Field of Science

Today's work on the RNA-seq samples

The Co-op tech has pressed half the samples through the RNeasy kit spin columns, and will probably get the rest done today.

But she also is still working on the PCR checks (problems getting amplification from the colony-DNA material) and redoing some of the transformation assays (using the frozen cells I'd saved for each sample) because two of her original tests gave surprisingly low transformation frequencies.

I'm hoping we can also get started on the next steps today - checking the RNA concentrations using the Nanodrop and running aliquots in a gel to check that the rRNA bands are intact.  First step for this is to clean up a gel box and comb and see if we have any RNA-loading dye made up.

Other tasks for the soon-to-be-departing Co-op tech

The Co-op tech will be leaving us at the end of the month.  Last week she gave an excellent bab meeting presentation, and this revealed a couple of loose ends and interesting possibilities that should be cleared up before she leaves.

First, she and the sabbatical visitor (now back home in Regina) isolated a new mutation in rpoD that causes hypercompetence.  But this strain hasn't yet been added to our formal collection of frozen strains and the associated 'Strain List' database.  This is essential and urgent.

Second, the mutant hunt turned up a couple of other 'possibly hypercompetent' mutants whose phenotypes haven't been checked out.  These need to be checked in competence time courses, and added to the Strain List collection if they turn out to be genuinely hypercompetent.

A weird result that I'd forgotten about is the finding that cells transformed with a mixture of NovR and KanR DNA fragments (generated by PCR) showed a much LOWER cotransformation frequency than expected.  We suspect this reflects some chromosome-level interaction between these linked segments, but we'd first need to reproduce the result.

RNA-seq progress

I've collected and frozen all the 24 samples for our make-up RNA-seq run. (not the Trizol-prep ones - they've been deep-sixed).  And this morning the co-op tech learned to prep RNA from each sample. She's done the first 4, and will do the rest over the next couple of days.

The next steps are:

  1. Complete the PCR tests of strain genotypes and the analysis of transformation frequency data. She's still working on the PCR tests, but so far everything looks OK.
  2. Check the RNA concentration using the Nanodrop
  3. Run aliquots of the samples in a gel to check integrity of the rRNA bands (surrogate for integrity of the mRNA).
  4. Treat 5 µg of each sample with DNA-free.  We found our stock from last year, and there's still enough to treat all our samples.
  5. The former RA says we can take the DNA-free-treated samples directly to the RiboXero ste; we don't need to first do a 'clean-up' step with the RNeasy Minelute kit spin-columns.
  6. Treat an aliquot of each sample with RiboZero.  We will use only half as much RNA as recommended, and only 1/4 as much of the other reagents (in 1/4 of the recommended volume, of course). This will let us treat 24 samples with a 6-treatment RiboZero kit.
  7. Give the samples to the former RA in her new lab for library preparation and sequencing.

What can I recover from an old failed experiment

About 18 months ago I did a big mutagenesis experiment, intending to isolate new hypercompetent mutations.  I made several mistakes and the experiment was a failure, but I did freeze stocks of intermediate cultures.   At the time I thought that some of these could be used in a future attempt, because they came from stages before the mistakes were made.

I still want to repeat this experiment, and I just found the stocks in the -80 °C freezer.  Now I need to decide which are potentially useful, and throw out the rest.  Here's photos of what I found:


The letters A-G refer to different strains, each with a wildtype version of a gene known to give rise to hypercompetence-causing mutations, and to different levels of mutagenesis
  • A, B & C: wildtype cells, incubated in 0 (A), 0.05 (B) and 0.08 (C) M solutions of the mutagen EMS.
  • D & E: strain RR514, which has a Streptomycin-resistance mutation (StrR) close to the wildtype sxy gene, incubated in 0.05 (D) and 0.08 (E) M solutions of EMS.
  • F & G: strain RR805, which has a chloramphenicol cassette (CmR) inserted within a few kb of (= closely linked to) the wildtype murE gene, incubated in 0.05 (F) and 0.08 (G) M solutions of EMS.

The big tubes turn out to be useless, since they contain cells that were incubated with the wrong DNA after the mutagenesis.  Most of the small tubes also are from stages that have been incubated with DNA, (e.g. label 'F DNA'), but others (the ones labeled '90') were frozen after 90 min of post-mutagenesis growth, before the DNA addition step.  These ones I can use.

The first step now is to do a test I didn't do in the original experiment, to check that the EMS mutagenesis did indeed cause mutations by plate some of the cells on low-concentration novobiocin. I'll do this test on the wildtype cells (B & C), so not to unnecessarily use up the more valuable cells in the marked strains (D-F).  I don't have tubes of the control A culture, so I'll just use normal wildtype cells.

If this test shows that the mutagenesis worked, I have two alternatives.  1.  I could isolate DNA from the mutagenized marked cultures and use it to transform wildtype cells to StrR (D & E, to enrich for cells with sxy mutations) or CmR (F & G, to enrich for cells with murE mutations).  Then I'd enrich these transformants for hypercompetent mutants by transforming them with the PCR'd NovR fragment (after first testing that this works well).  2.  I could do the hypercompetence-selection transformation first, and then isolate DNA and transform wildtype cells with selection for the linked marker.  The advantage of 1 is that I can pool many thousands (millions?) of transformants, maintaining whatever genetic diversity my mutagenesis has created in the gene of interest.

New RNA-seq work

(OK, I just checked, RNA-seq should be hyphenated.)

I've made a couple of posts about plans for new RNA-seq work on the Sense Strand blog: and

Now it's time to get down to work.

Here's the planned samples.  We have 26 on the list, but a standard run will only be 24, so two need to be dropped.  Conveniently, the two KW20-in-Trizol samples might not be needed, depending on the available small-RNA data for H. influenzae, so I won't consider those right now.

For the rest, we have 6 mutant strains.  Given the mixups that have occurred so far, it would be prudent to check these every way we can.

We are checking antibiotic resistances and will transform each strain when we grow its culture and collect the samples for the RNA preps. One of the Honours Zoology undergrads has already checked the toxin and antitoxin strains by PCR (she's the one who suffered most from the mixup), and we'll use the same primers to check the cultures we're sampling from.  We need the former RA's help to find the primers for the ∆hfq mutant, but luckily its phenotype is quite distinctive - in our whole collection I can only think of one other mutant that has its competence down 10-fold.  RR753's phenotype is also distinctive, hypercompetent, but not very.  The crp and sxy mutants have the same phenotypes (same drug resistance, same complete lack of competence.  We have lots of sxy PCR primers, but we'll have to check to see which ones will work with this insertion mutant.  And, for both crp and sxy there's the additional problem that miniTn10 insertions do not amplify well because of their end repeats.  I wonder if we have an internal primer for miniTn10kan.

The co-op tech has checked antibiotic resistances and frozen fresh stocks of all the strains.  She's inoculated the 4 strains for the MIV-competence preps, and tomorrow we'll toy to collect all those samples.  (No, I haven't done anything in preparation yet.)