Cyclooxygenases 1 and 2 and thromboxane synthase in kidneys of Lyon hypertensive rats

AJH

2000;13:404 – 409

Cyclooxygenases 1 and 2 and Thromboxane Synthase in Kidneys of Lyon Hypertensive Rats Nicole Bernard, Joelle Sacquet, Daniel Benzoni, and Jean Sassard

Compared with their two normotensive (LN and LL) controls, genetically hypertensive rats of the Lyon strain (LH) exhibit increased renal vascular resistance and a blunted pressure natriuresis function as well as an increased urinary excretion of vasoconstrictor prostanoids. The aim of this study was to assess in the kidneys of these animals the synthesis of vasoconstrictor or sodiumretaining prostanoids. The relative abundance of the mRNAs of cyclooxygenases (COX) 1 and 2 and of thromboxane A2 synthase (TXS), was measured by reverse-transcription polymerase chain reaction (RT-PCR) in renal cortex and medulla dissected in groups of male LH, LN, and LL rats either in baseline conditions or after 1 week of salt loading (1.5% NaCl in the drinking water). In basal conditions, at 3 and 11 weeks of age COX1 was expressed in the kidneys of all rats more markedly in medulla than in cortex. COX2 was poorly

G

enetically hypertensive rats of the Lyon strain (LH) differ from their normotensive (LN) and low blood pressure (LL) controls1 by a preglomerular vasoconstriction and a shift of the pressure natriuresis curve towards high blood pressure (BP) levels.2 We previously ob-

Received April 1, 1999. Accepted August 17, 1999. From the De´partement de Physiologie et Pharmacologie Clinique, CNRS ESA, Faculte´ de Pharmacie, Lyon, France. Address correspondence and reprint requests to Dr N. Bernard, De´partement de physiologie et pharmacologie clinique, Faculte´ de pharmacie, 8 Avenue Rockefeller, F-69373 Lyon Cedex 08, France; e-mail: [email protected]

© 2000 by the American Journal of Hypertension, Ltd. Published by Elsevier Science, Inc.

expressed in the whole kidney. TXS expression was usually too low to be quantified. No difference could be observed among LH, LN, and LL rats. After salt loading, the expression of COX1 was enhanced in the medulla and that of COX2 reduced in the cortex. LH rats differed from controls by a significantly more marked increase in medullary COX1 expression. The present work excludes any primary generalized increase in the renal expression of the genes that control the synthesis of vasoconstrictor prostanoids in LH rats, but suggests that medullary COX1 is upregulated by salt in these animals. Am J Hypertens 2000; 13:404 – 409 © 2000 American Journal of Hypertension, Ltd.

KEY WORDS:

Hypertension, kidney, prostaglandins, salt, thromboxane.

served that higher amounts of thromboxane (TX) B2, the stable form of the vasoconstrictor TXA2, were excreted in the urines of 5- and 9-week– old LH rats,3 and that a chronic treatment with an antagonist of the prostaglandin (PG) H2-TXA2 (TP) receptor delayed and partially prevented their hypertension.4 These data suggested that vasoconstrictor prostanoids could participate in the renal alterations seen in LH rats. Because these autacoids are undergoing a complex metabolism and are unevenly distributed in tissues,5,6 we thought it of interest to assess, at the kidney level, the expression of the genes that control their synthesis. Cyclooxygenase (EC.1.14.99.1), the rate-limiting enzyme of the pathway leading to prostaglandins, exists 0895-7061/00/$20.00 PII S0895-7061(99)00211-3

AJH–APRIL 2000 –VOL. 13, NO. 4, PART 1

in two forms, one constitutive (COX1) and one inducible (COX2). COX1 predominates under physiologic conditions in numerous structures of the kidney such as vascular, glomerular, and tubular structures and its abundance is increased by medullary osmolality.7 Moreover, it has been shown to be overexpressed in aortic cells from spontaneously hypertensive rats (SHR) compared with Wistar Kyoto (WKY) controls.8 COX2 is transiently and intensively induced in macrophages and endothelial cells during inflammatory processes.9 However, it is also present in restricted areas of the kidney, mainly the macula densa, where it is downregulated by high salt intake.10 –12 In the present work, the gene expression of COX1 and COX2 and of the renal-specific TX synthase (TXS) was studied separately in kidney cortex and medulla so as to take into account regional differences in prostanoid synthesis.5,6 The study was conducted in the three strains of Lyon rats at two ages that characterize the development of hypertension in the Lyon model, ie, 3 and 11 weeks of age. In addition, 8-week– old animals, which, due to their young age, remain devoid of marked renal lesions,13 were also studied after an oral salt load. MATERIALS AND METHODS Animals The study was conducted in 12 groups of six LH, LN, or LL male rats. They were housed in controlled conditions of light (8 –20 h), temperature (21 ⫾ 1°C), and humidity (60⫾10%) and fed a standard diet (0.3% NaCl, A03, UAR, F-Villemoisson s/Orge, France) with tap water ad libitum. In addition, 7-week– old LH, LN, and LL rats were salt loaded for 7 days by adding 1.5% NaCl in drinking water. Other 7-week– old rats served as controls as they drank tap water. Systolic blood pressure (SBP) was measured by tail-cuff plethysmography (Narco Biosystems, Houston, TX) in preheated (15 min at 37°C) unrestrained animals, except in 3-week– old rats because, at this age, plethysmography is not very accurate. Animals were killed by an overdose of phenobarbital intraperitoneally. Kidneys were immediately removed, cleaned from connective tissue, and rinsed with sterile physiologic saline. After cooling by aerosol freezer (Cryolab, Labonord, F-Villeneuve d’Asq, France), the kidneys were hemisectioned, then cortex and medulla were separated macroscopically—including the inner part of medulla without papilla— weighed, frozen in liquid nitrogen, and stored at ⫺80°C until extraction. The study was conducted in accordance with our institutional guidelines for animal care. RNA Determination RNA contents were measured by extraction of RNA followed by semiquantitative reverse transcription-polymerase chain reaction (RT-

PROSTANOID SYNTHESIS IN LH RATS

405

PCR). Tissues were disrupted in a mortar filled with liquid nitrogen and homogenized by Dounce in extraction buffer. For the determination of COX1 at 3 and 11 weeks the extraction step led to total RNA. For COX2 and for the experiments using salt-loaded rats we used a specific mRNA extraction method. Extraction of total RNA was achieved by a method derived from Chomczynski et al14 in 4.23 mol/L guanidine isocyanate, 0.75 mol/L sodium citrate, 10% N-lauryl sarcosine, and 14.2 mol/L 2␤-mercapto-ethanol. A dilution of homogenized RNA was quantified by UV absorbance at 260 nm and expressed as total RNA per weight of tissue. RNA quality was checked by 18S and 28S RNA ribosomal fraction profile after agarose electrophoresis in denaturing conditions. Extraction of mRNA used magnetic separation of particles bearing streptavidin coupled to biotinylated polyT probe seeking all polyA mRNA termination (PolyATtract System, Promega, F-Charbonnie`res, France). An RNAse inhibitor (RNAsin, Promega) was added in the medium along all steps of RT-PCR to protect mRNA from degradation. Two microliters (1 ␮g of the total RNA extract or mRNA from 100 mg of tissue) were reverse transcribed by 100 units of Murine Moloney Leukemia Virus (MMLV) reverse transcriptase (Promega) using either oligo dT15 or specific antisense primers in a PCR buffer under a 20-␮l total volume. The cDNA was submitted to three serial dilutions (1/10, 1/20, and 1/50 for COX1; 1/2, 1/4, and 1/8 for COX2 and TXS; and 1/100, 1/200, and 1/500 for glyceraldehyde phosphate deshydrogenase, GAPDH) and amplified by 0.5 ␮mol/L specific primers with 1 unit Taq polymerase (Oncor-Applige`ne, Illkirch, France), 1.5 mmol/L MgCl2, and 0.5 mmol/L dNTP. After an initial 5-min denaturation period at 94°C, the DNA was amplified for 30 cycles for COX1 and 35 cycles for COX2 and TXS, each cycle consisting of 1 min denaturation at 94°C, primer annealing at 60°C for 1 min, and extension at 72°C for 90 s, followed by 3 min, 30 s of final extension at 72°C. Primers used are in Table 1. They were chosen according to published sequences of these genes in Sprague Dawley rats, the strain of origin for LH, LN, and LL animals. PCR products were electrophoretically separated on agarose and visualized by ethidium bromide staining. They were quantified by densitometry (KDS-1D system Amersham, F-Les Ulis, France) in comparison to known concentrations of pure murine cDNA (SpiBio, Cayman Chemicals, Gif s/Yvette, France) amplified in the same run. Contamination by genomic DNA was checked by negative RT. The RT-PCR method was checked for linearity with pure cDNA and serial dilutions of mRNA reverse transcript in different experimental conditions. The results are expressed as the ratio of light intensity of a given mRNA controlled for the

406

AJH–APRIL 2000 –VOL. 13, NO. 4, PART 1

BERNARD ET AL

TABLE 1. SEQUENCES OF PRIMERS USED Gene

Sense Primer

Antisense Primer

Reference Number

COX1 COX2 TXS GAPDH

5⬘CTCACAGTGCGGTCCAAC3⬘ 5⬘GAAATGGCTGCAGAGTTG3⬘ 5⬘TCCAGAGGTGTTACTGCTGT3⬘ 5⬘GTGAAGGTCGTGTCAACGGATTT3⬘

5⬘CCAGCACCTGGTACTTAAG3⬘ 5⬘GGAATTCTCATCTAGTCTGGAGAGTTG3⬘ 5⬘GAAGCATGACAAACATTTATTC3⬘ 5⬘CACAGTCTTCTGAGTGGCAGTGAT3⬘

15, 16 10 17, 18 19

expression of the housekeeping gene GAPDH amplified in the same tissues. Statistical Analysis Results are mean ⫾ SEM. Comparisons between groups used two-way ANOVA with strain and diet or three-way ANOVA with strain, age, and tissue localization as factors of variation. RESULTS Basal Conditions Systolic blood pressure established at 135 ⫾ 4, 124 ⫾ 2, and 108 ⫾ 3 mm Hg at 7 weeks; 147 ⫾ 4, 124 ⫾ 2, and 116 ⫾ 2 mm Hg at 8 weeks; and 152 ⫾ 7, 124 ⫾ 4, and 113 ⫾ 3 mm Hg at 11 weeks of age in LH (n ⫽ 6), LN (n ⫽ 6), and LL (n ⫽ 6) rats, respectively. Figure 1 shows typical PCR products in one LH rat. For COX1, COX2, and TXS the electrophoretic bands were of the expected length and did not differ in nature when amplified with different nested primers in the three strains of rats (data not shown). The expression of COX1 could be quantified in all the samples. Figure 2 shows that it was three- to fivefold higher in the medulla than in the cortex, and that it significantly decreased with age in the cortex but not in the medulla. No significant difference could be disclosed among the three strains, although COX1 expression tended to be lower in LH than in LN and LL kidneys of 11-week– old rats. COX2 was expressed at a much lower level than COX1 in both cortex and

FIGURE 1. Typical products of RT-PCR for COX1, COX2, TXS, and GAPDH in cortex (C) and medulla (M) of kidneys of Lyon rats aged 3 and 11 weeks, or 8-week– old control or salt loaded kidneys (one representative LH kidney).

medulla of the three strains. Due to the different number of amplification cycles used for COX1 and COX2, a more precise comparison of the expression of the two enzymes would be inaccurate. COX2 expression significantly decreased with age in the cortex of LH kidneys only. Levels of TXS could not be quantified because most of the amplification products were below the limit of detection. When a nested amplification (with twice 30 cycles) was applied, the identification was easier but the quantification was no longer linear. When the lowest dilution (1/2) of cDNA used led to a discernable electrophoretic band, the expression of TXS was considered as positive. Table 2 summarizes the number of positive data and indicates that TXS expression tended to be higher in 3-week– old than in 11-week– old rats of the three strains, without obvious interstrain difference.

FIGURE 2. Expression (ratio to GAPDH) of COX1 and COX2 in cortex (C) and medulla (M) of kidneys of Lyon hypertensive (LH f, n ⫽ 6), normotensive (LN 3, n ⫽ 6), and low blood pressure (LL 䡺, n ⫽ 6) rats aged 3 and 11 weeks. *P ⬍ .05, v 3 weeks.

AJH–APRIL 2000 –VOL. 13, NO. 4, PART 1

PROSTANOID SYNTHESIS IN LH RATS

407

TABLE 2. NUMBER OF DISCERNABLE THROMBOXANE SYNTHASE EXPRESSIONS IN KIDNEYS OF LYON HYPERTENSIVE (LH), NORMOTENSIVE (LN), AND LOW BLOOD PRESSURE (LL) RATS Age, weeks 3 11

Tissue

LH (n ⴝ 6)

LN (n ⴝ 6)

LL (n ⴝ 6)

Cortex Medulla Cortex Medulla

4 6 4 3

5 5 2 3

4 4 2 4

Effect of Salt Intake One week of salt loading induced rises in SBP, which did not differ significantly among LH, LN, and LL 7-week– old rats (Table 3). As already observed, the weight gain decreased during saline intake, this decrease reaching significance in LH rats only. As shown in Figure 3, salt loading enhanced COX1 expression, the increase being significant only in the medulla of LH kidneys. COX2 expression was not significantly changed in the cortex and medulla of the three strains of rats.

FIGURE 3. Effects of a 1-week salt loading on the expression (ratio to GAPDH) of COX1 and COX2 in cortex (C) and medulla (M) of kidneys of Lyon hypertensive (LH f, n ⫽ 6), normotensive (LN 3, n ⫽ 6), and low blood pressure (LL 䡺, n ⫽ 6) 8-week– old rats. *P ⬍ .05, v controls.

DISCUSSION This study assessed the expression of the enzymes that control the synthesis of the most important vasoconstrictor prostanoids in the kidneys of LH rats and of their LN and LL controls. The major findings are that the expression of COX1, COX2, and TXS does not differ among the strains in baseline conditions, whereas salt loading increased COX1 expression in the medulla of LH kidneys only. The role of autocrine and paracrine regulatory compounds in cardiovascular diseases is increasingly being recognized.20,21 However, their importance remains difficult to determine precisely for several reasons, among which are the difficulty of their quantification and the lack of fully specific inhibitors of their synthesis. In previous experiments we observed that young LH rats, compared with their normotensive LN and LL controls, excreted higher amounts of TXB2 3 and that a chronic treatment with a TP receptor

antagonist delayed the development of their hypertension.4 In addition, using isolated perfused kidneys we observed that TP receptor antagonists decreased the preglomerular vasoconstriction seen in LH kidneys while leaving untouched the flow in LN and LL kidneys.1 Taken as whole, these results suggested that an increased production of PGH2 or TXA2 could be involved in the high blood pressure of LH rats, either directly by elevating the peripheral resistances or indirectly through a renal vasoconstriction that leads to a rightward shift of the pressure natriuresis. Therefore, in the present work we assessed, at the kidney level, the expression of the genes that control the synthesis of these two vasoconstrictor prostanoids. As a preliminary study showed us that Northern blot analysis was not sensitive enough, this was achieved using a RT-PCR detection of corresponding mRNAs. RNA was protected by a rapid dissection and freezing

TABLE 3. CHANGES IN SYSTOLIC BLOOD PRESSURE (⌬ SBP) AND BODY WEIGHT (⌬ BW) FROM 7 TO 8 WEEKS OF AGE IN LYON HYPERTENSIVE (LH), NORMOTENSIVE (LN), AND LOW BLOOD PRESSURE (LL) RATS DRINKING TAP WATER (CONTROLS) OR 1.5% NaCl (SALINE) LH (n ⴝ 6)

⌬ SBP (mm Hg) ⌬ BW (g)

LN (n ⴝ 6)

LL (n ⴝ 6)

Controls

Saline

Controls

Saline

Controls

Saline

12 ⫾ 7 49 ⫾ 6

21 ⫾ 6 21 ⫾ 7*

1⫾3 39 ⫾ 3

16 ⫾ 9 31 ⫾ 7

8⫾5 53 ⫾ 6

17 ⫾ 8 36 ⫾ 9

* P ⬍ .01 v controls; two-way ANOVA: factor strain, NS; factor diet, P ⬍ .05 for ⌬ SBP, P ⬍ .005 for ⌬ BW.

408

AJH–APRIL 2000 –VOL. 13, NO. 4, PART 1

BERNARD ET AL

of the kidney samples in the presence of an RNAse inhibitor. Quantification of cDNA was allowed by comparison with standards but the efficiency of reverse transcription could not be directly controlled, as no pure RNA was available for amplification. Constitutive COX1 expression was easily detectable. It was more important and stable with age in the medulla than in the cortex. On the contrary, COX2 expression was equally distributed at a low level in the cortex and medulla. It decreased with age, this decrease being significant in the cortex of LH rats only. Such a low level of expression might be due to the macrodissection technique of sampling that we used, which did not allow for separating vascular, interstitial, and tubular cells. However, our data are in good accordance with biochemical measurements of the COX activity in kidneys of developing Wistar rats22 and with the corticopapillary distribution of COX1 and 2 mRNA in Sprague Dawley rats.10 The level of expression of TXS was either undetectable or too low to be quantified. This result could indicate that PGH2 is poorly transformed into TXA2 in rat kidneys. Otherwise it suggests that the production of TXA2 does not require a de novo synthesis of the enzyme or is more dependent upon enzymes that control the amount of its immediate precursor such as phospholipase A2.23,24 Finally, one cannot exclude a stabilization of the mRNA or a direct oxidative activation of the enzyme,25 which might induce a higher enzymatic activity with a similar content in mRNA. In any case, the lack of differences between LH rats and their LN and LL controls demonstrates that their previously described increased renal synthesis of TXA23 occurs at a posttranscriptional level or may be restricted to the vessel walls, thus being indiscernible in a cortex or medulla total extract. A 1-week salt load produced a similar slight increase in SBP in the three strains. It increased COX1 in the medulla of all kidneys, the increase being significant in LH rats only. The response to salt loading reflected an increase only in COX1 in medulla; it is unlikely this response is due to the production of vasoconstrictor PGH2 or TXA2, but is probably due to the vasodilatory prostaglandin PGE2 which is predominant at this level. The ability of the LH rat to produce prostaglandins to eliminate the salt excess is not different from that of its controls and cannot explain its lower natriuresis. Levels of COX2 were not significantly modified by salt loading in the three strains. These divergent responses of COX1 and COX2 to salt loading are in good agreement with in vivo and in vitro experiments, describing their opposite sensitivity to osmolality.7,10 –11 In conclusion, the present work excludes that a generalized increase in the expression of COX1, COX2, and TXA2 synthase plays a primary role in the devel-

opment of hypertension in Lyon rats, but demonstrates that an upregulation of COX1 by sodium exists in the renal medulla of LH animals. This may be of importance for their blood pressure increase and their elevated sensitivity to salt. REFERENCES 1.

Vincent M, Sacquet J, Sassard J: The Lyon strains of hypertensive, normotensive and low blood pressure rats, in de JongW (ed): Handbook of Hypertension, Volume 4. Elsevier Science Co., Amsterdam, 1984, pp 314 –327.

2.

Liu KL, Benzoni D, Sassard J: Prostaglandin H2-thromboxane A2 and renal functions in the Lyon hypertensive rats. Am J Physiol 1994;266:R1530 –R1536.

3.

Geoffroy J, Benzoni D, Vincent M, Sassard J: Thromboxane A2 and development of genetic hypertension in the Lyon rat strain. Hypertension 1990;16:655– 661.

4.

Geoffroy J, Benzoni D, Sassard J: Anti-hypertensive effect of thromboxane A2 receptor blockade in genetically hypertensive rat of the Lyon strain. J Hypertens 1989;7(suppl 6):S272–S273.

5.

Bonvalet JP, Pradelles P, Farman N: Segmental synthesis and actions of prostaglandins along the nephron. Am J Physiol 1987;253:F377–F387.

6.

Badr KF, Jacobson HR: Arachidonic acid metabolites and the kidney, in Brenner BM, Rector FC (eds): The Kidney, Volume 1. WB Saunders, Philadelphia, 1991, pp 584 – 619.

7.

Cowley DB, Muessel MJ, Douglass D, Wilkins W: In vivo and in vitro osmotic regulation of HSP-70 and prostaglandin synthase gene expression in kidney cells. Am J Physiol 1995;269:F854 –F862.

8.

Ge T, Hughes H, Junquero DC, Wu KK, Vanhoutte PM, Boulanger CM: Endothelium-dependent contractions are associated with both augmented expression of prostaglandin H2 synthase-1 and hypersensitivity to prostaglandin H2 in the SHR aorta. Circ Res 1995;76: 1003–1010.

9.

Meade EA, Smith WL, DeWitt DL: Differential inhibition of prostaglandin endoperoxide synthase (cyclooxygenase) isozymes by aspirin and other non-steroidal anti-inflammatory drugs. J Biol Chem 1993;268: 6610 – 6614.

10.

Jensen BL, Kurtz A: Differential regulation of renal cyclooxygenase mRNA by dietary salt intake. Kidney Int 1997;52:1242–1249.

11.

Yang T, Singh I, Pham H, Sun D, Smart A, Schnermann JB, Briggs JP: Regulation of cyclooxygenase expression in the kidney by dietary salt intake. Am J Physiol 1998;274:F481–F489.

12.

Harding P, Sigmon DH, Alfie ME, Huang PL, Fishman MC, Beierwaltes WH, Carretero OA: Cyclooxygenase-2 mediates increased renal renin content induced by lowsodium diet. Hypertension 1997;29:297–302.

13.

Blanc-Brunat N, Vincent M, Sassard J: Evolution des le´sions re´nales pre´coces chez trois souches de rats ge´ne´tiquement hypertendus, normotendus et normotendus bas. Arterial Wall 1980;6:145–156.

14.

Chomczynski P, Sacchi N: Single-step method of RNA

AJH–APRIL 2000 –VOL. 13, NO. 4, PART 1

PROSTANOID SYNTHESIS IN LH RATS

sory role of whole-body autoregulation in the pathogenesis of hypertension. Am J Hypertens 1989;2:575–585.

isolation by acid guanidium thiocyanate-phenol-chloroform extraction. Anal Biochem 1987;162:156 –159. 15.

16.

Kawaguchi H, Yavari R, Stover ML, Rowe DW, Raisz G, Pilbeam CC: Measurement of interleukin-1 stimulated constitutive prostaglandin G/H synthase (cyclooxygenase) mRNA levels in osteoblastic MC3T3-E1 cells using competitive reverse transcriptase polymerase chain reaction. Endocr Res 1994;20:219 –233. Kitzler J, Hill E, Hardman R, Reddy N, Philpot R, Eling TE: Analysis and quantitation of splicing variants of the TPA-inducible COX-1 mRNA in rat tracheal epithelial cells. Arch Biochem Biophys 1995;316:856 – 863.

17.

Tone Y, Miyata A, Hara S, Yukawa S, Tanabe T: Abundant expression of thromboxane synthase in rat macrophages. FEBS Lett 1994;340:241–244.

18.

Zhang L, Chase MB, Shen RF: Molecular cloning and expression of murine thromboxane synthase. Biochem Biophys Res Commun 1993;194:741–748.

19.

Hughes AK, Padilla E, Kutchera WA, Michael JR, Kohan DE: Endothelin-1 induction of cyclooxygenase-2 expression in rat mesangial cells. Kidney Int 1995;47: 53– 61.

20.

Guyton AC: Dominant role of the kidneys and acces-

409

21.

Cowley AW Jr: Genetic and nongenetic determinants of salt sensitivity and blood pressure. Am J Clin Nutr 1997;65:587S–593S.

22.

Fernandez-Tome MC, D’Antuono C, Kahane VL, Speziale EH, Sterin-Speziale NB: Compartmental study of rat renal prostaglandin synthesis during development. Biol Neonate 1996;70:235–245.

23.

Nanayama T, Hara S, Inone H, Yokoyama C, Tanabe T: Regulation of two isozymes of prostaglandin endoperoxide synthase and thromboxane synthase in human monoblastoid cell line U937. Prostaglandins 1995;49: 371–382.

24.

Matijevic-Aleksic N, Sanduja SK, Wang LH, Wu KK: Differential expression of TXA synthase and prostaglandin H synthase in megakaroycytic cell line. Biochim Biophys Acta 1995;1269:167–175.

25.

Davidge ST, Baker PN, McLaughlin MK, Roberts JM: Nitric oxide produced by endothelial cells increases production of eicosanoids through activation of prostaglandin H synthase. Circ Res 1995;77:274 –283.