VIRAL IMMUNOLOGY Volume 5, Number 2, 1992 Mary Ann Liebert, Inc., Publishers Pp. 163-172
Construction of a Recombinant Bacterial Human CD4 Expression System Producing a Bioactive CD4 Molecule DANIEL E. MCCALLUS,' KENNETH E. UGEN,1 ALICE I. SATO,1'2 2 WILLIAM V. WILLIAMS,2 3 and DAVID B. WEINER1
ABSTRACT
protein expressed on helper T lymphocytes is a restriction element for major histocompatibility class II immune responses. This molecule is also used by the human immunodeficiency virus as its specific cellular receptor facilitating binding of virus to cells. As soluble forms of CD4 inhibit HIV infection in tissue culture, attention has focused on this molecule. Bacterially produced CD4 would facilitate studies of the biology of the CD4 molecule. However, bacterially expressed CD4 must be refolded for assumption of its interaction with conformâtionally dependant anti-C 1)4 monoclonal antibodies as well as the HIV-1 envelope protein gpl20. We report here the engineering of an external domain construct of the CD4 gene into a novel expression vector containing the nucleotide sequence encoding the pelB leader peptide of Envinia carotovara (pDAB, ), to facilitate correct folding of CD4 in bacteria. Monoclonal antibodies specific for important conformational epitopes of the CD4 molecule were able to bind bacterial colonies containing the pDABL/CD4 vector but not colonies with vector alone. Importantly, recombinant gpl20 produced in baculovirus bound specifically to bacterial colonies expressing the CD4 recombinant molecule. This system presents a simple screening mechanism for molecules that bind to the external domain of the CD4 glycoprotein. Vectors such as pDABL will also facilitate the production of large amounts of biologically active proteins in bacteria.
The CD4
the surface of certain human T lymphocytes, is a 55,000 molecular for interaction with class II major histocompatibility (MHC) molecules of antigen-presenting cells (24). The CD4 molecule has also been shown to be the cellular receptor for the human immunodeficiency virus (HIV), which binds CD4 through the HIV envelope glycoprotein gpl20 (8,14,18). Interaction of CD4+ permissive cells with cells displaying the envelope glycoproteins of HIV (e.g., HIV-1-infected cells) results in the formation of multinucleated giant cells (syncytia). This is initiated
molecule, expressed Theweight glycoprotein responsible CD4
on
vr
'iv1
'The Wistar Institute of Anatomy and Biology, the department of Medicine, University of Pennsylvania School of Medicine, and ^Children's Hospital of Philadelphia, Philadelphia, Pennsylvania. 163
McCALLUS ET AL.
by the binding of surface gpl20 on infected cells with the CD4 molecule expressed on uninfected cells (17). The region of gp 120 that binds CD4 is found in the carboxy-terminal end of the molecule and conserved amino acid sequences have been implicated in this binding (27). Site directed mutagenesis defines three discontinuous sites as required for binding to CD4. The CD4 molecule is composed of extracellular, transmembrane and intracellular segments. The extracellular segment is 370 amino acids in length and can be separated into four domains with homology to immunoglobulin variable domains (6,19). It has been found that the most membrane distal domain is sufficient for binding to gpl20 (7,21). However, constructs containing the first two domains appear to mediate more efficient binding to gpl20 in some systems. Attempts have been made to use soluble forms of the CD4 molecule (sCD4), lacking the transmembrane and cytoplasmic domains, to inhibit HIV infection and/or syncytia formation (9,10,13,26,29). Both full length sCD4, containing the four external domains, as well as sCD4 containing only the first one or two amino-terminal domains have been shown to bind gpl20 and block both HIV infection and syncytium formation. Several studies have also demonstrated that recombinant sCD4 produced by bacteria are able to bind HIV (5,12,28). However, bacterial expression systems are hampered by the fact that CD4, like many mammalian transmembrane proteins, is incorrectly processed by the bacterial cell such that the CD4 recombinant protein must be denatured and refolded to recover biological activity. Production of a more natural folding CD4-like molecule would present a number of advantages for the analysis of CD4 biology as well as investigations of CD4 reactivity with HIV envelope proteins. We have used the polymerase chain reaction (PCR) to amplify CD4 from human T lymphocyte lines and to insert the amplified molecule into a bacterial expression vector. This vector contained the pelB leader peptide of Erwinia carotovora. Expression of the CD4 gene segment in a biologically relevant form in Escherichia coli was achieved, as demonstrated by binding of conformationally dependent anti-CD4 monoclonal antibodies (mAb) as well as reactivity with recombinant gpl20 molecules. MATERIALS AND METHODS Bacterial strains. Escherichia coli DH5-alpha competent cells (BRL, Gaithersburg, MD) were used for transformation. Bacteria were grown in Luria broth containing 100 mg/ml ampicillin (LB/amp) (1,23). Enzymes and oligonucleotides. Restriction endonucleases and T4 DNA ligase were purchased from BRL. Enzyme reaction conditions were according to that of the supplier. Oligonucleotides for PCR primers and for Southern blotting were synthesized by the DNA Synthesis Facility of the Wistar Institute. Antibodies. Anti-CD4 mAbs SIM.2 (murine IgG2b) and SIM.4 (murine IgG,) were from the National Institute of Allergy and Infectious Diseases AIDS Research and Reference Reagent Program (Rockville, MD). Leu3a (murine IgG,) was from Becton-Dickinson (MountainView, CA). The Leu3a and SIM.4 are specific for the HIV-1 gpl20 binding epitope on CD4, and both block HIV-induced syncytium formation. SIM.2 binds a distinct epitope, but also blocks HIV-induced syncytium formation. Antireovirus type 3 mAb 9B.G5 (murine IgG2a) was the kind gift of Mark I. Greene (Department of Pathology and Laboratory Medicine, University of Pennsylvania, Philadelphia, PA). Plasmid construction. The DNA coding for the pectate lyase (pelB) signal peptide of Erwinia carotovara was synthesized according to standard methodology according to the published sequence (15) and made double-stranded by PCR. Restriction sites for Hindlll and EcoRl were incorporated into the primers for insertion into pUC 19. Positive transformants were selected and plasmid preps were made. ThepelB insert was excised by digestion with Hindlll and EcoRl and identified by agarose gel electrophoresis. CD4 amplification and cloning. RNA from SupTl cells (30) was reverse transcribed into cDNA using 200 U of MMLV reverse transcriptase (BRL) in a 20 u.1 reaction. The first domain of CD4 was amplified by PCR using primers corresponding to codons ( )4 to ( + )5 (primer A: ccagtcgacactcagggaaacaaagtg) and codons 116 to 125 (primer B: accgaattcgctctacaaggtcagggtcag) in a Programmable Thermal Controller (MJ Research; Cambridge, MA). Primer A contained a Sail restriction digestion site (underlined) while primer B contained an EcoRl restriction site (underlined) and also an added stop codon. The amplification program was 94°C for 60 sec, 52°C for 90 sec, and72°C for 120 sec. Following 30 cycles, the temperature was held at 72°C for 5 min. Positive amplification was determined by agarose gel electrophoresis and ethidium bromide —
164
BIOACTIVE CD4 IN BACTERIA CD4 Insert
Diagram of pDABL vector with CD4 insert. The leader sequence of thepelB gene of Erwinia caratovara was amplified by PCR using primers containing the the polylinker site. The amplified DNA was digested with Hindlll and EcoRI and ligated into pUC19 which had been digested with the same endonucleases and treated with calf intestinal alkaline phosphatase. The amplified CD4 gene was inserted into the polylinker using the Sail and EcoRI restriction sites. FIG. 1.
staining. The CD4 DNA and plasmid DNA was cut with the appropriate endonucleases and plasmid DNA was treated with calf intestinal phosphatase (Boehringer Mannheim, Indianapolis, IN). The CD4 DNA was ligated into pDABL using 1 U of T4 DNA ligase overnight at 16°C. Ligation mixtures were transformed into E. coli DH5-alpha competent cells as described by the manufacturer. Southern hybridization, restriction digestion and amplification. Competent E. coli transformed with the amplified CD4 gene segment were plated on LB/amp plates as described above. An internal oligonucleotide probe for the CD4 gene segment was used which corresponded to the antisense sequence of amino acids 20-25 (ctcttcttctggtaatc). The probe (100 ng) was labeled with 32P for 30 min using T4 polynucleotide kinase (1,23). Nitrocellulose filters (0.45 pun; Schleicher & Schuell, Keene, NH) were used to lift the transformed bacteria. Following alkaline lysis of the bacteria the filters were incubated with the labeled probe for 2 hr at 55°C. The filters were then washed two times at room temperature with 2x SSPE(1,23) and 0.1% SDS. The nitrocellulose filters were exposed to film for 48 hr. A colony that hybridized with the probe was selected for further study and was grown overnight in LB/amp. Plasmid DNA from this culture was prepared and was subjected to restriction digestion to liberate the CD4 gene segment with or without the pelB signal sequence gene. Plasmid DNA from the transformant was also amplified by PCR using the CD4 primers and primers for the 5' end of the pelB signal sequence and the 3' end of the CD4 gene segment. Protein expression. A bacterial clone possessing the CD4 gene segment inserted into pDABL was plated onto one half of LB/amp plates. The other side of the plates contained E. coli containing pUC 19. Following overnight growth, a 0.45-u.m nitrocellulose filter was placed on the bacterial plate. Filters were lifted to other LB/amp plates on which 100 p.1 of isopropyl-ß-thiogalactopyranoside (IPTG; 25 mg/ml; Stratagene, La Jolla, CA) had been spread and were then incubated for 4 hr at 37°C. Filters were then exposed to chlorofoam vapor to fix the bacteria for 15 min and incubated overnight (with shaking) in lysis buffer ( 100 mM Tris-Cl, pH 7.8, 150mMNaCl,5 mMMgCl2, 1.5% bovine serum albumin (BSA), 1 p-g/ml pancreatic DNAse I, and 40 p-g/ml lysozyme]. Filters were then blocked for 4 hr with 4% nonfat dry milk and 1% BSA in Tris-buffered saline (TBS; 20 mM Tris, 500 mM NaCl, pH7.5). Filters were incubated for 2 hr at room temperature with the 165
McCALLUS ET AL.
Amplification of CD4 by PCR. The CD4 gene was amplified by PCR from SupTl cell cDNA (lane 2), but was not amplified from HSB cell cDNA (lane 3) or Leu3a cell cDNA (lane 4). Primers for murine immunoglobulin variable region genes did not amplify SupTl cell cDNA (lane 5) but did amplify Leu3acDNA (lane 6). Size markers were phage k/Hindlll digest (lane 1) and X174///aeIII digest (lane 7). FIG. 2.
appropriate monoclonal antibody (mAb) diluted to 7.4 p.g/p.1 in TBS containing 1 % BSA and 0.1 % Tween-20 (TBS/BSA). Filters were washed extensively with TBS/BSA, then incubated for 2 hr at room temperature with goat anti-mouse immunoglobulin (horseradish peroxidase-labeled) (Fisher Biotech, Pittsburgh, PA) diluted to 0.8 p.g/ml in TBS/BSA. Filters were washed as above and the substrate, TMBlue (TSI, Worcester,
MA), was added for 5-10 min at room temperature. After development, filters were rinsed in deionized water and photographed. For binding of l25I-labeled gpl20, blocked filters were incubated for 2 hr at room the temperature with 125I-labeled gpl20 (MicroGeneSys, CT), at 500,000-1,000,000 cpm/ml, labelled by washed Filters Tween-20 were 0.1% 1% in and BSA (TBS/BSA). chloramine T method (31) TBS containing extensively with TBS/BSA and autoradiographed (Kodak XRP film) for 2-24 hr. RESULTS Plasmid construction. The signal sequence for the pelB protein of Erwinia carotovara was inserted into pUC 19 as shown in Fig. 1. A poly linker containing restriction sites for Xbal, Sail, and EcoRl was added to the end of this signal sequence to facilitate cloning of the desired gene(s). CD4 amplification and cloning. CD4 was amplified by PCR from Supt 1 cells but was not amplified from CD4~ HSB cells (Fig. 2). In addition, primers for immunoglobulin variable region genes did not amplify the CD4 gene segment. Amplified CD4 DNA was inserted into pDABL. The presence of this DNA was verified by Southern hybridization (Fig. 3) and restriction digestion of the transformant and by amplification of the plasmid mini-prep using the original CD4 primers as described below (Fig. 4). 166
BIOACTIVE CD4 IN BACTERIA
•
>•
*
I*» i
Southern hybridization to pDABL/CD4 transformants. Nitrocellulose lifts were made from LB/amp plates bacteria with pDABL/CD4 (A) or bacteria with the pDABL vector alone (B). The lifts were probed with 32P-labeled oligonucleotide specific for an internal sequence of the V1 loop of CD4. FIG. 3.
containing
Southern hybridization, restriction digestion and amplification. Following exposure to film for 48 hr, was observed that transformants that received the CD4 gene segment hybridized with the internal oligonucleotide probe, while transformants receiving plasmid alone did not react with this probe (Fig. 3). Positive colonies was grown and plasmid DNA prepared. The DNA was digested with Sail and EcoRI to liberate the CD4 insert or with Hindlll and £coRI to liberate the pelB signal sequence with the CD4 gene segment (see Fig. 1 for location of restriction sites). Amplification of this DNA with the CD4 primers yielded a 330-bp band identical in size to the liberated CD4 gene segment, while amplification of the DNA with the 5' pelB primer and the 3' CD4 primer yielded a 480-bp band identical in size to the liberated pelB/CD4 gene it
segment (Fig. 4).
Protein expression. Nitrocellulose lifts of colonies of E. coli containing pDABL/CD4 were probed with several anti-CD4 mAbs and an irrelevant mAb 9BG5 [directed against reovirus type 3 (31)]. The nitrocellulose lifts were prepared, cut into strips, and reacted with the anti-CD4 mAbs (Fig. 5). Positive reactions with pDABL/CD4 were observed with anti-CD4 mAbs Leu3a, SIM2, and SIM4. This was observed for E. coli either lysed or intact. Minimal background binding was observed with pUC19, while strong binding was observed with pDABL/CD4 expressing cells. The mAB 9BG5, anti-mouse-conjugated secondary antibody, or the TMB substrate alone showed no reactivity with either pUC19 or pDABL/CD4. Next, the ability of CD4 expressed in bacteria to react with gpl20 was directly assayed. E. coli was transformed with pUC19 or pDABL/CD4, bacterial colonies grown on LB/amp plates, and filters were lifted and induced with IPTG. l25I-labeled gpl20 was incubated with the filters, and autoradiography performed. No significant binding of l25I-labeled gpl20 was seen to the pUC19-transformed negative control filters (Fig. 6A). In marked contrast, extensive binding of l25I-labeled gpl20 to pDABL/CD4-transformed E. coli was observed (Fig. 6B). These results support and extend the results obtained using anti-CD4 antibodies. It is likely that CD4 expressed on the surface of the pDABL/CD4-transformed E. coli is expressed in a conformationally relevant form. 167
McCALLUS ET AL.
FIG. 4. Restriction digestion and amplification of pDAB[ /CD4 plasmid. The pDABL/CD4 construct was restricted with various enzymes and the restriction fragments run on a 2% agarose gel. Lane I: X174///aeIII MW markers; Lane 2: PCR amplification product of pDABL/CD4 plasmid with CD4 primers (see Materials and Methods); Lane 3: Sail and EcoRl digest of pDABL /CD4 plasmid to liberate the CD4 gene; Lane 4: PCR amplification product of pDABL/CD4 plasmid with the 5' primer for pelB and the 3' primer for CD4; Hindlll and EcoRl digest of pDABL/CD4 plasmid to liberate the CD4 gene; phase \/HindIII MW markers.
DISCUSSION The use of sCD4 as a therapeutic modality for the treatment of AIDS has received a great deal of study during recent years. Although sCD4 has been shown to inhibit the ability of HIV to infect in vitro and has been shown to block syncytium formation, there have been as yet no reports that have demonstrated clinical benefits from treatment with sCD4. The possibility that HIV uses other cellular receptors to gain entrance to permissive cells may abrogate the effectiveness of sCD4 as a treatment for AIDS. However, promising data from the in vitro experiments provide impetus for further studies into the protective effects of sCD4. CD4 has
also been observed to function as a restriction element on T cells for their interaction with MHC Class II bearing antigen-presenting cells. Interaction of human CD4 with class II has been demonstrated, but the exact residues involved in binding remain controversial (reviewed in ref. 11). Further investigation of ligandreceptor interactions between CD4 and MHC class II will enhance our understanding of antigen responsiveness and T-cell activation in a general sense. The forms in which sCD4 has been produced for in vitro studies have varied. Therefore, caution should be taken when comparing the different studies involving sCD4. Early studies using recombinant sCD4 lacking 168
BIOACTIVE CD4 IN BACTERIA
pUC
19
pDAB/CD4
ImJà Sin* 2 Sim 4
anti-mQp$e CQnjyjgate •
«.
substrate alone FIG. 5. Reactivity of bacteria containing pDABL/CD4 with anti-CD4 mAbs. Colonies of bacteria transformed with pUC19 (left side of strips) or pDABL/CD4 (right side of strips) were lifted from LB/amp plates with nitrocellulose filters. The filters were cut into strips and reacted with the mAbs shown. Specific reactivity with the anti-CD4 mAbs Leu3a,
SIM2, and SIM4 is evident, while control mAb 9B.G5 or second antibody alone demonstrates background binding to pUC19 is present with some of the mAbs.
no
binding. Minimal
the transmembrane and cytoplasmic domains (9,10,13,26,29) showed that sCD4 possessed anti-HIV properties. Linking the first two external domains of sCD4 to immunoglobulin Fc domains increased the serum half-life of sCD4. These "immunoadhesions" also showed anti-HIV properties and were able to bind to cellular Fc receptors and cross primate placentas (3,4). A retroviral vector has been used to transduce the gene segment for sCD4 into cells thus enabling these cells to produce their own copies of sCD4 (20). A source for large amounts of sCD4 will be necessary if the therapeutic promise shown by sCD4 is achieved. The production of sCD4 by bacteria would enable researchers and clinicians to obtain these quantities of sCD4 at a comparatively inexpensive cost. Conversely, the production of bacterial cell lines that express sCD4 would allow many laboratories to conduct research into the protective activity of sCD4. Toward this end, experiments have been done in which gene segments for CD4 has been used to transform E. coli. Chao et al. (5) produced a single domain sCD4 molecule in E. coli that was isolated from inclusion bodies. Garlick et al. (12) produced a sCD4 molecule consisting of the 183 N-terminal amino acids that was also
B 1
'I
'
Binding of 25I-labeled gp 120 to pDABL/CD4 expressing bacteria. E. coli were transformed with pUC 19 (A) or pDABu/CD4 (B), and spread on LB/amp plates. Following incubation overnight, colonies were lifted onto filters, induced with IPTG, lysed, and probed with l25I-labeled gpl20. Following binding, the filters were washed and autoradiographed. A 4-hr exposure is shown. FIG. 6.
169
McCALLUS ET AL. isolated from inclusion bodies. Hybrid proteins consisting of the N-terminal 177 amino acids of CD4 linked the E. coli maltose-binding protein were produced by Szmelcman et al. (28), which were purified from the periplasm by binding to maltose. The CD4 gene segment used in the experiments reported here was obtained by PCR amplification of SupTl cells using primers designed to include the VI loop. The amplified DNA was then ligated into a bacterial vector that included a ribosome binding site and signal sequence. This signal sequence was added to aid in the secretion of the expressed protein into the periplasmic space of the bacteria. Bacterial fractionation of E. coli transformed with pDABL/CD4 revealed the reactivity of anti-CD4 mAbs to be with the cytoplasmic/ membrane bound fraction (data not shown). Therefore, the bacterially expressed CD4 protein was either anchored in the cytoplasmic membrane or was still present in the cytoplasm itself. However the ability of the anti-CD4 mAbs and l25I-labeled gpl 20 to bind to these bacteria indicates that the expressed CD4 protein was folded correctly to form the correct antigenic determinants. The signal sequence-directed translocation of proteins through the cytoplasmic membrane of gramnegative bacteria can be problematic. Simply attaching a signal sequence to a protein does not guarantee transport through the membrane. Charged amino acid residues and the three-dimensional conformation of the mature protein also influence whether proteins will enter into the periplasm (2,16). Furthermore, the function of the signal sequence is to direct the transport of proteins through the cytoplasmic membrane and has no direct effect on the secretion of proteins through the outer membrane to the outside of the bacteria. A series of periplasmic and outer membrane proteins are responsible for the final secretion of the protein to the exterior of the cell (22,25). The ability of pDABL and similar vectors to direct the secretion of inserted proteins to the periplasm and/or the surrounding medium will thus be influenced by the protein itself. The expression of CD4 in bacteria, which we have shown by binding of anti-CD4 mAbs and reactivity with 125I-labeled gpl20, is a further demonstration of the ability to produce this molecule by prokaryotic organisms. Use of the pDABL vector, with possible refinements, will allow the production of other important proteins such as immunoglobulin variable regions. This type of genetic manipulation make it easier to obtain proteins important in the study of receptor-ligand interactions and protein research in general. Furthermore, bacterially produced CD4 may be a useful research tool for analysis of HIV-receptor binding as well as of interactions with class II MHC molecules. to
ACKNOWLEDGMENTS The authors wish to thank Daniel Eldridge for technical assistance and M. Beth for her helpful comments. This work was supported by grants from NIH, the Lupus Foundation, and the Scleroderma Federation to W. V.W.; grants from the American Foundation for AIDS Research, the Council for Tobacco Research, and NIH to D.B.W.; grants from the American Foundation for AIDS Research and from NIH to T.K.E.; and grants from the Pennsylvania Lupus Foundation and the American Foundation for AIDS Research to K.E.U. K.E.U. is an American Foundation for AIDS Research/Pediatric AIDS Foundation Scholar. We also acknowledge the contribution of reagents by the AIDS Reference Reagent Program.
REFERENCES 1. Ausubel, F., R. Brent, R. molecular biology. Greene 2. 3.
Kingston, D. Moore, J. Seidman, J. Smith, and K. Struhl. 1989. Current protocols in Publishing Associates and Wiley-Interscience, John Wiley, New York. Beckwith. 1990. The role of charged amino acids in the localization of secreted and membrane
Boyd, D., and J. proteins. Cell 62:1031-1033. Byrn, R.A., J. Mordenti, C. Lucas, D. Smith. S.A. Marsters, J.S. Johnson, P. Cossum, S.M. Chamow, F.M. Wurm, T. Gregory, J.E. Groopman, and D.J. Capon. 1990. Biological properties of a CD4 immunoadhesin. Nature
(London) 344:667-670. 4. Capon, D.J., S.M. Chamow, J. Mordenti, S.A. Marsters, T. Gregor, H.J. Mitsuya, R.A. Byrn, C. Lucas, F.M. 170
BIOACTIVE CD4 IN BACTERIA
Wurm, J.E. Groopman, S. Broder, and D.H. Smith. 1989. Designing CD4 immunoadhesins for AIDS therapy. Nature (London) 337:525-531. 5. Chao, B.H., D.S. Costopoulos, T. Curiel, J.M. Bertonis, P. Chisholm. C. Williams, R.T. Schooley, J.J. Rosa, R.A. Fisher, and J.M. Maraganore. 1989. A 113-amino acid fragment of CD4 produced in Eschericha coli blocks human immunodeficiency virus-induced cell fusion. J. Biol. Chem. 264:5812-5817.
6. Clark, S.J., W.A. Jefferies, A.N. Barclay, J. Gagnon, and A.F. Williams. 1987. Peptide and nucleotide sequences of rat CD4 (W3/25) antigen: Evidence for derivation from a structure with four immunoglobulin-related domains. Proc. Nati. Acad. Sei. U.S.A. 84:1649-1653. 7.
Clayton, L.K., R.E. Hussey, R. Steinbrich, H. Ramachandran, Y. Husain, andE.L. Reinherz. 1988. Substitution of murine for human CD4 residues identifies amino acids critical for HIV-gpl20 binding. Nature (London) 335: 363-366.
8.
9.
Dalgleish, A.G., P.C.L. Beverley, P.R. Clapham, D.H. Crawford, M.F. Greaves, and R.A. Weiss. 1984. TheCD4 (T4) antigen is an essential component of the receptor for the AIDS retrovirus. Nature (London) 312:763-767. Deen, K.C., J.S. McDougal, R. Inacker, G. Folena-Wasserman, J. Arthros. J. Rosenberg, P.J. Maddon, R. Axel, and R.W. Sweet. 1988. A soluble form of CD4 82-84.
(T4) protein inhibits AIDS virus infection. Nature (London) 331:
10. Fisher, R.A., J.M. Bertonis, W. Meier, VA. Johnson, D.S. Costopoulos, T. Liu, R. Tizard, B.D. Walker, M.S. Hirsch, R.T. Schooley, and R.A. Flavell. 1988. HIV infection is blocked in vitro by recombinant soluble CD4. Nature (London) 331:76-78. 11.
12.
Fleury,S.,D. Lamarre, S. Melodie, S.-E. Ryu, C. Cantin, W.A. Hendrickson, and R.-P. Sekaly. 1991. Mutational analysis of the interaction between CD4 and Class II MHC: Class II antigens contact CD4 on a surface opposite the gpl20-binding site. Cell 66:1037-1049. Garlick, R.L., R.J. Kirschner, F.M. Eckenrode, W.G. Tarpley, and C.-C. Tomich. 1990. Escherichia coli expression, purification, and biological activity of a truncated soluble CD4. AIDS Res. Hum. Retroviruses 6: 465-479.
13.
Hussey, R.E., N.E. Richardson, M. Kowalski, N.R. Brown, H.-S. Chang, R.F. Siliciano, T. Dorfman. B. Walker, J. Sodroski, and E.L. Reinhrez. 1988. A soluble CD4 protein selectively inhibits HIV replication and syncytium formation. Nature (London) 331:78-81.
14.
Klatzmann, D., E. Champagne, S. Chamaret, J. Gruest, D. Guetard, T. Hercend, J.-C. Gluckman, and L. Montagnier. 1984. T lymphocyte T4 molecule behaves as the receptor for human retrovirus L. Nature (London) 312:767-768.
15. Lei, S.-P., H.-C. Lin, S.-S. Wang, J. Callaway, and G. Wilcox. 1987. Characterization of the Erwinia carotovara pelB gene and its product pectate lyase. J. Bacteriol. 169:4379-4383.
16. Li, P., J. Beckwith, and H. Inouye. 1988. Alteration of the amino terminus of the mature sequence of a periplasmic protein can severly affect export in Escherichia coli. Proc. Nati. Acad. Sei. U.S.A. 85:7685-7689. 17.
Lifson, J.D., G.R. Reyes, M.S. McGrath, B.S. Stein, and E.G. Engleman. 1986. AIDS retrovirus induced
cytopathology:
Giant cell formation is mediated
by
cell-cell fusion
1123-1127.
involving
the CD4 molecule. Science 232:
18. Maddon, P., A.G. Dalgleish, J.S. McDougal, P.R. Clapham, R.A. Weiss, and R. Axel. 1986. The T4 gene encodes the AIDS virus receptor and is expressed in the immune system and the brain. Cell 47:333-348.
19. Maddon, P.J., D.R. Littman, M. Godfrey, D.E. Maddon, L. Chess, and R. Axel. 1985. The isolation and nucleotide sequence of a cDNA encoding the T cell surface protein T4: A new member of the immunoglobulin gene family. Cell 42:93-104. 20.
Morgan, R.A., D.J. Looney, D.D. Muenchau, F. Wong-Staal, R.C. Gallo, and W.F. Anderson. 1990. Retroviral vectors expressing soluble CD4: A potential gene therapy for AIDS. AIDS Res. Hum. Retroviruses 6:183-191.
21. Peterson, A., and B. Seed. 1988. Genetic lymphocyte antigen CD4. Cell 54:65-72.
analysis of monoclonal antibody 171
and HIV
binding
sites
on
the human
McCALLUS ET AL. 22.
Pugsley, A.P., C, d'Enfert, I. Reyss, and M.G. negative bacteria. Annu. Rev. Genet. 24:67-90.
Kornacker. 1990. Genetics of extracellular secretion
23. Sambrook, J., E. Fritch, and T. Maniatis. 1989. Molecular Laboratory Press, Cold Spring Harbor, NY. 24. Sattentau,
cloning.
A
laboratory
manual. Cold
by
Spring
gramHarbor
Q.J., and R.A. Weiss. 1988. The CD4 antigen: Physiological ligand and HIV receptor. Cell 42:631-633.
25. Schatz, P.J., and J. Beckwith. 1990. Genetic 24:215-248.
analysis
of
protein export
in Escherichia coli. Annu. Rev. Genet.
26. Smith, D.H., S.A. Marsteers, T. Gregory, J.E. Groopman, andD.J. Capon. 1988. a soluble secreted form of the CD4 antigen. Science 238:1704-1707.
Blocking of HIV-1 infectivity by
27. Sodroski, J., W. Goh, C. Rosen, K. Campbell, and W. Haseltine. 1986. Role of the HTLV-III/LAV syncytium formation and cytopathicity. Nature (London) 322:470-474.
envelope
in
28. Szmelcman, S., J.-M. Clement, M. Jehanno, O. Schwartz, L. Montagnier, and M. Hofnung. 1990. Export and one-step purification from Escherichia coli of a malE-CD4 hybrid protein that neutralizes HIV in vitro. J. Acq. Immune Def. Synd. 3:859-872. 29. Traunecker, A., W. Luke, and K. Kajalainen. 1988. Soluble CD4 molecules neutralise human virus type 1. Nature (London) 331:84-86. 30. Weiner, D.B., K. Huebner, W.V. Williams, and M.I. Greene. 1991. Non-CD4 HIV-1 infection of human cells. Pathobiology 4:1-20.
immunodeficiency
encoding human genes
facilitate
31. Williams, W.V., H.R. Guy, D.H. Rubin, M.F. Robey, J.N. Myers, T. Kieber-Emmons, D.B. Weiner, and M.I. Greene. 1988. Sequences of the cell-attachment sites of reovirus type 3 and its anti-idiotypic/antireceptor antibody: Modeling of their three-dimensional structures. Proc. Nati. Acad. Sei. U.S.A. 85:6488-6492.
University
111
Address reprint requests to: David B. Weiner, Ph.D. Division of Rheumatology of Pennsylvania School of Medicine 570 Maloney, 3600 Spruce Street Philadelphia, PA 19104
Construction of a Recombinant Bacterial Human CD4 Expression System Producing a Bioactive CD4 Molecule DANIEL E. MCCALLUS,' KENNETH E. UGEN,1 ALICE I. SATO,1'2 2 WILLIAM V. WILLIAMS,2 3 and DAVID B. WEINER1
ABSTRACT
protein expressed on helper T lymphocytes is a restriction element for major histocompatibility class II immune responses. This molecule is also used by the human immunodeficiency virus as its specific cellular receptor facilitating binding of virus to cells. As soluble forms of CD4 inhibit HIV infection in tissue culture, attention has focused on this molecule. Bacterially produced CD4 would facilitate studies of the biology of the CD4 molecule. However, bacterially expressed CD4 must be refolded for assumption of its interaction with conformâtionally dependant anti-C 1)4 monoclonal antibodies as well as the HIV-1 envelope protein gpl20. We report here the engineering of an external domain construct of the CD4 gene into a novel expression vector containing the nucleotide sequence encoding the pelB leader peptide of Envinia carotovara (pDAB, ), to facilitate correct folding of CD4 in bacteria. Monoclonal antibodies specific for important conformational epitopes of the CD4 molecule were able to bind bacterial colonies containing the pDABL/CD4 vector but not colonies with vector alone. Importantly, recombinant gpl20 produced in baculovirus bound specifically to bacterial colonies expressing the CD4 recombinant molecule. This system presents a simple screening mechanism for molecules that bind to the external domain of the CD4 glycoprotein. Vectors such as pDABL will also facilitate the production of large amounts of biologically active proteins in bacteria.
The CD4
the surface of certain human T lymphocytes, is a 55,000 molecular for interaction with class II major histocompatibility (MHC) molecules of antigen-presenting cells (24). The CD4 molecule has also been shown to be the cellular receptor for the human immunodeficiency virus (HIV), which binds CD4 through the HIV envelope glycoprotein gpl20 (8,14,18). Interaction of CD4+ permissive cells with cells displaying the envelope glycoproteins of HIV (e.g., HIV-1-infected cells) results in the formation of multinucleated giant cells (syncytia). This is initiated
molecule, expressed Theweight glycoprotein responsible CD4
on
vr
'iv1
'The Wistar Institute of Anatomy and Biology, the department of Medicine, University of Pennsylvania School of Medicine, and ^Children's Hospital of Philadelphia, Philadelphia, Pennsylvania. 163
McCALLUS ET AL.
by the binding of surface gpl20 on infected cells with the CD4 molecule expressed on uninfected cells (17). The region of gp 120 that binds CD4 is found in the carboxy-terminal end of the molecule and conserved amino acid sequences have been implicated in this binding (27). Site directed mutagenesis defines three discontinuous sites as required for binding to CD4. The CD4 molecule is composed of extracellular, transmembrane and intracellular segments. The extracellular segment is 370 amino acids in length and can be separated into four domains with homology to immunoglobulin variable domains (6,19). It has been found that the most membrane distal domain is sufficient for binding to gpl20 (7,21). However, constructs containing the first two domains appear to mediate more efficient binding to gpl20 in some systems. Attempts have been made to use soluble forms of the CD4 molecule (sCD4), lacking the transmembrane and cytoplasmic domains, to inhibit HIV infection and/or syncytia formation (9,10,13,26,29). Both full length sCD4, containing the four external domains, as well as sCD4 containing only the first one or two amino-terminal domains have been shown to bind gpl20 and block both HIV infection and syncytium formation. Several studies have also demonstrated that recombinant sCD4 produced by bacteria are able to bind HIV (5,12,28). However, bacterial expression systems are hampered by the fact that CD4, like many mammalian transmembrane proteins, is incorrectly processed by the bacterial cell such that the CD4 recombinant protein must be denatured and refolded to recover biological activity. Production of a more natural folding CD4-like molecule would present a number of advantages for the analysis of CD4 biology as well as investigations of CD4 reactivity with HIV envelope proteins. We have used the polymerase chain reaction (PCR) to amplify CD4 from human T lymphocyte lines and to insert the amplified molecule into a bacterial expression vector. This vector contained the pelB leader peptide of Erwinia carotovora. Expression of the CD4 gene segment in a biologically relevant form in Escherichia coli was achieved, as demonstrated by binding of conformationally dependent anti-CD4 monoclonal antibodies (mAb) as well as reactivity with recombinant gpl20 molecules. MATERIALS AND METHODS Bacterial strains. Escherichia coli DH5-alpha competent cells (BRL, Gaithersburg, MD) were used for transformation. Bacteria were grown in Luria broth containing 100 mg/ml ampicillin (LB/amp) (1,23). Enzymes and oligonucleotides. Restriction endonucleases and T4 DNA ligase were purchased from BRL. Enzyme reaction conditions were according to that of the supplier. Oligonucleotides for PCR primers and for Southern blotting were synthesized by the DNA Synthesis Facility of the Wistar Institute. Antibodies. Anti-CD4 mAbs SIM.2 (murine IgG2b) and SIM.4 (murine IgG,) were from the National Institute of Allergy and Infectious Diseases AIDS Research and Reference Reagent Program (Rockville, MD). Leu3a (murine IgG,) was from Becton-Dickinson (MountainView, CA). The Leu3a and SIM.4 are specific for the HIV-1 gpl20 binding epitope on CD4, and both block HIV-induced syncytium formation. SIM.2 binds a distinct epitope, but also blocks HIV-induced syncytium formation. Antireovirus type 3 mAb 9B.G5 (murine IgG2a) was the kind gift of Mark I. Greene (Department of Pathology and Laboratory Medicine, University of Pennsylvania, Philadelphia, PA). Plasmid construction. The DNA coding for the pectate lyase (pelB) signal peptide of Erwinia carotovara was synthesized according to standard methodology according to the published sequence (15) and made double-stranded by PCR. Restriction sites for Hindlll and EcoRl were incorporated into the primers for insertion into pUC 19. Positive transformants were selected and plasmid preps were made. ThepelB insert was excised by digestion with Hindlll and EcoRl and identified by agarose gel electrophoresis. CD4 amplification and cloning. RNA from SupTl cells (30) was reverse transcribed into cDNA using 200 U of MMLV reverse transcriptase (BRL) in a 20 u.1 reaction. The first domain of CD4 was amplified by PCR using primers corresponding to codons ( )4 to ( + )5 (primer A: ccagtcgacactcagggaaacaaagtg) and codons 116 to 125 (primer B: accgaattcgctctacaaggtcagggtcag) in a Programmable Thermal Controller (MJ Research; Cambridge, MA). Primer A contained a Sail restriction digestion site (underlined) while primer B contained an EcoRl restriction site (underlined) and also an added stop codon. The amplification program was 94°C for 60 sec, 52°C for 90 sec, and72°C for 120 sec. Following 30 cycles, the temperature was held at 72°C for 5 min. Positive amplification was determined by agarose gel electrophoresis and ethidium bromide —
164
BIOACTIVE CD4 IN BACTERIA CD4 Insert
Diagram of pDABL vector with CD4 insert. The leader sequence of thepelB gene of Erwinia caratovara was amplified by PCR using primers containing the the polylinker site. The amplified DNA was digested with Hindlll and EcoRI and ligated into pUC19 which had been digested with the same endonucleases and treated with calf intestinal alkaline phosphatase. The amplified CD4 gene was inserted into the polylinker using the Sail and EcoRI restriction sites. FIG. 1.
staining. The CD4 DNA and plasmid DNA was cut with the appropriate endonucleases and plasmid DNA was treated with calf intestinal phosphatase (Boehringer Mannheim, Indianapolis, IN). The CD4 DNA was ligated into pDABL using 1 U of T4 DNA ligase overnight at 16°C. Ligation mixtures were transformed into E. coli DH5-alpha competent cells as described by the manufacturer. Southern hybridization, restriction digestion and amplification. Competent E. coli transformed with the amplified CD4 gene segment were plated on LB/amp plates as described above. An internal oligonucleotide probe for the CD4 gene segment was used which corresponded to the antisense sequence of amino acids 20-25 (ctcttcttctggtaatc). The probe (100 ng) was labeled with 32P for 30 min using T4 polynucleotide kinase (1,23). Nitrocellulose filters (0.45 pun; Schleicher & Schuell, Keene, NH) were used to lift the transformed bacteria. Following alkaline lysis of the bacteria the filters were incubated with the labeled probe for 2 hr at 55°C. The filters were then washed two times at room temperature with 2x SSPE(1,23) and 0.1% SDS. The nitrocellulose filters were exposed to film for 48 hr. A colony that hybridized with the probe was selected for further study and was grown overnight in LB/amp. Plasmid DNA from this culture was prepared and was subjected to restriction digestion to liberate the CD4 gene segment with or without the pelB signal sequence gene. Plasmid DNA from the transformant was also amplified by PCR using the CD4 primers and primers for the 5' end of the pelB signal sequence and the 3' end of the CD4 gene segment. Protein expression. A bacterial clone possessing the CD4 gene segment inserted into pDABL was plated onto one half of LB/amp plates. The other side of the plates contained E. coli containing pUC 19. Following overnight growth, a 0.45-u.m nitrocellulose filter was placed on the bacterial plate. Filters were lifted to other LB/amp plates on which 100 p.1 of isopropyl-ß-thiogalactopyranoside (IPTG; 25 mg/ml; Stratagene, La Jolla, CA) had been spread and were then incubated for 4 hr at 37°C. Filters were then exposed to chlorofoam vapor to fix the bacteria for 15 min and incubated overnight (with shaking) in lysis buffer ( 100 mM Tris-Cl, pH 7.8, 150mMNaCl,5 mMMgCl2, 1.5% bovine serum albumin (BSA), 1 p-g/ml pancreatic DNAse I, and 40 p-g/ml lysozyme]. Filters were then blocked for 4 hr with 4% nonfat dry milk and 1% BSA in Tris-buffered saline (TBS; 20 mM Tris, 500 mM NaCl, pH7.5). Filters were incubated for 2 hr at room temperature with the 165
McCALLUS ET AL.
Amplification of CD4 by PCR. The CD4 gene was amplified by PCR from SupTl cell cDNA (lane 2), but was not amplified from HSB cell cDNA (lane 3) or Leu3a cell cDNA (lane 4). Primers for murine immunoglobulin variable region genes did not amplify SupTl cell cDNA (lane 5) but did amplify Leu3acDNA (lane 6). Size markers were phage k/Hindlll digest (lane 1) and X174///aeIII digest (lane 7). FIG. 2.
appropriate monoclonal antibody (mAb) diluted to 7.4 p.g/p.1 in TBS containing 1 % BSA and 0.1 % Tween-20 (TBS/BSA). Filters were washed extensively with TBS/BSA, then incubated for 2 hr at room temperature with goat anti-mouse immunoglobulin (horseradish peroxidase-labeled) (Fisher Biotech, Pittsburgh, PA) diluted to 0.8 p.g/ml in TBS/BSA. Filters were washed as above and the substrate, TMBlue (TSI, Worcester,
MA), was added for 5-10 min at room temperature. After development, filters were rinsed in deionized water and photographed. For binding of l25I-labeled gpl20, blocked filters were incubated for 2 hr at room the temperature with 125I-labeled gpl20 (MicroGeneSys, CT), at 500,000-1,000,000 cpm/ml, labelled by washed Filters Tween-20 were 0.1% 1% in and BSA (TBS/BSA). chloramine T method (31) TBS containing extensively with TBS/BSA and autoradiographed (Kodak XRP film) for 2-24 hr. RESULTS Plasmid construction. The signal sequence for the pelB protein of Erwinia carotovara was inserted into pUC 19 as shown in Fig. 1. A poly linker containing restriction sites for Xbal, Sail, and EcoRl was added to the end of this signal sequence to facilitate cloning of the desired gene(s). CD4 amplification and cloning. CD4 was amplified by PCR from Supt 1 cells but was not amplified from CD4~ HSB cells (Fig. 2). In addition, primers for immunoglobulin variable region genes did not amplify the CD4 gene segment. Amplified CD4 DNA was inserted into pDABL. The presence of this DNA was verified by Southern hybridization (Fig. 3) and restriction digestion of the transformant and by amplification of the plasmid mini-prep using the original CD4 primers as described below (Fig. 4). 166
BIOACTIVE CD4 IN BACTERIA
•
>•
*
I*» i
Southern hybridization to pDABL/CD4 transformants. Nitrocellulose lifts were made from LB/amp plates bacteria with pDABL/CD4 (A) or bacteria with the pDABL vector alone (B). The lifts were probed with 32P-labeled oligonucleotide specific for an internal sequence of the V1 loop of CD4. FIG. 3.
containing
Southern hybridization, restriction digestion and amplification. Following exposure to film for 48 hr, was observed that transformants that received the CD4 gene segment hybridized with the internal oligonucleotide probe, while transformants receiving plasmid alone did not react with this probe (Fig. 3). Positive colonies was grown and plasmid DNA prepared. The DNA was digested with Sail and EcoRI to liberate the CD4 insert or with Hindlll and £coRI to liberate the pelB signal sequence with the CD4 gene segment (see Fig. 1 for location of restriction sites). Amplification of this DNA with the CD4 primers yielded a 330-bp band identical in size to the liberated CD4 gene segment, while amplification of the DNA with the 5' pelB primer and the 3' CD4 primer yielded a 480-bp band identical in size to the liberated pelB/CD4 gene it
segment (Fig. 4).
Protein expression. Nitrocellulose lifts of colonies of E. coli containing pDABL/CD4 were probed with several anti-CD4 mAbs and an irrelevant mAb 9BG5 [directed against reovirus type 3 (31)]. The nitrocellulose lifts were prepared, cut into strips, and reacted with the anti-CD4 mAbs (Fig. 5). Positive reactions with pDABL/CD4 were observed with anti-CD4 mAbs Leu3a, SIM2, and SIM4. This was observed for E. coli either lysed or intact. Minimal background binding was observed with pUC19, while strong binding was observed with pDABL/CD4 expressing cells. The mAB 9BG5, anti-mouse-conjugated secondary antibody, or the TMB substrate alone showed no reactivity with either pUC19 or pDABL/CD4. Next, the ability of CD4 expressed in bacteria to react with gpl20 was directly assayed. E. coli was transformed with pUC19 or pDABL/CD4, bacterial colonies grown on LB/amp plates, and filters were lifted and induced with IPTG. l25I-labeled gpl20 was incubated with the filters, and autoradiography performed. No significant binding of l25I-labeled gpl20 was seen to the pUC19-transformed negative control filters (Fig. 6A). In marked contrast, extensive binding of l25I-labeled gpl20 to pDABL/CD4-transformed E. coli was observed (Fig. 6B). These results support and extend the results obtained using anti-CD4 antibodies. It is likely that CD4 expressed on the surface of the pDABL/CD4-transformed E. coli is expressed in a conformationally relevant form. 167
McCALLUS ET AL.
FIG. 4. Restriction digestion and amplification of pDAB[ /CD4 plasmid. The pDABL/CD4 construct was restricted with various enzymes and the restriction fragments run on a 2% agarose gel. Lane I: X174///aeIII MW markers; Lane 2: PCR amplification product of pDABL/CD4 plasmid with CD4 primers (see Materials and Methods); Lane 3: Sail and EcoRl digest of pDABL /CD4 plasmid to liberate the CD4 gene; Lane 4: PCR amplification product of pDABL/CD4 plasmid with the 5' primer for pelB and the 3' primer for CD4; Hindlll and EcoRl digest of pDABL/CD4 plasmid to liberate the CD4 gene; phase \/HindIII MW markers.
DISCUSSION The use of sCD4 as a therapeutic modality for the treatment of AIDS has received a great deal of study during recent years. Although sCD4 has been shown to inhibit the ability of HIV to infect in vitro and has been shown to block syncytium formation, there have been as yet no reports that have demonstrated clinical benefits from treatment with sCD4. The possibility that HIV uses other cellular receptors to gain entrance to permissive cells may abrogate the effectiveness of sCD4 as a treatment for AIDS. However, promising data from the in vitro experiments provide impetus for further studies into the protective effects of sCD4. CD4 has
also been observed to function as a restriction element on T cells for their interaction with MHC Class II bearing antigen-presenting cells. Interaction of human CD4 with class II has been demonstrated, but the exact residues involved in binding remain controversial (reviewed in ref. 11). Further investigation of ligandreceptor interactions between CD4 and MHC class II will enhance our understanding of antigen responsiveness and T-cell activation in a general sense. The forms in which sCD4 has been produced for in vitro studies have varied. Therefore, caution should be taken when comparing the different studies involving sCD4. Early studies using recombinant sCD4 lacking 168
BIOACTIVE CD4 IN BACTERIA
pUC
19
pDAB/CD4
ImJà Sin* 2 Sim 4
anti-mQp$e CQnjyjgate •
«.
substrate alone FIG. 5. Reactivity of bacteria containing pDABL/CD4 with anti-CD4 mAbs. Colonies of bacteria transformed with pUC19 (left side of strips) or pDABL/CD4 (right side of strips) were lifted from LB/amp plates with nitrocellulose filters. The filters were cut into strips and reacted with the mAbs shown. Specific reactivity with the anti-CD4 mAbs Leu3a,
SIM2, and SIM4 is evident, while control mAb 9B.G5 or second antibody alone demonstrates background binding to pUC19 is present with some of the mAbs.
no
binding. Minimal
the transmembrane and cytoplasmic domains (9,10,13,26,29) showed that sCD4 possessed anti-HIV properties. Linking the first two external domains of sCD4 to immunoglobulin Fc domains increased the serum half-life of sCD4. These "immunoadhesions" also showed anti-HIV properties and were able to bind to cellular Fc receptors and cross primate placentas (3,4). A retroviral vector has been used to transduce the gene segment for sCD4 into cells thus enabling these cells to produce their own copies of sCD4 (20). A source for large amounts of sCD4 will be necessary if the therapeutic promise shown by sCD4 is achieved. The production of sCD4 by bacteria would enable researchers and clinicians to obtain these quantities of sCD4 at a comparatively inexpensive cost. Conversely, the production of bacterial cell lines that express sCD4 would allow many laboratories to conduct research into the protective activity of sCD4. Toward this end, experiments have been done in which gene segments for CD4 has been used to transform E. coli. Chao et al. (5) produced a single domain sCD4 molecule in E. coli that was isolated from inclusion bodies. Garlick et al. (12) produced a sCD4 molecule consisting of the 183 N-terminal amino acids that was also
B 1
'I
'
Binding of 25I-labeled gp 120 to pDABL/CD4 expressing bacteria. E. coli were transformed with pUC 19 (A) or pDABu/CD4 (B), and spread on LB/amp plates. Following incubation overnight, colonies were lifted onto filters, induced with IPTG, lysed, and probed with l25I-labeled gpl20. Following binding, the filters were washed and autoradiographed. A 4-hr exposure is shown. FIG. 6.
169
McCALLUS ET AL. isolated from inclusion bodies. Hybrid proteins consisting of the N-terminal 177 amino acids of CD4 linked the E. coli maltose-binding protein were produced by Szmelcman et al. (28), which were purified from the periplasm by binding to maltose. The CD4 gene segment used in the experiments reported here was obtained by PCR amplification of SupTl cells using primers designed to include the VI loop. The amplified DNA was then ligated into a bacterial vector that included a ribosome binding site and signal sequence. This signal sequence was added to aid in the secretion of the expressed protein into the periplasmic space of the bacteria. Bacterial fractionation of E. coli transformed with pDABL/CD4 revealed the reactivity of anti-CD4 mAbs to be with the cytoplasmic/ membrane bound fraction (data not shown). Therefore, the bacterially expressed CD4 protein was either anchored in the cytoplasmic membrane or was still present in the cytoplasm itself. However the ability of the anti-CD4 mAbs and l25I-labeled gpl 20 to bind to these bacteria indicates that the expressed CD4 protein was folded correctly to form the correct antigenic determinants. The signal sequence-directed translocation of proteins through the cytoplasmic membrane of gramnegative bacteria can be problematic. Simply attaching a signal sequence to a protein does not guarantee transport through the membrane. Charged amino acid residues and the three-dimensional conformation of the mature protein also influence whether proteins will enter into the periplasm (2,16). Furthermore, the function of the signal sequence is to direct the transport of proteins through the cytoplasmic membrane and has no direct effect on the secretion of proteins through the outer membrane to the outside of the bacteria. A series of periplasmic and outer membrane proteins are responsible for the final secretion of the protein to the exterior of the cell (22,25). The ability of pDABL and similar vectors to direct the secretion of inserted proteins to the periplasm and/or the surrounding medium will thus be influenced by the protein itself. The expression of CD4 in bacteria, which we have shown by binding of anti-CD4 mAbs and reactivity with 125I-labeled gpl20, is a further demonstration of the ability to produce this molecule by prokaryotic organisms. Use of the pDABL vector, with possible refinements, will allow the production of other important proteins such as immunoglobulin variable regions. This type of genetic manipulation make it easier to obtain proteins important in the study of receptor-ligand interactions and protein research in general. Furthermore, bacterially produced CD4 may be a useful research tool for analysis of HIV-receptor binding as well as of interactions with class II MHC molecules. to
ACKNOWLEDGMENTS The authors wish to thank Daniel Eldridge for technical assistance and M. Beth for her helpful comments. This work was supported by grants from NIH, the Lupus Foundation, and the Scleroderma Federation to W. V.W.; grants from the American Foundation for AIDS Research, the Council for Tobacco Research, and NIH to D.B.W.; grants from the American Foundation for AIDS Research and from NIH to T.K.E.; and grants from the Pennsylvania Lupus Foundation and the American Foundation for AIDS Research to K.E.U. K.E.U. is an American Foundation for AIDS Research/Pediatric AIDS Foundation Scholar. We also acknowledge the contribution of reagents by the AIDS Reference Reagent Program.
REFERENCES 1. Ausubel, F., R. Brent, R. molecular biology. Greene 2. 3.
Kingston, D. Moore, J. Seidman, J. Smith, and K. Struhl. 1989. Current protocols in Publishing Associates and Wiley-Interscience, John Wiley, New York. Beckwith. 1990. The role of charged amino acids in the localization of secreted and membrane
Boyd, D., and J. proteins. Cell 62:1031-1033. Byrn, R.A., J. Mordenti, C. Lucas, D. Smith. S.A. Marsters, J.S. Johnson, P. Cossum, S.M. Chamow, F.M. Wurm, T. Gregory, J.E. Groopman, and D.J. Capon. 1990. Biological properties of a CD4 immunoadhesin. Nature
(London) 344:667-670. 4. Capon, D.J., S.M. Chamow, J. Mordenti, S.A. Marsters, T. Gregor, H.J. Mitsuya, R.A. Byrn, C. Lucas, F.M. 170
BIOACTIVE CD4 IN BACTERIA
Wurm, J.E. Groopman, S. Broder, and D.H. Smith. 1989. Designing CD4 immunoadhesins for AIDS therapy. Nature (London) 337:525-531. 5. Chao, B.H., D.S. Costopoulos, T. Curiel, J.M. Bertonis, P. Chisholm. C. Williams, R.T. Schooley, J.J. Rosa, R.A. Fisher, and J.M. Maraganore. 1989. A 113-amino acid fragment of CD4 produced in Eschericha coli blocks human immunodeficiency virus-induced cell fusion. J. Biol. Chem. 264:5812-5817.
6. Clark, S.J., W.A. Jefferies, A.N. Barclay, J. Gagnon, and A.F. Williams. 1987. Peptide and nucleotide sequences of rat CD4 (W3/25) antigen: Evidence for derivation from a structure with four immunoglobulin-related domains. Proc. Nati. Acad. Sei. U.S.A. 84:1649-1653. 7.
Clayton, L.K., R.E. Hussey, R. Steinbrich, H. Ramachandran, Y. Husain, andE.L. Reinherz. 1988. Substitution of murine for human CD4 residues identifies amino acids critical for HIV-gpl20 binding. Nature (London) 335: 363-366.
8.
9.
Dalgleish, A.G., P.C.L. Beverley, P.R. Clapham, D.H. Crawford, M.F. Greaves, and R.A. Weiss. 1984. TheCD4 (T4) antigen is an essential component of the receptor for the AIDS retrovirus. Nature (London) 312:763-767. Deen, K.C., J.S. McDougal, R. Inacker, G. Folena-Wasserman, J. Arthros. J. Rosenberg, P.J. Maddon, R. Axel, and R.W. Sweet. 1988. A soluble form of CD4 82-84.
(T4) protein inhibits AIDS virus infection. Nature (London) 331:
10. Fisher, R.A., J.M. Bertonis, W. Meier, VA. Johnson, D.S. Costopoulos, T. Liu, R. Tizard, B.D. Walker, M.S. Hirsch, R.T. Schooley, and R.A. Flavell. 1988. HIV infection is blocked in vitro by recombinant soluble CD4. Nature (London) 331:76-78. 11.
12.
Fleury,S.,D. Lamarre, S. Melodie, S.-E. Ryu, C. Cantin, W.A. Hendrickson, and R.-P. Sekaly. 1991. Mutational analysis of the interaction between CD4 and Class II MHC: Class II antigens contact CD4 on a surface opposite the gpl20-binding site. Cell 66:1037-1049. Garlick, R.L., R.J. Kirschner, F.M. Eckenrode, W.G. Tarpley, and C.-C. Tomich. 1990. Escherichia coli expression, purification, and biological activity of a truncated soluble CD4. AIDS Res. Hum. Retroviruses 6: 465-479.
13.
Hussey, R.E., N.E. Richardson, M. Kowalski, N.R. Brown, H.-S. Chang, R.F. Siliciano, T. Dorfman. B. Walker, J. Sodroski, and E.L. Reinhrez. 1988. A soluble CD4 protein selectively inhibits HIV replication and syncytium formation. Nature (London) 331:78-81.
14.
Klatzmann, D., E. Champagne, S. Chamaret, J. Gruest, D. Guetard, T. Hercend, J.-C. Gluckman, and L. Montagnier. 1984. T lymphocyte T4 molecule behaves as the receptor for human retrovirus L. Nature (London) 312:767-768.
15. Lei, S.-P., H.-C. Lin, S.-S. Wang, J. Callaway, and G. Wilcox. 1987. Characterization of the Erwinia carotovara pelB gene and its product pectate lyase. J. Bacteriol. 169:4379-4383.
16. Li, P., J. Beckwith, and H. Inouye. 1988. Alteration of the amino terminus of the mature sequence of a periplasmic protein can severly affect export in Escherichia coli. Proc. Nati. Acad. Sei. U.S.A. 85:7685-7689. 17.
Lifson, J.D., G.R. Reyes, M.S. McGrath, B.S. Stein, and E.G. Engleman. 1986. AIDS retrovirus induced
cytopathology:
Giant cell formation is mediated
by
cell-cell fusion
1123-1127.
involving
the CD4 molecule. Science 232:
18. Maddon, P., A.G. Dalgleish, J.S. McDougal, P.R. Clapham, R.A. Weiss, and R. Axel. 1986. The T4 gene encodes the AIDS virus receptor and is expressed in the immune system and the brain. Cell 47:333-348.
19. Maddon, P.J., D.R. Littman, M. Godfrey, D.E. Maddon, L. Chess, and R. Axel. 1985. The isolation and nucleotide sequence of a cDNA encoding the T cell surface protein T4: A new member of the immunoglobulin gene family. Cell 42:93-104. 20.
Morgan, R.A., D.J. Looney, D.D. Muenchau, F. Wong-Staal, R.C. Gallo, and W.F. Anderson. 1990. Retroviral vectors expressing soluble CD4: A potential gene therapy for AIDS. AIDS Res. Hum. Retroviruses 6:183-191.
21. Peterson, A., and B. Seed. 1988. Genetic lymphocyte antigen CD4. Cell 54:65-72.
analysis of monoclonal antibody 171
and HIV
binding
sites
on
the human
McCALLUS ET AL. 22.
Pugsley, A.P., C, d'Enfert, I. Reyss, and M.G. negative bacteria. Annu. Rev. Genet. 24:67-90.
Kornacker. 1990. Genetics of extracellular secretion
23. Sambrook, J., E. Fritch, and T. Maniatis. 1989. Molecular Laboratory Press, Cold Spring Harbor, NY. 24. Sattentau,
cloning.
A
laboratory
manual. Cold
by
Spring
gramHarbor
Q.J., and R.A. Weiss. 1988. The CD4 antigen: Physiological ligand and HIV receptor. Cell 42:631-633.
25. Schatz, P.J., and J. Beckwith. 1990. Genetic 24:215-248.
analysis
of
protein export
in Escherichia coli. Annu. Rev. Genet.
26. Smith, D.H., S.A. Marsteers, T. Gregory, J.E. Groopman, andD.J. Capon. 1988. a soluble secreted form of the CD4 antigen. Science 238:1704-1707.
Blocking of HIV-1 infectivity by
27. Sodroski, J., W. Goh, C. Rosen, K. Campbell, and W. Haseltine. 1986. Role of the HTLV-III/LAV syncytium formation and cytopathicity. Nature (London) 322:470-474.
envelope
in
28. Szmelcman, S., J.-M. Clement, M. Jehanno, O. Schwartz, L. Montagnier, and M. Hofnung. 1990. Export and one-step purification from Escherichia coli of a malE-CD4 hybrid protein that neutralizes HIV in vitro. J. Acq. Immune Def. Synd. 3:859-872. 29. Traunecker, A., W. Luke, and K. Kajalainen. 1988. Soluble CD4 molecules neutralise human virus type 1. Nature (London) 331:84-86. 30. Weiner, D.B., K. Huebner, W.V. Williams, and M.I. Greene. 1991. Non-CD4 HIV-1 infection of human cells. Pathobiology 4:1-20.
immunodeficiency
encoding human genes
facilitate
31. Williams, W.V., H.R. Guy, D.H. Rubin, M.F. Robey, J.N. Myers, T. Kieber-Emmons, D.B. Weiner, and M.I. Greene. 1988. Sequences of the cell-attachment sites of reovirus type 3 and its anti-idiotypic/antireceptor antibody: Modeling of their three-dimensional structures. Proc. Nati. Acad. Sei. U.S.A. 85:6488-6492.
University
111
Address reprint requests to: David B. Weiner, Ph.D. Division of Rheumatology of Pennsylvania School of Medicine 570 Maloney, 3600 Spruce Street Philadelphia, PA 19104
