Mutational Analysis of PHEX Gene in
PETER H. DIXON†, PAUL T. CHRISTIE, CAROL WOODING, DOROTHY TRUMP‡,
MARVIN GRIEFF, INGRID HOLM, JOSEPH M. GERTNER, JORG SCHMIDTKE,
BINITA SHAH, NICHOLAS SHAW, COLIN SMITH, CHRISTINA TAU,
DAVID SCHLESSINGER, MICHAEL P. WHYTE, AND RAJESH V. THAKKER
Medical Research Council Molecular Endocrinology Group, Medical Research Council Clinical
Sciences Centre (P.H.D., P.T.C., C.W., D.T., R.V.T.), Imperial College School of Medicine,
Hammersmith Hospital, London W12 ONN, United Kingdom; Renal Division (M.G.), Department of
Molecular Microbiology and Center for Genetics in Medicine (D.S.), and Metabolic Research Unit,
Shriners Hospital for Children and Division of Bone and Mineral Diseases (M.P.W.) ,Washington
University School of Medicine (M.G.), St. Louis, Missouri 63110; Division of Genetics, Children’s
Hospital (I.H.), Boston, Massachusetts 02115; Serono Laboratories, Inc. (J.M.G.), Norwell,
Massachusetts 02061; Medizinische Hochschule Hannover, Abteilung Humangenetik, Zentrum
Kinderheilkunde und Humangenetik (J.S.), Hannover D-30625, Germany; Childrens Medical Center of
Brooklyn, Kings County Hospital Centre, University Hospital of Brooklyn (B.S.), Brooklyn, New York
11203-2098; Department of Growth and Endocrinology, The Birmingham Children’s Hospital National
Health Service Trust (N.S.), Ladywood, Birmingham B16 8ET, United Kingdom; Alder Hey Children’s
Hospital (C.S.), Liverpool, L12 2AR, United Kingdom; and Hospital de Pediatria Garrahan,
Laboratoria de Metabolismo Calccio y Oseo, Endocrinologia (C.T.), Buenos Aeres, Argentina.
Hypophosphatemic rickets is commonly an X-linked dominant dis-
order (XLH or HYP) associated with a renal tubular defect in phos-
phate transport and bone deformities. The XLH gene, referred to as
PHEX, or formerly as PEX (phosphate regulating gene with homol-
ogies to endopeptidases on the X-chromosome), encodes a 749-amino
acid protein that putatively consists of an intracellular, transmem-
brane, and extracellular domain. PHEX mutations have been ob-
served in XLH patients, and we have undertaken studies to charac-
patients with nonfamilial XLH by single stranded conformational
polymorphism and DNA sequence analysis. We identified 31 muta-
tions (7 nonsense, 6 deletions, 2 deletional insertions, 1 duplication,
region), of which 30 were scattered throughout the putative extra-
cellular domain, together with 6 polymorphisms that had heterozy-
gosity frequencies ranging from less than 1% to 43%. Single stranded
conformational polymorphism was found to detect more than 60% of
these mutations. Over 20% of the mutations were observed in non-
familial XLH patients, who represented de novo occurrences of PHEX
mutations. The unique point mutation (a3g) of the 5?untranslated
region together with the other mutations indicates that the dominant
XLH phenotype is unlikely to be explained by haplo-insufficiency or
a dominant negative effect. (J Clin Endocrinol Metab 83: 3615–3623,
number 307800) is the commonest inherited form of rickets
(1, 2) and is usually transmitted as an X-linked dominant
disorder (3), although autosomal forms have also been ob-
served (4–6). The disorder is clinically characterized by
childhood rickets, which is unresponsive to physiological
doses of vitamin D, growth retardation, and poor dental
YPOPHOSPHATEMIC (vitamin D-resistant) rickets
(HYP or XLH, Mendelian Inheritance in Man (MIM)
development (2). In addition, extraskeletal ossification, os-
teoarthritis, limitation of joint mobility, and occasionally spi-
nal cord compression may develop (7). Affected individuals
have hypophosphatemia because of a renal tubular defect,
decreased intestinal absorption of calcium, and inappropri-
ately low serum 1,25-dihydroxy-vitamin D concentration;
the serum concentrations of calcium, PTH, and 25-hydroxy-
vitamin D are normal (2). Prepubertal girls and boys are
affected with equal severity (8), and this dominant pheno-
type, which is not generally observed in other X-linked dis-
orders, is an unusual feature of XLH. Interestingly, XLH in
adult men shows a more severe phenotype to that observed
in women, and this difference may be caused by the effects
(or HYP) gene may help to elucidate this unusual feature.
The XLH gene had been localized by family linkage stud-
ies to Xp22.1 (9–11) within a 500,000-bp region flanked cen-
tromerically by DXS365 and telomerically by DXS1683 (12).
The establishment of a YAC contig of this region (13, 14)
Received April 16, 1998. Revision received June 29, 1998. Accepted
July 6, 1998.
Address all correspondence and requests for reprints to: R. V. Thak-
ker, Medical Research Council Molecular Endocrinology Group, Med-
ical Research Council Clinical Sciences Centre, Imperial College School
of Medicine, Hammersmith Hospital, Du Cane Road, London W12
ONN, United Kingdom. E-mail: firstname.lastname@example.org.
* This work was supported by the Medical Research Council (MRC),
Grant 15958 from The Shriners Hospital for Children (to M.P.W.).
† An MRC PhD student.
‡ An MRC Training Fellow.
Journal of Clinical Endocrinology and Metabolism
Copyright © 1998 by The Endocrine Society
Vol. 83, No. 10
Printed in U.S.A.
facilitated the isolation of candidate genes and the identifi-
cation of the PHEX, formerly referred to as PEX (phosphate-
regulating gene with homologies to endopeptidases on the
X-chromosome) gene, which was found to harbor mutations
in XLH (15). The human PHEX gene (Fig. 1) consists of 22
exons that encode a 749-amino acid protein. PHEX gene
expression, as a 6.6-kilobase (kb) transcript, has been re-
ported by Northern blot analysis in adult ovary and fetal
lung, and to a lesser extent in adult lung and fetal liver
(15–17), indicating that only 35% of the PHEX messenger
remaining 65% representing untranslated regions (UTRs).
gene family (18–20), which includes neutral endopeptidase,
Kell antigen, and endothelin-converting enzyme 1. Members
of this family have a small amino-terminal intracellular tail,
a single transmembrane domain, and a large carboxy-termi-
nal extracellular domain that contains 10 conserved cysteine
many zinc peptidases (21). Neutral endopeptidase and
endothelin-converting enzyme cleave peptide bonds and al-
ter the activity of angiotensin and vasopressin and big en-
dothelin, respectively (20, 22), and it is postulated that other
family members may have similar functions. Disorders as-
sociated with mutations of neutral endopeptidase, Kell an-
tigen, or endothelin-converting enzyme have not yet been
identified, but mutations of PHEX, which are likely to result
in a functional loss, have been demonstrated to be associated
with XLH. A characterization of such mutations will help to
elucidate further the important functional domains of PHEX
and thereby the role of this putative endopeptidase in phos-
phate homeostasis. Therefore, we performed mutational
analysis of the PHEX gene in patients with familial and
nonfamilial (sporadic) forms of hypophosphatemic rickets.
Materials and Methods
The families of 68 unrelated XLH probands were ascertained and
members assessed. The diagnosis of XLH from among the various types
of rickets was based on a consistent medical history and physical ex-
amination, radiological evidence of rachitic disease, unremarkable se-
rum calcium and electrolyte concentrations, hypophosphatemia caused
by selective renal phosphate wasting for which no other etiology was
found, and a family history consistent with multigenerational or spo-
radic (i.e. nonfamilial) occurrence of XLH. Two of the probands suffered
from other putative X-linked disorders in addition to XLH; one affected
female from the Indian subcontinent suffered from XLH and congenital
adrenal hypoplasia (23), and one affected male from Argentina suffered
from XLH and Duchenne Muscular Dystrophy. Patients with suspected
tumoral rickets had been identified and were excluded from the study.
A family history of XLH could be established in 46 of the probands, and
there were 172 affected members (62 males and 110 females) and 140
unaffected members (90 males and 50 females). A familial basis for XLH
could not be established in 22 of the XLH probands (8 males and 14
females). Venous blood samples were obtained, after informed consent,
from 159 affected (62 males and 97 females) and 97 unaffected (52 males
and 45 females) members of the families of the 68 XLH probands. Of the
68 probands and their families, 62 were of northern European origin, 3
were of African-American origin, 1 was of Saudi Arabian origin, 1 was
studies had received approval from the Ethical Committee of The Ham-
mersmith Hospital, London and from the Human Studies Committee of
the Washington University School of Medicine, St. Louis, MO.
FIG. 1. Schematic representation of genomic organization of PHEX gene. Human PHEX gene consists of 22 exons that span more than 200
kb of genomic DNA and encodes a 749-amino acid peptide that has significant homology to neutral endopeptidase family (15). Filled in regions
represent 2.2 kb of coding region, and 5?- and 3?UTRs of exons 1 and 22, respectively, are indicated by open regions; 5? 72 bp of exon 1 encode
and 3? 40 bp of exon 2 together with exons 3–22 encode 700 amino acids of extracellular domain. Exon 17 contains zinc-binding motif (Zn), which
is a pentapeptide characteristic of such zinc metalloproteases. Location of 10 conserved cysteine (C) residues that are characteristic of neutral
endopeptidase family are indicated. Sites of 31 novel mutations (7 nonsense, 6 deletions, 5 deletional insertions (including duplications and
insertions), 4 splice site, 8 missense, and 1 UTR) and 6 different polymorphisms (numbers 32–37) are shown below, and details of each of these
are provided in Table 1.
3616DIXON ET AL.
JCE & M • 1998
Vol 83 • No 10
DNA sequence analysis of PHEX gene
DNA from leukocytes was prepared by standard methods, and RNA
was extracted from Epstein-Barr virus transformed lymphoblastoid cell
lines obtained from the peripheral blood cells of affected individuals
from each family, using methods previously described (24–26). DNA
sequence abnormalities were initially sought in each of 36 probands (30
familial and 6 sporadic) by RT-PCR amplification using 12 pairs of
nested PHEX-specific primers (our unpublished observations; details
available on request, from R.V.T.) and lymphoblastoid RNA, as de-
scribed (26). The PCR products were then gel purified, and the DNA
sequences of both strands were determined by Taq polymerase cycle
sequencing and a semiautomated detection system (ABI 373A se-
quencer, PE Applied Biosystems, Foster City, CA) (27, 28). DNA se-
analysis of genomic PCR products obtained by the use of appropriate
primers, or by sequence-specific oligonucleotide (SSO) hybridization
analysis or by agarose gel electrophoresis (27, 28). In addition, the DNA
sequence abnormalities were confirmed and demonstrated to cosegre-
gate with the disorder and to be absent as common polymorphisms in
the DNA obtained from 72 unrelated normal individuals (34 males, 38
females). Microsatellite polymorphism analysis at D11S533, D1S422,
D13S260, D3S1303, and a variable number tandem repeat (VNTR) at the
PTH-related peptide (PTHrP) locus were used to exclude nonpaternity
as described previously (28). Southern blot hybridization analysis (29)
was used to investigate the genomic deletions (data not shown).
The sensitivity and specificity of single-stranded conformational
polymorphism (SSCP) analysis for the detection of mutations was ini-
tially investigated by assessing the detection rate of the identified ab-
and 6 unrelated normal individuals was used with the appropriate
primers (our unpublished observations; details available on request,
from R.V.T.) for PCR amplification, and the PCR products analyzed by
SSCP using the Phast electrophoresis system (Pharmacia Biotech, Upp-
sala, Sweden) and silver staining, as previously described (28). In ad-
dition, another 32 XLH probands (16 familial and 16 sporadic) were
investigated solely by SSCP analysis for PHEX mutations. The DNA
sequence of abnormal SSCPs was determined and confirmed by restric-
tion endonuclease and SSO analysis as described above.
Mutations in PHEX
An analysis of the 2247-bp coding sequence of PHEX and
of the 508-bp 5?UTR in the 68 XLH probands revealed 31
mutations (Table 1). Thus, 7 nonsense, 6 deletions, 2 dele-
tional insertions, 2 insertions, 1 duplication, 4 splice site, 1
5?UTR, and 8 missense mutations were detected (Fig. 1). The
nonsense, insertional, duplication, and majority of the dele-
tional mutations were associated with premature termina-
tion codons, and the splice site mutations were either asso-
ciated with exon skipping or with use of cryptic splice sites,
which resulted in frameshifts that included premature ter-
mination codons (Table 1). Approximately 70% of these mu-
tations are likely to result in a truncated PHEX protein and
thus be inactivating. Twenty four (77%) of the PHEX muta-
tions were from the 46 probands with familial XLH, and 7
(23%) of the PHEX mutations were from the 22 probands
with sporadic XLH. Thus, PHEX mutations were likely to be
observed more often in probands with established X-linked
dominant inheritance rather than the nonfamilial forms.
However, PHEX mutations were not observed in 22 of the
probands, and in 5 of these families cosegregation of XLH
and PHEX had been established by studies using X-linked
flanking markers (10, 30). Each of the mutations in the 31
XLH probands was confirmed, and in the 24 familial XLH
patients was demonstrated to cosegregate with the disease,
either by restriction enzyme analysis (Figs. 2 and 3), or SSO
hybridization analysis (Fig. 4) or gel electrophoresis (Table
1). In addition, the absence of the DNA sequence abnormal-
ities in 110 alleles from 72 unrelated normal individuals (34
males and 38 females) established that these abnormalities
were mutations and unlikely to be polymorphisms, which
would be expected to occur at a frequency of more than 1%
in the general population. The 31 PHEX mutations, two of
which occurred more than once, were observed in different
in families of northern European origin, 2 in African-Amer-
ican families, 1 in a Saudi Arabian family, 1 in a southeast
Asian family, and 1 in a family from the Indian subcontinent
(Table 1). A more detailed examination of the mutations
revealed several interesting findings.
the 3? portion of exon 2 through to exon 22 (Fig. 1), and no
mutations were observed in the putative transmembrane and
intracellular domains. In addition, point mutations altering the
were found to occur more than once and in unrelated families.
The R747X mutation was observed in a Saudi Arabian family
ern European XLH family and in a nonfamilial XLH patient of
African-American origin (Table 1). Third, 2 of the 7 PHEX
mutations observed in the sporadic XLH patients were con-
firmed to have arisen de novo (Fig. 3 and Table 1). Haplotype
analysis using microsatellite polymorphisms (data not shown)
the initiation (ATG) site, was detected in one family. This mu-
tation involved an a3g transversion and was found to coseg-
the 72 unrelated normal individuals, 33 from 20 unrelated in-
other XLH probands). In addition, this nucleotide (a) is evolu-
indicate that this 5?UTR a3g transversion is not likely to be a
that may alter translation efficiency (32) (Table 1).
Polymorphisms in PHEX gene
Six polymorphisms that were all detected by SSCP anal-
restriction endonuclease analysis and SSO hybridization
analysis, were observed (Table 1 and Fig. 1) in the PHEX
gene. Two of these polymorphisms occurred in introns 17
and 18, two occurred in the 5?UTR, and the other two, which
occurred in exon 5, involved the third base of a codon that
did not lead to an alteration of the encoded amino acid. The
heterozygosity frequencies of these polymorphisms ranged
from less than 1% to 43%. The polymorphisms in intron 17,
which involved a poly T tract, and the polymorphism in
intron 18 have been previously observed (33, 34). The im-
portance of these polymorphisms lies in their recognition
and distinction from the SSCP and DNA sequence abnor-
malities that represent PHEX mutations. In addition, the g/a
MUTATIONAL ANALYSIS OF PHEX GENE 3617
polymorphism in the 5?UTR, which has a heterozygosity
frequency of 43%, may be of potential help in segregation
studies in the 55% of XLH families in whom a PHEX mu-
tation cannot be identified, but who may require presymp-
tomatic diagnosis for some family members.
Mutation detection by SSCP analysis
Eleven (?60%) of the 18 PHEX mutations in the XLH
probands detected by DNA sequence analysis of RT-PCR
products that encompassed the 2247 bp of the coding se-
Table 1: PHEX mutations and polymorphisms in XLH patients
Exon Codon Mutationk
GAA 3 TAA
TAC 3 TAA
CGA 3 TGA
TCG 3 TGA
CGA 3 TGA
CGA 3 TGA
CGA 3 TGA
Glu 3 stop
Tyr 3 stop
Arg 3 stop
Trp 3 stop
Arg 3 stop
Arg 3 stop
Arg 3 stop
Del ? 9.5 kb
Del ? 18 kb
Del ? 123 bpl
fs, 6 aa, stop
Lose exon 6
Lose exon 12
Lose exons 16,17
fs, 39 aa, stop
20687 Ins 5 bpfs, 0 aa, stopGEY
Ins 8 bp
fs, 11 aa, stop
fs, 20 aa, stop
Del 7 bp ins AA
fs, 1 aa, stop
fs, 11 aa, stop
agTT 3 ggTT
CTgt 3 CTct
86 bp insertion
Del 8 bp
Partial exon 9 skip
Exon 12 skip
Exon 21 skip
TAC 3 TTC
CCG 3 CTG
CCG 3 CTG
GGA 3 AGA
CAG 3 CGG
GCA 3 ACA
TTT 3 TAT
TGG 3 CGG
Tyr 3 Phe
Pro 3 Leu
Pro 3 Leu
Gly 3 Arg
Gln 3 Arg
Ala 3 Thr
Phe 3 Tyr
Trp 3 Arg
?429a 3 g Translation efficiency?SSOY
g 3 a
c 3 t
TCA 3 TCG
GTT 3 GTA
c 3 t
Ser 3 Ser, ?1%m
Val 3 Val, ?1%m
Intron 18 31%m
aNumber refers to location of mutation as illustrated in Fig. 1.
bAnalysis confirmed by restriction enzyme (RE), SSO hybridization, or gel electrophoresis (GE). NT, not tested.
dIndicates demonstration of de novo mutation.
eNot tested as abnormality involves regions not amplified by genomic PCR primers.
fPatient also has congenital adrenal hypoplasia.
gPatient also has Duchenne Muscular Dystrophy.
hIndicates patient of African-American origin.
iIndicates patient of Saudi Arabian origin.
jIndicates patient of Indian subcontinent origin.
kUpper and lower case indicates exon and intron sequence, respectively.
lActual genomic deletion size in female XLH patient not possible to assess by Southern blot.
mIndicates heterozygosity frequencies (not possible to determine for poly (T) tract).
nY ? Yes, N ? No.
oIndicates patient of Southeast Asian origin.
3618 DIXON ET AL.
JCE & M • 1998
Vol 83 • No 10
quence were correctly identified by SSCP analysis. In addi-
tion, SSCP analysis helped to identify 11 further mutations
(Table 1), and the detection of 5 of these is demonstrated in
Fig. 5. SSCP analysis has been reported to be more reliable
for the detection of mutations in PCR products that are
smaller than 250 bp, and 15 of the 24 total pairs of primers
used amplified products less than 250 bp. However, the
primers used for exon 22, which harbored 6 of the PHEX
mutations (Table 1), yielded fragments of 343 bp, and this
may particularly account for the lower (?66%) detection rate
for mutations in this exon. SSCP proved to be reliable in the
detection of more than 60% of all the PHEX mutations, and
a redesigning of primers to amplify smaller fragments may
help to improve this.
Our results, which have identified 31 different PHEX mu-
tations (Figs. 1–4 and Table 1) in 68 unrelated XLH patients
and their families are comparable with the combined results
FIG. 2. Detection of mutation in exon 22 in family B by restriction enzyme analysis. DNA sequence analysis of affected male IV-5 revealed a
G3A transversion at codon 720 (a), thus altering wild-type (WT) sequence, GCA, encoding an alanine (Ala) to mutant (m) sequence, ACA,
encoding a threonine (Thr). This missense mutation also resulted in gain of an RsaI restriction enzyme site (GT/ACA), and this facilitated
detection of mutation in other affected members (I-2, II-1, II-2, III-2, III-5, III-7, III-8, and IV-1) of this family (b). Following PCR amplification
and RsaI digestion, one product of 343 bp is obtained from WT (normal) sequence, but two products of 277 bp and 66 bp (not shown in b) are
obtained from m sequence (c). Cosegregation of this mutation (Ala720Thr, no. 28 in Table 1) in family B and its absence from 110 alleles of
72 unrelated normal individuals (34 males, 38 females) (N1-N3 shown), thereby indicating that it is not a common DNA sequence polymorphism
(III-6, IV-2, IV-3, IV-4) were found only to have WT sequence and none of m sequence. However, all of affected females (I-2, II-1, II-2, III-2,
as unaffected males male (▫), affected males (?), unaffected females (?), and affected females (●). DNA size standard (s) markers, which were
a 1-kb ladder, are indicated.
MUTATIONAL ANALYSIS OF PHEX GENE3619
from three recent studies (33–35) that consisted of nonsense
[23% vs. 25%; this report vs. combined studies (33–35)], de-
vs. 4%), deletional insertions (6% vs. 0%), splice site (13% vs.
14%), and missense (26% vs. 22%) mutations. The mutations
are scattered throughout the PHEX gene, and the majority
(?70%) of the 31 mutations are nonsense or frameshift de-
letions, duplications, insertions, or splice site abnormalities
that are likely, if translated, to result in a functional loss of
the PHEX protein. Interestingly, of all the PHEX abnormal-
ities reported to date, including this study, only one has
involved the putative transmembrane region (33–35). In ad-
dition, the independent and multiple occurrences of the
P534L and R747X mutations, which have also been observed
four and three times, respectively, in other studies (33, 34),
indicate that these may represent PHEX codons or regions
that are particularly prone to mutations. Our findings of two
confirms the occurrence of such mutations together with
their inheritance in subsequent generations. Five of the 7
mutations identified in the sporadic XLH patients occurred
in exons 21 and 22 and 2 of them represented the only
indicate that 22% of PHEX mutations may have arisen de
novo, and this is in agreement with the previously reported
18% estimate for de novo mutations derived from clinical
Our study is the first to report a PHEX mutation of the
5?UTR, which was observed in a female patient from the
Indian subcontinent with XLH (Table 1) and congenital ad-
to an alteration in the binding sites for ribosomal and other
FIG. 3. Detection of de novo deletional
insertion mutation in exon 21 in family
A, by restriction enzyme analysis. DNA
sequence analysis of individual III-1 re-
vealed a 7-bp deletion (CCT CAG T) at
codons 713–715 together with an inser-
tion of two bases (AA) (a). This dele-
tional insertion has resulted in a frame-
shift in which there is a Pro713Asn
substitution and a stop (TAG) at codon
714. This mutation (no. 17, Table 1 and
Fig. 1) results in loss of a DdeI restric-
tion enzyme site (A/GTTT) from normal
(WT) sequence (a), and this has facili-
tated detection of this mutation in af-
fected mother, II-2 (b). Following PCR
amplification and DdeI digestion, two
products of 103 bp and 56 bp are ob-
tained from WT sequence, but only one
product of 159 bp is obtained from m
sequence (c). Affected mother (II-2) is
heterozygous (WT/m) and her affected
son (III-1) is hemizygous for mutation.
However, this mutation was absent in
unaffected parents of affected individ-
ual (II-2), thereby demonstrating that
mutation has arisen de novo. Symbols
representing individuals are as de-
scribed in Fig. 2. In addition, absence of
mal unrelated individuals (34 male, 38
female), N1-N3 shown, indicates that it
is not a common DNA sequence poly-
3620 DIXON ET AL.
JCE & M • 1998
Vol 83 • No 10
(32). Thus, PHEX expression may be reduced and haplo-
insufficiency may represent a possible mechanism in the
etiology of XLH in this female patient. However, the R747X
and W749R mutations, which, if translated, result in almost
intact PHEX proteins, suggest that haplo-insufficiency is un-
likely to be an explanation for the dominant nature of PHEX
mutations in XLH females. An alternative possibility is that
by dimerization. However, this possibility is equally un-
likely, because X-chromosome inactivation in each female
FIG. 4. Detection of mutation in exon 9 in family C by SSO hybridization analysis. DNA sequence analysis of affected female II-1 revealed an
A3T transversion at codon 317 (a), thus altering WT sequence, TAC, encoding a tyrosine, to m sequence, TTC, encoding a phenylalanine.
Cosegregation of this mutation (Tyr317Phe, no. 23 in Table 1) in this family and its absence from 110 alleles from normal individuals (N1-N3
shown) was demonstrated by SSO hybridization analysis, because it was not associated with an alteration of a restriction enzyme site (b). Thus,
three unrelated normal individuals were found to have only WT sequence. However, two affected identical twins and their affected mother have
both WT and m sequences, and this heterozygosity indicates dominant nature of this mutation. Same mutation (Tyr317Phe) in identical twins
was associated with markedly different bone deformities (photograph). Thus, one sister (left) has genu valgum, whereas other (right) has genu
varum. These findings suggest that phenotypic expression of mutation may depend on several other factors that may involve different genes
(5, 43–46) or environmental influences such as mechanical stress and physical activity (8).
FIG. 5. Detection of five mutations in exons 9 (A) and 21 (B) by SSCP. A, Results of SSCP analysis of 8-bp insertion at codon 323 (lane d) and
C insertion at codon 329 (lane e), together with three unrelated normals (lanes a, b, and c). Mutant (m) bands, lanes d and e, differed from two
WT bands. B, Results of SSCP analysis of 8-bp splice site deletion (lane c), deletional insertion mutation at codons 713–715 (lane d), and
Arg702Stop mutation (lane e), together with two unrelated normals (lanes a and b). Mutant (m) bands differed from WT bands. SSCP analysis
was successful in detecting more than 60% of PHEX mutations.
MUTATIONAL ANALYSIS OF PHEX GENE3621
cell is likely to lead to the expression of only a wild-type or
a mutant form of PHEX, and not both, thereby precluding
any interaction between wild-type and mutant PHEX pro-
phenotype in females remains to be elucidated.
The mutational diversity within the 2247 bp of the PHEX
coding region, splice site regions and 5?UTR sequences
makes mutational screening by a direct DNA sequencing
approach in patients considered to suffer from XLH time
consuming and impractical. We have therefore explored the
use of the SSCP technique for the more rapid screening of
PHEX mutations. Our results demonstrate that SSCP was
successful in the detection of more than 60% of the PHEX
mutations, and that redesigning of some primers to amplify
DNA fragments less than 250 bp in exons 1, 5, 9, 11, 12, 18,
22, and the 5?UTR may help to increase this detection rate.
However, our DNA sequence analysis of the RT-PCR prod-
ucts from the coding regions and the 5?UTR did not detect
it is important to note that only 2.25 kb of the approximately
6.6-kb PHEX mRNA transcript has been investigated, and
that a more likely explanation for the lack of mutations in
these XLH patients is that they may harbor mutations within
the remaining 4.4-kb mRNA that contains the 3?UTR and
probable additional 5?UTR. In addition, this failure to detect
genetic heterogeneity with the possible involvement of other
X-linked genes, for example, the voltage-gated chloride
channel, CLCN5, that is located on Xp11.2 and mutated in
some patients with X-linked recessive hypophosphatemic
rickets (26), or alternatively, some of the sporadic cases may
represent autosomal forms of hypophosphatemia (4–6).
Expression of the PHEX gene has been detected by North-
ern blot analysis in mouse osteoblasts (16, 36) and in human
lung and ovary (17). Its expression in other tissues appears
to be low and detected only by RT-PCR (15, 36, 37). The
manner in which a functional loss of the putative PHEX
enzyme in these tissues leads to the anatomically remote
renal tubular defects of phosphate transport together with
function may be analogous to neutral endopeptidase, which
(18, 22, 38), or to that of endothelin-converting enzyme,
which activates its substrate, big endothelin (20). It has been
postulated that the substrate for PHEX may be phosphato-
nin, which is the putative tumour-derived phosphaturic fac-
tor from mixed mesenchymal tumours (39–42). Thus, PHEX
may inactivate phosphatonin by a possible paracrine action
(43), and a loss of PHEX function caused by mutation may
lead to increased phosphatonin activity appearing in the
circulation that may alter the activity of the sodium-phos-
phate cotransporter (NPT2) (44) and hence lead to phospha-
turia and hypophosphatemia in XLH (45, 46). The identifi-
cation of a substrate for PHEX together with the functional
expression of PHEX mutants (Table 1 and Fig. 1) will help to
1. Albright F, Butler AM, Bloomberg E. 1937 Rickets resistant to vitamin D
therapy. Am J Dis Child. 54:529–547.
2. Thakker RV, O’Riordan JL. 1988 Inherited forms of rickets and osteomalacia.
Baillieres Clin Endocrinol Metab. 2:157–191.
3. Winters RW, Graham JB, Williams TF, McFalls VW, Burnett CH. 1958 A
genetic study of familial hypophosphataemia and vitamin D resistant rickets
with a review of the literature. Medicine. 37:97–142.
4. Bianchine JW, Stambler AA, Harrison HE. 1971 Familial hypophosphataemic
rickets showing autosomal dominant inheritance. Birth Defects. 7:287–294.
5. Econs MJ, McEnery PT. 1997 Autosomal dominant hypophosphataemic rick-
ets/osteomalacia: Clinical characterisation of a novel renal phosphate wasting
disorder. J Clin Endocrinol Metab. 82:674–681.
aetiology of the ’vitamin D-resistant’ syndrome. Arch Dis Child. 51:360–365.
7. Adams JE, Davies M. 1986 Intraspinal new bone formation and spinal cord
compression in familial hypophosphataemic vitamin D resistant osteomalacia.
Q J Med. 61:1117–1129.
8. Whyte MP, Schranck FW, Armamento R. 1995 X-Linked Hypophosphatae-
expression in children. J Clin Endocrinol Metab. 81:4075–4080.
9. Read AP, Thakker RV, Davies KE, et al. 1986 Mapping of human X-linked
hypophosphataemic rickets by multilocus linkage analysis. Hum Genet.
position of ? linked hypophosphataemic rickets. J Med Genet. 24:756–760.
11. Rowe PSN, Goulding J, Read A, et al. 1993 New markers for linkage analysis
of ? linked hypophosphataemic rickets. Hum Genet. 91:571–575.
12. Econs MJ, Rowe PSN, Francis F, et al. 1994 Fine structure mapping of the
human X-linked hypophosphataemic rickets gene locus. J Clin Endocrinol
13. Francis F, Rowe PS, Econs MJ, et al. 1994 A YAC contig spanning the hy-
pophosphataemic rickets disease gene (HYP) candidate region. Genomics.
14. Trump D, Pilia G, Dixon PH, et al. 1996 Construction of a YAC contig and
an STS map spanning 3.6 megabase pairs in Xp22.1. Hum Genet. 97:60–68.
15. The HYP consortium. 1995 A gene (PEX) with homologies to endopeptidases
is mutated in patients with X-linked hypophosphataemic rickets. Nature Ge-
16. Beck L, Soumounou Y, Martel J, et al. 1997 Pex/PEX tissue distribution and
evidence for a deletion in the 3? region of the Pex gene in X-linked hypophos-
phataemic mice. J Clin Invest. 99:1200–1209.
X-linked hypophosphataemia gene cDNA. Biochem Biophys Res Commun.
18. D’Adamio L, Shipp MA, Masteller EL, Reinherz EL. 1989 Organization of the
gene encoding common acute lymphoblastic leukemia antigen (neutral en-
dopeptidase 24.11): multiple miniexons and separate 5? untranslated regions.
Proc Natl Acad Sci USA. 86:7103–7107.
19. Lee S, Zambas ED, Marsh WL, Redman CM. 1991 Molecular cloning and
primary structure of Kell blood group protein. Proc Natl Acad Sci USA.
20. Xu D, Emoto N, Giaid A, et al. 1994 ECE-1:a membrane-bound metallopro-
tease that catalyzes the proteolytic activation of big endothelin-1. Cell.
21. Turner AJ, Tanzawa K. 1997 Mammalian membrane metallopeptidases: NEP,
ECE, Kell, and PEX. FASEB J. 11:355–364.
22. Shipp MA. 1990 Down regulation of enkephalin mediated inflammatory re-
sponses by CD10/neutral endopeptidase 24.11. Nature. 347:394–396.
23. Shah BR, Fiordalisi I, Sheinbaum K, Finberg L. 1988 Familial glucocorticoid
deficiency in a girl with familial hypophosphataemic rickets. Am J Dis Child.
24. Sambrook J, Fritsch EF, Maniatis T. 1989 A laboratory manual, 2nd ed. Cold
Spring Harbor, NY: Cold Spring Harbor Press; 7.19–7.22.
25. Thakker RV, Pook MA, Wooding C, Boscaro M, Scanarini M, Clayton RN.
1993 Association of somatotrophinomas with loss of alleles on chromosome 11
and with gsp mutations. J Clin Invest. 91:2815–2821.
26. Lloyd SE, Pearce SHS, Fisher SE, et al. 1996 A common molecular basis for
three inherited kidney stone diseases. Nature. 380:445–449.
27. Pearce SHS, Williamson C, Kifor O, et al. 1996 A familial syndrome of
hypocalcaemia with hypercalciuria due to mutations in the calcium-sensing
receptor. N Engl J Med. 335:1115–1122.
28. Pearce SHS, Trump D, Wooding C, et al. 1995 Calcium-sensing receptor
mutations in familial benign hypercalcemia and neonatal hyperparathyroid-
ism. J Clin Invest. 96:2683–2692.
29. Southern EM. 1975 Detection of specific sequences among DNA fragments
separated by gel electrophoresis. J. Mol Biol. 98:503–517.
30. Thakker RV, Davies KE, Read AP, et al. 1990 Linkage analysis of two cloned
DNA sequences, DXS197 and DXS207, in hypophosphataemic rickets families.
31. Thakker, RV, Farmery MR, Sakati NA, Milner RD. 1992 Genetic linkage
Endocrinol (Oxf). 37:338–343.
3622 DIXON ET AL.
JCE & M • 1998
Vol 83 • No 10
32. Pain VM. 1996 Initiation of protein synthesis in eukaryotic cells. Europ J Bio- Download full-text
33. Rowe PSN, Oudet CL, Francis F, et al. 1997 Distribution of mutations in the
PEX gene in families with X-linked hypophosphataemic rickets (HYP). Hum
Mol Genet. 6:539–549.
PEX gene mutated in X-linked hypophosphataemic rickets. Genome Research.
35. Holm IA, Huang X, Kunkel LM. 1997 Mutational Analysis of the PEX Gene
in Patients with X-linked Hypophosphataemic Rickets. Am J Hum Genet.
36. Du L, Desbarats M, Viel J, Glorieux FH, Cawthorn C, Ecarot B. 1996 cDNA
cloning of the murine Pex gene implicated in X-linked hypophosphataemia
and evidence for expression in bone. Genomics. 36:822–837.
37. Guo R, Quarles LD. 1997 Cloning and sequencing of human PEX from a bone
cDNA library: evidence for its developmental stage-specific regulation in
osteoblasts. J Bone Miner Res. 12:1009–1017.
peptides and regulates the growth of small cell carcinomas of the lung. Proc
Natl Acad Sci USA. 88:10662–10666.
39. Cai Q, Hodgson SF, Kao PC, et al. 1994 Brief report: inhibition of renal
phosphate transport by a tumor product in a patient with oncogenic osteo-
malacia. N Engl J Med. 330:1645–1649.
40. Econs MJ, Drezner MK. 1994 Tumor-induced osteomalacia—unveiling a new
hormone. N Engl J Med. 330:1679–1681.
41. Kumar R, Haugen JD, Wieben ED, Londowski JM, Cai Q. 1995 Inhibitors of
renal epithelial phosphate transport in tumor-induced osteomalacia and ure-
mia. Proc Assoc Am Physicians. 107:296–305.
42. Wilkins GE, Granleese S, Hegele RG, Holden J, Anderson DW, Bondy GP.
1995 Oncogenic osteomalacia: evidence for a humoral phosphaturic factor.
J Clin Endocrinol Metab. 80:1628–1634.
43. Grieff M. 1997 New insights into X-linked hypophosphataemia. Curr Opin
Nephrol Hypertens. 6:15–19.
44. Miyamoto K, Tatsumi S, Yamamoto H, et al. 1995 Chromosome assignments
of genes for human Na(?)-dependent phosphate co-transporters NaPi-3 and
NPT-1. Tokushima J Exp Med. 42:5–9.
45. Econs MJ. 1996 Positional cloning of the HYP gene: a review. Kidney Int.
MUTATIONAL ANALYSIS OF PHEX GENE3623