Ankyrin-independent membrane protein-binding sites for brain and erythrocyte spectrin

Brain spectrin reassociates in in vitro binding assays with protein(s) in highly extracted brain membranes quantitatively depleted of ankyrin and spectrin. These newly described membrane sites for spectrin are biologically significant and involve a protein since (a) binding occurs optimally at physiological pH (6.7-6.9) and salt concentrations (50 mM), (b) binding is abolished by digestion of membranes with alpha-chymotrypsin, (c) Scatchard analysis is consistent with a binding capacity of at least 50 pmol/mg total membrane protein, and highest affinity of 3 nM. The major ankyrin-independent binding activity of brain spectrin is localized to the beta subunit of spectrin. Brain membranes also contain high affinity binding sites for erythrocyte spectrin, but a 3-4 fold lower capacity than for brain spectrin. Some spectrin-binding sites associate preferentially with brain spectrin, some with erythrocyte spectrin, and some associate with both types of spectrin. Erythrocyte spectrin contains distinct binding domains for ankyrin and brain membrane protein sites, since the Mr = 72,000 spectrin-binding fragment of ankyrin does not compete for binding of spectrin to brain membranes. Spectrin binds to a small number of ankyrin-independent sites in erythrocyte membranes present in about 10,000-15,000 copies/cell or 10% of the number of sites for ankyrin. Brain spectrin binds to these sites better than erythrocyte spectrin suggesting that erythrocytes have residual binding sites for nonerythroid spectrin. Ankyrin-independent-binding proteins that selectively bind to certain isoforms of spectrin provide a potentially important flexibility in cellular localization and time of synthesis of proteins involved in spectrin-membrane interactions. This flexibility has implications for assembly of the membrane skeleton and targeting of spectrin isoforms to specialized regions of cells.     

Spectrin subtypes in mammalian brain

Mammalian neural cells contain at least two forms of brain spectrin: brain spectrin (240/235) which is located primarily in the axons and presynaptic terminals of neurons, and brain spectrin (240/235E) which is found in the cell bodies, dendrites and postsynaptic terminals of neurones. Brain spectrin (240/235E) is also found in certain glial cell types. Antibodies against red blood cell spectrin detect only brain spectrin (240/235E), while antibodies against brain spectrin isolated from axonal and synaptic membranes detect brain spectrin (240/235). Previous apparent discrepancies in the literature concerning brain spectrin localization at the light microscope level were undoubtedly due to different laboratories detecting distinct brain spectrin subtypes, based on the particular antibody being utilized for immunohistochemistry. In this review we (1) discuss the data supporting the presence of at least two distinct subtypes of mammalian brain spectrin, (2) explain how these results reconcile previous discrepancies concerning the localization of spectrin within neural cells, and (3) suggest the future implications of these findings.     

A cyclic AMP-dependent phosphorylation of spectrin dimer

In contrast to the properties of spectrin obtained from [³²P]phosphate-labeled red cells, purified spectrin dimer could be phosphorylated by a cAMP-dependent protein kinase from bovine heart. Both spectrin bands were phosphorylated. Spectrin band 2 contained in addition to autophosphorylated peptides several phosphopeptides that were distinct from autophosphorylated ones. The cAMP-dependent phosphorylation of spectrin band 1 was modulated by reducing agent and the concentration of spectrin. At high concentrations spectrin band 2 was predominantly labeled. The cAMP-dependent phosphoform of spectrin band 2 had a pI slightly higher than that of autophosphorylated spectrin band 2, but lower than that of ankyrin.     

Sequential degradation of alphaII and betaII spectrin by calpain in glutamate or maitotoxin-stimulated cells

Calpain-catalyzed proteolysis of II-spectrin is a regulated event associated with neuronal long-term potentiation, platelet and leukocyte activation, and other processes. Calpain proteolysis is also linked to apoptotic and nonapoptotic cell death following excessive glutamate exposure, hypoxia, HIV-gp120/160 exposure, or toxic injury. The molecular basis for these divergent consequences of calpain action, and their relationship to spectrin proteolysis, is unclear. Calpain preferentially cleaves II spectrin in vitro in repeat 11 between residues Y1176 and G1177. Unless stimulated by Ca++ and calmodulin (CaM), betaII spectrin proteolysis in vitro is much slower. We identify additional unrecognized sites in spectrin targeted by calpain in vitro and in vivo. Bound CaM induces a second II spectrin cleavage at G1230*S1231. BetaII spectrin is cleaved at four sites. One cleavage only occurs in the absence of CaM at high enzyme-to-substrate ratios near the betaII spectrin COOH-terminus. CaM promotes II spectrin cleavages at Q1440*S1441, S1447*Q1448, and L1482*A1483. These sites are also cleaved in the absence of CaM in recombinant II spectrin fusion peptides, indicating that they are probably shielded in the spectrin heterotetramer and become exposed only after CaM binds alphaII spectrin. Using epitope-specific antibodies prepared to the calpain cleavage sites in both alphaII and betaII spectrin, we find in cultured rat cortical neurons that brief glutamate exposure (a physiologic ligand) rapidly stimulates alphaII spectrin cleavage only at Y1176*G1177, while II spectrin remains intact. In cultured SH-SY5Y cells that lack an NMDA receptor, glutamate is without effect. Conversely, when stimulated by calcium influx (via maitotoxin), there is rapid and sequential cleavage of alphaII and then betaII spectrin, coinciding with the onset of nonapoptotic cell death. These results identify (i) novel calpain target sites in both alphaII and betaII spectrin; (ii) trans-regulation of proteolytic susceptibility between the spectrin subunits in vivo; and (iii) the preferential cleavage of alphaII spectrin vs betaII spectrin when responsive cells are stimulated by engagement of the NMDA receptor. We postulate that calpain proteolysis of spectrin can activate two physiologically distinct responses: one that enhances skeletal plasticity without destroying the spectrin-actin skeleton, characterized by preservation of betaII spectrin; or an alternative response closely correlated with nonapoptotic cell death and characterized by proteolysis of betaII spectrin and complete dissolution of the spectrin skeleton.     

Some viscoelastic properties of human erythrocyte spectrin networks end-linked in vitro

We have succeeded in making macroscopic networks of end-linked human erythrocyte spectrin. The network junctions were made using erythrocyte protein 4.1 irreversibly attached to 5 nm (diameter) colloidal gold particles. Rotary shadowing electron microscopy verifies that the protein 4.1-labelled colloidal gold particles bind only to the tail end of the spectrin molecules. Electron micrographs of protein 4.1-labelled colloidal gold particles incubated at 4 degrees C with spectrin dimers reveal that 1-5 spectrin dimers attach to each protein 4.1-labelled colloidal gold particle yielding a spider-like appearance of these complexes. Incubation with a low concentration of spectrin tetramers instead of dimers leads to extensive formation of spectrin microaggregates whereas use of spectrin concentrations higher than 3 mg/ml and a molar ratio between spectrin tetramers and protein 4.1/Au of 4 leads to formation of macroscopic spectrin networks. We have quantitated the viscoelastic properties of such end-linked macroscopic spectrin networks using a gravitational pendulum viscoelastometer. We find that in vitro end-linked spectrin networks can be described by linear viscoelastic theory. The dynamic storage modulus increases almost linearly with the spectrin-protein 4.1/gold particle concentration when the spectrin concentration exceeds about 3 mg/ml and the molar ratio between spectrin tetramers and protein 4.1/Au is 4. At a spectrin concentration of 6 mg/ml and the same ratio between spectrin and protein 4.1/Au, we find a dynamic storage modulus at low frequency of about 80 dyn/cm2. This is in adequate agreement with what is predicted by simple elastomer theory.     

Caenorhabditis elegans beta-G spectrin is dispensable for establishment of epithelial polarity, but essential for muscular and neuronal function

The Caenorhabditis elegans genome encodes one alpha spectrin subunit, a beta spectrin subunit (beta-G), and a beta-H spectrin subunit. Our experiments show that the phenotype resulting from the loss of the C. elegans alpha spectrin is reproduced by tandem depletion of both beta-G and beta-H spectrins. We propose that alpha spectrin combines with the beta-G and beta-H subunits to form alpha/beta-G and alpha/beta-H heteromers that perform the entire repertoire of spectrin function in the nematode. The expression patterns of nematode beta-G spectrin and vertebrate beta spectrins exhibit three striking parallels including: (1) beta spectrins are associated with the sites of cell-cell contact in epithelial tissues; (2) the highest levels of beta-G spectrin occur in the nervous system; and (3) beta spectrin-G in striated muscle is associated with points of attachment of the myofilament apparatus to adjacent cells. Nematode beta-G spectrin associates with plasma membranes at sites of cell-cell contact, beginning at the two-cell stage, and with a dramatic increase in intensity after gastrulation when most cell proliferation has been completed. Strikingly, depletion of nematode beta-G spectrin by RNA-mediated interference to undetectable levels does not affect the establishment of structural and functional polarity in epidermis and intestine. Contrary to recent speculation, beta-G spectrin is not associated with internal membranes and depletion of beta-G spectrin was not associated with any detectable defects in secretion. Instead beta-G spectrin-deficient nematodes arrest as early larvae with progressive defects in the musculature and nervous system. Therefore, C. elegans beta-G spectrin is required for normal muscle and neuron function, but is dispensable for embryonic elongation and establishment of early epithelial polarity. We hypothesize that heteromeric spectrin evolved in metazoans in response to the needs of cells in the context of mechanically integrated tissues that can withstand the rigors imposed by an active organism.     

Mechanisms of cytoskeletal regulation: functional and antigenic diversity in human erythrocyte and brain beta spectrin

A study of human erythrocyte and brain spectrin with particular emphasis on the beta subunits revealed a structural homology but functional dissimilarity between these two molecules. Six monoclonal antibodies raised to human erythrocyte beta spectrin identify three of the four proteolytically defined domains of erythrocyte beta spectrin. Five of these monoclonal antibodies cross-react with human brain spectrin. None of a previously identified set of alpha erythrocyte spectrin monoclonal antibodies [Yurchenco et al: J Biol Chem 257:9102, 1982] reacted with brain spectrin. A domain map generated by limited tryptic digestion shows that brain spectrin is composed of proteolytically resistant domains analogous to erythrocyte spectrin, but the brain protein is more basic. The binding of brain spectrin to erythrocyte ankyrin, both in solution and on erythrocyte IOVs, yielded an association constant approximately 100 time weaker than for erythrocyte spectrin. The binding of azido-calmodulin under native conditions was specific for the erythrocyte beta subunit but was not calcium dependent. In contrast, azido-calmodulin bound only to the alpha subunit of brain spectrin in a calcium-dependent manner. The similarity of structure but modified functional characteristics of the brain and erythrocyte beta spectrins suggest that these proteins serve different cellular roles.     

Expression and assembly of the erythroid membrane-skeletal proteins ankyrin (goblin) and spectrin in the morphogenesis of chicken neurons

The membrane-skeleton of adult chicken neurons in the cerebellum and optic system is composed of polypeptides structurally and functionally related to the erythroid proteins spectrin and ankyrin, respectively. Neuronal spectrin comprises two distinct complexes that share a common alpha subunit (Mr 240,000) but which have structurally distinct polymorphic subunits (beta' beta spectrin; Mr 220/225,000; gamma spectrin, Mr 235,000); the brain-specific form (alpha gamma spectrin or fodrin) and an erythrocyte-specific form (alpha beta' beta spectrin). Two structurally related isoforms of ankyrin have also been identified and are termed alpha (Mr 260,000) and beta (Mr 237,000) ankyrin. Immunofluorescence demonstrates that the variants of spectrin and ankyrin, respectively, have different distributions within neurons. On the one hand, alpha gamma spectrin and beta ankyrin are present throughout the neuron, in the perikaryon, dendrites, and axon, whereas alpha beta' spectrin and alpha ankyrin are localized exclusively in the perikaryon and dendrites where they are actively segregated from alpha gamma spectrin and other components of axonal transport. This asymmetric distribution of spectrin and ankyrin isoforms is established in distinct stages during neuronal morphogenesis. Early in cerebellar and retinal development, alpha gamma spectrin is expressed in mitotic cells. Subsequently beta ankyrin and alpha gamma spectrin are coexpressed in postmitotic cells and gradually accumulate on the plasma membrane in a uniform pattern throughout the neuron during the phase of cell growth. At the onset of synaptogenesis and the cessation of cell growth, their levels of synthesis decline sharply while the assembled proteins remained as stable membrane components. Concomitantly, there is a dramatic induction in the accumulation of alpha ankyrin and alpha beta' spectrin, whose assembly is limited to the plasma membrane of the perikarya and dendrites. These results demonstrate that two successive, developmentally regulated programs of ankyrin and spectrin expression and patterning on the plasma membrane are involved in the assembly of the spectrin-based asymmetry in the neuronal membrane-skeleton, and that their asymmetric distribution is actively maintained throughout the life of the neuron.     

Cell type-specific association between two types of spectrin and two types of intermediate filaments

We have demonstrated a differential association between two types of spectrin, from erythrocytes and brain, with two types of intermediate filaments, vimentin filaments and neurofilaments. Electron microscopy showed that erythrocyte spectrin promoted the binding of vimentin filaments to red cell inside-out vesicles via lateral associations with the filaments. In vitro binding studies showed that the association of spectrin with vimentin filaments was apparently saturable, increased with temperature, and could be prevented by heat denaturation of the spectrin. Comparisons were made between erythrocyte and brain spectrin binding to both vimentin filaments and neurofilaments. We found that vimentin filaments bound more erythrocyte spectrin than brain spectrin, while neurofilaments bound more brain spectrin than erythrocyte spectrin. Our results show that both erythroid and nonerythroid spectrins are capable of binding to intermediate filaments and that such associations may be characterized by differential affinities of the various types of spectrin with the several classes of intermediate filaments present in cells. Our results also suggest a role for both erythroid and nonerythroid spectrins in mediating the association of intermediate filaments with plasma membranes or other cytoskeletal elements.     

Immunoprecipitation of nonerythrocyte spectrin within live cells following microinjection of specific antibodies: relation to cytoskeletal structures

The intracellular precipitation of nonerythrocyte spectrin has been achieved by the microinjection into cells of either a monoclonal antibody (IgM) directed against the alpha chain of nonerythrocyte spectrin or an affinity-purified polyclonal antibody raised against bovine brain spectrin (fodrin). This antibody-induced precipitation of spectrin was observed in fibroblastic and epithelial cell types, including embryonic bovine tracheal fibroblasts, a bovine kidney epithelial cell line (MDBK), Hela cells, gerbil fibroma cells, and fibroblast lines of human and mouse origins. The precipitation of the spectrin was specific and two proteins with a similar distribution to the nonerythrocyte spectrin were not induced to co-precipitate in the spectrin aggregates. Comparing the two types of antibody microinjected, the affinity-purified polyclonal antibody resulted in more compact aggregates of spectrin and these were frequently aligned with microfilament bundles. The rate at which the spectrin aggregates were cleared into presumptive lysosomes varied with different cell types: in some such as the bovine kidney epithelial cells, this appeared complete within 3 h after microinjection, whereas in some of the fibroblasts the spectrin aggregates were prominent in the cytoplasm at 24 and even 48 h after microinjection. Microfilament bundles appeared unaffected by the aggregation of spectrin. We conclude that the integrity of the actin microfilament bundles does not require nonerythrocyte spectrin and that most probably these structures are linked at their termini to the membrane through proteins other than nonerythrocyte spectrin. No effect of the intracellular spectrin precipitation was observed on cell shape, or on the distribution of coated vesicles or microtubules. The aggregation of the nonerythrocyte spectrin, however, did affect the distribution of the vimentin type of intermediate filaments in most of the cell types studied. These filaments became more distorted and condensed, but generally did not collapse around the nucleus as occurs following microtubule disruption induced by colchicine treatment. The clumped intermediate filaments were frequently seen to coincide with regions of aggregated spectrin. This aggregation of intermediate filaments was not induced by microinjection of irrelevant antibodies, nor was it induced by the monoclonal antibody against spectrin in cells with which it did not cross-react.(ABSTRACT TRUNCATED AT 400 WORDS)     

Transgene Rescue Identifies an Essential Function for Drosophila β Spectrin in the Nervous System and a Selective Requirement for Ankyrin-2–binding Activity

The protein spectrin is ubiquitous in animal cells and is believed to play important roles in cell shape and membrane stability, cell polarity, and endomembrane traffic. Experiments here were undertaken to identify sites of essential β spectrin function in Drosophila and to determine whether spectrin and ankyrin function are strictly linked to one another. The Gal4-UAS system was used to drive tissue-specific overexpression of a β spectrin transgene or to knock down β spectrin expression with dsRNA. The results show that 1) overexpression of β spectrin in most of the cell types studied was lethal; 2) knockdown of β spectrin in most tissues had no detectable effect on growth or viability of the organism; and 3) nervous system-specific expression of a UAS-β spectrin transgene was sufficient to overcome the lethality of a loss-of-function β spectrin mutation. Thus β spectrin expression in other cells was not required for development of fertile adult males, although females lacking nonneuronal spectrin were sterile. Previous data indicated that binding of the DAnk1 isoform of ankyrin to spectrin was partially dispensable for viability. Domain swap experiments here uncovered a different requirement for neuronal DAnk2 binding to spectrin and establish that DAnk2-binding is critical for β spectrin function in vivo.     

Possible identity of a membrane-bound with a soluble cyclic AMP-independent erythrocyte protein kinase that phosphorylates spectrin

A soluble casein kinase isolated and purified to homogeneity from the human erythrocyte cytosol by phosphocellulose and Sephadex G-200 chromatographies is indistinguishable from the membrane-bound casein (spectrin) kinase according to physical and site-specificity criteria. The soluble enzyme shows an Mr of about 30000 by gel filtration and comigrates with the purified membrane spectrin kinase as a single polypeptide of 32000 Da on sodium dodecyl sulfate polyacrylamide gels. The soluble kinase phosphorylates spectrin in situ in spectrin kinase-depleted ghosts and catalyzes the in vitro phosphorylation of partially dephosphorylated spectrin with saturation kinetics identical to those displayed by the membrane spectrin kinase. When component 2 of spectrin that had been phosphorylated with [gamma-32P]ATP by either the soluble or the membrane kinases was subjected to limited proteolysis, the same 21500 Da papain-generated phosphopeptide was found to have been produced by the two enzymes. The same 21500 Da phosphopeptide was identified after papain digestion of spectrin isolated from intact cells that had been incubated with 32Pi. However, this particular peptide was not labeled in spectrin that had been phosphorylated in vitro by the catalytic subunit of cyclic AMP-dependent protein kinase. Identical phosphopeptide patterns were obtained by gel filtration and two-dimensional peptide maps of trypsin-cleaved component 2 of spectrin that had been labeled in situ, in intact ghosts or in spectrin kinase-depleted ghosts supplemented with the soluble kinase. These findings indicate a possible identity of the soluble with the membrane-bound casein (spectrin) kinase.     

Spectrin and calmodulin in spreading mouse blastomeres

The role of spectrin and its association with calmodulin in spreading mouse blastomeres was investigated. Embryonic spectrin binds 125I-calmodulin in a calcium-dependent fashion in the blot overlay technique. Double-labeling experiments show coordinate redistribution of spectrin and calmodulin in blastomeres preparing to undergo active spreading movement. At this stage cortical spectrin staining is lost from the region of cell-substrate contact and spectrin and calmodulin become concentrated in two structures closely associated with the contacted region: a group of spherical bodies located on the cytoplasmic side of the cortical layer and a subcortical ring that marks the perimeter of the contacted region. The localization pattern of spectrin and calmodulin is also coordinated with that of actin and myosin. The results suggest that spectrin plays a role in the spreading of blastomeres and that this function may involve linkage of spectrin, calmodulin, and the cortical contractile apparatus.     

Tryptophan phosphorescence study of the internal dynamics of human erythrocyte membrane proteins upon spectrin modification

Room-temperature tryptophan phosphorescence has been used analyze the slow (millisecond) internal dynamics of proteins in isolated native human erythrocyte membranes, after removal of 95% of spectrin, and after thermal denaturation of spectrin or medium acidification to pH 6.0–4.0, as well as the internal dynamics of spectrin extracted from the membrane in solution. The integral membrane proteins prove to differ sharply from spectrin in their structural and dynamic state. The millisecond movements of structural elements in integral proteins are considerably hindered as compared with spectrin. Removal of the bulk of spectrin from membranes leads to amplification of slow fluctuations in the structure of integral proteins. This suggests involvement of spectrin in the control of the structural and dynamic state of the erythrocyte membrane proteins. The acidification of the medium to pH 6.0–4.0 decreases the internal dynamics of native membrane proteins, which is explained by the pH-induced aggregation of spectrin. After thermal denaturation of spectrin, there is no pH-induced increase in the rigidity of the structure of membrane proteins.     

Biochemical characterization of complex formation by human erythrocyte spectrin, protein 4.1, and actin

Ternary complex formation between the major human erythrocyte membrane skeletal proteins spectrin, protein 4.1, and actin was quantified by measuring cosedimentation of spectrin and band 4.1 with F-actin. Complex formation was dependent upon the concentration of spectrin and band 4.1, each of which promoted the binding of the other to F-actin. Simultaneous measurement of the concentrations of spectrin and band 4.1 in the sedimentable complex showed that a single molecule of band 4.1 was sufficient to promote the binding of a spectrin dimer to F-actin. However, the molar ratio of band 4.1/spectrin in the complex was not fixed, ranging from approximately 0.6 to 2.2 as the relative concentration of added spectrin to band 4.1 was decreased. A mole ratio of 0.6 band 4.1/spectrin suggests that a single molecule of band 4.1 can promote the binding of more than one spectrin dimer to an actin filament. Saturation binding studies showed that in the presence of band 4.1 every actin monomer in a filament could bind at least one molecule of spectrin, yielding ternary complexes with spectrin/actin mole ratios as high as 1.4. Electron microscopy of such complexes showed them to consist of actin filaments heavily decorated with spectrin dimers. Ternary complex formation was not affected by alteration in Mg2+ or Ca2+ concentration but was markedly inhibited by KCl above 100 mM and nearly abolished by 10 mM 2,3-diphosphoglycerate or 10 mM adenosine 5'-triphosphate. Our data are used to refine the molecular model of the red cell membrane skeleton.     

Brain spectrin(240/235) and brain spectrin(240/235E): two distinct spectrin subtypes with different locations within mammalian neural cells

Adult mouse brain contains at least two distinct spectrin subtypes, both consisting of 240-kD and 235-kD subunits. Brain spectrin(240/235) is found in neuronal axons, but not dendrites, when immunohistochemistry is performed with antibody raised against brain spectrin isolated from enriched synaptic/axonal membranes. A second spectrin subtype, brain spectrin(240/235E), is exclusively recognized by red blood cell spectrin antibody. Brain spectrin(240/235E) is confined to neuronal cell bodies and dendrites, and some glial cells, but is not present in axons or presynaptic terminals.     

Spectrin in mouse gametogenesis and embryogenesis

Antibodies to nonerythroid alpha spectrin (p 230) were used to study the distribution of this polypeptide in mouse germ cells, zygote, and early embryonic cells. In the primordial germ cells, fetal oocytes, and spermatogonia, spectrin was found predominantly in the form of a narrow condensed subplasmalemmal band, as in all other somatic cells. During spermatogenesis, spectrin is condensed into the supraacrosomal cytoplasm and is lost during the reduction of the cytoplasm of the maturing spermatozoa. The postnatal growth of the oocyte is accompanied by a loss of the dense cortical band of spectrin and its redistribution in the cytoplasm. Zygotes also contain granular dispersed spectrin. Cortical condensation of spectrin filaments gradually reappears in the blastomeres at the two-cell stage and in the secondary polar body. Cortically condensed filaments represent thereafter the predominant form of spectrin in all preimplantation stage embryonic cells. Trophoblastic cells spreading out from explanted blastocysts are devoid of the cortically condensed spectrin and contain, instead, spectrin arrays in the cytoplasm. Trophoblastic cells, which surround the implanted embryo in vivo, also show diffuse cytoplasmic spectrin which subsequently undergoes subplasmalemmal condensation. These data show that spectrin is present in all stages of gametogenesis and embryogenesis, except in mature spermatozoa; and that it undergoes cytoplasmic redistribution during morphogenesis.     

Biosynthesis of spectrin and its assembly into the cytoskeletal system of Friend erythroleukemia cells

Friend erythroleukemia cells, grown in the presence of dimethyl sulfoxide for 3 d, synthesize unequal amounts of the two chains (alpha and beta) of spectrin with approximately 15-30% more beta than alpha spectrin. When cells were ruptured by nitrogen cavitation, nascent alpha and beta spectrin were found to be associated with a membranous cell fraction and were not detected in the soluble cytoplasmic cell fraction. Nascent membrane-bound spectrin appeared not to be protected by membranes, since it was susceptible to trypsin degradation in the absence of detergent. On fractionation of cells with 1% Triton X-100, more (1.75-fold) nascent spectrin was found in the Triton-soluble fraction than in the Triton-insoluble fraction (cytoskeleton). In the Triton-soluble fraction, there was 55% more nascent beta spectrin than alpha spectrin, while the cytoskeleton contained nearly equal amounts of alpha and beta spectrin. Cells were pulse-labeled with L-[35S]methionine for 2 min and chase incubated for varying periods of time from 15 to 90 min with nonradioactive L-methionine. Radioactive spectrin accumulated in the Triton-soluble fraction for the first 15 min of chase incubation and then dropped by 25% in the next hour. By contrast, the amount of radioactive spectrin in the Triton-insoluble fraction rose gradually for 1 h of the chase period. This indicates that, in Friend erythroleukemia cells, a pool of membrane-bound spectrin containing an excess of the beta polypeptide is used to form the cytoskeletal system which is composed of equal molar amounts of alpha and beta spectrin. The location of spectrin was determined by immunoelectron microscopy. Small amounts of spectrin were detected in cells not treated with dimethyl sulfoxide and in these cells it was located on the surface membrane and within the cytoplasm. On treatment with dimethyl sulfoxide, complex vacuolar structures containing viruses appeared in the cells. In cells treated with dimethyl sulfoxide for 3 d 30% of the spectrin was near the outer membrane and 25% was associated with vacuolar structures, whereas in cells treated for 5 and 7 d the majority of spectrin (57-61%) was located in the vacuolar areas.