ABSTRACT
An important step toward the understanding of the beating mechanism in ciliary movement would be a quantitative analysis of the mechanical properties of the cilium. As a matter of fact, several measurements have been reported concerning the work done by ciliated tissues (Maxwell, 1905) or the force exerted by the whole organism in ciliate protozoa (Jensen, 1893) and flagellated bacteria (von Angerer, 1919). The present paper will deal with the force which a single molluscan cilium can produce during its effective stroke.
MATERIAL AND METHODS
The material used was the large abfrontal gill cilia of the marine bivalve, Mytilus edulis. As is clear in Gray’s micrographs (1930, 1931), cilia of this kind are convenient on account of their extremely large size (about 50 μ.) and sparse distribution ; because of the latter condition, the effects of neighbouring large cilia can be neglected. However, surrounding the base of the abfrontal cilia, there are innumerable small cilia covering the entire abfrontal surface of the gill, the influence of which will be dealt with separately later. One large abfrontal cilium consists of several fine subunits which are usually packed together, beating in unison so that the cilium might more appropriately be called a cirrus. Although the subunits can beat independently of each other when they are separated, such separated subunits were excluded from the present experiments.
Kinosita & Kamada (1939), using a glass micro-needle to arrest the movements of single abfrontal cilia, made the following observations.
(1) When a needle is applied at a point near the tip of the cilium (i.e. at a point more than 80% of the distance from base to tip, Text-fig. 1, a), the cilium passes under the needle by elastic backward bending and can complete the effective stroke.
(2) When a needle is applied at the proximal part of the cilium, the distal part of the beating cilium bends over the needle and the effective stroke is again completed (Text-fig. 1, b). Kinosita & Kamada called such a region which could bend over the needle to complete the stroke the’active region’ and estimated its length as 10-20% of the length of the cilium from base to tip ; the present investigation indicates that the active region occupies about 30% of the length of the cilium from base to tip.
(3) If the needle is applied to the mid-part of the cilium (between 30 and 80% of the length) the cilium is arrested without further bending (Text-fig. 1, c) and remains motionless until the recovery stroke makes it reverse its motion. Usually the shape of the arrested cilium is straight, as in Pl. 9, fig. 1 a, although sometimes the proximal part is slightly curved backward (Pl. 9, fig. 1 b).
Now taking advantage of situation (3)—motion arrested when cilium is held by mid-region—the force exerted by the cilium can be measured using a flexible needle whose bending characteristic has previously been calibrated (Text-fig. 1,c’) This is the principle followed in the present investigation. It is, however, admitted that the application of the method is restricted to the mid-region of the cilium.
Calibration of the bending balance
In a word, the bending characteristic of the micro-needle is calibrated against an actual weight. However, since it is impossible to put a small enough weight on a fine needle whose bending characteristic matches that of the cilium, calibration is done indirectly in several steps by employing a set of needles which differ progressively in compliance. Although this method is tedious, it enables us to determine the absolute value of the compliance of the needles as well as the range of maximum possible error. The procedure for calibration is as follows :
First, several horse-shoe-shaped pieces of copper wire are prepared, and their weights (1−3 mg.) are calculated from their length (1−4 mm.) on the basis of the weight of a known length of the wire (37·8 mg./50 cm.). These are the calibrating weights (error within 0·6%).
Secondly, three needles which are more rigid than the micro-needles to be used as bending balances are bent by putting the calibrating weights on their tips. The bending characteristics of the needles are determined from the movement of their tips. This first lot of needles will be called ‘reference needles’ (error within 2%).
Thirdly, many needles, whose compliance varies between that of the reference needles and that of the bending balances, are prepared and calibrated by making them push against each other and against the reference needles in various combinations. In this way the bending characteristics of all the needles can be found (error of this matching process within 15%).
Lastly, three of the needles, calibrated as above, are matched against the bending balances which will be used to arrest the cilia (error within 2·5 %).
The range of strengths of the bending balances used is 2·5−6·0 × 10−5 dynes per 1 μ displacement, the error for this figure being ± 16% (√ [0·62 + 2·02 + 15·02± 2·52] = 15·4). Deviation from Hooke’s Law is less than 2% for a movement of loop, and the mechanical asymmetry around the needle axis is less than 2·5 % ; both being well within the error of calibration.
Experimental set-up
As shown in Text-fig. 2, a gill filament (F) is laid across a slit of 5 mm. between two thin glass plates (P, P’). The filament is fastened in position by placing a coverglass (C) over the slit. This whole assemblage is fitted in a trough (T) in such a way that the cover glass is level with the sea water filling the trough. Although placing the filament on its flat side usually brings the beating plane of the abfrontal cilium to a horizontal direction (Gray, 1930), a careful selection is made of a cilium which is beating strictly in a focal plane.
A thick glass rod (R) (about 10 cm. in length and 7 mm. in diameter) is tapered at one end and fused with a bending balance (B) which is bent at right angles at a point about 5 mm. proximal to the tip. The glass rod now bearing the bending balance is so mounted on a manipulator (H) as to bring the needle tip upright. By illuminating the basal end of the rod with a microscopic light source (L), it is possible to make the tip of the bending balance shine brilliantly as a spot of 1.5–2.5 μ in diameter against the dark background. The movement of the tip caused by the ciliary activity can thus be recorded photographically. In actual records, photographic images of the cilium and that of the loci of the needle tip in the dark field were superimposed by successive exposures on one plate. For the bright field, a bright contrast-phase microscope equipped with a condenser of a long focal length for manipulation was used. Camera lucida drawings which were sometimes made gave essentially the same results as the photographic method.
RESULTS
By arresting a cilium at a point between 30 and 80% of its base-tip length, the bending balance is forced to move by several microns and throughout the period of arrest, of from a fraction of a second to several seconds, the cilium and the needle tip are stationary. It is therefore concluded that the cilium continues to exert a constant force on the needle tip throughout the period of arrest and that this force is equal and opposite to the force resulting from the displacement of the needle tip. Such a force will be referred to as ‘arresting force’. Occasionally a slight curving of the cilium itself (Pl. 9, fig. 1,b) was noticed which, however, was considered to be a passive elastic bending.
In order to make measurements, a photographic film is first exposed to obtain images of a cilium and the gill surface (see Pl., fig. 2, cilium). Secondly, the field is darkened and the shining tip of the bending balance is applied close to the base of the cilium (about 15 μ. distant from the base, A in Pl. 9, fig. 2). The spot of light is now oscillating between the arrested (A’) and the free (A) positions as the cilium moves. The shutter of the camera is then opened and the needle is slowly pulled toward the distal end of the cilium along the straight line previously set (A-C in Pl. 9, fig. 2). As the bending balance is pulled, the cilium beats against it and intermittently pushes the needle sideways till equilibrium is reached. The light track on the film shows the successive loci of the needle tip, and consequently the arresting forces at various levels, provided the base-line is given. When the needle is pulled clear of the tip of the cilium (B), the light track becomes straight (B-C). The preparation is then removed entirely and the needle is moved back along the same straight line once again to obtain the base-line (C-D). If the new straight line (C-D) failed to coincide with the former line (B-C), which sometimes did happen owing to the presence of mucus and other obstacles, the data were discarded.
Text-fig. 3 a represents four typical results. It is shown that the arresting force decreases as the position of the arresting needle shifts towards the tip of the cilium.
One point in connexion with this procedure requires serious consideration. As was just shown, the magnitude of force which a single cilium can produce varies from one end of the cilium to the other. Consequently, if a bending balance is applied at various levels of a cilium by simply sliding the needle along the cilium, the equilibrium positions of the cilium at various levels change in proportion to the forces exerted, and thus the basal angle a (see Text-fig. 1, c’) of the arrested position varies each time. Since it is desirable to unify the data in relation to a more or less constant value of α, the direction of the sliding motion of the needle necessary to stop the cilium within the 10 ° of the upright position has been previously found empirically. Such line slants toward the end of the effective stroke (line C-D in Pl. 9, fig. 2). Although this procedure may seem rather arbitrary in a physical sense, there is a possibility that larger biological complications may be involved if a is not kept constant.
In the introduction the active region of the cilium which is able to bend over the needle was mentioned. However, since the large abfrontal cilium remains nearly straight throughout the effective stroke (Gray, 1930, 1931), under the conditions of these measurements the situation can be looked upon, in simplified form, as follows; the movement of a cilium is due to a bending motion of the cilium restricted to its extreme base on the level of the gill surface. When the cilium is arrested, the force exerted by the cilium is also generated at this basal plane and is in equilibrium with the restoring force of the bent needle by way of an inert straight shaft of the cilium. If this simplified scheme is allowed, the forces which are being generated at the base give rise to a torque, whose magnitude is the product of the arresting force (which is known) and the distance between the level of arrest and the base (which can be measured).
This torque, as illustrated in Text-fig. 3 b, is found to remain constant irrespective of the position of arrest (cilium III and IV). The curves sometimes show a slight upward convexity (cilium I and II), but since this trend falls within the limit of error in taking measurements from the photographic records (± 5 %, see below), it does not invalidate the statement that the torque is constant irrespective of the position of arrest. It is notable that downward convexity has never been observed.
The results are summarized in Table 1. The mean value of the torque is 3·9 × 10−7 dyne.cm., with some variation (2.8 × 10−7 dyne.cm.). Measurements of photographic records involve a relative error of 5 %. Since the error in the bending characteristic of the balance is 16%, as stated above, the torque has an error of about 16% (√ [15·42 + 52] = 16·1). In fact, in one case, when the arresting force of the same cilium was measured by two different bending balances, the two series of values coincided very well with each other, as shown in Text-fig. 4.
Effect of small cilia of the abfrontal gill surface
In order to examine the effective range of the stream caused by small cilia, one experiment was carried out in which the tip of a bending balance was gradually brought close to a part of the gill surface where no large cilium existed. It was found that the range where the tip undergoes detectable shift is limited to within 10μ from the gill surface. It is therefore safe to say that small cilia have no effect upon the values obtained in the present experiments.
Distribution of force-generating capacity within the active region
In the preceding section, the force of the cilium is calculated on the basis of a simplification and expressed in terms of the torque produced by the extreme base of the cilium. But since the proximal 30 % of the total ciliary length has the capacity to bend, the active region must be able to produce torque at any level. Theoretically this can be analysed by immobilizing the active region itself at different levels, by applying a rigid needle (as in Text-fig. 1,b), and measuring with a bending balance the force which the remaining distal part of the active region exerts. The maximum value of torque obtained when the active region was immobilized at a point 9μ distal to the base (20% of the ciliary length) was about one-third of the torque obtained at the extreme base. Although technical difficulties made it impossible to inquire further into the activity of the more proximal part of the active region, it can be imagined that the force-generating capacity increases continuously from the distal end of the active region to the extreme base of the cilium.
DISCUSSION
The force which a single cilium can exert has been expressed as a torque acting at the extreme base of the cilium. This is justified in connexion with free ciliary beating because the cilium remains straight all through the effective stroke, the extreme base being the only part which bends.
On the other hand, it has been shown that there is a distal-proximal gradient in the ability of the active region to generate torque. However, from such an observation alone it does not necessarily follow that ah the components of the active region are contributing force to the free beating of the cilium. When the cilium is arrested, the active region usually remains straight or, possibly, is curved back as shown in Pl., fig. 1 b. Under this condition, it can also be said that the only part which is actively bending is the extreme base. To say the least, as long as the shape of the active region is maintained constant, the beating force is properly defined as a torque acting at the extreme base of the cilium.
The torque so far defined is that which a cilium can generate when it is arrested during the effective stroke. But whether or not the same torque is generated during movement is a different question. As in the case of isometric and isotonic contraction of muscle, it is very likely that the force developed in free movement would be much smaller and would increase as the viscosity of the medium is raised, reaching a maximum when the cilium is arrested. The author will report on this subject in a subsequent paper.
SUMMARY
A bending balance made of a flexible glass micro-needle was prepared, and its bending characteristic was determined by direct matching with calibrated needles.
The force produced by a large abfrontal cilium of Mytilus was measured by allowing the bending balance to arrest the effective stroke of the cilium.
The bending force generated by the cilium is described in terms of a torque, referred to the base of the cilium, amounting to 2−8 × 10−7 dyne.cm. (mean, 4 × 10−7 dyne. cm.).
ACKNOWLEDGEMENTS
The author expresses his thanks to Prof. Katsuma Dan for his advice throughout the work and kind help in preparation of the manuscript, and also to Prof. Hamo Kinosita of the Tokyo University for his valuable criticism.