Gene 576 (2016) 182–188

Contents lists available at ScienceDirect

Gene

journal homepage: www.elsevier.com/locate/gene

Short communication

A novel common large genomic deletion and two new missense

mutations identified in the Romanian phenylketonuria population
Corinne Gemperle-Britschgi a, Daniela Iorgulescu b, Monica Alina Mager c, Dana Anton-Paduraru d,
Romana Vulturar e,f,⁎, Beat Thöny a,g,h,i,j,⁎⁎
a
Division of Clinical Chemistry and Biochemistry, Department of Pediatrics, University of Zürich, Zürich, Switzerland
b
Center of Newborn Screening, Department of Pediatrics, Institute for Mother and Child Care, Bucharest, 120 Lacul Tei Blv., 020395, Romania
c
“I. Hatieganu” University of Medicine and Pharmacy — Cluj-Napoca, Department of Neurology, Romania
d
“Gr.T.Popa” University of Medicine and Pharmacy, Newborn Screening Center Iasi, 3rd Clinic of Pediatrics, Romania,
e
“I.Hatieganu” University of Medicine and Pharmacy — Cluj-Napoca, Department of Molecular Sciences, Cluj-Napoca, Romania
f
Cognitive Neuroscience Laboratory, Babes-Bolyai University — Cluj-Napoca, Romania
g
Division of Metabolism, Department of Pediatrics, University of Zurich, Steinwiesstrasse 75, Zurich CH-8032, Switzerland
h
Neuroscience Centre Zürich, University of Zürich, Zürich, Switzerland
i
Neuroscience Centre Zürich, ETH Zürich (ZNZ), Zürich, Switzerland
j
Children's Research Centre (CRC), Zürich, Switzerland

a r t i c l e i n f o a b s t r a c t

Article history: The mutation spectrum for the phenylalanine hydroxylase (PAH) gene was investigated in a cohort of 84 hyper-
Received 4 March 2015 phenylalaninemia (HPA) patients from Romania identified through newborn screening or neurometabolic inves-
Received in revised form 2 June 2015 tigations. Differential diagnosis identified 81 patients with classic PAH deficiency while 3 had tetrahydropterin-
Accepted 12 October 2015
cofactor deficiency and/or remained uncertain due to insufficient specimen. PAH-genetic analysis included a
Available online 21 October 2015
combination of Sanger sequencing of exons and exon–intron boundaries, MLPA and NGS with genomic DNA,
Keywords:
and cDNA analysis from immortalized lymphoblasts. A diagnostic efficiency of 99.4% was achieved, as for one al-
Hyperphenylalaninemia lele (out of a total of 162 alleles) no mutation could be identified. The most prevalent mutation was p.Arg408Trp
Phenylketonuria which was found in ~38% of all PKU alleles. Three novel mutations were identified, including the two missense
Mutation analysis mutations p.Gln226Lys and p.Tyr268Cys that were both disease causing by prediction algorithms, and the
large genomic deletion EX6del7831 (c.509 + 4140_706 + 510del7831) that resulted in skipping of exon 6
based on PAH-cDNA analysis in immortalized lymphocytes. The genomic deletion was present in a heterozygous
state in 12 patients, i.e. in ~ 8% of all the analyzed PKU alleles, and might have originated from a Romanian
founder.
© 2015 Elsevier B.V. All rights reserved.

1. Introduction
Hyperphenylalaninemia caused by autosomal recessively inherited
mutations in the PAH gene encoding the phenylalanine hydroxylase
(PAH; E.C. 1.14.16.1), also termed phenylketonuria (PKU; OMIM
#261600)(Blau et al., 2010), presents with a large but continuously
growing spectrum of DNA mutations based on numerous studies in
many countries and various ethnic populations (for a recent overview
Abbreviations: PAH, phenylalanine hydroxylase; HPA, hyperphenylalaninemia; PKU,
phenylketonuria; BH4, tetrahydrobiopterin; L-Phe, phenylalanine; MLPA, multiplex
ligation-dependent probe amplification; NGS, next generation sequencing.
⁎ Correspondence to: R. Vulturar, “I. Hatieganu” University of Medicine and Pharmacy —
Cluj-Napoca, Department of Molecular Sciences, Cluj-Napoca, Romania.
⁎⁎ Correspondence to: B. Thöny, Division of Metabolism, Department of Pediatrics,
University of Zurich, Steinwiesstrasse 75, Zurich CH-8032, Switzerland.
E-mail addresses: vulturar.romana@umfcluj.ro, romanavulturar@yahoo.co.uk
(R. Vulturar), beat.thony@kispi.uzh.ch (B. Thöny).
(Blau et al., 2014)). While over 98% hyperphenylalaninemias are due
to mutation in the PAH gene (*612349), the remaining variants are
due to rare mutation in the tetrahydrobiopterin (BH4) cofactor biosyn-
thesis or regeneration (see (Blau et al., 2010; Werner et al., 2011)).
BH4 cofactor deficiency should not be confounded with cofactor respon-
siveness where (oral) supplementation with synthetic BH4 in so-called
loading tests revealed that in up to about 20% of patients with PKU, liver
PAH activity is stimulated and thus is lowering plasma L-Phe concentra-
tions (Blau et al., 2010). Apart from the over 850 allelic variants known
today, estimations on incidence based on newborn screening is as broad
as from 1:2,700 in Sicily (Guldberg et al., 1993) compared to 1:100,000
in Finland (Guldberg et al., 1995), with a European-Caucasian average of
1:10,000. However, a much lower frequency is observed in the Asian
population (for an overview see (Donlon et al., 2010)).
Newborn screening for PKU for the Romanian population with 19.9
million inhabitants was implemented in the year 1979 and is currently
being carried out in five centers. Screening for PKU started in two

http://dx.doi.org/10.1016/j.gene.2015.10.020
0378-1119/© 2015 Elsevier B.V. All rights reserved.
 C. Gemperle-Britschgi et al. / Gene 576 (2016) 182–188
centers, Bucharest and Cluj, and after that it was also implemented in
Iasi and two much smaller centers, Timisoara and Tirgu Mures. In
1998, the incidence of hyperphenylalaninemia in the Romanian popula-
tion was reported to be 1:8,000 based on an 18 year newborn screening
(Popescu et al., 1998a). Since 2011, the program is covering the entire
country on a national level with three main screening centers including
Bucharest (southern part), Cluj (west part) and Iasi (east part). Accord-
ing to the report by Bodamer and coworkers (Bodamer et al., 2007)
there is no information on a screening center in Romania and on the in-
cidence of PKU available. On the other hand, it was published in 1998
that an 18-year HPA screening in Transylvania demonstrated a PKU in-
cidence of 1:8,000 in newborn and 1:110 in retarded children (Popescu
et al., 1998a). Although the incidence of confirmed PKU patients in
Romania is officially not published, the Bucharest National Authority
for “Neonatal screening, diagnostic confirmation and specific treatment
in PKU and congenital hypothyroidism”, financed by the Romanian
Ministry of Health, reported that the calculated incidence, after the eval-
uation of the results from the entire country in 2013, is 1 case in 10,000
births (Dr. Nanu, personal communication).
Genotyping of PAH mutations in patients is primarily confirmatory
of diagnosis but with essential implications for the understanding
of the molecular basis and pathophysiology of PKU (for a broader
discussion see (Blau et al., 2010)). Standard molecular analysis includes
Sanger sequencing of all exon and exon-boundaries that can be
complemented, in case of incomplete or unclear analysis, with e.g. mul-
tiplex ligation-dependent probe amplification (MLPA) analysis
(Schouten et al., 2002). Alternatively, genomic DNA analysis for
the PAH gene can also be expanded to cDNA analysis isolated from
EBV-immortalized lymphocytes to identify or confirm transcriptional
abnormalities (Abadie et al., 1993). Recently, high-throughput, next
generation sequencing methods were introduced for analysis of PKU
patients mainly as a rapid alternative to Sanger sequencing but also to
solve insertions, deletions and/or larger rearrangement in PKU, that
cannot be easily identified with conventional molecular analyses (Cao
et al., 2014; Gu et al., 2014; Trujillano et al., 2014).
The aim of this study was to elucidate the spectrum of PKU
genotypes in the current Romanian population. For this, we have used
a combination of several approaches, i.e. Sanger and next generation
sequencing, MLPA and cDNA analyses, to achieve N99% diagnostic effi-
ciency to identify the mutation spectrum in 81 Romanian PKU patients.
Besides the novel deletion Ex6del7831, we report on two new missense
mutations, p.Gln226Lys and p.Tyr268Cys, which are predicted to be dis-
ease causing.
2. Subjects and methods
2.1. Patients
The 84 index patients with hyperphenylalaninemia, known to the
University of Medicine and Pharmacy Cluj-Napoca (for biochemical
and clinical evaluation in Department of Molecular Sciences and Depart-
ment of Neurology), and from Bucharest and Iasi PKU screening centers
were contacted and investigated for this study between January 2012
and July 2014. These patients presenting in Bucharest, Iasi and in Cluj-
Napoca, where identified either through newborn screening by measur-
ing L-phenylalanine (L-Phe) levels in dried blood spots using the
victor2d fluorometer (Perkin Elmer, Fluoroskan ascent neonatal ANI
Labsystems Thermo Scientific, Finland), or using the chromatography
method (see below). Informed consent was obtained from all parents
or legal representatives.
2.2. Metabolic and genetic analyses
Blood L-Phe concentrations were confirmed in Cluj from venous
blood using a method based on two-dimensional thin layer chromatog-
raphy (Wadman et al., 1969) coupled with video densitometry (using
183
BioRad videodensitometer GS-700 and Molecular Analyst software
(Vulturar et al., 2005)). In Bucharest L-Phe concentration was confirmed
from venous blood using an HPLC method. Pterin concentrations were
measured from dried blood spots using a HPLC method for pterins
(Opladen et al., 2011). DHPR activity was measured from dried blood
spots by enzyme assay (Arai et al., 1982). DNA was isolated from pe-
ripheral blood using phenol-chloroform-extraction, the MasterPure
DNA Purification Kit for Blood (Epicenter, Madison, USA) or the Arrow
Blood DNA kit plus the extraction device Arrow (Nordiag, Bergen,
Norway).
Genetic analysis included PCR amplification and “Sanger” genomic
DNA sequencing carried out on a Biometra T3000 PCR System and a
3130xl Genetic Analyzer (Applied Biosystems Inc., Switzerland), respec-
tively. For each patient, the PAH gene was screened for mutations step
by step, starting with the exons most probably carrying a mutation
and finishing analysis, as soon as two mutant alleles were found. The
corresponding primers for PCR amplification of genomic DNA isolated
from peripheral blood samples, as well as for DNA sequence analysis,
were used according to our published method (Supplementary
Table S1 in (Kostandyan et al., 2011). The standard amplification condi-
tions we used can be obtained upon request. Reference accession num-
ber for normal alleles is ENSG00000171759.2 for PAH.
Mutation designation was according to the official mutation nomen-
clature (http://www.hgvs.org/mutnomen/). To validate nomenclature
of mutations, we used the program Mutalyzer 2.0 β-2 (http://www.
mutalyzer.nl/2.0/).
Diagnosis of the EX6del7831 deletion described here in the
Romanian cohort was carried out by conventional PCR using genomic
DNA and the breakpoint-flanking primers PAH_delE6F (5′-CATCATTG
CTGGCTACCACA-3′) and PAH_delE6R (5′-GAGGCTTGCGACTCAATGTA
CT-3′; see also Fig. 1). Initial diagnosis and molecular characterization
of this deletion were done by a combination of MLPA, next generation
sequencing and Sanger sequencing with genomic DNA, and with
cDNA analysis from immortalized lymphoblasts. The only specimen,
i.e. cDNA from immortalized lymphoblasts available for these analyses
was from an unrelated patient of Turkish origin with the identical geno-
mic deletion as found in the Romanian subjects (TUR-304 (Dobrowolski
et al., 2011)). MLPA was carried out according to the manufacturer's
protocol using the kit no. SALSA P055-C1 (MRC Holland, The
Netherlands). Next generation sequencing was performed following
the Nimblegen SeqCap EZ Library LR User's Guide using a custom
Nimblegen SeqCap EZ Choice design covering the whole PAH gene (as
far as specific targeting was possible). Sequencing was done on a
Roche GS Junior and analyzed with the Roche GS Reference Mapper
(Roche Diagnostics, Switzerland). The mean coverage of 20 was in this
case sufficient to see the deletion range. RT-PCR was done with RNA iso-
lated from EBV-transformed lymphoblasts according to a published
protocol (Abadie et al., 1993). In order to avoid allelic dropout, frag-
ments B and C were amplified as one fragment with a length of
729 bp using the forward primer 5′-GTCCATGAGCTTTCACGAGATAAG-
3′ and the reverse primer 5′-AAGGCATATGGTGCTGGGCTCCTGT-3′ (for
more details see ref. (Abadie et al., 1993)).
3. Results and discussion
Information on the mutation spectrum in Romanian PKU patients
was limited to basically the two reports from 1998 where the haplo-
types of 22 Romanian families plus a Romanian Gypsy subject were de-
scribed (Popescu et al., 1998a, 1998b). In the total 27 patients studied by
Popescu and coworkers about 11% of the PAH alleles remained undiag-
nosed (see Table 1 in ref. (Popescu et al., 1998a)). A communication
from 2010 presented the distribution of three types of mutations,
p.Arg408Trp, p.Arg261Gln and p.Leu48Ser, previously found as being
frequent in Central Europe, identified in 16 alleles of a small cohort of
PKU Romanian patients, using the PCR-RFLP method (Vulturar and
Barecki, 2010). To obtain a more thorough overview, we collected
184 C. Gemperle-Britschgi et al. / Gene 576 (2016) 182–188

Fig. 1. Breakpoint analyses of genomic deletion EX6del7831 resulting in exon 6 skipping of the PAH-mRNA. (A) Schematic representation of the 7831 bp deletion encompassing exon 6 of
the human PAH gene with break points downstream of exon 5 (c.509 + 4140) and of exon 6 (c.706 + 510). Various repeating elements in intron 5 and intron 6 of the PAH gene are
depicted in different colors. These elements were identified using the RepeatMasker program that screens DNA sequences for interspersed repeats and low complexity DNA sequences
(http://www.repeatmasker.org/). (B) Amplification and Sanger sequencing of the of the genomic exon 6 deletion. The indicated primers flanking the deletion, as indicated also in (A),
were used to amplify the 649 bp PCR product. Sequence analysis of the 649 bp fragment shows the details of the break point with a GTAT repeat boxed in red. (C) PAH transcript analysis
by RT-PCR from EBV-transformed B-lymphoblasts established exon 6 deletion (P, patient; C, control). Note that the EBV-transformed B-lymphoblasts for the analysis presented were from
an unrelated PKU patient of Turkish origin.

samples available from hyperphenylalaninemia subjects, adult and
child PKU patients, between January 2012 and July 2014 from the
Bucharest and Iasi screening centers, and from Cluj (University of Med-
icine), and performed molecular genetic investigations. Among the total
of 84 subjects initially analyzed in the present study (see below), 12
were overlapping with previous analyses; however, it was not possible
for the authors to unequivocally identify or sort out which were the
overlapping subjects.
Following standard differentiation testing on dried blood spots, in-
cluding measurement of DHPR activity and analysis of pterins (Blau
and Van Spronsen, 2014), we identified 81 patients with classic PAH de-
ficiency while 3 had tetrahydrobiopterin (BH4)-cofactor deficiency but
unfortunately remained unclear for the specific defect due to insuffi-
cient specimen available for further work-up. We continued with muta-
tion analysis by Sanger sequencing of PAH-exons and exon–intron
boundaries from 81 subjects. Whenever possible, we confirmed muta-
tion analysis by genotyping the corresponding parents or siblings (not
shown). The results from genotyping, including blood L-Phe are summa-
rized in Table 1. We achieved an overall diagnostic efficiency of 99.4%, as
for one allele out of a total of 162 alleles no mutation could be identified
(case #F24C in Table 1). Here we applied besides Sanger sequencing,
MLPA and NGS, the latter with coverage of 20, and detected various
SNPs but none of them could be assigned to a potential disease causing
mutation. Unfortunately, data on BH4-cofactor responsiveness are not
available for any of these patients presented here.
Among the 162 alleles analyzed, the most prevalent variant in this
population was the c.1222C N T (p.Arg408Trp) mutation which we
found in 61 alleles, accounting for 37.7% of all PKU alleles analyzed,
and which was present homozygously in 16 out of 81 subjects
(19.8%). These numbers are somewhat comparable to what was
reported before in an analysis of the 22 Romanian families with PKU pa-
tients (47.7%; (Popescu et al., 1998a)) but significantly higher than re-
ported in the BIOPKU database, where the protein variant p.Arg408Trp
was the most prevalent allele (22.6%) and p.Arg408Trp;p.Arg408Trp
with the highest genotype frequency (10.8%) (Blau et al., 2014). Accord-
ing to a review article on PKU mutations in Europe, Zschocke (2003)
also described the variant p.Arg408Trp as the most common mutation
in Eastern Europe. The second most common allele we identified was
the variant c.143TNC (p.Leu48Ser) which was found in 15 patients, ac-
counting for 9.3% all the PKU chromosomes. Together with the deletion
allele EX6del7831 (see below), which was found in a heterozygous state
in 12 patients, i.e. in 7.4% of all alleles, these allelic variants p.R408W,
p.Leu48Ser and EX6del7831 represent over 54% of all the PKU chromo-
somes in the Romanian population analyzed thus far.
As mentioned above, we found the common but novel large genomic
deletion EX6del7831 in a heterozygous state in 12 PKU subjects by a
combination of MPLA and/or breakpoint PCR-sequencing of genomic
DNA. The detailed molecular analysis is shown in Fig. 1, where we
used a combination of MLPA, next generation sequencing and PAH-
cDNA analysis in immortalized lymphocytes, to define the exact geno-
mic deletion c.509 + 4140_706 + 510del7831. Although this deletion
was also detected in one Turkish PKU patient with a hitherto incomplete
molecular diagnosis (CB and BT, unpublished; see also chapter “Subjects
and methods”), the high frequency in the PKU population analyzed here
might hint towards a Romanian founder and/or a heterozygous advan-
tage. According to Zschocke (2003) there have been numerous inde-
pendent mutation events for PKU in Europe, and several mutations
have independently recurred in different founders. Furthermore, it
was proposed that a potential heterozygous advantage by an “over-
dominant selection” must have played an important role during the
 C. Gemperle-Britschgi et al. / Gene 576 (2016) 182–188 185

Table 1
Spectrum of Romanian PKU patients.

Blood L–Phe level at

Case # Allele 1 Allele 2 newborn–screening
(µmol/L)

F1C c.143T>C, p.Leu48Ser, missense c.1222C>T, p.Arg408Trp, missense 1212

F2C c.472C>T, p.Arg158Trp, missense c.1208C>T, p.Ala403Val, missense 333

F3C c.1066–11G>A, splicing c.1066–3C>T, splicing 727

F4C c.842C>T, p.Pro281Leu, missense c.533A>G, p.Glu178Gly, missense 424

F5C c.1208C>T, p.Ala403Val, missense c.898G>T, p.Ala300Ser, missense 224

F6C c.143T>C, p.Leu48Ser, missense c.916A>G, p.Ile306Val, missense 333

F7C c.168+5G>C, splicing c.1066–11G>A, splicing 1091

F8C c.1222C>T, p.Arg408Trp, missense c.1169G>A, p.Glu390Gly, missense 364

F9C c.803A>G, p.Tyr268Cys, missense c.533A>G, p.Glu178Gly, missense, 485

F10C c.165delT, p.Phe55Leufs*6, deletion c.1066–11G>A, splicing 1515*

F11C c.1222C>T, p.Arg408Trp, missense c.1222C>T, p.Arg408Trp, missense 1394 *

F12C c.842C>T, p.Pro281Leu, missense c.673C>A, p.Pro225Thr, missense 2000

F13C c.842C>T, p.Pro281Leu, missense c.898G>T, p.Ala300Ser, missense 279

F14C c.782G>A, p.Arg261Gln, missense c.1222C>T, p.Arg408Trp, missense 1091

F15C c.533A>G, p.Glu178Gly, missense c.967_969delACA, p.Thr323del, deletion 727

F16C c.1222C>T, p.Arg408Trp, missense c.782G>A, p.Arg261Gln, missense 788

F17C c.1222C>T, p.Arg408Trp, missense c.1222C>T,p.Arg408Trp, missense 849

F18C c.782G>A, p.Arg261Gln, missense c.1169G>A, p.Glu390Gly, missense 727

F19C c.143T>C, p.Leu48Ser, missense c.842C>T, p.Pro281Leu, missense 606

F20C c.168+5G>C, splicing c.676C>A, p.Gln226Lys, missense 1212

F21C c.143T>C, p.Leu48Ser, missense c.1222C>T, p.Arg408Trp, missense 546

F22C c.143T>C, p.Leu48Ser, missense c.143T>C, p.Leu48Ser, missense 261

F23C c.782G>A, p.Arg261Gln, missense c.1066–11G>A, splicing 727

F24C c.842+4A>G, splicing no mutation detected 303

F25C c.1222C>T, p.Arg408Trp, missense c.1222C>T, p.Arg408Trp, missense 455

F26C c.143T>C, p.Leu48Ser, missense c.1222C>T, p.Arg408Trp, missense 485

F27C c.143T>C, p.Leu48Ser, missense c.782G>A, p.Arg261Gln, missense 1091

F28C c.143T>C, p.Leu48Ser, missense c.916A>G, p.Ile306Val, missense 267

F29C c.1222C>T, p.Arg408Trp, missense c.1169G>A, p.Glu390Gly, missense 364

F30C c.1066–11G>A, splicing c.1066–11G>A, splicing 424

F31C c.1066–11G>A, splicing c.1066–11G>A, splicing 1273

F32C c.143T>C, p.Leu48Ser, missense c.472C>T, p.Arg158Trp, missense 788

(continued on next page)

186 C. Gemperle-Britschgi et al. / Gene 576 (2016) 182–188

Table 1 (continued)

F33C c.842C>T, p.Pro281Leu, missense c.533A>G, p.Glu178Gly, missense, 333

F34C c.1222C>T, p.Arg408Trp, missense c.754C>T, p.Arg252Trp, missense 491

F35C c.735G>A, p.Val245Ala, missense c.472C>T, p.Arg158Trp, missense 242

F36C c.1222C>T, p.Arg408Trp, missense c.1222C>T, p.Arg408Trp, missense 1818

F38C c.1066–11G>A, splicing c.1315+1G>T, splicing 1091

F39C c.1222C>T, p.Arg408Trp, missense c.1222C>T, p.Arg408Trp, missense 758

F40C c.754C>T, p.Arg252Trp, missense c.842C>T, p.Pro281Leu, missense 1515 *

F41C c.1089delG, p.Lys363Asnfs*37, nonsense c.1089delG, p.Lys363Asnfs*37, nonsense 2121

F42C c.143T>C, p.Leu48Ser, missense c.1222C>T, p.Arg408Trp, missense 1455 *

F43C c.1222C>T, p.Arg408Trp, missense c.1222C>T, p.Arg408Trp, missense 667

F44C c.168+5G>C, splicing c.1222C>T, p.Arg408Trp, missense 1212 *

F45C c.1222C>T, p.Arg408Trp, missense c.842+4A>G, splicing 1697 *

F46C c.842C>T, p.Pro281Leu, missense c.509+4140_706+510del7831 1697 *

F47C c.1222C>T, p.Arg408Trp, missense c.1315+1G>T, splicing 909 *

F48C c.1222C>T, p.Arg408Trp, missense c.1222C>T, p.Arg408Trp, missense 1394 *

F49C c.1222C>T, p.Arg408Trp, missense c.1089delG, p.Lys363Asnfs*37, nonsense 1515 *

F50C c.1222C>T, p.Arg408Trp, missense c.592_613del22, p.Tyr198Serfs*136, nonsense 849 *

F51C c.1222C>T, p.Arg408Trp, missense c.1222C>T, p.Arg408Trp, missense 1818

F52C c.782G>A, p.Arg261Gln, missense c.842C>T, p.Pro281Leu, missense 1121 *

F53C c.143T>C, p.Leu48Ser, missense c.1066–11G>A, splicing 546 *

F54C c.1222C>T, p.Arg408Trp, missense c.673C>A, p.Pro225Thr, missense 1394

F55C c.1222C>T, p.Arg408Trp, missense c.1222C>T, p.Arg408Trp, missense 1637

F57C c.143T>C, p.Leu48Ser, missense c.1241A>G, p.Tyr414Cys, missense 1091

F58C c.1222C>T, p.Arg408Trp, missense c.1222C>T, p.Arg408Trp, missense 1333 *

F59C c.1222C>T, p.Arg408Trp, missense c.509+4140_706+510del7831 2909 *

F60C c.1222C>T, p.Arg408Trp, missense c.1222C>T, p.Arg408Trp, missense 1273 *

F61C c.782G>A, p.Arg261Gln, missense c.842C>T, p.Pro281Leu, missense 1333 *

F63C c.1222C>T, p.Arg408Trp, missense c.1222C>T, p.Arg408Trp, missense 1515

F64C c.1315+1G>T, splicing c.842+5G>A, splicing 1394

F65C c.1222C>T, p.Arg408Trp, missense c.509+4140_706+510del7831 1091

F66C c.1222C>T, p.Arg408Trp, missense c.1222C>T, p.Arg408Trp, missense 1273 *

F67C c.782G>A, p.Arg261Gln, missense c.1222C>T, p.Arg408Trp, missense 2121

F68C c.143T>C, p.Leu48Ser, missense c.1222C>T, p.Arg408Trp, missense 1939

F69C c.1222C>T, p.Arg408Trp, missense c.509+4140_706+510del7831 1333

 C. Gemperle-Britschgi et al. / Gene 576 (2016) 182–188 187

Table 1 (continued)

F70C1 c.1222C>T, p.Arg408Trp, missense c.509+4140_706+510del7831 1030 *

F70C2 c.1222C>T, p.Arg408Trp, missense c.509+4140_706+510del7831 1091 *

F70C3 c.1222C>T, p.Arg408Trp, missense c.509+4140_706+510del7831 1091 *

F71C c.1222C>T, p.Arg408Trp, missense c.1222C>T, p.Arg408Trp, missense 1546 *

F72C c.1222C>T, p.Arg408Trp, missense c.1222C>T, p.Arg408Trp, missense 1333 *

F73C1 c.1222C>T, p.Arg408Trp, missense c.509+4140_706+510del7831 739 *

F73C2 c.1222C>T, p.Arg408Trp, missense c.509+4140_706+510del7831 666 *

F74C c.1222C>T, p.Arg408Trp, missense c.509+4140_706+510del7831 1818

F75C c.1222C>T, p.Arg408Trp, missense c.509+4140_706+510del7831 849

F76C c.782G>A, p.Arg261Gln, missense c.1315+1G>T, splicing 1818

F77C c.1089delG, p.Lys363Asnfs*37, nonsense c.1315+1G>T, splicing 1273 *

F78C c.441+5G>T, splicing c.441+5G>T, splicing 1212

F79C c.1222C>T, p.Arg408Trp, missense c.1238G>C, p.Arg413Pro, missense 788 *

F80C c.1222C>T, p.Arg408Trp, missense c.1222C>T, p.Arg408Trp, missense 1576

F81C c.143T>C, p.Leu48Ser, missense c.1222C>T, p.Arg408Trp, missense 1576 *

* l–Phe level at diagnosis, i.e. no newborn screening

Key: No mutation detected

Novel mutation

history of PKU in European population, although the exact nature of this
effect remains unknown (for a broader discussion of these topics, see
Zschocke, 2003 and Krawczak and Zschocke, 2003).
Besides the novel deletion two new missense mutations were iden-
tified, c.676CN A (p.Gln226Lys) in exon 6 and c.803ANG (p.Tyr268Cys)
in exon 7 (see Table 1). The potential functional effects of the two
novel PAH variants identified in the PKU-affected individuals were
predicted via SIFT, PolyPhen2, Mutation Taster, and PROVEAN, where
both amino acid alterations were suggested to be intolerant, probably
damaging, disease causing and deleterious, respectively (see Table 2).
To further predict the effects of the identified amino acid substitutions
on PAH, the enzyme stability algorithm FoldX was also used for disease
causing predictions. Based on the same parameters and conditions as in
(Wettstein et al., 2014), both mutations are predicted to be destabilizing
(ΔΔG N 1 kcal/mol) with p.Tyr268Cys (ΔΔG 3.02 ± 0.03 kcal/mol) being
the more severe destabilized mutant enzyme than p.Gln226Lys (ΔΔG
1.08 ± 0.04 kcal/mol; Table 2). Furthermore, multiple mutations in
the codons of the PAH gene are not only common but also present in co-
dons His226 and Tyr268 where we identified the two new mutations.
The variants c.678GN C; p.Gln226His and c.802TNC; p.Tyr268His have
both been identified in the PAH gene of PKU patients and were both de-
scribed as catalytic missense mutations with destabilizing, deleterious
and probably damaging effects (see also http://www.biopku.org). In
summary, we conclude that both variants are disease causing by these
various function-prediction algorithms.
In summary, we have investigated the mutation spectrum for the
PAH gene in a cohort of 81 PKU patients from Romania and found
besides two new missense mutations, p.Gln226Lys and p.Tyr268Cys,

Table 2
Functional predictions of the two novel PAH variants.

Nucleotide change (exon) Amino acid change SIFTa PolyPhen2b Mutation Tasterc PROVEANd FoldXe
(ΔΔG in kcal/mol)

c.676CNA (exon 6) p.Gln226Lys Intolerant Probably damaging Disease causing Deleterious Destabilizing
(1.08 ± 0.04)
c.803ANG (exon 7) p.Tyr268Cys Intolerant Probably damaging Disease causing Deleterious Destabilizing
(3.02 ± 0.03)
a
SIFT (Sorting Intolerant From Tolerant; http://sift.bii.a-star.edu.sg/) predicts whether an amino acid substitution affects protein function based on sequence homology and the physical
properties of amino acids (Kumar et al., 2009).
b
PolyPhen2 (Polymorphism Phenotyping v2; http://genetics.bwh.harvard.edu/pph2/) is a tool which predicts possible impact of an amino acid substitution on the structure and function of
a human protein and utilizes multiple alignments of 45 vertebrate genomes (Adzhubei et al., 2010).
c
MutationTaster (http://www.mutationtaster.org) is an application for rapid evaluation of the disease-causing potential of DNA sequence alterations, and analyses comprise, among
other features, evolutionary conservation (Schwarz et al., 2010).
d
PROVEAN (Protein Variation Effect Analyzer; http://provean.jcvi.org/index.php) is a software tool which predicts whether an amino acid substitution or indel has an impact on the
biological function of a protein (Choi et al., 2012).
e
FoldX values were based on the same parameters and conditions as in (Wettstein et al., 2014). Variants leading to a change of ΔΔG N1 kcal/mol are predicted to be destabilizing.
188 C. Gemperle-Britschgi et al. / Gene 576 (2016) 182–188
the novel but common deletion EX6del7831 which might originate
from a Romanian founder.
Conflict of interest
The authors declare no conflict of interest.
Acknowledgments
We thank Anahita Rassi and Jacqueline Marti for pterin and DHPR
analyses on dried blot spots, JoAnne Locher for editorial help, and Drs.
Jarl Underhaug and Aurora Martinez for performing the FoldX analysis.
R.V. was supported by a grant from the Romanian National Authority
for Scientific Research (151-2005) and grant CNCS–UEFISCDI, project
number PNII-ID-PCCE-2011-2-0045.
References
Abadie, V., Jaruzelska, J., Lyonnet, S., Millasseau, P., Berthelon, M., Rey, F., Munnich, A., Rey,
J., 1993. Illegitimate transcription of the phenylalanine hydroxylase gene in lympho-
cytes for identification of mutations in phenylketonuria. Hum. Mol. Genet. 2, 31–34.
Adzhubei, I.A., Schmidt, S., Peshkin, L., Ramensky, V.E., Gerasimova, A., Bork, P.,
Kondrashov, A.S., Sunyaev, S.R., 2010. A method and server for predicting damaging
missense mutations. Nat. Methods 7, 248–249.
Arai, N., Narisawa, K., Hayakawa, H., Tada, K., 1982. Hyperphenylalaninemia due to
dihydropteridine reductase deficiency: diagnosis by enzyme assays on dried blood
spots. Pediatrics 70, 426–430.
Blau, N., Van Spronsen, F.J., 2014. In: Blau, N., Duran, M., Michael Gibson, K. (Eds.), Disor-
ders of Phenylalanine and Tetrahydrobiopterin metabolism. Physicans's Guide to the
Diagnosis, Treatment, and Follow-Up of Inherited Metabolic Diseases. Carlo Dionisi-
Vici Springer Heidelberg, New York, pp. 3–21.
Blau, N., van Spronsen, F.J., Levy, H.L., 2010. Phenylketonuria. Lancet 376, 1417–1427.
Blau, N., Shen, N., Carducci, C., 2014. Molecular genetics and diagnosis of phenylketonuria:
state of the art. Expert. Rev. Mol. Diagn. 14, 655–671.
Bodamer, O.A., Hoffmann, G.F., Lindner, M., 2007. Expanded newborn screening in Europe
2007. J. Inherit. Metab. Dis. 30, 439–444.
Cao, Y.Y., Qu, Y.J., Song, F., Zhang, T., Bai, J.L., Jin, Y.W., Wang, H., 2014. Fast clinical molec-
ular diagnosis of hyperphenylalaninemia using next-generation sequencing-based
on a custom AmpliSeq panel and ion torrent PGM sequencing. Mol. Genet. Metab.
113, 261–266.
Choi, Y., Sims, G.E., Murphy, S., Miller, J.R., Chan, A.P., 2012. Predicting the functional effect
of amino acid substitutions and indels. PLoS One 7, e46688.
Dobrowolski, S.F., Heintz, C., Miller, T., Ellingson, C., Ellingson, C., Ozer, I., Gokcay, G.,
Baykal, T., Thony, B., Demirkol, M., Blau, N., 2011. Molecular genetics and impact of
residual in vitro phenylalanine hydroxylase activity on tetrahydrobiopterin respon-
siveness in Turkish PKU population. Mol. Genet. Metab. 102, 116–121.
Donlon, J., Levy, H. and Scriver, C.R., 2010. Hyperphenylalaninemia: phenylalanine
hydroxylase deficiencies. In: The Online Metabolic and Molecular Bases of Inherited
Diseases (McGraw-Hill, New York), eds: Valle D., Beaudet, A., Vogelstein, B., Kinzler,
K., Antonorakis, S., Ballabio, S. eds: Scriver, CR, Child, B., Sly, WS, (emeritus Eds.)
Gu, Y., Lu, K., Yang, G., Cen, Z., Yu, L., Lin, L., Hao, J., Yang, Z., Peng, J., Cui, S., Huang, J., 2014.
Mutation spectrum of six genes in Chinese phenylketonuria patients obtained
through next-generation sequencing. PLoS One 9, e94100.
Guldberg, P., Romano, V., Ceratto, N., Bosco, P., Ciuna, M., Indelicato, A., Mollica, F., Meli, C.,
Giovannini, M., Riva, E., et al., 1993. Mutational spectrum of phenylalanine hydroxy-
lase deficiency in Sicily: implications for diagnosis of hyperphenylalaninemia in
southern Europe. Hum. Mol. Genet. 2, 1703–1707.
Guldberg, P., Henriksen, K.F., Sipila, I., Guttler, F., de la Chapelle, A., 1995. Phenylketonuria
in a low incidence population: molecular characterisation of mutations in Finland.
J. Med. Genet. 32, 976–978.
Kostandyan, N., Britschgi, C., Matevosyan, A., Oganezova, A., Davtyan, A., Blau, N.,
Steinmann, B., Thony, B., 2011. The spectrum of phenylketonuria genotypes in the
Armenian population: identification of three novel mutant PAH alleles. Mol. Genet.
Metab. 104, S93–S96.
Krawczak, M., Zschocke, J., 2003. A role for overdominant selection in phenylketonuria?
Evidence from molecular data. Hum. Mutat. 21, 394–397.
Kumar, P., Henikoff, S., Ng, P.C., 2009. Predicting the effects of coding non-synonymous
variants on protein function using the SIFT algorithm. Nat. Protoc. 4, 1073–1081.
Opladen, T., Abu Seda, B., Rassi, A., Thony, B., Hoffmann, G.F., Blau, N., 2011. Diagnosis of
tetrahydrobiopterin deficiency using filter paper blood spots: further development
of the method and 5 years experience. J. Inherit. Metab. Dis. 34, 819–826.
Popescu, T., Blazkova, M., Kozak, L., Jebeleanu, G., Popescu, A., 1998a. Mutation spectrum
and phenylalanine hydroxylase RFLP/VNTR background in 44 Romanian phenylke-
tonuric alleles. Hum. Mutat. 12, 314–319.
Popescu, T., Popescu, A., Kozak, L., Blazkova, M., Jebeleanu, G., 1998b. R252W/E178G phe-
nylalanine hydroxylase genotype in a Romanian non-PKU hyperphenylalaninaemic
patient. Acta Paediatr. 87, 1103–1104.
Schouten, J.P., McElgunn, C.J., Waaijer, R., Zwijnenburg, D., Diepvens, F., Pals, G., 2002. Rel-
ative quantification of 40 nucleic acid sequences by multiplex ligation-dependent
probe amplification. Nucleic Acids Res. 30, e57.
Schwarz, J.M., Rodelsperger, C., Schuelke, M., Seelow, D., 2010. MutationTaster evaluates
disease-causing potential of sequence alterations. Nat. Methods 7, 575–576.
Trujillano, D., Perez, B., Gonzalez, J., Tornador, C., Navarrete, R., Escaramis, G., Ossowski, S.,
Armengol, L., Cornejo, V., Desviat, L.R., Ugarte, M., Estivill, X., 2014. Accurate molecular di-
agnosis of phenylketonuria and tetrahydrobiopterin-deficient hyperphenylalaninemias
using high-throughput targeted sequencing. Eur. J. Hum. Genet. 22, 528–534.
Vulturar, R., Barecki, M., 2010. Molecular and phenotypical aspects in a group of
Romanian patients with phenylketonuria. J. Inherit. Metab. Dis. 33, S115.
Vulturar, R., Benga, I., Benga, G., 2005. Biochemical phenotypes in patients with HPA/PKU
established by videodensitometry original method on 2D-TLC plates. J. Inherit. Metab.
Dis. 28, 27.
Wadman, S.K., de Jonge, H.F., de Bree, P.K., 1969. Rapid, high-resolution, two-dimensional
amino acid chromatography on micro scale chromatograms. Clin. Chim. Acta 25,
87–90.
Werner, E.R., Blau, N., Thony, B., 2011. Tetrahydrobiopterin: biochemistry and pathophys-
iology. Biochem. J. 438, 397–414.
Wettstein, S., Underhaug, J., Perez, B., Marsden, B.D., Yue, W.W., Martinez, A., Blau, N.,
2014. Linking genotypes database with locus-specific database and genotype–pheno-
type correlation in phenylketonuria. Eur. J. Hum. Genet. 23, 302–309.
Zschocke, J., 2003. Phenylketonuria mutations in Europe. Hum. Mutat. 21, 345–356.