Gene expression profiling of cerebellar development with high-throughput functional analysis
Sakae Saito1,
Kimi Honma2,
Hiroko Kita-Matsuo1,
Takahiro Ochiya3 and
Kikuya Kato4
1 Taisho Laboratory of Functional Genomics, Nara Institute of Science and Technology, and Core Research for Evolutional Science and Technology, Japan Science and Technology Corporation, Tokyo, Japan
2 Koken Bioscience Institute, Tokyo, Japan
3 National Cancer Center Research Institute, Tokyo, Japan
4 Research Institute, Osaka Medical Center for Cancer and Cardiovascular Diseases, Osaka, Japan
 |
ABSTRACT
|
---|
We measured the expression levels of 450 genes during mouse postnatal cerebellar development by quantitative PCR using RNA purified from layers of the cerebellar cortex. Principal component analysis of the data matrix demonstrated that the first and second components corresponded to general levels of gene expression and gene expression patterns, respectively. We introduced 288 of the 450 genes into PC12 cells using a high-throughput transfection assay based on atelocollagen and determined the ability of each gene to promote neurite outgrowth or cell proliferation. Five genes induced neurite outgrowth, and seven genes enhanced proliferation. Evaluation of the functional data and gene expression patterns showed that none of these genes exhibited elevated expression at maturation, suggesting that genes characteristic of mature neurons are not likely to participate in neuronal development. These results demonstrate that functional data can facilitate interpretation of expression profiles and identification of new molecules that participate in biological processes.
adaptor-tagged competitive polymerase chain reaction; cell transfection array
 |
INTRODUCTION
|
---|
DEVELOPMENTAL PROCESSES are executed by intrinsic programs encoded in the genome with modulation by extrinsic factors. Because the execution of such programs is, for the most part, dependent on gene expression, it should be possible to directly elucidate the programs by analyzing gene expression. Conventional approaches such as genetic studies based on mutant analysis have so far identified genes responsible for particular developmental processes. Many of these genes are transcription factors such as homeotic genes; however, many of the molecular events downstream of the transcription factors still remain to be elucidated.
Although gene expression profiling is expected to shed light upon downstream molecular events, the lack of valid information regarding gene functions makes it difficult to interpret the expression data. Complementary methods of analysis include bioinformatics of the scientific literature (24) and proteomics (7, 10). However, direct analysis of gene function is more desirable. Recently, two techniques for high-throughput cell transfection have been described: reverse transfection (33) and a cell transfection array using atelocollagen (16). These two techniques enable large-scale experiments to study the overexpression or suppression of genes.
In this report, we measured the expression levels of 450 cerebellar genes using RNA purified from each layer of the cerebellar cortex. In parallel, we analyzed 288 of the 450 genes by cell transfection array using PC12 cells. The cerebellar cortex is a classic model system in developmental biology (2). The postnatal development of the cerebellum includes most of the events characteristic of nervous system development including axon elongation, cell proliferation, cell migration, synapse formation, and apoptosis. PC12 cells are also a classic model system of neuronal differentiation (13). This cell line is derived from the rat pheochromocytoma, and the cells extend neurites under appropriate stimulation, such as treatment with nerve growth factor (NGF). Combined use of these classical systems has allowed the description of both the transcriptional and functional aspects of nervous system development, demonstrating the correlation between gene expression and function.
 |
MATERIALS AND METHODS
|
---|
Genes subjected to analysis.
We have previously performed gene expression profiling of postnatal cerebellar development using the whole cerebellum (22). This experiment analyzed 1,869 genes selected by descending order of abundance from our expressed sequence tag collection (21). Of these 1,869 genes, 768 genes, including 384 genes whose expression changed during development and 384 genes that were previously reported to be involved in brain development, were selected for the present study (22). Because RNA quantitation using samples recovered by laser capture microdissection (LCM) is demanding, we succeeded in analyzing only 450 of the 768 genes. The full-length cDNAs of 288 of these genes were subcloned into expression vectors. It should be noted that the selected genes may be biased toward abundance but should not be biased to other factors such as gene function or annotation.
LCM.
With the use of a PixCell LCM microscope (Arcturus Engineering), the granule cell layers corresponding to the distinct areas described in the main text were dissected from hematoxylin-eosin-stained 12-µm frozen tissue sections of 4-, 8-, 12-, and 21-day-old mice. Three mice were used for LCM at each age. Experimental details are described in protocols for microdissection and RNA preparation on the NIH LCM web page (http://dir.nichd.nih.gov/lcm/lcm.htm). Approximately 100 ng of total RNA were obtained from 1,000 transfer shots of the external granular layer (EGL) and the internal granular layer (IGL).
Adaptor-tagged competitive PCR and data analysis.
The adaptor-tagged competitive PCR (ATAC-PCR) protocol is essentially the same as described previously (18, 22, 30). Total RNA from the cerebrum at postnatal week 6 was used as a standard. The relative expression levels versus the 6-wk cerebrum were calculated. Among the 768 genes subjected to ATAC-PCR, 450 genes had no missing values. This data matrix was normalized as previously described (25). In brief, the data were first divided by the median of the samples and converted into a logarithmic scale. The cutoff values were set at 20 and 0.05 before the logarithmic conversion. Principal component analysis (PCA) was performed using GeneMath 2.0 (AppliedMath).
The chance of biased distribution of functional genes (P) was calculated using the hypergeometric distribution, described as follows:
where N is the number of genes in the total population, M is the number of functional genes in the total population, n is the number of genes in a particular group, and m is the number of functional genes in the group.
Culture and differentiation of PC12 cells.
PC12 cells, obtained from the American Type Culture Collection, were grown in DMEM with 10% charcoal-stripped horse serum (Biofluid), 5% dialyzed FBS (GIBCO-BRL), 10 mM HEPES (pH 7.4), 50 mg/ml streptomycin, and 50 U/ml penicillin. To differentiate PC12 cells, 1 x 105 cells were seeded on poly-L-orinithine (Sigma)-coated 25-cm2 culture flasks. The following day, cells were incubated in PC12 medium containing 100 ng/ml of 2.5S NGF (Collaborative Biomedical Products) for 1 wk. Medium was replaced every 3 days. On day 7, neurite extension was clearly observed in >90% cells. Passage 24 clones were used in all the experiments.
Atelocollagen complex and gene transfer.
The full-length open reading frames of 288 genes were amplified from 8-day-old mouse cerebellar cDNA. Because the 8-day-old mouse cerebellum contains both the EGL and the IGL, the selected genes were well amplified and subcloned into pDEST26 using Gateway Technology (Invitrogen). Each purified pDEST26x plasmid and a pCMV-EGFP plasmid were mixed with Atelocollagen (Koken) and dissolved in PBS() buffer at a fixed concentration of 80 µg/ml. Each plasmid DNA was used at a concentration of 5 µg/ml. The mixture was incubated for 20 min at 4°C. A small aliquot of the mixture, up to 50 µl, was spotted into a flat-bottom 96-well plate and precoated. PC12 cells were precultured and seeded at 2 x 104 cells/well for transfection of pDESTx and pCMV-EGFP. Green fluorescent protein (GFP) gene expression was monitored by fluorescence microscopy. The transfection efficiency calculated by counting the GFP-positive cells was 29.3 ± 4.51%. Neurite outgrowth was investigated using the Neurite Outgrowth HitKit assay (Cellomics). The signal intensity of the fluorescent stain was detected by the ArrayScan II system (Cellomics) and analyzed using Neurite Outgrowth Application software (Cellomics) under the default parameter setting. The results of the computer analysis were confirmed by microscopic observation. Cellular proliferation was investigated using the TetraColor One cell proliferation assay reagent (Seikagaku) according to the recommended protocol. The colorimetric reaction was assessed by measuring the absorption of each well at 450 nm with a reference wavelength at 650 nm and a ultraviolet microplate reader (Bio-Rad Laboratories). Most of the genes exhibited absorbances around 0.4 (absorbance 450/630 nm). Absorption values of the genes exhibiting growth promotion properties exceeded 0.7.
 |
RESULTS
|
---|
Gene expression profiling using RNA purified from granule cell layers.
In rodents, the development of the cerebellar cortex begins just after birth; the peak of granule cell proliferation occurs during the first week, and the peak of cell migration and axon elongation occurs during the second week (2). The developing granule cell layer typically consists of two sublayers: the EGL and the IGL. Furthermore, the EGL and IGL typically consist of two areas. The proliferating granule cells are located in the superficial area of the EGL (represented as EGLa in Fig. 1), whereas postmitotic and migrating granule cells are located in the internal area of the EGL (EGLb). These granule cells migrate to the IGL, extending parallel fibers. The synaptogenic granule cells are located in the IGL adjacent to the Purkinje cell layer (IGLa), and the mature granule cells are located in the internal area of the IGL (IGLb). The morphological changes in these cells are complete by the third week (2) and regarded as the mature cerebellum. The 21-day cerebellum therefore only consists of IGLb and has the structure as the same as that of an adult. Using LCM, we dissected the granule cell layers from 4-, 8-, 12-, and 21-day-old mice, collecting a total of 12 samples (4-day EGLa, 4-day EGLb, 4-day IGL, 8-day EGLa, 8-day EGLb, 8-day IGLa, 8-day IGLb, 12-day EGL, 12-day IGLa, 12-day IGLb, 21-day IGLa, and 21-day IGLb). It should be noted that we were not able to discriminate among the four layers in some samples because the fraction of EGL decreases and the fraction of IGL increases as development proceeds. The relative expression levels of the 450 genes were obtained using ATAC-PCR. ATAC-PCR is an advanced version of quantitative PCR and has been established as a mature expression profiling technique in several studies (30). The data matrix consisting of 450 genes x 12 samples is supplied as supplementary data (http://physiolgenomics.physiology.org/cgi/content/full/00142.2004/DC1).

View larger version (26K):
[in this window]
[in a new window]
|
Fig. 1. Gene expression profiling during cerebellar granule cell differentiation; 450 genes (vertical axis) are arranged in order of the factor score of the first component (A) or the second component (B) of principal component analysis (PCA). Expression levels are indicated by color, with the scale shown at bottom right. Crossbars designate genes whose overexpression resulted in phenotypic changes (B). From top to bottom, blue bars (neurite outgrowth) indicate potassium channel subfamily K member 2 (TREK-1), retinoblastoma 1 (Rb1), ring finger protein 13 (Rnf13), eukaryotic translation initiation factor 4A1 (eIF4A1), and neuronatin (Nnat), and red bars (cell proliferation) indicate hepatoma-derived growth factor (Hdgf), fibroblast growth factor-inducible gene 14 (FIN14/Ddx3x), immediate-early response gene 2 (Ier2/pip92/ETR109), mitogen-activated protein kinase 6 (Mapk6/Erk3), Eph receptor A3 (Epha3/Hek/Mek4), signal-induced proliferation-associated gene 1 (Sipa1/Spa1), Finkel-Biskis-Reilly murine sarcoma virus (FBR-MuSV) ubiquitously expressed gene (Fau/MNSFß). EGLa, superficial area of the external granular layer (EGL); EGLb, internal area of the EGL; IGLa, internal granular layer (IGL) adjacent to the Purkinje cell layer; IGLb, internal area of the IGL.
|
|
We applied several unsupervised feature extraction methods to the data matrix. PCA, a statistical technique to summarize a multivariate data set using a few components, was successful. This analysis demonstrated that the first and second components corresponded to general levels of gene expression and to gene expression patterns, respectively (Fig. 1, A and B). Sorting of genes by the factor scores of the second component ordered the genes from those elevated during development to those elevated at maturation. In contrast to the prominent temporal changes, differences between the layers were obscure.
Cell transfection array.
Postnatal cerebellar development includes cell proliferation and axon elongation. Genes involved in these processes are expected to induce phenotypic changes in PC12 cells; 288 of the 450 cerebellar genes that were subjected to gene expression analysis were also transfected into PC12 cells. To enable high-throughput functional analysis, we employed a cell transfection array using atelocollagen. Atelocollagen is a biocompatible natural polymer that is free of telopeptides and is obtained by pepsin treatment of highly purified type I collagen. Atelocollagen can be used to condense and deliver DNA, antisense oligodeoxynucleotides, virus vectors, and siRNAs into cells (27, 28).
Cells transfected with five genes, including neuronatin (Nnat), retinoblastoma 1 (Rb1), potassium channel subfamily K member 2 (TREK-1), ring finger protein 13 (Rnf13), and eukaryotic translation initiation factor 4A1 (eIF4A1), exhibited spontaneous neurite outgrowth. The morphologies of these transfected cells are shown in Fig. 2. Transfection of seven genes promoted cell growth. These growth-promoting genes were fibroblast growth factor (FGF)-inducible gene 14 (Fin14/Ddx3x), immediate-early response 2 gene (Ier2/pip92/ETR101), signal-induced proliferation-associated gene 1 (Sipa1/Spa1), Eph receptor A3 (Epha3/Hek/Mek4), hepatoma-derived growth factor (Hdgf), mitogen-activated protein kinase 6 (Mapk6/Erk3), and Finkel-Biskis-Reilly murine sarcoma virus (FBR-MuSV) ubiquitously expressed gene (Fau/MNSFß).

View larger version (46K):
[in this window]
[in a new window]
|
Fig. 2. Cell transfection array. PC12 cells transfected with enhanced green fluorescent protein alone (A and B) and PC12 cells differentiated into neurons by treatment with 100 ng/ml neuron growth factor (NGF; C). Fluorescent micrographs show neurite outgrowth of cells that overexpressed Nnat (D), Rb1 (E), TREK-1 (F), Rnf13 (G), and eIF4A1 (H) in the absence of NGF.
|
|
Correlation between gene expression patterns and functions.
To explore the relationship between gene expression and function, we aligned genes in the functional categories identified by the cell transfection array to the gene expression patterns sorted by the second component (Fig. 1B). The genes that functioned in neurite outgrowth and cell proliferation had characteristic distribution; none of these genes exhibited elevated expression at maturation. Figure 1B includes genes that were not analyzed in transfection experiments. In Fig. 3, the expression patterns of the 288 genes analyzed by transfection are shown in a more quantitative manner: the quantitative differences between the expression levels in 8-day and 12-day/3-week cerebella are plotted by the order of factor scores of the second component. There are no genes that functioned either in neurite outgrowth or cell proliferation distributed in the region where the indicator exhibited negative values.

View larger version (22K):
[in this window]
[in a new window]
|
Fig. 3. The characteristics of the expression-function correlation. Vertical axis, the log-transformed ratio of the mean value of expression levels in 8-day cerebella to the mean value of expression levels in 12-day and 3-wk cerebella. Horizontal axis, the 288 genes subjected to transfection were sorted by the factor score of the second component of PCA. Crossbars designate genes whose overexpression resulted in phenotypic changes. From left to right, blue bars (neurite outgrowth) indicate TREK-1, Rb1, Rnf13, eIF4A1, and Nnat, and red bars (cell proliferation) indicate Hdgf, FIN14/Ddx3x, Ier2/pip92/ETR109, Mapk6/Erk3, Epha3/Hek/Mek4, Sipa1/Spa1, and Fau/MNSFß.
|
|
We calculated the chance probability of the biased gene distribution. Because Nnat-3 is located nearest to the right end, we calculated the probability that all genes of a category are located on the left side of the Nnat-3 position. The P values for all genes, cell proliferation genes, and neurite outgrowth genes are 0.002, 0.028, and 0.267, respectively.
 |
DISCUSSION
|
---|
Identification of transcriptional activation or deactivation of functional gene classes would facilitate the understanding of molecular mechanisms underlying biological processes. Exploration of the correlations between gene expression patterns and functions has therefore been a central issue of transcriptome analysis. However, these analyses have usually been performed by subjective interpretation of individual scientists without valid evidence. Such arguments have often been based on previously published scientific literature rather than on information derived from the analysis of experimental systems. Experimental data regarding gene functions should provide more valid information. In this report, we described the simultaneous analysis of gene expression and function to test whether an expression-function correlation could be experimentally identified. One important finding herein is that none of the genes in the functional groups exhibited elevated expression at maturation. This result agrees with the gene expression-function correlation previously identified from the literature (20, 22). Taken together, these data imply that molecules that are functionally active in mature neurons are not likely to be active participants in the developmental process. Although confirmation with a larger number of genes is needed, our approach will provide a rigid basis for future studies of gene expression-function correlations.
The rat pheochromocytoma cell line PC12 is a well-characterized model for the study of neuron-like differentiation and signal transduction. There are many cases where genes involved in cerebellar development are induced phenotypic changes in PC12 cells. In response to expression of Wnt-1, a secreted signaling factor required for development of mammalian cerebellum, PC12 cells adopt a flat morphology reminiscent of epithelial cells, form extensive cell-cell contacts, and become refractory to neuronal differentiation induced by NGF (31). Moreover, NCAM, a member of the immunoglobulin superfamily implicated in cerebellar granule cell migration, has stimulated neurite outgrowth from PC12 cells (8). Nnat protein is expressed much more highly in fetal and neonatal brains than in the adult brain (9) and is expressed in a segment-specific pattern during early hindbrain development (32), suggesting involvement in nervous system development. Rb1 is known to be a tumor suppresser gene and is hypophosphorylated in PC12 cells by NGF through the Ras signaling pathway (19). Hypophosphorylated Rb1 is essential for neurite outgrowth, as was demonstrated by inhibition with a monoclonal antibody against hypophosphorylated Rb (19). The related protein Rb2/p130 was reported to induce neurite outgrowth in the absence of NGF (29). Including our results, there is enough evidence to support the involvement of Rb1 in differentiation of PC12 cells. However, the role of Rb1 in cerebellar development is unknown. There have been very few reports of other genes that induced neurite outgrowth. Our findings should be a starting point for elucidating the roles of these genes in this process.
Most of the genes we identified as promoting cell growth have been described in the literature as having similar functions. Hdgf was found to have activity in hepatocytes (26) but may also be involved in nervous system development (1). Fin14/Ddx3x was initially identified as a FGF-inducible gene (14). Ier2/pip92/ETR101 was identified in FGF-stimulated hippocampal cells during neuronal differentiation (5) and in PC12 cells treated with cell proliferation agents, neuronal differentiation agents, and membrane depolarization agents (3). Sipa1/Spa1 is transcriptionally induced in murine lymphoid cells after mitogenic stimulation (15). The expression pattern of Epha3/Hek/Mek4 suggests its participation in guidance of cell migration and axonal growth cones during development (6, 12). Epha3/Hek/Mek4 displayed a lobe-dependent variability in expression pattern in the developing chick cerebellum (17), although its role in cerebellar development is poorly understood. Mapk6/Erk3 is localized in the nucleus in exponentially growing, quiescent, and growth factor-stimulated cells (4). Fau/MNSFß, a lymphokine product of a T cell hybridoma, was originally isolated from a radiation-induced mouse osteosarcoma (11) and possibly acts as a tumor suppressor gene (23). Among the genes selected by the cell transfection assay, only a few genes, i.e., Nnat, Ier2/pip92/ETR101, and Epha3/Hek/Mek4, have direct evidence of gene expression in neuronal development. However, this does not exclude the possibility of other genes involvement in neuronal development.
There may be criticism that functionality in PC12 cells and cerebellar development are different biological processes. Thus activity in PC12 cells may not reflect processes occurring in cerebellar granule cells. However, another interpretation of our system is that this heterogeneity implies stringency of our experimental conditions: the genes we have identified are active enough to induce phenotypic changes in different types of cells, i.e., both PC12 and granule cells. The identification of functional genes depends on the assay system, and our current procedure is the most stringent as far as using this heterogeneous system. More sensitive assays such as transfection under weak stimulation by growth factors, e,g, NGF, may identify other genes. It should be noted that sensitive assays may increase pseudopositive hits, and it will be important to choose optimal conditions. Despite several reservations, we stress that the cell transfection array or comparable large-scale functional analysis is per se a strong approach to finding new molecules involved in a particular biological process.
 |
GRANTS
|
---|
This study was supported by the Program for Promotion of Fundamental Studies in Health Sciences of the Organization for Pharmaceuticals and Medical Devices Agency.
 |
FOOTNOTES
|
---|
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).
Address for reprint requests and other correspondence: K. Kato, Research Institute, Osaka Medical Center for Cancer and Cardiovascular Diseases, 1-3-2 Nakamichi, Higashinari-ku, Osaka 537-8511, Japan (E-mail: katou-ki{at}mc.pref.osaka.jp).
10.1152/physiolgenomics.00142.2004.
 |
REFERENCES
|
---|
- Abouzied MM, Baader SL, Dietz F, Kappler J, Gieselmann V, and Franken S. Expression patterns and different subcellular localization of the growth factors HDGF (hepatoma-derived growth factor) and HRP-3 (HDGF-related protein-3) suggest functions in addition to their mitogenic activity. Biochem J 378: 169176, 2004.[CrossRef][ISI][Medline]
- Altman J and Bayer SA. Development of the Cerebellar System: In Relation to its Evolution, Structure, and Functions. Boca Raton, FL: CRC 1996.
- Charles CH, Simske JS, O'Brien TP, and Lau LF. Pip92: a short-lived, growth factor-inducible protein in BALB/c 3T3 and PC12 cells. Mol Cell Biol 10: 67696774, 1990.[ISI][Medline]
- Cheng M, Boulton TG, and Cobb MH. ERK3 is a constitutively nuclear protein kinase. J Biol Chem 271: 89518958, 1996.[Abstract/Free Full Text]
- Chung KC, Shin SW, Yoo M, Lee MY, Lee HW, Choe BK, and Ahn YS. A systemic administration of NMDA induces immediate early gene pip92 in the hippocampus. J Neurochem 75: 917, 2000.[CrossRef][ISI][Medline]
- Connor RJ, Menzel P, and Pasquale EB. Expression and tyrosine phosphorylation of Eph receptors suggest multiple mechanisms in patterning of the visual system. Dev Biol 193: 2135, 1998.[CrossRef][ISI][Medline]
- de Wildt RM, Mundy CR, Gorick BD, and Tomlinson IM. Antibody arrays for high-throughput screening of antibody-antigen interactions. Nat Biotechnol 18: 989994, 2000.[CrossRef][ISI][Medline]
- Doherty P, Ashton SV, Moore SE, and Walsh FS. Morphoregulatory activities of NCAM and N-cadherin can be accounted for by G protein-dependent activation of L- and N-type neuronal Ca2+ channels. Cell 67: 2133, 1991.[CrossRef][ISI][Medline]
- Dou D and Joseph R. Cloning of human neuronatin gene and its localization to chromosome-20q 11.212: the deduced protein is a novel "proteolipid". Brain Res 723: 822, 1996.[CrossRef][ISI][Medline]
- Falsey JR, Renil M, Park S, Li S, and Lam KS. Peptide and small molecule microarray for high throughput cell adhesion and functional assays. Bioconjug Chem 12: 346353, 2001.[CrossRef][ISI][Medline]
- Finkel MP, Reilly CA Jr, and Biskis BO. Pathogenesis of radiation and virus-induced bone tumors. Recent Results Cancer Res 54: 92103, 1976.[Medline]
- Flenniken AM, Gale NW, Yancopoulos GD, and Wilkinson DG. Distinct and overlapping expression patterns of ligands for Eph-related receptor tyrosine kinases during mouse embryogenesis. Dev Biol 179: 382401, 1996.[CrossRef][ISI][Medline]
- Greene LA and Tischler AS. Establishment of a noradrenergic clonal line of rat adrenal pheochromocytoma cells which respond to nerve growth factor. Proc Natl Acad Sci USA 73: 24242428, 1976.[Abstract/Free Full Text]
- Guthridge MA, Seldin M, and Basilico C. Induction of expression of growth-related genes by FGF-4 in mouse fibroblasts. Oncogene 12: 12671278, 1996.[ISI][Medline]
- Hattori M, Tsukamoto N, Nur-e-Kamal MS, Rubinfeld B, Iwai K, Kubota H, Maruta H, and Minato N. Molecular cloning of a novel mitogen-inducible nuclear protein with a Ran GTPase-activating domain that affects cell cycle progression. Mol Cell Biol 15: 552560, 1995.[Abstract]
- Honma K, Ochiya T, Nagahara S, Sano A, Yamamoto H, Hirai K, Aso Y, and Terada M. Atelocollagen-based gene transfer in cells allows high-throughput screening of gene functions. Biochem Biophys Res Commun 21: 10751081, 2001.
- Karam SD, Burrows RC, Logan C, Koblar S, Pasquale EB, and Bothwell M. Eph receptors and ephrins in the developing chick cerebellum: relationship to sagittal patterning and granule cell migration. J Neurosci 20: 64886500, 2000.[Abstract/Free Full Text]
- Kato K. Adaptor-tagged competitive PCR: a novel method for measuring relative gene expression. Nucleic Acids Res 25: 46944696, 1997.[Abstract/Free Full Text]
- Li H, Kawasaki H, Nishida E, Hattori S, and Nakamura S. Ras-regulated hypophosphorylation of the retinoblastoma protein mediates neuronal differentiation in PC12 cells. J Neurochem 66: 22872294, 1996.[ISI][Medline]
- Matoba R, Kato K, Kurooka C, Maruyama C, Sakakibara Y, and Matsubara K. Correlation between gene functions and developmental expression patterns in the mouse cerebellum. Eur J Neurosci 12: 13571371, 2000.[CrossRef][ISI][Medline]
- Matoba R, Kato K, Saito S, Kurooka C, Maruyama C, Sakakibara Y, and Matsubara K. Gene expression in mouse cerebellum during its development. Gene 241: 125131, 2000.[CrossRef][ISI][Medline]
- Matoba R, Saito S, Ueno N, Maruyama C, Matsubara K, and Kato K. Gene expression profiling of mouse postnatal cerebellar development. Physiol Genomics 4: 155164, 2000.[Abstract/Free Full Text]
- Michiels L, Van der Rauwelaert E, Van Hasselt F, Kas K, and Merregaert J. fau cDNA encodes a ubiquitin-like-S30 fusion protein and is expressed as an antisense sequence in the Finkel-Biskis-Reilly murine sarcoma virus. Oncogene 8: 25372546, 1993.[ISI][Medline]
- Mootha VK, Lindgren CM, Eriksson KF, Subramanian A, Sihag S, Lehar J, Puigserver P, Carlsson E, Ridderstrale M, Laurila E, Houstis N, Daly MJ, Patterson N, Mesirov JP, Golub TR, Tamayo P, Spiegelman B, Lander ES, Hirschhorn JN, Altshuler D, and Groop LC. PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes. Nat Genet 34: 267273, 2003.[CrossRef][ISI][Medline]
- Muro S, Takemasa I, Oba S, Matoba R, Ueno N, Maruyama C, Yamashita R, Sekimoto M, Yamamoto H, Nakamori S, Monden M, Ishii S, and Kato K. Identification of expressed genes linked to malignancy of human colorectal carcinoma by parametric clustering of quantitative expression data. Genome Biol 4: R21, 2003.[CrossRef][Medline]
- Nakamura H, Izumoto Y, Kambe H, Kuroda T, Mori T, Kawamura K, Yamamoto H, and Kishimoto T. Molecular cloning of complementary DNA for a novel human hepatoma-derived growth factor. J Biol Chem 269: 2514325149, 1994.[Abstract/Free Full Text]
- Ochiya T, Takahama Y, Nagahara S, Sumita Y, Hisada A, Itoh H, Nagai Y, and Terada M. New delivery system for plasmid DNA in vivo using atelocollagen as a carrier material: the minipellet. Nat Med 5: 707710, 1999.[CrossRef][ISI][Medline]
- Ochiya T, Nagahara S, Sano A, Itoh H, and Terada M. Biomaterials for gene delivery: atelocollagen-mediated controlled release of molecular medicines. Curr Gene Ther 1: 3152, 2001.[Medline]
- Paggi MG, Bonetto F, Severino A, Baldi A, Battista T, Bucci F, Felsani A, Lombardi D, and Giordano A. The retinoblastoma-related Rb2/p130 gene is an effector downstream of AP-2 during neural differentiation. Oncogene 20: 25702578, 2001.[CrossRef][ISI][Medline]
- Saito S, Matoba R, and Kato K. Adapter-tagged competitive PCR (ATAC-PCR)a high-throughput quantitative PCR method for microarray validation. Methods 31: 326331, 2003.[CrossRef][ISI][Medline]
- Shackleford GM, Willert K, Wang J, and Varmus HE. The Wnt-1 proto-oncogene induces changes in morphology, gene expression, and growth factor responsiveness in PC12 cells. Neuron 11: 865875, 1993.[CrossRef][ISI][Medline]
- Wijnholds J, Chowdhury K, Wehr R, and Gruss P. Segment-specific expression of the neuronatin gene during early hindbrain development. Dev Biol 171: 7384, 1995.[CrossRef][ISI][Medline]
- Ziauddin J and Sabatini DM. Microarrays of cells expressing defined cDNAs. Nature 411: 107110, 2001.[CrossRef][ISI][Medline]
Copyright © 2005 by the American Physiological Society.