Genetic Rescue of Leishmania Deficiency in Porphyrin Biosynthesis Creates Mutants Suitable for Analysis of Cellular Events in Uroporphyria and for Photodynamic Therapy*

Leishmania was found deficient in at least five and most likely seven of the eight enzymes in the heme biosynthesis pathway, accounting for their growth requirement for heme compounds. The xenotransfection of this trypanosomatid protozoan led to their expression of the mammalian genes encoding δ-aminolevulinate (ALA) dehydratase and porphobilinogen deaminase, the second and the third enzymes of the pathway, respectively. These transfectants still require hemin or protoporphyrin IX for growth but produce porphyrin when ALA was supplied exogenously. Leishmania is thus deficient in all first three enzymes of the pathway. Uroporphyrin I was produced as the sole intermediate by these transfectants, further indicating that they are also deficient in at least two porphyrinogen-metabolizing enzymes downstream of porphobilinogen deaminase, i.e. uroporphyrinogen III co-synthase and uroporphyrinogen decarboxylase. Pulsing the transfectants with ALA induced their transition from aporphyria to uroporphyria. Uroporphyrin I emerged in these cells initially as diffused throughout the cytosol, rendering them sensitive to UV irradiation. The porphyrin was subsequently sequestered in cytoplasmic vacuoles followed by its release and accumulation in the extracellular milieu, concomitant with a reduced photosensitivity of the cells. These events may represent cellular mechanisms for disposing soluble toxic waste from the cytosol. Monocytic tumor cells were rendered photosensitive by infection with uroporphyric Leishmania, suggestive of their potential application for photodynamic therapy.

Leishmania, like other trypanosomatid protozoa, are among the rare examples of aerobic organisms, which depend on oxidative phosphorylation (for Leishmania mexicana and Leishmania amazonensis, see Refs. 1 and 2), but are defective in the synthesis of heme (3) required for electron transport respiratory complexes. This peculiar defect in tetrapyrrole biosynthesis is manifested as a nutritional requirement for hemin by these organisms in chemically defined medium (for review, see Ref. 3). In nature, these parasitic protozoa must acquire protoporphyrin IX or heme exogenously from their hosts as a nutritional factor (3). Exceptional are several entomophilic nonpathogenic Crithidia species that harbor ␤-proteobacteria as endosymbionts presumably to help them complete the heme biosynthetic pathway, thereby sparing their nutritional requirement for hemin as an essential growth factor (4).
In the present studies, the heme biosynthetic pathway was found far more defective in trypanosomatids than expected as determined by genetic complementation of naturally symbiontfree Leishmania. These organisms normally infect mammalian macrophages as intracellular parasites and acquire heme via the activity of their host cells (3). We report here that transgenic Leishmania with alad and pbgd became highly porphyric when supplied with ALA, indicative of their deficiencies in ␦-aminolevulinate synthase in addition to ALAD and PBGD. The production of uroporphyrin I as the sole intermediate under these conditions also indicates the deficiency of porphyrinogen-modifying enzymes further downstream of the pathway. Supplying the transfectants with exogenous ALA caused cellular accumulation of uroporphyrin I followed by its release. The infection of monocytic tumor cells in vitro with these porphyric Leishmania followed by UV irradiation resulted in their * The work is supported in part by UHS/Chicago Medical School and National Institutes of Health Grants AI 20486 (to K. P. C.) and DK32890 (to S. S.). 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  cytolysis, suggestive of its potential application for photodynamic therapy of various diseases.

EXPERIMENTAL PROCEDURES
Cell Cultures-Wild type L. amazonensis (LV78) promastigotes (clone 12-1) were grown at 25°C in Hepes-buffered Medium199 to pH 7.4 and supplemented with 10% heat-inactivated fetal bovine serum. Transfectants were grown under similar conditions with different concentrations of selective pressure, i.e. G418 and/or tunicamycin. Cells were also adapted to grow in a chemically defined medium (11). To initiate such cultures, cells were washed twice with the defined medium by centrifugation at 3500 ϫ g for seeding at 2-5 ϫ 10 6 cells/ml. Cells were counted using a hemacytometer. Macrophages (J774A1) were grown in RPMI 1640 medium-supplemented with 10 or 20% heatinactivated fetal bovine serum at 35°C. Cultures of all cells rendered porphyric were kept in the dark to avoid cytolysis because of photosensitivity.
Porphyrin Fluorescent Microscopy-For all microscopic examinations of Leishmania, living cell suspension in 5-10 l aliquots was placed on a glass slide and then covered with an 18 mm 2 glass coverslip. For routine examinations, the preparations were viewed under phase contrast for cellular structures in conjunction with epifluorescence for porphyrins using a filter set consisting of D405/10X (405 nm exciter), 485DCXR (485 nm dichroic) and RG610LP (610 nm emitter) (Chroma Tech Co., Brattleboro, VT) in a Zeiss standard microscope with super pressure mercury lamp (HBO 50 W, Osram). Images were obtained by confocal microscopy using an Olympus FluoView confocal microscope equipped with a krypton/argon-mixed gas laser. Specimens were illuminated with the 488 nm excitation line. The specific fluorescent emission of the porphyrin was collected by a photomultiplier tube after passing through a 605 nm bandpass emission filter. Differential interference contrast images were simultaneously collected using a transmission field detector coupled to a photomultiplier tube. Detection settings were determined using a negative control by adjusting the gain and offset settings to eliminate background. Images were collected using a ϫ100 oil immersion objective (NA 1.40) with an electronic zoom of ϫ3. The confocal aperture was set to 5 mm to maximize the depth of field within the specimen. Digital image acquisition took ϳ7 s/frame, resulting in movement-induced blurring of the flagella in viable specimens. Images were composed in Adobe Photoshop. Only differential interference contrast images were adjusted for brightness.
Nucleic Acid Techniques-The cDNA of rat pbgd (1038 bp) (Gen-Bank TM accession number X06827) (12) was obtained by digesting the plasmids with BamHI. The human alad (993 bp) (GenBank TM accession number M13928) (13) was PCR-amplified from a cDNA cloned in pGEM vector using a high fidelity Taq polymerase (Expand Hi Fi, Roche Molecular Biochemicals). The forward and reverse primers used were 5Ј-TGCCCACTGGATCCCCGCCATG-3Ј and 5Ј-CACTGGGATCCAT-CATTCCTCC-3Ј. To facilitate cloning into the Leishmania expression vectors, the primer sequences were designed to include BamHl sites (underlined) flanking the PCR products. The amplified products of alad was first cloned in pGEM-T for expansion and then gel-purified after BamHI digestion for cloning into Leishmania-specific vector, pX-neo (14). The rat pbgd was cloned into p6.5 with N-acetylglucosamine-1phosphate transferase gene for tunicamycin-resistance (15)(16)(17). The clones with the inserts in correct orientation were identified by restriction mapping. Promastigotes were transfected with pX-alad and/or p6.5-pbgd (see Fig. 1) by electroporation as described earlier (18) and selected initially for resistance of up to 10 g of tunicamycin/ml or 20 g of G418/ml or a combination of both. Stable transfectants emerged in 8 -10 days and were subsequently passaged continuously in media with appropriate drug pressures.
Enzyme and Porphyrin Assays-Cells were harvested by centrifuga-tion for 5 min at 3500 ϫ g, resuspended in phosphate-buffered saline (pH 7.4), and lysed by three cycles of freezing-thawing in dry ice/acetone bath. Cell lysates equivalent to 20 -50 ϫ l0 6 cells and to 2-5 ϫ 10 6 cells were used for ALAD and PBGD assays, respectively. The activity of ALAD was assayed by monitoring the absorption at 553 nm of the color salt of porphobilinogen using the modified Ehrlich reagent as described previously (19). PBGD activities were assayed by a microfluorometric method (20). Porphyrin levels were determined fluorometrically using 5 l of cell suspensions (2-5 ϫ 10 6 cells/ml) and 0.5 ml of 1 M perchloric acid/methanol (1:1, v/v) as described previously (21). Samples were assayed for proteins using Coomassie Blue R-250 binding dye. The type of porphyrins produced was determined by thin-layer chromatography of relevant samples using porphyrin ester chromatographic marker kit as the standard (Porphyrin Products Co., Logan, UT). Cells were grown in porphyrin-free chemically defined medium to 3-4 ϫ 10 8 cells. Porphyrins were extracted from the cell pellets, methylated, and analyzed by thin-layer chromatography as described previously (22).
Western Blot Analysis-Stable transfectants grown in Medium199 supplemented with heat-inactivated fetal bovine serum and selected with appropriate drugs were assessed for the presence of ALAD and PBGD by Western blot analysis. Protein samples each equivalent to 20 ϫ l0 6 cells were subjected to SDS-PAGE using MiniProtean II (Bio-Rad) and blotted to nitrocellulose. The primary anti-PBGD and anti-ALAD antisera were generated by immunization of rabbits with purified enzymes (23). Both were used at 1:10 5 dilution. Peroxidaseconjugated goat anti-rabbit IgG (Sigma) was used as the secondary antibody. Immunoblots were subsequently developed with the ECL reagent and exposed to x-ray films.
UV Sensitivity Assays-For these experiments, transfectants with alad and pbgd and those with pbgd alone were grown in chemically defined medium supplemented with up to 1.6 mM ALA to generate different levels of porphyria. Cell suspensions in 24-well microtiter plates (10 7 promastigotes/ml/well or 5 ϫ 10 6 promastigotes ϩ 5 ϫ 10 5 J774A1 macrophages/ml/well) were irradiated after infection or immediately at room temperature under a long wave UV lamp (254 -366 nm multi-bands, Mineralight Lamp, Model UVSL-58, Ultraviolet Products, Inc, San Gabriel, CA) placed ϳ5 cm above the cell layers. Porphyric Leishmania prepared under other conditions and their spent media with different concentrations of released porphyrins were also examined for their effects on J774A1 cells. After illumination for variable time periods, cells were microscopically examined immediately. Cells of the monocytic tumor line were counted using a hemacytometer 1-2 days after irradiation. All experiments were repeated at least twice.   The results thus indicate that both genes are expressed at the protein level individually in different transfectants and simultaneously in the same one using different vectors.

Expression of PBGD and ALAD Only in Leishmania Trans
Both ALAD and PBGD Expressed in the Transfectants Were Enzymatically Active-Both ALAD and PBGD activities are absent in wild type cells (data not shown) and present only in transfectants with the genes of relevance (Table II). The specific activities in pmol products/mg protein/h fall within the range of ϳ2500 to ϳ9500 and ϳ400 to ϳ1400 for ALAD and PBGD, respectively. The variations in the specific activities among different experiments seen may be accounted for by differences introduced inadvertently in the culture and selective conditions used. Clearly, both enzymes are fully functional alone or in combination in the transgenic Leishmania cells.
␦-Aminolevulinate-inducible Uroporphyria in Transfectants with pbgd and alad-Whereas both ALAD and PBGD were expressed and fully active in Leishmania transfected with the respective gene, the transfectants produced no detectable porphyrins (see Figs. 3, lanes 3 and 6, and 4, panel N, 0 M ALA) unless ALA was provided to those with both transgenes (Figs.  3, lanes 2 and 5, and 4, panel N, 125-1000 M ALA). However, this porphyric Leishmania along with all other transfectants resembled nontransfected wild type cells in that they grew continuously only in the defined medium supplemented with either hemin or protoporphyrin IX (data not shown). Deletion of the heme compound from this medium resulted in the eventual cessation of their growth in all cases after several passages. Heme biosynthesis pathway thus remains incomplete in these transgenic Leishmania clearly because of additional enzymatic defect(s) downstream of PBGD.
Uroporphyrin I Is the Sole Intermediate Detected in Porphyric Leishmania-This finding was originally suggested by the fluorescence emission spectra of porphyrins extracted from porphyric Leishmania observed (data not shown) and confirmed by thin-layer chromatography analysis of these samples (Fig. 3). Thin-layer chromatography of porphyrins extracted by standard procedures from porphyric Leishmania and their spent medium revealed only a single UV-fluorescent species (Fig. 3,  lanes 2 and 5), which co-migrated with uroporphyrin I octamethyl esters (lanes 1, 4, and 7). This finding indicates that only uroporphyrin I was produced by these cells. No porphyrin bands were visible in samples prepared simultaneously from controls, e.g. transfectants with one or the other gene and their culture supernatants (Fig. 3, lanes 3 and 6). The cells used for sample extraction were grown in porphyrin-free defined medium, eliminating the possibility that the porphyrin species detected may have derived from an exogenous source.
Emergence and Cellular Localization of Uroporphyrin I in Porphyric Leishmania-The porphyrins emerged only in the double transfectants after the addition of ALA into their culture media. Porphyrin-specific signals were followed by epifluorescent microscopy and imaged by confocal fluorescent microcopy (see "Experimental Procedures" for the settings used). By differential interference contrast microscopy, living cells under all conditions used appeared granulated with anterior flagella (Fig. 4, panels A, D, G, and J). Under the settings for confocal microscopy used for porphyrin, fluorescence signals emerged only in the double transfectants (Fig. 4, panels H and K) but not in the control cells, e.g. the single transfectants with pbgd alone (panels B and E). When the two sets of images from the same preparations were merged, porphyrin fluorescent signals appeared to be diffused in the cytosol (Fig. 4, panel I) as well as localized in cytoplasmic vacuoles (panels I and L).
Intracellular Accumulation of Uroporphyrin Followed by Its Extracellular Release-Porphyric Leishmania released uroporphyrin I into the medium, independent of cytolysis. This was demonstrated under two different conditions to generate modest and high levels of uroporphyria. Cells were handled gently to avoid inadvertent cytolysis. The kinetics of uroporphyrin accumulation in and release from porphyric Leishmania was quantitatively assessed fluorometrically. Initially used were cells grown in a chemically defined medium with a modest selective pressure of 2 g of tunicamycin and 10 g of G418/ml in conjunction with increasing but low concentrations of ALA from 0 to 200 M (Fig. 5). Under all these conditions, cells grew from 2.5 ϫ 10 6 to ϳ10 7 /ml in a period of 3 days (Fig. 5, left  panels), except the one with the highest ALA concentration of 200 M in which case the cell density decreased on day 3 (Fig.  5, bottom left panel). In the absence of ALA, porphyrin was detected neither in cells nor in their spent media throughout the period of cell growth (Fig. 5, top middle and right panels).
In the presence of ALA, the cells produced uroporphyrin in an ALA dose-dependent manner, namely an increase from ϳ3 to ϳ8 pmol uroporphyrin/10 6 cells in the presence of 25 to 200 M ALA during the first day (Fig. 5, middle panels). The cellular levels of uroporphyrin declined in these cells from days 2 to 3, concomitant with its release also in an ALA dose-dependent manner from 5 to 28 pmol uroporphyrin/ml in the culture medium (Fig. 5, right panels). In a separate set of experiments, cells were grown in Medi-um199 plus heat-inactivated fetal bovine serum under the optimal conditions for uroporphyria, i.e. a 10-fold increase of the selective pressure (20 g of tunicamycin and 100 g of G418/ ml) and a 5-to 8-fold increase of the substrate (up to 1.0 -1.6 mM ALA provided exogenously). Under these conditions, both cellular and released uroporphyrin levels were considerably enhanced (Fig. 4, panels N, 125-1000 M ALA), the latter reaching a level as much as ϳ2 M. Cytolysis was observed in Ͻ1% of these cells that did not account for the level of porphyrin release seen.
The results from both sets of the experiments indicate that uroporphyria is induced in an ALA dose-dependent fashion, which is marked by initial cellular accumulation of uroporphyrin followed by its release and accumulation in the culture medium.

UV Sensitivity of Porphyric Leishmania and Porphyric Leishmania-infected Monocytic Tumor Cells-Porphyric
Leishmania remained motile and thus viable under all culture and selective conditions used, except when they were subjected to UV irradiation. This sensitivity was indicated by the immediate cessation of the motility of the early porphyric cells after exposure to illumination under the setting for epifluorescent microscopy or with the long wave UV lamp. Late porphyric cells exposed to ALA 2 days or longer were less sensitive, whereas nonporphyric cells were totally insensitive to UV irradiation under these conditions as indicated by their motility.
The monocytic tumor cells, J774A1, were also rendered sen-sitive to long wave UV irradiation after infection with porphyric Leishmania. Used for these experiments were double transfectants with both alad and pbgd and single transfectants with only pbgd grown under the same conditions. Uroporphyria was generated only in the double transfectants. The results (Fig. 6) showed that UV irradiation lysed only the macrophages infected with porphyric Leishmania and that the cytolysis was proportional to the porphyric levels of the latter modulated by prior exposure to different ALA concentrations (Fig. 6, PBGD/ ALAD). The nonporphyric Leishmania produced no such effect (Fig. 6, PBGD) regardless of their exposure to ALA and UV irradiation under the same conditions. There was also no cytolysis of the tumor cells when irradiated immediately after mixing them with the porphyric Leishmania or in the presence of their spent media containing uroporphyrin I. The results obtained from these experiments were similar to the control in Fig. 6 (data not shown).

DISCUSSION
In this study, both alad and pbgd from mammalian sources were successfully expressed in Leishmania by transfection (Fig. 1), yielding products of expected size (Fig. 2) with enzymatic activities (Table II). Significantly, the episomal transgenes in two different vectors can be selected appropriately to co-express both enzymes with activities. These activities are at least 10 times higher than those normally found in the mammalian cells, e.g. macrophages (3), and more comparable to those in murine Friend virus-transformed leukemia cells induced for erythroid differentiation with a heightened level of heme biosynthesis (23). Both mammalian genes thus appear to express adequately and produce functionally active products in a xenotransgenic system with Leishmania as the recipient.
Despite the fact that functionally active ALAD and PBGD were made available in abundance to these transgenic Leishmania, their nutritional dependence on hemin or protoporphyrin IX was not spared, indicative of more extensive deficiencies of this pathway than previously expected. Transfection of Leishmania with pbgd alone was originally expected to produce the desired phenotype, because PBGD was thought to be the only enzyme that is deficient in this group of protozoa and supplied to several Crithidia species by their endosymbionts (5). Because ␦-aminolevulinate synthase activity was detected previously in Leishmania (7,8) and other trypanosomatids (6), additional transfection of these organisms with alad would be expected to provide the first three enzymes of the pathway sufficient to produce, at the very least, one or more of the porphyrin intermediates if not heme as the final product (see Table I). However, the double transfectants produced porphyrins only when ALA was supplied exogenously (Figs. 3-5). The deficiency of Leishmania in ␦-aminolevulinate synthase in addition to ALAD and PBGD is thus apparent. This conclusion is supported by the fact that the substrates for ALA synthesis are not limiting, i.e. glycine and alanine amply supplied in the   FIG. 3. Thin layer chromatogram showing cellular and released uroporphyrin in porphyric L. amazonensis. Cells transfected with p6.5-pbgd/pX-alad and p6.5-pbgd/pX were grown for 2 days in a chemically defined medium with 1 mM ALA. Cells and spent medium were separated by centrifugation and subjected to porphyrin extraction for thin layer chromatography (see "Experimental Procedures" for details). The porphyrin standards include type I uroporphyrin, which can be separated from type III uroporphyrin (data not shown) under the chromatographic conditions used. Lanes 1, 4, and 7, porphyrin carboxymethyl ester standards; lanes 2 and 5, cellular and released porphyrins extracted from transfectants with pbgd/alad and their culture supernatants, respectively; lanes 3 and 6, absence of porphyrins from transfectants with pbgd alone and their culture medium, respectively. COOMe, carboxymethyl groups. * Transfectants with the pX vector alone in addition to p6.5-pbgd. 1 Grown to stationary phase in a defined medium and harvested for enzyme assays as described under "Experimental Procedures." See Fig. 1 for the plasmid constructs used for the transfection.
culture medium, and succinyl-CoA in abundance (6) from the active TCA cycle in these cells. The loss of alas and alad from Leishmania is of interest, because their products are thought to have additional functions in other eukaryotes besides heme biosynthesis, i.e. ALA for the formation of corin ring in cobalamin biosynthesis and ALAD as CF-2 inhibitor of proteasome ATP-and ubiquitin-dependent proteolysis (24). The emergence of uroporphyrin I in the porphyric Leishmania entails the occurrence of cellular events in the following order: (a) transport of exogenously supplied ALA into the cytosol where it is converted by ALAD into porphobilinogen, which in turn is changed into hydroxymethylbilane by PBGD, and (b) spontaneous polymerization of hydroxymethylbilanes into uroporphyrinogen I, which undergoes autooxidation to form uroporphyrin I. The   FIG. 4. Cellular localization and ALA dose-dependent release of porphyrin from porphyric L. amazonensis. Panels A-L, cells transfected with p6.5-pbgd/pX-alad and p6.5-pbgd/pX were examined by confocal microscopy for the presence of cellular porphyrins after exposure to 1 mM ALA for 2 days. See "Experimental Procedures" for the settings used for differential interference (DIC) (panels A, D, G, and J), porphyrin fluorescence (Porphyrin) (panels B, E, H, and K), and merged images of DIC and porphyrin (Merged) (panels C, F, I, and L). Panels A-F, cells transfected with the control P6.5-pbgd/pX; panels G-L, cells transfected with P6.5-pbgd/pX-alad. Note that the porphyrin signals diffused in the cytosol and localized more intensely to cytoplasmic vacuoles (panels H, I, K, and L) in some cells and in the cytosol as a diffused pattern (panel I) in others. Panels M and N, transfectants with P6.5-pbgd/pX-alad (PBGDϩALAD) and the control with P6.5-pbgd/pX (PBGD only) were exposed to 0 -1 mM ALA for 4 days. accumulation of uroporphyrin I as the sole porphyrin species (Fig. 3) indicates the absence or deficiency of the porphyrinogen-modifying enzymes, i.e. uroporphyrinogen co-synthase, uroporphyrinogen decarboxylase, co-proporphyrinogen oxidase, and protoporphyrinogen oxidase (Table I), which would otherwise catalyze the formation of uroporphyrinogen III and co-proporphyrinogen III and protoporphyrinogen IX and protoporphyrin IX, respectively. Because all of these enzymes downstream of PBGD catalyze a cascade of ring-modifying reactions specific to various porphyrinogens (Table I), their loss may not be unexpected in the absence of the upstream substrate. Curiously, ferrochelatase, the last enzyme of the pathway, remains functional in Leishmania for heme biosynthesis. The presence of this enzyme was previously demonstrated biochemically by detecting its catalytic activity (6, 7) as well as nutritionally by the substitution of hemin in the culture medium with protoporphyrin IX for cell growth (5). Nonenzymatic formation of heme from ferric iron and protoporphyrin IX has not been reported, except under nonphysiological conditions (25). It awaits further study to determine whether ferrochelatase may have additional function beyond heme biosynthesis in trypanosomatids. The extensive defects of Leishmania in this pathway reported here further underscore the importance of heme compounds as the most unique of their essential nutritional requirements. Restricting the availability of this nutrient may thus potentially contribute to the regulation of Leishmania virulence in natural infection as well as serve as a potential strategy to design therapeutic drugs for treating leishmaniasis.
Finding the extensive defects of Leishmania in heme biosynthesis and their partial rectification by genetic complementation inadvertently provides a novel model suitable for elucidating the cellular response to porphyria. The key feature is the apparent absence of ␦-aminolevulinate synthase, which renders this model substrate-inducible with exogenously supplied ALA in an unregulated manner. The accumulation of different porphyrins has been reported in transfectants or mutants of porphyrin-modifying enzymes from mammalian cells (26) and other lower eukaryotes, e.g. yeast (27)(28)(29) and Chlamydomonas (30). However, these mutants develop porphyria of modest levels with "background noises" apparently because of feedback regulation and transcriptional and allosteric control of their endogenous heme biosynthesis pathway. This observation was especially notable when ALA was used in attempt to increase porphyrin levels of tumor cells for photodynamic therapy (31)(32)(33). In contrast, xenotransgenic Leishmania with dysregulated expression of alad and pbgd responded proportionally to increasing levels of exogenous ALA. As a result, uroporphyria was developed in these cells, making it possible to follow the emergence, progression, and consequence of this event at the cellular level (Figs. [3][4][5]. Several cellular events were documented for the first time in the development of uroporphyria with the Leishmania model. Uroporphyrin was initially found to distribute throughout the cells indicating that it is synthesized in the cytosol and diffused in this compartment, which is consistent with the hydrophilic property of this porphyrin. The cell-wide distribution of uroporphyrin apparently renders these early porphyric cells more sensitive to UV irradiation presumably as a result of the generation of free radicals via oxidation of this porphyrin (34), accounting for their rapid paralysis. The subsequent condensation of diffused porphyrin into cytoplasmic vacuoles may be carrier-mediated or mediated by a proton pump of the vacuolar membrane. Interestingly, exogenous uroporphyrin is poorly internalized by living Leishmania, 2 suggesting that the mechanism responsible for vacuolar condensation of the intracellular uroporphyrin is not operational in the plasma membrane for porphyrin uptake. The "porphyrinosomes" (Fig. 4, panels H and K) subsequently formed in the porphyric Leishmania may be the lysosomal or endosomal compartment of these organisms, because uroporphyrin has been reported to accumulate in acidic vacuoles of other cells (35). Of special interest is the disappearance of the cytosolic fluorescence with the vacuolar condensation of uroporphyrin. This cannot be accounted for by the cessation of uroporphyrin synthesis in the system due to substrate limitation or inactivation of the enzymes involved, because there was an abundant supply of the exogenous ALA of up to 2 mM and cells were kept in total darkness to minimize photosensitization of intracellular porphyrin to generate protein-denaturing free radicals. The kinetics of the cellular events suggests that uroporphyrin is removed from the cytosol by an increased efficiency of the vacuolar transport mechanism and/or its efflux out of the cells. Vacuolar condensation of uroporphyrin appears to occur slightly ahead of its extracellular efflux, suggesting that the former event may produce a signal to trigger the occurrence of the latter. Interestingly, the suspension of Leishmania in uroporphyrin-containing medium does not render them photosensitive, 2 suggesting that the efflux of porphyrins from cells may represent a significant mechanism of toxic waste disposal. A number of membrane proteins have been described in Leishmania as efflux systems related to their drug resistance (36 -38), and putative receptors/transporters have also been described for heme (39) and hemoglobin (40). Further investigation is needed to determine whether any of these transport molecules or a novel one may be related to the release of the porphyrins seen. The cellular events of uroporphyrin I emergence, accumulation, and release seen in Leishmania may be relevant to the disposal of "uroporphyria" also found in other conditions, such as porphyria cutanea tarda (41). Further studies of these cellular events in Leishmania may shed light on the mechanisms of cytopathology in and management for this and other types of human porphyria.
The availability of porphyric Leishmania presents the opportunity of considering their use in concept to deliver exogenous porphyrins for photodynamic therapy, still an evolving idea for treating cancers and other malignant diseases (42). We tested this possibility with cells of J774 murine monocytic tumor line of macrophage origin. Uroporphyric and aporphyric Leishmania infect these cells in vitro equally well, but only the uroporphyric Leishmania render them sensitive to cytolysis by UV irradiation (Fig. 6). Intracellular residence of the porphyric Leishmania is required, because the tumor cells were not photosensitized when mixed with the porphyric Leishmania without infection or when suspended in their spent media with abundant uroporphyrin. Thus, the uroporphyrin responsible for photosensitivity of the infected cells must be from "within," but not from outside the tumor cells, consistent with the observations on the porphyric Leishmania themselves (see above). The porphyric Leishmania are expected to lodge in the phagolysosomes where these organisms are known to reside normally in the infected cells (43). Thus, cytolysis of the monocytic tumor cells results probably from a combined action of the free radicals from sensitized porphyrins plus the lysosomal enzymes of the target cells and possibly additional lytic factors from disrupted Leishmania in this vacuolar compartment.
Leishmania have evolved tissue-tropism toward the reticuloendothelial system to achieve intralysosomal parasitism of macrophages (43). Especially attractive are the potential use of the nonpathogenic or avirulent species/strains attained via "molecular attenuation" (44) to deliver pro-drugs for lysosomal activation and antigens for vaccination. When further genetically grafted to express these agents, porphyric Leishmania may be used as a "suicidal capsule" for their timely release after pulsing with ALA as a trigger followed by UV irradiation for cytolysis. Along the same vein, cell and tissue specificity of 2 K.-P. Chang, unpublished observation.
FIG. 6. Cytolysis of J774A1 monocytic tumor cells exposed to porphyric Leishmania. Macrophages J774A1 were grown to confluence in 24-well microtiter plate to ϳ10 6 cells/well. They were infected 10:1 with control (Leishmania transfectants containing only the alad gene) and tested (Leishmania transfectants with both alad and pbgd genes) promastigotes grown in the defined medium supplemented with 400 and 1600 M ALA. Infection was allowed to proceed overnight followed by irradiation with long wave UV for 1 h. After further incubation overnight, macrophages were stripped and counted. The value for each time point was the average ϩ S.D. from duplicate samples for each of two independent experiments. other parasites may be further exploited to design different targeting strategies. The feasibility of this concept to deploy porphyric parasites for cell-and tissue-specific photodynamic prophylaxis and therapy must await further experimental evaluation in vivo.