Characterization of the Renal Phenotype of Transgenic Rats Expressing the Human Endothelin-2 Gene
We have previously established a transgenic rat model termed TGR(hET-2)37 overexpressing the human endothelin-2 (ET-2) gene with high renal transgene expression. This renal overexpression is of pathophysiological interest because a long-term activated paracrine renal endothelin system has been implicated in chronic renal failure due to progressive glomerular injury. Therefore, our aim in the present study was to analyze renal transgene expression in detail and address the question of whether transgene expression causes phenotypic and functional changes in the kidney. We used reverse transcription-polymerase chain reaction and in situ hybridization techniques for transgene expression analysis. Tissue ET-2 concentrations were measured with a specific radioimmunoassay. For histological evaluation of renal tissue, all samples were subjected to hematoxylin-eosin and periodic acid-Schiff staining. Renal tissue ET-2 concentrations were significantly increased in TGR(hET-2)37 rats. Using in situ hybridization, we found that the human ET-2 gene was almost exclusively expressed within the glomeruli. The glomerular transgene expression resulted in a significantly increased glomerular injury score and likewise in a significantly increased protein excretion, whereas glomerular filtration rate was not altered. Blood pressure was similar in TGR(hET-2)37 rats and age-matched controls, suggesting that the local changes in the kidney were correlated with paracrine endothelin actions. In conclusion, our study revealed that the major renal expression site of the human ET-2 transgene in TGR(hET-2)37 rats was within the glomeruli and caused the development of glomerulosclerosis with significantly increased protein excretion that is independent of blood pressure. We suggest that TGR(hET-2)37 rats are a new monogenetic animal model for study of the paracrine renal endothelin system and its involvement in renal pathophysiology.
Endothelins are among the most potent vasoconstrictor substances active in vivo. ET-1 was the first of the isopeptides to be characterized in the supernatant of cultured vascular endothelial cells.1 The biological effects of endothelins are mediated by plasma membrane-bound receptors that belong to the family of rhodopsin-like receptors with seven-transmembrane domains coupled to different G proteins. Three endothelin receptor subtypes (ETA, ETB, and recently, ETC) have been cloned.2 3 4 They differ in their binding affinity to endothelin isopeptides (ETA: ET-1≥ET-2>>ET-3; ETB: ET-1=ET-2=ET-3; ETC: ET-3>>ET-2=ET-1). The ETC receptor, however, has not been described in mammals to date. In addition to their vasoactive properties, endothelins have been shown to elicit a wide variety of biological activities in nonvascular tissues, including the kidney. It is now generally accepted that the paracrine renal endothelin system is involved in the regulation of renal blood flow, glomerular filtration rate, and tubular water and sodium reabsorption.5 Apart from these effects of endothelin on renal function, it has been shown that endothelins are able to stimulate vascular smooth muscle cell and mesangial cell proliferation.6 Furthermore, the endothelin system appears to be involved in matrix protein production in the kidney (for review, see Reference 7). This implies a possible involvement of the paracrine endothelin system in the pathogenesis of glomerulosclerosis. Interestingly, several reports8 9 10 show a correlation between glomerulonephritis/glomerulosclerosis and an activated renal endothelin system. However, it remains to be clarified whether a primary, endogenous activation of the paracrine endothelin system is a cause or consequence of glomerular injury. The best way to answer this question is to analyze animals with an activated renal paracrine endothelin system without any other stimuli that may contribute to glomerulosclerosis. We therefore used human ET-2 transgenic rats11 that were generated with a microinjection technique for fertilized oocytes as described by Paul et al.12 The genomic DNA construct contained the entire human ET-2 gene under the control of its own promoter. Northern blot analysis has shown that these rats exhibit overexpression of human ET-2 mRNA, with high levels in the kidney.13 Since these rats have normal blood pressure, hypertensive renal changes could be excluded. In the present study, we analyzed the renal ET-2 gene expression in further detail using in situ hybridization, RT-PCR, and measurement of tissue immunoreactive endothelin concentrations. Furthermore, we analyzed whether human ET-2 gene expression in rats resulted in renal injury, altered blood pressure, and impaired renal function.
All experiments were performed in 12-month-old female Sprague-Dawley rats heterozygous for the human ET-2 transgene (line 37) that we named TGR(hET-2)37. As controls we used nontransgenic littermates. Southern blot and PCR analyses with human-specific ET-2 probes were used for detection of rats that carried the transgene.
Blood Pressure Measurement
Blood pressure was measured as recently described.14 Briefly, 1 week before measurements, the rats (n=7 in each group) were anesthetized with ether, and femoral artery and vein catheters were implanted. Arterial blood pressure and heart rate were measured in conscious rats via the arterial line with a pressure transducer (Statham P23Db) and a pressure processor (Gould Instruments) coupled to a recorder (Gould Brush 2600).
Serum Values (Clinical Chemistry)
Plasma and urinary levels of potassium, sodium, creatinine, and protein were determined with the appropriate commercial kits in an automatic analyzer. Blood samples for determination of serum concentrations were drawn after urine was collected over 24 hours. The absolute excretion of either sodium, potassium, or protein was calculated by the equation Ux×V24, where Ux is the concentration of sodium, protein, or potassium and V24 is the amount of urine excreted in 24 hours.
Tissue immunoreactive endothelin levels were measured with a method described recently.15 After thawing, kidney samples were suspended in 4 mL/g wet wt of acetic acid (0.2 mol/L)/NaCl (0.5 mol/L), disrupted with a polytron, and subsequently homogenized. The homogenate was centrifuged at 4°C for 15 minutes at 25 000g. The supernatant was retained for endothelin radioimmunoassay.
Immunoreactive ET-2 was extracted from tissue supernatant with 500-mg C2 columns (Amprep, Amersham). The columns were equilibrated with 2 mL methanol (100%) and 2 mL distilled water. Samples (1-mL) were acidified with 0.25 mL of 2 mol/L HCl, centrifuged at 10 000g for 5 minutes at room temperature, and loaded onto the columns. The columns were washed twice with 2 mL distilled water plus 0.1% trifluoroacetic acid. The adsorbed peptide was eluted with 2 mL of 80% methanol in water plus 0.1% trifluoroacetic acid. The eluates were collected in polypropylene tubes and dried under a stream of nitrogen.
The samples were reconstituted in 250 μL radioimmunoassay buffer (20 mmol/L borate buffer, pH 7.4, containing 0.1% sodium azide) and analyzed with a commercial 125I-endothelin radioimmunoassay kit (Endothelin-1,2 [high sensitivity] 125I Assay System, Amersham). Separation of the antibody-bound fraction was produced by magnetic separation with an Amerlex-M Separator (Amersham). This assay reacts 100% with ET-2 and cross-reacts 7.7% with ET-1. Cross-reactivity with ET-3 was less than 0.1%.
RT-PCR was performed according to Paul et al16 with the following human-specific oligonucleotide primers described by O'Reilly et al17 : sense: 5′-AGCGTCCTCATCTCATGCCC-3′; antisense: 5′-TCTCTTCCTCCACCTGGAATG-3′; fragment length, 436 bp.
RNA was prepared from kidneys with TRIzol reagent (GIBCO-BRL); RNA samples were reverse transcribed (Moloney murine leukemia virus [MMLV] reverse transcriptase, GIBCO-BRL); and the resultant cDNA of 0.5 μg total RNA was amplified with a Programmable Thermal Controller (PTC-100, MJ Research, Inc). After pretreatment (94°C, 3 minutes), a step program (94°C, 30 seconds; 56°C, 45 seconds; 72°C, 60 seconds; 40 cycles) was performed followed by a final extension reaction (72°C, 6 minutes). Taq DNA polymerase was purchased from GIBCO-BRL.
In Situ Hybridization
Kidney samples were snap-frozen in liquid nitrogen and stored at −70°C. Cryostat sections (5 μm) were placed on siliconized slides treated with 3-aminopropyltriethoxysilane for improved adherence and dried on a hot plate at 80°C for 3 minutes. Tissue sections were fixed in 4% paraformaldehyde/PBS (pH 7.4) for 20 minutes and dehydrated in graded ethanol.
We used a human ET-2 fragment subcloned in pBluescript 2+ SK plasmid (Stratagene). The plasmid was linearized by digestion with HindIII and EcoRI. Single-stranded RNA probes, complementary (antisense probe) or anticomplementary (sense probe, negative control) to cellular RNA, were obtained by runoff transcription with T7 or T3 RNA polymerase (transcription kit from Boehringer-Mannheim). 35S-UTP was used for labeling of RNA probes. The specific activity of the probes was 1.0×109 to 1.5×109 cpm/mg RNA.
In Situ Hybridization
Prehybridization, hybridization, washing, and RNAse A digestion for removal of nonspecifically bound probes as well as autoradiography were performed as recently described,17 18 with minor modifications. Tissue sections were treated with 0.2 mol/L HCl for 20 minutes, digested in 0.125 mg/mL pronase (Boehringer-Mannheim) for 10 minutes at 22°C, rinsed in 0.1 mol/L glycine/PBS, and washed in PBS. After an additional fixation step in 4% paraformaldehyde/PBS for 15 minutes, tissues were acetylated in a solution of acetic anhydride/triethanolamine (0.1 mol/L, pH 8.0), rinsed in PBS, dehydrated in graded ethanol, and finally air-dried. Each slide was covered with 0.025 mL hybridization mixture containing 1×105 to 3×105 cpm labeled RNA probe in a solution of 50% formamide, 10% dextran sulfate, 10 mmol/L dithiothreitol, 10 mmol/L Tris-HCl (pH 7.5), 10 mmol/L Na2HPO4, 0.3 mol/L NaCl, 5 mmol/L EDTA, and 0.2 mg/mL yeast tRNA. Sections were sealed with a siliconized coverslip. After 18 hours of incubation at 50°C in a humidified chamber, slides were washed for 4 hours at 50°C in a solution of 50% formamide, 10 mmol/L dithiothreitol, and 1× SALTS (300 mmol/L sodium chloride; 10 mmol/L Tris-HCl, pH 7.5; 10 mmol/L sodium phosphate, pH 6.8; 5 mmol/L EDTA, pH 8.0; in diethylpyrocarbonate-treated water) and for 15 minutes in TES buffer (10 mmol/L Tris-HCl [pH 7.5], 0.5 mol/L NaCl, 1 mmol/L EDTA) at 37°C; they were then digested with RNAse A to reduce background caused by nonspecific binding. Subsequently, tissue sections were washed in TES buffer for 30 minutes and finally rinsed in 2× SSC (1× SSC = 150 mmol/L sodium chloride and 15 mmol/L sodium citrate, pH 7.0), 0.1× SSC, and 0.05× SSC for 20 minutes each at 22°C. The sections were dehydrated in graded ethanol, air-dried, and dipped in K5 photoemulsion (Ilford Mobberley). After exposure for 10 to 28 days at 4°C, sections were developed for 2.5 minutes with D 19 developer (Eastman Kodak), rinsed in 1% acetic acid, fixed in Kodak fixer for 2.5 minutes, washed in water, and counterstained with hematoxylin-eosin. All tissues were simultaneously processed with the same probes and reagents (n=6 in each group). Specific binding was defined as the silver granules per cell nucleus in the antisense probes minus the silver granules per cell nucleus in the corresponding sense probes. We have analyzed transgene expression in renal blood vessels, glomeruli, and renal tubules according to this protocol.
For pathohistological evaluation of the renal tissue, all samples were subjected to hematoxylin-eosin and PAS staining. The extent of glomerulosclerosis was analyzed according to Raij et al.18 Glomerulosclerosis was defined by the presence of increased amounts of PAS-positive material within the glomeruli. To examine differences in the degree of glomerulosclerosis, we used a semiquantitative score. We analyzed both kidneys of six transgenic rats and six age-matched controls. A minimum of 80 glomeruli in each specimen were examined, and the degree of lesion was graded from 0 to 4+ according to the percentage of glomerular involvement. Thus, a 1+ lesion represented an involvement of 25% of the glomerulus, and a 4+ lesion indicated that 100% of the glomerulus was PAS positive. An injury score was then obtained by multiplying the degree of damage (0 to 4) by the percentage of the glomeruli with the same degree of injury. The extent of injury for each individual tissue specimen was then obtained by addition of these scores. For example, if 10 of 20 glomeruli had a lesion of 1+ and 10 of 20 had a lesion of 3+, the final injury score for that specimen would be 10×1/20+10×3/20=2. All tissue samples were evaluated blindly by two independent investigators.
One-way ANOVA was performed for determination of whether there were any differences between the two groups of data. Subsequently, unpaired Student's t test was used for determination of the statistical difference of group means. Results were considered significantly different at a value of P<.05.
We could demonstrate by RT-PCR that all human ET-2 transgenic rats (line 37) exhibited expression of human ET-2 mRNA in the kidney, whereas nontransgenic littermates showed no ET-2 mRNA signal (Fig 1⇓). With in situ hybridization (Fig 2⇓), significantly increased ET-2 mRNA signal densities were detectable in the glomeruli of heterozygous TGR(hET-2)37 rats only (sense, 0.71±0.43 silver grains per cell nucleus; antisense, 4.94±0.79 silver granules per cell nucleus; P<.05; n=6). The human ET-2 gene was homogeneously expressed throughout the glomeruli in the cortex and medulla. On the other hand, the ET-2 signal in the smooth muscle cells of intrarenal arteries and renal tubules showed only weak signal densities. Furthermore, transgene expression in the kidneys resulted in a significantly increased tissue concentration of immunoreactive endothelin (Fig 3⇓), indicating that the transgene is correctly transcribed and translated.
Heart rate (data not shown) and systolic pressure (Fig 4⇓) were similar in female 12-month-old heterozygous TGR(hET-2)37 rats and age-matched littermates.
Despite normal blood pressure, heterozygous TGR(hET-2)37 rats developed significant morphological changes in the kidney compared with age-matched controls. Transgenic rats had a significantly increased amount of PAS-positive material within the glomeruli (Fig 5⇓). Using the glomerular injury score according to Raij et al,18 we found a significantly increased glomerulosclerosis index in TGR(hET-2)37 rats (Fig 6⇓). The affected glomeruli were homogeneously distributed within the kidneys. In addition to glomerulosclerosis, perivascular infiltration of mononuclear cells was detectable in the kidneys of transgenic rats.
Body weight was similar in 12-month-old heterozygous female TGR(hET-2)37 rats and corresponding controls. Kidney weight was not significantly decreased in transgenic rats. Heart weight of transgenic rats showed a nonsignificant increase compared with that of nontransgenic littermates (Table 1⇓). In addition, creatinine clearance was similar in heterozygous TGR(hET-2)37 rats compared with corresponding controls, whereas urinary protein concentration and protein excretion were significantly increased in 12-month-old heterozygous female TGR(hET-2)37 rats compared with corresponding controls (Table 2⇓). Serum potassium concentration was also increased in human ET-2 transgenic rats. However, this difference was not significant (P=.12) (Table 2⇓).
Expression of Human ET-2 Transgene in Rat Kidney
Using RT-PCR, we could not detect ET-2 mRNA in the kidneys of nontransgenic control rats (Fig 1⇑). These data are in agreement with a report demonstrating the absence of ET-2 gene expression in normal rat kidney.19 Therefore, generation of a transgenic rat model using the human ET-2 gene could allow the study of renal morphological and functional changes caused by ET-2 overexpression. Although the almost similar effects of ET-1 and ET-2 in vitro and in vivo should be borne in mind, this transgenic line serves as a pharmacological model for the study of chronic overexpression of endothelins in the kidney.
The increased ET-2 mRNA signal in the transgenic rats seen by RT-PCR is clearly due to transgene expression within the glomeruli (Fig 2⇑), resulting in a significantly increased tissue immunoreactive ET-2 concentration (Fig 3⇑). The detection of a basal immunoreactive ET-2 tissue concentration in nontransgenic littermates is most probably due to the cross-reactivity of the ET-2 radioimmunoassay used in our study, because Northern blotting revealed no ET-2 gene expression in the rat kidney of nontransgenic rats (data not shown).
In general, extrarenal tissues of heterozygous TGR(hET-2)37 rats had lower human ET-2 expression, as recently shown by Liefeldt et al.13 The mechanisms leading to tissue differences in ET-2 transgene expression remain unknown. Further studies therefore are necessary for elucidation of which transacting factors are differentially expressed in different rat tissues, especially within the different glomerular cell types (eg, mesangial cells, podocytes, and glomerular endothelial cells).
Induction of Glomerulosclerosis Without Hypertension in ET-2 Transgenic Rats
There is accumulating evidence that an activated renal paracrine endothelin system is involved in the pathogenesis of glomerulosclerosis.8 9 10 Several studies have demonstrated a correlation between an activated endothelin system and glomerular injury. However, clear evidence showing that glomerular endothelin overexpression results in glomerular injury has not been reported previously. Therefore, the transgenic approach has been used to answer this question because the primary event of the cascade leading to glomerulosclerosis (glomerular human ET-2 gene overexpression) is well defined. However, it is important to note in this context that overexpression of the human ET-2 gene in transgenic rats represents an animal model of a primary activated paracrine renal endothelin system without known direct counterparts in human diseases. In humans, a secondary activation of the renal endothelin system (namely, ET-1) due to hypoxia, cyclosporine treatment, or inflammatory kidney diseases such as lupus nephritis seems to be more important.20 Despite this, the ET-2 transgenic rat model offers a unique opportunity for analysis of the effect of an activated renal endothelin system (either primary or secondary activated) on renal pathology and function. The ET-2-induced development of glomerulosclerosis leads to impaired renal function, as indicated by significantly elevated proteinuria. This is in accordance with glomerular injury in humans, where proteinuria is an early clinical indicator of this alteration, preceding the decrease of glomerular filtration rate by months or years. However, further studies that use more-sensitive techniques for the detailed analysis of renal function and matrix protein synthesis are necessary.
The observation that heterozygous TGR(hET-2)37 rats are normotensive is surprising because ET-2 is a strong vasoconstrictor in vivo.21 22 23 24 This may be due to the fact that TGR(hET-2)37 rats have a very low rate of transgene expression in extrarenal vascular beds and the heart (L.L., unpublished data, 1996).
The development of blood pressure-independent glomerulosclerosis in heterozygous TGR(hET-2)37 rats demonstrates that the human ET-2 gene can have a blood pressure-independent growth-promoting effect on the rat glomerulus. It is important to note that these effects occur despite a rather low expression rate of the ET-2 gene, indicating that ET-2 is a potent peptide that induces matrix protein synthesis in vivo. Endothelin growth-promoting effects appear to be mediated by the ETA receptor subtype in the rat kidney, as it has been shown in cultured mesangial cells that ETA receptor blockade reduces ET-1-stimulated cell growth and extracellular matrix formation.25 Furthermore, Orisio et al9 recently reported that renal preproET-1 gene is upregulated in rats with renal mass reduction and that time-dependent increases in the urinary excretion of ET-1 correlate with renal disease progression. ETA receptor blockade reduced the abnormal permeability to proteins, limited glomerular injury, and prevented renal functional impairment,26 thus confirming the hypothesis that the ETA receptor subtype is involved in such chronic glomerular diseases. Interestingly, the renal ETA receptor is not downregulated in TGR(hET-2)37 rats,13 indicating that the progression of glomerular injury in these rats may also be mediated by ETA. The endothelin-stimulated, ETA-mediated mitogenesis of rat mesangial cells is induced at least in part via induction of c-fos and c-myc expression.27 28
In summary, our study revealed that the major renal expression site of the human ET-2 transgene of heterozygous transgenic rats was within the glomeruli. Glomerular expression of the human ET-2 gene in rats resulted in the development of a blood pressure-independent glomerulosclerosis with significantly increased urinary protein excretion, suggesting that TGR(hET-2)37 rats are a new monogenetic animal model of glomerulosclerosis. Further studies are necessary for analysis of which matrix proteins are involved in this process and which endothelin receptor subtype mediates the ET-2 growth effects in the rat glomerulus, for example, by use of specific endothelin receptor antagonists. With the latter, the amount of proteinuria might be an in vivo indicator of the prevention of glomerulosclerosis.
Selected Abbreviations and Acronyms
|PCR||=||polymerase chain reaction|
B. Hocher and L. Liefeldt have equally contributed to this study and are listed in alphabetical order. The study was supported by grants from the Deutsche Forschungsgemeinschaft (Pa 332/2-1) as well as by the Transgeneur program of the European Community (BIOMED) to M.P. and Fonds der Chemischen Industrie to B.H. and C.B. The technical assistance of H. Marquardt, B. Schwaneberg, and S. Schiller is greatly appreciated. We would like to thank Dr Julie Yu for critically reading the manuscript.
- Received February 22, 1996.
- Revision received March 5, 1996.
- Accepted March 5, 1996.
Yanagisawa M, Kurihara H, Kimura S, Tomobe Y, Kobayashi M, Mitsui Y, Yazaki Y, Goto K, Masaki T. A novel potent vasoconstrictor peptide produced by vascular endothelial cells. Nature. 1988;32:411-415.
Haendler B, Hechler U, Schleuning WD. Molecular cloning of human endothelin (ET) receptors ETA and ETB. J Cardiovasc Pharmacol. 1992;20(suppl 12):S1-S4.
Kumar C, Mwangi V, Nuthulaganti P, Wu HL, Pullen M, Brun K, Aiyar H, Morris RA, Naughton R, Nambi P. Cloning and characterization of a novel endothelin receptor from Xenopus heart. J Biol Chem. 1994;269:13414-13420.
Gross P, Renn C, Waldherr R, Seifert M, von Baehr R, Hocher B. Potential role of endothelin in the physiological and pathological regulation of kidney function. Endothelium. 1993;1:71-83.
Paul M, Rettig R, Talsness CE, Zintz M, Yanagisawa M. Transgenic rats expressing the human endothelin-2 gene: a new model to study endothelin regulation in vivo. J Hypertens. 1994;12:S72. Abstract.
Liefeldt L, Böker W, Schönfelder G, Zintz M, Paul M. Regulation of the endothelin system in transgenic rats expressing the human endothelin-2 gene. J Cardiovasc Pharmacol. 1995;26(suppl 3):S32-S33.
Paul M, Wagner J, Dzau VJ. Gene expression of the renin-angiotensin system in human tissues: quantitative analysis by the polymerase chain reaction. J Clin Invest. 1993;91:2058-2064.
O'Reilly G, Charnock-Jones DS, Cameron IT, Smith SK, Davenport AP. Endothelin-2 mRNA splice variants detected by RT-PCR in cultured human vascular smooth muscle and endothelial cells. J Cardiovasc Pharmacol. 1993:22(suppl 8):S18-S21.
Firth JD, Ratcliff PJ. Organ distribution of the three rat endothelin RNAs and the effects of ischemia on renal gene expression. J Clin Invest. 1992;90:1023-1031.
Guimaraes CL, Calixto JB, Rae GA. Potent constrictor actions of endothelin-1, endothelin-2, and endothelin-3 in rat isolated portal vein. Hypertension. 1992;19(suppl II):II-79-II-86.
Ishimura E. Shouji S. Nishazawa Y, Mori H, Kashgarian M. Regulation of mRNA expression of extracellular matrix (ECM) by cultured rat mesangial cells. J Am Soc Nephrol. 1991;2:546.
Simonson MS, Wann S, Mene P, Dubyak GR, Kester M, Nakazato Y, Sedor JR, Dunn MJ. Endothelin stimulates phospholipase C Na+/H+ exchange, c-fos expression, and mitogenesis in rat mesangial cells. J Clin Invest. 1989;83:708-712.
Simonson MS, Herman WH. Protein kinase C and protein tyrosine kinase activity contribute to mitogenic signaling by endothelin-1: cross-talk between G protein-coupled receptors and pp60c-src. J Biol Chem. 1993;268:9347-9357.