Ascorbic acid deficiency in porphyria cutanea tarda

Ascorbic acid deficiency in porphyria cutanea tarda PETER R. SINCLAIR, NADIA GORMAN, STEVEN I. SHEDLOFSKY, CHARLES P. HONSINGER, JACQUELINE F. SINCLAIR, MARGARET R. KARAGAS, and KARL E. ANDERSON WHITE RIVERJUNCTION, VERMONT, HANOVER, NEW HAMPSHIRE,LEXINGTON, KENTUCKY,and GALVESTON,TEXAS

Porphyria cutanea tarda (PCT), the most common form of porphyria, is manifested as skin photosensitivity caused by excess hepatic production of uroporphyrin and heptacarboxylporphyrin. In experimental animal models, ascorbic acid modulates chemically induced uroporphyrin accumulation. The purpose of this study was to determine whether ascorbic acid is decreased in the plasma of patients with PCT. Plasma was obtained after an overnight fast from 21 PCT patients, 16 of whom were infected with hepatitis C virus (HCV), and from a separate group of 9 patients with HCV infection but not PCT. Thirteen PCT patients were studied when they had active disease and 8 after treatment-induced remission. Plasma ascorbic acid was low (<23 l~mol/L) in 11 (85%) of the 13 untreated PCT patients and deficient (<11 l~mol/L) in 8 (62%). Two patients with normal ascorbic acid levels (45 and 62 ~mmol/L) had consumed multivitamins. In 2 patients with deficient ascorbic acid, plasma levels returned to normal after phlebotomy treatment. Of the 8 patients studied during remission, 4 had normal ascorbic acid values and 4 were deficient (5 to 8 ~mol/L). Plasma ascorbic acid values were normal for all patients who had HCV but no PCT. These data suggest that plasma ascorbic acid concentrations are commonly low in PCT, but this decrease is unrelated to HCV infection. Ascorbic acid deficiency may be one of the factors that contributes to the pathogenesis of PCT. (J Lab Clin Med 1997;130:197-201)

Abbreviations: AU = alanine aminotransferase; AST = aspartate aminotransferase; HCV = Hepatitis C virus; PCT= porphyria cutanea tarda; URO = uroporphyrin

From the Veterans Administration Medical Center, White River Junction; the Departments of Biochemistry, PharmacologyfFoxicology, and Community/Family Medicine, Dartmouth Medical School, Hanover; the Veterans AdministrationMedical Center and the Department of Medicine, Universityof KentuckyMedical Center, Lexington; and the Department of Preventive Medicine and CommunityHealth, Universityof TexasMedicalBranch, Galveston. Supported by Grant ES-06263 from the National Institutes of Health, research funds from the Department of Veterans Affairs, United States Food and Drug Administration Office of Orphan Product Development (FD-R-000710), the American Porphyria Foundation, the National Center for Research Resources, National Institutes of Health (MO1 RR-00073), and the American Cancer Society (ACS-SIG17). Submitted for publication Aug. 20, 1996; revision submitted Feb. 18, 1997; accepted Feb. 25, 1997. Reprint requests: Peter R. Sinclair, PhD, Veterans Administration Medical Center (151), White River Junction, VT 05009. 0022-2143/97 $5.00 + 0 5/1/81991

p

orphyria cutanea tarda is the most c o m m o n form of porphyria in humans and is manifested by chronic, blistering lesions on areas of skin exposed to sunlight. 1'2 It is associated with a deficiency of uroporphyrinogen decarboxylase activity in the liver. Uroporphyrinogen and the intermediate products, hepta-, hexa- and penta-carboxylporphyrinogen, are substrates of uroporphyrinogen decarboxylase, but, if oxidized to the corresponding porphyrins, they cannot be further metabolized to heine. 3 Following massive accumulation of these porphyrins in the liver, excess amounts of porphyrins appear in plasma and urine. The usual treatment of P C T is repeated phlebotomy; this lowers the content of liver iron, which is thought to have a role in the pathogenesis of the disease. After phlebotomy, there is synthesis of new active uroporphyrinogen decarboxylase. 3 Alcohol use, estrogen use, 197

198

J Lab Clin Med August 1997

Sinclair et al.

and HCV infection are associated with PCT. 1'2 This porphyria is commonly an acquired disorder. In some cases there is an inherited (autosomal dominant) deficiency in uroporphyrinogen decarboxylase which can predispose to the development of the disease. The mechanism of inactivation of hepatic uroporphyrinogen decarboxylase in PCT is poorly understood. 3 However, in the biochemically similar rodent model produced by polyaromatic hydrocarbons, there is evidence for a role of a cytochrome P450 (specifically CYP1A2)-catalyzed oxidation of uroporphyrinogen. 3 This oxidation both depletes the substrate and appears to produce an inhibitor of uroporphyrinogen decarboxylase (see reference 3 for review). Iron administration accelerates the development of this uroporphyria. Recently it was found that ascorbic acid can inhibit CYPIA2-catalyzed oxidation of uroporphyrinogen in vitro and prevent the accumulation of U R O in cultured hepatocytes in which C Y P I A is induced. 4 Ascorbic acid deficiency and CYP1A2 induction in an ascorbic acid-requiring strain of rats result in massive U R O accumulation, whereas ascorbic acid intake sufficient to restore hepatic concentrations to normal prevents U R O accumulation. 5 The aim of this pilot study was to determine whether ascorbic acid deficiency is found in patients with PCT, as assessed by plasma concentrations of the vitamin. Since many PCT patients have concomitant H C V infection, 6 we also determined whether H C V infection in the absence of PCT is associated with low plasma concentrations of ascorbic acid. METHODS

Plasma samples were obtained from 21 patients with well-documented PCT from three centers, and from 9 patients with HCV but without PCT, from two centers. Plasma samples were also obtained from 24 healthy volunteers. All PCT patients had characteristic skin findings and the diagnosis of PCT was established by standard methods, including increased urinary URO and 7-carboxylporphyrin, increased plasma porphyrins, and increased fecal isocoproporphyrins. Urinary porphyrin excretions ranged from 1900 to 6600 p~gm/day. Heparinized blood samples for plasma ascorbic acid determinations were obtained after an overnight fast. Of the 21 patients with PCT, 11 were studied when they had active disease, 8 after treatment had produced remission, and 2 both before and after treatment. All patients experienced lowering of plasma and urinary porphyrin concentrations and remission of dermatologic symptoms after treatment by phlebotomy (19 patients) or by low-dose chloroquine (1 patient); 1 patient declined treatment. For each PCT patient sample, plasma was also obtained from a healthy subject at the same time to control for ascorbic

acid extraction and preservation techniques. Samples were protected from light and processed by centrifugation, separation of plasma, and extraction with four volumes of methanol in 1 mmol/L aqueous ethylenediamine tetraacetic acid (9:1). 7 Ascorbic acid was measured by high-performance liquid chromatography with electrochemical detection,a The coefficient of variation was <5% within assays and <10% between assays. Ascorbic acid in the methanol extract was stable at -80 ° C for at least 6 months. HCV infection was established by commercially available second-generation enzyme-linked immunosorbent assays. Addition of URO to normal plasma (1 ~mol/L URO, corresponding to the high concentrations found in PCT patients), followed by exposure to partial sunlight for 15 or 30 minutes at room temperature, resulted in 33% and 65% decreases in ascorbic acid concentration, respectively. At 0.1 txmol/L URO, there was negligible loss of ascorbic acid. During the preparation of plasma extracts containing 1 ixmol/L URO, there was negligible loss of ascorbic acid after 10 minutes' exposure to laboratory fluorescent lighting. One-way analysis of variance was used to assess differences in plasma ascorbic acid level among PCT patients, HCV patients, and assay controls. The protocol was approved by the institutional review boards of the participating centers. Written informed consent was obtained from all patients. RESULTS

According to accepted interpretive guidelines, human plasma ascorbic acid levels are defined as "deficient" when they are <11.4 ixmol/L (0.2 mg/dl) and as "low" when they are <23 ixmol/L (0.4 mg/dl). 9 Of the 13 patients with active PCT, plasma ascorbic acid levels were "low" in 11 (85%), "deficient" in 8 (62%), and normal for 2 patients (15%) at 45 and 62 txmol/L (Fig. 1). The two patients with normal ascorbic acid levels had been taking multivitamins before the study began. Ascorbic acid values for all subjects who had H C V but no PCT were normal (see Fig. 1). Although PCT patients were not systematically examined for early signs of ascorbic acid deficiency, none had obvious signs of scurvy. Major clinical features and plasma ascorbic acid concentrations in the individual PCT patients with active disease are shown in Table I. Most of the PCT patients in this study were infected with HCV, as has been reported in some other locations. 6 Three of the PCT patients were known to be infected with the human immunodeficiency virus, as has been reported previously by others) ° Seventy percent of the PCT patients smoked tobacco, and 46% drank alcoholic beverages. Three patients were taking phenytoin, and two of them were not consuming alcoholic beverages. Mild increases in AST and A L T were

J Lab Clin Med Volume 130, Number 2

found in almost all patients. One patient also had hemochromatosis, as indicated by a serum ferritin concentration of 2259 p.gm/L, and an increased transferrin saturation of 64%. All other patients had transferrin saturations lower than 60%. Serum ferritin was normal (<400 txgm/L) in six, and 400 to 1200 p~gm/L in three out of the ten patients for whom serum ferritin values were available. There was a weak inverse correlation of serum ferritin and plasma ascorbic acid (r 2 = 0.30; p = 0.10). The serum ferritin concentrations did not correlate with ALT values. Two patients who were studied both before and after remission had ascorbic acid concentrations of 3.4 and 1.3 ixmol/L before phlebotomy (see Table 1) and 27 and 54 txmol/L, respectively, after phlebotomy led to clinical remission. Of the 8 PCT patients studied only in remission, 4 (50%) had normal ascorbic acid values (42 to 60 ixmol/L), and 4 were deficient (5 to 8 txmol/L). These 8 patients in remission who had normal or low plasma ascorbic acid levels had no differences in clinical features. Ascorbic acid levels were also measured in the household companions of four of the PCT patients in Table 1, each of whom had low ascorbic acid levels. All four companions were found to have low plasma ascorbic acid concentrations (7 to 18 ~mol/L). To determine if whole blood or plasma from a PCT patient contained a substance other than URO that oxidizes or destroys ascorbic acid, the blood or plasma from one patient with active PCT was mixed 1:1 with blood or plasma of a control subject. The mixtures were then immediately processed for ascorbic acid content in the same way as other samples. The results were close to the average ascorbic acid levels of the samples that were mixed. Therefore, no substance in the patient's blood directly affected the ascorbic acid concentration. DISCUSSION

In this study plasma ascorbic acid levels were very low in a substantial number of PCT patients with active disease (11 of 13, or 84%). None of these patients had obvious signs or symptoms of ascorbic acid deficiency. Possible explanations for the low values include inadequate intake, impaired absorption or increased turnover of the vitamin, faulty re-reduction of oxidized ascorbic acid, and an effect of excess iron on ascorbic acid metabolism. The finding of normal ascorbic acid concentrations in 9 patients with HCV infection but without PCT suggests that this viral infection alone did not account for the low ascorbic acid concentrations in the PCT patients.

Sinclair et al.

120

0 0

199

*

0

II:

o z O o

60

O O

I- F

O

O O O

_m_

8

o

8 O

o

O

o

9

PCT

HCV

Assay Control

DEF

Fig. 1. Plasma ascorbic acid concentrations in patients with active PCT (n = 13), in patients with HCV and no PCT (n = 9), and in assay control subjects (n = 24). The HCV-infected patients had the following characteristics: all were male, ages ranged from 38 to 73 years, 3 were smokers, and 4 consumed alcohol at >1 drink per day. The two highest values for PCT patients and the two highest for HCV patients were from individuals who had been taking vitamin supplements. *Plasma ascorbic acid values for PCT patients were significantly different from those for HCV and control groups (p < 0.01), as determined by one-way analysis of variance.

Plasma ascorbic acid may be decreased in PCT in view of the recognized role of iron in the disease. Classic studies from South Africa showed that massive iron loading caused by consumption of alcoholic beverages brewed in iron vessels was associated with low plasma ascorbic acid levels and overt scurvy. 11 In our study, there was only a weak inverse correlation of serum ferritin with plasma ascorbic acid, and the majority of patients with low ascorbic acid values had normal serum ferritin levels. In [3-thalassemia patients with iron overload secondary to multiple transfusions, plasma ascorbic acid levels are significantly decreased, 12but not to the low concentrations found in the PCT patients in this study. As noted in this and other studies, 13 many PCT patients have little if any iron overload. The lack of correlation of iron stores and plasma ascorbic acid in humans is consistent with our finding of no effect of iron loading on plasma and liver ascorbic acid levels in ascorbic acid-requiring rats that were massively iron loaded by two different methods (Gorman N, Sinclair P, unpublished observations). We conclude that excess iron is unlikely to be the explanation for the low plasma ascorbic acid concentrations we observed in PCT patients. Low plasma levels were also found in the apparently healthy spouses, who probably consumed diets similar to those of their PCT companions. This suggests that the low plasma ascorbic acid levels in PCT patients may have resulted from an inadequate diet.

200

J Lab Clin M e d A u g u s t 1997

Sinclair e t a [ .

Table I. Characteristics of PCT patients with active disease

Case

Age (yr)

Sex

Alcohol

Smoker

Serum antibody to HCV

1 2 3 4 5 6 7 8 9 10 11 12 13

41 36 64 49 39 41 32 41 39 29 43 44 40

M M F M M M M F M F M M M

+ + + + -

+ + + ÷ ÷

+ + ÷ + +

+

+

+

+ + -

+ ÷ + -

+ + + +

Fe (i~gm/dl)

TIBC (~gm/dl)

%Fe Saturation

Ferritin (ng/ml)

AST (U/L)

ALT (U/L)

Plasma Asc (lxmol/L)

93 190 136 ND 105 99 60 116 72 124 152 190 178

264 373 315 ND 234 376 273 221 304 228 322 299 ND

35 51 43 ND 45 26 22 52 24 54 47 64 ND

211 1024 ND 198 294 489 137 631 233 303 1062 2259 ND

45 58 ND 40 73 61 73 65 31 42 85 79 72

60 89 ND 38 98 105 95 89 76 33 73 128 110

3.4 2.0 6.6 1.3 9.0 4.4 21 10 15 13 0.6 45 62

All patients were studied before treatment or early in treatment and had active disease as indicated by marked increases in urinary URO, 7-carboxylate porphyrins, and, when measured, plasma porphyrins. Positive smoking is defined as >0.5 pacWday and positive alcohol consumption as >1 drinWday. The range of plasma ascorbic acid was 25 to 101 i~mol/L in assay controls. Patients 3, 4, and 10 were being treated with phenytoin for coexisting seizures disorders. Patient 12 also had hemochromatosis and diabetes mellitus. Patients 7, 10, and 11 had positive antibodies for the human immunodeficiency virus. Asc, ascorbic acid; hiD, not done; TIBC, total iron-binding capacity.

A role for ascorbic acid in hepatic URO accumulation was first observed in experimental systems in which ascorbic acid was found to inhibit oxidation of uroporphyrinogen by the CYP1A2 isoform of cytochrome P450. 4'5 Furthermore, ascorbic acid was found to prevent the chemically induced accumulation of URO in cultured hepatocytes in which CYP1A was chemically induced.4 Massive URO accumulation occurs in an ascorbic acid-requiring rat mutant when CYP1A2 is chemically induced, but this accumulation is prevented by ascorbic acid intake sufficient to produce a normal concentration of hepatic ascorbic acid. 5 Studies with cultured hepatocytes and whole animals also suggest that modulation of uroporphyrinogen oxidation may be a previously unrecognized biologic function of ascorbic acid. Although no role for any P450 isoform has yet been established in human PCT, smoking, a recognized inducer of hepatic CYP1A2 in humans, ~4 was common among the patients studied. Smoking is also associated in humans with decreased plasma ascorbic acid levels and increased ascorbic acid turnover. 9 Alcohol consumption decreases plasma ascorbic acid levels, even when intake of the vitamin is normal. ~5 Although smoking and alcohol may have contributed to the development of PCT in many of our patients, the decreases in ascorbic acid concentration were much greater than those reported in association with smoking and alcohol consumption, suggesting that additional mechanisms, such as those discussed above, are responsible for the low

ascorbic acid levels in patients with PCT. Recently, others have reported low serum levels of antioxidant carotenoids in PCT patients, suggesting a role of oxidant "stress" in the pathogenesis of this disease. 16 The findings in this study suggest that treatment with ascorbic acid could be beneficial in PCT. Although ascorbic acid does not chemically reduce URO to uroporphyrinogen, it could inhibit further P450-catalyzed oxidation of uroporphyrinogen to URO and possibly prevent further oxidative damage to uroporphyrinogen decarboxylase.3 Ascorbic acid administration is noninvasive, inexpensive, and more convenient than phlebotomy or low-dose chloroquine. However, these latter treatments are highly effective and there is no clinical evidence that ascorbic acid therapy alone can produce a remission of the disease. Moreover, there may be adverse effects of ascorbic acid administration in individuals with excess body iron stores. 17 Ascorbic acid may also enhance iron absorption from the intestine. 9 Some physicians routinely prescribe low doses of ascorbic acid (100 rag/day) to patients with PCT or hemochromatosis during phlebotomy, to enhance iron mobilization, without apparent ill effect (Richard Hift, University of Capetown, personal communication). Therefore, it remains uncertain whether use of ascorbic acid is safe and efficacious in the treatment of PCT, and further investigation is required. In conclusion, the results of this study indicate that plasma ascorbic acid concentrations are low in patients with active PCT and that insufficient intake or altered metabolism of ascorbic acid may be one

J Lab Clin Med Volume 130, Number 2

of several predisposing factors for the development of this disease. Prevention of uroporphyrinogen oxidation may be a previously unrecognized biologic function of ascorbic acid. We thank Y. Maheshwari for providing blood samples from HCV patients and R. Tosheva, N. Alcock, D. Goeger, and M. Goeger for preparation of blood extracts. We also thank N. Azimi, D. Balestra, G. Elder, and R. Labbe for comments on the manuscript. REFERENCES

1. Anderson KE. The porphyrias. In: Zakim D, Boyer TD, editors. Hepatology. 3rd ed. Philadelphia: WB Saunders, 1996:417-63. 2. Elder GH. Porphyria cutanea tarda: a multifactorial disease. In: Champion RH, Pye R J, editors. Recent advances in dermatology 8. Edinburgh: Churchill Livingstone, 1990:55-69. 3. Elder GH, Roberts AG. Uroporphyrinogen decarboxylase. J Bioenerg Biomembr 1995;27:207-14. 4. Sinclair P, Gorman N, Walton HS, Bement WJ, Jacobs JM, Sinclair JF. Ascorbic acid inhibition of cytochrome P450catalyzed uroporphyrin accumulation. Arch Biochem Biophys 1993;304:464-70. 5. Sinclair P, Gorman N, Sinclair JF, Walton HS, Bement WJ, Lambrecht RW. Ascorbic acid inhibits chemically induced uroporphyria in ascorbate-requiring rats. Hepatology 1995; 22:565-72. 6. Fargion S, Piperno A, Cappellini MD, Sampietro M, Fracanzani AL, Romano R, et al. Hepatitis C virus and porphyria cutanea tarda: evidence of a strong association. Hepatology 1992;16:1322-6. 7. Levine M, Conry-Cantilena C, Wang Y, Welch RW, Washko PW, Dhariwal KR, et al. Vitamin C pharmacokinetics in healthy volunteers: evidence for a recommended dietary allowance. Proc Natl Acad Sci U S A 1996;93:3704-9.

Sinclair et al.

201

8. Berger J, Shepard D, Morrow F, Taylor AJ. Relationship between dietary intake and tissue levels of reduced and total vitamin C in the nonscorbutic guinea pig. J Nutr 1989;119: 734-40. 9. Jacob RA. Vitamin C. In: Roe J, editor. Modem nutrition in health and disease. Philadelphia: Lea & Febiger, 1994:43248. 10. Wissel PS, Sordillo P, Anderson KE, Sassa S, Savillo RL, Kappas A. Porphyria cutanea tarda associated with the acquired immune deficiency syndrome. Am J Hematol 1987; 25:107-13. 11. Lynch SR, Seftel HC, Torrance JD, Charlton RW, Bothwell TH. Accelerated oxidative catabolism of ascorbic acid in siderotic Bantu. Am J Clin Nutr 1967;20:641-7. 12. Livrea MA, Tesoriere L, Pintaudi AM, Calabrese A, Maggio A, Freisleben H-J, et al. Oxidative stress and antioxidant status in [3-thalassemia major: iron overload and depletion of lipid-soluble antioxidants. Blood 1996;88:3608-14. 13. Ratnaike S, Blake D, Campbell D, Cowen P, Varigos G. Plasma ferritin as a guide to the treatment of porphyria cutanea tarda by venesection. Australas J Dermatol 1988;29: 3-7. 14. Butler MA, Lang NP, Young JF, Caporaso NE, Vineis P, Hayes RB, et al. Determination of CYPIA2 and NAT2 phenotypes in human populations by analysis of caffeine urinary metabolites. Pharmacogenetics 1992;2:116-27. 15. Lecomte E, Herbeth B, Pirollet P, Chancerelle Y, Arnaud J, Musse N, et al. Effect of alcohol consumption on blood antioxidant nutrients and oxidative stress indicators. Am J Clin Nutr 1994;60:255-61. 16. Rocchi E, Stella M, Cassanelli M, Borghi A, Nardella N, Seinm Y, et al. Liposoluble vitamins and naturally occurring carotenoids in porphyria cutanea tarda. Eur J Clin Invest 1995;25:510-4. 17. Herbert V, Shaw S, Jayatilleke E. Vitamin C-driven free radical generation from iron. J Nutr 1996;126:1213S-20S.