Search

The clinical success of mitomycin C (1) and its associated toxicities and resistance have led to efforts to prepare semisynthetic analogues (i.e., KW-2149 (3), BMS-181174 (4)) that have improved pharmacological profiles. In this study, we report the preparation and evaluation of the novel 7-N-(1'-amino-4',5'-dithian-2'-yl)porfiromycin C(8) cyclized imine (6) and its reference compound, 7-N-(1'-aminocyclohex-2'-yl)porfiromycin C(8) cyclized imine (13). Porfiromycin 6 contains a disulfide unit that, upon cleavage, may provide thiol(s) that affect drug reactivity. We demonstrated that phosphines dramatically accelerated 6 activation and solvolysis in methanolic solutions ("pH 7.4") compared with 13. Porfiromycins 6 and 13 efficiently cross-linked EcoRI-linearized pBR322 DNA upon addition of Et3P. We found enhanced levels of interstrand cross-link (ISC) adducts for 6 and 13 compared with porfiromycin (7) and that 6 was more efficient than 13. The large Et3P-mediated rate enhancements for the solvolysis of 6 compared with 13 and a N(7)-substituted analogue of 1, and the increased levels of ISC adducts for 6 compared with 13 and 7 are attributed to a nucleophile-assisted disulfide cleavage process that permits porfiromycin activation and nucleophile (MeOH, DNA) adduction. The in vitro antiproliferative activities of 6 and 13 using the A549 tumor cell line (lung adenocarcinoma) were determined under aerobic and hypoxic conditions and then compared with 7. Both 6 and 13 were more cytotoxic than 7, with 13 being more potent than 6. The C(8) iminoporfiromycins 6 and 13 displayed anticancer profiles similar to 3.

Porfiromycin (PM), a bioreductive alkylating agent, is currently under development for the treatment of head and neck cancers as an adjunct to radiation therapy in phase III clinical trials. After i.v. administration of a single dose of PM to patients at 40 mg/m2, urinary metabolites were isolated by HPLC and identified by atmospheric pressure chemical ionization mass spectrometry. In dogs, [methyl-3H]PM was administered i.v. to three Beagle dogs at a single dose of 2 mg/kg. Urinary excretion of radioactivity and PM at different times was determined by liquid scintillation counting and by HPLC, respectively. An average of 48.0% of total radioactivity given to the dogs was cumulatively excreted in urine over a period of 7 days. Unchanged parent drug excreted in urine accounted for 10.8% of the administered dose over the same period of time. The results indicated that the majority of excreted dose in dog urine was in the form of metabolites. Three phase I and four phase II metabolites of PM were identified in human and dog urine. The phase I metabolites are 2-methylamino-7-aminomitosene, 1,2-cis and 1,2-trans-1-hydroxy-2-methylamino-7-aminomitosenes. The phase II metabolites are a pair of isomeric N-acetylcysteine S-conjugates and a pair of isomeric cysteine S-conjugates of mitosenes at the C-1 and C-10 positions. Most of the identified metabolites were confirmed by comparison with synthetic reference standards using HPLC and liquid chromatography/mass spectrometry (LC/MS). The identification of mercapturic acids and cysteine S-conjugates in urine indicates that the metabolism of PM may be through GSH conjugation.

The isolation and identification of the major metabolites of porfiromycin formed in the presence of a rat liver preparation under aerobic conditions were performed with high-performance liquid chromatography and electrospray ionization mass spectrometry. Porfiromycin was extensively metabolized by the rat liver preparation in an aqueous 0.1 M potassium phosphate buffer (pH 7.4) containing an NADPH generating system at 37 degrees C. A total of eight metabolites was identified as mitosene analogs. Of these, three primary metabolites are 2-methylamino-7-aminomitosene, 1,2-cis and 1,2-trans-1-hydroxy-2-methylamino-7-aminomitosene, which are consistent with those previously observed in hypoxia using purified rat liver NADPH-cytochrome c reductase. Interestingly, 2-methylamino-7-aminomitosene is a reactive metabolite, which undergoes further activation at the C-10 position by the loss of carbamic acid and then links with the 7-amino group of the primary metabolites to yield two dimeric adducts. In addition, three phosphate adducts, 10-decarbamoyl-2-methylamino-7-aminomitosene-10-phosphate, 1,2-cis and 1,2-trans-2-methylamino-7-aminomitosene-1-phosphate, were also identified in the incubation system. The configurations of the diastereoisomeric metabolites were determined with (1)HNMR and phosphatase digestion. On the basis of the metabolite profile, we propose in vitro metabolic pathways for porfiromycin. The findings provide direct evidence for understanding the reactive nature and hepatic metabolism of the drug currently in phase III clinical trials., (Copyright 2000 Wiley-Liss, Inc.)

Purified NADPH:cytochrome c (P-450) reductase (FpT; NADPH-ferrihemoprotein oxidoreductase, EC 1.6.2.4) can reductively activate mitomycin antibiotics through a one-electron reduction to species that alkylate DNA. To assess the involvement of FpT in the intracellular activation of the mitomycins, transfectants overexpressing a human FpT cDNA were established from a Chinese hamster ovary cell line deficient in dihydrofolate reductase (CHO-K1/dhfr-). The parental cell line was equisensitive to the cytotoxic action of mitomycin C under oxygenated and hypoxic conditions. In contrast, porfiromycin was considerably less cytotoxic to wild-type parental cells than was mitomycin C in air and markedly more cytotoxic under hypoxia. Two FpT-transfected clones were selected that expressed 19- and 27-fold more FpT activity than the parental line. Levels of other oxidoreductases implicated in the activation of the mitomycins were unchanged. Significant increases in sensitivity to mitomycin C and porfiromycin in the two FpT-transfected clones were seen under both oxygenated and hypoxic conditions, with the increases in toxicity being greater under hypoxia than in air. These findings demonstrate that FpT can bioreductively activate the mitomycins in living cells and implicate FpT in the differential aerobic/hypoxic toxicity of the mitomycins.

Porfiromycin (methyl mitomycin C) has been shown in laboratory studies to have increased preferential cytotoxicity to hypoxic cells and therefore may provide enhanced therapeutic efficacy over mitomycin C when used in combination with radiation therapy (RT). The purpose of the two clinical studies reported here is to evaluate the concomitant use of porfiromycin with RT in the management of squamous cell carcinoma of the head and neck. Between October 1989 and July 1992, 21 patients presenting with locally advanced stage III/IV squamous cell carcinoma of the head and neck were entered into a phase I toxicity trial evaluating porfiromycin as an adjunct to RT. Patients were eligible if they had biopsy documented squamous cell carcinoma of the head and neck with a low probability of cure by conventional means. Patients were treated with standard fractionated daily RT to a total median dose of 63 Gy, with porfiromycin administered on days 5 and 47 of the course of RT. Upon completion of this phase I trial, a phase III trial was initiated in November 1992 randomizing patients with squamous cell carcinoma of the head and neck to RT with mitomycin C vs. RT with porfiromycin. There is no radiation only arm in this current trial. To date, 75 patients have been entered on this trial and acute toxicity data are available on 67 patients (34 porfiromycin, 31 mitomycin C) who have completed their entire course of treatment. Median follow-up of the 21 patients enrolled in the phase I porfiromycin trial is 58.5 months. Of the 21 patients, 5 were treated at a dose of 50 mg/M2, 4 at 45 mg/M2, and the final 12 at 40 mg/M2, which appeared to result in acceptable acute hematological and nonhematological toxicities. As of December 1995, 14 of the 21 patients have died with disease and 7 remain alive and free of disease, resulting in a 5-year actuarial survival of 32%. Of the patients enrolled to date in the phase III randomized trial of mitomycin C vs. porfiromycin, there have been no statistically significant differences between the two arms with respect to white blood cell count (WBC), platelet, or hemoglobin nadirs. Acute nonhematological toxicities including mucositis, epidermitis, odynophagia, and nausea have also been comparable. Two patients in this current randomized trial died during treatment, apparently of nondrug-related causes. We conclude that the bioreductive alkylating agent porfiromycin has demonstrated an acceptable toxicity profile to date. Final analysis of the phase I trial, which revealed a 5-year no evidence of disease survival rate of 32% in patients with locally advanced disease and a low probability of cure, appears encouraging. We anticipate completion of the current ongoing trial comparing mitomycin C to porfiromycin in the next 2 years. Further investigations, including large-scale multiinstitutional trials employing bioreductive alkylating agents or other hypoxic cell cytotoxins as adjuncts to RT, are warranted.

Hypoxic cells in solid tumors represent a therapeutically resistant population that limits the curability of many solid tumors by irradiation and by most chemotherapeutic agents. The oxygen deficit, however, creates an environment conducive to reductive processes; this results in a major exploitable difference between normal and neoplastic tissues. The mitomycin antibiotics can be reductively activated by a number of oxidoreductases, in a process required for the production of their therapeutic effects. Preferential activation of these drugs under hypoxia and greater toxicity to oxygen-deficient cells than to their oxygenated counterparts are obtained in most instances. The demonstration that mitomycin C and porfiromycin, used to kill the hypoxic fraction, in combination with irradiation, to eradicate the oxygenated portion of the tumor, produced enhanced cytodestructive effects on solid tumors in animals has led to the clinical evaluation of the mitomycins in combination with radiation therapy in patients with head and neck cancer. The findings from these clinical trials have demonstrated the value of directing a concerted therapeutic attack on the hypoxic fraction of solid tumors as an approach toward enhancing the curability of localized neoplasms by irradiation.

Radiobiological data and measurements with O2 microelectrodes show that EMT6 tumors implanted into aged mice have a higher proportion of radioresistant, hypoxic cells than do tumors implanted into young adult animals; radiation is less effective in killing cells in tumors in old mice than in tumors in young adult mice. The studies reported here examine the effects of porfiromycin (POR), a bioreductive alkylating agent shown previously to be preferentially toxic to hypoxic EMT6 cells in vitro and in solid tumors in young adult mice. POR was effective in attacking the hypoxic cells of tumors in aged mice; regimens combining POR with x-rays overcame the radioresistance of tumors in the old animals. Comparisons of the distribution of 3H-labeled POR in young and old mice showed that tumors in aged mice had a slightly larger proportion of areas with necrotic features, which bound higher levels of tritiated POR than did healthy tumor regions without necrotic features. Studies of histology, lissamine green distributions, binding of tritiated POR, and radiation and POR cytotoxicity suggested that tumors in old mice contained a larger proportion of poorly perfused tumor cells, and that cells in these regions were resistant to radiation and sensitive to POR. Studies of the distribution of POR in normal tissues and of the toxicity of POR to bone marrow progenitor cells (CFU-GM) revealed no differences between young and old animals, showing that the differences observed in tumors reflected differences in the microenvironments within the tumors, rather than differences in the processing of drug in young and old animals.

A mitomycin C (MMC)- and porfiromycin (PFM)-resistant subline of the HCT 116 human colon-cancer cell line was isolated after repeated exposure of HCT 116 cells to increasing concentrations of MMC under aerobic conditions. The MMC-resistant subline (designated HCT 116-R30A) was 5 times more resistant than the parent cells to MMC and PFM under aerobic conditions. Both the MMC-resistant cells and the parent HCT 116 cells accumulated similar amounts of PFM by passive diffusion, but levels of macromolecule-bound PFM were about 50% lower in the resistant cell line, implying a decrease in PFM reductive activation in the resistant cells. The finding that microsomes from either sensitive or resistant cells showed an equal ability to reduce MMC and PFM indicated that the activity of NADPH cytochrome P-450 reductase (EC 1.6.2.4) was not changed in the resistant subline. Soluble extracts of HCT 116 cells reduced MMC and PFM more effectively at pH 6.1, and NADH and NADPH were utilized equally well as electron donors under both aerobic and anaerobic conditions. These data suggest that quinone reductase (EC 1.6.99.2; DT-diaphorase) in soluble extracts is responsible for the reduction of MMC. Quinone reductase activities in soluble extracts of HCT 116-R30A cells for the reduction of dichlorophenol indophenol (DCPIP) and menadione-cytochrome c at optimal pHs were decreased by 95% as compared with those obtained in parent cells. However, the MMC-reducing activity of HCT 116-R30A soluble extracts was only 50% lower than that of the parent cell extracts. The kinetic constants (Km, Vmax) found for quinone reductase in the two cell lines with respect to the substrates DCPIP and menadione differed. Two species of mRNA for quinone reductase (2.7 and 1.2 kb) were detected in both cell lines, and there was no detectable difference between parent and resistant cells in the steady-state level of either of these mRNA species. Furthermore, incubation with the quinone reductase inhibitor dicoumarol rendered HCT 116 cells more resistant to MMC. Alteration of the quinone reductase activity in HCT 116-R30A cells appears to be the mechanism responsible for their resistance to MMC and PFM.

A series of cyclic acetal derivatives of mitomycin C (MC) and porfiromycin (POR) were tested for their ability to kill hypoxic and oxygenated EMT6 tumor cells. Amino methyl acetal and thioacetal substitutions at C-7 of MC and POR dramatically increased the cytotoxicity of the compounds to hypoxic EMT6 tumor cells in vitro but had little effect on the aerobic toxicities. In contrast, a methyl substitution at N1a markedly decreased the aerobic cytotoxicities of the compounds but did not alter the hypoxic cytotoxicities. The POR acetal, BMY-42355, had the largest differential between hypoxic and aerobic cytotoxicities yet observed among MC analogs. Preliminary studies in mice showed that BMY-42355 had good antineoplastic activity when used alone or in combination with radiation and was less toxic than POR; the therapeutic ratio of this compound in these initial studies was higher than those of either MC or POR.

X-ray, NMR, and molecular mechanics studies on antitumor antibiotic porfiromycin (C16H20N4O5), a covalent binder of DNA, have been carried out to study the structure, conformation, and theoretical interactions with DNA. The crystal structure was solved by direct methods and refined to an R value of 0.052. The configurations at C(9), C(9a), C(1), and C(2) are S, R, S, and S, except for the orientation of the aziridine ring and (carbamoyloxy)methyl side chain. The five-membered ring attached to the aziridine ring adopts an envelope conformation. The solution conformation is similar to that observed in the solid state except for the (carbamoyloxy)methyl side chain. Monovalent and cross-linked models of the drug bound to DNA have been energetically refined by using molecular mechanics. The results indicate that, in the case of monocovalent binding, the drug clearly prefers a d(CpG) sequence rather than a d(GpC) sequence. In the case of the cross-linked model there is no clear-cut preference of d(CpG) over d(GpC), indicating that the binding preference of the drug may be kinetic rather than thermodynamic.

The cytotoxicity, metabolism, and DNA alkylation of porfiromycin (PFM) under aerobic and hypoxic conditions were evaluated in P388 murine leukemia cells. Clonogenic assays showed that the IC50 value for a 1-h exposure to PFM was 4 microM for aerobic cells and 0.5 microM for hypoxic cells. After a 1-h exposure to concentrations of 1, 5, and 10 microM [14C]-PFM, the accumulation of total radioactivity in hypoxic cells was 10 to 20 times that in aerobic cells. The disposition of radioactivity in cells that had been treated for 1 h with 5 microM PFM under aerobic or hypoxic conditions showed that (a) under either condition, internal free-PFM concentration equalled the external drug concentration; (b) DNA-, RNA-, and protein-bound radioactivity were at least 10 times greater in hypoxic cells than in aerobic cells; and (c) known metabolites and unidentified radioactive products were also generated in greater amounts in hypoxic cells than in aerobic cells. Thus, the increased amounts of radioactivity accumulated by hypoxic P388 cells after exposure to [14C]-PFM resulted from the accumulation of nonexchangeable protein and nucleic-acid adducts and metabolites rather than free PFM. Determinations of DNA adducts formed in P388 cells revealed five possible adducts: (1) N2-(2'-deoxyguanosyl)-7-methylaminomitosene, (2) a second monofunctional PFM-guanine adduct, (3) a PFM cross-linked dinucleotide, (4) possibly a nucleoprotein-related adduct, and (5) an unknown. We conclude that the enhancement of PFM-induced cytotoxicity by hypoxia appears to be primarily due to increased alkylation of macromolecules.

DNA adduct formation by enzyme-activated antibiotics, mitomycin C (MMC) or porfiromycin (PFM), at pH 7.6 or pH 6.0 under anaerobic conditions was analyzed by a 32P-postlabeling method. Antibiotic activation by rat liver NADPH-cytochrome P-450 reductase (EC 1.6.2.4) and bovine milk xanthine oxidase (EC 1.2.3.2) produced similar results. Five 32P-labeled MMC adducts were separated by thin layer chromatography and high performance liquid chromatography from DNA alkylated at either pH. Four of the radioactive spots separated by thin layer chromatography were identified as two monofunctional monoadducts [1" alpha and 1" beta forms of N2-(2" beta,7"-diaminomitosen-1"-yl)-2'-deoxyguanylic acid], one bifunctional monoadduct [N2-(10"-decarbamoyl-2",7"-diaminomitosen-1" alpha-yl)-2'-deoxyguanylic acid], and one cross-linked adduct [N2-(2" beta,7"-diamino-10"-deoxyguanyl-N2-yl-mitosen- 1" alpha-yl)-2'-deoxyguanylic acid]. One minor radioactive spot was not identified. By comparing DNA alkylated at the two pH values, based on equal amounts of 32P radioactivity, similar amounts of cross-links were detected. However, the DNA showed different ratios of the alpha and beta isomers of the monofunctional monoadduct. Furthermore, the DNA alkylated at pH 6.0 showed more bifunctional monoadducts than did the DNA alkylated at pH 7.6. Analysis of alkylated DNA by enzyme-activated PFM showed a similar spectrum of DNA adduct formation. The effect of pH on the distribution of the five PFM-DNA adducts was similar to that observed for the five MMC-DNA adducts. The distribution of adducts in DNA alkylated at the same pH was similar irrespective of which enzyme activated MMC or PFM. The pH of the reaction during DNA and MMC interaction was the determining factor for the quantitative distribution of the adducts. This pH effect may be important for the cytotoxicity of MMC and PFM in tumor cells that have high levels of reductive enzymes with low optimal pH values.

Laboratory studies and clinical trials are exploring the use of hypoxia-directed cytotoxic agents as adjuncts to radiotherapy. Because hypoxia and the microenvironmental inadequacies associated with hypoxia in solid tumours inhibit cell proliferation, an essential requirement for the successful use of hypoxia-directed drugs in cancer therapy is that these drugs be toxic to quiescent tumour cells, as well as tumour cells progressing rapidly through the cell cycle. The experiments reported here compared the cytotoxicities of mitomycin C and porfiromycin to exponentially growing and plateau phase cultures of EMT6 mouse mammary tumour cells. The proliferative status of the cultures did not influence the cytotoxicity of mitomycin C under either aerobic or hypoxic conditions, or the cytotoxicity of porfiromycin in air. Exponentially growing cultures were slightly more sensitive than plateau phase cultures to porfiromycin in hypoxia, but the difference between the sensitivities of proliferating and quiescent cells was much smaller than the difference between aerobic and hypoxic cells. No evidence for repair of potentially lethal damage was found after treatment with porfiromycin in air or in hypoxia; this is in agreement with previous findings for mitomycin C. Mitomycin C and porfiromycin therefore exhibit the toxicity to quiescent cells needed for effective use as hypoxia-directed drugs for the treatment of solid tumours.

Mitomycin C and porfiromycin were found to inactivate rat hepatic DT-diaphorase. Inactivation was pH dependent; little inactivation was detected at pH 5.8, but inactivation increased as the pH was raised to 7.8. Inactivation was concentration and time dependent and displayed pseudo-first-order kinetics. Inactivation was NADH dependent, indicating that reductive metabolism was necessary for inhibition. [3H]Mitomycin C was covalently bound to DT-diaphorase during inhibition, and the stoichiometry for inactivation of DT-diaphorase by mitomycin C was approximately 0.8 nmol of mitomycin C bound/nmol of enzyme. A higher molecular mass product (60 kDa) was detected by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and Western blot analysis of DT-diaphorase preincubated with NADH and mitomycin C at pH 7.8, suggesting that mitomycin C is capable of cross-linking DT-diaphorase. The kinetics of inhibition, requirement for NADH for inhibition, covalent binding of [3H] mitomycin C to DT-diaphorase, and approximate 1:1 stoichiometry suggest that this inactivation process may be mechanism based. Inhibition of DT-diaphorase by mitomycin C and porfiromycin is not limited to a cell-free system and could also be observed in HT-29 cells in culture at pH 7.2. Bioactivation of mitomycin C or porfiromycin by DT-diaphorase is favored at lower pH, whereas at higher pH values enzyme alkylation and inactivation of DT-diaphorase occur. These data suggest that the success of attempts to exploit the elevated DT-diaphorase content of certain human tumors for improved chemotherapeutic response using mitomycin C or porfiromycin will depend on intracellular pH.

[3H]-(N-la-methyl) Porfiromycin (POR) was employed to detect and identify the radiolabeled mono- and bis-adducts formed in living EMT6 mouse mammary tumor cells under different conditions. To provide authentic standard adducts, calf-thymus DNA was treated with POR under reductive activation, then digested to nucleosides and POR-nucleoside adducts. The three major adducts formed were isolated by HPLC and authenticated. Two were mono-adducts, composed of deoxyguanosine linked at its N2-position to C-1 of POR and of 10-decarbamoyl POR. The third was a bis-adduct, in which POR was crosslinked to two deoxyguanosines at their N2-positions. DNA from [3H]-POR treated EMT6 cells was digested an analyzed by HPLC. DNA-associated label was located in thymidine and in two mono-adducts and one bis-adduct identical to those described above. Label in thymidine resulted from N-demethylation of POR and reincorporation of label into new thymidylate residues. Adducts were formed more abundantly in hypoxia than in air. In addition, the mono-adduct to crosslink ratios were different, approximately 1:1 and 2:1 for hypoxic and aerobic cells, respectively. The different patterns of alkylation in air and hypoxia may be related to the greater toxicity of POR in hypoxia. When cells were treated simultaneously with POR and dicumarol, adduct levels were lower, and a new, unknown adduct was observed primarily under hypoxia; these changes may be related to the altered toxicity of POR in the presence of dicumarol. The HPLC assay detected simultaneously the full array of stable mono- and bis-adducts in DNA with good sensitivity (greater than or equal to 2 x 10(6) adducts/nucleotide) and excellent reproducibility. This assay should be generally applicable to all cells and tissues when MC or POR with high specific radioactivity can be employed.

The mechanism of aerobic resistance to the quinone-containing anti-tumour agents mitomycin C (MMC) and porfiromycin (PM) has been investigated using non-transformed human cells. One of the cell strains used (3437T) was derived from an afflicted member of a cancer-prone family. This cell strain had been shown previously to be six times more resistant to the cytotoxic effects of these agents under aerobic but not hypoxic conditions when compared to a cell strain derived from an unrelated, normal donor (GM38). Differences could not be detected in the ability of cell sonicates prepared from either cell strain to produce alkylating species under aerobic conditions using a 4-(p-nitrobenzyl)pyridine assay. However, using 3H-labelled PM to monitor rapid drug uptake and subsequent accumulation due to drug metabolism, results were obtained indicating that the resistant cell strain (3437T) was deficient in an enzymatic pathway capable of metabolizing these compounds under aerobic but not hypoxic conditions. Dicumarol, an inhibitor of the quinone reductase DT-diaphorase (EC 1.6.99.2), decreased aerobic drug accumulation and cytotoxicity in the control cell strain, but did not alter the lack of accumulation noted in the resistant cell strain. Under hypoxic conditions, dicumarol increased cytotoxicity and drug accumulation in both cell strains. The mechanism of this enhanced cytotoxicity remains unclear. These results suggested that the resistant cells were deficient in the enzyme DT-diaphorase, a potential activator of PM. Enzymatic assays confirmed this and revealed no alterations in cytochrome P450 reductase (EC 1.6.2.4) activity or glutathione content. No protein characteristic of DT-diaphorase was detected in the resistant cell strain using a polyclonal rabbit-anti-rat antibody raised against this enzyme. Southern blot analysis using a rat DT-diaphorase cDNA probe demonstrated differences between the normal and resistant cell strains in the restriction fragment patterns. The present results are consistent with the hypothesis that decreased DT-diaphorase levels are causally associated with PM and MMC resistance in these cells under aerobic exposure conditions.

The distribution of porfiromycin was studied in BALB/c mice bearing EMT6 mammary tumors. The levels of 3H in blood and most tissues peaked approximately 15 minutes after intraperitoneal injection of [3H]porfiromycin. The levels of radioactivity present in most of the tissues and in the tumors were similar at 4 hours and 24 hours after administration. Most of the normal tissues showed uniform, low grain densities when analyzed by autoradiography; the liver and the small intestine had the highest labeling densities. Only kidney, bladder, and tumor showed differential distributions of grains from [3H]porfiromycin. In the kidney, higher grain counts were found in cortex than in medullary regions; grains were uniformly distributed within each region. In the bladder, the highest labeling densities were found in regions near the lumen. Tumor regions that had some necrotic features or regions of necrosis that included some viable cells showed higher labeling intensities than healthy-looking tumor regions, probably because the abnormal microenvironments in these regions led to increased rates of activation of porfiromycin to electrophilic species. These findings show that porfiromycin can reach and be activated in tumor regions containing cells resistant to many chemotherapeutic agents and to x rays. The results also support the concept that agents such as porfiromycin can target cells in specific microenvironmental subpopulations of solid tumors.

A wild type Chinese hamster cell line (AA8) and three repair-deficient sublines of AA8 (EM9, UV4, and UV5) were used to study the nature of the cytotoxic lesions produced by the bioreductive alkylating agents mitomycin C and porfiromycin under aerobic and hypoxic conditions. The sensitivities of the repair-deficient sublines to the drugs varied markedly: EM9 was similar to AA8, whereas UV4 was exquisitely sensitive and UV5 was of intermediate sensitivity. Moreover, both the relative toxicities of the two drugs and the relative toxicities of each drug under aerobic and hypoxic conditions varied for the different cell lines. These data suggest that there are differences in the spectra of toxic lesions produced by mitomycin and porfiromycin and that there are differences in the lesions produced by these drugs under aerobic and hypoxic conditions.

In an effort to develop effective combination treatments for use with radiation against solid tumors, the cytotoxic effects of the addition of mitomycin C or porfiromycin on treatment with Fluosol-DA/carbogen (95% O2/5% CO2) breathing and radiation in the FSaIIC tumor system were studied. In vitro mitomycin C and porfiromycin were both preferentially cytotoxic toward hypoxic FSaIIC cells. After in vivo exposure, however, the cytotoxicity of mitomycin C toward single cell tumor suspensions obtained from whole tumors was exponential over the dose range studied, but for porfiromycin a plateau in cell killing was observed. With Fluosol-DA/carbogen breathing and single dose radiation, addition of either mitomycin C or porfiromycin increased the tumor cell kill achieved at 5 Gy by approximately 1.2 and 1.0 logs, respectively. Less effect was seen with addition of the drugs at the 10 and 15 Gy radiation doses. In tumor growth delay experiments, the addition of either mitomycin C or porfiromycin to Fluosol-DA/carbogen breathing and radiation resulted in primarily an additive increase in tumor growth delay. The survival of Hoechst 33342 dye-selected tumor cell subpopulations indicated that Fluosol-DA/carbogen breathing increased the cytotoxicity of radiation (10 Gy) more in the bright cell subpopulation (4-fold) than in the dim cell subpopulation (2-fold) resulting in an overall 4-fold sparing of the dim subpopulation. Mitomycin C and porfiromycin were both more toxic toward the dim cell subpopulations. Addition of mitomycin C or porfiromycin to Fluosol-DA/carbogen breathing and radiation (10 Gy) resulted in a primarily additive effect of the drugs and radiation killing in both tumor cell subpopulations. Thus, with mitomycin C/Fluosol-DA/carbogen and radiation there was a 2-fold sparing of dim cells and with porfiromycin in the combined treatment a 1.6-fold sparing of the dim cell population. Our results indicate that treatment strategies directed against both oxic and hypoxic tumor subpopulations can markedly increase the tumor cell kill achieved by radiation.

Porfiromycin was given to a group of patients with a variety of solid tumors. Of 114 patients admitted to the study, 103 yielded evaluable data. The following dosage schedules were used to determine the toxicity of porfiromycin when given in multiple doses by intravenous injection: 0.2 mg/kg x 5 days, 0.3 mg/kg x 5 days, 0.35 mg/kg x 5 days, 0.4 mg/kg x 5 days, 0.24 mg/kg x 10 days and 0.6 mg/kg weekly. Toxic effects noted were mainly leukopenia, thrombocytopenia, and, when injected paravenously, local tissue necrosis. Biological effects were noted at all dosage levels and were more severe at the higher dosages. The data suggest that profiromycin administered intravenously at a dose of 0.35 mg/kg daily for 5 days results in moderate hermatological toxicity and clinical evaluation in a Phase II study at this dosage level is indicated.

New mitomycin C and porfiromycin analogues were prepared by treating mitomycin A and N-methylmitomycin A with a variety of amines, including aziridines, allylamines, propargylamines, chloroalkylamines, hydroxyalkylamines, glycine derivatives, aralkylamines, and heterocyclic amines. All analogues were evaluated against P-388 murine leukemia and selected ones were examined for their leukopenic properties. Certain analogues were found to be superior to mitomycin C in potency, efficacy, and therapeutic ratio in the P-388 assay. The most active substituents at the mitosane 7 position included aziridine, 2-methylaziridine, propargylamine, furfurylamine, methyl glycinate, and 3-aminopyridine. Mitomycin A and the 7-aziridino, 7-(2-methylaziridino), and 3-aminopyridine analogues were less leukopenic than mitomycin C. Certain other analogues, including propargylamino and methyl glycinate, were highly leukopenic. The three compounds tested against B-16 melanoma in mice were significantly more effective than mitomycin C in this assay. Previously established structure--activity relationships were found inadequate to account for all of the new data.

The effects of dicoumarol (DIC) on the cytotoxicity of porfiromycin (POR) were studied in vitro using EMT6 mammary tumor cells in monolayer cultures and in vivo using solid EMT6 tumors and bone marrow stem cells. In vitro, POR was more toxic to hypoxic EMT6 cells than to aerobic cells. Exposure of aerobic cultures to DIC protected against POR; in contrast, DIC sensitized hypoxic cells to POR. Treatment of mice with DIC produced a slight increase in the toxicity of POR to cells in solid tumors. The toxicity of POR to marrow stem cells (CFU-GM and CFU-MK) was not altered by DIC. Pretreatment of mice with DIC therefore produced a small improvement in the therapeutic ratio.

Mitomycin C and its methylated analogue porfiromycin (Por) have significant potential as adjuncts to regimens presently used for treating solid tumors because of their preferential toxicity to cells existing in an hypoxic environment. An understanding of the factors producing the differential activity of these drugs under aerobic and hypoxic conditions would facilitate the development of new agents of this class. Previous studies have focused on the enzymes that reductively activate the mitomycins and on the interaction of these drugs with DNA; none of these studies has fully explained the differences in cytotoxicity observed under hypoxic and aerobic conditions. The present investigation demonstrates that the rate of Por uptake is directly correlated with cytotoxicity under both aerobic and hypoxic conditions. Uptake of Por into hypoxic cells is more rapid than into aerobic cells at equal drug concentrations. Hypoxic cells also accumulate drug in concentrations well in excess of those in the extracellular medium; this is apparently a reflection of drug sequestration in these cells. This sequestration of Por, which affects the rate and extent of uptake in hypoxic cells, does not take place in aerobic cells. The failure of aerobic cells to sequester drug is evidenced by the very rapid efflux of Por from these cells upon removal of extracellular Por and by the fact that aerobic cells attain a state of equilibrium between the intracellular and extracellular drug concentrations. The findings demonstrate that differences in the uptake and retention of Por are associated with the preferential toxicity of Por to hypoxic cells.

Porifiromycin has been tested in many groups of patients over the past several years (1-7). This report is an analysis of greater than 100 patients with epidermoid and urinary tract transitional cell carcinomas treated with this agent. A review of these data and the available literature on this agent shows that porfiromycin offers definite usefulness in disseminated squamous cell carcinomas of the cervix, and definite but lesser effectiveness in epidermoid carcinomas of the lung, the head and neck region, and other sites. Responsiveness was also demonstrated in transitional cell carcinomas of the urinary tract. Toxicity was tolerable and consisted primarily of myelosuppression and local skin necrosis in sites where extravasation had occurred.

The mechanism of uptake and efflux of porfiromycin (PFM) by HCT 116 human colon carcinoma cells or freshly obtained human RBC was investigated. The time course of uptake of radioactivity upon exposure of HCT 116 cells to [14C]PFM showed one fast and one slow phase of linear increase. The initial phase of PFM uptake was not saturable with external drug concentrations from 2 to 100 microM. PFM accumulation was temperature dependent with a temperature coefficient (Q10 24-37 degrees C) of 2.3 +/- 0.3. PFM uptake was not affected either by individual inhibitors such as 1 mM 2,4-dinitrophenol, sodium azide, iodoacetic acid, ouabain, 0.02 mM oligomycin, p-hydroxylmercuribenzoate, 0.2 mM N-ethylmaleimide, or by combinations of inhibitors. PFM uptake did not demonstrate competitive inhibition by unlabeled PFM and mitomycin C. Efflux of cellular radioactivity was not affected by the above mentioned inhibitors or by verapamil, diltiazem, or trifluoperazine. Only aliphatic alcohols accelerated the initial influx rate. The RBC, however, only exhibited the initial fast accumulation of [14C]PFM, and all the 14C accumulated by RBC was exchangeable. These data demonstrate that the uptake and the efflux of PFM in HCT 116 cells and RBC comprise a passive diffusion process.

Aspects of the degradations of mitomycin and porfiromycin were studied. The initial degradation processes of the compounds in an acidic medium were investigated. Influences of pH, buffers, and other additives such as halogenides and dioctyl sodium sulfosuccinate [sodium 1,4-bis(2-ethylhexyl)sulfosuccinate] were studied. The hydrogen ion catalyzes the degradation of both the uncharged and the protonated species. Anions also promote the degradation of the compounds in an acidic medium. Rate constants for all of the catalytic reactions could be determined. From the pH profiles, after correction for buffer influences, accurate pKa values for the aziridine nitrogens could be obtained. The protective influence of the dioctyl sulfosuccinate ion could be explained. From the data obtained a plausible mechanism for the initial acidic degradation reactions was developed.

Porfiromycin was reductively metabolized by NADPH cytochrome P-450 reductase and xanthine oxidase under anaerobic conditions. The production of metabolites varied with the pH and the contents of the reaction buffer. In Tris buffer, two major metabolites were produced at pH 7.5 and above, whereas one major metabolite was produced at pH 6.5. The three major metabolites were separated and isolated by HPLC. Identification by californium-252 plasma desorption mass spectrometry showed that the two major metabolites from pH 7.5 were (trans) and (cis)-forms of 7-amino-1-hydroxyl-2-methylaminomitosene and the major metabolite from pH 6.5 was 7-amino-2-methylaminomitosene. All three major metabolites showed substitutions at the C-1 position. DNA was alkylated readily by enzyme-activated porfiromycin. Digestion of porfiromycin-alkylated DNA by DNase, snake venom phosphodiesterase, and alkaline phosphatase resulted in an insoluble nuclease-resistant fraction and a soluble fraction. The nuclease-resistant fraction reflected a high content of cross-linked adducts. Upon HPLC analysis, the solubilized fraction contained two monofunctionally linked porfiromycin adducts and a possibly cross-linked dinucleotide. The major adduct was isolated by HPLC and identified by NMR, as N2-(2'-deoxyguanosyl)-7-amino-2-methylaminomitosene. The N2 position of deoxyguanosine appeared as the major monofunctional alkylating site for DNA alkylation by porfiromycin. Thus, mitomycin C and porfiromycin (which differs from mitomycin C only by the addition of a methyl group to the aziridine nitrogen) share the same enzymatic activating mechanism that leads to the formation of the same types of metabolites and the same specificity of DNA alkylation.

Mitomycin C (1) is the prototypical bioreductive alkylating agent. Studies have shown that mitomycin C and its derivatives selectively alkylate guanine residues within di- and trinucleotide DNA sequences. This investigation sought to improve the selective DNA bonding properties of the mitomycins by coupling them with antisense oligodeoxynucleotides. Two procedures were developed that allowed the attachment of a phosphorothioate oligodeoxynucleotide containing a hexylamino spacer at the 5' terminus with a C(10)-activated mitomycin. In the first procedure, decarbamoylation of 1 (NaOCH3/ benzene) afforded 10-decarbamoylmitomycin C (10), which was treated with either dimethyl sulfate or methylthiochloroformate and base to yield 10-decarbamoylporfiromycin (11) and N(1a)-[(methylthio)-carbonyl]-10-decarbamoylmitomycin C (12), respectively. Activation of the C(10) site in 11 and 12 with 1,1'-carbonyldiimidazole or with 1,1'-thiocarbonyldiimidazole provided the N(1a)-substituted mitomycin 10-decarbamoyl-10-O-carbonylimidazoles (5, 7) and 10-decarbamoyl-10-O-thiocarbonylimidazoles (6, 8), respectively. Compounds 5-8 were reacted with glycine methyl ester hydrochloride (17) and base in both methylene chloride and aqueous buffered solutions to determine the ease and efficiency in which these C(10)-activated mitomycin derivatives coupled to amines. It was found that 5-8 all reacted with 17 in methylene chloride to give the coupled products 18-21 but that improved amine coupling yields in water were observed for the 10-decarbamoyl-10-O-thiocarbonylimidazoles 6 and 8 as compared with the 10-decarbamoyl-10-O-carbonylimidazoles 5 and 7. This finding led to the coupling of the phosphorothioate oligodeoxynucleotide, H2N(CH2)6-P(S)(OH)-GGCCCCGTG-GTGGCTCCAT (22) to 8. Compound 22 complemented a 19-base sequence in the translation initiation region of the human A-raf-1 gene. Use of excess 8 (28 equiv) with 22 gave only a 36% yield of the coupled product 23, which proved difficult to separate from 22. In the second procedure, phosphorothioate oligodexynucleotides that contained a hexylamino spacer at the 5'termini were coupled to 10-des(carbamoyloxy)-10-isothiocyanatoporfiromycin (9). Compound 9 was prepared in four steps from 11. Mesylation (methanesulfonyl chloride/pyridine) of 11 gave the C(10) mesylate 13, which was then treated with NaN3 (dimethylformamide, 90 degrees C) to give 10-des(carbamoyloxy)-10-azidoporfiromycin (14). Catalytic reduction (PtO2, H2) of 14 in pyridine afforded C(10) amine 15. Treatment of 15 with di-2-pyridyl thionocarbonate provided the desired 10-des(carbamoyloxy)-10-isothiocyanatoporfiromycin (9). Compound 9 readily coupled with 17 and base in both methylene chloride and aqueous buffered solutions to give 25. Use of the 5'hexylaminophosphorothioate oligodeoxynucleotides 32-35 in place of 17 gave the conjugated adducts 28-31, respectively, in a 12% to near-quantitative yield. The products were purified by semipreparative HPLC. Antisense agents 28-31 were designed to target a 30-base-long region from the coding region of the human FGFR1 gene. One adduct, 29, reduced the number of FGFR1 receptors in human aortic smooth cells for bFGF on the cell surface, which suggested down-regulation of FGFR1 gene expression. Further, 29 inhibited cultured human aortic smooth muscle cell proliferation and was less cytotoxic than porfiromycin (2). The biological assay data suggest that the phosphorothioate oligodexynucleotide porfiromycin conjugates may be more target selective and less toxic than either mitomycin or porfiromycin and thus be promising therapeutic agents.

Cyanide and phenethyl alcohol greatly enhance the lethal action of mitomycin C, porfiromycin, and streptonigrin on an exponentially growing culture of Escherichia coli. Dinitrophenol similarly enhances the lethal action of mitomycin C and porfiromycin, but only slightly that of streptonigrin. Phenethyl alcohol may be functioning in these experiments as an inhibitor of electron transport.

Mitomycins and porfiromycin, generally nonreactive in the natural oxidized state, behave as bifunctional "alkylating" agents upon chemical or enzymatic reduction, followed by spontaneous loss of the tertiary methoxy (hydroxyl) group and formation of an aromatic indole system. Thus activated, mitomycins and porfiromycin react in vitro with purified DNA, linking its complementary strands. A high content of guanine and cytosine favors this cross-linking reaction, which is the basis of the lethal effect in vivo of these antibiotics. The activation and cross-linking reactions are discussed in terms of reactive sites on the mitomycin and DNA molecules.

High-performance liquid chromatography (HPLC) and thermospray mass spectrometry were combined for the analysis of DNA adducts formed from the interaction of the anticancer drugs mitomycin C, porfiromycin and thiotepa with calf thymus DNA. The adducts formed from reaction of mitomycin C and porfiromycin with DNA were separated from unmodified nucleosides by HPLC on a C18 column and identified by thermospray mass spectrometry. Thiotepa DNA adducts readily depurinated from DNA and were chromatographed and identified by thermospray liquid chromatography-mass spectrometry as the modified bases without the ribose moiety attached. The utility of thermospray mass spectrometry for the identification of microgram quantities of nucleoside adducts and depurinated base adducts of these anticancer drugs was demonstrated.

The bioreductive alkylating agent porfiromycin (POR) is more toxic to EMT6 cells that are hypoxic at the time of treatment than to aerobic cells. The toxicity of POR to hypoxic EMT6 cells in vitro was similar to that of mitomycin C (MC): the aerobic toxicity of POR was considerably less than that of MC. Treatment of cells in vitro with POR before and during irradiation did not sensitize either hypoxic or aerobic cells to X rays; instead, only additive cytotoxicity was produced. In contrast, treatment of solid EMT6 tumors in vivo with POR plus radiation produced supra-additive cytotoxicity, as assessed by analyses of the complete dose-response curves for the killing of tumor cells by radiation alone or by POR alone. The supra-additivity of the combination regimens appeared to reflect the preferential killing by each agent of those tumor cells which were in an environment conferring resistance to the other agent. In contrast, combinations of POR and X rays produced only additive cytotoxicities to marrow CFU-GM. Supra-additive antineoplastic effects were obtained at doses of POR which produced little hematologic or other host toxicity. The complementary cytotoxicities of radiation and POR to cells in different microenvironments in solid tumors and the absence of a similar effect in normal tissue make optimized regimens combining radiotherapy and POR unusually promising for the treatment of solid tumors.

Solid neoplasms may contain deficient or poorly functional vascular beds, a property that leads to the formation of hypoxic tumor cells, which form a therapeutically resistant cell population within the tumor that is difficult to eradicate by ionizing irradiation and most existing chemotherapeutic agents. As an approach to the therapeutic attack of hypoxic cells, we have measured the cytotoxicity and DNA lesions produced by the bioreductive alkylating agents mitomycin C and porfiromycin, two structurally similar antibiotics, in oxygen-deficient and aerobic cells. Mitomycin C and porfiromycin were preferentially cytotoxic to hypoxic EMT6 cells in culture, with porfiromycin producing a greater differential kill of hypoxic EMT6 cells relative to their oxygenated counterparts than did mitomycin C. Chinese hamster ovary cells were more resistant to these quinone antibiotics; although in this cell line, porfiromycin was significantly more cytotoxic to hypoxic cells than to aerobic cells, and the degree of oxygenation did not affect the toxicity of mitomycin C. Alkaline elution methodology was utilized to study the formation of DNA single-strand breaks and DNA interstrand cross-links produced by mitomycin C and porfiromycin in both EMT6 and Chinese hamster ovary cells. A negligible quantity of DNA single-strand breaks and DNA interstrand cross-links were produced in hypoxic and aerobic Chinese hamster ovary cells by exposure to mitomycin C or porfiromycin, a finding consistent with the considerably lower sensitivity of this cell line to these agents. In EMT6 tumor cells, no single-strand breaks appeared to be produced by these antitumor antibiotics under both hypoxic and aerobic conditions; however, a significant number of DNA interstrand cross-links were formed in this cell line following drug treatment, with substantially more DNA interstrand cross-linking being produced under hypoxic conditions. Mitomycin C and porfiromycin caused the same amount of cross-linking under conditions of oxygen deficiency; however, mitomycin C produced considerably more DNA cross-linking than did porfiromycin in oxygenated cells. DNA interstrand cross-links were observed in hypoxic EMT6 cells throughout a 24-h period following removal of mitomycin C and porfiromycin, with a decrease in DNA interstrand cross-links observed at 24 h. An increase in DNA interstrand cross-links occurred in aerobic EMT6 cells treated with mitomycin C and porfiromycin at 6 h after drug removal, with a decrease in these lesions being observed by 24 h, suggesting that the rate of formation of the cross-links may be slower and the removal of cross-links more rapid under aerobic conditions.(ABSTRACT TRUNCATED AT 400 WORDS)

Some factors affecting the cytotoxicity of porfiromycin (PM), an analogue of mitomycin C (MMC), were investigated in suspension cultures of wild-type (AA8-4) and repair-deficient (UV-20) Chinese hamster ovary cells. Oxygen was an important modulator of PM toxicity in AA8-4 cells. The aerobic toxicity was significantly less, and toxicity under extremely hypoxic conditions was significantly greater for PM than MMC. Porfiromycin cytotoxicity at intermediate O2 levels was similar to that observed previously for MMC. While the aerobic/hypoxic ratio was greater for PM than MMC, survival at intermediate oxygen concentrations could limit the therapeutic utility of these drugs as adjuncts to radiotherapy. Ascorbic acid was found to increase the aerobic, but not hypoxic, cytotoxicity of PM in AA8-4 cells, as was observed previously for MMC. Investigation of various exposure times and drug concentrations revealed that drug toxicity for both aerobic and hypoxic cells was dependent on the product of drug concentration and time, and that the aerobic/hypoxic differential observed in AA8-4 cells was constant over a broad range of exposure conditions. The sensitivity of UV-20 cells was also a linear function of concentration and time, but no aerobic/hypoxic differential was observed in these cells. It is suggested that the sensitivity of UV-20 to PM and MMC, and its lack of an hypoxic/aerobic differential could result from lethality being due to a different lesion than in wild-type cells.

Hypoxic cells may limit the curability of solid tumors by conventional chemotherapeutic agents and radiotherapy. Agents which are preferentially toxic to cells with low oxygen contents could therefore be useful as adjuncts to the regimens now used to treat these cancers. To date, the best agent of this type that we have tested is porfiromycin. Porfiromycin is similar to mitomycin C in its toxicity to hypoxic EMT6 cells in vitro but has much less toxicity than mitomycin C to well-oxygenated EMT6 cells. EMT6 cell sonicates reduce mitomycin C and porfiromycin to reactive electrophiles at similar rates under hypoxic conditions, a finding that correlates with cytotoxicity, whereas the rate of production of reactive species from both drugs is very slow under aerobic conditions. We also show that porfiromycin is capable of killing hypoxic radiation-resistant cells in solid EMT6 tumors. Appropriate regimens combining porfiromycin (which preferentially kills hypoxic cells) and radiation (which preferentially kills aerated cells) may therefore be especially efficacious for the treatment of solid tumors.