Main

cln3 was recently identified as the gene defective in juvenile Batten disease (1). This is an inherited neurodegenerative disease of childhood in which neurons die by a process of apoptosis or programmed cell death (2). The cln3 gene encodes a 438-amino acid protein (1). Recently, it has been determined that overexpression of the cln3 protein increased survival of human NT2 cells in the face of serum withdrawal and treatment with proapoptotic drugs but not ceramide. Measurement of ceramide in affected brain increased 40–200% (3). Also, when cln3 is overexpressed, both endogenous and ceramide levels decrease, suggesting cln3 acts upstream of ceramide and hence impacts ceramide generation (3). We had also determined that Bcl-2 was up-regulated in the brains of patients affected with the juvenile form of Batten disease both at the protein and RNA level. When double immunolabeling experiments were performed and apoptotic and/or Bcl-2 overexpressing neurons were counted, it was determined that surviving neurons were overexpressing Bcl-2 but apoptotic neurons were not. This suggests that the cln3 gene may play a role in neuronal survival by suppressing cell death.

The brain is a complex organ consisting of a diverse population of neuronal cells (4). There are at least 50% more neurons formed than will actually make up the content of the adult brain (5). Excess cells die by a process of programmed cell death or apoptosis (6–10). Naturally occurring or programmed cell death occurs in almost all cell populations in the vertebrate nervous system in the brain. There is little understanding of how cell number and diversity may be controlled, particularly at the genetic level. Neuronal cells have to achieve a complex balance of cell growth and cell death. This balance is inherent within cells and is controlled by many elements including regulatory genes and trophic factors (11). We hypothesized that cln3 may be present in developing neurons and may play an important role in this balance of cell proliferation and death in neuronal development.

In the present study, we characterize the expression of cln3 in two models of development. We describe temporal cln3 expression across development of NT2 neuronal precursor cells into mature nondividing neurons. We also describe the expression of cln3 during neonatal rat brain development.

NT2 cells are committed CNS neuronal precursor cells (12). These are clonally derived, pluripotent human embryonal carcinoma cells. These cells differentiate in response to retinoic acid into nondividing neurons. Once mature, these cells express endogenous proteins of the neuronal cytoskeleton characteristic for postmitotic terminally differentiated neurons and are a convenient model to study human neuronal cell development. We evaluate the expression of cln3 in these cells across 7 wk of differentiation.

Development of the neonatal rat brain is predominantly a postnatal event (13). This makes it an excellent model to infer the development of the otherwise inaccessible human fetal brain. Neurons in rat brain develop during the last week of gestation. They then migrate and develop synaptic connections during the postnatal period (13, 14). The most rapid period of brain growth in the rat is during the first week of life (15). We describe the expression of cln3 in brain tissue of preterm, 1-wk, and 30-wk-old rats.

METHODS

Growth of neuronal cells (NT2 cells).

NT2 cells were grown in plastic flasks (Corning) at 37°C in 5% CO2 atmosphere in Dulbecco modified Eagle medium (GIBCO) supplemented with 10% FCS and 100 U each of penicillin and streptomycin. Cells were grown to confluence and then enzymatically harvested (trypsin EDTA for mRNA preparation). Others were fixed on coverslips with a solution containing 2% formaldehyde and 0.2% glutaraldehyde. Samples were collected at wk 0 (before induction treatment). Induction into neuronal cells was performed using a 10−5 M solution of all-trans-retinoic acid twice each week on d 2 and 5 for 4 consecutive weeks. Cells were harvested weekly for mRNA preparation. Cells plated onto coverslips during wk 0 were also treated with fresh 10−5 M retinoic acid and fixed weekly. After 4 wk, cells were replated at a lower density and grown for 2 d without retinoic acid. Cells were then differentially harvested and treated with mitotic inhibitors. Cells during these weeks were plated onto coverslips pretreated with poly-D-lysine and matrigel to allow neuronal processes to extend over the length of the coverslips. Cells were prepared weekly as above for mRNA preparation and coverslip fixation each week for a total of 7 wk.

Harvest of rat tissue.

Sprague-Dawley time-dated rats were used for these experiments. Five developmental ages were studied. Group 1 included rats delivered 1 d before their expected due date and was labeled preterm. Group 2 included rats evaluated on the day they were born and was labeled P0. Group 3 included rats evaluated on d 4 of life and was labeled P4. Group 4 included rats evaluated on d 8 of life and was labeled P8. The last group was evaluated on d 30 of life and was labeled P30. Animals were killed with pentobarbital. Brain tissue was harvested immediately and placed in liquid nitrogen or preserved in 10% formaldehyde for paraffin embedding. All animal-related procedures were in compliance with the Institutional Animal Care and Use Committee guidelines.

Immunostaining of NT2 cells.

Fixed cells were permeabilized with 1.0% Triton X-100 at 25°C and then incubated with 3% BSA in 10% PBS. For 12 h, the cells were exposed to human-cln3 antiserum at a dilution of 1:2500. A complementary section was exposed to rabbit IgG at the same concentration under the same conditions. Cells were washed and then incubated with a 1:500 dilution of biotin conjugated goat anti-rabbit IgG for 1 h. The cln3 antibody used in the present study was a polyclonal antibody raised against the peptide sequence AAHDILSHDRTSGNQSHVDP corresponding to amino acids 58–77 of the cln3 protein (Research Genetics, Huntsville, AL). This peptide sequence probably represents the first extracellular loop from the N-terminal end of the cln3 protein (3). This was followed with exposure to avidin conjugated horseradish peroxidase 1:500 for 30 min. The cells were stained with 3,3′-DAB in PBS and counterstained with hematoxylin blue. Coverslips were washed in xylene and mounted to glass slides. To show that mature neurons were present, additional coverslips were exposed to antisera to MAP2B and NFH. The dilution of each was 1:1000.

Immunostaining of rat brain tissue.

Brain tissue from preterm, term, and adult rats was preserved in 10% formaldehyde, embedded in paraffin, and sectioned into 5-μm specimens on glass slides. Just before analysis, brain sections were deparaffinized in xylene and rehydrated in graded alcohols. After blocking in 10% FBSA in PBS for 1 h at 25°C, the sections were incubated in anti-human cln3 antibody 1:100 in PBS containing 5% FBSA for 15 h at 25°C. A complementary set of slides was exposed to rabbit IgG at the same concentration and under the same conditions. Cells were washed and then incubated with a 1:300 dilution of biotin conjugated anti-rabbit IgG for 1 h. Next, specimens were exposed to conjugated horseradish peroxidase-streptavidin in PBS for 30 min. Tissues were color developed with DAB at 1 mg/mL in PBS containing H2O2. The sections were then counterstained with hematoxylin blue. Tissues were gradually dehydrated, placed in xylene at 25°C for 1 h, and then mounted with glass coverslips for viewing with a light microscope.

RT-PCR of neuronal cells and rat brain tissue.

Cells were harvested on ice before differentiation and each week during the 7-wk period. Frozen brain tissue was ground under liquid nitrogen. mRNA was isolated from the cells or tissue by the oligo-dT-cellulose method with the Quick Prep Micros mRNA Purification kit (Pharmacia Fine Chemicals, Piscataway, NJ). The mRNA was converted to cDNA by the RT reaction (16). For each sample, one fourth of the RNA obtained was used for the reaction. The RT reaction was performed in buffer containing 50 mM Tris-HCl, pH 8.3, 40 mM KCl, 6 mM MgCl2, 1 mM DTT, 0.5 mM each of dATP, dCTP, dGTP, and dTTP, 16 μM random primers (hexamers), and 20 U Rnasin (Promega Corp., Madison, WI) in a final volume of 30 μL. After addition of 1 μL of Superscript reverse transcriptase (Life Technologies, Gaithersburg, MD), the sample was incubated at 20°C for 10 min followed by 50 min at 42°C. The reaction was terminated by adding 70 μL of diethylpyrocarbonate water and incubating at 94°C for 5 min. The PCR reaction was set up with 10 μL of the first strand cDNA, 1 U of Taq polymerase (Perkin Elmer Biosystems, Foster City, CA), and 5 μCi of 32P-dCTP in each reaction. The reactions were performed in Taq buffer (Perkin Elmer Biosystems, Foster City, CA) containing 1.5 mM MgCl2, 2.5 mM dCTP, and 5 mM each of dATP, dGTP, and dTTP. A homologue of cln3 for rat was developed (Pane and Boustany, unpublished) with degenerate primers based on known sequences from human, mouse, and dog (1, 17, 18). The reaction conditions used were 1 min at 94°C, 1 min at 50°C, and 2 min at 72°C for 20 cycles. The PCR amplified products were analyzed on an 8% nondenaturing acrylamide gel. The gel was dried and visualized by autoradiography. The amplified signals were quantified using a PhosphorImager (Molecular Dynamics Inc., Sunnyvale, CA).

RESULTS

Differentiation of NT2 cells.

NT2 neuronal precursor cells were induced with retinoic acid to develop into mature nondividing neuronal cells over a 7-wk period. Cells with neuronal morphology were first observed in cultures 12 to 14 d after the first exposure to retinoic acid. During the period of differentiation, it was noted that the size of the cell bodies increased. The cells developed extensive axonal and dendritic processes during differentiation. After 2 wk of treatment with mitotic inhibitors, 95% of the cells were differentiated neurons. The NT2 precursor cells are shown in Figure 1A. The fully mature neuronal cells are shown in Figure 1B. To show maturation into mature neurons, these cells were stained with NFH and MAP2B. MAP2B is a high molecular weight protein that promotes nucleation and elongation of microtubules. It is found in the dendrites of differentiating and mature neurons (19). NFH is a triplet protein expressed in mature neurons. It is a prominent cytoskeletal element found most abundantly in large myelinated axons (20). The staining was prominent in neuritic processes. Mature neurons during wk 7 showed the greatest amount of NFH staining (Fig. 2). The expression of MAP2B is shown in mature neuronal cells in Figure 3. Expression of MAP2B and NFH in fully differentiated neurons is consistent with previous reports.

Figure 1
figure 1

Differentiation of NT2 cells. NT2 cells (A) were induced with retinoic acid to develop into mature nondividing postmitotic neurons (B) over a 7-wk period. DAB, ×150.

Full size image
Figure 2
figure 2

Differentiated NT2 cells express NFH. NT2 cells incubated with IgG (A) stain light blue. NT2 cells before differentiation do not stain for the cytoskeletal protein NFH (B). Postmitotic wk-7 neurons incubated with IgG (C) and those incubated with and staining prominently for NFH (D) are shown. DAB, ×150.

Full size image
Figure 3
figure 3

Differentiated NT2 cells express MAP2B. NT2 cells incubated with IgG are (A) stained light blue. NT2 cells before differentiation stain lightly in the cytoplasm for the cytoskeletal protein MAP2B (B). Postmitotic wk-7 neurons incubated with IgG (C) and MAP2B highlighting dendritic processes (D) are shown. DAB, ×150.

Full size image

cln3 expression in NT2 cells.

cln3 is expressed in the neuronal precursor cells across the 7 wk of development (Fig. 4). Relative mRNA expression of cln3 in wk-6 cells is increased over 1200 times that of cln3 expressed in previous weeks, then returns sharply to predifferentiation levels during wk 7. This increase is confirmed by immunostaining of these cells with anti-human cln3. Staining is predominantly seen within the nuclei of the cells. The most intense staining is noted in wk 6 [slides were examined blind by multiple observers and all fields were scanned (Fig. 5)]. The degree of protein overexpression does not entirely correlate with RNA expression as seen by RT-PCR in Figure 4. RT-PCR is a much more sensitive measure than immunocytochemistry. There are also other factors that influence degree of detection by immunocytochemistry such as variable regulation of protein degradation across neuronal differentiation of a specific protein and/or compartmental localization. cln3 is differentially expressed over time.

Figure 4
figure 4

Relative expression of cln3 by RT-PCR in NT2 cells during a 7-wk period of differentiation. Expression of cln3 in untreated NT2 cells is represented in wk 0. Expression of cln3 in cells treated twice weekly with retinoic acid (10−5 M) and mitotic inhibitors is represented in wk 1, 2, 3, and 4–5. cln3 expression in mature neurons is shown in wk 6 and 7. The greatest expression of cln3 is noted in wk 6 before returning to predifferentiated levels in wk 7. Expression of cln3 is compared with standard expression of human cyclophilin in each sample. Results were performed in triplicate and are reported as mean ± SEM. ★p< 0.05. au, arbitrary units.

Full size image
Figure 5
figure 5

Expression of cln3 in NT2 cells by immunohistochemistry over a 7-wk period of differentiation. A, NT2 cells (wk 0) incubated with IgG. B, NT2 cells (wk 0) incubated with cln3. NT2 cells treated with retinoic acid during wk 1, 2, and 3, incubated with IgG, are shown in (C), (E), and (G), respectively. NT2 cells treated with retinoic acid during wk 1, 2, and 3, incubated with cln3, are shown in (D), (F), and (H). NT2 cells treated with retinoic acid and mitotic inhibitors during wk 4 to 5, incubated with IgG, are shown in (I). NT2 cells treated with retinoic acid and mitotic inhibitors during wk 4 to 5, incubated with cln3, are shown in (J). Mature neurons during wk 6 and 7, incubated with IgG, are shown in (K) and (M). Mature neurons during wk 6 and 7, incubated with cln3, are shown in (L) and (N). DAB, ×100.

Full size image

cln3 expression in neonatal rat brain.

Immunostaining of tissue samples from rat brain cortex is shown in Figure 6. cln3 expression is detected in all samples. The most intense staining is seen in P0 brain sections. Less staining is noted in the preterm, P4, P8, and P30 rat brain specimens. This is supported by the RT-PCR results, a semiquantitative measure of RNA expression, and is shown in Figure 7. The greatest amount of cln3 mRNA is noted on P0 in which the expression is double that seen in the P4, P8, and P30 samples. There is a decrease in amount of cln3 expression in P4 and P8, with the lowest amount noted in P30.

Figure 6
figure 6

Expression of cln3 in neonatal rat brain by immunohistochemistry. Brain sections from regions of cortex sampled in P-1, P0, P4, P8, and P30 rats are shown in the left panels (A, B, C, D, and E, respectively). Sections from the same animals incubated with IgG are shown in the middle panels (F, G, H, I, and J, respectively). Sections from P-1, P0, P4, P8, and P30 rats incubated with anti-cln3 are shown in the right panels (K, L, M, N, and O, respectively). The most intense staining is noted in P0. DAB, ×150.

Full size image
Figure 7
figure 7

Relative expression of cln3 by RT-PCR during neonatal rat brain development. Expression of cln3 in P-1, P0, P4, P8, and P30 rats is compared with standard expression of rat cyclophilin in all samples. Greatest expression of cln3 is noted in P0 rat brains. Results were performed in triplicate and reported as mean ± SEM. ★p< 0.05.

Full size image

DISCUSSION

The present study describes the differential expression of cln3 during neuronal NT2 cell and rat brain development. cln3 plays a role in apoptotic pathways in neuronal NT2 cells and is predicted to play a role in human brain (2, 3). Apoptosis is known to play a major role in neuronal development (6) and is also involved in normal brain development (21).

Excess neurons formed or those making faulty synapses are programmed to die by apoptosis (5, 11, 22). In the case of the developing nervous system, 50 to 70% of neurons die late in the maturation process (5). There is increasing evidence to suggest that this is genetically programmed. This has been best studied in the nematode Caenorhabditis elegans (23). Cell death genes give this organism its distinct number of cells, morphologic shape, and function.

In this study, we describe expression of cln3 during neuronal development in two model systems. Each was chosen to reflect development in the human CNS. Human NT2 cells, an in vitro system of neuronal precursor cells, were studied as they differentiated into mature postmitotic neurons (12). We also studied expression of the rat homologue to cln3 in neonatal rat brain development in the first week of life.

A distinguishing characteristic of neurons is their complex and highly differentiated function and form that are mirrored by NT2 cells after they mature into postmitotic neurons. The maturation into nondividing neurons was confirmed by the presence of the structural proteins MAP2B and NFH in the mature cells. NFH is neuron specific and plays a major role in radial growth of axons (20). In development, NFH is first seen in human neurons at 20 wk gestational age (24). MAP2B, which is most prominent in dendritic processes (20), is first seen in human neurons at 9 wk gestational age (24).

cln3 was present in the neuronal precursor cells across 7 wk of development and as they matured into nondividing postmitotic neurons. The greatest expression of cln3 was noted during wk 6, after which levels returned to predifferentiation levels in wk 7. This implies that cln3 may play an important role in development and that wk 6 is a crucial time period for cellular differentiation. Cells during this phase may be in an active phase of the cell cycle. The sharp rise in cln3 may indicate the imminent exit of cells from the cell cycle.

Injury may occur to the brain during critical periods of development. It is conceivable that cln3 is a neuroprotective gene that is up-regulated during vulnerable periods of brain development. In fact, it has been shown that a population of neuronal cells with stable expression of cln3 may be protected from apoptosis when cln3 is overexpressed. To test this hypothesis in vivo, we isolated the homologue of human cln3 in rat. Corticogenesis in the rat begins d 18–20 of gestation and continues through the first week of life (13, 15). Apoptosis is prominent in the proliferating neuroepithelium of the rat cerebral cortex at the end of the embryonic period and early postnatal period (10). This is in contrast to the human brain in which the full complement of cortical neurons is acquired during the first half of gestation (4, 25). Despite this important difference, development in the rat brain has been compared with development in the human brain (13). During the first week of life a neonatal rat brain is comparable to that of a 20- to 24-wk human fetus (15). At the end of the first week, the neonatal rat is comparable to a 30-wk gestation infant; by the end of the second week, a 35–36-wk infant; and by the end of the third week, a term infant. The rat brain grows so rapidly that it is comparable to an adult human brain at the end of the first month. Although neurons are generated toward the end of the gestational or embryonic period in the rat, neuronal migration and synaptogenesis occur during the first week. Cell growth (15) and protein synthesis (26) peak during this period.

cln3 was noted to be expressed and to vary in expression during rat brain development. The highest level was noted on 0. A decline was noted on d 4 to levels similar to the preterm rat. Expression continued to decrease in d 8 and 30 animals. cln3 expression decreases as development proceeds. It has been shown that the number of apoptotic cells in the developing cortex of the rat is high at birth (10) and that the number of these apoptotic cells increases from birth to the first postnatal week, with a peak noted between P5 and P8 (27). Decrease in expression of genes implicated in apoptotic death in the cortex of the developing nervous system at the time of birth has been reported. Proapoptotic genes p53, p27Kip1, p16Ink4α, and bax α are colocalized in the proliferating zones of the rat cortex and are noted to be variably expressed. They are high at birth, then decrease during postnatal development (28). Bcl-2, an antiapoptotic gene, is variably expressed during development of the cortical nervous system in rat (29). Similarly to cln3, bcl-2 levels are high initially and decline during postnatal development of the cerebral cortex. We speculate that levels of cln3 decrease after neurons exit the cell cycle and the appropriate numbers of neurons migrate to their final destinations in the cerebral cortex.

Although the brain continues to develop in the human long after birth, brain maturation is subject to differential insults depending on critical gestational periods. Intraventricular hemorrhage is more likely to occur in the more premature infant (30). Although many factors may be important in development, antiapoptotic genes such as cln3 may have a specific neuroprotective role. Additionally, well-orchestrated cell loss in the developmental period is an important event. When regulation of this event is disrupted, brain malformation and pathology may ensue. Therefore, elucidation of specific pathways of apoptosis in brain development should allow identification of strategies aimed at neuroprotection, resulting in prolonged neuronal survival.