The allochthonous communities from different pollution sources can affect the ambient microbial communities of water bodies. We recorded the variation on bac-terial diversity, taxonomic profile, and the abundance of PHRB on different sample groups, namely municipal influent, municipal effluent, industrial effluent, mine
66
runoff, surface water, and DWPP. Such differences may be due to variations in the ecological conditions of different sample groups. For example, surface water may have a high dissolved oxygen concentration due to continuous aeration from the atmosphere and possible photosynthetic production. GW may have stable ecologi-cal conditions, such as anoxic, dark, oligotrophic, and a constant temperature. The low alpha-diversity in DWPP samples may be due to a lack of organic carbon for bacterial growth. A negative relationship between bacterial diversity and energy input has been reported earlier (Gülay et al. 2016). In the case of industrial and mu-nicipal effluent, low diversity can be due to the toxic effect of possible biocides or other toxic compounds (Hubbe et al. 2016). In contrast to our conclusion, Korajkic et al. (2015) reported a higher Shannon index in sewage effluent than in river water.
The omnipresence of some taxonomic groups (phyla Proteobacteria and teroidetes) can be due to their ability to survive in wide metabolic conditions. Bac-teria within these phyla are among some of the dominant microbes found in natural ecosystems (Zeglin, 2015). The detection of Firmicutes and Fusobacteria phyla sole-ly in sewage samples (influents and effluents) can impsole-ly the monitoring of water quality; having these communities in surface water can indicate a possible faecal contamination. Further, these bacterial groups of faecal origin may be highly dilut-ed in surface water, and thus were not detectdilut-ed in our analysis. Phyla Actinobacte-ria was highly abundant in the surface water and pretreated samples of DWPP.
Phyla Acidobacteria was abundant in GW samples. An earlier study reported a distinct taxonomic variation in different sample groups among groundwater, sur-face water, treated drinking water, and household tap water (Zeglin, 2015). Earlier studies have also reported the high abundance of phyla Proteobacteria, Actinobac-teria and Bacteroidetes in surface water (Wang et al. 2018, Yu et al. 2014, Ibekwe et al. 2012). The taxonomy and diversity of the microbial communities in an aquatic ecosystems are dynamic due to changes in various environmental factors, such as temperature, light, UV radiation, pH, the available concentration of oxygen, nitro-gen, phosphorous and metal ions (Staley et al. 2014), predators and bacteriophages (Korajkic et al. 2015), and land-use patterns of the catchment area (Wang et al. 2018).
The seasonal effect on bacterial diversity, taxonomic profile, and PHRB abun-dance was noted only in surface water. The major seasonal dissimilarity factors in Finland are the ambient temperature, amount of solar radiation and snow cover.
The solar radiation can affect photosynthesis and UV inactivation. The lower alpha-diversity of surface water during the summer and seasonal variations in the taxo-nomic profile has been also reported earlier elsewhere (Yu et al. 2014, Wang et al.
2018, Liu et al. 2019). The predation of protozoa may control bacterial diversity in an aquatic ecosystem (Korajkic et al. 2015). The Finnish river ecosystem becomes pho-tosynthetically more active in the summer and may increase the eukaryotic preda-tor communities that graze the bacteria. Another explanation may be due to the inactivation effect of high-intensity solar radiation due to the long daylight hours.
The major bacterial community of municipal influent samples, Arcobacter spp.
(Epsilonproteobacteria) and Moraxellaceae family (Gammaproteobacteria) are a
po-67 tential pathogenic group (Huys G., 2014, Lastovica and Zhang 2014). Municipal sewage was rich in gut microbiota communities, such as bacterial groups from Fusobacteriia, Bacilli, Clostridia, and Bacteroides. The significant reduction of these communities during the sewage treatment process can be because these groups are from an anaerobic gut environment, but wastewater treatment with activated sludge is an aerobic process. These groups have also been reported earlier from sewage samples (Korajkic et al. 2015, Wang et al. 2018). However, the detection of these communities, also in mine effluent, indicates the inclusion of sanitary wastewater together with runoff water from the mining area.
Industrial effluents were rich in the bacterial groups having wide metabolic ca-pacities, such as the Comamonadaceae family (Betaproteobacteria). The members of this group can act as an aerobic organotroph, anaerobic denitrifier, iron reducer, hydrogen oxidizer, photoautotroph, photo-heterotroph or a fermenter (Willems, 2014). Further, industrial effluent has family Rhodocyclaceae (Betaproteobacteria), photo-heterotrophs, plant-associated nitrogen-fixing aerobes. These bacteria have a wide metabolic capacity and utilize a wide source of organic carbons as their ener-gy sources (Oren 2014). Industrial effluent also contained anaerobic members of the community, such as Weeksellaceae family (Flavobacteriia) and facultative anaerobes Aeromonadaceae (Gammaproteobacteria). The Weeksellaceae has been reported earlier in sewage and activated sludge (Pan et al. 2016, Ahmed et al. 2018). The Aeromona-daceae spp. can be a potential pathogenic, commensal or environmental origin (Huys 2014).
Surface water samples were rich in heterotrophic and mesophilic families Oxa-lobacteraceae and restricted facultative methylotrophs Methylophilaceae of Betaprote-obacteria. Oxalobacteraceae can have a wide range of phenotypic and ecological properties and is mostly aerobic or micro-aerobic, and some may be facultative anaerobic (Baldani et al. 2014). Methylophilaceae can utilize methanol or methylamine as a source of carbon and energy and was previously detected in activated sludge and other freshwater sources (Doronina et al. 2014). Further, a Gram-positive group ACKM1 family (Actinobacteria) detected in a surface water sample are het-erotrophic or symbiotic nitrogen-fixing with plants (Goodfellow et al. 2012).
DWPP samples had a high read proportion of Alphaproteobacteria, mainly the Rhodospirillaceae family. These bacteria are mainly anaerobic, chemoheterotrophic in dark conditions and heterotrophic under aerobic conditions (Baldani et al. 2014).
This group of bacteria is normally smaller than most other bacteria (Morris et al.
2002). Further, GW samples had Gram-positive and oligotrophic classes Acidobac-teria-5 and Acidobacteria-6. These groups are also smaller than most other bacteria and were previously reported in soil samples (Kielak et al. 2016). The poor classifi-cation of the Betaproteobacteria group in groundwater samples may imply that these groups are new and not reported previously or potentially unique in Finland.
The share of potential health-related bacterial (PHRB) reads was higher in ef-fluent and surface water samples than in the samples collected from DWPP. The
68
sewage treatment process significantly reduced the read proportion of PHRB. The lowest reads of PHRB in summer samples can be due to the solar inactivation effect of the 18 hours daylight of Nordic summer. Another explanation could be that the detection of PHRB bacteria was masked with high abundance of autochthonous bacteria in samples collected during that period. The high mean value of Mycobacte-rium and Legionella on environmental samples was not surprising, as these genera are independent of faecal contamination (Pond 2005).
The higher predicted function in industrial effluents can be indicating the bac-terial communities might need more active pathways for their existence than in other sample groups. The possible presence of toxic compounds in industrial efflu-ents may promote the production of secondary metabolites on this sample group.