Journal Logo

Article

Clinical and Biochemical Characteristics and Genotype-Phenotype Correlation in 143 Finnish and Russian Patients With Acute Intermittent Porphyria

Fraunberg, Mikael von und zu MD, PhD; Pischik, Elena MD; Udd, Lina; Kauppinen, Raili MD, PhD

Author Information
doi: 10.1097/01.md.0000152455.38510.af
  • Free

Abstract

INTRODUCTION

Acute intermittent porphyria (AIP) is an inherited metabolic disease with an autosomal dominant pattern of inheritance12. The disease is caused by a partial deficiency of porphobilinogen deaminase (PBGD, also known as uroporphyrinogen-1 synthase or hydroxymethylbilane synthase, EC 4.3.1.8, MIM 176000 [See footnote citing database sources.]), the third enzyme in the heme biosynthetic pathway. PBGD catalyzes tetramerization of porphobilinogen (PBG) yielding a linear tetrapyrrole, preuroporphyrinogen (hydroxymethylbilane), which is converted to cyclic uroporphyrinogen III in a reaction catalyzed by uroporphyrinogen synthase, or nonenzymatically to uroporphyrinogen I.

The PBGD gene (GeneBank M95623) has been assigned to chromosome 11q24. The gene is 10 kb in size and the coding region (1.3 kb) is spread over 15 exons4,9,31,33,44. At least 3 isoforms, arising from 2 different promoters, are transcribed and processed by alternative splicing4,10. The mRNA of the housekeeping (nonerythropoietic) isoform, containing exons 1 and 3 to 15, codes for an enzyme of 361 amino acids (M ~ 42,000), whereas the erythroid isoforms, encoded by exons 2 to 15, lack the first 17 amino acids of the amino terminus and alternatively include the 176 bp intron 2 (M ~ 40,000).

At the time of this writing, 214 PBGD mutations had been reported worldwide (http://www.uwcm.ac.uk/uwcm/mg/hgmd0.html)16. Fifty-three (25%) are small insertions or deletions that introduce a frameshift and a premature stop codon, 7 (3%) are small in-frame deletions, 72 (34%) are missense mutations, 54 (25%) change invariant nucleotides at splice sites, 23 (11%) produce stop codons, 3 (1%) are gross insertions or deletions, and 2 (1%) reside in the PBGD promoter region. Mutations are dispersed quite evenly throughout the gene, with the number of mutations in exons 10, 12, and 14 exceeding the others mainly due to their size16.

In AIP, PBGD activity is usually decreased by about 50% in erythrocytes, liver samples, fibroblasts, and lymphocytes24,35,36,39. Due to alternative splicing of the erythroid and housekeeping isoforms, erythrocyte PBGD activity is normal in a variant form of AIP, in which patients have a mutation in the first exon or intron16,26. Other biochemical abnormalities in AIP patients include overproduction and increased urinary excretion of porphyrins and their precursors, PBG, δ-aminolevulinic acid (ALA), and uro- and coproporphyrins16.

Biochemical and immunologic studies of PBGD conducted before DNA analysis suggested that AIP was caused by a heterogeneous group of mutations, and according to the ratio of enzyme polypeptide concentration and enzyme activity in erythrocytes, AIP was divided into 2 subtypes: cross-reacting immunologic material (CRIM) negative or CRIM positive1. CRIM-positive patients were further divided into 2 groups according to the relative amount of inactive protein5.

AIP patients experience acute attacks including abdominal pain, constipation, nausea and vomiting, and tachycardia and hypertension, which are signs of sympathetic activity. The symptoms are occasionally accompanied by acute motor polyneuropathy, seizures, and mental disorder8,27,38. Acute attacks are often precipitated by various exogenous factors such as drugs, alcohol, and infections, and endogenously by fasting and the menstrual cycle14. (For more information on acute porphyric attacks, see the report by Hift and Meissner 10a in this same issue.)

The phenotype of AIP varies within families, and the cause of the variable penetrance is unknown. Based on clinical manifestations, the morbidity varies both within and between families, and thus, we were interested to investigate whether a genotype-phenotype correlation exists among heterozygous AIP patients. Variation in the stability and residual activity of different PBGD mutants has been demonstrated in vitro3,6,30,32 and could support our hypothesis. As a few AIP patients with compound heterozygosity have survived20, the relatively high residual activity of the 2 mutations (R167W and R167Q) identified in these patients has had clinical implications6.

According to our hypothesis, PBGD mutants express different levels of the enzyme in the liver, which is the major organ related to the pathogenesis of an acute attack (Figure 1). Thus, individuals with certain mutations may have a higher susceptibility to manifest disease. The current prevailing theory concerning the pathogenesis of an acute attack is that precipitating factors induce aminolevulinic acid synthase (ALAS), the rate-limiting enzyme in the heme biosynthetic pathway7, resulting in accumulation of porphyrins and their precursors in the liver, from where they pass into plasma. The circulating precursors then affect other tissues such as neurons, causing impairment of the peripheral and central nervous systems23. Heme compounds, which act via a negative feedback mechanism in the liver, have been shown to be effective in alleviating biochemical abnormalities and clinical symptoms during an acute attack28.

F1-4
FIGURE 1:
Relationship between PBGD genotype, urinary excretion of porphyrins and their precursors, and severity of AIP. Precipitating factors and modifying genes play a major role in determining the phenotype. PBGD activity in the liver determines the plasma level of porphyrin metabolites, which cause neuronal damage. Measurement of urinary ALA and PBG indirectly reflects the severity of the biochemical abnormalities in plasma and liver tissue. Thick solid arrow = cause-effect relationship; thin solid arrow = observed correlation; dashed arrow = no correlation.

In the current study, we investigated the correlation between PBGD genotype and clinical phenotype of patients with 8 genotypes common in Europe, a genotype (1073delA) identified in the Koltta population located in northern Finland, and a 333 bp Alu insertion observed in Finland29. We studied 1) whether the occurrence of acute attacks correlates with the mutation type, 2) whether the patients' biochemical characteristics differ depending on the mutation type, and 3) whether the occurrence of symptoms can be predicted by gender, mutation type, and biochemical tests in remission.

PATIENTS AND METHODS

Patients and Biochemical Analyses

Since 1966, we have kept a registry of all Finnish patients known to have AIP27. After diagnosis of AIP in an index case was confirmed, family members were enrolled in the study between 1966 and 2003. A total of 287 AIP patients were diagnosed, representing 45 Finnish AIP families with 26 different mutations. In most families, ancestors could be traced back to at least the 19th century using church registers27. During 1995-2002, 12 Russian patients were admitted to the departments of neurology in the city hospitals of Saint Petersburg or local hospitals in northwestern Russia, and 11 additional family members were diagnosed. Nine different mutations in the PBGD gene were each revealed in 1 Russian family.

To investigate a genotype-phenotype correlation of different mutations, families sharing a mutation were pooled, with the material limited to those mutations where data were available from at least 6 patients for each mutation. Thus, 10 different mutations and 23 AIP families including 190 patients were selected. Of these, clinical data were not available from 6 patients, and an additional 5 patients were younger than 15 years at the time of the study. Thirty-six patients had died before 1966, and they were excluded.

The study population consisted of 143 heterozygous AIP patients: 88 patients were asymptomatic individuals who had never experienced an acute attack requiring hospitalization, and 55 patients had experienced 1 or several attacks (Figure 2). Diagnosis of AIP was based on mutation analysis (n = 81), characteristic clinical symptoms with elevated urinary PBG excretion (n = 90), low erythrocyte PBGD activity (n = 97), family history and typical clinical symptoms of AIP (n = 8), or pedigree analysis (n = 5). Erythrocyte PBGD activity was determined from red cells separated by centrifugation from a venous blood sample with heparin as an anticoagulant25,40. Urinary PBG and ALA were measured using the ALA/PBG Column test (Bio-Rad, CA), based on Mauzerall and Granick21. Urinary excretions of uro- and coproporphyrin and fecal excretions of copro- and protoporphyrin were measured according to Rimington34 and Holti et al11 until 1988, and thereafter using high-pressure liquid chromatography18,19. All biochemical measurements were performed during remission of acute symptoms (more than 6 mo after an acute attack) in adolescence or adulthood (patients aged 15-83 yr).

F2-4
FIGURE 2:
Graphic presentation of the groups studied at the time of diagnosis of AIP and during the follow-up period.

Biochemical data were not obtained from 24 patients (6 symptomatic and 18 asymptomatic), of whom 13 were screened only by mutation analysis. Information concerning acute attacks was obtained from hospital records for all patients who had required hospitalization since 1929 and/or by interviewing each patient. The criteria for an acute attack were the acute nature of symptoms, urinary excretion of PBG at least 4 times above the upper normal limit, and severe abdominal or other pain associated with 1 or more additional symptoms of acute porphyria28.

For mutation analysis (Table 1), DNA was isolated from blood leukocytes; amplified by polymerase chain reaction; and analyzed either by restriction enzyme digestion, whenever a specific enzyme was available, or by direct sequencing16. The analyses were repeated at least twice for each sample in the presence of negative and positive controls. Informed consent was obtained for all DNA testing, and the study protocol was approved by the Ethics Committee of the Department of Medicine, University Central Hospital of Helsinki, Helsinki, Finland.

T1-4
TABLE 1:
AIP Patients and 10 PBGD Mutations Commonly Identified in Finland and Northwestern Russia

Statistical Methods

The Fisher exact test was used to compare categorical variables. Continuous variables were analyzed with the Mann-Whitney U test when 2 groups were compared, or with Kruskal-Wallis 1-way ANOVA when more than 2 groups were compared simultaneously. Logistic regression with maximum likelihood estimation as optimization criteria was used to evaluate the association between dichotomous outcome variables (the occurrence of acute attacks) and covariates (mutation, gender, age at diagnosis, and results of biochemical tests). Statistical calculations were performed with SPSS version 10.04 and NCSS 2000.

RESULTS

Clinical Manifestations of AIP, 1966-2002

The study group included 143 patients, of whom 55 (38%) had experienced 1 or several acute attacks requiring hospitalization and 88 (62%) had been symptom-free throughout their lives (Tables 1 and 2). Thirty-seven patients died during the follow-up(1966-2002), 7 of them due to an acute attack (before 1980) and 7 due to hepatoma. Of the 7 patients who died during an acute attack, 3 died during a subsequent attack following diagnosis of AIP but without receiving proper counseling, 3 died during their first severe attack, which was at first misdiagnosed, and 1 experienced recurrent attacks with no diagnosis. In the latter patient, AIP was confirmed only later by pedigree analysis and typical clinical symptoms.

T2-4
TABLE 2:
Clinical Characteristics of AIP Patients in Each Mutation Group

Of the 55 symptomatic patients, 46 were symptomatic at the time of diagnosis, 5 were symptom-free, 3 died during their first severe attack, and 1 patient was diagnosed postmortem. After diagnosis of AIP was confirmed and information concerning precipitating factors was provided, 23 patients (42%) experienced recurrent attacks, resulting in severe neurologic symptoms with paresis in 6 cases (26%) (see Figure 2). Precipitating factors causing recurrent attacks were alcohol, 26% (3 men, 3 women); various drugs, 13% (1 man, 4 women); and menstrual cycle, 26% (6 women). Of the 46 patients who were symptomatic at the time of diagnosis, 18 (39%) experienced recurrent attacks. In contrast, of the 93 patients who were symptom-free at the time of diagnosis, only 5 (5%) experienced 1 or more attacks during the follow-up period (p < 0.0001).

Thirteen patients had experienced paresis before diagnosis of AIP, and 8 of these (61%) experienced subsequent acute attacks after diagnosis, resulting in paresis in 3 cases (38%). Three additional patients experienced paresis after diagnosis was confirmed: 1 was symptom-free at the time of diagnosis and 2 had previously experienced acute attacks without paresis. The high number of patients with previous and subsequent paresis suggests a higher risk, although not significant, of neurologic manifestations. Of the precipitating factors causing paresis in recurrent attacks, alcohol was involved in 1 case, menstrual cycle in 2 cases, and anticonvulsant and sulfonamides in 1 case, indicating that both endogenous and exogenous factors may be provocative. All except 1 recurrent attack complicated with paresis occurred before 1980, and none had been treated with heme arginate.

Of the 81 women, 38 (47%) had experienced acute attacks, whereas only 17 of 62 men (27%) had been symptomatic (p = 0.02) (Table 3). Of the symptomatic women, 17 (45%) had experienced recurrent attacks after diagnosis, resulting in paresis in 4 (24%) individuals. Of the symptomatic men, 6 (35%) had experienced recurrent attacks after diagnosis, resulting in paresis in 2 (33%) individuals.

T3-4
TABLE 3:
Clinical Manifestations in AIP Patients by Gender and Year of Diagnosis

The proportion of patients with acute symptoms decreased dramatically from 49% to 17% among patients diagnosed before and after 1980, respectively (see Table 3). The decrease in acute symptoms was even more prominent in male patients, since only 2 of 21 men (10%) diagnosed after 1980 experienced acute attacks requiring hospitalization. One of these had a single attack in 1979, while the other experienced several attacks after 1980 and died of hepatoma in 1999.

Correlation Between Genotype and Clinical Symptoms

The gender and age of patients at the time of diagnosis differed subtly among the mutation groups (median age, 22-43 yr; percentage of female patients, 33%-73%; see Tables 1 and 2). Twenty-two (13%) patients, including 1 symptomatic and 21 symptom-free patients, were more than 55 years of age at the time of diagnosis, and the percentage varied among the mutation groups (0%-44%). The high percentage of late-diagnosed patients with mutations R167W or R225G (9 patients) indicates that these patients were more likely to be symptom-free and to live longer. Median age at the onset of symptoms was 23 years, but was remarkably high, 45 years (range, 40-46 yr), among patients with the R167W mutation. The proportion of patients who experienced 1 or more acute attacks varied from 11% for patients with the R225G mutation to 67% for patients with the 97delA mutation (see Table 2).

Patients with normal erythrocyte PBGD activity26 were chosen as a reference group and a multivariate logistic regression analysis was performed, adjusted for age, gender, and year of diagnosis. The analysis revealed that the risk for acute attacks was highest for patients with the 97delA, R149X, and R173Q mutations, whereas the risk of attacks was lowest for patients with the R167W and R225G mutations. No significant differences existed among the other mutation groups. In a model including only mutation type as an explanatory variable, Cox & Snell R2 and Nagelkerke R2 were 0.135 and 0.183, respectively, indicating that the genotype determines less than 20% of the variation in clinical penetrance. Recurrent attacks were rare for patients with the R167W and R225G mutations, occurring in only 2 of 6 symptomatic patients, none of whom experienced paresis. In contrast, 17 (50%) of 34 patients with the R149X, R173W, R173Q, and R225X mutations experienced subsequent attacks after diagnosis of AIP, resulting in severe neurologic symptoms and paresis in 6 (36%).

The gender of the parent with AIP had no effect on transmission of the affected gene in the family. Of the 61 patients with paternal inheritance, 27 (44%) experienced clinical manifestations, while of the 60 patients with maternal inheritance, 16 (27%) manifested the disease. Thus, paternal inheritance predicted higher, although not significant (p = 0.057), risk for a manifest disease.

Correlation Between Genotype and Biochemical Characteristics

Table 4 shows the biochemical data of patients with different mutations. Urinary PBG and ALA were elevated in 90% and 61% of patients with AIP in remission. Urinary copro- and uroporphyrin were elevated in 65% and 75% of patients, respectively. Urinary coproporphyrin III (70%-90%) exceeded isoform I in remission, indicating that PBG was mainly converted into uroporphyrinogen III and coproporphyrinogen III enzymatically. Fecal copro- and protoporphyrin were elevated in only 1% and 10% of the patients, respectively (data not shown). No significant difference existed between these excretions in male and female patients.

T4-4
TABLE 4:
Mean Urinalysis Values (±SEM) in AIP Patients in Remission with Different Mutations

Urinary excretions of porphyrins and their precursors in remission varied significantly by mutation type (Figure 3). Patients with the R167W and R225G mutations resulted in significantly lower urinary PBG excretion in remission (PBG pooled: 47 ± 10 vs. 163 ± 21 μmol/L, p < 0.001). In patients older than 55 years old at the time of the diagnosis, urinary PBG excretion was >30 μmol/L in 9 patients with the 33G→T, R149X, R167W, R173W, and R173Q mutations; 9-30 μmol/L in 7 patients with either the R167W or R225G mutation; and <9 μmol/L in 2 patients with the R225X or InsAlu333 mutations. Similarly, urinary ALA, uroporphyrin, and coproporphyrin excretions were significantly lower in patients with the R167W and R225G mutations (ALA pooled, 40 ± 6 vs. 104 ± 11 μmol/L, p < 0.001; uroporphyrin pooled, 130 ± 40 vs. 942 ± 183 nmol/d, p < 0.001; coproporphyrin pooled, 399 ± 77 vs. 610 ± 54 nmol/d, p = 0.01).

F3-4
FIGURE 3:
Urinary excretions (mean ± SEM) of porphyrins and their precursors in AIP patients with the 10 most common mutations in Finland and northwestern Russia (n = 104). Individual values were calculated as the mean of 1-9 single measurements in adulthood. Dashed line denotes the normal value limit.

Correlation Between Urinary PBG Excretion and Clinical Symptoms

To investigate the correlation among urinary excretions, erythrocyte PBGD activity, and the occurrence of clinical symptoms, patients were divided into 2 groups: Group A with low PBG excretion, including patients with the R167W and R225G mutations, and Group B with high PBG excretion, including patients with the 33G→T, 97delA, InsAlu333, R149X, R173W, R173Q, R225X, and 1073delA mutations (Figure 4).

F4-4
FIGURE 4:
Erythrocyte PBGD activities (mean ± SEM) in AIP patients with the 10 most common mutations in Finland and northwestern Russia (n = 78). Individual values were calculated as the mean of 1-2 single measurements in adulthood. Dashed line denotes the normal value limit. *P value calculated without the 33G→T mutation.

Low PBG excretion correlated with less severe disease, as only 6 (17%) of 36 patients experienced acute attacks in Group A, whereas 49 (46%) of 107 patients were symptomatic in Group B (p = 0.003). Furthermore, subsequent attacks were less frequent in Group A, occurring in only 2 of 6 symptomatic patients (33%), none of whom experienced paresis. In contrast, 22 (45%) of 49 symptomatic patients experienced subsequent attacks in Group B, resulting in paresis in 8 patients (36%). Patients with the 97delA mutation had the highest clinical penetrance (67%) accompanied by a 3-fold excretion of urinary PBG, but recurrent attacks were rare. Regardless of mutation type, all patients with normal urinary PBG excretion (<9 μmol/L, n = 13) were symptom-free.

No Correlation Among Erythrocyte PBGD Activity, Urinary PBG Excretion, and Clinical Symptoms

Erythrocyte PBGD activities varied with mutation type (see Figure 4). Mean erythrocyte PBGD activities were 36 ± 1 and 39 ± 2 nmol/mg prot/h for Groups B (n = 49) and A (n = 29), respectively (p = 0.3). Patients with the 33G→T mutation, resulting in normal erythrocyte PBGD activity, were excluded from this comparison.

Among patients with the classical form of AIP, 19 (36%) of 53 patients with erythrocyte PBGD activity <40 nmol/mg prot/h experienced acute symptoms, whereas 11 (44%) of 25 with erythrocyte PBGD activity ≥40 nmol/mg prot/h were symptomatic. This indicates that the high erythrocyte expression level of the normal allele, together with the potential residual activity of the mutated allele, could not predict freedom from clinical manifestations. We compared clinical manifestations of the group including the 97delA and R173Q mutations, having the lowest erythrocyte PBGD activities, with the group including the other mutations. In the former group, 15 (44%) of 34 patients experienced acute attacks, whereas in the latter group 40 (37%) of 109 patients were symptomatic. This indicates that in patients with the classical form of AIP, erythrocyte PBGD activity does not correlate with PBG excretion in remission or with the severity of disease. Acute attacks were less frequent among patients with the variant form of AIP, occurring in only 2 of 12 (17%) patients with the 33G→T mutation, whereas 53 of 131 (40%) patients with the classical form of AIP were symptomatic.

CRIM-positive mutations, detectable by PBGD-specific antibodies1, were pooled and compared with CRIM-negative mutations, but no significant differences in urinary excretions of porphyrin precursors or enzyme activities were observed (data not shown).

Prediction of Clinical Symptoms

To predict the risk of clinical symptoms in symptom-free family members, we constructed a logistic regression model where the occurrence of clinical symptoms was explained by gender, mutation type, and/or biochemical tests. Since the patients' biochemical tests were correlated with each other, all urinary tests except PBG excretion, together with the mutation type, were excluded from the model as nonsignificant. The model was adjusted for gender and age at time of diagnosis, since women are known to be more prone to acute symptoms and patients diagnosed at a late age are more likely to be mutation-screened symptom-free individuals (Table 5).

T5-4
TABLE 5:
Statistical Results for the Stepwise Logistic Regression Giving the Best Model for Predicting the Occurrence of Acute Attacks in AIP Patients

Patients with the increased urinary PBG excretion were more prone to experience acute symptoms, and the likelihood of symptoms increased further if the excretion was markedly elevated (>100 μmol/L). According to our data, the likelihood of symptoms was reduced for patients with the 33G→T mutation causing the variant form of AIP.

DISCUSSION

A systematic follow-up of Finnish AIP patients conducted from the 1960s, together with extensive mutation screening, provided an excellent opportunity for genotype-phenotype analysis in AIP. Patients with the common mutations were recruited from northwestern Russia, where a systematic follow-up of AIP patients began in 1995. Since the biochemical and DNA analyses of the patients were carried out in the same laboratory, it was possible to unite Finnish and Russian AIP patients in 1 study group. Our series includes both symptomatic patients (38%) and phenotypically normal carriers (62%), and thus provides information about the clinical and biochemical outcome among AIP patients in general.

Clinical Characteristics of AIP Patients, 1966-2002

The proportion of symptomatic patients (38%) in the current series is comparable to that reported in previous family studies in which 20%-50% of patients experienced acute attacks2,37. We have shown that the occurrence of acute attacks requiring hospitalization has decreased markedly over the last 2 decades. The decline has been even more prominent among men, suggesting that the avoidance of exogenous precipitating factors is effective. The higher prevalence of clinical manifestations in women has been described also in other series37,38, and the onset of disease after puberty indicates the importance of endogenous factors such as sex hormones in the pathogenesis of acute symptoms22. The median age at the onset of clinical symptoms was 23 years, which is lower than described previously37,38, and may indicate an early diagnosis of AIP among our patients. This early diagnosis of AIP by biochemical analysis or mutation screening among family members along with proper counseling about the precipitating factors was instrumental in preventing subsequent attacks in 60% of previously symptomatic patients and in 95% of patients symptom-free at the time of diagnosis. The most common precipitating factor for a subsequent attack was heavy drinking in men and the menstrual cycle in women, but in many cases provocation was multifactorial (14). In the current series, approximately half of the symptomatic patients experienced subsequent attacks during follow-up but only a quarter of them experienced paresis, usually due to endogenous factors, indicating that avoiding exogenous precipitating factors does not fully prevent neurologic manifestations.

Mortality due to porphyria has declined during the last 2 decades, and death during an acute attack, which was previously due to late or incorrect diagnosis, has become rare. In contrast, 20% of deaths were due to hepatocellular carcinoma13, and that should be taken into account when AIP patients are followed.

Correlation Between Genotype and Phenotype

In the current study we demonstrate that urinary PBG excretion together with gender, age at time of diagnosis, and mutation type can predict the likelihood of acute attacks in AIP patients. Two mutations, R167W and R225G, resulted in a milder phenotype in terms of both clinical manifestations and biochemical abnormalities. The late onset of symptoms in our patients with the R167W mutation was comparable to 3 manifest cases among 24 Swedish AIP patients with the same mutation2, which may indicate that these patients are less vulnerable to endogenous factors than others.

The milder mutations manifested with lower penetrance, reduced recurrence rate, and absence of severe neurologic symptoms during an acute attack. In the Swedish study2, patients with the R167W mutation also experienced a lower penetrance (13%) and recurrence rate compared to those with the major Swedish mutation (W198X) and the R173W mutation. In contrast, milder mutations did not protect from hepatoma in the current series, indicating that different pathogenic mechanisms may result in hepatoma and acute attacks.

Patients with the R167W and R225G mutations had significantly lower urinary excretion of porphyrin precursors than the others. This was particularly evident among the symptomatic patients in remission. Our findings are in accordance with the results of symptomatic Swedish AIP patients with the R167W mutation, who excreted one-fourth the amount of urinary PBG and ALA than patients with the W198X mutation in remission2. Based on our results, low urinary excretion of porphyrin precursors predicts a mild phenotype, and normal urinary PBG excretion in the previous unselected series predicted freedom from acute attacks14,16. This supports the current prevailing theory that porphyrin precursors are crucial in the pathogenesis of clinical manifestations23.

Based on structural and functional analyses of PBGD mutants, substitutions of arginines 167 and 225, which display 5% and 16% residual activities when expressed in COS-1 cells30, could cause milder phenotypes, since these residues cause no dramatic changes in the catalytic active site or in the stability of the polypeptide. Although patients with the R225G mutation belonged to the same large family living in the capital region of Finland or abroad, it is unlikely that they would be less exposed to environmental precipitating factors than others. A more likely explanation is that high residual PBGD activity together with common modifying genes protects these patients from acute attacks.

In contrast, the R173W and R173Q mutations, which are located in the catalytic site of PBGD, show <1% residual activity30, indicating a more profound effect on enzyme function and presumably resulting in a more severe disease. However, the 1073delA mutation with 50% residual activity in vitro30 had low PBGD activity in vivo, indicating that the structure and function of the mutant polypeptide expressed in vitro may not be functionally identical with PBGD in vivo.

CONCLUSIONS

Based on the current patient series, a correlation among mutation type, severity of clinical manifestations, and urinary excretions of precursors can be established (see Figure 1). In contrast, no correlation between erythrocyte PBGD activity and these parameters could be demonstrated. Previously, PBGD activity was measured from a few AIP patients' liver samples, showing 42% of normal activity in remission and 11% during an acute attack39. Furthermore, the regulation of heme biosynthesis in bone marrow differs vastly from that of the liver and is controlled by the erythroid-specific ALAS43. Our current results and the previous findings suggest that erythrocyte PBGD activity does not reflect the enzyme activity in the liver, especially during an acute attack.

The phenotype of AIP is not determined only by the PBGD genotype, which may account for about 20% of clinical manifestations, but precipitating factors and modifying gene(s) also play a major role in the pathogenesis of acute attacks. Genes involved in sex hormone and drug metabolism as well as those involved in modulation of heme biosynthesis and porphyrin metabolism, for example, ALAS induction and porphyrin excretion, are good candidates for genes modifying the phenotype of acute porphyria42. Cytochrome P450 genes and genes encoding ATP-binding cassette efflux proteins-for example, multidrug resistance protein 2 (MRP2) and P-glycoprotein, which transport various substances across hepatocyte membranes into bile-are polymorphic17,41 and may explain in part individual variation in the hepatic clearance of porphyrins and their precursors.

Accumulation of porphyrins and their precursors in the liver is determined mainly by the level of PBGD activity and ALAS induction. Currently, this accumulation can be routinely measured only indirectly from urine or plasma samples. However, the threshold for the development of neuronal damage and acute symptoms depends not only on the plasma level of porphyrin metabolites, but also on the individual vulnerability of nerve cells in different neuronal systems.

REFERENCES

1. Anderson PM, Reddy RM, Anderson KE, Desnick RJ. Characterization of the porphobilinogen deaminase deficiency in acute intermittent porphyria. Immunologic evidence for heterogeneity of the genetic defect. J Clin Invest. 1981;68:1-12.
2. Andersson C, Floderus Y, Wikberg A, Lithner F. The W198X and R173W mutations in the porphobilinogen deaminase gene in acute intermittent porphyria have higher clinical penetrance than R167W. A population-based study. Scand J Clin Lab Invest. 2000;60:643-648.
3. Chen CH, Astrin KH, Lee G, Anderson KE, Desnick RJ. Acute intermittent porphyria: identification and expression of exonic mutations in the hydroxymethylbilane synthase gene. An initiation codon missense mutation in the housekeeping transcript causes "variant acute intermittent porphyria" with normal expression of the erythroid-specific enzyme. J Clin Invest. 1994;94:1927-1934.
4. Chretien S, Dubart A, Beaupain D, Raich N, Grandchamp B, Rosa J, Goossens M, Romeo PH. Alternative transcription and splicing of the human porphobilinogen deaminase result either in tissue-specific or in house-keeping expression. Proc Natl Acad Sci U S A. 1988;85:6-9.
5. Desnick RJ, Ostasievicz LT, Tishler PA, Mustajoki P. Acute intermittent potphyria: characterization of a novel mutation in the structural gene for porphobilinogen deaminase. Demonstration of noncatalytic enzyme intermidiates stabilized by bound substrate. J Clin Invest. 1985;76:865-874.
6. Edixhoven-Bosdijk A, de Rooij FW, de Baar-Heesakkers E, Wilson JH. Residual activity of human porphobilinogen deaminase with R167Q or R167W mutations: an explanation for survival of homozygous and compound heterozygous acute intermittent porphyrics. Cell Mol Biol (Noisy-le-grand). 2002;48:861-866.
7. Fraser DJ, Podvinec M, Kaufmann MR, Meyer UA. Drugs mediate the transcriptional activation of the 5-aminolevulinic acid synthase (ALAS1) gene via the chicken xenobiotic-sensing nuclear receptor (CXR). J Biol Chem. 2002;277:34717-34726.
8. Goldberg A. Acute intermittent porphyria. A study of 50 cases. Q J Med. 1959;28:183-209.
9. Grandchamp B, de Verneuil H, Beaumont C, Chretien S, Walter O, Nordmann Y. Tissue-specific expression of porphobilinogen deaminase: two isoenzymes from a single gene. Eur J Biochem. 1987;162:105-110.
10. Gubin AN, Miller JL. Human erythroid porphobilinogen deaminase exists in 2 splice variants. Blood. 2001;97:815-817.
10a. Hift RJ, Meissner PN. An analysis of 112 acute porphyric attacks in Cape Town, South Africa: evidence that acute intermittent porphyria and variegate porphyria differ in susceptibility and severity. Medicine (Baltimore). 2005;84:48-60.
11. Holti R, Rimington C, Tate BC, Thomas C. An investigation of porphyria cutanea tarda. Q J Med. 1958;27:1-17.
12. Kappas A, Sassa S, Galbraith RA, Nordmann Y. The porphyrias. In: Scriver CR, Beaudet A, Sly WS, Valle D, eds. The Metabolic and Molecular Bases of Inherited Diseases. 7th ed. New York: McGraw-Hill; 1995:2116-2127.
13. Kauppinen R, Mustajoki P. Acute hepatic porphyria and hepatocellular carcinoma. Br J Cancer. 1988;1:117-120.
14. Kauppinen R, Mustajoki P. Prognosis of acute porphyria: occurrence of acute attacks, precipitating factors, and associated diseases. Medicine (Baltimore). 1992;71:1-13.
15. Kauppinen R, Mustajoki S, Pihlaja H, Peltonen L, Mustajoki P. Acute intermittent porphyria in Finland: 19 mutations in the porphobilinogen deaminase gene. Hum Mol Genet. 1995;4:215-222.
    16. Kauppinen R, von und zu Fraunberg M. Molecular and biochemical studies of acute intermittent porphyria in 196 patients and their families. Clin Chem. 2002;48:1891-1900.
    17. Kerb R, Hoffmeyer S, Brinkmann U. ABC drug transporters: hereditary polymorphisms and phramacological impact in MDR1, MRP1 and MRP2. Pharmacogenomics. 2001;2:51-64.
    18. Li F, Lim CK, Peters TJ. Analysis of urine and faecal porphyrins by HPLC coupled to an advanced automated sample processor. Biomed Chromatogr. 1986;1:93-94.
    19. Lim CK, Peters TJ. Urine and faecal porphyrin profiles by reversed-phase high-performance liquid chromatography in the porphyrias. Clin Chim Acta. 1984;139:55-63.
    20. Llewellyn DH, Smyth SJ, Elder GH, Hutchesson AC, Rattenbury JM, Smith MF. Homozygous acute intermittent porphyria: compound heterozygosity for adjacent base transitions in the same codon of the porphobilinogen deaminase gene. Hum Genet. 1992;89:97-98.
    21. Mauzerall D, Granick S. The occurrence and determination of delta-aminolevulinic acid and porphobilinogen in urine. J Biol Chem. 1956;219:435-436.
    22. McColl KE, Wallace AM, Moore MR, Thompson GG, Goldberg A. Alterations in haem biosynthesis during the human menstrual cycle: studies in normal subjects and patients with latent and active acute intermittent porphyria. Clin Sci (Lond). 1982;62:183-191.
    23. Meyer UA, Schuurmans MM, Lindberg RL. Acute porphyrias: pathogenesis of neurological manifestations. Semin Liver Dis. 1998;18:43-52.
    24. Meyer UA, Strand LJ, Doss M, Rees AC, Marver HS. Intermittent acute porphyria-demonstration of a genetic defect in porphobilinogen metabolism. N Engl J Med. 1972;286:1277-1282.
    25. Mustajoki P. Red cell uroporphyrinogen I synthetase in acute intermittent porphyria. Ann Clin Res. 1976;8(Suppl 17):133-138.
    26. Mustajoki P. Normal erythrocyte uroporphyrinogen I synthase in a kindred with acute intermittent porphyria. Ann Intern Med. 1981;95:162-166.
    27. Mustajoki P, Koskelo P. Hereditary hepatic porphyrias in Finland. Acta Med Scand. 1976;200:171-178.
    28. Mustajoki P, Nordmann Y. Early administration of heme arginate for acute porphyric attacks. Arch Intern Med. 1993;153:2004-2008.
    29. Mustajoki S, Ahola H, Mustajoki P, Kauppinen R. Insertion of Alu element responsible for acute intermittent porphyria. Hum Mutat. 1999;13:431-438.
    30. Mustajoki S, Laine M, Lahtela M, Mustajoki P, Peltonen L, Kauppinen R. Acute intermittent porphyria: expression of mutant and wild-type porphobilinogen deaminase in COS-1 cells. Mol Med. 2000;6:670-679.
    31. Namba H, Narahara K, Tsuji K, Yokoyama Y, Seino Y. Assignment of human porphobilinogen deaminase to 11q24.1 - q24.2 by in situ hybridization and gene dosage studies. Cytogenet Cell Genet. 1991;57:105-108.
    32. Ong PM, Lanyon WG, Graham G, Hift RJ, Halket J, Moore MR, Connor JM. Acute intermittent porphyria: the in vitro expression of mutant hydroxymethylbilane synthase. Mol Cell Probes. 1997;11:293-296.
    33. Raich N, Romeo PH, Dubart A, Beaupain D, Cohen-Solal M, Goossens M. Molecular cloning and complete primary sequence of human erythrocyte PBG deaminase. Nucleic Acids Res. 1986;14:5955-5968.
    34. Rimington C. Quantitative determination of porphobilinogen and porphyrin in urine and faeces. Association of Clinical Pathology Broadsheet No. 21. Association of Clinical Pathology; 1958.
    35. Sassa S, Solish G, Levere RD, Kappa A. Studies in porphyria. IV. Expression of the gene defect of acute intermittent porphyria in cultured human skin fibroblasts and amniotic cells: prenatal diagnosis of the porphyric trait. J Exp Med. 1975;142:722-731.
    36. Sassa S, Zalar GL, Kappas A. Studies in porphyria. VII. Induction of uroporphyrinogen-I synthase and expression of the gene defect of acute intermittent porphyria in mitogen-stimulated human lymphocytes. J Clin Invest. 1978;61:499-508.
    37. Schuurmans M, Schneider-Yin X, Rüfenacht U, Schnyder C, Minder C, Puy H, Deybach J, Minder E. Influence of age and gender on the clinical expression of acute intermittent porphyria based on molecular study of porphobilinogen deaminase gene among Swiss patients. Mol Med. 2001;7:535-542.
    38. Stein JA, Tschudy DP. Acute intermittent porphyria: a clinical and biochemical study of 46 patients. Medicine (Baltimore). 1970;49:1-16.
    39. Strand LJ, Felsher BF, Redeker AG, Marver HS. Heme biosynthesis in intermittent acute prophyria: decreased hepatic conversion of porphobilinogen to porphyrins and increased delta aminolevulinic acid synthetase activity. Proc Natl Acad Sci U S A. 1970;67:1315-1320.
    40. Strand LJ, Meyer UA, Felsher BF, Redeker AG, Marver HS. Decreased red cell uroporphyrinogen I synthetase activity in intermittent acute porphyria. J Clin Invest. 1972;51:2530-2536.
    41. van der Weide J, Steijns LS. Cytochrome P450 enzyme system: genetic polymorphisms and impact on clinical pharmacology. Ann Clin Biochem. 1999;36:722-729.
    42. von und zu Fraunberg M, Timonen K, Mustajoki P, Kauppinen R. Clinical and biochemical characteristics and genotype-phenotype correlation in Finnish variegate porphyria patients. Eur J Hum Genet. 2002;10:649-657.
    43. Wada O, Sassa S, Takaku F, Yano Y, Uratta G, Nakao K. Different responses of the hepatic and erythropoietic delta-aminolevulinic acid synthetase of mice. Biochim Biophys Acta. 1967;148:585-587.
    44. Yoo HW, Warner CA, Chen CH, Desnick RJ. Hydroxymethylbilane synthase: complete genomic sequence and amplifiable polymorphisms in the human gene. Genomics. 1993;15:21-29.
    © 2005 Lippincott Williams & Wilkins, Inc.