Purification and characterization of recombinant spider silk expressed in Escherichia coli

A partial cDNA clone, from the 3' end of the dragline silk gene was isolated from Nephila clavipes major ampullate glands. This clone contains a 1.7-kb insert, consisting of a repetitive coding region of 1.4-kb and a 0.3-kb nonrepetitive coding region; 1.5-kb of the 1.7-kb fragment was cloned into Escherichia coli and a 43-kDa recombinant silk protein was expressed. Characterization of the purified protein by Western blot, amino acid composition analysis, and matrix-assisted laser desorption ionization/time-of-flight mass spectrometry confirms it to be spider dragline silk.

Spontaneous oligomerization of a staphylococcal alpha-hemolysin conformationally constrained by removal of residues that form the transmembrane beta-barrel

Staphylococcal alpha-hemolysin is a water soluble, monomeric, bacterial exotoxin, which forms heptameric pores in membranes. The rate determining step in assembly is the conversion of a heptameric prepore to the fully assembled pore in which the central glycine-rich domain of each subunit inserts into the membrane to form a 14 strand beta barrel. Barrel formation is accompanied by a conformational change in which each N terminus latches onto an adjacent subunit. In the monomer in solution, the central domain is loosely organized and exposed to solvent. In this study, 25 amino acids of the central domain were removed and replaced with the sequence Asp-Gly, which favors the formation of a type I' beta-turn, to yield a mutant devoid of hemolytic activity. Within minutes after synthesis in the absence of membranes, the mutant polypeptide spontaneously assembled into heptamers, as demonstrated by atomic force microscopy. Limited proteolysis suggested that the N termini of the subunits in the heptamers were in the fully assembled pore conformation rather than the prepore conformation. Based on these findings, the deletion is proposed to constrain the central domain and thereby force the creation of a shortened beta barrel, which in turn induces the additional structural changes that normally accompany pore formation. The truncated pore might make a useful framework for the construction of designed membrane active macromolecules.

The heptameric prepore of a staphylococcal alpha-hemolysin mutant in lipid bilayers imaged by atomic force microscopy

We have used atomic force microscopy to study the oligomeric state of a genetically engineered mutant of staphylococcal alpha-hemolysin (alphaHL-H5) that can be arrested as a "prepore" assembly intermediate. AFM images of alphaHL-H5 on supported bilayers of a fluid-phase lipid, egg-yolk phosphatidylcholine (egg-PC), under conditions that lock alphaHL-H5 into the prepore state, clearly show a heptameric structure for many individual oligomers. The central dent of the prepore has a diameter of 3.2 +/- 0.2 nm. The distance between the centers of mass of neighboring subunits is 2.8 +/- 0.3 nm. The heptamer has an average diameter of 8.9 +/- 0.6 nm. These results support a recently proposed pathway for the assembly of alpha-hemolysin.

Designed protein pores as components for biosensors

BACKGROUND

There is a pressing need for new sensors that can detect a variety of analytes, ranging from simple ions to complex compounds and even microorganisms. The devices should offer sensitivity, speed, reversibility and selectivity. Given these criteria, protein pores, remodeled so that their transmembrane conductances are modulated by the association of specific analytes, are excellent prospects as components of biosensors.

RESULTS

Structure-based design and a separation method that employs targeted chemical modification have been used to obtain a heteromeric form of the bacterial pore-forming protein staphylococcal alpha-hemolysin, in which one of the seven subunits contains a binding site for a divalent metal ion, M(II), which serves as a prototypic analyte. The single-channel current of the heteromer in planar bilayers is modulated by nanomolar Zn(II). Other M(II)s modulate the current and produce characteristic signatures. In addition, heteromers containing more than one mutant subunit exhibit distinct responses to M(II)s Hence, a large collection of responsive pores can be generated through subunit diversity and combinatorial assembly.

CONCLUSIONS

Engineered pores have several advantages as potential sensor elements: sensitivity is in the nanomolar range; analyte binding is rapid (diffusion limited in some cases) and reversible; strictly selective binding is not required because single-channel recordings are rich in information; and for a particular analyte, the dissociation rate constant, the extent of channel block and the voltage-dependence of these parameters are distinguishing, while the frequency of partial channel block reflects the analyte concentration. A single sensor element might, therefore, be used to quantitate more than one analyte at once. The approach described here can be generalized for additional analytes.

Structure of staphylococcal alpha-hemolysin, a heptameric transmembrane pore

The structure of the Staphylococcus aureus alpha-hemolysin pore has been determined to 1.9 A resolution. Contained within the mushroom-shaped homo-oligomeric heptamer is a solvent-filled channel, 100 A in length, that runs along the sevenfold axis and ranges from 14 A to 46 A in diameter. The lytic, transmembrane domain comprises the lower half of a 14-strand antiparallel beta barrel, to which each protomer contributes two beta strands, each 65 A long. The interior of the beta barrel is primarily hydrophilic, and the exterior has a hydrophobic belt 28 A wide. The structure proves the heptameric subunit stoichiometry of the alpha-hemolysin oligomer, shows that a glycine-rich and solvent-exposed region of a water-soluble protein can self-assemble to form a transmembrane pore of defined structure, and provides insight into the principles of membrane interaction and transport activity of beta barrel pore-forming toxins.

Tumor protease-activated, pore-forming toxins from a combinatorial library

We describe a library of two-chain molecular complementation mutants of staphylococcal alpha-hemolysin that features a combinatorial cassette encoding thousands of protease recognition sites in the central pore-forming domain. The cassette is flanked by a peptide extension that inactivates the protein. We screened the library to identify alpha-hemolysins that are highly susceptible to activation by cathepsin B, a protease that is secreted by certain metastatic tumor cells. Toxins obtained by this procedure should be useful for the permeabilization of malignant cells thereby leading directly to cell death or permitting destruction of the cells with drugs that are normally membrane impermeant.

An intermediate in the assembly of a pore-forming protein trapped with a genetically-engineered switch

BACKGROUND

Studies of the mechanisms by which certain water-soluble proteins can assemble into lipid bilayers are relevant to several areas of biology, including the biosynthesis of membrane and secreted proteins, virus membrane fusion and the action of immune proteins such as complement and perforin. The alpha-hemolysin (alpha HL) protein, an exotoxin secreted by Staphylococcus aureus that forms heptameric pores in lipid bilayers, is a useful model for studying membrane protein assembly. In addition, modified alpha HL might be useful as a component of biosensors or in drug delivery. We have therefore used protein engineering to produce variants of alpha HL that contain molecular triggers and switches with which pore-forming activity can be modulated at will. Previously, we showed that the conductance of pores formed by the mutant hemolysin alpha HL-H5, which contains a Zn(II)-binding pentahistidine sequence, is blocked by Zn(II) from either side of the lipid bilayer, suggesting that residues from the pentahistidine sequence line the lumen of the transmembrane channel.

RESULTS

Here we show that Zn(II) can arrest the assembly of alpha HL-H5 before pore formation by preventing an impermeable oligomeric prepore from proceeding to the fully assembled state. The prepore is a heptamer. Limited proteolysis shows that, unlike the functional pore, the prepore contains sites near the amino terminus of the polypeptide chain that are exposed to the aqueous phase. Upon removal of the bound Zn(II) with EDTA, pore formation is completed and the sites near the amino terminus become occluded. Conversion of the prepore to the active pore is the rate-determining step in assembly and cannot be reversed by the subsequent addition of excess Zn(II).

CONCLUSIONS

The introduction of a simple Zn(II)-binding motif into a pore-forming protein has allowed the isolation of a defined intermediate in assembly. Genetically-engineered switches for trapping and releasing intermediates that are actuated by metal coordination or other chemistries might be generally useful for analyzing the assembly of membrane proteins and other supramolecular structures.

Subunit stoichiometry of staphylococcal alpha-hemolysin in crystals and on membranes: a heptameric transmembrane pore

Elucidation of the accurate subunit stoichiometry of oligomeric membrane proteins is fraught with complexities. The interpretations of chemical cross-linking, analytical ultracentrifugation, gel filtration, and low-resolution electron microscopy studies are often ambiguous. Staphylococcal alpha-hemolysin (alpha HL), a homooligomeric toxin that forms channels in cell membranes, was believed to possess six subunits arranged around a sixfold axis of symmetry. Here, we report that analysis of x-ray diffraction data and chemical modification experiments indicate that the alpha HL oligomer is a heptamer. Self-rotation functions calculated using x-ray diffraction data from single crystals of alpha HL oligomers show a sevenfold axis of rotational symmetry. The alpha HL pore formed on rabbit erythrocyte membranes was determined to be a heptamer by electrophoretic separation of alpha HL heteromers formed from subunits with the charge of wild-type alpha HL and subunits with additional negative charge generated by targeted chemical modification of a single-cysteine mutant. These data establish the heptameric oligomerization state of the alpha HL transmembrane pore both in three-dimensional crystals and on a biological membrane.

Differential phosphorylation of neuronal substrates by catalytic subunits of Aplysia cAMP-dependent protein kinase with alternative N termini

cAMP-dependent protein kinase (PKA) is an important participant in neuronal modulation: the ability of neurons to change their properties in response to external stimuli. In Aplysia mechanosensory neurons, PKA plays roles in both short and long term presynaptic facilitation, which is a simple model for learning and memory. PKA in Aplysia is a collection of structurally and functionally diverse regulatory and catalytic (C) subunits. We have argued that this diversity may in part account for the ability of the enzyme to take part in neuronal events that are spatially and temporally separated. Here, we add credence to this hypothesis by showing that C subunits of Aplysia PKA with alternative N termini target different substrates in subcellular fractions from Aplysia neurons, despite their similar actions on synthetic peptide substrates. Purified recombinant CAPL-AN1A1, which has an N terminus that is homologous to the myristylated sequence described in mammals, catalyzes the formation of two phosphoproteins of 24 and 8 kDa more rapidly than CAPL-AN2A1, which has a distinct N terminus weakly related to that of the yeast TPK1 gene product. The 24-kDa phospoprotein, but not the 8-kDa species, is detected in taxol-stabilized microtubules, suggesting that it is associated with the cytoskeleton. CAPL-AN2A1, in contrast, generates a 55-kDa phosphoprotein that is not observed with CAPL-AN1A1. The 55-kDa species is found in the detergent supernatant of the cytoskeleton fraction. Differential targeting of substrates by C subunits of PKA may therefore contribute to the ability of this kinase to play multiple roles in neuronal modulation.

Type II regulatory subunits of cAMP-dependent protein kinase and their binding proteins in the nervous system of Aplysia californica

Two type II regulatory (R) subunits of cAMP-dependent protein kinase (PKA) of 50 and 47 kDa have been identified in Aplysia neurons by several criteria which include phosphorylation by the catalytic subunit of PKA and nanomolar affinity for a peptide fragment of the human thyroid protein Ht 31, properties that in mammals distinguish type II from type I R subunits. The neuronal type II R subunits are differentially localized within cells. For example, the 50-kDa polypeptide is enriched in taxol-stabilized microtubules. In addition, at least seven high molecular mass neuronal RII-binding proteins ranging in mass from 110 to 420 kDa have been demonstrated by a blot overlay technique, which uses 32P-labeled bovine RII alpha as a probe. The RII-binding proteins also exhibit discrete patterns of subcellular localization. For example, the 420 kDa species is enriched in taxol-stabilized microtubules and therefore may serve to anchor the 50-kDa RII subunit. The localization of PKA through the association of RII subunits with the binding proteins may anchor the multifunctional kinase close to key substrates and thereby contribute to the spatial organization required to mediate the orderly phosphorylation events that underly neuronal modulation.

Phosphorylated baculovirus p10 is a heat-stable microtubule-associated protein associated with process formation in Sf9 cells

Insect ovarian Sf9 cells extend processes with complex morphologies when infected with a recombinant baculovirus encoding the catalytic subunit of protein kinase A. Within the shafts of the processes are abundant microtubules, which, in contrast to those in Sf9 cells expressing the microtubule-associated protein tau, are generally not organized into parallel bundles. During infection the late viral polypeptide p10 becomes phosphorylated by the protein kinase A catalytic subunit at its penultimate residue, Ser92. The expression or phosphorylation of other major host cell or viral polypeptides does not change, compared with polypeptides from a wild-type viral infection. Once phosphorylated, p10 associates with microtubules in the infected cells and may thereby play a role in process formation.

A regulatory subunit of the cAMP-dependent protein kinase down-regulated in aplysia sensory neurons during long-term sensitization

Binding of cAMP by the five neuronal isoforms (N1-5) of the regulatory (R) subunit of the Aplysia cAMP-dependent protein kinase is diminished in sensory neurons stimulated to produce long-term presynaptic facilitation. To determine how the cAMP-binding activity of the R subunits is lost, we isolated cDNAs encoding N4, which is a homolog of mammalian RI. Immunoblots with antisera raised against the R protein overexpressed in E. coli show that the diminished binding activity, which occurs in long-term facilitation, results from coordinate loss of R protein isoforms. No change was detected in the amount of transcripts for R subunits, suggesting that the down-regulation results from enhanced proteolytic turnover.

Kinetics and regulation of two catalytic subunits of cAMP-dependent protein kinase from Aplysia californica

CAPL-A1 and CAPL-A2, two catalytic subunits of Aplysia cAMP-dependent protein kinase, are encoded by mRNAs generated by alternative splicing of transcripts of a gene that contains two mutually exclusive exon cassettes. The subunits are identical except for amino acids 142-183 of the 352 residues, which differ at 10 of 42 positions. CAPL-A1 and CAPL-A2 have now been expressed in insect cells and purified to homogeneity. The subunits differ in their catalytic properties, which have been determined with a series of synthetic peptide substrates. For example, kcat and Km values for the peptide LRRASLG (kemptide) are 42 s-1 and 36 microM and 28 s-1 and 17 microM for CAPL-A1 and CAPL-A2, respectively. CAPL-A1 and CAPL-A2 have different substrate specificities. For example, (kcat/Km)peptide-T/(kcat/Km)kemptide is 9.1 x 10(-3) for CAPL-A1 and 15 x 10(-3) for CAPL-A2, where peptide-T is the kemptide homologue LRRATLG. The subunits also differ in regulation as determined by their interactions with a purified type I regulatory subunit, which has an IC50 for CAPL-A1 that is 3.5 times higher than the IC50 for CAPL-A2. These modest differences reinforce accumulating evidence that the physiological state of a cell depends upon a spectrum of protein kinases with overlapping substrate specificities and regulatory properties.

Nonmuscle and muscle tropomyosin isoforms are expressed from a single gene by alternative RNA splicing and polyadenylation

The molecular basis for the expression of rat embryonic fibroblast tropomyosin 1 and skeletal muscle beta-tropomyosin was determined. cDNA clones encoding these tropomyosin isoforms exhibit complete identity except for two carboxy-proximal regions (amino acids 189 to 213 and 258 to 284) and different 3'-untranslated sequences. The isoform-specific regions delineate the troponin T-binding domains of skeletal muscle tropomyosin. Analysis of genomic clones indicates that there are two separate loci in the rat genome that contain sequences complementary to these mRNAs. One locus is a pseudogene. The other locus contains a single gene made up of 11 exons and spans approximately 10 kilobases. Sequences common to all mRNAs were found in exons 1 through 5 (amino acids 1 to 188) and exons 8 and 9 (amino acids 214 to 257). Exons 6 and 11 are specific for fibroblast mRNA (amino acids 189 to 213 and 258 to 284, respectively), while exons 7 and 10 are specific for skeletal muscle mRNA (amino acids 189 to 213 and 258 to 284, respectively). In addition, exons 10 and 11 each contain the entire 3'-untranslated sequences of the respective mRNAs including the polyadenylation site. Although the gene is also expressed in smooth muscle (stomach, uterus, and vas deferens), only the fibroblast-type splice products can be detected in these tissues. S1 and primer extension analyses indicate that all mRNAs expressed from this gene are transcribed from a single promoter. The promoter was found to contain G-C-rich sequences, a TATA-like sequence TTTTA, no identifiable CCAAT box, and two putative Sp1-binding sites.

Translational control in murine hepatitis virus infection

High multiplicity infection of mouse fibroblast L-2 cells with mouse hepatitis virus (MHV) resulted, within 6 h, in a decline in total protein synthesis to about 7% of that observed in uninfected cells. The amount of intracellular total translatable RNA, however, increased approximately threefold, as a result of the accumulation of virus-encoded mRNAs. MHV-infected cells could be superinfected with vesicular stomatitis virus, demonstrating that MHV infection did not irreversibly alter the cellular translational machinery to the exclusion of non-MHV mRNAs. Comparative polysome analysis from MHV-infected and uninfected L-2 cells showed that MHV infection resulted in an increase in single 80S ribosomes and in a shift from longer to shorter polysomes. These observations suggest first, that MHV infection inhibits total protein synthesis at a very early stage, as evidenced by the increase in 80S ribosomes, and, second, that the increased number of viral mRNAs produced after infection compete with cellular mRNAs for cellular ribosomes. In vitro translation of RNA extracted from MHV-infected and mock-infected cells suggested that levels of cellular mRNAs were decreased after infection. This suggestion was confirmed by demonstrating the loss of cellular actin mRNA, using a radiolabelled cDNA probe, as a consequence of MHV infection.

Attenuation of murine coronavirus infection by ammonium chloride

Ammonium chloride at a concentration of 20 mM delayed by 4-5 hr the production of virus progeny in mouse L-2 cells infected at high multiplicity with mouse hepatitis virus (MHV). This delay was seen in the production of both intracellular and extracellular virus. However, the final titers were similar to those produced by MHV-infected cells maintained in normal medium. The manifestation of virus-induced cell fusion was similarly found to be delayed, but not otherwise decreased in severity, when ammonium chloride was present in the culture medium. Ammonium chloride caused similar delays in production of virus-specific, positive-sense RNAs and of viral polypeptides. The relative proportions and apparent molecular weights of viral RNAs and polypeptides were similar to those found in MHV-infected cells cultured in normal medium. In vitro translation of endogenously produced viral RNAs in cell extracts, prepared from MHV-infected cells, was not inhibited by ammonium chloride. Thus, ammonium chloride has no specific, inhibitory effect on viral protein synthesis. Ammonium chloride did not reduce the number of virus-infected cells in culture, as monitored by infectious center assay. Analysis of early events in MHV infection showed that ammonium chloride did not affect adsorption or internalization of MHV by L-2 cells. However, the subsequent eclipse phase, as monitored by decline in infectivity of internalized virus inoculum proceeded less efficiently in the presence of ammonium chloride. On the basis of the known inhibitory effects of ammonium chloride on lysosomal/endosomal functions, the results suggest an endosomal mechanism of MHV uncoating. Thus the primary effect of ammonium chloride on MHV infection of L-2 cells is to attenuate virus uncoating, thereby chronologically displacing all subsequent virus-encoded functions.

A reproducible microanalytical method for the detection of specific RNA sequences by dot-blot hybridization

A rapid, microanalytical procedure for the reproducible isolation of RNA from small cultured cell samples and application to dot-blot hybridization is described. The procedure employs guanidine hydrochloride solubilization of whole cells, disruption by syringing, and selective precipitation of RNA with ethanol. The method can be performed in a single tissue culture tube and obviates the need for removal of nuclei or for organic solvent extractions. Recovery of RNA from small cell samples (10(6) cells) is 51%, while 97% of the DNA and 99% of the protein are eliminated by the procedure. Detection of specific RNA by dot-blot hybridization using a labeled probe demonstrates high reproducibility of recovered RNA and lack of "masking" with up to a 10-fold excess of starting cell material. Applicability of the procedure to detection of virus-specific RNA in cells persistently infected with mouse hepatitis virus is described.

Fusion resistance and decreased infectability as major host cell determinants of coronavirus persistence

Mouse hepatitis virus persists in cultures of a subline (designated LM-K) of mouse LM cells but produces a lytic infection in L-2 cells. Persistence in the LM-K cells was not accompanied by production of ts mutants or of soluble anti-MHV factors. Infectious center assay demonstrated an approximately 500-fold lower level of infectibility by MHV of the LM-K cells as compared to L-2 cells. On an infected cell basis, production levels of infectious progeny and viral RNA were comparable between the two cell lines. The extent of virus-induced cell-cell fusion, however, was markedly reduced in the LM-K cells. Cell-mixing experiments showed that both infected L-2 and LM-K cells have the capacity of fusing with neighboring uninfected L-2 cells but not with uninfected LM-K cells. This suggests that the decreased level of fusion observed in the LM-K infection is due not to absence of viral fusion protein at the cell surface, but rather to an inherent resistance of the LM-K cell membrane to MHV-induced fusion. It is believed that such fusion resistance in LM-K cells moderates virus dissemination throughout the culture, thereby contributing to a state of virus persistence.

RNA and polypeptide homology among murine coronaviruses

Using a 32P complementary DNA (cDNA) prepared against the A59 nucleocapsid protein messenger RNA, we have investigated the extent of homology between A59 and four other strains of murine hepatitis virus (MHV). Analysis by hybridization kinetics of the annealing between A59 [32P]cDNA and infected cell RNA from the other four MHV strains demonstrated 70–80% homology. By gel transfer analysis, the A59 [32P]cDNA was able to detect subgenomic-size virus-specific RNAs in cells infected with all of the five MHV strains. A similar pattern of seven viral RNAs was detected in cells infected with A59, MHV-1, MHV-3, and JHM. In contrast, cells infected with MHV-S contained seven virus-specific RNAs, of which only the two smallest species comigrated with RNAs from the other four strains. The results suggest that as previously shown with A59 (S. Cheley, R. Anderson, M. J. Cupples, E. C. M. Lee Chan, and V. L. Morris (1981) Virology , 112, 596–604), all MHV strains tested encode a nested set of subgenomic RNAs. Analysis of [35S]methionine-labeled viral proteins by SDS-polyacrylamide gel electrophoresis revealed that each strain of MHV specified four major viral polypeptides with apparent molecular weights very similar to those previously reported for the E2, N, El, and PEI polypeptides of A59. The strong degree of interstrain homology among the five MHV strains investigated was confirmed by comparative chymotryptic peptide mapping of the viral N proteins. A majority of the chymotryptic peptides from each of the [35Sknethionine-labeled N proteins was found to coelute by high-performance liquid chromotography. Moreover, this technique of peptide mapping indicated a particularly strong relatedness between MHV-1 and MHV-S and among MHV-3, JHM, and A59.

Intracellular murine hepatitis virus-specific RNAs contain common sequences

A major polyadenylated viral RNA of approximately 0.8 × 10⁶ daltons was isolated from murine hepatitis virus (A59)-infected cells by preparative polyacrylamide gel electrophoresis in formamide. This RNA was shown to encode the viral nucleocapsid protein by direct in vitro translation in a cell-free, reticulocyte-derived system. Single stranded ³²P-labeled complementary DNA was prepared from this RNA and was demonstrated to be virus specific. Using this complementary DNA in a Northern blotting procedure, we were able to identify six major virus-specific intracellular RNA species with estimated molecular weights of 0.8, 1.1, 1.4, 1.6, 3, and 4 × 10⁶ daltons. All of these RNA species were polyadenylated. Our results support the idea that coronavirus-infected cells contain multiple intracellular polyadenylated RNAs which share common sequences.