Toward a unified model for the role of sphingolipids in apoptosis
It’s complicated and context dependent! A truer statement has never been spoken, especially when it comes to the wild world of sphingolipid metabolism and the regulation of cell-stress responses by sphingolipids.
In the past, many laboratories focused on only one (or perhaps a small handful) of the thousands of known sphingolipids and declared that their sphingolipids of interest were the all-powerful modulators of whichever stress responses they were studying that day.
However, just when we think we have triangulated the identity or role of a lipid species involved in a particular cell-stress response, the lipid itself teaches us a valuable lesson about just how slippery it really is. Even small manipulations of a single sphingolipid entity can alter metabolites dramatically or flux through the entire metabolic pathway, making it difficult (if not impossible) to attribute the phenotype to the originally targeted sphingolipid.
Furthermore, the field is just now beginning to understand that sphingolipid metabolism is so important to cells that when the expression of one enzyme isoform is altered the system will compensate by altering the expression of other isoforms to maintain homeostasis. Likewise, sphingolipid enzymes have been shown to heterodimerize, which appears to be important for their activity. This was recently reinforced in publications from Lina Obeid’s laboratory at Stony Brook University and Tony Futerman’s laboratory at the Weizmann Institute of Science. It’s complicated!
When it comes to the regulation of apoptosis by sphingolipids, we have fallen victim to this type of narrow-minded tunnel vision, which has led us to overemphasize single roles for single lipid species. For example, the dogma in the field has been for years that a simple balance between pro-apoptotic sphingoplipids such as ceramide and anti-apoptotic sphingolipids such as sphingosine-1-phosphate could dictate cellular life-versus-death decisions. However, recent data from the laboratories of Jerry Chipuk at The Mount Sinai Hospital, Douglas Green at St. Jude Children’s Research Hospital and Holger Wesche at Amgen have called on the field to reevaluate the dogma that S1P’s sole function is purely pro-survival and that decreasing its levels is sufficient to induce cell death. It’s (perhaps) context dependent!
It long has been known that there is interplay between the BCL2-like proteins and sphingolipids in the regulation of apoptosis. The first examples of this were in the early 1990s, when it was shown that overexpression of BCL2 or BCLxL blocks ceramide-induced apoptosis (see Fang et al and Martin et al). Molecular mechanisms for interactions between the BCL2-like proteins and sphingolipids have been proposed by many groups. For example, Marco Colombini’s laboratory at the University of Maryland College Park showed in model systems that ceramide channel formation is inhibited by BCLxL via binding ceramide in its hydrophobic pocket. In addition, both the Colombini laboratory at the University of Maryland and Richard Kolesnick’s laboratory at Memorial Sloan-Kettering Cancer Center showed synergism between ceramide and BAX in permeabilization of mitochondria during apoptosis.
Recent data by our laboratories at the University of Louisville shed additional light on cross-talk between these two families. We show that pro-apoptotic BAK regulates ceramide generation during apoptosis via activation of a ceramide synthase. Importantly, activated BAK is a more potent activator of CerS. Further, we show that the anti-apoptotic BCL2 proteins directly interfere with BAK activation of CerS by binding and inhibiting BAK. This intricate system of cross-talk is complicated further by the fact that the six different anti-apoptotic BCL2-like proteins preferentially interact with BAK, suggesting that expression levels and availability of both the pro- and anti-apoptotic BCL2-like proteins need to be taken into account when considering how this system regulates ceramide metabolism. Data from the laboratories of Chipuk and Green also suggest that metabolites of ceramide are required for the full potential of BAX and BAK to induce apoptosis after certain stimuli.
The combination of these data, as well as myriad not discussed herein, makes it clear that we no longer can live in an isolated laboratory where we care about only our favorite protein or lipid species of interest. In addition, we must not be quick to pass judgment on whether newly published data or findings fit perfectly within the prevailing dogma, but rather we need to view the field as a whole and let the data guide us to new dogmas.
The only way we will be able to achieve this lofty goal is to change the way we think about our field, the people in our field and the goals of the field. The idea of a unified model for the role of sphingolipids in regulating apoptosis may be just a pipedream of the narrow-minded scientist who is interested in viewing the world through tunnel vision. Perhaps it is more likely that each apoptotic stimuli or each cell type will throw us a few curve balls that go against the model. This does not mean that the research is right or wrong, but it could be just as correct as the current model. In either case, we need to remember that it’s complicated and context dependent!
Leah J. Siskind (firstname.lastname@example.org) is an associate professor at the University of Louisville Medical Center’s pharmacology and toxicology department. Levi J. Beverly (email@example.com) is an assistant professor in the medicine department and the pharmacology and toxicology department. Both are members of the James Graham Brown Cancer Center.