Double-stranded DNA can be translocated across a planar membrane containing purified mitochondrial porin

^a CNR Unit for Biomembranes, Department Biomedical Sciences, University of Padova, Italy
^b Laboratory of Biomembranes, ERS CNRS 571, University of Paris-sud, France
^c Institute of Biochemical and Pharmaceutical Sciences, University of Catania, Italy

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
The transport of genetic material across biomembranes is a process of great relevance for several fields of study. However, much remains to be learned about the mechanisms underlying transport, one of which implies the involvement of proteic DNA-conducting pores. Entry of genetic material into mitochondria has been observed under both physiological and pathological conditions. We report here that double-stranded DNA can move through a planar bilayer membrane containing isolated mitochondrial porin (voltage-dependent anion channel). The transport is driven by the applied electrical field, and the presence of DNA is associated with a decrease of current conduction by the pores. The passage of DNA does not take place if the bilayer has not been doped with any protein or in the presence of both reconstituted porin and anti-porin antibody. Translocation does not occur if the bilayer contains Shigella sonnei maltoporin, gramicidin A channels, or a 30 pS anion-selective channel plus other proteins. These results show that mitochondrial porin is capable of mediating the transport of genetic material, revealing a new property of this molecule and futher confirming the idea that DNA can move through proteic pores.—Szabò, I., Bàthori, G., Tombola, F., Coppola, A., Schmehl, I., Brini, M., Ghazi, A., De Pinto, V., Zoratti, M. Double-stranded DNA can be translocated across a planar membrane containing purified mitochondrial porin. FASEB J. 12, 495–502 (1998)

Key Words: DNA transport • channels • VDAC • mitochondria

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
THE TRANSLOCATION of genetic material across biomembranes is a process of fundamental as well as medical interest about which much remains to be learned. Different mechanisms apply depending on the system being considered. Endocytotic processes are involved in most vector-mediated transport through the plasma membrane of eukaryotic cells, a crucial step in gene therapy, with details depending on the vector system used. On the other hand, large channels seem to be able to mediate the transport of proteins and genetic material in both prokaryotic and eukaryotic organisms. The existence of protein-conducting pores was documented earlier in a few cases [e.g., in the endoplasmatic reticulum (1) and in the nuclear membrane (2)]. It has been suggested that in prokaryotes, nucleic acid transport may use pore-forming proteins (3, 4). Some of the most interesting cases concern bacterial phage receptors—FhuA in particular, which forms a large aqueous channel upon binding phage T5 in a reconstituted system (5). Channels formed by filamentous phage proteins Pb2 (6, 7) and G3p (8) have also been associated with the injection of DNA into parasitized cells. In eukaryotic systems, studies of pore-mediated translocation of RNA and DNA have concentrated on processes involving the nuclear pore complex and, in plants, plasmodesmata (9). Cotransport with proteins takes place in these instances.

Direct evidence for pore-mediated nucleic acid transport has been obtained only recently. Movement of single-stranded DNA through isolated Staphylococcus aureus -hemolysin pores has been detected by Kasianowicz et al. (10). We have shown that fusion of Bacillus subtilis plasma membrane vesicles containing high-conductance channels with an artificial membrane can make the latter permeable to double-stranded DNA (4), suggesting that some bacterial ion channels might be involved in DNA transport during processes such as transformation or horizontal gene transfer.

Mitochondria are widely believed to have originated as endosymbiotic bacteria (11). Genetic material can be transported across mitochondrial membranes, despite the presence of a matrix-negative transmembrane voltage gradient, both in physiological and pathological conditions. One of the most interesting findings is that HIV-1 RNA accumulates in the mitochondria of infected lymphocytes. Using an inducible cell line, it was found that RNA reached the mitochondria within 1 h after stimulation of viral expression (12). Nuclear-encoded tRNAs, which have an L-shaped, coiled 3-dimensional structure, are imported into the mitochondria of plants and protozoa (13–16). In mammalian mitochondria, a well-documented case is that of the RNA subunit of mitochondrial RNA processing endonuclease (17, and references therein). Chimeras formed by a leader peptide and an oligonucleotide can be imported into isolated, energized yeast mitochondria (18, 19). The machinery responsible for these translocations is unknown, but at least sometimes the protein import apparatus appears to be involved (15–19), implying transport via proteic channels.

A macromolecule transport function in mitochondria might be fulfilled by porin [also called voltage-dependent anion channel (VDAC)]⁵ (20–22), presumably in association with proteins of the inner membrane, since it forms a large, anion-selective pore (approx. 3–4 nm) (21, 23), is localized in the outer membrane but also in the contact sites of mitochondria (24), can form supramolecular heterocomplexes (24–26), and allows the passage of ATP, a nucleotide (27). VDAC is the only mitochondrial channel available as isolated protein in highly purified form, and has been characterized from molecular, genetic, and biophysical points of view. Upon reconstitution into artificial membranes, the porin exhibits well-documented properties: a full conductance of about 450 pS in 100 mM KCl; a variety of conductance states; and a reduction of the average conductance with increasing voltages of either sign (21, 22).

Porin is also present in the plasma membrane of various eukaryotic cells, where its function is unknown (28). We have proved (G. Bàthori et al., unpublished results) its presence in caveolas, as suggested by earlier results of Lisanti and colleagues (29). Caveolas are specialized regions of the plasma membrane forming characteristic invaginations and are involved in solute transport by potocytosis. They have been suggested to be a site of entry for oligonucleotides (30). Porin is suspected to provide the pathway for solute efflux from the potocytotic vesicles (29), and may thus be involved in the entry of genetic material into cells.

The focus of the present work was to find out whether DNA might be translocated through porin molecules in vitro so as to shed light on some possible mechanisms of the processes mentioned above. We applied an experimental procedure that allowed us to follow DNA translocation and the electrophysiological behavior of porin molecule (or molecules) simultaneously.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
Materials
Beef heart mitochondrial porin was purified following established procedures (31, 32), using Triton X-100 or octylglucoside as detergents. The purity of the preparation was ascertained by gel electrophoresis and silver staining (not shown). Antiserum containing polyclonal antibodies against purified porin was obtained from rabbits (33), and the antibodies were purified after the procedure in ref 34. Maltoporin (LamB) was purified from Escherichia coli Pop 154, a strain carrying the malB region of S. sonnei 3070 (35), according to the procedure in ref 36. A 633 bp, double-stranded DNA segment encoding the cDNA of aequorin, purified from the recombinant plasmids mtAEQ-pcDNAI (37) or cytAEQ-pcDNAI (38) (referred to as `DNA') or solutions of the parent 4.2 Kb plasmids themselves, was used as the DNA to be translocated. The plasmid preparations contained linearized and/or fragmented DNA produced during handling of the plasmid solutions. These materials were chosen because of their availability in our department.

DNA translocation experiments
Use was made of a standard electrophysiological planar bilayer apparatus (39). Bilayers of approx. 300 pF capacity were prepared by painting a chloroform or decane solution of purified soybean azolectin (Sigma, St. Louis, Mo.) across a smoothed hole in a Teflon film (25 µm thickness, Goodfellow) separating two chambers (3 ml) carved in a Teflon block. The standard experimental medium was 100 mM KCl, 0.1 mM CaCl₂, 20 mM Hepes/K⁺, pH 7.2. Connections to the electrodes were provided by agar bridges. All voltages reported are those of the cis chamber, zero being assigned by convention to the trans (grounded) side. The contents of both chambers were stirred by magnetic bars when needed. A few micrograms of purified porin were added to the cis chamber; after incorporation of one or a few channels, the cis chamber was perfused with 20 ml of medium to prevent further incorporation. The activity of the channels was monitored for several minutes and only channels exhibiting stable activity at ± 40 or 50 mV were used for DNA translocation experiments. In some experiments, a sample of medium was taken from the trans side for future use as a control against contamination (referred to as -1'). Aequorin cDNA (633 bp; see above) at subnanomolar final concentrations (see figure legends) or plasmid (final DNA concentration: 0.8–16 µg/ml) was added to the cis compartment; samples were taken at regular intervals from the trans side while channel activity was monitored and recorded on tape for off-line analysis (Axon Pclamp 6.0.2 program set). The voltage was turned to 0 while the samples were taken. Samples (referred to as `final cis' and `final trans') from both compartments were taken 10–30 min after the eventual membrane collapse for use as positive controls in subsequent polymerase chain reaction (PCR) amplifications (see below). The samples were stored at -20°C until needed. Stringent precautions were necessary to avoid contamination artifacts. For example, after each experiment the chambers and stirrers were left overnight under sulfochromic mixture and the agar bridges were discarded.

DNA detection
Aliquots of the samples withdrawn during each experiment were subjected to 35- or sometimes 40-cycle PCR amplification. The primers were designed to lead to amplification of the 633 bp segment. By-products running slightly ahead of the desired product could rarely be observed, probably indicating the formation of fragmented DNA. Amplification of DNA standards in 1 M KCl gave irreproducible results due to the high salt concentration. Samples (20 µl) of the amplified mixtures were electrophoresed on 1% agar gels and Southern blotted onto nylon (Boehringer-Mannheim, Mannheim, Germany). The blots were processed using the Boehringer-Mannheim digoxigenin-based, nonradioactive DNA labeling and detection reagents, with CSPD as a substrate for chemiluminescent detection (Kodak Biomax or X-Omatic film). In amplifications to test the sensitivity of the DNA detection protocol, a clear band (fainter ones were often observed) was produced by samples containing about 10 molecules of plasmid DNA, indicating that our lower detection limit was 6000 amplification templates in the trans chamber (not shown; see ref 4).

Extreme precaution was taken to avoid contamination artifacts. For example, variable-volume pipettes (kept segregated by type of use) were routinely washed (HCl/NaOH/UV light) every few days. Only filter-protected tips were used throughout this work. Water samples were routinely included in PCR runs as a control.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
DNA translocation
DNA translocation across a planar bilayer doped with VDAC channels was examined. DNA (a 633 bp segment) was added on one side (cis) of the membrane, and samples were subsequently taken from the opposite (trans) chamber while channel activity was recorded. During the experiments, a cis side-negative transmembrane voltage (usually -40 or -50 mV) was applied in order to provide a driving force for translocation of the DNA, a polyanion, from the cis to the trans side. The collected samples were then subjected to PCR amplification and Southern blotting to verify whether they contained translocated DNA.

The blot shown in Fig. 1A presents both a control experiment (lanes 1–4) and an experiment with VDAC (lanes 6–9) that were processed together. In the control experiment, only the DNA was added to the chamber, i.e., no proteins were incorporated into the bilayer. No amplification product could be detected in any samples. In this particular control experiment, a leak current developed after the 60 min sample was taken, progressing from random `fast noise' ( Fig. 1B, lower trace) to a steady conductance in the 5 nS range at the end of the experiment. Despite this, no DNA was detected in the samples taken 80 and 100 min after plasmid addition, indicating that nonspecific conductances cannot mediate the transport. The same result was obtained from nine other such experiments, displaying no `leaks', lasting up to 240 min.

View larger version (39K): [in this window] [in a new window]   Figure 1. Translocation of double-stranded DNA across a mitochondrial porin-doped but not across an `empty' planar bilayer. A) Southern blot of the amplified samples from a control experiment (lanes 1–4) and an experiment with VDAC (lanes 6–9). The numbers above the lanes refer to the time (in minutes) after DNA addition in cis, when the samples were taken from the trans chamber. Lanes 5 and 10: `PCRed' water. Lane 11: mass markers (Boehringer-Mannheim DIG-labeled type II, 564 bp band). Lanes 12 and 13: `final trans' and `final cis' (see Experimental Procedures) samples of the control experiment, respectively (the amount of amplified mixture loaded onto the gel was 1/20 and 1/100 that of the other lanes, respectively). DNA concentrations in cis were 300 pM (control) and 540 pM (VDAC). The direction of migration was toward the top. B) Representative current traces from the control experiment. The upper trace was recorded immediately after taking the first sample, the lower one about 10 min after the second sample. The spikes at left signal the transition from 0 to -40 mV applied potential (cis side). Filter: 0.5 KHz. Digital sampling frequency: 2 KHz. C) Current traces illustrating VDAC activity during the experiment shown in Fig. 1A. The lower trace shows an often-visited substate level. V(cis): -50 mV. Filter: 0.5 KHz. Digital sampling rate: 2 KHz.

In sharp contrast, bands were produced by all four samples taken in the experiment with VDAC ( Fig. 1A, lanes 6–9), in agreement with the outcome of 19 other analogous experiments. In three experiments, translocation could not clearly be detected. The electrophysiological activity of VDAC from the same experiment of Fig. 1A (after the addition of DNA) is exemplified in Fig. 1C.

To check the idea that DNA translocation was essentially an electrophoretic process, we performed experiments with incorporated VDAC in which a transmembrane voltage difference of opposite polarity (cis side-positive) was applied. No DNA was detected (in five of six experiments) in samples from the trans side. One such experiment is illustrated in Fig. 2 A, B.

View larger version (92K): [in this window] [in a new window]   Figure 2. DNA translocation does not take place in the presence of a cis side-positive voltage gradient (A, B) or anti-VDAC antibodies (C, D). A) Southern blot from a representative DNA translocation experiment conducted at V(cis) +50 mV. Cis-side DNA concentration: 150 pM. Final cis: the amount of amplified mixture loaded onto the gel was 1/200th that of the other lanes. B) Current trace from the experiment shown in panel A. Filter: 0.5 KHz. Digital sampling rate: 2 KHz. C) Southern blot. After incorporation of porin and perfusion of the cis chamber, purified rabbit anti-VDAC polyclonal antibodies (1.25 µg/ml) were added to both chambers; after 10 min, DNA (final concentration: 200 pM) was added in cis. Mass markers: Boehringer-Mannheim DIG-labeled type VI. Final trans: the amount of amplified mixture loaded onto the gel was 1/1000th that of the other lanes. V(cis): -40 mV. D) Representative current trace from the experiment shown in panel C (0/-40 mV transition after taking the 30 min sample). Leaks developed before the last sample was taken (not shown). Filter 0.4 KHz. Digital sampling rate: 2 KHz.

Translocation could also be abolished if purified anti-VDAC antibodies (in seven experiments of seven) or antiserum containing polyclonal antibodies raised against purified VDAC (in seven of nine experiments) was added before introduction of DNA into the cis compartment ( Fig. 2C, D). Preincubation of anti-VDAC antibodies with VDAC molecules prevented porin incorporation into the membrane (n=4; not shown; see also ref 43).

To verify that DNA translocation was not an unspecific process caused by the presence of any protein in the membrane, we incorporated another porin into the planar bilayer: S. sonnei maltoporin. The structure of this inducible porin and phage receptor of the outer membrane of Gram-negative bacteria has been determined by crystallographic methods (40, 41): the constriction area of the pore has a diameter of only 0.5 nm. No DNA translocation took place in two experiments, lasting 140 and 280 min, with several LamB channels in the bilayer. Figure 3 illustrates one of these experiments. The same result was obtained in analogous control experiments using gramicidin A exhibiting 25 pS conductance in 150 mM KCl (n=4) and a mammalian plasma membrane preparation containing several proteins, but exhibiting activity only by a 30 pS anion-selective channel (n=2) (not shown, see ref 4).

View larger version (41K): [in this window] [in a new window]   Figure 3. DNA is not translocated in the presence of Shigella sonnei maltoporin. A) Southern blot. DNA concentration: 2 nM. Mass markers: Boehringer-Mannheim DIG-labeled type VI. Final cis: the sample was diluted 1 million-fold before amplification. V(cis): -40 mV. B) Representative current trace. Filter: 0.5 KHz. Digital sampling rate: 2 KHz.

Electrophysiological effects
One major advantage of our experimental approach is the opportunity to simultaneously follow DNA translocation and electrophysiological activity. Records obtained from the translocation experiments in our standard medium (100 mM KCl) did not provide clear-cut evidence either for or against DNA-induced changes in channel activity. VDAC displays a variety of conductance substates, often closely spaced, with changes in channel behavior on a time scale of minutes. Since in these experiments we recorded only relatively short periods before the addition of DNA, in order to maximize translocation time, meaningful comparisons generally were not feasible.

We specifically investigated this aspect in DNA translocation experiments conducted in 1 M KCl to allow a better resolution of conductance changes (n=5). Channel activity was monitored for at least 10 min before and after the addition of DNA, at a constant voltage (-40 or -50 mV). Analysis was based on the comparison of total current amplitude histograms: the average probability of a given current level during the period examined is proportional to the height of the corresponding bin in the histogram. In four of five cases, the histograms obtained from the traces recorded after addition of DNA (`+DNA') were significantly shifted toward zero: reduction or disappearence of some of the higher current levels occurred, with a corresponding increased occupancy of lower ones. In three cases, the amplitude peaks were less broad, indicating that high current levels were visited less often and/or for shorter periods. One of these experiments, in which the main current levels decreased by approx. 40% in the presence of DNA, is illustrated in Fig. 4 A, B. Inspection of the current traces showed that in some cases the channels exhibited relatively slow (several millisecond range) and large (nS range) closures, which became more prevalent after the addition of DNA. During the same time course, we never observed spontaneous rundown of VDAC activity in the absence of DNA in control experiments.

View larger version (18K): [in this window] [in a new window]   Figure 4. Effect of DNA on VDAC channel activity. Current amplitude histograms and exemplificative current record segments. A) No addition. The histogram was obtained from a 658 s-long record. B) In the presence of 300 pM DNA in the cis chamber. The histogram was based on a 581 s-long record. A, B) Medium: 1 M KCl, 20 mM Hepes/K+, 0.1 mM CaCl2, pH 7.2. V(cis): -50 mV. Filter: 0.5 kHz. Digital sampling rate: 2.

To exclude the possibility that the observed electrophysiological changes could be due to a nonspecific charge-screening effect of the DNA polyanion and to ascertain the correlation between DNA translocation and channel behavior, we performed experiments under the same conditions with reconstituted LamB (n=3). In none of these experiments did we observe a significant shift in the amplitude histograms after DNA addition ( Fig. 5 A, B).

View larger version (15K): [in this window] [in a new window]   Figure 5. Lack of DNA effect on maltoporin activity. Current amplitude histograms and representative current traces before (A) and after (B) addition of DNA (2 nM end concentration) into the cis chamber. The histograms were based on 800 s-long and on 2100 s-long records, respectively. Conductance steps of approx. 60, 120, and 180 pS were observed in 1 M KCl, consistent with the gating of one, two, or three subunits of the LamB trimers. Medium: as in Fig. 4. V(cis): -40 mV. Filter: 0.5 kHz. Digital sampling rate: 2 kHz.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 
This work is the first demonstration that isolated and reconstituted mitochondrial porin can mediate the in vitro transport of genetic material—in particular, of a double-stranded helix—in an electrophoretic process without the necessity of accessory proteins. The results indicate a new property of this well-characterized molecule with potentially important implications and further confirm that DNA can move through proteic pores. The study integrates electrophysiological methods with molecular biology techniques, offering a methodology for investigations of DNA/RNA transport in various systems.

Concerning the specificity of the pathway followed by DNA, we show that it cannot cross the bilayer itself, since translocation does not take place through empty bilayers even when they show a leak conductance. The results of control experiments with LamB porins, gramicidin A channels, and vesicles containing several proteins as well as a 30 pS channel indicate that translocation is specifically linked to VDAC, and it is not just due to the presence of proteins in the bilayer. Another observation indicating the specificity of the process is that DNA translocation is inhibited by anti-VDAC antibodies, whereas current conduction is not. The binding of antibodies to porin would be expected to physically reduce the accessibility of the pore to macromolecules, but it does not necessarily affect the motion of the small current-conducting ions (42). Monoclonal antibodies against the amino-terminal of plasma membrane porin `31HL' have also been reported to have no effect on current conduction by that channel (43).

The size of the VDAC pore is compatible with DNA translocation through it. Using a similar experimental approach, Kasianowicz et al. (10) recently showed that single-stranded DNA can move through the pore formed by S. aureus -hemolysin, which has a conductance of 1 nS in 1 M KCl and a diameter of 1.4 nm at its narrowest point (44). No structural information is available about the B. subtilis channels shown to mediate double-stranded DNA transport (4), but their conductance is in the same range (0.1–1.5 nS in 100 mM KCl) as porin's. VDAC has a maximal conductance of 450 pS in 100 mM KCl; estimates of the pore diameter based on imaging studies range up to 4 nm (21, 23), sufficient to allow the passage of an uncoiled, partly desolvated DNA double helix (diameter: approx. 2 nm). Even if the presence in our experiments of a transmembrane voltage gradient caused the channels to occupy mostly lower-than-maximal conductance states, DNA translocation could be observed. The conductance of a channel is not necessarily correlated with the size of the molecules capable of moving through it (45). The existence of multiple substates and gating patterns for VDAC indicate that it can adopt a variety of different conformations: it is a very flexible and `adaptable' channel (21).

The ß-barrel structure might be intrinsically suitable for DNA transport: the S. aureus -hemolysin channel is a ß-barrel (44), and VDAC is believed to be a ß-barrel pore (23) with a 13 (46) or 16 strands (47) structure. The bacterial phage receptor FhuA, hypothesized to be involved in the transport of phage genetic material, also contains ß-strands and is predicted to be a ß-barrel by modeling studies (48).

Concerning the electrophysiological behavior of the channel during the DNA translocation, the major observed effect of the presence of DNA was the reduced occupancy or disappearance of the higher current levels, along with the increased occupancy of lower ones. A similar phenomenon was also observed in the case of B. subtilis channels after the addition of DNA (4). This effect might well be due to the presence of DNA in the lumen of the pores, causing a reduction in the flow of ions. In some parts of the VDAC experiments we observed gating that resembled a relatively slow and partial block of the pore. At any rate, the translocation of bulky molecules through pores need not necessarily lead to a complete block of the channel (49). Indeed, in the case of B. subtilis, a flickering behavior became prominent (4), whereas clearly identifiable blocking events were recorded in the S. aureus -hemolysin channel (10). Thus, an electrophysiological effect was detectable in all three cases, but not with identical characteristics. The differences might be due simply to the different sizes and properties of the pores and of the transported DNA species. No electrophysiological effect of DNA was observed in the case of LamB, in which translocation did not take place. These results thus correlate DNA translocation with changes in channel behavior.

The finding that VDAC is able to allow the passage of DNA is not only of biophysical interest, but may also have applications. Mutations and deletions of mitochondrial DNA are known to be at the origin of several pathologies (50, 51), and their role in aging has been suggested (52, 53). Gene therapy of these pathologies appears to be problematic because "no suitable transformation systems for reintroducing mitochondrial genes into the organelle are available" (54). The possibility that VDAC might be used to create a pathway for the introduction of genetic material into mitochondria should be considered. Under physiological conditions, the presence of a matrix-negative transmembrane potential in mitochondria argues against the possibile import of negatively charged genetic material mediated exclusively by porin. For such a transport to occur, other proteins would have to be involved. The problem of importing a negative macromolecule against an unfavorable voltage gradient has been solved in the prokaryotic kingdom, where phages have exploited multifunctional, multiprotein systems whose normal function is to transport macromolecules other than DNA. In mitochondria, such a helper function might be fulfilled by the proteins that form complexes with porin. For example, VDAC is a component of the so-called mitochondrial benzodiazepine receptor (55), one of whose proposed functions is to import heme precursors (56). VDAC also form complexes with the adenine nucleotide translocator and hexo- or cretaine kinases that span the mitochondrial membrane system and serve as conduits for ATP (24). Under appropriate conditions, these complexes have both been proposed to form the large multiprotein permeability transition pore (57), which would be suitable for the translocation of macromolecules.

Porin is present not only in the mitochondrial membranes, but also in the plasma membrane of practically all cells examined (28). As mentioned, caveolas presumably involved in oligonucleotide transport (30) also contain VDAC (G. Bàthori et al., unpublished results). Work is under way in our laboratory to clarify the possible involvement of porin in DNA transport in this system, opening new perspectives for gene therapy applications.

	ACKNOWLEDGMENTS

 
We thank Prof. G. Schatz for critical reading of the manuscript and for helpful comments. We are grateful to Prof. T. Pozzan and Dr. R. Rizzuto for useful discussions, advice on technical problems, and support. We also thank Drs. C. Di Francesco, M. Scarpa, and L. Dalla Valle for help with DNA detection procedures, Prof. D. Wolff for help with the bilayer apparatus, Dr. M. Sargiacomo for plasma membrane preparation, Dr. C. Berrier for isolating maltoporin, and Dr. A. Messina for help with isolation of mitochondrial porin. Dr. C. Cola performed some of the initial experiments. We thank A. Milani and M. Santato for technical help. I.Sz. holds a Magyary Zoltan postdoctoral fellowship from the Foundation for the Hungarian Higher Education and Research. A.C. participated as a trainee supported by a CNR fellowship. I.Sz. holds a postdoctoral fellowship from the University of Padova. G.B.'s participation was financed in part by an EMBO short-term fellowship and a Telethon fellowship. We are especially grateful to Telethon-Italy for financial support (grants A.44 and A.59 to M.Z. and 711 to V.D.P.).

	FOOTNOTES

 
¹ The first two authors contributed equally to the study and therefore share first authorship.

³ Permanent address: Department of Physiology, Semmelweis Medical University, Budapest, Hungary.

⁴ Permanent address: Department of Atomic Physics, Eötvös University, Budapest, Hungary.

¹ Correspondence: CNR Centro Biomembrane, Dip. Scienze Biomediche, Viale G. Colombo, 3, 35121 Padova, Italy. E-mail: zoratti{at}civ.bio.unipd.it

⁵ Abbreviations: PCR, polymerase chain reaction; LamB, maltoporin; VDAC, voltage-dependent anion channel (or channels).

Received for publication November 4, 1997. Accepted for publication December 2, 1997.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

 

1. Simon, S. M., and Blobel, G. (1991) A protein-conducting channel in the endoplasmic reticulum. Cell 65, 371–380[Medline]
2. Bustamante, J. O., Hanover, J. A., and Liepins, A. (1995) The ion channel behavior of the nuclear pore complex. J. Membr. Biol. 146, 239–251[Medline]
3. Dreiseikelmann, B. (1994) Translocation of DNA across bacterial membranes. Microbiol. Rev. 58, 293–316[Abstract/Free Full Text]
4. Szabò, I., Bàthori, G., Tombola, F., Brini, M., Coppola, A., and Zoratti, M. (1997) DNA translocation across planar bilayers containing Bacillus subtilis ion channels. J. Biol. Chem. 272, 25275–25282[Abstract/Free Full Text]
5. Bonhivers, M., Ghazi, A., Boulanger, P., and Letellier, L. (1996) FhuA, a transporter of the Escherichia coli outer membrane, is converted into a channel upon binding of bacteriophage T5. EMBO J. 15, 1850–1856[Medline]
6. Feucht, A., Schmidt, A., Benz, R., Schwarz, H., and Heller, K. J. (1990) Pore formation associated with the tail-tip protein pb2 of bacteriophage T5. J. Biol. Chem. 265, 18561–18567[Abstract/Free Full Text]
7. Guihard, G., Boulanger, P., and Letellier, L. (1992) Involvement of phage T5 tail proteins and contact sites between the outer and inner membrane of Escherichia coli in phage T5 DNA injection. J. Biol. Chem. 267, 3173–3178[Abstract/Free Full Text]
8. Glaser-Wuttke, G., Keppner, J., and Rasched, I. (1989) Pore-forming properties of the adsorption protein of filamentous phage fd. Biochim. Biophys. Acta 985, 239–247[Medline]
9. Citovsky, V., and Zambryski, P. (1993) Transport of nucleic acids through membrane channels: snaking through small holes. Annu. Rev. Microbiol. 47, 167–197[Medline]
10. Kasianowicz, J. J., Brandin, E., Branton, D., and Deamer, D. W. (1996) Characterization of individual polynucleotide molecules using a membrane channel. Proc. Natl. Acad. Sci. USA 93, 13770–13773[Abstract/Free Full Text]
11. Schwartz, R. M., and Dayhoff, M. O. (1978) Origins of prokaryotes, eukaryotes, mitochondria and chloroplasts. Science 199, 395–403[Free Full Text]
12. Somasundaran, M., Zapp, M. L., Beattie, L. K., Pang, L., Byron, K. S., Bassell, G. J., Sullivan, J. L., and Singer, R. H. (1994) Localization of HIV RNA in mitochondria of infected cells: potential role in cytopathogenicity. J. Cell Biol. 126, 1353–1360[Abstract/Free Full Text]
13. Small, I., Maréchal-Drouard, L., Masson, J., Pelletier, G., Cosset, A., Weil, J. H., and Dietrich, A. (1992) In vivo import of a normal or mutagenized heterologous transfer RNA into the mitochondria of transgenic plants: towards a way of influencing mitochondrial gene expression? EMBO J. 11, 1291–1296[Medline]
14. Hauser, R., and Schneider, A. (1995) tRNAs are imported into mitochondria of Trypanosoma brucei independently of their genomic context and genetic origin. EMBO J. 14, 4212–4220[Medline]
15. Tarassov, I., Entelis, N., and Martin, R. P. (1995) Mitochondrial import of cytoplasmic lysine-tRNA in yeast is mediated by cooperation of cytoplasmic and mitochondrial lysyl-tRNA synthetases. EMBO J. 14, 3461–3471[Medline]
16. Tarassov, I. A., Entelis, N. S., and Martin, R. P. (1995) An intact protein translocation machinery is required for mitochondrial import of a yeast cytoplasmic tRNA. J. Mol. Biol. 245, 315–323[Medline]
17. Li, K., Smagula, C. S., Parsons, W. J., Richardson, J. A., Gonzalez, M., Hagler, H. K., and Williams, R. S. (1994) Subcellular partitioning of MRP RNA assessed by ultrastructural and biochemical analysis. J. Cell Biol. 124, 871–882[Abstract/Free Full Text]
18. Vestweber, D., and Schatz, G. (1989) DNA-protein conjugates can enter mitochondria via the protein import pathway. Nature (London) 338, 170–172[Medline]
19. Seibel, P., Trappe, J., Villani, G., Klopstock, T., Papa, S., and Reichmann, H. (1995) Transfection of mitochondria: strategy towards a gene therapy of mitochondrial DNA diseases. Nucleic Acids Res. 23, 10–17[Abstract/Free Full Text]
20. Benz, R. (1994) Permeation of hydrophilic solutes through mitochondrial outer membranes: review on mitochondrial porins. Biochim. Biophys. Acta 1197, 167–196[Medline]
21. Colombini, M. (1994) Anion channels in the mitochondrial outer membrane. Curr. Top. Membr. 42, 73–101
22. Colombini, M., Blachly-Dyson, E., and Forte, M. (1996) VDAC, a channel in the outer mitochondrial membrane. In Ion Channels (Narahashi, T., ed) Vol. 4., pp. 169–202, Plenum Press, New York
23. Mannella, C. A., Forte, M., and Colombini, M. (1992) Toward the molecular structure of the mitochondrial channel, VDAC. J. Bioenerg. Biomembr. 24, 7–19[Medline]
24. Brdiczka, D. (1991) Contact sites between mitochondrial envelope membranes. Structure and function in energy- and protein-transfer. Biochim. Biophys. Acta 1071, 291–312[Medline]
25. McEnery, M. W., Snowman, A. M., Trifiletti, R. R., and Snyder, S. H. (1991) Isolation of the mitochondrial benzodiazepine receptor: association with the voltage-dependent anion channel and the adenine nucleotide carrier. Proc. Natl. Acad. Sci. USA 89, 3170–3174[Abstract/Free Full Text]
26. Zoratti, M., and Szabò, I. (1994) Electrophysiology of the inner mitochondrial membrane. J. Bioenerg. Biomembr. 26, 543–553[Medline]
27. Rostovtseva, T., and Colombini, M. (1997) VDAC channels mediate and gate the flow of ATP: implications for the regulation of mitochondrial function. Biophys. J. 72, 1954–1962[Medline]
28. Thinnes, F. P. (1992) Evidence for extramitochondrial localization of VDAC/porin channel in eukaryotic cells. J. Bioenerg. Biomembr. 24, 71–75[Medline]
29. Lisanti, M. P., Scherer, P. E., Vidugiriene, J., Tang, Z., Hermanowsky-Vosatka, A., Tu, Y.-H., Cook, R. F., and Sargiacomo, M. (1994) Characterization of caveolin-rich membrane domains isolated from an endothelial-rich source: implication for human disease. J.Cell. Biol. 126, 111–126[Abstract/Free Full Text]
30. Zamecnik, P., Aghajanian, J., Zamecnik, M., Goodchild, J., and Witman, G. (1994) Electron micrographic studies of transport of oligodeoxynucletides across eukaryotic cell membranes. Proc. Natl. Acad. Sci. USA 91, 3156–3160[Abstract/Free Full Text]
31. De Pinto, V., Prezioso, G., and Palmieri, F. (1987) A simple and rapid method for the purification of the mitochondrial porin from mammalian tissues. Biochim. Biophys. Acta 905, 499–502[Medline]
32. De Pinto, V., Benz, R., and Palmieri, F. (1989) Interaction of non-classical detergents with the mitochondrial porin. A new purification procedure and characterization of the pore-forming unit. Eur. J. Biochem. 183, 179–187[Medline]
33. De Pinto, V., Thinnes, F., Link, T., and Palmieri, F. (1991) Peptide-specific antibodies and proteases as probes of the transmembrane topography of the bovine heart mitochondrial porin. Biochemistry 30, 10191–10200[Medline]
34. Perosa, F., Carbone, R., Ferrone, S., and Dammaco, F. (1990) Purification of human immunoglobulins by sequential precipitation with caprylic acid and ammonium sulphate. J. Immunol. Methods 74, 385–394
35. Roa, M., and Scandella, D. (1976) Multiple steps during interaction between coliphage lambda and its receptor protein in vitro. Virology 72, 182–194[Medline]
36. Francis, G., Brennan, L., and Ferenci, T. (1991) Affinity chromatographic purification of cysteine-substituted maltoporin variants: thiol reactivity and cross-linking in an outer membrane protein of E. coli. Biochim. Biophys. Acta 1067, 89–96[Medline]
37. Rizzuto, R., Brini, M., Bastianutto, C., Marsault, R., and Pozzan, T. (1995) Photoprotein-mediated measurement of calcium ion concentration in mitochondria of living cells. Methods Enzymol. 260, 417–428[Medline]
38. Brini, M., Marsault, R., Bastianutto, C., Alvarez, J., Pozzan, T., and Rizzuto, R. (1995) Transfected aequorin in the measurement of cytosolic Ca²⁺ concentration ([Ca^2+]_c). A critical evaluation. J. Biol. Chem. 270, 9896–9903[Abstract/Free Full Text]
39. Alvarez, O. (1986) How to set up a bilayer system. In Ion Channel Reconstitution (Miller, C., ed) pp. 115–130, Plenum Press, New York
40. Schirmer, T., Keller, T. A., Wang, Y.-F., and Rosenbusch, J. P. (1995) Structural basis for sugar translocation through maltoporin channels at 3.1 Å resolution. Science 267, 512–514[Abstract/Free Full Text]
41. Meyer, J. E. W., Hofnung, M., and Schulz, G. E. (1997) Structure of maltoporin from Salmonella typhimurium ligated with a nitrophenyl maltotrioside J. Mol. Biol. 266, 761–775[Medline]
42. Dermietzel, R., Hwang, T. K., Buettner, R., Hofer, A., Dotzler, E., Kremer, M., Deutzmann, R., Thinnes, F. P., Fishman, G. I., Spray, D. C., and Siemen, D. (1994) Cloning and in situ localization of a brain-derived porin that constitutes a large-conductance anion channel in astrocytic plasma membranes. Proc. Natl. Acad. Sci. USA 91, 499–503[Abstract/Free Full Text]
43. Benz, R., Maier, E., Thinnes, F. P., Gotz, H., and Hilschmann, N. (1992) studies on human porin. VII. The channel properties of the human B-lymphocyte membrane-derived `porin 31 HL' are similar to those of mitochondrial porin. Biol. Chem. Hoppe-Seyler 373, 295–303[Medline]
44. Song, L. Z., Hobaugh, M. R., Shustak, C., Cheley, S., Bayley, H., and Gouaux, J. E. (1996) Structure of staphylococcal alpha-hemolysin, a heptameric transmembrane pore. Science 274, 1859–1866[Abstract/Free Full Text]
45. Korchev, Y. E., Bashford, C. L., Adler, G. M., Kasianowicz, J. J., and Pasternak, C. A. (1995) Low conductance states of a single ion channel are not `closed'. J. Membr. Biol. 147, 233–239[Medline]
46. Song, J., and Colombini, M. (1996) Indications of a common folding pattern for VDAC channels from all sources. J. Bioenerg. Biomembr. 28, 153–161[Medline]
47. Rauch, G., and Moran, O. (1994) On the structure of mitochondrial porin and its homologies with bacterial porins. Biochem. Biophys. Res. Commun. 200, 908–915[Medline]
48. Koebnik, R., and Braun, V. (1993) Insertion derivatives containing segments up to 16 amino acids identify surface- and periplasm-exposed regions of the FhuA outer membrane receptor of Escherichia coli K-12. J. Bacteriol. 175, 826–839[Abstract/Free Full Text]
49. Bezrukov, S. M., Vodyanoy, I., and Parsegian, V. A. (1994) Counting polymers moving through a single ion channel. Nature (London) 370, 279–281[Medline]
50. Brown, M. D., and Wallace, D. C. (1994) Molecular basis of mitochondrial DNA disease. J. Bioenerg. Biomembr. 26, 273–289[Medline]
51. Wallace, D. C., Shoffner, J. M., Trounce, I., Brown, M. D., Ballinger, S. W., Corral-Debrinski, M., Horton, T., Jun, A. S., and Lott, M. T. (1995) Mitochondrial DNA mutations in human degenerative diseases and aging. Biochim. Biophys. Acta 1271, 141–151[Medline]
52. Linnane, A. W., Marzuki, S., Ozawa, T., and Tanaka, M. (1989) Mitochondrial DNA mutations as an important contributor to aging and degenerative diseases. Lancet 1, 642–645[Medline]
53. Ozawa, T. (1995) Mechanism of somatic mitochondrial DNA mutations associated with age and diseases. Biochim. Biophys. Acta 1271, 177–189[Medline]
54. Lander, E. S., and Lodish, H. (1990) Mitochondrial diseases: gene mapping and gene therapy. Cell 61, 925–926[Medline]
55. McEnery, M. W., Snowman, A. M., Trifiletti, R. R., and Snyder, S. H. (1991) Isolation of the mitochondrial benzodiazepine receptor: association with the voltage-dependent anion channel and the adenine nucleotide carrier. Proc. Natl. Acad. Sci. USA 89, 3170–3174
56. Taketani, S., Kohno, H., Furukawa, T., and Tokunaga, R. (1995) Involvement of peripheral-type benzodiazepine receptors in the intracellular transport of heme and porphyrins. J. Biochem. Tokyo 117, 875–880[Abstract/Free Full Text]
57. Zoratti, M., and Szabò, I. (1995) The mitochondrial permeability transition. Biochim. Biophys. Acta 1241, 139–176[Medline]