Full-term pregnancy early in reproductive life is protective against breast cancer in women. The protective effects of parity have variously been attributed to the differentiation that accompanies pregnancy and lactation, alterations in ovarian hormone receptor levels, and altered sensitivity to ovarian hormones. Butyrate, a short-chain fatty acid,induces differentiation in breast cancer cell lines and decreases hormone receptor expression. Butyrate also inhibits proliferation in breast cancer cell lines and modulates expression of key cell cycle-regulatory proteins including cyclin D1. Given these properties,butyrate could be considered a promising agent for breast cancer prevention. Therefore, this study aimed to determine the effects of butyrate on normal human breast epithelial cells and to compare the effects of two stable butyrate derivatives with more favorable pharmacological properties: phenylacetate and its p.o. active precursor phenylbutyrate. Treatment with each agent resulted in concentration-dependent growth inhibition in a normal breast epithelial cell line and two breast cancer cell lines (MCF-7 and MDA-MB-231). Phenylbutyrate and butyrate inhibited proliferation to a similar extent, but phenylacetate was less effective in all of the cell lines. All three of the agents induced differentiation (accumulation of lipid droplets) in normal as well as in breast cancer cells and caused a decrease in estrogen receptor (ER) mRNA in MCF-7 cells. The butyrates decreased expression of cyclin D1, increased expression of p21Waf1/Cip1, and hypophosphorylated pRB in the normal mammary epithelial cells. The effects on cyclin D1 expression correlated with the effects on cell proliferation, which suggests that modulation of cyclin D1 expression may underpin the antiproliferative effects of butyrates. We have shown that butyrate and butyrate-like agents are able to decrease proliferation and induce differentiation in normal breast cells as well as in malignant breast cells (ER-positive and ER-negative) and, as such, may be considered as candidate chemopreventative agents for women at high risk of developing breast cancer.
Ovarian hormones are required for normal mammary development. Female mice lacking ERα3 have underdeveloped mammary glands, with only rudimentary ducts present at the nipple (1), and PR-null mice fail to establish the lobular-alveolar system essential for lactation (2). Ovarian hormones also influence susceptibility to mammary carcinogenesis: women without ovaries have a low risk of breast cancer;and prolonged ovarian activity during life (early menarche or late menopause) is associated with increased risk (3). Sequential rounds of proliferation and cell death in concert with the menstrual cycle (4), during which genetic alterations can accumulate, are thought to underlie this increased risk of breast cancer associated with cumulative hormone exposure. Sensitivity of mammary cells to proliferative effects of the ovarian hormones is,therefore, likely to be a key requirement for their activity as tumor promoters.
Full-term pregnancy early in reproductive life is protective against breast cancer in women. The protective effects of parity have been attributed to the differentiation that accompanies pregnancy and lactation (5). However, differentiation alone is not always accompanied by protection against mammary carcinogenesis (6). Alternative views on the mechanisms underlying the protective effects of parity include alterations in ovarian hormone receptor levels and altered sensitivity to ovarian hormones. Mimicking the pregnancy-induced changes that occur in the mammary gland using pharmacological agents may serve as a form of chemoprevention for breast cancer (7).
A promising preventative agent in human cancer is the short-chain fatty acid butyrate. Butyrate is a naturally occurring fatty acid derived from a high-fiber diet and produced in the large bowel by bacterial fermentation. Butyrate is also present at low levels in many fruit and vegetables but its richest source is milk fat (8, 9). Colonic generation of butyrate has been associated with the protective effect of dietary fiber for colon cancer (8, 10). Butyrate decreases expression of ER and PR (11, 12, 13), arrests growth, and induces differentiation in a variety of normal and malignant cell lines in culture, including normal mammary epithelial cells (9, 12, 14, 15, 16, 17, 18). Although the molecular mechanisms by which butyrate induces differentiation and cell cycle arrest are not well understood, it is known that butyrate induces a variety of changes within the nucleus, including histone hyperacetylation (19). Butyrate increases expression of the cyclin-dependent kinase inhibitor p21Waf1/Cip1 through a process thought to involve histone hyperacetylation, and it has been suggested that p21Waf1/Cip1 expression is a requirement for butyrate-mediated growth arrest (16),although this has been disputed (20). Induction of p21Waf1/Cip1 by butyrate is independent of p53,an advantage in potential anticancer and chemopreventive agents,because p53 mutations occur frequently in neoplasia (21). Butyrate also causes a decrease in cyclin D1 expression at the level of gene transcription (20, 22). This is relevant to the potential use of butyrate in chemoprevention, because cyclin D1 has been strongly associated with proliferation and carcinogenesis in the mammary gland (23, 24, 25, 26).
Another cell cycle-related gene implicated in butyrate action is pRB. pRB inhibits cell proliferation by repressing a subset of genes that are controlled by the E2F family of transcription factors and which are involved in progression from G1 to the S-phase of the cell cycle. The histone deacetylase HDAC1 physically interacts and cooperates with pRB, and there is evidence that this complex is a key element in the control of proliferation and differentiation (27). Butyrate treatment leads to pRB phosphorylation, and this is likely to be involved in the mechanisms underpinning butyrate inhibition of cell proliferation, although it may be distal to down-regulation of cyclin D1 (20, 28). Other mechanisms of butyrate action have been postulated including inhibition of protein prenylation; activation of peroxisome proliferator-activated receptors (29); DNA hypomethylation; and depletion of circulating glutamine (30, 31, 32).
Despite the demonstrated antiproliferative actions of butyrate,its usefulness as a cancer preventative agent in humans is limited,because it is rapidly metabolized and excreted in vivo. These limitations have potentially been circumvented by the development of stable derivatives or prodrugs of butyrate, such as phenylacetate and its p.o. active precursor phenylbutyrate, which have more favorable pharmacological properties (9). These butyrate derivatives have been used clinically for the treatment of diseases such as urea cycle disorders, β-thalassemia (33, 34), and sickle cell anemia (35). Phenylbutyrate has also been used in the treatment of leukemia; and Tributyrin, a triglyceride containing three butyrate moieties esterified to glycerol, has been tested in Phase I trials in patients with a variety of solid tumors (36). Phenylacetate has also been shown to block 5-aza-2′-deoxycytidine-induced neoplastic transformation of fibroblasts in vivo and in vitro (37).
Although butyrates show promise as potential chemopreventative agents,their actions on normal breast cells must be clarified before a role for these agents in breast cancer prevention is contemplated, in particular the action of the longer-acting butyrate analogues that can be p.o. administered, such as phenylbutyrate. The antiproliferative and differentiation effects of butyrate have previously been investigated in breast cancer cell lines, and, in some cases, differences in butyrate action have been noted in ER-positive compared with ER-negative cell lines (11, 12, 38). Therefore, a comparison of the effects of butyrate and its analogues on proliferation and differentiation in normal and malignant breast cell lines (ER-positive and ER-negative), as a first step in determining whether this class of drugs could be effective in prevention of breast cancer in women who are at high risk, forms the focus of studies reported here.
MATERIALS AND METHODS
The human breast cancer cell lines MCF-7 and MDA-MB-231 were obtained from EG & G Mason Research Institute (Worcester, MA). The normal mammary epithelial cell line, BRE-80, was obtained from Dr. L. Huschtscha (Children’s Medical Research Institute, Sydney, Australia). This cell line was derived from primary culture of a tissue reduction mammoplasty, obtained with informed consent, from a 36-year-old female patient (39).
MCF-7 and MDA-MB-231 cells were grown in RPMI 1640, supplemented with 10% v/v FCS, 0.25 units/ml insulin (Actrapid, MC Neutral Injection,human (100 units/ml); CSL-NOVO, North Rocks, NSW, Australia), 6 mm l-glutamine, buffered with sodium bicarbonate (5.6% w/v), and 20 mm HEPES. BRE-80 cells were grown in MCDB.170 media supplemented with bovine pituitary extract(Life Technologies, Inc., Melbourne, VIC, Australia) and maintained as described previously (39). Cell lines were routinely tested for Mycoplasma contamination using the Gen Probe T. C. Rapid Detection System (Bio Mediq, Doncaster, VIC, Australia)and found to be negative.
Exponentially growing breast cell lines were plated at the following cell densities: 1.6 × 103cells/well for MCF-7 and 1 × 103 cells/well for MDA-MB-231 and BRE-80, in 96-well plates in 50 μl of their respective media.
Cells were grown in a humidified incubator chamber in 5%CO2 at 37°C until logarithmic growth was established (∼2–3 days). Stock solutions (0.5 m) of sodium butyrate (from Sigma-Aldrich Pty Ltd, Castle Hill, NSW,Australia), sodium phenylacetate, and sodium phenylbutyrate(Fyrklovern Scandinavia AB, Kopingsvik, Sweden) were diluted in the appropriate medium immediately prior to use. Cell proliferation was measured using a colorimetric assay (Cell Titer 96 Non-Radioactive Cell Proliferation/Cytotoxicity assay obtained from Promega,Annandale, NSW, Australia) according to the manufacturer’s instructions.
Levels of lipid accumulation, a classic differentiation marker,were measured in breast cell lines by histochemical analysis using oil-red-O staining (12). Cells were grown in Lab-Tek Chamber slides (eight chambers per slide, 0.9 cm2per chamber; obtained from Nunc Inc., Naperville, IL) at 37°C in a humidified chamber in 5% CO2. After treatment with butyrate, phenylacetate, phenylbutyrate, or vehicle (deionized water), cells were rinsed in calcium- and magnesium-free PBS and fixed in 10% formol calcium (10 min, room temperature). Slides were rinsed in water and then in 60% (v/v) isopropyl alcohol before staining for 10 min in oil-red-O (2.5 mg of oil-red-O/ml isopropyl alcohol, diluted 6:4 in distilled water, and filtered through No. 42 Whatman filter paper; Ref. 40). Slides were washed briefly in 60%isopropyl alcohol and then in deionized water, counterstained in Harris’ hematoxylin (BDH Chemicals Ltd, Poole, England), and mounted using Aquamount (Fronine Histo Labs, Riverstone, NSW, Australia).
Cells were individually assigned with levels of intensity of oil-red-O staining on a scale of 0 to 3 with 0 denoting no lipid accumulation within a cell and 3 denoting maximal lipid accumulation, determined by light microscopy. The proportions of cells with differing levels of intensity were determined by counting random high power (×400) fields of the cell monolayer with the aid of the Image Analyser (consisting of Zeiss Axioskop microscope and the Optimas 5.1a software) until at least 1000 cells were counted. Histoscores were derived by calculating the percentage of cells with a particular score and multiplying by the value issued to that score, i.e., maximal lipid accumulation in all of the cells was, therefore, calculated to have a histoscore score of 300 (score of 3 × 100%).
Northern Blot Analysis
All of the chemicals used in the isolation and purification of RNA were of analytical or molecular biology grade. Exponentially growing cells were treated with butyrate, phenylacetate, and phenylbutyrate, or vehicle control, in 150-cm2 flasks. Total cellular RNA was isolated using Tri-reagent [purchased from Molecular Research Center (Cincinnati, OH) and Sigma-Aldrich] according to the manufacturers’ instructions. RNA (20 μg/sample) was separated by formaldehyde/agarose gel electrophoresis and transferred to Zeta-Probe nylon membranes (Bio-Rad, Richmond, CA) as described previously (41).
The human ER cDNA probe was a 0.8 kb EcoRI/HindIII fragment encoding the 5′ end of the open reading frame (41, 42). Equivalent loading of RNA samples for Northern blot analysis was confirmed by hybridization with an end-labeled oligonucleotide probe complementary to a highly conserved region of the 18S rRNA as described previously (41).
The ER cDNA probe was labeled by random priming with[α-32P]dCTP (Redivue, 370 MBq/ml, 10 mCi/mmol; Amersham Life Sciences, Ryde, NSW, Australia) using a Prime-a-Gene Kit (Promega), and oligonucleotides were end-labeled with[γ-32P]dATP (370 MBq/ml, 10 mCi/mmol;Amersham) using a Terminal Kinasing Kit (Promega) to specific activities of 108-109dpm/μg. Northern blot hybridization was performed as described previously (41). Membranes were exposed to Kodak Scientific Imaging Film, resulting bands were quantitated using a Molecular Dynamics densitometer and analyzed using Image Quant software(Molecular Dynamics, Sunnyvale, CA).
Western Blot Analysis
After 24 h treatment with butyrate, phenylacetate and phenylbutyrate, or vehicle control, cell lysates were prepared using lysis buffer (50 mm HEPES, 150 mm NaCl,1.5 mm MgCl2, 1 mm EDTA,10% glycerol, 1% Triton X-100, 50 mm NaF, 20 mm sodium PPi, 1 mmvanadate, and 1 mm DTT) containing protease inhibitors:aprotinin (10 μg/ml), leupeptin (10 μg/ml), and phenylmethylsulfonyl fluoride (1 mm), obtained from Sigma-Aldrich. Twenty μg of total protein/lane was separated by SDS PAGE—6% for pRB, 10% for cyclin D1, and 15% for p21Waf1/Cip1, each with a 3% (w/v) acrylamide stacking gel—and then transferred to nitrocellulose. Membranes were incubated (1 h, room temperature) with the primary antibody diluted in TBS [20 mm Tris/HCl, 500 mm NaCl(pH 7.5)] + 5% (w/v) nonfat skim-milk, as follows: cyclin D1 (1:300;Novocastra Laboratories Ltd, Newcastle upon Tyne, United Kingdom);p21Waf1/Cip1 rabbit polyclonal antibody(1:1000); or purified mouse IgG1 antihuman pRB antibody (1:1000; Santa Cruz Biotechnology Inc., Santa Cruz, CA).
The membranes were then washed three times in TBBS [20 mm Tris/HCl, 500 mm NaCl (pH 7.5), and 0.05%(v/v) Tween 20], and were incubated (1 h, room temperature) with secondary antibodies [peroxidase-conjugated goat antirabbit IgG (0.25 g/liter) or peroxidase-conjugated goat antimouse IgG (1.0 g/liter),supplied by DAKO, Denmark] diluted 1:4000 in TBS + 5% (w/v) nonfat skim-milk powder. Immunoreactive bands were visualized using Enhanced Chemiluminescence (ECL) Reagent Plus kit (Amersham, Life Sciences). The protein abundance was quantitated by densitometric analysis within the linear range of the autoradiograph film using a Molecular Dynamics densitometer and Image Quant Software.
A normal breast cell line derived from primary culture of normal breast tissue, designated BRE-80 (39), was used in this study. Normal mammary epithelial cells grown in culture are almost invariably ER-negative (43), and we found the BRE-80 cell line also to be ER-negative by Northern blot analysis (not shown). The two breast cancer cell lines used differed in ER status (MCF-7 cells are ER-positive; MDA-MB-231 are ER-negative; Ref. 41). The doubling times of the three cell lines were similar: 36 h for BRE-80; 38 h for MCF-7; and 39 h for MDA-MB-231.
Butyrate and Butyrate Analogues Decrease Proliferation
Treatment with butyrate, phenylacetate, and phenylbutyrate resulted in concentration-dependent growth inhibition of the normal breast cell line BRE-80 (Fig. 1). The ER-positive and ER-negative breast cancer cell lines were also growth inhibited by all of the agents. To compare the effectiveness of each agent, the IC60 after 72 h was calculated(Fig. 1 D). Phenylbutyrate was most effective in inhibiting proliferation in BRE-80 cells, whereas butyrate and phenylbutyrate were similarly effective in the breast cancer cell lines. Phenylacetate was the least effective agent in inhibiting proliferation in all of the three cell lines.
Overall, MCF-7 was the most sensitive of the cell lines to butyrate and butyrate analogues, followed by MDA-MB-231 and BRE-80, the latter being least sensitive. The IC60 values for butyrate and phenylbutyrate were in the low mm range, consistent with physiological levels of butyrate in the colon and with pharmacologically achievable levels of phenylbutyrate in the circulation (35).
Induction of Differentiation by Butyrate and Butyrate Analogues
Butyrates are known to induce differentiation in human breast cell lines (12, 17, 38); therefore, the effects of butyrate,phenylbutyrate, and phenylacetate on lipid accumulation—a classical marker of differentiation in breast cells—were compared using oil-red-O staining (40).
Butyrate and both butyrate analogues induced concentration-dependent accumulation of lipid droplets in all of the three cell lines (Fig. 2). All of the three butyrates were similarly effective in inducing differentiation in the normal cell line BRE-80. In MCF-7 cells, half maximal lipid accumulation was achieved by the same concentration of all of the three agents, although butyrate induced some lipid accumulation at a lower concentration than the other two butyrates (Fig. 2 B). The greatest variation in lipid accumulation between the three agents was found in the ER-negative breast cancer cell line, MDA-MB-231. In this cell line, phenylbutyrate was the most effective lipid-inducing agent,followed by phenylacetate and then butyrate.
Overall, at 48 h, the MDA-MB-231 and MCF-7 cell lines were more sensitive to the differentiating effects of butyrate and phenylbutyrate than the normal cell line BRE-80. However, after 72 h in butyrate,the BRE-80 cell line had accumulated levels of lipid that were similar to those of the other cell lines (Fig. 2 D), which suggested that lipid accumulation in response to butyrate was time-dependent, as well as concentration-dependent. A similar increase in lipid accumulation was observed in BRE-80 after a 72-h treatment with phenylbutyrate and phenylacetate (data not shown).
A comparison of the concentrations required for cell-growth inhibition(Fig. 1) and for differentiation (Fig. 2) showed that despite the different IC60 values observed for each agent,the effects on differentiation were similar, at least for BRE-80 and MCF-7 cells. This may suggest that the induction of differentiation was maximally effective at a lower threshold concentration than that required for maximal growth inhibition in these cell lines.
Effect of Butyrate and Butyrate Analogues on Levels of ER mRNA
We have previously shown that butyrate directly down-regulates ERα expression at the level of gene transcription (11). However, it was not known whether butyrate analogues would achieve the same effect. Therefore, the effects of butyrate, phenylacetate, and phenylbutyrate on ERα mRNA levels were measured by Northern blot analysis. Each of the agents caused a concentration-dependent down-regulation of ERα expression (Fig. 3). Butyrate was the most effective in down-regulating ERα: ERα mRNA levels were decreased to below 40%of control at 1 mm and were diminished markedly after treatment with 3 mm. Three mm phenylbutyrate reduced ER mRNA levels to 40% of control and 10 mmresulted in a decrease in ERα mRNA to below the level of detection. Phenylacetate was the least effective of the three agents in decreasing ERα expression. The concentrations required for ER down-regulation were similar to those required for growth inhibition.
Effects of Butyrate and Butyrate Analogues on Cell Cycle-related Proteins
Although the key regulatory molecules underlying the mechanism of action of butyrates are not well delineated, some of the potential effectors of butyrate action on differentiation and cell cycle arrest,were compared in normal breast cells and breast cancer cell lines. The expression of cyclin D1, p21Waf1/Cip1, and pRB was examined by Western blot analysis after exposure to butyrate,phenylacetate, and phenylbutyrate.
Butyrate and Butyrate Analogues Modulate Expression of Cyclin D1.
Down-regulation of cyclin D1 by butyrate has previously been described in breast cancer cell lines (20). In agreement with these data, butyrate and phenylbutyrate caused a concentration-dependent decrease of cyclin D1 expression in both of the breast cancer cell lines. In the normal, BRE-80 breast cell line (Fig. 4,A), butyrate and phenylbutyrate decreased cyclin D1 levels, although an increase at low phenylbutyrate levels was noted. Phenylacetate decreased cyclin D1 levels in BRE-80 and MCF-7 cells but was less effective than the other two agents. The effectiveness of each of the agents in modulating cyclin D1 expression correlated well with their effects on cell proliferation, in that phenylbutyrate and butyrate were similarly effective in decreasing proliferation, and they modulated cyclin D1 expression to an equal extent; phenylacetate was the least effective antiproliferative agent and modulated cyclin D1 expression to a lesser extent. Furthermore, a linear relationship was found between reduction of cyclin D1 levels and reduction in proliferation by butyrate as shown in Fig. 4 D.
Effects of Butyrate and Butyrate Analogues on p21Waf1/Cip1 and pRB.
Treatment with butyrate and both butyrate analogues resulted in increased expression of p21Waf1/Cip1 in all of the three cell lines (Fig. 5). Butyrate caused the greatest-fold induction of p21Waf1/Cip1 in each of the lines, followed by phenylacetate, and then phenylbutyrate. The greatest induction of p21Waf1/Cip1 was seen in the ER-negative cancer cell line MDA-MB-231. No direct correlation was observed between the fold induction of p21Waf1/Cip1 and the extent of growth inhibition or induction of differentiation by butyrate and the butyrate analogues.
In MCF-7 and BRE-80 cells butyrate, phenylbutyrate and, to a lesser extent, phenylacetate resulted in concentration-dependent hypophosphorylation of pRB, shown in Fig. 6. pRB hypophosphorylation in these cell lines was concordant with growth inhibition, in that the effects of butyrate and phenylbutyrate were similar, but phenylacetate was less effective. However, the MDA-MB-231 cell line did not show hypophosphorylation of pRB at 24 h, despite being sensitive to the antiproliferative and differentiating effects of all three of the agents.
Cloning of breast cancer predisposition genes including BRCA1 and BRCA2 (44, 45, 46) has allowed identification of women at increased risk of breast cancer and has focused attention on the need for preventative strategies. The success of the National Surgical Adjuvant Breast and Bowel Project (NSABP)Breast Cancer Prevention Trial (BCPT), reporting a 45% reduction in breast cancer incidence among the high risk participants on tamoxifen,has shown that chemoprevention in breast cancer is possible (47). However, predominantly ER-positive tumors were inhibited by tamoxifen in the NSABP BCPT. Tamoxifen reduced the occurrence of ER-positive tumors by 69%, but no difference in the occurrence of ER-negative tumors was seen (47). Unfortunately, BRCA1 tumors are predominantly ER-negative (48); hence, tamoxifen may not be effective in the very group at highest risk of developing breast cancer (49). Therefore, there is a need to develop novel strategies to prevent ER-negative as well as ER-positive breast cancers.
The efficacy of new candidate preventative agents, such as the butyrates, in the normal breast is largely unexplored, and the effectiveness of physiologically and therapeutically relevant formulations of such agents require further investigation in the normal breast and breast cancer cells. In this study, we have sought to investigate the effects of butyrate and butyrate-like agents on normal breast cells and ER-positive and ER-negative breast cancer cells. Although the majority of human breast cancers are ER-positive,ER-positive cells are in the minority in the normal breast; and the greater proportion of normal cells proliferating under the influence of ovarian hormones are ER-negative (50, 51). This suggests that ovarian hormones control proliferation in the normal breast in a paracrine fashion involving, as yet incompletely defined, sets of growth factors. A preventative strategy, aiming for a reduction in proliferation of normal mammary cells and preneoplastic cells, would,therefore, need to encompass the inhibition of both ER-positive and ER-negative cells.
The rationale behind the use of butyrates in the reduction of mammary cell proliferation is that butyrate is known to decrease the expression of both ER and cyclin D1 (11, 17, 20, 22). Ovarian hormones, known to be critical for normal mammary growth and development, may have a role in modulating carcinogenesis (1, 3), and ovarian hormone receptors are expressed in the majority of breast cancers. Cyclin D1 has also been implicated in mammary carcinogenesis: overexpression of cyclin D1 in transgenic mice results in the development of mammary hyperplasia and adenocarcinomas (52); cyclin D1 is overexpressed in premalignant lesions in the human breast (25); and cyclin D1 is overexpressed in almost 50% of human breast cancers (26). Cyclin D1 and ER are, therefore, rational targets for mammary chemoprevention.
Butyrate itself is rapidly metabolized and excreted in vivo;and, therefore, stable derivatives or prodrugs of butyrate, such as phenylacetate and phenylbutyrate, with more favorable pharmacological properties have been developed (9). Butyrate derivatives,having a low incidence of side effects, are considered to be safe and are used in children for the treatment of diseases such as urea cycle disorders, β-thalassemia (33, 34), and sickle cell anemia (35). Administration of p.o. phenylbutyrate in patients with sickle cell anemia has resulted in plasma levels in the mm range (up to 1.95 mm),which is within the range of concentrations resulting in growth inhibition in the in vitro experiments reported here (35). Although the currently available oral formulations of phenylbutyrate still require frequent dosing with large numbers of capsules or tablets to sustain effective plasma drug concentrations,alternative derivatives and delivery strategies are being developed that may overcome these limitations (53). Although the action of butyrate has been investigated in a number of cell lines,including breast cancer cells, there has been little or no information on the comparative efficacy of the butyrate analogues phenylacetate and phenylbutyrate in breast cancer cells, nor has the effect of these analogues in normal breast cells been known.
Overall, butyrate and its analogues inhibited normal-breast epithelial cell growth and growth of breast cancer cells. The effect of the stable analogue phenylbutyrate on cell proliferation was similar to that of butyrate, whereas phenylacetate was less effective at inhibiting proliferation. These results are in agreement with those found in human prostate cancer cell lines, in which phenylbutyrate was more active than phenylacetate and similar in activity to butyrate (54), and in myeloid leukemia cells, in which butyrate but not phenylacetate had significant growth-inhibiting and differentiating effects (55).
Lipid accumulation, measured by oil-red-O staining, was measured as a marker of mammary cell differentiation (40). The butyrates induced differentiation in the normal cell line BRE-80 and in the two breast cancer cell lines: there was a time- and concentration-dependent accumulation of lipid droplets in ER-negative and ER-positive breast cell lines after treatment with butyrate and the two butyrate derivatives. One previous study has shown lipid accumulation in breast cancer cells after butyrate treatment (16), but, in contrast to the results of this study, the effect was confined to ER-positive cells. The results of this study indicate that these agents are effective differentiation agents in ER-negative as well as ER-positive breast cells and in normal as well as malignant breast cell lines.
In the MCF-7 cell line, reduction in ER mRNA paralleled inhibition of proliferation, with sodium butyrate being the most effective agent,followed by phenylbutyrate and then phenylacetate. However, as discussed above, the butyrates inhibited proliferation in both ER-positive and ER-negative cells. Although it is possible that butyrate and butyrate analogues inhibit proliferation in ER-positive and ER-negative cells by different mechanisms, there is little evidence that this is the case. Therefore, it is likely that a reduction in ER is not the primary mechanism underpinning the antiproliferative effects of butyrate and the effectors are potentially cell cycle-related genes downstream of ER.
Butyrate and phenylacetate have been shown to increase expression of the cyclin dependent kinase inhibitor p21Waf1/Cip1. It has been suggested that p21Waf1/Cip1 is a critical effector of butyrate-induced growth arrest and that cells lacking p21Waf1/Cip1 are insensitive to these agents (16, 21). However, Vaziri et al.(20) found that fibroblasts lacking p21Waf1/Cip1 underwent butyrate-induced G1 arrest at doses similar to those of wild-type cells and suggested that perturbation of cyclin D1/RB phosphorylation are more likely to be the key effectors of butyrate inhibition of growth rather than p21Waf1/Cip1. Although p21Waf1/Cip1 was induced by all of the three agents in this study, phenylbutyrate and phenylacetate were not as effective as butyrate in inducing p21Waf1/Cip1 expression. p21waf1/Cip1is likely to play some part in butyrate effects on proliferation. Increases in cyclin D1 expression at lower concentrations of butyrates did not translate into an increase in proliferation, and this effect is most likely to be attributable to a concomitant increase in p21Waf1/Cip1.
Butyrate, phenylbutyrate, and phenylacetate induced pRB hypophosphorylation in MCF-7 and BRE-80 cells. However, none of these agents hypophosphorylated pRB in MDA-MB-231 cells, despite a reduction in cyclin D1, an induction of p21Waf1/Cip1, and an inhibition of proliferation, which suggested that perturbation of pRB per se is not required for butyrate effects on cell growth.
However, in all three cell lines, and with each agent, a reduction in cyclin D1 expression correlated with inhibition of cell proliferation. These results suggest that the reduction in expression of cyclin D1 is the most likely mechanism to underpin the inhibition of proliferation by butyrate.
The effects of butyrate on gene expression have been attributed to histone hyperacetylation (19) and DNA hypomethylation (30). As described above, butyrate increases expression of p21Waf1/Cip1 and decreases expression of cyclin D1. Other histone deacetylase inhibitors also increase expression of p21Waf1/Cip1 and decrease expression of cyclin D1 (56), which suggests that these actions of butyrate are mediated by histone deacetylase inhibition. The effects of butyrate on the expression of cyclin D1 are mediated via defined sequences in the cyclin D1 promoter (22). It has been suggested that butyrate may modulate gene expression via the GTPase Rac1 signaling pathway (57). A fragment of a novel butyrate-induced transcript (B-ind1) has been shown to block Rac1-mediated NF-κB activity (57). It has previously been shown that activation of Rac1 enhances expression of cyclin D1 through a NF-κB site in the proximal promoter (58). Therefore, butyrate may decrease cyclin D1 expression by interrupting Rac1 signaling.
Agents that decrease hormone responsiveness and down-regulate cyclin D1 may be potential breast cancer preventative agents. We have shown that butyrate and butyrate-like agents have these effects and are able to decrease proliferation and induce differentiation in normal and malignant breast cells and, because of this, may be considered as candidate chemopreventative agents. This is further supported by our demonstration that phenylbutyrate decreases proliferation of normal cells in the murine mammary gland in vivo.4 The effectiveness of butyrate analogues in normal breast cells and in both ER-positive and ER-negative breast cancer cells suggests that this class of compounds warrants serious consideration as preventative agents in women at particularly high risk of developing breast cancer in their lifetime.
Supported by the National Breast Cancer Foundation, the New South Wales Cancer Council, and the Australian Cancer Foundation.
The abbreviations used are: ER, estrogen receptor; PR, progesterone receptor; pRB, retinoblastoma protein;IC60, the concentration at which 60% of growth was inhibited; TBS, Tris-buffered saline.
C. Kennedy, C. L. Clarke, and A. deFazio. Proliferation in the normal mouse mammary gland and inhibition by a butyrate analogue, manuscript in preparation.