The Therapeutic Impact of Biased Signaling:
In general, prior art suggests that identifying agonists that stabilize unique receptor states is not a difficult task. However, the linking of these exemplar molecules to attainable biased signaling states of benefit in therapy is a more formidable challenge. As a preface to considering this, it is useful to review the current impact of bias on drug therapy.
Retrospective Analyses of Known Drugs: Over the past years, literature reports have linked favorable therapeutic profiles of drugs retrospectively with identified bias. An example of this is the drug carvedilol, an inverse agonist for Gαs-mediated cAMP production in heart failure where beneficial effects have been attributed to β-arrestin–mediated partial agonist activity for activation of ERK1/2 (Wisler et al, 2007; Kim et al, 2008). This shared profile also has been cited for the useful responses to propranolol (Azzi et al, 2003; Baker et al, 2003), nebivolol (Erickson et al, 2013), and alprenolol (Baker et al, 2003). Further, the reduced respiratory depression potential of the opioid analgesic levorphanol (over morphine) has been attributed to its biased signaling profile (lack of β-arrestin2 recruitment) (Le Rouzic et al, 2019). Interestingly, the dependence liability of oxycodone, hydrocodone/paracetamol, and hydromorphone has been attributed to biased signaling to G protein over β-arrestin (Johnson et al, 2017). Other reportedly unique biased profiles for established drugs have been reported for the adenosine 2b receptor agonist capadenoson (cardioprotective and cardiac fibrosis–modulating properties through biased cAMP activity; Baltos et al, 2017) and the β2-adreoceptor agonist fenoterol (biased activation of Gαs protein over nonspecific dual Gαs and Gαi activation (Jozwiak et al, 2010). While it is not clear whether the noted biased signaling for these molecules is the discerning property in their beneficial therapeutic effects, it is notable that beneficial profiles are associated with biased profiles in these drugs. These drugs were developed before bias was considered a viable drug property; since then, molecules consciously have been developed with known biased signaling properties in attempts to improve therapeutic profiles.
Current Biased Molecules: As mentioned previously, the first biased molecule designed to enter the therapeutic arena was TRV027 for congestive heart failure. This molecule has a unique lack of efficacy for angiotensin-mediated vasoconstriction (via Gαq protein) and a residual and measurable efficacy for angiotensin receptor-mediated β-arrestin activation. Notably, an acute randomized double-blind placebo-controlled Phase IIB dose-ranging trial (BLAST-AHF) did not show superior efficacy for TRV027 illustrating the difficulties in bias translation from the in vitro to in vivo arena; in this case, the trial design could have been the reason for failure. Specifically, a molecule such as TRV027 would be expected to enhance beneficial cardiac remodeling over many months. In contrast, the trial design was a short term 24-28 hr infusion of TRV 027 followed by a 30 day endpoint of symptoms and survival which really would not detect the possible beneficial effects of this molecule. In addition , a possible lack of activation of the renin-angiotensin system in the heart failure patients could have caused a negative background for this molecule coupled with shortcomings in study design and duration of treatment (Pang et al, 2017; Sugihara et al, 2017).
Data from early experiments with morphine and other opioids had suggested that bias away from β-arrestin could be beneficial (Raehal and Bohn, 2011; Raehal et al, 2005; Bohn et al, 2003; Urs and Caron, 2014) and this led to continued research into biased opioid agonists for pain management (Le Rouzic et al, 2019; Conibear et al, 2019). This is a very active area of research and presently there is a viable biased agonist opioid agonist (TRV130) approved for moderate to severe pain (Lambert et al, 2020; Viscusi et al, 2019).
A recent and very active research into biased signaling is in the field of incretin agonists for metabolic diseases such as obesity and diabetes. Several peptides are under development for GLP-1R and either the glucagon receptor (GCGR) and/or glucose-dependent insulinotropic peptide receptor (GIPR) as dual- or tri peptide agonists. It has been shown that many novel peptides have distinct biased agonism profiles relative to either natural agonists or each other (Darbalaei et al, 2020). Examples of this are biased signaling for GIP mono-agonists (Pro3GIP, Lys3GIP) towards ERK1/2 phosphorylation (pERK1/2) relative to cAMP accumulation at GIPR, the dual GIPR/GLP-1R agonist, LY3298176, biased towards pERK1/2 relative to cAMP accumulation at both GIPR and GLP-1R (vs endogenous ligands), and the triple agonist GLP-1R/GCGR/GIPR tirzepatide biased towards pERK1/2 relative to β-arrestin2 recruitment. In addition, studies have shown reduced β-arrestin recruitment for tirzepatide leading to less GLP-1 receptor desensitisation and downregulation (Yuliantie et al, 2020). While there is an abundance of data on the biased signaling for these agonists, it is not yet clear to what extent the bias contributes to their beneficial action vs dual and tripartide agonism (Jones, 2021).