Measuring kinetic isotope effects in enzyme reactions using time-resolved electrospray mass spectrometry

Article abstract of DOI:10.1021/ac400191t

Kinetic isotope effect (KIE) measurements are a powerful tool for studying enzyme mechanisms; they can provide insights into microscopic catalytic processes and even structural constraints for transition states. However, KIEs have not come into widespread use in enzymology, due in large part to the requirement for prohibitively cumbersome experimental procedures and daunting analytical frameworks. In this work, we introduce time-resolved electrospray ionization mass spectrometry (TRESI-MS) as a straightforward, precise, and inexpensive method for measuring KIEs. Neither radioisotopes nor large amounts of material are needed and kinetic measurements for isotopically labeled and unlabeled species are acquired simultaneously in a single competitive assay. The approach is demonstrated first using a relatively large isotope effect associated with yeast alcohol dehydrogenase (YADH) catalyzed oxidation of ethanol. The measured macroscopic KIE of 2.19 ± 0.05 is consistent with comparable measurements in the literature but cannot be interpreted in a way that provides insights into isotope effects in individual microscopic steps. To demonstrate the ability of TRESI-MS to directly measure intrinsic KIEs and to characterize the precision of the technique, we measure a much smaller ¹²C/ ¹³C KIE associated specifically with presteady state acylation of chymotrypsin during hydrolysis of an ester substrate.

Full text of DOI:10.1021/ac400191t

Article
pubs.acs.org/ac
Measuring Kinetic Isotope Effects in Enzyme Reactions Using Time-
Resolved Electrospray Mass Spectrometry
Peter Liuni, Ekaterina Olkhov-Mitsel,§ Arturo Orellana, and Derek J. Wilson*
Department of Chemistry, York University, Toronto, ON, M3J 1P3, Canada
S
* Supporting Information
ABSTRACT: Kinetic isotope effect (KIE) measurements are a powerful
tool for studying enzyme mechanisms; they can provide insights into
microscopic catalytic processes and even structural constraints for
transition states. However, KIEs have not come into widespread use in
enzymology, due in large part to the requirement for prohibitively
cumbersome experimental procedures and daunting analytical frameworks.
In this work, we introduce time-resolved electrospray ionization mass
spectrometry (TRESI-MS) as a straightforward, precise, and inexpensive
method for measuring KIEs. Neither radioisotopes nor large amounts of
material are needed and kinetic measurements for isotopically “labeled”
and “unlabeled” species are acquired simultaneously in a single
“competitive” assay. The approach is demonstrated first using a relatively
large isotope effect associated with yeast alcohol dehydrogenase (YADH)
catalyzed oxidation of ethanol. The measured macroscopic KIE of 2.19 0.05 is consistent with comparable measurements in
the literature but cannot be interpreted in a way that provides insights into isotope effects in individual microscopic steps. To
demonstrate the ability of TRESI-MS to directly measure intrinsic KIEs and to characterize the precision of the technique, we
measure a much smaller 12C/13C KIE associated specifically with presteady state acylation of chymotrypsin during hydrolysis of
an ester substrate.
effects associated with hydride transfer and a well justified
inetic isotope effects (KIEs) have a broad range of uses in
Kthe study of enzyme mechanisms, providing insights that interest in hydrogen tunneling but also because reduction/
cannot be obtained by other techniques.1−4 KIEs have been
oxidation of the nicotinate moiety is optically detectable
employed to determine rate limiting microscopic steps,5−7 link
(NADH has a characteristic absorption band at λ = 340
nm).6,18−20 Optical analysis by UV/visible absorption or
fluorescence also generally precludes competitive assays, since
isotopically “labeled” and “unlabeled” species are indistinguish-
able. More recently, MS- and NMR-based methods have
emerged that address these limitations somewhat.21,22 NMR
approaches that allow for the measurement of KIEs from
natural abundance 13C are especially promising since they do
not require isotopic labeling.23−25 However, these techniques
bring their own drawbacks which can include the requirement
for a large amount of material,23,25 limited applicability, and
lower precision quantitation of reactants and products.
catalysis to dynamics in the active site,8−11 and structurally
characterize transition states.12−14 However, in spite of their
obvious usefulness and a long history development, KIE
measurements have not come into widespread use in the
broader field of enzymology. There are a number of potential
reasons for this, including analytical frameworks that can appear
quite daunting even for moderately complex reaction
mechanisms, and the ambiguity arises from drawing con-
clusions about microscopic processes from what are usually
steady state measurements.15,16 This is in addition to the
experimental procedure themselves, which are often arduous
and expensive.
The ability to distinguish multiple isotopes simultaneously at
high sensitivity makes mass spectrometry well suited to the
study of isotope effects, as exemplified by the well-established
field of isotope ratio mass spectrometry.26 MS-based
approaches have been used to measure KIEs in enzyme
systems, typically by quenching the reaction midphase and
noting the relative quantities of specifically labeled and
unlabeled product.21,22 This approach is in principle broadly
Scintillation counting was initially the dominant method for
measuring KIEs, but the requirement for radioisotopes has
resulted in a marked decrease in use. Currently, radiolabeling is
used mainly for hydrogen tunneling analyses that compare
1H/3H and 2H/3H isotope effects, as well as competitive assays
for heavy atom KIEs.17 The majority of KIE measurements are
made using optical techniques, but this has severely limited the
subset of systems studied due to the requirement for a
chromophoric change on turnover. For example, NAD(P)+
dependent enzymes represent the overwhelming majority of
KIE studies in the literature, partly because of the large isotope
Received: January 18, 2013
Accepted: March 5, 2013
Published: March 5, 2013
© 2013 American Chemical Society
3758
dx.doi.org/10.1021/ac400191t | Anal. Chem. 2013, 85, 3758−3764

Analytical Chemistry
Article
applicable but relies on a single time-point measurement and
numerous sample handling steps between quenching and
analysis, which can make it difficult to achieve the needed level
of precision.
In this work, we introduce a new, straightforward approach
for measuring KIEs based on time-resolved electrospray
ionization mass spectrometry (TRESI-MS). TRESI-MS is an
electrospray-coupled rapid mixing technique that enables
monitoring of solution phase processes on the ms time
scale.27 The data are analogous to conventional stopped-flow
intensity−time profiles except that no chromophore is required
and isotopically labeled and unlabeled species are distinguish-
able by mass. To establish the feasibility of the approach, we
first measure a large, well-studied KIE in the yeast alcohol
dehydrogenase (YADH)-catalyzed hydride transfer from
ethanol to NAD+. We then test the limits of the technique
by measuring a small 12C/13C isotope effect associated with the
presteady state acylation of chymotrypsin by the ester substrate
para-nitrophenyl acetate (pNPA).
Figure 1. Schematic depiction of the TRESI-MS source.27 Solution
containing enzyme is injected into the “inner capillary” from Syringe 1
(green) via infusion pump. This pump, Syringe 1, and the inner
capillary comprise an assembly that can be withdrawn from the source
region (gray arrow), resulting in withdrawal of the inner capillary from
the end of the outer capillary and a correspondingly increased delay
between mixing and ionization. Syringe 2 (red) injects substrate
solution through the annular intercapillary space. Mixing occurs when
solution from the inner capillary flows into the intercapillary space
though a notch cut 2 mm from the end of the inner capillary. Freshly
mixed solution then passes into a delay volume which is determined by
the position of the inner capillary within the outer capillary and then
into a static mixer where electrospray-enhancing “makeup solvent” is
added (pink syringe) immediately prior to the onset of ESI.
EXPERIMENTAL SECTION
■
Materials. α-Chymotrypsin, yeast alcohol dehydrogenase,
β-nicotinamide adenine dinucleotide, ammonium hydroxide,
ammonium acetate, and 4-nitrophenol acetate (p-NPA) were
obtained from Sigma-Aldrich (St. Louis, MO). Deuterium
oxide, 1,1′-13C acetic anhydride (99%), and 1,1-D2 ethanol
were purchased from Cambridge Isotope Laboratories (And-
over, MA). HPLC-grade methanol, ethanol, acetonitrile, and
hydrochloric acid were purchased from Caledon Laboratories
(Georgetown, ON, Canada). Ultrapure water was generated
from the in-house Milli-Q system (EMD Millipore, Billerica,
MA). All solutions were pH adjusted using the S20 SevenEasy
pH meter (Denver Instruments, Bohemia, NY). For all
experiments, infusion and continuous pullback of the inner
capillary was conducted using Harvard 11+ infusion syringe
pumps (Holliston, MA, USA).
inner capillary back from the end of the outer capillary,
resulting in the acquisition of steadily later reaction times. For
chymotrypsin experiments, an additional static mixer was added
to the end of the outer capillary in order to supply an
electrospray-enhancing “makeup solvent” consisting of 10% (v/
v) methanol and 6 mM HCl (pH 2) in water immediately prior
to ionization. Kinetics were measured from 42 ms to 22 s of
reaction time, corresponding to 30 min of inner capillary
withdrawal and data acquisition.
KIE Measurements. For all experiments, isotopically
labeled and unlabeled substrates were mixed and reacted
simultaneously in an internal competition assay. In alcohol
dehydrogenase experiments, Syringe 1 contained a 2 μM
solution of YADH in 10 mM ammonium acetate (pH 8.4).
This was mixed with a series of equimolar ethanol and 1,1-D2
ethanol solutions, (10 mM-200 mM) each containing 400 μM
NAD+ prepared in 10 mM ammonium acetate (pH 8.4).
Samples were injected at a total flow rate of 3.0 μL/min. The
intensity of NADH and NADD at 666 m/z and 667 m/z,
respectively, was monitored as a function of time, and kinetic
data was acquired from extracted ion currents at their respective
masses. For the purpose of extracting V0, the 667 m/z signal
was assumed to contain a 20% component from NADH due to
the M + 1 isotopic peak.
Synthesis of 13C-Labeled pNPA. To a solution of p-
nitrophenol (1.0 equiv.) in freshly distilled dichloromethane
was added a catalytic amount of dimethylaminopyridine (0.05
equiv.) in one portion, followed by a slight excess of acetic
anhydride (1.05 equiv.). The reaction was monitored by thin-
layer chromatography (TLC) analysis. Once complete, the
reaction was diluted with diethyl ether and extracted with
water. The ether layer was washed with brine, dried with
MgSO4, and concentrated under vacuum. The crude material
was purified by column chromatography using silica gel. The
final product was characterized by mass spectrometry and UV−
vis spectroscopy. 13C-labeled labeled p-nitrophenyl acetate was
prepared using the same procedure and using 1,1′-13C2-labeled
acetic anhydride.
TRESI-MS. Time-resolved measurements were carried out
using an electrospray mass spectrometry-coupled capillary
mixer described previously (Figure 1).27,28 Briefly, the device
is based on a concentric capillary design. In both YADH and
chymotrypsin experiments, the enzyme solution was injected
through the polyimide-coated fused silica inner capillary
(Polymicro, Pheonix, AZ). Solutions containing 50% labeled
and unlabeled substrate were injected through the stainless
steel outer capillary (Small Parts, Logansport, IN). Mixing
occurred when enzyme solution was released into the
intercapillary space from a notch cut 2 mm from the inner
capillary end, with the mixed solution then passing into a delay
volume. Reaction profiles were acquired by steadily moving the
For chymotrypsin experiments, a 25 μM chymotrypsin
solution at pH 8.4 is loaded into Syringe 1. A solution
containing both 2.5 mM 12C-p-NPA and 2.5 mM 13C-p-NPA
was prepared in 40% methanol, adjusted to pH 7.0, and loaded
into Syringe 2. A 6 mM HCl solution in 10% methanol is
loaded into Syringe 3 as a makeup solvent. Samples were
injected at flow rates of 3.5 μL/min for all 3 syringes (10.5 μL/
3759
dx.doi.org/10.1021/ac400191t | Anal. Chem. 2013, 85, 3758−3764

Analytical Chemistry
Article
Figure 2. Schematic depiction of the semisequential bi-bi YADH mechanism proposed by Dickinson and Monger in 1973, with release of NADH
being the rate-limiting process in the oxidation of ethanol (EtOH) to acetaldehyde (Al).
Figure 3. Reaction progress curves for the oxidation of ethanol by YADH at two different ethanol concentrations (a) 200 mM and (b) 10 mM,
acquired by monitoring the accumulation of NAD(H/D). Dark circles represent the NADH signal intensity, while gray circles correspond to NADD.
Lines are single exponential fits to the data to aid visualization (not used in the analysis). Insets show the linear portions of the progress curves used
for the extraction of V0.
min total flow rate). The makeup solvent also acts as a chemical
quencher for the reaction; therefore, the dead time for the
mixing tee is negligible. Extracted ion current profiles from
peaks corresponding to the 12+, 11+, 10+, and 9+ charge states
of the free and acyl-enzyme forms of chymotrypsin,
respectively, were acquired and used in the calculation of KIEs.
All experiments were conducted with mild-moderate
declustering potentials on a QSTAR Elite Qq-TOF mass
spectrometer (AB Sciex, Framingham, MA). ESI voltages were
optimized between +4500 and +5500 V (positive ion mode)
prior to the acquisition of data. The Analyst QS 2.0 software
package was used to control, acquire, and view mass spectral
information, and protein mass spectra were deconvoluted using
the BioAnalyst software suite (AB Sciex, Framingham, MA).
catalyzed hydride transfer from benzyl alcohol to NAD+ was the
first reaction in which evidence of hydrogen tunneling was
reported.31 There remains nonetheless substantial disagreement
on the extent to which hydrogen tunneling can be inferred from
the conventional approach of characterizing the pressure or
temperature dependence of the KIE.34 The reliability of KIE-
based evidence for the contribution of enzyme dynamics to
catalysis has also recently come into question.16 In the case of
YADH, much of the difficulty in interpreting the hydride
transfer primary KIE arises from ambiguities in our under-
standing of the mechanism. Recent studies generally assume the
semisequential bi-bi mechanism proposed by Dickinson and
Monger in 1973 (Figure 2), with the rate limiting step in the
“oxidation-of-alcohol” direction being release of the nucleotide
product.35
RESULTS AND DISCUSSION
YADH has frequently been used as a protein complex
standard for electrospray mass spectrometry.36,37 Under
optimal electrospray conditions, the YADH spectrum is
dominated by the intact tetramer, shown in Figure S1,
Supporting Information. The addition of saturating concen-
■
Yeast Alcohol Dehydrogenase Oxidation of Ethanol.
KIEs associated with hydride transfer are among the most well-
studied isotope effects, serving as a model for hydrogen
tunneling29−31 and catalysis-linked dynamics.32,33 The YADH-
3760
dx.doi.org/10.1021/ac400191t | Anal. Chem. 2013, 85, 3758−3764

Analytical Chemistry
Article
trations of ethanol has little impact on the quality of the
spectrum, and specific binding to the tetramer is detected via a
184 Da mass shift (i.e., 1 molecule per active site). However,
mM concentrations of NAD+ have a strongly deleterious effect
on spectral quality, eliminating the possibility of directly
monitoring active enzyme complexes in the steady state.
Kinetic measurements were therefore carried out by monitoring
the product NAD(H/D), with adjustment to the extracted
parameters for the ∼20% contribution of the NADH M + 1
(13C) peak to the monoisotopic NADD peak. Typical reaction
progress curves from competition experiments involving 50%
labeled ethanol (CH3CD2OH) at 10 mM (a) and 200 mM (b)
are shown in Figure 3. The insets show the linear portions of
the curves used to extract initial velocities V0 for a steady state
analysis.
YADH mechanism is too complex to allow for the
determination of specifically which processes are affected
from a single macroscopic measurement. Moreover, there are
a number of other factors that complicate any attempt at a
quantitative interpretation of the observed KIE, in particular an
undefined 2° KIE from the deuterium that is not transferred
and an untested assumption that catalysis at each active site is
fully independent. Random binding of labeled and unlabeled
substrate at each active site generates a heterogeneous set of
active enzyme complexes, which could distort the observed KIE
if the activity of one site is influenced by the activity of another.
It is worth noting that some of this ambiguity arises from an
inability to directly detect enzyme complexes that are populated
in the steady state (i.e., the E·NAD(H/D) complex, where E =
enzyme) or the presteady state (e.g., the ternary
E·NAD+·CH3C(H/D)2OH complex). In the case of YADH,
the detection of active enzyme complexes is complicated by the
presence of NAD+ as discussed earlier; however, this challenge
may prove surmountable with careful adjustment of solvent and
ionization conditions. The challenge of distinguishing labeled
from unlabeled complexes in competition experiments, on the
other hand (i.e., a 2−8 Da difference on a roughly 148 kDa
complex), would be daunting even on an ultrahigh resolution
instrument. Nonetheless, for smaller proteins, direct measure-
ment of microscopic KIEs via enzyme complex intermediates is
possible using TRESI-MS, as demonstrated in the following
section.
As expected, these data show a significant KIE, with hydride
transfer from the labeled ethanol to form NADD being
substantially slower than the formation of NADH from the
unlabeled substrate. The average “observed” kinetic isotope
effect KIEobs extracted directly from these data is V0(NADH)
/
V0(NADD) = 1.8 0.4; however, this analysis assumes that the
isotopic substitution has no influence on any microscopic
process apart from the “rate limiting” step, which seems
unlikely given that hydride transfer (where the isotopic
substitution generates a primary KIE) is not rate limiting.38
The observed KIE is therefore a macroscopic phenomenon
which is better defined in terms of the apparent specificity
Primary 12C/13C KIE in Chymotrypsin Acylation. In
contrast to KIEs involving isotopes of protium, heavy atom
KIEs rarely exceed more than a few percent and are therefore
much more challenging to measure in terms of achieving the
necessary level of precision. Chymotrypsin catalyzed hydrolysis
of para-nitrophenyl acetate (mechanism shown in Figure S3,
Supporting Information) is among the relatively few enzymatic
reactions for which a heavy atom KIE has been reported.40 This
reaction also includes an observable presteady state charac-
terized by the population of an acyl-enzyme intermediate.
Acylation of the enzyme is fully rate limiting in the presteady
state, and isotopic substitution at the carbonyl carbon has no
impact on the substrate binding equilibrium.40 The primary
12C/13C KIE associated with acyl-enzyme formation is therefore
attributable specifically to the irreversible acylation step.
The release of chromophoric para-nitrophenolate upon
acylation makes the chymotrypsin/p-NPA presteady state
accessible to optical analysis; however, the optical equivalency
of labeled and unlabeled para-nitrophenol rules out a
competition approach for measuring KIEs. By TRESI-MS,
presteady state acylation of chymotrypsin is monitored by
direct detection of the acyl-enzyme intermediate.28 Competi-
tion experiments are therefore possible in principle but will
require the differentiation of a 1 Da difference on a 25 kDa
protein, complicated still further by overlapping natural
abundance 13C isotope distributions. In terms of fwhm
resolution, this requirement is well beyond the capabilities of
a typical Q-TOF instrument; however, a competitive KIE can
still be measured provided that the “labeled” and “unlabeled”
peaks are distinguishable and the extent of overlap between the
peaks is known. To determine if the labeled and unlabeled
enzyme complex were distinguishable, we collected mass
spectra of chymotrypsin in the presence of unlabeled (Figure
5a) and labeled substrate (Figure 5b). In both cases, the free
constant: KIEspec = (kcat(NADH)/KM(NADH))/(kcat(NADD)
/
15
D
KM(NADD)), usually abbreviated as (V/K).
Michealis−Menten plots showing the initial rates of
formation of NADH and NADD as a function of ethanol
concentration are provided in Figure 4. With the NAD+
Figure 4. Steady state analysis of YADH catalyzed oxidation of
ethanol. Filled squares represent the initial velocities of NADH
production, and open circles indicate the same for NADD. After fitting
to the Michealis−Menten equation, the data yield a D(V/K) value of
2.19 0.05. Error bars are from 5 replicates.
concentration held constant at a saturating level of 20 mM,
the V0 vs. [ethanol] profile exhibits classical Michealis−Menten
saturation kinetics. Extracted values of kcat and KM yield a
KIEspec of 2.19 0.05 (R2 NADH = 0.82, R2 NADD = 0.88),
which is in line with comparable literature values, including the
initial measurement by Mahler and Douglas2 and more recent
work by Park et al.39 The significant difference between KIEobs
and KIEspec suggests a substantial shift in one or more
microscopic equilibria upon isotopic substitution, but the
enzyme appears at 25446
2 Da after deconvolution,
corresponding to the δ′ form of the enzyme.28 The mass of
3761
dx.doi.org/10.1021/ac400191t | Anal. Chem. 2013, 85, 3758−3764

Analytical Chemistry
Article
Figure 5. Matched pair of deconvoluted mass distributions for labeled and unlabeled acyl-chymotrypsin. (a) unlabeled; (b) labeled. The peak at 25
446 Da corresponds to free δ′-chymotrypsin while the acyl-enzyme is observed at 25 488 Da in the case of the unlabeled substrate and 25 489 Da in
the case of the labeled substrate. The spectra also include a number of sodium and potassium adducts common in commercially supplied samples.
(c) A model of the 10+ m/z peak for labeled and unlabeled acyl-chymotrypsin (50/50 mixture) illustrating the challenge of differentiating the two
peaks at the m/z level due to overlap. Peaks for 12C are shown in black, while the 13C peaks are shown in gray. Small peaks appearing above and
below the m/z axis are difference peaks (12C−13C). Gray areas indicate the ion current extraction window for 12C acyl-enzyme (left) and 13C acyl-
enzyme (right), resulting in a 60% contribution from the desired species.
the acyl-enzyme, however, was 25488 3 Da for the unlabeled
substrate and 25489 3 Da for the labeled substrate. In terms
of absolute mass, these values are identical to within error;
however, in each matched set of measurements (n = 5),
acylation with labeled substrate resulted in a mass difference
that was always 1 Da more than acylation with unlabeled
substrate. This is a testament to the power of deconvolution
over multiple peaks to measure mass differences in the absence
of an internal standard, and it confirms that there is a
measurable difference in peak position between labeled and
unlabeled acyl-chymotrypsin.
The challenge in differentiating the labeled and unlabeled
acyl-enzyme in individual m/z peaks is illustrated in Figure 5c,
in which the labeled and unlabeled 10+ peaks are modeled at
100 000 fwhm. Kinetics taken from overlapping regions of the
distribution will yield a weighted average rate for the two
species (see Supporting Information). It is therefore desirable
to acquire kinetics from regions of the peak where overlap is
lowest, corresponding to the outer edges of the distribution.
However, this must be balanced with the need for sufficient
signal-to-noise. Ultimately, a region corresponding to 60%
contribution from the desired species (Figure 5c, shaded
regions) was selected. Effectively, this means that the observed
KIE will be only 20% of the actual KIE (see Supporting
Information for details), which makes for a very challenging
measurement. Nonetheless, the enhanced precision of measur-
ing the KIE in a single competitive assay appears to outweigh
the challenges; TRESI-based experiments in which the labeled
and unlabeled rates were measured separately failed to yield a
reliable KIE (i.e., the measurement error was greater than the
measurement; data not shown).
Extracted ion currents (XICs) from the appropriate regions
of the m/z peaks yield intensity−time profiles dominated by
either 12C- or 13C-acylated chymotrypsin as described above
(also see methods and Figure S2, Supporting Information).
After normalization to the total ion current (TIC) to remove
artifacts from sensitivity drift, and averaging over all m/z peaks
with sufficient signal-to-noise, we obtain low-dispersion kinetic
profiles, an example of which is shown in Figure 6a. To ensure
that the observed small rate difference was not the result of
systematically changing peak shape or interference from nearby
peaks, we performed control experiments in which an identical
procedure was used to generate “left-side-of-peak” and “right-
side-of-peak” kinetic profiles from samples containing only
unlabeled substrate, shown in Figure 6b. These control profiles
were identical to within error.
Least-squares fits to the data using a single exponential
expression yielded an average 12C acylation rate of 1.10 0.21
s−1 and an average 13C acylation rate of 1.06 0.19 s−1, based
on five independent runs. The error associated with these
independent measurements is too great to allow for the
extraction of a reliable KIE; however, this is largely a
consequence of run-to-run variability in the absolute rates.
Since in a competition experiment KIEs are calculated from
matched rates (i.e., paired 12C and 13C rates extracted from
individual runs), the issue of run-to-run variability is eliminated,
resulting in a substantial decrease in the error. After adjusting
for peak overlap, the average KIE derived from these
3762
dx.doi.org/10.1021/ac400191t | Anal. Chem. 2013, 85, 3758−3764

Analytical Chemistry
Article
details. This material is available free of charge via the Internet
at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*Phone: +1 416-736-2100 x 20786. E-mail: dkwilson@yorku.
ca.
Present Address
§E.O.-M.: Samuel Lunenfeld Research Institute, Mount Sinai
Hospital, 60 Murray St., Toronto, ON, Canada M5G 1X5.
Figure 6. Intensity−time profiles drawn from acyl-chymotrypsin ion
currents extracted as shown in Figure 4. Filled squares represent
intensities taken from the right side of the peak; open triangles
correspond to the intensity on the left side. (a) A typical 12C/13C
profile for KIE measurement yielding an intrinsic acylation KIEobs of
1.017 0.008, which is 20% of the actual value due to peak overlap
(details on the data analysis are provided in the Supporting
Information). (b) A 12C/12C negative control to ensure that KIE
measurements are not an artifact from time-dependent changes in
peak shape. The data are fit using least-squares to a single exponential
expression yielding a “KIE” of 1.00 0.005.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Natural Sciences and
Engineering Council of Canada (NSERC) Discovery Grant
program and an Ontario Ministry of Research and Innovation
Early Researcher Award.
REFERENCES
■
experiments (n = 5) is 1.09
0.02. To the best of our
(1) Northrop, D. B. Methods 2001, 24, 117−124.
(2) Mahler, H. R.; Douglas, J. J. Am. Chem. Soc. 1957, 79, 1159−
1166.
knowledge, this represents the first competitive heavy-atom
KIE measurement in the presteady state. While there are no
directly comparable measurements in the literature, this value is
within the range determined for deacylation of chymotrypsin
using a single-turnover approach (≤1.08, depending on ethanol
concentration).41 The value is significantly higher than the
1.030 0.002 steady state 13(V/K) value reported by Hess and
co-workers;40 however, that measurement would have been a
convolution of the primary KIEs for acylation and deacylation
and would also have been influenced by any change in the ratio
of the acylation/deacylation rates resulting from isotopic
substitution. The magnitude of our acylation-specific KIE
measurement indicates a late transition state for loss of para-
nitrophenolate from the tetrahedral acyl-enzyme complex,
consistent with strong stabilization of the oxyanion.
(3) Whittaker, M. M.; Ballou, D. P.; Whittaker, J. W. Biochemistry
1998, 37, 8426−8436.
(4) Papajak, E.; Kwiecien, R. A.; Rudzinski, J.; Sicinska, D.; Kaminski,
R.; Szatkowski, L.; Kurihara, T.; Esaki, N.; Paneth, P. Biochemistry
2006, 45, 6012−6017.
(5) Nesheim, J. C.; Lipscomb, J. D. Biochemistry 1996, 35, 10240−
10247.
(6) Rendina, A. R.; Hermes, J. D.; Cleland, W. W. Biochemistry 1984,
23, 6257−6262.
(7) Kim, K. H.; Isin, E. M.; Yun, C. H.; Kim, D. H.; Guengerich, F. P.
FEBS J. 2006, 273, 2223−2231.
(8) Knapp, M. J.; Klinman, J. P. Eur. J. Biochem. 2002, 269, 3113−
3121.
(9) Sikorski, R. S.; Wang, L.; Markham, K. A.; Rajagopalan, P. T. R.;
Benkovic, S. J.; Kohen, A. J. Am. Chem. Soc. 2004, 126, 4778−4779.
(10) Hay, S.; Sutcliffe, M. J.; Scrutton, N. S. Proc. Natl. Acad. Sci. U. S.
A. 2007, 104, 507−512.
(11) Swanwick, R. S.; Maglia, G.; Tey, L.; Allemann, R. K. Biochem. J.
2006, 394, 259−265.
(12) Schramm, V. L.; Horensteins, B. A.; Kline, P. C. J. Biol. Chem.
1994, 269, 18259−18262.
CONCLUSIONS
■
We have demonstrated a novel TRESI-MS-based approach for
the measurement of kinetic isotope effects in enzyme catalyzed
reactions. The main advantage of TRESI-MS in this context is
that is allows for competition experiments in a broad range of
systems without the need for radioactive labeling or large
amounts of material. Interpretation of the data may also be
simplified in many cases, particularly when an enzyme complex
is directly observable. Finally, by enabling competition
experiments in the presteady state, TRESI-MS greatly increases
the number of enzymatic reactions in which it is possible to
measure microscopic KIEs (i.e., KIEs that are attributable to a
single step in the reaction mechanism). The competition
approach was found to be essential in achieving the necessary
level of precision for heavy atom KIE measurements. In
summary, TRESI-MS represents a powerful alternative for
measuring KIEs, with the potential to yield new insights into
enzymatic mechanisms, transition states, and catalysis-linked
dynamics.
(13) Schramm, V. L. Curr. Opin. Struct. Biol. 2005, 15, 604−613.
(14) Berti, P. J. Methods Enzymol. 1999, 308, 355−397.
(15) Northrop, D. B. Biochemistry 1975, 14, 2644−2651.
(16) Glowacki, D. R.; Harvey, J. N.; Mulholland, A. J. Biochem. Soc.
Trans. 2012, 40, 515−521.
(17) McCann, J. A. B.; Berti, P. J. J. Am. Chem. Soc. 2007, 129, 7055−
7064.
(18) Kohen, A. Prog. React. Kinet. Mech. 2003, 28, 119−156.
(19) Sutcliffe, M. J.; Masgrau, L.; Roujeinikova, A.; Johannissen, L.
O.; Hothi, P.; Basran, J.; Ranaghan, K. E.; Mulholland, A. J.; Leys, D.;
Scrutton, N. S. Philos. Trans. R. Soc., B: Biol. Sci. 2006, 361, 1375−
1386.
(20) Klinman, J. P. Biochemistry 1976, 15, 2018−2026.
(21) Bates, C.; Kendrick, Z.; McDonald, N.; Kline, P. C.
Phytochemistry 2006, 67, 5−12.
(22) Hunt, C.; Gillani, N.; Farone, A.; Rezaei, M.; Kline, P. C. BBA-
Proteins Proteomics 2005, 1751, 140−149.
ASSOCIATED CONTENT
* Supporting Information
Raw ESI mass spectrum of native YADH (Figure S1). Data
acquisition and handling procedure (Figure S2). Data analysis
■
(23) Brecker, L.; Kogl, M. F.; Tyl, C. E.; Kratzer, R.; Nidetzky, B.
Tetrahedron Lett. 2006, 47, 4045−4049.
S
(24) Chan, J.; Lewis, A. R.; Gilbert, M.; Karwaski, M. F.; Bennet, A. J.
Nat. Chem. Biol. 2010, 6, 405−407.
3763
dx.doi.org/10.1021/ac400191t | Anal. Chem. 2013, 85, 3758−3764

Analytical Chemistry
Article
(25) Singleton, D. A.; Thomas, A. A. J. Am. Chem. Soc. 1995, 117,
9357−9358.
(26) Werner, R. A.; Brand, W. A. Rapid Commun. Mass Spectrom.
2001, 15, 501−519.
(27) Wilson, D. J.; Konermann, L. Anal. Chem. 2003, 75, 6408−6414.
(28) Wilson, D. J.; Konermann, L. Anal. Chem. 2004, 76, 2537−2543.
(29) Powell, M. F.; Bruice, T. C. J. Am. Chem. Soc. 1983, 105, 7139−
7149.
(30) Northrop, D. B.; Cho, Y. K. Biochemistry 2000, 39, 2406−2412.
(31) Cha, Y.; Murray, C. J.; Klinman, J. P. Science 1989, 243, 1325−
1330.
(32) Bahnson, B. J.; Colby, T. D.; Chin, J. K.; Goldstein, B. M.;
Klinman, J. P. Proc. Natl. Acad. Sci. U. S. A. 1997, 94, 12797−12802.
(33) Kohen, A.; Cannio, R.; Bartolucci, S.; Klinman, J. P. Nature
1999, 399, 496−499.
(34) Hay, S.; Johannissen, L. O.; Hothi, P.; Sutcliffe, M. J.; Scrutton,
N. S. J. Am. Chem. Soc. 2012, 134, 9749−54.
(35) Dickinson, F. M.; Monger, G. P. Biochem. J. 1973, 131, 261−
270.
(36) Zhang, H.; Cui, W. D.; Wen, J. Z.; Blankenship, R. E.; Gross, M.
L. J. Am. Soc. Mass Spectrom. 2010, 21, 1966−1968.
(37) Loo, J. A. J. Mass Spectrom. 1995, 30, 180−183.
(38) Klinman, J. P. J. Biol. Chem. 1972, 247, 7977−7987.
(39) Park, H.; Kidman, G.; Northrop, D. B. Arch. Biochem. Biophys.
2005, 433, 335−340.
(40) Hess, R. A.; Hengge, A. C.; Cleland, W. W. J. Am. Chem. Soc.
1998, 120, 2703−2709.
(41) Mishra, A. K.; Klapper, M. H. Biochemistry 1986, 25, 7328−
7336.
3764
dx.doi.org/10.1021/ac400191t | Anal. Chem. 2013, 85, 3758−3764