|
|
 |
 |
 |
 |
|
Steve Spiker |
||
 |
|
The thrust of the research in my laboratory is the role of chromatin structure in the control of gene expression in higher plants. In order for genes to be expressed, transcription factors and RNA polymerases must bind to control sequences. These sequences are normally inaccessible due to the structure of chromatin. What changes in chromatin structure are necessary to unblock the control sequences? How do these changes occur? In order to approach these questions, we study the physical properties of histones and interactions of these proteins, the major protein components of chromosomes, with DNA. We also study a group of chromosomal proteins, the HMG proteins, which have been implicated as structural proteins of transcriptionally active chromatin. For the last several years we have focused on the MAR (matrix attachment region) DNA sequences that anchor chromatin fibers to the nuclear matrix and generate loop domains that can have either a transcriptionally active or inactive structure as depicted in panel A of the figure just above. In this figure the solid blue bar represents a fiber of the nuclear matrix. The yellow boxes represent MAR sequences. In the domain on the left, DNA is organized into nucleosome fibers to form a relatively open structure that is accessible to transcription factors. In the domain on the right, the nucleosome fibers are supercoiled to form an inaccessible, transcriptionally inactive chromatin fiber. The chromatin domains shown in panel A are relevant to the "position effect" model to explain the low and variable expression of introduced genes in transgenic plants and animals. According to this model, if a transgene becomes incorporated into an inactive domain (like the one on the right), it would be transcribed at a very low rate or not at all. To test this model, we have made DNA constructs in which a transgene is flanked by a cloned MAR from tobacco plants (red boxes in panel B). This construct is then introduced into tobacco cells in culture using the "gene gun." Expression of the transgene is measured and compared to expression from a control construct, which lacks the flanking MAR sequences. According to our model as depicted in panel B, the cloned MAR sequences would allow the transgene to form an independent and transcriptionally active domain that would be free of the influence of the chromatin domain into which it became incorporated. In support of this model, we observe that the level of gene expression in the MAR-flanked transgene is much greater than that of controls lacking MARs. We are now investigating the mechanism of how such transcriptionally active, independent domains might be formed, and we are exploiting MAR sequences to make plants that express transgenes at high levels. Some images involving the nuclear matrix are shown at the top of the page. On the left (bottom four panels) are fluorescence images of an ethidium bromide stained tobacco cell, an isolated tobacco nucleus, a nucleus in which histones have been extracted to release the coiling restraints on the DNA and allow it to spill out to form a nuclear "halo" and the nuclear matrix that results from removing DNA in the "loop domains" leaving only the DNA directly attached to the nuclear matrix (MARs by operational definition). The top four panels are the same images but are viewed by differential interference contrast (DIC) rather than by fluorescence. These images were provided by Bekir Ülker, while he was a postdoc in the lab. The next two panels contain electron microscope images of the nuclear matrix. The image on the left is a transmission electron micrograph of a section of a human nuclear matrix obtained from Capco et al. (1982) Cell 29, 847-858. The image on the right is a scanning electron micrograph of a tobacco nuclear matrix. The matrix was isolated by Tom Phelen when he was a graduate student in the lab, and the scanning electron micrograph of the matrix was made by Tuyen Nguyen working in Bill Thompson's lab. The ethidium bromide-stained image on the right was provided by Bekir Ülker. Histones were added back to tobacco nuclear halos collapsing the DNA in the loop domains and making them visible by light microscopy. The image beside the diagram of the loop domain model is an autoradiogram of a matrix binding assay. A plasmid vector carrying a DNA insert of a suspected MAR is cut with restriction enzymes to release the vector (V) and the insert (I). The two DNA molecules are radioactively end labeled and an aliquot of the total preparation (T) is run on an agarose gel. Another aliquot is mixed with isolated tobacco nuclear matrices. The insoluble matrices along with any bound DNA are collected as a pellet (P) by centrifugation. DNA molecules with a strong affinity for the nuclear matrix are found in the pellet with the matrix and are by operational definition MARs (matrix attachment regions). DNA molecules with low affinity for the nuclear matrix remain in the supernatant (S). In this assay the MAR in the matrix binding assay is pS116-1 isolated by Susan Michalowski while she was a graduate student in the lab. She nicknamed the MAR "Big Dog" because of its extremely high affinity for the nuclear matrix. The MAR fractionates almost entirely in the pellet bound to the nuclear matrix, while the vector DNA remains in the supernatant. |
|
