| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Biotium, Inc., Hayward, California, USA; and
Department of Molecular and Cell Biology, University of California, Berkeley, California, USA
1Correspondence: Biotium, Inc., 3423 Investment Blvd., Ste. 8, Hayward, CA 94545, USA. E-mail: dcen{at}biotium.com
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: apoptosis • fluorescence • imaging • nucleic acid • nucleus
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Therefore, cell lysis prevents the observation of heterogeneous changes that occur in individual cells within a population, and the assay results are simply a population-averaged measurement, often with greatly reduced sensitivity. Methods such as microinjection and hypotonic shock have been employed to load enzyme substrates into cells (2
, 3)
. However, these procedures tend to be cumbersome and disturb the normal cell cycle progression and viability of cells. Thus, they are not suitable for kinetic monitoring of enzyme activation in live cells.
Several attempts have been made to construct highly cell-permeable substrates whose signals are readily retained in the cell to detect intracellular enzyme activity in intact cells. One well-established approach is to add a lipophilic tail to the substrate, such as the lipophilic analogs of fluorescein di-β-D-galactopyranoside (FDG), a β-galactosidase substrate (4)
. These nonfluorescent FDG analogs can pass freely through the cellular membrane under normal physiological or culture conditions, likely by incorporation of their lipophilic tails into the cell membrane. The cleaved fluorescent products are believed to remain on the membrane and, therefore, are retained in the cell. A similar strategy was employed to derive a caspase-3 substrate, N-Ac-DEVD-N'-octyloxycarbonyl-rhodamine 110 (5)
. The cell penetration and retention properties of this substrate are enhanced compared to (Z-DEVD)2-rhodamine110, which has poor cell permeability due to six carboxylic acid groups present in the two peptides. However, N-Ac-DEVD-N'-octyloxycarbonyl-rhodamine110 requires prolonged incubation (2 h at 37°C) for intracellular caspase-3 detection, either due to inefficient cleavage by caspase-3 or insufficient cellular permeability. Thus, this substrate is not suitable for kinetic measurements of caspase-3 activity, especially for cell types such as Jurkat whose apoptotic induction is normally completed within 4 h.
The intracellular retention of the cleaved fluorescent product of a substrate can be increased by incorporation of a mildly nucleophile-reactive group into the substrate. After introduction of this type of substrate in the cell, the nucleophile-reactive group reacts with various nucleophiles and results in covalent attachment to intracellular peptides or proteins. This reaction should increase retention of the cleaved fluorescent product within the cell and prevent its diffusion out of the cell. Based on this theory, caspase-3 substrates with a nucleophile-reactive group, such as N-Ac-DEVD-N'-(polyfluorobenzoyl) rhodamine110, were developed (6)
. However, when used to detect caspase-3 activity in live cells, this substrate required a long incubation time with relatively low signal-to-noise ratio.
A recently developed caspase-1-cleavable peptide substrate [Gly-Trp-Glu-His-Asp-Gly-Lys (fluorescein)-Cys-NH2] contains an N-terminal near-infrared fluorochrome, Cy5.5, attached to a biocompatible, partially pegylated poly-L-lysine, which creates a cell-permeable caspase-1 near-infrared fluorescent probe (7)
. On average, each polymer molecule contained 18 reporter dye molecules that efficiently autoquenched fluorescence. This probe was shown to successfully stain apoptotic cells in cell culture and in vivo with a relatively short incubation time. However, the complexity and lack of uniformity of the probe is not ideal for experimental use.
Due to the limitations of currently available reagents for live cell detection of intracellular enzyme activity, we set out to develop an innovative molecular imaging agent that could be used for real-time detection and quantification of intracellular caspase-3 activity. In this study, we report the synthesis and characterization of DEVD-NucView488, a novel class of caspase-3 substrate.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, was dissolved in 10 ml N,N-dimethylformamide. Triethylamine (24 µl) and O-(N-succinimidyl)-1,1,3,3–tetramethyluronium tetrafluoroborate (50 mg, 0.166 mmole, from Aldrich, Milwaukee, WI, USA) were added successively. After the solution was stirred for 20 min at room temperature, N-[Ac-Asp(OBu-t)-Glu(OBu-t)-Val-Asp(OBu-t)]ethylenediamine (100 mg, 0.137 mmole) (Biotium, Hayward, CA, USA) was added. The resulting solution was stirred at room temperature overnight and then poured into 30 ml water. The orange solid was collected by suction filtration and then purified by silica gel column using methanol/chloroform (10–15%) to give the protected substrate. To a stirred suspension of the substrate (
130 mg) in 8 ml cooled dichloromethane, 4 ml 50% trifluoroacetic acid in dichloromethane was added dropwise. The final product was stirred continuously in an ice-water bath for 1 h and then for 5 h at room temperature. The solvent was removed by rotary evaporation. The orange residue was stirred in ethyl acetate and collected by filtration, and the final product was dried under vacuum to give pure DEVD-NucView488 (80 mg). DEVD-NucView488 was analyzed by thin-layer chromatography (TLC), HPLC, and liquid chromatography-mass spectrometry (LC-MS). NucView488 was prepared using the same method as described above except that the protected tetrapeptide was replaced with N-t-Boc-ethylenediamine (Quanta Biodesign, Powell, OH, USA).
Cell culture
The human T cell lymphoblast-like cell line Jurkat [American Type Culture Collection (ATCC), Rockville, MD, USA] was maintained in RPMI 1640 medium supplemented with 10% fetal bovine serum, antibiotics, 10 mM HEPES, 1 mM sodium pyruvate, 4.5 g/L glucose, and 1.5 g/L sodium bicarbonate. The human epithelial carcinoma cell line HeLa (ATCC) was cultured in Dulbecco modified Eagle medium (DMEM) supplemented with 10% fetal bovine serum, antibiotics and 30 mM HEPES. All cells were maintained in 5% CO2 at 37°C.
DNA binding assays
DNA binding assays were conducted in black 96-well plates (Nunc, Rochester, NY, USA). Titration of DNA for its binding to DEVD-NucView488 or NucView488 was conducted by incubating 1 µM of each dye with 0.049 to 50 µg/ml double-stranded salmon sperm DNA (Sigma, St. Louis, MO, USA) in 100 µl caspase-3 assay buffer (50 mM HEPES, pH 7.4; 100 mM NaCl; 1 mM EDTA; 0.1% CHAPS; 10 mM dithiothreitol (DTT); 1 mM PMSF; 10% glycerol) for 30 min at 25°C. Each DNA concentration was prepared in triplicate. Titration of DEVD-NucView488 and NucView488 for DNA binding was conducted by incubating 200 µg/ml double-stranded salmon sperm DNA with DEVD-NucView488 or NucView488 from 10 µM to 0.02 µM in a series of 2-fold dilution in 100 µl caspase-3 assay buffer for 30 min at 25°C. Each concentration of DEVD-NucView488 and NucView488 was prepared in triplicate. Fluorescence was detected by a SpectraMax Gemini XS microplate reader (Molecular Devices, Mountain View, CA, USA) set with a 488 nm excitation and a 520 nm emission wavelength.
Kinetic assay on DEVD-NucView488 cleavage by caspase-3
One unit of caspase-3 (BioVision, Mountain View, CA, USA) was added into 100 µl caspase assay buffer containing 10 µM DEVD-NucView488 and 3.3 µg/ml salmon sperm DNA in a black 96-well plate. Fluorescence signal was collected every 5 min as described above. Control samples were performed in the same composition but in the absence of capase-3. Inhibition studies were performed with or without 50 µM DEVD-CHO in the presence of caspase-3 and DEVD-NucView488 and measured after 60 min.
Determination of Km for DEVD-NucView488
DEVD-NucView488 was prepared in a series of 3-fold dilutions (5 µM, 1.67 µM, 0.56 µM, 0.18 µM, 0.06 µM) in caspase-3 assay buffer in duplicate. Caspase-3 was then added into each sample at a concentration of 0.67 U/ml to generate different reaction solutions. After every 1.25 min, a 50 µl sample was pipetted out of each reaction solution and added into its corresponding well in a separate black 96-well plate that contained 50 µl caspase-3 assay buffer with 200 µg/ml salmon sperm DNA and 10 µM caspase-3 inhibitor DEVD-CHO to create a time point for each concentration of the substrate during the enzymatic reaction. After seven time points were obtained for each concentration of the substrate, fluorescence signal was measured as described above. The average fluorescence signal of each data point was then calculated into the amount of DEVD-NucView488 converted into NucView488 (picomoles) according to the standard curve for each concentration of the substrate. The value derived at each time point was plotted against time and curve-fitted. V0 for each substrate concentration was obtained from each curve and then used for the Lineweaver-Burk plot to derive Km for the substrate.
Caspase-3 induction
Staurosporine was added directly to 0.5–2 x 106 Jurkat cells/ml suspensions to a final concentration of 1 µM. Likewise, 80–100% confluent HeLa cells were treated with 200 µM indole-3-carbinol (I3C) to activate intracellular caspase-3. Depending on the assay, induction times varied from 0.5 to 5 h.
Population-averaged caspase-3 assay
Jurkat cells (6x105 cells/ml) were treated with 1 µM staurosporine for 0.5, 1, 2, 3, or 4 h. Untreated Jurkat cells were incubated in parallel as controls. At the end of treatments, 100 µl Jurkat cell suspension was mixed with lysis/assay solution from the Caspase-3 DEVD-R110 Fluorometric & Colorimetric Assay Kit (Biotium, Hayward, CA, USA) and incubated at room temperature for 1.5 h. Caspase-3 activity was measured by the fluorescence generated by the cleavage of R110 from DEVD-R110. Fluorescence was detected by a SpectraMax Gemini XS microplate reader (Molecular Devices, Mountain View, CA, USA) set with a 470 nm excitation and a 520 nm emission wavelength.
Live cell caspase-3 assay
For live cell caspase-3 detection by DEVD-NucView488 in HeLa, 80–100% confluent HeLa cells on 8-chamber slides were treated with 200 µM I3C, a known inducer of apoptosis in reproductive cancer cell-lines (12)
. DEVD-NucView488 was then added into the culture medium to a final concentration of 5 µM. For some HeLa cell samples, 10 µM of the caspase-3 inhibitor DEVD-CHO was added 15 min before the addition of the DEVD-NucView488 substrate. After 30 min of incubation with DEVD-NucView488, HeLa cells were fixed with 3.75% formaldehyde for 15 min and permeabilized with 0.5% Triton X-100 for 10 min. The cells were blocked with 3% BSA/PBS for 1 h and then stained with sulforhodamine 101 (Texas Red) -phalloidin (Biotium) diluted 1:100 in 1% BSA/PBS for 30 min. The slides were mounted with glass coverslips using Vectashield with 4',6'-diamidino-2-phenylidole (DAPI) (Vector Laboratories, Burlingame, CA, USA) and sealed with nail polish. Microscopy was performed at x40 on a Zeiss Axiophot epifluorescence microscope (Carl Zeiss, Thornwood, NY, USA) with a 3.3 MPix QImaging MicroPublisher CCD camera. Images were processed using QCapture software (QImaging, Surrey, BC, Canada).
For live cell caspase-3 assay by DEVD-NucView488 in Jurkat cells, 1–10 µM DEVD-NucView488 was added directly into 0.5–1 ml staurosporine induced or uninduced Jurkat cell (0.5–1x106 cells/ml) suspension cultures. After 30 min of incubation, NucView488 stained cells were then subjected to flow cytometry analysis, either with or without washes. NucView488 stained cells were detected by the FL1 channel of an EPICS XL-MCL flow cytometer (Beckman-Coulter, Fullerton, CA, USA). For dual detection with Annexin V-PE, NucView488 stained cells were spun down and resuspended in binding buffer containing Annexin V-PE from the Annexin V-PE Apoptosis Detection Kit I (BD Biosciences, San Jose, CA, USA). Fluorescence was detected by the FL1 and FL2 channels. Data were analyzed using FCS Express v. 3 (DeNovo Software, Thornhill, ON, Canada).
For live cell caspase-3 assay by PhiPhiLuxG1D2, 0.5–1 x 106 staurosporine-induced or uninduced Jurkat cells were spun down and resuspended in 75 µl of 10 µM PhiPhiLuxG1D2 solution (OncoImmunin, Gaithersburg, MD, USA). For live cell caspase-3 assay by fluorochrome-labeled inhibitors of caspases (FLICA) in Jurkat cells, 1.7 µl of FLICA (FITC-DEVD-FMK) of CaspGLOWTM Fluorescein Active Caspase-3 Staining Kit (BioVision) was added directly into 0.5 ml of a staurosporine-induced or uninduced Jurkat cell (0.5–1x106 cells/ml) culture. After 30 min incubation with either PhiPhiLuxG1D2 or FLICA, cells were washed once and resuspended in Annexin V-PE binding buffer, and fluorescence was detected by the FL1 and FL2 channels as described above.
Assay for cytotoxicity
After incubation with DEVD-NucView488, Jurkat cells were spun down to remove supernatant and resuspended in 100 µl Annexin V binding buffer. Then 5 µl of Annexin V-PE and 5 µl 7-AAD from the Annexin V-PE Apoptosis Detection Kit I (BD Biosciences) were added into the solution. After 15 min of incubation, 400 µl of Annexin V binding buffer was added. The Annexin V-PE and 7-AAD stained cells were analyzed by the FL2 and FL4 channels as described above.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. As most DNA-binding dyes carry one or more positive charge (9)
, which presumably contributes to DNA binding via electrostatic interaction, we envisioned that attaching a highly negatively charged group, such as Ac-DEVD, to NucView488 should inactivate the dye’s DNA binding ability. However, enzymatic cleavage of the resulting substrate should release NucView488, thus restoring its functionality (see Fig. 1
A). Like thiazole orange itself, NucView488 is not fluorescent until it binds to nucleic acids such as DNA in cell nuclei. Because NucView488 is a green fluorescent dye excitable by an argon laser at 488 nm, and it stains the nuclei of caspase-3-activated cells, we named the substrate DEVD-NucView488.
|
The synthesis of DEVD-NucView488 is shown in Fig. 1B
. In the first step, a carboxyl thiazole orange derivative compound A, prepared according to a known procedure (10)
, is coupled to a t-butyl ester protected tetrapeptide compound B to give the protected substrate compound C. Standard deprotection of the protected substrate with trifluoroacetic acid yielded the final product DEVD-NucView488.
DNA binding properties of DEVD-NucView 488 and caspase-3 released NucView488
To serve as a sensitive fluorogenic substrate for detection of caspase-3 in live cells, the uncleaved form of the substrate should have no or minimal binding to DNA to ensure low fluorescence background after its entry into cells. At the same time, the caspase-3 cleaved form of the substrate should have a high affinity for DNA to generate a strong fluorescence signal in apoptotic cells within a short period of time. To determine whether DEVD-NucView488 possesses such characteristics, the substrate was tested for DNA binding against the parent dye NucView488, the nucleic acid binding moiety released after cleavage by caspase-3. As shown in Fig. 2
A, DEVD-NucView488 did not have any significant DNA binding activity while NucView488 efficiently bound to DNA. In the presence of 1 µM NucView488, the fluorescence signal reached its peak at a DNA concentration of 25 µg/ml and was at a plateau afterward, indicating saturation of 1 µM NucView488 by 25 µg/ml DNA. However, the maximum signal (DNA bound NucView488) to noise (DNA bound DEVD-NucView488) ratio occurred when the DNA concentration was 6.25 µg/ml. These DNA binding results confirmed that the covalently attached DEVD moiety effectively prevents NucView488 from binding to DNA.
|
As shown in Fig. 2B
, in the presence of 200 µg/ml DNA, the fluorescence signal generated through binding of NucView488 to DNA increases linearly from 0.1 to 10 µM, whereas DEVD-NucView488 does not significantly bind to DNA at any tested concentration of the dye. This result indicates that 200 µg/ml DNA provides sufficient binding sites for NucView488 even at the highest concentration of 10 µM.
DEVD-NucView488 is efficiently cleaved by activated caspase-3
To determine whether DEVD-NucView488 serves as an efficient fluorogenic substrate for caspase-3, DEVD-NucView488 was subjected to caspase-3 mediated cleavage in a cell-free assay. In the presence of DNA, incubation of caspase-3 with DEVD-NucView488 led to an increased fluorescence signal over time, whereas no significant change of fluorescence signal was observed in the absence of caspase-3 (Fig. 2C
). The increase of fluorescence signal was blocked by the addition of DEVD-CHO, an inhibitor of caspase-3 (Supplemental Fig. 1
), which suggests that the substrate is specific for caspase-3.
A caspase-3 kinetics study was carried out to determine the Km of DEVD-NucView488. The substrate hydrolysis rate was determined at various concentrations of DEVD-NucView488. Samples of each enzymatic reaction were taken out at different time points and immediately mixed with a large excess of DNA and a sufficient amount of a caspase-3 inhibitor to "freeze" the reaction. The resulting reactions were measured for their fluorescence intensities, which were then converted into the amount of product formation at the various time points. These values were then used to derive V0 for each substrate concentration. Analysis of the reciprocal plot of 1/V vs. 1/S revealed that the Km of DEVD-NucView488 is 2.3 µM (Fig. 2D
). This Km value suggests that DEVD-NucView488 has a higher affinity for caspase-3 than the conventional DEVD-AMC substrate (Km=9.7 µM) (11)
.
DEVD-NucView488 is a highly cell-permeable fluorogenic caspase-3-specific substrate whose cleaved dye labels nuclei
Because apoptotic mechanisms are well-established in HeLa, an adherent human cervical carcinoma cell line, it was used as a model system to test whether DEVD-NucView488 can detect caspase-3 activity within live cells by generating a fluorescence signal through binding of the cleaved product to nuclear DNA. Cultured cells were treated with 200 µM I3C, a known phytochemical inducer of apoptosis in many human cancer cell lines (12)
. We found that DEVD-NucView488 effectively stained I3C treated apoptotic HeLa cells while control cells remained unstained (Fig. 3
). Caspase-3 activated cells exhibited fluorescent green nuclear staining as predicted from the spectrophometric property of the cleaved product. The fluorescent staining of the nucleus was prevented by pretreatment with the caspase-3 inhibitor DEVD-CHO, indicating that the staining is specific to caspase-3 activity. In addition, we found that the fluorescent nuclear staining can be subjected to formaldehyde fixation for further processing of cell samples. Figure 3
also shows staining of apoptotic and nonapoptotic cells by Texas Red-conjugated phalloidin (Biotium) after NucView488 fixation.
|
DEVD-NucView488-based live cell caspase-3 assay is more sensitive than population-averaged caspase-3 assays
Jurkat cells are widely used in apoptotic studies, particularly in flow cytometry applications. Thus, Jurkat cells were used to compare the sensitivity of the DEVD-NucView488 substrate with a conventional population-averaged caspase-3 assay. Cells were treated with or without staurosporine, a protein kinase inhibitor that is known to induce caspase-3 activity and apoptosis in Jurkat cells (1)
, to determine whether DEVD-NucView488 can detect caspase-3 activation within a live cell population earlier than the conventional population-averaged caspase-3 assay using cell lysates. For the population-averaged assay, cell extracts were made from cells at 0, 0.5, 1, 2, and 4 h after induction of apoptosis by staurosporine. For the live cell assay, DEVD-NucView488 was added directly to Jurkat cells and, after 30 min of incubation, the cells were subjected to flow cytometry analysis. As shown in Fig. 4
A, DEVD-NucView488 detected an increase in apoptotic cell number as early as 30 min after staurosporine induction. When caspase-3 activity was measured using the highly sensitive caspase-3 substrate DEVD-R110 in a population-averaged assay (Fig. 4B
), activation of caspase-3 was detected after 2 h, which is 1.5 h later than detection by DEVD-NucView488. Thus, the necessary substrate incubation time with DEVD-NucView488 is as short as 30 min and takes 1/3 of the assay time needed for the population-averaged caspase-3 assay.
|
DEVD-NucView488 is not cytotoxic and does not interfere with apoptosis progression
To determine whether DEVD-NucView488 has any cytotoxic effects, staurosporine-induced and uninduced Jurkat cell populations were incubated with the following concentrations of DEVD-NucView488 for 30 min: 0, 1, 2.5, 5, or 10 µM. Jurkat cell populations were then stained with PE conjugated Annexin V (Annexin V-PE) for identification of apoptotic cells and 7-AAD for identification of dead cells by flow cytometry. Annexin V is a 35 kDa Ca2+-dependent phospholipid protein with a high affinity for phosphatidylserine (PS), which is translocated from the inner to the outer surface of a cell’s plasma membrane during apoptosis (13)
. Fluorescently labeled Annexin V is commonly used to identify apoptotic cells by binding to PS exposed on the outer membrane leaflet. 7-AAD is a nucleic acid dye that is not cell-permeable and stains only dead or late-stage apoptotic cells that lose plasma membrane integrity. As shown in Fig. 5
A, DEVD-NucView488 did not cause any significant change in the profile of Annexin V and 7-AAD stained cell populations. This result indicates no significant cellular toxicity is associated with this substrate within the 30-min assay time.
|
DEVD-NucView488 could be ideal for studying the kinetics of caspase-3 activation in live cells since it can be added directly into cell culture medium for fluorescence detection by flow cytometry or microscopy without further manipulation. To determine whether DEVD-NucView488 is suitable for long-term kinetic studies, staurosporine-induced and uninduced Jurkat cells were incubated with or without 1 µM DEVD-NucView488 for various times. The percentage of apoptotic cells was determined before apoptosis induction and 2 or 4 h after staurosporine treatment. As shown in Fig. 5B
, the percentage of caspase-3 activated cells remained the same between control and DEVD-NucView488 incubated cell populations during the course of apoptosis progression. DEVD-NucView488 does not appear to interfere with the progression of apoptosis in Jurkat cells.
Comparison of DEVD-NucView488, PhiPhiLuxG1D2, and FLICA for detection of activated caspase-3 in live cells
To further delineate the unique features of DEVD-NucView488, the performance of this substrate was compared directly to two other agents capable of detecting live apoptotic cells, PhiPhiLuxG1D2 and FLICA. PhiPhiLuxG1D2 is also a cell-permeable caspase-3 substrate (14)
but lacks any cell retaining functional group. FLICA is a fluorochrome-labeled inhibitor of caspases used to detect caspase activation within live cells by affinity labeling the enzyme active center (15)
.
To compare the sensitivity of these three agents and their correlation with Annexin V staining, uninduced and staurosporine-induced apoptotic Jurkat cells were incubated with DEVD-NucView488, PhiPhiLuxG1D2, or FLICA for 30 min, followed by staining with Annexin V-PE. As shown in Fig. 5C
, among the three agents tested, DEVD-NucView488 generated the strongest fluorescence shift of staurosporine-induced apoptotic cell population from the uninduced cell population. The ratio between the maximum fluorescence signal of induced and uninduced cell populations (signal-to-noise ratio) for DEVD-NucView488 was 8-fold, while the ratios for PhiPhiLuxG1D2 and FLICA were 4- and 1.2-fold, respectively.
As shown in Fig. 5C
, Annexin V-PE detected an almost identical increase in apoptosis after staurosporine induction in the Jurkat cell populations: 93% for DEVD-NucView488, 91.2% for PhiPhiLuxG1D2, and 93% for FLICA. However, the percentage of apoptotic cells detected by these three agents varied dramatically: 86.6% for DEVD-NucView488, 55.3% for PhiPhiLuxG1D2, and 16.3% for FLICA. These data suggest that the percentage of apoptotic cells detected by DEVD-NucView488 most closely matched the number of apoptotic cells as determined by Annexin V-PE staining.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
17
18
19
20)
. Most notably, DEVD-NucView488 offers a unique assessment of apoptosis: real-time kinetic activation of caspase-3 and visual observation of nuclear apoptotic morphology. This substrate may be particularly useful in cells where apoptosis is difficult to measure.
First, DEVD-NucView488 is highly cell-permeable, and the cleaved dye NucView488 is retained in cells due to its affinity for DNA. In contrast, PhiPhiLuxG1D2 is also cell-permeable but lacks any cell-retaining functional group. As a result, the cleaved substrate lacks the ability to stay within apoptotic cells if the integrity of the plasma membrane is compromised and can slowly diffuse out of cells even if the membrane remains intact. Therefore, PhiPhiLuxG1D2 is not an ideal substrate for assessing caspase-3 activation in cells that lose plasma membrane integrity in the late stages of apoptosis. The enzyme recognition sequence DEVD within DEVD-NucView488 contains three negative charges that neutralize the positive charge present on the NucView488 chromophore and renders the substrate molecule to be negatively charged. This negative charge minimizes the DNA binding of DEVD-NucView488 compared to the parent dye NucView488. An increased DNA concentration does lead to a small increase of DEVD-NucView488 binding to DNA (Fig. 2A
). However, in view of the high concentration of DNA content within cells (12–348 mg/ml) (21)
, we anticipated a smaller signal-to-noise ratio (NucView488 stained caspase-3 activated cells vs. DEVD-NucView488 stained caspase-3 inactivated cells) than predicted based on cell-free DNA binding assays (Fig. 2B
). Based on our live cell studies, the signal-to-noise ratio is 8 using 5 µM DEVD-NucView488 (Fig. 5C
); this is much smaller than the predicted ratio of 53 at the same concentration in the presence of 200 µg/ml DNA (Fig. 2B
). Nevertheless, many factors such as incomplete cleavage of DEVD-NucView488 could contribute to the decreased signal-to-noise ratio in live cell staining in comparison to the DNA binding assay.
Second, DEVD-NucView488 is efficiently cleaved by activated intracellular caspase-3, and its cleaved product NucView488 efficiently labels nuclear DNA. We observed, however, that DNA can inhibit the cleavage of DEVD-NucView488 by caspase-3 to some degree (data not shown). For this reason, the cleavage rate of DEVD-NucView488 by caspase-3 in Fig. 2D
was measured in the absence of DNA before termination of the reaction by a capase-3 inhibitor. A large excess of DNA was then added to quantify the amount of NucView488 formed from the reaction at each specified time point. We found that the Km of DEVD-NucView488 is 2.3 µM (Fig. 2D
). This Km value suggests that DEVD-NucView488 has a higher affinity for caspase-3 than the conventional DEVD-AMC substrate (Km=9.7 µM) (11)
. The effect of DNA on the enzyme kinetics should not be of major concern for intracellular caspase-3 detection because DNA is mostly located within cell nuclei, whereas caspase-3 is localized in the cytoplasm. Thus, the majority of DEVD-NucView488 molecules are expected to encounter caspase-3 before diffusing into nuclei. The DNA concentration in the nuclei of mammalian cells ranges from 12 to 348 mg/ml (21)
; this ample amount of DNA in cell nuclei should provide a large dynamic range for detecting caspase-3 or other intracellular enzyme activity using a similarly designed functional substrate.
Third, in live Jurkat cell studies, DEVD-NucView488 did not appear to be cytotoxic or interfere with the progression of apoptosis. Therefore, a low concentration (1 µM) of DEVD-NucView488 could be used to kinetically monitor live cell caspase-3 activation for several hours. This is a significant benefit for real-time studies in intact cells that may take several hours to complete and offers a new approach for live cell imaging.
Fourth, DEVD-NucView488 is more sensitive than currently available reagents for detection of caspase-3 activity. As shown in Fig. 4A, B
, DEVD-NucView488 can detect caspase-3 activation in staurosporine-treated Jurkat cells much earlier than the conventional population-averaged caspase-3 assay using cell lysates. As shown by Smolewski et al. (1)
, cells within a population undergo apoptosis at different rates. For example, in TNF-
-treated HL-60 cells, 50% of cells underwent apoptosis during the initial 6 h of treatment at an approximate rate of 8% of cells per hour. The remaining cells underwent apoptosis between 6 and 24 h at a rate of 2.5% of cells per hour. Conventional assay systems require cell lysis for caspase-3 detection due to the poor permeability of caspase-3 substrates. Following cell lysis, the detected caspase activity becomes the average activity of caspase-3 from a heterogeneous cell population. In addition, the population-averaged assay may prove challenging for detecting a relatively small increase in caspase activity within a cell population. In some cell populations that undergo spontaneous apoptosis, a significant background will likely accompany population-averaged methods of detecting caspase activity. Consequently, when a small percentage of cells within the population undergo apoptosis after induction, the total increase of caspase-3 activity from the newly apoptotic cells may be too small to be detected due to the presence of existing background. With DEVD-NucView488, caspase-3 activity can be determined on an individual cell basis, and the spontaneously apoptotic cells can be gated-out through flow cytometry. Thus, with DEVD-NucView488, researchers can profile a heterogeneous cell population based on caspase-3 activity. Furthermore, researchers will also be able to correlate caspase activity in individual cells with their apoptotic progression based on morphological changes in cell nuclei and expression level and location of a specific protein through subsequent immunostaining.
Lastly, in addition to the sensitivity of DEVD-NucView488, another key advantage of DEVD-NucView488 over reagents such as PhiPhiLuxG1D2 or FLICA is the ease of application. For example, no washing steps are required before addition of DEVD-NucView488 (prewash) or after incubation of the substrate (postwash). Both pre- and postwashes are required with PhiPhiLuxG1D2, and postwashing is required with FLICA. Apoptotic cells are typically fragile and could be easily damaged or lost during these washes, especially when wash procedures involve centrifugation and resuspension steps. Thus, any result obtained after the washing step may not reflect the initial apoptotic profile of the cell population. Moreover, postwashing also prevents application of the reagent for kinetic enzyme assays. Flow cytometry results from this study with DEVD-NucView488, PhiPhiLuxG1D2, and FLICA, revealed that the percentage of Annexin V positive cells most closely matched that of DEVD-NucView488. These data suggest that DEVD-NucView488 is the most sensitive and accurate agent among the three tested for detection of caspase-3 activation in intact cells. In view of these considerations, among the three agents tested, DEVD-NucView488 is most suitable for real-time detection of caspase-3 in live cells.
In principle, any enzyme-recognition moiety capable of inactivating a functional fluorescent dye after being linked to the dye via a cleavable bond is amenable to the construction of a detectable substrate for the enzyme. For example, the peptide Ac-DEVD may be readily replaced by another negatively charged peptide or nonpeptide enzyme-recognition moiety for the design of a different fluorogenic substrate specific for the enzyme of choice. Likewise, NucView488 may be replaced by another functional dye, including a nucleic acid dye of a different color or a dye of entirely different functionality. Similar to DEVD-NucView488, substrates of this design could measure the activity of an enzyme and at the same time detect another target. Because of the ability to retain fluorescence signal within cells, this type of substrate is especially useful for live cell applications. The success of this strategy will depend on the cell membrane permeability of the individual substrate. Nevertheless, the fact that a highly negatively charged peptide, such as Ac-DEVD, can be brought into cells by a dye such as NucView488 suggests that a large number of fluorogenic enzyme substrates could be designed in this manner. The principle utilized in the successful design of this substrate may be generally applicable to the development of fluorogenic substrates for other intracellular enzymes and provides a novel method for real-time detection of intracellular enzyme activity in living cells.
|
ACKNOWLEDGMENTS |
|---|
Received for publication October 3, 2007. Accepted for publication January 10, 2008.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, NucView488; 
, DEVD-NucView488. B) Titration of DEVD-NucView 488 and NucView488 for DNA binding. Two-fold titrations of DEVD-NucView 488 and NucView488 from 10 to 0.01 µM were applied to graph the DNA binding curve. The means of triplicate samples at each concentration of DEVD-NucView488 or NucView488 are plotted. Error bars represent SD of mean values. 
