reveal pathotype-specific genes and SNPs useful for molecular - PowerPoint PPT Presentation

Comparative genome analyses of Synchytrium endobioticum representing six different pathotypes reveal pathotype-specific genes and SNPs useful for molecular diagnostics Hai Nguyen, Ph.D. Research Scientist

How Canada became involved with potato wart? In North America, the disease • was first detected in Newfoundland in 1909. In 1912, Ottawa forbade the • movement of Newfoundland soil into Canada. In 2000, potato wart was found • in Prince Edward Island. US prohibited the movement of seed and table stock potatoes from Prince Edward Island. New requirements are now in • place for export of potato from PEI to the United States AAFC/CFIA (only 2 scientists) • with focused program on potato wart

ladder water control PEI contaminated soil Triparticalcar arcticum Spizellomyces plurigibbosus Gaertneriomyces semiglobiferus Kochiomyces dichotomus Powellomyces sp Entophlyctis conferrae-glomeratae Podochytrium sp Blastocladiella britannica Rhizoclosmatium globosum Chytriomyces hyalinus Rhizophydium patellarium Spizellomyces acuminatus Rhizophydium haynaldii S. endobioticum (1944) S. endobioticum (1903) S. plantagineum S. aureum S. collapsum S. perforans Real time PCR tests: quantification of spores, viability S. papillatum PCR test: tests Presence/absence CA, NFL (3 samples), P? CA, NFL, P? CA, PEI (2 samples), P6 DE, P18 CA, PEI, P6 NL, P18 Microsatellites. Each group creates a CA, PEI, P6 DE, P2 NL, P1 NL, P2 UK, P1 DE, P1 unique pattern and strains could be DE, P2 UK, P1 NL, P1 UK, P? differentiated from each other NL, P6 DE, P6 UK, P?

Pathotypes of S. endobioticum • Isolates of S. endobioticum grouped into pathotypes based on interaction with a differential set of resistant potato varieties • 39 pathotypes (and maybe more)

Pathotypes of S. endobioticum • Pathotypes manifest themselves as a result of different R genes present in potato varieties and their cognate Avr genes in S. endobioticum – Presence of Sen1 gene in potato gives it resistance to pathotype 1 isolates – Sen1 recognizes the effector AvrSen1 present only in S. endobioticum pathotype 1 isolates and it triggers HR in the potato, which gives the potato resistance to S. endobioticum pathotype 1

None of the tests that exist so far can tell difference between various pathotypes of S. endobioticum The benefit of knowing (or predicting) the pathotype gives producers a tool to mitigate economic risk

• In 2018, we published a paper on the mitochondrial genome of S. endobioticum • NGS data generated for 30 strains of S. endobioticum • Phylogeny of concatenated mito genes does not show any distinction between S. endobioticum strains

Haplotype network based on 141 polymorphic sites shows some groupings corresponding perhaps to certain pathotypes

Would differences in the nuclear genome allow us to separate pathotypes of S. endobioticum ?

Would differences in the nuclear genome allow us to differentiate between pathotypes? Technically yes because the pathotype is a manifestation of the pathogen’s genotype and its interaction with the host’s genotype, and there is more (relevant) information in the nuclear genome compared to the mitochondrial genome 3 STRATEGIES 6 1 2 8 Pathotype phylogeny Site specific variation in Core genome / pathotype- represented by genes in the genome (SNPs, specific genome the core genome indels, struct. variants) Presence/absence of genes

1. Read mapping to reference: building a pan genome Pan genome Pan genome LEV6574 (Canadian) MB42 (Dutch)

1. Read mapping to reference: only some of the 30 datasets have enough coverage to proceed for in depth comparative genome analyses Strain Coverage Reference Location 1D1_MB42 NA van de Vossenberg et al. (2018) Langenboom, the Netherlands 2G1_MB08 9.9 van de Vossenberg et al. (2018) Mussel, the Netherlands 2G1_SE4 20.4 van de Vossenberg et al. (2018) Germany St. Eleanor ’ s, Prince Edward Island, Canada 6O1_LEV6574 NA van de Vossenberg et al. (2018) 6O1_LEV6602 19.5 van de Vossenberg et al. (2018) Augustine Cove, Prince Edward Island, Canada 6O1_LEV6687 40.4 van de Vossenberg et al. (2018) New Annan, Prince Edward Island, Canada 6O1_CHY1000 41.7 new Carmanville, Newfoundland, Canada 6O1_CHY1001 55.1 new St. Philips, Newfoundland, Canada 6O1_SE5 63.9 van de Vossenberg et al. (2018) the Netherlands 6O1_SE6 93.7 van de Vossenberg et al. (2018) Germany 8F1_DEN01 32.7 van de Vossenberg et al. (2018) Jylland, Denmark 8F1_CHY1002 79.1 new Avondale, Newfoundland, Canada 18T1_SE7 48.6 van de Vossenberg et al. (2018) Borgercompagnie, the Netherlands 38Nevsehir_MB56 12.2 van de Vossenberg et al. (2018) Nevsehir, Turkey

2. De novo genome assembly Number of contigs (>= 1000 bp) 3654 3500 3434 3372 3325 3367 3341 3275 3238 3226 3102 3105 3079 3077 1285 786 … e d 8 4 e - 2 7 0 1 5 6 1 2 7 6 e 5 c e 0 E c 0 8 0 0 E E 0 0 E n r l B n 6 6 0 0 S S N S B S 0 e b _ _ e 6 6 1 1 _ _ 1 _ M M 4 E r m 1 r V V Y Y 1 1 1 D Y e e 7 _ G O O T _ f e E E H H H f 5 _ 1 8 r e 2 L L C C 6 6 C s e 6 1 i G 1 h r s _ _ r V _ _ F _ _ 1 1 a 2 _ 1 1 e 8 1 2 - E O O 4 O O s e L F 4 v 7 r _ 6 6 8 B 6 6 e 5 _ 1 N M 2 6 O V 8 4 _ 6 3 B E 1 M L D _ 1 1 _ O 1 D 6 1 Total length of assembly (>= 1000 bp) 23209196 21483073210010962119336621383118 21492804 214316142127949521316832 21296695 211960742104956021249904 20608936 20174921 20008646 … e e d 8 4 2 7 0 1 5 6 1 2 7 6 - e c e 0 c 0 8 0 0 E E 0 0 E E 5 n n r B 0 0 S S N S B l S 6 6 0 b e _ e _ 6 6 _ _ 1 _ M 4 1 1 E M m r 1 r V V Y Y 1 1 1 D Y e e 7 _ G O O H T _ e E E H H f f 5 _ 1 8 r e s 2 e L L C C 6 6 C 6 1 i s G 1 r r _ _ h V _ _ F _ _ a _ 1 1 2 1 1 e 8 1 2 - 4 E O O O O s e L F 4 7 v r _ 6 6 8 B 6 6 e _ 5 1 6 N M 2 O V 4 8 _ 6 B E 3 1 L M D _ 1 1 _ O 1 D 6 1

1 1 D D 1 1 _ _ M M B B 4 4 2 2 1 _ 1 _ r r D D e e 1 1 f f 2. Genome annotation e e _ _ r r M M e e B n B n 8031 85.1 c c 4 4 e e 2 2 _ _ r r e e - - a a s s s s e e m m b b l 6675 l e e d d 82 2 2 G G 1 1 _ _ M M B B 6696 0 0 81.7 8 8 6 6 2 2 O O G G 1 1 1 1 _ _ _ _ L L E S E S 6668 E E 82.8 V V 4 4 6 6 5 5 7 7 4 4 _ _ r r e e f f e e r r e e n n 6 8671 6 84.2 c c O O e e Number of validated gene models 1 1 _ _ L L E E V V 6 6 5 5 7 7 4 4 _ _ 6562 r r 83.8 e e - - … … 6 6 Complete BUSCO (%) O O 1 1 _ _ L L E E V V 6 6 6 6 6688 0 0 82.1 2 2 6 6 O O 1 1 _ _ L L E E V V 6 6 6 6 6579 8 8 83.1 7 7 6 6 O O 1 1 _ _ C C H H Y Y 1 1 0 0 6642 0 0 82.1 0 0 6 6 O O 1 1 _ _ C C H H Y Y 1 1 0 0 6585 0 0 83.4 1 1 6 6 O O 1 1 _ _ S S 6558 82.7 E E 5 5 6 6 O O 1 1 _ _ S S 6624 E E 84.1 6 6 8 8 F F 1 1 _ _ D D E E N N 6646 0 0 1 1 82 8 8 F F 1 1 _ _ C C H H Y Y 1 1 0 0 7342 84.5 0 0 2 2 1 1 8 8 T T 1 1 _ _ S S 3 3 6584 E E 81.7 8 8 7 7 N N e e v v s s e e h h i i r r _ _ M M B B 6712 5 5 82.8 6 6

Orthogroup analysis • Validated gene models were pooled and clustered into gene families (orthogroups) with orthofinder2 • 8012 orthogroups were found in total (pan genome) • 4625 orthogroups were considered single copy (large subset of the core genome)

• Orthogroups of interest are those that are present in a given pathotype but absent in all other pathotypes analysed Orthogroup Se_MB42.1D1 Se_MB08.2G1 Se_SE4.2G1 Se_LEV6574.6O1 Se_LEV6602.6O1 Se_LEV6687.6O1 Se_CHY1000.6O1 Se_CHY1001.6O1 Se_SE OG0007119 0 1 1 0 0 0 0 0 OG0007125 0 1 1 0 0 0 0 0 OG0007127 0 1 1 0 0 0 0 0 OG0007134 0 1 1 0 0 0 0 0 Pathotypes (* represented by a single strain) • 1(D1)* – 47 orthogroups • 2(G1) – 4 orthogroups • • 8(F1) – 2 orthogroups • 18(T1)* – 60 orthogroups • 38(Nevsehir)* – 24 orthogroups • 6(O1) – 5 orthogroups Caveat: not all strains positive for all 5, they usually have 3 out of the 5 • • Note: Some pathotype specific orthogroups are going to be strain specific because there was only 1 representative genome for that pathotype

4625 single copy gene neighbour joining phylogeny (amino acid level) Alignment Length = 2.4M columns (PEI, Canada) Variable sites = 1.6% (38.4k columns) (PEI, Canada) (Newfoundland, Canada) (Newfoundland, Canada) (Newfoundland, Canada) (PEI, Canada) (Langeboom, The Netherlands) (The Netherlands) (Germany) (Germany) (Mussel, The Netherlands) (Borgercompagnie, The Netherlands) (Nevsehir, Turkey) (Jylland, Denmark)

Look for genes that have similar relationships as our hypothesis of the pathotyping system • Search for relationships in single copy genes that resemble the pathotyping groupings • Calculate the symmetric difference between two trees (Robinson-Foulds) • No good candidates were found, but might need more technical development

SNP analysis Map read fragments to ref. genome, detect variation (SNPs)