In this study, we provide the first description of the bacterial communities of cryoconite holes from South American glaciers, in particular from both small high-elevation glaciers of the Central Andes in the Santiago Metropolitan Region (Chile), and from the tongues of two large glaciers in Patagonian Andes that reach low altitudes. These pieces of information fill a large geographical gap in our knowledge of glacier environments because this is the first description of the microbial communities of supraglacial environments in South America, a continent with about 30,000 km2 covered by ice29. Results showed that the large Patagonian glaciers (Exploradores and Perito Moreno) had the highest oxygen concentrations, while Iver and East Iver had the lowest ones and Morado an intermediate value. This pattern could be related to the different altitudes of the glaciers. Indeed, since water temperature in cryoconite holes is always quite low and stable at all altitudes, oxygen solubility in these environments is related to the atmospheric partial pressure of oxygen that decreases at increasing altitude30. This result is consistent with [O2] values we found in our samples. Indeed, Exploradores and Perito Moreno are located in Patagonia at low altitudes (< 200 m a.s.l.), while Iver and East Iver are the highest ones among those we investigated (samples were collected at about 4000 m a.s.l.) and Morado is at an intermediate value (3400 m a.s.l.).
pH values seem to follow a different pattern: the highest values were recorded on Exploradores and Morado, while the lowest ones were on East Iver and Iver. However, all the mean pH values of the five glaciers were basic (between 8.57 and 10.47) and differences in pH among glaciers are not easily explained also considering the lithology of the surrounding environments. Indeed, Perito Moreno lies between rhyolitic and undifferentiated volcanic bedrock, Morado, Iver and East Iver are located between alkaline and andesitic igneous rocks, and Exploradores is characterized by a granitic bedrock31. We note, however, that we measured water pH rather than that of the cryoconite because the first one was already reported to influence bacterial communities in cryoconite holes4.
The most abundant orders were (in decreasing order): Betaproteobacteriales, Cytophagales, Chitinophagales, Acetobacterales, Frankiales, Armatimonadales, Sphingobacteriales, Rhizobiales, Bacteroidales, Sphingomonadales, and Micrococcales. These orders belong to the phyla Proteobacteria, Actinobacteria, Bacteroidetes, and Armatimonadota. Most of these orders are typical of cryoconite holes and dominate bacterial communities in these environments in all the geographical areas investigated so far: Arctic32, Antarctica5, Europe10,12, and Asia33. Betaproteobacteriales are a quite heterogeneous order where diverse members showed resistance to stressful conditions including oxidative stress and also degradation of aromatic compounds34. Interestingly, 48.12% of Betaproteobacteriales sequences in our samples belong to the genus Polaromonas, a bacterium well adapted to glacier environments35,36. According to our results, Polaromonas was quite abundant on all the glaciers but on Perito Moreno. This is an important genus because it is well adapted to harsh environments because of its ability to survive long-range atmospheric transport and thanks to its versatile metabolism (can use both organic and inorganic electron donors)35. Cytophagales are Gram-negative bacteria that are known to be less resistant to UV radiation than Gram-positive bacteria37. In addition, 91.68% of Cytophagales sequences belong to Hymenobacter, a heterotrophic genus composed of mostly heterotrophic, aerobic, Gram-negative bacteria which have been described in extreme environments like Antarctica, where a strain with extremely high resistance to solar radiation was isolated38,39. Armatimonadota and Frankiales were never reported as abundant inhabitants of cryoconite holes in other geographical areas. Interestingly, Armatimonadota were present with a high relative abundance on Perito Moreno only. This phylum includes both aquatic and terrestrial genera, surprisingly described in hot springs and geothermal soils40. Maybe, the presence of pigments, oligotrophy and the ability to degrade polysaccharides provide them with the ability to survive in such harsh environments40. Frankiales, together with Micrococcales, belong to the phylum of Actinobacteria and are Gram positive bacteria41. They were abundant on all glaciers, except for Perito Moreno and their high relative abundance may be related to their resistance to UV radiation42. cyanobacteria, which are considered important components of cryoconite hole bacterial communities and even ecosystem engineers in these environments9, were only 0.5% in our samples and no order was included among the most abundant ones. In summary, cryoconite holes of Andean glaciers seem to host bacterial communities dominated by the taxa typical of these environments worldwide, but with a low abundance of cyanobacteria, and a rather high abundance of Frankiales in all glaciers except for Perito Moreno, where a high abundance of Armatimonadota, another unusual taxon in cryoconite holes, was observed.
Despite this general similarity of the dominant orders of bacterial communities at a global scale, a closer inspection at the ASV level revealed that the bacterial communities of the different glaciers differed from one another. In addition, the RDA biplot showed that the three small and high-elevation glaciers clustered quite close to one another, whereas Perito Moreno and Exploradores were separate from one another and the other glaciers. This pattern may be due to geographic distance, different altitudes, or other ecological and environmental differences among glaciers. Indeed, bacterial communities in high-elevation glaciers are exposed to higher UV radiation and, therefore, to high oxidative stress3 as well as lower partial oxygen pressure and lower nutrient availability from aeolian deposition. Patagonian glaciers are 440 km from one another and c.ca 2000 km from the other glaciers we sampled, which, in turn, are less than 60 km from one another. In addition, it has been demonstrated that the main source of cryoconite bacterial communities is the local environment surrounding glacier2. The high-elevation glaciers we sampled are small and above the tree line, and generally in areas with sparse vegetation. In contrast, the Patagonian glaciers are large and their tongues (where we collected the samples) are well below the tree line and therefore exposed to very different potential sources of microorganisms with respect to the other glaciers.
The RDA also showed that bacterial communities of cryoconite holes varied significantly with ΔpH and Δ[O2]. We stress that these variables represent the difference between the pH and oxygen concentration of each cryoconite hole from the respective mean value of all the holes of that glacier. They therefore do not account per se for the difference in mean pH and oxygen concentration among glaciers highlighted in the analyses discussed above, because in this analysis this effect was accounted for by the glacier factor, which accounts for any difference between glaciers. Thus, this analysis suggests that bacterial communities vary consistently according to pH and oxygen concentration gradients present on each glacier, even if different glaciers show on average different pH and oxygen concentration values. In other words, for example, an increase of two pH units seems related to a similar variation in bacterial communities independently from the absolute pH value, at least in the range of variability of pH values recorded in this study. In addition, no effect of ΔpH on alpha diversity resulted from GLSs. pH is known to affect bacterial communities of cryoconite holes4, but little is known about the mechanisms underlying these relationships. Indeed, water pH may also be partially affected by the metabolic activities of the bacterial communities, so it is still unclear whether water pH affects cryoconite hole bacterial communities or, at least, partially, also the reverse occurs. Indeed, it has been reported that the photosynthetic activity of both bacteria and algae can alter CO2 concentration which, in turn, can alter pH43. In addition, in cryoconite holes, this activity can also be balanced by that of the heterotrophic bacterial community and other heterotrophic taxa (e.g. tardigrades, rotifers, fungi)44. Furthermore, analyses of the most abundant orders within each glacier revealed that some varied according to pH. Different studies already investigated the effect of both water and sediment pH on bacterial community composition, proving that they influence the communities of different types of river sediments45. Nonetheless, so far only the decrease of Acidobacteria at basic pH values was already reported in literature data45 but was not detected in our study, probably because the pH values of our samples varied from 7.25 to 12.71 i.e. in a range where the abundance of Acidobateria is always low46.
Cryoconite holes are aerobic environments in the water, thanks to the high O2 solubility in cold environments, the release of air bubbles from the ice that melts because of the presence of the dark sediment, and to photosynthesis47. On these glaciers, we observed on average lower oxygen concentrations in the high-altitude cryoconite holes probably because water oxygen depends on the equilibrium with the atmospheric oxygen. In addition, variation in oxygen concentration in the water seems to play a role in explaining the structure of bacterial community structures, albeit none of the most abundant orders seemed to vary significantly according to [O2] concentrations within a glacier. Interestingly, no differences in cyanobacteria abundance were observed, indeed it would be expected that their abundance was higher when also [O2] was higher13, because of their photoautotrophic activity, but no evidence emerged about it. In any case, the mechanisms that link oxygen concentration in the water to bacterial community structure in the sediment are not easy to explain. Indeed, [O2] in the sediment can largely differ from that in the water, with the rapid formation of anoxic layers under a few millimetres of cryoconite48,49.
The Iver and East Iver glaciers are the two geographically closest glaciers (only 1.2 km apart), and they show also similar pH and [O2] values (Fig. 2a, b) and similar alpha diversity values. In addition, their bacterial communities, although significantly different, were close to one another in the RDA biplot. Interestingly, also the Morado glacier clustered close to them in the RDA biplot, even though it is about 60 km from them. In contrast, the two Patagonian glaciers are separated from one another and the other glaciers on the same plot. These results, on the one side, support the hypothesis that altitude and geographic position play an important role in defining bacterial community composition. On the other side, however, the three small glaciers in the central Andes are also at a similar elevation and they are surrounded by very similar environments (R. Ambrosini and F. Pittino, personal observation). These three high mountain glaciers have also an important anthropic input of aerosol and black carbon that act as sources of cryoconite. Indeed, a high percentage of black carbon deposition on ice in the Central Andes close to Santiago has been associated with emissions from the city (> 40%), whereas mining is also an additional important black carbon source50. Their similarity can therefore derive also from being exposed to the same general ecological conditions, including high UV radiation, oxidative stress, anthropic pressures, and probably, also from similar sources of bacteria. These results therefore highlight that correlative studies like the present ones can hardly disentangle the effects of geographical positions and ecological conditions on the structure of cryoconite hole bacterial communities, and further studies should be designed to add insight into this still open question.
Analyses of alpha diversity indices indicated that cryoconite holes on Exploradores glacier showed the highest richness and evenness. Samples on the Exploradores were collected close to the glacier terminus, surrounded by a rich evergreen broadleaf vegetation, and in an area with abundant supraglacial debris and frequented by tourists. The higher biodiversity of this large, low-altitude glacier, compared to that of the small, high-altitude Iver and East Iver glaciers is not surprising, as the rich evergreen broadleaf forest that surrounds the tongue of the first glacier can be the source of a richer and more diverse bacterial community than the bare ground surrounding the other ones. However, it is more surprising that the alpha biodiversity of the large, low-altitude Perito Moreno was intermediate and similar to that of the Morado glacier. Interestingly, Perito Moreno was the southernmost glacier among those we collected, and was surrounded by a less diverse forest, dominated by southern beeches, Nothofagus ssp. than that of Exploradores, while Morado was the glacier where samples were collected at the lowest altitude among the three glaciers near Santiago. We may therefore speculate that a broad gradient related to altitude and general climate conditions of the area surrounding the glacier may somehow affect its biodiversity. For instance, among the most abundant orders, Cytophagales were more abundant on high than on low-elevation glaciers (Fig. 5b). A similar pattern was observed for the Micrococcales and Chitinophagales (Fig. 5c–k) with the only exception of Iver.
In summary, we provide the first-ever description of the bacterial communities of cryoconite holes of glaciers in South America, specifically in the Southern Andes. This study thus fills an important gap of knowledge as almost no information was previously available on the cryoconite holes of this continent, and opens the possibility of future biogeography analyses including samples from almost every important glacial area of the world. The five glaciers we investigated are still a too small sample for thoroughly assessing the ecological processes that control cryoconite hole bacterial communities, and a larger set of environmental variables should also be considered, but we hope this study can be the basis for further investigations aiming at a deeper understanding of these extreme environments.
Source: Ecology - nature.com
