bio‐Repository Database. MH‐GRID is a study of hypertension in self‐identified Black individuals, aged 30 to 55 years. The data included in this analysis are from an MH‐GRID subset of samples from the Morehouse School of Medicine (Atlanta, GA). A description of the baseline characteristics of the 179 MH‐GRID study samples is outlined in Table [45]1. In this analysis, MH‐GRID study data were used to replicate differential expression analysis and random forests analysis conducted in the GENE‐FORECAST study dataset. In both data sets, individuals with systolic blood pressure (SBP) ≥140 mm Hg or diastolic blood pressure (DBP) ≥90 mm Hg or on ≥1 high blood pressure medications were considered hypertensive or case, whereas those with SBP/DBP ≤120/80 mm Hg without blood pressure medication were considered normotensive or control. SBP and DBP were adjusted for medication using the approach by Tobin et al.[46] ^14 Renal function was measured as estimated glomerular filtration rate (eGFR) based on serum creatinine and was computed using the MDRD (Modification of Diet in Renal Disease) Study equation,[47] ^15 which includes a term to adjust for African ancestry to generate an eGFR measure referred to as eGFR‐AFR. Transcriptome Data The transcriptome data consist of the mRNA sequencing of whole blood samples. RNA extraction: total RNA extraction was performed using MagMAXTM for Stabilized Blood Tubes RNA Isolation Kit, as recommended by the vendor (Life Technologies, Carlsbad, CA). Library preparation: total RNA samples were converted into indexed cDNA sequencing libraries using Illumina's TrueSeq kits. Ribosomal RNA was removed. Sequencing platform: the GENE‐FORECAST study samples were pair end sequenced, on the Illumina HiSeq2500 and HiSeq4000 platforms, with a sequencing depth of at least 50 million reads per sample. For the MH‐GRID study samples, Illumina paired end sequencing was performed on the HiSeq2000 analyzer (Illumina) with a sequencing depth of at least 50 million per sample. Expression quantification: mRNA expression was quantified using a bioinformatics pipeline developed by the Broad Institutes and used by the genotype‐tissue expression. The pipeline is detailed in the GitHub software development platform.[48] ^16 Transcripts that did not achieve an expression of 2 read counts per million in at least 3 samples were excluded. The expression data were normalized using the trimmed mean of M‐values, an optimal method for read count data.[49] ^17 Principal component analysis was conducted to identify and exclude sample and gene outliers. After those quality control filters, 17 948 protein‐coding mRNAs and 9646 lncRNAs were included in the statistical analyses described in the next section. Statistical Analysis The analysis was undertaken in mainly 3 steps, described graphically in Figure [50]1. Figure 1. Overview of the analyses undertaken. Figure 1 [51]Open in a new tab (1) First, mRNAs differentially expressed by hypertension status were identified in GENE‐FORECAST (Genomics, Environmental Factors and the Social Determinants of Cardiovascular Disease in African‐Americans Study) and replicated in the MH‐GRID (Minority Health Genomics and Translational Research Bio‐Repository Database) study. (2) Then, the association between each mRNA and long noncoding RNAs (lncRNAs) in the cis region (within 1 Mb) was evaluated, in both data sets, to uncover lncRNAs that modulate the expression of the mRNA. (3) Finally, the relationship between hypertension and lncRNAs associated with mRNA expression was assessed using random forests. In the initial step, differential expression analysis was performed with the list of 17 948 protein‐coding mRNAs using the R library edgeR, which fits a negative binomial model to mRNA read counts and computes likelihood ratio tests for the coefficients in the model. The model was adjusted for age and sex in both the discovery data (GENE‐FORECAST study) and the replication dataset (MH‐GRID study). mRNAs were considered significantly differentially expressed between hypertensive and normotensive individuals if the Benjamini and Hochberg[52] ^18 multiple testing adjusted P value was ≤0.05 in the discovery data set and the nominal P value was ≤0.05 in the replication data set and the log fold change is in the same direction. The association between the differentially expressed mRNAs and SBP, DBP, mean arterial pressure (MAP), eGFR, and body mass index (BMI) was also investigated in the larger data set (GENE‐FORECAST study). Subsequently, an adapted version of the R library MatrixEQTL was used to identify lncRNAs in the cis region (1 megabase around the mRNA) associated with each differentially expressed mRNA and referred to as lncRNA quantitative trait loci (lncrQTL). MatrixeQTL was used to conduct individual tests for each lncRNA–mRNA pair; multiple testing was addressed using the false discovery rate correction proposed by Benjamini and Hochberg.[53] ^18 The model was adjusted for age and sex in both the discovery data (GENE‐FORECAST study) and the replication data (MH‐GRID study). A lncrQTL‐mRNA association was considered statistically significant and replicated if the multiple testing adjusted P value was ≤0.05 in the discovery data and the nominal P value was ≤0.05 in the replication data, and the β value is in the same direction, in both datasets. Then, random forests (RF) were used to establish the relationship between the lncrQTL and hypertension in the 2 datasets. This ensemble technique was used to assess the cumulative impact of lncrQTL on hypertension by modulating the expression of mRNAs associated with hypertension. RF is a tree‐based machine learning technique that makes no assumptions about the relationship (eg, linear) between the predictors and the outcome and can capture higher‐order interactions that cannot be readily included in regression models.[54] ^19 , [55]^20 Unlike linear models, RF predictive power is not much affected by multicollinearity because of the random sampling of features and predictor variables, which ensures a different set of data points is modeled for each tree.[56] ^21 For this project, the R library randomForest, an R implementation of the algorithm developed by Breiman and Cutler,[57] ^21 was used to identify the set of lncRNAs that classifies hypertension status in both datasets. Finally, pathway enrichment analysis was performed to identify molecular pathways enriched in the list of mRNAs associated with cis lncRNA. This analysis was conducted using the R algorithm pathfindR,[58] ^22 a tool that defines subnetworks in a cluster of genes and identifies pathways overrepresented in those subnetworks. RESULTS Differential expression analysis was conducted to discern mRNA expression variations between hypertensive and normotensive individuals across 2 data sets: GENE‐FORECAST study (386 hypertensive versus 111 normotensive individuals) and MH‐GRID study (101 hypertensive versus 78 normotensive individuals). The analysis identified 129 mRNAs exhibiting statistically significant differential expression in the GENE‐FORECAST study, a finding replicated in the MH‐GRID study. A detailed inventory of these differentially expressed mRNAs, along with their respective log fold change values and corresponding raw and adjusted P values, is provided in Table [59]S1. Additionally, Figure [60]2 illustrates a graphical representation of this differential expression, highlighting the top 10 mRNAs associated with cis‐regulatory lncRNAs. Figure 2. Top 10 differentially expressed mRNAs that are associated with 1 or more long noncoding RNAs. Figure 2 [61]Open in a new tab The log fold change (logFC) and false discovery rate–adjusted (adj.) P value of the difference between the 2 groups are provided below the name of the mRNA. CEBPB indicates CCAAT enhancer binding protein beta; CHST12, carbohydrate sulfotransferase 12; ISG15, ubiquitin like modifier; LY6E, lymphocyte antigen 6 complex, locus E; MAP2K3, mitogen‐activated protein kinase kinase 3; POLR2J2, RNA polymerase II subunit J2; SIGLEC14, sialic acid binding Ig like lectin 14; SLC38A5, solute carrier family 38 member 5; UCP2, uncoupling protein 2; and VAMP2, vesicle‐associated membrane protein2. We conducted tests on all lncRNAs located within the cis‐regulatory region (1 Mb) of the 129 differentially expressed mRNAs to evaluate their correlation with mRNA expression levels. This analysis revealed 271 distinct lncRNA‐mRNA associations that demonstrated statistically significant correlations after test in the GENE‐FORECAST study (n=497) and replication in the MH‐GRID study (n=179). These associations involved 249 unique lncRNAs, termed lncrQTLs, linked to 102 unique mRNAs distributed across all chromosomes except for chromosomes 4 and 13. Among the identified associations, 128 featured lncRNAs located upstream of the corresponding mRNA, whereas 119 associations involved lncRNAs located downstream of the mRNA. Furthermore, 20 associations revealed lncRNAs overlapping with the 5′ untranslated region of the mRNA. In 3 instances, the lncRNA overlapped with the 3′ untranslated region, whereas in 1 instance, the lncRNA was positioned within the mRNA sequence. Detailed results of the cis‐regulatory lncrQTL analysis are available in Table [62]S2. Table [63]2 reports the lncRNAs associated with the top 10 differentially expressed mRNAs, and Table [64]3 provides a comprehensive breakdown of the lncrQTL‐mRNA associations per chromosome. Table 2. cis lncRNA Associated With the Top 10 mRNAs Differentially Expressed Between Normotensive and Hypertensive Individuals lncRNA mRNA Chromosome lncRNA location β (GENE‐FORECAST) Adjusted Pvalue (GENE‐FORECAST) β (MHGRID) P value (MH‐GRID) LINC02904 LY6E 8 15.4 kb Upstream 116.99 1.52e‐28 27.15 1.71e‐06 MINCR 256.3 kb Upstream −20.56 3.63e‐03 −15.52 1.67e‐02 SIGLEC5 SIGLEC14 19 5′ UTR 23.07 4.18e‐140 14.91 2.01e‐14 HSALNG0127325 864.5 kb Upstream −108.48 2.86e‐05 −23.19 3.98e‐02 HSALNG0127309 653.9 kb Upstream −11.27 8.07e‐06 −9.90 8.59e‐04 HSALNG0127257 124.5 kb Upstream −6.13 4.83e‐14 3.73 1.89e‐03 HSALNG0085699 UCP2 11 472.5 kb Upstream −61.52 4.80e‐22 7.57 3.19e‐02 HSALNG0085635 578.2 kb Downstream −20.38 9.63e‐08 12.62 1.12e‐02 HSALNG0085617 732.3 kb Downstream 35.27 2.95e‐31 4.49 4.70e‐03 HSALNG0143018 0.6 kb Upstream −117.29 2.98e‐10 49.36 3.00e‐02 LINC01270 CEBPB 20 100 kb Upstream 13.17 7.25e‐47 1.34 1.86e‐12 PELATON 55 kb Upstream 2.16 5.07e‐182 0.45 5.36e‐29 HSALNG0130783 378.3 kb Upstream 19.93 1.50e‐74 1.84 3.53e‐02 LINC01271 107.2 kb Upstream 56.08 8.87e‐42 9.36 3.56e‐11 LINC01275 24.6 kb Downstream −227.59 3.34e‐07 55.80 4.96e‐02 HSALNG0115230 MAP2K3 17 284.3 kb Downstream 554.73 3.05e‐25 159.25 6.56e‐04 HSALNG0114480 VAMP2 17 804 kb Upstream 53.16 3.85e‐20 4.91 2.68e‐02 HSALNG0134398 CHST12 7 553.7 kb Downstream 6.57 7.62e‐194 1.21 3.81e‐02 HSALNG0055776 634.8 kb Upstream 19.83 2.74e‐40 2.64 7.68e‐04 LINC01128 ISG15 1 141.7 kb Downstream −0.52 1.67e‐03 −0.65 1.63e‐03 HSALNG0000121 235.2 kb Upstream 35.35 2.80e‐04 17.39 7.93e‐03 HSALNG0000065 3.1 kb Downstream 58.19 1.52e‐29 80.21 7.23e‐25 HSALNG0137873 SLC38A5 X 802.794 kb Upstream 5.99 1.30e‐33 2.62 1.05e‐02 HSALNG0151199 809.417 kb Upstream 55.70 1.85e‐26 12.15 4.06e‐03 HSALNG0060170 POLR2J2 7 230.588 kb Downstream 18.93 1.04e‐03 12.08 4.60e‐02 HSALNG0060138 840.639 kb Downstream 21.98 1.26e‐64 4.98 3.19e‐03 HSALNG0150018 510.905 kb Downstream 24.51 8.70e‐41 6.92 7.56e‐03 HSALNG0149758 703.474 kb Downstream 6.18 1.49e‐137 1.78 1.60e‐04 HSALNG0060189 Intragenic 5.44 5.05e‐40 1.44 1.01e‐03 [65]Open in a new tab The β and P value of the association between the lncRNA and the mRNA are provided for each of the 2 datasets. To be considered replicated, an association had to achieve a false discovery rate–adjusted P value of ≤0.005 in the discovery data (GENE‐FORECAST) and a nominal P value of ≤0.05 in the replication data (MH‐GRID). CEBPB, CCAAT enhancer binding protein beta; GENE‐FORECAST indicates Genomics, Environmental Factors and the Social Determinants of Cardiovascular Disease in African‐Americans Study; lncRNA, long noncoding RNA; MAP2K3, mitogen‐activated protein kinase kinase 3; MH‐GRID, Minority Health Genomics and Translational Research Bio‐repository Database; and UTR, untranslated region. Table 3. Summary of the Number of Significant and Replicated lncRNA‐mRNA Associations per Chromosome Chromosome Associations lncRNA mRNA mRNA symbol(s) 1 18 18 11 MOV10, IFI6, ISG15, NIBAN1, WASF2, TRIM58, ELAPOR1, ITLN1, RHD, S100A10, MPL 2 18 14 7 ODC1, PPP3R1, RSAD2, CMPK2, SPATS2L, RPL37A, TUBA4A 3 5 5 3 GPX1, RTP4, CISH 5 2 2 2 GCNT4, SPARC 6 16 13 6 MPIG6B, HSPA1A, VNN2, H2BC4, ETV7, GMPR 7 10 7 3 CHST12, POLR2J2, POLR2J3 8 7 7 4 LY6E, ASPH, CA2, MTSS1 9 4 4 1 TRIM14 10 7 6 4 TCF7L2, ANKRD22, IFIT3, GSTO1 11 18 18 6 LMO2, IFITM3, SERPING1, HMBS, UCP2, FTH1 12 28 25 10 MAP1LC3B2, MANSC1, OASL, YBX3, PCBP2, ATF7, MYL6, ARF3, OAS1, CHPT1 14 12 12 3 NIN, CCNK, RPS29 15 9 9 2 MCTP2, MAP1A 16 27 25 6 FUS, NPIPB5, AHSP, HBM, ARHGAP17, SULT1A3 17 34 31 11 ZACN, CTDNEP1, UBB, EIF5A, EIF1, JUP, TBX21, ALOX15, MAP2K3, LGALS9, VAMP2 18 3 3 2 DSC2, CDH2 19 22 20 6 SIGLEC14, MYADM, LENG9, ENSG00000269242, NIBAN3, CLC 20 12 12 5 CEBPB, TLDC2, MAFB, PLAGL2, SIGLEC1 21 7 6 4 KCNJ15, MX1, MX2, ENSG00000249624 22 9 9 4 OSBP2, ENSG00000284554, USP18, APOL1 X 3 3 2 SLC38A5, RPL39 [66]Open in a new tab The table also reports the number of unique lncRNAs and mRNAs involved in the associations in each chromosome along with the names of the unique mRNAs. Subsequently, we assessed the correlation between the 102 differentially expressed mRNAs associated with cis‐regulatory lncRNAs and blood pressure variables (SBP, DBP, and MAP), as well as hypertension‐related traits, including eGFR and BMI. Detailed results of this analysis are documented in Table [67]S3. Overall most of the mRNAs differentially expressed in hypertensive individuals showed significant associations with SBP, MAP, and eGFR, whereas the majority showed no significant correlations with DBP and BMI. To investigate the association between the set of 249 lncrQTLs and hypertension, we used a RF classification model in both the GENE‐FORECAST (386 hypertensive and 11 normotensive individuals) and the MH‐GRID (101 hypertensive and 78 normotensive individuals) datasets. In this RF model, the 249 lncrQTLs served as predictor variables in a forest comprising 1000 trees. The analysis underwent 10 permutations to calculate the variable importance measures for each individual lncrQTL. Notably, this comprehensive set of 249 lncRNAs demonstrated predictive capability for hypertension, yielding area under the curve values of 0.79 and 0.71 in the GENE‐FORECAST and MH‐GRID data, respectively. We conducted enrichment analysis to identify molecular pathways enriched within the set of 102 mRNAs associated with lncrQTLs. Our analysis revealed 22 pathways relevant to the vascular system and blood pressure, as detailed in Table [68]4, which were enriched in the list of 102 mRNAs associated with lncrQTLs. Table 4. Molecular Pathways Related to the Vascular System and Blood Pressure and Enriched in the List of 102 mRNAs Differentially Expressed by Hypertension Status and Associated With cis‐lncrQTLs Pathway identifer Term description P value Differentially expressed mRNAs in the pathway hsa04530 Tight junction 0.00003 ARHGAP17, YBX3, MYL6, TUBA4A hsa05412 Arrhythmogenic right ventricular cardiomyopathy 0.00004 TCF7L2, JUP, CDH2, DSC2 hsa03010 Ribosome 0.00004 RPS29, RPL37A, RPL39 hsa04668 TNF signaling pathway 0.0004 MAP2K3, CEBPB hsa04137 Mitophagy 0.001 UBB, MAP1LC3B2 hsa04130 SNARE interactions in vesicular transport 0.002 VAMP2 hsa05417 Lipid and atherosclerosis 0.003 MAP2K3, PPP3R1, HSPA1A hsa04216 Ferroptosis 0.003 ALOX15, PCBP2, FTH1, MAP1LC3B2 hsa03020 RNA polymerase 0.003 POLR2J3, POLR2J2 hsa04978 Mineral absorption 0.007 FTH1 hsa04010 MAPK signaling pathway 0.007 PPP3R1, MAP2K3, HSPA1A hsa04520 Adherens junction 0.016 WASF2, TCF7L2 hsa04924 Renin secretion 0.016 PPP3R1 hsa04622 RIG‐I–like receptor signaling pathway 0.016 ISG15 hsa04720 Long‐term potentiation 0.016 PPP3R1 hsa04962 Vasopressin‐regulated water reabsorption 0.018 VAMP2 hsa04664 Fc ε RI signaling pathway 0.028 MAP2K3 hsa04657 IL‐17 signaling pathway 0.031 CEBPB hsa05131 Shigellosis 0.032 WASF2, MAP1LC3B2, UBB hsa04666 Fc γ R‐mediated phagocytosis 0.033 WASF2 hsa04612 Antigen processing and presentation 0.034 HSPA1A hsa04721 Synaptic vesicle cycle 0.035 VAMP2 [69]Open in a new tab CEBPB, CCAAT enhancer binding protein beta; IL‐17 indicates interleukin 17; lncrQTL, long non‐coding RNA quantitative trait loci; MAPK, mitogen‐activated protein kinase; PPP3R1, protein phosphatase 2B regulatory subunit 1; TNF, tumor necrosis factor; and VAMP2, vesicle‐associated membrane protein 2. DISCUSSION In the present investigation, we explored the interconnections between lncRNAs, mRNAs, and the intricate molecular underpinnings of hypertension, in a total sample set of 676 Black individuals and uncovered several significant relationships. Specifically, our differential expression analysis revealed 129 mRNAs with significant difference in expression levels between hypertensive and normotensive individuals, consistent across both GENE‐FORECAST and MH‐GRID study cohorts. Notably, 271 distinct lncRNA‐mRNA associations were identified, involving 249 unique lncRNAs (referred to as lncrQTL) and 102 unique mRNAs. Most of the associations featured lncRNAs upstream or downstream of the respective mRNA. The differentially expressed mRNAs exhibited significant correlations with SBP, MAP, and eGFR, whereas no significant associations were observed with DBP and BMI. Moreover, a random forests classification model using the 249 lncrQTLs predicted hypertension with area under the curve values of 0.79 and 0.71 in the GENE‐FORECAST and MH‐GRID datasets, respectively. Enrichment analysis revealed 22 molecular pathways relevant to the vascular system and blood pressure enriched in the list of 102 mRNAs associated with lncrQTLs. Prior investigations have underscored the pivotal role of lncRNAs in cardiovascular pathophysiology.[70] ^23 For example, Jeong et al elucidated the influence of lncRNAs on vascular calcification,[71] ^6 whereas other studies have probed into how lncRNAs modulate endothelial functions in coronary artery disease.[72] ^24 Our findings align with and extend this literature, emphasizing the multifaceted roles of lncRNAs in the interconnected molecular pathways that modulate blood pressure. Integrating Vesicle Trafficking, VAMP2 Functionality, and lncRNA in the Renin‐Angiotensin System The perturbation of vesicle trafficking and the functioning of VAMP2 (vesicle‐associated membrane protein 2) can exert a profound influence on the release of elements within the renin‐angiotensin system, thereby disrupting vascular tone[73] ^25 and potentially playing a role in the genesis of hypertension.[74] ^26 Within the context of blood vessels, the vesicle trafficking process assumes particular significance, as it governs the release of vasoactive substances that govern vessel constriction and dilatation.[75] ^27 VAMP2, 1 of the principal actors in vesicle trafficking, orchestrates the fusion of vesicles with the cellular membrane, liberating their contents into the extracellular milieu.[76] ^27 Within the renin‐angiotensin system, a cascade of processes modulate blood pressure and fluid equilibrium.[77] ^28 An essential event in this sequence is the secretion of renin, an enzyme synthesized by specialized renal cells, into the bloodstream.[78] ^29 Renin, in turn, catalyzes the cleavage of angiotensinogen, yielding angiotensin I, subsequently transformed into angiotensin II by angiotensin‐converting enzyme.[79] ^30 The nexus between vesicle trafficking, VAMP2, and renin‐angiotensin system components revolves around renin release.[80] ^31 Our pathway analysis unveiled the enrichment of the renin secretion pathway, notably implicating protein phosphatase 2B regulatory subunit 1, an mRNA found to be overexpressed in subjects with hypertension in the 2 independent data sets we examined. lncRNAs associated with protein phosphatase 2B regulatory subunit 1 expression, including 3 (HSALNG0015689, LINC01890, and HSALNG0015679) identified in our lncrQTL analysis (Table [81]S4), may have an influence on the functionality of the renin‐angiotensin system, a pivotal modulator in the regulation of blood pressure. Perturbations in the regulatory dynamics of this system can precipitate hypertensive conditions. It is postulated that VAMP2‐mediated vesicle trafficking may modulate the exocytosis of renin‐containing vesicles from renal renin‐secreting cells.[82] ^25 VAMP2 plays a critical role in the exocytosis process, and any dysfunction in this protein is likely to result in inadequate renin release.[83] ^25 This impairment, stemming from VAMP2 dysfunction, could disrupt the entire cascade of the renin‐angiotensin‐aldosterone system, leading to diminished production of angiotensin II, a potent vasoconstrictor crucial for maintaining blood pressure homeostasis.[84] ^32 Although this scenario initially suggests a potential decrease in blood pressure attributable to reduced vasoconstriction, the body's homeostatic mechanisms are expected to counteract this change.[85] ^33 Specifically, a decline in angiotensin II levels may trigger compensatory upregulation of the renin‐angiotensin‐aldosterone system, characterized by increased renin synthesis and heightened sensitivity of angiotensin II receptors, ultimately maintaining or even elevating blood pressure despite the initial disruption.[86] ^34 Additionally, prolonged low renin levels could prompt alterations in aldosterone secretion, resulting in sodium and water retention, elevated plasma volume, and subsequent hypertension development.[87] ^35 , [88]^36 This intricate interplay of mechanisms likely contributes to the prevalence of low‐renin hypertension observed in certain populations, such as Black individuals.[89] ^37 , [90]^38 Dysregulation of VAMP2 function could impede vesicle fusion with the cell membrane, resulting in decreased renin release into circulation.[91] ^39 This reduced renin release would subsequently diminish angiotensin I levels, thereby impacting angiotensin II production.[92] ^32 Disruption in angiotensin II synthesis could alter blood vessel tone, as angiotensin II plays a pivotal role in vasoconstriction.[93] ^40 Inadequate vessel constriction may compromise blood pressure regulation, potentially contributing to hypertension development.[94] ^41 Although a reduction in renin levels would typically result in decreased angiotensin II production and subsequent blood pressure reduction, the prevalence of low‐renin hypertension in Black individuals presents a paradoxical scenario.[95] ^37 , [96]^42 Despite lower angiotensin II levels, individuals of African descent often exhibit sustained or even elevated blood pressure.[97] ^43 In this context, the body's compensatory mechanisms may include the upregulation of angiotensin II type 1 receptors, enhancing sensitivity to circulating angiotensin II.[98] ^44 Moreover, alternative vasoconstrictor pathways and alterations in aldosterone signaling leading to volume expansion and sodium retention may contribute to blood pressure maintenance or elevation in this population.[99] ^45 Our investigation reveals a notable relationship between the lncRNA HSALNG0114480 and VAMP2, a pivotal participant in vesicular transport.[100] ^46 The consistent manifestation of this association in both the GENE‐FORECAST and MH‐GRID data implies a potential regulatory impact of HSALNG0114480 on VAMP2 expression. HSALNG0114480 is located 804 kilobases upstream of VAMP2, suggesting the likelihood of long‐range chromatin interactions that could modulate VAMP2 transcription. This regulatory mechanism holds the potential for profound ramifications on the functional aspects of VAMP2 in critical cellular processes. The dysregulation of VAMP2, potentially attributed to aberrant modulation by regulatory elements, such as HSALNG0114480, may disrupt the exocytosis of renin‐containing vesicles.[101] ^25 Such impairment in renin release has the potential to initiate a cascade of effects within the renin‐angiotensin system, consequently influencing blood pressure regulation and contributing to the pathophysiology of hypertension.[102] ^32 The intricate interaction among these mechanisms may contribute to the disproportionately high rates of hypertension observed in Black individuals, alongside other genetic and environmental factors that likely play a role.[103] ^47 , [104]^48 The Intersection of Inflammation, Atherosclerosis, and Vascular Health Through lncRNA‐mRNA Relationships Three pathways enriched within our list of mRNAs associated with lncRNA (namely, the mitogen‐activated protein kinase signaling pathway, the tumor necrosis factor signaling pathway, and the interleukin 17 signaling pathway) are intricately entwined with inflammatory responses. Notably, mitogen‐activated protein kinase kinase 3 (MAP2K3), a pivotal constituent of the mitogen‐activated protein kinase pathway,[105] ^49 and CCAAT enhancer binding protein beta (CEBPB), implicated in tumor necrosis factor and interleukin 17 signaling,[106] ^50 demonstrate significant associations with lncRNAs within their cis‐regulatory regions, in our results. We observed an association between CEBPB and several lncRNAs, including LINC01270, PELATON, HSALNG0130783, LINC01271, and LINC01275. PELATON, a monocyte‐ and macrophage‐specific lncRNA, exhibits upregulation in unstable atherosclerotic plaques.[107] ^51 Atherosclerosis is a condition characterized by the accumulation of plaques in the arterial walls.[108] ^52 These plaques are often composed of cholesterol, immune cells (such as monocytes and macrophages), and other cellular debris.[109] ^53 The upregulation of PELATON is associated with unstable atherosclerotic plaques, suggesting its potential role in the progression of atherosclerosis.[110] ^51 Two pathways significantly enriched in our list of mRNAs associated with lncRNAs (lipid and atherosclerosis and ferroptosis) are relevant to vascular health. Atherosclerosis, influenced by lipid metabolism and cellular death processes, like ferroptosis,[111] ^54 involves genes, such as MAP2K3 and MAP1LC3B2, associated with lncRNAs predicting hypertension, a main risk factor for atherosclerosis. Furthermore, the inflammation accompanying plaque instability contributes to endothelial dysfunction, which impairs blood vessels' capacity to regulate blood pressure.[112] ^40 The suppression of PELATON through knockdown adversely affects cellular functions associated with plaque progression.[113] ^55 In vitro studies have demonstrated that PELATON knockdown leads to reduced uptake of CD36, a pivotal mediator in the phagocytic oxidized low‐density lipoprotein uptake during atherosclerotic plaque formation and progression.[114] ^51 Interplay of Inflammation, MYC proto‐oncogene (MYC)‐Induced lncRNA, and LINC02904 in the Modulation of Lymphocyte Antigen 6 Complex, Locus E and Its Implications for Hypertension Pathophysiology Inflammation plays a key role in the genesis and progression of hypertension.[115] ^56 Inflammatory cytokines exert a profound influence on blood vessel functionality, instigating increased vascular resistance and elevated blood pressure.[116] ^57 Lymphocyte antigen 6 complex, locus E (LY6E) is recognized for its involvement in immune cell signaling modulation and its impact on the response to infections.[117] ^58 Our lncrQTL analysis unveiled 2 lncRNAs (LINC02904 and MYC‐induced lncRNA [MINCR]) intricately associated with LY6E. MINCR, 1 of the lncRNAs associated with LY6E, is recognized for its capacity to modulate the transcriptional activity of the MYC gene.[118] ^59 Given the well‐documented role of MYC in cancer and cell proliferation,[119] ^60 the induction of MINCR by MYC suggests a nuanced interplay wherein MYC may exert an indirect influence on LY6E expression through MINCR. A pivotal observation in our investigation is the observed negative association between MINCR and LY6E, implying that MINCR potentially governs LY6E expression, thereby implicating hypertension‐related pathways. Furthermore, the documented involvement of LY6E in cell proliferation,[120] ^58 a process intricately regulated by MYC, accentuates the conceivable significance of this interaction in the pathophysiology of hypertension.[121] ^61 Additionally, a noteworthy correlation between the lncRNA LINC02904 and LY6E was identified. The close genomic proximity of LINC02904, positioned 15.4 kilobases upstream of LY6E, lends support to the hypothesis that LINC02904 may function as a regulatory factor for LY6E. lncRNAs are known to modulate the expression of neighboring genes through diverse mechanisms, including chromatin remodeling, transcriptional interference, and serving as molecular scaffolds.[122] ^62 This inference raises the prospect that LINC02904 may play an important role in orchestrating the activity of LY6E. Strengths and Limitations Although our study offers promising insights, it bears some limitations. The study's cross‐sectional design restricts our ability to draw causal conclusions. The MDRD Study equation used to calculate eGFR tends to underestimate high values and has limited generalizability as it was primarily developed and validated in populations with chronic kidney disease. Only a limited number of demographic and traditional risk factors were available for this study. To establish the temporal relationship between lncRNA expression and hypertension development, longitudinal investigations are imperative. Experimental studies are also warranted to elucidate the mechanisms underpinning the influence of the lncRNAs on blood pressure regulation. Although this study used eGFR‐AFR, adjusting for African ancestry, race‐based eGFR calculations are discouraged, partly because they may perpetuate the misconception that race is a biological rather than a social construct. Nevertheless, the study boasts several strengths. The analysis leveraged a substantial sample size of Black individuals, a population disproportionately affected by hypertension and underrepresented in current genomic data. It harnessed high‐throughput sequencing technologies to conduct an unbiased transcriptome‐wide analysis. The validation of the findings in 2 independent datasets provides robust results that could be reliably built on to design in vivo experiments that will provide insights into the molecular mechanisms behind the observed relationships. CONCLUSIONS In summary, our research offers fresh insights into the involvement of lncRNAs in hypertension among Black. We identified several differentially expressed mRNAs and several lncRNAs that predict hypertension and unveiled a network of lncRNA‐mRNA relationships linked to hypertension, SBP, eGFR, and MAP. Our study aligns with and extends the burgeoning evidence highlighting the pivotal role of lncRNAs in the development of cardiovascular diseases, including hypertension. Understanding the role of these lncRNAs in hypertension may pave the way for novel therapeutic strategies targeting these lncRNAs, potentially serving as diagnostic markers or therapeutic targets for hypertension. Sources of Funding This research was supported by the National Human Genome Research Institute, National Institutes of Health. Disclosures None. Supporting information Tables S1–S4 [123]JAH3-13-e034417-s001.pdf^ (1.1MB, pdf) Acknowledgments