Norrie disease gene polymorphism in Indonesian infants with retinopathy of prematurity

Introduction

Retinopathy of prematurity (ROP) is a major cause of blindness in preterm infants. The ROP incidence in infants with gestational ages of more than 26 weeks has been decreasing in developed countries due to improved quality of neonatal care. In contrast, an incidence is seen in infants born before 26 weeks.1 2 We showed that the incidence of ROP is higher in preterm infants born in Indonesia as compared with developed countries, and is also seen in infants with gestational ages of up to 34 weeks.3 Our results are in accordance with other studies in low-income and middle-income countries.4

ROP is a multifactorial disease. Gestational age, birth weight and use of supplemental oxygen are known risk factors. In Indonesia, we found that other factors such as sepsis, asphyxia, patent ductus arteriosus and multiple blood transfusions might also play a role.3 However, none of the presently identified risk factors can explain why some infants develop ROP while others do not. It is also not known why ROP regresses in most of the infants and progresses to severe forms in others.

A number of studies have suggested that genetic factors influence the development and severity of ROP.5–14 A study of monozygotic and dizygotic twins showed that genetic factors accounted for 70% of the variance in liability for ROP.5 Cooke et al 13 found that a mutation in the VEGF gene was related to the progression of ROP to threshold levels. Preterm infants who were homozygous for six alleles in the VEGF gene were twice as likely to develop threshold ROP.13 Mohamed et al 14 studied the genetic polymorphisms with candidate genes and genes previously genotyped for associations with preterm delivery. They found that IHH, AGTR1, TBX5, CETP and GP1BA were all associated with the development of ROP.14 Most studies described an association of mutations in the Norrie disease (ND) gene on the X chromosome with the progression of ROP.6–12 However, other studies could not confirm these results. ND, familial exudative vitreoretinopathy (FEVR) and ROP are retinal diseases which share similar phenotypes. ND is caused by a mutation in the Norrie (ND) gene, a condition which leads to a deficiency of Norrie protein.15 FEVR is characterised by reduced retinal vascularisation, retinal retraction and retinal detachment.

There are indications that racial differences exist in the development of ROP. Studies have shown a lower incidence of ROP in African–American infants and a higher incidence in Asian infants.16 That race might be involved in the development and progression of ROP was also found in studies in different populations in the USA.17

No study has evaluated if the higher number of ROP incidence in Indonesia might be related to the presence of mutations in the ND gene. Therefore, we investigated if mutations in the ND are present in infants with ROP in Indonesia.

Methods

This is a case–control study conducted in four hospitals (Harapan Kita Women and Children Hospital, Budi Kemuliaan Hospital, Awal Bros Tangerang Hospital, Royal Taruma Hospital) in Greater Jakarta, Indonesia. All infants admitted to the neonatal intensive care unit of each hospital and screened for ROP were included in the study. Most patients came from Jakarta, and the rest from West Java, Central Java, Sumatra and Kalimantan. Infants with birth weight less than 1500 g and/or gestational age of less than 32 weeks, and infants with a higher birth weight and/or gestational ages who needed respiratory support, were screened for ROP and included in this study.

Infants diagnosed with ROP type 2 or type 1, according to the Early Treatment for Retinopathy of Prematurity criteria,18 were included as patients, while infants without ROP were included as controls. A venous blood sample or a buccal swab was obtained from each infant.

Results

A total of 182 infants were enrolled in this study. The DNA amplification was performed successfully in 162 samples. The clinical characteristics of the infants are presented in table 2.

Birth weight and low socioeconomic status were slightly higher (p<0.05) in the control group, while other characteristics were not different between the groups (p>0.05). Of the infants with ROP, 24 had a birth weight of less than 1000 g and 15 a gestational age of less than 28 weeks. Forty-seven infants with ROP had a birth weight of 1000–1500 g and 43 infants a gestational age of 28–32 weeks. Twelve infants had a birth weight of more than 1500 g and 25 infants had a gestational age of above 32 weeks (table 3). Fifty-seven infants showed type 2 ROP, while 26 showed type 1 (table 3).

The PCR-RFLP process showed that all samples have wild-type ND genes (table 4). Mutations or polymorphism of C597A, L108P, R121W and A105T were not found in any of the samples collected both from infants with ROP and the control group.

DNA sequencing was performed on 36 samples to verify the results of PCR-RFLP and to detect V60E and C110G mutations where the PCR-SSP failed. Samples from 26 infants with ROP and 10 infants from the control group (without ROP) were processed in DNA sequencing. Observed sequences were compared with sequences in GenBank using the BLAST program. The results of BLAST showed 99% homology with Homo sapiens’ ND, RefSeqGene on chromosome X with an access code NG 009832.1. Furthermore, the sequences were aligned with GenBank’s sequences using ClustalW multiple alignment program. We found no differences in the arrangement of bases compared with the BLAST of the ND gene in six mutation sites, that is, C597A, L108P, R121W, A105T, V60E and C110G, in either cases or controls, in exon 3 (table 5).

Discussion

We did not find any mutations in the ND gene in the 83 infants with ROP as well as in the 79 infants in the control group. This result does not support the hypothesis that this gene may be involved in the development or progression of ROP in infants from Indonesia. Our result cannot explain why severe ROP is more frequent in Indonesian infants than in infants from other racial backgrounds.

There are two explanations for our findings. First, there might be variations in the presence of mutations among different countries in Asia. Our results are in line with a study from Korea which also found no mutations in the Norrie gene in infants with ROP.9 A study from Hiraoka et al 12 found one infant with a heterozygous mutation in the 5’ untranslated region (UTR) of exon 1 ND gene in 17 Japanese infants with advanced ROP and an insertion in the signal peptide of LRP5 in another infant.12 In another study Hiraoka et al 11 failed to find mutations in exons 2 and 3 and the 3’ UTR of the ND gene. They found in 100 patients with severe ROP two different mutations in exon 1.11 We did not study these mutations because these mutations were reported only once. Further studies are needed to confirm this possibility.

Second, previous studies have shown inconsistent results regarding the mutations in the ND gene and the development or progression of ROP.6–8 10 11 19 27 28 Shastry et al 6 described missense mutations (R121W and L108P) in 4 out of 16 infants with stage 4 or 5 ROP. In a follow-up study, they found one patient with an insertion in exon 1 and one with a deletion in the same exon in 100 infants with ROP.11 Haider et al 7 initially failed to find a relation between mutations in R121W and L108P in infants from Kuwait. Later, he found a polymorphism in the ND gene C597A in 23 out of 24 infants with advanced stage ROP.19 Talks et al 27 found two patients with a deletion in exon 1 of the ND gene in a group of 22 patients with stage 4 or 5 ROP. On the other hand, Dickinson et al 10 found no increased incidence of ND mutations in infants with ROP. Hutcheson et al 28 found five novel nucleotide changes in 143 infants with severe ROP, one in the 5’ UTR region of exon 2 and four in the 3’ UTR region of exon 3. They concluded that ND gene polymorphism might play a role in the pathogenesis of ROP, but does not appear to be a major causative factor. Recently, Dailey et al 8 found an FZD4 gene mutation, instead of mutations of the ND gene, to be associated with severe ROP. Hartnett29 recently noted a lack of consensus in reports of genetic variants associated with ROP, including VEGF, EPAS1, SOD and the WNT signalling pathway as well as the ND gene. Hartnett et al 30 reported a large study of two cohorts (817 and 543 infants) of US preterm infants with birth weight <1000 g and found an single nucleotide polymorphism (SNP) in the BDNF gene that encodes for brain-derived neurotrophic factor to be significantly associated with severe (threshold) ROP. No mutations in the ND gene were found. We also did not find any mutations in the ND gene in the infants with ROP.

Finally, DNA sequencing was performed on 36 infants. The samples were selected based on a simple random sampling method. The purpose of the sequencing was to detect polymorphisms in all mutations, including V60E and C110G. The χ² goodness of fit was used to identify the consistency between the characteristics of the 36 infants and those of the whole group. All variables showed a p value of >0.05, indicating that there were no significant differences between these 36 infants and the whole group. It can be concluded that the 36 samples are representative of the whole group.

A limitation of our study is that we limited the search for mutations in the ND gene to mutations that were described more than once in the literature. We realise that we may have missed mutations, mutations that might be found especially in infants in Asia. More studies will be needed to screen for more mutations than we did.

Based on our results and previously published studies, we conclude that it is very unlikely that the six polymorphisms in exon 3 of the ND gene studied in this paper are involved in the development or progression of ROP. Further studies must be conducted to find definitive explanations as to why ROP regresses in some patients, while it progresses in others.