Post-Dive Blood Lactate Concentrations in Emperor Penguins, Aptenodytes Forsteri

Emperor penguins Aptenodytes forsteri can dive as deep as 534 m (Kooyman and Kooyman, 1995) and for as long as 22 min (Robertson, 1995). Frequency analysis of diving durations during foraging trips at sea has revealed that 21 % of all dives are longer than 5 min in duration, and that 3 % exceed 8 min in duration (Kooyman and Kooyman, 1995). The proportional contributions of aerobic and anaerobic metabolism to the energy requirements of diving are unknown, but they may be significant in determining the durations of dives, surface intervals and diving bouts. An aerobic dive limit (ADL), the diving duration beyond which the post-dive lactate concentration increases above resting levels (Kooyman, 1985), has been used in other species to evaluate which dives are defined as aerobic, i.e. those dives in which there is no net lactate production. Although an ADL has been measured only in the Weddell seal Leptonychotes weddellii (Kooyman et al. 1980, 1983), estimates of ADLs, based either on O₂ store/metabolic rate calculations or on analyses of diving duration frequencies, have been applied to many species (Gentry and Kooyman, 1986; Kooyman, 1989; Kooyman et al. 1992a; Le Boeuf et al. 1988; Lydersen et al., 1992). This limit for the emperor penguin has been calculated as 5 min on the basis of O₂ stores and an estimated diving metabolic rate (Kooyman and Ponganis, 1990). More recent analysis of emperor penguin diving duration versus surface interval relationships has inferred a behavioral ADL of 8 min (Kooyman and Kooyman, 1995). Because of the differences in these ADL estimations as well as the potential effects of metabolism on diving behavior, we attempted to determine an ADL in emperor penguins by measuring post-dive lactate concentrations in free-diving birds.

In McMurdo Sound, Antarctica, an experimental diving hole, penguin corral and camp (Kooyman et al. 1992b) were set up on sea ice over a water depth of 600 m. Seventeen adult emperor penguins Aptenodytes forsteri (G. R. Gray) of 21–25 kg body mass (m_b) were collected locally and held in the corral for 30–40 days. They were allowed daily access to the diving hole. While diving, they fed on fish, probably Pagothenis borchgrevinki.

The penguins typically left the water after dives and stood on the ice prior to the next dive. This behavior was the basis of the blood sampling protocol. Birds selected for catheterization were those most approachable and calm while in the close presence of researchers.

Penguins were anesthetized using isoflurane and catheterized percutaneously as described previously (Kooyman and Ponganis, 1994). The catheter, placed either in a foot vein or in a wing vein, was secured using Coban (3M) wrap and Loctite epoxy glue. A 15 cm extension tube and stopcock extended from the catheter either to the posterior aspect of the foot or to the back near the axilla. The dead space in the catheter tubing was filled with 33 % ethanol/normal saline (0.9 %) solution prior to returning the animal to the corral. Tessa tape secured the distal end of the tubing to the feathers. Before recovery from anesthesia, a Mark 4.5 time–depth recorder (TDR, Wildlife Computers) was glued with Loctite onto the lower back.

Four hours after recovery from anesthesia, the birds were allowed free access to the diving hole. Diving durations and surface intervals were recorded using a stopwatch and verified using the TDR records. Blood sampling occurred over a few hours to 2 days following implantation of the catheter. The duration of the sampling period was dependent on the functioning of the catheter. Catheters and TDRs were removed under anesthesia at the end of the sampling period. Observations of diving behavior were made from a sub-ice observation chamber and during SCUBA diving.

Blood collection was accomplished using a slow, calm approach to the bird so as to minimize disturbance and stress. Stopcock manipulation and blood collection were achieved from behind the upright bird once it had been surrounded by two or three kneeling persons. Most samples were collected within 5 min after a dive. Usually, only one post-dive sample was obtained. On five occasions, multiple post-dive blood samples were drawn over periods of 7–60 min. Control samples were drawn during rest periods in the corral.

Blood samples (1 ml) were drawn into heparinized syringes and were immediately analyzed for hematocrit by microcentrifugation and for blood lactate concentration using a YSI sport lactate analyzer (model 1500, sensitivity 0.01 mmol l⁻¹).

Graphic analysis of the data was conducted using Table Curve (Jandel).

Control blood lactate concentrations ranged from 1.2 to 2.7 mmol l⁻¹; the mean value was 1.7±0.71 mmol l⁻¹ (mean ± S.D., N=3). On four occasions, serial samples after dives showed that the blood lactate concentration peaked within 5 min after a dive. In one animal, lactate level peaked at 12 min post-dive. This animal had made three consecutive 6–6.5 min dives before leaving the water and shivered continuously for 31 min after surfacing. In three cases where multiple samples were obtained after the peak lactate value, 7–12 min was required for blood lactate concentration to decrease from 5– 8 mmol l⁻¹ to below 2.5 mmol l⁻¹ (Fig. 1). Mean time (± S.D.) between the end of a dive and the first sample was 3.6±1.78 min (N=22).

Blood lactate determinations were made after 22 dives in four birds. Two-phase regression analysis (Yeager and Ultsch, 1989) of the post-dive blood lactate concentration and diving duration data (Fig. 2) yielded a transition at 5.6 min between the two regression equations with a minimum residual sum of squares (RSS) of 81. Continuous linear and exponential function analyses resulted in greater RSS values. These included: y=0.70+0.82x, r²=0.63, RSS=98; y=1.20+0.39x^1.32, r²=0.68, RSS=93; y=−11.8+12.8^(x/20.28), r²=0.67, RSS=95.

The catheter did not impede the motion of the wing and, from underwater observations, did not appear to encumber the animal significantly during diving. The birds often dived as a group, and the catheterized birds typically returned in the midst of the group.

Maximum diving depths during this study were less than 85 m. Diving durations ranged from 0.5 to 12 min. Feeding during most of these diving bouts was determined from underwater observations of capture of Pagothenia borchgrevinki and inferred from (1) differences in the color of guano when the birds began to dive after resting periods during storms, (2) observed abdominal distention after diving bouts, and (3) constant or increased body mass whenever measurements were made.

We consider the baseline mean lactate concentration of 1.7 mmol l⁻¹ as evidence that significant elevation of lactate concentrations due to stress did not occur as a result of the sampling procedure. These values are similar to the 2.0–2.6 mmol l⁻¹ resting blood lactate concentrations reported in chronically catheterized geese adapted to handling for 2–3 weeks (Le Maho et al. 1981). In addition, they are in the same range as plasma lactate concentrations in emperor penguins swimming at low workloads in a flume (Kooyman and Ponganis, 1994).

Since lactate concentrations at rest ranged from 1.2 to 2.7 mmol l⁻¹, we only considered post-dive lactate concentrations greater than 3 mmol l⁻¹ as definitive evidence of an increase in post-dive blood lactate level. Using this criterion, no diving durations of less than 5 min were associated with an increase in lactate level. The two-phase regression analysis resulted in a transition at 5.6 min and indicated that post-dive lactate accumulation was not a continuous linear function of diving duration. After dives lasting longer than 7 min, there was always an increase in lactate level. However, for dives of between 5 and 7 min duration, even for those by the same individual, the lactate response was variable. We suggest that this range of diving durations is the transition zone towards greater reliance on anaerobic metabolism. Within this range, factors such as high swimming speed and the associated increase in oxygen consumption could influence the onset of anaerobiosis. The variability of the lactate response in this diving duration range results in the large RSS values using either curve-fitting or two-phase regression analysis and broadens the threshold for lactate accumulation. We conclude from this analysis that the ADL in emperor penguins ranges between 5 and 7 min. This measured ADL is similar to that previously calculated on the basis of estimated O₂ stores and metabolic rates.

Although the O₂ store/metabolic rate calculation appears to predict the dive duration after which post-dive lactate concentration increases, this does not mean that all O₂ stores have been depleted within that period. The status of the O₂ stores at the end of a dive, the O₂ store depletion rate during the dive and the site(s) of tissue lactate production are unknown.

Oxygen store management within the breath-hold period remains one of the important issues in diving physiology.

It is also apparent from Fig. 1 that elevated lactate concentrations decrease to baseline levels within 7–12 min following a dive. Although the number of data points is small, this decline in lactate concentration appears to be more rapid (Fig. 1) than equivalent changes in post-dive lactate concentrations of Weddell seals (Castellini et al. 1988; Kooyman et al. 1980). This may be due to several differences between the two species, including the relative diving duration and the quantity of lactate produced, the blood volume, the volume of distribution of lactate and lactate clearance rate. Nonetheless, such decreases in lactate concentration may occur during the surface intervals and/or shallow dives recorded after extended dives of emperor penguins (Kooyman and Kooyman, 1995) and king penguins (A. patagonicus; Kooyman et al. 1992a). These rapid recoveries and the variable lactate response after 5–7 min dives may account for the 8 min behavioral ADL for emperor penguins, which was inferred from analysis of diving durations and surface intervals (Kooyman and Kooyman, 1995).

This research was supported by NSF grant OPP 92-19872.