Inulin Supplementation Lowered the Metabolic Defects of Prolonged Exposure to Chlorpyrifos from Gestation to Young Adult Stage in Offspring Rats

Increasing evidence indicates that chlorpyrifos (CPF), an organophosphorus insecticide, is involved in metabolic disorders. We assess the hypothesis whether supplementation with prebiotics from gestation to adulthood, through a modulation of microbiota composition and fermentative activity, alleviates CPF induced metabolic disorders of 60 days old offspring. 5 groups of Wistar rats, from gestation until weaning, received two doses of CPF pesticide: 1 mg/kg/day (CPF1) or 3.5 mg/kg/day (CPF3.5) with free access to inulin (10g/L in drinking water). Then male pups received the same treatment as dams. Metabolic profile, leptin sensitivity, insulin receptor (IR) expression in liver, gut microbiota composition and short chain fatty acid composition (SCFAs) in the colon, were analyzed at postnatal day 60 in the offspring (PND 60). CPF3.5 increased offspring’s birth body weight (BW) but decreased BW at PND60. Inulin supplementation restored the BW at PND 60 to control levels. Hyperinsulinemia and decrease in insulin receptor β in liver were seen in CPF1 exposed rats. In contrast, hyperglycemia and decrease in insulin level were found in CPF3.5 rats. Inulin restored the levels of some metabolic parameters in CPF groups to ranges comparable with the controls. The total bacterial population, short chain fatty acid (SCFA) production and butyrate levels were enhanced in CPF groups receiving inulin. Our data indicate that developmental exposure to CPF interferes with metabolism with dose related effects evident at adulthood. By modulating microbiota population and fermentative activity, inulin corrected adult metabolic disorders of rats exposed to CPF during development. Prebiotics supply may be thus considered as a novel nutritional strategy to counteract insulin resistance and diabetes induced by a continuous pesticide exposure.


Introduction
human neonate during its development. We evaluated the effect of the different treatments in rats on lipid and glucose metabolism, insulin and leptin, gut microbiota composition and SCFA production at 60 days of age. The supplementation with inulin is relevant in this study because it may help to identify new properties of prebiotics in both the mother and the child exposed to pesticides, and to develop new strategies against metabolic programming in later life.

Protein extraction from liver
Proteins were extracted from 20 mg of liver tissue according to methods described in our previous studies [26,27]. Briefly, tissues were lysed in 1X Ripa Buffer containing a cocktail of protease inhibitors (Thermo Scientific, France). Lysis was performed using a Tissue Lyser device (Qiagen, France). Samples were incubated for 15-20 min at 4°C and centrifuged (14000g; 15 min; 4°C). The supernatant was removed and stored at -80°C. Protein concentrations were measured using a Pierce BCA Protein Assay kit according to the manufacturer's instructions.

Gel electrophoresis and Western Blot analysis
For gel electrophoresis, the protein samples were resuspended and heated for 5 min at 95°C and loaded on a 4-12% SDS-PAGE (Criterion XT Bis-Tris Gel, Biorad, France). Proteins were transferred onto a nitrocellulose membrane (Biorad, Fance) after electrophoresis. Membranes were blocked with TBS-T/5% milk and incubated overnight at 4°C with the primary antibody anti insulin receptor β (IRβ) (rabbit monoclonal, #3020, Cell Signaling, diluted 1/1000). The blot was then incubated with peroxidase conjugated secondary antibody (Abcam, diluted 1/ 5000). The protein signal was detected using the ECL kit (Amersham Biosciences). Proteins were analyzed using anti-Signals on autoradiographic films were quantified by scanning densitometry using ImagQuant 350 (GE Healthcare, France).

DNA Extraction and 16S RNA qPCR analysis
Total bacterial count and specific bacterial profile were evaluated by quantitative PCR analyses targeting bacterial group-specific 16S rRNA genes using the Rotor-Gene system (Qiagen, France). Total DNA was extracted from 20 to 25 mg of colon content using the Qiagen QIAamp Fast DNA stool kit according to the manufacturer's instructions (Qiagen, France). PCR inhibitions were tested with TaqMan1 Exogenous Internal Positive Control and the Taq-Man1 universal master Mix (Life Technologies S.A, France). No PCR inhibition was detected using 10 −3 dilutions for the feces samples. DNA from each sample was amplified using selected primers and probe sets given in Table 1 [28,29]. For bacterial SYBR-green amplification studies, a melting curve was added to show the amplification specificity and the following PCR profile was used: 1 cycle at 95°C for 12 min, followed by 40 cycles of 95°C for 15 s, 60°C for 30 s, 72°C for 30 s. Cycle amplification data were quantified according to standard curves. Data were expressed as log10 (copy number)/g of feces.

Short Chain Fatty Acid (SCFA) analysis
The concentration of acetate, propionate and butyrate in the fecal material were determined after water extraction of acidified samples using gas-liquid chromatography (Thermo Scientific, Focus GC-AutoInjector AI 3000) as described in our previous study [30].

Results
Dam's body weight, food and drink intake (GD 0 and LD 21) Dam's body weight, food intake and drink intake were measured every 3 days. Neither CPF nor inulin significantly affected maternal weight gain, food intake and drink intake (S1 Table). No signs of cholinergic toxicity such as tremor, salivation or diarrhea were observed in animals during the whole experiment.

Offspring's body weight and growth, food and drink intake
Offspring's body weight, growth and food intake were recorder every two days from PND1 to PND60. The global growth curve of different groups did not appear to be affected by treatments except CPF3.5 animals. Indeed, a "drop" of the curve for these exposed rats to the highest dose of CPF was observed at PND53 (Fig 1A). The mean BW of control offspring group at birth and PND 60 were 6.77 ± 0.12 g and 355.9 ± 8.2 g, respectively. Inulin alone did not affect offspring BW and BW gain but animal's weight at PND 60 decreased with inulin supplementation (p<0.05) (Fig 1D). Gestational exposure to CPF significantly increased BW at birth (p<0.001) (Fig 1B). BW gain and BW at PND 60 were decreased in both CPF-exposed groups (p<0.05) (Fig 1C and 1D). This reduction in weight gain was not associated with changes in either food or water intake (S2 Table). The supplementation of inulin in CPF-exposed groups induced a decrease of BW at birth as compared with CPF animals. However, no effects of inulin Table 1. Primers and probes used in this study.

Traget Primers and Probes Sequences (5'-3')
TaqMan System All bacteria a F-Bact 1369 [28] b [29] doi:10.1371/journal.pone.0164614.t001 on BW gain and BW at PND 60 were observed in CPF-exposed animals ( Fig 1C and 1D). Thus CPF exposure modified the body weight growth pattern of the offspring's from birth to weaning. In both CPF exposed rats, inulin supply induced a decrease of birth body weight and restored the BW at PND 60 to control levels.

Metabolic parameters
The effect of the different treatments on metabolic parameters was compared between the different groups. Inulin alone did not affect metabolic parameters while leptin level at PND 60 decreased when inulin was supplied in the drinking water (p<0.05) (Fig 2). CPF exposure during pregnancy and lactation induced different effects on the metabolic parameters depending on the dose used: CPF3.5 group showed a significant higher increase in glycaemia at PND 60 (p<0.05) compared to the control ( Fig 3A). A slight increase in plasma insulin level (p = 0.08) at PND 60 was observed in CPF1-exposed offspring's while a slight decrease in insulin secretion was observed in animals exposed to the higher dose (p = 0.08). When supplemented with inulin, the levels of insulin in both CPF1 and CPF3.5 groups were close to the values observed in control animals (p<0.05) (Fig 3B). The lipid status (i.e., the total cholesterol, LDL and HDL) was not affected by inulin or CPF exposure alone except for the level of TG which was lower in CPF3.5-exposed rats as compared to control animals (p<0.05). The TG level was recovered when animals were co-exposed with Control groups (CPF0, circles), CPF-exposed groups (CPF1: 1 mg/kg/day, squares; CPF3.5: CPF3.5 mg/kg/day, triangles) or inulin groups (black symbol). Signification *p<0.05; **p<0.01,***p<0.001, i = 0.058. inulin and CPF3.5 in comparison to CPF3.5 animals alone (p<0.01) (Fig 4). Fig 5A shows an increase in ALT enzyme for the CPF3.5-exposed groups (p<0.05) and a slight increase (p = 0.07) in ALT in CPF1 exposed animals compared to the control. The difference in ALT level did not persist when animals received both CPF3.5 and inulin. Meanwhile, no changes in AST level were observed with the different treatments ( Fig 5B). Thus CPF modified the metabolic status of the offspring's and inulin restored the levels of some metabolic parameters in CPF groups to ranges comparable with the controls.

Insulin protein expression
Western blot analysis evidenced a significant (p<0.05) decrease of IRβ protein expression in the liver of CPF1 as compared to controls. When supplemented with inulin, IRβ was significantly (p<0.01) recovered only in CPF1-exposed group (Fig 6A and 6B).

Gut microbiota
Total bacterial count and specific bacterial profile were evaluated by quantitative PCR analyses targeting bacterial group-specific 16S rRNA. Perinatal exposure to inulin alone significantly increased the global bacterial population (p<0.05) (Fig 7A) and total SCFA production (p<0.01) with a significant increase in the proportion of butyric acid (p<0.05) ( Table 2). A significant decrease in the population of Firmicutes (Fig 7B) was observed in CPF-exposed animals (p<0.05) for CPF1 and a slight decrease (p = 0.08) for CPF3.5 respectively. Perinatal exposure to CPF alone significantly decreased C. coccoides group (p<0.05) and tended to reduce C. leptum count too (Fig 8A and 8B). In the CPF3.5+inu group, C. coccoides group was significantly (p<0.01) higher than CPF3.5-exposed rats (Fig 8A). The Firmicutes/Bacteroidetes ratio decreased only in the CPF1 group in comparison to the control (p<0.05) (Fig 7D). In the CPF1+inu group, the total count of Firmicutes population and C. leptum group were still significantly lower when compared to the control group (p<0.05) (Figs 7B and 8B). For the production of SCFAs, supplementation of CPF1 animals with inulin increased the total SCFAs production (p<0.05) as well as the proportion of butyric acid (p = 0.058) compared to CPF1 animals without inulin. However, in CPF3.5 group, inulin supplementation only increased the proportion of butyric acid (p = 0.058) when compared to CPF3.5 without inulin. Of note, a high decrease of total SCFAs was observed in CPF3.5 animals supplemented with inulin as compared to the group receiving only inulin (305 ± 16 μmol/g vs. 393 ± 27,1 μmol/g respectively, p<0.01). Thus, CPF decreased Firmicutes population. Inulin supplementation enhanced the total bacterial population, SCFA production and butyrate levels in CPF groups.

Discussion
This study demonstrates that CPF exposure during a critical window of development induced permanent defects in normal newborn male rats in terms of their phenotype and their metabolic regulation during adulthood. The exposure to CPF during gestation led to an increase in BW when the offspring were born. However, a long term exposure extending from gestation to adulthood induced a decrease in BW gain and BW in the offspring at day 60. Perturbations in the perinatal environment can impair metabolic programming that can consequently increase susceptibility to T2D in adulthood. Several studies addressing the role of developmental exposure to pesticides with diabetes have rapidly expanded over the past years suggesting that they contribute to metabolic programming [9,31]. In our study we demonstrate that a continuous exposure to a "safe" dose of CPF beginning in the womb until adult life impairs adult metabolism of the offspring's leading to insulin and lipid dysregulation. Based on these findings, one can assume that the exposure to CPF during gestation, lactation and even during other sensitive periods of development, as pre-puberty, impairs developmental programming. Although our design does not able determining precisely which of the different targets sensitive windows were involved in the "programmed" metabolic defects, it complements the Barker hypothesis which makes a link between low birth weight and diabetes [32], extending the same outcomes into the situation to developmental exposure to pesticides, even in the absence of growth retardation. The specific pattern of birth weight increase and BW loss after pre-and postnatal exposure to CPF differs from studies published by others on exposure to CPF [33] or Parathion [34], which is another organophosphorous in neonates. This discrepancy can be explained by the different doses used, the time window, the type (acute or chronic) of exposure and the mode of administration (oral, subcutaneous) in different studies.
The major drastic effects of CPF exposure on glucose and lipid metabolism seen in the 60 days old offspring rats were observed in those subjected to the higher dose of 3.5 mg/kg/day CPF. Indeed, an increase in glycaemia concomitant with a drastic reduction of insulin level was observed. This result may reflect an alteration of the pancreas integrity and functionality and as a consequence, the inability of the pancreas to control glycaemia in CPF-exposed rats. Moreover, the alteration of glucose metabolism observed for the higher dose of CPF resulted in a significant increase in serum alanine aminotransferase reflecting hepatic injuries. Indeed, Mansour et al., showed that CPF caused oxidative damage leading to impaired liver cell membrane permeability and thus the release of hepatic enzymes [35]. Furthermore, in association with the decrease in insulin secretion, a subsequent decrease in triglycerides was seen in CPF3.5-exposed rats. As insulin is an important regulatory factor of lipid metabolism [36], one can assume that an alteration in insulin secretion perhaps modified the triglyceride content. Other mechanisms can also contribute to the disturbance of triglyceride synthesis, notably the ones targeting adipose tissue and/or liver signaling that are essential for homeostasis regulation [37,38]. We recently reported that CPF exposure increased para-cellular permeability in the Early Prebiotic and Pesticide Co-Exposure and Metabolic Disorders small intestine [39] and delayed intestinal epithelial maturation [25], which may suggest a modification of intestinal absorption of triglycerides in the blood stream.
In agreement with the studies by Slotkin [9,10], CPF1-exposed rats presented a hyperinsulinemia. This result showed that an apparently "safe" exposure was probably maintained in the CPF1-group by the compensatory hypersecretion of insulin. This situation is close to what was observed in a pre-diabetic state. We also reported that the liver of CPF1 animals contain reduced amounts of IRSβ, which support our hypothesis and reflect a compensatory response to chronically elevated insulin level and alteration of insulin sensitivity [40].These findings highlighted a mechanistic link between insulin signaling in the liver and the subsequent emergence of hyperinsulinemia and hyperlipidemia.
The original approach used in this study was based on the use of inulin, acting as a prebiotic to counteract the side effects of CPF on metabolism. Prebiotics are known to improve the host's health by inducing favorable changes in intestinal microbiota [41]. Inulin selectively stimulates beneficial Bifidobacteria and Lactobacilli in vitro [42,43] and also in human subjects and rodents [44][45][46].
Intestinal SCFA concentration, especially butyrate, the preferred energy source of colonocytes, increases when inulin is consumed [45,46]. In agreement with this, inulin supplementation resulted in an increase in the general population of bacteria and the total SCFA production. An increase in SCFA, observed in supplemented groups, may explain the reduction of insulin level noticed in CPF1-exposed rats because of the role of SCFAs in the activation of G-protein-coupled free fatty acid receptor (GPR43) in the adipose tissue [47]. Moreover, the observed increase in butyrate in rats consuming inulin is probably due to a cross-feeding phenomenon such that the butyrate-producing bacteria belonging to C. coccoides (cluster XIVa) and C. leptum group (cluster IV) such as Roseburia intestinalis or Faecalibacterium Early Prebiotic and Pesticide Co-Exposure and Metabolic Disorders prausnitzii or others clostridial clusters such as I, III, XI, XV, XVI [48] which use acetate produced by Bifidobacteria [49]. To our knowledge, we are the first to describe a decrease in Firmicutes, C. coccoides and C. leptum group in particular, after a perinatal CPF exposure in rats. This microbiota pattern resembles what we observed in the Human diabetic subjects with a decrease of Firmicutes and particularly butyrate-producing Clostridiales [50,51], Bacteroides vulgates, and Bifidobacteria [52]. It is well known that aberrant intestinal microbiota can induce a translocation of bacterial fragments and the development of "metabolic endotoxemia", leading to systemic inflammation and insulin resistance [17]. Access to inulin may counteract endotoxemia in CPF-exposed rats as it alleviates the decrease in the phylum Firmicutes and clostridial clusters XIVa (C. coccoides) and IV (C. leptum).
In addition to the beneficial effect of inulin in CPF-exposed rats on microbiota and SCFA production, our results showed that it modified the metabolic status of the CPF-exposed rats according to the dose used. In CPF3.5 group, free access to inulin induced an increase in insulin, triglycerides and alanine transferase in the serum to a level close to those of control groups, showing that inulin restored some features of the disturbed metabolic profile in CPF3.5-exposed animals. Conversely, in CPF1-exposed rats, inulin decreased insulin secretion, initially high in CPF1-exposed rats, to a level comparable to those of control animals. Moreover, prebiotics increased the expression of IRβ in CPF1-exposed rats suggesting an enhancement of insulin sensitivity and a decrease of insulin resistance in these animals. Thus, altogether, our results show that our study is the first to demonstrate that bringing a continuous source of prebiotic to the mother and the infant, allowed young rats to alleviate the side effect of long-term exposure to CPF on their metabolism during their adult lives.

Conclusions
In summary, the current results indicate that CPF exposure during pre-and postnatal period may be a limiting factor, at least in rodents, for the onset of a normal regulation of metabolism in the offspring during their adult life. It is suggested that organophosphate insecticides can increase the risk of diabetes mellitus. We propose that prebiotics, which have the ability to alter the microbiota in a positive manner, is a safe and cost-effective nutritional strategy to counteract CPF insulin resistance and diabetes in later life. Human clinical trials should be undertaken to confirm these effects. However, additional basic research is necessary to better understand the crosstalk between microbiota and the host in order to elucidate the exact mechanism by which, microbiota alleviates the metabolic defects induced by CPF in adults. Experiments are in progress to determine whether the gut/brain axis is also involved in such regulation.