Worm Breeder's Gazette 14(5): 16 (February 1, 1997)
These abstracts should not be cited in bibliographies. Material contained herein should be treated as personal communication and should be cited as such only with the consent of the author.
Carnegie Institution of Washington, Department of Embryology, 115 West University Parkway, Baltimore, Md. 21210 USA
We are preparing a supplementary vector kit for release February 1
1997. If you are interested in obtaining these kits or documentation
concerning their use, please email us (fire@mail1.ciwemb.edu), and we
will return the relevant release forms.
The 1997 supplementary kit contains the following goodies:
I. Blue GFP constructs
Heim and Tsien [Cur. Biol. 6:178] reported that an efficient blue
fluorescing variant of gfp could be obtained by making two amino acid
substitutions: tyr66>his and tyr145>phe. We have produced these
substitutions in the context of our intron-containing gfp and gfp::lacZ
cassettes. These give a blue fluorescent signal using standard "DAPI"
filter sets (standard GFP version gives a green signal with DAPI filter
sets). Under these illumination conditions, the Y66HY145F blue
fluorescence fades much more rapidly than standard GFP. For our
applications, we are thus somewhat unsure about the usefulness of this
variant for double labeling.
The blue gfp constructs are in test plasmids driven by the unc-54
and myo-3 promoters. Standard restriction sites flanking and within gfp
allow easy swap of these coding regions with equivalent regions in
existing vectors or fusion constructs.
II. Other GFP variants
We have produced a number of other amino-acid sequence variants in
the context of our intron-rich gfp coding region. The amino-acid
variants include some based on published literature, and a few
combinatorial mutations. From the literature, some of these might be
expected to have different spectral properties and thus be advantageous
in specific applications. We have no simple quantitative way of
comparing GFP spectra. Nonetheless we would like to encourage any
enthusiastic laboratory to generate such data.
As above, these GFP mutations have been produced in test constructs
carrying the gfp variant driven in body wall muscle by the unc-54 or
myo-3 promoters. Any of the variants can be exchanged into existing
vectors or fusion constructs by a simple restriction fragment swap.
Table: Variant GFP forms in 1997 kit
(Original 1995 kit had S65T [gf2] and S65C [gf3])
F64L S65T [gf4]
S65A V68L S72A [gf5]
S65G S72A [gf6]
Y66H [gf7]
Y66W [gf8]
F64L S65T Y145F [gf9]
F64L S65T N146I [gf10]
S65C M153A [gf11]
F64L S65T N146I M153A [gf12]
F64L S65T N146I M153T V163A [gf13]
Y66H N146I M153T V163A [gf14]
Y66W N146I M153T V163A [gf15]
Y66H N146I [gf16]
Y66W N146I [gf17]
F64L S65T Y145F M153A [gf18]
F64L S65T Y145F M153T V163A [gf19]
Y66H Y145F M153T V163A [gf20]
Y66W Y145F M153T V163A [gf21]
Y66H Y145F [gf22]
Y66W Y145F [gf23]
S65T Y145F [gf24]
F64L S65T M153A [gf25]
S65T M153A [gf26]
F64L S65T M153T V163A [gf27]
Y66H M153T V163A [gf28]
Y66W M153T V163A [gf29]
Y66H N146I M153A [gf30]
Y66W N146I M153A [gf31]
S65T N146I M153A [gf32]
S65T N146I M153T V163A [gf33]
S65T N146I [gf34]
Y66H Y145F M153A [gf35]
Y66W Y145F M153A [gf36]
S65T Y145F M153A [gf37]
S65T Y145F M153T V163A [gf38]
Y66H M153A [gf39]
Y66W M153A [gf40]
S65T M153T V163A [gf41]
Notes: We now have some experience with variants 2, 3, 4, 5, 6, 13,
15, 19, 22, 23, 25, 26. All of these show marked improvement in (as
observed by standard FITC illumination) over wild-type gfp. All forms
with S65 modifications cause an excitation "red shift", reducing
fluorescence with near UV (e.g. dapi, Hoechst filter sets), and yielding
a surprisingly strong "red signal" using rhodamine filter sets (green
illumination). We saw comparable initial activity levels from
gf2,3,4,5,6 & 25. The gf3, gf4, and gf25 variants exhibited the best
photo-stability, while the gf6 variant was markedly less stable to
photobleaching than the others. gfp modifications : Cormack et al.,
Gene 173:33; Heim et al., PNAS 91, 12501; Heim & Tsien, Cur. Biol. 6:178
III. Vectors for insertional tagging of C. elegans exons with gfp
In many cases, it is desirable to study a functional gene by the
insertion of gfp coding sequences. In general, this is carried out by
inserting gfp coding sequences in frame without losing any of the
original sequences. In many cases the resulting GFP-tagged gene
products can retain the function of the original gene while acquiring
fluorescence from the GFP component. GFP appears remarkably well suited
to such a tagging approach, since it can retain its fluorescence
properties in the context of both C-terminal and N-terminal amino acid
extensions. In addition, GFP can confer fluorescence properties in a
wide variety of cellular compartments, including [but by no means
limited to] cytoplasm, nucleus, mitochondria, and extracellular spaces.
Construction of in-frame fusions can be carried out by several
means. The most general is to choose optimal points within protein
sequence for GFP insertion, then use site directed mutagenesis and PCR
to insert gfp precisely at those sites. In other cases, there are
existing unique restriction sites in the coding region that can be used
for direct insertion of gfp. In the latter case, PCR could be used to
generate GFP coding sequences with appropriate linker sequences for
in-frame insertion. This requires rather careful analysis of the final
product to rule out mutations in the oligonucleotides used for PCR, or
in the intervening material.
As an alternative, we have produced a set of six clones with
pre-made gfp cassettes flanked with numerous restrictions sites placed
in all different reading frames. The current set of vectors is
sufficiently complete that virtually any common restriction site in any
reading frame can be insertion-tagged with gfp.
_-------------------------------------------------------------
Figure 1: General structure of exon tagging vectors
---R1-R2-R3-R4-R5---GfpCodingSequences---R1-R2-R3-R4-R5---
Where R1-Rn are restriction enzyme sites with different overhanging
ends. Vectors are designed so that cutting with any single enzyme
produces a single reading frame entering and leaving gfp. Six different
vectors with distinct multiple-cloning-site regions have been produced
and are sufficient to tag virtually any common restriction site.
_-------------------------------------------------------------
IV. Vectors for insertional tagging of C. elegans introns with gfp.
In some cases where a unique restriction site is found in an
intron, it has been possible to "tag" the gene of interest by inserting
gfp with flanking splice junctions. The geometry of this is shown below.
_-------------------------------------------------------------
Figure 2. Construction of gfp-tagged genes using intron insertion
A. Original Gene
EXON-[5'splice site]-INTRON-[3'splice site]-EXON
|
unique restriction site [RX]
B. gfp segment for insertion into unique restriction site [RX].
RX-[3' Splice site]-GFP CODING SEQUENCE-[5' Splice Site]-RX
_-------------------------------------------------------------
We provide three such vectors, covering all reading frames. A
variety of restriction sites (including EcoRI, BamHI, XmaI, SalI, Hin3,
MluI, NotI, SfcI, XbaI, ClaI, SacI, ApaI, PstI, and several blunt sites)
are duplicated both upstream and downstream of the splice junctions
flanking gfp. These are otherwise unique, so that each is available for
constructing insertion clones. To the extent that splice junctions can
occur at hinges or domain boundaries in protein coding sequences, it
might be expected that this scheme will often yield functional chimeric
proteins.
V. Vectors for construction of gfp-tagged constructs using PCR.
For PCR-based tagging schemes, we have constructed a set of gfp
vectors which are slightly modified versions of currently available
vectors. The modifications involve the addition of restriction sites in
N-terminal and C-terminal coding regions flanking gfp. These vectors
save several steps in PCR-based construction of chimeric genes in which
the coding region of interest is to be placed either 5', 3', or
surrounding gfp. In particular these vectors should be of significant
utility in cases where the sequence of a gene is available (e.g., from
the sequencing project) before the availability of a well defined
subclone.