Kinetics of reactions of the Actinomadura R39 DD -Peptidase with specific substrates

Article abstract of DOI:10.1021/bi101760p

The Actinomadura R39 dd-peptidase catalyzes the hydrolysis and aminolysis of a number of small peptides and depsipeptides. Details of its substrate specificity and the nature of its in vivo substrate are not, however, well understood. This paper describes the interactions of the R39 enzyme with two peptidoglycan-mimetic substrates 3-(d-cysteinyl)propanoyl-d-alanyl-d-alanine and 3-(d-cysteinyl)propanoyl-d-alanyl-d-thiolactate. A detailed study of the reactions of the former substrate, catalyzed by the enzyme, showed dd-carboxypeptidase, dd-transpeptidase, and dd-endopeptidase activities. These results confirm the specificity of the enzyme for a free d-amino acid at the N-terminus of good substrates and indicated a preference for extended d-amino acid leaving groups. The latter was supported by determination of the structural specificity of amine nucleophiles for the acyl-enzyme generated by reaction of the enzyme with the thiolactate substrate. It was concluded that a specific substrate for this enzyme, and possibly the in vivo substrate, may consist of a partly cross-linked peptidoglycan polymer where a free side chain N-terminal un-cross-linked amino acid serves as the specific acyl group in an endopeptidase reaction. The enzyme is most likely a dd-endopeptidase in vivo. pH-rate profiles for reactions of the enzyme with peptides, the thiolactate named above, and β-lactams indicated the presence of complex proton dissociation pathways with sticky substrates and/or protons. The local structure of the active site may differ significantly for reactions of peptides and β-lactams. Solvent kinetic deuterium isotope effects indicate the presence of classical general acid/base catalysis in both acylation and deacylation; there is no evidence of the low fractionation factor active site hydrogen found previously in class A and C δ-lactamases.

Full text of DOI:10.1021/bi101760p

376 Biochemistry 2011, 50, 376–387
DOI: 10.1021/bi101760p
Kinetics of Reactions of the Actinomadura R39 DD-Peptidase with
Specific Substrates†
S. A. Adediran,‡ Ish Kumar,‡, Rajesh Nagarajan,‡,^ Eric Sauvage,§ and R. F. Pratt*,‡
‡Department of Chemistry, Wesleyan University, Lawn Avenue, Middletown, Connecticut 06459, United States, and
§
ꢀ
ꢀ
ꢁ
ꢁ
Centre d’Ingeniere des Proteines, Universite de Liege, B-4000 Sart Tilman, Liege, Belgium. Present address: School of
Natural Sciences, Fairleigh Dickinson University, Teaneck, NJ 07666. ^Present address: Department of
Chemistry and Biochemistry, Boise State University, 1910 University Drive, Boise, ID 83725-1520
Received November 2, 2010; Revised Manuscript Received December 6, 2010
ABSTRACT: The Actinomadura R39 DD-peptidase catalyzes the hydrolysis and aminolysis of a number of
small peptides and depsipeptides. Details of its substrate specificity and the nature of its in vivo substrate
are not, however, well understood. This paper describes the interactions of the R39 enzyme with two
peptidoglycan-mimetic substrates 3-(
panoyl- -alanyl- -thiolactate. A detailed study of the reactions of the former substrate, catalyzed by the
enzyme, showed DD-carboxypeptidase, DD-transpeptidase, and DD-endopeptidase activities. These results
confirm the specificity of the enzyme for a free -amino acid at the N-terminus of good substrates and
indicated a preference for extended -amino acid leaving groups. The latter was supported by
D
-cysteinyl)propanoyl-
D
-alanyl-
D
-alanine and 3-(
D
-cysteinyl)pro-
D
D
D
D
determination of the structural specificity of amine nucleophiles for the acyl-enzyme generated by
reaction of the enzyme with the thiolactate substrate. It was concluded that a specific substrate for this
enzyme, and possibly the in vivo substrate, may consist of a partly cross-linked peptidoglycan polymer
where a free side chain N-terminal un-cross-linked amino acid serves as the specific acyl group in an
endopeptidase reaction. The enzyme is most likely a DD-endopeptidase in vivo. pH-rate profiles for
reactions of the enzyme with peptides, the thiolactate named above, and β-lactams indicated the presence
of complex proton dissociation pathways with sticky substrates and/or protons. The local structure of the
active site may differ significantly for reactions of peptides and β-lactams. Solvent kinetic deuterium
isotope effects indicate the presence of classical general acid/base catalysis in both acylation and
deacylation; there is no evidence of the low fractionation factor active site hydrogen found previously
in class A and C β-lactamases.
The DD-peptidases represent a family of enzymes that
catalyze the final steps in bacterial cell wall biosynthesis.
Their substrates are peptide elements of peptidoglycan.
They catalyze nucleophilic double displacement reactions
at N-acyl-DD-dipeptide moieties leading, via formation of an
acyl-enzyme intermediate, to carboxypeptidase, endopepti-
dase, and transpeptidase reactions (Scheme 1).
The latter of these reactions leads to the cross-linking of
peptidoglycan to form the final mature polymer while the former
two are believed to be important in control of the degree of cross-
linking (1, 2). These reactions are clearly important to bacteria,
but they are also important to people since the enzymes catalyz-
ing them, being unique to bacteria, represent excellent targets for
antibacterial drugs. The best known of the latter are of course
the β-lactams, which inhibit DD-peptidases by formation of stable
acyl-enzymes (3, 4). The increasing resistance displayed by
bacteria to all antibiotics, including β-lactams, encourages
further research on the relevant targets (5, 6).
The DD-peptidases have been classified into groups based on
molecular mass and amino acid homology (7, 8). One important
division is between the high molecular mass (HMM)1 enzymes,
usually 60-90 kDa, and low, of molecular mass less than 50 kDa
(LMM). The former enzymes are believed to act as transpepti-
dases in vivo while the latter are carboxypeptidases and
endopeptidases (1, 2). The HMM group can be subdivided
into classes A and B and the LMM group into classes A, B,
and C, again based on amino acid sequence homology (7, 8).
Curiously, the former enzymes, in vitro, do not catalyze, at any
significant rate, the cleavage of small, peptidoglycan-mimetic
peptides (9). The reasons for this are not known. The
LMM group generally does catalyze the hydrolysis and
aminolysis of such peptides and, in certain cases at least (see
†This research was supported by National Institutes of Health Grant
AI-17986 (R.F.P.) and in part by the Belgian Program on Interuniver-
sity Poles of Attraction initiated by the Belgian State, Prime Minister’s
Office, Science Policy programming (IAP no. P6/19), the Fonds de la
Recherche Scientifique (IISN4.4505.00, IISN4.4509.09, FRFC2.4.-
508.01.F, FRFC9.4.538.03.F, FRFC2.4.524.03), and the University of
1Abbreviations: AMPSO, 3-[(1,1-dimethyl-2-hydroxyethyl)amino]-2-
hydroxypropanesulfonic acid; CAPS, 3-(cyclohexylamino)-1-propane-
sulfonic acid; DMSO, dimethyl sulfoxide; ESMS, electrospray mass
spectrometry; HMM, high molecular mass; LMM, low molecular mass;
MES, 2-(N-morpholino)ethanesulfonic acid; MOPS, 3-(N-morpholi-
no)propanesulfonic acid; NMR, nuclear magnetic resonance; TAPS,
3-[[tris(hydroxymethyl)methyl]amino]-1-propanesulfonic acid; TFA,
trifluoroacetic acid; THF, tetrahydrofuran.
ꢁ
ꢀ
ꢀ
Liege (Fonds speciaux, Credit classique, 2009).
*Corresponding author: telephone, 860-685-2629; e-mail, rpratt@
wesleyan.edu; fax, 860-685-2211.
r
pubs.acs.org/biochemistry
Published on Web 12/23/2010
2010 American Chemical Society

Article
Biochemistry, Vol. 50, No. 3, 2011 377
Scheme 1
efficiency, specificity, and mechanism of action of the enzyme
emerges.
below), shows dramatic specificity for elements of peptidoglycan
structure.
MATERIALS AND METHODS
The R39 DD-peptidase was expressed and purified as described
previously (19). The following compounds were purchased
One very well studied example of these enzymes, and one
showing pronounced substrate specificity, is the class B
LMM DD-peptidase of Streptomyces R61. The stem peptide
of this organism has the structure 1. It has been demon-
strated that the peptide 2 is a very specific substrate of
this enzyme (kcat/Km = 8.7 ꢀ 106 s-1 M-1) (10). On the
commercially and used as received: N-R-Ac-Lys-
N-ε-Ac- -Lys, N-ε-Ac-
-Lys, and 4,40-biphenyl-
(ChemImpex), N-R-Ac- -Lys- -Ala- -Lys(Ac) (New England
Peptide), -alanine and -alanine (Fluka), -lactic acid
(Bachem), (R)-2-amino-1-propanol ( -alaninol) (Acros), and
-Lys[Ac-Ala( -iGln)] (NeoMPS). -R-Phenylglycine, -homo-
phenylalanine, -tyrosine, -leucine, -phenylalanine, -pheny-
lactic acid, -norleucine, and 2,6-diaminopimelic acid were all
D-Ala-D-Ala,
D
L
D
-alanine
L
D
D
D
L
D
D
other hand, the peptide 3 is quite a poor substrate (kcat/Km
=
270 s-1 M-1) (11). This enzyme, therefore, appears to
specifically recognize the free side chain N-terminus of the
stem peptide in choosing a substrate. The structural basis for
this specificity has been clearly demonstrated by crystal
structures of the R61 DD-peptidase in complexes with a
peptide, a phosphonate inhibitor, and a β-lactam, each
bearing the specific side chain of 2 (12-14). In another
D
D
D
D
D
D
D
D
D
supplied by Sigma-Aldrich. Cefotaxime (16) and 7-β-aminoce-
phalosporanic acid (17) were gifts from Merck and Eli Lilly and
Co., respectively. 6-R-
was available from previous studies in this laboratory (11). 3-(
Cysteinyl)-propanoyl- -alanyl- -alanine (6) and 3-( -cysteinyl)-
propanoyl- -alanyl- -thiolactate (7) were prepared using a gen-
D-Aminopimelyl-ε-D-alanyl-D-alanine (4)
D-
example, 4 is an excellent substrate (kcat/Km = 5.6 ꢀ 106 s-1 M-1
)
D
D
D
D
D
of the DD-peptidase of Actinomadura R39, a LMM class C enzyme,
eral method described previously (21); details of these syntheses
are provided in Supporting Information.
again mimicking the free side chain N-terminus of the relevant
stem peptide 5, while 3 is a relatively poor substrate (kcat
/
Km = 2.9 ꢀ 103 s-1 M-1) (11). Here also the structural basis
of the specificity has been demonstrated crystallographi-
cally (15, 16).
Kinetic Methods. All kinetic measurements were carried out
in 20 mM MOPS buffer at pH 7.5 and 25 ꢀC, unless otherwise
noted. A mixed buffer was employed for the pH profiles and
certain other experiments. This buffer contained 20 mM each of
acetate, MES, MOPS, TAPS, AMPSO, and CAPS, and ionic
strength was maintained at 1.0 M with sodium chloride. Absorp-
tion spectra and spectrophotometric reaction rates and progress
curves were obtained by means of a Hewlett-Packard 8452A
spectrophotometer. The enzyme-catalyzed reactions were mon-
itored spectrophotometrically at 230 nm (Δε = 57 cm-1 M-1
)
(4), 230 nm (Δε = 100 cm-1 M-1) (6), 240 nm (Δε = 4980 cm-1
M-1) (7), 230 nm (Δε = 76 cm-1 M-1) (11), and 230 nm (Δε =
92 cm-1 M-1) (15). Steady-state kinetic parameters for hydro-
lysis of the above substrates by the R39 DD-peptidase were
obtained from measurements of initial rates as a function of
substrate concentrations. These were fitted to the Michae-
lis-Menten-Henri equation by a nonlinear least-squares proce-
dure (Kaleidograph).
1
ꢁ
Although the Liege group has studied various aspects of the
The reactions of 4 and 6 were also monitored by H NMR
R39 enzyme quite extensively over many years (17-20), a
systematic study of its reaction with specific substrates has not
been made. This is important since the active site of the R39
enzyme has the SerXAsn motif rather than the more unusual
(confined to LMM class B enzymes) TyrXAsn combination of
the more intensively studied R61 enzyme. We present here kinetic
studies of 6 and 7, thia analogues of 4, reacting with the R39
enzyme, with emphasis on the transpeptidase and endopeptidase
reactions. We also present pH-rate profiles for the enzyme
catalysis and the results of solvent kinetic isotope effect measure-
ments. From these experiments, a clearer picture of the catalytic
spectroscopy. Reaction mixtures, in D2O, contained the substrate
(6.3 mM), enzyme (30 nM), and buffer (50 mM sodium
bicarbonate). Spectra were recorded immediately after addition
of enzyme and at appropriate times thereafter.
The rates of formation and decay of intermediates in the
reactions of 6 with the R39 DD-peptidase were quantitatively
monitored by a HPLC method. Reaction mixtures of 6 (300 μL,
2.0 mM) and enzyme (0.25 μM) in 20 mM mixed buffer, pH 7.50,
were prepared and incubated at 25 ꢀC. At appropriate times,
aliquots (20 μL) were withdrawn and injected into a Varian
Prostar 210 HPLC system equipped with a Varian 340 UV/vis

378 Biochemistry, Vol. 50, No. 3, 2011
Adediran et al.
detector. Separation was achieved on a Macherey-Nagel Nucleo-
sil 5 C-18 column with a mobile phase of 1% (v/v) acetoni-
trile/water, also containing 0.05% TFA. A flow rate of 0.9
mL/min was employed, and the effluent was monitored at
215 nm. Under these conditions, retention times of the
substrate 6, the hydrolysis product 8, and the two inter-
mediates, S2 and S3, were 5.8, 4.9, 26.7, and 36.7 min,
respectively. The fractions containing S2 and S3 were sub-
jected to electrospray mass spectroscopy (Thermofinnigan
LCQ Advantage). This experiment was repeated with 4
where retention times of the substrate and hydrolysis
product were 7.1 and 5.9 min, respectively.
Kinetics of turnover of 6 were studied in a similar experiment.
In this case, 20 μL aliquots of the reaction mixture were taken at
appropriate times. In each aliquot, the reaction was quenched by
its addition to 20 μL of an aqueous solution containing 1%
acetonitrile, 0.05% TFA, and 0.4 mM
was then frozen immediately and later analyzed by HPLC as
described above. The -phenylalanine (retention time 13.5
D-phenylalanine. Each
D
min) was added as a concentration reference. The integrated
HPLC peak areas as a function of time were taken as
proportional to the concentrations of reactants, products,
and intermediates as a function of time and fitted to appro-
priate reaction schemes by means of the KinTek version
2.0 (22) and Dynafit version 3.28 (23) programs. The former
was particularly useful since it allowed the fixing of rate
constant ratios. The two programs did, however, when
comparably restrained, produce essentially the same fit with
the data from the above experiments.
In order to obtain rate data directly for the intermediates S2
and S3, freeze-dried HPLC fractions of these species, isolated
as described above, were dissolved in 350 μL of the mixed
buffer, and enzyme, to final concentrations of 5.0 and 61 nM,
respectively, was added to 100 μL portions. The subsequent
reaction of each species was then monitored by HPLC, as
described above, and the resulting progress curves were
analyzed by means of the Dynafit program (23). In the case
of S2, the absolute concentration of the peptide was obtained
by analysis of D-alanine (24) in the final reaction mixture after
complete reaction.
The pH profiles of kcat/Km for 7, 11, and 15 were obtained
directly from spectrophotometric experiments run under [S]
, Km conditions as a function of pH in the mixed buffer.
Reactions of the R39 DD-peptidase with the cephalosporins
16 and 17 were studied fluorometrically (λex = 280 nm, λf =
340 nm) (25). The reaction mixture, in mixed buffer, con-
tained 16 (5.0 μM) or 17 (30 μM) and enzyme (0.57 μM).
For both the peptides and cephalosporins, time traces
were fitted to an exponential function to obtain pseudo-
first-order rate constants that were converted to second-
order constants (kcat/Km or k2/K1) by dividing them by the
relevant peptide/cephalosporin concentration. The pH de-
pendences of these parameters were fitted to eq 1 for a double
dissociation model to obtain pKa values. In this equation, k2
represents kcat/Km when all enzyme is present in the mono-
protonated form.
F
IGURE 1: Total progress curves for reaction of 7 (100 μM) (top), 6
(2.0 mM) (middle), and 4 (1.0 mM) (bottom) with the R39 DD-
peptidase (0.24 μM). Data are shown at 240, 230, and 230 nm,
respectively.
log kcat ¼ log½ðk1h þ k2KaÞ=ðh þ KaÞꢁ
ð2Þ
Solvent Kinetic Isotope Effects. The solvent deuterium
kinetic isotope effects DV/K and DV for hydrolysis of 7 and 15
by the R39 DD-peptidase were determined from spectrophoto-
metric rates in 20 mM MOPS buffer, pH/pD 7.5, as previously
described (26). Substrate concentrations of ca. 0.1 Km and 5 Km
were used for DV/K and DV determinations, respectively. Appro-
priate corrections for the degree of saturation were made in each
case (27). A common concentrated enzyme stock solution in 1/1
(v/v) H2O/D2O MOPS buffer, pH meter reading 7.3, was used.
Substrate solutions were separately prepared in H2O or D2O.
k
cat=Km ¼ k2Ka1h=ðh2 þ Ka1h þ Ka1Ka2
Þ
ð1Þ
The pH dependence of kcat for 7 was fitted to eq 2, where k1
represents a low pH plateau.

Article
Biochemistry, Vol. 50, No. 3, 2011 379
FIGURE 2: HPLC traces of samples taken at 3 min (A) and 7 min (B) from the reaction mixture of 6 (2 mM) and the R39 DD-peptidase (0.25 μM).
The signals are proportional to the absorbance at 215 nm. D-Phenylalanine (0.4 mM) was added as a concentration reference. Retention times are
noted above the peaks.
Aminolysis Kinetics. Spectrophotometric initial rate mea-
surements were also used to obtain kinetic parameters for
aminolysis of 7 (50 μM) by the R39 DD-peptidase (40 nM) in
MOPS buffer at pH 7.5 (29). These data were then analyzed in
terms of Scheme 7.
reaction between 6 and the R39 DD-peptidase. Figure 1C shows
a comparable trace for 4. The reaction of 6 appears to contain
two separate phases, the faster involving an increase in absor-
bance at 230 nm and the slower the anticipated (Figure 1C)
decrease accompanying peptide hydrolysis. Experiment showed
that both phases were enzyme-catalyzed, and thus the first phase
was not an artifact.
RESULTS AND DISCUSSION
In order to understand the complex kinetics of reaction of 6,
the composition of a reaction mixture (initial concentrations of
enzyme and 6 were 0.25 μM and 2.0 mM, respectively) was
monitored as a function of time by HPLC, as described in the
Materials and Methods section. Panels A and B of Figure 2 show
absorbance chromatograms at 3.0 and 7.0 min, respectively. The
chromatograms show 6 (marked as S1) and the expected hydro-
lysis product 8 (marked as P) but also what are apparently
intermediate species S2 and S3 that form and decay as a function
of time. Figure 3A shows the time dependence of concentrations
of 6, S2, S3, and 8. 1H NMR spectra of a reaction mixture (not
shown) also indicated the presence of at least one accumulating
intermediate. Upon completion of all reactions, the NMR
spectrum was consistent with that of a mixture of 8 and
Peptide 4 is an excellent substrate of the R39 DD-peptidase,
most likely because of its resemblance to a specific element of
the stem peptide 5, as described in the introduction. The
analogues 6 and 7 were prepared to facilitate synthesis and
kinetics studies. Thia derivatives of 2 have previously been
used in studies of the R61 DD-peptidase (21, 31). The reac-
tions of D-thiolactates are more readily monitored spectro-
photometrically than those of peptides, and the leaving group
ability of the thiolate often ensures that deacylation is rate-
determining (29-31).
Spectrophotometric initial rates of turnover of 7, like those of
4 (11), were readily obtained (Figure 1A). Interesting complica-
tions arose immediately, however, on investigation of 6.
Figure 1B shows a typical spectrophotometric trace of the

380 Biochemistry, Vol. 50, No. 3, 2011
Adediran et al.
F
IGURE 3: (A) Variation of concentration (the signal is proportional
to the absorbance at 215 nm) of S1 (O), S2 (0), S3 (]), and P (4) as a
function of time on reaction of 6 (2 mM) in the presence of the R39
DD-peptidase (0.25 μM). (B) The same reaction followed directly
spectrophotometrically at 230 nm. Concentrations of 6 were 1.0 mM
(O), 1.5 mM (0), 2.0 mM (]), 2.5 mM (3), 3.0 mM (ꢀ), and 4.0 mM
(4), and the enzyme concentration was 0.13 μM. The points are
experimental, and the lines represent a simultaneous fit of all data to
Scheme 6 (see text).
F
IGURE 4: (A) Variation of the concentration (the signal is propor-
tional to the absorbance at 215 nm) of S2 (O), S1 (0), and P (4) as a
functionoftimeon reactionofS2 (0.37mM) in the presence oftheR39
DD-peptidase (5.0 nM). (B) Disappearance of S3 in the presence of the
R39 DD-peptidase (61 nM). The points are experimental, and the lines
represent a fit of the data to Schemes 4 and 5, respectively (see text).
Scheme 2
more quantitative data for the first phase of reaction (see
above).
D-alanine. At short reaction times, the S2 species was observed to
arise first, followed by S3; the simplest interpretation, therefore,
would suggest that S3 is a product of reaction of S2. The
hydrolysis product of 6 (viz., 8), however, appears to increase
in concentration directly from time zero and thus cannot solely
be a product of S3. In the NMR experiment, D-alanine was also
In a preparative experiment, samples of S2 and S3, formed on
reaction of 6 as described above, were isolated by HPLC and
subjected to ES(þ) mass spectral analysis. The molecular masses
observed to increase directly from time zero. Figure 3B shows
the results of spectrophotometric experiments designed to get

Article
Biochemistry, Vol. 50, No. 3, 2011 381
Scheme 3
Table 2: Steady-State Rate Parameters for Hydrolysis of Substrates by the
R39 DD-Peptidasea
substrate
kcat (s-1
)
Km (μM)
kcat/Km (s-1 M-1
)
4
6
5.2
4.9
1.1 ꢀ 106
7.3 ꢀ 104
3.0 ꢀ 105 b,c
8.0 ꢀ 105
1.05 ꢀ 106
4.9 ꢀ 104
4.5 ꢀ 104
1.9 ꢀ 105
4.3 ( 0.2 (9.5 ( 0.5)b
59 ( 5
7
3.0 ( 0.1
9.5
3.8 ( 0.4
9.5
9b,c
10b,c
11
9.5
190
11.8 ( 0.4
26.5 ( 0.8
260 ( 40
140 ( 30
Scheme 4
15
aAll represent hydrolysis of terminal
D-alanine or D-thiolactate except
for 9 and 10 (see text). bDetermined in mixed buffer. cCalculated from
Table 1.
4) indicate that S3 proceeds directly to the final hydrolysis
product P (8), as would be expected; no S2 was observed. On
the other hand, reaction of S2 appeared to produce, first, S1 and
then P; no S3 was observed on reaction of S2, despite expectations
based on the simplest interpretation of Figure 3A. This phenom-
enon cannot arise because S3 reacts faster than S2 and thus is
unobserved in Figure 4B, because this is contrary to the
observations of Figure 3A. The conclusion must be that S3 is
not a product of reaction of S2, at least not at a rate observable
under the conditions of these experiments. These observations
therefore cannot be accommodated by Scheme 2 but can be, most
simply, by Scheme 3.
Scheme 5
Scheme 6
Scheme 3 shows that the R39 DD-peptidase is able to
catalyze the carboxypeptidation (C) reaction of S1 and the
transpeptidation reaction (T) of S1 with either itself or P as
nucleophile. The other notable feature is that the enzyme
prefers to act as a transpeptidase or endopeptidase (E) in its
reaction with S2 rather than as a carboxypeptidase; this may
be relevant to its role in vivo (see below).
To proceed further, the data of Figure 4 were quantitatively
fitted to Schemes 4 and 5, respectively, where the endopeptidase
(E) reactions of both S2 and S3 are noted.
With values of kcat/Km for S2 and S3 in hand from these fits,
the full Scheme 6 was then applied to the data of Figure 3 with
the restraints listed in Table 1. In this scheme, E.S1, E.S2,
and E.S3 represent noncovalent complexes of the enzyme
with S1, S2, and S3, respectively, and ES1 is the R-D-amino-
pimelyl acyl-enzyme.
This procedure yielded the fitted lines shown in Figure 3 and
the rate constant data of Table 1. It might be noted in passing
here that omission of specific E.S2 and E.S3 complexes from
Scheme 6 led to a noticeably poorer fit.
Table 1: Steady-State Kinetic Parameters for Reaction of the R39 DD
-
Peptidase with Peptide 6a
parameter
value
provenance
k1 (s-1 M-1
k2/k-1
)
2.0 ꢀ 108
assumed, fixed
1.5 ꢀ 10-3
500
experiment,b fixed
experiment, fixed
experiment, fixed
fitted
k-2 (s-1 M-1
)
k3 (s-1
)
9.5
The first important results coming from the above quantitative
treatment are the calculated kcat/Km values for 6 [k1k2/(k-1 þ k2)],
9 [k-4k-5/(k-4 þ k5)] and 10 [k-6k-7/(k-6 þ k7)] in the reactions
k4 (s-1 M-1
k-4/k5
)
(1.4 ( 0.1) ꢀ 104
5.0 ꢀ 10-3
2.0 ꢀ 108
(3.0 ( 0.1) ꢀ 103
2.5 ꢀ 10-4
2.0 ꢀ 108
experiment,b fixed
assumed, fixed
fitted
experiment,b fixed
assumed, fixed
k-5
shown in Scheme 6, viz., 3.0 ꢀ 105 s-1 M-1, 1.0 ꢀ 106 s-1 M-1
,
k6 (s-1 M-1
k-6/k7
)
and 5.0 ꢀ 104 s-1 M-1, respectively. This shows, as noted above
k-7 (s-1 M-1
)
from qualitative assessment of the results, that the R39 DD
-
peptidase is an effective endopeptidase as well as a carboxypep-
tidase. In particular, the extended peptide 9 is a much better
substrate (in the endopeptidase reaction) than 6 (in the carbox-
ypeptidase reaction) or, as noted above, 9 itself (in the carbox-
ypeptidase reaction). Similarly, k4 > k6 > k-2, showing the
aDetermined from experiments in mixed buffer. bCalculated from
kcat/Km values for S1 (6), S2 (9), and S3 (10).
determined, 582.2 and 511.2 respectively, indicated that S2 and S3
are 9 and 10, respectively. Thus, Scheme 2 seemed reasonable,
with 9 and 10 arising from transpeptidation reactions of 6 with
itself and with 8, respectively. In separate experiments, enzyme
was added to solutions of isolated S2 and S3, and the subsequent
reaction in each case was followed by HPLC. The results (Figure
greater effectiveness of extended nucleophiles (S1 > P >
in the transpeptidase reaction of ES1.
D-Ala)
As noted above, the kinetic complexity that led to the
discovery of Scheme 6 was not observed, superficially at least,

382 Biochemistry, Vol. 50, No. 3, 2011
Adediran et al.
Scheme 7
Table 3: Kinetic Parameters (Scheme 7) for Aminolysis of 7 by the R39 DD
-
Peptidase
entry
acyl acceptor
k4 (s-1 M-1
)
Ki (mM)
1
D
-alanine
(5 ( 3) ꢀ 102
no (300)a
9.9 ( 0.4
23 ( 3
70
2
L-alanine
no (300)b
no (200)b
100
3
D
D
D
D
D
D
-alaninol
4
-lactate
5
-phenylalanine
-tyrosine
(7.7 ( 1.6) ꢀ 103
(2.0 ( 0.3) ꢀ 103
(5.4 ( 3.6) ꢀ 103
(1.7 ( 0.3) ꢀ 103
(9.3 ( 1.7) ꢀ 103
(1.5 ( 1.0) ꢀ 103
(7.1 ( 2.5) ꢀ 103
no (300)a
no (25)b
no (4.7)b
9.5
no (9.4)b
no (0.4)b
43
no (94)b
no (300)b
4.5
6
7
-R-phenylglycine
-homophenylalanine
8
9
4-biphenyl-
D-alanine
10
11
12
13
14
15c
16
D
-norleucine
N-ε-acetyl-
N-ε-acetyl-
D-lysine
L-lysine
2,6-meso-diaminopimelic acid
-Lys(Ac- -Ala- -iGln)
(1.0 ( 0.2) ꢀ 103
(1.8 ( 0.5) ꢀ 103
(1.4 ( 0.1) ꢀ 104c
(3.0 ( 0.1) ꢀ 103c
D
L
D
2.3
S1 (6)
P1 (8)
no (2.0)b
0.48
ano, no rate acceleration observed at the concentrations (mM) employed
(in parentheses). bno, no inhibition observed at the highest concentration
employed (in parentheses). cFrom Table 1.
Steady-state rate parameters for the hydrolysis of 4, 6, 7, 9, 10,
and 11, a generic, nonspecific peptide substrate for comparison,
are presented in Table 2.
It is noticeable, first, that 6 is some 15 times poorer as a substrate
(kcat/Km) than 4. This difference is very similar in magnitude to the
effect of the analogous structural change in substrates of the R61
DD-peptidase and is probably due to the effect of thia substitution
on the hydrophobic effect (21). The specificity of 4 as a substrate is
best seen in comparison with 11, where the effect of the pimelyl
carboxylate is evident. The thiolactate 7 is some 10 times more
reactive than 6, reflecting the better leaving group, which would be
expressed in the acylation rate and thus in kcat/Km (=k2/Ks,
F
IGURE 5: Effect of D-phenylalanine (A) and D-alanine (B) on turn-
over of 7 (50 μM), catalyzed by the R39 DD-peptidase (13.3 and
40 nM, respectively). The points are experimental, and the lines
represent the fit of the data to Schemes 7 and 8 (see text).
in the reactions of 4 and 7 under the same solution conditions.
In the latter case, this is most likely because the reactions of 7
were studied at lower concentrations (e100 μM) because
of the low Km value of this substrate (Table 2). The case of 4 is
more intriguing. HPLC and 1H NMR experiments, which
directly showed the presence of at least one intermediate
during the hydrolysis of 6, revealed a clean hydrolysis of 4
under the same conditions. Either 4 is a much poorer acyl
acceptor than 6, which seems unlikely, or the analogues
of 9 and 10 derived from 4 are considerably better sub-
strates of the enzyme than 4 itself and thus do not accu-
mulate in solution as 4 reacts. The latter explanation is
supported by the observation of more complex kinetics of
reaction of 4 at pH 9.0, analogous to those of 6 illustrated in
Figure 1B. This also has interesting implications for the
nature of the in vivo substrate of the R39 DD-peptidase (see
below).
Scheme 7). A similar effect is observed in the analogous R61 DD
-
peptidase substrates (29).
Initial rates of hydrolysis of 7 increased markedly on addition
of
D-amino acids. Figure 5A, for example, shows the effect of
D-phenylalanine. This result is typical of rate-determining dea-
cylation (k2 > k3[H2O], Scheme 7), where the acyl-enzyme can be
attacked by both water and an amine nucleophile and where
addition of the latter would be expected to increase the observed
steady-state rate (28, 29). The increase in rate slows at high
nucleophile concentration (Figure 5A) as k4[R0NH2] þ k3[H2O]
approaches k2 and acylation takes over as the rate-determining
step. The fit of the data of Figure 5A to Scheme 7 led to a k4 value
of (7.7 ( 1.6) ꢀ 103 s-1 M-1 for
D-phenylalanine. Thus, it seems
likely that during hydrolysis of 7 under conditions of saturation
of the R39 DD-peptidase, deacylation (k3) is rate-determining.
The close similarity between the kcat values of 6 and 7 suggests
that this is true for the peptide 6 as well.

Article
Biochemistry, Vol. 50, No. 3, 2011 383
Scheme 8
different acyl acceptor binding sites adjacent to the acyl
donor site in this enzyme. Some evidence for this idea has
been obtained (29). Certainly, both
D-amino acids and Gly-
L
-X dipeptides can play the role of acyl acceptor (29). This
The specificity of the amino acid acceptor R0NH2 is of interest
since, assuming the simple Scheme 7, it should also be an
indicator of leaving group specificity in the peptide substrate
RCONHR0 of the endopeptidase reaction. The experiment
may also be the case with respect to the R39 DD-peptidase
where there is also asymmetry, but here both leaving group
(D-alanine) and nucleophile (D-aminopimelyl) are D-amino
acid derivatives and thus, in principle, only one binding site
for the acceptor may be needed. Some further investigation of
described above for D-phenylalanine was repeated with a number
of potential amino acceptors with the results shown in Table 3.
Note that the issue of whether the bound species here is the amino
acid anion or the zwitterion is moot; the zwitterion would
certainly be the dominant species in solution at pH 7.5, but the
reactive and thus, perhaps, bound form may be the anion. It
should also be noted that with several acyl acceptors an initial
increase in rate at low concentration, as in Figure 5A, was
followed by a rate decrease at higher concentrations, as seen with
this issue would be of interest. Second, Gly-L-Lys(Ac), a good
analogue of a hypothetical natural R61 acceptor (see 1), was
a poor acyl acceptor in experiments with 14 and the R61
DD-peptidase [considerably less effective than
D-alanine and
Gly- -Ala, for example (29)]. This was taken to be an
L
indication that the R61 enzyme, in vivo, may largely be a
carboxypeptidase.
D
-alanine in Figure 5B, for example. This phenomenon was
interpreted as reflecting competitive inhibition by the -amino
acid (Scheme 8) and incorporated into Scheme 7 for data fitting.
It is certainly not unreasonable to find that a -amino acid could
compete with the -aminopimelyl terminus of 6 for the acyl
D
D
D
donor site. Similar inhibition was observed in studies of the R61
DD-peptidase (10, 29).
The first four entries of Table 3 affirm the importance of
D
-stereochemistry of the amino acid in an effective acceptor. The
The converse result in the present case could therefore be
interpreted to mean that the R39 DD-peptidase is a transpeptidase
rather than a carboxypeptidase. Alternatively, and probably
more likely in view of the above results, the R39 enzyme may
predominantly be an endopeptidase. In cases where the stem
peptide has the structure 5, this DD-endopeptidase reaction is of
course formally also a DD-carboxypeptidase reaction although it
differs in consequence since it leads to cleavage of the peptido-
next six demonstrate that incorporation of hydrophobic moieties
adjacent to the polar terminus is beneficial. Entries 11-13
represent analogues of what would be a natural acyl acceptor
in a transpeptidation reaction in vivo (Scheme 1). Compounds 12
and 13 are probably the closest analogues to a natural acceptor,
and it is seen that N-ε-acetyl-D-lysine 12 is significantly more
effective than the acceptor of entry 14 (13).
glycan cross-links rather than just clipping terminal
residues.
D-alanine
In view of the above results, the peptide 15 was designed as
a useful “endopeptidase” substrate (anticipated cleavage site
indicated). It contains the putative “natural” leaving group
analogue, N-ε-Ac-D-Lys, and has a simple ammonium ion at
the side chain N-terminus. The latter will probably decrease
its reactivity as an acyl donor (compare 4 and 11), but it will
also remove the ability of 15 to be an acyl acceptor (N-R-Ac-
L
-Lys is a very poor acceptor, Table 3). Peptide 15, in fact,
turned out to be a quite effective endopeptidase substrate of
the R39 DD-peptidase (Table 2) although not as reactive as 4.
1H NMR monitoring of the reaction of 15 with the R39
enzyme showed clean cleavage to N-R-Ac-
N-ε-Ac- -Lys, with no sign of intermediates. Studies with
-alanine as an alternative acceptor (29, 32) showed that
L-Lys-D-Ala and
As in the acyl donor (11), there seems little particular
D
affinity for the
L
-Ala-
D
-iGln moiety in an acyl acceptor (13).
D
The kinetics of turnover of 6, discussed above, show that 6 is
a better acceptor than the others of Table 3, including 8. The
final hydrolysis product 8 was, at high concentration, also an
inhibitor, with a Ki value of (0.48 ( 0.05) mM. This is the
strongest inhibition observed, but 8 has the side chain
N-terminus closest in structure to that of the good substrates
4 and 6.
It is interesting to compare the results above with those of
comparable experiments with the R61 DD-peptidase (28, 29, 32).
First, the transpeptidation reaction of the R61 enzyme is asym-
deacylation was rate-determining under active site saturation
conditions.
The data from the “transpeptidase” reactions of 7 thus show
the preference of the R39 DD-peptidase for an extended acyl
acceptor, as would be expected in both the transpeptidase and
endopeptidase reactions of a natural substrate.
pH-Rate Profiles. The kcat/Km and kcat pH-rate profiles
for hydrolysis of 7, catalyzed by the R39 DD-peptidase, are shown
in Figure 6. The former shows a bell-shaped profile with maximal
activity at around pH 7.4 and pKa values of 6.1 and 8.4. The kcat
metric in that, although the leaving group would be
D-alanine, as
profile is quite different with a plateau at low pH (kcat ≈ 5 s-1
)
usual, the incoming nucleophile would be an N-terminal glycyl-
and rising, apparently linearly (log kcat), with pH (slope 1) at
peptide (see 1). This suggests that there must be two rather
pH >7; a best fit of these data does not require an acid

384 Biochemistry, Vol. 50, No. 3, 2011
Adediran et al.
Scheme 9
from substrates, however, the acyl-enzymes derived from β-
lactams are inert to hydrolysis (25).
In a simple three-step reaction sequence (Scheme 9), the pH
dependence of kcat/Km should reflect the pKa(s) of free enzyme
and free substrate (33, 34). The enzyme-related values should,
therefore, be independent of the identity of the substrate. It is
noticeable from Table 4, however, that the pKa1 (and pKa2)
values do vary and well beyond experimental uncertainty. The
most likely explanation for this would involve the substrates
being “sticky” (35). If the acylation of the enzyme is represented
by Scheme 9, as in most simple cases, the observed pKa1 of kcat
/
app
Km is given by eq 3 (35); pKa1
will, therefore, be lower than
pKa1 and particularly noticeably so when kb > k-a. It is therefore
possible that the more specific substrates 7, 11, and 15 are stickier
than the β-lactam cefotaxime, and thus pKa1 g 7.1. Since kb is
likely to be larger for thioester 7 than the peptides 11 and 15, the
app
lower apparent pKa1 for 7 is reasonable. The raising of pKa2
for 15, 11, and 7 in a mirror image fashion can also be understood
in terms of stickiness at pH >7 and suggests that pKa2 e 7.6, i.e.,
very close, perhaps, to pKa1 (26). Sticky protons may also lead to
apparent pKa variation (35, 36).
F
IGURE 6: (A) pH-rate profile (kcat/Km) for the reaction of 7 with
app
pKa1
¼ pKa1 - logð1 þ kb=k - a
Þ
ð3Þ
the R39 DD-peptidase. The points are experimental, and the line
represents the fit of the data to eq 1 (see text). (B) pH-rate profile
(logarithmic) of kcat for reaction of 7 with the R39 DD-peptidase. The
points are experimental, and the line represents the fit of the data to
eq 2 (see text).
More striking in Table 4 is the low pKa1 for 17 (which was
included to assess the effect of the β-lactam side chain) and the
high pKa2 values for the cephalosporins. The low pKa1 for 17 does
not reflect the pKa of the protonated 7-amino group of 17
[4.47 (37)] and must therefore represent the result of complex
binding kinetics. The pKa2 values for 16 and 17 (>10) are
intriguingly much higher than those from the substrate profiles.
Interpretation of these pKas in terms of active site structure can
only be rather speculative. The most obvious dissociable active
site residues are the two conserved lysine residues, one from the
SXXK active site motif (Lys 52 in the R39 sequence) and the
second from the KT(S)G motif (Lys 410). It is possible that at
least one of these residues, most likely, Lys 52, has a low pKa,
corresponding to pKa1 and is required as a general base in
catalysis (9, 16). Interpretation of pKa2 is more problematic,
especially in view of the disparate values for substrates and
β-lactams noted above. Although both peptide and β-lactam (but
not thioester) hydrolysis probably require a general acid catalyst
to assist (amine) leaving group departure, it seems unlikely from
the pKa2 values that they employ the same one, if indeed the
dissociation of this general acid is observed in the pH range
covered (4.5-9.5). It seems more likely that the lower pKa2,
Table 4: pH Dependence of Second-Order Rate Constants for Reaction of
Various Substrates and β-Lactams with the R39 DD-Peptidasea
substrate
k2 (s-1 M-1
)
pKa1
pKa2
7
(7.3 ( 0.7) ꢀ 105
(3.9 ( 0.5) ꢀ 104
(4.7 ( 0.5) ꢀ 104
(1.35 ( 0.07) ꢀ 103
(3.6 ( 0.1) ꢀ 102
6.08 ( 0.16
6.42 ( 0.15
6.56 ( 0.11
7.07 ( 0.20
5.78 ( 0.09
8.42 ( 0.16
8.18 ( 0.33
7.65 ( 0.14
10.86 ( 0.11
11.32 ( 0.12
11
15
16
17
aDetermined in mixed buffer.
dissociation above pH 7 at less than 9.5, the highest pH for
which data were obtained. Further perspective is provided by the
kcat/Km profiles of the peptides 11 and 15 and the β-lactams
cefotaxime (16) and 7-aminocephalosporanic acid (17) (Table 4).
The β-lactams, at low concentration, acylate the enzyme in a
second-order fashion. These second-order rate constants should
be comparable to the kcat/Km parameter for substrates. Unlike

Article
Biochemistry, Vol. 50, No. 3, 2011 385
group, is probably not required, unlike with the amine leaving
Table 5: Solvent Deuterium Kinetic Isotope Effects
group from the peptide 15 (46).
substrate
DV/K
DV
The significantly different DV/K values between the R39 DD
-
peptidase and the other enzymes mentioned above do correlate
nicely with differences in active site structure. The R61 DD
7
2.0 ( 0.4
2.8 ( 0.3
2.4 ( 0.2
2.4 ( 0.2
-
15
peptidase, like class C β-lactamases, has a Tyr/Lys hydrogen-
bonded couple at the active site, which may exist in zwitterionic
form in the free enzyme. It has been proposed that the hydrogen
involved in this hydrogen bond may be the low fractionation
factor contributor to the isotope effect (26, 45). Similarly, the
class A β-lactamases have a Glu/Lys couple that may contribute
analogously. The tyrosine and glutamate residues at these
β-lactamase active sites are thought to be required for efficient
deacylation of acyl-enzymes derived from β-lactams (4). The
DD-peptidases, in general, apart from low molecular mass class B
enzymes, such as that from Streptomyces R61, do not have such
functionality and thus, perhaps, do not have the characteristic
low fractionation factor hydrogen; there is, nonetheless, quite a
lot of intraprotein hydrogen bonding of the active site compo-
nents in DD-peptidases in general, including the R39 enzyme (19).
Conclusions. Previous studies of the substrate specificity of
the R39 DD-peptidase suggested great specificity in the acyl donor
observed in peptide hydrolysis, reflects the dissociation of
another residue, one not in direct contact with the substrate at
the active site but which must be required for peptide and
thioester hydrolysis and efficient departure of the leaving group.
This residue may not be required for acylation of the enzyme by
β-lactams where the higher pKa2 may reflect dissociation of Lys
410. Similar pH profiles have been obtained for a few other DD
-
peptidases (38-41) and generally interpreted in terms of the pKas
of the two lysines.
Interestingly different from the above profiles is that of
Escherichia coli PBP5 as a catalyst of hydrolysis of N,N0-
diacetyl-L-lysyl-D-alanyl-D-alanine (42, 43) where maximal activ-
ity in both kcat/Km and kcat profiles occurs around a pH of 10,
with associated pKas of 9.1 and 10.8. These were interpreted in
terms of dissociations of the two lysines. Both kcat/Km and kcat
profiles, however, also show a small shoulder around pH 7 (with
a reported acid dissociation of pKa 6.1), which may correspond to
the activity maximum observed in the present work and in other
cases (38-41). It might be noted that, conversely, a small rise at
higher pH is clearly visible in the kcat/Km profile of 11 (not
shown). The difference between the relative heights of these
maxima, which is seen between the R39 (and other) DD-pepti-
dases and E. coli PBP5, may reflect the existence of the latter
predominantly in solution, as in the crystal (15), in a less active
conformation; dissociation of a second lysine (or other functional
group) at higher pH does, however, promote the reaction.
Nonetheless, the reported activity of PBP5, even against N,N0-
for a free N-terminal
-R-aminopimelyl N-terminus of 4 obviously closely resembles
the side chain N-terminus of the monomeric stem peptide 5. The
above-referenced paper also showed that the N-acyl- -alanyl-
-isoglutamyl “north fork” of the stem peptide did not convey
D-amino acid, as in the peptide 4 (11). The
D
L
D
reactivity to model substrates such as 3 and therefore probably
did not have a specific binding site on the enzyme. The experi-
ments described in this paper confirm the specificity of the R39
DD-peptidase for a free D-amino acid terminus in the acyl donor
through the reactivity of 6, 7, 9, and 10 (Table 2). Also striking in
this regard is the poor (not observed) reactivity of 9 as a
carboxypeptidase substrate in contrast to its high activity as an
endopeptidase substrate. A considerable part of this difference
must come from the acylation of the N-terminus in 6 when
considered as a carboxypeptidase substrate; the additional acyl
group would probably interact unfavorably with Trp 139 and
Asp 142 in the important side chain N-terminal specificity
site (15).
The complex reactions observed for 6 (Scheme 6) (and their
likely presence with 4) showed that the R39 DD-peptidase can act
as a transpeptidase and endopeptidase. This result is indicated by
the formation and breakdown of the condensation products 9
and 10 in addition to formation of the expected carboxypeptidase
product 8. Finally, the much higher kcat/Km value for the
extended peptide 9 (and, most likely, for its analogue derived
from 4) in the endopeptidase reaction than that for either 6
(carboxypeptidase reaction) or 10 (endopeptidase reaction) sug-
gests that the highest activity of the R39 enzyme would be as an
endopeptidase cleaving a peptidoglycan polymer adjacent to a
free side chain N-terminus (Figure 7). This conclusion is sup-
ported by the effectiveness of extended nucleophiles, N-ε-acetyl-
diacetyl-L-Lys-D-Ala-D-Ala, is very low [kcat/Km values around
1 s-1 M-1 at pH 7 and 11 s-1 M-1 at pH 10 (43)].
Solvent Deuterium Kinetic Isotope Effects. These were
measured for the thioester 7 and the peptide 15 (rather than for 6,
because of the complex kinetics of reaction of the latter; see
above) and are shown in Table 5. The immediate conclusion is
that proton transfer, likely arising from general acid/base cata-
lysis, is present in the transition states for both acylation (V/K)
and deacylation (V). Such catalysis does indeed appear in most
mechanisms that have been proposed (9, 16). Values of DV are
very similar to those previously determined for the Streptomyces
R61 DD-peptidase (44), which were interpreted similarly. Strik-
ingly different, however, are the values of DV/K. For the R61
enzyme and for the structurally related class A and, more partic-
ularly, class C β-lactamases, which also catalyze the hydrolysis of
acyclic thioesters, the DV/K solvent isotope effects are small and
can be either normal or inverse (0.8-1.2) (26, 44). This has been
ascribed to the presence of a low fractionation factor hydrogen in
the free enzyme (44, 45), whose effect is offset by proton transfer
during acylation. The DV/K values obtained here give no indica-
tion of such a hydrogen in the R39 enzyme active site. The large
solvent kinetic isotope effects obtained speak only of general base
catalysis, as noted above. The slightly lower value for the
thioester suggests that transition states for both formation and
breakdown of the tetrahedral intermediate of the acylation
reaction contribute to the observed rate constant and kinetic
isotope effect, of 7 at least, since general acid catalysis of
breakdown of the intermediate from 7, with a thiolate leaving
D
-lysine and, particularly, 6 itself against the acyl-enzyme gener-
ated from 6 or 7 (Table 3). The lesser effectiveness of -Lys(N-Ac-
-Ala- -iGln) as a nucleophile suggests that the presence of the
D
L
D
“north fork” in a substrate leaving group (endopeptidase re-
action) confers little benefit. The much greater effectiveness of the
extended nucleophiles mentioned above than D-alanine (Table 3)
indicates that the R39 DD-peptidase is probably a more effective
endopeptidase than carboxypeptidase. This conclusion agrees
with most current views of the in vivo role of LMM class C

386 Biochemistry, Vol. 50, No. 3, 2011
Adediran et al.
FIGURE 7: Endopeptidase activity of the R39 DD-peptidase. Circles represent N-acetylglucosamine, squares N-acetylmuramic acid, and vertical
lines peptide cross-links. When not cross-linked, the peptides have free side chain N- and C-termini, one of each per N-acetylmuramic acid. The
enzyme cleaves cross-links that have a free side chain N-terminus on the acyl side.
DD-peptidases (2), including E. coli PBP4, for example (20, 46).
Recently, Duez et al. have demonstrated endopeptidase activity
against a small peptidoglycan-mimetic peptide in another such
enzyme, Bacillus subtilis PBP 4a (47). An endopeptidase substrate
15, which is not susceptible to the carboxypeptidase or transpep-
9. Pratt, R. F. (2008) Substrate specificity of bacterial DD-peptidases
(penicillin-binding proteins). Cell. Mol. Life Sci. 65, 2138–2155.
10. Anderson, J. W., and Pratt, R. F. (2000) Dipeptide binding to the
extended active site of the Streptomyces R61 D-alanyl-D-alanine
peptidase: the path to a specific substrate. Biochemistry 39, 12200–
12209.
11. Anderson, J. W., Adediran, S. A., Charlier, P., Nguyen-Disteche, M.,
ꢁ
Frere, J.-M., Nicholas, R. A., and Pratt, R. F. (2003) On the substrate
tidase (here defined as beginning by
reactions, has been described.
D-alanine displacement)
ꢁ
specificity of bacterial DD-peptidases: evidence from two series of
peptidoglycan-mimetic peptides. Biochem. J. 373, 949–955.
12. McDonough, M. A., Anderson, J. W., Silvaggi, N. R., Pratt, R. F.,
Knox, J. R., and Kelly, J. A. (2002) Structures of two kinetic
intermediates reveal species specificity of penicillin-binding proteins.
J. Mol. Biol. 322, 111–122.
The pH dependence of kcat/Km for the carboxypeptidase
reactions of 7, 11, and 15 with the R39 DD-peptidase gave
evidence of complex proton-dissociation schemes and/or sticky
substrates or protons. The pH dependence of reaction of the
β-lactams 16 and 17 with the enzyme may also reflect complicated
phenomena since they do not mimic the pH dependence of the
peptide substrates (Table 4). The reactions of peptides and
β-lactams may occur at active sites with locally different protein
conformations. Finally, the solvent kinetic deuterium isotope
effects DV/K and DV indicate straightforward general acid/base
catalysis in both enzyme acylation and deacylation steps of
turnover.
13. Silvaggi, N. R., Anderson, J. W., Brinsmade, S. R., Pratt, R. F., and
Kelly, J. A. (2003) The crystal structure of a phosphonate-inhibited
D
-Ala-D-Ala peptidase reveals an analogue of a tetrahedral transition
state. Biochemistry 42, 1199–1208.
14. Silvaggi, N. R., Josephine, H. R., Kuzin, A. P., Nagarajan, R., Pratt,
R. F., and Kelly, J. A. (2005) Crystal structures of complexes between
the R61 DD-peptidase and peptidoglycan-mimetic β-lactams: a non-
covalent complex with a “perfect penicillin. J. Mol. Biol. 345, 521–
533.
15. Sauvage, E., Powell, A. J., Heilemann, J., Josephine, H. R., Charlier,
P., Davies, C., and Pratt, R. F. (2008) Crystal structures of complexes
of bacterial DD-peptidases with peptidoglycan-mimetic ligands; the
substrate specificity puzzle. J. Mol. Biol. 381, 383–393.
16. Dzhekieva, L., Rocaboy, M., Kerff, F., Charlier, P., Sauvage, E., and
Pratt, R. F. (2010) Crystal structure of a complex between the
Actinomadura R39 DD-peptidase and a peptidoglycan-mimetic bor-
onate inhibitor: interpretation of a transition state analogue in terms
of catalytic mechanism. Biochemistry 49, 6411–6419.
The methods employed above may, in principle, be applied to
any bacterial DD-peptidase in order to better define its substrate
specificity.
SUPPORTING INFORMATION AVAILABLE
Details of the syntheses of the substrates 6 and 7. This material
is available free of charge via the Internet at http://pubs.acs.org.
ꢁ
17. Ghuysen, J.-M., Frere, J.-M., Leyh-Bouille, M., Coyette, J., Dusart,
J., and Nguyen-Disteche, M. (1977) Use of model enzymes in
ꢁ
determination of the mode of action of penicillins and Δ3-cephalos-
porins. Annu. Rev. Biochem. 48, 73–101.
18. Zhao, G.-H., Duez, C., LePage, S., Forceille, C., Rhazi, N., Klein, D.,
REFERENCES
ꢁ
Ghuysen, J.-M., and Frere, J.-M. (1997) Site-directed mutagenesis of
the Actinomadura R39 DD-peptidase. Biochem. J. 327, 377–381.
1. Vollmer, W., and Bertsche, U. (2007) Murein (peptidoglycan) struc-
ture, architecture and biosynthesis in Escherichia coli. Biochim.
Biophys. Acta 1778, 1714–1734.
ꢁ
19. Sauvage, E., Herman, R., Petrella, S., Duez, C., Bouillenne, F., Frere,
2. Sauvage, E., Kerff, F., Terrak, M., Ayala, J. A., and Charlier, P.
(2008) The penicillin-binding proteins: structure and role in peptido-
glycan biosynthesis. FEMS Microbiol. Rev. 32, 234–258.
3. Waxman, D. J., and Strominger, J. L. (1983) Penicillin-binding
proteins and the mechanism of action of β-lactam antibiotics. Annu.
Rev. Biochem. 52, 825–869.
J.-M., and Charlier, P. (2005) Crystal structure of the Actinomadura
R39 DD-peptidase reveals new domains in penicillin-binding proteins.
J. Biol. Chem. 280, 31249–31256.
20. Granier, B., Duez, C., Lepange, S., Englebert, S., Dusart, J., Dideberg,
ꢁ
O., Van Beeumen, J., Frere, J.-M., and Ghuysen, J.-M. (1992) Primary
and predicted secondary structures of the Actinomadura R39 extra-
cellular DD-peptidase, a penicillin-binding protein (PBP) related to
Escherichia coli PBP4. Biochem. J. 282, 781–788.
4. Pratt, R. F. (2002) Functional evolution of the serine β-lactamase
active site. J. Chem. Soc. Perkin 2, 851–861.
5. Payne, D. J., Gwynn, M. N., Holmes, D. J., and Pompliano, D. L.
(2007) Drugs for bad bugs: confronting the challenges of antibacterial
discovery. Nat. Rev. Drug Discov. 6, 29–40.
6. Boucher, H. W., Talbot, G. H., Bradley, J. S., Edwards, T. E., Gilbert,
D., Rice, L. B., Scheld, M., Spellberg, B., and Bartlett, J. (2009) Bad
bugs, no drugs: no ESKAPE! An update from the Infectious Diseases
Society of America. Clin. Infect. Dis. 48, 1–12.
21. Nagarajan, R., and Pratt, R. F. (2004) Synthesis and evaluation of
new substrate analogues of Streptomyces R61 DD-peptidase: dissec-
tion of a specific ligand. J. Org. Chem. 69, 7472–7478.
22. Johnson, K. A., Simpson, Z. B., and Blom, T. (2009) A new computer
program for dynamic simulations and fitting of kinetic data. Anal.
Biochem. 387, 20–29.
23. Kuzmic, P. (1996) Program DYNAFIT for the analysis of enzyme
kinetic data: application to HIV proteinase. Anal. Biochem. 237, 260–
273.
7. Ghuysen, J.-M. (1991) Serine β-lactamases and penicillin-binding
proteins. Annu. Rev. Microbiol. 45, 37–67.
ꢀ
24. Frere, J.-M., Leyh-Bouille, M., Ghuysen, J.-M., Nieto, M., and
Perkins, H. R. (1976) Exocellular DD-carboxypeptidase-transpepti-
8. Goffin, C., and Ghuysen, J.-M. (1998) Multimodular penicillin-
binding proteins: an enigmatic family of orthologs and paralogs.
Microbiol. Mol. Biol. Rev. 62, 1079–1093.
dases from Streptomyces. Methods Enzymol. 45, 610–636.

Article
Biochemistry, Vol. 50, No. 3, 2011 387
ꢀ
ꢀ
ꢀ
ꢀ
€ €
ꢀ
25. Fuad, N., Frere, J.-M., Ghuysen, J.-M., Duez, C., and Iwatsubo, M.
(1974) Mode of interaction between β-lactam antibiotics and the
exocellular DD-carboxypeptidase-transpeptidase from Streptomyces
R39. Biochem. J. 155, 623–629.
37. Koczian, K., Szakacs, Z., Kokosi, J., and Noszal, B. (2007) Site-
specific protonation micro equilibria of penicillin and cephalosporin
beta-lactam core molecules. Eur. J. Pharm. Sci. 32, 1–7.
ꢁ
ꢁ
38. Varetto, L., Frere, J.-M., Nguyen-Disteche, M., Ghuysen, J.-M., and
Houssier, C. (1987) The pH dependence of the active site serine DD-
26. Adediran, S. A., Deraniyagala, S. A., Xu, Y., and Pratt, R. F. (1996)
β-Secondary and solvent deuterium kinetic isotope effects on
β-lactamase catalysis. Biochemistry 35, 3604–3613.
27. Stein, R. L. (1983) Catalysis by human leukocyte elastase: substrate
structural operation of the charge relay system. J. Am. Chem. Soc.
105, 5111–5116.
28. Adediran, S. A., Kumar, I., and Pratt, R. F. (2006) Deacylation
transition states of a bacterial DD-peptidase. Biochemistry 45, 13074–
13082.
29. Kumar, I., and Pratt, R. F. (2005) Transpeptidation reactions of a
specific substrate catalyzed by the Streptomyces R61 DD-peptidase:
characterization of a chromogenic substrate and acyl acceptor design.
Biochemistry 30, 9971–9979.
peptidase of Streptomyces R61. Eur. J. Biochem. 162, 525–531.
€
39. Stevanova, M. E., Tomberg, J., Olesky, M., Holtje, J.-V., Gutheil,
W. G., and Nicholas, R. A. (2003) Neisseria gonorrhoeae penicillin-
binding protein 3 exhibits exceptionally high carboxypeptidase and
β-lactam binding activities. Biochemistry 42, 14614–14625.
40. Stefanova, M. E., Tomberg, J., Davies, C., Nicholas, R. A., and
Gutheil, W. G. (2004) Overexpression and enzymatic characterization
of Neisseria gonorrhoeae penicillin-binding protein 4. Eur. J. Biochem.
271, 23–32.
41. Thomas, B., Wang, Y., and Stein, R. L. (2001) Kinetic and mechan-
istic studies of penicillin-binding protein 2x from Streptococcus
pneumoniae. Biochemistry 40, 15811–15823.
ꢁ
30. Adam, M., Damblon, C., Plaitin, B., Christiaens, L., and Frere, J.-M.
42. Stefanova, M. E., Davies, C., Nicholas, R. A., and Gutheil, W. G.
(2009) pH, inhibitor and substrate specificity studies on Escherichia
coli penicillin-binding protein 5. Biochim. Biophys. Acta 1597, 292–
300.
(1990) Chromogenic depsipeptide substrates for β-lactamases and
penicillin-sensitive DD-peptidases. Biochem. J. 270, 525–529.
ꢁ
31. Jamin, M., Adam, M., Damblon, C., Christiaens, L., and Frere, J.-M.
(1991) Accumulation of acyl-enzymes in DD-peptidase-catalyzed reac-
tions with analogues of peptide substrates. Biochem. J. 280, 499–506.
32. Kumar, I., and Pratt, R. F. (2005) Transpeptidation reactions of a
specific substrate catalyzed by the Streptomyces R61 DD-peptidase:
the structural basis of acyl acceptor specificity. Biochemistry 44, 9961–
9970.
43. Zhang, W., Shi, Q., Meroueh, S. O., Vakulenko, S. B., and Mobash-
ery, S. (2007) Catalytic mechanism of penicillin-binding protein 5 of
Escherichia coli. Biochemistry 46, 10113–10121.
44. Adediran, S. A., and Pratt, R. F. (1999) β-Secondary and solvent
deuterium isotope effects on catalysis by the Streptomyces R61
DD-peptidase: comparisons with
a structurally similar class C
33. Peller, L., and Alberty, R. A. (1959) Multiple intermediates in steady
state enzyme kinetics. I. The mechanism involving a single substrate
and product. J. Am. Chem. Soc. 81, 5907–5914.
34. Knowles, J. R. (1976) The intrinsic pKa values of functional groups in
enzymes: improper deductions from the pH-dependence of steady
state parameters. CRC Crit. Rev. Biochem. 4, 165–173.
35. Cleland, W. W. (1977) Determining the chemical mechanisms of
enzyme-catalyzed reactions by kinetics studies. Adv. Enzymol. 45,
273–387.
β-lactamase. Biochemistry 38, 1469–1477.
45. Page, M. I., Vilanova, B., and Layland, N. J. (1995) pH dependence of
and kinetic solvent isotope effects on the methanolysis and hydrolysis
of β-lactams catalyzed by class C β-lactamase. J. Am. Chem. Soc. 117,
12092–12095.
46. Korat, B., Mottl, H., and Keck, W. (1991) Penicillin-binding protein 4
of Escherichia coli: molecular cloning of the dac B gene, controlled
overexpression, and alterations in murein composition. Mol. Micro-
biol. 5, 675–684.
ꢀ
47. Duez, C., Zervosen, A., Teller, N., Melkonian, R., Banzubaze, E.,
36. Hardy, L. W., Nishida, C. H., and Kirsch, J. F. (1984) Anomalous pH
dependence of the reactions of carbenicillin and sulbenicillin with
Bacillus cereus β-lactamase I. Influence of the R-substituent charge on
the kinetic parameters. Biochemistry 23, 1288–1294.
ꢁ
Bouillenne, F., Luxen, A., and Frere, J.-M. (2009) Characterization
of the proteins encoded by the Bacillus subtilis yoxA-dacC operon.
FEMS Microbiol. Lett. 300, 42–47.