Experimental Cell Research 269, 117–129 (2001) doi:10.1006/excr.2001.5297, available online at http://www.idealibrary.com on
A Verotoxin 1 B Subunit-Lambda CRO Chimeric Protein Specifically Binds Both DNA and Globotriaosylceramide (Gb 3) to Effect Nuclear Targeting of Exogenous DNA in Gb 3 Positive Cells Linda M. Facchini and Clifford A. Lingwood 1 Division of Infection, Immunity, Injury and Repair, Research Institute, The Hospital for Sick Children, and Departments of Laboratory Medicine and Pathobiology, and Biochemistry, University of Toronto, Toronto, Canada
INTRODUCTION Inefficient nuclear incorporation of foreign DNA remains a critical roadblock in the development of effective nonviral gene delivery systems. DNA delivered by traditional protocols remains within endosomal/lysosomal vesicles, or is rapidly degraded in the cytoplasm. Verotoxin I (VT), an AB 5 subunit toxin produced by enterohaemorrhagic Escherichia coli, binds to the cell surface glycolipid, globotriaosylceramide (Gb 3) and is internalized into preendosomes. VT is then retrograde transported to the Golgi, endoplasmic reticulum (ER), and nucleus of highly VT-sensitive cells. We have utilized this nuclear targeting of VT to design a unique delivery system which transports exogenous DNA via vesicular traffic to the nucleus. The nontoxic VT binding subunit (VTB) was fused to the lambda Cro DNA-binding repressor, generating a 14kDa VTB-Cro chimera. VTB-Cro binds specifically via the Cro domain to a 25-bp DNA fragment containing the consensus Cro operator. VTB-Cro demonstrates simultaneous specific binding to Gb 3. Treatment of Vero cells with fluorescent-labeled Cro operator DNA in the presence of VTB-Cro, results in DNA internalization to the Golgi, ER, and nucleus, whereas fluorescent DNA alone is incorporated poorly and randomly within the cytoplasm. VTB-Cro mediated nuclear DNA transport is prevented by brefeldin A, consistent with Golgi/ER intracellular routing. Pretreatment with filipin had no effect, indicating that caveoli are not involved. This novel VTB-Cro shuttle protein may find practical applications in the fields of intracellular targeting, gene delivery, and gene therapy. © 2001 Academic Press
Key Words: retrograde transport; vesicular transport; receptor-mediated endocytosis; glycolipid; membrane translocation; confocal microscopy.
1 To whom reprint requests should be addressed at Division of Infection, Immunity , Injury and Repair, The Hospital for Sick Children Research Institute, 555 University Avenue, Toronto, Canada M5G 1X8. Fax: (416) 813-5993. E-mail:
[email protected].
Nonviral gene delivery systems which safely introduce exogenous DNA into eukaryotic cells are under vigorous development for use in gene therapy. However, existing polycation, lipid-based systems typically yield transient and low-level expression of foreign genes. This results, in part, from inefficient nuclear transport of the exogenous DNA. Although most cells readily uptake exogenous DNA delivered by polycationic lipids, the nucleic acids localize to endosomal vesicles and are degraded in lysosomes in all but a small percentage of cells [1]. Another major factor limiting the efficacy of transfection is the relatively short half life (50 –90 min) of plasmid DNA in the cytosol [2]. While viral vectors effectively exploit the innate transport mechanism of the virion to circumvent these barriers, they often invoke strong inflammatory and immunological responses in vivo [3, 4]. Alternative strategies have incorporated membrane destabilizing viral and bacterial toxin components into polycation– lipid conjugates to facilitate DNA release from endosomes or the ER, [5– 8]; however these systems do not address the instability of DNA in the cytosol. Recent studies on the intracellular localization of the B subunit of the E. coli-derived verotoxin 1 (VT) have identified a novel retrograde transport pathway whereby the toxin–receptor complex is internalized from the cell surface via early endosomes to the Golgi in cells of medium sensitivity, but to the ER, nuclear membrane even within the nucleus in cells of high VT sensitivity [9 –12]. This latter pathway may provide a route for targeting exogenous DNA from the cell surface to the nucleus. The cytotoxicity of VT is a function of the 32-kDa A subunit, an RNA (and DNA [13]) N-glycanase which inhibits protein translation by removing a specific adenine residue from ribosomal RNA [14]. The 7.7-kDa B subunits are nontoxic [15] and self-aggregate into a doughnut-shaped pentamer [16] which mediates binding of the holotoxin to its cell surface receptor, globotri-
117
0014-4827/01 $35.00 Copyright © 2001 by Academic Press All rights of reproduction in any form reserved.
118
FACCHINI AND LINGWOOD
aosyl ceramide (Gal␣(1 3 4)-Gal(1 3 4) glucosyl ceramide (Gb 3)) [17]. The VTB pentamer and the holotoxin bind Gb 3 with equal affinity [16, 18]. Upon binding to Gb 3, VTB pentamer or holotoxin is internalized via clathrin-coated pits [19, 20] into preendosomes and the Golgi network [21, 22]. In cells highly sensitive to verotoxin, the VTB is further transported to the ER, nuclear envelope [19, 21], and discrete regions within the nucleus [9, 10, 12]. Conversely, cells which are less sensitive to VT, internalize VTB exclusively to the Golgi compartment [9, 10, 23]. While this retrograde transport mechanism is not well understood, differential subcellular targeting of the VTB subunit correlates with fatty acid chain length of Gb 3 [9, 11]. Both highly sensitive and less sensitive cells express equivalent levels of surface Gb 3; however, in the more sensitive cell lines, the Gb 3 molecules contain a greater percentage of C16,18 fatty acids, while the less-responsive cells contain more C22,24 fatty acids. Sensitivity to verotoxin and nuclear localization of the VTB/Gb 3 complex can be induced in cell lines of lower VT sensitivity by culture with sodium butyrate, correlating with an increase in C16,18 fatty acid-containing Gb 3. Coupling the VTB subunit to a DNA binding protein may allow retrograde transit of exogenous DNA to the nucleus to increase transfection efficiency. The 66-aminoacid Cro repressor (Cro) is the smallest DNA-binding protein. This helix–turn– helix protein binds as a dimer in the major groove of DNA [24, 25]. Cro specifically recognizes and tightly binds three similar 17-base-pair DNA consensus sequences, OR1, OR2, and OR3 [26]. Binding of Cro to these sites in the phage genome represses transcription from adjacent P R and P RM promoters through steric hindrance [27]. We describe here the construction of a novel fusion protein, VTB–Cro. We demonstrate that the fusion protein retains both Gb 3-binding capacity and the ability to simultaneously bind to a consensus Cro DNA recognition sequence. More importantly, VTB–Cro mediates targeting of exogenous DNA to the nucleus of Gb 3positive cells. MATERIALS AND METHODS Constructs The VTB gene fragment was amplified by PCR from plasmid pJLB 120 [18] using primers 5⬘-CCGTCGACCTGCAGGTCGC-3⬘ and 5⬘GCTATTCTGAATTCACGAAAAATAACTTCGC-3⬘ and subcloned into the PstI–EcoRI sites of pBluescript KS (Stratagene) to generate pBS-VTB. The Cro gene fragment was amplified from plasmid pCroRS (generous gift from M. Mossing, University of Notre Dame, Indiana) by PCR using the primers 5⬘-GGAGAATTCATGGAACAACGC-3⬘ and 5⬘-CCAAGCTTGCATGCCTGC-3⬘ and subcloned into the EcoRI–HindIII sites of pBS-VTB to generate pBS-VTB–Cro. The DNA fragment encoding the fusion protein VTB–Cro was excised from pBS-VTB–Cro with PstI and subcloned into the PstI site
of pKK223-3 to generate pVTB–Cro. Colonies were screened by immunoblot analysis for production of VTB–Cro fusion protein. Conservation of reading frame was confirmed by DNA sequencing. Bacterial Cultures and Cell Extracts Host E. coli strain TB1 was transformed with pVTB–Cro. Cultures were grown in LB broth supplemented with 100 g/ml ampicillin. Overnight cultures of TB1 and TB1 pVTB–Cro were harvested by centrifugation at 3000g, 4°C, for 10 min. To each gram of cell pellet was added 3 ml of cold extraction buffer (50 mM Tris, pH 8.0, 1 mM EDTA, 100 mM NaCl, 100 M phenylmethylsulfonyl fluoride (PMSF), 5 g/ml aprotinin, 5 g/ml leupeptin). Periplasmic extracts. Extraction buffer was supplemented with 20 mg/ml colymycin. Samples were vortexed at high speed and incubated at 37°C, 250 rpm, for 1 h. Cells were recovered by centrifugation and the pellet was resuspended in fresh extraction buffer plus colymycin and extracted in the same manner. Supernatants were pooled, concentrated 12⫻, and dialyzed overnight against TEN (50 mM Tris, pH 8.0, 1 mM EDTA, 100 mM NaCl). Soluble extracts. Extraction buffer was supplemented with 100 mg/ml lysozyme. Samples were vortexed at high speed and incubated at room temperature for 20 min. Deoxycholic acid (1.3 mg/ml) was added and incubated at 37°C for 1 h with gentle shaking until the mixture became viscous. DNase I (5 g/ml) was added and incubated at room temperature for 30 min. The lysate was clarified by centrifugation, and the supernatant was concentrated 6⫻ and dialyzed overnight against TEN. Immunoblot Analysis Total cell extracts were prepared by boiling cell pellets for 10 min in 1⫻ reducing SDS–sample buffer [28]. Equal samples of proteins were separated on 12% SDS–tricine gels and transferred to nitrocellulose membranes, blocked with 3% (w/v) nonfat milk in 50 mM Tris-buffered saline (TBS) plus 0.05% Tween 20 for 1 h, and then washed in TBS–Tween. Membranes were incubated rabbit polyclonal anti-VT1 B antiserum 6869 [29] in TBS–Tween for 1 h, washed, then incubated goat anti-rabbit horseradish peroxidase conjugated antibody (Bio-Rad) for 0.5 h. Membranes were washed and then developed with 4-chloro-1-naphthol in 50 mM TBS plus 3% (v/v) hydrogen peroxide. Localization Experiments Overnight cultures of TB1 pVTB–Cro were harvested and pellets were resuspended in 1/10 culture volume of sucrose solution (30 mM Tris, pH 8.0, 1 mM EDTA, 20% (w/v) sucrose) and incubated at room temperature for 10 min. Cells were harvested by centrifugation as above, gently resuspended in 1/10 culture volume of 5 mM MgSO 4, and incubated for 10 min at 4°C. Cells were centrifuged at 3000g, 4°C, for 20 min and supernatant retained as periplasmic fluid. The pellet was resuspended in 1/10 volume TEN containing 100 M PMSF and sonicated. Cells were centrifuged and supernatant retained as soluble fraction. The pellet was resuspended in 1/50 culture volume SDS–sample buffer and boiled for 10 min with vortexing and collected as soluble plus insoluble fraction. Periplasmic fluid and soluble fraction extracts were lyophilized overnight and resuspended in ddH 2O. Sample protein concentration was determine by the BioRad method and characterized by immunoblot analysis. Protein Purification Soluble extracts were dialyzed overnight against 10 mM sodium phosphate, pH 7.2, loaded onto a hydroxyapatite (BIOGEL-HT, BioRad) column, and washed with 2 column volumes 10 mM sodium phosphate, followed by a 250 mM sodium phosphate wash. Proteins were eluted with 500 mM sodium phosphate, and VTB–Cro-contain-
VTB-CRO CHIMERA TARGETS DNA TO THE NUCLEUS ing fractions were pooled and dialyzed into chromatofocusing start buffer (0.025 M ethanolamine, pH 9.4). Extracts were loaded onto a Polybuffer exchanger 94 gel (Pharmacia) in start buffer. Protein was eluted in Polybuffer 74 –HCl, pH 5.3, following the manufacturers’ protocol, and collected as flow-through fractions. Appropriate fractions were pooled and concentrated through a centrifugal membrane filter having a 30,000 MW cutoff (Millipore). Electrophoretic Mobility Shift Assays Electrophoretic mobility shift assays (EMSAs) were performed essentially as described [25]. Oligonucleotides containing the Cro binding site (C-OR3) were as described [25]. Oligonucleotides used to control for nonspecific binding (CM1) were modified from Blackwell [30] as follows: 5⬘-AGCTCCCCCACCACGTGGTGCCTGA-3⬘ and 5⬘AGCTTCAGGCACCACGTGGTGGGGG-3⬘. Oligonucleotides were annealed, end-labeled with 32P-dATP to a specific activity of 50,000 cpm/l and purified on NICK spin columns (Pharmacia). Cell extracts were diluted into 10 l KP200 assay buffer (20 mM potassium phosphate, pH 7.0, 200 mM potassium chloride, 1 mM EDTA, 5% (v/v) glycerol, 150 g/ml bovine serum albumin) or into assay buffer plus 1:10 dilution of anti-VTB or anti-hsp70 polyclonal antisera, and incubated at room temperature for 10 min. Diluted extracts were mixed with 10 l assay buffer containing approximately 50,000 cpm radiolabeled probe and incubated at room temperature for 30 min. Samples were loaded onto a 12.5% polyacrylamide gel and electrophoresed for 2 h at 200 V. Gels were dried under vacuum and exposed to autoradiography film overnight. Gb 3 Binding Assays Overlay assay. Purified galactosyl ceramide or globotriaosyl ceramide (2 g) (dissolved in chloroform: methanol, 2:1) was adsorbed to silica-coated plates (Macherey-Nagel). Plates were blocked overnight in 1% (w/v) gelatin at 37°C and then washed in 50 mM TBS. Plates were incubated 1.5 h at room temperature with 4 g VTB or a 1:30 dilution of chromatofocusing fractions 1– 4 in TBS. Plates were washed and then incubated with a 1:1000 dilution of polyclonal anti-VTB for 1.5 h, washed again, and incubated for 1.5 h with a 1:2000 dilution of goat anti-rabbit horseradish peroxidase-conjugated antibody (Bio-Rad). Plates were developed in 4-chloro-1-naphthol as described above. Matrix assay. Periplasmic and soluble extracts were diluted in KP200 assay buffer and incubated with radiolabeled C-OR3 probe in a total volume of 20 l assay buffer as described above (for EMSA). DNA–protein samples were then mixed with a 10-l slurry of a control matrix or a Gb 3-coated matrix [31] for at least 1 h. Samples were washed three times with 20 l assay buffer, matrix was resuspended in 10 l assay buffer, and radioactivity quantitated in a  scintillation counter. The assay was repeated three times with similar results. Intracellular Localization C-OR3 oligonucleotides containing a poly(A) 5⬘ overhang were annealed, end-labeled with Oregon Green-dUTP (Molecular Probes), and purified on NICK spin columns (Pharmacia). Fluorescein isothiocyanate (FITC)-labeled COR3 oligonucleotide was custom prepared by Gibco/BRL. Labeled oligonucleotide (0.8 –1.0 g) was incubated in KP200 buffer (minus glycerol and bovine serum albumin), either alone or with 2.5 g VTB or approximately 2 g VTB–Cro, in a total volume of 20 l per sample, for 30 min at room temperature to allow anealing. Following incubation, samples were immediately added to Vero cells grown on coverslips overnight to approximately 80% confluency in ␣-MEM (minimal essential medium) supplemented with 5% fetal bovine serum and 40 g/ml gentamycin. Cells were incubated for 1 h at 37°C in the presence of 20 l VTB-Cro or VTB/DNA samples or with 1 g FITC-VTB [19] in 250 l medium
119
plus serum. Medium was removed, replenished with 2 ml fresh medium, and cells were incubated at 37°C for an additional 2 h. Cells were cooled to 4°C, washed with cold PBS, fixed in 4% formaldehyde/ PBS, washed again, and mounted onto glass microscope slides in fluorescent mounting medium (DAKO). Cells were examined by laser confocal microscopy using a TCS 4D confocal microscope (Leica) with krypton/argon laser source, blue excitation filter for FITC and Oregon green detection, and a 40⫻ objective lens. For each image, four multiple scanning passes were made and averaged [9] or a single central optical slice was selected. For triple labeling, after fixation, the cells were permeabilized with methanol at ⫺20°C for 3 min, washed with cold PBS, and immunostained for VTB. Cells were incubated with 1% bovine serum albumin (BSA) in PBS for 30 min at room temperature, washed, and then incubated with 10 g/mL PH1 Mab anti-VTB antibody [32] in 1% BSA/PBS for 30 min. Cells were washed and then incubated with TRITC-conjugated goat anti-mouse antibodies in 1% BSA/PBS for 30 min. Cells were washed well, incubated with the nuclear stain DAPI (4⬘-6-diamidino-2-phenylindole) (0.05 g/ml) for 2–5 min, washed, and fixed again in 4% formaldehyde/PBS. Each intracellular fluor (FITC, TRITC, and DAPI) was monitored using an Axiomat fluorescence microscope with appropriate filters using a digital camera. The images were superimposed to determine coincidence.
RESULTS
Molecular modeling of the VTB pentamer predicts two major cleft sites for Gb 3 interaction [33]. The VTB crystal structure reveals that the carboxyl terminus of each monomer protrudes from the pentameric surface opposing these proposed Gb 3 binding sites [34, 35]. Indeed, previous studies have indicated that VTB can accommodate the addition of small peptides to the carboxyl terminus without compromising Gb 3 binding [36, 37]. Likewise, crystal structure analyses of Cro dimers bound to DNA indicate that the amino terminus of the Cro monomer lies opposite to the ␣helices mediating DNA binding [24, 38]. Thus Cro protein could be coupled to the carboxyl terminus of VTB to generate a fusion protein with intact spatially separate (opposite) Gb 3 and DNA binding sites. The Cro gene encoded by plasmid pCroRS was subcloned downstream of, and in frame with, the gene encoding the VTB subunit (Fig. 1). The fusion replaces the stop codon for VTB with six nucleotides specifying an EcoRI restriction site. This translates into the insertion of two amino acids, glutamic acid and phenylalanine, between the VTB and Cro proteins. The VTB– Cro fusion gene was subcloned into the pKK 223-3 expression vector for protein production in E. coli. Total bacterial cell extracts from VTB–Cro producing E. coli clones were prepared and analyzed by immunoblot using polyclonal antiVTB (Fig. 2A). Expression of VTB–Cro in E. coli gave three major anti VTBreactive bands with apparent molecular weights of 8, 14, and 16 kDa (Fig. 2A, lane 2). The 8-kDa band migrates to a similar position as purified recombinant VTB (Fig. 2A, lane 1) and may represent either cleavage of the fusion protein, or a premature termination product.
120
FACCHINI AND LINGWOOD
mic and soluble fraction extracts prepared from E. coli-expressing VTB–Cro were analyzed by EMSA. Ten-fold dilutions of each extract were independently incubated with 32P-radiolabeled oligonucleotides corresponding to the OR3 Cro DNA-binding site (C-OR3). With each extract, two specific protein–DNA complexes which migrated more slowly than the unbound probe alone were formed: a single dominant protein–DNA complex (I) and a larger minor complex (II) (Fig. 4, cf. lanes 4 and 8 with lane 1). These complexes were not evident when the probe was incubated with control periplasmic or soluble extracts prepared from control E. coli (Fig. 4, lanes 3 and 7). Furthermore, these
FIG. 1. Schematic of the fusion construct, pVTB–Cro. Refer to Materials and Methods for details of construction. The fusion replaces the stop codon for VTB with an EcoRI restriction enzyme site (GAATTC) and results in the insertion of glutamic acid and phenylalanine into the amino acid sequence.
To localize fusion protein expression within bacteria, overnight cultures of VTB–Cro-producing cells were separated into periplasmic, soluble, and soluble plus insoluble fractions, and extracts were analyzed by immunoblot. The lower 8-kDa band appears in both periplasmic and soluble fractions (Fig. 2B, lanes 1 and 2), while the upper bands are expressed mainly in the soluble and soluble plus insoluble fractions (Fig. 2B, lanes 2 and 3). Therefore, the majority of VTB–Cro fusion protein remains intracellular, unlike VT1 holotoxin or VTB subunits, which localize to the periplasmic space [18]. To determine the Gb 3-binding capacity of the VTB– Cro fusion protein, VTB–Cro was separated from the truncated VTB product in crude soluble fraction extracts by chromatofocusing [18]. VTB–Cro, which has a predicted isoelectric point of approximately 9.3, readily separated from VTB, which has an observed isoelectric point of 5.6. Silica plates were spotted with either 2 g of galactosyl ceramide (GC) or 2 g of Gb 3 and incubated with purified recombinant VTB or with various chromatofocusing fractions. Binding of VTB and VTB– Cro to the glycolipids was visualized by overlay with polyclonal anti-VTB antibody. All fractions tested exhibited binding to Gb 3 but not GC (Fig. 3B). Binding of VTB–Cro to Gb 3 was comparable to binding observed with recombinant VTB alone. Immunoblot analysis revealed the presence of VTB–Cro in all four fractions and the absence of VTB in fractions 1 and 2 (Fig. 3A). Thus, VTB–Cro fusion protein exhibits specific binding to Gb 3. Wild-type Cro repressor binds strongly to DNA as a dimer [24]. To investigate whether the VTB–Cro fusion protein retains DNA binding capacity, crude periplas-
FIG. 2. Fusion of Cro to VTB generates a soluble 14-kDa chimeric protein. (A) Immunoblot analysis of total cell extract from TB1 E. coli transformed with the pVTB–Cro plasmid (VTB–Cro, lane 2). Purified, recombinant VTB subunit is shown for comparison (VTB, lane 1). VTB and VTB–Cro were detected using rabbit polyclonal anti-VTB. Molecular weight standards are indicated on the left. (B) Localization of VTB–Cro protein within the bacterial cell. Immunoblot analysis of subcellular fraction extracts (15 g) from E. coli expressing VTB–Cro. Periplasmic fluid extracted from cells subjected to osmotic shock (P, lane 1). Cells from (P) were resuspended, and sonicated, and supernatant was collected as a soluble extract (S, lane 2). Cells from (S) were homogenized in SDS sample buffer and collected as soluble plus insoluble extract (S ⫹ IN, lane 3). Molecular weight standards are indicated on the right.
VTB-CRO CHIMERA TARGETS DNA TO THE NUCLEUS
FIG. 3. VTB–Cro binds to Gb 3, similar to wild-type VTB. (A) Immunoblot assay of chromatofocusing fractions 1– 4 (lanes 2–5) compared with purified recombinant VTB (lane 1). Soluble extracts from E. coli expressing VTB–Cro were separated by chromatofocusing. Fractions 1 and 2 contain VTB–Cro fusion protein, while fractions 3 and 4 contain VTB–Cro as well as VTB. (B) Overlay assay demonstrating binding of VTB and VTB–Cro to the glycolipid globotriaosyl ceramide (Gb 3). Purified VTB and fractions 1– 4 containing VTB–Cro were incubated with 2 g of GC or Gb 3 glycolipid adsorbed to silica plates. Glycolipid binding of VTB or VTB–Cro was visualized using polyclonal anti-VTB antiserum. The control plate was incubated with anti-VTB antiserum alone.
complexes were supershifted when probe and extracts were coincubated with polyclonal anti-VTB antiserum (Fig. 4, lane 5). Incubation with an unrelated antihsp70 antiserum did not change the original complexes (Fig. 4, lane 6). To confirm that binding of the VTB–Cro fusion protein was occurring through the Cro DNA-binding domain, radiolabeled C-OR3 oligonucleotides were incubated with 500 ng purified VTB subunit. No specific protein–DNA complexes were formed, indicating that the complexes recognized by the anti-VTB antibody contained VTB–Cro fusion protein and that DNA binding occurs through the Cro domain (Fig. 4, lane 2). To distinguish between specific binding of VTB–Cro to C-OR3 versus nonspecific DNA binding, extracts containing VTB–Cro fusion protein were incubated with radiolabeled oligonucleotides containing a mammalian transcription factor binding site (CM1). No specific pro-
121
tein–DNA complexes were observed (Fig. 4, lanes 10 and 11). In addition, incubation of extracts with radiolabeled C-OR3 plus an equal amount of unlabeled competitor C-OR3 probe led to a twofold reduction in protein–DNA complex formation (Fig. 4, lanes 8 and 9). Taken together, these results suggest that VTB–Cro binds DNA specifically at the Cro binding site. In order to target DNA to a specific subpopulation of cells, the VTB–Cro fusion protein must be able to bind DNA and Gb 3 simultaneously. To assay for this interaction, periplasmic, soluble, and control cell extracts were tested for the ability to bind a Gb 3 affinity matrix. To monitor DNA–protein complex formation, cell extracts were incubated with radiolabeled C-OR3 probe. The binding reactions were then incubated with the Gb 3 affinity matrix. As a control for nonspecific binding, binding reactions were also incubated with a control matrix. Following rinsing in binding buffer, the matrix was suspended in 10 l of buffer and total  emissions were monitored. To control for nonspecific adherence of the radiolabeled DNA probe to the matrix, C-OR3 probe was incubated with Gb 3 and control matrices in the absence of cellular extracts. A small amount of DNA probe bound nonspecifically to both the control and the Gb 3 matrices (Fig. 5, probe alone). Trace binding radioactivity on Gb 3 or control matrices incubated with control cellular extracts established the nonspecific background (Fig. 5, ⫺). When the radiolabeled DNA probe was incubated with extracts containing the VTB–Cro fusion protein, near background binding to the control matrix was observed (Fig. 5, ⫹, gray bars). By comparison, there was a significant, 7- to 28-fold increase in signal when the VTB–Cro–DNA probe samples were incubated with matrix containing Gb 3 (Fig. 5, ⫹, black bars, and data not shown). Thus, in the presence of VTB–Cro, the C-OR3 DNA probe binds to Gb 3. The Vero cell line expresses Gb 3 and is highly sensitive to the cytotoxic effects of VT1 holotoxin [39, 40]. To ascertain the capacity of VTB–Cro to deliver exogenous DNA to the cell nucleus, Vero cells were incubated in the presence of purified VTB–Cro and fluorescent-labeled C-OR3 DNA oligonucleotide or fluorescent COR3 DNA alone. To compare intracellular DNA localization by VTB–Cro with intracellular targeting of native VTB, cells were also incubated in the presence of recombinant VTB conjugated to fluorescein isothiocyanate (FITC–VTB). Although FITC–VTB staining of Vero cells was reported as Golgi-restricted [23], this phenotype is variable according to cell batch. Nearly all Vero cells exhibited perinuclear FITC–VTB fluorescent staining, consistent with Golgi/ER localization, as previously described (Fig. 6, panel A) [23]. In a subpopulation of cells, particularly following prolonged (3 h) incubation, FITC–VTB also localized to the nucleus (Figs. 6A and
122
FACCHINI AND LINGWOOD
FIG. 4. VTB–Cro binds specifically to a DNA fragment containing the consensus Cro binding site. Electrophoretic mobility shift assays on extracts from VTB-Cro expressing E. coli. Two specific fusion protein–DNA complexes are formed (I and II). The radiolabeled probe C-OR3 (lanes 1–9) corresponds to the Cro OR3 DNA binding consensus sequence. The probe CM1 (lanes 10 and 11), contains an E-box element for the Myc transcription factor. Lane 1, radiolabeled C-OR3 probe alone. Lanes 2–9, radiolabeled C-OR3 probe incubated with the following: lane 2, purified VTB protein; lane 3, soluble fraction extract from untransformed E. coli; lane 4, soluble fraction extract from E. coli transformed with pVTB–Cro; lane 5, as in lane 4 plus anti-VTB antiserum; lane 6, as in lane 4 plus anti-hsp70 antiserum; lane 7, periplasmic fraction extract from untransformed E. coli; lane 8, periplasmic extract from E. coli transformed with pVTB–Cro; lane 9, as in lane (8) plus 1:1 unlabeled competitor C-OR3 probe. Lanes 10 and 11, radiolabeled CM1 probe incubated with: lane 10, periplasmic fraction extract from untransformed E. coli; lane 11, periplasmic fraction extract from E. coli transformed with pVTB–Cro.
6B) as previously observed for cells of high VT sensitivity [9, 11]. Cells treated with fluorescent C-OR3 DNA alone demonstrated significantly less, cytoplasmic, staining (Figs. 6E and 6F), reflecting “passive” internalization of the oligonucleotide. However, cells treated with fluorescent C-OR3 DNA preincubated with VTB–Cro demonstrated a fluorescent staining pattern similar to that seen with FITC-VTB (Figs. 6C and 6D). Specifically, fluorescence-labeled DNA localized to the perinuclear region and within the nucleus of a proportion of cells. Treatment of cells with fluorescent C-OR3 DNA preincubated with native VTB resulted in a weak, diffuse staining pattern similar to passive DNA internalization (data not shown). Thus, VTB–Cro facilitates internalization of DNA from the cell surface to the nucleus. To verify the nuclear location of VTB-Cro mediated DNA transport, triple labeling of cells was performed. Low background incorporation of exogenous DNA la-
beling in VTB treated cells was not coincident with the intracellular retrograde targeting of VTB (Fig. 7B). In contrast, cells cotreated with DNA and VTB–Cro showed a marked increase in DNA internalization and the intracellular DNA was entirely coincident with the internalized VTB–Cro (Fig. 7A). The intracellular labeling was typical of ER and around the nucleus and distinct, localized, coincident staining for both DNA and VTB–Cro was seen within the DAPI-labeled nuclei. Not all cells within the field are labeled with VTB (or VTB–Cro ⫹ DNA). The retrograde trafficking of VT though the Golgi/ER is prevented by brefeldin A (BFA) [41] which collapses the Golgi structure [42]. Internalization of VT can occur through both clathrin-dependent and caveolin-independent mechanisms [43]. One such caveolin-dependent mechanism may bypass the Golgi and deliver the toxin directly to the ER. This mechanism is operative at higher VT concentrations and is sensitive to filipin
VTB-CRO CHIMERA TARGETS DNA TO THE NUCLEUS
FIG. 5. VTB–Cro mediates DNA binding to Gb 3. Periplasmic and soluble fraction extracts from untransformed (⫺) and pVTB–Cro transformed (⫹) E. coli were preincubated with radiolabeled C-OR3 DNA probe. VTB–Cro–DNA complexes or C-OR3 DNA probe alone were incubated with either a control matrix (stippled bars) or a Gb 3 matrix (black bars), which was washed, and counted.
[44], which complexes with cholesterol within caveoli at the cell surface [45]. Confocal microscopy of BFA- or filipin-pretreated cells (Fig. 8) shows that the VTB–Cro mediated intranuclear localization of DNA is prevented in BFA-, but not filipin-treated cells. After BFA treatment of cells, VT is concentrated in the collapsed Golgi/TGN [44]. The highly restricted, intense DNA fluorescence seen after BFA treatment (Figs. 8B and 8D) is consistent with such a location. DISCUSSION
We have shown that a fusion protein linking VTB with the Cro DNA binding protein retains specific binding to both the VTB cell surface receptor, Gb 3, and the consensus Cro operator DNA sequence, OR3. This study demonstrates that both VTB and Cro proteins can accommodate the addition of up to 70 amino acids at their carboxyl and amino termini, respectively, without substantial loss of protein function. Individually, these molecules may have practical applications for the construction of a wide array of chimeric proteins designed to (i) deliver conjugates to intracellular Golgi, ER, and nuclear compartments or (ii) to bind conjugates to specific DNA molecules. Moreover, we have demonstrated that the VTB–Cro chimera mediates delivery of exogenous DNA to the nuclei of Gb 3 positive cells, suggesting a potential alternative DNA delivery system to complement existing viral and cationic lipidbased vectors. Although the mechanism of VT transit from the ER/nuclear envelope to the nucleus has yet to be defined, it is possible that this transit avoids a cytosolic step. VTB contains no nuclear localization
123
motifs. It may be that the toxin translocates across the nuclear membrane as other toxins transit the ER/endosome [46]. In selecting the DNA binding protein to be fused to VTB, several criteria were considered. First, given the small size of the VTB subunit (69 amino acids), the chimera should be small, so as to limit steric interference with VTB–receptor binding. Second, to ensure that DNA remains tightly associated with the delivery protein during adherence, entry, and intracellular trafficking, the candidate protein should exhibit high-affinity DNA binding. This binding should be specific for a consensus DNA sequence with little homology to known mammalian regulatory DNA elements to ensure that the protein will be biologically inert in mammalian cells. The Cro repressor protein fulfils all of the above criteria. The Cro monomer, at only 66 amino acids in length, is the smallest known DNA-binding protein. The affinity of Cro dimers for the OR1, OR2, and OR3 DNA sites is approximately 10 10–10 12 M ⫺1 [47]. By comparison, the affinity of the VTB pentamer for immobilized Gb 3 is approximately 10 8–10 9 M ⫺1 [16]. Fusion of Cro to VTB may even enhance the Cro–DNA interaction, as prevention of Cro dimer dissociation results in increased DNA binding affinity [48]. Although Cro functions as a transcriptional repressor of bacteriophage promoters containing adjacent Cro binding sites, a BLAST search failed to detect any of the complete 17-bp consensus sites in known human DNA sequences. In addition, we observed that VTB–Cro did not bind to an unrelated mammalian transcription factor binding element, suggesting that delivery of Cro to human cells should not globally repress transcription. More importantly it is the N-terminal domain of Cro which binds DNA [24] while the C-terminus of VTB can be modified without affecting receptor binding [36]. Thus it was possible that the chimera as designed could retain both binding specificities. Expression of VTB–Cro in E. coli yielded two protein products: the VTB–Cro fusion protein and a smaller VTB-like protein. This smaller protein may result from premature termination of translation, or proteolytic cleavage. It is not known whether stable Cro-like peptides were also formed. Despite such potential interfering components, we detected one major and one minor VTB–Cro–DNA complex by EMSA. The presence of two complexes suggests different stoichiometric combinations of VTB-Cro and, possibly, VTB proteins in these aggregates. While Cro dimerization is a prerequisite for high-affinity DNA interaction, it is not known whether B subunit pentamerization is a requirement for Gb 3 binding. The chimera was able to bind DNA and Gb 3 simultaneously, indicating that either the monomer of each species is able to bind its respective ligand, that the conjugation serves to provide the ad-
124
FACCHINI AND LINGWOOD
FIG. 6. VTB–Cro targets DNA to nuclei of Gb 3-positive cells. Fluorescent confocal micrographs of Vero cells incubated in the presence of fluorescein isothiocyanate-conjugated recombinant VTB (FITC-VTB) (A and B), with purified VTB-Cro pre-incubated with Oregon greenlabeled C-OR3 DNA (OG-DNA) (C and D), or with OG-DNA alone (E and F). (A, C, and E) Composite images of all optical sections through the cells. Arrows point to cells depicted at right. (B, D, and F) A single optical section through the approximate center of the cell, showing nuclear localization of the fluorescent molecules. C and E are matched exposures (original magnifications: A, C, E, 100⫻; B, F, 500⫻; E, 360⫻).
ditional interactions necessary to stabilize the binding site as in the natural oligomers or that the conjugate achieves a multimeric state to satisfy such requirements. Further investigation of the stoichiometry of VTB–Cro aggregates may yield insights into the glycolipid and DNA binding capacities of VTB and Cro, respectively. The single confocal optical sections bissecting the nuclei, clearly show that exogenous DNA is transported by VTB–Cro to a restricted region within the nucleus. Separate studies indicate that this is the nucleolus [12]. VT holotoxin is a ribosomal RNA glycanase and targeting the site of RNA processing may increase efficacy. However, DNA may also be a target [13, 49]. In terms of effective DNA delivery for transfection, this may require dissociation from the Cro element. The size of the DNA sequence that can be transported into cell nuclei has yet to be determined, and hence the ability to transcribe VTB–Cro targeted
DNA has yet to be determined. However, nuclear targeting of short (e.g., anti-sense) sequences is useful. While all nuclei are not labeled, it is likely that ER 3 nuclear transit is a continuous process and incubation beyond 3 h should increase nuclear labeling efficiency. Moreover, cells initially unstained should internalize the VTB-Cro/DNA complex during subsequent cell cycle transit [50]. Our finding that BFA inhibits DNA nuclear entry shows that the DNA reaches the nucleus by retrograde transit via the Golgi, within the endomembrane system, and is thus protected from cytosolic nucleases. The factors which determine nuclear transit of VT are as yet unclear. VT does not contain nuclear localization or Golgi to ER retrieval motifs [51, 52]. The retrograde transport is likely a function of the glycolipid receptor rather than the toxin per se. On ligation, CD19, a B cell marker containing N-terminal sequence similarity to VTB and which binds Gb 3, is also trans-
VTB-CRO CHIMERA TARGETS DNA TO THE NUCLEUS
125
FIG. 7. Triple labeling of nuclei, internalized VTB, and exogenous DNA. Vero cells were treated with VTB or VTB-Cro together with FITC-DNA (green). After 3 h at 37°C cells were permeabilized with methanol and immunostained with anti VTB using RITC conjugated anti species antibodies (red). Nuclei were stained with DAPI (blue). Coincident staining of FITC–DNA and VTB–Cro (or VTB) is yellow. (A) Cells treated with VTB–Cro; arrows indicate cells with nuclear labeling. (B) Cells treated with VTB; arrow indicates cell with noncoincident DNA and VTB immunostaining (original magnification, 180⫻).
ported to the ER/nucleus but only in Gb 3-containing cells [53]. Retrograde transport of VT to the ER/nucleus is associated with an increased content of short fatty acid containing Gb 3 isoforms and high VT susceptilility [9]. These isoforms are greatly increased in
drug-resistant Gb 3 expressing tumour cells [9, 11], suggesting that drug-resistant tumors might be amenable to VTB–Cro/DNA-mediated therapy. Longer fatty acid containing Gb 3 isoforms are associated with reduced VT susceptibility and retrograde traffic to the Golgi
FIG. 8. Effect of BFA and/or filipin on VTB–Cro-mediated nuclear targeting of exogenous DNA monitored by confocal microscopy. Vero cells were pretreated with BFA or filipin for 30 min prior to addition of VTB–Cro and FITC-labeled DNA. Internalization of DNA after a further 3 h was monitored by confocal fluorescence microscopy and a single (central) optical section is shown. (A) Cells pretreated with MEM alone; (B) cells pretreated with BFA; (C) cells pretreated with filipin; (D) cells pretreated with BFA and filipin; (E) VTB–Cro, FITC–DNA-treated cells without preincubation; (F) cells treated with FITC–DNA without VTB–Cro. Labeling in (B) and (D) is not within the nucleus (original magnification, 500⫻). 126
VTB-CRO CHIMERA TARGETS DNA TO THE NUCLEUS
only [9]. This trafficking discrimination between Gb 3 isoforms may relate the physical restrictions of vesicle formation [54]. The use of modified bacterial toxins as delivery systems in eukaryotic cells is becoming increasingly popular (reviewed in [46]). In their natural forms, many toxins have developed distinct mechanisms to gain entry to the cytosol where they exert their cytotoxic effects. Nontoxic derivatives of Pseudomonas exotoxin A (exoA), pertussis toxin B oligomer, anthrax toxin protective antigen, and verotoxin B subunit have successfully presented heterologous peptide epitopes to major histocompatibility complex class I molecules [37, 55– 58]. These toxin–peptide conjugates show promise as new vaccine strategies for the induction of T lymphocyte immunity. Similarly, toxin domains have also been used to improve exogenous DNA delivery to cell lines. Fusion proteins combining the translocation domains of either diptheria toxin (DT) or exoA with the GAL4 DNA binding domain and an ErbB2-specific antibody fragment augmented DNA delivery and reporter gene expression in cells expressing ErbB2 tumor antigen [5, 7]. Likewise, incorporation of the DT translocation domain [6] or cholera toxin (CT) B subunit [8] in DNA–polycation complexes increased transfection efficiencies of exogenous DNA. These increases in reporter gene expression likely involve toxin-mediated release of DNA from endosomal (DT) or ER (exoA, cholera toxin) compartments into the cytosol. VTB, unlike DT, CTB, or exoA, undergoes retrograde transport, in part, to the nucleus. Thus, VTB–Cro has the potential to deliver exogenous DNA via vesicular retrograde traffic to the nucleus and thus not access the nuclease containing cytosol [2]. Considerations such as the upper size limit of exogenous DNA which can be efficiently delivered to the nucleus by VTB–Cro, separation of the Cro/DNA complex and the possible requirement for condensation of larger DNA molecules may impact on the levels of transgene expression. Nevertheless, the VTB–Cro fusion protein may find potential applications in intracellular targeting, DNA vaccine, and gene delivery research. We thank M. Mossing (University of Notre Dame, Indiana) for the kind gift of pCroRS plasmid. We thank P. Tam for assistance with confocal imaging. This work was supported by CHIR Grant MT13073.
Lukacs, G. L. (1999). Metabolic instability of plasmid DNA in the cytosol: A potential barrier to gene transfer. Gene Ther. 6, 482– 497. 3.
Kay, M. A., Liu, D., and Hoogerbrugge, P. M. (1997). Gene therapy. Proc. Natl. Acad. Sci. USA 94, 12,744 –12,746.
4.
Anderson, W. F. (1998). Human gene therapy. Nature 392, 25–30.
5.
Fominaya, J., and Wels, W. (1996). Target cell-specific DNA transfer mediated by a chimeric multidomain protein. Novel non-viral gene delivery system. J. Biol. Chem. 271, 10,560 – 10,568.
6.
Fisher, K. J., and Wilson, J. M. (1997). The transmembrane domain of diphtheria toxin improves molecular conjugate gene transfer. Biochem. J. 321, 49 –58.
7.
Uherek, C., Fominaya, J., and Wels, W. (1998). A modular DNA carrier protein based on the structure of diphtheria toxin mediates target cell-specific gene delivery. J. Biol. Chem. 273, 8835– 8841.
8.
Barrett, L., Logan, A., Berry, M., Ying, W., Gonzalez, A., Baird, A., and LW., S. (1999). Targeted transfection of neuronal cells using a poly(D-lysine)-cholera-toxin b chain conjugate. Biochem. Soc. Trans. 27, 851– 857.
9.
Arab, S., and Lingwood, C. (1998). Intracellular targeting of the endoplasmic reticulum/nuclear envelope by retrograde transport may determine cell hypersensitivity to Verotoxin: Sodium butyrate or selection of drug resistance may induce nuclear toxin targeting via globotriosyl ceramide fatty acid isoform traffic. J. Cell Physiol. 177, 646 – 660.
10.
Lingwood, C. A., Khine, A. A., and Arab, S. (1998). Globotriaosyl ceramide (Gb 3) expression in human tumour cells: Intracellular trafficking defines a new retrograde transport pathway from the cell surface to the nucleus, which correlates with sensitivity to verotoxin. Acta Biochem. Polonic. 45, 351–359.
11.
Lala, P., Ito, S., and Lingwood, C. A. (2000). Transfection of MDCK cells with the MDR1 gene results in a major increase in globotriaosyl ceramide and cell sensitivity to verocytotoxin: Role of P-gp in glycolipid biosynthesis. J. Biol. Chem. 275, 6246 – 6251.
12.
Falguieres, T., Baron, C., Mallard, F., Lingwood, C., Goud, B., Salamero, J., and Johannes, L. (in press). Cell type differences in sensitivity to Shiga toxin-mediated inhibition of protein biosynthesis can be correlated with the existence of two distinct toxin transport pathways. Mol. Biol. Cell., in press.
13.
Brigotti, M., Accorsi, P., Carnicelli, D., Rizzi, S., Vara, A. G., Montanaro, L., and Sperti, S. (2001). Shiga toxin 1: Damage to DNA in vitro. Toxicon 39, 341–348.
14.
Endo, Y., Tsurugi, K., Yutsudo, T., Takeda, Y., Ogasawara, K., and Igarashi, K. (1988). Site of the action of a vero toxin (VT2) from Escherichia coli O157:H7 and a Shiga toxin on eukaryotic ribosomes. Eur. J. Biochem. 171, 45–50.
15.
Austin, P. R., and Hovde, C. J. (1995). Purification of recombinant shiga-like toxin type I B subunit. Protein Expression Purif. 6, 771–779.
16.
Head, S., Karmali, M., and Lingwood, C. A. (1991). Preparation of VT1 and VT2 hybrid toxins from their purified dissociated subunits: Evidence for B subunit modulation of A subunit function. J. Biol. Chem. 266, 3617–3621.
17.
Lingwood, C. A. (1993). Verotoxins and their glycolipid receptors. In “In: Sphingolipids. A. Functions and Breakdown Products. Advances in Lipid Research” (R. Bell, Y. A. Hannun, and A. Merrill, Jr., Eds.), Academic Press, San Diego.
18.
Ramotar, K., Boyd, B., Tyrrell, G., Gariepy, J., Lingwood, C. A., and Brunton, J. (1990). Characterization of Shiga-like toxin 1 B
REFERENCES 1.
2.
Chowdhury, N. R., Wu, C. H., Wu, G. Y., Yerneni, P. C., Bommineni, V. R., and Chowdhury, J. R. (1993). Fate of DNA targeted to the liver by asialoglycoprotein receptor-mediated endocytosis in vivo. Prolonged persistence in cytoplasmic vesicles after partial hepatectomy. J. Biol. Chem. 268, 11,265–11,271. Lechardeur, D., Sohn, K.-J., Haardt, M., Joshi, P. B., Monck, M., Graham, R. W., Beatty, B., Squire, J., O’Brodovich, H., and
127
128
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
FACCHINI AND LINGWOOD subunit purified from overproducing clones of the SLT-1 B cistron. Biochem. J. 272, 805– 811. Khine, A. A., and Lingwood, C. A. (1994). Capping and receptor mediated endocytosis of cell bound verotoxin(Shiga-like toxin). 1. Chemical identification of an amino acid in the B subunit necessary for efficient receptor glycolipid binding and cellular internalization. J. Cell Physiol. 161, 319 –332. Sandvig, K., Ryd, M., Garred, Ø., Schweda, E., and Holm, P. K. (1994). Retrograde transport from the Golgi complex to the ER of both Shiga toxin and the nontoxic Shiga B-fragment is regulated by butyric acid and cAMP. J. Cell Biol. 126, 53– 64. Sandvig, K., Garred, Ø., Prydz, K., Kozlov, J., Hansen, S., and van Deurs, B. (1992). Retrograde transport of endocytosed Shiga toxin to the endoplasmic reticulum. Nature 358, 510 – 512. Mallard, F., Antony, C., Tenza, D., Salamero, J., Goud, B., and Johannes, L. (1998). Direct pathway from early/recycling endosomes to the Golgi apparatus revealed through the study of shiga toxin B-fragment transport. J. Cell Biol. 143, 973–990. Kim, J. H., Lingwood, C. A., Williams, D. B., Furuya, W., Manolson, M. F., and Grinstein, S. (1996). Dynamic measurement of the pH of the Golgi complex in living cells using retrograde transport of the verotoxin receptor. J. Cell Biol. 134, 1387–1399. Brennan, R. G., Roderick, S. L., Takeda, Y., and Matthews, B. W. (1990). Protein-DNA conformational changes in the crystal structure of a lambda Cro-operator complex. Proc. Natl. Acad. Sci. USA 87, 8165– 8169. Jana, R., Hazbun, T. R., Mollah, A. K., and Mossing, M. C. (1997). A folded monomeric intermediate in the formation of lambda Cro dimer-DNA complexes. J. Mol. Biol. 273, 402– 416. Takeda, Y., Folkmanis, A., and Echols, H. (1977). Cro regulatory protein specified by bacteriophage lambda. Structure, DNA-binding, and repression of RNA synthesis. J. Biol. Chem. 252, 6177– 6183. Johnson, A., Meyer, B. J., and Ptashne, M. (1978). Mechanism of action of the cro protein of bacteriophage lambda. Proc. Natl. Acad. Sci. USA 75, 1783–1787. Laemmeli, U. K. (1970). Cleavage of structural proteins during the assembly of head of bacteriophage T4. Nature 227, 680 – 685. Boyd, B., Richardson, S., and Gariepy, J. (1991). Serological responses to the B subunit of Shiga-like toxin 1 and its peptide fragments indicate that the B subunit is a vaccine candidate to counter the action of the toxin. Infect. Immun. 59, 750 –757. Blackwell, T. K., Kretzner, L., Blackwood, E. M., Eisenman, R. N., and Weintraub, H. (1990). Sequence-specific DNA binding by the c-Myc protein. Science 250, 1149 –1151. Armstrong, G. D., Fodor, E., and Vanmaele, R. (1991). Investigation of Shiga-like toxin binding to chemically synthesized oligosaccharide sequences. J. Infect. Dis. 164, 1160 –1167. Boulanger, J., Petric, M., Lingwood, C. A., Law, H., Roscoe, M., and Karmali, M. (1990). Neutralization receptor-based immunoassay (NeutrELISA) for detection of neutralizing antibodies to Escherchia coli verocytotoxin 1. J. Clin. Microbiol. 28, 2830 – 2833. Nyholm, P. G., Magnusson, G., Zheng, Z., Norel, R., BinningtonBoyd, B., and Lingwood, C. A. (1996). Two distinct binding sites for globotriaosyl ceramide on verotoxins: Molecular modelling and confirmation by analogue studies and a new glycolipid receptor for all verotoxins. Chem. Biol. 3, 263–275. Stein, P. E., Boodhoo, A., Tyrrell, G. J., Brunton, J. L., and Read, R. J. (1992). Crystal structure of the cell-binding B oligomer of verotoxin-1 from E. coli. Nature 355, 748 –750.
35.
36.
37.
38.
39.
40.
41. 42.
43.
44.
45.
46.
47.
48.
49.
50.
51.
52.
53.
Fraser, M., Chernaia, M., Kozlov, Y., and James, M. (1994). Crystal structure of the holotoxin from Shigella dysenteriae at 2.5A resolution. Struct. Biol. 1, 59 – 64. Johannes, L., Tenza, D., Anthony, C., and Goud, B. (1997). Retrograde transport of KDEL-bearing B-fragment of Shiga toxin. J. Biol. Chem. 272, 19,554 –19,561. Lee, R. S., Tartour, E., van der Bruggen, P., Vantomme, V., Joyeux, I., Goud, B., Fridman, W. H., and Johannes, L. (1998). Major histocompatibility complex class I presentation of exogenous soluble tumor antigen fused to the B-fragment of Shiga toxin. Eur. J. Immunol. 28, 2726 –2737. Anderson, W. F., Ohlendorf, D. H., Takeda, Y., and Matthews, B. W. (1981). Structure of the cro repressor from bacteriophage lambda and its interaction with DNA. Nature 290, 754 –758. Karmali, M. A., Petric, M., Lim, C., Fleming, P. C., and Steele, B. T. (1983). Escherichia coli cytotoxin, hemolytic uremic syndrome, and hemorrhagic colitis. Lancet 2, 1299 –1300. Pudymaitis, A., Armstrong, G., and Lingwood, C. A. (1991). Verotoxin resistant clones are deficient in the glycolipid globotriosyl ceramide: Differential basis of mutant phenotype. Arch. Biochem. Biophys. 286, 448 – 452. Donta, S., Tomicic, T., and Donohue-Rolfe, Ø. (1995). Inhibition of Shiga-like toxins by Brefeldin A. J. Infect. Dis. 171, 721–724. Lippincott-Schwartz, J., Yuan, L., Bonifacino, J., and Klausner, R. (1989). Rapid redistribution of Golgi proteins into the ER in cells treated with brefeldin A: evidence for membrane cycling from Golgi to ER. Cell 56, 810 – 813. Schapiro, F., Lingwood, C. A., Furuya, W., and Grinstein, S. (1998). pH-independent targeting of glycolipids to the Golgi complex. Am. J. Physiol. 274, 319 –332. Khine, A. A., and Lingwood, C. (submitted for publication). An alternative pathway of verotoxin-1/globotriaosyl ceramide internalization: Caveolar-mediated, Golgi-independent VT1/receptor transport. Rothberg, K. G., Heuser, J. E., Donzell, W. C., Ying, Y., Glenney, J. R., and Anderson, R. G. W. (1992). Caveolin, a protein component of caveolae membrane coats. Cell 68, 673– 682. Goletz, T. J., Klimpel, K. R., Leppla, S. H., Keith, J. M., and Berzofsky, J. A. (1997). Delivery of antigens to the MHC class I pathway using bacterial toxins. Hum. Immunol. 54, 129 –136. Kim, J. G., Takeda, Y., Matthews, B. W., and Anderson, W. F. (1987). Kinetic studies on Cro repressor– operator DNA interaction. J. Mol. Biol. 196, 149 –158. Jana, R., Hazbun, T. R., Fields, J. D., and Mossing, M. C. (1998). Single-chain lambda Cro repressors confirm high intrinsic dimer-DNA affinity. Biochemistry 37, 6446 – 6455. Brigotti, M., Carnicelli, D., Accorsi, P., Rizzi, S., Montanaro, L., and Sperti, S. (2000). 4-aminopyrazolo [3,4-d] pyrimidine (4APP) as a novel inhibitor of the RNA and DNA depurination induced by Shiga toxin 1. Nucleic Acids Res. 28, 2383–2388. Pudymaitis, A., and Lingwood, C. A. (1992). Susceptibility to verotoxin as a function of the cell cycle. J. Cell Physiol. 150, 632– 639. Lord, J. M., and Roberts, L. M. (1998). Toxin entry: Retrograde transport through the secretory pathway. J. Cell Biol. 140, 733–736. Jackson, M., Simpson, J., Girod, A., Pepperkok, R., Roberts, L., and Lord, J. (1999). The KDEL retrieval system is exploited by Pseudomonas exotoxin A, but not by Shiga-like toxin-1, during retrograde transport from the Golgi complex to the endoplasmic reticulum. J. Cell Sci. 112, 467– 475. Khine, A. A., Firtel, M., and Lingwood, C. A. (1998). CD77-dependent retrograde transport of CD19 to the nuclear membrane:
VTB-CRO CHIMERA TARGETS DNA TO THE NUCLEUS Functional Relationship between CD77 and CD19 during germinal center B-cell apoptosis. J. Cell Physiol. 176, 281–292. 54. Lingwood, C. A. (2000). Glycolipids and Bacterial Pathogenesis. In “Oligosaccharides in Chemistry and Biology-a Comprehensive Handbook” (B. Ernst, P. Sinay, and G. Hart, Eds.), Wileyvch, Wenheim, Germany. 55. Donnelly, J. J., Ulmer, J. B., Hawe, L. A., Friedman, A., Shi, X. P., Leander, K. R., Shiver, J. W., Oliff, A. I., Martinez, D., and Montgomery, D. (1993). Targeted delivery of peptide epitopes to class I major histocompatibility molecules by a modified Pseudomonas exotoxin. Proc. Natl. Acad. Sci. USA 90, 3530 – 3534. 56. Goletz, T. J., Klimpel, K. R., Arora, N., Leppla, S. H., Keith, J. M., and Berzofsky, J. A. (1997). Targeting HIV proteins to the major histocompatibility complex class I processing pathway Received March 20, 2001 Revised form received June 12, 2001 Published online August 6, 2001
129
with a novel gp120-anthrax toxin fusion protein. Proc. Natl. Acad. Sci. USA 94, 12,059 –12,064. 57.
Carbonetti, N. H., Irish, T. J., Chen, C. H., O’Connell, C. B., Hadley, G. A., McNamara, U., Tuskan, R. G., and Lewis, G. K. (1999). Intracellular delivery of a cytolytic T-lymphocyte epitope peptide by pertussis toxin to major histocompatibility complex class I without involvement of the cytosolic class I antigen processing pathway. Infect. Immun. 67, 602– 607.
58.
Haicheur, N., Bismuth, E., Bosset, S., Adotevi, O., Warnier, G., Lacabanne, V., Regnault, A., Desaymard, C., Amigorena, S., Ricciardi, C. P., Goud, B., Fridman, W. H., Johannes, L., and Tartour, E. (2000). The B subunit of Shiga toxin fused to a tumor antigen elicits CTL and targets dendritic cells to allow MHC class I-restricted presentation of peptides derived from exogenous antigens. J. Immunol. 165, 3301–3308.
