A faster way to make GFP-based biosensors: Two new transposons for creating multicolored libraries of fluorescent fusion proteins
© Sheridan and Hughes; licensee BioMed Central Ltd. 2004
Received: 19 April 2004
Accepted: 18 August 2004
Published: 18 August 2004
There are now several ways to generate fluorescent fusion proteins by randomly inserting DNA encoding the Green Fluorescent Protein (GFP) into another protein's coding sequence. These approaches can be used to map regions in a protein that are permissive for GFP insertion or to create novel biosensors. While remarkably useful, the current insertional strategies have two major limitations: (1) they only produce one kind, or color, of fluorescent fusion protein and (2) one half of all GFP insertions within the target coding sequence are in the wrong orientation.
We have overcome these limitations by incorporating two different fluorescent proteins coding sequences in a single transposon, either in tandem or antiparallel. Our initial tests targeted two mammalian integral membrane proteins: the voltage sensitive motor, Prestin, and an ER ligand gated Ca2+ channel (IP3R).
These new designs increase the efficiency of random fusion protein generation in one of two ways: (1) by creating two different fusion proteins from each insertion or (2) by being independent of orientation.
Biosensors based on GFP-fusion proteins are powerful tools for observing real-time events within living cells. Insertion of GFP within another protein has produced biosensors capable of signaling intracellular events through intrinsic fluorescence changes [1, 2], fluorescence resonance energy transfer (FRET) [3, 4], and changes in sub-cellular localization . The difficult task of finding the right insertion site to produce a biosensor can be accelerated by screening libraries of random GFP insertions [6–8]. The insertional strategies described to date, however, are limited in two ways. First, each insertion produces only one kind, or color, of fluorescent fusion protein. Creating the multicolored libraries necessary for co-expression or FRET analyses requires either separate rounds of insertion and screening for each fluorescent protein or additional subcloning to exchange fluorophores. Second, the efficiency of any random approach is limited to a maximum of 1:6 because a fusion protein can only be produced if the GFP coding sequence lands in the correct orientation and reading frame with respect to the target coding sequence. We reasoned that it might be possible to overcome these limitations by placing two different fluorescent protein coding sequences in a single transposon, either in tandem or antiparallel. Here we present the results of our initial tests with these designs.
Results and Discussion
To verify that <EYOR> could be used to generate full-length fusions with either YFP or CFP unique in-frame clones were digested in parallel with Srf I or Asc I and re-ligated. The resulting fusion constructs were transiently expressed in HEK-293 cells and screened for YFP and CFP fluorescence (Figure 2B,2C). All 13 unique insertion sites produced fluorescent full-length fusions with both CFP and YFP and all were localized to intracellular membranes.
The <EYOR> design could be expanded for a wide range of protein tagging applications by replacing the secondary CFP cassette with another open reading frame. With such a transposon, YFP fluorescence would be used as a reporter to rapidly identify random in-frame insertions. Subsequent digestion with Asc I and re-ligation could then generate fusion proteins that might otherwise be difficult to screen for such as epitope tags, protease cleavage sites, or even a new N-terminus complete with a secretory signal peptide. Several groups have reported similar strategies based on multi-domain transposons for the random insertion of small peptide tags [reviewed in:]. Like <EYOR>, these transposons utilize a primary reporter domain to identify in-frame insertions. Subsequent excision of the reporter domain (and selectable marker) then restores the full-length target coding sequence with an inserted peptide tag. The <EYOR> design is unique, however, in that its overlapping pairs of Asc I and Srf I restriction sites, allow the user to create identical full-length fusion proteins from both the reporter domain and the secondary coding sequence.
The second transposon design, the Double-Barrel transposon (<DBT>, Figure 1B), encodes green and red fluorescent proteins (GFP and DsRed) in opposite orientations. This means that any <DBT> insertion within another protein coding sequence has a 1:3 chance of being in-frame regardless of its orientation. Therefore, <DBT> should double the efficiency of random fusion protein generation, by producing equal numbers of GFP and DsRed fusions.
In addition to their antiparallel orientation, the GFP and DsRed coding sequences in <DBT> each use a different relative reading frame through the Tn5 MEs. As in <EYOR>, GFP fusion proteins are created by insertions after the third nucleotide of a target codon. The DsRed coding sequence, however, has been shifted by 1 nucleotide relative to the Tn5 MEs. Therefore, DsRed fusion proteins are generated by transposon insertions between the second and third nucleotides. Using different reading frames for GFP and DsRed doubles the total number of insertion sites in the target coding sequence from which fusion proteins could potentially be made. While this does not alter the frequency of in-frame insertions, it does reduce the screening cost of saturating a target clone by increasing the probability of recovering unique in-frame insertions.
We chose GFP and DsRed to build the <DBT> transposon because their coding sequences are so dissimilar. Our concern was that if we chose two similar sequences, CFP and YFP for example, the antiparallel orientation of these coding sequences could produce extensive mRNA hybridization and secondary structure that would inhibit protein translation. It appears however, that DsRed is not well suited for insertion within other proteins. Indeed, DsRed has not been reported as a fusion protein in the middle of another protein, and even N- and C-terminal fusions with DsRed can be problematic , perhaps due to its being an obligate multimer . Despite the low yield of DsRed fusions, these results demonstrate that <DBT> can be used to simultaneously generate full-length fusion proteins in two different reading frames. As novel fluorescent proteins are isolated from new species, or old ones are altered, this type of bi-directional transposon could potentially double the output of the screening process.
The transposons described here should greatly accelerate the creation of multicolored libraries of fluorescent fusion proteins. By creating identical full-length YFP and CFP fusion proteins from each in-frame insertion, the <EYOR> transposon not only facilitates the generation potential FRET pairs, it enables the direct comparison of different fluorophores in otherwise identical fusion proteins. The <DBT> design, on the other hand, has the capacity to double both the throughput of fusion protein generation by virtue of its bi-directionality as well as the total output of novel fusion proteins through the simultaneous use of multiple reading frames. Ultimately, the ability to generate large numbers of novel fusions proteins in days rather than months, should shift the limiting rate at which novel fluorescent protein biosensors are identified to functional screening rather than protein design and construction.
PCR and standard subcloning procedures were used to create the plasmids encoding <EYOR> (pBNJ55b.1), <DBT> (pBNJ38.5) and Prestin (pBNJ12.5) (full sequences in supplementary material). The fluorescent protein coding sequences used were Venus (YFP) , ECFP-N164H (CFP), EGFP (GFP) and DsRed2 (Clontech). The Kan R gene was obtained from pUniV5-His-TOPO (Invitrogen). The construction of the IP3 receptor expression plasmid, pCMVI-9, was previously described .
Tn5 transposition and plasmid isolation
Transposons were amplified from their host plasmids via PCR with a single primer complementary to the 19 bp Tn5 ME (5'-CTGTCTCTTATACACATCT-3') and purified as previously described . Purified transposon and target concentrations were each quantified against an independent DNA standard using a DynaQuant 200 fluorimeter. The transposition reaction was performed according to manufacturer's recommendations (Epicentre Technologies, Madison, WI) with 200 ng of target DNA and a molar equivalent of purified transposon. Electrocompetent XL-10 Gold E. coli (Stratagene, La Jolla, CA) were transformed with 0.5 μL of the transposition reaction and plated on LB agar with Ampicillin (100 μg/mL) and Kanamycin (50 μg/mL). Parallel plating of the transformation on LB agar with Ampicillin alone was used to establish the transposition efficiency.
Transposed plasmids were isolated in a 96-well format from 1.25 mL LB cultures with Eppendorf PerfectPREP-96 Vac Direct Bind miniprep kits on a PerkinElmer MultiPROBE II HT liquid handling robot and eluted in 70 μL of ddH2O.
HEK-293 cells (American Type Culture Collection CRL-1573) were plated 24 hr prior to transfection, in 96-well glass bottom tissue culture plates (NalgeNUNC) at 6 × 104 cells in 100 μL of MEM-E with 10% fetal bovine serum. Transfections were performed with ~300 ng of plasmid DNA and 0.3 μL of Lipofectamine 2000™ (Gibco BRL) in a total volume of 50 μL of Opti-MEMI (Gibco BRL) per well. The cells were screened for fluorescence 24 hr after transfection with a 20× objective on a Zeiss inverted microscope with excitation and emission filter sets optimized for CFP/YFP or GFP/DsRed imaging (Omega, Brattleboro, VT).
Sequencing and generation of full-length fusion proteins
Exact insertion sites were identified for all fluorescent transposed clones by sequencing 5' out of the transposon with a primer complimentary to the <EYOR>YFP coding region (5'-CTGCAGGCCGTAGCC-3') or <DBT>GFP coding region (5'-TGGCCGTTTACGTCGCCGTCCA-3'). To generate full-length fusion proteins, plasmids with unique in-frame insertions were digested and re-ligated. After restriction digestion (100 ng of plasmid DNA and 0.5 U of Asc I or Srf I in 10 μL total volume), 1 μL of the digest reaction (~20 ng DNA) was re-ligated with Fast-Link™ ligase (Epicentre Technologies) for 15 min at room temperature (0.5 mM ATP, 1X Fast-Link™ buffer, 1 U ligase, 7.5 μL total volume). After heat inactivation (70°C for 15 min.), XL-10 Gold E. coli were transformed with 0.5 μL of the ligation reaction and plated on LB agar with Ampicillin. The following day, colonies were co-inoculated in LB with Ampicillin and Ampicillin + Kanamycin to verify loss of the Kan R prior to plasmid isolation.
The authors would like to thank Martina Brauns and the members of the Friday Afternoon Lab Meeting for their invaluable input on this project. We would also like to thank Atsushi Miyawaki for the Venus (YFP) coding sequence, Catherine Berlot for the ECFP-N164H (CFP) coding sequence, Joseph Santos-Sacchi for the Prestin cDNA and Barbara Ehrlich for pCMVI-9. We are grateful to the staff of the HHMI Biopolymer/Keck Foundation Biotechnology Research Laboratory for their rapid and reliable DNA sequencing. This work was supported by NINDS R21 NS044883-01. D.S. is a Howard Hughes Medical Institute Predoctoral Fellow.
- Siegel MS, Isacoff EY: A genetically encoded optical probe of membrane voltage. Neuron. 1997, 19: 735-741. 10.1016/S0896-6273(00)80955-1.View ArticleGoogle Scholar
- Ataka K, Pieribone VA: A genetically targetable fluorescent probe of channel gating with rapid kinetics. Biophys J. 2002, 82: 509-516.View ArticleGoogle Scholar
- Sakai R, Repunte-Canonigo V, Raj CD, Knopfel T: Design and characterization of a DNA-encoded, voltage-sensitive fluorescent protein. Eur J Neurosci. 2001, 13: 2314-2318. 10.1046/j.0953-816x.2001.01617.x.View ArticleGoogle Scholar
- Miyawaki A, Griesbeck O, Heim R, Tsien RY: Dynamic and quantitative Ca2+ measurements using improved cameleons. Proc Natl Acad Sci U S A. 1999, 96: 2135-2140. 10.1073/pnas.96.5.2135.View ArticleGoogle Scholar
- Barak LS, Ferguson SS, Zhang J, Caron MG: A beta-arrestin/green fluorescent protein biosensor for detecting G protein-coupled receptor activation. J Biol Chem. 1997, 272: 27497-27500. 10.1074/jbc.272.44.27497.View ArticleGoogle Scholar
- Merkulov GV, Boeke JD: Libraries of green fluorescent protein fusions generated by transposition in vitro. Gene. 1998, 222: 213-222. 10.1016/S0378-1119(98)00503-4.View ArticleGoogle Scholar
- Biondi RM, Baehler PJ, Reymond CD, Veron M: Random insertion of GFP into the cAMP-dependent protein kinase regulatory subunit from Dictyostelium discoideum. Nucleic Acids Res. 1998, 26: 4946-4952. 10.1093/nar/26.21.4946.View ArticleGoogle Scholar
- Sheridan DL, Berlot CH, Robert A, Inglis FM, Jakobsdottir KB, Howe JR, Hughes TE: A new way to rapidly create functional, fluorescent fusion proteins: random insertion of GFP with an in vitro transposition reaction. BMC Neurosci. 2002, 3: 7-10.1186/1471-2202-3-7.View ArticleGoogle Scholar
- Zhou M, Bhasin A, Reznikoff WS: Molecular genetic analysis of transposase-end DNA sequence recognition: cooperativity of three adjacent base-pairs in specific interaction with a mutant Tn5 transposase. J Mol Biol. 1998, 276: 913-925. 10.1006/jmbi.1997.1579.View ArticleGoogle Scholar
- Goryshin IY, Reznikoff WS: Tn5 in vitro transposition. J Biol Chem. 1998, 273: 7367-7374. 10.1074/jbc.273.13.7367.View ArticleGoogle Scholar
- Zheng J, Shen W, He DZ, Long KB, Madison LD, Dallos P: Prestin is the motor protein of cochlear outer hair cells. Nature. 2000, 405: 149-155. 10.1038/35012009.View ArticleGoogle Scholar
- Mehdi H, Ono E, Gupta KC: Initiation of translation at CUG, GUG, and ACG codons in mammalian cells. Gene. 1990, 91: 173-178. 10.1016/0378-1119(90)90085-6.View ArticleGoogle Scholar
- Manoil C, Traxler B: Insertion of in-frame sequence tags into proteins using transposons. Methods. 2000, 20: 55-61. 10.1006/meth.1999.0905.View ArticleGoogle Scholar
- Mignery GA, Newton CL, Archer B. T., 3rd, Sudhof TC: Structure and expression of the rat inositol 1,4,5-trisphosphate receptor. J Biol Chem. 1990, 265: 12679-12685.Google Scholar
- Mizuno H, Sawano A, Eli P, Hama H, Miyawaki A: Red fluorescent protein from Discosoma as a fusion tag and a partner for fluorescence resonance energy transfer. Biochemistry. 2001, 40: 2502-2510. 10.1021/bi002263b.View ArticleGoogle Scholar
- Sacchetti A, Subramaniam V, Jovin TM, Alberti S: Oligomerization of DsRed is required for the generation of a functional red fluorescent chromophore. FEBS Lett. 2002, 525: 13-19. 10.1016/S0014-5793(02)02874-0.View ArticleGoogle Scholar
- Nagai T, Ibata K, Park ES, Kubota M, Mikoshiba K, Miyawaki A: A variant of yellow fluorescent protein with fast and efficient maturation for cell-biological applications. Nat Biotechnol. 2002, 20: 87-90. 10.1038/nbt0102-87.View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.