A facile method of purification of neocarzinostatin, an antitumor protein.

A three-step column chromatographic method for the purification of neocarzinostatin (NCS) from a crude preparation was described. The purified material was homogeneous by acrylamide gel electrophoresis, isoelectric focusing, amino terminal analysis, and immunologic criteria. Purified NCS was 40 times as active in the inhibition of growth of Sarcina lutea and twice as active against CCRF-CEM human leukemia cells in vitro as was the starting material. When assayed against P388 and L1210 mouse leukemias in vivo, the purified material showed a median increase in life-span of 119 and 72%, respectively.

Radioimmunoassay of neocarzinostatin, an antitumor protein.

Antibodies directed toward the antitumor protein neocarzinostatin (NCS) have been produced in a rabbit by immunization with a highly purified NCS preparation. The antiserum was monospecific and reversed the antibacterial activity of NCS against Sarcina lutea. It cross-reacted with chemically modified derivatives of NCS and mitomalcin but failed to cross-react with macromomycin. A radioimmunoassay procedure has been developed utilizing the antiserum and a biologically active 125I-labeled derivative of NCS. The lower limit of detection by this radioimmunoassay, which involves a double antibody technique for the separation of antibody-bound and free antigen, was 1 X 10(-13) mole. The sensitivity of the assay is such that serum levels of NCS can be determined accurately after administration of the drug to rats at a single dose of 2 mg/kg. Since NCS is now undergoing clinical trial, the radioimmunoassay of the drug will be a valuable tool in clinical pharmacological studies.

Auromomycin-induced DNA damage and repair in human leukemic lymphoblasts (CCRF-CEM cells).

An alkaline elution procedure was used to study the nature of DNA damage induced by auromomycin, an antitumor protein, in human leukemic lymphoblasts (CCRF-CEM cells). The filter elution of drug-treated cells at pH 12.2 and 9.6 showed induction of both single and double strand DNA breaks. The DNA strand scission activities were linear in relation to drug concentration. The frequency of single strand breaks was higher than that of the double strand breaks. Protein-associated DNA single strand breaks were also detected in alkaline elution of drug-treated cells when a proteinase K digestion step was included in the assay protocol. The auromomycin-induced single strand breaks were repaired to almost completion within 8 h of postincubation of DNA-damaged cells whereas the repair of double strand breaks was not detected.

Calcium, calmodulin, and protein content of adriamycin-resistant and -sensitive murine leukemic cells.

Anticalmodulin and calcium channel blockers have been shown to reverse Adriamycin resistance by reducing the drug efflux from resistant cells. Since cellular calcium and calmodulin levels are probably related to these effects, we have measured the total, membrane-bound, and intracellular calcium levels in Adriamycin-resistant (P388/R) and -sensitive (P388/S) leukemia cells. In P388/R cells, total calcium was approximately 1.4-fold higher than that of P388/S cells. Membrane-bound and intracellular calcium levels were also higher in P388/R cells. No major difference was observed in the calmodulin content of these cells. The P388/R cells had a higher (approximately 1.4-fold) protein content. When calculated on the basis of per unit protein, P388/S and P388/R cells had similar total calcium but a higher intracellular free calcium and calmodulin content in P388/S cells. Thus our studies suggested that the lower drug efflux and increased drug retention in P388/R cells may not be related to calcium and calmodulin levels but may be due to some other membrane-related factors.

In vitro inhibition of human leukemic cells (CCRF-CEM) by agarose-immobilized neocarzinostatin.

Neocarzinostatin (NCS) is an acidic protein (molecular weight, 10,700) isolated from Streptomyces carzinostaticus that has antitumor activity both in model rodent systems and in humans. In vitro it inhibits the growth of a human lymphoblastic leukemic cell line (CCRF-CEM) at a very low concentration (the amount of drug that causes a 50% inhibition of growth compared to control cultures as extrapolated from a dose-response curve (ID50), 2.4 X 10(-9) M). We covalently coupled NCS to the N-hydroxysuccinimide ester of agarose and obtained a product that, by a variety of biochemical and immunological criteria, has been demonstrated to be devoid of any free or loosely bound NCS. Agarose-bound NCS, which is unable to enter cells because of its size, retains a significant amount of inhibitory activity (ID50, 6 to 15 X 10(-9) M) and is also capable of inhibiting tritiated deoxythymidine incorporation into CCRF-CEM cells. Since agarose-bound NCS cannot enter mammalian cells, the above findings indicate that NCS is able to exert its toxic effects by binding to or reacting with receptors on the cell membrane.

Experimental pharmacology of auromomycin in L1210 tumor cells in vitro and in vivo.

The antibiotic protein auromomycin (AUR) is composed of a chromophore and an apoprotein. The in vitro cytotoxic activity of AUR is associated with the chromophore, whereas the apoprotein is not cytotoxic. Exposure to AUR blocks cell cycle traverse in G2-M phase. Both AUR and the chromophore exhibit antitumor activity against murine L1210 leukemia, both in vitro and in vivo. Similarly, neocarzinostatin and its chromophore also show antitumor activity against L1210 tumor cells in vitro and in vivo. AUR and neocarzinostatin apoproteins have no antitumor activity. The pharmacokinetics of 125I-labeled AUR in L1210 tumor-bearing mice shows a triexponential drug decay and initial rapid clearance of drug from the blood. The distribution of AUR-associated radioactivity into various tissues is rapid, and at the end of 24 hr, there is little drug accumulation in the tissues. During the first hr, greater than 50% of the administered drug is eliminated in the urine. As the chromophores alone possess antitumor activity, the effect of these antitumor proteins in cancer chemotherapy might be significantly increased by the administration of protein preparations containing biologically active and stable chromophores.

Effects of neocarzinostatin on mammalian nuclei: release of nucleosomes.

When Chinese hamster cell nuclei (Line CHO) were reacted with neocarzinostatin and its DNA was analyzed on non-denaturing agarose gel electrophoresis, a series of bands with a multiplicity of 175 base pairs was obtained. A similar result was also obtained when the DNA samples were electrophoresed under denaturing gels. Our results suggest that the linker DNA between nucleosomes is the susceptible site to the drug and that neocarzinostatin can be used to study the chromatin structure.

Proteasome participates in the alteration of signal transduction in T and B lymphocytes following trauma-hemorrhage.

Proteasomes are essential components of the cellular protein degradation machinery. They are nonlysosomal and their participation is critical for (1) the removal of short lived proteins involved in metabolic regulation and cell proliferation, (2) the control of the activities of regulators involved in gene transcription, such as nuclear factor-kappa B (NF-kappa B) and signal transducer and activator of transcription (STAT1), and (3) processing of antigenic peptides for MHC class I presentation. Trauma-hemorrhage induces profound immunosuppression which is characterized by reduced splenocyte proliferation, interleukin (IL)-2 and interferon (IFN)-gamma productive capacity, increased activation of transcription factors NF-kappa B and STAT1 in splenic T lymphocytes, reduced macrophage antigen presentation capacity and inordinate release of proinflammatory cytokines, such as IL-6 and tumor necrosis factor-alpha. Furthermore, it appears that the activity of several regulatory proteins involved in immune function is altered by trauma-hemorrhage. Since proteasomes are involved in regulation and removal of regulatory proteins, we hypothesized that trauma-hemorrhage alters proteasomal activity in splenic lymphocytes. The data showed that activities of 26s proteasome from CD3+CD4+ and CD3+CD8+ splenic T lymphocytes were enhanced following trauma-hemorrhage which was associated with increased expression of NF-kappa B and STAT1. On the other hand, trauma-hemorrhage attenuated the activity of 26s proteasome from splenic B lymphocytes which was restored upon IFN-gamma stimulation and correlated with increased expression of NF-kappa B. These studies indicate a potential role for proteasomes in the regulation of signal transduction in splenic T and B lymphocytes following trauma-hemorrhage, and also suggest them as potential therapeutic targets for attenuation of immune suppression associated with this form of injury.

Structural study of crystalline auromomycin. Preliminary X-ray diffraction study of auromomycin and macromomycin.

Auromomycin and macromomycin from the organism Streptomyces macromomyceticus have been crystallized. The X-ray diffraction pattern of crystals of each molecule is consistent with space group P2(1)2(1)2 with cell parameters a = 46.45 A, b = 54.34 A and c = 42.03 A for auromomycin, and a = 46.45 A, b = 54.52 A and c = 41.54 A for macromomycin. Diffraction analysis of auromomycin is in progress.

Thermal injury alters macrophage responses to prostaglandin E2: contribution to the enhancement of inducible nitric oxide synthase activity.

Prostaglandin E2 (PGE2) and macrophage (Mphi)-derived reactive nitrogen intermediates (RNI) have been implicated in T cell dysfunction after thermal injury. Normally, Mphi inducible nitric oxide synthase (iNOS) activity can be regulated by PGE2, however, it is unknown whether PGE2 modulates Mphi iNOS activity after thermal injury. Splenic Mphi isolated from mice 7 days after thermal injury produced higher levels of RNI than Mphi from sham mice when stimulated with lipopolysaccharide (LPS) or interferon-gamma (IFN-gamma) and tumor necrosis factor alpha (TNF-alpha) in combination. PGE2, when added concurrently with LPS, suppressed RNI production by Mphi from sham mice, whereas Mphi from injured mice were unaffected. When Mphi were pretreated with PGE2 before LPS, RNI production was suppressed in both populations. RNI production in response to IFN-gamma or IFN-gamma and TNF-alpha in combination was enhanced by PGE2 in both populations, however, the effect was markedly greater in Mphi from injured mice. The PGE2-mediated changes in RNI production were paralleled by similar changes in iNOS protein expression, suggesting that the effect of PGE2 was at the level of enzyme expression rather than activity. Dibutryl cAMP induced similar effects as PGE2, suggesting the response to PGE2 after thermal injury is independent of potential changes in PGE2-induced adenylate cyclase activity and is cAMP-mediated. The results indicate that Mphi from burned mice display an altered sensitivity to PGE2, resulting in enhanced iNOS activity. Thus, PGE2, which is elevated after thermal injury and can directly suppress T cell function, may also contribute to immune dysfunction through the enhancement of Mphi iNOS activity.

Divergent immune responses in male and female mice after trauma-hemorrhage: dimorphic alterations in T lymphocyte steroidogenic enzyme activities.

Immune responses are suppressed in males, but not in proestrous females, after trauma-hemorrhage. Testosterone and 17beta-estradiol appear to be responsible for divergent immune effects. There is considerable evidence to suggest sex steroid hormone involvement in immune functions. As formation of active steroid depends on the activity of androgen- and estrogen-synthesizing enzymes, expression and activity of 5alpha-reductase, aromatase, and 3beta- and 17beta- hydroxysteroid dehydrogenases were determined in spleen and T lymphocytes of male and proestrous female mice after trauma-hemorrhage. All of the enzymes were present in spleen, specifically in T lymphocytes. 5alpha-Reductase expression and activity increased in male T lymphocytes, whereas aromatase activity, but not expression, increased in female T lymphocytes. Increased 5alpha-reductase activity in male T lymphocytes is immunosuppressive because of increased 5alpha-dihydrotestosterone synthesis, whereas in females increased aromatase activity triggering 17beta-estradiol synthesis is immunoprotective. This study also demonstrates the importance of 17beta-hydroxysteroid dehydrogenase oxidative and reductive functions. The immunoprotection of proestrous females is associated with enhanced reductase function of the enzyme. In males, decreased expression of oxidative isomer type IV, which impairs catabolism of 5alpha-dihydrotestosterone, probably augments immunosuppression. This study provides evidence for the involvement of intracrine sex steroid synthesis in gender dimorphic immune responses after trauma-hemorrhage.

Gender dimorphism in immune responses following trauma and hemorrhage.

Previous studies have shown marked immunosuppression in males after trauma and hemorrhage. By contrast, immunosuppression was not evident in proestrous female animals. Further support for these findings came from studies demonstrating the immunosuppressive effects of male sex hormones and immunoenhancing effects of female sex hormones. In this regard, administration of 17beta-estradiol or prolactin to male animals after trauma and hemorrhage significantly improved immune function. Furthermore, castration or androgen receptor blockade with flutamide after trauma and hemorrhage in male mice showed similar beneficial effects. Thus, regulation of sex hormone synthesis or their receptor activity appears to be a useful therapeutic approach for patients to cope with the deleterious sequelae of severe trauma and hemorrhage.

Anthracycline-induced DNA breaks and resealing in doxorubicin-resistant murine leukemic P388 cells.

Energy-dependent drug efflux is believed to be a major factor in cellular resistance to doxorubicin (DOX). However, recent studies have shown that decreased retention alone cannot account for anthracycline resistance, and possibly other factors, such as drug metabolism, free radical scavengers, and altered DNA damage/repair, may be involved. We have measured DOX-induced DNA damage and its repair in P388 cells sensitive (P388/S) and resistant (P388/R) to DOX. Our studies show 2- to 5-fold less DNA damage, measured as protein-associated single-strand DNA breaks, in P388/R cells when compared to similarly treated P388/S cells. The repair of DNA in whole cells, expressed as percent DNA rejoined, was complete in 4 hr in P388/R, whereas no repair was seen in P388/S cells until 20 hr. No difference in repair of DNA lesions was observed when nuclei were used in repair experiments. The absence of repair in sensitive whole cells may be due to high retention or slow drug efflux. Increase of cellular DOX retention by exposure of cells to trifluoperazine (TFP) or verapamil (VPL) did not result in the increase of DNA damage in P388/R cells. DOX analogs, N-trifluoroacetyladriamycin-14-valerate (AD 32), 4'-O-tetrahydropyranyladriamycin (THP-adriamycin), and N-benzyladriamycin-14-valerate (AD 198), induced 2- to 4-fold more DNA damage than DOX in resistant cells. There was no difference in the poly(ADP-ribose) synthesis of P388/S and P388/R cells exposed to DOX or AD 32. Since ADP-ribose polymer synthesis is associated with free radical-induced DNA damage and is indicative of DNA repair by an excision-repair mechanism, data from these studies suggest that DNA breaks in anthracycline-exposed cells may not be due to free radical production but rather to other mechanisms, such as inhibition of DNA topoisomerase II activity. The present studies, in addition to emphasizing the role of DNA damage in resistance, also underscore the relative importance of DNA topoisomerase II function in anthracycline cytotoxicity.

Anthracycline resistance in murine leukemic P388 cells. Role of drug efflux and glutathione related enzymes.

Energy-dependent drug efflux is a major factor in cellular resistance of P388/R84 mouse leukemic cells to anthracyclines such as doxorubicin (DOX), and blocking of efflux increases sensitivity. However, efflux does not play a significant role in resistance to N-trifluoroacetyladriamycin-14-valerate (AD 32), a DOX analog. Since drug efflux alone cannot account for resistance to anthracyclines, we have, in the present study, measured cellular glutathione (GSH) content and activity of GSH cycle related enzymes to determine their role in resistance. Cellular GSH content was similar in DOX-sensitive and -resistant mouse leukemic cells (P388 and P388/R84). GSH peroxidase, glucose-6-phosphate dehydrogenase and glutathione reductase activities were 1.36-, 1.58- and 1.14-fold higher in P388/R84 cells. Incubation of P388/R84 cells with 100 microM buthionine-S,R-sulfoximine (BSO) for 24 hr reduced cellular GSH content to 6% of control and reduced their resistance to DOX [dose modification factor (DMF) 3.9]. GSH depletion had no significant effect on the cytotoxicity of AD 32 (DMF 1.5). Exposure of P388/R84 cells to BSO (for GSH depletion) and trifluoperazine (for efflux blocking) further reduced their resistance to DOX (DMF 14). These results indicate that DOX resistance in P388/R84 cells is multifactorial and that changes in GSH cycle related enzymes such as GSH peroxidase may also contribute to their resistance.

Doxorubicin-induced DNA breaks, topoisomerase II activity and gene expression in human melanoma cells.

We have analyzed five human melanoma cell lines, displaying variable doxorubicin resistance (1- to 6-fold), for drug-induced DNA breaks, topoisomerase II activity and mRNA expression. Enhanced drug efflux was not the reason for doxorubicin resistance of these tumor cells although they overexpressed the transmembrane 170 kDa P-glycoprotein. Doxorubicin-induced DNA lesions (2-fold) and topoisomerase II activity (7-fold) were higher in HM-1 and G361 cells than in the less doxorubicin-sensitive NH and FCCM-9 cells. Topoisomerase II mRNA expression was also 2-fold higher in HM-1 and G361 cells. Doxorubicin-induced DNA breaks and topoisomerase II activity inversely correlated with the degree of doxorubicin sensitivity. Southern blot analysis showed variation in the hybridization pattern of topoisomerase II gene in doxorubicin-resistant cells when compared to sensitive cells. This study portrays the low doxorubicin sensitivity of NH and FCCM-9 cells as "atypical" and emphasizes the importance of DNA damage and topoisomerase II activity in cellular low doxorubicin resistance.

Trauma-hemorrhage activates signal transduction pathways in mouse splenic T cells.

Severe impairment in the functions of immune-competent cells has been observed following trauma and hemorrhage. Inappropriate release of cytokines during trauma and hemorrhagic shock disrupt T lymphocyte functions and enable cells to activate genes whose products are detrimental for maintaining a much-needed humoral and cell-mediated immunity. The intracellular events for gene activation are mediated by cytoplasmic transcription factors present as nascent (signal transducer and activator of transcription 1 (STAT 1)) or as a complex (nuclear factor kappaB (NF-kappaB)). Receptor-initiated phosphorylation activates these transcription factors prior to their nuclear translocation and binding to cognate DNA sequences. Because T cell functions are critical to efficient functioning of the immune system, we investigated whether expression of transcription factors, STAT1 and NF-kappaB, is perturbed in splenic T cells following trauma and hemorrhage. To study this, enriched T cells harvested from spleens (pooled from three or four mice per group) of sham, trauma (consisting of midline laparotomy), sham+trauma, hemorrhage (blood pressure maintained at approximately 30 mmHg for 90 min followed by adequate fluid resuscitation), and trauma+hemorrhage groups at 16-18 h after surgical procedure were probed for signal expressions in the presence and absence of interferon-gamma using electrophoretic mobility shift and Western immunoblot assay procedures. Hemorrhage with or without trauma induced activation of Janus kinase 1, STAT1, and NF-kappaB in T cells. Stimulation of T cells with interferon-gamma led to activation of all these signals in all groups including experimental controls. STAT1 activation was accompanied by Janus kinase 1 phosphorylation, whereas NF-kappaB activation was mediated by phosphorylation and rapid degradation of IkappaBalpha. These studies demonstrate that hemorrhagic shock, with or without laparotomy, is sufficient to induce activation of transcription factors in splenic T cells. Thus, attempts to prevent the activation of transcription factors following hemorrhage by pharmacologic means might be helpful for maintaining cell-mediated immunity under these conditions.

Androgen and estrogen receptors in splenic T lymphocytes: effects of flutamide and trauma-hemorrhage.

The endogenous sex steroids, testosterone and beta-estradiol, play a major role in inflammatory processes. They regulate several cytokine genes by interaction with their intracellular receptors that are, essentially, transcription factors. Because T-lymphocyte functions are altered following trauma-hemorrhage in male mice, we investigated whether (i) receptors for androgen (AR) and estrogen (ER) are present in splenic T lymphocytes, (ii) receptor expressions are altered following trauma-hemorrhage, and (iii) pretreatment of male mice with the AR antagonist, flutamide, alters receptor expressions and IL-6 release. Analysis of nuclear extracts indicated the presence of AR and ER in splenic T lymphocytes. No difference in receptor expressions between males and females or following trauma-hemorrhage was observed. Pretreatment of males with flutamide, however, led to increased ER expression in T lymphocytes of sham and trauma-hemorrhaged animals. This suggested that flutamide is capable of inducing the expression of another receptor belonging to a different gonadal steroid. Because response elements for AR and ER are present in the promoter region of the IL-6 gene, release of IL-6 and expression of signal transducer and activator of transcription 3 (STAT3) were analyzed as functional parameters in splenic T lymphocytes. Trauma-hemorrhage decreased IL-6 release by T lymphocytes and the release was restored to sham levels with flutamide pre-treatment. Similarly, STAT3 expression was decreased in T lymphocytes following trauma-hemorrhage and the expression was restored by flutamide pre-treatment. These data collectively demonstrate the importance of gonadal steroids in the regulation of splenic T-lymphocyte functions.

The aromatase inhibitor, 4-hydroxyandrostenedione, restores immune responses following trauma-hemorrhage in males and decreases mortality from subsequent sepsis.

Studies have shown that immune responses are depressed in male mice, but not in proestrus females after trauma-hemorrhage (TH), resulting in increased mortality from subsequent sepsis in male mice compared with female mice. These gender-specific alterations in immune function are believed to be due to differences in sex steroid levels. Aromatase is a key enzyme in the sex steroid biosynthesis. Although earlier studies have shown that aromatase inhibitors prevent thymic atrophy in aged male rats, it remains unknown whether the use of 4-hydroxy-androstenedione (4-OHA) after TH in male mice has any salutary effects on the depressed immune responses. Male C3H/HeN mice were sham operated or subjected to trauma (i.e., midline laparotomy) and hemorrhagic shock (30+/-5 mmHg for 90 min) followed by adequate fluid resuscitation. 4-OHA (5 mg/kg) or vehicle was administrated s.c. just before resuscitation. At 2 h after resuscitation, the mice were killed, and spleens were harvested. Splenocyte proliferation, interleukin (IL-2), interferon (IFN-gamma), and IL-10 release and expression of androgen (AR) and estrogen receptors (ER)-alpha and -beta by immunoblot and reverse transcription-polymerase chain reaction (RT-PCR) were assessed. In another group, sepsis was induced by cecal ligation and puncture (CLP) 3 days after resuscitation, and survival was measured over a period of 10 days. A significant decrease in splenocyte proliferation, IL-2, and IFN-gamma release and increased release of IL-10 were observed in vehicle-treated mice. Animals treated with 4-OHA showed increased splenocyte proliferation, IL-2, and IFN-gamma release, and decreased IL-10 release. Immunoblot analysis showed decreased expression of the cytosolic AR, but no significant difference in the cytosolic and nuclear ER-alpha and -beta expression was observed in the vehicle-treated group after TH. In addition, AR and ER-beta mRNA expression was increased, whereas ER-alpha expression decreased in the vehicle-treated group after TH. ER-alpha expression decreased and ER-beta expression increased in the nucleus of 4-OHA treated mice as determined by immunoblot. There was no difference in the cytosolic AR expression in the 4-OHA-treated group after TH. AR and ER-beta mRNA expression was unaffected, whereas ER-alpha expression increased under such conditions. In additional groups, the increased mortality rate after TH and subsequent sepsis was significantly reduced by 4-OHA treatment. Thus, 4-OHA seems to be a novel and useful adjunct for restoring the depressed immune functions in males after TH and for decreasing mortality rates from subsequent sepsis.

Androgen-mediated modulation of macrophage function after trauma-hemorrhage: central role of 5alpha-dihydrotestosterone.

Androgens have been implicated as the causative factor for the postinjury immune dysfunction in males; however, it remains unknown whether androgens directly affect macrophages. To study this, male mice were sham operated or subjected to trauma (i.e., midline laparotomy) and hemorrhagic shock (mean arterial pressure, 30 +/- 5 mmHg for 90 min and then resuscitated). The mice received the 5alpha-reductase inhibitor 4-hydroxyandrostenedione (4-OHA) before resuscitation. Plasma TNF-alpha, IL-6, and IL-10 levels were elevated after trauma-hemorrhage and normalized by 4-OHA. TNF-alpha and IL-6 production by splenic macrophages was decreased after injury, whereas Kupffer cell production of these mediators was enhanced. 4-OHA normalized cytokine production. Androgens suppressed cytokine production by splenic macrophages from hemorrhaged mice, whereas it enhanced TNF-alpha and IL-6 production by Kupffer cells. The addition of 4-OHA in vitro normalized cytokine production by cells treated with testosterone, but it had no effect on dihydrotestosterone-treated cells. These results indicate that androgens directly affect macrophage function in males after trauma and hemorrhagic shock and that the intracellular conversion of testosterone to dihydrotestosterone is of particular importance in mediating the androgen-induced effects.

Phase I and pharmacokinetics studies of prochlorperazine 2-h i.v. infusion as a doxorubicin-efflux blocker.

In an earlier phase I study, we reported that the maximal tolerated dose (MTD) of prochlorperazine (PCZ) given as a 15-min i.v. infusion was 75 mg/m2. The highest peak plasma PCZ concentration achieved was 1100 ng/ml. The present study was conducted to determine if PCZ levels high enough to block doxorubicin (DOX) efflux in vitro could be achieved and sustained in vivo by increasing the duration of i.v. infusion from 15 min to 2 h. The treatment schedule consisted of i.v. prehydration with at least 500 ml normal saline (NS) and administration of a fixed standard dose of 60 mg/m2 DOX as an i.v. bolus over 15 min followed by i.v. doses of 75, 105, 135, or 180 mg/m2 PCZ in 250 ml NS over 2 h. The hematologic toxicities attributable to DOX were as expected and independent of the PCZ dose. Toxicities attributable to PCZ were sedation, dryness of mouth, anxiety, akathisia, hypotension, cramps, and confusion. The MTD of PCZ was 180 mg/m2. Large interpatient variation in peak PCZ plasma levels (91-3215 ng/ml) was seen, with the plasma half-life (t1/2 alpha) being approximately 57 min in patients given 135-180 mg/m2 PCZ. The volume of distribution (Vd), total clearance (ClT), and area under the curve (AUC) were 350.1 +/- 183.8 1/m2, 260.7 +/- 142.7 l m2 h-1 and 1539 +/- 922 ng ml h-1, respectively, in patients given 180 mg/m2 PCZ and the respective values for patients receiving 135 mg/m2 were 48.9 +/- 23.76 l/m2, 33.2 +/- 2.62 l m2 h-1, and 4117 +/- 302 ng ml h-1. High PCZ plasma levels (> 600 ng/ml) were sustained in all patients treated with 135 mg/m2 PCZ for up to 24 h. DOX plasma elimination was biphasic at 135 and 180 mg/m2 PCZ, and a > 10-ng/ml DOX plasma level was maintained for 24 h. Partial responses were seen in three of six patients with malignant mesothelioma, in two of ten patients with non-small-cell lung carcinoma, and in the single patient with hepatoma. Our data show that PCZ can be safely given as a 2-h infusion at 135 mg/m2 with clinically manageable toxicities. The antitumor activity of the combination of DOX and PCZ needs to be confirmed in phase II trials.