What is S-palmitoylation?
S-palmitoylation is a reversible post-translational protein modification process, whereby palmitic acid forms a thioester bond with the sulphur atom of the cysteine residue of a protein.
How is this relevant in cell signaling?
S-palmitoylation is involved in efficient trafficking and anchoring of multiple proteins to the plasma membrane amongst other targets. These include GPCRs, G-proteins, eNOS and multiple kinases. S-palmitoylation acts as a ‘switch’ to regulate receptor expression, intracellular protein activity and retrograde signals in neurons.
What if this applies to my research?
The CAPTUREomeTM S-Palmitoylated Protein Kit can be used to determine whether this fatty acid modification is occurring in your experimental systems.
Click below to find out more!
A simple animation to show the thioester bond between palmitic acid and a sulphur atom in the side chain of a cysteine residue
The post-translational fatty-acid acylation of some receptors and signalling molecules is a transient modification that alters the interaction of these signalling proteins with cell membranes. It provokes the relocation of specific individual proteins to the plasma membrane, controls their residence time in the plasma membrane, and their activity often by regulating the macromolecular complexes they can form. The most common lipid modification is S-palmitoylation1, which is the addition of palmitic acid to the sulphur atom in the side chain of a cysteine residue, through a thioester bond.
A unique feature of S-palmitoylation is the reversibility of the modification, causing it to act as a switch that regulates cell signalling by altering membrane localisation, subcellular trafficking and the binding capacity of different proteins2. Extensive research has elucidated the regulatory roles of S-palmitoylation throughout the body, with a focus on transmembrane proteins and signalling molecules.
The modification of receptors by S-palmitoylation in the Endoplasmic Reticulum (ER) and Golgi apparatus (GA) initiates the trafficking of receptors to the plasma membrane. Removal of these palmitate groups at the plasma membrane often leads to internalisation of the protein3. This cycle of S-palmitoylation-depalmitoylation of certain receptors causes a switch from high-level to low-level cell signalling.
One good example of this is the S-palmitoylation of G-protein coupled receptors (GPCRs), a process thought to be mediated by zinc finger-DHHC type proteins. These include ZDHHC-5 that has been shown to S-palmitoylate somatostatin receptor subtype 54. Research shows S-palmitoylation of GPCRs leads to efficient trafficking of the receptors to the plasma membrane. When S-palmitoylation of the δ opioid receptor was inhibited, cell surface expression of the GPCR fell by 61.4%; suggesting S-palmitoylation contributes to, but is not absolutely necessary for, the targeting of GPCRs to the membrane and/or trafficking vesicles5.
Additionally, data from the chemokine receptor CCR56 and the A1 adenosine receptor6 suggest many GPCRs that do not undergo S-palmitoylation, are degraded. Some hypothesise this may be due to a misfolding of the protein due to free cysteine sulphydryl group(s)3. This leads to a fall in concentration of membrane-bound GPCRs, and a decrease in cell signalling.
Another feature of S-palmitoylation is its ability to mediate the level of GPCR phosphorylation by PKA, and therefore internalisation of the receptor. A non-S-palmitoylated β2-adrenergic receptor mutant showed phosphorylation levels ~4-times that of wild type (WT), with the authors suggesting depalmitoylation of the receptor exposes additional phosphorylation sites8. Increased levels of phosphorylation may lead to internalisation and/or desensitisation of GPCRs, resulting in a reduced level of cell signalling3.
Not only does S-palmitoylation regulate GPCRs, but it also exhibits effects on the G-proteins themselves. All Gα subunits, excluding transducin, have been shown to be S-palmitoylated at the Golgi apparatus. This fatty acid modification, in combination with myristoylation and βγ association, is necessary for the targeting and binding of the α-subunit to the plasma membrane3. Research suggests S-palmitoylation also increases the affinity of the α-subunit for the βγ-complex, which in turn regulates the activity of the α-subunit and the intracellular signalling pathway3.
Evidence suggests S-palmitoylation regulates the activity of some signalling proteins, for example endothelial nitric oxide synthase (eNOS). eNos is dually acylated, first with the co-translational addition of a myristoyl group at the N-terminal (glycine-2), followed by the post-translational addition of two palmitate groups at cysteines 15 and 269. Five DHHC domain proteins have been associated with eNOS S-palmitoylation: DHHC-2, 3, 7, 8, and 21; significantly DHHC-21 co-localises with eNOS in vascular endothelial cells (the principal site of eNOS activity)10.
Firstly, co-translational myristoylation and S-palmitoylation of eNOS at the Golgi apparatus causes the enzyme to be trafficked to caveolae11,12, a similar mechanism to that of aforementioned GPCRs. A non-S-palmitoylated eNOS mutant exhibited deficient localisation at the plasma membrane, with a significant increase in concentration at the Golgi apparatus and cytoplasm11. Additionally, each acylation process caused a 10-fold increase in trafficking of eNOS to cavolae, suggesting both modifications are essential for optimal membrane targeting12.
Furthermore, myristoylation followed by S-palmitoylation increases the affinity of eNOS for the caveolae membrane13. Myristoylation results in hydrophic interactions between eNOS and the lipid membrane, however these interactions were found to be readily reversible14 and consequently would need stabilising, potentially by the two palmitate groups. This supposition is supported by data that indicate the loss of two palmitate groups causes a 2-fold decrease in membrane-bound eNOS, with an additional loss of the myristoyl group removing any remaining eNOS-membrane associations13.
Finally, S-palmitoylation contributes to the subcellular trafficking and membrane localisation of eNOS, which results in regulation of the enzyme via caveolin-1 (Cav1) and calmodulin (CaM). eNOS activity, that mediates the conversion of L-arginine to NO and L-citrulline10, was undetectable in plasma membrane fractions that did not contain caveolae. Furthermore, eNOS activity was 9-fold greater in caveolae in relation to the whole plasma membrane15. This is due to signalling proteins (Cav1, CaM), which contribute to the acute regulation of eNOS, also residing within caveolae. These include Cav1, which causes a direct, reversible inhibition of eNOS, due to the disruption of electron flow that prevents the production of NO. Inhibition is relieved in the presence of excess CaM, causing electron transport to resume and subsequently NO synthesis11. If eNOS does not undergo S-palmitoylation, it will not be trafficked and anchored to the cavealae membrane, and therefore cannot be regulated.
A simplified schematic showing the regulation of dually acylated eNOS by Cav1 and CaM.
There are over 500 protein kinases in the human genome and to date 20 have been reported to be S-palmitoylated16. This includes LIM kinase-1 (LIMK1), an actin-binding kinase that inactivates cofilin and subsequently promotes actin polymerisation. S-palmitoylation occurs near the N-terminus of LIMK1 at cysteines 7 and 8 and contributes to the regulation of LIMK1-dependent cytoskeletal dynamics in dendritic spines17.
S-palmitoylation is necessary for the targeting of LIMK1 to dendritic spines, shown by a significant reduction in localisation of a non-S-palmitoylated LIMK1 mutant17. S-palmitoylation also contributes to actin turnover in spines and activity-dependent spine enlargement. LIMK1 knock down in hippocampal neurons led to a reduction in actin turnover and activity-dependent spine enlargement. These phenotypes could be rescued by WT LIMK1, but not a non-S-palmitoylated mutant17.
In addition, S-palmitoylation is essential for LIMK1 activation, via its upstream kinase: p21-activated kinase-3 (PAK3), in vivo. When tested in vitro mutation of cysteines 7 and 8 had no significant effect on the activation or activity of LIMK1. However, when tested within neurons, the loss of palmitoyl-LIMK1 led to a 10-fold decrease in phosphorylation by PAK317. This may be because the non-S-palmitoylated LIMK1 mutant cannot co-localise with PAK3 on the dendritic spine membrane.
A simplified schematic to show LIMK1 localisation to the plasma membrane by S-palmitoylation. Here it is phosphorylated by PAK, and in turn phosphorylates cofilin. Activated cofilin then acts to reassemble actin filaments.
Abnormal spine morphology is associated with various cognitive disabilities, such as Autism Spectrum Disorder or schiozphrenia. These data indicate that S-palmitoylation of LIMK1 is involved in the regulation of neuronal actin dynamics and therefore impairments in S-palmitoylation may lead to various neurological disorders.
Dual leucine-zipper kinase (DLK) is another kinase that is S-palmitoylated, leading to the facilitation of the enzyme’s retrograde signalling. One role for DLK is to mediate signalling from the axon to the soma in response to peripheral nerve injury, which is essential for both axon degeneration and regeneration17. Activation of DLK in peripheral axons sequentially activates the c-Jun N-terminal Kinase (JNK)18. This results in the phosphorylation of the transcription factor c-Jun (p-c-Jun). However, DLK does not alter basal physiological JNK activity16 and therefore a mechanism by which DLK propagates these retrograde signals was largely unknown…until now!
Holland et al. (2016) concluded that S-palmitoylation exerts three distinct effects on DLK-dependent signalling: enzyme location, macromolecular complex formation and enzyme activity20.
Firstly, S-palmitoylation targets DLK to axonal trafficking vesicles, providing an explanation as to how the soluble enzyme propagates long-range signals. Using fluorescent tagging, DLK was shown to accumulate in the axonal puncta in cultured sensory neurons, whilst a non-S-palmitoylated mutant form of DLK was distributed throughout the axon20. Therefore, at a cellular level S-palmitoylation contributes to the localisation of the enzyme, by targeting palmitoyl-DLK to axonal trafficking vesicles, resulting in a retrograde signal.
Secondly, S-palmitoylation is essential for oligomerisation with MAP2K4, MAP2K7 and JNK-interacting Protein-3 (JIP3). Mutating the DLK S-palmitoylation site led to a disruption of DLK association with JIP3, MAP2K4 and MAP2K7, but did not alter DLK-DLK homodimerisation20. Preceding evidence suggests JIP3 is involved in dynein-based retrograde transport of JNK and lysosomes21. Together these data suggest S-palmitoylation facilitates the formation of multiprotein complexes that bind to microtubules, aiding retrograde transport.
A simplified model to show axonal retograde signalling by palmitoylated DLK-JNK pathway kinases.
Thirdly, S-palmitoylation may regulate DLK activity. Non-S-palmitoylated mutant DLK showed significantly reduced phosphorylation of JNK3 and downstream MAP2Ks20. These data indicate that S-palmitoylation contributes to DLK activity, possibly by preventing an auto-inhibitory interaction of the acidic N-terminal domain with the basic kinase domain20.
Looking towards the future, these findings point towards an exciting therapeutic target in palmitoyl-DLK. Molecules that inhibit S-palmitoylation could be used to reduce DLK-mediated neurodegeneration and conversely, molecules that inhibit depalmitoylation enzymes could be a target for DLK-mediated neuronal regeneration20. Additionally, further research addressing which other MAP3Ks contain S-palmitoylation sites and consequently the role of S-palmitoylation on these kinases may be of interest.
Author: Eleanor Eisenstadt
Landing page image: credit to suitesculturelles.wordpress.com