Methylphosphonate mapping of phosphate contacts critical for RNA recognition by the human immunodeficiency virus tat and rev proteins.

The HIV-1 regulatory proteins tat and rev are both RNA binding proteins which recognize sequences in duplex RNA which are close to structural distortions. Here we identify phosphate contacts which are critical for each binding reaction by use of a new method. Model RNA binding sites are constructed carrying substitutions of individual phosphodiesters by uncharged methylphosphonate derivatives isolated separately as Rp and Sp diastereoisomers and tested for protein binding by competition assays. In the binding of tat to the trans-activation response region (TAR), three phosphates, P21 and P22 which are adjacent to the U-rich bulge and P40 on the opposite strand, are essential and in each case both isomers inhibit binding. Similarly, in the interaction between the HIV-1 rev protein and the rev-responsive element (RRE) both methylphosphonate isomers at P103, P104, P124 and P125 interfere with rev binding. At P106, only the Rp methylphosphonate isomer is impaired in rev binding ability and it is proposed that the Rp oxygen is hydrogen-bonded to an uncharged amino acid or to a main chain hydrogen atom. Synthetic chemistry techniques also provide evidence for the conformations of non-Watson-Crick G106:G129 and G105:A131 base-pairs in the RRE 'bubble' structure upon rev binding. Almost all functional groups on the 5 bulged residues in the bubble have been ruled out as sites of contact with rev but, by contrast, the N7-positions of each G residue in the flanking base-pairs are identified as sites of likely hydrogen-bonding to rev. The results show that both tat and rev recognize the major groove of distorted RNA helixes and that both proteins make specific contacts with phosphates which are displaced from the sites of base-pair contact.

Characterization of the protein-protein interactions determining the heat shock protein (hsp90.hsp70.hsp56) heterocomplex.

We have reported previously that the three heat shock proteins hsp56, hsp70, and hsp90 exist together in a heterocomplex in human lymphocyte cytosol (Sanchez, E. R., Faber, L. E., Henzel, W. J., and Pratt, W. B. (1990) Biochemistry 29, 5145-5152). All three of these proteins also exist in the native glucocorticoid receptor heterocomplex isolated from WCL2 cell cytosol and we have recently shown that the three heat shock proteins are present when immunopurified mouse glucocorticoid receptor is reconstituted into a heterocomplex by rabbit reticulocyte lysate (Hutchison, K. A., Scherrer, L. C., Czar, M. J., Ning, Y., Sanchez, E. R., Leach, K. L., Deibel, M. R., Jr., and Pratt, W. B. (1993) Biochemistry 32, 3953-3957). In this work, we show that highly purified mouse hsp90 binds in a reversible equilibrium to immunopurified rabbit hsp56, but hsp56 does not bind to purified mouse hsp70. In contrast to the equilibrium binding of hsp90 to hsp56, purified hsp90 binds poorly or not at all to purified hsp70 unless a third factor from reticulocyte lysate is present to permit complex formation. This hsp70.hsp90 complex-forming factor is heat-labile, and in the presence of this factor and ATP, a heat shock protein heterocomplex can be reconstituted from purified mouse hsp90 and hsp70 and rabbit hsp56 that is present in the factor preparation. Our data are consistent with a model in which hsp56 and hsp70 bind to different sites on hsp90 but do not interact with each other. The presence of hsp56 in the heat shock protein heterocomplex is not stabilized by molybdate but hsp56 is stabilized if the glucocorticoid receptor is present in addition to hsp90 and hsp70.