Kidney International, Vol. 57 (2000), pp. 1493–1501

Angiotensin II blockade and low-protein diet produce additivetherapeutic effects in experimental glomerulonephritis

HARM PETERS, WAYNE A. BORDER, and NANCY A. NOBLE

Fibrosis Research Laboratory, University of Utah School of Medicine, Salt Lake City, Utah, USA

progression in humans with chronic renal failure may be slowedAngiotensin II blockade and low-protein diet produce additivemore effectively when Ang II blockade and low-protein diettherapeutic effects in experimental glomerulonephritis.are combined. Since maximal pharmacological Ang II inhibi-Background. Transforming growth factor-b (TGF-b) over-tion was used, it is likely that dietary protein restriction furtherexpression plays a key role in the accumulation of extracellularreduces pathological TGF-b overexpression by mechanismsmatrix in acute and chronic renal diseases. Recent studies havedifferent from those of enalapril or losartan.suggested that the degree of reduction in pathological TGF-b

overexpression can be used as a therapeutic index to evaluatethe antifibrotic potential of pharmacological angiotensin II(Ang II) blockade in renal disease. Using this target, we found

Despite major differences in the underlying disorder,that treatment with the angiotensin I-converting enzyme inhibi-the histologic picture of chronic renal disease is uni-tor enalapril or the Ang II type 1 receptor antagonist losartan

reduced TGF-b overexpression more effectively at doses clearly formly characterized by progressive accumulation of ex-higher than those required to control blood pressure. However, tracellular matrix proteins that obliterates renal functionboth forms of Ang II blockade were only partially effective in and leads to organ failure [1]. A large body of work nownormalizing TGF-b expression. This study investigated whether

supports the hypothesis that overexpression of the cyto-a greater antifibrotic, TGF-b–reducing benefit can be achievedkine transforming growth factor-b (TGF-b) plays a keywhen Ang II blockade is combined with dietary protein re-

striction. role in the progressive matrix accumulation and tissueMethods. Mesangioproliferative glomerulonephritis was in- fibrosis in all organs in which fibrosis occurs, including

duced in male Sprague-Dawley rats on a normal-protein diet. the kidney [1, 2]. A number of factors known to be injuri-Treatment with a low-protein diet and/or maximally effectiveous to the kidney at the cellular or molecular level havedoses of enalapril or losartan was started one day after diseasenow been shown to increase TGF-b expression [re-induction. On the fifth day, 24-hour urine protein excretion

was measured. On the sixth day, cortical kidney tissue was viewed in 2]. These include high glucose levels, hypoxia,taken for periodic acid-Schiff staining. Isolated glomeruli were immune complexes, advanced glucosylated end prod-used for mRNA extraction or were placed in culture for deter-

ucts, mesangial cell stretch, protein trafficking, platelet-mination of production of TGF-b1, the matrix protein fibro-derived growth factor, basic fibroblast growth factor, andnectin, and the protease inhibitor plasmin activator inhibitor

type 1 (PAI-1) by enzyme-linked immunosorbent assay. TGF-b itself. TGF-b overproduction in turn mediatesResults. Compared with untreated nephritic animals on a local expansion of extracellular matrix through increased

normal-protein diet, a single treatment with enalapril, losartan, matrix production, decreased matrix degradation, andor low-protein diet significantly reduced glomerular TGF-b

increased local expression of matrix-binding receptors.production, albeit to a similar degree of approximately 45%.Blocking TGF-b overexpression with neutralizing anti-A moderate, but significant further reduction in pathological

TGF-b expression of a total of 65% for enalapril and 60% for bodies has now been shown to reduce matrix expansionlosartan was achieved when these drugs were combined with in a number of models of fibrotic diseases in tissues,low-protein feeding. This reduction in TGF-b overexpression including kidney, artery, joint, lung, liver, and the centralparalleled decreased proteinuria, glomerular matrix accumula-

nervous system [1, 3, 4].tion, and overproduction of fibronectin and PAI-1.Blockade of angiotensin II (Ang II) with angiotensin-1–Conclusions. Ang II blockade and low-protein diet have

additive effects on disease reduction, suggesting that disease converting enzyme (ACE) inhibitors or the newly avail-able class of Ang II type 1 (AT1) receptor antagonistssignificantly limits renal matrix expansion in numerousKey words: TGF-b, angiotensin, low-protein diet, kidney, extracellular

matrix, progressive renal disease. experimental acute and chronic renal diseases [2, 5]. Inpatients with chronic renal failure, ACE inhibition re-Received for publication November 4, 1998duces proteinuria, slows the fall in the glomerular filtra-and in revised form October 18, 1999

Accepted for publication November 1, 1999 tion rate, and delays the need for dialysis or transplanta-tion. The octapeptide Ang II is the principal effector 2000 by the International Society of Nephrology

1493

Peters et al: Additive therapeutic effects in GN1494

molecule of the renin-angiotensin system (RAS) and has that additional pathways must be targeted. Supportingthis conclusion are recent data from Fern et al, wholong been known as a potent vasoconstrictor. Besides

increasing hypertensive damage to the kidney, several showed that fibrosis is reduced by only about 50% inmice with obstructive nephropathy in which the angio-recent studies have shown that Ang II also has important

blood pressure-independent actions leading to matrix ac- tensinogen gene has been disrupted [14].A prime candidate to further reduce renal TGF-b over-cumulation [2]. Direct pressure-independent induction of

TGF-b overexpression has emerged as a central feature expression and tissue fibrosis in the presence of Ang IIblockade is restriction of dietary protein intake. Theof this work [2, 6, 7]. In cultured mesangial cells, the

induction of matrix proteins such as collagens, fibronec- protective effects of lowering protein intake were firstdemonstrated in a model of nephrotoxic serum nephritistin, and proteoglycans by Ang II is preceded by induction

of TGF-b expression and can be blocked by a TGF-b– in the late 1930s [15]. Subsequently, low protein intakehas been shown to improve disease course and limitneutralizing antibody [8]. In further studies, we have

shown that the sustained effect of Ang II on synthesis matrix expansion in virtually all renal diseases [16]. Incontrast, increased protein intake has been shown toof the key protease inhibitor, plasmin activator inhibitor

type 1 (PAI-1) is also mediated via induction of TGF-b, worsen experimental diseases [17]. Despite continuedcontroversy and questions about the mechanism of ac-while the initial short-term effect appeared to be induced

by Ang II directly [9]. Comparable work on pressure- tion, as recently reviewed by Heidland, Sebekova, andLing [18], several clinical trials have suggested that pro-independent Ang II effects has been done using proximal

tubular cells [7, 10]. In addition, Ang II induces several tein restriction effectively slows the progressive loss ofrenal function in patients with chronic renal failure. Stud-chemokines that lead to infiltration of monocytes/macro-

phages [7, 11], effects that may be mediated by TGF-b. ies have shown that low-protein feeding significantly re-duces TGF-b overexpression following the induction ofExciting preliminary data further indicate that Ang II

enhances TGF-b’s fibrogenic activities by up-regulation ATS GN [19]. These results were later confirmed byothers in models of puromycin- and adriamycin-inducedof TGF-b receptor expression, thereby increasing the

actions of exogenous TGF-b to increase matrix protein GN [20, 21], and are consistent with the concept thatantifibrotic effects are mediated through a reduction insynthesis (abstracts; Kanai et al, J Am Soc Nephrol

8:518A,1997; J Am Soc Nephrol 9:498A,1998). In experi- TGF-b overexpression.In the present study, we tested the hypothesis that themental kidney diseases, in turn, the decrease in renal

matrix expansion by Ang II blockade is paralleled by a antifibrotic effect of maximally effective doses of Ang II-blocking drugs can be enhanced by simultaneously feed-reduction in TGF-b overexpression, suggesting that its

therapeutic effects are mediated through reduction of ing a low-protein diet. Again, using ATS GN in the rat,we show that treatment with enalapril or losartan or low-TGF-b [2]. Human data are beginning to emerge show-

ing that ACE inhibition lowers TGF-b levels in chronic protein diet alone reduces glomerular TGF-b productionapproximately 45%. A significant further reduction inhuman kidney diseases [12].

Our laboratory has recently been involved in testing glomerular TGF-b overexpression and all disease mea-sures is achieved when enalapril or losartan treatment isthe hypothesis that the efficacy of current or promising

antifibrotic drugs can be determined by their potential combined with dietary protein restriction. A major differ-ence between these and earlier studies with a low-proteinto reduce TGF-b overexpression. We have further asked

whether the addition of a second therapy to one drug diet is the attempt to quantitate therapeutic efficacy.given at doses that maximize TGF-b reduction alonecan enhance efficacy [13]. Using this approach, we have

METHODSrecently shown that treatment with either the ACE in-

Materialshibitor enalapril or the AT1 receptor blocker losartanproduces a dose-dependent reduction in glomerular If not otherwise indicated, materials, chemicals, or

culture media were purchased from Sigma Chemical Co.TGF-b overexpression and matrix accumulation in theantithymocyte serum (ATS) model of mesangioprolifer- (St. Louis, MO, USA).ative glomerulonephritis (GN) [13]. Although maximal

Experimental designdisease reduction was seen at doses that were clearlyhigher than those known to control blood pressure, In rats following the induction of acute GN, we tested

the hypothesis that the antifibrotic effect of Ang II-block-TGF-b overproduction and other measures of diseasecould not be reduced by more than 50%, even with ing drugs can be enhanced by additional low-protein

restriction. GN was induced in rats fed a normal-proteinextremely high doses. Furthermore, no additive effectswere seen when both forms of Ang II blockade were diet. Treatment was started 24 hours after antibody injec-

tion when mesangial cell lysis is complete and the fibroticgiven. The data strongly suggest that pharmacologicalblockade of Ang II alone will not halt renal fibrosis and response starts. Single therapy with enalapril, losartan,

Peters et al: Additive therapeutic effects in GN 1495

or a low-protein diet was compared side by side with that the doses used in these experiments maximally re-duced glomerular TGF-b overexpression and matrix ac-combined treatment with low-protein feeding and enala-

pril or losartan. Renal tissues were harvested six days cumulation in this model [13]. Sixty percent of the firstdaily dose of enalapril or losartan (based on 40 mL drink-after disease induction, at a time point when previous

time-course studies had shown that the fibrotic response ing volume per day) was administered by gavage 24 hoursafter OX-7 injection. Water and drug intake was moni-was maximal.

A reduction in TGF-b overexpression served as the tored daily. Diets included a normal-protein diet (22%casein) and a low-protein diet (6% casein, Teklad dietprinciple measure of antifibrotic action. A histologic ma-

trix score was used to determine whether the changes 86551; Teklad Premier Laboratory Diets). The dietswere identical in fat and mineral content and were madein TGF-b expression reflect the actual renal matrix accu-

mulation. Measurements of the matrix component fi- isocaloric (14.6 J/g) by the addition of sucrose.One day after the induction of GN, treatment wasbronectin and the protease inhibitor PAI-1 were used

as sensitive markers of TGF-b’s action. TGF-b1, fibro- started. Rats (8 per group) were assigned to the followingtherapies: (1) controls on normal protein, (2) GN onnectin, and PAI-1 expression were measured at the pro-

tein level in the supernatant of cultured glomeruli and normal protein, (3) GN on low protein, (4) GN on nor-mal protein plus enalapril, (5) GN on normal proteinat the mRNA level by Northern blotting.plus losartan, (6) GN on low protein plus enalapril, and

Animals (7) GN on low protein plus losartan.Male Sprague-Dawley (SD) rats (200 to 270 g) ob-

Measurement of blood pressure and proteinuriatained from Sasco (Omaha, NE, USA) were fed a normalprotein diet (22% protein, Teklad no. 86550; Teklad Systolic blood pressure was measured in conscious

animals five days after disease induction by the tail-cuffPremier Laboratory Diets, Madison, WI, USA) for atleast three days before the start of the experiment to method [13]. During the following 24 hours, rats were

individually housed in metabolic cages for urine collection.allow equilibration.Proteinuria was measured using the Bradford method and

Disease induction is expressed as mg protein/24 h.Glomerulonephritis was induced by tail-vein injection

Sacrificeof the monoclonal anti-Thy 1.1 antibody OX-7 (1.5 mg/kg)under light methoxyflurane anesthesia. In the kidney, Animals were anesthetized with ether six days after

the induction of GN. Following a midline abdominalthe monoclonal antibody binds to a Thy 1-like epitopeon the surface of mesangial cells and causes complement- incision, 5 to 10 mL of blood were drawn from the lower

abdominal aorta, and the kidneys were subsequently per-dependent cell lysis. Control animals were injected withequal volumes of phosphate buffered saline (PBS). Body fused with 30 mL ice-cold PBS. For histologic examina-

tion, cortical tissue was fixed in 10% neutral-bufferedweight, food, and water intake were monitored individu-ally at the beginning and end of each experiment. formalin. Glomeruli from individual rats were isolated

by a graded sieving technique (150, 125, 106, and 75 mmOX-7 production mesh metal sieves) as described previously [13].

OX-7 was produced as previously described [13]. TheLight microscopyhybridoma cells were obtained from American Type

Culture Collection (Rockville, MD, USA). Briefly, the All microscopic examinations were performed in ablinded fashion. Three micrometer sections of paraffin-hybridoma was grown in ascitic fluid of adult female

BALB/c mice (Charles River, Wilmington, MA, USA) embedded tissue were stained with periodic acid-Schiff(PAS). Glomerular matrix expansion was evaluated asand was purified using a Protein G column (Pierce, Rock-

ford, IL, USA). The eluant was dialyzed extensively previously described [13]. Briefly, in 30 glomeruli fromeach rat, the percentage of matrix occupying each glo-against PBS (pH 7.4). The antibody concentration was

adjusted to 1 mg/mL based on the absorbance at 280 nm merulus was rated as follows: 0 5 0%, 1 5 25%, 2 550%, 3 5 75%, and 4 5 100%.and was stored at 2708C until use.

Treatments Production of TGF-b1, fibronectin, and PAI-1 byglomeruli in cultureAngiotensin II blockade was performed as previously

described [13]. The ACE inhibitor enalapril (Enal, 100 Glomeruli from individual rats were resuspended inDulbecco’s modified Eagle’s medium supplemented withmg/L) and the AT1 antagonist losartan (Los, 500 mg/L)

were given in the drinking water. Losartan was kindly 0.1 U/mL insulin, 100 U/mL penicillin, 100 mg/mL strep-tomycin, and 25 mmol/L HEPES buffer. After 72 hoursprovided by DuPont Merck Pharmaceuticals Co. (Wil-

mington, DE, USA). In a previous study, we have shown of incubation at 378C/5% CO2, the supernatant was har-

Peters et al: Additive therapeutic effects in GN1496

vested and stored at 2708C until analysis of glomerularTGF-b1, fibronectin, and PAI-1 production. TGF-b1production of cultured glomeruli was measured after acidactivation using a commercially available enzyme-linkedimmunosorbent assay (ELISA) kit (Quantikinee; R&DSystems, Minneapolis, MN, USA) according to the man-ufacturer’s instructions. Fibronectin and PAI-1 synthesiswere measured with modified inhibitory ELISAs ac-cording to published methods [22]. Three samples fromeach rat were analyzed.

RNA preparation and Northern hybridization forTGF-b1, fibronectin, PAI-1, and GAPDH

Total RNA was extracted immediately from freshly iso-lated glomeruli by a guanidinium isothiocyanate methodusing Trizolt Reagent according to the manufacturer’s

Fig. 1. Twenty-four–hour urinary protein excretion measured in sam-instructions. Isolated glomeruli from two to four ratsples collected from days 5 and 6 of mesangioproliferative glomerulone-

were pooled for subsequent RNA extraction. For North- phritis (GN). The animals were fed either a normal-protein (NP, 22%casein) or a low-protein diet (LP, 6% casein). Enalapril (Enal, 100 mg/L)ern analysis, RNA was denatured and fractionated byor losartan (Los, 500 mg/L) was given in the drinking water (*P , 0.01electrophoresis through a 1.8% agarose gel (30 mg/lane) vs. NP; #P , 0.05 vs. LP, NP 1 Enal, or NP 1 Los).

and transferred to a Hybond nylon membrane (Amers-ham Corp., Little Chalfont, UK). Nucleic acids wereimmobilized by ultraviolet irradiation. Membranes were

animals. Treatment with Ang II blockade or low-proteinprehybridized with 50% formamide, 10% Denhardt’sdiet did not significantly change systolic blood pressure.solution, 0.1% sodium dodecyl sulfate (SDS), 5 3 stan-

dard saline citrate (SSC), and 200 mg/mL denaturedEffect of disease inductionsalmon sperm DNA and hybridized with DNA probes

As shown in Figure 1, control animals excreted 13 6labeled with 32P-dCTP by random oligonucleotide prim-2 mg of protein in the 24 hours before sacrifice. In theing (Boehringer Mannheim Biochemicals, Indianapolis,untreated nephritic rats, proteinuria increased to 41 6IN, USA). The blots were washed in 2 3 SSC, 0.1%5 mg/24 h. Injection of OX-7 resulted in an 8.6-foldSDS at room temperature for 15 minutes, and 0.1 3increase in TGF-b1 production, a 16.3-fold increase inSSC, 0.1% SDS at 558C for 30 minutes. DNA probes usedfibronectin production, and a 15.5-fold increase in glo-were (1) rat glyceraldehyde-3-phosphate dehydrogenasemerular PAI-1 production by glomeruli isolated at day(GAPDH) cDNA (a gift form Drs. P. Kondaiah and6 (Fig. 2). At the mRNA level, corrected for GAPDHM.B. Sporn) [23], (2) TGF-b1 cDNA (kindly providedexpression, TGF-b1 expression increased 3.6-fold, fi-from Dr. D.L. Moses) [24], (3) fibronectin-EDA cDNAbronectin expression 18.9-fold, and PAI-1 expression by(a generous gift from Dr. R.O. Hynes) [25], and (4)24.7-fold, respectively (Fig. 3). The glomerular matrixPAI-1 cDNA (kindly provided by Dr. T.D. Gelehrter)score in the PBS-injected controls was 1.12 6 0.04 and[26]. Three blots per probe were performed.increased to 2.56 6 0.04 after OX-7 injection (Fig. 2).

Statistical analysisEffect of single treatment with enalapril, losartan orData are expressed as means 6 SEM. Statistical analy-low proteinsis between the groups was performed by analysis of

variance (ANOVA) and subsequent t-testing with Bonf- Compared with the disease control animals, treatmenterroni correction for multiple comparison. A P value , of nephritic rats with either a low-protein diet, enalapril,0.05 was considered as significant. or losartan resulted in a statistically significant, but very

similar, reduction in all measures of disease. Proteinuriain the 24 hours before sacrifice was equally decreased

RESULTSin all three groups with single treatment. From untreated

Basic data disease group values of 41 6 5 mg, the mean was 19 65 mg protein per 24 hours in low-protein group, 20 6 3Body weight, food consumption, and drinking volume

were not significantly different between the groups. Sys- mg in the enalapril, and 17 6 3 mg in the losartan groups,respectively (Fig. 1). Compared with the disease controltolic blood pressure was 112 6 9 mm Hg in normal

control rats and 108 6 11 mm Hg in untreated nephritic animals (2.56 6 0.04), glomerular matrix score decreased

Peters et al: Additive therapeutic effects in GN 1497

Fig. 2. Glomerular matrix accumulation (A), glomerular production of transforming growth factor-b1 (TGF-b1; B), fibronectin (C), and plateletactivator inhibitor type 1 (PAI-1; D) six days after the induction of mesangioproliferative glomerulonephritis (GN). The animals were fed eithera normal-protein (NP, 22% casein) or a low-protein diet (LP, 6% casein). Enalapril (Enal, 100 mg/L) or losartan (Los, 500 mg/L) was given inthe drinking water (*P , 0.01 vs. NP; #P , 0.05 vs. LP, NP 1 Enal, or NP 1 Los).

to 2.12 6 0.04 in low-protein–treated, to 2.08 6 0.08 in restriction by 46%, with enalapril treatment by 43%, andwith losartan treatment by 36%, while glomerular PAI-1the enalapril-treated, and to 2.08 6 0.12 in losartan-

treated and nephritic animals (Fig. 2). The changes in production decreased by 41, 33, and 36%, respectively(Fig. 2). These data indicate that similar reductions inTGF-b1 production paralleled glomerular matrix accu-

mulation and were also similarly reduced in all three TGF-b1 expression with different drugs lead to similarreductions in matrix protein and PAI-1 production. Nonegroups, indicating that glomerular TGF-b production is

a valid indicator of histologic glomerular matrix expan- of these parameters were significantly different betweenthe nephritic animals treated with low-protein diet, ena-sion. TGF-b1 production was reduced with dietary pro-

tein restriction by 49%, with enalapril by 42%, and with lapril, or losartan alone.Northern blotting was performed to analyze whetherlosartan treatment by 42% (Fig. 2). The differences be-

tween these groups were not significantly different. the reduction of TGF-b1, fibronectin, and PAI-1 achievedby enalapril, losartan, or low-protein diet at the proteinChanges in TGF-b1 production were also closely par-

alleled by changes in glomerular fibronectin and PAI-1 level were related to decreased mRNA expression ofthese components. As shown in Figure 3, the expressionproduction, both sensitive indicators of TGF-b’s action.

Fibronectin production was reduced with dietary protein of the glomerular TGF-b1, fibronectin, and PAI-1 mRNA,

Peters et al: Additive therapeutic effects in GN1498

Fig. 3. Relative glomerular mRNA expression of TGF-b1 (A), fibronectin (B), and PAI-1 (C) six days after the induction of mesangioproliferativeglomerulonephritis (GN). (D) A representative Northern blot (one representative out of three). The animals were fed either a normal-protein(NP, 22% casein) or a low-protein diet (LP, 6% casein). Enalapril (Ena, 100 mg/L) or losartan (Los, 500 mg/L) was given in the drinking water(*P , 0.05 vs. NP). Molecular size markers are shown on the right.

corrected for GAPDH, showed the same pattern as their at least in part through reduced expression of TGF-b,production by nephritic glomeruli, although the relative fibronectin, and PAI-1 mRNA.reduction was less pronounced than the results in the

Effect of combined treatment with angiotensin IIprotein production (Fig. 2). Compared with the mRNAblockade and low-protein dietexpression seen in the disease control animals, TGF-b

expression was reduced 23% with dietary protein restric- As shown in the right-hand bars of Figure 1, cotreat-ment with enalapril or losartan and dietary protein re-tion, 24% with enalapril, and 22% with losartan (P ,

0.05). Glomerular mRNA expression of fibronectin was striction resulted in a significantly greater reduction inurinary protein excretion than single therapy (P , 0.05).reduced by 38, 32, and 47%, while glomerular PAI-1

mRNA expression decreased by 31, 25, and 39% with Similarly, compared with the animals treated with enala-pril only (2.08 6 0.08), adding low protein reduced thelow protein, enalapril, and losartan, respectively, (P ,

0.05). The differences between the three groups were not matrix score to 1.76 6 0.08. In the losartan-treated ani-mals, additional dietary protein restriction reduced thesignificant. Thus, the data suggest that these therapies act

Peters et al: Additive therapeutic effects in GN 1499

matrix score from 2.08 6 0.12 to 1.64 6 0.04 (Fig. 2). dietary protein restriction significantly reduced patho-logical TGF-b overexpression following induction ofCotreatment also resulted in a significantly greater re-

duction in glomerular TGF-b (Fig. 2). Compared with GN, but none of the three therapies were superior. How-ever, a moderate but significant additional reduction inenalapril alone (42%), an addition of low-protein diet

reduced TGF-b overexpression by 65%. The addition TGF-b overexpression and most other measures of dis-ease was achieved when Ang II blockade and low proteinof low-protein diet to losartan therapy reduced TGF-b

overexpression from 42 to 60%. intake were combined.The results presented here for the low-protein dietConsistent with data on single therapy, the changes

in glomerular fibronectin and PAI-1 production closely group differ somewhat from our earlier data using theATS model of GN [19]. While we cannot be sure of thefollowed the changes seen in TGF-b production (Fig.

2), indicating comparable reductions in TGF-b’s action. reasons for an apparently more pronounced effect ofprotein restriction seen earlier, a number of possibilitiesCompared with disease control animals, a combination

of low-protein diet and enalapril reduced glomerular exist. Earlier studies did not aim to quantitate therapeuticeffect but merely to show a significant effect. We nowfibronectin production by a total of 72% and glomerular

PAI-1 synthesis by a total of 61%. Similarly, losartan use highly sensitive ELISA techniques to quantitate invitro glomerular production of proteins of interest. Pilotplus a low-protein diet resulted in a total reduction of

63% in fibronectin and 59% in PAI-1 production. experiments have shown that TGF-b, PAI-1, and fibro-nectin production are linear for both normal and diseasedGlomerular mRNA analysis of TGF-b1, fibronectin,

and PAI-1 expression corrected for GAPDH showed glomeruli for 72 hours, after which it falls (unpublisheddata). ELISA data are also obtained from individual rats,the same pattern as the production of corresponding

proteins by nephritic glomeruli, although they did not whereas Northern blotting currently requires pooling ofRNA from three to four rats. The largest differencereach statistical significance. Compared with the mRNA

expression found in untreated nephritic rats, therapy between this and the earlier study of low protein is inthe data for PAS-stained tissue. However, quantitationwith enalapril and low-protein diet reduced glomerularof PAS staining is by far the most variable of the mea-TGF-b1 mRNA by a total of 30%, fibronectin mRNAsures presented both because of variability in tissue sec-by 58%, and PAI-1 mRNA by 45%. In the losartan-tioning, staining intensity, and variability between scores.treated animals fed a low-protein diet, total reductionsBecause we have obtained very similar data in otherin TGF-b1, fibronectin, and PAI-1 mRNA expressionstudies using Ang II blockade [13] and low protein [27],were 27, 73, and 56%, receptively. None of these parame-we believe the data presented here more precisely repre-ters were significantly different between the rats treatedsent the true efficacy of these therapies in this modelwith enalapril and a low-protein diet and those treatedthan earlier published work.with losartan and a low-protein diet.

An earlier study reported an additive therapeutic ef-fect of captopril and low protein in unilaterally nephrec-

DISCUSSION tomized rats [28]. The present report confirms and ex-The present study demonstrates that the combination pands this earlier report in a number of important ways.

of Ang II blockade and a low-protein diet reduces glo- First, this study goes to the molecular level by showingmerular TGF-b overexpression and matrix accumulation that the decrease in matrix accumulation is likely tomore effectively than treatment with low-protein diet, result from both reduced matrix synthesis, as exemplifiedenalapril, or losartan alone. The data further validate by data on fibronectin, and less inhibition of matrix deg-our approach of evaluating, contrasting, and maximizing radation, as shown by the PAI-1 data. Second, the usethe antifibrotic actions of currently established and of AT1 receptor antagonist losartan shows that additivepromising new therapies for progressive fibrotic renal effects shown with ACE inhibition and low-protein dietdiseases by examining their potential to reduce patholog- also apply to receptor antagonists. Third, the doses ofical TGF-b expression [13]. The changes in histologic enalapril or losartan used here are those that have re-matrix accumulation closely followed the reductions in cently been shown to produce maximal therapeutic ef-TGF-b overexpression, as did the in vivo indicators of fects in this model [13]. The use of maximally effectiveTGF-b’s action: fibronectin and PAI-1. Because TGF-b doses of enalapril and losartan is important. If drugsproduction by glomeruli in vitro is many-fold higher in that have the same mechanism of action are given atnephritic animals, its use as a measure of disease severity submaximal doses, additive effects can result simply fromprovides a much larger window to contrast effects of a dosing effect. However, if maximal doses are used andsingle or combined therapeutic interventions than con- additivity is found, different mechanisms of action of theventional variables such as histologic score or protein- drugs are suggested. Thus, the finding that dual therapyuria. As expected from previous studies [13, 19], we results in further reductions of TGF-b overproduction

and a further reduction in TGF-b’s fibrogenic actionsfound that single therapy with Ang II-blocking drugs or

Peters et al: Additive therapeutic effects in GN1500

strongly suggests that these therapies act, to some extent, intake. The addition of low protein to Ang II blockadedoes significantly boost efficacy, but the effect is smallon different stimuli to TGF-b overexpression.

Consistent with our previous report, the present study relative to differences between no treatment and singletherapy. Thus, in the complex situation that exists indemonstrates that treament with maximally effective

doses of enalapril or losartan reduces glomerular TGF-b human clinical trails, such a small effect may be difficultto detect. Supporting this concept is a human study byoverexpression and matrix accumulation to a similar ex-

tent [13]. At first glance, this may favor the notion that Ihle et al showing that the therapeutic effect of low-protein diet can be easily demonstrated in a small num-their main therapeutic action is mediated through inhibi-

tion of Ang II as the principal effector molecule of the ber of patients when ACE inhibitors are not used [37].The ATS model of GN is a well-characterized modelRAS. However, there is increasing evidence that addi-

tional effects of ACE inhibition such as bradykinin/nitric of acute tissue injury followed by fibrotic repair. Onemay argue that chronic progressive disease differs fromoxide accumulation or additional effects of receptor an-

tagonism due to accumulation of Ang II breakdown acute reversible disease and that renal disease in rats dif-fers from that seen in humans. A strong counter argumentproducts may be operating [13, 29, 30]. High circulating

levels of Ang II resulting from receptor antagonism may is that the similarities are far greater than the differences.TGF-b overproduction has been shown in essentially allactivate AT2 receptors or produce increases in angioten-

sin breakdown products that may have important biolog- fibrotic diseases in which it has been investigated [2, 6].Furthermore, the degree of TGF-b overexpression isical actions [30, 31]. Both activation of AT2 receptors and

increased levels of angiotensin fragments have recently generally correlated with the rate of accumulation of theextracellular matrix. Rapidly progressing disease such asbeen linked to increases in the bradykinin/nitric oxide

pathway [13]. HIV nephropathy or anti-glomerular basement mem-brane disease show marked overexpression of TGF-b,Compared with Ang II blockade, the mechanisms un-

derlying the antifibrotic action of low-protein diet are while slowly progressing diabetic nephropathy showsmore slowly increasing TGF-b overexpression [38–40].far less understood. Studies by Rosenberg et al have

shown that low-protein feeding markedly reduces renal Therefore, we argue that the present study may be rele-expression of renin and angiotensinogen, suggesting that vant for treatment of patients with chronic renal failure,a low-protein diet, similar to Ang II blockade, acts through and suggest that use of a combination of Ang II blockadea reduction of renal RAS [5]. However, the finding here and protein restriction in chronic human renal diseasesthat dietary protein restriction further reduces pathologi- might be beneficial.cal TGF-b overexpression in the presence of maximal In conclusion, the present study shows that doublepharmacological Ang II inhibition strongly implies that therapy with Ang II blockers and a low-protein diet sig-a low-protein diet has therapeutic actions separate from nificantly enhances therapeutic efficacy in renal fibrosis.those of enalapril or losartan. It is conceivable that these The results suggest that low protein acts by pathways inactions involve RAS. For example, a low-protein diet addition to blocking Ang II and supports the conceptmay reduce the very high plasma renin levels caused by that the rate of progression can be retarded more effi-Ang II blockade. Low protein may also act on pathways ciently when Ang II blockade is combined with dietaryindependent of RAS. We have recently reported that a protein restriction in patients with chronic renal failure.lower intake of the semiessential amino acid l-argininemay be a key factor in the effect of low-protein diet [32, ACKNOWLEDGMENTS33]. Evidence for a considerable number of other actions This work was supported by National Institute of Diabetes andof dietary protein restriction was recently reviewed [18]. Digestive and Kidney Diseases grant DK 49342 to N.A.N. and DK

43609 and DK 49374 to W.A.B. H.P. was supported by a matchingAlthough still controversial, several clinical studiesfellowship from the National Kidney Foundation of Utah. We thankhave suggested that dietary protein restriction slows theMs. Linda Hoge, Ms. Diane Miller, and Mr. Trung Le for excellent

rate of progression of chronic renal failure [34–37]. The technical assistance.recent Modification of Diet in Renal Disease (MDRD)

Reprint requests to Nancy A. Noble, Ph.D., Fibrosis Research Labo-study encountered difficulties demonstrating efficacy ofratory, University of Utah School of Medicine, Salt Lake City, Utah

dietary protein restriction despite a large number of study 84132, USA.E-mail: Nancy.Noble@hsc.utah.edusubjects [35, 36]. In that study, the rate of loss of the

glomerular filtration rate and the beneficial effect of low-REFERENCESprotein diet were less than predicted from earlier studies.

However, the majority of patients were being treated 1. Border WA, Noble NA: Transforming growth factor-b in tissuefibrosis. N Engl J Med 331:1286–1292, 1994with ACE inhibitors. An important concept, brought out

2. Noble NA, Border WA: Angiotensin II in renal fibrosis: Shouldby the data presented here, is that the majority of theTGF-b rather than blood pressure be the therapeutic target? Semin

therapeutic effect is seen with single therapy, whether Nephrol 17:455–466, 19973. Border WA, Okuda S, Languino LR, Sporn MB, RuoslahtiACE inhibition, receptor antagonism, or low protein

Peters et al: Additive therapeutic effects in GN 1501

E: Suppression of experimental glomerulonephritis by antiserum zyme-linked immunoassay (ELISA) for connective tissue compo-nents. Anal Biochem 104:205–214, 1980against transforming growth factor beta-1. Nature 346:371–374,

1990 23. Fort P, Marty L, Piechaczyk M, Sabrouty SE, Dani C, JeanteurPH, Blanchard JM: Various rat adult tissues express only one4. Border WA, Noble NA, Yamamoto T, Harper JR, Yamaguchi

Y, Pierschbacher MD, Ruoslahti E: Natural inhibitor of trans- major mRNA species from the glyceraldehyde-3-phosphate-dehy-drogenase multigenic family. Nucleic Acids Res 13:1431–1442, 1985forming growth factor-b protects against scarring in experimental

kidney disease. Nature 360:361–364, 1992 24. Derynck R, Jarrett A, Chen EY, Goeddel DV: The murinetransforming growth factor-b precursor. J Biol Chem 261:4377–5. Rosenberg ME, Smith LJ, Correa Rotter R, Hostetter TH: The

paradox of the renin-angiotensin system in chronic renal disease. 4379, 198625. Schwarzbauer JE, Tamkun JW, Lemischka IR, Hynes RO: ThreeKidney Int 45:403–410, 1994

6. Border WA, Noble NA: Interactions of TGF-b and angiotensin different fibronectin mRNAs arise by alternative splicing withinthe coding region. Cell 35:421–431, 1983II in renal fibrosis. Hypertension 31:181–188, 1998

7. Wolf G: Molecular mechanisms of angiotensin II in the kidney: 26. Zeheb R, Gelehrter TD: Cloning and sequencing of cDNA forthe rat plasminogen activator inhibitor-1. Gene 73:459–468, 1988Emerging role in the progression of renal disease: Beyond haemo-

dynamics. Nephrol Dial Transplant 13:1131–1142, 1998 27. Peters H, Border WA, Noble NA: Tandem antifibrotic actionsof l-arginine supplementation and low protein diet during the8. Kagami S, Border WA, Miller DE, Noble NA: Angiotensin II

stimulates extracellular matrix protein synthesis through induction repair phase of experimental glomerulonephritis. Kidney Int55:2264–2273, 1999of transforming growth factor-b expression in rat glomerular mes-

angial cells. J Clin Invest 93:2431–2437, 1994 28. Cartwright ME, Jaenke RS: Effects of dietary protein and capto-pril on glomerular permselectivity in rats with unilateral nephrec-9. Kagami S, Kuhara T, Okada K, Kuroda Y, Border WA, Noble

NA: Dual effects of angiotensin II on the plasminogen/plasmin tomy. Lab Invest 59:492–499, 198829. Ichikawa I: Will angiotensin II receptor antagonists be renoprotec-system in rat mesangial cells. Kidney Int 51:664–671, 1997

10. Wolf G, Ziyadeh FN: Renal tubular hypertrophy induced by tive in humans? Kidney Int 50:684–692, 199630. Campbell DJ, Kladis A, Valentijn AJ: Effects of losartan onangiotensin II. Semin Nephrol 17:448–454, 1997

11. Wolf G, Ziyadeh FN, Thaiss F, Tamaszewski J, Caron RJ, Wen- angiotensin and bradykinin peptides and angiotensin-convertingenzyme. J Cardiovasc Pharmacol 26:233–240, 1995zel U, Zahner G: Angiotensin II stimulates expression of the

chemokine RANTES in rat glomerular endothelial cells. J Clin 31. Siragy HM, Carey RM: The subtype-2 (AT2) angiotensin receptorregulates renal cyclic guanosine 39,59-monophosphate and AT1Invest 100:1047–1058, 1997

12. Sharma K, Ziyadeh FN, Alzahabi B, McGowan TA, Kapoor S, receptor-mediated prostaglandin E2 production in conscious rats.J Clin Invest 97:1978–1982, 1996Kurnik BRC, Kurnik PB, Weisberg LS: Increased renal produc-

tion of transforming growth factor-b1 in patients with type II diabe- 32. Narita I, Border WA, Ketteler M, Ruoslahti E, Noble NA:l-arginine may mediate the therapeutic effects of low protein diets.tes. Diabetes 46:854–859, 1997

13. Peters H, Border WA, Noble NA: Targeting TGF-b overexpres- Proc Natl Acad Sci USA 92:4552–4556, 199533. Peters H, Noble NA: Dietary l-arginine in renal disease. Seminsion in renal disease: Maximizing the antifibrotic action of angioten-

sin II blockade. Kidney Int 54:1570–1580, 1998 Nephrol 16:567–575, 199634. Rosman JB, Ter-Wee PM, Meijer S, Piers-Becht TP, Sluiter14. Fern RJ, Yesko CM, Thornhill BA, Kim H-S, Smithies O, Cheva-

lier RL: reduced angiotensinogen expression attenuates renal in- WJ, Donker AJ: Prospective randomised trial of early dietaryprotein restriction in chronic renal failure. Lancet 2:1291–1296,terstitial fibrosis in obstructive nephropathy in mice. J Clin Invest

103:39–46, 1999 198435. Klahr S, Levey AS, Beck GJ, Caggiula AW, Hunsicker L, Kusek15. Farr LE, Smadel JE: The effect of dietary protein on the course

of nephrotoxic nephritis in rats. J Exp Med 70:615–627, 1939 JW, Striker G: The effects of dietary protein restriction and blood-pressure control on the progression of chronic renal disease: Modi-16. Rosenberg ME, Chmielewski D, Hostetter TH: Effect of dietary

protein and rat renin and angiotensin gene expression. J Clin Invest fication of Diet in Renal Disease Study Group. N Engl J Med330:877–884, 199485:1144–1149, 1992

17. Hostetter TH, Meyer TW, Rennke HG, Brenner BM, Noddin 36. Levey AS, Adler S, Caggiula AW, England BK, Greene T,Hunsicker LG, Kusek JW, Rogers NL, Teschan PE: Effects ofJA, Sandstrom DJ: Chronic effects of dietary protein in the rat

with intact and reduced renal mass. Kidney Int 30:509–517, 1986 dietary protein restriction on the progression of advanced renaldisease in the modification of diet in renal disease study. Am J18. Heidland A, Sebekova K, Ling H: Effect of low-protein diets on

renal disease: Are non-haemodynamic factors involved? Nephrol Kidney Dis 27:652–663, 199637. Ihle BU, Becker GJ, Whitworth JA, Charlwood RA, KincaidDial Transplant 10:1512–1514, 1995

19. Okuda S, Nakamura T, Yamamoto T, Ruoslahti E, Border WA: Smith PS: The effect of protein restriction on the progression ofrenal insufficiency. N Engl J Med 321:1773–1777, 1989Dietary protein restriction rapidly reduces transforming growth

factor b1 expression in experimental glomerulonephritis. Proc Natl 38. Iwano M, Kubo A, Nishino T, Sato H, Nishioka H, Akai Y,Kurioka H, Fujii Y, Kanauchi M, Shiiki H, Dohi K: QuantificationAcad Sci USA 88:9765–9769, 1991

20. Eddy AA: Protein restriction reduces transforming growth factor- of glomerular TGF-b1 mRNA in patients with diabetes mellitus.Kidney Int 49:1120–1126, 1996beta and interstitial fibrosis in nephrotic syndrome. Am J Physiol

266:F884–F893, 1994 39. Yamamoto T, Noble NA, Cohen AH, Nast CC, Hishida A, GoldLI, Border WA: Expression of transforming growth factor-b iso-21. Nakayama M, Okuda S, Tamaki K, Fujishima M: Short- or long-

term effects of a low protein diet on fibronectin and transforming forms in human glomerular diseases. Kidney Int 49:461–469, 199640. Yamamoto T, Noble NA, Miller DE, Gold LI, Hishida A, Na-growth factor-beta synthesis in ariamycin-induced nephropathy.

J Lab Clin Med 127:29–39, 1996 gase M, Cohen AH, Border WA: Increased levels of TGF-b inHIV-associated nephropathy. Kidney Int 55:579–592, 199922. Rennard SI, Berg R, Martin GR, Foidart JM, Robey PG: En-