Characterization and evaluation of matrix attachment regions in rice (Oryza sativa L.)

Jacqueline Peterson Rosales, Purdue University

Abstract

Matrix attachment regions (MARs) are non-coding, non-homologous, and A-T-rich segments of DNA. They are hypothesized to secure loops of chromatin to the proteinaceous matrix of the eukaryotic nucleus. Although they have been shown to be cis-acting to the gene(s) within the loops they border, the particular activity MARs impose cannot be generalized. In some observations, the surrounded genes are insulated from the influence of nearby chromatin. Often, the expression of these genes is enhanced. A more thorough understanding of the function and a clearer definition of these DNA elements are of great interest to both basic and applied scientists. In this work, matrix attachment regions (MARs) from the 5′ region of the rice actin 1 (Act1) gene as well as from Oryza sativa L. (rice) nuclei were isolated and cloned. DNA-matrix binding assays positively identified these six clones as MARs, and were used to observe comparative binding of selected MARs to the proteins of the nuclear matrix. DNA fragments with both an A+T content greater than 60% and presence of numerous MAR motifs were associated with specific matrix binding; most of these MARs also contained repetitive DNA observed in other non-coding regions of the rice genome. A hypothetical 4.1 kilobase-pair “loop” domain was cloned and described, in terms of its contents: two MARs that were positioned 5 ′ and 3′ to a putative floral (carboxyl methyltransferase) gene and a 3′ adjoining long terminal repeat (LTR). These MARs, when surrounding a CaMV 35S promoter-gusA -nos construct, appeared to impart anther-specific GUS expression to flowers of transgenic rice plants. Transgene expression as reflected by both transient GUS activity from rice protoplasts and GUS activity from stably-transformed rice plants was observed. Upon stable transformation, when the 5′ Act1 MAR was placed both 5′ and 3′ of the Act1 promoter-gusA-nos construct, mean GUS activity was less than that observed with the Act1 promoter-gusA-nos construct with no MAR, but three times as high as that seen in the Act1 promoter-gusA-nos construct with the 5′ Act1 MAR in the 5′ position only. This was in reverse order of the relative GUS activities shown in transient assays, when the 5'′ Act1 MAR-Act1 promoter-gusA-nos construct resulted in the highest GUS activity. When the CaMV 35S promoter was substituted for the Act1 promoter, flanking 5′ and 3′ (5 ′ Act1) MARs were associated with mean GUS activity four times that seen in the CaMV 35S promoter-gusA-nos construct with no MAR in stably-transformed plants. The very low and similar transient GUS activity observed with both of these constructs increased relative to the other constructs upon stable transformation. Despite these observations, the relatively small population studied and factors other than chromatin domain that were involved, such as multiple integration sites and rearrangement of the constructs within many of the transformants, prevent any final conclusion being made on the role that the 5′ Act1 MAR played in gene expression.

Degree

Ph.D.

Advisors

Hodges, Purdue University.

Subject Area

Molecular biology|Cellular biology

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS