• HEPATOCELLULAR CARCINOMA

Primary liver cancer is the third-leading cause of cancer-related mortality globally, accounting for over 700 000 deaths each year.1 In the last decade, the contribution of diet has been unravelled as a significant driver of hepatic oncogenesis.2 Specifically, diets high in fat have been associated with increased risk for the development of metabolic dysfunction-associated steatohepatitis (MASLD) and hepatocellular carcinoma (HCC) in many epidemiological studies,3 4 and these findings have been recapitulated in various in vitro and in vivo models. Although the link between diet and tumourigenesis has been convincingly established, the underlying mechanisms remain poorly understood.

In Gut, Bu et al hypothesised that exogenous metabolites from the diet might play a direct regulatory role in protein signalling and specifically in the activation of protein kinase B (AKT).5 The PI3K-AKT pathway is a well-characterised transducer of extracellular signals and promotes tumourigenesis by integrating environmental stimuli.6 Specifically in HCC, the role of PI3K-dependent AKT signalling has been an enduring focus of mechanistic and therapeutic investigations.7 In particular, it has been demonstrated that overexpression of an activated/myristoylated form AKT in the mouse liver via hydrodynamic gene delivery is sufficient to drive liver tumour development in a fatty acid-dependent manner.7 Post-translational modifications of AKT are a central regulatory component of the PI3K-AKT signalling pathway, with ubiquitination, methylation, and acetylation all uniquely modulating the function of AKT. More recently, two reports have shown that AKT can be palmitoylated in the context of lipogenesis and adipocyte differentiation8 and macrophage-induced colitis,9 but the mechanism of and role of AKT palmitoylation needs to be more thoroughly investigated.

Physiologically, the palmitoylation post-translational modification affects the activation, localisation and stability of numerous proteins. In addition, palmitoylation has been implicated in various pathological conditions, such as tumour progression, viral infections and neurological disorders.10 At the biochemical level, the process of palmitoylating a protein involves the conversion of palmitic acid to palmitoyl-Co-A and the subsequent conjugation of palmitoyl-Co-A to a cysteine residue on a protein by a palmitoyl acyltransferase, belonging to the zinc finger DHHC domain-containing (ZDHHC) protein family. Conversely, palmityl modifications are reversible and can be removed from proteins by palmitoyl thioesterases, such as APT1 and APT2. These palmityl groups are derived from palmitic acid found in the diet, primarily in animal fats.11 Still, palmitic acid can also be produced endogenously through citrate in the tricarboxylic acid cycle (TCA), further modified by fatty acid synthase (FASN).12

Using an unbiased screening approach, Bu et al identified palmitic acid as a metabolite that can activate AKT, aligning with previous reports.5 The authors proceeded to systematically and meticulously assess this mechanism in the context of MASLD and HCC. First, they confirmed their metabolomics screen and demonstrated that palmitic acid from a high-fat diet (HFD) can activate AKT and promote MASLD and HCC. Next, they found that AKT is palmitoylated on C77 and C224, and these modifications are responsible for the oncogenic function of AKT. Interestingly, they demonstrated that the palmitoylation of AKT at C77 promotes the membrane localisation of AKT, concurrent with other reports. In particular, the authors showed that this relocalisation of AKT occurs in a PI3K-independent manner. They also revealed that the palmitoylation of AKT at C224 inhibits the formation of inactive AKT aggregates, thus leaving more AKT in its singular, activated form. The effects of AKT palmitoylation at these two unique cysteine residues both induce the broad pro-tumourigenic functions of AKT. Transitioning to the regulation of AKT palmitoylation, Bu et al found that ZDHHC17 and ZDHHC24 proteins drive the palmitoylation of AKT and the progression of MASLD and HCC. Conversely, APT2 controls the depalmitoylation of AKT and suppresses its oncogenic activity. The detailed mechanistic investigation pursued in this article elucidates a novel oncogenic pathway with the potential for broad applications in numerous future studies.

Immediate next steps derived from this study are numerous. First, this group focuses explicitly on AKT1, although they did note that both AKT2 and AKT3 undergo similar palmitoyl modifications. Broadening their investigation to include these other isoforms may help to elucidate redundant or resistance-promoting mechanisms. Additionally, the in vivo modelling performed in this investigation is highly compelling, but an evaluation of the ability of HFD to promote liver disease in liver-specific Akt1 knockout mice, Akt-C77S knockin mice or Akt-C224S knockin mice would further enhance the link between HFD, AKT and HCC. Another lingering question involves the requirement of palmitoylation for full AKT activation. Thus, investigations should be conducted to assess the degree of AKT activation with and without palmitoylation. Finally, most prior studies investigating the oncogenic role of AKT showed PI3K-dependent signalling mechanisms and attempted to target the AKT-PI3K axis. The present identification of PI3K-independent functions of AKT introduces new signalling mechanisms to be investigated.

Using metabolomics screening approaches to identify protein-activating substrates in this study can also have widespread implications for other investigations. First, palmitic acid was one of seven different metabolites identified by this screen, and the functions of the other six on AKT specifically have yet to be explored. Also, this method, in theory, can be applied to proteins other than AKT to identify countless other metabolite-induced interactions. The appreciation of the direct role of diet in disease progression is expanding rapidly, and this system provides an unbiased way to assess more than 500 metabolites on direct protein activation. More broadly, the finding that the direct incorporation of exogenous factors derived from nutrients may contribute to MASLD and HCC progression has the potential for infinite applications.

Given the role of palmitoylation across numerous disease states, potential therapeutic exploitation at several points along this mechanism may expand far beyond just HCC. Bu et al cleverly repurposed the already-approved FASN inhibitor, orlistat, and demonstrated that the upstream inhibition of this pathway by suppressing the production of palmitic acid could inhibit HCC tumour growth, further increasing the potential benefits of this study.5 However, given the significant role that the TCA plays across the body, the off-target effects, subsequent side effects, and the potential for resistance development due to the incomplete mechanism inhibition limit this approach’s efficacy. Instead, a small molecule inhibitor directed toward the ZDHHC class of proteins could be the next step. In this regard, several DHHC inhibitors with promising effects are currently in preclinical investigation.13 Furthermore, the adoption of a diet low in palmitic acid could function as a potential nutraceutical option for those with an increased risk for MASLD or HCC. Finally, Bu et al used peptides to mimic the AKT cysteine residue-containing regions to compete with endogenous AKT, effectively reducing AKT palmitoylation. The use of this therapeutic strategy seems promising and warrants further study.

In summation, HFD-induced palmitoylation of AKT by ZDHHC17/24 proteins is responsible for the activation of AKT in a PI3K-independent manner and drives HCC development (figure 1). The characterisation of this novel PA-ZDHHC-AKT axis represents a breakthrough in the field of HCC with exciting future implications for HCC patients and beyond.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Description of PA-ZDHHC-AKT axis in MASLD and HCC. PA ingested from HFD or produced endogenously by TCA and FASN has been shown to activate AKT. The conjugation of PA to AKT by ZDHH17/24 has been shown to promote the activation and oncogenicity of AKT and lead to MASLD and HCC. Conversely, the depalmitolyation of AKT by APT2 has been shown to reduce the oncogenic function of AKT. Several points on this pathway can be exploited for therapeutic intervention, including but not limited to dietary restriction of PA, inhibition of PA production and inhibition of PA conjugation. See Bu et al for details.5 Created with BioRender.com. APT2, acyl-protein thioesterase 2; FASN, fatty acid synthase; HCC, hepatocellular carcinoma; HFD, high-fat diet; MASLD, metabolic dysfunction-associated steatohepatitis; PA, palmitic acid; TCA, tricarboxcylic acid cycle; ZDHHC17/24, zinc finger DHHC domain-containing protein 17 and 24.