[Point of clarification: I was delighted to use this post to congratulate my friend and blogging colleague, Dr Chris Patil, on his contributions to this paper from the laboratory of Dr Judith Campisi discussed below. As the formal press release notes, "[c]o-authoring the paper with Campisi were Jean-Philippe Coppé and Christopher Patil, members of Campisi’s research group in Berkeley Lab’s Life Sciences Division, Joshua Goldstein, now with the Novartis Research Foundation; Francis Rodier and Denise Muñoz of the Buck Institute; and Peter Nelson and Yu Sun from the Fred Hutchinson Cancer Research Center in Seattle.” I asked Chris for an interview since he is an e-colleague, was kind enough to pick one of my posts for his new blog carnival (Hourglass), and is a top-tier scientist with a substantial presence in the online science communications and open access communities. It’s great fun to learn the backstory on what I predict will become a Citation Classic/The Scientist Hot Paper. Thanks again, Chris, for being so generous with your time and sharing your thoughts with our readers.]
I’ve kept a passing interest in senescence (cellular, not personal) over the last 20 years or so because I’ve always felt that attempts to increase longevity in a multicellular organism would also increase the risk of cancer (more seasoned readers may recognize this as the Hayflick phenomenon of replicative cellular senescence first identified in 1965). As a newly-minted PhD looking for postdoc fellowship, including in a yeast lab focused on aging, I thought that harnessing the senescence pathway might be a valuable strategy for treating cancer.
Yesterday’s PLoS Biology paper from Coppé et al., builds on a newly-described cellular response that throws a wrench in that simplistic idea. Cells induced to undergo senescence with DNA-damaging agents exhibit a secretory phenotype, termed the senescence-associated secretory phenotype (SASP), whereby molecules involved in inflammation and metastasis are released into the local environment. In younger individuals, this mechanism could prevent the development of cancer but in older individuals could increase the risk of cancer. In cancer, for example, therapies that causes cellular growth arrest and senescence may be of limited utility unless those senescent cells are removed. My takehome message from this paper is that we may have to rethink the benefit of cancer therapies that are cytostatic.
Using an antibody microarray technique optimized to provide a wide, quantitative dynamic range, the authors identified in human epithelial and stromal cells approximately 120 factors secreted by cells arrested with X-irradiation. There was significant overlap in the proteins across cell types and could also be observed in primary human prostate carcinoma cells surgically excised from patients before and after mitoxantrone chemotherapy (mitoxantrone is a natural product-derived antitumor drug that poisons type II DNA topoisomerases causing protein-linked DNA strand breaks.). The latter observation was crucial in demonstrating that the SASP profile was not an artifact of in vitro cell culture but could be observed cells exposed in vivo to a DNA-damaging, senescence-inducing drug.
Perhaps the most important finding is that cells with loss of p53 or gain of oncogenic RAS exhibited a far greater abundance of secreted proteins than with normal cells, leading the authors to conclude:
Our findings define a central feature of genotoxic stress-induced senescence. Moreover, they suggest a cell-nonautonomous mechanism by which p53 can restrain, and oncogenic RAS can promote, the development of age-related cancer by altering the tissue microenvironment.
I refer those to the paper for the identity of the secreted proteins which are likely to include one or more of your favorites:
The SASPs were complex, and their biological effects could not be predicted a priori. SASP components included inflammatory and immune-modulatory cytokines and chemokines (e.g., IL-6, IL−7, and IL−8, MCP-2, and MIP-3a). They also included growth factors (e.g., GRO, HGF, and IGFBPs), shed cell surface molecules (e.g., ICAMs, uPAR, and TNF receptors), and survival factors (Figure 1A and Table S2). However, the SASP was not a fixed phenotype. Rather, it was a wide-ranging profile because each cell strain also displayed unique quantitative or qualitative features.
The laboratories conducting the work were led by well-known biology of aging researcher, Dr Judith Campisi at Lawrence Berkeley National Laboratory and the Buck Institute for Age Research. One of the authors from the Campisi group was Dr Chris Patil, the science blogger who writes at Ouroboros and who founded the Hourglass blog carnival on the field of biogerontology.
Chris was kind enough to select one of my posts for the inaugural edition of Hourglass so I figured I’d dial him up to congratulate him on his work and get his impression of the significance of the work and gain some insight as to why the team selected the open-access journal, PLoS Biology, for such a comprehensive and ground-breaking report. The formal press release with Dr Campisi’s quotes is quite nice but it’s great to talk with a blogging scientist – I asked two short questions and had more than enough for two blog posts (thanks, Chris).
Abel: I’d love to note what you personally find most significant about your paper…
Chris Patil: For me, these results have really driven home the intimate connection between cancer and aging: Senescent cells accumulate in tissues with age. Their secretory behavior, which we’ve characterized in this study, provides a unifying principle that could potentially explain both age-related decline in tissue function and age-related increase in cancer rates. The SASP (the factors secreted by senescent fibroblasts) seem to be able to accomplish both deleterious outcomes simultaneously.
Beyond that, this story is an excellent illustration of the at-times-exasperating logic of the evolution of aging; this is especially true when you consider it alongside other work published earlier in the year. A few years ago, most of us were convinced that senescence was a cell-autonomous tumor-suppressor mechanism whose tissue-level effects (if any) were related to diminution of regenerative capacity. Now we know that, via the SASP, senescent cells are exerting deleterious effects on nearby tissue — so we have to invoke antagonistic pleiotropy, i.e., the idea that this could have been selected because it was good for you early in life (because it does stop damaged cells from becoming cancerous) despite being bad for you late in life (because the accumulation of senescent cells will ultimately damage the tissue and encourage metastatic tumor progression).
The story gets even more complex when you take into account some excellent recent papers from other groups. Work from Scott Lowe’s lab implies that the secretory profile may be a mechanism (albeit imperfect) to encourage immune clearance of senescent cells, whereas work from the Gil and Peeper labs strongly suggest that one of the functions of the SASP is to reinforce the senescent growth arrest. So now the SASP is at least partially good for you, in addition to being demonstrably bad for you.
On a personal level, I really enjoyed getting the quantitative protein array method up and running. That was a very close collaboration between JP (the first author) and myself, and we had a lot of fun doing it. By now, the same service (or a close equivalent) is commercially available (and non-radioactive, to boot), so we’re not going to follow up on it very much, but for a while it was nice to know we were basically the only ones in the world with a really robust, quantitative antibody array technology. The challenges of developing that method taught me a lot about tool building, and about what it means to really convince yourself of a new result obtained with a new tool.
Abel: Why did you choose to publish this work in an OA journal like PLoS Biology?
Chris Patil: Speaking for myself: I published my first paper PLoS Biology paper several years ago, when it was first starting out and it wasn’t clear whether the model was going to work. I chose PLoS Biology for several reasons: For one, much of the early ideological activism in support of OA had been going on at Bay Area universities, especially UCSF and Stanford. As a consequence, many of my role models (including my doctoral advisor, Peter Walter) were outspoken supporters of the OA movement — so I had a lot of encouragement and support, and I was exposed relatively early to the excellent arguments in favor of the OA model. Furthermore, I had substantial objections to the business practices of certain large for-profit corporate entities in the science publishing field, and I knew I didn’t want to be part of that system (anymore). In short, everything lined up pointing the same direction, so when I was deciding where to send my final paper from my graduate lab, it wasn’t a very hard decision. I think access to scientific information should be free, especially when it’s being paid for by the taxpayers but even when it isn’t.
Regarding this current study: I wasn’t the first author of this paper, so the decision about where to send the manuscript wasn’t entirely mine. I initially advocated for PLoS Biology but I was overruled; in the end, we sent this paper to another journal before PLoS — I’m not going to say where, except to say that it is a privately owned journal, and that your first guess is almost definitely right [I guessed - APB] — and while that wasn’t my choice, it also wasn’t really my call, so I didn’t lose a lot of sleep over it. At least, not at first.
As it turns out, we had an absolutely horrible experience at that other journal, bordering on editorial malpractice. I can’t say a ton about it, for what are probably obvious reasons — if I were the only author or the lead author, I would feel differently, but there are a lot of other authors and I don’t want to burn bridges for them.
After that experience, we pulled the paper and had a meeting about where to send it, and at that point the other authors were much more agreeable to the idea of sending the paper to PLoS. With my 2004 paper, I’d had a great experience with the review process there, and it was almost as smooth this time around. The quality and competence of the editorial staff made a huge difference with this second round of submission. During the prior submission to the closed-access journal, the editor of the journal were either unwilling or unable to override reviewers, even when they were being unreasonable or asking the impossible. In contrast, PLoS editors are very strong and empowered, and they’re sophisticated consumers of the input they receive from reviewers. Most of the comments we received from reviewers were constructive, but sometimes they weren’t, and the PLoS editors were self-confident and knowledgeable enough to recognize the difference.
So in addition to the ideological reasons for supporting OA, the practical end of it was also very nice, and we couldn’t be more pleased with the way things went (OK, I lied: it could have been faster, but it was by no means any slower than the competition).
Thanks again, Chris – it’s nice to have an inside view of the work and the publication process, especially for early-career researchers who may be less familiar with the peer-review process and journal editing.
This is definitely an area where I need to think more deeply about my own research so you can rest assured that I’ll be sitting with this paper and its multitude of supplementary data for the rest of the week. You can too because it’s Open Access.
Jean-Philippe Coppé, Christopher K. Patil, Francis Rodier, Yu Sun, Denise P. Muñoz, Joshua Goldstein, Peter S. Nelson, Pierre-Yves Desprez, Judith Campisi (2008). Senescence-Associated Secretory Phenotypes Reveal Cell-Nonautonomous Functions of Oncogenic RAS and the p53 Tumor Suppressor PLoS Biology, 6 (12) DOI: 10.1371/journal.pbio.0060301