Discontinuous Carbon Dioxide Release and Metabolic Depression in Dormant Land Snails

Pulmonate land snails respond to dehydrating conditions by entering a state of quiescence (dormancy) in which the rates of oxygen consumption and evaporative water loss are greatly reduced. The ability to survive for months or years in this inactive condition permits snails to live in seasonally arid habitats in spite of their need for moisture during activity and feeding (reviewed by Nopp, 1974; Machin, 1975).

Ventilation of the snail lung occurs by diffusion across a closable pore, the pneumostome (Krogh, 1941). in the lung and haemolymph of Otala lactea increases and decreases during dormancy, indicating that specific ventilation (i.e. ventilation relative to metabolic rate) is reduced (Barnhart, 1986b). External gas exchange becomes discontinuous as the pneumostome, which is open nearly continuously in active snails, opens only intermittently during dormancy (Nopp, 1971). In addition, of dormant helicid snails shows large fluctuations in apparently quiescent and undisturbed individuals (Schmidt-Nielsen, Taylor & Shkolnik, 1971; Kratochvil, 1976). These bursts of O2 consumption are sustained for hours or days and are distinct from short-term changes in O2 uptake that result from intermittent breathing.

The retention of CO₂ in dormant snails strongly depresses extracellular pH (Burton, 1976; Barnhart, 1986a). One goal of the present study was to investigate the apparent correlation between CO₂ retention and reduction of metabolic rate during dormancy in Otala. Previous results indicated that fluctuations of in dormant Otala might be correlated with a cycle of CO₂ retention and release (Barnhart, 1986b). In the present study, flow-through respirometry was used to investigate the temporal patterns of O₂ uptake and CO₂ release and to estimate the changes which occur in whole-body CO₂ content during dormancy.

A second objective of this study was to compare the pattern of gas exchange of dormant snails with that-of insects. Certain large insects exhibit a discontinuous pattern of CO₂ release during rest or diapause, to which has been attributed a role in reduction of respiratory water loss (Miller, 1974, 1981). Although they are quite different morphologically, snails and insects encounter similar problems in resisting desiccation and starvation during prolonged dormancy. Comparison of the patterns of gas exchange in insects and snails is therefore of interest.

Snails were collected from an introduced population at Playa del Rey in Los Angeles County, California. The snails were transported to the University of Calgary and maintained in the laboratory on a diet of cabbage, carrots and Purina rat chow. Blackboard chalk was provided as a calcium source. Water and food were provided for approximately 1 week of each month. On this regimen, juvenile snails matured rapidly and only 5 of 160 adult specimens had died after 1 year of captivity. At least 2 weeks prior to respirometry, dormancy was induced by withholding food and water. Only adult snails having a fully formed shell were used for measurements. Whole mass (including shell) ranged between 7 and 10g (Table 1).

An open mask flow-through system was used for respirometry. A transparent plastic mask was sealed over the aperture of the shell with dental wax, leaving a space between the mask and the mantle collar of the snail. Air was drawn through the mask at 10 ml min⁻¹, and then successively through desiccant (Drierite), CO₂ analyser, O₂ analyser and pump. Stainless-steel tubing was used throughout, except for the glass desiccant tube (3 ml) and a 6-cm length of narrow-bore Tygon tubing connecting the mask and desiccant tube.

The instruments used were an Applied Electrochemistry S-3A O₂ analyser, CD-3A CO₂ analyser and R-2 flow controller. The gas analysers were calibrated with air and with O₂/CO₂/N₂ standards analysed with a Scholander device (Scholander, 1947). Outputs from the gas analysers were sampled at a frequency of about 50 Hz, digitized by a 12-bit a./d. converter, averaged over 5-s intervals and displayed and recorded automatically by a microcomputer. Air flow rate (10mlmin⁻¹) was measured downstream of the desiccant tube by means of a bubble flowmeter and was constant to within ±2% during the measurement periods.

Gas exchange was recorded in constant conditions of humidity, temperature and illumination. The snail with mask was placed on a lump of dental wax and covered by an inverted beaker, which was flushed continuously with air. Humidity in the beaker was controlled at 32 ± 2 % by bubbling the air through a saturated solution of MgC1₂ (Winston & Bates, 1960). Temperature was monitored with a thermocouple and remained at 24 ± 1 °C.

Changes in and of the airstream due to respiration were usually less than 0·3 vol%. Over several hours, calibration of the 0·3A may drift by up to 0·1 vol% at an of 20·95 vol%, so frequent checking of (baseline) was important for accuracy. The composition of air drawn through the mask when the pneumostome was closed did not differ measurably from ambient. Therefore, and during each ‘breath’ (opening of the pneumostome) were compared with adjacent records of and between breaths, (pneumostome closed) in calculating gas (exchange.

After baseline subtraction and synchronization, the and records were treated in two ways. First, in order to represent the rate of metabolic oxygen consumption more closely, a smoothed record of gas exchange was derived by integrating the data over successive, non-overlapping intervals which were of at least 30 min duration and included at least three breaths. Second, instantaneous rates (i.e. corrected for system washout characteristics) were calculated over the course of individual breaths in some records, in order to examine the dynamic aspects of diffusion gas exchange. Equations used were those of Bartholomew, Vleck & Vleck (1981). These calculations were also used to estimate the percentage of time that the pneumostome was open.

Following respirometry, the lung gas volume of each snail was measured by the method of Jones (1961). Briefly, the animal was placed in a closed, water-filled chamber connected with a manometer and a micrometer burette. The volume of the chamber was reduced by a known amount with the burette and the resulting change in pressure read from the manometer. The apparatus was calibrated with measured volumes of air. Precision was about ±5 μl

was often elevated for a day or two following placement of a dormant snail in the respirometry system. After this period of acclimation, long-term records of show a pattern of periodic elevation above a relatively stable minimum (Fig. 1). The regularity and frequency of the fluctuations varied considerably within and between individuals, but the minimum or ‘basal’ varied little (Table 1). Two individuals secreted calcareous epiphragms during respirometry (see below). In both cases, the periodic bursts of oxygen consumption appeared to be less frequent following this event (Fig. 1).

Simultaneous records of and other respiratory parameters in a representative individual are shown in detail in Fig. 2. When was high R remaine relatively close to RQ and calculated changes of whole-body CO₂ content were minor (Fig. 2C,D). As declined, R dropped below RQ and increased. Sub sequent periods of elevated always commenced with elevation of R and a large burst of net CO₂ release. During the longest periods of reduced ⁠, the increase of limited by small bursts of CO₂ release, which were accompanied by only a slight change of (e.g. last quarter of record in Fig. 2).

During periods of reduced and CO₂ accumulation the pneumostome was open only about 3–5% of the time (Fig. 2E). During these periods, breaths were brief (30–60s) and breath frequency was low (2–3 h⁻¹) (Fig. 3). In contrast, during periods of elevated and CO₂ release, breaths were more frequent and of longer duration (Fig. 3) and the pneumostome was open about 70% of the time (Fig. 2E).

Periods of increased ventilation and net CO₂ release began with a series of prolonged breaths (Fig. 3). Comparison of instantaneous and during sustained opening of the pneumostome (Fig. 3) shows that is initially highet than but declines much more rapidly than CO₂ release continues at a high rate due to the high capacitance of the body fluids for CO₂. Thus, brief breaths are associated with net CO₂ accumulation, and prolonged breaths with net CO₂ release.

was inversely and linearly related to the logarithm of as these parameters varied over time (Fig. 4). The correlation was highly significant (P< 0·001) in each of seven individuals tested. However, further analysis is difficult because the slope of y on logx depends on the absolute values of x. Absolute values of x (i.e. of ) in this case are unknown, because the method used for calculating measures only change and not absolute values (see Materials and Methods). varied about five-fold as ranged over about 4·3 mmol I⁻¹ (Table 1).

The effects of elevated external CO₂ (2%) and O₂ (42%) were tested in three individuals during periods of basal and CO₂ accumulation. Results were qualitatively similar in each individual. During exposure to 2% CO₂, was reduced by about half compared to in air (Fig. 5). This change apparently results from the reduced gradient for diffusion of CO₂ out of the lung. breath frequency and breath duration were not affected, indicating that 2% CO₂ had no immediate (2h) effect on the pattern of ventilation (i.e. pneumostome opening).

When snails were exposed to 42% O₂, during the interval including the first breath increased approximately three-fold, which may be attributed to the increased gradient for O₂ uptake (Fig. 6). Breath frequency then dropped by about half but O₂ uptake per breath remained elevated, so returned to the previous level. Breath duration was essentially unchanged. decreased by about half owing to reduced breath frequency. On return to air, O₂ uptake during the first breath dropped by about two-thirds, then returned to the previous level within a few breaths. was unchanged during the first breath, then rose gradually. Breath frequency increased two-to three-fold on return to air.

Similar tests during periods of elevated were inconclusive because of the high variability of respiration over time. However, in contrast to the results during low in 2 % CO₂ there was a large (about five-fold) increase in breath duration when was elevated.

Two instances of epiphragm secretion occurred during respirometry. Both events occurred during periods of elevated (Fig. 1) and in both instances gas exchange was apparently blocked for 1·5–2 h following secretion as the epiphragm dried (Fig. 7, see Discussion). The pattern of instantaneous gas exchange was otherwise apparently similar whether epiphragms were present or absent.

Periodic CO₂ release in dormant Otala was previously inferred from measurements of lung gas tensions and from closed-chamber respirometry (Barnhart, 1986b). In a broad sense, periodic or intermittent CO₂ release evidently occurs in a wide variety of animals including hibernating mammals (Bickler, 1984), reptiles (Ackerman & White, 1979; Glass & Johansen, 1979; Hicks & Riedesel, 1983) and in resting or dormant insects (see reviews by Miller, 1974; Edney, 1978) as well as in snails. Accumulation and release of CO₂ result from changes in the level of ventilation relative to the rate of aerobic metabolism (‘specific ventilation’), which alters partial pressures and concentrations of O₂ and CO₂ in the body fluids. Equal changes of partial pressure involve a much larger amount of CO₂ than O₂ because capacitance of body fluids is much higher for CO₂ (Dejours, 1981). Therefore, a change in specific ventilation may only slightly affect body O₂ content but cause a prolonged period of net CO₂ accumulation or release from the body fluids.

Periodic CO₂ release is most thoroughly described in diapause pupae of large saturniid moths, especially Hyalophora cecropia (see reviews by Miller, 1974; Edney, 1978). These pupae have a constant but release CO₂ in discrete bursts, often hours apart. The spiracles open slightly and close at high frequency (‘flutter’) during CO₂ accumulation but open widely and continuously during CO₂ release (Schneiderman, 1960). This pattern apparently results from differential sensitivity of ventilation to O₂ and CO₂. Low at the abdominal ganglia causes spiracle flutter (Burkett & Schneiderman, 1974) which keeps tracheal O₂ at about 3·5% (Levy & Schneiderman, 1966a). In contrast, CO₂ acts directly at the spiracles and triggers prolonged opening when tracheal CO₂ exceeds about 6% (Levy & Schneiderman, 1966a; see Miller, 1981, for discussion of CO₂ and ventilation in insects).

Gas exchange of pupae and snails appears similar in several respects. First, periodic CO₂ release in each is associated with dormancy and reduced ⁠. CO₂ release is relatively continuous when is high (Schneiderman & Williams, 1953, 1955; see Fig. 2). Second, ventilation of both pupae and snails evidently responds primarily to O₂ and not CO₂ during periods of CO₂ accumulation. Neither spiracle movement nor tracheal of pupae changes appreciably as CO₂ accumulates and tracheal increases (Schneiderman, 1960; Levy & Schneiderman, 1966a). Likewise, the pneumostome activity of Otala apparently remains constant as CO₂ accumulates during periods of low (Fig. 2E). Effects of experimental hypercapnia in Otala also suggest that ventilation is insensitive to CO₂ during periods of basal 2 % CO₂ impeded CO₂ diffusion out of the lung but had no obvious effect on O₂ uptake, breath frequency or duration (Fig. 5). In contrast, hyperoxia caused an immediate reduction of breath frequency and, consequently, of the rate of CO₂ release (Fig. 6).

In both pupae and snails, hypoventilation accompanies reduced -Thus, interburst tracheal in pupae decreases as decreases (Levy & Schneiderman, 1966a). In Otala, values in the lung gas and the haemolymph are greatly reduced when is low (Barnhart, 1986b). Interestingly, tarantulas also exhibit lower arterial at rest (28 Torr) than following activity (74 Torr) when is presumably higher (Angersbach, 1978). It appears that a change in specific ventilation may be a common mechanism for matching of O₂ transport to metabolic demand in terrestrial invertebrates. The rate of internal O₂ transport in these animals may be diffusion-limited, so that in the lung or tracheal gas affects the rate of O₂ transport. In Otala, O₂ transfer from lung to arterial haemolymph is strongly diffusion-limited (Barnhart, 1986b).

Based on the above observations, periodic CO₂ release in snails and pupae might be explained as follows. O₂ and CO₂ may both influence ventilation, to some degree independently. Fall of internal to a threshold level stimulates a pattern of ventilation (flutter in pupae, brief breaths in Otala) which maintains at or near the threshold level. When is low, the lung or tracheal necessary to supply respiration is reduced and hypoventilation ensues. Consequently, slowly increases towards a new equilibrium level as CO₂ accumulates. If exceeds a critical level, prolonged opening of the spiracles or pneumostome is induced and may reduce well below the triggering level. A periodic pattern of CO₂ release ensues. When is high, high is required in the lung or tracheae and the specific ventilation is therefore elevated. The equilibrium at this high specific ventilation will be low, and if it lies near or below the threshold, CO₂ release will be continuous.

An interesting question is whether the sensitivity of ventilation changes with metabolic state. varies between about 25 and 100Torr in dormant Otala, but is apparently controlled rather precisely at a much lower level in active snails. Moreover, specific ventilation of active snails changes with temperature, so and haemolymph pH change with temperature in the ‘alphastat’ pattern (Barnhart, 1986a,b).

Major episodes of CO₂ release in Otala were usually followed by prolonged increase of (Fig. 2). In contrast, saturniid pupae maintain constant (Schneiderman & Williams, 1953, 1955; Buck & Keister, 1955). Certain other insect pupae show infradian cycles of during diapause, but CO₂ release has not been investigated in those species (Denlinger, Willis & Fraenkel, 1972; Beck, 1980). Semiperiodic bursts of O₂ uptake appear to be characteristic of dormant helicid snails (see Schmidt-Nielsen et al. 1971; Kratochvil, 1976). Dormant snails are quite sensitive to disturbance and may increase in response to a variety of factors including vibration, light, temperature and humidity (Machin, 1975; Herreid & Rokitka, 1976). In specimens kept in constant conditions, simply passing a shadow across the shell affects respiration (M. C. Barnhart, personal observations) so that it is very difficult to rule out external influences. However, Schmidt-Nielsen et al. (1971) believed that bursts are an inherent feature of snail metabolism and not necessarily a response to external stimuli. Possibly bursts of oxygen uptake represent occasional attempts to arouse from dormancy, perhaps necessary for a full assessment of external conditions. The reduction of burst frequency following epiphragm secretion (Fig. 1) might indicate that frequent bursts are not a physiological necessity.

The mean and range of in dormant Otala (Table 1) are similar to those in diapausing cecropia pupae, i.e. 16·3μlg⁻¹ h⁻¹, 6·8–38μlg⁻¹ h⁻¹ at 25°C (Schneiderman & Williams, 1953). The ‘basal’ of dormant Otala is about 7% that of active snails at similar temperature (85 μlg⁻¹ h⁻¹; Barnhart, 1986zz). Peak of active Otala may be as high as 140μlg⁻¹ h⁻¹ (calculated from Herreid, 1977), so the factorial aerobic scope is about 24. Given the often low frequency of breaths, gas exchange should be used only cautiously as a measure of metabolic rate in dormant pulmonates. Care must be taken that the respirometry measurement period includes several breaths. Closed-chamber respirometry over 5-h periods gave results similar to those reported here (mean ⁠; Barnhart, 1986b).

During epiphragm formation a complex mucous layer is secreted on the mantle collar, then separated from it by expulsion of air from the lung. The mucus dries to form a membrane across the aperture of the shell and may attach the shell to the substrate. Epiphragms of Otala include a porous area which permits the diffusion of gases (Barnhart, 1983). However, the perforations form only as the epiphragm dries, and the freshly secreted mucus evidently prevents gas exchange (Fig. 7). The epiphragm has no obvious effect on the pattern of gas exchange after drying, as was predicted on the basis of its high permeability (Barnhart, 1983).

A negative correlation between and whole-body CO₂ content (Fig. 4) might arise by at least two mechanisms. First, reduced metabolic rate may induce hypoventilation and thereby cause CO₂ accumulation (see above). Second, accumulation of respiratory CO₂ might depress cellular metabolism, by affecting intracellular pH. Possibly both mechanisms operate simultaneously. Acid-base changes are associated with metabolic transitions in a variety of cells and organisms (Busa & Nuccitelli, 1984; Bickler, 1984). The idea that dormancy in snails might involve ‘CO₂ narcosis’is not new (Picher, 1971, and references therein) but little direct evidence is available. of non-dormant Otala is depressed by about half on exposure to 10% CO₂, coincident with depression of intracellular pH (M. C. Barnhart, unpublished).

In H. cecropia pupae, the volume deficit that results from low R causes negative pressure in the tracheal system. Opening of the spiracles during flutter is sufficiently small and brief that the pressure gradient is maintained and causes air to flow into the spiracles (suction or passive ventilation) which may further enhance retention of CO₂ and water vapour (Levy & Schneiderman, 1966b ; Kestler, 1985). The degree to which CO₂ retention is enhanced by suction ventilation is not clear. Assuming diffusion only, the lowest R predicted in cecropia pupae is about 0’3 (calculated from tracheal partial pressures in table 2 of Levy & Schneiderman, 1966a). Measured R of cecropia pupae during CO₂ accumulation is 0·17 and ranges from 0·01 to 0·56 (RQ = 0·78) (Schneiderman & Williams, 1955).

Reduction of lung volume probably occurs in dormant Otala between breaths (Barnhart, 1986b). Compared to insect spiracles, however, the pneumostome opens very widely (about 1–3 mm²; M. C. Barnhart, personal observations) and breath duration is seldom less than 30 s. Any pressure gradient that develops during apnoea must be abolished almost immediately after the pneumostome opens. The lowest R observed in the present study was about 0–22, which is similar to the lowest predicted R based on measurements of and an the lung of dormant Otala and assuming gas exchange by diffusion alone (Barnhart, 1986b). Thus, it appears that suction ventilation may be negligible in snails.

Dormancy probably first developed in littoral marine ancestors of land snails (see Little, 1983) and primitively appears to be a strategy for avoiding desiccation. Respiration is significant in this regard. In Otala, the lung is a major potential site of water loss; the total rate of water loss in dormancy would be at least 54% higher without hypoventilation, and 460 % higher if the aerobic metabolic rate were not also reduced (Barnhart, 1986b).

It is often stated that suction ventilation reduces water loss in insects by retarding efflux of water vapour from the spiracles while enhancing O₂ uptake, and that this effect provides an adaptive advantage for periodic CO₂ release (Buck, 1958; Miller, 1974, 1981 ; Edney, 1978). Such an interpretation is probably misleading, however. It should be emphasized that, considering the entire respiratory cycle, the need for CO₂ release, not O₂ uptake, appears to limit the extent to which ventilation and consequent water loss can be reduced. The maximum partial pressure gradient for CO₂ loss is only about one-third that for O₂ uptake in cecropia pupae (Levy & Schneiderman, 1966a). Apparently, therefore, the minimum conductance necessary for CO₂ release (i.e. to keep tolerably low) is roughly three times greater than that needed for O₂ uptake (assuming RQ =1). Suction ventilation does not enhance CO₂ release relative to water loss; water vapour and CO₂ presumably encounter similar convective and diffusive conductance across the spiracles. The ratio of water vapour loss to CO₂ release must depend directly on their respective partial pressure gradients. Assuming a constant pressure gradient for water loss, the ratio will be low when the gradient is large. In this respect, there is no inherent advantage to suction ventilation or periodic CO₂ release. In fact, a constant minimum level of ventilation that maintained the highest tolerable internal would theoretically result in less water loss than the periodic pattern, in which specific ventilation fluctuates (see Kestler, 1985, p. 183).

There appear to be only two basic mechanisms for reducing respiratory water loss. The rate of aerobic metabolism may be reduced, and/or the specific ventilation may be reduced, with consequent decrease of internal and elevation of Intermittent CO₂ release in insects and snails appears to be a consequence of these changes and of characteristics of ventilatory control.

The technical advice of J. R. B. Lighten is gratefully acknowledged. MCB was supported by a postdoctoral fellowship from the Alberta Heritage Foundation for Medical Research.