UEDONO H, TSUDA A, ISHIMURA E, NAKATANI S, KURAJOH M, MORI K, UCHIDA J, EMOTO

M, NAKATANI T, & INABA M. (2017). U-shaped relationship between serum uric acid levels and

intrarenal hemodynamic parameters in healthy subjects. American Journal of Physiology. Renal Physiology. 312, F992-F997.

U-shaped relationship between serum uric

acid levels and intrarenal hemodynamic

parameters in healthy subjects

Hideki Uedono, Akihiro Tsuda, Eiji Ishimura, Shinya Nakatani,

Masafumi Kurajoh, Katsuhito Mori, Junji Uchida, Masanori

Emoto, Tatsuya Nakatani, Masaaki Inaba

Citation American Journal of Physiology-Renal Physiology, 312(6); F992-F997

Issue Date 2017-06-01

Type Journal Article

Textversion author

Right

The following article has been accepted by American Journal of Physiology-Renal

Physiology. After it is published, it will be found at

https://doi.org/10.1152/ajprenal.00645.2016

URI http://dlisv03.media.osaka-cu.ac.jp/il/meta_pub/G0000438repository_15221466-312-

6-F992

DOI 10.1152/ajprenal.00645.2016

SURE: Osaka City University Repository

http://dlisv03.media.osaka-cu.ac.jp/il/meta_pub/G0000438repository

1

U-shaped relationship between serum uric acid levels and 1

intrarenal hemodynamic parameters in healthy subjects 2

3

Hideki Uedono, Akihiro Tsuda, Eiji Ishimura, Shinya Nakatani, Masafumi Kurajoh, 4

Katsuhito Mori, Junji Uchiada, Masanori Emoto, Tatsuya Nakatani, Masaaki Inaba 5

6

Department of Nephrology, Department of Metabolism, Endocrinology, Molecular Medicine, 7

and Department of Urology Osaka City University Graduate School of Medicine 8

9

H.U., A.T., and E.I. contributed equally to this work. 10

11

Running Head: Uric acid and afferent arteriolar resistance 12

Key words: uric acid, inulin clearance, para-aminohippurate clearance, afferent arteriolar 13

resistance, renal hemodynamics 14

15

Word count: 3090 words 16

Number of figures and tables: 2 figures and 2 tables 17

18

Correspondence should be sent to: Akihiro Tsuda., Department of Metabolism, 19

Endocrinology and Molecular Medicine, Internal Medicine, Osaka City University Graduate 20

School of Medicine, 1-4-3, Asahi-machi, Abeno-ku, Osaka 545-8585, Japan 21

Tel: +81-6-6645-3806, Fax: +81-6-6645-3808 22

E-mail address: naranotsudadesu@infoseek.jp 23

24

2

Disclosure Statements: The authors have nothing to disclose. 25

26

3

Abstract 27

BACKGROUND: Hyperuricemia has been reported to affect renal hemodynamics. In a 28

recent study, both low and high levels of serum uric acid (SUA) were found to be associated 29

with loss of kidney function. 30

Objective: The goal of this study was to evaluate the relationship between SUA levels and 31

intrarenal hemodynamic parameters in healthy subjects, utilizing plasma clearance of 32

para-aminohippurate (CPAH) and inulin (Cin) 33

Subjects and Methods: Renal and glomerular hemodynamics were evaluated by 34

simultaneous measurements of CPAH and Cin in 48 healthy subjects (54.6 ± 13.4 years). 35

Intrarenal hemodynamic parameters, including efferent (Re) and afferent (Ra) arteriolar 36

resistance, were calculated using Gomez's formulae. Relationships of SUA levels with these 37

intrarenal hemodynamic parameters were examined. 38

Results: In quadratic regression analysis, SUA levels had a significant, inverse U-shaped 39

relationship with Cin (p < 0.0001, R2 = 0.350) and CPAH (p = 0.0093, R2 = 0.188), and a 40

U-shaped relationship with Ra (p = 0.0011, R2 = 0.262). In multiple regression analysis with 41

normal (3.0 to 6.0 mg/dL) and mildly low or high (<3.0 or >6.0 mg/dL) SUA levels entered 42

as dummy variables of 0 and 1, respectively, mildly low or high SUA levels were 43

significantly and independently associated with Ra (β = 0.230, p = 0.0403) after adjustment 44

for several factors. (R2 = 0.597, p < 0.0001). 45

Conclusions: Both mild hyperuricemia and mild hypouricemia are significantly associated 46

with increased Ra, although weakly. The increase in Ra in subjects with mild hyperuricemia 47

or hypouricemia may be related to renal hemodynamic abnormalities, possibly leading to a 48

decline in renal function. 49

50

4

Higher uric acid levels are well known to be associated with reduced glomerular filtration 51

rate (GFR) (3, 12, 13, 39). In rat models, the mechanism underlying renal dysfunction 52

induced by elevated uric acid has been explained by endothelial dysfunction, which causes a 53

decrease in nitric oxide synthesis and a subsequent increase in resistance of the renal artery 54

(10, 24). Glomerular ischemia can be induced by hyperuricemia in a rat model (19), and 55

increased resistance of the renal artery and decreased renal plasma flow may occur due to 56

hyperuricemia in humans (40). We recently found a significant association between higher 57

serum uric acid levels and increased afferent arteriolar resistance in humans with normal 58

renal function (37). 59

Hypouricemia may be a risk factor for development of acute renal failure with loin pain 60

and patchy renal ischemia after anaerobic exercise (ALPE) induced by a non-myoglobinuric 61

status (16). This complication may be due to the decreased antioxidant potential caused by 62

loss of uric acid leading to kidney injury by reactive oxygen species (ROS) (22). Patchy 63

vasoconstriction of the renal vessels may be induced by oxidative imbalance in ALPE, 64

given the wedge-shaped distribution and reversibility (14). However, idiopathic renal 65

hypouricemia is a rare disorder with an incidence of 0.15% in Japan (9). Thus, the cause of 66

acute kidney injury in ALPE is unclear (15). 67

Kanda et al. recently found that low or high levels of serum uric acid are associated with 68

loss of kidney function in a community-based prospective cohort study with an observational 69

period of up to 9 years (21). Verdecchia et al. showed that serum uric acid levels have a 70

U-shaped association with cardiovascular disease in essential hypertension (38), and Sugihara 71

et al. found that hyper- and hypouricemia cause endothelial dysfunction (33). These data 72

suggest that both hyper- and hypouricemia are risk factors for organ failure. However, 73

hypouricemia is normally defined as severe at a uric acid level of <2 mg/dL, and there are no 74

data supporting a relationship of mild hypouricemia with intrarenal hemodynamic 75

5

parameters. 76

In humans, it is not possible to measure glomerular hemodynamic variables directly. 77

However, Gomez published a series of formulae for indirect evaluation of human glomerular 78

hemodynamics (7). These formulae have been used to calculate glomerular hemodynamics in 79

various conditions, including untreated and treated essential hypertension (7, 18), 80

renovascular hypertension (26), primary aldosteronism (31), and supraventricular tachycardia 81

(28). We have also utilized the formulae for evaluation of intrarenal hemodynamics in 82

diabetic and non-diabetic subjects (34, 36). The aim of the present study was to evaluate the 83

relationships of serum uric acid levels, including mild hyper- and hypouricemia, with 84

intrarenal hemodynamic parameters in healthy subjects, utilizing data for clearance of 85

para-aminohippurate (CPAH) and inulin (Cin). 86

87

SUBJECTS AND METHODS 88

Subjects 89

The subjects were 48 healthy individuals (19 males and 29 females, age 54.6 ± 13.4 years 90

old) who were scheduled to provide a kidney for transplantation at Osaka City University 91

Hospital between January 2010 and March 2016. All subjects were admitted to the Urology 92

Department of our Hospital for examinations to confirm the suitability for providing a kidney 93

for transplantation, including oral glucose tolerance test and several blood tests, including 94

measuring prostate specific antigen (PSA) in male subjects. All subjects underwent medical 95

checkup by expert urologists and nephrologists, who found no urination disorder in any 96

subject during the clearance study. PSA was normal in all of the male subjects. None of the 97

subjects had received treatment with any drug such as antihypertensive agents. All subjects 98

showed normal glucose tolerance. Urinalysis showed normal findings in all subjects. During 99

the course of admissions including the present study period, all participants took hospital 100

6

food, including salt 6 g/day and protein 60-70 g/day. The clearance study, as described below, 101

was performed in the morning after overnight fasting (approximately 12 hours of fasting). 102

The study protocol was approved by the Ethics Committee of Osaka City University 103

Graduate School of Medicine (# 1444). Written informed consent was obtained from each 104

subject. 105

106

Measurements of Cin and CPAH, and calculation of intrarenal hemodynamic parameters 107

Renal plasma flow (RPF) and glomerular filtration rate (GFR) were determined by the 108

constant input clearance technique using PAH and inulin, respectively (34, 36). Continuous 109

intravenous infusion of 1% inulin and 0.5% PAH via the antecubital vein was performed in 110

the morning after an overnight fast, based on the method of Horio et al. (11). CPAH and Cin 111

were measured simultaneously using a simple method based on a single urine collection. In 112

brief, subjects received 500 mL of water orally 15 min before infusion. As a priming dose of 113

PAH and inulin, the rates of infusion were set at 300 mL/h for the first 30 min and at 100 mL/h 114

for the remaining time. As instructed by the physicians, subjects completely emptied their 115

bladder 45 min after the start of the test and urine was collected for measurement of urinary 116

PAH and inulin. The urine collection period was set at 90 min to increase the accuracy of the 117

clearance study. We did not determine whether the bladder of the subjects was emptied, i.e. 118

through the use of an indwelling bladder catheter or ultrasonography, since the clearance 119

study was performed according to the Japanese guidelines, as has been performed previously 120

in several studies (11, 20, 34-37). Blood samples for measurements of serum PAH and inulin 121

were taken at the beginning and end of the clearance period. Blood samples were obtained 122

from the antecubital vein after an overnight fast, and serum was separated by a routine, 123

standard clinical laboratory method within an hour. Serum uric acid and total protein were 124

measured by standard routine clinical laboratory methods of uric acid/uricase assay and 125

7

Biuret test, respectively. 126

CPAH and Cin were calculated by the UV/P method (U: concentration in urine, V: urine 127

volume [mL/min], P: concentration in plasma) using the mean serum PAH and inulin 128

concentrations at the beginning and end of the clearance period of 90 min. Plasma PAH and 129

inulin were determined colorimetrically using the N-(1-naphthyl)ethylenediamine and 130

anthrone methods, respectively, with a Corning 258 spectrophotometer (5, 6, 25). 131

Formulae introduced by Gomez (7) for indirect assessment of glomerular hemodynamics 132

(7, 8) were utilized. These formulae were designed for quantitative estimation of filtration 133

pressure across the glomerular capillaries (ΔPF), glomerular hydrostatic pressure (Pglo), and 134

afferent and efferent arteriolar resistances (Ra and Re, respectively), using measured blood 135

pressure, GFR as measured by Cin, RPF as measured by CPAH, hematocrit, and plasma protein 136

concentrations, under the assumptions that (1) intrarenal vascular resistances can be divided 137

into three compartments; afferent, efferent and venular; (2) hydrostatic pressures in the 138

venules, interstitium, renal tubules, and Bowman’s space (PBow) are in equilibrium at about 10 139

mmHg; (3) the gross filtration coefficient (KFG) is 0.0406 mL/s per mmHg per kidney; and 140

(4) filtration disequilibrium is postulated along the glomerular capillaries (7, 8). 141

The Gomez formulae were calculated from the original paper as follows: 142

ΔPF = GFR/KFG 143

Pglo = ΔPF + PBow + πG 144

πG = 5·(CM - 2) 145

CM = TP/FF·ln (1/(1-FF)) 146

where ΔPF is the filtration pressure across the glomerular capillary. KFG (gross filtration 147

coefficient) was estimated as 0.0406 mL/s · mmHg per kidney, and PBow (hydrostatic 148

pressure in Bowman's space) as 10 mmHg. πG (oncotic pressure in the glomerular 149

capillaries) can be obtained from CM (plasma protein concentration in the glomerular 150

8

capillaries), which is calculated from total protein concentration (TP) and filtration fraction 151

(FF) (FF = glomerular filtration rate (GFR)/renal plasma flow (RPF). 152

From Ohm's law: 153

Ra = ((MBP-Pglo)/RBF) ·1328 154

Re = (GFR/KFG ·(RBF-GFR)) ·1328 155

RBF can be calculated from RPF and hematocrit (Ht), using the standard formula: 156

RBF = RPF/(1-Ht) 157

In these equations, 1328 is the conversion factor to dyne · s · cm−5; GFR (glomerular 158

filtration rate), RPF (renal plasma flow), and RBF (renal blood flow) are expressed in mL/s; 159

and the mean blood pressure (MBP) is calculated as (2 × diastolic BP + systolic BP)/3. In the 160

present study, the Gomez formulae were applied in subjects with Cin > 60 mL/min/1.73m2. Ht 161

denotes hematocrit. 162

163

Statistical analysis 164

Results are expressed as the mean ± standard deviation (SD). Correlations between two 165

variables were examined using single and quadratic regression analyses. Multiple regression 166

analyses were performed to evaluate relationships between general hemodynamic parameters 167

and other clinical parameters. In the analysis, normal (3.0 to 6.0 mg/dL) and mildly low or 168

high (<3.0 or >6.0 mg/dL) uric acid levels were entered as dummy variables of 0 and 1, 169

respectively. All analyses were performed using StatView 5 for Windows (SAS Institute Inc., 170

Cary, NC, USA). The level of significance was set at p < 0.05. 171

172

Results 173

Baseline characteristics 174

Baseline characteristics of the subjects are shown in Table 1. The mean GFR measured by 175

9

Cin was 95.8 ± 22.6 mL/min/1.73m2, and GFR was ≥60 mL/min/1.73m2 in all subjects. The 176

mean serum uric acid level was 4.9 ± 1.1 mg/dL, and uric acid was <7.0 mg/dL in all 177

subjects. 178

179

Relationships between serum uric acid and renal hemodynamic parameters 180

Relationships between serum uric acid levels and renal hemodynamic parameters were 181

examined in all subjects. Serum uric acid levels showed a significant negative correlation 182

with GFR measured as Cin (r = - 0.390, p =0.0061), but no significant correlation with renal 183

plasma flow (r = 0.191, p = 0.1931) or filtration fraction (r = 0.019, p = 0.8984). Quadratic 184

regression analysis showed that serum uric acid levels had a significant inverse U-shaped 185

relationship with GFR measured as Cin (R2 = 0.350, p < 0.0001) and with RPF measured as 186

CPAH (R2 = 0.188, p = 0.0093) (Figure 1); a significant U-shaped relationship with Ra, (R2 = 187

0.262, p = 0.001), but not with Re (R2 = 0.0026, p = 0.5549); and an inverse U-shaped 188

relationship with Pglo with significance (R2 = 0.188, p = 0.0105). 189

Relationship between Ra, GFR and serum uric acid was examined separately in men (n = 190

19) and in women (n= 29) by quadratic regression analyses. In both men and women, serum 191

uric acid levels had a significant U-shaped relationship with Ra (R2=0.347, p=0.0332 in men, 192

and R2=0.306, p=0.0087 in women). In both men and women, serum uric acid levels had a 193

significant inverse U-shaped relationship with GFR measured as Cin (R2=0.461, p=0.0072 in 194

men, and R2=0.264, p=0.0187 in women). 195

196

Effect of mild hyperuricemia and mild hypouricemia on renal hemodynamics 197

Multiple regression analyses were performed to examine independent associations of 198

mildly low or high uric acid levels with GFR, RPF, Ra, and Pglo. Because uric acid levels had 199

a relationship with several hemodynamic parameters in quadratic regression analyses, 200

10

multiple regression analyses were performed with uric acid included as an independent 201

dummy variable: i.e., normal (3.0 to 6.0 mg/dL) and low or high (< 3.0 or >6.0 mg/dL) uric 202

acid levels were entered as dummy variables of 0 and 1, respectively. In the multiple 203

regression analyses (Table 2), mildly low or high serum uric acid levels (low and high) were 204

independently associated with GFR (β = -0.314, p = 0.0256) and RPF (β = -0.270, p = 205

0.0551) with borderline significance after adjustment for age, gender, body mass index, and 206

systolic blood pressure, in which β is standard partial regression coefficient of multiple 207

regression analysis. In a second multiple regression analysis (Table 2), mildly low or high 208

serum uric acid levels (low and high) were significantly and independently associated with 209

increased Ra (β = 0.230 p = 0.0403) and decrease Pglo (β = -0.259 p = 0.0838) after 210

adjustment of the above confounders, but not with Re (β = -0,234 p = 0.1459). 211

212

Discussion 213

In this study, we examined relationships between serum uric acid levels and renal 214

hemodynamic parameters in healthy subjects. Quadratic regression analyses showed that 215

these levels had a significant inverse U-shaped relationship with GFR and RPF, a significant 216

U-shaped relationship with Ra, and an inverse U-shaped relationship with Pglo, but no 217

relationship with Re. In multiple regression analyses, mildly abnormal serum uric acid levels 218

(low and high) were independently associated with GFR and RPF, and significantly and 219

independently associated with Ra and Pglo after adjustment for several confounders, whereas 220

Re. 221

The relationship between uric acid and kidney disease in humans is complicated by many 222

confounding variables. Serum uric acid levels are associated with other risk factors for kidney 223

disease, such as hypertension, insulin resistance and microalbuminuria (4, 32), and potential 224

mechanisms underlying kidney damage by uric acid include induction of afferent arteriopathy, 225

11

inflammation, and activation of the renin-angiotensin system (4, 23). Recently, we examined 226

the relationship between serum uric acid and intrarenal hemodynamic parameters in humans, 227

utilizing the plasma clearance of para-aminohippurate (CPAH ) and inulin (Cin), and found that 228

higher serum uric acid levels are significantly associated with Ra in subjects with Cin >60 229

mL/min/1.73m2 (37). Using renal biopsy samples, Kohagura et al. found that hyperuricemia 230

was significantly and independently associated with renal arteriolar hyalinosis and higher 231

grade wall thickening, which suggests that higher uric acid may cause progression of chronic 232

kidney disease (CKD) through an effect on arterioles (27). Weiner et al. reported that serum 233

uric acid is a modest independent risk factor for incident CKD in the general population (39), 234

and Bellomo et al. also showed that the serum uric acid level is an independent risk factor for 235

decreased kidney function (2). The effects of elevated uric acid on incident CKD are thought 236

to be due to a direct toxic effect of uric acid on the kidney, via exacerbation of hypertension 237

(2, 27, 39). In humans, the highest tertile of serum uric acid levels is associated with incident 238

kidney injury, but not with progression of kidney injury (29). However, Iseki et al. reported 239

that hyperuricemia is associated with progression of CKD to end-stage renal disease (13). 240

Our finding that mild hyperuricemia is associated with increased Ra and decreased GFR, RPF 241

and Pglo is consistent with these previous studies relating hyperuricemia to kidney dysfunction. 242

Kanda et al. recently described a U-shaped association between the risk of loss of 243

kidney function and serum uric acid level in healthy subjects, showing that hypouricemia 244

is also a risk factor for kidney dysfunction (21). Some studies have also shown that uric 245

acid has a U-shaped association with cardiovascular disease and endothelial dysfunction. 246

Sugihara et al. found that depletion of uric acid due to a loss-of-function mutation in 247

SLC22A12 (URAT1) causes endothelial dysfunction in hypouricemia (33). Uric acid serves 248

as an antioxidant in vascular endothelial cells (33) and hypouricemia is a risk factor for ALPE 249

induced by non-myoglobinuria. Ishikawa et al. reported exercise-induced acute renal failure 250

12

in three patients with renal hypouricemia (17). The cause of acute kidney injury in ALPE is 251

unclear, but several hypotheses have been proposed (15). A plausible explanation is that the 252

decreased antioxidant potential due to the reduced uric acid level leads to reduction of renal 253

plasma flow by increased ROS (1, 30). Patchy vasoconstriction of renal vessels may then 254

be induced by the oxidative imbalance in ALPE (14). 255

These reports suggest that reduction of renal blood flow is promoted by endothelial 256

dysfunction of the kidney in patients with hypouricemia. In this study, we showed that 257

mild hyper- and hypouricemia are both associated with decreased GFR and RPF and 258

increased Ra although weakly. The increase in Ra in subjects with mildly higher or lower uric 259

acid levels may lead to dysfunction of glomerular perfusion. To the best our knowledge, this 260

is the first description of the relationships between serum uric acid levels and intrarenal 261

hemodynamic parameters in healthy subjects. The mechanisms underlying our results may be 262

associated with vasoconstriction of afferent arterioles in both mild hyper- and hypouricemia. 263

There are some limitations in this study. First, the measurements were performed in a 264

relatively small number of Japanese subjects, and a large-scale study would be needed to 265

confirm whether hyper- or hypouricemia increases Ra. In particular, the generalizability of 266

the findings is not known as this was a study performed in a relatively small number of 267

Japanese adults, which may not be relevant to other ethnicities, who may exhibit different 268

frequencies for URAT1 mutations. Second, we did not measure renal hemodynamics directly, 269

since this is very difficult in humans. Instead, we used Gomez’s formulae to calculate renal 270

hemodynamics. These formulae are based on several assumptions, but many studies have 271

validated their clinical utility (26, 28, 31, 36). Third, there could be differences in renal 272

hemodynamics based on the mechanism underlying the low uric acid, such as diet (although, 273

during the course of admissions including the present study period, all participants took 274

hospital food, including salt 6 g/day and protein 60-70 g/day), or the genetic background on 275

13

the fractional excretion of uric acid and endothelial function in response to lower serum uric 276

acid. Future studies will be needed to determine if the mechanism for low uric acid is related 277

to the hemodynamic findings. Fourth, since we did not measure nitric oxide levels, plasma 278

renin activity, angiotensin II levels or URAT1 gene expression/mutation, we could not 279

determine the mechanism underlying how and to what extent the lower and higher serum uric 280

acid levels affect renal hemodynamic abnormalities. Fifth, since, in the present study, as in 281

the previous studies by ourselves and others (11, 20, 34-37), the bladder was not confirmed as 282

being empty after the subjects had been instructed to completely void their bladder at the 283

specified times, it is possible that a very small amount of residual urine might have affected 284

the data of the clearance study. However, none of the previous studies by others or ourselves 285

utilized methods to confirm voiding of the bladder in human subjects, i.e., by indwelling 286

bladder catheter or ultrasonography. As such, we cannot confirm that the emptiness of the 287

bladder was thoroughly examined at the instructed times in the present study, or in the 288

previous studies (11, 20, 34-37). Sixth, we could not follow-up the data on the same subjects 289

at some interval to determine the predictive value of these measurements. It is a future study 290

needed to explore the significance of these parameters on renal dysfunction. Lastly, we did 291

not examine the mechanisms through which Ra is affected by mildly abnormal serum uric 292

acid levels. Approaches such as comparison of renal hemodynamics in the presence of 293

various vasoactive substances in urine and blood are needed to reveal these mechanisms. 294

In conclusion, the findings in this study show for the first time that mildly lower and 295

mildly higher serum uric acid levels in healthy subjects are both significantly associated with 296

decreased GFR and RPF, and with increased Ra. The increase in Ra in subjects with lower or 297

higher uric acid levels may be related to abnormalities in renal hemodynamics, and may lead 298

to dysfunction of glomerular perfusion. 299

300

14

Declaration of interest 301

The authors declare that there is no conflict of interest that could be perceived as 302

prejudicing the impartiality of the reported research. This study was not funded by a grant 303

from a funding agency in the public, commercial or not-for-profit sector. 304

305

Author contributions 306

H.U., A.T. and E.I. generated and analyzed the data and wrote the manuscript. H. U., A.T., 307

E.I., S.U., S.N., J.U.,K.M., M.E., T.N., and M.I. contributed to the discussion and reviewed 308

the manuscript. H.U.., A.T. and E.I. had full access to all of the study data and take 309

responsibility for the integrity of the data and the accuracy of the data analysis. 310

311

15

References 312

1. Ames BN, Cathcart R, Schwiers E, and Hochstein P. Uric acid provides an antioxidant defense in 313 humans against oxidant- and radical-caused aging and cancer: a hypothesis. Proceedings of the National 314 Academy of Sciences of the United States of America 78: 6858-6862, 1981. 315 2. Bellomo G, Venanzi S, Verdura C, Saronio P, Esposito A, and Timio M. Association of uric acid 316 with change in kidney function in healthy normotensive individuals. Am J Kidney Dis 56: 264-272, 2010. 317 3. Chonchol M, Shlipak MG, Katz R, Sarnak MJ, Newman AB, Siscovick DS, Kestenbaum B, 318 Carney JK, and Fried LF. Relationship of uric acid with progression of kidney disease. Am J Kidney Dis 50: 319 239-247, 2007. 320 4. Fathallah-Shaykh SA and Cramer MT. Uric acid and the kidney. Pediatric nephrology 29: 321 999-1008, 2014. 322 5. Fliser D, Dikow R, Demukaj S, and Ritz E. Opposing effects of angiotensin II on muscle and renal 323 blood flow under euglycemic conditions. J Am Soc Nephrol 11: 2001-2006, 2000. 324 6. Fliser D, Zeier M, Nowack R, and Ritz E. Renal functional reserve in healthy elderly subjects. J Am 325 Soc Nephrol 3: 1371-1377, 1993. 326 7. Gomez DM. Evaluation of renal resistances, with special reference to changes in essential 327 hypertension. J Clin Invest 30: 1143-1155, 1951. 328 8. Guidi E, Cozzi MG, Minetti EE, Civati G, Busnach G, and Brando B. Effect of familial 329 hypertension on glomerular hemodynamics and tubulo-glomerular feedback after uninephrectomy. Am J 330 Hypertens 14: 121-128, 2001. 331 9. Hisatome I, Ogino K, Kotake H, Ishiko R, Saito M, Hasegawa J, Mashiba H, and Nakamoto S. 332 Cause of persistent hypouricemia in outpatients. Nephron 51: 13-16, 1989. 333 10. Hong Q, Qi K, Feng Z, Huang Z, Cui S, Wang L, Fu B, Ding R, Yang J, Chen X, and Wu D. 334 Hyperuricemia induces endothelial dysfunction via mitochondrial Na+/Ca2+ exchanger-mediated mitochondrial 335 calcium overload. Cell calcium 51: 402-410, 2012. 336 11. Horio M, Imai E, Yasuda Y, Hishida A, and Matsuo S. Simple sampling strategy for measuring 337 inulin renal clearance. Clin Exp Nephrol 13: 50-54, 2009. 338 12. Iseki K, Ikemiya Y, Inoue T, Iseki C, Kinjo K, and Takishita S. Significance of hyperuricemia as a 339 risk factor for developing ESRD in a screened cohort. Am J Kidney Dis 44: 642-650, 2004. 340 13. Iseki K, Oshiro S, Tozawa M, Iseki C, Ikemiya Y, and Takishita S. Significance of hyperuricemia 341 on the early detection of renal failure in a cohort of screened subjects. Hypertens Res 24: 691-697, 2001. 342 14. Ishikawa I. Acute renal failure with severe loin pain and patchy renal ischemia after anaerobic 343 exercise in patients with or without renal hypouricemia. Nephron 91: 559-570, 2002. 344 15. Ishikawa I. Comments on 'a young man with acute kidney injury after exercise'. Kidney Int 78: 345 1047-1048; author reply 1048, 2010. 346 16. Ishikawa I, Nakagawa M, Hayama S, Yoshida S, and Date T. Acute renal failure with severe loin 347 pain and patchy renal ischaemia after anaerobic exercise (ALPE) (exercise-induced acute renal failure) in a 348 father and child with URAT1 mutations beyond the W258X mutation. Nephrol Dial Transpl 20: 1015-1015, 349 2005. 350 17. Ishikawa I, Sakurai Y, Masuzaki S, Sugishita N, Shinoda A, and Shikura N. Exercise-induced 351 acute renal failure in 3 patients with renal hypouricemia. Nihon Jinzo Gakkai Shi 32: 923-928, 1990. 352 18. Isshiki T, Amodeo C, Messerli FH, Pegram BL, and Frohlich ED. Diltiazem maintains renal 353 vasodilation without hyperfiltration in hypertension: studies in essential hypertension man and the 354 spontaneously hypertensive rat. Cardiovasc Drugs Ther 1: 359-366, 1987. 355 19. Johnson RJ, Kang DH, Feig D, Kivlighn S, Kanellis J, Watanabe S, Tuttle KR, 356 Rodriguez-Iturbe B, Herrera-Acosta J, and Mazzali M. Is there a pathogenetic role for uric acid in 357 hypertension and cardiovascular and renal disease? Hypertension 41: 1183-1190, 2003. 358 20. Kakuta Y, Okumi M, Ichimaru N, Abe T, Nonomura N, Okuyama A, Kojima Y, Isaka Y, 359 Takahara S, Imai E, and Horio M. Utility of the Japanese GFR estimation equation for evaluating potential 360 donor kidney function. Clinical and experimental nephrology 14: 63-67, 2010. 361 21. Kanda E, Muneyuki T, Kanno Y, Suwa K, and Nakajima K. Uric acid level has a U-shaped 362 association with loss of kidney function in healthy people: a prospective cohort study. PloS one 10: e0118031, 363 2015. 364 22. Kaneko K, Taniguchi N, Tanabe Y, Nakano T, Hasui M, and Nozu K. Oxidative imbalance in 365 idiopathic renal hypouricemia. Pediatric nephrology 24: 869-871, 2009. 366 23. Kang DH and Chen W. Uric acid and chronic kidney disease: new understanding of an old problem. 367 Seminars in nephrology 31: 447-452, 2011. 368 24. Khosla UM, Zharikov S, Finch JL, Nakagawa T, Roncal C, Mu W, Krotova K, Block ER, 369 Prabhakar S, and Johnson RJ. Hyperuricemia induces endothelial dysfunction. Kidney Int 67: 1739-1742, 370

16

2005. 371 25. Kimata S, Mizuguchi K, Hattori S, Teshima S, and Orita Y. Evaluation of a new automated, 372 enzymatic inulin assay using D-fructose dehydrogenase. Clin Exp Nephrol 13: 341-349, 2009. 373 26. Kimura G, London GM, Safar ME, Kuramochi M, and Omae T. Glomerular hypertension in 374 renovascular hypertensive patients. Kidney Int 39: 966-972, 1991. 375 27. Kohagura K, Kochi M, Miyagi T, Kinjyo T, Maehara Y, Nagahama K, Sakima A, Iseki K, and 376 Ohya Y. An association between uric acid levels and renal arteriolopathy in chronic kidney disease: a 377 biopsy-based study. Hypertension research : official journal of the Japanese Society of Hypertension 36: 43-49, 378 2013. 379 28. Kojima S, Fujii T, Ohe T, Yoshitomi Y, Kuramochi M, Shimomura K, and Omae T. Glomerular 380 hemodynamics during supraventricular tachycardia. Jpn Heart J 36: 429-437, 1995. 381 29. Madero M, Sarnak MJ, Wang X, Greene T, Beck GJ, Kusek JW, Collins AJ, Levey AS, and 382 Menon V. Uric acid and long-term outcomes in CKD. Am J Kidney Dis 53: 796-803, 2009. 383 30. Murakami T, Kawakami H, Fukuda M, and Furukawa S. Patients with renal hypouricemia are 384 prone to develop acute renal failure--why? Clinical nephrology 43: 207-208, 1995. 385 31. Nagai T, Kimura G, Matsuoka H, Sanai T, Imanishi M, Kawano Y, Kojima S, Yoshida K, Abe H, 386 Ashida T, and et al. [Estimation of the intrarenal hemodynamics in patients with primary aldosteronism]. Nihon 387 Jinzo Gakkai Shi 31: 235-241, 1989. 388 32. Nashar K and Fried LF. Hyperuricemia and the progression of chronic kidney disease: is uric acid a 389 marker or an independent risk factor? Advances in chronic kidney disease 19: 386-391, 2012. 390 33. Sugihara S, Hisatome I, Kuwabara M, Niwa K, Maharani N, Kato M, Ogino K, Hamada T, 391 Ninomiya H, Higashi Y, Ichida K, and Yamamoto K. Depletion of Uric Acid Due to SLC22A12 (URAT1) 392 Loss-of-Function Mutation Causes Endothelial Dysfunction in Hypouricemia. Circ J 79: 1125-1132, 2015. 393 34. Tsuda A, Inaba M, Ichii M, Ochi A, Ohno Y, Nakatani S, Yamada S, Mori K, Tahara H, and 394 Ishimura E. Relationship between serum TSH levels and intrarenal hemodynamic parameters in euthyroid 395 subjects. Eur J Endocrinol 169: 45-50, 2013. 396 35. Tsuda A, Ishimura E, Ohno Y, Ichii M, Nakatani S, Machida Y, Mori K, Uchida J, Fukumoto S, 397 Emoto M, Nakatani T, and Inaba M. Poor glycemic control is a major factor in the overestimation of 398 glomerular filtration rate in diabetic patients. Diabetes care 37: 596-603, 2014. 399 36. Tsuda A, Ishimura E, Ohno Y, Ichii M, Nakatani S, Mori K, Fukumoto S, Emoto M, and Inaba 400 M. Significant association of poor glycemic control with increased resistance in efferent arterioles--study of 401 inulin and para-aminohippuric acid clearance in humans. Diabetes Res Clin Pract 104: 234-240, 2014. 402 37. Uedono H, Tsuda A, Ishimura E, Yasumoto M, Ichii M, Ochi A, Ohno Y, Nakatani S, Mori K, 403 Uchida J, Nakatani T, and Inaba M. Relationship Between Serum Uric Acid Levels and Intrarenal 404 Hemodynamic Parameters. Kidney & blood pressure research 40: 315-322, 2015. 405 38. Verdecchia P, Schillaci G, Reboldi G, Santeusanio F, Porcellati C, and Brunetti P. Relation 406 between serum uric acid and risk of cardiovascular disease in essential hypertension. The PIUMA study. 407 Hypertension 36: 1072-1078, 2000. 408 39. Weiner DE, Tighiouart H, Elsayed EF, Griffith JL, Salem DN, and Levey AS. Uric acid and 409 incident kidney disease in the community. J Am Soc Nephrol 19: 1204-1211, 2008. 410 40. Zhou X, Matavelli L, and Frohlich ED. Uric acid: its relationship to renal hemodynamics and the 411 renal renin-angiotensin system. Current hypertension reports 8: 120-124, 2006. 412 413

17

Figure Legends 414

Figure 1. Relationships of serum uric acid levels with glomerular filtration rate (GFR) 415

measured by inulin clearance, renal plasma flow (RPF) measured by para-aminohippurate 416

clearance, and filtration fraction (FF) in all subjects using quadratic regression analysis (n = 417

48). Serum uric acid had significant and inverse U-shaped relationships with GFR and RPF, 418

but not with FF. 419

420

Figure 2. Relationships of serum uric acid levels with resistance of the afferent (Ra) and 421

efferent (Re) arterioles and glomerular hydrostatic pressure (Pglo) in quadratic regression 422

analysis (n = 48). Serum uric acid had a significant and U-shaped relationship with Ra, no 423

significant relationship with Re, and an inverse U-shaped relationship with Pglo that had 424

borderline significance. 425

426

427

18

Table 1. Clinical characteristics of the subjects 428

Item Value

Age (years) 54.6 ± 13.4

Male / female 19 / 29

Body mass index (kg/m2) 23.1 ± 3.6

Mean blood pressure (mmHg) 89 ± 12

Systolic blood pressure (mmHg) 124 ± 17

Diastolic blood pressure (mmHg) 72 ± 11

Hemoglobin (g/dl) 13.8 ± 1.1

Serum albumin (g/dl) 4.1 ± 0.3

Serum uric acid (mg/dl) 4.9 ± 1.1

Plasma glucose (mg/dl) 89.3 ± 7.1

Hemoglobin A1c (%) 5.5 ± 0.2

Glomerular filtration rate (ml/min/1.73m2) 95.8 ± 22.6

Renal plasma flow (ml/min) 425.1 ± 144.3

Renal blood flow (ml/min) 699.5 ± 228.5

Filtration fraction 0.22 ± 0.04

Glomerular hydrostatic pressure 57.4 ± 5.5

Afferent arteriolar resistance (dyne · s · cm-5) 4010.3 ± 1964.1

Efferent arteriolar resistance (dyne · s · cm-5) 2560.5 ± 823.5

429

19

Table 2. Factors associated with glomerular filtration rate (GFR), renal plasma flow (RPF), 430

afferent arteriola resistance (Ra), and glomerular hydrostatic pressure (Pglo) in multiple 431

regression analysis. 432

433

Parameter GFR*1 RPF*2 Ra*3 Pglo

*4

β p β p β p β p

Age (years) -0.391 0.0053 -0.463 0.0019 0.365 0.0014 -0.422 0.0040

Gender (male =0, female = 1) 0.131 0.3233 0.017 0.9004 -0.001 0.9936 -0.071 0.6034

Body mass index (kg/m2) -0.148 0.2560 0.098 0.4535 -0.008 0.9389 0.158 0.2453

Systolic blood pressure (mmHg) 0.108 0.4173 0.218 0.1091 0.486 <0.0001 0.186 0.1815

Uric acid (mg/dl) *5 -0.314 0.0256 -0.270 0.0551 0.230 0.0403 -0.259 0.0838

R2 / p 0.370 / 0.0012 0.358 / 0.0017 0.597 / <0.0001 0.315 / 0.0057

434

*1 GFR (mL/min/1.73m2): glomerular filtration rate measured by inulin clearance 435

*2 RPF (mL/min): renal plasma flow measured by para-aminohippurate clearance 436

*3 Ra: afferent arteriolar resistance 437

*4 Pglo: glomerular hydrostatic pressure 438

*5 Dummy variables: 0 for uric acid 3.0 to 6.0 mg/dL, and 1 for uric acid <3.0 or >6.0 mg/dL 439

30

60

90

120

150

180

2.0 3.0 4.0 5.0 6.0 7,0 8.0

0

200

400

600

800

1000

2.0 3.0 4.0 5.0 6.0 7.0 8.0 0

0.2

0.4

0.6

2.0 3.0 4.0 5.0 6.0 7.0 8.0

GF

R m

L/m

in/1

.73m

2

Uric acid (mg/dL)

RP

F m

L/m

in

Uric acid (mg/dL) Uric acid (mg/dL)

FF

R2 = 0.350 p < 0.0001

R2 = 0.188 p = 0.0093

R2 = 0.012 p = 0.7652

Figure 1

Ra再計算

0

4000

8000

12000

2.0 3.0 4.0 5.0 6.0 7.0 8.0

Re再計算

0

2000

4000

6000

8000

2.0 3.0 4.0 5.0 6.0 7.0 8.0

pglo再計算

40

50

60

70

80

2.0 3.0 4.0 5.0 6.0 7.0 8.0

Affere

nt A

rteriola

r R

esis

tance

dyn・s・cm^-5

Effere

nt A

rteriola

r R

esis

tance

dyn・s・cm^-5

Pglo

mmHg

Uric acid (mg/dL) Uric acid (mg/dL) Uric acid (mg/dL)

R2 = 0.262 p = 0.0011

R2 = 0.026 p = 0.5549

R2 = 0.183 p = 0.0105

Figure 2