The formation of pyrroline and tetrahydropyridine rings in amino acids catalyzed by pyrrolysine synthase (PylD)

Article abstract of DOI:10.1002/anie.201402595

The dehydrogenase PylD catalyzes the ultimate step of the pyrrolysine pathway by converting the isopeptide L-lysine-Nε-3R-methyl-D-ornithine to the 22nd proteinogenic amino acid. In this study, we demonstrate how PylD can be harnessed to oxidize various isopeptides to novel amino acids by combining chemical synthesis with enzyme kinetics and X-ray crystallography. The data enable a detailed description of the PylD reaction trajectory for the biosynthesis of pyrroline and tetrahydropyridine rings as constituents of pyrrolysine analogues.

Full text of DOI:10.1002/anie.201402595

.
Angewandte
Communications
DOI: 10.1002/anie.201402595
Pyrrolysine Analogues
The Formation of Pyrroline and Tetrahydropyridine Rings in Amino
Acids Catalyzed by Pyrrolysine Synthase (PylD)**
Felix Quitterer, Philipp Beck, Adelbert Bacher, and Michael Groll*
Abstract: The dehydrogenase PylD catalyzes the ultimate step
of the pyrrolysine pathway by converting the isopeptide l-
lysine-Ne-3R-methyl-d-ornithine to the 22nd proteinogenic
amino acid. In this study, we demonstrate how PylD can be
harnessed to oxidize various isopeptides to novel amino acids
by combining chemical synthesis with enzyme kinetics and X-
ray crystallography. The data enable a detailed description of
the PylD reaction trajectory for the biosynthesis of pyrroline
and tetrahydropyridine rings as constituents of pyrrolysine
analogues.
Scheme 1. Biosynthesis of pyrrolysine (1) starting from two l-lysine
molecules. C-terminal and N-terminal parts of 1a and 1 are depicted
in gray and black, respectively. SAM=S-adenosylmethionine.
P
yrrolysine (1) is the 22nd genetically encoded amino acid,
which is incorporated into some proteins by means of
ribosomal read-through of an amber stop codon (UAG) in
certain methanogenic archaea and some eubacteria, including
[
1]
the human pathogen Bilophila wadsworthia. Pyrrolysine
comprises a 4-methylpyrroline-5-carboxylate, which is linked
[
1b]
by an isopeptide bond to Ne of l-lysine.
Recent studies
[
2]
revealed that the gene cluster pylBCDST orchestrates the
[3]
biosynthesis of 1 (Scheme 1) and its insertion into the
proteins MtmB, MtbB, and MttB, which function in the
[
4]
[5]
breakdown of methylamines.
Crystal structures of PylD in complex with the substrate
and product surrogates l-lysine-Ne-d-ornithine (0a) and
pyrroline-carboxy-lysine (0) revealed that the C-terminal
lysine moiety is well coordinated, whereas the N-terminal unit
only partially occupies the hydrophobic active site cavity,
[
3e]
which is filled by a cluster of water molecules. The aim of
the presented work was to expand the scope of PylD
substrates and harness the enzyme to oxidize isopeptides to
novel amino acids (Scheme 2).
[
+]
[+]
[
*] F. Quitterer, P. Beck, Prof. Dr. A. Bacher, Prof. Dr. M. Groll
Center for Integrated Protein Science Munich (CIPSM)
Lehrstuhl fꢀr Biochemie, Technische Universitꢁt Mꢀnchen
Lichtenbergstrasse 4, 85748 Garching (Germany)
Scheme 2. Isopeptide substrates (0a–6a) and their respective prod-
ucts (0–3).
E-mail: michael.groll@tum.de
+
[
] These authors contributed equally to this work.
[
**] We thank the staff of the beamline X06SA at the Paul Scherrer
Institute, Swiss Light Source, Villigen (Switzerland) for their help
with data collection and Katrin Gꢁrtner for excellent technical
assistance. Kinetic measurements were carried out by F.Q. at the
Biological & Organometallic Catalysis (BOC) Laboratories of
Prof. Dr. Jçrg Eppinger, King Abdullah University of Science and
Technology (KAUST), Thuwal, Saudi Arabia. This work was sup-
ported by the Hans-Fischer-Gesellschaft, by Award No. FIC/2010/07
from KAUST, and by the Deutsche Forschungsgemeinschaft (DFG,
grant GR1861/7-1).
Since structural and functional data of PylD with its
natural product pyrrolysine (1) have not been reported so far,
we synthesized the substrate l-lysine-Ne-3R-methyl-d-orni-
thine (1a, Figure S1). Compound 1a was cocrystallized with
PylD and the complex structure was solved to 2.2 ꢀ
resolution (Figure 1a, for details see the Supporting Infor-
mation). Inspection of the electron density map revealed that
1 has been formed. The structure displays that the l-lysine
moiety of the isopeptide is enclosed in a hydrophobic channel,
which is comprised by Leu3, Ile60, Phe63, and Ala103. The
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201402595.
8150
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 8150 –8153

Angewandte
Chemie
Inspired by the tolerated substrate spectrum of PylD, we
designed further derivatives and synthesized l-lysine-Ne-d-
lysine (3a) and l-lysine-Ne-l-lysine (4a) in which the N-
terminal moiety of 0a is extended by a CH unit. The PylD:3
2
complex structure (Figure 1c, 2.2 ꢀ resolution) demonstrates
that the d-lysine side chain is converted into a tetrahydropyr-
idine ring (Figure S2c), whereas conversion of l-lysine-Ne-l-
lysine (4a) is not catalyzed by PylD as shown by the complex
structure (Figure 1d, 2.1 ꢀ resolution) and by kinetic meas-
urements (Table 1). The distance of the free e’-amino group of
Table 1: Kinetic parameters for the conversion of PylD substrates. A
detailed table, including K , kcat, kcat/K , and v for 0a and 1a, is
m
m
max
provided in the Supporting Information (Table S2).
Substrate
Specific activity
v0 at 8 mM
[nmolmin
Rel. activity
at 8 mm [%]
À1
À1
À1
[nmolmin mg
]
]
[
3e]
0
1
2
3
4
5
6
a
a
a
a
a
a
a
54.5
61.6
10.2
14.1
3.4
72
100
24
0.6
<0.01
<0.01
<0.01
0.08
<0.001
<0.001
<0.001
4
a to C4 of the coenzyme appears perfectly suitable for
Figure 1. Active site of PylD in complex with a) 1, b) 2, c) 3, d) 4a,
e) 5a, and f) 6a. The 2F ÀF electron density map is contoured at
hydride transfer (3.3 ꢀ, Figure S3a); however, an analysis of
the rotational degrees of freedom of the terminal l-lysine side
chain suggests that ring closure after preceding oxidation of
the side chain would be hampered by steric interference with
the walls of the active site cavity.
o
c
1
.0s. The N-terminus (amino acids 1–11) is shown in red and the loop
connecting b-strand 1 and a-helix 3 (amino acids 55–59) is depicted in
blue. The active-site residues and ligands are presented as stick
models and water molecules are drawn as red spheres. Hydrogen
bonds are indicated by dashed lines. Stereo views are presented in the
Supporting Information (Figure S4).
Since the C-terminal parts, including the isopeptide bonds,
of the ligands 1a–4a perfectly match each other by a root
mean square deviation of less than 0.35 ꢀ, we started to
investigate its impact on enzyme catalysis. We placed l-
ornithine as the C-terminal unit, resulting in l-ornithine-Nd-
d-ornithine (5a) and l-ornithine-Nd-d-lysine (6a). Interest-
ingly, the crystal structures of PylD:5a (Figure 1e, 2.2 ꢀ
resolution) and PylD:6a (Figure 1 f, 2.2 ꢀ resolution) depict
both compounds at the active site, thus demonstrating that the
driving force for binding strongly depends on the coordina-
tion of the carboxy and amino groups of the C-terminus. Since
the C-terminal end of these analogues occupies the same
position as in the previous examples, the isopeptide motif is
forced to alter its location, orientation, and coordination as
shown in Figure S3b,c. This is conducive to major alterations
in the conformation of the amide substituents, which result in
an unfavorable distance (6.3 ꢀ) of the d’-amino group from
carboxy and amino groups of the ligand are coordinated by
direct and indirect hydrogen bonds to the backbone of PylD
(
for details see Figure S2a). The isopeptide motif is stabilized
by an indirect hydrogen bond between its NH group and
Asp104O via a defined water molecule, whereas the carbonyl
oxygen interacts with Leu4NH of the N-terminal loop,
[
3e]
participating in the induced-fit mechanism. The N-terminal
unit of 1 is oriented towards NADH with C5 of the pyrroline
ring in proximity (4.0 ꢀ) to C4 of the coenzymeꢁs pyridine
part. The 3R-methyl group of the pyrroline moiety displays
van der Waals interactions with Phe63, Phe108, and Leu247,
while the aliphatic ring is stabilized by the Leu4 side chain.
The structural data revealed that the closed active site
3
+
cavity around the N-terminal part of 1 is spacious (ca. 450 ꢀ ).
the pyridine nucleotide coenzyme to NAD in the case of 5a,
Hence, we were interested in the stereospecificity of the
enzyme and synthesized the diastereomer l-lysine-Ne-3S-
methyl-l-ornithine (2a) and cocrystalized it with PylD. The
thus preventing catalysis.
So far, all described structures of PylD in complex with
a ligand were obtained in the closed conformation. Hence, it
was surprising that 6a is fully defined in the electron density
map, although the enzyme adopts its open state. Similar as in
2
F ÀF electron density map of the PylD:2 complex (Fig-
o
c
ure 1b, 2.2 ꢀ resolution) shows the conversion of 2a, result-
ing in a tilted pyrroline ring in S,S configuration. Notably, the
[
3e]
the PylD:holo structure, the N-terminal residues (amino
acids 1–11) do not cover the active site but extend the N-
terminal helix. Furthermore, residues 55–59 of the loop
coordinating the carboxy and amino groups of the C-terminal
3S-methyl group points towards the side chain of Leu4 and
the nicotinamide ring of the coenzyme (for details see
Figure S2b).
Angew. Chem. Int. Ed. 2014, 53, 8150 –8153
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8151

.
Angewandte
Communications
moiety are structurally distorted. Though the terminal e’-
of any activating amino acid in the active site, it appears
reasonable that the terminal steps of PylD catalysis, in
particular the addition of the a-amino group to the imine
motif followed by the release of ammonia, are facilitated by
the cluster. A related reaction, the conversion of l-ornithine
amino group is correctly orientated towards the coenzyme
+
(
4.3 ꢀ to C4 of NAD ), 6a was not transformed. These
findings demonstrate that the closed state, which is not
feasible in case of 6a due to clashes with the PylD side chains
of Leu3 and Leu4 (Figure S5), is crucial for catalysis. On the
basis of the structural results, Leu4 seems to play a major role
in the coordination and orientation of the N-terminal
substrate moiety. In line with that, an L4A mutant did not
show any detectable turnover of 1a and therefore confirmed
the major impact of this amino acid in the closed state.
A closer inspection of the PylD complex structures
identified a set of organized water molecules in proximity to
the N-terminal isopeptide part (Figure S6). Notably, this
cluster is in contact with the heterocyclic nitrogen of all
products through a defined solvent molecule at a distance of
1
to D -pyrroline-5-carboxylate, is catalyzed by the enzyme
ornithine d-aminotransferase in plants. In contrast to the
PylD substrates, ornithine is oxidized by transamination with
[6]
2-oxoglutarate.
The presented data on PylD complexes with substrate
analogues in their open and closed conformations provide
detailed insights into the reaction mechanism of PylD
(Scheme 3):
a) Ligand binding occurs in the open conformation, mainly
driven by interactions of the C-terminal substrate part
with the protein.
2
.6–3.1 ꢀ (Figure 2a), which is absent in the electron density
b) The correct positioning of the ligandꢁs isopeptide bond
initiates the enzymatic induced fit, including amino acids
maps of the unreacted surrogates (Figure 2b). Due to the lack
1
–11 and 55–59.
c) The substrate sensor Leu4NH forms a defined H-bond
with the isopeptidic carbonyl oxygen of the ligand,
resulting in an appropriate orientation of the N-terminal
substrate moiety within the active site cavity.
d) In addition, the Leu4 side chain restricts the active site
cavity and forces the terminal amino group of the
+
substrate towards NAD to enable hydride transfer. The
resulting Schiff base is poised for a nucleophilic attack by
the ligandꢁs a amino group, resulting in a 2-aminopyrro-
lidinium intermediate.
e) The closed active site encompasses a fixed network of
water molecules that act as the proton shuttle, thus
enabling proton release from the positively charged amino
Figure 2. Superposition of PylD complex structures. The active site is
shown for PylD:1 as a surface representation (gray). Hydrophobic
amino acid side chains of PylD are colored in light green. The N-
terminus of the protein is depicted as a transparent coil. a) Product
structures 0–3. Solvent molecules in contact with the heterocyclic
nitrogen are shown as spheres; interactions are indicated by dashed
lines (distances in ꢂ). b) Substrate structures 0a and 4a–6a. Stereo
views are presented in the Supporting Information (Figures S7 and S8,
respectively).
nitrogen. In particular, a single small molecule, H O or
2
NH , interacting with the pyrroline or pyrrolidine nitro-
3
gen, was identified only in the product structures. This
solvent molecule is in contact with the water cluster and
therefore might initiate a proton transfer cascade, trigger-
ing the formation and release of 1.
Scheme 3. Schematic representation of crucial substrate interactions and the reaction trajectory of PylD catalysis. Blue spheres represent fixed
water molecules in proximity to the N-terminal substrate or product part. The purple sphere symbolizes a protonated water molecule.
8152
www.angewandte.org
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2014, 53, 8150 –8153

Angewandte
Chemie
The turnover rates and substrate affinities for 1a reveal
e) M. A. Gaston, R. Jiang, J. A. Krzycki, Curr. Opin. Microbiol.
011, 14, 342 – 349.
[2] D. G. Longstaff, R. C. Larue, J. E. Faust, A. Mahapatra, L. Zhang,
2
that PylD only possesses moderate activity (K = 1.6 mm
m
À1
À1
Æ 0.18 mm, k = 1.72 s Æ 0.07 s , Table S1). These findings
cat
K. B. Green-Church, J. A. Krzycki, Proc. Natl. Acad. Sci. USA
are in good agreement with the proposed mechanism, since
the enzyme lacks activating residues in the active site as well
as elaborate specificity for the coordination of the N-terminal
isopeptide moiety. Furthermore, PylD is subject to major
conformational rearrangements, which so far is a singular case
in the large family of dehydrogenases.
2007, 104, 1021 – 1026.
[
3] a) M. A. Gaston, L. Zhang, K. B. Green-Church, J. A. Krzycki,
Nature 2011, 471, 647 – 650; b) S. E. Cellitti, W. Ou, H. P. Chiu, J.
Grunewald, D. H. Jones, X. Hao, Q. Fan, L. L. Quinn, K. Ng, A. T.
Anfora, S. A. Lesley, T. Uno, A. Brock, B. H. Geierstanger, Nat.
Chem. Biol. 2011, 7, 528 – 530; c) F. Quitterer, A. List, W.
Eisenreich, A. Bacher, M. Groll, Angew. Chem. 2012, 124,
Nowadays, unnatural amino acids are indispensable in
1
367 – 1370; Angew. Chem. Int. Ed. 2012, 51, 1339 – 1342; d) F.
Quitterer, A. List, P. Beck, A. Bacher, M. Groll, J. Mol. Biol. 2012,
24, 270 – 282; e) F. Quitterer, P. Beck, A. Bacher, M. Groll,
[7]
protein engineering as well as high-throughput technologies.
We could show that primary amines with different stereo-
chemical properties are well tolerated in the active site cavity
of PylD. Thus, in vivo incorporation of pyrrolysine analogues
into defined target proteins is also possible by utilizing the
PylDST machinery, hereby obviating cost- and time-intensive
chemical synthesis.
4
Angew. Chem. 2013, 125, 7171 – 7175; Angew. Chem. Int. Ed.
2013, 52, 7033 – 7037; f) J. A. Krzycki, Curr. Opin. Chem. Biol.
2013, 17, 619 – 625.
[
4] J. A. Soares, L. Zhang, R. L. Pitsch, N. M. Kleinholz, R. B. Jones,
J. J. Wolff, J. Amster, K. B. Green-Church, J. A. Krzycki, J. Biol.
Chem. 2005, 280, 36962 – 36969.
[
5] a) S. A. Burke, J. A. Krzycki, J. Biol. Chem. 1997, 272, 16570 –
Received: February 19, 2014
Published online: June 10, 2014
16577; b) D. J. Ferguson, N. Gorlatova, D. A. Grahame, J. A.
Krzycki, J. Biol. Chem. 2000, 275, 29053 – 29060; c) D. J. Ferguson,
J. A. Krzycki, J. Bacteriol. 1997, 179, 846 – 852.
Keywords: amino acids · dehydrogenases · enzyme catalysis ·
[6] J. Strꢂnskꢂ, D. Kope cˇ n y´ , M. Tylichovꢂ, J. Snꢃgaroff, M. Sˇ ebela,
.
protein crystallography · pyrrolysine
Plant Signaling Behav. 2008, 3, 929 – 935.
[
7] a) B. H. Geierstanger, W. Ou, S. Cellitti, T. Uno, T. Crossgrove,
H. P. Chiu, J. Grꢄnewald, X. Hao, International patent applica-
tion WO/2010/048582, 2010; b) S. Lepthien, L. Merkel, N. Budisa,
Angew. Chem. 2010, 122, 5576 – 5581; Angew. Chem. Int. Ed. 2010,
[
1] a) G. Srinivasan, C. M. James, J. A. Krzycki, Science 2002, 296,
459 – 1462; b) B. Hao, W. Gong, T. K. Ferguson, C. M. James,
1
49, 5446 – 5450; c) E. Kaya, M. Vrabel, C. Deiml, S. Prill, V. S.
J. A. Krzycki, M. K. Chan, Science 2002, 296, 1462 – 1466; c) S. K.
Blight, R. C. Larue, A. Mahapatra, D. G. Longstaff, E. Chang, G.
Zhao, P. T. Kang, K. B. Green-Church, M. K. Chan, J. A. Krzycki,
Nature 2004, 431, 333 – 335; d) C. Hertweck, Angew. Chem. 2011,
Fluxa, T. Carell, Angew. Chem. 2012, 124, 4542 – 4545; Angew.
Chem. Int. Ed. 2012, 51, 4466 – 4469; d) C. H. Kim, M. Kang, H. J.
Kim, A. Chatterjee, P. G. Schultz, Angew. Chem. 2012, 124, 7358 –
7361; Angew. Chem. Int. Ed. 2012, 51, 7246 – 7249.
123, 9712 – 9714; Angew. Chem. Int. Ed. 2011, 50, 9540 – 9541;
Angew. Chem. Int. Ed. 2014, 53, 8150 –8153
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
8153