Kidney International, Vol. 63 (2003), pp. 2088–2093

Microvascular and tubulointerstitial injury associated withchronic hypoxia-induced hypertension

MARILDA MAZZALI, J. ASHLEY JEFFERSON, ZEHMIN NI, NOSRATOLA D. VAZIRI,and RICHARD J. JOHNSON

Division of Nephrology, Baylor College of Medicine, Houston, Texas; Division of Nephrology, State University of Campinas,Campinas, Brazil; Division of Nephrology, University of Washington, Seattle, Washington; and Division of Nephrology,University of California at Irvine, Irvine, California

Microvascular and tubulointerstitial injury associated with chronic hypertension [2]. However, recent studies have sug-hypoxia-induced hypertension. gested that the induction of preglomerular arteriolar dis-

Background. Rats submitted to chronic hypoxia develop hy- ease can lead to persistent hypertension, especially inpertension that persists despite cessation of the hypoxia orthe presence of a high salt diet [3]. The mechanism iscorrection of the hematocrit. We examined whether chronic hy-thought to be mediated by ischemia of the outer medullapoxia might induce subtle renal injury since studies in other

animal models of hypertension suggest this may cause persis- and juxtamedullary cortex, resulting in a local inflamma-tent hypertension. tory response that alters the local vasoconstrictive/vaso-

Methods. Chronic hypoxia was induced in rats by placementdilator balance favoring sodium retention [3]. This path-in a hypobaric chamber for up to 24 days. Blood pressure andway has been shown in a variety of animal models,kidney biopsies were performed at baseline, 6 hours, 24 hours,

and 24 days of hypoxia. including that induced by transient infusion of angioten-Results. Chronic hypoxia induced hypertension and erythro- sin II, l-arginine analog N-nitro-l-arginine methyl ester

cytosis at 24 days. Acute hypoxia was associated with endothe- (L-NAME) (a nitric oxide synthesis inhibitor), and cyclo-lial cell swelling in arterioles (6 and 24 hours), followed by thick-sporine [4–6].ening of the arterioles at 24 days. Subtle tubulointerstitial injury

Vaziri and Wang [7] recently reported a new modeland inflammation occurred and was progressive. The influx ofmacrophages increased steadily over the 24 days and was asso- of chronic hypertension induced by systemic hypoxiaciated with a progressive increase in interstitial collagen III using hypobaric chambers. Specifically, rats maintaineddeposition. Hypoxia was associated with increased tubular ex-

in a hypobaric chamber (at 1⁄2 atmosphere) for 24 dayspression of osteopontin as early as 6 hours, the same perioddevelop hypertension that persists despite correction ofwhen an increase of proximal tubular cell proliferation oc-

curred. Interstitial cell proliferation peaked twice, at 6 hours the polycythemia and return to normoxic conditions. Weand at 24 days. Glomerular hypertrophy was manifest at 24 days. hypothesized that the persistence of hypertension in this

Conclusion. Both afferent arteriolar disease and subtle tu- model might be due to the development of microvascularbulointerstitial inflammation and injury occur early in hypoxicand renal injury. We therefore examined the renal tissuesrats. These changes may predispose these animals to persistent

hypertension. of rats maintained in hypoxic chambers at different timepoints up to 24 days.

Goldblatt [1] suggested that the pathogenesis of essen-METHODStial hypertension is due to primary disease of the renal

Male Sprague-Dawley rats (Charles River, Wilming-arterioles (arteriolosclerosis), and that this would leadton, MA, USA) weighing 220 to 320 g were used in theseto renal ischemia and an increase in blood pressure. Thisexperiments. Rats were allowed free access to food (Pur-hypothesis was largely discarded because many investi-ina Mills, Inc., Brentwood, MO, USA) and water duringgators considered the renal changes to be secondary tothe study period. The animals were placed in a hypobaricchamber in which the air pressure was kept at 390 mm HgKey words: hypoxia, arteriolosclerosis, glomerular hypertrophy, hyper-

tension, interstitial disease. using a continuous vacuum pump and an adjustable in-flow valve. The interior of the chamber was maintained

Received for publication August 6, 2002at ambient air temperature. A normal interior light cycleand in revised form November 15, 2002, and December 18, 2002

Accepted for publication January 17, 2003 was accommodated through glass windows in the cham-ber’s structure, as previously described [7]. The hypoxic 2003 by the International Society of Nephrology

2088

Mazzali et al: Hypoxia-induced hypertension 2089

group was kept under hypobaric conditions for 6 hours with the combination of periodic acid and Schiff’s re-agent, and proximal tubules identified by the presence(group 6 hours), 24 hours (group 24 hours) and 24 days

(group 24 days), respectively. A control group comprised of brush border.eight rats sacrificed at baseline (control).

Quantification of morphologyArterial blood pressure was measured using a tail sphyg-momanometer (Harvard Apparatus, South Natick, MA, All quantification was performed blinded.USA) at baseline (day 0) and at various time points dur-

Afferent arteriolopathying the study. Blood samples were obtained by orbitalsinus puncture under light anesthesia at appropriate in- Only vessels adjacent to glomeruli in the outer cortex

were selected. Afferent arterioles were distinguishedtervals. Serum creatinine was measured by a kit fromSigma Chemical Co. (St. Louis, MO, USA), according from efferent arterioles by the presence of an internal

elastic lamina and by thin, flattened endothelial cells.to manufacturer’s protocol. Serum uric acid was mea-sured by a modified phosphotungstic acid method [8]. Using immersion fixed tissue, afferent arteriolar wall

thickness was measured by computer image analysis byRenal histology a blinded observer. For each arteriole, the outline of the

vessel and its internal lumen (excluding the endothe-Renal biopsies obtained at sacrifice were fixed in methyl-carnoys, 10% formalin, or Bouin’s fixative and embed- lium) was generated using computer analysis to calculate

the total medial area (outline � inline), in a minimumded in paraffin. Sections (4 �m) of methyl-Carnoy fixedtissue were stained with periodic acid-Schiff (PAS) re- of 12 arterioles per biopsy. Vessels that were cross-sec-

tioned or not sectioned transversally, providing an asym-agent or analyzed by indirect immunoperoxidase stainingwith the following primary antibodies: goat anti-osteo- metrical wall, were excluded from the present study [9].

Tubulointerstitial injury score (0 to 5) was based on PASpontin (OP199, a gift from C. Giachelli from the Univer-sity of Washington), monoclonal mouse anti-rat macro- staining as previously described [10].phage and monocyte (ED-1) (Serotec, Haleigh, NC, USA),

Glomerular hypertrophygoat anti-human type III collagen (Southwestern Bio-tech, Birminghan, AL, USA), monoclonal mouse anti- Glomerular tuft area in the outer cortex was measured

by computer image analysis (Optimas V6.2, Media Cy-alpha smooth muscle actin (�-SMA) (IA4, Sigma Chemi-cal Co.) and mouse monoclonal antibody to a 70 kD bernetics, Silver Springs, MD, USA). Only glomeruli

with an intact vascular pole were considered for analysis.antigen expressed by rat endothelial cells (JG-12, a giftfrom Dontscho Kerjaschki from the University of Vi- Tubular expression of osteopontin (OPN) was calcu-

lated as the percent of renal cortex and medulla occupiedenna). Sections were incubated with a secondary anti-body followed by horseradish peroxidase–conjugated av- by OPN-positive tubules. Utilizing computer image anal-

ysis, the percent of area occupied by OPN-positive tu-idin D (Vector Laboratories, Burlingame, CA, USA),diaminobenzidine (DAB) (Sigma Chemical Co.) with or bules per 4 mm2 field at a magnification of �50 was

measured and the median percent area calculated forwithout nickel chloride as a chromogen, and counter-stained with methyl green. An irrelevant primary anti- each biopsy [8].

Cortical and medullary macrophage infiltration werebody of the same isotype was used as a negative control.measured by counting the number of ED-1–positive cells

Detection of proximal and distal tubular in selected 0.01 mm2 grids (20 in the cortex and 10 incell proliferation the medulla) at �200 magnification. The median number

of macrophages per biopsy was calculated and correctedTo detect cell proliferation, each rat was injected with10 �L/g body weight 5-bromo-2�-deoxyuridine (BrdU) to a final value of ED-1–positive cells/mm2.

Cortical and medullary deposition of type III collagen4 hours before sacrifice. Kidney tissue sections under-went an antigen retrieval step, which comprised boiling were measured by computer image analysis, at �50 mag-

nification, and expressed as the percentage of renal cor-in 0.01 mol/L sodium citrate buffer for a total of 10 min-utes. Sections were then incubated with a murine immu- tex or medulla occupied by collagen III.

Proximal and distal tubular cell proliferation was mea-noglobulin G2a (IgG2a) monoclonal antibody againstBrdU (Cell Proliferation Kit, Amersham Pharmacia Bio- sured by counting the number of BrdU-positive cells

(distinguishing between proximal and distal segments, bytech, Bucks, UK) at room temperature for 60 minutes,followed by a peroxidase-conjugated goat anti-mouse PAS counterstaining) in 20 sequentially selected 0.1 mm2

grids, in the renal cortex, at �200 magnification, andIgG2a at room temperature for 30 minutes. A blackcolor was developed using DAB with nickel chloride expressed as the number of cells per mm2. Cortical and

medullary interstitial cell proliferation was measured byas the chromogen. In order to identify and accuratelyquantify proximal versus distal tubular epithelial cell counting the number of BrdU-positive cells in 20 sequen-

tially selected 0.1 mm2 grids in renal cortex and 10 inproliferation and apoptosis, slides were counterstained

Mazzali et al: Hypoxia-induced hypertension2090

for alpha actin, the arteriolar smooth muscle cells ap-peared normal during the first 24 hours; however, anincrease in �-SMA staining was observed at 24 daysconsistent with the increased arteriolar thickness mea-surements (Table 2). An increase in BrdU-positive cellswas not appreciated in the arterioles, but an increase inBrdU-positive cells was observed in the interlobular andarcuate arteries at 24 days.

Hypoxia was associated with the development of earlyand subtle interstitial inflammation

Despite the normal appearance of kidneys by PASstaining at 24 hours and 24 days, careful examinationshowed subtle tubulointerstitial inflammation. A signifi-cant influx of macrophages was noted by 24 hours inboth the cortex and medulla and was further increased

Fig. 1. Time course of systolic blood pressure in rats exposed to sys- at 24 days (Table 3). OPN, a macrophage adhesive factor,temic hypoxia. Hypertension was noted as early as 4 days, and increased was increased in both cortical and medullary tubulesprogressively thereafter. Symbols are: (�), control; (�), hypoxia. *P �

as early as 6 hours. The presence of macrophages was0.05 vs. control.associated with an increase in the interstitial type IIIcollagen deposition at 24 days (Table 3).

Tubular and interstitial cell proliferation occurs as anrenal medulla, at �200 magnification, and expressed asearly response to hypoxiathe number of cells per mm2.

The number of proliferating BrdU-positive cells wasStatistical analysis significantly increased in both proximal and distal tubu-

Data are expressed in mean � standard deviation. lar segments following 6 hours of hypoxia (Table 4).Groups were compared by analysis of variance (ANOVA) However, while the number of positive distal cells re-with appropriate corrections for multiple comparisons. turned toward normal levels after 24 hours of hypoxia,Statistical significance was considered if P � 0.05. the proximal tubular cell proliferation remained in-

creased at 24 hours. Interstitial cell proliferation peakedat 6 hours. The specific type of interstitial cells that wereRESULTSproliferating was not identified (Table 4).

Chronic hypoxia induces hypertension, erythrocytosis,and hyperuricemia in rats

DISCUSSIONRats maintained in hypobaric hypoxic chamber wereIn this study, we have investigated the renal changeshypertensive by 4 days, with blood pressure increasing

that occur with chronic hypoxia-associated hypertension.progressively thereafter (Fig. 1). Hematocrit was ele-Our major findings were that hypoxia was associatedvated at 24 days. Serum uric acid was increased as earlywith the development of glomerular hypertrophy, anas 24 hours, and was further increased at 24 days. Noarteriolopathy of the afferent arteriole, and mild tubulo-changes in renal function were observed, as reflected byinterstitial inflammation and fibrosis. Accompanying theserum creatinine (Table 1).tubulointerstitial inflammation was a modest prolifera-

Chronic hypoxia is associated with afferent tion of proximal and distal tubular cells.arteriolopathy and glomerular hypertrophy The first changes noted at 6 hours were primarily con-

fined to the tubulointerstitium, with an increase in tubu-Chronic hypoxia was associated with the developmentof glomerular hypertrophy and arteriolar thickening that lar cell proliferation, OPN expression, and early macro-

phage infiltration. These inflammatory and proliferativewas manifest at 24 days compared to controls (Fig. 2 andTable 2). The afferent arterioles showed an increase in changes occurred before any change in blood pressure,

serum creatinine, or serum uric acid. It is likely that thesemedial afferent arteriolar area, suggesting the occurrenceof vascular remodeling. Acute hypoxia (6 and 24 hours) changes are mediated by hypoxia. It is known that the

outer medulla of the kidney is particularly vulnerable towas associated with some endothelial swelling in the ar-terioles (as demonstrated by prominent cytoplasm when ischemic injury as it is normally borderline hypoxic due

to the countercurrent circulation and the high metabolicendothelial cells were stained with the JG-12 antibody)that resolved by day 24 (Fig. 2). In contrast, when stained demands of the tubules (medullary thick ascending limb

Mazzali et al: Hypoxia-induced hypertension 2091

Table 1. Chronic hypoxia induces hypertension, erythrocytosis, and hyperuricemia

Baseline 6 hours 24 hours 24 days

Number of rats 8 8 8 8Blood pressure mm Hg 123�4.3 121�4.5 126�3.9 195�7.4a,b,c

Hematocrit % 47.5�1.1 46.2�0.5 47.6�0.7 67.5�2.6a,b,c

Serum creatinine mg/dL 0.69�0.13 0.75�0.13 0.68�0.22 0.86�0.22Serum uric acid mg/dL 0.98�1.0 1.04�0.8 2.0�0.42a,b 3.21 �1.0a,b,c

a P � 0.05 vs. baseline; b P � 0.05 vs. 6 hours; c P � 0.05 vs. 24 hours

Fig. 2. Glomerular and arteriolar changes induced by hypoxia. Shown are periodic acid-Schiff (PAS)-stained sections of glomeruli at baseline(A), 24 hours (B), and 24 days (C ), and sections stained for endothelial cells with the JG-12 antibody at baseline (D), 24 hours (E), and 24 days(F ) of hypoxia. Glomerulomegaly is present at 24 days (C). Arteriolar changes (arrows) consist of early endothelial cell swelling (E) and endothelialprominence (B) at 24 hours, followed by arteriolar thickening at 24 days (C).

Table 2. Chronic hypoxia is associated with afferent arteriolopathy and glomerular hypertrophy

Baseline 6 hours 24 hours 24 days

Glomerular tuft area lm2 7659.1�467.5 7295.9�474.1 7391.2�224.1 8730.6�692.4a,b,c

Afferent arteriolar area lm2 119.2�7.4 127.0�14.7 129.6�5.4 147.3�11.9a,b,c

aP � 0.05 vs. baseline; b P � 0.05 vs. 6 hours; c P � 0.05 vs. 24 hours

and S3 segment of the proximal tubule) [11]. In turn, tubular cell proliferation in response to hypoxia. Impor-tantly, the early renal changes observed at 6 and 24 hourshypoxia is known to be a potent stimulus for tubular

and interstitial cell proliferation in vitro and for OPN cannot be ascribed to the effects of hypertension as theyprecede its development.expression [12–14]. It is interesting that the OPN expres-

sion correlated with the tubular cell proliferation, as The most evident glomerular change that was notedwas a marked increase in glomerular tuft size at 24 days.Sahai et al [15] have reported that OPN mediates the

Mazzali et al: Hypoxia-induced hypertension2092

Table 4. Tubular and interstitial cell proliferation occurs asTable 3. Hypoxia was associated with the development of early andsubtle interstitial injury in cortical and medullary areas an early response to hypoxia

Cortex Baseline 6 hours 24 hours 24 daysBaseline 6 hours 24 hours 24 days

Cortex Interstitial cells 3.5�0.7 16.7�3.3a 2.6 �0.8b 9.3�5.2b

Proximal tubule 1.7�0.9 14.1�3.7a 10.8 �3.3a 3.8�1.5b,cPAS score (TI) 0.05�0.01 0.15�0.05 0.10�0.05 0.30�0.18OPN % area 0.46�0.09 0.75�0.19a 0.53�0.10b 0.71�0.12a,c Distal tubule 2.8�1.1 7.8�2.3a 4.6 �2.1b 1.6�0.8b,c

Medullary tubules 8.6�0.2 58.6�10.6a 20.8 �5.7b 15.3 �5.3bED-1/mm2 27.8�2.8 49.2�4.2a 58.4�0.6a,b 70.3�2.7a,b,c

Col III % area 1.5�0.07 1.9�0.1 2.1�0.03 2.3�0.5aa P � 0.05 vs. baseline; b P � 0.05 vs. 6 hours; c P � 0.05 vs. 24 hours

MedullaOPN % median 2.9�0.4 3.9�0.6a 3.7�0.8b 5.0�0.8a,b,c

ED-1/mm2 39.0�1.4 52.8�15.8 78.4�3.4a 93.6�9.0a,b

Col III % area 3.6�0.4 4.7�1.4 5.2�0.7 5.7�0.8a

proliferation. In addition, hypoxia itself can induce vas-Abbreviations are: PAS, periodic acid-Schiff; OPN, osteopenia; Col III, colla-gen type III. cular smooth muscle cell proliferation [21].

a P � 0.05 vs. baseline; b P � 0.05 vs. 6 hours; c P � 0.05 vs. 24 hoursWhile the arteriolopathy and interstitial fibrosis noted

at 24 days are likely the consequence of the systemichypoxia, it is not clear if they represent the direct effectsof hypoxia or whether they are secondary to the effects ofIt is known that persons exposed to chronic hypoxiahypoxia to induce hypertension, polycythemia, or even(such as patients with congenital cyanotic heart diseasealterations in systemic pH. An additional possibility isand subjects chronically living at high altitude) developthat the increase in uric acid that occurs with hypoxiaglomerulomegaly [16, 17]. The mechanism for this hyper-may be involved in the renal changes. Recently, we havetrophic response is not known, but may relate to in-reported that mild hyperuricemia in rats induces hyper-creases in glomerular hydrostatic pressure in associationtension, glomerular hypertrophy, mild tubulointerstitialwith the increase in filtration fraction that occurs withinflammation and fibrosis, and arteriolopathy [9, 22–24].erythrocytosis [18].We have also reported that subjects chronically exposedOne of the more striking findings was the thickeningto high altitude develop erythrocytosis, hypertension,of the afferent arteriole. The earliest changes involve theand hyperuricemia, and evidence of mild renal injury (asvascular endothelium, often with encroachment of themanifested by microalbuminuria) [25]. The mechanismcapillary lumen. However, over time the most prominentby which uric acid induces these changes appears tochange was that of thickening of the afferent arterioles,involve activation of the renin-angiotensin system andwhich by immunostaining appeared to be due to an in-direct stimulatory effects of uric acid on vascular smoothcrease in �-actin–positive smooth muscle cells.muscle cells [9, 22–24].The thickening of the afferent arteriole, coupled with

the tubulointerstitial changes, might have a role in thepersistence of hypertension in this animal model [7]. We CONCLUSIONhave recently shown that similar structural changes can We have documented that chronic hypoxia can inducepredispose animals to persistent hypertension in other microvascular (arteriolopathy), glomerular (hypertro-experimental models [3]. Thus, in models of renal injury phy), and tubulointerstitial (inflammation and fibrosis)such as that induced by infusing angiotensin II, we were changes in the kidney. These changes may predisposeable to show that the development of preglomerular these animals to the persistent hypertension that theyarteriolopathy and tubulointerstitial inflammation pre- are known to develop. This could be potentially relevantdispose animals to salt-sensitive hypertension even after to humans at high altitude who develop hypertensionthe angiotensin II infusion has been stopped [4]. It is [25]. Further studies are necessary to investigate thethus of interest that Vaziri and Wang [7] have shown mechanisms involved in chronic hypoxia-induced renalthat hypertension is also persistent for up to 84 days injury, and the consequences of these renal structural

changes on long-term blood pressure control and saltin this model after discontinuing use of the hypobaricsensitivity.chamber. This hypoxia-induced hypertension was not

due to erythrocytosis since it occurred in iron-deficientACKNOWLEDGMENTSanimals, which showed no rise in erythrocyte mass [7].

The initial endothelial cell swelling raises the possibil- Support for this project was provided by NIH DK-47659 andHL-68607. Dr. Mazzali was supported by a postdoctoral grant fromity of release of factors from the endothelium that couldFAPESP (Fundacao de Amparo a Pesquisa do Estado de Sao Paulo,influence smooth muscle cell proliferation [19]. A local Brazil).

loss of nitric oxide and an increase in endothelin hasReprint requests to Richard J. Johnson, M.D., Division of Nephrol-also been shown in the kidneys of rats in this model

ogy, Baylor College of Medicine, One Baylor Plaza, Alkek N530, Hous-before the establishment of hypertension [20], and this ton, Texas 77030.E-mail: rjohnson@bcm.tmc.eduimbalance would also favor vascular smooth muscle cell

Mazzali et al: Hypoxia-induced hypertension 2093

tubular cell matrix production via a TGF-beta1-independent mech-REFERENCESanism. Kidney Int 52:637–647, 1997

14. Kleinman JG, Worcester EM, Beshensky AM, et al: Upregula-1. Goldblatt H: The renal origin of hypertension. Physiol Rev 27:120–165, 1947 tion of osteopontin expression by ischemia in rat kidney. Ann N

Y Acad Sci 76:321–323, 19952. Perera GA: Hypertensive vascular disease: Description and natu-ral history. J Chron Dis 1:33–42, 1955 15. Sahai A, Mei C, Schrier RW, Tannen RL: Mechanisms of chronic

hypoxia-induced renal cell growth. Kidney Int 56:1277–1281, 19993. Johnson RJ, Herrera-Acosta J, Schreiner GF, Rodriguez-Iturbe B: Subtle acquired renal injury as a mechanism of salt 16. Faustinella F, Uzoh C, Sheikh-Hamad D, et al: Glomerulomeg-

aly and proteinuria in a patient with idiopathic pulmonary hyper-sensitive hypertension. N Engl J Med 346:913–923, 20024. Franco M, Tapia E, Santamaria J, et al: Renal cortical vasocon- tension. J Am Soc Nephrol 8:1966–1970, 1997

17. Winterborn MH, Bradwell AR, Chesner IM, Jones GT: Thestriction contributes to development of salt-sensitive hypertensionafter angiotensin II exposure. J Am Soc Nephrol 12:2263–2271, 2001 origin of proteinuria at high altitude. Postgrad Med J 63:179–

181, 19875. Quiroz Y, Pons H, Gordon KL, et al: Mycophenolate mofetilprevents salt-sensitive hypertension resulting from nitric oxide syn- 18. Thorn CD, Cheng J, Leiter JC, Ou LC: Renovascular adapta-

tive changes in chronic hypoxia polycythemia. Kidney Int 54:2014–thesis inhibition. Am J Physiol Renal Physiol 281:F38–F47, 20016. Andoh TF, Johnson RJ, Lam T, Bennett WM: Subclinical renal 2020, 1998

19. Brij SO, Peacock AJ: Cellular response to hypoxia in the pulmo-injury induced by cyclosporine exposure is associated with salt-sensitive hypertension. Am J Transplant 1:222–227, 2001 nary circulation. Thorax 53:1075–1079, 1998

20. Ni Z, Bemanian S, Kivlignh SD, Vaziri ND: Role of endothelin7. Vaziri ND, Wang ZQ: Sustained systemic arterial hyperten-sion induced by extended hypobaric hypoxia. Kidney Int 49:1457– and nitric oxide imbalance in the pathogenesis of hypoxia-induced

arterial hypertension. Kidney Int 54:188–192, 19981463, 19968. Mazzali M, Kim YG, Suga S, et al: Hyperuricemia exacerbates 21. Sodhi CP, Phadke SA, Battle D, Sahai A: Hypoxia stimulates

osteopontin expression and proliferation of cultured vascularchronic cyclosporine nephropathy. Transplantation 71:900–905, 20019. Mazzali M, Kanellis J, Han L, et al: Hyperuricemia induces a smooth muscle cells: Potentiation by high glucose. Diabetes 50:

1482–1490, 2001primary arteriolopathy in rats by a blood pressure independentmechanism. Am J Physiol Renal Physiol 282:F991–F997, 2002 22. Mazzali M, Hughes J, Kim YG, et al: Elevated uric acid increases

blood pressure in the rat by a novel crystal independent mechanism.10. Kang D-H, Kim Y-G, Andoh TF, et al: Post-cyclosporine-medi-ated hypertension and nephropathy: Amelioration by vascular Hypertension 38:1101–1106, 2001

23. Watanabe S, Kang D-H, Feng L, et al: Uric acid, hominoid evolu-endothelial growth factor. Am J Physiol Renal Physiol 280:F727–F736, 2001 tion and the pathogenesis salt-sensitivity. Hypertension 40:355–

360, 200211. Epstein FH: Oxygen and renal metabolism. Kidney Int 51:381–385, 1997 24. Nakagawa T, Mazzali M, Kang D-H, et al: Hyperuricemia causes

glomerular hypertrophy in the rat. Am J Nephrol 23:2–7, 200312. Fine LG, Orphanides C, Norman JT: Progressive renal disease:The chronic hypoxia hypothesis. Kidney Int 53(Suppl 65):S74– 25. Jefferson JA, Escudero E, Hurtado ME, et al: Hyperuricemia,

hypertension and proteinuria associated with high-altitude poly-S78, 199813. Orphanides C, Fine LG, Norman JT: Hypoxia stimulates proximal cythemia. Am J Kidney Dis 39:1135–1142, 2002