Field of Science

Who writes this drivel?

My institution has a new fundraising campaign.

Its slogan is 'Start an Evolution'*.  The text accompanying it reads: 
UBC generates ideas that start evolutions.  Ideas that change the way people think and the way the world works.  We see this change as an evolution, one that improves upon what has come before and inspires the generations that follow.
So much for scientific literacy in the Development Office. 

 - - - - - - - - - - - - - - - - - - - - - - - - - - -

*Oops, 'start an evolution'

Let's try growing GFAJ-1 in glass flasks at 28 °C and 30 °C

For credible DNA preps I need reproducible growth with and without 40 mM arsenate, and I need to be able to grow large volumes because the cultures with only 3 µM added phosphate won't reach a very high density and thus won't yield much DNA per ml.

One practical limitation to my GFAJ-1 experiments is that it's optimum growth temperature is 28 °C but we only have two shaking water-baths, one that's always at 37 °C and one that currently often needed at 30 °C.  But Halomonas bacteria typically tolerate a wide range of growth temperatures, and there's 30 °C room down the hall with shakers in it.  So tonight I'm setting up pairs of replicate cultures without arsenate, one to shake at 28 °C and one to shake at 30 °C.  If they grow at about the same rate and to the same density I'll do my growth tests in the 30 °C room and only switch to our own shaking water-bath at 28 °C when the arsenate-resistance problem has been sorted out and I'm ready to generate good growth curves and DNA preps.

So...  Make up a fresh batch of medium, add cells from a freshly thawed tube of GFAJ-1, and put into replicate flasks with different amounts of phosphate.  Do I have enough glass flasks?  Yes, there are 12 on the sterile-glassware shelf.

Could the glass be sequestering the arsenate? (I doubt it, but...)

Commenters on my latest post suggest that the glass tubes might be somehow sequestering the arsenate.  I've been assuming that the surface area of the glass is much too small to bind up all the arsenate in a 40 mM solution, but maybe I should do a calculation to check this.

First, how many arsenate ions would it take to completely coat the inner surface of one of the glass tubes I'm using?  The tubes are about 10 cm long, with an inner circumference of about 2 cm, so that's about 20 cm^2 of glass surface.  This paper says that the ionic radius of arsenate ions is 0.248 nm, so I'll assume a diameter of 0.5 nm.  If the ions were to pack squarely side-by-side onto the glass surface, there would be about 2x10^3 per µm, or about 2 x 10^6 per mm.  That's about 4 x 10^12 per mm^2, or 4x10^14 per cm^2.  So if the inner surface of a glass tube were densely coated with a monolayer of arsenate ions, it would sequester about 8x10^15 ions.

Next, how many arsenate ions are in 5 ml of a 40 mM solution?
6x10^23 ions/mole times 0.04 moles/liter times 0.005 liter = 1.2x10^20 ions.
So less than 0.01% of the arsenate ions in my medium could be tightly packed in a monolayer on the inner surface of a glass tubes.   The assumption of dense packing is very conservative, so I don't think sequestration of arsenate by the glass can be the explanation for the GFAJ-1 growth I'm seeing.  But thank you to the commenters for prompting me to do the calculation.

Latest GFAJ-1 results

The design of this experiment is described in the previous post.  

The upper graph shows that, in the absence of arsenic, growth is phosphate-limited and nicely reproducible.  The red lines are replicate cultures with no added phosphate (two in glass screw-capped tubes, one in plastic (square symbols) ); the blue lines are replicates with 3 µM phosphate added (three in glass, one in plastic) and the purple ones are replicates with 1500 µM phosphate added (two in glass, one in plastic).  

The lower graph shows that growth is much less reproducible in the presence of 40 mM sodium arsenate.  The blue, red and purple lines are the same phosphate treatments as in the left panel; again square symbols indicate cultures in plastic rather than glass tubes.  The most striking result is, as seen before, that cultures in the plastic tubes (the three lowest lines) grew very little, even with abundant phosphate.  The two no-added-phosphate cultures in glass tubes (red) grew identically and slightly better than their no-arsenate controls, as did two of the three 3 µM phosphate cultures (blue).  The third 3 µM phosphate culture grew to only half the density of the others.  The two 1500 µM phosphate cultures (purple) grew to high density, one slower than the other.

The growth differences between replicates are unlikely to be due to differences in inoculum size, since all cultures began with 10^5 cells/ml.

I have no idea why growth in polypropylene tubes makes cells arsenic-sensitive.  Googling 'arsenate' plus 'polypropylene' didn't suggest anything.

Maybe now I'll try cultures in glass flasks.

Why doesn't this post have a title?

Regular readers will know that my attempts to grow GFAJ-1 in medium with 40 mM arsenate have given very inconsistent results (Expts. 1-3Expts. 1-4Expt. 5Expt. 6).

The Wolfe-Simon et al paper reported that these cells were resistant to 40 mM arsenate, but in my experiments so far the only time the cells really appeared to be resistant to 40 mM arsenate was Expt. 5, when I grew them in screw-capped glass tubes.

Even if I set aside all the other experiments (in foil-capped flasks or in screw-capped plastic tubes), this experiment was compromised by tube-to-tube inconsistencies in final cell density, probably due to the presence of some limiting nutrient (perhaps not phospate) contaminating some of the tubes.  So I've now acid-washed all the glass tubes and caps, rinsed them lavishly in distilled water, and re-autoclaved them, so I can see if this experiment's result is reproducible.  These conditions are closest to those used by Wolfe-Simon et al., so if I can consistently get growth in 40 mM arsenate I can prepare DNA from arsenate-grown and control cultures for mass-spectrometry analysis.

I'm going to streamline the number of conditions (just ± 40 mM arsenate, combined with no added phosphate, 3 µM phosphate or 1500 µM phosphate) and do two replicates in glass tubes and one in plastic tubes.  I'll make up 50 ml of no-phosphate medium, add cells (from my frozen stock of phosphate-depleted GFAJ-1) to about 10^4 cfu/ml, and split this in two.  I'll add arsenate to one half (and water to the other).  Then I'll put 5 ml into each no-phosphate tube, add phosphate to 3 µM to the rest (both parts), and put 5 ml into each 3 µM PO4 tube (using 3 replicate glass tubes).  Then I'll add more phosphate to the rest, to 1500 µM, and put 5 ml into each 1500 µM tube.  Then I'll put all the cells gently rocking in the 28 °C incubator.

Not as busy as I'd like to be

I had expected to spend the first half of September frantically polishing our latest resubmission of our  grant proposal on DNA uptake by Haemophilus influenzae.  But I missed a pre-registration deadline, and the Canadian Institutes for Health Research (CIHR) offers no recourse.

Applicants are supposed to register their pending proposals a month in advance, including a one-page summary of the proposal and suggestions for the appropriate review panel.  In the past I've always done this at the same time I signed up to have a draft of the proposal go through the internal review process my university offers.  But this time around I didn't think it needed another round of internal review - the writing and presentation is already very good, and the only weakness the reviewers found was in the proposed experiments.  So I didn't sign up for internal review, and didn't remember the pre-registration step (due August 15) until earlier this week, when I went online to start working on revisions to the budget. 

I've now spoken to our Research Services office, to a CIHR administration, to a local colleague with lots of CIHR administration expertise,a nd to a distant colleague who made the same mistake for a previous deadline.  All agree that there's no recourse - I'll have to wait until the March 1 2012 deadline.

Well, this gives us plenty of time to generate the new preliminary data that will address the reviewer's request that we use point-mutation mutagenesis to analyze the functions of competence genes.

Science Online London workshop on "Beyond Scholarly Publication"

The blurb for this workshop says:
This workshop will tie together a number of concepts raised at last January’s “Beyond the PDF” conference, looking at how we can move beyond a static PDF journal article and can redefine both our writing tools and the format of the scholarly paper. This workshop will showcase Scholarly HTML and participants will learn to use blogging tools to write content that is interesting, enriched with multimedia, collaborative, and semantically enhanced.
OK, they're giving us a Wordpress blog to work on, or we can work on our own blogs (but they say that maybe some tools won't work in Blogger...)  I'm trying to use the Wordpress version (the 'sandbox) but it's just hanging even though there are only about 75 people in the room.  Blogger has no problem connecting, so maybe it's just this many people trying to access a single blog.

I'm also not clear about what we're supposed to be doing.  Writing content about Spinal Muscular Atrophy, I think.
I could write about how I might use SMA as an example in my genetics class.  It's excellent in many ways, because it can be presented simply as an autosomal recessive, but then I can introduce complications that lead to a more complex understanding of the relationships between genotype and phenotype.

Here's an image taken from a website (yes, without permission).  It shows how the amount of SML protein (the  gene product needed for normal phenotype) depends on which alleles a person has, for two genes, SNM1 and SMN2, one of which, SMN2, is poorly expressed and may be duplicated.

Unfortunately the figure is not very clear (I think it must have been prepared for a different context, one with some explanatory text).  The first two genotypes are OK - the healthy individual is shown as having two functional copies of SML1 and of SML2, and the Type I as having two defective copies of SML1 and one defective and one functional copies of SML2.  But  the phenotypes don't make sense.  The amount of SML1 protein in a normal person is shown as 100%, but the amount of SML2 protein is shown as 20%.  100% of what, and 20% of what?  If the SML1 level is 100% of normal SML1 protein, then the SML2 protein level in a normal person should be 100% too.  If we instead think they might mean 100% of total (SML1 + SML2) protein, then the total SML1 protein can't be 100% because there's also a contribution from SML2 (the total would have to be 120%).

It gets more confusing when we look at the Type II individual.  Here the defective allele of SML1 is shown aligned with and thus allelic to a functional copy of SML2.  Well, maybe this genotype arose by a gene conversion event that replaced one SML1 allele with a paralogous SML2 allele...  Indeed, when I looked at a pdf about SMA genetics, written for parents, I found statements that gene conversion is frequent and that many patients have replaced one or both copies of the SML1 gene withSML2.

While I've been writing this the moderator of the session has been showing us how to collect and insert a reference list into a Wordpress blog.  I wasn't paying enough attention to see how he did this - he used a Wordpress plugin, but there was quite a bit of muttering about the site being slow.

Now he's going to show us another plugin.  e-Pub???  The text on the screen is just too small to read - he embiggened it once, but it seems to have shrunk back.  He says ePub is nicer than pdfs, but I've no idea what it does or where to find the plugin.

Preparing for Science Online London

In a few hours I'll be part of a Science Online London panel titled:
Linking with the Literature – the Arsenic Story.  How to engage with the peer-reviewed literature: strategies for fellow researchers, science journalists and bloggers. A look at how this was handled in a controversial example from last year – the #arseniclife story.
I 'm trying to figure out what I'll say in my introductory bit.  Our leader has given us four questions to guide our preparations.

The first question is:
1. The #arseniclife story - what deficiencies in engaging with the literature did this case make obvious?
So I think I'll start by considering what researchers might hope to accomplish by writing online about specific peer-reviewed papers.  We can spotlight work we think deserves more attention, and point out problems with work we think flawed.  We might hope to change how the published results affect future work, or what the public thinks science has discovered.

Then I'll do a very quick reminder of the #arseniclife story, emphasizing not the science but the responses of the media, the public, and other researchers. 

I should end my introductory bit by considering the 'deficiencies' of the online responses.  This requires hindsight - what do we wish the online responses had accomplished?  Since everyone but the paper's authors agrees that the big claimed result is false, the responses should have counteracted the media blitz (corrected public understanding), and discouraged researchers from following up on the result.  I think they only partly succeeded at both of these.  Only a very small fraction of the original audience will have seen the criticisms and taken them seriously.  And lots of scientists still refuse to think about material that isn't peer-reviewed.  Is this due to deficiencies on the online responses???

The next two questions will be easier for me to answer:
2. (briefly) Is this only an issue in the life sciences, what are the experiences in the physical sciences?
I think I'm off the hook for this one.
3. Is the peer-reviewed system employed by journals too slow and/or not open enough to deal with such cases? (this might be particularly the case for retractions, which e.g. often are not motivated by editors).  Similarly, what are the online alternatives that have been developed?
Too slow to accomplish what?  It's  not too slow to sort out bad science from good, but it might be too slow to save researchers from wasting their time building on the faulty results.  It's definitely far too slow to counteract the kind of media blitz that occurred with #arseniclife.

The final question is much too big for my limited brain.
4. Possible scenarios for the future. What are the implications and consequences for the various stakeholders (scientists, publishers, journalists/bloggers)?
Can I come up with even one consequence of online engagement with the peer-reviewed literature?  Well, if you're a scientist hoping to incorporate an exciting new paper's results into your planned research you should begin by checking the online response to it.  If you're a publisher (or Google Scholar), providing an easy way to track the online responses to articles would add a lot of value to papers (supplement the 'Cited by' list of links with a 'Discussed by' list).  If you're a blogger, make your posts easy to find.