Betatrophin—Promises Fading and Lessons Learned

Abstract

The discovery of a potent mitogen for insulin-producing pancreatic β cells, optimistically termed “betatrophin,” excited researchers and laypeople alike, promising a new therapeutic approach to diabetes. A recent “Matters Arising” report by Gusarova and colleagues (2014) places serious doubts on whether “betatrophin” plays any major role in β cell replication.

A 2013 report published in Cell identified the circulating protein angiopoietin-like 8 (ANGPTL8, also known as “lipasin,” “RIFL,” or “Gm6484”), produced primarily by liver and adipose tissue, as a novel hormone promoting β cell proliferation— renamed “betatrophin” by the authors— and raised high hopes in the scientific and lay community alike that an entirely new treatment option for diabetes might be on the horizon (Yi et al., 2013). In fact, given the impressive publicity received by the study in the lay press and social media, thousands of patients approached their physicians with questions about the new “wonder drug,” and philanthropists rescinded their support for diabetes research centers. Unfortunately, several recent articles now show that the celebration was premature, and that ANGPTL8, or “betatrophin,” plays little to no role as a β cell mitogen (Gusarova et al., 2014; Wang et al., 2013).

How was ANGPTL8 implicated as a β cell mitogen in the first place? Melton and colleagues used a potent antagonist of the insulin receptor developed at Novo Nordisk termed S961 (Schäffer et al., 2008) to acutely cause insulin resistance and increase plasma glucose levels to induce β cell replication (Yi et al., 2013). It had been known for decades that insulin resistance causes β cell mass expansion in rodents, as seen, for instance, in mice with genetic inhibition of peripheral insulin signaling in the liver (Michael et al., 2000; Withers et al., 1998). The existence of a circulating factor mediating mitogenic effects on β cells had been proposed by others based on increased β cell replication in islets grafted under the kidney capsule of insulin-resistant mice (Flier et al., 2001). Thus, the finding that S961 also causes cell-cycle entry of β cells by itself was not surprising. Nonetheless, the insulin receptor antagonist was established in this study as a powerful tool with which to search for any possible downstream β cell mitogen.

In order to identify such a circulating factor originating from the liver of insulin-resistant mice, Melton and coworkers performed expression profiling on the livers of S961-treated mice to identify potential secreted proteins that might mediate the mitogenic effects. They found steady-state mRNA levels of ANGPTL8 increased by approximately 4-fold. ANGPTL8, being a secreted protein, made an excellent candidate as mediator linking hepatic insulin resistance to β cell replication. Yi and colleagues then proceeded to overexpress ANGPTL8 in the liver of mice via tail-vein hydrodynamic plasmid injection and reported an astonishing 17-fold increase in β cell replication rate, leading within a few days to increased β cell mass (Yi et al., 2013). In light of these exciting results, the authors renamed ANGPTL8 “betatrophin,” befitting its proposed function.

Missing from this initial publication was the genetic proof that loss of ANGPTL8/betatrophin would in fact attenuate or prevent β cell mass expansion in any insulin-resistant setting, or, alternatively, hints on the molecular mechanism of action, in particular given that application of betatrophin to cultured islets did not trigger replication. This level of support for any conclusion is frequently required for high-impact publication of mouse mechanistic studies and has proven to be the Achilles’ heel of the initial optimistic report. Just months after Yi et al. published their findings, Wang and colleagues reported that mice null for ANGPTL8/betatrophin had entirely normal glucose metabolism, even when made insulin resistant, an effect that requires expansion of β cell mass (Wang et al., 2013). In a more recent issue of Cell, Gusarova and colleagues went even further, showing that neither genetic ablation of ANGPTL8 nor its overexpression affect β cell mass in mice (Gusarova et al., 2014). They also repeated ANGPTL8 overexpression studies, again using hydrodynamic delivery of plasmids via the tail vein, and showed that while this manipulation is sufficient to increase plasma triglyceride levels, a known effect of ANGPTL8 on lipid metabolism, there was no change in β cell mass. Perhaps most strikingly, when Gusarova and colleagues treated ANGPTL8 mutant mice with the insulin receptor antagonist S961, β cell mass was expanded to the same degree as in control mice, demonstrating that ANGPTL8 is not required to mediate the effect of insulin resistance on β cells in mice. In a Correspondence piece in the same issue of Cell, the authors of the original betatrophin paper confirmed that their own null mouse model was similarly unaffected in terms of β cell mass expansion in the setting of insulin resistance (Yi et al., 2014). They also report that the discrepancy with their 2013 paper appears to have been caused by high variability of the procedure employed to increase ANGPTL8 expression in the liver, which caused a small number of mice to respond strongly, but many not at all (Yi et al., 2014).

What are the lessons learned from betatrophin? An important question appears to be whether rodents are in fact the best model system to identify β cell mitogens that might be effective in increasing β cell mass in type 2 diabetics. Indeed, pancreatectomy leads to diabetes, but does not increase β cell regeneration in humans (Menge et al., 2008), and our group recently showed that whatever the stimulus for β cell replication in mice treated with S961, it had no effect on transplanted human β cells (Jiao et al., 2014). Perhaps human β cells are less responsive to the insulin-resistant state and need additional “prodding” to reenter the cell cycle, such as, for instance, through suppression of cell-cycle inhibitors (Avrahami et al., 2014). What is clear from the betatrophin saga is that the problem of increasing β cell replication and mass is complicated. Referees and editors must weigh the excitement of a new finding versus the level of proof required to make it public. While the scientific enterprise is benefitted by enthusiasm in the lay public, we must also weigh the negative consequences of not being able to keep promises of widely publicized discoveries.