Polyadenylation of mRNA: minimal substrates and a requirement for the 2' hydroxyl of the U in AAUAAA.

mRNA-specific polyadenylation can be assayed in vitro by using synthetic RNAs that end at or near the natural cleavage site. This reaction requires the highly conserved sequence AAUAAA. At least two distinct nuclear components, an AAUAAA specificity factor and poly(A) polymerase, are required to catalyze the reaction. In this study, we identified structural features of the RNA substrate that are critical for mRNA-specific polyadenylation. We found that a substrate that contained only 11 nucleotides, of which the first six were AAUAAA, underwent AAUAAA-specific polyadenylation. This is the shortest substrate we have used that supports polyadenylation: removal of a single nucleotide from either end of this RNA abolished the reaction. Although AAUAAA appeared to be the only strict sequence requirement for polyadenylation, the number of nucleotides between AAUAAA and the 3' end was critical. Substrates with seven or fewer nucleotides beyond AAUAAA received poly(A) with decreased efficiency yet still bound efficiently to specificity factor. We infer that on these shortened substrates, poly(A) polymerase cannot simultaneously contact the specificity factor bound to AAUAAA and the 3' end of the RNA. By incorporating 2'-deoxyuridine into the U of AAUAAA, we demonstrated that the 2' hydroxyl of the U in AAUAAA was required for the binding of specificity factor to the substrate and hence for poly(A) addition. This finding may indicate that at least one of the factors involved in the interaction with AAUAAA is a protein.

Alteration by prolactin of surface charge and membrane fluidity of rat 13762 mammary ascites tumor cells.

Intraperitoneal injection of ovine prolactin (100 micrograms/d) in Fischer 344 rats bearing transplantable 13762 mammary ascites tumor (MAT) cells modifies the surface charge density and membrane fluidity of the tumor cells. In each of five experiments the mean electrophoretic mobility (epm) of MAT cells taken from prolactin-treated rats was significantly lower than that of cells from nonhormone-treated controls. Prolactin concentrations were increased in vivo by (a) direct intraperitoneal injection of ovine prolactin; (b) subcutaneous implantation of diethylstilbestrol-containing silastic capsules to produce pituitary prolactin secreting tumors; or (c) a single subcutaneous injection of polyestradiol phosphate, a long-acting estrogen. In an effort to establish that the prolactin effect was a direct one, two in vivo protocols were used: (a) MAT cells were coincubated with anterior pituitary halves obtained from nontumor-bearing littermates; or (b) rat or ovine prolactin was added to serum-free culture media containing MAT cells. In both protocols, the epm of the prolactin-treated cells was significantly lower. The isoelectric focusing pH of whole cells was increased by prolactin treatment from 4.93 to 5.12, consistent with a reduction in the number of surface carboxyl groups. The fluidity of membranes of treated cells was drastically increased, as measured by spin-label probe rotation rates. These combined results imply that the hormone exerts its effect by stimulating events in the cell that lead to a reduction of the average density of carboxylic acid residues on the tumor cell surface.