Disparate impact of butyroyloxymethyl diethylphosphate (AN-7), a histone deacetylase inhibitor, and doxorubicin in mice bearing a mammary tumor

Abstract

The histone deacetylase inhibitor (HDACI) butyroyloxymethyl diethylphosphate (AN-7) synergizes the cytotoxic effect of doxorubicin (Dox) and anti-HER2 on mammary carcinoma cells while protecting normal cells against their insults. This study investigated the concomitant changes occurring in heart tissue and tumors of mice bearing a subcutaneous 4T1 mammary tumor following treatment with AN-7, Dox, or their combination. Dox or AN-7 alone led to inhibition of both tumor growth and lung metastases, whereas their combination significantly increased their anticancer efficacy and attenuated Dox- toxicity. Molecular analysis revealed that treatment with Dox, AN-7, and to a greater degree, AN-7 together with Dox increased tumor levels of γH2AX, the marker for DNA double-strand breaks and decreased the expression of Rad51, a protein needed for DNA repair. These events culminated in increased apoptosis, manifested by the appearance of cytochrome-c in the cytosol. In the myocardium, Dox-induced cardiomyopathy was associated with an increase in γH2AX expression and a reduction in Rad51 and MRE11 expression and increased apoptosis. The addition of AN-7 to the Dox treatment protected the heart from Dox insults as was manifested by a decrease in γH2AX levels, an increase in Rad51 and MRE11 expression, and a diminution of cytochrome-c release. Tumor fibrosis was high in untreated mice but diminished in Dox- and AN-7-treated mice and was almost abrogated in AN-7+Dox-treated mice. By contrast, in the myocardium, Dox alone induced a dramatic increase in fibrosis, and AN7+Dox attenuated it. The high expression levels of c-Kit, Ki-67, c-Myc, lo-FGF, and VEGF in 4T1 tumors were significantly reduced by Dox or AN-7 and further attenuated by AN-7+Dox. In the myocardium, Dox suppressed these markers, whereas AN-7+Dox restored their expression. In conclusion, the combination of AN-7 and Dox results in two beneficial effects, improved anticancer efficacy and cardioprotection.

Figure 1. Effect of Dox, AN-7, and AN-7+Dox in the 4T1 murine mammary carcinoma model.

(A) In the preliminary experiment, female Balb/c mice carrying 75–180 mm³ subcutaneous 4T1 tumors were randomly divided into three groups (10 mice/group) and were injected ip once a week with either vehicle, 3 or 5 mg/kg Dox. Tumor volume was measured twice a week up to day 14 when the experiment was terminated (after 2 Dox treatments). Mean±SE of tumor volume was: *p<0.05, untreated vs. the other groups; #p<0.05, Dox 5 mg/kg vs. Dox 3 mg/kg. (B) Mean±SE of tumor weight and the ratio of heart-weight to body-weight (HW/BW, g/g), measured at the experiment termination point were: *p<0.01, Dox 5 mg/kg vs. all other groups. (C) In the pivotal experiment, the Percent Failure Free mice bearing 75–180 mm³ subcutaneous 4T1 tumors, was assessed by a Kaplan-Mier graph. The mice were assigned to the following treatments: saline vehicle ip (n = 7) or po (n = 7); ip Dox, 5 mg/kg once/week (n = 14); po AN-7, 50 mg/kg 3 times/week (n = 14); po AN-7+ip Dox (n = 14). The arrows on the x-axis indicate the point in time of the second and the third Dox treatment. (D) Mean±SE of heart-weight to body-weight ratios (HW/BW, g/g) at the above experiment end-point was ^#p<0.01, Dox 5 mg/kg vs. all other groups. (E) Tumor weight at the termination point (mean±SE) was: Dox 5 (total) represents the tumor weight of all the mice treated with 5 mg/kg Dox (n = 14), as measured at their individual end-points; Dox 5 (>21) represents the tumor weight of the mice (n = 5) treated with 5 mg/kg Dox and survived beyond day 21 of treatment. *p<0.02 all groups vs. untreated control; ^#p<0.05 AN-7+Dox vs. AN-7 or Dox 5 (total) or p<0.01 Dox 5 (>21). (F) Mean±SE of lung lesions for the various treatment groups is shown. *p<0.02, untreated vs. all treatment groups; ^#p<0.05, AN-7+Dox vs. AN-7 or vs. Dox.

Figure 2. Effect of AN-7 on the activity and expression of cellular HDACs classes I and II.

(A) The cells were exposed to AN-7 for 3 h in the presence of the HDAC fluorogenic substrate. The average % of HDAC activity as a function of 0–100 µM concentration of AN-7 of a representative experiment conducted in quadruplicates is shown. (B) The average IC₅₀±SE of AN-7 for each of the cell types is calculated from three independent experiments. (C) Lysates of cells were loaded (35 µg protein/well) on 10% SDS gels and were subjected to Western blot analysis. HDAC1, HDAC2, HDAC5 or actin were detected with the appropriate antibodies. (D) HDAC activity of three experiments (in triplicate) conducted with 4T1 (cancer) and H9C2 (non-cancer) cells treated with AN-7 (50 µM) as a single agent for a total of 3 h, 1 h prior to the addition of the fluorogenic substrate and 2 h in its presence; Dox (200 nM) as a single agent treatment, or with the combination AN-7+Dox where 50 µM AN-7 was added 1 h prior to the addition of Dox and the fluorogenic substrate and then incubated for an additional 2 h. (E) The viability of the cells treated as described in (D) was determined by a Hoechst assay after 23 or 24 h for cells treated with Dox. (F) Cells were treated with AN-7 (3 or 24 h), Dox (2 or 23 h) or AN-7 (3 or 24 h)+Dox (2 or 23 h), as described (D). The lysates of the cells were subjected to Western blot analysis as described (C).

Figure 3. Effect of AN-7, DOX and AN-7+Dox on HDAC1 and HDAC2 expression in tumor or heart.

(A) Extracts of tumors from 8 mice and hearts from 6 mice per treatment group were loaded (20 µg protein/well) on 10% SDS gels, and subjected to Western blot analysis. HDAC1, HDAC2 or actin were detected with the appropriate antibodies. (B) Expression of HDAC1, HDAC2 or actin in tumors and hearts was analyzed by Odyssey 2.1. The intensity ratio of HDACs to actin (mean±SE) is presented. *p<0.05, tumors of untreated or Dox-treated vs. AN-7 or AN-7+Dox; ^#p<0.007, HDAC2 expression in the hearts of Dox treated vs. all other treatment groups.

Figure 4. Effect of AN-7, DOX and their combinations on apoptosis, DNA damage and repair and HO-1.

Sections of tumors (left hand-side) and hearts (right hand-side) from mice were stained with the specified antibodies and counter stained with hematoxylin. For the detection of markers by Western blot analysis, the samples were resolved on SDS gels. Lysates of tumors (20 µg protein) or hearts (40 µg protein) were loaded on the gels (left hand-side). Mean±SE of the ratios of bands intensity, each normalized to actin, is shown for tumors (n = 7), hearts (n = 7) and for 4 hearts of naive mice (the right hand-side). (A) IHC with anti cytochrome c (Nova Red). The arrows mark the necrotic areas; Bar = 200 µm, the 4-fold magnified picture in the insert was taken from the area indicated by the square. (B) 15% SDS gel stained for γH2AX; (C) IHC with anti Rad51 (DAB); Bar = 100 µm, the 2-fold magnified picture in the insert was taken from the area indicated by the square. (D) 10% SDS gel stained for Rad51 and MRE11; (E) 12% SDS gel stained for HO-1. ^#p<0.05 vs. untreated control; *p<0.05, Dox vs. AN-7 and AN-7+Dox.

Figure 5. Effect of AN-7, DOX and their combinations on markers of proliferation and fibrosis.

Sections of tumors and hearts were stained (DAB) for: (A) Ki-67; (B) c-Kit and both were counter stained with hematoxylin. The arrowhead marked the c-Kit positively staind cells. Bar = 200 µm, the 4-fold magnified picture in the insert was taken from the area indicated by the square. (C) Lysates samples, as described above, were resolved on 10% SDS gel and stained for c-Myc (left). The Mean±SE ratio of c-Myc/actin expression is shown (right). ^#p<0.05, as specified, vs. untreated control; *p<0.05, as specified, vs Dox. (D) Picrosirius red and fast-green for the visualization of interstitial fibrosis. Bar = 200 µm, the 2-fold magnified picture in the insert was taken from the area indicated by the square.

Figure 6. Effect of AN-7, DOX and their combinations on angiogenesis.

Sections of tumors and hearts were stained for (A) lo-FGF (DAB); (B) VEGF (Nova Red) and counter stained with hematoxylin. Bar = 200 µm, the 4-fold magnified picture in the insert was taken from the area indicated by the square. (C) Lysates samples (as described above) were resolved on 7.5% SDS gels and stained for TSP-1(left). Mean±SE of the ratio of TSP-1/actin expression is shown (right), ^#p<0.05, as specified, vs. untreated control; * p<0.05, as specified, vs. Dox; ^&p<0.05, AN-7+Dox vs. AN-7.