Vazquez.indd 63Vazquez & Mac Cormack 2002: Polar Research 21(1), 63–71 Enzymes constitute an important industrial prod- uct and have reached an annual marketing figure of US$ 1.6 billion. Proteolytic enzymes repre- sent nearly 60 % of this market and have a broad spectrum of applications, mainly in the food in- dustry, laundry products and leather processing (Demain 1999). As enzymes showing high cat- alytic efficiency under extreme environmental conditions could be useful for many purposes, extremophilic microorganisms became the centre of an important research effort conducted to ob- tain enzymes adapted to work under high or low temperatures, and in extremely acidic or alkaline media and high salinity (Aguilar 1996). Among extremophiles, psychrophilic and psychrotolerant bacteria, which are adapted to grow at 0 °C, con- stitute a relevant source of enzymes potentially useful in industrial processes in which working at low temperature represents an advantage (Ger- day et al. 2000). As the majority of the industrial proteases currently used are produced by meso- philic microorganisms and show optimum activ- ity around 55 - 60 °C (Kamal et al. 1995; Suharto- no et al. 1997; Varela et al. 1997; Asakawa et al. Effect of isolation temperature on the characteristics of extracellular proteases produced by Antarctic bacteria Susana C. Vazquez & Walter P. Mac Cormack Protease-producing psychrotolerant bacteria were isolated from Antarctic biotopes on casein agar plates using different incubation temperatures. Most of the isolates were non-spore-forming Gram-negative motile rods with catalase activity, 30 % were pigmented and none of them were glu- cose fermenters. All the strains were grown in liquid cultures at 20 °C and protease secretion was tested using the azocasein method. Despite their capacity for production of a clear zone of hydrolysis in agar plates, some strains did not produce detectable levels of proteolytic activity in liquid cultures. The lowest apparent optimum temperature for protease activity found in culture supernatants was 40 °C. Almost all the strains showed activation energy values about 10 - 20 kJ·mol-1 lower than that observed for a mesophilic Subtilisin. Most of the proteases showed optimal activity at neutral or alkaline pH values and developed a multiple-band profile on gelatine-SDS-PAGE. It was observed that the lower the strain isolation temperature was, the more stongly cold-adapted—in terms of optimal temperature and activation energy—were the proteases produced by them. This dependence of the characteristics of the proteases on the isola- tion temperature is an important factor to take into account in the design of screening programmes directed towards the isolation of psychrotoler- ant bacteria able to produce proteases strongly or weakly adapted to work in the cold. The Antarctic area explored proved to be a promising source of proteolytic bacteria with potential use in industrial processes to be car- ried out at low or moderate temperatures. S. C. Vazquez, Cátedra de Microbiología Industrial y Biotecnología, Facultad de Farmacia y Bioquímica, Universidad de Buenos Aires, Junín 956 6º Piso (1113) Buenos Aires, Argentina; W. P. Mac Cormack, Instituto Antártico Argentino, Departamento de Biología, Cerrito 1248 (1010) Buenos Aires, Argentina. 64 Proteases from Antarctic bacteria 65Vazquez & Mac Cormack 2002: Polar Research 21(1), 63–71 1998; Han & Damodaran 1998), there is a gener- al interest to find new proteases with lower opti- mum temperatures, which would result in a con- siderable energy saving (Horikoshi 1995; Aguilar 1996). In Antarctica, one of the coldest areas in the world, prokaryotes are the dominant organisms and they are mainly responsible for the hydro- geochemical cycles and the mineralization of or- ganic matter (Karl 1993; Rivkin et al. 1996). In addition, the Antarctic continent is one of the less explored places on Earth and accounts for the high proportion of new taxa that are described when biochemical and molecular techniques are applied to the analysis of bacteria present in sam- ples taken from these environments (Irgens et al. 1996; Reddy et al. 2000). Therefore, Antarctica is one of the most promising sources of new bacte- rial isolates able to produce cold-adapted prote- ases with lower optimum temperature than those currently used in industry. Environments in the Antarctic Peninsula are exposed to different tem- perature ranges. Some of them, such as soil and rock surfaces, are exposed to significant solar ir- radiation and show a broad range of temperatures which extends from below the freezing point to over 25 °C (Vishniac 1993). On the other hand, environments such as seawater, marine sediments and sea ice have constant temperature values lower than 4 °C (Schloss et al. 1998). Antarctic microorganisms’ ecophysiological and biochem- ical adaptations could be reflecting the different temperature regimes they have to withstand. It is known that isolation temperature strongly in- fluences the psychrophilic/psychrotolerant ratio of the bacteria isolated from a sample (Delille & Perret 1989; Delille 1992). Thus, even among the psychrotolerant bacteria, a low isolation tem- perature could determine a higher probability to obtain strains with low optimum temperature not only for growth but also for the activity of their extracellular enzymes. Although cloning of the genes involved in pro- tease production could be a necessary step in the optimization of the production process at in- dustrial scale, the screening of new strains and the characterization of the proteases produced by them represent an important initial activity. Since, from a biotechnological point of view, an industrial process based on psychrotolerant bac- teria is more simple and cost-efficient than one based on psychrophiles, the influence of the iso- lation temperature on the optimum temperature for growth and protease activity is an important point to consider to improve the search for this kind of strains. In the present work, our aim was to isolate neu- tral or alkaline protease-producing psychrotoler- ant bacteria from Antarctic environments and compare the effect that isolation temperature has on the basic properties of these extracellular pro- teases. Materials and methods Sampling area The samples were taken during Argentine sum- mer Antarctic Research Expedition (ARE) 1995–96 near Jubany scientific station (62° 14’ S, 58° 40’ W) on King George Island, South Shet- land Islands. Samples were collected from soil, fresh and marine waters, marine sediments and remains of organic matter of animal and plant or- igin. In addition, strains isolated during the ARE 1991–92 (Vazquez et al. 1995) were included for comparison and further study. Screening and isolation of psychrotolerant bacterial strains Samples from soil and organic matter were placed in a screw-capped bottle containing 5 g of sterile sand and 15 ml of sterile diluent (1 g·l-1 of peptone solution) and shaken for 15 min. After shaking, serial 10-fold dilutions were prepared in the same diluent and 0.1 ml of each dilution was spread on the surface of nutrient agar plates (Merck) for counting total heterotrophic bacteria. Proteolytic bacteria were screened by spreading 0.1 ml of di- lutions on casein agar plates at pH 7.0, in accor- dance with Hoshino et al (1997). Proteolytic ac- tivity was detected as clear zones of hydrolysis around the colony. Samples from marine origin were processed in the same way but diluent and media were prepared using 75 % v/v seawater. Samples were incubated at 20 °C and at 4 - 6 °C for 20 days, in order to further analyse the effect of incubation temperature on the characteristics and growth of the isolated strains and on the ac- tivity of the proteases produced by them. Differ- ent proteolytic colonies were isolated and purified by re-streaking twice. General characteristics of previously obtained proteolytic strains isolated at 10 - 13 °C during ARE 1991–92 (Vazquez et 64 Proteases from Antarctic bacteria 65Vazquez & Mac Cormack 2002: Polar Research 21(1), 63–71 al. 1995) are included for comparison in the Re- sults section. Characterization of bacterial strains All proteolytic bacteria were characterized from 48 h old pure cultures grown at 20 ºC. The strains were Gram stained and shape and size of the cells were examined under the light microscope. Their mobility was investigated in hanging drops. The colour of the colonies was observed on nutrient agar plates. Catalase, cytochrome oxidase activ- ity and aerobic and anaerobic utilization of glu- cose (Hugh-Leifson medium) were also tested. Liquid culture conditions Growth experiments were performed in nutrient broth (Oxoid CM1) supplemented with 0.3 g·l-1 CaCl2.2H2O, pH 8.0. For marine strains the me- dium was re-hydrated with 100 % v/v artificial seawater (Lyman & Fleming 1940). Experiments were carried out in 300 ml Erlenmeyer flasks with 60 ml of medium and incubated in a rotatory shaker at 240 rpm at 20 °C. Inocula were grown in the same medium, adjusted to an optical densi- ty at 580 nm (OD580) of 0.100 and added to broth at 1 % v/v proportion (final OD under the detec- tion limit). Samples were taken after 24, 48 and 72 h of incubation and used to measure growth, pH and proteolytic activity. All the experiments were performed in duplicate. Growth was mon- itored by measuring the OD580 of the samples in an UV-Visible spectrophotometer and expressed as dry weight using previously tested calibration curves. Dry weight was determined (in triplicate) by weighting the dried biomass (48 h at 105 °C) after centrifugation of aliquots of the culture. Protease assay Proteolytic activity was measured by digestion of azocasein, in accordance with Charney & To- marelli (1947). An appropriate dilution of culture supernatant (400 µl) was incubated with 400 µl of 1 % w/v azocasein in 0.1 M Tris/HCl buffer (pH 8.0) and 0.5 mM CaCl2.2H2O at 20 °C for 30 min. The reaction was stopped by adding 800 µl of 5 % w/v trichloroacetic acid. After centrifugation of the reaction mixture, absorbance of the superna- tant was measured at 340 nm. Samples were as- sayed in duplicate and the activity was expressed in relative enzymatic units (EU). One EU was de- fined as the amount of enzyme that produces an increase of 0.100 in A340 under the assay condi- tions. Effect of pH and temperature on activity of proteases The effect of pH and temperature on protease ac- tivity was determined by using the protease assay described above. Determination of the optimum pH was performed at 20 °C with the following buffer systems (0.1 M each): KH2PO4/Na2HPO4 (pH 5 - 7); Tris/HCl (pH 8 - 9) and Na2HPO4/ NaOH (pH 10 - 12). For determination of the opti- mum temperature, the reaction was carried out at different incubation temperatures (between 0 °C and 60 °C) and pH 8.0. Activation energies were calculated from the linear part of Arrhenius plots as described by Pirt (1985). Polyacrilamide gels Extracellular protease profiles were analysed by SDS-PAGE minigels containing gelatine as copo- lymerized substrate (Heussen & Dowdle 1980). Proteolytic activity was evident as bands deplet- ed of gelatine. Results Even though we sampled in biological systems where natural changes in the bacterial popula- tion make comparison among samples difficult, a common pattern was observed (Table 1). Samples from ecosystems exposed to wide thermal oscil- lation (like soil surface and organic matter ex- posed to solar radiation) showed higher total bac- terial counts at 20 °C when compared with those obtained at 4 °C. Conversely, samples taken from areas subjected to low and almost constant tem- perature (like seawater or marine sediments) had lower counts at 20 °C than at 4 °C. The number of Gram-positive strains varied with the different isolation temperatures, corresponding to 6 - 14 % when the isolation was made at 13 °C or lower, but rising to 40 % when the strains were isolat- ed at 20 °C. The presence of spore-forming bacte- ria was observed when the screening was carried out at 20 °C (Table 2) but they were not isolated at 4 °C, probably because this temperature was not suitable for the germination of the spores which were likely to be present. 66 Proteases from Antarctic bacteria 67Vazquez & Mac Cormack 2002: Polar Research 21(1), 63–71 On the basis of the formation of a clear zone in casein agar, 89 proteolytic strains (40 isolated at 20 °C and 49 at 4 - 6 °C) were selected. In Table 2, we summarize the general characteristics of the selected protease-producing psychrotolerant strains isolated during ARE 1995–96 and of other strains previously isolated during ARE 1991–92 (Vazquez et al. 1992). All the strains were capa- ble of growth at 0 °C but had an optimum growth temperature higher than 15 °C, corresponding to psychrotolerant microorganisms (Morita 1975). The majority of them were non-spore form- ing Gram-negative rods, though some strains were Gram variable. This was probably due to the presence of certain components on their cell walls which may interfere with the cell permea- bility to the stain. The pigmented strains (30 % of total isolates) were mostly yellow and orange co- loured. Motile strains with cytochrome oxidase activity predominated. Almost all the strains had catalase activity and none of them were glucose fermenters. When all the newly-isolated 89 proteolyt- ic strains plus the 34 additional strains selected in a previous screening were cultured in nutri- ent broth, some showed a protease secretion pat- tern associated with growth (Fig. 1b), while other strains presented their enzyme secretion partially associated with growth (starting at the early sta- tionary phase) or not associated with growth at all (Fig. 1a). Some strains were discarded for further stud- Fig. 1. Examples of two different patterns of growth and pro- tease production kinetics observed in the selected Antarctic strains. When strains were cultivated in liquid culture, prote- ase production showed two different behaviours. (a) Protease secretion delayed with respect to growth, starting at the early stationary phase (P-96-34). (b) Enzyme secretion associated with growth (P95-16). (a) (b) Table 1. Heterotrophic bacterial counts incubated at 20 °C and 4 - 6 °C from different samples taken during summer Antarctic Research Expedition 1994–95. Sample origin pH CFUa 20 °C 4 - 6 °C Lichens (Usnea antarctica) 5.5 1.8·106 ± 5.1·104 1.0·105 ± 4.5·104 Moss mat (near pond coast) 6.0 4.3·106 ± 4.6·105 3.0·104 ± 4.5·103 Dead elephant seal skin (Potter Cove coast) ndb 4.0·106 ± 4.9·105 3.5·104 ± 4.1·103 Seawater (Potter Cove) Surface 7.8 1.2·102 ± 6.9·10 3.8·103 ± 6.9·102 5 m deep 7.9 7.6·101 ± 7.0·10 2.9·103 ± 6.5·102 10 m deep 7.9 8.3·101 ± 6.2·10 3.1·103 ± 6.0·102 20 m deep 7.8 3.5·102 ± 1.5·101 2.7·103 ± 3.4·102 30 m deep 7.8 1.2·102 ± 9.0·10 2.9·103 ± 5.4·102 Soil (near Fourcade glacier) ndb 2.6·105 ± 4.1·104 1.0·104 ± 3.7·103 Algae (Desmarestia antarctica) 7.0 5.0·105 ± 5.3·104 4.6·105 ± 5.7·104 Sediment Pond (1 m deep) 7.2 7.5·103 ± 7.4·102 1.7·103 ± 5.7·102 Potter Cove (30 m deep) 7.7 5.9·103 ± 7.0·102 2.4·104 ± 4.1·103 a Colony forming units of the cultivable heterotrophic bacteria expressed as gram of dry weight except for the seawater samples, which are expressed as ml of sample. b Not determined. 66 Proteases from Antarctic bacteria 67Vazquez & Mac Cormack 2002: Polar Research 21(1), 63–71 ies, as they showed no proteolytic activity at any time in liquid culture, measured by the azocasein method. After centrifugation of cells, the proteas- es present in culture broth supernatants were par- tially characterized. When the strains isolated at 20 °C were tested, only 14 % of cell-free superna- tants showed optimal activity at 40 °C (the lowest apparent optimum temperature found). Neverthe- less, when the strains were isolated at 10 - 13 °C or at 4 - 6 °C, the percentages raised to 28 % and 46 %, respectively (Table 2). On the basis of their apparent optimum temperature for substrate uti- lization (Table 3), 25 protease-containing culture supernatants were selected. The isolation temper- ature of the 25 protease-producing strains is in- dicated in Table 4. These supernatants showed their maximal activity at 40 °C (Table 3) and con- tained neutral and alkaline proteases with regards to their optimum pH for activity (Table 4). The activation energy values for proteolytic activity were, in general, about 10 - 20 kJ·mol-1 lower than those observed for the mesophilic protease Carls- berg Subtilisin (Table 3). The development of proteolytic activity of cul- ture supernatants on SDS-PAGE with gelatine as a copolymerized substrate showed mainly mul- tiple-band profiles. Among the 25 protease-con- taining supernatants with the lowest optimum temperature, only 8 developed a single proteo- lytic band. Discussion Microbiological studies related to Antarctic en- vironments are frequently focused on the role of the bacterial community in heterotrophic produc- tion, in the cycling of matter, or in the relation- ship between bacterial activity and the dynam- ics of food webs. However, the majority of these studies do not discriminate between the different groups of bacteria that constitute the total flora responsible for one particular activity. When analyses of the bacterial flora of particular Ant- arctic environments were carried out, Gram-neg- ative bacteria proved to be highly predominant in marine habitats (Delille 1993), coastal zones (Tearle & Richard 1987; Line 1988) and terres- trial habitats covered with vegetation (Heal et al. 1967). This is consistent with the results present- ed here and could explain the fact that, in spite of the known capacity of Gram-positive bacteria to produce and secrete proteases, few of the selected strains were Gram-positive. Among marine iso- lates, Gram-positive bacteria were found most- ly in sediments while among non-marine iso- lates they were predominant in soil, mosses and lichens. Also, the number of spore-forming and pigmented strains was higher when the incuba- tion temperature was 20 °C. The predominance of these strains during the summer could be re- flecting their capability to withstand the rela- tively high temperatures and levels of UV radia- tion that prevail in this period, especially taking into account that the explored area is affected by the thinning of the ozone layer during spring and summer (Tong & Lighthart 1997). It is worth mentioning that most microorganisms, especial- ly in aquatic environments, are not culturable on conventional agar media and using conventional techniques. The considerations expressed above relate to the part of the bacterial community that can be isolated with the culture conditions used, corresponding to the aerobic chemoorganotro- phic psychrophilic and psychrotolerant cultur- able bacteria. The differences observed in the total counts ob- tained at 4 - 6 °C and at 20 °C with samples from environments having broad and narrow rang- es of temperature reflect the different degrees of adaptation showed by bacteria living in Ant- arctic areas with or without thermal oscillations. Table 2. Characteristics of the psychrotolerant Antarctic pro- teolytic bacteria, collected during two Antarctic Research Expeditions (ARE), isolated on casein agar using three different incubation temperatures. ARE 1995–96 ARE 1991–92 ARE 1995–96 Isolation Isolation Isolation 20 °C 10 - 13 °C 4 - 6 °C Pigmented strains 40 % 15 % 22 % Spore-forming 13 % 1 % 0 % Form rod-shaped 87 % 97 % 88 % cocci-shaped 13 % 3 % 2 % Gram stain positive 40 % 6 % 14 % negative 60 % 94 % 86 % Detection of proteases in liquid culture at 20 ºC 53 % 73 % 67 % Optimum temperature for proteolytic activity 40 °C 14 % 28 % 46 % 45 °C 29 % 7 % 46 % >50 °C 57 % 65 % 8 % Number of isolated strains 40 34 49 68 Proteases from Antarctic bacteria 69Vazquez & Mac Cormack 2002: Polar Research 21(1), 63–71 Since psychrotolerants grow faster at 20 °C than at 4 °C and we previously tested that the incuba- tion time was long enough to allow the develop- ment of psychrophiles and psychrotolerants colo- nies, the higher counts obtained with non-marine samples when the incubation was made at 20 °C might be related to the predominance of this type of microorganism in areas with a wide daily and seasonal variations in temperature (Delille 1990). On the contrary, with seawater and marine sedi- ment the counts obtained at 4 °C were higher than those obtained at 20 °C, reflecting the appropri- ateness of these constantly cold environments for the predominance of psychrophiles. The selected strains showed variable proteo- lytic activity levels in liquid cultures. Even when some of the proteases from psychrotolerants had their maximal activity at similar temperatures than their corresponding mesophile, the majori- ty of them showed curves of activity as a function of temperature shifted towards low temperatures, with apparent optimum values between 10 and 15 °C lower than Carlsberg Subtilisin. Other au- thors found similar results for proteases (Helmke & Weyland 1991; Turkiewicz et al. 1999) and other enzymes (Feller et al. 1997; Hoyoux et al. 2001) produced by marine bacteria from perma- nently cold environments. This observation re- veals a clear adaptation of the enzyme-producing strains to their habitats. The low activation ener- gies shown by many of the crude proteases also represents a way of adaptating to work in the cold and constitute a requirement for substrate hydro- lysis in cold environments and, thus, for surviv- al (Margesin et al. 1991; Feller & Gerday 1997). These enzymes compensate the low kinetic ener- gy in their environments by reducing the activa- tion energy barrier. In addition, all proteases with the lower op- timum temperature (40 °C) had a neutral or al- kaline optimum pH, probably influenced for the neutral pH of the isolation culture media or the pH of the habitats explored. Although the broad pH range in which some of our proteases showed maximum activity agree with the ones report- ed for other cold-adapted proteases (Hamamoto Table 3. Optimum temperature (OT), percent of relative activity measured at 20 °C with respect to the activity at optimum temperature (RA) and activation energy for azocasein hydrolysis (Eact) of crude proteases from 79 Antarctic psychrotolerant strains. Strain OT RA Eact Strain OT RA Eact Strain OT RA Eact (°C) (%) (kJ·mol-1) (°C) (%) (kJ·mol-1) (°C) (%) (kJ·mol-1) Ele-2 40 34 40 P95-8 45 17 45 P96-20 45 21 51 Ele-3 40 19 54 P95-9 45 22 46 P96-23 45 19 47 TCN-2 45 13 51 P95-16 45 20 49 P96-27 45 14 55 GUD-5 50 12 50 P95-17 45 26 49 P96-28 45 9 52 GUD-8 50 17 52 P95-18 45 29 49 P96-29 40 29 53 ANT-1-1 55 6 55 P95-19 45 24 44 P96-33 40 21 52 ANT-3-1 50 6 50 P95-20 >50 12 52 P96-35 40 35 40 ANT-7-1 55 6 55 P95-21 >50 25 47 P96-37 50 22 46 YOA-3 60 5 60 P95-24 40 30 36 P96-38 45 37 39 814 40 32 40 P95-27 40 10 52 P96-39 45 34 42 PIEL-1 40 29 40 P95-28 >50 10 52 P96-41 40 14 54 91 50 35 50 P95-29 >50 21 38 P96-43 40 31 47 435 50 30 50 P95-31 >50 27 33 P96-44 50 26 38 337 50 9 50 P95-32 >50 9 55 P96-45 45 14 50 273 50 13 50 P95-33 >50 13 41 P96-46 45 7 55 A1 50 6 50 P95-37 >50 25 51 P96-47 45 27 47 Prot-2 50 17 50 P95-38 >50 15 49 P96-48 45 21 50 Prot-4 50 20 50 P95-39 >50 19 55 P96-49 40 23 50 Prot-5 45 18 45 P95-40 >50 9 48 P96-50 40 35 48 Prot-8 40 34 40 P96-1 45 30 37 P96-51 40 16 87 Prot-9 40 13 40 P96-3 45 28 38 P96-52 40 22 74 Prot-10 50 11 45 P96-4 45 29 39 P96-53 40 24 55 Prot-11 40 14 40 P96-5 40 23 48 P96-54 40 22 76 Prot-12 55 7 55 P96-6 50 11 52 P96-55 45 40 29 Prot-14 45 29 45 P96-7 40 33 51 P96-56 45 21 49 P95-1 40 21 42 P96-12 40 13 51 P95-6 40 15 47 P96-18 40 24 50 Subtilisin 55 12 60 68 Proteases from Antarctic bacteria 69Vazquez & Mac Cormack 2002: Polar Research 21(1), 63–71 et al. 1994), this characteristic differs from that observed by other authors for low-temperature active proteases (Davail et al. 1994; Kärst et al. 1994) where a discrete value of optimum pH was found. In other cases, even though a discrete op- timum pH value was reported, there was a broad range where the proteases showed more than 80 % of their maximum activity (Margesin & Schinner 1992). The pH values for maximum ac- tivity of the studied proteases make them suitable for use in detergents and other industrial products requiring alkaline proteases, though the stabili- ty of the proteases at the pH of such products and processes must be confirmed. By developing proteolytic activity of culture supernatants on gelatine-SDS-PAGE, it was pos- sible to observe that some strains produced only one protease. These strains were selected for fur- ther studies (when their optimum temperature for activity was of interest) because they are easier to purify and characterize than a mixture of pro- teins. However, the majority of the strains pro- duced a multiple-band profile, which might have been the result of the activity of more than one protease produced by the strain, isoenzymes or the active fragments of self-digestion. The syn- thesis of more than one protease, which was ob- served mainly in marine strains, could be un- derstood as a strategy to better cope with the fluctuating supply of nutrients as well as to en- hance the uptake of proteins in oligotrophic envi- ronments. Although references to the number of proteolytic bands developed in SDS-PAGE from the culture supernatants of cold-adapted bacteria are infrequent in the literature, Turckiewicz et al (1999) recently found two different protein bands active against denatured haemoglobin working with an Antarctic marine strain of Sphingomo- nas paucimobilis. Given the relatively high cost and complexity of biotechnological processes based on the cul- ture of psychrophiles, which must be kept very cool, psychrotolerant organisms will be more suitable than psychrophiles to support an en- zyme production process at moderate tempera- tures (20 - 25 °C) while producing enzymes with high activity at low temperatures. Such a process would yield an enzyme with considerable activi- ty at temperatures significantly lower than those shown by the enzymes currently in use. On the basis of this concept, we chose 20 °C as the more suitable temperature to cultivate and produce proteases. We consider that this is an adequate temperature to carry out a process in a template climate zone, as it can be maintained without ex- pending costly energy in cooling the reactor. We also point out the relevance of selecting the proper conditions when designing a screen- ing programme for the selection of proteolytic bacteria in the explored area. Isolation tempera- ture was shown to be an important factor to take into account to direct the screening towards the isolation of bacteria producing the desired type of proteases. To conclude, the results presented here suggest that the selected psychrotolerant strains are po- tentially useful for industrial applications, and that further studies will be necessary to opti- mize the natural capability of these strains to pro- duce and secrete proteases. Moreover, we would like to emphasize the relevance of the Antarc- tic environment as a source of psychrotolerant microorganisms whose physiological character- istics are often unusual in comparison to the cor- responding mesophiles (Shivaji et al. 1989) and which offer new possibilities in biotechnology. Acknowledgements.—This study was supported by the In- stituto Antartico Argentino and the Universidad de Buenos Aires (grant FA 059). We wish to thank C. Ferreiro for cor- recting the English manuscript. References Aguilar, A. 1996: Extremophile research in the European Union: from fundamental to industrial applications. FEMS Table 4. pH values for maximal activity of crude proteases from Antarctic strains showing optimum temperature for proteolytic activity at 40 °C. Strain pH Strain pH Ele-2*,1 8 P96-18*,3 7 - 9 Ele-31 8 - 11 P96-293 7 - 9 Prot-81 8 - 9 P96-333 8 Prot-9*,1 6 - 11 P96-35*,3 7 - 9 Prot-11*,1 8 P96-413 7 - 9 8141 8 - 9 P96-433 7 - 9 PIEL-11 8 - 9 P96-493 8 P95-1*,2 7 - 9 P96-503 9 P95-21*,2 7 - 9 P96-513 8 P95-24*,2 7 - 10 P96-523 8 - 10 P96-53 8 P96-533 7 - 9 P96-73 9 P96-543 9 P96-123 7 *Single proteolytic band developed in gelatin-SDS-PAGE. 1,2,3 Strain isolated at 10 - 13 ºC, 20 ºC and 4 - 6 ºC, respectively. 70 Proteases from Antarctic bacteria 71Vazquez & Mac Cormack 2002: Polar Research 21(1), 63–71 Microbiol. Rev. 18, 89–92. Asakawa, M., Sadakata, Y., Araki, T., Sumi, T. & Nakagawa, H. 1998: Purification and characterization of the alkaline serine protease produced by Bacillus sp. N4 strain from fish skin mucus. Fish. Sci. 64, 793–797. Charney, J. & Tomarelli, R. M. 1947: A colorimetric method for the determination of the proteolytic activity of duodenal juice. J. Biol. Chem. 171, 501–507. Davail, S., Feller, G., Narinx, E. & Gerday, C. 1994: Cold adaptation of proteins. Purification, characterization, and sequence of the heat labile subtilisin from the Antarctic psychrophile Bacillus TA41. J. Biol. Chem. 269, 1–6. Delille, D. 1990: Seasonal changes of sub-Antarctic hetero- trophic bacterioplankton. Arch. Hydrobiol. 119, 267–277. Delille, D. 1992: Marine bacterioplankton at the Weddel Sea ice edge, distribution of psychrophilic and psychrotrophic populations. Polar Biol. 12, 205–210. Delille, D. 1993: Seasonal changes in the abundance and com- position of marine heterotrophic bacterial communities in an Antarctic coastal area. Polar Biol. 13, 463–470. Delille, D. & Perret, E. 1989: Influence of temperature on the growth potential of southern polar marine bacteria. Micro- biol. Ecol. 18, 117–123. Demain, A. L. 1999: Stunning achievements of industrial microbiology. ASM News 65, 311–316. Feller, G. & Gerday, C. 1997: Psychrophilic enzymes: molec- ular basis of cold adaptation. Cell. Mol. Life Sci. 53, 830– 841. Feller, G., Zekhnini, Z., Lamotte-Brasseur, J. & Gerday, C. 1997: Enzymes from cold-adapted microorganisms. The class C β-lactamase fron the Antarctic psychrophile Psy- chrobacter immobilis A5. Eur. J. Biochem. 244, 186–191. Gerday, C., Aittaleb, M., Bentahir, M., Chessa, J. P., Claver- ie, P., Collins, T., D’Amico, S., Dumont, J., Garsoux, G., Georlette, D., Hoyoux, A., Lonhienne, T., Meuwis, M. A. & Feller, G. 2000: Cold-adapted enzymes: from fundamen- tals to biotechnology. Trends Biotechnol. 18, 103–107. Hamamoto, T., Kaneda, M., Horikoshi, K. & Kudo, T. 1994: Characterization of a protease from a psychrotroph, Pseu- domonas fluorescens 114. Appl. Environ. Microbiol. 60, 3878–3880. Han, X. Q. & Damodaran, S. 1998: Purification and charac- terization of protease Q: a detergent- and urea-stable serine endopeptidase from Bacillus pumilus. J. Agric. Food Chem. 46, 3596–3606. Heal, O. W., Bailey, A. D. & Latter, P. M. 1967: Bacteria, fungi and protozoa in Signy Island soils compared with those from a temperate moorland . Philos. Trans. R. Soc., Lond., Ser. B252, 191–197. Helmke, E. & Weyland, H. 1991: Effect of temperature on extracellular enzymes occurring in permanently cold marine environments. In G. Rheinheimer et al. (eds): Dis- tribution and activity of microorganisms in the sea. Pro- ceedings of the 4th European Marine Microbiology Sym- posium. Kiel, 1990. Pp. 198–204. Institute for Marine Research, University of Kiel. Heussen, C. & Dowdle, E. B. 1980: Electrophoretic analysis of plasminogen activators in polyacrilamide gels contain- ing sodium dodecyl sulfate and copolymerised substrates. Anal. Biochem. 102, 196–202. Horikoshi, K. 1995: Discovering novel bacteria, with an eye to biotechnological applications. Curr. Opin. Biotechnol. 6, 292–297. Hoshino, T., Ishizaki, K., Sakamoto T., Kumeta, H., Yumoto, I., Matsuyama, H. & Ohgiya, S. 1997: Isolation of a Pseu- domonas species from fish intestine that produces a pro- tease active at low temperature. Lett. Appl. Microbiol. 25, 70–72. Hoyoux, A., Jennes, I., Dubois, P., Genicot, S., Dubail, S., François, J. M., Baise, E. Feller, G. & Gerday, C. 2001: Cold-adapted β-galactosidase from the Antarctic psychro- phile Pseudoalteromonas haloplanktis. Appl. Environ. Microbiol. 67, 1529–1535. Irgens, R. L., Gosink, J. J. & Staley, J. T. 1996: Polaromonas vacuolata gen. nov., sp. nov., a psychrophilic, marine, gas vacuolate bacterium from Antarctica. Int. J. Syst. Bacteri- ol. 46, 822–826. Kamal, M., Hoeoeg, J. O., Kaiser, R., Shafgat, J., Razzaki, T., Zaidi, Z. H. & Joernvall, H. 1995: Isolation, characteriza- tion and structure of subtilisin from a thermostable Bacil- lus subtilis isolate. FEBS Lett. 374, 363–366. Karl, D. M. 1993: Microbial processes in the southern oceans. In E. I. Friedman (ed.): Antarctic microbiology. Pp. 1–63, New York: Wiley Press. Kärst, U., Woehl, M., Czempinski, K. & Schmid, R. D. 1994: Characterization of extracellular hydrolases from marine psychrophilic bacteria. In L. Alberghina et al. (eds.): Prog- ress in biotechnology: proceedings of the 6th European Congress on Biotechnology. Florence, 1993. Pp. 263–266. Amsterdam: Elsevier Science. Line, M. A. 1988: Microbial flora of some soils of Mawson Base and the Vestfold Hills, Antarctica. Polar Biol. 8, 421– 427. Lyman, J. & Fleming, R. H. 1940: Composition of seawater. J. Mar. Res. 3, 134–146. Margesin, R., Palma, N., Knauseder, F. & Schinner, F. 1991: Proteases of psychrotrophic bacteria isolated from glaciers. J. Basic Microbiol. 31, 377–383. Margesin, R. & Schinner, F. 1992: A comparison of extra- cellular proteases from three psychrotrophic strains of Pseudomonas fluorescens. J. Gen. Appl. Microbiol. 38, 209–225. Morita, R. Y. 1975: Psychrophilic bacteria. Bacteriol. Rev. 39, 144–167. Pirt, S. J. 1985: Effects of temperature. In S. J. Pirt (ed): Principles of microbial and cell cultivation. Pp. 137–142. London: Blackwell. Reddy, G. S. N., Aggarwal, R. K., Matsumoto, G. I. & Shivaji, S. 2000: Arthrobacter flavus sp. nov., a psychrophilic bac- terium isolated from a pond in McMurdo Dry Valley, Ant- arctica. Int. J. Syst. Evol. Microbiol. 50, 1553–1561. Rivkin, R. B., Anderson, M. R. & Lajzerowicz, C. 1996: Microbial processes in cold oceans. I: relationships between temperature and bacterial growth rate. Aquat. Microb. Ecol. 10, 243–254. Schloss, I., Ferreyra, G. & Klöser, H. 1998: Seasonal varia- tion of the conditions for phytoplankton growth in Potter Cove. Ber. Polarforsch. 299, 59–66. Shivaji, S., Shyamala, R. N., Saisree, L., Sheth, V., Reddy, G. S. N. & Bhargava, P. H. 1989: Isolation and identification of Pseudomonas spp from Schirmacher Oasis, Antarctica. Appl. Environ. Microbiol. 55, 767–770. Suhartono, M. T., Satiawihardja, B., Widjaja, I. P. N. & Han- doko, T. W. T. 1997: Characterization of protease enzyme produced from Bacillus pumilus Y1 isolated from liquid tofu waste. Asian Pac. J. Mol. Biol. Biotechnol. 5, 72–80. Tearle, P. V. & Richard, K. J. 1987: Ecophysiological group- ing of Antarctic environmental bacteria by API20 and fatty acid fingerprints. J. Appl. Bacteriol. 63, 497–503. Tong, Y. & Lighthart, B. 1997: Solar radiation in shown to 70 Proteases from Antarctic bacteria 71Vazquez & Mac Cormack 2002: Polar Research 21(1), 63–71 select for pigmentation bacteria in the ambient outdoor atmosphere. Photochem. Photobiol. 65, 103–106. Turkiewicz, M., Gromek, E., Kalinowska, H. & Zielinska, M. 1999: Biosynthesis and properties of an extracellu- lar metalloprotease from the Antarctic marine bacterium Sphingomonas paucimobilis. J. Biotechnol. 70, 53–60. Varela, H., Ferrari, M. D., Belobrajdic, L., Vazquez, A. & Loperena, M. L. 1997: Skin unhairing proteases of Bacillus subtilis: production and partial characterization. Biotech- nol. Lett. 19, 755–758. Vazquez, S. C., Rios Merino, L. N., Mac Cormack, W. P. & Fraile, E. R. 1995: Protease-producing psychrotrophic bac- teria isolated from Antarctica. Polar Biol. 15, 131–135. Vishniac H. S. 1993: The microbiology of Antarctic soils. In: E. I. Friedmann (ed.): Antarctic microbiology. Pp. 292–341. New York: Wiley-Liss Press.