Today, we’re including a special perspective written by Buck postdoc Chris Wiley on his recent publication in Cell Metabolism (“Mitochondrial Dysfunction Induces Senescence with a Distinct Secretory Phenotype.” Cell Metab. 2016 Feb 9;23(2):303-14.) which covers a unique and unexpected connection between mitochondria and the aging process. The full publication can be found here.

Frequently, science is as much about determining the right question to ask as it is answering questions. Forty-one years after the National Institute on Aging was created, the approaches to the question “What causes aging?”, have evolved dramatically. Research has shown us that “what is aging?” is just as important a question to answer. In fact, the more we’ve learned about aging, the more it appears that there are not only multiple causes of aging, but also multiple different aging processes. Today, I’m going to tell you about one potential cause of some parts of aging (mitochondrial dysfunction), and how our recent paper links it to a basic cellular process that is associated with aging (cellular senescence).


Addition of a simple sugar–pyruvate–can repair the aging phenotypes of senescent cells with damaged mitochondria. This suggests there’s a specific process that might be targeted with a drug in the future to prevent this kind of cellular decline.

Mitochondria are the primary source of energy (largely in the form of ATP) for most of our cells. They also more closely resemble bacteria than they do other parts of the cell. In fact, they have their own unique genome that, much like bacteria, is circular. This genome encodes a small number of proteins and RNAs that are essential for healthy mitochondrial function. As we age, the mitochondrial genome acquires mutations. Moreover, our mitochondrial DNA acquires mutations much more rapidly than our nuclear genome – due to a combination of weaker DNA repair and close proximity to reactive oxygen species (ROS) produced by respiration. These and other factors result in a state in which mitochondria become less and less functional as we age, a term generically called “mitochondrial dysfunction”.


In our recent paper, we show that cells with mitochondrial dysfunction undergo cellular senescence – a tumor-suppressive process that permanently halts cell division. More than simple arrest of proliferation, senescent cells undergo several changes – including change of shape, size, and (most notably) the biologically active molecules they secrete. This can result in loss of tissue regeneration, as well as disruptive effects on other non-senescent cells nearby. For more on senescence, and why it is important, please see our previous blog post: “Aging Fundamentals: Cellular Senescence”, here at SAGE.

In the case of mitochondrial dysfunction, cells adopt a senescent state, but they lack many of the secretory features of the other types of senescence that we and others have studied. Cells that undergo mitochondrial dysfunction-associated senescence (MiDAS) secrete many biologically active factors, but they don’t produce many of the typical inflammatory molecules produced by other forms of senescence. Instead, these cells secrete their own unique blend of biologically active factors that prevent adipogenesis and promote skin cell differentiation. In a model of mice that age prematurely due to mitochondrial mutations, MiDAS cells accumulate in fat deposits and skin, causing the mice to lose fat, lose hair, and develop very thin skin as they age.


Mechanistically, MiDAS occurred due to decreased cytosolic NAD+/NADH ratios. Mitochondria oxidize NADH to NAD+ as part of normal respiration, so when mitochondria are compromised NADH levels rise in the cell. As a consequence of lower NAD+/NADH ratios, AMP and ADP rise, leading to activation of AMP-activated protein kinase (AMPK), which then phosphorylates and activates p53 – a major mediator of senescence. Therefore, senescence is a natural outcome of metabolic stress following mitochondrial compromise.


NADH can be oxidized by alterative means – and addition of factors such as pyruvate to the culture media allowed mitochondria-independent enzymes to oxidize NADH, restoring the NAD+/NADH ratio. When cultured in the presence of these compounds, cells with mitochondrial dysfunction grew normally and did not senesce. Surprisingly, these non-senescent cells had a secretome that largely resembled the canonical SASP! Upon pyruvate withdrawal, these cells underwent senescence and lost their SASP-like secretome.


Many questions are still unanswered. Do MiDAS cells accumulate during normal aging? If so, where and when do they do so? Are NAD-targeted therapeutics still beneficial if they allow secretion of inflammatory factors? More importantly, can we target these cells to prevent or even cure some of the disorders associated with aging? Now that we know that MiDAS exists, we are positioned to answer these important questions.