Introduction

Top Abstract Introduction Experimental Procedures Results Discussion References

Previously, a series of 1,2-diacyl-4,4-diethyl-3,5-pyrazolidinediones demonstrated antineoplastic activity due to the marked suppression of DNA, RNA, and protein syntheses in Tmolt₃ leukemia cells. It was determined that Tmolt₃ tumor cell death was induced by the inhibition of multiple enzyme activities involved with nucleic acid synthesis (Hall et al., 1995). Subsequently, a series of 3,3-disubstituted-1,5-diazabicyclo[3.1.0]hexane-2,4-diones were synthesized as second generation compounds and their antineoplastic activity examined to establish chemical modifications on the pyrazolidine-3,5-dione ring that may play a role in specificity to a target enzyme (Barnes et al., 2000, 2001). The antiproliferative activity of these agents was determined in murine and human tumor cell lines demonstrating potent ED₅₀ values of 0.7 to 13.0 µM (Barnes et al., 2000, 2001). Specifically, the 6-ethoxycarbonyl-substituted derivatives were significantly active against the growth of tumor cells derived from human leukemias, glioma, and MCF-7 breast effusion (Barnes et al., 2001).

A mode of action study in Tmolt₄ acute lymphoblastic leukemia cells showed that the inclusion of a tricyclic ring system [3.1.0] composed of a five-membered pyrazolidine-3,5-dione ring fused to a substituted diaziridine resulted in agents with exclusive effects on DNA and de novo purine biosynthesis (Barnes et al., 2001). The major cellular target of these three compounds was IMPDH [EC 1.1.1.205], one of the rate-limiting enzymes in de novo purine biosynthesis. Isolation of the type I and II IMPDH isoforms, originating from Tmolt₄ leukemia cells, led to studies that revealed that the effects of the agents on IMPDH activity were strictly due to the inhibition of type II IMPDH activity (Barnes et al., 2000, 2001). The agents acted as competitive inhibitors of IMPDH type II with respect to the endogenous substrate, IMP.

The significance of this finding is due to the recent elucidation of two IMPDH isoforms (I and II). The type I and II IMPDH proteins have been shown to be indistinguishable in their catalytic activities, substrate affinities, and K_i values for known inhibitors (Collart and Huberman, 1988; Natsumeda et al., 1990; Hager et al., 1995). The two mRNA transcripts are differentially regulated in a confounding manner, where type I is constitutively expressed and is the predominant species in normal cells, while type II is selectively up-regulated in neoplastic and replicating cells and emerges as the dominant species (Konno et al., 1991; Nagai et al., 1991, 1992; Senda and Natsumeda, 1994). This remarkable difference in the regulation of the two enzymes allows for the design of therapeutic agents that could be directed to the type II isoform specifically in cancer cells. In an attempt to improve and explore the selective inhibition of type II IMPDH activity by the 1,5-diazabicyclo[3.1.0]hexane-2,4-diones, a pentamethylene substitution has been made at position C-6, resulting in the 3,3-disubstituted-6,6-pentamethylene-1,5-diazabicyclo[3.1.0]hexane-2,4-diones. The effects of these agents on tumor cell growth studied in cell lines derived from human neoplasms and IMPDH activity are reported herein.

	Experimental Procedures

Top Abstract Introduction Experimental Procedures Results Discussion References

Materials. The gene for type I and II human IMPDH was cloned and expressed in Escherichia coli as previously described (Barnes et al., 2000). All radioisotopes were purchased from PerkinElmer Life Sciences (Boston, MA) unless otherwise indicated. Radioactivity was determined in Fisher Scintiverse scintillation fluid with corrections for quenching. Substrates and cofactors were obtained from Sigma Chemical Co. (St. Louis, MO).

Synthesis of Compounds 2 and 3. The general procedure for synthesis of compounds 1 to 3 was previously reported (Barnes et al., 2000).

3,3-Diethyl-6,6-pentamethylene-1,5-diazabicyclo[3.1.0]hexane-2,4-dione (1). Analytical results were previously reported (Barnes et al., 2000).

3-Ethyl-3-phenyl-6,6-pentamethylene-1,5-diazabicyclo[3.1.0]hexane-2,4-dione (2). 10%: m.p. 66-68.5°C; IR (Nujol) 1737 cm¹; ¹H NMR (CDCl₃)  0.86 and 0.95 (2 t, 3 H), 1.5 to 1.7 and 1.7 to 1.88 (2 br m, 10 H), 1.92 and 2.41 (2 q, 2 H), 7.27 to 7.45 (m, 3 H), 7.68 (m, 2 H); MS (rel int) m/z 284 (50). Anal. calcd. for C₁₇H₂₀N₂O₄: C 71.8, H 7.0, N 9.8. Found: C 72.0, H 7.3, N 9.8.

3-Ethyl-3-(4-methylphenyl)-6,6-pentamethylene-1,5-diazabicyclo [3.1.0]hexane-2,4-dione (3). 35%: m.p. 65-68°C; IR (Nujol) 1747 cm¹; ¹H NMR (CDCl₃)  0.87 and 0.95 (2 t, 3 H), 1.5 to 1.7 and 1.7 to 1.85 (2 br m, 10 H), 1.9 and 2.39 (2 q, 2 H), 2.34 (s, 3 H), 7.19 to 7.28 (m, 3 H), 7.53 to 7.57 (m, 1 H). Anal. calcd. for C₁₈H₂₂N₂O₄: C 72.5, H 7.4, N 9.4. Found: C 72.4, H 7.2, N 9.4.

Cell Proliferation Assay. Compounds 1 to 3 were tested for antiproliferative activity by methods previously described (Hall et al., 1995; Barnes et al., 2001). The following tumor cell lines, derived from a variety of murine and human neoplasms, were maintained by literature techniques (Geran et al., 1972) and the growth medium and conditions were according to American Type Culture Collection protocols: murine L₁₂₁₀ lymphoid leukemia and P₃₈₈ lymphocytic leukemia, human Tmolt₃ and Tmolt₄ acute lymphoblastic leukemia, HL-60 promyelocytic leukemia, HUT-78 lymphoma, THP-1 acute monocytic leukemia, HeLa-S³ suspended cervical carcinoma, HeLa solid cervical carcinoma, KB epidermoid nasopharynx, SkMel-2 malignant melanoma, colorectal adenocarcinoma SW480, HCT-8 ileocecal adenocarcinoma, lung bronchogenic MB-9812, A549 lung carcinoma, Saos-2 osteosarcoma, breast MCF-7, clear cell renal Caki-1, A-431 skin epidermoid carcinoma, glioma U87 MG, and normal lung fibroblast WI-38. Values for antiproliferative activity were expressed as ED₅₀ (µg/ml and µM), i.e., the concentration of compound that inhibits 50% of cell growth, where values less than 4 µg/ml were required for significant activity of cell growth inhibition by National Institutes of Health standards. Standard antineoplastic agents 6-mercaptopurine, etoposide, and MPA were also examined.

Incorporation Studies. The effects of agents 1 to 3 at 25, 50, and 100 µM on the incorporation of radiolabeled [³H]thymidine, [³H]uridine, or [³H]leucine into DNA, RNA, or protein, respectively, for 10⁶ human Tmolt₄ cells were determined at 60-min incubations (Hall et al., 1995).

Enzyme Assays. The effects of compounds 1 to 3 on Tmolt₄ nucleic acid metabolism were analyzed at concentrations of 25, 50, and 100 µM after a 60-min incubation period. DNA polymerase  [E.C. 2.7.7.7] activity was determined in cytoplasmic extracts using a protocol by Sawada et al. (1974). Messenger, ribosomal, and transfer RNA polymerase nuclei enzymes [E.C. 2.7.7.6] were isolated and individual RNA polymerase activities measured (Hall et al., 1995). The following enzyme activities were assayed using Tmolt₄ homogenates. Ribonucleoside reductase [E.C. 1.17.4.1] activity was measured using [¹⁴C]cytidine-5'-diphosphate with dithioerythritol and carbamyl phosphate synthetase [E.C. 6.3.5.5] activity was determined (Hall et al., 1995). Aspartate transcarbamylase [E.C. 2.1.3.2] activity and the product carbamyl aspartate were measured colorimetrically (Hall et al., 1995). Thymidylate synthase [E.C. 2.1.1.45] activity was analyzed as previously described (Hall et al., 1995). Thymidine, thymidine-5'-monophosphate, and thymidine-5'-diphosphate kinase [E.C. 2.7.1.21] activities were determined using [³H]thymidine (58.3 mCi/mmol) in the medium of Maley and Ochoa (1958). Dihydrofolate reductase [E.C. 1.5.1.3] activity was assayed by monitoring the disappearance of NADH at 340 nm (Hall et al., 1995). Amidophosphoribosyltransferase [E.C. 2.4.2.14] activity was determined by the method of Martin (1972) and IMPDH activity was analyzed with 8-[¹⁴C]IMP (54 mCi/mmol) (Hall et al., 1995). Effects of compounds 1 to 3 on IMPDH enzymatic activity were determined using whole cell Tmolt₄. The activity of ribavirin was measured on a cellular homogenate. Protein content was determined for the enzymatic assays by the Lowry technique.

DNA Studies. After deoxyribonucleoside triphosphates were extracted, deoxyribonucleoside triphosphate levels were assayed by the method of Hunting and Henderson (1981) and Hall et al. (1995). The effects of compounds 1, 2, and 3 on DNA strand scission were determined by methods described previously (Hall et al., 1995). Thermal calf thymus DNA denaturation studies, changes in DNA UV absorption from 220 to 340 nm, and DNA viscosity studies were conducted after incubation of compounds 1 to 3 at 100 µM, 37°C for 24 h (Zhao et al., 1987).

Guanosine Recovery Studies. Cytotoxicity and DNA synthesis studies were conducted in Tmolt₄ cells as indicated previously at the respective ED₅₀ values for each compound (Barnes et al., 2001). Exogenous guanosine from 5 to 50 µM was coincubated in the assays (Lee et al., 1985; Yu et al., 1989)

Recombinant IMPDH Type I and II Enzyme Assay. Recombinant human type I and II IMPDH were prepared as previously described (Barnes et al., 2000) and specific activities were 1.1 and 0.9 U/mg, respectively, in the standard assay buffer (50 mM Tris-Cl, pH 8.0, 100 mM KCl, 3 mM EDTA, 1 mM diothiothreitol, 10-200 µM IMP, 30-500 µM NAD, and 0.2-0.4 µM purified enzyme) at 37°C. IMPDH activity was determined spectrophotometrically by methods previously described (Barnes et al., 2000, 2001). Steady-state apparent kinetic parameters were evaluated by the direct fit of initial velocity data versus substrate concentration to the Michaelis-Menten equation using a weighted nonlinear regression method in the program Enzyme Kinetics (Trinity Software, Campton, NH).

IMPDH Type I and II Enzyme Inhibition. IC₅₀ values for compounds 1, 2, and 3 were determined using purified recombinant type I and type II IMPDH as previously described (Barnes et al., 2000, 2001). IC₅₀ values were estimated from a semilog plot of inhibitor concentration versus percentage of inhibition of enzyme activity for type I or type II IMP dehydrogenase.

Kinetic Analysis for Type II Inhibition. Kinetic studies were conducted at 37°C using the standard assay buffer but with variable concentration of inhibitor and one of the substrates, while the second substrate concentration was either saturating or below saturation (Barnes et al., 2000, 2001). In addition to examining binding at the IMP active site, studies were conducted to test interactions of the inhibitors with the NAD binding site by using a fixed concentration of IMP as the nonvariable substrate and NAD at varying concentrations in the presence of increasing concentrations of the inhibitors. For the determination of K_i values, the apparent K_m and V_max at each inhibitor concentration were determined by nonlinear regression assuming Michaelis-Menten kinetics. Calculations were carried out using Enzyme Kinetics from Exeter. MPA and ribavirin were used as positive controls for all inhibition studies. Reactions for K_i determinations of MPA were performed at 14 nM enzyme.

Statistical Analysis. Data are displayed in tables and figures as the means ± standard error of the mean expressed as percentage of control. N is the number of samples per group. The Student's t test was used to determine the probable level of significance (p) between test samples and control samples.

	Results

Top Abstract Introduction Experimental Procedures Results Discussion References

Synthesis of Inhibitors. Compounds 2 and 3 were successfully synthesized and purified as derivatives of parent compound 1 (Fig. 1). The identity and purity of these compounds were confirmed by NMR spectroscopy, IR, MS, and elemental analysis.

View larger version (11K): [in this window] [in a new window]   Fig. 1.   Chemical structures: (1) 3,3-diethyl-6,6-pentamethylene-1,5-diazabicyclo[3.1.0]hexane-2,4-diones, (2) 3-ethyl-3-phenyl-6,6-pentamethylene-1,5-diazabicyclo[3.1.0]hexane-2,4-diones, and (3) 3-ethyl-3-(4-methylphenyl)-6,6-pentamethylene-1,5-diazabicyclo[3.1.0]hexane-2,4- diones.

Antiproliferative Activity and Enzyme Inhibition. To further define substitutions of the 1,5-diazabicyclo[3.1.0]hexane-2,4-dione ring that are necessary for selective inhibition of IMPDH activity leading to the targeted death of cells derived from murine and human neoplasms, two new derivatives of the chemical class 3,3-disubstituted-6,6-pentamethylene-1,5-diazabicyclo[3.1.0]hexane-2,4-dione were synthesized and examined for antiproliferative activity. The effects of agents 1 to 3 on a number of murine and human tumor cell lines were studied along with an assortment of enzymes involved with nucleic acid metabolism to establish a molecular mechanism of action. Compounds 1 to 3 were all significantly active in the majority of suspended tumor cell screens with ED₅₀ values <4 µg/ml. In human Tmolt₃ and Tmolt₄ acute lymphoblastic leukemia, HUT-78 lymphoma, and HeLa-S³ uterine carcinoma screens, all three compounds were significantly active with ED₅₀ values of 3.3 to 13.6 µM (Table 1). However, in the HL-60 promyelocytic and THP-1 acute monocytic leukemia screens only compounds 1 and 2 demonstrated good activity.

View larger version (19K): [in this window] [in a new window]   Fig. 2.   Effects of 1, 2, and 3 on human Tmolt4 lymphoblastic leukemia cell growth over 8 days. Cells (100 µl) were removed from incubated tubes at each time point and stained with trypan blue to determine cell viability and counted using a hemocytometer. Points on the lines are representative of the mean of three experimental values obtained. For each value point, p < 0.05.

View this table: [in this window] [in a new window]   TABLE 3 Effects of agents on human Tmolt4 lymphoid leukemia cell metabolism at 60 min Mean values ± S.E.M. (N  4) versus specified Tmolt4 cellular synthesis, enzyme activity, or nucleotide pool levels. Control values for 106 Tmolt4 cells incubated 60 min.

Effects on Cellular IMPDH Activity. Taken together, results from the mode of action studies suggest that the molecular target of the 3,3-disubstituted-6,6-pentamethylene-1,5-diazabicyclo[3.1.0]hexane-2,4-diones is the IMPDH enzyme. To elucidate whether this inhibition was specific to IMPDH alone leading to the observed antiproliferative activity, more detailed studies were undertaken with Tmolt₄ cellular IMPDH and the purified recombinant isoenzymes. IC₅₀ values for the inhibition of whole cell IMPDH activity were 55, 65, and >150 µM, respectively, for compounds 1, 2, and 3. A time-dependent study monitoring crude IMPDH activity over 2 h was performed at the IC₅₀ values for individual agents illustrating that the inhibition of the enzyme was immediate (Fig. 3). Furthermore, the addition of guanosine to the medium of Tmolt₄ cells coincubated with the agents should bypass the metabolic block incurred at IMPDH if it is due to the specific inhibition of IMPDH activity. The effects of agents on cell growth (at 3 days) and DNA synthesis (at 60 min) could be observed when 5 µM guanosine was added to the medium (Fig. 4, A and B). These studies illustrated that Tmolt₄ cell growth inhibition could be reversed back to control levels of cell growth by coincubation of guanosine. In addition, effects of agents on suppressing DNA synthesis could be reversed back to levels of DNA synthesis imposed by guanosine alone.

View larger version (18K): [in this window] [in a new window]   Fig. 3.   Effects of 1, 2, and 3 on the inhibition of whole cell Tmolt4 IMPDH activity over 120 min. Agents were incubated with Tmolt4 cells and activity was measured by the radioisotope method described under Experimental Procedures. Values are represented as percentage of the control and each point on the line was calculated as the mean of three experimental values obtained, p < 0.01.

View larger version (33K): [in this window] [in a new window]   Fig. 4.   Effects of exogenous guanosine on the inhibition of human Tmolt4 cell growth and DNA synthesis by agents 1 to 3. A, compounds were incubated at their ED50 values for growth inhibition and guanosine was supplemented at 5 µM. The number of viable cells was determined at the end of the incubation period as described under Experimental Procedures. B, compounds were incubated at their ED50 values for inhibition of Tmolt4 cell growth and guanosine added to obtain a final concentration of 5 µM. The amount of DNA synthesis was determined as described under Experimental Procedures. A and B, results are expressed as the mean of three experimental values and for each column, p < 0.01.

Specificity to Type I and II IMPDH. Studies were undertaken to examine the effects of inhibitors 1, 2, and 3 on human recombinant type I and II IMPDH activity to determine whether the inhibition of cellular IMPDH activity was due to a selectivity toward either isoform. Effects of inhibitor 1 on type I and II IMPDH activity was previously determined (Barnes et al., 2000). Lineweaver-Burke plots indicated that type II IMPDH activity was inhibited competitively by 1 and 2 with respect to the IMP substrate (Fig. 5). IC₅₀ values for compounds 1, 2, and 3 were 63, 98, and 106 µM, respectively. The K_i value for compound 1 was 44.2 µM and 62 µM for compound 2. Compounds 1 to 3 were not effective inhibitors of the type I isoform at concentrations as high as 200 to 500 µM. Instead, type I IMPDH activity was slightly stimulated ~10% (S.E.M. ±0.3) in the presence of agents. The compounds did not display inhibition of IMPDH type II activity when NAD was used as the variable substrate, indicating that inhibitors 1 to 3 do not interact with the enzyme at the cofactor binding site.

View larger version (17K): [in this window] [in a new window]   Fig. 5.   Kinetics of inhibition for compound 2. The data are presented as Lineweaver-Burk Plots of 1/ (s1) as a function of 1/[IMP] (µM1) at different inhibitor concentrations. Measurements were made as competitive inhibitors with respect to IMP in the IMPDH reaction unless noted, while NAD was held constant at 170 µM. The plot inset shows a replot of Km, app./Vmax (µM · s1) versus different concentrations of 2 (µM): , 0 µM; , 10 µM; , 50 µM; and , 100 µM. Points on the lines are representative of the mean of three experimental values obtained with p < 0.01.

	Discussion

Top Abstract Introduction Experimental Procedures Results Discussion References

Results reported herein demonstrate that addition of a three-membered fused ring to the parent disubstituted-pyrazolidine-3,5-dione ring structure, leading to the 3,3-disubstituted-6,6-pentamethylene-1,5-diazabicyclo[3.1.0]hexane-2,4-diones, resulted in compounds with similar antiproliferative activity against the growth of cell lines derived from human leukemias, lymphomas, glioma, and breast tumors, as the previously examined 6-ethoxycarbonyl-3,3-disubstituted-1,5-diazabicyclo[3.1.0]hexane-2,4-diones (Barnes et al., 2001). However, the 6,6-pentamethylene substituted compounds were not as active in the broad spectrum of solid tumor cell screens as the previously studied 6-ethoxycarbonyl-substituted IMPDH inhibitors. The 3,3-disubstituted-6,6-pentamethylene-1,5-diazabicyclo[3.1.0]hexane-2,4-diones appeared to be antimetabolites of the purine biosynthetic pathway, inhibiting the key regulatory enzyme IMPDH. This finding was supported by results obtained from Tmolt₄ cell growth studies over 8 days indicating that the antiproliferative effects of agents was cytostatic. The thesis that IMPDH is the major target of the compounds was supported by the rapid inhibition of cellular IMPDH activity at 15 min and the selective reduction in deoxyguanosine triphosphate pool levels at 60 min. Furthermore, the metabolic block in de novo purine synthesis incurred by the agents due to inhibition of IMPDH activity should be circumvented by the addition of exogenous guanosine to the medium. This level of examination is an indication of the degree of specificity for inhibition of IMPDH activity. The reversal of the inhibitory effects by coincubation of the agents with guanosine indicated that the inhibitors are interacting with IMPDH and excess guanosine is allowing the cells to bypass the need for IMPDH activity. The affects produced by exogenous guanosine on Tmolt₄ cell proliferation and DNA synthesis correspond well with results published in the literature (Kiguchi et al., 1990).

After testing the effects of agents on recombinant type I and II IMPDH previously obtained by reverse transcription-polymerase chain reaction from Tmolt₄ cell RNA (Hager et al., 1995; Barnes et al., 2000), it could be observed that this new class of 3,3-disubstituted-1,5-diazabicyclo[3.1.0]hexane-2,4-diones contained selective inhibitors of type II IMPDH activity with no inhibitory effects on the type I IMPDH isoform. Like the previously tested 1,5-diazabicyclo[3.1.0]hexane-2,4-diones (Barnes et al., 2000, 2001), inhibitors 1 to 3 were competitive with respect to the endogenous substrate IMP. Agents did not inhibit the enzyme activity when tested with NAD as the variable substrate. The standard IMPDH inhibitor ribavirin, also a competitive inhibitor with respect to IMP, showed no significant selectivity for either type I or II IMPDH with IC₅₀ values of 121 and 79 µM, respectively. Furthermore, in our hands, MPA was not significantly selective to either isoform, yielding IC₅₀ values of 0.1 and 0.08 µM for type I and II, respectively. Both ribavirin and MPA were cytotoxic in the human fibroblast cell line WI-38, where inhibitors 1 to 3 were ineffective in reducing cell growth, indicating that selectivity to the type II isoform may decrease affects on normal rapidly growing cells during chemotherapeutic treatment. Most importantly, the antiproliferative activity of compound 1 correlated well with the IC₅₀ value for type II IMPDH inhibition. Inhibitor 1 was the most active agent in the tumor cell screens and demonstrated the lowest K_i value. Previous findings suggest that the differences in ED₅₀ and IC₅₀ values obtained for individual inhibitors are most likely due to metabolism of the agents occurring in the Tmolt₄ cellular system compared with the in vitro system with recombinant enzyme (Barnes et al., 2001).

Even though the 3,3-disubstituted-6,6-pentamethylene-1,5-diazabicyclo[3.1.0]hexane-2,4-diones 1 to 3 were not as potent IMPDH inhibitors as the previously studied derivatives (Barnes et al., 2000, 2001), selectivity for the type II isoenzyme was maintained. Based on these new results and results from previous studies, information regarding a structure-activity relationship could be observed for selective IMPDH type II inhibition. We now know that substitution at position C-6 is not only important for selectivity (Barnes et al., 2000) to the type II isoform but also may be important for potency of inhibition. By replacing the hydrophilic characteristic given by the ethoxycarbonyl or benzoyl moiety with lipophilic substitution (cyclohexane), the K_i values for IMPDH inhibition were increased 4- to 8-fold. Further exploration of the 1,5-diazabicyclo[3.1.0]hexane-2,4-dione ring system is currently underway to determine essential requirements for isoform selectivity leading to more potent derivatives.