X hits on this document

65 views

0 shares

0 downloads

0 comments

4 / 22

1048 Chemical Reviews, 1998, Vol. 98, No. 3

isomerase, that catalyze the unscrambling of nonna- tive disulfide bonds. 107,108

A more readily tractable issue in the folding of reduced RNase A involves the particular disulfide bond between Cys65 and Cys72. Along the polypep- tide chain, Cys58 and Cys72 are equidistant from Cys65. According to polymer theory alone,105,109 the stability of a disulfide bond between Cys58 and Cys65 should be equal to that of a disulfide bond between Cys65 and Cys72. Yet, in the M-peptide (which encompasses residues 50-79110), the Cys65-Cys72 disulfide bond is 3.6-fold more stable than is the Cys58-Cys65 disulfide bond.111 This bias is consis- tent with the loop structure formed by the Cys65- Cys72 disulfide being a nucleation site for the folding

of RNase A under oxidizing conditions.104,111-114

Un-

der reducing conditions, however, the Cys65-Cys72 disulfide bond is vulnerable. That bond and the Cys40-Cys95 disulfide bond are the first in native RNase A to suffer reduction by dithiothreitol. 115

B. Prolyl Peptide Bond Isomerization

RNase A exhibits a slow kinetic phase in its refolding (that is, its folding with native disulfide bonds intact).116 The existence of this second kinetic phase is due to the presence of at least two distinct

forms of enzyme is

unfolded RNase A.117-120 If the native unfolded rapidly and then allowed to refold

immediately,

all

of

the

molecules

refold

rapidly.

But

if refolding is delayed, 80% slowly. The simplest kinetic

of the molecules refold scheme that is consis-

tent with these data is

Us

slow} U

f

9 8 f a s t N

(2)

where N is the native enzyme, Uf are fast refolding species, and Us are slow refolding species. The trans isomer of a typical peptide bond is greatly favored over the cis isomer. In contrast, a trans bond preceding a proline residue is only slightly favored, and its conversion to cis can be slow on the time scale of protein folding. In native RNase A, the peptide bonds to Pro42 and Pro117 are trans and those to Pro93 and Pro114 are cis. The isomerization of one or both of the cis peptide bonds may be responsible for the slow kinetic phase observed during the refolding of RNase A. The conservation of Pro93 and Pro114 in pancreatic ribonucleases from different vertebrates,20,22 which is particularly rare for residues in a surface loop,121 corroborates the importance of a cis peptide bond at these positions. 122

The role of prolyl peptide bond isomerization in the refolding of RNase A has been probed by site-directed mutagenesis. The refolding rate of P42A RNase A is similar to that of the wild-type enzyme, indicating that cis-trans isomerization of the Pro42 peptide bond does not hinder refolding.123,124 Nonetheless, a hydrogen bond from the side chain of Tyr97, which has the least mobile side chain of the six tyrosine and three phenylalanine residues,67 to the Pro42 peptide bond enhances stability substantially.125,126 The re- folding kinetics of P93A, P114A, and P117A RNase A differ significantly from that of the wild-type enzyme.124 This difference has allowed for an elabo- ration of the scheme in eq 2 to include additional species. 124

Raines

A major conclusion from work on the refolding of RNase A is that the Pro93 peptide bond is trans in the slowest refolding species.124,127,128 In other words, the trans-to-cis isomerization of that bond is the slowest step in the refolding of the fully denatured enzyme. The kinetics of refolding suggest that the analogous peptide bond is cis in P93A RNase A. Yet in the three-dimensional structure of crystalline P93G RNase A, this bond is trans because Gly93 allows the formation of a type II -turn. 124,129 130

V. RNA Binding

The forces that lead to the binding of proteins to double-stranded DNA are becoming apparent. By comparison, the forces that lead to the affinity and specificity of proteins for single-stranded RNA are relatively unknown.135 RNase A is being used to reveal detailed information on the binding of proteins to RNA. 131-134

A. Subsites

The number of lysine (10) and arginine (4) residues in RNase A exceeds that of aspartate (5) and glutamate (5) residues. Accordingly, RNase A is cationic (pI ) 9.3136) at physiological pH. RNase A has been shown to destabilize double-stranded DNA by binding to single strands.137 Moreover, cation titration138 suggests that RNase A can occlude eleven nucleotides of a single-stranded nucleic acid139 and that binding involves seven Coulombic interac- tions.140 These results suggest that the interaction between the enzyme and a single-stranded nucleic

acid extends

well beyond

the

scissile bond.

Structural74-78

and functional

(vide

infra) data di-

vulge the existence of several enzymic (Figure 2). The subsites of RNase A have subject of recent reviews. 141,142

subsites been the

Figure 2. Apparent interactions between the subsites in RNase A and a bound molecule of RNA. The 12 indicated residues have been shown by site-directed mutagenesis to make a contribution to substrate binding or turnover (or

158,159 146-148 158,159 159 153,270 37 149 Gln11, both). These residues are Lys7, Arg10, 154 148 153,155-157 149 His12, Lys41, Thr45, Lys66, Asn71, His119,154,46 Glu111, Asp83, and Asp121. Phe120

is also likely to contribute to the P1 subsite (via its main chain) and the B1 subsite (via its side chain).164 The numbers in parentheses refer to the conservation of the

indicated residues in pancreatic ribonucleases.

20,22

Document info
Document views65
Page views65
Page last viewedSat Dec 03 05:10:52 UTC 2016
Pages22
Paragraphs1374
Words20315

Comments