Molecular picture of cobalamin C/D defects before and after newborn screening era

Abstract

Objective

Birth prevalence of Cobalamin (Cbl) C or D defects in Portugal is an estimated 1:85,000, one of the highest worldwide. We compared the genotype/phenotype of patients identified with CblC or CblD before and after the implementation of expanded newborn screening.

Methods

Twenty-five Portuguese CblC/D patients, 14 symptomatic and 11 identified through screening, were diagnosed using gas chromatography or tandem mass spectrometry. Molecular characterization was performed through the study of MMACHC and MMADHC genes.

Results

The most common MMACHC mutation, c.271dupA, was present in 100% of MMACHC alleles of all CblC screened patients, in contrast with the 61% identified before expanded newborn screening. All studied cases (except one, who presented a CblD deficiency) presented a CblC defect. More CblC late-onset patients were diagnosed before the introduction of newborn screening than in the post newborn screening era, probably because some early onset patients died without a definitive diagnosis.

Conclusion

The molecular data found in this cohort contribute to the improvement of screening and diagnosis of Cbl defects and would enable a confirmatory diagnosis of these patients, reducing the need for complex, costly, laborious, and time-consuming biochemical/enzymatic tests.

Keywords

Expanded newborn screening cobalamin blood spots Vitamin B12

Introduction

Acquired from the diet, cobalamin (Cbl) or vitamin B12 is metabolized within the cell into methylcobalamin and adenosylcobalamin,¹ essential coenzymes to methionine synthase and methylmalonyl-CoA mutase. Inborn errors of intracellular cobalamin metabolism, resulting from impaired conversion of dietary vitamin B12 to its two metabolically active forms, cause methylmalonic aciduria (MMA) and homocystinuria (HC), and a decrease of methionine.^2,3 In Portugal, only CblC and CblD deficiencies have been detected.

CblC (MIM# 277400), the most common inherited disorder of vitamin B12 metabolism, has two distinct phenotypes: (i) early onset (EO), a multisystemic disease that presents with MMA and HC, and clinical symptoms such as feeding difficulties, hypotonia, hydrocephalus, progressive developmental delay, seizures, mild dysmorphic features, anaemia, and haemolitic uremic syndrome appearing within the first year of life, and (ii) late onset (LO), which presents later in life, with a relatively milder clinical phenotype, progressive neurological symptoms, and behavioural disturbances.⁴ The treatment of both forms with hydroxycobalamin, betaine, and folic acid usually decreases metabolite levels.² Early treatment is clearly advantageous for patients with LO CblC defect, and for those with EO CblC treatment improves both survival and non-neurological symptoms, but its effect on neurocognitive development remains controversial.⁵

CblC defects are caused by mutations in the MMACHC gene (ID: 25974; MIM# 609831), located in the 1p34.1 locus.⁶ Three common mutations have been identified: c.271dupA, c.394C > T, and c.331C > T.^6–8 The c.271dupA and c.331C > T mutations have been associated with EO disease, and c.394C > T primarily with LO disease. Some MMACHC mutations have been observed to correlate with age of onset and others cluster according to ethnicity, probably as a result of different founder effects.^6,7,9 MMACHC is responsible for early processing of both cyanocobalamin (decyanation) and alkylcobalamins (dealkylation) in mammalian cells.¹⁰

Patients with the CblD genetic defect (MIM# 277410) can have three distinct biochemical phenotypes: (i) isolated homocystinuria (CblD-HC), (ii) isolated methylmalonic aciduria (CblD-MMA), and (iii) combined homocystinuria and methylmalonic aciduria (CblD-MMA/HC).¹¹ CblD-HC patients present with developmental delay, ataxia, and megaloblastic anaemia; CblD-MMA patients show respiratory distress, cranial haemorrhage, seizures, and an abnormal electroencephalogram; CblD-MMA/HC patients present with developmental delay, seizures, hypotonia, lethargy, and megaloblastic anaemia.^11,12 Regardless of their clinical phenotype, patients with CblD defects usually respond well to betaine, folic acid, and OH-cobalamin treatment, even though they remain neurologically impaired. MMADHC (MIM 611935) is the gene (chromosome region 2q23.2)¹² responsible for the CblD defect of vitamin B12 metabolism. Usually, individuals with CblD-HC carry missense mutations causing amino acid substitutions towards the C-terminal part of the protein, and patients with CblD-MMA carry mutations causing a premature stop of translation towards the N-terminal part of the protein, on at least one allele. The combined phenotype (CblD-MMA/HC) is associated with deleterious nonsense and missense mutations in the middle of the protein.¹² Recent studies demonstrate that MMADHC interacts directly with MMACHC, and the MMACHC–MMADHC complex is involved in Vitamin B12 trafficking.¹³ Genotype/phenotype correlation analysis of the three CblD variants suggests that the N- and C-termini of MMADHC have specific functions in the mitochondrion and cytoplasm, respectively.^14,15

Our molecular findings on CblC/D Portuguese patients indicate the type of patients who may be detected in the neonatal period. While there is no good published evidence of the benefits of screening for CblC/D defects, there is a great deal of anecdotal evidence suggesting that some EO patients do much better after early diagnosis.¹⁶ Improving the newborn screening diagnosis therefore seems worthwhile.

Methods

Of the 25 Portuguese CblC/D patients in our study (12 males, 13 females), 14 symptomatic patients were identified up to 2004 by urinary organic acids and plasmatic amino acid analysis (group 1). The remaining 11 patients were identified after 2004, when expanded newborn screening was established (group 2), and were detected through acylcarnitines and amino acids analysis in dried blood spots. The age at diagnosis ranged between 25 days and 16 years for group 1, and 3–7 days, for group 2. Informed consent was obtained from all patients included in this study.

Two API2000 triple quadrupole tandem mass spectrometers (ABSciex) were used to perform routine MS/MS neonatal screening, including the analysis of amino acids and acylcarnitines as butyl esters.¹⁷ Screening for CblC/D was included in the Portuguese Newborn Screening Program in 2004 and has been performed at national level since 2006. It is based on the detection of a high propionyl/acetylcarnitine ratio (C3/C2; normal <0.25), alongside elevated levels of propionylcarnitine (C3; normal <6.3 µM) and an increased propionylcarnitine/methionine ratio (C3/Met; normal <0.3), as well as normal to low levels of methionine (Met; normal >7.0 µM). More recently, the cut-off value of C3 metabolite was modified to 5.3 µM and new markers were introduced: propionylcarnitine/free carnitine (C3/C0; normal <0.4) and methionine/phenylalanine (Met/Phe; normal >0.3).

Urinary organic acids were extracted by a standard ethylacetate procedure, derivatized with Bis-trimethylsilyltrifluoroacetamide and pyridine, and analysed on a gas chromatography–mass spectrometry system (GC/MS Shimadzu QP 2010 Plus). Plasma amino acid analysis was performed by ion-exchange chromatography (Biochrom 30 Gomensoro Instrumentation Cientifica).

Genomic DNA was extracted from peripheral blood; where patients were already deceased, DNA was extracted from Guthrie cards, using standard procedures. The coding exons and exon–intron boundaries of MMACHC and MMADHC genes were amplified by polymerase chain reaction as previously described,^6,12 and the products were sequenced using the BigDye Terminator Cycle Sequencing Version 3.1 (Applied Biosystems, Foster City, CA), and analysed on an ABI 3130XL DNA Analyser.

All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975 (revised 2000).

Results

Over the past 25 years, our centre (National Institute of Health, Porto, Portugal) has been involved in CblC/D defect diagnosis and newborn screening. Diagnosis in the pre-newborn screening era was based on the identification of urinary and circulating metabolites and, whenever possible, confirmed by either fibroblast complementation assays in the early stages of the programme¹⁸ or, more recently, by molecular studies.^6,11 Based on the results of 10 years of expanded newborn screening, the birth prevalence of CblC/D defects in Portugal is 1:85,000, one of the highest known prevalence in the world.¹⁹

As shown in Table 1, 9/14 subjects (five male, four female) (64%) from group 1, presented the landmark description of CblC defect within the first year of life (EO) and 5/14 (three male, two female) (36%) showed the disease later in life (LO). Six patients from this group (two male, four female) died (three EO, three LO). Cause of death for all was a metabolic-related event.

Table 1.

Molecular and biochemical data of CblC patients, diagnosed in pre newborn screening era.

	Patient	Sex	Onset	Age of diagnosis	MMACHC genotype		Biochemical data
	Patient	Sex	Onset	Age of diagnosis	Allele 1	Allele 2	MMA (U)^⋄ µmol/mmol creat	Total homocysteine (P)^# µmol/l
Group 1 (Before NBS)	1	M	EO	1.5Mo	c.271dupA (p.R91Kfs*14)	c.565C > A (p.R189S)	5706	143
	2	F	EO	3Mo	c.271dupA (p.R91Kfs*14)	c.394C > T (p.R132*)	2027	292
	3	M	EO	1Y	c.271dupA (p.R91Kfs*14)	c.544T > C (p.C182R)	2100	96
	4^†	M	EO	1Mo	c.271dupA (p.R91Kfs*14)	c.271dupA (p.R91Kfs*14)	850	142
	5^†	F	EO	33D	c.271dupA (p.R91Kfs*14)	c.271dupA (p.R91Kfs*14)	2350	152
	6	F	EO	25D	c.271dupA (p.R91Kfs*14)	c.271dupA (p.R91Kfs*14)	560	133
	7	M	EO	28D	c.271dupA (p.R91Kfs*14)	c.271dupA (p.R91Kfs*14)	568	160
	8	M	EO	1Y	c.271dupA (p.R91Kfs*14)	c.565C > A (p.R189S)	4304	157
	9^†	F	EO	1Mo	c.271dupA (p.R91Kfs*14)	c.271dupA (p.R91Kfs*14)	5517	146
	10^†	F	LO	15Y	c.394C > T (p.R132*)	c.394C > T (p.R132*)	539	66
	11	M	LO	9Y	c.271dupA (p.R91Kfs*14)	c.394C > T (p.R132*)	1674	86
	12	M	LO	16Y	c.271dupA (p.R91Kfs*14)	c.394C > T (p.R132*)	1560	66
	13^†	M	LO	16Y	c.394C > T (p.R132*)	c.394C > T (p.R132*)	1009	42
	14^†	F	LO	4Y	c.271dupA (p.R91Kfs*14)	c.394C > T (p.R132*)	2420	72

creat: creatinine; D: days; EO: early onset; F: female; LO: late onset; M: male; ^†: died (patients 4, 5, 9, and 14 died at diagnosis time; patients 10 and 13 died at 21 and 26 years old, respectively); MMA: methylmalonic acid; Mo: months; P: plasma; U: urine; Y: years; ^⋄: MMA reference values: 0–1 Mo <15.6 µmol/mmol creat, 1Mo–5Y <6.2 µmol/mmol creat, 5Y–18Y <3.3 µmol/mmol creat; ^#: total homocysteine reference values: 4.0–14.0 µmol/l; *: stop codon.

The group 2 patients (Table 2) detected by expanded newborn screening were EO and presented a CblC defect, except for one (patient 25), who presented with a CblD defect with combined homocystinuria and methylmalonic aciduria (CblD-MMA/HC). This was a false negative case of expanded newborn screening. The screening results for this patient (age five days) were: C3–6.1 (N < 6.3 µM), methionine − 15.2 (N > 7.0 µM), C3/C2 − 0.4 (N < 0.25), C3/Met − 0.4 (N < 0.3), C3/C16 − 2.0 (N < 2.5), and C3/C0 − 0.3 (N < 0.4). Sequencing analysis identified MMACHC mutations in 96% of patients (24 of 25). Only one patient from group 2 presented with a mutation in the MMADHC gene: the previously reported c.748C > T (p.R250X) nonsense mutation, in homozygosity. These findings suggest that a defect in the MMACHC gene is the major cause of disease among Portuguese patients with combined MMA and HC. The c.271dupA mutation accounted for 100% of MMACHC alleles in group 2, a significantly different percentage from that detected in patients from group 1 (61%).

Table 2.

Molecular and biochemical data of CblC/D patients, diagnosed in newborn screening era.

	Patient	Sex	Age of screening	MMACHC genotype		NBS biochemical data						Diagnosis biochemical data
	Patient	Sex	Age of screening	Allele 1	Allele 2	Met (N > 7.0 µM)	C3 (N < 5.3 µM)	C3/C2 (N < 0.25)	C3/Met (N < 0.3)	C3/C16 (N < 2.5)	C3/C0 (N < 0.4)	MMA (U) (N < 6.2 µmol/ mmol creat)	Total homocysteine (P) (N:4–14 µmol/l)
Group 2 (After NBS)	15	M	5D	c.271dupA (p.R91Kfs*14)	c.271dupA (p.R91Kfs*14)	6.8	4.9	0.3	0.75	1.9	0.2	6048	180
	16	M	4D	c.271dupA (p.R91Kfs*14)	c.271dupA (p.R91Kfs*14)	14.0	9.2	0.4	0.7	3.2	0.4	1792	250
	17	F	7D	c.271dupA (p.R91Kfs*14)	c.271dupA (p.R91Kfs*14)	6.3	7.6	0.7	1.4	3.4	1.1	>10,000	310
	18	F	4D	c.271dupA (p.R91Kfs*14)	c.271dupA (p.R91Kfs*14)	7.4	7.5	0.4	1.0	1.9	0.4	3626	195
	19	F	5D	c.271dupA (p.R91Kfs*14)	c.271dupA (p.R91Kfs*14)	25.4	7.9	0.5	0.3	2.5	0.3	1551	158
	20	F	4D	c.271dupA (p.R91Kfs*14)	c.271dupA (p.R91Kfs*14)	6.5	8.4	0.5	1.3	2.4	0.4	3210	180
	21	M	5D	c.271dupA (p.R91Kfs*14)	c.271dupA (p.R91Kfs*14)	7.2	15.3	0.7	2.1	6.0	0.8	5187	192
	22	F	4D	c.271dupA (p.R91Kfs*14)	c.271dupA (p.R91Kfs*14)	13.3	14.4	0.7	1.1	4.3	0.4	2510^♦	130
	23	F	4D	c.271dupA (p.R91Kfs*14)	c.271dupA (p.R91Kfs*14)	11.9	13.5	0.7	1.1	5.5	0.4	2830^♦	145
	24	M	5D	c.271dupA (p.R91Kfs*14)	c.271dupA (p.R91Kfs*14)	6.5	6.1	0.5	0.9	3.9	0.7	1714^♦	98
				MMADHC genotype			Diagnosis biochemical data
	Patient	Sex	Age of diagnosis	Allele 1	Allele 2	Met^⋄ (N > 14.0 µM)	C3^⋄ (N < 3.6 µM)	C3/C2^⋄ (N < 0.2)	C3/Met^⋄ (N < 0.17)	C3/C16^⋄ (N < 4.0)	C3/C0^⋄ (N < 0.1)	MMA (U) (N < 6.2 µmol/ mmol creat)	Total homocysteine (P) (N:4–14 µmol/L)
	25	F	9Mo	c.748C > T (p.R250*)	c.748C > T (p.R250*)	7.1	3.6	0.3	0.5	5.1	0.22	2393	151

C0: free carnitine; C2: acetylcarnitine; C3: propionylcarnitine; C16: palmitoylcarnitine; D: days; F: female; M: male; Met: methionine; Mo: months; ^⋄: reference values: 1Mo–1Y.

The c.271dupA mutation was the most common MMACHC disease allele (77%) in our study. Mutations c.394C > T (p.R132X), c.544T > C (p.C182R), and c.565C > A (p.R189S) were found in 16, 2, and 4% of the total MMACHC alleles, respectively. As previously noted, the c.271dupA, c.544T > C, and c.565C > A mutations were associated with EO disease, while the c.394C > T mutation underlies LO forms of the disease.

Discussion

Defects in the MMACHC gene are the major cause of combined MMA and HC among Southern European (Portugal, Spain, and Italy)⁷ and Chinese patients.²⁰ The 14 diagnosed symptomatic patients in our study demonstrated significant molecular and clinical heterogeneity, with different genotypes and both EO and LO clinical forms. However, since the implementation of expanded newborn screening in 2004, all CblC defect patients we have identified share the same genotype: the c.271dupA mutation in a homozygous state. These data confirm that newborn screening allows the detection of the most severe EO cases. No LO patients were identified by newborn screening during this period. This may be related with the low birth prevalence of LO forms, but these patients are more difficult to identify by newborn screening. To ensure that all cases are identified we could decrease the cut-off markers, but this would substantially increase the number of false positives, the cost of the newborn screening for these diseases, and also parental anxiety. The implementation of second-tier tests for total homocysteine and methylmalonic acid determination may help to overcome this problem and allow the diagnosis of LO cases in our population. Another study has suggested that the C3/C2 ratio (and perhaps other ratios such as C3/C0) should be used as markers for disorders of propionate metabolism when the total carnitine level is low, because it may lead to a lower C3 acylcarnitine level. Therefore, depending on screening cut-offs, CblC patients with LO forms, mild mutations, and low total carnitine levels could be missed in newborn screening.²¹

Our study allowed for the recognition of strong genotype/phenotype correlations, mainly in CblC defects, as already outlined in a large French-Canadian study.⁶ When a correlation to the clinical phenotype was attempted, homozygosity for c.271dupA appears to be uniformly associated with an EO phenotype. In particular, we found that in the pre-neonatal screening era, 61% of EO alleles harboured the c.271dupA (five patients of this group homozygous for this mutation), whereas the c.394C > T mutation accounted only for 4% of EO alleles. In the LO cases diagnosed in the pre-neonatal screening era, these frequencies were quite different: the c.271dupA and the c.394C > T mutations were found in 30 and 70% of the alleles, respectively. The latter mutation was never detected in homozygosity in EO patients, whereas it was found in compound heterozygosity with c.271dupA in one case with early disease, and in three LO cases. In the post-newborn screening era, all CblC defects presented the c.271dupA mutation in a homozygous state. The reduced number of EO patients identified in the pre-newborn screening era was probably due to the patients’ death without a definitive diagnosis.

These genotype/phenotype observations are consistent with the clinical heterogeneity observed in CblC disease.²² While age at onset has been used to subcategorize CblC individuals into EO and LO groups, individuals with LO CblC disease may have had disease-related symptoms, which went unrecognized, before they were diagnosed.²¹

Molecular strategies based on the simultaneous detection of the two most common mutations (i.e. c.271dupA (EO) and c.394C > T (LO)) would be advantageous as they might characterize almost the whole set of Portuguese patients, enabling a cost- and time-effective confirmatory diagnosis, and reducing the need for more laborious and time-consuming biochemical testing in cultured cells. The two other mutations found in our study suggest that the likelihood of detecting new Southern European variants in MMACHC gene is low. A quick testing for the most common c.271dupA mutation allowed the rapid confirmation of the MMACHC infants detected by expanded newborn screening, as all were homozygous for this mutation. This genotype is associated with the most severe phenotype of EO presentation and contributed to the high prevalence of EO patients detected by newborn screening compared with that among patients identified in the pre-newborn screening era, already in a symptomatic stage of the disease. This may be explained by the death, without a definitive diagnosis, of patients with severe EO forms in early infancy.

We detected one patient, in the post-newborn screening era, who presented with a CblD defect. This case, diagnosed at age nine months, was a neonatal screening false negative, because the C3 cut-off at that time was higher than that currently in use (6.3 µM instead of 5.3 µM). This patient presented with EO, but not a severe form of the disease, thus corroborating our hypothesis that milder EO and LO forms of CblC/D disease may not be identified at newborn screening. Additionally, during the newborn screening era, we have identified secondary cobalamin defects in two neonates of vegetarian mothers, and two other children whose mothers had pernicious anaemia.

Conclusion

Comparison of the two groups in our cohort of 25 ClbC/D cases demonstrates that LO forms were more frequent in the pre-newborn screening era, probably because many EO patients died without a definitive diagnosis. Clinicians should be aware that LO patients may not be detected by newborn screening. To avoid newborn screening false positives and false negatives for ClbC/D defects, we are now implementing second tier tests that will allow the detection of the increased levels of total homocysteine and methylmalonic acid, and lower C3/C2 and C3 cut-off levels, respectively. Our data corroborate the importance of molecular genetic testing for an accurate genetic counselling and for a confirmatory diagnosis of CblC/D, thus reducing the need for complex, costly, laborious, and time-consuming biochemical/enzymatic tests.

Footnotes

Acknowledgements

We are grateful to the European Network and Registry for Homocystinurias and Methylation Defects – EHOD project (N°2012_12_02) for the partial support of this work. We are also grateful to Dr Francisca Coutinho for editorial assistance and to clinicians and patients.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Banerjee

Gherasim

Padovani

. The tinker, tailor, soldier in intracellular B12 trafficking. Curr Opin Chem Biol 2009; 13: 477–484.

Rosenblatt

Fenton

Inherited disorders of folate and cobalamin transport and metabolism. In: Scriver

Beaudet

Sly

(eds). The metabolic and molecular bases of inherited disease, 8th ed. New York: McGraw-Hill, 2001, pp. 3897–3933.

Watkins

Rosenblatt

. Inborn errors of cobalamin absorption and metabolism. Am J Med Genet C Semin Med Genet 2011; 157C: 33–44.

Rosenblatt

Aspler

Shevell

. Seashore, Clinical heterogeneity and prognosis in combined methylmalonic aciduria and homocystinuria (CblC). J Inherit Metab Dis 1997; 20: 528–538.

Huemer

Kožich

Rinaldo

. Newborn screening for homocystinurias and methylation disorders: systematic review and proposed guidelines. J Inherit Metab Dis 2015; 38: 1007–1019.

Lerner-Ellis

Tirone

Pawelek

. Identification of the gene responsible for methylmalonic aciduria and homocystinuria, CblC type. Nat Genet 2006; 38: 93–100.

Nogueira

Aiello

Cerone

. Spectrum of MMACHC mutations in Italian and Portuguese patients with combined methylmalonic aciduria and homocystinuria, CblC type. Mol Genet Metab 2008; 93: 475–480.

Fischer

Huemer

Baumgartner

. Clinical presentation and outcome in a series of 88 patients with the CblC defect. J Inherit Metab Dis 2014; 37: 831–840.

Morel

Lerner-Ellis

Rosenblatt

. Combined methylmalonic aciduria and homocystinuria (CblC): phenotype–genotype correlations and ethnic-specific observations. Mol Genet Metab 2006; 88: 315–321.

10.

Hannibal

Kim

Brasch

. Processing of alkylcobalamins in mammalian cells: a role for the MMACHC (cblC) gene product. Mol Genet Metab 2009; 97: 260–266.

11.

Suormala

Baumgartner

Coelho

. The CblD defect causes either isolated or combined deficiency of methylcobalamin and adenosylcobalamin synthesis. J Biol Chem 2004; 279: 42742–42749.

12.

Coelho

Suormala

Stucki

. Gene identification for the CblD defect of vitamin B12 metabolism. N Engl J Med 2008; 358: 1454–1464.

13.

Froese

Kopec

Fitzpatrick

. Structural insights into the MMACHC-MMADHC protein complex involved in vitamin B12 trafficking. J Biol Chem 2015; 290: 29167–29177.

14.

Stucki

Coelho

Suormala

. Molecular mechanisms leading to three different phenotypes in the CblD defect of intracellular cobalamin metabolism. Hum Mol Genet 2012; 21: 1410–1418.

15.

Miousse

Watkins

Coelho

. Clinical and molecular heterogeneity in patients with the CblD inborn error of cobalamin metabolism. J Pediatr 2009; 154: 551–556.

16.

Weisfeld-Adams

Bender

Miley-Åkerstedt

. Neurologic and neurodevelopmental phenotypes in young children with early-treated combined methylmalonic acidemia and homocystinuria, cobalamin C type. Mol Genet Metab 2013; 110: 241–247.

17.

Rashed

Ozand

Bucknall

. Diagnosis of inborn errors of metabolism from blood spots by acylcarnitines and amino acids profiling using automated electrospray tandem mass spectrometry. Pediatr Res 1995; 38: 324–331.

18.

Baumgartner

The cobalamins. In: Hall

(ed). Methods in hematology, Edinburgh, New York: Churchill Livingstone, 1983, pp. 181–181.

19.

Weisfeld-Adams

Morrissey

Kirmse

. Newborn screening and early biochemical follow-up in combined methylmalonic aciduria and homocystinuria, CblC type, and utility of methionine as a secondary screening analyte. Mol Genet Metab 2010; 99: 116–123.

20.

Liu

Yang

Chang

. Mutation spectrum of MMACHC in Chinese patients with combined methylmalonic aciduria and homocystinuria. J Hum Genet 2010; 55: 621–626.

21.

Ahrens-Nicklas

Serdaroglu

Muraresku

. Cobalamin C disease missed by newborn screening in a patient with low carnitine level. JIMD Rep 2015; 23: 71–75.

22.

Huemer

Scholl-Bürgi

Hadaya

. Three new cases of late-onset CblC defect and review of the literature illustrating when to consider inborn errors of metabolism beyond infancy. Orphanet J Rare Dis 2014; 9: 161–161.