Neuronal expression and subcellular localization of cholesterol 24-hydroxylase in the mouse brain

Denise M O Ramirez, Stefan Andersson, David W Russell, Denise M O Ramirez, Stefan Andersson, David W Russell

Abstract

Cholesterol 24-hydroxylase is a cytochrome P450 (CYP46A1) that is selectively expressed in the brain and is responsible for the majority of cholesterol turnover in the central nervous system. Mice deficient in 24-hydroxylase exhibit impaired learning and defective hippocampal long-term potentiation, suggesting that the metabolism of cholesterol by this enzyme is required for learning and memory formation. To determine where in the neuron cholesterol turnover was taking place, monoclonal antibodies directed against 24-hydroxylase were generated by immunization of mice with recombinant protein and used to detect the enzyme in brain homogenates, cultured neurons, and histological sections. 24-Hydroxylase was localized to the endoplasmic reticulum and was distributed throughout the cell bodies and dendrites of multiple types of neurons; the enzyme was not detected in axon terminals or in the cells of 24-hydroxylase knockout mice. 24-Hydroxylase was highly expressed in pyramidal neurons of the hippocampus and cortex, in Purkinje cells of the cerebellum, and in hippocampal and cerebellar interneurons. Within the retina, 24-hydroxylase was detected in ganglion cells and some but not all cells of the inner nuclear layer. These findings reveal the microsomal localization of 24-hydroxylase and provide subcellular insight into cholesterol turnover in the brain.

(c) 2008 Wiley-Liss, Inc.

Figures

Fig. 1
Fig. 1
Assessment of monoclonal antibody specificity. A: Monoclonal antibodies 1A7, 1F4, and 1F11 were used in immunoblotting experiments of whole-brain homogenates from wild-type and 24-hydroxylase knockout mice as described in Materials and Methods. 1A7 and 1F11 specifically detected a protein with the predicted molecular weight of 24-hydroxylase (~50,000), whereas 1F4 did not. Immunoblotting with an antiserum recognizing βIII-tubulin showed that equal amounts of protein were present in each sample. B: Immunoblotting of 24-hydroxylase in transfected CHO-K1 cells. Increasing amounts of total cell homogenate from cells expressing the mouse 24-hydroxylase (m24, lanes 2–6), human 24-hydroxylase (h24, lanes 7–11), and homogenate from mock-transfected cells (lane 1) were separated by SDS-PAGE and subjected to immunoblotting with antibodies 1A7 and 1F11. Both antibodies recognized the mouse and human 24-hydroxylases. C: Cross-species recognition of 24-hydroxylase by monoclonal antibodies 1A7 and 1F11. Transfected (TF) cell homogenates and whole-brain homogenates from mouse, rat, and rabbit were subjected to immunoblotting with antibodies 1A7 and 1F11. Both antibodies detect 24-hydroxylase from each of the three species. Immunoblotting with an antiserum recognizing α-tubulin showed that equal amounts of protein were present in each sample. D: Immunodetection of 24-hydroxylase in mouse embryonic brain tissue and cultured neurons. Microsomal membrane proteins were isolated from dissected embryonic day 16 cortex and hippocampal regions or from neurons prepared from these regions and cultured for 14 days in vitro. Immunoblotting experiments with 1A7 and 1F11 revealed the 24-hydroxylase protein in both wild-type (WT) preparations but not in the samples prepared from 24-hydroxylase knockout (KO) mice. Immunoblotting with an antiserum recognizing P450 reductase showed that equal amounts of protein were present in each sample analyzed. In the detection of 24-hydroxylase, lanes marked E16 Brain were exposed to film for 1.5 minutes; those marked 1° Neuron were exposed to film for 30 seconds.
Fig. 2
Fig. 2
Localization of 24-hydroxylase in transfected cells by immunofluorescence. CHO-K1 cells were transfected with a mouse 24-hydroxylase expression plasmid and stained with the DNA binding dye DAPI and an irrelevant monoclonal antibody, 3C11 (A), or monoclonal antibodies 1F11 (B) or 1F4 (C) that recognize 24-hydroxylase. Green fluorescence represents 24-hydroxylase, and nuclear DNA is shown in blue. A reticular staining pattern is observed in expressing cells with 1F11 and 1F4 but not with the irrelevant antibody 3C11. In D–F, CHO-K1 cells were cotransfected with mouse 24-hydroxylase and rat NADPH-cytochrome P450 reductase (P450 Red), then stained with 24-hydroxylase monoclonal antibody 1A7, a rabbit polyclonal antibody recognizing P450 Red, and DAPI. 24-Hydroxylase staining is shown in green, P450 reductase staining in red, and DNA staining in blue. The merged image (F) shows strong colocalization, indicated by a yellow color, between exogenously expressed 24-hydroxylase and P450 reductase. In G–I, CHO-K1 cells transfected with the mouse 24-hydroxylase plasmid were stained with 1A7, a rabbit polyclonal antiserum recognizing endogenous protein disulfide isomerase (PDI), and DAPI. DNA staining appears in blue, 24-hydroxylase staining in green, and PDI in red. Yellow staining in the merged image (I) indicates that 24-hydroxylase colocalized with endogenous PDI. Scale bar = 5 μm.
Fig. 3
Fig. 3
Localization of 24-hydroxylase in cultured neurons by immunofluorescence. Primary neurons prepared as described in Materials and Methods from wild-type (WT) or 24-hydroxylase knockout (KO) mouse embryos as indicated were stained with DAPI and 1A7 (A,B), or 1F4 (C,D), or 1F11 (E,F), or 1A7 and an antiserum recognizing brain lipid binding protein (BLBP), an astrocyte marker (G,H). DAPI staining appears in blue, 24-hydroxylase staining in green, and BLBP staining in red. 1A7, 1F4, and 1F11 reveal 24-hydroxylase signal in cell bodies and processes of cultured wild-type neurons. No 24-hydroxylase staining is detected in neurons from knockout mice. 1A7 staining is present in wild-type neurons but not in BLBP-positive astrocytes. The morphology of astrocytes prepared from knockout mice is indistinguishable from that of wild-type astrocytes as demonstrated by BLBP staining. Scale bar = 10 μm.
Fig. 4
Fig. 4
Coexpression of 24-hydroxylase with dendritic MAP2 but not presynaptic Rab3A in cultured neurons. Primary neurons prepared as described in Materials and Methods from wild-type (WT) or 24-hydroxylase knockout (KO) mouse embryos were stained with DAPI, 1A7, and a polyclonal antiserum recognizing MAP2 (A–F) or with DAPI, 1A7, and a polyclonal antiserum recognizing Rab3A (G–L). DAPI staining appears in blue, 24-hydroxylase staining in green, and MAP2 and Rab3A staining in red. The merged images from wild-type neurons show overlap between 24-hydroxylase and MAP2 signals (C) but no colocalization of 24-hydroxylase and Rab3A (I). No 24-hydroxylase staining is detected in neurons from knockout mice (D,J). Neurons from the mutant mice show normal morphology as demonstrated by MAP2 and Rab3A staining (E,K). Scale bar = 5 μm.
Fig. 5
Fig. 5
Colocalization of 24-hydroxylase with a known endoplasmic reticulum protein in cultured mouse neurons. A–C: Wild-type (WT) primary neurons were prepared as described in Materials and Methods and stained with DAPI, monoclonal antibody 1A7, and a polyclonal antiserum recognizing the NADPH-cytochrome P450 reductase (P450 Red). 24-Hydroxylase staining is shown in green and P450 reductase staining in red. Nuclei are shown in blue. The merged image shows strong colocalization of the two staining patterns in the cell body and somewhat less colocalization in the dendrites. D–F: 24-Hydroxylase knockout (KO) neurons were stained with DAPI, 1A7, and P450 reductase antibodies. No staining of 24-hydroxylase (green fluorescence) was detected, although the P450 reductase (red fluorescence) staining pattern was the same as that observed in wild-type neurons. Scale bar = 5 μm.
Fig. 6
Fig. 6
24-Hydroxylase is expressed in the endoplasmic reticulum, which extends throughout the dendrites of cultured mouse neurons. Wild-type neurons were prepared as described in Materials and Methods and stained with the 1A7 monoclonal antibody recognizing 24-hydroxylase and a polyclonal antiserum recognizing NADPH-cytochrome P450 reductase (P450 Red; A–E) or 1A7 and a rabbit polyclonal antiserum recognizing synaptophysin (F–J). Sections of proximal dendrites are shown. Merged images on the left (C–E) show many areas of colocalization between 24-hydroxylase (green) and P450 reductase (red), whereas merged images on the right (H–J) show little or no colocalization between 24-hydroxylase (green) and synaptophysin (red). Portions of the dendrites (boxed insets C,H) were rendered in three dimensions, and then enlarged four times (D,I) or 10 times (E,J). Colocalized pixels are shown in white in these three-dimensional images. The reticular patterns of 24-hydroxylase and P450 reductase are visible. Scale bars = 1 μm in A (applies to A–C,F–H); 0.25 μm in D (applies to D,I); 0.1 μm in E (applies to E,J).
Fig. 7
Fig. 7
Expression of 24-hydroxylase in the hippocampus. A–F: Cryosections were prepared from female wild-type (WT) and 24-hydroxylase knockout (KO) mouse brains and stained with the monoclonal antibody 1A7, and immunocomplexes were visualized by diaminobenzidine-streptavidin-horseradish peroxidase (DAB-HRP) staining. Strong expression was detected in the CA1 region (A), in which 24-hydroxylase staining was visible in the cell bodies and dendrites of pyramidal neurons. Lower levels of expression were detected in the granule neurons of the dentate gyrus, whereas hilar interneurons had high levels of 24-hydroxylase expression (C). No staining was detected in hippocampal sections from knockout mouse brains (B,D). Higher magnification (×40) images of wild-type sections demonstrate 24-hydroxylase staining in CA1 dendrites (E) and those of an interneuron in the stratum radiatum (F). Scale bars = 100 μm in A (applies to A,B); 50 μm in C (applies to C,D); 25 μm in E (applies to E,F).
Fig. 8
Fig. 8
Expression of 24-hydroxylase in the cerebellum. Cryosections were prepared from female wild-type (A,C) and 24-hydroxylase knockout (B,D) mouse brains and stained with the monoclonal antibody 1A7, and immunocomplexes were visualized by DAB-HRP staining. Strong 24-hydroxylase staining was detected in Purkinje cells and their associated dendritic trees as well as in some granule cell layer interneurons, which were provisionally identified as Golgi cells (A,C). Cerebellar sections from knockout brains were devoid of 24-hydroxylase staining (B,D). Scale bars = 100 μm in A (applies to A,B); 50 μm in C (applies to C,D).
Fig. 9
Fig. 9
Expression of 24-hydroxylase in the cortex. Cryosections were prepared from female wild-type (WT) and 24-hydroxylase knockout (KO) mouse brains and stained with the monoclonal antibody 1A7, and immunocomplexes were visualized by DAB-HRP staining. 24-Hydroxylase was detected in the somata of cortical neurons within layers II/III, V, and VI but not in nonpyramidal neurons of layer IV or those of layer I (A). A diffuse staining of the neuropil was present in all layers of the cortex. No staining was visible in cortical sections from knockout brains (B). Scale bar = 100 μm.
Fig. 10
Fig. 10
Expression of 24-hydroxylase in the retina. Cryosections were prepared from male wild-type (A,C) and 24-hydroxylase knockout (B,D) mouse eyes, stained with monoclonal antibodies 1A7 (A,B) or 1F11 (C,D), and immunocomplexes were visualized by DAB-HRP treatment. Strong 24-hydroxylase staining was observed with both monoclonal antibodies in retinal ganglion cells of the ganglion cell layer (GCL). A lower level of staining was detected with both antibodies in some but not all cells located at the edge of the inner nuclear layer (INL) and by the 1A7 antibody in the retinal pigmented epithelial layer (RPE). No staining was visible in other retinal layers or in retinal sections from knockout eyes. IPL, inner plexiform layer; OPL, outer plexiform layer; ONL, outer nuclear layer; RL, receptor layer. Scale bar = 50 μm.

Source: PubMed

3
Abonnieren