- Email: [email protected]
Apolipoprotein D gene expression in the rat brain and light and electron microscopic immunocytochemistry of apolipoprotein D expression in the cerebellum of neonatal, immature and adult rats
Pergamon PII:
Neuroscience Vol. 90, No. 3, pp. 913–922, 1999 Copyright 䉷 1999 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306-4522/99 $20.00+0.00 S0306-4522(98)00507-7
APOLIPOPROTEIN D GENE EXPRESSION IN THE RAT BRAIN AND LIGHT AND ELECTRON MICROSCOPIC IMMUNOCYTOCHEMISTRY OF APOLIPOPROTEIN D EXPRESSION IN THE CEREBELLUM OF NEONATAL, IMMATURE AND ADULT RATS W. Y. ONG,* C. P. LAU,† S. K. LEONG,† U. KUMAR,‡ S. SURESH§ and S. C. PATEL§k *Department of Anatomy, National University of Singapore, 119260 Singapore †Department of Anatomy, Oxford University, Oxford, U.K. ‡Fraser Laboratories, McGill University, Montreal, Quebec, Canada, H3A 1A1 §Neurobiology Research Laboratory, VA Connecticut Healthcare System, 555 Willard Avenue, Newington, CT 06111, U.S.A.
Abstract—Apolipoprotein D gene and protein expression were investigated in the rat brain and cerebellum, respectively, during development. Apolipoprotein D gene expression was first observed in embryonic day 12 rat brain, with a moderate increase in apolipoprotein D messenger RNA levels towards the later part (embryonic days 15–17) of gestation. In the postnatal rat brain, a marked induction of apolipoprotein D messenger RNA occurred at postnatal day 10, with progressively higher levels of apolipoprotein D messenger RNA observed up to postnatal day 20. Somewhat lower, but none the less high, levels of apolipoprotein D messenger RNA continued to be present in brains of adult animals. In the immature cerebellum (day 3 up to one- to two-week-old rats), there were many densely labeled apolipoprotein Dimmunoreactive cells that had features of oligodendrocyte precursors. Purkinje neurons showed apolipoprotein D immunoreactivity in one- to two-week-old animals, after which there appeared to be some decrease in staining. Oligodendrocytes in the cerebella of two-week-old animals were strongly apolipoprotein D positive, with immunoreactivity declining in older animals. These results reveal a maturation-associated induction of apolipoprotein D gene expression in the rat brain, and expression of apolipoprotein D in glial (immature oligodendrocyte) cells in the immature cerebellum, followed by specific expression of apolipoprotein D in Purkinje neurons. 䉷 1999 IBRO. Published by Elsevier Science Ltd. Key words: apolipoprotein D, cerebellum, development, glia, Purkinje neuron, rat brain.
Apolipoprotein D (apoD), a member of the lipocalin superfamily of ligand transporters 5,6 and an apoprotein component of plasma high-density lipoproteins, 10 is a widely expressed protein in both neural and peripheral tissues. 2,20,24 Both human and mouse apoD are 27,000–31,000 mol. wt proteins encoded by single genes located on chromosomes 3 and 16, respectively. 5,27 ApoD has been shown to bind several small hydrophobic molecules, including cholesterol, progesterone, porphyrins and arachidonic acid. 3,8,11,13,17 This diversity of ligands transported by apoD suggests that the protein may have wide-ranging cellular functions. Several lines of evidence have suggested a role for apoD in the nervous system in degenerative as well as regenerative processes. For instance, apoD gene kTo whom correspondence should be addressed. Abbreviations: apoD, apolipoprotein D; E, embryonic day; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; NP-C, Niemann–Pick disease type C; P, postnatal day; PBS, phosphate-buffered saline. 913
expression is known to be induced to high levels, 25 and there is a remarkable 500-fold increase in the amount of apoD protein 3 in the regenerating rat sciatic nerve following a crush injury. Another example concerns the inherited neurodegenerative disorder, Niemann–Pick disease type C (NP-C), which is linked to an intracellular cholesterol transport defect that causes an abnormal accumulation and storage of cholesterol in lysosomes. 14,18 In studies on the NP-C mouse, an animal model of the human disorder, we found a pattern of alterations in apoD gene expression in the brain and deficiencies in cellular processing of apoD in cultured NP-C astrocytes. 26 The highest levels of accumulation of apoD protein were found in NP-C cerebellum, 26 which characteristically shows progressive Purkinje cell degeneration. 7 Finally, in an experimental model of excitotoxic neuronal injury, namely kainate-induced neurodegeneration in the hippocampus in rats, we found that apoD expression is induced in pyramidal neurons destined for
914
W. Y. Ong et al.
Fig. 1. Developmental expression of apoD mRNA in rat brain. RNA was extracted from the brains of rats at E12– E17 and at P1, P3, P5, P10, P15, P20 and P50, as described in the Experimental Procedures. Total RNA (20 m g) from whole brain was fractionated on 1% agarose–formaldehyde gels and transferred to Nytran membranes. The membranes were hybridized with [ 32P]dCTP-labeled rat apoD cDNA and the membranes stripped and reprobed with [ 32P]dCTP-labeled rat GAPDH cDNA as control. The 0.9-kb apoD mRNA signals (A) were quantified by densitometry and apoD gene expression determined as relative intensity of apoD/GAPDH mRNA signals (B).
subsequent cell death. 12 These observations suggest that a better understanding of the role of apoD in neural tissues could provide insights into diseaseassociated changes with this lipocalin. Thus, in the present study, we have investigated developmental changes in apoD gene expression in the rat brain and studied, using light and electron microscopic immunocytochemistry, the distribution of apoD in the developing and adult cerebellum.
adult were used for northern blot analysis. A total of 28 rats at P1, P3, P7, P14, P21 and adult ( ⬎ 50 days old) were used for light and electron microscopic immunocytochemical studies. Animals 50 days or older are referred to as “adults” in this study. All animal procedures were in accordance with the USPHS Guide for the Care and Use of Laboratory Animals. All efforts were made to minimize animal suffering, to reduce the number of animals used and to utilize alternatives to in vivo techniques.
Northern blot analysis EXPERIMENTAL PROCEDURES
Animals
Rat fetuses from timed-pregnant Sprague–Dawley rats at embryonic days (E) 12–17 and postnatal day (P) 1 rats to
RNA from the brains of embryonic (E12–E17) and postnatal (P1, P3, P5, P10, P20 and adult) rats was isolated by the acid guanidinium isothiocyanate–phenol–chloroform procedure. 4,16 Concentration and purity of the RNA were assessed by measurement of absorbance at 260 and
Fig. 2. Sections of the cerebellar cortex and white matter from three- (A) and seven day-old (B–D) rats. (A) Purkinje neurons do not show expression of apoD in three-day-old rats, but there is staining of scattered glial cells in the white matter (arrows). (B–D) Seven-day-old rats show dense immunoreactivity in the cell bodies and stem dendrites of Purkinje neurons (arrows in B), and in glial cells within the white matter (arrows in C and D). The latter were often observed to extend processes towards neighboring axons (AX), and bundles of axons which were outlined by the processes of the apoD-positive glial cells were observed in the white matter. M, molecular layer. Scale bar 80 m m (A, B), 40 mm (C), 16 mm (D).
Fig. 2.
Fig. 3.
Apolipoprotein D expression in rat brain
280 nm. Total RNA (20 m g) was electrophoresed on 1% agarose–formaldehyde gels, transferred to Nytran membranes (Schleicher and Schuell, Keene, NH, U.S.A.) and the blots hybridized for 16 h at 42⬚C with 50% formamide. The hybridization probe consisted of a 290-bp Hind III-BamH I fragment of rat apoD cDNA (pSP64-77), 25 labeled with [ 32P]dCTP using a random primer labeling kit (Gibco-BRL, Gaithesburg, MD, U.S.A.). Autoradiograms were prepared by exposing the membranes to XAR-5 film using intensifying screens (Eastman Kodak, Rochester, NY, U.S.A.) at ⫺ 70⬚C for three to four days. Hybridization signals were quantified by densitometry, as described previously. 16 Two internal controls were included to confirm equal recovery of mRNA in gels and to validate the specificity of apoD mRNA levels: (i) ribosomal bands visualized by ethidium bromide or their shadows detected using a photosensitive plate were photographed and analysed densitometrically; and (ii) apoD mRNA was compared with mRNA for GAPDH used as a control. Perfusion
A total of 28 rats at P1, P3, P7, P14, P21 and adult (⬎50 days old) were deeply anesthetized by intraperitoneal injection of chloral hydrate, and perfused through the left cardiac ventricle with either a solution of 4% paraformaldehyde in 0.1 M phosphate buffer or 4% paraformaldehyde and 0.1% glutaraldehyde in 0.1 M phosphate buffer. 12 The brains were removed, and a block consisting of the cerebellum and brainstem dissected out for immunocytochemical processing. Immunocytochemistry
Immunocytochemistry on frozen sections was carried out on blocks which had been fixed in 4% paraformaldehyde and kept in a solution of 30% sucrose in 0.1 M phosphate buffer for two days at 4⬚C. The blocks were sectioned coronally at 40 mm thickness. The sections were incubated overnight (18 h) with a rabbit polyclonal antiserum against rat apoD, whose characterization and specificity have been described previously. 2,3 Sections were washed in three changes of phosphate buffer, and incubated for 1 h at room temperature in a 1:200 dilution of biotinylated goat anti-rabbit immunoglobulin G (Vector Labs, Burlingame, CA, U.S.A.). This was followed by three changes of phosphate buffer to remove unreacted secondary antibody. Sections were then reacted for 1 h at room temperature with an avidin-biotinylated horseradish peroxidase complex. The reaction was visualized by treatment for 5 min in 0.1% 3,3 0 -diaminobenzidine tetrahydrochloride solution in phosphate buffer containing 0.05% hydrogen peroxide. The color reaction was stopped with several washes of phosphate buffer. The sections were lightly counterstained with Methyl Green before coverslipping. Control sections were incubated with phosphate-buffered saline (PBS) or preimmune rabbit serum instead of primary antibody, and showed complete absence of immunostaining. Immunocytochemistry was also carried out on 100-mm sections prepared using an Oxford Vibratome. For these purposes, blocks which were fixed in 4% paraformaldehyde and 0.1% glutaraldehdye were used. The sections were washed for 3 h in PBS to remove any traces of fixative,
917
and immersed for 1 h in a solution of 1% defatted dry milk in PBS (PBS–milk) to block non-specific binding of antibodies. They were then incubated overnight with rabbit antibodies against rat apoD. The rest of the immunocytochemical procedure was similar to that of the frozen sections. Control sections were incubated with PBS or preimmune rabbit serum and showed no immunostaining. Electron microscopy
Electron microscopy was carried out by subdissecting the apoD-immunostained sections from the two-week postnatal and adult animals, or PBS or preimmune serum-incubated control sections into smaller rectangular portions. Thin sections were obtained from the first 5 m m of the sections, mounted on copper grids coated with Formvar, and stained with uranyl acetate and lead citrate. They were viewed using a Jeol 1200 EX or a Philips CM120 electron microscope. Purkinje neurons had similar features to those described previously. 19 RESULTS
Northern blot analysis ApoD gene expression was first observed at days 12–14 in embryonic rat brain, and continued expression at similar or slightly higher levels could be seen in E17 brain (Fig. 1A, B). In postnatal brain, apoD was expressed at low levels in P1, P3 and P5 brain. Expression increased markedly, however, by P10, and high levels of apoD mRNA were found in P15 and P20 brain. Maximal expression appeared to occur around P15. The adult rat brain also showed high levels of apoD gene expression, although somewhat decreased by comparison with the marked induction observed towards the end of the second week postnatally. Light microscopic immunocytochemistry of the cerebellum Three-day-old rats (Fig. 2A). Small numbers of apoD-immunoreactive cells were observed in the cerebellar white matter. These had rounded or oval outlines, and were occasionally observed to give off thin and straight processes. They were putatively identified as oligodendrocyte precursors. Purkinje neurons in the cerebellar cortex were not immunolabeled for apoD at this age, in contrast to later time intervals. One-week-old rats (Fig. 2B–D). ApoD-positive Purkinje neurons were first observed in the cerebellar cortex at this age (Fig. 2B). In some folia, almost all the Purkinje neurons were labeled. In others, unlabeled neurons were present amongst the labeled
Fig. 3. Sections of the cerebellar cortex and white matter from two-week-old (A), three-week-old (B) and adult (C, D) rats. (A) ApoD immunoreactivity is present in the cell bodies of almost all the Purkinje neurons (arrows), and in their fine dendritic branches in the molecular layer (M) at this time. (B, C) Three-week-old and adult rats show some unlabeled Purkinje cell bodies (double arrows) amongst the labeled Purkinje cells (single arrows). There is less intense staining of the fine branches of dendrites of Purkinje neurons in the molecular layer (M) compared to the two-week-old rats. (D) Fewer apoD-positive glial cells are present in the white matter (arrows), compared to the two-week-old rats. Their processes are also less densely labeled (cf. Fig. 2C, D). Scale bar 80 mm (A, D), 130 mm (B), 160 mm (C).
918
W. Y. Ong et al.
Fig. 4.
Apolipoprotein D expression in rat brain
neurons. Reaction product was mostly confined to the cell bodies and large stem dendrites of the Purkinje neurons at this age. Neurons in the granule cell layer and other types of neurons in the cerebellar cortex were unlabeled for apoD. Larger numbers of apoD-positive glial cells were observed in the white matter, compared to the P3 rats (Fig. 2C, D). These had rounded, oval or elongated shapes, and ranged from 10 to 15 mm in diameter. They gave off thin and straight processes that could be traced to nearby axons, and bundles of axons were found in the white matter, which were outlined by the processes of apoD-positive oligodendrocyte-like cells (Fig. 2C). Two-week-old rats (Fig. 3). ApoD immunoreactivity was observed in the majority of Purkinje neurons of the cerebellum at this time (Fig. 3A). Label was observed in the cell bodies and fine dendritic branches of the Purkinje neurons (Fig. 3A), and not confined to the cell bodies and stem dendrites, unlike the one-week-old rats. Oligodendrocyte-like cells in the white matter were densely labeled for apoD, as at earlier time intervals. Three-week-old and adult rats (Fig. 3B–D). In contrast to the two-week-old rats, where most Purkinje neurons were labeled (Fig. 3A), some unlabeled Purkinje cell bodies were present amongst the labeled cell bodies at these times (Fig. 3B, C). ApoD was distributed in the fine branches of the dendrites of Purkinje neurons in the molecular layer, although the density of staining was less than in two-week-old rats. Fewer apoD-positive glial cells were present in the white matter, compared to the two-week-old rats (Fig. 3D). Their processes were also less densely labeled, and although individual axons were outlined by these processes, outlines of bundles of axons seen at earlier ages were not observed (Fig. 3D). Electron microscopy Many densely stained Purkinje neuronal cell bodies were observed in the cerebellar cortex of two-week-old and adult rats. These had similar features to Purkinje neurons described previously. 19 The nucleus was regular in outline and contained evenly dispersed fine heterochromatin (Fig. 4A). The cytoplasm contained large numbers of organelles, including mitochondria, interconnected
919
cisternae of rough endoplasmic reticulum, smooth endoplasmic reticulum and prominent Golgi apparatus. ApoD immunoreactivity was associated with free ribosomes and the rough and smooth endoplasmic reticulum, but was absent from the nucleus, the Golgi apparatus and the interior of mitochondria (Fig. 4A). Labeled dendrites of Purkinje neurons were observed in the molecular layer. Reaction product was present in the cytosol, but was not particularly concentrated at postsynaptic densities of the dendrites. The cerebellar white matter of two-week-old rats contained predominantly unmyelinated axons, while that of adult rats had mostly myelinated axons. Oligodendrocyte-like cells were frequently observed in the white matter in both the 14-day-old and adult rats (Fig. 3A, C, D). The nucleus contained evenly distributed fine heterochromatin clumps. The cytoplasm was present as a thin rim around the nucleus, except at one side, where fairly large numbers of cytoplasmic organelles, including free ribosomes, profiles of smooth endoplasmic reticulum and prominent Golgi apparatus, were present. Reaction product was associated with free ribosomes, ribosomes of the rough endoplasmic reticulum and the exterior of the smooth endoplasmic reticulum (Fig. 4B, 5), but was absent from the Golgi apparatus. Small-diameter processes containing prominent microtubules were sometimes seen emerging from the cell bodies. The oligodendrocyte-like cells were sometimes observed to extend processes towards unmyelinated axons in the 14-day-old rats, and myelinated axons in the adult rats. DISCUSSION
The present study aimed to elucidate developmental changes in apoD gene expression in the rat brain and to investigate, immunocytochemically, expression of apoD in the cerebellum. An important finding of this study is the marked induction of apoD gene expression in 10-day-old rats and its continued high level of expression in the adult rat brain. The increase in apoD mRNA levels in the second postnatal week coincides with the period of active myelination in the rodent brain and may be a reflection of enhanced apoD gene expression by oligodendroglia, a cell type that is known to express apoD. 2,23 Our previous studies have revealed a close functional link between apoD and cholesterol: (i) studies on apoD in astroglia in culture have revealed that astrocytes
Fig. 4. (A) Electron micrograph of an apoD-immunoreactive Purkinje neuron in the cerebellar cortex of a twoweek-old rat. The cytoplasm contains large numbers of organelles, including mitochondria (M), cisternae of rough endoplasmic reticulum and prominent Golgi apparatus (G). Reaction product is associated with free ribosomes and ribosomes of the rough endoplasmic reticulum (arrows), but is absent from the nucleus (N), Golgi apparatus and the interior of mitochondria. (B) Electron micrograph of apoD-immunolabeled oligodendrocyte-like cells from the cerebellar white matter of a two-week-old rat. The nucleus (N) contains evenly distributed fine heterochromatin clumps. Reaction product is associated with free ribosomes (arrows) in the cytoplasm. Scale bar 1 mm (A), 1.4 mm (B).
920
W. Y. Ong et al.
Fig. 5. Electron micrograph of an apoD-immunoreactive oligodendrocyte-like cell, within the cerebellar white matter of an adult rat. The nucleus (N) contains evenly-distributed fine heterochromatin clumps. The cytoplasm contains large numbers of apoD-immunolabeled free ribosomes and attached ribosomes of the rough endoplasmic reticulum (arrows).
synthesize and secrete apoD, and that the secreted apoD is associated with lipids including cholesterol; 15 (ii) apoD secretion from astrocytes is subject to regulation by oxysterols and apoD synthesis appears to be regulated by sterols; 26 (iii) biophysical studies with apoD purified from human breast cyst fluid have revealed binding of cholesterol to apoD; 13 (iv) NP-C mouse astrocytes have coordinate deficiencies in cholesterol homeostasis and apoD metabolism. 26 Thus, the high levels of apoD produced during the period of active myelination in rodent brain, as shown by the current results, could be important functionally for the transport of lipids such as cholesterol for myelin synthesis. Many densely labeled apoD-immunoreactive glial cells were observed in the cerebellar white matter, especially in the early neonatal and immature rats. Although there was a decline in apoD immunoreactivity in glial cells with maturation, apoD immunoreacitvity remained high, even in the adult. Electron microscopy showed that the glial cells had features of oligodendrocytes. Although glial cells in the three-day- to two-week-old rats resembled oligodendrocytes, they are more likely to be oligodendrocyte precursors, since mature oligodendro-
cytes are generally observed only after postnatal week 2 in the rat brain. 21,22 This notion is also supported by our electron microscopic study, which showed that very few myelinated axons were present in the white matter of the cerebellum before 14 days, further suggesting that the majority of the oligodendrocyte-like cells in the white matter at these time intervals were immature and not mature oligodendrocytes. It is possible that some of the glial cells in the older animals were also oligodendrocyte precursors and not mature oligodendrocytes. In addition to glial cells of the cerebellar white matter, apoD immunoreactivity was also observed in Purkinje neurons in the cerebellar cortex. ApoDpositive Purkinje neurons were observed at P7, and a continued high level of expression was observed in adult animals. This temporal profile of apoD expression in the cerebellum, as revealed by immunocytochemistry, parallels the developmental pattern of apoD gene expression in the rat brain (Fig. 1), which also shows a progressive increase in apoD mRNA levels postnatally, with the highest levels observed around P12–P15. The temporal pattern of apoD expression in Purkinje neurons is of interest in the light of synapse formation on these neurons by
Apolipoprotein D expression in rat brain
basket axons, which also occurs in increasing numbers from P10 onwards. 1 Considering the role of apoD as a transporter of lipophilic molecules, one possible function of apoD in glial cells and Purkinje neurons could be in the transport of lipophilic ligands, such as cholesterol. Cholesterol, which comprises 20% of the myelin sheath, would be a critical component required for myelin synthesis and assembly by the apoD-positive oligodendrocyte-like cells, as well as for the formation and maintenance of cell membranes on the dendrites of Purkinje neurons. In contrast to Purkinje neurons, granule cells were conspicuous by their absence of apoD immunoreactivity. This observation is of interest with regards to the selective degeneration of Purkinje neurons, but complete sparing of granule neurons in the NP-C mouse. 7 One possiblity suggested by this pattern of selective neurodegeneration is that expression of apoD in susceptible neurons, together with a second factor, e.g., a mutant NPC1 protein 9 in NP-C disease may be the necessary trigger for cell death. By the same token, the lack of apoD expression as observed in granule neurons may account for their lack of susceptibility to cell death. ApoD mRNA and apoD protein are significantly elevated in NP-C cerebellum compared to normal mouse cerebellum, 26 further implicating a possible association between increased apoD expression in the cerebellum and Purkinje cell death. In view of the fact that apoD is induced in degenerating neurons of the hippocampus as a result of kainate-induced excitotoxic injury, 12 the question arose as to whether the apoD-positive Purkinje neurons might be degenerating neurons or neurons dying as a result of programmed cell death (apoptosis). Electron microscopy, however,
921
confirmed that the apoD-positive neurons had features of viable and not apoptotic or degenerating necrotic neurons. CONCLUSION
In summary, the current results have revealed a temporal pattern of apoD gene expression in the developing rat brain, with a high level of expression coincident with the period of active myelination in rodent brain. In addition, the rat cerebellum shows apoD expression in oligodendroglial precursor cells in the early neonatal period and by mature oligodendrocytes in adult animals, as well as expression in Purkinje neurons starting in the second postnatal week. These observations suggest that apoD may play a functional role in myelination, as well as in synaptogenesis, in the cerebellum. The restricted expression of apoD in Purkinje neurons normally, and the observation that these neurons degenerate in NP-C, whereas granule neurons which do not express apoD and also escape cell death, suggest that apoD may play a role in the selective neurodegeneration observed in this genetic disorder. The current results reveal a maturation-associated induction of apoD gene expression in the rat brain and expression of apoD in glial (oligodendrocytelike) cells in the immature cerebellum, followed by specific expression of apoD in Purkinje neurons.
Acknowledgements—This work was supported by grants from the National University of Singapore (RP920367) and the National Institutes of Health (NS34339). C.P.L. is a medical student on attachment from Oxford University.
REFERENCES
1. Altman J. (1972) Postnatal development of the cerebellar cortex in the rat. II. Phases in the maturation of Purkinje cells and of the molecular layer. J. comp. Neurol. 145, 399–464. 2. Boyles J. K., Notterpek L. M., Wardell M. R. and Rall Jr S. C. (1990) Identification, characterization, and tissue distribution of apolipoprotein D in the rat. J. Lipid Res. 31, 2243–2256. 3. Boyles J. K., Notterpek L. M. and Anderson L. J. (1990) Accumulation of apolipoproteins in the regenerating and remyelinating mammalian peripheral nerve. Identification of apolipoprotein D, apolipoprotein A-IV, apolipoprotein E, and apolipoprotein A-I. J. biol. Chem. 265, 17805–17815. 4. Chomzynski P. and Saachi N. (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extraction. Analyt. Biochem. 161, 156–159. 5. Drayna D., McLean J., Wion K., Trent J., Drabkin H. A. and Lawn R. (1987) Human apolipoprotein D gene: gene sequence, chromosome localization and homology to a 2m -globulin superfamily. DNA 6, 199–204. 6. Flower D. R. (1996) The lipocalin protein family: structure and function. Biochem. J. 318, 1–14. 7. Higashi Y., Pentchev P. G., Murayama S. and Suzuki K. (1991) Pathology of Niemann–Pick type C: studies of murine mutants. In Neuropathology in Brain Research (ed. Ikuta F.), pp. 85–102. Elsevier Science, Amsterdam. 8. Lea O. A. (1988) Binding properties of progesterone-binding cyst protein, PBCP. Steroids 52, 337–338. 9. Loftus S. K., Morris J. A., Carstea E. D., Gu J. Z., Cummings C., Brown A., Ellison J., Ohno K., Rosenfield D. A., Tagle D. A., Pentchev P. G. and Pavan W. J. (1997) Murine model of Niemann–Pick C disease: mutation in a cholesterol homeostasis gene. Science 277, 232–235. 10. McConathy W. J. and Alaupovic P. (1973) Isolation and partial characterization of apolipoprotein D: a new protein moiety of the human plasma lipoprotein system. Fedn Eur. biochem. Socs Lett. 37, 178–182. 11. Morais-Cabral J. H., Atkins G. L., Sanchez L. M., Lopez-Boado Y. S., Lopez-Otin C. and Sawyer L. (1995) Arachidonic acid binds to apolipoprotein D: implications for the protein’s function. Fedn Eur. biochem. Socs Lett. 366, 53–56. 12. Ong W. Y., He Y., Suresh S. and Patel S. C. (1997) Differential expression of apolipoprotein D and apolipoprotein E in the kainic acid-lesioned rat hippocampus. Neuroscience 79, 359–367. 13. Patel R. C., Lange D., McConathy W. J., Patel Y. C. and Patel S. C. (1997) Probing the structure of the ligand binding cavity of lipocalins by fluorescence spectroscopy. Protein Eng. 10, 101–105.
922
W. Y. Ong et al.
14. Patel S. C. and Pentchev P. G. (1989) Genetic defects of lysosomal function in animals. A. Rev. Nutr. 9, 395–416. 15. Patel S. C., Asotra K., Patel Y. C., McConathy W. J., Patel R. C. and Suresh S. (1995) Astrocytes synthesize and secrete the lipophilic ligand carrier apolipoprotein D. NeuroReport 6, 653–657. 16. Patel Y. C., Liu J.-L., Warszynska A., Kent G., Papachristou D. N. and Patel S. C. (1995) Differential stimulation of somatostatin but not neuropeptide Y gene expression by quinolinic acid in cultured cortical neurons. J. Neurochem. 65, 998–1006. 17. Peitsch M. C. and Boguski M. S. (1990) Is apolipoprotein D a mammalian bilin-binding protein? New Biologist 2, 197–206. 18. Pentchev P. G., Vanier M. T., Suzuki K. and Patterson M. C. (1995) Niemann–Pick disease, type C: a cellular cholesterol lipidosis. In The Metabolic and Molecular Bases of Inherited Disease (eds Scriver C. R., Beaudet A. L., Sly W. S. and Valle D.), pp. 2625–2640. McGraw-Hill, New York. 19. Peters A., Palay S. L. and Webster H. de F. (1991) The Fine Structure of the Nervous System. Neurons and their Supporting Cells, p. 24. Oxford University Press, New York. 20. Provost P. R., Weech P. K., Tremblay N. M., Marcel Y. L. and Rassart E. (1990) Molecular characterization and differential mRNA tissue distribution of rabbit apolipoprotein D. J. Lipid Res. 31, 2057–2065. 21. Reynolds R. and Wilkin G. P. (1988) Development of macroglial cells in rat cerebellum. II. An in situ immunohistochemical study of oligodendroglial lineage from precursor to mature myelinating cell. Development 102, 409–425. 22. Reynolds R. and Hardy R. (1997) Oligodendroglial progenitors labelled with the O4 antibody persist in the adult rat cerebral cortex in vivo. J. Neurosci. Res. 47, 455–470. 23. Schaeren-Weimers N., Schaefer C., Vallenzuela D. M., Yancopoulos G. D. and Schwab M. E. (1995) Identification of new oligodendrocyte and myelin-specific genes by a differential screening approach. J. Neurochem. 65, 10–22. 24. Seguin D., Desforges M. and Rassart E. (1995) Molecular characterization and differential mRNA tissue distribution of mouse apolipoprotein D. Molec. Brain Res. 30, 242–250. 25. Spreyer P., Schaal H., Kuhn G., Rothe T., Unterbeck A., Olek K. and Muller H. W. (1990) Regeneration-associated high level expression of apolipoprotein D mRNA in endoneurial fibroblasts of peripheral nerve. Eur. molec. Biol. Org. J. 9, 2479–2484. 26. Suresh S., Yan Z., Patel R. C., Patel Y. C. and Patel S. C. (1998) Cholesterol storage in the Niemann–Pick type C mouse is associated with increased expression and defective processing of apolipoprotein D. J. Neurochem. 70, 242–251. 27. Warden C. H., Diep A., Taylor B. A. and Lusis A. J. (1992) Localization of the gene for apolipoprotein D on mouse chromosome 16. Genomics 12, 851–852. (Accepted 24 August 1998)
Neuroscience Vol. 90, No. 3, pp. 913–922, 1999 Copyright 䉷 1999 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306-4522/99 $20.00+0.00 S0306-4522(98)00507-7
APOLIPOPROTEIN D GENE EXPRESSION IN THE RAT BRAIN AND LIGHT AND ELECTRON MICROSCOPIC IMMUNOCYTOCHEMISTRY OF APOLIPOPROTEIN D EXPRESSION IN THE CEREBELLUM OF NEONATAL, IMMATURE AND ADULT RATS W. Y. ONG,* C. P. LAU,† S. K. LEONG,† U. KUMAR,‡ S. SURESH§ and S. C. PATEL§k *Department of Anatomy, National University of Singapore, 119260 Singapore †Department of Anatomy, Oxford University, Oxford, U.K. ‡Fraser Laboratories, McGill University, Montreal, Quebec, Canada, H3A 1A1 §Neurobiology Research Laboratory, VA Connecticut Healthcare System, 555 Willard Avenue, Newington, CT 06111, U.S.A.
Abstract—Apolipoprotein D gene and protein expression were investigated in the rat brain and cerebellum, respectively, during development. Apolipoprotein D gene expression was first observed in embryonic day 12 rat brain, with a moderate increase in apolipoprotein D messenger RNA levels towards the later part (embryonic days 15–17) of gestation. In the postnatal rat brain, a marked induction of apolipoprotein D messenger RNA occurred at postnatal day 10, with progressively higher levels of apolipoprotein D messenger RNA observed up to postnatal day 20. Somewhat lower, but none the less high, levels of apolipoprotein D messenger RNA continued to be present in brains of adult animals. In the immature cerebellum (day 3 up to one- to two-week-old rats), there were many densely labeled apolipoprotein Dimmunoreactive cells that had features of oligodendrocyte precursors. Purkinje neurons showed apolipoprotein D immunoreactivity in one- to two-week-old animals, after which there appeared to be some decrease in staining. Oligodendrocytes in the cerebella of two-week-old animals were strongly apolipoprotein D positive, with immunoreactivity declining in older animals. These results reveal a maturation-associated induction of apolipoprotein D gene expression in the rat brain, and expression of apolipoprotein D in glial (immature oligodendrocyte) cells in the immature cerebellum, followed by specific expression of apolipoprotein D in Purkinje neurons. 䉷 1999 IBRO. Published by Elsevier Science Ltd. Key words: apolipoprotein D, cerebellum, development, glia, Purkinje neuron, rat brain.
Apolipoprotein D (apoD), a member of the lipocalin superfamily of ligand transporters 5,6 and an apoprotein component of plasma high-density lipoproteins, 10 is a widely expressed protein in both neural and peripheral tissues. 2,20,24 Both human and mouse apoD are 27,000–31,000 mol. wt proteins encoded by single genes located on chromosomes 3 and 16, respectively. 5,27 ApoD has been shown to bind several small hydrophobic molecules, including cholesterol, progesterone, porphyrins and arachidonic acid. 3,8,11,13,17 This diversity of ligands transported by apoD suggests that the protein may have wide-ranging cellular functions. Several lines of evidence have suggested a role for apoD in the nervous system in degenerative as well as regenerative processes. For instance, apoD gene kTo whom correspondence should be addressed. Abbreviations: apoD, apolipoprotein D; E, embryonic day; GAPDH, glyceraldehyde-3-phosphate dehydrogenase; NP-C, Niemann–Pick disease type C; P, postnatal day; PBS, phosphate-buffered saline. 913
expression is known to be induced to high levels, 25 and there is a remarkable 500-fold increase in the amount of apoD protein 3 in the regenerating rat sciatic nerve following a crush injury. Another example concerns the inherited neurodegenerative disorder, Niemann–Pick disease type C (NP-C), which is linked to an intracellular cholesterol transport defect that causes an abnormal accumulation and storage of cholesterol in lysosomes. 14,18 In studies on the NP-C mouse, an animal model of the human disorder, we found a pattern of alterations in apoD gene expression in the brain and deficiencies in cellular processing of apoD in cultured NP-C astrocytes. 26 The highest levels of accumulation of apoD protein were found in NP-C cerebellum, 26 which characteristically shows progressive Purkinje cell degeneration. 7 Finally, in an experimental model of excitotoxic neuronal injury, namely kainate-induced neurodegeneration in the hippocampus in rats, we found that apoD expression is induced in pyramidal neurons destined for
914
W. Y. Ong et al.
Fig. 1. Developmental expression of apoD mRNA in rat brain. RNA was extracted from the brains of rats at E12– E17 and at P1, P3, P5, P10, P15, P20 and P50, as described in the Experimental Procedures. Total RNA (20 m g) from whole brain was fractionated on 1% agarose–formaldehyde gels and transferred to Nytran membranes. The membranes were hybridized with [ 32P]dCTP-labeled rat apoD cDNA and the membranes stripped and reprobed with [ 32P]dCTP-labeled rat GAPDH cDNA as control. The 0.9-kb apoD mRNA signals (A) were quantified by densitometry and apoD gene expression determined as relative intensity of apoD/GAPDH mRNA signals (B).
subsequent cell death. 12 These observations suggest that a better understanding of the role of apoD in neural tissues could provide insights into diseaseassociated changes with this lipocalin. Thus, in the present study, we have investigated developmental changes in apoD gene expression in the rat brain and studied, using light and electron microscopic immunocytochemistry, the distribution of apoD in the developing and adult cerebellum.
adult were used for northern blot analysis. A total of 28 rats at P1, P3, P7, P14, P21 and adult ( ⬎ 50 days old) were used for light and electron microscopic immunocytochemical studies. Animals 50 days or older are referred to as “adults” in this study. All animal procedures were in accordance with the USPHS Guide for the Care and Use of Laboratory Animals. All efforts were made to minimize animal suffering, to reduce the number of animals used and to utilize alternatives to in vivo techniques.
Northern blot analysis EXPERIMENTAL PROCEDURES
Animals
Rat fetuses from timed-pregnant Sprague–Dawley rats at embryonic days (E) 12–17 and postnatal day (P) 1 rats to
RNA from the brains of embryonic (E12–E17) and postnatal (P1, P3, P5, P10, P20 and adult) rats was isolated by the acid guanidinium isothiocyanate–phenol–chloroform procedure. 4,16 Concentration and purity of the RNA were assessed by measurement of absorbance at 260 and
Fig. 2. Sections of the cerebellar cortex and white matter from three- (A) and seven day-old (B–D) rats. (A) Purkinje neurons do not show expression of apoD in three-day-old rats, but there is staining of scattered glial cells in the white matter (arrows). (B–D) Seven-day-old rats show dense immunoreactivity in the cell bodies and stem dendrites of Purkinje neurons (arrows in B), and in glial cells within the white matter (arrows in C and D). The latter were often observed to extend processes towards neighboring axons (AX), and bundles of axons which were outlined by the processes of the apoD-positive glial cells were observed in the white matter. M, molecular layer. Scale bar 80 m m (A, B), 40 mm (C), 16 mm (D).
Fig. 2.
Fig. 3.
Apolipoprotein D expression in rat brain
280 nm. Total RNA (20 m g) was electrophoresed on 1% agarose–formaldehyde gels, transferred to Nytran membranes (Schleicher and Schuell, Keene, NH, U.S.A.) and the blots hybridized for 16 h at 42⬚C with 50% formamide. The hybridization probe consisted of a 290-bp Hind III-BamH I fragment of rat apoD cDNA (pSP64-77), 25 labeled with [ 32P]dCTP using a random primer labeling kit (Gibco-BRL, Gaithesburg, MD, U.S.A.). Autoradiograms were prepared by exposing the membranes to XAR-5 film using intensifying screens (Eastman Kodak, Rochester, NY, U.S.A.) at ⫺ 70⬚C for three to four days. Hybridization signals were quantified by densitometry, as described previously. 16 Two internal controls were included to confirm equal recovery of mRNA in gels and to validate the specificity of apoD mRNA levels: (i) ribosomal bands visualized by ethidium bromide or their shadows detected using a photosensitive plate were photographed and analysed densitometrically; and (ii) apoD mRNA was compared with mRNA for GAPDH used as a control. Perfusion
A total of 28 rats at P1, P3, P7, P14, P21 and adult (⬎50 days old) were deeply anesthetized by intraperitoneal injection of chloral hydrate, and perfused through the left cardiac ventricle with either a solution of 4% paraformaldehyde in 0.1 M phosphate buffer or 4% paraformaldehyde and 0.1% glutaraldehyde in 0.1 M phosphate buffer. 12 The brains were removed, and a block consisting of the cerebellum and brainstem dissected out for immunocytochemical processing. Immunocytochemistry
Immunocytochemistry on frozen sections was carried out on blocks which had been fixed in 4% paraformaldehyde and kept in a solution of 30% sucrose in 0.1 M phosphate buffer for two days at 4⬚C. The blocks were sectioned coronally at 40 mm thickness. The sections were incubated overnight (18 h) with a rabbit polyclonal antiserum against rat apoD, whose characterization and specificity have been described previously. 2,3 Sections were washed in three changes of phosphate buffer, and incubated for 1 h at room temperature in a 1:200 dilution of biotinylated goat anti-rabbit immunoglobulin G (Vector Labs, Burlingame, CA, U.S.A.). This was followed by three changes of phosphate buffer to remove unreacted secondary antibody. Sections were then reacted for 1 h at room temperature with an avidin-biotinylated horseradish peroxidase complex. The reaction was visualized by treatment for 5 min in 0.1% 3,3 0 -diaminobenzidine tetrahydrochloride solution in phosphate buffer containing 0.05% hydrogen peroxide. The color reaction was stopped with several washes of phosphate buffer. The sections were lightly counterstained with Methyl Green before coverslipping. Control sections were incubated with phosphate-buffered saline (PBS) or preimmune rabbit serum instead of primary antibody, and showed complete absence of immunostaining. Immunocytochemistry was also carried out on 100-mm sections prepared using an Oxford Vibratome. For these purposes, blocks which were fixed in 4% paraformaldehyde and 0.1% glutaraldehdye were used. The sections were washed for 3 h in PBS to remove any traces of fixative,
917
and immersed for 1 h in a solution of 1% defatted dry milk in PBS (PBS–milk) to block non-specific binding of antibodies. They were then incubated overnight with rabbit antibodies against rat apoD. The rest of the immunocytochemical procedure was similar to that of the frozen sections. Control sections were incubated with PBS or preimmune rabbit serum and showed no immunostaining. Electron microscopy
Electron microscopy was carried out by subdissecting the apoD-immunostained sections from the two-week postnatal and adult animals, or PBS or preimmune serum-incubated control sections into smaller rectangular portions. Thin sections were obtained from the first 5 m m of the sections, mounted on copper grids coated with Formvar, and stained with uranyl acetate and lead citrate. They were viewed using a Jeol 1200 EX or a Philips CM120 electron microscope. Purkinje neurons had similar features to those described previously. 19 RESULTS
Northern blot analysis ApoD gene expression was first observed at days 12–14 in embryonic rat brain, and continued expression at similar or slightly higher levels could be seen in E17 brain (Fig. 1A, B). In postnatal brain, apoD was expressed at low levels in P1, P3 and P5 brain. Expression increased markedly, however, by P10, and high levels of apoD mRNA were found in P15 and P20 brain. Maximal expression appeared to occur around P15. The adult rat brain also showed high levels of apoD gene expression, although somewhat decreased by comparison with the marked induction observed towards the end of the second week postnatally. Light microscopic immunocytochemistry of the cerebellum Three-day-old rats (Fig. 2A). Small numbers of apoD-immunoreactive cells were observed in the cerebellar white matter. These had rounded or oval outlines, and were occasionally observed to give off thin and straight processes. They were putatively identified as oligodendrocyte precursors. Purkinje neurons in the cerebellar cortex were not immunolabeled for apoD at this age, in contrast to later time intervals. One-week-old rats (Fig. 2B–D). ApoD-positive Purkinje neurons were first observed in the cerebellar cortex at this age (Fig. 2B). In some folia, almost all the Purkinje neurons were labeled. In others, unlabeled neurons were present amongst the labeled
Fig. 3. Sections of the cerebellar cortex and white matter from two-week-old (A), three-week-old (B) and adult (C, D) rats. (A) ApoD immunoreactivity is present in the cell bodies of almost all the Purkinje neurons (arrows), and in their fine dendritic branches in the molecular layer (M) at this time. (B, C) Three-week-old and adult rats show some unlabeled Purkinje cell bodies (double arrows) amongst the labeled Purkinje cells (single arrows). There is less intense staining of the fine branches of dendrites of Purkinje neurons in the molecular layer (M) compared to the two-week-old rats. (D) Fewer apoD-positive glial cells are present in the white matter (arrows), compared to the two-week-old rats. Their processes are also less densely labeled (cf. Fig. 2C, D). Scale bar 80 mm (A, D), 130 mm (B), 160 mm (C).
918
W. Y. Ong et al.
Fig. 4.
Apolipoprotein D expression in rat brain
neurons. Reaction product was mostly confined to the cell bodies and large stem dendrites of the Purkinje neurons at this age. Neurons in the granule cell layer and other types of neurons in the cerebellar cortex were unlabeled for apoD. Larger numbers of apoD-positive glial cells were observed in the white matter, compared to the P3 rats (Fig. 2C, D). These had rounded, oval or elongated shapes, and ranged from 10 to 15 mm in diameter. They gave off thin and straight processes that could be traced to nearby axons, and bundles of axons were found in the white matter, which were outlined by the processes of apoD-positive oligodendrocyte-like cells (Fig. 2C). Two-week-old rats (Fig. 3). ApoD immunoreactivity was observed in the majority of Purkinje neurons of the cerebellum at this time (Fig. 3A). Label was observed in the cell bodies and fine dendritic branches of the Purkinje neurons (Fig. 3A), and not confined to the cell bodies and stem dendrites, unlike the one-week-old rats. Oligodendrocyte-like cells in the white matter were densely labeled for apoD, as at earlier time intervals. Three-week-old and adult rats (Fig. 3B–D). In contrast to the two-week-old rats, where most Purkinje neurons were labeled (Fig. 3A), some unlabeled Purkinje cell bodies were present amongst the labeled cell bodies at these times (Fig. 3B, C). ApoD was distributed in the fine branches of the dendrites of Purkinje neurons in the molecular layer, although the density of staining was less than in two-week-old rats. Fewer apoD-positive glial cells were present in the white matter, compared to the two-week-old rats (Fig. 3D). Their processes were also less densely labeled, and although individual axons were outlined by these processes, outlines of bundles of axons seen at earlier ages were not observed (Fig. 3D). Electron microscopy Many densely stained Purkinje neuronal cell bodies were observed in the cerebellar cortex of two-week-old and adult rats. These had similar features to Purkinje neurons described previously. 19 The nucleus was regular in outline and contained evenly dispersed fine heterochromatin (Fig. 4A). The cytoplasm contained large numbers of organelles, including mitochondria, interconnected
919
cisternae of rough endoplasmic reticulum, smooth endoplasmic reticulum and prominent Golgi apparatus. ApoD immunoreactivity was associated with free ribosomes and the rough and smooth endoplasmic reticulum, but was absent from the nucleus, the Golgi apparatus and the interior of mitochondria (Fig. 4A). Labeled dendrites of Purkinje neurons were observed in the molecular layer. Reaction product was present in the cytosol, but was not particularly concentrated at postsynaptic densities of the dendrites. The cerebellar white matter of two-week-old rats contained predominantly unmyelinated axons, while that of adult rats had mostly myelinated axons. Oligodendrocyte-like cells were frequently observed in the white matter in both the 14-day-old and adult rats (Fig. 3A, C, D). The nucleus contained evenly distributed fine heterochromatin clumps. The cytoplasm was present as a thin rim around the nucleus, except at one side, where fairly large numbers of cytoplasmic organelles, including free ribosomes, profiles of smooth endoplasmic reticulum and prominent Golgi apparatus, were present. Reaction product was associated with free ribosomes, ribosomes of the rough endoplasmic reticulum and the exterior of the smooth endoplasmic reticulum (Fig. 4B, 5), but was absent from the Golgi apparatus. Small-diameter processes containing prominent microtubules were sometimes seen emerging from the cell bodies. The oligodendrocyte-like cells were sometimes observed to extend processes towards unmyelinated axons in the 14-day-old rats, and myelinated axons in the adult rats. DISCUSSION
The present study aimed to elucidate developmental changes in apoD gene expression in the rat brain and to investigate, immunocytochemically, expression of apoD in the cerebellum. An important finding of this study is the marked induction of apoD gene expression in 10-day-old rats and its continued high level of expression in the adult rat brain. The increase in apoD mRNA levels in the second postnatal week coincides with the period of active myelination in the rodent brain and may be a reflection of enhanced apoD gene expression by oligodendroglia, a cell type that is known to express apoD. 2,23 Our previous studies have revealed a close functional link between apoD and cholesterol: (i) studies on apoD in astroglia in culture have revealed that astrocytes
Fig. 4. (A) Electron micrograph of an apoD-immunoreactive Purkinje neuron in the cerebellar cortex of a twoweek-old rat. The cytoplasm contains large numbers of organelles, including mitochondria (M), cisternae of rough endoplasmic reticulum and prominent Golgi apparatus (G). Reaction product is associated with free ribosomes and ribosomes of the rough endoplasmic reticulum (arrows), but is absent from the nucleus (N), Golgi apparatus and the interior of mitochondria. (B) Electron micrograph of apoD-immunolabeled oligodendrocyte-like cells from the cerebellar white matter of a two-week-old rat. The nucleus (N) contains evenly distributed fine heterochromatin clumps. Reaction product is associated with free ribosomes (arrows) in the cytoplasm. Scale bar 1 mm (A), 1.4 mm (B).
920
W. Y. Ong et al.
Fig. 5. Electron micrograph of an apoD-immunoreactive oligodendrocyte-like cell, within the cerebellar white matter of an adult rat. The nucleus (N) contains evenly-distributed fine heterochromatin clumps. The cytoplasm contains large numbers of apoD-immunolabeled free ribosomes and attached ribosomes of the rough endoplasmic reticulum (arrows).
synthesize and secrete apoD, and that the secreted apoD is associated with lipids including cholesterol; 15 (ii) apoD secretion from astrocytes is subject to regulation by oxysterols and apoD synthesis appears to be regulated by sterols; 26 (iii) biophysical studies with apoD purified from human breast cyst fluid have revealed binding of cholesterol to apoD; 13 (iv) NP-C mouse astrocytes have coordinate deficiencies in cholesterol homeostasis and apoD metabolism. 26 Thus, the high levels of apoD produced during the period of active myelination in rodent brain, as shown by the current results, could be important functionally for the transport of lipids such as cholesterol for myelin synthesis. Many densely labeled apoD-immunoreactive glial cells were observed in the cerebellar white matter, especially in the early neonatal and immature rats. Although there was a decline in apoD immunoreactivity in glial cells with maturation, apoD immunoreacitvity remained high, even in the adult. Electron microscopy showed that the glial cells had features of oligodendrocytes. Although glial cells in the three-day- to two-week-old rats resembled oligodendrocytes, they are more likely to be oligodendrocyte precursors, since mature oligodendro-
cytes are generally observed only after postnatal week 2 in the rat brain. 21,22 This notion is also supported by our electron microscopic study, which showed that very few myelinated axons were present in the white matter of the cerebellum before 14 days, further suggesting that the majority of the oligodendrocyte-like cells in the white matter at these time intervals were immature and not mature oligodendrocytes. It is possible that some of the glial cells in the older animals were also oligodendrocyte precursors and not mature oligodendrocytes. In addition to glial cells of the cerebellar white matter, apoD immunoreactivity was also observed in Purkinje neurons in the cerebellar cortex. ApoDpositive Purkinje neurons were observed at P7, and a continued high level of expression was observed in adult animals. This temporal profile of apoD expression in the cerebellum, as revealed by immunocytochemistry, parallels the developmental pattern of apoD gene expression in the rat brain (Fig. 1), which also shows a progressive increase in apoD mRNA levels postnatally, with the highest levels observed around P12–P15. The temporal pattern of apoD expression in Purkinje neurons is of interest in the light of synapse formation on these neurons by
Apolipoprotein D expression in rat brain
basket axons, which also occurs in increasing numbers from P10 onwards. 1 Considering the role of apoD as a transporter of lipophilic molecules, one possible function of apoD in glial cells and Purkinje neurons could be in the transport of lipophilic ligands, such as cholesterol. Cholesterol, which comprises 20% of the myelin sheath, would be a critical component required for myelin synthesis and assembly by the apoD-positive oligodendrocyte-like cells, as well as for the formation and maintenance of cell membranes on the dendrites of Purkinje neurons. In contrast to Purkinje neurons, granule cells were conspicuous by their absence of apoD immunoreactivity. This observation is of interest with regards to the selective degeneration of Purkinje neurons, but complete sparing of granule neurons in the NP-C mouse. 7 One possiblity suggested by this pattern of selective neurodegeneration is that expression of apoD in susceptible neurons, together with a second factor, e.g., a mutant NPC1 protein 9 in NP-C disease may be the necessary trigger for cell death. By the same token, the lack of apoD expression as observed in granule neurons may account for their lack of susceptibility to cell death. ApoD mRNA and apoD protein are significantly elevated in NP-C cerebellum compared to normal mouse cerebellum, 26 further implicating a possible association between increased apoD expression in the cerebellum and Purkinje cell death. In view of the fact that apoD is induced in degenerating neurons of the hippocampus as a result of kainate-induced excitotoxic injury, 12 the question arose as to whether the apoD-positive Purkinje neurons might be degenerating neurons or neurons dying as a result of programmed cell death (apoptosis). Electron microscopy, however,
921
confirmed that the apoD-positive neurons had features of viable and not apoptotic or degenerating necrotic neurons. CONCLUSION
In summary, the current results have revealed a temporal pattern of apoD gene expression in the developing rat brain, with a high level of expression coincident with the period of active myelination in rodent brain. In addition, the rat cerebellum shows apoD expression in oligodendroglial precursor cells in the early neonatal period and by mature oligodendrocytes in adult animals, as well as expression in Purkinje neurons starting in the second postnatal week. These observations suggest that apoD may play a functional role in myelination, as well as in synaptogenesis, in the cerebellum. The restricted expression of apoD in Purkinje neurons normally, and the observation that these neurons degenerate in NP-C, whereas granule neurons which do not express apoD and also escape cell death, suggest that apoD may play a role in the selective neurodegeneration observed in this genetic disorder. The current results reveal a maturation-associated induction of apoD gene expression in the rat brain and expression of apoD in glial (oligodendrocytelike) cells in the immature cerebellum, followed by specific expression of apoD in Purkinje neurons.
Acknowledgements—This work was supported by grants from the National University of Singapore (RP920367) and the National Institutes of Health (NS34339). C.P.L. is a medical student on attachment from Oxford University.
REFERENCES
1. Altman J. (1972) Postnatal development of the cerebellar cortex in the rat. II. Phases in the maturation of Purkinje cells and of the molecular layer. J. comp. Neurol. 145, 399–464. 2. Boyles J. K., Notterpek L. M., Wardell M. R. and Rall Jr S. C. (1990) Identification, characterization, and tissue distribution of apolipoprotein D in the rat. J. Lipid Res. 31, 2243–2256. 3. Boyles J. K., Notterpek L. M. and Anderson L. J. (1990) Accumulation of apolipoproteins in the regenerating and remyelinating mammalian peripheral nerve. Identification of apolipoprotein D, apolipoprotein A-IV, apolipoprotein E, and apolipoprotein A-I. J. biol. Chem. 265, 17805–17815. 4. Chomzynski P. and Saachi N. (1987) Single-step method of RNA isolation by acid guanidinium thiocyanate–phenol–chloroform extraction. Analyt. Biochem. 161, 156–159. 5. Drayna D., McLean J., Wion K., Trent J., Drabkin H. A. and Lawn R. (1987) Human apolipoprotein D gene: gene sequence, chromosome localization and homology to a 2m -globulin superfamily. DNA 6, 199–204. 6. Flower D. R. (1996) The lipocalin protein family: structure and function. Biochem. J. 318, 1–14. 7. Higashi Y., Pentchev P. G., Murayama S. and Suzuki K. (1991) Pathology of Niemann–Pick type C: studies of murine mutants. In Neuropathology in Brain Research (ed. Ikuta F.), pp. 85–102. Elsevier Science, Amsterdam. 8. Lea O. A. (1988) Binding properties of progesterone-binding cyst protein, PBCP. Steroids 52, 337–338. 9. Loftus S. K., Morris J. A., Carstea E. D., Gu J. Z., Cummings C., Brown A., Ellison J., Ohno K., Rosenfield D. A., Tagle D. A., Pentchev P. G. and Pavan W. J. (1997) Murine model of Niemann–Pick C disease: mutation in a cholesterol homeostasis gene. Science 277, 232–235. 10. McConathy W. J. and Alaupovic P. (1973) Isolation and partial characterization of apolipoprotein D: a new protein moiety of the human plasma lipoprotein system. Fedn Eur. biochem. Socs Lett. 37, 178–182. 11. Morais-Cabral J. H., Atkins G. L., Sanchez L. M., Lopez-Boado Y. S., Lopez-Otin C. and Sawyer L. (1995) Arachidonic acid binds to apolipoprotein D: implications for the protein’s function. Fedn Eur. biochem. Socs Lett. 366, 53–56. 12. Ong W. Y., He Y., Suresh S. and Patel S. C. (1997) Differential expression of apolipoprotein D and apolipoprotein E in the kainic acid-lesioned rat hippocampus. Neuroscience 79, 359–367. 13. Patel R. C., Lange D., McConathy W. J., Patel Y. C. and Patel S. C. (1997) Probing the structure of the ligand binding cavity of lipocalins by fluorescence spectroscopy. Protein Eng. 10, 101–105.
922
W. Y. Ong et al.
14. Patel S. C. and Pentchev P. G. (1989) Genetic defects of lysosomal function in animals. A. Rev. Nutr. 9, 395–416. 15. Patel S. C., Asotra K., Patel Y. C., McConathy W. J., Patel R. C. and Suresh S. (1995) Astrocytes synthesize and secrete the lipophilic ligand carrier apolipoprotein D. NeuroReport 6, 653–657. 16. Patel Y. C., Liu J.-L., Warszynska A., Kent G., Papachristou D. N. and Patel S. C. (1995) Differential stimulation of somatostatin but not neuropeptide Y gene expression by quinolinic acid in cultured cortical neurons. J. Neurochem. 65, 998–1006. 17. Peitsch M. C. and Boguski M. S. (1990) Is apolipoprotein D a mammalian bilin-binding protein? New Biologist 2, 197–206. 18. Pentchev P. G., Vanier M. T., Suzuki K. and Patterson M. C. (1995) Niemann–Pick disease, type C: a cellular cholesterol lipidosis. In The Metabolic and Molecular Bases of Inherited Disease (eds Scriver C. R., Beaudet A. L., Sly W. S. and Valle D.), pp. 2625–2640. McGraw-Hill, New York. 19. Peters A., Palay S. L. and Webster H. de F. (1991) The Fine Structure of the Nervous System. Neurons and their Supporting Cells, p. 24. Oxford University Press, New York. 20. Provost P. R., Weech P. K., Tremblay N. M., Marcel Y. L. and Rassart E. (1990) Molecular characterization and differential mRNA tissue distribution of rabbit apolipoprotein D. J. Lipid Res. 31, 2057–2065. 21. Reynolds R. and Wilkin G. P. (1988) Development of macroglial cells in rat cerebellum. II. An in situ immunohistochemical study of oligodendroglial lineage from precursor to mature myelinating cell. Development 102, 409–425. 22. Reynolds R. and Hardy R. (1997) Oligodendroglial progenitors labelled with the O4 antibody persist in the adult rat cerebral cortex in vivo. J. Neurosci. Res. 47, 455–470. 23. Schaeren-Weimers N., Schaefer C., Vallenzuela D. M., Yancopoulos G. D. and Schwab M. E. (1995) Identification of new oligodendrocyte and myelin-specific genes by a differential screening approach. J. Neurochem. 65, 10–22. 24. Seguin D., Desforges M. and Rassart E. (1995) Molecular characterization and differential mRNA tissue distribution of mouse apolipoprotein D. Molec. Brain Res. 30, 242–250. 25. Spreyer P., Schaal H., Kuhn G., Rothe T., Unterbeck A., Olek K. and Muller H. W. (1990) Regeneration-associated high level expression of apolipoprotein D mRNA in endoneurial fibroblasts of peripheral nerve. Eur. molec. Biol. Org. J. 9, 2479–2484. 26. Suresh S., Yan Z., Patel R. C., Patel Y. C. and Patel S. C. (1998) Cholesterol storage in the Niemann–Pick type C mouse is associated with increased expression and defective processing of apolipoprotein D. J. Neurochem. 70, 242–251. 27. Warden C. H., Diep A., Taylor B. A. and Lusis A. J. (1992) Localization of the gene for apolipoprotein D on mouse chromosome 16. Genomics 12, 851–852. (Accepted 24 August 1998)