![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Department of Comparative Biosciences, University of Wisconsin-Madison, Madison, Wisconsin
(Received September 16, 2003; Accepted December 10, 2003)
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
; Skipper et al., 1954; Clarke et al., 1958; Moore et al., 1968; Elion, 1989; Elfarra et al., 1995). However, the clinical use of the thiopurines against solid tumors has been limited by their bone marrow and intestinal toxicity (Frank and Tornyos, 1962; Regelson et al., 1964; Moore et al., 1968). Of late, we have been characterizing the biological activity of cis-6-(2-acetylvinylthio)purine (cis-AVTP; Fig. 1) and trans-6-(2-acetylvinylthio)guanine (trans-AVTG), which are novel prodrugs of 6-MP and 6-TG, respectively. cis-AVTP and trans-AVTG are 
,ß-unsaturated conjugates of the thiopurines that were shown to react rapidly with cellular thiols such as glutathione (GSH) to yield the respective thiopurines as the major metabolites (Gunnarsdottir et al., 2002a). Because overproduction of GSH has been observed in many tumor cells, including colon, breast, lung, ovarian, and renal tumor cells (Mekhail-Ishak et al., 1989; Perry et al., 1993; Oberli-Schrämmli et al., 1994; Joncourt et al., 1998; Lusini et al., 2001), the prodrugs cis-AVTP and trans-AVTG may more selectively deliver the thiopurine moiety to a tumor cell than is achieved using the thiopurine itself. Thus, these prodrugs may limit the systemic toxicity associated with thiopurine use. Additionally, because up-regulated levels of GSH are associated with chemotherapeutic drug resistance, the prodrugs may also deliver a thiopurine moiety to a tumor cell that has become resistant to chemotherapeutic drugs because of up-regulated GSH levels.
|
When the in vivo toxicity of the prodrugs cis-AVTP and trans-AVTG was assessed in mice, we observed less bone marrow and intestinal toxicity after multiple treatments with the prodrugs than after equivalent treatments with 6-TG. In particular, minimal toxicity was observed after treatment with cis-AVTP (Gunnarsdottir et al., 2002b). Additionally, treatment with cis-AVTP, trans-AVTG, and 6-TG yielded distinct patterns of metabolite composition and quantities in mouse tissues, suggesting differences in tissue distribution and metabolism among the compounds (Gunnarsdottir and Elfarra, 2003). Also, when the in vitro cytotoxicities of cis-AVTP and trans-AVTG were assessed in the renal cell carcinoma cell lines ACHN and A-498, the prodrugs exhibited similar cytotoxicities that were greater than the cytotoxicity of 6-MP but comparable with that of 6-TG after a 72-h drug incubation (Gunnarsdottir et al., 2002a). Likewise, studies comparing the cytotoxicities of cis-AVTP and 6-MP after different incubation times revealed that even after as little as 24 h of drug incubation, extensive cytotoxicity was detected in incubation with cis-AVTP, whereas no cytotoxicity was observed in 6-MP incubations (Gunnarsdottir et al., 2002a). These findings suggested that the in vitro cytotoxic potency among the thiopurines and their prodrugs was different. Because the in vitro cytotoxicity studies were only carried out in two renal cell carcinoma cell lines, it was of interest to assess the activity of the prodrugs against cancer cell lines derived from other tissue types to determine whether differences in sensitivity existed among different cell lines and different tissue types. Therefore, the cytotoxicities of cis-AVTP and trans-AVTG were assessed in the National Cancer Institute's (NCI's) anticancer screening program and compared with the cytotoxicities of 6-MP and 6-TG, which have also been evaluated in the same screen. Analyses using the pattern-recognition algorithm COMPARE (Paull et al., 1989; http://itbwork.nci.nih.gov) were also carried out for the prodrugs.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
).
The NCI's Anticancer Drug Screen. The protocol used for the cytotoxicity assessment in the NCI anticancer screening program has been described in detail (Monks et al., 1991, 1997). Briefly, tumor cell lines derived from leukemia, lung, colon, brain, melanoma, ovary, kidney, prostate, and breast were grown in RPMI 1640 medium containing 5% fetal bovine serum and 2 mM L-glutamine. Cells (5000–40,000 cells/well) were plated into 96-well microtiter plates and allowed to grow for 24 h at 37°C in a humidified atmosphere supplemented with 5% CO2. cis-AVTP or trans-AVTG, dissolved in dimethyl sulfoxide, was then added to the cells at final concentrations of 0.01 µM to 100 µM, after which the cells were incubated for another 48 h. At the end of the incubation period, the cells were fixed in situ and stained with the protein-staining dye sulforhodamine B. After solubilization of the dye, the optical density of the stain was measured at 515 nm. Three dose-response parameters were calculated for the prodrugs: GI50, the drug concentration resulting in a 50% reduction in the net protein increase compared with control cells during the drug incubation; TGI, the drug concentration resulting in total growth inhibition, and LC50, the concentration of drug resulting in a 50% reduction in the measured protein at the end of the drug treatment compared with that at the beginning, thus indicating a net loss of cells following treatment. These three parameters were calculated if the level of cytotoxicity was reached, whereas if the effect was not reached or was exceeded, the value was listed as greater or less than the maximum or minimum concentration tested.
The thiopurines 6-MP and 6-TG are in the NCI's standard agent database and, thus, their in vitro cytotoxicity has been determined using the assay described above. The results from these assays can be accessed from the NCI's website (http://itbwork.nci.nih.gov). The data on the cytotoxicity of 6-MP and 6-TG used for the analyses described in this paper were obtained from the NCI's website in July 2003. Because the highest drug dilution used to assess the cytotoxicity of 6-MP and 6-TG was higher than 100 µM, any GI50, TGI, and LC50 value obtained for 6-MP and 6-TG that was above 100 µM is listed as greater than 100, to simplify the comparison between the thiopurines and their prodrugs.
COMPARE Analyses. The pattern-recognition algorithm COMPARE (Paull et al., 1989; http://itbwork.nci.nih.gov) was used to search databases of screened compounds for agents similar to cis-AVTP and trans-AVTG in their pattern of activity against the cell lines used in the NCI screen. Similarity in pattern may indicate similarity in molecular structure or mechanism of action. The analyses were carried out using the data obtained for cis-AVTP and trans-AVTG as seeds, searching all public data using all available cell lines.
Statistical Analyses. Statistical analyses were carried out using the software package R (version 1.4.1; Free Software Foundation Inc., Boston, MA).
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
Treatment with cis-AVTP yielded GI50 values ranging from less than 0.01 µM to 6.94 µM in the cell lines used to assess the cytotoxicity of cis-AVTP. Additionally, whereas most TGI values obtained after cis-AVTP treatment were within the range of 0.24 µM to 27.4 µM, the leukemia cell lines K-562, MOLT-4, and SR, and the colon cell line SW-620 had TGI values greater than 100 µM. Similarly, the LC50 values obtained after cis-AVTP treatment ranged from 2.85 µM to more than 100 µM. Treatment with trans-AVTG yielded a greater range in the obtained GI50 and TGI values than was obtained after cis-AVTP treatment. The lowest and highest GI50 values obtained after trans-AVTG treatment were less than 0.01 µM and 18.7 µM, respectively, whereas the TGI values were generally within 0.05 µM to 51.4 µM, except for the cell lines K-562 and MOLT-4 (leukemia), HOP-62 (non-small cell lung, NSCL), SNB-75 (central nervous system, CNS), UACC-257 (melanoma), and OVCAR-8 (ovarian), which had TGI values greater than 100 µM. The LC50 values of trans-AVTG ranged from 7.86 µM to more than 100 µM.
Cell lines that showed a consistently good response toward cis-AVTP treatment at all three parameters assessed were COLO 205, HCT-116, and HT29 (colon), SF-539 (CNS), M14, SK-MEL-28, SK-MEL-5, and UACC-62 (melanoma), OVCAR-3 and OVCAR-5 (ovarian), ACHN, CAKI-1, TK-10, and UO-31 (renal), and MDA-MB-435 and BT-549 (breast). These 16 cell lines had GI50, TGI, and LC50 values below 2, 5, and 15 µM, respectively. Additionally, the leukemia cell lines CCRF-CEM, HL-60(TB), and RPMI-8226 had low GI50 and TGI values. Cell lines that show a consistently good response toward trans-AVTG treatment are NCI-H522 (NSCL), COLO 205 and HCT-116 (colon), IGR-OV1 and OVCAR-3 (ovarian), and CAKI-1 (renal). These 6 cell lines also had GI50, TGI, and LC50 values below 2, 5, and 15 µM, respectively. Furthermore, the leukemia cell lines CCRF-CEM and RMPI-8226 exhibited good response toward trans-AVTG treatment.
When the responses obtained after cis-AVTP treatment were compared with those obtained after treatment with the respective thiopurine 6-MP, it was revealed that cis-AVTP treatment yielded GI50, TGI, and LC50 values that were lower than or equal to those obtained after 6-MP treatment in all the cell lines used to assess the cytotoxicity of cis-AVTP (Tables 1 and 3). Thus, the median GI50, TGI, and LC50 values obtained after treatment with cis-AVTP were lower than those obtained after 6-MP treatment (Table 5). On the other hand, treatment with trans-AVTG yielded GI50, TGI, and LC50 values lower than or equal to those obtained after treatment with 6-TG in 28, 45, and 51 cell lines, respectively, of the 51 cell lines used to assess trans-AVTG cytotoxicity (Tables 2 and 4). However, the median values obtained for all cell lines after trans-AVTG treatment were nevertheless always lower than those of 6-TG (Table 5).
|
To examine whether cell lines of different histotypes respond differently to treatment with a prodrug compared with its respective thiopurine, the median GI50, TGI, and LC50 values for cell lines belonging to the same histotype were calculated for each drug treatment and compared between the prodrug and its respective thiopurine. For the nine different histotypes included in the screen, the median GI50, TGI, and LC50 values obtained after cis-AVTP treatment were always lower than or equal to those obtained after 6-MP treatment (Table 6). Similar comparisons carried out for trans-AVTG and 6-TG revealed that higher median GI50 values were obtained after trans-AVTG treatment than after 6-TG treatment in leukemia, NSCL, colon, CNS, and prostate cell line panels, whereas the melanoma, ovarian, renal, and breast cancer cell line panels were more sensitive toward trans-AVTG treatment than toward 6-TG treatment (Table 7). Contrary to the results obtained for the GI50 endpoint, the median TGI and LC50 values obtained for all the histotypes after trans-AVTG treatment were always lower than or equal to those obtained after treatment with 6-TG.
|
|
The greater cytotoxicity of the prodrugs compared with the respective thiopurines could be due to increased incorporation of thiopurine-containing nucleotides into DNA after prodrug treatment, but such incorporation is considered a primary mechanism of the cytotoxicity of the thiopurines (Elion, 1989). Thus, a correlation may exist between thiopurine cytotoxicity and the cellular doubling time. Alternatively, because the prodrugs are bioactivated by GSH, a correlation may exist between intracellular GSH concentrations and the cytotoxicity observed after prodrug treatment. In preliminary analyses examining the relationship between GI50 values obtained after prodrug treatment and the doubling time for the cell lines included in the screen, no significant correlation was found between these two parameters (data not shown). Similarly, no significant correlation was found between the GI50 values and the GSH concentrations reported for the cell lines (Tew et al., 1996; data not shown).
COMPARE analyses were carried out for the parameter GI50 obtained for cis-AVTP and trans-AVTG. When cis-AVTP was used as the seed, 13 compounds were found to have correlation coefficients ranging from 0.550 to 0.699; 6 of these had correlation coefficients above 0.600. Interestingly, 12 of the 13 compounds with correlation coefficients greater than 0.550 contained a thiopurine moiety; among them, 6-MP (data not shown). The compound whose cytotoxicity profile correlated best with that of cis-AVTP was the heterocyclic thioguanine conjugate, NSC 348401 (Fig. 1). When trans-AVTG was used as the seed, 17 compounds had correlation coefficients ranging from 0.550 to 0.661 but only 2 of these had correlation coefficients greater than 0.600. Interestingly, only 3 of these 17 compounds contained a thiopurine moiety, and 6-TG was not among them. The most common structural feature of compounds whose activity correlated highly with that of trans-AVTG were heterocyclics (data not shown). More intriguingly, the compound with the highest correlation coefficient was NSC 348401. Thus, the cytotoxicity responses obtained after treatment with both cis-AVTP and trans-AVTG correlate best with the response obtained after treatment with NSC 348401. The median GI50, TGI, and LC50 values calculated using data reported for NSC 348401 on NCI's website, were 1.01, 11.7, and 42.0 µM, respectively. These GI50 and TGI values are comparable to those obtained after trans-AVTG treatment but higher than those obtained after cis-AVTP treatment, whereas the LC50 value is comparable to that obtained after cis-AVTP treatment.
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
).
cis-AVTP is more cytotoxic than its respective thiopurine in all the cell lines included in the screen for all three endpoints, GI50, TGI, and LC50. Similarly, whereas the cytotoxicity of trans-AVTG was comparable to that of 6-TG for the endpoint GI50, trans-AVTG was in general more cytotoxic than 6-TG at the endpoints TGI and LC50. The mechanism responsible for the increase in cytotoxicity observed after treatment with the prodrugs compared with their respective thiopurine is currently not fully understood. Previously, we demonstrated that the in vitro uptake of trans-AVTG in the renal cancer cell lines ACHN and A-498 was greater than that of 6-TG, resulting in higher intracellular 6-TG concentration after trans-AVTG treatment than after equimolar 6-TG treatment (Gunnarsdottir et al., 2002a). It is likely that the cellular uptake of cis-AVTP is also greater than that of 6-MP, resulting in higher intracellular 6-MP concentrations. Alternatively, we have demonstrated that the intracellular metabolism of trans-AVTG to 6-TG decreases the intracellular GSH concentrations and thus disrupts the cellular GSH homeostasis (Gunnarsdottir et al., 2002a). Furthermore, we showed that by modulating the intracellular GSH concentrations, we could modulate the intracellular 6-TG concentrations after treatment with trans-AVTG (Gunnarsdottir et al., 2002a). It has been suggested that disruption of intracellular GSH status can affect the intracellular redox state and initiate apoptosis (Coffey et al., 2000; Yang et al., 2000a,b; Davis et al., 2001). Therefore, it is conceivable that the perturbation of intracellular GSH homeostasis that occurs when the prodrugs are bioactivated may enhance the cytotoxic effects of the thiopurines. This hypothesis is supported by previous studies examining the in vivo efficacy of BW 57-323, a 6-TG prodrug and a structural analog of trans-AVTG. In these studies, it was discovered that administration of sulfhydryl compounds prior to or simultaneously with B.W. 57-323 administration decreased the antitumor activity of B.W. 57-323, suggesting that something other than thiopurine release might play a role in the antitumor activity of the compound (Elion et al., 1960). However, in preliminary statistical analyses reported here, no significant correlation between the GI50 values obtained after prodrug treatment in this study and the cellular doubling time or the cellular GSH concentrations was found, suggesting that neither the doubling time nor GSH concentration alone can adequately explain sensitivity of a particular cell line toward prodrug treatment. These analyses do not exclude that a combination of the doubling time, GSH concentration, and other biochemical factors such as drug transport into and out of cells and DNA repair may be able to predict cellular response toward treatment with the prodrugs.
Several findings are worth further consideration when the responses of the different histotypes toward cis-AVTP treatment are examined. It is of interest to point out that after treatment with cis-AVTP, leukemic cells had the lowest median GI50 value but the highest median TGI and LC50 values of all the tissue types included in the screen (Table 6). Similar response was obtained after treatment with the parent thiopurine, 6-MP; leukemic cells had the lowest median GI50 value of the tissue types, whereas the TGI and LC50 values were above 100 µM. In light of the fact that 6-MP is successfully used as an antileukemic agent, these low GI50 but high TGI and LC50 values obtained for leukemic cells show that, at least in some cases, a clinically useful drug can have high TGI and LC50 values. Based upon the widespread use of 6-MP as an antileukemic agent, it would be of considerable interest to examine whether cis-AVTP also exhibits antileukemic activity in vivo and whether the antileukemic activity of cis-AVTP surpasses that of its corresponding thiopurine.
After treatment with cis-AVTP, the panel of breast cancer cells exhibited very low median GI50 and TGI values, whereas their median LC50 value was above 100 µM. Interestingly, even though the median LC50 values were high, two cell lines had LC50 values below 10 µM. Thus, there appears to be a bimodal response toward cis-AVTP in the treatment of breast cancer cells when LC50 is used as the endpoint, most likely because of unidentified inherent differences among the breast cancer cell lines used in the screen.
The response of melanoma and renal cancer cell line panels toward treatment with cis-AVTP is of considerable significance. These two cell line panels had the second and third lowest GI50 values, and the lowest and second lowest TGI and LC50 values, suggesting that they are very sensitive toward cis-AVTP treatment. The sensitivity of renal cancer cells is of importance because no chemotherapeutic agent is currently effective against renal cell carcinoma (Hartmann and Bokemeyer, 1999; Motzer and Russo, 2000). At this point, the mechanism responsible for the sensitivity of melanoma and renal cancer cell lines toward cis-AVTP is unclear. However, it has previously been shown that incubation of isolated rat renal cortical cells with 6-MP at 0.5 and 1 mM concentrations for 2 h significantly increases cell death as measured by the release of lactate dehydrogenase compared with buffer-only incubations. This increase in lactate dehydrogenase release could be partially blocked by allopurinol, an inhibitor of xanthine oxidase that catalyzes the oxidation of 6-MP to thiouric acid and superoxide anion (Lash et al., 1997). These findings suggest that the toxicity of 6-MP could be, in part, due to the generation of reactive oxygen species and oxidative stress, and further indicate that in isolated renal cells, 6-MP toxicity occurs via multiple pathways. Thus, cis-AVTP toxicity may also occur via multiple pathways.
The finding that the GI50 and TGI values of trans-AVTG are comparable to or better than those of 6-TG suggests that trans-AVTG may possess similar or better antitumor activity than 6-TG. Similar to what was observed after cis-AVTP and 6-MP treatment, leukemia cells were most sensitive toward treatment with trans-AVTG and 6-TG when the endpoint GI50 was considered, but not at the endpoints TGI or LC50. When the GI50 values obtained after trans-AVTG and 6-TG treatment are examined, it is of interest to note that five of six ovarian cancer cell lines are more sensitive toward trans-AVTG than toward 6-TG, whereas the opposite is true for colon cancer cells. These observations suggest that the mechanism of action of trans-AVTG and 6-TG may be different. The results of the COMPARE analyses support these observations.
In general, melanoma and ovarian cancer cells appear to be the most sensitive cell line panels toward trans-AVTG treatment. The sensitivity of ovarian cancer cells toward trans-AVTG treatment is of importance because currently, there is a need for new chemotherapeutic drugs effective against ovarian cancers (Piccart et al., 2001). Both the melanoma and ovarian cell line panels have low GI50 and TGI values, whereas their LC50 values are above 50 µM, possibly suggesting that trans-AVTG may act as a cytostatic compound rather than a cytotoxic compound in these cells. However, a trimodal response is seen when the LC50 values for individual cell lines making up the panel of ovarian cells is examined. Two cell lines have low LC50 values (7.86 and 12.1 µM) and two cell lines have LC50 values around 50 µM, whereas two cell lines have LC50 values above 100 µM. Thus, trans-AVTG may be useful against a subset of ovarian cancers.
It is noteworthy that COMPARE analyses of the cytotoxicity profiles of both cis-AVTP and trans-AVTG revealed that they correlate best with that of the heterocyclic conjugate of thioguanine, NSC 348401. To the best of our knowledge, no further characterization of NSC 348401 and its mechanism of cytotoxicity has been published. Because of this similar cytotoxicity profile, it would be of considerable interest to examine whether NSC 348401 is also a thiopurine prodrug and whether it is bioactivated by reaction with GSH similar to cis-AVTP and trans-AVTG.
Taken together, our results suggest that cis-AVTP and trans-AVTG show superior in vitro cytotoxicity compared with their respective thiopurine. Additionally, the prodrugs exhibit both similar and distinct activities against the cell lines included in the NCI anticancer screen. Both compounds appear active against leukemias; this finding is consistent with the use of the parent thiopurines as antileukemic agents. Furthermore, melanoma cells appear relatively sensitive toward treatment with both compounds. However, cis-AVTP appears much more effective against renal cell lines, whereas trans-AVTG is more effective against ovarian cancer cell lines.
|
Footnotes |
|---|
Preliminary results from this work have previously been published in the Ph.D. thesis of Sjofn Gunnarsdottir, University of Wisconsin-Madison, 2002.
1 Abbreviations used are: 6-MP, 6-mercaptopurine; 6-TG, 6-thioguanine; GSH, glutathione; trans-AVTG, trans-6-(2-acetylvinylthio)guanine; cis-AVTP, cis-6-(2-acetylvinylthio)purine; NCI, National Cancer Institute; GI50, drug concentration reducing tumor cell growth by 50% compared with untreated controls; LC50, drug concentration required to decrease tumor cell numbers by 50% compared with untreated controls; TGI, drug concentration required to inhibit tumor cell growth; NSCL, non-small cell lung; CNS, central nervous system; B.W. 57-323, 2-amino-6-[(1-methyl-4-nitro-5-imidazolyl)thio]purine; NSC 348401, 6-[(7-nitro-2,1,3-benzoxadiazol-4-yl)thio]-9H-purin-2-amine. 
Address correspondence to: Dr. Adnan A. Elfarra, Department of Comparative Biosciences, School of Veterinary Medicine, University of Wisconsin-Madison, 2015 Linden Drive, Madison, WI 53706. E-mail: elfarraa{at}svm.vetmed.wisc.edu
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|
 |
 |
 |
