Protein Knotting

REFERENCE: Mallam and Jackson. “Knot formation in newly translated proteins is spontaneous and accelerated by chaperonins.” Nature Chemical Biology (2012) 8, pgs 147 – 153.

                YibK and YbeA are Escherichia coli proteins that are known to form trefoil knots (Figure 15.1) before folding to their full native structures.  Previous work has shown that these proteins will denature without losing their trefoil knot and spontaneously refold to a functional protein once placed in renaturing conditions.  Mallam and Jackson address in a recent paper published in Nature Chemical Biology how this trefoil knot is initially formed within the cell.

                The authors first worked in a cell free system using a coupled transcription/translation set up that only included the necessary components (tRNAs, rNTPs, enzymes, etc) and determined that in vitro translated protein behaved the same as bacterially expressed protein.  Using a pulse-proteolysis experimental set up, a time course of folding for the protein was determined.  Interestingly, a 10 – 20 minute lag time was seen between the translation and final protein folding and rate constants of 0.05 – 0.09 min^-1 were elucidated.  

                Curious if bacterial chaperones could help folding, GroEL in complex with GroES was added.  Strikingly, no delay was noted from translation to folded protein and rate constants increased by almost 20-fold.  Adding the chaperonin complex to denatured YibK and YbeA did not substantially increase the rate of folding leading to the conclusion that GroEL-GroES specifically facilitate trefoil knot formation.  An excellent overview of their work is shown in Figure 15.2, which is directly from the authors’ paper.

                Even under highly denaturing conditions, the undoing of the trefoil knot has never been observed.  It seems that proteins knot only once in their lifetimes.  This work gives scientists a step towards understanding protein folding.

               

How to: Autophosphorylation

REFERENCES:

Malecka & Peterson. “Face-to-Face, Pak-to-Pak.” (2011) Structure 19(12), pgs 1723 – 1724.

Wang et al. “Structural Insights into the Autoactivation Mechanism of p21-Activated Protein Kinase.” (2011) Structure 19(12), pgs 1752 – 1761. 

                Many kinases require the phosphorylation of a residue within their active site to help maintain a conformation that is compatible with substrate binding/kinase activity.  While some kinases are able to phosphorylate other kinases, many times it is the kinase itself which performs autophosphorylation.  However, it is an interesting question: if the kinase requires phosphorylation to function but it itself must provide the phosphorylation, how does that work?  It is nearly a “chicken or the egg” problem.

                In the issue of Structure published on December 6^th, two papers discuss the de novo phosphorylation of the Pak1 (p21-activated kinase 1) protein.  A commentary offered by Malecka and Peterson discuss the history of common mutations often used by bench scientists to achieve “kinase-dead” proteins that are useful for crystallization as well as an overview of other kinases whose crystal structures reveal dimeric structures where the active loop of one subunit is placed within the active site of the other subunit.  The authors discuss two ways in which kinases can transiently adopt active conformations within the dimer to achieve autophosphorylation in trans: symmetric and asymmetric.  Symmetric structures, such as those seen for the kinases Chk2 and Ire1, place each others activation loops in their active sites while asymmetric structures, such as DAPK3 and IGF1R, have only one subunit place its activation loop within the active site of the other subunit.  

                Pak1, as reported by Wang et al, adopts a symmetric trans-autophosphorylation structure.  Interestingly, the authors also report that the common lysine mutations made within the active site are not 100% kinase-dead so caution should be used when trusting them as such.