Two Brothers With Mild Congenital Erythropoietic Porphyria Due to a Novel Genotype FREE

Congenital erythropoietic porphyria (CEP, or Günther disease) is a rare autosomal recessive disease caused by the deficient activity of uroporphyrinogen III synthase (URO-synthase), the fourth enzyme in the heme biosynthetic pathway.¹ Normally, URO-synthase metabolizes hydroxymethylbilane to uroporphyrinogen III, the tetrapyrrole precursor of heme.^{2 - 3} The deficiency of URO-synthase activity leads to nonenzymatic conversion of hydroxymethylbilane to uroporphyrinogen I,a nonphysiologic substrate that is converted to coproporphyrinogen I, and then these porphyrinogen I isomers are oxidized to uroporphyrin I (URO-I) and coproporphyrin I (COPRO-I), which are phototoxic compounds. Elevated levels of URO-I and COPRO-I in erythrocytes result in massive hemolysis, and the released porphyrins accumulate in peripheral blood, skin, bone, and teeth.

The tissue deposition of URO-I and COPRO-I is responsible for the clinical manifestations of CEP. These manifestations include transfusion-dependent chronic hemolysis and secondary hypersplenism, erythrodontia, osteolytic skeletal changes, and cutaneous photosensitivity with blistering and extensive scarring. Although most reported CEP cases have presented with early-onset severe disease, there has been notable variation and genetic heterogeneity in their phenotypic presentation.³ This ranges from moderate forms of the disease presenting in childhood to milder adult-onset disease with mild cutaneous manifestations.^{3 - 5}

Recent advances in genetic analysis have shed light on the genetic basis of the phenotypic heterogeneity. Since the identification and localization of the URO-synthase gene,⁶ more than 35 different CEP-causing mutations have been identified.⁷ These include missense and/or nonsense mutations, splicing defects, small and large deletions or insertions, and complex rearrangements. Except for 4 mutations in the promoter region, all the mutations occurred in the coding region of the gene.^{3 ,7} Most mutations occurred in one or a few unrelated families, with the exception of the severe C73R missense mutation, which occurred in nearly 30% of the disease alleles.^{3 ,8} In vitro studies³ that characterized the residual URO-synthase enzymatic activity expressed by different mutations revealed that the level of residual activity directly correlated with the disease severity; namely, higher residual URO-synthase activity levels resulted in milder CEP phenotypes, and lower residual enzymatic activity levels resulted in a more severe phenotype.

In this article, we describe 2 brothers who have a mild CEP phenotype and a previously unreported genotype, −76G→A/G225S. The mild phenotype in our patients further supports previous work indicating a strong genotype-phenotype correlation in this disease. Although the residual URO-synthase activity expressed by the severe G225S missense exonic mutation was only 1.2% of normal,⁹ the residual enzyme activity associated with −76G→A promoter substitution was 54% of mean activity expressed by the normal allele in vitro, substantially higher than that seen in severe cases.^{3 ,10}

A 16-year-old white adolescent boy was referred to the Department of Dermatology, Henry Ford Hospital, Detroit, Mich, for skin fragility, blister formation, and mild scarring of the face, arms, and legs since age 3 years. He reported having mild cutaneous photosensitivity and frequent reddish discoloration of the urine. Occasional splenomegaly had been detected, but no frank episodes of hemolysis had occurred. He had been evaluated for occasional abdominal pain, but no cause was found. No joint pains or neuropsychiatric symptoms were reported. His younger brother had a similar condition (see case 2), whereas his parents and other siblings had no symptoms of photosensitivity or any cutaneous findings.

Physical examination showed hypertrichosis over the periorbital area (Figure 1), several shallow scars on the forehead, and scattered crusted erosions with mild nonmutilating scarring over the dorsum of the hands (Figure 2) and extensor surfaces of the forearms. There was no hepatosplenomegaly.

On the basis of these findings, cutaneous porphyria was suspected, and the patient’s plasma, erythrocyte, urine, and fecal porphyrin levels and erythrocyte uroporphyrinogen decarboxylase activity were evaluated. The results were consistent with the diagnosis of CEP because a marked increase in total urinary, plasma, erythrocyte, and fecal porphyrin levels was detected, with a predominance of URO-I and COPRO-I isomers in the urine and plasma and COPRO-I in the stool (Table). Uroporphyrinogen decarboxylase activity was normal, thus excluding the diagnosis of porphyria cutanea tarda or hepatoerythropoietic porphyria. Analysis of genomic DNA revealed a novel combination of mutant URO-synthase alleles, a paternally inherited promoter mutation (−76G→A) and maternally inherited missense exonic mutation (G225S).^{9 - 11}

The 4-year-old brother of the patient in case 1 was referred to the Department of Dermatology, Henry Ford Hospital, complaining of pruritic and painful erosions and blistering with mild scarring on sun-exposed skin since 18 months of age. He was the product of a full-term pregnancy and uncomplicated delivery. He was born healthy and had not had any hemolysis or organomegaly. He had had no abdominal or joint pains. With a notable exception of his 16-year-old brother (case 1), his parents and 2 other siblings, including a fraternal twin, did not have any similar complaints.

Physical examination revealed crusted erosions on the nasal bridge and forehead (Figure 3). Erosions, superficial scars, hyperpigmentation, and milia were observed on the nape of the neck, dorsum of the hands (Figure 4), and extensor extremities. Hypertrichosis was noted on the face, forearms, and legs; no lesions were noted in the non–sun-exposed areas. There was no hepatosplenomegaly.

Similar to case 1, cutaneous porphyria was suspected and the patient’s plasma, erythrocyte, urine, and fecal porphyrin levels and erythrocyte uroporphyrinogen decarboxylase activity were evaluated. The results were consistent with a diagnosis of CEP; a marked increase in total urinary, plasma, erythrocyte, and fecal porphyrin levels was detected with a predominance of URO-I and COPRO-I isomers in the urine and plasma and COPRO-I in the stool (Table). Erythrocyte uroporphyrinogen decarboxylase activity was normal. As shown in Figure 5, genetic testing further supported the diagnosis of CEP because the −76G→A/G225S genotype was identified.

The results of mutation detection studies of all family members are shown in Figure 5. One of the father’s URO-synthase alleles had the −76G→A promoter mutation, and one of the mother’s URO-synthase alleles contained the G225S missense mutation. A 15-year-old male sibling of the affected brothers carried the G225S mutation, whereas the 4-year-old fraternal twin brother of the patient in case 2 carried the promoter mutation. Except for the 2 affected siblings, none of the family members had any signs or symptoms of CEP.

Both patients were advised to avoid sun exposure and to use inorganic sunscreens that contained zinc oxide or titanium dioxide. They were also provided with inorganic sunscreen formulated as a paste with coloring that could be matched to skin color.¹² This was more acceptable and practical for use. The patients and their family members also received genetic counseling. They subsequently reported better control of their cutaneous eruptions. Except for limiting sun exposure, the affected brothers have been able to lead normal lives.

Congenital erythropoietic porphyria is an inborn error of heme biosynthesis that results from the markedly deficient, but not absent, activity of URO-synthase and the resulting accumulation of uroporphyrinogen I and coproporphyrinogen I, which are oxidized to the nonphysiologic porphyrin I isomers, URO-I and COPRO-I.^{1 - 3} Light in the Soret region band (400-410 nm) activates the phototoxic URO-I and COPRO-I isomers, resulting in skin fragility, blisters, milia, and mutilating scars. Other clinical manifestations include erythrodontia, alopecia, and hypertrichosis. Patients with severe CEP present with mutilating cutaneous involvement starting in the neonatal period, are transfusion dependent, and have hypersplenism. Patients with late-onset disease usually present with only mild cutaneous lesions^{4 - 5} and may develop thrombocytopenia and myelodysplasia^{5 ,13 - 14} ; they are not transfusion dependent. On the basis of the in vitro expression of the individual mutant alleles found in cases of CEP, it has been shown that the residual URO-synthase activity in milder cases is higher than the residual URO-synthase activity in severe cases.^{3 ,15 - 16}

To date, more than 35 URO-synthase mutations are listed in the Human Gene Mutation Database.^{3 ,7} Patients have been described who are homoallelic or heteroallelic for different mutations. For example, the most common mutation, C73R, has occurred in severely affected individuals when homoallelic (C73R/C73R) and in moderately affected patients when heteroallelic (C73R/A104V).³ Thus, the genotype is directly correlated with the amount of expressed URO-synthase activity, as previously shown by in vitro studies, and is also directly correlated with disease severity.^{3 ,9 - 10} In our patients, the novel genotype consisted of 2 different mutations, a promoter region point mutation (−76G→A) in one allele and an exonic missense mutation (G225S) in the other. Previous in vitro expression of the −76G→A allele and the G225S allele in Escherichia coli revealed residual URO-synthase activities of 53.9% and 1.2%, respectively, of the mean in vitro wild-type expressed activity.^{9 - 10} This suggests that the mild phenotype seen in our patients is due to the compensatory residual activity of the URO-synthase activity produced primarily by the −76G→A allele. The delay in onset of disease in our patients (3 years of age in case 1 and 18 months of age in case 2) may be due to the slow rate of accumulation of the porphyrin isomers in reaching a threshold pathogenic level. A similar phenomenon has been postulated for patients with myelodysplastic disease who develop late-onset CEP, which is usually mild.^{5 ,13 - 14}

The management of the cutaneous manifestations of CEP consists of strict sun avoidance, use of inorganic sunscreens, and prompt treatment of secondary bacterial skin infections. Inorganic sunscreens that contain zinc oxide or titanium dioxide are best if applied as a paste, but this may be impractical and cosmetically unappealing for most patients. Other treatment modalities with limited efficacy include oral beta carotene, intravenous hematin, plasmapheresis, hydroxychloroquine, oral charcoal, and other porphyrin binders.^{2 ,17} For severely affected patients who are transfusion dependent, erythrocyte transfusion,¹⁸ hydroxyurea,¹⁹ and splenectomy²⁰ appear effective, at least until adolescence; stem cell or bone marrow transplantation has been curative.^{20 - 21} In vitro gene transfer studies have demonstrated restoration of URO-synthase activity, suggesting the future potential effectiveness of gene therapy.^{22 - 24}

In summary, we describe 2 brothers with CEP who have an unusually mild phenotype due to a novel heterozygous genotype that consists of a promoter mutation (−76G→A) and an exonic missense mutation (G225S). The mild disease in these patients is a further example of the clinical heterogeneity seen in CEP and is additional proof that in vitro enzyme expression studies can correlate genotype to phenotype.

Correspondence: Henry W. Lim, MD, Department of Dermatology, Henry Ford Medical Center, New Center One, 3031 W Grand Blvd, Suite 800, Detroit, MI 48202 (hlim1@hfhs.org).

Accepted for Publication: August 21, 2005.

Author Contributions:Study concept and design: Berry, Desnick, and Lim. Acquisition of data: Berry, Desnick, Astrin, Shabbeer, Lucky, and Lim. Analysis and interpretation of data: Desnick, Lucky, and Lim. Drafting of the manuscript: Berry, Desnick, and Lim. Critical revision of the manuscript for important intellectual content: Berry, Desnick, Astrin, Shabbeer, Lucky, and Lim. Obtained funding: Desnick. Administrative, technical, and material support: Berry, Desnick, Shabbeer, and Lim. Study supervision: Desnick, Astrin, and Lim.

Financial Disclosure: None.

Funding/Support: This work was supported in part by grants from the National Institutes of Health, Bethesda, Md, including research grant RO1 DK26824 and grant 5 MO1 RR00071 to the Mount Sinai General Clinical Research Center from the National Center for Research Resources.