July 2012: New update on high efficiency guanine-specific RVD NH and mammalian TALE transcription repressor (Cong et al., Nature Communications 2012 and Free full text available from PubMed Central).

June 2012: Have a question? Check out the new TALE Discussion Forum.

March 2012: Using PlasmidSafe before transforming ligation products greatly increases the yield of correctly assembled TALEs. Updated online protocol here.


Introduction to TAL Effectors

Transcription activator like effectors (TALEs) are natural type III effector proteins secreted by nutmerous species of Xanthomonas to modulate gene expression in host plants and to facilitate bacterial colonization and survival (Boch et al., Annu Rev Phytopathol 2010; Bogdanove et al., Curr Opin Plant Biol 2010). Recent studies of TALEs have revealed an elegant code linking the repetitive region of TALEs with their target DNA-binding site (Boch et al., Science 2009; Moscou et al., Science 2009). Common among the entire family of TALEs is a highly conserved and repetitive region within the middle of the protein, consisting of tandem repeats of mostly 33 or 34 amino acid segments. Repeat monomers differ from each other mainly in amino acid positions 12 and 13 (repeat variable di-residues), and recent computational and functional analyses have revealed a strong correlation between unique pairs of amino acids at positions 12 and 13 and the corresponding nucleotide in the TALE-binding site: NI to A, HD to C, NG to T, NN to G (and to a lesser degree A) (Boch et al., Science 2009Moscou et al., Science 2009; Miller et al., Nat. Biotech 2011Zhang et al., Nat. Biotech 2011).

The simplicity of the TALE code opens many opportunities for biological applications (Boch, Nat. Biotechnol 2011; Rusk, Nat. Methods 2011). Recent studies have tested the modularity of the TALE DNA binding code and have demonstrated that custom TALEs can be designed to recognize specific DNA sequences in a number of different cell types including plant and mammalian cells (Scholze et al., Virulence 2010; Morbitzer et al., PNAS 2010; Miller et al., Nat. Biotech 2010; Zhang et al., Nat. Biotech 2011). Custom TALEs can be tethered to a variety of effector domains to modulate transcription of endogenous genes in the genome (Morbitzer et al., PNAS 2010; Miller et al., Nat. Biotech 2010; Zhang et al., Nat. Biotech 2011) as well as generate site-specific double strand break to catalyze homologous recombination for genome engineering applications (Christian et al., Genetics 2010; Miller et al., Nat. Biotech 2010; Li et al. Nucleic Acids Res 2011; Mahfouz et al., PNAS 2011).

These pages include protocols as well as reagents for constructing custom TALEs as well as deploying TALEs in experimental applications (Sanjana, Cong, Zhou et al., Nature Protocols 2012Zhang, Cong et al., Nature Biotechnology 2011). You will also find a full list of reviews and relevant studies in the references section.


75 Responses to TAL Effectors

  1. Luke McLaughlin says:


    My name is Luke McLaughlin, I’m a life scientist (BSc) and currently finishing up my masters studies in molecular biology. I have always been fascinated by precision gene targeting techniques and have been a keen follower of all the lastest methods from ZFN’s, Meganucleases and now TAL effectors.

    I have a brief question if that’s alright, there’s been a lot of concern recently over patenting areas of the human genome and methods used to study it.

    Would TAL nucleases also end up being the sole patent of a single entity or would this technology be permitted for free use by any laboratory? I ask as i have recently read that a French biotech company Collectis has been given exclusive licensing rights for any technology associated with TAL nuclease use, is this another monopoly situation?

    I thank you for any answers you might provide.



    • congle says:

      Dear Luke, we have always been working very hard to make the TALE technology 100% open and free for the research community and We are confident that with this website and our open sources of vectors and constructs, the TALE technology is now and will be available to every researcher. Thanks a lot for your comments!

      - Le Cong

  2. Manoj Kumar.R says:

    Respected Sir,
    I did M.Sc.,Biotechnology.currently am doing my project work in TAL effector AvrBs3 genes.i want to know all gene sequence of Avrbs3 gene family.thats only will help for my further analysis.iso,i need a infermation about all avrBs3 gene family.am waiting for your reply.

    Thanking you,

    Manoj .R

    • congle says:

      Hi Manoj,

      I believe if you would like to full sequences of this gene family, the best and most up-to-date way is to use BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi) to search for homologous genes in the database. Specifically, you could use the repeat region amino acid sequence to perform a protein blast. Hope this is helpful to you. Thanks.

  3. Kim Anthony says:


    I was wondering where does the VP64 transactivation domain originate from? It it from HSV?

    Thank you,


  4. Behcet says:

    I did M.Sc.,Biotechnology. for my now project; ı want transform AvrBs3 ( xanthomonas sp. ) into a proper plant in order to acquire transgenic plant. my aim is triggering AvrBs3 the BS3(Resistance gene ) gen in plant. eventually stimulation the HR ( hypersensitive response )
    how can ı do that ? by the way ı want use 35S promoter and ı want you design a TALEN for cleavage my promoter or any region of my gene ( BS3)? is it possible ? I request your help plz
    thank you. ı wait reply

  5. guozili says:


    My name is Kok Zhi. I am currently working with the TALEN to generate KO mice. I have several question about the TALEN.

    1.Does the spaces between two TALEN monomers matter?
    During searching the off-target binding site of my desired TALEN, I found that one of the of target separates about 100bp with one mismatch. Could this off-target site been excised by TALEN?

    2.If only a monomer of the TALENs have the perfect match with the off-target site, could the monomer excised the DNA and cause double strand break?

    I wait for your reply.Thanks

    Kok Zhi.

    • neville says:

      Hi Kok Zhi:
      1. Yes, the spacing between the 2 TALENs used to generate the double-strand break does matter. As we mention in the paper, a distance of about 14-20 bases between the two TALENs’ binding sites seems to be optimal. If you find a match for both TALENs separated by 100bp, we would expect to see no cutting though off-target activity is always possible and thus it is important to screen your mouse ES colonies carefully.

      2. If only one of the two TALENs binds, it cannot cause a double-strand break because the FokI must dimerize to cut the DNA. Of course, it is always possible that there is some off-target activity (esp. since we are often greatly overexpressing our TALEN constructs). I think it is wise to check all regions in the genome that are perfect matches for either TALEN (left or right) and make sure only the intended target is modified.

  6. guozili says:


    It is Kok Zhi again. I have faced some problem when I use this web


    to search the potential off-target.

    I do not understand the output, particular the “target sequence” and “plus strand sequence”. What is the relationship between them and to the locus?

    P/S: I have already check the Help pages though it shows nothing with those technical terms there.

    I wait for your reply.Thanks

    Kok Zhi.

  7. thomas wood says:

    Hi there,

    Beautiful site, really good to see an effort to keep TALEN tech open. I’ve been trying to make some A and B arrays and, on digest, very few are the correct length. I wonder if you might know if there’s anything to do to increase the chance of having a correct length in my construct, aside from picking white colonies, and whether this has happened before or if I’m doing something wrong. I’m plating onto spectinomycin and using electroporation for transformation.

    Many thanks for any help,


    • neville says:

      Thanks Thomas. Based on your description, it doesn’t sound like you are using our protocol for TALE assembly. Please check out our protocol here or here for more information.

  8. Simon says:

    Hi all,
    I am based in Europe and wondered if anyone has identified a European supplier for T7 DNA ligase. I have only found Enzymatics and MCLAB on the web and where I am based, ordering from the States can be complicated.

    • congle says:

      Hi Simon, we have not found a eu company that sella t7 ligase, as we order from the us. The good news is the golden gate reaction essentially requires a high concentration ligase but not limited to t7, so you could use a high concentration t4, but because t7 ligase is more specific for sticky end, you might want to increase the cycle number of golden gate ligation to achieve similar efficiency. You could refer to the the high concentration version of t4 ligase from neb as a standard when you pick yore ligase.

  9. Davide says:


    I’m interested in the use of TALENs together with a repair template, to induce/correct/modify point mutations or SNPs.
    I found no or little information on the amount of homology necessary to achieve HR after TALENs induced double-strand break.
    I understand that with the increased efficiency, selection markers are unnecessary, but I have no idea if reducing homology arms to 1-2kbs would work.
    Any feedback is wellcome



    • neville says:

      Hi Davide: In general, longer homology arms are better. Practically, extremely long HR arms can make PCR screening difficult. We and others have had success with HR arms in the 500-1000 bp range (for each side). There are recent reports in the literature of success with extremely short HR arms (eg. oligos), which may also work for your needs. Hope that helps!

  10. lorenzo says:

    hi there,

    i’m trying to create some TALE nucleases plasmid following the instructions of the nature protocols and the plasmids from addgene. i’m at the last step of the plasmid creation but when i sequence the plasmids (and blast them with the target sequence generated by your website), they are full of mutations and also some deletions… i sequences more than 10 clones and non of them is correct…
    have you experienced some troubles like that? any suggestion for overcoming this issue?

    thank you so much for the help


    • neville says:

      Lorenzo: We haven’t experienced problems like you describe. Sequence error is pretty general, so I think you need to narrow it down a bit more… Before sequencing, did you do colony PCR or a restriction digest? It is a good idea to quickly screen clones before sequencing. There is detailed troubleshooting information in our Nature Protocols paper. Using that, you should be able to narrow down at which step in the protocol you are having problems. Good luck!

    • congle says:

      Hi Lorenzo,

      In addition to Neville’s comment, I would like to add that you might want to check two things: (1) check that the control plate from your ligation (which does not contain hexamer) has no colonies; and (2) it looks like you have insertion of repeats but they are not correct? If that is the case, I would suggest you look into if the primers for monomer amplification are correct and also you are using a high fidelity enzyme for monomer/hexamer amplification, such as Herculase II. Please let us know more details so we could help you troubleshooting, and as Neville said the trouble shooting guide in our paper should also be helpful. :)


    • Zhang Qiang says:

      Dear lorenzo,

      I’m constructing TALENs and TALE-TFs and I’ve also met the problems just as yours. What’s more, there are always some repeats deletions in my constructed pLenti plasmids. Have your problems been solved? How? Any reply is welcome.

      • neville says:

        Qiang: pLenti plasmids are not part of the current TALE Toolbox. Can you make sure you’re using the current set of plasmids? Please see our Jan 2012 Nature Protocols paper (linked on the front page). It is important to use the correct plasmids for the assembly as we’ve changed enzyme usage and other aspects of the cloning since the earlier Nature Biotech publication.

        • Zhang Qiang says:

          Hi neville,

          Thanks for your reply, sir. I’ve seen and studied both of your papers on 2011 and 2012 Nature Protocols. Since we bought the pLenti backbones from addgene and ordered primers as your 2011 one last year, we are still using your previous method. But I omit your gel-stab PCR step because we don’t own that. We amplify tetramer just using 4 monomers assemble product even omiting purification because of low sensitivity of EtBr Staining gel under a UV light. But I wonder how can the incorrectly assembled TALE with wrong overhangs ligated with the backbones. Thank you.

      • lorenzo says:

        hi there,

        i solve the problem using the same Taq suggested in the paper,
        although the taq i used previously was also proof-reading, maybe the herculase is better!

  11. Vinita says:

    How and when were TALES discovered and how have them been used and what future directions are being taken to exploit their utility?

    • congle says:

      Please refer to our reference section for a panel of papers and reviews about TALE’s history and development. I would recommend the reviews in particular.

  12. Greg says:

    Hi –

    Quick question about the primers in your Nat. Prot. paper: Must they be PAGE purified as IDT recommends, or can you get away with standard desalting?



    • congle says:

      Hi Greg,

      We don’t use PAGE purified primer, and it has been working fine for us. So you can just do standard desalting.



  13. jaouad says:

    I have questions about the therapeutic application of the TALE-proteins;
    It’s possible to use a TALE to increase the expression of a gene down in some disease by injecting the TALE-specific protein in the blood?
    What are the potential and the limitations of the therapeutic application of the TALE-proteins?
    TALEs can be designed to recognize specific DNA sequences. How much is the specificity, 100%? or we can affect the expression of other genes also?


    • neville says:

      Hi Jaouad, It might be possible to make an blood-deliverable TALE but it would be necessary to find a way to deliver them across cell membranes (eg. virus). Also, TALEs bind to the genome and thus need to get into the nucleus. Depending on their target sequence, TALEs can have off-target effects but present characterizations of these effects indicate that they are less than with existing DNA targeting proteins (eg. zinc fingers). I think it’s fantastic to think about the possible therapeutic potential of TALE proteins: Please check out some of the reviews listed on our references page for more information.

  14. Greg says:


    Is it okay to use a nanodrop (spectrophotometer) to standardize amplified monomer concentrations across wells of the plate, or should it be gel electrophoresis for all 72 purified PCRs?



    • Feng says:

      It should be okay to use nanodrop but it is better to use gel. If using nanodrop, make sure the final concentration of each monomer is no less than 30ng/ul. If you happen to have a couple of reactions with low yield, repeat those reactions to get more so that you don’t have to dilute all of the other monomer to too low of a concentration.

  15. Michele says:

    Dear all,

    I want to use the TALEN system but I prefer to be sure to one point: When I design the system using your software, I obtain two sequences, for example:

    I have to follow this order for both? So NG in the last position means that I have to use the NG TALEN vector and at position 1 For R-tale I have NI?

    Thanks a lot

    • zhang_f says:

      Yes, the last RVD is encoded by the plasmid backbone. This means L-Tale NI NN NN NG NI NN NG NI NN NN NN NG will be targeting (5′-TAGGTAGTAGGGT-3′). Notice that the first T is specified by the N-term of the TALE.

  16. Henri says:

    I have a question about homologous recombination, after inducing a double strand break using TALENs. I have read that TALENs are a good means to insert a piece of DNA by means of homologous recombination. However, I do not want to insert a piece of DNA, but simply to swap a DNA sequence with a homologous sequence (so, end product should contain the exact nr of basepairs as the starting material). Can this be accomplished using TALENs, and if yes, how?



    • neville says:

      Henri: That should be no problem. HR can be used to delete, insert, or, as you want, to have the same number of bases as before HR. You simply need a donor construct with 5′ and 3′ homology to the cut site. Good luck!

  17. Justin says:


    I as wondering what information you have about the use of the TALEN’s after verification in the 293FT cells? I am interested in making TALEN’s to create knockout human cell lines for various genes. Do you use the 293FT cells with some packaging system such as Invitrogens ViraPower? Or, can you just use the supernatent from the cells that demonstrated success with the Surveyor assay? Any help would be appreciated.


    • neville says:

      Justin: I’m not sure I understand your question. After verification of cutting in 293FT cells (eg. via Surveyor assay), you can then proceed to use the TALENs for NHEJ and HR. If your intended cell type for your application is significantly different from 293s (eg. not human, etc.), you might want to just use that cell for the Surveyor assay. We are working in human cells so 293s are a natural, easy to culture/transfect cell line. For us, TALEN delivery into 293s is via lipofection. I am not sure what you’d do with the supernatant of TALEN-transfected cells… I think you might be confusing making viruses (also often done in 293s) with our functional validation (eg. Surveyor).

  18. Zhang Qiang says:


    I’ve constructed some TALE-TFs using pLenti backbones depending on your previous paper(2011 Nature Protocols). But I met a problem on cell transfection. I transfected pLenti expression plasmids only, but extremly few green florescences can be seen 24h after transfection. I’ve tried 293T cells, primary MEF cells and NIH 3T3 cells. Liposometransfection(liposome2000, inventrogen) and electrotransformation have been used. Other plasmids are transfected well in our lab. So how do you achieve high efficiency transfection with pLenti? Any reply is welcome.


    • neville says:

      Zhang: It should be fairly easy to get high (90%+) transfection in 293 cells. Please consult the standard Lipo2000 protocol, which is what we use.

  19. surya says:


    I am working on neurobiology using zebrafish as a model organism. Recently I came to know about the TALENs and the TALE toolbox. I would like to generate some transgenic Gal4 lines using your toolbox. I do have some doubts regarding the TALENs. Please clarify my doubts.

    They are:
    1) Is there any software you would suggest to design the TALEN target sites.

    2) My lab is new and we do not have any tissue culture equipment. Is it necessary to test the specificity of TALEN pairs in vitro before using in vivo. If it is necessary, can you please let me the alternatives without using cell lines.

    3) Can you please let me how can you find the off-targets that has been cleaved by TALENs in vivo. Is it possible to get transgenic without any off-target modifications by TALENs.

    Thanking you,


    • neville says:

      1: Not yet. We are working on something like this now. For design suggestions, please see our Jan 2012 Nature Protocols paper. It isn’t too hard to pick TALEN sites by hand.
      2: You could try the in vitro SELEX assay or simply go straight to in vivo… I am unfamiliar with work in zebrafish, so it’s hard to advise you here.
      3: Off-target modification is definitely a concern, although longer TALEN binding sites (like the 18mers we use in the paper) should have minimal off-target effects. You could BLAST for similar sites to the TALEN binding sites and then sequence those regions of the genome to make sure there is no cutting at those sites.
      4: Again, I’m not an expert on zebrafish but both could work… I think mRNA is favored in zebrafish due to relative speed of expression (ie. faster) and uniformity. I believe that plasmid DNA often results in later expression (not so good for studying development) and mosaic expression. If you do in vitro transcription, you might want to first subclone your TALENs into a T3/T7/SP6 promoter vector or whatever is standardly used for zebrafish.

  20. Yang says:

    Hei,I am thinking about apply talen transform filamentous fungi,and I have some questions:
    1,how fast can talen work in the cell(for me it maybe the spore)after introduce the plasmid into the cell. (because my fungal spores contains some nucleus, if we want to get the pure knockouts, maybe it is good to edit the genome before cell divition)
    2,can the talen proteins tranport from cell to the nebour cells?
    3,can i use it for insert GFP? or some kind protein tagetting?
    4,how to select the transforments (in the other methords,we usually use antibiotic)
    5,can i design my own Customed TAlEN by your kits?

    • congle says:

      Hi Yang, I am not very familiar with the filamentous fungi system, but since TALENs works well in yeast and number of species, I would expect that it should work in your system. As for your questions, I could only provide my personal comments, as I have not worked with this system, please take them at your own discretion.

      (1) Usually after 48 – 72h of the expression of TALENs, the genomic DNA would be cleaved. I would recommend you try 72h after transformation as a start, and increase the time if needed.
      (2) I think TALEN proteins usually would have a NLS (Nuclear Localization Signal), so I would expect it might not have too much localization in the cytoplasm, hence transportation across different cells might not be very efficient. However, if there is plasmid DNA or mRNA transportation between the neighbor cells, that would be sufficient for expression TALENs in a group of cells.
      (3) Yes, I think you could, for this you would need a insertion construct containing GFP (and flanked by homologous recombination arms if you hope to integrate into a particular locus in the genome) and maybe some selection marker. You could then cotransform this construct with your TALEN pairs to have insertion.
      (4) For bacteria we use Amp in the TALEN vector, and there is also hygromycin resistance gene in the TALEN backbone. You could clone in your own antibiotic marker too.
      (5) Yes. you could synthesize TALENs targeting any target sequences less than 24bp in length with our kit.

      Hope this helps.

      Le Cong

  21. Justin says:

    Thanks for the reply…you are correct in my confusion. The 293FT cell line made me assume Lentivirus was being produced and that was the end goal. I understand now that virus is only needed if I want to increase the efficiency of the TALEN getting into the cell. I intend to just use the TALEN as you describe in the protocol. One question though, what would be the disadvantage of skipping the 293 cells and moving directly into my cells of interest? I could still do the surveyor assay on those cells correct? Also, would you select for hygromycin before subcloning the cells to eliminate those without any TALEN at all? Thank you so much for your assistance and the effort you all have put forth to make this an open, and responsive, community.

    • congle says:

      Hi Justin, I didn’t follow your previous questions. But I would like to let you know that the TALEN vectors in the TALE toolkit is not a lenti-vector so there would be NO lentivirus being produced in any situation with this vector.

      I think there is no disadvantage of trying TALEN directly in your cell of interest as long as you could either (1) efficiently deliver TALEN expression, or (2) have some selection methods post delivery.

      We generally do not select with hygromycin as we typically get very good transfection efficiency. However, it would be a good idea to perform the selection if efficiency of delivery is low.

      Also, in a previous paper, we have a lenti-viral version of the TALE and TALEN backbone (see our paper in Nature Biotechnology). I have deposited the vectors in addgene and you should be able to find them.

      Thanks for your comments and questions.

      Le Cong

  22. Gabor says:


    I am a postdoc and we decided to use TALENs in order to generate genetically modified mice. I have one question.
    I have my gene of interest and after the stop codon i would like to introduce and ires gfp. I have a talen pair which binds in the 3′ region of my gene(the binding site includes the stop codon of my gene) and 15 bp later in the 3′ UTR. this is perfect for me in order to introduce my construct. The question is: is it possible to delete the stop codon with this TAlEN pair and introduce a P2A cassette which follows my gene just by designing different homologous arms?
    I hope my description makes sense!
    Thank you very much for your answer in advance.


    • congle says:

      Hi Gabor,

      Thanks a lot for your question. Because a pair of TALENs cut in between (roughly in the middel) of their two binding sites, it seems for your purpose, to delete the stop codon you could design a TALEN that flanks the stop codon, and place the binding site so that the stop codon is right in the middle of the spacer (14-22bp) between the two binding sites. This would usually allow you to eliminate the stop codon if you introduce a homologous recombination (HR) template together with the pair of TALENs. Basically, if the TALENs could cut the target, then the integration of the HR template would replace the part of DNA around the stop codon. I think the plan you described should work if the TALENs works fine and the HR efficiency is not too low in your cell of interest. Hoep this helps and let me know if you have additional ones.


  23. Stefan says:

    Thanks for your site and the support.
    I have difficulties to find appropriate sites in the locus of interest because of the first binding domain in the N-term of your TALE. Have you any experiences with deleting this first binding domain or the RVD.

    • zhang_f says:

      The 5′ thymine requirement is not strict so you can try targeting sequences that do not start with a 5′T. Efficiency will likely depend on the specific site so you should try a couple of different targeting sites. Hope this helps.

  24. Pieterjan says:

    Hi Le Cong,
    I’m currently cloning some TALENs via the addgene Golden Gate TALEN assembly kit.
    I was just wondering if there is a proper software (open access) to make a vector map of the TALEN constructs, because with Lasergene I didn’t find a function from which you can clone 10 plasmid in 1 reaction. With serial cloner it works, but it takes a very long time, because the ligation reactions only work with 3 plasmids each time. Since this kit is based on restriction Type II enzymes, it is even not possible to clone (in SerialCloner) it in several steps, since these restriction enzymes do not leave a recognition site after cutting. It is only possible if after the virtual ligation, you ligate some extra flanking recognition sites back to the products.
    We don’t have a licence for Vector NTI, so we cannot use that software either.
    Do you know if there’s a software that can do this at once and how this works, and is preferably open access. (I’m a mac user)

    Thank you very much in advance!


    • neville says:

      PJ: If you just need a reference sequence of your TALEN (either in FASTA or GenBank format), please use our reference sequence generator tool here: http://taleffectors.com/tools . Click on TALEN and enter the DNA target sequence of the TALEN (including the 5′ thymine and the 0.5 repeat at the end) and it will do the rest for you! Hope that helps, -Neville

      • Pieterjan says:

        Thanks Neville,
        I’ve tried it and aligned it to my reference sequence (made by SerialCloner), but there are many many mismatches. I’ve sequenced the TALENs and they nicely align with my ref. sequence (made by SerialCloner). I don’t know what could be wrong, but maybe the Sanjana tool kit is not using the same coding sequence for the same Amino Acid repeats as the Cermak (Golden Gate) tool kit is using?

        • neville says:

          PJ: Yes, the TALEN reference sequence generator on our website is only for TALEN/TALE-TF constructs built using our plasmids & primers (Sanjana, Cong, Zhou et al., Nature Protocols, 2012). It will not work for the Cermak tool kit. Please contact the authors of your TALE-making kit for information on generating reference sequences for their constructs. Best, – Neville

          • Pieterjan says:

            I will contact them. Still, it appears strange to me that the code can be different.

          • neville says:

            PJ: Synthetic TALEs can be based on different backbones (N and C term conserved regions) and can use different codons for the monomer repeats… it is very likely that the nucleotide sequence will differ when using different construction protocols. Hope that helps! – Neville

  25. Stefan says:

    Thank you for the fast response, the binding of the 5´T at position 0 (or N) seems to be strange.
    In a recent study from Zhaos group (optimized TAL effector nucleases (TALENs) for use in treatment of sickle cell disease) they nicely showed the impact of the 5´T. But they also found a variant with reduced preference for T. This seems to be also dependent on the C term, however have you experiences with similar N and C term variants that support this finding or is it in this particular case just a lucky coincidence.
    But back to my initial problem.
    I could bypass my problem to find proper binding site with an 5´T by using shorter or longer TALEs. This should be possible by generating unusual monomers using alternativ primer combinations.
    Generating longer TALEs (more repeats) seems also possible, if I understood correctly your toolbox or better your recommended primers allow building up to 4 hexamers. Is this correct? And if, is there a reason to use just 3 hexamers? Are longer TALEs less efficient or is the golden gate reaction to assembling 4 hexamers not working?


  26. Markus Grompe says:

    We are trying to make a TALEN. We bought the Addgene cloning kit with all of the plasmids pTALEN v2 (NI) etc. (http://www.addgene.org/32189/). Looking at the sequence provided for the plasmid by Addgene, there appears to be no FokI domain in the plasmid. The plasmid is supposed to be ready to clone the TALE assembly into it.
    Where is the FokI domain containing plasmid in this toolkit?

  27. Martin says:

    Hey all-

    Question. What is the minimal transcription unit to encode TALE? Are the entire N- and C- termini required? Essentially, I would like to generate a simple TALE-NLS-activation domain construct.

    Thanks in advance,


    • neville says:

      Hi Martin- Please see Figure 2c and d of our lab’s paper in Nature Biotechnology (Zhang et al., 2011) for an analysis of N and C term truncations.

  28. Jamie says:

    Different TALEN assembly protocols use different TALE backbones (N and C amino acids that surround the repeats), and these (anecdotally) appear to affect their effectiveness in various systems. I am having trouble figuring out which backbone your system uses, and why. I note some minimization experiments from Zhang et al., 2011. My two questions are:

    1. is it correct to assume that the Sanjana et al. 2012 protocol assembles into the N1-C0 hax3 architecture?

    2. What went into the choice to use hax3, and have you compared this backbone to others?

    Thanks, appreciate all you’ve done to make this kit so great,


    • congle says:

      Hi Jamie, thanks a lot for your question. Although there are different versions of the N and C term truncations used, the version from our protocol is one of the most commonly-adopted TALEN designs (please refer to other TALEN papers, such as Miller et al., and Hockemeyer et al. NBT), which have shown to be quite robust in terms of activity for several targets across different organisms. And as you noted, our designs stems from the truncation study we performed in Zhang et al. 2011. For your questions:
      1. The TALEN backbones in the 2012 protocol paper is based on N3-C3 of the 2011 paper, only that the N-term was slightly extended.
      2. The TALE N/C-term sequences are highly conserved, so generally it is not expected that the activity of TALEs would differ significantly depending on which natural TALE the backbone is based upon. We chose hax3 because it’s one of the most active, well-characeterized natural TALE that were successfully applied to make customized TALEs with various repeat lengths.

      Hope this helps. :)


  29. Song Wang says:

    How can we design TALEN when a target gene have multiple Transcript Variant?
    Is it enough to choose the longest transcript as the target?

    Song Wang

    • neville says:

      If the transcription start site is the same (which is sometimes the case), then the TALE-TF should in theory activate multiple variants. If they have different 5′ UTRs/different transcription start sites, you will likely need to build separate TALE-TFs for targeting different transcript variants.

  30. Yuanchun says:

    I plan to make transgene addition by TALENs in human ES cells or iPS cells. I noticed that both your TALEN cloning backbone and the TALE monomer templates vectors on the website contain CMV promoter. Since it has been shown that CMV promoter is most likely to be silenced or inactive in human ES or iPS cells, I wonder if you have tried the TALEN in human ES or iPS cells? Did they work well in these cells?

    • neville says:

      We and other labs have had success with the CMV TALENs from our toolbox in human PSCs. Give it a try!

  31. Brock says:

    Great site and it’s very exciting that these methods are in wide use. I have a question about where the field stands on the sequence requirements of TALE repeat modules. Is there consensus on whether the most C terminal TALE in the array must bind to T in order for the TALEN to function? I understand that the 5′ base ought to be a T. How about the most 3′? Does anyone have information on this? Good luck to all with your cloning and applications…

  32. Martin says:

    I’m unable to get amplification of positive clones by colony PCRs (or minipreps) using the TALE-SEQ-F1 and TALE-SEQ-R1 primers. Interestingly, the primers work great for sequencing. Are the 2kb repeats difficult to amplify?

  33. lucabio says:

    Hi all,
    I was wondering if the TALEs binidng to DNA is affected by epigentics marks, like metilation. I suppose yes but wanted to double check with you.
    BTW, complimets fo rthe website.
    Kind regrds,

    • congle says:

      Hi Lucabio, Thanks a lot for your question. We have yet to see a convincing and comprehensive report on the effect of epigenetic marks on TALE binding. We definitely agree with you that it is very likely that it could affect TALE binding. Hopefully there will be some studies soon to elucidate this issue.

  34. Alun says:

    Hi Le,

    Is there a possibility for a private company to order the TALEN kit and use it to create GM animals freely?


  35. Ika says:

    Hi and thanks for this pretty site. Our questions are: any news of successful use of TALEs in drosophila? If TALE is used to target an enzymatic activity, do you see feasible to ‘cover’ the chosen locus/regulatory element with Tandem TALEs excluding (some/most)TF-binding sites? (Dense coverage of a defined genomic area). Which effector domains (apart from nucleases and transcr activators) were successfully tethered with TALE?

    Many thanks (also for giving me a break from those zinc fingers),

    Kind regards.

    • congle says:

      Hi Ika,

      Thank you for your question and kind words. Yes, I believe TALENs have been applied in drosophila successfully and please refer to the following paper: http://www.sciencedirect.com/science/article/pii/S167385271200077X.

      As for the TFs application, it’s definitely an intriguing idea and with high-throughput TALE assembly systems available now it should be quite feasible. Currently there are mostly two type of domains showed consistent activity when fused to TALEs, transcriptional activation domain and nuclease domain. We believe there will be more functional domains that enable more diversified activity of TALEs and we will keep updating our website, so please check back regularly. Thanks a lot.



Set your Twitter account name in your settings to use the TwitterBar Section.