INTRODUCTION

Two major apoptosis pathways have been defined in mammalian cells, the Fas/TNF-R1 death receptor pathway and the mitochondria pathway1, 2. TRAIL (TNF-Related Apoptosis Inducing Ligand), also called Apop2 Ligand, belongs to the TNF super-family. In 1995, TRAIL was identified and characterized as a member of the TNF family of death-inducing ligand3. As a characteristic type II transmembrane protein, TRAIL shows the highest homology with Fas-L, sharing 28% amino acid identity in the extracellular receptor-binding motif. Unlike most other members of the TNF ligand family, TRAIL has a very short cytoplasmic domain, comprising only 17 amino acids. In addition, no conservation exists within the cytoplasmic region between the human and murine sequences, suggesting that this domain is unlikely to be involved in transmitting cellular signals3.

TRAIL is widely distributed in normal tissues and cells 3. A recent study reported that TRAIL is present on the cell surface of mouse T and B cells. CD3+ T cells were found to express TRAIL upon activation with the mitogen phorbol myristate acetate plus ionomycin or via the T cell receptor (anti-CD3) in an activation-dependent manner4. Similarly, activation of human T cells with PMA plus ionomycin, anti-CD3, or super antigen was able to induce the expression of TRAIL3. Initially it was reported that T cell-enriched cultures of peripheral blood lymphocytes stimulated with interleukin-2 were sensitive to TRAIL-mediated killing5. However, these findings have not been reproduced3. So far, the only reproducible effects of TRAIL in mediating apoptosis have been reported in transformed cell line3, 4, 5 and hematological malignancies4, 7. Other researches of TRAIL were mainly focused on its signaling pathways and its selectivity to induce tumor cell apoptosis8, 9, 10, 11. To the best of our knowledge, there have been no reports as to how it affects the biophysical properties of tumor cells.

As it is well known, RevTet-On gene expression system12 containing Dox possesses excellences such as : the strict open/close function, the strong singularity, the high inducement ratio and expression level and the low toxicity. So the system provides a powerful tool to study eukaryotic gene expression. In the present study, we constructed a Tet-On controlled gene expression system of TRAIL in the human leukemic cell line Jurkat, which was used as a cell model to study the changes in biophysical properties of the Jurkat cells.

MATERIALS AND METHODS

Construction of retroviral plasmids

The RevTet-On system, which includes pRevTet-On and pRev-TRE vectors and the RetroPack PT67 cell line, was purchased from CLONTECH, Palo Alto, California, USA. TRAIL gene was obtained from a human placenta cDNA library (CLONTECH) by PCR. The PCR primers were: 5′-AAGCTTATGGCTATGATGGAGGT-CCAGGGGGGAC-3′ and 5′-AAGCTTTTAGCCAACTAAAAA-GGCCCCGAAAAAACTGGC-3′. The PCR product was cloned into an intermediate vector PCR-2 (Invitrogen, Carlsbad, California, USA) and then linked into pLNCX to form vector pLNCX-TRAIL. The Tet-regulated vector pRev-TRE-TRAIL was constructed in the same manner.

Cell culture

The retroviral packaging cell line PT67 and the human leukemic cell line Jurkat were maintained in Dulbecco's modified essential medium and RPMI 1640 containing 10% fetal bovine serum.

Transfections of packaging cells PT67

The packaging cells PT67 were cultured in a 60-mm plate at a density of 50-80%, 24 h before transfection. The cells were washed with DMEM twice and incubated with 2 ml DMEM before transfection with 10 μg of plasmid DNA by Lipofectin method (Life Technology, GIBCO BRL). To obtain stable virus-producing cell lines, the packaging cells transfected with retroviral plasmids were placed in selection medium 48 h later. The regulatory vector Tet-On carrying the neomycin gene served as a selectable marker. For G418 selection, cells were cultured in presence of G418 (0.4 mg/ml, GIBCO) for two weeks. The cells carrying pRev-TRE and pRev-TRE-TRAIL were selected in the presence of hygromycin B (0.06 mg/ml, SIGMA) for two weeks.

Establishment of RevTet-On TRAIL gene expression system in Jurkat cell line11

Jurkat cells were kept for 24 h before infection. The medium from packaging cells containing virus were collected, filtered through a 0.45 μm filter, and added to the Jurkat cells in the presence of Polybrene (4 μg/μl, SIGMA). The medium was replaced 4 h later. 3-6 serial infections were performed to increase the efficiency of infection. 48 h after infection, the cells were subjected to G418 or hygromycin selection. The Jurkat cells, which were serially infected with virus-containing media, pRev-Tet-On virus and pRevTRE-TRAIL. Hence they possessed stable resistance to G418 and hygromycin, and were called Jurkat-RevTet-On-TRE-TRAIL(Jurkat3T).

Induction of gene expression and measurement of apoptosis by the bi-parameter method

Cells mentioned above were added to a medium containing Dox (0, 0.01, 0.1, 1, 5, 10 μg/ml, SIGMA). After 48 h, TRAIL protein expression was detected by Western blot. The apoptosis ratio of cells was measured by Flow cytometry.

Confocal microscopy (CLSM) analysis

2×106 cells were washed twice with PBS (phosphate-buffered saline, pH 7.4). The sample was fixed in 3.7% formaldehyde for 10 min at room temperature. After washing twice in PBS, the cells were resuspended in 0.1% Triton X-100/PBS for 5 min at room tem-perature. Then, the cells were incubated in 1% BSA/PBS for 30 min to reduce the nonspecific binding and observed with a confocal laser scanning microscope (Leica Lasertechnik, Germany). A 580 nm long pass filter was used in the fluorescence detection path. The images were collected using a 100×oil immersion objective. Series of 2× optically zoomed confocal sections, 1-2 mm apart, were scanned, with each image averaged by 16 line scans.

Measurement of cell electrophoretic mobility

Jurkat and Jurkat-RevTet-On-TRE-TRAIL induced with Dox (0-10 μg/ml) for 48 h were collected by centrifugation (1000 rpm, 5min). The cell concentration was adjusted to 2×106 cells/ml. The cell electrophoretic mobility was measured with a LIANG-100 electrophoresis meter (Shanghai Medical University, Shanghai, China). 100 cells were randomly selected for measurement.

Measurement of membrane fluidity

Jurkat and Jurkat-RevTet-On-TRE-TRAIL induced with Dox (0-10 μg/ml) for 48 h were collected by centrifugation (1000 rpm, 5 min). After washing twice with PBS, the cell concentration was adjusted to 2×106 cells/ml. Each cell solution was mixed with the same volume of DPH solution (2×10−6M). After 30 min incubation (37°C) and centrifugation (1000 rpm, 5 min), the supernatant was removed and the cells were resuspended in PBS. The fluorescence polarization angle P of each tube was measured with a Fluorescence Spectrophotometer within 2 h and calculated according to following equation:

P=(IVV -GIVH)/(IVV +GIVH)

where IVV was the fluorescence identity when polarizer and polarization checker were both at a vertical angle. IVH was the fluorenscence identity when the polarizer was at a vertical angle and the polarization checker was at a horizontal angle. G was the correction factor. According to Azumi13, G=IHV/IHH, where IHH was the fluorescence identity when the polarizer and polarization checker were both at horizontal angles. IHV was the fluoresence identity when the polarizer was at a horizontal angle and the polarizaition checker was at a vertical angle. The microviscosity η, which represents the membrane fluidity, was then calculated from the value of P13:

η=2P/(0.46-P).

Measurement of the change of secondary structure in cell protein

Jurkat and Jurkat-RevTet-On-TRE-TRAIL induced with Dox (0-10 μg/ml, SIGMA) for 48 h were collected by centrifugation (1000 rpm, 5 min). Then distilled H2O was added to each tube. 30 min later, the supernatant of each tube was removed by centrifugation (10,000 rpm, 5 min). 100 μl of heavy water was added to each tube. 30 μl solution of each tube was taken out for measurement to obtain the proportion of secondary structure in different cell proteins by using an FT-IR Spectrometer (Bio-RAD Company). The scan solution was 4cm−1 and the scan range was 1500-1800cm−1.

Measurement of cell viscoelastic properties13

Jurkat and Jurkat-RevTet-On-TRE-TRAIL induced with Dox (0-10 μg/ml) for 48 hours were collected by centrifugation (1000 rpm, 5 min). Cell viscoelastic properties were measured with micropipette aspiration system, which was composed of an inverted microscope, a micromanipulator, a video recorder, a pressure control and recorder system, and a pipette. 0.5 ml of the cell suspension (1×106 cells/ml) was placed in a chamber located on the specimen stage of the microscope. The internal diameters of the micropipette were 2.4–3.1 μm. The time course of cell deformation was continuously recorded on the video recorder. The sequential photographs were taken from the recorded video image during single-frame replay on the video monitor every 120 ms. The length of the cell tongue aspirated into the micropipette was determined as a function of time. The time history of deformation typically showed an initial rapid phase followed by a slow creep, similar to the behavior of peripheral blood leukocytes studied previously13. A standard solid viscoelastic model was used to fit the experimental data. The equation of this model was as follows:

Where σ and ɛ are the stress and the strain, and ∂σ/∂t and ∂ɛ/∂t are partial derivatives of the stress and the strain as functions of time, respectively. K1 and K2 are elastic elements, and μ is a viscous element.

Each sample was scanned 200 times. Student t-test and ANOVA were used for the statistical analysis of experiment data with the statistic analysis software SPSS.

RESULTS

Establishment of RevTet-On TRAIL gene expression system and induction of TRAIL in Jurkat cell line

From recombinant pRevTet-On after purification, the 7.6 kb fragment was obtained using Bam HI, and from recombinant Plasmids pRev-TRE, the 3.6 kb and 2.8 kb fragments were obtained using SspI and Hind III. This was consistent with the restriction enzyme map of Plasmids in Fig 1A.

Figure 1
figure 1

Plasmids construction. (A) 1. Marker; 2. pRev-Tet-On with Bam HI; 3. pRev-TRE-TRAIL with Ssp I+ Hind III. (B) 1.Marker 2. pLNCX-TRAIL with Hind III. 3. pRev-TRE-TRAIL with Hind III; 4. pRev-TRE-TRAIL with Ssp I.

Full size image

From recombinant plasmid, 6.5 kb and 845 bp were obtained after Hind III enzyme cutting. It showed that the objective gene TRAIL has been cloned into the recombinant Plasmid. Moreover, in order to identify the orientation, from recombinant Plasmid, the 4.0 kb, 2.9 kb and 413 bp fragments were obtained after SspI enzyme cutting, as shown in Fig 1B, which showed that the result was consistent with the restriction enzyme map of the plasmids. This fact indicated that the objective gene TRAIL has been correctly linked to vector pRev-TRE and it was called pRev-TRE-TRAIL.

By plasmids transfection and antibiotics selection, we constructed a Tet controlled gene expression system of TRAIL in Jurkat. Fig 2 shows the growth curves of Jurkat, Jurkat2T and Jurkat3T. And as shown in this figure, there was no significant difference between the three cell lines.

Figure 2
figure 2

Growth curves of Jurkat, Jurkat 2T and Jurkat 3T.

Full size image

Fig 3 shows the results of Western blot analysis of TRAIL protein expression. Lane 1 indicates the protein markers. Protein in lane 2 was isolated from Jurkat-RevTet-On–TRE-TRAIL induced with Dox (10 μg/ml) for 48 h and that in lane 3 was isolated from Jurkat -RevTet-On–TRE-TRAIL and kept for 48 h without Dox induction (0 μg/ml). The band in lane 2 indicates that the TRAIL protein (molecular weight about 37 kD) was expressed when Jurkat -RevTet-On–TRE-TRAIL was induced with Dox (10 μg/ml) for 48 h. The fact that this band was absent in lane 3 indicated that TRAIL protein was not expressed when Jurkat -RevTet-On–TRE-TRAIL had not been treated with Dox.

Figure 3
figure 3

Western blot analysis of TRAIL protein expression. Lane 1: protein markers, Lane 2: protein isolated from Jurkat RevTet-On-TRE-TRAIL induced with Dox (10 μg/ml) for 48 h, Lane 3: protein isolated from Jurkat-RevTet-On-TRE-TRAIL not induced with Dox (0 μg/ml) and kept for 48 h.

Full size image

As shown in Fig 4, compared with Jurkat3T untreated with Dox (0 μg/ml), the apoptosis ratio of Jurkat3T induced with Dox increased significantly (p<0.01) in a dose-dependent manner. The apoptosis ratio had a correlation coefficient of 0.78 with the logarithm of Dox concentration.

Figure 4
figure 4

Apoptosis ratio of Jurkat3T induced with different concentrations of Dox measured by flow cytometer. The ratio increased significantly in a dose-dependent manner. Compared with Dox (0 μg/ml) : ** indicates p<0.01

Full size image

Compared with Jurkat3T untreated with Dox (0 μg/ml), Jurkat3T cells expressing TRAIL showed distinct changes. The cells chromatin were collected under nucleus membrane and the nucleus became into crescent shape or petal, and even appeared apoptotic bodies. Furthermore, the apoptosis ratio of Jurkat3T induced with Dox increased significantly in a dose-dependent manner (Fig 5).

Figure 5
figure 5

Observation of cell nucleus undergoing apoptosis by confocal laser scanning microscopy. The nucleus of Jurkat3T treated with Dox (0 μg/ml) were regular and integral (A). However, in the cells treated with Dox (0.01 μg/ml), the chromatin were collected under nucleus membrane and the nucleus became into crescent shape or petal (B). In the cells treated with Dox (1 or 10 μg/ml), even appeared apoptotic bodies (C, D).

Full size image

As shown in Fig 6, after treating Jurkat with 1 μg /ml concentration of Dox, the Jurkat death ratio increased with time, and the death ratio of Jurkat reached maximum at 60 h.

Figure 6
figure 6

The relation curve between cell death ratio and time after treating Jurkat3T with Dox (1 μg/ml). The death ratio increased with time, and reached maximum at 60 h.

Full size image

These facts indicated that the Tet-controlled gene expression system had been constructed successfully and there was no significant influence on cell growth of Jurkat after transgene. However, the apoptosis ratio of Jurkat3T increased significantly with Dox.

Cell electrophoretic mobility

Tab 1 shows the results of measurement of the electrophoretic mobility of different groups of cells. There was no significant difference between Jurkat cells (Group 1) and Jurkat3T cells untreated with Dox (Group 2). Jurkat3T induced with different Dox concentrations (0.01, 0.1, 1, 5, 10 μg/ml in Groups 3-7, respectively) have a dose-dependent decrease in electrophoretic mobility (P<0.01). The correlation coefficient with the logarithm of the Dox concentration is -0.86.

Table 1 The electrophoretic mobility of different groups of cells (x̄ ± SD)
Full size table

Membrane fluidity

Tab 2 shows the results of microviscosity of different groups of cells. There was no significant difference between Jurkat cells (Group 1) and Jurkat3T cells untreated with Dox (Group 2). Jurkat3T induced with different Dox concentrations (0.01, 0.1, 1, 5, 10 μg/ml in Groups 3-7, respectively) have a dose-dependent increase in micro-viscosity (p<0.01); the correlation coefficient with the logarithm of the Dox concentration is 0.91.

Table 2 The membrane microviscosity of different groups of cells (x̄ ± SD)
Full size table

The change of secondary structure in cell proteins

There was no significant difference between Jurkat cells (Group 1) and Jurkat3T cells untreated with Dox (Group 2). While in Jurkat3T cells induced with different Dox concentrations (0.01, 0.1, 1, 5, 10 μg/ml in Groups 3-7, respectively) the proportion of α-helix decreased significantly (P<0.01). No apparent orderliness was found in proportions of β-pleated sheets and β bend after the induction with Dox (Tab 3).

Table 3 The proportion of secondary structure in different types of cell proteins of different groups of cells
Full size table

Viscoelastic coefficients of different groups of Jurkat cells

The elastic coefficient K2 and the viscosity coefficient were not significantly changed among all groups. The elastic coefficient K1 is not significantly different between Jurkat cells (Group 1) and Jurkat3T cells untreated with Dox (Group 2). Jurkat3T induced with different Dox concentrations (0.01, 0.1, 1, 5, 10 μg/ml in Groups 3-7, respectively), however, have a dose-dependent increase in K1 (P<0.01) (Tab 4).

Table 4 Viscoelastic coefficients of different groups of cells (x̄ ± SD).
Full size table

DISCUSSION

TRAIL is known to have a number of biological actions,including the induction of apoptosis. However, the influence of TRAIL on biophysical properties of cells such as Jurkat is still unknown.

Treatment with Dox caused significant alterations in the membrane microstructure of the Jurkat3T cells. These changes include the decrease in the proportion of α-helix in cell protein, and increase in membrane microviscosity. Thus, the induction of TRAIL by Dox, binding with the ligand-receptors15, 16 on the membrane of Jurkat, caused these biophysical changes in Jurkat3T in addition to its capability to increase apoptosis. The important role of TRAIL in causing these biophysical changes is further demonstrated by the dose dependency of these changes in relation to the Dox concentration.

The decrease in cell electrophoretic mobility following TRAIL expression indicates a decrease in the negative surface potential on the cell surface, which was caused by acetylneuraminate of glycoprotein17. The negative charge of cells mainly came from the carboxyl groups of sialic acids on the cell surface18. When TRAIL was expressed, the ensuing ligand-receptor binding may affect the conformation and function of the molecules on membrane. For example, the binding may lead to partial removal of sialic acids from the membrane and thus the surface charge was reduced18. This is manifested not only in the electrophoretic mobility, but also in the changes in the secondary structure proportions in membrane proteins. These alterations in the structure and function of the cell may signal the cell to undergoing apoptosis.

When the TRAIL gene is expressed and the TRAIL protein binds to its receptors on the membrane, it would influence the fluidity of membrane lipid to cause a decline of membrane fluidity15. The membrane fluidity is inversely related to the microviscosity η. The Jurkat cell membrane microviscosity increased, i.e., the fluidity decreased, with the TRAIL expression level. The membrane fluidity is mainly determined by the membrane lipid composition (cholesterol, the degree of saturation and chain length of the fatty acids, and the proportion of lecithin and sphingomyelin), as well as membrane protein-lipid interactions. One of the plasma membrane alterations in the early stages of apoptosis is the translocation of phosphatidylserine (PS) from the inner side of the plasma membrane to the outer layer, by which PS becomes exposed at the external surface of the cell20, 21. Moreover, some biological molecules, such as molecules in cytoskeleton protein were disintegrated during the apoptosis. These changes can also influence the fluidity of the membrane lipid in Jurkat22.

α-helix structures in cell protein play a significant role in transducing the extracellular signals into the cell, and the secondary structure in proteins of the cell may change during this process. The decrease in α-helix proportion following TRAIL expression may reflect the effect of its binding to the related receptor on the cell. When the apoptosis signal transmitted into Jurkat after TRAIL expression, it would lead to the changes of α-helix structure which may turn into β-pleated sheets or β-bend in the cell proteins and accelerate the apoptosis of cancer cells.

The increase in the elastic coefficient K1 after TRAIL expression indicates that the cells' maximum deformation became smaller, i.e., the cells became more rigid.

The membrane biomechanics is closely related to the microstructure. The membrane elastic coefficient of Jurkat increased with the TRAIL expression level. The membrane elastic coefficient is mainly determined by the composition of cytoskeleton protein and the membrane lipid, as well as the interactions between the molecules of protein and lipid in the membrane. When the TRAIL gene was expressed and binded to its receptors on the membrane, it would change the structure of membrane molecules to cause a decline of membrane biomechanics properties17, 23.

On all accounts, from above experiments results, we can see that the changes of biophysical properties (including the increase of the membrane microviscosity η, the membrane elastic coefficient K1 as well as the decrease in cell electrophoretic mobility) of Jurkat may come from the variations of acetylneuraminate composition in the glycoprotein and the protein's proportions after the TRAIL expression, but it should be manifested by further experiments.

Our new findings can be summarized as follows. First, we constructed the RevTet-On TRAIL gene expression system on human leukemic cell line Jurkat. The TRAIL protein was expressed in the presence of the tetracycline derivative doxycycline, and, importantly, this expression was able to induce apoptosis. Second, we have elucidated the changes of the membrane microstructure of the Jurkat cell following the TRAIL expression. These changes may be related to the apoptosis induced by TRAIL and also the inhibiting effect of the TRAIL on tumor cells. The present study has laid the foundation for the potential application of TRAIL in the gene therapy of cancers. Third, we combined the biophysical with molecular biology to study the influence of the TRAIL gene on the microstructure of the tumor cells as well as the changes of the biophysics properties of tumor cells before and after apoptosis by regulation of gene expression. We have investigated the biophysical effects of p53 and TFAR19 (TF-1 cells apoptosis-related gene 19) gene on murine erythroleukemia cell line (MEL)24, 25, and the current work will provide further evidence for the relationship between the biophysics and the tumor gene therapy.