Introduction

The planet Earth’s surface is unevenly distributed with environments and some of them are termed “Extreme Environments”, which may be detrimental to the biosphere. Extreme environments once thought to be lethal for life existence, be brimming with organisms (Gilichinsky et al. 2007). Extremes exist in physical (temperature, pressure, or radiation) and geochemical (salinity, desiccation, pH, redox potential, and oxygen species) forms, and occupants include Archaea, Bacteria, and some Eukaryotes (Rothschild and Mancinelli 2001). MacElroy (1974) coined the word “Extremophile” for organisms that thrive in a such hostile environment (MacElroy 1974). Extremophiles have adapted themselves to get through and retain their existence in extreme environments through evolution. Based on such environmental conditions for bacteria to thrive in extreme environments, extremophiles are described as psychrophilic (0 to 20 °C), thermophilic (41 to 122 °C), hyperthermophilic (80 °C to > 122 °C), acidophilic (pH 1–5), alkaliphilic (≥ pH 9), piezophilic (elevated hydrostatic pressure, up to 50 MPa), halophilic (salt concentration up to 5 M), oligotrophic (grow in trace nutrients environment), endolithic (nourish within mineral pores and rock) and xerophilic (low availability of free water) (Gupta et al. 2014; Kamekura 1998; Kato et al. 1998).

Considering cold extreme environments on planet earth, approximately 90% of oceans and 26% of terrestrial ecosystems are exposed to low temperatures (Zhang et al. 2009), which stretch across the Arctic to Antarctic plates, as well as from high mountain regions to cavernous of oceans. An utmost portion of the deep sea (90% oceans) have < 5 °C, snow covers 35% of the land surface, permafrost comprises 24% of the land surface, sea ice 13% of the earth’s surface and glaciers occupy 10% of the land surface (Margesin and Miteva 2011; Cowan et al. 2007). Glaciers are moving gigantic masses of stable ice bodies that consist of highly pressurized and re-crystallized snow which displays oblique slope topology under the influence of gravitational force (Takeuchi et al. 2006). Besides the north and south poles, a massive reservoir of tropical glaciers lies in the region of Karakoram and Himalaya (third pole of the world), with more than 54,252 glaciers that occupy an area of 60,000 km2, mainly with < 0 °C annual temperature (Rafiq et al. 2019). Like other extreme environments, glaciers also harbor all life forms and microorganisms that have successfully colonized permanently frozen habitats favorable for their survival (Anesio and Laybourn, 2012; Morita, 1975).

A diverse range of bacterial flora from cold environments such as glaciers, cold soil, subsoil, frozen lakes, and cold caves have been explored across the globe (Rafiq et al. 2019, 2017; Anesio et al. 2017; Anesio and Laybourn, 2012; Margesin and Miteva 2011; Amato et al. 2007; Takeuchi et al. 2006). From glaciers located in the Himalaya region, researchers have reported bacterial species such as Pedobacter himalayensis (Shivaji et al. 2005), Exiguobacterium indicum (Chaturvedi and Shivaji 2006), Dyadobacter hamtensis (Chaturvedi et al. 2005), Leifsonia pindariensis, Bacillus cecembensis, Cryobacterium roopkundense, Cryobacterium pindariense (Reddy et al. 2010, 2008). Paenibacillus glacialis (Kishore et al. 2010), psychrotrophic proteolytic bacteria such as Bacillus liceniformis, Bacillus subtilis and Pseudomonas aeruginosa (Baghel et al. 2005). On the contrary, algae and fungi are reported in low abundance in cold environments (Lutz et al. 2015; Margesin and Miteva 2011). Globally, glaciers have been found with drastic variations in form of microbial ecology, community structure, and biomass (Liu et al. 2009; Zhang et al. 2007, 2009), which is under the main influences of environmental and climatic factors like geographic location (Takeuchi et al. 2004; Mueller and Pollard 2004), sunlight intensity, winds speed and direction, nutrients and liquid water availability (Bhatia et al. 2006).

Organisms inhabiting the earth’s cryosphere thrive in extremely inhospitable environments for existence and are therefore exposed to nutrient-deficient conditions, characterized by low temperatures, free water availability, freeze-thaw cycles, crystallization, and increased salinity. To endure such challenges, microorganisms adopt intracellular mechanisms in the form of elevated expression of exopolysaccharides, unsaturated fatty acids, transcription and translation of cold-active enzymes, ice nucleating and antifreeze proteins (Margesin and Miteva 2011; Casanueva et al. 2010; Stibal et al. 2008; Sakamoto and Murata 2002). For life continuity, the glacier microbiota performs several requisite roles in biogeochemical cycling, subglacial weathering, mineralization, sustaining food web, and immobilization of numerous xenobiotic compounds (Anesio et al. 2017; Montross et al. 2013; Dong et al. 2006; Takeuchi and Koshima 2004). Moreover, to cope with other competitors for survival, indigenous microbiota produces antimicrobial compounds and violet pigments that exert selective pressure on resident microbes dwelling in cold environments (Allen et al. 2010; Giudice et al. 2007; Nakamura et al. 2003).

Antibiotics production and resistance are natural phenomena with primordial origins (D’Costa et al. 2011; Allen et al. 2010; Martinez 2008; Hall and Barlow 2004) and presently the expeditious spread of antibiotic resistance among bacteria is the preeminent global threat, imposing an emerging problem on public and animal’s health and this occurrence is not only confined to health care settings but has also been majorly observed in cold pristine environments (Van Goethem et al. 2018; Perron et al. 2015; Segawa et al. 2013; Ushida et al. 2010; Wright 2010). Previous research has notably highlighted the phenomenon of antibiotic resistance in clinical settings, but the surge in community-acquired infections by antibiotic-resistant bacteria has called the attention of scientists to search for antibiotic-resistant determinants in natural environments (Gibson et al. 2015). Environmental bacteria inhabiting soil, glacier, and marine water are thought to be the vigorous reservoir of antibiotic-resistant genes (ARGs) (Van Goethem et al. 2018; Perron et al. 2015; Segawa et al. 2013; Ushida et al. 2010; Allen et al. 2009; De Souza et al. 2006) that are massively circulating in clinical settings since bacteria in the natural environment have evolved mechanisms of resistance against antimicrobial compounds for billions of years in the absence of anthropogenic effect (Blair et al. 2015; Forsberg et al. 2012; D’Costa et al. 2006). The discovery of natural reservoirs of antibiotic-resistant determinants manifests the vital role of the natural environment in the dissemination of ARGs (Whitman et al. 1998). Bacteria can effectively resist intrinsically to antibiotics in a natural environment under stressful conditions as shown for Escherichia coli which evolved innate resistance against vancomycin drug (Russell 2003) and innate efflux pump mechanism present in other Gram-negative bacteria (Blair et al. 2015). In addition to acquired mechanisms of antibiotic resistance via mutation and exchange of plasmids, horizontal gene transfer (HGT) of integrons and transposons represent additional acquired resistance mechanisms (Blair et al. 2015; Martinez et al. 2008).

Anthropogenic impact on natural environments, due to misuse of antibiotics for medical and agricultural purposes is mainly responsible for the origination and dissemination of antibiotic resistance in natural environments (Hawkey 2008; Cabello 2006; Chee-Sanford et al. 2001; Goni- Urriza et al. 2000). Bacteria harboring ARGs usually thrive in antibiotic dense habitats such as human and livestock gut, agricultural soil and antibiotic-resistant bacteria (ARB) resulting from atmospheric deposition and aquatic circulation. The dissemination of ARGs occurs across adjoining environments (Gibbs et al. 2006; Levy and Marshall 2004; Lighthart and Shaffer 1995). Such adjacent environments are found to be abundant with ARB compared to remote sites (Malik et al. 2008; Furushita et al. 2003). In contrast, several studies revealed the presence of ARB-carrying ARGs in geographically secluded habitats free of anthropogenic effect (Van Goethem et al. 2018; Perron et al. 2015; Ushida et al. 2010; Allen et al. 2009).

Worldwide, researchers have reported numerous ARGs among bacterial communities thriving in remote cold habitats like; the Gulkana glacier (Eastern Alaska, USA), Ürümqi glacier (Xinjiang, China) (Ushida et al. 2010), Central Asia (China, Tajikistan, and Kyrgyzstan), Himalayas (Bhutan and Nepal), Africa (Uganda), South America (Chile), North America (Alaska), Arctic (Greenland), Antarctica (Segawa et al. 2013) and Mackay glacier (Antarctica) (Van Goethem et al. 2018). The existence of AR bacteria harbored with ARGs in a natural antibiotic-free environment like a glacier is a curious and quintessential case for ARGs (Ushida et al. 2010). Globally, polar cold habitats are extensively searched for microbial biodiversity, antibiotic resistance, biogeochemical processes, and the effect of climate changes (Glaring et al. 2015; Lanoil et al. 2009; Segawa et al. 2010; Christner et al. 2002). On the contrary, the non-polar cold habitats have not been addressed properly. A decade ago, scientists has started to explore these regions (Karakorum Himalaya) for microbial diversity and antibiotic resistance (Rafiq et al. 2017; Shen et al. 2012; Reddy et al. 2010; Zhang et al. 2010; Liu et al. 2009; Chaturvedi et al. 2005). Considering the culturable bacterial diversity and antibiotic resistance in glaciers and other pristine cold environments, the main objectives of this research were to screen the unexplored non-polar glacier of Pakistan for bacterial diversity and antibiotic resistance genes. In this study, bacterial isolates from Passu glacier were isolated and evaluated for culturable diversity, antibiotic susceptibility, and abundance of ARGs.

Materials and methods

Site description of study isolates

The Passu glacier is located in Gilgit-Baltistan, Northern Pakistan, corresponding to the most largely populated non-polar glacier’s region of the Karakorum mountains range covering an area of 72,971 square kilometers that lies at latitude 36° 27’ 59.99” N and longitude 74° 53’ 59.99” E (Fig. 1).

Fig. 1
figure 1

Sampling site location: Passu glacier, Northern Pakistan, Karakorum Mountains Range (a) Photograph of Passu glacier from where samples were collected (Latitude 36° 27’ 59.99” North and longitude 74° 53’ 59.99” East)

Full size image

Sample collection

The bacterial isolates were collected from glacial ice, meltwater, and sediment from three random distant locations (250–300 m apart) during March and April 2017. With the help of a ceramic knife sterilized by ethanol and UV exposure, more than 0.01-meter dept samples were taken to avoid the influence of mammals, birds, and humans. All the samples from three sources were collected and placed in sterile Whirl-Pak Bags (Nasco, USA) with extremely careful measurements taken during sample collection and packaging to avoid any sort of contamination by gloves, masks, and suits. Finally, the collected samples were transferred to the Applied, Environmental and Geomicrobiology Lab, Department of Microbiology, Quaid-i-Azam University Islamabad, and stored at -70ºC for subsequent uses.

Chemicals and reagents

Media (R2A medium, Nutrient Agar, Nutrient Broth, Lauria-Bertani Broth, Mueller Hinton Agar), glycerol, and normal saline (0.85%NaCl) from Sigma Chemicals (St. Lious, MO, USA), were used for bacterial isolation while antibiotic discs were obtained from Liofilchem, Italy.

Total viable count, isolation, and morphological identification of bacteria

Both aqueous and sediment samples were diluted 1:10 (1mL sample and 9 mL of sterile saline solution or 1 g of sediment in 9 mL of sterile saline solution). Volumes of 200 µL were spread over R2A medium agar plates and incubated at 4 and 15 °C for 40 to 60 days to isolate both psychrophilic and psychrotrophic isolates. Afterward, examining different colonies on R2A plates, the selection, isolation, and characterization of culturable bacteria were performed by following Zhang et al. criteria (Zhang et al. 2013). The total viable count of isolates was calculated by evaluating colony-forming units (CFU) per gram (g) or milliliter (mL) of the corresponding sample. Isolates were identified based on colony morphology, growth requirements, and 16S rRNA gene sequencing. All isolates were constantly cultured on Lauria-Bertani agar and R2A agar media and were preserved in glycerol and LB broth at -20 °C for further study.

Antibiotic susceptibility testing

The antibiotic susceptibility profile of the isolated bacteria was determined in vivo by using the Kirby Bauer agar disc diffusion method (Bauer et al. 1996) and standardized by guidelines of the Clinical and Laboratory Standards Institute (CLSI 2018). A total of 29 antibiotic groups from 10 classes were used for antibiotic susceptibility evaluation. Gram-positive bacteria were tested against 28 antibiotics while Gram-negative bacteria were against 21 antibiotics. The antibiotics panel included; Fluoroquinolones (levofloxacin 5 µg, ciprofloxacin 5 µg, ofloxacin 5 µg, norfloxacin 10 µg and nalidixic acid 30 µg), Penicillins (amoxicillin 10 µg, penicillin G 10 µg, carbenicillin 100 µg, ticarcillin 75 µg, oxacillin 1 µg and piperacillin 100 µg), Cephalosporins (ceftriaxone 30 µg, cefotetan 30 µg, cephalexin 30 µg, cephalothin 30 µg, cefepime 30 µg, ceftazidime 30 µg, cephazolin 30 µg, cefpodoxime 10 µg and ceftaroline 30 µg), Carbapenems (meropenem 10 µg, imipenem 10 µg and ertapenem 10 µg), Monobactam (aztreonam 30 µg), Oxazolidinone (linezolid 30 µg), Macrolide (erythromycin 5 µg), Glycopeptide (vancomycin 30 µg) Glycycline (tigecycline 15 µg) and Nitroimidazole (metronidazole 5 µg).

Multiple antibiotic resistance index

Multiple antibiotic resistance (MAR) indexes of antibiotic-resistant bacteria were determined by utilizing the following formula equation, suggested by Krumperman (1983).

MAR index = a/b

‘a’ shows number of resistant antibiotics

‘b’ refers to total number of tested antibiotics

Bacteria with MAR index values ≥ 0.2 are thought to be originated from contaminated sources with a high-risk (Krumperman 1983).

Genomic DNA extraction and PCR amplification

Chromosomal DNA was extracted using DNA Purification Kit # Ko721 (Gene JET Genomic, Thermo Scientific) by following the manufacturer’s guidelines. PCR amplification of 16S rRNA gene, ARGs, and class1, 2 and 3 integron integrase genes were carried out in Eppendorf Gradient Thermocycler in a final volume of 30 µL reaction which contained 15 µL of Dream Taq green PCR 2x Master Mix (Thermo Scientific), followed by 1 µL of each oligonucleotide primer (1µM) (Integrated DNA Technologies), 8 µL of PCR grade nuclease-free water (Thermo Scientific) and 5 µL of template DNA. The forward and reverse primers sequences of 16S rRNA gene, ARGs, and integron integrase class 1, 2 and 3, expected amplicon size, and PCR cycling conditions are summarized in (Supplementary File 1). For visualization of desired amplified products, electrophoresis was carried out on agarose gel (1.5%), stained with ethidium bromide, and combined with GeneRuler 100 bp DNA Ladder (Thermo Scientific) as a standard marker, followed by band analysis through gel documentation system (Alpha Innotech, Biometra).

Sanger sequencing

PCR products of 16S rRNA, ARGs, and class 1 integron integrase were purified and quantified using PCR Purification Kit (Gene JET, Thermo Scientific) and Qubit 3 Fluorometer (Invitrogen), respectively. Qubit Fluorometer dsDNA HS Assay Kit was used to quantify the final purified products. Bi-directional Sequencing of 16 S rRNA (27 F, 1492R), ARGs, and class 1 integron integrase was performed through the University of Arizona Genetic Core (https://uagc.arl.arizona.edu/services/services/dna-sequencing), on Applied Biosystems 3730XL DNA Analyzer, with 600 bases sequence per reading in one direction. The 16S rRNA sequences of the bacterial isolates obtained in this study were compared for sequence homology search through the Basic Local Alignment Search Tool (BLAST) program (https://blast.ncbi.nlm.nih.gov/Blast.cgi) for further evaluation. The gene sequences were submitted to the GeneBank Nucleotide Sequence Database, NBCI (National Center for Biotechnology Information), and the corresponding accession numbers are provided in (Supplementary File 2).

ARGs and class 1 integron integrase sequence analysis

The nucleotide sequences of ARGs and integron integrase class1 were further scrutinized through BLAST search and compared to the relative available sequences in the NCBI database. Different variants of ARGs were determined from The Comprehensive antibiotic Resistant Database (CARD) (https://card.mcmaster.ca/analyze/blast). For variant calling, each gene nucleotide homology sequences were downloaded from relative reference genes available in the NCBI database and translated to amino acids with the ExPASy Translate Tool (https://web.expasy.org/tools/translate/dna.html). Clustal W was used to align the sequences. To evaluate the amino acid variation, alignment was performed with CLC Main Workbench 8 software. To evaluate the evolutionary relationship of ARGs and intI1 MEGA X software was used (Kumar et al. 2018).

Results

Description and retrieval of bacteria from glacier

In this study, a total of 65 bacterial isolates were collected from Passu glacier, of which 43 (66.1%) were retrieved from glacial sediment, followed by 14 (21.5%) from ice and 8 (12.3%) from meltwater, respectively. Based on the R2A agar plate examination, in terms of CFU g− 1 and CFU mL− 1, bacteria-enriched colonies were obtained from glacial sediment at 4 and 15 °C (Table 1). Among isolated species, 43 (66.15%) were Gram-negative bacteria whereas 22 (33.84%) were Gram-positive bacteria. Out of the total retrieved isolates, only 5 (7.69%) retained their growth at 37 °C. Morphological and growth characteristics of retrieved bacterial isolates are documented in (Supplementary File 3).

Table 1 Total viable count of retrieved isolates in term of CFUmL− 1 or g− 1
Full size table

Antibiotic susceptibility testing

The tested bacterial isolates showed significantly varying phenotypic resistance to a panel of 29 antibiotics representing 10 classes. The susceptibility profile revealed that Gram-positive and Gram-negative bacteria both exhibited significant resistance to the majority of antibiotics. More precisely, Gram-positive bacteria exhibited more resistance to Fluoroquinolones than Gram-negative bacteria, whereas Gram-negative bacteria were more resistant to other antibiotic classes. Among Gram-positive bacteria that exhibited resistance to Fluoroquinolones, resistance was maximum against norfloxacin, while levofloxacin was a more effective drug with the least resistance. The penicillin group showed the least effective against tested bacteria except for piperacillin with only 1 isolate showing resistance to this antibiotic. Out of the Cephalosporins class, the isolated bacteria showed high sensitivity to ceftriaxone and resistance to cephazolin. Resistance to Carbapenems was maximum against ertapenem and minimum against imipenem. The resistance against linezolid, erythromycin, vancomycin, and metronidazole was also significant among Gram-positive bacteria.

Among the Gram-negative isolates, Fluoroquinolone resistance was also prevalent but to a letter extent against levofloxacin and more against norfloxacin and nalidixic acid. Gram-negative bacteria also revealed prevalent resistance to Cephalosporins with a maximum against cephalothin and a minimum against ceftaroline and ceftriaxone. Gram-negative bacteria were slightly resistant to Carbapenem group antibiotic imipenem while this group exhibited significant resistance against meropenem and ertapenem. Considering Monobactam, Macrolide, Glycycline, and Nitroimidazole, the most effective drug was tigecycline, whereas 100% of the Gram-negative isolates were resistant to metronidazole. The antibiotic susceptibility profile of Gram-negative and Gram-positive bacterial isolates is illustrated in (Table 2).

Table 2 Antibiotic susceptibility profile of Gram-positive and Gram-negative bacteria isolated from glacier
Full size table

Multiple antibiotic resistance (MAR) index

MAR indexes for AR bacteria were calculated through the formula designed by Krumperman (Krumperman 1983). When MAR indices were calculated, 87.9% of tested isolates yielded MAR index value > 0.2. Among Gram-positive isolates, 86.04% have > 0.2 MAR index whereas 90.9% Gram-negative isolates exhibited > 0.2 MAR index. Gram-positive species showing these patterns of MAR included Staphylococcus equorum (HP19), Leucobacter aridicollis (HP22), Leucobacter komagatae (HP51) and Arthrobacter sp. (LP2) while Gram-negative bacteria included Brevundimonas diminuta (HP21), Rahnella inusitata (HP3) and Alcaligenes faecalis (HP55) that showed highest MAR index value corresponding to 0.7. Multiple antibiotic resistance indexes and resistance patterns of bacteria are recorded in (Table 3).

Table 3 Antibiotics resistance pattern and MAR indexes of bacteria isolated from glacier
Full size table

PCR amplification and sequencing of 16S rRNA

The amplification of specific 1350 bp bands was visualized under the UV trans-illuminator (Fig. 2). Based on 16S rRNA gene sequencing, the bacterial isolates detected belong to the phyla Proteobacteria 40 (62%), Actinobacteria 12 (18%), Firmicutes 9 (14%) and Bacteroidetes 4 (6%). The relative abundances of bacterial phyla in three different sample sources isolated from the glacier are represented in (Fig. 3). Bacterial communities and their relative abundance, distribution plus structural community in glacial sediment, ice, and meltwater are shown in (Fig. 4).

Fig. 2
figure 2

PCR amplicons of 16S rRNA gene on 1.5% agarose gel. Lane 1 to 17: 27 F 1492R (1350 bp), B: Blank, M: 100 bp DNA Ladder (Invitrogen)

Full size image
Fig. 3
figure 3

Bacterial Phyla relative abundance in glacier samples (a) shows the whole bacteria abundance in glacier (b) bacterial phyla in glacial sediment (c) glacial ice and (d) glacial meltwater

Full size image
Fig. 4
figure 4

Bacterial communities and their relative abundance, distribution plus structural community in glacial sediment, ice and meltwater identified by amplicon sequencing of 16S rRNA gene

Full size image

Molecular detection of ARGs and class 1 integron integrase

The presence of ARGs and class1 integron integrase was revealed by PCR amplification using primers and cycling conditions described in the methods. The ARGs detected ratios were blaCTXM-15 (21.53%), blaNDM-1 (24.61%), blaAmpC (9.23%), blaVIM-1 (7.69%), blaSHV (7.69%), blaTEM-1 (18.46%), blaOXA-1 (1.53%), blaGES (6.15%), blaCMY-4 (6.15%), blaDHA (7.69%), sul1 (21.5%), sul2 (24.6%), tet(A) (12.30%), tet(B) (7.69%), qnrB (15.38%), aac(6)-Ib3 (29.23%), mecA (18.46%), cat (15.38%), qepA (15.38%), gyrA (24.6%) and intl1 (24.6%), respectively. All tested strains were negative for blaKPC, qnrA, vanA, ermA, ermB, intl2, and intl3. The representative amplified bands of desired genes are illustrated in (Fig. 5). The PCR analysis showed that aac(6)-Ib3, blaCTXM-15, blaNDM-1, sul1, sul2, and gyrA genes were consistently present among tested isolates, while single isolate harbored blaOXA-1. Moreover, ARGs and integron integrase class 1 were more prevalent in Gram-negative bacteria from glacial sediment than Gram-positive bacteria.

Fig. 5
figure 5

PCR amplicons of ARGs and intI1 on 1.5% agarose gel. Lane 1: CTX-M (1000 bp), Lane 2: blaNDM-1 (214 bp). Lane 3: blaOXA (813 bp), Lane 4: blaAmpC (189 bp). Lane 5: blaCMY (1432 bp), Lane 6: blaSHV (300 bp), Lane 7: blaVIM (189 bp), Lane 8: blaTEM (1150 bp), Lane 9: blaDHA (404 bp), Lane 10: blaGES (890 bp), Lane 11: gyrA (344 bp), Lane 12: sul1 (163 bp), Lane 13: sul2 (191 bp), Lane 14: aac(6’) (482 bp), Lane 15: tetA (210 bp), Lane 16: tetB (659 bp), Lane 17: qnrB (476 bp), Lane 18: intl1 (196 bp), Lane 19: qepA (596 bp), Lane 20: cat (122 bp), Lane 21: mecA (527 bp), M: 100 bp DNA Ladder (Invitrogen)

Full size image

Among the Gram-negative bacterial strains, Brevundimonas diminuta (HP21) and Rahnella inusitata (HP3) harbored intl1 plus 19 and 17 ARGs, respectively. Thirteen ARGs were identified in Alcaligenes faecalis (HP55), however, intl1 was not associated with this bacterium. Meanwhile, intl1 plus 11 and 8 ARGs were identified in the Gram-positive bacterial strains Staphylococcus equorum (HP19) and Leucobacter aridicollis (HP22). Arthrobacter sp. (LP2) was found positive for 7 ARGs and negative for intl1. The distribution of ARGs and intI1 among the tested bacteria are categorized in (Table 4) and the relative abundance of ARGs in 3 different sample sources is summarized in (Fig. 6).

Table 4 Distribution of antibiotic resistant genes and class 1 integron integrase among Gram-negative and Gram-positive bacteria isolated from glacier
Full size table
Fig. 6
figure 6

Relative abundance of ARGs and class 1 integron integrase among glacier sediment, ice and meltwater

Full size image

Considering the bacterial Phyla, Proteobacteria harbored the largest number of ARGs and intI1, followed by Actinobacteria, Firmicutes, and Bacteroidetes, respectively. The abundance of ARGs and intI1 among bacterial phyla is listed in (Table 5).

Table 5 Abundance of antibiotic resistant genes and class 1 integron integrase among 4 bacterial phyla
Full size table

ARGs and class 1 integron integrase sequence analysis and protein alignment

The BLAST sequence homology search was evaluated with > 98% similarity for all ARGs and intI1 except blaOXA-1 (97.36%) and aac(6)-Ib3 (97.86%) and different variants of ARGs were determined through the CARD database. When compared to the reference gene, blaOXA-1 amino acids alignment showed variations at terminal positions (Leu160→ Val, 161Gln→ Trp, 162Asn→ Glu, 163Gly→ Asn, 165Phe→ Cys, 166Glu→ Arg, 170Ile→ His). The blaNDM-1 gene was found with single amino acid variation (61Asp→ Tyr), blaSHV with 2 amino acids (80Val→ Leu, 84Arg→ Pro), whereas, (tetA) with 3 amino acids variations (4Pro→ Arg, 52Leu→ Val, 54Phe→ Ile) and sul2 with 2 amino acids variations (4Ala→ Thr and 62Pro→ Thr). Moreover, qnrB gene amino acids alignment demonstrated 3 amino acids variations at positions (6Ile→ Asn, 103Thr→ Ser, 128Arg→ Pro), gyrA by 2 amino acids (5Phe→ Tyr, 46Asp→ Glu), and intI1 protein alignment illustrated the amino acids coverage and consensus with 2 amino acids variation at positions (60Val→ Asp, 61Phe→ Leu). The complete description and evolutionary relationship details of ARGs and class 1 integron integrase are summarized in (Supplementary File 4).

Discussion

This research work includes a comprehensive assessment of antibiotic resistance and prevalence of ARGs and class 1 integron integrase among culturable bacteria isolated from a non-polar glacier (Passu, Pakistan) in the Karakoram (Himalaya) region. A total of 65 bacterial species were isolated from glacial sediment, ice, and meltwater. 16S rRNA gene sequencing identified different bacteria belonging to multiple phyla including Flavobacteria, Firmicutes, Actinobacteria, Bacteroidetes, and Proteobacteria. From various cold and frozen habitats worldwide such as glaciers, permafrost, ice sheets, and lakes, bacterial phyla belonging to Proteobacteria, Bacteroidetes, Actinobacteria, Firmicutes, Chlroflexi, and Acidobacteria have been frequently reported (Margesin and Miteva 2011; Junge et al. 2011). Early studies from Hindu Kush (Tirich Mir, Chitral, Pakistan) identified Firmicutes, Actinobacteria, Bacteroidetes, Flavobacteria, and Proteobacteria in Siachen glaciers and samples collected from multiple sources including glacial ice, sediment, and meltwater (Rafiq et al. 2019, 2017). In our study, (66.1%) of isolates were retrieved from glacial sediment, followed by (21.5%) from ice and (12.3%) from meltwater. Glacial sediment is rich in nutrients with slightly higher temperatures in comparison to glacial ice and meltwater which favors diverse microbial flora (Parnell and McMahon 2016). Our findings revealed a high prevalence of Gram-negative bacteria (66.15%) when compared to Gram-positive (33.84%). These bacterial communities were dominated by Proteobacteria (61.53%), followed by Actinobacteria (18.46%), Bacteroidetes (6.15%), and Firmicutes (13.84%), similar to studies conducted previously from Siachen glacier, Greenland ice Sheet and Finnish Lapland (Rafiq et al. 2017; Musilova et al. 2015; Männistö and Häggblom 2006) which reported Proteobacteria (Alphaproteobacteria, Betaproteobacteria, Gammaproteobacteria) and Gram-negative bacteria as most predominant bacteria. Boetius et al. (2015) reported Gram-positive bacteria as the most dominating strains from Arctic ice sheets and Antarctic glaciers which contrasts with our findings. Seasonal variations in cold environments and geographical locations may contribute to bacterial diversity dominancy and variation because of the counter-selection of more adoptable bacteria (Zhang et al. 2010, 2007; Bhatia et al. 2006).

Bacteria were evaluated for antibiotic resistance for a panel of 29 antibiotics and the results revealed significant antibiotic resistance. The antibiotic susceptibility profile revealed that Gram-positive bacteria were comparatively more resistant to antibiotic class Fluoroquinolones than Gram-negative bacteria, whereas Gram-negative bacteria showed more resistance to the rest of antibiotic classes. This might be due to the inherited less susceptibility of Gram-positive bacteria towards Fluoroquinolones where a single mutation can confer antibiotic resistance. Staphylococcus equorum (HP19) was resistant to nearly 78.57% of tested antibiotics (including more than one antibiotic from the same class), followed by Leucobacter aridicollis (HP22) and Arthrobacter sp. (LP2) that were resistant to 75% antibiotics each, Leucobacter komagatae (HP51) to 71.42% and Bacillus pumilus (HP18) to 67.85% antibiotics. Carnobacterium maltaromaticum (HP36) showed a minimum of 21.42% antibiotic resistance. Among the Gram-negative bacteria, Brevundimonas diminuta (HP21) was resistant to a maximum of 76.19% antibiotics, Rahnella inusitata (HP52) and Alcaligenes sp. (HP55) to 71.42% each, while Flavobacterium antarcticum (HP20) and Serratia marcescens (HP50) were resistant to 66.66% each. Alcaligenes faecalis (HP2) and Acinetobacter calcoaceticus (LP5) were resistant to 52.38% of overall tested antibiotics. Our results were in contradiction to reports by Rafiq et al. (2017) from the Siachen glacier (Karakoram, Pakistan), which reported increased antibiotic resistance among Gram-positive bacteria corresponding to ofloxacin 11.76%, ceftriaxone 58.82%, cefotaxime 76.47%, imipenem, and vancomycin each 64.70% resistance, whereas among Gram-negative bacteria resistance pattern was ofloxacin 24.24%, nalidixic acid 15.15%, cefotaxime 45.45% and imipenem 51.51%, among nearly similar genus isolates (Rafiq et al. 2017). Meanwhile, we reported ofloxacin 27.27%, ceftriaxone 31.81%, nalidixic acid 36.36%, cefpodoxime 77.27%, imipenem 9.09% and vancomycin 81.81% among Gram-positive bacteria while among Gram-negative ofloxacin 9.3%, nalidixic acid 18.6%, cefotaxime 62.79% and imipenem 11.62% were resistant. In another study, conducted by James and Wong (2015), the highest resistance was reported against vancomycin (64.2%), metronidazole (92.8%) and ceftazidime (42.8%) among Antarctic isolates, including Arthrobacter and Pedobacter species and these are in close association with our isolates’ resistance pattern.

In addition, Tomova et al. (2015) have reported higher degrees of antibiotic resistance delineated in bacteria procured from sediment and soil samples of Antarctic islands, with 100% multiple antibiotic resistance among Proteobacteria and overall, 79% multi-drug resistance, most frequently towards cephazolin (75%), erythromycin (62%) and vancomycin (58%), while Acinetobacteria were all sensitive to tested antibiotics. There are no major differences in resistance profile in comparison to our study as among our tested isolates, Proteobacteria showed a higher degree of multiple antibiotic resistance while strain Rahnella inusitata (HP56) was the most sensitive strain. The area of Passu glacier is usually free of anthropogenic effects and still, the resistance pattern of currently tested isolates was comparable to bacterial resistance inhibiting antibiotic-rich environments. The same scenario has been reported by James and Wong (2015) who evaluated that bacteria living in anthropogenic free environment exhibit significant antibiotic resistance same as in habitats exposed to major human activities since pristine Antarctic bacteria were resistant to multiple antibiotics compared to tropical bacteria. On the other hand, Miller et al. (2009) described the lower frequency of multiple antibiotic resistances among native Antarctic bacteria which increased with an increase in anthropogenic activities. Usually, it is believed that the effect of anthropogenic activities on pristine environments plays a vital evolutionary role in the origination of multiple AR bacteria (Ushida et al. 2010; Cabello 2006; Goni-Urriza et al. 2000). On the contrary, different physiological states like trace nutrients, low-high temperature, pressure, presence of secondary metabolites might be involved to stimulate stress signaling pathways in the upset bacterial genome resulting in resistant traits (antibiotics resistance, heavy metals resistance) in bacteria thriving in hostile environments. Moreover, resistance mechanisms can be acquired by gaining additional resistant determinants (genes, transposons, integrons, gene cassettes). However, the phenomenon of multiple antibiotic resistance among bacteria in a pristine environment like Passu glacier is strange and not clear as these habitats are usually free of anthropogenic effects.

Tropical glaciers are of major interest for researchers to study, regarding antibiotic resistance, as tropical glaciers interact differently with climate changes in comparison to glaciers located in mid and high latitudes (Kaser 1999). Passu glacier is also located in tropical regions of Pakistan and is exposed to drastic seasonal variations. A study on tropical glacier located in Venezuelan Andes, by Ball et al. (2014) reported an elevated degree of multiple resistant strains with maximum resistance against ampicillin (64.44%), followed by chloramphenicol, nalidixic acid, and penicillin (57.77% each). streptomycin, kanamycin, and tetracycline resistances were (24.44%, 22.22%, and 4.4%), respectively, whereas in the current study resistant ratios for nalidixic acid (36.36% Gram-positive, 18.60% Gram-negative) and penicillin were (86.36% Gram-positive). It is generally reported that antibiotic resistance and heavy metals resistance coexist and are common traits expressed by bacteria trapped in glacier environments (Sherpa et al. 2020). From Antarctic shallow sediments, nearly free of antibiotics, chemicals, and heavy metals pollution, Giudice et al. (2013) reported multiple AR bacteria with remarkable resistance to heavy metals and chemicals, thus confirming the presence of coexistence of antibiotics and metals resistance genes on the same genome and this is not only confined to glaciers environment but observed ubiquitously (De Souza et al. 2006).

MAR index provides useful information to better address the risks associated with exposure to environments loaded with antibiotic-resistant bacteria. An arbitrary value of > 0.2 is an indicator of a high-risk source of contamination where antibiotics are frequently used (Krumperman 1983). A significant number of bacterial isolates (87.9%) from this current study showed MAR index value > 0.2. Among Gram-positive bacteria, 86.04% had > 0.2 MAR index value, whereas 90.9% Gram-negative isolates were found with > 0.2 MAR index value, emphasizing high degrees of risk source of pollution or bacteria are previously exposed to antimicrobials. Among Gram-positive strains, Staphylococcus equorum (HP19), Leucobacter aridicollis (HP22), Leucobacter komagatae (HP51), and Arthrobacter sp. (LP2) while Brevundimonas diminuta (HP21), Rahnella inusitata (HP3) and Alcaligenes sp. (HP55) from Gram-negative showed highest MAR index value 0.7. From Siachen glacier, bacteria reported by Rafiq et al. (2017), were found to possess the highest MAR index value of 0.8 for Bevibacterium sp., followed by Rhodococcus sp., Arthrobacter sp. and Bacillus simplex (0.7 each) which is comparable to our results. Environmental bacteria thriving in cold environments are found to be excessively producing antimicrobial, antibacterial compounds (Giudice et al. 2007) and antibacterial violet pigments (Nakamura et al. 2003). Psychrophilic bacteria have also the potential to biosynthesize extracellularly highly stable silver nanoparticles at low temperatures, with remarkable antimicrobial activity (Shivaji et al. 2011). In natural environments, under the selective pressure of antimicrobial compounds, bacteria may evolve to resist the action of these compounds. Psychrophilic bacteria previously isolated from Pakistan (Karakorum region), Passu glacier (Rafiq et al. 2016), Hindu Kush range glacier (Rafiq et al. 2019), and Siachen glacier (Rafiq et al. 2017) were found to possess strong antimicrobial potential.

In this current study, PCR amplification revealed that tested bacteria were found positive for 21/28 (75%) tested ARGs (ß lactam, non- ß lactam, and integron integrase). Gram-negative bacteria were found with maximum ARGs (67.85%, 19 ARGs) while among Gram-positive bacteria maximum (39.28%, 11 ARGs) were detected. Scientists from all over the world have reported numerous ARGs from cold environments. A diverse set of 117 ARGs showing resistance to ß-lactam antibiotics, aminoglycosides, macrolides, rifampicin, and tetracycline have been reported in Mackay glacier (Antarctica), among which 71% ARGs were detected in Gram-negative bacteria while only 9% were detected in Gram-positive bacteria (Van Goethem et al. 2018), which is following our results. Moreover, Ushida et al. (2010) detected blaCTX-M, ampC, cmrA, msrA, msrB, aacC1, aph6, tetD, tetE and tetG from Gulkana glacier in Eastern Alaska, (USA) and Ürümqi glacier in Xinjiang (China). In addition, a massive collection of ARGs including; blaIMP, blaOXA, vanA, blaCMY1, blaCMY2, ampC2, strA, msrA, msrB, aacC1, aacC2, aac2’Ic, aac3, aac(6), ermA, ermC, ermM, ermML, ermTR, aadA, aadB, aadE, aadK, aph6, aph3’, cat1, catA4, catB3B4, catB5B8, catB7, tetD, tetG, tetL, tetM, tetO1, tetO2, tetS1, tetS2, tetX, tetW, cmlA, cmlV, cmrA, mefA, mefE and cmx were detected in remote pristine glaciers of Central Asia (China, Tajikistan and Kyrgyzstan), the Himalayas (Bhutan and Nepal), Africa (Uganda), South America (Chile), North America (Alaska), Arctic (Greenland) and Antarctica regions, thus accentuating the worldwide distribution of ARGs across the cryosphere (Segawa et al. 2013). In our study, 81.81% of Gram-positive bacteria were phenotypically resistant to vancomycin but all remained negative for vanA gene PCR amplification, which was the strange thing among our findings. This might be due to the innate active efflux mechanisms in bacteria (Blair et al. 2015).

In a recent study conducted in Greenland, Svalbard, and Caucasus glaciers, Makowska et al. (2020) documented different ß-lactamase ARGs (blaCTX-M, blaSHV, blaOXA, blaGES, blaTEM, blaCMY, blaVEB, blaDHA) plus intI1 among 138 strains. More precisely, 4.6% corresponded to blaCMY, 16.3% to intI1 from Greenland, 4.6% corresponded to blaOXA, and 2.2% each to blaSHV and blaTEM, plus 14.8% to intI1 from Georgia. Moreover, from Spitsbergen (High Arctic Island, Norwegian territory), 1.2% were associated with blaOXA, and 1.1% with intI1, nearly under our findings. Another report by Shen et al. (2019) found a prevalence of 11.3% of ß-lactamase from Urumqi glacier China. Allen et al. (2010) reported blaNDM-1, blaSHV-1, blaKPC, and blaGES-1 ARGs from the Arctic and sub-Arctic environment. In addition, McCann et al. (2019) reported 8% ß-lactamase genes from high Arctic environments. Previous reports highlighted the presence of ß-lactamase genes and resistance against numerous ß-lactam antibiotics (ampicillin, tetracyclines, chloramphenicol, sulfamethoxazole, and trimethoprim) among AR bacteria isolated from bird’s feces samples from Arctic environment which resolutely involucrated migrating birds in the propagation of ARGs and integrons (Literak et al. 2014). Moreover, through clonal expansion, dissemination of AR bacteria, and exchange of mobile genetic elements integrated with resistant determinants via HGT, ARGs can frequently propagate among bacterial communities in a new environment (Van Elsas and Bailey 2002).

The detection of integrons in glacier environments regardless of location is associated with biotic pollution in nature (Makowska et al. 2020) and can be used as an indicator of anthropogenic impacts on pristine environments (Ushida et al. 2010). Besides, the plausible modes of transmission of AR bacteria and integrons through hydrological systems, wind, migrating birds, vertebrate feces, tourists, and airborne bacteria may induce the geographical dispersal of ARGs and AR bacteria (Makowska et al. 2020; Literak et al. 2014; Segawa et al. 2013). Moreover, the mixing of glacier melt and rainwater induces bacterial dispersion and promotes HGT which contributes to ARGs dissemination (Makowska et al. 2020). Initially, integrons were only confined to clinical bacteria but studies conducted in natural pristine environments indicated their ubiquitous presence in environmental microbiomes and their emergence under the influence of selective pressure caused by antibiotics and other pollutants (Cambray et al. 2010). Our integron-harbored bacteria showed resistance to multiple antibiotics when compared to intI1-negative bacteria which manifest the key role of HGT among diverse glaciated bacteria which is supported by other studies (Makowska et al. 2020; Zhang et al. 2018) .

In this current study, we further evaluated ARGs and intI1 protein alignment with counter-clinical-based references to find out amino acid variations between natural and clinical bacteria. The ARGs and intI1 showed > 97% BLAST similarities and the majority showed aligned identities to Gram-negative bacteria. For blaOXA-1, blaNDM-1, blaSHV, sul2, (tetA), qnrB, gyrA and class 1 integron integrase amino acid variations were detected, whereas blaCTX-M15, blaCMY-4, blaTEM-1, aac(6)-Ib3, mecA and tetB were negative for amino acid variations (Supplementary File 4). The sequence similarity (99%) for ARGs from glacier bacteria isolated from North Sikkim glaciers has been reported by Sherpa et al. (2020) where aligned identities belonged to Gram-negative bacteria including clinical-based Escherichia coli and Acinetobacter baumanii. In our current study, Staphylococcus species (S. saprophyticus (HP7) and S. equorum (HP19)) were positive for mecA gene beside other ARGs which correlated with the study conducted by Kashuba et al. (2017), where the ancient Staphylococcus haemolyticus strain isolated from Mammoth Mountain permafrost was positive for mecA gene and showed 99% homology and 100 query length in both cases. Unfortunately, the available literature data is limited regarding the amino acids-based alignment of ARGs and integrons among glaciated bacteria as well with over the counter clinical-based ARGs which need more extensive research for intricate genome-wide comparative analysis.

Conclusion

In an antibiotic-free glacier environment, the occurrence of indigenous AR bacteria with class 1 integron integrase is a quintessential scenario and an archetypal case of antibiotic-resistant elements in an antibiotic-free environment. The current study also reflects the same where the culturable glacier bacteria was found with significant antibiotic resistance, integrated with various ARGs plus class 1 integron integrase and was more prevalent among Proteobacteria, Gram-negative from sediment source. The Brevundimonas diminuta (HP21) and Staphylococcus equorum (HP19) were the most resistant strains, yielded the highest MAR indexes, and harbored maximum ARGs. The sequence alignment of some ARGs and intI1 showed amino acids variation and were found in associations with the clinical-based counterparts and it might be inferred that the presence of ARGs and intI1 in glacier bacteria is an indication of the evolutionary process of opportunistic as well as relevant clinically resistant species. In a glacier environment, native bacteria are exposed to an eternal selective pressure of elevated UV radiation doses, daily freeze-thaw cycles, oligotrophy, radionuclides, and heavy metals which evolves antibiotics production and counter-resistance mechanisms. The existence of AR bacteria in glaciers can be indicated as an adaptation to survive hostile environments. Under the influence of climatic changes like; global warming, the glaciers will favor the expansion of ecosystems and it is crucial to mention the risks that are associated with the dissemination of glacial AR bacteria to new habitats when there is an elimination of the naturally existing ecological filters along with the migration of animals and human activities near non-polar glaciers.

As a final remark, it’s worth mentioning that; (1) The existence of AR bacteria possessed with ARGs and class 1 integron integrase abundance among non-polar glaciated bacteria, (2) The rapid melting of glaciers with emancipation and reactivation of dormant bacteria, and (3) The promising donor behavior of environmental bacteria in sequential episodes of HGT, is of significant concern regarding public health. Therefore, precautionary measures should be taken to monitor glacier melting water as a potentially hazardous source, especially in areas where the only source of domestic fresh water is from glaciers.