Synthesis of combretastatin A-4 O-alkyl derivatives and evaluation of their cytotoxic, antiangiogenic and antitelomerase activity

Título artículo / Títol article: Synthesis of combretastatin A-4 O-alkyl derivatives and evaluation of their cytotoxic, antiangiogenic and antitelomerase activity Autores / Autors Torijano Gutiérrez, Sandra Adela ; Díaz Oltra, Santiago ; Falomir Ventura, Eva ; Murga, Juan ; Carda Usó, Miguel ; Marco, J. Alberto Revista: Bioorganic and Medicinal Chemistry, 2013, Vol. 21, no. 23 Versión / Versió: Preprint del autor Cita bibliográfica / Cita bibliogràfica (ISO 690): TORIJANO-GUTIÉRREZ, Sandra, et al. Synthesis of combretastatin A-4< i> O</i>-alkyl derivatives and evaluation of their cytotoxic, antiangiogenic and antitelomerase activity. Bioorganic & medicinal chemistry, 2013, vol. 21, no 23, p. 7267-7274 url Repositori UJI: http://hdl.handle.net/10234/83448 Graphical Abstract To create your abstract, type over the instructions in the template box below. Fonts or abstract dimensions should not be changed or altered. Synthesis of combretastatin O-alkyl derivatives and evaluation of their cytotoxic, antiangiogenic and antitelomerase activity Sandra Torijano-Gutiérrez, Santiago Díaz-Oltra, Eva Falomir, Juan Murga, Miguel Carda and J. Alberto Marco Dept. de Q. Inorgánica y Orgánica, Univ. Jaume I, E-12071 Castellón; Dept. Q. Orgánica, Univ. Valencia, E-46100 Burjassot, Valencia, Spain Leave this area blank for abstract info. Bioorganic & Medicinal Chemistry journal homepage: www.elsevier.com Synthesis of combretastatin O-alkyl derivatives and evaluation of their cytotoxic, antiangiogenic and antitelomerase activity* Sandra Torijano-Gutiérrez,a Santiago Díaz-Oltra,a Eva Falomir,a,*1 Juan Murga,a Miguel Cardaa and J. Alberto Marcob,** aDepart. de Q. Inorgánica y Orgánica, Univ. Jaume I, E-12071 Castellón, Spain bDepart. de Q. Orgánica, Univ. de Valencia, E-46100 Burjassot, Valencia, Spain ——— * Dedicated to Prof. G. Asensio, University of Valencia, on the occasion of his 65th birthday. ** Corresponding authors. Fax numbers: +34-964-728214, +34-96-3543880; E-mail addresses: [email protected], [email protected] 1. Introduction It is widely known that cancer, one leading cause of death in developed countries,1 may be induced by a plethora of both external and internal factors, including genetic mutations. Accordingly, a number of types of therapeutic attack has been investigated.2,3 One of these involves the use of cytotoxic drugs, which exert their effect in many cases by means of inducing cell apoptosis.4 A second type of therapeutic strategy against cancer is based on the use of compounds with vascular-targeting properties. These may be due to their ability to either inhibit the formation of new blood vessels (antiangiogenic agents) or else to promote the destruction of existing ones (antivascular agents).5,6 Tumor angiogenesis is a very complex process and involves the tight interplay of many factors.7 One of these is a protein called vascular endothelial growth factor (VEGF), a key regulator of angiogenesis which promotes endothelial cell survival, proliferation and migration while increasing vascular permeability.8 In fact, overexpression in the production of VEGF has been reported to occur in various types of tumors.9 Not unexpectedly, VEGF has become one further key target molecule in cancer therapy.10,11 A third therapeutic line follows the path of the chromosomal telomeres.12 Telomeres are the terminal zones of chromosomes and display a special structure that fulfills at least two essential functions: a) they must be recognized as functional domains, thus distinguishing them from random chromosomal breaks that would stimulate the onset of repair mechanisms; b) they must prevent these DNA ends from fusing with other DNA ends.13 To comply with these functions, an appropriate amount of repetitive DNA sequences (telomeres) must be added to the ends of the chromosomes. This task is fulfilled by a special type of ribonucleoprotein complex called telomerase, an enzyme with reverse transcriptase activity. The expression of this enzyme is restricted or absent in normal human somatic cells and so telomeres progressively shorten during cell lifespan.12 This triggers a DNA damage response which culminates in cell senescence or apoptosis.14 Shortening of telomeres during cell division provides a barrier for tumor progression. Indeed, cancer cells have evolved the ability to overcome senescence by using mechanisms capable of maintaining telomere lengths, such as expressing telomerase, which enables them to divide indefinitely. This uncontrolled cell growth is a key feature of cancer malignancy and has been found to correlate with telomerase reactivation.15 For this reason, and ARTICLE INFO ABSTRACT Article history: Received Received in revised form Accepted Available online We here report the synthesis and biological evaluation of several combretastatin derivatives alkylated at the phenol hydroxyl group. Some of these derivatives contain an (E)-arylalkene fragment reminiscent of that present in some natural stilbenes like resveratrol. The cytotoxicities towards one human healthy kidney embryonic and two tumoral cell lines were determined. In addition, the ability of these compounds to inhibit the production of the vascular endothelial growth factor (VEGF) was measured. Finally, the expression of genes controlling the production of telomerase was measured. Some of the compounds were found to have an activity comparable or higher than that of combretastatin A4 in at least one of the aforementioned biological properties. The compounds with the (E)-arylalkene fragment were in general terms more active than the simple O-alkyl derivatives. However, no clear structure/activity correlations were perceived when comparing the observed compound activities across the three biological properties. This points out the existence of marked differences between the mechanisms responsible for their cytotoxicity. 2013 Elsevier Ltd. All rights reserved. Keywords: Anticancer drugs Cytotoxic compounds Antiangiogenic compounds VEGF Telomerase Combretastatin A-4 derivatives since telomerase has been detected in about 90% of all malignant tumors,12 drugs with ability to inhibit telomerase activity are potentially useful weapons in the fight against cancer, aging and other diseases, including some related to premature telomere shortening.16 The emergence of resistances to existing drugs has led to a continuous need of developing new bioactive compounds that overcome such problems. Even though first observed in the case of antibiotics, resistances have been reported to therapies with various types of cytotoxic,17 antiangiogenic18 and antitelomerase agents.19 The discovery and investigation of new members of these compound classes therefore constitutes an important goal in chemistry and pharmacology. Among the new drug types, the combretastatin family, which belongs like the well-known resveratrol to the stilbene class of natural products,20 has acquired an outstanding status in the last years and their members have been found utility in various pharmacological applications.21 The knowledge of the chemistry, biology, and medical potential of this particular compound class, discovered about 30 years ago in the African tree Combretum caffrum, has continued to advance.22 Preclinical and clinical developments over the last decade have been rapidly accelerating for drugs such as combretastatin A-4 (1a), most particularly in the form of its phosphate prodrug CA4P (1b), and for combretastatin A-1 (1c) as promising cancer vascular-disrupting and ophthalmology drugs (Fig. 1).22,23 Indeed, these encouraging developments have stimulated a variety of efforts devoted to the synthesis and biological evaluation of numerous combretastatin structural modifications.24 Recent reports include SAR studies that provide varying levels of cancer cell growth inhibition.25 Figure 1. Structures of some combretastatins (1a-1c) and resveratrol. 2. Research purpose In the present paper, we are disclosing our results in the synthesis and biological evaluation of a number of O-alkyl derivatives of combretastatin A-4 1a. These are arranged in two series as shown in Figure 2. One includes compounds 2a-e having O-alkyl residues of the allyl and benzyl type. The second series encompasses compounds 2f-n, which contain (E)-5arylpent-4-enyl residues with various aryl groups. This part of the structure displays an arylalkene fragment which shows similarity to a part of the structure of antiangiogenic stilbenes of the resveratrol type (Fig. 1).20,26 Accordingly, it may be expected that these compounds could not only exhibit cytotoxicity but also antiangiogenic activity. Thus, we here present the results of our measurements of the cytotoxicities of compounds 2a-n but also of their ability to inhibit the production of the VEGF. Finally, and in order to cover all three aforementioned lines of therapeutic attack, we also have tested their ability to inhibit telomerase activity. In our belief, the observed results might be possibly helpful in aiding to clear the mechanisms of action of these compounds. Figure 2. Structures of combretastatin O-alkyl derivatives 2a-2n used in this study (MOM = MeOCH2). 3. Chemical results Combretastatin A-4 1a, the starting material for all compounds discussed here, was prepared according to a literature procedure.27 Conversion into O-alkyl derivatives 2a-2n was performed as depicted in Scheme 1 using bromides 3a-3n (Scheme 2) as the alkylating reagents. Yields are given in the experimental part. Scheme 1. Synthesis of combretastatin O-alkyl derivatives 2a-2n. From the alkyl bromides mentioned in Scheme 1, 3a-3e are commercially available. Bromides 3f-3n were synthesized by means of ruthenium-catalyzed cross metathesis28 between the commercially available bromide 4 and styrene derivatives 5-13 (Scheme 2). Yields are given in the experimental part. Scheme 2. Synthesis of alkyl bromides 3f-3n. 4. Biological results 4.1. Cytotoxicity of combretastatine derivatives We first carried out a measurement of the cytotoxicity of the synthetic combretastatin derivatives 2a-2n. To this purpose, MTT assays were performed using two tumoral cells, the human colon HT-29 and the breast adenocarcinoma MCF-7 cell lines, as well as one normal cell line, the human kidney embryonic cell line, HEK-293.29 Cytotoxicity values, expressed as the compound concentration (µmol/L) that causes 50% inhibition of cell growth (IC50), are shown in Table 1. The observed values are in the low to medium micromolar range, with compound 2i showing the lowest values, not very different of those of combretastatin A4 for these two particular cell lines. In addition, it is worth mentioning that some of the synthetic compounds are much more toxic for tumoral cells than for normal ones, an obviously desirable feature. This can better appreciated with the α and β coefficients, obtained by dividing the IC50 values of the normal cell line by those of one or the other tumoral cell line (see footnote in the Table). The highest value of either coefficient, the highest the therapeutic safety margin of the compound in the corresponding cell line. Thus, combretastatin A4 shows high values of both coefficients, most particularly in the case of the MCF-7 cell line. This turns out also to be the case of compounds 2f and 2i. Compounds 2m and 2n show a good selectivity only in the case of the HT-29 line (α > 4) whereas compounds 2b, 2h and 2l show a good selectivity in the specific case of the MCF-7 line (β > 4). Table 1. Cytotoxicity of combretastatin A4 derivatives 2a-2n.a Compound HT-29 MCF-7 HEK-293 αb βc CoA4 4.2 ± 0.5 1 ± 0.2 25 ± 3 5.9 25 2a 29 ± 3 12.0 ± 0.4 43.5 ± 0.8 1.5 3.6 2b 161 ± 5 5.4 ± 0.4 135 ± 13 0.8 25 2c 39 ± 6 59 ± 2 15 ± 3 0.4 0.3 2d 47 ± 5 48 ± 1 64 ± 7 1.4 1.3 2e 108 ± 1 29 ± 1 45 ± 4 0.4 1.6 2f 86 ± 6 21 ± 2 >400 >4.6 >19 2g 25 ± 3 41 ± 3 39 ± 2 1.6 1.0 2h 47 ± 7 16 ± 3 88 ± 6 1.9 5.5 2i 9.6 ± 0.2 4.4 ± 0.8 106 ± 9 11 24.1 2j 59 ± 7 8 ± 3 30 ± 2 0.5 3.7 2k 7 ± 0.4 42 ± 7 25 ± 4 3.6 0.6 2l 98 ± 8 16 ± 2 115 ± 5 1.2 7.2 2m 18 ± 3 40 ± 5 111 ± 3 6.2 2.8 2n 11 ± 2 37 ± 8 55 ± 5 5.0 1.5 aIC50 values include those of combretastatin A4 itself and are expressed as the compound concentration (µmol/L) that causes 50% inhibition of cell growth, and are the average (± s.d.) of three different measurements (described in the Material and Methods section). bα = IC50 (HEK-293) / IC50 (HT-29). cβ = IC50 (HEK-293) / IC50 (MCF-7). Values of α and β have been rounded off to a decimal figure. The lowest IC50 values (for tumoral cells) and the highest values of the α and/or β coefficients (> 4) have been highlighted in italics. 4.2. Effect of combretastatin A4 derivatives on VEGF production According to that discussed in the Introduction section, we next investigated whether combretastatin derivatives 2a-n were able to inhibit or at least decrease the production of the VEGF protein in HT-29 tumoral cells. Figure 3 shows the results of VEGF production obtained by means of ELISA measurements after treatment of HT-29 tumoral cells with combretastatin A4 and with compounds 2a-n dissolved in DMSO, which was the control substance in all experiments (in all cases, concentrations values below IC50 were used). With the control substance, the observed VEGF production was standardized to 100%, the other values being then referred to it. Thus, compound 2g caused the highest degree of inhibition in the VEGF production, which underwent a reduction to 38% of the control value. This is an even stronger effect than that caused by combretastatin A4, where VEGF production was reduced to 47% of the control value. Similarly strong effects were observed in the cases of compounds 2b (44%) and 2i (42%). The remaining compounds showed less favourable values, with compound 2e being essentially inactive (99%). Figure 3. VEGF protein production from HT-29 cells treated with DMSO (control), combretastatin A4 (CoA4) and derivatives 2a-n (at least three measurements were performed in each case). Bars represent mean values of VEGF expression (in ng/mL) and error bars indicate standard errors of the mean. The statistical significance was evaluated using one-sample t-tests (P < 0.001). While the results depicted in Fig. 3 point out that compounds 2a-n cause inhibition of the VEGF production, they do not say anything about the precise phase of the VEGF generation process with which they interfere. In order to deepen into the knowledge of this issue, we proceeded to determine whether the compounds under study were able to control protein production at the transcriptional level. With this idea in mind, we performed a reverse transcriptase/polymerase chain reaction (RT-PCR) analysis. For this purpose, we selected six of the combretastatin A4 derivatives with the highest anti-VEGF activities (2b, 2g, 2i, 2j, 2k and 2n). As shown in Figure 4, treatment of HT-29 tumoral cells with these derivatives and with combretastatin A4 in DMSO did in fact cause a reduction of the transcription of VEGF mRNA as compared with control cells (the values of these are standardized to 100%). The most active derivatives turned out to be 2b and 2k, which proved able to reduce the expression of the corresponding gene to less than 70% of the control value. These effects are thus stronger than that of combretastatin A4 itself, which only causes a slight reduction of the transcription level of VEGF mRNA (91%). Interestingly, compound 2g, which showed the highest degree of inhibition of VEGF production (Fig. 3), leaved the transcription of VEGF mRNA practically unaltered (97%). Table 2 present the same results depicted in Figures 3 and 4 for the aforementioned six compounds (2b, 2g, 2i, 2j, 2k and 2n) although in the form of percentages of inhibition of VEGF production (100 ‒ % VEGF production) and of gene expression (100 ‒ % gene expression): thus, the highest values correspond to the strongest inhibitory effect. There is a visible lack of parallelism between the data of protein inhibition and those of gene inhibition. This suggests that these combretastatin A4 derivatives exert the control of VEGF production at a phase different from that of gene transcription, perhaps during the posttranslational stage. Table 2. Inhibition of VEGF protein and of gene expression by combretastatin A4 and some of its O-alkyl derivatives.a Compound % Protein inhibitionb % Gene inhibitionc CoA4 53 9 2b 56 31 2g 62 3 2i 58 20 2j 49 10 2k 49 34 2n 52 27 aConcentrations used were below the IC50 values. bValues obtained by subtracting from 100 the percentages of VEGF secretion values in Fig. 3. cValues obtained by subtracting from 100 the percentages of VEGF gene inhibition values in Fig. 4. Figure 4. Agarose gel profile of products resulting from RT-PCR amplification. The total RNA of HT-29 cells previously treated with the appropriate combretastatin A4 derivative was isolated, converted into cDNA, and amplified by PCR as described in the Materials and Methods section (primers used for the RT-PCR are shown in Table 3). Gene expression of VEGF and β-actin was quantified using the Image J program and normalized to that of the housekeeping gene β-actin. At least three measurements were performed in each case. Bars shown represent mean activations of VEGF gene expression and error bars indicate standard errors of the mean. The statistical significance was evaluated using one-sample t-tests (P < 0.001). 4.3. Effect of combretastatin A4 derivatives on telomerase production We were also interested in designing compounds with the ability to inhibit the expression of telomerase in tumoral cells. The role of this important ribonucleoprotein has been referred to in the Introduction.12-16 Human telomerase contains an RNA component (hTR) that serves as a template for the addition of the repeat nucleotide sequences and a protein subunit (hTERT) which catalyzes the nucleotide polymerization process. In addition, there are other associated protein factors, the role of which has not yet been completely elucidated. Human telomerase is regulated during development and differentiation, mainly through transcriptional control of the hTERT gene, the expression of which is restricted to cells that exhibit telomerase activity. This indicates that hTERT is the rate limiting factor of the enzyme complex.16a For the expression of the hTERT gene, two transcriptional factors called c-Myc and Sp1, among others, have been found to play an important role through upregulation of the mRNA encoding the hTERT protein subunit of telomerase.30 Thus, and as a preliminary study of the potential anti-telomerase activity of combretastatin derivatives, we have investigated their ability to inhibit the expression of the hTERT and c-myc genes. Figure 5. Agarose gel profile of products resulting from RT-PCR amplification. The total RNA of HT-29 cells previously treated with the appropriate compound was isolated, converted into cDNA, and amplified by PCR as described in the Materials and Methods section (primers used for the RT-PCR are shown in Table 3). Gene expression of hTERT, c-myc and βactin was quantified using the Image J program and normalized to that of the housekeeping gene β-actin. At least three measurements were performed in each case. Bars shown represent mean activations of hTERT and c-myc gene expression and error bars indicate standard errors of the mean. The statistical significance was evaluated using one-sample t-tests (P < 0.001). In order to determine whether O-alkyl derivatives of combretastatin A4 were able to regulate the expression of the hTERT and c-myc genes, we have performed a RT-PCR analysis using HT-29 tumoral cells. For that purpose, we selected the same group of derivatives (2b, 2g, 2i, 2j, 2k and 2n) previously investigated for their antiangiogenic activity. The results, depicted in Figure 5, show that treatment of HT-29 cells with combretastatin A4 and the aforementioned derivatives dissolved in DMSO leads in fact to various degrees of reduction in the transcription of hTERT and c-myc mRNA as compared with control cells. The most active derivative was found to be 2n, which proved able to reduce the expression of both genes to less than 40% of the control value. For the sake of comparison, combretastatin A4 caused a reduction to 74% of the control value in the case of the hTERT gene and proved practically inactive in the case of the c-myc gene. Comparison of the results for the expression of the hTERT gene with those of the c-myc gene reveals a good correlation between them (Fig. 5). We may conclude therefore that the compounds under study downregulate the expression of the hTERT gene by lowering the transcription of the c-myc gene. 5. Summary A series of O-alkyl derivatives of combretastatin A4 has been prepared and evaluated in relation to three types of biological properties: cytotoxicity, antiangiogenesis and telomerase inhibition. One healthy (human kidney embryonic, HEK-293) and two tumoral cell lines (human colon HT-29 and breast adenocarcinoma MCF-7) were used for the assays. No clear correlations are perceived between structure and activity in the compounds under study. For instance, the strongest cytotoxicities (lowest IC50 values) and the highest α or β values are found almost exclusively among the compounds containing the (E)-arylalkene fragment, i.e., within the 2f-2n group (Table 1). However, the observed IC50 values do not bear a close relationship with the α/β coefficients. Only in the case of compound 2i was an appreciable cytotoxicity towards both tumoral cell lines accompanied by high α and β values. In other cases (e.g. 2f, 2h, 2n), these desirable features were observed in only one of the two cell lines. The inhibition of the VEGF production, as a measure of the antiangiogenic activity, did not show a very marked relation with the structural type, either. Except for 2c and 2e, which were clearly less active, all other compounds displayed comparable activities, with 2g being the most active. The most outstanding aspect here was that some of the compounds (2b, 2g and 2i) proved even more active in this property than combretastatin A4. Again, these belong to the 2f-2n group. From these, 2i also showed favourable cytotoxicity indexes, as commented above. The ability to inhibit the expression of the hTERT and c-myc genes was also found in the investigated combretastatin A4 derivatives. The profile was, however, clearly different from that observed in the two other biological properties. For example, the strongest activity was found here in compound 2n, with compounds 2b, 2g and 2i, which had favourable antiangiogenic features, being much less active. Nonetheless, it is worth noting that several of the compounds displayed a higher activity than combretastatin A4. Once again, the compounds of the 2f-2n group were found more active than the simple O-alkyl derivatives. In summary, some of the investigated combretastatin A4 Oalkyl derivatives show an activity comparable or higher than that of combretastatin A4 itself in at least one of the three examined biological properties. While it seems that compounds containing the (E)-arylalkene fragment (2f-2n) display in general stronger activities than the simple O-alkyl derivatives (2a-2e), no clear structure/activity correlations were perceived, however, when comparing the observed compound activities across the three biological properties. This points out the existence of marked differences between the mechanisms responsible for their cytotoxicity. 6. Materials and Methods 6.1. Chemistry: general procedures General features. NMR spectra were recorded at 500 MHz (1H NMR) and 125 MHz (13C NMR) in CDCl3 solution at 25 °C, if not otherwise indicated, with CDCl3 signals as internal reference. 13C NMR signal multiplicities were determined with the DEPT pulse sequence. Mass spectra were run in the electrospray (ESMS) mode. Reactions which required an inert atmosphere were carried out under dry N2 with flame-dried glassware. Commercial reagents were used as received. Dichloromethane was freshly distilled from CaH2. Column chromatography was performed on a silica gel column (60-200 µm) with elution with the indicated solvent mixtures. 6.2. Reaction conditions 6.2.1. Styrene derivatives Styrene derivatives 5-7 and 11-13 are commercially available. Styrenes 8-10 were prepared as reported31 by means of Wittig methylenation of the appropriate commercial benzaldehydes (yields in the range 60-70%): 8: oil; 1H NMR (500 MHz) δ 7.54 (1H, dd, J = 7.7, 1.5 Hz), 7.26 (1H, td, J = 7.7, 1.5 Hz), 7.20-7.10 (2H, m), 7.04 (1H, td, J = 7.5, 1 Hz), 5.80 (1H, dd, J = 17.7, 1.5 Hz), 5.33 (1H, dd, J = 11, 1.5 Hz), 5.25 (2H, s), 3.53 (3H, s); 13C NMR (125 MHz) δ 154.4, 127.6 (C), 131.5, 128.8, 126.4, 121.9, 114.8 (CH), 114.5, 94.7 (CH2), 56.0 (CH3). 9: oil; 1H NMR (500 MHz) δ 7.26 (1H, t, J = 7.9 Hz), 7.12 (1H, br t, J ∼ 2 Hz), 7.08 (1H, br d, J ∼ 7.6), 6.97 (1H, ddd, J = 7.9, 2.4, 1 Hz), 6.71 (1H, dd, J = 17.5, 10.8 Hz), 5.77 (1H, dd, J = 17.5, 1 Hz), 5.27 (1H, dd, J = 10.8, 1 Hz), 5.21 (2H, s), 3.51 (3H, s); 13C NMR (125 MHz) δ 157.6, 139.2 (C), 136.7, 129.5, 120.1, 115.7, 114.0 (CH), 114.3, 94.5 (CH2), 56.0 (CH3). 10: oil; 1H NMR (500 MHz) δ 7.38 (2H, br d, J ∼ 8.5 Hz), 7.05 (2H, br d, J ∼ 8.5 Hz), 6.71 (1H, dd, J = 17.6, 11 Hz), 5.77 (1H, d, J = 17.6 Hz), 5.21 (2H, s), 5.20 (1H, d, J = 11 Hz), 3.52 (3H, s); 13C NMR (125 MHz) δ 157.0, 131.6 (C), 136.2, 127.3 (x 2), 116.3 (x 2) (CH), 112.0, 94.4 (CH2), 55.9 (CH3). 6.2.2. Synthesis of bromides 3f-3n by means of cross metathesis Representative example: a solution of bromide 4 (60 µL, 75 mg, ca. 0.5 mmol) and styrene 6 (235 mg, 1.75 mmol) in dry, degassed CH2Cl2 (50 mL) was treated with Ru-II catalyst (150 mg, 0.175 mmol). The reaction mixture was then stirred at reflux under N2 for 24 h. Subsequently, the mixture was treated with DMSO (600 µL)32 and stirred overnight at room temperature. The reaction mixture was then evaporated under reduced pressure, and the residue was carefully chromatographed on silica gel (elution with hexane-Et2O, 400:1)33. This gave 3g (55 mg, 43% based on 4) together with the stilbene derivative generated by homodimerization of 6 (major product). In the other examples, yields were in the range 35-50% except for bromides 3l-3n, which could not be purified and were obtained only in admixture with variable percentages of the homodimerization products. The mixtures were then used for the alkylation step. 3f: oil; 1H NMR (500 MHz) δ 7.43 (1H, br d, J ∼ 7.8 Hz), 7.22 (1H, br t, J ∼ 7.8 Hz), 6.94 (1H, t, J = 7.8 Hz), 6.88 (1H, br d, J ∼ 7.8 Hz), 6.79 (1H, d, J = 16 Hz), 6.20 (1H, dt, J = 16, 7 Hz), 3.87 (3H, s), 3.48 (2H, t, J = 7 Hz), 2.42 (2H, br q, J ∼ 7 Hz), 2.07 (2H, br quint, J ∼ 7 Hz); 13C NMR (125 MHz) δ 156.4, 126.4 (C), 129.2, 128.1, 126.5, 126.0, 120.6, 110.9 (CH), 33.2, 32.4, 31.8 (CH2), 55.5 (CH3); HR EIMS m/z (rel. int.) 254.0351 (M+, 40), 239 (M+−Me, 100). Calcd. for C12H15 79BrO, 254.0306. 3g: oil; 1H NMR (500 MHz) δ 7.23 (1H, t, J = 7.8 Hz), 6.96 (1H, br d, J ∼ 7.8 Hz), 6.91 (1H, br t, J ∼ 2 Hz), 6.79 (1H, dd, J = 7.8, 2.5 Hz), 6.44 (1H, d, J = 16 Hz), 6.18 (1H, dt, J = 16, 7 Hz), 3.83 (3H, s), 3.47 (2H, t, J = 7 Hz), 2.40 (2H, br q, J ∼ 7 Hz), 2.05 (2H, br quint, J ∼ 7 Hz); 13C NMR (125 MHz) δ 159.8, 138.9 (C), 131.2, 129.5, 128.8, 118.7, 112.8, 111.4 (CH), 33.1, 32.2, 31.2 (CH2), 55.2 (CH3); HR EIMS m/z (rel. int.) 254.0336 (M+, 80), 175 (M+−Br, 100), 147 (M+−Br−C2H4, 98). Calcd. for C12H15 79BrO, 254.0306. 3h: oil; 1H NMR (500 MHz) δ 7.30 (2H, br d, J ∼ 8.5 Hz), 6.86 (2H, br d, J ∼ 8.5 Hz), 6.41 (1H, d, J = 16 Hz), 6.04 (1H, dt, J = 16, 7 Hz), 3.82 (3H, s), 3.46 (2H, t, J = 7 Hz), 2.38 (2H, br q, J ∼ 7 Hz), 2.05 (2H, br quint, J ∼ 7 Hz); 13C NMR (125 MHz) δ 158.9, 128.6 (C), 130.7 (x 2), 127.1 (x 2), 126.3, 114.0 (CH), 33.2, 32.4, 31.3 (CH2), 55.3 (CH3); HR EIMS m/z (rel. int.) 254.0362 (M+, 45), 175 (M+−Br, 12), 147 (M+−Br−C2H4, 100). Calcd. for C12H15 79BrO, 254.0306. 3i: oil; 1H NMR (500 MHz) δ 7.45 (1H, br d, J ∼ 7.8 Hz), 7.19 (1H, br t, J ∼ 7.8 Hz), 7.10 (1H, br d, J ∼ 7.8 Hz), 6.98 (1H, br t, J ∼ 7.8 Hz), 6.81 (1H, d, J = 16 Hz), 6.18 (1H, dt, J = 16, 7 Hz), 5.23 (2H, s), 3.52 (3H, s), 3.49 (2H, t, J = 7 Hz), 2.42 (2H, br q, J ∼ 7 Hz), 2.07 (2H, br quint, J ∼ 7 Hz); 13C NMR (125 MHz) δ 154.1, 127.4 (C), 129.2, 128.1, 126.4, 125.8, 122.0, 115.0 (CH), 94.8, 33.2, 32.4, 31.7 (CH2), 56.1 (CH3); HR EIMS m/z (rel. int.) 284.0428 (M+, 60), 239 (M+−CH2OMe, 100). Calcd. for C13H17 79BrO2, 284.0412. 3j: oil; 1H NMR (500 MHz) δ 7.23 (1H, t, J = 7.8 Hz), 7.05 (1H, br t, J ∼ 2 Hz), 7.01 (1H, br d, J ∼ 7.8 Hz), 6.92 (1H, br dd, J ∼ 7.8, 2.2 Hz), 6.43 (1H, d, J = 15.8 Hz), 6.18 (1H, dt, J = 15.8, 6.8 Hz), 5.20 (2H, s), 3.50 (3H, s), 3.47 (2H, t, J = 6.8 Hz), 2.39 (2H, br q, J ∼ 6.8 Hz), 2.05 (2H, br quint, J ∼ 6.8 Hz); 13C NMR (125 MHz) δ 157.5, 139.0 (C), 131.1, 129.5, 128.9, 119.8, 115.1, 113.7 (CH), 94.4, 33.0, 32.2, 31.2 (CH2), 55.9 (CH3); HR EIMS m/z (rel. int.) 284.0416 (M+, 100), 254 (M+−CH2O, 36), 175 (M+−CH2O −Br, 65). Calcd. for C13H17 79BrO2, 284.0412. 3k: oil; 1H NMR (500 MHz) δ 7.29 (2H, br d, J ∼ 8.5 Hz), 6.99 (2H, br d, J ∼ 8.5 Hz), 6.41 (1H, d, J = 15.9 Hz), 6.05 (1H, dt, J = 15.9, 6.8 Hz), 5.18 (2H, s), 3.49 (3H, s), 3.46 (2H, t, J = 6.8 Hz), 2.37 (2H, br q, J ∼ 6.8 Hz), 2.04 (2H, br quint, J ∼ 6.8 Hz); 13C NMR (125 MHz) δ 156.5, 131.5 (C), 130.6, 127.1 (x 2), 126.8, 116.3 (x 2) (CH), 94.5, 33.1, 32.3, 31.2 (CH2), 55.9 (CH3); HR EIMS m/z (rel. int.) 284.0353 (M+, 85), 254 (M+−CH2O, 76), 147 (C5H8Br+, 100). Calcd. for C13H17 79BrO2, 284.0412. 6.2.3. Synthesis of combretastatin A-4 derivatives 2a-2n Representative example: A solution of 1a (63 mg, 0.2 mmol) in dry DMF (3 mL) was treated under N2 in the dark at room temperature with K2CO3 (70 mg, ca. 0.5 mmol) and stirred for 1 h. Subsequently, 3a (52 µL, 0.6 mmol) was added and the stirring was continued for 24 h under the same conditions. The reaction mixture was then poured into saturated aqueous NH4Cl and extracted three times with Et2O (3 x 15 mL). The organic layer was dried over anhydrous Na2SO4, filtered and evaporated under reduced pressure. Column chromatography of the residue on silica gel (elution with hexane-EtOAc, 9:1) provided 2a (50 mg, 70%). In the other cases, yields were in the range 40-70%. Bromides 3l-3n were used in excess as the mixtures with the homodimerization products (see above). Caution: reaction, work-up and purification procedures should be performed under minimization of exposure to light, due to the ease of photoinduced Z/E isomerization in the combretastatin moiety. 2a: solid, mp 107-108 °C; 1H NMR (500 MHz) δ 6.86 (1H, dd, J = 8.3, 1.7 Hz), 6.84 (1H, d, J = 1.7 Hz), 6.79 (1H, d, J = 8.3 Hz), 6.52 (2H, s), 6.50 (1H, d, J = 12.2 Hz), 6.46 (1H, d, J = 12.2 Hz), 5.95 (1H, ddt. J = 17.2, 11.4, 5.5 Hz), 5.27 (1H, dq, J = 17.2, 1.5 Hz), 5.20 (1H, dq, J = 11.4, 1.5 Hz), 4.40 (2H, dt, J = 5.5, 1.5 Hz), 3.86 (3H, s), 3.84 (3H, s), 3.71 (6H, s); 13C NMR (125 MHz) δ 153.0 (x 2), 148.7, 147.5, 137.2, 133.0, 129.8 (C), 133.1, 129.7, 128.8, 122.3, 113.9, 111.3, 106.0 (x 2) (CH), 117.9, 69.7 (CH2), 60.9, 56.0 (x 3) (CH3); HR ESMS m/z 379.1518 (M+Na+). Calcd. for C21H24NaO5, 379.1521. 2b: oil; 1H NMR (500 MHz) δ 6.86 (2H, m), 6.74 (1H, d, J = 8.5 Hz), 6.53 (2H, s), 6.50 (1H, d, J = 12.1 Hz), 6.44 (1H, d, J = 12.1 Hz), 5.75-5.60 (2H, m), 4.30 (2H, br d, J ∼ 5.8 Hz), 3.85 (3H, s), 3.84 (3H, s), 3.72 (6H, s), 1.70 (3H, br dd, J ∼ 6.5, 1.5 Hz); 13C NMR (125 MHz) δ 153.0 (x 2), 148.7, 147.6, 137.2, 133.0, 129.8 (C), 130.8, 129.7, 128.7, 125.9, 122.0, 113.6, 111.1, 106.0 (x 2) (CH), 69.5 (CH2), 60.9, 56.0 (x 3), 17.7 (CH3); HR ESMS m/z 393.1677 (M+Na+). Calcd. for C22H26NaO5, 393.1678. 2c: solid, mp 83-84 °C; 1H NMR (500 MHz) δ 7.35-7.25 (5H, br m), 6.90-6.85 (2H, m), 6.80 (1H, d, J = 8.3 Hz), 6.51 (2H, s), 6.47 (1H, d, J = 12.2 Hz), 6.43 (1H, d, J = 12.2 Hz), 4.94 (2H, s), 3.87 (3H, s), 3.85 (3H, s), 3.71 (6H, s); 13C NMR (125 MHz) δ 153.0 (x 2), 148.9, 147.7, 137.2, 136.9, 133.0, 129.8 (C), 133.1, 129.6, 128.4 (x 2), 127.7, 127.2 (x 2), 122.4, 114.5, 111.5, 106.0 (x 2) (CH), 70.9 (CH2), 60.9, 56.0, 55.9 (x 2) (CH3); HR ESMS m/z 429.1674 (M+Na+). Calcd. for C25H26NaO5, 429.1678. 2d: solid, mp 62-63 °C; 1H NMR (500 MHz) δ 7.24 (2H, br d, J ∼ 8.7 Hz), 6.95-6.85 (3H, m), 6.78 (2H, br d, J = 8.7 Hz), 6.52 (2H, s), 6.47 (1H, d, J = 12.2 Hz), 6.43 (1H, d, J = 12.2 Hz), 4.87 (2H, s), 3.85 (3H, s), 3.84 (3H, s), 3.81 (3H, s), 3.71 (6H, s); 13C NMR (125 MHz) δ 159.4, 153.0 (x 2), 149.0, 147.8, 137.2, 133.0, 129.9, 129.1 (C), 129.7, 129.0 (x 2), 128.8, 122.4, 114.7, 113.9 (x 2), 111.5, 106.0 (x 2) (CH), 70.7 (CH2), 60.9, 55.9 (x 3), 55.3 (CH3); HR ESMS m/z 459.1783 (M+Na+). Calcd. for C26H28NaO6, 459.1784. 2e: oil; 1H NMR (500 MHz) δ 7.44 (2H, br d, J ∼ 8.4 Hz), 7.18 (2H, br d, J = 8.4 Hz), 6.88 (1H, dd, J = 8.3, 1.7 Hz), 6.83 (1H, d, J = 1.7 Hz), 6.80 (1H, d, J = 8.3 Hz), 6.50 (2H, s), 6.44 (2H, s), 4.87 (2H, s), 3.86 (3H, s), 3.85 (3H, s), 3.71 (6H, s); 13C NMR (125 MHz) δ 153.0 (x 2), 149.0, 147.4, 137.2, 136.0, 133.0, 129.8, 121.6 (C), 131.5 (x 2), 129.5, 128.8, 128.7 (x 2), 122.8, 114.6, 111.5, 106.0 (x 2) (CH), 70.2 (CH2), 60.9, 56.0 (x 3) (CH3); HR ESMS m/z 507.0782 (M+Na+). Calcd. for C25H25 79BrNaO5, 507.0783. 2f: oil; 1H NMR (500 MHz) δ 7.41 (1H, dd, J = 7.7, 1.5 Hz), 7.19 (1H, td, J = 7.7, 1.5 Hz), 6.91 (1H, td, J = 7.7, 1 Hz), 6.90-6.85 (3H, m), 6.78 (1H, d, J = 8 Hz), 6.74 (1H, br d, J ∼ 15.8 Hz), 6.53 (2H, s), 6.50 (1H, d, J = 12.2 Hz), 6.45 (1H, d, J = 12.2 Hz), 6.22 (1H, dt, J = 15.8, 6.8 Hz), 3.90 (2H, t, J = 6.8 Hz), 3.86 (3H, s), 3.84 (6H, s), 3.70 (6H, s), 2.36 (2H, br q, J ∼ 7 Hz), 1.94 (2H, br quint, J ∼ 7 Hz); 13C NMR (125 MHz) δ 156.3, 153.0 (x 2), 148.8, 148.1, 137.2, 133.0, 130.0, 126.8 (C), 130.4, 129.8, 128.8, 127.9, 126.5, 125.2, 122.0, 120.7, 113.9, 111.5, 110.8, 106.0 (x 2) (CH), 68.4, 29.8, 28.8 (CH2), 60.9, 56.1, 56.0 (x 2), 55.5 (CH3); HR ESMS m/z 513.2247 (M+Na+). Calcd. for C30H34NaO6, 513.2253. 2g: oil; 1H NMR (500 MHz) δ 7.21 (1H, t, J = 7.9 Hz), 6.94 (1H, br d, J ∼ 7.7 Hz), 6.90-6.85 (3H, m), 6.78 (1H, d, J = 8.6 Hz), 6.76 (1H, dd, J = 8.3, 2.5 Hz), 6.53 (2H, s), 6.50 (1H, d, J = 12 Hz), 6.45 (1H, d, J = 12 Hz), 6.38 (1H, br d, J ∼ 15.8 Hz), 6.23 (1H, dt, J = 15.8, 6.8 Hz), 3.88 (2H, t, J = 6.7 Hz), 3.86 (3H, s), 3.84 (3H, s), 3.82 (3H, s), 3.70 (6H, s), 2.34 (2H, br q, J ∼ 7 Hz), 1.92 (2H, br quint, J ∼ 7 Hz); 13C NMR (125 MHz) δ 160.0, 153.0 (x 2), 148.8, 148.0, 139.2, 137.2, 133.0, 130.0 (C), 130.4, 130.1, 129.7, 129.4, 128.8, 122.1, 118.7, 113.8, 112.5, 111.5 (x 2), 106.0 (x 2) (CH), 68.2, 29.4, 28.7 (CH2), 60.9, 56.0 (x 3), 55.2 (CH3); HR ESMS m/z 513.2255 (M+Na+). Calcd. for C30H34NaO6, 513.2253. 2h: oil; 1H NMR (500 MHz) δ 7.28 (2H, br d, J ∼ 8.5 Hz), 6.906.85 (4H, m), 6.79 (1H, d, J = 8.2 Hz), 6.54 (2H, s), 6.50 (1H, d, J = 12 Hz), 6.46 (1H, d, J = 12 Hz), 6.36 (1H, br d, J ∼ 15.8 Hz), 6.09 (1H, dt, J = 15.8, 6.8 Hz), 3.89 (2H, t, J = 6.7 Hz), 3.86 (3H, s), 3.85 (3H, s), 3.82 (3H, s), 3.71 (6H, s), 2.33 (2H, br q, J ∼ 7 Hz), 1.92 (2H, br quint, J ∼ 7 Hz); 13C NMR (125 MHz) δ 158.7, 153.0 (x 2), 148.8, 148.0, 137.2, 133.0, 130.6, 130.0 (C), 129.8, 129.7, 128.7, 127.5, 127.1 (x 2), 122.0, 113.9 (x 2), 113.7, 111.4, 106.0 (x 2) (CH), 68.2, 29.4, 28.8 (CH2), 60.9, 56.0 (x 3), 55.2 (CH3); HR ESMS m/z 513.2256 (M+Na+). Calcd. for C30H34NaO6, 513.2253. 2i: oil; 1H NMR (500 MHz) δ 7.44 (1H, dd, J = 7.9, 1.8 Hz), 7.16 (1H, td, J = 7.9, 1.8 Hz), 7.07 (1H, dd, J = 8, 1 Hz), 6.97 (1H, td, J = 7.6, 1 Hz), 6.90-6.85 (2H, m), 6.78 (1H, d, J = 8 Hz), 6.75 (1H, br d, J ∼ 15.8 Hz), 6.53 (2H, s), 6.50 (1H, d, J = 12 Hz), 6.45 (1H, d, J = 12 Hz), 6.22 (1H, dt, J = 15.8, 6.8 Hz), 5.20 (2H, s), 3.90 (2H, t, J = 6.8 Hz), 3.86 (3H, s), 3.84 (3H, s), 3.70 (6H, s), 3.48 (3H, s), 2.38 (2H, br q, J ∼ 7 Hz), 1.94 (2H, br quint, J ∼ 7 Hz); 13C NMR (125 MHz) δ 154.0, 153.0 (x 2), 148.8, 148.1, 137.2, 133.0, 130.0, 127.7 (C), 130.5, 129.8, 128.8, 127.9, 126.4, 125.0, 122.0, 121.9, 114.9, 113.8, 111.5, 106.0 (x 2) (CH), 94.8, 68.3, 29.8, 28.8 (CH2), 60.9, 56.1, 56.0, 55.9 (x 2) (CH3); HR ESMS m/z 543.2362 (M+Na+). Calcd. for C31H36NaO7, 543.2359. 2j: oil; 1H NMR (500 MHz) δ 7.21 (1H, t, J = 7.8 Hz), 7.02 (1H, br s), 7.00 (1H, br d, J ∼ 8 Hz), 6.90-6.85 (3H, m), 6.78 (1H, d, J = 8.4 Hz), 6.54 (2H, s), 6.50 (1H, d, J = 12 Hz), 6.46 (1H, d, J = 12 Hz), 6.38 (1H, br d, J ∼ 15.7 Hz), 6.23 (1H, dt, J = 15.7, 6.8 Hz), 5.19 (2H, s), 3.88 (2H, t, J = 6.6 Hz), 3.86 (3H, s), 3.84 (3H, s), 3.70 (6H, s), 3.49 (3H, s), 2.34 (2H, br q, J ∼ 7 Hz), 1.91 (2H, br quint, J ∼ 7 Hz); 13C NMR (125 MHz) δ 157.5, 153.0 (x 2), 148.8, 148.0, 139.2, 137.2, 133.0, 129.9 (C), 130.3, 130.2, 129.7, 129.4, 128.8, 122.0, 119.8, 114.8, 113.8, 113.7, 111.4, 106.0 (x 2) (CH), 94.4, 68.2, 29.4, 28.7 (CH2), 60.8, 55.9 (x 4) (CH3); HR ESMS m/z 543.2359 (M+Na+). Calcd. for C31H36NaO7, 543.2359. 2k: oil; 1H NMR (500 MHz) δ 7.26 (2H, br d, J ∼ 8.5 Hz), 6.97 (2H, br d, J ∼ 8.5 Hz), 6.90-6.85 (2H, m), 6.78 (1H, d, J = 8.2 Hz), 6.53 (2H, s), 6.50 (1H, d, J = 12.3 Hz), 6.45 (1H, d, J = 12.3 Hz), 6.36 (1H, br d, J ∼ 15.7 Hz), 6.10 (1H, dt, J = 15.7, 6.8 Hz), 5.17 (2H, s), 3.88 (2H, t, J = 6.8 Hz), 3.86 (3H, s), 3.84 (3H, s), 3.70 (6H, s), 3.48 (3H, s), 2.32 (2H, br q, J ∼ 7 Hz), 1.91 (2H, br quint, J ∼ 7 Hz); 13C NMR (125 MHz) δ 156.4, 153.0 (x 2), 148.8, 148.0, 137.2, 133.0, 131.8, 130.0 (C), 129.8, 129.7, 128.8, 128.1, 127.1 (x 2), 122.0, 116.3 (x 2), 113.8, 111.4, 106.0 (x 2) (CH), 94.5, 68.3, 29.4, 28.8 (CH2), 60.9, 56.0 (x 4) (CH3); HR ESMS m/z 543.2360 (M+Na+). Calcd. for C31H36NaO7, 543.2359. 2l: oil; 1H NMR (500 MHz) δ 7.30 (2H, m), 6.98 (2H, m), 6.906.85 (2H, m), 6.79 (1H, d, J = 8.3 Hz), 6.53 (2H, s), 6.50 (1H, d, J = 12.1 Hz), 6.45 (1H, d, J = 12.1 Hz), 6.37 (1H, br d, J ∼ 15.8 Hz), 6.14 (1H, dt, J = 15.8, 6.8 Hz), 3.88 (2H, t, J = 6.8 Hz), 3.86 (3H, s), 3.84 (3H, s), 3.70 (6H, s), 2.33 (2H, br q, J ∼ 7 Hz), 1.91 (2H, br quint, J ∼ 7 Hz); 13C NMR (125 MHz) δ 162.9/161.0 [d, 1JC−F ~ 244 Hz], 153.0 (x 2), 148.8, 148.0, 137.2, ∼133.9 [d, 4JC−F ~ 3 Hz], 133.0, 129.9 (C), 129.7, 129.5, 129.3, 128.8, ∼127.4 [d, 3JC−F ~ 8 Hz] (x 2), 122.1, 115.3/115.2 [d, 2JC−F ~ 21 Hz], (x 2), 113.7, 111.4, 106.0 (x 2) (CH), 68.2, 29.3, 28.7 (CH2), 60.9, 56.1, 56.0 (x 2) (CH3); HR ESMS m/z 501.2052 (M+Na+). Calcd. for C29H31FNaO5, 501.2053. 2m: oil; 1H NMR (500 MHz) δ 7.26 (4H, br s), 6.88 (1H, dd, J = 8.2, 1.7 Hz), 6.85 (1H, d, J = 1.7 Hz), 6.79 (1H, d, J = 8.2 Hz), 6.53 (2H, s), 6.50 (1H, d, J = 12 Hz), 6.45 (1H, d, J = 12 Hz), 6.36 (1H, br d, J ∼ 15.8 Hz), 6.21 (1H, dt, J = 15.8, 6.8 Hz), 3.87 (2H, t, J = 6.8 Hz), 3.85 (3H, s), 3.84 (3H, s), 3.70 (6H, s), 2.34 (2H, br q, J ∼ 7 Hz), 1.91 (2H, br quint, J ∼ 7 Hz); 13C NMR (125 MHz) δ 153.0 (x 2), 148.8, 147.9, 137.2, 136.2, 133.0, 132.5, 129.9 (C), 130.5, 129.7, 129.3, 128.8, 128.5 (x 2), 127.2 (x 2), 122.1, 113.7, 111.4, 106.0 (x 2) (CH), 68.2, 29.4, 28.6 (CH2), 60.9, 56.0 (x 3) (CH3); HR ESMS m/z 517.1761 (M+Na+). Calcd. for C29H31 35ClNaO5, 517.1758. 2n: oil; 1H NMR (500 MHz) δ 7.41 (2H, br d, J = 8.5 Hz), 7.20 (2H, br d, J = 8.5 Hz), 6.87 (1H, dd, J = 8.2, 1.8 Hz), 6.85 (1H, d, J = 1.8 Hz), 6.79 (1H, d, J = 8.2 Hz), 6.53 (2H, s), 6.50 (1H, d, J = 12 Hz), 6.45 (1H, d, J = 12 Hz), 6.35 (1H, br d, J ∼ 15.8 Hz), 6.22 (1H, dt, J = 15.8, 6.8 Hz), 3.87 (2H, t, J = 6.8 Hz), 3.85 (3H, s), 3.84 (3H, s), 3.70 (6H, s), 2.33 (2H, br q, J ∼ 7 Hz), 1.90 (2H, br quint, J ∼ 7 Hz); 13C NMR (125 MHz) δ 153.0 (x 2), 148.8, 147.9, 137.2, 136.6, 133.0, 131.5, 129.9 (C), 131.4 (x 2), 130.7, 129.7, 129.4, 128.8, 127.5 (x 2), 122.1, 113.7, 111.4, 106.0 (x 2) (CH), 68.2, 29.4, 28.6 (CH2), 60.9, 56.1, 56.0 (x 2) (CH3); HR ESMS m/z 561.1248 (M+Na+). Calcd. for C29H31 79BrNaO5, 561.1253.