Proteomic evaluation of neonatal exposure to 2,2 ,4,4 ,5-pentabromodiphenyl ether

Abstract

Exposure to the brominated flame retardant 2,2 ,4,4 ,5-pentabromodiphenyl ether (PBDE-99) during the brain growth spurt disrupts normal brain development in mice and results in disturbed spontaneous behavior in adulthood. The neurodevelopmental toxicity of PBDE-99 has been reported to affect the cholinergic and catecholaminergic systems. In this study we use a proteomics approach to study the early effect of PBDE-99 in two distinct regions of the neonatal mouse brain, the striatum and the hippocampus. A single oral dose of PBDE-99 (12 mg/kg body weight) or vehicle was administered to male NMRI mice on neonatal day 10, and the striatum and the hippocampus were isolated. Using two-dimensional fluorescence difference gel electrophoresis (2D-DIGE), we found 40 and 56 protein spots with significantly (p < 0.01) altered levels in the striatum and the hippocampus, respectively. We used matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-ToF-MS) to determine the protein identity of 11 spots from the striatum and 10 from the hippocampus. We found that the levels of proteins involved in neurodegeneration and neuroplasticity (e.g., Gap-43/neuromodulin, stathmin) were typically altered in the striatum, and proteins involved in metabolism and energy production [e.g., alpha-enolase; gamma-enolase; ATP synthase, H+ transporting, mitochondrial F1 complex, beta subunit (Atp5b); and alpha-synuclein] were typically altered in the hippocampus. Interestingly, many of the identified proteins have been linked to protein kinase C signaling. In conclusion, we identify responses to early exposure to PBDE-99 that could contribute to persistent neurotoxic effects. This study also shows the usefulness of proteomics to identify potential biomarkers of developmental neurotoxicity of organohalogen compounds.

Figure 1

Experimental design and pooling procedure. Individual control (1–9) and treated (10–18) tissue samples were separately pooled into three control (1–3) and treated (5–7) pools per sample type (striatum and hippocampus). In addition to these biological replicates, we made a technical replicate of pools 1 and 5 (4 and 8, respectively). Control pools (1–4) were labeled with Cy3, treated pools (5–8) with Cy5, and a mixture of all pools (pools 1–8) with Cy2 as internal standard. Four gels were run in parallel, as indicated, with extra unlabeled protein added to gel 4.

Figure 2

2D-DIGE image of fluorescently labeled proteins from striatum (A) and hippocampus (B) of PD11 mice. Proteins were extracted from the striatum (A) and hippocampus (B) of mice 24 hr after they received a single oral dose of PBDE-99 (treated) or vehicle (control) on PD10. Proteins were labeled with Cy3 (control), Cy5 (treated), and Cy2 (mixture of control and treated as internal reference; see “Material and Methods”) and separated on pH 4–7 IPG strips in the first dimension. This procedure was followed by separation on 12.5% SDS–polyacrylamide gels in the second dimension, and the gels were scanned with a Typhoon 9400 fluorescent scanner. The arrows point to the 9 (A) and 10 (B) spots found to be regulated and identified spots, respectively. The numbers correspond to the numbers in Table 1.

Figure 3

Box plots of the spot ratios from the four gels from the (S) striatum and the (H) hippocampus. The box limits are the first and third quartile, closely representing 50% of the data. The whiskers extend to the most extreme data point, which is no more than 1.5 times the length of the box away from the box. The overall average and SD of the log₂ 95% ratios averages 0.141 ± 0.016 for the striatum and 0.165 ± 0.022 for the hippocampus, corresponding to log₂ 95% prediction limits of 0.199 and 0.242, respectively. S1–S4, gels 1–4 (see Figure 1) for striatum; H1–H4, gels 1–4 for hippocampus.

Figure 4

Measurement of relative expression versus average log intensity (MA) plot showing the result of the statistical analysis of the striatum proteins. The log₂ values of spot intensity on the y-axis are plotted against log₂ values of fold change on the x-axis. The intensity was calculated as the log₂ of the raw intensity of the average of Cy3 and Cy5 across all four gels. The spots correspond to the striatum proteins identified on all four DIGE gels. The spots marked in blue represent the differently regulated proteins, whereas the squared blue spots represent the nine proteins picked and identified from the striatum gel 4.

Figure 5

MS/MS identification of Gap-43 (neuromodulin). MS/MS spectrum for spot 10 in Figure 2A. The parent ions m/z 1986.91 and 1217.56 were used for MS/MS identification.

Figure 6

Western blot validation of Gap-43 expression. Gap-43 and the endogenous reference protein β-tublin were simultaneously detected in control (−) and PBDE-99–exposed (+) striatum (S) and hippocampus (H) samples after SDS–PAGE and electroblotting to nitrocellulose.