MCB 201 Gene Expression - Spring Semester 2003
Lecture 15 (Regulation of Transcription Initiation cont.)
Section 10.5 (Eukaryotic Transcription Activators and Repressors)
1. Introduction: both prokaryotic and eukaryotic cells have nucleotide sequences in their DNA genomes that serve as binding sites for trans-acting regulatory proteins, i.e. transcription factors. In this section, eukaryotic transcriptional activators will be used as examples, although eukaryotic repressors are also known.
2. In mammals and other vertebrates, transcription factors are usually identified biochemically, since the genetic systems of these organisms are difficult to manipulate in the laboratory.
-Figure 10-35a: use of chromatography to purify a transcription factor, including the very powerful purification step of sequence-specific DNA affinity chromatography. For this technique, the DNA element to which the factor binds must be known. Weakly bound proteins (low affinity) are removed by washing the column with a buffer containing a low concentration of KCl (low salt buffer). Proteins that bind tightly to this sequence can be removed only with high concentrations of KCl.
-Figure 10-35b: How can we assay the proteins separated on the chromatography columns for their ability to bind to a known regulatory element? The method shown here is called DNase I footprinting, described in more detail below:
-Figure 10-6: DNase I footprinting. For this method, DNA fragments containing the control element are labeled on their 5' ends using radioactive ATP(gamma 32P-labeled) and polynucleotide kinase. The labeled DNA is then incubated in either the presence or absence of the column fraction or protein sample, followed by addition of a low concentration of DNase I (chosen so that each DNA molecule is cut once), which cuts the phosphodiester bonds in the polynucleotide chains at random, except where they are protected by the bound transcription factor. The protein is then removed from the DNA, which is denatured to single strands, and the DNA pieces are separated by gel electrophoresis. Note the protected area, or footprint.
-Figure 10-7: Another method is called electrophoretic mobility shift assay (EMSA) for DNA-binding proteins. This method also uses radioactively labeled DNA fragments. The physical basis of the method is that fragments of DNA to which protein is bound, i.e. the transcription factor, migrate more slowly during gel electrophoresis than unbound DNA.
-Figure 10-36: The final test is to determine if the isolated protein can stimulate transcription. Here an in vitro transcription reaction is used to test the activity of protein thought to be the transcriptional activator SP1. The assay includes template DNA containing the regulatory element (here, a GC-rich element), a HeLa cell nuclear extract containing RNA polymerase II and general transcription factors, labeled ribonucleoside triphosphates. The reactions are incubated with or without the protein being tested. The readout of this assay is the amount of radioactive RNA produced. The RNA products are first separated by gel electrophoresis and then visualized by autoradiography.
3. Table 4-2 (page 118, Genetic Code): How do we go from an isolated transcription factor to isolating the gene that encodes it? A common strategy is to obtain some amino acid sequence from the protein. Using the genetic code, it is possible to work back from amino acid sequence to RNA codons, but due to the degeneracy of the code, more than one RNA sequence is possible to translate into the same amino acid sequence. DNA copies of these sequences can be used as probes to find the gene by complementary base pairing and screening of either cDNA or genomic libraries. Once identified by the screen, the piece of DNA of interest can be sequenced and also tested to see if it actually encodes a transcription factor using the following assay:
Figure 10-37: In vivo assay for transcription factor activity. Two plasmids are required: one carries the gene thought to encode a transcription factor, and the second carries a reporter gene and the control element to which the transcription factor binds. Control cells receive only the plasmid carrying the reporter gene. The host cell is chosen so that it does not have the transcription factor gene being tested and it does not have the reporter gene. The idea is that the transcription factor gene will be transcribed into mRNA which will be translated into the putative transcription factor. If this is truly a transcriptional activator, then the rate of transcription of the reporter gene will increase when this factor is bound to its control element.
4. Figure 10-38: Evidence that transcriptional activators are modular proteins composed of distinct functional domains. The yeast Gal4 protein is used here as an example. This is an activator of genes that encode proteins used to break down the sugar galactose. This experiment requires a reporter gene DNA construct containing the lacZ gene (encoding E. coli b-galactosidase), a eukaryotic TATA box (promoter, RNA polymerase II binding site) and the control element to which Gal4 binds, called UASgal. UAS = Upstream Activating Sequence, an enhancer-like sequence. The second DNA construct required carries the gene encoding Gal4, and this is actually a series of DNA constructs that carry a series of deletions of the parts of the Gal4 gene. The host cells carry a mutated Gal4 gene, so they do not produce this factor, which would hopelessly complicate the assay. B-galactosidase activity is measured and also the abilities of the mutated Gal4 proteins to bind to the UASgal DNA element are measured. The data indicated that deletions that result in loss of amino acids from the N-terminal region of Gal4 disrupt binding to the UAS sequence, whereas, loss of sequences from the C-terminal end of Gal4 result in loss of activator activity as judged by decreased production of B-galactosidase.
5. Figure 10-39: We can represent the two separate Gal4 domains by the top diagram. Notice that there can be different arrangements of activation domains and DNA binding domains in different transcriptional activators. Usually, there is only one DNA binding domain, but there can be multiple activator domains. Note the relatively unstructured, flexible domains that connect these functional domains. These regions make transcriptional activators very sensitive to proteases, and this may allow cells to downregulate their levels quickly by protein degradation.
6. DNA binding domains are classified into a variety of structural types based on the domains of these proteins that bind to the DNA, usually in the major groove. In this region, hydrogen bonds and van der Waals interactions occur between atoms in the protein and in the DNA. Contacts in the minor groove are also made by some transcription factors. X-ray crystallography has given us detailed information on these complexes between transcription factor and DNA.