The function(s) of the Biogenesis of Lysosome-related Organelles Complex 1 (BLOC-1) during brain development is hitherto largely unknown. Here, we investigated how its absence alters the trajectory of postnatal brain development using as model the pallid mouse. Most of the defects observed early postnatally in the mutant mice were more prominent in males than in females and in the hippocampus. Male mutant mice, but not females, had smaller brains as compared to sex-matching wild-types at postnatal day 1 (P1), this deficit was largely recovered by P14 and P45. An abnormal cytoarchitecture of the pyramidal cell layer of the hippocampus was observed in P1 pallid male, but not female, or juvenile mice (P45), along with severely decreased expression levels of the radial glial marker GLAST. Transcriptomic analyses showed that the overall response to the lack of functional BLOC-1 was more pronounced in hippocampi at P1 than at P45 or in the cerebral cortex. These observations suggest that absence of BLOC-1 renders males more susceptible to perinatal brain maldevelopment and although most abnormalities appear to have been resolved in juvenile animals, still permanent defects may be present, resulting in faulty neuronal circuits, and contribute to previously reported cognitive and behavioural phenotypes in adult BLOC-1-deficient mice.

Histomorphometrical and Immunohistochemical Analyses

P1 male and female, and P45 male, wild-type and pallid mice were anesthetized with isoflurane (30-32%) and transcardially perfused with phosphate-buffered saline (PBS, 0.1 M, pH 7.4) containing 4% (w/v) paraformaldehyde (Electron Microscopy Sciences, Hatfield, PA). The brains were rapidly dissected out, post-fixed overnight in 4% (w/v) paraformaldehyde at 4°C, and cryoprotected in 15% (w/v) sucrose. Coronal sections were cut on a cryostat (Leica, Buffalo Grove, IL) collected sequentially, and paired along the anterior-posterior axis before further processing.

For Nissl Staining, coronal brain sections (20 μm) were stained with a 1% (w/v) cresyl violet (Sigma-Aldrich Corp., St. Louis, MO) solution as previously reported (Lee et al., 2018). Photographs were acquired on a Zeiss Axioskop equipped with a Zeiss colour or monochrome Axiocam using the AxioVision software (Zeiss, Pleasanton, CA) and used to determine the thickness of the cerebral cortex (layers I-VI) in rostral and caudal sections, the outer border and total sectional area of the hippocampus, and the cell density and cytoarchitecture in the Cornu Ammonis (CA)1 subfield of the hippocampus. Measurements were performed by two observers masked to the genotype, sex and age of the animals, from which each histological section has been generated, with the aid of the Zeiss Axiovision or the NIH Image Software (ImageJ, http://rsb.info.nih.gov/ij/). To control and reduce variations due to staining intensity, for the analysis of CA1 cell density, images were acquired using a monochromatic camera (AxioCam; Zeiss). For image analysis of the CA1 cell layer, the “vertical profile plot analysis” feature of ImageJ was used as follows: a grid set at 16,000 square pixels was over-imposed onto each image, and three identical rectangles of height equal to 800 pixels and width equal to 400 pixels (156 µm x 88 µm) were set at a fixed distance from the lateral ventricle (third, fifth, and seventh columns of the grid). The rectangles were positioned such that the long axis would cross completely the pyramidal cell layer, and used to automatically calculate a profile plot as the average pixel intensity per row (i.e., across the short horizontal axis) represented as a function of the position on the long vertical axis. The plots (three plots for each of the left and right hippocampi in each section) were then averaged to obtain one profile per section. To average the profiles of all the sections per animal (8-12 consecutive sections per animal), sixth-order polynomial curves were fitted to each profile per image and used to automatically estimate the mid-point of the cell layer, which in turn was used to align and average the profiles from the different sections to yield one average profile per animal. Data are shown as the mean ± SEM of 4-5 animals per sex, genotype and age.

total RNA extractions for Biochemical Analyses

Hippocampi from P1 and P45 wild-type and pallid mice were rapidly dissected, and the two halves frozen separately to be used for protein or total RNA extraction (Ghiani et al., 2010; Lee et al., 2018). Total RNA was extracted using the Invitrogen™ TRIzol™ reagent (ThermoFisher; Carlsbad, CA) following the manufacturer’s protocol from P1 wild-type, pallid and sandy, as well as P45 wild-type and pallid hippocampi. Samples were further purified by treatment with Ambion® TURBO DNA-free™ (Life Technologies; Waltham, MA), followed by a second extraction with phenol/chloroform. Sample concentrations and purity were assessed using a ThermoScientific™ NanoDrop™ One Microvolume UV-Vis Spectrophotometer (Canoga Park, CA).

Microarray Analysis

Microarray hybridization was performed at the Southern California Genotyping Consortium. Prior to hybridization, the quality of RNA was further monitored using micro-capillary electrophoresis (Bioanalizer 2100, Agilent Technologies; Santa Clara, CA). Total RNA (10 ng/ml) was amplified, labelled, and hybridized to the Illumina MouseRef-8 v2.0 expression array (Illumina; San Diego, CA) overnight at 58°C. The microarrays were then scanned using an iScan reader, and the signal compiled using BeadStudio software (Illumina). Raw microarray data were analysed using Bioconductor packages and online tools [Database for Annotation, Visualization and Integrated Discovery (DAVID) Bioinformatics Resources, http://david.abcc.ncifcrf.gov/]. Quality assessment was performed by examining the inter-array Pearson correlation and clustering based on the top variant genes. Contrast analysis of differential expression was performed by using the LIMMA (Linear Models for Microarray and RNA-Seq Data) package (Ritchie et al., 2015). After linear model fitting, a Bayesian estimate of differential expression was calculated at a false discovery rate (FDR) of 5% or less. To identify genes with differential expression in P1 hippocampus potentially due to BLOC-1 deficiency, as opposed to strain-specific effects, average mutant-to-wild-type signal ratios (in logarithmic scale base 2) were calculated for each probe yielding a strong signal in at least one genotype (quartile >0.66) in both BLOC-1-deficient mutants pallid and sandy (relative to their matched wild-type controls), and only those probes resulting in a greater than 50% signal increase (logarithm base 2 > 0.6) in both mutants were selected for further analyses. Top enriched biological functions were inferred by means of the Gene Set Enrichment Analysis (GSEA), using the online GEne SeT AnaLysis Toolkit (WebGestalt, http://www.webgestalt.org) and the datasets comprising the relative signals obtained in each BLOC-1-deficient mutant and its wild-type control for all microarray probes yielding relatively strong signals (defined as quartile >0.66 in at least one sample), using standard parameters (5-2000 genes per category, Benjamini-Hochberg procedure to control for multiple comparison).