original article http://www.kidney-international.org & 2009 International Society of Nephrology see commentary on page 1253 Overexpression of P450 4F2 in mice increases 20-hydroxyeicosatetraenoic acid production and arterial blood pressure Xiaoliang Liu1,4, Yanyan Zhao1,4, Luzeng Wang2, Xianghong Yang3, Zhihong Zheng2, Yuanyuan Zhang1, Fangjie Chen1 and Hong Liu1

1Department of Medical Genetics, China Medical University, Shenyang, Liaoning, China; 2Department of Laboratory Animal Center, China Medical University, Shenyang, Liaoning, China and 3Department of Pathology, the Second Affiliated Hospital, China Medical University, Shenyang, Liaoning, China

Cytochrome P450 4F2 (CYP4F2) activity is thought to be a The human 4F2 (CYP4F2) gene (MIM factor in the pathogenesis of through its no. 604426) encodes an o-hydroxylase that catalyzes the bioactive metabolite 20-hydroxyeicosatetraenoic acid (AA) to 20-hydroxyeicosatetraenoic acid (20-HETE). We previously found that a gain-in-function (20-HETE).1 Previous studies have demonstrated that CYP4F2 variant in a Chinese cohort was associated with 20-HETE participates in the development of hypertension by elevated urinary 20-HETE and hypertension. To further regulating vascular and renal tubular functions.2–5 On one explore this association we generated a transgenic mouse hand, 20-HETE increases peripheral vascular resistance by model expressing CYP4F2 driven by a modified mouse potently constricting small arteries or by sensitizing the vascular kidney androgen-regulated protein promoter. This smooth muscle cells to constrictor and myogenic stimuli. On heterologous promoter regulated the expression of luciferase the other hand, 20-HETE causes natriuresis by inhibiting and his-tagged CYP4F2 in transfected HEK 293 cells. In the sodium reabsorption. Recently, we found that a functional kidney of transgenic mice, CYP4F2 was localized to renal haplotype of CYP4F2 with increased transcriptional activity was proximal tubule epithelia and was expressed at a higher level associated with elevated urinary 20-HETE and hypertension in than in control mice, leading to increased urinary a Chinese population.6 A similar result was reported that the 20-HETE excretion. Assessment of CYP4F2 activity by an variant V433M in CYP4F2 was associated with the increase of arachidonic acid hydroxylation assay showed that 20-HETE 20-HETE excretion and systolic blood pressure (SBP) in a white production was significantly higher in kidney microsomes of cohort.7 However, a different result showed that the variant transgenic mice compared to control mice, as was their F434S in CYP4A11, another human 20-HETE synthase gene, systolic blood pressure. There was a positive correlation of with reduced catalytic activity was associated with hypertension blood pressure with urinary 20-HETE levels. Our results show in white populations.8 Therefore, CYP4F2 transgenic approach that increased expression of CYP4F2 in mice enhanced would be a pivotal and irreplaceable strategy to clarify the 20-HETE production and elevated blood pressure. correlation between 20-HETE and blood pressure. Kidney International (2009) 75, 1288–1296; doi:10.1038/ki.2009.67; Heterologous promoters have been utilized to target published online 11 March 2009 transgene expression to expected tissues at controllable level KEYWORDS: cytochrome P450 4F2; 20-hydroxyeicosatetraenoic acid; in transgenic models. Native CYP4F2 is expressed at a high hypertension; transgenic mice level in the proximal tubules of the human kidney where it accounts for the majority of renal 20-HETE production.1 Mouse kidney androgen-regulated protein (KAP) is one of the most abundant proteins expressed in the proximal tubules in response to different hormones such as androgen and estrogen.9 Experimental evidence from many previous transgenic models has shown that a 1542-bp fragment of the KAP promoter could successfully regulate transgene expres- 10–14 Correspondence: Yanyan Zhao, Department of Medical Genetics, China sion in the kidney. Moreover, Soler’s recent study has Medical University, No 92 Bei Er Road, Shenyang 110001, China. documented that a 224-bp truncated fragment of the KAP E-mail: [email protected] promoter was sufficient to drive androgen-dependent 15 4These authors contributed equally to this work. transactivation in the kidney. Herein, we chose the Received 7 August 2008; revised 7 January 2009; accepted 27 January truncated KAP promoter to highlight kidney-dominant and 2009; published online 11 March 2009 inducible expression of CYP4F2 in transgenic mice.

1288 Kidney International (2009) 75, 1288–1296 X Liu et al.: CYP4F2 transgenic mice original article

This study aims to develop a transgenic mouse model 2.6-fold of cross-reacting signal over the pGL3-enhancer expressing CYP4F2 under the control of the heterologous control using anti-CYP4F2 antibody, wherein endogenous KAP promoter, and to shed light on the impact of 20-HETE CYP4F2 was included. The recombinant CYP4F2 expression on blood pressure through CYP4F2 gain-in-function. was further verified by anti-his antibody. These results confirmed the 224-bp KAP promoter activity in driving the RESULTS expression of the heterologous CYP4F2 gene in HEK293 cells. KAP promoter activity in HEK293 cells To identify the KAP promoter activity, we amplified the 224- Generation of CYP4F2 transgenic mice bp fragment from 50-flanking of the KAP gene to construct The linearized pKAP-CYP4F2/his (Figure 2a) was micro- pKAP-LUC and pKAP-CYP4F2/his expression vectors. The injected into FVB/N mouse to generate transgenic mice. We expression of luciferase and his-tagged CYP4F2 was detected routinely screened the mice by PCR (data not shown) for the after transient transfection into HEK293 cells. As illustrated integration of CYP4F2 using specific primers, and further in Figure 1a, the luciferase activity of cells transfected with evaluated the presence of the transgene by Southern blot the pKAP-LUC, after being normalized to Renilla luciferase analysis (Figure 2b). Three transgenic founders, F0-6 activity, was threefold over those with the pGL3-enhancer (female), F0-16 (female), and F0-56 (male), were successfully control. Western blot analysis for CYP4F2 expression in cells bred to establish independent transgenic mouse lines. There transfected with the pKAP-CYP4F2/his (Figure 1b) showed was no difference in the number of offsprings between male and female, indicating the CYP4F2 gene may insert into the autosomes of the three founders. Line F0-16 seemed to be 3 less reproducible and thereby the number of positive 2.5 transgenic pups per brood was fewer than the others.

2 CYP4F2 expression profile in transgenic mice by Western blot analysis 1.5 CYP4F2 was expressed in all kidney samples of the three 1 transgenic lines, among which line F0-16 exhibited the highest expression level (Figure 3a). Detailed expression 0.5

Relative luciferase activity luciferase Relative profile of female and male transgenic offspring in line F0-16 0 was illustrated in Figure 3b and c, respectively, demonstrating pGL3-enhancer pKAP-LUC a wide expression spectrum of CYP4F2. Densitometry scanning (Figure 3d) revealed that the highest expression (–) (+) (–) (+) level was in the kidney of male mice, about 2.5-fold over CYP4F2 female littermates. In addition, the extra-renal expression of CYP4F2 was prominent in reproductive organs. These results GAPDH

0.6 KpnI Bgl II Pst I Xba I BamHI 0.5 0.4 P1 P2 0.3 KAP promoter CYP4F2 cDNA 6×his tag SV40 enhancer 0.2

Densitometry F0-6 F0-16 WT F0-56 0.1 87654321 91011 0 (–) (+) (–) (+) 5.3 kb

Anti-CYP4F2 Anti-his 4.1 kb Figure 1 | KAP promoter activity in transfected HEK293 cells. 3.7 kb (a) Relative luciferase activity assay. Using the Dual Luciferase Reporter assay system, the relative luciferase activity is presented as the ratio of luciferase activity of firefly to that of Renilla as Figure 2 | Schematic map of the linear pKAP-CYP4F2/his and internal control. (b) Western blot analysis of CYP4F2 expression. Southern blot analysis of CYP4F2 in transgenic mice.(a) The CYP4F2 level was determined by anti-CYP4F2 antibody and anti- fragment for microinjection contained the 224-bp KAP promoter, his antibody in cells transfected with the pKAP-CYP4F2/his ( þ ) CYP4F2 cDNA, 6 Â his tag, and SV40 enhancer. Arrow-indicated P1 and the pGL3-enhancer (À). The 53-kD band represented CYP4F2 and P2 are the primers specific for CYP4F2 to screen positive mice expression, and the 36-kD GAPDH was shown as internal control. by PCR. (b) Southern blot analysis. The hybridization bands of tail Data are presented as mean±s.d. All experiments were DNA with specific CYP4F2 probe confirmed the integration of the performed three times independently. transgene in the three transgenic lines. WT: wild-type.

Kidney International (2009) 75, 1288–1296 1289 original article X Liu et al.: CYP4F2 transgenic mice

1.2 1 0.8 CYP4F2 0.6 0.4 GAPDH Densitometry 0.2 WT F0-6 F0-16 F0-56 0 WT F0-6 F0-16 F0-56

Kidney Liver Brain Heart Lung Spleen Pancreas Stomach Intestine Uterus Ovary Muscle Bladder

CYP4F2

GAPDH

Kidney Liver Brain Heart Lung Spleen Pancreas Stomach Intestine Testis Epididymis Muscle Bladder

CYP4F2

GAPDH

3.5 Female 3 Male 2.5 2 1.5 1 Densitometry 0.5 0 Kidney Liver Brain Heart Lung Spleen Pancreas Stomach Intestine Uterus/ Ovary/ Muscle Bladder testis epididymis Figure 3 | CYP4F2 expression by Western blot analysis. 53-kD CYP4F2 was detected using anti-his antibody, and 36-kD GAPDH is shown as an internal control. (a) CYP4F2 expression in the kidney of three transgenic lines and wild-type (WT) control. (b) CYP4F2 expression profile of female offspring in line F0-16. (c) CYP4F2 expression profile of male offspring in line F0-16. (d) Scanning densitometry of b and c. Data are presented as mean±s.d. All experiments were performed three times independently. indicate that human CYP4F2 was successfully expressed in generated an AA hydroxylation reaction system in vitro. the kidney of transgenic mice, and the expression was Renal microsomes isolated from transgenic line F0-16 and influenced by sexual hormones, especially by androgen. wild-type mice were incubated with AA. As illustrated in Figure 5a, the 20-HETE production was significantly CYP4F2 expression by immunohistochemistry higher in the transgenic group than in the wild-type group Immunohistochemistry was performed in the kidney, liver, (660 (408–1073) versus 298 (224–398) pmol/min/mg micro- brain, and uterus from line F0-16 using anti-his antibody. some, P ¼ 0.003). Moreover, Western blot analysis for The immunostaining signal was abundant in the kidney and CYP4F2 in the renal microsomes demonstrated that the uterus, moderate in the liver, and not in the brain (Figure 4), cross-reacting signal of transgenic mice was 3.5-fold over which was consistent with the results of Western blot analysis. wild-type controls (Figure 5b). From these results, we In the kidney, positive staining was predominantly localized conclude that the overexpressed CYP4F2 in the renal in the proximal tubules, indicating a high level of CYP4F2 microsomes contributed to increased 20-HETE production expression; and there was no immunostaining signal in the in transgenic mice. glomerulus. The liver sections exhibited positive immunos- taining in some hepatocytes of the liver lobule. An intense Increased urinary 20-HETE excretion in transgenic mice staining could be observed throughout the uterus sections The urinary 20-HETE level of transgenic mice was deter- apart from the endomembrane. mined to evaluate the activity of CYP4F2 in vivo. 20-HETE ELISA assay was performed for the measurement according 16 Increased x-hydroxylation activity of transgenic renal to Grates’s report. As shown in Table 1, both female and microsomes male transgenic mice from all the independent lines had To address whether the overexpression of renal CYP4F2 was increased urinary 20-HETE compared with wild-type con- in correspondence to increased 20-HETE synthesis, we trols (Po0.05). The urinary 20-HETE of male mice from line

1290 Kidney International (2009) 75, 1288–1296 X Liu et al.: CYP4F2 transgenic mice original article

P = 0.003 700 600 500 400 300 200

20-HETE production 100

(pmol/min per mg microsome) 0 WT TG

WT TG WT TG CYP4F2

GAPDH

0.8

0.6

0.4

0.2 Densitometry 0 WT TG WT TG

Anti-CYP4F2 Anti-his Figure 5 | The x-hydroxylation analysis of renal microsomes. (a) Increased 20-HETE production by transgenic renal microsomes. In the AA o-hydroxylation reactions, the 20-HETE production was significantly higher in the transgenic (TG) group (660 (408–1073)) than in the wild-type (WT) group (298 (224–398)). (b)CYP4F2 expression in renal microsomes by Western blot analysis. The 53-kD band of CYP4F2 was detected by anti-CYP4F2 antibody and anti-his antibody, and the 36-kD GAPDH was shown as an internal control. All experiments were performed three times independently.

Table 1 | Urinary 20-HETE of CYP4F2 transgenic mice Group n 20-HETE (ng/ml) P-value WT Female 6 8.99 (3.17–25.47) Male 7 12.45 (7.88–19.66) Figure 4 | CYP4F2 expression by immunohistochemistry. (b, d, f and h) Kidney, liver, brain, and uterus of transgenic (TG) mice, Line F0-16 and (a, c, e and g) wild-type (WT) controls, respectively. Anti-his Female 6 40.23 (22.43–72.16) 0.0001 antibody was used as primary antibody and positive stainings are Male 6 73.28 (42.23–127.18) 0.000005 indicated by arrows, using Diaminobenzidin as the chromagen. Micrographs were taken at  200 magnification. The bar Line F0-56 represents 100 mm. (1) glomerulus; (2) proximal tubule; (3) lobular Female 8 33.27 (20.67–53.53) 0.0003 central vein; (4) hepatocyte of the lobule; (5) endometrium. Male 7 35.55 (23.95–52.77) 0.003

Line F0-6 Female 7 23.91 (14.24–40.13) 0.007 Male 10 41.36 (25.40–67.31) 0.0003 WT, wild-type. F0-16 was the highest of all. Male transgenic mice had higher As urinary 20-HETE was not normally distributed, it underwent logarithmic urinary 20-HETE than their female littermates (P ¼ 0.046), transformation for analysis of variance and was shown as geometric mean (95% CI). which was in accordance with higher renal CYP4F2 expression in male than in female mice. These results demonstrate that the overexpressed CYP4F2 in transgenic Elevated blood pressure in transgenic mice mice functioned and lead to an augment in urinary 20-HETE To determine the effect of 20-HETE overproduction on blood excretion. pressure, we measured the mouse blood pressure by external

Kidney International (2009) 75, 1288–1296 1291 original article X Liu et al.: CYP4F2 transgenic mice

tail pulse detection. As depicted in Table 2, the SBP of higher level of SBP than female transgenic littermates transgenic mice was elevated with variable magnitude (P ¼ 0.001). The changes in SBP basically paralleled the compared with wild-type controls. The most conspicuous changes in urinary 20-HETE shown in Table 1, and their increment of SBP was in line F0-16. Figure 6a better shows positive correlation was shown in Figure 6b with statistical the comparison. Male transgenic mice had significantly significance (R ¼ 0.872, P ¼ 0.005), indicating that the 20-HETE overproduction contributed to elevated blood pressure in transgenic mice. To record the development of Table 2 | SBP of CYP4F2 transgenic mice hypertension in transgenic mice, we performed time course Group n SBP (mm Hg) P-value measurement and found that the blood pressure was elevated since early lives and was maintained at a high level throughout WT Female 12 109.47±7.79 late lives (Figure 7). In addition, the hypertensive phenotype of Male 10 117.06±8.11 transgenic mice was further verified by intra-arterial blood pressure measurement (Supplementary Information). Line F0-16 Female 8 123.30±4.33 0.00008 Male 7 135.49±3.02 0.000002 Blood biochemical assays Finally, we performed blood biochemical assays including Line F0-56 serum creatinine, blood urea nitrogen (BUN), and plasma Female 11 123.18±6.95 0.00002 aldosterone determination to exclude some other factors Male 11 129.36±5.59 0.0002 associated with the development of hypertension. As shown Line F0-6 in Table 3, all these parameters were at equal levels between Female 6 112.83±7.08 NS transgenic mice and wild-type controls. Therefore, the Male 9 122.56±10.93 NS development of hypertension in our model was not related WT, wild-type; NS, not significant. to kidney failure or hyperaldosteronism, which further Data are presented as mean±s.d. confirmed the causative role of 20-HETE in the development of hypertension.

150 Female Male 140 * * * 130 * 140 TG 120 WT 130 SBP (mm Hg) 110 120 100 WT F0-16 F0-56 F0-6 SBP (mm Hg) 110 140 R = 0.872 P = 0.005 100 8 1216202428323640 130 Time (week) Figure 7 |Time course blood pressure measurement. The 120 blood pressure of transgenic (TG) mice in line F0-16 (n ¼ 12) and wild-type (WT) controls (n ¼ 9) aged 8 to 40 weeks was measured at four-week intervals to record the development of hypertension. SBP (mm Hg) 110 The SBP of TG mice was elevated in early life and maintained at higher levels than WT controls throughout late life.

100 0.8 1 1.2 1.4 1.6 1.8 2 Logarithmic urinary 20-HETE (ng/ml) Table 3 | Blood biochemical assays of CYP4F2 transgenic mice Figure 6 | SBP of CYP4F2 transgenic mice. (a) Comparison of Parameter WT (n=6) TG (n=5) P-value SBP in CYP4F2 transgenic mice. The SBPs of transgenic mice from the three transgenic lines were compared with the wild-type (WT) Creatinine (mg per 100 ml) 0.31±0.16 0.39±0.13 0.372 controls. Data are presented as mean±s.d. *Po0.001 in the BUN (mg per 100 ml) 28.54±3.94 27.78±0.99 0.664 analysis of variance. (b) Correlation of logarithmic urinary 20-HETE Aldosterone (pg per 100 ml) 335.60±92.41 321.47±85.12 0.800 with SBP. Pearson’s correlation coefficient and P-value are TG, transgenic; WT, wild-type. indicated on the graph. Data are presented as mean±s.d.

1292 Kidney International (2009) 75, 1288–1296 X Liu et al.: CYP4F2 transgenic mice original article

DISCUSSION results supported that a prohypertensive effect of 20-HETE This study offers a novel transgenic mouse model over- was in dominance. After excluding some other factors expressing human CYP4F2 under the control of the associated with the development of hypertension, such as heterologous KAP promoter. CYP4F2 was proved to be kidney failure and hyperaldosteronism, we deduce 20-HETE overexpressed, leading to an augment of 20-HETE produc- derived from the overexpressed CYP4F2 in the tubular system tion, which, in consequence, contributed to elevated blood of nephron may diffuse into the local and systemic pressure in CYP4F2 transgenic mice. circulation where the vasoactive molecule elicited powerful Native CYP4F2 is predominantly expressed in the kidney vasoconstriction action in the transgenic mice. In addition, and plays a pivotal role in the regulation of blood pressure 20-HETE synthesized in extra-renal organs may also increase through the vasoactive and natriuretic effects of 20-HETE the peripheral vascular tone. The prohypertensive effect of metabolite.2–5 Based on the anatomical and functional 20-HETE was evident in the early lives of transgenic mice and characters of CYP4F2 in the kidney, the heterologous KAP lasted throughout their late lives. As to the antihypertensive promoter was used to control CYP4F2 expression in a action of 20-HETE by inhibiting salt reabsorption, we kidney-dominant pattern in transgenic mice. CYP4F2 was presumed it was not overt enough to override the successfully expressed in the kidney of transgenic mice, and prohypertensive effect. Previous investigations indicated that its renal localization was quite similar to the spatial 20-HETE plays an important role in the renal adaptation to disposition of native CYP4F2 in the human kidney. To assess elevated Na þ intake: Laffer et al.19 demonstrated that urinary the activity of the overexpressed CYP4F2, we performed the 20-HETE excretion was 66% higher during salt loading than hydroxylation assay in isolated renal microsomes. We found during salt depletion in humans; another study also showed 20-HETE production was significantly higher in samples that urinary 20-HETE excretion increased by 20% when rats isolated from transgenic mice than in those from control were fed a high salt diet.20 In this regard, we hypothesized mice, and the elevated 20-HETE production was in that our transgenic mice fed with normal salt diet did not respondence to the increased CYP4F2 expression level in elicit strong ion reabsorption inhibition. The prohypertensive isolated renal microsomes from transgenic mice, indicating effect of 20-HETE in our transgenic model through CYP4F2 the overexpressed CYP4F2 resulted in a high level of AA gain-in-function was in agreement with our recent report o-hydroxylation into 20-HETE. The elevated urinary that a functional haplotype of CYP4F2 with increased 20-HETE excretion could further substantiate this notion in transcriptional activity was associated with elevated urinary vivo. In addition, the CYP4F2 expressed in other extra-renal 20-HETE and enhanced susceptibility of hypertension,6 and tissues may also contribute to some extent to the increased was also in agreement with Ward et al.’s report7 that the 20-HETE level in the systemic circulation, which in return variant V433M in CYP4F2 was associated with the increase exacerbated urinary 20-HETE excretion. The quantification of 20-HETE excretion and SBP. Many previous animal of 20-HETE in this study was performed using Grates’s models also supported that increased renal CYP activity and 20-HETE ELISA method,16 by which the 20-HETE concentra- 20-HETE promoted hypertension: the delivery of CYP4A1 tion was comparatively higher than that by GC-MS techni- cDNA into Sprague–Dawley rat increased blood pressure, and ques17 and was somewhat lower than that by fluorescent HPLC the transfer of antisense CYP4A1 cDNA into spontaneously analysis.18 However, the comparison between transgenic and hypertensive rat attenuated the development of hyperten- their wild-type littermates under the same background was sion;21 other researches on spontaneously hypertensive rat valid to reflect the effects of the overexpressed CYP4F2. The reported that the CYP4A2 gene was overexpressed in the aforementioned data suggest that the transgenic model with an kidney,22 with increased renal 20-HETE, and the use of 20- overexpression of CYP4F2 increased in 20-HETE production, HETE synthesis inhibitor attenuated hypertension.23 and could be used for clarifying the effect of CYP4F2 on blood In addition, the CYP4F2 expression in our transgenic mice pressure through the 20-HETE pathway. was found in a wide spectrum and was prominent in 20-HETE has both prohypertensive and antihypertensive reproductive organs other than in the kidney. The extra-renal properties.2–5 20-HETE promotes hypertension by vasocon- transgene expression was also found in other transgenic striction in many microcirculatory districts that regulates the models controlled by the 1542-bp KAP promoter: Ding’s vascular tone and homeostasis. On the contrary, 20-HETE investigation of KAP-hATG transgenic mice revealed that attenuates hypertension by the inhibition of ion reabsorption human angiotensinogin was restricted to the kidney and that regulates the salt balance. In our study, two lines of epididymis of a male line10 and was induced by testosterone transgenic mice (F0-16 and F0-56) had significantly higher treatment in female transgenic mice;11 human renin was SBP than the control. Transgenic line F0-16 with the highest expressed in the kidney of all KAP-hREN transgenic lines and renal CYP4F2 expression and urinary 20-HETE had the was variable in other tissues;12 the reporter luciferase gene highest SBP level, which could likely account for the lowest displayed high expression levels in the kidney and reproduc- reproductivity in this line; whereas line F0-6 with relatively tive organs of KAP-luciferase transgenic mice.13 The CYP4F2 low level of CYP4F2 and urinary 20-HETE failed to change expression was higher and wider in our model than in those blood pressure. The average SBP positively correlated with previous models. This may be explained by the truncation of urinary 20-HETE level with statistical significance. Thus, our the 224-bp promoter, or by the potential role of the

Kidney International (2009) 75, 1288–1296 1293 original article X Liu et al.: CYP4F2 transgenic mice

introduced SV40 enhancer in the pKAP-CYP4F2/his con- (an introduced 6 Â his tag was underlined) after reverse transcription struct as enhancer element was suggested to be required for of human liver RNA. The amplified CYP4F2 cDNAwasclonedinto the KAP promoter activity.24 As to the prominent expression the PstI/XbaIsitesofthepKAP-LUC to generate pKAP-CYP4F2/his of CYP4F2 in reproductive organs of the transgenic mice, it construct. All constructs were confirmed with sequencing. may be explained that the truncated KAP promoter was capable of driving sex hormone responsive expression as Transient transfection and luciferase assay native KAP is expressed in response to different hormones, Kidney-derived human embryo kidney 293 (HEK293) cells were such as androgen and estrogen.25 cultured and co-transfected with the pKAP-LUC construct and the pRL-TK plasmid (Promega), encoding Renilla luciferase as internal Furthermore, we also found sex-dimorphous phenomena control for luciferase assay, or transfected with the pKAP-CYP4F2/ in CYP4F2 expression, renal 20-HETE formation and SBP in his construct for Western blot analysis, using lipofectamine 2000 our transgenic model. The transgene expression in the kidney reagent (Invitrogen, Carlsbad, CA, USA). Cells were harvested 24 h was higher in male than in female mice; urinary 20-HETE after transfection. Luciferase assay was performed using Dual was higher in male transgenic mice than in female Luciferase Reporter assay system (Promega) and a Lumat LB 9507 littermates; and coincidently the average SBP was signifi- luminometer (Bethold Technologies, Bad Wildbad, Germany). cantly higher in male transgenic mice than in female littermates. As to the nontransgenic mice, 20-HETE and Generation of CYP4F2 transgenic mice SBP had male mice dominance as well. The disparities The pKAP-CYP4F2/his plasmid was digested by KpnI and BamHI to between gender in our study may be explained from two a linear DNA fragment containing the 224-bp KAP promoter, aspects: one was that the 224-bp KAP promoter regulated CYP4F2 cDNA and SV40 enhancer, and microinjected into FVB/N CYP4F2 expression in an androgen-responsive way; the other mouse one-cell fertilized embryos to generate CYP4F2 transgenic was that androgen-mediated 20-HETE synthesis directly. founders (Figure 3a). Mice were fed with standard mouse chow Several animal models have suggested a link between and water ad libitum, and bred in a 12:12 h light–dark cycle system. 20-HETE production and androgen: Holla et al.17 found Transgenic mice of 14- to 20-week old were used for data collection unless noted elsewhere, with age- and sex-matched wild-type that androgen can induce the expression of mice cyp4a12, mice as controls. All experiments conformed with the Guide for 20-HETE formation and hypertension; Sprague–Dawley rats the Care and Use of Laboratory Animals published by the US showed an increase in the renal 20-HETE and SBP level after National Institutes of Health (NIH Publication No. 85-23, revised 26 androgen induction. Our previous population study 1996). including 132 participants had shown that the urinary Transgenic mice were identified by PCR and Southern blot. PCR 20-HETE level in men was slightly higher than in women was performed using DNA isolated from tail biopsies with primers (P ¼ 0.041).6 Another research involving 66 subjects docu- 50-TGGACCTACGCCTTCTATGA-30 and 50-CCACTTGTCACCAG mented that men had significantly higher levels of urinary CACTCA-30 specific to human CYP4F2 cDNA. For Southern blot, 20-HETE than women (P ¼ 0.007).27 In addition, a recent the CYP4F2 cDNA sequence was isolated from the pKAP-CYP4F2/ 32 study also reported men had significantly higher levels his by digestion with PstI and XbaI, labeled by [a- P]dCTP (Furui of 20-HETE excretion when compared with women.7 Taken Biotechnology, Beijing, China) with random primer DNA labeling kit (TaKaRa Biotech, Japan), and then hybridized with genomic together, our findings suggested that androgen partially DNA digested with HindIII. contributed to the overexpressed CYP4F2, increased 20-HETE production and elevated blood pressure. In summary, we established for the first time a CYP4F2 Western blot analysis transgenic mouse model. The overexpression of CYP4F2 Total protein was extracted from transfected cells or tissues of successfully participated in the overproduction of 20-HETE, transgenic mice and the concentration was determined with the and consequently resulted in elevated arterial blood pressure. Bradford method. Denatured protein was separated by electrophor- esis and transferred onto poly (vinylidene fluoride) membranes at The novel CYP4F2 transgenic model provides a valid tool to 41C. Membranes were subsequently incubated with anti-CYP4F2 further elucidate the mechanism of CYP4F2–20-HETE path- antibody (1:1000) (Fitzgerald, Concord, MA, USA), anti-his anti- way on blood pressure. body (1:1000) (Sigma-Aldrich, St Louis, MO, USA), or anti- GAPDH antibody (1:10,000) (KangChen Bio-Tech, Shanghai, China) as primary antibody, followed by alkaline phosphatase- METHODS conjugated IgG (1:2000) as the secondary antibody. The final Plasmid constructions detection reaction was performed with b-naphthyl acid phosphate The promoter fragment from À224 to þ 1 of the KAP gene was and Fast Blue B salt (Sigma-Aldrich). obtained by PCR using mouse genomic DNA as template and primers: 50-GCGGTACCGGGTGGTGCACCTAGTCAG-30 and 50-CCAGATCTCCAAAAAGAGTCCTCCTGG-30.Theamplifiedpro- Immunohistochemistry duct was cloned into the KpnI/HindIII sites of the pGL3-enhancer Mice were killed and immediately perfused transcardiacally with ice- vector (Promega, Madison, WI, USA) to generate pKAP-LUC cold PBS, followed by 4% paraformaldehyde in PBS. Organs were plasmid. The CYP4F2 cDNA was amplified by PCR with primers isolated, fixed at 41C for additional 4 h, and then immersed in 30% 50-TGCCCTGCAGGATGTCCCAGCTGA-30 and 50-CTCTCTAGAT sucrose solution at 41C overnight. The samples were embedded in CAATGATGATGATGATGATGGCTCAGGGGCTCCACCCGCAG-30 low temperature paraffin wax, and 4-mm thick sections were then

1294 Kidney International (2009) 75, 1288–1296 X Liu et al.: CYP4F2 transgenic mice original article

prepared and incubated with anti-his antibody (1:300) at room DISCLOSURE temperature for 6 h. With Streptavitin Peroxidase-Conjugated (SP) All the authors declared no competing interests. method, Diaminobenzidin (Maxmim Bio-Tech, Fuzhou, China) was utilized as the chromagen to localize peroxidase activity. Photo- ACKNOWLEDGMENTS This study was supported by grants from the National Natural micrographs were taken by an OLYMPS IX51 inverted microscope Science Foundation of China (30571027) and the Natural Science with the OLYMPUSMicro software. Foundation of Liaoning Province (2007225004-4).

Preparation of renal microsomes and AA hydroxylation assay SUPPLEMENTARY INFORMATION Renal microsomes were prepared according to the method described Supplementary information accompanies the paper on the Kidney before1 for Western blot analysis and AA hydroxylation assay. The International website (http://www.nature.com/ki). conversion of AA to 20-HETE was assessed in a reaction mixture of 300 ml volume containing 100 mM potassium phosphate buffer (pH REFERENCES 7.4), 3.3 mM MgCl2,80mM AA (Cayman Chemical, Ann Arbor, MI, 1. Lasker JM, Chen WB, Wolf I et al. Formation of 20-hydroxyeicosatetraenoic USA), 1 mM NADPH (Roche Applied Science, Basel, Switzerland) acid, a vasoactive and natriuretic , in human kidney. Role of and 0.6 mg/ml mouse renal microsomes. After violent vortex and CYP4F2 and CYP4A11. J. Biol Chem 2000; 275: 4118–4126. 2. McGiff JC, Quilley J. 20-HETE and the kidney: resolution of old problems 5 min preincubation at 371C, NADPH was added to start the and new beginnings. Am J Physiol 1999; 277: R607–R623. reaction at 371C for 30 min, then the 20-HETE production was 3. Miyata N, Roman RJ. Role of 20-hydroxyeicosatetraenoic acid (20-HETE) in measured. vascular system. J Smooth Muscle Res 2005; 41: 175–193. 4. Roman RJ. P-450 metabolites of arachidonic acid in the control of cardiovascular function. Physiol Rev 2002; 82: 131–185. 20-HETE ELISA assay 5. Sarkis A, Roman RJ. Role of cytochrome P450 metabolites of arachidonic 20-HETE ELISA kits (Detroit R&D, Detroit, MI, USA) were used in the acid in hypertension. Curr Drug Metab 2004; 5: 245–256. 6. Liu H, Zhao Y, Nie D et al. Association of a functional cytochrome P450 AA hydroxylation assay of renal microsomes and urinary 20-HETE 4F2 haplotype with urinary 20-HETE and hypertension. J Am Soc Nephrol measurement following manufacturer’s instructions. Briefly, samples 2008; 19: 714–721. were loaded to the 96-well plate that was coated with anti-20-HETE 7. Ward NC, Tsai IJ, Barden A et al. A single nucleotide polymorphism in the antibody. The plate was read at 450 nm OD using an automatic ELISA CYP4F2 but not CYP4A11 gene is associated with increased 20-HETE excretion and blood pressure. Hypertension 2008; 51: 1393–1398. plate reader (Bio-Rad Model 680, Hercules, CA, USA), and the 20- 8. Gainer JV, Bellamine A, Dawson EP et al. Functional variant of CYP4A11 HETE concentrations were calculated from the standard curve. 20-hydroxyeicosatetraenoic acid synthase is associated with essential hypertension. Circulation 2005; 111: 63–69. 9. Meseguer A, Catterall JF. Mouse kidney androgen-regulated protein Blood pressure measurement messenger ribonucleic acid is expressed in the proximal convoluted The blood pressure of conscious mice was measured by tail-cuff tubules. Mol Endocrinol 1987; 1: 535–541. method using IITC Life Science Model 1631 tail pulse detection 10. Ding Y, Davisson RL, Hardy DO et al. The kidney androgen-regulated system (IITC Life Science Model 1631, Woodland Hills, CA, USA) protein promoter confers renal proximal tubule cell-specific and highly androgen-responsive expression on the human angiotensinogen gene in according to the manufacturer’s instructions. Mice were trained for transgenic mice. J Biol Chem 1997; 272: 28142–28148. 5 days and acclimated mice were warmed at 331C in the heating 11. Ding Y, Sigmund CD. Androgen-dependent regulation of human container for 5 min before measurement. One measurement session angiotensinogen expression in KAP-hAGT transgenic mice. Am J Physiol involved 10 repetitions and at least three sessions were performed on Renal Physiol 2001; 280: F54–F60. 12. Lavoie JL, Lake-Bruse KD, Sigmund CD. Increased blood pressure in each mouse to calculate the average. transgenic mice expressing both human renin and angiotensinogen in the renal proximal tubule. Am J Physiol Renal Physiol 2004; 286: F965–F971. Blood biochemical assays 13. Malstrom SE, Tornavaca O, Meseguer A et al. The characterization and Blood samples were collected for biochemical assays including hormonal regulation of kidney androgen-regulated protein (Kap)- serum creatinine, BUN, and plasma aldosterone determination. luciferase transgenic mice. Toxicol Sci 2004; 79: 266–277. Serum creatinine and BUN measurement was performed on the 14. Sachetelli S, Liu Q, Zhang SL et al. RAS blockade decreases blood pressure and proteinuria in transgenic mice overexpressing rat angiotensinogen Vitros 950 analyzer (Ortho-Clinical Diagnostics, Raritan, NJ, USA) gene in the kidney. Kidney Int 2006; 69: 1016–1023. with the VITROS Chemistry Products CREA and BUN Slides 15. Soler M, Tornavaca O, Sole´ E et al. Hormone-specific regulation of the according to the manufacturer’s instructions. Plasma aldosterone kidney androgen-regulated gene promoter in cultured mouse renal was determined by radioimmunoassay (Active Aldosterone Ria Kit, proximal-tubule cells. Biochem J 2002; 366: 757–766. 16. Grates HE, McGowen RM, Gupta SV et al. Quantification of DSL 8600, Diagnostic System Laboratories, Webster, TX, USA) using 20-hydroxyeicosatetraenoic acid by colorimetric competitive Gama Immunoassay Counter (USTC Chuangxin Co., Ltd. Zonkia linked immunosorbent assay. J Biosci 2003; 28: 109–113. Branch, Hefei, China). 17. Holla VR, Adas F, Imig JD et al. Alterations in the regulation of androgen- sensitive cyp 4a cause hypertension. Proc Natl Acad Sci USA 2001; 9: 5211–5216. Statistical analysis 18. Maier KG, Henderson L, Narayanan J et al. Fluorescent HPLC assay for Statistical analysis was run with SPSS version 10.0 for Windows 20-HETE and other P-450 metabolites of arachidonic acid. Am J Physiol (SPSS, Chicago, IL, USA). 20-HETE concentration was not normally Heart Circ Physiol 2000; 279: H863–H871. distributed and was presented by geometric mean (95% CI). Mouse 19. Laffer CL, Laniado-Schwartzman M, Wang MH et al. Differential regulation of natriuresis by 20-hydroxyeicosatetraenoic Acid in human SBP, serum creatinine, BUN, and plasma aldosterone were presented salt-sensitive versus salt-resistant hypertension. Circulation 2003; 107: by mean±s.d. Comparison between logarithmic urinary 20-HETE 574–578. and SBP was performed by one-way ANOVA, and their correlation 20. Hoagland KM, Flasch AK, Roman RJ. Inhibitors of 20-HETE formation was identified by Pearson’s method. Serum creatinine, BUN, and promote salt-sensitive hypertension in rats. Hypertension 2003; 42: 669–673. plasma aldosterone were compared by independent-sample t-test. 21. Zhang F, Chen CL, Qian JQ et al. Long-term modifications of blood Statistical significance was defined at Po0.05. pressure in normotensive and spontaneously hypertensive rats by gene

Kidney International (2009) 75, 1288–1296 1295 original article X Liu et al.: CYP4F2 transgenic mice

delivery of rAAV-mediated cytochrome P450 arachidonic acid 25. Meseguer A, Catterall JF. Cell-specific expression of kidney androgen- hydroxylase. Cell Res 2005; 15: 717–724. regulated protein messenger RNA is under multihormonal control. Mol 22. Iwai N, Inagami T. Isolation of preferentially expressed genes in the Endocrinol 1990; 4: 1240–1248. kidneys hypertensive rats. Hypertension 1991; 17: 161–169. 26. Nakagawa K, Marji JS, Schwartzman ML et al. The androgen-mediated 23. Su P, Kaushal KM, Kroetz DL. Inhibition of renal arachidonic acid omega- induction of the kidney arachidonate hydroxylases is associated with the hydroxylase activity with ABT reduces blood pressure in the SHR. Am J development of hypertension. Am J Physiol Regul Integr Comp Physiol Physiol 1998; 275: R426–R438. 2003; 284: R1055–R1062. 24. Bianco RA, Keen HL, Lavoie JL et al. Untraditional methods for targeting 27. Ward NC, Rivera J, Hodgson J et al. Urinary 20-hydroxyeicosatetraenoic the kidney in transgenic mice. Am J Physiol Renal Physiol 2003; 285: acid is associated with endothelial dysfunction in humans. Circulation F1027–F1033. 2004; 110: 438–443.

1296 Kidney International (2009) 75, 1288–1296