**7. Polyphasic taxonomy**

genes are also shared between all the three domains. The genes required for core cellular functions are the ones which are necessary for survival of a cell and could have arisen from

The divergence of the organisms represents the differences in genetic sequences which could have become fixed in each group as they evolved. It is also postulated that earlier, HGT played a key role in transfer of genes between organisms early in the evolutionary history [6]. It occurs as a response to any change in the environment and provides for better adaptation ([9] and references therein). Subsequently, reproductive isolation could have prevented extensive

The bacteria and archaea have evolved along different lines though both are essentially prokaryotic. The archaea are considered to be the most primitive and are common inhabitants of the so-called extreme habitats (hot springs, deep sea hydrothermal vents, alkaline and acidic habitats). Though the bacteria and archaea share certain common features, the archaea also share similarities with eukaryotes which are further exemplified by the 16S rDNA-based

An array of diverse definitions have been proposed to describe microbial species. Currently, a polyphasic approach is used to define a microbial species using phenotypic and genotypic properties [1, 10]. Whenever a new taxon is proposed, it is essential that the organism be isolated in pure culture and its characteristic features be tested under standard conditions [11]. Whether an organism constitutes a member of a common species is primarily based on whether its DNA-DNA re-association values are more than 70% and melting temperature (ΔTm) is less than 5ºC, the experiments being performed under standard conditions [12]. All the strains within a species must show similar phenotypes. A designated type strain of a species consti‐ tutes the reference specimen for that species [13]. A species description must preferably be based on the characteristics of more than one type strain. To be assigned a different species name, members must show at least one and is governed by the [12]. If the 16S rDNA sequences of organisms are ≤ 98.7% or ≤ 97% identical, they are members of different species. This is considered even in the absence of DNA-DNA hybridization experiments since this level of divergence in 16S rDNA sequences constitutes less than 70% DNA-DNA similarity [14]. Uncultured microbes cannot be assigned to a definite species since their phenotype is not known; however, they can be assigned a '*Candidatus*' designation provided their 16S rRNA sequence subscribes to the principles of identity with known species [15]. A concept applying to a taxon lower than that of the strain is the ecotype – those microorganisms that occupy an ecological niche and are adapted to the conditions of that niche [16]. It is important to remember

exchange of genes, though it continues to occur amongst prokaryotes.

**5. The prokaryotic microbes — Bacteria and archaea**

the common ancestor.

54 Biodiversity - The Dynamic Balance of the Planet

phylogenetic analyses.

**6. The species concept**

While the species is accepted as the basic unit of taxonomy (see Section 6), sub-species, strains and ecotypes occupy lower distinctive taxonomic levels for certain groups of organisms and are not mandatory for all. When classifying a new taxon, it is essential to describe phenotypic, genotypic and phylogenetic information as accurately as possible. This constitutes the polyphasic approach of taxonomy [1] and is shown in Figure 2. The phenotypic information comes from the colony characteristics, cell type, cell wall-type, pigmentation patterns, proteins and other chemotaxonomic markers while genotypic features are derived from the nucleic acids (DNA / RNA). Phylogenetic information is obtained from studying sequence similarities of the 16S rRNA or 23 S rRNA genes in case of bacteria and 18S rRNA in case of fungi. Many types of molecules are used for delineating and describing a taxon; some are mandatory (16S rRNA genes, phenotypes, chemotaxonomy) while others are optional (amino-acid sequencing of certain protein products, DNA-DNA hybridization), unless required for appropriate description.

(A or B), mode of cross-linking (whether it is directly linked or via interpeptide bridge and with amino acids in the bridge), and the composition of amino acids (especially the diami‐ noacid) of the side chain [18]. While the mode of cross-linkage can vary within a species and also between strains, the amino acid composition is common to all species within a genus. In archaea, pseudomurein is present where N-acetyl muramic acid is replaced by N-acetyl

Modern Taxonomy for Microbial Diversity http://dx.doi.org/10.5772/57407 57

Different types of lipids are present in bacterial cells. Polar lipids are present in the lipid bilayer of the cytoplasmic membrane. The diversity of polar lipids is known to be large and many are yet to be structurally elucidated. While in archaea, polar lipids are of types phospholipids, aminophospholipids, glycolipids and phosphoglycolipids, in bacteria, apart from the ones seen in archaea, there are also lipids derived from amino acids, capnines, sphingolipids (glycol or phosphosphingolipids) and hopanoids [18]. In Gram-negative bacteria, lipopolysaccharides are present in the outer membranes. The type of sugar present and the fatty acid type, the linkage of the fatty acid to the sugar (amide or ester linkage) provide information on charac‐ teristic of the cell [23−25]. However, determination of lipopolysaccharides is not routinely used in recent times. The total cellular fatty acids are extracted, esterified and the methyl ester content is analyzed by gas chromatography. Under standard conditions, this provides a reliable estimate of taxonomy up to genus and sometimes species level. The technique has been automated and the Sherlock MIS system (MIDI Inc.) has developed an extensive database.

Though incomplete, it still is the most widely used system in recent times [18].

The respiratory isoprenoid quinones are components that occur in cytoplasmic membranes of prokaryotes (archaea as well as bacteria). The naphthoquinones (with sub-types phylloqui‐ none and menaquinone) and benzoquinones (ubiquinones, rhodoquinones and plastoqui‐ nones) are the major types of quinones. The variability they depict in their side chains in terms of length (5-15 isoprenoid units known till date), degree and position of saturation are of taxonomic significance and help in characterization to various levels of genus and species [26]. These features generally also mirror the 16S rDNA groupings. Isoprenoid ether-linked side chains are present in members of the archaea. They are of different types such as diethers, hydroxylated diethers, macrocyclic diethers, tetraethers, and polyol derivatives of the tetraether [18]. Non-isoprenoid based-ether-linked lipids are present in bacteria and can be straight chain or with simple branched side chains or with mono-unsaturated derivatives [18].

Other than the principal diagnostic methods described above, other techniques used to lesser

**a.** Whole cell protein analyses, wherein protein is extracted from the cells and analysed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). This can help in

levels or for comparison between species or strains comprise of the following:

talosuronic acid [22].

*7.1.4. Fatty acid analyses*

*7.1.5. Isoprenoid quinones*

*7.1.6. Other diagnostic methods*

#### **7.1. Phenotypic techniques**

The phenotypic methods are all those that do not include the DNA/RNA sequencing or their typing methods. Study of morphological characteristics and chemotaxonomic profiles is broadly associated with phenotypic characterization.

#### *7.1.1. Classical: Colony characteristics, biochemical and physiological analyses*

The phenotypic features are the foundation for description of taxa. The morphological, biochemical and physiological characteristics provide in-depth information on a taxon. The morphology can include the colony characteristics (colour, shape, pigmentation, production of slime etc.). Further, the features of the cell are described as to shape, size, Gram reaction, extracellular material like capsule, presence of endospores, flagella presence and location, motility and inclusion bodies. Light microscopy is generally used to describe the broad cell features; however electron microscopy is recommended for high resolution images [18]. The biochemical and physiological features describe growth of the organism at different ranges of temperature, pH, salinity and atmospheric conditions, growth in presence of anti-microbial agents, production of various enzymes and growth in presence of different sole carbon and nitrogen sources [1]. These tests have to be carried out using standardized procedures to obtain results that are reproducible within and between laboratories [19].

#### *7.1.2. Numerical taxonomy*

Analyses of huge volumes of phenotypic data to derive meaningful relationships amongst a large number of microorganisms can be carried out using computer programs [20]. This system of analyses is called numerical taxonomy. Giving numerical weightage to each trait is followed by analyses of the data by the computer programs generating data matrices between each pair of isolates according to the degree of similarity. Based on the similarity data, cluster analyses are carried out (based on different algorithms) and dendrograms ('trees') are generated showing the overall pattern of similarity/dissimilarity amongst the various organisms being studied. While 16S rDNA sequences have garnered attention in recent times as sole means of bringing out the uniqueness of a species; numerical taxonomy (based on phenotypic traits of a large number of species) compares favourably with that of genotypic data and, indeed, is in alignment with the latter [1].

#### *7.1.3. Cell wall composition*

The peptidoglycan component of cell walls of bacteria does not provide much information except for classifying into Gram-positive, Gram-negative and acid-fast bacterial types. However, those in Gram-positive cells contain different types of peptidoglycan depending on the genus or species [21]. The peptidoglycan structure can be analysed by determining its type (A or B), mode of cross-linking (whether it is directly linked or via interpeptide bridge and with amino acids in the bridge), and the composition of amino acids (especially the diami‐ noacid) of the side chain [18]. While the mode of cross-linkage can vary within a species and also between strains, the amino acid composition is common to all species within a genus. In archaea, pseudomurein is present where N-acetyl muramic acid is replaced by N-acetyl talosuronic acid [22].

### *7.1.4. Fatty acid analyses*

rRNA genes, phenotypes, chemotaxonomy) while others are optional (amino-acid sequencing of certain protein products, DNA-DNA hybridization), unless required for appropriate

The phenotypic methods are all those that do not include the DNA/RNA sequencing or their typing methods. Study of morphological characteristics and chemotaxonomic profiles is

The phenotypic features are the foundation for description of taxa. The morphological, biochemical and physiological characteristics provide in-depth information on a taxon. The morphology can include the colony characteristics (colour, shape, pigmentation, production of slime etc.). Further, the features of the cell are described as to shape, size, Gram reaction, extracellular material like capsule, presence of endospores, flagella presence and location, motility and inclusion bodies. Light microscopy is generally used to describe the broad cell features; however electron microscopy is recommended for high resolution images [18]. The biochemical and physiological features describe growth of the organism at different ranges of temperature, pH, salinity and atmospheric conditions, growth in presence of anti-microbial agents, production of various enzymes and growth in presence of different sole carbon and nitrogen sources [1]. These tests have to be carried out using standardized procedures to obtain

Analyses of huge volumes of phenotypic data to derive meaningful relationships amongst a large number of microorganisms can be carried out using computer programs [20]. This system of analyses is called numerical taxonomy. Giving numerical weightage to each trait is followed by analyses of the data by the computer programs generating data matrices between each pair of isolates according to the degree of similarity. Based on the similarity data, cluster analyses are carried out (based on different algorithms) and dendrograms ('trees') are generated showing the overall pattern of similarity/dissimilarity amongst the various organisms being studied. While 16S rDNA sequences have garnered attention in recent times as sole means of bringing out the uniqueness of a species; numerical taxonomy (based on phenotypic traits of a large number of species) compares favourably with that of genotypic data and, indeed, is in

The peptidoglycan component of cell walls of bacteria does not provide much information except for classifying into Gram-positive, Gram-negative and acid-fast bacterial types. However, those in Gram-positive cells contain different types of peptidoglycan depending on the genus or species [21]. The peptidoglycan structure can be analysed by determining its type

description.

**7.1. Phenotypic techniques**

56 Biodiversity - The Dynamic Balance of the Planet

*7.1.2. Numerical taxonomy*

alignment with the latter [1].

*7.1.3. Cell wall composition*

broadly associated with phenotypic characterization.

*7.1.1. Classical: Colony characteristics, biochemical and physiological analyses*

results that are reproducible within and between laboratories [19].

Different types of lipids are present in bacterial cells. Polar lipids are present in the lipid bilayer of the cytoplasmic membrane. The diversity of polar lipids is known to be large and many are yet to be structurally elucidated. While in archaea, polar lipids are of types phospholipids, aminophospholipids, glycolipids and phosphoglycolipids, in bacteria, apart from the ones seen in archaea, there are also lipids derived from amino acids, capnines, sphingolipids (glycol or phosphosphingolipids) and hopanoids [18]. In Gram-negative bacteria, lipopolysaccharides are present in the outer membranes. The type of sugar present and the fatty acid type, the linkage of the fatty acid to the sugar (amide or ester linkage) provide information on charac‐ teristic of the cell [23−25]. However, determination of lipopolysaccharides is not routinely used in recent times. The total cellular fatty acids are extracted, esterified and the methyl ester content is analyzed by gas chromatography. Under standard conditions, this provides a reliable estimate of taxonomy up to genus and sometimes species level. The technique has been automated and the Sherlock MIS system (MIDI Inc.) has developed an extensive database. Though incomplete, it still is the most widely used system in recent times [18].

#### *7.1.5. Isoprenoid quinones*

The respiratory isoprenoid quinones are components that occur in cytoplasmic membranes of prokaryotes (archaea as well as bacteria). The naphthoquinones (with sub-types phylloqui‐ none and menaquinone) and benzoquinones (ubiquinones, rhodoquinones and plastoqui‐ nones) are the major types of quinones. The variability they depict in their side chains in terms of length (5-15 isoprenoid units known till date), degree and position of saturation are of taxonomic significance and help in characterization to various levels of genus and species [26]. These features generally also mirror the 16S rDNA groupings. Isoprenoid ether-linked side chains are present in members of the archaea. They are of different types such as diethers, hydroxylated diethers, macrocyclic diethers, tetraethers, and polyol derivatives of the tetraether [18]. Non-isoprenoid based-ether-linked lipids are present in bacteria and can be straight chain or with simple branched side chains or with mono-unsaturated derivatives [18].

#### *7.1.6. Other diagnostic methods*

Other than the principal diagnostic methods described above, other techniques used to lesser levels or for comparison between species or strains comprise of the following:

**a.** Whole cell protein analyses, wherein protein is extracted from the cells and analysed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE). This can help in comparison between related strains ([1] and references therein) and also shows congru‐ ency with DNA-DNA hybridization results [27]. However, since the identity of the protein bands is not known, this technique suffers from a drawback that is not associated with fatty acid analysis.

[37], in combination with a data classification approach [38] and can also provide information

Modern Taxonomy for Microbial Diversity http://dx.doi.org/10.5772/57407 59

Modern taxonomy has been influenced by genetic methods and indeed, much of the classifi‐ cation and identification is predicated on specific gene sequences. All the techniques involving

The technique, which is very nearly a gold standard for taxonomic purposes today, is se‐ quencing of the 16S rRNA gene of bacteria. The 23S rRNA gene sequence is also considered in many studies but lack of comprehensive databases for comparison is a drawback. Since the 16S rRNA is present in all bacteria, is functionally constant and is composed of conserved and variable regions, it has consistently served as a good taxonomic marker for deriving taxonomic relationships [1]. While it has proved to be the foundation for modern taxonomy, there are certain caveats and it has to be considered along with other techniques for formal identification purposes, especially at the species level. Generally universal sets of primers are used to amplify the 16S rRNA gene product gene (~1.5 kB). Subsequently, the amplified product is sequenced and quality of the sequence checked. The sequence is then aligned against high quality sequence data from curated databases [ARB (www.arb-home.de), RDP (http:// rdp.cme.msu.edu/), SILVA (www.arb-silva.de) and LTP (www.arb-silva.de/projects/living‐ tree/)]. Multiple alignment programs such as CLUSTAL\_X, CLUSTAL W, CLUSTAL X2, CLUSTAL W2, MEGA, T-COFFEE, MUSCLE) can also be used with manual editing [18].

As mentioned earlier, it has been shown that < 97% similarity of two 16S rDNA sequences implies a different species ([18] and references therein). This cut-off value is generally consid‐ ered for ecological studies [41]. For actual taxonomic studies, a 98.5% cut-off value is consid‐ ered [42]. However, the values should be based on high-quality and almost full-length sequences. When the similarity values are ~95%, other methods must also be included to justify the creation of a new genus. The descriptions must also provide information on the differences between the potentially new and existing genera. Subsequent to alignment, phylogenetic trees or dendrograms have to be constructed to reveal the taxonomic position of an organism. Different treeing algorithms such as maximum-parsimony, maximum-likelihood methods are preferred for evaluation of taxonomic position. Inclusion of type strain or type species is

Determination of moles percent guanosine and cytosine constitutes a classical method of establishment of genomic content. This is now being used along with other genotyping methods to establish taxonomic position of an organism [1]. Within species, the G+C content ranges within 3% and within genera 10% [42]. Overall, the G+C content ranges from 24-76%

about the Gram-type of a bacterium [39] as well as relate to moles G + C content [40].

**7.2. Genotypic techniques**

*7.2.1. 16S rDNA-based analyses*

essential.

in bacteria.

*7.2.2. DNA base content*

DNA or RNA fall under genotypic methods.


#### *7.1.7. Advanced spectroscopy and spectrometric methods*

Many advanced analytical techniques such as Pyrolysis Mass Spectrometry, Fourier Transform Infrared spectroscopy and UV Resonance Raman Spectroscopy have been employed to determine chemical composition of bacterial cells, primarily the bioactive metabolites from drug discovery point of view and relate it to characteristics of microbes from which the metabolites are obtained [30, 31]. Pyrolysis Mass Spectrometry is a high-resolution technique, wherein microbial colonies are carefully picked and placed onto a iron-nickel foil, vacuum desiccated, heated rapidly and the pyrolysate bombarded with low-energy electrons. The ionized fragments are separated on the basis of their mass to charge ratio (*m/z*), detected and amplified by an electron multiplier [30]. Metabolites with *m/z* ratio of 51-200 constitute degradation products useful for taxonomical discrimination, since those with *m/z* less than 50 are produced by most biological materials and those with *m/z* ratio greater than 200 are not useful in discriminating taxons. The multivariate data is further analysed by Principal Components Analysis (PCA) to understand the variance while reducing the dimensionality of the data.

Fourier Transform Infrared spectroscopy (FTIR) is a simple and cost-efficient method and has been applied to identify discriminative features of strains. Most cellular components (fatty acids, proteins, carbohydrates and nucleic acids) can be analysed by this method to reveal strain-specific features [32]. Five IR spectral regions or 'windows': W1 (3000–2800 cm–1) for fatty acids, W2 (1700–1500 cm–1) for amide I and II bands of proteins and peptides, W3 (1500– 1200 cm–1) for a mixed region of fatty acid bending vibrations, proteins, and phosphatecarrying compounds, W4 (1200–900 cm–1) for carbohydrates of cell walls and W5 (900–700 cm–1) which is the 'fingerprint region' with unique absorbances specific for different taxa [33]. The differences in spectra are resolved using multivariate tools such as cluster analysis, discriminant analysis etc. [34].

UV Resonance Raman Spectroscopy (UVRR) uses a frequency of Raman spectra when biological materials are not subjected to fluorescent background while using IR or visible excitation [35]. Where an IR spectroscopy needs hundreds of cells, Raman spectroscopy does not require this [33, 36]. It can also be used with single cells, as demonstrated by Rösch et al. [37], in combination with a data classification approach [38] and can also provide information about the Gram-type of a bacterium [39] as well as relate to moles G + C content [40].

#### **7.2. Genotypic techniques**

comparison between related strains ([1] and references therein) and also shows congru‐ ency with DNA-DNA hybridization results [27]. However, since the identity of the protein bands is not known, this technique suffers from a drawback that is not associated with

**b.** Polyamines are a group of compounds in the cytoplasm that provide stability to the DNA and maintain osmolarity in the cell. They are useful to distinguish above genus level and

**c.** Cytochromes are associated with the cytoplasmic membrane and are involved in respi‐ ratory and photosynthetic electron transfer. They are 'heme' proteins with a 'heme' prosthetic group attached to a protein. These are not used alone in identification since

Many advanced analytical techniques such as Pyrolysis Mass Spectrometry, Fourier Transform Infrared spectroscopy and UV Resonance Raman Spectroscopy have been employed to determine chemical composition of bacterial cells, primarily the bioactive metabolites from drug discovery point of view and relate it to characteristics of microbes from which the metabolites are obtained [30, 31]. Pyrolysis Mass Spectrometry is a high-resolution technique, wherein microbial colonies are carefully picked and placed onto a iron-nickel foil, vacuum desiccated, heated rapidly and the pyrolysate bombarded with low-energy electrons. The ionized fragments are separated on the basis of their mass to charge ratio (*m/z*), detected and amplified by an electron multiplier [30]. Metabolites with *m/z* ratio of 51-200 constitute degradation products useful for taxonomical discrimination, since those with *m/z* less than 50 are produced by most biological materials and those with *m/z* ratio greater than 200 are not useful in discriminating taxons. The multivariate data is further analysed by Principal Components Analysis (PCA) to understand the variance while reducing the dimensionality

Fourier Transform Infrared spectroscopy (FTIR) is a simple and cost-efficient method and has been applied to identify discriminative features of strains. Most cellular components (fatty acids, proteins, carbohydrates and nucleic acids) can be analysed by this method to reveal strain-specific features [32]. Five IR spectral regions or 'windows': W1 (3000–2800 cm–1) for fatty acids, W2 (1700–1500 cm–1) for amide I and II bands of proteins and peptides, W3 (1500– 1200 cm–1) for a mixed region of fatty acid bending vibrations, proteins, and phosphatecarrying compounds, W4 (1200–900 cm–1) for carbohydrates of cell walls and W5 (900–700 cm–1) which is the 'fingerprint region' with unique absorbances specific for different taxa [33]. The differences in spectra are resolved using multivariate tools such as cluster analysis,

UV Resonance Raman Spectroscopy (UVRR) uses a frequency of Raman spectra when biological materials are not subjected to fluorescent background while using IR or visible excitation [35]. Where an IR spectroscopy needs hundreds of cells, Raman spectroscopy does not require this [33, 36]. It can also be used with single cells, as demonstrated by Rösch et al.

fatty acid analysis.

of the data.

discriminant analysis etc. [34].

at species level too [28].

58 Biodiversity - The Dynamic Balance of the Planet

there are limited types of distinct cytochromes [29].

*7.1.7. Advanced spectroscopy and spectrometric methods*

Modern taxonomy has been influenced by genetic methods and indeed, much of the classifi‐ cation and identification is predicated on specific gene sequences. All the techniques involving DNA or RNA fall under genotypic methods.

#### *7.2.1. 16S rDNA-based analyses*

The technique, which is very nearly a gold standard for taxonomic purposes today, is se‐ quencing of the 16S rRNA gene of bacteria. The 23S rRNA gene sequence is also considered in many studies but lack of comprehensive databases for comparison is a drawback. Since the 16S rRNA is present in all bacteria, is functionally constant and is composed of conserved and variable regions, it has consistently served as a good taxonomic marker for deriving taxonomic relationships [1]. While it has proved to be the foundation for modern taxonomy, there are certain caveats and it has to be considered along with other techniques for formal identification purposes, especially at the species level. Generally universal sets of primers are used to amplify the 16S rRNA gene product gene (~1.5 kB). Subsequently, the amplified product is sequenced and quality of the sequence checked. The sequence is then aligned against high quality sequence data from curated databases [ARB (www.arb-home.de), RDP (http:// rdp.cme.msu.edu/), SILVA (www.arb-silva.de) and LTP (www.arb-silva.de/projects/living‐ tree/)]. Multiple alignment programs such as CLUSTAL\_X, CLUSTAL W, CLUSTAL X2, CLUSTAL W2, MEGA, T-COFFEE, MUSCLE) can also be used with manual editing [18].

As mentioned earlier, it has been shown that < 97% similarity of two 16S rDNA sequences implies a different species ([18] and references therein). This cut-off value is generally consid‐ ered for ecological studies [41]. For actual taxonomic studies, a 98.5% cut-off value is consid‐ ered [42]. However, the values should be based on high-quality and almost full-length sequences. When the similarity values are ~95%, other methods must also be included to justify the creation of a new genus. The descriptions must also provide information on the differences between the potentially new and existing genera. Subsequent to alignment, phylogenetic trees or dendrograms have to be constructed to reveal the taxonomic position of an organism. Different treeing algorithms such as maximum-parsimony, maximum-likelihood methods are preferred for evaluation of taxonomic position. Inclusion of type strain or type species is essential.

#### *7.2.2. DNA base content*

Determination of moles percent guanosine and cytosine constitutes a classical method of establishment of genomic content. This is now being used along with other genotyping methods to establish taxonomic position of an organism [1]. Within species, the G+C content ranges within 3% and within genera 10% [42]. Overall, the G+C content ranges from 24-76% in bacteria.

#### *7.2.3. DNA-DNA hybridization*

This method is an indirect measurement of sequence similarity between genomes. A cut-off value of 70% similarity is considered for establishment of species [1]. However, the method has to be reproducible between laboratories and performed under standardized conditions, which is often a drawback. Hence it is applied only where 16S rRNA gene sequences show similarity values above 98%. There have been reports where 16S rRNA gene sequence has shown 99% similarity and yet DNA-DNA hybridization values have been 60% or less. Hence, this method has to be used with caution and performed under highly standardized conditions. show that 95% ANI corresponds roughly to 70% DNA-DNA similarity values [49]. However,

Modern Taxonomy for Microbial Diversity http://dx.doi.org/10.5772/57407 61

Fungi are important from industrial point of view as well as the increasing numbers of pathogens that are arising. Primarily, fungi are classified on the basis of appearance, the structure appearing above ground. The below-ground vegetative structures are difficult to identify [50]. The focus is on the asexual stage. The concept of species is somewhat difficult to interpret in the case of fungi since sexual mating does not occur in all fungi; meiosis occurs only in sexual fungi and even where mating occurs, the product of fusion could either be sterile or fertile. The biological concept of species can therefore be applied to sexual fungi. In the case of asexual fungi, similarities in characteristics provide a system for classification. Modern developments such as analyses of DNA by sequencing have brought about a paradigm shift in fungal taxonomy. The phylogenetic species concept is being favoured over the other definitions, especially where asexual fungi are considered. The evolutionary relationships amongst fungi have not been well-delineated. Traditionally, fungal phylogeny has been classified using morphology (involving primarily the fruiting body), cell wall composition [51], cell ultra-structure [52], cytology [53] and metabolism [54] and even based on the study of fossils [55]. Nomenclature also creates confusion since fungi can exist in sexual and asexual stages and can develop at different times in different substrates and relationship between the

Currently, modern developments underscoring the importance of phylogeny in bacterial classification has also been used for fungal taxonomy. Molecular genotyping methods such as restriction analysis of internal transcribed spacer (ITS) region, 18S rDNA and RFLP are being used to classify fungi [50]. However, the datasets are not extensive enough to permit effective identification [56]. Hibbett et al. [57] proposed a phylogenetic-based classification for fungi with 195 taxa. The 5S, 5.8S, 18S and 25S rDNAs are considered for phylogenetic studies. Of these, the 18S rDNA has been more extensively used for filamentous fungi [58]. The D1 and D2 variable regions of 25S rDNA are used for yeasts. Limited datasets are available where both 18S and 25 S rDNAs have been sequenced leading to difficulties in comparison [59]. The 25S rDNA allows for comparison to species level [60] while 18S rDNA has to be nearly completely sequenced to obtain pertinent information. The ITS region is suitable to reveal close relation‐

Metagenomics has emerged as a promising field of interest, where identification of uncultured microorganisms is attempted. Since 99% of the microbial population is considered to be uncultivable, metagenomics assumes importance [63]. Next generation Sequencing (NGS) has

ships [61]. The 5S sequence provides information suitable for order level [62].

more datasets are required to implement this method.

**8. The eukaryotic microbes — Fungi**

two states has to be established first.

**9. Uncultured microorganisms**

#### *7.2.4. Other genotyping methods*

Earlier, sub-typing was done on the basis of biochemical profile (biotyping), serological profile (serotyping), phase susceptibility (phage typing) or antibiotic susceptibility. But currently DNA-typing methods are preferred due to their reproducibility, ease of performance and high level of discrimination between strains [1]. Genotyping methods such as Restriction Fragment Length Polymorphism (RFLP), Randomly Amplified Polymorphic DNA (RAPD), Amplified Fragment Length Polymorphism (AFLP), Amplified Ribosomal DNA Restriction Analysis (ARDRA), Repetitive Element-Polymerase Chain Reaction (REP-PCR), Ribotyping and Multi Locus Sequence Analyses (MLSA) are some of the newer methods to characterize a taxon.

RFLP was one of the earliest methods to be used and consisted of extraction of whole-genome DNA, restriction digestion using specific restriction enzymes and visualization of the DNA bands using gel electrophoresis. However, complex patterns can be generated, making comparison difficult. AFLP makes use of restriction analyses followed by PCR amplification to select for particular DNA fragments. Two restriction enzymes are used and fragments are amplified which can be genus and species-specific. Thus the method,though tedious, can be used for identification and typing purposes [1]. A PCR-based methodology makes use of random oligonucleotide primers 10 bases in length (RAPD), followed by amplification under specified conditions and analyses of the bands for similarity in size after gel electrophoresis. Ribotyping is another technique in which total genomic DNA is extracted, followed by restriction digestion and separation by electrophoresis [43]. Subsequently, the fragements are hybridized with a radiolabeled ribosomal operon probe (genes for 16S, 23S, tRNA and 5S rRNA) and analysed for presence of bands by autoradiography [44]. A simpler technique is amplifying the 16S rDNA (with or without spacer regions) using universal primers and restriction analysis (ARDRA). It can be used to discriminate at species level ([45]; [1] and references therein). Ribotyping and ARDRA have been shown to produce reproducible and congruent results in *Lactobacillus sp.* [46]. Consensus sequences complementary to repetitive sequences in the genomes of bacteria are used as primers, PCR-amplified and visualized as distinctive bands (REP-PCR). This method is discriminatory and rapid up to species or intraspecific level but is based on a library-based approach [47]. MLSA focuses on sequencing of housekeeping (6-8 protein-coding) genes and phylogenetic analyses of the same. It is a new method for characterizing a species [48]; however, databases are limited for realising the fullextent of utility of this method. Average Nucleotide Identity (ANI) of all orthologous genes in complete genome sequences has been proposed as a method to define species. Limited studies show that 95% ANI corresponds roughly to 70% DNA-DNA similarity values [49]. However, more datasets are required to implement this method.

### **8. The eukaryotic microbes — Fungi**

*7.2.3. DNA-DNA hybridization*

60 Biodiversity - The Dynamic Balance of the Planet

*7.2.4. Other genotyping methods*

This method is an indirect measurement of sequence similarity between genomes. A cut-off value of 70% similarity is considered for establishment of species [1]. However, the method has to be reproducible between laboratories and performed under standardized conditions, which is often a drawback. Hence it is applied only where 16S rRNA gene sequences show similarity values above 98%. There have been reports where 16S rRNA gene sequence has shown 99% similarity and yet DNA-DNA hybridization values have been 60% or less. Hence, this method has to be used with caution and performed under highly standardized conditions.

Earlier, sub-typing was done on the basis of biochemical profile (biotyping), serological profile (serotyping), phase susceptibility (phage typing) or antibiotic susceptibility. But currently DNA-typing methods are preferred due to their reproducibility, ease of performance and high level of discrimination between strains [1]. Genotyping methods such as Restriction Fragment Length Polymorphism (RFLP), Randomly Amplified Polymorphic DNA (RAPD), Amplified Fragment Length Polymorphism (AFLP), Amplified Ribosomal DNA Restriction Analysis (ARDRA), Repetitive Element-Polymerase Chain Reaction (REP-PCR), Ribotyping and Multi Locus Sequence Analyses (MLSA) are some of the newer methods to characterize a taxon.

RFLP was one of the earliest methods to be used and consisted of extraction of whole-genome DNA, restriction digestion using specific restriction enzymes and visualization of the DNA bands using gel electrophoresis. However, complex patterns can be generated, making comparison difficult. AFLP makes use of restriction analyses followed by PCR amplification to select for particular DNA fragments. Two restriction enzymes are used and fragments are amplified which can be genus and species-specific. Thus the method,though tedious, can be used for identification and typing purposes [1]. A PCR-based methodology makes use of random oligonucleotide primers 10 bases in length (RAPD), followed by amplification under specified conditions and analyses of the bands for similarity in size after gel electrophoresis. Ribotyping is another technique in which total genomic DNA is extracted, followed by restriction digestion and separation by electrophoresis [43]. Subsequently, the fragements are hybridized with a radiolabeled ribosomal operon probe (genes for 16S, 23S, tRNA and 5S rRNA) and analysed for presence of bands by autoradiography [44]. A simpler technique is amplifying the 16S rDNA (with or without spacer regions) using universal primers and restriction analysis (ARDRA). It can be used to discriminate at species level ([45]; [1] and references therein). Ribotyping and ARDRA have been shown to produce reproducible and congruent results in *Lactobacillus sp.* [46]. Consensus sequences complementary to repetitive sequences in the genomes of bacteria are used as primers, PCR-amplified and visualized as distinctive bands (REP-PCR). This method is discriminatory and rapid up to species or intraspecific level but is based on a library-based approach [47]. MLSA focuses on sequencing of housekeeping (6-8 protein-coding) genes and phylogenetic analyses of the same. It is a new method for characterizing a species [48]; however, databases are limited for realising the fullextent of utility of this method. Average Nucleotide Identity (ANI) of all orthologous genes in complete genome sequences has been proposed as a method to define species. Limited studies Fungi are important from industrial point of view as well as the increasing numbers of pathogens that are arising. Primarily, fungi are classified on the basis of appearance, the structure appearing above ground. The below-ground vegetative structures are difficult to identify [50]. The focus is on the asexual stage. The concept of species is somewhat difficult to interpret in the case of fungi since sexual mating does not occur in all fungi; meiosis occurs only in sexual fungi and even where mating occurs, the product of fusion could either be sterile or fertile. The biological concept of species can therefore be applied to sexual fungi. In the case of asexual fungi, similarities in characteristics provide a system for classification. Modern developments such as analyses of DNA by sequencing have brought about a paradigm shift in fungal taxonomy. The phylogenetic species concept is being favoured over the other definitions, especially where asexual fungi are considered. The evolutionary relationships amongst fungi have not been well-delineated. Traditionally, fungal phylogeny has been classified using morphology (involving primarily the fruiting body), cell wall composition [51], cell ultra-structure [52], cytology [53] and metabolism [54] and even based on the study of fossils [55]. Nomenclature also creates confusion since fungi can exist in sexual and asexual stages and can develop at different times in different substrates and relationship between the two states has to be established first.

Currently, modern developments underscoring the importance of phylogeny in bacterial classification has also been used for fungal taxonomy. Molecular genotyping methods such as restriction analysis of internal transcribed spacer (ITS) region, 18S rDNA and RFLP are being used to classify fungi [50]. However, the datasets are not extensive enough to permit effective identification [56]. Hibbett et al. [57] proposed a phylogenetic-based classification for fungi with 195 taxa. The 5S, 5.8S, 18S and 25S rDNAs are considered for phylogenetic studies. Of these, the 18S rDNA has been more extensively used for filamentous fungi [58]. The D1 and D2 variable regions of 25S rDNA are used for yeasts. Limited datasets are available where both 18S and 25 S rDNAs have been sequenced leading to difficulties in comparison [59]. The 25S rDNA allows for comparison to species level [60] while 18S rDNA has to be nearly completely sequenced to obtain pertinent information. The ITS region is suitable to reveal close relation‐ ships [61]. The 5S sequence provides information suitable for order level [62].
