In situ and multisubstrate detection of elastase enzymatic activity external to microdialysis sampling probes using LC-ESI-MS

Article abstract of DOI:10.1021/ac702047w

Extracellular proteases play significant roles in mammalian development and disease. Enzymatic activity external to a microdialysis sampling probe can be determined by infusing judicious choices of substrates followed by collecting and measuring the products. Porcine pancreatic elastase was used as a model enzyme with two substrates possessing different cleavage sites, N-methoxysuccinyl-Ala-Ala-Pro-Val-7-amino-4-methylcoumarin (FL-substrate) and N-succinyl-Ala-Ala-Ala-p-nitroanilide (UV-substrate). These substrates were infused through the microdialysis sampling probe to a solution containing elastase. The resulting four products and the remaining two substrates were collected into the dialysate and were subsequently analyzed off-line using liquid chromatography-mass spectrometry (LC-MS) with electrospray ionization (ESI). All analytes were identified using extracted ion chromatograms of m/z 628 (FL-substrate), m/z 452 (UV-substrate), m/2 471 (N-methoxysuccinyl-Ala-Ala-Pro- Val, FL-NTP), m/z 332 (N-succinyl-Ala-Ala-Ala, UV-NTP), m/z 176 (7-amino-4-methylcoumarin, AMC), and m/z 139 (p-nitroaniline, pNA). FL-NTP and FL-substrate exhibited 10-fold higher ion production as compared to AMC with equimolar standards. Microdialysis sampling combined with LC-ESI-MS detection allowed for in situ determination of the enzymatic activity of a protease external to the microdialysis probe when using different peptide-based substrates.

Full text of DOI:10.1021/ac702047w

Anal. Chem. 2008, 80, 2050-2057
In Situ and Multisubstrate Detection of Elastase
Enzymatic Activity External to Microdialysis
Sampling Probes Using LC-ESI-MS
Ying Wang, Dmitri V. Zagorevski, and Julie A. Stenken*,†
Department of Chemistry and Chemical Biology, Center for Biotechnology and Interdisciplinary Studies, Rensselaer
Polytechnic Institute, Troy, New York 12180
receptors, and cell-cell and cell-matrix adhesion molecules.3
MMPs have known involvement with numerous disease states
including arthritis, cancer, and cardiovascular disease. Elucidating
the roles that MMPs play in vivo via chemical analysis is
challenging since their endogenous substrates are typically
insoluble proteins. The multiplicity of MMP substrates makes it
difficult to effectively measure the activity of a single enzyme.4
Additionally, once MMPs cleave their targeted substrate, the
resulting product can be cleaved by additional MMPs or extra-
cellular proteases. Many analytical methods have been described
to assay the clinically relevant MMPs ex vivo from tissue biopsies.5
Immunochemical methods give quantitative information of total
enzyme concentrations, but they cannot discriminate between
zymogens and active enzymes.6 MMPs are synthesized as inactive
zymogens (inactive proteins) and require enzymatic activation that
is tightly regulated via tissue inhibitors of metalloproteinases
(TIMPs). Zymography is a common method to separate and
quantify the concentrations between the proenzyme and the active
enzyme.7 Although zymography allows visualization of enzyme
activities, it is a tedious method and is difficult to quantify.8 A
variety of fluorescence and colorimetric assays have been applied
to quantify the active enzyme based on changes in spectroscopic
properties of the resulting products after substrate cleavage.9-13
Few methods are available to determine in vivo MMP activity.
Assessing MMP enzymatic activity is important for identification
and localization of activated MMPs during various disease states.
Additionally, such assays would be useful in pharmaceutical
studies designed to inhibit targeted MMPs. Considerable efforts
have been made in the areas of fluorescent-based imaging
Extracellular proteases play significant roles in mam-
malian development and disease. Enzymatic activity ex-
ternal to a microdialysis sampling probe can be deter-
mined by infusing judicious choices of substrates followed
by collecting and measuring the products. Porcine pan-
creatic elastase was used as a model enzyme with two
substrates possessing different cleavage sites, N-meth-
oxysuccinyl-Ala-Ala-Pro-Val-7-amino-4-methylcoumarin
(FL-substrate) and N-succinyl-Ala-Ala-Ala-p-nitroanilide
(UV-substrate). These substrates were infused through the
microdialysis sampling probe to a solution containing
elastase. The resulting four products and the remaining
two substrates were collected into the dialysate and were
subsequently analyzed off-line using liquid chromatogra-
phy-mass spectrometry (LC-MS) with electrospray ion-
ization (ESI). All analytes were identified using extracted
ion chromatograms of m/z 628 (FL-substrate), m/z 452
(UV-substrate), m/z 471 (N-methoxysuccinyl-Ala-Ala-Pro-
Val, FL-NTP), m/z 332 (N-succinyl-Ala-Ala-Ala, UV-NTP),
m/z 176 (7-amino-4-methylcoumarin, AMC), and m/z
139 (p-nitroaniline, pNA). FL-NTP and FL-substrate
exhibited 10-fold higher ion production as compared to
AMC with equimolar standards. Microdialysis sampling
combined with LC-ESI-MS detection allowed for in situ
determination of the enzymatic activity of a protease
external to the microdialysis probe when using different
peptide-based substrates.
Extracellular proteinases play essential roles in the degradation
of extracellular proteins allowing for multicellular organs to
develop and function normally.1 Matrix metalloproteinases (MMPs),
a family of zinc-dependent proteinases, have been recognized for
their ability to cleave extracellular matrix (ECM) structural
proteins including collagen, laminin, and fibronectin.2 MMPs
regulate cell behavior via cleavage of cell surface molecules and
other pericellular nonmatrix proteins including proteinases, pro-
teinase inhibitors, clotting factors, chemotactic molecules, latent
growth factors, growth factor binding proteins, cell surface
(3) Sternlicht, M. D.; Werb, Z. Annu. Rev. Cell Dev. Biol. 2001, 17, 463-516.
(4) Lombard, C.; Saulnier, J.; Wallach, J. Biochimie 2005, 87, 265-272.
(5) Matrix Metalloproteinase Protocols Methods in Molecular Biology; Clark, I.
M., Ed.; Humana Press: NJ, 2001; Vol. 151.
(6) Capper, S. J.; Verheijen, J.; Smith, L.; Sully, M.; Visser, H.; Hanemaaijer, R.
Ann. N.Y. Acad. Sci. 1999, 878, 487-490.
(7) Leber, T. M.; Negus, R. P. M. Methods Mol. Med. 2000, 39, 509-514.
(8) Leber, T. M.; Balkwill, F. R. Anal. Biochem. 1997, 249, 24-28.
(9) Fields, G. B. In Matrix Metalloproteinase Protocols, Methods in Molecular
Biology; Clark, I. M., Ed.; Humana Press: NJ, 2001; Vol. 151, pp 495-518.
(10) Sasaki, M.; Yoshikane, K.; Nobata, E.; Katagiri, K.; Takeuchi, T. J. Biochem.
1981, 89, 609-614.
(11) Sklar, L. A.; McNeil, V. M.; Jesaitis, A. J.; Painter, R. G.; Cochrane, C. G. J.
* Corresponding author. Phone: 479-575-7018. E-mail: jstenken@uark.edu.
† Present Address: Department of Chemistry and Biochemistry, University
of Arkansas, Fayetteville, AR, 72701.
Biol. Chem. 1982, 257, 5471-5475.
(12) Rust, L.; Messing, C. R.; Iglewski, B. H. Methods Enzymol. 1994, 235, 554-
562.
(1) Stamenkovic, I. J. Pathol. 2003, 200, 448-464.
(2) Nagase, H.; Woessner, J. F. J. Biol. Chem. 1999, 274, 21491-21494.
(13) Castillo, M. J.; Nakajima, K.; Zimmerman, M.; Powers, J. C. Anal. Biochem.
1979, 99, 53-64.
2050 Analytical Chemistry, Vol. 80, No. 6, March 15, 2008
10.1021/ac702047w CCC: $40.75 © 2008 American Chemical Society
Published on Web 02/16/2008

techniques for in vivo assessment of MMP activity.14-16 However,
the target specificity for these approaches is presently limited by
the number of fluorophores available that would provide nonover-
lapping emission bands for the enzymatic products.
Microdialysis sampling is a well-established separation tech-
nique for continuous collection of analytes, such as neurotrans-
mitters, drugs, and their metabolites, both in vivo and in vitro.17,18
It is accomplished by using a probe consisting of a semipermeable
hollow-fiber dialysis membrane affixed to inlet and outlet tubing.
A perfusion solution is passed through the probe at microliter
per minute flow rates and collected for analysis. Analytes that are
smaller than the membrane pores can diffuse freely through the
membrane and are carried to the outlet by the perfusion fluid.
Larger analytes will be rejected by the membrane pores so that
their recovery is negligible. Microdialysis sampling can also be
used to deliver certain substances with a molecular weight smaller
than the membrane molecular weight cutoff (MWCO) to the fluid
outside the probe thus allowing localized metabolism to be
monitored,19 including MMP-2 and MMP-9 activity in breast
cancer,20 phenol metabolism from multiple sites in the liver,21 and
substance P metabolism at the blood-brain barrier.22 During
microdialysis sampling, sampling effectiveness is determined via
the bidirectional extraction efficiency (EE), which can be calcu-
lated as shown in eq 1,
Figure 1. Microdialysis sampling schematic. FL-substrate and UV-
substrate diffuse to the surrounding medium react with elastase
forming products that diffuse back into the probe.
products, we chose to focus on using the N-terminal peptide as
an analytical target. Peptides are easily ionized using ESI condi-
tions and the use of chromatographic separation prior to the
detection allows quantitative measurements for each individual
compound. Such an approach would ultimately allow multiplexed
analysis of various MMP or protease activities at their site of
action. An advantage for using LC-MS is that it allows simulta-
neous monitoring of multiple analytes in a single experiment as
long as the substrate and product species exhibit mass to charge
ratios (m/z) that can be spectroscopically resolved.24,25 LC-MS
analysis of peptide products for various in vitro studies of
enzymatic activity and profiles have been reported.26-29 Recently,
an LC-MS approach has been described for peptide biomarker
studies of MMP-13 activity in synovial fluid and urine.30
To determine the elastase enzymatic activity external to the
microdialysis probe, a fluorogenic elastase substrate and a
colorimetric elastase substrate with different N-terminal peptides
were coperfused and delivered through the dialysis probe as
shown in Figure 1. The organic and N-terminal peptide products
were detected and quantified along with the substrates using LC-
ESI-MS.
Cd - C
Ce - Ci
EE (%) )
i × 100
(1)
where Cd, Ce, and Ci stand for the analyte concentration in
dialysate, external sample medium, and perfusion fluid, respec-
tively.23 This equation allows the calculation of extraction for a
delivered substrate that is locally infused through the dialysis
probe. Additionally, when the analyte concentration in the perfu-
sion fluid is zero, the equation simplifies to the percentage of
analyte concentration in the dialysate divided by the analyte
concentration in the external sample medium and is usually
termed relative recovery (RR). Here, loss of substrate from the
EXPERIMENTAL SECTION
probe will be termed EEloss and recovery will be termed EErec
.
Chemicals. Fluorescent Series Chemicals. Self-quenched fluo-
In this work, in vitro microdialysis sampling and liquid
chromatography-electrospray ionization mass spectrometry (LC-
ESI-MS) was applied to determine the activity of porcine elastase
external to a microdialysis sampling probe. Elastase exhibits
similar substrate specificity to neutrophil elastase (MMP-12) at a
substantially reduced cost. Since many of the enzymatic substrates
for different MMPs contain similar colorimetric or fluorescent
rogenic elastase substrate V, N-methoxysuccinyl-Ala-Ala-Pro-Val-
7-amino-4-methylcoumarin (MeOSuc-AAPV-AMC, FL-substrate),
was obtained from Calbiochem (San Diego, CA), N-methoxysuc-
cinyl-Ala-Ala-Pro-Val (MeOSuc-AAPV, fluorescent substrate N-
terminal peptide, FL-NTP) was purchased from Bachem (Buben-
dorf, Switzerland), and 7-amino-4-methylcoumarin (AMC) was
purchased from Sigma (St. Louis, MO).
Colorimetric Series Chemicals. N-Succinyl-Ala-Ala-Ala-p-nitroa-
nilide (Suc-AAA-pNA, UV-substrate) and p-nitroaniline (pNA) were
both purchased from Sigma (St. Louis, MO).
(14) Funovics, M.; Weissleder, R.; Tung, C. Anal. Bioanal. Chem. 2003, 377,
956-963.
(15) Bremer, C.; Tung, C.; Weissleder, R. Nat. Med. 2001, 7, 743-748.
(16) Deguchi, Jun.; Aikawa, M.; Tung, C.; Aikawa, E.; Kim, D.; Ntziachristos, V.;
Weissleder, R.; Libby, P. Circulation 2006, 114, 55-62.
(17) Stenken, J. A. Microdialysis Sampling Encyclopedia of Medical Devices and
Instrumentation, 2nd ed.; Webster, J. G., Ed.; John Wiley & Sons, Inc.:
Hoboken, NJ, 2006; Vol. 4, pp 400-420.
(18) Bourne, J. A. Clin. Exp. Pharmacol. Physiol. 2003, 30, 16-24.
(19) Stenken, J. A.; Holunga, D. M.; Decker, S. A.; Sun, L. Anal. Biochem. 2001,
290, 314-323.
(24) Hempen, C.; Liesener, A.; Karst, U. Anal. Chim. Acta 2005, 543, 137-142.
(25) Pi, N.; Leary, J. A. J. Am. Soc. Mass. Spectrom. 2004, 15, 233-243.
(26) Liesener, A.; Karst, U. Anal. Bioanal. Chem. 2005, 382, 1451-1464.
(27) Gerber, S. A.; Scott, C. R.; Turee`ek, F.; Gelb, M. H. Anal. Chem. 2001, 73,
1651-1657.
(28) Basile, F.; Ferrer, I.; Furlong, E.; Voorhees, K. Anal. Chem. 2002, 74, 4290-
4293.
(20) Nilsson, U. W.; Dabrosin, C. Cancer. Res. 2006, 66, 4789-4794.
(21) Davies, M. I.; Lunte, C. E. Life Sci. 1996, 59, 1001-1013.
(22) Freed, A. L.; Audus, K. L.; Lunte, S. M. Electrophoresis 2001, 22, 3778-
3784.
(29) Boyer, A. E.; Moura, H.; Woolfitt, A. R.; Kalb, S. R.; McWilliams, L. G.;
Pavlopoulos, A.; Schmidt, J. G.; Ashley, D.; Barr, J. R. Anal. Chem. 2005,
77, 3916-3924.
(30) Nemirovskiy, O. V.; Dufield, D. R.; Sunyer, T.; Aggarwal, P.; Welsch, D. J.;
Mathews, W. R. Anal. Biochem. 2007, 361, 93-101.
(23) Bungay, P. M.; Morrison, P. F.; Dedrick, R. L. Life Sci. 1990, 46, 105-119.
Analytical Chemistry, Vol. 80, No. 6, March 15, 2008 2051

Porcine pancreatic elastase (type I, aqueous suspension) was
purchased from Sigma (St. Louis, MO). One unit of elastase will
hydrolyze 1.0 µmol of Suc-AAA-pNA per min, pH 8.0 at 25 °C. All
other chemicals were reagent grade or better. HPLC-grade water
was purchased from Fisher Scientific. Stock solutions (2 mM) of
FL-substrate, FL-NTP, AMC, and UV-substrate were prepared in
DMSO and stored at 4 °C. All other solutions were prepared with
HPLC-grade water.
High-Performance Liquid Chromatography-Mass Spec-
trometry System (LC-MS). For separation and detection, an
Agilent 1100 series high-performance liquid chromatograph
(HPLC) with the MSD 1100 ion trap mass analyzer (Agilent
Technologies, CA) operating in the ESI mode was used. All
samples were injected using an autosampler with an injection
volume of 2 µL. An Alltima 5 µm C18 column (150 mm × 4.6 mm
i.d.) with guard column was employed to separate the substrates
and products. A binary pump was used to deliver the solvents at
0.5 mL/min flow rate. Solvent A contained 0.2% formic acid in
water, and solvent B contained 0.2% formic acid in acetonitrile. A
linear gradient with solvent B changing from 25% (v/v) to 40%
over a 20 min period was used to separate the analytes.
Mass spectrometric measurements were performed in the
positive ion mode. Mass spectra were recorded over the range
from m/z 100 to 700. The ESI source conditions were as follows:
capillary voltage 4.5 kV, nebulizer gas (N2) pressure 25.0 psi, dry
gas (N2) flow rate 10 L/min, dry temperature 350 °C.
Microdialysis Sampling. Microdialysis sampling was per-
formed with a 1 mL Hamilton syringe (Hamilton Company, Reno,
NV) with plastic tip and a CMA-102 syringe pump (CMA/
Microdialysis, North Chelmsford, MA). CMA 20/04 microdialysis
probes with a 10 mm 20 kDa MWCO polycarbonate/polyether
(PC) membrane (CMA/Microdialysis, North Chelmsford, MA)
were used in all experiments. All experiments were performed in
quiescent solutions at 37 °C maintained by a sand bath (Fisher
Scientific, U.S.A.).
The extraction efficiency for the FL-series substrate and
products in the absence of enzyme was obtained at different flow
rates. For AMC and FL-NTP, microdialysis was performed in
recovery mode with water perfused through a probe to a solution
containing 15 µM AMC or 15 µM FL-NTP, respectively. For the
FL-substrate, microdialysis was performed in delivery mode with
15 µM FL-substrate perfused through a probe immersed into
water. Dialysates (60 µL) were collected in triplicate at different
flow rates (0.5, 1.0, 2.0, 3.0 µL/min) and analyzed by the LC-
ESI-MS.
Next, enzyme was included in the microdialysis system. The
solution external to the microdialysis probe contained 0.5 units/
mL elastase in water. In the single-substrate microdialysis assay,
the perfusion fluid passing through the dialysis probe contained
50 µM FL-substrate in HPLC-grade water. For the multisubstrate
microdialysis assay, the perfusion fluid contained 50 µM FL-
substrate and 50 µM UV-substrate in HPLC-grade water. The
microdialysis sampling perfusion fluid flow rate was 1 µL/min.
Dialysate collection was initiated 25 min post probe immersion
into the enzyme solution to allow for product concentrations to
reach detectable levels. Samples were collected for 60 min. Three
replicate samples were collected during single-substrate experi-
ments. Four replicate samples were collected during multisub-
strate experiments. Statistical analysis required pooling the data
due to differences in replicate sample numbers.
Calibration Curves and Analyte Quantitation. Standard
solutions of FL-substrate (0.5-50 µM), FL-NTP (1-20 µM), AMC
(2-20 µM), and UV-substrate (0.5-50 µM) were prepared by
appropriately diluting the corresponding 2 mM stock solution with
water. Standards (2 µL) were analyzed by the LC-ESI-MS system
in triplicate. Peak areas of extracted ion chromatograms (EIC) of
m/z 628 (FL-substrate), 471 (FL-NTP), 176 (AMC), 452 (UV-
substrate) versus concentration were used to create calibration
curves.
Due to the presence of enzyme outside the probe, the
extraction efficiency of products may differ from the values
obtained when there is no enzyme present. Standard addition
experiments were performed to test the EErec of FL-NTP and AMC
in the presence of enzyme. In this experiment, water was perfused
to microdialysis probes at 1 µL/min flow rate. Elastase was added
to the FL-substrate solution to achieve a concentration of 0.5 units/
mL elastase in 25 µM FL-substrate, and the enzymatic reaction
was initiated. This reaction mixture was vortexed and incubated
at 37 °C for 15 min. Fluorescence at Ex 360 nm, Em 465 nm was
monitored to ensure reaction completion. Then microdialysis
probes were immersed in this reaction mixture, and 10 µL of
dialysate was collected. Next, FL-NTP solution was added to the
reaction mixture to achieve an increase of FL-NTP concentration
of 20 µM. The reaction mixture was well-mixed, and 10 µL of
dialysate was collected. Then AMC solution was added to the
reaction mixture to achieve an increase of AMC concentration of
20 µM. The reaction mixture was well-mixed, and 10 µL of
dialysate was collected. All dialysates were subsequently analyzed
by LC-ESI-MS.
Samples (2 µL) were analyzed in duplicate by the LC-ESI-
MS system. Base peak chromatograms (BPC) of m/z 100-700
and EIC m/z 628 (FL-substrate), 471 (FL-NTP), 452 (UV-
substrate), 332 (UV-substrate N-terminal peptide, UV-NTP), and
139 (pNA) were monitored. Each peak was characterized by their
specific mass spectrum. Peak areas from the above EIC were used
for quantitation.
Analyte Stability. The stability of FL-substrate, FL-NTP, and
AMC were tested by storing 50 µM FL-substrate, 20 µM FL-NTP,
and 20 µM AMC (diluted from 2 mM stock solution by water,
respectively) in the dark at room temperature for various time
periods. Samples were analyzed by LC-ESI-MS using the above
conditions and quantified by using the calibration curves.
High-Performance Liquid Chromatography-Fluorescence
system (LC-FL). In order to compare with the LC-MS detec-
tion, separation and quantitation were also performed on a LC-
FL system, which includes a Shimadzu SIL-10ADvp autoinjector,
LC-10ADvp pumps, SCL-10vp system controller, and RF-551
fluorescence HPLC detector. The same column, flow rate, injection
volume, and gradient were used as in the LC-MS system.
Fluorescence detection at an excitation (Ex) wavelength of 370
nm and emission (Em) wavelength of 460 nm was chosen for AMC
detection.
Kinetics. A Tecan SPECTRAFluor plate reader (Tecan Group
Ltd., M¨annedorf, Switzerland) was used for all the kinetics
measurements. The KM, kcat, and kcat/KM values for both substrates
2052 Analytical Chemistry, Vol. 80, No. 6, March 15, 2008

Figure 2. LC chromatograms and mass spectra for the FL-substrate and its products, FL-NTP and AMC. (A) Extracted ion chromatogram
(EIC) of FL-NTP (m/z 471), AMC (m/z 176), and FL-substrate (m/z 628). (B) Mass spectra of FL-NTP, AMC, and FL-substrate in full scan mode
from m/z 100 to 700. Peaks at m/z 493 and 650 are Na+ adducts and were not quantified.
were determined by plotting 1/v versus 1/[S] in a Lineweaver-
Burk plot.
least stable. It was stable for at least 10 h and was reduced to 90%
of its initial concentration after 24 h at room temperature.
Collection followed by detection of dialysates was often less than
24 h suggesting that no more than 10% variation in sample
concentration would be expected under these conditions.
Figure 2 shows a typical LC chromatogram and mass spectra
for the single-substrate dialysate analyzed by LC-ESI-MS. Separa-
tion was achieved on a C18 column with FL-NTP, AMC, and FL-
substrate eluting at 5.4, 14, and 19 min, respectively. The retention
behavior of the FL-substrate and products was consistent with a
reversed-phase separation mechanism as the N-terminal peptide
is more hydrophilic than AMC. ESI-MS spectra showed peaks of
[M + H]+ ions of FL-NTP (m/z 471), AMC (m/z 176), and FL-
substrate (m/z 628). Sodium ion adducts, [M + Na]+ were
observed for FL-NTP (m/z 493) and FL-substrate (m/z 650)
despite the addition of formic acid to suppress sodium adduct
formation; however, their intensities are much lower compared
to the ions of the protonated molecules. Only the EIC peak area
of [M + H]+ ions was used for quantitation.
Elastase kinetics for the UV-substrate was determined by
adding 50, 63, 83, 100, 125, 167, 250, and 333 µM UV-substrate to
0.01 units/mL elastate. The reaction progress was monitored at
405 nm every 30 s for 8 min using a pNA calibration curve between
0 and 400 µM. At 405 nm, the product of the reaction, pNA, has
significant molar absorbtivity and the UV-substrate exhibits no
absorbance.
Kinetics of the FL-substrate in the presence of elastase were
determined by adding 67, 75, 86, 100, 120, 150, and 200 µM FL-
substrate to an enzyme concentration of 0.2 units/mL. The
production of AMC from the cleavage reaction was measured
using a 360 nm for excitation and 465 nm for emission every 30
s for 8 min. The calibration curve of AMC concentration ranging
from 0 to 250 µM was obtained using fluorescence Ex 360 nm,
Em 465 nm.
RESULTS AND DISCUSSION
Analyte Stability and Method Optimization. At room tem-
perature, the FL-substrate was found to be stable for 24 h; the
FL-NTP was stable for greater than 12 h and was reduced to 96%
of its initial concentration after 24 h. AMC was found to be the
Linear calibration curves were achieved for FL-substrate (0.5-
50 µM), FL-NTP (1-20 µM), and AMC (2-20 µM) (figures not
shown), based on the peak area from EIC of [M + H]+ ions versus
Analytical Chemistry, Vol. 80, No. 6, March 15, 2008 2053

Table 1. FL-Substrate and UV-Substrate Kinetic
Parameters
Table 2. FL-Substrate, FL-NTP, and AMC Extraction
Efficienciesa
FL-substratea
UV-substrateb
flow rate
AMC
FL-NTP
EErec
FL-substrate
EEloss
(µL/min)
EErec
%
%
%
KM (µM)
79.62
1.59
0.02
65.78
22.74
0.35
kcat (min-1
)
0.5
1.0
2.0
3.0
83.7 ( 1.3
76.3 ( 1.7
68.2 ( 8.3
56.0 ( 1.8
18.3 ( 0.3
11.5 ( 0.2
8.7 ( 1.8
7.6 ( 0.2
63.6 ( 5.5
48.5 ( 2.6
32.8 ( 3.9
23.0 ( 0.2
kcat/KM (min-1 µM-1
)
a FL-substrate kinetics were measured using 0.2 units/mL elastase.
b UV-substrate kinetics were measured using 0.01 units/mL elastase.
a EErec% denotes extraction efficiency in recovery mode. EEloss
%
denotes extraction efficiency in delivery mode. All experiments were
performed in quiescent solutions at 37 °C. Data represent mean ( SD,
n ) 3.
concentration. The limit of quantitation (LOQ) was used as the
lowest concentration standard for the calibration curve and was
achieved at the concentration that yields a signal-to-noise ratio
(S/N) of 10:1.
For equimolar concentrations of all three analytes, the signal
of AMC was about 10 times lower and it exhibited poor S/N due
to the low ionization efficiency compared to that of the peptides.31,32
Several methods were applied to optimize AMC signal, including
tuning the capillary voltage to a relatively high value of 4.5 kV,31,32
using a narrow scan range (m/z 170 to ∼190), or using selected-
reaction monitoring (SRM) for m/z 176. A narrow scan range can
relatively increase the sensitivity, but the S/N is poor; SRM can
significantly increase S/N though it did not cause an obvious
signal increase. Additionally, MS/MS detection (FL-NTP m/z 471
f m/z 215, AMC m/z 176 f m/z 117, FL-substrate m/z 628 f
m/z 443) can reach significantly lower LOQ for all analytes: 100
nM for FL-NTP, 500 nM for AMC, and 50 nM for FL-substrate.
Yet, the calibration curve for AMC using MS/MS did not exhibit
a linear concentration relationship above a concentration of 5 µM,
but was linear for the FL-substrate and FL-NTP up to 20 µM.
Extracted ion chromatograms obtained from the mass spectra
recorded over the range from m/z 100 to 700 were used for
quantitation of dialysates.
The difference in ionization efficiency between the peptide
product and AMC also suggests the advantage for using peptide
substrates for the protease analysis due to the high ionization
efficiency typically obtained for peptides. Although two different
peptide products may exhibit differences in their ionization
efficiency because of their different amino acid sequences,
generally their ESI signal will be much higher than typically
observed for organic products such as AMC that are used in
conventional protease assays.31,32
Microdialysis Sampling Extraction Efficiency without
Elastase. In vitro probe calibration for all three analytes without
elastase was performed, and the determined EE values are shown
in Table 2. At 1 µL/min, the EEloss of FL-substrate in water was
48.5% ( 2.6% (n ) 3) and the UV-substrate was 42.8 ( 2.1 (n )
3). The in vitro EErec of FL-NTP at 1 µL/min flow rate was 11.5 (
0.2 (n ) 3), which is much lower than the EEloss for the
FL-substrate. FL-NTP standard prepared in water instead of from
the 2 mM DMSO standard and diluted to an external concentra-
tion of 15 µM gave an EErec of 17.5 ( 0.4 (n ) 3), indicating that
use of the DMSO solvent did not cause the significantly lower
EErec for the FL-NTP compared to the FL-substrate. Microdialysis
Figure 3. FL-substrate delivery and collected AMC concentrations
as a function of perfusion fluid flow rate. The probes were perfused
with 50 µM FL-substrate to a quiescent 0.5 units/mL elastase solution
at 37 °C. Symbols and error bars denote mean ( SD, n ) 3.
sampling was also performed in delivery mode where 15 µM FL-
NTP was included in the perfusion fluid and perfused to water at
1 µL/min flow rate. In the delivery mode, the EEloss of FL-NTP
was 14.7 ( 3.6 (n ) 3) indicating that FL-NTP mass transport
characteristics to and from the microdialysis probe are similar as
would be expected.33
Single-Substrate Microdialysate Analysis in the Presence
of Enzyme. The flow-rate-dependent FL-substrate EEloss to a
solution containing elastase and the collected AMC concentrations
are determined by LC-FL and shown in Figure 3. As the perfusion
flow rate increased, both the concentration of AMC collected into
the probe and the FL-substrate EEloss decreased. The time-
dependent delivery of FL-substrate and collected FL-NTP con-
centrations are determined by LC-MS and shown in Figure 4.
Through the duration of the collection time, the FL-substrate EEloss
remained constant and the recovered concentration of FL-NTP
increased. These results are consistent with microdialysis sam-
pling mass transport principles. It is well-established that mass
transport rates in the dialysate are affected by the analyte aqueous
solution diffusion coefficient and the volumetric flow rate (Qd) of
the microdialysis perfusion fluid; an increase in Qd leads to a
decrease in the microdialysis EE.34 With a localized infusion
(31) Kollroser, M.; Schober, C. Clin. Chem. 2002, 48, 84-91.
(32) Concannon, S.; Ramachandran, V. N.; Smyth, W. F. Rapid Commun. Mass
Spectrom. 2000, 14, 1157-1166.
(33) Song, Y.; Lunte, C. E. Anal. Chim. Acta 1999, 379, 251-262.
(34) Stenken, J. A. Anal. Chim. Acta 1999, 379, 337-358.
2054 Analytical Chemistry, Vol. 80, No. 6, March 15, 2008

Table 4. Single-Substrate versus Multisubstrate
Comparisona,b
delivered
recovered
recovered
FL-NTP
EEenzyme of FL-substrate FL-NTP concn
FL-substrate (pmol/min) (µM) in the (pmol/min) in
perfusate
(%)
outside probe
dialysate
4.3 ( 0.7*
11.0 ( 0.5
5.6 ( 0.1*
the dialysate
4.3 ( 0.7*
11.0 ( 0.5
5.6 ( 0.1*
FL-substrate
50 µM
66.5 ( 4.6* 33.3 ( 2.3*
62.1 ( 0.9* 62.1 ( 0.9
67.3 ( 3.6* 33.7 ( 1.8*
FL-substrate
100 µM
FL-substrate
50 mM and
UV-substrate
50 µM
a An “*” denotes no statistical significance difference at the 95%
confidence level between each designated value and the FL-substrate
50 µM value. b Microdialysis was performed at 1 µL/min with quiescent
conditions at 37 °C. Sample collection (60 µL) was initiated 25 min
post probe insertion. Data represent mean ( SD, n ) 3 for single-
substrate experiments and n ) 4 for multisubstrate experiments.
Figure 4. Time-dependent delivery of FL-substrate and FL-NTP
recovery during single-substrate dialysis. FL-substrate (50 µM) was
perfused at 1 µL/min to a quiescent solution containing 0.5 units/mL
elastase at 37 °C. FL-NTP concentrations in the 10 and 20 min
dialysate samples were below the LOQ. Symbols and error bars
denote mean ( SD, n ) 3.
the enzyme removes the substrate as it diffuses from the probe.
When either 50 or 100 µM FL-substrate was perfused, the FL-
substrate delivered outside the probe was 33.3 ( 2.3 and 62.1 (
0.9 pmol/min. The FL-NTP collected in the dialysate was 4.3 (
0.7 and 11.0 ( 0.5 pmol/min. Combining this information, it shows
a higher percentage conversion of substrate to product outside
the probe when 100 µM FL-substrate was perfused as compared
to when 50 µM FL-substrate was perfused. This increase is
expected considering the enzyme kinetics, that when more
substrate is present, the reaction will proceed at a faster rate, and
thus, within a certain collection time more substrate is converted
to products and the conversion rate is higher.
Multisubstrate Microdialysate Analysis in the Presence
of Elastase. On the basis of all the previous MS setup, a
multisubstrate microdialysis assay scheme was developed to
determine the potential for multiple substrate infusion followed
by product detection. FL-substrate (50 µM) and UV-substrate (50
µM) were coperfused through the dialysis probe to a surrounding
solution containing 0.5 units/mL elastase. Four different products,
FL-NTP, AMC, UV-NTP, and pNA diffused back to the probe
combined with the two substrates remaining in the dialysate. With
the use of the same HPLC gradient, all six analytes can be resolved
and quantified based on the EIC of their [M + H]+ ions. Figure
5 shows the LC-ESI-MS chromatograms and mass spectra
obtained during the multisubstrate infusion to elastase.
When a single enzyme acts on two different substrates, each
substrate will work as a competitive inhibitor to the other.35 The
two substrates will compete for the binding site, and the enzyme
exhibits kinetics that make it appear as if the KM increases for
each substrate. Comparing the kinetics data shown in Table 1,
elastase has a weaker affinity to the FL-substrate (KM 79.62 µM)
than to the UV-substrate (KM 65.78 µM), and the kcat is slower for
the FL-substrate. Considering the kcat/KM value, which allows the
direct comparison of the effectiveness of the enzymatic conversion
toward different substrates, it is much more effective for elastase
to cleave the UV-substrate than to cleave the FL-substrate.
The FL-substrate and FL-NTP were quantified by comparing
with the calibration curves. The EEloss and product amounts
Table 3. Estimated Extraction Efficiencies in Elastasea
dialysate
product concn
(µM)
product concn
(µM) after spike
to external soln
EEenzyme
(%)
FLNTP
AMC
4.8 ( 0.7
12.4 ( 2.4
9.3 ( 1.8
22.3 ( 2.9
22.3
49.5
a Microdialysis samples (1 µL/min) were collected after 25 µM FL-
substrate was incubated with 0.5 units/mL at 37 °C for 15 min. Then
FL-NTP and AMC were spiked such that the external concentration
was increased by 20 µM. EEenzyme was calculated by dividing the
difference of recovered product concentration by 20 µM. Data represent
mean ( SD, n ) 3 for 10 µL dialysates.
combined with product collection, the concentration of products
collected would be expected to increase with an increasing
infusion into a quiescent system. Additionally, the enzymatic
reaction external to the probe causes the FL-substrate to reach
an EEloss steady state.
To determine approximate EErec values of the products in the
presence of elastase (EEenzyme) values were determined using a
standard addition experiment. Table 3 shows that although AMC
EErec values decreased under these experimental conditions from
76.3% (water) to 49.5% (enzyme solution), the FL-NTP EErec
slightly increased in the presence of the enzyme to 22.3%.
Although it is possible that AMC binds nonspecifically to the
elastase enzyme causing a reduction in its EErec, it is not clear
why the FL-NTP has an increased EErec value under these
conditions.
In the single-substrate microdialysis experiment, FL-substrate
(50 or 100 µM) was perfused through the probe immersed in a
medium containing 0.5 units/mL elastase. The EE data for the
FL-substrate and collected product concentrations are listed in
Table 4. With a 50 µM FL-substrate infusion, the EEloss was 66.6%
( 4.6% (n ) 4) using the LC-MS detection method, which is
comparable to 69.4% ( 2.2% (n ) 3) using LC-fluorescence
detection method (p < 0.05 level). The FL-substrate EEloss is
greater for probes immersed in an elastase solution as compared
to solutions with no enzyme (48.5%) as would be expected since
(35) Segel, I. H. Enzyme Kinetics. Behavior and Analysis of Rapid Equilibrium
and Steady-State Enzyme Systems; John Wiley & Sons, Inc.: 1975.
Analytical Chemistry, Vol. 80, No. 6, March 15, 2008 2055

Figure 5. LC chromatograms and mass spectra of multisubstrate dialysate experiments. FL-substrate (50 µM) and UV-substrate (50 µM)
were coperfused at 1.0 µL/min to a quiescent medium containing 0.5 units/mL elastase at 37 °C. (A) EIC of FL-NTP (m/z 471), AMC (m/z 139),
FL-substrate (m/z 628), and UV-NTP (m/z 332), pNA (m/z 139), UV-substrate (m/z 452). (B) Mass spectra of FL-NTP, AMC, FL-substrate, and
UV-NTP, pNA, UV-substrate in full scan mode from m/z 100 to 700. Peaks at m/z 354, 474, 493, and 650 are Na+ adducts and were not
quantified.
recovered between the multisubstrate versus the single-substrate
dialysate are shown in Table 4. There is no significant difference
(p < 0.05 level) between the EEloss value of FL-substrate in the
multisubstrate dialysate and in the single-substrate dialysate, when
either 50 or 100 µM FL-substrate was perfused. This indicates
that the EE value is not as sensitive to minor changes in kinetics
2056 Analytical Chemistry, Vol. 80, No. 6, March 15, 2008

external to the probe. The recovered concentrations of FL-NTP
during the multisubstrate versus the single-substrate infusion
exhibited no significant differences (p < 0.05 level). Despite the
higher kcat/KM value of UV-substrate as compared to the FL-
substrate, there appeared to be no potential inhibition of the
elastase catalytic activity toward the FL-substrate. This is most
likely due to the higher enzyme concentrations used in these
studies. Finally, the collected recovered concentrations of FL-NTP
were doubled when 100 µM FL-substrate was perfused.
dialysate. In comparison to LC-FL detection, the use of LC-ESI-
MS in enzymatic bioassay offers many distinct advantages. The
main advantage with the LC-MS approach is that selected peptide
substrates can be used allowing for a multiplexed approach toward
determining enzymatic activity.
ACKNOWLEDGMENT
We gratefully acknowledge NIH EB 001441 for funding the
research and NSF CHE-0091892 for funds to purchase the Agilent
MSD Trap Instrument.
CONCLUSIONS
Received for review October 2, 2007. Accepted December
28, 2007.
Enzymatic activity external to a microdialysis sampling probe
can be detected by infusing proper enzymatic substrates through
the dialysis probe followed by analysis of the products in the
AC702047W
Analytical Chemistry, Vol. 80, No. 6, March 15, 2008 2057