- Email: [email protected]
Increased Susceptibility to Apoptosis of Human Hepatocarcinoma Cells Transfected with Antisense N-Acetylglucosaminyltransferase V cDNA
Biochemical and Biophysical Research Communications 264, 509 –517 (1999) Article ID bbrc.1999.1303, available online at http://www.idealibrary.com on
Increased Susceptibility to Apoptosis of Human Hepatocarcinoma Cells Transfected with Antisense N-Acetylglucosaminyltransferase V cDNA Hua-Bei Guo, Fei Liu, and Hui-Li Chen Key Laboratory of Glycoconjugate Research, Ministry of Health, Department of Biochemistry, Shanghai Medical University, Shanghai, 200032, China
Received August 2, 1999
The antisense cDNA of N-acetylglucosaminyltransferase V (GnT-V, EC 2.4.1.155) was constructed as pcDNA3/GnT-V-AS plasmid and transfected into 7721 cells, a human hepatocarcinoma cell line. The transfection was confirmed with Northern blot. By using HPLC and HRP-lectin staining, it was found that the cells transfected with pcDNA3/GnT-V-AS (GnT-V-AS/7721) expressed less GnT-V activity and b-1,6-GlcNAc branching in the cell glycoproteins compared with the cells mock-transfected with the vector pcDNA3 (pcDNA3/7721). The growth rate of GnT-V-AS/7721 was decreased in serum-containing medium, while the cell death was accelerated in serum-free medium. The GnT-V-AS/7721 cells were more susceptible to the apoptosis induced by ATRA than the mock-transfected cells. This was evidenced by the obvious appearance of a hypoploid sub-G 1 fraction in the DNA histogram using FCM analysis, the more condensed new moon-type nuclei under morphological observation, and the more intensive TUNEL reaction for assaying the fragmented DNA. At the same time as GnT-V down-regulation by GnTV-AS, an increase of another N-aceylglusaminyltransferase, GnT-III (EC 2.4.1.144), was observed, and the biological significance of this finding was discussed. © 1999 Academic Press Key Words: human hepatocarcinoma cell line; apoptosis; N-acetylglucosaminyltransferase V (GnT-V); antisense; all-trans-retinoic acid.
Abbreviations used: GnT-V, N-acetylglucosaminyltransferase V; GlcNAc, N-acetylglucosamine; Man, mannose; ATRA, all-transretinoic acid; FCM, flow cytometry; HPLC, high performance liquid chromatography; L-PHA, leucoagglutinating phytohemagglutinin; ConA, concanavalin A; HRP, horseradish peroxidase; FCS, fetal calf serum; BSA, bovine serum albumin; TUNEL, terminal deoxynucleotidyltransferase-mediated dUTP nick end labeling; N-glycan, asparagine linked glycans; HPF, high power field.
UDP-N-acetylglucosamine:a-D-mannoside b-1,6-Nacetyglucosaminyltransferase V or N-acetylglucosaminyltransferase V (GnT-V, EC 2.4.1.155) is a key enzyme in the processing of asparagine-linked glycans (N-glycans) during the synthesis of glycoproteins. It is located in the Golgi apparatus and catalyses the transfer of a GlcNAc residue from UDP-GlcNAc to the a-1,6 mannoside arm in the acceptor glycans to form a b-1,6 branched structure (GlcNAcb1,6Mana1,6) in the products tri- or tetra-antennary N-glycans (1). GnT-V is well known as a tumorigenesis and metastasis-associated enzyme, since its product, GlcNAcb1,6Mana1,6-structure, is the preferred substrate for the synthesis of polylactosamine (repeating Galb1,4GlcNAcb1,3) by b1,4 galactosyltransferse and a subtype of b1,3 GlcNAcT, and is closely associated to the metastatic potential of many cancers (2– 4). In our laboratory, it was found that GnT-V activity was elevated in human liver cancer and pancreatic cancer, but unchanged in renal carcinoma (5–7). The effects of the transfection of GnT-V expression vector into Mv1Lu cells, an immortalized lung epithelial cell line, were reported by Demetriou et al. (8). The transfectant exhibited characteristics of cellular transformation, including loss of contact-inhibition, decreased adhesion to some adhesion molecules, such as fibronectin and collagen type IV, and increased ability of migration and tumorigenecity in nude mice, as well as accelerated cell death by apoptosis. These changes were suggested to be related to the larger apparent molecular weight of the adhesion molecules, owing to their increased branching of the N-glycans synthesized by the expressed GnT-V. However, the effect of the transfection of the antisense cDNA of GnT-V, whether it is opposite to the effect GnT-V cDNA transfection, was not known. The present investigation reported that a human hepatocarcinoma cell line, 7721, was also showed acceleration of cell death, and was more susceptible to the induced apoptosis by ATRA after transfection of antisense cDNA of GnT-V.
509
0006-291X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
Vol. 264, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
MATERIALS AND METHODS Materials. The 7721 hepatocarcinoma cell line was obtained from the Institute of Cell Biology, Academic Sinica. RPMI 1640 medium was purchased from Gibco/BRL. pcDNA3FluHuTV plasmid containing fulllength of GnT-V cDNA (2.3 kb) was generously provided by Prof. M. Pierce at Georgia University. The cDNA of 3-phosphoglyceraldehyde dehydrogenase (GAPDH) was a gift from Prof. I. Shimizu at Tokushima University. DNA restriction endonucleases, RNAase, Trypan blue, Hoechst 33 258 and random primer labeling kit were from Promega. Hybond-N1 nylon membrane and [a- 32P]-dATP were from Amersham Corp. UDP-GlcNAc, GlcNAc, 2-(N-morpholino)ethanesulfonic acid (MES), MOPS, 3-[N-morpholino]propanesulfonic acid; ATRA, HRP labeled lectin, neuraminidase (clostridium perfringens) and propidium iodide were purchased from Sigma. In situ cell death detection kit, AP (TUNEL) was from Boehringer Mannheim. The common acceptor substrate for GnT assay, fluorescence-labeled biantennary N-glycan (GlcNAc 2Man 3GlcNAc 2-pyridylamino), was prepared by our laboratory as described in the published paper (9). Other reagents were commercially available in China. Construction of the plasmid of antisense GnT-V. About 0.8 kb fragment near 59 end of GnT-V cDNA was cut out from pcDNA3FluHuTV with EcoRI. This fragment was selected, purified and then subcloned into the eukaryotic expression plasmid pcDNA3 which was also cut with EcoRI. The direction of the inserted fragment (0.8 kb) can be determined by BamHI digestion of the recombinant plasmid (BamHI mapping), because there was a BamHI site near the 39 end of the inserted fragment, and about 0.6 kb (if inserted in normal direction) or 0.2 kb (if inserted in inverse direction) apart from the up-stream another BamHI site located in multiple cloning sites of pcDNA3 vector (Fig. 1). After the plasmid with inversely inserted fragment (0.8 kb) was verified by BamHI digestion, this plasmid was selected, purified and named pcDNA3/GnT-V-AS. Agarose electrophoresis. 1–2 mg DNA was added onto 1% agarose prepared in TAE (pH 7.6) containing 8 mg ethidium bromide, and electrophoresis was carried out in TAE buffer at 60 – 80 V for 1 h. The separated fragments were observed under UV light. Transfection of pcDNA3/GnT-V-AS. The constructed pcDNA3/ GnT-V-AS plasmid (6.4 kb) or vector pcDNA3 (5.4 kb) was transfected to 7721 cells by using electroporation method (250 V/0.4 cm and 1000 mF). Transfected 7721 cells were incubated in RPMI 1640 medium containing G418 (0.8 mg/ml). Neomycin-resistant cells were obtained after 2 to 3 weeks and recloned by serial dilution. The pcDNA3/GnT-V-AS and vector transfected cells were titled as GnTV-AS/7721 and pcDNA3/7721, respectively. Cell culture. Cells were cultured at 37°C, 5% CO 2 in RPMI-1640 medium containing 10% FCS, penicillin and streptomycin as previously described by our laboratory (9). Preparation of GnT-V cDNA probe. The 0.8 kb fragment near the 59 end of GnT-V cDNA was cut out from pcDNA3FluHuTV with EcoRI, then separated and recovered from agarose electrophoresis. After purification and quantification, it was labeled with [a- 32P]dATP using random primer labeling kit from Promega according to the instruction described in the manual. Determination of GnT-V mRNA. The isolation of cell total RNA and the Northern blot of RNA were carried out according to the method described by Sagerstrom (10). Briefly, cell total RNA was extracted with Trizol, followed by formaldehyde denatured electrophoresis, transferred to Hybond-N1 nylon membrane, and prehybridization. Hybridization was performed using [a- 32P]-labeled cDNA of GnT-V as the probe. The cDNA of GAPDH was labeled with the same method and used as an intrinsic standard during the Northern blot. Enzyme preparation and assay of GnT. The preparation of crude GnT enzyme and the assay of GnT were performed according to the
methods described in our previous paper (9). A routine HPLC method in our laboratory for GnT-V assay was adopted, except for the concentration of acceptor substrate was increased to 800 mmol/L. The GnT-V and GnT-III were assayed in separate tubes. Briefly, the reaction mixture (50 ml) contained 100 mM MES (pH 6.25), 800 mM fluorescence-labeled biantennary glycan (GlcNAc 2Man 3GlcNAc 2pyridylamino), 50 mM UDP-GlcNAc, 0.2 M GlcNAc, 1% Triton X-100 and 80 –100 mg enzyme protein. In the tube of GnT-III assay, 10 mM of MnCl 2 was added. After incubation at 37°C for 5 h, the reactions were stopped by heating at 100°C for 3 min and the samples were centrifuged at 5000 rpm in an Eppendorf tube for 15 min. An aliquot (20 ml) of each sample was applied to ODS C 18 column to separate the acceptor substrate and the products of GnT-V by HPLC method. All the samples were assayed in duplicate. The enzyme activities were calculated according to the peak areas of the products and expressed as pmol of GlcNAc transferred/h. mg protein. Cell staining with HRP–lectin complexes. Cells coated on the microscopic plates were treated with PBS for 10 min, and neuraminidase (0.03 unit) for 5 h at 37°C to remove terminal sialic acids residues. After washing with PBS, 0.3% H 2O 2 and 1% BSA were added sequentially and incubated for 30 min at each step. After washing, the plates were incubated with 4 mg/ml HRP-L-PHA or HRP-ConA complex and incubated for 2 h at 37°C, then stained with diaminobenzene (DAB). Finally the plates were treated with gradient alcohol, xylol and neutral gel for microscopic observation. The intensity of the staining was quantified with imagine analysis, and expressed as light index. Cell growth and viability. Cells in log phase were harvested with trypsin/EDTA in PBS, re-suspended in RPMI-1640 medium and seeded into cell culture flasks with 2 3 10 5 cells/per flask. One group of the flasks was cultured in serum-containing medium, and the growth curves were determined by daily counting of the cells in triplicate. The other group of the flasks was cultured in serum-free medium, then the cells were harvested daily with trypsin/EDTA in PBS. An aliquot of the cells was diluted 1:5 with 0.4% trypan blue, and the viable cells were counted in triplicate as described by Demetriou et al. (8). Cell growth in soft agar. Assay of cell growth in soft agar was performed using 24-well culture plate (11). The wells were coated with two layers of agar in different concentrations, the lower layer was coated with of 0.8% agar in 0.9% sodium chloride, while the upper layer was 0.3% soft agar in complete culture medium. 4 3 10 4 cells were added into the upper layers of the wells. Plates were incubated at 37°C in 5% CO 2 for 2–3 weeks. Then the numbers of the colonies that developed in soft agar were counted under microscope in 8 HPFs (3200). The data were expressed by the mean value of cells per HPF in triplicate with two independent experiments. Measurement of cell cycle and sub-G1 fraction. Cells were harvested after treatment of 80 mM ATRA for 24 h (12, 13), and fixed with 75% ethanol at 4°C, then centrifuged at 1500 3 g for 5 min. The cells were re-suspended in 1% Triton X-100 (pH 7.8) for 10 min, centrifuged at 1500 3 g for 5 min again, and digested with 0.1 mg/ml RNAase in PBS at 37°C for 10 min, followed by staining with 20 mg/ml propidium iodide. Finally, the cell samples were analyzed by FCM (Becton-Dickinson, CA). Analysis of nuclear morphology by fluorescence staining. The method reported by Schartz (14) was adopted. After treatment of 80 mM ATRA at 37°C for 24 h, the cells grown on the cover glass were fixed with methanol/glacial acetic acid (3:1) at 4°C for 5 min. After washing with triplicate distilled water, the cells were stained with Hoechest 33 258 at room temperature for 10 min. Then the cover glasses were observed under Nikon phase microscope with fluorescence attachment after fixed with citrate buffer (20 mmol/L citrate, 50 mmol/L monosodium phosphate, 50% glycerol, pH 5.5). The green fluorescence was detected between 500 and 525 nm. The cells exhib-
510
Vol. 264, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Construction of antisense GnT-V plasmid (pcDNA3/GnT-V-AS).
Vol. 264, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Characterization of antisense GnT-V with restriction endonuclease (1) Marker: lDNA/HinddIII 1 EcoRI. (2) Plasmid without digestion with restriction endonuclease. (3) Plasmid digested with EcoRI. (4) Plasmid digested with BamHI. The experimental procedure was described under Materials and Methods.
iting bright green fluorescent with condensed new moon-type or fragmented nuclei were counted as apoptotic cells. Cell death detection in situ by TUNEL method. Apoptosisinduced DNA strand breaks were detected by terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate-biotin nick end labeling (15). In short, cells were cultured in 96-well culture plates in the complete culture medium containing 80 mM ATRA at 37°C for 24 h. After washed with PBS and air dried, cells were fixed with freshly prepared 4% paraformaldehyde in PBS (pH 7.4) at room temperature for 1 h, and incubated in permeabilization solution (0.1% Triton X-100, 0.1% sodium citrate) for 2 min on ice after washing with PBS. Cells were rinsed twice with PBS and dried, followed by addition of 50 ml TUNEL reaction mixture and incubation in a humidified chamber at 37°C for 1 h. After washing, 50 ml Converter-Alkaline phosphatase was added and incubated at 37°C for another 30 min. Then the cells were stained with NTB/BCIP staining kit. Label solution (without terminal transferase) was used to the control sample instead of TUNEL reaction mixture. Finally the intensity of the staining was quantified with imagine analysis, and expressed as light index.
FIG. 3. Northern blot of antisense GnT-V mRNA in 7721 cells transfected with antisense GnT-V cDNA. (1) Parental cells. (2) pcDNA3/7721 mock-transfected cells. (3) 7721 cells transfected with pcDNA3/GnT-V-AS. The experimental procedure was described under Materials and Methods.
Northern blot for identification of the pcDNA3/GnTV-AS transfection. After the expression plasmid of pcDNA3/GnT-V-AS or the vector pcDNA3 were transfected into 7721 hepatocarcinoma cells. The transfectants were selected by G418 and verified by Northern blot. The pcDNA3/GnT-V-AS transfectants (GnT-VAS/7721) were demonstrated to express about 0.8 kb mRNA corresponding to the length of externally transferred cDNA (Fig. 3, lane 3), which was undetectable in the parental and vector transfectants (pcDNA3/7721) (Fig. 3, lanes 1 and 2). Changes in GnT activities in GnT-V-AS/7721 cells. As shown in Table 1, the specific activity of GnT-V in the mock-transfected pcDNA3/7721 cells was equal to TABLE 1
Activities of GnT-V, GnT-III in 7721 Cells Transfected with Antisense GnT-V
Protein determination. Lowry method (16) was used.
Cell line
Activity (pmol/h/mg)
Relative activity (%)
268.9 6 12.4 265.0 6 9.6 117.5 6 8.5*
101.5 100.0 44.3*
60.3 6 5.8 62.5 6 6.8 106.8 6 13.8
96.5 100.0 170.9*
RESULTS Identification of pcDNA3/GnT-V-AS plasmid. A fragment of about 0.8 kb near the 59 end of GnT-V cDNA was inversely inserted into the polyclonal EcoRI site of pcDNA3 expression vector. Two fragments of 0.8 kb and 5.4 kb (vector) were appeared on the agarose electrophoresis when the plasmid was digested with EcoRI (Fig. 2, lane 3). The inverse direction of the inserted fragment was evidenced by the BamHI digestion of the plasmid, since a fragment of about 0.2 kb was observed on the agarose electrophoresis in addition to 6.0 kb fragment (Fig. 2, lane 4).
GnT-V 7721 pcDNA3/7721 GnT-V-AS/7721 GnT-III 7721 pcDNA3/7721 GnT-V-AS/7721
Note. Data were expressed as mean value 6 SD of three independent experiments. pcDNA3/7721, 7721 cells transfected with pcDNA3 (mock-transfected cells). GnT-V-AS/7721, 7721 cells transfected with pcDNA3/GnT-V-AS. * p , 0.05, compared with mocked group.
512
Vol. 264, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. Decrease of GnT-V product (GlcNAcb1,6Mana1,6) in N-glycans on 7721 cell surface after GnT-V-AS transfection. (A) Cells on the glass plate were treated with HRP-L-PHA or HRP-ConA, and the color was developed with DAB. (B) The staining colors were analyzed with Imagine analyzer, and expressed as light index (n 5 5). *p , 0.01 compared with the light index of the mock-transfected cells. White column: Mocked, 7721 cells mock-transfected with pcDNA3 vector. Black column: GnT-V-AS transfected, 7721 cells transfected with pcDNA3/ antisense GnT-V.
that of parental cells, while that in the GnT-V-AS/7721 cells was decreased about 56% compared with the mock-transfected cells. It was interested to show that the activity of GnT-III, another N-acetylglucosaminyltransferase, was increased to 171% of the mocktransfected cells after GnT-V-AS transfection. Decreased b1,6 branching on glycans in GnT-V-AS/ 7721 cells. HRP-lectin staining was performed to analyze the alterations of the glycan structure in the transfectants. The transfectants were treated with HRP-L-PHA or HRP-ConA, followed by staining and imagine analysis of the color intensity. It was found that the light index was much lower in GnT-V-AS/7721 cells with HRP-L-PHA staining (p , 0.01), while it was higher when HRP-ConA was used (p , 0.01) compared with the pcDNA3/7721 mock-transfected cells (Fig. 4). Growth and viability of GnT-V-AS/7721 cells. GnTV-AS/7721 cells showed decreased cell growth in serum-containing complete cultured medium (Fig. 5A); but accelerated cell death in serum-free medium when compared with mock-transfected pcDNA3/7721 cells (Fig. 5B). Cell growth on soft agar was used to evaluate the effect of GnT-V-AS expression on cell proliferation and malignancy. The colony-forming ability of GnT-V-AS/ 7721 cells was slightly reduced as compared with pcDNA3/7721 mock-transfected cells (Fig. 6). FCM analysis of induced apoptotic GnT-V-AS/7721 cells. After the cells were treated with 80 mM ATRA, a remarkable fraction of “sub-G 1” cells appeared in the
DNA histogram of the GnT-V-AS/7721 cells when analyzed with FCM, whereas the G 2/M peak was lower in the histogram of GnT-V-AS/7721 cells than in the pcDNA3/7721 mock-transfected cells. The sub-G 1 fraction is an index of the presence of hypoploid cell population, confirming that a significant increase of the apoptotic cells in the GnT-V-AS/7721 group than in the pcDNA3/7721 mock-transfected group (Fig. 7). Observation of nuclear morphology of the induced apoptotic cells. The nuclear morphology of the transfectants after 24 h induction of ATRA were observed after fluorescent staining with Hoechest 33 258. The bright green fluorescent condensed nuclei, which were shaped like a new moon, and the nuclear fragments, were very apparent in the GnT-V-AS/7721 cells, while there were very few in the pcDNA3/7721 mocktransfected cell (Fig. 8). TUNEL analysis of induced apoptotic cells. Induced apoptosis of the cells after treatment with 80 mM ATRA for 24 h was evaluated by determining the broken strand of DNA with TUNEL method. The TUNELpositive cells, which showed a prominent dark staining of the nuclei, were stained intensively in GnT-V-AS/ 7721 cells, but slightly in pcDNA3/7731 mocktransfected cells (Fig. 9). DISCUSSION In this study, the success of the transfection of antisense GnT-V cDNA (pcDNA3/GnT-V-AS) into 7721
513
Vol. 264, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 5. Growth and viability of 7721 cells after transfection with GnT-V-AS in different medium. (A) Growth curve of cells cultured in serum-containing medium. (B) Viability of cells cultured in serumfree medium. Open circle: pcDNA/7721 mock-transfected cells. Closed circle: pcDNA3/GnT-V-AS transfected (GnT-V-AS/7721) cells. The experimental procedure was described under Materials and Methods. Each point represented the mean values 6 SD of 5 HPFs in each of 3 independent experiments.
cells was verified by the decrease of GnT-V activity and b-1.6 GlcNAc branching as shown in Table 1 and Fig. 4. The decrease of the latter was assessed by lectin blotting, since the lectin L-PHA specifically bound to glycans containing b-1.6 GlcNAc branching, while Con A bound to biantannary glycans without b-1.6 and b-1,4 GlcNAc branching (17). Our findings revealed that the proliferation ability and malignancy of human hepatocarcinoma cell line was declined after transfection with antisense GnT-V cDNA, as evidenced by the lower growth rate in serumcontaining culture medium and less colony formation in soft agar when compared with the mock-transfected cells. In contrast, the GnT-V-AS transfected 7721 cells were more labile and susceptible to apoptosis induced by 80 mM ATRA. This conclusion was consistent with the lower viability of the cells in serum-free medium and more positive appearance in the tests of apoptosis, including the sub-G 1 fraction of hypoploid cell population in FCM histogram, the new moon-typed nuclei and nuclear fragments, as well as the TUNEL test. GnT-V is well known as an enzyme correlated to the proliferation and tumorigenesis of many cells. For example, GnT-V activity was higher in G 2/M phase of 7721 cell cycle (to be published), and elevated gradually during
rat hepatocarcinogenesis chemically induced by N-nitrosodiethylamine (DEN) (18). When the expression of GnT-V in the cells was blocked by its anti-sense cDNA, it was reasonable to find that the proliferation and malignant behaviors of the cells were decreased. Apoptosis is an opposite biological phenomenon to cell growth and malignant transformation. Therefore, it is also logical to consider that the transfection of anti-sense GnT-V will induce the cells to be more sensitive to the stimulating signals of cell death or apoptosis, such as the absence of nutrition in serum-free medium or the treatment of ATRA. Our laboratory has reported that ATRA is a potent differentiation-inducer of 7721 cells at a concentration of 10 mM (19). It decreased the expression of GnT-V activity in the cells and b-1,6 GlcNAc branching on the surface glycans (20), as well as increasing the expression of c-myc (21), which was known as an apoptosis-associated gene (22). Recently, our department also found that 80 mM ATRA could induce the apoptosis of 7721 cells after 36 – 60 h treatment in a time-dependent manner in the absence of coated fibronectin, but proportion of the apoptotic cells were not significantly increased after 24 h treatment of ATRA. Transfection of GnT-V-AS increased the sensitivity of 7721 cells to the apoptosis-induced effect of ATRA. As indicated in Fig. 7 and Fig. 8, the sub-G 1 peak in the DNA histogram and the abnormal nuclear morphological appearance were not obvious in the mocked cells, but very apparent in the GnT-V-AS transfected cells after treated with 80 mM ATRA for 24 h. The TUNEL test also showed that the GnT-V-AS/ 7721 cells were more susceptible to ATRA than the mock-transfected cells after the same condition of ATRA treatment. We have not found the DNA ladder in the apoptotic cells, as it were reported by some authors that DNA ladder was not always observed in apoptosis (23, 24). However, transfection of GnT-V-AS per se cannot induce a remarkable apoptosis of the cells, because cell apoptosis is a very complex event, which requires alterations in the expression of many genes other than that of GnT-V.
FIG. 6. Growth of 7721 cells on soft agar after transfection with GnT-V-AS. The experimental procedure was described under Materials and Methods. Data represented the mean values 6 SD of 5 HPFs in each of 3 independent experiments.
514
Vol. 264, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 7. Flow cytometry analysis of ATRA induced apoptosis of 7721 cells after transfection with GnT-V-AS. pcDNA3/7721: Cells mock-transfected with pcDNA3 vector. GnT-V-AS/7721: Cells transfected with pcDNA3/GnT-V-AS.
On the other hand, GnT-V and its product, b-1,6 GlcNAc branching on the cell surface glycans were positively correlated with the metastasis-potential of some malignancies (2– 4). We found that the expressions of some phenotypes related to cell metastasis were also changed by the transfection of GnT-V-AS, including the increased adhesion to fibronectin, reduced cell migration and cell invasion through
matrigel (to be published). These findings were just opposite to the effects of GnT-V cDNA transfection reported by Dennis’ group (8). It were also consistent with the result shown in Fig. 6, that the malignancy of 7721 cells were reduced after GnT-V-AS transfection, since the anchorage independent growth on soft agar to form colonies is a measure of cell malignancy.
515
Vol. 264, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 8. Fluorescent microscope observation of ATRA induced apoptosis of 7721 cells after transfection with GnT-V-AS. pcDNA3/7721: Cells mock-transfected with pcDNA3 vector. GnT-V-AS/7721: Cells transfected with pcDNA3/GnT-V-AS.
The mechanism of GnT-V-AS inducing the higher sensitivity to cell death and apoptosis is not resolved, but it seems to be different from the mechanism of the apoptosis induced in the GnT-V over-expressed cells. It is possible that the decrease of GnT-V and b-1,6 branching on N-glycans will change the sorting of some glycoproteins, such as growth factor receptors and adhesion molecules to the membrane, or reduce the affinity of their ligands, resulting in the attenuation of
signal transduction for cell survival or augmentation of the signaling for cell death and apoptosis. We were very interested to discover that the activity of another N-acetylglucosaminyltransferase-III (GnT-III), which catalyzed the synthesis of the bisecting b-1,4GlcNAc attached to the b-mannose of N-glycan, was elevated in the GnT-V-AS/7721 cells (Table 1). The increased GnTIII can also modify the structure of N-glycans on some glycoproteins, such as receptor of epidermal growth
FIG. 9. TUNEL analysis of ATRA induced apoptosis of 7721 cells after transfection with GnT-V-AS. (A) Cells were stained with NTB/BCIP staining kit as described under Materials and Methods. (B) The intensity of the staining was quantified with imagine analysis, and expressed as light index (n 5 6). *p , 0.01 compared with the light index of the mock-transfected pcDNA3/7721 cells. pcDNA3/7721: Cells mock-transfected with pcDNA3 vector. GnT-V-AS/7721: Cells transfected with pcDNA3/GnT-V-AS. 516
Vol. 264, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
factor (EGF) (25) and nerve growth factor (NGF) (26). These changes will lead to the inhibition of growth factor function or signaling dysfunction, including decreased receptor binding, dimerization and tyrosine autophosphorylation, by the over-expression of bisecting GlcNAc on the receptor glycans. We found that the expression of some growth factor receptors on the surface of 7721 cells was reduced by the treatment of ATRA (27, 28), which may be one of the mechanisms for inducing the cell apoptosis. Moreover, GnT-V-AS transfection and ATRA treatment showed many similar effects, both of them reduced the expression of GnT-V and b-1,6 branching of N-glycan (20), as well as the growth rate, migration and invasion ability of 7721 cells (unpublished data). Therefore, the previous reduction of the number or affinity of growth factor receptors on cell surface by GnT-V-AS transfection may increase the sensitivity of the cells to the apoptotic activity of ATRA. ACKNOWLEDGMENTS This work is supported by a grant from National Natural Science Foundation of China (No. 39630080). We thank Dr. Rudd of the Glycobiology Institute, Department of Biochemistry, University of Oxford, for her constructive suggestions on the manuscript.
REFERENCES 1. Brockhausen, I., Carver, J. P., and Schachter, H. (1989) Biochem. Cell Biol. 66, 1134 –1151. 2. Dennis, J. W., Lafertes, S., Waghorne, C., Breitman, M. L., and Kerbel, R. S. (1987) Science 236, 582–588. 3. Fernandes, B., Sagman, U., Auger, M., Demetriou, M., and Dennis, J. W. (1991) Cancer Res. 51, 718 –723. 4. Pierce, M., Buckhaults, P., Chen, L., and Fregien, N. (1997) Glycoconj. J. 14, 623– 630. 5. Yao, M., Zhou, D. P., Jiang, S. M., Wang, Q. H., Zhou, X. D., Tang, Z. Y., and Gu, J. X. (1998) J. Cancer Clin. Oncol. 124, 27–30. 6. Nan, B. C., Shao, D. M., Chen, H. L., Huang, Y., Gu, J. X., Zhang, Y. B., and Wu, Z. G. (1999) Glycoconj. J. 15, 1033–1037. 7. Zhu, T. Y., Chen, H. L., and Gu, J. X. (1997) J. Cancer Res. Clin. Oncol. 123, 296 –299.
8. Demetriou, M., Nabi, I. R., Coppolino, M., Dedhar, S., and Dennis, J. W. (1995) J. Cell Biol. 30, 383–392. 9. Ju, T. Z., Chen, H. L., Gu, J. X., and Qin, H. (1995) Glycoconj. J. 12, 767–772. 10. Sagerstrom, C. G., and Sive, H. L. (1996) in A Laboratory Guide to RNA: Isolation, Analysis, and Synthesis. (Krieg, P. A., Ed.), pp. 83–103, Wiley-Liss Inc. London, New York, Tokyo. 11. Casalini, P., Menard, S., Malandrin, S. M., Rigo, C. M., Colnaghi, M. I., Cultraro, C. M., and Segal, S. (1997) Int. J. Cancer 72, 631– 636. 12. Kim, D. G., Jo, B. H., You, K. R., and Ahn, D. S. (1996) Cancer Lett. 107, 149 –159. 13. Ponzoni, M., Bocca, P., and Chiesa, V. (1995) Cancer Res. 55, 853– 861. 14. Schartz, G. K., Redwood, S. M., Ohnuma, T., Holland, J. F., Droller, M. J., and Liu, B. C. (1995) J. Natl. Cancer Inst. 87, 1394 –1399. 15. Gavrieli, Y., Sherman, Y., and Ben-Sasson, S. A. (1992) J. Cell Chem. 119, 493–501. 16. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265–271. 17. Osawa, T., and Tsugi, T. (1987) Ann. Res. Biochem. 56, 21– 42. 18. Gu, J. X., Meng, X. W., Ren, J. X., Tang, H., and Chen, H. L. (1993) Chin. J. Biochem. 9, 653– 658. 19. Ai, Z. W., Zha, X. L., and Chen, H. L. (1990) J. Tumor Mark. Oncol. 5, 59 –70. 20. Chen, H. L., Dong, S. C., and Ju, T. Z. (1995) J. Cancer Res. Clin. Oncol. 121, 397– 401. 21. Chai, X. Y., and Chen, H. L. (1994) Acta Oncol. Sinica 4, 229 – 231. 22. Kauffmann-Zeh, A., Rodriguez-Vicina, P., Ulrich, E., Gilbert, C., Coffer, P., Downward, J., and Evan, G. (1997) Nature 365, 544 – 548. 23. Schulze-Osthoff, K., Walczak, H., Droge, W., and Krammer, P. H. (1994) J. Cell Biol. 127, 15–20. 24. Cohen, G. M., Sun, X. M., Snowden, R. T., Dinsdale, D., and Skillete, D. N. (1992) Biochem. J. 268, 331–334. 25. Rebbaa, A., Yamamoto, H., Saito, T., Meuillet, E., Kim, P., Kersey, D. S., Bremer, E. G., Taniguchi, N., and Moskal, J. R. (1997) J. Biol. Chem. 272, 9275–9279. 26. Ihara, Y., Sakamoto, Y., Mihara, M., Shimizu, K., Taniguchi, N. (1997) J. Biol. Chem. 272, 9629 –9634. 27. Chai, X. Y., Chen, H. L., Zhou, X. M., Qian, L. F., Chen, S. H., Jiang, H. Q., and Gu, J. R. (1994) Chin. J. Cancer Res. 6, 3– 8. 28. Qi, W. W., Cui, W., Feng, Y. M., Chen, M. M. (1995) Acta Biochim. Biophys. Sinica 27, 355–360.
517
Increased Susceptibility to Apoptosis of Human Hepatocarcinoma Cells Transfected with Antisense N-Acetylglucosaminyltransferase V cDNA Hua-Bei Guo, Fei Liu, and Hui-Li Chen Key Laboratory of Glycoconjugate Research, Ministry of Health, Department of Biochemistry, Shanghai Medical University, Shanghai, 200032, China
Received August 2, 1999
The antisense cDNA of N-acetylglucosaminyltransferase V (GnT-V, EC 2.4.1.155) was constructed as pcDNA3/GnT-V-AS plasmid and transfected into 7721 cells, a human hepatocarcinoma cell line. The transfection was confirmed with Northern blot. By using HPLC and HRP-lectin staining, it was found that the cells transfected with pcDNA3/GnT-V-AS (GnT-V-AS/7721) expressed less GnT-V activity and b-1,6-GlcNAc branching in the cell glycoproteins compared with the cells mock-transfected with the vector pcDNA3 (pcDNA3/7721). The growth rate of GnT-V-AS/7721 was decreased in serum-containing medium, while the cell death was accelerated in serum-free medium. The GnT-V-AS/7721 cells were more susceptible to the apoptosis induced by ATRA than the mock-transfected cells. This was evidenced by the obvious appearance of a hypoploid sub-G 1 fraction in the DNA histogram using FCM analysis, the more condensed new moon-type nuclei under morphological observation, and the more intensive TUNEL reaction for assaying the fragmented DNA. At the same time as GnT-V down-regulation by GnTV-AS, an increase of another N-aceylglusaminyltransferase, GnT-III (EC 2.4.1.144), was observed, and the biological significance of this finding was discussed. © 1999 Academic Press Key Words: human hepatocarcinoma cell line; apoptosis; N-acetylglucosaminyltransferase V (GnT-V); antisense; all-trans-retinoic acid.
Abbreviations used: GnT-V, N-acetylglucosaminyltransferase V; GlcNAc, N-acetylglucosamine; Man, mannose; ATRA, all-transretinoic acid; FCM, flow cytometry; HPLC, high performance liquid chromatography; L-PHA, leucoagglutinating phytohemagglutinin; ConA, concanavalin A; HRP, horseradish peroxidase; FCS, fetal calf serum; BSA, bovine serum albumin; TUNEL, terminal deoxynucleotidyltransferase-mediated dUTP nick end labeling; N-glycan, asparagine linked glycans; HPF, high power field.
UDP-N-acetylglucosamine:a-D-mannoside b-1,6-Nacetyglucosaminyltransferase V or N-acetylglucosaminyltransferase V (GnT-V, EC 2.4.1.155) is a key enzyme in the processing of asparagine-linked glycans (N-glycans) during the synthesis of glycoproteins. It is located in the Golgi apparatus and catalyses the transfer of a GlcNAc residue from UDP-GlcNAc to the a-1,6 mannoside arm in the acceptor glycans to form a b-1,6 branched structure (GlcNAcb1,6Mana1,6) in the products tri- or tetra-antennary N-glycans (1). GnT-V is well known as a tumorigenesis and metastasis-associated enzyme, since its product, GlcNAcb1,6Mana1,6-structure, is the preferred substrate for the synthesis of polylactosamine (repeating Galb1,4GlcNAcb1,3) by b1,4 galactosyltransferse and a subtype of b1,3 GlcNAcT, and is closely associated to the metastatic potential of many cancers (2– 4). In our laboratory, it was found that GnT-V activity was elevated in human liver cancer and pancreatic cancer, but unchanged in renal carcinoma (5–7). The effects of the transfection of GnT-V expression vector into Mv1Lu cells, an immortalized lung epithelial cell line, were reported by Demetriou et al. (8). The transfectant exhibited characteristics of cellular transformation, including loss of contact-inhibition, decreased adhesion to some adhesion molecules, such as fibronectin and collagen type IV, and increased ability of migration and tumorigenecity in nude mice, as well as accelerated cell death by apoptosis. These changes were suggested to be related to the larger apparent molecular weight of the adhesion molecules, owing to their increased branching of the N-glycans synthesized by the expressed GnT-V. However, the effect of the transfection of the antisense cDNA of GnT-V, whether it is opposite to the effect GnT-V cDNA transfection, was not known. The present investigation reported that a human hepatocarcinoma cell line, 7721, was also showed acceleration of cell death, and was more susceptible to the induced apoptosis by ATRA after transfection of antisense cDNA of GnT-V.
509
0006-291X/99 $30.00 Copyright © 1999 by Academic Press All rights of reproduction in any form reserved.
Vol. 264, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
MATERIALS AND METHODS Materials. The 7721 hepatocarcinoma cell line was obtained from the Institute of Cell Biology, Academic Sinica. RPMI 1640 medium was purchased from Gibco/BRL. pcDNA3FluHuTV plasmid containing fulllength of GnT-V cDNA (2.3 kb) was generously provided by Prof. M. Pierce at Georgia University. The cDNA of 3-phosphoglyceraldehyde dehydrogenase (GAPDH) was a gift from Prof. I. Shimizu at Tokushima University. DNA restriction endonucleases, RNAase, Trypan blue, Hoechst 33 258 and random primer labeling kit were from Promega. Hybond-N1 nylon membrane and [a- 32P]-dATP were from Amersham Corp. UDP-GlcNAc, GlcNAc, 2-(N-morpholino)ethanesulfonic acid (MES), MOPS, 3-[N-morpholino]propanesulfonic acid; ATRA, HRP labeled lectin, neuraminidase (clostridium perfringens) and propidium iodide were purchased from Sigma. In situ cell death detection kit, AP (TUNEL) was from Boehringer Mannheim. The common acceptor substrate for GnT assay, fluorescence-labeled biantennary N-glycan (GlcNAc 2Man 3GlcNAc 2-pyridylamino), was prepared by our laboratory as described in the published paper (9). Other reagents were commercially available in China. Construction of the plasmid of antisense GnT-V. About 0.8 kb fragment near 59 end of GnT-V cDNA was cut out from pcDNA3FluHuTV with EcoRI. This fragment was selected, purified and then subcloned into the eukaryotic expression plasmid pcDNA3 which was also cut with EcoRI. The direction of the inserted fragment (0.8 kb) can be determined by BamHI digestion of the recombinant plasmid (BamHI mapping), because there was a BamHI site near the 39 end of the inserted fragment, and about 0.6 kb (if inserted in normal direction) or 0.2 kb (if inserted in inverse direction) apart from the up-stream another BamHI site located in multiple cloning sites of pcDNA3 vector (Fig. 1). After the plasmid with inversely inserted fragment (0.8 kb) was verified by BamHI digestion, this plasmid was selected, purified and named pcDNA3/GnT-V-AS. Agarose electrophoresis. 1–2 mg DNA was added onto 1% agarose prepared in TAE (pH 7.6) containing 8 mg ethidium bromide, and electrophoresis was carried out in TAE buffer at 60 – 80 V for 1 h. The separated fragments were observed under UV light. Transfection of pcDNA3/GnT-V-AS. The constructed pcDNA3/ GnT-V-AS plasmid (6.4 kb) or vector pcDNA3 (5.4 kb) was transfected to 7721 cells by using electroporation method (250 V/0.4 cm and 1000 mF). Transfected 7721 cells were incubated in RPMI 1640 medium containing G418 (0.8 mg/ml). Neomycin-resistant cells were obtained after 2 to 3 weeks and recloned by serial dilution. The pcDNA3/GnT-V-AS and vector transfected cells were titled as GnTV-AS/7721 and pcDNA3/7721, respectively. Cell culture. Cells were cultured at 37°C, 5% CO 2 in RPMI-1640 medium containing 10% FCS, penicillin and streptomycin as previously described by our laboratory (9). Preparation of GnT-V cDNA probe. The 0.8 kb fragment near the 59 end of GnT-V cDNA was cut out from pcDNA3FluHuTV with EcoRI, then separated and recovered from agarose electrophoresis. After purification and quantification, it was labeled with [a- 32P]dATP using random primer labeling kit from Promega according to the instruction described in the manual. Determination of GnT-V mRNA. The isolation of cell total RNA and the Northern blot of RNA were carried out according to the method described by Sagerstrom (10). Briefly, cell total RNA was extracted with Trizol, followed by formaldehyde denatured electrophoresis, transferred to Hybond-N1 nylon membrane, and prehybridization. Hybridization was performed using [a- 32P]-labeled cDNA of GnT-V as the probe. The cDNA of GAPDH was labeled with the same method and used as an intrinsic standard during the Northern blot. Enzyme preparation and assay of GnT. The preparation of crude GnT enzyme and the assay of GnT were performed according to the
methods described in our previous paper (9). A routine HPLC method in our laboratory for GnT-V assay was adopted, except for the concentration of acceptor substrate was increased to 800 mmol/L. The GnT-V and GnT-III were assayed in separate tubes. Briefly, the reaction mixture (50 ml) contained 100 mM MES (pH 6.25), 800 mM fluorescence-labeled biantennary glycan (GlcNAc 2Man 3GlcNAc 2pyridylamino), 50 mM UDP-GlcNAc, 0.2 M GlcNAc, 1% Triton X-100 and 80 –100 mg enzyme protein. In the tube of GnT-III assay, 10 mM of MnCl 2 was added. After incubation at 37°C for 5 h, the reactions were stopped by heating at 100°C for 3 min and the samples were centrifuged at 5000 rpm in an Eppendorf tube for 15 min. An aliquot (20 ml) of each sample was applied to ODS C 18 column to separate the acceptor substrate and the products of GnT-V by HPLC method. All the samples were assayed in duplicate. The enzyme activities were calculated according to the peak areas of the products and expressed as pmol of GlcNAc transferred/h. mg protein. Cell staining with HRP–lectin complexes. Cells coated on the microscopic plates were treated with PBS for 10 min, and neuraminidase (0.03 unit) for 5 h at 37°C to remove terminal sialic acids residues. After washing with PBS, 0.3% H 2O 2 and 1% BSA were added sequentially and incubated for 30 min at each step. After washing, the plates were incubated with 4 mg/ml HRP-L-PHA or HRP-ConA complex and incubated for 2 h at 37°C, then stained with diaminobenzene (DAB). Finally the plates were treated with gradient alcohol, xylol and neutral gel for microscopic observation. The intensity of the staining was quantified with imagine analysis, and expressed as light index. Cell growth and viability. Cells in log phase were harvested with trypsin/EDTA in PBS, re-suspended in RPMI-1640 medium and seeded into cell culture flasks with 2 3 10 5 cells/per flask. One group of the flasks was cultured in serum-containing medium, and the growth curves were determined by daily counting of the cells in triplicate. The other group of the flasks was cultured in serum-free medium, then the cells were harvested daily with trypsin/EDTA in PBS. An aliquot of the cells was diluted 1:5 with 0.4% trypan blue, and the viable cells were counted in triplicate as described by Demetriou et al. (8). Cell growth in soft agar. Assay of cell growth in soft agar was performed using 24-well culture plate (11). The wells were coated with two layers of agar in different concentrations, the lower layer was coated with of 0.8% agar in 0.9% sodium chloride, while the upper layer was 0.3% soft agar in complete culture medium. 4 3 10 4 cells were added into the upper layers of the wells. Plates were incubated at 37°C in 5% CO 2 for 2–3 weeks. Then the numbers of the colonies that developed in soft agar were counted under microscope in 8 HPFs (3200). The data were expressed by the mean value of cells per HPF in triplicate with two independent experiments. Measurement of cell cycle and sub-G1 fraction. Cells were harvested after treatment of 80 mM ATRA for 24 h (12, 13), and fixed with 75% ethanol at 4°C, then centrifuged at 1500 3 g for 5 min. The cells were re-suspended in 1% Triton X-100 (pH 7.8) for 10 min, centrifuged at 1500 3 g for 5 min again, and digested with 0.1 mg/ml RNAase in PBS at 37°C for 10 min, followed by staining with 20 mg/ml propidium iodide. Finally, the cell samples were analyzed by FCM (Becton-Dickinson, CA). Analysis of nuclear morphology by fluorescence staining. The method reported by Schartz (14) was adopted. After treatment of 80 mM ATRA at 37°C for 24 h, the cells grown on the cover glass were fixed with methanol/glacial acetic acid (3:1) at 4°C for 5 min. After washing with triplicate distilled water, the cells were stained with Hoechest 33 258 at room temperature for 10 min. Then the cover glasses were observed under Nikon phase microscope with fluorescence attachment after fixed with citrate buffer (20 mmol/L citrate, 50 mmol/L monosodium phosphate, 50% glycerol, pH 5.5). The green fluorescence was detected between 500 and 525 nm. The cells exhib-
510
Vol. 264, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 1. Construction of antisense GnT-V plasmid (pcDNA3/GnT-V-AS).
Vol. 264, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 2. Characterization of antisense GnT-V with restriction endonuclease (1) Marker: lDNA/HinddIII 1 EcoRI. (2) Plasmid without digestion with restriction endonuclease. (3) Plasmid digested with EcoRI. (4) Plasmid digested with BamHI. The experimental procedure was described under Materials and Methods.
iting bright green fluorescent with condensed new moon-type or fragmented nuclei were counted as apoptotic cells. Cell death detection in situ by TUNEL method. Apoptosisinduced DNA strand breaks were detected by terminal deoxynucleotidyl transferase-mediated deoxyuridine triphosphate-biotin nick end labeling (15). In short, cells were cultured in 96-well culture plates in the complete culture medium containing 80 mM ATRA at 37°C for 24 h. After washed with PBS and air dried, cells were fixed with freshly prepared 4% paraformaldehyde in PBS (pH 7.4) at room temperature for 1 h, and incubated in permeabilization solution (0.1% Triton X-100, 0.1% sodium citrate) for 2 min on ice after washing with PBS. Cells were rinsed twice with PBS and dried, followed by addition of 50 ml TUNEL reaction mixture and incubation in a humidified chamber at 37°C for 1 h. After washing, 50 ml Converter-Alkaline phosphatase was added and incubated at 37°C for another 30 min. Then the cells were stained with NTB/BCIP staining kit. Label solution (without terminal transferase) was used to the control sample instead of TUNEL reaction mixture. Finally the intensity of the staining was quantified with imagine analysis, and expressed as light index.
FIG. 3. Northern blot of antisense GnT-V mRNA in 7721 cells transfected with antisense GnT-V cDNA. (1) Parental cells. (2) pcDNA3/7721 mock-transfected cells. (3) 7721 cells transfected with pcDNA3/GnT-V-AS. The experimental procedure was described under Materials and Methods.
Northern blot for identification of the pcDNA3/GnTV-AS transfection. After the expression plasmid of pcDNA3/GnT-V-AS or the vector pcDNA3 were transfected into 7721 hepatocarcinoma cells. The transfectants were selected by G418 and verified by Northern blot. The pcDNA3/GnT-V-AS transfectants (GnT-VAS/7721) were demonstrated to express about 0.8 kb mRNA corresponding to the length of externally transferred cDNA (Fig. 3, lane 3), which was undetectable in the parental and vector transfectants (pcDNA3/7721) (Fig. 3, lanes 1 and 2). Changes in GnT activities in GnT-V-AS/7721 cells. As shown in Table 1, the specific activity of GnT-V in the mock-transfected pcDNA3/7721 cells was equal to TABLE 1
Activities of GnT-V, GnT-III in 7721 Cells Transfected with Antisense GnT-V
Protein determination. Lowry method (16) was used.
Cell line
Activity (pmol/h/mg)
Relative activity (%)
268.9 6 12.4 265.0 6 9.6 117.5 6 8.5*
101.5 100.0 44.3*
60.3 6 5.8 62.5 6 6.8 106.8 6 13.8
96.5 100.0 170.9*
RESULTS Identification of pcDNA3/GnT-V-AS plasmid. A fragment of about 0.8 kb near the 59 end of GnT-V cDNA was inversely inserted into the polyclonal EcoRI site of pcDNA3 expression vector. Two fragments of 0.8 kb and 5.4 kb (vector) were appeared on the agarose electrophoresis when the plasmid was digested with EcoRI (Fig. 2, lane 3). The inverse direction of the inserted fragment was evidenced by the BamHI digestion of the plasmid, since a fragment of about 0.2 kb was observed on the agarose electrophoresis in addition to 6.0 kb fragment (Fig. 2, lane 4).
GnT-V 7721 pcDNA3/7721 GnT-V-AS/7721 GnT-III 7721 pcDNA3/7721 GnT-V-AS/7721
Note. Data were expressed as mean value 6 SD of three independent experiments. pcDNA3/7721, 7721 cells transfected with pcDNA3 (mock-transfected cells). GnT-V-AS/7721, 7721 cells transfected with pcDNA3/GnT-V-AS. * p , 0.05, compared with mocked group.
512
Vol. 264, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 4. Decrease of GnT-V product (GlcNAcb1,6Mana1,6) in N-glycans on 7721 cell surface after GnT-V-AS transfection. (A) Cells on the glass plate were treated with HRP-L-PHA or HRP-ConA, and the color was developed with DAB. (B) The staining colors were analyzed with Imagine analyzer, and expressed as light index (n 5 5). *p , 0.01 compared with the light index of the mock-transfected cells. White column: Mocked, 7721 cells mock-transfected with pcDNA3 vector. Black column: GnT-V-AS transfected, 7721 cells transfected with pcDNA3/ antisense GnT-V.
that of parental cells, while that in the GnT-V-AS/7721 cells was decreased about 56% compared with the mock-transfected cells. It was interested to show that the activity of GnT-III, another N-acetylglucosaminyltransferase, was increased to 171% of the mocktransfected cells after GnT-V-AS transfection. Decreased b1,6 branching on glycans in GnT-V-AS/ 7721 cells. HRP-lectin staining was performed to analyze the alterations of the glycan structure in the transfectants. The transfectants were treated with HRP-L-PHA or HRP-ConA, followed by staining and imagine analysis of the color intensity. It was found that the light index was much lower in GnT-V-AS/7721 cells with HRP-L-PHA staining (p , 0.01), while it was higher when HRP-ConA was used (p , 0.01) compared with the pcDNA3/7721 mock-transfected cells (Fig. 4). Growth and viability of GnT-V-AS/7721 cells. GnTV-AS/7721 cells showed decreased cell growth in serum-containing complete cultured medium (Fig. 5A); but accelerated cell death in serum-free medium when compared with mock-transfected pcDNA3/7721 cells (Fig. 5B). Cell growth on soft agar was used to evaluate the effect of GnT-V-AS expression on cell proliferation and malignancy. The colony-forming ability of GnT-V-AS/ 7721 cells was slightly reduced as compared with pcDNA3/7721 mock-transfected cells (Fig. 6). FCM analysis of induced apoptotic GnT-V-AS/7721 cells. After the cells were treated with 80 mM ATRA, a remarkable fraction of “sub-G 1” cells appeared in the
DNA histogram of the GnT-V-AS/7721 cells when analyzed with FCM, whereas the G 2/M peak was lower in the histogram of GnT-V-AS/7721 cells than in the pcDNA3/7721 mock-transfected cells. The sub-G 1 fraction is an index of the presence of hypoploid cell population, confirming that a significant increase of the apoptotic cells in the GnT-V-AS/7721 group than in the pcDNA3/7721 mock-transfected group (Fig. 7). Observation of nuclear morphology of the induced apoptotic cells. The nuclear morphology of the transfectants after 24 h induction of ATRA were observed after fluorescent staining with Hoechest 33 258. The bright green fluorescent condensed nuclei, which were shaped like a new moon, and the nuclear fragments, were very apparent in the GnT-V-AS/7721 cells, while there were very few in the pcDNA3/7721 mocktransfected cell (Fig. 8). TUNEL analysis of induced apoptotic cells. Induced apoptosis of the cells after treatment with 80 mM ATRA for 24 h was evaluated by determining the broken strand of DNA with TUNEL method. The TUNELpositive cells, which showed a prominent dark staining of the nuclei, were stained intensively in GnT-V-AS/ 7721 cells, but slightly in pcDNA3/7731 mocktransfected cells (Fig. 9). DISCUSSION In this study, the success of the transfection of antisense GnT-V cDNA (pcDNA3/GnT-V-AS) into 7721
513
Vol. 264, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 5. Growth and viability of 7721 cells after transfection with GnT-V-AS in different medium. (A) Growth curve of cells cultured in serum-containing medium. (B) Viability of cells cultured in serumfree medium. Open circle: pcDNA/7721 mock-transfected cells. Closed circle: pcDNA3/GnT-V-AS transfected (GnT-V-AS/7721) cells. The experimental procedure was described under Materials and Methods. Each point represented the mean values 6 SD of 5 HPFs in each of 3 independent experiments.
cells was verified by the decrease of GnT-V activity and b-1.6 GlcNAc branching as shown in Table 1 and Fig. 4. The decrease of the latter was assessed by lectin blotting, since the lectin L-PHA specifically bound to glycans containing b-1.6 GlcNAc branching, while Con A bound to biantannary glycans without b-1.6 and b-1,4 GlcNAc branching (17). Our findings revealed that the proliferation ability and malignancy of human hepatocarcinoma cell line was declined after transfection with antisense GnT-V cDNA, as evidenced by the lower growth rate in serumcontaining culture medium and less colony formation in soft agar when compared with the mock-transfected cells. In contrast, the GnT-V-AS transfected 7721 cells were more labile and susceptible to apoptosis induced by 80 mM ATRA. This conclusion was consistent with the lower viability of the cells in serum-free medium and more positive appearance in the tests of apoptosis, including the sub-G 1 fraction of hypoploid cell population in FCM histogram, the new moon-typed nuclei and nuclear fragments, as well as the TUNEL test. GnT-V is well known as an enzyme correlated to the proliferation and tumorigenesis of many cells. For example, GnT-V activity was higher in G 2/M phase of 7721 cell cycle (to be published), and elevated gradually during
rat hepatocarcinogenesis chemically induced by N-nitrosodiethylamine (DEN) (18). When the expression of GnT-V in the cells was blocked by its anti-sense cDNA, it was reasonable to find that the proliferation and malignant behaviors of the cells were decreased. Apoptosis is an opposite biological phenomenon to cell growth and malignant transformation. Therefore, it is also logical to consider that the transfection of anti-sense GnT-V will induce the cells to be more sensitive to the stimulating signals of cell death or apoptosis, such as the absence of nutrition in serum-free medium or the treatment of ATRA. Our laboratory has reported that ATRA is a potent differentiation-inducer of 7721 cells at a concentration of 10 mM (19). It decreased the expression of GnT-V activity in the cells and b-1,6 GlcNAc branching on the surface glycans (20), as well as increasing the expression of c-myc (21), which was known as an apoptosis-associated gene (22). Recently, our department also found that 80 mM ATRA could induce the apoptosis of 7721 cells after 36 – 60 h treatment in a time-dependent manner in the absence of coated fibronectin, but proportion of the apoptotic cells were not significantly increased after 24 h treatment of ATRA. Transfection of GnT-V-AS increased the sensitivity of 7721 cells to the apoptosis-induced effect of ATRA. As indicated in Fig. 7 and Fig. 8, the sub-G 1 peak in the DNA histogram and the abnormal nuclear morphological appearance were not obvious in the mocked cells, but very apparent in the GnT-V-AS transfected cells after treated with 80 mM ATRA for 24 h. The TUNEL test also showed that the GnT-V-AS/ 7721 cells were more susceptible to ATRA than the mock-transfected cells after the same condition of ATRA treatment. We have not found the DNA ladder in the apoptotic cells, as it were reported by some authors that DNA ladder was not always observed in apoptosis (23, 24). However, transfection of GnT-V-AS per se cannot induce a remarkable apoptosis of the cells, because cell apoptosis is a very complex event, which requires alterations in the expression of many genes other than that of GnT-V.
FIG. 6. Growth of 7721 cells on soft agar after transfection with GnT-V-AS. The experimental procedure was described under Materials and Methods. Data represented the mean values 6 SD of 5 HPFs in each of 3 independent experiments.
514
Vol. 264, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 7. Flow cytometry analysis of ATRA induced apoptosis of 7721 cells after transfection with GnT-V-AS. pcDNA3/7721: Cells mock-transfected with pcDNA3 vector. GnT-V-AS/7721: Cells transfected with pcDNA3/GnT-V-AS.
On the other hand, GnT-V and its product, b-1,6 GlcNAc branching on the cell surface glycans were positively correlated with the metastasis-potential of some malignancies (2– 4). We found that the expressions of some phenotypes related to cell metastasis were also changed by the transfection of GnT-V-AS, including the increased adhesion to fibronectin, reduced cell migration and cell invasion through
matrigel (to be published). These findings were just opposite to the effects of GnT-V cDNA transfection reported by Dennis’ group (8). It were also consistent with the result shown in Fig. 6, that the malignancy of 7721 cells were reduced after GnT-V-AS transfection, since the anchorage independent growth on soft agar to form colonies is a measure of cell malignancy.
515
Vol. 264, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
FIG. 8. Fluorescent microscope observation of ATRA induced apoptosis of 7721 cells after transfection with GnT-V-AS. pcDNA3/7721: Cells mock-transfected with pcDNA3 vector. GnT-V-AS/7721: Cells transfected with pcDNA3/GnT-V-AS.
The mechanism of GnT-V-AS inducing the higher sensitivity to cell death and apoptosis is not resolved, but it seems to be different from the mechanism of the apoptosis induced in the GnT-V over-expressed cells. It is possible that the decrease of GnT-V and b-1,6 branching on N-glycans will change the sorting of some glycoproteins, such as growth factor receptors and adhesion molecules to the membrane, or reduce the affinity of their ligands, resulting in the attenuation of
signal transduction for cell survival or augmentation of the signaling for cell death and apoptosis. We were very interested to discover that the activity of another N-acetylglucosaminyltransferase-III (GnT-III), which catalyzed the synthesis of the bisecting b-1,4GlcNAc attached to the b-mannose of N-glycan, was elevated in the GnT-V-AS/7721 cells (Table 1). The increased GnTIII can also modify the structure of N-glycans on some glycoproteins, such as receptor of epidermal growth
FIG. 9. TUNEL analysis of ATRA induced apoptosis of 7721 cells after transfection with GnT-V-AS. (A) Cells were stained with NTB/BCIP staining kit as described under Materials and Methods. (B) The intensity of the staining was quantified with imagine analysis, and expressed as light index (n 5 6). *p , 0.01 compared with the light index of the mock-transfected pcDNA3/7721 cells. pcDNA3/7721: Cells mock-transfected with pcDNA3 vector. GnT-V-AS/7721: Cells transfected with pcDNA3/GnT-V-AS. 516
Vol. 264, No. 2, 1999
BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS
factor (EGF) (25) and nerve growth factor (NGF) (26). These changes will lead to the inhibition of growth factor function or signaling dysfunction, including decreased receptor binding, dimerization and tyrosine autophosphorylation, by the over-expression of bisecting GlcNAc on the receptor glycans. We found that the expression of some growth factor receptors on the surface of 7721 cells was reduced by the treatment of ATRA (27, 28), which may be one of the mechanisms for inducing the cell apoptosis. Moreover, GnT-V-AS transfection and ATRA treatment showed many similar effects, both of them reduced the expression of GnT-V and b-1,6 branching of N-glycan (20), as well as the growth rate, migration and invasion ability of 7721 cells (unpublished data). Therefore, the previous reduction of the number or affinity of growth factor receptors on cell surface by GnT-V-AS transfection may increase the sensitivity of the cells to the apoptotic activity of ATRA. ACKNOWLEDGMENTS This work is supported by a grant from National Natural Science Foundation of China (No. 39630080). We thank Dr. Rudd of the Glycobiology Institute, Department of Biochemistry, University of Oxford, for her constructive suggestions on the manuscript.
REFERENCES 1. Brockhausen, I., Carver, J. P., and Schachter, H. (1989) Biochem. Cell Biol. 66, 1134 –1151. 2. Dennis, J. W., Lafertes, S., Waghorne, C., Breitman, M. L., and Kerbel, R. S. (1987) Science 236, 582–588. 3. Fernandes, B., Sagman, U., Auger, M., Demetriou, M., and Dennis, J. W. (1991) Cancer Res. 51, 718 –723. 4. Pierce, M., Buckhaults, P., Chen, L., and Fregien, N. (1997) Glycoconj. J. 14, 623– 630. 5. Yao, M., Zhou, D. P., Jiang, S. M., Wang, Q. H., Zhou, X. D., Tang, Z. Y., and Gu, J. X. (1998) J. Cancer Clin. Oncol. 124, 27–30. 6. Nan, B. C., Shao, D. M., Chen, H. L., Huang, Y., Gu, J. X., Zhang, Y. B., and Wu, Z. G. (1999) Glycoconj. J. 15, 1033–1037. 7. Zhu, T. Y., Chen, H. L., and Gu, J. X. (1997) J. Cancer Res. Clin. Oncol. 123, 296 –299.
8. Demetriou, M., Nabi, I. R., Coppolino, M., Dedhar, S., and Dennis, J. W. (1995) J. Cell Biol. 30, 383–392. 9. Ju, T. Z., Chen, H. L., Gu, J. X., and Qin, H. (1995) Glycoconj. J. 12, 767–772. 10. Sagerstrom, C. G., and Sive, H. L. (1996) in A Laboratory Guide to RNA: Isolation, Analysis, and Synthesis. (Krieg, P. A., Ed.), pp. 83–103, Wiley-Liss Inc. London, New York, Tokyo. 11. Casalini, P., Menard, S., Malandrin, S. M., Rigo, C. M., Colnaghi, M. I., Cultraro, C. M., and Segal, S. (1997) Int. J. Cancer 72, 631– 636. 12. Kim, D. G., Jo, B. H., You, K. R., and Ahn, D. S. (1996) Cancer Lett. 107, 149 –159. 13. Ponzoni, M., Bocca, P., and Chiesa, V. (1995) Cancer Res. 55, 853– 861. 14. Schartz, G. K., Redwood, S. M., Ohnuma, T., Holland, J. F., Droller, M. J., and Liu, B. C. (1995) J. Natl. Cancer Inst. 87, 1394 –1399. 15. Gavrieli, Y., Sherman, Y., and Ben-Sasson, S. A. (1992) J. Cell Chem. 119, 493–501. 16. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265–271. 17. Osawa, T., and Tsugi, T. (1987) Ann. Res. Biochem. 56, 21– 42. 18. Gu, J. X., Meng, X. W., Ren, J. X., Tang, H., and Chen, H. L. (1993) Chin. J. Biochem. 9, 653– 658. 19. Ai, Z. W., Zha, X. L., and Chen, H. L. (1990) J. Tumor Mark. Oncol. 5, 59 –70. 20. Chen, H. L., Dong, S. C., and Ju, T. Z. (1995) J. Cancer Res. Clin. Oncol. 121, 397– 401. 21. Chai, X. Y., and Chen, H. L. (1994) Acta Oncol. Sinica 4, 229 – 231. 22. Kauffmann-Zeh, A., Rodriguez-Vicina, P., Ulrich, E., Gilbert, C., Coffer, P., Downward, J., and Evan, G. (1997) Nature 365, 544 – 548. 23. Schulze-Osthoff, K., Walczak, H., Droge, W., and Krammer, P. H. (1994) J. Cell Biol. 127, 15–20. 24. Cohen, G. M., Sun, X. M., Snowden, R. T., Dinsdale, D., and Skillete, D. N. (1992) Biochem. J. 268, 331–334. 25. Rebbaa, A., Yamamoto, H., Saito, T., Meuillet, E., Kim, P., Kersey, D. S., Bremer, E. G., Taniguchi, N., and Moskal, J. R. (1997) J. Biol. Chem. 272, 9275–9279. 26. Ihara, Y., Sakamoto, Y., Mihara, M., Shimizu, K., Taniguchi, N. (1997) J. Biol. Chem. 272, 9629 –9634. 27. Chai, X. Y., Chen, H. L., Zhou, X. M., Qian, L. F., Chen, S. H., Jiang, H. Q., and Gu, J. R. (1994) Chin. J. Cancer Res. 6, 3– 8. 28. Qi, W. W., Cui, W., Feng, Y. M., Chen, M. M. (1995) Acta Biochim. Biophys. Sinica 27, 355–360.
517