Page 1 of 6
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
(
89@E=:=75E=A@ A: E<9 %AC?H7=@ ꢀ !=ADH@E<9E=7 &9@9 ">FDE9C
:
CA? S-+#*-)'/!#, &ꢀ(%$ꢀ+ꢀ#(,%, (>>FDEC5E9D E<9 (@E9CB>5H 69-
EG99@ !=A>A;=75> .HC5IA>ABHC=?=8=@9 %AC?5E=A@ 5@8 "# ().)
FC=@9 !=ADH@E<9D=D
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
.
†
,à
†,à
±
¶
†
¶,§
±
Shao-An Wang, Yeonjin Ko, Jia Zeng, Yujie Geng, Daan Ren, Yasushi Ogasawara, Seema Irani, Yan
Zhang, Hung-wen Liu*
±
,¶,†
†
±
¶
Department of Chemistry, Department of Molecular Biosciences, and Division of Chemical Biology and Medicinal Chemistry, College of Phar-
macy, The University of Texas at Austin, Austin, TX, 78712
Supporting Information Placeholder
of pur-like genes (i.e., the for-genes) (see Table S4) are likely in-
volved in the biosynthesis of formycin A, since the pyrazolopyrimi-
dine moiety of formycin A may be formed in a similar manner as the
imidazolopyrimidine group in adenosine. This hypothesis was sup-
ported by the in vitro demonstration that the purA-like (forA) and
purB-like (forB) gene products can catalyze the conversion of
formycin B 5'-phosphate to formycin A 5’-phosphate (7bf8bf9b,
ABSTRACT: Formycin A is a potent purine nucleoside antibiotic
with a C-glycosidic linkage between the ribosyl moiety and the pyra-
zolopyrimidine base. Herein, a cosmid is identified from the Strepto-
myces kaniharaensis genome library that contains the for gene cluster
responsible for the biosynthesis of formycin. Subsequent gene dele-
tion experiments and in vitro characterization of the forBCH gene
products established their catalytic functions in formycin biosynthe-
sis. Results also demonstrated that PurH from de novo purine bio-
synthesis plays a key role in pyrazolopyrimidine formation during bi-
osynthesis of formycin A. The participation of PurH in both path-
ways represents a good example of how primary and secondary me-
tabolism are interlinked.
1
2
Figure 1). In the present work, the full for gene cluster for formycin
A biosynthesis is identified from S. kaniharaensis. Moreover, investi-
gation of the encoded enzymes indicates that the cluster alone is in-
sufficient for the biosynthesis of formycin A and requires involve-
ment of PurH from de novo purine biosynthesis.
5
52
25
42
5
8
9
42
8
9
8 42
9
2
(/=?.
B?.
2B?-
(/=?-
5
5
2
5
42
5
42%
Formycin A (1), also known as 8-aza-9-deazaadenosine, is a purine
6=5
6
=5
6=5
5
5
42%
5
1
2
42%
42%
nucleoside antibiotic isolated from Nocardia interforma, Streptomy-
25
52
25
52
25
52
3
"-CA -/6
<E?:B:D;
8
3
4
ces kaniharaensis SF-557 and Streptomyces sp. MA406-A-1. It is an
isomer of adenosine (2); however, it is a C-nucleoside with a pyra-
zolopyrimidine instead of an imidazolopyrimidine nucleobase mak-
ing it distinctly different from typical nucleosides. Formycin A is a
potent inhibitor of adenosine-utilizing enzymes such as bacterial pu-
rine nucleoside phosphorylase and the adenosine kinase involved in
the purine salvage pathway in Mycobacterium tuberculosis. It is cyto-
toxic to Leishmania species due to the incorporation of the phos-
phorylated form of formycin A into RNA. Formycin A also exhibits
8ꢀ,ꢀ.!ꢀ9ꢀ,ꢀ4!ꢀ.,13ꢀ(%4 -76
ꢀ,ꢀ4!ꢀ9ꢀ,ꢀ.ꢀ(%5
6=
8ꢀ,ꢀ.!ꢀ9ꢀ,ꢀ4ꢀ(&4
8ꢀ,ꢀ4!ꢀ9ꢀ,ꢀ.ꢀ(&5
8ꢀ,ꢀ.!ꢀ9ꢀ,ꢀ4!ꢀ,1.,3ꢀ('4
8ꢀ,ꢀ4!ꢀ9ꢀ,ꢀ.ꢀ('5
8
42
8
42
2B?0
(/=?0 6=5
5
5
2B?,
9
2
B?0
5
9
5
(/=?,
5
6=5
4
42
2
4
42%
52 .25
25
52
4$#"<@B?F>" 720
720
25
%
2 5
8ꢀ,ꢀ.!ꢀ9ꢀ,ꢀ4ꢀ
1<=@:<8"'D"2ꢀ()4
ꢀ,ꢀ4!ꢀ9ꢀ,ꢀ.ꢀ
/=?;C6:<ꢁ-"'D"2ꢁ()5
3"-CA 1/6
1
76
6=
6
8ꢀ,ꢀ.!ꢀ9ꢀ,ꢀ4!ꢀ/,1.,3ꢀ(ꢀ4
8ꢀ,ꢀ4!ꢀ9ꢀ,ꢀ.ꢀ(ꢀ5
8
7
5
52
25
5
*
8
)
42
8
9
42
8
8
9
42
5
5
+
2
B?-
42% 29=@>94"
A4@8 25
9
&
ꢁ
42%
5
42
'
(
/=?-
6=5
9
6=5
4
4
4 %
4
antiviral activity against influenza virus A1 and human immunode-
ficiency virus type 1. Although the biological functions of formycin
A have been well documented, little is known about how it is assem-
4
4
25
52
25
52
2
5
52
1
0
<
E?:B:D;
8ꢀ,ꢀ.!ꢀ9ꢀ,ꢀ4ꢀ
78<=@:<8"'D"2ꢀ(+4
8
ꢀ,ꢀ.!ꢀ9ꢀ,ꢀ4!ꢀ,78<=@:<8ꢁ($
8ꢀ,ꢀ.!ꢀ9ꢀ,ꢀ4ꢀ(*4
8ꢀ,ꢀ4!ꢀ9ꢀ,ꢀ.ꢀ(*5
,
8ꢀ,ꢀ4!ꢀ9ꢀ,ꢀ.!ꢀ/=?;C6:<ꢁ,ꢁ(#
8ꢀ,ꢀ4!ꢀ9ꢀ,ꢀ.ꢀ
4
,11
/=?;C6:<ꢁ,"'D"2ꢁ(+5
bled in nature.
Figure 1. Later steps of the biosynthetic pathway for adenosine (2)
and the proposed pathway for formycin A (1) formation.
To investigate the biosynthesis of formycin A, the genome of S.
1
2
kaniharaensis was sequenced. Preliminary analysis revealed the
presence of two sets of purA, purB, purC, and purH genes. The purA,
purB, purC and purH gene products are enzymes responsible for the
conversion of carboxyaminoimidazole ribonucleotide (CAIR, 3a)
to adenosine 5'-phosphate (9a, see Figure 1) during the biosynthesis
To identify the formycin A gene cluster, a cosmid library from the
genome of S. kaniharaensis was constructed and those cosmids har-
boring the pur-like genes were screened by PCR amplification with
primers based on the sequences of the second set of purA-like, purC-
like, and purH-like genes (i.e., forA, forC and forH). This effort led to
the identification of a single cosmid, designated K24C, harboring the
13
of purine nucleosides. The discovery of two sets of pur homologous
genes in the genome of S. kaniharaensis suggested that the second set
ACS Paragon Plus Environment