Pflügers Arch - Eur J Physiol (2002) 444:286–290 DOI 10.1007/s00424-002-0813-1
O R I G I N A L A RT I C L E
Jose Omar Bustamante
Nuclear pore ion channel activity in live syncytial nuclei
Received: 17 November 2001 / Accepted: 28 January 2002 / Published online: 3 April 2002 © Springer-Verlag 2002
Abstract Nuclear pore complexes (NPCs) are important nanochannels for the control of gene activity and expression. Most of our knowledge of NPC function has been derived from isolated nuclei and permeabilized cells in cell lysates/extracts. Since recent patch-clamp work has challenged the dogma that NPCs are freely permeable to small particles, a preparation of isolated living nuclei in their native liquid environment was sought and found: the syncytial nuclei in the water of the coconut Cocos nucifera. These nuclei have all properties of NPC-mediated macromolecular transport (MMT) and express foreign green fluorescent protein (GFP) plasmids. They display chromatin movement, are created by particle aggregation or by division, can grow by throwing filaments to catch material, etc. This study shows, for the first time, that living NPCs engaged in MMT do not transport physiological ions – a phenomenon that explains observations of nucleocytoplasmic ion gradients. Since coconuts are inexpensive (less than US$1/nut per litre), this robust preparation may contribute to our understanding of NPCs and cell nucleus and to the development of biotechnologies for the production of DNA, RNA and proteins. Keywords Nuclear pore · Ion channel · Cell nucleus · Nuclear transport · Transcription-translation · Protein-RNA-DNA expression system
Introduction Nuclear pore complexes (NPCs) are supramolecular nanochannels located at the double membrane that deJ.O. Bustamante (✉) The Nuclear Physiology Lab-ITP-UNIT, The Millenium Institute of Nanosciences, Brazilian Ministry of Science and Technology, Praia Aruana, Aracaju, Sergipe 49037-610, Brazil e-mail:
[email protected] Tel.: +55-79-99884320, Fax: +1-209-8854343
limits the cell nucleus: the nuclear envelope (NE). For this reason, NPCs play an important role as regulators of the signals that control gene activity and expression (reviewed in [23]). Much of our knowledge of NPC function has been derived from preparations not in their natural state, e.g. isolated nuclei and permeabilized cells in cell lysates/extracts. Since 1990, patch-clamp investigations have challenged the established dogma that NPCs are freely permeable to physiological ions and other small particles [23]. Since I have been one of the opponents of this dogma (e.g. [6, 10]), I sought a natural preparation that would provide isolated and living nuclei floating in their native environment. Surprisingly, nature still offers systems that provide a syncytium of free-floating nuclei and that have not been exploited in nuclear transport research: insect embryos, trophoblasts, endosperms, etc. [4, 13, 14, 16, 24, 27, 28]. Of these, the non-committed nuclei of the coconut syncytial endosperm (i.e. those nuclei found in the coconut water) seemed the simplest alternative because of the large volume of liquid that can be extracted from each nut. My extensive tests on over 500 coconuts confirmed that this preparation had all the essential characteristics required for research in this field. The syncytial nuclei of the coconut water displayed all properties of NPC macromolecular transport. Furthermore, the preparation has important biotechnological applications because the coconut water expresses foreign plasmids. Many of the non-committed syncytial nuclei showed chromatin movement and it was possible to observe the creation of the syncytial nuclei by nuclear division or particle aggregation. They also could grow by projecting tentacle-like filaments to catch material. These and other interesting properties of coconut syncytial nuclei made them unique to substantiate their intact state. Furthermore, the inexpensive coconut water syncytium (less than US$1/nut per litre) expressed foreign plasmid DNA (e.g. that for the green fluorescent protein, GFP) and thus is a novel and inexpensive system for the study of the biology of the cell nucleus and for the biotechnology of proteins and RNA production.
287
The present paper is the first to report that living NPCs engaged in macromolecular transport do not transport physiological ions. This explains the common observations of nucleocytoplasmic ion gradients (see [7]). Syncytial NPCs also display ion channel activity whenever they are not transporting macromolecules.
Materials and methods With the exception of the details for the new preparation, the experiments followed procedures developed earlier [5, 10, 11, 12]. Briefly, the experiments consisted of monitoring with fluorescence microscopy the expression of enhanced GFP (EGFP) plasmids and the nuclear translocation of various fluorescent probes (e.g. [11, 12]). I also measured nucleocytoplasmic ion flow using the patchclamp technique [11, 12]. Some fluorescent and patch-clamp observations were made simultaneously. All experiments were conducted at room temperature (24–28 °C), a temperature that is agreeable with our local coconut trees.
Results and discussion I used freshly picked (5–10 min) young coconuts from my laboratory garden. Depending on their size, 100– 600 ml of liquid could be drawn from each coconut. This abundant volume was useful for conditioning all material (laboratory ware, syringes, bath, electrodes, etc.) in contact with the syncytium of nuclei uncommitted to the endosperm. The nuclear density (103–106/ml) was weather dependent. About 200 of the coconuts did not contain a significant density of intact, free-floating nuclei (500, ten coconuts). This plasmid drives GFP expression and does not enter the nucleus [15, 31]. In contrast, as Fig. 1a–c shows, adding 50–100 pM pEGFP-C1 (3.1 MDa, Clontech) resulted in all nuclei (dividing and non-dividing) displaying EGFP fluorescence after 5 h (n>200, ten coconuts). Only after prolonged exposure (>8 h) was there a small, but measurable, signal detectable outside the nucleus, in the syncytial fluid (ten coconuts). This signal could have resulted from EGFP released from dividing nuclei. The plasmid pEGFP-C1 contains the SV40 enhancer that allows transcription factors to attach and target it to the nucleus [15, 31]. I did not observe a fluorescence signal on adding pEGFP-C1 to nucleus-free syncytium, even
288
after 24 h exposure and up to 1 nM pEGFP-C1 (n>200, ten coconuts). I obtained the nucleus-free fluid by filtration of the coconut water through a sterile 0.2-µm Teflon syringe filter (Nalgene, 199-2020). That EGFP was expressed when I incubated the nuclei with pEGFP-C1 but not with pGFP∆SV40 indicates that the plasmids are imported into the nuclei in a sequence-specific manner as has been shown for mammalian cells [15, 31]. These observations suggest that, like prokaryotes, syncytial nuclei have all the machinery for protein synthesis. The plasmid-transporting capacity of the syncytial nuclei may be related to natural intrinsic mechanisms, because acridine orange fluorescence showed the whole syncytium to contain DNA and RNA (100% of the 12 coconuts tested). The capacity of the nuclei to import macromolecules was not confined to plasmids, because nuclear-targeted B-phycoerythrin (240 kDa) also enters the nucleus [10]. My observations on NPC-mediated ion flow also provide new insights on nuclear transport phenomena. Firstly, patch-clamp pipettes attached to non-dividing syncytial nuclei recorded no ion flow, even with potentials of up to +150 or –150 mV between the pipette and the bath. Several important conclusions can be drawn from this observation. First, non-committed and non-dividing nuclei readily allow the formation of tight seals (gigaseals) between the pipette tip and the nuclear surface membrane. In 50 nuclei of 50 coconuts, mean seal resistance was 17.6±4.2 GΩ (pooled values from experiments in which the pipette was filled with filtrate or saline and the bath with syncytium, filtrate or saline). Regardless of how these conditions were combined, a gigaseal was formed. Figure 1d shows the typical lack of ion flow observed during 2–24 h recordings when the nuclei were in their syncytium. The lack of ion flow does not support the idea of putative NPC peripheral channels (e.g. [18, 23, 26]). This, along with my observation 10 years ago of the lack of consistency between the eight-fold NPC symmetry and quantitative analyses of nuclear ion channel behaviour (e.g. [5]), indicates alternative explanations for findings purporting to support the existence of peripheral channels (e.g. incorrect interpretation of leakage in macroscopic technique – see [26]). My conclusion of the existence of only a single NPC transport channel is based on independent electron and fluorescence microscopy studies [17, 21]. Atomic force microscopy (AFM) observations of unplugged NPCs with transport substrates also suggest that NPCs use a single diffusional channel to transport ions [29]. Furthermore, the lack of ion flow also does not support the existence of any other type of channel in the outer nuclear membrane of these syncytial nuclei [23]. This is so because when the NPCs were plugged by macromolecular transport, the shortest pathway for ions to the bath electrode would have been through these putative ion channels and thus, ion current should have been recorded under patch-clamp conditions. Secondly, ion channel activity appeared and EGFP fluorescence disappeared after replacing the syncytium
or its filtrate with HEPES-buffered saline containing 1 µM CaCl2 plus the ATP-regenerating system [11], (6–10 h recordings, n=10, ten coconuts). The [Ca2+] used was in the order of estimates obtained with Oregon Green 488 (BAPTA-1 and Rhod-5N, Molecular Probes). Figure 1e shows that after 1 h of saline replacement, open-closed channel gating was detected. Thus, my experiments show that NPC ion transport was possible only when the natural conditions of the syncytium were replaced with saline. Under these unnatural conditions the ion conductance of the NPC channel was 335±42 pS (n=10, ten coconuts), in agreement with reported values for other preparations 10, [12]. Figure 1g shows that after 2 h saline exposure, channel closures were rarely seen and the number of ion-conducting channels reached a maximum of 11±3 (n=5, five coconuts). This number corresponds to that observed in other preparations with ~1-µm-diameter pipette tips [10]. The tight seal was maintained in these experiments and could be measured during simultaneous closing of the channels, which occurred with a probability of less than 0.1% (~104 records). The lack of channel opening under syncytium or filtrate is reminiscent of previous observations made using TnT (Promega), a commercial transcription-translation coupled medium [7, 10]. In the present experiments, however, the lack of channel opening persisted for as long as the nuclei were in their syncytium or the syncytial filtrate (monitoring for up to 24 h). The high throughput reported for NPC-mediated macromolecular transport [25] explains why, despite their large diameter, NPCs are capable of restricting nucleocytoplasmic transport of small ions and molecules. Since particles other than macromolecules (importins, exportins, etc.) shuttle between the two sides of the NPCs, they must ride “piggyback” on the larger macromolecules or have transport mechanisms and pathways not shared with physiological ions such as K+ (e.g. transporters, pumps), otherwise ions would flow through these pathways and current would be detected with the patch-clamp apparatus. Although I used a relative low gain of the patch-clamp amplifier (20 mV/pA) to record the large NPC ion flow, calculations showed that channels other than NPCs did not participate in the observed phenomena. This is because I could apply over 100 mV between the pipette and bath electrodes. Since a current of 5 pA could be detected with my patch-clamp settings, I estimate a maximal conductance of less than 50 pS for all non-NPC pathways. Clearly, even if these alternative pathways do exist, they are insignificant compared with the large pathway for the NPC channel. To check whether patch-clamp measurements of ion flow reflected the true state of NPCs, I tested the nuclear entry of fluoroisothiocyanate (FITC)-labelled dextrans and dendrimers [12]. In agreement with the patch-clamp observations, these fluorescent labels entered the nuclei only under saline conditions that facilitated ion channel activity. Dextrans and dendrimers entered the salinebathed non-dividing nuclei only when their size was smaller than 20 kDa and 10 nm, respectively. That is,
289
contrary to the present dogma of NPC transport, not even a 4-kDa dextran could enter the nucleus when the latter was intact and in its native environment. Dividing nuclei, however, allowed the passage of larger particles in either saline or syncytial filtrate. Although the cell biology of syncytial endosperm has progressed [4, 13, 14, 16, 24], only a handful papers have appeared on coconut syncytial nuclei. The first cell biology papers appeared in the 1950s (e.g. [13, 14]) and the first patch-clamp study in 1992 [22]. The latter inspired my investigation of syncytial nuclei (both plant and animal). The short life of those pioneer efforts may have resulted from the lack of commercialization of fresh coconut in developed nations and, thus, from the accompanying complicated logistics created by this commercial void. An example of this difficulty is the recent single trial in Germany with one batch of eight coconuts, after “express” transportation (1 week) plus over 1 week of sub-optimal storage. The AFM images showed the NE covered with many filaments – typical of nuclei already committed to endosperm formation (personal communication, H. Oberleithner, L. Albermann, I. Buchholz, C. Schäfer, V. Shahin). These results are consistent with my observations (more than 500 coconuts during 1999–2001) and sheds light on why I chose freshly isolated coconuts and the current location of my laboratory. The rationale for the present preparation derives from that employed for the development of equally challenging techniques [8, 9]. Briefly, fast but delicate processing of fresh material should secure a responsive preparation. I tried successfully nuts from other palms of the Arecaceae/Palmae family (e.g. Orbignya cohune). However, the easiest species to work with was our garden dwarf Cocos nucifera. The small size of the palms allowed easy hand picking of their normal-size nuts. Coconuts remained functional for up to 2 weeks after picking when kept at non-freezing temperatures. Furthermore, the syncytium recovers after freezing at –80 °C (100% of C. nucifera and O. cohune, six nuts each). This, along with the capacity for synthesizing proteins and nucleic acids, renders coconut water a very attractive alternative for cell/molecular biology and biotechnology of expression systems. My results demonstrate that live NPCs are plugged during their transport of DNA, RNA, transcription factors and other macromolecules. NPC plugging, in turn, causes nucleocytoplasmic electrical fields and concentration gradients (e.g. [1]). NPCs are therefore able to restrict the flow of K+, Na+, Ca2+ and other monoatomic ions. Figure 2 illustrates the simplest model explaining the observed results. Along with the molecular “Coultercounter principle” (e.g. [2, 3]), patch-clamp and fluorescence microscopy can be used to study the translocation of key molecules that control gene activity and expression. Electrically sensitive plugging by DNA, RNA and other macromolecules is a feature of nanochannels [19, 20, 30]. Syncytial NPCs may shed information that can
Fig. 2a, b The simplest model explaining the experimental observations. a Under normal conditions, the high throughput of NPCmediated macromolecular transport (MMT) prevents small particles (ions, dextrans, dendrimers, etc.) from translocating along the NPC channel. Both fluorescence microscopy and patch-clamp detect no translocation of the small particles and nucleocytoplasmic gradients of these particles develop. b In the presence of physiological saline, MMT ends due to lack of cargo macromolecules and essential substrates. Under these conditions, some NPCs are unplugged and others are plugged. Fluorescence microscopy and patch-clamp detect translocation of the small particles
be used in developing new strategies for cloning, gene manipulation and gene therapy. Acknowledgements I tender my thanks to Prof. Dr. Werner R. Loewenstein (Marine Biological Lab) for suggesting in 1995 the use of a simple preparation; to Drs. Marjori A. and Antonius J.M. Matkze (Institute of Molecular Biology of the Austrian Academy of Sciences) for information on their coconut water preparation; to Dr. David A. Dean (Northwestern University) for comments on plasmid transport mechanisms and for the gift of GFP plasmids; to Dr. Walter Stühmer, (Max-Planck Institute for Experimental Medicine) for constructive criticism of the gigaseal, patch-clamp; to Dr. Gregory Beckler (Promega) for discussions on TnT and other cell lysates and to Dr. Elen R.F. Michelette (The NPL) for discussions on insect syncytial nuclei. I am a Research Scholar of CNPq (MCT-Brazil) and currently supported through The Millenium Institute of Nanosciences-CNPq. Some of the chemicals, materials and equipment used in the study were contributed by Axon Instruments, Molecular Probes, Omega Optical, Roche-Boehringer and Sigma Chemical.
References 1. Badminton MN, Kendall JM, Rembold CM, Campbell AK (1998) Current evidence suggests independent regulation of nuclear calcium. Cell Calcium 23:79–86 2. Bezrukov SM (2000) Ion channels as molecular Coulter counters to probe metabolite transport. J Membr Biol 174:1–13 3. Bezrukov SM, Vodyanoy I, Parsegian VA (1994) Counting polymers moving through a single ion channel. Nature 370: 279–281 4. Boisnard-Lorig C, Colon CA, Bauch M, Hodge S, Doerner P, Bancharel E, Dumas C, Haseloff J, Berger F (2001) Dynamic analyses of the expression of the HISTONE::YFP fusion protein in Arabidopsis show that syncytial endosperm is divided in mitotic domains. Plant Cell 13:495–509 5. Bustamante JO (1992) Nuclear ion channels in cardiac myocytes. Pflugers Arch 421:473–485 6. Bustamante JO (1993) Restricted ion flow at the nuclear envelope of cardiac myocytes. Biophys J 64:1735–1749 7. Bustamante JO (1994) Nuclear electrophysiology. J Membr Biol 138:105–112 8. Bustamante JO, McDonald TF (1983) Sodium currents in segments of human heart cells. Science 220:320–321
290 9. Bustamante JO, Watanabe T, McDonald TF (1981) Single cells from adult mammalian heart: isolation procedure and preliminary electrophysiological studies. Can J Physiol Pharmacol 59:907–910 10. Bustamante JO, Hanover JA, Liepins A (1995) The nuclear pore ion channel activity. J Membr Biol 146:239–251 11. Bustamante JO, Michelette ERF, Geibel JP, Dean DA, Hanover JA, McDonnell TJ (2000) Calcium, ATP and nuclear pore channel gating. Pflugers Arch 439:433–444 12. Bustamante JO, Michelette ERF, Geibel JP, Hanover JA, McDonnell TJ, Dean DA (2000) Dendrimer-assisted patchclamp sizing of nuclear pores. Pflugers Arch 439:829–837 13. Cutter VM Jr, Freeman B (1952) Development of the syncytial endosperm of Cocos nucifera. Nature 173:827–828 14. Cutter VM Jr, Wilson KS, Dubé GR (1952) The isolation of living nuclei from the endosperm of Cocos nucifera. Science 115:58–59 15. Dean DA, Dean BS, Muller S, Smith LC (1999) Sequence requirements for plasmid nuclear import. Exp Cell Res 253:713– 722 16. Eckardt NA (2001) A brief tour of the cell cycle. Plant Cell 13:449–451 17. Feldherr CM, Akin D (1997) The location of the transport gate in the nuclear pore complex. J Cell Sci 110:3065–3070 18. Hinshaw JE, Carragher BO, Milligan RA (1992) Architecture and design of the nuclear pore complex. Cell 69:1133–1141 19. Howorka S, Cheley S, Bayley H (2001) Sequence-specific detection of individual DNA strands using engineered nanopores. Nat Biotechnol 19:636–639 20. Kasianowicz JJ, Brandin E, Branton D, Deamer DW (1996) Characterization of individual polynucleotide molecules using a membrane channel. Proc Natl Acad Sci USA 93:13770– 13773
21. Keminer O, Peters R (1999) Permeability of single nuclear pores. Biophys J 77:217–228 22. Matzke AJM, Behensky C, Weiger T, Matzke MA (1992) A large conductance ion channel in the nuclear envelope of a higher plant cell. FEBS Lett 32:81–83 23. Mazzanti M, Bustamante JO, Oberleithner H (2001) Electrical dimension of nuclear envelope. Physiol Rev 81:1–19 24. Otegui MS, Mastronarde DN, Kang BH, Bednarek SY, Staehelin LA (2001) Three-dimensional analysis of syncytial-type cell plates during endosperm cellularization visualized by high resolution electron tomography. Plant Cell 13:2033–2051 25. Ribbeck K, Görlich D (2001) Kinetic analysis of translocation through nuclear pore complexes. EMBO J 20:1320–1330 26. Shahin V, Danker T, Enss K, Ossig R, Oberleithner H (2001) Evidence for Ca2+- and ATP-sensitive peripheral channels in nuclear pore complexes. FASEB J 15:1895–1901 27. Simpson RA, Mayhew TM, Barnes PR (1992) From 13 weeks to term, the trophoblast of human placenta grows by the continuous recruitment of new proliferative units: a study of nuclear number using the dissector. Placenta 13:501–512 28. Stafstrom JP, Staehelin LA (1984) Dynamics of the nuclear envelope and of nuclear pore complexes during mitosis in the Drosophila embryo. Eur J Cell Biol 34:179–189 29. Stoffler D, Goldie KN, Aebi U (1999) Calcium-mediated structural changes of native nuclear pore complexes monitored by time-lapse atomic force microscopy. J Mol Biol 287:741–752 30. Szabò I, Bathori G, Tombola F, Brini M, Coppola A, Zoratti M (1997) DNA translocation across planar bilayers containing Bacillus subtilis ion channels. J Biol Chem 272:25275–25282 31. Wilson GL, Dean BS, Wang G, Dean DA (1999) Nuclear import of plasmid DNA in digitonin-permeabilized cells requires both cytoplasmic factors and specific DNA sequences. J Biol Chem 274:22025–22032
