Kidney International, Vol. 67 (2005), pp. 2422–2433

Rapamycin inhibits human renal epithelial cell proliferation:Effect on cyclin D3 mRNA expression and stability

NICOLAS PALLET, ERIC THERVET, DELPHINE LE CORRE, BERTRAND KNEBELMANN,PATRICK NUSBAUM, CELINE TOMKIEWICZ, PAUL MERIA, JEAN-PIERRE FLINOIS, PHILIPPE BEAUNE,CHRISTOPHE LEGENDRE, and DANY ANGLICHEAU

Unite INSERM UMR S490, Molecular Toxicology, Centre Universitaire des Saints-Peres, Universite Rene Descartes, Paris, France;Service de Nephrologie et de Transplantation Renale, Hopital Saint Louis, Paris, France; Unite INSERM U507, Hopital Necker,Paris, France; and Service d’Urologie, Hopital Saint Louis, Paris, France

Rapamycin inhibits human renal epithelial cell proliferation:Effect on cyclin D3 mRNA expression and stability.

Background. Recent data have suggested that rapamycin useduring the initial period after transplantation is associated withprolonged delayed graft function (DGF). Because of the knowneffects of rapamycin in other cell types, we speculated thatthis action may be secondary to human renal epithelial cells(HRECs) inhibition of proliferation.

Methods. Primary cultures of HRECs were incubated withvarious concentrations of rapamycin. Cell proliferation wasevaluated by cytotoxicity assays. The cell cycle was analyzedby flow cytometry. Protein expression levels were assessed byWestern blot. Cyclin D3 mRNA levels were measured by quan-titative real-time polymerase chain reaction (PCR). The tran-scriptional activity of the cyclin D3 gene was evaluated usingtransient transfection.

Results. Rapamycin exerted a significant concentration-dependent antiproliferative effect on growing HRECs byinhibiting the G1 to S transition. The p70S6 kinase pathway lead-ing to cell cycle progression was found to be active, and lowconcentrations of rapamycin dramatically reduced p70S6 kinasephosphorylation. Rapamycin completely inhibited the increasein cyclin D3 protein expression and mRNA accumulation in-duced by fetal calf serum, but did not affect cyclin E or cdk-inhibitor expression levels. This regulation of cyclin D3 proteinexpression is mainly due to a destabilization of its mRNA. Ra-pamycin reduced the mRNA half-life by 26% (4.8 ± 1.3 hoursvs. 6.5 ± 1.0 hours, P < 0.001).

Conclusion. Rapamycin inhibits the proliferative responseof HRECs to mitogenic stimuli, and causes cell cycle arrest inthe early G1 phase, not only by a nonspecific process due toinhibition of the p70S6k pathway, but also by a direct effect oncyclin D3 mRNA stability.

Key words: rapamycin, p70S6 kinase, cyclin D, delayed graft function,renal transplantation.

Received for publication July 10, 2004and in revised form October 26, 2004Accepted for publication December 20, 2004

C© 2005 by the International Society of Nephrology

Rapamycin is a new potent immunosuppressive agentused as an immunosuppressive drug [1]. Rapamycin actsby inhibiting the proliferation and clonal expansion ofinterleukin-2–stimulated T cells. When complexed withFK-binding protein-12, its cellular receptor, rapamycinbinds to and inhibits the kinase activity of mTOR (mam-malian target of rapamycin) [2], a central controller ofcell growth. In response to amino acids and growth fac-tors, mTOR controls the mammalian translation ma-chinery by activating the protein kinase p70S6 (p70S6k)and inhibiting the eIF4E inhibitor 4E-BP1. Activationof p70S6k and the resulting phosphorylation of the 40Sribosomal protein S6 drive the mRNAs translation thatprimarily encodes ribosomal proteins and componentsof the translational apparatus. Thus, under favorablegrowth conditions, phosphorylated p70S6k up-regulatesthe translational machinery. TOR also phosphorylates4E-BP1, a protein associated with the translation initi-ation factor eIF4E. Phosphorylation of 4E-BP1 inducedby various mitogens releases 4E-BP1 from this complex,allowing eIF4E to initiate the translation process [3].EIF4E controls the translation of mRNAs such as thetranscripts encoding cyclin D1 and the c-myc oncogene[3]. By inhibiting TOR, rapamycin reduces the transla-tional process and arrests proliferating T-cell growth inthe mid-to-late G1 phase [4]. Rapamycin has also beenshown to block the G1 to S phase transition in other celltypes [5–7].

After renal transplantation, delayed graft function(DGF), defined as the need for at least one hemodial-ysis session during the first week after transplantation,has been associated with both long-term and short-termdecrease of graft survival. DGF is secondary to ischemia-reperfusion injury, and is associated with acute tubularnecrosis [8]. Recent data have shown that, despite theabsence of specific nephrotoxicity, rapamycin might in-crease the prevalence, length, and severity of DGF whenused during the initial phase after transplantion [9–11].

2422

Pallet et al: Antiproliferative effect of rapamycin on human renal epithelial cells 2423

The rate of recovery of tubular cell function is thought todepend on a balance between tubular cell death and theability of injured cells to enter the cell cycle and prolif-erate [12, 13]. We therefore postulated that the negativeeffect of rapamycin on recovery from DGF may be sec-ondary to its antiproliferative action. Since this actionhad never been reported in human renal epithelial cells,the aim of this study was to investigate the effects of ra-pamycin on cell proliferation, and to explore the mecha-nisms of these effects, using human renal epithelial cells(HRECs) in primary culture.

METHODS

Reagents and antibodies

Rapamycin was a gift from Wyeth Research(Monmouth Junction, NJ, USA). All other chemi-cals were obtained from Sigma-Aldrich (Saint QuentinFallavier, France). Rapamycin was diluted in ethanol. Thecell culture medium and the other cell culture productswere from Life Technologies (Cergy Pontoise, France).Antibodies (Abs) specific for p16INK4

, cyclin D1, cy-clin D2, cyclin D3, and cyclin E were purchased fromBD Biosciences (Le Pont de Claix, France). Phospho-p70S6k (Thr389 and Thr421/Ser424) and total p70S6k Abswere from Cell Signaling Technology (Saint Quentinen Yvelines, France), and Abs specific for p21Cip1 andp27Kip1 were from Santa Cruz Biotechnology (Le Perrayen Yvelines, France). Anti-actin Ab was obtained fromSigma-Aldrich.

Cell culture methods

Normal human renal epithelial cells (HRECs) wereharvested from human nephrectomy specimens removedfor renal cell carcinoma, and isolated according to pre-viously published methods, with minor modifications[14, 15]. Fragments of nontumoral renal cortex wereminced and digested with collagenase IV (250 IU/mL)for three hours at 37◦C. Cells were centrifuged andthe pellets washed three times with phosphate-bufferedsaline (PBS). Cells were then cultured in Dulbecco’smodified Eagle’s medium (DMEM) containing 5 lg/mLinsulin, 10 lg/mL human apotransferrin, 500 ng/mL hy-drocortisone, 10 ng/mL EGF, 6.5 ng/mL triiodothyronin,5 ng/mL sodium selenite, 1% fetal calf serum (FCS),25 IU/mL penicillin, 25 lg/mL streptomycin, and 10mmol/L HEPES buffer. Cells were then incubated at37◦C in 5% CO2 and 95% air. The characterization of thiscellular model was performed by immunocytochemicalperoxydase analysis and flow cytometry analysis (data notshown). Immunocytochemical analysis revealed a pos-itive staining for cytokeratin (clone AE1/AE3) and a6

integrin (clone NKI-GoH3), two markers of epithelialcells [16, 17], by the vast majority of cells, thus confirmingtheir epithelial nature. A small subset of cells also stained

positively for epithelial membran antigen (clone E29), amarker of distal tubular and collecting duct cells [17], sug-gesting their distal origin. FACS analysis showed a strongstaining for alanine aminopeptidase (CD13, clone WM-47), a cell surface marker that distinguishes proximalfrom distal tubular epithelial cells, but not from fibrob-lasts [18], and the absence of staining for CD90 (clone5E10), which is present on fibroblasts, but not on tubularepithelial cells [19]. These results confirmed the proximaldescent of the vast majority of the cultured tubular ep-ithelial cells. Experiments were not performed with cellsbeyond the third passage because it has been shown thatno phenotypic changes occur up to this passage number[20].

Cell proliferation and cytotoxicity assay

HRECs (103 cells/well) were seeded in 96-well plates,starved for another 24 hours in serum-free medium, andthen treated with rapamycin and FCS. Six days later, therelative number of living cells per well was determined onthe basis of mitochondrial integrity by assay with [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sul-fophenyl)-2H-tetrazolium (MTS) (Promega,Charbonniers, France), according to the manufacturer’sinstructions.

Cell death was determined by measuring the amount oflactate dehydrogenase (LDH) released into the mediumafter treatment with rapamycin. LDH activity was deter-mined by the CytoTox 96 non-radioactive cytotoxicity as-say (Promega), and quantified by measuring absorbanceat 490 nm.

Protein isolation and Western blots

All experiments were done with subconfluent cells. Inall, 2 × 106 cells were washed once with cold PBS, har-vested with a rubber policeman, and lysed for 30 minutesat 4◦C in lysis buffer containing 10 mmol/L Tris-HCl (pH7.5), 10% glycerol, 1% Triton X-100, 0.5% Nonidet P40,1 mmol/L EDTA, 1 mmol/L EGTA, 150 mmol/L NaCl,50 mmol/L NaF, 1 mmol/L sodium orthovanadate, anda cocktail of protease inhibitors. Lysates were spun at12,000 rpm for four minutes, and the supernatant wascollected. The protein concentration was assessed by thebicinchoninic acid method (Pierce Chemicals, Rockford,IL, USA).

Total protein extracts (50 lg/lane) were separated by9% to 12% sodium dodecyl sulfate-polyacrylamide gelelectophoresis (SDS-PAGE) under denaturing and re-ducing conditions and transferred to polyvinylidene fluo-ride (PVDF) membrane (Amersham Pharmacia Biotech,Orsey, France). Membranes were blocked for 1 hourat room temperature with TTBS (20 mmol/L Tris, 150mmol/L NaCl, and 0.1% Tween 20) containing 5% non-fat dry-milk. Blots were incubated with the primary anti-body for 1 hour at room temperature, or overnight at 4◦C,

2424 Pallet et al: Antiproliferative effect of rapamycin on human renal epithelial cells

Table 1. Primer sequences used for real-time quantitative polymerase chain reaction

GenBankRNAs accession number Primer sequences Fragment length

Human cyclin D3 mRNA NM001760 5′-TGGATGCTGGAGGTATGTG-3′ 205 bp5′-CGTGGTCGGTGTAGATGC-3′

Human ribosomal protein L13a mRNA BC001836 5′-CCTGGAGGAGAAGAGGAAAGAGA-3′ 124 bp5′-GAGGACCTCTGTGTATTTGTCAA-3′

18S rRNA K03432 5′-TCCCCCAACTTCTTAGAGG-3′ 234 bp5′-CTTATGACCCGCACTTACTG-3′

and then for 30 minutes with a horseradish peroxidase-conjugated secondary antibody (1/10,000 in TTBS).Proteins were stained by chemoluminescence with theenhanced chemiluminescence (ECL) system (AmershamPharmacia Biotech), followed by exposure to x-ray film.Blots were then probed with an anti-actin polyclonalantibody to check the amount of protein loaded ontothe gel.

Flow cytometry analysis

Cells were seeded in 100 mm dishes. At 50% con-fluence, they were shifted to serum-free conditions for24 hours. Cells were then incubated with 20 nmol/L ra-pamycin for one hour, and stimulated with 10% FCS.Eighteen hours later, they were labeled with 10 lmol/Lbromodeoxyuridine (BrdU) for one hour, harvestedby trypsinization, and washed once with PBS. Fixationand permeabilization were performed according to theBD Pharmingen BrdU Flow Kit staining protocol (BDPharmingen, Le Pont de Claix, France). After perme-abilization, cells were incubated with 300 lg/mL DNAseto expose incorporated BrdU, and then incubated for 20minutes at room temperature in the dark with fluores-cein isothiocyanate (FITC)-conjugated anti-BrdU anti-bodies. Cells were resuspended with the fluorescent dye7-amino actinomycin D for staining of total DNA. Up to10,000 events/sample were analyzed. DNA was quanti-fied by FACScan (Becton Dickinson, Franklin Lakes, NJ,USA).

RNA isolation and cDNA synthesis

Total RNA was extracted from approximately 5 × 106

cells with the RNA Easy Midi Kit (Qiagen, Valencia,CA, USA). Reverse transcription was performed on eachRNA sample (2 lg) using oligo dT and the High-CapacitycDNA Archive Kit (Applied Biosystems, Foster City, CA,USA) according to the manufacturer’s instructions.

Quantitative real-time PCR

The mRNA levels of cyclin D3 were measured bySYBR green real-time polymerase chain reaction (PCR),and quantified using an ABI PRISM 7900 sequence de-tector system (Applied Biosystems). Primers were usedto amplify a 205-base fragment of the human cyclin D3gene mRNA (Table 1). This sequence was confirmed to

be specific to cyclin D3 mRNA (not homologous with theother D-type cyclins mRNAs). Results were normalizedto the endogenous control human ribosomal protein L13amRNA for the study of the transcriptional activity of thecyclin D3 gene, and to the 18S rRNA gene for the eval-uation of the cyclin D3 mRNA half-life. Previous experi-ments were conducted to test the impact of rapamycinon 18S rRNA, and found no alteration of 18S rRNAexpression by rapamycin. We did not use an mRNA asa endogenous control but a rRNA because 5,6-dichloro-1b-D-ribofuranosylbenzimidazole (DRB) is known as aspecific inhibitor for RNA polymerase II transcription[21]. It then affects the mRNA transcription, but has lim-ited effect on rRNA, whose transcription is governedby RNA polymerase I in eukaryotes [22]. DRB, thus,allowed the analysis of the cyclin D3 mRNA expres-sion profile and, finally, cyclin D3 mRNA half-life. Real-time PCR was performed in triplicate reactions with 20ng of cDNA in a final volume of 10 lL containing 1XSYBR Green PCR Master Mix (Applied Biosystems)and 300 nmol/L of both primers. Samples were heatedfor 10 minutes at 95◦C, followed by 40 cycles of amplifi-cation for 15 seconds at 95◦C, and 1 minute at 60◦C.

To compare the expression levels among different sam-ples, the relative expression of cyclin D3 mRNA levelswas calculated using the comparative delta Ct (thresholdcycle number) method as described by Livak et al [23].

Plasmid, transfection, and luciferase assay

The cyclin D3 promotor-luciferase construct pGL3-hCyclinD3 was generously provided by Dr. Cress(Moffitt Cancer Center, FL, USA) [24]. The transcrip-tional factor E2F1 expression plasmid pCMV-E2F1 waskindly provided by Dr. Transy (INSERM UMR-S490,Paris, France) [25, 26]. Plasmid DNA was prepared us-ing QIAfilter columns (Qiagen).

For transfection of the reporter construct, 60 × 105

HRECs were seeded per well in 12-well plates. The nextday, cells were starved in serum-free medium for 24 hours.They were then transfected with 1 lg of plasmid pGL3-hCyclinD3 using the mixed lipofection reagent FuGene6 (Roche Molecular Biochemicals, Meylan, France) ata ratio of 6 lL of FuGene to 1 lg of DNA in serum-free medium. In some experiments, 100 ng of the ex-pression plasmid pCMV-E2F1 (or the corresponding

Pallet et al: Antiproliferative effect of rapamycin on human renal epithelial cells 2425

empty expression vector pCMV) was cotransfected.Twenty-four hours later, they were stimulated with FCSin the presence or absence of 20 nmol/L of rapamycinfor 24 hours. After lysis of the cells in Reported LysisBuffer (Promega), luciferase content was quantified byluciferase assay (Promega).

Statistical analysis

All data were expressed as mean ± SEM of threedifferent experiments, unless otherwise specified. Sta-tistical significance was tested by nonparametric tests(Mann Whitney U test for unpaired comparisons andWilcoxon test for paired comparisons). Relative densit-ometric values were estimated using NIH Image 1.63software (National Institutes of Health, Bethesda, MD,USA). Statistical analyses were performed using Statview5.0.1® software (Cary, NC, USA). P values of less than0.05 were considered significant.

RESULTS

Rapamycin inhibits cell proliferation

We first evaluated the impact of increasing concentra-tions of rapamycin on cell viability, and found that ra-pamycin reduces viable cell numbers in a dose-dependentmanner compared to untreated cells (Fig. 1A). Thisreduction occurred even with low concentrations ofrapamycin. At the concentration of 10 nmol/L, usuallyobserved in the blood of patients treated with rapamycinin clinical practice, rapamycin reduced cell viability by upto 40% compared to untreated cell viability (P < 0.05).

To establish whether this reduced viability resultedfrom cell death or from decreased proliferation, we mea-sured the amount of LDH released into the supernatant.Such release occurs after cell death, whether the latter iscaused by necrosis or apoptosis. As shown in Figure 1B,the total LDH content, indirectly reflecting the total cellnumber, decreased in a dose-dependent manner. With 10nmol/L rapamycin, the total LDH content decreased by50% compared to untreated cells, confirming the resultobserved with the cell proliferation assay (Fig. 1A). Anadditional cytotoxic effect may be present with the higherrapamycin concentrations because the LDH released inthe medium remained stable in spite of a decreased totalcell number.

When cell growth was stimulated with increasing con-centrations of FCS, the antiproliferative effect of ra-pamycin persisted. For all concentrations, the number ofcells decreased by 30% to 40% when they were incubatedwith 20 nmol/L of rapamycin (Fig. 1C). The antiprolifer-ative effect of rapamycin was not detectable in quiescentcells cultured in a serum-free medium, even after 6 days.This finding confirms that the decrease of cell viability in-duced by rapamycin is secondary to its inhibition of cellproliferation (Fig. 1C).

100

80

60

40

20

00 5 10 15 20

C

Num

ber

of v

iabl

e ce

lls,

% o

f m

axim

um

Fetal calf serum, %

Untreated cellsRapamycin 20 nmol/L

Abs

orba

nce,

490

nm

0 0.1 1 10 100 1000

B4

3

2

1

0

Rapamycin, nmol/L

Total LDHSupernatant LDH

Cel

l via

bilit

y, %

of

untr

eate

d ce

lls 100

80

60

40

20

00 0.01 0.1 1 10 100 1000

A

Rapamycin, nmol/L

*********

Fig. 1. Inhibition of human renal epithelial cell proliferation by ra-pamycin. (A and B) Serum-starved HRECs, seeded in 96-well plates,were stimulated with complete medium containing 10% FCS in the pres-ence of rapamycin concentrations ranging from 0 to 1000 nmol/L. Afterincubation at 37◦C for six days, the relative number of living cells wasdetermined by the MTS assay in (A), and cytotoxicity was quantifiedby LDH release into the culture medium (B). The effect of rapamycinis expressed as % cell viability compared to untreated cells (A) ab-sorbance at 490 nm (B). Results are expressed as mean ± SEM of 10and eight independent experiments for (A) and (B), respectively. (C)Serum-starved HRECs were stimulated for six days in medium con-taining various concentrations of FCS, in the presence or absence of20 nmol/L rapamycin. Cell viability was then measured and expressedas% of viable cells compared to the maximum viability observed in un-treated cells. Results are expressed as mean ± SEM of 3 independentexperiments. P values were evaluated vs. untreated cells: ∗P ≤ 0.05; ∗∗P≤ 0.01.

Rapamycin inhibits G1 to S transition in cycling tubularcells

To establish the effect of rapamycin on cycling HRECs,the cell cycle was analyzed by flow cytometry, usingBrdU/7-AAD staining. Whereas after 36 hours of star-vation in serum-free medium only 17.6 ± 4.0% of cellswere in the S/G2/M phases, stimulation with 20% FCS for

2426 Pallet et al: Antiproliferative effect of rapamycin on human renal epithelial cells

100

80

60

%

40

20

0Starved

cellsFCS + rapamycin

20 nmol/LFCS

S/G2/MG0/G1

§ #

§

§

§#

Fig. 2. Rapamycin inhibits G1 to S transition in cycling human renalepithelial cells. Serum-starved HRECs were stimulated with completemedium containing 20% FCS, in the presence or absence of 20 nmol/Lrapamycin for 18 hours. Cell cycle was then analyzed after BrdU label-ing as described in Methods (mean ± SEM of five independent experi-ments). #P < 0.05 compared to FCS treated cells. §P < 0.05 comparedto starved cells.

18 hours caused significant cell entry into the cycle, with34 ± 3.7% of cells in the S/G2/M phases (P = 0.0003).When 20 nmol/L rapamycin was added in the culturemedium one hour before stimulation with FCS, the pro-portion of cells that progressed from the G0/G1 to theS/G2/M phase decreased significantly to 25.2 ± 8.0%(P = 0.02) (Fig. 2). Conversely, when rapamycin wasadded, the proportion of cells in the G0/G1 phases wassignificantly higher than in untreated cells (71.2 ± 10.2%vs. 60.9 ± 7.0%, P < 0.02). These results demonstrate thatrapamycin induces G1 phase arrest in cycling HRECs.

Rapamycin inhibits p70S6 kinase phosphorylation in nor-mal human epithelial cells

Rapamycin treatment of mouse proximal tubular cellswas previously found to cause virtual abrogation of p70S6k

[21]. The activation of p70S6k by various mitogenic stim-uli such as growth factors or nutrients requires phos-phorylation at several serine and threonine residues. Todate, p70S6k activation has not been described in HRECs.We therefore studied the phosphorylation of p70S6k attwo different sites in HRECs stimulated with 20% FCS,with or without 20 nmol/L rapamycin, using Abs spe-cific for p70S6k phosphorylated at threonine 389 and thre-onine 421/serine 424. Figure 3A shows the kinetics ofp70S6k phosphorylation after mitogen stimulation. Lowlevels of phosphorylation were observed in quiescentcells. After stimulation, a rapid increase in phospholy-lated p70S6k is observed at 60 minutes. Rapamycin addedbefore stimulation totally inhibited p70S6k phosphoryla-tion at all the time points measured (Fig. 3A). As shown inFigure 3B, rapamycin concentrations as low as 1 nmol/Linhibit the p70S6k phosphorylation in serum-stimulatedHRECs.

B

00 0.01 0.1 1 10 100

A

Rapamycin(nmol/L)

Rapamycin 20 nmol/L

Time (hours) 0 1 3 12

FCS

Phospho-p70S6k

(Thr 389)Phospho-p70S6k

(Thr 421/Ser 424)

Total p70S6k

Actin

Phospho-p70S6k

(Thr 389)Phospho-p70S6k

(Thr 421/Ser 424)

Total p70S6k

Actin

– + + + + + +

– – – –+ + +

Fig. 3. Rapamycin inhibits p70S6 kinase phosphorylation in normalhuman epithelial cells. Serum-starved HRECs were stimulated withcomplete medium containing 20% FCS, in the presence or absence ofrapamycin. The expression of phosphorylated p70S6k at two differentphosphorylation sites, and of total p70S6k, was evaluated by Westernblot, using specific polyclonal antibodies. Blots are representative of atleast three independent experiments, performed with cells from threedifferent kidneys. Cell extracts were prepared after serum stimulationwith or without 20 nmol/L rapamycin at the times indicated (A), orstimulated with 20% FCS for three hours in the presence of variousconcentrations of rapamycin (B), and then subjected to Western blotanalysis.

Rapamycin inhibits cyclin D3 protein expression inproliferating tubular cells

The entry of quiescent cells (G0) into early G1 requiresthe expression of D1, D2, and D3 cyclins. The level of cy-clin E is increased in late G1, and is associated with cdk-2,thereby playing a pivotal role in G1/S progression [28].To determine whether rapamycin affects the expressionof cyclin D and/or cyclin E, total HRECs lysates weresubjected to Western blotting after stimulation with 20%FCS with or without 20 nmol/L rapamycin. Rapamycindramatically reduced cyclin D3 expression in HRECs(Fig. 4A and D), and almost completely inhibited theFCS-induced increase in cyclin D3 expression. It may alsoinduce smaller inhibition of cyclin D1 expression, but didnot significantly affect the expression of cyclin D2 or cy-clin E (Fig. 4). These results suggest that rapamycin mightcause cell cycle arrest in the early G1 phase by reducingcyclin D protein expression.

p70S6k is a downstream regulator of the phosphatidyli-nositol 3 (PI3) kinase signaling pathway [29]. In orderto explore if the effect of rapamycin on cyclin D3 ex-pression is mediated by a linear pathway, we compared

Pallet et al: Antiproliferative effect of rapamycin on human renal epithelial cells 2427

A

Rapamycin 20 nmol/L

Time (hours) 0 12 24 36

Cyclin D1

Cyclin D2

Cyclin D3

Cyclin E

– – – –+ + +

B

Cyc

lin D

1/ac

tin,

fold 4

3

2

1

00 12 24 36

C

Cyc

lin D

2/ac

tin,

fold 4

3

2

1

00 12 24 36

D

Cyc

lin D

3/ac

tin,

fold 4

3

2

1

00 12 24 36

Time after serum stimulation, hours

E

Cyc

lin E

/act

in,

fold

4

3

2

1

00 12 24 36

Fig. 4. Rapamycin prevents the serum-stimulated increase in cyclin D proteinsduring G1 progression. HRECs were de-pleted of growth factors for 36 hours. Theywere then stimulated with complete mediumcontaining 20% FCS, in the presence orabsence of 20 nmol/L rapamycin. Cell extractswere prepared at the times indicated, andsubjected to SDS-PAGE. The level of proteinexpression was visualized by Western blot.(A) Representative Western blots of at leastfour independent experiments, performedwith cells from three different kidneys. (B, C,D, E) Cyclin levels were normalized againstactin (mean ± SEM of five independentexperiments). Cells were treated with vehicle(black bars), or rapamycin (open bars). ∗P <

0.05.

the consequences on cyclin D3 expression of two specificPI3-kinase inhibitors, LY294002 and wortmannin, andof a treatment by rapamycin. LY294002 and rapamycincompletely prevented the FCS-induced increase in cyclinD3 expression (Fig. 5). These data were considered PI3kinase-specific, as these results were reproduced by wort-mannin, which suppress in a dose-dependent manner theFCS-induced cyclin D3 expression (Fig. 5). This similar ef-fect of LY294002, wortmannin, and rapamycin suggesteda PI3 kinase/mTOR/p70S6k pathway-dependent regula-tion of cyclin D3 expression.

Effect of rapamycin on cyclin-dependent kinase inhibitorexpression in proliferating tubular cells

Cyclin-cdk complexes are negatively regulated by cdkinhibitors, which bind to and inhibit specific cyclin-cdkcomplexes. Some of these inhibitors, such as p16INK4,p21Cip1, and p27Kip1, are involved in the repression of G1

phase progression in many cell types. We found that 20

nmol/L rapamycin did not affect the expression of thesethree inhibitors in proliferating HRECs (Fig. 6).

Effect of rapamycin on cyclin D3 mRNA in proliferatingepithelial cells

As cyclin D3 was the main cyclin affected by rapamycintreatment, we then focused the study on this protein. Ithas been previously shown in other cell types that ra-pamycin may affect D-type cyclins expression by differ-ent mechanisms [5]. It has been shown in other modelsthat rapamycin treatment not only delays the accumu-lation of cyclin D1 mRNA during progression throughG1 in NIH 3T3 cells, but also affect stability of the tran-script and accelerates the degradation of cyclin D pro-tein [5]. Therefore, we first analyzed the transcriptionalactivation of cyclin D3 mRNA. As shown in Figure 7A,serum increased cyclin D3 mRNA levels, and treatmentof cells with rapamycin dramatically reduced the cy-clin D3 mRNA accumulation. Noteworthy, 5 hours after

2428 Pallet et al: Antiproliferative effect of rapamycin on human renal epithelial cells

300

200

100

0

Cyc

lin D

3/ac

tin,%

Cyclin D3

Actin

FCS 20%

Rapamycin (nmol/L)

LY294002 (µmol/L)

Wortmannin (nmol/L)

– + + + + + + + +– – 20 – – – – – –

– – – 0.5 5 50 – – –

– – – – – – 0.5 5 50

Fig. 5. The inhibition by rapamycin of the serum-induced cyclin D3protein is mediated by the PI3 kinase/mTOR/p70S6k pathway. Serum-starved HRECs were stimulated with complete medium containing20% FCS, in the presence or absence of 20 nmol/L rapamycin, or thetwo PI3-kinase inhibitors LY294002 and wortmannin at the indicatedconcentrations for 24 hours. Cell extracts were then prepared and sub-jected to SDS-PAGE. The level of protein expression was visualized byWestern blot (mean ± SEM of five independent experiments).

Time (hours)

Rapamycin 20

nmol/Lp16

p21

p27

– – + – + – +

120 24 36

Fig. 6. Rapamycin does not affect cyclin-dependent kinase inhibitorexpression in HRECs. HRECs were depleted of growth factors for36 hours. They were then stimulated with complete medium contain-ing 20% FCS, in the presence or absence of 20 nmol/L rapamycin.Cell extracts were prepared at the times indicated, and subjected toSDS-PAGE. The level of protein expression was visualized by West-ern blot. These Western blots are representative of five independentexperiments, performed with cells from three different kidneys.

stimulation, cyclin D3 mRNA levels increased similarlyin untreated and rapamycin-treated cells. This delayed ef-fect of rapamycin suggests that this was not simply due toa transcriptional inhibition. To distinguish between ef-fects of rapamycin on transcription and on the stabil-ity of the mRNA, quiescent HRECs were stimulatedas described above, and treated either in the presenceor the absence of the specific RNA polymerase II in-hibitor DRB. The results obtained indicate that the sta-bility of cyclin D3 mRNA is clearly affected by rapamycin(Fig. 7B). Rapamycin reduced the cyclin D3 mRNA half-life from more than 26% (4.8 ± 1.3 hours vs. 6.5 ±1.0 hours, P < 0.001), suggesting an accelerated mRNAdegradation.

Effect of rapamycin on the transcriptional activity of thehuman cyclin D3 gene

To explore the hypothesis of a transcriptional natureof the control of the cyclin D3 gene expression by ra-pamycin, we investigated the potential of its upstreamsequences to induce transcription in the HRECs in thepresence or absence of rapamycin. We used a plasmidcontaining 1023 bp of the 5′-flanking region of the humancyclin D3 gene (from −1017 to +6 relative to the ATGcodon at +1), cloned in front of a luciferase marker gene[24], and analyzed using classic transient transfection as-says. This upstream sequence of 1023 bp contains all ofthe positive elements required for maximal transcription,which are localized between nucleotides –366 and –167[30]. As the transcription factor E2F1 is a well-recognizedtransactivator of the cyclin D3 gene, an expression plas-mid containing the E2F1 cDNA (or the correspondingempty vector) was cotransfected as an internal positivecontrol. As expected from previous study, E2F1 [24] in-duced the cyclin D3 transcriptional activity (Fig. 8). More-over, increasing doses of rapamycin from 0 to 100 nmol/Ldo not alter the effects of serum-stimulation on cyclin D3transcriptional activity.

Effect of rapamycin on cyclin D3 protein stability incycling epithelial cells

As rapamycin has been suggested to accelerate thedegradation of cyclin D1 protein by the proteasome [5],we studied the effect of rapamycin on cyclin D3 proteinhalf-life in HRECs. Quiescent HRECs were stimulatedas described above, and treated either in the presence orthe absence of the translation inhibitor cycloheximide.Cyclin D3 protein level was evaluated by Western blot.As shown in Figure 9, the protein half-life was about 1.6hours and was not affected by rapamycin.

DISCUSSION

As far as we know, this work is the first to investigatethe antiproliferative effect of rapamycin on the growthof normal HRECs in response to mitogenic stimuli, andof its action on the signaling pathways that convey ex-ternal growth signals to the machinery regulating the cellcycle. We show here that, in HRECs: (1) rapamycin in-hibits the proliferative response to mitogenic stimuli andcauses cell cycle arrest in the G1 phase; (2) this inhibitoryeffect is associated with a specific inhibition of the cyclinD3 expression; (3) this cyclin D3 expression inhibition ispartially independent of the inhibition of the p70S6k path-way, but is secondary to a direct effect on cyclin D3 bydestabilizing its mRNA.

In a nonhuman cellular model, Lieberthal et al demon-strated that rapamycin impaired recovery from acute re-nal failure by blocking mouse tubular cell proliferation,

Pallet et al: Antiproliferative effect of rapamycin on human renal epithelial cells 2429

3

2

1

00 5 10 2015

Fol

d, %

of c

ontr

ol

Cyc

lin D

3 m

RN

A

Cyclin D3 mRNAHalf-life (h)

Untreatedcells

Rapamycin 20nmol/L P

6.5 ± 1.0 4.8 ± 1.3 < 0.001

1.00

0.10

0.01

B

0 6 12 18 24Time after DRB addition, hours

Untreated cells

Rapamycin 20 nmol/L

Time after FCS stimulation, hours

A Untreated cells Rapamycin 20 nmol/L

*

* *

Fig. 7. Effect of rapamycin on cyclin D3mRNA accumulation and degradation in cy-cling human renal epithelial cells. HRECswere depleted of growth factors for 36 hours.They were then stimulated with completemedium containing 20% FCS, in the pres-ence or absence of 20 nmol/L rapamycin.Total RNA was prepared and cDNA weresynthetized for quantitative PCR analysis asdescribed in Methods. (A) Quantitation ofcyclin D3 mRNA levels. mRNA levels werestandardized against the mRNA amount be-fore stimulation (mean ± SEM of threeindependent experiments from three differ-ent kidneys). (B) Effect of rapamycin oncyclin D3 mRNA half-life. 5,6-dichloro-1b-D-ribofuranosylbenzimidazole (DRB) (100lmol/L) was added three hours’ post-stimulation. Total RNA was then extracted atthe indicated time points for quantitative PCRanalysis. RNA levels are expressed as relativeto the level of expression before BRD addi-tion (mean ± SEM of three independent ex-periment from three different kidneys). ∗P <

0.05.

an effect that seems to be partly mediated by the inhibi-tion of p70S6k [27], whose activity is essential for G1 cycleprogression [31]. We demonstrated here that rapamycininhibits both p70S6k activation and cell proliferation instimulated HRECs. Our findings suggest that p70S6k orone of its downstream molecules plays a pivotal role incell cycle progression at the G0/G1 transition. Althoughthe role of rapamycin in this progression has been mostclearly established in mitogen-stimulated lymphocytes,other human cell types are also potential targets for itsantiproliferative effects, including fibroblasts, endothelialcells, smooth muscle cells, hepatocytes, and a broad rangeof human tumor cell lines [5, 27, 32–36].

Rapamycin can block the G1 to S-phase transition ofthe cell cycle. In HRECs, we demonstrate that rapamycin-induced cell cycle arrest in the G1 phase is associated

with a dramatic reduction of cyclin D3 expression, andof cyclin D1, to a smaller extent. This reduction wasnot secondary to a general down-regulation of the pro-teins involved in the G0/G1 to S-phase transition, be-cause we could not detect any change in other cyclinsand cdk inhibitors expression. This regulation of cyclinD3 by rapamycin seems to be tissue-specific becauserapamycin was found to be responsible for cyclin D3down-regulation in a breast cancer cell line [37], inprimary cultures of spermatogonia [38], and in cyclingLLC-B cells [6], but did not affect cyclin D3 expressionin activated T cells [39]. We also show here that rapamycinhas no effect on the cdk inhibitor p27Kip1 in HRECs. Thisresult agrees with the results reported for other types ofepithelial cells [35, 40], whereas rapamycin was shown toup-regulate p27Kip1 expression in T cells [6, 7]. Although

2430 Pallet et al: Antiproliferative effect of rapamycin on human renal epithelial cells

400

300

200

100

0

**

pCMV-E2F1

pCMV nul

pGL3-hCyclinD3

FCS 10%

Rapamycin

+ – – – –

– + + + ++ + + + +

+ +

– –

+ + +

1nmol/L

10nmol/L

100nmol/L

hcyclinD3 (−1017/+6) bp Luc

Fig. 8. Rapamycin has no effect on the transcriptional activity of thehuman cyclin D3 gene. HRECs were transiently transfected with a plas-mid containing 1023 bp of the 5′-flanking region of the human cyclinD3 gene (see Methods). Twelve hours later, cell cultures were serum-stimulated in the presence of various amounts of rapamycin as indicated,or with the solvent vehicle alone (ethanol 0.1% v/v) for 24 hours. Lu-ciferase activity was assayed after 24 hours of serum stimulation. Theresults were expressed relative to the luciferase activity measured inthe serum-stimulated sample (second bar), which was set arbitrarily at100%. Results are expressed as luciferase activity, standardized againstthe activity in cells treated with the rapamycin vehicle (mean ± SEM offive independent experiment from five different kidneys). ∗∗P = 0.01.

all cells use the same repertoire of cyclins and CKIs, theyapparently regulate them differently, in a cell-type andstimulus-specific manner [41].

We tried to determine the mechanism by which ra-pamycin repress cyclin D3 in HRECs. We showed firstthat the two unrelated specific PI 3-kinase inhibitorsLY294002 and wortmannin, on the one hand, and ra-pamycin, on the other, inhibit similarly the serum-induced cyclin D3 expression, suggesting that the effectof rapamycin on p70S6k phosphorylation and cyclin D3expression is mediated by the PI3-kinase/mTOR/p70S6k

pathway.It has been previously reported in other cell types that

rapamycin may affect D-type cyclins expression by delay-ing the accumulation of their mRNA during progressionthrough G1, but also by affecting stability of the transcriptand of the protein [5]. Unlike members of the cyclin fam-ily, cyclins D1, D2, and D3 have a cell- and tissue-specificpattern of expression and regulation [30, 42]. Since theregulation of D-type cyclins is unknown in renal epithe-lial cells, we studied the effect of rapamycin on the cyclinD3 gene transcriptional activity, cyclin D3 mRNA accu-mulation and stability, and cyclin D3 protein turnover.

Serum stimulation led to a rapid increase in cyclin D3mRNA and protein [30]. Rapamycin also led to a de-crease in cyclin D3 mRNA and protein accumulation.This effect may be explained by an increased cyclin D3mRNA turnover since rapamycin reduced by more than25% the cyclin D3 mRNA half-life. Recent data have fo-cused the role of TOR in the regulation of numerous geneexpressions at a post-transcriptional level in both yeastand humans [43–45]. Indeed, the TOR pathway has beenshown to regulate mRNA turnover in the yeast Saccha-romyces cerevisiae [43]. In human, rapamycin has beenshown to destabilize interleukin-3 (IL-3) mRNA in a mastcell tumor line [44]. This effect has been attributed to thepresence in the 3′untranslated region (3′UTR) of the IL-3transcript of an AU-rich element (ARE), IL-3 transcriptslacking the AREs in the 3′UTR being resistant to ra-pamycin treatment [44]. ARE are cis elements containedwithin the 3′UTR of many short-lived mRNAs and con-stitute repetitive AU motifs, which target the transcriptfor rapid deadenylation and degradation [46]. Interest-ingly, we found that the cyclin D3 mRNA carries in its3′UTR two canonical AUUUA motifs. The destabilizingeffect of rapamycin on the cyclin D3 transcript is thenlikely to be due to the presence of these AREs. Furtherwork will be required to confirm this hypothesis.

Serum stimulation led to a very similar increase of cy-clin D3 protein and mRNA levels These data suggestthat serum induction of cyclin D3 expression does notresult only from enhanced translation of its mRNA inHRECs, as previously proposed in other cell type [37].Moreover, this parallelism between cyclin D3 mRNA andprotein regulation by rapamycin rules out the possibilityof an inhibition of the cyclin D3 mRNA translation byrapamycin, as previously mentioned for an essential classof mRNAs that contain an oligopyrimidine tract at theirtranscriptional start (5′TOP), most notably mRNAs en-coding ribosomal proteins and elongation factors [47].

Finally, using a reporter construct, we showed that ra-pamycin had no effect on the transcriptional activity ofthe cyclin D3 gene. Taken together, our results suggestthat the decrease by rapamycin of the cyclin D3 mRNAand protein levels in serum-stimulated HRECs is due pri-marily to post-transcriptional events such as stabilizationof the mRNA. However, we cannot formally rule out thepossibility that sequences outside of the 1023 bp constructtested are required for transcriptional induction of the cy-clin D3 gene.

In summary, We have shown that rapamycin inhibitsHRECs proliferation, and we have explored possiblemechanisms. These effects might be clinically relevantbecause tubular lesions leading to DGF constitute a com-mon complication after renal transplantation, and therate of recovery of tubular function depends on the abil-ity of injured cells to enter the cell cycle and proliferate[12]. Delayed graft function complicates post-transplant

Pallet et al: Antiproliferative effect of rapamycin on human renal epithelial cells 2431

1.00

0.10

0.010 1 2 3 4

Time after cycloheximide addition, hours

Cyc

lin D

3/ac

tin, %

Untreated cellsRapamycin 20 nmol/L

B

ACycloheximide (hours)

Repamycin 20 nmol/L

Cyclin D3

Actin

0 0.5 1 2 3 4 0 0.5 1 2 3 4– – – – – – + + + + + +

Fig. 9. Effect of rapamycin on cyclin D3 protein half-life. (A) Serum-starved HRECs were incubated in the absence or presence of rapamycin, asdescribed in the legend of Figure 2. Cycloheximide (CHX, 20 lg/mL) was added 12 hours post-stimulation. Cell extracts were then prepared at thetimes indicated, and subjected to SDS-PAGE. The level of protein expression was visualized by Western blot. (B) Denstitometric quantitation ofcyclin D3 protein level over time (mean ± SEM of three independent experiments, performed with cells from three different kidneys).

management, and increases both the duration of initialhospital stay and the cost of hospitalization [48]. Al-though conflicting conclusions have been reported aboutthe impact of DGF on the long-term outcome of renaltransplantation [49, 50], increasing evidence suggests thatDGF might influence long-term graft and patient sur-vival [51, 52]. Recovery from ischemic acute renal fail-ure requires the stimulation of tubular epithelial cell pro-liferation by several cytokines and growth factors [12].Agents that impair the ability of the renal epithelium toproliferate may induce prolonged periods of acute re-nal failure, or the failure of recovery [53]. In preclinicalstudies, no substantial nephrotoxicity has been demon-strated for rapamycin, and clinical trials in transplant pa-tients also failed to show that it has any adverse effectson the kidney, except for the exacerbation of renal in-jury, when rapamycin is used together with calcineurin in-hibitors [54]. However, recent clinical observations haveshown enhanced injury and delayed graft function recov-ery when rapamycin is given at the time of transplanta-tion [9, 10]. This apparent discrepancy suggests that ra-pamycin only has harmful effects on tubular epithelialcells when cell regeneration is needed to repair ischemicinjuries.

We demonstrate here that rapamycin has a significantantiproliferative effect on human renal epithelial cells invitro, at concentrations of 0.1 nmol/L corresponding to

less than 0.1 ng/mL, which is below the usual blood con-centrations observed in clinical pratice.

CONCLUSION

The present work is the first to demonstrate this an-tiproliferative effect of rapamycin on cycling HRECs invitro. Rapamycin inhibits the proliferation of HRECs byinducing cell cycle arrest in the G0/G1 phase. It inhibitsp70S6k activation in these cells, thereby blocking intracel-lular signaling that leads to cell multiplication in responseto mitogenic factors. Moreover, unexpectedly, this an-tiproliferative effect is not only mediated by a nonspecificinhibition of the p70S6k pathway, but also by a direct ef-fect of rapamycin on cyclin D3, a critical D-type cyclin forthe cell cycle progression into early G1, by a destabiliza-tion of its transcript. This antiproliferative effect, whichis important for clinical effect on lymphocytes, may alsoexplain the prolongation of delayed graft function afterrenal transplantation.

ACKNOWLEDGMENTS

Dany Anglicheau was awarded a grant by the Groupe Cooperatifde Transplantation d’Ile-de-France. This work was supported by theDelegation a la Recherche Clinique-Hopitaux de Paris, and by the In-stitut National de la Sante et de la Recherche Medicale. Nicolas Palletwas awarded a grant by the Societe Francaise de Nephrologie. These

2432 Pallet et al: Antiproliferative effect of rapamycin on human renal epithelial cells

funding sources were not involved in study design, in the collection,analysis, and interpretation of data, or in the writing of this report. Renalspecimens were obtained from the Urology department of the HopitalSaint-Louis, with the help of Profs. Teillac and Desgrandchamps, andDrs. Mongiat-Artus and Meria.

Reprint requests to Dany Anglicheau, Unite INSERM UMR S490,Centre Universitaire des Saints-Peres, 45, rue des Saints-Peres, 75006Paris, France.E-mail: Dany.Anglicheau@univ-paris5.fr

REFERENCES

1. JOHNSON RW, KREIS H, OBERBAUER R, et al: Sirolimus allows earlycyclosporine withdrawal in renal transplantation resulting in im-proved renal function and lower blood pressure. Transplantation72:777–786, 2001

2. SEHGAL SN: Sirolimus: Its discovery, biological properties, andmechanism of action. Transplant Proc 35:S7–S14, 2003

3. SONENBERG N, GINGRAS AC: The mRNA 5′ cap-binding proteineIF4E and control of cell growth. Curr Opin Cell Biol 10:268–275,1998

4. SCHMELZLE T, HALL MN: TOR, a central controller of cell growth.Cell 103:253–262, 2000

5. HASHEMOLHOSSEINI S, NAGAMINE Y, MORLEY SJ, et al: Rapamycininhibition of the G1 to S transition is mediated by effects on cyclinD1 mRNA and protein stability. J Biol Chem 273:14424–14429, 1998

6. DECKER T, HIPP S, RINGSHAUSEN I, et al: Rapamycin-induced G1arrest in cycling B-CLL cells is associated with reduced expressionof cyclin D3, cyclin E, cyclin A, and survivin. Blood 101:278–285,2003

7. KAWAMATA S, SAKAIDA H, HORI T, et al: The upregulation ofp27Kip1 by rapamycin results in G1 arrest in exponentially growingT-cell lines. Blood 91:561–569, 1998

8. GEDDES CC, WOO YM, JARDINE AG: The impact of delayed graftfunction on the long-term outcome of renal transplantation. JNephrol 15:17–21, 2002

9. SMITH KD, WRENSHALL LE, NICOSIA RF, et al: Delayed graft functionand cast nephropathy associated with tacrolimus plus rapamycinuse. J Am Soc Nephrol 14:1037–1045, 2003

10. MCTAGGART RA, GOTTLIEB D, BROOKS J, et al: Sirolimus prolongsrecovery from delayed graft function after cadaveric renal trans-plantation. Am J Transplant 3:416–423, 2003

11. DAVIS C: Sirolimus delays renal allograft recovery. Am J Transplant3:363–365, 2003

12. HUMES HD, MACKAY SM, FUNKE AJ, et al: Acute renal failure:Growth factors, cell therapy, and gene therapy. Proc Assoc AmPhysicians 109:547–557, 1997

13. WITZGALL R, BROWN D, SCHWARZ C, BONVENTRE JV: Localizationof proliferating cell nuclear antigen, vimentin, c-Fos, and clusterinin the postischemic kidney. Evidence for a heterogenous geneticresponse among nephron segments, and a large pool of mitoti-cally active and dedifferentiated cells. J Clin Invest 93:2175–2188,1994

14. TANG S, LEUNG JC, ABE K, et al: Albumin stimulates interleukin-8expression in proximal tubular epithelial cells in vitro and in vivo.J Clin Invest 111:515–527, 2003

15. DETRISAC CJ, SENS MA, GARVIN AJ, et al: Tissue culture of humankidney epithelial cells of proximal tubule origin. Kidney Int 25:383–390, 1984

16. JOLY D, MOREL V, HUMMEL A, et al: Beta4 integrin and laminin 5 areaberrantly expressed in polycystic kidney disease: Role in increasedcell adhesion and migration. Am J Pathol 163:1791–1800, 2003

17. HELBERT MF, DAUWE SE, DE BROE ME: Flow cytometric immun-odissection of the human distal tubule and cortical collecting ductsystem. Kidney Int 59:554–564, 2001

18. BAKKER RC, VAN KOOTEN C, VAN DE LAGEMAAT-PAAPE ME, et al:Renal tubular epithelial cell death and cyclosporin A. Nephrol DialTransplant 17:1181–1188, 2002

19. LEIS M, MARSCHALL M, STAMMINGER T: Downregulation of the cel-lular adhesion molecule Thy-1 (CD90) by cytomegalovirus infectionof human fibroblasts. J Gen Virol 85:1995–2000, 2004

20. PHILLIPS AO, STEADMAN R, TOPLEY N, WILLIAMS JD: Elevated D-glucose concentrations modulate TGF-beta 1 synthesis by humancultured renal proximal tubular cells. The permissive role of platelet-derived growth factor. Am J Pathol 147:362–374, 1995

21. ZANDOMENI R, WEINMANN R: Inhibitory effect of 5,6-dichloro-1-beta-D-ribofuranosylbenzimidazole on a protein kinase. J BiolChem 259:14804–14811, 1984

22. GRUMMT I: Regulation of mammalian ribosomal gene transcriptionby RNA polymerase I. Prog Nucleic Acid Res Mol Biol; 62:109–154,1999

23. LIVAK KJ, SCHMITTGEN TD: Analysis of relative gene expression datausing real-time quantitative PCR and the 2−��CT method. Method25:402–408, 2001

24. MA Y, YUAN J, HUANG M, et al: Regulation of the cyclin D3 promoterby E2F1. J Biol Chem 278:16770–16776, 2003

25. BERGAMETTI F, SITTERLIN D, TRANSY C: Turnover of hepatitis B virusX protein is regulated by damaged DNA-binding complex. J Virol76:6495–6501, 2002

26. HELIN K, HARLOW E, FATTAEY A: Inhibition of E2F-1 transactiva-tion by direct binding of the retinoblastoma protein. Mol Cell Biol13:6501–6508, 1993

27. LIEBERTHAL W, FUHRO R, ANDRY CC, et al: Rapamycin impairs re-covery from acute renal failure: Role of cell-cycle arrest and apop-tosis of tubular cells. Am J Physiol Renal Physiol 281:F693–706,2001

28. SHANKLAND SJ, WOLF G: Cell cycle regulatory proteins in renal dis-ease: Role in hypertrophy, proliferation, and apoptosis. Am J Phys-iol Renal Physiol 278:F515–529, 2000

29. SHEPHERD PR, NAVE BT, O’RAHILLY S: The role of phosphoinosi-tide 3-kinase in insulin signalling. J Mol Endocrinol 17:175–184,1996

30. BROOKS AR, SHIFFMAN D, CHAN CS, et al: Functional analysis of thehuman cyclin D2 and cyclin D3 promoters. J Biol Chem 271:9090–9099, 1996

31. LANE HA, FERNANDEZ A, LAMB NJ, THOMAS G: p70s6k function isessential for G1 progression. Nature 363:170–172, 1993

32. VINALS F, CHAMBARD JC, POUYSSEGUR J: p70 S6 kinase-mediatedprotein synthesis is a critical step for vascular endothelial cell pro-liferation. J Biol Chem 274:26776–26782, 1999

33. MARX SO, JAYARAMAN T, GO LO, MARKS AR: Rapamycin-FKBPinhibits cell cycle regulators of proliferation in vascular smooth mus-cle cells. Circ Res 76:412–417, 1995

34. NELSEN CJ, RICKHEIM DG, TUCKER MM, et al: Amino acids regulatehepatocyte proliferation through modulation of cyclin D1 expres-sion. J Biol Chem 278:25853–25858, 2003

35. GREWE M, GANSAUGE F, SCHMID RM, et al: Regulation of cell growthand cyclin D1 expression by the constitutively active FRAP-p70s6Kpathway in human pancreatic cancer cells. Cancer Res 59:3581–3587,1999

36. MISAWA A, HOSOI H, TSUCHIYA K, SUGIMOTO T: Rapamycin inhibitsproliferation of human neuroblastoma cells without suppression ofMycN. Int J Cancer 104:233–237, 2003

37. MUISE-HELMERICKS RC, GRIMES HL, BELLACOSA A, et al: Cyclin Dexpression is controlled post-transcriptionally via a phosphatidyli-nositol 3-kinase/Akt-dependent pathway. J Biol Chem 273:29864–29872, 1998

38. FENG LX, RAVINDRANATH N, DYM M: Stem cell factor/c-kit up-regulates cyclin D3 and promotes cell cycle progression via thephosphoinositide 3-kinase/p70 S6 kinase pathway in spermatogo-nia. J Biol Chem 275:25572–25576, 2000

39. TURNER JM: IL-2-dependent induction of G1 cyclins in primary Tcells is not blocked by rapamycin or cyclosporin A. Int immunol5:1199–1209, 1993

40. KENERSON HL, AICHER LD, TRUE LD, YEUNG RS: Activated mam-malian target of rapamycin pathway in the pathogenesis of tuber-ous sclerosis complex renal tumors. Cancer Res 62:5645–5650,2002

41. EKHOLM SV, REED SI: Regulation of G1 cyclin-dependent kinasesin the mammalian cell cycle. Curr Opin Cell Biol 12:676–684, 2000

42. SHERR CJ: Mammalian G1 cyclins. Cell 73:1059–1065, 199343. ALBIG AR, DECKER CJ: The target of rapamycin signaling pathway

regulates mRNA turnover in the yeast Saccharomyces cerevisiae.Mol Biol Cell 12:3428–3438, 2001

Pallet et al: Antiproliferative effect of rapamycin on human renal epithelial cells 2433

44. BANHOLZER R, NAIR AP, HIRSCH HH, et al: Rapamycin destabi-lizes interleukin-3 mRNA in autocrine tumor cells by a mechanismrequiring an intact 3′ untranslated region. Mol Cell Biol 17:3254–3260, 1997

45. GHOSH P, BUCHHOLZ MA, YANO S, et al: Effect of rapamycin onthe cyclosporin A-resistant CD28-mediated costimulatory pathway.Blood 99:4517–4524, 2002

46. CHEN CY, SHYU AB: AU-rich elements: Characterization and im-portance in mRNA degradation. Trends Biochem Sci 20:465–470,1995

47. JEFFERIES HB, FUMAGALLI S, DENNIS PB, et al: Rapamycin suppresses5′TOP mRNA translation through inhibition of p70s6k. EMBO J16:3693–3704, 1997

48. ROSENTHAL JT, DANOVITCH GM, WILKINSON A, ETTENGER RB: Thehigh cost of delayed graft function in cadaveric renal transplanta-tion. Transplantation 51:1115–1118, 1991

49. MARCEN R, OROFINO L, PASCUAL J, et al: Delayed graft function does

not reduce the survival of renal transplant allografts. Transplanta-tion 66:461–466, 1998

50. WOO YM, JARDINE AG, CLARK AF, et al: Early graft function andpatient survival following cadaveric renal transplantation. KidneyInt 55:692–699, 1999

51. BOOM H, MALLAT MJ, DE FIJTER JW, et al: Delayed graft functioninfluences renal function, but not survival. Kidney Int 58:859–866,2000

52. PROMMOOL S, JHANGRI GS, COCKFIELD SM, HALLORAN PF: Timedependency of factors affecting renal allograft survival. J Am SocNephrol 11:565–573, 2000

53. MCCAULEY J: The nephrotoxicity of FK506 as compared with cy-closporine. Curr Opin Nephrol Hypertens 2:662–669, 1993

54. KAHAN BD: Efficacy of sirolimus compared with azathioprine forreduction of acute renal allograft rejection: A randomised multi-centre study. The Rapamune US Study Group. Lancet 356:194–202,2000