The putative Diels-Alderase macrophomate synthase is an efficient aldolase

Article abstract of DOI:10.1021/ja8017994

We find that the putative Diels-Alderase macrophomate synthase (MPS) catalyzes the addition of pyruvate enolate, generated by decarboxylation of oxaloacetate, to a variety of aldehydes. Alkyl, aryl, and heteroaryl aldehydes are accepted as substrates, providing γ-hydroxy-α-ketoacids in 35-95% yield with modest levels of stereochemical control. These aldol products, which are difficult to synthesize by other methods, are formed with efficiency comparable to that of macrophomate. Our results thus provide evidence that a two-step Michael-aldol pathway is a plausible alternative to the postulated [4 + 2] cycloaddition in the MPS-catalyzed addition of pyruvate enolate to 2-pyrones. They are also relevant to understanding the divergent evolution of type II pyruvate aldolases. Copyright

Full text of DOI:10.1021/ja8017994

S1
The Putative Diels-Alderase Macrophomate Synthase is an Efficient
Aldolase
Jörg M. Serafimov, Dennis Gillingham, Simon Kuster and Donald
Hilvert*
Laboratory of Organic Chemistry, ETH Zurich, CH-8093 Zurich, Switzerland
E-mail: hilvert@org.chem.ethz.ch
Supporting Information
Materials and General Methods. All reactions were conducted in oven− (135 °C) or
flame-dried glassware under an inert atmosphere of dry N2 unless otherwise stated. NMR
were recorded on Bruker AV600 (1H 600 MHz, 13C 150.9 MHz), Bruker DRX 500 (1H
500 MHz, 13C 125 MHz), Bruker DRX (1H 400 MHz, 13C 100 MHz), ARX 300 (1H 300
MHz, 13C 75 MHz), or Varian Gemini 300 (1H 300 MHz, 13C 75 MHz) NMR
1
spectrometers. All 13C-NMR spectra are H-broadband decoupled and were measured at
13
room temperature if not stated otherwise, and the C resonance of the chloroform was
used as an internal standard (13CDCl3: 77.16). Chemical shifts are reported in ppm from
tetramethylsilane with the solvent resonance resulting from incomplete deuteration as the
internal standard (CDCl3: 7.26). Data are reported as follows: chemical shift, multiplicity
(s = singlet, d = doublet, t = triplet, q = quartet, br = broad, m = multiplet or combinations
thereof), coupling constants, and integration. Multiplicity was assigned according to a
1
method described by Hoye. 1 Enantiomer ratios were determined by H NMR of
derivatized adducts (see the individual substrate entries for details) or chiral HPLC
analysis. HPLC: Daicel AD column (4.6 mm x 250 mm). Mass spectra were recorded on
a Finnigan TSQ7000 Triple-Quad mass spectrometer (ESI) or on a HiRes-ESI IonSpec
Ultima FTMS-spectrometer (Ionspec, Lake Forest, CA, USA). ESI spectra of small
organic molecules were usually measured in MeOH. Calculated masses were based on
average isotope composition or on single isotope masses for high resolution spectra.
Chemicals were purchased from Sigma, Acros, ABCR, Merck, Aldrich or Fluka and
were used without further purification, unless otherwise noted. Buffer salts were
purchased from Sigma, J.T. Baker, or Fluka and used without further purification.
Oxaloacetic acid was purchased from Fluka or Sigma and recrystallized from ethyl
acetate before use. NMR solvents and internal standards, 3-trimethylsilyl-proprionate-d4
sodium salt (TSP) and tetramethyl silane, were purchased from Armar (Doettingen,
Switzerland) and Cambridge Isotope Laboratories (MA, USA), respectively. Silica gel
chromatography was driven with compressed air and performed with silical gel 60 from
Fluka (grain size 40-63 µm) at room temperature.

S2
Solvents were dried as follows: THF was freshly distilled over Na/benzophenone
before use. Et2O was filtered through a basic alumina plug directly before use. CH2Cl2
was distilled over CaH2. DMF was stirred over KOH pellets overnight and then distilled
under reduced pressure over CaH2 at 40 °C.
Plasmids. The pETMPS plasmid encoding the gene for MPS2 was provided by Prof.
Hideaki Oikawa, Hokkaido University, Japan. A commercially available pET22b(+)
vector (Novagen) was used for construction of the pETMPS22-2 plasmid. Experimental
details for construction and introduction of the plasmid into BL21 cells have already been
described.3
Protein production. Competent E. coli cells (BL21) were transformed with
pETMPS22-2 or the corresponding mutant plasmid. The proteins were overproduced
upon induction with 2.5 mM IPTG in 250 ml cultures grown at 25 ºC in LB medium
supplemented with ampicillin (200 mg/l). Cell pellets were suspended in 7.5 ml loading
buffer (50 mM sodium phosphate 300 mM NaCl, pH 7.0), 2 mg of the protease-inhibitor
PEFA-bloc were added and the cells were lysed by incubation with 10 mg lysozyme at
4 °C followed by ultrasonication. Cleared cell lysates were incubated with 4 ml Ni2+-
NTA-agarose resin (Qiagen), which was pre-equilibrated with loading buffer containing
10 mM imidazole, pH 7.0. After 10 min incubation, this suspension was loaded onto Ni-
chromatography columns that were then washed with 30 ml each of 10, 20, and 40 mM
imidazole in loading buffer. The desired protein was eluted as a pure fraction with 12 ml
of buffer containing 200 mM imidazole. Protein purity was assessed by 20% SDS-PAGE
to be >98%, protein identity was confirmed by LC-MS and plasmid sequencing after a
plasmid miniprep of a 5 mL portion of the cell culture. Typically, a 250 ml culture
yielded between 20 and 40 mg of purified protein, depending on the mutant. A typical
yield for the wild type enzyme was ca. 50 mg of pure protein per 250 ml of cell culture.
The proteins were dialyzed into 20 mM potassium phosphate buffer, pH 7.0, at 4 ºC,
sterile filtered and then stored in the cold prior to use.
LC-MS was performed on a Spectra System HPLC connected to a diode array
detector (UV6000LP, Thermo Separation Products) and an ion-trap mass spectrometer
(LCQdeca, Finnigan) with the same columns as for analytical RP-HPLC (see below).
Two buffer systems were routinely used for reactions with macrophomate synthase.
Buffer system A was used for the kinetic analysis of decarboxylation of oxaloacetate, and
buffer system B was used for all analytical and preparative scale aldol reactions except
when stated otherwise.
MPS Reaction buffer A (50 mM PIPES, 5 mM MgCl2, pH 7.0): 16.2 g of PIPES
monosodium salt and 1.015 g of MgCl2 • 12 H2O were dissolved in 1 L UPW. The pH
was adjusted to 7.0 with 1 M NaOH. The buffer was sterile filtered and stored at 4 °C.
MPS Reaction buffer B (50 mM phosphate buffer, 5 mM MgCl2, pH 7.0): 0.413 g
NaH2PO4 • 2 H2O, 17.07 g Na2HPO4 • 12 H2O and 1.015 g of MgCl2 • 12 H2O were
dissolved in 1 L UPW. If necessary, the pH was adjusted to 7.0 with 1 M NaOH or 1 M
HCl. The buffer was sterile filtered and stored at 4 °C.
UV data were collected on
a
Perkin-Elmer Lambda series UV/VIS
spectrophotometer (Lambda 16, Lambda 20, or Lambda 40). Measurements were usually

S3
carried out at 30 °C, unless stated otherwise. Scan speed was 480 nm*s-1 and the
bandwidth was set to 2 nm.
Decarboxylation kinetics. The MPS-catalyzed decarboxylation of oxaloacetate was
assayed by monitoring the absorption decrease at 305 nm (Δε305 = 173 M-1cm-1).
Reactions were carried out in 50 mM PIPES buffer, pH 7.0 with 5 mM MgCl2 at 30°C.
Enzyme concentration was 150 nM. The enzyme solution was pre-incubated for 3 min at
30° C and then the reaction was initiated by addition of oxaloacetate. Oxaloacetate
concentration varied between 25 μM and 4 mM. Initial velocities were determined by
linear regression and data was fitted to the Michaelis-Menten equation v0 = kcat
[E][S]/(Km + [S]), where v0 is the initial velocity, [S] is substrate concentration and [E] is
total enzyme concentration.
RP-HPLC was performed on a Waters HPLC system with 220 & 254 nm UV
detection. For analytical runs, a C8 column (Macherey-Nagel Nucleosil 250 mm × 4.6
mm, 300 Å, 5 μm), a C18 column (Macherey-Nagel Nucleosil 250 mm × 4.6 mm, 100 Å,
5 μm), or a short C18 column (Waters Polarity 100 mm × 4.6 mm, 100 Å, 3 μm) at a flow
rate of 1 ml/min were used. Peptides and small organic molecules were eluted with linear
gradients of solvents A and B (A = acetonitrile containing 0.05% TFA, B = H2O
containing 0.1% TFA). Preparative RP-HPLC separations were performed using a C8
column (Macherey-Nagel Nucleosil 250 mm × 21 mm, 300 Å, 7 μ) or a C18 column
(Vydac 250 mm × 22 mm, 300 Å, 10 μ) at a flow rate of 10 or 8 ml/min. Linear gradients
of solvents A and B (A = CH3CN, B = H2O containing 0.1% TFA) were used. Gradients
for analytical and preparative RP-HPLC are given as %A/%B to %A/%B followed by the
time. If not stated otherwise, quoted retention times refer to the gradients 10/90 to 35/65
in 25 min at a flow rate of 1 ml / min for analytical HPLC and 10/90 to 45/55 in 55 min at
a flow rate of 10 ml / min for preparative HPLC. In cases where it was desirable to
maximize the yield after preparative RP-HPLC the products were eluted with pure water
to prevent undesired side reactions such as elimination to 6a or cyclization to the lactone.
Test for MPS-catalyzed elimination of 5a to give 6a. A sample of 2-oxo-4-
hydroxy-4-phenyl-butanoic acid (5a) was dissolved in 50 mM phosphate buffer, 5 mM
MgCl2, pH 7.0 and was treated with MPS (300 nM). The signal increase at 340 nm
(conjugated enone 6a absorbs strongly in this region while 5a is silent) was monitored
over 1 h and compared to a control reaction that contained none of the enzyme. There
was a slow and identical increase in absorbance in both the enzymatic and control
reactions indicating no catalysis by MPS.
Procedure for the aldol kinetics outlined in Figure 1. The reaction of benzaldehyde
1
and oxaloacetate in the presence of MPS was monitored by H-NMR spectroscopy on a
Bruker AV600 instrument set to 300 K. Kinetic experiments were carried out in 50 mM
phosphate buffer, 5 mM MgCl2, pH 7.0 in 9:1 H2O:D2O. For each kinetic run 4 μl of a 20
mM solution of sodium-3-trimethylsilyltetradeuteriopropionate (TSP) (in reaction buffer)
and the appropriate amounts of the benzaldehyde and oxaloacetate solutions (prepared as
stock solutions in the reaction buffer) were added to a clean high-precision Wilmad 520-
PP NMR tube; the total volume of each sample was 800 μl. The sample was introduced
into the spectrometer and allowed to equilibrate for five minutes before shimming. Water
suppression was then optimized on the equilibrated sample using presaturation and a
spectrum was recorded to verify the initial conditions. The sample was removed from the

S4
spectrometer and the enzyme solution added (6.4 μl of a 12.5 μM solution for a final
concentration of 100 nM) followed by six careful inversions (t0 set to first inversion). The
sample was introduced again and data collection was immediately initiated. Consumption
of benzaldehyde and the appearance of product were quantified by measuring the change
in the integral of the aromatic protons of both the benzaldehyde starting material and the
aldol product 5a relative to the integral of the internal standard (TSP, s, 9H, 0 ppm, 0.1
mM) (see Figure S1 for details). Initial velocities were determined by linear regression
using the program Kaleidagraph and the data were fitted to the Michaelis-Menten
equation: v0 = kcat [E][S]/(Km + [S]).
Figure S1. Representative set of NMR spectra used to determine initial velocities
for aldol kinetics
5H’s of Benzaldehyde
4H’s of Product
b0165-4 4 1 C:\Bruker\TOPSPIN Dennis
t = 0 s
t = 185 s
t = 280 s
8.0
7.8
7.6
[ppm]
Retro-aldol activity of MPS with 5a. The retro-aldol reaction was analyzed with an
NADH/lactate dehydrogenase (LDH) coupled assay. Eight kinetic runs were collected in
a concentration range for 5a of 25 μM to 900 μM and an enzyme concentration of 300
nM. The assay was run in MPS reaction buffer A (50 mM PIPES buffer, see General
notes section for further details) containing 0.2 mM NADH and 1.5 μg of LDH. Ketoacid
5a was added to the reaction buffer containing the assay components and the sample was
monitored for 3 minutes to insure no background was observed. The enzyme was then
added to bring the total volume to 800 μL and the absorption increase at 340 nm was
monitored for five minutes. Initial velocities were determined by linear regression using
the program Kaleidagraph and the data were fitted to the Michaelis-Menten equation: v0

S5
= kcat [E][S]/(Km + [S]). The background reaction was determined by NMR analysis
under the same conditions (except that the buffer contains 10% D2O) by measuring the
appearance of products (benzaldehyde and the elimination product) and the
disappearance of starting material as a function of time. Concentrations were determined
by comparison with the known concentration of the internal standard TSP (sodium-3-
trimethylsilyltetradeuteriopropionate, [TSP] = 2 mM).
Compounds prepared by MPS catalyzed aldol reactions. NOTE: Although drawn in
the acid form, those compounds not purified by acidic RP-HPLC were typically
isolated as the sodium salt.
Analytical scale aldol reactions. The reaction of oxaloacetate with various
aldehydes in the presence and absence of MPS was carried out in 50 mM phosphate
buffer with 5 mM MgCl2 at pH 7.0 at room temperature. The two reactions were run in
parallel, one with enzyme and a control containing an equivalent volume of buffer. A 1
M solution of aldehyde in acetonitrile (10 μl) was added to 710 μl of MPS reaction buffer.
Then, 80 μl of a 10 μM MPS solution in MPS reaction buffer was added to one sample
(final MPS concentration 800 nM) and 80 μl of buffer to the control, and the samples
were incubated for 10 min at room temperature. The reaction was initiated by addition of
100 μl of a 60 mM solution of oxaloacetate in MPS reaction buffer and both reaction
vessels were gently shaken. After 1 h another 100 μl of 60 mM oxaloacetate were added
to give the final reaction volume of 1 ml. After 2 h, both samples were frozen in liquid
nitrogen and lyophilized. The remaining solid was dissolved in 200 μl of 49.5%
acetonitrile, 49.5% water and 1% TFA. This solution was then analyzed by analytical RP-
HPLC, using a linear gradient starting from 90% water and 10% acetonitrile, going to
65% water in 15 min and then to 10% water in another 10 min and also by LC-MS
starting from 90% water going to 40% water in 50 min. For analysis of these reactions,
the analytical RP-HPLC eluants (H2O and CH3CN) contained 0.5% TFA in order to
improve separation of the 2-oxo-carboxylic acid derivatives which are the products of the
aldol reaction. This leads to some loss of water to give the elimination products 6 and this
was considered when determining conversion.
Preparative scale aldol reactions procedure. 0.5 ml of a 1 M aldehyde solution in
acetonitrile was added to 17.5 ml of 50 mM phosphate buffer, 5 mM MgCl2, pH 7.0. 1 ml
of a 100 μM enzyme solution (in the same buffer) was added to give a final MPS
concentration of 5 μM and the mixture was incubated for 10 min at room temperature.
The reaction was then initiated by addition of 1 ml of a 60 mM solution of oxaloacetate
(in the reaction buffer). Additional aliquots of oxaloacetate were added every 60 min. In
total, 7 ml of oxaloacetate solution were added over 7 h. At the end of the reaction, the
sample was frozen and lyophilized. The solid was then dissolved in methanol and filtered
to remove inorganic salts. The ethanol was evaporated and the solid was dissolved in
CD3OD for NMR analysis and quantitation by comparing with an internal standard. The
products were also weighed to verify the NMR yield (typically 10% lower yields by
NMR indicating possible NMR-silent impurities in the dry-mass). If analytically pure
products were required preparative RP-HPLC (100% water, C-18 column, 8 mL/min,
retention time: 12.5 min) could be used. The use of an acidic mobile phase led to both
elimination and cyclization side-products.

S6
(S)-4-Hydroxy-2-oxo-4-phenylbutanoic acid (5a). Benzaldehyde
and oxaloacetate were reacted as described above to deliver 5a in 72 % yield. A labelled
1
1H-NMR spectrum of the crude reaction mixture in CD3OD is shown below. H-NMR
(300 MHz, CD3OD): 7.2-7.4 (m, 5H), 5.2 (dd, J = 8.4, 4.5 Hz, 1H), 3.2 (dd, J = 36, 8.4
Hz, 1H), 3.1 (dd, J = 36, 4.5 Hz, 1H). HRMS-EI(+): Calc. for C10H8O4 [M+H]: 193.0501.
1
Found: 193.0505. The elimination product 6a was also characterized. H-NMR (300
MHz, CD3OD): 7.69 (d, J = 16.5 Hz, 1H), 7.2-7.4 (m, 4H), 7.65 (m, 1H), 6.98 (d, J =
16.5 Hz, 1H). 13C-NMR (75 MHz, CD3OD): 119.9, 129.7, 129.9, 129.9, 130.0, 130.0,
132.5, 144.2, 171.6, 196.2. MS HRMS-EI(+): Calc. for C10H8O3 [M+]: 176.0473. Found:
176.0473. Analytical RP-HPLC retention times (conditions as described in General Notes
section): (5a): 13.8 min. (cyclization): 15.5 min. (6a): 18.5 min.
4-(4-fluorophenyl)-4-Hydroxy-2-oxobutanoic acid (5b). 4-
fluorobenzaldehyde and oxaloacetate were reacted as described above to deliver a 9:1
1
mixture of 5b and 6b in 52 % yield. The identity of 5b was confirmed by MS and H
NMR. 1H-NMR (300 MHz, CD3OD): 7.33 (dd, J = 8.7, 5.5 Hz, 2H) 7.09 (dd, J = 8.7, 8.7
Hz, 2H), 5.17 (dd, J = 8.3, 4.6 Hz, 2H), 3.13 (dd, J = 16.2, 8.4 Hz, 1H), 3.03 (dd, J = 16.2,
4.7 Hz, 1H). HRMS-EI(+): Calc. for C10H8O4F [M-H]: 211.0407. Found: 211.0412.
Analytical RP-HPLC retention times (conditions as described in General Notes section):
(6b): 19.9 min. (4-fluoro-benzaldehyde, 4b): 16.7 min.

S7
4-Hydroxy-4-(4-methoxyphenyl)-2-oxobutanoic acid (5c) and
elimination product (6c). 4-Methoxybenzaldehyde and oxaloacetic acid were combined
as described above to deliver a 1:3 mixture of 5c and 6c in 30% yield along with 30%
recovered starting material. The small amount of 5c obtained precluded its
characterization.
4-(furan-2-yl)-4-Hydroxy-2-oxobutanoic
acid
(5d).
2-
furancarboxaldehyde and oxaloacetic acid were combined as described above to deliver a
3:1 mixture of 5d and 6d in 49% yield. 5d: 1H-NMR (300 MHz, CD3OD): 7.42 (dd, J =
1.8, 0.8 Hz, 1H), 6.33 (dd, J = 3.2, 1.8 Hz, 1H), 6.28 (dd, J = 3.2, 0.8 Hz, 1H), 5.18 (dd, J
= 7.4, 5.7 Hz, 1H), 3.23 (d, J = 7.5 Hz, 1H), 3.22 (d, J = 5.7 Hz, 1H). 13C-NMR (75 MHz,
CD3OD): 46.7, 64.2, 107.0, 111.2, 143.2, 157.2, 172.0, 205.0.
(E)-4-Hydroxy-2-oxo-6-phenylhex-5-enoic
acid
(5e).
Cinnamaldehyde and oxaloacetate were combined as described earlier to deliver the
1
product 5e in 35% yield (50% conv). H-NMR (300 MHz, CD3OD): 7.59-7.05 (m, 5H),
6.63 (d, J = 15.9 Hz, 1H), 6.29 (dd, J = 15.9, 6.1 Hz, 1H), 4.77 (dddd, J = 6.6, 6.1, 6.1,
1.3 Hz, 1H), 3.00 (d, J = 6.6 Hz, 1H), 3.00 (d, J = 6.1 Hz, 1H).
(S)-4-Hydroxy-2-oxo-4-(pyridin-2-yl)butanoic acid (5f). 2-
Pyridinecarboxaldehyde and oxaloacetate were combined as described earlier to deliver
the product 5f in 95% yield. This compound was identical in all respects to the reported
characterization data.4 The enantioselectivity (36%, see Figure S2) was determined by
first treating 5f with a 1:1 mixture of ethanethiol and fuming HCl (to give 0.2 M 5f) and
the resulting lactone was separated by chiral HPLC using a Daicel Chiralpak® AD
column (97:3 Hexane : Isopropanol, 1 mL/min, 220 nm detection) Retention times:
Major: 18.9 min, Minor: 22.2 min. The absolute stereochemistry was determined by
comparison of the optical rotation to the known value: [α]Dlit = -39, [α]D = -15.8 (sample
of 36% ee, c = 0.17).

S8
Figure S2. Left: Authentic racemic lactone. Right: Product obtained from
enzymatic reaction, 36% ee.
(E)-4-Hydroxy-2-oxododeca-9,11-dienoic acid (5g). 4g and
oxaloacetate were combined as described earlier to deliver the product 5g in 48% yield.
Analytical RP-HPLC retention times (conditions as described in General Notes section):
1
(5g): 23.1 min (4g): 24.3 min. H-NMR (300 MHz, CD3OD): 6.29 (ddd, J = 17.0, 10.2,
10.2 Hz, 1H), 6.15-5.98 (m, 1H), 5.78-5.62 (m, 1H), 5.05 (d, J = 17.0 Hz, 1H), 4.96 (d, J
= 10.2 Hz, 1H Note: partially obscured by methanol solvent resonance), 4.05 (m, 1H),
2.86 (dd, J = 15.1, 4.0 Hz, 1H), 2.79 (dd, J = 15.1, 6.5 Hz, 1H), 2.15-2.03 (m, 2H), 1.55–
1.33 (m, 6H). 13C-NMR (75 MHz, CD3OD): 26.3, 30.4, 33.6, 38.3, 48.7, 68.6, 115.0,
132.5, 136.1, 138.7, 170.6, 205.3. HRMS-ESI(-): calc. 225.1132; found 225.1131.
5-Ethyl-4-hydroxy-2-oxoheptanoic acid (5h). 4h and oxaloacetate
were combined as described earlier to deliver the product 5h in 45% yield. 1H-NMR (300
MHz, CD3OD): 4.15 (ddd, J = 8.5, 4.3, 4.3 Hz, 1H), 2.83 (dd, J = 15.2, 3.5 Hz, 1H), 2.76
(dd, J = 15.2, 7.5 Hz, 1H), 1.50-1.20 (m, 4H), 0.97-0.86 (m, 6H).
(S)-4-((R)-2,2-Dimethyl-1,3-dioxolan-4-yl)-4-hydroxy-2-
oxobutanoic acid (8a)
(S)-4-((S)-2,2-dimethyl-1,3-dioxolan-4-yl)-4-hydroxy-2-
oxobutanoic acid (8b). The general procedure was followed with glyceraldehyde

S9
derivatives 7a or 7b to deliver 8a or 8b respectively. Their spectral properties were
identical to the literature report.5
(1) T. R. Hoye, H. Zhao, J. Org. Chem. 2002, 67, 4014–4016.
(2) Watanabe, K.; Oikawa, H.; Yagi, K.; Ohashi, S.; Mie, T.; Ichihara, A.; Honma, M. J. Biochem. 2000,
127, 467-473.
(3) Serafimov, J.; Lehmann, H. C.; Oikawa, H.; Hilvert, D. Chem. Commun. 2007, 1701-1703.
(4) Henderson, D. P; Shelton, M. C.; Cotterill, I. C.; Toone, E. J. J. Org. Chem. 1997, 62, 7910-7911.
(5) Lamble, H. J.; Danson, M. J.; Hough, D. W.; Bull, S. D. Chem. Commun. 2005, 124-126.