Positional cloning of a mutant using the classical mapping strategy remains a bottleneck. Recent advent of whole genome sequencing offers solution to the part of the problem. However, typical whole genome sequencing provides hundreds and hundreds of sequence variations that it is tantalizing to associate every molecular lesion with the phenotype under investigation. Under these circumstances, an investigator would attempt to rescue the mutant by introducing genomic DNAs (Cosmids, fosmids or Yeast Artificial Chromosomes) corresponding to the locus of mutation.
Generating transgenic line harboring a genomic DNA is a challenging task for two reasons: first, there is a bit of a learning curve in mastering the art of microinjection. Second, not every lab has an access to needle puller, micromanipulator and other rather expensive paraphernalia required to generate transgenic worms. This is especially true in the research institutes where predominantly undergraduates participate in research. In fact, the very requirement of having to generate transgenic worms may dissuade such institutes to carry out any forward genetic screen in the first place.
Availability of transgenic lines carrying defined pieces of genomic DNA would circumvent both of the aforementioned problems and would be an invaluable tool for the C. elegans community. One can simply cross transgenic line with the mutant of interest (or vice versa) and obtain a homozygous mutant carrying the wildtype copy of the transgene to see if it can rescue the mutant phenotype.
Many labs around the world generate several transgenic worms carrying defined piece(s) of genomic DNA for the rescue experiment. Typically, when a researcher X generates several lines, only one (or subset of them) rescues the mutant phenotype and rest of the lines are either thrown outright into a trash can or kept frozen in their lab freezer. Let us assume researcher A finds that the locus of his/her mutant lies within the same region. It would be far more convenient for the researcher A to borrow the lines from researcher X than to repeat the same process all over again.
I suggest a system that would facilitate the transaction of transgenic animals carrying defined piece(s) of genomic DNA between labs. First, we should create a website CTGC (Caenorhabditis TransGenics Center). If a ‘donor lab’ has a transgenic line, s/he should visit CTGC website and enter details such as strain name, genotype, description, etc. The submitted information will be stored in a searchable database. The consumer lab can search this database by using key word, such as Cosmid name, which can then be directly requested through CTGC by completing a simple form.
Creating CTGC website is relatively simple process. However, with the steady demand of transgenic strains from across the globe, many PIs may become reluctant to provide this service in long run, which is quite understandable. Between writing grants, papers, discussing with students while the tenure / promotion clock ticks, thawing out that transgenic strain and shipping it to some other lab is understandably not a priority. Should some transgenic worm become as popular as daf-16 or daf-2 mutant, it is easy to imagine the work-load of the lab of its origin!
So, I suggest a policy in which, by submitting strain information, the donor lab is expected to distribute the strain to only 3 labs. Any subsequent request for the strain will be redirected to the recipient labs. The lab that made the strain won’t be held responsible for distributing the strain indefinitely! It is obligatory that the lab that receives the transgenic strain should ship it forward to at least 3 of the subsequent requests. The strain request will be handled in this way for the rest of the requests.
In order to keep track of the strain distribution, the entire strain request has to be processed through CTGC, which automatically registers the details of all the researchers who received this strain.
In the hypothetical example, the transgenic line is originated from Singson lab. Singson lab has distributed this strain to 3 labs (Ahnn, Soto and Koushika labs). Therefore, the system will no longer show Singson lab as a potential source for this strain. I believe that C. elegans community will be immensely benefitted from CTGC and I look forward to your comments and suggestions to improve this concept.
Guna,
just a few observations on your suggested transgenics registry. As someone now in the process of spending a considerable amount of money setting up a biolistic system to generate transgenic worms I empathise with researchers on limited funding faced with the same decisions. So, thanks for your positive ideas.
To my observations on your text;
-
This topic sits a little uncomfortably between scientific discussion and methods, so not sure whether this section is the best to attract lots of replies…
-
Not sure of the logic behind the first paragraph. If a PI/group has identified the location of their mutant locus by GWS then they likely have the funds to take on the next stage of creating rescues don’t they? I can think of other situations where this database would be more useful (see last paragraph).
-
I agree that not all labs have the equipment and/or the expertise necessary to consider microinjection and the former made me choose biolistic bombardment. Any database of transgenic worm strains should also include those generated by biolistic bombardment.
-
We’re talking about integrated transgenic strains here? I wonder to what extent any transgenic strains not rescuing the mutant phenotype would be chracterised further or even processed by a lab such that it was suitable for shipping to other groups? If the choice rests between throwing the non-rescuing transgenic strain out or spending an additional week or so setting up and testing for an integrated but ultimately not useful transgenic strain?
However, I can see the usefulness of characterised, integrated lines being available to rule out particular regions of genomic DNA as the locus for newly isolated mutant phenotypes.
- The ‘3 labs and you’re out’ idea is a good one. Perhaps this would encourage more labs to play ball and send out transgenic worms (sorry, too many baseball puns).
Even where (according to the CGTC) the original lab is not sending the transgenic strain out anymore, the entry should carry a note of the originating lab for future reference. This avoids the last lab to send the strain out being cited as the originating lab.
That said;
I think the idea of systematising the current ‘Lab A contacts Lab B’ for a transgenic strain covering a specific genomic sequence is a good one as it complements the CGC and offers labs on restricted funds or those designing undergraduate teaching the access to both mutant and rescue strains that can be further characterised using readily available and affordable techniques.
Steve
I don’t think this is a very sensible system. I’m all in favor of sharing reagents, but there are a whole bunch of places where this falls down. I’ve focused on the genomic-rescue side, as I believe the designed-transgene side is a bit moot. WormBase curators already try to generate database records of those transgenes (typically gfp, targeted expression constructs, and the like) that are described in the literature, and anyone so publicly minded as to report the existence of a transgene they’ve created for inclusion in WormBase can do so via a handy online form.
And, focusing mainly on the genomic clones, there are a number of issues here:
- It’s a big genome out there. I don’t have the number of clones in the sequenced (ie minimally redundant) physical map ready to hand, but you can take for example a paper from Ann Rose’s lab in which they use 250 cosmids to cover less than 1/12 of the genome. Sure, the odds get a bit higher that someone will have used a relevant cosmid before if you’re in one of the gene-rich clusters, and you are disproportionately likely to be in a cluster. Even so, we’re talking about a database that’s unlikely to be useful unless people report to it hundreds if not thousands of lines containing cosmids.
- If they have injected a cosmid you’re interested in, they probably did so for a reason; they probably injected into their mutant background. Getting their transgene into your mutant background is going to be work - quite likely more work than doing the injections, if you can get access to a microinjection setup.
- Speaking of which, I don’t think it’s really fair to cite the learning curve for microinjection; access to the equipment can be a real problem, but assuming you do have access you can learn to microinject in one thoroughly miserable afternoon, or perhaps in two. You have to figure that any strain request you make is at least an hour of someone’s time, especially as lines containing extrachromosomal transgenes require some extra attention prior to shipping, so an attempt to avoid taking the time to learn microinjection amounts to trying to get others to spend their time, out of their generosity, rather than spending time of your own.
- Effort spent making such a resource looks backwards technologically. Time was, physically linked mapping markers were sparse and so people had to inject pools of cosmids and hope for the best. Now we have essentially unlimited mapping resolution with SNPs, and whole-genome sequencing gets more affordable all the time. You’re less likely to hope for the best and try all the clones covering a large region than you once were; this means people are less likely to inject multiple cosmids, or possibly they may inject no cosmids at all, preferring to map to a small number of genes or to identify candidate mutations and to amplify a candidate gene or genes by PCR. The disincentives to generate significant numbers of cosmid lines are only going to increase.
- The reporting requirement to make this work is time-consuming, and problematic. When do you report in the database that you’ve injected a cosmid? Do you do it before you’ve published the relevant work, when you might not want to say which cosmids you’re working with, let alone to distribute your mutation? Do you do it after publication, long after the lines were generated? How many people are going to bother to report their injections to the database; even less likely, is anyone going to dig through their lab’s list of frozen transgenic lines and note which are available? And if we did want to generate a registry of available genomic rescue lines, wouldn’t it make more sense to do so through WormBase, where the lines could easily be linked to the genomic clones they contain?
- Inverting the previous point, if you have a particular interest in a region, why not just write a post asking people if they have any lines containing genomic clones in that area, or can suggest someone who might? Why not spend some time looking up the papers reporting the cloning of genes close to your locus, and contact the responsible labs to see if they saved the transgenic lines they used?
- The three-labs-and-out distribution model is fine, and similar methods have been used in the past, for example to distribute the Fire Vector kits before Addgene existed. But there exists already a central resource to store and distribute C. elegans strains. The CGC has the facilities, the expertise, the database, and the funding - and it actually helps them make their case for continued funding if they can demonstrate they’re fulfilling a need. Cobbling together a distribution network to circumvent the CGC seems like misplaced priorities. and any clone that is in sufficient demand to require a second tier of distributing labs would probably be better made available through the CGC.
I salute the community spirit behind this idea, and I hope people are saving the transgenic lines they generate in case they are later useful to someone, to anyone. But I don’t really think the idea of making a new and separate database and a new and separate distribution system is going to work.
Given all the issues that Hillel’s post raises, perhaps his last point about supporting the CGC via some sort of list of strains containing genomic fragments held by them is the most important to focus efforts on
We are all used to the CGC being there to hand out strains for a still unbelievably low price. For newer labs, the CGC is the only obvious source of mutant and wildtype strains. If the CGC goes under then we’re back in the dark ages…
Perhaps we can work towards getting such a list on the CGC website?
Steve