 |
MB 451 Microbial Diversity |
 Audio recording of this lecture |
Other methods for determining phylogenetic relationships
Although ssu-rRNA molecular phylogenetic analysis is generally the most useful method for determining the phylogeny of an organisms, it is not always the best method.
The biggest issues with resolution of rRNA-based trees are confidence in particular branching arrangments and resolution of close and distant relatives. We dealt already with how the reliability of the branching order in a tree can be asessed via bootstrap analysis. The resolution of both very close and very distant relatives is more of a problem. The branching arrangement of the deeepest branches of rRNA trees are still subject to some question, but more clearly, close relatives cannot be distinguished using rRNA sequences. There are two reasons for this: sometimes phenotypically-distinct closely-related "species" can have very similar or even identical rRNA sequences, and on the other hand often even the different rRNA operons in the same organism vary to some extent. In a very few cases, different rRNA operons in the same organism may be quite different, either because the gene conversion process has failing to keep them in line, or because the organisms is actually a hybrid of two species. This is part of the larger problem of how a prokaryotic species is defined.
If ssu-rRNA sequences are too close in sequence (greater than about 97%), trees generated by these sequences are not reliable. The ability to distinguish very closely related organisms is very important for medical and food microbiologists; for example, different strains of Staphylococcus aureus are very different in their pathogenicity, and must be carefully identified. Likewise, in food fermentations using natural microbes, such as in cheeze, different very closely-related strains make the difference between a perfect Wensleydale and sour milk.
So, while ssu-rRNA is the standard method for broad-scale phylogenetic analysis, other methods are needed for fine-scale analyses. Some of these methods are:
- DNA:DNA hybridization
- DNA base composition
- Serology
- Lipid profiling
- rRNA spacer sequence analysis
- Other gene phylogenetic analysis
- Ribotyping
- RFLP analysis
- Phenotypic markers
DNA:DNA hybridization is a method that was commonly used in the past. The extent that the genomic DNAs of 2 species will hybridize is a general measure of how much sequence similarity there is between the genomes, and therefore how closely related they are. This method is widely used to define bacterial species - in general, two organisms are considered to be the same species if the DNA:DNA hybridization is 70% or greater, or different species of the same genus if they have measurable hybridization less than 70%.
DNA base composition is also an older technology. There are several methods for determining the amount of A, G, C and T in an organisms DNA. Because every G is paired to a C, and every A to a T, the ratios of G and C are always the same, and likewise for A and T. And because the sum of all 4 is 100%, there is only one degree of freedom in base composition, and it is usually expressed as %G+C. Two closely related strains are considered to be in the same species if their DNA base composition (%G+C) are very close. This method is rapidly losing favor over more informative methods - it turns out that DNA base compositions can change very rapidly and unpredictably in evolution.
Serology is used primarily to identify very closely-related clinical isolates, usually different strains of a single species. This method uses antisera developed from various strains of Bacteria to identify which strain a new isolate is. For example, when Salmonella is isolated from a patient, a bank of antisera is used to determine which of the hundreds of serotypes that particular isolate is. This is an old but still widely used method, since the antisera are easy to make and the assay is very quick, easily automated, & reliable. The commonly-used ELISA is a serological method.
Lipid profiling is also a fairly quick & easy method. A culture is extracted with an organic solvent to collect the membrane lipds, & a sample of this is analyzed by gas chromatography. Certain types and ratios of lipids, especially fatty acids, are indicative of certain species- the profile from the unknown species is compared to a bank of standards and analyzed using tree-building methods. However, cultures have to be grown under strictly controlled conditions, since cells alter their membrane lipids in response to growth conditions.
rRNA spacer sequence analysis is essentially the same as the ssu-rRNA analysis we've talked so much about, but takes advantage of the fact that, in most organisms, the small and large subunit rRNAs are right next to each other in the genome, and so it is easy to PCR amplify & sequence this spacer using primers that hybridize to conserved sequences the end of the ssu-rRNA and the beginning of the lsu-rRNA. This small spacer sequence evolves very quickly, and can be used to distinguish different closely-related species or strains of the same species. This sequence is often used to analyze the relationships between animal species, using the spacer sequence from the mitochondrial DNA, which evolves much faster than the nuclear genome. The main disadvantage of this approch is that there are usually many copies of the rRNA gene cluster and the spacer sequences aren't usually the same in the different clusters. For example, E. coli has 7 rRNA operons, 3 of which have one spacer sequence and 4 have another sequence. So you have to be sure to compare the homologous spacers for the analysis to work. This can be done - these spacers often have tRNA genes in them and these can serve to indicate which spacer sequence corresponds to which.
The use of other genes for molecular phylogenetic analysis. In cases where ssu-rRNA sequences are too close to give reliable trees, it is common practice to use other sequences in the same way to create phylogenetic trees. For example, DNA polymerase genes, RNA ploymerase genes, etc. But it is also possible to use RNA-encoding genes that retain the advantages of RNA sequences (alignment by structure) that evolve faster than rRNA. In the bacterial world, the gene encoding the RNA subunit of RNase P has been used, for example, to distinguish strains of Legionella, Vibrio, and Chlamydiae.
Restriction fragment length polymorphism (RFLP) analysis is a method widely used by animal and plant geneticists, to determine paternity, to identify the source of forensic tissue samples, etc. In these sorts of analyses it is used primarily to distinguish individuals, but in microbial world can be used to differentiate very closely related strains of the same species. DNA from the organism (or blood, hair follicle, etc) is digested with restriction enzymes and separated by size on a gel. The gel is then hybridized with a set of oligonucleotide probes complementary to repeated sequences in the genome. Variation in the sequences results in differences in presence/absence of the restriction sites and the lengths of the fragments that hybridize to the probes. If the probes are carefully designed, the RFLP banding patterns are unique genetic fingerprints of the individual or strain. The patterns of bands are compared to other strains (or parents, suspects, etc) to test identity or specific genetic relationships. Although not strictly speaking RFLP, a related method is to use arbitrary primers or primers targeting specific variable genes to get banding profiles of the same sort as an RFLP
Ribotyping is a form of RFLP analysis using the rRNA operons. In a ribotype analysis, the probe is labeled rRNA from a type-strain of the species or genus rather than oligonucleotides. The advantages of ribotyping over other RFLP methods are that 1) since both highly-variable and conserved sequences are highlighted by the probe, it can be used to type strains over a broad phylogenetic range, and 2) the same process can be used for analysis of strains within just about any species.
Phenotypic markers are what most people think of when they compare different microorganisms. What carbon sources can it use? Is it motile or non-motile? Is it aerobic, anaerobic, or facultative? What is it's shape & size? What is it's optimal growth temperature? And so forth & so on. While gross phenotypic markers aren't always very useful in determining phylogeny, they are still perfectly viable markers for taxonomies, and in many cases, what the organism is like is as important (or more so) than it's phylogeny. After all, until recently that's all microbiologists could do! More recently, these are often done in microtiter plates - you can test 96 phenotypes on each plate:
Questions for thought:
- Can you predict some of the properties of ES-2 based on its phylogenetic placement?
- rRNA sequence divergence is a measure of phylogenetic relatedness, i.e. biological diversity. How is this related to what people usually think of as biological diversity?
- What other types of phylogenetic analysis can you think of?