Field of Science

Spectrophotometer bulbs

The visible-light spectrophotometer we use to measure culture densities has been giving erratic readings lately so I decided to try changing the bulb.  I was happy to find 5 spare bulbs in the folder with the instruction manual (the Redfield lab is PREPARED!), but then very unhappy to discover that, although the new bulb had the right specifications, its filament couldn't be lined up with slit through which the light enters.  The old and new bulbs are the same shapes but their filaments are in different places.

After a bit of cursing I discovered that two of the five bulbs were made by Phillips, and these have the filament in the right place.  The other bulbs are made by Osram; I know the spec bulbs are expensive ($50 or more), so maybe we can send them back.

My work is laid out for me

I have to make competent cell preparations of a lot of our knockout mutants, and do transformation assays on almost all of them.  They're mostly replicates of ones I've already done at least once, so I know what transformation frequency to expect (not so I can fudge the data, but so I can optimize the dilutions I plate to get the most information.

I also have to freeze four tubes of each competent preparation so the postdoc can use them in DNA uptake assays.  He thoughtfully had one of the undergrads prepare all the freezer tubes for me (label them and add 250 µl of 80% glycerol).

Planning the experiments with our new knockout mutants

In an overview post last week I described some experiments we should add to the paper describing our new collection of knockout mutants.  Now I want to think more about these, laying out what we should do and maybe how the work might be divided up.  We don't want to take on a lot of work at this stage, just enough to make the paper more than a list of mutant phenotypes.  I mentioned in the earlier post the three mutants that have specific points of interest, but now I also want to consider the mutations that have no phenotype at all.

A.  The ATP-dependent periplasmic DNA ligase:   Repeat the undergrad's experiment testing the sensitivity to nicks, using DNase I and confirming presence of nicks by running the DNA with and without denaturing.  Test the ability of wildtype and mutant cells to ligate a plasmid cut at one site by a restriction enzyme, by recovering periplasmic DNA from rec2- and rec2- lig- periplasms and transforming the DNA into E. coli.

B.  The periplasmic protein ComE1:  Repeat the uptake assays of comEI- and comE1- rec2- (and comE1- recF-?) mutants, to see if the comE1- residual uptake is dependent on translocation of DNA across the inner membrane.

C.  The potential regulator HI0659:  Test the effect of added cAMP on this mutant - if this increases competence then we know HI0659 is needed for cAMP production.  Test competence of HI0659- in sxy1 (and murE749?) hypercompetent mutants - if this increases competence then we know... what?  That HI0659 helps with induction of gene expression?

D.  The proteins that don't appear to contribute anything to competence or transformation (HI0660, HI1631, RadC, and again the ligase):  Look carefully at growth properties (from the Bioscreen time courses we're doing for all the mutants) - are they at all different from wildtype cells?  The genes may not be expressed in log-phase growth, so look at growth properties in sxy1 mutants with added cAMP (so the genes are always on).  Make double mutants - do these still grow and transform normally?  Are they UV sensitive (this would imply a repair deficit.

How much work would all this be, and who would do what?  I could do most of it, although I probably shouldn't because I've got a lot of other stuff to do.  I could easily make the double mutants (rec2- lig- and various sxy1), selecting for the SpcR cassette in the marked knockouts.  (The RA might even have DNA from these mutants already made.)  And I can make new competent preps of the comE1and rec2 combinations for the postdoc to test DNA uptake.  I can also do the competence assays of HI0659 with sxy1 and with cAMP.  The postdoc needs to do the uptake assays.  The Bioscreen assays are easy to set up but take time to analyze - this is something the RA will do better than me.  The ligase assays should probably be a group effort - I can do the transformations with nicked chromosomal DNA, and the postdoc should do the H. influenzae transformations and plasmid-recovery, passing the recovered DNA on to the RA for E. coli transformation assays.  We can discuss these experiments at Tuesday's lab meeting and divide up the work.

Analysis of the ATP-dependent DNA ligase

I spent most of yesterday poking around in databases, hoping to get a better understanding of H. influenzae's competence-induced DNA ligase.  Specific questions: Are some or all of the H. influenzae alleles non-functional?  What other bacteria have homologs of this gene?  Are the homologs typically targeted to the periplasm?  Is there any other evidence associating them with DNA uptake?  What is known about their function?  I spent a lot of time chasing red herrings but ended up with some results.

Two homologs have been characterized biochemically, one from H. influenzae and one from Neisseria meningitidis.  Both were found to be in the adenylated state when isolated from cells (I think this is common for DNA ligases), and both catalyzed typical ATP-dependent ligation reactions when purified - they could ligate nicked double-stranded DNA but could not catalyze joining of blunt ends.  As I've mentioned before, this activity is puzzling for a protein that is predicted to not be cytoplasmic, since there's no ATP in the periplasm.

There's a concern about the functionality of the H. influenzae alleles, because the original Rd sequence annotation had two ligase-related genes, but these were the result of a frameshift mutation and restart at a downstream in-frame ATG.  This was an 'authentic' frameshift present in the Rd genomic DNA, not a sequencing error.  The ligase gene had been cloned and the protein shown biochemically to be a functional ATP-dependent DNA ligase, so we initially assumed that the translation machinery just skipped through the frameshift.  However, none of the other H. influenzae sequences has this frameshift, and when we realized yesterday that the protein tested by the biochemists didn't come from the Rd strain we started worrying that maybe the Rd allele is nonfunctional.  But even the independently determined sequence of another Rd strain lacks the frameshift, and when the postdoc checked his sequence of our Rd he found that it lacks the frameshift too.  So the mutation must be only in the version of Rd that was originally sequenced by TIGR.

The search for homologs was complicated by annotation inconsistencies, but the basic result is that the distribution of this gene is odd.  Many but not all Pasteurellaceae have it; to a first approximation I'll say that members of the 'Hin-clade' have it and members of the 'Apl-clade' don't.  (It's hard to be sure because the genus names are a mess - they don't correspond with the real phylogeny.)  The Enterobacteriaceae don't have it.  All the Vibrios do, and all the Campylobacters.  Most or all of the Neisseriaceae have it (Neisseria, Kingella, Eikenella).  Shewanella and relatives have it.  A few other groups, and a lot of species I've never heard of that aren't on the Wu and Eisen tree.

The red dots show the locations of the tree of species or groups that have this gene. It's in the epsilon-proteobacteria, in a few families of the gamma-, beta- and delta-proteobacteria.  It might be in a few other groups, or these might be misassignments.  The next-closest homologs are in phage and eukaryotes; the other bacterial ATP-dependent DNA ligases are in other ligase groups.  Could this distribution have arisen by multiple horizontal transfer events, perhaps by phage?  I don't think it makes sense as massive gene loss from a bacterial common ancestor.

The annotation inconsistencies are a pain in the butt.  First, VanWagoner et al. called this gene ligA, but this name is widely used for the ubiquitous distantly-homologous NAD-dependent DNA ligase that's  essential in all bacteria, and there's no standard name for this ATP-dependent ligase.  I guess we should suggest one when we discuss this gene in our knockout-mutant paper (ligB? ligC-for-competence? ligK because it's a k-family ligase like those of eukaryotes?).  The problem is made worse because some annotators have erroneously labelled ATP-dependent ligases as NAD-dependent ligases, and I suspect that some bacteria have genus names that don't reflect their true relationships (certainly in the Pasteurellaceae).

There's also uncertainty about the N-terminus of the protein.  It's important to have the right start codon because the N-terminus sequence determines the localization predictions.  Some Pasteurellaceae homologs are predicted to use a start codon that's 36 aa upstream of the predicted H. influenzae start;  H. influenzae can't use this upstream start codon because there's an intervening stop codon.

I used PSORTB to look for evidence of targeting to the cell envelope for 9 representative proteins.  Four were strongly predicted to be cytoplasmic: all are Pasteurellaceae (G. anatis, H. somnus, H. parainfluenzae and A. aphrophilus).  The other five were targeted to the envelope, but the program could only assign a specific location for Shewanella putrefaciens (cytoplasmic membrane); the other four could have been ineither membrane, in the periplasm, or extracellular (H. influenzae, V. cholerae, N. gonorrhoeae and N. meningitidis).

The gene has been knocked out only in H. influenzae, first by VanWagoner et al. and now also by us.  The mutants grow normally; VanWagoner et al. reported a small transformation defect but we saw none, both in our new marked and unmarked mutants and when we introduced their mutation into our KW20 background.  My Google searching turned up a poster from a research group at AstraZeneca; they showed that the H. influenzae NAD-dependent ligase is essential, and concluded that its function can't be replaced by the ATP-ligase.  But maybe the results would have been different if they had overexpressed the ATP-ligase, as it's under CRP-S regulation and levels may be very low in noncompetent cells.

Bottom line:  We don't know what function(s) this protein serves in any bacterium and its phylogenetic distribution is weird.  Our finding that it's in the H. influenzae competence regulon is the only clue to its function in any species, and figuring out what it contributes to competence would be a big advance.  So we're going to try to complete the former undergrad's transformation tests.

A refresher class on streaking for single colonies

I think we might benefit from an explanation of how to streak bacterial cells on an agar plate to get single colonies.

The problem:  You want to begin your experiment with cells from a single colony so you know they're a genetically pure clone.  You have a source of bacterial cells whose density you don't know (an overnight culture, a colony or patch of cells on an agar surface, a frozen stock), and you want to spread them out on a fresh agar plate so some of them will grow up to be well-separated individual colonies.

The solution:  You do what's really a solid-phase, non-quantitative serial dilution.  You start with an agar plate and a quantity of the cells on a sterile inoculating loop or the tip of a glass pipette.  It really doesn't matter how many cells you start with - anything between a whole colony and a drop of dilute culture will work fine.  You spread these cells onto a small area of the plate; the objective is not to spread out all the cells; just some.   You then pick up some of the spread cells by passing the resterilized loop or fresh pipette through this area, and then wipe some of these cells onto a fresh area of the plate by streaking the loop in a zigzag.  Again the objective is to transfer a small fraction of the cells (less than 1%) onto the new area.  You repeat this once or twice more.  

The grey lines in the drawing show the path of each pass, and the blue is the cells after overnight incubation.  It doesn't really matter how dense the cells originally were; if the transferred fractions are small enough, the third pass will spread out only a few cells, and these will grow into well-separated colonies overnight.  If the cells were not very dense originally, the third pass may not give any colonies, but the second pass will give well-separated colonies.

A common error:  Beginners often forget that the point is to drastically decrease the number of cells being spread in each pass.  In the photo above we see that the overly meticulous researcher has done seven (!) sequential passes and still wound up with many crowded colonies in the last streak area.  The problem is that each pass picked up a large fraction of the cells spread by the previous pass - you can see that there's very little decrease in colony density from one pass to the next.  The reason each pass picked up so many cells is that each 'zig' of the zig-zag went back into the previous pass, picking up more cells each time.

Later:  here's an example of a good streak.  Note that this and the above are both done on small (60 mm) Petri dishes rather than on the standard 90 mm ones.

A norte on what tool to use for streaking:  I use sterile 0.1 ml glass pipettes, rotating 180° after the first pass and then using the still-sterile other side.  (We got lots of these free because of a supplier error.) As The Lorax points out in the Comments below, you can do the same thing with skinny wooden dowels (like long cylindrical toothpicks), but these and toothpicks have rough surfaces and don't slide as smoothly over the agar surface.  Disposable plastic loops work fine, but you can go through a lot if you're doing serious genetics.  The best non-disposable loops are made of platinum - these are quite expensive ($50?) but retain their smooth surfaces through many thousand cycles of flame-sterilization.  The cheap steel loops become very rough after even a few flamings.  As a grad student I developed a custom handle for my platinum loop to get exactly the right degree of flexibility and balance of weight (yes, I did a lot of streaking then).

Back to the bench

OK, the RA has sent in the revisions of her E. coli paper, the former visiting grad student is working on her G. anatis revisions, and the phosphate measurements of the #arseniclife culture medium are underway.  But we still need to do quite a bit of work for the manuscript about our new collection of H. influenzae competence-gene knockout mutants, which we want to submit before the RA goes on a few months' parental leave in six weeks.

First, we want to use the lab next door's BioScreen machine to do growth curves on all the 'unmarked' (clean deletion with no insertion) mutants.  The plan is to dilute cells directly from fresh colonies into culture medium that's out into the wells of the BioScreen plate, but before we set up big screens with lots of mutants I need to check that this will work as planned.  So this morning I'm going to dilute a few wildtype colonies and measure the cfu/ml (by plating); we'd like to start with about 10^6 cfu/ml, as this should give dense growth overnight if the growth rates in the BioScreen are like those in our incubator.   On Thursday I'll streak out a few mutants and on Friday set them up in the BioScreen.

Second, we need to do a few more transformation assays and a lot more DNA uptake assays.  For these I've promised to make all the competent cell preparations - I have a big list above my bench.  I should be able to get these done over the next week or so, if I get my act together.

Third, we'd like to be able to include some follow-up analysis of interesting mutants - otherwise the paper is just a dry list.  One interesting mutant is comE1.  Homologs of this protein are present in all competent species, and it's always found to be essential for DNA uptake and transformation.  But the H. influenzae mutant has only a 5-10-fold defect.  (Tenfold sounds big, but transformation can usually be measured over at least several orders of magnitude.)  We don't know why we see so much residual DNA uptake.  We'll present a detailed analysis of the homologs in the different species, and an experiment testing whether the Rec2 function is responsible for the residual uptake.  The postdoc and I did one experiment testing this last month, so I'll make more competent cells and we'll replicate it. 

Another interesting mutant is the ATP-dependent DNA ligase that's predicted to be in the periplasm (see and and  We have some data from experiments done by a former student in the Honours program, working with a mutant that had been created by another lab.  She tested whether transformation of the ligase mutant was more sensitive to nicks in the DNA (it isn't), and did all the preliminary work for testing whether ligation occurs in the periplasm.  In this experiment competent H. influenzae rec2-mutant cells are given a USS-containing E. coli plasmid that's been cut with a restriction enzyme, and the plasmid DNA that's been taken up is then recovered from the H. influenzae cells and used to transform E. coli.  Only plasmid that's been ligated will give E. coli transformants.   She didn't have time to do the final experiment, testing the effect of the ligase mutation, and we can probably make the experiment much more sensitive by using the postdoc's periplasmic DNA prep and using very competent E. coli.

We've received the #arseniclife reviews from Science

They're too long to include in a blog post so I've posted them here.

Bottom line:  the reviews are largely favourable so our manuscript is provisionally accepted!

The main concern of the referees is the growth issues I've written about here: the cells would not grow in the medium specified by the original authors (I had to add glutamate) and the medium I used was supplemented with 3 µM phosphate and it's basal phosphate contamination had not been measured.

Most of the issues can either be just clarified in the text or declared to be beyond the scope of this work, but we're going to try to directly measure the basal phosphate contamination in the medium.  (We're worried that previous analyses of phosphate-buffered materials may have decreased the sensitivity of the LC-MS system we used.)

Of course we'll also address the comments on the manuscript posted by readers of this blog.  And we'll post the complete Response to Reviewers here.

Publication progress...

We still haven't heard back from Science about our January 30 #arseniclife submission.

We haven't heard back from PNAS about the postdoc's February 3 uptake-bias submission.

But we have good news about the RA's E. coli-competence submission to PLoS One.  It's officially a 'major revision' but she's already done the control experiment they ask for, and the rest is just minor rewriting.

And we have good news about the visiting grad student's Gallibacterium anatis-competence submission to Applied and Environmental Biology.  The reviewers said nice things and asked for only very minor changes.

And we have lots of progress on the manuscript front:  The RA has given me a draft of her new manuscript describing all her work creating a complete set of competence-gene knockouts for H. influenzae, and the postdoc has given me a draft of his new manuscript on H. influenzae recombination tracts.  And I'm nearly finished with my article for PLoS Biology about teaching genetics in the 21st century.

Any day now I'll get back in the lab and start making competent cultures of the remaining knockout mutants (the RA and postdoc have posted a big list above my bench).

A long post about Hfq and Sxy

I've been doing a lot of digging around and reading about Hfq's possible involvement in post-transcriptional regulation of sxy.  Now I need to write it all down here or I'll forget what I've done.

1.  H. influenzae Sxy is a great in vivo system for investigating Hfq activity:  A commenter on the previous post (thanks, Mike!) recommended doing in vitro assays for Hfq's ability to bind to sxy mRNA.  But, although we've done in vitro RNA work in the past, it certainly isn't our strength.  Fortunately, our in vivo transformation assays let us detect changes in Sxy activity over at least 6 orders of magnitude, because fully competent cells have a transformation frequency of ~10^-2 (higher if we use a pure marker fragment rather than chromosomal DNA), and the threshold of detection is <10^-8.  The secondary structure has been confirmed experimentally, and we also have mutations that affect the secondary structure of sxy mRNA, some that destabilize the main stem and increase transformation frequency dramatically and some that stabilize the stem and reduce transformation frequency dramatically.

2.  In Vibrio cholerae, post-transcriptional activation of the sxy homolog tfoX depends on both a small RNA (tfoR) and Hfq (Yamamoto et al. 2011. J. Bacteriol. 193:1953-1965, doi:10.1128/JB.01340-10).  This is consistent with Hfq's usual role in facilitating adaptation to stress, and predicts that loss of Hfq in H. influenzae will decrease (or even eliminate) transformation.  This couldn't be tested directly in V. cholerae because of other harmful effects of the Hfq knockout.  I'm hopeful that this won't be a big problem in H. influenzae, since Hfq is unlikely to affect as many processes.  H. influenzae has a much smaller genome; it doesn't do quorum sensing and doesn't regulate catabolite repression with a Hfq-dependent spot42 small as E. coli and V. cholerae do.

I've gone through the list of H. influenzae RNA sequences on the Rfam server (thanks, Paul, for the link!).  There are only about 100 of these, and most are rRNAs and tRNAs.  The rest all have predicted functions that are unlikely to affect competence.  We don't  have any RNA-seq results yet (though this is planned to happen soon), but there's a published report of RNA-seq analysis in the related Haemophilus somnus (renamed Histophilus somni by the evil Danish group) which includes a list of small RNAs, tagged with their distributions.  I grabbed the sequences of the eleven of these that are present in other Pasteurellaceae, and tested them for complementarity to H. somnus sxy mRNA using the IntaRNA server from U. Freiburg.  One of them (HS29) showed quite strong pairing.  HS29 has homologs in most Pasteurellaceae but not in H. influenzae, which is a drag, but at least this analysis suggests that a similar small RNA might interact with sxy mRNA in H. influenzae.

So the RA has just ordered the oligos she'll need to make a knockout mutant of hfq (HI0411).  I'll transform the mutation into wild type cells and into cells that are hypercompetent due to mutations that destabilize the main stem of sxy mRNA.  The V. cholerae precedent predicts a decrease in competence; if I don't see this, Hfq probably doesn't play any role.  If competence is down I may be able to rule out effects just due to changes in growth properties by comparing the effects in wildtype and hypercompetent backgrounds.  I might use the murE hypercompetent background as a control, though it's also possible that a murE-associated small RNA mediates the Hfq effect.   Just be sure I'll also check for increases in competence using one of our low-expression sxy mutants.

Pause while I spend two hours trying to see whether part of the murE mRNA, or an antisense RNA to it, could pair with the regulatory part of sxy mRNA.  Hmm, tantalizingly. it does.  This would be more exciting if the pairing segment included the site of one of the mutations known to cause hypercompetence...  Anyway, here's a figure:

Might Hfq regulate sxy translation?

I've just returned from a visit to the University of Western Ontario, where I gave two talks, one about #arseniclife and one about competence.  In the competence talk I briefly described how sxy translation is limited by the secondary structure of its mRNA, and said that we don't have a good understanding of the mechanism.

Later I had lunch with some grad students, and a couple of them told me that they'd noticed a potential binding site for the protein Hfq in the sxy mRNA secondary structure.  Hfq contributes to gene regulation by helping small regulatory RNAs (sRNAs) find and bind to their target mRNAs.  My earlier discussion with their supervisor had reminded me that we've never investigated whether Hfq plays a role in sxy regulation.  What the students noticed is that a loop in the structure ('A' in the figure) exposes the sequence AAUAAU, which is most of the Hfq motif identified by a recent SELEX study (citation below).

I think the first thing we should do is knock out the H. influenzae hfq gene.  The authors of the SELEX study say that "Hfq is not essential for growth, but the adaptation to changing environmental conditions is hampered in the absence of Hfq."  That would fit with Sxy's function.  I could imagine the mutation either increasing or decreasing sxy expression and thus the transformation frequency, depending on how Hfq interacts with the sxy mRNA.  Provided it doesn't have dramatic effects on viability or cell growth, we should be able to detect specific effects on transformation frequency.  We could then mutate one or more of the positions Hfq is predicted to interact with, but that would be more difficult, and first I'll need to read a lot more about Hfq.  Of course, long strings of As and Us are very common in H. influenzae mRNAs because of its AT-rich base composition, so I won't get my hopes too far up until we see a mutant phenotype.

The students' supervisor gave me the url for a web site that will, I think, search a genome for sRNAs complementary to a given mRNA.  I've lost the url, and all the websites I can find do the complementary search (start with small RNA, identify possible targets), so I can't check this until he sends me the url again by email. 
C. Lorenz et al.  Genomic SELEX for Hfq-binding RNAs identifies genomic aptamers predominantly in antisense transcripts Nucl. Acids Res. (2010) 38(11): 3794-3808 first published online March 26, 2010 doi:10.1093/nar/gkq032