1.Clinical features of the patients
- Clinical features of c.1663G>A group
The detailed information on each patient carrying c.1663G>A (p.A555T) mutation is summarized in table 1. Up to now, their median age was 2.8 years old, ranging from seven months to 14 years old. Among these patients, 26 cases were diagnosed by using a positive newborn screening, 2 patients were diagnosed because of onset of the disease (P18, P30), and 2 cases (P9, P13) were diagnosed because of sibling MMA diagnosis (P8, P12). Only seven cases encountered disease onset, among which four cases were subjected to newborn screening (P15, P19, P24, P29) and 3 cases were not (P9, P18, P30).
1.1.1 Clinical manifestation
Among the seven cases with a symptomatic presentation, 3 cases showed acute disease onset (P19, P29, P30), while four cases presented developmental delay or intellectual impairment progressively, without acute symptoms (P9, P15, P18, P24). As for the three cases with acute disease onset, the symptoms showed no specificity, including difficult feeding, vomiting, diarrhea, muscle weakness, lethargy, and convulsion. The median age of disease onset was 16 months old, ranging from three days to 22 months old.
The patient P19 was diagnosed by newborn screening and was treated by L-carnitine since one month of age. An acute attack of metabolic acidosis induced by upper respiratory tract infection, showing the symptoms of vomiting, diarrhea, lethargy, and muscle weakness happened at 22 months of age. After symptomatic treatment, the patient gradually recovered in two weeks. P29 had undergone newborn screening at birth. However, the individual did not accept treatment until a disease attack induced by respiratory tract infection at 16 months old, manifested with fever, vomiting, diarrhea, muscle weakness, lethargy, convulsion, and coma. Auxiliary examination indicated metabolic acidosis. P30 was observed to present difficult feeding, poor weight gain, muscle weakness, and metabolic acidosis at 1 month of age.
As for the four patients that showed progressive developmental delay or intellectual impairment, P9 was diagnosed because the younger sibling (P8) was confirmed with MMA in newborn screening, and then, detection of the gene confirmed the diagnosis of P9. A mild developmental delay was observed when diagnosis was confirmed at 23 months old. The patient could not speak until the age of 26 months. The patient P15 showed developmental delay and could not walk until the age of 16 months. P18 was not subjected to newborn screening at birth 8 years ago. The individual could not walk until 24 months old, could not speak until 36 months old, and was diagnosed with MMA at 57 months old because of intellectual impairment. P24 was diagnosed by newborn screening and was treated irregularly. This patient presented progressive developmental delay without attack of metabolic acidosis. Furthermore, the child could not walk properly at the age of 17 months.
1.1.2 Treatment and vitamin B12 responsiveness
All the 30 patients accepted treatment after diagnosis. More than half of patients (16/30) received the treatment before being two months old. Most patients were compliant with B12 treatment. Up to December 2019, except for P1 whose parents rejected further treatment after 4 months of treatment, and two patients (P20,P30) who was lost of follow-up, all the other 27 patients received treatment with hydroxocobalamin and/or L-carnitine. A total of nine patients accepted the intramuscular injection with vitamin B12 only, with hydroxocobalamin the single dose of 1–10mg, and the frequency of once every two days to once every two weeks. A total of 13 patients accepted oral L-carnitine only, with the dose of 50–100mg/(kg·d). A total of five patients accepted the treatments with both hydroxocobalamin and L-carnitine, with similar doses as above. Nearly half of the patients were maintained on the specialized diet therapy.
The treatment of the seven cases with symptomatic presentation were shown as follows. P9 were treated with L-carnitine and hydroxocobalamin in the first year after diagnose and then with hydroxocobalamin only. P15 was diagnosed by newborn screening and was adhered to a low protein diet. The vitamin B12 and L-carnitine were added until 1 year old. P18 has adhered to the treatment with L-carnitine and hydroxocobalamin regularly after diagnose. The patients P19 and P24 were diagnosed by newborn screening and treated with L-carnitine and hydroxocobalamin since the age of about one month. L-carnitine was kept at a dose of 100mg/(kg·d) routinely. The hydroxocobalamin treatment for P24 was stopped unexpectedly when the patient was two months old. As for the patient P29, hydroxocobalamin and L-carnitine were used immediately at the disease onset. Of note, this patient recovered quickly and showed no attacks after treatment. As for P30, Vitamin B12 and L-carnitine were used before the patient gave up the treatment at 2.5 months of age.
With regard to vitamin B12 responsiveness, nine patients have completed the vitamin B12 loading test in our clinic, eight with hydroxocobalamin and one with methylcobalamin (P12). According to C3/C2 ratio and methylmalonic acid before and after the loading test, five cases met the completely responsive type, with C3/C2 ratio and methylmalonic acid decreased more than 50%; three cases met the partly responsive type, with C3/C2 ratio and methylmalonic acid decreased but less than 50%; one case (P16) had inconsistent changes trend in C3/C2 ratio (decrease) and methylmalonic acid (increase).
There were remaining 21 patients did not undergo vitamin B12 loading test. There were two reasons: 1)Some patients were transferred to our clinic after vitamin B12 treatment had began in other hospitals. 2) The C3/C2 ratio or methylmalonic acid was as low as normal range or slightly above the normal range. Except for P13 who never used vitamin B12, we evaluated of the therapeutic effect of B12 for the other 20 patients, by referring to the judgment standard of the vitamin B12 loading test. The C3/C2 ratio and methylmalonic acid of the patients on the most recent were compared with the two indexes on presentation. As results, 11 cases showed C3/C2 ratio and methylmalonic acid decreased by more than 50% (11/21); six cases showed C3/C2 ratio and methylmalonic acid decreased but less than 50% (6/21); three case (P6, P25, P28), as well as P13, had inconsistent changes trend in C3/C2 ratio (decrease) and methylmalonic acid (increase).
Because of many confounding factors, such as health status, treatment situation, urine concentration, the diet on the detect day and so on, the urine metabolites fluctuates easily. We consider the changes in blood indicators (C3/C2 ratio) are more meaningful. So we deemed the four (P16, P6, P25, P28) patients with decreased blood C3/C2 ratio and increased urine methylmalonic acid were also responsive to vitamin B12. Therefore, all the nine patients finished the loading test was responsive to vitamin B12tand all the 20 patients received vitamin B12 treatment responded well. There was an obvious overall trend that vitamin B12 treatment is highly effective for these patients (100%, 29/29).
1.1.3 Prognosis
As for the current health condition, following up until December 2019, 2 patients could not be followed up (P20, P30). Twenty-four patients (24/30, 80%) were healthy and lived a normal life asymptomatically. Four patients showed progressive developmental delay or intellectual impairment (P9, P15, P18, and P24). P15 and P24 were diagnosed in newborn screening while P9 and P18 did not undergo newborn screening. The Gesell developmental schedule scores of P9 at 22 months old were gross motor 86, fine motor 66, adaptive 93, language 44, personal-social 62. Similarly, the Gesell developmental schedule scores of P15 at 25 months old were gross motor 65, fine motor 63, adaptive 58, language 69, personal-social 56. The Gesell developmental schedules scores of P24 at 17months old were gross motor 78, fine motor 69, adaptive 78, language 49 and personal-social 49. After treatment, the intelligence of P18 improved slowly. The WISC developmental schedule scores of this patient at 7.2 years were verbal IQ 44, performance IQ <40, and total IQ<40.
The patients P19 and P29 who had experienced disease onsets display currently normal intelligence after treatment. The diagnosis of MMA was achieved, and vitamin B12 was used when P30 was 2.5 months old. Unfortunately, individual’s parents gave up the treatment at 3 months old and we failed to confirm the current health status of P30.
1.2 Clinical features of non-c.1663G>A group
The detailed information on the enrolled patients of the control group is summarized in table 2. There were 23 boys and 13 girls in the control group, with a median age of 2.9 years old, ranging from 12 months to 12.5 years old. Twenty-two patients did not perform MS/MS expanded newborn screening and were diagnosed because of the onset of the disease. Further,14 patients were diagnosed by newborn screening, in which 10 patients showed disease onset during their next treatment. The median age for disease onset of these patients was 3 months old. The symptoms were manifested in varied forms, including difficult feeding, vomiting, diarrhea, poor weight gain, muscle weakness, dyskinesia, lethargy, convulsion, coma, mental retardation, jaundice, anemia, metabolic acidosis, and progressive developmental delay. All the 36 patients accepted treatment after diagnosis. Their median age of beginning treatment was 2 months old. Except for 1case lost during the follow up (C17) and 3 cases of death prior to vitamin B12 treatment (C1, C7, and C8), the remaining 32 patients were subjected to the vitamin B12 loading test. It was seen that 12 patients were responsive to vitamin B12, while the other 20 patients were unresponsive to vitamin B12, yielding a total vitamin B12 responsive rate of 38%, which is significantly lower than that for the c.1663G>A group. As for the current health condition under treatment, 24 patients showed developmental delay or intellectual impairment (67%), five patients are living healthy lives asymptomatically, and six patients died from disease onset at ages ranging from seven days to 18 months.
1.3 Comparison of clinical features in two groups
The detailed clinical features comparison of c.1663 G>A group and non-c.1663 G>A group are summarized in table 3. There were significant differences in presentation and clinical severity between the two groups. The proportions of disease onset were 7/30 (23%) in c.1663 G>A groupand 32/36 (89%) in non-c.1663 G>A group, with a significant difference in the incidence rate between the two groups (P< 0.0001). As for the treatment, the vitamin B12 responsive rate was 100% (29/29) in c.1663G>A group and only 38% (12/32) in non-c.1663G>A group (P< 0.0001). As for the prognosis, most of the patients carrying c.1663G>A (p.A555T) remained asymptomatic under treatment (24/30). In contrast, most of the patients carrying other mutations manifested developmental delay or intellectual impairment (24/36), and six patients died. A significant difference was also detected in the prognosis of the two groups (P< 0.0001). In conclusion, compared with patients carrying other mutations, c.1663G>A (p.A555T) - coding patients exposed lower morbidity, later disease onset, milder clinical phenotype, better vitamin B12responsiveness, and thereby better prognosis.
- Biochemical features of the patients
As the biochemical makers, the blood C3, blood C3/C2 ratio, urinary methylmalonic acid, urinary methylcitric acid before and after treatment in c.1663G>A and non-c.1663G>A groups are presented in table 1 and table 2. The comparative results between c.1663G>A and non-c.1663G>A groups are summarized in table 3.
On the primary state before treatment, C3, C3/C2, methylmalonic acid, methylcitric acid in c.1663G>A group showed a slight increase over the normal range. In contrast, the 4 biochemical indexes of the non-c.1663G>A group showed a prominent increase than the normal range in most patients.
All the 4 biochemical markers in c.1663G>A group before treatment were much lower than those in non-c.1663G>A group, with a significant statistical difference (table1, table 2, table 3). Similar changes were observed in the two groups after the treatment. The levels of C3, C3/C2 ratio, methylmalonic acid, methylcitric acid in c.1663G>A group after treatment were much significantly lower than those in non-c.1663G>A group. As for c.1663G>A (p.A555T) carrying patients, the levels of C3, C3/C2 ratio, methylmalonic acid, methylcitric acid decreased remarkably after treatment, compared with those before treatment. However, in non-c.1663G>A group, C3, methylcitric acid decreasedwhile C3/C2 ratio, methylmalonic acid increased after treatment, compared with those before treatment, respectively (table1, table 2, table 3). These data indicated that the therapeutic effect in c.1663G>A (p.A555T) carrying patients was much better than that in non-c.1663G>A (p.A555T) carrying patients.
- Geographical distribution of c.1663G>A (p.A555T)
In order to explore the geographical distribution of the mutation c.1663G>A (p.A555T), we analyzed the origin of the 328 patients harboring mutations in the MMUT gene, and calculated the mutation frequency of c.1663G>A (p.A555T) in different regions. As shown in table 4, the mutation frequency varies notably depending on the region. The population of the Shandong province was found to display the highest mutation frequency, followed by Hebei province and Henan province.
- Pathogenic effects of mutation c.1663G>A (p.A555T)
We assessed the potential pathogenicity of the mutationc.1663G>A (p.A555T) by MutationTaster, PolyPhen-2, Proven and SIFT software. It was predicted to be “disease-causing”, “probably damaging”, “deleterious”, and “damaging”, respectively. Furthermore, we evaluated its pathogenicity by the WinterVar database (http://wintervar.wglab.org), according to the ACMG 2015 guideline. It was defined as “likely pathogenic” with the score “PM1+PM2+PP3+PP5”.
The mutationc.1663G>A (p.A555T) leads to analanine into a threonine change at position 555 in the MCM protein. The mutant threonine residue is larger and less hydrophobic than the wild-type alanine residue. The website “HOPE” (https://www3.cmbi.umcn.nl/hope) was used to model the conceivable 3D conformations of wild-type and mutant MCM proteins.. The wild-type residue 555 is located in an α-helix ranging from amino acid 548 to 557, and is buried in the core of the protein. The mutated residue threoninedoes not prefer α-helices as a secondary structure for the steric hindrance effect and the loss of hydrophobic interactions, and thereby affects the function of the protein [21].