Brain Spectrin Research Articles

Divalent cation binding to erythrocyte spectrin

Erythrocyte spectrin dimers and separated alpha- and beta-spectrin chains bound 45Ca2+ after electrophoresis on native or sodium dodecyl sulfate-polyacrylamide gels, blotting, and 45Ca2+ overlay. Flow dialysis and equilibrium dialysis revealed two binding components: high-affinity, Ca(2+)-specific sites with kd = 4 x 10(-7) M and n = 100 +/- 20 per dimer and a low-affinity (millimolar) divalent cation component. Whereas brain spectrin had only four high-affinity sites [Wallis, C. J., Wenegieme, E. F., & Babitch, J. A. (1992) J. Biol. Chem. 267, 4333-4337], erythrocyte spectrin had 25-fold more sites per dimer. In addition to possibly modifying spectrin interactions with calcium-dependent protease and actin, as suggested by previous work on the interaction of Ca2+ with brain spectrin, the approximately two high-affinity sites per repeating segment of erythrocyte spectrin appear to stabilize a folded conformation of repeat structures indicated by an entropy increase upon binding. These data support the hypothesis that divalent cation binding to erythrocyte spectrin has become specialized to stabilize the membrane skeletal network and the cell, making them flexible but resistant to shear under the stressful conditions of blood circulation.

Solubility of cytoskeletal proteins in immunohistochemistry and the influence of fixation.

For accurate and quantitative immunohistochemical localization of antigens it is crucial to know the solubility of tissue proteins and their degree of loss during processing. In this study we focused on the solubility of several cytoskeletal proteins in cat brain tissue at various ages and their loss during immunohistochemical procedures. We further examined whether fixation affected either solubility or immunocytochemical detectability of several cytoskeletal proteins. An assay was designed to measure the solubility of cytoskeletal proteins in cryostat sections. Quantity and quality of proteins lost or remaining in tissue were measured and analyzed by electrophoresis and immunoblots. Most microtubule proteins were found to be soluble in unfixed and alcohol fixed tissues. Furthermore, the microtubule proteins remaining in the tissue had a changed cellular distribution. In contrast, brain spectrin and all three neurofilament subunits were insoluble and remained in the tissue, allowing their immunocytochemical localization in alcohol-fixed tissue. Synapsin I, a protein associated with the spectrin cytoskeleton, was soluble, and aldehyde fixation is advised for its immunohistochemical localization. With aldehyde fixation, the immunoreactivity of some antibodies against neurofilament proteins was reduced in axons unveiling novel immunogenic sites in nuclei that may represent artifacts of fixation. In conclusion, protein solubility and the effects of fixation are influential factors in cytoskeletal immunohistochemistry, and should be considered before assessments for a quantitative distribution are made.

The complete amino acid sequence for brain β spectrin (β fodrin): relationship to globin sequences

The amino acid sequence of mouse brain β spectrin (β fodrin), deduced from the nucleotide sequence of complementary DNA clones, reveals that this non-erythroid β spectrin comprises 2363 residues, with a molecular weight of 274,449 Da. Brain β spectrin contains three structural domains and we suggest the position of several functional domains including f-actin, synapsin I, ankyrin and spectrin self association sites. Analysis of deduced amino acid sequences indicated striking homology and similar structural characteristics of brain β spectrin repeats β11 and β12 to globins. In vitro analysis has demonstrated that heme is capable of specific attachment to brain spectrin, suggesting possible new functions in electron transfer, oxygen binding, nitric oxide binding or heme scavenging.

A dystrophin-immunoreactive protein in mammalian brain.

Dystrophin is expressed only in muscle and brain, but is absent from all tissues of the adult mdx mouse, a mutant with a single base substitution in the dystrophin gene. The brains of both normal and mdx mice contain a protein of approximately 230 kDa that is recognised by anti-dystrophin antibodies raised to the N-terminal region of the rod-like domain. Although the N-terminal and central rod regions of dystrophin share structural homologies with spectrin, the 230-kDa protein represents neither of the presently described forms of brain spectrin by a variety of criteria (molecular weight, cerebellar localisation, and developmental regulation) and is distinct from the product of the dystrophin gene. Studies of mdx and normal mouse brain show different postnatal developmental regulation of the 230-kDa dystrophin-immunoreactive protein.

Reorganization of brain spectrin (fodrin) during differentiation of PC12 cells

Fodrin has been shown to redistribute dynamically between cytoplasmic and plasma membrane-associated compartments upon the differentiation of T lymphocytes. We studied the changes of distribution of fodrin in PC12 cells upon neuronal differentiation induced by nerve growth factor. To visualize preferentially the elements that were tightly associated with cytoskeletal structures, we performed immunofluorescence and immunoelectron microscopy on saponin-extracted cells. In undifferentiated PC12 cells, fodrin was distributed mostly underneath the plasma membrane. However, after the administration of nerve growth factor, perinuclear spot-like aggregates of fodrin appeared. Double-labeling immunofluorescence revealed that the cytoplasmic fodrin spot was co-localized with the intermediate filament proteins, peripherin and neurofilament. Immunogold electron microscopy showed that fodrin and neurofilament were localized in close association in the perinuclear regions enriched with intermediate filaments. With prolonged exposure to nerve growth factor, fodrin and intermediate filaments spread to the cytoplasm and neurites. These results suggest that there is a dynamic reorganization of fodrin during differentiation of PC12 cells, and that fodrin is first recruited in the perinuclear region closely associated with intermediate filaments. This dynamic reorganization of fodrin may represent important, previously unrecognized aspects of the morphological differentiation of neurons.

Developmental expression of brain β-spectrin isoform messenger RNAs

We have investigated the expression of brain βSpIIa and βSpIb (previously referred to as the β-subunits of brain spectrin (240/235) and brain spectrin (240/235E), respectively) during mouse brain development. The 9 kb transcript which encodes βSpIIa is present in fetal mouse brain tissue and increases to a maximal level in a 30-day-old mouse. There is a coordinate accumulation of the 7.8 kb αSpIIa mRNA (with βSpIIa) during mouse brain development. The coordinate expression of αSpIIa and βSpIIa at the mRNA and protein level allows formation of (αSpIIa/βSpIIa) 2 tetramers (brain spectrin(240/235)) early in premitotic neuronal development; and avoids turnover of unassembled α and β-subunits. An 11 kb transcript which encodes βSpIb is not produced in embryonic tissue, and is first seen in a 6-day-old mouse. The protein translation products βSpIIa and βSpIb have previously been demonstrated by our laboratory to first appear in fetal mouse brain tissue and at postnatal day 6–8, respectively [ J. Neurosci., 7 (1987) 864–874]. The expression of βSpIb mRNA on postnatal day 6–8, and the appearance of brain spectrin(240/235E) in postmitotic and postmigratory neurons of the cerebellum at this same time; suggests that brain spectrin(240/235E) is involved in differentiated functions of the neuron (formation of cell-cell contacts, formation of dendritic processes and postsynaptic contacts). Thus, the data from the present study demonstrates that the expression of these two neuronal β-spectrin isoforms is regulated at the level of mRNA expression.

Identification of an amelin isoform located in axons

A new axonal isoform of amelin, an analogue of the erythrocyte spectrin binding protein termed protein 4.1 has been identified in mouse brain. This new isoform has a molecular weight of 93 kDa, and migrates to a more acidic pH (pH 7.5–8.0) than the previously described amelin E (pH 8.5) on two dimensional NEPHGE-SDS PAGE. The 93 kDa protein looks nearly identical to amelin E on two dimensional chymotryptic iodopeptide mapping, and both share partial homology with rbc protein 4.1. The new isoform is located in axons, and the soma of neurons in mouse cerebellum, while amelin E is located in neuronal soma and dendrites. The axonal amelin antibody detects a 97 kDa protein in embryonic tissue which diminishes during development; and a 93 kDa protein which is first seen at postnatal day 1 of mouse brain ontogeny, increasing constantly to its adult concentrations. This time course of expression is quite different than amelin E, which is present at embryonic day 15 and diminishes constantly reaching its lowest concentration in the adult brain. We hypothesize that axonal amelin and amelin E may play important roles in the interaction of brain spectrin(240/235) and brain spectrin (240/235E) with f-actin and neuronal membranes.

Dystrophin immunoreactivity in normal and Duchenne human fetal neurons in culture.

Dystrophin, the protein product defective in Duchenne muscular dystrophy (DMD), is present in all types of muscle and in the brain. The function of the protein is unknown and its role in the brain is unclear, although 30% of DMD patients show nonprogressive mental retardation. We have therefore studied the localisation of dystrophin in cultures of normal and DMD human fetal neurons using antibodies raised to different regions of the protein. Dystrophin immunoreactivity was demonstrated in the soma and axon hillock of normal neurons and appeared to be associated with the inner part of the cell membrane, although some intracellular staining was also observed. Positive dystrophin staining was present only in cells with fully developed neuronal features, although not all the neurons were positive. Glial cells were always negative for the antigen. Immunostaining with antibodies to the brain spectrins indicate that the dystrophin antibodies did not crossreact with these proteins. The possibility of cross-reactivity with other proteins is discussed. Studies of cells cultured from a DMD fetus also showed specific dystrophin immunostaining in neurons, although the muscle was generally negative for dystrophin. However, the localisation of dystrophin immunostaining and that of the brain spectrins and neurofilaments appeared abnormal, as did the overall morphology of the cells. This suggests that dystrophin may play a role during brain development and dystrophin deficiency results in abnormal neuronal features. This would be consistent with the nonprogressive nature of the mental retardation observed in DMD patients.

Actin and neurofilament binding domain of brain spectrin beta subunit.

Tryptic digestion of brain spectrin generates a number of fragments from alpha and beta subunits; when these fragments are incubated with F-actin or neurofilament light subunit, four of them with molecular masses below 30 kDa sediment with the cytoskeleton structures. A selective purification of these fragments by ammonium sulfate fractionation and butyl-Sepharose chromatography has been achieved. Two fragments with molecular masses of 28 and 23 kDa bind to F-actin. Native brain spectrin causes half-maximal inhibition of the association at a concentration of 3 microM. Protein sequencing indicates that the actin-binding domain is contained in the beta subunit, in a stretch of amino acids at the N terminus from Ala43 (28-kDa fragment) or from Met104 (23-kDa fragment) and terminate probably at the C-terminal Lys288 or Lys284. Amino acids are numbered by reference to the sequence of the Drosophila beta subunit. The 28-kDa fragment also binds to the low-molecular-mass subunit of neurofilaments; brain spectrin heterodimer disrupts this binding. Hence, spectrin binds to F-actin and to neurofilaments via a common binding domain.

Characterization of calcium binding to brain spectrin.

Brain spectrin alpha and beta chains bind 45Ca2+, as shown by the calcium overlay method. Flow dialysis measurements revealed eight high affinity binding sites/tetramer that comprise two binding components (determined by nonlinear regression analysis). The first component has one or two sites (kd = 2-30 x 10(-8) M), depending on the ionic strength of the binding buffer, with the remaining high affinity sites in the second component (kd = 1-3 x 10(-6) M). In addition, there is a variable, low affinity binding component (n = 100-400, kd = 1-2 x 10(-4) M). Magnesium inhibits calcium binding to the low affinity sites with a K1 = 1.21 mM. Proteolytic fragments from trypsin or chymotrypsin digests of brain spectrin bind 45Ca2+ if they include alpha domain IV, alpha domain III, or the amino-terminal half of the beta chain (but more than 25 kDa from the amino-terminal). These data suggest that calcium ions bind with high affinity to the putative EF-hands in alpha domain IV and to one site in the amino-terminal half of the beta chain that is associated with alpha domain IV in the native dimer. The localization is consistent with a direct calcium modulation of the spectrin-actin-protein 4.1 interaction. In addition, there appears to be one high affinity site near the hypersensitive region of alpha brain spectrin. All four proposed binding sites occur near probable calmodulin-binding or calcium-dependent protease cleavage sites.

Development and Maintenance of the Neuronal Cytoskeleton in Aggregated Cell Cultures of Fetal Rat Telencephalon and Influence of Elevated K+ Concentrations

Serum-free aggregating cell cultures of fetal rat telencephalon were examined by biochemical and immunocytochemical methods for their development-dependent expression of several cytoskeletal proteins, including the heavy- and medium-sized neurofilament subunits (H-NF and M-NF, respectively); brain spectrin; synapsin I; beta-tubulin; and the microtubule-associated proteins (MAPs) 1, 2, and 5 and tau protein. It was found that with time in culture the levels of most of these cytoskeletal proteins increased greatly, with the exceptions of the particular beta-tubulin form studied, which remained unchanged, and MAP 5, which greatly decreased. Among the neurofilament proteins, expression of M-NF preceded that of H-NF, with the latter being detectable only after approximately 3 weeks in culture. Furthermore, MAP 2 and tau protein showed a development-dependent change in expression from the juvenile toward the adult form. The comparison of these developmental changes in cytoskeletal protein levels with those observed in rat brain tissue revealed that protein expression in aggregate cultures is nearly identical to that in vivo during maturation of the neuronal cytoskeleton. Aggregate cultures deprived of glial cells, i.e., neuron-enriched cultures prepared by treating early cultures with the antimitotic drug cytosine arabinoside, exhibited pronounced deficits in M-NF, H-NF, MAP 2, MAP 1, synapsin I, and brain spectrin, with increased levels of a 145-kDa brain spectrin breakdown product. These adverse effects of glial cell deprivation could be reversed by the maintenance of neuron-enriched cultures at elevated concentrations of KCl (30 mM). This chronic treatment had to be started at an early developmental stage to be effective, a finding suggesting that sustained depolarization by KCl is able to enhance the developmental expression and maturation of the neuronal cytoskeleton.

CAMP and Ca(2+)-mediated secretion in parotid acinar cells is associated with reversible changes in the organization of the cytoskeleton.

The potential involvement of actin and fodrin (brain spectrin) in secretory events has been assessed in primary cultured guinea pig parotid acinar cells, using as a tool affinity purified anti-alpha-fodrin antibody, phalloidin, and immunofluorescence techniques. In resting parotid acinar cells fodrin and actin appeared as a continuous ring under the plasma membrane of most of the cells. Upon stimulation with secretagogues fodrin and actin labeling at the level of the plasma membrane disappeared almost completely. To establish a correlation between secretion and cytoskeletal changes at the individual cell level, anti-alpha-amylase-antibodies were used to label secreted amylase exposed at the surface of secreting cells. The number of cells expressing alpha-amylase on their surface followed bulk secretion of alpha-amylase. A strict correlation between secretion and alteration of the actin-fodrin labeling was observed at the individual cell level. The cytoskeletal changes occurred in parallel with secretion independently of the secretagogue used (carbamoylcholine in the presence of Ca2+, isoproterenol in presence or absence of Ca2+, forskolin, or dibutyryl-cyclic-AMP). The changes were reversible upon removal of the secretagogue. Since Ca2+, as well as cAMP-mediated secretion, was associated with the same kind of cytoskeletal changes, a reorganization of the cytoskeleton may play an essential part in regulated secretion.

Actin-binding and microtubule-associated proteins in the organ of Corti

Actin-binding and microtubule-associated proteins regulate microfilament and microtubule number, length, organization and location in cells. In freeze-dried preparations of the guinea pig cochlea, both actin and tubulin are found in the sensory and supporting cells of the organ of Corti. Fodrin (brain spectrin) co-localized with actin in the cuticular plates of both inner and outer hair cells and along the lateral wall of the outer hair cells. Alpha-actinin co-localized with actin in the cuticular plates of the hair cells and in the head and foot plates of the supporting cells. It was also found in the junctional regions between hair cells and supporting cells. Profilin co-localized with actin in the cuticular plates of the sensory hair cells. Myosin was detected only in the cuticular plates of the outer hair cells and in the supporting cells in the region facing endolymph. Gelsolin was found in the region of the nerve fibers. Tubulin is found in microtubules in all cells of the organ of Corti. In supporting cells, microtubules are bundled together with actin microfilaments and tropomyosin, as well as being present as individual microtubules arranged in networks. An intensely stained network of microtubules is found in both outer and inner sensory hair cells. The microtubules in the outer hair cells appear to course throughout the entire length of the cells, and based on their staining with antibodies to the tyrosinated form of tubulin they appear to be more dynamic structures than the microtubules in the supporting cells. The microtubule-associated protein MAP-2 is present only in outer hair cells within the organ of Corti and co-localizes with tubulin in these cells. No other MAPs (1,3,4,5) are present. Tau is found in the nerve fibers below both inner and outer hair cells and in the osseous spiral lamina. It is clear that the actin-binding and microtubule-associated proteins present in the cochlea co-localize with actin and tubulin and that they modulate microfilament and microtubule structure and function in a manner similar to that seen in other cell types. The location of some of these proteins in outer hair cells suggests a role for microfilaments and microtubules in outer hair cell motility.

Spectrin isoforms in the mammalian retina

Spectrin is a major component of the mammalian neuronal cytoskeleton. In the CNS, three isoforms of brain spectrin are known to exist: a cellular and dendritic isoform, (240/235E), related to neurons and glia; a cellular and axonal isoform, (240/235), related to neurons; and an isoform specific for astrocytes, (240/235A). In the present study, brain spectrins (240/235E) and (240/235) were localized within the mouse retina and optic nerve. Immunoblot analyses of proteins isolated from mouse retinas utilizing polyclonal antibodies to either brain spectrin (240/235) or brain spectrin (240/235E) revealed that these spectrins are present in the retina and that the two isoforms are the same molecular weights as those found in the brain. Immunocytochemical studies revealed that spectrin (240/235E) was localized in cell bodies of the inner nuclear, outer nuclear, and ganglion cell layers, and processes arborizing within the inner and outer plexiform layers. Spectrin (240/235) was distributed diffusely within the retina, lightly staining neurons in both the inner nuclear and outer nuclear layers, and the ganglion cell layer. In contrast to the situation found in the brain, spectrin (240/235) was but one of the axonal forms in the retina. We found that spectrin (240/235E) was also present in the axon-rich fiber layer and in the optic nerve and was often associated with fibrous elements. Spectrin (240/235) was also detected in the nerve fiber layer and optic nerve, but this isoform was not localized to fibers.(ABSTRACT TRUNCATED AT 250 WORDS)

Spectrin breakdown products increase with age in telencephalon of mouse brain

Calcium activated proteolysis of brain spectrin produces characteristic breakdown products (BDPs), the concentrations of which increase markedly in many instances of brain pathology. Results reported here indicate that levels of the BDPs rise with age (3–30 months) in the telencephalon but not in the hindbrain of Balb/c mice. These observations suggest that spectrin breakdown is a pathologic biochemical marker which increases with age in some but not all brain regions.

In vitro proteolysis of brain spectrin by calpain I inhibits association of spectrin with ankyrin-independent membrane binding site(s).

This report demonstrates that specific proteolysis of brain spectrin by a calcium-dependent protease, calpain I, abolishes association of brain spectrin with the ankyrin-independent binding site(s) in brain membranes. Calpain I cleaves the beta subunit of spectrin at the N-terminal end leaving a 218-kDa fragment and cleaves the alpha subunit in the midregion to produce 150- and 130-kDa fragments. Calpain-proteolyzed spectrin almost completely loses the capacity to displace binding of intact spectrin to membranes. Spectrin digested by calpain I under conditions that almost completely destroyed membrane-binding remained associated as a tetramer and retained about 60% of the ability to associate with actin filaments. Cleavage of spectrin occurred at sites distinct from the membrane-binding site which is located on the beta subunit since the isolated 218-kDa fragment of the beta subunit as well as a reconstituted complex of alpha and 218-kDa beta subunit fragment partially regained binding activity. Moreover, cleavage of the alpha subunit alone reduced the affinity of spectrin for membranes by 2-fold. A consequence of distinct sites for calpain I cleavage and membrane-binding is that calpain I can digest spectrin while spectrin is complexed with other proteins and therefore has the potential to mediate disassembly of a spectrin-actin network from membranes.

The effect of synapsin I phosphorylation upon binding of synaptic vesicles to spectrin

We have previously demonstrated that brain spectrin is attached to small spherical synaptic vesicles via synapsin I. These studies utilized a novel microfiltration assay in which 125I-labelled synaptic vesicles were incubated with brain spectrin which was covalently attached to cellulosic membranes. In these studies purified dephosphosynapsin I was demonstrated to competitively inhibit the binding of the synaptic vesicles to the immobilized brain spectrin with a K I = 45 nM. In the current study we demonstrate that phosphorylation of synapsin I site 1 (0.74 mol P i/mol synapsin I) with cAMP-dependent protein kinase and sites 2 and 3 (2.0 mol P i/mol synapsin I) with Ca 2+-calmodulin kinase II had little effect upon its interaction with brain spectrin. cAMP-dependent protein kinase phosphorylated synapsin I and Ca 2+-calmodulin kinase II phosphorylated synapsin I both inhibited the binding of 125I-labelled synaptic vesicles to immobilized brain spectrin with a K I of 23 nM and 34 nM respectively. We conclude that phosphorylation of synapsin I does not down-regulate the interaction of synaptic vesicles with brain spectrin.

Abnormal brain spectrin immunoreactivity in sprouting neurons in Alzheimer disease

Brain spectrin is a major membrane skeleton protein that participates in cellular transport, cell morphogenesis, neurotransmitter release and growth cone adhesion. The present study showed that in Alzheimer disease (AD) neuropil, brain spectrin immunoreactivity is co-localized with synaptophysin in the presynaptic boutons. At the ultrastructural level, brain spectrin immunoreactivity was observed in the presynaptic terminals and in the axoplasm of some myelinated and unmyelinated fibers. In addition to this normal localization of brain spectrin in the AD brain, we also found brain spectrin immunoreactivity associated with abnormal patchy lesions in the AD neuropil. Confocal laser imaging and immunoelectron microscopy revealed that these lesions corresponded to thick cellular processes derived from neurons. The findings that these structures were anti-neurofilament positive but anti-glial fibrillary acidic protein (GFAP) and Ricinus communis agglutinin I (RCA-I) negative confirm their neuronal origin, and rule out the possibility of glial origin. These structures could represent either atypical axonal or dendritic processes derived from sprouting neurons or the accumulation of brain spectrin degradation products in degenerating neurons.

Purification and Properties of p 103, a Novel 103-kDa Component of Postsynaptic Densities

A 103-kDa protein present in membrane cytoskeletal preparations from bovine brain has been identified. We have purified this protein to greater than 95% homogeneity using gel filtration and ion-exchange chromatography. This protein, p103, is an asymmetric dimer in dilute solution and has two major variants that can be distinguished by isoelectric focussing, pI 5.60 and 5.75. Using subcellular fractionation, it is most enriched in postsynaptic densities. Immunolocalization with anti-p103-specific antibodies reveals that it is confined to the dendrites and perikarya; it is apparently absent from spinal cord axons. It coextracts from brain membrane-skeletal preparations with brain spectrin and actin, but in vitro, it does not interact with them.

Identification of a mouse brain β-spectrin cDNA and distribution of its mRNA in adult tissues

A mouse brain β-spectrin of cDNA was identified within a λGt11 expression library using an antibody which specifically binds with the 235 kDa spectrin β-subunit. Restriction mapping and DNA sequencing analyses of the brain cDNA revealed that this clone contained 1185 bp of sequence, of which a 999 bp single open reading frame encoding 333 amino acids was determined. The deduced amino acid sequence exhibited homology with β-spectrins, demonstrating the characteristic 106 amino acid repeating unit. The homology between our mouse brain sequence and human RBC β-spectrin was ~56% beginning at the β15 repeat unit and extending to the C-terminus of sequence elucidated for human RBC sequence. An additional 62 amino acids were found at the C-terminus of the 235 kDa brain β-spectrin subunit not seen in the human RBC sequence. The ~1.2 Kb brain spectrin cDNA insert hybridized with a single 9 Kb mRNA transcript in various adult mouse tissues, with the most abundant hybridization demonstrated in RNA isolated from brain tissue. This mRNA was found to be present at high levels in heart tissue and at lower levels in spleen and skeletal muscle tissue. The 9 Kb mRNA was different in content and in size to mRNAs which hybridized with a cDNA encoding the mouse erythroid β-spectrin subunit, demonstrating that the brain spectrin cDNA is a distinct gene product and represents the first known sequence of a nonerythroid β-spectrin subunit.