Molecular Cancer Therapeutics Molecular Diagnostics in Cancer Therapeutic Development: Fulfilling the Promise of Personalized Medicine Targeting the PI3-Kinase Pathway in Cancer
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
Molecular Cancer Therapeutics 6, 1831-1839, June 1, 2007. doi: 10.1158/1535-7163.MCT-07-0045
© 2007 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wu, F.
Right arrow Articles by Gehring, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wu, F.
Right arrow Articles by Gehring, H.

Research Articles: Therapeutics, Targets, and Development

New transition state–based inhibitor for human ornithine decarboxylase inhibits growth of tumor cells

Fang Wu, Doris Grossenbacher and Heinz Gehring

Department of Biochemistry, University of Zurich, Zurich, Switzerland

Requests for reprints: Heinz Gehring, Department of Biochemistry, University of Zurich, Winterthurerstrasse 190, CH-8057 Zurich, Switzerland. Phone: 41-44-6355572; Fax: 41-44-6355907. E-mail: gehring{at}bioc.unizh.ch

Abstract

Pyridoxal 5'-phosphate (PLP)–dependent ornithine decarboxylase (ODC) is the key enzyme in polyamine synthesis. ODC is overexpressed in many tumor cells and thus a potential drug target. Here we show the design and synthesis of a coenzyme-substrate analogue as a novel precursor inhibitor of ODC. Structural analysis of the crystal structure of human ODC disclosed an additional hydrophobic pocket surrounding the {varepsilon}-amino group of its substrate ornithine. Molecular modeling methods showed favorable interactions of the BOC-protected pyridoxyl-ornithine conjugate, termed POB, in the active site of human ODC. The synthesized and purified POB completely inhibited the activity of newly induced ODC activity at 100 µmol/L in glioma LN229 and COS7 cells. In correlation with the inhibition of ODC activity, a time-dependent inhibition of cell growth was observed in myeloma, glioma LN18 and LN229, Jurkat, COS7, and SW2 small-cell lung cancer cells if DNA synthesis and cell number were measured, but not in the nontumorigenic human aortic smooth muscle cells. POB strongly inhibited cell proliferation not only of low-grade glioma LN229 cells in a dose-dependent manner (IC50 ~50 µmol/L) but also of high-grade glioblastoma multiforme cells. POB is much more efficient in inhibiting proliferation of several types of tumor cells than {alpha}-DL-difluoromethylornithine, the best known irreversible inhibitor of ODC. [Mol Cancer Ther 2007;6(6):1831–9]

Introduction

Polyamines, such as putrescine, spermidine, and spermine, are multicharged aliphatic cations and exist in all organisms (1). Ornithine decarboxylase (ODC), a pyridoxal 5'-phosphate (PLP)–dependent enzyme, catalyzes the conversion of ornithine to putrescine, the first and rate-limiting step in polyamine synthesis (Fig. 1A ; ref. 2). ODC expression is rapidly induced after exposing cells with cell proliferation–stimulating agents (2–4). High levels of ODC observed in cancer cells are closely related to tumor promotion (1, 5). Overexpression of ODC induces transformation of cells (6) and inhibition of ODC abolishes transformation and is associated with tumor suppression (2, 7). Even modest reductions in ODC activity can lead to a marked resistance in tumor development (2, 7). ODC-overexpressing cells are tumorigenic and expression of ODC in hair follicle cells is linked to growth and maintenance of hair follicles (8). Thus, ODC is an interesting target for anticancer or chemopreventive drugs (2, 5).


Figure 1
View larger version (6K):
[in this window]
[in a new window]

  		Figure 1. A, reaction mechanism of ODC with L-ornithine. E-PLP, ODC-PLP (internal aldimine); ES-PLP, enzyme-coenzyme-substrate intermediate (external aldimine). B, POB, the precursor analogue of phosphopyridoxyl-ornithine (see ES-PLP).

 
Several inhibitors of ODC are known and the most widely used compound is {alpha}-DL-difluoromethylornithine (DFMO), an enzyme-activated irreversible inhibitor, which was introduced more than 20 years ago. DFMO irreversibly binds into the active site of ODC by covalently attaching to Cys360 after it is activated by ODC itself (5). Although DFMO has been shown to be effective in some experimental models and therapeutic approaches, it seems not to be as effective in vivo against tumor cells (9). One explanation of why DFMO is not sufficiently blocking ODC activity in vivo might be that DFMO is displaced, due to its relatively low affinity (Ki ~40 µmol/L), by endogenous ornithine (Km ~ 90 µmol/L), the substrate of ODC (5). The concentration of ornithine in human plasma seems to be in the range of this Km value. Fast clearance, inefficient uptake into cells, and development of resistance, such as up-regulation of ODC expression and polyamine transporter, are further drawbacks of DFMO (5, 10, 11). Thus, other ODC inhibitors which have higher potency, better pharmacokinetics, and the ability to overcome the cell resistance are demanded (5, 10). A lot of attempts have been made by modifying the ornithine skeleton structure of DFMO to improve the efficiency (5). Although the modified compounds got improved binding affinity for ODC in vitro, they had no favorable effect in inhibiting cell proliferation in vitro and in vivo (5).

PLP is the functional cofactor of PLP-dependent enzymes which catalyze a wide variety of amino acid transformations (12, 13). An obligatory step in all these reactions is the intermediary formation of an aldimine (i.e., a Schiff base of PLP with the amino group of the amino acid substrate; Fig. 1A; refs. 12, 13). Reduction of the Schiff base produces phosphopyridoxyl-amino acids, which are analogues of the covalent coenzyme-substrate adducts. These compounds are a kind of transition-state analogues that bind to the apoprotein with high affinity (14–16). They cannot be transformed by the enzymes and thus inhibit them very potently. A further advantage of these analogues of coenzyme-substrate adducts is the high degree of specificity. Phosphopyridoxyl-ornithine was found to be a very potent inhibitor of ODC without affecting S-adenosylemethionine decarboxylase (15).

Successful exposure of newly synthesized cellular ODC to phosphopyridoxyl-ornithine would inhibit ODC activity completely and specifically (no mammalian {alpha}-transaminase for ornithine is known) and thus induce the inhibition of cell proliferation. However, the drawback of these compounds in their in vivo application is the impermeability of the cell membrane for these substances due to their negative charges.

Here we report the design and synthesis of a phosphopyridoxyl-ornithine–based precursor inhibitor for human ODC, which can be taken up by cells. N-(4'-Pyridoxyl)-ornithine(BOC)-OMe·HCl (POB; Fig. 1B) inhibited ODC activity in cells and inhibited proliferation of many types of tumor or transformed cells much more effectively than DFMO.

Materials and Methods

Chemicals and Cell Lines
Pyridoxal·HCl was purchased from Merck. H-Orn(BOC)-OMe·HCl and DFMO were from Bachem. L-[1-14C]Ornithine and [3H]thymidine were from American Radiolabeled Chemicals, Inc.

NIH 3T3 cells were purchased from American Type Culture Collection. Glioma cell line LN18, LN229 (both from brain), and primary tumor cells from lung cancer metastasis in brain and glioblastoma multiforme were generous gifts from Dr. K. Frei (Department of Neurosurgery, University Hospital Zurich, Zurich, Switzerland). COS7 cells (monkey kidney, SV40 T antigen transformed) and a murine myeloma cell line were generous gifts from Prof. P. Sonderegger (Department of Biochemistry, University Zürich, Zurich, Switzerland). Jurkat cells (a tumor T lymphoma cell line) and a human small-cell lung cancer cell line (SW2) were kindly provided by Dr. J. Kemler-Carraneo (Abteilung Klinische Immunologie, University Hospital Zurich, Zurich, Switzerland). Human aortic smooth muscle cells were a gift from Dr. C. Dumrese (Anatomische Institut, Universität Zürich, Zurich, Switzerland).

Synthesis of POB
POB (Fig. 1B) was synthesized as described for the synthesis of phosphopyridoxyl-ornithine or pyridoxyl-ornithine (15). Briefly, pyridoxal (1 mmol) and KOH (2 mmol) were dissolved in 5-mL methanol as was H-Orn(BOC)-OMe·HCl (1 mmol) and KOH (1 mmol) in 8-mL methanol. Both solutions were mixed at 0°C and stirred at room temperature for 20 min. Then, 125 mg of NaBH4 were added in small portions and the solution was put on ice for further 30 min. Acetic acid (100%) was added until pH 5 was reached to stop the reaction. The solvents were removed with a vacuum dryer and the residue was dissolved in water and subjected to high-performance liquid chromatography for purification and analysis.

High-Performance Liquid Chromatography Analysis and Purification of POB
The compound was analyzed and purified with a Jasco high-performance liquid chromatography equipped with a Nova Pack C18 column (3.9 x 150 mm, 5 µm; Millipore) and a multiwavelengh detector. The sample was loaded and separated with a flow rate of 0.5 mL/min using the following gradient: 0 to 12 min, 100% solvent A (3% acetonitrile/0.1% trifluoroacetic acid) and 0% solvent B (80% acetonitrile/0.1% trifluoroacetic acid); 12 to 22 min, 0% to 80% B; 22 to 92 min, 80% to 100% B, followed by 100% B. The detection was carried out at 290 and 215 nm. Under these conditions, the synthetic compounds eluted at a retention time of 21.4 min and the mass spectrophotometric peak of this purified compound was 398.22 Da (purity >95%; Supplementary Fig. S1).1

The extinction coefficient used for this compound was 8.400 (mmol/L)–1 cm–1 at 296 nm under acidic conditions (17). A semi-prep column (Aquapore octyl, 10 x 250 mm, 20 µm) was used under similar conditions to prepare larger amounts of POB. The purified POB was vacuum dried several times after adding water and stored at a concentration of ~10 mmol/L (water stock solution) in –20°C.

Cell Culture
LN18, LN229, and primary tumor cells from lung cancer metastasis in brain and glioblastoma multiforme were maintained in DMEM supplemented with 1 g/L glucose, 10% fetal bovine serum (FBS; Life Technologies), and 20 µg/mL gentamicin (Fluka) in a humidified 5% CO2 atmosphere at 37°C. Jurkat cells were grown in RPMI 1640 (Sigma) supplemented with 10% NBCS (Life Technologies), 15 mmol/L HEPES (Life Technologies), 2 mmol/L L-glutamine, 50 µmol/L ß-mercaptoethanol, and 1% (w/v) penicillin and streptomycin (Life Technologies) in a humidified 5% CO2 atmosphere at 37°C. SW2 cells were grown in RPMI 1640 supplemented with 10% FBS. NIH 3T3 and myeloma cells were maintained in DMEM with 10% NBCS or FBS. The human aortic smooth muscle cells were cultured in DMEM containing 10% FBS, 20 mmol/L HEPES buffer solution in a humidified atmosphere (5% CO2).

For measuring cell growth, cells (2.5 x 104–5 x 104 per well) were seeded and incubated in 0.5-mL medium containing 5% FBS or NBCS and 1% penicillin/streptomycin in a 24-well plate. After 1 day, cells were incubated in the absence or presence of the indicated concentrations of POB or DFMO for the indicated times. Cells were counted with a Coulter counter (ZM Coulter Electronics) as described (18). Data in the figures represent a typical experiment of at least two independent experiments.

[3H]Thymidine Incorporation
Cells were seeded and pretreated as described above. Then, [3H]thymidine (0.25 µCi) was added into each well and incubated together with cells for the indicated times. Radioactivity of incorporated [3H]thymidine was measured in a Wallac 1450 MicroBeta liquid scintillation counter according to the procedure described by Koga et al. (19). The numbers in the figures are the mean of at least three wells.

ODC Activity
ODC activity was measured by the released 14CO2 from labeled ornithine as described (20). For measuring the serum-induced ODC activity, cells were cultured in a six-well plate at a density of ~1.6 x 105 per well with 5% FBS or NBCS for 1 day and then treated in the absence or presence of POB for the indicated times. The cells were then induced by fresh medium containing 10% FBS for 6 h, collected after trypsine/EDTA treatment, and washed twice with ice-cold PBS (pH 7.4). The collected cells were lysed twice by freezing (liquid nitrogen) and thawing (37°C, 2 min) in 320-µL lysis buffer [1 mmol/L EDTA, 100 µmol/L pyridoxal phosphate, 2.5 mmol/L DTT, and protease inhibitor cocktail (Roche) in 50 mmol/L PBS (pH 7.4)]. The reaction was started by adding 30 µL of L-[1-14C]ornithine (100 µmol/L, 0.09 µCi) to 270-µL supernatant of lysed cells. After 75-min incubation, the reaction was stopped by injecting 200 µL of 4 N H2SO4 and kept for 1 h at room temperature to ensure complete absorption of released CO2 in the capture reagent (tissue solubilizer NCSII, Amersham). The captured 14CO2 was measured by scintillation counting using a Wallac 1450 MicroBeta liquid scintillation counter. The protein concentration in the lysed supernatant was determined with the Bio-Rad protein assay kit. The numbers in the figures are the mean of at least two wells.

Western Blotting
LN229 cells (~4 x 105) were seeded in 25-cm2 flasks and treated in the absence or presence of POB for 3 days and lysed as described above. Cell lysates were fractionated on 10% SDS-PAGE and subjected to immunoblotting analysis as described (21).

Polyamine Analysis
LN229 cells were seeded in a six-well plate at a density of ~1.6 x 105 per well with 5% FBS for 1 day and then treated in the absence or presence of POB (100 µmol/L) for the indicated times. Cells were collected, counted, and lysed. Polyamines were extracted with 0.2 mol/L perchloric acid and measured by Dr. L. Persson and coworkers (Department of Experimental Medical Research, Lund University, Lund, Sweden) as described (22). The numbers in the figures are the mean of triplicates.

Polyamine Oxidase
Polyamine oxidase activity was determined as described (23). LN229 cells were seeded in a six-well plate at a density of ~1.6 x 105 per well with 5% FBS for 1 day and treated in the absence or presence of POB (200 µmol/L) or DFMO (500 µmol/L) for 2 days. Then, cells were lysed and polyamine oxidase activities measured. The numbers in the figures are the mean of triplicates.

Molecular Modeling
The model containing human ODC, PLP, and putrescine was generated by superposition of the crystal structure of the human ODC (PDB code: 1D7K; Fig. 2A ; ref. 24) with the crystal structure of Trypanosoma brucei ODC (PDB code: 1F3T; ref. 25) containing PLP and putrescine as ligands. To optimize the structure of the complex formed with PLP, putrescine, and human ODC protein, energy minimization was done by using InsightII software (version 2000, Accelrys Corp.) under amber force field. The putative binding sites of human ODC in this model were analyzed with InsightII binding site module and prepared for further docking analysis by using DOCK4 (USCF) without considering the solvent effect. DOCK4 was used for predicting the geometry of the ligands bound to the protein (26). Electrostatic and van der Waals energies were evaluated on a grid with a 0.3-Å spacing in the region spanning all residues within 6 Å of phosphopyridoxyl-putrescine adduct (see Figs. 1A and 2B). The structures of phosphopyridoxyl-ornithine (Schiff base analogue; Fig. 1) and the putative active form of POB (Fig. 2C) were drawn based on the known three-dimensional coordinates of the phosphopyridoxyl-putrescine conjugate followed by energy minimization. The optimized structures were rigidly docked into the modeled active site of ODC using the program DOCK4 and the complexes of ODC and analogues were further optimized by energy minimization (InsightII; Fig. 2C). The de novo design program LUDI (27) of InsightII was used to retrieve putative fragments which can fit into the binding pocket (Fig. 2B). The standard default variable and a fragment library supplied with the program were used for the LUDI search.


Figure 2
View larger version (49K):
[in this window]
[in a new window]

  		Figure 2. Modeling the phosphopyridoxyl-ornithine analogues in the active site of human ODC. A, stereoview of the dimeric human ODC. Green, subunit A; blue, subunit B; yellow, cofactor PLP in the active site of human ODC shown in CPK mode. B, the Connolly surface of part of the active site is displayed as gray solid surface (generated by residues TyrA331, TyrA389, AlaA393, TyrB323, and PheB397) together with the reaction intermediate phosphopyridoxyl-putrescine adduct (see Fig. 1A) in yellow (phosphopyridoxyl moiety) and green (putrescine moiety). Black circle, additional pocket. C, stereoview of the putative binding mode of the active form of POB in the active site of human ODC. The Connolly surface of whole active site is displayed by dots. Blue, positive charged surface; red, negative; white, neutral. The phosphopyridoxyl-ornithine-BOC conjugate is displayed in yellow (phosphopyridoxyl moiety) and green (ornithine-BOC moiety).

 
Results

Design and Modeling of an Inhibitor for Human ODC
The phosphopyridoxyl-ornithine precursor, which we designed, synthesized, purified (see Materials and Methods), and expected to be able to pass the cell membrane, is shown in Fig. 1B. In this pyridoxyl-ornithine derivative, the 5'-phosphate group is eliminated and the carboxyl group is esterified with methanol. The crystal structure of human ODC (PDB code: 1D7K) contains the cofactor PLP but no substrate or product (24). To study the potential interaction between inhibitors and human ODC, we built a model composed of the reaction intermediate phosphopyridoxyl-putrescine adduct and the active site of human ODC (Figs. 1A and 2B; see Materials and Methods). Based on this model and by using the binding site analysis module of InsightII, we found a hydrophobic pocket, where the {varepsilon}-amino group of ornithine was situated, which is formed by the aromatic residues TyrA389, TyrA331, PheB397, and TyrB323 between the two ODC subunits (Fig. 2B and C). Only the dimer of ODC is catalytically active, as the active sites are constructed of residues from subunits A and B. Applying the structure-based drug design software LUDI (InsightII) indicated that fragments such as 1-methylaminoethanol, 2-aminopyrrole, triethylamine, and trimethylamine would fit into this additional pocket. Indeed, when the hydrophobic BOC group was added to the {varepsilon}-amino group of ornithine (Fig. 1B), the interaction was more favorable if analyzed by the software DOCK or InsightII.

Inhibition of Cellular ODC Activity
To prove whether POB (for synthesis and purification, see Materials and Methods) could enter cells and is there converted to an active inhibitor of newly expressed ODC activity, COS7 and glioma LN229 cells were subjected to POB. In COS7 cells, nearly complete inhibition of the activity of ODC was measured after 2 days of POB treatment at a concentration of 100 µmol/L (Fig. 3A ). A strong inhibition was also observed after 3 days of treatment of LN229 cells with POB at 100 µmol/L, comparable with the inhibition of DFMO at a concentration of 500 µmol/L (Fig. 3B). To show whether the diminished ODC activity was indeed due to the inhibition of enzymic activity and not to a decreased expression of ODC, Western blot was done to estimate the amount of ODC in LN229 and COS7 cells after treating them with POB for 3 or 2 days, respectively (Fig. 3C; for COS7, see Supplementary Fig. S2). The result showed that the amount of ODC (53 kDa) was similar in POB-treated and control cells. Although ODC activity was almost completely inhibited by POB, the amount of ODC seemed not to be affected by the inhibition. Thus, POB was taken up by cells and acted as proposed as an efficient inhibitor of ODC in cells.


Figure 3
View larger version (16K):
[in this window]
[in a new window]

  		Figure 3. Inhibition of ODC activity in cells by POB. A, COS7 cells were treated with POB (100 µmol/L) for 2 d. Cells were collected and lysed, and ODC activity was measured immediately by the release of labeled CO2 from ornithine (see Materials and Methods). B, LN229 cells were treated with POB (100 µmol/L) or DFMO (500 µmol/L) for the indicated times. Cells were collected, lysed, and ODC activity of untreated cells (gray columns) or treated cells (black columns) was measured immediately. C, Western blot of ODC after treatment of LN229 cells with POB (100 µmol/L) for 3 d. Cells were collected, lysed, and 2-fold Laemmli buffer was added to aliquots of the supernatant. The same amount of total proteins (47 µg) was subjected to SDS-PAGE followed by immunoblotting and Ponceau S staining (see Materials and Methods). The main band of 53 kDa represents ODC (Control 100%, Treated 106%). Bars, SD.

 
Effect of POB on Cellular DNA Synthesis
As POB was found effective in inhibiting activity of ODC, and ODC is associated with cell proliferation and tumor growth (see Introduction), the effect of POB on cancer cell proliferation was investigated with cell lines such as myleoma, human small-cell lung cancer cells (SW2), glioma LN229 cells, and, recently, isolated primary tumor cells derived from lung cancer metastasis in brain, as well as glioblastoma multiforme primary tumor cells (see Materials and Methods). The result showed that in these cells, the DNA synthesis was strongly inhibited on POB treatment (Fig. 4 ). Myeloma cells were highly sensitive to POB (Fig. 4A). Even after adding 50 µmol/L POB only for 1 day, DNA synthesis measured after a 13-h pulse with [3H]dThd was less than 20% of untreated cells. In SW2 cells, the DNA synthesis of cells treated with 200 µmol/L POB was reduced to ~15%, whereas the incubation with DFMO at the same concentration had no influence. DNA synthesis of LN229 was decreased to ~25% of the control when exposing them to 100 µmol/L POB for 1 day (Fig. 4C) and that of primary tumor cells from lung cancer metastasis in brain and glioblastoma multiforme cells to ~33% and 60% of the control, respectively, after 4 days of treatment (Fig. 4D).


Figure 4
View larger version (21K):
[in this window]
[in a new window]

  		Figure 4. Inhibition of cell DNA synthesis by POB. A, myeloma cells were pretreated with POB at the indicated concentrations for 1 d; then, [3H]dThd was added, incubated for 13 h, and the incorporated [3H]dThd was measured (see Materials and Methods). B, SW2 cells were pretreated with POB or DFMO for half a day and then incubated with [3H]dThd for 25 h. C, LN229 cells were pretreated with 100 µmol/L POB for 1 d, and [3H]dThd was added and incubated for the indicated times. D, lung cancer metastasizing cells in brain (LCMB) or glioblastoma multiforme cells (GBM) were pretreated with 100 µmol/L POB for 4 d, and [3H]dThd was added and incubated for the indicated times. Gray columns, without treatment; black columns, with POB treatment. Bars, SD.

 
Inhibiting Tumor Cell Growth
Cell growth of several different types of cell lines in culture was investigated by counting the cell number. After treating them with POB or DFMO for the indicated time and concentration (Fig. 5A–C ), POB at 100 µmol/L was a stronger inhibitor of cell growth than DFMO even at 5- or 10-fold higher concentrations in LN229 and COS7 cells, respectively. Whereas in myeloma cells, POB acted at the indicated concentration without a delay; in other cells, growth decreased drastically after a delay of 1 or 2 days. POB also inhibited other cells from different tissues, for instance, Jurkat (T cell line), LN18 (brain), and NIH 3T3 (fibroblast) cells, whereas nontumorigenic human aortic smooth muscle cells were resistant to inhibition with POB (Fig. 5D; the cell number of human aortic smooth muscle cells increased 2.5-fold during the indicated times). The dose and time dependence of inhibition of cell growth by POB varied and was dependent on the cells used and the treatment period. For LN229 cells, the IC50 value is ~50 µmol/L when estimated by [3H]dThd incorporation or by cell counting after 5 or 6 days of POB exposure (Fig. 6 , data of 6-day treatment not shown). The cell inhibition with DFMO was less effective than that with POB at the indicated concentrations (Fig. 6).


Figure 5
View larger version (13K):
[in this window]
[in a new window]

  		Figure 5. Cell growth inhibition of POB and DFMO. Myeloma cells (A), glioma LN229 cells (B), and COS7 cells (C) were treated without (bullet) or either with POB ({blacksquare}, 100 µmol/L) or with DFMO ({blacktriangleup}, 100 µmol/L for myeloma cells, 500 µmol/L for LN229, and 1 mmol/L for COS7 cells) for the indicated times. D, time-dependent inhibition of primary cultures of human aortic smooth muscle cells ({blacksquare}) and of LN18 ({blacktriangleup}), NIH 3T3 (bullet), and Jurkat ({blacktriangledown}) cells subjected to 100 µmol/L POB. Bars, SD.

 

Figure 6
View larger version (11K):
[in this window]
[in a new window]

  		Figure 6. Dose-dependent cell growth inhibition by POB and DFMO. Glioma LN229 cells were incubated with different concentrations of POB ({blacksquare}) or with DFMO ({square}) for 5 d, counted (—), and compared with the untreated controls. For measuring [3H]dThd incorporation (- - -), LN229 cells were incubated with [3H]dThd for 15 h after 2 d of pretreatment with POB (bullet) or DFMO ({circ}). Bars, SD.

 
Treatment of cells with POB affects the morphology of cells in different ways. Only few dead cells were found in untreated LN229 cells but a high fraction of cells detached after treatment with POB (100 µmol/L) for 3 days. DFMO-treated (500 µmol/L) LN229 cells showed higher cell density with less detached cells (Supplementary Fig. S3A). The number of detached cells after POB treatment seemed to depend on the dose of POB. If COS7 cells were treated with POB, many cells died and detached (round floating cells) and a substantial percentage (~40%) of attached COS7 cells could be stained with trypan blue indicating dying cells (Supplementary Fig. S3B). Only few dead cells were found among untreated attached COS7 cells if stained with trypan blue (data not shown). SW2 cells tend to aggregate and form large clumps. As shown by the inhibition of DNA synthesis, treatment with POB inhibited cell proliferation of SW2 and substantially decreased the size and amount of cell aggregates (Supplementary Fig. S3C). The microscopic cell images indicate that inhibition of ODC arrested cell proliferation, which is followed by cell death. The change in morphology of cells might reflect the importance of ODC activity for maintaining the cell shape (28, 29).

Discussion

The present study describes the successful action of POB, a newly developed inhibitor of ODC, on various tumor cells. The development strategy bases on the external aldimine, an obligatory intermediate in the mechanism of action of ODC. In the pyridoxyl-ornithine derivative POB, the 5'-phosphate group is eliminated and the carboxyl group is esterified with methanol. The rationale behind the design was that this compound together with the additional hydrophobic BOC group of the {varepsilon}-amino group of ornithine has no negative net charge and very likely crosses the cell membrane as it was reported for several pyridoxyl-amines (30, 31) and pyridoxyl-methionine analogues (32). Intracellular pyridoxal kinase, an enzyme that seems to tolerate considerable variation in C4' position of the pyridoxyl moiety (31, 33), is expected to phosphorylate this compound after it has been taken up by the cells whereas the carboxymethyl ester will be hydrolyzed intracellularly. The phosphopyridoxyl-ornithine analogue formed in cells will bind with very high affinity to the newly synthesized apo-ODC and affecting only proliferating cells in which ODC is obligatorily induced.

POB, the synthesized precursor of a covalent coenzyme-substrate adduct of ODC, is taken up by the cells. Intracellularly, the compound is transformed and binds, as deduced and shown by the inhibition of ODC activity, to newly synthesized ODC. The response time of POB varied in different cell lines; with LN229 cells, a lag of more than 1 day had been observed. The delayed inhibition might be due to the necessary transformation, i.e., ester hydrolysis and phosphorylation (as discussed above), of POB to an efficient ligand of ODC, processes that might well vary in different cells. Additionally, one has to consider that the active compound has to compete with intracellular pyridoxal phosphate (~20–50 µmol/L; ref. 34), a necessity for a cell to survive. Although the prominent ODC inhibitor DFMO acts directly with active ODC, less pronounced growth inhibition effects were observed even at higher concentrations of DFMO. The treatment of cells with POB neither up-regulates nor down-regulates significantly ODC expression whereas an up-regulation was reported with DFMO (5) and was also observed in our hands (Supplementary Fig. S2).

Consistent with the inhibition of ODC activity by POB is the inhibition of cell proliferation in the investigated tumor cells. Again, cells respond differently with dose and time of exposure to POB. The IC50 of POB is, under serum culture condition, in the micromolar range (50 µmol/L for LN229). Myeloma cells respond immediately, and COS7 or LN229 cells stopped cell proliferation after exposing them to POB for 1 to 2 days, whereas human aortic smooth muscle cells are not significantly affected by POB.

The proposed strategy suggests that the intracellularly transformed POB derivative binds most likely only to newly synthesized apo-ODC because competing bound PLP out of holo-ODC might be very unlikely to happen in the short life cycle of ODC due to the intrinsic high affinity of PLP. In addition to that, the transformed POB derivative seems to bind specifically ODC and does not inhibit human histidine decarboxylase, another highly induced PLP-dependent enzyme in cells (data not show).

DFMO, however, will interact only with active ODC. Adding POB or DFMO to cell extracts and measuring subsequently ODC activity showed indeed that DFMO (500 µmol/L) inhibited ODC activity whereas POB (100 µmol/L) had only a minor effect (data not shown). Although DFMO was able to suppress the activity of ODC in vitro and in cells, it was less efficient in preventing cancer cell proliferation. This indicates a different mechanism of action of these two inhibitors in inhibiting cell proliferation. This conclusion is further supported by the observation that cells supplied with excess of polyamines in culture medium responded differently if treated with POB or DFMO, respectively. The inhibition of DFMO-treated cells was reversed by addition of polyamines (Fig. S4), an already well-known effect (35, 36). Conversely, POB-treated cells did not show much differences in the presence or absence of additional polyamines (Fig. S4).

This different behavior could be explained by the observed intracellular polyamine pool. In POB-treated cells, putrescine declined at most after 2 days, when the inhibition of ODC caused by POB became effective (Supplementary Fig. S5 and Fig. 3B). However, the polyamine level is not lower than that of the controls, whereas DFMO was reported to deplete the putrescine, spermidine, but not spermine, pool already after a day (37) and was also observed in our hands with LN229 that were treated with 500 µmol/L DFMO for 2 days (putrescine, 0.09 nmol/106 cells; spermidine, 0.06 nmol/106 cells; spermine, 2.5 nmol/106 cells). This clearly indicates that the inhibitory action caused by POB is different from that of DFMO and is not simply caused by putrescine and spermidine depletion. Possible explanations could be that POB derivatives might act as a kind of polyamine analogue and interfere with the polyamine conversion (as indicated by a 2-fold increase of polyamine oxidase activity; see Supplementary Fig. S6) and transport, a complex interplay whose regulation is not yet understood (23, 38, 39).

Modeling the inhibitor into the active site of human ODC revealed a relatively large hydrophobic pocket formed by the two subunits of human ODC at the position of the {varepsilon}-amino group of ornithine, suggesting to us not to remove the BOC-protecting group of POB (Fig. 1B). And indeed, the BOC group contributes substantially to the successful inhibitions as we could not find a comparable inhibition if it was removed. Besides the favorable binding suggested from the molecular modeling experiment, the improved bioavailability might also contribute to the efficiency of POB. The calculated logP (log octanol/water partition coefficient, which is a theoretical tool for measuring the lipophilicity of a compound) of POB is 1.72 and without the BOC group, –0.50,2 which is a very strong indication that POB can more efficiently cross the cell membrane (40).

The attempt to develop a cell growth inhibitor by targeting ODC was successful and proved the proposed strategy. The newly designed and synthesized transition state-based analogue (POB) was very effective in inhibiting newly induced cellular ODC activity and exceeded the potency of DFMO, the best and well-characterized inhibitor of ODC thus far, in inhibiting proliferation of many types of tumor cells. POB inhibits the proliferation of a broad variety of tumor cell lines, such as myeloma, glioma LN18 and LN229, Jurkat, COS7, and SW2 cells, but not the nontumorigenic human aortic smooth muscle cells. Glioblastoma multiforme is the most aggressive form of primary brain tumors known collectively as gliomas. Very few therapeutic options exist for the treatment of human glioblastoma (41). Here, we report that POB can suppress the proliferation not only of low-grade (IC50 ~50 µmol/L) but also of high-grade glioblastoma cells. Future experiments will have to show whether the promising results with POB in culture studies can be confirmed in animal or even clinical studies. Structure modification of this inhibitor might improve its action and can be proposed and done on the basis of the interaction model of POB with ODC.

In conclusion, the present study shows that selected PLP-dependent enzymes in cells can be targeted with the proposed transition state-based inhibitors, and the designed strategy might serve to develop inhibitors of this type for other PLP-dependent enzymes of pharmacologic interest.

Acknowledgments

We thank Dr. S. Bienz (Institute of Organic Chemistry, University Zurich) for providing excellent chemistry equipments; Dr. L. Persson and coworkers for measuring the polyamines, Dr. K. Frei for kindly providing glioma cell lines, primary tumor cells from lung cancer metastasis in brain and from glioblastoma multiforme; Dr. C. Dumrese and coworkers for providing the experiments with human aortic smooth muscle cells; and Dr. P. Christen (Department of Biochemistry, University Zurich, Zurich, Switzerland) for his valuable advices and critical reading of the manuscript.

We would like to dedicate this article to Livio Besio who did initial work but died in an accident.

Footnotes

Grant support: COST Switzerland, Action 922, grant C02.0017.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 Supplementary data for this article are available at Molecular Cancer Therapeutics Online (http://mct.aacrjournals.org/). Back

2 Calculated values were obtained from http://www.logp.com. Back

Received 1/19/07; revised 3/27/07; accepted 5/ 2/07.

References

  1. Gerner EW, Meyskens FL, Jr. Polyamines and cancer: old molecules, new understanding. Nat Rev Cancer 2004;4:781–92.[CrossRef][Medline]
  2. Pegg AE. Regulation of ornithine decarboxylase. J Biol Chem 2006;281:14529–32.[Abstract/Free Full Text]
  3. Nickel KP, Belury MA. Inositol hexaphosphate reduces 12-O-tetradecanoylphorbol-13-acetate-induced ornithine decarboxylase independent of protein kinase C isoform expression in keratinocytes. Cancer Lett 1999;140:105–11.[CrossRef][Medline]
  4. Glikman P, Manni A, Demers L, Bartholomew M. Polyamine involvement in the growth of hormone-responsive and -resistant human breast cancer cells in culture. Cancer Res 1989;49:1371–6.[Abstract/Free Full Text]
  5. Seiler N. Thirty years of polyamine-related approaches to cancer therapy. Retrospect and prospect. Part 1. Selective enzyme inhibitors. Curr Drug Targets 2003;4:537–64.[CrossRef][Medline]
  6. Auvinen M, Paasinen A, Andersson LC, Holtta E. Ornithine decarboxylase activity is critical for cell transformation. Nature 1992;360:355–8.[CrossRef][Medline]
  7. Nilsson JA, Keller UB, Baudino TA, et al. Targeting ornithine decarboxylase in Myc-induced lymphomagenesis prevents tumor formation. Cancer Cell 2005;7:433–44.[CrossRef][Medline]
  8. Wheeler DL, Ness KJ, Oberley TD, Verma AK. Inhibition of the development of metastatic squamous cell carcinoma in protein kinase C epsilon transgenic mice by {alpha}-difluoromethylornithine accompanied by marked hair follicle degeneration and hair loss. Cancer Res 2003;63:3037–42.[Abstract/Free Full Text]
  9. Vlastos AT, West LA, Atkinson EN, et al. Results of a phase II double-blinded randomized clinical trial of difluoromethylornithine for cervical intraepithelial neoplasia grades 2 to 3. Clin Cancer Res 2005;11:390–6.[Abstract/Free Full Text]
  10. Casero RA, Jr., Frydman B, Stewart TM, Woster PM. Significance of targeting polyamine metabolism as an antineoplastic strategy: unique targets for polyamine analogues. Proc West Pharmacol Soc 2005;48:24–30.[Medline]
  11. Marton LJ, Pegg AE. Polyamines as targets for therapeutic intervention. Annu Rev Pharmacol Toxicol 1995;35:55–91.[CrossRef][Medline]
  12. Eliot AC, Kirsch JF. Pyridoxal phosphate enzymes: mechanistic, structural, and evolutionary considerations. Annu Rev Biochem 2004;73:383–415.[CrossRef][Medline]
  13. Christen P, Mehta PK. From cofactor to enzymes. The molecular evolution of pyridoxal-5'-phosphate-dependent enzymes. Chem Rec 2001;1:436–47.[CrossRef][Medline]
  14. Raso V, Stollar BD. The antibody-enzyme analogy. Characterization of antibodies to phosphopyridoxyltyrosine derivatives. Biochemistry 1975;14:584–91.[CrossRef][Medline]
  15. Heller JS, Canellakis ES, Bussolotti DL, Coward JK. Potent inhibition of ornithine decarboxylase by N-(5'-phosphopyridoxyl)-ornithine. Biochim Biophys Acta 1975;403:197–207.[Medline]
  16. Khomutov RM, Dixon HB, Vdovina LV, et al. N-(5'-Phosphopyridoxyl)glutamic acid and N-(5'-phosphopyridoxyl)-2-oxopyrrolidine-5-carboxylic acid and their action on the apoenzyme of aspartate aminotransferase. Biochem J 1971;124:99–106.[Medline]
  17. Elbert A, Peterson HAS. Preparation of Crystalline Phosphorylated Derivatives of Vitamin B6. J Am Chem Soc 1954;76:169–75.[CrossRef]
  18. Geistlich A, Gehring H. CDGF (chicken embryo fibroblast-derived growth factor) is mitogenically related to TGF-ß and modulates PDGF, bFGF, and IGF-I action on sparse NIH/3T3 cells. Exp Cell Res 1993;204:329–35.[CrossRef][Medline]
  19. Koga T, Nakano S, Nakayama M, et al. Identification and partial purification of a low-molecular-weight growth inhibitor formed by density-inhibited, tumorigenic V79 Chinese hamster cells. Cancer Res 1986;46:4431–7.[Abstract/Free Full Text]
  20. Gerhauser C, Mar W, Lee SK, et al. Rotenoids mediate potent cancer chemopreventive activity through transcriptional regulation of ornithine decarboxylase. Nat Med 1995;1:260–6.[CrossRef][Medline]
  21. Belyanskaya LL, Delattre O, Gehring H. Expression and subcellular localization of Ewing sarcoma (EWS) protein is affected by the methylation process. Exp Cell Res 2003;288:374–81.[CrossRef][Medline]
  22. Nasizadeh S, Myhre L, Thiman L, et al. Importance of polyamines in cell cycle kinetics as studied in a transgenic system. Exp Cell Res 2005;308:254–64.[CrossRef][Medline]
  23. Dai H, Kramer DL, Yang C, et al. The polyamine oxidase inhibitor MDL-72,527 selectively induces apoptosis of transformed hematopoietic cells through lysosomotropic effects. Cancer Res 1999;59:4944–54.[Abstract/Free Full Text]
  24. Almrud JJ, Oliveira MA, Kern AD, et al. Crystal structure of human ornithine decarboxylase at 2.1 A resolution: structural insights to antizyme binding. J Mol Biol 2000;295:7–16.[CrossRef][Medline]
  25. Jackson LK, Brooks HB, Osterman AL, Goldsmith EJ, Phillips MA. Altering the reaction specificity of eukaryotic ornithine decarboxylase. Biochemistry 2000;39:11247–57.[CrossRef][Medline]
  26. Kuntz ID. Structure-based strategies for drug design and discovery. Science 1992;257:1078–82.[Abstract/Free Full Text]
  27. Bohm HJ. A novel computational tool for automated structure-based drug design. J Mol Recognit 1993;6:131–7.[CrossRef][Medline]
  28. Heiskala M, Zhang J, Hayashi S, Holtta E, Andersson LC. Translocation of ornithine decarboxylase to the surface membrane during cell activation and transformation. EMBO J 1999;18:1214–22.[CrossRef][Medline]
  29. Pomidor MM, Ruhl KK, Zheng P, et al. Relationship between ornithine decarboxylase and cytoskeletal organization in cultured human keratinocytes: cellular responses to phorbol esters, cytochalasins, and {alpha}-difluoromethylornithine. Exp Cell Res 1995;221:426–37.[CrossRef][Medline]
  30. Zhang Z, McCormick DB. Uptake and metabolism of N-(4'-pyridoxyl)amines by isolated rat liver cells. Arch Biochem Biophys 1992;294:394–7.[CrossRef][Medline]
  31. Zhang ZM, McCormick DB. Uptake of N-(4'-pyridoxyl)amines and release of amines by renal cells: a model for transporter-enhanced delivery of bioactive compounds. Proc Natl Acad Sci U S A 1991;88:10407–10.[Abstract/Free Full Text]
  32. Ogier G, Chantepie J, Deshayes C, et al. Contribution of 4-methylthio-2-oxobutanoate and its transaminase to the growth of methionine-dependent cells in culture. Effect of transaminase inhibitors. Biochem Pharmacol 1993;45:1631–44.[CrossRef][Medline]
  33. Tang L, Li MH, Cao P, et al. Crystal structure of pyridoxal kinase in complex with roscovitine and derivatives. J Biol Chem 2005;280:31220–9.[Abstract/Free Full Text]
  34. Robson LC, Schwarz MR. in Vitamin B6 metabolism and role in growth. Tryfiates GP, editor. Westport (CT): Food and Nutrition Press; 1980. p. 205.
  35. Wallace HM, Fraser AV. Inhibitors of polyamine metabolism: review article. Amino Acids 2004;26:353–65.[Medline]
  36. Delcros JG, Tomasi S, Carrington S, et al. Effect of spermine conjugation on the cytotoxicity and cellular transport of acridine. J Med Chem 2002;45:5098–111.[CrossRef][Medline]
  37. Qu N, Ignatenko NA, Yamauchi P, et al. Inhibition of human ornithine decarboxylase activity by enantiomers of difluoromethylornithine. Biochem J 2003;375:465–70.[CrossRef][Medline]
  38. Vujcic S, Halmekyto M, Diegelman P, et al. Effects of conditional overexpression of spermidine/spermine N1-acetyltransferase on polyamine pool dynamics, cell growth, and sensitivity to polyamine analogs. J Biol Chem 2000;275:38319–28.[Abstract/Free Full Text]
  39. Duranton B, Holl V, Schneider Y, et al. Cytotoxic effects of the polyamine oxidase inactivator MDL 72527 to two human colon carcinoma cell lines SW480 and SW620. Cell Biol Toxicol 2002;18:381–96.[CrossRef][Medline]
  40. Eugene Kellogg G, Abraham DJ. Hydrophobicity: is LogP(o/w) more than the sum of its parts? Eur J Med Chem 2000;35:651–61.[CrossRef][Medline]
  41. Hui AM, Zhang W, Chen W, et al. Agents with selective estrogen receptor (ER) modulator activity induce apoptosis in vitro and in vivo in ER-negative glioma cells. Cancer Res 2004;64:9115–23.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 FASEB J.Home page
F. Wu, J. Yu, and H. Gehring
Inhibitory and structural studies of novel coenzyme-substrate analogs of human histidine decarboxylase
FASEB J, March 1, 2008; 22(3): 890 - 897.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wu, F.
Right arrow Articles by Gehring, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wu, F.
Right arrow Articles by Gehring, H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online