Synthesis and antiproliferative activity of novel erythromycin derivatives

Kai Bao; Foxiao Qiao; Long Liang; Hanbing Li; Huajun Zhu; Weige Zhang; Yingliang Wu

doi:10.1039/C0MD00103A

View PDF VersionPrevious ArticleNext Article

DOI: 10.1039/C0MD00103A (Concise Article) Med. Chem. Commun., 2010, 1, 294-298

Show CompoundsShow Chemical TermsShow Biomedical Terms

Synthesis and antiproliferative activity of novel erythromycin derivatives†

Kai Bao ^a, Foxiao Qiao ^b, Long Liang ^c, Hanbing Li ^a, Huajun Zhu ^c, Weige Zhang *^a and Yingliang Wu *^b
^aKey Laboratory of Structure-Based Drug Design & Discovery, Ministry of Education, Shenyang Pharmaceutical University, Shenyang 110016, P. R. China. E-mail: zhangweige2000@sina.com; Fax: +86 24 23986393; Tel: +86 24 23986422
^bSchool of Pharmacy, Shenyang Pharmaceutical University, Shenyang 110016, P. R. China. E-mail: yingliang_1016@163.com; Fax: +86 24 23986278; Tel: +86 24 23986278
^cSichuan Kelun Pharmaceutical Research Ltd, Chengdu, 610072, P. R. China

Received 8th July 2010 , Accepted 10th August 2010

First published on 8th September 2010

Abstract

A new series of dimers of de(N-methyl) erythromycin derivatives, joined by a -COCH₂- or -CH₂CH₂- linker, was synthesized and their in vitro antiproliferative activity against three human cancer cell lines was evaluated. Structure–activity relationships indicated that transformation of the linker -COCH₂- into -CH₂CH₂- led to an increase in the potency of the compound; removal of cladinose and conversion of the carbonyl group at the C9 position to 9(S)-hydroxyl- or 9(S)-amino-containing groups resulted in a marked decrease in activity. Among the compounds synthesized, compounds 3d and 3e, both had C9 oxime groups and showed strong inhibitory activity against the three cell lines. Flow cytometry studies on HT-1080 cells indicated that the mechanism underlying the dimers' antiproliferative activity involved inducing cell cycle arrest in the G₀/G₁ phase. It was discovered that the percentage of HT-1080 cells arrested depended on the concentration of the erythromycin derivative dimer and on the length of time of treatment.

Erythromycin A (EM-A), a typical macrolide antibiotic, has been used extensively in the clinic for more than 50 years.¹ In the past few decades, EM-A and its derivatives have attracted a great deal of attention because they have been demonstrated to have other therapeutic capabillities, such as gastrointestinal prokinetic activity and anti-inflammatory activity.^2–9 Moreover, pharmacological experiments have found that EM-A has anticancer potential by indirectly enhancing the tumoricidal activity of macrophages and natural killer cells or by inhibiting tumor angiogenesis.^10–14

In our study of modification and bioactivity of erythromycin derivatives, we treated de(N-methyl) EM-A with chloroacetyl chloride (1.1 equiv.) in anhydrous dichloromethane in the presence of N,N-diisopropylethylamine (3.0 equiv.). Besides the desired de(N-methyl)-N-chloracetyl EM-A derivative, dimer of de(N-methyl) EM-A linked by -COCH₂- (1a, Fig. 1) was obtained as the byproduct (11% yield) and was characterized by MS, ¹H NMR and ¹³C NMR. To our surprise, compound 1a was found to show potent antiproliferative activity against human tumor cell lines.


	Fig. 1 Structures of dimers of EM-A derivatives.

There are several reports of EM-A derivatives demonstrating potential as cancer treatment, however, the direct antiproliferative activity of EM-A derivatives on tumor cells has never been reported. This paper presents the first results of such a study. We evaluated the in vitro antiproliferative effects of a series of dimers of EM-A derivatives against three human cancer cell lines by 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay.

Compound 1a is a symmetrical dimer (homodimer) of two identical de(N-methyl) EM-A units joined by a -COCH₂- linker. Previous structure–activity relationship (SAR) studies indicate that modifications at the C3, C6 and C9 positions on the lactone ring of EM-A derivatives have significant influences on bioactivity. We thus also designed homodimers 1b (which has an O-methyl substitution at position 6) and 1c (which has an etheroxime group at position 9). To further expand the diversity of the dimers of EM-A derivatives, heterodimers (compounds bearing two different monomers) 2a–2e with modifications at the C3, C6 and C9 positions were also considered. Moreover, to investigate the effect modifications of the linker may have, we transformed the -COCH₂- unit into -CH₂CH₂- linker and synthesized homodimers 3a–3p and heterodimers 4a–4i (Fig. 1). Lastly, the carbonyl groups at position 9 were converted to hydroxyl-, amino-, oxime- and O-alkyloxime-containing groups in these series of dimers, to assess the importance of modifications at the C9 position to biological activity.

The target compounds were synthesized by acylation and/or alkylation of the de(N-methyl) EM-A derivatives 5a–5r (Scheme 1). Compounds 5a–5c, 5f–5o and 5r were obtained, starting from EM-A, clarithromycin and other EM-A derivatives modified at the C9 position, by N-demethylation with iodine and sodium acetate in methanol;¹⁵ the cladinose moieties of compounds 5b, 5c, 5f and 5h were selectively removed upon treatment with dilute hydrochloric acid in aqueous solution to give compounds 5d, 5e, 5p and 5q, respectively.¹⁶


	Scheme 1 Reagent and conditions: (a) ClCOCH₂Cl (0.5 equiv.), DIPEA, CH₂Cl₂, r.t.; (b) ClCOCH₂Cl (1.0 equiv.), DIPEA, CH₂Cl₂, r.t.; (c) 5a, 5b or 5c (1.0 equiv.), DIPEA, CH₂Cl₂, r.t. (d) BrCH₂CH₂Cl (20 equiv.), DIPEA, DMF, r.t.

The synthesis of compounds 1a–1c and 2a–2e was accomplished through acylation and alkylation of the de(N-methyl) EM-A derivatives 5a–5e. Starting from compounds 5a–5c, dimers 1a–1c were prepared by using 0.5 eq. chloroacetyl chloride in the presence of N, N-diisopropylethylamine in anhydrous dichloromethane (68% to 76% yields). Treatment of compounds 5a, 5b, 5d or 5e with 1.0 eq. chloroacetyl chloride gave the corresponding chloracetyl intermediates 5a′, 5b′, 5d′ or 5e′, which we reacted with compounds 5a, 5b or 5c (molar ratio, 1 [thin space (1/6-em)] :1) to give the desired dimers 2a–2e, respectively (63% to 77% yields).

Dimers 3a–3p were obtained by N-alkylation of compounds 5a–5p with 1-bromo-2-chloroethane. To obtain dimer 4a, compounds 5b and 5d were N-alkylated using 1-bromo-2-chloroethane in the presence of N,N-diisopropylethylamine (24% yield) in DMF. The same reaction also yielded dimers 3b and 3n in 19% and 22% yields, respectively. We used the same reaction sequence to produce dimers 4b–4i, starting from the two related N-demethylated erythromycin derivatives.

All the final dimers produced in this study were fully characterized by MS, ¹H NMR, ¹³C NMR and element analysis as described in the experimental section.

The antiproliferative activities of all target compounds were assessed in vitro against three human cancer cell lines: SGC-7901 gastric carcinoma, HT-1080 fibrosarcoma carcinoma and KB oral squamous carcinoma. We used colorimetric MTT assay with cisplatin as positive control. The antiproliferative activities were expressed as IC₅₀ values. The measured IC₅₀ values for all compounds are summarized in Table 1.

Table 1 In vitro antiproliferative activities of compounds 1a–4i against three human cancer cell lines^a

Compd.	IC₅₀/μM			Compd.	IC₅₀/μM
Compd.	SGC-7901	KB	HT-1080	Compd.	SGC-7901	KB	HT-1080
a The IC₅₀ values are defined as the concentration of the compound needed to cause a 50% inhibition in activity of the cell lines relative to a control cell line. Data are expressed as the means ± SE from the dose response curves of at least three independent experiments.
1a	26.5	>100	>100	3j	>100	>100	>100
1b	16.2	>100	30.9	3k	26.5	7.6	9.5
1c	13.3	28.7	33.4	3l	28.4	8.3	3.9
2a	>100	>100	>100	3m	20.0	20.1	10.3
2b	65.9	27.6	>100	3n	>100	>100	>100
2c	>100	>100	>100	3o	>100	>100	>100
2d	>100	>100	>100	3p	>100	>100	>100
2e	>100	>100	>100	4a	>100	97.0	>100
3a	18.3	37.3	>100	4b	12.7	38.2	42.8
3b	20.0	17.7	8.1	4c	39.8	28.5	20.4
3c	61.6	44.9	>100	4d	57.8	26.4	17.8
3d	21.7	6.2	8.6	4e	45.2	37.8	21.9
3e	7.8	4.4	5.3	4f	17.4	30.2	41.5
3f	>100	5.5	3.5	4g	7.2	18.6	16.2
3g	>100	17.5	27.7	4h	>100	>100	>100
3h	30.8	>100	54.1	4i	22.1	14.7	6.3
3i	>100	>100	>100	Cisplatin	9.3	0.58	12.7

As shown in Table 1, transformation of the linker -COCH₂- into -CH₂CH₂- did not lead to loss of biological activity, but to an increase in potency for compounds with cladinose moieties. For example, compound 3a showed IC₅₀ values of 18.3 and 37.3 against the SGC-7901 and KB cell lines, respectively, compared with 26.5 and >100 for compound 1a. In the -CH₂CH₂- linked series, compound 3e with C9 oxime groups was found to have significant activity against all three cell lines, suggesting that the structural modifications had a beneficial effect.

Comparing the IC₅₀ values of compounds 3b, 3c, 3e and 3f with their decladinose and/or di-decladinoses relatives, 4a and 3n, 3o, 4b, and 3p, it was clear that the removal of cladinose caused a marked decrease in the inhibition of growth of the cell lines. The di-decladinoses compounds 3n–3p were found to have no activity against the three cell lines.

From our preliminary investigation, we found that modifications at the C9 position influenced the biological activity of the compound. Conversion of the carbonyl group at the C9 position to 9(S)-hydroxyl- or 9(S)-amino-containing groups resulted in a marked decrease in the activity against SGC-7901, KB, and HT-1080 cell lines (e.g., 3a VS 3c, 3b VS 4d). The 9-oxime dimers, such as compounds 3d and 3e, which has unsubstituted oxime groups, were found to be more potent than their O-alkyloxime dimer relatives. Compound 3e exhibited a 2 to 10-fold increase in activity against the three cell lines compared to the O-alkyloximes-containing compounds 3h, 3k and 3m. The addition of an alkyl group to the C9 oxime groups in compound 3d (e.g., 3f, 3g, 3i, 3j and 3l) led to a decrease in antiproliferative activity.

The influence of the symmetry of the structure on activity was studied. In the case of dimers linked by a -COCH₂- unit, one general trend we observed was that the homodimers were more potent than the heterodimers. However, no substantial differences in activity between the homodimers and heterodimers were found for dimers linked by a -CH₂CH₂- unit, indicating that the symmetry of the structure was not crucial for good activity for these compounds.

To better understand the mechanism driving the antiproliferative activity of the prepared compounds, we analyzed using flow cytometry the cell cycles of HT-1080 cells treated for 24–48 h with the selected active compound 3d at 0.5×, 1.0× and 2.0 × IC₅₀ concentrations (Fig. 2, A and B). Treatment with compound 3d at 4.5 and 9.0 μM for 24 h caused a concentration-dependent increase in the number of HT-1080 cells in the G₀/G₁ phase. The percentage of HT-1080 cells in the G₀/G₁ phase rose from 49.18% to 65.96% after treatment with compound 3d at 4.5 μM and to 71.21% after treatment at 9.0 μM. This increase was associated with a decrease in the percentage of cells in the S and G₂/M phase. In addition, treatment of HT-1080 cells with compound 3d at 4.5 μM for 48 h induced cell cycle arrest at the G₀/G₁ phase from 41.60% to 80.04%. At higher concentrations (18.0 μM) of compound 3d, we observed an increase in the number of cells in the subdiploid peak, indicating apoptosis or necrosis of the cells.


	Fig. 2 Results for our cell cycle analysis. HT-1080 cells were treated with 0.5×, 1.0× and 2.0 × IC₅₀ concentrations of compounds 3d and 3e, stained with propidium iodide and processed at 24 and 48 h for flow cytometric analysis. Percentages of cells in the G₀/G₁ phase, S phase, G₂/M phase and SubG₁ phase cells are shown. Data refer to a representative result of three independent experiments.

Similar results were obtained for HT-1080 cells treated with compound 3e (Fig. 2, C and D). Treatment of cells with concentrations corresponding to 0.5×, 1.0× and 2 × IC₅₀ for 24 and 48 h resulted in a concentration- and time-dependent increase in the percentage of cells in the G₀/G₁ phase, although the concentration-dependent effect is not so obvious as compound 3d.

In conclusion, a new series of dimers of de(N-methyl) EM-A derivatives was prepared and their antiproliferative activity was evaluated. The present study revealed that their activity is influenced by the type and position of substituents on the lactone ring. Removal of cladinose and conversion of the carbonyl group at the C9 position to hydroxyl- or amino-containing groups resulted in a marked decrease in activity. Among the compounds synthesized, compounds 3d and 3e both had C9 oxime groups and showed potent activity against the three cell lines SGC-7901 gastric carcinoma, HT-1080 fibrosarcoma carcinoma and KB oral squamous carcinoma. From flow cytometry studies on HT-1080 cells, we concluded that the mechanism underlying the dimers' antiproliferative activity involved inducing cell cycle arrest in the G₀/G₁ phase. We found that the percentage of HT-1080 cells arrested depended on the concentration of the EM-A derivative and on the length of time of treatment. More detailed studies on the mechanism of inhibitory activity as well as in vivo cytostatic activity analyses are under way and will be reported in due course.

Notes and references

J. M. McGuire, R. L. Bunch, R. C. Anderson, H. E. Boaz, E. H. Flynn, H. M. Powell and J. W. Smith, Antibiot. Chemother., 1952, 2, 281 Search PubMed.
S. Omura, K. Tsuzuki, T. Sunazuka, S. Marui, H. Toyoda, N. Inatomi and Z. Itoh, J. Med. Chem., 1987, 30, 1941 CrossRef CAS.
W. E. Steinmetz, B. L. Shapiro and J. J. Roberts, J. Med. Chem., 2002, 45, 4899 CrossRef CAS.
T. Mercurio, A. Mereu, G. Morazzoni, E. Meriggi, M. Napoletano and F. Pellacini, Drugs Fut., 2004, 29, 387.
J. T. Randolph, P. Waid, C. Nichols, D. Sauer, F. Haviv, G. Diaz, G. Bammert, L. M. Besecke, J. A. Segreti, K. M. Mohning, E. N. Bush, C. D. Wegner and J. Greer, J. Med. Chem., 2004, 47, 1085 CrossRef CAS.
J. L. Kellenberger, S. R. Shapiro, S. Mathews, P. Guerry and P. J. N. Barbier, WO2006084410, 2006.
A. Mereu, E. Moriggi, M. Napoletano, C. Regazzoni, S. Manfredini, T. P. Mercurio and F. Pellacini, Bioorg. Med. Chem. Lett., 2006, 16, 5801 CrossRef CAS.
E. J. Giamarellos-Bourboulis, Int. J. Antimicrob. Agents, 2008, 31, 12 CrossRef CAS.
M. Bukvic-Krajacic, A. Hutinec, G. Kragol, N. Kujundzic and Z. Marusic-Istuk, WO2009130189, 2009.
E. Hofsli and J. Nissen-Meyer, Int. J. Cancer, 1989, 43, 520 CrossRef CAS.
J. Yatsunami, Y. Fukuno, M. Nagata, N. Tsuruta, S. Aoki, M. Tominaga, M. Kawashima, S. Taniguchi and S. Hayashi, Cancer Lett., 1999, 147, 17 CrossRef CAS.
E. Chlebda, A. Merwid-Ląd, D. Szumny, M. Trocha, L. Fereniec-Gołębiewska, E. Gębarowska, P. Kowalski and A. Szeląg, Pharmacol. Rep., 2007, 59, 269 Search PubMed.
K. Hamada and E. Kita, Chemotherapy, 1995, 41, 59 CrossRef CAS.
J. Yatsunami and S. Hayashi, Anticancer Res., 2001, 21, 4253 CAS.
R. Faghih, H. N. Ncllans, P. A. Lartey, A. Petersen, K. Marsh, Y. L. Bennani and J. J. Plattner, Bioorg. Med. Chem. Lett., 1998, 8, 805 CrossRef CAS.
C. Agouridas, A. Denis, J. M. Auger, Y. Benedetti, A. Bonnefoy, F. Bretin, J. F. Chantot, A. Dussarat, C. Fromentin, S. G. D'Ambrières, S. Lachaud, P. Laurin, O. L. Martret, V. Loyau and N. Tessot, J. Med. Chem., 1998, 41, 4080 CrossRef CAS.

Footnote

† Electronic supplementary information (ESI) available: Full experimental procedures. See DOI: 10.1039/c0md00103a

Click here to see how this site uses Cookies. View our privacy policy here.