Suspension Feeding in, Ciliated Protozoa: Structure and Function of

Arch. Protistenk. 123 (1980): 239-260

Department of Ecology and Genetics, University of Aarhus, Aarhus, Denmark

Suspension Feeding in Ciliated Protozoa: Structure and Function of Feeding Organelles By TOM

FENCHEL

With 16 Figures

Summary The function of mouth organelles in suspension feeding ciliates has been investigated. Ciliary membranelles propel water in a direction parallel and posterior to the individual membranelles (perpendicular to the membranellar band). In most oligohymenophores (hymenostomes and peri. trichs) the feeding currents are forced through the ciliary paroral membrane which acts as a sieve, retaining food particles. In polyhymenophores (spirotrichs), the membranellar band propels water out of the buccal cavity and the band itself also functions as a filter, retaining particles which are larger than the free space between two adjoining membranelles. Due to the torsion of the band, the retained particles are concentrated in the vicinity of the cytostome. The filtering activity of the mouth apparatus may concentrate suspended particles by a factor of several thousand; in the newly formed feeding vacuoles, ingested particles are fm-ther concen· trated by a factor of 3 - 4 due to the absorption of water from the vacuole. On the basis of simple physical considerations and observational data, it is concluded that most retention mechanisms previously suggested to be important for filter feeders cannot be of significance in the case of ciliates. Quantitative estimates of the amounts of water transported through the ciliary filters, in conjunction with data on the clearing rate for suspended latex beads show that foJ' a certain range of particle sizes, retention is nearly 100 %. Further, there is a correlation between the minimum particle size retained and the free space between adjoining cilia of the ciliary filters. It is therefore concluded, that in suspension feeding ciliates, particle retention takes place by a sieving mechanism. The evolution of suspension feeding in ciliates is discussed. Primitive ciliates are predominantly raptorial macrophages. The mouth of advanced (oligo- and polyhymenophoran) ciliates evolved as an adaptation to microphagy via some Vestibulifera-like ancestor. Many hymenostomes and spirotrichs have secondarily reverted to feed on large food particles, but they have remained suspension feeders and employ the same basic mechanisms for the concentration of food particles as do their microphagous relatives.

Introduction The feeding on particulate material by phagocytosis was probably among the first basic properties of the eukaryote cell and still the vast majority of heterotrophic eukaryotes make their living in this way. These forms show a variety of adaptations which serve to find or concentrate food particles. Species which feed on relatively small, freely suspended particles are called "suspension" or "filter feeders". They typically possess an apparatus which transports large volumes of water past structures which in some way are capable of retaining suspended food particles. This mode of 16

Arch. Protistenk. Ed. 123

240

T.FENCHEL

feeding is characterized bya high degree of automatization; that is, constant feeding rates and an inability to discriminate between different kinds of particles except for their mechanical properties (JORGENSEN 1966, 1980). Among ciliates, suspension feeding is widely distributed among the hymenostomes and the spirotrichs. The study of suspension feeding in ciliates is of interest from physiological, ecological as well as of evolutionary points of view. This paper is one in a series treating various aspects of this topic. FENCHEL (1980a, b) discusses food uptake as a function of environmental food particle concentration and particle size selection, and in relation to meta bolic demand and natural concentration of food particles. In FENCHEL (1980c), some general properties of suspension feeding are discussed, and finally FENCHEL and SMALL (1980) treat the membranelle function of Glaucoma in detail. This paper centers on the functional morphology of ciliate suspension feeding. This kind of feeding implies three different functions: 1) the generation of water currents which bring the organism in contact with large volumes of water; 2) some principle which allows the organism to remove suspended particles from the water; and 3) the phagocytosis of the concentrated particles. With respect to the two first mentioned components of filter feeding, our physiological understanding is still incomplete in the case of ciliates as well as for many other types of metazoan filter feeders (JORGENSEN 1980). In ciliates, the elaborate ciliary specializations of the mouth have long been subject to detailed morphological studies from the view·point of the systematist (CORLISS 1979), but with few exceptions (e. g., MACHEMER 1966, 1974; SLEIGH and AIELLO 1972; SLEIGH and BARLOW 1976) the functional aspects have been neglected. The mechanisms utilized for the retention of particles are still poorly understood and much literature on the subject is speculative and sometimes not in agreement with the basic laws of physics. One of the main objectives of this study is to contribute to a mechanistic description of particle retention in suspension feeders. In contrast to this, there are several thorough studies on phagocytosis in ciliates (e. g., MUELLER et al. 1965; NILSSON 1972; RASMUSSEN 1976) and this paper will only contribute little to this subject.

Material and Methods This study is based on seventeen species of ciliates. Glaucoma 8cintillans EHRB., Oolpidium campylum (STOKES) Paramecium caudatum EHRB., P. trichium (STOKES), Oyclidium sp., Lemba· dion sp. (most likely L. lucen8 (MASKELL)), Blephari8ma americanum SUZUKI, Bur8aria truncatella OFM, and the marine Euplote8 moebiu8i KAHL were isolated from ponds and streams or (in the case of Euplote8) in "Eel Pond" in the vicinity of Woods Hole, Mass., U. S. A. Oolpidium colpoda (EHRB.), Tetrahymena 8etoBa (SCHEWIAKOFF), Stentor coeruleuB EHRB., Stylonychia mytilu8 EHRB., and Euplote8 patella (OFM) were isolated from ponds in the Botanical Garden in Aarhus, Denmark. Zoothamnion sp. was collected as an epibiont on the amphipod, Gammaru8 lacuBtris, from a stream in northern Jutland. Oolpoda cucullus OFM and O. steini MAuPAs were isolated from soil of potted plants in the laboratory of the author in Denmar k. All species (except Zoothamnium) were grown in the laboratory on wheat or rice grain infusions or fed Ohilomona8, Glaucoma, or Oolpidimn according to the preferences of the different species. Methods empoyed for the quantification of clearance and particle size selection (based on the ingestion rates of suspended latex beads with different diameters) and measurements of cell volu-

241

Structure and Function of Feeding Organelles

mes are described in FENCHEL (1980a). Directions of water currents in the mouth of ciliates were determined by observing feeding ciliates in suspensions of latex b eads. Water velocities were esti· mated from photographs of such prepara tions in stroboscopic light so that positions of indivi fual beads could be determined at known time intervals. An alternative method was to photograph ciliates in p article suspensions at exposure times of 1/60 or 1/20 sec in constant light. This allows t he measurement of streaks on the film form ed by the moving p articles. B eat frequen cies of ciliary membranelles were estimated with stroboscopic light (Transistor Strobotorch 1202D, Dawe Instruments Ltd.). In this paper, all rates of ciliary acti vity, clearing r ates, et c., refer to temperatures of 20-22 cC. Mouth morphology was studied on prota rgol stained specimens (TUFFRAU 1967, as modified by BROWNLEE, personal communication). The free space between paror al cilia was estimated by counting the number of cilia over a known distance and assuming a ciliary diameter of 0.2 I'm. In a ddition, a number of species w ere examined with SEM (AMR 1000 A) following SMALL and MAU· GEL (1978). Tra nsmission EM (SIEMENS 101) was c arried out on cells fixed in 25 % gluta ra ldehyde followed by a buffered 1% OS04 solution, imbedded in Westopa l, sectioned on a LKB microtom e a nd stained with ura nyl acetate. Terminolog y of mouth organelles and of systematic groups fol· lows CORLISS (1979).

Results and Discussion 1. General Properties of the Feeding Apparatus

A qua ntitative description of food particle uptake has been given elsewhere (FENcHEL 1980a, b) but for the sake of the following discussion it will shortly be 8

.-

I

-; .c 0

~ ~

~E

•

~

0

"~

/. I

V

• 0

part iete- c.onc.f'nfration 0

4

8

~",\ 108 ",1- 1 10

F ig. 1. The ingestion r ate of 0.36 I'm latex p articles by Colpidium campylum as a function oi particle concentration (measured as total volume ml- 1 ). The m a ximum clearance (the slope of the straight line) is 4.08 X 10-5 mlh- 1 and the maximum uptake r a t e is 8.4 X 103 f'm 3 h - 1 • 16'

242

T.]'ENCHEL

o Bvrsorlo trl..lnc.oleolla

• EuPlote-S rnof'bilJSoi ~

Blf'pharisma ornf'ric. a num

50

10

o (1)

,~

,

E

• a

Colpoda tuc.ullus Glauc.oma sc.intillans Cyc.lidium glauc.orna Colpidium c.olp oda

,.

C. c.ampylum

o

100

"0 >

'"c o ~

a

E

10

],ig. 2. The size spectra of ingested p articles for three species of spirotrichs (above) a nd for five species of holotrichs (below). The clearing r a tes are expressed as multiples of the cell volumes of the different species.

reviewed here. The "functional response", that is the uptake rate as a function of environmental particle concentration, can be fitted to a h yperbolic function (Fig. 1) which is characterized by two parameters: the "maximum clearing rate" (dimension: ml h -1) which is the volume of water per unit time which the ciliate can clear at low particle concentration (viz., the slope of the straight line through the origin in Fig. 1) and the "maximum uptake rate" which is the number or total volume of food particles


243

which the real uptake rate assymtotically approaches at high particle concentrations. This relationship is explained as the result of a constant rate of water ventilation in conjunction with a time constant (which equals the reciprocal of the maximum uptake rate) which measures the time taken to ingest a unit of food and during which the mouth apparatus is blocked for further ingestion. Both parameters are functions of particle size (Fig. 2). In particular, there is a minimum particle size below which the filtration efficiency is zero. Most hymenostome bacterivores retain particles down to ,",,0.2f-l m while none of the studied spirotrichs can retain particles < 1-2 f-lm. The clearance is a parameter which is easy to measure as the rate of uptake divided by the environmental concentration, and is useful in an ecological context since it gives information on the concentration of food particles needed by the organism in question. However, this measure does not have a trivial relation to the volume of water moved by the organism during feeding. When a ciliate is transporting water through its feeding organelles it will, due to viscous forces, move considerably larger volumes than those actually cleared for particles. A meaningful measure of the transport rate and of the absolute efficiency of particle retention (that is, the fraction of particles retained from the water actually passing the "filter") requires an understanding of the identity and function of the structures responsible for particle retention. 2. Feeding Currents In higher ciliates (viz., the Oligohymenophora and the Polyphymenophora) the water currents for feeding are formed by ciliary membranelles found along the left side of the mouth. In the former mentioned group there are three, and in the latter many (in some cases ,",,100) membranelles. In most cases, each membranelle consists of three parallel rows of dense cilia which in some cases may be fused, and they beat in unison (CORLISS 1979; DIDIER 1976; PECK 1978). The key observation with respect toO the functional role of the membranelles is that they propel water in a direction which is nearly parallel to themselves, in spite of the fact that the main velocity component of the individual membranelles seems to be perpendicular. This has previously been observed in isolated cases. MACHEMER (1966) found this to be the case in Stylonychia. He maintained that the ciliary beat is perpendicular to the membranelles, but that the leading edge of the membranelle is stronger (with one extra row of cilia which is terminated in the middle of the membranelle) giving rise ro a "turbine effect". SLEIGH and AIELLO (1972) working with Stentor, however, found that the effective stroke of the membranellar cilia is not perpendicular to the membra nelles, but that the main velocity component is directed outwards and parallel to the membranelle, thus explaining the direction of the water flow. There is hardly any fundamental difference between the function of the mem branelles of Stentor, Stylonychia or of other ciliates. My own observations on the beat of Stylonychia membranelles, including photographs taken with electronic flash, show that there is a distal velocity component dm-ing the effective stroke. Furthermore, the cilia within each membraneUe do not beat synchronously but form metachronal waves which may drive the water currents. This is e,q,sily observed with stroboscopic light

244

T.FENCHEL

Fig. 3. A generalized representation of the feeding CWTents in the mouth of a n oligohymenophoran (left) and a polyhymenopho!"an ciliate (right).

____I I

Ii

Fig. 4. Left: feeding currents in the mouth of Colpidium colpoda (scale: 5 ,'m). Right, above: feeding currents in the mouth of Tetrahymena setifera (scale: 10 pm). Below: feeding currents of Paramecium trichium.


245

Fig. 5. Oyclidium sp. drawn on the basis of SEM graphs (scale: 5,um).

in forms with long membranelles such as Paramecium or peritrichs. TAYLOR (1951) showed that a sheet immersed in a viscous fluid and down which lateral displacements are propagated will drive water in the direction of the waves. While the complex shapes and beat patterns of the membranelles do not allow to use TAYLORS model for a quantitative prediction of water velocity in terms of the wave parameters, the model does explain the direction of the water currents. A generalized representation of the water currents in the mouth of an oligohymenophore ciliate is shown in Fig. 3 (left). The cytostome is situated in the posterior end of the buccal cavity. The membranelles are situated in the bottom and along the left margin of the cavity and accelerate water in the direction of the cytostome. Here the water currents are forced upwards and anteriorly, but before leaving the mouth the water is intercepted by the paroral membrane (haplokinety), a single row of cilia found along the right margin of the mouth. These cilia which cover the right side of the buccal cavity are assumed to sieve particles larger than the free space between the cilia. The particles are thus concentrated in the space between the paroral membrane and the cytostome and eventually phagocytized.

246

T.FENCHEL

I

llOpm

Fig. 6. Feeding currents of Zoothamnium sp. Only currents above the plane are indicated (scale:

10 f4Ill).

This general description covers forms such as Tetrahymena and Colpirlium (Fig. 4). In Glaucoma, the water currents are basically identical to what has been described above, but the inconspicuous paroral membrane, which is situated outside the right, raised rim of the buccal cavity does not seem to play any role for particle retention which is carried out by the third (rightmost) membranelle (FENCHEL and SMALL 1980). The function of the mouth of Paramecium is highly reminiscent of e. g. Colpirlium, but in Paramecium, the buccal cavity is much deeper and more narrow. The three membranelles (the "ventral" and "dorsal peniculus" and the "quadrulus") drive water along the left and dorsal walls of the buccal cavity and the water returns along the ventral and right walJs to be strained through the paroral membrane close to the entrance of the mouth (Fig. 4). It is possible, that the third membranelle (quadrulus), which has much more widely spaced cilia, also acts as a filter in a way similar to what is found in Glaucoma, but this could not be observed directly.


247

..

Fig. 7. A g eneralized r epresentation of the feeding currents in the infundibulum of a peritrich. The incoming current (black arrows) is drawn along the right side of the polykineties and e ventually driven through the parol'a l membrane.

In scuticociliates, the paroral membrane is much larger and its cilia are directed distally and nearly perpendicular to the shallow buccal cavity (Fig. 5). In the scuticociliates, the membranelles generate water currents which are directed posteriorl y and to the ri~ht and directly towards the extended paroral membrane which retain food particles. Due to the posterior velocity component of the water, the filtered particles are moved in a posterior direction along the membrane and towards the cytostome. In Cyclidium, the membranelles beat constantly and with a frequency of about 15 Hz while feeding. The paroral membrane is immobile most of the time but with intervals of several seconds, metachronal waves pass down the membrane. This may aid the transport of filtered particles to the cytostome. The feeding currents of sessile peritrichs were well described by SLEIGH and BARLOW (1976) and my observations on a Zoothamnium sp. confirm their study (Fig. 6). The large polykinety, presumably homologous to membranelle no 1 (LOM 1964), encircles the oral pole and terminates in the bottom of the infundibulum. It beats with a frequency of 33-42 Hz forming metachronal waves which move in the direction of the membranelle and towards the infundibulum. The wave length is ,....,25,um

248

T.FENCHEL

/

,-

/

/

/

/

Fig. 8. The water currents form ed by the membranelle zone or Euplotes moebiusi as seen from the dorsal·left and ventral· left sides (scale: 10 pm).

so the metachronal waves move with a velocity of about 1 mm sec-I. The paroral membrane runs distally to and parallel with the polykinety but its cilia are directed more horizontally. It beats with the same pattern as the polykinety but its phase lags slightly (I/S-Ih wave length) behind the polykinety. Seen from the side, water is dra wn towards the oral pole. Observed from the oral pole it can be seen that the water currents have a counter clockwise velocity component directed towards the infundibulum and which is generated by the metachronal waves. Particles entering between the polykinety and the paroral membrane are retained by the latter and directed towards the infundibulum. The exact morphology of the small (,...,..,lO,um) infundibulum is very hard to make out in details and the schematic drawing (Fig. 7) is mainly based on LOM (1964). The water currents, however, could be described with latex beads.

249

Structure and Function of F eed ing Organelles

,

•

150 iJ m t

, ~-; ;

1000 pm

Sf"-'

I

10flm

Fig. 9. Above: tracing of a photogra ph of Stylonychia myti lus in a suspension of latex beads and exposed for 1/60 sec. The arrows in the a nterior and posterior ends of the membranelle zone underestimate the velocity of the water since the currents in these regions are not parallel to the plane. Below, left: the feeding currents in the mouth of Blepharisma americanum. B elow, right: the feeding currents in the mouth of Stylonychia mytil~18 (scales: 10 ,urn).

250

T.FENCHEL I J

\

\

100pm Fig. 10. The feeding current,s of Bw'saria tt'uncatella (scale: 100,urn).

Water with concentrated food particles is drawn by and parallel to the polykineties (of which nos 2 and 3 are also represented in the infundibulum). The haplokinety follows the counter clockwise spiralling down along the wall of the infundibulum, but at a relatively larger distance from the membranelles than outside the infundibulum. At the end of the infundibulum, where the cytostome is situated, the water is forced upwards again and perpendicularly through the paroral membrane so that retained particles are concentrated in front of the cytostome. The exhalant water current moves upwards on the opposite side of the polykineties and emerges from t.he infundibulum on t.he proximal side of the polykinety. The basic mechanism of peritrich suspension feeding is not different from that of hymenostomes. In the peritrichs, the torsion of the buccal ciliature allows for a much larger paroral membrane and thus a much larger filter area. The polyhymenophorans show a basically different system, since in these forms the membranelles not only generate the feeding currents but also constitute the filter (Fig. 3). A paroral membrane may be present or absent but it does not playa role

251


t

i~ .

....

"'~-

.~

....

Fig. 11. The feeding currents of Stentor coeruleU8.

for the interception of food particles in any of the species studied since its cilia are much too densely situated to explain the particle size selection of these forms. The parora] membrane may assist in the posterior propagation of water currents by the metachronal beating of its cilia or simply function as an extension of the ventral floor of the buccal cavity. As mentioned, it is lacking or strongly reduced in several species of spirotrichs.

252

T. FENcHEL

I ,:.,;.;;:§{~::;?F .~:; ~:'~:~ ~.~. : ,':..?~ J"':~:""~

IOpm

,;

,t.,·I, ',01'., .': ,1 .. " , '

"

:', ,,1/ '-

\\

'

., ... .

'.~

' . " . . . '~. ,'" ", \'~

..... .

'.

il~A.( Fig. 12. Golpoda cucullua and the feeding cw-rents of the mouth (scale: 10 I,m).

The membranelles are arranged nearly perpendicularly in a band which extends from the anterior end to the cytostome bordering the left margin of the buccal cavity. From the anterior to the posterior end it is twisted about 180°. The membranelles propel water in a direction distal and perpendicular to the band. In the anterior end of the animal, water is driven in a posterior direction thus creating swimming currents and also accelerating water toward::: the buccal cavity. More posteriorly, the water is pumped out from the buccal cavity between the membranelles and the velocity components get a more distal and eventually vent.ral direction as the band approaches the cytostome (Figs. 8-9). Food particles, which cannot pass between two adjoining membranelles are retained within the buccal cavity. Due to the posteriorly directed component of the water current in the buccal cavity, retained particles are washed posteriorly along the membranelle zone to be concentrated in front of the cytost.ome. The membranelles of spirotrichs typically beat with frequencies of about 40 Hz and water leaves the buccal cavity with velocities of 0.5-2 mm sec - 1 (Fig. 9). This description of fpeding currents covers forms like Euplotes, 8tylonychia, Blepharrisma, and probably most other heterotrich, oligotrich, and hypotrich ciliates. Some variations of this basic patterns were also studied. In Bursaria truncatella (Fig. 10), the right margin of the mouth covers most of the buccal cavity which has a characteristic horn shape. The ventral floor of the buccal cavity has a window for exhalant water. The water currents of the mount are shown in Fig. 10. In this giant ciliate, the retention of food particles can easily be observed directly. The free space between neigh-


253

bouring membranelles is 8-1O,u m and small ciliates like Cyclidium which enter with the inhalant current are rarely retained by the membranelle zone. Larger food particles (e. g., Colpidium campylum) cannot pass between the membranelles but are drawn towards the cytostome by the posterior velocity component of the water current. The beating of the cilia situated in a field on the right margin, immediately before the cytostome, may assist in squeezing individual food particles into the forming vacuoles. SLEIGH and AIELLO (1972) and LIEBSCH (1976) have given good descriptions ofthe water currents produced by the membranelles of Stentor, but they did not offer a satisfactory explanation of the filtration mechanism. The membranellar zone spirals clockwise (seen from the oral pole) 720 0 downwards to the cytostome (Fig. 11). Everywhere the band accelerates water distally and away from the cytostome and this creates a compensatory, central water current directed towards the cytostome. Particles, which cannot pass between the membranelles, are therefore retained in the buccal cavity. They are often pushed back into the central waterflow and thus retained at lower levels of the membranelle zone and eventually squeezed against the cytostome by the incoming water current and the innermost membranelles. Finally,and out of taxonomic context, the feeding currents of the vestibuliferan (trichostome) genus Colporla will be shortly described. In these forms, the cytostome is found in the bottom of a depression, the vestibulum (Fig. 12). In the vestibulum, two fields of dense rows of cilia are found. The small sizes of the studied species (C. cucullu8 and C. steini) and the fact that they swim while feeding, made direct observations difficult and the exact function of the vestibular ciliature could not be made out. The feeding currents are shown in Fig. 12. It may be assumed, that the left field accelerates water out of the vestibulum, thus drawing water from the anterior and left opening of the vestibulum and past the cytostome, and that the right, or both fields retain suspended particles. But the details of the process remain conjectural. 3. Comparisons between Water Transport and Clearing Rates As previously mentioned, the clearance is an easily defined and measured parameter whereas a meaningful definition of the transport rate requires an understanding of the mechanism of filtration. Even so, it is difficult to obtain accurate estimates of the transport rate. However, such estimates may contribute to the understanding of the mouth function. Transport rates are most easily estimated for spirotrich ciliates. Assume that the part of the membranelle zone which accelerates water out of the buccal cavity constitutes the filter and that the filter area is determined by the height of the individual membranelles times the length of the band. This area times the velocity of the water expelled from the buccal cavity would then be a meaningful measure of the water transport. In Stylonychia, the height of the membranelles is about 15,um and the length of the membranelle zone along the left margin of the buccal cavity is about 90,um. The water leaves the membranelles with velocities of somewhat more than 1 mm sec-1 which yields a transport rate of 1.36 X 106 ,um3 sec - lor 4.9 X 10-3 mlh-l. Experiments with the ingestion of 5.7 ,um latex beads yielded a maximum clearance of about 5.5 X

254

T.FENCHEL

10-3 mlh-l (FENCHEL 1980b) or a quite close agreement. In Stentor coeruleus, WENZEL and LrEBscH (1975) found a maximum clearing rate (for suspended Tetrahyrnena as food particles) of 5.2 X 10-2 mlh-l. The total membranellar area (in the sense discussed above) of a S. coeruleus is about 1.5 X 1()4 f-lm 2 and with water velocities of about I mm sec-I, this yields a water transport of 5.4 X 10-2 mlh-l. In Bursaria truncatella, the inflow velocity through the anterior opening of the buccal cavity is also close to I mm sec-I. The area of this opening is 1.2 X 105 f-lm 2 which yields a water transport rate of 0.424 mlh-l. The maximum clearance (measured with Colpiilium campylum cells) was found to be 0.433 mlh-l. As described in FENCHEL (1980a), it is possible in Bursaria to observe what the above estimates indicate: that for certain particle sizes, the retention of particles from water passing the filtering apparatus is close to 100 %. Similar considerations can be made for oligohymenopherans, but estimates of water velocities are less accurate. In Cycliilium sp. the paroral membrane measures about 20 f-lm 2 • Water velocities in the buccal cavity were (with some uncertainty) estimated to be 45 f-lm sec-l which gives a transport rate of 3.24 X 10-6 mlh-l. The maximum clearing rate (measured with 0.36 f-lm latex beads) was found to be 2.86 X 10-6 mlh-l. In Glaucoma scintillans, crude estimates suggest that between I and 2 X 10-5 ml are passed through the buccal cavityper h. The maximum clearance (measured with 0.36 f-lm latex beads) is 1.3 X 10-5 mlh-l. In general, the low frequencies (7-15 Hz) and low velocities of feeding currents of tetrahymnine ciliates is in agreement with the low maximum clearing rates characteristic of forms which can retain very small «0.5 f-lm) food particles (FENCHEL 1980b, c). 4. The Mechanism of Particle Retention In the previous sections it is implied that particle retention in suspension feeding cilia tes is based on sieving through ciliary structures of the mouth. Since the mechanism of particle retention in suspension feeders in many cases still is controversial, some further evidence will be given here. There are basically three mechanisms which could explain particle retention in suspension feeders. These are I) mechanisms based on inertial or gravitational forces;

Fig. 13. The acceleration of a particle along a semi-circular path with the tangential velocity component Vo and the perpendicular component vp which is due to inertial forces.

255


10

o

•

E

.

:1-

0

50'/, of max c.1e-oranc.E' min. size' reotai nf'd

.9

.;;;N 5

9

~

~

"

::;

0

Cl.

8 0

7

• •7

8 60 ~O

0.5

20

3 0 05 10 3

0.1

...

6.

• •4 .5

fi

2 0.1

free'spac.e' bt'twf'fln c.ili a of filtf"'r

0.5

•

~m

10

Fig. 14. The relationship between the free space between the cilia of the organelles assumed to retain particles and the sizes of particles retained with an efficiency of 50 % and the minimum size retained. 1: Colpidium colpoda (paroral membrane), 2: Glaucoma scintillans (membranelle 3), 3: Cyclidium sp. (paroral membrane), 4: Paramecium caudatum (paroral membrane), 5: Golpoda cucullus and 6: C. steini (both the distance between rows of right vestibular ciliary field), 7: Euplotes moebiusi, 8: Blepharisma americanum, and 9: Bursaria truncatella (all the free space between adjoining membranelles in the posterior end of the zone).

2) various hydrodynamic mechanisms in conjunction with feeding surfaces to which food particles tend to stick; and 3) sieving. Particle retention based on interial forces depends on suspended particles tending to leave flow lines when these are curved (accelerated) around some structure. Particles could in this way be forced against a sticky surface or towards the mouth. Although this principle has often been suggested to explain particle concentration, it can definitely be stated that it cannot play any role for suspension feeding animals. The reason is that with the dimensions and velocities in question, viscous forces totally dominate over intertial ones. This can be seen from the following argument. Assume a particle is accelerated around a semicircle with a radius r and a tangential velocity, v 0 (Fig. 13). The perpendicular velocity component of the particle, vp , can be calculated by equating the centrifugal force, Fe, with the frictional force, Ff , experienced when travelling perpendicular to the flow lines. The centrifugal force equals [nd3 (ep - em) v2o]/6r, where is the particle diameter and ep and em are the densities of the particle and the medium, respectively. The frictional force will be .3nrJavp, where rJ is the visco-

a

17 Arch. Protistenk. Bd. 123

256

.

1:

T.

0.109

FENCllE;L

0.356

~m

~m

60 "-

'" ~

"0 ~

v

o

>

20

10

'.c 30

•

~

"0

3o >

60 'Eo

:L

20

1.09

~m

"';~

60

t

"0

"~

20

Il)

20

]'ig. 15. Rate of vacuole formation and vacuole volume as a function of environmental p a rticle concentration for four different particle sizes in Glaucoma scinti llans. Note the different scales on the horizontal axises; particle r et ention is very inefficient for the O.1091-'m particles.

sity of the medium. From these equations we have vp = [cfl(!?p - !?m}v2o] J18rrJ. The length of the semi-circular path in Fig. 13 is nr which will take nrJvo time units to pass. During this time, the particle will have travelled nrvpJvo length units away from the flow line. As an example, consider a particle with a diameter of 2 Inn and a density of 1.2 accelerated along a semi-circular path with a radius of 5 ftm and with a tangential velocity of 1 mm sec-1 . The viscosity and density of water is taken to be 0.01 Poise and 1, respectively. The particle will then move only ",10-4 ftm relative to the flow line, and anyset of reasonable parameter values for cilia tes (or metazoan filter feeders) similarly shows that inertial forces cannot be of significance. Gravitationa,l force (settling) was suggested to play a role for food concentration in Stentor by LIEBSCH (1976). However, arguments quite similar to those above shows that this is impossible. According to STOKE'S law, the settling velocity of a spherical particle isgd 2 (!?p - !?m) JJ8r;, or (assuming the above values for densities and water viscosity) 3 X 10-2 ftm sec-1 for a 0.5 ftm particle and about 10 ftm sec-1 for a 10 ftm particle. Since feeding currents typically have velocities of the order of 1 mm


257

sec -1 (and also since most food particles can move themselves faster than their settling velocity) gravitational forces can be ruled out as a mechanism for particle concentration. RUBENSTEIN and KOEHL (1977) suggested a number of retention mechanisms based on adhesion or sticking of particles to surfaces. Of these, "direct interception", viz., the collection of particles which move along stream lines within a distance of a particle radius from the adhesive surface, may well be important in na ture. However, the mechanism implies predictions, which are not fulfilled in the case of ciliates. Thus, one would not expect the rapid decrease in retentien efficiency with decreasing particle size as shown in Fig. 2, and maximum clearance rates would be significantly lower than water transport rates as defined here. Although NILSSON (1972) found mucus in the feeding vacuoles of Tetrahymena, the fact that ingested small particles (bacteria, latex beads) initially show Brownian movements inside feeding vacuoles, shows that particles are not trapped in mucus prior to ingestion. Similarly, prey protozoa often swim freely in the feeding vacuoles of e. g., Bursaria or Stylonyckia for some minutes after ingestion. The strongest argument in favor of a sieving mechanism in suspension feeding ciliates is the close agreelllent between the free space between cilia of the filter and the minimum sizes of particles retained (Fig. 14). Although very accurate measurements of ciliary distance as well as of the minimum particle sizes retained (based on Fig. 2 and on FENCHEL 1980a) are difficult to obtain, the graph does support the assumption that particle retention is based on a simple sieving mechanism. 5. Food Vacuole Formation The ingestion rate is evenhwJly determined by the formation rate and the volume of feeding vacuoles. As previously shown (e. g., MUELLER et al. 1965; :KILSSON 1972), particulate material is necessary to induce t.he formation of feeding vacuoles. When the concentration of particles in the environment is increased, the rate of formation , as well as the volume, is increased. The exact response depends on the particle size (Fig. 15). In this example it is seen that the larger particles have a higher tendency to induce vacuole formation so that a higher number of smaller vacuoles are formed than when smaller particles are filtered at similar ingestion rates. While the exact form of the results shown in Fig. 15 may be influenced by the artificial nature of the particles, the figure does show tha.t the maximum rate of vacuole formation is rapidly achieved with increasing environmental particle concentration and that further increase in ingestion rate is mainly due to an increase in vacuole volume. The filtration apparatus of the mouth plays by far the greatest role in the concentration of particles from the environment. However, some concentration also takes place after the formation of feeding vacuoles since they shrink during the first couple of minutes. A quantitative example can be given for Golpiai1lm colpoaa allowed to feed for 15 min in a suspension of 0.36,urn latex beads (108 ,urn3 or 4.2 X 109 particles ml - 1 ). The rate of vacuole formation was 12.8 h -1. The older vacuoles were completely spherical and densely packed and with an average volume of 469,um3 of which 66 % or

258

T.F!';NCHEL

312 11m3 is made up of latex. The last formed vacuoles, however, had a volume of 1,760I1m3. Assuming they also contained 312 flm 3 latex, it can be calculated, that the mouth apparatus had concentrated the environmental particle suspension by a factor of 1.8 X loa and that a subsequent concentration by factor of 3.8 had taken place in the vacuole due to sorbtion of water from the vacuole. This result, of course, depends on the environmental concentration of particles. A lower (more natural) concentration would yield a correspondingly higher concentration factor for the mouth apparatus but a relatively unaltered vacuole concentration factor.

6. The Evolution of Suspension Feeding in Ciliates This section does not pretend to contribute to the understanding of ciliate phylogeny in general; rather the intention is to offer a model, based on functional and morphological properties of extant ciliates, which can explain the origin and evolution of the complex mouth structures found in the oligo- and polyhymenophoran ciliates. There is a general consent that the gymnostomes comprise the most primitive extant ciliates (CORLISS 1979). In these forms, the cytostome is situated at the surface of the cell. They are basically macrophages, viz., they feed on large particles (algae, other protozoans) which come in contact with the mouth of the animal (DRAGESCO 1962 ; FENCHEL 1968). In principle these forms secure their food by "direct interception". Although various mechanisms, such as chemotaxis (as in histophagous ciliates) or rapid swimming (as in Diainium and Lacrymaria) ma y enhance encounters with food particles, they can only efficiently utilize food particles which are relatively large.

Fig. 16. Hypothetical stages in the evolution of feeding currents of suspension feeding ciliates represented by two generalized vestibuliferan ciliates.

An alternative to rapid swimming is to transport large volumes of water past the cytoetome. Such evolution has taken place in the vestibuliferan (trichostome) ciliates. In these forms, the cytostome has descended below the body surface in the bottom of a vestibulum. A primitive version of this is represented by the hypothetical ciliate


259

in Fig. 16 (left) which somewhat resembles Coelosornides. In such forms, the vestibulum is lined by somatic kineties. If their cilia beat their normal direction, they will tend to drive water out of the vestibulum along the walls and thus create a compensatory, central in-going flow of water. A further development of this system is seen in forms like Balantidiurn or Sonderia where a posterior migration of the cytostome has allowed a considerable enlargement of the mouth (Fig. 16, right). These forms probably depend mainly on relatively large food particles; at least, the plagiopylids feed on quite large bacteria and on microalgae (FENcHEL 1968). To what extent the cilia of the vestibulum function not only by drawing water with food particles to the cytostome but also for the retention of suspended particles is nct known. It seems that the more specialized vestibular cilia of Colpoda have this function as well. From such forms, the basic, asymmetric system of the oligohymenophorans must have evolved: ciliary membranelles on the left side of the mouth which propel water past the cytostome and a set of more or less motionless cilia on the right side of the mouth which filter suspended particles from the water, as an adaptation to feed on very small food particles. From this system to one in which the membranelles themselves act as the filter, the step is small; it requires a turning of the membranelles so that the formed water currents are directed to the left and ventrally rather than directly against the cytostome. Such a system ma y function in some (not studied) oligohymenophores. It is the basic food gathering system in the polyhymenophorans where it is associated with a polymerization of the membranelles leading to an enlarged filter area. The evolution of the higher ciliates may therefore be seen as the result of adaptations to feed on small food particles. The reven:e evolution has taken place many times, but these macrophagous hymenostomes and spirotrichs often remain suspension feeders retaining the basic mechanisms of particle retention. An extreme example is L€rnbadion Although it feeds on ciliates which may be more than 10 % of its own volume these are caught by accelerating water with prey organisms through the large buccal cavity with its membranelles. Similarly, within the polyhymenophorans, many groups have evolved forms which feed on relatively large prey organisms (e. g., Bursaria and Stentor). Although these forms share food particles with the primitive raptorial ciliates, the mechaniEms by which the prey is Eecured are in principle different.

Acknowledgements Parts of these studies were carried out during a sabbatical stay at Marine Biological Laboratory, Woods Hole. I am grateful for hospitaly and help during my stay. The SEM studies and some of the other work on ciliate morphology was carried out at the Department of Zoology, University of Maryland at College Park. My gratitude is to my colleagues there and in particular to Drs. D. BROWNLEE and E. B. SMALL for generous help and discussions. I am also grateful to Professor C. BARKER JORGENSEN, Department of Zoo physiology, University of Copenhagen, for discussions and constructive criticism of this paper and to Ms. H. ADLER-FENCHEL for linguistic improvements In my laboratory in Aarhus, Mr. P. G. SORENSEN and Ms. A. SOLLING aRsisted me with photo graphic work and the preparation of specimens for EM.

260

T. FENCHEL, Structure and Function of F eeding Organelles

Literature CORLISS, J. 0.: The Ciliated Protozoa. 2nd ed. Oxford 1979. DIDIER, P.: Organisation ultrastructurale des cils des membranelles de quelques cilies oligohymeno. phora. Corps paraxonemiens et coalescence ciliaire. Protistologica 12 (1976): 37-47. DRAGESCO, J.: Capture et ingestion des proies chez les infusoires cilies. Bull. BioI. Fr. Belg. 46 (1962): 123-167. FENCHEL, T.: The ecology of m arine microbenthos. II. The food of marine benthic ciliates. Ophelia 5 (1968): 73-121. Suspension feeding in cilia t ed protozoa: functional response a nd particle size selection. Microb. E col. 6 (1980a): I-II. Suspension feeding in ciliat ed protozoa: feeding rates and ecological significance. Microb. E col. 6 (1980b): 13 - 25. R elation between particle size selection and clearance in suspension feeding ciliates. Limnol. Oceanogr. 25 (1980c): 735-740. and SMALL, E. B.: Structure and function of the oral cavity and its organelles in the hymenostome ciliate Glaucoma. Trans. Amer. Microsc. Soc. 99 (1980): 52 - 60. J0RGENSEN, C. B.: Biology of Suspension Feeding. Oxford 1966. - Suspension Feeding. In: Handbook of Nutrition and Food, Section E, (in press) Cleveland 1980. LIEBSCH, H.: Untersuchungen zur Nahrungsselektion von Stentor coeruleU8 (EHRENBERG). Zoo!. Anz., Jena 196 (1976): 1 - 8_ LOM, J.: The morphology and morphogenesis of the bucca l cilia ry organ elles in some peritrichous ciliates. Arch. Protistenk. 107 (1964): 131-162. MACHEMER, H.: Zur Koordination und Wirkungsweise del' Membranellen von Stylonychia mytilu8. Arch. Protistenk. 109 (1966): 257 - 277. - Ciliary activity and m etachronism in protozoa. In: Cilia and Flagella, SLEIGH, M. A., (ed.), 199-286, London 1974. MUELLER, M., ROHLICH, P., a nd TaRO, I.: Studies on feeding a nd digestion in protozoa. VII. Ingestion of polysterene latex par ticles and its early effect on a cid phosphatase in Paramecium multimicronucleatum and T etrahymena pyriformis. J. Proto zoo!. 12 (1965): 27 - 34. NILSSON, J. R.: Further studies on vacuole formation in Tetrahymen a pyriform'is GL. C. R. trav. Lab. Carlsberg 39 (1972): 83 - 11 O. PECK, R. K.: Ultrastructure of the somatic and buccal cortex of the tetrahymine hymenostome Glaucoma chattoni J . Protozool. 25 (1978): 186-198. RASMUSSEN, L.: Nutrient upta ke in Tetrahymena pyriformis. Carlsberg Res. Comm. 41 (1976): 143-168. RUBENSTEIN, D. 1., and KOEHL, M. A. R.: The mechanisms of filter feeding: some theoretical considerations. ArneI'. Nat. 111 (1977): 981-994. SLEIGH, M. A., and AIELLO, E.: The movement of water by cilia . Acta Protozool. 11 (1972): 265277. - and BARLOW, D.: Collection of food by Vorticella. Trans. Amer. Micr os. Soc. 95 (1976): 482 - 486. SMALL, E. B., and MAUGEL, T. K.: Observations on the permanence of protozoan preparations for scanning electron microscopy. Scanning Electron Microscopy 2 (1978): 123-127. T AYLOR, G.: Analysis of the swimming of microscopic organisms. Proc. Royal Soc. London A, 20, (1951): 447-461. T UFFRAU, M.: Perfectionnements et practique de la technique d i' mpregnation au protargol des infusoires cilies. Protistologica 3 (1967): 91-98. W ENZEL, F., and LIEBSCH, H.: Qua ntitative Untersuchungen zur N a hrungsaufnahme von Stentor coeruleus EHRENBERG. Zoo!. Anz., Jena 195 (1975): 319-337. Author's address: TOM FENCHEL, Department of Ecology and Genetics, Ny Munkegade, University of Aarhus, DK - 8000 Aarhus C, Denmark.