Summary of:
Ito, T. K., Yokoyama, M., Yoshida, Y., Nojima, A., Kassai, H., Oishi, K., & … Minamino, T. (2014). A crucial role for CDC42 in senescence-associated inflammation and atherosclerosis. Plos One, 9(7). doi:10.1371/journal.pone.0102186
Trevor Griesman and Alissa Meister
Atherosclerosis is a chronic inflammatory disease affecting medium-large arteries beginning at birth, with the progression depending on many factors. The main risk factors for atherosclerosis include hypertension, hyperlipidemia, diabetes mellitus and others such as age, sex, smoking, and sedentary lifestyle. Possible complications from atherosclerosis include coronary artery disease, cerebrovascular diseases, and peripheral artery disease. Although these complications are responsible for over half of the yearly world mortality, they often occur late in the progression of the disease with limited early diagnoses.
The first molecular event in atherosclerosis is endothelial dysfunction in the arteries as a result of injury or inflammation. This dysfunction takes the form of either cell senescence, apoptosis, or activation. This paper focuses on senescence, when cells stop replicating. Atheroscletotic plaques result from an accumulation of lipids and smooth muscle cell proliferation. In response, endothelial cells over-express adhesion molecules and increase the recruitment of inflammatory cells. These inflammatory cells further release cytokines, causing a cytokine-mediated progression of atherosclerosis and LDL oxidation. Plaques deteriorate the cell wall and cause thickening of the surrounding muscles. Accumulation of these plaques can limit the flow of oxygen and nutrients to the rest of the body, leading to serious consequences depending on where these plaques form. Plaques forming in the coronary arteries cause Coronary Artery Disease (CAD), limiting the blood flow to the heart and increasing the risk for blood clots. Similar diseases can develop if these plaques form in the periphery, carotid artery, etc. Symptoms of atherosclerosis also depend on where these plaques form, and may include chest pain, weakness, numbness in the periphery, headache, kidney disease, and many more.
CDC42 is a GTPase in the Rho family responsible for cell cycle regulation functions such as migration, endocytosis, morphology, and cell-cycle progression. Additionally, CDC42 regulates the organization and proliferation of the actin cytoskeleton. Previous studies have shown that CDC42 may act in the senescence and inflammation of cells. The CDC42 pathway has been implicated as a potential therapeutic target for inflammation reduction in aging individuals, however this has yet to be studied in an in vivo model. Ito et al. (2014) set out to study the link between CDC42 and inflammation in senescent cells, and connect this to a potential role of CDC42 in atherosclerosis. The researchers specifically investigated the relationship between CDC42 and the inflammatory NF-κB pathway.
EXPERIMENTAL METHODS
In order to create a model for senescence, the researchers cloned the genes for p16 and p21 (cyclin-dependent kinase inhibitors) into retroviral vectors, and infected cells. This resulted in the integration of the gene DNA into the DNA of the cells. In order to determine the effects of different CDC42 pathway components, they knocked down genes using siRNA. They then used RT-qPCR (as we did in class) to measure the mRNA levels of three genes: the cytokine CCL2, the endothelial-leukocyte molecule E-selectin (SELE), and Vascular Cell Adhesion Molecule 1 (VCAM1). Western blotting was used to quantify translated levels of these proteins. For some experiments, other proteins, such as a deactivated form of CDC42 were upregulated in cells by infection with an adenoviral vector which contained the DNA that coded for those proteins. To measure innate immune response levels, induced-senescent cells transfected with various siRNAs were treated with LPS or TNF-α, and levels of CCL2, SELE, and VCAM1 mRNA were measured by RT-qPCR.
The researchers also studied mouse and roundworm models. Mice were made into conditional knockouts using a Cre-Lox system, in which special sequences called lox sequences are inserted into the genome flanking the target gene. The protein Cre cuts at the lox regions. A Cre protein was used that is inactive until treated with a drug, in this case tamoxifen. When the researchers introduced tamoxifen into their Cre-Lox mice, they effectively knocked out the gene between the lox sequences. The researchers targeted either CDC42, Mdm2 (a negative regulator of p53), or some combination of those genes. They also used Apolipoprotein E knockout mice. In these mice, they investigated protein expression in tissue sections using immunostaining. They recorded mRNA levels using RT-qPCR, and measured atherosclerotic lesion areas using tissue section staining. Finally, the researchers studied C. elegans worms, specifically the nol-6 strain, which has high innate immune expression due to a p53 dependant pathway. The researchers measured levels of immunity-gene mRNA using RT-qPCR in siRNA treated worms, and tracked the survival of the worms over time.
KEY RESULTS
The researchers found that in senescence-induced cells, the NF-κB pathway upregulates pro-inflammatory gene expression (Figure 1). This was shown through the creation of senescent like cells by introducing pathway kinase inhibitors, which led to an increase in inflammatory cytokine and adhesion molecule expression. Furthermore siRNA directly targeting this pathway decreased the inflammatory cytokines. The researchers additionally found that CDC42 also regulates pro-inflammatory gene expression, by testing knockdowns of the CDC42 pathway and observing a decrease in inflammatory molecules (Figure 2). After these two findings surrounding NF-κB and CDC42, the researchers further found that CDC42 up-regulates pro-inflammatory cytokines by activating NF-κB. However, when cells were activated by either LPS or TNF-α, CDC42 knockdowns had had cytokine levels higher than NF-κB knockdowns, implying that NF-κB is downstream of CDC42, as it functions with other methods of activation (Figure 3).
The researchers then decided to test these finding using in vivo models, including several strains of mice and C. elegans. An increase in pro-senescence signalling in atherosclerotic plaques was found, linking atherosclerosis and inflammation (Figure 4). Additionally, CDC42 deletion was found to reduce aortic infiltration by macrophages in atherosclerosis. Overall, using these in vivo models, the researchers found that CDC42 is a mediator of chronic inflammation, which leads to endothelial senescence.
BROADER CONTEXT
In senescent cells, replication is halted as cells age to prevent the onset of cancer or other genetic malfunctions. Senescent cells secrete inflammatory cytokines, possibly as a way of signalling to immune cells that will destroy the senescent cell before it becomes cancerous. As we discussed in class, one of the ways that endothelial cells can respond to stress is to become senescent. This can lead to the formation of plaques, as lipids and then immune cells infiltrate the tunica intima. The researchers investigated CDC42 because of its involvement in inflammation following senescence. In the context of plaque formation, if the senescent endothelial cells did not secrete inflammatory cytokines, less immune cells would invade the plaque, slowing the progression of the plaque and possibly preventing the formation of a fibrous cap. However, clinical treatments knocking out CDC42 are very far off, as the inhibition of all senescent inflammatory signalling could result in the buildup of senescent cells, and there is no way to target CDC42 specific siRNA to plaque sites. Additionally, the experiments were done in cell models of senescence, but other factors may contribute to senescence in vivo.
References
Mitrovska, S. (2009). Atherosclerosis : Understanding Pathogenesis and Challenge forTreatment. New York: Nova Biomedical Books.
What Is Atherosclerosis? (2014, August 4). Retrieved March 6, 2015, from http://www.nhlbi.nih.gov/health/health-topics/topics/atherosclerosis
In the discussion, the authors point out that it is interesting that deletions of CDC42 signaling had a markedly weaker influence on acute inflammation than on chronic inflammation. In fact, all of their evidence demonstrates the importance of the CDC42 pathway in regulating CHRONIC inflammation induced by cellular aging signals. My question: Why isn’t acute inflammation equally influenced by inhibition of CDC42 signaling?
You touched on this in the blog post, but I wanted to discuss further the clinical application of this study. It seems to me that a cell cycle-regulating protein as CDC42 would have many roles in diverse types of cells. It would be difficult, as you mentioned, to specifically target this one role of the molecule. Clearly they have expanded the body of knowledge about the mechanism of disease, but is there much hope for a potential novel therapy to arise out of this research?
Towards the end of the discussion section, the authors mention that CDC42 does not necessarily mediate NF-κB and that their in vivo experiments had slightly different results than the in vitro experiments. They also say that CDC42’s effects could be cell type and context dependent, suggesting that there may be other players involved in this inflammatory response. What other factors could possibly influence this pathway to have similar outcomes?
Midway through this paper and in reference to Figure 2E the authors state that the data suggestsCDC42-induced up-regulation of pro-inflammatory genes was unrelated to cell cycle regulation. Then, towards the end of the paper, they state that their results suggest that CDC42 signaling contributes more specifically to inflammation associated with senescence, defined by the authors as ‘irreversible growth arrest occurring after accumulation of DDR’. To me it is confusing to state that CDC42 signaling is related to senescence but not cell cycle regulation.
Additionally, early on in the paper the authors state that the pro-inflammatory signals emitted by senescent cells may help to prevent cancer development through the elimination of cells with oncogenes and that this inflammation could also promote tumor progression. I am curious how these processes differ and what allows them to create such different outcomes.
One of the interesting things that I realized while reading through this paper and your post is that many of the pathways, proteins, and drugs used in these experiments are similar to those identified in cancerous processes and treatments. As you discussed, the researchers observed increased inflammation and cytokine levels in their experiments, leading to further understanding of the pathway connecting CDC42 and NF-κB. Based on these findings, I wonder if there is further connection between atherosclerosis treatment and cancer treatments, and how someone with both of these illnesses would be affected through manipulation of CDC42.
You addressed a few of the difficulties with the clinical aspects of CDC42 knockout such as increased senescent cells. But, taking into consideration the large number of genes contributing to inflammation, how can the authors guarantee that through knockout of this particular protein, a desirable therapeutic outcome will result? Also, it does not seem possible to target one specific protein without disrupting other gene expression or cell processes.
One of the interesting i thought of after reading this paper was how a treatment using these findings could be delivered. As with many molecular targets, one concern is off target effects. With the advent of specific gene targeting through systems like TALENs and CRISPR, I wonder if there would be some way to specifically target CDC42 in the plaques formed in the tunica intima. In addition ( as mentioned by other classmates), it would be interesting to see if the complete understanding of the pathway that leads to inflammatory process in atherosclerosis could provide clues to the understanding of other diseases like cancer.
Great comments everyone! As always, we will have too much to discuss on Friday. First, I want to say that I think this paper was very well-written and easy to follow-a good model for writing papers! My second comment is about the data. I think the whole paper was connected very well to senescence, then they jumped to the ApoE model for atherosclerosis in mice and tried to make a senescence connection, which I thought was weak. It was not clear to me that the atherosclerosis induced in ApoE KO mice fed a high fat diet had a senescence component. However, it was clear that cdc42 was playing a role in plaque formation based on Figure 4E (quite a drastic reduction in atherosclerotic plaque area in the ApoE/cdc42 KO mice). To what extent does senescence contribute to atherosclerosis in the ApoE model? The plaque reduction was so great in the ApoE/cdc42 KO mice that I wonder if cdc42 is contributing to atherosclerotic signaling pathways other than senescence-induced pathways.
It was clear how they quantified the data from immunohistochemistry. I really like the part on the paper where it states the reason why they used worms and Apoe KO mice. I am also interested what other elements can be responsible for causing senescence in vivo other than knocking out cdc42 as well. In the meantime, I was wondering how knocking out sym-1 expression can block the inflammation from figure 5? sym-1 is important for p53-dependent activation of innate immunity, but I wasn’t sure blocking sym-1 expression will cause deactivation of all senescence-associated stimuli.
I feel that figure 3 is very important to the study. Though proving that the NF-κB pathway upregulates pro-inflammatory gene expression in senescence-induced cells was key to this paper’s findings, suggesting that NF-κB may be activated downstream of CDC42 may be vital treatment of ALS in the future. Furthermore, by using in vivo and in vitro models, this paper significantly links CDC42 to endothelial senescence.
As we discussed, this article did a nice job explaining the link between CDC42 and inflammation in senescence cell through human, mouse, and worm studies. I would be interested to see if other RhoGTPases, such as RhoA or Rac1 contributed to this inflammatory response as well!
Overall, I understand the purpose of CDC42 with regards to regulation of the inflammatory system. Is CDC42 being one of the predominate factors active in the relevant stress response a good enough reason for the authors to insinuate any future role in treatment of atherosclerosis? This conclusion seems a little off-base as a more efficient treatment would not be preventing growth of the plaque but rather preventing this unnecessary breakdown in the endothelial wall that allows for entrance of foreign bodies in the first place.
–> The main thing that I want to gain a better understanding of is how much wiggle room is there in GOOD science when making conclusions and connections about the importance of any research findings?