Mutation of Plekha7 Attenuates Salt-Sensitive Hypertension in the Rat

Mutation of Plekha7 attenuates salt-sensitive hypertension in the rat

Bradley T. Endresa,b, Jessica R. C. Priestleya, Oleg Palygina, Michael J. Flistera,b, Matthew J. Hoffmana, Brian D. Weinberga, Michael Grzybowskia,b, Julian H. Lombarda, Alexander Staruschenkoa, Carol Morenoa,b, Howard J. Jacoba,b,c, and Aron M. Geurtsa,b,d,1

Departments of aPhysiology and cPediatrics, bHuman and Molecular Genetics Center, and dCardiovascular Center, Medical College of Wisconsin, Milwaukee, WI 53226

Edited by Richard P. Lifton, Yale University School of Medicine, New Haven, CT, and approved July 23, 2014 (received for review June 24, 2014)

PLEKHA7 (pleckstrin homology domain containing family A mem- hypertension and renal injury by selective ablation of adaptive im- em1Mcwi em1Mcwi ber 7) has been found in multiple studies as a candidate gene for mune cells in the SS-Rag1 and SS-Cd247 knockout human hypertension, yet functional data supporting this associa- rats (16, 17) and reduced hypertension and renal injury in the em1Mcwi tion are lacking. We investigated the contribution of this gene SS-Ncf2 (p67phox) null model exhibiting reduced medul- to the pathogenesis of salt-sensitive hypertension by mutating lary oxidative stress (18). Additionally, we have recently demon- Plekha7 in the Dahl salt-sensitive (SS/JrHsdMcwi) rat using zinc- strated multiple genes at a single hypertension GWAS-nominated finger nuclease technology. After four weeks on an 8% NaCl diet, locus (Agtrap-Plod1 locus) can have additive or subtractive effects homozygous mutant rats had lower mean arterial (149 ± 9 mmHg on blood pressure and renal function when mutated in the SS rat vs. 178 ± 7 mmHg; P < 0.05) and systolic (180 ± 7 mmHg vs. 213 ± (19). These previous studies highlight the utility of this model 8 mmHg; P < 0.05) blood pressure compared with WT littermates. system for testing the roles of GWAS candidate human disease Albumin and protein excretion rates were also significantly lower genes by disrupting their specific rat orthologs using ZFN technology (20). in mutant rats, demonstrating a renoprotective effect of the mu- A single-nucleotide polymorphism (SNP) (rs381815, minor tation. Total peripheral resistance and perivascular fibrosis in the

Plekha7 allele frequency 0.26) in intron 1 of the pleckstrin homology GENETICS heart and kidney were significantly reduced in mutant domain containing family A member 7 (PLEKHA7) gene, was animals, suggesting a potential role of the vasculature in the at- identified by five independent GWAS to be associated with el- tenuation of hypertension. Indeed, both flow-mediated dilation evated systolic blood pressure and hypertension in multiple and endothelium-dependent vasodilation in response to acetyl- populations (5, 6, 8, 21, 22). The associated locus contains only cholinewereimprovedinisolatedmesenteric resistance arteries the PLEKHA7 gene (5); however, the genetic mechanism(s) of Plekha7 mutant rats compared with WT. These vascular underlying this locus have not yet been functionally character- improvements were correlated with changes in intracellular cal- ized. PLEKHA7 is highly expressed in the kidney and heart, cium handling, resulting in increased nitric oxide bioavailability where it may be involved in formation and maintenance of the in mutant vessels. Collectively, these data provide the first func- apical junction complex of epithelial cells (23). However, limited tional evidence that Plekha7 may contribute to blood pressure data on PLEKHA7 function are available to extrapolate its regulation and cardiovascular function through its effects on the potential role(s) in the pathogenesis of hypertension. Here we vasculature. used ZFN mutagenesis to obtain the first evidence to our knowledge in any model system that Plekha7 has a functional role in genetic | GWAS | adherens junction | physiology | endothelial cells several hypertension-associated phenotypes in the rat. We found that mutation of Plekha7 in the SS rat attenuated salt-induced ypertension is a complex disease that is characterized by Hincreased blood pressure, renal damage, and vascular dys- Significance function which collectively increase risk of atherosclerosis, stroke, heart disease, and renal failure in one-quarter of all Zinc-finger nuclease (ZFN)-mediated mutagenesis has now en- adults worldwide (1–3). Because there is strong evidence of abled researchers to manipulate specific genes to test their heritability in hypertension (2, 4, 5), considerable effort has been function in animal models other than mice. Applying ZFNs to put toward identifying novel candidate genes and their molecular rats, we can now test the role of specific human genome-wide mechanisms. Genome-wide association studies (GWAS) have association studies (GWAS)-nominated genes for hypertension identified many potential hypertension loci, which shed light on in a well-characterized hypertensive rat model, the Dahl salt- the genetic complexity of this disease (5–8) but have provided sensitive rat. This study provides the first functional evidence little mechanistic insight. As such, validation and elucidation of that the GWAS-nominated gene Plekha7 plays an essential role the functional roles and disease mechanisms for these gene can- in blood pressure regulation and cardiovascular function by didates are the next important challenges (4). modulating vascular function. Our results indicate that Plekha7 Because hypertension is a complex disease (i.e., multiple plays a role in the regulation of intracellular calcium, nitric variants of small effect sizes contributing to disease risk), we oxide bioavailability, and the response of the vasculature to hypothesized candidate gene targeting on a genetically sensitized increased flow. background would reveal functional role(s) of genetic disease modifiers. The Dahl salt-sensitive (SS) rat is an inbred genetic Author contributions: B.T.E., J.R.C.P., O.P., M.J.F., M.J.H., B.D.W., M.G., J.H.L., A.S., C.M., H.J.J., model of salt-sensitive hypertension that displays hypertension- and A.M.G. designed research; B.T.E., J.R.C.P., O.P., M.J.F., M.J.H., B.D.W., M.G., and C.M. induced renal damage, cardiac hypertrophy and vascular dys- performed research; B.T.E., J.R.C.P., O.P., M.J.F., B.D.W., J.H.L., A.S., C.M., H.J.J., and A.M.G. function (9–11). These phenotypes are induced by exposing SS contributed new reagents/analytic tools; B.T.E., J.R.C.P., O.P., M.J.F., J.H.L., C.M., and A.M.G. rats to a high-salt diet, which results in rapid induction of hy- analyzed data; and B.T.E., J.R.C.P., O.P., M.J.F., J.H.L., A.S., and A.M.G. wrote the paper. pertensive phenotypes that closely resemble salt-induced hyper- The authors declare no conflict of interest. tension seen in humans (12–15). Knockout of specific genes in This article is a PNAS Direct Submission. this disease model using zinc-finger nuclease (ZFN) technology 1To whom correspondence should be addressed. Email: [email protected]. have revealed the importance of key mechanisms contributing This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. to hypertension risk, such as the protection from salt-induced 1073/pnas.1410745111/-/DCSupplemental.

www.pnas.org/cgi/doi/10.1073/pnas.1410745111 PNAS | September 2, 2014 | vol. 111 | no. 35 | 12817–12822 Downloaded by guest on September 27, 2021 hypertension, reduced renal damage, and improved cardiac SS-Plekha7 mutant rats on a 0.4% NaCl diet but was significantly function. We also show that Plekha7 modulates calcium handling lower in the mutant rats on 8% NaCl diet compared with WT and nitric oxide (NO) bioavailability, both of which are required (Fig. S2A). We found no differences in BP variability between for normal vascular health. Collectively, these studies provide WT and mutant rats. Additionally, heart rate (HR) and HR significant mechanistic insight to the role of Plekha7 in salt- variability were not different between WT and mutant rats fed sensitive hypertension. a 0.4% NaCl diet or an 8% NaCl diet (Fig. S2B). Cumulatively, these data suggest that mutation of Plekha7 attenuates the de- Results velopment of salt-sensitive hypertension in the SS rat model. Targeting Plekha7 for Mutation Using ZFNs. To investigate the role of Plekha7 in hypertension, we used ZFNs to target genomic SS-Plekha7 Mutant Rats Have Reduced Renal Damage Compared with sequences encoding the pleckstrin homology (PH) functional WT Littermates. To examine whether renal damage under path- domain of Plekha7 in the SS rat (Fig. S1A). Injection of ZFNs ogenic conditions was affected by Plekha7 mutation, urine was targeting exon 8 of the Plekha7 gene into SS embryos resulted in collected and analyzed in male 9-wk-old WT and mutant rats a 19-bp deletion mutation in the coding sequence. This frame-shift (n = 10 per group) that were initially fed a 0.4% NaCl diet and mutation is predicted to cause the translation of 18 nonsense then switched to an 8% NaCl diet for a total of 28 d. Urinary amino acids and introduce a premature stop codon, yielding a albumin and protein excretion did not differ on the 0.4% NaCl truncated, nonfunctional protein consisting of 231 amino acids that diet (Fig. 2). After 28 d of the 8% NaCl diet, mutant rats had lacks the full PH domain (Fig. S1D). Verification of the mutation significantly lower urinary excretion of albumin (110 ± 10 mg/d; in homozygous SS-Plekha7 mutant animals was confirmed in P < 0.05) and protein (190 ± 16 mg/d; P < 0.05) compared with cDNA by RT-PCR and sequencing (Fig. S1 B and C). Quantitative WT levels of albuminuria (211 ± 22 mg/d) and proteinuria PCR revealed an approximated 50% reduction in Plekha7 ex- (381 ± 31 mg/d) (Fig. 2). In contrast, no differences in urine flow pression within the mutant animals compared with WT, indicating rate, urinary sodium, or urinary potassium concentrations be- nonsense-mediated decay of the transcript in the mutant rats (Fig. tween the mutant and WT rats were detected at the measured B S1 ). Homozygous mutant and WT control experimental rats were time points (Table S1). Body weight of WT (371 ± 5 g) and SS- generated by heterozygous brother–sister mating. There were no Plekha7 mutant rats (379 ± 4g) on the 8% NaCl diet for 4 wk was clear developmental abnormalities in the mutant animals com- not statistically different, suggesting that food and salt intake were pared with WT littermates, and mutant rats were similar in weight to WT at 9 wk of age (264 ± 9vs.250± 8g,respectively). similar between WT and mutant rats. Collectively, these data indicate that the reduction in albuminuria and proteinuria in the Protection Against Salt-Sensitive Hypertension in Plekha7 Mutant mutant rat are not due to differences in salt consumption or Rats. To test whether mutation of Plekha7 affected the de- natriuresis. velopment of hypertension in the SS rat, blood pressure (BP) was The extent of renal damage was also assessed histologically in measured by telemetry in conscious male WT and SS-Plekha7 the kidneys of WT and mutant rats after 28 d of the 8% NaCl = diet. There was no change in kidney weight/body weight ratio mutant littermates on a 0.4% NaCl diet for 31 d (n 4 per ± ± ± ± group) or on a 0.4% NaCl diet for 3 d followed by an 8% NaCl (4.31 0.10 vs. 4.56 0.12), cortical volume (450 33 vs. 497 14 3 ± ± 3 diet for 28 d (n = 8–10 per group). On the 0.4% NaCl diet, no mm ), medullary volume (386 51 vs. 406 85 mm ), or nephron × 3 ± × 3 × 3 ± × 3 significant differences in BP were detected between WT and number (34.33 10 0.48 10 vs. 33.45 10 2.08 10 mutant rats at any time point measured (Fig. 1). However, after nephrons) between mutant and WT littermates, respectively. 28 d of the 8% NaCl diet, mean arterial pressure (MAP) in- However, mutant kidneys showed reduced glomerular and tubular creased in WT (+Δ73 mmHg; P < 0.001) and in mutants (+Δ40 damage compared with WT animals on the 8% NaCl diet for 28 d mmHg; P < 0.01), but to a significantly lesser extent in mutants (Fig. 3 A–C). To determine whether the differences in renal compared with WT (Fig. 1A). Similarly, the salt-dependent in- damage coincided with renal vascular changes, mutant and WT crease in systolic blood pressure in WT (+Δ81 mmHg; P < 0.001) kidneys were stained for alpha-smooth muscle actin to examine was significantly attenuated in mutants (+Δ48 mmHg; P < 0.01) vessel wall thickening. Compared with WT renal vessel media (Fig. 1B), whereas diastolic blood pressures between mutants (9,878 ± 798 μm2), the renal vessel media was significantly less and WT did not significantly differ (Fig. 1C) after 28 d of 8% in the kidneys of SS-Plekha7 mutant rats (6,187 ± 527 μm2; P < NaCl diet. Diurnal MAP was also not different between WT and 0.001) (Fig. 3 D and E). Together, these data demonstrate that

0.4% 0.4% 0.4% ABCNaCl 0.4 or 8% NaCl NaCl 0.4 or 8% NaCl NaCl 0.4 or 8% NaCl

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Fig. 1. Blood pressure was measured by telemetry in WT and mutant rats fed a 0.4% NaCl diet for 31 d (n = 4 per group) or in WT and mutant rats fed a 0.4% NaCl diet for 3 d followed by a 8% NaCl diet for 28 d (n = 8–10 per group). (A) There was no difference in BP between WT and mutant rats fed a 0.4% NaCl diet for 31 d; however, there was a significant attenuation in 8% NaCl-induced increases in MAP in the Plekha7 mutant animals from day 11 to day 28 compared with WT littermates. (B) Plekha7 mutant rats also displayed an attenuation in 8% NaCl-induced elevations in systolic BP from day 3 to day 28 compared with WT. (C) Diastolic BP was lower in the Plekha7 mutant animals compared with WT but was not statistically significant. *P < 0.05. Error bars represent SEM.

12818 | www.pnas.org/cgi/doi/10.1073/pnas.1410745111 Endres et al. Downloaded by guest on September 27, 2021 Plekha7 0.4% 0.4% SS- Mutant Vessels Generate More NO. Because mutation of A 8% NaCl B 8% NaCl NaCl NaCl Plekha7 improved FMD in SS rats and increased sensitivity to ACh-mediated vessel dilation, we next examined levels of NO, a downstream mediator of FMD and ACh signaling (27), in isolated aortas of WT and mutant rats fed a 0.4% NaCl diet. Compared with WT in normal physiological salt solution (PSS) buffer, DAF-FM diacetate staining revealed a 1.65-fold (P < 0.05) increase in NO production in mutant aortas stimulated with ACh, which was completely abrogated by L-NAME treatment (Fig. 6 A and B), suggesting that mutation of Plekha7 -3 71421 28 -3 7 14 21 28 improves NO bioavailability. Time (Days) Time (Days) + Previous studies suggest that NO production is Ca2 -dependent Fig. 2. Renal function analysis of WT and mutant animals. (A)Urinary (28, 29) and that PH domain-containing proteins interact with + microalbumin excretion rate and (B) protein excretion rate was measured in phosphatidylinositol members to regulate intracellular Ca2 sig- Plekha7 WT and mutant rats (n = 10 per group) throughout the BP mea- naling (30, 31). Because Plekha7 contains a PH domain, we hy- surement protocol. There was a significant attenuation in 8% NaCl-induced pothesized that this protein influences NO production by increases in urinary albumin excretion rate and total urinary protein in the 2+ mutant animals compared with the WT. *P < 0.05. Error bars represent SEM. modulating endothelial-specific Ca signaling. To test this hy- pothesis, we examined ACh-mediated release of intracellular + Ca2 in isolated vessels of SS-Plekha7 mutant and WT rats, using + SS-Plekha7 mutant rats exhibit reduced renal and vascular the fluorescent Ca2 indicator (Fluo-4) and two-photon confocal damage that are likely secondary to changes in BP. microscopy. As shown in Fig. 6C, endothelial cells of the bisected aorta exhibit a strong green fluorescence, indicating sufficient SS-Plekha7 Mutant Rats Exhibit Increased Cardiac Output and loading of the Fluo-4 dye. In response to ACh in normal PSS, SS- Reduced Systemic Vascular Resistance. To determine whether re- Plekha7 mutant endothelial cells displayed a marked increased + duced BP and renal damage coincided with changes in cardiac Ca2 response compared with WT as measured by fluorescent function, the hearts of WT and SS-Plekha7 mutant rats on the intensity (Fig. 6 D and E). To determine whether this effect GENETICS 8% NaCl diet for 28 d were examined by echocardiography and exhibited by the mutant endothelial cells was due to extracellular + + histology. Compared with WT, the hearts of mutant rats had Ca2 influx or intracellular Ca2 storage, we carried out similar significantly increased stroke volume and cardiac output (Fig. 4 2+ 2+ A and B), which corresponded to a 44% (P < 0.05) decrease in studies in Ca -free solution. In Ca -free buffer, we found that the mutant endothelial cells continued to exhibit a 2.24-fold (P < systemic vascular resistance in mutant rats (Fig. 4C). The im- + 0.05) increased Ca2 response to ACh compared with WT, in- proved cardiac function in mutant rats also coincided with a re- 2+ duction in interstitial fibrosis and a decrease in perivascular dicating that intracellular Ca storage was higher in the mutant – cells (Fig. 6F). Collectively, these data indicate that mutation of fibrosis compared with WT rats (Fig. 4 D F). Moreover, mutant + rats had significantly reduced heart weight (1.54 ± 0.08 vs. 1.71 ± Plekha7 results in improved Ca2 handling leading to increased 0.04g; P < 0.05) and mean septal wall thickness (0.17 ± 0.01 vs. NO production within the vasculature. 0.25 ± 0.01cm; P < 0.001) compared with WT. Taken together, these results indicate that the SS-Plekha7 mutant rats have improved cardiac function and significantly lower vascular resistance after the 8% NaCl diet treatment compared with WT rats. A Plekha7 WT Plekha7 Mutant B

Plekha7 Modulates Vascular Dilation. Because Plekha7 has been shown to localize to the vasculature (24) and SS-Plekha7 mutant rats had decreased systemic vascular resistance (Fig. 4C), we hy-

pothesized that Plekha7 might directly mediate vascular 1 mm 1 mm function. Vasoreactivity was examined in isolated second-/third- Trichrome C order mesenteric resistance arteries from male rats fed a 0.4% NaCl diet at a point where there were no differences in MAP between WT and mutant rats (108 ± 1 vs. 107 ± 2 mmHg, respectively). Compared with WT, mutant arteries demonstrated a greater sensitivity to the endothelium-dependent relaxation ag- Trichrome 100 μm 100 μm E onist acetylcholine (ACh), whereas no changes were observed in D log EC50 values for sodium nitroprusside or contractile responses to either phenylephrine (PE) or serotonin (5-HT) (Fig. 5A). Ad- ditionally, we found no differences in maximal dilation responses to ACh or sodium nitroprusside or maximal contractile responses to PE or 5-HT between WT and mutant arteries. Interestingly, α-SMA 100 μm 100 μm however, we found that Plekha7 mutant arteries dilated by ∼7% in response to increased flow, whereas WT arteries did not dilate at Fig. 3. Analysis of renal damage. Kidneys were sectioned and stained with all (Fig. 5B). The lack of flow-mediated dilation (FMD) in WT trichrome stain. (A) Histological sections demonstrate that mutant rats are animals is consistent with the impaired endothelium-dependent less damaged than WT littermates. (B) There was a significant reduction in dilation reported previously in the SS rat (25, 26). Moreover, the tubular damage as measured by protein casting and (C) a significant reduction in glomerular sclerosis in the mutants compared with WT. (D) Rep- observed FMD response in Plekha7 mutant arteries could be resentative images of renal vessels from WT and mutant animals stained blocked by the eNOS inhibitor NG-nitro-L-arginine methyl ester with alpha-smooth muscle actin. (E) Vessel wall thickening (vessel media, (L-NAME) (Fig. 5B). Because L-NAME blocked FMD, whereas square micrometers) was calculated by measuring the outer circumference sodium nitroprusside (a NO donor) had no differential effect be- of the vessel minus the inner circumference of the lumen (20 random images tween WT and Plekha7 mutant (Fig. 5A), these data suggest that within the cortex per rat; n = 3 rats per group). There was a significant re- Plekha7 mediates endothelial NO production rather than down- duction in vessel media of the Plekha7 mutant renal arteries compared with stream NO effects on the vascular smooth muscle. WT. ***P < 0.001. Error bars represent SEM.

Endres et al. PNAS | September 2, 2014 | vol. 111 | no. 35 | 12819 Downloaded by guest on September 27, 2021 AB C and fibrosis (Figs. 3D and 4F) and reduced renal damage (Fig. 3A). We drew this conclusion because isolated vessels from normotensive SS-Plekha7 mutant rats had an increased ability to dilate in response to ACh (Fig. 5A) and flow (Fig. 5B) and exhibited elevated bioavailability of NO (Fig. 6A), a key vasodilator and currently the primary indicator of peripheral vascular health in humans (34–36). We also observed decreased perivascular fibrosis in hypertensive mutant rats compared with WT (Fig. 4F), suggesting that decreased vascular stiffening may be an D additional factor in reducing systemic vascular resistance in the SS-Plekha7 mutant rat (37). Multiple other BP mediators have effects on vascular remodeling that may contribute to hypertension [e.g., AngII (38) and ET-1 (39)], suggesting that the Plekha7-mediated BP effects are not mechanistically unique, but Plekha7 WT Trichrome 2 mm 200 μm 100 μm rather one of the few known genetic risk factors to date that contribute to hypertension risk by affecting vascular function and remodeling. Future studies using an endothelial-specific knockout of Plekha7 will be necessary to demonstrate whether it mediates hypertension risk solely through the vascular endothelium or has other nonendothelial roles in BP regulation. 2 mm 200 μm 100 μm Plekha7 Mutant Plekha7 Modulates Vasodilator Responses via Endothelial Calcium EF Signaling. Our whole-animal phenotyping suggested that Plekha7 has a vascular role in BP regulation, prompting us to begin to elucidate the downstream pathways that might mediate Plekha7 signaling. Plekha7 is a cell-surface protein with a PH domain that we hypothesize interacts with phospholipids to affect intracellular levels + of Ca2 (30, 31, 40), the central mediator of vascular tone (41). (% of total heart)

(% of vessel area) Knockdown of HADP1, the closest known orthologous protein to Interstitial Fibrosis 2+ Perivascular Fibrosis PLEKHA7 in zebrafish, was shown to regulate intracellular Ca Fig. 4. Assessment of heart function. Echocardiography was performed in signaling by interaction with the phosphatidylinositol 4-kinase = pathway (31). Furthermore, other PLEKHA family members di- WT and mutant rats (n 5 per group). Plekha7 mutant rats demonstrated 2+ a significant increase in (A) stroke volume and (B) cardiac output compared rectly interact with phosphoinositide members and influence Ca with WT. (C) Mutant rats also had a significant reduction in total peripheral release from intracellular stores, suggesting that Plekha7 may be resistance. (D) Histology of hearts indicated that Plekha7 mutant rat hearts working through similar mechanisms (42, 43). Our data now pro- 2+ were less damaged compared with WT. Mutant hearts showed a significant vide evidence that Plekha7 is involved in Ca signaling and in- 2+ reduction in (E) interstitial fibrosis as well as a significant reduction in (F) tracellular Ca storage in vertebrate mammals as well. Using the 2+ perivascular fibrosis. *P < 0.05. Error bars represent SEM. fluorescent Ca indicator (Fluo-4) and two-photon confocal mi- + croscopy, we observed a significant increase in intracellular Ca2 in response to ACh in isolated vessels from SS-Plekha7 mutant Discussion rats compared with WT littermates in both normal PSS buffer and 2+ 2+ Our goal was to test whether mutation of the GWAS-nominated Ca -free buffer (Fig. 6 D–F). The increased influx of Ca in Plekha7 mutant vessels coincided with elevated NO bioavailability (Fig. 6 A gene affected the development of hypertension in a rat 2+ model of salt-sensitive hypertension. Compared with WT, Ple- and B), fitting with previous reports of Ca -mediated regulation kha7 mutation altered MAP, systolic BP, vasoreactivity, and total of NO bioavailability in the endothelium (28, 29, 44, 45). Because endothelial NO synthase (eNOS) activity is highly dependent on peripheral resistance, suggesting that the vasculature is a major 2+ contributor to Plekha7-mediated BP reduction. This was sup- Ca and inhibition of eNOS with L-NAME blocked FMD and ported by in-depth analysis of renal and cardiac vascular histol- ACh-mediated NO release in the mutant rats, we propose that ogy, isolated vascular contractility, and FMD experiments, which Plekha7 is signaling through this pathway; however, additional indicated that Plekha7 mutation increases endothelial NO bio- + availability driven by increased intracellular storage of Ca2 .We also examined multiple other parameters of hypertension risk (e.g., natriuresis and nephron number) that did not change between WT and Plekha7 mutant animals. However, in the case of natriuresis, daily measurements of sodium handling might provide additional insights that were potentially missed by our weekly measurements. Collectively, these findings provide the first functional evidence to our knowledge that Plekha7 modulates BP by mediating systemic vascular resistance in a hypertensive rat strain. Moreover, to our knowledge these are the first mechanistic data to support multiple GWAS that have nominated PLEKHA7 as a hypertension risk gene (5, 6, 8, 21, 22) but had yet to demonstrate causality. Fig. 5. Assessing vascular reactivity. To determine whether vascular reactivity PLEKHA7 was affected by mutation of Plekha7, mesenteric arteries were isolated from Role of the Vasculature in -Mediated Hypertension Risk. The WT and mutant rats fed a 0.4% NaCl diet (n = 5pergroup).(A) Mutant arteries role of vascular resistance in hypertension is well established (32, displayed an increased sensitivity to the endothelium-dependent dilator ACh. 33), but the genetic mechanism(s) underlying this relationship There was no difference in logEC50 values for sodium nitroprusside, PE, or 5-HT have not been fully elucidated. Here, we demonstrated that in mutants compared with WT. (B) Mutant arteries also displayed improved mutation of Plekha7 in the rat attenuated salt-sensitive hyper- dilatory response to increased flow compared with WT. The flow-mediated tension (Fig. 1), which was likely due to reduced systemic vas- dilation observed in the mutant arteries was completely abolished by the cular resistance (Fig. 4C), leading to decreased vascular damage addition of L-NAME (100 nM). *P < 0.05. Error bars represent SEM.

12820 | www.pnas.org/cgi/doi/10.1073/pnas.1410745111 Endres et al. Downloaded by guest on September 27, 2021 A PSS + ACh D PSS + ACh WT WT Plekha7 50μm Plekha7 20μm Mutant Mutant Plekha7 Plekha7 + Fig. 6. Measurement of NO and Ca2 handling in the vasculature. (A and B) Aortas were extracted from WT and mutant rats fed a 0.4% NaCl diet B Plekha7 WT E F = – 2+ (n 4 5 per group) and were incubated in the NO Plekha7 Mutant Ca -Free Buffer ACh labeling indicator DAF-FM DA. Mutant aortas dis- ACh ACh played an increased fluorescent response to the L-NAME endothelium-dependent dilator ACh. The fluorescent response was abolished in WT and mutant cells ACh by incubating the vessels with L-NAME (100 nM). (C) Aortas from WT and mutant rats (n = 3–5 per group) were extracted, opened longitudinally, labeled with the Ca2+ dye, Fluo-4, and imaged using a two- photon confocal microscope. Endothelial cells exhibited a strong green fluorescent signal indicating that + C Plekha7 WT Vessel - Angled View Ca2 was present in these cells and the cells were loaded properly. (D) Endothelial cells were imaged every 1.11 s in a normal PSS buffer and in response to ACh. (E and F) Mutant endothelial cells displayed GENETICS an increased response to ACh in normal PSS and 2+

Z-Stack Ca -free buffer compared with WT. Arrows point 10μm to the endothelial cells. *P < 0.05. Error bars represent SEM.

studies will be needed to determine the connection between Ple- understand the role of specific risk alleles in regulating Plekha7 kha7 and eNOS. Collectively, these data show for the first time to expression or function. With the growing number and sophisti- + our knowledge that Plekha7 regulates Ca2 signaling in mammals cation of gene manipulation technologies now available and and modulates endothelium-dependent vasodilator responses via applicable to both whole-animal and human cell model systems, + the Ca2 and NO signaling axis in a disease model of salt-sensitive we will now be able to specifically test how individual candidate + hypertension. Because Ca2 can also regulate PGI2 and a number SNPs regulate gene function using multiple approaches. of different processes, more studies are needed to identify the + contribution of Plekha7 and Ca2 to other endothelium-dependent Materials and Methods control mechanisms, such as prostanoid release. All animal procedures were approved by the Institutional Animal Care and Use Committee at the Medical College of Wisconsin. Perspectives. Although GWAS can nominate candidate genes, they are unable to establish causality owing to the difficulty in Generation of Mutant Rats. ZFN constructs were designed, assembled, and investigating disease mechanism(s) for candidate genes directly validated by Sigma-Aldrich. mRNA encoding the Plekha7 ZFNs was diluted in in humans, especially for complex physiological phenotypes such microinjection buffer (1 mM Tris and 0.1 mmol/L EDTA, pH 7.4) and injected as the control of blood pressure. As such, we have undertaken into one-cell SS/JrHsdMcwi rat embryos as described previously (48). See SI identifying and functionally testing candidate genes on disease- Materials and Methods and Fig. S1. susceptible rat models. The foundation of this approach has been Blood Pressure, Renal Function, Urine/Serum Biochemistry, and Echocardiography. the newly developed gene-editing techniques (e.g., ZFN, TAL- + All rats were initially fed a 0.4% NaCl diet (113755; Dyets Inc.). Urine volume, Na , ENS, CRISPR/Cas9, etc.) that have recently been applied to the + rat (20, 46, 47). The advantages of this approach are multifold: K , creatinine, protein, and microalbumin were measured as described pre- (i) Disease-susceptible strains already harbor multiple disease viously (49). BP was measured as described previously (49). Urine measurements alleles that are necessary to manifest the modest effect sizes of were taken weekly. Echocardiography was performed on experimental rats complex disease alleles; (ii) >10 disease-susceptible strains have using a GE Vivid 7 ultrasound machine (GE Healthcare) as described previously now been modified (see http://rgd.mcw.edu/wg/physgenknockouts), (50). Cardiac output was calculated in anesthetized rats by multiplying stroke volume (measured by echocardiography) by the HR (measured by telemetry). demonstrating that current gene-editing approaches are directly Total peripheral resistance was calculated by dividing the MAP (measured by applicable to different backgrounds enabling candidate risk telemetry) by the cardiac output. See SI Materials and Methods. genes to be tested in multiple disease models and under various experimental conditions; and (iii) gene editing is possible on Kidney and Heart Histology. Kidneys and hearts were sectioned, mounted on pure inbred strains with uniform genetic backgrounds, limiting microscope slides, and stained with trichrome. Immunohistochemistry (IHC) confounding genetic heterogeneity. These considerations, com- was performed using an antibody against alpha-smooth muscle actin using bined with a controlled experimental setting and ability to per- standard IHC methodology (49). Vessel media was calculated by measuring form mechanistic studies, have enabled significant advances in the outer circumference of the vessel minus the inner circumference of the characterizing complex disease candidates, such as PLEKHA7. lumen (20 random vessels within the cortex). See SI Materials and Methods. Based on the success of testing Plekha7 function in the SS- + Plekha7 mutant rat model, we propose that the next step to Vascular Reactivity Studies and NO and Ca2 measurements. Mesenteric arteries translating these data to human hypertension will be to better were either mounted on a wire myograph (DMT) as described previously (51) or

Endres et al. PNAS | September 2, 2014 | vol. 111 | no. 35 | 12821 Downloaded by guest on September 27, 2021 cannulated as described previously (25, 26). For vasodilation studies, vessels ACKNOWLEDGMENTS. We thank Dr. Debebe Gebremedhin, Dr. Fara Saez, Dr. were preconstricted with 10 μM PE and drug-response curves to either ACh or Daria Ilatovskaya, Dr. Jennifer Strande, Dr. Chris McDermott-Roe, Leanne sodium nitroprusside were measured. Drug-response curves for contractile Harmann, Christine Duris, Lisa Henderson, Camille Torres, Jennifer Phillips, force were performed for PE and 5-HT. Flow studies were carried out as de- Rebecca Schilling, Jason Klotz, Jocelyn Miller, Liz Schneider, and Laura Utschig- Samuels for their excellent technical support and tremendous help with this scribed previously (52). NO levels were measured in aortas using DAF-FM project. We thank Dr. Rebecca Martinie for excellent help with proofreading diacetate and imaging the fluorescent response to the addition of 2 mM ACh, 2+ and editing the manuscript. We also thank Dr. Marie Schulte, Dr. William as described previously (45). Ca transient studies were performed on aortas Cashdollar, and the Northwestern Mutual Foundation Imaging Center at the loaded with Fluo-4. Fluorescence was imaged in vessels (n = 3–5ratspergroup) Blood Research Institute of Wisconsin for help with the imaging studies. This + after addition of 2 mM ACh. Average Ca2 transients of endothelial cells (n = study was supported by National Institutes of Health Grants RC2-HL101681 (to 25–60 cells per group) were calculated. See SI Materials and Methods. H.J.J.), HL108880 (to A.S.), and DP2-OD008396 (to A.M.G.).

1. Kotchen TA, Cowley AW, Jr, Frohlich ED (2013) Salt in health and disease—a delicate 26. Frisbee JC, Roman RJ, Murali Krishna U, Falck JR, Lombard JH (2001) Altered mech- balance. N Engl J Med 368(13):1229–1237. anisms underlying hypoxic dilation of skeletal muscle resistance arteries of hyper- 2. Padmanabhan S, Newton-Cheh C, Dominiczak AF (2012) Genetic basis of blood tensive versus normotensive Dahl rats. Microcirculation 8(2):115–127. pressure and hypertension. Trends Genet 28(8):397–408. 27. Cockcroft JR (2005) Exploring vascular benefits of endothelium-derived nitric oxide. 3. Cowley AW, Jr (2006) The genetic dissection of essential hypertension. Nat Rev Genet Am J Hypertens 18(12 Pt 2):177S–183S. 7(11):829–840. 28. Fleming I, Busse R (1999) Signal transduction of eNOS activation. Cardiovasc Res 43(3): 4. Coffman TM (2011) Under pressure: the search for the essential mechanisms of hy- 532–541. pertension. Nat Med 17(11):1402–1409. 29. Busse R, Fleming I (1995) Regulation and functional consequences of endothelial ni- 5. Levy D, et al. (2009) Genome-wide association study of blood pressure and hyper- tric oxide formation. Ann Med 27(3):331–340. tension. Nat Genet 41(6):677–687. 30. Carlisle RE, et al. (2012) TDAG51 mediates epithelial-to-mesenchymal transition in 6. Hong KW, et al. (2010) Recapitulation of two genomewide association studies on human proximal tubular epithelium. Am J Physiol Renal Physiol 303(3):F467–F481. blood pressure and essential hypertension in the Korean population. J Hum Genet 31. Wythe JD, et al. (2011) Hadp1, a newly identified pleckstrin homology domain pro- 55(6):336–341. tein, is required for cardiac contractility in zebrafish. Dis Model Mech 4(5):607–621. 7. Ehret GB, et al.; International Consortium for Blood Pressure Genome-Wide Associ- 32. Luft FC (2012) Molecular mechanisms of arterial stiffness: New insights. J Am Soc ation Studies; CARDIoGRAM consortium; CKDGen Consortium; KidneyGen Consor- Hypertens 6(6):436–438. tium; EchoGen consortium; CHARGE-HF consortium (2011) Genetic variants in novel 33. Blacher J, Safar ME (2005) Large-artery stiffness, hypertension and cardiovascular risk pathways influence blood pressure and cardiovascular disease risk. Nature 478(7367): in older patients. Nat Clin Pract Cardiovasc Med 2(9):450–455. – 103 109. 34. Green DJ, Dawson EA, Groenewoud HM, Jones H, Thijssen DH (2014) Is flow-mediated 8. Fox ER, et al.; International Consortium for Blood Pressure Genome-wide Association dilation nitric oxide mediated?: A meta-analysis. Hypertension 63(2):376–382. Studies (ICBP-GWAS) (2011) Association of genetic variation with systolic and diastolic 35. Green DJ, Jones H, Thijssen D, Cable NT, Atkinson G (2011) Flow-mediated dilation and blood pressure among African Americans: The Candidate Gene Association Resource cardiovascular event prediction: Does nitric oxide matter? Hypertension 57(3):363–369. – study. Hum Mol Genet 20(11):2273 2284. 36. Levy AS, Chung JC, Kroetsch JT, Rush JW (2009) Nitric oxide and coronary vascular 9. Mattson DL, et al. (2004) Influence of diet and genetics on hypertension and renal endothelium adaptations in hypertension. Vasc Health Risk Manag 5:1075–1087. – disease in Dahl salt-sensitive rats. Physiol Genomics 16(2):194 203. 37. Intengan HD, Schiffrin EL (2001) Vascular remodeling in hypertension: Roles of apo- 10. Cowley AW, Jr (1997) Genetic and nongenetic determinants of salt sensitivity and ptosis, inflammation, and fibrosis. Hypertension 38(3 Pt 2):581–587. blood pressure. Am J Clin Nutr 65(2, Suppl):587S–593S. 38. Torres VE, Harris PC (2014) Strategies targeting cAMP signaling in the treatment of 11. Mattson DL, et al. (2008) Chromosome substitution reveals the genetic basis of Dahl salt- polycystic kidney disease. J Am Soc Nephrol 25(1):18–32. sensitive hypertension and renal disease. Am J Physiol Renal Physiol 295(3):F837–F842. 39. Khalil RA (2011) Modulators of the vascular endothelin receptor in blood pressure 12. Campese VM (1994) Salt sensitivity in hypertension. Renal and cardiovascular im- regulation and hypertension. Curr Mol Pharmacol 4(3):176–186. plications. Hypertension 23(4):531–550. 40. Yamada K, Nomura N, Yamano A, Yamada Y, Wakamatsu N (2012) Identification and 13. Cowley AW, Jr, Roman RJ (1996) The role of the kidney in hypertension. JAMA characterization of splicing variants of PLEKHA5 (Plekha5) during brain development. 275(20):1581–1589. Gene 492(1):270–275. 14. Grim CE, et al. (1990) Blood pressure in blacks. Twin studies in Barbados. Hypertension 41. Adams DJ, Barakeh J, Laskey R, Van Breemen C (1989) Ion channels and regulation of 15(6 Pt 2):803–809. intracellular calcium in vascular endothelial cells. FASEB J 3(12):2389–2400. 15. Lackland DT, Keil JE (1996) Epidemiology of hypertension in African Americans. Semin 42. Marshall AJ, Krahn AK, Ma K, Duronio V, Hou S (2002) TAPP1 and TAPP2 are targets Nephrol 16(2):63–70. of phosphatidylinositol 3-kinase signaling in B cells: Sustained plasma membrane 16. Mattson DL, et al. (2013) Genetic mutation of recombination activating gene 1 in recruitment triggered by the B-cell antigen receptor. Mol Cell Biol 22(15):5479–5491. Dahl salt-sensitive rats attenuates hypertension and renal damage. Am J Physiol Regul 43. Krahn AK, Ma K, Hou S, Duronio V, Marshall AJ (2004) Two distinct waves of membrane- Integr Comp Physiol 304(6):R407–R414. proximal B cell antigen receptor signaling differentially regulated by Src homology 2- 17. Rudemiller N, et al. (2013) CD247 modulates blood pressure by altering T-lymphocyte – infiltration in the kidney. Hypertension 63(3):559–64. containing inositol polyphosphate 5-phosphatase. J Immunol 172(1):331 339. 18. Feng D, et al. (2012) Increased expression of NAD(P)H oxidase subunit p67(phox) in 44. Kida M, Sugiyama T, Yoshimoto T, Ogawa Y (2013) Hydrogen sulfide increases nitric the renal medulla contributes to excess oxidative stress and salt-sensitive hyperten- oxide production with calcium-dependent activation of endothelial nitric oxide syn- – sion. Cell Metab 15(2):201–208. thase in endothelial cells. Eur J Pharm Sci 48(1-2):211 215. 19. Flister MJ, et al. (2013) Identifying multiple causative genes at a single GWAS locus. 45. Zhu J, et al. (2006) Role of superoxide and angiotensin II suppression in salt-induced + Genome Res 23(12):1996–2002. changes in endothelial Ca2 signaling and NO production in rat aorta. Am J Physiol – 20. Geurts AM, et al. (2009) Knockout rats via embryo microinjection of zinc-finger nu- Heart Circ Physiol 291(2):H929 H938. cleases. Science 325(5939):433. 46. Tesson L, et al. (2011) Knockout rats generated by embryo microinjection of TALENs. 21. Ho JE, et al. (2011) Discovery and replication of novel blood pressure genetic loci in Nat Biotechnol 29(8):695–696. the Women’s Genome Health Study. J Hypertens 29(1):62–69. 47. Li D, et al. (2013) Heritable gene targeting in the mouse and rat using a CRISPR-Cas 22. Lin Y, et al. (2011) Genetic variations in CYP17A1, CACNB2 and PLEKHA7 are associ- system. Nat Biotechnol 31(8):681–683. ated with blood pressure and/or hypertension in She ethnic minority of China. Ath- 48. Geurts AM, et al. (2010) Generation of gene-specific mutated rats using zinc-finger erosclerosis 219(2):709–714. nucleases. Methods Mol Biol 597:211–225. 23. Pulimeno P, Bauer C, Stutz J, Citi S (2010) PLEKHA7 is an adherens junction protein 49. Moreno C, et al. (2011) Creation and characterization of a renin knockout rat. Hy- with a tissue distribution and subcellular localization distinct from ZO-1 and E-cad- pertension 57(3):614–619. herin. PLoS ONE 5(8):e12207. 50. Flister MJ, et al. (2012) Identification of hypertension susceptibility loci on rat chro- 24. Lee MC, et al. (2014) Expression of the primary angle closure glaucoma (PACG) sus- mosome 12. Hypertension 60(4):942–948. ceptibility gene PLEKHA7 in endothelial and epithelial cell junctions in the eye. Invest 51. Zhu J, Huang T, Lombard JH (2007) Effect of high-salt diet on vascular relaxation and Ophthalmol Vis Sci 55(6):3833–3841. oxidative stress in mesenteric resistance arteries. J Vasc Res 44(5):382–390. 25. Frisbee JC, Roman RJ, Krishna UM, Falck JR, Lombard JH (2001) 20-HETE modulates 52. Zhang DX, et al. (2012) H2O2-induced dilation in human coronary arterioles: Role of myogenic response of skeletal muscle resistance arteries from hypertensive Dahl-SS protein kinase G dimerization and large-conductance Ca2+-activated K+ channel rats. Am J Physiol Heart Circ Physiol 280(3):H1066–H1074. activation. Circ Res 110(3):471–480.

12822 | www.pnas.org/cgi/doi/10.1073/pnas.1410745111 Endres et al. Downloaded by guest on September 27, 2021