In our studies, various stereoisomers of catechin [(+), (-)] and (-)-and (+)-epicatechin have been oligomerized using Horseradish Peroxidase (HRP), derived from the roots of the horseradish plant, in water-ethanol mixtures. A typical enzymatic oligomerization occurs in aqueous media buffered at pH 7 with the monomeric epicatechin, a catalytic amount of the enzyme and hydrogen peroxide, to initiate the reaction (Figure 2).

The reaction mixture was stirred overnight under ambient conditions. The resulting oligomer is separated from the monomer and lower molecular weight compounds through simple methods like dialysis and centrifugation. Figure 3 shows the UV-Visible spectra of (-)-epicatechin monomer and the resulting oligo(epicatechin) synthesized at pH 7.

As seen in the figure, the monomer shows significant absorption in the range of 250-300 nm and no absorption beyond 300 nm. The initiation of oligomerization by the addition of H₂O₂ leads to the appearance of a dark-red brown solution and a new broad absorption peak in the 325-550 nm range with a maxima around 390 nm. Initial MALDI-TOF studies indicate the formation of oligomers (up to 2035 a.m.u) indicating the presence of at least 7 repeat units (data not shown).

Stereoisomers of catechins have been known to exhibit characteristic Circular Dichroism (CD) [19]. CD spectroscopy can also be used to study the stereochemistry and/or the secondary structure of the oligomers. The CD spectra of oligomeric forms of (+) and (-) catechins indicate that the oligocatechins adopt unique secondary structures that are exact mirror images of each other (Figure 4b).

The oligomer of (-)-epicatechin also exhibits a unique CD spectrum [20]. Further studies are underway to deconvolute and carry out specific peak- assignment. We hypothesize that oligocatechins with distinct secondary structures can be capable of mimicking specific ligand(s), accessing growth control pathways with the possibility for enhancing the therapeutic activity.

We have conducted proliferation studies on normal and malignant human breast cancer cells using this new-class of oligoepicatechins. A series of human cancer cell lines (high and low metastatic breast cancer cells, colorectal cancer, and nasopharyngeal cancer) were selected and the efficacy of these oligomeric compounds in inhibiting the growth of these human cancer cells was then analyzed in a dose response study. The experimental means were compared to the means of untreated cells harvested in parallel and the data was pooled for replicate experiments. Among all oligomerized forms of the catechins, oligo(epicatechin) in ethanol (Oligo EC/EtOH) proved to be the most efficient in growth inhibiting the cancer cells without effecting normal cell growth at low doses. At effective doses in which the oligo(EC/EtOH) inhibits the growth of cancer cells, the monomer and HRP did not show any activity. The oligomer was tested over a dose range of 0.1 to 5 μg/ml in parallel with EGCG tested at 5 μg/mL and 9.2 μg/mL. These doses of EGCG were selected based on the literature showing EGCG is not effective at inhibiting human breast cancer cells in-vitro at doses below 5 µg/mL [21]. In our studies, we also found that EGCG at 5 μg/mL was a poor inhibitor of both the breast cancer cell lines (Figure 5a, Figure 5b) and it also growth inhibited normal mammary epithelial cells to the same extent as the cancer cells (Figure 6).

Though there have been numerous reports on the ability of EGCG to inhibit cancer cell growth, the inhibitory activity requires a very high concentration of EGCG (>20 μM). While these concentrations are achievable in-vitro, they may not be achievable in-vivo [22]. EGCG failed to inhibit MCF-7 cells (low metastatic breast adenocarcinoma) (Figure 5a) and only inhibited the high metastatic breast cancer cells by 22% (p<0.05) (Figure 5b) at a 5 µg/mL dose when tested in parallel. This is consistent with previously published results [22]. EGCG also growth inhibited MCF-12A cells (normal mammary epithelial cells) exhibiting no selectivity for cancer cells. EGCG is known to inhibit MCF-7 cells by 38% and MDA-MB-231 cells (high metastatic breast adenocarcinoma) by 34% over a seven day treatment when used at a dose of 9.2 µg/mL. Lower doses were found to be ineffective [23]. Our results therefore indicate that EGCG does not exhibit any therapeutic differential in growth inhibition of cells. In striking contrast, the growth of MCF-12A cells was not significantly affected when treated with our new oligomer namely oligo(EC/EtOH) at most doses (Figure 5c). At the same doses, MCF-7 cells were 75% growth inhibited (Figure 4a) whereas MDA-MB-231 cells show a dose response at 1.0 and 5.0 μg/mL (Figure 5b).

Moreover, the oligoepicatechin we have synthesized enzymatically outperforms and is more selective than EGCG. Furthermore, the oligocatechin exhibits a specific inhibitory effect for cancer cells, is soluble in aqueous solutions and very stable. This new oligomeric compound also retained complete activity after being stored in solution for 3 months. It is worthwhile to point out that EGCG was found to degrade completely by 3 hours when incubated in buffered solutions.

Different strategies have been adopted for the synthesis of oligomeric catechins [24]. Kozikowski et al. have reported the chemical synthesis of epicatechin oligomers and the ability of these oligomers to growth inhibit human breast cancer cell lines (MDA MB 231) [13]. But the dose required for growth inhibition (100 μg/mL) is at least 20 times higher than those observed with our enzymatically synthesized oligomeric epicatechins (>75% inhibition even at 5 µg/mL dose). Recently oligonol, a compound containing catechin-type monomers and oligomeric proanthocyanidins has also been shown to be effective in growth inhibiting MCF-7 and MDA-MB-231 cells at 20 μg/mL [25].

In addition, Oligo(EC/EtOH) was also found to be effective in inhibiting the growth of human colorectal and naso-pharyngeal human cancer cell lines at similar low doses (data not shown). Studies are currently underway to identify the mechanism of growth inhibition of human cancer cells by oligo(epicatechins). It does not appear to be through a mechanism of inducing apoptosis but does achieve a cell cycle arrest in the G2 part of the cell cycle.

Initial NMR studies on the oligomeric mixture were not conclusive. Peroxidase-catalyzed polymerization of phenolic compounds is carried out in the presence of H₂O₂, which acts as an oxidant, and the peroxidase cycle involves a two-electron oxidation step and two one-electron reduction steps. The free radicals formed undergo oxidative coupling to produce dimers and other oligomeric/polymeric products. HRP catalyzed oligomerization reactions are hence quite complex, since the radicals formed can couple in numerous ways. There have been considerable efforts in understanding the nature of coupling in the HRP catalyzed modification of substituted phenols [26] and other stereoisomers of catechin [27]. The reactions have been quenched and even at initial stages of the reaction, multiple products have been isolated and identified using 1D and 2D NMR.

In this research work, we have identified oligo(epicatechin) to be the most potent oligomer from among the family of catechins in inhibiting growth of human breast cancer cells. High Performance Liquid Chromatography (HPLC) studies indicate the oligomer to be a mixture comprising of three different fractions. Based on absorption of the fractions at 210 nm, the lower molecular weight compounds are present to the extent of around 30%. HPLC studies on the HRP catalyzed oligomerization of (+)-catechin has been reported to result in the formation of oligomeric(catechins) as well [28]. At the current time, separation of these oligomeric epicatechin fractions and testing of anti-proliferative activity (both in vitro and in vivo) are currently being pursued. We plan to investigate the structure of the most potent fraction in-vitro and in-vivo. The nature of the coupling and the mechanism of the reaction are under investigation.

Horseradish peroxidase (HRP, EC 1.11.1.7) type II, 150-200 units/mg solid and all catechin monomers was purchased from Sigma chemicals Co. (St. Louis, MO). Hydrogen peroxide (30 wt%) were purchased from Aldrich Chemicals Inc., Milwaukee, WI and was diluted with water to make 0.3% H₂O₂ solution. All other chemicals were purchased from Aldrich and were of reagent grade or better. UV-Visible spectra were obtained using a Perkin-Elmer Lambda 9 UV-Visible-near-IR spectrophotometer. Circular Dichroism (CD) spectra were obtained using a Jasco CD spectrometer J-720.

Synthesis of oligo(epicatechin): The oligomerization reactions were carried out in 95:5 (v/v) mixtures of phosphate buffer (pH 7) and ethanol. For a typical reaction, (-)-epicatechin (3.44 mM) was dissolved in ethanol (500 μL) and then phosphate buffer (8 mL) was added to the above solution. The pH was adjusted to be between 6-6.5. A solution of HRP (4 mg in 0.5 mL deionized water) was to the reaction mixture and the reaction was initiated by the dropwise addition of 0.3% H₂O₂ (1500 μL). The reaction was allowed to proceed for 24 hours. The product formed was dialyzed to remove unreacted monomers and oligomers. The oligomer was then dried under vacuum for 72 hours. The gravimetric yield was between 65-70%. A similar procedure was followed for the oligomerization of other stereoisomers of catechin.

Proliferation assays: Cells (1,000 per well) were plated into 48-well assay plates and proliferation studies were performed using MCF-7 (HTB-22, low metastatic breast adenocarcinoma), MDA-MB-231 (HTB-26, high metastatic breast adenocarcinoma) and MCF-12A [CRL#10782, normal mammary epithelial cells obtained from the American Type Culture Collection (ATCC, Manassas, VA)]. Stock cultures of MCF-7 cells were maintained in MEM with 10% Fetal Calf Serum, 1% fungizone, 1% GPS, NEAA, Sodium pyruvate and bovine insulin. GPS solutions contained 200 mM glutamine, 10K U penicillin, 10 mg/mL streptomycin solution. A stock solution of insulin was prepared (10 mg/mL) and was used at 10 µL/mL. MDA-MB-231 cells were in DMEM with 10% Calf Serum, 1% fungizone and 1% GPS. MCF-12A cells were in DMEM/ F-12 with 10% Horse Serum, 20 ng/ml hEGF, 100 ng/ml cholera toxin, 500 ng/ml hydrocortisone and bovine insulin in DMEM with 10% calf serum containing glutamine and penicillin streptomycin, according to the manufacturers instructions. Colorectal cancer cells HT29 (ATCC, HTB-38) are maintained in McCoy’s 5A medium with 10% fetal bovine serum, 1% fungizone and 1% glutamine, penicillin, streptomycin solution. Nasopharyngeal cancer cells FaDu (ATCC, HTB-43, pharynx squamous cell carcinoma) are maintained in MEM with 10% fetal bovine serum, NEAA, Fgz, and 100 mM sodium pyruvate (diluted 1:100 for cell culture). All these cell lines are maintained in a humidified incubator at 37°C, with 5% CO₂.

T₀ counts were taken at 24 hours after plating, and cells were refed with growth media, with or without vehicle alone, EGCG or oligo(catechins), at various doses. Every three days, in triplicate, per treatment per time point, wells were harvested using trypsinization, and cell numbers determined electronically using a Coulter counter. Remaining wells were re-fed growth media, with or without treatments. Experiments were performed a minimum of three times and only experiments were included where control cells exhibited a minimum of 2.5 cell population doubling over the first 6 day period. The experimental means are compared to the means of untreated cells harvested in parallel and the data is pooled for replicate experiments, standard deviations and statistical differences calculated using student T-tests (p values <0.05(*) or <0.001(**) are considered significant).