1. Journal of Surgical Oncology 2002;80:52–60
Mutations in the von Hippel-Lindau (VHL) Gene Refine
Differential Diagnostic Criteria in Renal Cell Carcinoma
NANDITA BARNABAS, PhD,1,2 MITUAL B. AMIN, MD,1 KIRIT PINDOLIA, MS,1 RESHMA NANAVATI, MD,1
MAHUL B. AMIN, MD,3
1,4*
AND MARIA J. WORSHAM, PhD, FACMG
1Department of Pathology, Henry Ford Health System, Detroit, Michigan
2Department of Urology Research, Henry Ford Health System, Detroit, Michigan
3Department of Pathology, Emory University Hospital, Atlanta, Georgia
4Department of Otolaryngology, Henry Ford Health System, Detroit, Michigan
Background and Objectives: Renal cell carcinomas (RCC) with abundant
granular cytoplasm include oncocytomas, eosinophillic variants of chromo-phobe
RCC, papillary RCC, collecting duct carcinoma, and some conven-tional
(clear cell) RCC. Tumors with predominantly clear cell cytoplasm
include typical chromophobe RCC and conventional (clear cell) RCC. The
objective of this study was to determine if mutations in the VHL gene can
serve as auxiliary diagnostic criteria in refining histology based subtyping
of renal epithelial neoplasia.
Methods: The study cohort of 67 cases included 24 conventional RCC, 14
chromophobe RCC, 14 papillary RCC, and 15 oncocytomas. Single strand
conformational polymorphism (SSCP) was used as a screening procedure
for mutations followed by automated sequencing to identify mutations.
Results: Thirteen of the 14 mutations identified were novel, seven of which
were in the coding region. In chromophobe RCC, mutations clustered in
the 50UTR/promoter region and have not been previously reported. Exon 3
appeared to favor conventional (clear cell) RCC and correlated with a more
aggressive phenotype. Mutations were absent in the papillary and onco-cytoma
RCC subtypes.
Conclusions: Exon 3 mutations permitted a morphological distinction
between conventional (clear cell) RCC and chromophobe RCC with clear
cells. Mutations in the VHL gene refine histologic diagnostic criteria in
RCC serving as adjuncts to the present morphology based diagnosis of
RCC.
J. Surg. Oncol. 2002;80:52–60. 2002 Wiley-Liss, Inc.
KEYWORDS: clear cell; chromophobe; papillary; oncocytoma; exon 3; 50UTR
INTRODUCTION
Renal cell carcinoma (RCC), the most common malig-nancy
of the adult kidney, occurs in more than 27,000
individuals annually and is responsible for over 11,000
deaths in the United States each year. RCC most com-monly
occurs in adults between 50–70 years of age,
although it has been reported in children as young
as 3 years [1]. Just over a decade ago, renal neoplasms
were broadly classified as clear renal cell carcinomas
or granular renal cell carcinomas, based purely on the
tinctorial characteristics of the cytoplasm. Subsequent
studies saw the emergence and redefining of several
subtypes and several classification schemes were pro-posed
to include this evolving histologic spectrum [2–4].
Contract grant sponsor: NIH; Contract grant number: RO1 CA 70923;
Contract grant sponsor: ACS; Contract grant number: RPG-96-093.
*Correspondence to: Maria J. Worsham, PhD, FACMG, Department of
Otolaryngology, Henry Ford Health System, 1 Ford Place 1D, Detroit,
MI 48202. Fax: (313) 874-1079. E-mail: mworsha1@hfhs.org
Accepted 2 February 2002
DOI 10.1002/jso.10086
Published online in Wiley InterScience (www.interscience.wiley.com).
2002 Wiley-Liss, Inc.

2. Contribution of recent new and important information
regarding the origin, progression, and characteristics of
these malignancies has come from cytogenetic and mole-cular
genetic markers that provide improved classifica-tion
and diagnosis [5]. These markers hold the potential
of identifying alterations at the genetic level that may
precede morphologic change [6]. A key outcome of such
novel diagnostic concepts is reflected by the emergence
of biologically distinct entities, each characterized by a
specific combination of genetic changes.
Germline mutations of the von Hippel-Lindau (VHL)
gene cause a hereditary cancer syndrome in humans
characterized by the development of retinal and central
nervous system hemangioblastomas [7]. The VHL gene
has been localized to chromosomal region 3p25–3p26
[8]. Somatic VHL gene mutations and allele loss are
frequent events and can be detected in approximately
50% of sporadic renal cell carcinomas [9–12]. The VHL
gene has been implicated in the carcinogenesis of the
clear cell type of primary renal carcinomas [13–15]. The
majority of conventional (clear cell) RCC is associated
with losses of chromosome 3p and mutations in the VHL
gene [11]. These mutations are part of a spectrum of
familial renal cancers, arising as part of the VHL syn-drome
and the majority of sporadic renal carcinoma [16].
The present study was undertaken to assess whether
VHL mutations provide a molecular fingerprint that can
refine the current histology based differential diagnosis of
sporadic subtypes of renal neoplasia.
MATERIALS AND METHODS
Study Population
A total of 67 cases of adult renal epithelial neoplasia
seen in the Department of Pathology at Henry Ford
Hospital, Detroit, MI were selected for this study cohort.
Cases of sporadic renal cell carcinomas were chosen after
carefully excluding familial cases and cases with end
stage renal disease.
Cases were selected consecutively, purely based on
availability of sufficient paraffin embedded material for
each type and subtype of tumor, regardless of other clini-cal
parameters. The clinico-pathologic parameters used
VHL Gene Mutations in Renal Cell Carcinoma 53
for correlation in this study included, age, sex, size of
primary tumor, Fuhrman nuclear grade, TNM stage [3,4],
and follow-up. The cohort included 24 conventional
RCCs consisting of 13 clear cell, 4 mixed variants, and
7 granular NOS (not otherwise specified), 14 chromo-phobe
RCC, 14 papillary RCC, and 15 oncocytomas.
The 14 chromophobe RCCs included 6 classic and 8
eosinophillic variants.
DNA Extraction
DNA was isolated from sections of paraffin embedded
tissue using the QIAamp Kit (Qiagen, Inc. Chatsworth,
CA). Tissue was microdissected from three to five slides
of 5-micron thick sections.
Single Strand Conformational
Polymorphism (SSCP)
SSCP analysis was performed on paraffin extracted
DNA in order to identify VHL gene mutations. Exon 1, 2,
and 3 of the VHL gene were evaluated using single strand
conformational analysis. Specific primers were obtained
from published data in order to amplify exons 1–3 [16].
Fragment 1A amplifies the promoter/ 50UTR (Fig. 1)
along with the start codon. Nucleotides 565–670 encom-pass
the promoter region and nucleotides 643–714 the
50UTR region [8]. Fragments 1B, 1C, and 1D amplify
overlapping regions of exon 1 from codons 1–34, 12–80,
and 74–113, respectively (Fig. 1). Fragments 2A and 3A
(Fig. 1) amplify regions between codons 114–155 and
156–213 in exon 2 and 3, respectively [16]. The 30UTR
region was not analyzed. PCR products were internally
labeled by direct incorporation of a- 32 P dCTP during the
PCR reaction. PCR was carried out using a Perkin Elmer
9600 thermocycler (1 cycle: 11 min 958C; 30 cycles:
50 sec at 948C, 1 min at 568C, 2 min at 728C; 1 cycle:
5 min 728C). Electrophoresis was carried out on a 6%
non-denaturing gel with 5% glycerol at 8 W for 20 hr at
room temperature. Autoradiography was carried out for
12–24 hr.
DNA Sequencing
Samples with SSCP shifts were re-amplified and a
fraction of the product checked on an agarose gel. PCR
Fig. 1. PCR fragments generated using primer sequences reported by Gnarra et al. [16] representing the Promoter ‘P’/ 50UTR (1A), and exons 1
(1B, 1C, 1D), 2 (2A), and 3 (3A) are shown. The VHL gene with nucleotide numbers is indicated, based on the VHL gene sequence published in
Gene Bank by Latif et al. [8].

3. 54 Barnabas et al.
products were purified using chromatography columns
(Biospin 30, Biorad, Hercules, CA). Sequencing was
performed using the Big Dye terminator cycle sequenc-ing
kit (PE Applied Biosystems, NJ) on the ABI Prism
377 sequencer. Verification of DNA sequence changes
was attempted by sequencing in both directions. A se-quence
change remained a mutation in those cases that
failed bidirectional sequencing. Oncocytoma samples
that did not show any SSCP shifts were used as wild type
controls. Analysis was carried out using the ABI se-quencer
analysis 3.0 software and comparison with the
VHL sequence accession number AF010238 [8].
RESULTS
Histology
Oncocytomas had a nested or tubulocyotic architecture
with cells having abundant fine eosinophilic granularity
(Fig. 2A). Conventional RCC cases had extensive areas
of the tumor composed of small nests of uniformly clear
cells with low to intermediate grade nuclei separated by a
delicate and intricate connecting vascularity (Fig. 2B).
Some of the conventional RCC cases had appreciable
areas of granular eosinophilic cytoplasm. Papillary RCCs
were encapsulated with an exclusive tubulopapillary
architecture, variable nuclear grade, necrosis, and foam
cells. Chromophobe RCC cases were either clear cell
(typical) or eosinophillic variants with a nested or
alveolar growth pattern and koilocytoid nuclear atypia
(Fig. 2D).
VHL Mutations
There were 32 samples with SSCP shifts (Table I,
Fig. 3). None of the papillary RCCs or renal oncocytomas
showed SSCP shifts and were excluded from sequencing
analysis. Cases from oncocytoma and papillary subtypes
that did not show SSCP shifts were randomly chosen as
negative controls for sequencing. Of the 32 samples with
SSCP shifts, 20 mutations were identified in 18 samples.
Fourteen mutations were confirmed by bidirectional
sequencing in 13 cases (Case CL1 had two mutations).
In five cases, a mutation indicated by sequencing in one
direction could not be confirmed by sequencing in the
opposite direction (Table I). Of the conventional RCC
patients, 50% (12/24) had mutations in the VHL gene
with most of the mutations being in the 3A fragment of
exon. Mutations found in 43% of the chromophobe RCC
(6/14) were located in the fragments 1A and 1D regions
(Table I). Of the five frameshift mutations, three occurred
in codon 33, 36, and 161 (Table II, codon 161 illustrated
in Fig. 4) as early truncations and the two found in codon
185 and 206 as late terminations. These mutations have
not been previously identified in the VHL database [17].
Of the three missense mutations located at codon 33, 155,
and 156, respectively (Table II) only the 156 Tyr to Asp
transition has been previously reported [17]. Of the six
Fig. 2. Hematoxylin and eosin stained sections of various renal epithelial neoplasms. A: Renal oncocytoma, (B) Conventional renal cell
carcinoma (RCC), (C) Papillary RCC, and (D) Chromophobe RCC.

4. VHL Gene Mutations in Renal Cell Carcinoma 55
TABLE I. Clinical and Overall Molecular Data for the Sporadic Renal Epithelial Neoplasia(REN) Cohort
mutations identified in the non-coding promoter/50UTR
region of fragment 1A, two were present in the promoter
region and four in the 50UTR region of the VHL gene
(Table II). Some of these alterations may represent
normal polymorphisms. These are new additions to the
VHL database with as yet unknown functional signifi-cance
for these mutations.
Tumor Size and VHL Mutations
Tumors were grouped into three categories (0–3.9 cm,
4–7.9 cm, 8 cm) based on size and a correlation was
attempted between tumor size and type of VHL mutation.
Preliminary observations place tumors of smaller size and
improved survival with fragment 1A mutations (Table
III). Exon 3 mutations appeared to favor increased tumor
size (4 cm representing 80% of mutations) and more
aggressive disease.
Fuhrman Nuclear Grade (FNG)
and VHL Mutations
FNG is known to be a prognostic indicator in RCC.
A distinct trend was observed between FNG and VHL
mutations (Table III). In this group of cases, mutations in
the promoter/50UTR appear to be associated with lower
FNG. This pattern was also observed for exon 1 muta-tions.
In contrast, all five cases with exon 3 mutations had
tumors with high FNG (Grade 3 and 4) and were con-ventional
RCC.
Case Code Diagnosis Subclass
SSCP
shifts
VHL
mutation Age Sex Size Grade Stage
Survival
status
Follow-up
(year)
CL1 CL21 Conventional Clear 1a, 1b/1c 1a, 1c 66 M 2.8 2 II ARF 6
CL2 CL07 Conventional Clear 1b/1c 1c 41 M 5.2 2 II ARF 2
CL3 CL03 Conventional Clear 1a, 1d, 3a 3a 61 M 5.5 4 II DOD 1
CL4 CL14 Conventional Clear 3a 3a 81 F 13 4 III ARF 4.5
CL5 CL01 Conventional Clear 1c, 3a a3a 45 F 4 3 II ARF 11
CL6 CL05 Conventional Clear 1d None 72 M 2 1 I ARF 11
CL7 CL06 Conventional Clear 1a, 3a None 76 M 3.5 2 II DOTD 5
CL8 CL08 Conventional Clear 1c None 45 M 7 2 III ARF 5
CL9 CL15–17 Conventional Clear 0 None 82 M 5 2 II AFR 7
CL10 CL18 Conventional Clear 0 None 55 M 4 2 II ARF 1.5
CL11 CL23 Conventional Clear 1a None 57 M 2.5 1 I ARF 5
CL12 CL24 Conventional Clear 3a None 42 M 4.5 1 II ARF 5
CL13 CL25 Conventional Clear 3a None 57 F 8.6 4 III DOTD 1
CL14 CL19–20 Conventional Mixed 1b/1c 1c 48 M 6 2 III ARF 7
CL15 CL02 Conventional Mixed 3a 3a 52 M 3.5 3 II DOTD 13
CL16 CL09 Conventional Mixed 1c None 74 F 8 3 III ARF 4.5
CL17 CL22 Conventional Mixed 1a None 56 F 3.8 3 II ARF 5.3
GR1 CL10 Granular-NOS Granular 1a, 3a 1a 68 M 2.5 2 I ARF 9
GR2 CL27 Granular-NOS Granular 1a, 1d 1a, a1d 65 M 13 4 IV AWD 4
GR3 CL12 Granular-NOS Granular 1a, 1c, 3a 3a 69 F 12 3 III ARF 5.7
GR4 CL13 Granular-NOS Granular 3a 3a 57 M 10 3 III ARF 6
GR5 CL28 Granular-NOS Granular 1b, 1d a1b 58 F 6.2 3 II ARF 15
GR6 CL11 Granular-NOS Granular 2a, 3a None 62 F 7 3 II ARF 7.8
GR7 CL26 Granular-NOS Granular 1a, 1c None 46 M 11.4 4 III DOD 1.2
CP1 CP12 Chromophobe Eosinophilic 1a, 2a 1a 82 F 3.3 3 II ARF 2.5
CP2 CP13 Chromophobe Eosinophilic 1a 1a 77 M 8.5 2 II ARF 1
CP3 CP14 Chromophobe Eosinophilic 1a, 2a 1a 42 M 2.5 2 I ARF 2
CP4 CP08 Chromophobe Eosinophilic 1a a1a 77 M 2 3 I ARF 5.2
CP5 CP01 Chromophobe Eosinophilic 1d None 30 M 8.5 3 II ARF 3
CP6 CP09 Chromophobe Eosinophilic 1a, 1c None 61 M 5 2 II ARF 5
CP7 CP11 Chromophobe Eosinophilic 3a None 78 F 9.8 3 III ARF 3
CP8 CP10 Chromophobe Eosinophilic 0 None 80 M 5 2 II ARF 4
CP9 CP17 Chromophobe Classic 1d a1d 81 F 12.8 2 II ARF 1
CP10 CP04 Chromophobe Classic 1a a1a 59 M 8 3 II ARF 6.2
CP11 CP16 Chromophobe Classic 2a None 60 M 15 3 II ARF 2
CP12 CP05 Chromophobe Classic 0 None 32 M 5.9 3 II ARF 5.4
CP13 CP06-7 Chromophobe Classic 0 None 33 F 10 3 II ARF 5.8
CP14 CP15 Chromophobe Classic 0 None 59 F 11.5 2 II ARF 2
aBidirectional sequencing of mutations difficult to confirm due to quality of paraffin template.
ARF, alive recurrence free; DOD, dead of disease; DOTD, dead other disease; AWD, alive with disease.

5. 56 Barnabas et al.
Fig. 3. SSCP analysis of exon 3 demonstrating a shift in samples GR3 and GR4 (arrows).
TNM Stage and VHL Mutations
Although cases with mutations in the promoter /50UTR
region (fragment 1A) had a range of disease from TNM
Stage I to TNM Stage IV (Tables I and III), there was a
tendency to low and intermediate stage occurrence and
better prognosis. The only Stage IV case was alive with
disease after 4 years of initial diagnosis (Table I). Cases
with mutations in exon 3 had a tendency to be inter-mediate
to high stage with 60% (3/5) of patients having
Stage III disease (Table III).
Correlations between exon 3 mutations, aggressive-ness,
stage, and grade are trends implicated in this study.
A more comprehensive analysis with a larger number of
patients in each subgroup would be required to establish
the functional significance of such changes.
DISCUSSION
VHL disease is a dominantly inherited multisystem
family cancer syndrome, which predisposes individuals
to several cancers including renal carcinoma [1,7,18,19].
The VHL gene located at human chromosome 3p25–
3p26 [8] contains three exons and codes a putative VHL
protein (pVHL) of 213 amino acids. Loss of hetero-zygosity
(LOH) reported in 98% of tumors indicates that
the VHL gene is a tumor suppressor gene [16].
It is highly likely that the C-terminus of the VHL gene
product, pVHL, contributes to the tumor suppressor
function because many VHL mutations result in C-terminal
truncated versions of pVHL [10]. The VHL
protein has been recently shown to bind specifically to
elongins B and C, which inhibit in vitro transcription
TABLE II. Summary of the VHL Gene Mutations (Bidirectionally Verified) in the Sporadic Renal Cell Neoplasia Cohort
Case Code Position
Nucleotide
location
Nucleotide
change Codon Consequence
Amino acid
change
VHL
database*
GR3 CL27 Promoter 614 Ins G None Unknown None No
CP1 CP12 Promoter 635 Ins G None Unknown None No
GR1 CL10 50UTR 644 Ins C None Unknown None No
CP3 CP14 50UTR 650 CT None Unknown None No
CP2 CP13 50UTR 657 GC None Unknown None No
CL1 CL21 50UTR 677 CT None Unknown None No
CL14 CL19–20 Exon 1 812 TCGGCG 33 Missense Ser to Glu No
CL2 CL7 Exon 1 812 Ins G 33 Frameshift ET 131 No
CL1 CL21 Exon 1 821 4bp ins GTG 36 Frameshift ET 131 No
CL3 CL3 Exon 3 8,667 GGC/GAC 155 Missense Val to Asp/Gly No
CL15 CL2 Exon 3 8,669 TATGAT 156 Missense Tyr to Asp Yes
GR4 CL13 Exon 3 8,686 Ins G 161 Frameshift ET172 No
CL4 CL14 Exon 3 8,757 Ins G 185 Frameshift LT No
GR3 CL12 Exon 3 8,819 Ins A 206 Frameshift LT No
ET, early termination or truncations; LT: late termination.
*Beroud et al. [17].

6. Fig. 4. DNA sequencing analysis in a clear cell RCC case GR4 demonstrating an ins G (see arrow) at codon 161 resulting in a frameshift
mutation in exon 3.
activity. This suggests that the pVHL could be part of the
transcription regulatory network in direct link with its
tumor suppressor function [20–22]. The elongin binding
domain of the VHL protein is represented by codon 157–
189 and many germline and somatic mutations are
predicted to affect the integrity of this region function
[20–22]. Four of the exon 3 mutations in this report are in
the elongin binding domain of the VHL protein.
The VHL protein has also been shown to regulate
angiogenesis by downregulating VEGF (vascular endo-thelial
growth factor), to regulate ubiquitination, and
possibly protein degradation [23]. The currently accepted
mechanism of VHL gene regulation of VEGF and res-ponses
to hypoxia is via the direct regulation of the
transcription factor HIF-1 (hypoxia inducible factor)
[24]. It downregulates carbonic anhydrases 9 and 12 and
binds at least indirectly to fibronectin associated with
the endoplasmic reticulum [15]. Others have shown an
indirect relationship between VHL mutations and TGFa
expression [25]. Recently a new tumor suppressor func-tion
of VHL with its unique kinase inhibitory domain was
proposed [26].
There have been several large studies on the VHL gene
in RCC [22,27,28]. Recently, somatic mutations of the
VHL gene were detected only in clear cell RCC that
overexpressed HIF-1 [24]. Alterations of the VHL gene
have been shown to be one of the underlying factors in
sporadic tumorigenesis especially in the clear cell type of
RCC [19]. The current report reinforces these findings.
The majority of the sporadic cell renal epithelial neo-
TABLE III. Clinical Correlations for Patients With Verified VHL Gene Mutations
Mutation location
Size of primary tumor Fuhrman Nuclear Grade TNM Stage
0–4 cm 4–8 cm 8 cm 1 2 3 4 I II II IV
1a a4 2 a4 1 1 1 a2 2 1
1b/1c a1 2 a3 a2 1
1d
2
3 1 1 3 3 2 2 3
Total 5 3 5 0 6 4 3 1 5 6 1
aOne case of conventional RCC had two mutations.
VHL Gene Mutations in Renal Cell Carcinoma 57

7. plasia examined to date either lack the wild type VHL
gene or fail to produce a VHL mRNA transcript [29].
In the case of clear cell RCC, VHL inactivation has been
detected at the early renal cyst pre-malignant stage
[30,31] and suggests that inactivation of the VHL gene
may be an early event in the pathogenesis of clear cell
RCC [29].
Chromosome 3p losses and mutations in the VHL gene
have been reported predominantly in familial renal cell
cancers [22] and the majority of sporadic conventional
(clear cell) renal carcinomas [11,16]. Our study of VHL
mutations in sporadic RCC supports these observations.
Zbar et al. [10] showed that somatic VHL mutations and
hypermethylation of the VHL gene occurred in 75–80%
of the sporadic RCCs. The spectrum of mutations in the
VHL gene is large for a small gene and includes missense
and nonsense mutations, deletions of one to several
nucleotides, frameshift mutations, splice site mutations,
and even deletions of the entire gene [17,32]. A larger
patient database should permit genotype-phenotype cor-relates
with respect to specific VHL mutations [32].
SSCP screening showed more shifts than was con-firmed
by sequencing, indicating its limited role as a
stand-alone diagnostic tool. Its usefulness as a screening
tool allows the selection of subsets rather than entire
study sets for sequencing. SSCP shifts that did not in-dicate
a mutation on DNA sequencing might be partly
due to the degraded quality of formalin-fixed paraffin
tissue DNA (as compared to whole genomic DNA).
Though empirical rules on good separation of sequence
variants by SSCP are emerging, whether a certain muta-tion
can be detected in a given condition is not predic-table
[33]. It is therefore possible that a mutation would
be missed by SSCP screening and that the quoted number
of genetic alterations are likely underestimates. However,
sensitivity of PCR–SSCP is generally believed to be high
in short fragments [33], as in case of the VHL gene.
VHL mutations are extremely heterogeneous, widely
distributed throughout the coding sequence [22], and
occur rarely in the first 50 codons [22]. The current
study reports three new mutations in the first 50 codons
(Table III). VHL mutations were observed in 50% of
conventional RCC in this study group. The frequency of
VHL mutations in conventional RCC reported in the
literature range from 46–80% [6,9,10,13]. Some authors
have reported mutations in the 30UTR region [34,35].
The current study did not examine this region.
Mutations in chromophobe RCC cases occurred in the
promoter/50UTR non-coding region of the VHL gene.
Identification of promoter/50UTR mutations in chromo-phobe
RCC as distinct from exon 3 mutations in conven-tional
clear cell RCC have not been previously reported
and offer a molecular based differential diagnostic cri-teria
for these RCC groups. Earlier studies performed on
the chromophobe RCC renal subtype did not report any
mutations [12,36,37]. None of the papillary RCCs and
renal oncocytomas showed mutations, which is also con-cordant
with literature reports [27].
The current study reports 14 verified (bidirectional)
mutations, 13 of which are new observations. Only one of
the eight mutations identified in the coding region (codon
156, Tyr to Asp, exon 3) has been recorded in the
database by Beroud et al. [17], which compiles mutations
in the coding region of the VHL gene. Six of the 13 new
mutations in this study occurred in the promoter/50UTR
region of the VHL gene. There is no record in the litera-ture
of mutations in the promoter/50UTR of the VHL
gene, although reports have alluded to their presence
[19,23].
Germline and somatic mutations appear to be equally
distributed through out the VHL genome [9]. Exon 2
is alternatively spliced resulting in two distinct sizes of
transcribed messenger RNA [8]. The smaller of the
mRNA product in VHL patients with exon 2 germline
deletions indicate that the shorter transcript lacks full
functionality [16]. Somatic mutations are relatively more
frequent in exon 2 and yield a higher proportion of
truncations [9,11,12]. No mutations were found in exon 2
in the present study. That VHL mutations resulting in
truncated proteins might lead to a higher risk of RCC in
VHL patients has been suggested by Gallou et al. [27] in a
large study of sporadic renal cell carcinoma. The current
study had three early and two late truncating mutations
(Table II), and did not appear to discriminate conven-tional
clear cell and granular NOS. The latter favor a
biological relatedness for granular NOS and conventional
clear cell subtypes.
The VHL database [17] of over 600 mutations conti-nues
to expand. This growing database of new mutations,
many of them non-specific, points not so much to the
nature of these mutations but rather suggests that their
occurrence in the VHL gene may be a necessary des-tabilizing
step in tumor formation.
Mutations in the promoter and 50UTR region of the
VHL gene have not been studied for various reasons and
may have clinical relevance. Studies in prostate cancer
have shown that cases with polymorphisms in the 50UTR
of the CYP3A4 gene have a higher clinical stage [38].
Similarly, ovarian cancers with polymorphisms in the
promoter region of the matrix metalloproteinase-1
(MMP-1) gene have shown an increased expression of
MMP-1, in association with aggressive tumor behavior
[39]. There has been one report of a polymorphism in
exon 1 in the VHL gene, which encodes either a glycine
or a serine in the VHL protein [16]. Other studies indicate
base substitutions in the 50UTR and intron 1 in sporadic
RCC [14,40]. Many of these substitutions were novel,
similar to the current report, and were found in conven-
58 Barnabas et al.

8. tional (clear cell) RCC. These alterations may be DNA
sequence variants that represent normal polymorphisms
in the population and require cautious attribution to any
clinical significance in the absence of evidence that these
DNA sequence variants impair function. In the current
study, mutations in the promoter and 50UTR occurred in
conventional (clear cell) RCC and chromophobe RCCs,
and appear to be associated with a smaller size primary
tumor, a tendency for lower Fuhrman nuclear grade, and
lower stage disease. In the spectrum of adult renal epi-thelial
neoplasms, chromophobe RCCs are known to have
a relatively good prognosis [41], suggesting that muta-tions
in the promoter/50UTR appear to favor improved
disease outcomes. The latter suggest that VHL mutations
can differentiate RCC subtypes asso-ciated with better
prognosis. In our study, in the absence of mutations
elsewhere in the VHL gene, mutations in the promoter/
50UTR region suggest an etiologic contribution to the
underlying tumorigenesis of RCC in the chromophobe
subtypes. Exon 1 mutations reported in this study appear
to ascribe similar tumorigenesis potential as mutations in
the promoter/50UTR.
Mutations in exon 3 of the VHL gene appear to be
associated with a more aggressive phenotype within spo-radic
clear cell RCC as indicated by a larger size primary
tumor, a higher Fuhrman nuclear grade and a tendency to
a higher stage of disease. Zbar et al. [22] have identified
hot spots in exon 1 (codon 76 and 78) and exon 3 (codon
161), the latter being an Arg to a stop codon. The current
report also had one case with a codon 161 mutation. An
earlier study of 110 cases by Gnarra [16] included mostly
Stage III and IV tumors and showed mostly exon 3
mutations. These findings lend further support for the
association of this region of the VHL gene with more
aggressive phenotypes. All of the five mutations shown in
exon 3 in this report were located in the elongin binding
domain [20–22].
Silencing of the VHL gene via hypermethylation has
been demonstrated in clear cell RCC and in clear cell
RCC cell lines [42,43]. The methylation status of the
VHL gene was not investigated in this study.
CONCLUSIONS
Despite a small sample size, archival tissue DNA
resources, and exclusion of the study for 3p deletions,
methylation status, and for mutations in the 30UTR
region, histologic and clinical discrimination associated
with lack of mutations and with mutations when present
in the promoter/50UTR and exon 3 regions of the VHL
gene is striking. In sporadic RCC, we show that mutations
in exon 3 of the VHL gene appear to be associated with
aggressive behavior when compared to tumors with
mutations in the promoter/50UTR and in exon1. Absence
VHL Gene Mutations in Renal Cell Carcinoma 59
of mutations in papillary RCC and renal oncocytoma
subtypes indicate that mutations in the VHL gene can
discriminate between renal subtypes and can provide a
molecular basis for the differential diagnosis of conven-tional
and chromophobe subtypes. The study demon-strates
that mutations in the VHL gene refine histologic
diagnostic criteria in renal cell carcinoma and can be
useful as adjuncts to the present morphology based diag-nosis
of RCC.
REFERENCES
1. Linehan WM, Lerman MI, Zbar B: Identification of the von
Hippel-Lindau (VHL) gene. Its role in renal cancer. JAMA
1995;273:564–570.
2. Storkel S, Eble JN, Adlakha K, et al.: Classification of renal cell
carcinoma: Workgroup No. Union Internationale Contre le
Cancer (UICC) and the American Joint Committee on Cancer
(AJCC). Cancer 1997;80:987–989.
3. Tickoo SK, Amin MB, Linden MD, et al.: Antimitochondrial
antibody (113-1) in the differential diagnosis of granular renal cell
tumors. Am J Surg Pathol 1997;21:922–930.
4. Tickoo SK, Lee MW, Eble JN, et al.: Ultrastructural observations
on mitochondria and microvesicles in renal oncocytoma,
chromophobe renal cell carcinoma, and eosinophilic variant of
conventional (clear cell) renal cell carcinoma. Am J Surg Pathol
2000;24:1247–1256.
5. Bugert P, Kovacs G: Molecular differential diagnosis of renal cell
carcinomas by microsatellite analysis. Am J Pathol 1996;149:
2081–2088.
6. Steiner G, Sidransky D: Molecular differential diagnosis of renal
carcinoma: from microscopes to microsatellites. Am J Pathol
1996;149:1791–1795.
7. Maher ER, Kaelin WG, Jr.: von Hippel-Lindau disease. Medicine
1997;76:381–391.
8. Latif F, Tory K, Gnarra J, et al.: Identification of the von Hippel-
Lindau disease tumor suppressor gene. Science 1993;260:1317–
1320.
9. Foster K, Prowse A, van den Berg A, et al.: Somatic mutations
of the von Hippel-Lindau disease tumour suppressor gene in
non-familial clear cell renal carcinoma. Hum Mol Genet 1994;3:
2169–2173.
10. Zbar B: von Hippel-Lindau disease and sporadic renal cell
carcinoma. Cancer Surv 1995;25:219–232.
11. Fleming S: Genetics of kidney tumours. Forum (Genova) 1998;8:
176–184.
12. Shuin T, Kondo K, Torigoe S, et al.: Frequent somatic mutations
and loss of heterozygosity of the von Hippel-Lindau tumor
suppressor gene in primary human renal cell carcinomas. Cancer
Res 1994;54:2852–2855.
13. Suzuki H, Ueda T, Komiya A, et al.: Mutational state of von
Hippel-Lindau and adenomatous polyposis coli genes in renal
tumors. Oncology 1997;54:252–257.
14. Yang K, Lindblad P, Egevad L, et al.: Novel somatic mutations
in the VHL gene in Swedish archived sporadic renal cell car-cinomas.
Cancer Lett 1999;141:1–8.
15. Zbar B, Kaelin W, Maher E, et al.: Third international meeting on
von Hippel-Lindau disease. Cancer Res 1999;59:2251–2253.
16. Gnarra JR, Tory K, Weng Y, et al.: Mutations of the VHL tumour
suppressor gene in renal carcinoma. Nat Genet 1994;7:85–90.
17. Beroud C, Joly D, Gallou C, et al.: Software and database for the
analysis of mutations in the VHL gene. Nucleic Acids Res 1998;
26:256–258.
18. Richards FM,Webster AR, McMahon R, et al.: Molecular genetic
analysis of von Hippel-Lindau disease. J Intern Med 1998;243:
527–533.
19. Decker HJ, Weidt EJ, Brieger J, et al.: The von Hippel-Lindau
tumor suppressor gene. A rare and intriguing disease opening new

9. insight into basic mechanisms of carcinogenesis. Cancer Genet
Cytogenet 1997;93:74–83.
20. Kibel A, Iliopoulos O, DeCaprio JA, et al.: Binding of the von
Hippel-Lindau tumor suppressor protein to Elongin B and C.
Science 1995;269:1444–1446.
21. Kishida T, Stackhouse TM, Chen F, et al.: Cellular proteins that
bind the von Hippel-Lindau disease gene product: mapping of
binding domains and the effect of missense mutations. Cancer Res
1995;55:4544–4548.
22. Zbar B, Kishida T, Chen F, et al.: Germline mutations in the von-
Hippel Lindau disease (VHL) gene in families from North
America, Europe, and Japan. Hum Mutat 1996;8:348–357.
23. Kaelin WG, Jr.: Cancer. Many vessels, faulty gene. Nature 1999;
399:203–204.
24. Weisener MS, Munchenhagen PM, Berger I, et al.: Constitutive
activation of hypoxiinducible genes related to overexpression of
hypoxia-inducible factor-1a in clear cell renal carcinoma. Cancer
Res 1998;61:5215–5222.
25. Knebelmann B, Ananth S, Cohen HT, et al.: Transforming growth
factor alpha is a target for the von Hippel-Lindau tumor
suppressor. Cancer Res 1998;58:226–231.
26. Datta K, Nambudripad R, Pal S, et al.: Inhibition of insulin-like
growth factor-I-mediated cell signaling by the von Hippel-Lindau
gene product in renal cancer. J Biol Chem 2000;275:20700–
20706.
27. Gallou C, Joly D, Mejean A, et al.: Mutations of the VHL gene in
sporadic renal cell carcinoma: definition of a risk factor for VHL
patients to develop an RCC. Hum Mutat 1999;13:464–475.
28. Bailly M, Bain C, Favrot MC, et al.: Somatic mutations of von
Hippel-Lindau (VHL) tumor-suppressor gene in European kidney
cancers. Int J Cancer 1995;63:660–664.
29. Kaelin WG, Iliopoulos O, Lonergan KM, et al.: Functions of the
von Hippel-Lindau tumour suppressor protein. J Intern Med 1998;
243:535–539.
30. Lubensky IA, Gnarra JR, Bertheau P, et al.: Allelic deletions of
the VHL gene detected in multiple microscopic clear cell renal
lesions in von Hippel-Lindau disease patients. Am J Pathol 1996;
149:2089–2094.
31. Zhuang Z, Bertheau P, Emmert-Buck MR, et al.: A microdis-section
technique for archival DNA analysis of specific cell
populations in lesions 1 mm in size. Am J Pathol 1995; 146:
620–625.
32. Neumann HP, Bender BU: Genotype-phenotype correlations
in von Hippel-Lindau disease. J Intern Med 1998;243:541–
545.
33. Hayashi K, Kukita Y, Inazuka M, et al.: Single strand con-formation
polymorphism analysis. In Hames BD (ed): Mutation
Detection: A Practical Approach. 1st edition. New York: Oxford
University Press, 1998:7–24.
34. Payne SJ, Richards FM, Maher ER, et al.: A PCR generated AccI
RFLP in the 30 untranslated region of the von Hippel-Lindau
disease (VHL) tumour suppressor gene. Hum Mol Genet 1994;3:
390.
35. Jackson RJ: Cytoplasmic regulation of mRNA function: the
importance of the 30 untranslated region. Cell 1993;74:9–14.
36. Kenck C, Wilhelm M, Bugert P, et al.: Mutation of the VHL gene
is associated exclusively with the development of non-papillary
renal cell carcinomas. J Pathol 1996;179:157–161.
37. Brauch H, Weirich G, Brieger J, et al.: VHL alterations in human
clear cell renal cell carcinoma: association with advanced tumor
stage and a novel hot spot mutation. Cancer Res 2000;60:1942–
1948.
38. Rebbeck TR, Jaffe JM,Walker AH, et al.: Modification of clinical
presentation of prostate tumors by a novel genetic variant in
CYP3A4. J Natl Cancer Inst 1998;90:1225–1229.
39. Kanamori Y, Matsushima M, Minaguchi T, et al.: Correlation
between expression of the matrix metalloproteinase-1 gene in
ovarian cancers and an insertion/deletion polymorphism in its
promoter region. Cancer Res 1999;59:4225–4227.
40. Chen F, Kishida T, Yao M, et al.: Germline mutations in the von
Hippel-Lindau disease tumor suppressor gene: correlations with
phenotype. Hum Mutat 1995;5:66–75.
41. Amin MB, Amen M, Tamboli P, et al.: Prognostic impact of
histologic subtyping of adult Renal Epithelial Neoplasms: An
experience of 405 cases. Am J Surg Pathol 2002 (in press).
42. Clifford SC, Prowse AH, Affara NA, et al.: Inactivation of the
von Hippel-Lindau (VHL) tumour suppressor gene and
allelic losses at chromosome arm 3p in primary renal cell
carcinoma: evidence for a VHL-independent pathway in clear
cell renal tumourigenesis. Genes Chromosomes Cancer 1998;22:
200–209.
43. Herman JG, Latif F, Weng Y, et al.: Silencing of the VHL tumor-suppressor
gene by DNA methylation in renal carcinoma. Proc
Natl Acad Sci USA 1994;91:9700–9704.
60 Barnabas et al.