Added value of next generation gene panel analysis for patients with elevated methylmalonic acid and no clinical diagnosis following functional studies of vitamin B12 metabolism Mihaela Pupavac, Xia Tian, Jordan Chu, Guoli Wang, Yanming Feng, Stella Chen, Remington Fenter, Victor W. Zhang, Jing Wang, David Watkins, Lee-Jun Wong, David S. Rosenblatt PII: DOI: Reference:
S1096-7192(16)30007-5 doi: 10.1016/j.ymgme.2016.01.008 YMGME 6007
To appear in:
Molecular Genetics and Metabolism
Received date: Accepted date:
21 January 2016 22 January 2016
Please cite this article as: Pupavac, M., Tian, X., Chu, J., Wang, G., Feng, Y., Chen, S., Fenter, R., Zhang, V.W., Wang, J., Watkins, D., Wong, L.-J. & Rosenblatt, D.S., Added value of next generation gene panel analysis for patients with elevated methylmalonic acid and no clinical diagnosis following functional studies of vitamin B12 metabolism, Molecular Genetics and Metabolism (2016), doi: 10.1016/j.ymgme.2016.01.008
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Added value of next generation gene panel analysis for patients with elevated methylmalonic acid and no clinical diagnosis following functional studies of
PT
vitamin B12 metabolism Mihaela Pupavaca, Xia Tianb, Jordan Chua, Guoli Wangb, Yanming Fengb, Stella Chenb,
RI
Remington Fenterb, Victor W Zhangb,c, Jing Wangb,c, David Watkinsa, Lee-Jun Wongb,c,
SC
David S. Rosenblatta
NU
Affiliations:
Department of Human Genetics, McGill University, Montreal, Quebec, Canada
b
Baylor Miraca Genetics Laboratories, Houston, Texas, United States.
c
Department of Molecular and Human Genetics, Baylor College of Medicine, Houston,
MA
a
D
Texas, United States.
David S. Rosenblatt
AC CE P
McGill University
TE
Corresponding Author
Research Institute of the McGill University Health Centre Glen Site, 1001 Décarie Boulevard Block E, M0.2220
Montreal, Québec, H4A 3J1
[email protected] 514-934-1934 ext. 42978
1
ACCEPTED MANUSCRIPT ABSTRACT Next Generation Sequencing (NGS) based gene panel testing is increasingly available as a molecular diagnostic approach for inborn errors of metabolism. Over the past 40
PT
years patients have been referred to the Vitamin B12 Clinical Research Laboratory at McGill University for diagnosis of inborn errors of cobalamin metabolism by functional
RI
studies in cultured fibroblasts. DNA samples from patients in which no diagnosis was
SC
made by these studies were tested by a NGS gene panel to determine whether any molecular diagnoses could be made. 131 DNA samples from patients with elevated
NU
methylmalonic acid and no diagnosis following functional studies of cobalamin metabolism were analyzed using the 24 gene extended cobalamin metabolism NGS
MA
based panel developed by Baylor Miraca Genetics Laboratories. Gene panel testing identified two or more variants in a single gene in 16/131 patients. Eight patients had pathogenic findings, one had a finding of uncertain significance, and seven had benign
D
findings. Of the patients with pathogenic findings, five had mutations in ACSF3, two in
TE
SUCLG1 and one in TCN2. Thus, the NGS gene panel allowed for the presumptive
assays.
Key words:
AC CE P
diagnosis of 8 additional patients for which a diagnosis was not made by the functional
Next generation sequencing gene panel, methylmalonic acid, cobalamin metabolism, vitamin B12, diagnostic test, functional studies.
2
ACCEPTED MANUSCRIPT Conflict of interest notification The Canadian authors offer fee-based functional somatic cells testing for cobalamin
PT
disorders in a not for profit setting. The Baylor Miraca Genetics Laboratories offer extensive fee-based genetic tests including the use of next generation sequencing for
AC CE P
TE
D
MA
NU
SC
RI
molecular analyses of cobalamin metabolic pathway related disorders.
3
ACCEPTED MANUSCRIPT
1. Introduction
PT
Methylmalonic acidemia/aciduria characterizes a group of genetically
RI
heterogeneous disorders [1]. Isolated methylmalonic aciduria is most frequently
SC
associated with mutations affecting methylmalonylCoA mutase (mut), and is also associated with mutations affecting synthesis of the adenosylcobalamin cofactor
NU
required for its activity (cblA and cblB). Combined methylmalonic aciduria and
MA
homocystinuria is associated with mutations that affect synthesis of methylcobalamin as well as adenosylcobalamin (cblC, cblD, cblF, cblJ and cblX), and with mutations
D
affecting intestinal uptake or blood transport of cobalamin (intrinsic factor deficiency,
TE
Imerslund-Gräsbeck syndrome, and transcobalamin deficiency). Elevations in serum
AC CE P
methylmalonic acid (MMA) are seen in patients with mutations affecting the subunits of succinyl-CoA synthetase (SUCLA2, SUCLG1). Mildly elevated MMA occurs in patients with mutations affecting methylmalonyl-CoA epimerase (MCEE) [2, 3]; and in patients with mutations affecting the ACSF3 gene [4, 5], in which malonic acid levels may be elevated as well.
For the past 40 years the Vitamin B12 Clinical Research Laboratory at McGill University has been one of two international referral centers for the diagnosis of inborn errors of vitamin B12 metabolism, using functional studies of cobalamin metabolism in cultured patient skin fibroblasts (Table 1) [6]. Patients with suspected inborn errors of cobalamin metabolism are referred because of an elevation of MMA, homocysteine, or both in their blood and/or urine. Although the underlying disorder could be identified in the majority of the referred patients, there remained a number of patients with no 4
ACCEPTED MANUSCRIPT identified cause of disease. This may be because a potentially causal gene is not expressed in cultured skin fibroblasts (GIF, AMN, CUBN) or, in the case of ACSF3, the
PT
mutations do not result in any changes in the parameters tested at the McGill
RI
laboratory. Mutations affecting other gene products (CD320, SUCLG1, SUCLA2, TCN2) can be detected in fibroblasts, but the tests required to identify these disorders are not
SC
part of the standard work-up for patient cells at the McGill laboratory.
NU
In the past, molecular diagnosis of disorders resulting in methylmalonic
MA
acidemia/aciduria required individual candidate gene sequencing to identify pathogenic variants in suspected causal genes. Recently, Baylor Miraca Genetics Laboratories
D
developed an extended cobalamin metabolism gene panel that contains 24 genes
TE
known to cause elevated MMA, homocysteine, or both (Table 2). DNA from 131 patients referred to the McGill University center, in which no diagnosis had been established
AC CE P
after routine functional studies of cobalamin metabolism, was tested with the Baylor Miraca Genetics Laboratories Cbl pathway related gene panel. These studies demonstrated that the NGS panel allowed presumptive diagnosis of a number of patients that had not been diagnosed by functional studies, and that the functional studies had not failed to diagnose any patients with biallelic mutations in any of the inborn errors of cobalamin that they were designed to identify. 2. Material and Methods 2.1. Patients Genomic DNA was extracted from 131 fibroblast lines from patients referred to the diagnostic laboratory at the Vitamin B12 Clinical Research Laboratory (Division of 5
ACCEPTED MANUSCRIPT Medical Genetics, Department of Medicine, McGill University Health Centre) using the FlexiGene DNA Kit (Qiagen, Canada). These patients were referred to the McGill
PT
laboratory because of elevated levels of methylmalonic acid in blood or urine, but they
RI
remained undiagnosed after fibroblast functional studies of cobalamin metabolism, as depicted in flow chart of study design (Figure S1). The laboratory has cells from over
SC
200 such patients that could not be diagnosed by fibroblast functional studies. The 131
NU
patient DNA samples were chosen because of the availability of extracted DNA in the laboratory. Samples do not include patients where a nutritional deficiency was
MA
suspected due to reported low serum cobalamin in the patient or in the mother.
D
However, cobalamin status was not reported for all patients at time of referral.
TE
2.2. Functional Studies of Cobalamin Metabolism
AC CE P
All fibroblast lines that were submitted to the laboratory at McGill University were subjected to a common panel of tests that allowed diagnosis of all inborn errors of cobalamin metabolism known at the time of submission (Table 1). In cases in which [14C]propionate incorporation was decreased compared to controls, a specific diagnosis was attempted by cellular complementation analysis using established methods [7]. The 131 patients in the panel consisted of patients with fibroblast [14C]propionate incorporation within reference range (n= 123) and patients with decreased fibroblast [14C]propionate incorporation in which results of complementation analysis did not identify a cause (n=8). Clinical findings in patients were ascertained from written clinical reports submitted at the time of referral. This study was approved by the Research Ethics Board of the Royal Victoria Hospital, Montreal, Quebec, Canada.
6
ACCEPTED MANUSCRIPT 2.3. Gene Panel Sequencing and Data Analysis DNA samples were sequenced by the Baylor Miraca Genetics Laboratories
PT
“Cobalamin Metabolism Panel + Severe MTHFR Deficiency by Massively Parallel
RI
Sequencing” with three additional genes: AMN, CUBN, and SLC46A1 (“extended
SC
cobalamin metabolism gene panel”, Table 2). Target sequences of a panel of the 24 genes were enriched by using custom designed NimbleGen SeqCap probe
NU
hybridization (Roche NimbleGen Inc., Madison, WI, USA). The captured target
MA
sequences include all coding exons and 20 bp of their flanking intronic regions. DNA template libraries were prepared according to the manufacturer’s recommendation.
D
Equal molar ratios of 10 indexed samples were pooled to be loaded onto each lane of
TE
the flow cells for sequencing on a HiSeq2000 (Illumina Inc., San Diego, CA, USA) with 100 cycle single-end reads. Raw data in base call files (.bcl format) were converted to
AC CE P
qseq files before demultiplexing with CASAVA v1.7 software (Illumina Inc., San Diego, CA, USA). Demultiplexed data were processed further by NextGENe software for alignment (SoftGenetics, State College, PA, USA). Average depth of coverage of the NGS analysis was 500-1000X. All exons were covered at sufficient depth. The coverage-based depth analysis using NGS data has been previously reported [8]. 2.4. Evaluation of variant pathogenicity Potential pathogenicity of variants was determined by examining the frequency of the variant in the ExAC and dbSNP databases; the predicted pathogenicity by SIFT, PolyPhen2, and MutationTaster algorithms; the type of variant change (i.e. missense, indel, or premature termination); the presence of the variant in patients previously
7
ACCEPTED MANUSCRIPT described in literature; and by comparing clinical phenotypes of previously reported patients with the ones in this study. Variants that did not satisfy the criteria for
PT
pathogenicity were considered benign and deemed noncontributory to the patient’s
RI
phenotype.
SC
3. Results
NU
Two heterozygous variants or an apparently homozygous variant in a single gene were found in 16 of the 131 patients (Table 3). In 8 of these, both detected variants
MA
were classified as pathogenic, allowing for a presumptive clinical diagnosis. None of these were disorders that would have been diagnosed by routine functional studies of
D
cobalamin metabolism. One patient had findings of uncertain significance, and 7 had
TE
benign findings. Two potentially deleterious mutations in separate genes associated
AC CE P
with elevations in MMA were identified in a single patient (MMAB and MMACHC), (Supplementary Tables 1 and 2). Although synergistic heterozygosity has been proposed as an explanation for patient metabolic disorders [9, 10], there was not sufficient evidence to prove that the combination of heterozygous variants affecting proteins in separate cellular compartments caused this patient’s illness. This study was performed on patient DNA samples collected over many decades; thus it was not possible to determine the phase of the two variants identified in the 16 patients by segregation analysis in family members. Clinical findings in patients in whom the NGS panel provided a presumptive diagnosis are reviewed below. 3.1. Patient 1 / TCN2 8
ACCEPTED MANUSCRIPT This Bahraini patient, whose parents are first cousins, was homozygous for a c.679C>T (p.R227*) variant in TCN2 (Table 3), which was deemed pathogenic. Depth of
PT
coverage analysis revealed no evidence of copy number variation in the region. He had
RI
feeding difficulties and transient tachypnea with mild respiratory distress and jaundice during the newborn period, and developed recurrent vomiting and diarrhea that were
SC
attributed to acute gastroenteritis (Table 4). He also had acidosis, pancytopenia (platelet
NU
count of 1600/µL, hemoglobin of 10 g/dL, white blood cell count of 3.2/µL) and elevated urine and serum MMA. Serum total homocysteine, methionine, cobalamin and folate
MA
were within reference limits. Treatment with vitamin B12 and carnitine resulted in resolution of pancytopenia and acidosis, as well as normalization of his MMA levels.
TE
D
Long term treatment was with weekly vitamin B12 injections and carnitine (1 mg weekly, initially IV and later orally). Studies of cultured patient fibroblasts showed incorporation
AC CE P
of both [14C]propionate and of [14C]methyltetrahydrofolate within the reference range, with adequate synthesis of both adenosylcobalamin and methylcobalamin. 3.2. Patients 2 and 3 / SUCLG1 Patient 2 had two heterozygous variants (c.677T>G [p.I226S] and c.776G>A [p.G259D]) in the gene SUCLG1 (Table 3), which were evaluated to be pathogenic. The patient is a female who presented with metabolic acidosis, vomiting, and elevated urine MMA at 20 hours of age (Table 4). She was adopted, and family history was not available. Her [14C]propionate incorporation was 1.6 nmol/mg protein/18h (control 10.8±3.7 nmol/mg protein/18h). Her fibroblasts complemented those of other known inborn errors of cobalamin metabolism.
9
ACCEPTED MANUSCRIPT Patient 3 had two heterozygous variants (c.97+1delG and c.788A>G [p.E263G]) in SUCLG1 (Table 3) that were evaluated to be pathogenic. The patient is a Hispanic male
PT
from El Salvador. He was hospitalized at 5 months for diarrhea, emesis, and
RI
dehydration. He had metabolic acidosis with high blood lactate, and a history of elevated MMA (Table 4). He had elevated propionic acid and other organic acids
SC
including lactic, pyruvic, and 2-hydroxyglutaric, and TCA cycle intermediates. He also
NU
had developmental delay, generalized muscle weakness, and failure to thrive. His [14C]propionate incorporation was 1.8 nmol/mg protein/18h (control 10.8±3.7 nmol/mg
MA
protein/18h). His fibroblasts complemented those of other known inborn errors of
TE
3.3. Patients 4 to 8 / ACSF3
D
cobalamin metabolism.
AC CE P
Two pathogenic variants in ACSF3 were identified in 5 patients with elevated MMA and no diagnosis following functional studies of Cbl metabolism (Table 3). All patients had [14C]propionate incorporation within the reference range, and varied clinical findings, with 3 out of 5 patients having neurological manifestations (Table 4). Patient 4 had two heterozygous variants (c.1672C>T [p.R558W] and c.1673G>A [p.R558Q]) in ACSF3 (Table 3). She was a 2-year-old with elevated MMA that did not respond to vitamin B12 treatment. She had mild developmental delay, a seizure disorder, blue sclera, and a maternal history of osteogenesis imperfecta. Patient 5 had three heterozygous variants (c.634G>C [p.V212L], c.781G>T [p.G261*], and c.854C>T [p.P285L]) in ACSF3 (Table 3). He was a 73-year-old Icelandic male who had a metabolic workup because of a late-onset neurologic 10
ACCEPTED MANUSCRIPT syndrome characterized by progressive generalized weakness of unknown cause (Table 4). There was no cognitive decline or psychiatric problems. As part of this
PT
workup his urine MMA was found to be grossly elevated (~689 µmol/mmol Cr); this
SC
serum B12 levels were within the reference ranges.
RI
decreased to ~300 µmol/mmol Cr after vitamin B12 treatment. His homocysteine and
Patient 6 had a homozygous c.1470G>C (p.E490D) variant in ACSF3 (Table 3).
NU
Depth of coverage analysis revealed no evidence of copy number variation in the
MA
region. She was a 10-year-old girl of Pakistani ethnicity with a history of significant developmental and speech delays, as well as elevated methylmalonic acid levels in
D
plasma and urine (Table 4). Her plasma MMA was 1239 times the upper limit of the
TE
reference range. Homocysteine levels were within the reference range, but lactate
AC CE P
levels were reported to be elevated.
Patient 7 had a heterozygous c.1239+2T>G variant affecting a splice site, and a heterozygous c.1672C>T (p.R558W) variant in ACSF3 (Table 3). She was an 8-monthold in generally good health with elevated MMA on a newborn screen (Table 4). Plasma homocysteine and plasma amino acids levels were within the reference range. The MMA levels responded partially to cobalamin treatment. Patient 8 had a heterozygous c.1672C>T (p.R558W) variant and a heterozygous 1.7 Mb deletion on chromosome 16q24 (Chr16: 87441993:89171912) that encompasses ACSF3 and several other genes. This deletion was first detected in two exons of the gene by coverage depth analysis, and later confirmed using exon targeted array CGH. The patient had elevated MMA, recurrent vomiting from birth to 8 months,
11
ACCEPTED MANUSCRIPT and febrile seizures (Table 4), but no other problems and was developmentally appropriate. Her MMA was 4341 nmol/L (reference range 73-27 nmol/L).
PT
3.4. Patient with Findings of Uncertain Significance
RI
Patient 9 had three heterozygous variants in CUBN: c.196G>A (p.G66R),
SC
c.1088A>G (p.D363G), and c.6260C>G (p.S2087C) (Table 3), which were deemed to
NU
be of uncertain significance. The patient was a 7 week old female of Puerto Rican, Irish, and German descent (Table 4). She had prenatal cardiomyopathy, and post-natal non-
MA
compaction cardiomyopathy. A metabolic evaluation showed a low free carnitine of 8 µmol/L (reference range 15 µmol/L to 55 µmol/L) and low total carnitine of 20 µmol/L
D
(reference range 21 µmol/L to 83 µmol/L). After carnitine supplementation, her free and
TE
total carnitine levels normalized, but there was mildly increased excretion of
AC CE P
methylmalonic acid on urine organic acids (46 µmol/mmol Cr, serum 4.12 µmol/L) and some lactic/pyruvic aciduria. The findings were categorized as uncertain significance because the clinical findings in the patient were not consistent with those typically found in patients with Imerslund-Gräsbeck syndrome. 3.5. Patients Lacking Two Pathogenic Variants in a Single Gene Patient 10 had c.262_264delGAG (p.E88del) and c.658G>A (p.G220R) variants in CD320 (Table 3). Uptake by patient fibroblasts of radiolabeled cyanocobalamin bound to transcobalamin was within the reference range, indicating that function of the transcobalamin receptor, the product of CD320, was not impaired (Table S3). Patients 11 to13 each had two heterozygous variants affecting CUBN, but all were either outside the region of the gene associated with Imerslund-Gräsbeck syndrome or were 12
ACCEPTED MANUSCRIPT annotated as tolerated or benign (Table 3, Table S3). Patient 14, a male, was hemizygous for a c.4442C>T (p.T1481M) variant in HCFC1 that lies outside the kelch
PT
domain of the protein, which is associated with abnormal cobalamin metabolism in cblX
RI
patients (Table 3, Table S3). Patient 15 had a c.824delC (p.A275Vfs) and a c.1186A>G (p.T396A) variant in the SUCLG2 gene (Table 3). Patients with mutations in this gene
SC
have not been previously reported in literature. The c.1186A>G variant is deemed to be
NU
benign by PolyPhen2 and MutationTaster, and has a 0.098 frequency in the general population, indicating that it cannot be causal of a rare disorder (Table S3). Patient 16
MA
had c.1532G>A (p.R511Q) and c.205A>G (p.I69V) variants in the MUT gene (Table 3). The p.R511Q variant was annotated as deleterious/damaging; the p.I69V variant was
TE
D
annotated as benign by PolyPhen2 but deleterious by SIFT. Results of fibroblasts
AC CE P
studies in this patient were not consistent with a diagnosis of mut (Table S3).
4. Discussion
Over the past four decades, it has been possible to identify the genetic cause of metabolic abnormalities in 64% of patients referred to the Vitamin B12 Clinical Research Laboratory at McGill University using functional studies of cobalamin metabolism in cultured patient fibroblasts. In the present study, genomic DNA from 131 patients, referred to the Laboratory on the basis of elevated MMA in blood or urine, with no diagnosis after functional studies, was analyzed using the Baylor Miraca Genetics Laboratories extended cobalamin metabolism gene panel. The study is a retrospective analysis on samples collected over many years, and thus follow-up with referring
13
ACCEPTED MANUSCRIPT physicians was not possible. The study is also limited by the amount of clinical information provided at the time of referral. These studies identified two heterozygous
PT
mutations in the same gene, an apparently homozygous mutation, or a hemizygous
RI
mutation in an X-linked gene, in 16 patients. Of these, eight patients had 2 predicted pathogenic variants, resulting in a presumptive diagnosis of the cause of elevated MMA.
SC
The remainder of the patients, for whom neither the functional studies nor the Cbl gene
NU
panel were able to provide a diagnosis may have non-genetic causes of elevated MMA, including nutritional deficiencies, or they may have yet undiscovered genetic defects.
MA
These might be elucidated by whole exome or genome approaches.
D
The NGS gene panel analyzes a larger spectrum of disorders than those tested
TE
for by the standard analysis of cultured fibroblasts. This allowed identification of variants in genes that do not encode cobalamin metabolic genes but can result in elevation of
AC CE P
MMA (ASCF3, SUCLG1 and SUCLA2). Pathogenic variants in SUCLG1 were identified in two subjects. This disorder results in decreased propionate incorporation, as seen in the two patients, and the diagnosis of mitochondrial depletion syndrome 9 can potentially be confirmed by complementation analysis. However, it has proven difficult in practice to interpret the results of complementation, and complementation with SUCLG1 and SUCLA2 deficient cells is not included as part of the panel of tests that are performed on all samples. The NGS diagnosis of transcobalamin deficiency in Patient 1 was confirmed by measurement of uptake of radiolabeled cyanocobalamin in the absence of exogenous transcobalamin, which showed that patient fibroblasts did not produce functional transcobalamin. However, this test is not part of the standard panel and is performed only when specifically requested, which was not done in this case.
14
ACCEPTED MANUSCRIPT Establishing the diagnosis in patients with ACSF3 variants is challenging because the clinical phenotype of patients is known to be variable. Patients with ACSF3
PT
mutations typically have elevation of malonic and methylmalonic acid, but some patients
RI
have been identified with elevation of only MMA (unpublished data). The phenotypes of all 5 patients presented in this study were within the range of phenotypes seen in other
SC
patients with this disorder, and the variants identified were predicted to be pathogenic.
NU
Biochemical testing of whether transduction of ACSF3 into patient fibroblasts decreases the MMA excretion by the fibroblasts, as was done by Sloan et al. [4] , would provide
MA
functional confirmation of variant pathogenicity.
D
Cubilin is not expressed in fibroblasts, and there are no in vitro tests for cubilin
TE
function. Patient 9 had three sequence variants in the CUBN gene. Of these, two (p.G66R and p.D363G) affect the C-domain of cubilin, the portion of the membrane
AC CE P
protein that supports binding to the intrinsic factor-cobalamin complex and its endocytosis [11]; the third lies outside the region of the gene that has been associated with Imerslund-Gräsbeck syndrome. However, the clinical findings in the patient were not compatible with Imerslund-Gräsbeck syndrome. Many of the patients referred for analysis at the McGill laboratory have had elevated MMA identified as the result of a positive newborn screen in the absence of clinical findings, or in combination with findings not typical of methylmalonic aciduria. In this context, diagnosis of a cause for elevated MMA levels may not represent diagnosis of a cause for the clinical findings in a patient. For example, it is not clear that that mutations affecting ACSF3 function have any consistent associated clinical picture, and decreased function of the CD320 gene product has not had any clinical consequences 15
ACCEPTED MANUSCRIPT beyond elevated MMA levels in most cases. The two potentially deleterious mutations in MUT identified in patient 18 may be the cause of the persistent borderline elevated
PT
MMA identified in this subject, but results of fibroblasts studies make it clear that the
RI
individual does not have the mut disorder. This finding is similar to that in a previously described patient with elevated MMA and two MMAB variants but without clinical
SC
findings of cblB disease [12].
NU
This study demonstrates the high sensitivity of functional studies of cobalamin
MA
metabolism for diagnosis of inborn errors of cobalamin metabolism and the ability of the NGS panel to provide presumptive diagnoses for a number of patients with disorders
D
that had not been diagnosed by functional studies. In addition to identifying pathogenic
TE
variants in patients with known disease presentations, NGS-based panel analysis has the power to expand phenotypes and genotypes of known diseases. Going forward, as
AC CE P
NGS-based Cbl gene panels gain popularity as first line tests for clinical diagnosis, somatic cell testing can still be useful to aid in the evaluation of pathogenicity of novel variants.
Acknowledgments The authors would like to thank Jeehye Jung for her assistance with gathering the patient data.
16
ACCEPTED MANUSCRIPT References [1]
I. Manoli, C.P. Venditti, Methylmalonic acidemia GeneReviews (2010).
RI
PT
[2] C.M. Dobson, A. Gradinger, N. Longo, X. Wu, D. Leclerc, J. Lerner-Ellis, M. Lemieux, C. Belair, D. Watkins, D.S. Rosenblatt, R.A. Gravel, Homozygous nonsense mutation in the MCEE gene and siRNA suppression of methylmalonyl-CoA epimerase expression: a novel cause of mild methylmalonic aciduria Mol Genet Metab 88 (2006) 327-333.
NU
SC
[3] A.B. Gradinger, C. Belair, L.C. Worgan, C.D. Li, J. Lavallee, D. Roquis, D. Watkins, D.S. Rosenblatt, Atypical methylmalonic aciduria: frequency of mutations in the methylmalonyl CoA epimerase gene (MCEE) Hum Mutat 28 (2007) 1045.
MA
[4] J.L. Sloan, J.J. Johnston, I. Manoli, R.J. Chandler, C. Krause, N. CarrilloCarrasco, S.D. Chandrasekaran, J.R. Sysol, K. O'Brien, N.S. Hauser, J.C. Sapp, H.M. Dorward, M. Huizing, B.A. Barshop, S.A. Berry, P.M. James, N.L. Champaigne, P. de Lonlay, V. Valayannopoulos, M.D. Geschwind, D.K. Gavrilov, W.L. Nyhan, L.G. Biesecker, C.P. Venditti, Exome sequencing identifies ACSF3 as a cause of combined malonic and methylmalonic aciduria Nat Genet 43 (2011) 883-886.
AC CE P
TE
D
[5] A. Alfares, L.D. Nunez, K. Al-Thihli, J. Mitchell, S. Melancon, N. Anastasio, K.C. Ha, J. Majewski, D.S. Rosenblatt, N. Braverman, Combined malonic and methylmalonic aciduria: exome sequencing reveals mutations in the ACSF3 gene in patients with a non-classic phenotype J Med Genet 48 (2011) 602-605. [6] D. Watkins, D.S. Rosenblatt, Lessons in biology from patients with inborn errors of vitamin B12 metabolism Biochimie 95 (2013) 1019-1022. [7] D. Watkins, N. Matiaszuk, D.S. Rosenblatt, Complementation studies in the cblA class of inborn error of cobalamin metabolism: evidence for interallelic complementation and for a new complementation class (cblH) J Med Genet 37 (2000) 510-513. [8] Y. Feng, D. Chen, G.L. Wang, V.W. Zhang, L.J. Wong, Improved molecular diagnosis by the detection of exonic deletions with target gene capture and deep sequencing Genet Med 17 (2015) 99-107. [9] C. Chery, A. Hehn, N. Mrabet, A. Oussalah, E. C. Besseau, J.-M. B Alberto, I. Gross, T. Josse, P. Josse, P. Gérard, R.M. Guéant-Rodriguez, J.-N. Freund, J. Devignes, F. Bourgaud, L. Peyrin-Biroulet, F. Feillet, J.-L. Guéant,.Gastric intrinsic factor deficiency with combined GIF heterozygous mutations and FUT2 secretor variant Biochimie 95 (2013) 995-1001 [10] J. Vockley, P. Rinaldo, M.J. Bennett, D. Matern, G.D. Vladutiu, Synergistic heterozygosity: disease resulting from multiple partial defects in one or more metabolic pathways Mol Genet Metab 71 (2000) 10-18.
17
ACCEPTED MANUSCRIPT [11] S.M. Tanner, A.C. Sturm, E.C. Baack, S. Liyanarachchi, A. de la Chapelle, Inherited cobalamin malabsorption. Mutations in three genes reveal functional and ethnic patterns Orphanet J Rare Dis 7 (2012) 56.
AC CE P
TE
D
MA
NU
SC
RI
PT
[12] M.L. Illson, L. Dempsey-Nunez, J. Kent, Q. Huang, A. Brebner, M.L. Raff, D. Watkins, B.M. Gilfix, C.T. Wittwer, D.S. Rosenblatt, High resolution melting analysis of the MMAB gene in cblB patients and in those with undiagnosed methylmalonic aciduria Mol Genet Metab 110 (2013) 86-89.
18
ACCEPTED MANUSCRIPT Table 1. Functional studies of Cbl metabolism routinely offered and performed on fibroblasts from patients in this study. Is the functional test
Gene in Cbl
functional studies of
part of routine
panel
cobalamin metabolism
diagnostic analysis?
1
MMAA
cblA
Yes
2
MMAB
cblB
Yes
3
MMACHC
cblC
4
MMADHC
cblD
5
MTRR
cblE
6
LMBRDI
cblF
7
MUT
mut
8
ABCD4
cblJ
9
MTR
cblG
10
HCFC1
cblX*
11
TCN2
Transcobalamin II deficiency
12
MTHFR
13
IVD
14
ACSF3
15
MCEE
16
GIF
17
SC
RI
PT
Disease diagnosis by
Yes
TE
D
MA
NU
Yes
Homocystinuria due to
Yes Yes Yes Yes Yes
No – must be requested No – must be requested
AC CE P
MTHFR deficiency
Yes
N/A
No
N/A
No
N/A
No
N/A
No
SLC46A1
N/A
No
18
CUBN
N/A
No
19
AMN
N/A
No
20
CBS
N/A
No
21
CD320
N/A
No
22
SUCLA2
SUCLA2 deficiency
No – must be requested
23
SUCLG1
SUCLG1 deficiency
No – must be requested
24
SUCLG2
N/A
No
* cblX patients have a cellular phenocopy of cblC disease.
19
ACCEPTED MANUSCRIPT Table 2. List of genes tested by Baylor Miraca Genetics Laboratories extended cobalamin metabolism gene
GenBank
symbol
Accession
OMIM
Phenotype
1
MMAA
NM_172250.2
251100
Methylmalonic aciduria cblA type, vitamin B12 responsive
2
MMAB
NM_052845.3
251110
Methylmalonic aciduria cblB type, vitamin B12 responsive
3
MMACHC
NM_015506.2
277400
Methylmalonic aciduria and homocystinuria, cblC type
4
MMADHC
NM_015702.2
277410
Methylmalonic aciduria and homocystinuria, cblD type
5
MTRR
NM_002454.2
236270
Homocystinuria-megaloblastic anemia, cblE type
6
LMBRDI
NM_018368.3
277380
Methylmalonic aciduria and homocystinuria, cblF type
7
MUT
NM_000255.3
251000
Methylmalonic aciduria due to methylmalonyl-CoA mutase
8
ABCD4
NM_005050.3
614857
9
MTR
NM_000254.2
250940
10
HCFC1
NM_005334.2
309541
11
TCN2
NM_000355.3
275350
12
MTHFR
NM_005957.4
236250
13
IVD
NM_002225.3
14
ACSF3
NM_174917.4
15
MCEE
NM_032601.3
16
GIF
17
NU
SC
RI
PT
Gene
TE
panel.
deficiency, mut type
D
MA
Methylmalonic aciduria and homocystinuria, cblJ type Homocystinuria-megaloblastic anemia, cblG type Methylmalonic acidemia and homocysteinemia, cblX type Transcobalamin II deficiency Homocystinuria due to MTHFR deficiency Isovaleric acidemia
614265
Combined malonic and methylmalonic aciduria
251120
Methylmalonyl-CoA epimerase deficiency
NM_005142.2
261000
Intrinsic factor deficiency
SLC46A1
NM_080669.5
229050
Hereditary folate malabsorption
18
CUBN
NM_001081.3
261100
Megaloblastic anemia
19
AMN
NM_030943.3
261100
Megaloblastic anemia
20
CBS
NM_000071.2
236200
Homocystinuria due to cystathionine beta-synthase
21
CD320
22
SUCLA2
23
SUCLG1
24
SUCLG2
AC CE P
243500
deficiency
NM_016579.3
Methylmalonic aciduria due to transcobalamin receptor
613646
defect NM_003850.2
612073
Mitochondrial DNA depletion syndrome 5 (encephalomyopathic with or without MMA)
NM_003849.3
245400
Mitochondrial DNA depletion syndrome 9 (encephalomyopathic type with MMA)
NM_001177599.1
N/A
N/A
20
ACCEPTED MANUSCRIPT Table 3. Variants identified in patients with MMA that had 2 variants in a single gene on the Cbl gene panel.
Pathogenic findings c.679C>T 1 TCN2 2 SUCLG c.677T>G 1 c.776G>A
Amino Acid Change
Zygosit y
dbSNP ID
ExAC frequency
p.R227* p.I226S
hom het
NA NA
NA NA
p.G259D
het
NA
NA
3
SUCLG 1
c.97+1delG c.788A>G
Splice p.E263G
het het
NA NA
4
ACSF3
c.1672C>T
p.R558W
het
c.1673G>A
p.R558Q
het
c.634G>C
p.V212L
het
rs14109014 3 rs14032814 2 NA
0.00004191
c.781G>T
p.G261*
het
NA
0.00002479
c.854C>T
p.P285L
het
0.008748
hom
NA NA
SC
NU
ACSF3
MA
5
SIFT
PT
Variant Call
RI
Patien Gene t
0.002739
0.00005408
PolyPhen Mutatio 2 n Taster
NA deleteriou s deleteriou s splicing deleteriou s deleteriou s tolerated
NA damaging
DC DC
damaging
DC
NA damaging
DC DC
damaging
DC
damaging
DC
deleteriou s deleteriou s deleteriou s tolerated
damaging
DC
NA
DC
damaging
DC
damaging
DC
NA deleteriou s deleteriou s NA
NA damaging
DC DC
damaging
DC
NA
NA
deleteriou s deleteriou s deleteriou s
damaging
DC
benign
P
damaging
DC
-
het
rs14379350 2 rs14753837 0 NA rs14109014 3 rs14109014 3 -
p.G66R
het
rs12259370
0.006806
p.D363G
het
0.0004614
p.S2087 C Findings lacking two pathogenic variants in a single gene 10 CD320 c.262_264delGA p.E88del G c.658G>A p.G220R
het
rs13804440 9 rs37664172 5
rs15038417 1 rs2336573
0.008838
NA
NA
DC
0.06352
damaging
P
c.2594G>A
p.S865N
het
0.007454
benign
P
c.6469A>G
het
damaging
DC
damaging
DC
c.6788T>G
p.N2157 D p.G3587 R p.F2263C
het
rs13808352 2 rs14436024 1 rs20148426 6 rs2271460
deleteriou s tolerated
damaging
P
c.8741C>T
p.A2914
het
rs45551835
0.01203
damaging
DC
ACSF3
c.1470G>C
p.E490D
7
ACSF3
c.1239+2T>G c.1672C>T
splice p.R558W
het het
8
ACSF3
c.1672C>T
p.R558W
het
AC CE P
TE
D
6
Chr16: 87441993 : 89171912 deletion (1.7 Mb spanning ACSF3) Findings of uncertain significance c.196G>A 9 CUBN c.1088A>G
c.6260C>G
11
12
13
CUBN
CUBN
CUBN
c.10759G>A
het het
het
0.0002489 NA 0.002739 0.002739 NA
0.00002475
0.005628 0.001326 0.01841
deleteriou s deleteriou s deleteriou s deleteriou
21
ACCEPTED MANUSCRIPT
14
HCFC1
c.10265C>T c.4442C>T
15
SUCLG2
c.824delC c.1186A>G
V p.T3422I p.T1481 M p.A275Vf s p.T396A
0.01832 0.001013
s tolerated tolerated
damaging damaging
P P
het
rs1801230 rs19979802 9 NA
NA
NA
NA
DC
het
rs35494829
0.09769
het hemi
SC
RI
PT
deleteriou damaging P s c.1532G>A p.R511Q 16 MUT het NA 0.00000834 deleteriou damaging DC 7 s c.205A>G p.I69V het rs11592355 0.002043 deleteriou benign DC 6 s hom = homozygous, het = heterozygous, hemi = hemizygous. *Patient also carries MTHFR c.203G>A. #Patient also carries MTR c.3650G>T. DC = disease causing, P = polymorphism. Variants not previously reported in patients in literature are bolded.
m
2
f
4 yrs
8.7
Bahraini
3
m
4
f
5
m
AC CE P
TE
1
Ethnicity
7 days
1.6
Clinical findings
MA
Gender
Propionate incorporation (nmol/mg protein/18h)
D
Patient
Age at biopsy
NU
Table 4. Clinical findings in patients with elevated MMA that had 2 variants in a single gene on the Cbl panel.
Black
Elevated MMA, poor feeding, vomiting, diarrhea, metabolic acidosis, developmental delay.
Acidosis at 20 hours of age, vomiting, urine showed MMA. Metabolic acidosis. Hospitalized at 5 months for diarrhea, emesis, and dehydration. Has been acidotic with high blood lactate. History of elevated MMA. Elevated propionic acid and other organic acids including lactic, pyruvic and 2-hydroxyglutaric acid, and TCA cycle intermediates. Not doing well on MMA diet. Developmental delay. Generalized muscle weakness. Failure to thrive.
Family History Parents are first cousins. Mother has sickle cell disease. No known family members with metabolic disease. Patient is adopted.
No consanguinity. Patient has two healthy siblings.
9 mts
1.8
Hispanic – El Salvador
2 yrs
5.3
White
Mild developmental delay, seizure disorder, methylmalonic aciduria.
Patient has an autistic brother.
White
Elevated MMA in urine, late-onset neurologic syndrome characterized by progressive weakness of unknown cause, rare axonal degeneration of segmental myelin thinning.
N/A No consanguinity, patient has healthy siblings.
73 yrs
13.2
6
f
10 yrs
13.9
Pakistani
History of significant speech and development delays, elevated MMA in plasma and urine, elevated lactates, normal homocysteine.
7
f
8 mts
6.3
White
Persistent elevation of MMA.
N/A
Elevated MMA, recurrent vomiting from birth to 8 months, febrile seizures.
N/A
Non-compaction cardiomyopathy, elevated
No
8
f
3 yrs
7.2
English, Irish, Cherokee
9
f
7 wks
19.2
Puerto
22
ACCEPTED MANUSCRIPT
10
m
1 mt
5.5
N/A
11
m
9 mts
3.0
White
MMA in urine. Elevated lactate and pyruvate.
Elevated glycine, elevated C3 on NBS, elevated MMA. Probable benign MMA. Seizures and developmental delay.
PT
Rican, Irish, German
consanguinity. Patient has a healthy older brother. N/A N/A Mom, dad, and 2 older sisters were born before NBS. Sibling diagnosed as cblG..
m
4 mts
11.4
Pakistani
Detected on NBS with elevated C3. Elevated MMA in plasma and urine.
13
f
2 mts
8.0
N/A
Tested negative for cblG
14
m
17 mts
4.8
White
Mild developmental delay, low plasma carnitine, seizure-like episodes.
N/A
15
m
12 mts
9.1
White
Infantile spasms, elevated MMA, constipation.
N/A
7.2
White (mixed Caucasian European)
Positive NBS for MSUD, but well. Borderline elevated MMA. Slightly elevated homocysteine.
N/A
SC
NU
MA
D
2 mts
TE
f
AC CE P
16
RI
12
23