Genotype in the diagnosis of 21-hydroxylase deficiency: who should undergo CYP21A2 analysis?

AIMS

to confirm the diagnosis of 21-hydroxylase deficiency (21-OHD) by the analysis of CYP21A2 gene in infants with clinical and/or biochemical features of 21-OHD in order to clarify which patients to submit to genetic analysis; to analyze the genotype-phenotype concordance in these infants.

SUBJECTS AND METHODS

We studied 25 children with clinical and/or biochemical features of 21-OHD. All of them and their parents were submitted to genetic analysis of CYP21A2. Patients were classified in 3 groups according to mutations' severity: severe (group A), moderate (group B) or mild (group C).

RESULTS

CYP21A2 gene mutations were found in 17 children. Whereas all infants of groups A and B presented a classical form of 21- OHD, children of group C had a non-classical form of 21-OHD. Four infants resulted heterozygotes and 4 children were wildtype. A girl clinically presenting a non-classical form of 21-OHD resulted compound heterozygote with one of the mutations not described in literature (R25W) and whose residual enzymatic activity is not already known. All affected children presented a 17-OHP level after ACTH stimulation greater than 100 nmol/l. We found an optimal concordance between 17-OHP levels after ACTH test and genotype.

CONCLUSIONS

CYP21A2 analysis permitted to confirm the diagnosis of 21-OHD in 68% of our children. To improve this percentage we suggest to perform the CYP21A2 analysis only when 17-OHP after ACTH test is greater than 100 nmol/l. Moreover, we found an optimal genotype-phenotype concordance in the 21-OHD patients.

[Influence of thyroid morphology on psychomotor development in patients with congenital hypothyroidism during 8 year follow-up]

AIM

The aim of this paper was to evaluate the impact of thyroid morphology on auxological and neuropsychological development in children affected by congenital hypothyroidism (CH), treated with levothyroxine, up to 8 years of age.

METHODS

Fifty-three children affected by CH divided into 3 groups on the basis of thyroid morphology determined at birth: patients with athyreosis (N=17), with ectopic gland (N=23), with in situ thyroid (N=13). The developmental quotient (DQ) was evaluated by the Brunet-Lezine test up to 3 years, and intelligent quotient (IQ) by the Terman-Merril test after 3 years of age.

RESULTS

DQs at one year in athyreotic patients are lower (P<0,05) as compared to those determined in patients with other thyroid morphology. Later on these patients still showed lower DQ and IQ values than in other groups, although statistically not significant.

CONCLUSION

Thyroid morphology seems to be fundamental in psychomotor development, in fact patients with athyreosis show a transient impairment at one year of age. This difference could be transient or to have repercussions on adult. Individualization of the starting dose of levothyroxine on the basis of thyroid morphology, could be useful.

A technique of mRNA extraction and labeling from circulating lymphocytes of children treated with growth hormone replacement therapy for microarray analysis

GH replacement therapy exhibits a wide spectrum of response in terms of growth. Nevertheless, standardized doses are still given in clinical practice. In order to optimize the therapy, it is necessary to identify its markers of responsiveness. Given the presence of GH receptors in the circulating lymphocytes, accessible by means of a simple blood withdrawal, blood becomes the tissue of choice as a source of RNA for in vivo gene expression analysis. Hence, the purpose of the present paper is to develop a method of preparation of RNA from lymphocytes suitable for microarray analysis, focusing on the reduction of the blood volume withdrawal in order to perform the analysis on pediatric subjects. After lymphocyte isolation and total RNA extraction from 6 ml of blood, we carried out an amplification procedure preserving the relative abundance of each transcript. Thereafter, we hybridized the labeled amplified RNA on an oligo chip (Human 30K A, MWGBiotech), but the unsuccessful detection of a good signal to noise ratio indicates that labeled RNA is still insufficient. Therefore, we suggest performing pools of total RNA from different subjects with similar responsiveness to the therapy. It can be speculated that, upon comparison of the obtained data with those derived from pools of controls properly responding to the therapy, specific hallmarks of the condition of low responsiveness, devoid of inter-individual variability, will be evidenced.

Lack of association between thyrotropin receptor gene polymorphisms and subclinical hypothyroidism in children

Subclinical hypothyroidism is defined as a serum TSH level above the statistically set reference range, associated to normal free thyroid hormone concentrations. Genetic and environmental factors contribute to the inter- and intra-individual biological variations of TSH levels, sometimes leading to uncertainty of treatment in the clinical practice, especially when moderate elevations above the upper limit of the reference range are considered (5< TSH <10 mIU/l). In this view, the study of association between subclinical hypothyroidism and possible molecular effectors, such as polymorphisms in the TSH receptor (TSHR) gene, could be interesting. In this paper, we analyzed the TSHR gene polymorphisms in 103 hyperthyrotropinemic infants. A control group of 120 newborns of the same ethnic background was used to evaluate the frequencies of each polymorphism in the population. We found a statistically significant difference in the allelic frequency of the P52T polymorphism, being that the T variant was more represented in the control group (p=0.03). However, no significant results have been obtained in the analysis of the association between genotypes and serum TSH levels. In conclusion, we analyzed 7 polymorphic variants of TSHR gene in subclinical hypothyroidism. The only significant result refers to the allelic frequency of A in the P52T polymorphism, which is statistically reduced when compared with that of a control group.

Osteogenesis imperfecta: clinical, biochemical and molecular findings

Mutations in COL1A1 and COL1A2 genes, encoding the alpha1 and alpha2 chain of type I collagen, respectively, are responsible for the vast majority of cases of osteogenesis imperfecta (OI) (95% of patients with a definite clinical diagnosis). We have investigated 22 OI patients, representing a heterogeneous phenotypic range, at the biochemical and molecular level. A causal mutation in either type I collagen gene was identified in 20 of them: no recurrent mutation was found in unrelated subjects; 15 out of 20 mutations had not been reported previously. In two patients, we could not find any causative mutation in either type I collagen gene, after extensive genomic DNA sequencing. Failure of COL1A1/COL1A2 mutation screening may be due, in a few cases, to further clinical heterogeneity, i.e. additional non-collagenous disease loci are presumably involved in OI types beyond the traditional Sillence's classification.

Genetic analysis carried out on blood-spots of phenylalanine hydroxylase-deficient newborns detected by northeastern Italian neonatal screening

The aim of this work was to perform genetic analysis on 18 different blood-spot samples collected from neonates detected as hyperphenylalaninemic by Northeastern Italian screening program. DNA was extracted from blood-spots. Exons/introns of PAH gene were amplified by polymerase chain reaction (PCR), and PCR products were purified and sequenced with both forward and reverse primers. The most frequent mutations were IVS12nt1g>a (16.7%) and R408W, P281L and L48S (all together 11.1%). As expected, compound heterozygosity was the usual finding; homozygosity was found only in two patients with R158Q and IVS2nt5g>c mutations. The V230I mutation was reported for the first time in Italy. We found six previously described polymorphisms (V245V, IVS4nt47c>t, IVS2nt19t>c, IVS3nt-22c>t, IVS5nt-54a>g, and E280>Q280). To our knowledge, four genotypes were not previously described: R158Q/V230I present in one patient with classical PKU; and L48S/R408Q, A403V/IVS2nt-13t>g, and G272X/V230I present in patients showing HPA phenotype. Most of the mutations were located in the exons 12 and 7 and in exon/intron 2 (83.3% detection of total mutations in PKU or HPA patients of Northeastern Italy). From a practical viewpoint, the genetic analysis of blood-spots collected on Guthrie cards for neonatal screening for PKU could be a simple method to establish the genotype of neonates. Consequently, the genotype/phenotype correlation could lead to a more accurate diagnosis and prognosis for families.

Thyroxine hair content in congenital hypothyroidism and hyperthyroidism

Using the determination of thyroxine (T4) hair content, we studied 16 hypothyroid newborns diagnosed by means of our regional screening program, and five hypothyroid infants, undetected at birth, at diagnosis and after 3 months of substitutive therapy (8-10 microg/kg/day L-thyroxine in newborns; 15 microg/kg/day in infants), and 13 hyperthyroid adults. Hair T4 content was similar at diagnosis in hypothyroid newborns (2.6 +/- 2.3 pg/mg hair) and in infants undetected at birth (2.4 +/- 1.7 microg/mg hair), but very high only in the latter after therapy (23.2 +/- 3.9 microg/mg hair). Untreated hyperthyroid adults surprisingly evidenced lower hair T4 (0.4 +/- 0.2 microg/mg hair) than controls (1.5 +/- 0.3 microg/mg hair). We suggest these findings are due to differential tissue storage of thyroid hormone, related to the different blood T4 concentration. Therefore, T4 hair assay could be a non-invasive method to further assess thyroid status.

Determination of thyroxine in the hair of newborns by radioimmunoassay with high-performance liquid chromatographic confirmation

Hair analysis is often used in forensic toxicology to study, retrospectively, chronic exposure of individuals to drugs, and consequently newborn hair may become an ideal sample to study intrauterine exposure to xenobiotics as well as to endogenous compounds. As a tool to investigate a supposed maternal thyroxine (T4) supply to the congenital hypothyroid fetus, we devised to use the analysis of T4 extracted from newborn hair. In the present paper, the analytical method based on T4 extraction from hair followed by a radioimmunoassay is described. To verify the nature of the T4-like immunoreactive material present in newborn hair, it was further studied by HPLC fractionation with radioimmunoassay of the eluted fractions. On the basis of a clear correspondence between retention times of T4 standard and T4-immunoreactive compound extracted from hair, we assigned this immunoreactive material to T4. Then, we determined T4 hair concentrations in 19 control newborns at birth and 12 congenital hypothyroid infants at 22 days of life. Values obtained from hypothyroid infants (31.47+/-8.8 pg/mg(hair), mean+/-S.D.) were not significantly lower than those obtained from healthy newborns at birth (36.10+/-13.2 pg/mg(hair)). Such results are in agreement with the hypothesis of a maternal supply of thyroxine to the fetus through placental crossing.