Cytotoxic Action of Carboxyborane Heterocyclic Amine Adducts

The heterocyclic carboxyborane amines were found to be potent cytotoxic agents in the murine L1210 lymphoid leukemia and human HeLa suspended carcinoma cells. These agents were observed to inhibit HeLa DNA topoisomerase II activity ~ 200 μM and L1210 topoisomerase II activity ≥ 100 μM. These agents did not cause DNA protein linked breaks themselves, but upon incubation for 14-24 hr did enhance the ability of VP-16 to cause cleavable complexes. The heterocyclic amineboranes inhibited DNA synthesis and caused DNA strand scission. They were additive with VP-16 in affording these results as well as inhibiting colony growth of L1210 cells after co-incubation for 1 hr. The agents inhibited in vitro PKC phosphorylation of both L1210 lymphoid leukemia and human topoisomerase II enzyme.


Introduction
Previously, a series of cyano-, carboxy-, carboxymethoxy-, and carbamoylborane adducts of heterocyclic amines were shown to be potent antineoplastic agents in the Lewis lung and P388 lymphocytic leukemia screens at 20 mg/kg/day in mice. Furthermore, they demonstrated significant cytotoxic activity in human Tmolt3 lymphoblastic leukemia, HeLa-S uterine carcinoma, KB nasopharynx, colorectal adenocarcinoma SW480, and osteosarcoma and murine L1210 lymphoid leukemia cells [1 ]. A mode of action study in L1210 cells demonstrated that DNA synthesis and purine syntheses were preferentially inhibited with significant inhibition at regulatory enzymatic sites such as PRPP-amido transferase, IMP dehydrogenase, dihydrofolate reductase, nucleoside kinases which resulted in a reduction of deoxyribonucleotide pools. Furthermore, LI210 DNA strand scission was evident after 24 hr. incubation with the agents, but the compounds did not alkylate the bases of DNA, intercalate between base pairs of DNA or cause cross linking of DNA strands ]. Thus, the purpose of the following study is to characterize the reasons why the drugs cause DNA strand scission.

Methods
Compounds 1-7 were previously synthesized and the chemical and physical characteristics reported [3] [ Table ]. All radioisotopes were purchased from New England Nuclear (Boston, MA) unless otherwise indicated. Radioactivity was determined in a Fisher Scintiverse scintillation fluid with correction for quenching. Substrates and cofactors were obtained from Sigma Chemical Co. (St. Louis, MO).

Cytotoxicity
Compounds 1-7 were tested for cytotoxic activity by homogenizing drugs in a mg/ml solution in 0.05% Tween 80/H_O. These solutions were sterilized by passing them through an acrodisc (0.45 t). The murine L20 lymphoid leukemia and the human HeLa-S uterine carcinoma cell lines were maintained culture following standard literature techniques. Geran et al.'s protocol [2] was used to assess the cytotoxicity of the compounds and standards in each cell line. Values for cytotoxicity were expressed as EDso tg/ml, i.e. the concentration of the compound inhibiting 50% of cell growth. EDso values were determined by the trypan blue exclusion technique. A value of less than 4 tg/ml was required for growth inhibition activity to be considered significant. P4-Phage Knotted DNA P4-phage knotted DNA was isolated by the method of Liu and Davis [7]. Cultures of C-117 E. coli were grown overnight in 5 ml modified Luria Broth (1% bactotryptone, 0.5% yeast extract, 1% NaC1, 1.6 mM MgC12, 0.5 mM CaC12, and 1% dextrose). After 18 hr, P4 phage 1:4000 stock dilution (4 ml,-lxl0 Ix pfu/ml) was used to inoculate the 5 ml overnight culture. Phage was allowed to adsorb by sitting at room temperature for 10 min. Two L of modified Luria Broth was then inoculated with 4 ml phage infected overnight culture. The 2 L inoculate was incubated at 37C with vigorous aeration for 4 hr and 20 min, after which 20 ml 0.5 M EGTA (pH 8) was added. After hr of stirring at 37C, chloroform (4 ml) was added to ensure complete cell lysis. The entire 2 L of inoculated broth was centrifuged, in 40 ml aliquots, at 10,000 rpm and 4C, for 10 min, to remove cell debris. The final volume of supernatant collected was measured and sufficient NaC1, PEG-8000, and MgCI2 were added to make a solution of 0.5 M NaC1, 10% PEG-8000, and 80 mM MgCI2. This mixture was allowed to precipitate at 4C overnight. After 18 hr the precipitate was pelleted by centrifugation at 10,000 rpm for 10 min. Pellets were combined and resuspended in 15 ml P4 diluent 1% NH4OAc, 80 mM MgCI2, 10 mM Tris-HC1 (pH 7. 2)] and repelleted. This process was repeated. Supernatants from the previous procedure were collected and centrifuged at 25,000 rpm for 4 hr. The resulting pellet, containing phage capsids and phage heads, was resuspended in 45 ml of P4 diluent and the phage solution was weighed. For every gram of phage solution, 0.632 g ultra pure CsC1 was added. The CsC1/phage solution was placed in ultracentrifuge tubes and topped off with mineral oil. The CsCl/phage solution was centrifuged at 24,000 rpm, 4C, for 3 days. Two "milky" colored bands were apparent after 3 days; upper band P4 phage capsids, lower band P4 heads (containing knotted DNA). The bands were collected via a syringe and were dialyzed with P4 diluent (2 L) using 18 mm dialysis tubing. P4 heads were extracted with an equal volume of ultrapure buffered phenol. The aqueous layer was separated and 2 volumes of 99 % ethanol were added and samples were placed at -20C overnight to precipitate the knotted DNA. The precipitate was pelleted at 10,000 rpm for 5 min and was resuspended in 10 mM Tris (pH 8), lmM EDTA, 0.1 M NaC1.
DNA-Topoisomerase II L1210 or HeLa-S DNA-topoisomerase II was isolated by the method of Miller, et a/. [8] by pelleting approximately 108 cells by centrifugation at 2500 rpm for 4 min. This pellet was then washed by resuspending in sterile PBS (pH 7.4) and pelleting again. Next, the pelleted cells were resuspended in buffer solution containing 0.25 M sucrose, 20 mM potassium phosphate (pH 7.5), 2 mM MgC12, mM dithiothreitol (DTT), mM spermidine, 0.1 mM EDTA, 0.1 mM phenylmethylsulfonyl fluoride (PMSF), and mM Na2S2Otl], at 4C. Cell membranes were lysed by addition of a volume of Triton X-100, equivalent to 1/100 of the total cell suspension, in concert with Dounce homogenization. Complete cell lysis was determined by visualization of a small aliquot of cell homogenate stained with trypan blue (0.4%). An equal volume of buffer solution containing 1.75 M sucrose was added to cell suspension and mixed by gently swirling. After mixing the total volume was loaded on a sucrose cushion, 1.4 M sucrose, 20 mM potassium phosphate (pH 7.5), 5 mM MgC12, mM DTT, 0.1 mM EDTA, 0.1 mM PMSF, and centrifuged at 18,000 rpm for 45 min, at 4C. The sucrose cushion was removed via vacuum aspiration and the remaining pellet was resuspended in buffer containing 20 mM potassium phosphate (pH 7.5), 2 mM MgC12, mM DTT, 0.1 mM EDTA, 0.1 mM PMSF, mM D-mercaptoethanol, 10% glycerin, and 100 mM NaC1. The suspension was incubated at 4C for 30 min and then was centrifuged at 2,400 rpm for 5 min. Supernatant was collected and the pellet was resuspended in buffer containing 150 mM NaC1 and incubated again for 30 minutes at 4C, followed by centrifugation. The process was repeated with buffer containing 400 mM NaC1. Fractions containing active enzyme, as determined by the topoisomerase unknotting assay, were kept and stored, after the addition of glycerin to make a 50% v/v solution, at -70C. Active fractions were diluted to U/tl (1 U was defined as the amount of enzyme required to completely unknot 150 ng knotted DNA in hr at 37C) with buffer solution.

DNA Unknotting Assay
The effects of the agents on isolated L1210 or HeLa-S DNA topoisomerase II activity was determined by the method of Miller et a/. [8]. The reaction mixture was prepared to contain the following components at the following concentrations: 0.05 M Tris (pH 7.5), 0.1 M KC1, 0.01 M MgC12, 30 ktg/ml bovine serum albumin, 0.5 mM EDTA, 1.0 mM DTT, 1.0 mM ATP. A "4 x Reaction Mix" was prepared, meaning the components were four times as concentrated as the previously described mix. To prepare samples a premix was made containing 2.5 tl "4 x Reaction Mix" per sample, 0.25 tl knotted DNA/sample, and enough autoclaved, distilled water to bring the volume of each sample up to 8.0 tl. To each sample reaction 1.0 tl sample compound (at desired concentration) and 1.0 tl topoisomerase II (1 U/tl) was added so that the final volume of each sample reaction was 10 tl. Samples were warmed to 65C in a water bath and each was syringed eight times, to sheer DNA fragments to uniform size, using a 3 ml syringe equipped with a 22 gauge needle. Aliquots, 10 tl, were taken from each control tube containing no drugs, and were precipitated on glass-fiber filters (GF/A) with ice cold 5% TCA, to determine the total number of dpm per sample. To precipitate SDS, 0.25 ml 325 mM KC1 was added and mixed by inversion. The samples were placed on ice for 10 min. Samples were centrifuged for min on a Fisher Scientific micro centrifuge (model 235C) to pellet the precipitate. The supernatant was removed via vacuum aspiration followed by addition of a wash buffer 10 mM Tris (pH 7.5), 2 mM EDTA (pH 8), 100 mM KC1, 0.1 mg/ml calf thymus DNA].
The precipitate was resuspended by vortexing and the samples were placed in a 65C water bath for 10 min. After the precipitate was completely dissolved, samples were placed on ice. After 5 min samples were vortexed, and returned to ice for another 5 min. Microcentrifugation for another minute was followed by removal of supernatant. The remaining pellets were then dissolved in distilled water at 65C and transferred to scintillation vials. Scintiverse BD(R), 4 ml, was added to each scintillation vial and samples were counted on a Packard Tricarb 4000 liquid scintillation spectrometer, and corrected for quenching.

Interference Assay
The interference assay followed the procedures described for protein-linked DNA breaks, only differing in that after the hr incubation with sample compound, an equal volume of VP-16 (80 tM) plus an equimolar concentration of the appropriate sample compound was added to each sample [9]. The cells were allowed to incubate at 37C for an additional hr after the addition of VP-16 solution. The cells were then lysed and processed exactly as described above.

Cleavage-interference in vitro
The method used to assess interference with drug-induced topoisomerase II-mediated DNA cleavage in vitro, was based on the procedure of Harker, et al. [ 10]. Following incubation, cells were pelleted by centrifugation at 600 x g for 5 min. The media was removed via aspiration, and the cell pellet was resuspended in fresh media of a volume equivalent to the original (2 ml). The cells were diluted by serial dilution to a concentration of x 103 per milliliter. The final dilution being made in sterile test tubes with a 0.13% agar solution of growth media, heated to 45C. Agar dilutions were mixed by inversion and placed in test tube racks for incubation at 37C. Following 5 days of incubation, a solution of 0.01% neutral red (200tl) in growth media was gently layered on the surface of each sample to diffuse into the colonies. On day 7, samples were decanted into 6-well plates and colonies were counted using a colony counter over a 2 mm x 2mm grid. Results were expressed as the percent of the number of colonies in untreated control.
Phosphorylation of Topoisomerase II Phosphorylation of L1210 DNA or human topoisomerase II enzyme was measured by the method of DeVore et al. [ 12]. Reactions  The mean and standard deviation are designated by "X + SD." The probable level of significance (p) between test and control samples was determined by the Student's "t" test with the raw data.

Results
Piperidine carboxyborane 1, piperazine carboxyborane 2, morpholine carboxyborane 3, Nmethylmorpholine carboxyborane 4, 4-phenyl piperidine carboxyborane 5, 4-methyl piperidine carboxyborane 6, and N-methyl imidazole carboxyborane 7 were effective cytotoxic agents in the L1210 and HeLa-S tumor screens. All of the EDs0 values were less than 4 pg/ml with the exception of compound 3 in the L1210 screen [ Table 2]. Generally the compounds were more effective in the HeLa-S uterine carcinoma screen and were very comparable with standard antineoplastic agents used in these screens. The compounds were effective in reducing DNA synthesis greater than 50% in both the HeLa-S uterine carcinoma and L1210 lymphoid leukemia cells over 60 min [ Fig. and 2].  Fig. 3 and 4].
Using the partially purified HeLa-S DNA topoisomerase II enzyme, subsequent studies demonstrated that the compounds 1-7 were inhibitors at 200 tM. [Fig 5]. Using the L1210 partially purified DNA topoisomerase enzyme was inhibited at _> 100 tM for the compounds. An in-depth study was performed to determine if the compounds affected DNA topoisomerase II activity in a concentration range consistent with the concentration needed for DNA synthesis inhibition. Compound 2 was selected for this detailed study. affording additional breaks at 15 and 24 hr which were significantly greater in magnitude than with VP-16 alone [ Fig. 8 and 9]. These studies suggested that the heterocyclic amine carboxyboranes functioned The Hind III-cut PBR322 DNA end-labeled with [ct-aZP] dCTP study also confLrmed that compound 2 was not a direct inhibitor of DNA topoisomerase II activity from 1 to 100 lxM, but it does appear to be enhancing the action of VP-16 causing DNA breaks at 10 and 100 tM suggesting that the heterocyclic amines function by some other mechanism than binding to the same site on the enzyme as VP-16 [ Fig  10]. Combination studies of VP-16 with compound 2 demonstrate that use of both agents together produce a significant increases in small fragments of DNA in the gradient [Fig 15]. VP-16 alone caused DNA fragmentation with medium size DNA fragments, whereas when the drug was added in combination this resulted in smaller DNA fragments. Nevertheless, the drug alone was capable of producing small fragments of DNA. The phosphorylation by protein kinase C of L1210 DNA topoisomerase II as well as human DNA topoisomerase II was found to be reduced by compound 2 whereas VP-16 caused an elevation of the phosphorylation by PKC. The magnitude of reduction afforded by compound 2 was similar to known inhibitors of protein kinase C activity [  The heterocyclic amine carboxyborane adducts have more than one mode of action in affording tumor cell death. Obviously the reported inhibition of enzyme activities involved in the regulation of the purine pathway by the compounds from 25 to 100 tM in L1210 cells will result in cell death by reducing deoxyribonucleotide pools ]. DNA strand scission has also been linked to tumor cell death or apoptosis.
Since the L1210 growth study demonstrated that longer than 24 hr was required to observe significant changes in cell number in the presence of these compounds, this suggests that the heterocyclic amine carboxyboranes are altering cellular biochemical events later than 60 min. Thus, there are secondary events afforded by the drugs which are also important in causing cancer cell death. These agents do not appear to be L1210 DNA topoisomerase II inhibitors at 100 tM concentrations consistent with the concentration of drug for the observed inhibition of DNA synthesis and DNA fragmentation. The agents were additive with the effects of VP-16 in inducing DNA protein linked breaks after 6 hr incubation in a concentration dependent manner. The study would suggest that the heterocyclic amine carboxyboranes did not bind to the same site on DNA topoisomerase II enzyme as VP-16 since they produced no cleavage product when used alone from to 24 hr. The agents appeared to be acting in a manner other than VP-16 on the DNA topoisomerase II enzyme to enhance the ability of VP-16 to afford cleavable product, i.e. DNA protein linked breaks. VP-16 is passively taken up by cancer cells but its efflux is by the p-glycoprotein transporter. This transporter can be blocked by a variety of agents, i.e. verapamil, to block the efflux of VP-16 and increase cell death. However, the heterocyclic amine carboxyboranes did not cause increased accumulation of VP-16 in L 1210 cells. Agents which block phosphorylation by PKC of DNA topoisomerase II reduce its catalytic action and cause more DNA protein linked breaks followed by apoptosis [12,[14][15]. These studies have demonstrated that P-16 actually stimulates phosphorylation of L1210 and human DNA topoisomerase II whereas the heterocyclic amine carboxyboranes cause a reduction in phosphorylation of the enzyme which is related to increase in cleavable products and cell death.