JBC Avanti Polar Lipids

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M212685200 on August 20, 2003

J. Biol. Chem., Vol. 278, Issue 46, 45555-45562, November 14, 2003
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
278/46/45555    most recent
M212685200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Akhtar, M. K.
Right arrow Articles by Kaderbhai, M. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Akhtar, M. K.
Right arrow Articles by Kaderbhai, M. A.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Export of a Heterologous Cytochrome P450 (CYP105D1) in Escherichia coli Is Associated with Periplasmic Accumulation of Uroporphyrin*

M. Kalim Akhtar, Naheed N. Kaderbhai, David J. Hopper, Steven L. Kelly, and Mustak A. Kaderbhai{ddagger}

From the Institute of Biological Sciences, Cledwyn Building, University of Wales, Aberystwyth, Ceredigion, Wales SY23 3DD, United Kingdom

Received for publication, December 12, 2002 , and in revised form, August 15, 2003.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
This report suggests an important physiological role of a CYP in the accumulation of uroporphyrin I arising from catalytic oxidative conversion of uroporphyrinogen I to uroporphyrin I in the periplasm of Escherichia coli cultured in the presence of 5-aminolevulinic acid. A structurally competent Streptomyces griseus CYP105D1 was expressed as an engineered, exportable form in aerobically grown E. coli. Its progressive induction in the presence of 5-aminolevulinic acid-supplemented medium was accompanied by an accumulation of a greater than 100-fold higher amount of uroporphyrin I in the periplasm relative to cells lacking CYP105D1. Expression of a cytoplasm-resident engineered CYP105D1 at a comparative level to the secreted form was far less effective in promoting porphyrin accumulation in the periplasm. Expression at a 10-fold molar excess over the exported CYP105D1 of another periplasmically exported hemoprotein, the globular core of cytochrome b5, did not substitute the role of the periplasmically localized CYP105D1 in promoting porphyrin production. This, therefore, eliminated the possibility that uroporphyrin accumulation is merely a result of increased hemoprotein synthesis. Moreover, in the strain that secreted CYP105D1, uroporphyrin production was considerably reduced by azole-based P450 inhibitors. Production of both holo-CYP105D1 and uroporphyrin was dependent upon 5-aminolevulinic acid, except that at higher concentrations this resulted in a decrease in uroporphyrin. This study suggests that the exported CYP105D1 oxidatively catalyzes periplasmic conversion of uroporphyrinogen I to uroporphyrin I in E. coli. The findings have significant implications in the ontogenesis of human uroporphyria-related diseases.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Porphyrin metabolism in humans has been extensively investigated, particularly in the occurrence of a group of clinically related diseases known as porphyrias, arising from genetic defects (1) of the enzymes in heme biosynthesis causing accumulation of various heme precursors (2, 3). The commonest porphyric disorder, porphyria cutanea tarda, and its related forms are characterized by massive hepatic accumulation and increased excretion of uroporphyrin (URO)1 and to a lesser extent other types of porphyrins such as coproporphyrins and isocoproporphyrins (4). A number of etiological factors such as iron overload (5), ascorbic acid deficiency (6), hepatitis c infection (7), alcohol abuse (8), and induction of CYP isoenzymes (9) have been suggested to be the cause of the sporadic form of the diseases. Of the many possible URO isomers, only types I and III occur in nature. Both isomers arise from spontaneous oxidation of the heme metabolic intermediates, uroporphyrinogens (UROgens) I and III. UROgen III synthase catalyzes the conversion of hydroxymethylbilane to UROgen III, which involves rearrangement of the propionate and acetate groups of ring IV of the tetrapyrrole, whereas UROgen I is formed nonenzymatically. UROgens I and III can be further metabolized by UROgen decarboxylase to yield coproporphyrinogens I and III, respectively, which can also undergo spontaneous oxidation to coproporphyrins I and III.

Uroporphyric-like states that lead to a significant build-up of URO have been experimentally induced in eukaryotic cells exposed to CYP inducers such as chlorinated biphenyls (10) and methylcholanthrene (11). Since such xenobiotics are classic inducers of CYP (12), a link has been suggested between susceptibility to uroporphyria (porphyrias characterized by URO and UROgen overload) and CYP-mediated oxidization of heme intermediates. The in vitro observation of CYP-catalyzed conversion of UROgen to URO (13) has been suggested as a major potential factor in the initiation of uroporphyria. Although there is strong, but indirect, evidence for the participation of a variety of CYPs in the etiology of uroporphyrias including the CYP1A subfamily (14), CYP1A2, and particularly CYP3A (10), the metabolic contribution of the hemoprotein in vivo has not been firmly established.

Most recently, the soluble CYP105D1 of Streptomyces griseus has been successfully expressed in Escherichia coli as an exportable form that was efficiently localized to the periplasm (15). Expression of the periplasmic CYP105D1 was coupled with generation of a highly fluorescent compound identified as URO I. We now show that the uroporphyric state induced in the engineered E. coli, which has a requirement for ALA to be present, is associated in vivo with the exported, catalytically active CYP105D1 and arises from URO I accumulation. The formation of such products has not been observed before in prokaryotic cells heterologously expressing CYP and has in turn important implications in the ontogenesis of the related diseases in human disorders.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Unless stated otherwise, all chemicals were purchased from Sigma. Clotrimazole, triadimefon, and miconazole were kindly donated by Dr. E. I. Mercer (University of Wales, Aberystwyth). UROgen III was obtained from Frontier Scientific. Electrophoresis reagents were purchased from Bio-Rad and BDH (Poole, UK).

Heterologous Expression, Quantification, and Isolation of CYPs—In the present study, we employed E. coli TB1 (F, ara{Delta} (lac-proAB)rps {phi} 80d lacZ{Delta} M15hsdR17 ()) strain harboring pLi-CYP105D1 encoding the secretory form of CYP105D1, pAA-CYT coding for the secretory form of mammalian cytochrome b5 (16, 17), pIN-CYP105 expressing the cytosolic CYP105D1, and the control progenitor plasmid pLi-Q (15). Plasmid pIN-CYP105D1 is a derivative of pLi-CYP105D1 and lacks the signal sequence. The vector was constructed by an inverse PCR strategy using pLi-CYP105D1 as the template and the following primers: 5'-ATGACGGAATCCACGACGG and 5'-TTTATTTTCTCCATGTACAAATAC.

The amplified DNA duplex was circularized by T4 DNA ligase and introduced into E. coli TB1. For induction of recombinant proteins, a 2% (v/v) inoculum of Luria-Bertani broth-grown, saturated culture was added to a phosphate-limited (0.1 mM) MOPS medium (16) containing 75 µg/ml ampicillin. The culture was batch-cultivated at 35 °C with orbital agitation at 125 rpm for periods specified elsewhere. Unless stated otherwise, 5-aminolevulinic acid (ALA) at 1 mM was routinely included in the MOPS growth medium.

All media used for subcellular fractionation or isolation of URO contained 5 mM ascorbic acid in order to reduce adventitious auto-oxidative generation of porphyrin. Periplasm extracts were isolated as described previously (15). To verify that authentic isolation of the periplasm had occurred, marker enzymes known to be specific for each of the subcellular fraction were routinely assayed as described (16). CYP content in biological samples was estimated according to the procedure described by Omura and Sato (18) using an absorption coefficient of 91 mM-1 cm-1. The oxidized cytochrome b5 was quantified from the Soret absorption peak at 413 nm using the absorption coefficient of 115 mM-1 cm-1 (19). Protein content was estimated employing the Bio-Rad protein assay (Bio-Rad) using bovine serum albumin as the standard (20).

The CYP105D1 enzyme was purified from the periplasmic extract of E. coli CYP105D1 (1 liter of culture) cultivated for 24 h at 28 °C with 1 mM ALA supplementation. Following 2-fold dilution of the periplasmic extract with 10 mM Tris acetate (pH 8) buffer (TA) containing 1 mM EDTA, the sample was applied twice through a TA-pre-equilibrated column of DEAE CL-6B-Sepharose (5-ml bed volume). The column was successively washed with 5 ml of (i) TA, (ii) 0.1 M NaCl in TA, and (iii) 0.3 M NaCl in TA. The enzyme was eluted with 25 ml of 0.5 M NaCl in TA. The solution was concentrated by pressure filtration to 2.5 ml, and the CYP105D1 was further purified by fast protein liquid chromatography by passage through a Superose column (1.5 x 65 cm) with 50 mM Tris-HCl (pH 7.5) buffer. The flow rate was 1 ml/min, and fractions of 2 ml were collected. CYP105D1-enriched fractions were pooled and concentrated using an Amicon Centricon 10 microconcentrator (Amersham Biosciences).

Isolation and Estimation of URO—Periplasmic and cytoplasmic extracts were derived from E. coli cultured in 100 ml of MOPS medium supplemented with ALA and 75 µg/ml ampicillin. Each extract, finally recovered in 5 ml of TA, was acidified with HCl to 0.2 M final concentration (pH <2) and left on ice for 5 min. The deproteinated supernatant was recovered by centrifugation at 13,000 x g for 2 min. URO amounts in the supernatant fractions after appropriate dilution were estimated spectroscopically from absorbance at 406 nm using the absorption coefficient 541 mM-1 cm-1 (21). The values were standardized to nmol/liter of culture.

For estimation of URO in the "spent" growth medium, 20 ml of the growth medium was applied onto a DEAE CL-6B-Sepharose column (1-ml bed volume). After washing the column with 5 ml of 0.3 M NaCl in TA, the pigments were eluted with 2 ml of 1 M NaCl in TA, acidified, and spectroscopically quantified as described above.

For the analysis and estimation of URO by HPLC (22), the periplasmic fraction (5-ml volume recovered from 100-ml culture) was first deproteinated with 0.2 M HCl as described above. The pH of the supernatant was adjusted to 7.5 by titration with Trizma base. Following 5-fold dilution with distilled water, the solution was applied onto a DEAE-Sepharose CL-6B column (1-ml bed volume). After washing the column with 5 ml of TA, the pigments were eluted with 1 ml of 1 M NaCl in TA. HPLC was performed on an ODS2 column (250 x 4.6 mm; 5-µm article size) eluted at a flow rate of 1 ml/min with an acetonitrile-1 M ammonium acetate (pH 5.16) (14.3:85.7 (v/v)) as the mobile phase. The amounts of porphyrins in each peak were calculated by comparison with the standard curve derived from known amounts of URO I and III. Compounds were detected by measuring the absorbance at 406 nm in a continuous flow cell, and the data were processed by Thermochrom II software (LDC Analytical).

Heme Solution—A 10 mM heme solution was prepared as follows. Bovine hemin, after dissolving in one-twentieth of the final volume of freshly prepared 1 M NaOH, was sequentially mixed with an equal volume of 1 M Tris-HCl (pH 7) and an eight-tenths final volume of ethylene glycol. The pH was adjusted to 8.2 with 1 M HCl. The solution was made up to the final volume with distilled water, filter-sterilized by passage through a 0.2-µm filter, and stored at 4 °C. To test the ability of the E. coli TB1 strain to utilize exogenous heme, the stock heme solution was applied in the MOPS medium at 5-fold increasing concentrations ranging from 1 to 50 µM.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
A Fluorescent Compound Generated during Periplasmic Expression of CYP105D1 Is a URO—We recently reported that an E. coli strain harboring an engineered pLi-CYP105D1 plasmid expressed an exportable form of the Streptomyces griseus CYP105D1 under the transcriptional control of the tightly regulated phoA promoter (23). The periplasmic localization of CYP105D1 was facilitated by N-terminal appendage of a bacterial secretory signal that utilizes the sec-dependent translocon export pathway (24). The pre-CYP105D1 was correctly processed, folded, and heme-assembled to yield substantial quantities of the functional recombinant hemoprotein in the periplasm.

E. coli pLi-CYP105D1 cultured in the MOPS medium in the presence of 1 mM ALA for periods ranging from 12 to 36 h displayed a strong brown hue in contrast to comparably cultured control cells containing the progenitor plasmid pLi-Q lacking cyp105D1. The color was partly attributed to the presence of significant quantities of the holo-CYP105D1 that had accrued in the periplasm. Interestingly, UV illumination of E. coli pLi-CYP105D1 but not E. coli pLi-Q cells exhibited an intense red fluorescence emission. Preliminary subfractionation studies indicated that the fluorescent agent(s) was particularly concentrated in the periplasmic extract.

In order to isolate and characterize the fluorescent compound(s), the periplasmic extract from E. coli pLi-CYP105D1 was subjected to anion exchange separation by DEAE-Sepharose CL-6B chromatography. Following sequential washes with 0.1, 0.2, and 0.3 M NaCl in TA, an intense dark brown layer remained firmly bound to the column. A further wash with 0.5 M NaCl in TA eluted a light brown-colored fraction containing the CYP105D1 protein together with a pigment that showed predominant absorption peaks between 486 and 492 nm. A final wash with 1 M NaCl in TA displaced a rose-colored fraction that gave an intense red fluorescence upon UV light illumination. Spectrofluorometric analysis of the isolated compound(s) upon excitation at 398 nm displayed a major emission peak at 615 nm together with smaller peaks 642 and 678 nm (Fig. 1). Absorption spectroscopy in the visible region gave a major peak at 398 nm (Fig. 2A) with four smaller peaks at 502, 538, 560, and 612 nm (Fig. 2B); the spectrum matched well with that of authentic URO I (25). The etio-nature of the UV-visible spectrum suggested that the potential URO was in a metal-free state (26). Acidification of the solution with HCl (0.2 M final concentration) shifted the Soret absorption peak to 406 nm (Fig. 3A) and yielded two additional minor peaks at 550 and 592 nm (Fig. 3B). Further analysis of the pigment(s) by electrospray ionization spectrometry revealed a major, stable predominant species with a relative molecular mass of 831 (Fig. 4) that was not detectable in the E. coli pLi-Q extract. The spectral characteristics and the mass of the bacterially produced fluorescent compound were identical to a commercially obtained authentic URO I (Figs. 2, 3, 4). This could be indicative of URO I, but notably URO III, the counterpart physiological isomer, also has identical spectral characteristics. However, further comparison of the unknown pigment with the two authentic pigments by HPLC analysis (22) identified type I isomer as the major species, constituting about 85%. Identification was further confirmed by thin layer chromatography (27), whereby the relative mobility of the bacterial pigment matched with the authentic URO I. Under the present conditions of analysis, other intermediates of the pathway such as hepta-, hexa-, penta-, and tetracarboxylporphyrins were not detected.



View larger version (22K):
[in this window]
[in a new window]
  		FIG. 1.
The fluorescence scans of the pigment eluted from a DEAE CL-6B-Sepharose column following washing with 1 M NaCl in TA buffer (pH 8). The spectra were recorded across a 1-cm light path and at 1-nm wavelength intervals following excitation at 398 nm. The emission spectra were performed using a Shimadzu RF-5301PC spectrofluorophotometer.

 



View larger version (25K):
[in this window]
[in a new window]
  		FIG. 2.
Absorption spectra of the putative porphyric pigment isolated from E. coli periplasm compared with authentic URO I. The spectra were recorded in TA buffer (pH 8). The rose-colored pigment displayed a major Soret absorption peak at 398 nm (A) together with four minor peaks at 502, 538, 560, and 612 nm, shown more clearly in the inset (B).

 



View larger version (22K):
[in this window]
[in a new window]
  		FIG. 3.
Absorption spectra of the HCl-acidified putative porphyric pigment (pH <2) isolated from E. coli periplasm compared with authentic URO I. Spectra range in wavelength from 300-600 nm (A), and from 450-650 nm (B).

 



View larger version (12K):
[in this window]
[in a new window]
  		FIG. 4.
Electrospray ionization spectroscopy of the isolated pigment. A 70-µl portion of the 1 M NaCl TA buffer (pH 8)-eluted pigment from DEAE-Sepharose CL-6B was 100-fold prediluted with acetonitrile/water (50:50 (v/v)) containing 0.1% (w/v) formic acid and applied continuously through a Harvard syringe pump at a flow rate of 5 µl/min through a silicon tubing (0.7-µm inner diameter x 1 m) into an LCT-ES spectroscope (Micromass). Each scan was integrated over a 30-s period, and the data were deconvoluted using Mass Lynx software (version 3.2).

 
URO Accumulation Is Confined to the Periplasm and Concomitantly Coupled with the Exported Holo-CYP105D1—The quantities of URO in the periplasm and cytoplasm compartments and in the growth medium of E. coli strains harboring pLi-Q or pLi-CYP105D1 cultured in the presence of 1 mM ALA were monitored as a function of time (Table I). In both strains, the periplasmic URO accrued with the duration of culturing. However, the control pLi-Q strain showed only a 5-fold increase, ranging from 2 nmol/liter of culture at 12 h to about 10 nmol of URO/liter of culture at 36 h compared with an ~100-fold increase in URO in the recombinant strain. The latter produced 800 nmol of URO/liter by 36 h. In the pLi-Q strain, the URO profile followed a similar pattern to that seen in its periplasm. In contrast, the cytoplasmic level of URO in pLi-CYP105D1 increased substantially by nearly 10-fold, reaching 25 nmol of URO/liter of culture following 36 h of culturing. Given the high accumulation of URO in the periplasm, we cannot discount the likelihood that URO in the cytoplasmic fraction could be, at least partly, derived from contamination during cellular fractionation or in vivo leakage from the periplasm. In contrast, the URO found in the growth medium in the control and pLi-CYP105D1-expressing strains was negligible (Table I).


View this table:
[in this window]
[in a new window]
  		TABLE I
URO amounts in isolated periplasmic and cytoplasmic fractions and in the growth medium of E. coli strains containing derivatives of pLiQ expressing secreted CYP105D1 (pLi-CYP105D1), secreted cytochrome b5 (pAA-CYT), and cytoplasmic-resident CYP105D1 (pIN-CYP105D1) The values as nmol of URO/liter of culture or pmol of URO /mg of protein (in parentheses) are presented as means ± S.D. of triplicate measurements. UROs were monitored in deproteinatad. HCl-acidified (pH < 2) samples as described under "Experimental Procedures." Protein was not detectable in the growth medium. P, phoA promoter, SD, Shine-Dalgarno sequence; SS, signal sequence of alkaline phosphatase; b5, globular core of cytochrome b5.

 
We considered the possibility that overexpression and export of the apo-CYP105D1 may have induced synthesis and periplasmic export of heme. This possibility was explored by measuring the periplasmic and cytoplasmic pools of URO in an E. coli TB1 strain pAA-CYT that efficiently exports a mammalian recombinant cytochrome b5 under expression conditions comparable with CYP105DI (16). Interestingly, although cytochrome b5 is produced at about a 10-fold molar (24 h) excess over CYP105 (Table II), the periplasmic, cytoplasmic, and medium URO levels in this strain are similar to those in the control pLi-Q strain throughout the duration of culturing (Table I).


View this table:
[in this window]
[in a new window]
  		TABLE II
Holo forms of CYP105D1 and cytochrome b5 in E. coli extracts as a function of culture duration The estimations of hemoproteins were performed in the isolated periplasmic extracts as described under "Experimental Procedures." Data are presented as means ± S.D. of triplicate measurements.

 
To investigate whether URO production was an event specifically coupled with expression of the periplasmic form of CYP105D1, we constructed a cell line that expressed a nonsecreted, cytoplasmic CYP105D1. The cytoplasmic CYP105D1 was expressed more efficiently than the secretory form (Table II) and supported URO formation, although to a significantly lesser extent than that catalyzed by the periplasmic catalyst (Table I). Identities of the URO accumulated in the periplasmic fractions of the control and recombinant strain was confirmed by subjecting deproteinated extracts to HPLC analysis. Using the standard conditions described under "Experimental Procedures," UROs (I and III) were clearly resolved, and URO I was the dominant species. The overall pattern and yields of URO detected by HPLC were similar to that monitored spectroscopically in the periplasmic extracts.

URO Production Requires Exogenous ALA and Co-expression of Holo-CYP105D1—In view of the long periods (>12 h) required to generate significant URO production, it was of interest to explore whether this phenomenon was related to synthesis and/or assembly of CYP105D1 in the periplasm. Hence, this necessitated an investigation of ALA dependence on production of holo- and apo-CYP105D1 (Fig. 5). The periplasmically recovered soluble CYP105D1 was separated by nondenaturing gel electrophoresis. The apo and holo forms were detected and distinguished by a combination of heme staining and Western blotting using anti-CYP105D1. Apo- but not holo-CYP105D1 was detectable in the E. coli cultured for 12 h without ALA. By 24 h of culturing in the absence of ALA, apo-CYP105D1 almost doubled, of which approximately 6% was converted to the holo form. With supplementation of increasing ALA concentrations in the medium, conversion of the apo form to holo-CYP105D1 accrued hyperbolically and was saturated at 2 mM ALA. Substitution of ALA with heme in the medium did not enhance the production of holo-CYP105D1. Moreover, significant synthesis of URO in the periplasm of E. coli pLi-CYP105D1 occurred only in the presence of exogenously supplemented ALA (Fig. 6). Peak production of URO was reached at 0.25 mM ALA, and at higher doses inhibition was imposed. Interestingly, supplementation of the cultures with additional ALA at 12 h slightly enhanced the productivity of URO, but beyond 0.2 mM ALA a more drastic inhibition of URO was imposed (Fig. 6). In contrast, the formation of holo-CYP105D1 was more directly proportional to ALA concentration up to 0.8 mM such that at the highest tested concentration of 2 mM ALA, a 20-fold increase in holo-hemoprotein was reached (Figs. 5 and 6). These findings suggest that formation of both holo-CYP105D1 and URO production were clearly dependent upon ALA supplemented in the medium, except that in the latter case a higher concentration of ALA caused a decrease in URO accumulation for reasons that are not clear.



View larger version (23K):
[in this window]
[in a new window]
  		FIG. 5.
Expression of CYP105D1 detected in the periplasm extracts of E. coli pLi-CYP105D1 cultured without or with varying doses of ALA supplemented in the growth media. CYP105D1 visualized either by heme activity staining (49) (A) or immunodetection (B). Band I, apo-CYP105D1; band II, holo-CYP105D1. Periplasmic proteins were separated in a nondenaturing polyacrylamide gel (6% (w/v)), pH 6.5 (50), containing glycine and 0.2 M urea. For Western blot analysis, the gel was first gently agitated with the denaturing buffer (0.2% (w/v) SDS, 50 mM Tris-HCl (pH 8.2) for 2 h, and the proteins were transferred onto a nitrocellulose transfer membrane essentially as described (47). The membrane was probed with goat anti-CYP105D1 antiserum (1:400) followed by affinity-purified guinea pig anti-goat IgG-coupled to alkaline phosphatase. The antigen-antibody interaction was detected by alkaline phosphatase activity after incubating the blot with 0.5 mg/ml {alpha}-naphthyl pyrophosphate and 0.5 mg/ml 4-chloro-o-toluidine diazonium in 30 mM Tris-HCl (pH 9). The densitometrically scanned (Epson GT-12000) profiles were quantified (C) using Phoretix 1D Advanced software (version 3.01) operating under Microsoft Windows XPTM.

 



View larger version (18K):
[in this window]
[in a new window]
  		FIG. 6.
Effects of supplementing ALA in the growth medium on the periplasmic formations of URO I (A) and expression of CYP105D1 (B). E. coli pLi-CYP105D1 were cultured for 24 h, and ALA was supplemented either at the start (0 h, pretreated) or both at the start and again at 12 h (pre- and post-treated) of cultivation.

 
Selective Fungicides Inhibit URO Production—We investigated the effects of the fungal P450 inhibitors, clotrimazole, miconazole, and triadmefon, which interfere with the sterol biosynthesis (28) on the productivity of URO. These compounds are well known to exert their inhibitory effects by interacting with the heme iron of CYPs (29). Standard batch cultures were incubated with the inhibitors at concentrations ranging from 0 to 2.5 mM, and the periplasmic URO was monitored after 24 h of culturing (Fig. 7). Clotrimazole at 0.5 mM produced the most pronounced decline (65%) in URO accumulation. However, all of the inhibitors suppressed URO production in a dose-dependent manner and yielded a similar inhibition of about 60% at 2.5 mM azole. These inhibitors did not affect the growth of E. coli. Moreover, clotrimazole, which most potently inhibited URO formation in the recombinant cell line, did not significantly affect the production of URO in the control cell line. {beta}-Diethylaminoethyl diphenylpropyl acetate (proadifen; SKF525A), a nonimidazole CYP inhibitor that is structurally and mechanistically unrelated to azoles (30), had no significant effect on URO production in E. coli pLi-CYP105D1.



View larger version (25K):
[in this window]
[in a new window]
  		FIG. 7.
In vivo inhibition of URO I production in E. coli pLiCYP105D1 and E. coli pLiQ by azoles. The MOPS media containing the dissolved azoles were filter-sterilized. The E. coli strains were cultured for 24 h as described under "Experimental Procedures," and the URO was estimated in the isolated periplasmic fractions. The values are average determinations of three independent culture measurements.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
In the present study, we have investigated the formation of a fluorescent pigment in E. coli TB1 strain in ALA-treated cells expressing a secreted or periplasm-localized CYP105D1. The pigment has been firmly identified by absorption and fluorescence spectroscopy, electron spray ionization, and HPLC. Several pieces of evidence strongly point to catalytic involvement of CYP105D1 in the transformation of UROgen I to URO I. An over 100-fold higher level of URO was found in E. coli pLi-CYP105D1 than in the comparable control harboring pLi-Q or pAA-CYT (Table I). The azole compounds, known to be potent inhibitors of CYPs, considerably reduced the in vivo productivity of URO in the recombinant but not in the nonrecombinant cell lines (Fig. 7). In contrast to the control strain pLi-Q and another overexpressing, secreted hemoprotein, cytochrome b5, production of URO I was most effective in E. coli harboring pLi-CYP105D1 expressing the periplasmically localized CYP105D1. Nevertheless, a small but significantly elevated production of URO was also observed in the cytosolic expressing CYP105D1 cells (Table I). Thus, we conclude that periplasmic targeting of the catalyst was a prerequisite for generating a uroporphyric-like state. This raises interesting issues regarding the temporal nature of periplasmic CYP-catalyzed conversion of UROgen I to URO I in E. coli. The important points to address are (i) the derivation of the substrate from the heme biosynthetic pathway in the cytosol, (ii) the preferential generation of URO I rather than URO III isomer, (iii) the likely mechanism(s) by which UROgen I reaches the periplasm for the CYP105D1-catalyzed conversion to URO I, and (iv) the nature of electrons source and the likely mechanism of the transformation by CYP105D1.

ALA is the first committed precursor of tetrapyrrole synthesis and its intracellular concentration limits the rate of porphyrin synthesis in E. coli (31). In prokaryotes, ALA synthase is a regulatory enzyme whose activity can be repressed by heme feedback inhibition. However, exogenously supplied ALA can cause an overproduction of porphyrins (32, 33) and heme as shown in this study. However, it seems unlikely that uroporphyrin accumulation merely resulted from increased hemoprotein synthesis, since overexpression at a 10-fold molar excess over the exported CYP105D1 of the periplasmically exported hemoprotein, the globular core of cytochrome b5, failed to promote porphyrin production. Expression of the substantial amounts of periplasmic apo-CYP105D1 and its conversion to the holo form demands a higher than normal cellular heme supply, which is derived from exogenously supplied ALA in the growth medium (Figs. 5 and 6). In the absence of ALA, only a small proportion (<10%) of the expressed CYP105D1 was converted to the holo form (Fig. 5). Exogenous heme supplementation in the culture medium did not improve the generation of holo CYP105D1, suggesting that the prosthetic group was not made available in the periplasm of E. coli TB1 cells as previously reported for the secreted form of cytochrome b5 (34) and that the outer membrane of E. coli is normally impermeable to heme (35). Nevertheless, the alleviation of this heme deficiency by ALA supplemented in the culture medium (Figs. 5 and 6) indicated that heme, during the assembly of CYP in E. coli, was derived metabolically via the de novo pathway (36). This suggests that the limiting step in heme synthesis during overproduction of CYP105D1 occurs at or prior to ALA synthesis (37). Thus, excess ALA is required to produce maximal holo-CYP, which, when expressed periplasmically, produces massive URO accumulation.

The second issue relates to the question of preferential periplasmic accumulation of URO I rather than the III isomer with co-expression of periplasmic CYP105D1. Indeed, both URO I and URO III can accumulate either in the presence or absence of ALA under some circumstances in a variety of cells. Moreover, the UROgens, which are the true intermediates, also need to be oxidized to generate the porphyrins. The production of all heme intermediates would be expected to occur exclusively in the cytosol, where the complete hem-encoded pathway is localized. Hydroxymethylbilane synthase (porphobilinogen deaminase) condenses four porphobilinogen molecules to form the first linear tetrapyrrole, hydroxymethylbilane (38). Hydroxymethylbilane, being unstable in solution, can nonenzymatically cyclize to form UROgen I, which is a physiologically redundant end product. However, hydroxymethylbilane is normally rapidly converted to UROgen III by the action of UROgen III synthase (39). Both I and III forms of UROgens have the potential to be oxidized to their respective porphyrin forms in the cytosol, although this apparently does not occur to a large extent under the physiological conditions studied here. UROgen III would be expected to be removed by the action of UROgen decarboxylase, and its auto-oxidative conversion to URO III may be partly suppressed due to a higher reducing condition within the cytoplasm. Whereas UROgen I may also accumulate in the cytosol and/or periplasm, UROgen III can be removed by the action of UROgen III decarboxylase, to be eventually converted to heme. Although the accumulated UROgen I can also be decarboxylated by UROgen decarboxylase to give coproporphyrinogen I, we are unable to explain why the oxidized intermediate, coproporphyrin I, was not observed here. Nevertheless, the findings imply that UROgen III decarboxylase is absent in the periplasm. A plausible model of the likely metabolic events is depicted in Fig. 8.



View larger version (19K):
[in this window]
[in a new window]
  		FIG. 8.
Scheme showing UROgen I production from heme biosynthetic pathway, translocation of UROgen I to E. coli periplasm, and CYP105D1-catalyzed conversion of UROgen I to URO I. HelABCD, heme translocator; PPT, preprotein translocase; COPROgen, coproporphyrinogen; pre CYP105D1, precursor of CYP105D1; PBG, porphobilinogen; HMB, hydroxymethylbilane; IM, inner membrane; OM, outer membrane. The thickness of the arrowed lines indicates the extent of metabolite flow.

 
UROgen I must accumulate in the cytoplasm, but then how does it resist auto-oxidation within the cytoplasm and become available in the periplasm? Perhaps under the more reducing conditions within the cytoplasm, endogenous auto-oxidation of UROgen may be partially suppressed. Another possibility is that the overproduced heme intermediate may be translocated into the periplasm (40), where its conversion to URO I can be accounted for by catalytically enhanced oxidation by CYP105D1 in the more amenable oxidizing environment of the periplasm (Fig. 8).

The periplasmic reaction involving production of URO I would be expected to require reducing equivalent(s) and molecular oxygen for the P450 catalysis and a yet unidentified reductase system to transfer the reducing equivalents to CYP105D1. Reducing equivalents such as NADH or NADPH are unlikely to be present in periplasmic space, but they could be scavenged from the electron transport chain or other periplasmic oxido-reductant reactions. However, electron transfer to the heterologous CYP105D1 must require a reductase, most likely a ferredoxin reduction type system. Genomic analysis of E. coli suggests the presence of a periplasmic ferrodoxin-type reductase that uses a TAT-dependent export pathway (41). The removal of six hydrogen atoms from UROgen I, two from the pyrrolic nitrogens and four from the methene carbons, would result in the formation of URO I. As yet, the mechanism for the mediation of UROgen oxidation by CYP remains unknown.

Clearly, ALA-dependent URO I production can be induced in a recombinant E. coli expressing periplasmic CYP105D1, and we have made similar observations for a secreted form of human CYP17. Further examples are likely and will extend our understanding of such metabolic effects on the ontogenesis of porphyria. Exposure of eukaryotic cells with exogenous ALA also increases URO due to insufficiency of decarboxylase (42) which can be potentiated by pre-exposure to xenobiotics such as chlorinated biphenyls (43). The essential role of a methycholantherene-inducible (44) and other CYPs (13, 45, 46) in the oxidation of UROgen suggests that the CYP enzyme(s) catalytically participate in the conversion of the heme intermediate UROgen to its oxidized form, URO, thus implicating it as a major potential factor in the initiation of URO overload. The present findings are consistent with the more recent observations of CYP knock-out mice, which fail to accumulate URO upon exposure to the classic inducers of CYP1A2 and ALA (48). Nevertheless, this is the first study providing clear evidence that the highest levels of uroporphyrin accumulation are remarkably correlated with expression of a broad substrate-range CYP105D1 in E. coli.


   FOOTNOTES
 
* This work was supported in part by the Biotechnology and Biological Science Research Council. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

{ddagger} To whom correspondence should be addressed. Tel.: 44-1970-622294; Fax: 44-1970-622294; E-mail: mak{at}aber.ac.uk.

1 The abbreviations used are: URO, uroporphyrin; ALA, 5-aminolevulinic acid; CYP, cytochrome P450; TA, Tris acetate (10 mM) buffer (pH 8); UROgen, uroporphyrinogen; MOPS, 4-morpholinepropanesulfonic acid; HPLC, high pressure liquid chromatography. Back


   ACKNOWLEDGMENTS
 
We are most grateful to Dr. K. Abbas for kindly supplying anti-CYP105D1 serum and to Jim Heald for mass spectral analysis.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Sassa, S., and Kappas, A. (2000) J. Intern. Med. 247, 169-178[CrossRef][Medline] [Order article via Infotrieve]
  2. Moore, M. R. (1998) Clin. Dermatol. 16, 203-223[CrossRef][Medline] [Order article via Infotrieve]
  3. Sarkany, R. P. (1999) Adv. Exp. Med. Biol. 455, 235-241[Medline] [Order article via Infotrieve]
  4. Elder, G. H. (1998) Semin. Liver Dis. 18, 67-75[Medline] [Order article via Infotrieve]
  5. Sampietro, M., Fiorelli, G., and Fargion, S. (1999) Haematologica 84, 248-253[Abstract/Free Full Text]
  6. Sinclair, P. R., Gorman, N., Shedlofsky, S. I., Honsinger, C. P., Sinclair, J. F., Karagas, M. R., and Anderson, K. E. (1997) J. Lab. Clin. Med. 130, 197-201[CrossRef][Medline] [Order article via Infotrieve]
  7. Tsukazaki, N., Watanabe, M., and Irifune, H. (1998) Br. J. Dermatol. 138, 1015-1017[Medline] [Order article via Infotrieve]
  8. Doss, M. O., Kuhnel, A., and Gross, U. (2000) Alcohol Alcohol. 35, 109-125[Abstract/Free Full Text]
  9. Sinclair, P. R., Gorman, N., Dalton, T., Walton, H. S., Bement, W. J., Sinclair, J. F., Smith, A. G., and Nebert, D. W. (1998) Biochem. J. 330, 149-153[Medline] [Order article via Infotrieve]
  10. Franklin, M. R., Phillips, J. D., and Kushner, J. P. (2000) Biochem. Pharmacol. 60, 1325-1331[CrossRef][Medline] [Order article via Infotrieve]
  11. Sinclair, P., Lambrecht, R., and Sinclair, J. (1987) Biochem. Biophys. Res. Commun. 146, 1324-1329[CrossRef][Medline] [Order article via Infotrieve]
  12. Rendic, S. (2002) Drug Metab. Rev. 34, 83-448[CrossRef][Medline] [Order article via Infotrieve]
  13. Sinclair, P. R., Gorman, N., Tsyrlov, I. B., Fuhr, U., Walton, H. S., and Sinclair, J. F. (1998) Drug Metab. Dispos. 26, 1019-1025[Abstract/Free Full Text]
  14. Gorman, N., Walton, H. S., Sinclair, J. F., and Sinclair, P. R. (1998) Comp. Biochem. Physiol. C Pharmacol. Toxicol. Endocrinol. 121, 405-412[Medline] [Order article via Infotrieve]
  15. Kaderbhai, M. A., Ugochukwu, C. C., Kelly, S. L., and Lamb, D. C. (2001) Appl. Environ. Microbiol. 67, 2136-2138[Abstract/Free Full Text]
  16. Karim, A., Kaderbhai, N., Evans, A., Harding, V., and Kaderbhai, M. A. (1993) Bio/Technology 11, 612-617[Medline] [Order article via Infotrieve]
  17. Harding, V., Karim, A., Kaderbhai, N., Jones, A., Evans, A., and Kaderbhai, M. A. (1993) Biochem. J. 293, 751-756[Medline] [Order article via Infotrieve]
  18. Omura, T., and Sato, R. (1964) J. Biol. Chem. 239, 2370-2378[Free Full Text]
  19. Estabrook, R. W., and Werringloer, J. (1978) Methods Enzymol. 52, 212-220[Medline] [Order article via Infotrieve]
  20. Bradford, M. (1976) Anal. Biochem. 72, 248-254[CrossRef][Medline] [Order article via Infotrieve]
  21. Rimington, C. (1960) Biochem. J. 75, 620-623[Medline] [Order article via Infotrieve]
  22. Lim, C. K., and Peters, T. J. (1986) Methods Enzymol. 123, 383-389[Medline] [Order article via Infotrieve]
  23. Kikuchi, Y., Yoda, K., Yamasaki, M., and Tamura, G. (1981) Nucleic Acids Res. 9, 5671-5678[Abstract/Free Full Text]
  24. Kim, J., and Kendall, D. A. (2000) Cell Stress Chaperones 5, 267-275[CrossRef][Medline] [Order article via Infotrieve]
  25. Dawson, R. M. C., Elliott, D. C., Elliott, W. H., and Jones, K. M. (1987) Data for Biochemical Research, 3rd Ed., pp. 213-232, Clarendon Press, Oxford
  26. Hans-Peter, K. (1988) Fat-soluble Pigments, CRC Press, Inc., Boca Raton, FL
  27. Henderson, M. J. (1989) Clin. Chem. 35, 1043-1044[Abstract/Free Full Text]
  28. Lamb, D., Kelly, D., and Kelly, S. (1999) Drug Resistance Updates 2, 390-402[CrossRef][Medline] [Order article via Infotrieve]
  29. Mercer, E. I. (1993) Prog. Lipid Res. 32, 357-416[Medline] [Order article via Infotrieve]
  30. Bruce, J. I., and Elliott, A. C. (2000) Br. J. Pharmacol. 131, 761-771[CrossRef][Medline] [Order article via Infotrieve]
  31. Philipp-Dormston, W. K., and Doss, M. (1975) Z. Naturforsch. Sect. C Biosci. 30, 425-426
  32. Chartrand, P., Tardif, D., and Sasarman, A. (1979) J. Gen. Microbiol. 110, 61-66[Medline] [Order article via Infotrieve]
  33. Cox, R., and Charles, H. P. (1973) J. Bacteriol. 113, 122-132[Abstract/Free Full Text]
  34. Kaderbhai, N., Karim, A., Hankey, W., Jenkins, G., Venning, J., and Kaderbhai, M. A. (1997) Biotechnol. Appl. Biochem. 25, 53-61
  35. Surdeanu, A., and Horodniceanu, T. (1968) J. Bacteriol. 96, 1882-1884[Free Full Text]
  36. Beale, S. I. (1994) Ciba Found. Symp. 180, 156-168[Medline] [Order article via Infotrieve]
  37. Verderber, E., Lucast, L. J., Van-Dehy, J. A., Cozart, P., Etter, J. B., and Best, E. A. (1997) J. Bacteriol. 179, 4583-4590[Abstract/Free Full Text]
  38. Spencer, P., and Jordan, P. M. (1995) Biochem. J. 305, 151-158[Medline] [Order article via Infotrieve]
  39. Crockett, N., Alefounder, P. R., Battersby, A. R., and Abell, C. (1991) Tetrahedron 47, 6003-6014[CrossRef]
  40. Goldman, B. S., Beckman, D. L., Bali, A., Monika, E. M., Gabbert, K. K., and Kranz, R. G. (1997) J. Mol. Biol. 268, 724-738[CrossRef][Medline] [Order article via Infotrieve]
  41. Berks, B. C., Sargent, F., and Palmer, T. (2000) Mol. Microbiol. 35, 260-274[CrossRef][Medline] [Order article via Infotrieve]
  42. Constantin, D., Francis, J. E., Akhtar, R. A., Clothier, B., and Smith, A. G. (1996) Biochem. Pharmacol. 52, 1407-1413[CrossRef][Medline] [Order article via Infotrieve]
  43. Mylchreest, E., and Charbonneau, M. (1997) Toxicol. Appl. Pharmacol. 145, 23-33[CrossRef][Medline] [Order article via Infotrieve]
  44. Jacobs, J. M., Sinclair, P. R., Bement, W. J., Lambrecht, R. W., Sinclair, J. F., and Goldstein, J. A. (1989) Biochem. J. 258, 247-253[Medline] [Order article via Infotrieve]
  45. Sinclair, P. R., Gorman, N., Walton, H. S., Sinclair, J. F., Lee, C. A., and Rifkind, A. B. (1997) Drug Metab. Dispos. 25, 779-783[Abstract/Free Full Text]
  46. Jacobs, J. M., Sinclair, P. R., Lambrecht, R. W., Sinclair, J. F., and Jacobs, N. J. (1990) J. Biochem. Toxicol. 5, 193-199[CrossRef][Medline] [Order article via Infotrieve]
  47. Kaderbhai, M. A., He, M. Y., Beechey, R. B., and Kaderbhai, N. (1990) DNA Cell Biol. 9, 11-25[Medline] [Order article via Infotrieve]
  48. Gorman, N., Ross, K. L., Walton, H. S., Bement, W. J., Szakacs, J. G., Gerhard, G. S., Dalton, T. P., Nebert, D. W., Eisenstein, R. S., Sinclair, J. F., and Sinclair, P. R. (2002) Hepatology 35, 912-921[CrossRef][Medline] [Order article via Infotrieve]
  49. Goodhew, C. F., Brown, K. R., and Pettigrew, G. W. (1986) Biochim. Biophys. Acta 852, 288-294
  50. Gallagher, S. R. (1999) in Current Protocols in Molecular Biology (Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, A. G., and Struhl, K., eds) Vol. 2, pp. 10.12B.11-10.12B.11, John Wiley & Sons, Inc., New York

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 Protein Sci.Home page
M. A. Kaderbhai, H. M. Davey, and N. N. Kaderbhai
A directed evolution strategy for optimized export of recombinant proteins reveals critical determinants for preprotein discharge
Protein Sci., September 1, 2004; 13(9): 2458 - 2469.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
278/46/45555    most recent
M212685200v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Akhtar, M. K.
Right arrow Articles by Kaderbhai, M. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Akhtar, M. K.
Right arrow Articles by Kaderbhai, M. A.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME