Profiling microbial communities
with T-RFLP
(terminal restriction fragment length polymorphism)
Anne Fahy
b- Ti 01
a- Ti 02
c- Pi 02
d- Pi 01
e- Ud 02
f- Ud 01
why microbial ecology?
huge metabolic diversity
“higher” organisms dependent on microbial activities
applications: bioremediation and natural attenuation of pollutants in the environment
Microbial organisms occupy a peculiar place in the human view of life. Microbes receive little attention in our general texts of biology. They are
largely ignored by most professional biologists and are virtually unknown to the public except in the context of disease and rot. Yet, the workings of
the biosphere depend absolutely on the activities of the microbial world. (Pace, 1997)
between 0.001 and 1 % microorganisms are culturable (Amann 1995)
microbial communities are complex:
huge diversity
close interactions between organisms
highly dynamic
why culture-independent?
phylogenetic tree based on 16S rRNA : major phyla of the domain Bacteria (Rappé & Giovannoni, 2003)black = 12 original phyla described by Woese, 1987 white = 14 phyla with isolated representatives grey = 26 candidate phyla with no known isolates
why 16S rRNA as a phylogenetic marker ?
secondary structure of the Escherichia coli 16S rRNA molecule (Van de Peer, et al. 1996). colours variability between organisms: pink = highly conserved red = least conservedgrey = unaligned
- protein translation : universal
- no horizontal transfer (caveat: Wang & Zhang, 2000)
- convenient length : 1500 bp
- highly conserved regions as well as species-specific regions
- large databases (EMBL, NCBI, DDJB)
other phylogenetic markers
- proteins; difficult to identify homologous proteins (Demoulin, 1979)
- historically: also 5S, 23S rRNA
- ribosomal intergenic spacer: 16S – 23s
- 18S for Eukaryotes
T-RFLP(Terminal Restriction Fragment Length Polymorphism)
1 extraction of community DNA or RNA from environmental sample (need RT-PCR step with RNA)
2 PCR amplification of 16S rRNA gene with fluorescent primers
3 digestion of amplicons with restriction enzyme
4 detection and sizing of labelled terminal fragments by capillary or gel electrophoresis
T-RFLP (2)raw data
Red: internal size standard Blue: forward primer Green: reverse primer
T-RFLP (3)
Analysis of raw data with Genescan:
virtual filter: adjust overlap of fluorescence
sizing: standard curve
integration of peaks
size calling curve
T-RFLP (4)
relative peak
height
T-RF length in nucleotides
Electropherogram: a visual profile of the community.
In principle, the height and area of the peaks are representative of the abundance of the groups of organisms.Several groups of organisms may share the same T-RF.
Table: digital data can be further processed and used, for example, to generate dendrograms illustrating the relationship between bacterial communities.
- several groups of organisms may share the same T-RF
- T-RFs need to be within range of size standard
T-RFLP : resolution
resolution of T-RFLP depends on the choice of restriction enzyme / primer combination
Ribosome Database Project:
TAP-TRFLP application
enter choice of enzyme/primer
in silico digestion of 16S rRNA on the database
http://rdp8.cme.msu.edu/html/TAP-trflp.html#program
several digests + combine data increase resolution
T-RFLP : good technique
- reproducible technique
- relatively fast monitor community dynamics
- culture-independent
- digital data for further analyses
- link data to clone libraries
Rs + PO4
Rhodoferax antarcticus ( Proteobacteria, Comamonadaceae)
but…..…. need to look at data to avoid pitfalls and know the limitations
sources of biases (von Wintzingerode et al., 1997):
- experimental design
- sampling
- storage of sample
- DNA extraction
- PCR amplification (loads of literature!)
keep experimental procedures constant
PCR-based techniques provide information that is not obtainable through other methods
limitations inherent to T-RFLP
- glitches in the electrophoresis rerun sample
- incomplete digestion (partially single-stranded amplicons; Egert & Friedrich, 2003) be aware
clone M232: 3 hours digestion
clone M232: 15 hours digestion
limitations inherent to T-RFLP (2)
renaturation of sample:
salts in buffer
amount of DNA in sample
delay between denaturation and electrophoresis
rerun sample
renaturation of internal size standard
limitations inherent to T-RFLP (3)
overloading of the capillary rerun sample
12 seconds injection
3 seconds injection
6 seconds injection
d
d
limitations inherent to T-RFLP (4) sizing problems
discrepancy between:
expected T-RF (from in silico digestion of known sequence) apparent T-RF (from electrophoresis)
caution when interpreting T-RFLP profiles
limitations inherent to T-RFLP (5) sizing problems
causes:
- apparent size varies with the type of genetic analyser: a 142 nt fragment will measure 143.4 and 140.6 nt respectively on a gel or capillary genetic analyser (GeneScan Reference Guide)
- resolution: decreases as fragment length increases
- ROX label of internal standard migrates more slowly than the FAM label of the forward primer (Boorman et al., 2002)
- apparent size of fragment depends on its secondary structure
limitations inherent to T-RFLP (6) sizing problems
0
2
4
6
8
10
12
25 50 75 100 125 150 175 200 225 250 275 300 325 350 375 400 425
Expected fragment size in nucleotide (CfoI )
Dif
fere
nce
in s
ize
(nt)
- difference ± proportional to fragment length
- can vary from -2 to -4 nt for very similar length of fragment
- outside range of size standard: can’t size accurately
- abnormal migration
expected size apparent size difference
AluI 109 105.8 3.2
CfoI 114 103.0 11.0
limitations inherent to T-RFLP (7) sizing problems
- possible “hairpin” from secondary structure
- no such discrepancy from other clones with same sequence immediately preceding the restriction site
very abnormal migration from a specific clone T-RF:
in press: Nogales et al.
(a study of mobility anomalies of 16S rRNA gene fragments)
-----CGGAACGTGCCCAGTCGTGGGGGATAACGCAGC G ------CGGAACGTGCCCAGTCGTGGGGGATAACGCAGCGA ------CGGAACGTGCCCAGTCGTGGGGGATAA GCGTCGGA ------CGGAACGTGCCCAGTCGTGGGGGATAAGCGTCGA
limitations inherent to T-RFLP (8) Conclusions:
T-RFLP very reproducible (electropherograms need to be perfect)
comparison of data limited to studies using same type of genetic analyser
cannot predict phylogenetic affiliations from the length of the T-RFs
within its limitations, T-RFLP is a good culture-independent technique for profiling microbial communities!
many community profiling techniques
Techniques based on PCR of rDNACloning and sequencing of 16S rDNA DGGE (denaturing gradient gel electrophoresis)SSCP (single strand conformation polymorphism)RFLP (restriction fragment length polymorphism)LH-PCR (length heterogeneity analysis by PCR)ARISA (automated ribosomal intergenic spacer analysis)DGGE and T-RFLP also used for diversity of catabolic genes
Other approaches to community profilingHybridisation, FISH, PLFA, BIOLOG
Linking metabolic function to phylogenySIP (stable isotope probing)