Molecular, biochemical, and clinical analyses of five patients with carbamoyl phosphate synthetase 1 deficiency

Abstract Background Carbamoyl phosphate synthetase 1 deficiency (CPS1D) is a rare urea cycle disorder. The aim of this study was to present the clinical findings, management, biochemical data, molecular genetic analysis, and short‐term prognosis of five children with CPS1D. Methods The information of five CPS1D patients was retrospectively studied. We used targeted next‐generation sequencing to identify carbamoyl phosphate synthetase 1 (CPS1) variants in patients suspected to have CPS1D. Candidate mutations were validated by Sanger sequencing. In silico and structure analyses were processed for the pathogenicity predictions of the identified mutations. Results The patients had typically clinical manifestations and biochemical data of CPS1D. Genetic analysis revealed nine mutations in the CPS1 gene, including recurrence of c.1145C > T, five of which were firstly reported. Seven mutations were missense changes, while the remaining two were predicted to create premature stop codons. In silico and structure analyses showed that these genetic lesions were predicted to affect the function or stability of the enzyme. Conclusion We reported five cases of CPS1D. Five novel mutations of CPS1 gene were found. Mutations of CPS1 have private nature, and most of them are missense compound heterozygous. The mutation affecting residue predicted to interfere the catalytic sites, the internal tunnel, or the regulatory domain results in severe phenotype.

The mutations display a "private" nature between families, and specific relationship between genotype and phenotype is still under investigation, which further complicates the diagnosis. 3 Currently, most reported cases were newborns in their first few days of life or adults, whereas cases that have their initial symptoms starting from late neonates to puberty are scarce. 11,14,15 An approach to help establish diagnoses of CSP1D is needed from late neonate to puberty. In the present study, we report the clinical findings, management, biochemical data, molecular genetic analysis, and short-term prognosis of five children with CPS1D.

| Subjects
We retrospectively studied CPS1D patients from the Children's Hospital of Chongqing Medical University from January 2014 to January 2019. In total, 130 patients were detected to have HA >100 μmol/L. Among them, 33 patients were suspected to have congenital metabolic diseases and genetic analysis was conducted.
Eleven patients were diagnosed with OTC deficiency, and five of them were diagnosed with CPS1D. Their main clinical features and laboratory findings are summarized in Table 1. Blood taken for tandem mass spectrometry and urine gas chromatography tests were all in the acute phase. The study was conducted under the guidance of the Declaration of Helsinki 1975. All patients provided written informed consent for participation in the study, which was approved by the Children's Hospital of Chongqing Medical University Ethics Committee.

| Mutational analysis
Genomic DNA was extracted from peripheral blood lymphocytes using QIAamp DNA Blood Mini Kit (Qiagen). Targeted next-generation sequencing was performed according to experimental procedures previously described. 16    PolyPhen-2 grades the damaging effect of an amino acid substitution as "probably damaging" if the score is between 0.909 and 1, and "possibly damaging" if the score is between 0.447 and 0.908, and "benign" is the score is between 0 and 0.446.

| RE SULTS
b SIFT scores the substitution as ≤0.05 = damaging, which means that the change is predicted to affect protein function.
c MutPred2 scores the probability that the amino acid substitution is pathogenic. A score threshold of 0.50 would suggest pathogenicity.  Table 2. Sequencing data are shown in Figure S3. In total, nine mutations were detected, of which five were novel. None of the novel mutations we found have been listed in the public single nucleotide polymorphism databases (dbSNP) and have not been previously reported in HGMD and 1000 Genome Database. Of the total nine mutations, we identified seven missense mutations, one nonsense mutation, and one frameshift mutation. In silico prediction programs rate the changes as damaging.

| D ISCUSS I ON
In the present study, we presented our main clinical findings, bio-  should not be given for potential UCD patients with gastrointestinal symptoms or issues with consciousness. 17 We could not make a prognosis because of the short follow-up.
Waisbern et al reported that CPS1D was associated with global early developmental delays and autism spectrum disorder, but the sample size of the study was small. 27  loop, residues 1311-1333), which is a hub for NAG signal transmission to the catalytic machinery. 30 As predicted by Mutpred2, introduction of a hydrophobic residue at a positive charge position may alter ordered interface and changed DNA binding. Figure 2 shows the structure of the T'-loop; the mutations affect the protein structure by changing the hydrogen bonding and the spatial conformation. Therefore, the c.3949C > T (p.R1317W) and c.2865_c.2869delAAACT (p.T955Tfs*12) mutations may explain the neonatal onset and severe clinical manifestations observed in our patients.
p.V653G is located in the BPSD close to K-loop (L1β11-L1β12, residues 654-662). K-loop has a coordinated potassium ion in its center and is involved in intimate interactions with bound ADP. 30 The mutation is predicted to alter the metal binding, causing it to fail to catalyze the bicarbonate-dependent ATPase partial reaction. exhibits ≤2% of the activity at NAG saturation. 30 The loss of NAG affinity is pathogenic; therefore, this finding in patient 4 could explain his recurrent episodes and worst neuropsychological outcome.
In vitro glycerol has been proven to activate CPS1 without binding to the NAG-binding site, 28 indicating that the patients carrying this mutation could possibly be treated with designed pharmacological chaperones in the future.
The genetic lesion c.1271A > T (p.E379V) in Patient 5 located in the N-terminal domain. The substitution is predicted to alter the metal binding and interfere with the catalytic property of CSP1. In the second allele, we identified c.3928C > T (p.P1265S) at the heterozygous level. The lesion was located at the CPSD and adjacent to the possible carbamate tunnel. Carbamate migrates between the phosphorylation active centers through a water shielded tunnel to the active the center of the CPSD, where it is phosphorylated to carbamoyl phosphate. 30 Substitution of polar residues from non-polar residues may hamper the tunneling of intermediates. Neither mutations were interpreted to abolish the enzyme activity, which mirrored with the mild clinical observations from the patient 5. Longterm follow-up is still needed to evaluate the patient's biochemistry and neuropsychological changes.

| CON CLUS ION
In the present report, we present five children diagnosed with CPS1D. Genetic analysis revealed five new mutations that have not been previously reported, thereby expanding the mutational spectrum of the CPS1 gene. The overall clinical manifestations and biochemical data of our patient group were consistent with the variation of gene mutations, indicating that the current understanding of the CPS1 protein structure could be used to interpret unverified missense mutations. Our report aims to improve physician's awareness of CPS1D to aid early diagnosis and intervention, thus improving neurological outcomes in these patients.