First, Some Project Background

In order to get some background and understanding of the hematodinium work done in the Roberts lab, I chatted with Grace Crandall via Zoom for an hour or so.

As background, Grace is a member of the Roberts lab who just graduated. She’s been investigating how infected and uninfected Tanner crab respond to various temperature regimes by collecting and analyzing transcriptomes. Here’s a (very very) rough outline of the experimental protocol, which was conducted in partnership with Pam Jensen at the Alaska Department of Fish and Game (ADF&G):

  • 400 male Tanner crab were collected from Stephen’s Passage, near Juneau AK, and brought back to the ADF&G lab in Juneau

  • All were tested for hematodinium infection via cPCR, 180 were selected for temperature treatments (90 infected, 90 uninfected)

  • Crabs were divided into 9 tanks, with 10 infected and 10 uninfected in each

  • Tanks were at 3 different temperatures - 4°C (cool), 7.5°C (ambient), and 10°C (elevated)

  • Hemolymph samples were taken 3 times - initial sample, Day 2, and Day 17, with Day 0 being the start of the differential temp treatments

  • RNA was extracted from hemolymph samples

  • Gene expression differences over time and between treatments were analyzed

Notes on Conversation with Grace

The following notes are in no particular order

  • Grant money for project was released in fall 2017

  • Collections targeted immatures. Since crabs are more prone to infection while molting, and Tanner crabs have a terminal molt, immature crabs have a higher chance of having BCS

  • There’s a new version of the maturity ratio (chela height vs carapace width, used to determine maturity in Tanner crab) that’s specific to SE Alaska

  • Crabs were left in tanks for ~9 days to acclimate

  • Crabs weren’t fed for two reasons. First, it would’ve been an extra variable to manage. Second, because water quality was an issue

  • Prevalence of hematodinium in Kodiak/Cook Inlet (interested me because it’s between the high-infection areas of SE Alaska and the low-infection-rate Bering Sea): Grace isn’t sure, will send over papers

  • All tanks had header tanks that were fed into main tanks for temperature treatment. No water mixing between tanks

  • Some initially-uninfected crabs had extremely low infection levels at end of experiment. Likely because cPCR was used initially, qPCR was used at the end

  • There’s a ton of good qPCR data that’s largely unused - that’ll be great for looking at the hematodinium side of things. How much hemat is present in which crabs? What temp treatment were they?

  • No hemolymph samples were taken from dead crab

  • Heater likely broke during 9-day acclimation period, but doesn’t remember exactly

  • Several samples were taken from each crab at Day 17 as a safety net. 3 samples each from cold and ambient crab, 6 each from warm crab (since 95% of warm-treatment crab died by Day 10)

  • There are still a ton of hemolymph samples in the -80 freezer

  • Hematodinium/crab comparisons done with MEGAN (which assigns phylogenetic stuff to sequence data) and Busco. Steps on github wiki

  • Sequenced the crabs that fit with libraries we wanted. As am example: wanted pool of samples with Day 2 uninfected crabs. Picked ones that matched it with good RNA yields

  • Morado and Jeff Fields have a lot of good hematodinium research

  • Manifestations of hematodinium infection differ some between hosts. BCD might be unique for its bitter taste

  • Samples in the -80 freezer don’t have RNALater anymore - they were pelleted and the supernatant was extracted

  • In Grace’s paper draft, Supplemental I is all samples, Supplemental II is libraries

  • Later, figured how to do differential gene expression using package DESeq2

  • Certain stages of hematodinium may be more infectious than others. Dinospore may be more infectious

  • Libraries were limited by time, since so much time was spent figuring out extraction protocol

Biggest Obstacles

  • The biggest problem was finding a good method of RNA extraction that wouldn’t result in degraded or impure RNA. This problem took a year to solve! Eventually Pam Jensen found a method that worked.

  • GeneWiz was used for sequencing, since they could sequence smaller amounts of RNA and were much quicker

  • The large die-off (95% of warm-treatment crabs dying by Day 10) limited what could be analyzed

  • The NW Genomic Center (UW facility in Figgy Hall) took over 6 months to sequence samples

  • Since crab hemolymph is clear for uninfected crabs, it was sometimes difficult to determine whether you were getting seawater or hemolymph

  • When trying to pellet samples, was getting more of a slush. This might have been why the RNA wasn’t extracting well. After switching to a hemolymph extraction protocol, this was largely resolved

Next Steps

  • Looking at hematodinium. Have only looked at host response to temperature. What does hematodinium do inside host? Is it impacted by temperature?

  • Hemolymph was placed in RNALater, spun down, supernatant was pulled off and kept. Might be something to be extracted in there

  • Getting more of a time series. Won’t be super robust for warm treatment. Already partially done, but a more statistic-focused approach (on either crab or hematodinium) would be great