A unique cis -3-hydroxy- l -proline dehydratase in the enolase superfamily

Article abstract of DOI:10.1021/ja5103986

The genome of Labrenzia aggregata IAM 12614 encodes an uncharacterized member of the muconate lactonizing enzyme (MLE) subgroup of the enolase superfamily (UniProt ID A0NXQ8). The gene encoding A0NXQ8 is located between genes that encode members of the proline racemase superfamily, 4R-hydroxyproline 2-epimerase (UniProt ID A0NXQ7; 4HypE) and trans-3-hydroxy-l-proline dehydratase (UniProt ID A0NXQ9; t3LHypD). A0NXQ8 was screened with a library of proline analogues; two reactions were observed with cis-3-hydroxy-l-proline (c3LHyp), competing 2-epimerization to trans-3-hydroxy-d-proline (1,1-proton transfer) and dehydration to δ¹-pyrroline-2-carboxylate (β-elimination; c3LHyp dehydratase), with eventual total dehydration. The genome context encoding A0NXQ8 both (1) confirms its novel c3LHyp dehydratase function and (2) provides evidence for metabolic pathways that allow L. aggregata to utilize several isomeric 3- and 4-hydroxyprolines as sole carbon sources.

Full text of DOI:10.1021/ja5103986

Communication
pubs.acs.org/JACS
A Unique cis-3-Hydroxy‑L‑proline Dehydratase in the Enolase
Superfamily
Xinshuai Zhang,‡ Ritesh Kumar,‡ Matthew W. Vetting,§ Suwen Zhao,∥ Matthew P. Jacobson,∥
Steven C. Almo,§ and John A. Gerlt*,†,‡
‡
†Departments of Biochemistry and Chemistry and Institute for Genomic Biology, University of Illinois at Urbana−Champaign,
Urbana, Illinois 61801, United States
§Department of Biochemistry, Albert Einstein College of Medicine, Bronx, New York 10461, United States
∥Department of Pharmaceutical Chemistry, University of California, San Francisco, California 94143, United States
S
* Supporting Information
ABSTRACT: The genome of Labrenzia aggregata IAM
12614 encodes an uncharacterized member of the
muconate lactonizing enzyme (MLE) subgroup of the
enolase superfamily (UniProt ID A0NXQ8). The gene
encoding A0NXQ8 is located between genes that encode
members of the proline racemase superfamily, 4R-
hydroxyproline 2-epimerase (UniProt ID A0NXQ7;
4HypE) and trans-3-hydroxy-L-proline dehydratase (Uni-
Prot ID A0NXQ9; t3LHypD). A0NXQ8 was screened
with a library of proline analogues; two reactions were
observed with cis-3-hydroxy-L-proline (c3LHyp), compet-
ing 2-epimerization to trans-3-hydroxy-D-proline (1,1-
proton transfer) and dehydration to Δ1-pyrroline-2-
carboxylate (β-elimination; c3LHyp dehydratase), with
eventual total dehydration. The genome context encoding
A0NXQ8 both (1) confirms its novel c3LHyp dehydratase
function and (2) provides evidence for metabolic pathways
that allow L. aggregata to utilize several isomeric 3- and 4-
hydroxyprolines as sole carbon sources.
Figure 1. (A) SSN of the MLE subgroup displayed at a BLASTP e-
value of 10−50; the “Roseo” cluster is boxed. (B) SSN of the “Roseo”
cluster displayed at a BLASTP e-value of 10−80; 2PMQ, A0NXQ8, and
D7A0Y2 are denoted by the red node, orange node, and green node,
respectively.
cluster (boxed), includes an uncharacterized member (UniProt
ID A0NXQ8) from the bacterium Labrenzia aggregata IAM
12614.
Recently, our laboratories assigned the trans-4-hydroxy-L-
proline betaine 2-epimerase function to Uniprot ID Q0FPQ4
(PDB 2PMQ) in the “Roseo” cluster.4 A0NXQ8 shares 34%
sequence identity with Q0FPQ4. When the BLAST e-value
threshold is decreased to 10−80, the “Roseo” cluster segregates
into several presumed isofunctional families (Figure 1B), with
A0NXQ8 and Q0FPQ4 located in different clusters, suggesting
they have different in vitro activities and in vivo functions.
Genome neighborhood context can provide useful informa-
tion for deducing the functions of uncharacterized bacterial
enzymes by providing clues about the type of substrate; activity
screening of a focused physical library then may allow the
substrate to be identified.3 The genome context for A0NXQ8
reveals that its gene is located between genes encoding
members of the proline racemase superfamily (UniProt IDs
A0NXQ7 and A0NXQ9; Figure 2A). Their in vitro activities are
4R-hydroxyproline 2-epimerase (4HypE; A0NXQ7) and trans-
3-hydroxy-L-proline dehydratase (t3LHypD; A0NXQ9).6 The
genome context does not provide sufficient information to
assign a function to A0NXQ8; however, we used it to
hypothesize that a substituted proline is the substrate.
he members of the enolase superfamily (ENS) are
Tfunctionally diverse, catalyzing reactions initiated by
abstraction of the α-proton of a carboxylate substrate.1 The
reactions include 1,1-proton transfer (racemization and
epimerization) and β-elimination (dehydration and deamina-
tion).2 The UniProt database (release 2014_10) contains
sequences for 83 038 members of the ENS. Perhaps 50% of
these have unknown functions because they, or likely
orthologues, have not been functionally characterized.
The MLE subgroup of the ENS includes the cis,cis-muconate
cycloisomerase (MLE), o-succinylbenzoate synthase (OSBS),
N-succinyl-amino acid racemase (NSAR), dipeptide epimerase,
and 4R-hydroxyproline betaine 2-epimerase functions.3,4 In this
manuscript, we report the identification of the novel cis-3-
hydroxy-L-proline dehydratase (c3LHypD) function, the first
example of an amino acid dehydratase in the ENS.
Sequence−function relationships in enzyme superfamilies
can be visualized using sequence similarity networks (SSNs).5
When the SSN for the MLE subgroup is displayed at a BLAST
e-value threshold of 10−50 (Figure 1A), several clusters are
observed. One of the divergent clusters, designated the “Roseo”
Received: October 15, 2014
Published: January 21, 2015
© 2015 American Chemical Society
1388
DOI: 10.1021/ja5103986
J. Am. Chem. Soc. 2015, 137, 1388−1391

Journal of the American Chemical Society
Communication
Figure 4. Initial progress of the A0NXQ8-catalyzed reaction with
c3LHyp during the initial of the reaction showing formation of
Figure 2. (A) Genome context of A0NXQ8 from L. aggregata IAM
12614. (B) Genome context of orthologous D7A0Y2 from Starkeya
novella DSM 506. The genes encoding orthologues are highlighted
with the same color. DHDPS, member of the dihydrodipicolinate
synthetase superfamily.
1
t3DHyp (epimerization) and Pyr2C (dehydration) monitiored by H
NMR; the 1H NMR assay contained 10 mM c3LHyp, 10 mM MgCl2,
50 mM sodium phosphate, pH 8.0, and 0.14 μM A0NXQ8. Eventually,
only Pyr2C is observed.
We screened A0NXQ8 with a library of proline derivatives,
including L-proline and several stereoisomers of 3-hydroxypro-
line, 4-hydroxyproline and their mono- and di-N-methylated
derivatives (Figure S1). Only cis-3-hydroxy-L-proline (c3LHyp)
We assign the c3LHyp dehydratase (c3LHypD) function to
A0NXQ8. Four other reactions involving abstraction of the α-
proton of a proline derivative have been described for the
proline racemase superfamily (Pfam PF05544): proline race-
mase,7 trans-4-hydroxy-L-proline 2-epimerase (4HypE),8 trans-
3-hydroxy-L-proline 2-epimerase (t3LHypE), and trans-3-
hydroxy-L-proline dehydratase (t3LHypD).6,9 To the best of
our knowledge, A0NXQ8 is the first c3LHypD and the first
amino acid dehydratase in the ENS.3,4
The crystal structure of A0NXQ8 was determined at 2.2-Å
resolution with bound Mg2+ (PDB entry 4MGG). 4MGG has
the ENS bidomain structure (Figure S7); as expected for a
member of the MLE subgroup, the C-terminal (β/α)7β-barrel
domain contains two conserved general acid/base residues, Lys
165 and Lys 265, located on the opposite faces of the active site
at the ends of the second and sixth β-strands in the barrel
domain.3
1
was a substrate. The H NMR spectrum of the reaction with
c3LHyp performed in D2O revealed (1) exchange of the α-
proton with solvent, (2) 2-epimerization to trans-3-hydroxy-D-
proline (t3DHyp), and (3) dehydration to Δ1-pyrroline-2-
carboxylate (Pyr2C). All three reactions are initiated by
abstraction of the α-proton of the substrate, the common
initial partial reaction for members of the ENS (Figure 3). As
Mutagenesis studies were performed to investigate the
reaction mechanism. Using c3LHyp, no activity was observed
(change in optical rotation or exchange of the α-proton with
solvent deuterium by 1H NMR) for either the K165A or
K265A mutant. With t3DHyp prepared using A0NXQ8 in H2O
(as a 1:1:1 mixture with c3LHyp and Pyr2C; t3DHyp is not
commercially available), no dehydration or exchange of the α-
proton with solvent deuterium was observed for either mutant
(Figure S2). Therefore, the identities of the acid/base catalysts
cannot be specified. Nevertheless, the reaction is expected to
involve abstraction of the α-proton of c3LHyp (by either Lys
165 or Lys 265) to generate an enolate anion intermediate.
The promiscuity (competing dehydration and epimerization)
likely is a consequence of divergent evolution of function in the
enolase superfamily. Abstraction of the α-proton from the cis-3-
hydroxy-L-proline substrate by one active site Lys results in
formation of the enolate anion intermediate that partitions
between general acid catalysis of departure of the 3-OH group
to form Δ1-pyrroline-2-carboxylate (β-elimination/dehydra-
tion) and protonation of carbon-2 to form trans-3-hydroxy-D-
proline (1,1-proton transfer) by the second active site Lys. For
dehydration of c3LHyp, departure of the 3-OH group from the
intermediate would be general acid-catalyzed by the second
Lys; for epimerization, the intermediate is protonated on the
opposite face of carbon-2 by the same Lys (Figure 3). We have
observed similar examples of competing dehydration and
epimerization in acid sugar dehydratase in the ENS10,11 and
attributed these to the consequences of “hard-wired” active site
Figure 3. Proposed mechanism for the A0NXQ8-catalyzed reaction.
expected for irreversible dehydration and reversible epimeriza-
tion reactions that share a common intermediate, only the
Pyr2C dehydration product was observed when the reaction
was complete.
The kinetic constants for disappearance of c3LHyp were
measured using a polarimeter: kcat = 23 0.6 s−1, KM = 8.1
0.9 mM, and kcat/KM = 2800 M−1 s−1. The relative rates of
epimerization and dehydration were determined by monitoring
the reaction by 1H NMR using several concentrations of
c3LHyp: the resonances associated with C3 of c3LHyp,
t3DHyp, and Pyr2C have different chemical shifts (Figure
S3). As the reaction proceeds, the intensities of the signals
associated with c3LHyp decrease as the intensities of the signals
associated with the t3DHyp and Pyr2C products increase with
no detectable lag (Figure 4 and Figure S4). The rate of
epimerization is about 1.5-fold greater than that of dehydration
at substrate concentrations both below and above the value of
the KM (Figure S5); thus, the kcat for epimerization is ∼14 s−1
and that for dehydration is ∼9 s−1. The initial linear
dependence of product formation on time for both reactions
(absence of a lag for dehydration) indicates that the substrate
for the dehydration reaction is c3LHyp and not t3DHyp, the
product of the 2-epimerization reaction.
1389
DOI: 10.1021/ja5103986
J. Am. Chem. Soc. 2015, 137, 1388−1391

Journal of the American Chemical Society
Communication
Figure 5. Pathways for degradation trans-4-hydroxy-L-proline, trans-3-hydroxy-L-proline, and cis-3-hydroxy-L-proline by L. aggregata IAM 12614 and
S. novella DSM 506.
functional groups that can be accessed by substrates with
different structures as determined by changes in the specificity
determinants, thereby enhancing the evolution of new
functions.12
L. aggregata IAM 12614 utilizes c3LHyp as carbon source.
Transcriptomics using qRT-PCR revealed that the gene
encoding A0NXQ8 is upregulated 10.5-fold when L. aggregata
IAM 12614 is grown on c3LHyp as carbon source compared to
growth on glucose (Table S1), implicating c3LHypD in the
catabolic pathway.
of genome context to facilitate the assignment of function to
uncharacterized enzymes discovered in genome projects.
The X-ray coordinates and structure factors for c3HypD
from L. aggregata IAM 12614 with Mg2+ in the active site have
been deposited in the PDB (4MGG). This report describes the
characterization of in vitro enzymatic activities of proteins with
the following UniProt accession IDs: A0NXQ8, D7A0Y2,
D7A0Y0, D7A0Y1, and D7A0Y3.
ASSOCIATED CONTENT
■
S
The genome sequence of L. aggregata IAM 12614 is an
incomplete draft sequence. To identify the catabolic pathway in
which c3LHypD participates, we studied S. novella DSM 506 for
which a complete genome sequence is available; it encodes an
orthologue of A0NXQ8 (UniProt ID D7A0Y2; Figure 1B;
A0NXQ8 and D7A0Y2 share 75% sequence identity).
The genome context of D7A0Y2 is shown in Figure 2B. Its
gene is located between genes encoding orthologues of 4HypE
(UniProt ID D7A0Y3) and a t3LHypD (UniProt ID
D7A0Y1).6 As expected from the sequence identity, D7A0Y2
also catalyzes the c3LHypD reaction. The gene encoding a
member of the malate dehydrogenase-2 (MDH-2) superfamily
(UniProt ID D7A0Y0) is adjacent to those encoding t3LHypD
(D7A0Y1) and c3LHypD (D7A0Y2). We hypothesized that
D7A0Y0 catalyzes the reduction of Pyr2C generated by
dehydration of both c3LHypD and t3LHypD to L-proline;6,13
in vitro assays of D7A0Y0 confirmed that it is a NADPH-
dependent Pyr2C reductase, kcat = 26 1 s−1, KM = 2.8 0.4
mM, and kcat/KM = 9100 M−1 s−1. The L-proline product likely
is catabolized to L-glutamate via the bifunctional L-proline
dehydrogenase/Δ1-pyrroline-5-carboxylate dehydrogenase that
is encoded by the genome.14
Both L. aggregata IAM 12614 and S. novella DSM 506
encode pathways (Figure 5) for utilization of (1) trans-4-
hydroxy-L-proline [catalyzed by orthologues of 4HypE (proline
racemase superfamily), the FAD-dependent oxidoreductase
(c4DHyp oxidase), and the member of the DHDPS superfamily
(Δ1-pyrroline 4-hydroxy 2-carboxylate dehydratase/deaminase,
Pyr4H2C DD)],4 (2) trans-3-hydroxy-L-proline [catalyzed by
an orthologue of t3LHypD6 and the Pyr2C reductase (vide
supra)], and (3) cis-3-hydroxy-L-proline (catalyzed by
c3LHypD and Pyr2C reductase). This work illustrates the use
* Supporting Information
A description of the experimental procedures. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
j-gerlt@illinois.edu
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This research was supported by a program project grant and
three cooperative agreements from the U.S. National Institutes
of Health (P01GM071790, U54GM093342, U54GM074945,
and U54GM094662). Use of the Advanced Photon Source, an
Office of Science User Facility operated for the U.S.
Department of Energy (DOE) Office of Science by Argonne
National Laboratory, was supported by the U.S. DOE under
Contract No. DE-AC02-06CH11357. Use of the Lilly Research
Laboratories Collaborative Access Team (LRL-CAT) beamline
at Sector 31 of the Advanced Photon Source was provided by
Eli Lilly Company, which operates the facility. We thank Dr.
Xudong Guan, NMR specialist at the Institute for Genomic
Biology, University of Illinois at Urbana−Champaign for
1
assistance with the H NMR kinetic experiments.
REFERENCES
■
(1) Babbitt, P. C.; Hasson, M. S.; Wedekind, J. E.; Palmer, D. R.;
Barrett, W. C.; Reed, G. H.; Rayment, I.; Ringe, D.; Kenyon, G. L.;
Gerlt, J. A. Biochemistry 1996, 35, 16489−16501.
(2) Gerlt, J. A.; Babbitt, P. C.; Rayment, I. Arch. Biochem. Biophys.
2005, 433, 59−70.
1390
DOI: 10.1021/ja5103986
J. Am. Chem. Soc. 2015, 137, 1388−1391

Journal of the American Chemical Society
Communication
(3) Gerlt, J. A.; Babbitt, P. C.; Jacobson, M. P.; Almo, S. C. J. Biol.
Chem. 2012, 287, 29−34.
(4) Zhao, S.; Kumar, R.; Sakai, A.; Vetting, M. W.; Wood, B. M.;
Brown, S.; Bonanno, J. B.; Hillerich, B. S.; Seidel, R. D.; Babbitt, P. C.;
Almo, S. C.; Sweedler, J. V.; Gerlt, J. A.; Cronan, J. E.; Jacobson, M. P.
Nature 2013, 502, 698−702.
(5) Atkinson, H. J.; Morris, J. H.; Ferrin, T. E.; Babbitt, P. C. PLoS
One 2009, 4, e4345.
(6) Zhao, S.; Sakai, A.; Zhang, X.; Vetting, M. W.; Kumar, R.;
Hillerich, B.; Francisco, B. S.; Solbiati, J.; Steves, A.; Brown, S.; Akiva,
E.; Barber, A.; Seidel, R. D.; Babbitt, P. C.; Almo, S. C.; Gerlt, J. A.;
Jacobson, M. P. eLife 2014, DOI: 10.7554/eLife.03275.
(7) Buschiazzo, A.; Goytia, M.; Schaeffer, F.; Degrave, W.; Shepard,
W.; Gregoire, C.; Chamond, N.; Cosson, A.; Berneman, A.; Coatnoan,
N.; Alzari, P. M.; Minoprio, P. Proc. Natl. Acad. Sci. U.S.A. 2006, 103,
1705−1710.
(8) Goytia, M.; Chamond, N.; Cosson, A.; Coatnoan, N.; Hermant,
D.; Berneman, A.; Minoprio, P. PLoS One 2007, 2, e885.
(9) Visser, W. F.; Verhoeven-Duif, N. M.; de Koning, T. J. J. Biol.
Chem. 2012, 287, 21654−21662.
(10) Palmer, D. R.; Hubbard, B. K.; Gerlt, J. A. Biochemistry 1998, 37,
14350−14357.
(11) Yew, W. S.; Fedorov, A. A.; Fedorov, E. V.; Almo, S. C.; Gerlt, J.
A. Biochemistry 2007, 46, 9564−9577.
(12) Schmidt, D. M. Z.; Mundorff, E. C.; Dojka, M.; Bermudez, E.;
Ness, J. E.; Govindarajan, S.; Babbit, P. C.; Minshull, J.; Gerlt, J. A.
Biochemistry 2003, 42, 8387−8393.
(13) Muramatsu, H.; Mihara, H.; Kakutani, R.; Yasuda, M.; Ueda, M.;
Kurihara, T.; Esaki, N. J. Biol. Chem. 2005, 280, 5329−5335.
(14) Srivastava, D.; Schuermann, J. P.; White, T. A.; Krishnan, N.;
Sanyal, N.; Hura, G. L.; Tan, A.; Henzl, M. T.; Becker, D. F.; Tanner, J.
J. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 2878−2883.
1391
DOI: 10.1021/ja5103986
J. Am. Chem. Soc. 2015, 137, 1388−1391