A Respiratory Rhythm in Sea Anemones

Recent studies on the O₂ consumption of sea anemones have produced results that are inconsistent: Newell & Northcroft (1967) observed that Actinia equina consumed O₂ at variable rates at each experimental temperature used, whereas Sassaman & Mangum (1970) could find no evidence for different levels of metabolism in either Metridium senile, Haliplanella luciae or Diadumene leucolena. The former authors interpreted their results in terms of a ‘maintenance metabolism’, which was augmented at times by particular states of activity, the whole being termed ‘active metabolism’. Sassaman & Mangum, on the other hand, supported the concept of a continual state of spontaneous, more or less rhythmical activity. Both groups of research workers employed an O₂ electrode for measuring the O₂ depletion rate of the sea water containing the anemone in a sealed chamber, but whereas Newell & Northcroft took readings at intervals of 12 min, Sassaman & Mangum recorded the O₂ concentration continuously. Neither group detected any rhythmical changes in the O₂ level.

We have discovered a respiratory rhythm in Actinia equina and Metridium senile, somewhat similar to the rhythm obtained by Brafield & Chapman (1967) and by Chapman (1972) with pennatulid species. The discovery is of some importance, for it means that the results of the earlier investigations require reinterpretation. It also affords a possible explanation for the slow rhythmic contractions of the actinian body wall; slow rhythms have so far been ‘difficult to relate to definite functions in the animal’ (Ross, 1974, p. 283). Finally it helps to solve the problem of how O₂ gains access to the endoderm. Mangum & van Winkle (1973) state that only 35 % of the total surface area of Metridium is exposed to the ambient sea water, the remainder being in contact with coelenteric fluid.

Specimens of Actinia equina were allowed to settle individually in 800 ml glass beakers in a tank of running sea water at about 12 °C. For each experiment a beaker was selected in which the anemone was attached half way up the side. The water was replaced by 600 ml of aerated sea water and the beaker was placed on the platform of a Gallenkamp magnetic stirrer situated in a constant temperature room at 14 °C. During the experiment the sea water was stirred by a magnetic rod of dimensions 4·5 × 0·8 cm, rotating at approximately 90 rev min ^-1. A thermistor and a Protech-Simac O₂ electrode, calibrated by the Winkler method after setting the zero using 4 % sodium sulphite, were inserted near the side of the beaker, with the Teflon membrane of the electrode at the same level and on the same circumference as the stomodaeum of the anemone, and usually in the next quadrant on the downstream side of the latter. Twenty ml of liquid paraffin were slowly poured on the surface of the water to minimize gaseous diffusion between this and the air. The O₂ content of the water was recorded continuously on a pen-recorder set at a chart speed of 20 min in ^-1. In no case was the O₂ content allowed to fall below 75 % of air saturation. The O₂ consumption of the electrode in the sea water without the anemone was at the most only 0-04 ppm h ^-1. Control recordings using sea water from which the air had been evacuated showed that the paraffin was an effective barrier to the entry of O₂. The pH of the sea water did not materially change during any of the experiments. Some of the anemones were subsequently dried to constant weight at 105 °C. For the experiments with Metridium senile 1 1 beakers containing 800 ml sea water at 15 °C were used, with 50 ml paraffin covering the surface.

Three types of records of O₂ depletion were obtained, namely (a) rhythmic, (b) more or less linear and (c) irregular. In all some 60 % of the records for Actinia were of the first type. An example is shown in Fig. 1 A. There is a gradual decline in O₂ level with a regular series of small deflexions superimposed. The period of the rhythm is 34 min, but values of from 24 to 40 min have been obtained. Each deflexion in Fig. 1A consists of a fall of about 0·22 ppm lasting about 3 min, followed by a rise to nearly the same level as at the start of the fall. Rhythmical traces were also usually obtained with Metridium senile (Fig. 2A). Each deflexion comprised an initial fairly rapid drop, the rate of change then declining exponentially. After about 3 min this was followed by an exponential rise lasting from 3 to 6 min. The intervals between the deflexions varied from 15·0 to 36·2 (mean 27·9) min. The frequency became more uniform towards the end of the record, so that probably insufficient time (1 h) had been allowed for acclimation to the experimental conditions. The shape of the deflexions depended on the time required by the electrode to achieve equilibrium with tha O₂ concentration, as well as on the mixing conditions, but the deflexions would have been produced if the mouth had periodically opened for about 3 min, allowing some coelenteric fluid to be exhaled (Fig. 3). In anemones generally the water pressure in the coelenteron is slightly higher than that outside (Chapman, 1949; Batham & Pantin, 1950 a; Trueman, 1966), due to the apparently continuous slow intake via the siphonoglyph(s) (Parker, 1905a; Batham & Pantin, 1950a). The mouth is the only means of escape for the coelenteric fluid in Actinia, because it has no cinclides. Acontia were not protruded by Metridium during any of the experiments, so there is no reason to think that its cinclides opened. The electrode was situated at the level of the mouth and it is known that the coelenteric fluid in Metridium is lower in O₂ content than the sea water outside, except when the latter is almost completely de-aerated (Sassaman & Mangum, 1973).

The deflexions are probably the result of periodic exhalation from the coelenteron, but no change in the state or activity of the Actinia could be perceived when the deflexions were being recorded. No periodic muscular contractions or periodic ciliary currents could be detected, even when carmine particles were applied in the vicinity of the mouth. This was observed to inflate and open on occasions, but without obvious periodicity. Brafield & Chapman (1967) also failed to observe the rhythmic outflovi suggested by the O₂ depletion records they obtained with the pennatulid Pteroide griseum. Presumably, as they state, the volume and speed of exit of the fluid must be slight. Subsequently Brafield (1969) demonstrated that peristalsis causes a periodic release of water via a mid-ventral band of siphonozooids near the apex of the rachis.

Kymograph records of mouth movements of an Actinia showed that rhythmic patterns of movement can sometimes occur, with peaks of contraction at times at intervals of about 30 min, but the experiment is of dubious validity because the isotonic lever may have been influencing the rhythm. Spontaneous activity is certainly a feature of preparations of the radial muscles of the oral disc of Tealia, but the frequency of the contractions is often irregular and relatively high (about 40 h ^-1) (McFarlane & Lawn, 1972).

With Metridium there were times when the deflexions occurred in phase with peristaltic movements of the body-wall. For example, the deflexions in Fig. 2 A happened when the base of the column was constricted, the constrictions moving up the column slowly during the intervals in between.

The records show a more or less steadily declining O₂ concentration between the deflexions, presumably due to O₂ uptake via the ectoderm when the mouth was closed. In Fig. 1A the overall gradient of the trace is constant, at 0·2 ppm h ^-1. The total O₂ consumption was 0·15 μl mg ^-1 h ^-1 (/μl at N.T.P.). This rate includes both ecto- and endodermal O _a consumption. In Fig. 2A the Metridium (dry weight 1·03 g) was taking up O₂ at mean rates of from 79 to 330 μl g ^-1 h ^-1. The animal slowly contracted during the experiment, which may account for the decline in rate of uptake, the surface area for exchange progressively diminishing.

The smallest Actinia found so far to produce the rhythmic type of trace weighed 107 mg when dried.

The second type of trace (for example, Fig. 1B) showed a more or less constant O₂ depletion rate lasting several hours. Different rates were shown by different Actinia and even by the same anemone on different occasions. While this may seem to be at variance with the conclusions of Sassaman & Mangum (1970), it is possible that insufficient time was allowed for the anemones to adapt to the conditions of the experiment.

The third type, arhythmic and variable, is exemplified by Fig. 1C. Some six rates, ranging from 0·04 to 0·6 μl mg ^-1 h ^-1 can be calculated. The record resembles to some extent the graphs published by Newell & Northcroft (1967) for the same species. Two of the 8 records we obtained for unstimulated Actinia were of this type.

The three different patterns of O₂ uptake were not correlated with any particular appearance or activity of the Actinia. Some specimens showed the tendency to exhibit the rhythmic pattern better than others, but all three patterns were shown by most specimens. In one case (Fig. 1D) there was a change from the rhythmic to the arhythmic pattern during the experiment, yet here again no appreciable change in behaviour of the anemone occurred. Conversely, in another case the slope of the recording remained constant throughout, despite the anemone’s changing from a stationary and expanded state to one of crawling, then bending, then resting and finally tentacle deflation.

The possibility of artifact must not be neglected. The absence of deflexions and the variability in O₂ depletion rate may well be a consequence of unsuitable siting of the electrode relative to the position of the anemone. Much depends upon the mixing conditions. To be detected, water leaving via the stomodaeum must flow across the Teflon membrane before dispersing completely in the ambient sea water. The electrode should thus be close and at the right level. However, if too close, the sporadic movement of the tentacles and slight changes in shape of the anemone may affect the mixing of the general ambient sea water with the water in contact with the anemone on the leeward side, thereby producing apparent changes in the O₂ depletion rate. Siting the electrode for the optimal detection of rhythmic exhalations could thus diminish the accuracy of determination of the overall O₂ uptake rates, because this depends upon thorough mixing to ensure a uniform O₂ concentration. Following the discovery of the rhythm our experiments became biased towards the investigation of its properties, to the possible neglect of accuracy in the determination of the O₂ uptake rates.

Prodding the base of an expanded Actinia, or electrically stimulating at one shock s ^-1, elicits a contraction and closure response, but no increase in the O₂ uptake rate, even when the period of contraction is prolonged for 15 min. In five cases the interval between deflexions during which brief stimulation was applied was shortened. The stimulation was electrical in two of these cases and mechanical in the other three. In the example shown in Fig. 1E the intervals in order were 35, 32, 28 and 33 min. After the period of stimulation there was either no change in the O₂ depletion rate (6 records) or a prolonged decrease (4 records). Thus in Fig. 1F the slopes before and after stimulation were respectively 0·62 and 0·37 ppm h ^-1. Recording was continued for as long as 3 h after stimulation in one example. Presumably the anemones had not expanded to their initial state after the stimulation.

With Metridium the stimuli resulted in an augmented outflow from the coelenteron. Fig. 2B shows the effect of prodding a fully expanded specimen three times. The resulting deflexions had a depth of 0·32 ppm compared with the 0·17 ppm average for the rhythmic deflexions, suggesting that more fluid than usual was being exhaled as a result of the contraction following each stimulus. The contraction would have raised the internal hydrostatic pressure and thus increased the rate of outflow during exhalation. The duration of mouth opening, however, remained much the same (3– 4 min), suggesting that it is controlled by the anemone. Batham & Pantin (1950 a) refer to a mouth-opening radial mesenteric reflex which can operate when the internal pressure is raised. In Fig. 2B the second stimulus appears to have produced a supernumerary deflexion by this means. Such deflexions were not obtained with Actinia, which contracts less extensively than Metridium in response to stimulation; presumably Actinia keeps its mouth more tightly closed in between exhalations, which occur without obvious contractions of the body musculature. As with Actinia, mechanical stimulation of Metridium had no prolonged effect on the rhythm of deflexions. In Fig. 2 B the averages of the intervals before and after stimulation were not significantly different.

With Metridium electrical stimulation by 3 shocks at 1 s intervals resulted in an immediate severe contraction and a considerable deflexion of the trace (Fig. 2C), the mouth remaining open for a longer period than usual (20 min).

When a cube of fish muscle was captured and swallowed by Actinia, no immediate change became apparent on the O₂ depletion record when this was of the arhythmic type (b) or (c). With the rhythmic type, however, a small deflexion was obtained as the food entered the pharynx. Presumably the act of swallowing was accompanied by a leakage of some coelenteric fluid. The absence of a corresponding deflexion on the arhythmic records suggests that the electrode was not optimally sited for detecting exhaled coelenteric fluid in their case.

Three experiments showed that the act of ingestion resulted in no significant change in the duration of the intervals between the rhythmic exhalations. Thus in Fig. 1G the interval during which food was ingested (31 min) is shorter than those immediately before and after, but the difference from the average of the five other intervals (35 min) is not significant (P = 8·5 %). Feeding with fish cubes, however, did result in a prolonged increase in the O₂ depletion rate, and this was also noticeable in the arhythmic type of record.

When Actinia was given a suspension of particulate fish meal, there was no detectable change in the O₂ depletion rate, either immediately or in the long term. The experiments were continued for 2 h. Actinia is not a particle feeder like Metridium and presumably little ingestion, if any, was taking place. One would expect food to stimulate metabolism, but whether this is the explanation for the increased rate of O₂ uptake after the ingestion of fish cubes requires further investigation.

KCN, at a lethal concentration of 1·6 HIM, resulted in the abolition of spontaneous deflexions, when present, and the cessation of O₂ uptake after about 19 min (Fig. 1H). When the experiment was carried out at room temperature (about 18 °C) the O₂ consumption became zero after 13 min (Fig. 11). Cyanide relaxes the contractile tissues.

Experiments on the effect of temperature on the respiratory rhythm have not yet been carried out, but a comparison of the mean periods of the rhythms in Fig. 1H and I (respectively 33·75 and 21·4 min) suggests that the frequency of the deflexions increases with a rise in temperature, as does the rate of O₂ consumption (Sassaman & (Mangum, 1970). The period at room temperature corresponds with the 10– 20, or 25, min period usually given for rhythmic contractions of the body musculature on anemones in general (Batham & Pantin, 1950a, b, c;Needler & Ross, 1958; Ewer, 1960; McFarlane, 1973b).

Actinia equina does not contain any zooxanthellae (Smith, 1939), but it usually possesses red and brown lipochromes which are stated to act as optical sensitizers, producing photochemical action involving the release of O₂ to the animal (Elmhirst & Sharp, 1920). If so, one would expect the rate of O₂ uptake to be less in the light than in the dark. This was found to be the case for expanded Actinia by Smith (1939). Gompel (1937), however, found that theO₂ uptake was maximal at high water during the afternoon and minimal at low water during night time, despite the Actinia having been kept for 3 weeks in an aquarium beforehand. The effect of circadian and tidal rhythms on Actinia is complex (Piéron, 1906,1908; Bohn&Piéron, 1906; Bohn, 1906, 1907, 1910). In our experiments the Actinia used usually became more expanded in the darkness of the constant temperature room. Three of the records obtained were of the rhythmic type, and these revealed that the expanded condition was correlated with a decrease in the frequency of the rhythm. For example, in Fig. 1J the mean period during the light phase (a strip light was used, of intensity about 88 lux at the experimental bench) was 30 min in contrast to the 36·7 min of the dark phase. The difference is significant (P < 1 %). The O₂ depletion rates during the light and dark phases were respectively O’15 and 0·19 μ1 mg ^-1 h ^-1. Whether the increased rate during darkness was the result of the expansion (increasing the surface area) of the anemone, or of a cessation in release of O₂ by photochemical action, either by algae in the water or by lipochromes in the anemone, cannot be decided without further research. However, the sea water used had been continuously filtered and irradiated by ultraviolet light to diminish the presence of organisms in suspension.

Longitudinal bisection had a negligible effect on the respiratory rhythm (Figs. 1K, 2D). With Actinia a deflexion immediately followed each of the two cuts made, the second slightly shortening the interval since the preceding spontaneous exhalation. After bisection the two contracted halves tend to remain in contact. Water remains trapped in the cavities between the mesenteries and presumably this contributes to the fluid being exhaled. Also with Actinia the O₂ depletion rate was less after the bisection in both of the experiments carried out, possibly because of the state of contraction. The persistence of the rhythm is perhaps a reflexion of the occurrence of potential pacemakers in all sectors of the column (McFarlane, 1974b).

The expulsion of juveniles by a parent Actinia (Fig. 1L) was accompanied by a large deflexion on the O₂ depletion trace, presumably resulting from the release of a large quantity of coelenteric fluid.

Excess K+ has an excitatory effect on the contractile tissues of sea anemones (Ross & Pantin, 1940; Ross, 1960a,b), causing an increase in tonus and an increased frequency of contractions. Potassium, however, also influences the beating of cilia. Parker (1905 a) demonstrated that 2% KCl in sea water causes the reversal of the beat on the labial ridges of Metridium. It is possible that the rhythmic exhalations are caused by a stoppage or reversal of the beat of the pharyngeal cilia, and not by the rhythmical mouth opening or contractions of the body-wall musculature as hitherto assumed. However, this possibility has not been supported by direct observation. There are no reports of periodic changes, despite investigations of ciliary reversal in response to stimulation of the oral lips of Metridium (Parker, 1905 a, b, 1928; Parker & Marks, 1928) and of Actinia (Baba, 1968).

Fig. 2E shows the effect of replacing 200 ml of the sea water containing a Metridium by the same volume of 0·54 M KCl solution. There was no appreciable effect on either the respiratory rhythm or the depth of the deflexion during the 2 h for which the experiment was continued after the addition. However, the rate of O₂ depletion was reduced. The KCl caused the anemone to contract and remain contracted, thereby reducing its external surface area. O₂ uptake may also have been hindered by the secretion of mucus in the presence of excess KCl (Parker, 1905 a).

The K ⁺ concentration used (0·14 M) was greater than the 0·04 M found to enhance and prolong muscular contractions within 10 min in Calliactis by Ross & Pantin (1940). It also exceeded the 0· 08 M found by Ross (1960 a) to increase the frequency of inherent contractions of column preparations. It would thus appear that the rhythmical exhalations are not mediated by inherent muscular activity. The experiment was inconclusive as regards a possible ciliary reversal mechanism, because the K ⁺ concentration was only half that found by Parker (1905 a) to be most effective for inducing reversal. It was not possible to replace more than 200 ml of the sea water by isomotic KCl without the paraffin coming into contact with the anemone. Further experiments with more sophisticated apparatus need to be undertaken.

MgCl₂ has a powerful anaesthetizing action on sensory cells, the nerve net and neuromuscular junctions (Ross & Pantin, 1940; Ross, 1960a). According to Baba (1968) Yasuda found that the ciliary reversal induced by creatine on the pharyngeal lining of Actinia never occurs in sea water containing MgCl₂.

The effect of replacing 200 ml of the ambient water by 200 ml of 7 ’5 % MgCl _a solution (giving a final concentration of o·1 M-Mg ²+) is shown in Fig. 2F. The Metridium was contracted for the first 2 h, possibly accounting for the low O₂ depletion rate, but after the addition of the MgCl₂ it slowly relaxed. The O₂ uptake increased. It seems unlikely that an O₂ debt was being paid off after the period of contraction, for while this has been observed by Brafield & Chapman (1965) and Sassaman & Mangum (1972), in Metridium it only occurs after all detectable aerobic metabolism has ceased (Sassaman & Mangum, 1973). More probably the increase was the result of the change in body shape as the animal relaxed. Again there was no significant effect on the respiratory rhythm, but the depth of the deflexions was greater after the addition of the MgCl₂, perhaps because of an alteration in the spatial relationship between the mouth and electrode, or because of the relaxing action of Mg ²⁺. According to Ross (1960a) the inherent activity and responses of column preparations of Calliactis are inhibited to some extent by Mg ^24- at about o·1 M strength. While the absence of effect on the respiratory rhythm, even after h application, might suggest that the rhythm is not mediated by the nerve net, or by the SS2 system (McFarlane, 1973b), or even by ciliary reversal, it is probable that the animal was not fully anaesthetized by the 1:3 MgCl₂:sea water mixture used.

Our most interesting discovery is of the existence of a respiratory rhythm in Actinia and Metridium. These two anemones are dissimilar in that M. senile occurs sub-littorally and shows great variability in shape (Batham & Pantin, 1950a), whereas A. equina is a littoral species, which does not produce readily perceptible rhythmic contractions of the body-wall and which is not a particle feeder like Metridium. The occurrence of the rhythm in both species, with similar period, makes it likely that all epifaunal sea anemones would exhibit a similar rhythm. Whether the burrowing sea anemones do likewise is more difficult to predict. While the column of Haloclava producta shows the same rhythmic contractions as Metridium, it can irrigate its burrow by peristaltic movements so that much of the required O₂ is taken up through the relatively thin body-wall (Sassaman & Mangum, 1972). The tubicolous cerianthid Ceriantheopsis americanus, on the other hand, does not irrigate its tube, but appears to rely on its tentacular crown for most of its gaseous exchange with the environment (Sassaman & Mangum, 1974). Tentacular exchange was likewise concluded to be predominant in Metridium senile by Sassaman & Mangum (1972), in view of their inability to detect any periodic irrigation of the coelenteron. The existence of this factor puts some doubt on their conclusion. There is evidence that the gastrodermis of Anthozoa constitutes the major site of O₂ consumption (Brafield & Chapman, 1965; Sassaman & Mangum, 1972). One would thus expect the supply of O₂ to this tissue to be as direct as possible. The demonstration of the respiratory rhythm fulfils this expectation and brings sea anemones into line with the pennatulids, Pteroides griseum (Brafield & Chapman, 1967) and Renilla kolUkeri (Chapman, 1972). However, the period of the rhythm in these two species is shorter (7– 8 and 3– 6 min respectively) and the rhythm is only exhibited by well-expanded specimens.

In theory one would expect to be able to separate the two components of O₂ uptake, the ectodermal and the coelenteric on the rhythmical type of trace (Fig. 3). In practice the effect of mixing the exhaled and the ambient water is too slight to be precisely measurable. This is confirmed by calculation. Water enters the coelenteron of Metridium via the siphonoglyphs, the cilia of which beat continuously (Parker, 1905 a). Metridium usually has one (Stephenson, 1935) or two (Parker & Marks, 1928)siphonoglyphs. Assuming the latter, we calculate from the data given by Batham & Pantin (1950a) that the outflow per deflexion would be about 6 ml during the periods when the coelenteric volume remains constant. The O₂ concentration in the coelenteric fluid of a Metridium in fully aerated sea water (15 °C) would be 3·6 ppm according to the graph given by Sassaman & Mangum (1973). The calculated deflexion produced by mixing this fluid with the 800 ml of ambient water would be only 0·033 ppm, which is small when compared with the 0·25 ppm deflexions generally obtained.

However, the proportion of O₂ taken up via the coelenteron may not be negligible. If one assumes that the siphonoglyphal current is kept separate from the exhaled fluid, an input of 12 ml h ^-1 containing 8 ppm O₂ and an output of the same volume containing 3·6 ppm would supply 38 μlh ^-1 to the endoderm cells. This is about of the total uptake rate (80– 330μl g ^-1 h ^-1) obtained with a Metridium of 1 g dry weight. Differently directed currents can be kept separate in the coelenteric cavities of other coelenterates (Southward, 1955; Sassaman & Mangum, 1974). The depths of the deflexions on our records were much smaller than 3·6 ppm, but with an electrode diameter of 1 cm some mixing of the exhaled and ambient waters would have been inevitable.

The mechanism by which water is exhaled and the site of the pacemaker are quita unknown at present. It seems probable that the mouth opens slightly for 3·4 min and that, in Metridium at least, rhythmic contractions of the body-wall participate by raising the internal hydrostatic pressure. The significance of such contractions has not been established (McFarlane, 1974a). Needier & Ross (1958) observed on a specimen of Calliactis that the main contractions lasted for 3– 4 min and were repeated every 10– 15 min. From the changes in shape and volume they surmised that some water was expelled from the coelenteron at such times. This has yet to be proved. To date we have not attempted to record the contractions of the various muscles while also monitoring the respiratory rhythm. Nor have we determined the optimal conditions under which rhythmical exhalation occurs. Rhythmical contractile behaviour is complex, different patterns of activity being shown at different times depending upon the physiological state of the anemone and external stimulation (Batham & Pantin, 1950c; Needier & Ross, 1958). It would seem probable that the pattern of exhalations would likewise vary.

McFarlane (1973a) and Lawn (1976 a) have detected rhythmic changes in the frequency of action potentials associated with the SS2 system of Calliactis, with peaks of activity every 20– 25min (8– 14 °C). This system exhibits a marked increase in activity during the mouth opening and pharynx protrusion associated with feeding (McFarlane, 1975). In longitudinally bisected animals the SS2 system exhibits spontaneous activity (McFarlane, 1973 a). It could well be a candidate for the respiratory pacemaker.

The experiments with Actinia were initially undertaken to test whether the three or four distinct O₂ uptake rates observed by Newell & Northcroft (1967) could be correlated with particular types of activity. Various rates have been obtained, some in response to particular types of stimulation (for example, feeding or darkness). However, O₂ uptake is not necessarily the same as O₂ consumption in an animal capable of great changes in shape. Only when the anemones were fully acclimated and maintaining a state of uniform activity would this be true. Certainly there was no increase in O₂ depletion during contraction, which supports the contention of Sassaman & Mangum (1970) that the energy required would be too small to be detectable on records of O₂ uptake.