Microbial Synthesis of the Energetic Material Precursor 1,2,4-Butanetriol

Article abstract of DOI:10.1021/ja036391+

The lack of a route to precursor 1,2,4-butanetriol that is amenable to large-scale synthesis has impeded substitution of 1,2,4-butanetriol trinitrate for nitroglycerin. To identify an alternative to the current commercial synthesis of racemic D,L-1,2,4-butanetriol involving NaBH₄ reduction of esterified D,L-malic acid, microbial syntheses of D- and L-1,2,4-butanetriol have been established. These microbial syntheses rely on the creation of biosynthetic pathways that do not exist in nature. Oxidation of D-xylose by Pseudomonas fragi provides D-xylonic acid in 70% yield. Escherichia coli DH5α/pWN6.186A then catalyzes the conversion of D-xylonic acid into D-1,2,4-butanetriol in 25% yield. P. fragi is also used to oxidize L-arabinose to a mixture of L-arabino-1,4-lactone and L-arabinonic acid in 54% overall yield. After hydrolysis of the lactone, L-arabinonic acid is converted to L-1,2,4-butanetriol in 35% yield using E. coli BL21(DE3)/pWN6.222A. As a catalytic route to 1,2,4-butanetriol, microbial synthesis avoids the high H₂ pressures and elevated temperatures required by catalytic hydrogenation of malic acid. Copyright

Full text of DOI:10.1021/ja036391+

Published on Web 10/07/2003
Microbial Synthesis of the Energetic Material Precursor 1,2,4-Butanetriol
Wei Niu, Mapitso N. Molefe, and J. W. Frost*
Department of Chemistry, Michigan State UniVersity, East Lansing, Michigan 48824
Received May 28, 2003; E-mail: frostjw@cem.msu.edu
Scheme 1 a
Nitration of racemic D,L-1,2,4-butanetriol 1 affords the energetic
material D,L-1,2,4-butanetriol trinitrate 2, which is less shock
sensitive, more thermally stable, and less volatile than nitroglycerin.1
However, the limited availability of 1,2,4-butanetriol has impeded
substitution of 1,2,4-butanetriol trinitrate for nitroglycerin. Catalytic
hydrogenation of D,L-malic acid (Scheme 1) constitutes one
synthetic route to D,L-1,2,4-butanetriol. This account establishes
microbial synthesis as an alternative catalytic strategy. Enzymes
from three different microbes are recruited to create biosynthetic
pathways (Scheme 2) by which D-1,2,4-butanetriol 1a and L-1,2,4-
butanetriol 1b are derived from D-xylose 4a and L-arabinose 4b,
respectively.
a
Reaction conditions: 5000 psi H2, Ru on C, 135 °C, H2O.
Scheme 2 a
2
Commercial synthesis of D,L-1,2,4-butanetriol employs NaBH
4
3
reduction of esterified D,L-malic acid 3. For every ton of 1,2,4-
butanetriol synthesized, multiple tons of byproduct borates are
generated. D,L-Malic acid can also be hydrogenated over various
a
catalysts (Cu-Cr, Cu-Al, Ru-Re) at 2900-5000 psi of H
2
and
Enzymes (microbial source): (a) D-xylose dehydrogenase (P. fragi);
4
(a′) L-arabinose dehydrogenase (P. fragi); (b) D-xylonate dehydratase (E.
coli); (b′) L-arabinonate dehydratase (P. fragi); (c) benzoylformate decar-
boxylase (P. putida); (d) dehydrogenase (E. coli).
60-160 °C reaction temperatures. Yields of 1,2,4-butanetriol range
from 60% to 80%. A variety of byproducts are also formed during
4
high-pressure hydrogenation. These byproducts are not generated
3
when esterified malic acid is reduced using NaBH
4
.
D,L-Malic acid
(
77 g/L) in 70% yield. L-Arabinose was discovered to be similarly
oxidized in 54% overall yield to a mixture of L-arabino-1,4-lactone
40 g/L) and L-arabinonic acid 5b (15 g/L). The lactone was
subsequently hydrolyzed to L-arabinonic acid. Escherichia coli
constructs were then employed for the conversion of D-xylonic 5a
acid and L-arabinonic acid 5b into the enantiomers of 1,2,4-
butanetriol.
is synthesized from the n-butane component of liquefiable petroleum
gas via intermediacy of maleic anhydride.5
To examine product and byproduct yields in detail, Ru on C
was selected as the catalyst for reduction of malic acid due to its
commercial availability and use in lactic acid hydrogenations.6
(
2
Reduction of D,L-malic acid was optimized relative to H pressure,
temperature, concentration, and the ratio of catalyst to substrate.
Hydrogenation at 5000 psi and 135 °C of a 1 M aqueous solution
of malic acid using 1.3 mol % relative to substrate of 5 wt % Ru
on C afforded 1,2,4-butanetriol in 74% yield. Byproducts (Scheme
Previously undocumented catabolism of D-xylonic acid by E.
10
coli K-12 was discovered to coincide with D-xylonate dehydratase
expression. Transport of D-xylonic acid and its conversion into
-deoxy-D-glycero-pentulosonic acid 6a (Scheme 2) thus employed
enzymes native to E. coli. L-Arabinonic acid catabolism and
L-arabinonate dehydratase9
3
1) accounted for a total of 25% of the starting malic acid and
complicated purification of D,L-1,2,4-butanetriol using distillation.
Melting points, a key consideration in energetic material
formulations, differ for a single enantiomer relative to a racemic
mixture. Microbial syntheses were therefore needed for both 1,2,4-
b,11
activity needed for generation of
3
-deoxy-L-glycero-pentulosonic acid 6b (Scheme 2) were absent
in E. coli. This necessitated the isolation and heterologous expres-
sion of the encoding genes from a P. fragi genomic DNA library.
Three cosmids enabled E. coli K-12 to catabolize L-arabinonic acid.
From a 5.0 kb region shared between these cosmids, aadh-encoded
L-arabinonate dehydratase and an aatp-encoded L-arabinonate
transport protein were identified.
Attention then turned to identification of a 2-ketoacid decar-
boxylase capable of catalyzing the conversions of pentulosonic acid
6a to D-3,4-dihydroxybutanal 7a and pentulosonic acid 6b to L-3,4-
dihydroxybutanal 7b (Scheme 2). Only benzoylformate decarboxy-
lase expressed by Pseudomonas putida1 catalyzed these reactions.
Other 2-ketoacid decarboxylases screened but found to lack the
requisite activity included indole 3-pyruvate decarboxylase ex-
pressed by Erwinia herbicola1 and a variety of different pyruvate
7
butanetriol enantiomers. To address this challenge, the opposing
C-4 stereogenic centers of D-xylose and L-arabinose were exploited
(Scheme 2) as the basis for synthesis of D-1,2,4-butanetriol 1a and
L-1,2,4-butanetriol 1b, respectively. Both D-xylose and L-arabinose
are abundantly available in heteroxylans derived from corn fiber
8
and sugar beet pulp. Mixing of microbe-synthesized enantiomers
a and 1b (Scheme 2) would provide the equivalent of the racemic
1
D,L-1,2,4-butanetriol currently used to manufacture D,L-1,2,4-
butanetriol trinitrate.
Microbial synthesis (Scheme 2) of 1,2,4-butanetriol enantiomers
began with the reported oxidation of D-xylose using fermentor-
controlled cultures (1 L scale) of Pseudomonas fragi ATCC4973.9
D-Xylose (100 g/L) was oxidized at 30 °C to D-xylonic acid 5a
2a
2b
12998
9
J. AM. CHEM. SOC. 2003, 125, 12998-12999
10.1021/ja036391+ CCC: $25.00 © 2003 American Chemical Society

C O MMU N I C A T I O N S
Scheme 3 a
for a 3% yield of this byproduct, while E. coli BL21(DE3)/
pWN6.222A synthesized ethylene glycol (0.087 g/L) in 2% yield.
A key feature of the microbial synthesis of 1,2,4-butanetriol is
the substitution of a straightforward enzymatic reduction of an
aldehyde for the problematic catalytic reduction of a carboxylic
2
acid. The high H pressures and elevated temperatures required for
hydrogenation of malic acid are thus avoided. Byproduct formation
resulting from cleavage of carbon-carbon bonds is also substan-
tially reduced. Further metabolic engineering is clearly required to
increase product yields and concentrations. Nonetheless, microbial
catalysis is an intriguing alternative to catalytic hydrogenation for
the large-scale synthesis of 1,2,4-butanetriol needed for replacement
of nitroglycerin with 1,2,4-butanetriol trinitrate. The significance
of such a substitution is considerable given that nitroglycerin has
been used in industrial and military energetic materials since the
a
Plasmids (size): restriction enzyme maps. Sites are abbreviated as
follows: B ) BamHI, Bg ) BglII, E ) EcoRI, H ) HindIII, S ) ScaI.
Parentheses indicate that the designated enzyme site has been eliminated.
Lightface lines indicate vector DNA; boldface lines indicate insert DNA.
14
original dynamite formulations developed by Nobel.
1
2c
decarboxylases expressed by Zymomonas mobilis, Acetobacter
Acknowledgment. Research was funded by a contract from the
Office of Naval Research.
12d
12e
pasteurianus, Zymobacter palmae, and Saccharomyces cere-
Visiae.1
2f
Supporting Information Available: Hydrogenation of malic acid;
isolation of aadh and aatp; enzyme assays; microbial oxidation of
D-xylose and L-arabinose; microbial synthesis of D- and L-1,2,4-
butanetriol; enantiomer analysis (PDF). This material is available free
of charge via the Internet at http://pubs.acs.org.
Native dehydrogenase activity in E. coli was anticipated to be
adequate for the reduction of butanal 7a to D-1,2,4-butanetriol 1a
and butanal 7b to L-1,2,4-butanetriol 1b (Scheme 2). To test for
the needed dehydrogenase activity, intact E. coli DH5R/pWN5.238A
(
Scheme 3) expressing benzoylformate decarboxylase was incubated
in medium containing racemic D,L-3-deoxy-glycero-pentulosonic
acid. Accumulation of D,L-1,2,4-butanetriol indicated that E. coli
expressed the required dehydrogenase activity under aerobic culture
conditions.
References
(
1) CPIA/M3 Solid Propellant Ingredients Manual; The Johns Hopkins
University, Chemical Propulsion Information Agency: Whiting School
of Engineering, Columbia, MD, 2000.
(2) Cramer, R. J. Personal communication. Indian Head Division, Naval
Surface Warfare Division, Indian Head, Maryland, 2002.
With the required enzyme activities identified, E. coli constructs
were assembled for the conversions of D-xylonic acid 5a and
L-arabinonic acid 5b synthesized by P. fragi from D-xylose 4a and
L-arabinose 4b (Scheme 2). D-1,2,4-Butanetriol-synthesizing E. coli
DH5R/pWN6.186A (Scheme 3) carried a P. putida mdlC plasmid
insert encoding benzoylformate decarboxylase while relying on
native D-xylonate transport along with native D-xylonate dehydratase
and dehydrogenase activities. The required heterologous expression
of only a single gene was a consequence of E. coli catabolism of
D-xylonic acid. By contrast, L-1,2,4-butanetriol-synthesizing E. coli
BL21(DE3)/pWN6.222A (Scheme 3) carried the P. putida mdlC
plasmid insert encoding benzoylformate decaboxylase, a P. fragi
aadh plasmid insert encoding L-arabinonate dehydratase, and a P.
fragi aatp insert encoding an L-arabinonate transport protein.
Alcohol dehydrogenase activity was the only native E. coli enzyme
activity recruited for L-1,2,4-butanetriol synthesis.
Fermentor-controlled cultivation (1 L) of E. coli DH5R/
pWN6.186A at ambient pressure and 33 °C resulted in the
conversion of D-xylonic acid (10 g/L) into D-1,2,4-butanetriol (1.6
g/L) in 25% yield. Similar cultivation of E. coli BL21(DE3)/
pWN6.222A led to the conversion of L-arabinonic acid (10 g/L)
into L-1,2,4-butanetriol (2.4 g/L) in 35% yield. Stereochemical
assignments for microbe-synthesized products were based on the
conversion to Mosher esters and comparison with similarly deriva-
tized D- and L-1,2,4-butanetriol obtained from commercial sources.13
E. coli DH5R/pWN6.186A synthesized ethylene glycol (0.093 g/L)
(3) (a) Monteith, M. J.; Schofield, D.; Bailey, M. PCT Int. Appl. WO 98/
0
8793, 1998. (b) Ikai, K.; Mikami, M.; Furukawa, Y.; Ho, S. PCT Int.
Appl. WO 99/44976, 1999.
(4) (a) Adkins, H.; Billica, H. R. J. Am. Chem. Soc. 1948, 70, 3121. (b)
Mueller, H.; Mesch, W.; Broellos, K. U.S. Patent 4973769, 1990. (c)
Antons, S.; Tillling, A. S.; Wolters, E. U.S. Patent 6355848, 2002.
(
5) Szmant, H. H. Organic Building Blocks of the Chemical Industry; Wiley:
New York, 1989; p 362.
(6) (a) Zhang, Z.; Jackson, J. E.; Miller, D. J. Appl. Catal. A 2001, 219, 89.
(b) Jere, F. T.; Miller, D. J.; Jackson, J. E. Org. Lett. 2003, 5, 527.
(7) Chiral synthons can also be derived from 1,2,4-butanetriol enantiomers
and their biosynthetic precursors. See: (a) Wu, G.; Wong, Y. S.; Chen,
X.; Ding, Z. J. Org. Chem. 1999, 64, 3714. (b) Wu, G.-Z.; Chen, X.;
Wong, Y.-S.; Schumacher, D. P.; Steinman, M. U.S. Patent 5,886,171,
1999.
(
8) (a) Saulnier, L.; Marot, C.; Chanliaud, E.; Thibault, J.-F. Carbohydr.
Polym. 1995, 26, 279. (b) Micard, V.; Renard, C. M. G. C.; Thibault,
J.-F. Enzyme Microb. Technol. 1996, 19, 162.
9) (a) Buchert, J.; Viikari, L.; Linko, M.; Markkanen, P. Biotechnol. Lett.
1986, 8, 541. (b) Weimberg, R. J. Biol. Chem. 1961, 236, 629.
(
(
10) Dahms, A. S.; Donald, A. Methods Enzymol. 1982, 90, 302.
11) (a) Weimberg, R. J. Biol. Chem. 1959, 234, 727. (b) Schimz, K.-L.; Kurz,
G. Biochem. Soc. Trans. 1975, 3, 1087.
(
(
12) (a) Tsou, A. Y.; Ransom, S. C.; Gerlt, J. A. Biochemistry 1990, 29, 9856.
(
b) Brandl, M. T.; Lindow, S. E. Appl. EnViron. Microbiol. 1996, 62,
4
121. (c) Conway, T.; Osman, Y. A.; Konnan, J. I.; Hoffmann, E. M.;
Ingram, L. O. J. Bacteriol. 1987, 169, 949. (d) Raj, K. C.; Ingram, L. O.;
Maupin-Furlow, J. A. Arch. Microbiol. 2001, 176, 443. (e) Raj, K. C.;
Talarico, L. A.; Ingram, L. O.; Maupin-Furlow, J. A. Appl. EnViron.
Microbiol. 2002, 68, 2869. (f) Commercially available from Sigma.
13) Dale, J. A.; Mosher, H. S. J. Am. Chem. Soc. 1973, 95, 512.
14) Lindner, V. In Kirk-Othmer Encyclopedia of Chemical Technology;
Kroschwitz, J. I., Howe-Grant, M., Eds.; Wiley: New York, 1993; Vol.
10, p 46.
(
(
JA036391+
J. AM. CHEM. SOC.
9
VOL. 125, NO. 43, 2003 12999