No mortality occurred and there were no changes noted in terms of behaviour, activity, posture, gait, or external appearance for any of the different treatments for the duration of the two feeding studies. All animals in all treatment groups were healthy and appeared normal throughout the course of the study.
Effects on weight gain, food consumption and feed conversion efficiency (FCE) of male Wistar rats
7-Day rat feeding study
Body weight gain was measured daily over the 7-day period. Although there were small differences between the different treatment groups, these were not significant (Fig. 1a; Table S3; p = 0.859) and were within the normal limits for rats of this strain, age and sex. Those fed the commercial Purina chow gained more weight and those fed Reference diet 1 gained the least weight, whilst rats fed the parental near isogenic line (Control group) gained more weight compared to those fed MON810 (Test group), but these differences were small and non-significant. Similarly, whilst there were small differences in food intake at the end of the 7-day period, this being greatest on the control diet (283 g) compared to the Test group (267.8 g), again this was not significant between the five groups (Fig. 1b; p = 0.695). These results are reflected in the calculated feed conversion efficiency (FCE) with those fed Purina chow having the lowest FCE (i.e. considered the most efficient users of feed); but this also was not significant between the five diets (Fig. 1c, Table S4; p = 0.600).
28-Day rat feeding study
There was no significant differences in body weight gain of rats between the different treatment groups after 28 days (F4,15 = 1.35, p = 0.297 after 4 weeks). The residuals were not significantly different to a normal distribution (p = 0.682). The greatest increase was seen in the rats fed the Test diet (MON810), which was 35.7% greater than its near isogenic parental line (Fig. 2a; Table S5), followed by the commercial rodent diet. Food intake, calculated on a weekly basis, was the greatest on the control diet (332.5 g on week 4) compared to 294.75 g for the test group, however these differences between treatment groups were not significant (Fig. 2b; p = 0.450). The Feed conversion efficiency value was lowest for the Test group. However, as with all other parameters measured, differences in FCE between the five treatment groups were not significant (Fig. 2 c; Table S6; p = 0.270).
In vivo effects on differential gene expression in the epithelial cells of the small intestine of male Wistar rats
To obtain a better understanding of the underlying molecular responses in mammals to MON810, differential gene expression in the epithelial cells of the small intestine of rats fed formulated diets containing MON810, its near isogenic parental line, two conventional corn varieties, and a corn-based control diet were investigated using comparative proteomic profiling. Two-way (Figs. 3a, 4a) and five-way (Figs. 3b, 4b) comparisons of the differentially expressed protein spots were made between the different treatment groups both for the 7-day and 28-day studies. A Student t test p value less than 0.05 and the false discovery rate (FDR, from q-values) were used to qualify differentially expressed protein spots.
7-Day rat feeding study
Comparison of the proteome maps for the IECs of rats fed MON810 with those fed the near isogenic parental line (Control; designated Mon Conv) identified 1216 protein spots in common between the two groups, which were expressed at similar levels. However, four protein spots (Spot No. 1950, 2212, 1328, 2045) were shown to be upregulated in response to the MON810 diet, whereas six (Spot No. 2883, 2443, 2400, 2004, 1777, 2249) were upregulated in the control fed rats. LC–MS/MS analysis identified two of these upregulated (2.4-fold) protein spots in the MON810 group to be stress-related proteins (catalase and 60 kDa heat shock protein); no stress related proteins were upregulated in the Control group (Fig. 3a; Table 1). Similar trends were also identified when the MON810 group was compared with the other groups. When compared with the commercial rodent diet 1222 protein spots were expressed at similar levels, with both groups exhibiting significant upregulation of only two protein spots each. Those upregulated in the MON810 group (spot No. 1146, 1630) encoded three stress-related proteins (catalase, 60 kDa heat shock protein and stress-induced phosphoprotein 1), but again no-stress related proteins were upregulated in the commercial rodent diet group (Spot No. 2306 and 3258). Comparison between the MON810 group and the Reference diet 1 (Mon Garst) group showed that of the seven protein spots (Spot No. 2212, 3248, 2279, 3243, 1950, 2626, 2048) upregulated in the Test group, three were classified as stress-related proteins, (thioredoxin-dependent peroxide reductase, peroxiredoxin-6 and LDLR chaperone MESD precursor). No proteins appeared to be upregulated in this Reference group. When compared with the Reference Group 2 (Mon Gold), again the only upregulated proteins were in the Test group, but in this instance there was only one (Spot No. 2626), which was identified as representing two stress-related proteins, (peroxiredoxin-6 and LDLR chaperone MESD precursor); however, the level of upregulation was only 1.2-fold.
Not only were stress-related proteins identified in rats from the Test group when compared to the other treatments, but they were also shown to be upregulated when comparing the other diets with one another (Fig. 3a; Table 1). For example, when comparing the rodent diet (MCert) group with Reference 1, two stress-related proteins (thioredoxin-dependent peroxide reductase and protein DJ-1) were identified out of the seven upregulated proteins (Spot No. 3248, 1222, 2212, 3243, 2608, 1976, 1516) in the former group. When this group was compared with the Reference 2 group, there was only one upregulated protein and this again was in the rodent diet group; however, this protein was not associated with stress. The greatest number of differentially upregulated proteins was seen when comparing the rodent diet (MCert) group with the Control (near isogenic line of MON 810) group, where seven (Spot No. 1580, 2511, 2016, 2212, 2045, 1976, 1950) and five (Spot No. 2883, 2021, 2249, 1146, 2400) protein spots, respectively, were shown to be differentially upregulated. The 60 kDa heat shock protein was upregulated by threefold in the former group whilst stress-induced phosphoprotein 1 was upregulated in the latter. Comparisons between the Reference 1 group and Control group revealed that six protein spots were upregulated (Spot No. 2249, 1257, 1777, 2400, 2883, 2443) in the control and one of these, catalase, a stress-related protein was upregulated 2.6-fold. Comparison between the two reference groups only identified the stress-related protein T-complex protein 1 subunit beta as being upregulated.
In contrast to all the above two-way comparisons, no stress-related proteins were identified to be upregulated when the Control group was compared to the Reference 2 group. Whilst the expression of 1223 protein spots was not significantly different to one another, three (Spot No. 2883, 2249, and 2443) were upregulated in the control diet, but none of these were associated with stress-related proteins.
Analysis of the data in a five-way comparison showed that far fewer proteins were differentially expressed, this being five, four, and three protein spots for the rodent diet (of which one was the stress-related protein thioredoxin-dependent peroxide reductase), control and MON810 (of which two were the stress-related proteins LDLR chaperone MESD precursor and peroxiredoxin-6), respectively. There was no differential expression in either of the two reference groups (Fig. 3b; Table 1). Full details of protein spots in common, and those upregulated, between the treatments, with their fold-change, are presented in Fig. 3a, b, Table S7. What is interesting to note was the similarity between the different treatments with the number of protein spots in common ranging from 1214 to 1225; this very small variance illustrates the robustness of the data.
28-Day rat feeding study
Two-way and five-way comparisons were also carried out for the 28-day feeding studies. What is interesting to note was the overall similarity in the response at 7 days and 28 days in terms of differential regulation of stress-related proteins between different treatments. Overall, in two-way analyses, fewer proteins were differentially expressed and fewer of these were classified as stress-related (Fig. 4a, b; Table 2). Comparison of the MON810 group with its near isogenic line (Control) revealed that the number of upregulated protein spots in the former group had increased to nine (Spot No. 3320, 1485, 2550, 2222, 3190, 2179, 1707, 2232, 2442) whilst in the latter this number had decreased to three (Spot No. 3123, 3506, 3255); however the number of stress-related proteins in the Test group remained at two (stress-induced phosphoprotein 1 and peroxiredoxin-1), although these were different to those upregulated in the 7-day study; in this instance one stress-related protein, superoxide dismutase, was upregulated in the control group (Fig. 4a). In contrast to the 7-day study, no proteins were differentially expressed between the MON810 and rodent diet group in the 28-day study, with 1446 proteins exhibiting expression levels in common. When compared to Reference 1, and Reference 2, there were seven (Spot No. 2209, 1485, 1960, 1972, 2912, 4082, 4101) and four (Spot No. 2232, 2149, 1564, 1485) upregulated proteins, respectively and two (Spot No. 2893, 2965) and one (Spot No. 3305), respectively for the reference groups. Of those upregulated in the MON810 group vs Reference 1, two in the Test group were stress-related proteins (stress-induced phosphoprotein 1 and 60 kDa heat shock protein) and one in the comparator (T-complex protein 1 subunit beta). When compared to the Reference 2 group, only two stress-related proteins (both phosphoprotein 1) were upregulated in total, and both in the MON810 group. The only other stress-related protein to be upregulated between the other treatments in the two-way comparisons was peroxiredoxin-1 (2.8-fold) in the rodent diet when compared with the control diet (Fig. 4a; Table 2).
Comparisons of differentially expressed proteins between all five treatment-groups identified that only two protein spots were upregulated, both in the MON810 group (Spot No. 2912, 1485), one of which was stress-induced phosphoprotein 1 (upregulated 3.1-fold). Full details of protein spots in common, and those upregulated, between the treatments, with their fold-change, are presented in Fig. 4a, b, Table S8. There were slightly more proteins visualized on the proteome maps for the 28-day study, with the variance between treatments being even smaller (number of protein spots ranging from 1445 to 1447).
To assess the functional relevance of changes in the differentially expressed proteins identified, proteins were aligned according to their molecular functions, based on information provided by the online resource UniProt classification system. Some proteins were annotated manually, based on literature searches and closely related homologues. There were 28 groups for the 7-day study and 21 groups for the 28-day study (Figs. S1, S2).