Skip to main content
Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Targeted capture sequencing identifies genetic variations of GRK4 and RDH8 in Han Chinese with essential hypertension in Xinjiang

  • Wenxi Jiang ,

    Contributed equally to this work with: Wenxi Jiang, Xizi Wang

    Roles Conceptualization, Data curation, Funding acquisition, Project administration, Resources, Supervision, Validation, Writing – original draft, Writing – review & editing

    Affiliation Department of Medicine, The Fifth Affiliated Hospital of Xinjiang Medical University, Urumqi, China

  • Xizi Wang ,

    Contributed equally to this work with: Wenxi Jiang, Xizi Wang

    Roles Formal analysis, Methodology, Software, Visualization, Writing – original draft, Writing – review & editing

    Affiliation Joint Laboratory for Translational Medicine Research, Beijing Institute of Genomics, Chinese Academy of Sciences & Liaocheng People’s Hospital, Liaocheng, China

  • Ronghui Li,

    Roles Formal analysis, Methodology

    Affiliation CAS Key Laboratory of Genome Science and Information, Beijing Institute of Genomics, Chinese Academy of Sciences (CAS), Beijing, China

  • Panpan Wang,

    Roles Data curation, Investigation, Methodology

    Affiliation CAS Key Laboratory of Genome Science and Information, Beijing Institute of Genomics, Chinese Academy of Sciences (CAS), Beijing, China

  • Guangle Shan,

    Roles Formal analysis, Methodology

    Affiliation CAS Key Laboratory of Genome Science and Information, Beijing Institute of Genomics, Chinese Academy of Sciences (CAS), Beijing, China

  • Xiaodong Jia,

    Roles Data curation, Methodology

    Affiliation CAS Key Laboratory of Genome Science and Information, Beijing Institute of Genomics, Chinese Academy of Sciences (CAS), Beijing, China

  • Mingliang Gu

    Roles Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Supervision, Validation, Writing – review & editing,

    Affiliations Joint Laboratory for Translational Medicine Research, Beijing Institute of Genomics, Chinese Academy of Sciences & Liaocheng People’s Hospital, Liaocheng, China, CAS Key Laboratory of Genome Science and Information, Beijing Institute of Genomics, Chinese Academy of Sciences (CAS), Beijing, China


Essential hypertension is a common cardiovascular disease with complex etiology, closely related to genetic and environmental factors. The pathogenesis of hypertension involves alteration in vascular resistance caused by sympathetic nervous system (SNS) and renin angiotensin system (RAS). Susceptibility factors of hypertension vary with regions and ethnicities. In this study, we conducted target capture sequencing on 54 genes related to SNS and RAS derived from a collection of Han nationality, consisting of 151 hypertension patients and 65 normal subjects in Xinjiang, China. Six non-synonymous mutations related to hypertension were identified, including GRK4 rs1644731 and RDH8 rs1801058, Mutations are predicted to affect 3D conformation, force field, transmembrane domain and RNA secondary structure of corresponding genes. Based on protein interaction network and pathway enrichment, GRK4 is predicted to participate in hypertension by acting on dopaminergic synapse, together with interacting components. RDH8 is involved in vitamin A (retinol) metabolism and consequent biological processes related to hypertension. Thus, GRK4 and RDH8 may serve as susceptibility genes for hypertension. This finding provides new genetic evidence for elucidating risk factors of hypertension in Han nationality in Xinjiang, which in turn, enriches genetic resource bank of hypertension susceptibility genes.


Essential hypertension (EH) is traditionally defined as persistently high blood pressure (HBP) with systolic blood pressure ≥ 140 mmHg and diastolic blood pressure ≥ 90 mmHg. The prevalence of hypertension is expected to increase by 60% worldwide by 2025 [1]. High blood pressure is one of the leading causes of death in China. However, awareness, treatment and control rate of hypertension in China remains at a relatively low level (46.9%, 40.7% and 15.3%, respectively) [24].

Many pathogenic elements contribute to EH, including smoking, drinking, high salt, oil intake, air pollution [5] and genetic alterations. Genetic and environmental factors can affect intermediate phenotypes [6], including SNS and RAS, sodium excretion, vascular reactivity and cardiac contraction. Generally, SNS is the first defense against a wide range of environmental pressures, and neuro-modulatory factors respond very rapidly during blood pressure regulation. Among various genetic risk factors, over-activation of SNS can cause both hypertension and target organ damage [7, 8]. The main functions of RAS are to regulate blood pressure and maintain the homeostasis of water and electrolyte, which is an important mechanism involved in the pathogenesis of hypertension. Renin hydrolyzes angiotensinogen in the plasma to produce angiotensinogen I (Ang I). Ang I exerts a weak effect on vasoconstriction, however, when entering pulmonary circulation, it is converted into Ang II catalyzed by angiotensin-converting enzyme (ACE). As a very strong vasoactive substance, Ang II increases susceptibility to hypertension [9].

The development and progression of EH can be affected by different genetic factors in different populations and in different regions. At present, very few studies have been focused on genetic factors related to hypertension in Xinjiang Han nationality. The next-generation sequencing enables to explore pathogenic variation within whole genome or exome. For hypertension, a common disease with high genetic heterogeneity, targeted sequencing may greatly improve detection efficiency by directly correlating candidates with SNS or RAS [10, 11]. In this study, we aim to identify genetic factors of hypertension in Xinjiang Han population by targeted capture sequencing on 54 genes associated with SNS or RAS.

Materials and methods

Research objects

According to diagnostic criteria (Chinese Guidelines for Prevention and Treatment of Hypertension, 2009 Basic Version), in the absence of antihypertensive drugs, hypertension was diagnosed as systolic blood pressure ≥140 mm Hg and/or diastolic blood pressure ≥90 mm Hg measured three times on the same day. The patients with a history of hypertension and currently taking antihypertensive drugs were diagnosed as hypertension even with blood pressure lower than 140/90mm Hg. From March 2015 to December 2015, 151 patients with hypertension treated at the Fifth Affiliated Hospital of Xinjiang Medical University were randomly selected as “case” group, and 65 healthy adults without hypertension, metabolic diseases or family history of hypertension served as “control” group. All subjects were Han ethnicity, born in Xinjiang. This study was approved by Ethics Committee of the Fifth Affiliated Hospital of Xinjiang Medical University. Informed consent was signed by each subject before blood sample collection.

Selection of candidate genes

Both SNS and RAS regulate blood pressure through a variety of physiological and cellular pathways. We selected 54 coding genes involved in SNS or RAS for targeted sequencing, and adopted systematic analysis to deeply mine genetic factors related to hypertension (S1 Table).

Targeted capture and next-generation sequencing

Genomic DNA was extracted from peripheral blood samples using QIAamp DNA blood maxi kit. Then DNA was interrupted into ~200 bp fragments with Covaris S220. Breaking parameters were setup as follows: Duty factor 10%; Peak Incident Power 175; Cycles per Burst 200; Treatment time 360s; and Bath Temperature 4°C -8°C. Agilent 2100 quality control was performed on fragmented DNA. Agilent Sureselect DNA targeting sequence capture kit was used for library preparation (S2 Table). Firstly, fragmented DNA was end-repaired and purified using AMPure XP beads. The purified DNA was added with A at the 3 ’end and purified with AMPure XP beads. Subsequently, DNA was connected with an adaptor and purified with AMPure XP beads. Secondly, polymerase chain reaction (PCR) was performed to expand the linker. Library preparation was programed as follows: pre-denaturation at 94°C for 2 min; denaturation at 94°C for 30 s, annealing at 65°C for 40 s, extension at 72°C for 2 min, a total of 30 cycles of the above steps; extension at 72°C for 10 min; with a total amplification volume of 50 μl. Agilent 2100 was used for quality control. Finally, Illumina Nextseq500CN was applied for sequencing analysis. Using Trimmatomatic [12] to remove the original sequencing connector and low-quality sequence. Filtered sequence was aligned to Thousands Genome Reference Sequence (GRCh37) using BWA [13]. The output data were converted to BAM file and sorted with Samtools [14]. Sequences were deduplicated with Picard. Using GATK software [15], single nucleotide variation (SNV) and indel mutation (In/Del) were analyzed and filtered (S1 File). The called variants were mapped to dbSNP database and annotated with ANNOVAR [16].

Quality control (QC) and Allelic association

To obtain high-quality data for association analysis, SNPs were trimmed using the following criteria: (1) call rate of sample or SNP > 95%; (2) a threshold of 0.0001 for Hardy-Weinberg equilibrium (HWE). Generally, SNPs with minor allele frequencies (MAFs) > 0.01 were included in association analysis. Sequencing depth was counted by R program. Samples with sequencing depth less than 30 were removed. Using QC-passed SNPs to calculate allele frequency difference and allele superiority ratio between case and control, SNPs related to EH were included (P<0.05). QC and Allelic association were completed through Plink [17].

Protein 3D structure

Swiss-model was used to construct 3D configuration of GRK4 and RDH8 encoded proteins based on amino acid sequences. Then, SwissPDB viewer was used to analyze structural changes before and after mutation was introduced and to estimate alterations in force fields.

RNA secondary structure

RNA-fold sever and RNAstructure were used to predict RNA secondary structure of GRK4 and RDH8. The minimum free energy prediction model was applied. Dynamic changes were observed before and after the minimum free energy mutation was introduced.

PPI network construction and function enrichment analysis

STRING [18], IntAct [19], MINT [20], BioGRID [21], HTRIdb [22], InWeb_IM [23] and HPRD [24] were employed to query interaction proteins of GRK4 and RDH8. Protein-protein interaction (PPI) network was constructed by using Cytoscape. Functional enrichment was established with David (


SNP distribution

A total of 17,581 SNPs was obtained after comparing sequencing results with dbSNP database. These SNPs were annotated using ANNOVAR. Most SNPs were located in intergenic regions and introns. 213 SNPs were localized in exons, including 137 non-synonymous mutations within 32 targeted sequencing genes (Fig 1). Because of the presence of 2 kb extension flanking each sequence, some SNPs were located beyond targeted capture sequencing genes.

Fig 1. Distribution of non-synonymous mutations.

The horizontal, coordinate and vertical axis represents genes, chromosomes, and SNP number, respectively.

Through quality control parameter screening, 148 samples met enrollment criteria, including 109 cases and 39 controls carrying 8,761 eligible SNPs. Based on plink, 451 SNPs were related to hypertension (p <0.05). Through annotation, 6 SNPs (non-synonymous mutations) were located in exons (Table 1). The protein encoded by HPS3 may participate in organelle biogenesis associated with melanosomes, platelet dense granules and lysosomes. Related diseases include Hermansky Pudlak syndrome 3 and Hermansky Pudlak syndrome without pulmonary fibrosis. A family member of Fas activated serine/threonine kinase domain (FASTKD) encoded by FASTKD3 may be involved in apoptosis. The protein encoded by FASTKD3 interacts with certain components of mitochondrial respiration and translation. FASTKD3 is associated with Sengers syndrome. FIGNL1 encodes a protein belonging to AAA ATPase family. FIGNL1 is absorbed into DNA damage site and participates in DNA double strand break repair through homologous recombination. In addition, FIGNL1 can be located in centrosome, inhibit cilia formation, and regulate the proliferation and differentiation of osteoblasts. HPS3, FASTKD3 and FIGNL1 1 are not directly or indirectly associated with hypertension. The mutation frequencies of these three genes as well as GRK4 rs77833898 were low, with OR value of less than 1. Therefore, in this study, we focused on how RDH8 rs1644731 and GRK4 rs1801058 correlate with hypertension in Han people in Xinjiang. The functions of HPS3, FASTKD3 and FIGNL1 have no direct or indirect association with the pathogenesis of hypertension, and the existing studies have not proved their association with hypertension. The frequency of mutations in these three loci and GRK4 rs77833898 was low in the samples, and the OR values were all less than 1. Therefore, this study mainly studied how the two genes, RDH8 rs1644731 and GRK4 rs1801058, and their mutation sites play a role in the pathogenesis of hypertension in Han people in Xinjiang.

Table 1. SNPs significantly associated with hypertension.

G protein-coupled receptor kinase 4 (GRK4), located on chromosome 4, has 23 exons. The rs1801058 is located in exon 20. GRK4 regulates blood pressure by modifying dopamine signal in kidney [25]. A high level of phosphorylated dopamine D1 receptor (D1R) contributes to HBP. Notably, activities of dopamine receptor and AT1R are regulated by GRK4 and protein phosphatase mediated phosphorylation/dephosphorylation [9, 26]. All-trans retinol dehydrogenase coding gene (RDH8), also known as photoreceptor RDH, located on chromosome 19, has 6 exons. The rs1644731 is located in exon 5. Alcohol reduction is the first step of rhodopsin regeneration pathway [27, 28].

Protein 3D structure analysis

Mutations in coding regions may alter protein structure, and then affect protein function. Mutations in rs1644731 and rs1801058 caused amino acid 486 in GRK4 to change from valine (V) to alanine (A), whereas amino acid 222 in RDH8 from methionine (M) to serine (T), respectively. In order to decipher changes in configuration, tertiary structure of GRK4 or RDH8 was constructed by Swiss-model, whereas 3D structure by SwissPDB-viewer (Fig 2).

Fig 2. Schematic diagram of 3D protein structure and force field curves.

A, B light blue represents amino acid 486 Val / Ala before and after mutation in GRK4; C, D light blue represents amino acid 222 Met / Thr before and after mutation in RDH8. The green dotted line represents hydrogen bonding. E. The blue line represents RDH8 p. 222M. The red line represents RDH8 p. 222T; F. The blue line represents GRK4 p. 486V. The red line represents GRK4 p. 486A.

Potential energy changes in GRK4 (486 V →A) and RDH8 (222 M→T) were estimated by GROMOS96 field using SwissPDB-viewer (Fig 2). Mutations are predicted to cause instability of protein structure.

Analysis of transmembrane structure domain

TMpred was used to predict transmembrane domains of protein sequences upon GRK4 p.V486A and RDH8 p.M222T. One transmembrane domain model was predicted in GRK4, including one transmembrane helix. No significant change in transmembrane domain was predicted in response to individual mutation. Two models of transmembrane domains were predicted in RDH8, including three and two transmembrane helices, respectively. The amino acid as start- or end-position of transmembrane helix changed when sequence mutation was introduced (Table 2). Based on UniProt annotation, RDH8 is a multi-channel transmembrane protein. Amino acid changes may affect RDH8 transmembrane transport.

Table 2. Distribution of transmembrane domains before and after GRK4 and RKH8 site mutations.

RNA secondary structure prediction analysis

Using RNAstructure, secondary structure of GRK4 or RDH8 mRNA was constructed (S1 Fig). The minimum free energy of GRK4 m.1457T or m.1457C was -643.6 Kcal/mol or -645.1 Kcal/mol, respectively. The minimum free energy of RDH8 m.665T or m.665C was -779.0 Kcal/mol or -779.7 Kcal/mol, respectively. Mutations in GRK4 m.1457 and RDH8 m.665 did not change free energy of RNA secondary structure, which might have little effect on mRNA stability.

PPI network analysis for GRK4 and RDH8

In order to explore potential effects of GRK4 and RDH8 on hypertension pathogenesis, a PPI network was constructed. Firstly, proteins potentially interacting with GRK4 or RDH8 were mined through 7 protein-interaction databases (Fig 3). Totally, 24 candidates might interact with GRK4; whereas 11 candidates with RDH8. Among them, 5 candidates interacting with GRK4 were associated with hypertension. Secondly, pathways were explored to predict how interaction proteins participate in signal transduction. Interestingly, GRK4 interacting proteins yielded 7 pathways (p <0.01), of which 5 pathways related to Organismal Systems and 2 pathways related to Human Diseases. RDH8 interacting proteins yielded one pathway related to Metabolism (Fig 4). GRK4 and RDH8 may interact with these proteins involved in hypertension. For example, based on enrichment analysis of RDH8 interaction genes, LRAT, AOX1, CYP26A1, BCO1, PNPLA4, and RETSAT were involved in retinol metabolism. Downregulation of LRAT and a low level of serum retinol (VA) were independent predictors of EH. LRAT may affect blood pressure by down-regulating Ang II [29].

Fig 3. Protein-protein interaction network.

Red: genes have been reported to be related to hypertension. No evidence has been provided to correlate green genes with hypertension. The larger the circle, the more databases query this interaction relationship (CALM1 was queried in five databases; FSHR in four; RHO and IKBKG in two; whereas others in one database).

Fig 4. Enriched KEGG pathways.

A. Signaling pathways involved in GRK4 and its interacting proteins. Blue represents Organismal Systems, and green represents Human Diseases; B. Signal pathways involved in RDH8 and its interacting proteins. Red represents Metabolism.


Hypertension is attributed to multiple environmental (such as obesity and excessive drinking) and genetic factors [30, 31]. Accumulating evidence indicates that genetic risk factors for hypertension vary with ethnicity and regions. For example, CAMK4 rs10491334 is associated with hypertension in Uyghur population; whereas GRK4 rs180105 in East Asians and Europeans; and GRK4 rs2960306 in Europeans [32]. Abnormalities in SNS and RAS functions contribute to pathogenesis of hypertension. Therefore, among Han population in Xinjiang, SNS, RAS and candidate genes are hypothesized to play roles in etiology of hypertension.

In order to verify this conjecture, we have collected 151 hypertension cases and 65 controls, selected 54 genes related to SNS and RAS for targeted sequencing, and identified SNPs of genes related to hypertension. Notably, GRK4 rs1801058 (p.V486A) and RDH8 rs1644731 (p.M222T) were related to hypertension. Upon mutation was introduced, 3D protein structure, potential energy, RNA secondary structure, respectively, was evaluated. In order to explore functions of GRK4 and RDH8, seven databases, such as String and HPRD, were used to mine proteins interacting with GRK4 and RDH8, and to construct PPI network. Five proteins might interact with GRK4, such as A2ML1. Based on meta-analyses, A2ML1 was related to hypertension [33]; ADRB2 and NOS3 jointly acted on sympathetic nervous system [34]; while ADRB2 was related to hypertension in American Indians [35] and involved in pathophysiology of hypertension in Yi population [36]. In addition, transmembrane helix in RDH8 might change to a certain extent due to RDH8 p.M222A mutation, which might affect protein transmembrane transport.

GRK4 and ADRB2 may play synergistic roles in hypertension through dysregulating SNS. Then, functional enrichment was analyzed by DAVID, and proteins interacting with GRK4 were annotated to 7 KEGG pathways. Morphine Addiction process indirectly enhances the release of dopamine from dopaminergic synaptic terminals by relaxing inhibitory dopamine cells to discharge, while interacting proteins may affect the release of presynaptic dopamine terminals and enter postsynaptic cells in Dopaminergic Synapse. Dopamine receptor D1 is closely related to hypertension [37]. GRK4- interacting proteins may affect susceptibility to hypertension through morphine addiction and dopaminergic synaptic pathway.

RDH8 is involved in several GO terms such as retinol metabolism and NADP-retinol dehydrogenase catalysis. Retinol metabolism includes retinol chemical reactions and pathways. Retinol is required to produce vitamin A. NADP-retinol dehydrogenase catalyzes vitamin A binding to NADP+, and vitamin A is dehydrogenated into retinoids. In a recent GWAS study, pathways and biological processes related to hypertension were discovered, including retinoids metabolism [38], related to RDH8 and NADP-retinol dehydrogenase and thus vitamin A metabolism. NADP retinol dehydrogenase is a part of retinol metabolism, while retinol metabolism is a process of vitamin A metabolism. Low expression of LRAT, which may interact with RDH8, is an independent predictor of essential hypertension. RDH8 may participate in two biological processes, vitamin A metabolism and hypertension. Another interaction protein of RDH8, BCO1 (Beta-carotene 15,15’-dioxygenase), was related to lycopene [39], and thus hypertension [40].

In summary, GRK4 and RDH8 are genetic risk factors for essential hypertension in Han Chinese in Xinjiang. This finding provides new evidence for a better understanding of genetic factors susceptible for essential hypertension pathogenesis in Han Chinese in Xinjiang.

Supporting information

S1 Fig. Secondary structure of gene.

A. secondary structure of GRK4 m.1457T; B. secondary structure of GRK4 m.1457C; C. secondary structure of RDH8 m.665T; D. secondary structure of GRK4 m.665C.


S2 Table. Agilent company targeted sequence capture kit information.


S1 File. SNV and Indel data of 151 patients and 65 healthy adults targeted capture sequencing.



The authors would like to thank all patients who participated in this study.


  1. 1. Elliott W. J. Systemic hypertension. Current problems in cardiology 32, 201–259, (2007). pmid:17398315
  2. 2. Mortality G. B. D. & Causes of Death, C. Global, regional, and national age-sex specific all-cause and cause-specific mortality for 240 causes of death, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet 385, 117–171, (2015). pmid:25530442
  3. 3. Collaborators, G. B. D. C. o. D. Global, regional, and national age-sex specific mortality for 264 causes of death, 1980–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet 390, 1151–1210, (2017). pmid:28919116
  4. 4. Wang Z. et al. Status of Hypertension in China: Results From the China Hypertension Survey, 2012–2015. Circulation 137, 2344–2356, (2018). pmid:29449338
  5. 5. Munzel T. et al. Environmental stressors and cardio-metabolic disease: part I-epidemiologic evidence supporting a role for noise and air pollution and effects of mitigation strategies. European heart journal 38, 550–556, (2017). pmid:27460892
  6. 6. Carretero O. A. & Oparil S. Essential hypertension. Part I: definition and etiology. Circulation 101, 329–335, (2000). pmid:10645931
  7. 7. Beetz N. et al. Phosducin influences sympathetic activity and prevents stress-induced hypertension in humans and mice. The Journal of clinical investigation 119, 3597–3612, (2009). pmid:19959875
  8. 8. Light K. C. Hypertension and the reactivity hypothesis: the next generation. Psychosomatic medicine 63, 744–746, (2001). pmid:11573022
  9. 9. Yang J., Villar V. A., Jones J. E., Jose P. A. & Zeng C. G protein-coupled receptor kinase 4: role in hypertension. Hypertension 65, 1148–1155, (2015). pmid:25870190
  10. 10. Teekakirikul P., Cox S., Funke B. & Rehm H. L. Targeted sequencing using Affymetrix CustomSeq Arrays. Current protocols in human genetics Chapter 7, Unit7 18, (2011). pmid:21480183
  11. 11. Rehm H. L. Disease-targeted sequencing: a cornerstone in the clinic. Nature reviews. Genetics 14, 295–300, (2013). pmid:23478348
  12. 12. Bolger A. M., Lohse M. & Usadel B. Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30, 2114–2120, (2014). pmid:24695404
  13. 13. Li H. & Durbin R. Fast and accurate long-read alignment with Burrows-Wheeler transform. Bioinformatics 26, 589–595, (2010). pmid:20080505
  14. 14. Li H. et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics 25, 2078–2079, (2009). pmid:19505943
  15. 15. DePristo M. A. et al. A framework for variation discovery and genotyping using next-generation DNA sequencing data. Nature genetics 43, 491–498, (2011). pmid:21478889
  16. 16. Wang K., Li M. & Hakonarson H. ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. Nucleic acids research 38, e164, (2010). pmid:20601685
  17. 17. Purcell S. et al. PLINK: a tool set for whole-genome association and population-based linkage analyses. American journal of human genetics 81, 559–575, (2007). pmid:17701901
  18. 18. Szklarczyk D. et al. STRING v11: protein-protein association networks with increased coverage, supporting functional discovery in genome-wide experimental datasets. Nucleic acids research 47, D607–D613, (2019). pmid:30476243
  19. 19. Orchard S. et al. The MIntAct project--IntAct as a common curation platform for 11 molecular interaction databases. Nucleic acids research 42, D358–363, (2014). pmid:24234451
  20. 20. Zanzoni A. et al. MINT: a Molecular INTeraction database. FEBS letters 513, 135–140, (2002). pmid:11911893
  21. 21. Oughtred R. et al. The BioGRID interaction database: 2019 update. Nucleic acids research 47, D529–D541, (2019). pmid:30476227
  22. 22. Bovolenta L. A., Acencio M. L. & Lemke N. HTRIdb: an open-access database for experimentally verified human transcriptional regulation interactions. BMC genomics 13, 405, (2012). pmid:22900683
  23. 23. Li T. et al. A scored human protein-protein interaction network to catalyze genomic interpretation. Nature methods 14, 61–64, (2017). pmid:27892958
  24. 24. Keshava Prasad T. S. et al. Human Protein Reference Database--2009 update. Nucleic acids research 37, D767–772, (2009). pmid:18988627
  25. 25. Felder R. A. et al. G protein-coupled receptor kinase 4 gene variants in human essential hypertension. Proceedings of the National Academy of Sciences of the United States of America 99, 3872–3877, (2002). pmid:11904438
  26. 26. Villar V. A. et al. G protein-coupled receptor kinase 4 (GRK4) regulates the phosphorylation and function of the dopamine D3 receptor. The Journal of biological chemistry 284, 21425–21434, (2009). pmid:19520868
  27. 27. Saari J. C., Garwin G. G., Van Hooser J. P. & Palczewski K. Reduction of all-trans-retinal limits regeneration of visual pigment in mice. Vision research 38, 1325–1333, (1998). pmid:9667000
  28. 28. Rattner A., Smallwood P. M. & Nathans J. Identification and characterization of all-trans-retinol dehydrogenase from photoreceptor outer segments, the visual cycle enzyme that reduces all-trans-retinal to all-trans-retinol. The Journal of biological chemistry 275, 11034–11043, (2000). pmid:10753906
  29. 29. Liang X., Chen M., Wang D., Wen J. & Chen J. Vitamin A deficiency indicating as low expression of LRAT may be a novel biomarker of primary hypertension. Clinical and experimental hypertension 43, 151–163, (2021). pmid:33052059
  30. 30. Whelton P. K. et al. Primary prevention of hypertension: clinical and public health advisory from The National High Blood Pressure Education Program. Jama 288, 1882–1888, (2002). pmid:12377087
  31. 31. Caligiuri S. P. B. & Pierce G. N. A review of the relative efficacy of dietary, nutritional supplements, lifestyle, and drug therapies in the management of hypertension. Critical reviews in food science and nutrition 57, 3508–3527, (2017). pmid:27494115
  32. 32. Liu C. & Xi B. Pooled analyses of the associations of polymorphisms in the GRK4 and EMILIN1 genes with hypertension risk. International journal of medical sciences 9, 274–279, (2012). pmid:22639547
  33. 33. Surendran P. et al. Trans-ancestry meta-analyses identify rare and common variants associated with blood pressure and hypertension. Nature genetics 48, 1151–1161, (2016). pmid:27618447
  34. 34. Kumar R. et al. Interactions between the genes of vasodilatation pathways influence blood pressure and nitric oxide level in hypertension. American journal of hypertension 28, 239–247, (2015). pmid:25159081
  35. 35. Soudani N. Y., Fakhoury R. M., Kaissi S. S. & Zgheib N. K. The role of genetic polymorphisms in endothelial nitric oxide synthase and beta2-adrenergic receptors with risk of hypertension in a sample of Lebanese people. Saudi medical journal 35, 255–260 (2014). pmid:24623205
  36. 36. Valencia D. M. et al. [Association and interaction of AGT, AGTR1, ACE, ADRB2, DRD1, ADD1, ADD2, ATP2B1, TBXA2R and PTGS2 genes on the risk of hypertension in Antioquian population]. Biomedica: revista del Instituto Nacional de Salud 33, 598–614, (2013). pmid:24652215
  37. 37. Harris R. C. & Zhang M. Z. Dopamine, the kidney, and hypertension. Current hypertension reports 14, 138–143, (2012). pmid:22407378
  38. 38. Ji L. D. et al. Shared genetic etiology of hypertension and stroke: evidence from bioinformatics analysis of genome-wide association studies. Journal of human hypertension 32, 34–39, (2017). pmid:29176593
  39. 39. Crowe-White K. M. et al. Variation of Serum Lycopene in Response to 100% Watermelon Juice: An Exploratory Analysis of Genetic Variants in a Randomized Controlled Crossover Study. Current developments in nutrition 4, nzaa102, (2020). pmid:32695957
  40. 40. Li Z., Chen J. & Zhang D. Association between dietary carotenoid intakes and hypertension in adults: National Health and Nutrition Examination Survey 2007–2014. Journal of hypertension 37, 2371–2379, (2019). pmid:31356404