Free-floating brain sections were incubated in Perl’s solution (a 1:1 mixture of 2% potassium ferrocyanide and 2% HCl) for 30 minutes, washed in distilled water, and immersed for 15 min in 0.05% DAB in 0.1 M phosphate buffer (pH 7.4). might occur in a non-cell autonomous manner in em zi/zi /em rats. Aberrant extension of cellular processes and hypomyelination of oligodendrocytes were not due to the failure of the intrinsic program of oligodendrocytes, but rather, were caused by extrinsic factors that interrupt oligodendrocyte development, based on the facts that em zi/zi /em OPCs exhibit normal competence for proliferation and differentiation into mature oligodendrocytes in an em in vitro /em culture system. The immunohistochemical evidence showing the absence of atrn expression from the early stage of an oligodendrocyte lineage reinforced the idea of non-cell autonomous failure of oligodendrocyte development in em zi/zi /em rats. Such a non-cell autonomous defect of oligodendrocyte development may explain why the penetrance of abnormalities in oligodendrocyes is low ( em see Background section /em ). If em zi/zi /em oligodendrocytes developed under the control of an impaired intrinsic program, most of these cells should exhibit a common and synchronous abnormality. Nevertheless it should be noted that there was a small population of oligodendrocytes that were positive for atrn in the adult brain [this study, and ref. [29]]. Thus, it is also possible that atrn bears a critical BMS-1166 hydrochloride function in BMS-1166 hydrochloride some CACNB2 oligodendrocytes. Loss-of-function of atrn in this small population might be sufficient to initiate vacuolation in the em zi/zi /em rat. Alternatively, atrn may be transiently expressed in most oligodendrocytes at the time when its function is required. We might have failed to detect atrn expression in oligodendrocytes due to the rapid down-regulation of atrn in this cell lineage. A recent study demonstrated that there is a decline in the amount of plasma membrane lipid rafts in the liver and spleen cells prepared from atrn mutant mice, suggesting a cell-autonomous defect in the plasma membrane maintenance of atrn-deficient cells [44]. BMS-1166 hydrochloride Therefore, we cannot exclude the possibility that atrn transiently plays an intrinsic role in the maintenance/integrity of the plasma membrane or the myelin sheaths of oligodendrocytes. Another explanation for the low penetrance of myelin defects observed in em zi/zi /em rat is the redundancy of the biochemical pathways in oligodendrocyes in which atrn functions. Atrn and atrnl1 (attractin like-1) gene products were shown to share significant sequence similarity [45]. BMS-1166 hydrochloride A genetic study using transgenic and knock-out mice actually demonstrated that over-expression of em atrnl1 /em compensates for the loss of em atrn /em [45]. Given the similar patterns of expression of the two genes in the adult rodent CNS [45], endogenous atrnl1 might partially compensate to prevent development of a severe phenotype in the em zi/zi /em CNS. We should further note the possibility of inappropriate expression of the secreted form of atrn in the em zi/zi /em brain. In the rat brain, the em atrn /em gene is transcribed into two different mRNAs with sizes of 9.0 kb and 4.5 kb, by alternative splicing. The 9.0-kb transcript was deduced to encode membrane-type atrn, while the 4.5-kb transcript was deduced to encode the secreted type of atrn corresponding to the secreted form of the human em ATRN /em locus product [13,46]. Our current and recent studies have shown the absence of membrane-type atrn protein in em zi/zi /em brain using an antibody specific for the cytoplasmic tail of membrane-type atrn [this study and ref. [29]]; however, it remains to be determined whether an abnormal secreted form of atrn protein is generated in em zi/zi /em brain. However, we could exclude the possibility that the defect in em zi/zi /em oligodendrocytes is due to the inappropriate expression of the secreted form of atrn in the em zi/zi /em rat brain for the following three reasons. BMS-1166 hydrochloride (i) The deletion in the em atrn /em em zi /em allele has been identified at the splice donor site of the intron of the em atrn /em gene, which is expected to result in unstable transcripts. Certainly, previous Northern analysis showed a marked decrease in both secreted-type and membrane-type em atrn /em mRNAs in the em zi/zi /em brain [13]. Kuramoto em et. al. /em also mentioned the existence of faint and long multiple em atrn /em -related transcripts in em zi/zi /em brain, although it remains unclear whether or not these transcripts encode a functional protein [13]. (ii) Transgenic rescue experiments showed that membrane-type em atrn /em is responsible for the neuropathological phenotype in em zi /em / em zi /em rats, but that the secreted-type em atrn /em cannot complement this mutant phenotype [13]. This result indicated that the em zi/zi /em brain phenotype is attributable solely to the loss of membrane-type atrn, even though secreted-type atrn might be up-regulated in em zi/zi /em brain. (iii) The em mv/mv /em (myelin vacuolation) rat, a spontaneous mutant harboring a genomic deletion including exon 1 of the em atrn /em gene ( em atrn /em em mv /em ), showed no detectable expression of both.