Phosphorylation of the head domain of neurofilament protein (NF-M): a factor regulating topographic phosphorylation of NF-M tail domain KSP sites in neurons

In neurons the phosphorylation of neurofilament (NF) proteins NF-M and NF-H is topographically regulated. Although kinases and NF subunits are synthesized in cell bodies, extensive phosphorylation of the KSP repeats in tail domains of NF-M and NF-H occurs primarily in axons. The nature of this regulation, however, is not understood. As obligate heteropolymers, NF assembly requires interactions between the core NF-L with NF-M or NF-H subunits, a process inhibited by NF head domain phosphorylation. Phosphorylation of head domains at protein kinase A (PKA)-specific sites seems to occur transiently in cell bodies after NF subunit synthesis. We have proposed that transient phosphorylation of head domains prevents NF assembly in the soma and inhibits tail domain phosphorylation; i.e. assembly and KSP phosphorylation in axons depends on prior dephosphorylation of head domain sites. Deregulation of this process leads to pathological accumulations of phosphorylated NFs in the soma as seen in some neurodegenerative disorders. To test this hypothesis, we studied the effect of PKA phosphorylation of the NF-M head domain on phosphorylation of tail domain KSP sites. In rat cortical neurons we showed that head domain phosphorylation of endogenous NF-M by forskolin-activated PKA inhibits NF-M tail domain phosphorylation. To demonstrate the site specificity of PKA phosphorylation and its effect on tail domain phosphorylation, we transfected NIH3T3 cells with NF-M mutated at PKA-specific head domain serine residues. Epidermal growth factor stimulation of cells with mutant NF-M in the presence of forskolin exhibited no inhibition of NF-tail domain phosphorylation compared with the wild type NF-M-transfected cells. This is consistent with our hypothesis that transient phosphorylation of NF-M head domains inhibits tail domain phosphorylation and suggests this as one of several mechanisms underlying topographic regulation.     

Activation of mitogen-activated protein kinases (Erk1 and Erk2) cascade results in phosphorylation of NF-M tail domains in transfected NIH 3T3 cells.

Neurofilaments (NFs) are neuron-specific intermediate filaments, and are the major cytoskeletal component in large myelinated axons. Lysine-serine-proline (KSP) repeats in the tail domains of high molecular weight NF proteins (NF-M and NF-H) are extensively phosphorylated in vivo in the axon. This phosphorylation in the tail domain has been postulated to play an important role in mediating neuron-specific properties, including axonal caliber and conduction velocity. Recent studies have shown that the mitogen-activated protein kinases (extracellular signal-regulated kinases, Erk1 and Erk2) phosphorylate KSP motifs in peptide substrates derived from the NF-M and NF-H tail domains in vitro. However, it is not clear whether activation of the mitogen activated protein (MAP) kinase pathway is able to phosphorylate these domains in vivo. To answer this question, a constitutively active form of mitogen-activated Erk activating kinase (MEK1) was cotransfected with an NF-M expression construct into NIH 3T3 cells. The activated mutant, but not the dominant negative mutant, induced phosphorylation of NF-M. In addition, it was shown that epidermal growth factor, which induces the MAP kinase cascade in NIH 3T3 cells, also activated endogenous Erk1 and Erk2 and NF-M tail domain phosphorylation in the transfected cells. These results present direct evidence that in-vivo activation of Erk1 and Erk 2 is sufficient for NF-M tail domain phosphorylation in transfected cells.     

CDK-5-Mediated Neurofilament Phosphorylation in SHSY5Y Human Neuroblastoma Cells

Cyclin-dependent kinase-5 (CDK-5) has been shown to play important roles in neuronal development and neurogenesis. In vitro studies indicate a role of CDK-5 in phosphorylation of neurofilaments (NFs). In this study, we have chosen the human neuroblastoma cell line SHSY5Y as a model system to study the in vivo phosphorylation of NF proteins by CDK-5. Upon differentiation of SHSY5Y cells with retinoic acid, we found that the phosphorylation of high molecular mass (NF-H) and medium molecular mass (NF-M) NFs increased, whereas the CDK-5 protein level and kinase activity were unaffected. The role of CDK-5 in the phosphorylation of cytoskeletal proteins was studied by using antisense oligonucleotides (ONs) to inhibit the expression of the CDK-5 gene. We found that inhibition of CDK-5 levels by antisense ON treatment resulted in a decrease in phosphorylation of NF-H that correlated with a decline in neurite outgrowth. These results demonstrate that CDK-5 is a major proline-directed kinase phosphorylating the human NF-H tail domain.     

Role of phosphorylation on the structural dynamics and function of types III and IV intermediate filaments

Phosphorylation of types III and IV intermediate filaments (IFs) is known to regulate their organization and function. Phosphorylation of the amino-terminal head domain sites on types III and IV IF proteins plays a key role in the assembly/disassembly of IF subunits into 10 nm filaments, and influences the phosphorylation of sites on the carboxyl-terminal tail domain. These phosphorylation events are largely under the control of second messenger-dependent protein kinases and provide the cells a mechanism to reorganize the IFs in response to the changes in second messenger levels. In mitotic cells, Cdk1, Rho kinase, PAK1 and Aurora-B kinase are believed to regulate vimentin and glial fibrillary acidic protein phosphorylation in a spatio-temporal manner. In neurons, the carboxyl-terminal tail domains of the NF-M and NF-H subunits of heteropolymeric neurofilaments (NFs) are highly phosphorylated by proline-directed protein kinases. The phosphorylation of carboxyl-terminal tail domains of NFs has been suspected to play roles in forming cross-bridges between NFs and microtubules, slowing axonal transport and promoting their integration into cytoskeleton lattice and, in doing so, to control axonal caliber and stabilize the axon. The role of IF phosphorylation in disease pathobiology is discussed.     

A Unique Neurofilament from Torpedo Electric Lobe: Sequence, Expression, and Localization Analysis

A set of cDNA clones encoding a protein highly homologous to the mammalian middle-size class of neurofilaments (NF-M) was characterized. The amino acid similarity between the Torpedo and rat NF-M approaches 90% in the amino-terminal "rod-like" domain and is significantly lower in the carboxy-terminal tail. The Torpedo protein contains 13 tandem repeats of a unique six amino acid core, containing a Lys-Ser-Lys putative phosphorylation site. Surprisingly, the 3' untranslated region contains stretches of 80-90% nucleic acid homology with the mammalian, but not with the chicken sequences. This homology is greater than much of the coding region, suggesting that the 3' untranslated region of the message has an important functional role, perhaps governing RNA stability or localization. This Torpedo NF-M mRNA is expressed specifically in the electric lobe and was not detected in other tissues, including brain and spinal cord. A polyclonal antibody generated against a fusion protein synthesized in E. coli detects a 150-kDa protein in the electric lobe and organ, as well as a small amount of material in the brain. Cytochemical studies reveal immunoreactivity in electromotor neuron axons and terminals. Specific expression of neurofilament genes in subsets of central neurons may be important in determining the morphology and functional characteristics of specific neuronal subtypes.     

Integrins stimulate phosphorylation of neurofilament NF-M subunit KSP repeats through activation of extracellular regulated-kinases (Erk1/Erk2) in cultured motoneurons and transfected NIH 3T3 cells

Integrin-mediated interactions of cells with components of the extracellular matrix (ECM) regulate cell survival, cell proliferation, cell differentiation and cell migration through activation of multiple intracellular signal transduction pathways. In this study, we have demonstrated that integrin-matrix interactions promote KSP tail-domain phosphorylation of neurofilament medium molecular weight subunits (NF-M) in cultured rat spinal cord motoneurons and NF-M transfected NIH 3T3 cells. We found that laminin and fibronectin induce NF-M tail-domain phosphorylation in motoneurons and NIH 3T3 cells transfected with NF-M, respectively. This phosphorylation was selectively inhibited by PD98059, a specific MEK1 inhibitor. This suggests that laminin and fibronectin-induced MEK1 activation and the downstream targets Erk1 and Erk2 are involved in NF-M KSP tail-domain phosphorylation. This pathway appears to represent one of the mechanisms whereby integrin-extracellular matrix interactions are involved in phosphorylation of the NF-M KSP tail domain.     

Proline-Directed Protein Kinase (p34cdc2/p58cyclin A) Phosphorylates Bovine Neurofilaments

Proline-directed protein kinase (PDPK), a complex of p34cdc2 and p58cyclin A, phosphorylates bovine neurofilaments (NFs) in vitro. Incubation of intact filaments with PDPK led to strong labeling of the heavy (NF-H) and middle (NF-M) molecular weight NF proteins and weaker labeling of the low molecular weight protein (NF-L). All three proteins were phosphorylated in solution, with the best substrate being NF-H. Proteins that had been dephosphorylated by enzymatic treatment were better substrates than native proteins--as many as 6 mol of phosphate were incorporated per mole of NF-H. Partial proteolytic cleavage experiments combined with two-dimensional peptide mapping indicated that NF-H and NF-M were phosphorylated predominantly in the tail domains, with some phosphate also appearing in the heads. Soluble NF-L is phosphorylated on the head domain peptide L-3, whereas NF-L within intact filaments is phosphorylated only on the tail domain peptide L-1. Phosphorylation does not lead to filament disassembly. A possible role for PDPK in NF phosphorylation in vivo is discussed.     

Hyperphosphorylation and Accumulation of Neurofilament Proteins in Transgenic Mice with Alzheimer Presenilin 1 Mutation

Neurofilaments (NFs) are hyperphosphorylated and accumulate in Alzheimer’s disease (AD) brains. In this study, employing the transgenic mouse model, we explored the effect of presenilin 1 (PS-1) mutation on the phosphorylation and distribution of NFs. Western blot analysis showed that there was a significant increase in the phosphorylation of NF-H and NF-M subunits with a concomitant increase in phosphorylated c-Jun N-terminal protein kinase 1/2 (JNK1/2) mitogen-activated protein kinase (MAPK) in hippocampus of PS-1 transgenic mice compared to that of wild-type littermates. Immunohistochemical analysis revealed that phosphorylated NFs accumulated throughout the hippocampal neurons of the transgenic mice. These findings suggest that PS-1 mutation may induce hyperphosphorylation and accumulation of NFs via a JNK1/2-involved mechanism.     

Hypophosphorylated neurofilament subunits undergo axonal transport more rapidly than more extensively phosphorylated subunits in situ

Axonal transport of neurofilaments (NFs) has long been considered to be regulated by phosphorylation. We present evidence that in optic axons of normal mice, the rate of NF axonal transport is inversely correlated with the NF phosphorylation state. In addition to 200 kDa NF-H and 145 kDa NF-M, axonal cytoskeletons from CNS contained a range of phospho-variants of NF-H migrating between 160-200 kDa, and of NF-M migrating at 97-145 kDa. While 160 kDa phospho-variants of NF-H have been well characterized, we confirmed the identity of the previously-described 97 kDa species as a hypophospho-variant of NF-M since (1) pulse-chase metabolic labeling confirmed the 97 kDa species to be a new synthesis product that was converted by phosphorylation over time into a form migrating at 145 kDa, (2) the 97 kDa protein reacted with multiple NF-M antibodies, including one specific for hypophosphorylated NF-M, and (3) dephosphorylation converted NF-M isoforms to 97 kDa. Autoradiographic analyses following metabolic radiolabeling demonstrated that hypophosphorylated NF-H and NF-M isoforms underwent substantially more rapid transport in situ than did extensively phosphorylated isoforms, while NF-H subunits bearing a developmentally delayed C-terminal phospho-epitope transported at a rate slower than that of total 200 kDa NF-H. Differential transport of phospho-variants also highlights that these variants are not homogeneously distributed among NFs, but are segregated to some extent among distinct, although probably overlapping, NF populations, indicating that axonal NFs are not homogeneous with respect to phosphorylation state.     

Requirement of Heavy Neurofilament Subunit in the Development of Axons with Large Calibers

Neurofilaments (NFs) are prominent components of large myelinated axons. Previous studies have suggested that NF number as well as the phosphorylation state of the COOH-terminal tail of the heavy neurofilament (NF-H) subunit are major determinants of axonal caliber. We created NF-H knockout mice to assess the contribution of NF-H to the development of axon size as well as its effect on the amounts of low and mid-sized NF subunits (NF-L and NF-M respectively). Surprisingly, we found that NF-L levels were reduced only slightly whereas NF-M and tubulin proteins were unchanged in NF-H–null mice. However, the calibers of both large and small diameter myelinated axons were diminished in NF-H–null mice despite the fact that these mice showed only a slight decrease in NF density and that filaments in the mutant were most frequently spaced at the same interfilament distance found in control. Significantly, large diameter axons failed to develop in both the central and peripheral nervous systems. These results demonstrate directly that unlike losing the NF-L or NF-M subunits, loss of NF-H has only a slight effect on NF number in axons. Yet NF-H plays a major role in the development of large diameter axons.     

In vivo and in vitro phosphorylation at Ser-493 in the glutamate (E)-segment of neurofilament-H subunit by glycogen synthase kinase 3beta

Neurofilament (NF), a major neuronal intermediate filament, is composed of three subunits, NF-L, NF-M, and NF-H. All three subunits contain a well conserved glutamate (E)-rich region called "E-segment" in the N terminus of the tail region. Although the E-segments of NF-L and NF-M are phosphorylated by casein kinases, it has not been observed in NF-H. Using mass spectrometric analysis, we identified phosphorylation of the E-segment of NF-H, prepared from rat spinal cords, at Ser-493 and Ser-501 in the Ser-Pro sequences. The E-segment kinase was isolated from rat brain extract using column chromatography and identified as glycogen synthase kinase (GSK) 3beta. GSK3beta was shown to phosphorylate at Ser-493 in vitro by phosphopeptide mapping and site-directed mutagenesis, and in vivo in HEK293 cells using the phospho-Ser-493 antibody, but did not phosphorylate Ser-501. GSK3beta preferred Ser-493 to the KSP-repeated sequences for phosphorylation sites in the NF-H tail domain. Moreover, Ser-493 was a better phosphorylation site for GSK3beta than other proline-directed protein kinases, Cdk5/p35 and ERK. GSK3beta in the spinal cord extract was associated with NF cytoskeletons. Taken together, we concluded that Ser-493 in the E-segment of NF-H is phosphorylated by GSK3beta in rat spinal cords.     

Growth cones contain a dynamic population of neurofilament subunits

Neurofilaments (NFs) are classically considered to transport in a primarily anterograde direction along axons, and to undergo bulk degradation within the synapse or growth cone (GC). We compared overall NF protein distribution with that of newly expressed NF subunits within NB2a/d1 cells by transfection with a construct encoding green fluorescent protein (GFP) conjugated NF-M subunits. GCs lacked phosphorylated NF epitopes, and steady-state levels of non-phosphosphorylated NF subunits within GC were markedly reduced compared to those of neurite shaft as indicated by conventional immunofluorescence. However, GCs contained significant levels of GFP-tagged subunits in the form of punctate or short filamentous structures that in some cases exceeded that visualized along the shaft itself, suggesting that GCs contained a relatively higher concentration of newly synthesized subunits. GFP-tagged NF subunits within GCs co-localized with non-phosphorylated NF immunoreactivity. GFP-tagged subunits were observed within GC filopodia in which steady-state levels of NF subunits were too low to be detected by conventional immunofluorescence. Selective localization of fluorescein versus rhodamine fluorescene was observed within GCs following expression of NF-M conjugated to DsRed1-E5, which shifts from fluorescein to rhodamine fluorescence within hours after expression; axonal shafts contained a more even distribution of fluorescein and rhodamine fluorescence, further indicating that GCs contained relatively higher levels of the most-recently expressed subunits. GFP-tagged structures were rapidly extracted from GCs under conditions that preserved axonal structures. These short filamentous and punctate structures underwent rapid bi-directional movement within GCs. Movement of GFP-tagged structures within GCs ceased following application of nocodazole, cytochalasin B, and the kinase inhibitor olomoucine, indicating that their motility was dependent upon microtubules and actin and, moreover, was due to active transport rather than simple diffusion. Treatment with the protease inhibitor calpeptin increased overall NF subunits, but increased those within the GC to a greater extent than those along the shaft, indicating that subunits in the GC undergo more rapid turnover than do those within the shaft. Some GCs contained coiled aggregates of GFP-tagged NFs that appeared to be contiguous with axonal NFs. NFs extended from these aggregates into the advancing GC as axonal neurites elongated. These data are consistent with the presence of a population of dynamic NF subunits within GCs that is apparently capable of participating in regional filament formation during axonal elongation, and support the notion that NF polymerization and transport need not necessarily occur in a uniform proximal-distal manner.     

Phosphorylation of neurofilament H subunit at the tail domain by CDC2 kinase dissociates the association to microtubules

We sought the mammalian neurofilament tail domain-specific kinase. Several well known kinases including cAMP-dependent protein kinase, protein kinase C, Ca(2+)-calmodulin-dependent protein kinase II, casein kinase I, and casein kinase II phosphorylated the high (NF-H) and middle molecular mass subunit (NF-M) of bovine neurofilaments, but they did not reduced the electrophoretic mobility of the dephosphorylated form of NF-M and NF-H by phosphorylation nor was the amount of phosphorylation increased by dephosphorylation of NF proteins, indicating that the phosphorylation sites by these kinases are not major in vivo phosphorylation sites at the tail domain. In contrast, cdc2 kinase phosphorylated specifically the dephosphorylated form of NF-H. 4 mol of phosphates were incorporated per mol of NF-H and this phosphorylation returned the electrophoretic mobility of the dephosphorylated form of NF-H to the position of the isolated, fully phosphorylated form of NF-H. Furthermore, the phosphorylation by cdc2 kinase dissociated the binding of dephosphorylated NF-H to microtubules. Phosphorylation sites were located at the carboxyl-terminal tail domain. The KSPXK motif, but not KSPXX, in the repetitive sequence was suggested to be the phosphorylation site by using synthetic peptides.     

Selective accumulation of the high molecular weight neurofilament subunit within the distal region of growing axonal neurites.

Axonal maturation in situ is accompanied by the transition of neurofilaments (NFs) comprised of only NF-M and NF-L to those also containing NF-H. Since NF-H participates in interactions of NFs with each other and with other cytoskeletal constituents, its appearance represents a critical event in the stabilization of axons that accompanies their maturation. Whether this transition is effected by replacement of "doublet" NFs with "triplet" NFs, or by incorporation of NF-H into existing doublet NFs is unclear. To address this issue, we examined the distribution of NF subunit immunoreactivity within axonal cytoskeletons of differentiated NB2a/d1 cell and DRG neurons between days 3-7 of outgrowth. Endogenous immunoreactivity either declined in a proximal-distal gradient or was relatively uniform along axons. This distribution was paralleled by microinjected biotinylated NF-L. By contrast, biotinylated NF-H displayed a bipolar distribution, with immunoreactivity concentrated within the proximal- and distal-most axonal regions. Proximal biotinylated NF-H accumulation paralleled that of endogenous NF immunoreactivity; however, distal-most biotinylated NF-H accumulation dramatically exceeded that of endogenous NFs and microinjected NF-L. This phenomenon was not due to co-polymerization of biotin-H with vimentin or alpha-internexin. This phenomenon declined with continued time in culture. These data suggest that NF-H can incorporate into existing cytoskeletal structures, and therefore suggest that this mechanism accounts for at least a portion of the accumulation of triplet NFs during axonal maturation. Selective NF-H accumulation into existing cytoskeletal structures within the distal-most region may provide de novo cytoskeletal stability for continued axon extension and/or stabilization.     

Phosphorylation-Dependent Immunoreactivity of Neurofilaments Increases During Axonal Maturation and β,β''-Iminodipropionitrile Intoxication

The immunoreactivity of the high-molecular-weight neurofilament (NF) subunit toward antibodies that react with phosphorylation-related epitopes was determined at different anatomic sites in the PNS of rats during normal maturation and after intoxication with beta,beta'-iminodipropionitrile (IDPN). A maturational increase in the relative binding of phosphorylation-dependent antibodies compared to phosphorylation-inhibited antibodies occurred from age 3 to 12 weeks. An increase in phosphorylation-related immunoreactivity with increasing distance from the cell bodies was present in ventral and dorsal roots at all ages. The degree of phosphorylation-related immunoreactivity was greater for centrally directed axons in the dorsal roots of the L5 ganglion than for peripherally directed axons. IDPN, a toxin that impairs NF transport, caused a marked increase in reactivity toward the phosphorylation-dependent antibody. NFs from IDPN-treated rats also bound less of an antibody that is normally phosphorylation independent and this inhibition of binding was sensitive to phosphatase digestion. In each instance, greater degrees of phosphorylation-dependent immunoreactivity correlate with conditions known to exhibit slower net rates of axonal transport of NF proteins.     

A Neurofilament-Associated Kinase Phosphorylates Only a Subset of Sites in the Tail of Chicken Midsize Neurofilament Protein

Although neurofilaments are among the most highly phosphorylated proteins extant, relatively little is known about the kinases involved in their phosphorylation. The majority of the phosphates present on the two higher-molecular-mass neurofilament subunits are added to multiply repeated sequence motifs in the tail. We have examined the specificity of a neurofilament-associated kinase (NFAK) partially purified from chicken spinal cord that selectively phosphorylates the middle-molecular-mass neurofilament subunit, NF-M. Two-dimensional phosphopeptide mapping of ³²P-labeled NF-M shows that, in vitro, NFAK phosphorylates a subset of peptides phosphorylated in vivo in cultured neurons. The absence of a complete complement of labeled phosphopeptides following in vitro phosphorylation, compared with phosphorylation in vivo, is not due to a lack of availability of phosphorylation sites because the same maps are obtained when enzymatically dephosphorylated NF-M is used as an in vitro substrate. Phosphopeptide maps from in vitro-phosphorylated NF-M and those from a recombinant fusion protein containing only a segment of the tail piece of chicken NF-M reveal identical labeled peptides. The fusion protein lacks a segment containing 17 KXX(S/T)P putative phosphorylation sites contained in the tail of chicken NF-M but contains a segment that includes four KSPs and a KSD site also present in the intact tail. These results suggest (a) that NFAK mediates the phosphorylation of some, but not all, potential phosphorylation sites within the tail of NF-M and (b) that multiple kinases are necessary for complete phosphorylation of the NF-M tail.     

Subunit composition of neurofilaments specifies axonal diameter

Neurofilaments (NFs), which are composed of NF-L, NF-M, and NF-H, are required for the development of normal axonal caliber, a property that in turn is a critical determinant of axonal conduction velocity. To investigate how each subunit contributes to the radial growth of axons, we used transgenic mice to alter the subunit composition of NFs. Increasing each NF subunit individually inhibits radial axonal growth, while increasing both NF-M and NF-H reduces growth even more severely. An increase in NF-L results in an increased filament number but reduced interfilament distance. Conversely, increasing NF-M, NF-H, or both reduces filament number, but does not alter nearest neighbor interfilament distance. Only a combined increase of NF-L with either NF- M or NF-H promotes radial axonal growth. These results demonstrate that both NF-M and NF-H play complementary roles with NF-L in determining normal axonal calibers.     

Structure of the peripheral domains of neurofilaments revealed by low angle rotary shadowing

The structure of the peripheral domains of neurofilaments (NFs) was revealed by rotary shadowing electron microscopy. NFs were isolated from bovine spinal cords by Sepharose CL-4B gel filtration and examined by low angle rotary shadowing. The peripheral domains appeared as thin, flexible, filamentous structures projecting from the intermediate filament core, with a constant density along their entire length. The average length of the projections was approximately 85 nm and the width about 4 nm. These projections appeared from regularly distributed sites, at 22 nm spacing, which seemed to correspond to the typical repeat of the alpha-helix-rich rod domain of the core filament. The density of the projections was found to be 4.1 (+/- 0.6) per 22 nm. We performed reconstitution experiments using purified NF polypeptides to confirm that the projection was indeed the NF peripheral domain. Individual components of the NF triplet, i.e. NF-L, NF-M and NF-H, were purified by DE-52 and Mono-Q anion exchange chromatographies in the presence of 6 M-urea and were assembled in various combinations into filaments. Reassembled filaments were somewhat more slender than the isolated NFs and exhibited a distinct 22 nm axial periodicity. While prominent projections were not observed in the filaments assembled from NF-L alone, reconstructed filaments containing NF-L plus either NF-M or NF-H revealed many projections. The average length of the projections in the filaments reconstructed from NF-L and NF-H was about 63 nm. The projections of reconstructed filaments from NF-L and NF-M were about 55 nm in length. The difference in the lengths of the projections might reflect the difference in the length of the carboxy-terminal tail domain between NF-M and NF-H. The results are interpreted to show that the carboxy-terminal tail domains of NFs project in a regular pattern from the core filament, which is consistent with a half-staggered organization of the tetrameric subunits.     

Increasing neurofilament subunit NF-M expression reduces axonal NF-H, inhibits radial growth, and results in neurofilamentous accumulation in motor neurons

The carboxy-terminal tail domains of neurofilament subunits neurofilament NF-M and NF-H have been postulated to be responsible for the modulation of axonal caliber. To test how subunit composition affects caliber, transgenic mice were generated to increase axonal NF- M. Total neurofilament subunit content in motor and sensory axons remained essentially unchanged, but increases in NF-M were offset by proportionate decreases in both NF-H and axonal cross-sectional area. Increase in NF-M did not affect the level of phosphorylation of NF-H. This indicates that (a) in vivo NF-H and NF-M compete either for coassembly with a limiting amount of NF-L or as substrates for axonal transport, and (b) NF-H abundance is a primary determinant of axonal caliber. Despite inhibition of radial growth, increase in NF-M and reduction in axonal NF-H did not affect nearest neighbor spacing between neurofilaments, indicating that cross-bridging between nearest neighbors does not play a crucial role in radial growth. Increase in NF- M did not result in an overt phenotype or neuronal loss, although filamentous swellings in perikarya and proximal axons of motor neurons were frequently found.