Microbes present in the gastrointestinal tract have a tremendous effect on host physiology, including protection from pathogenic microorganisms [1], obtaining nutrients and energy and fermentation of non-digestible carbohydrates [2]. The structure of this microbial community contributes to the host in terms of modulating the immune system, metabolism and combating infection [3–5]. The composition and activities of intestinal bacteria may change in response to a number of environmental or lifestyle variables, one of the most important being diet [6]. In fact, it has been reported that dietary alterations are responsible for 57 % of the gut microbiota’s entire variation, whereas genetic background may only contribute to 12 % [7]. Furthermore, the disruption of the normal host microbial community can contribute to the development of chronic diseases such as obesity, diabetes and irritable bowel disease [8]. Diet is a major determining factor of the gut microbiota [3, 9, 10]. For example, previous study found that the ratio of Bacteroidetes to Firmicutes increased in the gut of people consuming either fat or carbohydrate restricted diets, and a higher ratio was associated with greater weight loss [11].

Deep frying in oil is one of the most popular cooking methods worldwide and provides an appealing flavor, color and texture to the foods. However, foods fried between 150 and 180 °C have been shown to absorb up to 8 to 25 % of the oil [12], and this is dependent on food types, oil properties and cooking methods, indicating that frying oil is becoming a major dietary component in the food formula. Thus, the safety of deep-fried edible oil (DFEO) is a concern because it is ingested with fried food. The continuous cooking at high temperature (160 ~ 190 °C) in the presence of air and moisture leads to a number of chemical reactions (e.g. oxidation, hydrolysis and polymerization, etc.) [13, 14]. These reactions alter the composition of the oil with the production of various types of oxidative products [15], such as oxidized fatty acids, polar compounds and polymeric products [16–18]. Previous studies confirmed that the consumption of deep oil fried foods contributes to chronic diseases such as obesity, atherosclerosis, hypertension and diabetes [19–21]. Furthermore, the intake of over-cooked edible oil could elevate blood pressure and impair vasorelaxiation in rats which maybe attribute to endothelial dysfunction [20], and the studies have also described the potential links between deep-fried oil and the increased risk of prostate cancer [22]. Previous work focused mainly on physical and chemical changes of the oils occurred during deep-frying process. To date there has not been sufficient studies to directly or indirectly investigate the ingestion of deep-fried edible oil (DFEO) on the changes in gut microflora. Therefore, the purpose of this study was to evaluate potential toxicity of DFEO on gastrointestinal environment through the regulation of gut microbiota structure. Moreover, the effect of DFEO consumption on energy metabolism, in particular lipid metabolism, was also investigated. To the best of our knowledge, this is the first attempt to reveal the potential links between energy metabolism and gut microbiota community structure.

This study found that DFEO consumption led the rats to having the lightest body weight, which may indicate that the toxic components from DFEO have a negative impact on the metabolism and thus decrease rat weight gain. The growth response of rats fed with DFEO in this study is consistent with the findings of López et al. who reported that diet containing deep-fried oil led to animals having less weight gain [23]. Considering that the significantly enriched KEGG pathways in control versus NEO groups are mainly associated with metabolism disorder, this study further provided the evidence that the repeated consumption of over-cooked edible oil may increase the risk of chronic diseases, which is consistent with other studies [22, 24]. In particular, only one significantly enriched metabolism KEGG pathway, glycerolipid metabolism, was detected in NEO versus DFEO group (Table 1), which may provide a better understanding of the mechanism involved in DFEO influence on lipids metabolism, because glycerolipid (GL)/free fatty acid (FFA) cycle is referred to as a “futile” cycle, in which it involves continuous formation and hydrolysis of GL with the release of energy at the expense of ATP [25]. Thus, this study may indicate that the DFEO toxicity impairs energy metabolism, in particular through regulating the pathway of glycerolipid metabolism. The inhibition of the pathway regarding glycerolipid metabolism led to the reduction in energy absorption in rats of DFEO group, which subsequently influence rats growing performance. Four genes expressions related to this pathway were also investigated, and the down-regulation of these genes, in particular, one of them, Lpin1, may play the key roles for regulating the energy metabolism, because gene Lpin1 encodes a magnesium-ion-dependent phosphatidic acid phosphohydrolase enzyme that catalyzes the penultimate step in triglyceride synthesis including the dephosphorylation of phosphatidic acid to yield diacylglycerol [26]. The finding in the lipids metabolism in this study is consistent with the decreased body weight of the rats in DFEO group compared to the rats in other two groups.

The influence of DFEO consumption on rats growing performance was further investigated from the damage induced by DFEO on intestinal mucosa, including directly destroy of gland tissue structure. These histological structural changes may be associated with direct damage induced by the toxic compounds in DFEO which may indirectly influence changes in the gut microbial profile. The bacterial profile study showed that more oil consumption in the diet led to a reduction of both Chao and Shannon indices, indicating the lower bacterial community richness and diversity in rats of NEO and DFEO groups. This was consistent with a previous study that ingestion of a high-fat diet in rats alters the gut microbiota, leading to a decrease in total bacterial density and diversity [27, 28], suggesting that diet is one of the most important factors for influencing the composition of gut microbiota. Furthermore, the lowest bacterial community diversity was found in the rats of DFEO group in this study, which further suggests that the chemical structure in DFEO may also significantly influence microbial structure. In general, the higher the microbial diversity, the more stable the ecosystem. Thus, from this study it might indicate that the consumption of DFEO reduced the intestinal microbial diversity, which may contribute to its potential negative effect on gut.

At phylum level, Firmicutes and Bacteroidetes were most dominant in all three groups, which were in agreement with previous studies [11, 29], and the oil consumption achieved a higher proportion of Firmicutes, which agrees with the previous reports that Firmicutes increased in fecal samples in mice following the feed of high fat diets [6, 11]. However, the decreased proportion of Bacteroidetes in DFEO group and increased in the NEO group compared with the control group may directly influence rats growing performance, and these bacterial patterns were consistent with their corresponding body weights found in this study.

At the genus level, Prevotella, this study found the proportion of Prevotella was decreased in the rats of NEO and DFEO groups, indicating that oil present in the diet may inhibit the proliferation of Prevotella, and induce the proliferation of bacteria which take part in the degradation and utilization of fat [30]. This result may be associated with a function of Prevotella, which has been implicated in the degradation and utilization of carbohydrate and simple sugars, and strongly related with long-term diet [31]. However, the lowest proportion of Prevotella in DFEO group may imply more complicated mechanism involved in this the interactions between diet and gut bacteria.

More importantly, rats in DFEO group were characterized with higher existence of genus Allobaculum (Firmicutes), but not found in control and NEO groups (Fig. 8). This difference may be used to characterize the gut microbiota in rats of DFEO group. Meanwhile, previous studies have reported that a high level of dietary fat is accompanied with a significant reduction in the abundance of lactic acid bacteria [32, 33]. However, in this study, the amount of Lactobacilli increased both in NEO and DFEO groups, especially in DFEO group. The plausible interpretation of these results may be that: the long-term administration of DFEO induced an immune stress on the intestine, which contributed to a loss of immune homeostasis. There is a direct relationship between the intestinal flora and the immune state of animals [34]. The immune system may have been primed to resist potential chronic stressors (e.g. various types of oxidative products in DFEO) by activating an increase of beneficial bacteria, such as Lactobacillus and Bifidobacterium. Meanwhile, the extra administration of oil caused the compensatory proliferation of Lactobacillus, as Lactobacillus can down-regulate lipogenesis and up-regulate lipolysis and fatty acid oxidation [35]. Alternatively, it has been reported that the longer the frying time, a lower pH is achieved, and the activity of H⁺ of DFEO becomes 100 times higher than that of non-heated oil [36]. Therefore, the administration of DFEO may induce a reduction of the gut pH value, in which the low pH environment further promotes the growth of Lactobacillus and Bifidobacterium. Nevertheless, the long-term effect, especially deep-fried oil needs to be further investigated. Considering the complication of reactions between diet and gut microbiotia, and in particular, there are no related studies describing different influences of non-heated and deep-fried oils on gut microbiota, the indicative action of these microbiota needs to be further verified.

Consumption of oil following deep-frying process had a different impact on body weight, lipid metabolism, gut microbiota and histological properties. This study found that the rats with feeding of DFEO had the lowest body weight among the three groups. More importantly, this is the first report to show that there was only one significantly enriched KEGG pathway, glycerolipid metabolism, detected in NEO versus DFEO group, indicating that the DFEO toxicity impairs energy metabolism, in particular through the regulation of the pathway of glycerolipid metabolism. The inhibition of the pathway regarding glycerolipid metabolism was highly down-regulated by gene of Lpin1, and this change may be associated with the decreased body weight of the rats in DFEO group. Compared with non-heated oil, the consumption of deep-fried oil has potential harmful effects on gut histological structure. As a pro-inflammatory potential, the consumption of DFEO was found to be associated with a decrease in Prevotella and an increase in Bacteroides. In particular, rats in DFEO group were also characterized with higher presence of Allobaculum (Firmicutes). Furthermore, the increased proportion of Lactobacillus and Bifidobacterium in the rats of DFEO group may be related with an immune stress induced by DFEO-containing diet.