|Home | About | Journals | Submit | Contact Us | Français|
The heat shock (HS) response is the major cellular defense mechanism against acute exposure to environmental stresses. The hallmark of the HS response, which is conserved in all eukaryotes, is the rapid and massive induction of expression of a set of cytoprotective genes. Most of the induction occurs at the level of transcription. The master regulator, heat shock transcription factor (HSF, or HSF1 in vertebrates), is responsible for the induction of HS gene transcription in response to elevated temperature. Under normal conditions HSF is present in the cell as an inactive monomer. During HS, HSF trimerizes and binds to a consensus sequence in the promoter of HS genes, stimulating their transcription by up to 200-fold. We have shown that a large, non-coding RNA, HSR1, and the translation elongation factor eEF1A form a complex with HSF during HS and are required for its activation.
The HS response is an evolutionarily ancient, conserved reaction of cells to unfavorable environmental conditions. During HS major changes in the pattern of gene expression occur, including the rapid induction of a specific set of genes. Most of these genes encode heat shock proteins (HSPs), a large family of molecular chaperones, which includes several functional and molecular weight sub-families. HSPs are essential for cell survival under normal conditions and are critical for cell survival during stress. Elevated expression of HSPs is a major cause of cancer cell resistance to therapies (1, 2). In some instances, HSPs were found to be critical for malignant transformation due to their ability to stabilize mutant, metastable forms of oncoproteins, such as ras or p53 (3). Expression of HSPs is regulated predominantly at the level of transcription by heat shock transcription factor (HSF) (4, 5). HSF is ubiquitously expressed as an inactive monomer. The activation of HSF during HS requires its trimerization, nuclear translocation, binding to a cognate DNA sequence in the promoter of HS genes, phosphorylation, and, finally, activation of transcription by rescue of transcriptionally engaged, arrested RNA polymerase II molecules in the 5' coding region of most of HSP genes (6, 7). We have shown that the first steps in the activation of HSF in response to HS require two positive regulators: a large, non-coding RNA, HSR1, which provides thermosensing capabilities, and translation elongation factor eEF1A, which promotes trimerization of HSF (8).
HSR1 is isolated by the affinity fractionation of a lysate of heat shocked cells on a column of HSF covalently attached to Sepharose . The functional characterization of HSR1 is performed in a reconstituted, in vitro system using recombinant HSF1 and eEF1A isolated from HeLa cells. Since HSF spontaneously trimerizes at high concentrations, it is expressed and purified by procedures that keep its concentration low. Activation of HSF is determined by electrphoretic mobility shift assays (EMSA).
Protein-denaturing SDS polyacrylamide gel electrphoresis (SDS-PAGE) is performed according to Laemmli, with minor modifications (9).
Purified HSF trimerizes spontaneously at high concentration (10)(11). Therefore the conditions for HSF expression and purifications are selected to minimze the spontaneous activation of HSF by keeping its concentration relatively low. Just before coupling to periodate-activated Sepharose the HSF is exchanged into high-pH borate buffer, which also helps to prevent spontaneous trimerization by keeping HSF molecules charged. The monomeric state of HSF to be bound to the activated Sepharose is essential for the successful capture of the eEF1A-HSR1 complex from the HS cell lysate.
Once HSR has been isolated and cloned (by a combination of 3' and 5' RACE or by another strategy of choice), it can be synthesized in large amounts by in vitro transcription by phage RNA polymerase. Of all the commercially available RNA polymerases, T7 RNA polymerase provides the greatest yield. The yield tends to be higher when transcribing from a linearized plasmid template, but care should be taken to avoid the introduction of as few extra nucleotides into the resulting transcript as possible.
The active trimeric form of HSF binds to the consensus heat shock element (HSE) sequence with high affinity. HSE is defined as an array of repeats of the sequence 5'-nGAAn-3' in head-to-tail orientation. The minimum of three HSE units is required for the high affinity binding of HSF (12–14). We routinely use four-element arrays of the consensus HSE element as a probe in gel shift experiments involving HSF activation. The in vitro activation is performed by incubating the RNA at the desired temperature, followed by the addition of the proteins (HSF and eEF1A), the probe, and the degree of HSF activation is assayed by EMSA. The activation is conveniently performed in a PCR cycler, which allows for the fast and precise control of the incubation temperature. To screen antisense oligonucleotides, the oligonucleotide to be tested is pre-mixed with HSR1 at 5–10 fold molar excess.
This procedure covalently couples HSF to the activated Sepharose via reaction of the epsilon-amino groups of lysine residues in the protein with reactive aldehyde groups of the periodate-treated Sepharose. The mild oxidation of Sepharose by periodate results in conversion of 1,2-cis-diol groups to the corresponding dialdehydes. The aldehydes then reacts with the amino groups to yield a Schiff's bases, which are stabilized by mild reduction with sodium borohydrate.
The S-100 lysate, which is the starting material for this procedure, can be obtained commercially or prepared according to the published method (15). This protocol is an adaptation of the original procedure described by Merrick (16). With minimal modifications the procedure can be easily adapted to use tissue, for example rat liver, as the source of the protein. This protocol assumes that 50 ml of the S-100 lysate is used; the column and gradient volumes can be scaled up or down if greater or lesser amounts are used.
This procedure has been used extensively with cells grown in monolayer, such as HeLa, BHK, or 3T3. However, it should be easily adaptable to the cells grown in suspension, if the ratio of cells to HSF-Sepharose is kept constant. Monolayer cells are grown in T-75 phenolic cap flasks (Corning) to ca. 75–90% confluency.
HSR1 is cloned in the pBluescript SK- vector (pBSKM-HSR1) in an orientation such that T7 RNA polymerase generates the sense transcript after digestion of the plasmid with SmaI and T3 RNA polymerase generates the antisense transcript after digestion of the plasmid with EcoRV. It is essential to cleave the plasmid with restriction enzymes that produce blunt ends, if possible, as this will minimize non-specific transcription initiation on protruding single stranded ends. Alternatively, a PCR product containing the T7 promoter sequence in one of the primers can be used as template to minimize the introduction of extra nucleotides in the RNA sequence.
|water:||to 200 μl|
|5× transcription buffer:||40 μl||1×|
|100 mM DTT:||10 μl||5 mM|
|25 mM rNTP (ROCHE):||30 μl||3.75 mM|
|T7 RNA polymerase:||1000–2000 U/ml|
The typical range of concentrations in which HSF is activated in this system is between 1 and 5 nM final. Higher concentrations, especially in excess of 100 nM, tend to cause spontaneous trimerization and activation of HSF. Because of the fluctuations in the quality of the recombinant HSF preparations, it may be necessary to titrate each new preparation of HSF to determine the concentration at which it is not activated spontaneously. The final concentration of eEF-1A is 100 nM. The concentration of proteins is determined by UV absorption at 280 nm in 6 M guanidin-HCl/20 mM phosphate buffer, pH 6.5 (e(HSF)=28670 l*M−2 and e(eEF-1A)=45755 l*M−2 in a 1 cm cuvette). The final concentration of HSR1 is 0.1 nM. A low concentration of HSR1 is crucial because at higher concentrations the equilibrium concentration of the HSF-activating conformation of HSR1 is high enough to cause activation to occur independent of incubation temperature. However, for the antisense oligo screen, this “constitutive” activation is beneficial as it considerably simplifies the experimental setup.
This work was supported by the NIH grant R01 GM069800 (E.N.)
1Ammonium persulfate solution is stable for up to two weeks if stored at −20°C and kept on ice when thawed.
2Prepare Coomassie staining solution by first dissolving the dye powder in methanol (stir on a magnetic stirrer for at least 1 h), then add acetic acid and water.
3Avoid heating urea-containing solutions to high temperature, as urea will decompose.
4Phosphocellulose P11 column is prepared as described at: http://mitchison.med.harvard.edu/protocols/tubprep.html
5The efficient cooling of the sample during sonication is essential. It is best achieved if the sonication is done in a metal vial (we routinely use tube adapters from an older Beckman ultracentrifuge) to ensure rapid and efficient dissipation of heat.
6The volumes and incubation times given are for a 4% gel in a Bio-Rad mini electrophoresis apparatus. They should be adjusted accordingly if using larger gels and/or higher concentration of acrylamide. Larger 4% gels may be difficult to handle.