Neurofilament Gene Research Articles

Interaction of multiple nuclear proteins with the promoter region of the mouse 68-kDa neurofilament gene.

Four brain-specific, DNase I hypersensitive sites (HSS) have been mapped to the 5' flanking region of the mouse 68-kDa neurofilament gene. These sites are contained within a 1.7-kb sequence that confers neuronal specificity of expression of this gene in transgenic mice. To identify DNA sequences that might be involved in gene regulation, the HSS situated near the promoter region has been analyzed by gel mobility shift assays and DNase I footprinting to investigate protein binding sequences. Of particular interest are two footprints localized to a 9-nucleotide sequence that flanks both the light and medium neurofilament gene in mouse and to a sequence that demonstrates partial homology to several promoter regions, including element-1, a motif required for neuron specificity in Drosophila. A prominent footprint was also detected at a sequence that contains a near-perfect palindrome centered at a PstI restriction site.

Origin of the two mRNA species for the human neurofilament light gene.

The human neurofilament light (hNF-L) gene yields two major species of mRNAs of 2.4 and 3.8 kilobases (kb) in size. To investigate the origin of these two mRNAs, we have analyzed in transgenic mice the expression of hNF-L DNA fragments including different lengths of 5'-flanking regions. The finding that the 3.8-kb mRNA species is produced by a hNF-L transgene that includes only the proximal promoter region (-0.3 kb) demonstrates that both the 2.4- and 3.8-kb mRNAs are derived from the same site of transcription initiation. Sequencing of the 3' untranslated region of the hNF-L gene revealed the presence of multiple AATAAA polyadenylation signals. We conclude from Northern blotting experiments using probes spanning various regions of the hNF-L gene that the 2.4- and 3.8-kb mRNAs originate from the selective use of polyadenylation signals located 1.4 kb apart.

Aluminum-induced neurofilamentous changes in cultured rat dorsal root ganglia explants

Intrathecal administration of aluminum (AI) salts to susceptible species causes prominent accumulations of neurofilaments (NFs) in neurons of the CNS. Involved nerve cells display abnormal phosphorylation of perikaryal NFs, impaired axonal transport of NFs, and reduced levels of mRNA for NF proteins. Further understanding of the pathogenesis of AI toxicity has been limited by difficulties inherent in the available in vivo systems. For this reason, we have developed a model to study the effects of AI on cultured sensory neurons. Explant cultures of rat dorsal root ganglia (DRG) were exposed to 1 mM aluminum lactate for 1 d, 3 d, or 7 d and then examined morphologically. Accumulations of NFs were noted as early as 1 d after exposure, and prominent masses of NFs were seen at 3 and 7 d. Northern analysis of mRNA extracted from the cultured ganglia showed that high, medium, and low molecular weight NF protein mRNA levels were markedly reduced compared to control values by 1 d of exposure. Class II beta-tubulin mRNA was also moderately decreased. Reversibility of toxicity was assessed by removing the aluminum lactate from the medium after a 3 d exposure and examining the cultures 1 week later. The perikaryal masses of NFs dispersed and the levels of mRNA coding for the NF proteins and class II beta-tubulin increased. The neurotoxic effects of AI on cultured DRG recapitulates the effects of intrathecal administration of AI on animals; this model produces similar changes in neuronal morphology with neurofilamentous masses and similar modifications of NF gene expression.(ABSTRACT TRUNCATED AT 250 WORDS)

Negative regulatory regions are present upstream in the three mouse neurofilament genes

We have cloned and examined the 5′ flanking regions of the heavy (NF-H), light (NF-L) and mid-sized (NF-M) mouse neurofilament (NF) genes in order to begin to characterize the regions of each gene that regulate NF transcription. Chimeric plasmids bearing the CAT reporter gene and deletion mutants of the upstream NF genes were transiently transfected into neuronal (PC12 and Neuro 2A) and non-neuronal (HeLa) cell lines. Constructs bearing upstream regions to −4000 in NF-H, to −5600 in NF-L and to −4500 in NF-M were expressed at low levels in neuronal and in non-neuronal cells. Progressive deletion of 5′ flanking sequence to −385 in NF-H, to −325 in NF-L and to −505 in NF-M caused a several-fold increase of transcription from the transfected plasmids. Increases of transcription by deletion mutants followed a similar pattern in neuronal and in non-neuronal cell lines. Negative upstream regions are located between −1314 and −385 in NF-H, between −936 and −325 in NF-L and between −874 and −505 in NF-M. Additional negative regions are present further upstream in NF-L and in NF-H. The negative regions of NF-H and of NF-L suppress transcription when placed in either orientation in front of the SV40 or a heterologous NF promoter. These studies demonstrate that the three mouse NF genes possess similar functional features, namely, that of a relatively strong and promiscuous promoter with negative upstream elements. The role of the negative elements in regulating NF expression remains unclear

Postnatal increases in neurofilament gene expression correlate with the radial growth of axons

Neurofilament gene expression appears to play an important role in regulating axonal calibre. In the dorsal root ganglia of mature mammals, large sensory neurons contain high levels of neurofilament mRNAs and give rise to large-calibre myelinated axons (with diameters up to 8 microns in rat), while small sensory neurons contain undetectable levels of neurofilament mRNAs and give rise to unmyelinated axons (with diameters less than 1 micron). In the present study we used a combination of morphological and molecular approaches to examine the relationships among postnatal increases in neurofilament gene expression, growth in perikaryal size, growth in axonal calibre and myelin formation in lumbar sensory neurons of rat. Using in situ hybridization, three populations of sensory neurons could be clearly distinguished at birth: (1) neurons containing relatively high levels of neurofilament mRNAs; (2) neurons containing low levels of neurofilament mRNAs; and (3) neurons containing undetectable levels of neurofilament mRNAs. Perikaryal size was greater for neurons with high levels of neurofilament mRNAs than for those with either low or undetectable levels. The proportion of neurons expressing high levels of neurofilament mRNAs increased from approximately 10% at birth to 30% by 28 days of age; increases in the abundance of neurofilament mRNAs in these neurons between 0 and 28 days of age, as documented by blot analyses of RNA purified from dorsal root ganglia, correlated with increases in perikaryal size. This postnatal rise in neurofilament gene expression also correlated with an increase in the cross-sectional areas of myelinated axons in the L5 dorsal root. As axons matured in their relationship to Schwann cells (polyaxonal pockets----ensheathed----segregated----myelinated), their cross-sectional areas increased. Thus, growth in both perikaryal size and axonal calibre correlated closely with increased neurofilament gene expression in these sensory neurons. These findings are consistent with the hypothesis that neurofilament expression plays an important role in the sequence of events leading to the radial growth and myelination of axons.

Corticospinal neurons exhibit a novel pattern of cytoskeletal gene expression after injury

We examined changes in the expression of major cytoskeletal protein mRNAs in adult hamster corticospinal neurons after axotomy. While a number of studies had determined that peripheral neurons exhibit major alterations in cytoskeletal gene expression after axotomy, no previous studies had addressed the question of whether or not intrinsic mammalian CNS neurons, which do not have the ability to successfully regenerate axons after injury, alter their expression of tubulin and neurofilament genes after injury. In the present study we used in situ hybridization methods to examine this issue. 35S-labeled cDNA probes for the low molecular weight neurofilament protein (NF-L) mRNA and an alpha-tubulin mRNA species (M alpha 1) were used for in situ hybridizations of sections of the sensorimotor cortex obtained 2, 7, and 14 days after unilateral axotomy of the corticospinal tract in the caudal medulla. Film as well as emulsion autoradiography showed dramatic decreases in both alpha-tubulin and NF-L mRNA levels within axotomized neurons in layer Vb of the sensorimotor cortex. Tubulin mRNA levels were decreased as early as 2 days after injury whereas NF-L mRNA levels were not decreased until later times. Ribosomal RNA (rRNA) levels in axotomized corticospinal neurons were also examined using in situ hybridization with a 35S-labeled rDNA probe. These studies showed only a slight decrease in rRNA levels in corticospinal neurons at 14 days after axotomy. Immunoblotting experiments of total protein from corticospinal axons in the medulla were performed to assess whether the axonal composition immediately proximal to the injury site reflected changes in cell body gene expression. Both alpha-tubulin and NF-L levels were found to decrease in corticospinal axons by 28 days after injury. These findings, to our knowledge, are the first to demonstrate that a class of mammalian CNS neurons have an intrinsically different cytoskeletal response to axonal injury than do PNS neurons. The failure to upregulate tubulin gene expression following injury may contribute to the ineffective regenerative response of these long-tract CNS neurons.

Squid low molecular weight neurofilament proteins are a novel class of neurofilament protein. A nuclear lamin-like core and multiple distinct proteins formed by alternative RNA processing

The primary structure of the major 60-kDa squid low molecular mass neurofilament protein (NF60) and a related 70-kDa neurofilament protein have been determined from cDNA clones isolated from a squid brain cDNA library. Structural analysis suggests that the squid NF60 and NF70 neurofilament genes and proteins are remarkably distinct from vertebrate neuronal intermediate filaments characterized previously. Both proteins are encoded on mRNAs generated by alternative RNA processing of the primary transcript of a single gene. Among the known intermediate filament proteins, NF60 and NF70 neurofilament proteins show highest similarity to an epithelial intermediate filament protein from Helix pomatia, a gastropod mollusk, and are less similar to vertebrate neurofilaments. The length of the alpha-helical rod domain in the NF60 and NF70 proteins was reminiscent of the vertebrate nuclear lamins, 6 heptads longer than is found in all known vertebrate cytoplasmic intermediate filaments, in particular the vertebrate neurofilaments. These distinct structural properties suggest that the vertebrate and invertebrate low molecular weight neurofilaments evolved independently from primordial intermediate filament proteins.

Retinoic acid can mimic endogenous signals involved in transformation of the xenopus nervous system

In the frog Xenopus laevis, signals from the mesoderm divert part of the ectoderm from an epidermal to a neural fate. In the course of neural induction, the neurectoderm also acquires anterior-posterior polarity. In this report, the early expression of two genes, XIHbox6 and the neurofilament gene XIF6, is examined. The pattern of expression of the two genes seen in the tailbud embryo develops progressively over a 4 hr period following gastrulation. Physiological concentrations of retinoic acid can mimic this effect in isolated embryonic explants, consistent with the involvement of retinoic acid, or a closely related molecule, in localizing gene expression along the anterior-posterior axis of the neural tube.

Anomalous Gene Expression in Alzheimer Disease: Cause or Effect

Altered chromatin conformation and increased amounts of aluminum have been observed in the brains of patients with Alzheimer disease. These factors have been shown to affect gene regulation. In this report, we describe how these changes may selectively alter the pool size of the human light chain neurofilament gene and play a fundamental role in the expression of this disease.

V- myc immortalization of early rat neural crest cells yields a clonal cell line which generates both glial and adrenergic progenitor cells

We describe the isolation and characterization of an immortal cell line derived by infection of rat neural crest cells with a v- myc-containing replication-defective retrovirus. This clonal cell line, called NCM-1, contains a majority cell population with antigenic and morphologic properties that suggest it may represent a peripheral glial progenitor. In conditioned or in serum-free medium, these NGF receptor-positive cells differentiate to an elongated, bipolar morphology resembling that of primary Schwann cells. This morphologic differentiation is prevented by TGF- β 1, which also acts as a mitogen for the cells. The NCM-1 line is also able to generate clonal derivatives which have extinguished expression of most or all glial markers. Once generated, such cells are stable and do not revert to the glial phenotype. At least some of these cells have acquired sympathoadrenal progenitor-like properties, as shown by their capacity to coexpress tyrosine hydroxylase (TH) and neurofilament (NF) in response to basic FGF and dexamethasone. These data imply that the NCM-1 line contains self-renewing cells with the potential to generate precursors in at least two of the sublineages that normally develop from the neural crest. This in turn suggests that the process of immortalization may preserve at least some of the developmental properties characteristic of multipotential neural crest cells. NCM-1 cells may prove useful for the study of neural crest cell lineage segregation, Schwann cell differentiation, and the mechanisms controlling the initial induction of TH and NF gene expression.

Decreased Glucocorticoid Receptor Activity Following Glucocorticoid Receptor Antisense RNA Gene Fragment Transfection

Depression is often characterized by increased cortisol secretion caused by hyperactivity of the hypothalamic-pituitary-adrenal axis and by nonsuppression of cortisol secretion following dexamethasone administration. This hyperactivity of the hypothalamic-pituitary-adrenal axis could result from a reduced glucocorticoid receptor (GR) activity in neurons involved in its control. To investigate the effect of reduced neuronal GR levels, we have blocked cellular GR mRNA processing and/or translation by introduction of a complementary GR antisense RNA strand. Two cell lines were transfected with a reporter plasmid carrying the chloramphenicol acetyltransferase (CAT) gene under control of the mouse mammary tumor virus long terminal repeat (a glucocorticoid-inducible promoter). This gene construction permitted assay of the sensitivity of the cells to glucocorticoid hormones. Cells were also cotransfected with a plasmid containing 1,815 bp of GR cDNA inserted in the reverse orientation downstream from either a neurofilament gene promoter element or the Rous sarcoma virus promoter element. Northern (RNA) blot analysis demonstrated formation of GR antisense RNA strands. Measurement of the sensitivity of CAT activity to exogeneous dexamethasone showed that although dexamethasone increased CAT activity by as much as 13-fold in control incubations, expression of GR antisense RNA caused a 2- to 4-fold decrease in the CAT response to dexamethasone. Stable transfectants bearing the GR antisense gene fragment construction demonstrated a 50 to 70% decrease of functional GR levels compared with normal cells, as evidenced by a ligand-binding assay with the type II glucocorticoid receptor-specific ligand [3H]RU 28362. These results validate the use of antisense RNA to GR to decrease cellular response to glucocorticoids.

Decreased glucocorticoid receptor activity following glucocorticoid receptor antisense RNA gene fragment transfection.

Depression is often characterized by increased cortisol secretion caused by hyperactivity of the hypothalamic-pituitary-adrenal axis and by nonsuppression of cortisol secretion following dexamethasone administration. This hyperactivity of the hypothalamic-pituitary-adrenal axis could result from a reduced glucocorticoid receptor (GR) activity in neurons involved in its control. To investigate the effect of reduced neuronal GR levels, we have blocked cellular GR mRNA processing and/or translation by introduction of a complementary GR antisense RNA strand. Two cell lines were transfected with a reporter plasmid carrying the chloramphenicol acetyltransferase (CAT) gene under control of the mouse mammary tumor virus long terminal repeat (a glucocorticoid-inducible promoter). This gene construction permitted assay of the sensitivity of the cells to glucocorticoid hormones. Cells were also cotransfected with a plasmid containing 1,815 bp of GR cDNA inserted in the reverse orientation downstream from either a neurofilament gene promoter element or the Rous sarcoma virus promoter element. Northern (RNA) blot analysis demonstrated formation of GR antisense RNA strands. Measurement of the sensitivity of CAT activity to exogeneous dexamethasone showed that although dexamethasone increased CAT activity by as much as 13-fold in control incubations, expression of GR antisense RNA caused a 2- to 4-fold decrease in the CAT response to dexamethasone. Stable transfectants bearing the GR antisense gene fragment construction demonstrated a 50 to 70% decrease of functional GR levels compared with normal cells, as evidenced by a ligand-binding assay with the type II glucocorticoid receptor-specific ligand [3H]RU 28362. These results validate the use of antisense RNA to GR to decrease cellular response to glucocorticoids.

Involvement of neurofilaments in the radial growth of axons

The control of radial growth of axons is of functional importance because caliber is a principal determinant of conduction velocity in myelinated nerve fibers. Neurofilaments, the major cytoskeletal protein in myelinated nerves, appear to be intrinsic determinants of caliber. Evidence supporting this derives first from the linear relationship between neurofilament content and axonal diameter. Further, following distal axonal injury in a peripheral nerve, caliber is reduced in the proximal axonal stumps. This reduction in caliber is itself due to selective suppression of neurofilament gene expression, thereby leading to lower levels of newly synthesized neurofilament subunits transported into the axon and a consequent decrease in axonal neurofilament content. To demonstrate directly the physiological consequence of altering normal neurofilament accumulation, we have elevated neurofilament expression by introducing additional genes into transgenic mice. The clear result is that increases in NF-L content alone are not sufficient to increase axonal caliber. To test the consequence of disruption of normal filament accumulation, we have identified dominant assembly-disrupting mutants in NF-L and NF-M and have used these to produce transgenic animals in which neurofilament assembly should be disrupted.

Localization and quantitation of 68 kDa neurofilament and superoxide dismutase-1 mRNA in alzheimer brains

The technique of in situ hybridization with tritiated RNA probes was used to study the expression of the 68 kDa neurofilament (NF68) gene and the superoxide dismutase-1 (SOD-1) gene in the brains of Alzheimer's disease (AD) patients. Messenger RNA (mRNA) for these proteins was localized and quantified in single cells of formalin-fixed, paraffin-embedded sections of 4 pairs of AD and Huntington's disease (HD) brains from patients matched for age at death and autopsy interval. The cerebellar cortex and hippocampal CA1 and CA2 regions were compared in these two groups of subjects, since in AD the CA2 region of the hippocampus and the cerebellum have been found to be relatively unaffected by the Alzheimer process in comparison to the hippocampal CA1 region. The amount of NF68 mRNA was reduced by approximately 50% in pyramidal cells of both the CA1 and CA2 of AD hippocampus ( P < 0.001), and by 15% in the Purkinje cells of AD cerebellum ( P < 0.05) relative to that of the HD individuals. SOD-1 mRNA was reduced by about 22% in the CA1 of AD brains ( P < 0.001) with no corresponding reduction in the CA2, and by only 5% in the AD cerebellum ( P > 0.5). The paired design of the study suggests that these results are not simply attributable to the effects of autopsy interval or the agonal process in each patient's death.

Structure of an invertebrate gene encoding cytoplasmic intermediate filament (IF) proteins: implications for the origin and the diversification of IF proteins.

The structure of the single gene encoding the cytoplasmic intermediate filament (IF) proteins in non-neuronal cells of the gastropod Helix aspersa is described. Genomic and cDNA sequences show that the gene is composed of 10 introns and 11 exons, spanning greater than 60 kb of DNA. Alternative RNA processing accounts for two mRNA families which encode two IF proteins differing only in their C-terminal sequence. The intron/exon organization of the Helix rod domain is identical to that of the vertebrate type III IF genes in spite of low overall protein sequence homology and the presence of an additional 42 residues in coil 1b of the invertebrate sequence. Intron position homology extends to the entire coding sequence comprising both the rod and tail domains when the invertebrate IF gene is compared with the nuclear lamin LIII gene of Xenopus laevis presented in the accompanying report of Döring and Stick. In contrast the intron patterns of the tail domains of the invertebrate IF and the lamin genes differ from those of the vertebrate type III genes. The combined data are in line with an evolutionary descent of cytoplasmic IF proteins from a nuclear lamin-like progenitor and suggest a mechanism for this derivation. The unique position of intron 7 in the Helix IF gene indicates that the archetype IF gene arose by the elimination of the nuclear localization sequence due to the recruitment of a novel splice site. The presumptive structural organization of the archetype IF gene allows predictions with respect to the later diversification of metazoan IF genes. Whereas models proposing a direct derivation of neurofilament genes seem unlikely, the earlier speculation of an mRNA transposition mechanism is compatible with current results.

Structure of the 68-kDa neurofilament gene and regulation of its expression.

The complete structure of the mouse 68-kDa neurofilament (NF-L) gene was elucidated. We cloned cDNAs corresponding to 3.5- and 2.3-kb NF-L mRNA, including their polyadenylation sites. Sequence analysis revealed that these NF-L mRNAs arose from the alternative use of two polyadenylation sites in exon 4. Promoter analysis using NF-L promoter-beta-galactosidase fusion plasmids determined regions responsible for its basic promoter activity, which were located between -328 and -36 base pairs from the transcription initiation site. These promoter fusion plasmids induced a significant level of beta-galactosidase in NF-nonproducing C6 cells as well as in NF-producing PC12h cells. The in vitro transcription assay using HeLa cell extract also showed that this promoter exhibited strong transcriptional activity. Little difference in NF-L mRNA stability was observed between the two cells. However, nuclear run-off assay revealed that the NF-L gene was not transcribed in NF-nonproducing C6 cells. These data suggest that the strong promoter activity of the NF-L gene is repressed in vivo at the transcription initiation level in a tissue-specific manner.

Nerve growth factor rapidly induces expression of the 68-kDa neurofilament gene by posttranscriptional modification in PC12h-R cells.

We have isolated a subclonal cell line of PC12, PC12h-R, which is characterized by its rapid response to nerve growth factor (NGF) in extending its neurites. The mRNA levels of the 68-kDa neurofilament (NF-L) increased severalfold as early as 3 h after the addition of NGF, but this was not accompanied by an increase in the transcription rate. Analysis of the stabilities of 3.5- and 2.3-kilobase (kb) NF-L mRNAs using actinomycin D revealed that 2.3-kb mRNA was degraded much faster than the 3.5-kb species and that NGF apparently stabilized NF-L mRNAs. These results show that the NF-L gene can be regulated posttranscriptionally, a finding that may be important in responding to the acute demand for neurofilament syntheses.

Differential regulation of peripherin and neurofilament gene expression in regenerating rat DRG neurons

Quantitative in situ hybridization and RNA blotting methods were used to define the time course and magnitude of changes in expression of mRNAs encoding peripherin and the neurofilament (NF) triplet proteins in rat dorsal root ganglion (DRG) neurons during axonal regeneration. mRNA levels in adult rat L4 and L5 DRGs were examined in autoradiograms after in situ hybridization with specific 35S-labeled cDNA probes 1-56 days following unilateral crush lesions of the sciatic nerve. The results of quantitative analyses indicated that peripherin mRNA levels were significantly increased in large-sized (greater than 1000 microns 2) DRG neurons at 7, 14, and 28 days after axotomy while the mRNA levels for each of the NF triplet proteins were significantly decreased at these same time points. The mRNA levels of the low (NF-L) and middle (NF-M) sized NF subunits were significantly decreased as early as 1 day postaxotomy but the mRNA level of the large NF subunit (NF-H) did not change until 7 days after axotomy. The maximal reduction in NF mRNA levels was observed at 14 days postaxotomy when NF-L mRNA levels were only 35% of those in large-sized, normal control neurons. Recovery toward normal levels of both NF and peripherin mRNAs was observed at 8 weeks postaxotomy. RNA blot analyses with total RNA obtained from DRGs at different postaxotomy times confirmed that NF-L mRNA levels were reduced in the DRG during the first 4 weeks after axotomy but, interestingly, failed to detect an increase in peripherin mRNA levels. This difference concerning peripherin mRNA levels in axotomized preparations obtained by RNA blotting vs. in situ hybridization was attributed to the fact that RNA blots utilized total DRG RNA which includes mRNAs from both small and large-sized DRG neurons. A recent in situ hybridization study showed that the small-sized DRG neurons which contain the majority of the peripherin mRNA in the DRG do not increase their peripherin mRNA levels 14 days after axotomy (Oblinger et al., 1989b). This may mask any change in the large neuron response when total RNA is examined. Overall, the results of this study demonstrate (1) that type III (peripherin) and type IV (NF) intermediate filament genes are regulated differently during axonal regeneration, and (2) that the three NF genes are down-regulated in a fairly coordinate manner during regeneration. These data suggest that an important component of the regeneration program is the alteration of the composition of the IF component of the cytoskeleton.(ABSTRACT TRUNCATED AT 400 WORDS)

Assembly properties of dominant and recessive mutations in the small mouse neurofilament (NF-L) subunit.

We have generated a set of amino- and carboxy-terminal deletions of the NF-L neurofilament gene and determined the assembly properties of the encoded subunits after coexpression with vimentin or wild-type NF-L. NF-L molecules missing greater than 30% (31 amino acids of the head) or 90% (128 amino acids of the tail) failed to incorporate into intermediate filament networks. Carboxy-terminal deletions into the rod domain yield dominant mutants that disrupt arrays assembled from wild-type subunits, even when present at levels of approximately 2% of the wild-type subunits. Even mutants retaining 55% of the tail (61 amino acids) disrupt normal arrays when accumulated above approximately 10% of wild-type subunits. Since deletion of greater than 90% of the head domain produces "recessive" assembly incompetent subunits that do not affect wild-type filament arrays, whereas smaller deletions yield efficient network disruption, we conclude that some sequence(s) in the head domain (within residues 31-87) are required for the earliest steps in filament assembly. Insertional mutagenesis in the nonhelical spacer region within the rod domain reveals that as many as eight additional amino acids can be tolerated without disrupting assembly competence.

Different regulation of mid-size neurofilament and N-myc mRNA expression during neuroblastoma cell differentiation induced by retinoic acid

1. Neuroblastoma (NB) is an unusual neuroectodermal tumor showing a high degree of spontaneous regression. NB cells can be induced to differentiate in vitro by various agents. Cell differentiation results in morphological changes characteristic of the mature neuronal phenotype, including outgrowth of neurite-like structures with several interconnections. 2. Recent experiments indicate that morphological differentiation of NB cells is associated with changes in expression of N-myc, c-myc, and c-myb oncogenes and synthesis of neurofilament proteins. However, little is known about the transcription of neurofilament genes during differentiation. 3. We have analyzed the expression of both the N-myc oncogene and mid-size neurofilament (NF) genes in the LAN-1 human NB cell line, cultured in the presence of retinoic acid (RA). Continuous treatment with RA induced morphological differentiation within 5-6 days. The transcription of N-myc was down-modulated within 24 hr of the initial exposure to RA. The mid-size NF mRNA was increased at this time. The expression of N-myc was not modified in serum-deprived LAN-1 cells, indicating that N-myc transcription is unaffected by the arrest of the cells in the G1 phase. 4. We conclude that new synthesis of mid-size NF mRNA and a decrease in N-myc transcription precede de novo formation of neurite-like processes and morphological cell differentiation of neuroblastoma cells.