The multiple-hit pathogenesis of non-alcoholic fatty liver disease (NAFLD)

Nonalcoholic fatty liver disease, or NAFLD, is one of the most common chronic liver conditions on the planet. Buzzetti, Pinzani, and Tsochatzis estimate that approximately 20 to 30 percent of adults are affected, with higher rates in industrialized countries. Most people who have it aren't aware, as it tends to be quiet, usually asymptomatic, and closely linked to obesity, high blood pressure, abnormal lipids, and insulin resistance. However, the disease covers a wide spectrum—from simple fat accumulation in liver cells, known as steatosis, all the way through nonalcoholic steatohepatitis, or NASH, to fibrosis, cirrhosis, and eventually liver cancer. Most individuals remain at the mild end of the spectrum. Only about 5 to 10 percent of patients with NAFLD develop NASH, and around 30 percent of those with NASH progress to cirrhosis. This uneven distribution is significant and raises an important question: why do some individuals progress while others do not? For decades, the prevailing answer was the two-hit hypothesis. The first hit was fat accumulation in the liver, driven by factors such as diet, sedentary behavior, obesity, and insulin resistance. This condition primed the liver. Then a second, later insult triggered inflammation and scarring. This idea was intuitive, and animal models supported it—leptin-deficient ob/ob mice accumulated fat without developing inflammation until an additional stress was applied. The theory appeared clean, linear, and satisfying. But it does not hold true in humans. Multiple parallel processes occur simultaneously in genetically predisposed individuals, and these processes interact in ways that a two-step model cannot capture. NASH can present initially, rather than as a downstream stage. Simple steatosis and NASH may actually represent distinct disease entities, not just different phases of the same process. The field has shifted to a multiple-hit framework, and Buzzetti and colleagues outline what those hits actually are. Starting with fat, triglycerides accumulate in liver cells from three sources: dietary fat, fatty acids released by adipose tissue lipolysis, and a process called de novo lipogenesis, or DNL, where the liver synthesizes new fatty acids from non-fat precursors like glucose. In cases of NAFLD, DNL is elevated compared to healthy controls; it does not shut down during fasting as it normally would, and transcription factors like SREBP-1c, ChREBP, and PPAR-gamma are overactive. Here’s the counterintuitive part: triglyceride accumulation itself may not be the primary issue. Animal experiments in the review indicate that blocking triglyceride incorporation into very low-density lipoprotein particles caused fat retention without liver injury. Additionally, inhibiting DGAT2, a key enzyme for producing triglycerides, reduced stored fat but worsened steatohepatitis because it forced more fatty acids down oxidative pathways that generated toxic byproducts. The visible fat in a biopsy may actually serve as a buffer, rather than being the cause of the problem. The real damage arises from lipotoxic intermediates: free fatty acids, free cholesterol, ceramides, and diacylglycerols. Their accumulation, known as lipotoxicity, drives two interconnected forms of cellular stress. The first is mitochondrial dysfunction, where the respiratory chain is overwhelmed, beta-oxidation falters, and reactive oxygen species build up. The second is endoplasmic reticulum stress, or ER stress, which occurs when unfolded proteins accumulate in the ER, triggering what’s called the unfolded protein response. If that response persists, it activates inflammatory and apoptotic signaling. In animal models, silencing a key stress kinase called JNK reduced both steatosis and steatohepatitis. These are not isolated molecular curiosities; they are the mechanisms that connect a fatty liver to a diseased one. Insulin resistance amplifies this situation. In healthy adipose tissue, insulin suppresses lipolysis, keeping fat contained within fat cells. When insulin resistance develops, that suppression fails, and free fatty acids flood into the portal circulation, reaching the liver in excess. Simultaneously, something paradoxical occurs in the liver itself. Hepatic insulin resistance disrupts glucose regulation, but insulin's fat-building program remains intact. SREBP-1c continues to drive DNL while beta-oxidation is suppressed. Fat accumulates from both sources at the same time. These excess fatty acids activate JNK and the IKK-NF-kappa-B axis, which are stress kinases that generate chronic inflammation. Here’s where the cycle closes in on itself: JNK and IKK-beta promote serine phosphorylation of insulin receptor substrates, disrupting downstream insulin signaling and worsening insulin resistance. Adipose insulin resistance increases fat delivery to the liver, which activates inflammatory kinases, which in turn worsen insulin resistance and increase fat delivery. Buzzetti and colleagues identify this self-reinforcing cycle as central to NAFLD progression. Two additional hits originate outside the liver entirely. The first arises from fat tissue itself in the form of adipokines—hormones secreted by adipocytes that have opposing effects on the liver. Leptin, a sixteen kilodalton hormone that typically suppresses appetite, increases in obesity as the brain becomes resistant to it. In the liver, it activates hepatic stellate cells through hedgehog and mTOR pathways and stimulates Kupffer cells to release TGF-beta-1, promoting fibrosis. Low doses of gut-derived endotoxin amplify leptin's effects, accelerating fibrosis in mouse models of NASH. Adiponectin exerts contrasting effects; it sensitizes cells to insulin, blocks NF-kappa-B, inhibits inflammatory cytokines like TNF-alpha and IL-6, and has direct antifibrotic effects possibly mediated through AMPK. Mice lacking adiponectin develop greater liver fibrosis, while mice administered recombinant adiponectin experience significantly improved steatohepatitis. Obesity shifts this balance: adiponectin levels decrease, leptin levels increase, and the liver exists in a hormonal environment that favors inflammation and scarring. The second external hit originates from the gut. The liver receives more than half of its blood supply from the splanchnic circulation, making it uniquely exposed to anything leaking from the intestine. Patients with NAFLD demonstrate increased gut permeability and higher rates of small-intestinal bacterial overgrowth compared to healthy controls. When the gut microbiome is dysfunctional, the intestinal barrier becomes leaky, allowing lipopolysaccharide—LPS, a fragment of bacterial cell walls—to enter the portal circulation. LPS activates toll-like receptors on liver cells, triggering JNK, p38, interferon regulatory factor three, and NF-kappa-B. The result is hepatic inflammation, fat accumulation, and insulin resistance. Dysbiosis also alters bile acid metabolism, depletes available choline, and can even increase endogenous alcohol production, each contributing to liver injury through different routes. So now you have a comprehensive picture: lipotoxic stress from excess fat intermediates, a self-amplifying insulin resistance cycle, an adverse adipokine environment from dysfunctional fat tissue, and a leaky gut delivering bacterial products directly to the liver. Multiple hits are acting simultaneously within the same organ. However, this still does not explain why some individuals are much more vulnerable than others under the same metabolic conditions. Genetics and epigenetics help fill that gap. The strongest common genetic signal for NAFLD is a variant in the PNPLA3 gene, known as patatin-like phospholipase domain-containing protein three. The I148M change, referred to as rs738409, is associated with dramatically higher hepatic fat. A meta-analysis of twenty-three studies found odds ratios of 3.41 for NAFLD and 4.44 for NASH associated with this variant. Carriers of the risk allele paradoxically show lower de novo lipogenesis but significantly higher liver fat because the mutant protein accumulates on lipid droplets and hampers fat mobilization. Knock-in mice with this change develop steatosis, confirming the mechanism. A second signal, TM6SF2, reduces very low-density lipoprotein secretion and traps fat in the liver. This was also linked to advanced fibrosis in two cohorts totaling over a thousand patients, although carriers appear somewhat protected from cardiovascular disease, likely because less fat exits the liver in lipoproteins. Next, consider epigenetics—changes in gene expression that do not alter the DNA sequence but are influenced by diet and environment. DNA methylation, histone modifications, and non-coding RNAs all regulate pathways relevant to NAFLD. A deficiency in methyl donors leads to hypomethylation and steatosis. SIRT1 deacetylase activity is reduced in NAFLD models. Some of these epigenetic marks may be passed across generations, and some are partially reversible; studies show that methylation changes improve after bariatric surgery. Inherited genetic variants and dynamic epigenetic states together explain why the same diet affects different livers in markedly different ways. The multiple-hit framework ultimately calls for a different approach to treatment. No single drug will resolve this issue. Liver biopsy remains the gold standard for distinguishing simple steatosis from NASH and for staging fibrosis, but it is impractical at a population scale, which is why the search for noninvasive markers is urgent. Two longitudinal studies mentioned in the review found that advanced fibrosis, rather than the NASH label itself, predicted overall mortality. This finding has significant implications for clinical trials: if fibrosis is the endpoint that truly matters, trial design and patient selection must reflect that. Buzzetti and colleagues conclude by calling for further characterization of the individual pathways, improved noninvasive markers, and targeted therapies that can address more than one hit at a time. The biology has become more complex, and the medicine will have to keep pace. This lecture was created by ennepō. Go to https://ennepo.ai to Discover, Create and Follow the latest research in your field. Read when you can. Listen when you want to.