Relaxin Improves Renal Function and Histology in Aging Munich Wistar Rats
Lee A. Danielson,
Angela Welford and
Alexis Harris
Department of Pathology, University of New Mexico School of Medicine, Albuquerque, New Mexico
Address correspondence to: Dr. Lee A. Danielson, University of New Mexico, MSC09 5250, Albuquerque, NM 87131-0001. Phone: 505-272-5509; Fax: 505-272-8079; E-mail: ldanielson{at}salud.unm.edu
Received for publication December 16, 2005.
Accepted for publication February 17, 2006.
Administration of recombinant human relaxin (rhRLX) to conscious,chronically instrumented rats increases GFR and effective renalplasma flow (ERPF) and decreases effective renal vascular resistance(ERVR) with no significant change in mean arterial pressure.The Munich Wistar albino rat shows progressive chronic nephrosiswith age and therefore was used to determine the functionaland histologic consequences of rhRLX on matrix remodeling inthe kidney of older rats. RLX-infused rats showed increasedGFR and ERPF with decreased ERVR. Furthermore, in a double-blindedexamination, the renal histology showed a significant decreasein glomerular and tubular collagen deposition in the rhRLX-infusedaged rats. During short-term rhRLX administration (24 h), gelatinaseactivity was found to be essential for renal vasodilation andhyperfiltration. Surprisingly, after 20 d, improved renal functionwas insensitive to the inhibition of gelatinase activity, suggestingthat collagen degradation in these rats had permanently alteredthe matrix of the renal vasculature. In conclusion, long-termadministration of rhRLX improves renal function and amelioratesrenal pathology in an aging rat model. The biphasic action ofrhRLX on the kidney indicates that, acutely, the vessels dilate,causing increased filtration and renal blood flow with decreasedvascular resistance as a result of upregulation of gelatinaseactivity. Subsequently, the renal vessels undergo alterationin supporting matrix, showing increased blood supply even inthe face of acute matrix metalloproteinase inhibition, mostlikely as a result of the inhibitory properties of RLX on collagenproduction or increased collagen breakdown.
Relaxin (RLX) has numerous biologic effects. Its role in thereproductive tract has been well documented (1). More recently,we have shown it to be a potent vasodilator in the kidney (2–5).In long-term instrumented conscious rats, we demonstrated thatRLX stimulates nitric oxide (NO)-dependent renal vasodilationand hyperfiltration in vivo, and reduces myogenic reactivityof small renal arteries in vitro via endothelin (ET) and theendothelial ETB receptor subtype (5). We also have documenteda role for the matrix metalloproteinases (MMP) in this phenomenon.Vascular gelatinase activity converts big ET to ET1-32, whichpotently activates the endothelial ETB receptor. Using a novelinhibitor of the gelatinases, cyclic CTTHWGFTLC (CTT), we demonstratedthat vascular gelatinase activity mediates renal vasodilatoryaction of RLX in rats (5). MMP-2 was specifically implicatedin this phenomenon because an MMP-2–neutralizing antibodyblocked the reduction in myogenic reactivity in isolated renalvessels that is typically caused by long-term RLX administrationin vivo (5).
RLX is widely known for its "matrix-degrading" properties thatare associated with the stimulation of MMP and inhibition ofcollagen expression (6–8). Specifically, RLX decreasessynthesis and secretion of interstitial collagens, increasesexpression of the procollagenase (MMP-1), and decreases productionof tissue inhibitor of metalloproteinases (TIMP) by human dermalfibroblasts (6). In a murine model of induced pulmonary fibrosis,RLX inhibits the formation of abnormal extracellular matrix(ECM) in vivo (7). In human lower uterine fibroblasts, RLX stimulatesthe production of MMP-2 (8). Long-term recombinant human RLX(rhRLX; BAS Medical, Palo Alto, CA) infusion upregulates MMP-2expression in isolated rat arteries (5). Furthermore, RLX decreasesrenal fibrosis in a model of chemically induced papillary necrosis,seemingly independent of systemic hemodynamic changes (9). Intwo models of renal mass reduction, RLX decreased renal injuryin the rat by at least two mechanisms: lowering BP and reducingglomerular and tubular sclerosis (10). In recent studies ofthe RLX gene knockout mouse, the kidney of male mice showedincreased kidney weight and size with a build-up of collagen(11). Despite these intriguing correlative studies implicatingRLX in the control of matrix remodeling, the mechanisms havenot been established. Because of the significant renal vasculareffects and matrix-remodeling properties of the hormone RLXand its connection with the metalloproteinases, especially gelatinaseactivity in the kidney, this study was designed to investigatea possible role for RLX and gelatinases in ameliorating age-relatedchanges that are seen in the kidney.
Animal Preparation
Munich-Wistar male rats that were 3, 6, and 12 mo of age werepurchased from Harlan Sprague Dawley (Frederick, MD) and providedTechlad Rodent Irradiated Diet 2018 that contained 0.23% sodium(Harlan-Techlad Feed, Madison, WI) and water ad libitum. Therats were maintained on a 12-h light/dark cycle in an InstitutionalAnimal Care and Use Committee–approved animal resourcefacility at the University of New Mexico. In preparation forexperimentation, the rats were trained for several hours inan appropriate-size Plexiglas restraining cage (Braintree ScientificCo., Braintree, MA) on at least five different occasions beforelong-term instrumentation.
Details of the surgical procedures have been described previously(3). A single-staged surgery was used in most rats; a double-stagesurgery was used on the rats that were prepared for acute MMPinhibition. Briefly, using isoflurane/oxygen gas for anestheticpurposes and aseptic technique, we implanted Tygon vascularcatheters in the abdominal aorta and inferior vena cava viathe femoral artery and vein, respectively. Then, a silastic-coveredstainless steel cannula was sewn into the urinary bladder witha purse-string suture and exteriorized through the ventral abdominalwall. For the infusion of metalloproteinase inhibitors, a catheterwas implanted in the right carotid artery. All vascular catheterswere tunneled subcutaneously and exteriorized between the scapulae.
Experimental Protocols Effects on Renal Function during Chronic Infusion of rhRLX.
Renal function and mean arterial pressure (MAP) were measuredin a pair of age-matched, chronically instrumented consciousrats on day 0 (baseline) and 1, 3, 5, and 10 d of rhRLX/vehicleinfusion. Before infusion of any fluids, blood was drawn forbaseline blood chemistries. MAP/heart rate (HR) was measuredby a Gould P23ID pressure transducer and Gilson ICT-2H Duograph(Middleton, WI). The obturator in the bladder catheter was removedand extended with a short piece of silastic tubing for collectionof urine. A bolus of inulin (IN; Sigma, St. Louis, MO; 0.1 mlof a 50% stock solution/100 g body wt) and para-aminohippurate(PAH; (Merck & Co., West Point, PA; 0.05 ml of a 4% workingsolution/100 g body wt) was given into the venous catheter followedby constant infusion (12 µl/min; Sage Instruments, Houston,TX) of the two reagents at 0.4 and 0.1 mg/min per 100 g bodywt, respectively. After a 60-min equilibration, three baselineurine collections of 30 min each with midpoint blood collectionsof 300 µl each were obtained to measure the renal clearancesof IN and PAH, which provide estimates of GFR and effectiverenal plasma flow (ERPF), respectively. After centrifugationand removal of plasma, all red cells were gently resuspendedin Ringer’s solution and returned to the rat.
After these baseline measurements were made, an osmotic minipump(Alzet model 2ML1 or 2ML2; Durect Corp., Cupertino, CA) thatcontained rhRLX (4 µg/h) or vehicle was implanted subcutaneouslyin the hindquarters, under isofluorane anesthesia in each rat.Renal function in each pair of rats again was measured at 1,3, 5, and 10 d during long-term infusion of rhRLX/vehicle usingthe aforementioned methods (Figure 1). After the last experiment,1 ml of blood was collected for measurement of rhRLX and bloodchemistries.
Figure 1. Longitudinal basic experimental protocol on a pair of age-matched chronically instrumented Munich Wistar rats for studying the effects of recombinant human relaxin (rhRLX) versus vehicle infusion.
Influence of MMP on Renal Function in Rats Treated Chronically with rhRLX/Vehicle.
Renal function and MAP were assessed in six pairs of age-matchedconscious rats (12 mo) on days 2 through 5 and day 20 of rhRLXinfusion using the aforementioned procedure. Infusion of RLXor vehicle was accomplished in this group of rats by replacingthe 14-d Alzet minipump on day 13 with a primed 7-d pump toensure continuous infusion for the full 20 d. After rhRLX/vehicle-influencedrenal function was assessed, continuous infusion of the specificgelatinase inhibitor cyclic CTT at 1.0 µg/min or its controlpeptide STTHWGFTLS (STT) at 1.0 µg/min was administeredvia the carotid arterial catheter (5). Each pair of rats wasrandomly assigned to receive either cyclic CTT or STT on day2 of rhRLX or vehicle administration. If they received cyclicCTT on day 2, then they were administered STT on day 5, andvice versa (Figure 2). Six 1-h renal clearances were obtainedduring the drug infusion as described. The general MMP inhibitorGM6001 (Ilomastat; rhRLX-treated rats, n = 4; vehicle-treatedrats, n = 4) or its control, dilute DMSO (rhRLX, n = 4; vehicle,n = 4), was administered at 30 ng/min (5).
Figure 2. Longitudinal experimental protocol on a pair of age-matched long-term instrumented Munich Wistar rats involving infusion of either cyclic CTTHWGFTLC (cCTT) or STTHWGFTLS (STT) in rhRLX/vehicle-infused rats.
MMP-2 Expression in Small Renal Arteries Isolated from rhRLX- and Vehicle-Treated Rats.
Kidneys were removed and processed using a validated techniquethat was described previously (5). Briefly, for gelatin zymography,the homogenates were prediluted in homogenizing buffer to equalizeprotein concentration with standard, combined with Novex Tris-GlycineSDS Sample Buffer (Invitrogen Co., Carlsbad, CA) and electrophoresedon Novex precast 10% Tris-Glycine gels that contained 0.1% gelatin.After staining/destaining, the gels were scanned using a Hewlett-PackardScan Jet 5370C scanner and HP PrecisionScan Pro v1.4 computerprogram (Palo Alto, CA). The images were digitized by UN-SCAN-ITgel automated digitizing system v4.3 (Silk Scientific Corp.,Orem, UT) (5).
Synthesis and Characterization of STT and CTT Peptides.
Peptides were synthesized by a solid-phase Pioneer automatedpeptide synthesizer (PE-Biosystem, Inc., Framingham, MA) usingthe FMOC synthesis protocol (5).
Analytical Techniques
Plasma osmolality was measured with freezing-point depressionosmometer (Advanced Instruments, Needham Heights, MA). Plasmaand urine IN and PAH were assayed by standard techniques (3).The levels of rhRLX in serum were measured by a validated ELISAimmunoassay (12). Microalbuminuria was assayed by PyrogallolRed Reagent (Sigma-Aldrich, St. Louis, MO). The Vitros DT-60photometer was used to measure blood chemistries (Johnson/JohnsonDiagnostics, Rochester, NY).
Histologic evaluation was performed for ultrastructural alterationof tissue. All histologic sections were reviewed by light microscopy.The degree of collagen deposition in the glomerulus and theinterstitium was graded as follows: 0, no collagen deposition;1+, mild; 2+ = moderate; 3+, moderately severe; and 4+, severe.The renal tissue was stained with two stains: Hematoxylin andeosin and Masson’s trichrome stain. The hematoxylin andeosin stain is used to look at general morphology of the glomerulus.The more specialized Masson’s trichrome stain dyes collagena deep blue for easier quantification. All stains were evaluatedin a double- blinded assessment.
Preparation of Drugs.
PAH and IN were freshly prepared on the morning of the experiment.The rhRLX (provided by Elaine Unemori, BAS Medical, San Mateo,CA), provided as a 1.4-mg/ml solution in 20 mM sodium acetate(pH 5.0), was diluted in the same buffer. GM6001 (Chemicon International,Temecula, CA) was prepared as a 10-mg/ml stock solution in sterile,endotoxin-free DMSO and diluted in Ringer’s solution,yielding a final DMSO concentration of 0.025% for infusion.Endotoxin-free, sterile DMSO was diluted in Ringer’s solutionas the vehicle control. Cyclic CTT and STT were prepared as1.0-mM stock solutions. Stock solutions were aliquotted andstored at –20°C.
Statistical Analyses
Data are presented as mean ± SEM. The data for the three30-min baseline renal clearances were averaged. Most of thedata were analyzed using two-factor repeated-measures ANOVA.When significant main effects or interactions were observed,group means were compared by Tukey test. Blood chemistries werecompared by unpaired t test. P < 0.05 was considered statisticallysignificant.
Three age groups—3, 6, and 12 mo—were selected tostudy the time course of the age-related changes and the effectsof RLX administration. The 3-mo-old rats showed no significantdecrease in renal function, abnormalities in blood chemistries,or significant renal pathology. Therefore, this age group wasnot investigated further except as comparison with older agegroups.
Rats gained weight as they aged: Body weights for 3-, 6-, and12-mo-old groups were 371 ± 5, 512 ± 8, and 591± 9 g, respectively. Therefore, renal function was normalizedto 100 g body wt for comparison among the three age groups.Baseline GFR was significantly higher in the 3-mo-old groupcompared with the 6- and 12-mo-old groups (Figure 3). After24 h of rhRLX infusion, there was a significant increase inGFR in all age groups over baseline. Days 3 and 5 renal functionwas not significantly different from day 10 values (one-factorrepeated measures ANOVA); therefore, all results were averaged.After 10 d of infusion, the 6- and 12-mo-old groups continuedto exhibit a lower GFR than the 3-mo-old group, but the oldestrats improved, showing the largest increase over baseline (350µl/min).
Figure 3. Effect of rhRLX infusion on GFR (µl/min per 100 g body wt) at baseline, 24 h of infusion, and 10 d of infusion in 3-, 6-, and 12-mo-old rats (n = 6 at each time point). *P 0.001, 3-mo-old baseline versus 6- and 12-mo-old baseline; **P 0.001 versus baseline in all age groups; #P 0.001, 12-mo-old 10-d versus 1-d rhRLX infusion.
In Figure 4, effective renal plasma flow showed that the youngestrats had a significantly higher ERPF than the two older agegroups. After 24 h of rhRLX infusion, ERPF increased significantlyin all age groups compared with baseline. However, the greatestincrease was observed in the 6- and 12-mo-old rats. After the10-d infusion, the 12-mo-old group showed a further sizableincrease in ERPF, thereby eliminating any difference in ERPFamong the age groups. The oldest rats showed the most robustresponse to rhRLX infusion (1260 µl/min). In all threegroups, there was a significant decrease in effective renalvascular resistance (ERVR) after 10 d of infusion of rhRLX (Figure 5).Here again, the oldest age group showed the largest decreasein ERVR. Baseline MAP (105 ± 3, 109 ± 4, and 112± 3 mmHg, respectively) did not change significantlyduring the course of rhRLX infusion in the three age groupseven after 10 d of infusion (103 ± 3, 112 ± 4,and 111 ± 3, respectively).
Figure 4. Effect of rhRLX infusion on effective renal plasma flow (ERPF; µl/min per 100 g body wt) at baseline, 24 h of infusion, and 10 d of infusion in 3-, 6-, and 12-mo-old rats (n = 6 at each time point). *P 0.001 3-mo-old baseline versus 6- and 12-mo-old baseline; **P 0.001 versus baseline in all age groups; #P 0.001 12-mo-old 10-d versus 1d rhRLX infusion.
Figure 5. Effect of rhRLX infusion (4 µg/min) for 10 d on effective renal vascular resistance (ERVR; mean arterial pressure [MAP]/ERPF per 100 g body wt) compared with vehicle infusion (R, rhRLX; V, vehicle). Each bar represents six animals in each age group (*P 0.001). Twelve-mo-old rats showed a significant decrease in ERVR compared with 3- and 6-mo-old rhRLX-infused groups (#P 0.001).
On histology, there was no statistical difference in collagendeposition in the glomerulus or the tubular interstitium ofthe 6-mo-old rats (data not shown). However, in the 12-mo-oldrats, there is considerable alteration in collagen depositionin both areas of the kidney. Using Fisher exact test, the 12-mo-oldvehicle-treated rats had significantly more glomerular and tubulointerstitialcollagen formation (P 0.001) than the rhRLX rats after a 10-dinfusion (Figure 6). Vehicle-infused rats had 1+ to 3+ abnormalcollagen depositions in the glomerulus, tubules, and interstitiumwith significant tubular distension and cast formation (Table 1).This finding was virtually absent in the rhRLX group.
Figure 6. (A) Normal glomerulus with Masson’s trichrome staining after 20 d of infusion of rhRLX. (B) Normal tubular back-to-back staining with Masson’s trichrome after 20 d of infusion of rhRLX. (C) Sclerotic, disorganized glomerulus with Masson’s trichrome staining after 20 d of infusion of vehicle. (D) Distended tubules with cast material shown with Masson’s trichrome after 20 d of infusion of vehicle. Magnification, x400.
Table 1. Effect of rhRLX versus vehicle infusion on collagen deposition in the glomerulus, tubules, and interstitium after 10 d treatment in the conscious rata
After 20 d of infusion, plasma levels of rhRLX were not significantlydifferent among the three age groups: 22.0 ± 2.5, 20.4± 1.6, and 27.7 ± 4.7 ng/ml in the 3-, 6-, and12-mo-old groups, respectively. Plasma chemistries were assayedon baseline samples and after 10 d of treatment with rhRLX orvehicle. Infusion of RLX significantly decreased plasma osmolality,sodium, and urea in both as compared with vehicle (Table 2).Cholesterol was increased significantly in the oldest age groupof RLX-infused rats compared with vehicle. Proteinuria, an indicationof a leaky and damaged glomerulus, was minimal in the 3-mo-oldgroup (5 ± 2 mg/24 h) but was significantly increasedin the 6- and 12-mo-old rats. The infusion of rhRLX in the 12-mo-oldrats lowered proteinuria, corresponding with the histologicfinding of decreased collagen deposition and improved glomerularfunction.
Table 2. Plasma and urine chemistry comparisons in conscious 6- and 12-mo-old rats after 10-d infusion of rhRLX or vehicle
Administration of the gelatinase inhibitor cyclic CTT for 6h reduced both GFR and ERPF and increased ERVR in the RLX-treatedbut not vehicle-infused rats in the 12-mo-old rats after 24to 48 h of rhRLX infusion (Figure 7). There was complete abrogationof renal vasodilation and hyperfiltration after the first hourof cyclic CTT infusion in RLX-treated rats. Because there wasno significant difference in the six 1-h renal clearances thatwere obtained during cyclic CTT infusion, they were combinedfor presentation (Figure 7). Administration of the control peptideSTT did not significantly affect GFR, ERPF, or ERVR in RLX-or vehicle-treated rats.
Figure 7. (A) Effect of gelatinase peptide inhibitor cyclic (c) CTT and the control peptide STT on GFR in conscious rats that received rhRLX or vehicle for rhRLX (20 mM sodium acetate [pH 5.0]) after the first 24 h. **P 0.005 versus all other rhRLX-infused groups; *P 0.001 versus vehicle. (B) Effect of gelatinase peptide inhibitor cCTT and the control peptide STT on ERPF in conscious rats that received rhRLX or vehicle for rhRLX (20 mM sodium acetate [pH 5.0]) after the first 24 h. **P 0.005 versus all other groups; *P 0.001 versus vehicle. (C) Effect of gelatinase peptide inhibitor cCTT and the control peptide STT on ERVR in conscious rats that received rhRLX or vehicle for rhRLX (20 mM sodium acetate [pH 5.0]) after the first 24 h. **P 0.005 versus all other groups; *P 0.001 versus vehicle. (D) Effect of gelatinase peptide inhibitor cCTT and the control peptide STT on MAP in conscious rats that received rhRLX or vehicle for rhRLX (20 mM sodium acetate [pH 5.0]) after the first 24 h. *P < 0.001 versus cCTT versus baseline.
To corroborate the results that were obtained with the gelatinaseinhibitor cyclic CTT, we also tested the nonspecific MMP inhibitorGM6001, a compound that is structurally dissimilar from cyclicCTT (5). Once again, at baseline, GFR and ERPF were significantlyhigher and ERVR was lower in the 2- to 5-d RLX-treated ratscompared with vehicle-infused rats. Administration of GM6001significantly reduced GFR and ERPF and increased ERVR in theRLX-treated but not the vehicle-infused rats (data not shown).Comparable to cyclic CTT, GM6001 completely reversed RLX-inducedincreases in GFR and ERPF, as well as decreases in ERVR, althoughthe time course of reversal was not as rapid. Administrationof the vehicle control, 0.025% DMSO, did not affect significantlyrenal parameters in RLX- or vehicle-treated rats. MAP was notaltered significantly by GM6001 or its vehicle in the two groupsof rats.
On further investigation of rhRLX infusion after 20 d in 12-mo-oldrats, it was surprising that administration of cyclic CTT didnot abolish the RLX-induced renal hyperfiltration or vasodilation(Figure 8). GFR and ERPF remained increased above baseline levelsaccompanied by decreased renal vascular resistance. MAP didnot change significantly. These results suggest that eithera higher dose of CTT may be needed in these animals or thereis an underlying permanent restructuring or change in environmentof the cortical matrix that allowed the vessels to remain dilatedin the face of acute MMP inhibition. This finding was confirmedby the use of the inhibitor GM6001 (Figure 9). Measurement ofserum rhRLX in these rats was comparable to previous reportsin which the same infusion rates were used (26.9 ± 1.7ng/ml) (12). No immunoreactivity was detected in vehicle-treatedcontrol rats.
Figure 8. (A) Effect of gelatinase peptide inhibitor cCTT and the control peptide STT on GFR in conscious rats that received 20 d of infusion of rhRLX or vehicle for rhRLX (20 mM sodium acetate [pH 5.0]). *P 0.001 versus vehicle. (B) Effect of gelatinase peptide inhibitor cCTT and the control peptide STT on ERPF in conscious rats that received 20 d of infusion of rhRLX or vehicle for rhRLX (20 mM sodium acetate [pH 5.0]). *P 0.001 versus vehicle. (C) Effect of gelatinase peptide inhibitor cCTT and the control peptide STT on ERVR in conscious rats that received 20 d of infusion of rhRLX or vehicle for rhRLX (20 mM sodium acetate [pH 5.0]). *P 0.001 versus vehicle. (D) Effect of gelatinase peptide inhibitor cCTT and the control peptide STT on MAP in conscious rats that received 20 d of infusion of rhRLX or vehicle for rhRLX (20 mM sodium acetate [pH 5.0]). *P < 0.001 versus cCTT versus baseline.
Figure 9. Effect of gelatinase peptide inhibitor GM6001 and control peptide DMSO on ERVR in conscious rats after 20 d of infusion of rhRLX or vehicle. *P 0.001 versus vehicle.
Figure 10 depicts a representative gelatin zymogram that isindicative of increased expression of pro- and active MMP-2in the interlobar arteries that were dissected from 12-mo-oldrats that were infused for 48 to 72 h with rhRLX or vehicle.Densitometric analysis revealed that in five of the six pairsof rats, one third more MMP-2 activity was detected in RLX-treatedrats.
Figure 10. Representative gelatin zymogram of gelatinase activities in interlobar arteries from rhRLX-infused and vehicle-infused rats (10 d). STD, recombinant human matrix metalloproteinase-2 (MMP-2) active MMP-2 and pro-MMP-2 standards. *P 0.05 rhRLX versus vehicle-infused rats by densitometry.
It is estimated that by the year 2050, 80 million people, onein every five Americans, will be more than 65 yr old. Half ofthe cases of chronic renal disease involving fibrosis occurin this age group. In humans, advancing age often is accompaniedby decreased renal function. Anatomic changes include decreasein renal mass and size that occurs mainly in the cortex withloss of functioning of cortical glomeruli, narrowing of therenal arteries as a result of collagen formation, formationof afferent-efferent arteriovenous fistula as a result of thisloss of glomeruli, and an increase in fibrosis in the cortexand renal pyramids (13). This fibrosis causes an increase inabnormal collagen deposition in the glomeruli and interstitiumECM, decreasing GFR and disrupting normal blood flow. The mechanismsunderlying age-associated glomerular atrophy and tubulointerstitialfibrosis remain uncertain (14).
The aging of the human kidney is paralleled in the Munich Wistaralbino rat (14,15), which demonstrates chronic renal nephrosisas described by Gray et al. (15), characterized by proteinuria,enlarged and sclerotic glomeruli with thickened basement membranes,and abnormal accumulation of collagen (15). Our study was designedto document the impact of rhRLX infusion on renal function andhistology in this animal model. We hypothesized that both short-and long-term administration of RLX would improve renal functionin the older rat. Furthermore, after long-term administration,RLX would reduce renal collagen deposition and improve renalhistology.
Our study showed that 24 to 72 h of rhRLX infusion caused hyperfiltrationand increased renal blood flow in all three age groups of rats,the largest increase occurring in the oldest age group. Thisacute improvement in renal function was reversed by an inhibitorof gelatinase, cyclic CTT. Isolation of the small renal arteriesfrom 10-d rhRLX-infused rats showed a significant increase inpro- and active MMP-2 activity. Examination of Masson’strichrome stain revealed a significant decrease in collagendeposition in the glomerulus, tubules, and interstitium in the12-mo-old rats after 10 d of rhRLX infusion. After 20 d of infusionof rhRLX in the 12-mo-old rats, cyclic CTT (1.0 µg/mininfusion) did not reverse the increase in GFR and ERPF. Theresults support our hypothesis that RLX works in a biphasicmanner, causing an initial acute vasodilation and improvementin renal function with an upregulation of gelatinase activityin the renal vessels followed by a permanent structural alteration.
Several studies have demonstrated that RLX reduces fibrosisin experimental animal models (6–8). RLX-deficient knockoutmice also demonstrate age-related progression of pulmonary,cardiac, and renal fibrosis (11). Our study suggests that inthe aging rat kidney, RLX has the ability to increase renalfunction quickly (24 to 48 h) and reduce renal fibrosis significantlyafter 10 d of infusion.
The delicate balance between synthesis and degradation of ECMin the kidney is disrupted by the aging process. Reduced bloodsupply, lack of oxygen to the kidney as a result of increasedECM production versus degradation, and reduced filtering capacityof sclerotic glomeruli can stimulate the influx of inflammatorycells, leading to the release of cytokines and growth factors,including TGF- (16). TGF- has been shown to differentiate renalfibroblasts into activated myofibroblasts that express the marker-smooth muscle actin. These activated cells increase matrixprotein gene expression, leading to increased matrix productionand resulting in tissue fibrosis (17).
Tubulointerstitial fibrosis is a widely recognized common pathwayof all progressive renal disease regardless of the cause. Interstitialfibroblasts are important effector cells in renal fibrosis (18).RLX plays an important role in the downregulation of these fibroblasts,in vitro, resulting in decreased production of TGF-. Mastersonet al. (19) recently showed that myofibroblast differentiationin vitro can be downregulated by RLX. Within 24 h, 25% of fibroblastsno longer exhibited -smooth muscle actin, suggesting an RLX-mediatedreversal of myofibroblast differentiation. This reversal causeda decrease in matrix production, increased collagenase synthesis,and an inhibition of collagen-I lattice contraction. RLX wasfound previously to abrogate the effects of TGF- (6,9), recognizedas a pivotal cause of glomerulosclerosis and tubulointerstitialfibrosis in renal disease by stimulation of matrix protein production(20) and increasing the contraction of surrounding ECM (21).
In our study, the oldest rats that showed the greatest decrementsin renal function were the most affected by RLX infusion. Thatis, they demonstrated the greatest increase in GFR and ERPFafter 24 to 48 h of rhRLX infusion. At least four possible pathwayscould explain these findings: (1) Increased action of the gelatinasesaffecting ECM and the vascular smooth muscle, (2) downregulationof the activating myofibroblasts in the first 24 h, (3) reducedTGF- in the ECM from downregulation of myofibroblasts, or (4)increased production of NO (with an upregulation of NO synthase).The first three mechanisms would seem unlikely for the explanationof the quick 24- to 48-h reversal that we have documented.
Our study explored the possibility of vascular gelatinase upregulation.In the first 24 to 48 h, inhibition of gelatinase by cyclicCTT or GM6001 caused a complete reversal of renal hyperfiltration,vasodilation, and decreased vascular resistance in the 6- and12-mo-old rhRLX-infused rats, implicating that an upregulationof gelatinase expression as an important mediator. Indeed, pro-and active forms of MMP were increased in isolated small renalarteries by gelatin zymography.
Because we also were interested in the long-term effects ofRLX infusion on renal histology and function, we continued theinfusion for 10 d. The 12-mo-old rats showed a further increasein renal function during this 10-d infusion period. This maybe explained by the significant decrease in renal collagen causedby the antifibrotic actions of RLX documented by histologicstaining in this group as compared with the younger rats. Becausethe ECM was altered, the vessels may be free to dilate to theirfull extent. A second possibility is decreasing sclerosis ofresistance arteries in glomeruli, leading to increased vasodilationand hyperfiltration. The histology showed fewer inflammatorycells and less collagen deposition in the glomeruli and tubulointerstitialspace as well as less tubular dilation containing cast material.
We further tested rhRLX infusion for 20 d in the 12-mo-old rats.To our surprise, the acute inhibition of gelatinase activitywith cyclic CTT (1.0 µg/min) did not affect the increaseof renal function that was induced by rhRLX. Because we verifiedthe results with GM6001, these data suggest either that a greaterconcentration of MMP inhibitors is needed to reverse the RLX-inducedvasodilation or that another mechanism of rhRLX action is responsiblefor the improved renal function after 20 d of infusion. Thisfinding possibly may be explained by downregulation of myofibroblasts,changing the cytokine milieu, or an increased production ofNO.
RLX infusion increases GFR and renal plasma flow and decreasesrenal vascular resistance in the aged Munich Wistar albino rat.This increase in the first 24 to 72 h is mediated by the actionof vascular gelatinase activity. After 20 d of infusion of RLX,inhibition of this gelatinase did not abolish the increasedrenal function that was induced by RLX. This suggests a permanentchange in the structure of the interstitium surrounding theglomerulus and tubules, possibly caused by the antifibroticeffect of rhRLX infusion. Further study is needed to examinethe mechanistic details of this change. Because RLX improvesrenal function and ameliorates the pathologic nephrosis/fibrosisin this rat model, there is a possibility that this remodelingagent could be used to reverse the fibrotic condition of thehuman kidney in our older population as well as in other diseasestates with fibrotic injury, such as diabetes and polycystickidney disease.
Acknowledgments
This project was supported in part by the Dedicated Health ResearchFunds of the University of New Mexico School of Medicine RACgrant C-2246-RAC.
We are grateful to Elaine Unemori, PhD (BAS Medical, San Mateo,CA), and BAS Medical for providing the rhRLX and the antibodiesfor the rhRLX ELISA. We extend our sincere thanks to Ketah Doty,in Dr. Kirk Conrad's laboratory at Magee Women’s ResearchInstitute, for work on the MMP-2 zymograms. Ms. Doty was supportedby National Institutes of Health grant R01 DK63321. We acknowledgeAngela Wandinger-Ness (Department of Pathology, University ofNew Mexico) for critical reading of the manuscript and providingsupport and processing and sectioning of kidney performed byAngela Welford in her laboratory.
Footnotes
Published online ahead of print. Publication date availableat www.jasn.org.
Sherwood DA: Relaxin’s physiological roles and other diverse actions.
Endocr Rev 25
: 205
–234, 2004[Abstract/Free Full Text]
Conrad KP: Mechanisms of renal vasodilation and hyperfiltration during pregnancy.
J Soc Gynecol Investig 11
: 438
–448, 2004[Abstract/Free Full Text]
Danielson LA, Sherwood OD, Conrad KP: Relaxin is a potent renal vasodilator in conscious rats.
J Clin Invest 103
: 525
–533, 1999[Medline]
Danielson LA, Kerchner LF, Conrad KP: Impact of gender and endothelin on renal vasodilation and hyperfiltration induced by relaxin in conscious rats.
Am J Physiol 279
: R1298
–R1304, 2000
Jeyabalan A, Novak J, Danielson LA, Kerchner LJ, Opett SL, Conrad KP: Essential role for vascular matrix metalloproteinases-2 in relaxin-induced renal vasodilation, hyperfiltration and reduced myogenic reactivity of small arteries.
Circ Res 93
: 1249
–1257, 2003[Abstract/Free Full Text]
Unemori EN, Amento EP: Relaxin modulates synthesis and secretion of procollagenase and collagen by human dermal fibroblasts.
J Biol Chem 265
: 10681
–10685, 1990[Abstract/Free Full Text]
Unemori EN, Pickford LB, Salles AL, Piercy CE, Grove BH, Erikson ME, Amento EP: Relaxin induces an extracellular matrix-degrading phenotype in human lung fibroblasts in vitro and inhibits lung fibrosis in a murine model in vivo.
J Clin Invest 98
: 2739
–2745, 1996[Medline]
Palejwala S, Weiss G, Goldsmith LT: Relaxin stimulates expression of matrix metalloproteinase 2 by human lower uterine segment fibroblasts.
J Soc Gynecol Invest 7
: 124A
, 2000
Garber SL, Mirochnik Y, Brecklin C, Unemori EN, Singh AK, Grove BH, Slobodskoy L, Arruda JAL, Dunea G: Relaxin decreases renal interstitial fibrosis and slow progression of renal disease.
Kidney Int 59
: 876
–882, 2001[CrossRef][Medline]
Garber SL, Mirochnik Y, Brecklin C, Slobodskoy L, Arruda JAL, Dunea G: Effect of Relaxin in two models of renal mass reduction.
Am J Nephrol 23
: 8
–12, 2002
Samuel CS, Zhao C, Bathgate R, Du X-J, Summers R, Amento EP, Walker LL, McBurnie M, Zhao L, Tregear GW: The relaxin gene-knockout mouse: A model of progressive fibrosis.
Ann N Y Acad Sci 1041
: 173
–181, 2005[CrossRef][Medline]
Danielson LA, Conrad KP: Time course and dose response of relaxin-mediated renal vasodilation, hyperfiltration, and changes in plasma osmolality in conscious rats.
J Appl Physiol 95
: 1509
–1514, 2003[Abstract/Free Full Text]
Baylis C, Corman B: The aging kidney: Insights from experimental studies.
J Am Soc Nephrol 9
: 699
–709, 1998[Abstract]
Clark B: Biology of renal aging in humans.
Adv Ren Replace Ther 7
: 11
–21, 2000[Medline]
Gray J: Chronic progressive nephrosis in the albino rats.
CRC Crit Rev Toxicol 5
: 115
–143, 1977[Medline]
Negri AL: Prevention of progressive fibrosis in chronic renal diseases: Antifibrotic agents.
J Nephrol 17
: 496
–503, 2004[Medline]
Sappino A, Schurch W, Gabbiani G: Differentiation of fibroblastic cells: Expression of cytoskeletal proteins as marker of phenotypic modulations.
Lab Invest 62
: 144
–161, 1990
Becker G, Hewitson T: The role of tubulointerstitial injury in chronic renal failure.
Curr Opin Nephrol Hypertens 9
: 133
–138, 2000[CrossRef][Medline]
Masterson R, Hewitson TD, Kelynack K, Martic M, Parry L, Bathgate R, Darby I, Becker G: Relaxin down-regulates renal fibroblast function and promotes matrix remodeling in vitro.
Nephrol Dial Transplant 19
: 544
–552, 2004[Abstract/Free Full Text]
Border W, Noble N: Transforming growth factor beta in tissue fibrosis.
N Engl J Med 331
: 1286
–1292, 1994[Free Full Text]
Kelynack K, Hewitson T, Pedagogos E, Nicholls K, Becker G: Renal myofibroblasts contract collagen I matrix lattices in vitro.
Am J Nephrol 19
: 694
–701, 1999[CrossRef][Medline]
This article has been cited by other articles:
P. Boor, K. Sebekova, T. Ostendorf, and J. Floege Treatment targets in renal fibrosis
Nephrol. Dial. Transplant.,
December 1, 2007;
22(12):
3391 - 3407.
[Full Text][PDF]
J. M Sasser, M. D. Lindheimer, and C. Baylis An Emerging Role for Relaxin as a Renal Vasodilator
J. Am. Soc. Nephrol.,
November 1, 2006;
17(11):
2960 - 2961.
[Full Text][PDF]
Top
Abstract
Introduction
350 µl/min). 
0.001) than the rhRLX rats after a 10-d infusion (Figure 6). Vehicle-infused rats had 1+ to 3+ abnormal collagen depositions in the glomerulus, tubules, and interstitium with significant tubular distension and cast formation (Table 1). This finding was virtually absent in the rhRLX group. 
(16). TGF-
-smooth muscle actin. These activated cells increase matrix protein gene expression, leading to increased matrix production and resulting in tissue fibrosis (17). 
