A Perspective on Metabolic Syndrome and Nonalcoholic Fatty Liver Disease

Introduction

Nonalcoholic fatty liver disease (NAFLD) has emerged as the most common chronic liver disease of this decade, which is in part correlated with the global epidemic of obesity, the metabolic syndrome, and type 2 diabetes mellitus (T2DM). ^1,2 Cardiovascular disease (CVD) is the leading cause of death in NAFLD, ³ followed by malignancy and liver failure, all of which are predominantly due to the progressive form of NAFLD, namely nonalcoholic steatohepatitis. ¹

The last few years have seen advances in many aspects of NAFLD, including molecular mechanisms involved in hepatic lipotoxicity and insulin resistance (IR), mechanisms of CVD risk, identification of novel gene associations, non-invasive assessment, and therapeutic agents. Here, we highlight some of these findings that are likely to have a major influence on the field.

Does NAFLD Predict the Development of T2DM and Metabolic Syndrome?

NAFLD is now increasingly diagnosed in individuals who do not have T2DM or metabolic syndrome. This supports the assertion that the conventional paradigm of NAFLD representing the mere “hepatic manifestation” of the metabolic syndrome is outdated, and it is now becoming clear that NAFLD is also a pathogenic determinant of metabolic syndrome. ¹ With regard to this evolving concept, there is a growing body of evidence strongly supporting the notion that NAFLD precedes the development of T2DM and metabolic syndrome. ⁴

There are now approximately 30 longitudinal, retrospective, and prospective studies that provide strong evidence for NAFLD as an independent predictor of the future development of T2DM. ^{4 –15} Moreover, more than a dozen of longitudinal studies with a relatively long duration of follow-up reported that NAFLD, as diagnosed by either ultrasonography or biochemistry, is independently associated with an increased incidence of several components of metabolic syndrome, not only of T2DM. ^{4,16 –24}

Amongst the prospective studies that have used ultrasonography to diagnose NAFLD and that have assessed the risk of developing T2DM, the risk of T2DM varied markedly from an approximately 60% increase to approximately a six-fold increase in risk. ^{7 –15} This wide interstudy variation in T2DM risk might reflect differences in NAFLD severity, as the study by Park et al. ¹³ showed that the incidence rate of T2DM increased sharply according to the ultrasonographic severity of NAFLD at baseline (normal, 7%; mild, 9.8%; moderate-to-severe, 17.8%). Even after adjustment for multiple confounders, the hazard ratios (HRs) for T2DM development were higher in the mild-NAFLD group [adjusted HR, 1.09; 95% confidence interval (CI) 0.8–1.5] and in the moderate-to-severe-NAFLD group (adjusted HR, 1.73; 95% CI 1.0–3.0) compared with the no-NAFLD group. ¹³

Of note, Sung et al. have investigated the impact of abdominal obesity, IR, and hepatic steatosis on the risk of T2DM at the 5-year follow-up in a cohort of over 12,000 Korean individuals. ¹⁴ These data showed that each of these three risk factors was associated with T2DM risk, and each risk factor was associated with an approximate doubling of the T2DM risk after adjustment for other established risk factors. Notably, in the same observational cohort of individuals, the authors examined the impact of resolution of hepatic steatosis on the risk of T2DM at the 5-year follow-up to establish whether NAFLD improvement was associated with T2DM risk reduction. These data demonstrated that there was a significant T2DM risk reduction in those subjects in whom hepatic steatosis on ultrasonography resolved over time. In particular, in those subjects, risk of T2DM decreased to the background risk of someone who had never had hepatic steatosis. ¹⁵ Conversely, the individuals in whom the severity of hepatic steatosis worsened over 5 years showed a marked increase in T2DM risk [adjusted odds ratio (OR)=6.13; 95% CI 2.5–14.6 compared with the risk in people with resolution of hepatic steatosis], further supporting the notion that more severe forms of NAFLD are associated with a greater risk of T2DM. ¹⁵

From these published data it is important to understand how mechanistically pre-existent NAFLD may lead to the development of T2DM and metabolic syndrome. With the exception of cases where NAFLD results from either familial hypobetalipoproteinemia or patatin-like phospholipase domain-containing 3 (PNPLA3) gene polymorphisms in which NAFLD is usually dissociated from IR, ^25,26 NAFLD directly causes hepatic IR in most cases. Molecular and genetic studies have recently supported the view that the pathogenetic mechanisms contributing to intrahepatocytic lipid compartmentation are determinants of whether fatty liver is or is not associated with IR and metabolic syndrome. ²⁷ Proposed mechanisms by which NAFLD causes hepatic IR implicate various lipid species, inflammatory signaling, and other cellular modifications. ²⁷ Experimental studies have elucidated a key role for hepatic diacylglycerol activation of protein kinase Cɛ (PKCɛ) in triggering hepatic IR. ²⁸ Alterations in the regulation of intrahepatic lipid droplet metabolism influence the intracellular compartmentation of diacylglycerol, which dictates whether or not PKCɛ will translocate to the plasma membrane so promoting lipotoxicity and hepatic IR. ^27,28

Are There Differences Between Metabolic Syndrome–Related NAFLD and PNPLA3 -Related NAFLD?

Although, as discussed above, NAFLD is strongly associated with metabolic syndrome features, ^1,2 there are some people who develop NAFLD but do not have metabolic syndrome. ²⁹ This suggests that other factors besides metabolic syndrome–specific mechanisms are involved in NAFLD pathogenesis.

To date, the strongest evidence that any single gene may cause NAFLD, independently of metabolic syndrome, is with a specific polymorphism in the PNPLA3 gene. Increased hepatic fat content ^30,31 and, in contrast to metabolic syndrome, low circulating serum triglycerides (TG) levels and a dissociation between hepatic steatosis and IR ^26,32 have been described with the GG allele (I148M) gene variant of PNPLA3 in people with NAFLD. Although there is modest attenuation of metabolic syndrome features, such as improved IR and decreased TG levels, with the GG genotype, it is uncertain whether this may also affect risk of CVD with NAFLD. Perhaps surprisingly, it was recently reported that in patients with NAFLD there was a greater prevalence of increased carotid intima media thickness among those with the GG genotype compared to those with the combined CG+CC genotypes. ³³ However, further studies are needed to elucidate this issue and determine whether the NAFLD phenotype occurring with the GG allele gene variant is associated with higher or lower risk of CVD.

The precise function of PNPLA3 is uncertain, but both hepatic lipogenic and lipolytic roles have been proposed. ³⁴ PNPLA3 has acyltransferase activity, converting lysophosphatidic acid to phosphatidic acid within the triacylglycerol synthesis pathway, and also has triacylglycerol hydrolysis activity. It has been suggested that homozygosity of a G → C substitution at position 148 by changing an isoleucine to a methionine increases lipogenic activity and decreases hydrolytic activity. However, elucidating the precise function of hepatic PNPLA3 in NAFLD and investigating the influence of I148M has proved difficult. Until recently, animal studies have also provided limited insight. Very recent work with the I148M PNPLA3 knockin mice has shown that, although these mice have normal liver fat levels on a chow diet, when fed a high-sucrose diet, their liver fat levels increased two- to three-fold. ³⁵ The increased liver fat content in these knockin mice was also accompanied by a 40-fold increase in PNPLA3 on hepatic lipid droplets, with no increase in hepatic PNPLA3 mRNA. ³⁵ These data provide the first direct evidence that physiological expression of the PNPLA3 148M variant causes NAFLD, and that the accumulation of catalytically inactive PNPLA3 on the surfaces of lipid droplets is very important and associated with the accumulation of lipid in the liver.

The strength of effect of the I148M variant on NAFLD disease severity across different populations has been analyzed in a systematic review and meta-analysis of 16 studies. ³⁶ In this analysis, homozygosity for I148M was associated with ∼75% higher hepatic fat content compared with the CC variant. The I148M GG variant was also associated with more severe NAFLD. For example, GG homozygotes had higher serum aminotransferase levels and an approximately three-fold greater risk of hepatic necro-inflammation and fibrosis when compared with CC homozygous individuals. Evaluation of the risk associated with heterozygosity for the I148M variant suggested that the additive genetic model best explained the effect of I148M on the susceptibility to develop NAFLD. Nevertheless, and surprisingly, carrying two G alleles did not seem to increase the risk of severe histological features. Interestingly, meta-regression showed an inverse association between male sex and the effect of I148M on hepatic fat content, ³⁶ suggesting that there may be dissimilar influences of I148M in men and women.

Although decreased serum TG levels ³² have been described with the GG variant, Qiao et al. showed in an animal model that PNPLA3 increased lipogenesis, inducing hepatic very-low-density lipoprotein (VLDL)–TG accumulation that, in turn, may be secreted into serum, thereby increasing serum TG levels. ³⁷ The mechanisms by which PNPLA3 regulates VLDL assembly and secretion are poorly understood. PNPLA3 might alter the availability of hepatic lipid for VLDL assembly and secretion, thereby affecting serum TG concentrations. Similarly by regulating hepatic lipolytic activity, PNPLA3 might alter the availability of hepatic TG for VLDL assembly and secretion, thereby also affecting serum TG levels.

There is some evidence that the PNPLA3 I148M GG genotype facilitates incorporation into liver lipid droplets of different types of fatty acid (e.g., palmitoleic acid), thereby altering the fatty acid composition of the lipid droplets. ³⁸ Whether the PNPLA3 genotype influences the fluxes of hepatic fatty acids to affect liver fat and VLDL assembly and secretion in NAFLD is uncertain. Recently, Nobili et al. showed that the PNPLA3 GG genotype diminished the beneficial effect of the long-chain omega-3 polyunsaturated fatty acid docosahexanoic acid (DHA) treatment on liver disease end points in children with NAFLD. ³⁹ DHA and eicosapentaenoic acid (EPA) are known to exert hypotriglyceridemic effects, ⁴⁰ and we have recently also shown a benefit of high tissue enrichment with DHA with omega-3 fatty acid treatment to decrease hepatic fat content in NAFLD. ⁴¹ The hypotriglyceridemic effects of omega-3 fatty acids have also been demonstrated in NAFLD patients, ⁴² and it is known that omega-3 fatty acids reduce hepatic VLDL–TG secretion via apolipoprotein B100 degradation, ⁴³ downregulation of de novo TG synthesis, ⁴⁴ enhancement of liver β-oxidation, and reduction of adipose tissue inflammation, thus reducing plasma nonesterified fatty acid delivery from adipose tissue to the liver. ⁴⁵ We suggest further work is needed to understand how PNPLA3 I148M modifies VLDL assembly (and/or secretion) and how PNPLA3 I148M modifies fluxes of omega-3 fatty acids in particular to affect liver disease severity and to affect metabolic syndrome features, such as increased serum TG concentrations in NAFLD.

Conclusions

The current evidence supports a key role of NAFLD in the pathogenesis of hepatic IR and T2DM. NAFLD may also have relevant clinical implications for diagnosing, preventing, and treating the metabolic syndrome. Additionally, we believe that considerable attention should be paid to the development of NAFLD in the absence of metabolic syndrome and in the absence of overweight/obesity (i.e., the so called “lean” NAFLD). A key unanswered question is whether “lean” NAFLD and PNPLA3-related NAFLD carry the same risk as NAFLD occurring with metabolic syndrome for the development of extrahepatic (CVD and noncardiovascular) complications. Future well-designed prospective studies will be needed to answer this question.