Mercury Reduces the Enzymatic Activity of Neprilysin in Differentiated SH-SY5Y Cells

Abstract

Levels of amyloid beta (Aβ) in the central nervous system are regulated by the balance between its synthesis and degradation. Neprilysin (NEP) is associated with Alzheimer's disease (AD) by its ability to degrade Aβ. Some studies have involved the exposure to mercury (Hg) in AD pathogenesis; therefore, our aim was to investigate the effects on the anabolism and catabolism of Aβ in differentiated SH-SY5Y cells incubated with 1-20 μM of Hg. Exposure to 20 µM of Hg induced an increase in Aβ-42 secretion, but did not increase the expression of the amyloid precursor protein (APP). Hg incubation (10 and 20 µM) increased NEP protein levels; however, it did not change NEP mRNA levels nor the levels of the amyloid intracellular domain peptide, a protein fragment with transcriptional activity. Interestingly, Hg reduced NEP activity at 10 and 20 µM, and circular dichroism analysis using human recombinant NEP showed conformational changes after incubation with molar equivalents of Hg. This suggests that the Hg-induced inhibition of NEP activity may be mediated by a conformational change resulting in reduced Aβ-42 degradation. Finally, the comparative effects of lead (Pb, 50 μM) were evaluated. We found a significant increase in Aβ-42 levels and a dramatic increase in APP protein levels; however, no alteration in NEP levels was observed nor in the enzymatic activity of this metalloprotease, despite the fact that Pb slightly modified the rhNEP conformation. Overall, our data suggest that Hg and Pb increase Aβ levels by different mechanisms.

Keywords: Alzheimer’s disease; amyloid precursor protein; beta-amyloid peptide; lead; mercury; neprilysin; retinoic acid.

FIG. 1.

RA promotes differentiation of SH-SY5Y cell line. SH-SY5Y cells (50 000) were seeded on cover-slips for 24 h and then treated with 10 µM RA for 7 days. A, Cell morphology was evaluated by contrast phase microscopy; a representative image is presented. B, Representative immunofluorescence microscopy images showing changes in the expression of proteins of neuronal markers: tubulin βIII (TRITC, red color) and synapsin (FITC, green color). The blue staining indicates the nuclei (DAPI). RA = retinoic acid.

FIG. 2.

Cell viability in differentiated SH-SY5Y cells treated with HgCl₂. Differentiated SH-SY5Y cells (200 000) were incubated with different concentrations of Hg (1, 5, 10, 20, 50, or 100 µM) for 48 h. A group without treatment was used as a negative control. A, MTT test was used to determine cell viability. B, Densitometric analysis of p-AKT protein levels. Total AKT was used as a loading control. The media of 3 independent experiments ± SE is presented. *P ≤ .05 compared with control cells according to 1-way ANOVA and Tukey’s post-hoc tests.

FIG. 3.

Effect of metals in Aβ-42 and APP levels. Differentiated SH-SY5Y cells (1.5 × 10⁶) were incubated with 1, 5, 10, or 20 µM of Hg or 50 µM of Pb for 48 h. Cells treated with 50 µM Pb were used as a positive control. The group without treatment was used as a control. A, Aβ-42 levels in total protein extracts. B, A representative blot of Hg and Pb effects on APP levels. C, Densitometric analysis of APP protein levels from 3 independent experiments; mean ± SE. Actin was used as a loading control. *P ≤ .05 different with respect to control cells according to 1-way ANOVA and Dunnet’s multiple comparison post-hoc tests. RA = retinoic acid.

FIG. 4.

Hg increases NEP protein but not mRNA levels in differentiated SH-SY5Y cells. Differentiated cells (1.5 × 10⁶) were exposed to 1, 5, 10, or 20 µM of Hg or 50 µM Pb for 48 h. The negative control was the group without treatment while cells treated with 50 µM Pb were used as a positive control. One hundred micrograms of total protein was used for electrophoresis and WB analysis. A, A representative blot image. B, Densitometric analysis of 3 independent experiments; mean ± SE. Actin was used as a loading control. C, A representative image of the end-point PCR assay. Actin was used as a housekeeping gene. D, Relative values of 2ΔCt data from q-PCR assay of 3 independent experiments expressed as the mean ± SE. HPRT was used as housekeeping gene. *P ≤ .05 different with respect to control cells according to 1-way ANOVA and Dunnet’s multiple comparison post-hoc tests. NEP = neprilysin.

FIG. 5.

Hg does not affect nuclear levels of AICD. SH-SY5Y cells (1.5 × 10⁶) were differentiated with RA for 7 days, and then treated with 0, 10, or 20 µM of Hg or 50 µM Pb for 48 h. Cells were fixed with 4% PFA and tested for AICD by immunofluorescence. A representative image is presented, the blue staining indicates nuclei labeled with DAPI and in green are nuclei positive to AICD peptide.

FIG. 6.

Hg reduces the enzymatic activity of NEP. SH-SY5Y cells (1.5 × 10⁶) were differentiated with RA for 7 days, and then treated with 10 or 20 µM of Hg or 50 µM Pb for 48 h. A, A kinetic evaluation of NEP activity. B, The area under the curve from each treatment is presented. Data represent the mean ± SE from 3 independent experiments. Phosphoramidon was used as a positive control of NEP activity inhibition. *P ≤ .05 different with respect to control cells according to 1-way ANOVA and Tukey’s post-hoc tests. NEP=neprilysin. Phospho = phosphoramidon.

FIG. 7.

NEP is the principal protease responsible for the fluorogenic substrate degradation. Nondifferentiated SH-SY5Y cells were transfected with siRNA NEP or siRNA control for 24 h. A, Protein expression 24 h after the transfection. B, A kinetic evaluation of NEP activity. C, The area under the curve is presented. Phosphoramidon, an inhibitor of NEP and related metalloproteases, was used as a positive control for inhibition of the enzymatic activity. *P ≤ .05 different with respect to control cells according to 1-way ANOVA and Tukey’s post-hoc tests. NEP = neprilysin. Phospho = phosphoramidon.

FIG. 8.

Hg changes the conformational structure of hrNEP. Recombinant hrNEP (0.5 µM in 50 µl of PBS pH 7) was used for titration with 2–20 molar equivalents of Hg or Pb dissolved in the same buffer. The spectrum was recorded every 2 nm in a CD spectrophotometer. Changes in the conformational structure of hrNEP induced by the addition of molar equivalents of Hg (A) or Pb (B) are presented. C, Effect of titration of molar equivalents of Hg or Pb at 222 nm in the hrNEP spectrum. The data were normalized to 1 when the rhNEP was free of any metal.