Surface ocean microbial communities differ systematically in their abilities to enzymatically hydrolyze high molecular weight organic matter, and thus to initiate remineralization of high molecular weight substrates (Arnosti et al., 2011). Pelagic microbial communities at high latitude are capable of hydrolyzing a narrower spectrum of soluble substrates than their temperate counterparts, although the underlying reasons for this substrate selectivity are unknown. This narrowing of microbial community function at high latitude parallels decreases in microbial diversity and/or species richness at high latitude (Baldwin et al., 2005; Pommier et al., 2007; Fuhrman et al., 2008; Wietz et al., 2010). Nevertheless, Arctic pelagic heterotrophic microbial communities are active players in the marine carbon cycle, transforming, and respiring a wide range of substrates, despite permanently cold temperatures (Wheeler et al., 1996; Kirchman et al., 2009a). Substrate consumption in Arctic microbial communities is affected by a wide range of factors, including season, location, and nutrient levels (Kirchman et al., 2009b; Nikrad et al., 2012). The functional difference in enzyme activities observed over latitudinal gradients, and the narrow spectrum of substrates hydrolyzed by Arctic microbial communities, may thus be driven by a lack of a nutrient or co-factor required to induce production of specific enzymes by organisms possessing them, or conversely by a lack of organisms with the genes to synthesize the enzymes required to hydrolyze specific substrates.

The supply of labile DOM has been suggested to play a key role in controlling bacterial growth at high latitude (e.g., Kirchman et al., 2009b). Substrate availability in ocean waters is also highly variable: transient events, such as phytoplankton blooms or physical mixing can rapidly change the set of organic substrates present, requiring changes in expression of genes related to organic matter metabolism (Teeling et al., 2012). Here, we investigated whether an expansion in enzymatic capabilities of pelagic microbial communities in waters of the Svalbard archipelago could be achieved through addition of a factor potentially lacking in the water column: mineral nutrients (nitrate and/or phosphate), labile organic carbon or nitrogen, or a specific inducer that might be expected to activate enzyme production. As a test case, we focused on trying to enhance the activity of enzymes that hydrolyze pullulan, a soluble, glucose-containing polysaccharide. Pullulan [α(1,6)-linked maltotriose; maltotriose is α(1,4)-linked glucose], was selected as a test case because in repeated investigations over the past decade, we have not detected pullulanase activity in the water column of the fjords of Svalbard, although it is readily hydrolyzed at a number of sites in temperate waters. Its basic constituent, glucose, is also is found in laminarin, a substrate readily hydrolyzed in Svalbard waters (Arnosti et al., 2005, 2011; Arnosti, 2008; Teske et al., 2011). The lack of hydrolysis of pullulan in the water column at Svalbard is particularly intriguing because pullulanase activity in underlying sediments is high (Arnosti, 2008; Teske et al., 2011), and bacterial isolates have been obtained from Svalbard sea-ice using pullulan as a growth substrate (Groudieva et al., 2004).

Heterotrophs capable of hydrolyzing polysaccharides were active in the pullulan induction experiments, as shown by maximum xylan hydrolysis rates (measured in a single replicate after 21 days incubation) of 4.3 nM L^-1 h^-1 in the 24 h pre-incubation no-addition samples and 6.8 nM L^-1 h^-1 in the 5-day pre-incubation no-addition samples (Table 1). Moreover, four distinct polysaccharide substrates were hydrolyzed in surface and bottom waters from Station J in Smeerenburgfjord (Figure 1), consistent with previous results that pelagic microbial communities in Svalbard are readily capable of accessing specific polysaccharides (Arnosti, 2008; Teske et al., 2011).

Efforts to stimulate use of pullulan as a substrate, however, showed that pullulan is microbially-inaccessible for most of the pelagic community (Table 1). Only a single set of treatments—the 5 days pre-incubation with maltotriose—led to hydrolysis of fl-pullulan during the subsequent incubation. Even in this case, substantial incubation time (21 days) was required before detectable hydrolysis of pullulan occurred. Addition of glucose, pullulan, nitrate, phosphate, nitrate, and phosphate, and free amino acids did not lead to measurable pullulanase activity, nor did addition of maltotriose on timescales shorter than the 5 days pre-incubation + 21 days incubation (Table 1).

The observation that addition of maltotriose led to measurable pullulanase activity after sufficient pre-incubation is consistent with biochemical control of pullulanase expression, since pullulan is composed of repeating units of maltotriose. Many extracellular polysaccharide hydrolases are regulated by a scheme in which the hydrolase is expressed constitutively at a low level (Béguin and Aubert, 1994). Oligomers of the polysaccharide induce further expression of the hydrolase, while monomers of the polysaccharide (glucose, in this example) repress expression of the homologous hydrolase. This regulation mechanism has been demonstrated for thermophilic bacteria that hydrolyze pullulan (Antranikian, 1990); the current data suggest that pullulanase activity in Svalbard waters may be regulated in a similar manner. The fact that hydrolysis was only detected after 21 days in the +maltotriose, 5-days pre-incubation treatment from 2005 (and was not detectable with shorter pre-incubation periods with either maltotriose or pullulan) further suggests that the microorganisms in fjord waters capable of hydrolyzing pullulan were likely slow-growing as well as initially rare.

Induction of enzymatic pathways can occur on timescales of hours or faster in the lab (Madigan et al., 2000); and less than 2 days for hydrolysis patterns consistent with pullulanase induction in seawater at lower latitudes (Steen et al., 2012). Substrates for which the corresponding extracellular enzyme must be induced may therefore be hydrolyzed quite rapidly, if the final activity of the induced enzyme is high. The combined requirements of growth of a relatively rare member of the microbial community and induction of enzymes, however, could lead to the pattern observed in this study. Slow growth of bacteria in Svalbard waters is consistent with the results of Vadstein (2011), who found no increase in cell numbers over 22 h bioassay experiments in which carbon and nutrients were added to Svalbard fjord water. It is also consistent with results of very long-term incubations of surface water from Station J, in which hydrolysis of Isochrysis extract was first observed only after 70 days’ incubation, activity also attributable to initially rare and slow-growing bacteria (Arnosti, 2008).

In many respects, the general absence of pullulanase activity in Svalbard waters is quite surprising, given that it is a soluble, linear polysaccharide composed of glucose. Furthermore, pullulanase functions as a debranching enzyme of starch (Ball and Morell, 2003), a common algal energy storage polysaccharide (Painter, 1983). Hydrolysis of pullulan in Svalbard sediments is rapid (Arnosti, 2008; Teske et al., 2011), and a diverse range of bacterial isolated from sea ice in Svalbard’s fjords can hydrolyze pullulan (Groudieva et al., 2004). In addition, Aureobasidium pullulans, a fungus which produces pullulan, has been isolated from Svalbard fjord water (Zalar et al., 2008), so there is very likely a source of pullulan to the water we studied. Furthermore, pelagic communities in Svalbard’s fjords can hydrolyze a number of other polysaccharide substrates, as demonstrated by the incubations carried out with water from Station J (Figure 1). The observation that pullulan and arabinogalactan were not hydrolyzed over the time course of incubation in the waters of Station J, however, demonstrates that specific soluble, carbohydrate-containing substrates are unavailable to pelagic microbial communities in these polar waters. The observation that these same substrates were also not hydrolyzed in incubations carried out in other years in this same fjord (Arnosti et al., 2011; Teske et al., 2011) demonstrates that the inability to hydrolyze specific substrates is a consistent feature of these communities.

Polar microbial communities are composed of different species than those at lower latitudes (Bano et al., 2004), and exhibit enzymatic activity patterns distinct from their temperate counterparts (Arnosti et al., 2011). In order to cope with the rhythms of primary productivity and plankton succession that are unique to high-latitude environments (Hop et al., 2002), Arctic microbial communities may specialize in terms of substrate sets or controlling mechanisms for key carbon cycling processes. The observation that some soluble substrates readily hydrolyzed in lower latitude environments are only microbially available under a very specific set of circumstances supports the view that most members of Arctic pelagic microbial communities focus on a more limited range of substrates than their temperate counterparts. These substrates are hydrolyzed at considerable rates (Figure 1; Arnosti et al., 2011), perhaps at the expense of the ability to access a wider range of substrates, including a fraction of freshly-produced phytoplankton-derived organic matter in Svalbard waters (Thingstad et al., 2008). This streamlining of metabolic function may help explain how polar communities are able to rapidly mineralize specific organic substrates (e.g., Rich et al., 1997), even at low temperatures.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.