BASIC RESEARCH www.jasn.org

GPR48 Increases Mineralocorticoid Expression

† † Jiqiu Wang,* Xiaoying Li,* Yingying Ke,* Yan Lu,* Feng Wang, Nengguang Fan,* ‡ Haiyan Sun,* Huijie Zhang,* Ruixin Liu,* Jun Yang,* Lei Ye,* Mingyao Liu, and † Guang Ning*

*Shanghai Clinical Center for Endocrine and Metabolic Diseases, Shanghai Institute of Endocrinology and Metabolism, Shanghai Key Laboratory for Endocrine Tumors and E-Institute of Shanghai Universities, Ruijin Hospital, Shanghai Jiaotong University School of Medicine, Shanghai, China; †Laboratory for Endocrine and Metabolic Diseases, Institute of Health Sciences, Shanghai Institutes for Biological Sciences (SIBS), Chinese Academy of Sciences, Shanghai, China; and ‡Institute of Biosciences and Technology, Texas A&M University Health Science Center, Houston, Texas

ABSTRACT Aldosterone and the mineralocorticoid receptor (MR) are critical to the maintenance of electrolyte and BP . Mutations in the MR cause aldosterone resistance known as pseudohypoaldosteronism type 1 (PHA1); however, some cases consistent with PHA1 do not exhibit known gene mutations, suggesting the possibility of alternative genetic variants. We observed that G –coupled receptor 48 (Gpr48/ Lgr4) hypomorphic mutant (Gpr48m/m) mice had hyperkalemia and increased loss and salt excretion despite elevated plasma aldosterone levels, suggesting aldosterone resistance. When we challenged the mice with a low-sodium diet, these features became more obvious; the mice also developed and increased renin expression and activity, resembling a mild state of PHA1. There was marked renal downregulation of MR and its downstream targets (e.g., the a-subunit of the amiloride-sensitive epithelial sodium channel), which could provide a mechanism for the aldosterone resistance. We identified a non- canonical cAMP-responsive element located in the MR promoter and demonstrated that GPR48 upregulates MR expression via the cAMP/ pathway in vitro. Taken together, our data demonstrate that GPR48 enhances aldosterone responsiveness by activating MR expression, suggesting that GPR48 contributes to homeostasis of electrolytes and BP and may be a candidate gene for PHA1.

J Am Soc Nephrol 23: 281–293, 2012. doi: 10.1681/ASN.2011040351

INTRODUCTION Na+ reabsorption by activated ENaC also enhances K+ excretion through the luminal potassium chan- Aldosterone plays critical roles in the control of nel, leading to kaliuresis in the collecting duct.4 salt homeostasis and BP through stimulating Na+ Loss-of-function mutations in the MR or ENaC reabsorption and K+ secretion.1 Upon binding to subunitgenesaccountforpseudohypoaldosteronism mineralocorticoid receptor (MR), the hormone– type 1 (PHA1),5,6 which is the principal form of aldo- receptor complex translocates into the nucleus sterone resistance and shows an autosomal-dominant and interacts with glucocorticoid receptor response element in specific promoter regions of the target Received April 7, 2011. Accepted October 6, 2011. , thereby activating their transcription, such Published online ahead of print. Publication date available at as a-subunit of the amiloride-sensitive epithelial www.jasn.org. sodium channel (aEnaC) in the aldosterone- Correspondence: Dr. Guang Ning, Shanghai Clinical Center for 2 sensitive distal . The functional relevance Endocrine and Metabolic Diseases, Ruijin Hospital, Shanghai of ENaC to aldosterone-dependent Na+ reabsorp- Jiaotong University School of Medicine, 197 Ruijin 2nd Road, tion, and thus to the regulation of extracellular fluid Shanghai 200025, China. Email: [email protected] volume and BP, is well established.3 Promotion of Copyright © 2012 by the American Society of Nephrology

J Am Soc Nephrol 23: 281–293, 2012 ISSN : 1046-6673/2302-281 281 BASIC RESEARCH www.jasn.org

(adPHA1; OMIM #177735) and a recessive (arPHA1; OMIM trap–mutated Gpr48 ES cells into C57BL/6 blastocysts.27 #264350) transmission.7 First described in 1958, this rare con- An insertion of the trap vector into intron 1 of the Gpr48 dition features renal resistance to aldosterone action.8 Early in gene resulted in approximately 90% knockdown efficiency infancy, patients with adPHA1 present with dehydration and in the and adrenal gland of adult Gpr48m/m mice (Fig- failure to thrive associated with salt wasting, hypotension, hy- ure 1, A–C). We observed that approximately half of Gpr48m/m perkalemia, and metabolic acidosis, despite increased plasma newborns died within 28 hours after birth, but no further aldosterone levels.6 These patients can improve with age, and deaths occurred in the following 20 hours (Figure 1D). some adult patients are usually asymptomatic and have fewer abnormal biochemical findings (e.g., only lifelong increases Increases in Water Intake and Urine Volume with in aldosterone or hyperkalemia).9,10 To date, approximately Partially Impaired Urine-Concentrating Ability 50 distinct mutations in the human MR gene and approxi- in Gpr48m/m Mice mately 20 mutations in ENaC genes responsible for PHA1 We explored the potential changes of water balance in Gpr48m/m have been described.11,12 However, some patients, especially mice using metabolic cages and found that Gpr48m/m mice those with sporadic PHA1, do not have genetic abnormalities showed a dramatic increase in 24-hour water and food intake in MR or ENaC.6,10,13–15 Because PHA1 is life-threatening and at 16 and 24 weeks compared with wild-type mice (Figure 2, many probands may be missed as a result of early death, addi- A and B). Their urine volume was also significantly increased tional genetic mechanisms might participate in its pathogene- during this time (Figure 2C). Meanwhile, urine osmolality sis. According to recent studies, aldosterone resistance may also declined markedly (Figure 2D). Female mice showed overall be associated with a reduction in MR expression, probably similar phenotypes (Supplemental Figure S1). We next per- mediated by transcriptional mechanisms.16–19 formed a water deprivation test to assess the urine-concentrating Gprotein–coupled receptor 48 (GPR48/ LGR4) belongs to the –coupled receptor superfamily. It has recently been re- ported to bind R-spondin and mediate its signaling in intestinal crypt cells,20,21 yet its function has not been well investigated.22 GPR48 is critical in development, and Gpr48- mutant mice display early neonatal lethality.23 Our group and others have demonstrated that Gpr48 deficiency results in impaired func- tion of the male reproductive tract through downregulation of estrogen receptor a ex- pression.24,25 GPR48 is also involved in colon carcinoma metastasis, development of ocular anterior segment, and bone formation through different downstream targets.26–28 However, the involvement of this protein in electrolyte balance has not been described. In this study, we find that Gpr48 hypo- morphic mutant mice display a significant aldosterone resistance, which mimics a mild state of adPHA1 disease. We also dem- onstrate that GPR48 regulates MR expres- sion through the cAMP/protein kinase A (PKA) pathway. This study elucidates the potential role of GPR48 in electrolyte ho- meostasis and aldosterone resistance. Figure 1. Residual levels of Gpr48 transcripts and neonatal survival rate of Gpr48m/m mice. (A) Approximately 10% of Gpr48 transcripts remained in the kidney of Gpr48m/m RESULTS mice at age 16 weeks according to quantitative PCR (n=12). (B) Reverse transcription PCR for Gpr48 expression in the kidney of Gpr48m/m mice at age 16 weeks (n=6). The corre- sponding cDNA length is 471 bp, and the PCR products were verified to be wild-type Gpr48 Homozygous Mutant Mice Gpr48 by sequencing. (C) Approximately 10% of Gpr48 transcripts remained in the adrenal Show Hypomorphic Features gland of Gpr48m/m mice at age 16 weeks according to qualitative PCR (n=12). 36B4 was We obtained Gpr48 hypomorphic mutant used as internal control. (D) Survival curve of neonatal wild-type (Gpr48+/+)(n=29) and (Gpr48m/m) mice by microinjecting gene Gpr48m/m (n=23) mice within 48 hours after birth. Error bars represent SEM. ***P,0.001.

282 Journal of the American Society of Nephrology J Am Soc Nephrol 23: 281–293, 2012 www.jasn.org BASIC RESEARCH

between the two genotypes, nor did pH, 2 HCO3 , or base excess (Table 1 and Sup- plemental Table S1). We next examined plasma mineralocor- ticoid levels, which play fundamental roles in regulating water and salt homeostasis. Surprisingly, plasma aldosterone was sig- nificantly higher in Gpr48m/m mice than in wild-type mice (302.6630.0 versus 155.3611.9 pg/ml; P,0.001) (Figure 4A). Their daily urinary aldosterone excretion was also increased (Figure 4B). However, this aldosterone excess did not have the expected effects of excreting K+ while pre- serving Na+ and water, suggesting aldoste- rone resistance in Gpr48m/m mice. Plasma renin activity (PRA) showed no difference but did have a tendency to increase in Gpr48m/m mice (Figure 4C). In addition, plasma corticosterone concentration did not show any difference (Figure 4, D–F). Morphologic analysis of the adrenal gland revealed no obvious changes in Gpr48m/m mice (Supplemental Figures S2 and S3A), nor did the gene expression of the key en- zymes in the adrenal gland and the plasma levels of adrenergic hormones (Supple- mental Figure S3, B–E), suggesting that the phenotypes observed in Gpr48m/m Figure 2. Adult Gpr48m/m mice show polydipsia, polyphagia, severe water loss, and re- mice are not secondary to defects in the duced urine osmolality. Male Gpr48m/m mice at age 16 and 24 weeks displayed increased adrenal gland. Taken together, these data (A) water intake, (B) food intake, and (C) urine volume but (D) decreased urine osmolality suggest that Gpr48m/m mice displayed so- compared with their age-matched wild-type littermates. At 8 weeks, n=4; at 16 weeks, +/+ dium and water loss as well as hyperkale- n=9; at 24 weeks, n=10. White bar, wild-type mice (Gpr48 ); black bar, Gpr48-mutant mia, despite elevated aldosterone, thus mice (Gpr48m/m). Error bars represent SEM. **P,0.01; ***P,0.001. BW, body weight. resembling the mild state of adPHA1.6,32 ability of Gpr48m/m mice. The urine osmolalities of both Enhanced adPHA1 Phenotype in Gpr48m/m Mice genotypes were markedly increased by a similar extent after on a Low-Sodium Diet water deprivation; however, that of Gpr48m/m mice still did To exclude the effects of compensatory salt intake on Na+ not reach the level of wild-type mice (Figure 3A), suggesting a balance, we challenged Gpr48m/m and wild-type mice with a partial urine-concentrating defect. However, neither geno- low-sodium diet proportionate to their body weight. Urinary type showed any difference in hypothalamic Na+ excretion was reduced in both groups (Figure 5A). How- expression or renal 2 and aquaporin 2 ever, except for a marked reduction on the first day after salt expression, which are all associated with diabetes insipidus deprivation, Gpr48m/m mice overall displayed limited changes (Figure 3, B–E).29–31 in Na+ excretion compared with wild-type mice and exhibited corresponding hyponatremia (Figure 5, A and B), suggesting Aldosterone Resistance Exhibited by Gpr48m/m Mice impaired Na+ conserving ability. Hyperkalemia persisted in Water and electrolyte homeostasis are tightly controlled in the Gpr48m/m mice as under normal chow (Figure 5C). Meanwhile, kidney. As shown in Table 1, the daily Na+ excretion was in- obvious depletion in extracellular fluid volume was seen in creased by 1.5 times in Gpr48m/m mice compared with excre- Gpr48m/m mice, indicated by increased hematocrit and blood tion in wild-type mice. Although urinary K+ and creatinine hemoglobin content (Figure 5, D and E). Consequently, the excretion showed no significant change, plasma K+ was strik- systolic, mean arterial, and diastolic BPs in Gpr48m/m mice ingly elevated in Gpr48m/m mice (6.760.3 versus 5.560.2 mM; were markedly decreased (Figure 5F). Gpr48m/m mice also P,0.001). The urinary Na+/K+ ratio was also higher. Plasma showed higher PRA and plasma aldosterone levels (Figure 5, 2 Na+,Cl , creatinine, and BUN concentrations did not differ G and H). Consistent with PRA changes, the renal renin

J Am Soc Nephrol 23: 281–293, 2012 GPR48 in Aldosterone Resistance 283 BASIC RESEARCH www.jasn.org

morphologic abnormalities or apparent pathologic defects were observed in the kidney of Gpr48m/m mice, except for re- duced kidney size (Figure 6, B–D). After an examination of the expression of genes that regulate water and salt balance, we found that both the mRNA and protein levels of MR were dramatically reduced in the kidney of Gpr48m/m mice compared with wild type, whereas glucocorticoid re- ceptor (GR) expression did not differ (Fig- ure 6, E–G). Accordingly, the expression of two MR downstream targets, aENaC and Na/K-adenosine triphosphatase (ATPase) 1a (Na/K-ATPase 1a), also significantly decreased, but the expression of inwardly rectifying K+ channel showed no difference between the genotypes (Figure 6, E–G). Furthermore, both the distal tubules and collecting ducts displayed significant de- creases in MR and Na/K-ATPase 1a immu- nostaining in Gpr48m/m mice (Figure 6H). These results suggest that the reduced ex- pression of MR and MR-driven genes could be the main reason for the electrolyte dis- turbance observed in Gpr48m/m mice.

GPR48 Regulates MR Gene Expression through the cAMP/PKA Pathway We isolated and established mouse embry- onic fibroblasts (MEFs) of both genotypes (Figure 7A), and MR expression was con- sistently decreased in Gpr48m/m MEFs (MEFGpr48m/m) compared with wild type (Figure 7B and Supplemental Figure S4A). We further investigated whether Figure 3. Partially impaired urine-concentrating ability in Gpr48m/m mice. (A) Urine GPR48 could directly regulate MR expres- Gpr48m/m osmolality of wild-type and Gpr48m/m mice aged 16 weeks was measured under 8-hour sion. As shown in Figure 7C, MEF water deprivation (n=11). (B–D) Relative mRNA expression levels of vasopressin (AVP) showed much lower MR promoter activity Gpr48+/+ in the (B) and (AVPR2) (C) and aquaporin 2 (AQP2) than wild-type MEFs (MEF ). More- (D) in the kidney (n=11). (E) Representative immunohistochemical staining of aqua- over, GPR48 overexpression increased MR porin2inrenalcollectingducts(scalebars,50mm). n.s., not significant. Error bars transcriptional activity in a dose-dependent represent SEM. manner (Figure 7D). The intracellular sig- naling and downstream targets of GPR48 expression was also increased in Gpr48m/m mice (Figure 5, are mediated by the cAMP/PKA pathway.27,28 As shown in I–K). These data demonstrate that Gpr48m/m mice displayed Figure 7, E and F, the adenylate cyclase agonist forskolin and more severe electrolyte and hormone abnormalities when de- the phosphodiester inhibitor 3-isobutyl-1-methyl-xanthine in- prived of salt than did wild-type mice. creased MR promoter activity in MEFGpr48m/m or HEK293 cells. Consistently, the PKA inhibitor H89 (Figure 7G and Sup- Morphologic and Gene Expression Changes plemental Figure S4B) and the adenylate cyclase antagonist in the Kidney of Gpr48m/m Mice M182 (Supplemental Figure S4C) inhibited MR promoter ac- Gpr48 was ubiquitously expressed in different tissues of mice, tivity in a dose-dependent manner. Moreover, H89 abolished with the highest expression in the kidney (Figure 6A), sug- GPR48-induced MR transcriptional activity (Figure 7H). The gesting its critical role in this organ. However, no gross classic cAMP-responsive element (CRE) reporter was used as a

284 Journal of the American Society of Nephrology J Am Soc Nephrol 23: 281–293, 2012 www.jasn.org BASIC RESEARCH

Table 1. Blood and urine measures of male wild-type and positive control (Figure 7I). These results show that GPR48 Gpr48m/m mice on the normal chow diet (n=10) regulates MR transcription through the cAMP/PKA pathway. Variable Gpr48+/+ Gpr48m/m By using two software programs (TFSEARCH and TESS), Blood we found a noncanonical CRE, TAGCGTCA, located be- Na+ (mM) 144.260.7 145.461.2 tween 2472 and 2465 bp in the MR promoter. Subsequently, K+ (mM) 5.560.2 6.760.3a one CRE mutant and three truncated MR promoter luciferase 2 Cl (mM) 117.261.1 120.062.3 constructs were generated and transfected into HEK293 cells. creatinine (mM) 16.961.6 19.562.6 The CRE mutant and 2321 truncated MR promoter con- BUN (mM) 9.260.4 10.861.0 structs displayed only half of the transcriptional activity of pH 7.2960.01 7.2760.02 the full-length MR promoter, and H89 inhibited the transcrip- 2 6 6 HCO3 (mM) 19.8 0.6 23.5 2.9 tional activity of the full-length MR promoter, but not the 2 6 2 6 BE (mM) 6.7 0.6 3.5 3.0 CRE-mutant or truncated promoters (Figure 8A). Moreover, hemoglobin (g/dl) 15.360.2 15.260.2 the MR promoter with the mutant CRE failed to be activated hematocrit (% packed cell volume) 44.960.7 44.860.7 Urine by GPR48 overexpression (Figure 8B). These results suggest Na+ (mmol/g body wt per day) 8.761.0 13.061.1b that CRE is essential for GPR48-activated MR transcription. K+ (mmol/g body wt per day) 13.961.7 18.361.5 We next examined the interaction of phospho-CRE binding 2 Cl (mmol/g body wt per day) 13.261.7 18.861.7b protein (pCREB) with the CRE sequence. As shown in Figure creatinine (pmol/g body wt per day) 0.2560.02 0.2960.02 8C, abundant DNA fragments contained the CRE sequence Na+/K+ ratio 0.6460.01 0.7160.02b from the MR promoter bound to pCREB in MEFGpr48+/+, Values are expressed as the mean 6 SEM. BE, base excess. whereas the binding was dramatically reduced in MEFGpr48m/m a , P 0.001. (Figure 8C). This CRE was confirmed in vitro using an electro- bP,0.01. phoresis mobility-shift assay (EMSA) with a wild-type or mu- tant CRE probe designed according to the mouse MR promoter (Figure 8D). The exact band was confirmed by gel shift assay with CREB antibody and a CRE-containing fragment from the mouse somatostatin promoter (positive control).33 These find- ings identify a functional CRE in the MR promoter that can directly bind to pCREB, thereby mediating GPR48-induced MR ex- pression (Figure 8E).

DISCUSSION

In this study, we demonstrated that Gpr48m/m mice displayed salt loss, hyponatremia, and hyperkalemia despite aldosterone ex- cess, which was associated with reduced renal expression of MR and its targets. These phenotypes, especially under a low- sodium diet challenge, resembled the char- acteristics exhibited by adPHA1 patients.6

Mild adPHA1 Phenotype Caused by a Gpr48 Hypomorphic Mutant-Induced MR Downregulation adPHA1 caused by MR mutations is char- acterized by salt wasting, hypotension, and Figure 4. Elevated plasma aldosterone levels and urine aldosterone excretion in hyperkalemia despite elevated plasma al- 6 Gpr48m/m mice. (A–C) Elevated aldosterone concentrations (n=9) (A), increased urinary dosterone levels. The appearance of salt aldosterone excretion (n=14) (B), and marginally unchanged plasma renin activity (n=9) loss as well as hormone abnormalities in m/m (C) in Gpr48m/m mice. (D–F) Unaltered plasma corticosterone was observed in Gpr48m/m Gpr48 mice resembled a mild state of mice under normal conditions (n=8) (D), 8-hour fasting (n=8) (E), and stimulation (n=5) adPHA1. These phenotypes were similar to (F). Error bars represent SEM. ***P,0.001. MR heterozygous mutant mice but less

J Am Soc Nephrol 23: 281–293, 2012 GPR48 in Aldosterone Resistance 285 BASIC RESEARCH www.jasn.org

Figure 5. Gpr48m/m mice exhibit adPHA1 features on a low-sodium diet. (A) Impaired salt-conserving ability was observed in Gpr48m/m mice fed the low-sodium diet. Mice were placed into metabolic cages initially, with normal chow for 3 days (from D23 to D0), followed by the low-sodium diet for 5 days (from D0 to D5). Salt deprivation started on D0. (B and C) Reduced plasma Na+ and increased K+ in Gpr48m/m mice (n=12 for Gpr48+/+; n=8 for Gpr48m/m). (D and E) Dehydration indicated by increased hematocrit and hemoglobin content was observed in Gpr48m/m mice (n=12 for Gpr48+/+; n=8 for Gpr48m/m). (F) Systolic (SBP), mean arterial (MBP), and diastolic (DBP) BP were markedly decreased in Gpr48m/m mice on the low-sodium diet (n=14 for Gpr48+/+; n=8 for Gpr48m/m). (G and H) Elevated PRA and al- dosterone in Gpr48m/m mice (n=8). (I–K) Renin mRNA (n=8) (I) and protein (two mice mixed as a pool, n=6–8) (J and K) were markedly elevated in the kidney of Gpr48m/m mice on the low-sodium diet. White bar, wild-type mice (Gpr48+/+); black bar, Gpr48-mutant mice (Gpr48m/m). Error bars represent SEM. *P,0.05; **P,0.01; ***P,0.001. severe than MR homozygous mutant mice,34,35 which are GPR48 in the regulation of renal MR transcription via the most likely attributable to some functional MR re- cAMP/PKA pathway and identified a functional CRE site in tained in the kidney of Gpr48m/m mice. Our results are also the MR promoter. This is consistent with previous reports on supported by the phenotypes of rats with MR knockdown, the mechanism underlying GPR48-mediated target gene ex- which exhibit Na+/K+ disturbance and a significant inverse pression.24,27 correlation between MR downregulation and plasma aldoste- rone level.19 There is also clinical evidence for the association Salt Deprivation Exacerbates Electrolyte and Hormone of low renal MR expression with aldosterone resistance in Abnormalities in Gpr48m/m Mice neonates.36 These findingssuggestthatinadditiontoMR Although Gpr48m/m mice fed a normal chow diet exhibited mutations, the factors that affect MR expression could be an- aldosterone resistance, we did not find any abnormalities in other cause of adPHA1. plasma Na+ level or PRA. This can also be observed in some The neural MR expression is tightly controlled by various adPHA1 carriers. The first index case of adPHA1 showed hormone stimuli,37,38 but the factors regulating renal MR ex- markedly elevated serum aldosterone levels and urinary Na+ pression are unknown. Herein, we identified a direct role for excretion but normal serum Na+ and PRA at 28 years of age.9

286 Journal of the American Society of Nephrology J Am Soc Nephrol 23: 281–293, 2012 www.jasn.org BASIC RESEARCH

Figure 6. Morphology and gene expression in the kidney of Gpr48m/m mice. (A) Relative expression pattern of Gpr48 in different tissues of C57BL/6 mice. (B and C) Kidney and body sizes of wild-type and Gpr48m/m mice (scale bar, 1 cm). (D) Representative hematoxylin and eosin staining of renal cortex and medulla in Gpr48m/m mice showed no defects at age 16 weeks (scale bar, 20 mm). (E–G) Relative mRNA and protein expression levels of genes involved in water and electrolyte reabsorption or excretion in both genotypes (n=8). (H) Representative immunohistochemical staining for MR and Na/K-ATPase 1a in cortical and medullary tubules of wild-type and Gpr48m/m mice (scale bar, 50 mm). White bar, wild-type mice (Gpr48+/+); black bar, Gpr48-mutant mice (Gpr48m/m). Error bars represent SEM. *P,0.05. AQP, aquaporin; ATPase, adenosine triphosphatase; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; GR, glucocorti- coid receptor; ROMK1, inwardly rectifying K+ channel; NHE3, Na(+)/H(+) exchanger 3; NCC, thiazide-sensitive Na-Cl cotransporter; NKCC2, Na-K-2Cl cotransporter.

In addition, Geller and colleagues have prospectively screened more sensitive serum electrolyte marker for mild adPHA1 individuals in two large Spanish pedigrees for various clinical animal models39–43 and adult patients with adPHA1.9,10,32 indices, and only elevated aldosterone was seen in adPHA1 The normal plasma Na+ level in Gpr48m/m mice on normal carriers with the R537X mutation in the MR gene.32 On the chow may be attributable to increased food intake, which has basis of our results and those of previous studies, we postulate been observed in rescued MR-null mice with NaCl substitu- that hyperkalemia, rather than hyponatremia, might be a tions;35 the normal Na+ is also consistent with the observation

J Am Soc Nephrol 23: 281–293, 2012 GPR48 in Aldosterone Resistance 287 BASIC RESEARCH www.jasn.org

Figure 7. GPR48 regulates MR expression via the cAMP/PKA pathway. (A) Morphology of MEFs derived from wild-type and Gpr48m/m embryos (scale bars, 100 mm). (B) Reduced MR protein in MEFGpr48m/m. (C) Decreased MR promoter transcriptional activity in MEFGpr48m/m. GAPDH, glyceraldehyde 3-phosphate dehydrogenase. (D) MR promoter transcriptional activity induced by hGPR48 overexpression (0.2, 0.5, 1.0, or 1.5 mg) in a dose-dependent pattern in HEK293 cells. (E) Forskolin (FSK, 10 mM) increased MR tran- scriptional activity in MEFGpr48m/m. (F and G) MR promoter transcriptional activity was induced by FSK plus 3-isobutyl-1-methyl-xanthine (IBMX) (5/125, 10/250, 15/375, 20/500, or 25/625 mM) treatment for 8 hours (F) and inhibited by H89 (5, 10, 20, 30, or 50 mM) treatment for 24 hours (G) in HEK293 cells in a dose-dependent pattern. (H) MR promoter transcriptional activity induced by hGPR48 over- expression was abrogated by H89. HEK293 cells were co-transfected with the MR promoter luciferase construct and pRL-TK Renilla luciferase construct, followed by treatment with FSK plus IBMX (20/500 mM) for 8 hours, or co-transfected with the hGPR48-expressing construct with or without 20 mM H89 treatment for 24 hours. In I, the classic CRE reporter was used as a positive control, which was subjected to the same treatments described for H. Each experiment was performed at least three times. Error bars represent SEM. that the symptoms of patients with adPHA1 are alleviated with survive after birth, with higher lethality than our strains, and age because of higher dietary salt intake from food.6 Expectedly, whose only surviving mouse showed dramatic morphologic when challenged with the low-sodium diet, Gpr48m/m mice defects in the kidney. In our study and Mazerbourg and col- displayed markedly decreased plasma Na+ concentration leagues’ study,23,27 Gpr48 gene expression was abolished by a compared with wild-type mice. Consequently, renal renin ex- secretory trap approach, whereas in Kato and coworkers’ fl fl pression and PRA were also augmented. These changes are study, Gpr48 was deleted by intercrossing Gpr48 oxed/ oxed consistent with the phenotypes of MRAQP2-cre mice deprived mice with CAG-Cre “transgenic general deleter” mice.44 The of salt.39 These findings confirm the role of GPR48 in the path- gene-trap method can lead to residual Gpr48 transcripts,22,45 ogenesis of adPHA1 and the regulation of electrolyte and hor- which was also seen in Gpr48m/m mice in our study, with ap- mone homeostasis. proximately 10% residual Gpr48 transcripts in the kidney. This low residual expression may prevent Gpr48m/m mice Unchanged Renal Morphology with Partial from dying and displaying severe morphologic defects in the Urine-Concentrating Defects in Gpr48m/m Mice kidney. We thus speculate that different gene-targeting strat- Gpr48m/m mice displayed no apparent pathologic defects in egies and the residual Gpr48 transcripts may have caused the the kidney except for the smaller size. These observations are discrepancies observed in different strains of Gpr48m/m mice. consistent with those in the previous study by Mazerbourg and Gpr48m/m mice displayed a partial urine-concentrating de- colleagues.23 However, the phenotypes reported by us and fect, which can also be observed in aldosterone deficiency those authors are different from the findings of Kato and as- conditions (e.g., aldosterone synthase knockout mice,46 adre- sociates,44 who found that Gpr48-knockout mice can barely nalectomized rats,47 and patients with chronic adrenal

288 Journal of the American Society of Nephrology J Am Soc Nephrol 23: 281–293, 2012 www.jasn.org BASIC RESEARCH

Figure 8. A functional CRE-like element is located in the mouse MR promoter. (A) HEK293 cells transfected with the pGL4.15-control, +176 trunMR, 2174 trunMR, 2321 trunMR, 21165 mutMR (CRE mutant), or 21165 wtMR promoter luciferase reporter construct were treated with H89 (20 mM) or an equivalent volume of H2O for 24 hours before measurement. The mouse MR transcription start site is indicated by +1. (B) mutMR promoter transcription failed to be activated by hGPR48. ***P,0.001. (C) Chromatin immunoprecipitation as- say. MEFGpr48+/+ and MEFGpr48m/m were subjected to fixation, lysis, sonication, and incubation with pCREB antibody. The immunoprecipitated DNA was amplified with specific primers. Primer 1 included the CRE region, whereas primer 2 was located in exon 9 as a negative control. IgG was used as a negative control for pCREB antibody. The relative binding level is indicated as the percentage of input DNA. (D) EMSA to analyze the binding activity of CREB to the mMR promoter in vitro. Lanes 1–4, nuclear extracts from mouse Leydig cells showed binding to the mMR-CRE probe; lanes 5 and 6, anti-CREB antibody (Ab) shifted the binding complex of the mMR-CRE probe; lanes 7–10, positive control with an mSom-CRE probe. Each experiment was performed at least three times. (E) Model illustrating how initial GPR48 activation promotes MR transcription in the kidney, which induces the expression of downstream targets (e.g., aEnaC) and leads to more salt reabsorp- tion and thus hypertension, theoretically. Conversely, inactivation or loss of GPR48 protein could lead to reduced expression of MR and its targets, and, thus, aldosterone resistance occurs. FSK, forskolin; mMR, mouse MR; mSom, mouse somatostatin. Error bars represent SEM.

J Am Soc Nephrol 23: 281–293, 2012 GPR48 in Aldosterone Resistance 289 BASIC RESEARCH www.jasn.org insufficiency48). Moreover, administration of the MR antago- autoanalyzer (Beckman). Blood gases and plasma K+ were measured nist spironolactone in patients can also increase diluted urine within 2 minutes of blood collection with a portable blood gas ana- production.49 These findings demonstrate that aldosterone lyzer (Abbott) using the EG7+ cartridge (i-STAT Corp.). signaling defects could be one origin of urine-concentrating RIAs for plasma renin activity (Immunotech) and plasma and defect. As a result of aldosterone resistance, Gpr48m/m mice urine aldosterone (Diagnostic Systems Laboratories) were performed showed reduced expression of aENaC and Na/K-ATP 1a and using appropriate kits according to the manufacturers’ protocols. thus impaired Na+ reabsorption in aldosterone-sensitive distal For plasma corticosterone measurement, mice were anesthetized nephron, which led to less water reabsorption and hence with 1% pentobarbital sodium (Sigma-Aldrich) before blood collec- more urine production. We further excluded as causative fac- tion between 1600 and 1800 hours (lights on 0700 hours). For the tors any abnormalities in the expression or cellular location fasting condition, mice were deprived of food for 8 hours before of hypothalamic vasopressin or renal vasopressin receptor 2 blood collection. For the restrained condition, mice were placed and aquaporin 2, which are important regulators in water in a clear acrylic tube for 30 minutes before blood collection. The reabsorption.29–31 On the basis of the preceding findings, al- RIA for plasma corticosterone (ICN Biomedicals) was performed by dosterone resistance due to MR deficiency can probably be following standard procedures. considered one major cause, if not the sole cause, of this m/m urine-concentrating defect in Gpr48 mice. BP measurement In summary, Gpr48m/m mice showed aldosterone resistance We measured BP as described elsewhere.50 The operating room was and resembled the manifestations of mild adPHA1, which kept quiet, and the room temperature was 28°C throughout the ex- could be explained by the reduced renal MR expression level. periment. We put the conscious mouse into a properly sized holder This study extends our understanding of GPR48 function in with a darkened nose cone to reduce its stress, and then the holder electrolyte homeostasis as well as BP control, and it provides was inserted into a warming chamber of 38°C. When the mouse was another potential pathogenic mechanism of adPHA1, espe- quiet for a while and its tail was fully extended, BP was measured cially in cases without known MR or ENaC mutations. Further with a tail cuff using a BP analyzer (Softron). Data were collected studies are needed to screen Gpr48 gene mutations in these after mice acclimated to the instrument for 3 days. All measurements patients to confirm the link between this gene and adPHA1. were completed at a fixed time by a skilled technician.

Tissue Collection, Quantitative PCR Analysis, and CONCISE METHODS Immunoblot Assay Mouse kidneys were collected and stored in liquid nitrogen until use. Mice One microgram of RNA was reverse-transcribed into cDNA using The generation of Gpr48m/m mice has been described in our previous the Reverse Transcription System (Promega), and quantitative PCR study.27 Three PCR primers were used for genotyping: the common was performed to verify gene expression in the kidney with a Roche upstream primer A: 59- CCA GTC ACC ACT CTT ACA CAA TGG LightCycler 480 Real-Time PCR System. Primers used in this study CTA AC-39; downstream primer B: 59-ATT CCC GTAGGA GATAGC are shown in Supplemental Table S2. Immunoblot assay was per- GTC CTA G-39; and downstream primer C: 59-GGT CTT TGA GCA formed using rabbit anti-MR (Santa Cruz) or monoclonal mouse CCA GAG GAC-39. Gpr48m/m mice and their wild-type littermates anti-MR (6G1, a gift from Dr. Celso E. Gomez-Sanchez), rabbit anti- were age-matched and produced by intercrosses of male and female GR (Santa Cruz), rabbit anti-aENaC (Santa Cruz), rabbit anti– + heterozygous mutant mice throughout the experiments. All proce- inwardly rectifying K channel (Alomone Labs), mouse monoclonal dures were approved by the Animal Care Committee of Shanghai anti-Na/K-ATPase a1 (Novus Biologicals), mouse monoclonal anti- Jiaotong University School of Medicine. In metabolic balance studies, renin (Santa Cruz), rabbit anti-CYP11B1 (Santa Cruz), goat anti- wild-type and Gpr48m/m mice were fed normal chow (0.25% sodium; HSD3B1 (Santa Cruz), rabbit anti-steroidogenic acute regulatory Shanghai Laboratory Animal Center) or a low-sodium diet (global protein (Santa Cruz), rabbit anti-tubulin (Cell Signaling Technology), sodium content , 0.05%; Shanghai Laboratory Animal Center) in and anti–glyceraldehyde 3-phosphate dehydrogenase antibodies metabolic cages (Tecniplast) to assess their water and salt balance. (Kangchen Bio-tech). For histologic examination, tissue sections were The water deprivation test was conducted for 8 hours, and spot urine stained with hematoxylin and eosin by routine methods. For immuno- samples were collected by bladder massage. Urine osmolality was histochemistry, sections were stained with the indicated antibodies: determined using the freezing point depression method according rabbit anti-aquaporin 2 (Santa Cruz), mouse monoclonal anti-MR, to standard procedures (Fiske Associates). and Na/K-ATPase a1. The Vectastain ABC system (Vector Laboratories) was used to detect the primary antibodies. Hormone and Biochemical Measurements in Plasma and Urine Cell Culture Blood samples of mice fed normal chow or the low-sodium diet were MEFs from both Gpr48m/m mice and their wild-type littermates were drawn through retro-orbital bleeding, and urine samples were col- isolated and established in our laboratory.24 These cells were grown lected with the metabolic cage for biochemical measurements. Plasma in DMEM supplemented with 15% FBS at 37°C with 5% CO2. and urine electrolytes were measured using a clinical biochemical HEK293 and Leydig cells (MA-10), which were used for luciferase

290 Journal of the American Society of Nephrology J Am Soc Nephrol 23: 281–293, 2012 www.jasn.org BASIC RESEARCH assays and EMSA, respectively, were cultured in DMEM supplemented ACKNOWLEDGMENTS with 10% FBS. Chemicals used in this study were all obtained from Sigma and dissolved in appropriate solvents. WearegratefultoImeldaLeeforcriticalrevisionforthemanu- script, Professor Celso E. Gomez-Sanchez (University of Mississippi Plasmids and Luciferase Reporter Assay Medical Center) for kindly providing MR antibodies, and Yan Ge The pcDNA3.1-hGPR48 construct containing cDNA for human and Jia Xu for immunohistochemical staining. GPR48 (hGPR48) has been described elsewhere.27 The CRE reporter This study is supported by the National Natural Science Foun- construct containing four tandem CREs was used to assess cAMP- dation of China (Grants 81030011 and 30725037). dependent gene transcription (Stratagene). The various lengths of mouse MR promoter constructs from –1165 to +235 were ampli- fied from mouse genomic cDNA and inserted into the pGL4- DISCLOSURE Basic vector (Promega). Mutations were introduced into the MR None. promoter using the QuikChange II Site-Directed Mutagenesis Kit (Stratagene). All constructs were verified by DNA sequencing. For the luciferase reporter assay, HEK293 or MEFs seeded in 24-well REFERENCES plates were co-transfected with the indicated MR promoter con- structs, pRL-TK (expressing Renilla luciferase) (Promega) and 1. White PC: Disorders of aldosterone biosynthesis and action. NEngl pcDNA3.1 or pcDNA3.1-hGPR48, followed by lysis and luciferase JMed331: 250–258, 1994 activity measurement using the Dual-Luciferase Reporter Assay 2. Masilamani S, Kim GH, Mitchell C, Wade JB, Knepper MA: Aldosterone- mediated regulation of ENaC alpha, beta, and gamma subunit proteins System (Promega). in rat kidney. J Clin Invest 104: R19–R23, 1999 3. Rossier BC, Pradervand S, Schild L, Hummler E: Epithelial sodium Chromatin Immunoprecipitation Assay channel and the control of sodium balance: Interaction between To investigate the interaction of pCREB with the MR promoter, genetic and environmental factors. Annu Rev Physiol 64: 877–897, chromatin immunoprecipitation assay was performed using a com- 2002 4. Adrogué HJ, Madias NE: Sodium and potassium in the pathogenesis of mercial kit (Upstate Biotechnology) according to the manufacturer’s hypertension. N Engl J Med 356: 1966–1978, 2007 instructions. In brief, wild-type and Gpr48-mutant MEFs were 5. Chang SS, Grunder S, Hanukoglu A, Rösler A, Mathew PM, Hanukoglu I, treated with 1% formaldehyde to cross-link the proteins and DNA, Schild L, Lu Y, Shimkets RA, Nelson-Williams C, Rossier BC, Lifton RP: followed by sonication in an ultrasound bath on ice. The chromatin Mutations in subunits of the epithelial sodium channel cause salt was incubated with rabbit anti-pCREB antibody (Cell Signaling wasting with hyperkalaemic acidosis, pseudohypoaldosteronism type – Technology) or rabbit IgG overnight at 4°C. The immunoprecipi- 1. Nat Genet 12: 248 253, 1996 6. Geller DS, Rodriguez-Soriano J, Vallo Boado A, Schifter S, Bayer M, tated DNA fragments containing or lacking a CRE sequence were Chang SS, Lifton RP: Mutations in the mineralocorticoid receptor gene detected using quantitative PCR with primers shown in Supplemental cause autosomal dominant pseudohypoaldosteronism type I. Nat Table S2. Genet 19: 279–281, 1998 7. Hanukoglu A: Type I pseudohypoaldosteronism includes two clinically EMSA and genetically distinct entities with either renal or multiple target organ defects. JClinEndocrinolMetab73: 936–944, 1991 Nuclear proteins were extracted from mouse Leydig cells using nu- 8. Cheek DB, Perry JW: A salt wasting syndrome in infancy. Arch Dis Child clear and cytoplasmic extraction reagents (Pierce Biotechnology). 33: 252–256, 1958 The DNA probe was derived from a 29-bp DNA fragment cov- 9. Armanini D, Kuhnle U, Strasser T, Dorr H, Butenandt I, Weber PC, ering the CRE region in the mouse MR promoter, and a mutant Stockigt JR, Pearce P, Funder JW: Aldosterone-receptor deficiency in probe contained eight nucleotide substitutions within the CRE pseudohypoaldosteronism. NEnglJMed313: 1178–1181, 1985 sequence (Supplemental Table S2). Oligonucleotides were labeled 10. Pujo L, Fagart J, Gary F, Papadimitriou DT, Claës A, Jeunemaître X, Zennaro MC: Mineralocorticoid receptor mutations are the principal with biotin (Invitrogen), and unlabeled probes were used to compete cause of renal type 1 pseudohypoaldosteronism. Hum Mutat 28: 33– fi for the speci c binding. EMSA was performed using the Lightshift 40, 2007 Chemiluminescent EMSA kit (Pierce Biotechnology). For the 11. Viengchareun S, Le Menuet D, Martinerie L, Munier M, Pascual-Le DNA supershift assay, rabbit anti-CREB antibody (Cell Signaling Tallec L, Lombes M: The mineralocorticoid receptor: insights into its Technology) was incubated on ice before being mixed with the la- molecular and (patho)physiological biology. Nucl Recept Signal 5: e012, 2007. beled probes. As a positive control, the consensus CRE of the mouse 12. Geller DS: Mineralocorticoid resistance. Clin Endocrinol (Oxf) 62: 513– 33 somatostatin promoter was prepared in parallel. Electrophoresis 520, 2005 binding reactions, transfer, cross-linking, and detection were per- 13. Viemann M, Peter M, López-Siguero JP, Simic-Schleicher G, Sippell WG: formed according to standard procedures. Evidence for genetic heterogeneity of pseudohypoaldosteronism type 1: Identification of a novel mutation in the human mineralocor- ticoid receptor in one sporadic case and no mutations in two auto- Statistical Analyses – 6 somal dominant kindreds. J Clin Endocrinol Metab 86: 2056 2059, All data are presented as mean SEM. The data were analyzed using 2001 m/m a two-tailed t test for Gpr48 mice versus wild-type mice. P,0.05 14. Arai K, Tsigos C, Suzuki Y, Listwak S, Zachman K, Zangeneh F, Rapaport R, was considered to indicate a statistically significant difference. Chanoine JP, Chrousos GP: No apparent mineralocorticoid receptor

J Am Soc Nephrol 23: 281–293, 2012 GPR48 in Aldosterone Resistance 291 BASIC RESEARCH www.jasn.org

defect in a series of sporadic cases of pseudohypoaldosteronism. JClin signal peptidase as a cause for familial central diabetes insipidus. JClin Endocrinol Metab 80: 814–817, 1995 Invest 91: 2565–2571, 1993 15. Sartorato P, Lapeyraque AL, Armanini D, Kuhnle U, Khaldi Y, Salomon R, 30. Rosenthal W, Seibold A, Antaramian A, Lonergan M, Arthus MF, Hendy Abadie V, Di Battista E, Naselli A, Racine A, Bosio M, Caprio M, Poulet- GN, Birnbaumer M, Bichet DG: Molecular identification of the gene Young V, Chabrolle JP, Niaudet P, De Gennes C, Lecornec MH, Poisson E, responsible for congenital nephrogenic diabetes insipidus. Nature Fusco AM, Loli P, Lombès M, Zennaro MC: Different inactivating mu- 359: 233–235, 1992 tations of the mineralocorticoid receptor in fourteen families affected 31. Deen PM, Verdijk MA, Knoers NV, Wieringa B, Monnens LA, van Os CH, by type I pseudohypoaldosteronism. J Clin Endocrinol Metab 88: van Oost BA: Requirement of human renal water channel aquaporin-2 2508–2517, 2003 for vasopressin-dependent concentration of urine. Science 264: 92–95, 16. Deppe CE, Heering PJ, Viengchareun S, Grabensee B, Farman N, 1994 Lombès M: Cyclosporine a and FK506 inhibit transcriptional activity 32. Geller DS, Zhang J, Zennaro MC, Vallo-Boado A, Rodriguez-Soriano J, of the human mineralocorticoid receptor: A cell-based model to in- Furu L, Haws R, Metzger D, Botelho B, Karaviti L, Haqq AM, Corey H, vestigate partial aldosterone resistance in kidney transplantation. En- Janssens S, Corvol P, Lifton RP: Autosomal dominant pseudohypoaldo- docrinology 143: 1932–1941, 2002 steronism type 1: Mechanisms, evidence for neonatal lethality, and 17. Heering PJ, Klein-Vehne N, Fehsel K: Decreased mineralocorticoid phenotypic expression in adults. JAmSocNephrol17: 1429–1436, receptor expression in blood cells of kidney transplant recipients un- 2006 dergoing immunosuppressive treatment: cost efficient determination 33. Montminy MR, Sevarino KA, Wagner JA, Mandel G, Goodman RH: by quantitative PCR. J Clin Pathol 57: 33–36, 2004 Identification of a cyclic-AMP-responsive element within the rat so- 18. Heering PJ, Kurschat C, Vo DT, Klein-Vehne N, Fehsel K, Ivens K: matostatin gene. Proc Natl Acad Sci USA 83: 6682–6686, 1986 Aldosterone resistance in kidney transplantation is in part induced by 34. Berger S, Bleich M, Schmid W, Cole TJ, Peters J, Watanabe H, Kriz W, a down-regulation of mineralocorticoid receptor expression. Clin Trans- Warth R, Greger R, Schütz G: Mineralocorticoid receptor knockout plant 18: 186–192, 2004 mice: Pathophysiology of Na+ metabolism. Proc Natl Acad Sci U S A 19. Lim HY, van den Brandt J, Fassnacht M, Allolio B, Herold MJ, Reichardt 95: 9424–9429, 1998 HM: Silencing of the mineralocorticoid receptor by ribonucleic acid 35. Bleich M, Warth R, Schmidt-Hieber M, Schulz-Baldes A, Hasselblatt P, interference in transgenic rats disrupts endocrine homeostasis. Mol Fisch D, Berger S, Kunzelmann K, Kriz W, Schütz G, Greger R: Rescue of Endocrinol 22: 1304–1311, 2008 the mineralocorticoid receptor knock-out mouse. Pflugers Arch 438: 20. de Lau W, Barker N, Low TY, Koo BK, Li VS, Teunissen H, Kujala P, 245–254, 1999 Haegebarth A, Peters PJ, van de Wetering M, Stange DE, van Es JE, 36. Martinerie L, Viengchareun S, Delezoide AL, Jaubert F, Sinico M, Prevot S, Guardavaccaro D, Schasfoort RB, Mohri Y, Nishimori K, Mohammed S, Boileau P, Meduri G, Lombès M: Low renal mineralocorticoid receptor Heck AJ, Clevers H: Lgr5 homologues associate with Wnt receptors expression at birth contributes to partial aldosterone resistance in and mediate R-spondin signalling. Nature 476: 293–297, 2011 neonates. Endocrinology 150: 4414–4424, 2009 21. Carmon KS, Gong X, Lin Q, Thomas A, Liu Q: R-spondins function 37. Munier M, Meduri G, Viengchareun S, Leclerc P, Le Menuet D, Lombès M: as ligands of the orphan receptors LGR4 and LGR5 to regulate Wnt/ Regulation of mineralocorticoid receptor expression during neuronal beta-catenin signaling. Proc Natl Acad Sci U S A 108: 11452–11457, differentiation of murine embryonic stem cells. Endocrinology 151: 2011 2244–2254, 2010 22. Mustata RC, Van Loy T, Lefort A, Libert F, Strollo S, Vassart G, Garcia MI: 38. Castrén M, Patchev VK, Almeida OF, Holsboer F, Trapp T, Castrén E: Lgr4 is required for Paneth cell differentiation and maintenance of Regulation of rat mineralocorticoid receptor expression in neurons by intestinal stem cells ex vivo. EMBO Rep 12: 558–564, 2011 progesterone. Endocrinology 136: 3800–3806, 1995 23. Mazerbourg S, Bouley DM, Sudo S, Klein CA, Zhang JV, Kawamura K, 39. Ronzaud C, Loffing J, Bleich M, Gretz N, Gröne HJ, Schütz G, Berger Goodrich LV, Rayburn H, Tessier-Lavigne M, Hsueh AJ: Leucine-rich S: Impairment of sodium balance in mice deficient in renal principal repeat-containing, G protein-coupled receptor 4 null mice exhibit in- cell mineralocorticoid receptor. JAmSocNephrol18: 1679–1687, trauterine growth retardation associated with embryonic and perinatal 2007 lethality. Mol Endocrinol 18: 2241–2254, 2004 40. Hummler E, Barker P, Talbot C, Wang Q, Verdumo C, Grubb B, Gatzy J, 24. Li XY, Lu Y, Sun HY, Wang JQ, Yang J, Zhang HJ, Fan NG, Xu J, Jiang JJ, Burnier M, Horisberger JD, Beermann F, Boucher R, Rossier BC: A Liu RY, Li DL, Liu MY, Ning G: G protein-coupled receptor 48 upreg- mouse model for the renal salt-wasting syndrome pseudohypoaldo- ulates estrogen receptor alpha expression via cAMP/PKA signaling steronism. Proc Natl Acad Sci U S A 94: 11710–11715, 1997 in the male reproductive tract. Development 137: 151–157, 2010 41. Wulff P, Vallon V, Huang DY, Völkl H, Yu F, Richter K, Jansen M, Schlünz M, 25. Mendive F, Laurent P, Van Schoore G, Skarnes W, Pochet R, Vassart G: Klingel K, Loffing J, Kauselmann G, Bösl MR, Lang F, Kuhl D: Impaired Defective postnatal development of the male reproductive tract in renal Na(+) retention in the sgk1-knockout mouse. J Clin Invest 110: 1263– LGR4 knockout mice. Dev Biol 290: 421–434, 2006 1268, 2002 26. Luo J, Zhou W, Zhou X, Li D, Weng J, Yi Z, Cho SG, Li C, Yi T, Wu X, Li XY, 42. Barker PM, Nguyen MS, Gatzy JT, Grubb B, Norman H, Hummler E, de Crombrugghe B, Höök M, Liu M: Regulation of bone formation and Rossier B, Boucher RC, Koller B: Role of gammaENaC subunit in remodeling by G-protein-coupled receptor 48. Development 136: liquid clearance and electrolyte balance in newborn mice. Insights 2747–2756, 2009 into perinatal adaptation and pseudohypoaldosteronism. J Clin Invest 27. Weng J, Luo J, Cheng X, Jin C, Zhou X, Qu J, Tu L, Ai D, Li D, Wang J, 102: 1634–1640, 1998 Martin JF, Amendt BA, Liu M: Deletion of G protein-coupled receptor 43. Pradervand S, Barker PM, Wang Q, Ernst SA, Beermann F, Grubb BR, 48 leads to ocular anterior segment dysgenesis (ASD) through down- Burnier M, Schmidt A, Bindels RJ, Gatzy JT, Rossier BC, Hummler E: regulation of Pitx2. Proc Natl Acad Sci U S A 105: 6081–6086, 2008 Salt restriction induces pseudohypoaldosteronism type 1 in mice 28. Gao Y, Kitagawa K, Hiramatsu Y, Kikuchi H, Isobe T, Shimada M, expressing low levels of the beta-subunit of the amiloride-sensitive Uchida C, Hattori T, Oda T, Nakayama K, Nakayama KI, Tanaka T, epithelial sodium channel. Proc Natl Acad Sci U S A 96: 1732–1737, Konno H, Kitagawa M: Up-regulation of GPR48 induced by down- 1999 regulation of p27Kip1 enhances carcinoma cell invasiveness and 44. Kato S, Matsubara M, Matsuo T, Mohri Y, Kazama I, Hatano R, Umezawa A, metastasis. Cancer Res 66: 11623– 11631, 2006 Nishimori K: Leucine-rich repeat-containing G protein-coupled receptor-4 29. Ito M, Oiso Y, Murase T, Kondo K, Saito H, Chinzei T, Racchi M, Lively MO: (LGR4, Gpr48) is essential for renal development in mice. Nephron, Possible involvement of inefficient cleavage of preprovasopressin by Exp Nephrol 104: e63–e75, 2006

292 Journal of the American Society of Nephrology J Am Soc Nephrol 23: 281–293, 2012 www.jasn.org BASIC RESEARCH

45. Hoshii T, Takeo T, Nakagata N, Takeya M, Araki K, Yamamura K: LGR4 49. van Vliet AA, Donker AJ, Nauta JJ, Verheugt FW: Spironolactone in regulates the postnatal development and integrity of male reproductive congestive refractory to high-dose loop diuretic and low- tracts in mice. Biol Reprod 76: 303–313, 2007 dose angiotensin-converting enzyme inhibitor. Am J Cardiol 71: 21A– 46. Makhanova N, Lee G, Takahashi N, Sequeira Lopez ML, Gomez RA, Kim 28A, 1993 HS, Smithies O: Kidney function in mice lacking aldosterone. Am 50. Krege JH, Hodgin JB, Hagaman JR, Smithies O: A noninvasive com- J Physiol Renal Physiol 290: F61–F69, 2006 puterized tail-cuff system for measuring blood pressure in mice. Hy- 47. Schwartz MJ, Kokko JP: Urinary concentrating defect of adrenal in- pertension 25: 1111–1115, 1995 sufficiency. Permissive role of adrenal steroids on the hydroosmotic response across the rabbit cortical collecting tubule. JClinInvest66: 234–242, 1980 48. Willson DM, Sunderman FW: Studies in Serum Electrolytes. Xii. The effect of water restriction in a patient with Addison’s disease receiving This article contains supplemental material online at http://jasn.asnjournals. sodium chloride. J Clin Invest 18: 35–43, 1939 org/lookup/suppl/doi:10.1681/ASN.2011040351/-/DCSupplemental.

J Am Soc Nephrol 23: 281–293, 2012 GPR48 in Aldosterone Resistance 293 Supporting Document

Figure S1. (A-C) Polydipsia, polyphagia and polyuria phenotypes were observed in Gpr48m/m female mice (n=7 for 8 weeks, n=3 for 16 weeks). (D) Urine osmolality was declined in Gpr48m/m female mice at age 8 weeks but not 16 weeks (n=7 for 8 weeks, n=3 for 16 weeks). (E-F) Urine osmolality of wild-type and Gpr48m/m mice at age 8 weeks was measured under 8 h water deprivation challenge, but no significant difference in the increase was observed between the two groups (n=5 for Gpr48+/+ and n=6 for Gpr48m/m). White bar, wild-type mice (Gpr48+/+); black bar,

Gpr48 mutant mice (Gpr48m/m).

1

Figure S2. Hematoxylin and eosin (HE) staining of the adrenal gland of 3-week-old and

8-week-old mice at low magnification (×10, A, B, E and F) or at higher magnification (×40, C, D,

G and H). M, adrenal medulla. X, zona X. D, adrenal cortex.

2

Figure S3. Mophological and functional analysis of the adrenal gland in Gpr48m/m mice. (A)

No defects in ultrastructure of zona glomerulosa were observed in Gpr48m/m mice (×5800). (B)

The mRNA levels of the key enzymes in the adrenal gland, including StAR, CYP11A1, HSD3b1,

HSD3b6, CYP21A1, CYP11B1 and CYP11B2 (n=12 for Gpr48+/+ and n=9 for Gpr48m/m). (C)

The protein levels of StAR, HSD3B1 and CYP11B1 were further examined by western blotting (4 adrenal glands were mixed as a pool, n=8). No differences were found between the two groups.

(D-E) No changes in plasma epinephrine and norepinephrine levels of Gpr48m/m mice under anesthesia condition (n=9 for Gpr48+/+ and n=6 for Gpr48m/m).

3

Figure S4. Gpr48 regulates MR expression via the cAMP/PKA pathway. (A) Reduced MR mRNA levels in MEFGpr48m/m. (B) H89 (20 µM) decreased MR transcription activity in

MEFGpr48+/+. (C) MR promoter transcription activity was suppressed by M182 (5-40 µM) in a dose-dependent pattern.

4 Table S1. Blood parameters of female wild-type and Gpr48 mutant mice under normal dietary conditions (n=4).

Gpr48+/+ Gpr48m/m Na (mM) 146.5 ± 1.6 145.5 ± 1.6 K (mM) 5.5 ± 0.5 7.2 ± 0.3a Cl (mM) 118.0 ± 1.1 119.7 ± 0.8 Cr (mM) 13.8 ± 1.1 19.6 ± 3.5 Protein (mg/mL) 55.5 ± 0.7 57.2 ± 0.8 pH (pH units) 7.32 ± 0.05 7.34 ± 0.02

PO2 (mm Hg) 42.3 ± 7.5 39.8 ± 1.4

PCO2 (mm Hg) 45.8 ± 7.5 36.4 ± 1.7

TCO2 (mm Hg) 24.3 ± 1.7 20.8 ± 1.9 HCO3- (mM) 22.8 ± 1.4 20.0 ± 1.8 BE (mM) -3.2 ± 1.1 -6.0 ± 2.1 Hb (g/dL) 14.0 ± 0.2 13.7 ± 1.5

Hct (%PCV) 41.0 ± 0.7 40.3 ± 4.4

a P < 0.05 vs. wild-type mice.

5 Table S2. Primers sequences for qRT-PCR, ChIP and EMSA.

primer Forward Reverse

AVP GCCAGGATGCTCAACACTACG TCTCAGCTCCATGTCAGAGATG AVPR2 TGACCGAGACCCGCTGTTA CGACCCCGTCGTATTAGGG AQP2 ATGTGGGAACTCCGGTCCATA ACGGCAATCTGGAGCACAG GPR48 CCGACTTCGCATTCACCA AGTCCAGAGTCCGCAGCAT STAR CCGGAGCAGAGTGGTGTCA GCCAGTGGATGAAGCACCAT CYP11A1 AAGGTACAGGAGATGCTGCG AGTGTCTCCTTGATGCTGGC HSD3B1 AGCATCCAGACACTCTCATC GGAGCTGGTATGATATAGGGTA HSD3B6 CATCCTTCCACAGTTCTAGC TGGTGTGAGATTAATGTACA CYP21A1 GCTGTGGCTTTCCTGCTTCAC GGCCCAGCTTGAGGTCTAACT CYP11B1 GTGAGCCCATCTTCTGACTTTC CAATGTGTCATGAGTGGTCATAG CYP11B2 GTTTTCCAATGGTCACTCCAG GCTTGCTGCCCCTACAAAC MR GAAGAGCCCCTCTGTTTGCAG TCCTTGAGTGATGGGACTGTG GR AGCTCCCCCTGGTAGAGAC GGTGAAGACGCAGAAACCTTG α-ENaC CCTTCTCCTTGGATAGCCTGG CAGACGGCCATCTTGAGTAGC β-ENaC CGGTCACCGTCTGCAATTC AGATGGTAAAGTTCAGGGTCCT γ-ENaC GTGGCCCTCATTATCTGGCAG GCCTGTTTTGTCTCACTGTCCA NHE3 TGAAAAGCAGGACAAGGAAATCT TTGGCCGCCTTCTTATTCTGG NCC GCAAGGTCGTGAGCTGACT ATGCAACGGATCATCACCCC NKCC2 TGTTAGGTGGCACAGAAGATACC CACGGTTACATTGCTTGTTTGTT ROMK1 TGTAGATGCACAGTCGAGGTT GCTACGACATACCACAGGAGA AQP1 AGGCTTCAATTACCCACTGGA GTGAGCACCGCTGATGTGA AQP3 GCTTTTGGCTTCGCTGTCAC TAGATGGGCAGCTTGATCCAG Na/K-ATPase 1α GGGGTTGGACGAGACAAGTAT CGGCTCAAATCTGTTCCGTAT Renin CTCTCTGGGCACTCTTGTTGC GGGAGGTAAGATTGGTCAAGGA ChIP-MR-promoter CGGCGGGAAGGTAACTCT CTGGCGACCACAGCAATC ChIP-MR-exon 9 GGGGTGTTCGTGAGACT ACAGATGTGGAGGGTGC EMSA-MR-WT GCTCTAGCGTCACCATGCTTTCCCCTTCT AGAAGGGGAAAGCATGGTGACGCTAGAGC EMSA-MR-MUT GCTCcctaacacCCATGCTTTCCCCTTCT AGAAGGGGAAAGCATGGgtgttaggGAGC EMSA-WT-Sst GCCTCCTTGGCTGACGTCAGAGAGAGAG CTCTCTCTCTGACGTCAGCCAAGGAGGC

6