Identification of microorganisms can be performed using classical microbiology techniques, including culture-dependent ^[1][21] techniques. These methods involve the use of normal, oligotrophic, or specialized culture media to count, purify, and identify microbial isolates ^[2][16]. However, a major challenge in culture-dependent studies is to find suitable culture conditions for the cultivation of different bacterial species ^[1][21]; in addition, this method provides limited information on community structure ^[3][30]. Molecular biology techniques have been successfully used in the detection of microorganisms in their environment. Such techniques, based on the detection of nucleic acids, allows the differentiation of microorganisms within complex microbial communities ^[4][31]. Detection of microorganisms is usually based on the sequences of the small subunit ribosomal RNA (rRNA) genes (16S for prokaryotes) because rRNA genes are highly conserved and contain a level of variability that allows the identification of microorganisms detected by their sequences and the possibility of performing phylogenetic analyses with their closest relatives. A variety of methods have been used to analyze these sequences, including polymerase chain reaction (PCR)-based fingerprinting methods, such as DGGE (denaturing gradient gel electrophoresis) ^[5][6][32,33], T-RFLP (terminal restriction fragment length polymorphism) ^[6][7][33,34], clone library construction ^[8][35], quantitative PCR assays (including those targeting functional genes of interest), sequencing, and the use of stable isotope probing methods ^[9][10][36,37]. DNA sequencing approaches are very useful for phylogenetic identification ^[11][38], and more recently, next-generation sequencing (NGS) tools on a variety of platforms, such as Roche FLX 454 pyrosequencing ^[12][39], Illumina ^[13][14][15][40,41,42], and SOLiD and Ion Torrent PGM ^[16][17][43,44], have been applied to the study of cave microorganisms ^[2][16].

New tools for understanding the microbial world have been provided by culture-independent methods ^[18][45], such as metagenomics, metaproteomics, metatranscriptomics, and metabolomics, which are fundamental for fully identifying microbial diversity and recognizing its interactions with biotic and abiotic factors ^[11][38]. Metagenomics approaches, as functional sequence-based analyses of the collective microbial genomes contained in an environmental sample ^[19][46], have also evolved in recent years. In the classical metagenomic approach, environmental DNA was cloned into vectors using ultracompetent host strains. The resulting clone libraries were then screened for either specific marker genes (sequence-driven approach) or metabolic functions (function-driven approach) ^[20][47]. Currently, metagenomics typically involves two specific sequencing strategies: amplicon sequencing, most commonly of the 16S rRNA gene as a phylogenetic markers, or shotgun sequencing, which captures the full range of DNA in a sample ^[14][16][21][41,43,48]. Typically, 16S rRNA gene amplicon sequencing is limited to taxonomic classification at the genus level, depending on the database and classifiers used, and provides limited functional information ^[22][49]. Shotgun metagenomics provides a more robust and reliable assessment of microbial diversity and has the advantage of classifying bacteria at the species and strain level. It also allows the functional relationships between hosts and bacteria to be studied by directly determining the functional content of samples and allows the exploration of previously unknown microbial life that would otherwise remain unclassified. However, the relatively high cost of shotgun metagenomics and more challenging bioinformatics have prevented its widespread use for microbiome analysis ^[14][16][41,43].

Rausch et al. 2019 ^[14][41] presented a study to systematically compare the experimental and analytical aspects of the two main technical approaches for microbial community characterization: 16S rRNA gene amplicon (variable regions V1, V2 and V3, V4) and shotgun sequencing. In addition, for each region, a one-step fusion PCR was compared with a two-step procedure, resulting in five different sequence profiles for each sample. The many aspects of bacterial community characterization are consistent when analyzed by different methods.

In another case, an investigation of the taxonomic composition of microorganisms in the Manao Pee Cave soil using high throughput metagenomic sequencing showed results consistent with the 16S rRNA study based on community structure. The shotgun metagenomic sequencing confirmed that Actinobacteria and Proteobacteria were the dominant bacterial phyla in the Manao Pee Cave community. Shotgun metagenomic sequencing provided higher resolution, allowing the detection of more microbial taxonomic profiles than 16S rRNA sequencing, especially of rare microorganisms. For example, at the family level, 123 bacterial families were identified by shotgun sequencing, but only 55 families were detected by amplicon sequencing ^[16][43].

To study the biodiversity, activity, and biodeterioration of the microbial populations thriving in the Escoural Cave (Portugal), NGS analyses were performed in different areas of the cave, revealing a predominant distribution of Proteobacteria (58%), Actinobacteria (19%), Firmicutes (7%), Acidobacteria (4%), Bacteroidetes (2%), Gemmatimonadetes (2%), Planctomycetes (2%), and Chloroflexi (1%) ^[23][50]. Miller et al. (2022) ^[24][4] analyzed DNA samples isolated from the Atacama Desert (Chile). The first (preliminary) microbiological results confirmed the presence of halophilic microorganisms, such as Salinisphaera sp. and Haloparvum sp., as well as other genera commonly found in saline environments, including Acinetobacter and Pseudomonas.

The identification and study of microorganisms in caves allows us to understand which species are in the majority and what roles they play in the diversity of these environments. Culture-independent methodologies are very useful in this identification, as they allow for a screening of all microorganisms present in a sample. The culture-dependent methods are interesting in the individual study of microorganisms, allowing their isolation for studies of metabolism and the production of compounds with bioactivity.

Pristine environments preserve large numbers of unstudied bacterial strains with specific metabolic pathways ^[25][20]. The unique characteristics of these environments give microorganisms the capacity to develop specific metabolisms and to produce new bioactive compounds with potential activities, such as antimicrobial, antifungal, antiviral, and anticancer ^[26][9].

In recent years, several strains isolated from this ecosystems have been proposed as new species based on a polyphasic taxonomic approach comprising chemotaxonomic, phylogenetic morphological, and physiological characterization, such as strain PO-11 (Arthrobacter cavernae sp. nov.) isolated from Karst cave sediments, Guizhou Province, China ^[27][51], strain MM109 (Streptomyces lunaelactis sp. nov.) isolated from moonmilk deposit from the cave ‘Grotte des Collemboles’, Belgium ^[28][52], strain SG1 (Streptosporangium becharense sp. nov.) isolated in a Saharan soil sample collected from Algeria ^[29][3], strain AG31 (Arthrobacter psychrophenolicus sp. nov.) isolated in an Alpine ice cave, Austria ^[30][53], among many others. For instance, other bacteria showing important biological activities have also been proposed as new species, such as strains LM 036 and LM044 (Saccharothrix violacea sp. nov.), identified for the first time in a gold mine cave, Kongju, Korea. These strains exhibit antibacterial activity against Bacillus subtilis, Micrococcus luteus, Staphylococcus aureus, and Streptomyces murinus and exhibit antifungal activity against Aspergillus niger, Candida albicans, and Saccharomyces cerevisiae ^[31][54]. Additionally, in Pha Tup Cave Forest Park, Nan province, Thailand, Nonomuraea monospora sp. nov. ^[32][55] was identified for the first time, exhibiting antibacterial activity against Bacillus cereus, methicillin-resistant Staphylococcus aureus, and Paenibacillus lavae and exhibiting antitumoral activity against KB (human oral epidermoid carcinoma) and NCI-H187 (human small cell lung) cell lines ^[33][56]. The strain MBRL 251 (Streptomyces hundungensis sp. nov.) isolated from limestone deposit sites, Manipur, India ^[34][57], showed antibacterial activity against Curvularia oryzae, Fusarium oxysporum, Helminthosporum oryzae, Pyricularia oryzae, Rhizoctonia oryzae-sativae, and Rhizoctonia solani ^[35][58].

NGS and field emission scanning electron microscopy were often used for bacterial characterization. For example, in the identification and the functional and morphological characterization in the lava tube cave, Fuente de la Canaria Cave, La Palma Island, was revealed a predominant abundance of Proteobacteria (37–89%), followed by Actinobacteria, Acidobacteria, and Candidatus Rokubacteria. In this study, the ecological role of the microbial communities was also predicted using bioinformatics software that estimated the functional profile from the 16S rRNA gene data, which obtained and predicted the metabolic pathways and enzymes involved in nitrogen, sulfur, methane cycles, and CO₂ fixation ^[36][59].

There are many other studies that report microorganisms from pristine environments with important biological activity. Table 1 outlines the bioactivity of various bacterial species from pristine environments. Most of these studies describe bacteria of the phylum Proteobacteria, Bacteroidetes, Firmicutes, and mainly, Actinobacteria.

Proteobacteria, the largest and most phenotypically diverse phylum, are Gram-negative bacteria ^[37][60] and one of the most abundant phyla in caves ^[38][39][19,61].

Microorganisms of this type are phenotypically diverse, aerobic or facultatively anaerobic chemoorganotrophs, often producing carotenoids and/or flexirubin, which confer yellow or orange colony coloration ^[40][27].

The most members of Firmicutes Phylum are Gram-positive, with low content of guanine and cytosine (G + C) in their genome ^[41][42][62,63]. The phylum is phenotypically diverse ^[43][64] and is one of the least robust taxonomic groups, with the taxonomic hierarchy of this phylum remaining weak ^[41][62]. Cells may be spherical, may have straight, curved, and helical rods or filaments, and may be with or without flagella and with or without heat-resistant endospores ^[43][64]. Firmicutes are abundant in soil and aquatic environments, where they participate in the decomposition and recycling of organic matter ^[41][62].

Actinobacteria are Gram-positive filamentous bacteria with a high content of guanine and cytosine (G + C) in their genome. They grow by a combination of tip elongation and hyphal branching ^[44][65]. Most Actinobacteria are saprophytic soil-dwelling organisms that spend most of their life cycle as semidormant spores, especially under nutrient-limited conditions. They are more abundant in soil than in other environments, especially in alkaline and organic soils, where they form a significant part of the microbial population and are found both on the surface and more than 2 m underground. However, the family has adapted to very different ecological environments: actinomycetes are also found in fresh and salt water and in air ^[44][65]. Actinobacteria are particularly known for their potential to produce bioactive compounds, such as antibiotics, antimetabolites, and antitumor agents, with the genus Streptomyces having the greatest potential ^[45][15]. Approximately 45% of known bioactive compounds are secreted by Actinobacteria, of which 85% originate from the Streptomyces genus ^[46][66]. Actinobacteria are the most studied in the search for bioactive compounds. Members of Actinobacteria are reported to be a dominant microbial population in several cave ecosystems.

In Shuanghe Cave, China, the dominant phylum was Actinobacteria (42.13–48.03%) ^[3][30]. In Helmcken Falls cave in Canada (volcanic cave), 400 sample were collected from rocks, wall, sediment, and speleothems inside of cave. Isolates were screened, and most of the tested cave actinomycetes demonstrated antimicrobial activities. The results show that bacteria can be the source of novel compounds that provide precursors of new drugs to combat Gram-negative antibiotic resistant bacteria. This study also suggests a high possibility of finding new antimicrobial agents from previously unknown actinomycetes in volcanic cave habitats ^[47][11]. Yücel and Yamaç (2010) ^[48][67] isolated 180 actinomycete from Turkish karstic and tested for antimicrobial activity, where 27% exhibited activity only against Gram-negative bacteria and 33% against Gram-positive bacteria. Active cave isolate ratios against overall bacteria, yeasts, and filamentous fungi were determined as 15%, 19%, and 15%, respectively. In another case, in Belgium, different genera (Agromyces, Amycolatopsis, Kocuria, Micrococcus, Micromonospora, Nocardia, Streptomyces, and Rhodococcus) were isolated from cave milk deposits, and 87% of the bacteria showed activity against Gram-positive and 59% against Gram-negative bacteria ^[49][68].

Gonzalez-Pimentel et al. (2022) ^[15][42] reported the isolation of two strains of the genus Crossiella, likely representing a new species, isolated from the Altamira Cave, Spain. In vitro and in silico analyses showed the inhibition of pathogenic Gram-positive and Gram-negative bacteria and fungi, as well as the taxonomic distance of both strains from their closest relative, Crossiella cryophile.

While many studies have been conducted to identify bioactivity in cave bacteria, some of them do not identify the compounds that have activity. This can be due to a variety of reasons, such as the complexity of the microbial community in caves or the limitations of the analytical techniques used. However, identifying the specific compounds that have activity is crucial for further research and development of potential applications. It would be beneficial for future studies to focus on identifying and characterizing the bioactive compounds produced by cave bacteria, as this can lead to a better understanding of their potential uses and applications.