J. Cell Sci. 25, 157-161 (1977) 157 Printed in Great Britain PORE-LIKE STRUCTURES IN BIOLOGICAL MEMBRANES L. ORCI, A. PERRELET, FRANCINE MALAISSE-LAGAE AND P. VASSALLI* Institute of Histology and Embryology, and *Department of Pathology, University of Geneva Medical School, Geneva, Switzerland SUMMARY In frccze-fracture replicas, biological membranes appear as smooth surfaces interrupted by random globular protrusions, the intramembrane particles. Smooth areas correspond to the membrane phospholipidic domain, while intramembrane particles are the morphological counterpart of membrane proteins. In the present work, examination of membranes in a variety of cell types reveals that a number of intramembrane particles contain an electron-dense spot. The spot is thought to correspond to a minute pit in the particle, filled by the platinum used in the freeze-fracture procedure. Similar images, described previously in intramembrane particles forming the specific array of the gap junction, were interpreted as hydrophilic channels bridging the interior and the exterior of the plasma membrane. Comparison between the gap junction particles and the non-junctional particles containing a dense spot suggests that these latter may too contain hydrophilic channels. The channels in random intramembrane particles would represent the morphological counterparts of the water-filled pores described in models of mem- brane permeability. INTRODUCTION The concept of pores in biological membranes was postulated to explain the diffusion of hydrophilic molecules across the membrane's lipid bilayer (Solomon, 1968; Woodbury, 1965). However, in spite of the fact that models of diffusion across bio- logical membranes imply the presence of numerous pores, their morphological counterparts are lacking. The only exception described so far concerns the gap junction or nexus. This specialized area of the cell membrane represents an ionic and metabolic communication pathway between the cytoplasm of two adjacent cells (McNutt & Weinstein, 1973; Payton, Bennett & Pappas, 1969) and appears mor- phologically as an aggregate of subunits bridging the plasma membranes (Kreutziger, 1968; McNutt & Weinstein, 1970, 1973). Thin-section (Revel & Karnovsky, 1967) and freeze-fracture (McNutt & Weinstein, 1970) electron microscopy have shown that the subunits contain minute circular densities which were interpreted as portions of hydrophilic channels piercing the centre of each subunit (McNutt & Weinstein, 1970, 1973). In freeze-fracture, gap junction subunits appear as regularly sized and closely packed intramembrane particles on the cytoplasmic leaflet of the membrane. As reported in the present paper, a re-examination of fracture faces in a wide variety of plasma membranes indicates that a sizable number of non-junctional intramembrane particles show densities at their top. Each density would represent the filling of a hole or channel in the non-junctional particles. II CEL 25
Transcript
J. Cell Sci. 25, 157-161 (1977) 157
Printed in Great Britain
PORE-LIKE STRUCTURES IN BIOLOGICAL
MEMBRANES
L. ORCI, A. PERRELET, FRANCINE MALAISSE-LAGAEAND P. VASSALLI*Institute of Histology and Embryology, and *Department of Pathology, University ofGeneva Medical School, Geneva, Switzerland
SUMMARY
In frccze-fracture replicas, biological membranes appear as smooth surfaces interrupted byrandom globular protrusions, the intramembrane particles. Smooth areas correspond to themembrane phospholipidic domain, while intramembrane particles are the morphologicalcounterpart of membrane proteins. In the present work, examination of membranes in a varietyof cell types reveals that a number of intramembrane particles contain an electron-dense spot.The spot is thought to correspond to a minute pit in the particle, filled by the platinum used inthe freeze-fracture procedure. Similar images, described previously in intramembrane particlesforming the specific array of the gap junction, were interpreted as hydrophilic channels bridgingthe interior and the exterior of the plasma membrane. Comparison between the gap junctionparticles and the non-junctional particles containing a dense spot suggests that these latter maytoo contain hydrophilic channels. The channels in random intramembrane particles wouldrepresent the morphological counterparts of the water-filled pores described in models of mem-brane permeability.
INTRODUCTION
The concept of pores in biological membranes was postulated to explain the diffusionof hydrophilic molecules across the membrane's lipid bilayer (Solomon, 1968;Woodbury, 1965). However, in spite of the fact that models of diffusion across bio-logical membranes imply the presence of numerous pores, their morphologicalcounterparts are lacking. The only exception described so far concerns the gapjunction or nexus. This specialized area of the cell membrane represents an ionic andmetabolic communication pathway between the cytoplasm of two adjacent cells(McNutt & Weinstein, 1973; Payton, Bennett & Pappas, 1969) and appears mor-phologically as an aggregate of subunits bridging the plasma membranes (Kreutziger,1968; McNutt & Weinstein, 1970, 1973). Thin-section (Revel & Karnovsky, 1967)and freeze-fracture (McNutt & Weinstein, 1970) electron microscopy have shown thatthe subunits contain minute circular densities which were interpreted as portions ofhydrophilic channels piercing the centre of each subunit (McNutt & Weinstein, 1970,1973). In freeze-fracture, gap junction subunits appear as regularly sized and closelypacked intramembrane particles on the cytoplasmic leaflet of the membrane. Asreported in the present paper, a re-examination of fracture faces in a wide variety ofplasma membranes indicates that a sizable number of non-junctional intramembraneparticles show densities at their top. Each density would represent the filling of a holeor channel in the non-junctional particles.
II CEL 25
158 L. Orci, A. Perrelet, F. Malaisse-Lagae and P. Vassalli
MATERIALS AND METHODS
Pieces of liver, adrenal, endocrine pancreas and bladder of laboratory animals (mostlyrodents) were used, as well as pellets of isolated mouse myeloma cells (MOPC 315), rabbiterythrocytes and mastocytes. Samples were fixed in phosphate-buffered glutaraldehyde forvarious periods of time, soaked for 0-5-2 h in 30 % phosphate-buffered glycerol and frozen inFreon 22 cooled with liquid nitrogen. Frozen tissue was freeze-fractured according to Moor &Miihlethaler (1963). Platinum-carbon replicas were observed in a Philips EM 300 electron-microscope operating at 80 kV with initial magnifications between 40000 and 60000 times.
RESULTS
As indicated in Materials and methods, samples of freeze-fractured membranes arederived from a wide variety of cells and tissues. Since the general morphology ofmembrane fracture faces is not - at least with the available material - cell- or tissue-specific, the following description applies to all material examined. As amply docu-mented in previous studies, the freeze-fracture appearance of biological membranesis that of smooth surfaces interrupted at random by globular protrusions, the intra-membrane particles (Branton, 1969, 1971)- The smooth areas correspond to themembrane phospholipidic domain, while the particles have been shown to correspondto membrane proteins (Singer, 1974; Singer & Nicolson, 1972; Vail, Papahadjopoulos& Moscarello, 1974). During freeze-fracturing, the membrane is split in 2 halvesalong the middle of the phospholipid bilayer, so that each leaflet of the membraneyields the above-described picture. By convention, the freeze-fractured inner leafletof the membrane is called the P-face, while the freeze-fractured outer leaflet is calledthe E-face (Branton et al. 1975). A prerequisite for the study of the fine structure ofparticles is the use of suitably high magnification. An initial magnification in theelectron microscope of x 40000, which can be further enlarged photographically by3-5 times, is adequate in most cases. In such conditions, examination of fracture facesof both plasma and/or intracellular membranes of all cell types studied in the presentwork revealed that certain intramembrane particles had a tiny black spot at their top(Fig. 1 A-K). The particles showing a black spot were not otherwise different from theremaining intramembrane particles, as far as their size and shape were concerned.The relative frequency of the spotted particles in P-fracture faces could be estimatedroughly to be 15-30 % of all intramembrane particles present in this face (data pooledfrom all the membrane types examined). No attempt was performed at a quantitativecorrelation between a given membrane type and its content of spotted particles. Thesize of the dense spot on the particles was found to vary from approximately 2 to 6 nm.The position of the dense spot on the particles seemed also subject to variation. Somedense spots appeared not to be situated at the top of the particle, but were placedsideways. In plasma membranes carrying gap junctions (Fig. 1 K) we could confirmthe previous finding of McNutt & Weinstein (1970) that many of the gap junctionparticles do show a small density at their top. As noted by these authors, randomparticles with a dense spot could be observed outside the junctional area as well (seeFig. IK) .
Pore-like structures in membranes 159
Fig. 1. Examples of membrane fracture faces (P- or E-faces) showing intramembraneparticles with pits. A pit is represented by a small dense spot present at the top of theparticle; some of the pits are outlined by white circles. Since all figures were arrangedwith the source of platinum evaporation at the lower part of the picture (shadows, inwhite, are upwards), one sees that pits have various orientations with respect to themembrane plane. In all figures, the bar represents 20 nm. A, Mouse myeloma cell:plasma membrane (P-face). x 300000. B, Mouse myeloma cell: plasma membrane (E-face). x 280000. c, Rabbit mastocyte: plasma membrane (P-face). x 300000. D, Toadbladder cell: plasma membrane (P-face). x 270000. E, Endothelial cell from the ratadrenal gland: plasma membrane (P-face). x 340000. F, Rat liver (hepatocyte): mem-brane from the endoplasmic reticulum (P-face). x 320000. G, Rabbit erythrocyte:plasma membrane (P-face). x 300000. H, Fish islet cell: plasma membrane (P-face).x 340000. 1, Rat islet cell: membrane from a secretory granule (P-face). x 360000.
j , Rat islet cell: plasma membrane (E-face). x 220000. K, Plasma membrane from apancreatic islet cell (P-face). This area of the membrane contains an aggregate ofclosely packed intramembrane particles (gap junction) as well as randomly scatteredparticles. Several aggregated and dispersed particles show distinct pits at their top.X 350000.
160 L. Orci, A. Perrelet, F. Malaisse-Lagae and P. Vassalli
DISCUSSION
The observations reported above suggest the existence, in freeze-fractured mem-branes, of a distinct population of intramembrane particles characterized by thepresence of a dense spot at their top. Given the conditions of oblique platinumshadowing necessary to obtain a freeze-fracture replica (Moor & Muhlethaler, 1963),electron-dense areas on positive prints invariably indicate an accumulation of plati-num. Practically, in the case of a single direction of platinum shadowing, an accumu-lation of this metal will occur'either on objects raised above the fracture plane (attheir side facing the source), or at the bottom of pits or cavities extending below thefracture plane. Distinction between raised objects and cavities is usually easy, sincethe former project a shadow while the latter do not. According to this reasoning, thedense spots situated on the top of intramembrane particles (these latter are raisedstructures projecting shadows) represent most likely minute pits filled by platinum.In view of the similarity of appearance between the pitted particles and gap junctionsubunits, it is tempting to suggest that the former may contain hydrophilic channels(pores) as well. If this is true, one might ask whether the random particles with pitsdo not simply represent free-floating gap junction particles, given the fluidity of themembrane (Singer, 1974; Singer & Nicolson, 1972) and the fact that gap junctions aresubmitted to assembly and disassembly (Decker & Friend, 1974), which may occurthrough aggregation or disaggregation of preformed subunits (particles). Althoughthis possibility cannot be excluded at present, the fact that pitted particles are presentin red blood cells, mastocytes, myeloma cells and leucocytes as well as in sperm cells(Friend & Fawcett, 1974), all of which are free cells which do not share gap junctions,renders it less likely.
In summary, the data indicate (a) that the presence of densities at the top of intra-membrane particles is in favour of the presence of channels (pores) in these particles;and (b) that the channels may represent the pores assumed to account for the diffusionof hydrophilic molecules across biological membranes.
This work was supported by grants no 3.553.7s and 3.2920.74 from the Swiss NationalScience Foundation.
REFERENCESBRANTON, D. (1969). Membrane structure. A. Rev. PL Physiol. 20, 209-238.BRANTON, D. (1971). Freeze-etching studies of membrane structure. Phil. Trans. R. Soc. Ser.
B 261, 133-138.BRANTON, D., BULLIVANT, S., GILULA, N. B., KARNOVSKY, M. J., MOOR, H., MUHLETHALER, K.,
NORTHCOTE, D. H., PACKER, L., SATIR, B., SATIR, P., SPETH, V., STAEHELIN, L. A., STEERER. L. & WEINSTEIN, R. S. (1975). Freeze-etching nomenclature. Science, N.Y. 190, 54-56.
DECKER, R. S. & FRIEND, D. S. (1974). Assembly of gap junctions during amphibian neuru-lation. J. Cell Biol. 62, 32-47.
FRIEND, D. S. & FAWCETT, D. W. (1974). Membrane differentiation in freeze-fracturedmammalian sperm. J. Cell Biol. 63, 641-664.
KREUTZICER, G. O. (1968). Freeze-etching of intercellular junctions in mouse liver. In Proc.26th Meet. Electron Micros. Soc. America, pp. 234-235. Baton Rouge, La.: Claitor's Publish-ing Division.
Pore-like structures in membranes 161
McNurr, N. S. & WEINSTEIN, R. S. (1970). The ultrastructure of the nexus. A correlated thin-section and freeze-cleave study. J. Cell Biol. 47, 666-688.
MCNUTT, N. S. & WEINSTEIN, R. S. (1973). Membrane ultrastructure at mammalian inter-cellular junctions. Progr. Biophys. molec. Biol. 26, 45-101.
MOOR, H. & MUHLETHALER, K. (1963). Fine structure in frozen-etched yeast cells. J. Cell Biol.17, 609-649.
PAYTON, B. W., BENNETT, M. V. L.& PAPPAS, G. (i969).Permeabilityandstructureof junctionalmembranes at an electrotonic synapse. Science, N.Y. 166, 1641-1643.
REVEL, J. P. & KARNOVSKY, M. J. (1967). Hexagonal array of subunits in intercellular junctionsof the mouse heart and liver. J. Cell Biol. 23, C7-C12.
SINGER, S. J. (1974). The molecular organization of membranes. A. Rev. Biochem. 43, 805-833.SINGER, S. J & NICOLSON, G. L. (1972). The fluid mosaic model of the structure of cell mem-
branes. Science, N.Y. 175, 720-731.SOLOMON, A. K. (1968). Characterization of biological membranes by equivalent pores. J. gen.
Physiol. 51, 335S-364S.VAIL, W. J., PAPAHADJOPOULOS, D. & MOSCARELLO, M. A. (1974). Interaction of a hydro-
phobic protein with liposomes: evidence for particles seen in freeze-fracture as being pro-teins. Biochim. biophys. Ada 345, 463-467.
WOODBURY, J. W. (1965). The cell membrane: ionic and potential gradients and active trans-port. In: Physiology and Biophysics, igth edn (ed. Th. C. Ruch & H. D. Patton), pp. 1-25.Philadelphia: Saunders.
