Ecophysiological studies on arthropods from Spits berg en TORE AUNAAS, JOHN G. BAUST AND KARL ERIK ZACHARIASSEN Aunaas, T . , Baust, J. G. & Zachariassen, K. E. 1983: Ecophysiological studies on arthropods from Spitsbergen. Polar Research 1 a s . , 235-240. The cold-hardiness, high temperature tolerance and metabolic activity of summer specimens of staphylinid beetles ( A t h e t a graminicola), collembolans (Onichiurus g r o e n l a n d i c w ) , spiders (Erigone arcrica), and prostigmatid mites (Molgus l i n o r a l b ) from Spitsbergen were investigated. The animals displayed cold- hardiness and haemolymph melting points within the normal ranges for summer insects from temperate regions, but were less tolerant to high temperatures. Haemolymph from spiders and from one species of collembolans ( l s o r o m a sp.) was found to contain thermal hysteresis factors. The beetles. collembolans. and mites were found to have oxygen consumption rates above the values of their relatives in other climatic zones, whereas the spiders had values within the range of temperate arachnoids. The study supports the view that polar arthropods have activation energy values lower than those of temperate animals. Tore A u n a a s and Karl Erik Zachariassen, Deparrmetir of Z o o l o g y , Uniuersiry of Trondheim. 7055 Dragvoll. N o r w a y ; John G. Baucr. Deparrmenr of Biology, Unioersiry of H O U S ~ O I I , H o u s t o n . Texas 77004, U . S . A . Introduction Several studies have revealed that mammals and birds from Spitsbergen a r e physiologically highly specialized (Krog et al. 1976; Grammeltvedt & Steen 1978; Ringberg & Reimers 1982). These specializations apparently reflect adaptations to the prevailing climatic conditions on the islands. Although the invertebrate fauna at Spitsbergen is fairly well known ( S ~ m m e 1979). very few studies have been made on the possible physio- logical adaptations of the Spitsbergen arthropods. The purpose of the present study was to inves- tigate the cold-hardiness and possible metabolic adaptations of the terrestrial arthropods in this area. Materials and methods The investigations were carried out o n insects and arachnoids collected in the vicinity of Longyear- byen in mid J u n e 1981 and in the vicinity of N y - h e s u n d at the end of July 1982. The animals were collected under stones o n the tundra or in the rich vegetation below bird cliffs in Kongsfjor- den. The animals were kept in small glass tubes at temperatures ranging from + 2 to -2°C for u p to ten days before they were used in the experi- ments. T h e experiments were carried out partly in a temporary laboratory established in a cabin in Adventdalen and partly at the research station of the Norwegian Polar Research Institute in Ny-Alesund. The supercooling points of the animals were determined by using the arrangement shown in Fig. l a . T h e animals were attached t o a DM thermistor probe of a G r a n t temperature recorder by means of silicone grease. They were cooled inside a thermos bottle containing a cold mixture made from snow and CaCI2.6H20. T h e super- cooling points were indicated as small inflections on the temperature curve, d u e to the release of the heat of fusion of water freezing. The melting points of the haemolymph of the animals were determined by using a Clifton nanolitre osmometer, in which t h e melting pro- cess of 3 0 n l samples of haemolymph could be observed in a microscope, while the temperature of the samples was regulated with an accuracy of 5 0.001"C. T h e temperature at which the last tiny ice crystal disappeared during slow warming of the sample was taken as the melting point. Samples of haemolymph were obtained from the animals by puncturing the cuticle with a thin glass capillary and sucking the haemolymph into the capillary by means of the capillary forces. When a sufficient amount of haemolymph could not be obtained from o n e animal, the samples were pooled from several individuals. T h e haemo- lymph samples were handled as described by Zachariassen et al. (1982). 236 T. Aunaas, J . G. Baust and K . E. Zacharias3 ;en / A B Fig. I . A : Arrangement of instruments for determination of supercooling points. B-animal. C.M.*old mixture. T% thermos. T.R.-Grant temperature recorder. B: Arrangement of instruments for determination of oxvgen consumption. %animal. A-CO? absorber (10% solution of KOH). R- respirometer. S.L.-Styrofoam lid. T-thermometer. TS-ther- mos. W-water. F-indicator fluid The presence in t h e haemolymph of thermal hysteresis factors ( T H F ) , which have the ability to separate the temperature of ice crystal growth upon cooling of a frozen sample from t h e melting point. was investigated o n the Clifton nanolitre osmometer as described by Zachariassen & Husby (1982a). T h e haemolymph samples were cooled with a tiny ice crystal present, and the temperature at which a rapid growth of the ice crystals was observed was taken as the hysteresis freezing point (HFP). The upper lethal temperatures of the animals were determined by exposing them to tempera- tures increased in steps of 5°C until they showed abnormal behaviour (uncoordinated walking) or died. The animals were exposed t o each temper- ature for 1 0 m i n . T h e experiments were per- formed with the animals kept within small glass tubes which were immersed in water baths of the desired temperature. T h e oxygen consumption of the animals was measured a t different temperatures by using Engelmann constant pressure respirometers, the inner capillary diameter of which is 0.5 mm (Engelmann 1963). I n order t o obtain constant temperatures, the respirometers were immersed in waterfilled thermos bottles as shown in Fig. l b . U p t o five animals were used in each respiro- meter. A piece of paper was put into the respi- rometers t o give the animals a convenient sub- stratum and thus keep their activity at a low level. In order t o adjust for possible variations in tem- perature o r atmospheric barometric pressure, an empty respirometer was used as a blank instru- ment. T h e oxygen consumption was calculated at N T P and in relation to t h e fresh body weight of the animals. Since n o balance was available in the field lab- oratories, the fresh body weight had t o be deter- mined by means of an indirect method after return to the University. Some of the animals, particu- larly the collembolans, became seriously dehy- drated during t h e respirometer experiments, and the fresh body weight was estimated from the dry weight. Immediately after the experiments the animals were transferred t o dry, clean glass tubes, in which they were dried t o a constant weight at +60"C. T h e relative water content of each species was determined o n specimens that were trans- ported alive to the University of Trondheim, and these data were used to estimate the fresh body weights of the experimental animals from the dry weight values. T h e relative water content of the animals varied from about 7 5 % for the collem- bolans and mites t o about 61% for the beetles. The oxygen consumption was calculated in rela- tion t o the estimated initial fresh body weights. Results The melting points, hysteresis freezing points, supercooling points, and upper lethal tempera- tures of the animals are shown in Table 1. T h e data reveal that all species investigated had haemolymph melting points in t h e range of from Ecophysiological studies on arthropods 237 Table I . Melting points, hysteresis freezing points, supercooling points and upper lethal temperatures of Spitsbergen arthropods. Values are Mean ? SD, and the number of measurements are given in parentheses. ~ Melting Hysteresis point freezing point Species (“C) (“C) Molgus littoralis -1.02 -1.02 Hypogastrura sp. - - Erigone arctica -1.06 -1.31 Onichiurus groenlandicus -0.31 -0.31 Isotoma sp. -0.75 -0.90 Atheta graminicola -1.03 -1.03 Supercooling point (“C) Upper lethal temperature (“C) -6.4 t 0.6 (6) -6.0 2 1.1 (7) -6.7 t 1.7 (11) -7.1 5 1.7 (10) - -5.3 2 1.1 (7) +35- +40 - +30- +35 - - +30- +35 Table 2 . Oxygen comsumption (mm3 O2/g.min) of Spitsbergen arthropods at different temperatures. Values are mean 2 S E . The numbers of measurements are given in parentheses. Species 0 Temperature (“C) + 5 + 10 + 15 Erigone arctica 1.1 2 0.4 (2) 2.1 2 0.3 (7) 1.7 2 0.5 (4) 3 4 2 1.2 (4) Onichiurus groenlandicus 5.9 2 0.8 (5) 9.9 2 2.4 (6) 14.0 ? 5.8 (5) - Molgus littoralis 1.4 ? 0.3 (3) 5.0 % 1.5 (3) 6.7 ? 1 . O (3) 14.0? 1.4 (3) Atheta graminicola 5.4 t 1.3 (3) 10.6 2 1.1 (4) 13.7 t- 1.8 (4) 26.8 -t 1.3 ( 4 ) about -1.2 to about -0.3”C. In most of the species the freezing point corresponds to the melt- ing point, indicating that THF are absent from their haemolymph. However, the haemolymph of the spiders and one species of collembolans (Isotorna sp.) showed a moderate hysteresis, revealing the presence of THF in the haemolymph of these species. All species had supercooling points within the range of from -5 to -7°C. None of the animals were tolerant to freezing, and thus, the super- cooling points correspond to the lower lethal tem- peratures of the animals. The upper lethal tem- peratures range from +30 to +35”C for the collembolans and the beetles, and from +35 to +40”C for the spiders. The oxygen consumption values of each species at different temperatures are listed in Table 2. The results show that the mites, collembolans, and beetles have an oxygen consumption con- siderably higher than that of spiders. In order to study the temperature dependence of the metabolic processes more exactly, the data are plotted in an Arrhenius plot (Fig. 2). The Arrhenius plot is based on the Arrhenius equation M = a . e-dRJR.T, where M is the metabolic rate (= oxygen consumption), a is a constant, p is the activation energy, R is the universal gas constant, and T is the temperature in O K . The Arrhenius equation can be expressed as In M =In a - (@) (1/T), where 1n M and 1/T are used directly in the plot by being linearly related variables. The terms In a and -p/R represent the ordinate inter- ception point and the slope, respectively, and can be determined by calculating the linear regression line of the values of In M and 1/T. 3 5 - 3 0 2 5 ~ 2 0 - x c - 1 5 1 0 - 0 5 - 0 3 50 3 55 3 60 3 65 I / T I O K - ~ ~ 1031 Fig. 2 . Arrhenius plot of the metabolic rates of four species of terrestrial arthropods from Spitsbergen. The lines are linear regression lines, calculated as described in the text. 238 T . Aunaas, J . G. Baust and K . E . Zachariassen Table 3. Qlo values and activation energy values of Spitsbergen arthropods. The correlation coefficient ( r ) and the number of observations for each species ( n ) are also given. Species Activation energy Qiil (kcab'mol) r n Erigone arctica 1.9 6.9 -0.41 17 Molgus littoralis 3.4 21.1 -0.91 12 Onichiurus groenlandicus 2.3 9.8 -0.41 16 Atheta graminicola 2.8 16.6 -0.92 15 The activation energy is calculated by multi- plying the slope of the regression line by the universal gas constant (1.98 cal/mol."K). The val- ues for the activation energy of the Spitsbergen arthropods are given in Table 3, together with the Q l o values of the oxygen consumption. The data show that the activation energy vanes from about 7kcal/mol for the spider to about 21 kcay'mol for the mite. The Q l o values vary from 1.9 for the spider to 3.4 for the mite. The values for the insect are between these extremes. Discussion The observed melting points of the haemolymph of the Spitsbergen arthropods (-1.2 to -0.3"C) are typical for insects and spiders lacking accu- mulated polyols in their body fluid (Zachariassen 1980). This conforms well with the observed supercooling points (-5 to -7°C). which are in the range characteristic of temperate summer adapted insects ( S ~ m m e & Conradi-Larsen 1977; Zachariassen 1980). Thus, the summer insects at Spitsbergen seem to be similar to summer insects in temperate and tropical regions. This pattern appears to be disturbed by the presence of thermal hysteresis factors (THF) in the haemolymph of the spiders and in one species of collembolans. The latter observation may sug- gest that several species of the Spitsbergen arthro- pods are able to survive prolonged exposures at moderately low subzero temperatures in the sum- mer, even when in direct contact with external ice (Zachariassen & Husby 1982b). Studies by Duman (1977, 1979) have revealed that the levels of T H F vary considerably over the year. from moderate levels or absence during the summer, to a hysteresis range of up to about 6°C in the winter. Since the Spitsbergen arthropods are likely to spend an extremely long period in a hibernating state, those which are sensitive to freezing probably depend heavily on THF to stabilize the unfrozen state and thus avoid a lethal freezing. Consequently, the T H F levels of hiber- nating Spitsbergen arthropods are likely to be considerably higher than those observed in the summer specimens. Furthermore, high levels of T H F may well be present during the winter in species lacking such substances in the summer. Block & Young (1978) found that Antarctic cryptostigmate and mesostigmate mites had oxygen consumption rates higher than those of temperate species, whereas the oxygen consump- tion of prostigmate mites was lower than the values of temperate mites. Block (1981) obtained results indicating that sub-Antarctic beetles have oxygen consumption rates somewhat below the values found in alpine temperate species (HBgvar & Bstbye 1974), but that sub-Arctic collembolans had rates which were somewhat higher than those of collembolans from temperate alpine regions (Conradi-Larsen 1974). In order to compare the oxygen consumption of the Spitsbergen arthro- pods with the values of arthropods from other I 1 0 0 E m . 0 E - a r r z g 1 0 5 Q z 0 u 7 3 1 t i I + 5 r 1 0 * I 5 TEMPERATURE i " C \ Fig. 3. Semilogarithmic plot of the oxygen consumption of arachnids from Spitsbergen and other regions as a function of temperature. --: Spitsbergen spiders Erigone arctica (0) and mites Molgus linoralis ( 0 ) ; ---: Temperate zone spiders from Finse (line a) (HHgvar & 0stbye 1974; Steigen 1976) and tem- perate cryptostigmatid mites (line b) (Young 1979); - -: Antarctic mite (Young 1979). The lines are linear regression lines. The lines representing values for Spitsbergen animals are calculated from the individual values forming the basis for the data in Table 2, whereas the other lines are calculated from the mean values tabulated in the articles referred to. The bars represent SE and the number of parallel measurements are indicated on the top of each bar. Ecophysiological studies on arthropods 239 deviating results obtained for the spiders may mean that these animals survive in the area due to other strategies, such as reduced development rate and ability to complete their development over several years. It has also been speculated whether a low oxygen consumption rate may cause more energy to be invested in growth, and thus that animals with a low oxygen consumption rate in fact grow and develop faster than animals oxidizing their food at a high rate (Block & Young 1978). How- ever, it should be kept in mind that protein syn- thesis and growth also require energy, and that a high growth rate is likely to be accompanied by a high rate of oxygen consumption. Considerably more data are needed on the biological features of the animals and the climatic conditions under which they live, before firm conclusions regarding the physiological adaptations of the terrestrial invertebrates can be drawn. Young (1979) and Block & Young (1978) cal- culated the activation energy of different Ant- arctic and sub-Antarctic arthropods, and found that the values were generally within the lower part of the range of values found for temperate animals. These authors concluded that the acti- vation energy values seem to vary systematically with the prevailing temperature in the habitat of the animals. Figure 5 shows the values of acti- vation energy of Spitsbergen arthropods, plotted in frequency distribution histogram, together with values of temperate, alpine temperate, sub-Ant- arctic, and Antarctic arthropods. The Spitsbergen animals seem to have activation energy values in the lower range, quite similar to the Antarctic arthropods studied by Block and Young. More- over, measurements carried out on tropical desert insects have shown that these insects have a + z t 6 100 2 I- I 0 0 z W > 0 0 .5 .I0 +15 TEMPERATURE ( O C I Fig. 4. Semilogarithmic plot of the oxygen consumption of insects from Spitsbergen and other regions as a function of temperature. -: Spitsbergen collembolans Onichiurus groen- h d i c w (0) and staphylinid beetles Arheru gruminicolu (0); beetles (line b) (Block 1981); ---: Alpine temperate zone beetles (line c) (five species from Finse studied by HBgvar & Bstbye 1974) and alpine temperate zone collembolans (line d) (two species from Finse studied by Conradi-Larsen 1974); _ _ : two species of curculionid beetles from Mount Kenyz (Zachariassen, unpublished). The bars represent SE and the number of parallel measurements are indicated on the top of each bar. ...... . Sub-Antarctic collembolans (line a) (Block 1981) and regions, the values have been plotted together in Figs. 3 and 4. Figure 3 shows that Erigone spiders from Spits- bergen have oxygen uptake rates which are lower than the values of temperate alpine spiders from the Finse high mountain plateau in Norway over most of the temperature range, but which agree fairly well with the temperature range where the Spitsbergen spiders perform their activity in nature, i.e. 0 - +5"C (personal observations). Mites from Spitsbergen, on the other hand, have oxygen consumption rates considerably higher than those of temperate mites: these results con- form with those obtained by Young (1979). Figure 4 shows that the Spitsbergen collem- bolans and beetles have oxygen consumption rates considerably above the values of insects from other regions. It is tempting to interpret these results to mean that the Spitsbergen insects are active at relatively low temperatures, and that they have to increase their metabolic rate in order to complete their development within the rela- tively short summer period. The somewhat i n I 9 6 3 4 m 2 0 15 20 25 10 5 ACTIVATION ENERGY I kcoi I m o l l Fig. 5. Frequency distribution of activation energy of polar and temperate arthropods. 0: Temperate mites (Young 1979); W : Alpine temperate arthropods (Conradi-Larsen 1974; HBgvar & Bstbye 1974; Steigen 1976); H Antarctic and sub-Antarctic arthropods (Young 1979; Block 1981); 88 Spitsbergen arthro- pods (Table 3 in present study). 240 T . Aunaas, J . G. Baust and K . E . Zachariassen extremely high activation energy values (Zacha- riassen pers. comm.). These observations provide further support to the view that the activation energy reflects the predominant temperature con- ditions under which a species lives. Acknowledgements. - This study was made possible by econ- omic support from the Norwegian Research Council for Science and the Humanities, the Norwegian Marshall Fund. the Norwegian Polar Research Institute. and the Department of Zoology. University of Trondheim. The authors express their gratitude to these institutions. We also thank Professor Arnol- dus Schytte Blix. Department of Arctic Biology. University of Tromsa. who provided accommodation and laboratory facilities in Adventdalen. and Drs. Erling Sendstad. Per Sveum. and Arne Lauvaas for help with classifying part of the material. References Block. W. 1981: Respiration studies on some South Georgian Coleoptera. Colloque sur les Ecosysremes Subanrarcriqrces. P a i m p o n f , C . H . F . R . A . No. 5 1 . 183-192. Block, W . & Young, S . R. 1978: Metabolic adaptations of antarctic terrestrial micro-arthropods. Comp. Biochern. Physiol. 6 1 A I 363-368. Conradi-Larsen. E. M. 1974: A constant pressure micro- respirometer for measurement of 02-consumption in Collem- bola. Nor. Ent. Tidsskr. 21. 187-189. Duman. J . G. 1977: The role of macromolecular antifreeze in the darkling beetle, Meracanrha contracta. 1. Comp. PhyJiol. 115. 279-286. Duman, J. G . 1979: Thermal hysteresis factors in over-wintering insects. J . Inrecr Physiol. 25. 805-810. Engelmann. M. D. 1963: A constant pressure respirometer for small arthropods. E N . News 74. 181-187. Grammeltvedt. R. & Steen, J . B . 1978: Fat deposition in Spitsbergen Ptarmigans. Arctic 31, 496498. Higvar. S . & Bstbye, E . 1974: Oxygen consumption, caloric values. water and ash content of some dominant terrestrial arthropods from alpine habitats at Finse. south Norway. N o r . Ent. Tidsskr. 21, 117-126. Krog, J . O . , Wika. M . , Lund-Larsen, T.. Nordfjell, J . & Myrnes. I. 1976: Spitzbergen Reindeer: Morphology, fat storage and weight in the late winter season. Nor. J . 2001. 2 4 . 407-417. Ringberg. T . & Reimers, E . 1982: An epic in high arctic survival. The Svalbard Reindeer. A c t a Physiol. Scand. Steigen. A . L. 1976: Energy metabolism in the wolf spider Pardosa palurtris (L.) (Araneae. Lycosidae) from Hardan- gervidda. southern Norway. Nor. J . Ent. 23, 51-60. S~imme, L. 1979: Insektliv pB Svalbard. Fauna 32, 137-144. Somme, L. & Conradi-Larsen, E . M. 1977: Cold-hardiness of collembolans and oribatid mites from windswept mountain ridges. Oikos 29, 118-126. Young. S . R . 1979: Respiratory metabolism of AIuskoretes antarcticus. J . Insect Physiol. 25. 361-369. Zachariassen. K. E. 1977: Ecophysiological studies on beetles from arid regions in East Africa. N o r . J . Ent. 24. 167-170. Zachariassen. K. E. 1980: The role of polyols and nucleating agents i n cold-hardy beetles. J . Comp. Physiol. 140. 227- 234. Zachariassen. K. E . . Baust. J . G. & Lee, R . E . 1982: A method for quantitative determination of ice nucleating agents in insect hemolymph. Crvobiology 19. 18@184. Zachariassen. K . E. & Husby. J . A . 1982a: Antifreeze effect of thermal hysteresis agents protects highly supercooled insects. Y a t u r e 298. 865-867. Zachariassen. K. E . & Husby. J . A . 1982b: Stabilization of highly supercooled insects by thermal hysteresis antifreeze agents. Abstract from Symposium on Invertebrate Cold-Har- diness. Oslo. August 1982. Cryo-Letters 3 , 316.