| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
Millennium Pharmaceuticals, Inc., 38 Sidney St., Cambridge, MA 02139.
a Author for correspondence. Fax 617-551-3747; e-mail elightcap{at}mpi.com
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionAppendix 1References |
|---|
Methods: Fluorogenic kinetic assays for both the chymotryptic and tryptic activities of the proteasome have been optimized for both whole blood and blood cells. Using the ratio of these activities and the catalytic mechanism of the proteasome, we developed a novel method of calculating percentage of inhibition, using two structurally unrelated inhibitors (PS-341 and lactacystin).
Results: This ratio method was demonstrated to be sensitive (detection limit of 13% inhibition with 10 µg of cell lysate), specific to the proteasome (PS-341 provides >98% inhibition), accurate (112% analyte recovery), and precise (0% ± 5% inhibition at 0 nmol/L PS-341 and 74.5% ± 1.7% inhibition at 200 nmol/L PS-341). Using these assays, we found that both erythrocytes and leukocytes contain proteasome at 3 µmol/L. Pharmacodynamic results for PS-341 obtained from the whole-blood ratio method were comparable to those using leukocytes determined by another method.
Conclusions: The described assay provides a reliable method for studying the pharmacodynamics of proteasome inhibitors and is now in use in concurrent phase I clinical trials with PS-341.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendix 1References |
|---|
). The ATP hydrolytic activity and the subunits that bind
ubiquitin in the 26S are contained within a protein complex known as
the 19S cap. The ATP-independent proteolytic activity of the proteasome
is contained within the central 20S core
(Mr = 730 000). The 20S proteasome is
a multicatalytic protease that has three well-characterized peptidase
activities—chymotryptic, tryptic, and postglutamyl peptide
hydrolytic—that are associated with three distinct subunits: ß5,
ß2, and ß1, respectively (8). Each site is defined by
its ability to hydrolyze peptide substrates in vitro, with hydrophobic,
basic, or acidic amino acids in the P1 position.
|
Inhibitors of the proteasome have been designed based on either the natural product, lactacystin (9), or synthetic peptidyl derivatives (10). PS-341, a peptidyl boronic acid, is a novel inhibitor of the chymotryptic site within the 20S proteasome (11). This molecule shows at least 500-fold selectivity for the proteasome over other enzymes and receptors and exhibits substantial potency in cell-free and in vitro cell-based assays (MDS PanLabs, Bothell, WA, unpublished results). PS-341 has shown significant activity in preclinical murine tumor models (3) where the antitumor activity was positively correlated with the degree of proteasome inhibition measured both in leukocyte and tumor biopsy material. PS-341 at 1 mg/kg body weight produced an ~60% reduction in the growth of PC-3 tumors in mice, which correlated with ~85% inhibition of proteasome activity in leukocyte lysates.
PS-341 currently is under evaluation as an anticancer agent in multiple phase I clinical trials in which blood concentrations of the drug are being measured by liquid chromatography–mass spectrometry to determine standard pharmacokinetic parameters (R.A. Newman and P.J. Elliott, personal communication). Based on animal models (including nonhuman primates), the drug is distributed rapidly and falls to near detection limits within minutes of intravenous dosing (P.J. Elliott, unpublished results). Whether this pattern will carry over to the clinical trials at the highest dose tolerated is still unknown. The pharmacodynamic profile of the drug could also be evaluated to assess the activity of PS-341 at its target site, the proteasome. As such, it would be possible not only to evaluate blood concentrations of PS-341 but also to record the extent of enzyme inhibition over time.
To explore the possibility that the proteasome activity assay, or a variation of it, could be used in future clinical studies, the present series of experiments were undertaken. The ex vivo assays reported here were developed utilizing the current knowledge of the catalytic activities within the 20S proteasome and a previous report by Stein et al. (12) on proteasome activity measurements. The focus of the method development was to obtain a simple assay that was both rapid and reproducible and could be used to determine accurately the proteasome activity in rodent blood samples treated with PS-341. To confirm that the tryptic and chymotryptic activities were attributable only to the proteasome, a proteasome inhibitor from a second structurally unrelated class was also used, clasto-lactacystin-ß-lactone (lactacystin). Using multiple assays, we cross-validated each format and determined the optimal assay methodology. Herein, we describe the development of these pharmacodynamic assays to record proteasome activity.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendix 1References |
|---|
Male Sprague–Dawley rats (180–220 g) were obtained from Harlan. All animal procedures were approved by the Institutional Animal Use and Care in accordance with NIH guidelines. Heparinized blood was obtained from anesthetized (8 mg of ketamine plus 1.2 mg of xylazine per 100 g of body weight) rats by cardiac puncture. Heparinized human blood was obtained from Millennium volunteers. 20S proteasome was purified from rabbit reticulocytes by published methods (12).
Stock solutions of PS-341 and clasto-lactacystin ß-lactone (lactacystin) were prepared in dimethyl sulfoxide (DMSO)2 for all in vitro studies. The final concentration of DMSO in all incubations was <1.0%. For in vivo work, PS-341 was prepared in vehicle (980 mL/L saline solution, 20 mL/L ethanol, 5 g/L ascorbic acid).
preparation of blood cell lysates from rat and human blood
Venous blood (10 mL) was collected from human subjects into
heparin-containing tubes and mixed, in accordance with the
manufacturer’s instructions. For rat samples, 20 µL of heparin
was mixed into 2 mL of collected venous blood. Blood samples were
prepared according to the method of Adams et al. (3) with
the following modifications: packed whole blood (PWB) lysates were
prepared without the intervening Nycoprep separation, and red blood
cell (RBC) lysates were prepared by collecting the pellet from the
Nycoprep separation, with careful avoidance of the granulocyte layer.
Complete lysis was achieved within 15 min.
assay of proteasome activities in blood lysates
Kinetic measurement of proteasome activity followed that described
by Adams et al. (3). PWB or RBC lysate (20–100 µg of
protein) or white blood cell (WBC) lysate (>10 µg) was added to the
cuvette. When the proteasome chymotryptic activity in PWB or RBC
lysates was measured, the same assay buffer was used except that the
sodium dodecyl sulfate (SDS) concentration was 0.5 g/L, except
as noted. The signal was linear between 3 and 60 min. Measurement of
the proteasome tryptic activity for all lysate types was performed in
60 µmol/L Bz-Val-Gly-Arg-7-amido-4-methylcoumarin (VGR-AMC) in 20
mmol/L HEPES, pH 8.0, containing 0.5 mmol/L EDTA and 10 mL/L DMSO.
Enzyme units are expressed as pmol AMC/s. This is equivalent to 6
x 10-5 U.
determination of the michaelis-menten constants
Concentrations of the substrate Suc-Leu-Leu-Val-Tyr-AMC (LLVY-AMC;
chymotryptic activity) were evaluated in 5 or 10 µmol/L increments up
to 100 µmol/L. VGR-AMC (tryptic activity) was evaluated in a similar
manner up to 400 µmol/L except that no SDS was included in the
buffer.
inhibition of proteasome chymotryptic and tryptic activities in
human blood lysate by lactacystin in vitro
Inhibitor (lactacystin, 0–4 µmol/L; or PS-341, 0–1 µmol/L)
was added to 30–50 µL of blood lysate (protein concentration, 1.3–7
g/L) and incubated at 37 °C for 30 min. The samples were then placed
on ice and assayed within 6 h. Proteasome activities were measured
as described above. The tryptic activity data for PS-341 were fitted to
the equation for general nonessential activation (13).
effect of ps-341 on the chymotryptic and tryptic activities in pwb
lysates prepared from ps-341 in vivo-treated rats
Rats received intravenously injections of 0–0.3 mg/kg PS-341 in
100 µL of vehicle (980 mL/L saline, 20 mL/L ethanol, 5 g/L ascorbic
acid). After 1 h, heparinized blood was obtained, and WBC
and PWB lysates were prepared for the chymotryptic and tryptic assays
as described above.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendix 1References |
|---|
and 2). By contrast, it was found that PS-341 also induces
activation of the tryptic catalytic activity within the 20S proteasome,
whereas lactacystin produces complete inhibition of the tryptic
activity, as determined using a specific peptide substrate for this
hydrolytic site, VGR-AMC (Fig. 2
).
|
When we used the published method to determine the extent of proteasome
inhibition, it was necessary to calculate the specific activity of the
chymotryptic site in the presence and absence of PS-341 and then apply
the following equation (Eq. 10 in the Appendix):
 |
where %I is the percentage of inhibition of the chymotryptic activity of the 20S proteasome, SpAI is the chymotryptic specific activity of the proteasome in the presence of inhibitor, and SpAU is the chymotryptic specific activity of the proteasome in the absence of inhibitor.
This required determination of the total protein concentration of the biological samples being assayed for proteasome activity. Such methodology was reliant on a constant number of proteasomes/mg of protein in a sample. However, the proteasome/protein ratio in a sample may vary depending on tissue type or reproducibility of ex vivo sample preparation. As such, determination of the extent of proteasome inhibition using specific activity measurements might be compromised.
To accurately determine the extent of proteasome inhibition, it was
desirable to determine the total proteasome activity in a sample,
regardless of whether PS-341 is bound, so that the percentage of
inhibition could be determined without the requirement for a separate
uninhibited "control" sample. This was achieved by exploring the
activity of the other catalytic sites within the proteasome. Studies
showed that PS-341 did not inhibit the tryptic activity within the
proteasome; rather, it enhanced the activity of this site (Fig. 2A
).
Therefore, it was possible to calculate the "total" proteasome
activity within the sample that correlated with the activity of the
tryptic site (vT), provided that the
activation of the tryptic site was taken into account. The ratio of the
inhibited chymotryptic activity (vC)
and the uninhibited tryptic activity
(vT) can then be used to determine an
accurate percentage of inhibition (derived in the Appendix,
Eq. 8 ):
 |
where kC and kT are the apparent rate constants for the chymotryptic and tryptic sites, respectively, under a defined set of assay conditions.
measurement of 20s proteasome activity in pwb lysates
Assay methods for the chymotryptic and tryptic activities of the
proteasome were developed and optimized for measurement of these
activities in PWB lysates. The chymotryptic assay presented here
involves the hydrolysis of LLVY-AMC to Suc-Leu-Leu-Val-Tyr-OH and AMC
by the chymotryptic activity (associated with subunit ß5) of the 20S
proteasome. AMC is a highly fluorescent molecule
(
ex = 380 nm; em =
440 nm), and its rate of release was measured over time. The addition
of 0.5 g/L SDS to the assay buffer (20 mmol/L HEPES, pH 8.0, 0.5 mmol/L
EDTA, 60 µmol/L LLVY-AMC, 10 mL/L DMSO) was required to activate the
20S core in PWB lysates (Fig. 3
). The exact mechanism by which this occurs is not fully
understood, but a recent review details the current hypotheses directed
at addressing this issue (7). When human PWB lysate was
used, the chymotryptic activity assay was linearly dependent on total
protein between 10 and 100 µg added to a 2-mL assay volume (Fig. 4
). The chymotryptic activity was inhibited by 98% by the
selective proteasome inhibitor, PS-341 (5 µmol/L). Under these
conditions, the tryptic activity was activated 1.2-fold over the
activity measured in the absence of PS-341. In addition, the activity
was inhibited 97% with another specific proteasome inhibitor,
lactacystin (1 µmol/L). Comparable data were produced using rat PWB
lysates (data not shown). Because two structurally distinct
proteasome-specific inhibitors gave essentially complete inhibition of
the activity, these results strongly support that the measured
hydrolytic activity was attributable to the proteasome.
|
|
Two critical modifications were introduced to the 20S proteasome assay
reported by Stein et al. (12) to allow for more precise
activity measurements in PWB lysates. (a) The SDS
concentration in the assay buffer was increased to 0.5 g/L from 0.35
g/L. When 0.35 g/L SDS was used in the assay buffer, the chymotryptic
activity was not linear with increasing RBC lysate added to the assay.
In fact, addition of RBC lysate under these conditions inhibited the
assay (Fig. 3
). The source and identity of this endogenous inhibition
is not known; however, the presence of such a proteinaceous proteasome
inhibitor in RBCs has been reported (14)(15)(16)(17)(18). By increasing
the concentration of SDS to 0.5 g/L, the inhibitory effects were
eliminated, perhaps because of denaturation of the endogenous
inhibitor. (b) The concentration of DMSO was maintained at
10 mL/L to improve substrate solubility. At 10 mL/L DMSO,
Michaelis-Menten kinetics were observed, which were not attainable at
lower DMSO concentrations (12). We did not observe evidence
for hysteresis in the assay at 10 mL/L DMSO. Under these new optimal
conditions, Vmax and
Km were 0.93 ± 0.05 pmol
AMC · s-1 · mg
protein-1 and 13.0 ± 1.9 µmol/L,
respectively.
The tryptic assay presented here involves the hydrolysis of VGR-AMC to
Bz-Val-Gly-Arg-OH and AMC by the tryptic activity (associated with
subunit ß2) of the proteasome. By contrast, no SDS was required in
this assay buffer (20 mmol/L HEPES, pH 8.0, 0.5 mmol/L EDTA, 60
µmol/L VGR-AMC, 10 mL/L DMSO) because no inhibition of the tryptic
activity was seen by PWB lysate. In addition, it was noted that SDS
precipitates the substrate at >10 µmol/L VGR-AMC. When the rat and
human PWB lysates were used, the tryptic assay was linear between 10
and 100 µg of protein added to a 2-mL assay (Fig. 4
).
PS-341 cannot be used to demonstrate that the proteasome accounted for
all of the tryptic activity within the PWB lysate because it did not
inhibit this activity (Fig. 2A
). Lactacystin is a selective inhibitor
of the proteasome (9) and provides 94% inhibition of
tryptic activity in human PWB lysate (8 µmol/L). Similar results were
obtained for rat PWB lysate. Therefore, all of the tryptic activity
present in rat and human PWB lysate appears to be attributable to the
proteasome. The tryptic activity also follows Michaelis-Menten kinetics
with a Vmax = 1.47 ± 0.15 pmol
AMC · s-1 · mg
protein-1 and
Km (VGR-AMC) = 130 ± 30
µmol/L.
The accuracy of both the chymotryptic and tryptic assays for human
samples was evaluated using an analytical recovery method, i.e., how
much of a known amount of enzyme activity was recovered after addition
to the appropriate biological sample. This was done by titrating
purified 20S proteasome into an assay containing human PWB lysate and
then using the slope to calculate the specific activity of the added
20S proteasome (Table 1
), effectively subtracting the PWB lysate proteasome activity.
The recoveries of the chymotryptic and tryptic activities when assayed
in the presence of PWB lysate were 131% and 117%, respectively. More
importantly, the ratio of the two activities,
vC/vT,
used in the calculation of percentage of inhibition in Eq. 8 , in the
presence of PWB lysate gave a more accurate recovery of 112% (3.4
± 0.5 for purified 20S; 3.8 ± 0.2 in PWB lysate), demonstrating
that the chymotryptic-to-tryptic ratio method is more accurate.
|
As an indication of the analytical sensitivity, the specific activity method has a calibration curve slope (%I vs SpA) of -80 and the ratio method has an approximate calibration curve slope (%I vs vC/vT) of -24. Thus, for a human PWB lysate, 0%, 30%, and 80% inhibition would correspond to specific activities of 1.25, 0.88, and 0.25 pmol AMC · s-1 · mg protein-1, respectively. For the ratio method, a human PWB lysate at 0%, 30%, and 80% inhibition would give vC/vT values of 3.96, 2.59, and 0.66, respectively.
measurement of 20s proteasome activity in wbc lysate
Clinical studies may require that proteasome measurements be
determined in blood cell subpopulations (e.g., leukemias) as well as in
whole blood samples. Optimization of assay conditions for both rat and
human WBC lysates produced a 20S chymotryptic assay identical to that
for PWB lysates except that the SDS concentration was lowered to 0.35
g/L, where maximal activity was seen. The proteasome tryptic assay for
WBC lysates was identical to that for PWB lysates. In general, the WBC
lysate assays were not as robust as the PWB lysate assays. A parallel
data set was collected to provide validation of the WBC lysate assay,
presented below.
Both the chymotryptic and tryptic assays were linearly dependent on enzyme concentration between 15 and 60 µg of protein from WBC lysate when added to a 2-mL assay. Proteasome activity accounted for all of the chymotryptic activity (96% inhibition at 500 nmol/L PS-341; 98% inhibition at 2 µmol/L lactacystin) and all of the tryptic activity (complete inhibition at 7 µmol/L lactacystin) in human WBC lysates, with similar results in rats (data not shown). The chymotryptic activity from human WBC lysate followed Michaelis-Menten kinetics with a Vmax = 35 ± 1 pmol AMC · s-1 · mg protein-1 and Km (LLVY-AMC) = 30.3 ± 1.8 µmol/L. A Michaelis constant for VGR-AMC in the tryptic activity from human WBC lysate could not be determined because of substrate insolubility even at 10 mL/L DMSO.
The analytical recovery of chymotryptic activity from purified 20S proteasome added to WBC lysate was 230%; the recovery of tryptic activity was 360%. The ratio of the two assays, vC/vT, in the presence of WBC lysate was 63% of that determined in the absence of WBC lysate (3.4 ± 0.5 for purified 20S, 2.15 ± 0.03 in WBC lysate). The activation effect that WBC lysate has on the 20S proteasome is not critical because the percentage of inhibition will be calculated using kC/kT determined for WBC lysate. Cross-validation of results from WBC lysate with those from PWB lysate demonstrated that the methods gave accurate percentages of inhibitions (see below). The specific activity method had a calibration curve slope (%I vs SpA) of -3.9, and the ratio method had an approximate calibration curve slope (%I vs vC/vT) of -30.
detection limits and precision
To determine the detection limits for these assays, the data from
the comparison of the four assay formats with blood samples obtained
from PS-341-treated rats were used (Fig. 5
and Table 2
). We evaluated the minimum percentage of inhibition
that could be determined using a 10-µg protein sample for our
determination of the detection limits. This provides the most critical
evaluation of limitations of the assay because the ability to
distinguish between two relatively similar rates at lower protein
concentrations is the most difficult to achieve. The percentage of
inhibition is more appropriate than drug concentration because this
assay is a pharmacodynamic assay. The detection limits for rat PWB
lysate methods, determined as 3 SD at a dose of 0.03 mg/kg (Fig. 5
),
was 17% inhibition for the specific activity method or 13% inhibition
for the ratio method. The detection limit for WBC lysate was similarly
16% inhibition for the specific activity method or 13% inhibition for
the ratio method.
|
|
The precision of inhibition results from the human PWB lysate ratio
method is presented in Table 3
. The CVs of both the chymotryptic and tryptic assays at
0, 10, 50, and 200 nmol/L PS-341 were all between 2.8% and 6.2% over
the 20 measurements (data not shown). The concentrations of PS-341
correspond to the following percentages of inhibitions ± SD: 0
nmol/L, 0% ± 5%; 10 nmol/L, 14% ± 4%; 50 nmol/L, 43% ± 3%; and
200 nmol/L, 74.5% ± 1.7%. Therefore, in in vitro assays under these
conditions, 10 nmol/L PS-341 represents the detection limit of this
assay.
|
comparison of assays and tissue sources
Comparison of the specific activity and ratio methods provided
evidence for the relative accuracy of the different methods. In
addition, comparison of results from WBC and PWB lysates demonstrated
differences in the pharmacodynamics of PS-341 between two sources, WBCs
and blood cellular components. Rats were treated with PS-341 (0.01–0.3
mg/kg body weight), and blood samples were withdrawn 1.0 h later.
WBC and PWB lysates were prepared and assayed for chymotryptic and
tryptic activity. Results show that the proteasome inhibition, observed
for each assay, was equivalent for each dose of PS-341 (Fig. 5
). These
data support that each assay is measuring true proteasome activity and
hence help cross-validate each method. In addition, these data
demonstrate that the presence of the endogenous inhibition in RBCs does
not impact the pharmacodynamic behavior of PS-341 in RBCs when assayed
under the described conditions. As such, variations in the endogenous
inhibition should not impact the interpretation of PS-341 assay
results.
assay robustness
In a clinical setting, the degree of proteasome inhibition in
blood could be evaluated in analytical laboratories distant from the
collection site. In such cases, it is important to understand the
limitations placed on the sample during transit, whether shipped on ice
or frozen. Studies may require the use of WBC or PWB samples.
Additionally, these lysed samples may need to be frozen and
subsequently reevaluated. The robustness of the WBC and PWB assays,
under these various conditions, was evaluated.
The stability of the proteasome inhibition by PS-341 in PWB lysates
over time was assessed. Human PWB titrated with PS-341 was stored at
0 °C, and PWB lysates were prepared and subsequently assayed at 1,
20, and 42 h. No differences were observed in the behavior of the
assays over these time points (Table 4
). Therefore, PS-341-treated whole blood can be stored on ice
for at least 42 h after collection before analysis begins.
Similarly, shipment of samples on dry ice did not adversely affect the
analysis (R.A. Newman and P.J. Elliott, personal communication).
|
Contamination during sample preparation could also be a key source of variability. The following sample preparation reagents were evaluated for their ability to affect the 20S activity in WBC lysates (concentration in assay buffer): Nycoprep Cell Separation Media (5 mL/L), phosphate-buffered saline (5 mL/L), heparin (25 USP units/mL), DMSO (5 mL/L), and EDTA (2.5 µmol/L). No significant effect was seen in either the tryptic or the chymotryptic activity.
Another source of variability for WBC lysate assays could be contamination by RBC lysate components. Only high concentrations of tested components produced any effects in our assays. Data showed that rat RBC lysate (17.4 µg) provided 33% inhibition of rat WBC lysate activity (using 3.2 µg of protein). Hemoglobin and methemoglobin (>12 µg) were found to activate the chymotryptic activity (40%). Bilirubin and biliverdin (0.3 µmol/L) were without effect. Hemin (0.3 µmol/L) provided 49% inhibition. However, complete conversion of 40 µg of hemoglobin would be required to produce this much hemin, an improbability given that typically only 20 µg of total protein was assayed. Although changes in the assays were seen, these conditions are unlikely to be encountered in biological samples and should be of little concern to future studies.
We examined repetitive freeze-thaw cycles of the PWB samples because reevaluation of samples may be necessary over time. The samples were measured for chymotryptic and tryptic activity over eight freeze-thaw cycles. The addition of 100 mL/L glycerol to the sample before freezing dramatically stabilized the activities over the freeze-thaw cycles: chymotryptic activity in the absence of glycerol was highly variable (CV = 24%), but much less variable with 100 mL/L glycerol (CV = 7%), and tryptic activity was even more dramatically affected (CV = 43% vs 6%). Because 100 mL/L glycerol could interfere with the hypotonic lysis of blood cells, it was added after cell lysis.
WBC and PWB samples from a homogeneous rat population provided
equivalent and constant data in all four assay variants (Fig. 5
). To
examine the effect of a heterogeneous human population on assay
parameters, we tested for the variance in healthy subjects
taken at random, by examining the specific activity of the proteasome
and the ratio of the chymotryptic to tryptic activity between and
within eight human volunteers (five males, three females; age range,
25–45 years) on six occasions over 6 months. Results indicated that
there was little intravariation for each subject during the testing
period in all four assays. Moreover, the parameters were constant
between individuals with SDs between 5% and 14% of the mean (data not
shown). For PWB lysate, the mean specific activity for eight volunteers
was 1.25 ± 0.18 pmol
AMC · s-1 · mg
protein-1, and
kC/kT
was 3.49 ± 0.16. For WBC lysate, the mean specific activity was
26 ± 2 pmol AMC · s-1 · mg
protein-1, and
kC/kT
was 2.10 ± 0.16. The values for
kC/kT
for PWB and WBC were different because the chymotryptic activity was
assayed under different assay conditions for the different lysate
sources (0.5 and 0.35 g/L SDS, respectively). Therefore, day-to-day and
individual-to-individual variability of baseline values can be compared
to calibrators as a judgement of the validity of results. However, we
have found that the determination of
kC/kT
for each individual before dosing is important for accurate assay
results. An individual’s
kC/kT
does not vary much with time.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendix 1References |
|---|
The pharmacokinetics of PS-341 currently are under investigation. Preliminary data suggest that it is rapidly removed from the vascular compartment and distributed widely, confirming data using radiolabeled PS-341 (3). As such, it is not yet possible to correlate plasma concentrations of the drug with the degree of proteasome inhibition in blood samples. However, the concentration of the proteasome in blood cells is ~3 µmol/L (see below), which is high relative to the Ki of PS-341 [0.62 nmol/L, compound 15 (11)]. Hence, it is expected that most of the drug that enters the cell will be bound by the proteasome. The calculations reported here demonstrate that the concentration of proteasome within red or white blood cells is equivalent (see below), and therefore, results from either biological sample should provide comparable data sets. Because the majority of whole blood (PWB) is RBCs (>99%), variations in WBC counts attributable to disease or diurnal rhythms will have insignificant impact on the final result. It should be noted that when using the degree of proteasome inhibition in WBCs as a pharmacodynamic measure over time, one should take into account the biological life span of those cells. If the cell type has a half-life of a few days, it is important to consider this when determining the rate of recovery of proteasome activity. In contrast, the half-life of RBCs is 15–17 weeks.
As such, under circumstances other than those where proteasome inhibition is being investigated in a particular cell type (e.g., in studies of leukemic cells), it is most likely that the simplest sample preparation would be used in the clinic. Thus, samples are likely to be collected, frozen without any preparation, and assayed at a later date. The current data support assays not only from isolated WBC populations but also from whole blood preparations containing mainly RBCs. Although Kahn et al. (19) previously have reported proteasome-like activity under similar assay conditions, this is the first report to clearly demonstrate quantitative proteasome activity in RBC lysates. This is highlighted by the fact that two chemically distinct proteasome inhibitors, PS-341 and lactacystin, both inhibit the chymotryptic activity in RBC lysates, which strongly suggests that these effects are attributable solely to the proteasome.
The methodology developed here is based on the fact that proteasome inhibitors affect the multiple catalytic sites differentially within the 20S core (20). Selection of appropriate peptidic substrates to determine activity at each active site was critical to this study along with the selective inhibitors of such sites. Whereas lactacystin inhibited all three sites (albeit with different affinities), PS-341 was shown to inhibit the chymotryptic activity (ß5) and actually increase tryptic activity. Based on the mechanism by which boronates inhibit serine proteases (21)(22)(23) and the mechanism by which aldehydes inhibit the proteasome (24), it is proposed that the boronate group within PS-341 interacts with the active site threonine within the proteasome. Presumably, it is this boronate:threonine binding that confers some conformational change to the tryptic subunit ß2 to induce the activation. Currently, it is not known whether the other subunits in the 20S core undergo any alteration in response to PS-341 binding to ß5.
The four assay formats, specific activity and the chymotryptic:tryptic
ratio method with either PWB or WBC lysates, provided an opportunity to
select the simplest, most sensitive, and precise assay to measure
proteasome inhibition. In addition, the development of the four assay
formats allowed cross-validation of the results. The data presented in
Fig. 5
illustrate that the percentage of inhibition, as determined by
each assay format, is internally consistent and that each assay will
give the same pharmacodynamic result, demonstrating four reliable
pharmacodynamic assays.
The variability of proteasome specific activity and the ratio of chymotryptic to tryptic activity between rats from the same population, or even between human volunteers, are quite low (CV, 5–14%). Assays developed with PWB lysates were no more variable than with WBC lysates, and preparation of PWB lysate is simpler and more reproducible than the preparation of WBC lysate. As such, the assays in PWB are likely to be chosen in future studies.
In the validation of this assay, we realized that nearly all of the
data necessary to calculate the concentration of 20S proteasome in
blood cells had been collected. Because cellular content of the 20S
proteasome is high (25), we were interested in how this
value would influence our expectations of the pharmacodynamics of
PS-341. The concentration of proteasome within a cell that can be
converted to the 20S proteasome form can be calculated if the specific
activity of purified 20S proteasome is known and if there are no
interfering inhibitors or activators in the crude lysate. Table 1
demonstrates that this assumption was accurate to within a factor of 2
for the chymotryptic activity. The protein content of blood cells
and the specific activity of each sample allowed the cellular content
of the 20S proteasome to be determined (Table 5
, row 3). The content determined for RBC lysate was
lower than the previously determined 0.28% ± 0.04%, which
was calculated based on immunoreactivity (25). The source of
this discrepancy is not understood, but the consistency of the results
detailed below lend confidence to the value derived in this report. The
numbers given within this report are within the same range as those
given by Tanaka et al. [0.07–1.0% (25)].
|
Although the proteasome content of several cell types, tissues, and
organs has been determined previously (25)(26),
the actual cellular concentration of 20S proteasome has been determined
only for yeast (27), which has a proteasome concentration in
the total cell of 0.4–0.7 µmol/L. Using the data from Table 5
and
the indicated values from the literature, we calculated that the
cellular concentration of 20S proteasome is 1–6 µmol/L in WBCs and
1.9–4.1 µmol/L in RBCs. Therefore, the concentration of 20S
proteasome in WBCs and RBCs is equivalent, and as such, data collected
from either cell type should reflect similar proteasome inhibition.
Finally, it is unlikely that full inhibition of the proteasome would be
observed with PS-341 because of the high cellular proteasome
concentrations. Thus, the cellular response to PS-341 will be dictated
by the availability of the drug rather than the proteasome
concentration within the cell.
In conclusion, a novel approach to determining the degree of inhibition has been applied to the proteasome in blood. The assay consists of measuring proteasome activity at two sites (chymotryptic and tryptic) within the 20S core of the proteasome and determining the degree of inhibition conferred by PS-341. Variations of the assay allow similar data sets to be calculated in subpopulations of blood cells, assuming that certain precautions are taken to avoid RBC contamination. Currently, the assays are being explored in phase I clinical trials to validate the methods and to determine the optimum conditions for collection, storage, and preparation of samples. To date, the assays have provided a source of real-time pharmacodynamics on individual patients dosed with PS-341. In the future, these results may be invaluable for correlation with drug blood concentrations, safety, and clinical activity data. In addition, the resolution of issues around measuring blood cell proteasome inhibition may be applicable to other biological material (e.g., tissue biopsy). Finally, the current cuvette-based assay appears readily adaptable to a high-throughput 96-well plate format.
|
Appendix 1 |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendix 1References |
|---|
The equations describing how the measurement of the two activities in a lysate treated with PS-341 can be used to determine the fraction of 20S bound by inhibitor are derived below. These equations apply to any enzyme that has at least two discrete catalytic activities that are differentially affected by the inhibitor of interest.
Let kC and
kT be the apparent rate constants for
the chymotryptic and tryptic sites, respectively, under a
defined set of assay conditions (in the absence of inhibitor):
 | (1) |
 | (2) |
where [20S]TOT is the total proteasome concentration.
In the presence of a proteasome modifier (e.g., an inhibitor) that
leads to formation of an E · I complex, the rate constant for
chymotryptic and tryptic sites may be altered by the single molecule of
modifier binding to an unidentified site. This effect can be
represented by 
CkC
and TkT, where
= 0 indicates total inhibition by the modifier (i.e.,
E · I complex has no activity); <1 indicates partial
inhibition (i.e., E · I complex is less active than E); =
1 indicates no inhibition (i.e., E · I complex is as active as E);
and >1 indicates activation (i.e., E · I complex is more
active than E).
At a given fraction of modified proteasome (f =
[E · I]/[E]TOT):
 | (3) |
 | (4) |
 | (5) |
 | (6) |
The parameter
kC/kT
is experimentally determinable. The parameter
kC/kT
is dependent on the assay conditions used for the measurement of the
chymotryptic and tryptic activities but is not dependent on the nature
of the inhibitor. The parameters C and
T are dependent on the identity of the
inhibitor. Once
kC/kT,
C, and T are known
for a defined set of assay conditions for a particular inhibitor, the
chymotryptic and tryptic activities of a crude sample can be used to
calculate the degree of proteasome inhibition by the inhibitor.
In the specific case described in the Results, where
C = 0:
 | (7) |
|
| (8) |
 | (9) |
where %I is the percentage of inhibition of the chymotryptic activity of the 20S proteasome.
For comparison, the percentage of inhibition (%I) of
the chymotryptic activity of the 20S proteasome was calculated
independently using specific activity from the following equation:
|
| (10) |
where SpAI is the chymotryptic specific activity of the proteasome in the presence of inhibitor, and SpAU is the chymotryptic specific activity of the proteasome in the absence of inhibitor.
In studies using blood samples from subjects treated in vivo with
PS-341, a baseline sample (untreated or treated with vehicle) from the
same study was used to determine the
kC/kT
with T = 1.28, as determined empirically with
human blood samples, unless otherwise indicated.
|
Footnotes |
|---|
2 Nonstandard abbreviations: DMSO, dimethyl sulfoxide; PWB, packed whole blood (all circulating blood cells); RBC, red blood cell; WBC, white blood cell (mononuclear cell); SDS, sodium dodecyl sulfate; AMC, 7-amido-4-methylcoumarin; LLVY-AMC, succinyl-Leu-Leu-Val-Tyr-AMC; and VGR-AMC, benzoyl-Val-Gly-Arg-AMC.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionAppendix 1References |
|---|
B in streptococcal cell wall-induced polyarthritis. Proc Natl Acad Sci U S A 1998;95:15671-15676.
-lytic protease and a tightly bound peptide boronic acid. Biochemistry 1987;26:7609-7614.
[Medline]
[Order article via Infotrieve]
-lytic protease complexes with analogues of reaction intermediates. Biochemistry 1989;28:7600-7609.
[Medline]
[Order article via Infotrieve]
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  R. L. Schilsky, G. Gordon, T. M. Gilmer, S. A. Courtneidge, L. M. Matrisian, O. Grad, W. G. Nelson, and on behalf of the Translational Research Working Gr The Translational Research Working Group Developmental Pathway for Anticancer Agents (Drugs or Biologics) Clin. Cancer Res., September 15, 2008; 14(18): 5685 - 5691. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Lu, J. Yang, X. Song, S. Gong, H. Zhou, L. Guo, N. Song, X. Bao, P. Chen, and J. Wang Point Mutation of the Proteasome {beta}5 Subunit Gene Is an Important Mechanism of Bortezomib Resistance in Bortezomib-Selected Variants of Jurkat T Cell Lymphoblastic Lymphoma/Leukemia Line J. Pharmacol. Exp. Ther., August 1, 2008; 326(2): 423 - 431. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. N. Utecht and J. Kolesar Bortezomib: A novel chemotherapeutic agent for hematologic malignancies Am. J. Health Syst. Pharm., July 1, 2008; 65(13): 1221 - 1231. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Kraus, E. Malenke, J. Gogel, H. Muller, T. Ruckrich, H. Overkleeft, H. Ovaa, E. Koscielniak, J. T. Hartmann, and C. Driessen Ritonavir induces endoplasmic reticulum stress and sensitizes sarcoma cells toward bortezomib-induced apoptosis Mol. Cancer Ther., July 1, 2008; 7(7): 1940 - 1948. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Shanker, A. D. Brooks, C. A. Tristan, J. W. Wine, P. J. Elliott, H. Yagita, K. Takeda, M. J. Smyth, W. J. Murphy, and T. J. Sayers Treating Metastatic Solid Tumors With Bortezomib and a Tumor Necrosis Factor-Related Apoptosis-Inducing Ligand Receptor Agonist Antibody J Natl Cancer Inst, May 7, 2008; 100(9): 649 - 662. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
) were fitted to
v = vu/(1 +
I/Ki,app), where
vu = 49 ± 3 fmol AMC/s, and
Ki,app = 24 ± 5 nmol/L PS-341.
The tryptic data (
) were fitted to v =
vu(1 +
