|
Gene silencing
has become a popular technique thanks to the
introduction of RNAi-based tools, but typical
approaches can’t be used to understand the role of
protein domains, posttranslational modifications, or
single residues in protein function. This technical
brief describes an RNAi-based technique that enables
classical genetic studies of single cells, by
substituting a normal gene product with a specifically
mutated protein.
Introduction
The immense capacity of
mutational analysis of protein function was first
limited to research on random organismal mutations
isolated through screening procedures. Modern
molecular genetics has greatly extended capacity to derive and work with
mutants, including recessive mutations, at the whole
organism level through knock-out and knock-in
procedures. Such analysis is largely limited to
nonlethal mutations. On the single-cell level,
analysis in mammalian cells is largely confined to
dominant mutations. Analysis is possible through use
of homologous recombination to examine recessive
mutations (1,2), or inducible expression following
silent selection to follow the fate of cells
expressing dominant mutants (3).
The recent introduction
of RNAi technology (4), silencing the translation of a
specific gene product, has enabled a revolution in
understanding the function of individual proteins.
Gene silencing, while of tremendous value, does not enable
understanding of the role of protein domains,
posttranslational modifications, or single residues in
protein function — the “meat” of classical
genetic analysis. The RNAi approach simply suppresses
the protein’s function.
Click to enlarge
We have developed an
RNAi-based technique that enables classical genetics
on a single-cell basis by substituting a normal gene
product by a specifically mutated protein (5). The
initial procedure we used permitted domain analysis of
the important mitotic protein Aurora B, and enabled us
to conclude that a small C-terminal region of the
protein was critical to its correct localization and
function (5).
The procedure involved
introducing two plasmids into a single cell, one
expressing shRNA, which would ablate the normal gene
product, and the other expressing the mutant
substitute. The success of the procedure depended on knowing, within a single
cell, that both the original protein was absent and
that the mutant protein was expressed. For this
purpose, we used an antibody that recognized a
discrete small sequence of the native protein whose
mutation was not consequential to function. We then
designed the mutant protein so that this sequence
contained mutant residues that eliminated recognition
of the mutant protein by the antibody. We then
introduced silent mutations in the siRNA binding
sequence of the cDNA so that the shRNA would not
silence the replacement protein when the native
protein was suppressed. Finally, we put a marker (such
as HA antigen) on the replacement protein so that its
expression would be evident in the cell (Figure 1).
This technique both
eliminated the native protein and introduced a
specific replacement. We were able to demonstrate, in
a single cell, both that the native protein was absent
and that the replacement was expressed. We then used
these cells for mutant analysis by immunofluorescence
microscopy.
Below, we describe some
pitfalls in applying RNAi technology and review
opportunities that our approach offers.
Overcoming Pitfalls
First, when the protein
is an enzyme, function may not be completely
suppressed even though only a small amount remains
expressed. In this case, expression of a recessive
mutant usually will adequately substitute, and the
replacement will be dominant in the case the native
protein is minimally present, whereas such substitute
function would not be evident were the native protein
to be fully expressed. Work with protein substitution
involving enzymes needs careful controls to ensure
that the native activity is absent.
Second, if there is
more than one isoform of the protein to be suppressed,
then all isoforms must be suppressed at once for
replacement to give meaningful data. Multiple siRNA
knockdowns have been applied to a single cell (6) and
are successful.
Third, there may be
difficulty in selecting the correct antibody against
the native protein. Antibodies raised against defined
internal sequence may be used successfully, as we have
reported. It is simpler, when practicable, to use
antibodies that recognize N- or C-terminal sequence.
Normally, antibodies can be successfully raised
against N- and C-terminal sequence and are specific to
the
sequence only when it is at the end; for example, the
C-terminal glutamate on the tubulin α subunit is
antigenic only when exposed at the end (7). Thus, when
the sequence is linked to a tag, it will no longer be
recognized by the antibody. Therefore, an experiment
can be designed so that the substitute protein can
contain an N- or C-terminal tag, which will both
permit unique recognition and suppress recognition by
an antibody that reacts with the native protein.
Fourth, when doing
high-throughput screens, it is not practical to mutate
the protein of interest. As a solution, one can target
siRNA to untranslated downstream sequence in the mRNA
and substitute the suppressed target with a mutant
message that does not contain the downstream sequence
(8). Thus, only the native protein will be ablated.
This approach requires simple truncation of the
message rather than multiple silent mutant
substitution within the translated sequence to enable
selective RNAi knockdown. Such an approach is
especially practical with sequence that is already
modified at the N-terminus to contain a tag, as
introduction of a single mutation would be sufficient
for both purposes.
Prospects
The approach we
describe here will prove to be of substantial value
for analyzing mutations important to human health.
Present protocols require analysis of cells long after
the impact of dominant expression has created
secondary alterations and responses in the host cell.
Proteins of interest to tumorigenesis, such as BRCA1,
Aurora A, p53, and APC are among the important factors
in tumorigenesis that are frequently mutant in tumors.
Analysis of the immediate impact of these mutations on
the cell cycle and on cell development will become
practical once one has, in hand, the capacity to
express the mutant in a cell following ablation of the
normal protein, as described here.
Similarly, studies to
determine drug specificity toward a particular protein
target will be greatly aided by generating control
cells that have been induced to express a
drug-resistant mutant target protein in the absence of the drug-sensitive protein.
Protein replacement
need not always be of the wild type by its mutant
counterpart. One can also do the inverse: rescue the
mutant by wild-type proteins. RNAi is increasingly
explored as a means of gene therapy. The prospect of
mutant suppression accompanied by wild-type rescue is,
in principal, quite appealing. For instance, as a
long-term prospect, inherited mutations that highly
predispose to specific tumors, such as in BRCA1 (9),
associated with breast and ovarian tumors, and in APC
(10), associated with colon carcinoma, could be
targeted for suppression and replacement by wild-type
counterparts within the affected tissues.
1 Sidney Kimmel Cancer Center,
10835 Road To The Cure, San Diego, CA 92121, USA
2 Institut Albert Bonniot, INSERM
U309, 38706 La Tronche Cedex, France
3 Institut de Biologie Structurale J-P.
Ebel 41, rue Jules Horowitz, 38027
Grenoble Cedex 1, France
References
1. Capecchi, M.R.
“Altering the genome by homologous recombination.”
Science 244, 1288-92; 1989.
2. Mortensen, R.M. et
al. “Production of homozygous mutant ES cells with a
single targeting construct.” Mol Cell Biol
12, 2391-5; 1992.
3. Terada, Y. et al.
“AIM-1: a mammalian midbody-associated protein
required for cytokinesis.” EMBO J 17, 667-76;
1998.
4. Elbashir, S.M. et
al. “Duplexes of 21-nucleotide RNAs mediate RNA
interference in cultured mammalian cells.” Nature
411, 494-8; 2001.
5. Scrittori, L. et al.
“A small C-terminal sequence of Aurora B is
responsible for localization and function.” Mol
Biol Cell 16, 292-305; 2005.
6. Toyoda, Y., and
Yanagida, M. “Coordinated requirements of human Topo
II and cohesin for metaphase centromere alignment
under Mad2-dependent spindle checkpoint
surveillance.” Mol Biol Cell 17, 2287-302;
2006.
7. Paturle, L. et al.
“Complete separation of tyrosinated, detyrosinated,
and nontyrosinatable brain tubulin subpopulations
using affinity chromatography.” Biochemistry
28, 2698-704; 1989.
8. Zhu, C. et al.
“Functional analysis of human microtubule-based
motor proteins, the kinesins and dyneins, in mitosis/cytokinesis
using RNA interference.” Mol Biol Cell 16,
3187-99; 2005.
9. Narod, S.A., and
Foulkes, W.D. “BRCA1 and BRCA2: 1994 and beyond.” Nat
Rev Cancer 4, 665-76; 2004.
10. Fodde, R. et al.
“Mutations in the APC tumour suppressor gene cause
chromosomal instability.” Nat Cell Biol 3,
433-8; 2001.
*Address correspondence
to Robert L. Margolis. E-mail: rmargolis@skcc.org.
Submit your comments
on this article:
|
