Cold adaptation in Arctic and Antarctic fungi


  • Clare H. Robinson

    Corresponding author
    1. Division of Life Sciences, King’s College, University of London, Franklin-Wilkins Building, 150 Stamford Street, London SE1 9NN, UK
      Author for correspondence: Clare H. Robinson Tel: +44 (0)20 7848 4352 Fax: +44 (0)20 7848 4500
    Search for more papers by this author

Author for correspondence: Clare H. Robinson Tel: +44 (0)20 7848 4352 Fax: +44 (0)20 7848 4500


Growth and activity at low temperatures and possible physiological and ecological mechanisms underlying survival of fungi isolated from the cold Arctic and Antarctic are reviewed here. Physiological mechanisms conferring cold tolerance in fungi are complex; they include increases in intracellular trehalose and polyol concentrations and unsaturated membrane lipids as well as secretion of antifreeze proteins and enzymes active at low temperatures. A combination of these mechanisms is necessary for the psychrotroph or psychrophile to function. Ecological mechanisms for survival might include cold avoidance; fungal spores may germinate annually in spring and summer, so avoiding the coldest months. Whether spores survive over winter or are dispersed from elsewhere is unknown. There are also few data on persistence of basidiomycete vs microfungal mycelia and on the relationship between low temperatures and the predominance of sterile mycelia in tundra soils. Acclimation of mycelia is a physiological adaptation to subzero temperatures; however, the extent to which this occurs in the natural environment is unclear. Melanin in dark septate hyphae, which predominate in polar soils, could protect hyphae from extreme temperatures and play a significant role in their persistence from year to year.


In 1967, Farrell and Rose wrote ‘the existence of microorganisms that are capable of growing well at near-zero temperatures … has been recognized for almost a century’ (Farrell & Rose, 1967), although this group of microorganisms was relatively neglected until the 1950s. There are several reasons for the continuing interest in low temperature tolerance in fungi. One is the importance of these organisms as agents of spoilage of refrigerated and frozen foods, and a second is their potential commercial value as sources of cold-active enzymes (Margesin & Schinner, 1994). These cold-tolerant fungi also have a biogeographical and ecological significance. Despite the general similarities of species spectra of decomposer microfungi between tundra and other biomes, it is the psychrophilic fungal component of some tundra areas that distinguishes them from other ecosystems (Flanagan & Scarborough, 1974).

The physiological and ecological mechanisms in cold-tolerant fungi that permit low temperature growth are still not fully understood (Russell, 1990; Smith, 1993; Cairns et al., 1995b; Snider et al., 2000; Weinstein et al., 2000), and have not been previously reviewed. Russell (1990) discussed mainly the biochemistry of adaptations of microorganisms to low temperatures with largely bacterial examples, whereas the aim here is to review the now quite large number of examples of growth and activity of Arctic and Antarctic fungi at low temperatures, and to outline the possible physiological and morphological mechanisms underlying fungal survival at such temperatures. This paper is also unique in collating examples from the Arctic and the Antarctic.


In this review, the definitions of psychrophiles and psychrotrophs follow those of Morita (1975) and Gounot (1986), developed mainly with reference to bacteria. Both psychrophilic and psychrotrophic fungi have the ability to grow at 0°C. Psychrophilic fungi have an optimum temperature for growth of c. 15°C or lower, and a maximum temperature for growth of 20°C or below, whereas psychrotrophic fungi have a maximum temperature for growth above 20°C.

Biologists generally accept the definition of the Arctic as those lands beyond the climatic limit of trees (Bliss & Matveyeva, 1992). Terrestrial ecosystems in Antarctica include a great variety of habitats from ice-free areas of the continent to the comparatively warm sub-Antarctic. Antarctic latitudes are colder than their equivalent northern counterparts (Convey, 1996), especially in summer. The cold Arctic consists of the northern fringes of Ellesmere Island and its ‘neighbours’, Svalbard, Franz Joseph Land, Novaya Zemlya and the New Siberian Islands, all in the region of 80–85° N, whereas the comparable cold (or maritime) Antarctic comprises the western side of the Antarctic Peninsula and islands of the Scotia Arc (South Shetland, South Orkney and South Sandwich Islands) at only 55–68° S. The remainder of the Antarctic land mass falls within the continental zone, which has mean monthly temperatures only rarely and locally exceeding 0°C in summer (Lewis Smith, 1984; Convey, 1996). A large proportion of the relatively small ice-free area of continental Antarctica consists of cold deserts, where microbial community development is mainly restricted to three types of habitat: endolithic communities inside rocks, freshwater communities in transient water bodies, and hypersaline ice-covered lakes (Wynn-Williams, 1990). These continental desert communities are not considered further here. There are no comparable terrestrial ecosystems in the Arctic, although similar temperatures are experienced on nunataks in the Greenland icecap (Convey, 1996). The current review encompasses studies of mycorrhizal and filamentous decomposer fungi plus yeasts from cold Arctic and Antarctic environments, therefore including basidiomycetes, ascomycetes and microfungi (but not lichenized fungi). The range of temperatures experienced in examples of cold Arctic and maritime Antarctic environments is shown in Fig. 1.

Figure 1.

Temperatures in examples of cold Arctic and maritime Antarctic environments. (a) Mean daily September 1993 to July 1994 soil temperatures at 3 cm depth at a polar semi-desert site at Ny-Ålesund, Svalbard, Norway, 79° N (after Coulson et al., 1995) (b) Daily surface soil temperature ranges (maximum and minimum values) during 1987 in the centre of a polygon at Jane Col, Signy Island, South Orkney Islands, 60° S (Davey et al., 1992).

Examples of growth, activity and viability of fungi at low temperatures


The majority of isolates tested for growth, activity and viability appear to be psychrotrophic (Table 1). Latter & Heal (1971) showed that strains from a single fungal species may differ in their physiology depending on their climatic origin. The predominance of psychrotrophy, rather than psychrophily, in Arctic and Antarctic environments may be because, while these fungi still have the ability to grow around 0°C, temperatures of substrata at some times of the year are much higher than low air temperatures. For example, Möller & Dreyfuss (1996) wrote ‘Although average air temperatures in the maritime Antarctic are around freezing point, local soil temperatures and microclimates may rise to 15°C through solar radiation’. Thus a relatively small percentage of fungal strains, estimated to be 10–20% of fungal species and strains from Alaskan tundra sites (Flanagan & Scarborough, 1974), and 10% of tested isolates from two Antarctic sites on King George Island, South Shetlands, Antarctica (Möller & Dreyfuss, 1996), appears to be truly psychrophilic. The low proportion of psychrophilic isolates in the strict sense from Arctic and Antarctic environments may be because the isolations have not been performed in winter when low temperatures prevail, or have not been performed from substrata experiencing solely low temperatures.

Table 1.  Examples of growth, activity and viability of fungi at low temperatures
TaxonTest temperature (°C) Psychrophile/psychrotroph?Isolation siteType of experimentReference
A. Examples of ‘growth’ experiments
(1) Arctic sites
54 isolates, mainly mitosporic fungiGrowth at 0 and 25, mainly psychrotrophicBarren soils, Franz Joseph Land, 79°50′ to 81°50′ NMalt extract agar mediumBergero et al. (1999)
Phoma herbarumGrowth at 2.5–20 Higher growth rate at 2.5 than 5, psychrotrophicSoil, Truelove Lowland, Devon Island Canada, 75°33′  N, 84°40′  WSand + glucose-mineral agar mediumWidden & Parkinson (1978)
(2) Antarctic sites
31 isolates, mainly mitosporic fungiGrowth at 4 and 35. All isolates grew at 4, most optima at 15 or 20, mainly psychrotrophicPlant and soil, Subantarctic Macquarie Island (55°30′  S, 158°57′  E); Antarctic Casey Station (64°17′  S, 100°32′  E)Tomato agar medium, Lilly & Barnett’s synthetic mediumKerry (1990)
Humicola marviniiGrowth at −2.5, optimum at 15, no growth at 20–22, psychrophilicFellfield soils, Signy Island, 60°43′  S, 45°38′  WMalt extract + mycological peptone nutrient agar mediumWeinstein et al. (1997)
14 taxa, mainly mitosporic fungiAll grew at 1, no higher temperatures tested, pyschrophilic/psychrotrophic?Soil, Signy Island, 60° S, 45°  WCzapek Dox + trace elements agar mediumLatter & Heal (1971)
35 isolates, mainly mitosporic fungiGrowth at 0–35, 33 strains grew at 5, 31 strains grew at 0. Most optima at 15 or 20, mainly psychrotrophicMainly soil and moss, Victoria Land, 72°30′  S to 77°52′  SPotato dextrose agar mediumZucconi et al. (1996)
B. Examples of ‘activity’ experiments
(1) Arctic sites
Phialophora hoffmanniCladosporium cladosporioidesOccurs at 1, optima between 18 and 20+, psychrotrophicSoil and litter, Point Barrow, 71°17′  N, 156°41′  win vitro pectinase activityFlanagan & Scarborough (1974)
Geomyces pannorumOccurs at −4, optima at 5, psychrophilicSoil and litter, Point Barrow, 71°17′  N, 156°41′  win vitro cellulase activityFlanagan & Scarborough (1974)
(2) Antarctic sites
Verticillium lecaniiActive at 5, 15 and 25, psychrotrophicMoss samples in continental Antarcticain vitro chitinase activityFenice et al. (1998)
C. Example of ‘viability’ experiment from an Antarctic site
Soil fungiGreater number of colonies when plates incubated at 10, rather than 25, psychrotrophic?Six sites at Signy Island, 60°  S, 45°  WSoil dilution and Warcup platesBailey & Wynn-Williams (1982)

An important point to note when obtaining fungal species from low temperature environments is that the isolation temperature may bias their frequency of isolation. Very different microfungal species spectra were obtained from forest soils in Rhode Island when two different isolation temperatures (0 and 25°C) were used (Carreiro & Koske, 1992). Most fungi isolated at 0°C were psychrotrophs, although some psychrophiles, mostly Mortierella and Mucor spp., were also isolated. Isolations at 25°C resulted in mostly mesophiles with growth minima between 5 and 10°C and maxima above 25°C. Cold temperature also appeared to have a selective effect on Geomyces pannorum because the isolation frequency increased as incubation of the soil and agar plates approached 0°C (Ivarson, 1973).

Physiological effects of low temperature and freezing on fungi

Low temperature is a relative term (Smith, 1993). In biology it is usually identified with subzero temperatures with a lower limit of −70°C, below which no life processes persist. From Fig. 1 it is apparent that the examples chosen can be defined as low temperature environments. Soil temperatures at 3 cm depth at a polar semi-desert site in Ny-Ålesund, Svalbard, Norway (78°56′  N, 11°50′  E) were below zero for 272 days, with minimum monthly winter temperatures ranging from −5.6 to −25.0°C (Coulson et al., 1995, Fig. 1a). Summer temperatures are likely to be more conducive to fungal growth, since at 5 cm depth, mean, minimum and maximum summer soil temperatures at a similar site were 6.1, 1.4 and 11.4°C (Wookey et al., 1993). Similarly, mean daily soil surface temperatures at Jane Col, Signy Island (60°43′ S, 45°38′  W) were below zero for a large part of the year (Fig. 1b), with 25 freeze-thaw cycles per annum. At 3 cm depth, the lowest monthly minimum temperature in the same polygon centre was −7.6°C in July, August and October (Davey et al., 1992).

Cooling to low temperatures reduces the rate at which chemical reactions occur, increases the viscosity of water, denatures protein and increases the relative permittivity of water thus reducing attraction between ions of opposite charge, and markedly affecting acidic and basic residues of proteins. In relation to the lower growth temperature limit of psychrophiles, there are no substantiated reports of microbial growth at temperatures below −12°C, which is consistent with the known physical state of aqueous solutions at subzero temperatures (Russell, 1990; Mazur, 1980). Dilute aqueous solutions will generally supercool to −10°C, occasionally to −20°C, and most cells remain unfrozen at −10 to −15°C even though these temperatures are 9–14 degrees below the freezing point of their cytoplasm and there is extracellular ice in the growth medium. Nucleation of the supercooled cytoplasmic water does not occur above this temperature because small ice-nuclei are barred from entering the cell by the plasma membrane. Supercooled water has a higher vapour pressure than that of extracellular ice, so water will move out of the cell, thereby concentrating the intracellular milieu. At temperatures below −10 to −15°C, the cell water begins to freeze, further concentrating intracellular salts up to 3 molal. The resulting ionic imbalance, altered pH and lowering of water activity have a toxic effect on the microorganism, which will either prevent it from functioning or possibly kill it. Thus the lower growth temperature limit of psychrophiles is fixed, not by the cellular properties of cellular macromolecules, but instead by the physical properties of aqueous solvent systems inside and outside the cell.

Cryoinjury is largely dependent on the rate of cooling, the cell type (whether it is sensitive or resistant to cold), and the composition of the suspending medium (Smith, 1993). Most work has been carried out with a view to using ultralow temperature for long-term preservation of viability and stability of fungi, aiming to minimize the two damaging factors of dehydration and intracellular freezing. Early studies established that slow cooling in the presence of a cryoprotectant such as glycerol or dimethyl sulfoxide (DMSO) and storage at temperatures below −139°C, quite often in liquid nitrogen or the vapour above it, gave best results when applied routinely (Smith, 1993).

Environmental factors related to low temperature: freeze-thaw and desiccation

It may not be adaptation to low temperature per se which influences fungal survival in such environments but rather, for example, to freeze-thaw cycles or desiccation. Temperature fluctuations around 0°C are characteristic of the substrata of Arctic and Antarctic environments. Indeed, Coulson et al. (1995) found that at 3 cm depth in a high Arctic polar semi-desert soil, 30 freeze-thaw cycles occurred during winter. Vishniac (1996) stated ‘Many investigators have examined the ability of their filamentous isolates to grow at near freezing temperatures, but not their ability to survive freeze-thawing. It is generally thought that unprotected hyphomycete hyphae do not survive freezing, while spores often do’. By contrast, Lyakh et al. (1984), cited in Wynn-Williams (1990), have shown that after periodic freezing to −13°C and thawing the Antarctic yeast Nadsoniella nigra var. hesuelica, 33% of the population was viable after one cycle and 10% was still viable after 10 cycles.

Desiccation is possibly as important an influence on survival as low temperatures, and the adaptations to both factors may be similar. Water (and substrate) availability rather than temperature is reported to regulate microbial activity in the maritime and sub-Antarctic (Wynn-Williams, 1990). Similarly, continental Antarctica is a cold desert and free water is only present intermittently (McRae & Seppelt, 1999).

Physiological mechanisms of cold tolerance

Several physiological mechanisms of cold-tolerance by fungi have been proposed, and it is possible that a combination of these strategies is employed. For example, Table 2 shows a selection of characteristics that would be advantageous to soil organisms for winter survival in Arctic soils (Hodkinson & Wookey, 1999), and it is likely that physiological mechanisms conferring these characteristics occur in fungi.

Table 2.  Ecophysiological characteristics of Arctic soil organisms exhibiting good winter survival (Hodkinson & Wookey, 1999)
Winter survival – high survival characteristics
Capacity to dehydrate
Produce antifreezes
High supercooling activity
High chill tolerance
Freeze tolerance
Behavioural selection of microhabitat
Live in habitats with snow cover
Survive anoxia (being encased in ice)

Trehalose and cryoprotectant sugars

Trehalose is an important storage compound in fungal vegetative cells and spores (Lewis & Smith, 1967), and is the most widely distributed disaccharide in fungi (Thevelein, 1984). In fungal vegetative structures, trehalose is usually found together with sugar alcohols and glycogen. This also occurs in reproductive structures, but in this case trehalose is often present in much higher concentrations than other storage carbohydrates (Thevelein, 1984). According to Cooke & Whipps (1993), trehalose appears to be a general stress protectant in the cytosol, and it is known to stabilize membranes during dehydration (Goodrich et al., 1988). More recently, several authors have also shown accumulations of trehalose in fungal hyphae in response to low temperatures. Concentrations of trehalose were shown to double in excised alpine mycorrhizal roots when they were exposed to low temperatures (Niederer et al., 1992) and comparative studies of arctic and temperate strains of Hebeloma spp. have indicated substantial accumulations of trehalose in the arctic species when grown at low temperature (Tibbett et al., 1998a). Similarly, Humicola marvinii, a psychrophile, isolated from fellfield soil at Jane Col, Signy Island in Antarctica, grown at 5°C and 15°C in liquid medium accumulated trehalose intracellularly to a greater extent at 5°C than at 15°C (Weinstein et al., 2000). In the same study, Mortierella elongata, a psychrotrophic fungus isolated as above and incubated at 5°C, showed intracellular trehalose concentrations which were increased by 75% compared with incubation at 15°C (Weinstein et al., 2000).


Glycerol and mannitol may increase in concentration to maintain turgor pressure against heat-mediated decreases in external water potential (Cooke & Whipps, 1993). Mannitol is thought to be important in protection against water stress (Lewis & Smith, 1967), and may be a cryoprotectant (Weinstein et al., 1997). Jennings (1984) thought of polyols as ‘… acting as “physiological buffering agents”[in fungi] in that they … probably maintain a suitable milieu for enzyme activity’. Evidence for a potential cryoprotectant role of polyols comes from a study by Weinstein et al. (1997), using an Antarctic isolate of Humicola marvinii compared with H. fuscoatra which had been isolated from Gossypium sp. and purchased from a culture collection. In still liquid culture incubated for 8 wk at 15°C, the quantity of total sugars produced by the two isolates was not significantly different. However, the amounts of individual sugars and polyols produced by the two species differed greatly (Table 3). The clearest differences were in mannitol, known for its cryoprotectant properties, which was found at high levels in H. marvinii but which was scarce in H. fuscoatra. Conversely, H. fuscoatra produced glucose and fructose, compounds not known for their cryoprotectant properties, while H. marvinii produced no fructose and very little glucose.

Table 3.  Polyol and sugar production (mg−1 100 mg d. wt) produced after 8 wk at 15°C by Humicola marvinii isolated from maritime Antarctic soil, and Humicola fuscoatra isolated from Gossypium and bought from a culture collection (Weinstein et al., 1997). Differences in polyol and sugar production by the two fungi are statistically different at the 99.8% confidence level in all cases
Polyol/sugarH. marviniiH. fuscoatra
Glycerol 0.350.80
Erythritol 0.270.11
Trehalose 7.764.51
Glucose 0.188.07
FructoseNot detected5.03
Arabitol 0.21Not detected


There is considerable evidence to suggest that membrane composition can determine the ability of fungi to grow over specific temperature ranges (Cooke & Whipps, 1993). The structure and composition of membranes is likely to affect the temperature at which their properties change from an inactive gel phase to an active, crystalline phase. In psychrophilic yeasts, for example species of Candida, Leucosporidium and Torulopsis, constituent fatty acids are more unsaturated than those of mesophiles and lowered incubation temperatures increase this degree of unsaturation (Kerekes & Nagy, 1980). A similar pattern of change has also been found for some Mucor species (Dexter & Cooke, 1984a,b). However, with other Mucor species, membrane phospholipids rather than general fatty acids (which can include storage lipids) differ between psychrophiles, mesophiles and thermophiles (Hammonds & Smith, 1986), with levels of membrane phospholipid unsaturation decreasing from psychrophile, to mesophile to thermophiles. Proteins and sterols within membranes can also influence their stability (Dexter & Cooke, 1985).

Experimental evidence for an increase in unsaturated lipid content with low temperature is provided by Weinstein et al. (2000). A psychrotrophic isolate of Geomyces pannorum grown at 5°C exhibited altered lipid composition compared with the same isolate grown at 15°C, with increases in unsaturated lipid content and overall unsaturation index. Mortierella elongata, also grown at 5°C in the same study, showed an absence of detectable ergosterol but presence of stearidonic acid, a fatty acid only previously reported in another species of psychrotrophic zygomycete (Weinstein et al., 2000). Snow moulds, which are pathogenic to winter cereals and ley grass in taiga and boreal zones characterized by a persistent snow cover on frozen soil throughout the winter and temperatures in the range −3 to +3°C (Jamalainen, 1974; Gaudet et al., 1999), must also have mechanisms to maintain the fluidity of their membrane structures and thereby grow actively at low temperatures. Since membrane fluidity varies with the degree of unsaturation of lipids, the abundance of polyunsaturated fatty acids (18 : 2 and 18 : 3) among the phospholipids of Microdochium nivale would enhance the ability of the organism to survive at lower temperatures (Istokovics et al., 1998).

Antifreeze proteins

Extracellular and intracellular antifreezes may allow fungi to be active at subzero temperatures and they may slow the growth of ice if crystallization does occur. Fungi require the maintenance of an aqueous environment for growth to secrete enzymes and absorb carbon and nutrients. Moreover antifreezes may be essential for inhibiting the recrystallization of ice and promoting fungal survival through freeze-thaw cycles. In addition to preventing hyphae from freezing at temperatures just below zero, antifreezes produced by fungi may keep substrates from freezing since these compounds would be otherwise unavailable for use (Snider et al., 2000). Antifreeze proteins (AFPs), which are thought to contribute significantly to survival at subzero temperatures by modifying the growth of ice, are found in bacteria (Xu et al., 1997), plants (Sidebottom et al., 2000), invertebrates (Duman et al., 1991) and fish (Griffith & Ewart, 1995). Although there appear to be no reports to date of AFPs or antifreeze activity in fungi from cold Arctic and Antarctic environments, an AFP showing epitopic homology to one found in the Atlantic winter flounder, was found in the hyphae of three psychrophilic snow moulds, the ascomycete Sclerotinia borealis, and two basidiomycetes, Coprinus psychromorbidus and Typhula incarnata (Newsted et al., 1994). However, the proteins from this study do not appear to have been characterized more fully, and the best recent evidence of antifreeze activity in snow-mould fungi comes from the work of Snider et al. (2000), from which the following information is taken. Isolates of Typhula incarnata, T. ishikariensis and T. phacorrhiza showed antifreeze activity in all fractions (the growth medium of 2% malt extract broth, the soluble hyphal fraction, and the insoluble hyphal fraction). No antifreeze activity was found in isolates with peak growth temperatures above 14°C (M. nivale, S. borealis and S. homeocarpa), whereas antifreeze activity was detected in isolates with peak growth temperatures at 4, 10 (Typhula species), or 14°C (Coprinus psychromorbidus). The antifreeze activity found in the growth medium of T. phacorrhiza isolate Tp94614 was shown to arise from protein molecules. The ice crystal structures associated with snow mould species showed different growth patterns from those previously observed, which suggests that these AFPs may bind to different planes of the ice crystal lattice than those found with AFPs from fish, insects and plants.

Enzyme activity at low temperatures

The growth of psychrotrophs and psychrophiles at low temperatures has led to the search for enzymes with psychrophilic or ‘cold active’ properties (Cairns et al., 1995b), and conversely isolation of cold-active enzymes may also contribute to an understanding of how these fungi thrive at low temperatures (Weinstein et al., 2000). From a further ecological viewpoint, interest in ‘cold active’ enzymes has been spurred by the fact that low diversity of fungal species in soils in continental Antarctica (McRae & Seppelt, 1999) and sub-Antarctic islands (Steiman et al., 1995) is hypothesized to be offset by a wide variety of enzymes produced by each species (Fenice et al., 1997).

The following section concerns examples of cold-active enzymes found in decomposer and mycorrhizal fungi in cold Arctic and Antarctic environments. Enzyme activity has been found at low temperatures in soil fungi isolated from Antarctica. In a study by Weinstein et al. (1997), using the psychrophile Humicola marvinii isolated from fellfield soil in Antarctica and H. fuscoatra isolated from the rhizosphere of Gossypium sp. in Nigeria, H. marvinii was capable of solubilizing inorganic phosphate and producing extracellular protease enzymes in agar media at 15°C, whereas H. fuscoatra was not. Using similar plate-screening techniques on soil fungi, 33 strains (23 mitosporic fungi, three ascomycetes, three sterile mycelial strains and four yeast or yeast-like fungi) isolated from different sites on Victoria Land, continental Antarctica, were tested for their ability to produce 12 extracellular enzymes at the relatively high temperature of 25°C (except for protease at 20°C), or at the species’ optimum growth temperature if not 25°C (Fenice et al., 1997). Lipases were generally present, and in high quantities in almost all strains. Polygalacturonase, as well as amylase and phosphatase, was common. Glucose oxidase, protease and DNAase appeared to be generally low or absent.

The optimum temperatures for phosphatase and proteolytic activity by Arctic strains of fungi of an ectomycorrhizal genus Hebeloma have been characterized in an excellent series of papers by Tibbett et al. (1998a,b, 1999). Twelve Hebeloma strains, two from arctic tundra in Svalbard, seven from near Fairbanks, Alaska, two from French forests and one from Scottish forest were tested for phosphatase activity (Tibbett et al., 1998a). At temperatures lower than or equal to 12°C, arctic strains produced more extracellular and wall-bound acid phosphatase, yet grew more slowly than the temperate strains. The authors suggested that low growth rates in arctic strains may be a physiological response to cold whereby resources are diverted into carbohydrate accumulation for cryoprotection. At near freezing temperatures, increased extracellular phosphatase production may compensate for a loss of enzyme activity at low temperature and serve to hydrolyse organic phosphorus in frozen soil over winter. In a second, related, paper, Tibbett et al. (1998b) grew Hebeloma strains of arctic and temperate origin at 22°C or 6°C, which were assayed for wall-bound and extracellular acid phosphomonoesterase (PNPPase) across a temperature range of 2–37°C. Only when grown at 6°C was a cold-active extracellular PNPPase induced in all the arctic strains and most of the temperate strains tested. Such enzymes are suggested to be an adaptation to low soil temperatures and may allow ectomycorrhizas access to soil PO43− monoesters at low temperatures. Cold-active proteases were also found to be produced in strains of Hebeloma representative of different climatic zones grown in axenic culture at either 2°C and 22°C or 6°C and 22°C (Tibbett et al., 1999). Culture filtrates were assayed between 0 and 37°C for proteolytic activity, with growth at low temperature inducing greater activity. Many of the strains produced protease(s), which retained significant activity at temperatures as low as 0°C, and had a thermal optimum between 0 and 6°C with a second optimum at a higher temperature. The results suggest the potential exists for continued nutrient acquisition by ectomycorrhizal fungi at low temperatures, since while ectomycorrhizal fungi remain viable below 0°C, their growth must be severely limited by the subzero temperatures and would be physically constrained in the frozen bulk soil.

From this account it is apparent that enzymes active at moderately low temperatures are produced by species of decomposer and mycorrhizal fungi from cold Arctic and Antarctic environments.

Do the properties of psychrophilic enzymes explain physiological psychrophily?

Cairns et al. (1995b) stated that, while there are examples of enzymes which exhibit impaired function at low temperatures, it appears that activity at 0–15°C at correspondingly lowered rates is common to many enzymes from all sources, psychrophilic and otherwise. The properties of these enzymes could maintain the metabolic fluxes necessary for reduced growth rates at low temperatures and explain this feature of psychrophily. Conversely, enzymes from psychrophiles show thresholds of thermal inactivation of 28°C and optimum temperatures for catalysis of 40–60°C. These temperatures do not match and cannot explain the low optimum temperature for psychrophilic growth of 9–12°C. Cairns et al. (1995b) also wrote that the invertase from the snow-mould M. nivalis showed no specifically cold-active properties and resembled invertases from mesophiles in all respects examined. Taken together with the growth data, a picture emerged of this species as a generally metabolically mesophilic organism. From the enzymological perspective, according to Cairns et al. (1995b), this also appears to be generally true of enzymes isolated from psychrophiles so far. For example, although 30% of the maximum activity of a polygalacturonase of the psychrophilic snow mould fungus Sclerotiniaborealis was observed at 5°C, the optimum temperature for activity was 40–50°C (Takasawa et al., 1997). The same mesophilic phenomenon is exhibited by a purified chitinase of Verticillium cfr. lecanii A3, isolated from continental Antarctica. The enzyme was active over a broad range of temperatures (5–60°C), and although at 5°C its activity was still 50% of that recorded at the optimal temperature, this was relatively high at 40°C (Fenice et al., 1998).

Against this metabolically mesophilic background, specific psychrophilic character (low optimum temperature for growth) must be conferred by some, potentially as few as one, temperature-sensitive limiting factors which inhibit growth above the low optimum. The factors conferring low optimum temperature for growth remain to be identified, but could, for example, be the loss of vital properties of intracellular proteins just above the upper limit of fungal growth (Cairns et al., 1995a; Hoshino et al., 1997). The available evidence does not implicate any enzyme as the sole determinant of obligate psychrophily, or psychrotrophy.

The above discussions of trehalose, polyols, lipids and proteins show that there is not one single mechanism which leads to tolerance of low temperatures by fungi (Weinstein et al., 2000). All the cell components of a psychrophile or psychrotroph must be functional for the fungus to grow at low temperatures, and cold adaptation must be an overall cellular phenomenon (Russell, 1990). By contrast, the upper growth temperature limit of psychrophiles and psychrotrophs can result from the lack of activity of a single enzyme.

It would be expected that, if the changes in cell constituents outlined above are adaptations to cold temperatures, there would be a seasonal pattern in these cell constituents. There are few studies at this fine scale, with the notable exception of Montiel (2000) who followed the concentrations of low molecular weight carbohydrates in winter, spring and summer from one alga, three mosses and three lichens. Seasonal changes in total soluble carbohydrate concentration were observed in all species. In most cryptogams exhibiting seasonal change, concentrations of soluble carbohydrates were higher in spring relative to winter. Montiel (2000) hypothesized that the relatively high concentrations of carbohydrates function as osmolytes in winter and physiological buffering agents in spring. No information was provided for free-living fungi.

Ecological mechanisms of cold avoidance and cold tolerance

Fungal survival in Arctic and Antarctic environments may occur because of cold avoidance, rather than cold tolerance. One method of cold avoidance would be for fungi to re-establish annually in spring or summer from spores in cold Arctic and maritime Antarctic areas, once the coldest period of the year has passed. This strategy may be of little benefit in continental Antarctica as positive air temperatures are experienced only rarely and locally, and at inland sites the mean daily temperature rarely rises above −10°C (Convey, 1996), although soil temperatures may be considerably higher than those recorded in air for brief periods.

Re-establishment every year from spores – cold avoidance rather than cold tolerance?

Bergero et al. (1999) stated that the major component of an Arctic soil mycoflora should be active in one or more short growing seasons interspersed with periods of prolonged dormancy. They thought only a minor component, such as the small group of psychroligotrophic fungi found in their study, may be expected to show continuous slow growth. Working with Antarctic fungi, Vishniac (1996) wrote ‘It is generally thought that unprotected hyphomycete hyphae do not survive freezing, while spores often do’. However, there is little work on whether fungi in Arctic and Antarctic soils and litter do re-establish annually from spores. It may be that this is a strategy used by microfungi, rather than basidiomycetes, to avoid periods of extreme winter cold. Also, there are few data on the longevity of basidiomycete mycelium over winter or several seasons, and there is little information on annual recruitment from basidiospores. Referring to basidiomycete mycorrhizas ‘Arctic and alpine environments probably select for traits such as longevity and mycelial spread of individual fungal genets’ (Gardes & Dahlberg, 1996), although significant recruitment from spores is suggested by the work of Matsumoto & Tronsmo (1995) concerning a basidiomycete snow mould, Typhula ishikariensis from Norway. The population structures of this fungus at 23 sites in meadows and pastures were determined, based on vegetative compatibility group. Except for two sites dominated by a single large vegetative incompatibility group, populations were generally diverse, regardless of the cropping history of the site.

Would these spores be able to survive low temperatures during winter to be able to act as a ‘reservoir’ to germinate during spring? In temperate agricultural soils AM fungi are thought to survive adverse environmental conditions as spores (Addy et al., 1997), although these authors showed that hyphae of Glomus spp. retained their infectivity following prolonged freezing and that spores were not an effective inoculum in bioassays. Gams & Stalpers (1994) stated,

From cryobiological experience … we know that temperatures between 0 and −40°C are not suited for long-term preservation of fungi. In a moist environment the surrounding water will freeze between 0 and −10°C. When a thin layer of water with a high osmotic value remains present, the spore will dehydrate. In this condition it can survive for several years, but at temperatures above −40°C there is still some metabolic activity and damage will occur to proteins and membranes. When spores freeze in dry conditions and no dehydration occurs, ice crystals will form inside the spores at temperatures between −20 and −40°C. These crystals will increase in size and the resulting damage is usually lethal … It is only at temperatures of liquid nitrogen or at least −130°C that a permanent preservation of living fungal material is now possible, and only after a carefully determined freezing protocol. While under dry conditions spores of  Chaetomium and some other ascomycetes, and also conidia of genera like Aspergillus and Penicillium are known to survive for some decades’.

Seeming to contradict Gams & Stalpers’ (1994) above statement that it is unlikely that frozen spores would survive subzero temperatures in the long term, viable fungi were recovered from all sections of Greenland ice-cores studied by Ma et al. (2000). Most isolates were recovered from 300- and 500-yr-old sections, and only a few from 600 to 140 000 yr before present. Thus, survival of viable spores is uncertain at the subzero temperatures experienced in cold Arctic and Antarctic environments.

Are these spores dispersed from elsewhere?

A further cold avoidance mechanism used by fungi in Arctic and Antarctic environments could be re-colonization in spring from spores, or hyphal fragments, dispersed from warmer climates elsewhere. According to Vishniac (1996), the problem of determining which fungi are indigenous is particularly troublesome with regard to the hyphomycetes reported from continental Antarctica, as these fungal spores are readily airborne. Are there enough spores transported in spring and summer from continental America each year to Antarctica, or from more temperate climates to the Arctic, to replace dead ‘frozen’ mycelia or spores? Marshall (1997) found spores of several fungi not normally native to Antarctica in air samples collected at Signy Island in the maritime Antarctic. Their presence was associated with a specific weather pattern, occurring with an estimated mean annual frequency of 1.5, which allowed wind-borne transfer of exotic biological particles to Antarctica from South America. Evidence from the occurrence of fungal propagules in the air suggested that most fungi that produce spores on Signy Island disperse their spores during the summer. Chlamydospores were exceptional in being dispersed in winter. These spores are much more resilient to stress than most, having thick pigmented walls and are therefore better protected to survive extreme cold of winter and the increased levels of UV radiation associated with ozone depletion. Marshall (1998) continued his research on the origin of fungal propagules at Signy Island by swabbing skuas to obtain keratinophilic fungi when they returned early in summer. Geomyces pannnorum was recovered in culture suggesting that these birds may act as vectors for the transport of microorganisms between Antarctica and more northern landmasses.

These results show that fungal spores are dispersed from elsewhere to Antarctica, and a similar pattern of transport of exotic airspora has been found at sites in the Arctic, although total concentrations of propagules were low in both the maritime Antarctic (Marshall, 1997) and the Arctic. Sampling of airborne pollen and spores at Ny-Ålesund on Svalbard in the summer of 1986 revealed only very low concentrations of air spora (Johansen & Hafsten, 1988). The maximum diurnal concentration of Cladosporium was merely 44 spores m−3 air, although of the fungal spores less than 2% belonged to Cladosporium spp. (the rest were unidentified). The maximum concentration measured in Tromsø, northern Norway, that season was about 80 and in Bodø about 600 spores m−3. The tendency towards increased concentration of Cladosporium spores during episodes of exotic pollen recording indicated that this spore type is subject to long-distance dispersal, perhaps from central Finland. Similarly, a survey of airspora collected on Jan Mayen between 24 April and 31 August 1988, revealed only very small concentrations (Johansen, 1991). The highest diurnal average of Cladosporium was 27 spores m−3 air. A cumulative diurnal mean of c. 2600 fungal spores m−3 air was recorded, constituting only one third of the total amount obtained on Svalbard in 1986. Cladosporium constituted 4.7% of the total fungus record. Alternaria made up less than 0.1% of the airborne fungal spores recorded during the sampling period, basidiospores 11.8% and unidentified fungal spores as much as 83.5%. The majority of these spores were small (2–4 µm) hyaline, unicellular globose spores, and hyaline fusiform spores. Detached hyphal fragments occurred, hyaline septate ones being more numerous than dematiaceous ones and conidiophores were occasionally observed. The levels of hyphal fragments on Jan Mayen in 1988 were small compared with, for instance diurnal average concentrations of 10–599 fragments m−3 air above grass in England (Pady & Gregory, 1963, cited in Johansen, 1991). Betula pollen indicated transport from Iceland and/or North America and Greenland.

In summary, long and short distance transport of fungal propagules is possible although the numbers of propagules are low. Thus, some fungi may survive in Arctic and Antarctic environments by avoiding the extreme low temperatures during winter through annual germination from airspora during spring and summer. There is evidence, however, from air samples of marked seasonality in spore dispersal. At the Svalbard site, spores of Cladosporium and other unspecified fungal taxa were recorded occasionally or very sparsely from the start of recording in late April and onwards, but did not become more frequent until the middle or later part of June. The seasonal maximum of Cladosporium spores was registered on 1 August (44 spores m−3), whereas the bulk of ‘total’ fungal spores occurred after 10 August when concentrations up to 600 spores m−3 were recorded. A similar pattern was observed in Jan Mayen (Johansen, 1991), with the seasonal peak in fungal spores being from the middle of July to a maximum on 27 August of 639 spores m−3 air. While sampling ceased on 31 August, fungal spores were still observed and Johansen (1991) believed that future investigations should include September to cover the whole season before the onset of winter. However, the importance of spring and summer germination of airspora as a cold-avoidance mechanism remains unclear since, in most areas of the world (except Signy Island where peak dispersal was in early summer), peak concentrations of Cladosporium spores and other epiphytic saprotrophic fungi tend to occur at the end of the growing season, when the maximum surface area of suitable host material is available for colonization. This is true even at high latitudes in the Northern hemisphere where the more diverse flora and larger numbers of vascular plants produce host material in larger volumes than the cryptogam-dominated Antarctic flora (Marshall, 1997). Thus, the ‘foreign’ propagules dispersed in autumn would have to survive the cold temperatures in winter, as would the ‘native’ propagules. More research is necessary on whether annual seasonal germination occurs, if it is a cold-avoidance mechanism, if this occurs in both microfungi and basidiomycetes, or if Arctic and Antarctic environments select for traits such as longevity and mycelial spread of individual fungal genets of basidiomycetes.

Effect of low temperature on spore production

Despite the now relatively large number of reports of extension rate of cold tolerant fungi at low temperatures (Table 1), there appear to be no data on spore production by psychrotrophic and psychrophilic fungi in response to a range of temperatures. Are more spores produced by cold-adapted fungi at low temperatures than temperate isolates of the same species? Such studies would also be useful in understanding the role of sterile mycelia in cold-tolerance (discussed below). In an experiment using a temperate isolate of Penicillium hirsutum grown on potato dextrose agar and kept at 20, 10, 4, 2, 0, −2 and −4°C, sporulation and germination were retarded at low temperatures, and at −4°C no germination occurred (Bertolini & Tian, 1996). It is necessary to repeat this type of test for psychrotrophic and psychrophilic fungi to see whether increased spore production occurs at low temperatures.

Mycelial acclimation to low temperatures in the field

From research in fungal cryopreservation, at fast rates of cooling hyphae do not have time to lose water and freeze internally. This is normally a lethal event (Smith, 1993). The conventionally employed cooling rate for the cryopreservation of fungi is 1°C min−1, although this is species dependent (Morris et al., 1988; Corbery & LeTacon, 1997). Another example of fungal acclimation to low temperatures in the laboratory has been provided by Robinson & Morris (1984). Hyphae of Fusarium oxysporum f. sp. lycopersici transferred from 25 to 7°C and maintained at this temperature for 2 h were more tolerant to subsequent cooling to −2°C than hyphae that had not been pretreated in this manner. Does this acclimation occur in Arctic and Antarctic fungi in the field and does it increase survival? There appear to be no directly relevant Arctic or Antarctic studies from the natural environment, but some evidence is provided from an experiment using blocks of temperate field soil containing AM (arbuscular mycorrhizal) fungi which were either slowly cooled (2°C d−1 to 5°C) or held at room temperature (20°C) before freezing at −12°C (Addy et al., 1998). Infectivity of AM fungi was greater in soil that was slowly cooled before freezing, hypothesized to result from cold acclimation of extra-radical hyphae. The hypothesis was tested using in vitro mycorrhizas, cultured in Petri plates with two compartments in which hyphae grew into a separate section. Metabolic activity of these hyphae following freezing at −5°C was assessed using a vital stain. The majority of cultures that were slowly cooled as above before freezing contained active hyphae, whereas hyphal activity was almost completely eliminated by freezing in cultures that were not precooled. The authors stated that, to their knowledge, this was the first report of acclimation to cold temperatures by AM fungi. The specific mechanisms conferring freezing tolerance on AM fungi remain to be determined, and it is unclear whether these temperate isolates would respond to freezing as Arctic and Antarctic isolates would, or whether acclimation occurs in natural high latitude environments.

Seasonal patterns in mycelial biomass of psychrophiles and psychrotrophs

‘The biomass of psychrophiles may vary seasonally from low values in spring to higher values in autumn’ (Flanagan & Scarborough, 1974). At one US tundra site, 67% of the fungal strains that grew from propagules isolated from soil in the autumn were psychrophilic (Flanagan & Scarborough, 1974). However, there are few studies concerning seasonal values of psychrophilic biomass from the field in cold environments, or of a series of seasonal isolations and response to a range of temperatures of the fungi obtained. As Smith & Read (1997) have written, more studies through the year in the natural environment of the fungi are much needed. They provide not only ecologically relevant information, but also valuable pointers for physiological investigations of hyphae and spores.

Other mechanisms of cold adaptation

Sterile mycelia and dark septate hyphae

Aspects of fungal life-history and morphology in Arctic and Antarctic fungi may be adaptations to cold tolerance. Examples of abbreviated life cycles are found in fungi existing in harsh environments, shown by the short-cycled rusts in the Canadian Arctic (Savile, 1953) and the predominance of sterile fungi in tundra soils (Widden & Parkinson, 1979). It is, however, not known whether these are responses to low temperature per se.

Fungi with dark septate hyphae dominate the soil microbial community in Antarctic, Arctic and alpine soils (Smith & Read, 1997). Melanins may protect dark septate hyphae from extreme temperatures and drought, and so broaden the ecological niche of these fungi (Jumpponen & Trappe, 1998). These authors stated that such resistance to cold and desiccation may play a significant role for persistence of hyphae from year to year.


The majority of fungi isolated from Arctic and Antarctic soils and litter are psychrotrophic. The physiological mechanisms conferring cold tolerance are complex and there is not one single adaptation as all the cell components of a psychrophile or psychrotroph must be functional for the fungus to grow at low temperatures. By contrast, the upper growth temperature limit of cold-tolerant fungi can result from the lack of activity of a single enzyme. With respect to ‘cold active’ enzymes, while there are examples of enzymes that exhibit impaired function at low temperatures, it appears that activity at 0–15°C at correspondingly lowered rates is common to many enzymes from all sources, psychrophilic or otherwise. Conversely, psychrophilic fungi (defined as such by their low optimum temperature for growth) may produce mesophilic enzymes. From an ecological viewpoint, more work is necessary on whether fungi germinate from spores annually in spring and summer as a strategy to avoid the coldest temperatures in Arctic and Antarctic environments, and the significance of long and short distance spore dispersal to this strategy is unclear. The longevity of basidiomycete mycelia in Arctic and Antarctic environments through periods of extreme cold is unknown, as is whether mycelial acclimation to low temperatures occurs in the field. More information is necessary about the response of sporulation in cold-tolerant fungi at low temperatures, and whether sterile mycelia and dark septate hyphae, both predominant in Arctic and Antarctic ecosystems, are adaptations to low temperatures.


Financial support from The Coalbourn Trust of the British Ecological Society for the study of the ecology of Arctic saprotrophic fungi, and from the Global Atmospheric Nitrogen Enrichment Thematic Programme of the Natural Environment Research Council, is gratefully acknowledged. The kind comments of Peter D. Moore and Ron I. Lewis Smith greatly improved earlier drafts of this manuscript.