A systematic investigation of the effect of acid on the denaturation of some 20 monomeric proteins indicates that several different types of conformational behavior occur, depending on the protein, the acid, the presence of salts or denaturant, and the temperature. Three major types of effects were observed. Type I proteins, when titrated with HCl in the absence of salts, show two transitions, initially unfolding in the vicinity of pH 3-4 and then refolding to a molten globule-like conformation, the A state, at lower pH. Two variations in this behavior were noted: some type I proteins, when titrated with HCl in the absence of salts, show only partial unfolding at pH 2 before the transition to the molten globule state; others of this class form an A state that is a less compact from of the molten globule state. In the presence of salts, these proteins transform directly from the native state to the molten globule conformation. Type II proteins, upon acid titration, do not fully unfold but directly transform to the molten globule state, typically in the vicinity of pH 3. Type III proteins show no significant unfolding to pH as low as 1, but may be caused to behave similarly to type I in the presence of urea. Thus, the exact behavior of a given protein at low pH is a complex interplay between a variety of stabilizing and destabilizing forces, some of which are very sensitive to the environment. In particular, the protein conformation is quite sensitive to salts (anions) that affect the electrostatic interactions, denaturants, and temperature, which cause additional global destabilization.(ABSTRACT TRUNCATED AT 250 WORDS)

The molecular chaperone proteins, particularly Hsp60 and Hsp70, have been implicated in essential cell functions under both normal and stress conditions (reviewed in refs 1-5). Members of the family of heat-shock proteins of M(r) 70K, Hsp70, bind to unfolded proteins and short peptides. Addition of Mg-ATP results in the dissociation of the substrate polypeptides from the chaperone, but as ATP-gamma S (an ATP analogue that is only slowly hydrolysable) cannot substitute for ATP in this reaction, it has been concluded that ATP hydrolysis is necessary to dissociate Hsp70-substrate protein complexes. By independently measuring the rates of ATP hydrolysis and substrate protein dissociation, we show here that Mg-ATP binding but not Mg-ATP hydrolysis is essential for substrate dissociation. We also show that there is an absolute requirement for K+ for the effect of Mg-ATP: only the combination of K+ and Mg-ATP will cause the conformational change in Hsp70 that is necessary for substrate dissociation. Moreover, in the absence of K+, Mg-ATP favours complex formation. We consider these results in terms of a G-protein-like model.

Circular dichroism and HPLC gel filtration were used to show that cytosolic hsp7O is thermally stable but undergoes a conformational transition (midpoint, 430C; 57C in the presence of ATP or ADP) leading to oligomerization. hsp7O binds to unfolded, but not to folded, proteins in a temperature-dependent manner; complex formation is significant only at physiologically relevant temperatures. hsp7O binds ADP more tightly than ATP to form a binary complex, which binds to the unfolded protein more rapidly than free hsp7O. ADP also inhibits the ATP-induced dissociation of the hsp7O-protein complex. A regulatory role for the hsp7O-nucleotide binary complexes is proposed.Recent studies indicate that protein folding and assembly events in vivo are mediated by intracellular components now being referred to as molecular chaperones (reviewed in refs. 1-3). One class of these chaperones is the 70-kDa heat shock protein, hsp70. The cytosolic forms of mammalian hsp70 are present in cells as two different gene products (4): a stressinducible form, hsp72, and a constitutive member, hsp73, also known as hsc70. hsp70 family members have been shown to interact with a number of proteins undergoing maturation in the cell (5-8). In addition to these nascent events, it has been proposed that hsp70 proteins can recognize and bind to mature polypeptides that may have become unfolded in the cell (e.g., following heat shock) and thereby facilitate their refolding (9). To date relatively little experimental evidence has been provided to support this idea. There are ample data demonstrating that nascent polypeptides, unable to properly fold or assemble, remain bound to their particular hsp70 chaperone (5, 10, 11). In vitro studies have shown that hsp70 proteins interact with a variety of short synthetic polypeptides (12).In the present study, we have utilized an in vitro system to examine the possible interaction of the mammalian hsp70 with both folded and unfolded polypeptides. We show that the cytosolic hsp73 will form stable complexes with a variety of unfolded protein targets, but not with their properly folded counterparts. Furthermore, we present data regarding the role of temperature and nucleotides in these interactions. MATERIALS AND METHODSMaterials. Bovine brain hsp73 and human hsp72/73 from HeLa cells were isolated and purified as described (13). Reduced carboxymethylated a-lactalbumin (RCMLA), bovine a-lactalbumin (type I) (a-LA), and horse heart cytochrome c (type VI) were obtained from Sigma. Reduced carboxamidomethylated ribonuclease A was prepared as described (14). Staphyloccocal nuclease (SNase) was purified from a cloned system provided by D. (15). Gel-filtration chromatography was performed with a Beckman HPLC instrument using a Bio SEC-250 silica column (600 x 7.5 mm;Bio-Rad) at 22°C; 20 mM sodium phosphate buffer/0.20 M KCl, pH 6.5 (buffer A), was used as the mobile phase; the flow rate was 1 ml/min. Typical elution volumes for hsp73, dimer, trimer, hsp73 soluble aggregates, reduced carboxymethylated a-lactalbu...

1,1'-Bis(4-anilino-5-naphthalenesulfonic acid) (bis-ANS), a hydrophobic fluorescent molecular probe which has been shown to bind to compact intermediate states of proteins (molten globules) and also to many nucleotide binding sites, induces a conformational change in DnaK by preferentially binding to its partially folded intermediate state (I) and thus shifting the equilibrium from favoring the native state (N) to favoring the I state. The conformational change was detected by CD, fluorescence emission, size exclusion chromatography, and small-angle X-ray scattering. The presence of bis-ANS significantly decreases the midpoint, Tm, of the initial transition (N-->I) in the thermal unfolding of DnaK, resulting in the apparent destabilization of the native state of DnaK. There is a linear correlation between the apparent free energy (reflected by Tm) of this transition and the concentration of bis-ANS. Bis-ANS does not affect the midpoint of the transition for DnaK from the intermediate to the unfolded state (U). An additional small transition from I to I*, a more expanded intermediate state, was observed, suggesting that the thermal denaturation of DnaK proceeds via a four-state (N-->I-->I*-->U) unfolding process. The addition of nucleotides, ADP or ATP, to the DnaK-bis-ANS complex causes a decrease in bis-ANS fluorescence emission due to the release of bound bis-ANS from the intermediate state of DnaK. This is due to preferential binding of the nucleotide to the native state of DnaK, resulting in a shift in the equilibrium from the intermediate toward the native state rather than the direct displacement of bis-ANS bound in the nucleotide binding site. Denaturation of DnaK induced by bis-ANS can be minimized by working at a temperature much lower than the Tm of the protein, at low dye concentration, and in the presence of nucleotide. Under these conditions, bis-ANS binds to the native state of DnaK.

2,4-Dinitroanisole, CTH6N2Os, Mr= 198, monoclinic, P21/n, a=8.772(2), b= 12.645 (2), c = 15.429 (4)A, fl= 81.89 (2) ° , V= 1694 A 3, Z= 8, D m not determined, Dx= 1.56 g cm -3, 2(MoK~q)= 0.7093 A, /t = 1.26 cm -~, F(000) = 816, T= 298 K, R r = 0.075 for 1916 observed reflections. 2,6-Dinitroanisole, CTH6N205, Mr=198, triclinic, P1, a= 3.854 (1), b = 7.736 (2), c= 14.476 (4) A, a= 89.25 (2), fl=86.97 (2), y= 75.59 (2) ° , V=417 A 3, Z = 2, D m not determined, D x= 1.58 g cm -3, 2(Mo Kch) = 0.7093 A, g = 1.28 cm -I, F(000) = 204, T= 298 K, R F = 0.069 for 1046 observed reflections. In the 2,4-dinitro compound, the two independent methoxy groups make torsion angles of 5 and 13 o with the ring planes, whereas in the 2,6-dinitro compound the angles are 72 and 79 ° . These results agree with those of related crystal structures. The rate of replacement of methoxy with an amino substituent appears to be correlated with out-of-plane twist of the methoxy group.