Gender-Dependent Hypoxic ToleranceMediated Via Gender-Specific Mechanisms

C.A.F. von Arnim, S.M. Etrich, M. Timmler, and M.W. Riepe*Department of Neurology, University of Ulm, Ulm, Germany

Primary hypoxic tolerance and preconditioning are gen-der dependent and modulated in females during theestrus cycle. The underlying mechanisms, however, re-main to be determined. mRNA of estrogen receptor-�(EAR), progesterone receptor (PR), and adenosine recep-tor subtypes A1 and A3 (A1R and A3R) were investigatedwith reverse transcriptase-PCR in hippocampi from con-trol male and female mice and animals treated in vivowith a single i.p. injection of 20 mg/kg body weight3-nitropropionate (3NP) 1 or 24 hr prior to preparation.Results were analyzed relative to expression in hip-pocampi from untreated males. mRNA levels of EAR andA1R were alike in males and females and unaltered bypreconditioning with 3NP. In contrast, PR mRNA levelswere alike in males and females during proestrus butlower during estrus and diestrus (85% � 15%, P � 0.05;and 80% � 10%, P � 0.05, respectively). Upon precon-ditioning, PR mRNA decreased to 67% � 19% (P � 0.05)and 56% � 13% (P � 0.05) during proestrus anddiestrus, respectively, but was unaltered during estrusand in males. On preconditioning, A3R mRNA decreasedfrom 115% � 16% to 86% � 29% (P � 0.05) duringdiestrus but remained at the control level duringproestrus and estrus. With low-level expression of PRs,as achieved upon preconditioning, hypoxic tolerance isincreased. Other than in males, adenosine A3 receptorsare not up-regulated upon preconditioning in females.Thus, not only is net hypoxic tolerance gender depen-dent but mechanisms conferring hypoxic tolerance aregender specific. © 2002 Wiley-Liss,Inc.

Key words: preconditioning; estrogen receptors; pro-gesterone receptors; adenosine receptors

Endogenously, the central nervous system can with-stand cerebral hypoxia or ischemia for a limited amount oftime, primary hypoxic tolerance. With an appropriatetime interval and dosage, a mild ischemic or chemicallyinduced hypoxic challenge of the central nervous systemincreases hypoxic tolerance, ischemic and chemical pre-conditioning (Kitagawa et al., 1990; Riepe et al., 1996).Increased hypoxic tolerance is associated with improvedenergy metabolism during hypoxia, decreased posthypoxicfree radical production, improved posthypoxic morphol-ogy, and preserved posthypoxic neuronal function (Riepeet al., 1996, 1997). Primary hypoxic tolerance has been

shown to be gender dependent (Alkayed et al., 1998;Kasischke et al., 1999), as has also been shown for theincrease of hypoxic tolerance by preconditioning (Kasis-chke et al., 1999).

In recent years, several studies have been performedinvestigating the effect of sex hormones. Findings, how-ever, remain ambiguous. It was reported that estrogenprotects against brain injury, and it was concluded thatestrogen contributes to the sex difference in ischemicvulnerability (Dubal et al., 1998; Rusa et al., 1999; Fukudaet al., 2000). The therapeutic range of exogenously ap-plied estrogen was reported to be narrow, and, unlike forthe male brain, single-injection estrogen exposure did notrescue ischemic tissue in the female brain (Rusa et al.,1999). It was, therefore, concluded that, although exoge-nous steroid therapy protected both male and femaleestrogen-deficient brain, the mechanisms may not beidentical and may depend on long-term hormone aug-mentation in females (Rusa et al., 1999). From experi-ments on knockout mice, it was concluded that the pro-tective effects of estrogen can be linked to estrogenreceptor-�, whereas estrogen receptor-� was found not toaffect the outcome (Dubal et al., 2001). The importance ofestrogen receptor-� has been called into doubt, however,in other studies (Culmsee et al., 1999; Sampei et al., 2000).Pharmacologic treatment with an estrogen receptor antag-onist worsened the postischemic outcome in the striatumof female animals via mechanisms unrelated to cerebralblood flow (Sawada et al., 2000). Direct application ofestrogen, on the other hand, has been shown to protectcortical but not striatal neurons (Dubal et al., 1998),whereas other modulators improved outcome in the basalganglia but not in cortex (Rossberg et al., 2000). Estrogenreceptor antagonists increased ischemic damage in femalesbut not in males (Sawada et al., 2000). However, all ofthese studies were performed in ovarectomized animalsand thus may not apply in general.

The first two authors contributed equally to this work.

*Correspondence to: Dr. Matthias W. Riepe, Department of Neurology,University of Ulm, Steinhovelstr. 1, 89075 Ulm, Germany.E-mail: matthias.riepe@medizin.uni-ulm.de

Received 11 October 2001; Revised 18 December 2001; Accepted 30December 2001

Published online 27 February 2002 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/jnr.10195

Journal of Neuroscience Research 68:84–88 (2002)

© 2002 Wiley-Liss, Inc.

Similarly to observations of neuroprotection uponapplication of estrogen, it was found that application ofprogesterone exerts protective effects against the neuronalcerebral damage induced by acute ischemia in female(Gonzalez-Vidal et al., 1998) and male (Jiang et al., 1996)animals. However, it also has been reported that exoge-nous progesterone therapy does not ameliorate ischemicinjury in female rats but rather may even exacerbate in-farction in subcortical regions (Murphy et al., 2000).Again, all of these studies have been performed on ovar-ectomized animals and thus may not apply in general.

Activation of adenosine receptors repeatedly wasshown to partake in mediating primary and induced hy-poxic tolerance (Heurteaux et al., 1995; Perez-Pinzon etal., 1996). In male animals it was shown that the adenosineA3 receptor is selectively involved in mediating the pro-tective effects of preconditioning (von Arnim et al., 2000).It was the goal of the present study to investigate gender-related mechanisms of primary and induced hypoxic tol-erance.

MATERIALS AND METHODS

Male and female CD-1 mice (25–35 g) were killed bycervical dislocation. In part animals received a single i.p. injec-tion of 20 mg/kg body weight 3-nitropropionate (3NP) 1 or24 hr prior to preparation. Similarly to a previous report (Ka-sischke et al., 1999), the estrus cycle was determined in femalemice as described by Allen (1922). In brief, the stages of theestrus cycle were determined by analyzing the number and typeof cells in vaginal smears. Vaginal smears in metestrus anddiestrus are very similar; therefore, data were pooled to avoidartifactual group assignment and overinterpretation. The cyclewas determined prior to i.p. injection. Both hippocampi fromeach animal were prepared and immediately submerged in liquidnitrogen and stored at –80°C.

RT-PCR Protocol

Total RNA was extracted from both hippocampi of oneanimal with PeqGold RNA pure (Peqlab, Erlangen, Germany)according to the manufacturer’s protocol. First-strand cDNAsynthesis was performed using 0.5 �g of total RNA, randomhexamers, and MuLv-reverse transcriptase (GeneAmp, RNA

PCR Kit; Perkin Elmer Roche Molecular Systems, Inc.,Branchburg, NJ) in a 21 �l reaction. Aliquots (1 �l) from eachreverse transcriptase reaction were used for PCR amplificationwith primer pairs for the mRNA of interest and the ubiquitouslyexpressed control sequence hypoxanthine phosphoribosyltrans-ferase (Hprt) to ensure integrity of cDNA, spanning an intron toshow putative contamination with genomic DNA. Each set ofreverse transcriptase PCR included control samples run withoutRNA.

PCR (Progene; Techne Ltd., Cambridge, United King-dom) cycling parameters are shown in Table I. Linearity for thecycle numbers used was shown in preexaminations (data notshown).

Reaction products were separated on a visigel separationmatrix (Stratgene, La Jolla, CA), to which ethidium bromidewas added. The bands were visualized under UV illumination(�exc � 302 nm) and photographed. The PCR products weresequenced to confirm their identity.

Relative optical density of the bands was determined usingan image analysis system (MultiAnalyst; Bio-Rad, Hercules,CA). All measurements were normalized to the relative opticaldensity of the Hprt band.

Statistical Analysis

Each experiment was conducted with hippocampi fromfive or six animals. Therefore, statistical testing was performedby Kruskal-Wallis one-way ANOVA, and multiple comparisonswere tested with Newman-Keuls test. Statistical significance wasaccepted at P � 0.05.

RESULTS

Estrogen Receptor-� and Progesterone ReceptorsNo significant changes of estrogen receptor-� were

observed in females during the estrus cycle or in males.Upon preconditioning, expression of estrogen receptor-�remained at control level (Fig. 1, Table II).

In untreated control animals, expression of proges-terone receptors is similar in males and females duringproestrus, whereas it is significantly lower during estrusand diestrus (85% � 15%, P � 0.05; and 80% � 10%, P �0.05, respectively). One hour after preconditioning, ex-

TABLE I. Primers and Cycling Parameters for Estrogen Receptor-� (EAR) and Progesterone Receptor (PR) mRNA, Adenosine A1Receptor (A1R) and Adenosine A3 Receptor (A3R) mRNA, and Hypoxanthine Phosphoribosyltransferase (Hprt) mRNA (Standard)

Primer Sequences (5� to 3�) forward/reverseProductsize (bp)

Temperature(°C) Cycles

EAR GGCCTGACTCTGCAGCAGCAG 300 62 36GTTGGGGAAGCCCTCTGCTTC

PR CCACAGGAGTTTGTCAAACTC 325 60 37TAACTTCAGACATCATTTCCGG

A1 ATTGCCTTGGTCTCTGTGC 637 59 32AGCTCCTTCCCGTAGTAC

A3 CAAGCTGACAGTCAGATATAG 362 60 38AACAGGGACTTAGCTGTCTTG

Hprt GCTGGTGAAAAGGACCTCT 249 61 36CACAGGACTAGAACACCTGC

Gender Specificity of Hypoxic Tolerance 85

pression is reduced in proestrus and diestrus (67% � 19%,P � 0.05; and 56% � 13%, P � 0.05, respectively; Fig. 1,Table II).

Adenosine A1 and A3 ReceptorsNo significant changes of adenosine A1 receptors

were observed in females during the estrus cycle or inmales. Upon preconditioning, expression of adenosine A1receptors remained at the control level (Table III).

In untreated control animals, expression of adenosineA3 receptors was similar in males and females during allstages of the estrus cycle. No change in expression wasobserved upon preconditioning with 1 and 24 hr intervalsduring proestrus and estrus. In diestrus, levels 1 and 24 hrafter preconditioning are reduced compared with control(86% � 29%, P � 0.05; and 85% � 26%, P � 0.05,respectively; Table III).

DISCUSSIONPrimary hypoxic tolerance and preconditioning have

been shown to be gender dependent (Alkayed et al., 1998;Kasischke et al., 1999). Subsequently, it was shown re-peatedly that endogenous or exogenous modulation ofestrogen and progesterone affects postischemic or posthy-poxic outcome (Alkayed et al., 2000; Fukuda et al., 2000).However, the results on the effects remained ambiguous.From several studies, it was concluded that estrogen pro-tects against brain injury and contributes to the sex differ-ence in ischemic vulnerability (Jiang et al., 1996; Dubal etal., 1998; Gonzalez-Vidal et al., 1998; Rusa et al., 1999;Fukuda et al., 2000). The therapeutic range was reportedto be narrow, and the mechanism of protection of thesehormones was concluded to be gender dependent (Rusa etal., 1999; Sawada et al., 2000). In contrast, in a time frameof weeks, neither estrogen withdrawal, nor replacement,

nor supplementation significantly affected the outcomefrom acute coronary occlusion (McNulty et al., 2000).Similarly, progesterone was reported to reduce ischemicbrain injury (Alkayed et al., 2000), whereas it exacerbatedischemic brain injury in other studies (Murphy et al.,2000).

It was shown that modulation of ischemic braininjury by estrogen or progesterone is unrelated to changesin cerebral blood flow (Alkayed et al., 2000; Fukuda et al.,2000). Commonly, it is thought that exogenous applica-tion of estrogen and progesterone reduces the conse-quences of the ischemic injury cascade by enhancing an-tioxidant mechanisms, reducing excitotoxicity or immuneinflammation, or providing neurotrophic support (Stein,2001). Investigation of these mechanisms is particularlyrelevant for the understanding of the effects of pharmaco-logic external dosage of estrogen or progesterone. Thepresent study for the first time investigates the endogenousexpression of estrogen receptor-� and progesterone recep-tors as endogenous modulators of gender and estrus cycledependence of primary and induced hypoxic tolerance infemale and male animals. In addition, expression of aden-osine receptors, known to contribute in males (von Arnimet al., 2000), has been investigated in females during thedifferent stages of the estrus cycle.

Progesterone receptors as well as estrogen receptor-�and adenosine receptors are down-regulated when thecorresponding hormone levels are high. Accordingly, itwas found that estrogen receptor-� is significantly lowerin males than in females during proestrus, resulting fromhigh estrogen levels during proestrus. Expression of estro-gen receptor-� is unchanged on preconditioning. This isin harmony with a previous study investigating hypoxictolerance in estrogen receptor-�-deficient mice (Sampeiet al., 2000). In that study, ischemic injury was found notto depend on activation of this receptor subtye.

Adenosine receptors in control male and female hip-pocampi during all stages of estrus cycle are alike. In aprevious study, it was shown that adenosine A1 receptormRNA remains unchanged in male animals on precondi-tioning (von Arnim et al., 2000). Similarly, it was observedin this study that adenosine A1 receptor mRNA remainsunchanged in all stages of the estrus cycle in female animalson preconditioning. With adenosine A3 receptor mRNA,a short-lasting increase is observed 1 hr after precondition-ing, with a return to baseline at 24 hr after preconditioning(von Arnim et al., 2000). The present study shows, how-ever, that up-regulation of adenosine A3 receptor mRNAon preconditioning is gender dependent and dependenton the estrus cycle in females. It remains unchanged dur-ing proestrus and estrus and even is down-regulated duringdiestrus.

From a previous study, it was concluded that pro-gesterone affects primary and induced hypoxic toleranceto a greater extent than estrogen (Kasischke et al., 1999).Accordingly, during stages of the estrus cycle with com-paratively high primary hypoxic tolerance, i.e., estrus anddiestrus, expression of progesterone receptors is low (cor-

Fig. 1. Typical set of RT-PCR analyses of control hippocampi andhippocampi prepared from male and female animals pretreated with asingle i.p. injection of 20 mg/kg body weight 3-nitropropionate (3NP)1 hr and 24 hr prior to brain preparation. PR, progesterone receptor;E�, estrogen receptor-�; Hprt, hypoxanthine phosphoribosyltrans-ferase (standard); bp, base pair marker.

86 von Arnim et al.

responding to high progesterone levels). Upon precondi-tioning, expression of progesterone receptors goes downduring proestrus, resulting in an improved induced hy-poxic tolerance (Kasischke et al., 1999). Although a similarobservation was made 1 hr after preconditioning duringdiestrus, at this time a decrease of adenosine A3 receptorswas observed. The beneficial effect of reduction of pro-gesterone receptors and the effect of a decrease of A3receptor thus levels off, corresponding to similar hypoxictolerance with and without preconditioning, as observedpreviously (Kasischke et al., 1999). However, the causalrelationship between hormone receptor regulation andadenosine receptor regulation should be investigated inmore detail. Gender-specific difference in ATP-sensitiveK channels may play an additional role (Ranki et al.,2001).

In summary, with low-level expression of progester-one receptors, as achieved upon preconditioning, hypoxictolerance is increased. Other than in males, adenosine A3receptors are not up-regulated upon preconditioning.Thus, not only is net hypoxic tolerance gender dependent

but underlying mechanisms are gender-specific. Both thenet difference in primary and induced hypoxic toleranceand the specificity of the mechanisms conferring the hy-poxic tolerance should be considered in the developmentof future gender-specific therapeutic strategies.

ACKNOWLEDGMENTSThe study was supported by a grant from the Deut-

sche Forschungsgemeinschaft (Ri583/2-2 and Ri583/2-3)to M.W.R.

REFERENCESAlkayed NJ, Harukuni I, Kimes AS, London ED, Traystman RJ, Hurn PD.

1998. Gender-linked brain injury in experimental stroke. Stroke 29:159–165.

Alkayed NJ, Murphy SJ, Traystman RJ, Hurn PD, Miller VM. 2000.Neuroprotective effects of female gonadal steroids in reproductively se-nescent female rats. Stroke 31:161–168.

Allen E. 1922. The oestrus cycle in the mouse. Am J Anat 30:297–371.

TABLE II. Estrogen Receptor-� and Progesterone Receptor mRNA (Mean � SD) in Hippocampi From Male and Female ControlAnimals and Animals Pretreated 1 or 24 hr Prior to Preparation With Single i.p. Injection of 3-Nitropropionate

Male Proestrus Significance vs. male Estrus Significance vs. male Diestrus Significance vs. male

Estrogen receptor-�Control 100 � 0 90 � 26 ** 107 � 24 n.s. 102 � 19 *Preconditioned 1 hr 100 � 25 83 � 32 ** 130 � 73 n.s. 89 � 10 *Significance vs. control n.s. n.s. n.s. n.s.Preconditioned 24 hr 115 � 21 113 � 54 n.s. 97 � 34 * 98 � 35 *Significance vs. control n.s. n.s. n.s. n.s.

Progesterone receptorControl 100 � 0 105 � 47 n.s. 85 � 15 * 80 � 10 **Preconditioned 1 hr 98 � 11 67 � 19 ** 75 � 19 ** 56 � 13 **Significance vs. control n.s. * n.s. *Preconditioned 24 hr 101 � 15 70 � 18 ** 71 � 19 ** 61 � 18 **Significance vs. control n.s. n.s. n.s. n.s.

*P � 0.05.**P � 0.01.

TABLE III. Adenosine Receptor Subtypes A1 and A3 mRNA (Mean � SD) in Hippocampi From Male and Female Control Animalsand Animals Pretreated 1 or 24 hr Prior to Preparation With Single i.p. Injection of 3-Nitropropionate

Male Proestrus Significance vs. male Estrus Significance vs. male Diestrus Significance vs. male

A1 receptorControl 100 � 0 98 � 16 n.s. 103 � 11 n.s. 107 � 14 n.s.Preconditioned 1 hr 102 � 15 94 � 23 95 � 12Significance vs. control n.s. n.s. n.s.Preconditioned 24 hr 93 � 10 95 � 18 98 � 17Significance vs. control n.s. n.s. n.s.

A3 receptorControl 100 � 0 112 � 18 n.s. 111 � 21 n.s. 115 � 16 n.s.Preconditioned 1 hr 102 � 20 109 � 16 86 � 29Significance vs. control n.s. n.s. *Preconditioned 24 hr 107 � 10 118 � 15 85 � 26Significance vs. control n.s. n.s. *

*P � 0.05.**P � 0.01.

Gender Specificity of Hypoxic Tolerance 87

Culmsee C, Vedder H, Ravati A, Junker V, Otto D, Ahlemeyer B, KriegJC, Krieglstein J. 1999. Neuroprotection by estrogens in a mouse modelof focal cerebral ischemia and in cultured neurons: evidence for areceptor-independent antioxidative mechanism. J Cereb Blood FlowMetab 19:1263–1269.

Dubal DB, Kashon ML, Pettigrew LC, Ren JM, Finklestein SP, Rau SW,Wise PM. 1998. Estradiol protects against ischemic injury. J Cereb BloodFlow Metab 18:1253–1258.

Dubal DB, Zhu H, Yu J, Rau SW, Shughrue PJ, Merchenthaler I, KindyMS, Wise PM. 2001. Estrogen receptor alpha, not beta, is a critical linkin estradiol-mediated protection against brain injury. Proc Natl Acad SciUSA 98:1952–1957.

Fukuda K, Yao H, Ibayashi S, Nakahara T, Uchimura H, Fujishima M, HallED. 2000. Ovariectomy exacerbates and estrogen replacement attenuatesphotothrombotic focal ischemic brain injury in rats. Stroke 31:155–160.

Gonzalez-Vidal MD, Cervera GM, Ruelas R, Escobar A, Morali G,Cervantes M. 1998. Progesterone: protective effects on the cat hippocam-pal neuronal damage due to acute global cerebral ischemia. Arch Med Res29:117–124.

Heurteaux C, Lauritzen I, Widmann C, Lazdunski M. 1995. Essential role ofadenosine, adenosine A1 receptors, and ATP-sensitive K channels in ce-rebral ischemic preconditioning. Proc Natl Acad Sci USA 92:4666–4670.

Jiang N, Chopp M, Stein D, Feit H. 1996. Progesterone is neuroprotectiveafter transient middle cerebral artery occlusion in male rats. Brain Res735:101–107.

Kasischke K, Huber R, Li H, Timmler M, Riepe MW. 1999. Primaryhypoxic tolerance and chemical preconditioning during estrus cycle inmice. Stroke 30:1256–1262.

Kitagawa K, Matsumoto M, Tagaya M, Hata R, Ueda H, Niinobe M,Handa N, Fukunaga R, Kimura K, Mikoshiba K, Kamada T. 1990.“Ischemic tolerance” phenomenon found in the brain. Brain Res 528:21–24.

McNulty PH, Jagasia D, Whiting JM, Caulin-Glaser T. 2000. Effect of6-wk estrogen withdrawal or replacement on myocardial ischemic toler-ance in rats. Am J Physiol Heart Circ Physiol 278:H1030–H1034.

Murphy SJ, Traystman RJ, Hurn PD, Duckles SP. 2000. Progesteroneexacerbates striatal stroke injury in progesterone-deficient female animals.Stroke 31:1173–1178.

Perez-Pinzon MA, Mumford PL, Rosenthal M, Sick TJ. 1996. Anoxicpreconditioning in hippocampal slices: role of adenosine. Neuroscience75:687–694.

Ranki HJ, Budas GR, Crawford RM, Jovanovic A. 2001. Gender-specificdifference in cardiac ATP-sensitive K() channels. J Am Coll Cardiol38:906–915.

Riepe MW, Niemi WN, Megow D, Ludolph AC, Carpenter DO.1996. Mitochondrial oxidation in rat hippocampus can be preconditionedby selective chemical inhibition of succinic dehydrogenase. Exp Neurol138:15–21.

Riepe MW, Esclaire F, Kasischke K, Schreiber S, Nakase H, Kempski O,Ludolph AC, Dirnagl U, Hugon J. 1997. Increased hypoxic tolerance bychemical inhibition of oxidative phosphorylation: “chemical precondi-tioning.” J Cereb Blood Flow Metab 17:257–264.

Rossberg MI, Murphy SJ, Traystman RJ, Hurn PD. 2000. LY353381.HCl,a selective estrogen receptor modulator, and experimental stroke. Stroke31:3041–3046.

Rusa R, Alkayed NJ, Crain BJ, Traystman RJ, Kimes AS, London ED,Klaus JA, Hurn PD. 1999. 17-Beta-estradiol reduces stroke injury inestrogen-deficient female animals. Stroke 30:1665–1670.

Sampei K, Goto S, Alkayed NJ, Crain BJ, Korach KS, Traystman RJ,Demas GE, Nelson RJ, Hurn PD. 2000. Stroke in estrogen receptor-alpha-deficient mice. Stroke 31:738–743.

Sawada M, Alkayed NJ, Goto S, Crain BJ, Traystman RJ, Shaivitz A,Nelson RJ, Hurn PD. 2000. Estrogen receptor antagonist ICI182,780exacerbates ischemic injury in female mouse. J Cereb Blood Flow Metab20:112–118.

Stein DG. 2001. Brain damage, sex hormones and recovery: a new role forprogesterone and estrogen? Trends Neurosci 24:386–391.

von Arnim CA, Timmler M, Ludolph AC, Riepe MW. 2000. Adenosinereceptor up-regulation: initiated upon preconditioning but not upheld.Neuroreport 11:1223–1226.

88 von Arnim et al.