Hi everyone, I would just like to apologize for the delay in posting the responses and say thank you to everyone who offered advice, it was very much appreciated. Deborah INITIAL EMAIL Hi, I have been developing microsatellites for 3 species of Galatheidae for the past year and although the developmental process went well and plasmid sequencing gave microsatellites and flanking regions which were sufficient to develop primers I am now having problems with optimisation and amplification. I am somehow amplifying product of incorrect size (as to what the primers were designed for- larger fragments) and am getting multiple bands and lots of background noise when trying to screen samples. Of the two micros that I have managed to optimise the screening concluded that they were uninformative (everything I screened including 19 adult individuals from a single area as well as a mother and 75 eggs where all found to have identical heterozygote bands!). I am confused by this result and was wondering if anyone has had similar problems with crustaceans or could offer advice as to possible problems that could be causing this or how to proceed. Thanking you in advance for any comments you could offer Deborah VERY KIND AND MUCH APPRECIATED RESPONSES: * I haven't much experience with crustaceans but I would've thought that mining non-unique sequences by chance was very unlucky. I realise that enrichment success is species dependant but I need to know more about your protocols before I can offer advice. Have you tried SSCP with your 'monomorphic' primers to check whether there's any conformational polymorphism? Perhaps you could look on the database for EST sequences to see whether you can design SSR markers from them. Jake Jake Clarke (EMR) [Jake.Clarke@emr.ac.uk] * Have you tried a HotStart Taq, touchdown, and touchup PCR??? Cheers, Fabio Fabio M. Diniz [F.M.DINIZ@soton.ac.uk] * I have experience with plant and insect microsatellites, but not Crustacean, so I can not really comment if your amplification problems are species-specific... Nevertheless, some hints perhaps: Although you have not mentioned the PCR conditions, I suppose you have tried selective conditions (e.g. higher annealing temperature), in order to amplify only specific bands. So, if you do get multiple bands or incorrect size, can be that your primers match with more sequences and so amplify more/different loci, in fact. Perhaps, the best to check on that is to cut out the suspicious bands from the gel and sequence them. Thus, you can precisely check whether you have amplified microsatellites. If so, you may then decide to try and re-design the primers. Also, you have not specified what is difference from expected and your observed length ...my own experience is that some microsatellite loci can amplify alleles within relatively broad range like 90 bp or so. Zavodna, Monica [M.Zavodna@nioo.knaw.nl] * you have to separate if the problem is: 1/ technical and general 2/ technical to do with this organism 3/ non-technical to do with this organism, ie a property of this organism 1/ can be ruled out if you can borrow some microsatellite primers and DNA from someone in the lab working on some other organism and get scoreable gels with your reagents and equipment. 2/ you might approach by seeing if you can get sensible results using nuclear primers that are supposed to work on your taxon, or at least related crustacea. If successful this will tell you that there is nothing too wrong with your DNA for example. Jarman SN, Ward RD, Elliott NG (2002) Oligonucleotide primers for the amplification of coelomate introns. /Marine Biotechnology/ *4*, 347-355. is a good potential source. You might develop some useful primers anyway that could supplement microsatellites 3/ some organisms are just hard. you probably know the crustacean literature. I seem to remember that Shane Lavery found that prawns have very long pure repeats that were stuttery, and other decapod studies have had a hard time too. There are lots of reasons for microsatellites to fail - invertebrates seem to be generally a lot more difficult and unpredictable in yielding good loci. In my group we have encountered some very difficult animals. My advice would be that if you work out that 3/ above applies, you have a few more goes at a few more microsatellites, then try alternative approaches (Garrick RC, Sands CJ, Rowell DM, Tait NN, Greenslade P, Sunnucks P (2004) Phylogeography recapitulates topography: very fine-scale local endemism of a saproxylic 'giant' springtail at Tallaganda in the Great Dividing Range of south-east Australia. /Molecular Ecology/ *13*, 3329-3344. contains some suggestions). Paul Sunnucks [paul.sunnucks@sci.monash.edu.au] * I had a wretched time getting micros to work on stone and blue crabs. I was having problems with the DNA itself, and I eventually figured out that if I boiled the tissue (claws or legs worked best, I boiled a whole leg in the shell) before I did the DNA prep, it denatured and inactivated whatever nasties were causing the problems. Hope this helps, j <><><><><><><><><><><><><> Judith Mank [jemank@uga.edu] * Concerning your questions on amplification difficulties using microat. in crustacean species, you seem to describe something similar to what a lot of people experienced when working with Lepidpotera. Most people find that designing primers for Lepidoptera are very difficult due to non-specific amplification bands, incorrect expected sizes for alleles, presence of null alleles, etc. There was a recent paper in Tree (Zhang, D.-X. 2004 Lepidopteran microsatellite DNA: redundant but promising, v19(10):507-509), where the author suggested that mobile/retro-/transposable elements may be responsible for the patterns in lepidopteran microsat markers. Have you check your sequences to see if this may be the case for your crustacean species? If this is the case, then you will have to probe your libraries for single copy markers. So, what you can do is: 1. design primers for the 5' and 3' flanking regions of your microsatellite clone excluding the repeat units, PCR amplify for the flanking regions, purify (either by gel extraction of the expected fragment or by direct column purification method if only the expected fragment amplified) and use it as DNA probe. If you gel extract the expected fragment then you must also sequence to confirm the identity of the fragment. 2. select a few individuals (6 - 10 maybe) randomly, cut each of these individuals with 4 - 6 different restriction enzymes (try to use 6-base rather than 4 base cutters), run out the samples and transfer them onto a nylon membrane for southern blot. 3. use individual fragments isolated from (1) as DNA probes for (2). What you should see: if the microsat flanking regions are unique, as you would expect for truly good microsat markers, you will see only one or two bands per sample (just like in microsat gels). If your flanking regions are highly repetitive within the genome of your crustaceans then you will see RAPD type profile on your southern blot membranes. From there, you can select for single copy microsat markers for your population study. Unfortunately I don't think there is any other quick and simple way to go about this. If you see multiple bands from many/most of your markers then it was always going to be trouble... however if you find out any other way of solving your problems easily please let me know. As to the problem of identical heterozygous alleles in all samples, I can't offer any good explanations but to speculate that if your markers were associated with low copy number mobile elements than that may explain for identical patterns in your populations, or if your organisms are clonal. Once again, you should carefully check your markers for retrotransposons, ME/TE (and virus + parasite) sequences. Finally, even after you have managed to isolate unique markers and apply them to your samples, you will have to include calculations for expected null alleles in your final analyses (as most of your primers are troublesome). An alternative to microsat. would be to look for SNPs in single copy genes.... but that will be many more months of work. Wee Tek Tay [wttay@unimelb.edu.au] * have just finished my PhD on moths and had exactly the same problems developing a library. It appears to occur all the time. I'm not sure if you'll have already done this but there is a BLASTn stand alone program that checks all your isolated sequences against each other. I found that the sequences flanking the repeats were repeated at numerous loci and that went some way to explain all the noise and lack of specificity. Emese Meglecz has written a paper on it for Lepidoptera which may help with some explanation. After getting rid of all the repeated regions I wound up with 4 that worked after a year's work. Microsats are fun. S.J.E.Anderson [S.J.E.Anderson@open.ac.uk] * Did you do a blast search to check for repeated flanking sequences (i.e. the sequences that flank the microsats)? This has proved to be a problem in lepidoptera and some other insects, and I suppose it's possible that it may also be occurring in crustaceans. A collaborator of one of my PhD students is currently preparing a manuscript describing this phenomenon. If you want further information on this, let me know Joanna J.R.Freeland [j.r.freeland@open.ac.uk] * We had similar result using PAGE for yellow perch, namely all heterozygotes and similarly spaced fragments. Aliquots of those same PCRs, when run on fluorescent genotyper, show very different result, namely, normal distribution of homozygotes and heterozygotes and allele sizes in the expected range. Never figured it out but now we don't use the data unless they're from fluorescent genotyping. PAGE is relied upon only for determining whether the PCR worked! Bonnie Brown [blbrown@vcu.edu] * Microsatellites are a quite heterogeneous grouping of DNA-sequences defined by there pattern of repeat length. In fact, only quite a few seem useful for population genetic analyses. I.e., those who are pure (only consisting of identical repeats) or simply compound (consisting of several pure simple repeats in an alternate pattern). If you picked satellites with quite complicate patterns their sizes might not follow a simple +1/-1 repeat change in individuals of your population. Hence your results contain no population genetical information. Another problem is primer specificity: regions flanking microsatellites might be highly repetitive as well, meaning that there can be several loci within the genome that can be primed by your oligonucleotides giving fragments of different sizes. In this case you should perform a virtual PCR with an appropriate programme (e.g. AMPLIFY for MacOS(1)) after having designed them (e.g. with PRIMER3(2)). If you have gotten more answers to your problems it would be very kind of you to forward them to the EvolDir-community as well. Thanks a lot, and good luck, indeed. Sébastien Wielgoss [Sebastien.Wielgoss@uni-konstanz.de] Deborah Bailie Fisheries Genetics and Evolutionary Ecology Lab School of Biology and Biochemistry Queens University of Belfast 97 Lisburn Road Belfast BT9 7BL Northern Ireland Tel: 028 90972247 Email: d.bailie@qub.ac.uk d.bailie@qub.ac.uk