In principle, the folded structure of any protein is determined by its primary sequence. Relatively small proteins can often refold from a denatured state into the correct folded state under suitable conditions without any help (Nobel Prize to Anfinsen for showing this with ribonuclease; we did this experiment in biochemistry student lab with lysozyme).
However, some proteins, and I think this becomes more true as the proteins become larger and more complex, need help to fold in a reasonable amount of time in vivo. This assistance is provided by chaperones, like Hsp70. Chaperones also help refold proteins after they become denatured by stresses such as heat (which is why Hsp stands for heat shock protein).
Larger proteins often contain multiple domains, each of which may fold as a separate unit, so that the large protein essentially consists of 2 or more smaller, connected proteins.
Some proteins are oligomers, either homooligomers (several identical subunits) or heterooligomers (2 or more different subunits). The formation of this quaternary structure must also be considered.
In principle, the folded structure of any protein is determined by its primary sequence. Relatively small proteins can often refold from a denatured state into the correct folded state under suitable conditions without any help (Nobel Prize to Anfinsen for showing this with ribonuclease; we did this experiment in biochemistry student lab with lysozyme).
However, some proteins, and I think this becomes more true as the proteins become larger and more complex, need help to fold in a reasonable amount of time in vivo. This assistance is provided by chaperones, like Hsp70. Chaperones also help refold proteins after they become denatured by stresses such as heat (which is why Hsp stands for heat shock protein).
Larger proteins often contain multiple domains, each of which may fold as a separate unit, so that the large protein essentially consists of 2 or more smaller, connected proteins.
Some proteins are oligomers, either homooligomers (several identical subunits) or heterooligomers (2 or more different subunits). The formation of this quaternary structure must also be considered.
The main difference is in complexity. As was pointed by Adam large proteins are often multidomain proteins. Thus, it is expected that folding mechanisms is characterized by the presence of intermediates.
Structural complexities often correlate with experimental difficulties. However, unfolding and reversible refolding of some large proteins –including a few membrane proteins- was successful in vitro (without the aid of chaperones). Some of them seem to fold at equilibrium following an apparent two-states model. At this point more experimental studies -mainly kinetic ones- are needed.
As Alexeii Finkelstein wrote in 2006: “Physics of large multi-domain proteins and protein complexes is far from maturity, though; but one may believe that its future development will be based on physics of single-domain proteins, like the development of chemistry is based on physics of atoms and electrons.”