STRUCTURAL DYNAMICS 4, 044003 (2017)

Structural enzymology using X-ray free electron lasers
Christopher Kupitz,1,a) Jose L. Olmos, Jr.,2,a) Mark Holl,3 Lee Tremblay,4
€r,5,6 Mark Hunter,7
Kanupriya Pande,5,6 Suraj Pandey,1 Dominik Oberthu
7
7
1
Mengning Liang, Andrew Aquila, Jason Tenboer, George Calvey,8
Andrea Katz,8 Yujie Chen,8 Max O. Wiedorn,5 Juraj Knoska,5 Alke Meents,6
Valerio Majriani,5,6 Tyler Norwood,1 Ishwor Poudyal,1 Thomas Grant,9
Mitchell D. Miller,2 Weijun Xu,2 Aleksandra Tolstikova,5 Andrew Morgan,5
Markus Metz,5 Jose M. Martin-Garcia,10 James D. Zook,10
Shatabdi Roy-Chowdhury,10 Jesse Coe,10 Nirupa Nagaratnam,10
Domingo Meza,10 Raimund Fromme,10 Shibom Basu,10 Matthias Frank,11
Thomas White,5 Anton Barty,5 Sasa Bajt,5 Oleksandr Yefanov,5
Henry N. Chapman,5,6 Nadia Zatsepin,3 Garrett Nelson,3 Uwe Weierstall,3
John Spence,3 Peter Schwander,1 Lois Pollack,8 Petra Fromme,10
Abbas Ourmazd,1 George N. Phillips, Jr.,2 and Marius Schmidt1,b)
1

Physics Department, University of Wisconsin-Milwaukee, 3135 N. Maryland Ave,
Milwaukee, Wisconsin 53211, USA
2
Department of BioSciences, Rice University, 6100 Main Street, Houston, Texas 77005, USA
3
Department of Physics, Arizona State University, Tempe, Arizona 85287, USA
4
Marbles Inc., 1900 Belvedere Pl, Westfield, Indiana 46074, USA
5
Center for Free-Electron Laser Science, DESY, Notkestrasse 85, 22607 Hamburg, Germany
6
University of Hamburg, Luruper Chaussee 149, 22761 Hamburg, Germany
7
Linac Coherent Light Source, Stanford Linear Accelerator Center (SLAC) National
Accelerator Laboratory, 2575 Sand Hill Road, Menlo Park, California 94025, USA
8
Department of Applied and Engineering Physics, Cornell University, 254 Clark Hall,
Ithaca, New York 14853, USA
9
Hauptman-Woodward Institute, State University of New York at Buffalo, 700 Ellicott Street,
Buffalo, New York 14203, USA
10
School of Molecular Sciences and Biodesign Center for Applied Structural Discovery,
Arizona State University, Tempe, Arizona 85287-1604, USA
11
Lawrence Livermore National Laboratory, Livermore, California 94550, USA
(Received 30 August 2016; accepted 29 November 2016; published online 15 December 2016)

Mix-and-inject serial crystallography (MISC) is a technique designed to image
enzyme catalyzed reactions in which small protein crystals are mixed with a
substrate just prior to being probed by an X-ray pulse. This approach offers several
advantages over flow cell studies. It provides (i) room temperature structures at near
atomic resolution, (ii) time resolution ranging from microseconds to seconds, and
(iii) convenient reaction initiation. It outruns radiation damage by using femtosecond
X-ray pulses allowing damage and chemistry to be separated. Here, we demonstrate
that MISC is feasible at an X-ray free electron laser by studying the reaction of M.
tuberculosis ß-lactamase microcrystals with ceftriaxone antibiotic solution. Electron
density maps of the apo-ß-lactamase and of the ceftriaxone bound form were
obtained at 2.8 Å and 2.4 Å resolution, respectively. These results pave the way to
study cyclic and non-cyclic reactions and represent a new field of time-resolved
C 2016
structural dynamics for numerous substrate-triggered biological reactions. V
Author(s). All article content, except where otherwise noted, is licensed under a
Creative Commons Attribution (CC BY) license (http://creativecommons.org/
licenses/by/4.0/). [http://dx.doi.org/10.1063/1.4972069]

a)

C. Kupitz and J. L. Olmos, Jr. contributed equally to this work.
Author to whom correspondence should be addressed. Electronic mail: smarius@uwm.edu. Tel.: 414-229-4338.

b)

2329-7778/2017/4(4)/044003/7

4, 044003-1

C Author(s) 2016.
V

044003-2

Kupitz et al.

Struct. Dyn. 4, 044003 (2017)

INTRODUCTION

ß-lactams are a class of antibiotics used frequently in the treatment of bacterial infection.1
They form one of the most important antibiotic families, including a broad range of molecules
such as penicillin derivatives, cephalosporins, and ß-lactamase inhibitors, as well as penems
and carbapenems.2–4 The broad range of activity of these antibiotics against both Gramnegative and Gram-positive pathogens renders ß-lactams some of the most widely used antibiotics worldwide.5 Evolutionary bacterial response has given rise to ß-lactamase enzymes able to
hydrolyze the ß-lactam ring, thereby conferring antibiotic resistance to bacteria.4,6 This increasing resistance now represents a serious challenge for the efficacy of this class of antibiotics.7
Tuberculosis (TB), a disease caused by the bacteria Mycobacterium tuberculosis (MTb), is
a major health concern worldwide with approximately 9.6  106 new cases in 2014, 1.5  106
deaths in the same year. TB remains a leading cause of death among AIDS patients.8 One reason tuberculosis has evaded treatment with b-lactams is the evolution of a singular ß-lactamase
(BlaC) found in MTb.9,10 Understanding the structural origins of binding modes of BlaC has
resulted in improved compounds for the treatment of not only TB but also Extensively Drug
Resistant (XDR-TB) strains of the disease.11
BlaC catalyzes the opening of the ß-lactam ring by nucleophilic attack via an active site
serine residue that generates an acyl-enzyme intermediate state. The intermediate formation is
followed by the generation of the ring-opened, inactive product through hydrolysis of the ester
bond.12 In addition, in cephalosporin substrates, the thiodioxotriazine moiety is a good leaving
group and likely dissociates in a concerted manner with the ring opening and formation of the
acyl intermediate.13–15 Previous studies have further shown that BlaC is a broad-spectrum ß-lactamase, hydrolyzing virtually all ß-lactam classes.16 Direct observations of this process would
lead to a better understanding of enzyme-drug interaction and could allow for novel ways to
bypass or inhibit the BlaC catalyzed ring hydrolysis, such as drug targeting of stable intermediate states. Many structures of M. tuberculosis BlaC have been solved with various substrates
via conventional X-ray crystallography. However, these are static structures showing only the
binding interactions and not the coordinated actions of the enzyme that result in binding and
catalysis.17–20
Time-resolved crystallography21 enables time-dependent structure determination to reveal
protein kinetics.22 Light-triggered studies at synchrotrons can provide up to 100 ps time resolution.23–25 However, one of the major challenges with time-resolved macromolecular crystallography is the study of reactions that do not “reset” for the next X-ray pulse. Reactions that are
not light-driven, like most enzymatic reactions, can be difficult or impossible to capture in crystals because of the difficulty in initiating reactions in a concerted way in the molecules throughout the volume of a large crystal needed for conventional crystallographic measurements.
Calculations show that this issue can be resolved using micron-sized crystals because the
reduced size drastically decreases the diffusion time of the substrate into the crystal.26–28 If diffusion times are much faster than the enzymatic turnover times, this provides an elegant and
straightforward way to initiate reactions in enzyme crystals.
Crystals of this small size (<10 lm) are difficult to examine at a synchrotron, due to radiation damage and the lack of well-established sample delivery methods. The introduction of Xray free electron lasers (XFELs) therefore presents new opportunities in several regards. XFELs
have enabled time-resolved structure determination of fast light-activated proteins after optical
illumination, allowing structures of protein conformational changes to be imaged on time scales
of hundreds of femtoseconds that have been previously impossible to reach.29–33 However, light
sensitive proteins are only one type of proteins whose dynamics are of interest. Many, if not
most, physiologically relevant reactions are initiated chemically rather than by light. The ability
to mix a protein with a substrate and directly observe the various structural intermediates associated with a chemical reaction would open an entirely new field in structural enzymology,
incorporating a new, more holistic perspective on substrate binding, significant intermediates,
bound products, and the kinetics of their appearance and disappearance. Such reactions need to
be studied at room temperature in order to explore relevant kinetics and discover the structure

044003-3

Kupitz et al.

Struct. Dyn. 4, 044003 (2017)

of the transitional states. In this paper, we present the evidence that such experiments are possible, using the mix-and-inject serial crystallography (MISC) approach at an XFEL. During the
review of this paper, Stagno et al. also reported successful diffusion of an amino acid (adenine)
into RNA with a 10 s mixing time using a similar setup, providing further credence that the
technique is a viable method for future enzymological studies.34
RESULTS AND DISCUSSION

Structures of BlaC with and without ceftriaxone (CEF), a cephalosporin antibiotic substrate,
were measured at room temperature using the Coherent X-ray Imaging (CXI) instrument at the
Linac Coherent Light Source (LCLS)35 at SLAC. The antibiotic was mixed with microcrystals of
BlaC using a T-junction (Fig. 1), which allowed for a diffusion/reaction time of about 2 s (see
supplementary material). The time-resolution of a mix-and-inject experiment also depends on the
rate of diffusion of the substrate into the crystals, limiting the time resolution that can be
achieved at synchrotron sources.26 Our crystals have average widths on the order of 3–10 lm in
two of the dimensions and thickness of 2–3 lm in the third. Diffusion calculations indicate that
the diffusion of the substrate into these microcrystals takes only about 1–15 ms.26 This is about
one hundred times smaller than the time required for the crystal substrate mixture to travel to the
injector and can be neglected in our experiment. Therefore, with improved mixing-injectors,36,37
this time scale should allow for a variety of enzyme reactions to be studied, from the long time
delays (seconds) to short time delays (500 ls).37 To demonstrate substrate binding with MISC as
a proof-of-principle, two separate room temperature data sets were collected using the serial femtosecond crystallographic (SFX) approach: an apo structure from 12 853 indexed diffraction patterns and a “mixed” structure from 22 646 indexed patterns with CEF added.
The data collected were used to determine the structures to a resolution of 2.8 Å for the apo
and 2.4 Å for the mixed form. Fig. 2 shows an overview of the BlaC asymmetric unit, as well as
the ligand binding pocket from both the apo and the BlaC mixed with CEF. The subunits labeled
A and C show no binding to ceftriaxone, highlighting the point that the kinetics of a catalytic
cycle in the crystal are not necessarily the same as in solution. The crystal packing of these subunits and their crystal contacts appear to prevent facile access to the binding pocket, showing only
the presence of phosphate. The density in the binding pocket of subunits D shows strong evidence of substrate accessibility with the active site electron density (ED) from subunit B being
less prominent, but still present. Given the long mixing times of this experiment along with the

FIG. 1. Data collection schematic showing the T-junction set-up used for a mixing time of about 2 s. The T-junction was
placed outside of the nozzle rod in our experiment but could also be engineered to fit inside closer to the interaction region
for shorter mixing times.

044003-4

Kupitz et al.

Struct. Dyn. 4, 044003 (2017)

FIG. 2. Electron density in the catalytic cleft of BlaC. (a) Refined model of the entire tetramer (r ¼ 1.1) in the asymmetric
unit after mixing. The mixed electron density (2Fo-Fc) is shown in blue in the binding pockets. Subunits A and C contain
phosphate while subunits B and D have a bound ceftriaxone, with the electron density of D being slightly stronger. (b)
Enlarged section of subunit D showing the unmixed ED, which corresponds to a bound phosphate. (c) Enlarged section of
subunit D showing the mixed ED (blue electron density) with ceftriaxone modelled in. Slightly different views of the same
subunit binding pocket are shown in (b) and (c); however, there are minimal changes to the ligand binding sphere (see supplementary material, Table S2).

measured K(cat) for ceftriaxone of 49 6 17/min (turn over time of 1.22 s),16 it is likely that on
average, at least one CEF molecule, perhaps two, was cleaved during the time between mixing
and actually interacting with the X-rays. Models of the uncleaved, acyl intermediate, and product
complex of the ceftriaxone molecule were modelled to attempt to fit the electron density (ED)
found in the BlaC active site. For the current time point of 1.4–2.6 s, we expect to observe the
superposition of several chemical intermediates in the ED. Since our time delay is more than 2 s
and the turnover time is on the order of seconds, a steady state has most likely formed. The
steady state consists of all states along the enzymatic pathway that are occupied according to
their free energies. The enzyme substrate complex, any putative intermediate, and the product
state may all be occupied and represented by the average electron density. Therefore, we
attempted to refine three different structures into the electron density (Figs. S3–S5, supplementary
material). Accordingly, the chemical form with the longest presence in the active site, i.e., the
rate-limiting step, will be most represented in the ED. From the modelling of our data, it appears
that we mainly observe a steady state. The Km value was previously measured at 520 lM.16
When considered in conjunction with the moderately slow turnover rate of approximately 1.22 s,
the data support that the rate-limiting step for this reaction has not been measured but is likely
the formation of the acyl intermediate (for a further discussion of the chemical kinetics see the
supplementary material). In this state, the enzyme requires time to achieve coordinated alignment
for the nucleophilic attack of the Ser70 residue on the b-lactam ring before the reaction can proceed to the acyl intermediate. As seen in other inhibitor complexes, other protein interactions we
observe with ceftriaxone (and the phosphate) include threonine (Thr237), glutamic acid (Glu166),
lysine (Lys73), and an additional serine residue (Ser128).
We next compare the active site structure presented here with the structure when the
antibiotic cefamandole is soaked into large BlaC crystals (PDB code 3N8S38). The structure

044003-5

Kupitz et al.

Struct. Dyn. 4, 044003 (2017)

of the Michaelis complex as well as the covalent acyl intermediate for mutants of BlaC was
solved at a synchrotron at cryogenic temperatures.38 In the cefamandole structure, the Ser70
nucleophilic oxygen is 2.8 Å from the b-lactam ring carbonyl carbon compared to 3.2 Å for
the ceftriaxone Ser70 oxygen. Both the complexes are positioned almost identically in the
active site. Interestingly, the Ser142, in our structure, which binds the carboxylate group, is
turned and the carboxylate is positioned over 3 Å away. This suggests that the ceftriaxone
molecule is bound more loosely and has a lower affinity in the active site than cefamandole;
however, it is impossible to fully separate the differences between the two due to the differences
in techniques used to solve the structure, especially the difference between room-temperature
and cryo-cooled data collection. For Thr253 and K250, this results in the ceftriaxone carboxylate group being rotated by nearly 90 . These observations explain the difference between the
observed kinetic parameters of cefamandole (Km ¼ 184 lM with kcat ¼ 3500 min1) and ceftriaxone (Km ¼ 520 lM and kcat ¼ 49 min1).16 The high turnover rate of cefamandole is why it
was deemed unsuitable for this initial proof-of-concept work. Since we see no clear density
connecting the Ser70 to the lactam ring, we have modeled the structure as the enzymesubstrate complex in this study (see Fig. 2); however, it is most likely a mixture of free
enzyme, enzyme substrate complex, and product bound to the enzyme (see supplementary
material for further discussion).
In conclusion, our data show clear electron density in the binding pockets, which changes
between the apo and mixed states. The electron density of the mixed sample matches ceftriaxone and its products fairly well, which shows that binding of the substrate to BlaC in the crystal
took place, and occurred extensively across the crystals so that electron density is clearly visible in the active site. The lack of strong density for the thiodioxotriazine ring also suggests that
the enzyme is turning over, although we cannot rule out the possibility that it is just disordered
in this context. This constitutes a proof-of-principle that room temperature MISC on enzymes is
possible. Future experiments with further increase in time resolution of the mixing process
should allow us to separate the events in time throughout the course of the catalytic cycle, free
of radiation-damage and photoelectron reduction effects. The potential impact of this method
on the field of enzymology and related disciplines is unquestionably exciting, since it allows
both the atomic structures of stable intermediates and the time scales of their inter-conversions
to be observed directly at atomic resolution.
SUPPLEMENTARY MATERIAL

See supplementary material for the further detailed discussion and analysis.
ACKNOWLEDGMENTS

This work was supported by NSF-STC “BioXFEL” (STC-1231306) and NIH Grant Nos.
GM098248 and GM109456 (Phillips for sample preparation) as well as NIH Grant No. GM095583
(to P.F. and members of her team). This material is based upon work supported by the National
Science Foundation Graduate Research Fellowship Program to J.L.O. under Grant No. 1450681.
A.O. was supported by the U.S. Department of Energy, Office of Science, Basic Energy Sciences
under Award No. DE-SC0002164 (algorithm design and development), and by the U.S. National
Science Foundation under Award Nos. STC 1231306 (joint initiation and design of experiment) and
1551489 (underlying analytical models). Use of the Linac Coherent Light Source (LCLS), SLAC
National Accelerator Laboratory, is supported by the U.S. Department of Energy, Office of Science,
Office of Basic Energy Sciences under Contract No. DE-AC02-76SF00515. Part of this work was
performed under the auspices of the U.S. Department of Energy by Lawrence Livermore National
Laboratory under Contract No. DE-AC52-07NA27344, and M.F. was supported by LLNL LabDirected Research and Development Project No. 012-ERD-031. G.N. was supported by NIH Award
No. GM097463. Parts of the sample injector used at LCLS for this research was funded by the
National Institutes of Health, P41GM103393, formerly P41RR001209. Support is acknowledged
from the Biodesign Center for Applied Structural Discovery at ASU, the Helmholtz Association
Virtual Institute Dynamic Pathways and the BMBF through Project Nos. 05K13GUK and

044003-6

Kupitz et al.

Struct. Dyn. 4, 044003 (2017)

05K14CHA. M.M. is supported by the European Union under the programme FP7-PEOPLE-2012ITN Nanomem, Grant No. 317079. We thank C. Conrad for helpful discussions.
1

K. Lewis, “Platforms for antibiotic discovery,” Nat. Rev. Drug Discovery 12, 371–387.
G. S. Singh, “Beta-lactams in the new millennium. Part-I: monobactams and carbapenems,” Mini Rev. Med. Chem. 4,
69–92 (2004).
3
M. G. Page, “Cephalosporins in clinical development,” Expert Opin. Invest. Drugs 13, 973–985 (2004).
4
M. R. Meini, L. I. Llarrull, and A. J. Vila, “Overcoming differences: The catalytic mechanism of metallo-betalactamases,” FEBS Lett. 589, 3419–3432 (2015).
5
T. Palzkill, “Metallo-beta-lactamase structure and function,” Ann. N.Y. Acad. Sci. 1277, 91–104 (2013).
6
K. Bush and J. F. Fisher, “Epidemiological expansion, structural studies, and clinical challenges of new beta-lactamases
from Gram-negative bacteria,” Annu. Rev. Microbiol. 65, 455–478 (2011).
7
M. Babic, A. M. Hujer, and R. A. Bonomo, “What’s new in antibiotic resistance? Focus on beta-lactamases,” Drug
Resist. Updates 9, 142–156 (2006).
8
World Health Organization, Global Tuberculosis Programme (Global Tuberculosis Programme, Geneva, 2015), Vol. 15.
9
C. N. Iland, “The effect of penicillin on the tubercle bacillus,” J. Pathol. Bacteriol. 58, 495–500 (1946).
10
L. W. Tremblay, F. Fan, and J. S. Blanchard, “Biochemical and structural characterization of Mycobacterium tuberculosis
beta-lactamase with the carbapenems ertapenem and doripenem,” Biochemistry 49, 3766–3773 (2010).
11
J. E. Hugonnet, L. W. Tremblay, H. I. Boshoff, C. E. Barry 3rd, and J. S. Blanchard, “Meropenem-clavulanate is effective
against extensively drug-resistant Mycobacterium tuberculosis,” Science 323, 1215–1218 (2009).
12
J. H. Jeon et al., “Structural basis for carbapenem-hydrolyzing mechanisms of carbapenemases conferring antibiotic
resistance,” Int. J. Mol. Sci. 16, 9654–9692 (2015).
13
V. Dubee et al., “Inactivation of Mycobacterium tuberculosis l,d-transpeptidase LdtMt(1) by carbapenems and cephalosporins,” Antimicrob. Agents Chemotherapy 56, 4189–4195 (2012).
14
E. Perez-Inestrosa et al., “Cephalosporin chemical reactivity and its immunological implications,” Curr. Opin. Allergy
Clin. Immunol. 5, 323–330 (2005).
15
D. B. Boyd and W. H. Lunn, “Electronic structures of cephalosporins and penicillins. 9. Departure of a leaving group in
cephalosporins,” J. Med. Chem. 22, 778–784 (1979).
16
J. E. Hugonnet and J. S. Blanchard, “Irreversible inhibition of the Mycobacterium tuberculosis beta-lactamase by clavulanate,” Biochemistry 46, 11998–12004 (2007).
17
F. Wang, C. Cassidy, and J. C. Sacchettini, “Crystal structure and activity studies of the Mycobacterium tuberculosis
beta-lactamase reveal its critical role in resistance to beta-lactam antibiotics,” Antimicrob. Agents Chemother. 50,
2762–2771 (2006).
18
C. Feiler et al., “Directed evolution of Mycobacterium tuberculosis beta-lactamase reveals gatekeeper residue that regulates antibiotic resistance and catalytic efficiency,” PloS One 8, e73123 (2013).
19
S. Hazra et al., “Kinetic and structural characterization of the interaction of 6-methylidene penem 2 with the betalactamase from Mycobacterium tuberculosis,” Biochemistry 54, 5657–5664 (2015).
20
L. W. Tremblay, J. E. Hugonnet, and J. S. Blanchard, “Structure of the covalent adduct formed between Mycobacterium
tuberculosis beta-lactamase and clavulanate,” Biochemistry 47, 5312–5316 (2008).
21
K. Moffat, “Time-resolved macromolecular crystallography,” Annu. Rev. Biophys. Biophys. Chem. 18, 309–332
(1989).
22
M. Schmidt, Structure Based Kinetics by Time-Resolved X-ray Crystallography, Biological and Medical Physics,
Biomedical Engineering (Springer, 2008), pp. 201–241.
23
F. Schotte et al., “Watching a signaling protein function in real time via 100-ps time-resolved Laue crystallography,”
Proc. Natl. Acad. Sci. U.S.A. 109, 19256–19261 (2012).
24
F. Schotte et al., “Watching a protein as it functions with 150-ps time-resolved x-ray crystallography,” Science 300,
1944–1947 (2003).
25
Y. O. Jung et al., “Volume-conserving trans-cis isomerization pathways in photoactive yellow protein visualized by picosecond X-ray crystallography,” Nat. Chem. 5, 212–220 (2013).
26
M. Schmidt, “Mix and inject: Reaction initiation by diffusion for time-resolved macromolecular crystallography,” Adv.
Condens. Matter Phys. 2013, 1–10.
27
S. Geremia, M. Campagnolo, N. Demitri, and L. N. Johnson, “Simulation of diffusion time of small molecules in protein
crystals,” Structure 14, 393–400 (2006).
28
J. Hajdu et al., “Catalysis in the crystal: Synchrotron radiation studies with glycogen phosphorylase b,” EMBO J. 6,
539–546 (1987).
29
T. R. Barends et al., “Direct observation of ultrafast collective motions in CO myoglobin upon ligand dissociation,”
Science 350, 445–450 (2015).
30
S. Boutet et al., “High-resolution protein structure determination by serial femtosecond crystallography,” Science 337,
362–364 (2012).
31
C. Kupitz et al., “Serial time-resolved crystallography of photosystem II using a femtosecond X-ray laser,” Nature 513,
261–265 (2014).
32
J. Tenboer et al., “Time-resolved serial crystallography captures high-resolution intermediates of photoactive yellow
protein,” Science 346, 1242–1246 (2014).
33
K. Pande et al., “Femtosecond structural dynamics drives the trans/cis isomerization in photoactive yellow protein,”
Science 352, 725–729 (2016).
34
J. R. Stagno et al., “Structures of riboswitch RNA reaction states by mix-and-inject XFEL serial crystallography,” Nature
(to be published).
35
M. Liang et al., “The coherent x-ray imaging instrument at the Linac coherent light source,” J. Synchrotron Radiat. 22,
514–519 (2015).
2

044003-7
36

Kupitz et al.

Struct. Dyn. 4, 044003 (2017)

D. Wang, U. Weierstall, L. Pollack, and J. Spence, “Double-focusing mixing jet for XFEL study of chemical kinetics,”
J. Synchrotron Radiat. 21, 1364–1366 (2014).
G. D. Calvey, A. M. Katz, C. B. Schaffer, and L. Pollack, “Mixing injector enables time-resolved crystallography with
high hit rate at X-ray free electron lasers,” Struct. Dyn. 3, 054301 (2016).
38
L. W. Tremblay, H. Xu, and J. S. Blanchard, “Structures of the Michaelis complex (1.2 A) and the covalent acyl intermediate (2.0 A) of cefamandole bound in the active sites of the Mycobacterium tuberculosis beta-lactamase K73A and
E166A mutants,” Biochemistry 49, 9685–9687 (2010).
37