Whodunit? – The Culprit of AMPK Phosphorylation in Insulin Resistance

Stephanie Spiegel

ChemBio Spotlight #4

Paper: PKD1 Inhibits AMPK2 through Phosphorylation of Serine 491 and Impairs Insulin Signaling in Skeletal Muscle Cells

http://www.jbc.org/content/291/11/5664.full.pdf+html

Whodunit? – The Culprit of AMPK Phosphorylation in Insulin Resistance

It is no secret that millions of people worldwide are affected by type II diabetes. People with this disease are unable to properly store glucose in their tissues due to insulin resistance. Insulin resistance means that the insulin receptors and/or their pathways are not functioning properly. Normally, when glucose is available in the bloodstream, insulin receptors are told to activate the downstream enzymes in their pathway to cause for metabolic changes in the tissues of the body. One of these changes can be to store the glucose present in the blood. However, in insulin resistance, one of these downstream enzymes, called AMPK, is inactivated by a phosphorylation at one of its serine residues. As a result, the pathway is disrupted and glucose cannot be stored. This leads to high blood glucose levels, a symptom seen in untreated diabetic patients. So who is causing this trouble by directly phosphorylating AMPK? A recent study by Coughlan et al. investigates two possible enzymes responsible: PKC and PKD1. Only one of the two enzymes, however, is proven guilty.

Figure 8a
Western Blot Analysis of AMPK Ser491 Phosphorylation in Three Different Conditions Reveals PKD1 Directly Phosphorylates AMPK

The authors first demonstrated that PMA, a PKC activator, caused an increase in AMPK phosphorylation when incubated with PKC or PKD1. This experiment served to establish that both enzymes have some sort of contribution to AMPK phosphorylation when stimulated to do so. In order to eliminate other enzymes as responsible for the direct phosphorylation of AMPK, the authors conducted an experiment with enzymes popularly mentioned in the literature to be involved in this insulin-receptor pathway. Inhibition of all three enzymes did not prevent PMA-induced phosphorylation of AMPK. If they were phosphorylating AMPK, their inhibition should have resulted in a decreased phosphorylation in this PMA environment. This made the authors sure PKC and PKD1 were the only two left in the running. The authors proceeded to test whether inhibitors of PKC and PKD1 would have any effect on AMPK phosphorylation. Inhibition of both enzymes prevented PMA-induced phosphorylation of AMPK’s serine. This further solidified the two enzymes’ role in the pathway disruption. Coughlan et al. went on to see the effects a knockdown of PKD1 has on the phosphorylation of AMPK versus IRS-1 (another intermediate in the insulin signaling pathway). They found that the PKD1 knockdown prevented PMA-induced phosphorylation of AMPK but not of IRS-1. IRS-1 phosphorylation is also known to impair insulin signaling. Thus the authors realized that PKD1 activation must diminish insulin signaling through an unprecedented mechanism independent of IRS-1. In order to exhibit that PKD1 is directly phosphorylating AMPK, the authors tested whether recombinant PKD1 phosphorylates AMPK in a cell-free system. A positive control of AMPK with Akt was compared to AMPK with PKD1 as well as AMPK alone. PKD1 did indeed phosphorylate AMPK in these conditions, suggesting PKD1 does so directly. Establishing PKD1 is directly responsible for this phosphorylation opens doors for drug-targeting approaches to combat insulin resistance.

References:

Coughlan, Kimberly A., Rudy J. Valentine, Bella S. Sudit, Katherine Allen, Yossi Dagon, Barbara B. Kahn, Neil B. Ruderman, and Asish K. Saha. 2016. “PKD1 Inhibits AMPKα2 through Phosphorylation of Serine 491 and Impairs Insulin Signaling in Skeletal Muscle Cells.” Journal of Biological Chemistry 291 (11): 5664–75. doi:10.1074/jbc.M115.696849.

Mancini, Arturo D., and Vincent Poitout. 2013. “The Fatty Acid Receptor FFA1/GPR40 a Decade Later: How Much Do We Know?” Trends in Endocrinology & Metabolism 24 (8): 398–407. doi:10.1016/j.tem.2013.03.003.

11 Replies to “Whodunit? – The Culprit of AMPK Phosphorylation in Insulin Resistance”

  1. Very interesting article. T2D is such a growing medical concern and this research has the potential to be applied to help combat the disease.

    In regards to directly inhibiting PKD1, the authors seem to be wary of this approach due to the widespread cellular phosphorylation activities it has. What other pathways is PKD1 involved in? Also, how is PKD1 activated? This would seems to be another potential target for therapeutics.

    1. You make a great point Elaine. Looking further back into PKD1’s pathway may also serve the same purpose. To answer your question, PKD1 is activated by either DAG or by PKC isoforms. Specifically, its activation loop is transphosphorylated at serine 744 or 748 and its C terminus is autophosphorylated at serine 916. So maybe DAG and PKC can be targeted as well. As for the other pathways PKD1 is involved in, it has been identified as a part of the signal transduction in gonadotropin-releasing hormone pathway as well as the suppression of colon cancer.

      References:

      Higa-Nakamine, Sayomi, Noriko Maeda, Seikichi Toku, and Hideyuki Yamamoto. 2015. “Involvement of Protein Kinase D1 in Signal Transduction from the Protein Kinase C Pathway to the Tyrosine Kinase Pathway in Response to Gonadotropin-Releasing Hormone.” Journal of Biological Chemistry, September, jbc.M115.681700. doi:10.1074/jbc.M115.681700.
      Sundram, Vasudha, Aditya Ganju, Joshua E. Hughes, Sheema Khan, Subhash C. Chauhan, and Meena Jaggi. 2014. “Protein Kinase D1 Attenuates Tumorigenesis in Colon Cancer by Modulating β-catenin/T Cell Factor Activity.” Oncotarget 5 (16): 6867–84. doi:10.18632/oncotarget.2277.

  2. Great article! You mentioned the fact that the authors were testing both PKC and PKD1 to see which was directly involved in the phosphorylation; however, the only knockdown test conducted was with PKD1. What caused the authors to believe that PKD1 was responsible rather than PKC? In addition, have there been other studies done to figure out the exact mechanism that PKD1 uses independent of IRS-1 for phosphorylation?

    1. I’m glad you liked it Tyler! You raise a really good point, which I actually wondered myself. It seems that authors use both a nonspecific and specific inhibition of both PKC and PKD1 to arrive to their conclusion. I believe that once the use of pharmacological inhibitors indicated PKD1 was involved in the PMA-induced AMPK serine phosphorylation, they followed that path more specifically to confirm this suspicion with the PKD1 knockdown experiment. In terms of other studies to figure out that exact mechanism, I could not find any that specifically investigate its kinase activity in insulin-signaling. However, this paper was just published March 11, 2016 so it may take some time.

  3. Really well written! It seems like DAG is also a culprit here (by activating PKD1). The authors also say that DAG is “activated” by a high glucose concentration but I’m not sure what they mean. I was wondering how DAG is synthesized? Does synthesis of DAG have any relationship or overlap with glucose metabolism?

    1. Thanks Elliott. You’re definitely correct in DAG’s responsibility in this whole process. I believe diacylglycerols (DAGs) are synthesized by the breakdown of triacylglycerols (TAGs) through lipases. I’m sure this relates to glucose metabolism because, well, isn’t all metabolism somehow related? Diabetes actually causes an error in the ability of insulin and glucose to stimulate TAG synthesis, which means glucose metabolism does indeed have a relationship with DAGs.

      References:

      Hopp, J. F., and W. K. Palmer. 1991. “Effect of Glucose and Insulin on Triacylglycerol Metabolism in Isolated Normal and Diabetic Skeletal Muscle.” Metabolism: Clinical and Experimental 40 (3): 223–25.

  4. Really interesting post! It was mentioned that PMA is a PKC activator, however inhibition of of PKC actually prevented PMA from phosphorylating. Does the relationship between the two go both ways? If not, why would inhibiting PKC decrease PMA’s ability to phosphorylate. Also, is this how the authors decided to only use PKD1 in the knockdown portion of the experiment? If not, why did they not try PKC as well?

    1. Nikki, so I think with all the “P” acronyms it might get a little confusing. PMA is only acting as a activator in this paper. It acts to stimulate PKC and PKD1. So you have the first part right. However, PKC inhibition does not affect PMA necessarily. PMA is not doing the phosphorylating, rather its presence has been found to decrease AMPK activity due to its ability to stimulate these kinases. So when the authors state that inhibitors prevent PMA-induced phosphorylation, it is not the PMA itself that is phosphorylating, but it causes PKC or PKD1 to do so. They are highlighting the fact that even tough PKC and PKD1 are in the presence of PMA (their stimulator) they can still be inhibited by Go6983 or CRT0066101.

  5. Very Interesting and well written. I can see how finding this link between AMPK activation and PKD1 is very important to the study of type 2 diabetes, as they indicate that PKD1 inhibition and knockdown completely prevent AMPK Ser 485/491 phosphorylation and subsequent inactivation. Does this suggest that PKC acts upstream of PKD1 in AMPK phosphorylation, since PKD1 knockdown prevents AMPK phosporylation?

    1. Thanks a lot Zach. You are exactly right. The authors do say that they have not completely eliminated novel PKC isoforms as upstream stimulants. But I’m not sure if it is for the reason that the PKD1 knockdown prevents AMPK phosphorylation. Although this is a true statement, the logic does not quite seem to connect.

  6. Fascinating article and great blog, Stephanie! As someone with several family members suffering from diabetes type II, this article really hits home for me. The methodical and systematic approach to effectively deduce that PKC and PKD1 were involved in AMPK phosphorylation and then conducting a knock out experiment with PKD1 was very interesting. The figure above had a column with the label “akt”. What was the relevance akt had with the experiment they were conducting?

Comments are closed.