Effect of iron and ascorbate on uroporphyria in ascorbate-requiring mice as a model for porphyria cutanea tarda§


  • See Editorial on Page 6

  • This is a US government work. There are no restrictions on its use.

  • §

    Potential conflict of interest: Nothing to report.


Excess hepatic iron is known to enhance both porphyria cutanea tarda (PCT) and experimental uroporphyria. Since previous studies have suggested a role for ascorbate (AA) in suppressing uroporphyria in AA-requiring rats (in the absence of excess iron), the present study investigated whether AA could suppress uroporphyria produced by excess hepatic iron. Hepatic URO accumulation was produced in AA-requiring Gulo(−/−) mice by treatment with 3,3′,4,4′,5-pentachlorbiphenyl, an inducer of CYP1A2, and 5-aminolevulinic acid. Mice were administered either sufficient AA (1000 ppm) in the drinking water to maintain near normal hepatic AA levels or a lower intake (75 ppm) that resulted in 70 % lower hepatic AA levels. The higher AA intake suppressed hepatic URO accumulation in the absence of administered iron, but not when iron dextran (300–500 mg Fe/kg) was administered. This effect of iron was not due to hepatic AA depletion since hepatic AA content was not decreased. The effect of iron to prevent AA suppression of hepatic URO accumulation was not observed until a high hepatic iron threshold was exceeded. At both low and high AA intakes, hepatic malondialdehyde (MDA), an indicator of oxidative stress, was increased three-fold by high doses of iron dextran. MDA was considerably increased even at low iron dextran doses, but without any increase in URO accumulation. The level of hepatic CYP1A2 was unaffected by either AA intake. Conclusion: In this mouse model of PCT, AA suppresses hepatic URO accumulation at low, but not high hepatic iron levels. These results may have implications for the management of PCT. (HEPATOLOGY 2007;45:187–194.)

Porphyria cutanea tarda (PCT) is the most common human porphyria and manifests as a skin disease. The disorder is characterized biochemically by massive hepatic production and accumulation of uroporphyrin (URO) (see Elder1 for review). The most commonly associated risk factors are consumption of alcoholic beverages and hepatitis C infections, but increased body iron stores also play a role.1 There is usually a mild hepatic siderosis in patients with PCT, and the disease is responsive to phlebotomy which presumably acts by removing iron.1 Some PCT patients have quite high levels of hepatic iron which are often associated with mutations in HFE as found in hereditary hemochromatosis.2, 3 Recently, it was reported that smoking, which induces CYP1A2, is highly prevalent in PCT patients.4

The accumulation of URO is attributed to uroporphyrinogen oxidation, which in rodents is catalyzed by a particular form of cytochrome P450, CYP1A2 (Fig. 1).5 This oxidation is considered to lead to the observed inactivation in the liver of the pathway enzyme, uroporphyrinogen decarboxylase,1 which converts uroporphyrinogen to coproporphyrinogen (Fig. 1). The inactivation of uroporphyrinogen decarboxylase is attributed to an inhibitor detected in the uroporphyric livers which may arise during oxidation of uroporphyrinogen (see Elder1 and Smith6 for reviews).

Figure 1.

Origin of hepatic URO accumulation from the heme synthetic pathway and the site of AA suppression of URO accumulation. The oxidation of uroporphyrinogen to uroporphyrin catalyzed by CYP1A2 is depicted by three arrows to represent the formation of partially oxidized porphyrinogen intermediates.7 AA inhibits the first step only.7 The specific steps stimulated by iron and the source and nature of the UROD inhibitor are not known. GLY, glycine; ALA, 5-aminolevulinate; UROgen, uroporphyrinogen; COPROgen, coproporphyrinogen; PROTOgen, protoporphyrinogen; PROTO, protoporphyrin; URO-D, uroporphyrinogen decarboxylase.

In work from this laboratory, ascorbic acid (AA) was found to prevent URO accumulation caused by treatment with 3,3′,4,4′-tetrachlorobiphenyl and 5-aminolevulinate in cultures of chick embryo liver cells7 and AA competitively inhibited microsomal CYP1A2-catalyzed oxidation of uroporphyrinogen (Fig. 1).7 Subsequently, in a spontaneous mutant rat that requires dietary AA, hepatic URO accumulation caused by treatment with 3-methylcholanthrene or hexachlorobenzene was found to be greatly enhanced when the animals were maintained on a very low AA dietary intake.8 In addition, a small clinical study showed that plasma AA levels were below normal in 11/13 patients with PCT.9 There was no correlation with body iron stores and the low AA levels were suggested to be due to dietary insufficiency.9 An earlier study also reported AA deficiency in PCT patients and lack of improvement with AA treatment.10

The effect of hepatic iron levels to modify the ability of AA to suppress uroporphyria was not examined in the previous rat study.8 Therefore, in the present study, we investigated the effect of iron-loading on development of uroporphyria in a line of AA-requiring mice in which part of the gene for the last enzyme of AA synthesis, gulonolactone oxidase, has been deleted.11 Administration of iron potentiates hepatic URO accumulation in many experimental protocols, including treatment with inducers of CYP1A2 such as 3,3′,4,4′,5-pentachlorobiphenyl (PCB126) (see Smith6 for review). Development of uroporphyria was compared in AA-requiring Gulo(−/−) mice consuming sufficient dietary AA to maintain near normal hepatic AA levels, or a lower intake that decreased hepatic AA levels by 70 %.


AA, ascorbic acid; CYP, cytochrome P450; MDA, malondialdehyde; PCB126, 3,3′,4,4′,5-pentachlorobiphenyl; PCT, porphyria cutanea tarda; URO, uroporphyrin.

Materials and Methods


Gulo(−/−) mice in C57BL/6 background were provided by the laboratory of Dr N. Maeda, University of N Carolina.11 Male Gulo(−/−) mice (8–10 weeks old, 18–25 g at the start of the experiment) were used, as indicated. Confirmation of the Gulo(−/−) genotype was performed by PCR with tail DNA using primers as described.11 Animals were housed in standard plastic cages and were maintained in 12-hour light/dark cycles, with free access to rodent chow (Teklad LM-485, containing 0.02 % Fe) from Harlan Teklad (Madison, WI). Gulo(−/−) mice were maintained on distilled water containing 1000 ppm AA prepared every 2 days. Some animals received 75 ppm or 300 ppm AA, as indicated.

5-aminolevulinic acid hydrochloride (Sigma, St Louis, MO) was dissolved in the drinking water at 2 mg/ml. Some mice were administered AA and 5-aminolevulinate together. No loss of 5-aminolevulinate, measured as described,12 or AA was detected after 24 hours at room temperature. Iron dextran solution (100 mg Fe/ml; Phoenix Pharmaceutical, St Joseph, MO), appropriately diluted in 0.9% NaCl, was injected at doses of 50–500 mg Fe/kg body weight ip, 3 days prior to starting the ALA treatment. PCB126 was from Ultrascientific, N. Kingston, RI and was injected (75 μg/kg body wt) after dissolving in minimal DMSO and diluting in corn oil. Figure 2 indicates the protocol used in all the experiments described in this paper. Pieces of liver for malondialdehyde (MDA) and AA determinations were kept at −80° C for up to one year prior to analysis.

Figure 2.

Experimental scheme for studies of the effect of AA and iron on uroporphyria in Gulo(−/−) mice. AA and 5-aminolevulinate are added in drinking water as described in Methods.

Liver homogenates and microsomes were prepared as described previously.13 Liver sections fixed in buffered formalin were stained by H and E and with Prussian Blue for histological analyses.

All animal protocols have been approved by the Dartmouth College and VA Animal Use Committees.


The porphyrin composition of liver homogenates was determined spectrofluorometrically.14 Protein concentrations were determined by the method of Lowry et al.15 using bovine serum albumin as a standard. Microsomal methoxyresorufin demethylase and uroporphyrinogen oxidation activities were measured as described previously.16 The presence of CYP1A1 and CYP1A2 proteins in liver homogenates was determined by immunoblotting, using a polyclonal antibody which detects both mouse CYP1A2 and CYP1A1 proteins (gift from Steve Wrighton, Lilly Research Labs, Indianapolis, IN), and was quantified as described.17 Non-heme Fe was measured by a modification of the method of Torrance and Bothwell, as described.17 MDA, as thiobarbituric acid-reactive substances, was measured fluorometrically in liver homogenates as described,18 using a standard curve of MDA generated by acid-catalyzed hydrolysis of 1,1,3,3 tetramethoxypropane (Aldrich, Milwaukee, WI).

Liver ascorbic acid was measured by reverse phase HPLC with electrochemical detection.19 Briefly, a 5 % (w/v) liver homogenate was prepared in 5 % (w/v) trichloroacetic acid/1 mM diethylenetriaminepentaacetic acid and, after centrifugation at 10,000 g for 5 minutes, AA was measured in the supernatant. All solutions were kept at 4°C and shielded from light. The mobile phase was 95 % 49 mM sodium acetate, 1 mM EDTA and 0.024 % octylamine, pH 4.0, and 5 % methanol. The column (C18 μBondapak, 3.9 × 300 mm, Waters, Milford, MA) was eluted at 0.9 ml/min. Detection was electrochemical (oxidation mode, + 0.7 V, carbon glassy electrode and Ag/AgCl reference electrode). Calculations of AA concentration were based on standard curves.

Statistical Analyses.

Results are presented as means ± SD. Significance was determined by one-way ANOVA, except where the SDs were significantly different, when a non-parametric test was used. P less than 0.05 was considered significant.


Dose Response of Dietary AA on Hepatic AA and URO Accumulation in Gulo(−/−) Mice in the Absence of Iron Administration.

We first determined the relationship between the doses of dietary AA and hepatic AA levels in mice treated as shown in Fig. 2, where URO accumulation was produced by treatment of Gulo(−/−) mice with 5-aminolevulinate and the CYP1A2 inducer, PCB126. Figure 3A shows that decreasing AA in the drinking water from the concentration used for maintaining and reproducing the Gulo(−/−) mouse colony (1000 ppm) to 75 ppm, resulted in a decrease of 70% in hepatic AA. Decreasing the AA intake to 75 ppm was associated with a 4-fold increase in hepatic URO accumulation (Fig. 3B). At the 75 ppm AA intake level, there were no overt signs of scurvy or toxicity, as indicated by loss of body weight or hepatic histopathology (data not shown). These results indicate that PCB126-mediated uroporphyria in Gulo(−/−) mice can be modulated by hepatic AA levels.

Figure 3.

Dose response of dietary AA on hepatic AA (A) and URO (B). Gulo(−/−) mice were treated with PCB126 (75 μg/kg body wt) and 5-aminolevulinate (2 mg/ml drinking water). The experimental design is shown in Fig. 2. Hepatic AA and URO were assayed as described in Materials and Methods. The values are means and SDs. The number of animals per group was 5–20. All groups were significantly different from each other (P < 0.05).

Effect of Fe Dextran Administration on Hepatic URO and AA in Gulo(−/−) Mice Maintained on Low (75 ppm) and High (1000 ppm) Dietary AA.

In experimental animals, administration of iron greatly enhances hepatic URO accumulation in mice in several protocols, including treatment with hexachlorobenzene, PCB126 and 2,3,7,8-tetrachlorodibenzo-p-dioxin.6 We examined the effects of iron dextran administration on the effectiveness of a high intake of AA to suppress URO accumulation caused by the combination of PCB126 and 5-aminolevulinate (Fig. 4). In the absence of iron dextran administration, much more URO accumulated at low AA intake than at high AA intake (Fig. 4A). The administration of iron dextran had no additional effect on URO accumulation in mice maintained on the low AA intake (Fig. 4A), which suggests that the amount of URO accumulation is independent of hepatic iron levels when mice are ascorbate-deficient. However, when iron dextran was administered to mice on the high AA intake, there was a 4-fold increase in URO. The amount of hepatic URO in mice given iron dextran and maintained on the high AA diet was similar to that in mice maintained on the 75 ppm AA intake, regardless of iron dextran administration (Fig 4A).

Figure 4.

Effect of iron dextran administration on hepatic URO (A), non-heme iron (B) and AA (C). Gulo(−/−) mice were treated with iron dextran (500 mg Fe/kg), PCB126 and 5-aminolevulinate as in Fig. 2. Low AA was 75 ppm and high AA was 1000 ppm in the drinking water. Non-heme iron was assessed as described in Materials and Methods. The values are means and SDs. The number of animals per group was 5–20. a indicates groups with significantly different values from each other (P < 0.05).

As expected, iron dextran administration produced large increases in hepatic non-heme iron and this was not affected by AA intake (Fig. 4B). Since iron can oxidize AA in vitro, we expected that the high Fe levels would decrease hepatic AA. However, there was no detectible effect of iron dextran administration on hepatic AA at either AA intake (Fig. 4C).

Dose Response Effects of Iron Dextran on Hepatic Non-Heme Iron Levels and URO Accumulation.

Figure 5 shows the results of an experiment designed to determine the minimal dose of iron dextran that prevented the suppression of URO accumulation in Gulo(−/−) mice maintained on the high AA intake. With doses of iron dextran up to 225 mg Fe/kg, there was no increase in URO accumulation, despite an 18-fold increase in hepatic non-heme iron (Fig. 5A). However, at 300 mg Fe/kg, URO accumulation was increased by 3.5-fold. Increasing the iron dextran dose from 300 to 500 mg Fe/kg did not further increase URO accumulation (Fig. 5B), even though hepatic non-heme iron levels continued to rise (Fig. 5A). Altogether, these results indicate that AA at the lower hepatic iron dextran range suppressed URO accumulation, whereas at high iron levels, this effect of AA was nullified.

Figure 5.

Dose response for iron dextran on hepatic non-heme iron (A) and hepatic URO (B). The experimental design is shown in Fig. 2. Mice on the high AA intake were treated with PCB126 and 5-aminolevulinate, and with increasing doses of iron dextran. The values are means and SDs. The number of animals per group was 4–20. All non-heme iron values (A) are significantly different from each other (P < 0.05). a indicates groups with significantly different values from each other (P < 0.05).

The hepatic localization of iron was also examined. Iron dextran is initially taken up by phagocytosis into Kupffer cells of the liver.20 Figure 6 shows Prussian Blue staining of iron in liver sections from some of the mice used in the dose response study shown in Fig. 5. At a dose of 50 mg Fe iron/kg, Kupffer cells were stained strongly and hepatocytes were stained weakly (Fig. 6B). At a dose of 225 mg Fe iron/kg, there was increased Kupffer cell staining with considerable staining in hepatocytes (Fig. 6C). At 300 mg Fe/kg, there was further increase in Kupffer cell staining and a dramatic increase in hepatocyte staining which was observed in all 5 livers examined (see a representative liver section shown in Fig. 6D). These results suggest that iron stimulation of URO accumulation requires doses of iron dextran that produce major accumulations of iron in the hepatocytes. Sections stained with H and E were also evaluated for histopathology. Although there was some inflammation around the Kupffer cells with increasing doses of iron dextran, there was no marked increase in hepatocyte toxicity as observed histologically when the dose was increased from 225 to 300 mg Fe/kg (data not shown).

Figure 6.

Localization of hepatic iron in mice treated with increasing doses of iron dextran. Mice were treated as described in the legend to Fig. 5. Following Perls staining, sections were counter-stained for only 15–20 sec to maximize detection of blue-stained iron.17 A typical liver section from one mouse is shown for each group. A, no iron dextran; B, Iron dextran: 50 mg Fe/kg; C, 225 mg Fe/kg; D, 300 mg Fe/kg. Original magnification, ×400.

Effect of AA Intake on Oxidative Stress Produced by Iron.

We examined the relationship between iron-induced oxidative stress and the effects of iron and AA on URO accumulation. MDA, a lipid peroxidation product, was selected as a measure of oxidative stress. In the absence of administered iron, hepatic MDA was the same at both AA intakes (Fig. 7A), whereas after treatment with a high dose of iron dextran, MDA was increased by about 3-fold at both AA intakes (Fig. 7A). This result indicates that AA did not modulate MDA formation. In the Fe dextran dose response experiment, hepatic MDA was increased by 1.5-fold at the lowest dose (50mg Fe/kg) and reached a plateau at 150 mg Fe/kg (Fig. 7B). Since the lower doses of iron dextran (up to 225 mg/kg) caused maximal increases in MDA, but did not increase URO accumulation in the same mice (Fig. 5), these results suggest that there is no direct relationship between oxidative stress and URO accumulation.

Figure 7.

Effect of iron dextran and AA on hepatic malondialdehyde (MDA). A. Mice were treated as described in the legend to Fig. 4. B. Mice were treated as described in the legend to Fig. 5. Liver MDA was assayed as thiobarbituric acid-reacting substances as described in Materials and Methods. The values are means and SDs. The number of animals per group was 3–7. Groups with the same letter were significantly different from each other (P < 0.05).

Effect of Dietary AA and Iron on Hepatic CYP1A2.

Figure 8 shows the results of an experiment designed to examine whether the effects of AA and iron on hepatic URO were due to differences in the levels of hepatic CYP1A2 following treatment with the inducer, PCB126. At the dose used here, this compound caused a 4–5-fold increase in CYP1A2, as observed previously.20 No differences in the levels of the induced CYP1A2 protein were detected immunochemically in animals maintained on either AA intake (Fig. 8A,B). Similarly, in other experiments where mice were treated with PCB126 and iron, but no 5-aminolevulinate, microsomal CYP1A2-catalyzed activities (methoxyresorufin demethylase and uroporphyrinogen oxidation) were not different in mice maintained on either AA intake (Fig. 8C,D). Thus, the differences in URO accumulation associated with changes in hepatic iron and AA were not due to changes in the level of induced CYP1A2.

Figure 8.

Effect of iron dextran and AA on hepatic CYP1A2. (A,B). Mice were treated with iron dextran (FeD), PCB126 and 5-aminolevulinate, as described in the legend to Fig. 4. CYP1A2 in liver homogenates (1.5 μg protein/lane) was measured by Western blotting as described in Methods. The ninth lane contained liver homogenate from a mouse treated with 3-methylcholanthrene to indicate location of CYP1A2. The upper band in this lane is CYP1A1. (C,D). Mice were treated as described in the legend to Fig. 4 except that no 5-aminolevulinate was administered. Hepatic microsomes were prepared as described in Materials and Methods and activities of methoxyresorufin demethylase (MROD, C) and uroporphyrinogen oxidation (UROX, D) were assayed as described in Materials and Methods. Uroporphyrinogen oxidation is an activity shown previously to be primarily catalyzed by CYP1A2.5 Even though methoxyresorufin demethylase activity can be catalyzed by CYP1A1, in addition to CYP1A2, at the PCB126 dose used here there was very little CYP1A1 induction as shown in the immunoblot (B). The values are means and SDs. The number of animals per group was 3–4. MROD and UROX values for untreated mice at the higher AA intake were 65–70 and 9–10 pmol/min/mg protein, respectively.


Previous studies have shown that dietary AA suppresses development of experimental uroporphyria in animal model systems.7, 8 Since increased hepatic iron is a well established enhancer of hepatic URO accumulation, both in experimental animals6 and in human PCT,1 the present study was designed to determine whether this effect of AA was modulated by hepatic iron levels. Surprisingly, when AA-requiring mice were treated with high doses of iron dextran, AA was no longer protective against hepatic URO accumulation caused by PCB126 and 5-aminolevulinate. The results of this study excluded the possibility that the effect of hepatic iron loading on URO accumulation was due to depletion of hepatic AA (Fig. 4C). An indicator of lipid peroxidation and oxidative stress, MDA, was increased by low doses of iron dextran whereas the iron-mediated stimulation of URO accumulation required much higher iron doses. This result suggests that iron-stimulated oxidative stress, as indicated by MDA accumulation, does not mediate URO accumulation in this model. Altogether, the results here demonstrate that AA can suppress development of hepatic uroporphyria in mice that have either normal levels of hepatic iron or even some elevation in hepatic iron stores. However, when the liver is highly iron loaded, AA cannot suppress URO accumulation.

Despite the large increases in hepatic non-heme iron in animals treated with high doses of iron dextran (where URO accumulation was increased), there were no detectable decreases in hepatic AA. This is a surprising result in the light of the well-known oxidation of AA by inorganic Fe in vitro. A previous study detected a decrease in hepatic AA in guinea pigs (which also require dietary AA) given very large doses of iron dextran, but this depletion could be prevented by administration of a large dose of AA.21 More recently, in another guinea pig study, that used a protocol similar to that used here with mice, no decreases were detected in hepatic AA caused by iron loading with iron dextran.22 There are several possibilities for explaining why the iron overload did not deplete hepatic AA in our study. The hepatocytes may have sequestered most of the iron derived from iron dextran as inert iron in ferritin. Alternatively, the rate of re-reduction of the oxidized products of AA, dehydroascorbate or ascorbate radical may have exceeded the rate of the iron-catalyzed oxidation. In addition, since the animals were maintained on dietary AA, there may have been rapid replenishment of the AA that was being oxidized by the excess iron.

The results of the present study may also contribute to our understanding of the interaction of iron and AA in the clinical disease, PCT. As noted earlier, in studies with a small number of patients, it has been reported that plasma AA is very low in PCT patients.9, 10 If these findings are substantiated in larger studies, they would be consistent with the effects of low AA intake to facilitate hepatic URO accumulation, as observed in this study with the AA-requiring mice (Fig. 3), and in a previous study with AA-requiring rats.8 Although Percy et al.10 reported that AA administration to PCT patients for one week had no effect on urinary URO excretion, these patients would most likely have had elevated hepatic iron stores since they had not been previously iron depleted by phlebotomy. The lack of response to AA of these patients10 is consistent with our findings that AA intake had no effect on hepatic URO accumulation in mice treated with high doses of iron dextran. Our mouse studies also suggest that the diet of AA-deficient PCT patients could be supplemented with small doses of AA to restore AA sufficiency and possibly prevent further accumulation of URO. However, our mouse studies suggest that AA suppression of URO accumulation would only be effective in patients with normal or mildly elevated hepatic iron. Although some authors have expressed concern that AA may act as a pro-oxidant in the presence of high levels of iron (see Gerster23 for review), others have concluded that vitamin C in vivo acts more as an anti-oxidant than as a pro-oxidant even at high hepatic iron levels23, 24 and is safe.25 There was no indication that AA increased uroporphyria at high hepatic iron levels (Fig. 4A).

In summary, uroporphyria induced by PCB126 and 5-aminolevulinic acid in AA-requiring mice was found to be increased by AA deficiency. However, after major iron loading, even an AA intake sufficient to restore hepatic AA levels to normal was ineffective in suppressing URO accumulation. No depletion of hepatic AA by the increased hepatic iron stores was detected even with major iron loading. Although oxidative stress, as indicated by MDA accumulation, was increased by iron-loading, this increase was not suppressed by AA. These results indicate that AA suppresses URO accumulation in experimental uroporphyria, but only when hepatic iron stores are normal or mildly elevated. The results suggest that increased AA consumption may suppress further URO accumulation in PCT patients following depletion of some hepatic iron by phlebotomy.


The authors thank Terry Mattoon and Linda Wilmot for histological preparation and staining of samples, and William J. Bement and Dawn Carbonneau for contributions to the animal work. We also thank Drs. George H. Elder and Robert A. Jacob for helpful discussions.