Preparation of new alkyne-modified ansamitocins by mutasynthesis

Article abstract of DOI:10.3762/bjoc.10.49

The preparation of alkyne-modified ansamitocins by mutasynthetic supplementation of Actinosynnema pretiosum mutants with alkyne-substituted aminobenzoic acids is described. This modification paved the way to introduce a thiol linker by Huisgen-type cycloaddition which can principally be utilized to create tumor targeting conjugates. In bioactivity tests, only those new ansamitocin derivatives showed strong antiproliferative activity that bear an ester side chain at C-3.

Full text of DOI:10.3762/bjoc.10.49

Preparation of new alkyne-modified ansamitocins
by mutasynthesis
Kirsten Harmrolfs1, Lena Mancuso1, Binia Drung1, Florenz Sasse2
and Andreas Kirschning*1
Full Research Paper
Open Access
Address:
Beilstein J. Org. Chem. 2014, 10, 535–543.
1Institute of Organic Chemistry and Center of Biomolecular Drug
Research (BMWZ), Leibniz University Hannover, Schneiderberg 1b,
30167 Hannover, Germany and 2Department of Chemical Biology,
Helmholtz Center for Infectious Research (HZI), Inhoffenstraße 7,
D-38124 Braunschweig, Germany
doi:10.3762/bjoc.10.49
Received: 30 November 2013
Accepted: 29 January 2014
Published: 03 March 2014
Email:
This article is part of the Thematic Series "Natural products in synthesis
and biosynthesis".
Andreas Kirschning* - andreas.kirschning@oci.uni-hannover.de
* Corresponding author
Guest Editor: J. S. Dickschat
Keywords:
© 2014 Harmrolfs et al; licensee Beilstein-Institut.
License and terms: see end of document.
ansamitocins; antibiotics; antitumor agents; click chemistry;
mutasynthesis; natural products
Abstract
The preparation of alkyne-modified ansamitocins by mutasynthetic supplementation of Actinosynnema pretiosum mutants with
alkyne-substituted aminobenzoic acids is described. This modification paved the way to introduce a thiol linker by Huisgen-type
cycloaddition which can principally be utilized to create tumor targeting conjugates. In bioactivity tests, only those new ansami-
tocin derivatives showed strong antiproliferative activity that bear an ester side chain at C-3.
Introduction
Although natural products and analogues cover a large portion nological strategies has seen increased interest lately [1-3]. The
of the drug market particularly in the treatment of cancer as concepts either rely on a concise understanding of the biosyn-
well as bacterial and viral infections their role in pharmaceu- thesis of natural products or simply individual enzymes for in
tical research has decreased over the past two decades. In part, vitro applications [4]. In this context, producer strains that are
this development is due to their often limited accessibility as genetically engineered in the biosynthesis of important and
well as their structural complexity. This situation hampers their complex natural products have shown to be powerful synthetic
use as lead structures for which access to small compound tools, e.g., knock out mutants are key players in mutational
libraries is essential in order to perform structure–activity rela- biosynthesis, or in short mutasynthesis. This technique relies on
tionship studies. Besides semisynthetic and total synthesis the cellular uptake of modified biosynthetic intermediates.
approaches the combination of chemical synthesis with biotech- Processing of these intermediates, sometimes called mutasyn-
535

Beilstein J. Org. Chem. 2014, 10, 535–543.
thons, can provide complex secondary metabolites specifically We extended these studies by combining mutasynthesis and
modified as planned by choice of the synthetic modification semisynthesis and thus accessed four new tumor specific folic
incorporated into the mutasynthon [5-7].
acid/ansamitocin conjugates [28] (Scheme 2). Bromo-ansami-
tocin 6 was obtained by mutasynthesis and was synthetically
In earlier work, we demonstrated that the ansamitocins modified to the complex folic acid/drug conjugate 7. The
(maytansinoids) 3–5 are an ideal showcase for creating small vitamin folic acid has become a promising ligand for selec-
libraries by mutasynthesis [8-10]. These secondary metabolites tively targeting the folate receptor (FR) in cancer tissues where
exert strong antiproliferative activity towards different leukemia the FR is known to be overexpressed [29,30]. Folic acid has a
cell lines as well as human solid tumors. Inhibitory concentra- high affinity for the FR (Kd = 10−10 M), even when conjugated
tions were as low as 10−3 to 10−7 µg/mL [11] which resulted to a cytotoxin such as maytansin. An important feature of these
from binding to β-tubulin monomers [12]. The ansamitocins are conjugates is the linker concept that connects the drug to the
produced by Actinosynnema pretiosum. For our mutasynthetic tumor-specific ligand. The linker is commonly designed in a
studies we utilized an AHBA blocked mutant of A. pretiosum way that a release mechanism of the cytotoxin is part of the
[13-21]. 3-Amino-5-hydroxybenzoic acid (1, AHBA) is the molecular architecture of the conjugate [31].
starter building block of the PKS type I that is responsible for
the biosynthesis of the ansamitocin backbone [22]. This PKS is Disulfide linkers have shown to be well suited when utilizing
a well studied megaenzyme complex and after the action of the reducing power between extra- and intracellular milieus
tailoring enzymes succeeding the PKS machinery ansamitocins which results in cleavage and liberation of the drug. Mitomycin
3–5 are formed [23-27].
conjugates [32] were one of the earliest examples of folate
disulfide–drug conjugates and after the conjugate is internal-
In the present case, 1 is loaded on the starter module of the ized by endocytosis, it was demonstrated that the endosomes
polyketide synthase (Scheme 1). The last PKS module releases exert reductive cleavage.
and cyclizes seco-proansamitocin, likely by an ansamycin
amide synthase (gene asm9) [23-27], that generates the For conjugate 7 we found that disulfide cleavage provided a
19-membered macrocyclic lactam proansamitocin (2). From thiol derivative of ansamitocin P-3 (4, AP-3) 8 with still strong
here a set of tailoring enzymes transform proansamitocin into antiproliferative activity (IC50 < 10 nM) for different cancer cell
the bioactive final metabolites 3–5 following a predetermined lines. Importantly, the intact conjugate showed strong antipro-
logic that is only flexible in part (Scheme 1).
liferative activity for a FR+ cancer cell line but was inactive
Scheme 1: Short representation of ansamitocin biosynthesis.
536

Beilstein J. Org. Chem. 2014, 10, 535–543.
Scheme 2: Structures of bromo-ansamitocin derivative 6, folate-ansamitocin P-3 conjugate 7 and thiol 8.
against a FR− cell line. Indeed, despite the substantial size of alkynes C by cross coupling chemistry. Previously, we success-
the substituent remaining at C19 after reductive cleavage, fully utilized this combination of mutasynthesis and semisyn-
ansamitocin derivative 8 still showed sufficient antiprolif- thesis for 19-brominated ansamitocin derivatives [13]. Alter-
erative activity. This is in line with our observation that struc- natively, also alkyne-, vinyl- and allyl-substituted aminoben-
tural changes at the aromatic ring do not affect the biological zoic acids can serve as mutasynthons so that the corresponding
properties of the ansamitocins to a great extent [13-21].
alkynyl- or respectively alkenyl-substituted ansamitocin deriva-
tives can be accessed directly after fermentation. Alkynes can
In order to broaden the opportunities of this approach for the be further used for Huisgen-type cycloadditions better known as
ansamitocins, we now describe alternative accesses towards “copper mediated Click-chemistry” while alkenes, especially
disulfide linked conjugates that are based on thiol-functional- vinyl- and allylbenzoic systems, are predestined for being
ized AP-3 derivatives. These are planned to be obtained by a utilized in cross metathesis reactions.
combined muta- and semisynthetic strategy using different
aminobenzoic acids as mutasynthons.
Therefore a series of aminobenzoic acids 9–20 (Figure 1) were
prepared (see Supporting Information File 1). We expected
them to serve our purposes, when fully processed after being
fed to the mutant strain of A. pretiosum blocked in the
Results and Discussion
Mutasynthetic experiments
As potential groups for introducing linker systems bound to the biosynthesis of the PKS starter unit AHBA (1). We found that
aromatic moiety that contain terminal thiol groups (B and D, several of these aminobenzoic acids, namely 10, 14 and
Scheme 3), we envisaged several activated functional groups 16–20 were either not loaded onto the PKS or not processed
(Hal, OH, NH2). Activation was planned to be achieved by by the polyketide synthase in A. pretiosum and thus no forma-
benzylic positioning as well as by choosing aryl bromides A tion of new ansamitocin derivatives was encountered in these
(Scheme 3) that can be modified to the corresponding aryl cases.
537

Beilstein J. Org. Chem. 2014, 10, 535–543.
Scheme 3: Strategies for introducing linker-based thiol groups to the aromatic moiety of ansamitocin P-3 for accessing tumor targeting conjugates
(CuAAC; Cu-mediated azide–alkyne cycloaddition).
Figure 1: m-Aminobenzoic acid derivatives 9–20 tested as mutasynthons.
In contrast, benzoic acid 11 provided Br-F-ansamitocin deriva- To evaluate the sterical and electronical properties of mutasyn-
tives 21a–d after being fed to a growing culture of the mutant thon 11 with respect to bioprocessing, 3-amino-5-bromoben-
strain as judged by HRMS (Scheme 4). The retention times in zoic acid (9) was next employed in feeding experiments. It was
LC and MS experiments clearly showed common patterns for well processed to a series of new ansamitocin derivatives 22a–f
ansamitocins. Additionally, the isotopic pattern provided evi- (Scheme 5).
dence for the incorporation of the bromo functionality (see
Supporting Information File 1). However, yields for each of the Again, all derivatives showed common fragmentation patterns
four ansamitocins were too small for practical scale-up.
and isotopic composition in the MS. Bromo-ansamitocin deriva-
538

Beilstein J. Org. Chem. 2014, 10, 535–543.
Scheme 4: Mutasynthetic transformation of aminobenzoic acid 11 with AHBA(−)-mutant of A. pretiosum; putative structures of ansamitocin-deriva-
tives 21a–d as judged from HRMS analysis.
Scheme 5: Mutasynthetic transformation of aminobenzoic acid 9 with AHBA(−)-mutant of A. pretiosum to bromo-ansamitocins 22a–f and detoxifica-
tion products 22g and 22h.
tives 22a, 22c and 22e were isolated after scale up to a 1000 mL Likewise, also feeding of propargyl-substituted aminobenzoic
culture broth. Structures, expected from HRMS analysis, were acid 13 furnished new propargyl-modified ansamitocins 24a–g.
corroborated by NMR spectroscopy. Yields were determined to Besides the expected AP-3 derivative 24a, also the formation of
be in the low mg/L range or 0.01% yield based on compound 9 the corresponding N-desmethyl ansamitocins 24b and 24c and/
(see Supporting Information File 1 for further details). In addi- or desepoxy-derivatives 24b–d were encountered. Finally, also
tion to the formation of the bromo-ansamitocin derivatives, also the propargyl-modified proansamitocin 24e and two truncated
the detoxification products 22g and 22h were isolated in 0.1% derivatives 24f and 24g were obtained. We had observed this
yield. The bromo-ansamitocins can be regarded to be an ideal unprecendeted defunctionalization at C7 in proansamitocin and
starting material for transition metal-catalyzed coupling reac- the possible biotransformation before for the mutasynthon
tions. Due to the low product yield obtained with mutasynthon 5-chloro-3-aminobenzoic acid [30]. Again, reduction of the keto
9, for which the steric demand of the bromo substituent can group at C9 occurred; but the relative configuration of C9 also
likely be made responsible, we switched the strategy towards remains unknown in the present example. All derivatives, which
the direct introduction of an alkyne moiety by a mutasynthetic could be isolated in preparative scale (24b–d, 24f and 24g)
approach. Feeding of alkynyl(amino)benzoic acid 12 to cultures were formed in 0.1% scale.
of the AHBA(−)-mutant of A. pretiosum yielded six new
alkynyl-modified ansamitocin derivatives 23a–f in yields in the In essence, the loading module of the ansamitocin PKS shows
lower mg/L range (Scheme 6). Among them, the major com- excellent flexibility towards aminobenzoic acids that bear the
pound 23f was obtained in 0.34% yield based on compound 12, slim alkynyl or propargyl substituent at C5. After upscaling of
furnishing enough mutasynthetic material (15 mg) for further the fermentation alkynyl-ansamitocins 23c, 23f, 24b–d and 24f
chemical transformations (see below).
and 24g were isolated and the structures were corroborated by
539

Beilstein J. Org. Chem. 2014, 10, 535–543.
Scheme 6: Mutasynthetic transformation of aminobenzoic acids 12 and 13 with AHBA(−)-mutant of A. pretiosum.
analysis of NMR and MS spectra. Derivatives 23a, 23b, 23d, 26a and 26b were prepared (see Supporting Information File 1)
23e and 24a, 24e were detected only by HRMS–MS analysis of and coupled with ansamitocin derivative 23f to yield the disul-
the crude extract.
fide dimer 27a and the thioacetate 27c (Scheme 8). The former
could be transformed into the thiol monomer 27b by dithio-
Mutasynthetic experiments with vinyl(amino)benzoic acid 15 threitol reduction.
supposedly provided ansamitocin derivative 25 as judged by
HRMS–MS analysis of the crude extract (Scheme 7). However, Biological testing
yields were too low for practical upscaling.
New ansamitocin derivatives 22a, 22c, 22e, 23c, 23f, 24b–d,
24f, 24g, 27b and 27c were subjected to in vitro biological
As mutasynthetic experiments with alkynyl- and propargyl- testing with different human cell lines and one murine cell line
(amino)benzoic acids gave best results, we decided to test the derived from tumors or from connective tissue, respectively.
Huisgen-type cycloaddition with alkynyl derivative 23f bearing The results are listed in Table 1 and are provided as values for
the biologically relevant ester side chain, the oxirane moiety the half-maximal inhibitory concentration of the respective
and the carbamoyl group for introducing a thiol moiety that ansamitocin derivatives. The most active derivatives were 23f
would be suited for further conjugation. Two linker elements and 27c. Expectedly, 24b and 24f are inactive, as they lack the
540

Beilstein J. Org. Chem. 2014, 10, 535–543.
Scheme 7: Mutasynthetic transformation of vinyl(amino)benzoic acid 15 with AHBA(−)-mutant of A. pretiosum.
Scheme 8: Preparation of thiofunctionalized ansamitocin derivatives 27 by Huisgen-type copper-mediated cycloaddition.
Table 1: Antiproliferative activity IC50 [ng/mL] of 22a, 22c, 22e, 23c, 23f, 24b–d, 24f, 24g, 27b and 27c in comparison to AP-3 (4). Values shown are
means of two determinations in parallel; human cell lines: KB-3-1 (cervix carcinoma), A-431 (epidermoid carcinoma), SK-OV-3 (ovary adenocarci-
noma), PC-3 (prostate adenocarcinoma), L-929 (connective tissue of a mouse).
cell line
KB-3-1
A-431
SK-OV-3
PC-3
L-929
compound
AP-3 4
22a
22c
22e
23c
23f
0.11
2.1
0.050
0.31
0.030
0.34
3.1
0.035
0.26
4.0
0.1
4.1
0.28
7.5
0.65
78
5.1
1.8
0.80
0.28
0.14
>10000
18
8.7
2.2
0.77
0.46
0.099
>10000
9.0
1.9
0.38
>10000
7.8
0.074
>10000
5.5
0.81
>10000
220
24b
24c
24d
24f
6.0
70
8.0
95
580
1800
50
2200
0.85
2500
70
3500
3.8
>10000
100
24g
27b
27c
2200
0.09
>10000
0.25
2800
0.09
>10000
0.28
>10000
0.33
ester side chain at C3, the key pharmacophore of the maytansi- group, and or the oxirane groups or not. This is in line with the
noids. All other derivatives show strong to moderate antiprolif- view obtained from structure–activity relationship studies that
erative activity, irrespective whether the lack of the N-methyl these tailoring modifications only modulate the biological
541

Beilstein J. Org. Chem. 2014, 10, 535–543.
4. Kirschning, A.; Hahn, F. Angew. Chem. 2012, 124, 4086–4096.
doi:10.1002/ange.201107386
activity of the maytansinoids. In the present study, the human
cancer cell lines were more sensitive than the mouse fibroblasts.
Some derivatives also seem to exert a specificity towards
certain cell lines, e.g. A-431 showed particular sensitivity for
24g.
Angew. Chem., Int. Ed. 2012, 51, 4012–4022.
doi:10.1002/anie.201107386
5. Weist, S.; Süssmuth, R. D. Appl. Microbiol. Biotechnol. 2005, 68,
141–150. doi:10.1007/s00253-005-1891-8
6. Kirschning, A.; Taft, F.; Knobloch, T. Org. Biomol. Chem. 2007, 5,
3245–3259. doi:10.1039/b709549j
The high antiproliferative activity of the “click” product 27c is
remarkable as it demonstrates that substantial structural changes
at C20 of the ansamitocins are tolerated and thus alkyne
substituents open up diverse opportunities to diversify that pos-
ition as reported including the introduction of tumor targeting
ligands. Noteworthy 27b is inactive which we ascribe to the fact
that it may have dimerized back to 27a.
7. Kennedy, J. Nat. Prod. Rep. 2008, 25, 25–34. doi:10.1039/B707678A
8. Cassady, J. M.; Chan, K. K.; Floss, H. G.; Leistner, E.
Chem. Pharm. Bull. 2004, 52, 1–26. doi:10.1248/cpb.52.1
9. Kirschning, A.; Harmrolfs, K.; Knobloch, T. C. R. Chim. 2008, 11,
1523–1543. doi:10.1016/j.crci.2008.02.006
10.Floss, H. G.; Yu, T.-W.; Arakawa, K. J. Antibiot. 2011, 64, 35–44.
doi:10.1038/ja.2010.139
11.Funayama, S.; Cordell, G. A. In Studies in Natural Product Chemistry;
Rahman, A.-U., Ed.; Elsevier Science B. V.: Amsterdam, 2000; Vol. 23,
pp 51–106. doi:10.1016/S1572-5995(00)80127-1
Conclusion
The combination of mutasynthesis and semisynthesis has great
potential for the production of more selective ansamitocin
derivatives. In this work we pursued different options to prepare
ansamitocin derivatives by mutasynthesis that are set for intro-
ducing linker systems to the aromatic moiety suited to create
conjugates composed of ansamitocin and tumor-specific
ligands.
12.Hamel, E. Pharmacol. Ther. 1992, 55, 31–51.
doi:10.1016/0163-7258(92)90028-X
13.Taft, F.; Brünjes, M.; Floss, H. G.; Czempinski, N.; Grond, S.;
Sasse, F.; Kirschning, A. ChemBioChem 2008, 9, 1057–1060.
doi:10.1002/cbic.200700742
14.Kubota, T.; Brünjes, M.; Frenzel, T.; Xu, J.; Kirschning, A.; Floss, H. G.
ChemBioChem 2006, 7, 1221–1225. doi:10.1002/cbic.200500506
15.Taft, F.; Brünjes, M.; Knobloch, T.; Floss, H. G.; Kirschning, A.
J. Am. Chem. Soc. 2009, 131, 3812–3813. doi:10.1021/ja8088923
16.Knobloch, T.; Harmrolfs, K.; Taft, F.; Thomaszewski, B.; Sasse, F.;
Kirschning, A. ChemBioChem 2011, 12, 540–547.
Supporting Information
doi:10.1002/cbic.201000608
The supporting information provide the synthesis of
building blocks, mutasynthetic experiments as well as
purification protocols after fermentations, a short
description of the cell proliferation assay, analytical
descriptions of new metabolites and copies of 1H and
13C NMR spectra.
17.Eichner, S.; Knobloch, T.; Floss, H. G.; Fohrer, J.; Harmrolfs, K.;
Hermane, J.; Schulz, A.; Sasse, F.; Spiteller, P.; Taft, F.; Kirschning, A.
Angew. Chem. 2012, 124, 776–781. doi:10.1002/ange.201106249
Angew. Chem., Int. Ed. 2012, 51, 752–757.
doi:10.1002/anie.201106249
18.Meyer, A.; Brünjes, M.; Taft, F.; Frenzel, T.; Sasse, F.; Kirschning, A.
Org. Lett. 2007, 9, 1489–1492. doi:10.1021/ol0702270
19.Frenzel, T.; Brünjes, M.; Quitschalle, M.; Kirschning, A. Org. Lett. 2006,
8, 135–138. doi:10.1021/ol052588q
Supporting Information File 1
20.Harmrolfs, K.; Brünjes, M.; Dräger, G.; Floss, H. G.; Sasse, F.; Taft, F.;
Kirschning, A. ChemBioChem 2010, 11, 2517–2520.
doi:10.1002/cbic.201000422
Additional experimental details and NMR spectra.
[http://www.beilstein-journals.org/bjoc/content/
supplementary/1860-5397-10-49-S1.pdf]
21.Knobloch, T.; Dräger, G.; Collisi, W.; Sasse, F.; Kirschning, A.
Beilstein J. Org. Chem. 2012, 8, 861–869. doi:10.3762/bjoc.8.96
22.Walsh, C. T.; Haynes, S. W.; Ames, B. D. Nat. Prod. Rep. 2012, 29,
37–59. doi:10.1039/c1np00072a
Acknowledgements
This work was supported by the Deutsche Forschungsgemein-
schaft (grant Ki-397, 13-1) and the Fonds der Chemischen
Industrie.
23.Spiteller, P.; Bai, L.; Shang, G.; Carroll, B. J.; Yu, T.-W.; Floss, H. G.
J. Am. Chem. Soc. 2003, 125, 14236–14237. doi:10.1021/ja038166y
24.Yu, T.-W.; Bai, L.; Clade, D.; Hoffmann, D.; Toelzer, S.; Trinh, K. Q.;
Xu, J.; Moss, S. J.; Leistner, E.; Floss, H. G.
Proc. Natl. Acad. Sci. U. S. A. 2002, 99, 7968–7973.
doi:10.1073/pnas.092697199
References
25.Arakawa, K.; Müller, R.; Mahmud, T.; Yu, T.-W.; Floss, H. G.
J. Am. Chem. Soc. 2002, 124, 10644–10645. doi:10.1021/ja0206339
26.Yu, T.-W.; Müller, R.; Müller, M.; Zhang, X.; Dräger, G.; Kim, C.-G.;
Leistner, E.; Floss, H. G. J. Biol. Chem. 2001, 276, 12546–12555.
doi:10.1074/jbc.M009667200
1. Cragg, G. M.; Grothaus, P. G.; Newman, D. J. Chem. Rev. 2009, 109,
3012–3043. doi:10.1021/cr900019j
2. Newman, D. J.; Cragg, G. M. J. Nat. Prod. 2007, 70, 461–477.
doi:10.1021/np068054v
3. von Nussbaum, F.; Brands, M.; Hinzen, B.; Weigand, S.; Häbich, D.
Angew. Chem. 2006, 118, 5194–5254. doi:10.1002/ange.200600350
Angew. Chem., Int. Ed. 2006, 45, 5072–5129.
27.Moss, S. J.; Bai, L.; Toelzer, S.; Caroll, B. J.; Mahmud, T.; Yu, T.-W.;
Floss, H. G. J. Am. Chem. Soc. 2002, 124, 6544–6545.
doi:10.1021/ja020214b
doi:10.1002/anie.200600350
542

Beilstein J. Org. Chem. 2014, 10, 535–543.
28.Taft, F.; Harmrolfs, K.; Nickeleit, I.; Heutling, A.; Kiene, M.; Malek, N.;
Sasse, F.; Kirschning, A. Chem.–Eur. J. 2012, 18, 880–886.
doi:10.1002/chem.201101640
29.D'Alincourt Salazar, M.; Ratnam, M. Cancer Metastasis Rev. 2007, 26,
141–152. doi:10.1007/s10555-007-9048-0
30.Low, P. S.; Antony, A. C. Adv. Drug Delivery Rev. 2004, 56,
1055–1058. doi:10.1016/j.addr.2004.02.003
31.Lee, R. J.; Wang, S.; Low, P. S. Biochim. Biophys. Acta 1996, 1312,
237–242. doi:10.1016/0167-4889(96)00041-9
32.Reddy, J. A.; Westrick, E.; Vlahov, I.; Howard, S. J.;
Santhapuram, H. K.; Leamon, C. P. Cancer Chemother. Pharmacol.
2006, 58, 229–236. doi:10.1007/s00280-005-0151-z
License and Terms
This is an Open Access article under the terms of the
Creative Commons Attribution License
(http://creativecommons.org/licenses/by/2.0), which
permits unrestricted use, distribution, and reproduction in
any medium, provided the original work is properly cited.
The license is subject to the Beilstein Journal of Organic
Chemistry terms and conditions:
(http://www.beilstein-journals.org/bjoc)
The definitive version of this article is the electronic one
which can be found at:
doi:10.3762/bjoc.10.49
543