Moss Tom. DNA-protein interactions: principles and protocols
Подождите немного. Документ загружается.
Genetic Analysis of DNA-Binding Proteins 441
digest with SalI and an enzyme recognizing the fusion junction between the DNA-
binding protein and the VP16 activation domain (see Note 10). A stock of this
gapped plasmid can be prepared in 10 mM Tris–Cl, pH 8.0, and 1 mM EDTA at
a concentration of 10–100 ng/µL and stored at –20°C.
2. Prepare competent yeast cells:
a. Grow a 50-mL culture of the yeast strain containing the reporter plasmid
(if mutations in the DNA-binding protein are to be analyzed) or expression
plasmid (if mutations in the DNA recognition site are to be analyzed) over-
night at 30°C until the OD
600
= 0.5–1.0.
b. Pellet the yeast cells by spinning for 5 min in a clinical centrifuge (1000g) at
room temperature.
c. Resuspend the cells in 10 mL of 100 mM lithium acetate, 1 mM EDTA, and
10 mM Tris–Cl, pH 8.0. Pellet again as in step 2b.
d. Resuspend again in 10 mL of 100 mM lithium acetate, 1 mM EDTA, and 10 mM
Tris–Cl, pH 8.0.
e. Incubate at 30°C for 30 min without agitation and pellet again as in step 2b.
f. Resuspend in 500 mL of 100 mM lithium acetate, 1 mM EDTA, and 10 mM
Tris–Cl, pH 8.0. Place on ice until used for transformation.
3. Combine 100 µL competent yeast cells with 40 µL of the error-prone PCR mixture
(Subheading 3.2.), 50 µg sonicated, denatured salmon sperm DNA, and 150 ng
gapped plasmid (see Notes 11 and 12).
4. Incubate at 30°C for 30 min without agitation.
5. Add 0.5 mL of 40% PEG (3350 mol. wt.) in 100 mM lithium acetate, 1 mM
EDTA, and 10 mM Tris–Cl, pH 8.0, and incubate further at 30°C for 60 min.
6. Incubate at 37°C for 5 min.
7. Centrifuge for 10 s at room temperature in a microcentrifuge at 14,000 rpm (16,000g).
8. Remove the supernatant and resuspend in 100 µL SOS solution.
9. Plate half of the resuspended transformed cells on each of the two C-agar plates
lacking the appropriate nutrients for selection of both expression and reporter
plasmids (e.g., C-ura-trp).
10. Incubate at 30°C for 2–3 d, until robust colony growth is obtained.
3.4. Screening for Mutants
1. Initial screen.
a. Replica plate the colonies obtained in step 10 of Subheading 3.3. onto selec-
tive C agar plates (C-ura-trp, for example) containing 40 µg/mL X-gal.
b. Place replica plates at 30°C for 2–3 d and then at room temperature for an
additional 2–3 d.
c. Identify potentially interesting mutants by color. These could include light
blue, white, or dark blue colonies; in each case, the color is in comparison to
that exhibited by colonies containing wild-type versions of both the expres-
sion and reporter plasmids.
2. Pick candidates for further study and respot on selective C-agar plates. Grow for
about 2 d at 30°C. Replica plate onto indicator media and score colony color
442 Setzer, Schulman, and Bumbulis
again after growth as in step 1. If the phenotype is consistent with that seen
initially, proceed to step 3.
3. Enrich for cells containing only the mutagenized plasmid.
a. Pick one or more colonies with an interesting phenotype from selective C-agar
plates without X-gal indicator. Grow in 2 mL liquid C-trp medium (if the
expression plasmid has been mutagenized; grow in C-ura medium if the
reporter plasmid has been mutagenized) overnight at 30°C.
b. Prepare a 1:200 dilution of this overnight culture in liquid medium and plate
10 µL of this diluted culture on a C-trp plate (assuming throughout that the
expression plasmid has been mutagenized).
c. After 3 d of growth at 30°C, replica plate onto C-ura-trp as well as onto C-trp.
d. Grow overnight at 30°C and compare the colonies that grew on the two rep-
lica plates. Choose several that grew on C-trp but not on C-ura-trp. These are
likely to have lost the reporter plasmid but not the expression plasmid. Use
these to inoculate 2-mL liquid cultures of C-trp and grow overnight at 30°C.
4. Isolate the mutated plasmid using Method 1 as described by Strathern and Higgins
(10) (see Note 13).
5. Use the isolated plasmid to transform E. coli. Pick individual colonies and per-
form standard plasmid minipreps. Analyze the plasmids thus obtained using
appropriate restriction digests to ensure that the plasmid isolated is the muta-
genized plasmid (e.g., the expression plasmid) and not the other plasmid compo-
nent of the system (e.g., the reporter plasmid).
6. Analyze isolated plasmids by retransformation.
a. Use the purified plasmid to retransform the appropriate yeast strain (already
containing the reporter plasmid) and obtain colonies on selective C-ura-trp
plates of cells with both reporter and expression plasmids.
b. Analyze transformants by replica plating onto indicator plates as described
in step 1. If the phenotype observed is consistent with that seen initially, then
one can conclude that the phenotype is plasmid dependent. If the expression
plasmid was mutagenized and it is the DNA-binding protein that is under
analysis, continue with step 7. If, instead, the reporter plasmid was muta-
genized and is being studied, one can move directly to Subheading 3.5.
7. Western blot screen to ensure the expressed fusion protein is full length and
expressed at normal levels
a. Inoculate 30 mL of C-ura-trp medium with a mutant strain and grow over-
night at 30°C.
b. Pellet the cells in a clinical centrifuge at (1000g) for 5 min. Resuspend the
pellet in 1 mL of 2% SDS.
c. Add 1 g glass beads and vortex vigorously for 15 min.
d. Transfer the mixture to another tube and spin in a microcentrifuge at top speed
(16,000g) for 15 s. Transfer the supernatant to another tube.
e. Determine protein concentration in the extract using the BCA method
(reagents from Pierce) (11) with bovine serum albumin as the standard.
Genetic Analysis of DNA-Binding Proteins 443
f. Using standard Western blot methods, analyze 20 µg total protein from each
mutant and compare to equivalent amounts of protein from a control strain
expressing wild-type DNA-binding protein as well as a second, negative con-
trol strain containing the parent plasmid from which the expression construct
was derived, but which lacks a coding sequence for the DNA-binding protein.
The primary antibody used in the Western blot should be specific for the
DNA-binding protein of interest; if such antibodies are unavailable, it may be
possible to obtain, from commercial sources, antibodies to the activation
domain used in the fusion. The point of the Western blot exercise is to exclude
from further consideration those mutants whose steady-state level of expres-
sion is different from that of wild-type protein or which are truncated as a
result of chain-termination or frame-shift mutations. Some truncation mutants
may be interesting, but recall that the activation domain will be completely
absent from truncated proteins if the activation domain was fused to the
C-terminus of the DNA-binding protein. We have found that a substantial
fraction of potential loss-of-function mutants (white phenotype) of TFIIIA
are truncation mutants and a Western blot screen was important in removing
them from further, more laborious analysis (5).
3.5. Analysis of Mutants
3.5.1. Quantitative Determination of
β
-Galactosidase Activity
We have found that relative DNA-binding affinities of various mutant forms
of TFIIIA can be predicted with reasonable precision from measurements
of β-galactosidase activity in vivo in yeast reporter strains expressing the
mutant protein fused to the activation domain of VP16. Whether this will
prove to be generally true remains to be seen, but β-galactosidase activities
can be readily measured using standard methods and may prove informative.
1. Inoculate 12.5 mL liquid C-ura-trp medium with the yeast strain to be analyzed.
In addition, prepare a similar culture of a control strain containing the same
reporter plasmid but with the parent vector (e.g., pG1) of the expression con-
struct, rather than the expression construct itself. This control will permit
determination of β-galactosidase activity in the absence of the DNA–protein
interaction of interest. Grow the cultures overnight at 30°C until the OD
600
=
0.3–1.0. Note the actual optical density.
2. Pellet cells by spinning at 1000g for 5 min in a clinical centrifuge. Resuspend in
3 mL of Z buffer.
3. Divide into three 1-mL fractions. In addition, process a tube containing 1 mL Z
buffer in parallel. This will serve as the blank for spectrophotometric determina-
tion of β-galactosidase activity. Add 30 µL chloroform and 20 µL of 0.1% SDS to
each and vortex for 10 s.
4. Incubate at 30°C for 5 min and add 200 µL ONPG at a concentration of 4 mg/mL
in Z buffer. Incubate at 30°C until a yellow color becomes apparent.
5. Add 500 mL Na
2
CO
3
and note the elapsed time since ONPG was added.
444 Setzer, Schulman, and Bumbulis
6. Transfer 800 µL of the reaction mixture to a microfuge tube and centrifuge at
16,000g for 2 min.
7. Measure the absorbance at 420 nm relative to that of the control processed in parallel.
8. Calculate the number of units of β-galactosidase activity according to the follow-
ing formula:
Absorbance at 420 nm × 100
(Assay duration [min]) (culture volume analyzed [mL]) (OD
600
of culture)
10. Average the three different determinations of activity to obtain a single value
(see Note 14).
11. Normalize the final average β-galactosidase activity relative to that of the nega-
tive control culture (see Note 14).
3.5.2. Sequence Determination
Ultimately, one must determine the sequence of the mutant DNA-binding
protein or recognition site in the expression or reporter plasmid, respectively.
The details of the sequencing strategy to be used will depend on the specific
system under analysis and the technology available to the investigator. We do
recommend that the complete set of screens described in Subheading 3.4. be
completed prior to sequence analysis, as the latter is typically laborious,
expensive, or both.
3.5.3. Biochemical Analysis of Mutants
Until correlations between in vivo phenotypes and DNA-binding affinity
and/or specificity can be verified in a particular system, the approach described
here can be applied best as a method for generating interesting mutants that can
be subjected to further biochemical analysis. In most cases, this will require
expression and purification of the mutant DNA-binding protein under analysis,
preferably without a fusion to the artificial activation domain. Again, the details
of how this can be done will depend on the specific interaction being studied.
We recommend, however, that the necessity of further sequence manipulation,
including subcloning into other plasmid vectors, be taken into account in the
initial design and construction of the yeast expression vector.
4. Notes
1. The genotype of strain BJ2168 is reported to be MATa ura3-52 trp1 leu2 pep4-3
prc1-407 prb1-1122 gal2 (6). A variety of alternative strains should also be suit-
able for use. The only relevant genotypes for the approach used here pertain to
the nutritional markers used for plasmid selection. If HIS3 were to be used as a
reporter, then it would also be important that the parent strain be his3- or that the
chromosomal HIS3 locus be placed under control of the hybrid promoter being
used in the analysis. Clearly, the mutant chromosomal alleles of the nutritional
Genetic Analysis of DNA-Binding Proteins 445
markers should be stable and not give rise to revertants at measurable frequen-
cies. BJ2168 also contains mutations in three vacuolar proteases (the PEP4,
PRC1, and PRB1 genes), but it is unlikely that these are relevant to the approach
described here. In fact, we have made use of other strains that presumably con-
tain wild-type alleles at these loci without detectably altering the results obtained
using the Xenopus TFIIIA/5S rRNA gene interaction.
2. A variety of reporter gene constructs probably can be used successfully. Plas-
mid-based reporters could be constructed using various parent plasmids contain-
ing different selectable markers, and alternative reporter genes might also be
chosen. In addition, the use of a chromosomally integrated reporter gene is pos-
sible; in fact, we have successfully used HIS3 as a chromosomal reporter gene in
this fashion to study the Xenopus TFIIIA–5S rRNA gene interaction. Advantages
of β-galactosidase as a reporter include the fact that it permits a range of pheno-
types to be scored and the ease with which quantitative measurements of activity
can be made using a simple spectrophotometric assay. HIS3, on the other hand, is
probably much more sensitive, permitting detection of weaker DNA–protein
interactions or use of a single-copy reporter when multicopy β-galactosidase
reporters are required. It also allows selection of rare gain-of-function mutants
that might occur at such a low frequency that identifying them would be difficult
or impossible using a screening protocol like that required with β-galactosidase
as a reporter. HIS3 expression does not provide a wide dynamic range of pheno-
types in a single-plate assay; however, this limitation can be overcome partially
by analyzing growth on plates containing variable 3-amino-triazole (3-AT)
concentrations. 3-AT is an inhibitor of the HIS3 gene product and can, therefore,
be used to adjust the level of HIS3 expression required to permit growth.
We have not investigated the use of different core promoter elements other
than that derived from CYC1. Even for CYC1, however, levels of reporter gene
expression potentially can be manipulated by adjusting the number of copies and
orientation of the DNA target sequence for the binding protein under study rela-
tive to the core promoter. We have found that both orientation and number of
TFIIIA-binding sites in the promoter can have substantial effects on reporter gene
activity when a TFIIIA-VP16 fusion protein is expressed. In our case, the orien-
tation dependence could be explained readily on the basis of the known polarity
of TFIIIA binding to the 5S rRNA gene (12). If similar information is available
for the DNA-binding protein of interest, it may be possible to design reporter
constructs rationally. Otherwise, it is probably advisable to generate multiple
constructs for testing. It is possible that homodimers binding to symmetrical sites
may exhibit little or no orientation dependence. With TFIIIA, we found that
increasing the number of binding sites from one to two in the reporter gene pro-
moter resulted in a doubling of β-galactosidase activity (5). For many natural Pol
II activators, the effect of having multiple binding sites is synergistic (13), so it is
possible that even larger effects will be observed when the number of binding
sites for other DNA-binding proteins is increased. We have not investigated the
effects of varying spacing between the artificial upstream activator sequence and
446 Setzer, Schulman, and Bumbulis
the core promoter, but others have shown that transcriptional activity is only
moderately affected by alterations in the wild-type spacing between these two
elements (14–16). Nonetheless, it is possible that attaining maximal activity in
any particular system may require optimizing the spatial relationship between
the DNA recognition site of interest and the core promoter. In the end, it may be
necessary or desirable to adjust several of these parameters to obtain a level of
reporter gene activity that is phenotypically detectable but that also provides for
reasonable sensitivity in the detection of mutants weakening the DNA–protein
interaction under analysis.
3. A variety of alternative expression plasmids are available, including several from
commercial sources that may be preferable to pG1 in some respects (see Sub-
heading 3.1.2.). We have used vectors other than pG1 and, in some cases, these
vectors are better engineered for the easy introduction of DNA fragments encod-
ing the DNA-binding protein of interest while producing an in-frame fusion to an
N-terminal activation domain. Choice of expression vector may be affected by
the selectable marker used, the presence of an activation domain in the vector,
whether the domain is to be N- or C-terminally fused to the DNA-binding domain,
and the strength of the promoter driving expression of the fusion protein. It is
also possible that regulated expression of the fusion protein will be desirable, in
which case a regulated promoter, like that of the GAL1 gene, might be chosen
over the constitutive GPD promoter of pG1.
4. The VP16 activation domain we have used (8) is probably the best studied acti-
vation domain acting to stimulate transcription by RNA polymerase II, and is
probably also the strongest one in common use. Depending on the system being
studied (see Note 5) and the choice of expression plasmid, an alternative activa-
tion domain might be chosen. We have used both the GAL4 activation domain
and the synthetic domain B42 in addition to that of VP16 to study the Xenopus
TFIIIA–5S rRNA gene interaction.
5. It may be necessary to adjust the system’s parameters to achieve a level of tran-
scriptional activity that affords a reasonable sensitivity of the assay to variations
in binding affinity. Potentially, activity may be affected not only by the affinity
of the interaction but also by the number, spacing, functional synergy between,
and orientation of the binding sites in the reporter construct (see Note 2), by the
level of expression of the fusion protein (see Note 3) and the fraction of protein
that is available for binding (effective in vivo concentration), by the copy number
of the reporter gene (see Note 2), by the choice of reporter gene (see Note 2), and
by the choice of activation domain (see Note 4). Our experience in analyzing the
Xenopus TFIIIA–5S rRNA gene interaction suggests that, for TFIIIA–VP16
under control of the GPD promoter on a 2 µ-containing plasmid, the effective in
vivo concentration of protein was much lower than the total in vivo concentra-
tion as determined by Western blot (5). Transcriptional activation in this system
could not be detected on X-gal plates when the K
d
was above 2 nM, using an E. coli
β-galactosidase reporter gene containing two TFIIIA-binding sites upstream of
the CYC1 promoter on a plasmid with a 2 µ origin of replication.
448 Setzer, Schulman, and Bumbulis
duced from the expression plasmid. To minimize this problem, we recommend
several steps. If possible, use two different restriction endonucleases to generate
the linear, gapped plasmid. Second, treat the cut plasmid with phosphatase to
dephosphorylate the 5' termini generated by restriction digestion. Third, gel-
purify the gapped plasmid to eliminate any small fragment that was removed by
the restriction digestion as well as to eliminate any uncut molecules that might
remain as a result of incomplete cleavage.
11. The quantities of gapped plasmid and error-prone PCR product that are optimal
for recovery of the maximal number of colonies containing plasmids with inte-
grated PCR products have not been carefully optimized, and probably should be.
We have some evidence that suggests that transformation efficiencies are
improved by reducing the amount of PCR product relative to the amount of
gapped plasmid recommended here.
12. We have routinely used the crude, unpurified products of the error-prone PCR
reaction in cotransformations with the gapped expression plasmid to generate
and study mutant forms of TFIIIA without encountering significant problems
(5). In some related applications, however, we have noted that a high background
of colonies can be produced as a consequence of poorly characterized recombi-
nation events that we believe derive from the production and integration of aber-
rant PCR products into the gapped target plasmid. It is most likely that these are
small DNA fragments, perhaps “primer dimers.” If these are produced at any-
thing other than a very low frequency and are capable of recombining with the
target plasmid, they can complicate the identification of loss-of-function mutants
in the DNA-binding protein of interest, because they result in the production of
white colonies on X-gal plates. If initial studies in a new system suggest that a
similar problem occurs with the use of the crude PCR product, then we recom-
mend purifying the mutagenized PCR product prior to recombination/transfor-
mation. Simply gel-isolating a product of the correct length should eliminate any
problems derived from the presence of unincorporated primers, primer dimers, or
other aberrant products of the PCR.
13. As minor modifications to the procedure of Strathern and Higgins (10), we have
isolated plasmids directly from yeast colonies in lieu of growing liquid cultures,
have used Prep-a-Gene (Bio-Rad) rather than Geneclean reagents, do not pre-
cipitate DNA after elution from Prep-a-Gene, and transform E. coli with 10% of
the DNA sample recovered from the Prep-a-Gene reagent (2 µL of 20 µL total).
14. We recommend repeating the triplicate assays described here for each mutant at
least three times with independent cultures to obtain an overall average value.
Normalization of activity to a standard seems to be important in correcting day-
to-day variability in the determination of β-galactosidase activity.
References
1. Brent, R. and Ptashne, M. (1985) A eukaryotic transcriptional activator bearing
the DNA specificity of a prokaryotic repressor. Cell 43, 729–736.
Genetic Analysis of DNA-Binding Proteins 449
2. Muhlrad, D., Hunter, R., and Parker, R. (1992) A rapid method for localized
mutagenesis of yeast genes. Yeast 8, 79–82.
3. Kuipers, O. P. (1996) Random mutagenesis by using mixtures of dNTP and dITP
in PCR. Methods Mol. Biol. 57, 351–356.
4. Ma, H., Kunes, S., Schatz, P. J., and Botstein, D. (1987) Plasmid construction by
homologous recombination in yeast. Gene 58, 201–216.
5. Bumbulis, M. J., Wroblewski, G., McKean, D., and Setzer, D. R. (1998) Genetic
analysis of Xenopus transcription factor IIIA. J. Mol. Biol. 284, 1307–1322.
6. Jones, E. W. (1990) Vacuolar proteases in yeast Saccharomyces cerevisiae.
Methods Enzymol. 185, 372–386.
7. Schena, M. and Yamamoto, K. R. (1988) Mammalian glucocorticoid receptor
derivatives enhance transcription in yeast. Science 241, 965–967.
8. Triezenberg, S. J., Kingsbury, R. C., and McKnight, S. L. (1988) Functional dis-
section of VP16, the trans-activator of Herpes simplex virus immediate early gene
expression. Genes Dev. 2, 718–729.
9. Rose, M. D., Winston, F., and Hieter, P. (1990) Methods in Yeast Genetics:
A Laboratory Course Manual. Cold Spring Harbor Laboratory, Cold Spring
Harbor, NY.
10. Strathern, J. N. and Higgins, D. R. (1991) Recovery of plasmids from yeast into
Escherichia coli: shuttle vectors. Methods Enzymol. 194, 319–329.
11. Smith, P. K. (1985) Measurement of protein using bicinchoninic acid. Anal.
Biochem. 150, 76–85.
12. Smith, D. R., Jackson, I. J., and Brown, D. D. (1984) Domains of the positive
transcription factor specific for the Xenopus 5S RNA gene. Cell 37, 645–652.
13. Carey, M., Lin, Y.-S., Green, M. R., and Ptashne, M. (1990) A mechanism for
synergistic activation of a mammalian gene by GAL4 derivatives. Nature 345,
361–364.
14. Struhl, K. (1982) The yeast his3 promoter contains at least two distinct elements.
Proc. Natl. Acad. Sci. USA 79, 7385–7389.
15. Brent, R. and Ptashne, M. (1984) A bacterial repressor protein or a yeast tran-
scriptional terminator can block upstream activation of a yeast gene. Nature 312,
612–665.
16. Guarente, L. and Hoar, E. (1984) Upstream activation sites of the CYC1 gene of
Saccharomyces cerevisiae are active when inverted but not when placed down-
stream of the “TATA” box. Proc. Natl. Acad. Sci. USA 81, 7860–7864.
17. Baudin, A., Ozier-Kalogeropoulos, O., Denouel, A., Lacroute, F., and Cullin, C.
(1993) A simple and efficient method for direct gene deletion in Saccharomyces
cerevisiae. Nuceic Acids Res. 21, 3329–3330.
18. Oldenburg, K. R., Vo, K. T., Michaelis, S., and Paddon, C. (1997) Recombi-
nation-mediated PCR-directed plasmid construction in vivo in yeast. Nucleic
Acids Res. 25, 451–452.
Assays for Transcription Factor Activity 451
451
From:
Methods in Molecular Biology, vol. 148: DNA–Protein Interactions: Principles and Protocols, 2nd ed.
Edited by: T. Moss © Humana Press Inc., Totowa, NJ
31
Assays for Transcription Factor Activity
Virgil Rhodius, Nigel Savery, Annie Kolb, and Stephen Busby
1. Introduction
Most transcription activator proteins have three important features that can
be probed at the molecular level: they bind to specific sequences near promot-
ers, they can be interconverted between active and inactive forms by covalent
or noncovalent modification, and, when bound at target promoters, they can
stimulate the initiation of transcription by RNA polymerase (1). This chapter is
concerned with in vitro methods for measuring the transcription activation
function of this important class of proteins. In most cases, these methods are
applied to activators that have been substantially purified, and for which the
target sequences are known and the binding sites characterized (other chapters
in this volume cover methods for locating and investigating the binding sites
for such activators). Here we are concerned with the measurement of the prod-
ucts of activation. Because Escherichia coli transcription activators have been
studied more than any others, we will take these as the paradigm, though, in
principle, the techniques can be applied to any organism for which in vitro
systems have been developed.
The starting point of the methodology was the observation, made in the early
1970s, that purified E. coli RNA polymerase could initiate transcription at pro-
moters in purified DNA (2). With improvements in RNA methodology, it was
found that the transcription start site in vitro, in many cases, was the same as in
vivo. Further, at a number of promoters, interactions with specific transcrip-
tion activators could be demonstrated and factor-dependent transcription in
vitro occurred. The literature is now full of instances of factor-dependent tran-
scription initiation with purified proteins and DNA, setting the scene for stud-
ies on the mechanism of transcription activation. However, as with all in vitro
techniques, it is worth noting that the conditions found in the plastic tube differ
452 Rhodius et al.
greatly from those in vivo and that, for some instances, the reconstituted tran-
scription system simply may not work.
Two principal methods can be used to monitor transcription activation: tran-
script assays and abortive initiation (Fig. 1). In transcript assays, RNA poly-
merase is incubated with a DNA template carrying the promoter of interest
either with or without the transcription activator protein. Radioactively labeled
nucleotides are then added and RNA polymerase molecules that have formed
transcriptionally competent complexes start to make RNA. RNA polymerase
molecules then “run” to a suitably placed downstream terminator (or to the end
of the fragment), thus making labeled RNA of a discrete length that can be
monitored and quantified (3). In abortive initiation assays, nucleotide precur-
sors corresponding just to the start of the message are added. RNA polymerase
forms transcriptionally competent complexes at the chosen promoter, but elon-
gation is prevented because some of the four nucleoside triphosphates are with-
held (4). The result is that RNA polymerase is “trapped” at the promoter and
can only synthesize a short oligonucleotide, which is released as the polymerase
cycles between a number of conformations in the abortive complex (it is termed
abortive because a longer transcript cannot be made). The consequence of the
cycling is that each polymerase molecule, trapped at a single promoter, synthe-
sizes the oligonucleotide continuously, and the appearance of the product can
be measured directly. The rate of product formation will be dependent on the
number of promoters that are occupied by polymerase in an open complex.
This, in turn, will usually depend on the activity of the transcription factor
under study. In most cases, transcript assays are used to locate transcription
starts and to monitor factor activity qualitatively, whereas abortive initiation
assays are exploited for quantitative and kinetic work.
2. Materials
1. E. coli RNA polymerase. This enzyme can be purified (5,6) or can be purchased
from Boehringer (Mannheim, Germany), Pharmacia (Piscataway, NJ), or other
molecular biology companies. The enzyme can also be reconstituted from indi-
vidual subunits that have been overproduced using overexpression vectors (7,8).
Purity can be easily checked by denaturing polyacrylamide gel electrophoresis
and activity can be verified using standard templates. Preparations of RNA poly-
merase are usually stored at –20°C in buffer containing 50% glycerol.
2. Transcription factors. These must be purified, at least partially, away from any
nuclease activities. Transcription assays can be used to monitor the purification.
3. Plasmids and DNA fragments. The DNA used for transcription assays should be
free from nicks and RNase activity. Short DNA sequences carrying different pro-
moters can be constructed using standard recombinant DNA methodology and
cloned into plasmid vectors such that a strong transcription terminator is located
downstream of the promoter (9). Transcript assays can be performed directly on