We read with interest the article in HEPATOLOGY by Spruss et al.1 The links between portal endotoxemia, Toll-like receptor 4 (TLR4) activity, and fatty liver disease are established, although they await full elucidation.2, 3 The article by Spruss et al. details elegant observations, but we remain uncertain of its interpretation. The article assesses two questions. First, does the presence of a TLR4 knockout influence hepatic metabolism, and second, does it alter the hepatic response to a high-fructose diet? The data, however, are presented in a manner that solely facilitates analysis of the first question and not the second.
TLR4 wild-type and knockout mice were given unrestricted access to either plain water or water with 30% fructose in addition to what we assume to be standard chow. Compared to the wild-type mice who drank water, the TLR4 knockout mice who drank water had significant reductions in liver weight and interferon regulatory factor 3, and nonsignificant reductions in hepatic triglyceride count, inducible nitric oxide synthase (iNOS), 4-hydroxynonenal, and elevated tumor necrosis factor alpha (TNFα) messenger RNA concentrations. TLR4 status therefore appears to influence hepatic metabolism, although no overall pattern clearly emerges.
Several factors limit the interpretation of the second research question. Critically, the two intervention arms of either water or water plus 30% fructose were not matched for either carbohydrate, energy, or prebiotic intake. As a result it is not possible to determine whether the observed changes are specific to the fructose, energy, or prebiotic aspects of the diets. Furthermore, there is a weight gain disparity between the two fructose-fed arms. Weight changes have been previously demonstrated by the same research group to reflect dietary energy intake patterns.4 The more than doubling of weight gain in the fructose-fed wild-type mice and the absence of significant weight change in the fructose-fed knockout mice implies significant differences in exposure to fructose. Consumption data are not presented.
A final issue is that the outcomes of the fructose-fed knockout mice were analyzed relative to the wild-type water-fed mice as opposed to their own knockout water-fed control. Such comparisons are not valid because hepatic metabolism clearly differs between wild-type and knockout mice. When we analyze the data presented in figure 1b of the article, the hepatic triglyeride ratio between the fructose-fed and water-fed wild-type mice appears to be 3.3 (values not presented), whereas the ratio for the fructose-fed and water-fed knockout mice is 3.0. The relative influence of TLR4 status on fructose-induced steatosis is therefore around 10%, which is in stark contrast to the presented absolute value of 40%. Similar differences between the absolute data presented and the relative changes observed are noted in terms of lipid peroxidation. When we analyze the data presented in figure 4, for 4-hydroxynonenal staining density, a ratio of 3.0 occurred between the fructose-fed and water-fed wild-type mice as compared to a ratio of 4.2 between the knockout mice. The corresponding ratios for iNOS were 3.3 and 2.6. Given the discrepancies in energy and fructose exposure between the two arms, TLR4 status appears to play little if any role in determining the hepatic triglyceride accumulation and lipid peroxidation effects of a high-fructose diet.
The authors drew two separate conclusions from this article: first, that it further demonstrates a clear link between TLR4 status and the development of hepatic steatosis and steatohepatitis, and second, that it has demonstrated a fructose-specific effect. We remain uncertain of either of these conclusions.