- Email: [email protected]
Oxygen supply and limiting oxygen pressures in an effect larva
121
Respiration Physiology (1985) 60, 121-134 Elsevier
OXYGEN SUPPLY AND LIMITING OXYGEN PRESSURES IN AN INSECT LARVA
S.M. T E N N E Y Department of Physiology, Dartmouth Medical School Hanover, NH 03756, U.S.A.
Abstract. Larvae of the moth, Carpocapsa saltitans, demonstrate a diurnal activity pattern of rhythmic twitching which, under conditions of controlled light and temperature, is characterized by a predictable frequency and regularity. The twitching activity is shown to be sensitive to the partial pressure of environmental oxygen, and it ceases altogether at a particular Po2 called 'critical'. Use is made of the 'critical' Po2 in normobaric and hypobaric conditions to deduce the roles of diffusion and convection in the larval oxygen transport mechanisms; and also as a value for the total decrement of Po2 from ambient air to mitochondria, in order to evaluate predicted values based on calculations of resistance to oxygen flow. For this latter study 'porosity' of the larva and the seed pod in which it is normally housed was inferred from measured rates of water vapor loss, and oxygen uptake rates of the larvae were measured by the manometric technique of Warburg. Applying these data to a model system the conclusion was reached that almost the total resistance to oxygen flow is at the spiracle.
Carpocapsa saltitans Critical Po2 Diurnal twitching activity
Insect Larva Metabolic rate
Oxygen diffusion Oxygen transport Water loss
Respiration in insects is sustained by a system of air-filled tubes which connect with the exterior of the organism through a valved aperture, the spiracle, and extend interiorly as a branching tree, the tracheal system, progressing to tracheoles and ending ultimately in air capillaries that may actually penetrate the interior of cells. Oxygen is transported in the gas phase over almost the entire distance to tissue level, except for the terminal branches which are liquid filled, and in non-flying insects the process is thought to be predominantly by diffusion (Miller, 1964). Analysis of the problem of gas exchange in insects has attracted the attention of physiologists for many years, but controversy remains regarding the magnitudes of oxygen gradients in the system, the extent to which convective air movement supplements diffusion, and the locations of the principal resistances to oxygen flow (Buck, 1962). Accepted for publication 19 January 1985 0034-5687/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)
122
s.M. TENNEY
The idea that experimental determinations of 'critical' oxygen pressures would provide a useful stratagem for approaching some problems of insect respiration was adopted with a rationale as follows. Although tissue Po2 is not known under normal conditions, if the ambient Po2 is progressively lowered to the point (the 'critical' Po2) that respiration cannot be sustained at a normal level it can be assumed that the Po2 in the microenvironment of mitochondria will have fallen to a value less than 0.1 torr (10-7 M O2) (Chance et aL, 1973). By this means two key values are known: that at the 'sink', i.e. mitochondria, which is approximately zero; that at the 'source', i.e., ambient, which is determinable under different experimental conditions. Thus, the total decrement of 0 2 pressure in the system is known when the organism is just at the 'critical' Po2. With that information, together with other measurable data, predictions based on several hypotheses can be evaluated. Further, the question of convective vs diffusive transport is amenable to experimental testing by determination of the 'critical' Po2 under both hypobaric and normobaric conditions. Because the diffusion coefficient of oxygen in the gas phase varies inversely with the barometric pressure, oxygen flow by diffusion can be maintained at a lower APo~ in hypobaria, and, therefore, the 'critical' Po2 ought to be lower in hypobaria than in normobaria. However, if the flow is by convection there should be no difference between the hypobaric and normobaric 'critical' Po~. The search for an appropriate species of insect was guided by the requirement that it manifest some trait that is reproducible, quantifiable and dependent on oxygen. The choice was the larva of the coddling moth, Carpocapsa saltitans, a parasitic inhabitant of the seed pods of a desert shrub, Sebastiana pringlei. The remarkable behavioral trait exhibited by this creature, relevant for the research questions posed, is a habit of regularly twitching its whole body, so as to throw it against the wall of the seed pod and cause it to move with a sudden jerk. These seemingly animated seeds, hopping on the desert floor, are called 'brincadores' by the small boys of Chihuahua, but are known in the United States as Mexican jumping beans. Preliminary experiments showed the hopping property of these larvae to be sensitive to their oxygen environment. The following studies were undertaken: (1)determine the normal characteristics of twitch frequency and the factors that influence it, such as light and temperature, in order to standardize these conditions when ambient Po2 was the experimental variable; (2) determine 'critical' Po2 for normal twitch behavior, comparing hypobaric and normobaric hypoxia; (3) determine oxygen uptake rates of larva; (4) determine 'porosity' of larva and seed pod from rates of evaporative water loss; (5) analyze theoretical models of insect oxygen delivery systems, incorporating physiological and morphological data, to predict oxygen gradients for comparison with measured values.
Materials and methods
Larvae of the moth Carpocapsa saltitans, encased in their natural habitat (see fig. 1) the seed pods of Sebastiana pringlei, were harvested in June and studied over the following
INSECT RESPIRATION
123
Fig. 1. Opened seed pod to show larva resident within. three months. Three populations, representing the c r o p o f each o f three consecutive years, were collected, a n d the experiments were c o n d u c t e d seasonally three years in a row. A v e r a g e physical characteristics are shown in table 1. A seed with its e n c a s e d larva was p l a c e d in a glass dish (2 c m diameter) that rested on a piezo crystal detector. E a c h twitch o f the larva created a signal in the crystal that p a s s e d through an integrating circuit a n d was r e c o r d e d as a blip on a strip-chart recorder. T h e integrator was designed to reset its count at zero after each fifty impulses, and amplitude was adjusted so that fifty impulses p r o d u c e d a full vertical scale deflection on the chart. A 10 L steel, sealed c h a m b e r h o u s e d all o f the a p p a r a t u s , except the strip-chart recorder, and it served either as an environmental c h a m b e r when it was flushed with TABLE 1 Average physical characteristics of larva and seed pod
Weight Length Radius Surface area Thickness
Larva
Seed pod
0.0445 g 0.85 cm 0.125 cm 0.67 cm2 0.005 cm (cuticle)
0.156 g 1.2 cm 0.6 cm 2.2 cm2 0.02 cm
124
S.M. TENNEY
gas mixtures of known composition, or as a low-pressure chamber when it was decompressed by a vacuum pump. The chamber was kept at constant temperature by a temperature-controlled water jacket, and at controlled light intensity by an external lamp whose light was directed through a plexiglass window also used for visual monitoring. Chamber pressure was monitored with a mercury manometer and chamber temperature with a thermistor. Chamber oxygen concentration was adjusted by regulating flow of oxygen and nitrogen from compressed sources through rotameters with periodic chemical checks with a Scholander micro-gas analyzer. Fo2 in the chamber was monitored directly with an S-3A oxygen analyzer (Applied Electrochemistry, Inc.). The procedure followed for determinations of 'critical' Po~ was first to establish in a preliminary trial run the approximate value; then, after 20 min of recovery in normoxia the chamber was decompressed moderately rapidly (about 20 min) to a pressure that represented a Po~ about 10 torr above the expected value of the 'critical' Po~. After a few minutes' pause at that stage to insure that the activity pattern continued normally, the chamber was slowly decompressed by stages of 5 torr (representing about 1 torr Po2) to watch for the fu'st signs of slowing of activity. When that was observed, the chamber pressure was held steady for sufficient time to determine whether activity would cease. Usually, it would be sustained, and thereupon the chamber was further decompressed and at one or two more stages activity would cease abruptly. That would be recorded as the 'critical' Po~. After a 10 rain wait, if there was no sign of recurrence of activity the chamber was recompressed. Recovery always ensued. The procedure for one determination required about 3 h. Experiments conducted under normobaric conditions were performed by slowly diluting the chamber air with nitrogen and watching for end points comparable to those observed under hypobaria. Oxygen consumption rate of larvae in air was determined on pooled samples of 10 at controlled temperatures in air by the manometric technique of Warburg (Dixon, 1943). Conductances for water vapor were inferred from measurements of weight loss of individuals maintained at 30 °C in an anhydrous jar. Measurements were made daily for 3 weeks. Larvae were divided into separate groups: one pool remained normally housed in their seed pods; the other pool were removed from their seed pods and studied naked. From each pool groups were composed for studies of weight loss when the jar was filled with air, or with hypoxic mixtures. Water conductance of the seed pod was determined after the larva had been removed via a small hole and replaced with water soaked cotton pledget. The hole was sealed with wax and rate of weight loss determined from daily weight measurements, as above. Calculations of rate of loss of water vapor, water conductance, equivalent pore area, and oxygen conductance followed the principles and equations established by Paganelli and Rahn for their studies of egg shells (Paganelli, 1980).
125
INSECT RESPIRATION
Results
General observations. Activity monitored over 24 h periods indicated a clearly established diurnal pattern with quiescence throughout the period of darkness (fig. 2). Artificial darkness in daytime arrested activity, and artificial light at nighttime initiated activity. During daylight hours activity averaged about 1100 twitches per hour with brief, periodic 'rest periods'. There was individual variation in the activity rate, but each insect maintained its inherent frequency throughout the day (+ 10%) and from day to day (+ 10%). Activity showed a strong temperature dependency, the Q~o of twitch frequency being 2.2 for the range 25-35 ° C. At still higher temperatures twitch frequency did not increase further. The pattern of reasonable constancy of twitch frequency and its reproducibility provided encouragement for the possibility that a major change of frequency attending some experimental manipulation would represent a significant event and not a random occurrence. 'Critical' Po:. When ambient Po2 was slowly reduced the influence of hypoxia on twitching behavior first became apparent as a slowing of frequency, followed by cessation of activity when the Po~ was lowered slightly further. Figure 3 illustrates a single experiment in which determinations were made on two larvae simultaneously, with progressive normobaric hypoxia. At an earlier stage an approximate 'critical' Po2 had been determined, and then the environment was returned to normoxic conditions. That portion of the record has been removed and is not shown in fig. 3. The illustration begins (at the left) where the ambient Po2 has been reduced again to about 50 torr 2000
T:25' C Natural Light
1500
." / 3.-" f.
I000 /
E
/
Z
: :
- "
/
500
\J
\!'
i
.: /
:
/ -
".\
_;.., ; I
o 5
i
i
i
o~
i
i
i
i
~ ~
i
i
i
i
~
~
i
i
i
o
o
i
i
i
i
oo o
i
i
i
i
o Clock Time ,~
Fig. 2. Diurnal activity pattern of Carpocapsa saltitans. Rateofjumping, quantitatedon the ordinate, reflects rate of whole body muscle twitching by larva. Activities of four individual specimens are portrayed, one of which was observed for only half a day.
126
s.M. TENNEY
./-- ......... S/
// / / J
NORMOBARIC
' L i i IVii~
T = 30°C PB = 736 TORR
O
i
TORR2O3o i
i
IOMIN
f~ iJ i
Fig. 3. Illustrative experimental determination of 'critical' Po2 at 30 °C. Tracings at top and bottom are kymograph records of integrated twitching activity of two larvae studied simultaneously. Each spike represents 50jumps. The trace of Po2 indicates values recorded in the chamber during the experiment. Activity is seen to slow at about 30 torr and to cease at 20 torr, which marks the 'critical' Po2. Recovery occurs when Po2 is returned to 30 torr.
(off-scale on the portion of the scale shown). As the Po2 is lowered further the larva, whose activity is recorded on the top tracing, begins to slow its twitch frequency at about 35 torr, while the larva in the bottom trace slows at about 30 torr, but in both activity ceases altogether at 20 torr. Recovery in both occurs at about 30 torr. The end points were unequivocal in all instances, and all specimens were tested for recovery following elevation of ambient Po2, either by raising pressure in the chamber in the hypobaric experiment, or adding oxygen to the inflow mixture of the chamber in the normobaric experiment. No specimen failed to recover, but the Po2 required to restart twitch activity was usually a few torr higher than that which had been found to arrest it. The group mean 'critical' Po~ ( + 1 SD) determined at 30 °C under hypobaric conditions was 17.4 + 13.5 torr (N = 81) and under normobaric conditions, 22.4 + 6.7 torr (N = 28). The difference of the means was not statistically significant. In 8 specimens both hypobaric and normobaric 'critical' Po~ values were measured in the same day, and the mean of the difference (hypobaric P o : - normobaric Po2) was found to be 8.8 + 7.0 torr. By Student's t-test for paired values this difference was significant at the 0.01 level. Determinations made at 25 °C and 40 °C failed to establish any clear relationship between 'critical' Po~ and temperature. Oxygen uptake rates (/Vlo2) were determined on larvae that had been removed from the seed pod (in this condition they do not twitch, but they do squirm) and on larvae encased in their seed pods (twitching continued, and the metabolic rate
Metabolic rate.
INSECT RESPIRATION
127
TABLE 2 Oxygen uptake rates of larvae of Carpocapsa saltitans Ambient temp. (° C)
N
1VIo2(#L'g ' •min- 1) (mean _+SE)
25 32 32 39 39
81 82 191 77 164
11.3 + 0.9 (encased) 15.0 + 1.4 (naked) 20.3 + 1.5 (encased) 14.7 + 1.0 (naked) 13.3 + 0.7 (encased)
was therefore, 'active'). T a b l e 2 presents the m e a s u r e d values./rio2 at 32 ° C o f e n c a s e d larvae was 35~o i n c r e a s e d over that m e a s u r e d on the n a k e d larvae, and p r o b a b l y indicative o f the increment in energy expenditure a s s o c i a t e d with increased m o t o r activity. T e m p e r a t u r e d e p e n d e n c e o f oxygen u p t a k e rate shows the peculiar pattern c o m m o n l y e n c o u n t e r e d in insect m e t a b o l i s m : an increase in the lower t e m p e r a t u r e range, but as the t e m p e r a t u r e is elevated further the rate decreases. This effect abolished the difference o f Nlo2, c o m p a r i n g e n c a s e d and n a k e d larvae at 39 °C. Rate o f water vapor loss. W a t e r loss (/VIH20) m e a s u r e d as weight loss at 30 °C was d e t e r m i n e d for the p u r p o s e o f estimating permeability and inferred 'porosity', p a r a meters n e e d e d for subsequent calculations with m o d e l s o f oxygen delivery. The primary d a t a are given in table 3. Average MH2 o (cm 3. sec - 1) ( S T P D ) o f larvae in 21 ~ 0 2 was 4.1 x 1 0 - 6 ; in 7~o 02, 4.7 x 1 0 - 6 ; a n d in 5~o 02, 6.0 × 10 - 6 . A s s u m i n g that the pressure o f w a t e r v a p o r in the a n h y d r o u s j a r was zero, a n d that the air in the organism at the i m m e d i a t e site o f entry was fully saturated at b o d y temperature, average c o n d u c t a n c e ( G , 2 o ) in air calculates to be 2.3 x 1 0 - 7 c m 3 . s e c - l . t o r r - 1 , and for larval surface area of 0.67 c m 2, specific conductance is, 3.4 x 1 0 - 7 c m 3 . sec - 1. c m - 2. t o r r - 1. TABLE 3 Rates of water loss by larvae of Carpocapsa saltitans (T = 30 °C) Ambient Po2 (torr)
N
155
40
55
30
37
30
/VIH20(rag. day- ', Mean + SD) (Max - Min) 0.287 +_0.082 (9.0 - ~0) 0.329 + 0.090 (8.9 - 1.6) 0.413 + 0.092 (9.4 - ~0)
128
s.M. TENNEY
No significant difference was observed in the rate of water loss of larvae kept normally housed in the seed pods, and those that were removed and studied thus artificially exposed. The essential observation regarding these values, which are the averages for a 30-day period, is that they are not representative of all time spans. It was a characteristic of every specimen to lose weight rapidly during the first 3 days in the anhydrous environment, and following this, the rate slowed pronouncedly, beginning at the time the larva had lost about 10~o body weight. The maximum and minimum values shown in table 3 indicate the observed range, but the finding was consistent that the maximum values were always early and the minimum, always late. The final rate of change was often just perceptible, but there was no question of the viability of the organisms. When the measurements were made in hypoxic environments the initial rates of weight loss were usually greater than observed in normoxia, but the major feature was that they remained elevated for many days beyond the time when the rates in normoxic specimens had begun to slacken. Water vapor loss from the seed pod alone was 8.5 x 10 4 cm 3 . sec i (STPD) at 30 °C, giving a conductance of 2.7 × 10 - 5 cm 3 . sec - ~ • torr - ~, and a specific conductance of 1.23 x 10- 5 cm 3 . sec - 1 cm 2. t o r r - 1 .
Discussion
General
The curious motor behavior of the larva of Carpocapsa saltitans is mentioned in the literature of natural history but seems not to have been the subject of any published physiological study. In the present investigation intriguing properties, like light sensitivity, were examined only insofar as it was necessary to appreciate their significance for controls in experiments designed to test sensitivity to hypoxia. The Q~o of 2.2 for twitch frequency is in a similar range to what is found for chirping, heart beat and respiratory frequencies in insects (Crozier and Stier, 1925). The extreme tolerance for hypobaric hypoxia observed in Carpocapsa saltitans is consistent with experimental observations on insects, generally, dating from the 17th Century (Boyle, 1670) to the present (Wellington, 1946; Mani, 1968). The point at which oxygen deficiency becomes manifest varies among insect species and depends also on stage of development and ambient temperature, but it is frequently observed that a beginning decline in oxygen consumption does not occur until oxygen concentrations as low as 3-5Yo are reached (Keister and Buck, 1964). The effect of oxygen concentration on oxygen uptake rate was not measured in the present study, but was inferred from a failure of rhythmic twitching. The average 'critical' Po2 of about 20 torr found by this means is in line with the Po2 at which oxygen consumption begins to fall, the conventional criterion, as cited above.
Assessment of convection and/or diffusion in 02 transport. Flying insects ventilate their airways, but diffusion is probably the sole mechanism of oxygen transport in non-flying
INSECT RESPIRATION
129
insects (Miller, 1964). The frequent gross body twitches of the larva of Carpocapsa saltitans, however, raise the question whether these movements serve effectively to ventilate the tracheae, as well as to stir the air in the seed pod, which is highly likely. An approach to the question about ventilation can be made by comparing the 'critical' Po2 in hypobaric and normobaric conditions. There should be no difference if convection is the sole means of oxygen transport in the gas phase, but a lower Po2 should be found in hypobaria if diffusion is the transport mechanism, because hypobaria increases the diffusion constant. A comparison of group mean values showed no significant difference, but when the same individual was tested under both conditions, and any difference between hypobaric and normobaric 'critical' Po2 was noted, the mean of those differences was positive and significant ( + 8.8 torr). The conservative conclusion to be drawn from these results is that diffusion in the gas phase is an operating mechanism for Oz transport, but a contributing role of ventilation is not excluded. Of course, transport in the liquid phase is not influenced by barometric pressure, hence these results have no bearing on that component of the oxygen delivery system. A major objective of this study was to determine the decrement of oxygen pressure from spiracle to mitochondria, based on the latter being close to zero when spiracular (ambient) Po2 was reduced to the 'critical' value, and with that total decrement known, to analyze the properties of the oxygen transport system. An ambient Po2 of 20 torr having been found the minimum compatible with normal motor behavior, the conclusion is that 20 torr is the total pressure loss in the oxygen conductance path, and the problem is to account for the resistances in the system that impede the flow of oxygen.
Oxygen uptake rate.
Measured IVlo2of the larvae of Carpocapsa saltitans (table 2) was slightly higher than reported for larvae of other species (Keister and Buck, 1964) all of which were larger and probably less active. There was a temperature effect observed in their experiments which showed an increased lVlo2 in the range 20-30 °C, but there was a maximum at 35 °C and thereafter, lVlo2 declined. This is almost identical with the results obtained in the present experiments. The metabolic rates reported in this paper were determined in air and are probably higher than what would be found in hypoxia. This characteristic of oxygen conformers would introduce a complication in subsequent calculations, but it is noteworthy that the activity of the larvae remained high to within a few torr of the 'critical' Po2, beyond which the metabolic rate has been assumed to approach zero. In any event, the subsequent studies reported in this work, based on I(/lo2 measured in air, and at 30 °C, represent the case for maximum oxygen demand.
A cylinder of tissue without airways.
The physical dimensions of a larva (table 1) and its level of oxygen requirement (table 2) lead immediately to the conclusion that oxygen supply by diffusion through the tissue, without benefit of an airway system, cannot suffice with normal external oxygen pressures. Treating the larva as a cylinder, and applying an appropriate diffusion equation (e.g., eq. 5 in Fenn, 1927) incorporating a
130
S.M. TENNEY
value of Krogh's diffusion constant for tissue of 2.3 × 10 - 7 ml. c m - 1 sec - ~• atm (Thorpe and Crisp, 1947) it is found that an external Po2 in excess of 3 atmospheres would be required to prevent the existence of an anoxic core. The situation with this hypothetical model becomes even worse with the knowledge that oxygen permeability of insect cuticle is less than 1/10th that of water or muscle tissue (Buck, 1962; Ebeling, 1964). An internal distribution system in which oxygen is transported principally in the gas phase is the universal solution adopted in insect respiratory system design. .
Porosity ofseedpod and larval suoCace. From the measurements of rate of water loss (MH~o) across seed pod surface or from the larva, pore area, Ap, can be calculated from the following equation (Paganelli, 1980): ]~IH20
=
Du~o. AP. APH2o = GH~o" PH20 RT
L
(1)
where DH2o is the diffusion coefficient for water, R is the gas constant, T the absolute temperature, L thickness of surface, and APH~o the difference of water vapor pressure between the gas space just inside the spiracle (saturated at body temperature) and the external environment. GH:O is water vapor conductance. For the seed pod, the measured MH2o, leads to a calculated value of Ap/L = 0.080cm, which, for average thickness (L) of 0.02cm, results in Ap = 16 x 10- 4 cm 2. For the estimated total surface area of an average seed pod, this means that 0.07~o is occupied by pores. By comparison, Ap of a chicken's egg is 23 x 10 3 cm 2 (Wangensteen et aL, 1970/71) which, for an average total shell surface area of 68 cm 2, indicates that 0.03 ~o of the total is occupied by pores. The seed pod is a relatively leaky sieve and it is not likely that it would provide any significant protection for the larva against dehydration. The data comparing rate of water loss of naked larvae with encapsulated larvae (table 3) bears this out, since the two rates are similar. Water loss from the larvae was rapid for the first 3 days, then the rate slowed gradually to very low values by the end of two or three weeks - a pattern similar to that noted for Agriotes larvae at 30 °C by Wigglesworth (1945). Using the average value of I~IH2o taken over the full time span (table 3), eq. (1) can be solved for Ap/L for the larva, and it is found to be 38.6 x 10-5 cm. Estimated thickness of the cuticle is 0.005 cm; therefore, calculated Ap = 1.93 x 10 - 6 c m 2, which is 0.0003 ~o of the total body surface area of the larva. Since l(/IH2o early in the period of observations was 2-3 times greater than the mean, Ap would calculate to be proportionately greater at that time. In fact, Ap becomes, 6.1 x 10 6 c m 2. Water loss through the cuticle is unlikely due to the impermeability of its wax to water (Wigglesworth, 1945) and therefore, the inference is that the spiracles are the sole port of exchange. Ap must be, by this argument, the total area of the spiracular apertures, and the variable dimensions calculated at different times (early vs late) reflects the activity of the muscular valve in the spiracle to control the size of the opening (Hoyle, 1959). Clearly, the valve closes as water is lost and
INSECT RESPIRATION
:V
131
\
6
E
~. 4 2 O0
~
20
410
a
60
L
80
L _
I00
i
120
J
140
Po2 (Tort)
Fig. 4. Calculated mean pore area (Ap) of larva (assumed to be total area of spiracular apertures) as influenced by ambient Po2 at time of measurement. Bars are + 1 SD. Dotted line is extrapolation of best-fit French curve that connected points. Measurements could not be made at lowest Po2 because of dehydration.
dehydration threatens. The larger Ap, deduced from the larger/VlH20, in hypoxia (fig. 4) is probably also a manifestation of the local control mechanism of the spiracle which is known to be responsive to low 0 2 (Hoyle, 1960; Miller, 1964). The highest value of Ap shown (at lowest Po2) was not measured, because the organisms could not be kept viable over a long enough period of time with that degree of hypoxia. The curve based on the measured points was therefore extrapolated to the minimal Po2. The calculated value can be compared with morphometric estimates of size of spiracle opening, which was 6 x 10- 3 cm 2. There is a 1000-fold difference in the two estimates, but the anatomical value is bound to be maximal, probably, supramaximal, because the spiracle is functionally controlled by a valve which normally reduces the opening to a narrow slit, but the morphometric value is from the diameter of the external port. The conclusion is that the larva must normally maintain the spiracles in an almost closed condition. An alternative model would portray the spiracle as intermittently opened and closed (discontinuous respiration) but since the experimental measurements lead to time-averaged 'equivalent spiracular area', no clear distinction can be made between a maintained slit and a winking aperture. Choice between the models does not seriously affect the subsequent calculations unless it is assumed that oxidative metabolism shuts down during the time of spiracular closure. Since the hopping frequency is usually much higher than that reported for spiracular opening intermittency (Buck and Keister, 1955) that possibility seems unlikely, although it does happen that interburst spiracular flutter may reach high frequencies (Schneiderman, 1960). The necessity for conserving water accounts for the organism effectively sealing itself off from the environment, even though the inward flow of oxygen will be simultaneously impeded.
Estimated gradients of Po2. Rewriting eq. (1) for /rio 2 _ Do2 ' Ap .APo 2 = Go2. APo2 RT L
0 2
instead of
H20"
(2)
132
S.M. TENNEY
and solving for APo2, using values of 1~1o2from table 2, and the value of Ap/L already calculated, the oxygen gradient across the seed pod is found to be 0.02 torr. The seed pod surface is, therefore, an insignificant resistance for oxygen flow from environment to the larva. It is also unlikely that any gradients exist in the air space of the seed since the frequent larval movements must stir the contained gas. The central finding of this study has been that an oxygen gradient, measured to be not less than 20 torr (the average 'critical' Po2) between environment and muscle mitochondria is required to sustain spontaneous motor activity. In a system operating by diffusion that gradient would be the consequence of resistances to oxygen flow located at the spiracular aperture, along the trachea and tracheolar airway system, and across the tissue barrier. The latter two have generally been regarded as negligible (Krogh, 1920; Thorpe and Crisp, 1947; Weis-Fogh, 1964) and recently, Scheid et al. (1982) measured the elimination of low solubility gases from the pupa of the silkworm and, on the basis of the monoexponential curves they obtained, concluded that the main gas phase diffusive resistance must be at the spiracular valve. The results of the present experiments point to the same conclusion, but it remains to establish a consistency between the quantitative data concerning oxygen flow through the spiracle and the measured APo2 of 20 torr. Introducing the measured values of larval 1~1o~and Ap/L in air into eq. (2) and solving for APo2 the calculated value is found to be 47 torr - a result clearly incompatible with the measured APo2 of 20 tort. However, the relevant Ap in the calculation ought to be the one based on the measurement of water vapor conductance in the extreme hypoxic environment, because the 'critical' Po2 was determined in those conditions. Extrapolation of the curve based on the experimental water vapor conductance data to an ambient Po2 of 20 torr (measurements were not possible) would indicate that hypoxia of that degree doubles the calculated Ap. The APo2 under hypoxic conditions then becomes 23 torr, which is a reasonable agreement between prediction and experiment and that would mean that the spiracles account for almost the entire decrement of pressure in the oxygen path. The fact that 1Vlo: in hypoxia is not known is troublesome, because if it is less than the normoxic value used the effect would be to lower the calculated APo~. There are no data to settle the doubt, but it has already been remarked that the activity pattern remains so lively to within a few torr of the 'critical' Po2 as to make it unlikely that 1Vlo~ was very much reduced to that point. No morphometric measurements of the conducting air channels in Carpocapsa saltitans were made in this study, and the only complete description in the literature is that of Nunome (1944) on larvae of Bombyx mori. Assuming that the branching arrangement and distribution of size found by Nunome (1944) in the silkworm is applicable to other species, and if it is generally the rule that tracheolar volume is about 48/~L. g 1 body mass (Bridges et al., 1980), an equivalent circuit can be designed and examined (e.g., with models described by Weis-Fogh, 1964) for pressure drop necessary to maintain longitudinal flow of oxygen at the rate observed in the metabolic studies on Carpocapsa saltitans. Over the distance from just inside the spiracle to center of the larva (0.125 cm) APo~ in the airway system calculates to be less than 2 toi'r. This trivial
INSECT RESPIRATION
133
gradient illustrates the efficiency of diffusion in the gas phase, but there is still the tissue barrier to be crossed, and remembering that the conductivity of oxygen in liquids is about 10 - 5 that in air, distances are of critical importance. The smallest tracheoles are probably filled with liquid, so that even the penetration of some cells by air capillaries is of limited significance (but the conductance of oxygen in water is higher than in tissue) and the question remains, what APo2 is required to maintain oxygen flow across that terminal resistance? The majority of workers have considered the tissue gradient to be negligible, and if Weis-Fogh's (1964) 'hole fraction' (i.e., summed cross-sectional area of air tubes per unit cross-sectional area of tissue) of 10 - ~to 10 - 3 is applicable to larval tissue of Carpocapsa saltitans, the interfacial area for diffusion becomes very large, and diffusional distances in the tissue parenchyma are much reduced. Intertracheolar distance is probably not more than 2 ktm (Buck, 1962) thus making the most remote target of oxygen consumption about 1/~m removed from the tracheolar source of oxygen supply in a symmetrical array oftracheoles. Radial diffusion of oxygen over this distance and for the highest metabolic rates shown in table 2 requires a APo2 much less than 1 torr, as calculated by the Krogh-Erlang equation. Even for a 10/~m separation the pressure gradient is still less than 1 torr. In summary, almost the entire loss of oxygen pressure head occurs at the spiracular aperture, and that loss is calculated to be 23 torr, which is in good agreement with the experimentally determined average 'critical' Po2 of about 20 torr, since no further significant decrement occurs in either the airway system or in the tissues.
Acknowledgements The author thanks Kathy Franklin for her expert technical assistance and the Chapparal Novelties Company of Alamogordo, NM, for their generous contribution. This work was supported by Grant H L 02888-27 of the National Heart, Lung and Blood Institute of the National Institutes of Health.
References Boyle, R. (1670). New pneumatical experiments about respiration. Trans. R. Soc. (London) 5:2011-2031. Bridges, C.R., P, Kestler and P. Scheid (1980). Tracheal volume in the pupa of the Saturniid Moth Hyalophora Cecropia determined with inert gases. Respir. Physiol. 40: 281-291. Buck, J. and M.L. Keister (1955). Cyclic CO2 release in diapausing Agapema pupae. Biol. Bull. Mar. Biol. Lab. Woods Hole 109: 144-163. Buck, J. (1962). Some physical aspects of insect respiration. Annu. Rev. EntomoL 7: 27-56. Chance, B., N. Oshino, T. Sugano and A. Mayevsky (1973). Basic principles of tissue oxygen determination from mitochondrial signals. Adv. Exp. Med. Biol. 37a: 277-292. Crozier, W.J. and T.B. Stier (1925). Critical thermal increments for rhythmic respiratory movements in insects. J. Gen. Physiol. 7: 429-447. Dixon, M. (1943). Manometric Methods. 2nd ed. Cambridge, Cambridge University Press; New York, MacMillan Co., 155 p.
134
S.M. TENNEY
Ebeling, W. (1964). The permeability of insect cuticle. In: The Physiology of Insecta. V. 1I, edited by M. Rockstein. New York - London, Academic Press, pp. 507-556. Fenn, W. O. (1927). The oxygen consumption of frog nerve during stimulation. J. Gen. Physiol. 10: 767-779. Hoyle, G. (1959). The neuromuscular mechanism of an insect spiracular muscle. J. Insect Physiol. 3: 378-394. Hoyle, G. (1960). The action of carbon dioxide gas on an insect spiracular muscle. J. Insect Physiol. 4: 63-79. Keister, M. and J. Buck (1964). Respiration: Some exogenous and endogenous effects on rate of respiration. In: The Physiology of lnsecta, vol. II, edited by M. Rockstein. New York - London, Academic Press, pp. 617-658. Krogh, A. (1920). Studien uber Tracheenrespiration. Ober Gasdiffusion in den Tracheen. Pfliigers Arch. Gesamte Physiol. Menschen Tiere 179:95-112. Mani, M. S. (1968). Ecology and Biogeography of High Altitude Insects. The Hague, W. Junk, 137 p. Miller, P. L. (1964). Respiration - aerial gas transport. In: The Physiology of Insecta. Vol. II, edited by M. Rockstein. New York - London, Academic Press, pp. 557-615. Nunome, Z. (1944). Studies on the respiration of the silkworm. 1. Diffusion of oxygen in the respiratory system of the silkworm. II. Several measurements in the respiratory organ. Bull. Sericulr Exp. Stn. Japan 12: 17-90. Paganelli, C.V. (1980). The physics of gas exchange across the avian egg shell. Am. Zool. 20: 329-338. Scheid, P., C. Hook and C. R. Bridges (1982). Diffusion in gas exchange in insects. Fed. Proc. 41: 2143-2145. S chneiderman, H. A. (1960). Discontinuous respiration in insects: role of the spiracles. Biol. Bull. Mar. Biol. Lab. Woods Hole 119: 494-528. Thorpe, W.H. and D.J. Crisp (1947). Studies on plastron respiration. II. The respiratory efficiency of the plastron in Aphelocheirus. J. Exp. Biol. 24: 270-303. Wangensteen, O. D., D. Wilson and H. Rahn (1970/71). Diffusion of gases across the shell of the hen's egg. Respir. Physiol. 11: 16-30. Weis-Fogh, T. (1964). Diffusion in insect wing muscle, the most active tissue known. J. Exp. Biol. 41: 229-256. Wellington, W. G. (1946). The effects of variations in atmospheric pressure upon insects. Can. J. Res. Zool. Soc. 24: 51-70. Wigglesworth, V.B. (1945). Transpiration through the cuticle of insects. J. Exp. Biol. 21:97-114.
Respiration Physiology (1985) 60, 121-134 Elsevier
OXYGEN SUPPLY AND LIMITING OXYGEN PRESSURES IN AN INSECT LARVA
S.M. T E N N E Y Department of Physiology, Dartmouth Medical School Hanover, NH 03756, U.S.A.
Abstract. Larvae of the moth, Carpocapsa saltitans, demonstrate a diurnal activity pattern of rhythmic twitching which, under conditions of controlled light and temperature, is characterized by a predictable frequency and regularity. The twitching activity is shown to be sensitive to the partial pressure of environmental oxygen, and it ceases altogether at a particular Po2 called 'critical'. Use is made of the 'critical' Po2 in normobaric and hypobaric conditions to deduce the roles of diffusion and convection in the larval oxygen transport mechanisms; and also as a value for the total decrement of Po2 from ambient air to mitochondria, in order to evaluate predicted values based on calculations of resistance to oxygen flow. For this latter study 'porosity' of the larva and the seed pod in which it is normally housed was inferred from measured rates of water vapor loss, and oxygen uptake rates of the larvae were measured by the manometric technique of Warburg. Applying these data to a model system the conclusion was reached that almost the total resistance to oxygen flow is at the spiracle.
Carpocapsa saltitans Critical Po2 Diurnal twitching activity
Insect Larva Metabolic rate
Oxygen diffusion Oxygen transport Water loss
Respiration in insects is sustained by a system of air-filled tubes which connect with the exterior of the organism through a valved aperture, the spiracle, and extend interiorly as a branching tree, the tracheal system, progressing to tracheoles and ending ultimately in air capillaries that may actually penetrate the interior of cells. Oxygen is transported in the gas phase over almost the entire distance to tissue level, except for the terminal branches which are liquid filled, and in non-flying insects the process is thought to be predominantly by diffusion (Miller, 1964). Analysis of the problem of gas exchange in insects has attracted the attention of physiologists for many years, but controversy remains regarding the magnitudes of oxygen gradients in the system, the extent to which convective air movement supplements diffusion, and the locations of the principal resistances to oxygen flow (Buck, 1962). Accepted for publication 19 January 1985 0034-5687/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)
122
s.M. TENNEY
The idea that experimental determinations of 'critical' oxygen pressures would provide a useful stratagem for approaching some problems of insect respiration was adopted with a rationale as follows. Although tissue Po2 is not known under normal conditions, if the ambient Po2 is progressively lowered to the point (the 'critical' Po2) that respiration cannot be sustained at a normal level it can be assumed that the Po2 in the microenvironment of mitochondria will have fallen to a value less than 0.1 torr (10-7 M O2) (Chance et aL, 1973). By this means two key values are known: that at the 'sink', i.e. mitochondria, which is approximately zero; that at the 'source', i.e., ambient, which is determinable under different experimental conditions. Thus, the total decrement of 0 2 pressure in the system is known when the organism is just at the 'critical' Po2. With that information, together with other measurable data, predictions based on several hypotheses can be evaluated. Further, the question of convective vs diffusive transport is amenable to experimental testing by determination of the 'critical' Po2 under both hypobaric and normobaric conditions. Because the diffusion coefficient of oxygen in the gas phase varies inversely with the barometric pressure, oxygen flow by diffusion can be maintained at a lower APo~ in hypobaria, and, therefore, the 'critical' Po2 ought to be lower in hypobaria than in normobaria. However, if the flow is by convection there should be no difference between the hypobaric and normobaric 'critical' Po~. The search for an appropriate species of insect was guided by the requirement that it manifest some trait that is reproducible, quantifiable and dependent on oxygen. The choice was the larva of the coddling moth, Carpocapsa saltitans, a parasitic inhabitant of the seed pods of a desert shrub, Sebastiana pringlei. The remarkable behavioral trait exhibited by this creature, relevant for the research questions posed, is a habit of regularly twitching its whole body, so as to throw it against the wall of the seed pod and cause it to move with a sudden jerk. These seemingly animated seeds, hopping on the desert floor, are called 'brincadores' by the small boys of Chihuahua, but are known in the United States as Mexican jumping beans. Preliminary experiments showed the hopping property of these larvae to be sensitive to their oxygen environment. The following studies were undertaken: (1)determine the normal characteristics of twitch frequency and the factors that influence it, such as light and temperature, in order to standardize these conditions when ambient Po2 was the experimental variable; (2) determine 'critical' Po2 for normal twitch behavior, comparing hypobaric and normobaric hypoxia; (3) determine oxygen uptake rates of larva; (4) determine 'porosity' of larva and seed pod from rates of evaporative water loss; (5) analyze theoretical models of insect oxygen delivery systems, incorporating physiological and morphological data, to predict oxygen gradients for comparison with measured values.
Materials and methods
Larvae of the moth Carpocapsa saltitans, encased in their natural habitat (see fig. 1) the seed pods of Sebastiana pringlei, were harvested in June and studied over the following
INSECT RESPIRATION
123
Fig. 1. Opened seed pod to show larva resident within. three months. Three populations, representing the c r o p o f each o f three consecutive years, were collected, a n d the experiments were c o n d u c t e d seasonally three years in a row. A v e r a g e physical characteristics are shown in table 1. A seed with its e n c a s e d larva was p l a c e d in a glass dish (2 c m diameter) that rested on a piezo crystal detector. E a c h twitch o f the larva created a signal in the crystal that p a s s e d through an integrating circuit a n d was r e c o r d e d as a blip on a strip-chart recorder. T h e integrator was designed to reset its count at zero after each fifty impulses, and amplitude was adjusted so that fifty impulses p r o d u c e d a full vertical scale deflection on the chart. A 10 L steel, sealed c h a m b e r h o u s e d all o f the a p p a r a t u s , except the strip-chart recorder, and it served either as an environmental c h a m b e r when it was flushed with TABLE 1 Average physical characteristics of larva and seed pod
Weight Length Radius Surface area Thickness
Larva
Seed pod
0.0445 g 0.85 cm 0.125 cm 0.67 cm2 0.005 cm (cuticle)
0.156 g 1.2 cm 0.6 cm 2.2 cm2 0.02 cm
124
S.M. TENNEY
gas mixtures of known composition, or as a low-pressure chamber when it was decompressed by a vacuum pump. The chamber was kept at constant temperature by a temperature-controlled water jacket, and at controlled light intensity by an external lamp whose light was directed through a plexiglass window also used for visual monitoring. Chamber pressure was monitored with a mercury manometer and chamber temperature with a thermistor. Chamber oxygen concentration was adjusted by regulating flow of oxygen and nitrogen from compressed sources through rotameters with periodic chemical checks with a Scholander micro-gas analyzer. Fo2 in the chamber was monitored directly with an S-3A oxygen analyzer (Applied Electrochemistry, Inc.). The procedure followed for determinations of 'critical' Po~ was first to establish in a preliminary trial run the approximate value; then, after 20 min of recovery in normoxia the chamber was decompressed moderately rapidly (about 20 min) to a pressure that represented a Po~ about 10 torr above the expected value of the 'critical' Po~. After a few minutes' pause at that stage to insure that the activity pattern continued normally, the chamber was slowly decompressed by stages of 5 torr (representing about 1 torr Po2) to watch for the fu'st signs of slowing of activity. When that was observed, the chamber pressure was held steady for sufficient time to determine whether activity would cease. Usually, it would be sustained, and thereupon the chamber was further decompressed and at one or two more stages activity would cease abruptly. That would be recorded as the 'critical' Po~. After a 10 rain wait, if there was no sign of recurrence of activity the chamber was recompressed. Recovery always ensued. The procedure for one determination required about 3 h. Experiments conducted under normobaric conditions were performed by slowly diluting the chamber air with nitrogen and watching for end points comparable to those observed under hypobaria. Oxygen consumption rate of larvae in air was determined on pooled samples of 10 at controlled temperatures in air by the manometric technique of Warburg (Dixon, 1943). Conductances for water vapor were inferred from measurements of weight loss of individuals maintained at 30 °C in an anhydrous jar. Measurements were made daily for 3 weeks. Larvae were divided into separate groups: one pool remained normally housed in their seed pods; the other pool were removed from their seed pods and studied naked. From each pool groups were composed for studies of weight loss when the jar was filled with air, or with hypoxic mixtures. Water conductance of the seed pod was determined after the larva had been removed via a small hole and replaced with water soaked cotton pledget. The hole was sealed with wax and rate of weight loss determined from daily weight measurements, as above. Calculations of rate of loss of water vapor, water conductance, equivalent pore area, and oxygen conductance followed the principles and equations established by Paganelli and Rahn for their studies of egg shells (Paganelli, 1980).
125
INSECT RESPIRATION
Results
General observations. Activity monitored over 24 h periods indicated a clearly established diurnal pattern with quiescence throughout the period of darkness (fig. 2). Artificial darkness in daytime arrested activity, and artificial light at nighttime initiated activity. During daylight hours activity averaged about 1100 twitches per hour with brief, periodic 'rest periods'. There was individual variation in the activity rate, but each insect maintained its inherent frequency throughout the day (+ 10%) and from day to day (+ 10%). Activity showed a strong temperature dependency, the Q~o of twitch frequency being 2.2 for the range 25-35 ° C. At still higher temperatures twitch frequency did not increase further. The pattern of reasonable constancy of twitch frequency and its reproducibility provided encouragement for the possibility that a major change of frequency attending some experimental manipulation would represent a significant event and not a random occurrence. 'Critical' Po:. When ambient Po2 was slowly reduced the influence of hypoxia on twitching behavior first became apparent as a slowing of frequency, followed by cessation of activity when the Po~ was lowered slightly further. Figure 3 illustrates a single experiment in which determinations were made on two larvae simultaneously, with progressive normobaric hypoxia. At an earlier stage an approximate 'critical' Po2 had been determined, and then the environment was returned to normoxic conditions. That portion of the record has been removed and is not shown in fig. 3. The illustration begins (at the left) where the ambient Po2 has been reduced again to about 50 torr 2000
T:25' C Natural Light
1500
." / 3.-" f.
I000 /
E
/
Z
: :
- "
/
500
\J
\!'
i
.: /
:
/ -
".\
_;.., ; I
o 5
i
i
i
o~
i
i
i
i
~ ~
i
i
i
i
~
~
i
i
i
o
o
i
i
i
i
oo o
i
i
i
i
o Clock Time ,~
Fig. 2. Diurnal activity pattern of Carpocapsa saltitans. Rateofjumping, quantitatedon the ordinate, reflects rate of whole body muscle twitching by larva. Activities of four individual specimens are portrayed, one of which was observed for only half a day.
126
s.M. TENNEY
./-- ......... S/
// / / J
NORMOBARIC
' L i i IVii~
T = 30°C PB = 736 TORR
O
i
TORR2O3o i
i
IOMIN
f~ iJ i
Fig. 3. Illustrative experimental determination of 'critical' Po2 at 30 °C. Tracings at top and bottom are kymograph records of integrated twitching activity of two larvae studied simultaneously. Each spike represents 50jumps. The trace of Po2 indicates values recorded in the chamber during the experiment. Activity is seen to slow at about 30 torr and to cease at 20 torr, which marks the 'critical' Po2. Recovery occurs when Po2 is returned to 30 torr.
(off-scale on the portion of the scale shown). As the Po2 is lowered further the larva, whose activity is recorded on the top tracing, begins to slow its twitch frequency at about 35 torr, while the larva in the bottom trace slows at about 30 torr, but in both activity ceases altogether at 20 torr. Recovery in both occurs at about 30 torr. The end points were unequivocal in all instances, and all specimens were tested for recovery following elevation of ambient Po2, either by raising pressure in the chamber in the hypobaric experiment, or adding oxygen to the inflow mixture of the chamber in the normobaric experiment. No specimen failed to recover, but the Po2 required to restart twitch activity was usually a few torr higher than that which had been found to arrest it. The group mean 'critical' Po~ ( + 1 SD) determined at 30 °C under hypobaric conditions was 17.4 + 13.5 torr (N = 81) and under normobaric conditions, 22.4 + 6.7 torr (N = 28). The difference of the means was not statistically significant. In 8 specimens both hypobaric and normobaric 'critical' Po~ values were measured in the same day, and the mean of the difference (hypobaric P o : - normobaric Po2) was found to be 8.8 + 7.0 torr. By Student's t-test for paired values this difference was significant at the 0.01 level. Determinations made at 25 °C and 40 °C failed to establish any clear relationship between 'critical' Po~ and temperature. Oxygen uptake rates (/Vlo2) were determined on larvae that had been removed from the seed pod (in this condition they do not twitch, but they do squirm) and on larvae encased in their seed pods (twitching continued, and the metabolic rate
Metabolic rate.
INSECT RESPIRATION
127
TABLE 2 Oxygen uptake rates of larvae of Carpocapsa saltitans Ambient temp. (° C)
N
1VIo2(#L'g ' •min- 1) (mean _+SE)
25 32 32 39 39
81 82 191 77 164
11.3 + 0.9 (encased) 15.0 + 1.4 (naked) 20.3 + 1.5 (encased) 14.7 + 1.0 (naked) 13.3 + 0.7 (encased)
was therefore, 'active'). T a b l e 2 presents the m e a s u r e d values./rio2 at 32 ° C o f e n c a s e d larvae was 35~o i n c r e a s e d over that m e a s u r e d on the n a k e d larvae, and p r o b a b l y indicative o f the increment in energy expenditure a s s o c i a t e d with increased m o t o r activity. T e m p e r a t u r e d e p e n d e n c e o f oxygen u p t a k e rate shows the peculiar pattern c o m m o n l y e n c o u n t e r e d in insect m e t a b o l i s m : an increase in the lower t e m p e r a t u r e range, but as the t e m p e r a t u r e is elevated further the rate decreases. This effect abolished the difference o f Nlo2, c o m p a r i n g e n c a s e d and n a k e d larvae at 39 °C. Rate o f water vapor loss. W a t e r loss (/VIH20) m e a s u r e d as weight loss at 30 °C was d e t e r m i n e d for the p u r p o s e o f estimating permeability and inferred 'porosity', p a r a meters n e e d e d for subsequent calculations with m o d e l s o f oxygen delivery. The primary d a t a are given in table 3. Average MH2 o (cm 3. sec - 1) ( S T P D ) o f larvae in 21 ~ 0 2 was 4.1 x 1 0 - 6 ; in 7~o 02, 4.7 x 1 0 - 6 ; a n d in 5~o 02, 6.0 × 10 - 6 . A s s u m i n g that the pressure o f w a t e r v a p o r in the a n h y d r o u s j a r was zero, a n d that the air in the organism at the i m m e d i a t e site o f entry was fully saturated at b o d y temperature, average c o n d u c t a n c e ( G , 2 o ) in air calculates to be 2.3 x 1 0 - 7 c m 3 . s e c - l . t o r r - 1 , and for larval surface area of 0.67 c m 2, specific conductance is, 3.4 x 1 0 - 7 c m 3 . sec - 1. c m - 2. t o r r - 1. TABLE 3 Rates of water loss by larvae of Carpocapsa saltitans (T = 30 °C) Ambient Po2 (torr)
N
155
40
55
30
37
30
/VIH20(rag. day- ', Mean + SD) (Max - Min) 0.287 +_0.082 (9.0 - ~0) 0.329 + 0.090 (8.9 - 1.6) 0.413 + 0.092 (9.4 - ~0)
128
s.M. TENNEY
No significant difference was observed in the rate of water loss of larvae kept normally housed in the seed pods, and those that were removed and studied thus artificially exposed. The essential observation regarding these values, which are the averages for a 30-day period, is that they are not representative of all time spans. It was a characteristic of every specimen to lose weight rapidly during the first 3 days in the anhydrous environment, and following this, the rate slowed pronouncedly, beginning at the time the larva had lost about 10~o body weight. The maximum and minimum values shown in table 3 indicate the observed range, but the finding was consistent that the maximum values were always early and the minimum, always late. The final rate of change was often just perceptible, but there was no question of the viability of the organisms. When the measurements were made in hypoxic environments the initial rates of weight loss were usually greater than observed in normoxia, but the major feature was that they remained elevated for many days beyond the time when the rates in normoxic specimens had begun to slacken. Water vapor loss from the seed pod alone was 8.5 x 10 4 cm 3 . sec i (STPD) at 30 °C, giving a conductance of 2.7 × 10 - 5 cm 3 . sec - ~ • torr - ~, and a specific conductance of 1.23 x 10- 5 cm 3 . sec - 1 cm 2. t o r r - 1 .
Discussion
General
The curious motor behavior of the larva of Carpocapsa saltitans is mentioned in the literature of natural history but seems not to have been the subject of any published physiological study. In the present investigation intriguing properties, like light sensitivity, were examined only insofar as it was necessary to appreciate their significance for controls in experiments designed to test sensitivity to hypoxia. The Q~o of 2.2 for twitch frequency is in a similar range to what is found for chirping, heart beat and respiratory frequencies in insects (Crozier and Stier, 1925). The extreme tolerance for hypobaric hypoxia observed in Carpocapsa saltitans is consistent with experimental observations on insects, generally, dating from the 17th Century (Boyle, 1670) to the present (Wellington, 1946; Mani, 1968). The point at which oxygen deficiency becomes manifest varies among insect species and depends also on stage of development and ambient temperature, but it is frequently observed that a beginning decline in oxygen consumption does not occur until oxygen concentrations as low as 3-5Yo are reached (Keister and Buck, 1964). The effect of oxygen concentration on oxygen uptake rate was not measured in the present study, but was inferred from a failure of rhythmic twitching. The average 'critical' Po2 of about 20 torr found by this means is in line with the Po2 at which oxygen consumption begins to fall, the conventional criterion, as cited above.
Assessment of convection and/or diffusion in 02 transport. Flying insects ventilate their airways, but diffusion is probably the sole mechanism of oxygen transport in non-flying
INSECT RESPIRATION
129
insects (Miller, 1964). The frequent gross body twitches of the larva of Carpocapsa saltitans, however, raise the question whether these movements serve effectively to ventilate the tracheae, as well as to stir the air in the seed pod, which is highly likely. An approach to the question about ventilation can be made by comparing the 'critical' Po2 in hypobaric and normobaric conditions. There should be no difference if convection is the sole means of oxygen transport in the gas phase, but a lower Po2 should be found in hypobaria if diffusion is the transport mechanism, because hypobaria increases the diffusion constant. A comparison of group mean values showed no significant difference, but when the same individual was tested under both conditions, and any difference between hypobaric and normobaric 'critical' Po2 was noted, the mean of those differences was positive and significant ( + 8.8 torr). The conservative conclusion to be drawn from these results is that diffusion in the gas phase is an operating mechanism for Oz transport, but a contributing role of ventilation is not excluded. Of course, transport in the liquid phase is not influenced by barometric pressure, hence these results have no bearing on that component of the oxygen delivery system. A major objective of this study was to determine the decrement of oxygen pressure from spiracle to mitochondria, based on the latter being close to zero when spiracular (ambient) Po2 was reduced to the 'critical' value, and with that total decrement known, to analyze the properties of the oxygen transport system. An ambient Po2 of 20 torr having been found the minimum compatible with normal motor behavior, the conclusion is that 20 torr is the total pressure loss in the oxygen conductance path, and the problem is to account for the resistances in the system that impede the flow of oxygen.
Oxygen uptake rate.
Measured IVlo2of the larvae of Carpocapsa saltitans (table 2) was slightly higher than reported for larvae of other species (Keister and Buck, 1964) all of which were larger and probably less active. There was a temperature effect observed in their experiments which showed an increased lVlo2 in the range 20-30 °C, but there was a maximum at 35 °C and thereafter, lVlo2 declined. This is almost identical with the results obtained in the present experiments. The metabolic rates reported in this paper were determined in air and are probably higher than what would be found in hypoxia. This characteristic of oxygen conformers would introduce a complication in subsequent calculations, but it is noteworthy that the activity of the larvae remained high to within a few torr of the 'critical' Po2, beyond which the metabolic rate has been assumed to approach zero. In any event, the subsequent studies reported in this work, based on I(/lo2 measured in air, and at 30 °C, represent the case for maximum oxygen demand.
A cylinder of tissue without airways.
The physical dimensions of a larva (table 1) and its level of oxygen requirement (table 2) lead immediately to the conclusion that oxygen supply by diffusion through the tissue, without benefit of an airway system, cannot suffice with normal external oxygen pressures. Treating the larva as a cylinder, and applying an appropriate diffusion equation (e.g., eq. 5 in Fenn, 1927) incorporating a
130
S.M. TENNEY
value of Krogh's diffusion constant for tissue of 2.3 × 10 - 7 ml. c m - 1 sec - ~• atm (Thorpe and Crisp, 1947) it is found that an external Po2 in excess of 3 atmospheres would be required to prevent the existence of an anoxic core. The situation with this hypothetical model becomes even worse with the knowledge that oxygen permeability of insect cuticle is less than 1/10th that of water or muscle tissue (Buck, 1962; Ebeling, 1964). An internal distribution system in which oxygen is transported principally in the gas phase is the universal solution adopted in insect respiratory system design. .
Porosity ofseedpod and larval suoCace. From the measurements of rate of water loss (MH~o) across seed pod surface or from the larva, pore area, Ap, can be calculated from the following equation (Paganelli, 1980): ]~IH20
=
Du~o. AP. APH2o = GH~o" PH20 RT
L
(1)
where DH2o is the diffusion coefficient for water, R is the gas constant, T the absolute temperature, L thickness of surface, and APH~o the difference of water vapor pressure between the gas space just inside the spiracle (saturated at body temperature) and the external environment. GH:O is water vapor conductance. For the seed pod, the measured MH2o, leads to a calculated value of Ap/L = 0.080cm, which, for average thickness (L) of 0.02cm, results in Ap = 16 x 10- 4 cm 2. For the estimated total surface area of an average seed pod, this means that 0.07~o is occupied by pores. By comparison, Ap of a chicken's egg is 23 x 10 3 cm 2 (Wangensteen et aL, 1970/71) which, for an average total shell surface area of 68 cm 2, indicates that 0.03 ~o of the total is occupied by pores. The seed pod is a relatively leaky sieve and it is not likely that it would provide any significant protection for the larva against dehydration. The data comparing rate of water loss of naked larvae with encapsulated larvae (table 3) bears this out, since the two rates are similar. Water loss from the larvae was rapid for the first 3 days, then the rate slowed gradually to very low values by the end of two or three weeks - a pattern similar to that noted for Agriotes larvae at 30 °C by Wigglesworth (1945). Using the average value of I~IH2o taken over the full time span (table 3), eq. (1) can be solved for Ap/L for the larva, and it is found to be 38.6 x 10-5 cm. Estimated thickness of the cuticle is 0.005 cm; therefore, calculated Ap = 1.93 x 10 - 6 c m 2, which is 0.0003 ~o of the total body surface area of the larva. Since l(/IH2o early in the period of observations was 2-3 times greater than the mean, Ap would calculate to be proportionately greater at that time. In fact, Ap becomes, 6.1 x 10 6 c m 2. Water loss through the cuticle is unlikely due to the impermeability of its wax to water (Wigglesworth, 1945) and therefore, the inference is that the spiracles are the sole port of exchange. Ap must be, by this argument, the total area of the spiracular apertures, and the variable dimensions calculated at different times (early vs late) reflects the activity of the muscular valve in the spiracle to control the size of the opening (Hoyle, 1959). Clearly, the valve closes as water is lost and
INSECT RESPIRATION
:V
131
\
6
E
~. 4 2 O0
~
20
410
a
60
L
80
L _
I00
i
120
J
140
Po2 (Tort)
Fig. 4. Calculated mean pore area (Ap) of larva (assumed to be total area of spiracular apertures) as influenced by ambient Po2 at time of measurement. Bars are + 1 SD. Dotted line is extrapolation of best-fit French curve that connected points. Measurements could not be made at lowest Po2 because of dehydration.
dehydration threatens. The larger Ap, deduced from the larger/VlH20, in hypoxia (fig. 4) is probably also a manifestation of the local control mechanism of the spiracle which is known to be responsive to low 0 2 (Hoyle, 1960; Miller, 1964). The highest value of Ap shown (at lowest Po2) was not measured, because the organisms could not be kept viable over a long enough period of time with that degree of hypoxia. The curve based on the measured points was therefore extrapolated to the minimal Po2. The calculated value can be compared with morphometric estimates of size of spiracle opening, which was 6 x 10- 3 cm 2. There is a 1000-fold difference in the two estimates, but the anatomical value is bound to be maximal, probably, supramaximal, because the spiracle is functionally controlled by a valve which normally reduces the opening to a narrow slit, but the morphometric value is from the diameter of the external port. The conclusion is that the larva must normally maintain the spiracles in an almost closed condition. An alternative model would portray the spiracle as intermittently opened and closed (discontinuous respiration) but since the experimental measurements lead to time-averaged 'equivalent spiracular area', no clear distinction can be made between a maintained slit and a winking aperture. Choice between the models does not seriously affect the subsequent calculations unless it is assumed that oxidative metabolism shuts down during the time of spiracular closure. Since the hopping frequency is usually much higher than that reported for spiracular opening intermittency (Buck and Keister, 1955) that possibility seems unlikely, although it does happen that interburst spiracular flutter may reach high frequencies (Schneiderman, 1960). The necessity for conserving water accounts for the organism effectively sealing itself off from the environment, even though the inward flow of oxygen will be simultaneously impeded.
Estimated gradients of Po2. Rewriting eq. (1) for /rio 2 _ Do2 ' Ap .APo 2 = Go2. APo2 RT L
0 2
instead of
H20"
(2)
132
S.M. TENNEY
and solving for APo2, using values of 1~1o2from table 2, and the value of Ap/L already calculated, the oxygen gradient across the seed pod is found to be 0.02 torr. The seed pod surface is, therefore, an insignificant resistance for oxygen flow from environment to the larva. It is also unlikely that any gradients exist in the air space of the seed since the frequent larval movements must stir the contained gas. The central finding of this study has been that an oxygen gradient, measured to be not less than 20 torr (the average 'critical' Po2) between environment and muscle mitochondria is required to sustain spontaneous motor activity. In a system operating by diffusion that gradient would be the consequence of resistances to oxygen flow located at the spiracular aperture, along the trachea and tracheolar airway system, and across the tissue barrier. The latter two have generally been regarded as negligible (Krogh, 1920; Thorpe and Crisp, 1947; Weis-Fogh, 1964) and recently, Scheid et al. (1982) measured the elimination of low solubility gases from the pupa of the silkworm and, on the basis of the monoexponential curves they obtained, concluded that the main gas phase diffusive resistance must be at the spiracular valve. The results of the present experiments point to the same conclusion, but it remains to establish a consistency between the quantitative data concerning oxygen flow through the spiracle and the measured APo2 of 20 torr. Introducing the measured values of larval 1~1o~and Ap/L in air into eq. (2) and solving for APo2 the calculated value is found to be 47 torr - a result clearly incompatible with the measured APo2 of 20 tort. However, the relevant Ap in the calculation ought to be the one based on the measurement of water vapor conductance in the extreme hypoxic environment, because the 'critical' Po2 was determined in those conditions. Extrapolation of the curve based on the experimental water vapor conductance data to an ambient Po2 of 20 torr (measurements were not possible) would indicate that hypoxia of that degree doubles the calculated Ap. The APo2 under hypoxic conditions then becomes 23 torr, which is a reasonable agreement between prediction and experiment and that would mean that the spiracles account for almost the entire decrement of pressure in the oxygen path. The fact that 1Vlo: in hypoxia is not known is troublesome, because if it is less than the normoxic value used the effect would be to lower the calculated APo~. There are no data to settle the doubt, but it has already been remarked that the activity pattern remains so lively to within a few torr of the 'critical' Po2 as to make it unlikely that 1Vlo~ was very much reduced to that point. No morphometric measurements of the conducting air channels in Carpocapsa saltitans were made in this study, and the only complete description in the literature is that of Nunome (1944) on larvae of Bombyx mori. Assuming that the branching arrangement and distribution of size found by Nunome (1944) in the silkworm is applicable to other species, and if it is generally the rule that tracheolar volume is about 48/~L. g 1 body mass (Bridges et al., 1980), an equivalent circuit can be designed and examined (e.g., with models described by Weis-Fogh, 1964) for pressure drop necessary to maintain longitudinal flow of oxygen at the rate observed in the metabolic studies on Carpocapsa saltitans. Over the distance from just inside the spiracle to center of the larva (0.125 cm) APo~ in the airway system calculates to be less than 2 toi'r. This trivial
INSECT RESPIRATION
133
gradient illustrates the efficiency of diffusion in the gas phase, but there is still the tissue barrier to be crossed, and remembering that the conductivity of oxygen in liquids is about 10 - 5 that in air, distances are of critical importance. The smallest tracheoles are probably filled with liquid, so that even the penetration of some cells by air capillaries is of limited significance (but the conductance of oxygen in water is higher than in tissue) and the question remains, what APo2 is required to maintain oxygen flow across that terminal resistance? The majority of workers have considered the tissue gradient to be negligible, and if Weis-Fogh's (1964) 'hole fraction' (i.e., summed cross-sectional area of air tubes per unit cross-sectional area of tissue) of 10 - ~to 10 - 3 is applicable to larval tissue of Carpocapsa saltitans, the interfacial area for diffusion becomes very large, and diffusional distances in the tissue parenchyma are much reduced. Intertracheolar distance is probably not more than 2 ktm (Buck, 1962) thus making the most remote target of oxygen consumption about 1/~m removed from the tracheolar source of oxygen supply in a symmetrical array oftracheoles. Radial diffusion of oxygen over this distance and for the highest metabolic rates shown in table 2 requires a APo2 much less than 1 torr, as calculated by the Krogh-Erlang equation. Even for a 10/~m separation the pressure gradient is still less than 1 torr. In summary, almost the entire loss of oxygen pressure head occurs at the spiracular aperture, and that loss is calculated to be 23 torr, which is in good agreement with the experimentally determined average 'critical' Po2 of about 20 torr, since no further significant decrement occurs in either the airway system or in the tissues.
Acknowledgements The author thanks Kathy Franklin for her expert technical assistance and the Chapparal Novelties Company of Alamogordo, NM, for their generous contribution. This work was supported by Grant H L 02888-27 of the National Heart, Lung and Blood Institute of the National Institutes of Health.
References Boyle, R. (1670). New pneumatical experiments about respiration. Trans. R. Soc. (London) 5:2011-2031. Bridges, C.R., P, Kestler and P. Scheid (1980). Tracheal volume in the pupa of the Saturniid Moth Hyalophora Cecropia determined with inert gases. Respir. Physiol. 40: 281-291. Buck, J. and M.L. Keister (1955). Cyclic CO2 release in diapausing Agapema pupae. Biol. Bull. Mar. Biol. Lab. Woods Hole 109: 144-163. Buck, J. (1962). Some physical aspects of insect respiration. Annu. Rev. EntomoL 7: 27-56. Chance, B., N. Oshino, T. Sugano and A. Mayevsky (1973). Basic principles of tissue oxygen determination from mitochondrial signals. Adv. Exp. Med. Biol. 37a: 277-292. Crozier, W.J. and T.B. Stier (1925). Critical thermal increments for rhythmic respiratory movements in insects. J. Gen. Physiol. 7: 429-447. Dixon, M. (1943). Manometric Methods. 2nd ed. Cambridge, Cambridge University Press; New York, MacMillan Co., 155 p.
134
S.M. TENNEY
Ebeling, W. (1964). The permeability of insect cuticle. In: The Physiology of Insecta. V. 1I, edited by M. Rockstein. New York - London, Academic Press, pp. 507-556. Fenn, W. O. (1927). The oxygen consumption of frog nerve during stimulation. J. Gen. Physiol. 10: 767-779. Hoyle, G. (1959). The neuromuscular mechanism of an insect spiracular muscle. J. Insect Physiol. 3: 378-394. Hoyle, G. (1960). The action of carbon dioxide gas on an insect spiracular muscle. J. Insect Physiol. 4: 63-79. Keister, M. and J. Buck (1964). Respiration: Some exogenous and endogenous effects on rate of respiration. In: The Physiology of lnsecta, vol. II, edited by M. Rockstein. New York - London, Academic Press, pp. 617-658. Krogh, A. (1920). Studien uber Tracheenrespiration. Ober Gasdiffusion in den Tracheen. Pfliigers Arch. Gesamte Physiol. Menschen Tiere 179:95-112. Mani, M. S. (1968). Ecology and Biogeography of High Altitude Insects. The Hague, W. Junk, 137 p. Miller, P. L. (1964). Respiration - aerial gas transport. In: The Physiology of Insecta. Vol. II, edited by M. Rockstein. New York - London, Academic Press, pp. 557-615. Nunome, Z. (1944). Studies on the respiration of the silkworm. 1. Diffusion of oxygen in the respiratory system of the silkworm. II. Several measurements in the respiratory organ. Bull. Sericulr Exp. Stn. Japan 12: 17-90. Paganelli, C.V. (1980). The physics of gas exchange across the avian egg shell. Am. Zool. 20: 329-338. Scheid, P., C. Hook and C. R. Bridges (1982). Diffusion in gas exchange in insects. Fed. Proc. 41: 2143-2145. S chneiderman, H. A. (1960). Discontinuous respiration in insects: role of the spiracles. Biol. Bull. Mar. Biol. Lab. Woods Hole 119: 494-528. Thorpe, W.H. and D.J. Crisp (1947). Studies on plastron respiration. II. The respiratory efficiency of the plastron in Aphelocheirus. J. Exp. Biol. 24: 270-303. Wangensteen, O. D., D. Wilson and H. Rahn (1970/71). Diffusion of gases across the shell of the hen's egg. Respir. Physiol. 11: 16-30. Weis-Fogh, T. (1964). Diffusion in insect wing muscle, the most active tissue known. J. Exp. Biol. 41: 229-256. Wellington, W. G. (1946). The effects of variations in atmospheric pressure upon insects. Can. J. Res. Zool. Soc. 24: 51-70. Wigglesworth, V.B. (1945). Transpiration through the cuticle of insects. J. Exp. Biol. 21:97-114.