Polyketide natural products are a structurally diverse
class of molecules isolated from bacteria, fungi, and plants
and represent a rich source of medicinally important
compounds.6 The complex structuresassociatedwithpoly-
ketides arise from the utilization of simple building blocks,
such as malonyl-CoA, and can undergo varying degrees of
β-keto reduction during each catalytic cycle on the PKS.7
As a result, enzyme-directed engineering approaches have
been aimed at diversifying building block and starter unit
selection as a way to enhance the structural variation of
polyketides and to create analogues containing nonnative
functional groups.8 Recently, enzyme engineering efforts
with bacterial acyltransferase (AT) domains9 and acyl-
CoA synthetases10 have been successful at assimilating
nonnatural moieties into polyketides using synthetic malo-
nic acid building blocks. While these synthetic biology
strategies are promising for enhancing the structural di-
versity of polyketides, an alternative method for introdu-
cing structural variation can involve the use of nonnatural
cofactors. Thus, given the reported promiscuous nature of
standalone methyltransferases, we wanted to investigate
whether MT domains found within PKSs could behave in
the same manner and facilitate the incorporation of dif-
ferent alkyl groups at the R-positions of polyketides.
To examine the substrate specificity of a PKS MT
domain, we turned our attention to the fungal highly-
reducing PKS (HR-PKS) CazF, which is involved in the
biosynthesis of the chaetoviridin and chaetomugilin
azaphilones.11 In vitro reconstitution experiments with re-
combinant CazF (284 kDa) expressed from Saccharomyces
cerevisiae BJ5464-NpgA12 verified that the MT domain
functions after the first chain-extending condensation step,
and the remaining domains are all catalytically active to
generate a small polyketide product.11 CazF therefore repre-
sented a good model system for exploring whether a MT
domain can be used to introduce chemical diversity into
polyketides. We selected the two unnatural SAM analogues
propargylic Se-adenosyl-L-methionine (ProSeAM, 2)2d,h and
keto-SAM (3)2d,13 for this study because both analogues can
provide a reactive handle that can be further modified with
mild, orthogonal chemistry. The selenium containing 2 was
selected over the sulfur-based propargylic SAM derivative
because it is more stable at physiological pH with a half-life of
approximately one hour.2d
Scheme 1. CazF-Mediated Transalkylation of 4 Using Its Natural
Cofactor 1 or Unnatural Analogues 2 or 3 as the Alkyl Donor
We first assayed the kinetics of the CazF MT domain
toward the unnatural cofactors. Recombinant CazF was
incubated with acetoacetyl-SꢀN-acetyl cysteamine
(acetoacetyl-SNAC, 4) in the presence of 1, 2, or 3
(Scheme 1). The production of alkylated acetoacetyl-SNAC
products were analyzed by liquid chromatography and mass
spectroscopy (LC-MS); and kinetic constants were calculated
by fitting initial velocity data at various concentrations of 1or
2 to MichaelisꢀMenten parameters using nonlinear least-
squares curve fitting. In the assays containing 3, no product
was detected suggesting that the MT domain would not
accept the ketone derivative. The alkylated 5 and 6 were
formed in the presence of 1 and 2, respectively. The steady-
ꢀ
ꢀ
(3) (a) Dalhoff, C.; Lukinavicius, G.; Klimasauskas, S.; Weinhold, E.
ꢀ
ꢀ
Nat. Chem. Biol. 2006, 2, 31. (b) Dalhoff, C.; Lukinavicius, G.; Kli-
masauskkas, S.; Weinhold, E. Nat. Protoc. 2006, 1, 1879.
(4) (a) Stecher, H.; Tengg, M.; Ueberbacher, B. J.; Remler, P.;
Schwab, H.; Griengl, H.; Gruber-Khadjawi, M. Angew. Chem., Int.
Ed. 2009, 48, 9546. (b) Zhang, C.; Weller, R. L.; Thorson, J. S.; Rajski,
S. R. J. Am. Chem. Soc. 2006, 128, 2760.
(5) Ansari, M. Z.; Sharma, J.; Gokhale, R. S.; Mohanty, D. BMC
Bioinf. 2008, 9, 454.
(6) O’Hagan, D. The Polyketide Metabolites; Ellis Howard: Chiche-
ster, 1991.
(7) (a) Fischbach, M. A.; Walsh, C. T. Chem. Rev. 2006, 106, 3468. (b)
Hertweck, C. Angew. Chem., Int. Ed. 2009, 48, 4688.
(8) (a) Pickens, L. B.; Tang, Y.; Chooi, Y. H. Annu. Rev. Chem.
Biomol. 2011, 2, 211. (b) Winter, J. M.; Tang, Y. Curr. Opin. Biotech.
2012, 5, 736. (c) Wilson, M. C.; Moore, B. S. Nat. Prod. Rep. 2012, 29, 72.
(d) Kennedy, J. Nat. Prod. Rep. 2008, 25, 25.
(9) (a) Sundermann, U.; Bravo-Rodriguez, K.; Klopries, S.; Kushnir,
S.; Gomez, H.; Sanchez-Garcia, E.; Schulz, F. ACS Chem. Biol. 2013, 8,
443. (b) Hughes, A. J.; Detelich, J. F.; Keatinge-Clay, A. T. Med. Chem.
Commun. 2012, 3, 956. (c) Koryakina, I.; McArthur, J. B.; Draelos,
M. M.; Williams, G. J. Org. Biomol. Chem. 2013, 11, 4449.
(10) Koryakina, I.; McArthur, J.; Randall, S.; Draelos, M. M.;
Musiol, E. M.; Muddiman, D. C.; Weber, T.; Williams, G. J. ACS
Chem. Biol. 2013, 8, 200.
state kinetic parameters were kcat = 0.025 minꢀ1 and KM
=
15.5 μM for1and kcat = 0.036 minꢀ1 and KM = 43.6 μMfor
2. Although the turnover rates were slow for both 1and 2, the
comparable kinetic parameters indicate that the CazF MT
displays similar preference toward 1 and 2. The slow turn-
over rate may be attributed to the use of a small molecule
SNAC-bound substrate instead of an acyl-carrier protein
(ACP)-tethered molecule.
(11) Winter, J. M.; Sato, M.; Sugimoto, S.; Chiou, G.; Garg, N. K.;
Tang, Y.; Watanabe, K. J. Am. Chem. Soc. 2012, 134, 17900.
(12) Lee, K. K.; Da Silva, N. A.; Kealey, J. T. Anal. Biochem. 2009,
394, 75.
(13) Lee, B. W. K.; Sun, H. G.; Zang, T.; Kim, B. J.; Alfaro, J. F.;
Zhou, Z. S. J. Am. Chem. Soc. 2010, 132, 3642.
B
Org. Lett., Vol. XX, No. XX, XXXX