Freeze-fracture replicas of the plasma membrane of unfixed, uncryoprotected Paramecium tetraurelia bear large rectilinear arrays of 11 nm particles arranged in 7–11 parallel rows. The arrays are of sufficient size to leave impressions in replicas of the underlying outer alveolar membrane, and are apparent as parallel ridges in replicas of the surface coat of deep-etched cells.

By noting the location of arrays in replicas of identified portions of the cortex of P. tetraurelia, it has been possible to map the distribution of arrays over the cell surface. The arrays are found primarily over the anterior surfaces of the cell, covering an area that extends from the preoral suture over the left adoral field and a large portion of the anterior dorsal surface.

Freeze-fracture analyses of cells taken from a number of different stages of a culture cycle suggest that the particle arrays are not replicated as an integral part of the cortex during cell division, but are assembled and oriented in the membrane as the cells mature. The appearance of small intramembranous particle complexes in the plasma membrane of cells in logarithmic growth phase supports this hypothesis, possibly representing an assembly stage in the formation of the larger particle arrays.

The facts that the particle arrays are apparent in replicas of the surface coat of cells, are found primarily at the anterior of the cell body, and have a highly specific orientation with respect to the cell surface, strongly suggest that they function as chemoreceptors in P. tetraurelia.

In recent years a variety of protozoa have been the subjects of freeze-fracture investigations and they have revealed a range of different intramembranous particle (IMP) complexes. The ciliate protozoa have proved particularly fruitful in this respect. Following an exhaustive study of 68 protozoan genera, Bardele (1981) described several types of ciliary membrane specialization, including particle rosettes, plaques, strands and necklaces. In two examples, the entire ciliary membrane was covered with orthogonally arranged particles. Some of these structures have been described previously, but by cataloguing the nature and origin of the specializations, Bardele was able to suggest that the different classes of arrays might be used to determine the phylogenetic relationships among ciliate protozoa.

Despite these prolonged and often repeated studies, the functions of the different particle assemblies remain largely unknown. Exceptions are the rings and rosettes of particles seen in replicas of the plasma membrane of Paramecium tetraurelia, now known to be trichocyst-docking and membrane-fusion sites (Plattner et al. 1973; Olbricht et al. 1984), and ciliary plaques, three rows of particles that lie adjacent to each of the nine peripheral microtubule doublets at the ciliary base. Although first observed in Tetrahymena (Wunderlich & Speth, 1972), plaques are common to a number of ciliate genera (Bardele, 1981), including Paramecium. Since voltagesensitive calcium channels are known to be located in the ciliary membrane of Paramecium (Dunlap, 1977), and the ciliary base exhibits enhanced sensitivity to calcium (Hamasaki & Naitoh, 1985), it has been suggested that the particle plaques represent calcium gates. This contention is supported by the finding that paramecia fixed in the presence of high concentrations of external calcium show cationic deposits in the plaque region (Plattner, 1975). If the plaques do represent calcium channels, mutants that are defective in calcium gating might be expected to show reduced numbers of plaque particles. Such mutants show apparently normal ciliary plaques, however (Bardele, 1981): the only cell line known to have abnormal or missing plaques is‘paranoiac’ (Byrne & Byrne, 1978), a sodium channel mutant of P. tetraurelia (Saimi & Kung, 1980). Thus the function of the ciliary plaques remains unclear.

Ordered arrays of particles are not unique to the ciliary membranes of protozoa. Allen (19786), Hufnagel (19816) and Bardele (1983) have described large rectilinear particle arrays in freeze-fracture replicas of the plasma membrane of three ciliate genera. Since these are restricted to the anterior of the cell body, it has been suggested that they might have a sensory function. The present study was undertaken to search for similar arrays of particles in freeze-fracture replicas of P. tetraurelia, to map their distribution over the surface of the cell, and to examine the possibility of their involvement in chemosensation.

Paramecium tetraurelia, stock 51s, was cultured on a rye-grass seed infusion, supplemented with 1 mmol I-1 CaCl2 and Imgl-1 stigmasterol. Cultures were inoculated with Aerobacter aerogenes 8–24 h before use and were maintained at 25°C.

Cells were harvested by centrifugation (150 g, 2min) and resuspended in 4mM-KCl, 1 mM-CaCl2, 1 mM-Tris-(hydroxymethyl)aminomethane buffer (pH 7·2). Ciliates were then pelleted by centrifugation and excess solution was removed from the pellet by suction. The specimens were layered between two 10·0 mm × 4·0 mm × 0·25 mm mica strips and immediately frozen by rapid immersion in a well of partly frozen Freon 22 cooled by a jacket of liquid nitrogen. The mica strips with enclosed cell suspension were loaded onto a monolayer freeze-fracture device (Balzers Union) under liquid nitrogen. This was modified (Newman, unpublished) to allow rapid attachment to the cold table of the freeze-fracture plant (Balzers 360M). Fracturing was achieved by parting the two strips, yielding a pair of complementary fracture faces. Fracturing occurred at a temperature of − 100°C under a vacuum of <6× 10−6 Torr (1 Torr ≈133·3 Pa), and was immediately followed by shadowing with platinum/carbon from an electron gun at an angle of 45 °, and then by deposition of a carbon backing layer. Etching, where applicable, was carried out at a temperature of–100°C for 1–2 min prior to shadowing.

Specimens were deciliated in preparation for the production of surface replicas by rapid agitation (2 min) in 5% (v/v) ethanol. The deciliated cells were then spread over single mica strips and frozen in freshly melted Freon 22. The strips were loaded onto the monolayer freeze-fracture device and placed in the freeze-fracture plant for deep-etching. Deep-etching was performed under a vacuum of <6×10−6 Torr at −90°C for a period of 40–60 min. After etching, the cells were shadowed at − 100°C with platinum/carbon and coated with carbon.

Replicas of both freeze-fractured and deep-etched cells were cleaned with IM-KOH (24 h), collected on uncoated, 300-mesh copper support grids, and examined using a Jeol JEM 100S electron microscope at an acceleration voltage of 80 kV.

General features of the cortex and cortical membranes of P. tetraurelia

The surface structures of both P. tetraurelia and Paramecium caudatum have been described in some detail (Janisch, 1972; Ehret & McArdle, 1974; Allen, 1978a). To summarize: the cortex of P. tetraurelia is organized into longitudinal rows of square or rectangular depressions called‘kinetosomal territories’ (Fig. 1). At the base of each depression, a cilium or ciliary pair emerges, and each territory is associated with an invagination of the plasma membrane called a‘parasomal sac’. The territories are bounded by ridges, comprising an underlying cytoplasmic fold and a vesicular alveolar system (Fig. 2). The alveolar sacs are enclosed by outer (OAM) and inner (IAM) alveolar membranes. Although continuous, the two membranes are morphologically distinct and together form an interconnected system that runs throughout the cortex of Paramecium (Allen, 1978a).

Fig. 1.

A freeze-fracture replica of the P-face of the plasma membrane of P. tetraurelia. The cortex comprises rows of square or rectangular depressions called kinetosomal territories. At the base of each territory an emergent cilium (c) and its associated parasomal sac (p) are seen. Trichocyst fusion sites are apparent as rings of IMPs on the crests of the cortical folds (t). ×29000. Shadowing direction is indicated in micrographs by an encircled arrow.

Fig. 1.

A freeze-fracture replica of the P-face of the plasma membrane of P. tetraurelia. The cortex comprises rows of square or rectangular depressions called kinetosomal territories. At the base of each territory an emergent cilium (c) and its associated parasomal sac (p) are seen. Trichocyst fusion sites are apparent as rings of IMPs on the crests of the cortical folds (t). ×29000. Shadowing direction is indicated in micrographs by an encircled arrow.

Fig. 2.

Diagrammatic representation of the surface membranes of P. tetraurelia.

Fig. 2.

Diagrammatic representation of the surface membranes of P. tetraurelia.

When the two mica strips with enclosed frozen suspension of ciliates are parted, the resultant fracture plane passes close to the surface of the upper strip and exposes large areas of the plasma membrane of those cells that contact the strip (Fig. 1). The P-face of the plasma membrane is characterized by a dense population of approximately 10–20nm intramembranous particles (IMPs). At the centre of those cortical folds lying perpendicular to the cell axis, concentric rings and rosettes of particles are observed, representing docking sites for attachment of underlying trichocysts (Plattner et al. 1973).

Rectilinear particle arrays

The P-face of the plasma membrane of unfixed, uncryoprotected P. tetraurelia bears large, rectilinear particle arrays (Fig. 3). These are located along the crests and sides of the cortical folds, and consist of several rows of ≈11 nm particles (measured perpendicular to the direction of shadowing). The rows are arranged in parallel, with an inter-row spacing of approximately 25 nm, and the particles within a row are tightly packed, with a centre-to-centre spacing of about 11 nm. The arrays usually consist of 7–11 rows, although there is considerable variation, which may be linked with culture age. Up to 18 rows have been recorded within a single complex (Fig. 3B). The length of the rows is also variable. Arrays in excess of 2μm spanning two adjacent kinetosomal territories have been observed, but more commonly their length is of the order of 1 μm.

Fig. 3.

Freeze-fracture replicas of the P-face of the plasma membrane bear rectilinear arrays of IMPs. Although the example shown in A contains only five particle rows, arrays containing as many as 18 rows have been observed (B). A, ×92500; B, ×88000.

Fig. 3.

Freeze-fracture replicas of the P-face of the plasma membrane bear rectilinear arrays of IMPs. Although the example shown in A contains only five particle rows, arrays containing as many as 18 rows have been observed (B). A, ×92500; B, ×88000.

Impressions of the rectilinear IMP arrays are present in replicas of the E-face of the plasma membrane (Fig. 4). E-face fractures exhibit characteristic furrows arranged in parallel, forming rectangular areas of similar size and with similar interrow spacing as the IMP arrays observed in P-face replicas of the plasma membrane.

Fig. 4.

The IMP arrays leave impressions in the unspecialized E-face of the plasma membrane. The parallel furrows are separated by a distance of 25 nm. ×138000.

Fig. 4.

The IMP arrays leave impressions in the unspecialized E-face of the plasma membrane. The parallel furrows are separated by a distance of 25 nm. ×138000.

Fractures through the outer alveolar membrane

Intact rectilinear particle arrays in P-face replicas of the plasma membrane are rare. The plane of fracture is thought to pass through the hydrophobic interior of membranes, following a line of least resistance (Branton et al. 1975). On encountering dense particle complexes, however, the fracture plane can frequently be deflected downward toward the outer alveolar membrane. A number of different combinations of fracturing faces can result from such a deviation (summarized in Fig. 5). An example of the commonest deviation is shown in Fig. 6 (see also Fig. 5E): the fracture plane has been deflected downward to the interior of the OAM prior to encountering a rectilinear IMP array, and has returned to the plasma membrane on reaching the far side of the particle complex. Thus the particle array has been ripped out of the plasma membrane on fracturing, taking with it the external leaflet of the OAM. The unspecialized E-face of the outer alveolar membrane can be seen through the rectangular‘scar’ in the P-face of the plasma membrane; parallel rows of shallow indentations in the OAM E-face reflect the existence of an overlying particle assembly prior to fracturing. Occasionally, the fracture plane exposes portions of an array in the plasma membrane before being deflected to the OAM (Fig. 7). Corresponding plasma membrane E-faces bear short parallel furrows with an attached block of membranous material, comprising a particle array and associated P-halves of the plasma and outer alveolar membranes (Fig. 8). In the example shown in Fig. 9 (see also Fig. 5G), particle rows can be seen through holes in the P-halves of the plasma membrane and the OAM, confirming the association of the rows with the plasma membrane E-face.

Fig. 5.

Diagrammatic representation of the surface membranes of P. tetraurelia and their fracturing characteristics.

A. The potential fracture planes of the plasma and outer alveolar membranes are denoted by broken lines through the interior of the membranes.

B. Etching a cell prior to replication reveals its external surface. Parallel ridges in the surface coat reflect the presence of an IMP array in the underlying plasma membrane.

C. A fracture through the plasma membrane reveals its P-face and an associated particle array.

D. The complementary E-face of the plasma membrane bears an impression of the IMP array in the form of parallel furrows.

E. Fracture through the OAM. The IMP arrays increase the resistance of the plasma membrane to fracturing. As a consequence, the cleavage plane is deflected towards the interior of the cell and passes beneath the array through the interior of the OAM. On reaching the far side of the array, the fracture plane returns to the plasma membrane and the IMP array is removed with the E-face of the plasma membrane on fracturing. Replicas of the P-face of the plasma membrane bear rectangular depressions, with the exposed E-face of the OAM retaining an impression of the particle arrays in the form of parallel indentations.

F. The complementary E-face of the plasma membrane bears a rectangular block of membranous material, comprising a particle array with associated P-faces of the plasma and outer alveolar membranes.

G. When pieces of the P-halves of the plasma and outer alveolar membranes are removed, the underlying particle rows are seen to be associated with the plasma membrane E-face.

H. An incomplete particle array on the P-face of the plasma membrane. The fracture plane has left the plasma membrane and passed on into the extracellular medium. On etching, the ridges of the surface coat are seen to be continuous with the particle rows.

I. Ridges of the etched surface coat are also seen to be continuous with the shallow indentations in the E-face of the outer alveolar membrane.

Fig. 5.

Diagrammatic representation of the surface membranes of P. tetraurelia and their fracturing characteristics.

A. The potential fracture planes of the plasma and outer alveolar membranes are denoted by broken lines through the interior of the membranes.

B. Etching a cell prior to replication reveals its external surface. Parallel ridges in the surface coat reflect the presence of an IMP array in the underlying plasma membrane.

C. A fracture through the plasma membrane reveals its P-face and an associated particle array.

D. The complementary E-face of the plasma membrane bears an impression of the IMP array in the form of parallel furrows.

E. Fracture through the OAM. The IMP arrays increase the resistance of the plasma membrane to fracturing. As a consequence, the cleavage plane is deflected towards the interior of the cell and passes beneath the array through the interior of the OAM. On reaching the far side of the array, the fracture plane returns to the plasma membrane and the IMP array is removed with the E-face of the plasma membrane on fracturing. Replicas of the P-face of the plasma membrane bear rectangular depressions, with the exposed E-face of the OAM retaining an impression of the particle arrays in the form of parallel indentations.

F. The complementary E-face of the plasma membrane bears a rectangular block of membranous material, comprising a particle array with associated P-faces of the plasma and outer alveolar membranes.

G. When pieces of the P-halves of the plasma and outer alveolar membranes are removed, the underlying particle rows are seen to be associated with the plasma membrane E-face.

H. An incomplete particle array on the P-face of the plasma membrane. The fracture plane has left the plasma membrane and passed on into the extracellular medium. On etching, the ridges of the surface coat are seen to be continuous with the particle rows.

I. Ridges of the etched surface coat are also seen to be continuous with the shallow indentations in the E-face of the outer alveolar membrane.

Fig. 6.

Impressions of the plasma membrane IMP arrays are also seen in replicas of the underlying OAM. P, plasma membrane P-face; Ea, E-face of the OAM. ×98000.

Fig. 6.

Impressions of the plasma membrane IMP arrays are also seen in replicas of the underlying OAM. P, plasma membrane P-face; Ea, E-face of the OAM. ×98000.

Fig. 7.

The fracture plane frequently exposes portions of an IMP array in the P-face of the plasma membrane (P) before being deflected downward to the OAM (Ea). ×84000.

Fig. 7.

The fracture plane frequently exposes portions of an IMP array in the P-face of the plasma membrane (P) before being deflected downward to the OAM (Ea). ×84000.

Fig. 8.

Corresponding plasma membrane E-face (E) replicas bear short furrows and a block of membranous material, comprising the P-half of the plasma membrane with associated particle array, and the P-face of the outer alveolar membrane (Pa). × 129000.

Fig. 8.

Corresponding plasma membrane E-face (E) replicas bear short furrows and a block of membranous material, comprising the P-half of the plasma membrane with associated particle array, and the P-face of the outer alveolar membrane (Pa). × 129000.

Fig. 9.

A freeze-fracture replica of the plasma membrane E-face (E) with an attached block of IMP array-membrane complex. Portions of the P-face of the plasma membrane and the OAM (Pa) have been lost on fracturing, exposing the underlying particle rows (see Fig. 5G). ×84000.

Fig. 9.

A freeze-fracture replica of the plasma membrane E-face (E) with an attached block of IMP array-membrane complex. Portions of the P-face of the plasma membrane and the OAM (Pa) have been lost on fracturing, exposing the underlying particle rows (see Fig. 5G). ×84000.

External manifestation of IMP arrays in the surface coat of etched cells

Hufnagel (1981a) reported that the plate-like arrays of intramembranous particles in the plasma membrane of Tetrahymena thermophila are apparent as ridges in replicas of the external surface of the cell. To determine whether the rectilinear particle complexes described in the present study are similarly manifest on the surface of P. tetraurelia, specimens were frozen, deep-etched and replicated. Deciliation of the cells prior to freezing permits unrestricted observation of the cell surface, but the surfaces of both deciliated and untreated ciliates show similar characteristics, suggesting that the observations reported below are not artifacts of the deciliation procedure.

The surface of P. tetraurelia, as revealed by deep etching, has a grainy, textured appearance without apparent specialization. However, certain regions of the cell surface bear rectilinear ridged units (Fig. 10), similar in size and appearance to the IMP arrays in the plasma membrane. The units lie along the crests and sides of the cortical folds, and adjacent ridges are separated by a gap of approximately 25 nm. To determine the relationship between the plasma membrane IMP arrays and the surface coat ridges, specimens were fractured and then etched briefly to expose the cell surface prior to replication. Examination of the resultant replica shows the exposed plasma membrane P-face to be contiguous with the surface coat. In regions of the cell where the fracture plane has exposed portions of a particle array before leaving the cell surface and traversing the extracellular space, the rows of particles are seen to coincide exactly with the ridges in the surface coat (Figs 11, 5H). Thus the ridges represent folds in the surface coat, caused by the underlying particle complexes. The surface coat ridges also coincide with the parallel depressions in the OAM (Fig. 12), confirming the relationship between the ridges in the surface of the cell and the plasma membrane particle complexes (see Fig. 5I).

Fig. 10.

The IMP arrays are apparent as a series of parallel ridges in replicas of the surface coat of deep-etched cells (arrowheads). ×43 000.

Fig. 10.

The IMP arrays are apparent as a series of parallel ridges in replicas of the surface coat of deep-etched cells (arrowheads). ×43 000.

Fig. 11.

Etching the plasma membrane P-face (P) prior to replication exposes the surface coat (sc) and shows that the surface coat ridges are caused by underlying particle rows (see Fig. 5H). ×144000.

Fig. 11.

Etching the plasma membrane P-face (P) prior to replication exposes the surface coat (sc) and shows that the surface coat ridges are caused by underlying particle rows (see Fig. 5H). ×144000.

Fig. 12.

Ridges in the surface coat (sc) are seen to be continuous with the rows of shallow indentations in the E-face of the outer alveolar membrane (Ea), confirming the relationship between the particle arrays, the depressions in the E-face of the OAM, and the ridges in the surface coat (Fig. 5I). ×93000.

Fig. 12.

Ridges in the surface coat (sc) are seen to be continuous with the rows of shallow indentations in the E-face of the outer alveolar membrane (Ea), confirming the relationship between the particle arrays, the depressions in the E-face of the OAM, and the ridges in the surface coat (Fig. 5I). ×93000.

Mapping and orientation of IMP arrays

Although much of the cortex of P. tetraurelia is covered by kinetosomal territories of similar size and shape, the cell surface possesses a number of unique features that are readily identified in freeze-fracture replicas of the plasma membrane. Knowledge of the relative locations of these features makes it possible to construct a map of the cortex, and to determine accurately the distribution of IMP arrays over the surface of the cell. From an examination of several hundred replicas of identified portions of the cortex, it is evident that the particle arrays have a restricted distribution in the plasma membrane and in most cases also have a specific orientation with respect to the kineties.

Fig. 13 is a diagrammatic representation of the surface of P. tetraurelia, the lines marking rows of kinetosomal territories, and each point representing an IMP array. The figure is the result of a cumulative study and is intended to provide an indication of the extent of array localization; it is not representative of the number of arrays possessed by a single cell. The rectilinear IMP complexes are located primarily over the anterior surfaces of the cell soma, extending from an area to the right of the preoral suture, over the left adoral field, the right lateral field of the anoral zone, and terminating 9–10 kineties into the left lateral field of the dorsal surface. Arrays were seldom found more than 1–2 kinetosomal territories posterior to the anterior contractile vacuole pore. After examining approximately 20 replicas of the membrane covering the posterior of the cell body, three arrays were found on the posterio-ventral surface, and a number observed in only two replicas of the posteriodorsal surface. The greatest density of IMP arrays was seen in replicas of the anterior pole of the cell.

Fig. 13.

Localization of IMP arrays on the surface of P. tetraurelia. The drawings are diagrammatic representations of the cortex of P. tetraurelia. Each point represents an IMP array whose location was accurately determined from replicas of identified portions of the plasma membrane. The figure is the result of a cumulative study of a number of individual replicas and is not representative of the number of arrays possessed by any one cell. Few arrays are found posterior to the anterior contractile vacuole pore (acvp).

Fig. 13.

Localization of IMP arrays on the surface of P. tetraurelia. The drawings are diagrammatic representations of the cortex of P. tetraurelia. Each point represents an IMP array whose location was accurately determined from replicas of identified portions of the plasma membrane. The figure is the result of a cumulative study of a number of individual replicas and is not representative of the number of arrays possessed by any one cell. Few arrays are found posterior to the anterior contractile vacuole pore (acvp).

The rectilinear IMP complexes lie along the sides and crests of any one of the four cortical folds that form the boundary of a kinetosomal territory in P. tetraurelia. They may lie along the summit of the folds, or they may be displaced from the crest by as much as 0·25 μm. Some arrays traverse two kinetosomal territories, and some cortical folds support two arrays, one on either side. However, after noting the orientation of arrays in a number of replicas of identified portions of the surface membrane of P. tetraurelia, it is apparent that ≈75–90 % of the arrays have identical locations on the cortical folds with respect to the long axis of the ciliate. Such arrays are located on the folds that run the length of the cell; and are displaced from the crest of the folds so as to face the external medium at an angle of approximately 45 ° to the cell surface. This displacement occurs in one direction only: when viewed from the cell posterior, the displacement is seen to be to the left. At the flattened anterior pole of the cell, the arrays are displaced so as to face the ventral surface.

Changes in IMP density and distribution with culture age

During the course of the studies described above, it became increasingly apparent that the likelihood of observing rectilinear IMP arrays in plasma membrane replicas is dependent on the age of the ciliate culture. In order to quantify these changes and to examine possible effects of culture age on IMP array structure, cell samples were taken for freeze-fracture replication on each of the first 6 days following inoculation of fresh culture medium with Paramecium.

The period of rapid cell division during the initial day in fresh culture medium was accompanied by the appearance of small rectilinear IMP arrays in the P-face of the plasma membrane of P. tetraurelia (Figs 14,15). These arrays are similar to the IMP arrays observed in replicas of mature cells, but are smaller, usually consisting of 3–4 rows of ≈11 nm particles with an inter-row spacing of about 24–28 nm. Each row possesses 6–14 tightly packed particles, so that the length of the small arrays was commonly of the order of 60–160nm. A notable feature of some of the small arrays was the presence of inter-row particles, forming diagonal links between the rows (Fig. 15A,B).

Fig. 14.

Replicas of logarithmic growth phase cells frequently show small IMP arrays, usually comprising two to seven rows of llnm particles with an inter-row spacing of 24–28nm. A, ×118500; B, ×180000; C, ×87500; D, ×83000.

Fig. 14.

Replicas of logarithmic growth phase cells frequently show small IMP arrays, usually comprising two to seven rows of llnm particles with an inter-row spacing of 24–28nm. A, ×118500; B, ×180000; C, ×87500; D, ×83000.

Fig. 15.

The rows within a small array are occasionally linked by inter-row particles (arrowheads in A,B). Small IMP arrays have been observed within kinetosomal territories, either at the base of territories next to an emergent cilium (C), or low on the cortical folds (D). A, × 116500; B, ×98500; C, ×59000; D, ×76500.

Fig. 15.

The rows within a small array are occasionally linked by inter-row particles (arrowheads in A,B). Small IMP arrays have been observed within kinetosomal territories, either at the base of territories next to an emergent cilium (C), or low on the cortical folds (D). A, × 116500; B, ×98500; C, ×59000; D, ×76500.

Culture-age-dependent changes in IMP array density were quantified by counting the number of arrays in a random selection of array-bearing replicas of ciliates fractured on each of 6 days after inoculating fresh growth medium with paramecia. This count included both complete and incomplete P-face IMP arrays, and also the number of exposed outer alveolar membrane E-faces bearing impressions of the P-face IMP arrays. The number of kinetosomal territories in each replica was also counted, so that IMP array densities may be expressed as the number of arrays per 100 territories. Fig. 16 shows that IMP array density gradually increases over the first 2 days of the culture, reaching a peak of 23·1 ±0·8 arrays/100 kinetosomal territories (mean ± S.E., n = 38) during day 3. No further increases in array density were observed during days 3–6 of the culture.

Fig. 16.

Culture-age-dependent changes in IMP array density. Samples of P. tetraurelia were fractured and replicated on each of the first 6 days of a culture cycle. IMP array density was estimated by counting the total number of arrays in a random selection of replicas from each sample and dividing this value by the total number of kinetosomal territories in the same replicas. Data points are the means ± S.E. of 15–38 determinations.

Fig. 16.

Culture-age-dependent changes in IMP array density. Samples of P. tetraurelia were fractured and replicated on each of the first 6 days of a culture cycle. IMP array density was estimated by counting the total number of arrays in a random selection of replicas from each sample and dividing this value by the total number of kinetosomal territories in the same replicas. Data points are the means ± S.E. of 15–38 determinations.

Since the IMP arrays have a restricted distribution in the surface membrane of P. tetraurelia, only a fraction of the total number of plasma membrane replicas examined bear P-face IMP arrays or impressions of such arrays in the E-face of the OAM. It is apparent that with increasing length of time in culture, there are marked changes in the relative numbers of array-bearing and array-free replicas, suggesting that there is a culture-age-dependent change in the spread of IMP arrays over the surface of the cells. These changes were analysed by counting the total number of array-bearing and array-free replicas of cells from each of the first 6 days in the life of the ciliate culture, and then the ratio of these numbers was plotted against culture age (Fig. 17). A chi-squared analysis shows that there is a significant increase in the ratio of array-bearing to array-free replicas during days 2–3 (P<0·01), and a significant decrease during days 3–4 (P<0·05). By counting the total number of arrays observed in replicas of the plasma membrane of P. tetraurelia and dividing this number by the total number of kinetosomal territories examined, a day 3 ciliate is estimated to possess 0·106 array/territory, or about one array for every nine kinetosomal territories examined.

Fig. 17.

Culture-age-dependent changes in the spread of IMP arrays over the surface of P. tetraurelia. Cell samples were fractured and replicated on each of the first 6 days of a culture cycle. A random selection of replicas from each of the six samples were then examined and the number of these replicas bearing IMP arrays was counted. These counts were divided by the number of replicas in the same selection that were not associated with IMP complexes. The resultant ratio (array-bearing/array-free replicas) provides an indication of the spread of IMP arrays over the cell surface.

Fig. 17.

Culture-age-dependent changes in the spread of IMP arrays over the surface of P. tetraurelia. Cell samples were fractured and replicated on each of the first 6 days of a culture cycle. A random selection of replicas from each of the six samples were then examined and the number of these replicas bearing IMP arrays was counted. These counts were divided by the number of replicas in the same selection that were not associated with IMP complexes. The resultant ratio (array-bearing/array-free replicas) provides an indication of the spread of IMP arrays over the cell surface.

While the studies presented above provide the first detailed description of rectilinear IMP arrays in the plasma membrane of P. tetraurelia, such particle complexes are not unique to this organism. Indeed, recent reports suggest that large rectilinear plaques of IMPs may be a common feature of the ciliate protozoa (Hufnagel, 19816; Bardele, 1983). The first report of rectilinear particle complexes in a protozoan membrane was provided by Allen (1976, 1978a,b) following a freeze-fracture investigation of the surface membranes of glutaraldehyde-fixed and cryoprotected P. caudatum. These particle plates, although larger, are otherwise identical to those described in the present study; rectilinear plaques of llnm particles with a centre-to-centre spacing of 11 nm and inter-row spacing of 25 nm, lying along the cortical folds in the anterior region of the cell body. Hufnagel (1980, 19816), Hufnagel & Lyon (1980) and Bardele (1983) have since undertaken freeze-fracture examinations of the plasma membranes of Tetrahymena pyriformis and Cyclidium glaucoma, respectively. Both authors report rectangular particle assemblies similar in appearance to those of Paramecium, but in both examples each row within an array comprises three subrows. In Tetrahymena, all three rows contain 3–4nm particles, while in Cyclidium, a central row of 8–9nm particles is flanked by rows of smaller (5–6 nm) IMPs. Bardele (1983) also reported similar particle plates in more than 20 ciliate species, and in all cases, the plates are found on the anterioventral surfaces the cells. While the relationship between the rectilinear particle complexes of Paramecium and the triplet-type plates of other ciliates has not been determined, the restricted distribution of both types of particle array may be indicative of a common function.

Fracturing characteristics of the IMP arrays and their relationship with the outer alveolar membrane

Since a plane of cleavage through frozen tissue follows lines of least resistance (usually the hydrophobic interior of membranes), examination of the behaviour of the IMP arrays on fracturing can yield information about the relative tenacity with which they are held by the two halves of the membrane bilayer. In the few examples in which complete IMP arrays have been exposed on the P-face of the plasma membrane by fracturing, complementary E-faces show only parallel furrows; the lack of associated particles suggests that the particles within an array are firmly anchored within the inner half of the plasma membrane. However, the fact that the arrays are manifest as ridges in the surface coat of the cell suggests that the particles traverse the entire width of the membrane and that a portion protrudes from the membrane surface, perhaps reflecting a functional requirement for contact with the external medium.

With increasing IMP density, the two halves of the plasma membrane become so tightly bonded together that the fracture plane is deflected downwards to the outer alveolar membrane. The presence of indentations in the exposed E-face of the OAM suggests that the arrays are located in regions where the plasma membrane and OAM are closely associated. In Tetrahymena, the distance between the two membranes is maintained at a constant 15–20 nm (Franke et al. 1971; Satir & Wissig, 1982) by densely staining, tubular struts. No such structures exist between the plasma and outer alveolar membranes in Paramecium, so that the inter-membrane distance varies from 15 to 50 nm. This gap must be considerably reduced in the regions of the IMP arrays for them to leave an impression on the OAM; thus it is puzzling that, despite an extensive search, no such close association could be found in ultrathin sections through the anterior of P. tetraurelia (unpublished results). Perhaps treating cells with fixatives and dehydrating agents in preparation for thin-section electron microscopy causes a dissociation of the plasma and outer alveolar membranes that form a close association in their natural state. The alternative explanation, that the two membranes are forced together during freezing, seems unlikely in the apparent absence of ice-crystal formation.

Culture-age-dependent changes in IMP array frequency and distribution

The number of IMP arrays possessed by P. tetraurelia is culture-age-dependent, significant increases in array density being observed during the logarithmic growth phase of the culture cycle (days 1–3). The rapid increase in cell numbers during this period is effected by binary fission. Although ovoid in shape, the resultant daughter cells possess essentially a full complement of kinetosomal territories (see Kaneda & Hanson, 1974), so that the lengthening and thinning that occurs during the postfission period results from expansion and contraction of the territories, not from duplication. Thus, the fact that IMP array density gradually increases as the rate of cell division decreases suggests that the arrays are not replicated as an integral part of the territories, but rather are assembled and inserted into the plasma membrane as the cell matures. This may explain the appearance of the small IMP arrays during the first 2 days of the culture, representing embryonic forms of the larger arrays to be later assembled and oriented along the cortical folds (Fig. 15D). There have been insufficient observations from which to draw firm conclusions about the relationship of the small IMP arrays to the larger forms, but it is apparent that they lack the specific orientation characteristic of the large arrays, and are occasionally found at the base of kinetosomal territories (Fig. 15C). This is one of the few regions of the cortex in which the plasma membrane is in direct contact with the cytoplasm without intervening alveolar sacs (Fig. 2), and may be the point at which the particle arrays are inserted into the plasma membrane following partial assembly in the cytoplasm.

When the relative numbers of array-bearing and array-free replicas are considered, it is apparent that there is a significant increase and subsequent decrease in the number of replicas bearing particle arrays during the initial days in the life of the culture (Fig. 17). This observation suggests that in addition to increases in array density in localized regions of the cell body of P. tetraurelia, there is also an expansion and subsequent contraction of the area of specialization during days 2–4 of the culture cycle.

Possible functions of the IMP arrays

The rectilinear IMP arrays are prominent features of the cortex of P. tetraurelia. Each array represents an aggregate of 1000 or more particles within an area of 0·3 μm2. A ciliate from a 3-day-old culture can be expected to bear approximately 400 of these arrays, a total estimated particle number of 400 000. However, if the particles observed in freeze-fracture replicas represent protein aggregates (see Zingsheim & Plattner, 1976), the actual number of functional macromolecules within the arrays may be several times that estimated from a count of particle numbers.

In previous studies (Allen, 1978b; Hufnagel, 1981b; Bardele, 1983) three possible functions for the IMP arrays have been suggested, on the basis of their restricted distribution over the surface of ciliates. Hufnagel (19816) speculated that two intramembranous particle plates, one on either side of a cytoplasmic ridge, may be linked via cytoskeletal connectives to help maintain the extreme topography of the anterior of Tetrahymena. It is difficult to assign a similar structural role to the IMP arrays of P. tetraurelia. Extensive examination of the subpellicular cytoplasm in regions of the cell where IMP arrays might be expected to occur has failed to reveal fibrous or cytoskeletal connectives of the type described by Hufnagel (1981b). Moreover, the size and shape of the pellicular folds remains fairly constant throughout the cortex of P. tetraurelia; if the IMP arrays had a structural role, they would not.be restricted to the anterior of the cell.

Secondly, Allen (1978b) suggested that the rectilinear IMP arrays seen in freezefracture replicas of P. caudatum may be involved in inter-cell fusion during mating in this unicell. This event is also a property of the anterior of Paramecium, but occurs at specific sites on the cell surface, an area that includes several kineties either side of the pre-oral suture and an area to the right of the post-oral suture (Watanabe, 1978). Inspection of Fig. 13 shows that IMP arrays are seldom found on the posterioventral surfaces of P. tetraurelia. The highest density of IMP arrays occurs to the left of the pre-oral suture and over the anteriodorsal surfaces, areas not involved in conjugation. Thus it is unlikely that the particle arrays seen in replicas of the plasma membrane of P. tetraurelia are related to mating in this ciliate.

A third function is suggested by the fact that the rectilinear particle arrays are manifest in the surface coat of P. tetraurelia, in contact with the external medium. This suggests that they have a sensory role. Paramecia are sensitive to at least two kinds of external stimuli: mechanical and chemical. Mechanical stimulation of the anterior of Paramecium triggers a characteristic‘avoidance reaction’; the cell swims backwards for a short period, turns and then resumes forward swimming (Jennings, 1906). Since the rectilinear IMP arrays are restricted to the anterior surfaces of Paramecium, Allen (1978b) suggested that they may be mechanoreceptors, triggering avoidance responses when cells collide with solid objects. Mechanoreception is not an exclusive property of the anterior of Paramecium, however. If a cell is stimulated posteriorly, it swims faster forwards for a short distance. Transduction of both anterior and posterior mechanostimulation is effected by changes in the membrane potential of Paramecium (Eckert et al. 1972; Naitoh & Eckert, 1973). Anterior stimulation triggers depolarization and reversed ciliary beating, while deformation of the cell posterior elicits a hyperpolarization and increased ciliary beat frequency. This observation questions the proposed mechanoreceptive function of the IMP arrays. If the complexes are sensitive to mechanical stimuli, they might be expected to mediate both membrane depolarization and hyperpolarization, the polarity of the potential change being a property of ion channels associated with the receptors. In addition, Ogura & Machemer (1980) showed that application of localized mechanical stimuli elicited responses from both ventral and dorsal surfaces of P. caudatum, yet Allen (1978b) reported that replicas of the dorsal surface of the cell are seldom seen to bear rectilinear particle arrays. If the IMP arrays were mechanoreceptive, they should be equally distributed over both ventral and dorsal surfaces.

The fact that the chemosensory properties of Paramecium may be restricted to the anterior of the cell was suggested by the early studies of Alverdes (1922). It has since been demonstrated that (1) chemoreception is a property of the cell body membrane (Preston, 1983), and (2) there is an anterior-posterior gradation of chemosensitivity in P. tetraurelia (Preston & Van Houten, unpublished results). The anterior of the cell body would provide an ideal location for a chemoreceptor complex, a position that would enable the cell to sample the chemical composition of its environment on first contact. This is especially important for rapid responses to noxious and possibly toxic chemicals; the cell could stop within 2μm of penetrating a repellent area (Roberts, 1981) and reverse before cellular damage occurs.

At the level of the kinetosomal territories, chemoreceptors would be expected to be situated on the crests of the cortical folds, fully exposed to the flow of medium over the cell surface. As previously described, the IMP arrays are indeed located on the crests of the cortical folds, and the majority are oriented parallel to the long axis of the cell and displaced from the crests so as to face the external medium at an angle of approximately 45° to the cell surface. The significance of this specific orientation is apparent when one considers that Paramecium rotates anticlockwise as it swims forwards. The IMP arrays are orientated so as to face full into the flow of bathing medium as the ciliate rotates along its spiral path.

The IMP arrays in the plasma membrane of P. tetraurelia bear a remarkable structural similarity to paracrystalline particle complexes associated with the post-synaptic region of the neuromuscular junction of a variety of invertebrates. Rosenbluth (1978) reported rectilinear arrays of IMPs in replicas of the post-synaptic membrane of the longitudinal musculature of an earthworm. The 7–15 nm particles are packed into parallel rows with an inter-row spacing of 17 nm. Since the arrays are found on the post-synaptic membrane, they are presumed to represent acetycholine receptors. Similarly, Lane (1979) reported rectilinear arrays of 10 nm particles in the neuropile of the moth Manduca sexta. The arrays are associated with axon terminals, again suggesting a chemoreceptive function. Rheuben & Reese (1978) have speculated that the plaques of 9nm particles in replicas of the post-synaptic membrane of moth flight muscles represent L-glutamate receptors. Although of unknown function, particle arrays similar to those of P. tetraurelia are observed in freeze-fracture replicas of spider muscle (Franzini-Armstrong, 1979), while replicas of the extrajunctional sarcolemma of crayfish (Smith et al. 1979) and locust (Newman & Duce, 1982) muscle bear parallel rows of IMPs that are in many respects similar to the triplet-type IMP complexes reported in Tetrahymena (Hufnagel, 1981b) and Cyclidium (Bardele, 1983). Thus, it is interesting to note that the development of denervation-induced extrajunctional chemosensitivity in locust retractor unguis muscle is accompanied by the breakdown of the larger IMP‘bars’ and the widespread appearance of smaller arrays (Newman & Duce, 1982), a process reminiscent of the appearance of small IMP arrays in replicas of the plasma membrane of P. tetraurelia.

A specific function cannot be assigned to the IMP arrays in the surface membrane of P. tetraurelia purely on the basis of their structural similarity to other systems, but when considered together with the observations of their location on the cortex and their close association with the external surface of the ciliate, these similarities would strongly suggest a chemoreceptive function for the IMP arrays.

This work was supported by a Research Studentship from the Science and Engineering Research Council of Great Britain.

Allen
,
R. D.
(
1976
).
The mosaic nature of the plasma membrane of Paramecium
.
J. Protozool
.
23
,
10
1 IS
.
Allen
,
R. D.
(
1978a
).
Membranes of ciliates: ultrastructure, biochemistry and fusion
. In
Membrane Fusion
(ed.
G.
Poste
&
G. L.
Nicolson
), pp.
657
763
.
Amsterdam
:
Elsevier/North Holland Biomedical Press
.
Allen
,
R. D.
(
1978b
).
Particle arrays in the surface membrane of Paramecium: junctional and possible sensory sites
.
J. Ultrastruct. Res
.
63
,
64
78
.
Alverdes
,
F.
(
1922
).
Zur lokalisation des chemischen und thermischen sinnes bei Paramaecium und Stentor
.
Zool. Anz
.
55
,
19
28
.
Bárdele
,
C. F.
(
1981
).
Functional and phylogenetic aspects of the ciliary membrane: a comparative freeze-fracture study
.
Biosystems
14
,
403
421
.
Bárdele
,
C. F.
(
1983
).
Mapping of highly ordered membrane domains in the plasma membrane of the ciliate Cyclidium glaucoma
.
J. Cell Sci
.
61
,
1
30
.
Branton
,
D.
,
Bullivant
,
S.
,
Gilula
,
N. B.
,
Karnovsky
M. J.
,
Moor
,
H.
,
Kühlethaler
,
K.
,
Northcote
,
D. H.
,
Packer
,
L.
,
Satir
,
B.
,
Satir
,
P.
,
Speth
,
V.
,
Staehlin
,
L. A.
,
Steer
,
R. L.
&
Neinstein
,
R. S.
(
1975
).
Freeze-etching nomenclature
.
Science
190
,
54
56
.
Byrne
,
B. J.
&
Byrne
,
B. C.
(
1978
).
An ultrastructural correlate of the membrane mutant‘paranoiac’ in Paramecium
.
Science
199
,
1091
1093
.
Dunlap
,
K.
(
1977
).
Localization of calcium channels in Paramecium caudatum
.
J. Physiol
.
271
,
119
133
.
Eckert
,
R.
,
Naitoh
,
Y.
&
Friedman
,
K.
(
1972
).
Sensory mechanisms in Paramecium. I. Two components of the electrical response to mechanical stimulation of the anterior surface../
,
exp. Biol
.
56
,
683
694
.
Ehret
,
C. F.
&
Mcardle
,
E. W.
(
1974
).
The structure of Paramecium as viewed from its constituent levels of organization
. In
Paramecium: a Current Survey
(ed.
W. J.
Van Wagtendonk
), pp.
263
338
.
Amsterdam
:
Elsevier
.
Franke
,
W. W.
,
Kartenbeck
,
J.
,
Zentgraf
,
H.
,
Scheer
,
U.
&
Falk
,
H.
(
1971
).
Membrane-to- membrane cross bridges../
.
Cell Biol
.
51
,
881
888
.
Franzini-Armstrong
,
C.
(
1979
).
Aggregates of particles on the plasmalemma of striated muscle from a spider
.
Tissue & Cell
11
,
209
215
.
Hamasaki
,
T.
&
Naitoh
,
Y.
(
1985
).
Localization of calcium-sensitive reversal mechanism in a cilium of Paramecium
.
Proc. Japan Acad. Ser. B
61
,
140
143
.
Hufnagel
,
L. A.
(
1980
).
Oriented particle assemblies in the plasma membrane of Tetrahymena: their deployment relative to cell surface topography, cellular morphogenesis and sensitivity to stimuli
.
Biol. Bull. mar. biol. Lab., Woods Hole
159
,
471
472
.
Hufnagel
,
L. A.
(
1981a
).
External manifestation of plate-like particle arrays in the plasma membrane of Tetrahymena
.
Cell Biol. Int. Rep
.
5
,
581
586
.
Hufnagel
,
L. A.
(
1981b
).
Particle assemblies in the plasma membrane of Tetrahymena: relationship to cell surface topography and cellular morphogenesis. J’
.
Protozoal
.
28
,
192
203
.
Hufnagel
,
L. A.
&
Lyon
,
M.
(
1980
).
Plate-like particle arrays in the plasma membrane of Tetrahymena: their association with cell surface topography and alveoli
.
J. Protozool
.
27
,
23S
.
Janisch
,
R.
(
1972
).
Pellicle of Paramecium caudatum as revealed by freeze etching
.
J. Protozool
.
19
,
470
472
.
Jennings
,
H. S.
(
1906
).
Behavior of the Lower Organisms
.
New York
:
Columbia University Press
.
Kaneda
,
M.
&
Hanson
,
E. D.
(
1974
).
Growth patterns and morphogenetic events in the cell cycle of Paramecium aurelia
. In
Paramecium: a Current Survey
(ed.
W. J.
Van Wagtendonk
), pp.
219
262
.
Amsterdam
:
Elsevier
.
Lane
,
N. J.
(
1979
).
Intramembranous particles in the form of ridges, bracelets or assemblies in Arthropod tissues
.
Tissue & Cell
11
,
1
—18.
Naitoh
,
Y.
&
Eckert
,
R.
(
1973
).
Sensory mechanisms in Paramecium. II. Ionic basis of the hyperpolarizing mechanoreceptor potential../
,
exp. Biol
.
59
,
53
65
.
Newman
,
T. M.
&
Duce
,
I. R.
(
1982
).
Membrane specialisation of the extra-junctional sarcolemma of normal and denervated insect muscle
.
Cell Tiss. Res
.
223
,
179
185
.
Ogura
,
A.
&
Machemer
,
H.
(
1980
).
Distribution of the mechanoreceptor channels in the Paramecium surface membrane
.
J. comp. Physiol. A
135
,
233
242
.
Olbricht
,
K.
,
Plattner
,
H.
&
Matt
,
H.
(
1984
).
Synchronous exocytosis in Paramecium cells. II. Intramembranous changes analysed by freeze-fracturing
.
Expl Cell Res
.
151
,
14
20
.
Plattner
,
H.
(
1975
).
Ciliary granule plaques: membrane-intercalated particle aggregates associated with Ca2+-binding sites in Paramecium
.
J. Cell Sci
.
18
,
257
269
.
Plattner
,
H.
,
Miller
,
F.
&
Bachmann
,
L.
(
1973
).
Membrane specializations in the form of regular membrane-to-membrane attachment sites in Paramecium. A correlated freeze-etching and ultrathin-sectioning analysis. J’
.
Cell Sci
.
13
,
687
719
.
Preston
,
R. R.
(
1983
).
Studies on the responses of Paramecium tetraurelia to amino acids. Ph.D. thesis, University of Nottingham
.
Rheuben
,
M. B.
&
Reese
,
T. S.
(
1978
).
Three-dimensional structure and membrane specializations of moth excitatory neuromuscular synapse
.
J. Ultrastruct. Res
.
65
,
95
111
.
Roberts
,
A. M.
(
1981
).
Hydrodynamics of protozoan swimming
.
In Biochemistry and Physiology of Protozoa
, vol.
4
,
2
nd edn (ed.
M.
Levandowsky
&
S. H.
Hutner
), pp.
5
56
.
London
:
Academic Press
.
Rosenbluth
,
J.
(
1978
).
Particle arrays in earthworm postjunctional membranes
.
J. Cell Biol
.
76
,
76
86
.
Saimi
,
Y.
&
Kung
,
C.
(
1980
).
A Ca-induced Na current in Paramecium
.
J. exp. Biol
.
88
,
305
325
.
Satir
,
B. H.
&
Wissig
,
S. L.
(
1982
).
Alveolar sacs of Tetrahymena: ultrastructural characteristics and similarities to subsurface cisterns of muscle and nerve
.
J. Cell Sci
.
55
,
13
33
.
Smith
,
D. S.
,
Evoy
,
W. H.
&
Cayer
,
M. L.
(
1979
).
Freeze-fracture studies on the plasma membrane of crayfish skeletal muscle fibers
.
Biol. Cell
34
,
17
22
.
Watanabe
,
T.
(
1978
).
A scanning electron-microscopic study of the local degeneration of cilia during sexual reproduction in Paramecium
.
J. Cell Sci
.
32
,
55
66
.
Wunderlich
,
F.
&
Speth
,
V.
(
1972
).
Membranes in Tetrahymena. I. The cortical pattern
.
J. Ultrastruct. Res
.
41
,
258
269
.
Zingsheim
,
H. P.
&
Plattner
,
H.
(
1976
).
Electron microscopic methods in membrane biology
. In
Methods in Membrane Biology
, vol.
7
(ed.
E. D.
Korn
), pp.
1
146
.
New York
:
Plenum
.