MOLECULAR GENETICS AND METABOLISM 63, 239–242 (1998)ARTICLE NO. GM972671

BRIEF COMMUNICATION

Studies on Human Porin XVI: Polyamines Reducethe Voltage Dependence of Human VDAC in PlanarLipid Bilayers—Spermine and Spermidine Inducing

Asymmetric Voltage Gating on the Channel

the channel. Correspondingly, it was demonstratedWe present the first data on the effects of poly- that albumin loaded with SDS works as a modulatoramines on human VDAC. Purified VDAC in lipid bi-

in the same sense (13). It then may be basically thelayers shows lowered voltage dependence wheneverpresentation of negatively charged groups in a spe-putrescine, cadaverine, spermine, spermidine, orcific distance or conformation that defines VDACthe histamine releaser Compound 48/80, respec-modulators showing increasing effects on the voltagetively, are applied. Only spermine and spermidine

induce an asymmetric reaction on the channel. dependence of the channel. From this point of viewHowever, we state a groupwise different reaction of biogenic, negatively charged, low-molecular-weightpolyamines, which are organic polycations, on hu- molecules interacting with varying carriers mayman porin. q 1998 Academic Press form part of channel closing modulators, regulated

Key Words: porin; VDAC; channel regulation; poly- themselves by the turnover of both constituents.amines; compound 48/80. As antagonists, low-molecular-weight molecules

showing positive charges in physiological situations,either on their own or attached to different carriers,There is increasing evidence of the extramitochon-may work on eukaryotic porin in the opposite direc-drial expression of eukaryotic porin or VDAC (volt-tion. Here, we measured the effects of four biogenicage-dependent anion-selective channel) (1–8). Thispolyamines—putrescine, cadaverine, spermine, andmakes regulation of the channel a key problem (9–spermidine, and also a synthetic polyamine Com-11). The voltage dependence of VDAC is known to bepound 48/80—on the channel characteristics ofsensitive to synthetic low-molecular-weight agonistshighly purified human type-1 porin integrated into(9–11) or to biogenic modulators (10,11), respec-artificial planar lipid bilayers. Polyamines are poly-tively. Endogenous VDAC modulators as they werecations that are protonated at physiological pH val-prepared from the intermembrane space of mito-ues. In each case a decreasing effect on the voltagechondria (12), as well as corresponding factors founddependence of the channel was seen.in mammalian cytosol or amniotic fluid (13), were

shown to be inactivated by proteases. The modula-MATERIALS AND METHODStors, on this experimental basis, are thought to

be intermediate-molecular-weight range proteins(12,13). Channel active human type-1 porin was highly

enriched by two steps of ion exchange chromatogra-To look at the modulation of eukaryotic porin ina different way might be more adequate, at least phy in nonionic detergent (2). Black membrane mea-

surement procedures were essentially the same ascomplementary. Several synthetic polyanions, e.g.,Konig’s polyanion, dextrane sulfate, and polyaspar- given in a recent study on the channel properties of

human type-1 porin (14). Here, we always measuredtic acid (10,11), increase the voltage dependence of

2391096-7192/98 $25.00

Copyright q 1998 by Academic PressAll rights of reproduction in any form reserved.

AID MGM 2671 / af10$$$121 04-07-98 23:53:00 bmmas AP: MGM

240 BRIEF COMMUNICATION

in 1 M KCl, applying human porin and polyaminesin identical concentrations to both chambers of theblack membrane apparatus. Concerning biogenicamines we used 50 mM each, while Compound 48/80, which is a synthetic polyamine, was present in62.5 mg/ml. The pH value was always adjusted to 7.2.After black membranes were formed by diphytanoylphosphatidylcholine, insertion was run initially at amembrane potential of 10 mV for 20–30 min, re-sulting in porin saturated membranes. Then voltagedependence of the channels was documented for10-mV steps up to 80 mV, changing the orientationof the field applied at each voltage step. We did 13to 15 measurements for each of the five agonists.

RESULTS

Figure 1 (top) demonstrates that the voltage de-pendence of black membrane-integrated humantype-1 porin is decreased by the biogenic poly-amines putrescine or cadaverine and also by thesynthetic histamine releaser Compound 48/80. Adirect comparison is given to control measure-ments in the absence of any modulating polya-mine. There is only a weak effect of putrescinewhile cadaverine and Compound 48/80 inducemore clear cut effects on the channel. There is no FIG. 1. Plots comparing the decreasing effects of the biogenic

polyamines putrescine, cadaverine, spermine, spermidine, andindication of any asymmetric effect of either of thethe synthetic polyamine Compound 48/80 on the voltage depen-three polyamines on the channel.dence of highly enriched human type-1 porin in planar artificialFigure 1 (bottom) correspondingly summarizes lipid bilayers. Biogenic polyamines were applied 50 mM, Com-

our data on the influences of the biogenic polyamines pound 48/80 at 62.5 mg/ml in 1 M KCl. The lipid used was diphyta-spermine and spermidine on the voltage dependence noyl phosphatidylcholine. Porin as well as the agonists were sym-

metrically applied to both chambers of the black membrane appa-of black membrane-integrated human type-1 porin,ratus (see inset). The voltage ramps were run changing thethis again in direct comparison to a control. Here, aorientation of the field at each voltage step (10 mV). *, Cathodepronounced voltage dependence (sidedness) of the chamber; %, anode chamber; PA, polyamines; VDAC, human

modulating effect of either polyamine (i.e., induction type-1 voltage-dependent anion-selective channel. (Top) Decreas-of rectification on the channel) is observed, this in ing effects of putrescine, cadaverine, and Compound 48/80 on

the voltage dependence of human type-1 porin are symmetric.spite of a symmetric application of the channel and(Bottom) Decreasing effects of spermine and spermidine on theof the modulators (see Fig. 1, inset).voltage dependence of black membrane-integrated human Type-1 porin are asymmetric.

DISCUSSION

Using a standard protocol, we show for the first ment for the first time a corresponding symmetriceffect of the histamine releaser Compound 48/80,time that the voltage dependence of a eukaryotic

type-1 porin is decreased by defined biogenic low- which is a synthetic polyamine, on the channel. Wedid not find any indication of altered channel conduc-molecular-weight modulators: the polyamines pu-

trescine, cadaverine, spermine, and spermidine. Pu- tance of human porin induced by the polyaminestested (data not shown). However, we demonstratetrescine and cadaverine work on the channel in a

symmetric way, while spermine and spermidine a groupwise different reaction of VDAC with poly-amines, which under physiological conditions areshow a highly voltage dependent (i.e., rectifying) ef-

fect on human type-1 VDAC. Furthermore, we docu- polycations that are protonated. Up to now, studies

AID MGM 2671 / af10$$$122 04-07-98 23:53:00 bmmas AP: MGM

241BRIEF COMMUNICATION

channel obtained from Paramecium mitochondria. J Membron VDAC regulation focused on anionic modulatorsBiol 30:99–120, 1976.of the channel, this group always inducing an in-

2. Thinnes FP, Gotz H, Kayser H, Benz R, Schmidt WE, Krat-crease of the voltage dependence on the channel.zin HD, Hilschmann N. Zur Kenntnis der Porine des

The concentration of the biogenic polyamines used Menschen I. Reinigung eines Porins aus menschlichen B-in our experiments was chosen under the influence Lymphozyten (Porin 31HL) und sein topochemischer Nach-

weis auf dem Plasmalemm der Herkunftszelle. Biol Chemof early studies on the effect of polyamines on bacte-Hoppe-Seyler 370:1253–1264, 1989.rial porin (15). However, biogenic polyamines often

3. Lisanti MP, Scherer PE, Vidugiriene J, Tang ZL, Herma-change amount with the cell compartment studied,nowski-Vosatka A, Tu Y-H, Cook RF, Sargiacomo M. Char-sometimes reaching low millimolar values. Further-acterization of Caveolin-rich membrane domains isolated

more, preincubation of the VDAC preparation used from an endothel-rich source: Implications for human dis-with 50 mM spermidine induced a corresponding ease. J Cell Biol 126:111–126, 1994.asymmetry on the gating of the channel (data not 4. Jakob C, Gotz H, Hellmann T, Hellmann KP, Reymann S,

Florke H, Thinnes FP, Hilschmann N. Studies on humanshown). We used Compound 48/80 in a concentrationporin: XIII. The Type-1 VDAC ‘‘Porin 31HL’’ biotinylated atonly about six times that applied in experimentsthe plasmalemma of Trypanblue excluding human B lym-with living cells.phocytes. FEBS Lett 368:5–9, 1995.

The differences concerning symmetric or asym- 5. Junankar PR, Dulhunty AF, Curtis SM, Pace SM, Thinnesmetric effects of the polyamines studied can be inter- FP. Porin-type 1 proteins in sarcoplasmic reticulum andpreted by the assumption that putrescine, cadav- plasmalemma of striated muscle fibres. J Muscle Res Cell

Motil 16:595–610, 1995.erine and Compound 48/80 work on both gating cen-6. Shoshan-Barmatz V, Hadad N, Feng W, Shafir I, Orr I, Var-ters of the channel (10,11), while spermine and

sanyi M, Heilmeyer LMG. VDAC/porin is present in sarco-spermidine react only with one of them. This hypoth-plasmic reticulum from skeletal muscle. FEBS Lett 386:esis should be testable by a site-directed mutagene- 205–210, 1996.

sis approach. The asymmetry in channel gating ob- 7. Reymann S, Florke H, Heiden M, Jakob C, Stadtmuller U,served furthermore indicates that channel insertion Steinacker P, Lalk VE, Pardowitz I, Thinnes FP. Furtheris always the same direction when native human evidence for multi-topological localization of mammalian

porin (VDAC) in the plasmalemma forming part of a chloridetype-1 porin spontaneously enters artificial lipid bi-channel complex affected in cystic fibrosis and encephalomy-layers, in agreement with recent data given by differ-opathy. Biochem Mol Med 54:75–87, 1995.ent laboratories (16–18).

8. Thinnes FP, Reymann S. New findings concerning verte-There is increasing evidence of the relevance of brate porin. Naturwissenschaften 84:480–498, 1997.

biogenic polyamines for the regulation of ion chan- 9. Thinnes FP, Schmid A, Benz R, Hilschmann N. Studies onnels. The corresponding data suggest that actually human porin III. Does the voltage-dependent anion channel

‘‘Porin 31HL’’ form part of the chloride channel complex,observed channel properties may often result fromwhich is observed in different cells and thought to be affectedendogenously acting polyamine modulators and mayin cystic fibrosis? Biol Chem Hoppe-Seyler 371:1047–1050,not be intrinsic to channel proteins (19–22). The1990.

data presented here, to our knowledge, are the first10. Benz R. Permeation of hydrophilic solutes through mito-

results on interactions of polyamines and eukaryotic chondrial outer membranes: review on mitochondrial porins.anion channels. Biochim Biophys Acta 1197:167–196, 1994.

11. Colombini M. Anion channels in the mitochondrial outerACKNOWLEDGMENTS membrane. Chloride Channels, Guest Editor William B.

Guggino. Curr Top Memb 42:73–101, 1994.We thank Prof. Dr. R. Benz and Dr. D. Gauss for advice concern-12. Liu MY, Torgrimson A, Colombini M. Characterization anding our experiments and Profs. Dr. N. Hilschmann and Dr. H.

partial purification of the VDAC-channel-modulating pro-Gotz for critically reading the paper.tein from calf liver mitochondria. Biochim Biophys Acta

Note added in proof. There is increasing evidence that plas- 1185:203–212, 1994.malemma-integrated VDAC is part of the ubiquitous volume-sen- 13. Heiden M, Kroll K, Thinnes FP, Hilschmann N. Proteins ofsitive anion channel complex (7,8). Meanwhile, we have first data cytosol and amniotic fluid increase the voltage dependenceindicating that the regulatory volume decrease (RVD) of Jurkat of human type-1 porin. J Bioenerg Biomembr 28:171–180,cells is influenced by incubating the cells with a-methylornithine, 1996.an inhibitor of the ornithine decarboxylase.

14. Benz R, Maier E, Thinnes FP, Gotz H, Hilschmann N. Stud-ies on human porin VII. The channel properties of the hu-REFERENCESman B-lymphocyte membrane-derived ‘‘Porin 31HL’’ aresimilar to those of mitochondrial porins. Biol Chem Hoppe-1. Schein SJ, Colombini M, Finkelstein A. Reconstitution in

planar lipid bilayers of a voltage-dependent anion-selective Seyler 373:295–303, 1992.

AID MGM 2671 / af10$$$122 04-07-98 23:53:00 bmmas AP: MGM

242 BRIEF COMMUNICATION

15. Delcour AH. MiniReview. Function and modulation of bacte- 21. Johnson TD. Modulation of channel function by polyamines.Trends Pharm Sci 17:22–27, 1996.rial porins: insights from electrophysiology. FEMS Microbiol

Lett 151:115–123, 1997. 22. Williams K. Review article, Interactions of polyamines withion channels. Biochem J 325:289–297, 1997.16. Elkeles A, Breiman A, Zizi M. Functional differences among

wheat voltage-dependent anion channel (VDAC) isoformsexpressed in yeast. J Biol Chem 272:6252–6260, 1997. Andrea Horn

17. Zizi M, Thomas L, Blachly-Dyson E, Forte M, Colombini M. Susanne ReymannOriented channel insertion reveals the motion of a trans- Friedrich P. Thinnes1membrane beta strand during voltage gating of VDAC. J

Department of ImmunochemistryMembr Biol 144:121–129, 1995.Max-Planck-Institute for Experimental Medicine

18. Xu X, Colombini M. Self-catalyzed insertion of proteins into Hermann-Rein-Strasse 3phospholipid membranes. J Biol Chem 271:23675–23682, D-37075 Gottingen1996. Germany

19. Scott RH, Sutton KG, Dolphin AC. Interactions of poly- Received November 13, 1997amines with neuronal ion channels. Trends NeuroSci 16:153–160, 1993.

1 To whom correspondence should be addressed. Fax: 49-551-20. Forsythe ID. Ion channels: A physiological function for poly-amines? Curr Biol 5:1248–1251, 1995. 3899-500. E-mail: thinnes@exmdi1.mpiem.gwdg.de.

AID MGM 2671 / af10$$$122 04-07-98 23:53:00 bmmas AP: MGM