Presentation to Washington Native Plants Society

Some Comments on PCR Results from Paul Talbert

November 18th and 20th, 2016

Thanks, Katie.

I guess I am relieved we don't find evidence for Phytophtora infection, but that leaves us with no good hypothesis.

I'll look forward to hearing about the bait experiments. Meanwhile, I guess we should try to think of other hypotheses.
 

So thinking out loud here, it looks like your positive control has about 16,000 times as much Phytophthora DNA as the healthy soil, which has about 4x as much as the affected soil. The latter difference may be within sampling variability/error. So there is no evidence that a Phytophthora is causing our problem, but we do apparently have Phytophthoras in our soils, since nothing came up in the water.

Assuming we get nothing of interest again in the bait experiment, would there be any point in repeating the experiment on growing roots in the spring when the fiddleheads are coming out? I don't want to chase phantoms or beat a dead horsetail (that was a fern ally joke), just wondering if we could be missing anything. You don't have to answer now; we have till spring to consider.

Katie Coats and Marianne Elliott of the WSU Plant Diagnostics Lab in Puyallup: a report after PCR assay of Phytophthora at the genus level

(received 18 November 2016)

(In early November, I collected samples from a healthy fern - fine roots and nearby soil - off the windfall trail, and from an affected surviving fern just west of ground zero, and delivered them to Katie in her lab - Paul Shannon)
 
I have qPCR results on the sword fern samples but they're not conclusive.

I extracted DNA from the following samples:

1. "healthy soil a" = 4 grams of soil from 12" away from healthy plant
2.  "healthy soil b" = 4 grams of soil from 24" away from healthy plant
3.  "healthy plant" = roots, surface sterilized in 10% bleach for 1 minute then rinsed twice in sterile water to assure qPCR is detecting infection inside the roots, photo attached
4.  "affected soil" = 4 grams of soil from near affected plant
5.  "affected plant a" = roots, surface sterilized in 10% bleach for 1 minute then rinsed twice in sterile water, photo attached
6.  "affected plant b" = split from sample 5, run as a replicate

The DNA samples were run in duplicate through the general Phytophthora (5.8S) qPCR assay along with a positive control (DNA from a pure culture of Phytophthora) and a negative control (water). All samples were multiplexed with an amplification control that checks for evidence of PCR inhibition, something that’s common in root and soil samples.

Results are shown in the bar graph below… it’s crude… I don’t usually use the Ct values, but I think it’s a quick way to get the results across to you. The lower the Ct value, the more target DNA is present… so it makes sense that the pure DNA sample has a Ct around 16.  The qPCR runs up to 42 cycles, but data above 40 cycles is dubious. If no phytophthora is present, then there will be no Ct value at all, which was the case with the water control. Theoretically the DNA doubles in each cycle, so a Ct value one more than another represents twice as much target DNA.

he bottom line is that the healthy soil and plant generally had lower Ct values (more Phytophthora DNA) than the affected soil and plant. I don’t have a soil sample from another location outside the park to give us an idea of how normal these levels of phytophthora are, but they’re counter what we expected...  One thing to keep in mind is that this test lumps all phytophthoras together, so it remains possible that a causal phytophthora is in higher concentration in the affected plant even though the total amount of phytophthora is lower.

Therefore, we will proceed with another strategy referred to as baiting. We’ll add some rhododendron leaves as baits to the individual plant and soil samples along with water… and then plate in culture medium to see if/how much/which phytophthoras may be present. That will take a couple weeks from start to finish… so more wait time.

I’ll send you another update once we know the timeline of those results; ballpark of early December.

I’m including Marianne in case she has other comments… and let me know if you have any questions.

Blog Overview

Introduction

Narrative Summary (including comments from NY Botanical Garden’s fern curator, 

  and world expert genus taxonomist  from the University of Vermont.

Informal final report from UW’s Tim Billo and his two undergraduate students

 Lab report from Dr. Marianne Elliott, WSU

LSU Professor Valverde, plant pathologist, 2008 paper on “new virus causing disease in [a related] fern”

Site visit by WSU pathologists, and the “introduced phytophthora” thesis

 PCR and Sequencing for Pathogen Detection (July 25th, 2016)

Pathogen Detection Strategies: PCR and Sequencing

We do not know what is causing the sword fern die-off.  Mountain beaver and drought may be contributing causes.   But the spatial pattern of the die-off strongly suggests a pathogen is the driving cause:

1. The die-off, unlike drought and mountain beavers, is limited to one region of the forest
2. This region is growing approximately radially, as would infection

Phytophthora is a plausible candidate, but repeated WSU and Ribeiro lab studies have been unable to provide confirmation (see earlier blog entries for full reports).  New approaches may be needed, among which are PCR, and next-generation sequencing (NGS).  

Concerning PCR (polymerase chain reaction), Paul Talbert suggests (email, 6/8/2016)

Not to discourage the sequencing idea, which is unbiased with respect to pathogens, but I wonder if a simple PCR assay for Phytophthora would be a simpler first step (although we don't a Phytophthora sample have a positive control). The genomes of at least 8 species of Phytophthora have been sequenced, including P. cinnamomi (see attached for  six species, plus P. infestans and P.ramorum were sequenced several years ago).  Genome sequences of six Phytophthora species associated with forests in New Zealand.

Jenny Glass of the WSU Puyallup Plant diagnostics lab used the  agdia ImmunoStrip for Phytophthora.  If I understand her, and Marianne Elliott's subsequent lab techniques, PCR was not otherwise employed.  

Speculatively, and with no guarantee of success, we may wish to consider both PCR (with appropriate primers) and unbiased NGS sequencing of affected and unaffected samples.   Marianne provides this poster as a good starting point:  

 Detection of Phytophthora in plants by the Polymerase Chain Reaction

NGS sequencing has proved successful in at least one somewhat-related field study:

Unfortunately, the genome of Western Sword Fern, polystichum munitum,  has not yet been sequenced:

Presentation to the Seattle Parks Field Committee, Thursday June 2nd 2016

At Lisa Cieko's invitation, I presented Kramer & Tristan's work to the Seatle Parks Field Committee.  UW's Tim Billo supervised this work, about which he said, 

"I think it would be worth reporting some of the stats from the analysis I did of Kramer and Tristan's data--so everyone knows that the decline over only a 6 month period is statistically significant and quite large (18% average decline). And it's also important to note that the decline is not equal across the whole area; in fact some areas actually showed an increase."

Note that the survey plots are limited to Ground Zero and the region of our informal survey of August 2015.

Twenty UW Monitoring Plots, Data Collection Ongoing

Twenty random-placed monitoring plots, in and around Ground Zero

• Initial and repeat surveys by Tristan Kramer and Tristan O'Mara, Tim Billo supervising.
• Third survey to be completed soon, after which we will summarize here