Generation of a Mouse Model of Von ... - Cancer Research · Renal cancers are highly aggressive and clinically challenging, but atransgenic mouse model to promote pathologic studies

Tumor and Stem Cell Biology

Generation of a Mouse Model of Von Hippel–Lindau KidneyDisease Leading to Renal Cancers by Expression of aConstitutively Active Mutant of HIF1a

Leiping Fu1,4, Gang Wang1,4, Maria M. Shevchuk2,4, David M. Nanus3,4, and Lorraine J. Gudas1,4

AbstractRenal cancers are highly aggressive and clinically challenging, but a transgenic mouse model to promote

pathologic studies and therapeutic advances has yet to be established. Here, we report the generation of atransgenic mouse model of von Hippel–Lindau (VHL) renal cancer termed the TRACK model (transgenicmodel of cancer of the kidney). TRACK mice specifically express a mutated, constitutively active HIF1a inkidney proximal tubule (PT) cells. Kidney histologies displayed by TRACK mice are highly similar tohistologies seen in patients with VHL disease, including areas of distorted tubular structure, cells withclear cytoplasm and increased glycogen and lipid deposition, multiple renal cysts, and early onset of clear cellrenal cell carcinoma (ccRCC). Distorted tubules in TRACK mice exhibit higher levels of CA-IX, Glut1, andVEGF than tubules in nontransgenic control mice. Furthermore, these tubules exhibit increased numbers ofendothelial cells, increased cell proliferation, and increased expression of the human ccRCC marker CD70(TNFSF7). Moreover, PT cells in kidney tubules from TRACK mice exhibit increased genomic instability, asmonitored by elevated levels of gH2AX. Our findings establish that activated HIF1a in murine kidney PT cellsis sufficient to promote cell proliferation, angiogenesis, genomic instability, and other phenotypic alterationscharacteristic of human VHL kidney disease, establishing the TRACK mouse as a valid preclinical model ofhuman renal cell carcinoma. Cancer Res; 71(21); 6848–56. �2011 AACR.

Introduction

Von Hippel–Lindau (VHL) disease is a hereditary cancersyndrome caused by germline mutations of the VHL tumorsuppressor gene (1). Patients with VHL disease have a greatlyincreased risk of developing various types of tumors, includinghemangioblastomas, clear cell renal cell carcinomas (ccRCC),and pheochromocytomas (1). The lifetime risk of developingccRCC in VHL disease patients is more than 70% by the age of60 years (2). Loss of expression or mutation of the VHL tumorsuppressor gene plays an etiologic role in this cancer syndrome(1). The associated, increased expression of 2 transcriptionfactors, the alpha subunits of hypoxia inducible factor 1(HIF1a) and HIF2a, may be critical for carcinogenesis (3–5).

VHL disease patient kidneys display cystic changes ordistortion of the tubular structure adjacent to cells, which canbe differentiated from normal renal parenchyma by histologic

evaluation (6). In addition to expressing HIF1a, cells in theseearly lesions overexpress HIF1a target genes, such as GlucoseTransporter 1(Glut-1), VEGF, and carbonic anhydrase IX(CA-IX;ref. 6). The majority of these early abnormalities are of the"clear" cell type, and a small percentage show changes innuclear morphology, nuclear to cytoplasm ratio, and tubulararchitecture. Surrounding areas show increased vasculariza-tion. Importantly, HIF1a, but not HIF2a activation occurs inthe earliest lesions (6), suggesting a role for HIF1a activation inearly renal carcinogenesis. However, this hypothesis remainsuntested, as many genes are aberrantly expressed in theabsence of VHL (1).

The HIF1a protein is regulated posttranslationally undernormoxic conditions (7) by interaction with and polyubiqui-tination by an E3 ubiquitin ligase complex containing VHLprotein(pVHL; ref. 8). The polyubiquitinated HIF1a is thentargeted to the 26S proteasome for degradation (8). Theinteraction of HIF1a with pVHL is mediated by hydroxylationof 2 proline residues (P402, 564) in the oxygen-dependentdegradation domain (9, 10). Under hypoxic conditions, theseproline residues are not hydroxylated. HIF1a, therefore, cannot be recognized by pVHL and becomes stabilized. Dimer-ization of the stabilized HIF1a with the constitutively activeHIF1b results in binding of the HIF heterodimer complex tohypoxia responsive elements (HRE, consensus sequence: 50-RCGTG-30; ref. 11). Under normoxic conditions HIF1a is alsohydroxylated at a conserved asparagine residue (N803) byfactor inhibiting HIF1a (FIH-1), whose activity is also oxygen

Authors' Affiliations: Departments of 1Pharmacology, 2Pathology andLaboratory Medicine, 3Division of Hematology and Medical Oncology,Department of Medicine, 4Weill Cornell Cancer Center, Weill Cornell Med-ical College (WCMC) of Cornell University, New York, New York

Corresponding Author: Lorraine J. Gudas, Department of Pharmaco-logy, Weill Cornell Medical College, 1300 York Avenue, New York,NY 10065. Phone: 212-746-6250; Fax: 212-746-8858; E-mail:[email protected]

doi: 10.1158/0008-5472.CAN-11-1745

�2011 American Association for Cancer Research.

CancerResearch

Cancer Res; 71(21) November 1, 20116848

on May 1, 2020. © 2011 American Association for Cancer Research. cancerres.aacrjournals.org Downloaded from

Published OnlineFirst September 9, 2011; DOI: 10.1158/0008-5472.CAN-11-1745

http://cancerres.aacrjournals.org/

dependent (12). The asparaginyl hydroxylation preventsrecruitment of p300 and the transcriptional coactivator pro-tein CBP, resulting in attenuated transcription of HIF1a targetgenes by HIF1a (13). Because the asparaginyl hydroxylation isalso oxygen dependent, under hypoxic conditions HIF1a is notinhibited by FIH-1 and can activate transcription. Mutationof these 3 key amino acids should mimic hypoxic conditionsby preventing HIF1a from being degraded by pVHL and byallowing HIF1a to recruit p300 and CBP, resulting in consti-tutive expression of HIF1a target genes.Few mouse models that exhibit the pertinent features of

human VHL disease in the kidney exist. Recapitulation ofhuman kidney carcinogenesis by inactivation of the humantumor suppressor geneVHL has not been achieved (14–16).Wecreated a triple mutant (P402A, P564A, N803A) human HIF1aconstruct using the kidney proximal tubule-specific type 1g-glutamyl transpeptidase (GGT or gGT) promoter (17, 18) todrive expression of this triple mutant, constitutively activeHIF1a in the proximal tubule cells (PTC). Mice expressing thistriple mutant, constitutively active HIF1a construct in thekidney exhibit "clear" cells, renal cysts, disorganized PTs, andcystic clear cell carcinoma, consistent with ccRCC.

Materials and Methods

Plasmid constructionMutated, constitutively active HIF1a cDNA was created by

site-directed mutagenesis (Invitrogen) of conserved prolineresidues (402, 564) and a conserved asparagine (803) intoalanine residues. GGT is specifically expressed in the PTs ofthe kidney starting at about 3 weeks (17, 18). The rat GGTpromoter (�1,930 to þ246) was amplified by PCR from aplasmid from Dr. Terzi (18). The GGT promoter, mutatedHIF1a, and b-globin poly-A were cloned into pBlueScript andnamed gGT-HIF1a triple mutant (g-HIF1a-M3).

Generation of transgenic miceDNA was prepared as a linearized ClaI-XbaI fragment (vec-

tor sequence removed) and was injected into pronuclei of one-cell embryos (C57BL/6 � C57BL/6) at the WCMC MouseGenetics Core. Southern Blot analysis was then done (19).

Tissue dissection, processing, and pathologic reviewTissues were fixed, processed, sectioned, and hematoxylin

and eosin stained following standard protocols (20). Slideswere reviewed blindly by Dr. Shevchuk, an experienced clinicalpathologist specializing in kidney cancer, and independentlyby a veterinary pathologist, Dr. S�ebastien Monette, from theLaboratory of Comparative Pathology, WCMC.

Reverse transcriptase PCRTotal RNA was extracted using mini-RNAeasy columns

(Qiagen). Reverse transcriptase PCR (RT-PCR) was then done(21).

ImmunostainingImmunohistochemistry and immunofluorescence were

done (22). Antibodies: HIF1a (610958, BD-Transduction);

CA-IX (sc-25600, Santa Cruz); Glut-1 (ab14683, Abcam); VEGF(sc-507, Santa Cruz); Cd-31 (550274, BD Pharmingen); PCNA(M0879, Dako) and gH2AX (9718S, Cell Signaling). Cd-31 wasstained on cryo-preserved, frozen sections. Periodic Acid/Schiff (PAS) stain was done on paraffin embedded (23) andcryo-preserved sections. Oil red O (ORO) staining was alsodone (24).

Statistical analysisResults are expressed as mean � SEM. Student t test was

used to determine the statistical significance of the gH2AXþ

and Ki67þ cell number differences between TGþ and TG�

mice.

Results

Generation of transgenic mice expressing mutated,constitutively active HIF1a

A high level of HIF1a protein is a prominent feature ofccRCC (8). Furthermore, increased expression of HIF1aand/or HIF2a, which has 48% amino acid homology withHIF1a, is thought to be a key event in ccRCC carcinogenesis(3). To examine the role of HIF1a in ccRCC carcinogenesis,we constructed a GGT-HIF1a triple mutant plasmid(g-HIF1a-M3, Fig. 1A). After confirmation of activity in cul-tured normal kidney proximal tubule cells, g-HIF1a-M3transgenic mice were generated from the gHIF1a-M3 plas-mid. Five of 51 founder mice harbored the integrated targetgene (Fig. 1B, founders #8 to #14). Four lines were evaluatedby RT-PCR, using a transgene-specific primer pair (primers1 and 2, Fig. 1A) for kidney, spleen, liver, heart, lung, intestine,skeletal muscle, and testis/ovary mutant HIF1a mRNA. Thetriple mutant HIF1a was expressed only in the kidneys oftransgenic positive (TGþ) lines #8, #25, #32, and #43 (Fig. 1C,panel 1, #43 as an example). The transgene was not expressedin other organs analyzed (Fig. 1D). The specific expression ofHIF1a in kidney proximal tubules (PT) was also confirmed byimmunohistochemistry (see below). VHL, endogenous HIF1a,and endogenous HIF2amRNA levels were not changed in thekidneys of TGþ compared with the kidneys of transgenicnegative (TG�) mice (Fig. 1C, panels 2, 3, 4). All 4 TGþ founderlines, g-HIF1a-M3-8, g-HIF1a-M3-25, g-HIF1a-M3-32, andg-HIF1a-M3-43, developed normally and passed the trans-gene to offspring following a Mendelian inheritance pattern.

g-HIF1a-M3-43 TGþ mice exhibit kidney lesions thathistologically resemble human VHL disease

Kidney, spleen, liver, heart, lung, intestine, skeletal muscle,and testis/ovary were removed from TGþ and TG� litter-mates, aged 6 to 7 months. Areas of distorted tubular struc-ture were identified with clusters of "clear" cells in the outercortex of TGþ kidneys from all 4 TGþ lines. These "clear" cellshave characteristic proximal tubule cell features, like abun-dant, acidophilic cytoplasm (Fig. 2A; Fig. 2B, TG�). Theg-HIF1a-M3-43 line exhibited the strongest phenotype (Fig.2A); the other TGþ lines had similar phenotypes, but theypossessed fewer PTs containing "clear" cells (see below). Thedistorted tubule cells showed moderate to marked cellular

Constitutively Active HIF1a Causes Renal VHL Disease

www.aacrjournals.org Cancer Res; 71(21) November 1, 2011 6849




swelling, cytoplasmic vacuolation (Fig. 2A, black arrows), andprominent cell membranes, a feature of human ccRCC(Fig. 2A). We identified 2 types of vacuolation: large round,discrete vacuoles displacing the nucleus (Fig. 2A, solid blackarrow), and vacuoles with pale, eosinophilic to clear featherycytoplasm without displacement of the nucleus (Fig. 2A,dashed black arrow). The morphology of the large round,discrete vacuoles is consistent with lipid accumulation, whilethe morphology of the feathery cytoplasm is consistent withglycogen accumulation and hydropic degeneration. These"clear" cell clusters were morphologically strikingly similarto early lesions reported to occur in the "normal" kidneys ofpatients with VHL disease (6). Because the GGT promoteronly drives the expression of the transgene in the PTs, we didnot expect abnormalities in the renal medulla or papilla anddid not observe any morphologic abnormalities in theseregions (not shown). Not all PTs contained these "clear" cellclusters; areas of "normal" PTs were present in the cortex, as

assessed by histologic morphology (Fig. 2A, yellow arrow)."Clear" cell PTs covered about 30% to 50% of the cortex in theg-HIF1a-M3-43 transgenic mice (see below). The majority ofhistologically abnormal PT cells in TGþ mice were large,simple cuboidal epithelial cells (Fig. 2A, black arrows) andwere surrounded by tubular basement membrane, suggestingthat these cells are still under proper growth control. Varia-tions in the sizes of the nuclei (anisokaryosis) were identifiedin these "clear" cells. HIF1a protein was overexpressed in"clear" cells of TGþ (Fig. 2C, arrow), but not TG� kidneys (Fig.2D) by immunohistochemistry.

In ccRCC, the clear cytoplasm is caused by deposition ofglycogen, phospholipids, and neutral lipids, particularly cho-lesterol esters (25). Similarly, the "clear" cells observed in ourTGþ mice contained large amounts of cytoplasmic glycogenand lipid, as shown by Periodic Acid/Schiff (PAS; Fig. 2E,arrow; Fig. 2F, TG�) and ORO staining (Fig. 2G, arrow;Fig. 2H, TG�), respectively. In normal PTs, we detected strong

HIF1α

TG HIF1α~300bp

HIF1αβ-Actin

Kid

ney

Sple

en

Liv

er

Heart

Lung

Inte

stin

e

Test

is

Ske

leta

l musc

les

Kid

ney

Sple

en

Liv

er

Heart

Lung

Inte

stin

e

Ova

ry

Ske

leta

l musc

les

Kid

ney

Sple

en

Liv

er

Heart

Lung

Inte

stin

e

Ova

ry

Ske

leta

l musc

les

Endogenous

HIF1α~400bp

Endogenous

HIF2α~300bp

Actin

~450bp

VHL

~350bp

TG+ TG–

TG-

~2.5 Kb

A

Cγ-HIF1α-M3

HIF1αGGT promoter β-Globin poly-A and intron

XbalClal–1,930 to +245 Primer 1 Primer 2

P564AP402A N803A

γ‐HIF1α‐M3-43 γ‐HIF1α‐M3-43

8 9 10 11 12 13 14

+ - - - - - +

D

B

Figure 1. Generation of g-HIF1a-M3transgenic lines. A, construction ofg-HIF1a-M3 plasmid. Threemutations (P402A, P564A, N803A),dashed arrows. Intron (shadedsquare) in b-globin poly-A. Primers1, 2 amplify the transgene by RT-PCR. B, Southern Blot of some TGþ

and TG� Founders. Founders #8and #14, TGþ; others are TG�. C,HIF1a transgene, endogenousVHL, HIF1a, and HIF2a RT-PCRg-HIF1a-M3-43 kidneys. HIF1atransgene, detected only in TGþ.Endogenous VHL, HIF1a, andHIF2a mRNAs are expressed atsimilar levels in TGþ and TG�.b-Actin, loading control. D, HIF1atransgene RT-PCR in multipleorgans of g-HIF1a-M3-43 mice.HIF1a transgene, detectedspecifically in g-HIF1a-M3-43 TGþ

kidneys. No transgene expressionin any organ of TG� mice. b-Actin,loading control.

Fu et al.

Cancer Res; 71(21) November 1, 2011 Cancer Research6850




PAS stain only in the basement membrane and the brushborder/luminal side of the PT cells, andnot in the cytoplasm. Incontrast, in PT cells from TGþ mice strong PAS stain wasidentified in the cytoplasm, consistent with glycogen accumu-lation. Similarly, we detectedORO staining in the cortex of TGþ

(Fig. 2G) but not TG� mice (Fig. 2H).The multiple renal cysts observed (26) in VHL disease

patients are believed to be precursors of ccRCC (1, 27).We identified cysts composed of a single layer of epithelialcells in 12-month TGþ mice (Fig. 3A). Dilated blood vesselsfilled with red blood cells were typically found near cysts.The majority of the cysts had clear spaces within, whilesome cysts contained amorphous, pale, eosinophilic mate-rial (Fig. 3A). Although both tubular and glomerular cystswere found in VHL conditional knockout mice (16), we

observed tubular cysts, but not glomerular cysts, in theg-HIF1a-M3-43 TGþ mice.

The PT cells from 7-month g-HIF1a-M3-43 TGþ mice showsimple, cuboidal epithelial cells with a "clear" morphology. Thenuclei of some "clear" cells were large and hyperchromatic,with conspicuous nucleoli, suggestive of a neoplastic change(Fig. 3B and C, arrow). In 14- to 20-month TGþ mice, weobserved abnormal PTs with multiple layers of epithelial cells(Fig. 3B andD, arrow). The normal proximal tubule structure iscompletely disrupted by these disorganized "clear" cells. Theseabnormal, disorganized "clear" cells form intratubular nests,consisting of multiple layers of cells, and are surroundedby cells with spindle-like nuclei, some of which are endothelialcells (see below), and red blood cells. This has also beenobserved in foci of VHL disease patient kidneys (6). Theseabnormal tubules display an intratubular proliferation ofatypical clear cells, indicating carcinoma in situ. We examined15 male g-HIF1a-M3-43 TGþ mice from 3 to 20 months old.Kidneys of all 15 mice show the expected phenotype. Theseverity of thesehistologic changes increaseswithage (Table 1).Twenty-two-month-old TGþ mice had cystic ccRCC (Fig. 3E).VHL kidney disease features such as "clear" cells, renal cysts,hyperchromatic nuclei, disorganized PTs, and cystic ccRCCwere not observed in TG� mice (Fig. 3F).

ATG+ TG–

B

C D

E F

G H

Figure 2. "Clear" cells in g-HIF1a-M3-43 mice. Representative images ofhistologicmorphology (A andB), HIF1a immunohistochemistry (C andD),PAS stain (E and F), and ORO stain (G and H) "clear" cells in 6 monthg-HIF1a-M3-43male TGþ kidneys (A, C, E, andG) and TG� kidneys (B, D,F, and H). "Clear" cells, TGþ tubule cells (A, solid, dashed arrows). Largeround vacuole displacing the nucleus, solid arrow; feathery cytoplasmwithout nucleus displacement, dashed arrow. "Normal" PTCs in TGþ

kidney (yellow arrow). Increased HIF1a protein in TGþ nuclei (C, arrow).Strong PAS (E, arrow), ORO (G, arrow) staining in cytoplasm of TGþ cells(E and G, arrows). Abnormal PTs with "clear" cells, A, C, E, and G. All PTsstained by PAS or ORO were abnormal PTs with "clear" cells (E and G).Scale bars, 100 mm.

A B

C D

E F

Figure 3. The g-HIF1a-M3-43 mice show morphologic changesconsistent with in situ ccRCC. Representative images of renal cysts(A), a "clear" cell with enlarged nuclei and conspicuous nucleoli (B andC), disorganized PTs (B and D), and cystic ccRCC (E) in g-HIF1a-M3-43 mice. A, two adjacent renal tubular cysts, 1 filled with amorphous,pale, eosinophilic material, 1 empty. These structures are notobserved in TG� mice (Fig. 2B and 3F). B, low magnification of cortexshowing a "clear" cell with enlarged nuclei and one conspicuousnucleolus (C, high magnification) and several TGþ PTs withdisorganized, multilayer "clear" cells (D, high magnification). E, cysticccRCC. F, TG� control. Scale bars, 100 mm.






g-HIF1a-M3-43 TGþ murine kidneys molecularlyresemble human VHL kidney disease

In ccRCC there is increased vascularization surroundingtumor cells (28). We detected increased CD31 staining sur-rounding the morphologically abnormal PTs in TGþ (Fig. 4A,arrow) as compared with TG�mice. In TG�mice, we detectedmany CD31þ endothelial cells (29) in glomeruli (Fig. 4B), butnot in blood vessels surrounding the PTs (Fig. 4B).

CA-IX (NP_647466), Glut-1 (NP_035530), and VEGF(NP_001020421) are HIF1a target genes that show increasedprotein expression in human ccRCC (30). Using immunohis-tochemistry (Fig. 4C–H),we observed strongCA-IX, Glut-1, andVEGF signals in the abnormal, but not in the morphologicallynormal, PTs of the same TGþ mice (Fig. 4C, E, G, TGþ),however, we saw weak or no CA-IX, Glut-1, and VEGF signalsin the PTs of TG� mice (Fig. 4D, F, H, TG�). The "clear" cells inthe abnormal PTs were identified by the clear cytoplasm withvacuoles (Fig. 4C, E, G).

CA-IX is the major marker used to diagnose ccRCC. CA-IXprotein expression increases in ccRCC carcinogenesis (6) andCA-IX protein is highly expressed in ccRCC (31). High CA-IXprotein expression is a diagnostic and prognostic marker forhuman ccRCC (32, 33). We detected greatly increased CA-IXexpression in the basal and lateral membranes of the abnor-mal PTCs in the g-HIF1a-M3-43 TGþ kidneys, but CA-IX wasnot easily identified in the luminalmembranes/brush borders(Fig. 4A, arrow). The staining pattern was membranous andcup-like. The cup-like Glut-1 and CA-IX staining patternssuggest that the brush borders of the abnormal PTCs areaberrant. By Western analysis, the CA-IX protein is elevated30- to 40-fold in the TGþ kidneys relative to TG� littermates

(Fig. 4I). CA-IX is a direct target of HIF1a, and CA-IXmRNA ishighly expressed in the kidneys of TGþ relative to TG�

littermates (Fig. 4J) but not in other tissues in TGþ mice(not shown). We also detected a much higher CD70 mRNAlevel in kidneys of TGþ versus TG� mice (Fig. 4J). CD70(TNFSF7) is a marker of human ccRCC (34).

We detected increased Glut-1 in the basal and lateralmembranes of the "clear" cells of abnormal PTs (Fig. 4C), butnot in the luminal membranes/brush borders in the g-HIF1a-M3-43 TGþ mice. We also observed a stronger Glut-1 signal insome membranes surrounding the clear vacuoles (Fig. 4C,arrow). The "normal" PTs in TGþ kidneys, which are similarin morphology to normal PTs in TG� kidneys, show muchweaker Glut-1 staining (Fig. 4D).

We detected increased VEGF staining, assessed by immu-nohistochemistry, in the cytoplasm of the "clear" cells in theabnormal PTs (Fig. 4E) in the g-HIF1a-M3-43 TGþ mice. Insome "clear" cells we detect intense VEGF staining close toempty vacuoles (Fig. 4E, arrow). The morphologically "normal"PT cells in TGþmice exhibit lowerVEGF staining, similar to thenormal PTs in TG� mice (Fig. 4F).

g-HIF1a-M3-43 TGþ mice express Ki67 and gH2AX,molecular markers of carcinogenesis

Proliferating Cell Nuclear Antigen (PCNA) is an essentialcomponent of DNA polymerase d (35) and is expressed duringS phase of the cell cycle; Ki67 is a marker for proliferation (36).We detected increased numbers of PCNAþ or Ki67þ cells in themorphologically abnormal PTs of TGþ mice (Fig. 5A, arrows),but not in the "normal" PTs. In TG� mice, we saw few PCNAþ

(not shown) or Ki67þ cells in the PTs (Fig. 5B, TG�).

Table 1. Summary of male TGþ mice showing clear cell PTs, renal cysts, or carcinoma in situ

Animal ID Age at sacrifice(mo)

Clear cellPT area (%)

Number ofhyperchromaticnuclei/300� field

Renal cysts(/section)

Carcinoma in situ(/section)

43-102 3 40.4 24.4 0 043 � 17-2 3 31.6 8.6 0 043-143 4 31.6 7.4 0 043-9 6 36.4 11.4 0 043-93 7 30.9 16 0 043-12 12 35.0 1.8 1 043-24 13 38.2 4.8 0 043-13 14 43.0 6.6 1 643 Founder 14 44.4 4.4 26 543-72 14 46.1 7.2 1 843-27 18 27.9 1.8 2 743-10 19 39.1 5.6 3 1343-21 20 46.5 6 6 1743-20 22 45.5 19.8 8 1943-22 22 52.4 22.2 5 18

NOTE: Results from15 TGþ and 15 TG�mice. A summary of TGþmice that exhibit clear cell PTs, hyperchromatic nuclei, renal cysts, orcarcinoma in situ. These abnormalities are not seen in male TG� littermates. Results from TG� littermates are not included.

Fu et al.





Genomic instability is another universal feature of tumorcells (37). Premalignant and malignant cells accumulate moremutations than normal cells (37). An increased DNA mutationrate/genomic instability may facilitate neoplastic transforma-tion (38). Because the DNA mutation rate is difficult tomeasure, we measured the numbers of DNA double-strandbreaks (DSB) in these "clear" cells. The serine(139) phosphor-ylated form of H2A histone family, member X (gH2AX) is awidely usedmarker for DSBs (39). Using a gH2AX antibody, we

showed that TGþmice show a greater number of gH2AXþ cellsin the kidneys, especially in the regions of "clear" cells (Fig. 5C,arrow) than TG�mice (Fig. 5D). Almost all of the gH2AXþ cellsin the TGþ mice were in clusters of "clear" cells in morpho-logically abnormal PTs (Fig. 5C). We detected 6 Ki67þ cells and8 gH2AXþ cells in the TGþmice compared with 1.5 Ki67þ cellsand 2 gH2AXþ cells per high-power field in the TG� mice (P <0.001; Fig. 5E).

Additional TGþ mouse lines show a similar phenotypeWe have 4 independently derived HIF1a TGþ lines,

g-HIF1a-M3-8, g-HIF1a-M3-25, g-HIF1a-M3-32, and g-HIF1a-M3-43. We analyzed the g-HIF1a-M3-43 line ingreatest detail because it expresses the highest level ofthe HIF1a transgene. The other 3 TGþ lines express a lowerlevel of the HIF1a transgene (Fig. 6). We observe "clear"cell morphologic abnormalities and increased expressionof CA-IX, and gH2AX in PT cells in these additional3 TGþ lines (Fig. 6A, B, C). We detect increased Ki67 in theg-HIF1a-M3-32 line, the line that expresses the secondhighest level of HIF1a protein (Fig. 6D). We also observe

ATG+ TG–

9

8

7

6

5

4

3

2

1

0γH2AX Ki67

WTHIF1α

Positiv

e c

ells

(#

pe

r fie

ld)

B

C

E

D

Figure 5. Increased gH2AX and Ki67 in g-HIF1a-M3-43 kidneys.Representative images of immunohistochemistry staining using Ki67antibody (A and B), and g-H2AX antibody (C and D) in 14-month g-HIF1a-M3-43 TGþ PTs (A and C) and TG� PTs (B and D). Increased Ki67 andgH2AX staining in TGþ (A and C, arrows). No Ki67 or gH2AX stain inTG� (B and D), but Ki67 stain in glomeruli of both TGþ and TG� kidneys(A and B). Arrows, highly stained Ki67þ (A) and gH2AXþ (C) "clear" cells.The Ki67þ and gH2AXþ cells in 10 random fields of sections of 4 TGþ

and 4 TG� were counted, E. All TGþ mice show a statistically significantincrease in Ki67þ and gH2AXþ cells (P < 0.001). Scale bars, 100 mm.

ATG+

TG+

CA-IX

β-Actin

CA-IX

CD70

β-Actin

TG–TG+ TG–

TG–

B

C D

E F

G H

I J

Figure 4. Expression of ccRCC markers in g-HIF1a-M3-43 kidneys.Representative images of immunostaining using CD31 (A and B),CA-IX (C and D), Glut-1 (E and F), and VEGF antibodies (G and H) in6-month g-HIF1a-M3-43 TGþ PTs (A, C, E, and G) and TG� PTs (B, D,F, and H). Increased staining of CD31 (A), CA-IX (C), Glut-1 (E), andVEGF (G) is only observed in the abnormal PTs of TGþ mice. Arrow(A), strong CD31 staining (red) around PT. Nuclei stained with DAPI(blue). Arrow (C), cup-like CA-IX staining of basal and lateralmembranes of "clear" cells. Arrow (E), strong Glut-1 staining aroundempty vacuole in "clear" cells. Arrow (G), strong VEGF stainingaround empty vacuole in "clear" cells. Representative Western Blot (I)of CA-IX protein and RT-PCR (J) of CA-IX and CD70 mRNA and b-actin(I and J), control. Scale bars, 100 mm.






renal cysts in older TGþ mice (Fig. 6E). Thus, the phenotypewe describe here is the result of mutant, constitutive HIF1aoverexpression.

Discussion

The kidneys of g-HIF1a-M3 transgenic mice exhibitfeatures of human VHL disease

Loss of the VHL gene plays an important role in the devel-opment of ccRCC in some VHL disease patients (3). VHLknockout mouse models have been established (14–16) butthey do not recapitulate early human ccRCCwell. For example,the "clear" cell phenotype has not been reported (14–16).Although some genes (VEGF, CA-IX) upregulated in humanccRCC showed increased expression and renal cysts wereobserved in oldmice (16), other obvious ccRCC characteristics,such as disorganized, "clear" cell proliferation in PTs, were notobserved in the kidneys of such mice.

In contrast, all TGþ mice that express the constitutivelyactive HIF1a construct show the characteristic human VHLdisease phenotype in kidney, including "clear" cells, abnor-mal vascularization, and extremely high CA-IX expression.We also observe some molecular abnormalities usually seenin tumor cells, such as increased cell proliferation and DNADSBs. Furthermore, we observe abnormalities characteristicof ccRCC, such as renal cysts, disorganized "clear" cell PTs,and CD70 overexpression (34). We observed no abnormal-ities in other organs of the g-HIF1a-M3 TGþ lines, mostlikely because the GGT promoter drives transgene expres-sion only in kidney PT cells, thought to be the cells of originof human ccRCC (40). We observed that only a subgroup ofall PTs exhibited "clear" cells; this may occur because thetruncated GGT promoter drives expression in certain seg-ments of PTs. For example, a previously reported, truncatedGGT promoter drives expression primarily in the S3 segmentof PTs (41).

HIF1a activation is responsible for the phenotype of VHLdisease in the kidney

This characteristic human ccRCC phenotype is most likelythe result of the increased expression of well-known HIF1atarget genes such as Glut-1, VEGF, and CA-IX (42, 43). Theupregulation of these genes is only seen in HIF1a expressing,"clear" cell PTs, indicating that these changes are directlyrelated to HIF1a overexpression in these "clear" cell PTs.Increased transcription of HIF1a and Glut-1 changes themetabolism and glucose uptake in HIF1a expressing kidneyPTs (11), which may account for the large increase in theaccumulation of glycogen and lipids in the cytoplasm ofPTCs in TGþ mice. Upregulation of VEGF induces the pro-liferation of endothelial cells (44), which then likely generatesthe enhanced vascularization around the abnormal PTs with"clear" cells. CA-IX is upregulated early in carcinogenesis,and CA-IX is now used as the major diagnostic and prognosticmarker of human ccRCC (32, 33). Increased expression ofCA-IX, observed in TGþmice and in early VHLdisease patients,may play a role in ccRCC carcinogenesis (45).

HIF1a activation can induce genomic instability andcell proliferation

A most interesting observation in the g-HIF1a-M3 TGþ

lines is the induction of DNA DSBs in a small percentage of"clear" cells, as indicated by gH2AX staining (Fig. 5C and E).DNA DSBs can induce genomic instability and carcinogen-esis (46). Almost every DNA DSB forms a gH2AXþ focus, andtherefore gH2AX is used to detect DNA DSBs (39). Theincreased gH2AX staining in some abnormal, "clear" cellsindicates increased numbers of DNA DSBs in these TGþ

cells. Why overexpression of a mutated, constitutively activeHIF1a is associated with increased numbers of DNA DSBs isnot known. RAD51, a key mediator of homologous recom-bination, shows reduced expression under hypoxic condi-tions (47). In preliminary experiments, we found that RAD51mRNA was lower in TGþ than in WT kidneys. Whether this isassociated with increased numbers of DSBs and/or genomicinstability is under investigation.

Increased numbers of PCNA or Ki67þ cells indicate thatmutated, constitutively active HIF1a is driving some cells toreenter the cell cycle and proliferate. This is very similar towhat has been observed in early kidney lesions of VHL diseasepatients (6). In thesepatients, 1% to 2%of the cells in theHIF1a-activated early lesions show Ki67 expression, which is about4- to 8-fold higher than normal cells.

HIF1a activation can induce VHL disease in the kidneyRenal cysts are observed in more than 60% of VHL disease

patients and are thought to be precursors of ccRCC (1). The"clear" cells in the disorganized PTs of TGþ mice (Fig. 3D)are between the stages of carcinoma in situ and frank carci-noma. However, we did not detect late stage ccRCC withinvasion and metastasis in TGþ mice. CD70 protein isupregulated in the majority of human ccRCC and is a markerof early human ccRCC (34). CD70 mRNA was higher in olderTGþ mice relative to TG� controls (Fig. 4I). This resultsuggests that there may be some neoplastically transformed,

TG+ #8 #25 #32 #43

H&E

CA-IX

Ki67

Renal

cysts

γH2AX

A

B

C

D

E

Figure 6. Staining in 4 independently derived TGþ lines. "Clear" cellmorphologic abnormalities are identified in all 4 TGþ lines (A). IncreasedCA-IX and gH2AX in all 4 TGþ lines (B and C). Increased Ki67 only ing-HIF1a-M3-43 and g-HIF1a-M3-32 (D). Renal cysts in all 4 TGþ lines (E).Scale bars, 100 mm.

Fu et al.





"clear" cells present. Cigarette smoking, obesity, and hyper-tension are 3 well-established risk factors for ccRCC (48). Weare testing these TGþ mice to determine whether exposureto these conditions results in more rapid neoplastic trans-formation of PT cells in our TGþ mice.Our results suggest that activated HIF1a functions as an

oncogene in clear cell renal carcinogenesis. Although increasedHIF1a protein expression is an early event in many solidtumors (4, 30), HIF1a has only been implicated as an etiologicfactor of ccRCC (3). The role of HIF1a in VHL disease,especially in ccRCC carcinogenesis, is of interest becauseHIF1a is directly regulated by pVHL, which is usually lost orsilenced in ccRCC (3). In an earlier publication, when pVHLexpression was restored in VHL�/� human ccRCC cells tumor-igenesis in a xenograft tumor model was blocked, and stabi-lization of HIF1a alone did not reproduce the tumor pheno-type in these ccRCC cells expressing ectopic pVHL (49).However, HIF1a is not fully activated in this model (49)because hydroxylation of the asparagine (N803) can still occur.Moreover, fully tumorigenic ccRCC cells were used in a xeno-graft model so the carcinogenesis process was not analyzed(49). In contrast, our TGþ mice allow us to follow the carci-nogenesis process over time and assess the early molecularchanges that occur.In summary, our results indicate that this constitu-

tively active HIF1a functions as an oncogene in renal

carcinogenesis. Expression of a mutated, constitutivelyactive HIF1a protein in kidney PTs results in a phenotypesimilar to that observed in patients with VHL disease,including premalignant lesions, multiple renal cysts, andccRCC.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

Acknowledgments

We thank Dr. Semenza for the HIF1a and Dr. Terzi for the GGT promoterconstruct.

Grant Support

This work was supported by WCMC, the Turobiner Kidney Cancer ResearchFund, and the Robert H. McCooey Genitourinary Oncology Research Fund. Fuholds the Robert H. McCooey Genitourinary Oncology Research Fellowship.Wang was supported by NCI-R25105012.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicate thisfact.

Received May 25, 2011; revised August 25, 2011; accepted September 6, 2011;published OnlineFirst September 9, 2011.

References1. Kaelin WG. Von hippel-lindau disease. Annu Rev Pathol 2007;2:145–

73.2. Maher ER, Yates JR, Harries R, Benjamin C, Harris R, Moore AT, et al.

Clinical features and natural history of von Hippel-Lindau disease. Q JMed 1990;77:1151–63.

3. Kaelin WG Jr. The von Hippel-Lindau tumor suppressor protein andclear cell renal carcinoma. Clin Cancer Res 2007;13:680–4s.

4. Semenza GL. Defining the role of hypoxia-inducible factor 1 in cancerbiology and therapeutics. Oncogene 2010;29:625–34.

5. Gordan JD, Simon MC. Hypoxia-inducible factors: central regulatorsof the tumor phenotype. Curr Opin Genet Dev 2007;17:71–7.

6. Mandriota SJ, Turner KJ, Davies DR, Murray PG, Morgan NV, SowterHM, et al. HIF activation identifies early lesions in VHL kidneys:evidence for site-specific tumor suppressor function in the nephron.Cancer Cell 2002;1:459–68.

7. Wang GL, Jiang BH, Rue EA, Semenza GL. Hypoxia-inducible factor 1is a basic-helix-loop-helix-PAS heterodimer regulated by cellular O2tension. Proc Natl Acad Sci U S A 1995;92:5510–4.

8. Maxwell PH, Wiesener MS, Chang GW, Clifford SC, Vaux EC, Cock-man ME, et al. The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature 1999;399:271–5.

9. Ivan M, Kondo K, Yang H, Kim W, Valiando J, Ohh M, et al. HIFalphatargeted for VHL-mediated destruction by proline hydroxylation: impli-cations for O2 sensing. Science 2001;292:464–8.

10. Jaakkola P, Mole DR, Tian YM, Wilson MI, Gielbert J, Gaskell SJ,et al. Targeting of HIF-alpha to the von Hippel-Lindau ubiquitylationcomplex by O2-regulated prolyl hydroxylation. Science 2001;292:468–72.

11. Semenza GL. Hypoxia-inducible factor 1 (HIF-1) pathway. Sci STKE2007;2007:cm8.

12. Mahon PC, Hirota K, SemenzaGL. FIH-1: a novel protein that interactswith HIF-1alpha and VHL to mediate repression of HIF-1 transcrip-tional activity. Gen Dev 2001;15:2675–86.

13. DamesSA,Martinez-YamoutM,DeGuzmanRN,DysonHJ,WrightPE.Structural basis for Hif-1 alpha/CBP recognition in the cellular hypoxicresponse. Proc Natl Acad Sci U S A 2002;99:5271–6.

14. Haase VH, Glickman JN, Socolovsky M, Jaenisch R. Vascular tumorsin livers with targeted inactivation of the von Hippel-Lindau tumorsuppressor. Proc Natl Acad Sci U S A 2001;98:1583–8.

15. Kleymenova E, Everitt JI, Pluta L, Portis M, Gnarra JR, Walker CL.Susceptibility to vascular neoplasms but no increased susceptibility torenal carcinogenesis in Vhl knockout mice. Carcinogenesis 2004;25:309–15.

16. Rankin EB, Tomaszewski JE, Haase VH. Renal cyst development inmice with conditional inactivation of the von Hippel-Lindau tumorsuppressor. Cancer Res 2006;66:2576–83.

17. Jacquemin E, Bulle F, Bernaudin JF, Wellman M, Hugon RN,Guellaen G, et al. Pattern of expression of gamma-glutamyl trans-peptidase in rat liver and kidney during development: study byimmunochemistry and in situ hybridization. J Pediatric Gastroen-terol Nutr 1990;11:89–95.

18. Terzi F, Burtin M, Hekmati M, Federici P, Grimber G, Briand P, et al.Targeted expression of a dominant-negative EGF-R in the kidneyreduces tubulo-interstitial lesions after renal injury. J Clin Invest2000;106:225–34.

19. Liu L, Gudas LJ. Disruption of the lecithin:retinol acyltransferase genemakes mice more susceptible to vitamin A deficiency. J Biol Chem2005;280:40226–34.

20. Sheren-Manoff M, Shin SJ, Su D, Bok D, Rando RR, Gudas LJ.Reduced lecithin:retinol acyltransferase expression in human breastcancer. Int J Oncol 2006;29:1193–9.

21. Gillespie RF, Gudas LJ. Retinoic acid receptor isotype specificity in F9teratocarcinoma stem cells results from the differential recruitment ofcoregulators to retinoic response elements. J Biol Chem 2007;282:33421–34.

22. Tang XH, Vivero M, Gudas LJ. Overexpression of CRABPI in supra-basal keratinocytes enhances the proliferation of epidermal basal






keratinocytes in mouse skin topically treated with all-trans retinoicacid. Exp Cell Res 2008;314:38–51.

23. SjolundJ, JohanssonM,MannaS,NorinC,PietrasA,BeckmanS, et al.Suppression of renal cell carcinoma growth by inhibition of Notchsignaling in vitro and in vivo. J Clin Invest 2008;118:217–28.

24. Tontonoz P, Hu E, Spiegelman BM. Stimulation of adipogenesis infibroblasts by PPAR gamma 2, a lipid-activated transcription factor.Cell 1994;79:1147–56.

25. Gebhard RL, ClaymanRV, PriggeWF, FigenshauR, StaleyNA, ReeseyC, et al. Abnormal cholesterol metabolism in renal clear cell carcinoma.J Lipid Res 1987;28:1177–84.

26. MotzerRJ,BanderNH,NanusDM.Renal-cell carcinoma.NEngl JMed1996;335:865–75.

27. Montani M, Heinimann K, von Teichman A, Rudolph T, Perren A,MochH. VHL-gene deletion in single renal tubular epithelial cells and renaltubular cysts: further evidence for a cyst-dependent progressionpathway of clear cell renal carcinoma in von Hippel-Lindau disease.Am J Surg Pathol 2010;34:806–15.

28. Haase VH. The VHL tumor suppressor in development and disease:functional studies in mice by conditional gene targeting. Semin CellDev Biol 2005;16:564–74.

29. Newman PJ, Berndt MC, Gorski J, White GC 2nd, Lyman S, PaddockC, et al. PECAM-1 (CD31) cloning and relation to adhesion moleculesof the immunoglobulin gene superfamily. Science 1990;247:1219–22.

30. Semenza GL. Targeting HIF-1 for cancer therapy. Nat Rev 2003;3:721–32.

31. Wykoff CC, Beasley NJ, Watson PH, Turner KJ, Pastorek J, Sibtain A,et al. Hypoxia-inducible expression of tumor-associated carbonicanhydrases. Cancer Res 2000;60:7075–83.

32. Bui MH, Seligson D, Han KR, Pantuck AJ, Dorey FJ, Huang Y, et al.Carbonic anhydrase IX is an independent predictor of survival inadvanced renal clear cell carcinoma: implications for prognosis andtherapy. Clin Cancer Res 2003;9:802–11.

33. Sandlund J,Oosterwijk E,Grankvist K,Oosterwijk-Wakka J, LjungbergB, Rasmuson T. Prognostic impact of carbonic anhydrase IX expres-sion in human renal cell carcinoma. BJU Int 2007;100:556–60.

34. Adam PJ, Terrett JA, Steers G, Stockwin L, Loader JA, Fletcher GC,et al. CD70 (TNFSF7) is expressed at high prevalence in renal cellcarcinomasand is rapidly internalised onantibodybinding.Br JCancer2006;95:298–306.

35. Prelich G, Tan CK, Kostura M, Mathews MB, So AG, Downey KM,et al. Functional identity of proliferating cell nuclear antigen anda DNA polymerase-delta auxiliary protein. Nature 1987;326:517–20.

36. Scholzen T, Gerdes J. The Ki-67 protein: from the known and theunknown. J Cell Physiol 2000;182:311–22.

37. CharamesGS, Bapat B. Genomic instability and cancer. Curr Mol Med2003;3:589–96.

38. Loeb KR, Loeb LA. Genetic instability and the mutator phenotype.Studies in ulcerative colitis. Am J Pathol 1999;154:1621–6.

39. Bonner WM, Redon CE, Dickey JS, Nakamura AJ, SedelnikovaOA, Solier S, et al. GammaH2AX and cancer. Nat Rev 2008;8:957–67.

40. WallaceAC,NairnRC.Renal tubular antigens in kidney tumors. Cancer1972;29:977–81.

41. DworniczakB, SkryabinB, Tchinda J, HeuckS, Seesing FJ,Metzger D,et al. Inducible Cre/loxP recombination in the mouse proximal tubule.Nephron Exp Nephrol 2007;106:e11–20.

42. Semenza GL. Hypoxia, clonal selection, and the role of HIF-1 in tumorprogression. Crit Rev Biochem Mol Biol 2000;35:71–103.

43. Grabmaier K, MC AdW, Verhaegh GW, Schalken JA, Oosterwijk E.Strict regulation ofCAIX(G250/MN) byHIF-1alpha in clear cell renal cellcarcinoma. Oncogene 2004;23:5624–31.

44. Leung DW, Cachianes G, Kuang WJ, Goeddel DV, Ferrara N. Vascularendothelial growth factor is a secreted angiogenic mitogen. Science1989;246:1306–9.

45. Parkkila S. Significance of pH regulation and carbonic anhydrases intumour progression and implications for diagnostic and therapeuticapproaches. BJU Int 2008;101(Suppl 4):16–21.

46. McKinnon PJ, Caldecott KW. DNA strand break repair and humangenetic disease. Annu Rev Genomics Hum Genet 2007;8:37–55.

47. Bindra RS, Schaffer PJ, Meng A, Woo J, Maseide K, Roth ME, et al.Down-regulation of Rad51 anddecreased homologous recombinationin hypoxic cancer cells. Mol Cell Biol 2004;24:8504–18.

48. Pascual D, Borque A. Epidemiology of kidney cancer. Adv Urol2008:782381.

49. Maranchie JK, Vasselli JR, Riss J, Bonifacino JS, Linehan WM, Klaus-ner RD. The contribution of VHL substrate binding and HIF1-alpha tothe phenotype of VHL loss in renal cell carcinoma. Cancer Cell2002;1:247–55.

Fu et al.





2011;71:6848-6856. Published OnlineFirst September 9, 2011.Cancer Res Leiping Fu, Gang Wang, Maria M. Shevchuk, et al.

αHIF1Mutant of Leading to Renal Cancers by Expression of a Constitutively Active

Lindau Kidney Disease−Generation of a Mouse Model of Von Hippel

Updated version

10.1158/0008-5472.CAN-11-1745doi:

Access the most recent version of this article at:

Cited articles

http://cancerres.aacrjournals.org/content/71/21/6848.full#ref-list-1

This article cites 47 articles, 15 of which you can access for free at:

Citing articles

http://cancerres.aacrjournals.org/content/71/21/6848.full#related-urls

This article has been cited by 10 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://cancerres.aacrjournals.org/content/71/21/6848To request permission to re-use all or part of this article, use this link



http://cancerres.aacrjournals.org/lookup/doi/10.1158/0008-5472.CAN-11-1745

http://cancerres.aacrjournals.org/content/71/21/6848.full#ref-list-1

http://cancerres.aacrjournals.org/content/71/21/6848.full#related-urls

http://cancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://cancerres.aacrjournals.org/content/71/21/6848


Documents

Generation of a Mouse Model of Von ... - Cancer Research · Renal cancers are highly aggressive and clinically challenging, but atransgenic mouse model to promote pathologic studies