Starch and Fiber Contents of Purified Control Diets Differentially Affect Hepatic Lipid Homeostasis and Gut Microbiota Composition

Abstract

BackgroundInterpretation of results from diet-induced-obesity (DIO) studies critically depends on control conditions. Grain-based chows are optimized for rodent nutrition but do not match the defined composition of purified diets used for DIO, severely limiting the comparability. Purified control diets are recommended but often contain high starch and only minor fiber amounts. It is unknown whether this composition leads to metabolic alterations compared with chow and whether the addition of refined fibers at the expense of starch affects these changes.MethodsIn this experiment, 6-week-old C57BL/6N mice were fed (i) a conventional purified control diet (high-starch, low-fiber; Puri-starch), (ii) an alternative, custom-made purified control diet containing pectin and inulin (medium-starch, higher-fiber; Puri-fiber), or (iii) grain-based chow for 30 weeks (N = 8–10).ResultsPuri-starch feeding resulted in significantly elevated levels of plasma insulin (p = 0.004), cholesterol (p < 0.001), and transaminases (AST p = 0.002, ALT p = 0.001), hepatic de novo lipogenesis and liver steatosis, and an altered gut microbiota composition compared with chow-fed mice. In contrast, Puri-fiber exerted only minor effects on systemic parameters and liver lipid homeostasis, and promoted a distinct gut microbiota composition.ConclusionCarbohydrate-rich purified diets trigger a metabolic status possibly masking pathological effects of nutrients under study, restricting its use as control condition. The addition of refined fibers is suited to create purified, yet physiological control diets for DIO research.

Oral Supplementation with the Polyamine Spermidine Affects Hepatic but Not Pulmonary Lipid Metabolism in Lean but Not Obese Mice

“…Insulin levels were analyzed in duplicates with an ultrasensitive mouse insulin ELISA (#90080, Chrystal Chem, Elk Grove Village, IL, USA). The homeostasis model assessment for insulin resistance (HOMA-IR) and 1/fasting insulin as a surrogate index for insulin resistance were calculated and normalized to body weight as described before [ 39 ]. HOMA-IR and 1/fasting insulin were normalized to body weight since this was shown to considerably enhance the correlation with glucose clamp-derived measures [ 40 ].…”

“…Briefly, left lung volumes were determined by Archimedes’ principle [ 42 ], and lung sampling was performed according to systematic uniform random sampling methods [ 43 ]. Tissue blocks were embedded in epoxy resin (Epon; Serva, Heidelberg, Germany) as described elsewhere [ 39 ] and analyzed by transmission electron microscopy.…”

“…Livers were isolated, weighed and cut into ~1 mm slices. Slices were randomly allocated to snap freezing and storage at −80 °C for mRNA expression analyses, or fixation in 4% paraformaldehyde/0.2 M HEPES buffer (pH 7.4) and embedding in epoxy resin as described elsewhere [ 39 ] for histological analysis.…”

“…Whole sections were digitalized with an AxioScan.Z1 Scanner (Zeiss, Oberkochen, Germany) at a primary magnification of 40×. Subsequently, automated image analysis was performed using the Tissuemorph DP software from Visiopharm (version 5.3.1.1723, Hoersholm, Denmark) as described before [ 39 ].…”

“…cDNA was synthesized from 1 µg mRNA using the iScript cDNA Synthesis Kit (BioRad, Hercules, CA, USA). Real-time PCR analysis was performed using a C1000 Thermal Cycler (CFX384 Real-Time System, BioRad) as described before [ 39 ]. Gene expression analysis was conducted in triplicates, and the primers used are listed in Table 2 .…”