Summary of previous lesson Janzen-Connol hypothesis; explanation of why diseases lead to spatial heterogeneity Diseases also lead to heterogeneity or changes through time –Driving succession –The Red Queen Hypothesis: selection pressure will increase number of resistant plant genotypes Co-evolution: pathogen increase virulence in short term, but in long term balance between host and pathogen Density dependance

Disease and competition Competition normally is conducive to increased rates of disease: limited resources weaken hosts, contagion is easier Pathogens can actually cryptically drive competition, by disproportionally affecting one species and favoring another

Janzen-Connol Regeneration near parents more at risk of becoming infected by disease because of proximity to mother (Botryosphaeria, Phytophthora spp.). Maintains spatial heterogeneity in tropical forests Effects are difficult to measure if there is little host diversity, not enough host-specificity on the pathogen side, and if periodic disturbances play an important role in the life of the ecosystem

Diseases and succession Soil feedbacks; normally it’s negative. Plants growing in their own soil repeatedly have higher mortality rate. This is the main reason for agricultural rotations and in natural systems ensures a trajectory towards maintaining diversity Phellinus weirii takes out Douglas fir and hemlock leaving room for alder

The red queen hypothesis Coevolutionary arm race Dependent on: –Generation time has a direct effect on rates of evolutionary change –Genetic variability available –Rates of outcrossing (Hardy-weinberg equilibrium) –Metapopulation structure

Diseases as strong forces in plant evolution Selection pressure Co-evolutionary processes –Conceptual: processes potentially leading to a balance between different ecosystem components –How to measure it: parallel evolution of host and pathogen

Rapid generation time of pathogens. Reticulated evolution very likely. Pathogens will be selected for INCREASED virulence In the short/medium term with long lived trees a pathogen is likely to increase its virulence In long term, selection pressure should result in widespread resistance among the host

More details on: How to differentiate linear from reticulate evolution: comparative studies on topology of phylogenetic trees will show potential for horizontal transfers. Phylogenetic analysis neeeded to confirm horizontal transmission

Phylogenetic relationships within the Heterobasidion complex Fir-Spruce Pine Europe Pine N.Am.

Geneaology of “S” DNA insertion into P ISG confirms horizontal transfer. Time of “cross-over” uncertain 890 bp CI>0.9 NA S NA P EU S EU F

Complexity of forest diseases At the individual tree level: 3 dimensional At the landscape level” host diversity, microclimates, etc. At the temporal level

Complexity of forest diseases Primary vs. secondary Introduced vs. native Air-dispersed vs. splash-dispersed, vs. animal vectored Root disease vs. stem. vs. wilt, foliar Systemic or localized

Stem canker on coast live oak

Progression of cankers Older canker with dry seep Hypoxylon, a secondary sapwood decayer will appear

Root disease center in true fir caused by H. annosum

HOST-SPECIFICITY Biological species Reproductively isolated Measurable differential: size of structures Gene-for-gene defense model Sympatric speciation: Heterobasidion, Armillaria, Sphaeropsis, Phellinus, Fusarium forma speciales

Phylogenetic relationships within the Heterobasidion complex Fir-Spruce Pine Europe Pine N.Am.

Recognition of self vs. non self Intersterility genes: maintain species gene pool. Homogenic system Mating genes: recognition of “other” to allow for recombination. Heterogenic system Somatic compatibility: protection of the individual.

Recognition of self vs. non self What are the chances two different individuals will have the same set of VC alleles? Probability calculation (multiply frequency of each allele) More powerful the larger the number of loci …and the larger the number of alleles per locus

INTERSTERILITY If a species has arisen, it must have some adaptive advantages that should not be watered down by mixing with other species Will allow mating to happen only if individuals recognized as belonging to the same species Plus alleles at one of 5 loci (S P V1 V2 V3)

MATING Two haploids need to fuse to form n+n Sex needs to increase diversity: need different alleles for mating to occur Selection for equal representation of many different mating alleles

SEX Ability to recombine and adapt Definition of population and metapopulation Different evolutionary model Why sex? Clonal reproductive approach can be very effective among pathogens

Long branches in between groups suggests no sex is occurring in between groups Fir-Spruce Pine Europe Pine N.Am.

Small branches within a clade indicate sexual reproduction is ongoing within that group of individuals 890 bp CI>0.9 NA S NA P EU S EU F

Index of association Ia= if same alleles are associated too much as opposed to random, it means sex is not occurring Association among alleles calculated and compared to simulated random distribution

SOMATIC COMPATIBILITY Fungi are territorial for two reasons –Selfish –Do not want to become infected If haploids it is a benefit to mate with other, but then the n+n wants to keep all other genotypes out Only if all alleles are the same there will be fusion of hyphae If most alleles are the same, but not all, fusion only temporary

The biology of the organism drives an epidemic Autoinfection vs. alloinfection Primary spread=by spores Secondary spread=vegetative, clonal spread, same genotype. Completely different scales (from small to gigantic) Coriolus Heterobasidion Armillaria Phellinus

OUR ABILITY TO: Differentiate among different individuals (genotypes) Determine gene flow among different areas Determine allelic distribution in an area

WILL ALLOW US TO DETERMINE: How often primary infection occurs or is disease mostly chronic How far can the pathogen move on its own Is the organism reproducing sexually? is the source of infection local or does it need input from the outside

Evolution and Population genetics Positively selected genes:…… Negatively selected genes…… Neutral genes: normally population genetics demands loci used are neutral Loci under balancing selection…..

Evolution and Population genetics Positively selected genes:…… Negatively selected genes…… Neutral genes: normally population genetics demands loci used are neutral Loci under balancing selection…..

Evolutionary history Darwininan vertical evolutionray models Horizontal, reticulated models..

Phylogenetic relationships within the Heterobasidion complex Fir-Spruce Pine Europe Pine N.Am.

Geneaology of “S” DNA insertion into P ISG confirms horizontal transfer. Time of “cross-over” uncertain 890 bp CI>0.9 NA S NA P EU S EU F

Because of complications such as: Reticulation Gene homogeneization…(Gene duplication) Need to make inferences based on multiple genes Multilocus analysis also makes it possible to differentiate between sex and lack of sex (Ia=index of association)

Basic definitions again Locus Allele Dominant vs. codominant marker –RAPDS –AFLPs

How to get multiple loci? Random genomic markers: –RAPDS –Total genome RFLPS (mostly dominant) –AFLPS Microsatellites SNPs Multiple specific loci –SSCP –RFLP –Sequence information Watch out for linked alleles (basically you are looking at the same thing!)

Sequence information Codominant Molecules have different rates of mutation, different molecules may be more appropriate for different questions 3rd base mutation Intron vs. exon Secondary tertiary structure limits Homoplasy

Sequence information Multiple gene genealogies=definitive phylogeny Need to ensure gene histories are comparable” partition of homogeneity test Need to use unlinked loci

Thermalcycler DNA template Forward primer Reverse primer

Gel electrophoresis to visualize PCR product Ladder (to size DNA product)

From DNA to genetic information (alleles are distinct DNA sequences) Presence or absence of a specific PCR amplicon (size based/ specificity of primers) Differerentiate through: –Sequencing –Restriction endonuclease –Single strand conformation polymorphism

Presence absence of amplicon AAAGGGTTTCCCNNNNNNNNN CCCGGGTTTAAANNNNNNNNN AAAGGGTTTCCC (primer)

Presence absence of amplicon AAAGGGTTTCCCNNNNNNNNN CCCGGGTTTAAANNNNNNNNN AAAGGGTTTCCC (primer)

RAPDS use short primers but not too short Need to scan the genome Need to be “readable” 10mers do the job (unfortunately annealing temperature is pretty low and a lot of priming errors cause variability in data)

RAPDS use short primers but not too short Need to scan the genome Need to be “readable” 10mers do the job (unfortunately annealing temperature is pretty low and a lot of priming errors cause variability in data)

RAPDS can also be obtained with Arbitrary Primed PCR Use longer primers Use less stringent annealing conditions Less variability in results

Result: series of bands that are present or absent (1/0)

Root disease center in true fir caused by H. annosum

Ponderosa pineIncense cedar

Yosemite Lodge 1975 Root disease centers outlined

Yosemite Lodge 1997 Root disease centers outlined

Are my haplotypes sensitive enough? To validate power of tool used, one needs to be able to differentiate among closely related individual Generate progeny Make sure each meiospore has different haplotype Calculate P

RAPD combination

Conclusions Only one RAPD combo is sensitive enough to differentiate 4 half-sibs (in white) Mendelian inheritance? By analysis of all haplotypes it is apparent that two markers are always cosegregating, one of the two should be removed

AFLP Amplified Fragment Length Polymorphisms Dominant marker Scans the entire genome like RAPDs More reliable because it uses longer PCR primers less likely to mismatch Priming sites are a construct of the sequence in the organism and a piece of synthesized DNA

How are AFLPs generated? AGGTCGCTAAAATTTT (restriction site in red) AGGTCG CTAAATTT Synthetic DNA piece ligated –NNNNNNNNNNNNNNCTAAATTTTT Created a new PCR priming site –NNNNNNNNNNNNNNCTAAATTTTT Every time two PCR priming sitea are within bp you obtain amplification

Dealing with dominant anonymous multilocus markers Need to use large numbers (linkage) Repeatability Graph distribution of distances Calculate distance using Jaccard’s similarity index

Jaccard’s Only 1-1 and 1-0 count, 0-0 do not count

Jaccard’s Only 1-1 and 1-0 count, 0-0 do not count A: AB= (1-AB) B: BC= C: AC=0.20.8

Now that we have distances…. Plot their distribution (clonal vs. sexual)

Now that we have distances…. Plot their distribution (clonal vs. sexual) Analysis: –Similarity (cluster analysis); a variety of algorithms. Most common are NJ and UPGMA

Now that we have distances…. Plot their distribution (clonal vs. sexual) Analysis: –Similarity (cluster analysis); a variety of algorithms. Most common are NJ and UPGMA –AMOVA; requires a priori grouping

AMOVA groupings Individual Population Region AMOVA: partitions molecular variance amongst a priori defined groupings

Now that we have distances…. Plot their distribution (clonal vs. sexual) Analysis: –Similarity (cluster analysis); a variety of algorithms. Most common are NJ and UPGMA –AMOVA; requires a priori grouping –Discriminant, canonical analysis

Results: Jaccard similarity coefficients Coefficient Frequency P. nemorosa P. pseudosyringae: U.S. and E.U. 0.3 Coefficient Frequency

Pp U.S. Pp E.U Jaccard coefficient of similarity 0.7 P. pseudosyringae genetic similarity patterns are different in U.S. and E.U.

P. nemorosa P. ilicis P. pseudosyringae Results: P. nemorosa

Results: P. pseudosyringae P. nemorosa P. ilicis P. pseudosyringae = E.U. isolate

Now that we have distances…. Plot their distribution (clonal vs. sexual) Analysis: –Similarity (cluster analysis); a variety of algorithms. Most common are NJ and UPGMA –AMOVA; requires a priori grouping –Discriminant, canonical analysis –Frequency: does allele frequency match expected (hardy weinberg), F or Wright’s statistsis

The “scale” of disease Dispersal gradients dependent on propagule size, resilience, ability to dessicate, NOTE: not linear Important interaction with environment, habitat, and niche availability. Examples: Heterobasidion in Western Alps, Matsutake mushrooms that offer example of habitat tracking Scale of dispersal (implicitely correlated to metapopulation structure)---

The “scale” of disease Dispersal gradients dependent on propagule size, resilience, ability to dessicate, NOTE: not linear Important interaction with environment, habitat, and niche availability. Examples: Heterobasidion in Western Alps, Matsutake mushrooms that offer example of habitat tracking Scale of dispersal (implicitely correlated to metapopulation structure)---

Have we sampled enough? Resampling approaches Saturation curves –A total of 30 polymorphic alleles –Our sample is either 10 or 20 –Calculate whether each new sample is characterized by new alleles

Saturation curves No Of New alleles

If we have codominant markers how many do I need IDENTITY tests = probability calculation based on allele frequency… Multiplication of frequencies of alleles 10 alleles at locus 1 P1=0.1 5 alleles at locus 2 P2=0,2 Total P= P1*P2=0.02

White mangroves: Corioloposis caperata

White mangroves: Corioloposis caperata Distances between study sites

Coriolopsis caperata on Laguncularia racemosa Forest fragmentation can lead to loss of gene flow among previously contiguous populations. The negative repercussions of such genetic isolation should most severely affect highly specialized organisms such as some plant- parasitic fungi. AFLP study on single spores

Using DNA sequences Obtain sequence Align sequences, number of parsimony informative sites Gap handling Picking sequences (order) Analyze sequences (similarity/parsimony/exhaustive/bayesian Analyze output; CI, HI Bootstrap/decay indices

Using DNA sequences Testing alternative trees: kashino hasegawa Molecular clock Outgroup Spatial correlation (Mantel) Networks and coalescence approaches

Pacifico Caribe

From Garbelotto and Chapela, Evolution and biogeography of matsutakes Biodiversity within species as significant as between species