Osteoporos Int (2016) 27:3607–3613

DOI 10.1007/s00198-016-3709-1

ORIGINAL ARTICLE

DNA sequence analysis in 598 individuals with a clinical

diagnosis of osteogenesis imperfecta: diagnostic yield
and mutation spectrum
G. Bardai 1 & P. Moffatt 1 & F. H. Glorieux 1 & F. Rauch 1

Received: 2 April 2016 / Accepted: 13 July 2016 / Published online: 11 August 2016
# International Osteoporosis Foundation and National Osteoporosis Foundation 2016

Abstract SERPINF1 (4.0 % of individuals with moderate to severe OI)

Summary We detected disease-causing mutations in 585 of and CRTAP (2.9 %).
598 individuals (98 %) with typical features of osteogenesis Conclusions DNA sequence analysis of currently known OI-
imperfecta (OI). In mild OI, only collagen type I encoding associated genes identifies disease-causing variants in almost
genes were involved. In moderate to severe OI, mutations in all individuals with a typical OI phenotype. About 20 % of
12 different genes were found; 11 % of these patients had individuals with moderate to severe OI had mutations in genes
mutations in recessive genes. other than COL1A1/COL1A2.
Introduction OI is usually caused by mutations in COL1A1 or
COL1A2, the genes encoding collagen type I alpha chains, but
Keywords Children . Fractures . Mutations .
mutations in at least 16 other genes have also been associated
Next-generation sequencing . Osteogenesis imperfecta
with OI. It is presently unknown what proportion of individ-
uals with clinical features of OI has a disease-causing muta-
tion in one of these genes.
Methods DNA sequence analysis was performed on 598 in- Introduction
dividuals from 487 families who had a typical OI phenotype.
OI type I was diagnosed in 43 % of individuals, and 57 % had Osteogenesis imperfecta (OI) is a heritable connective tissue
moderate to severe OI, defined as OI types other than type I. disorder that is mainly characterized by bone fragility. The
Results Disease-causing variants were detected in 97 % of typical clinical picture also includes extraskeletal findings
individuals with OI type I and in 99 % of patients with mod- such as blue or gray discoloration of the sclera and tooth
erate to severe OI. All mutations found in OI type I were abnormalities called dentinogenesis imperfecta [1].
dominant and exclusively affected COL1A1 or COL1A2. In The severity of bone fragility in OI varies widely. This is
moderate to severe OI, dominant mutations were found in captured in the traditional classification that distinguishes four
COL1A1/COL1A2 (77 %), IFITM5 (9 %), and P4HB phenotypic categories (OI types I to IV) [2]. OI type I repre-
(0.6 %). Mutations in one of the recessive OI-associated gene sents the mildest phenotype, OI type II is lethal in the neonatal
were observed in 12 % of individuals with moderate to severe period, OI type III is the most severe form of the disease in
OI. The genes most frequently involved in recessive OI were individuals surviving the neonatal period, and OI type IV is of
moderate severity [3]. Three more phenotypic OI types (V, VI,
Electronic supplementary material The online version of this article and VII) have been described, based on specific clinical char-
(doi:10.1007/s00198-016-3709-1) contains supplementary material,
acteristics [4]. More recently, a large number of additional OI
which is available to authorized users.
types (OI type VIII and higher) have been proposed on the
* F. Rauch
basis of genetic findings [1] and are listed in the OMIM data-
[email protected] base. However, the addition of genotypically defined OI types
to a phenotypic classification is controversial [5]. The 2015
1
Shriners Hospital for Children and McGill University, 1003 Decarie, Nosology and Classification of Genetic Skeletal Disorders
Montreal H3G 1A6, Québec, Canada exclusively uses phenotypic criteria to classify OI types [3].
3608 Osteoporos Int (2016) 27:3607–3613

It has been known for more than 30 years that OI pheno- OI type I, III, IV, V, VI, VII, or Cole-Carpenter syn-
types are most often caused by dominant alterations in drome [4]. Given that the study was performed at a
COL1A1 (MIM 120150) or COL1A2 (MIM 120160), the center for pediatric orthopedics, the study population
genes coding for collagen type I alpha chains [1]. The molec- did not include patients with OI type II, which leads
ular diagnosis of OI was initially performed by examining to death in the neonatal period.
collagen type I protein from skin fibroblasts [6] and later by The 598 individuals reported here were residents of a vari-
Sanger sequencing of COL1A1 and COL1A2. Both ap- ety of countries and geographic regions: Canada (n = 370),
proaches have a high diagnostic yield in OI with typical phe- USA (n = 154), Latin America (n = 44), Europe (n = 16),
notypes. For example, protein analysis in 132 individuals with Middle East (n = 10), and others (n = 4). The ethnic distribu-
Btypical nonlethal OI^ identified biochemical abnormalities in tion was as follows: Caucasian (n = 457), Hispanic (n = 68),
87 % of cases [7]. In the most detailed Sanger sequencing Arab (n = 27), Asian (n = 25), First Nations (n = 13), and
study to date, pathogenic COL1A1/COL1A2 sequence alter- others (n = 8).
ations were detected in 87 % of 142 children with Btypical OI^ The spectrum of disease severity varies between OI cen-
[8]. However, that study also showed that the diagnostic yield ters, which makes it difficult to compare the spectrum of mo-
varied with the phenotypic group. Pathogenic variants were lecular diagnosis results between studies. At the same time,
found in 94 % of individuals with OI type I, in 88 % of the diagnosis of specific OI types is somewhat subjective and
patients with OI type III but in only 63 % of children with can depend on the amount of available clinical information
OI type IV. (e.g., the diagnosis of OI type VI requires histological analysis
Starting in 2006, defects in at least 16 genes other than of bone tissue) or on the age of the patient (e.g., hyperplastic
COL1A1 and COL1A2 have been linked to OI phenotypes callus may not be evident in young children with OI type V).
[1]. Most of these gene defects lead to recessive forms of For easier comparison with other studies, we separated the
OI, even though two genes (IFITM5 [MIM 614757], P4HB present study cohort into two broad groups—Bmild OI^ (OI
[MIM 176790]) are associated with dominant OI, and defects type I) and Bmoderate to severe OI^ (all other OI types). The
in one gene (PLS3, [MIM 300131]) lead to X-linked bone main distinguishing characteristic between these two catego-
fragility. Inclusion of these Bnew^ genes in OI sequencing ries is the presence of lower extremity deformities in the mod-
panels can be expected to increase the clinical sensitivity of erate to severe OI group. The OI type I group only included
molecular diagnosis in OI, but it is presently not known what individuals who had Btypical extraskeletal features^ of OI.
proportion of individuals with a typical OI phenotype have This means that they had either discoloration of sclera or
identifiable pathogenic variants in one of the currently known dentinogenesis imperfecta or both. In the absence of such
OI genes. It is also not clear how frequently pathogenic vari- extraskeletal features, it can be difficult to distinguish OI from
ants in each of the newly discovered OI-associated genes other causes of increased fracture incidence in children and
cause the disease. Such data are needed to assess the scope adolescents. The diagnosis and clinical information were ob-
for further improvements in the molecular diagnosis of OI and tained by retrospective chart review.
will also be useful for estimating the number of individuals The study was approved by the Institutional Review Board
that might benefit from the future development of gene- of McGill University. For individuals in whom a disease-
specific treatment approaches. causing mutation had been found as part of the clinical diag-
In the present study, we therefore assessed the results of nostic workup, the present study was limited to a retrospective
molecular testing for the currently known OI genes in 598 chart review and no consent was required. Individuals in
individuals who had a typical OI phenotype and who had all whom Sanger sequencing had previously failed to detect a
been assessed at a single center. disease-causing variant were invited to have next-generation
sequencing (NGS) on a research basis. In these individuals,
consent was obtained if they were aged 14 years or older. For
Subjects and methods individuals below 18 years of age, parental consent was ob-
tained, as well as assent from participants over 8 years of age.
Subjects

The study population comprised all individuals with a Clinical analyses

typical OI phenotype who were evaluated at the
Shriners Hospital for Children in Montreal between Height was measured using a Harpenden stadiometer
December 2000 and February 2015 and in whom mo- (Holtain, Crymych, UK). Height and weight were con-
lecular diagnosis by sequence analysis of genomic DNA verted to age- and sex-specific Z-scores on the basis of
had been completed. One of the authors (F.G. or F.R.) reference data published by the Centers for Disease
assessed each patient clinically and made a diagnosis of Control and Prevention [9].
Osteoporos Int (2016) 27:3607–3613 3609

Sequence analysis sequencing run were processed using Torrent Suite software
(version 5.04; Life Technologies) for base calls, read align-
Sanger sequencing (Applied Biosystems 3100 DNA sequenc- ments, and variant calling using the reference genomic se-
er) was used until June 2013 for sequence analyses in genomic quence (hg19) of target genes. Called variants were annotated
DNA after PCR amplification of all exons and exon/intron using Ion Reporter (version 5.0). Variants were filtered out if
boundaries of target genes. Until 2007, sequence analysis they were present in the UCSC common SNPs database
was limited to those COL1A1 and COL1A2 exons, which (http://genome.ucsc.edu), which includes SNPs from dbSNP
were positive on heteroduplex screening [10]; later, all 138 that have a minor allele frequency equal to or higher than
COL1A1 and COL1A2 exons were sequenced without prior 1 %. Variants were also filtered out if they had been observed
screening step. Sequencing of CRTAP (2007), P3H1 (2008), in 4 or more of 45 control samples that had previously been
SERPINF1 (2011), and IFITM5 (2012) was added to the sequenced with the same methodology.
Sanger sequencing program, after variants in these genes The standards and guidelines of the American College of
had been identified as causes of OI. Medical Genetics and Genomics and the Association for
After June 2013, NGS was performed as the primary sequenc- Molecular Pathology were used to classify variants as patho-
ing methodology, using an Ion Torrent PGM device (Life genic or likely pathogenic [12]. Samples where no pathogenic
Technologies), as described [11]. Initially, a metabolic bone gene or likely pathogenic variant was found in the initial panel were
panel was assessed which included the following OI-related reanalyzed using the expanded panel. Mutations found by
genes: COL1A1 (NM_000088.3), COL1A2 (NM_000089.3), NGS were confirmed by Sanger sequencing.
B M P 1 ( N M _ 0 0 11 9 9 . 3 , N M _ 0 0 6 1 2 9 . 4 ) , C R TA P
(NM_006371.4), FKBP10 (NM_021939.3), IFITM5 Statistical analyses
(NM_001025295.2), P3H1 (NM_022356.3), PLOD2
(NM_000935.2), PPIB (NM_000942.4), SERPINF1 Differences between two groups were tested for significance
(NM_002615.5), SERPINH1 (NM_001235.3), SP7 using the unpaired Student’s t test. Differences between three
(NM_152860.1), and TMEM38B (NM_018112.2). groups were evaluated by analysis of variance (ANOVA). All
Subsequently, the panel was expanded to include additional genes tests were two-tailed, and throughout the study, P values
that are associated with OI or bone fragility disorders with a sim- <0.05 were considered significant. These calculations were
ilar phenotype (ALPL (NM_000478.4), CREB3L1 performed using the SPSS Statistics software version 19.0
(NM_052854.3), P4HB (NM_000918.3), PLS3 (SPSS Inc., Chicago, Illinois, USA).
(NM_001136025.4), RUNX2 (NM_001024630.3), SEC24D
(NM_014822.2), SPARC (NM_003118.3), and WNT1
(NM_005430.3). Results
A total of 10 ng DNA/sample were used for target enrich-
ment by multiplex PCR and sequencing on an Ion 316 Chip The study included 598 individuals with a clinical diagnosis
(Ion PGM Hi-Q Sequencing Kit; Life Technologies), which of OI who were from 487 families. OI type I was diagnosed in
was performed following the manufacturer’s instructions with 43 % of the study population, and 57 % were classified as
the 200-bp single-end run configuration. Data from the having one of the moderate to severe OI types (Table 1).

Table 1 Clinical characteristics of the study population

Mild OI Moderate to severe OI Total

I III IV V VI VII CC

N individuals (M/F) 256 (123/133) 100 (47/53) 189 (88/101) 30 (17/13) 14 (11/3) 7 (1/6) 2 (2/0) 598 (289/309)
N families 170 99 179 25 11 1 2 487
Family history (N pos./N total) 177/240 8/92 40/172 11/29 3/14 7/7 0/2 246/556
Age (years) 17.2 (11.4) 13.2 (6.8) 13.0 (7.4) 17.0 (10.8) 15.9 (9.7) 13.9 (9.6) 19.8 (1.5) 15.1 (9.6)
Height (Z-score) −1.0 (1.1) −7.9 (2.4) −3.9 (2.1) −3.3 (2.3) −4.6 (3.4) −2.7 (3.3) −10.1 (2.4) −3.4 (3.1)
Weight (Z-score) −0.2 (1.2) −2.9 (1.6) −1.4 (1.6) −1.2 (1.6) −1.4 (1.5) 1.2 (2.3) −4.4 (1.2) −1.1 (1.8)
Blue sclera (N pos./N total) 241/255 66/94 118/179 2/30 0/14 0/7 0/2 427/579
DI (N pos./N total) 32/249 79/93 105/173 0/30 0/14 0/7 1/2 217/569

Results represent N or mean (SD)

Abbreviations: CC Cole-Carpenter syndrome, OI osteogenesis imperfecta
3610 Osteoporos Int (2016) 27:3607–3613

Caucasians made up 86 % of the OI type I cohort and 70 % of exon 3 was <20. Mutations in this exon were excluded by
the moderate to severe OI group. The family history, where Sanger sequencing.
available, was positive for OI in 74 % of individuals with OI All mutations found in OI type I were dominant and exclu-
type I and in 22 % of individuals with moderate to severe OI. sively affected COL1A1 or COL1A2 (Fig. 1). COL1A1
In 518 individuals, Sanger sequencing had been performed haploinsufficiency (i.e., nonsense or frameshift) mutations
as the initial diagnostic methodology; pathogenic or likely were found in 126 of these patients, splice site mutations in
pathogenic variants were found in 483 of these subjects. In 55 patients (COL1A1, n = 52; COL1A2, n = 3), triple-helical
80 individuals, NGS was used as the initial diagnostic meth- glycine substitutions in 48 patients (COL1A1, n = 22;
odology. NGS was also performed on the 35 individuals in COL1A2, n = 26) and other types of COL1A1/COL1A2 muta-
whom Sanger sequencing had not detected a pathogenic mu- tions in 17 patients.
tation. This resulted in the detection of disease-causing vari- In moderate to severe OI, dominant mutations were found
ants in 27 individuals, of whom 6 had disease-causing variants in COL1A1/COL1A2, in IFITM5 (c.-14C > T in all individ-
in genes that had not been assessed by Sanger sequencing. uals) [14], and in P4HB (c.1178A > G in two individuals with
The other 21 individuals had mutations in COL1A1 or Cole-Carpenter syndrome) [15]. Mutations in one of the re-
COL1A2 that had not been identified by previous Sanger se- cessive OI-associated gene were observed in 40 individuals,
quencing for a variety of reasons, such as presence of a dele- who all had a moderate to severe phenotype (Table 2; Fig. 1).
tion of the entire COL1A1 gene [13] or polymorphisms at the The genes most frequently involved in recessive OI were
annealing site of Sanger sequencing primers. SERPINF1 and CRTAP. In two individuals with moderate to
Overall, disease-causing variants were detected in 97 % of severe OI, the sequencing result was classified as
individuals with OI type I and in 99 % of patients with moderate Binconclusive,^ because only one heterozygous mutation
to severe OI (Table 2). The 13 patients in whom sequence analysis was found in a gene associated with recessive OI
did not reveal a disease-causing variant were from 13 unrelated (SERPINF1, N = 1; SEC24D, N = 1).
families. The mean read depth in the samples from these 13 indi- The number of unique mutations per gene was 188 in
viduals was 744 (range 428 to 1196), similar to the samples where COL1A1 (66 haploinsufficiency, 52 triple-helical glycine sub-
a disease-causing variant was found (799 (range 367 to 1465)). In stitutions, 49 splice site, 16 C-propeptide, 5 other), 106 in
nine of the samples with negative NGS, coverage of COL1A2 COL1A2 (92 triple-helical glycine substitutions, 8 splice-site,

Table 2 Relationship between

clinical OI type and affected gene Mild OI Moderate to severe OI All moderate All
to severe OI
I III IV V VI VII CC

Dominant
COL1A1 219 45 90 – – – – 135 354
COL1A2 29 45 85 – – – – 130 159
IFITM5 – – – 30 – – – 30 30
P4HB – – – – – – 2 2 2
Recessive
CRTAP – 3 – – – 7 – 10 10
FKBP10 – – 3 – – – – 3 3
P3H1 – 5 1 – – – – 6 6
PLOD2 – 1 – – – – – 1 1
PPIB – – 1 – – – – 1 1
SERPINF1 – 1 – – 13 – – 14 14
SPARC – – 1 – – – – 1 1
WNT1 – – 4 – – – – 4 4
Inconclusive – – 1 – 1 – – 2 2
No pathogenic variant 8 – 3 – – – – 3 11
All 256 100 189 30 14 7 2 342 598
Positive 248 100 185 30 13 7 2 337 585

Values are N
Abbreviations: CC Cole-Carpenter syndrome, OI osteogenesis imperfecta
Osteoporos Int (2016) 27:3607–3613 3611

Fig. 1 Frequency of disease-

causing variants according to af-
fected gene

6 other), 1 in IFITM5, 1 in P4HB, 3 in CRTAP, 5 in FKBP10, 5 individuals with moderate to severe OI, but were not seen in
in P3H1, 2 in PLOD2, 2 in PPIB, 7 in SERPINF1, 1 in the context of mild OI.
SPARC, and 4 in WNT1 (Supplemental Tables 1 and 2). The present study was conducted in a specialized treatment
Overall, 49 of these mutations were not listed in the OI variant center for OI, and therefore, the study population comprised a
database and were therefore considered novel. relatively large proportion of individuals with moderate to
severe OI (57 %). In recent population-based studies, moder-
ate to severe OI types represented only 31 and 23 %, respec-
tively, of all individuals with OI [8, 16]. To compare mutation
Discussion findings between studies, it is therefore useful to separately
report results for mild OI and for moderate to severe OI. It
In this study, sequence analysis of genomic DNA identified must be acknowledged, however, that the present Bmoderate
disease-causing mutations in 585 of 598 (98 %) individuals to severe OI^ group may be skewed towards the more severe
with a typical OI phenotype. The percentage of patients in end of the spectrum, given the study setting in a center that
whom a disease-causing variant could be identified was sim- receives referrals for specialized care from many countries.
ilar for OI type I and for the moderate to severe OI types, but Population-based studies would be required to determine the
the mutation spectrum differed between these two groups. In OI mutations spectrum that is not affected by the inherent
OI type I, we observed only COL1A1 and COL1A2 mutations, biases of hospital-based analyses.
whereas in moderate to severe OI, mutations in COL1A1 and It is often assumed that >90 % of individuals with OI have
COL1A2 explained the disease in less than 80 % of patients, COL1A1 and COL1A2 mutations [17, 18]. While our results
and mutations in 12 different genes were found. Apart from in OI type I are consistent with this assumption, we found that
COL1A1/COL1A2, IFITM5 was the most frequently involved COL1A1/COL1A2 mutations explained the disease in only
gene in moderate to severe OI and was responsible for the 264 of the 342 individuals (77 %) with moderate to severe
disease in 9 % of individuals with moderate to severe OI. OI. Similarly, Lindahl et al. found COL1A1/COL1A2 muta-
Mutations in recessive genes were observed in 12 % of tions in 31 of 43 children (72 %) with OI types III and IV [8].
3612 Osteoporos Int (2016) 27:3607–3613

The notion that >90 % of individuals with OI have COL1A1/ The mild OI group of the present study included only indi-
COL1A2 mutations is mainly based on linkage studies in fam- viduals who had Btypical OI type I,^ which was defined as
ilies with dominant OI [19]. However, the results of linkage having blue/gray sclerae or dentinogenesis imperfecta, or
studies in dominant OI are not directly applicable to moderate both, in addition to bone fragility. For this reason, previously
to severe OI, because these individuals typically do not have a reported patients with mutations in PLS3 [25] or in the 3′-
positive family history and recessive OI is relatively common untranslated region of the short BMP1 transcript [26] were
in this group. excluded from the present analysis, because these individuals
Overall, we detected mutations in genes other than had bone fragility without either scleral discoloration or
COL1A1/COL1A2 in about a fifth of individuals with moder- dentinogenesis imperfecta.
ate to severe OI. Close to half of the individuals with muta- Overall, we failed to detect disease-causing variants in
tions in these Bother^ genes had the dominant c.-14C > T mu- about 2 % of our study population. It is possible that in some
tation in IFITM5, which is the single most prevalent mutation of these individuals, OI is caused by variants in the genes that
in the present series. This mutation leads to the addition of 5- are included in our panel but are located in areas that were not
amino acids to the N-terminus of BRIL, the protein encoded sequenced, such as deep intronic regions. However, it is likely
by IFITM5 [20]. No other IFITM5 mutations were found in that at least some of these individuals have disease-causing
the present study. variants in genes that have not yet been associated with OI. In
Biallelic mutations in genes linked to recessive OI were any case, our study suggests that these as yet undiscovered OI-
present in 12 % of individuals with moderate to severe OI related genes explain the disease in a very small proportion of
and about 7 % of the entire OI population at our institution. individuals with OI. Almost all individuals with a typical OI
In comparison, about 10 % of all individuals included in the phenotype have pathogenic variants in one of the known OI-
OI variant database have recessive OI (accessed 16 associated genes.
March 2016). However, as observations in newly described
genes may be more likely to be reported than variants in well- Acknowledgments We thank Patty Mason for technical assistance and
known genes, the proportion of individuals with recessive OI Mark Lepik for the preparation of the figures. F.R. received salary support
in the OI variant database is likely an overestimate of the true from the Chercheur-Boursier Clinicien program of the Fonds de
prevalence of recessive OI. The prevalence of recessive OI Recherche du Québec-Santé. This study was supported by the Shriners
of North America and the Fonds de recherche Québec-Santé.
may also vary between geographic areas, as recessive disor-
ders may be more prevalent in regions where consanguinity is Roles of the authors GB performed analyses. FHG contributed patient
common. information. PM revised manuscript content. FR conceptualized the pro-
SERPINF1 and CRTAP mutations were the most common ject, contributed patient information, finalized the report, and accepts
responsibility for the integrity of the data analysis. All authors have read
causes of recessive OI in the present study. This may be influ- and approved the final version of the manuscript.
enced by the presence of founder effects in the population
served by our institution. The SERPINF1 c.295C > T Conflicts of interest Ghalib Bardai, Pierre Moffatt, Francis H
(p.Arg99*) mutation was observed in six individuals from Glorieux, and Frank Rauch declare that they have no conflict of interest.
four apparently unrelated families of French Canadian origin
Web resources Exome Aggregation Consortium (ExAC) Browser:
[21]. A homozygous intronic c.472-1021C > G CRTAP muta- http://exac.broadinstitute.org/
tion that introduces a cryptic splice was found in seven mem- Online Mendelian Inheritance in Man (OMIM), http://www.omim.org
bers of a community in Northern Quebec who all were diag- Osteogenesis Imperfecta Variant Database: https://oi.gene.le.ac.uk/
nosed with OI type VII [22, 23]. Together, these two muta- UCSC database, version hg19: http://www.genome.ucsc.edu/
tions were responsible for close to one third of all occurrences
of recessive OI in the present study.
We detected disease-causing COL1A1/COL1A2 variants in References
97 % of individuals with OI type I, which is similar to the
results of Lindahl et al. who found COL1A1/COL1A2 muta- 1. Forlino A, Marini JC (2016) Osteogenesis imperfecta. Lancet 387:
1657–1671
tions in 94 % of such individuals [8]. It is noteworthy that we
2. Sillence DO, Senn A, Danks DM (1979) Genetic heterogeneity in
observed triple-helical glycine substitutions in 19 % of pa- osteogenesis imperfecta. J Med Genet 16:101–116
tients with an OI type I phenotype. Even though heterozygous 3. Bonafe L, Cormier-Daire V, Hall C et al (2015) Nosology and
COL1A1 null mutations are sometimes thought to be the sole classification of genetic skeletal disorders: 2015 revision. Am J
cause of OI type I [5, 18, 24], our observations highlight the Med Genet A 167A:2869–2892
fact that the phenotype caused by some triple-helical glycine 4. Rauch F, Glorieux FH (2004) Osteogenesis imperfecta. Lancet 363:
1377–1385
substitutions can be clinically indistinguishable from that 5. Van Dijk FS, Sillence DO (2014) Osteogenesis imperfecta: clinical
caused by COL1A1 haploinsufficiency. Similar observations diagnosis, nomenclature and severity assessment. Am J Med Genet
have been made by Lindahl et al. [8]. A 164A:1470–1481
Osteoporos Int (2016) 27:3607–3613 3613

6. Bonadio J, Holbrook KA, Gelinas RE, Jacob J, Byers PH (1985) Carpenter syndrome is caused by a heterozygous missense mutation
Altered triple helical structure of type I procollagen in lethal peri- in P4HB. Am J Hum Genet 96:425–431
natal osteogenesis imperfecta. J Biol Chem 260:1734–1742 16. Wekre LL, Froslie KF, Haugen L, Falch JA (2010) A population-
7. Wenstrup RJ, Willing MC, Starman BJ, Byers PH (1990) Distinct based study of demographical variables and ability to perform ac-
biochemical phenotypes predict clinical severity in nonlethal vari- tivities of daily living in adults with osteogenesis imperfecta.
ants of osteogenesis imperfecta. Am J Hum Genet 46:975–982 Disabil Rehabil 32:579–587
8. Lindahl K, Astrom E, Rubin CJ, Grigelioniene G, Malmgren B, 17. van Dijk FS, Byers PH, Dalgleish R, Malfait F, Maugeri A,
Ljunggren O, Kindmark A (2015) Genetic epidemiology, preva- Rohrbach M, Symoens S, Sistermans EA, Pals G (2012) EMQN
lence, and genotype-phenotype correlations in the Swedish popu- best practice guidelines for the laboratory diagnosis of osteogenesis
lation with osteogenesis imperfecta. Eur J Hum Genet 23:1042– imperfecta. Eur J Hum Genet 20:11–19
1050 18. Pepin MG, Byers PH (2015) What every clinical geneticist should
9. Ogden CL, Kuczmarski RJ, Flegal KM, Mei Z, Guo S, Wei R, know about testing for osteogenesis imperfecta in suspected child
Grummer-Strawn LM, Curtin LR, Roche AF, Johnson CL (2002) abuse cases. Am J Med Genet C Semin Med Genet 169:307–313
Centers for Disease Control and Prevention 2000 growth charts for 19. Sykes B, Ogilvie D, Wordsworth P et al (1990) Consistent linkage
the United States: improvements to the 1977 National Center for of dominantly inherited osteogenesis imperfecta to the type I colla-
Health Statistics version. Pediatrics 109:45–60 gen loci: COL1A1 and COL1A2. Am J Hum Genet 46:293–307
10. Korkko J, Ala-Kokko L, De Paepe A, Nuytinck L, Earley J, 20. Cho TJ, Lee KE, Lee SK et al (2012) A single recurrent mutation in
Prockop DJ (1998) Analysis of the COL1A1 and COL1A2 genes the 5′-UTR of IFITM5 causes osteogenesis imperfecta type V. Am J
by PCR amplification and scanning by conformation-sensitive gel Hum Genet 91:343–348
electrophoresis identifies only COL1A1 mutations in 15 patients 21. Rauch F, Husseini A, Roughley P, Glorieux FH, Moffatt P (2012)
with osteogenesis imperfecta type I: identification of common se- Lack of circulating pigment epithelium-derived factor is a marker of
quences of null-allele mutations. Am J Hum Genet 62:98–110 osteogenesis imperfecta type VI. J Clin Endocrinol Metab 97:
11. Rauch F, Lalic L, Glorieux FH, Moffatt P, Roughley P (2014) E1550–E1556
Targeted sequencing of a pediatric metabolic bone gene panel using 22. Ward LM, Rauch F, Travers R, Chabot G, Azouz EM, Lalic L,
a desktop semiconductor next-generation sequencer. Calcif Tissue Roughley PJ, Glorieux FH (2002) Osteogenesis imperfecta type
Int 95:323–331 VII: an autosomal recessive form of brittle bone disease. Bone 31:
12. Richards S, Aziz N, Bale S et al (2015) Standards and guidelines for 12–18
the interpretation of sequence variants: a joint consensus recom- 23. Morello R, Bertin TK, Chen Y et al (2006) CRTAP is required for
mendation of the American College of Medical Genetics and prolyl 3-hydroxylation and mutations cause recessive osteogenesis
Genomics and the Association for Molecular Pathology. Genet imperfecta. Cell 127:291–304
Med 17:405–424 24. Forlino A, Cabral WA, Barnes AM, Marini JC (2011) New perspec-
13. Bardai G, Lemyre E, Moffatt P, Palomo T, Glorieux FH, Tung J, tives on osteogenesis imperfecta. Nat Rev Endocrinol 7:540–557
Ward L, Rauch F (2016) Osteogenesis imperfecta type I caused by 25. Fahiminiya S, Majewski J, Al-Jallad H, Moffatt P, Mort J, Glorieux
COL1A1 deletions. Calcif Tissue Int 98:76–84 FH, Roschger P, Klaushofer K, Rauch F (2014) Osteoporosis
14. Rauch F, Moffatt P, Cheung M, Roughley P, Lalic L, Lund AM, caused by mutations in PLS3: clinical and bone tissue characteris-
Ramirez N, Fahiminiya S, Majewski J, Glorieux FH (2013) tics. J Bone Miner Res 29:1805–1814
Osteogenesis imperfecta type V: marked phenotypic variability de- 26. Fahiminiya S, Al-Jallad H, Majewski J, Palomo T, Moffatt P,
spite the presence of the IFITM5 c.-14C > T mutation in all patients. Roschger P, Klaushofer K, Glorieux FH, Rauch F (2015) A
J Med Genet 50:21–24 polyadenylation site variant causes transcript-specific BMP1
15. Rauch F, Fahiminiya S, Majewski J, Carrot-Zhang J, Boudko S, deficiency and frequent fractures in children. Hum Mol
Glorieux F, Mort JS, Bachinger HP, Moffatt P (2015) Cole- Genet 24:516–524
Reproduced with permission of copyright owner. Further reproduction
prohibited without permission.