Paper | Published:

Metabolic instability of plasmid DNA in the cytosol: a potential barrier to gene transfer

Gene Therapy 6, 482–497 (1999) | Download Citation

Subjects

Abstract

Inefficient nuclear delivery of plasmid DNA is thought to be one of the daunting hurdles to gene transfer, utilizing a nonviral delivery system such as polycation–DNA complex. Following its internalization by endocytosis, plasmid DNA has to be released into the cytosol before its nuclear entry can occur. However, the stability of plasmid DNA in the cytoplasm, that may play a determinant role in the transfection efficiency, is not known. The turnover of plasmid DNA, delivered by microinjection into the cytosol, was determined by fluorescence in situ hybridization (FISH) and quantitative single-cell fluorescence video-image analysis. Both single- and double-stranded circular plasmid DNA disappeared with an apparent half-life of 50–90 min from the cytoplasm of HeLa and COS cells, while the amount of co-injected dextran (MW 70000) remained unaltered. We propose that cytosolic nuclease(s) are responsible for the rapid degradation of plasmid DNA, since (1) elimination of plasmid DNA cannot be attributed to cell division or to the activity of apoptotic and lysosomal nucleases; (2) disposal of microinjected plasmid DNA was inhibited in cytosol-depleted cells or following the encapsulation of DNA in phospholipid vesicles; (3) generation and subsequent elimination of free 3′-OH ends could be detected by the terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling assay (TUNEL), reflecting the fragmentation of the injected DNA; and finally (4) isolated cytosol, obtained by selective permeabilization of the plasma membrane, exhibits divalent cation-dependent, thermolabile nuclease activity, determined by Southern blotting and 32P-release from end-labeled DNA. Collectively, these findings suggest that the metabolic instability of plasmid DNA, caused by cytosolic nuclease, may constitute a previously unrecognized impediment for DNA translocation into the nucleus and a possible target to enhance the efficiency of gene delivery.

Introduction

Liposome-mediated cellular transfer of plasmid DNA is a promising approach for gene therapy. However, despite the significant amount of lipid/DNA complexes internalized by the target cells, transgene expression remains undesirably low.1 Obstacles to nuclear accumulation of plasmid DNA include: the slow internalization process of the lipid/DNA complex in certain cells;2 the entrapment of DNA in the endolysosomal compartment;1,3,4 and the diffusional barrier of the nuclear envelope.5

The underlying mechanism of escape of internalized plasmid DNA from the endo-lysosomes is not fully understood. This process involves the destabilization of the limiting membrane of the endolysosomal compartment, the dissociation of the lipid/DNA complex and the release of plasmid DNA into the cytosol.6,7,8 Penetration of naked plasmid DNA into the cytosol was verified by using the T7 polymerase transfection system, which allowed cytosolic transcription of reporter genes controlled by the T7 promoter.9,10 Following the release of DNA into the cytosol, plasmid DNA appears to cross nuclear envelope via the nuclear pore complex (NPC), a notion which is supported by the enhanced transfectability of mitotic cells displaying a disassembled nuclear envelope.1,5,11,12,13 Consistent with the NPC theory of DNA transport, oligopeptides, containing nuclear localization sequence (NLS) were able to facilitate the nuclear uptake of nucleic acid fragments.14 These observations, collectively, define the intracellular itinerary of plasmid DNA from the endosomal compartment via the cytosol and the NPC into the nucleus. The invariable exposure of plasmid DNA to the cytoplasm raised the possibility that the cytosol may impose a presently unknown impediment on the nuclear targeting of DNA.

Considering that the metabolic fate of plasmid DNA in the cytoplasm has not been determined yet, our goal was to measure the stability of plasmid DNA in the cytosol. In order to bypass the plasma membrane and the endocytic compartments, plasmid DNA was delivered by microinjection into the cytosol. Subcellular distribution, turnover and nuclear uptake of DNA were followed by FISH in conjunction with quantitative fluorescence video-image analysis at the single-cell level. Our results suggest, for the first time, that naked but not encapsulated plasmid DNA has a rapid turnover in the cytoplasm of both COS and HeLa cells. In vitro studies, using isolated cytosol and intact or radioactively labeled plasmid DNA, confirmed our hypothesis that plasmid DNA is susceptible to degradation by cytosolic nuclease(s). We propose that this mechanism may represent a previously unrecognized metabolic barrier to DNA delivery into the nucleus and may serve to protect the genetic material of the cell.

Results

Turnover rate of the plasmid DNA in the cytosol

To monitor the intracellular fate of supercoiled, double-stranded pGL2, 1–2 × 103 copies of plasmid DNA were co-injected with tetramethylrhodamine-dextran (TRITC-dextran) into the cytosol of adherent HeLa cells and visualized by FISH and fluorescence microscopy. When the cells were fixed after microinjection, plasmid DNA remained concentrated around the injection site, in contrast to the TRITC-dextran, which spread homogeneously throughout the cytosol, implying that the lateral diffusional mobility of pGL2 is slower than that of TRITC-dextran (Figure 1a, 0 h). Incubation of the cells at 37°C led to a progressive disappearance of the FISH signal (Figure 1). Quantitative fluorescence video-image analysis of FISH signals derived from approximately 2500 injected cells revealed that the DNA was disposed from the cytoplasm with a half-life (t1/2) of approximately 90 min (Figure 2a). In contrast, the concentration of dextran remained at the initial level, implying that only a negligible amount of macromolecules was released from injected cells (Figure 1 and quantification on Figure 3b).

Figure 1
Figure 1

Distribution of plasmid DNA and TRITC-dextran in the cytoplasm of microinjected cells. Double-stranded (a) or single-stranded (0.05 μg/μl) (b) pGL2 plasmid (0.1 μg/μl) was co-injected with TRITC-dextran into the cytosol of HeLa cells and incubated for the indicated time under tissue culture conditions. Cells were processed for FISH and observed by fluorescence light microscopy. The distribution of TRITC-dextran is shown on the right panels. The left panels illustrate the FISH signals of the same cell population as shown on the right, using or DIG-dUTP labeled probes.

Full size image
Figure 2
Figure 2

Metabolic stability of plasmid DNA in the cytoplasm determined by FISH and quantitative single-cell fluorescence video-image analysis. (a) Comparison of the turnover of double-stranded (ds), single-stranded (ss) pGL2 and the prokaryotic expression vector pUC 6–9 (pUC) in HeLa cells. (b) Turnover of ds pGL2 in the cytoplasm of COS-1 and HeLa cells. (c) Turnover rate of biotinylated ds pGL2 in HeLa cells. The data for native pGL2 are replotted from panel a. Cells were microinjected with the indicated plasmid DNA (0.05 or 0.1 μg/μl) and incubated for 0–4 h at 37°C before FISH was performed (a, b). Since no significant difference could be observed in the turnover of pGL2 visualized by DIG-dUTP or bitin-dUTP labeled probe, the statistical analysis includes the results obtained with both probes. Circular ds pGL2 was covalently labeled with photoactivable biotin and detected with the same method as the biotinylated DNA probe in FISH (c). The FISH signals derived from single cells were quantified on images obtained by a cooled CCD camera using the Metafluor software and expressed as percentage of the initial fluorescence intensity of the cells observed immediately after injection (0 time-point). Individual time-points represent the means ± s.e. of at least three independent experiments, each comprising 150–200 injected cells.

Full size image
Figure 3
Figure 3

Depletion of the cytosol obliterates the elimination of plasmid DNA from the cytoplasm. Following the cytosolic injection of the pGL2 plasmid (0.1 μg/ml) into HeLa cells, the plasma membrane was permeabilized with digitonin (40 μg/ml in KH buffer). Permeabilized cells were maintained in ATP-regenerating buffer at 37°C for 1.5 h. A significant amount of TRITC-dextran was lost during the permeabilization procedure (0 h). (a) Fluorescence micrographs of the the TRITC-dextran and the FISH signals of pGL2 plasmid on intact and digitonin permeabilized cells before and after incubation. (b) Quantitative determination of pGL2 and TRITC-dextran content of intact and permeabilized cells. The fluorescence intensities of TRITC and FISH signals were measured in single cells by quantitative video-image analysis using the Metafluor software. Data are expressed as percentage of the initial fluorescence intensities. Values are means ± s.e. of three independent experiments.

Full size image

The rapid disappearance of plasmid DNA from the cytoplasm is independent of the detection method, the cell line or the plasmid DNA used. Similar turnover rates were measured when the half-life of the biotinylated pGL2 (Figure 2c) or fluorescein-labeled DNA (unpub- lished observation) was monitored, implying that the rapid decay of the FISH signal cannot be attributed to the inaccessibility of plasmid DNA by the probe. Furthermore, comparable turnover rate (t1/2 = 80 min) was obtained in COS-1 cells, implying that the metabolic instability of plasmid DNA is not specific to HeLa cells (Figure 2b). Finally, the apparent t1/2 of the prokaryotic expression vector pUC 6–9 was similar to that of pGL2, precluding the possibility that the pGL2 probes were hybridized to pGL2 mRNA (Figure 2a). Circular, single-stranded pGL2 displayed a somewhat shorter apparent t1/2 (50 min) than its double-stranded counterpart (Figures 1b and 2a), and the turnover of linearized pGL2 was indistinguishable from that of the circular form (data not shown). The disappearance rate of the pGL2 plasmid remained unaltered when cells were incubated in serum-free medium, showing that DNases present in serum cannot account for our observations (data not shown).

Multiple cellular mechanisms may be involved in the rapid elimination of DNA from the cytoplasm. Plasmids could be degraded by resident cytosolic nucleases or by organellar DNases after the sequestration of DNA. Alternatively, cytosolic DNA could be released by exocytosis and/or lost during cell division. The experiments described in the following sections were designed to examine each of these possibilities.

The cellular mechanism of the rapid turnover of cytosolic plasmid DNA

Sequestration of plasmid DNA by autophagy or facilitated translocation would promote its degradation by lysosomal DNases, exhibiting an acidic pH optima.5,15,16,17 However, it is highly unlikely that lysosomal nucleases are responsible for the rapid turnover of plasmid DNA, since dissipation of the acidic pH of the endolysosomes with ammonium chloride or chloroquine18,19 did not delay the disappearance of injected DNA (Table 1).

We speculated that disposal of plasmid DNA may proceed via exocytotic delivery of cytoplasmic vesicles that have accumulated DNA by a mechanism which is yet unknown. Since both homo and heterotypic membrane fusions require metabolic energy, the ATP-dependence was tested next. Depletion of the cellular ATP content by more than 95% had no significant effect on the turnover rate of plasmid DNA, indicating that the disposal of cytosolic DNA does not require vesicular transport (Table 1). A similar extent of ATP-depletion completely inhibited glycoprotein processing along the biosynthetic pathway.20

When the ambient temperature was reduced to 4°C, elimination of plasmid DNA was completely prevented (Table 1), implying that the disposal of DNA is temperature-dependent and may be mediated by an enzymatic process. Interestingly, inhibition of protein synthesis by cycloheximide slowed down the disappearance rate of plasmid DNA by 50% (Table 1), consistent with the notion that the disposal of cytosolic DNA may involve proteinaceous factor(s) with short biological half-life.

If a cytosolic nuclease is responsible for plasmid degradation, depletion of the soluble constituents of the cytoplasm would obliterate the disposal of DNA. To this end, depletion of the cytosol was achieved by selective permeabilization of the plasma membrane with digitonin, a well established technique to investigate vesicular transport and nuclear targeting of proteins and DNA in semi-intact cells.14,16,21,22 The nearly complete loss of microinjected TRITC-dextran from permeabilized cells but not from intact ones demonstrates the efficiency of the procedure (Figure 3a and b). In marked contrast to TRITC-dextran, plasmid DNA was retained in the cytoplasm at the initial concentration, as measured by quantitative single-cell fluorescence video imaging (Figure 3b), implying that the cytosol may contain either soluble nuclease(s) and/or cofactors that are indispensable for the hydrolysis of plasmid DNA. Meanwhile, permeabilization of the plasma membrane did not interfere with the light microscopic architecture of the ER, lysosomes or endosomes (data not shown).

If the degradation of DNA coincides with the accumulation of fragmented nucleic acids bearing free 3′-OH ends, these intermediates should be visualized by the terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL) assay. HeLa cells were injected with the pGL2 plasmid cytosolically and processed for the TUNEL assay after 0–6 h of incubation at 37°C. While the fluorescence signal was initially negligible (Figure 4, 0 h), it increased markedly after 1 h of incubation at 37°C (Figure 4, 1 h). In contrast, the signal remained at the initial level if the cells were kept at 4°C (data not shown). Upon extending the incubation period the TUNEL signal gradually disappeared, indicating that the degradation intermediates of plasmid DNA were eliminated from the cytoplasm (Figure 4, 6 h). These results suggest that transient accumulation of fragmented nucleic acids accompanies the cleavage of cytosolic plasmid DNA.

Figure 4
Figure 4

Fragmentation of microinjected plasmid DNA in the cytoplasm visualized by in situ end labeling (TUNEL) assay. pGL2 (1 μg/μl) and TRITC-dextran were delivered into the cytoplasm of HeLa cells by microinjection and cells were incubated for the indicated time at 37°C. Detection of free 3′ OH end was accomplished by the TUNEL assay using DIG-dUTP and indirect immunofluorescence. The TRITC-dextran and the TUNEL signals, derived from microinjected cells, are represented by the right and left panels, respectively.

Full size image

Encapsulation into lipid particles enhances the metabolic stability of plasmid DNA in the cytoplasm

Complex formation between DNA and polycationic macromolecules (eg cationic lipids or poly-L-Lys) is capable of protecting nucleic acids from degradation by endo- and exo-nucleases in vitro.23,24 A similar phenomenon could be envisioned in vivo, following the entry of DNA/polycation complexes into the cytosol. However, the aggregation tendency of DNA/polycation complexes prevented reproducible microinjection. To overcome this difficulty, pGL2 DNA was encapsulated into stable plasmid-lipid particles (SPLP).25 SPLP exhibits a uniform particle-size distribution of 60–90 nm and confers plasmid DNA resistance to cleavage by DNase I in vitro.25 The subcellular distribution and disposal rate of microinjected SPLP incorporating pGL2 was determined by FISH and quantitative single-cell fluorescence video imaging (Figure 5). Not more than 25% of the encapsulated plasmid was degraded after 4 h of incubation, in contrast to 75% of uncomplexed plasmid (Figures 3 and 5b), indicating that the degradation rate of encapsulated pGL2 is three-fold slower than that of the naked pGL2. This observation is fully consistent with the in vitro protective effect of SPLP against nucleolysis by DNase I.26 The incomplete protection of plasmid DNA by SPLP is presumably due to degradation of its lipid components in the cytosol and/or fusion of SPLP to compartments enriched in nuclease (eg lysosome).

Figure 5
Figure 5

Cytoplasmic turnover of encapsulated double-stranded plasmid DNA in SPLP. (a) TRITC-dextran and SPLP containing pGL2 plasmid (0.05 μg/μl) was co-injected into the cytosol of HeLa cells. Cells were processed for FISH after the indicated incubation period. The panels on the right show the distribution of TRITC-dextran fluorescence and the panels on the left are the FISH images of the corresponding cells, obtained with DIG-dUTP labeled probes. (b) Comparison of the degradation of naked and encapsulated pGL2 in the cytoplasm. HeLa cells were injected with naked or encapsulated pGL2 and processed as described on panel (a). Fluorescence intensities associated with single cells are means from approximately 350 injected cells, obtained in three independent experiments and were measured as described in Figure 2.

Full size image

Microinjection does not induce apoptosis and interfere with the integrity of organelles

In principle, nuclease activity associated with the cytosol can either be constitutively present or released from organelles, such as the nucleus, lysosomes, ER and the mitochondria upon microinjection. Several lines of evidence show that morphological and functional integrity of the intracellular organelles were preserved in microinjected cells. Firstly, injected TRITC-dextran (MW 70 kDa) was excluded from the nucleus (Figures 1 and 3a). Secondly, the immunostaining patterns of the ER, lysosomes and endosomes, visualized by the anti-protein disulfide isomerase (PDI), the anti-lysosomal associated membrane protein-2 (LAMP-2)27 antibodies and fluorescein-transferrin, respectively, were unperturbed in injected cells (Figure 6a). Finally, the transient elevation of the cytosolic Ca2+ concentration ([Ca2+]cy), evoked by the mechanical injury of the plasma membrane28,29 was completely normalized after 8–9 min of injection (Figure 6b), indicating that the Ca2+ permeability and transport capacity of the plasma membrane, ER and mitochondria were rapidly restored. These morphological and functional data indicate that the integrity of organelles is preserved and the release of macromolecules such as nucleases, from organelles into the cytosol, is highly unlikely.

Figure 6
Figure 6

Microinjection does not alter the morphology and Ca2+ transport ability of intracellular organelles or induce apoptosis. (a) Light microscopic morphology of the intracellular organelles upon microinjection. HeLa cells were injected into the cytoplasm with TRITC-dextran and fixed after 2 h of incubation. The ER and lysosomes of injected and intact cells were visualized by indirect immunostaining using anti-PDI and anti-LAMP-2 antibodies, respectively (right panels). Early endosomes were labeled with fluorescein-transferrin (Tf, 5 μg/ml, 45 min at 37°C). While the distribution of organelles are depicted on the right panels, the corresponding TRITC-dextran distribution indicates the microinjected cells. (b) Effect of the microinjection on the [Ca2+]cy. HeLa cells were loaded with the Ca2+-indicator Fura-2 and injection was carried out in Ca2+-containing or in Ca2+-free medium. The [Ca2+]cy was determined with fluorescence ratio video-imaging as described in Materials and methods. Cells were injected at arrows and the ratio values are the means of six to eight cells. The [Ca2+]cy of non-injected cells is stable (filled circle). (c) Microinjected cells do not undergo apoptosis. HeLa cells were microinjected with TRITC-dextran and fragmentation of chromosomal DNA was examined after 4 h of incubation with the TUNEL assay using DIG-dUTP and indirect immunofluorescence detection. Upper panel: TRITC-dextran distribution; lower panel: corresponding images of the TUNEL signal; insert: apoptotic nucleus of staurosporine (8 μM 4 h) treated HeLa cells are easily recognized with the TUNEL assay.

Full size image

Multiple signal-transduction pathways can provoke apoptosis with concomitant activation of Ca2+-dependent or Ca2+-independent nucleases such as DNase I, DNase II and the caspase-3-activated deoxyribonuclease (CAD).30,31,32,33,34,35,36,37 Therefore, it was imperative to preclude the possibility that microinjection initiated the enzyme cascade associated with apoptosis which culminates in the in situ activation of CAD and/or the release of other DNases into the cytosol.32 While we were unable to detect any evidence of chromosomal DNA fragmentation in injected cells by TUNEL (Figure 6c, lower panel), a strong signal was recorded from the nucleus of HeLa cells undergoing staurosporine-induced apoptosis (Figure 6c, insert). Furthermore, the instability of the plasmid was not affected when the [Ca2+]cy transient was attenuated by performing the experiment in Ca2+-free medium (Figure 6b and Table 1). Collectively, these data strongly support our hypothesis that plasmid DNA is degraded by nucleases constitutively present in the cytosol, rather than by nucleases either liberated from organelles or activated upon microinjection.

Characterization of the cytosolic DNase activity in vitro

In order to examine the enzymatic characteristics of the nuclease activity associated with the cytosol, DNA degradation was reconstituted in vitro. To obtain cytosol with minimal organellar contaminations, the plasma membrane of HeLa cells was selectively permeabilized with digitonin, and the extracellular medium, containing the soluble constituents of the cytoplasm, was collected. Contamination of the cytosolic extract with lysosomal enzymes was assessed by measuring the β-glucuronidase activity of the extract as well as the release of TRITC-dextran (MW 70000), a fluid-phase tracer of lysosomes, was determined. Both assays showed that no more than 3–4% of the luminal content of lysosomes was released into the extracellular medium.21,22,38

To visualize the nuclease activity, circular double-stranded DNA was incubated with isolated cytosol and cleavage products were examined by Southern blottting. Generation of relaxed and linearized double-stranded plasmid DNA demonstrates the endonucleolytic activity of the cytosol (Figure 7a). The endonuclease activity can be stimulated by divalent cations such as Mn2+, Ca2+ and Mg2+ and inhibited at 4°C or by heat treatment of the cytosol (Figure 7a and data not shown). Consistent with our in vivo studies, single-stranded circular and linearized double-stranded pGL2 were also degraded in vitro (data not shown). Besides the production of nicks and double-stranded breaks, smaller DNA fragments could be detected on the Southern blot (Figure 7a and b). In order to address the question whether this cleavage pattern could be due to the exonucleolytic activity of the cytosol, 32P release from 3′- or 5′-end-labeled, double-stranded plasmid DNA was measured with the TCA-precipitation method (Figure 7c–g). Although the time-dependent release of 32P is consistent with exonuclease activity (Figure 7b and d), it does not preclude the possibility that the assay reflects endonuclease activity of the cytosolic extract as well. The nuclease activity, measured by 32P release, was also stimulated by Mn2+, Mg2+ and Ca2+ and completely circumvented at 4°C or after heat treatment of the cytosol (Figure 7c, d and data not shown). Aurintricarboxylic acid and Zn2+, non-specific inhibitors of DNases32,39,40 obliterated the enzyme activity with a Ki of 2.5 and 0.2 μM, respectively, implying that the affinity of the cytosolic nuclease toward these inhibitors is 20- to 50-fold higher than that of pancreatic DNase I (Figure 7e and f). Finally, the pH dependence of the cytosolic nuclease was determined. Maximum nuclease activity was attained between pH 7 and 8 (Figure 7g), corroborating our previous conclusion that the cytosolic nuclease is distinct from DNase II, which displays an acidic pH optimum.17,32

Figure 7
Figure 7

Nuclease activity of isolated cytosol. (a,b) Southern blot analysis of the nuclease activity of cytosolic extract. Double-stranded circular (a) or linearized (b) pGL2 plasmid (25 ng) was incubated with isolated cytosol (approximately 100 μg protein) in the presence or absence of divalent cations (MnCl2, 1 mM; MgCl2, 1 mM and CaCl2, 1 mM) for the indicated time at 37°C. Following the precipitation of DNA, Southern blotting was performed with biotinylated probe (20 ng/ml) and enhanced chemiluminescence assay using horseradish-peroxidase conjugated streptavidin. r, relaxed; l, linearized; sc, supercoiled conformation of the pGL2 plasmid. For comparison, circular (a) and linearized (b) plasmid DNA (25 ng) was loaded on the agarose gel without precipitation (lane 1). (c–g) Measurements of the exonuclease activity of the cytosol. End-labeled plasmid DNA (2.5 fmol 3′-[32P]-labeled: panels c, e–g; and 10 fmol of 5′-[32P]-labeled DNA: panel d) was incubated with cytosolic extract (15 μg protein) for the indicated time or 10 min (panels e–g) either in HK-medium (panels c–f) or in medium buffered at the indicated pH (panel g). Nucleic acids were precipitated with TCA and the TCA-soluble radioactivity was measured. Panels c and d: incubation was carried out in the absence (open square) or presence (filled circle) of 1.0 mM MnCl2, at 4°C (diamond) or using heat-treated cytosol (10 min at 95°C, triangle). Panels e and f: comparison of the inhibitor susceptibility of the cytosolic nuclease and pancreatic DNase I. The nuclease activity of cytosol (15 μg protein, filled squares) and DNase I (1 U, open circles) was determined in the presence of ZnCl2 or aurintricarboxylic acid (ATA). All data points represent the mean of two to four independent experiments. Total amount of radioactivity liberated from 3′- and 5′-[32P]-labeled DNA was equivalent to 9000 c.p.m. and 900 c.p.m., respectively.

Full size image

Discussion

Introduction of foreign DNA into eukaryotic cells can be mediated by viral infection or by chemical vector. While viral DNA penetrates the cytosol either through the plasma membrane41 or the endosomal membrane,42 plasmid DNA complexed with its chemical vectors is invariantly internalized by endocytosis.1,4 The acidification of the luminal space of endosomes was proposed to play a critical role in the partial uncoating and penetration of viral particles.43 While, the role of acidification in the dissociation of the cationic lipid/DNA complex is not fully understood,7,8 dissipation of the endosomal pH gradient does not seem to impair the transfection efficiency.44 Nevertheless, the apparently inefficient liberation of plasmid DNA from the endolysosomal compartment implies that a significant fraction of internalized plasmid DNA remains entrapped and is degraded in lysosomes.44 This degradation could be partially circumvented either by dissipating the acidic pH of the lysosomes with lysomotropic agents, by obliterating endolysosome fusions by the depolymerization of the microtubule network or by promoting endosomal escape using endosome-disruptive agents.3,18,26,45 These treatments were able to extend the cellular persistence of plasmid DNA and to augment the expression of heterologous plasmid DNA.

Morphological investigations at the light and electron microscopy levels and functional studies using nuclear transport inhibitors provided convincing evidence that both cytosolic plasmid and viral DNA are taken up by the nucleus via the NPC.5,16,43 While the cytoplasmic movement of viral DNA towards the nucleus is facilitated by the microtubular network and viral proteins, such as polymerase or capsid proteins that remain associated with the nucleic acids,46,47 very little is known about the fate of plasmid DNA in the cytoplasm. Our study represents the first attempt to elucidate the metabolic stability of plasmid DNA in the cytoplasm using both morphological and functional approaches at the single cell level.

Comparison of the levels of luciferase expression upon nuclear and cytoplasmic injection of plasmid DNA allowed us to deduce that only approximately 0.1% of the cytosolic plasmid DNA enters the nucleus48 (DL and GL, unpublished observation). Although these results are consistent with the earlier notion that the nuclear envelope constitutes a major impediment against DNA translocation,5,16,18,49 the following observations suggest that the cytosol may account, in part, for the low transfection efficiency. First, plasmid DNA injected into the cytosol was localized at the site of injection, in contrast to the homogeneously distributed high molecular weight dextran (Figure 1). Presumably, both DNA–protein interactions and the limiting diffusional mobility of plasmid DNA (corresponding approximately in size to a 5 MDa protein) may account for this phenomenon, as observed for other macromolecules in the cytoplasm.50,51,52 Second, we have obtained direct evidence for the fast turnover of injected cytosolic DNA, using three different detection methods: FISH, covalently tagged plasmid with biotin, as well as nick-labeled DNA with fluorescein-dUTP (Figure 2 and unpublished observation). The similarly short half-lives (t1/2 1–2 h) of DNA, obtained with three different detection methods, demonstrate that naked plasmid DNA entering the cytoplasm is unstable. Considering that our detection techniques are unable to discriminate between the fluorescence signals derived from intact or fragmented plasmid DNA, the turnover data are likely to provide an overestimation of the half-life of intact plasmid DNA in the cytoplasm. Along these lines we can speculate that DNA-condensing agents, such as polylysine or polyethylenimine, are able to enhance the transfection efficiency by prolonging the half-life of plasmid DNA in the cytosol conferring them DNase resistance, similarly to that previously reported in vitro.24,53 On the other hand, the nuclease activity of the cytosol may provide an explanation for the inhibitory effect of cytoplasmic extract on DNA transport into the nucleus, observed in the permeabilized cell system.16

A panel of pharmacological treatments in conjunction with the measurement of DNA turnover allowed us to infer that neither exocytosis nor organellar sequestration can account for the disposal of cytosolic DNA. The following evidence argues for the pivotal role of cytosolic nuclease in the rapid degradation of plasmid DNA. First, time- and temperature-dependent generation of free 3′-OH ends could be detected by the TUNEL assay as a consequence of the in situ fragmentation of plasmid DNA (Figure 4). These degradation intermediates did not associate with any known subcellular organelles. Second, degradation of plasmid DNA was circumvented by depletion of the cytosol, following selective permeabilization of the plasma membrane with digitonin or substantially delayed by the inhibition of cellular protein synthesis with cycloheximide (Figure 3 and Table 1). Third, cytoplasmic elimination of plasmid DNA was also delayed upon encapsulation of nucleic acids into SPLP (Figure 5). Fourth, using indirect immunofluorescence of lysosomes, ER and endosomes as well as monitoring the [Ca2+]cyof microinjected cells, we could rule out the possibility that injection altered the permeability of these organelles in addition to the mitochondria and nuclear envelope (Figures 1 and 6). This is important, considering that apart from nuclear and mitochondrial DNases implicated in the metabolism of the genomic DNA,54,55 a number of other nucleases (eg DNase I and DNase II) have been associated with the biosynthetic compartments and the endolysosomes.17,30,37 Thus the above experiments strongly suggest that the release of compartmentalized nucleases is highly unlikely, which gives credence to our hypothesis that degradation of DNA is mediated by nuclease(s) present in the cytosol. This conclusion was substantiated by the nuclease activity of cytosolic extract, detected by Southern blot analysis and 32P-release of end-labeled plasmid DNA (Figure 7).

A panel of endo- and exonucleases have been invoked in chromosomal DNA degradation into nucleosomal units during apoptosis:56 (1) DNase I and DNase II, subsequent to their release into the cytosol and translocation into the nucleus;30,31,32,57,58,59 (2) specific apoptotic nucleases (eg cyclophilins);32,60 and (3) the caspase-3 activated DNase (CAD) following its activation and nuclear uptake.36,37 However, no sign of nuclear DNA fragmentation was detected by the TUNEL assay, indicating that none of the DNases invoked in chromosomal DNA degradation became stimulated in injected cells (Figure 6c). The distinct inhibitor and pH-sensitivity profiles of the cytosolic nuclease activity also consistent with the notion that neither DNase I nor DNase II contributes to the cleavage of injected plasmid DNA (Figure 7). Finally, experiments performed in the absence of extracellular Ca2+ revealed that the cytosolic plasmid DNA degradation cannot be attributed to the activation of Ca2+-sensitive DNases.33,34,35 Although the nuclease responsible for plasmid DNA disposal remains to be identified, we cannot preclude the possibility that the basal activity of CAD, expressed constitutively in the cytosol, is responsible for the elimination of plasmid DNA. This hypothesis has to be tested in future experiments.37

Despite the similarities of the intracellular transport route of lipid/DNA complexes and viral particles, the nuclear targeting efficiency of plasmid DNA appears to be substantially lower. We speculate that the highly efficient nuclear targeting of viral DNA stems from the synergistic effects of the following factors. Firstly, the viral genome penetrates the cytoplasm in association with viral proteins such as the viral polymerase or capsid proteins that may provide protection against cytosolic nucleases.46,47 Secondly, a number of viral proteins (eg viral capsid proteins) have been shown to encompass nuclear localization sequence (NLS) that facilitates nuclear uptake and conceivably diminishes the cytosolic residence of viral nucleic acids.46,47 Finally, NLS-independent interactions, such as the herpes virus association with the microtubular network, may promote the nuclear targeting of viral particles by preventing their entrapment in the cytoplasm.61 Strategies incorporating these naturally evolved mechanisms into the construction of chemical vectors may improve their transfection efficiency in the future.

In conclusion, our experiments provide the first measurement of the metabolic stability of cytosolic plasmid DNA at the single cell level. Based on the results obtained by in situ turnover measurements of plasmid DNA and in vitro demonstration of the nuclease activity of the cytosolic extract we propose that upon chemical vector-mediated gene transfer, the expression of heterologous DNA is not only hampered by the entrapment of DNA in the endolysosomal compartment and the physical barrier of the nuclear envelope, but also by the metabolic barrier represented by cytosolic nuclease(s). Identification of the cytosolic nuclease(s) and its specific inhibitors may open new approaches to enhance the efficiency of heterologous DNA delivery in gene transfer trials.

Materials and methods

Cell culture

HeLa (human epitheloid carcinoma, ATCC) cells were grown in α-MEM with 10% (v/v) fetal calf serum (FCS; Gibco BRL, Gaithersburg, MD, USA). COS-1 and COS-7 (SV40 transformed Green african monkey kidney cells, ATCC) were maintained in DMEM with 10% FCS. Suspension culture of HeLa cells was grown in Joklik’s modified alpha-MEM containing 5% calf serum. All cell cultures were incubated in 5% CO2 at 37°C.

Plasmid DNA

The pGL2 expression plasmid encoding the luciferase gene was purchased from Promega (Madison, WI, USA). The pUC (pUC 6–9) plasmid, encoding a chloroplast protein, was a gift of Dr J Hu (Hospital for Sick Children, Toronto). For microinjection, plasmids were purified by CsCl2 gradient centrifugation62 or with the Plasmid Maxi Kit (Qiagen). The two types of plasmid preparation yield identical results.

Circular single-stranded pGL2 was prepared as described.63 Biotinylation of circular double-stranded pGL2 was carried out with photoactivable biotin (Pierce, Rockford, IL, USA), according to the manufacturer’s instructions and its efficiency was verified by dot-blot analysis using streptavidin, conjugated to horseradishperoxidase (Amersham). Biotinylated pGL2 remained circular and susceptible to digestion with Bal31 nuclease and XhoI restriction enzyme as the native plasmid, implying that biotinylation does not alter plasmid DNA conformation significantly (data not shown).

Encapsulation of plasmid DNA into stable plasmid lipid particles

Plasmid DNA was encapsulated as stable plasmid lipid particles (SPLP) using the detergent dialysis method.26 The SPLP particles contained 42.5:42.5:15 mol% of dioleoyldimethylammonium chloride (DODAC), 1,2,- sn-dioleoylphosphatidylethanolamine (DOPE) and 1-O-(2′-(w-methoxypolyethylenglycol) succinoyl)-2-N-octylsphyngosine (PEG ceramide C-8), respectively.26 According to quasielastic light scattering measurements by a Nicomp 270 submicron particle sizer, the diameter of SPLP was 60–90 nm.26 More than 99.9% of the plasmid DNA was entrapped inside the SPLP particles as determined by the accessibility of plasmid DNA by the fluorescent DNA-dye, Picogreen (Molecular Probe, Eugene, OR, USA).

Microinjection

Microinjections were performed by a semi-automatic injection system (Eppendorf Transjector 5246, Hamburg, Germany) attached to the Eppendorf Micromanipulator 5171. Borosilicate thin-walled glass pipettes with filament (ID 0.78 mm, OD 1 mm) were fabricated with a P97 puller (Sutter Instrument, Novato, CA, USA). Plasmid DNA (0.05–1 mg/ml) and TRITC-dextran (2 mg/ml, MW 70 kDa, lysine-fixable, Molecular Probes) were microinjected in KG-medium (110 mM K+-glutamate, 30 mM NaCl, 2 mM MgCl2, pH 7.2) at 200 kPa over 0.1 s. Injections were performed at room temperature in bicarbonate free α-MEM, supplemented with 10 mM Na-HEPES, pH 7.3 and routinely delivered 3000 copies (0.1 mg/ml) of plasmid DNA for turnover determinations. Similar results were obtained when 10000 copies of plasmid DNA were injected into the cytoplasm. Microinjection introduced an average volume of 222 ± 91 fl, (n = 2145) into the cytosol as determined by the delivery of [3H]-dextran (MW 70 kDa, Amersham, Oakville, Ontario, Canada). The overnight survival rate of the cells subjected to cytosolic microinjection was 70–75%. Usually, 100–150 cells were microinjected on each slide over a 15–30 min period.

Fluorescence in situ hybridization (FISH)

Cells, grown on Labtek glass slides (Gibco BRL), were fixed with 3% paraformaldehyde in PBS (137 mM NaCl, 2.7 mM KCl, 4.3 mM Na2HPO4, 1.4 mM KH2PO4, pH 7.3) for 10 min at RT, rinsed with PBS supplemented with 2 mM glycine and permeabilized with 0.2% Tween-20 in PBS (5 min at RT). Slides were processed for hybridization with biotin- or digoxigenin-labeled probe as described.64 Probes were generated by nick translation of the target plasmid using biotin-dUTP or digoxigenin-dUTP (Boehringer Mannheim, Laval, Quebec, Canada) as recommended by the manufacturer. Detection of biotinylated probes was performed with fluorescein-avidin (Oncor, Gaithersburg, MD, USA) and the signal was amplified with biotinylated goat anti-avidin antibody (Ab, Oncor) and fluorescein-avidin. Detection of DIG-labelled probes was achieved by mouse anti-DIG-Ab, goat DIG-labeled anti-mouse-Ab (2 μg/ml) and FITC-conjugated anti-DIG-Ab (50 μg/ml). Visualization of the biotinylated plasmid DNA was done with the avidin-fluorescein/biotinylated anti-avidin Ab sandwich method. Dose-response curves revealed that the fluorescence intensity detected in single cell by FISH technique was linearly proportional with the amount of injected plasmid DNA between 100 and 10000 copies of ds pGL2 and the detection limit was 40–50 copies of pGL2 (data not shown).

In situ end-labeling of free 3′-OH ends of fragmented plasmid and chromosomal DNA

Labeling of the free 3′-OH ends of fragmented chromosomal DNA was performed with the terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling (TUNEL) technique, using the ApoAlert (Clontech, Palo Alto, CA, USA) kit with the following modification. In order to enhance the sensitivity, the substrate of the terminal deoxynucleotidyl transferase, fluorescein-dUTP, was replaced by DIG-dUTP (10 μM). Fluorescence detection of the DIG-labeled nucleic acids was performed as described for FISH. Fragmentation of microinjected circular plasmid DNA was detected with the same method.

Quantitative single-cell fluorescence video-image analysis

Cells were observed with a Zeiss Axiovert 100 (North York, Canada) inverted fluorescence microscope equipped with a 63 × /NA 1.4 Planachromat objective using a 100 W quartz halogen illumination system. For quantitative evaluation of the fluorescence intensities associated with single cells, images were captured with a cooled-CCD camera (Princeton Instrument, Trenton, NJ, USA). Image acquisition was controlled by the Metafluor software (Universal Imaging, Hollis, NH, USA), operating on a Pentium Dell computer (Dell, Canada). Acquisition time was adjusted to avoid saturation of the camera. The integrated fluorescence intensity of FISH or TRITC signals, derived from injected cells, were calculated from background substracted images with the Metafluor program. Immunofluorescence and FISH photographs were taken with a Contax camera on Kodak EpH 1600 film using fluorescein or rhodamine filter set (Zeiss). Slides were scanned and processed with the Adobe-Photoshop 3.0 software. Exposure time, scanning and image processing parameters were kept identical within individual experiments. Scale bar: 10 μM.

Monitoring the cytosolic free Ca2+ concentration

Cytosolic Ca2+ concentration ([Ca2+]cy) of Fura-2 loaded HeLa and COS-1 cells was determined with single-cell fluorescence ratio video-imaging. Cells were loaded with 1 μM Fura-2/AM (Calbiochem, La Jolla, CA, USA, 20 min at 37°C) in tissue culture medium and observed with a Fluar 40 × 1.3 (Zeiss) objective. Cells were illuminated alternatively at 340 and 380 nm light with a Sutter filter wheel for 200 ms. The ratio of emitted fluorescence light (>460 nm) at the two exitation wavelengths provides a measure of [Ca2+]cy.65 Image aquisition and analysis were performed with the Metafluor software.

Immunocytochemistry

Cells were fixed (3% paraformaldehyde in PBS, 15 min at RT), permeabilized (0.2% Triton X-100 in PBS, 5 min) and blocked (0.5% BSA in PBS, blocking buffer, BB) for 30 min. Immunostaining of the ER and lysosomes was performed with the murine monoclonal anti-protein disulfide isomerase Ab (Stressgene, Victoria, BC, Canada, 1:500 dilution in BB, 1 h at RT) and the murine monoclonal anti-LAMP-2 Ab (Hybridoma Facility, Johns Hopkins University, 1:20 dilution in BB), respectively. The secondary FITC-conjugated goat anti-mouse Ab (Calbiochem) was used at 1:1000 dilution (1 h at RT). All washes were performed in BB containing 0.1% Triton X-100. Alternatively, lysosomes were labeled with TRITC-dextran (3 mg/ml, MW 70 kDa, lysine-fixable, Molecular Probes) overnight and chased in dextran-free medium for 2 h at 37°C. Early endosomes were visualized by FITC-transferrin (Molecular Probes) as described.66

Digitonin permeabilization of plated cells

HeLa cells were grown and microinjected on Labtek glass chamber slides. Cells were washed with ice-cold KH-Mg buffer (110 mM K-acetate, 20 mM HEPES, 5 mM Mg-acetate at pH 7.2) and permeabilized with digitonin (40 μg/ml in KH buffer) for 5 min on ice. Incubation of cells was carried out in ATP-regenerating system (5 mM Na2-ATP, 5 mM creatine phosphate and 20 U/ml creatine phosphokinase in KH buffer).

Preparation of cytosol extract by digitonin permeabilization

Isolated cytosol was obtained from HeLa cells by selective permeabilization of the plasma membrane with digitonin. All isolation steps were carried out at 4°C. After extensive washes with PBS and with KH-buffer, cells (5 × 108) were resuspended in an equivalent volume of KH buffer, containing digitonin (0.15–0.20 mg/ml, Calbiochem) and protease inhibitors (pepstatin, 2 μg/ml; leupeptin, 2 μg/ml; and phenylmethyl sulphonyl fluoride, 1 mM). The efficiency of permeabilization was considered to be successful when 80% of the cells became Trypan-blue positive (5–8 min incubation). The extracellular medium was collected by sedimenting the cells twice at 1000 g for 2 min. To eliminate any particulate contamination, the cytosolic extract was centrifuged at 40000 g for 40 min. The supernatant was stored in aliquots at −80°C. The permeabilization efficiency was also assessed by the release of the lactate dehydrogenase (LDH) activity from the cells. LDH activity was measured according to the manufacturer’s recommendations (LDH kit, Sigma, Oakville, ON, Canada, No. 340LD). Enzyme markers of lysosomes and the Golgi compartment (β-glucuronidase and α-mannosidase II activity) were determined fluorometrically as described.67

Southern blot analysis

Following the incubation of the plasmid DNA (25–50 ng) with cytoplasmic extract (0.1–0.2 mg), the DNA was isolated with phenol/chloroform, chloroform extraction in the presence of 0.2% SDS and precipitated with ammonium acetate/ethanol in the presence of yeast tRNA (3 μg per sample). The tRNA was digested with RNase A (10 μg/ml, 1 h at 37°C) and the DNA was fractionated on 0.8% agarose gel, transferred on to a nylon membrane (in 0.4 M NaOH) and hybridized with biotinylated probe (20 ng/ml, at 65°C overnight), prepared as described for FISH. Biotin-labeled nucleic acids were visualized by enhanced chemiluminescence (Amersham).

Nuclease activity

Nuclease activity of the cytosol was determined using linearized end-labeled plasmid DNA. For 5′ end-labeling the DNA was treated with calf intestine phosphatase, purified with phenol/chloroform extraction and phosphorylated with T4 polynucleotide kinase in the presence of γ-32P-ATP (30 μCi; specific activity 3000 Ci/mol; Amersham). For 3′ end-labeling, plasmid DNA was treated with the Klenow enzyme in the presence of α-32P-CTP (20 μCi; specific activity 3000 Ci/mol; ICN, Montreal, Quebec, Canada). Radioactive DNA was purified on ProbeQuant G-50 columns (Pharmacia, Baie d’Urfe, Quebec, Canada). Nuclease activity of the cytosolic extract (15 μg protein) was measured in KH-buffer in the presence of 2.5 fmol of 3′-32P-labeled DNA or 10 fmol of 5′-32P-labeled DNA. The plasmid DNA was precipitated with 15% trichloroacetic acid (TCA) in the presence of 2 mg/ml BSA (20 min, on ice), separated by centrifugation (17500 g, 15 min, 4°C) and the radioactivity remaining in the supernatant was measured by a Beckman LS7500 scintillation counter.

Acknowledgements

The authors thank Drs L-T Tsui, M Buchwald, S Grinstein and D Koehler for valuable comments. Support for this work was provided by the Canadian Cystic Fibrosis Foundation (Sparx II) and by an intramural grant of the Hospital for Sick Children to GLL. Instrumentation was in part covered by a Block Term Grant of the Ontario Thoracic Society. The anti-LAMP antibody developed by JT August was obtained from the Developmental Studies Hybridoma Bank maintained by the The University of Iowa, Department of Biological Sciences, Iowa City, under the contract NO1-HD-7-3263 from the NICHD. DL was supported by a Postdoctoral Fellowship of the Canadian Cystic Fibrosis Foundation. GLL is a Scholar of MRC Canada.

References

  1. 1.

    et al. Cellular and molecular barriers to gene transfer by a cationic lipid J Biol Chem 1995 270: 18997–19007

  2. 2.

    et al. Complexes of adenovirus with polycationic polymers and cationic lipids increase the efficiency of gene transfer in vitro and in vivo J Biol Chem 1997 272: 6479–6489

  3. 3.

    et al. Microtubular disruption prolongs the expression of human bilirubin-uridinediphosphoglucuronate-glucuronyltransferase-1 gene transferred into Gunn rat livers J Biol Chem 1996 271: 2341–2346

  4. 4.

    , , . On the mechanism of DNA transfection: efficient gene transfer without viruses Gene Therapy 1997 4: 1313–1321

  5. 5.

    , , , . Plasmid DNA entry into post-mitotic nuclei of primary rat myotubes Proc Natl Acad Sci USA 1995 92: 4572–4576

  6. 6.

    , , . The role of dioleoyl phosphatidylethanolamine in cationic liposome mediated gene transfer Biochim Biophys Acta 1995 1235: 289–295

  7. 7.

    , . Mechanism of oligonucleotide release from cationic liposomes Proc Natl Acad Sci USA 1996 93: 11493–11498

  8. 8.

    , . Intracellular distribution and mechanism of delivery of oligonucleotides mediated by cationic lipids Pharm Res 1996 13: 1367–1372

  9. 9.

    , . Cytoplasmic expression of a reporter gene by co-delivery of T7 RNNA polymerase and T7 promoter sequence with cationic liposomes Nucleic Acids Res 1993 21: 2867–2872

  10. 10.

    et al. Effect of co-lipids in enhancing cationic lipid-mediated gene transfer in vitro and in vivo Gene Therapy 1997 4: 716–725

  11. 11.

    , . Nuclear localization signal of SV40 T antigen directs import of plasmid DNA into sea urchin male pronuclei in vitro Mol Reprod Develop 1996 45: 431–438

  12. 12.

    , , , . A low rate of cell proliferation and reduced DNA limit cationic lipid-mediated gene transfer to primary cultures of ciliated human airway epithelia Gene Therapy 1997 4: 1173–1180

  13. 13.

    et al. Efficacy of a peptide-based gene delivery system depends on mitotic activity Gene Theapy 1996 3: 1133–1142

  14. 14.

    . DNA vector chemistry: the covalent attachment of signal peptides to plasmid DNA Nature Biotechnol 1998 16: 80–85

  15. 15.

    . Autophagy and related mechanisms of lysosome-mediated protein degradation Trends Cell Biol 1994 4: 139–143

  16. 16.

    et al. Nuclear import of DNA in digitonin-permeabilized cells J Cell Sci 1997 110: 2323–2331

  17. 17.

    , , , . Deoxyribonuclease II purified from the isolated lysosomes of porcine spleen and from porcine liver homogenates. Comparison with deoxyribonuclease II purified from porcine spleen homogenates Biochim Biophys Acta 1989 1007: 15–22

  18. 18.

    , . High efficiency polyoma DNA transfection of chloroquine treated cells Nucleic Acids Res 1983 11: 1295–1308

  19. 19.

    , , . Inhibition of Semliki forest virus penetration by lysomotropic weak bases J Gen Virol 1982 58: 47–61.

  20. 20.

    et al. Conformational maturation of CFTR but not its mutant counterpart (ΔF508) occurs in the endoplasmic reticulum and requires ATP EMBO J 1994 13: 6076–6086

  21. 21.

    , , . Nuclear import in permeabilized cells requires soluble cytoplasmic factors J Cell Biol 1990 111: 807–816

  22. 22.

    , . Effect of tetanus toxin on catecholamine release from intact and digitonin permeabilized chromaffin cells J Neurochem 1988 51: 451–456

  23. 23.

    et al. Cationic lipids improve antisense oligonucleotide uptake and prevent degradation in cultured cells and in human serum Biochem Biophys Res Com 1993 197: 818–825

  24. 24.

    et al. Enhanced resistance to nuclease degradation of nucleic acids complexed to asialoglycoprotein–polylysine carriers Nucleic Acids Res 1994 22: 5439–5446

  25. 25.

    et al. Stabilized plasmid-lipid particles: construction and characterization Gene Therapy 1999 (in press)

  26. 26.

    et al. Liposome-mediated transfection of fetal lung epithelial cells: DNA degradation and enhanced superoxide toxicity Am J Physiol 1998 275: L452–L460

  27. 27.

    et al. Identification of two lysosomal membrane glycoproteins J Cell Biol 1985 101: 85–95

  28. 28.

    , . Vesicle accumulation and exocytosis at sites of plasma membrane disruption J Cell Biol 1995 131: 1737–1745

  29. 29.

    , , . Cell membrane resealing by a vesicular mechanism similar to neurotransmitter release Science 1994 263: 390–393

  30. 30.

    et al. Overexpression of deoxyribuonuclease I (DNase I) transfected into COS-cells: its distribution during apoptotic cell death Eur J Cell Biol 1993 62: 397–405

  31. 31.

    et al. Distribution of DNase I in rat tissues and its correlation to cellular turnover and apoptosis (programmed cell death) Eur J Cell Biol 1994 64: 200–210

  32. 32.

    , , . The apoptotic endonucleases: cleaning up after cell death? Trends Cell Biol 1994 4: 37–41

  33. 33.

    et al. An apoptotic endonuclease activated either by decreasing the pH or by increasing calcium J Cell Sci 1996 109: 2393–2399

  34. 34.

    et al. Association of high molecular weight DNA fragmentation with apoptotic or non-apoptotic cell death induced by calcium ionophore FEBS Lett 1995 364: 264–267

  35. 35.

    et al. Ionic regulation of endonuclease activity in PC12 cells Biochem J 1995 311: 1033–1038

  36. 36.

    , , . Cleavage of CAD inhibitor in CAD activation and DNA degradation during apoptosis Nature 1998 391: 96–99

  37. 37.

    et al. A caspase-activated DNase that degrades DNA during apoptosis, and its inhibitor ICAD Nature 1998 391: 43–50.

  38. 38.

    et al. Affinity purified tetanus toxin binds to isolated chromaffin granules and inhibits catecholamine release in digitonin-permeabilized chromaffin cells FEBS Lett 1989 253: 121–128

  39. 39.

    , . Internucleosomal DNA cleavage and neuronal cell death/survival J Cell Biol 1993 122: 523–532

  40. 40.

    , , . Aurintricarboxilic acid, a putative inhibitor of apoptosis, is a potent inhibitor of DNA topoisomerase II in vitro and in Chinese hamster fibrosarcoma cells Biochem Pharmacol 1995 49: 305–313

  41. 41.

    , . Penetration of cells by herpes simplex virus does not require a low pH-dependent endocytic pathway Virus Res 1991 18: 271–290

  42. 42.

    , , , . Stepwise dismantling of adenovirus 2 during entry into cells Cell 1993 75: 477–486

  43. 43.

    , , . Mechanism of virus uncoating Trends Microbiol 1994 2: 52–56

  44. 44.

    . Glycerol and polylysine synergise in their ability to rupture vesicular membranes: a mechanism for increase transferrin–polylysine-mediated gene transfer Exp Cell Res 1997 232: 137–145

  45. 45.

    et al. The influence of endosome-disruptive peptides on gene transfer using synthetic virus-like gene transfer systems J Biol Chem 1994 269: 12918–12924

  46. 46.

    et al. Association with capsid protein promotes nuclear targeting of simian virus 40 DNA Proc Natl Acad Sci USA 1996 93: 96–100

  47. 47.

    , , . In vitro model for the nuclear transport of the hepadnavirus genome J Virol 1997 71: 1310–1316

  48. 48.

    et al. The cytosol constitutes a metabolic barrier against transgene delivery to the nucleus Ped Pulm 1997 S14: 255

  49. 49.

    . High efficiency transformation by direct microinjection of DNA into cultured mammalian cells Cell 1980 22: 479–488

  50. 50.

    , , , . Hindered diffusion of inert tracer particles in the cytoplasm of mouse 3T3 cells Proc Natl Acad Sci USA 1987 84: 4910–4913

  51. 51.

    . Physical properties of cytoplasm Curr Opin Cell Biol 1994 6: 3–9

  52. 52.

    , , . Translational diffusion of macromolecule-sized solutes in cytoplasm and nucleus J Cell Biol 1997 138: 131–142

  53. 53.

    et al. Polyethylenimine but not the cationic lipids promotes transgene delivery to the nucleus in mammalian cells J Biol Chem 1998 273: 7507–7511

  54. 54.

    , , . Endogenous nucleases J Biol Chem 1982 257: 13009–13017

  55. 55.

    , , , . Purification and characterization of a potent endonuclease in extracts of bovine heart mitochondria J Biol Chem 1987 262: 2005–2015

  56. 56.

    . Glucocorticoid-induced thymocyte apoptosis is associated with endogenous endonuclease activity Nature 1980 284: 555–556

  57. 57.

    et al. Involvement of DNase II in nuclear degeneration during lens cell differentiation J Biol Chem 1995 270: 28579–28585

  58. 58.

    et al. Androgen ablation leads to an upregulation and intranuclear accumulation of DNase I in rat prostate epithelial cells paralleling their apoptotic elimination J Cell Biol 1997 137: 909–923

  59. 59.

    , . Identification of deoxyribonuclease II as an endonuclease involved in apoptosis Arch Biochem Biophys 1993 300: 440–450

  60. 60.

    , , . Native recombinant cyclophilins A, B, and C degrade DNA independently of peptidylprolyl cis-trans-isomerase activity J Cell Biol 1997 272: 6677–6684

  61. 61.

    , , . Microtubule-mediated transport of incoming herpes simplex virus 1 capsids to the nucleus J Cell Biol 1997 136: 1007–1021

  62. 62.

    , , . A dye-buoyant-density method for the detection and isolation of closed circular duplex DNA: the closed circular DNA in HeLa cells Proc Natl Acad Sci USA 1967 57: 1514–1521

  63. 63.

    , . Production of single-stranded plasmid DNA Meth Enzymol 1987 153: 3–11

  64. 64.

    . DNA localization in metaphase and interphase cells by fluorescence in situ hybridization Meth Cell Biol 1991 35: 3–35

  65. 65.

    , . Technique for in situ measurements of calcium in intracellular inositol 1,4,5-trisphosphate-sensitive stores using the fluorescent indicator mag-fura-2 Proc Natl Acad Sci USA 1993 90: 2598–2602

  66. 66.

    , , , . Mutagenesis of the human transferrin receptor: two cytoplasmic phenylalanines are required for efficient internalization and a second-site mutation is capable of reverting an internalization-defective phenotype J Cell Biol 1991 112: 853–861

  67. 67.

    , . Isolation of subcellular organelles Meth Enzymol 1990 182: 203–227

Download references

Author information

Affiliations

  1. Program in Cell and Lung Biology and Lung Gene Therapy, Hospital for Sick Children, Toronto, Ontario, Canada

    • D Lechardeur
    • , K-J Sohn
    • , M Haardt
    •  & G L Lukacs

  2. Inex Pharmaceuticals, Burnaby, BC, Canada

    • P B Joshi
    • , M Monck
    •  & R W Graham

  3. Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Canada

    • B Beatty
    • , J Squire
    •  & G L Lukacs

  4. Department of Pediatrics, University of Toronto, Toronto, Canada

    • H O’Brodovich

Authors

  1. Search for D Lechardeur in:

  2. Search for K-J Sohn in:

  3. Search for M Haardt in:

  4. Search for P B Joshi in:

  5. Search for M Monck in:

  6. Search for R W Graham in:

  7. Search for B Beatty in:

  8. Search for J Squire in:

  9. Search for H O’Brodovich in:

  10. Search for G L Lukacs in:

Corresponding author

Correspondence to G L Lukacs.

To obtain permission to re-use content from this article visit RightsLink.

About this article

Publication history

Received

Accepted

Published

DOI

https://doi.org/10.1038/sj.gt.3300867