Kidney International, Vol. 64, Supplement 86 (2003), pp. S27–S32

Apoptosis and NF�B activation are simultaneously induced inrenal tubulointerstitium in experimental hypertension

YASMIR QUIROZ, JANAURY BRAVO, JAIME HERRERA-ACOSTA, RICHARD J. JOHNSON, andBERNARDO RODRIGUEZ-ITURBE

Renal Service and Laboratory, Hospital Universitario and Instituto de Investigaciones Biologicas (INBIOMED),FUNDACITE-Zulia, Maracaibo, Venezuela; Department of Nephrology, Instituto Nacional de Cardiologıa“Ignacio Chavez,” Mexico, D.F.; and Division of Nephrology, Baylor Medical College, Houston, Texas

Apoptosis and NF�B activation are simultaneously induced inrenal tubulointerstitium in experimental hypertension.

Background. Oxidative stress is known to induce apoptosisand activation of pro-inflammatory transcription factor nuclearfactor kappa B (NF�B), which are biologic effects that mayplay a role in the renal damage associated with arterial hyper-tension. We investigated if increased apoptosis and NF�B acti-vation were present in experimental models of hypertension.

Methods. Sprague-Dawley rats fed with regular rodent chowand free access to water were studied. The Ang II group (N �6) received 435 ng/kg/min of angiotensin II during 2 weeks bysubcutaneous minipumps. The l-NAME group (N � 5) receivedN�-nitro-l-arginine-methyl-ester (l-NAME) in the drinking wa-ter (70 mg/100 mL) for 3 weeks. The control group consisted of6 rats. Systolic blood pressure (tail cuff plethysmography), serumcreatinine, and proteinuria were determined weekly. Kidneyswere examined for superoxide-positive cells (histochemistry)and for apoptosis [terminal deoxynucleotidyl transferase-medi-ated uridine triphosphate nick-end labeling (TUNEL)-positivecells], proliferation [proliferating cell nuclear antigen (PCNA)-positive cells), and activation of NF�B (p65 subunit) with theappropriate antibodies.

Results. As expected, hypertension developed in experimen-tal groups. Tubulointerstitial superoxide-positive cells were in-creased 7 times (P � 0.001), TUNEL-positive cells were in-creased 3 to 4 times (P � 0.001), PCNA-positive cells wereincreased 20 to 30 times (P � 0.001), and NF�B activation wasincreased 4 to 5 times (P � 0.001) in the experimental groups.NF�B expression correlated with the number of interstitiallymphocytes (r � 0.667, P � 0.01) and macrophages (r � 0.835,P � 0.001).

Conclusion. Angiotensin II infusion and l-NAME adminis-tration induce oxidative stress and increased apoptosis andactivation of the transcription factor NF�B. These effects mayparticipate in the development of progressive renal injury re-sulting from uncontrolled hypertension

Immunocompetent cells are involved in the mecha-nisms of progression of renal disease of immune-medi-ated as well as non-immune renal injury [1]. In most, if

2003 by the International Society of Nephrology

S-27

not all, models of arterial hypertension, there is infiltra-tion of lymphocytes and macrophages in the renal tubu-lointerstitium and recent investigations have shown thatreduction of this infiltrate prevents or ameliorates theblood pressure increment induced by a high-salt diet[2–5]. A constant feature of these experimental modelshas been the demonstration of increased oxidative stressin the kidney, a condition that results from the combinedeffects of renal vasoconstriction and interstitial inflam-mation [6–9].

The interrelation between oxidative stress and intersti-tial infiltrate has been emphasized by investigations thatshow that antioxidant strategies, which are known to im-prove hypertension in spontaneously hypertensive rats[10–12], are associated with a reduction in the infiltrationof lymphocytes and macrophages in tubulointerstitialareas [13, 16], and furthermore, that the intensity of theimmune infiltration and the degree of the oxidative stresscorrelate with the severity of hypertension [16].

The present work focuses on cellular events derivedfrom the increased oxidative stress in models of experi-mental hypertension. Increased generation of reactiveoxygen species (ROS) is known to trigger mechanisms ofprogrammed cell death, as well as inflammatory reactiv-ity, both of which could have considerable relevance inthe progressive deterioration of renal function that isassociated with uncontrolled hypertension. Therefore,we examined apoptosis and expression of activated nu-clear factor kappa B (NF�B) in the kidney in the experi-mental models of hypertension induced by chronic an-giotensin II infusion and nitric oxide synthase (NOS)inhibition.

METHODS

We have examined renal tissue from rats with salt-sensitive hypertension described in previous communica-tions [2, 3]. Male Sprague-Dawley rats (Instituto Vene-

Quiroz et al: Apoptosis and NFjB in hypertensionS-28

Fig. 1. After 2 weeks of angiotensin II (Ang II) infusion (black bar)and 3 weeks of N�-nitro-L-arginine-methyl-ester (L-NAME) administra-tion (gray bar), the blood pressure is significantly elevated. Values aremean � SD. ***P � 0.001.

zolano de Investigaciones Cientıficas, Los Teques, Vene-zuela) receiving a standard rat feed (Protinal, Valencia,Venezuela) were used in these studies. The followingexperimental groups were studied: (1) Ang II group(N � 6), constituted by rats treated with angiotensin II(Sigma Chemical Co., St Louis, MO, USA) for 2 weeks,administered at a rate of 435 ng�1 · kg�1 · min�1 by subcu-taneous minipumps (Alza Corp., Palo Alto, CA, USA);(2) N�-nitro-l-arginine-methyl-ester (l-NAME group)(N � 5) that received l-NAME in the drinking water(70 mg/100 mL) for 3 weeks; and (3) the Control group,which consisted of 6 rats. Systolic blood pressure (SBP)was determined weekly by tail-cuff plethysmography(IITC, Life Scientific Instruments, Woodland Hills, CA,USA). All rats were preconditioned to the procedureof blood pressure determination. Serum creatinine andproteinuria were determined weekly. At the end of theexperiments, rats were sacrificed under penthotal anes-thesia and the kidneys were harvested for microscopicstudies.

Histology and immunohistology

Light microscopy was used to study glomerular andtubulointerstitial damage using a semiquantitative glo-merular sclerosis index [17] and a tubulointerstitial dam-age score previously reported [2–4].

Cellular infiltration was evaluated by indirect immu-nohistology using monoclonal antibodies. Lymphocyteswere identified with anti-CD5 antibody (clone MRCOX19;Biosource, Camarillo, CA, USA) and macrophages withanti-ED1 antibody (Harlan Bioproducts, Indianapolis,IN, USA). Secondary antibodies were purchased fromAccurate Chemical and Scientific Co. (Westbury, NY,USA). Results were expressed as positive cells per glo-

Fig. 2. Superoxide-positive cells in tubulointerstitial areas of rats ofangiotensin II (Ang II) group (black bar), N�-nitro-L-arginine-methyl-ester (L-NAME) group (gray bar), and control (open bar). Values aremean � SD. ***P � 0.001.

merular cross section (gcs) and positive cells/mm2 in tu-bulointerstitial regions.

Superoxide production in renal cells was studied incryostat sections by the cytochemical method of Briggs[18] as previously described [19].

Frozen sections were used to detect apoptosis by theterminal deoxynucleotidyl transferase (TdT)-mediateddUTB-biotin nick-end labeling methodology (TUNEL)with commercial apoptosis detection kits (ApoTag;ONCOR, Gaithersburg, MD, USA). Proliferating cellswere identified with monoclonal antibody to proliferat-ing cell nuclear antigen (PCNA) obtained from ZymedLaboratories (San Francisco, CA, USA) utilizing avidin-biotin-peroxidase methodology. Specific details fromthese methods in our laboratories have been reportedpreviously [19, 20].

Activated NF�B was identified in 4-micron cryostatsections by immunoperoxidase methodology using anti-body specific for the 65 kD DNA-binding subunit (Zy-med Laboratories) as detailed in earlier work [15].

Statistical analysisComparisons between groups were done by analysis

of variance (ANOVA) and Tukey-Kramer post tests. Re-sults are expressed as mean � SD throughout the paperand two-tailed P � 0.05 were considered significant.

RESULTSSystolic blood pressure increased sharply as a result of

Ang II infusion and l-NAME administration. The bloodpressure levels observed at the end of the experiment areshown in Figure 1. Serum creatinine and urinary proteinexcretion increased slightly during the experiment butdid not exceed 0.7 mg/dL and 10 mg/day, respectively.

Superoxide-positive cells were seldom seen in the glo-meruli. Figure 2 shows that the accumulation of super-

Quiroz et al: Apoptosis and NFjB in hypertension S-29

Fig. 3. Apoptosis [terminal deoxynucleotidyl transferase-mediateduridine triphosphate nick-end labeling (TUNEL)-positive cells/mm2])(open bar) and proliferating cells [proliferating cell nuclear antigen(PCNA)-positive cells/mm2] (black bar) in tubulointerstitial areas incontrol, angiotensin II (Ang II) group, and N�-nitro-L-arginine-methyl-ester (L-NAME) group. ***P � 0.001 for all data in the correspondinggroup vs. the control group.

Fig. 4. Tubulointerstitial activation of nuclear factor kappa B (NF�B),represented as cells positive for the 65p subunit per mm2 (open bars),and accumulation of lymphocytes (CD5-positive cells/mm2) (black bars)and macrophages (ED1-positive cells/mm2) (striped bars) in experi-mental angiotensin II (Ang II) group, N�-nitro-L-arginine-methyl-ester(L-NAME) group and control group. Values are mean � SD. ***P �0.001 for all the data in each group vs. control group.

oxide-positive cells in both of the experimental groupswas more than 7 times that found in controls.

Apoptosis (TUNEL-positive cells) and proliferation(PCNA-positive cells) were essentially absent in glomer-uli but they were increased in tubulointerstitium 34 times(Ang II group) and 24 times (l-NAME group) as shownin Figure 3.

Expression of the 65 subunit of NF�B was limited alsoto the tubulointerstitial areas and it is similarly increasedin both experimental groups. Figure 4 shows these find-ings in association with the data showing accumulationof CD5-positive cells and ED1-positive cells. There weresignificant correlations between NK�B expression andthe number of interstitial lymphocytes (CD5-positive cells)

Fig. 5. Relationship (r � 0.835, P � 0.001) between the expression ofp65 nuclear factor kappa B (NF�B) and macrophage infiltration (ED1-positive cells. Control group (open circles); angiotensin II (Ang II)(black squares); N�-nitro-l-arginine-methyl-ester (l-NAME) group(black circles).

and (r � 0.667, P � 0.01) macrophages (r � 0.835, P �0.001) (Fig. 5). Representative microphotographs areshown in Figure 6.

DISCUSSIONThe association between oxidative stress and hyper-

tension is well recognized and has been demonstrated ingenetic [21, 22] as well as acquired forms of hypertension[22–28]. The pro-hypertensive effects of ROS are mainlydue to inactivation of nitric oxide and, thereby, endothe-lial dysfunction with impaired vasodilation. Since ROSgeneration is not increased in norepinephrine-inducedhypertension, the elevation of the blood pressure by itselfdoes not increase oxidant stress [29]. Several studies haveindicated that at least in the angiotensin II–dependentand angiotensin II–sensitive forms of hypertension, theformation of oxygen radicals in the vascular endotheliumis the result of angiotensin II activity and is mediatedby the NAD(P)H oxidase system [29–32].

An additional cause of increased ROS production, atleast in the kidney, is the interstitial inflammatory pro-cess that is a feature of conditions associated with salt-sensitive hypertension [6]. Lymphocyte and macrophageinfiltration, as well as evidence of oxidative stress, hasbeen found, among others, in the angiotensin II infusionmodel [2] and the l-NAME model [3]. Furthermore,recent evidence shows significant correlations betweenoxidative stress, immune cell infiltration, and blood pres-sure in the protein overload model [4], and in spontane-ously hypertensive rats treated with an antioxidant-richdiet [16]. In the present work, the tubulointerstitial in-filtration of lymphocytes and macrophages and the in-creased number of superoxide-positive cells is also dem-onstrated.

Quiroz et al: Apoptosis and NFjB in hypertensionS-30

Fig. 6. Microphotographs showing apoptosis[terminal deoxynucleotidyl transferase-mediateduridine triphosphate nick-end labeling (TUNEL)-positive cells (A and B ), proliferating cells [pro-liferating cell nuclear antigen (PCNA)-positivecells] (C and D ), and activated nuclear factorkappa B (NF�B) (p65-positive cells) (E and F )in biopsies from the control (A, C, and E) andexperimental group (B, D, and F).

Oxidative stress is a well-known stimuli of both apo-ptosis [33] and activation of NF�B [34], which under-scores the biological need to interrelate cellular prolif-eration with programmed cell death. In our experimentsit was demonstrated that NF�B, which is a rapid responsetranscription factor of proinflammatory genes [35], is ac-tivated by angiotensin II infusion as well as by l-NAMEadministration. In addition, there were significant cor-relations between the expression of the NF�B p65 sub-unit and the number of interstitial lymphocytes and mac-rophages (Fig. 5). NF�B mediates the synthesis of avariety of cytokines and, in addition, favors leuko-cyte transmigration since it mediates an increase in theexpression of E-selectin, Vascular cellular adhesion mole-cule-1 (VCAM-1) and intercellular adhesion molecule-1

(ICAM-1) [36, 37]. Consistent with the role of NF�Bin the pathogenesis of the inflammatory infiltration inhypertension is the recent demonstration that antioxi-dant treatments reduce NF�B activity, and inflammationin double transgenic rats for human and angiotensinogengenes [13] and in DOCA-salt hypertensive rats [14].

Therefore, NF�B is likely involved in the pathogenesisof the interstitial immune infiltration, possibly in associa-tion with other pathways that promote cell adhesion andproliferation that are activated by angiotensin II andl-NAME administration, such as the small GTPase Rhopathway [38, 39].

The possible effect of NF�B on apoptosis in the experi-mental conditions of this paper is less clear. The bestrecognized effect of NF�B is anti-apoptotic as a result

Quiroz et al: Apoptosis and NFjB in hypertension S-31

of its ability to suppress activation of caspase-8 [40].Furthermore, NF�B is involved in the mechanism ofplatelet-derived growth factor (PDGF) signaling [41],which is a survival factor. On the other hand, recentevidence indicates that in certain cell types NF�B maybe pro-apoptotic, since its activation may induce FasLexpression [42].

CONCLUSION

The increase in apoptosis and activation of NF�B dem-onstrated in experimental models of hypertension maybe relevant to the mechanisms of salt-sensitivity in hyper-tension. The loss of peritubular capillaries, which wouldimpair pressure natriuresis, and the interstitial immuneinfiltration are part of the pathogenesis of this condition[8, 9]. In addition, the interstitial infiltration of immunecells may play a determinant role in renal disease pro-gression [1] and, consequently, these events may be rele-vant to the risk of chronic renal failure associated withlong-term arterial hypertension.

ACKNOWLEDGMENTS

Investigations in Dr. Rodrıguez-Iturbe’s laboratory were supportedby grant S1-2001001097 of FONACYT, Venezuela, and from Asocia-cion de Amigos del Rinon, Maracaibo, Venezuela.

Reprint requests to Bernardo Rodrıguez-Iturbe, Sociedad Latinoam-ericana de Nephrologıa e Hipertension, Laboratorio de Nefrologia Uni-versidad Austral de Chile, Hospital Regional, Bueras No. 1003, Pisa 2,Valdivia, Chile.E-mail: bri@iamnet.com

REFERENCES

1. Rodrıguez-Iturbe B, Pons H, Herrera-Acosta J, Johnson RJ:The role of immunocompetent cells in non-immune renal diseases.Kidney Int 59:1626–1640, 2001

2. Rodrıguez-Iturbe B, Pons H, Quiroz Y, et al: Mycophenolatemofetil prevents the salt-sensitive hypertension resulting from an-giotensin II exposure. Kidney Int 59:2222–2232, 2001

3. Quiroz Y, Pons H, Gordon KL, et al: Mycophenolate mofetilprevents the salt-sensitive hypertension resulting from short-termnitric oxide synthesis inhibition. Am J Physiol Renal Physiol 281:F38–F47, 2001

4. Alvarez V, Quiroz Y, Nava M, Rodriguez-Iturbe B: Overloadproteinuria is followed by salt-sensitive hipertension caused byrenal infiltration of immune cells. Am J Physiol Renal Physiol283:F1132–F1141, 2002

5. Rodriguez-Iturbe B, Quiroz Y, Nava M, et al: Reduction ofrenal immune cell infiltration results in blood pressure control ingenetically hypertensive rats. Am J Physiol Renal Physiol 282:F191–F201, 2002

6. Rodrıguez-Iturbe B, Quiroz Y, Herrera-Acosta J, et al: The roleof immune cells infiltrating the kidney in the pathogenesis of salt-sensitive hypertension. J Hypertension 20(Suppl 3):S9–S14, 2002

7. Franco M, Tapia E, Santamarıa J, et al: Renal cortical vasocon-striction contributes to the development of salt-sensitive hyper-tension after angiotensin II exposure. J Am Soc Nephrol 12:2263–2271, 2001

8. Johnson RJ, Rodrıguez-Iturbe B, Schreiner G, Herrera-Acosta J: Hypertension: A microvascular and tubulointerstitialdisease. J Hypertens 20(Suppl 3):S1–S7, 2002

9. Johnson RJ, Herrera J, Schreiner G, Rodrıguez-Iturbe B: Ac-quired and subtle renal injury as a mechanism for salt-sensitive

hypertension: Bridging the hypothesis of Goldblatt and Guyton.N Engl J Med 346:913–923, 2002

10. Wilcox CS: Reactive oxygen species: Roles in blood pressure andkidney function. Curr Hypertens Rep 4:160–166, 2002

11. Berry C, Brosnan MJ, Fennell J, et al: Oxidative stress andvascular damage in hypertension. Curr Opin Nephrol Hypertens10:247–255, 2001

12. Vaziri ND, Zhemin N, Oveisi F, Trnavsky-Hobbs DL: Effect ofantioxidant therapy on blood pressure and NO synthase expressionin hypertensive rats. Hypertens 36:957–964, 2000

13. Muller D, Dechend R, Mervaala EMA, et al: NF-�B inhibitionameliorates angiotensin II-induced inflammatory damage in rats.Hypertension 35:193–201, 2000

14. Beswick RA, Zhang H, Marable D, et al: Long-term antioxidantadministration attenuates mineralocorticoid hypertension and re-nal inflammatory response. Hypertens 37:781–786, 2001

15. Nava M, Quiroz Y, Vaziri ND, Rodrıguez-Iturbe B: Melatoninreduces renal interstitial inflammation and improves hypertensionin spontaneously hypertensive rats. Am J Physiol Renal Physiol284:F447–F454, 2003

16. Rodrıguez-Iturbe B, Zhan Chang-D, Quiroz Y, et al: Antioxi-dant-rich diet improves hypertension and reduces renal immuneinfiltration in spontaneously hypertensive rats. Hypertens 41:341–346, 2003

17. Raij L, Azar S, Keane W: Mesangial immune injury, hypertensionand progressive damage in Dahl rats. Kidney Int 26:137–143, 1984

18. Briggs RT, Robinson JM, Karnovsky ML, Karnovsky MJ: Super-oxide production by polymorphonuclear leukocytes. A cytochemi-cal approach. Histochem 84:371–378, 1986

19. Nava M, Romero F, Quiroz Y, et al: Melatonin attenuates theacute renal failure and oxidative stress induced by mercuric chlo-ride in rats. Am J Physiol Renal Physiol 279:F910–F918, 2000

20. Soto H, Mosquera J, Rodrıguez-Iturbe B, et al: Apoptosis inproliferative glomerulonephritis: Decreased apoptosis expressionin lupus nephritis. Nephrol Dial Transplant 12:273–280, 1997

21. Schnackenberg CG, Welch WJ, Wilcox CS: Normalization ofblood pressure and renal vascular resistence in SHR with a mem-brane permeable superoxide dismutasemimetic: Role of nitricoxide. Hypertens 32:59–64, 1999

22. Swei A, Lacy F, De Lano FA, Schmidt-Schonbein GW: Oxida-tive stress in the Dahl salt-sensitive rat. Hypertension 30:1628–1633, 1997

23. Roberts CK, Vaziri ND, Liang K, Barnard JR: Reversibilityof experimental sindrome X by dietary modification. Hypertens37:1323–1328, 2001

24. Barton CH, Ni Z, Vaziri ND: Enhanced nitric oxide inactivationin aortic coarctation-induced hypertension. Kidney Int 60:1083–1087, 2001

25. Laursen JB, Rajagopalan S, Galis Z, et al: Role of superoxidein angiotensin II-induced but not cathecholamine-induced hyper-tension. Circulation 95:588–593, 1997

26. Roggensack AM, Zhang Y, Davidge ST: Evidence for peroxyni-trite formation in the vasculature of women with preeclampsia.Hypertension 33:83–89, 1999

27. Higashi Y, Sasaki S, Nakagawa K, et al: Endothelial functionand oxidative stress in renovascular hypertension. N Engl J Med346:1954–1962, 2002

28. Wilcox CS, Welch WJ: Oxidative stress: Cause or consequenceof hypertension. Exp Biol Med 226:619–620, 2001

29. Laursen JB, Rjagopalan S, Galis Z, et al: Role of superoxide inangiotensin II-induced but not in cathecholamine-induced hyper-tension. Circulation 95:588–593, 1997

30. Rajagopalan S, Kurz S, Munzel T, et al: Angiotensin II-mediatedhypertension in the rat increases superoxide production via mem-brane NADH/NADPH oxidation: Contribution to alterations ofvasomotor tone. J Clin Invest 97:1916–1923, 1996

31. Schnackenberg CG: Physiological and pathophysiological rolesof oxygen radicals in renal microvasculature. Am J Physiol Reg IntegrComp Physiol 282:R335–R342, 2002

32. Zou AP, Li N, Cowley AW, Jr: Production and actions of super-oxide in the renal medulla. Hypertens 37:547–553, 2001

Quiroz et al: Apoptosis and NFjB in hypertensionS-32

33. Gulbins A, Jekle K, Ferlinz H, et al: Physiology of apoptosis.Am J Physiol Renal Physiol 279:F605–F615, 2000

34. Schreck R, Rieber P, Baeuerle PA: Reactive oxygen intermedi-ates as apparently widely used messengers in the activation of theNF-�B transcription factor and HIV. EMBO J 10:2247–2258, 1991

35. Tak PP, Firestein GS: NF-�B: A key role in inflammatory diseases.J Clin Invest 107:7–11, 2001

36. Chen CC, Rosenbloom CL, Manning AM: Selective inhibitionof E-selectin, vascular cell adhesion molecule-1, and intercellularadhesion molecule-1 expression by inhibitors of I kappa B-alphaphosphorilation. J Immunol 155:3538–3545, 1995

37. Miagkow AV, Kovalenko DV, Brown CE, et al: NF-�appaBactivation provides the potential link between inflammation andhyperplasia in the arthritic joint. Proc Natl Acad Sci USA 95:13859–13864, 1998

38. Nanimiya S: The small GTPase Rho: Cellular functions and signaltransduction. J Biochem 120:215–228, 1996

39. Kataoka C, Egashira K, Inoue S, et al: Important role of theRho-kinase in the pathogenesis of cardiovascular inflammationand remodeling induced by long-term blockade of nitric oxidesynthesis. Hypertension 39:245–250, 2002

40. Wang CY, Mayo MW, Korneluk RG, et al: NF-�appa B antiapo-ptosis: Induction of TRAF1 and TRAF2 and c-IAPI and c-IAP2tosuppress caspase-8 activation. Science 281:1680–1683, 1998

41. Romashkova JA, Makarov SS: NF-�appa B is a target of AKTin anti-apoptotic PDGF signaling. Nature 401:33–34, 1999

42. Kasibhatla S, Brunner T, Genestier L, et al: DNA damagingagents induce expression of FAS ligand and subsequent apoptosisin T lymphocytes via activation of NF-�appa B and AP-1. Mol Cell1:543–551, 1998