Patent Application
20050048471
The present invention involves the utilization of albumin derivatives in the fabrication of therapeutic agents that can be used in the treatment of certain viral diseases and cancers. More precisely, this invention involves hybrid macromolecules characterized by the fact that they carry either the active domain of a receptor for a virus, or the active domain of a molecule which can bind to a virus, or the active domain of a molecule able to recognize the Fc fragment of immunoglobulins bound to a virus, or the active domain of a molecule able to bind a ligand that intervenes in a pathologic process, coupled to albumin or a variant of albumin. In the text that follows, the terms albumin derivatives or albumin variants are meant to designate all proteins with a high plasma half-life obtained by modification (mutation, deletion, and/or addition) via the techniques of genetic engineering of a gene encoding a given isomorph of albumin, as well as all macromolecules with high plasma half-life obtained by the in vitro modification of the protein encoded by such genes. Such albumin derivatives can be used as pharmaceuticals in antiviral treatment due to the high affinity of a virus or of an immunoglobulin bound to a virus for a site of fixation present on the albumin derivative. They can be used as pharmaceuticals in the treatment of certain cancers due to the affinity of a ligand, for example a growth factor, for a site of fixation present on the albumin derivative, especially when such a ligand is associated with a particular membrane receptor whose amplification is correlated with a transforming phenotype (proto-oncogenes). It should be understood in the text that follows that all functionally therapeutic albumin derivatives are designated indifferently by the generic term of hybrid macromolecules with antiviral function, or hybrid macromolecules with anticancer function, or simply hybrid macromolecules. In particular, the present invention consists in the obtention of new therapeutic agents characterized by the coupling, through chemical or genetic engineering techniques, of at least two distinct funtions:
In the present invention, the plasma transporter function, the therapeutic function, and a potential polymerization function, are integrated into the same macromolecule using the techniques of genetic engineering.
One of the goals of the present invention is to obtain hybrid macromolecules derived from HSA which can be useful in the fight against certain viral diseases, such as Acquired Immunodeficiency Syndrome (AIDS). Another goal is to obtain hybrid HSA macromolecular derivatives useful in the treatment of certain cancers, notably those cancers associated with genomic amplification and/or overexpression of human proto-oncogenes, such as the proto-oncogene c-erbB-2 (Semba K. et al., Proc. Natl. Acad. Sci. USA. 82 (1985) 6497-6501; Slamon D. J. et al., Science 235 (1987) 177-182; Kraus M. H. et al., EMBO J. 6 (1987) 605-610).
The HIV-1 virus is one of the retroviruses responsible for Acquired Immunodeficiency Syndrome in man. This virus has been well studied over the past five years; a fundamental discovery concerns the elucidation of the role of the CD4 (T4) molecule as the receptor of the HIV-1 virus (Dalgleish A. G. et al., Nature 312 (1984) 763-767; Klatzmann D. et al., Nature 312 (1984) 767-768). The virus-receptor interaction occurs through the highly specific binding of the viral envelope protein (gp120) to the CD4 molecule (McDougal et al., Science 231 (1986) 382-385). The discovery of this interaction between the HIV-1 virus and certain T lymphocytes was the basis of a patent claiming the utilization of the T4 molecule or its antibodies as therapeutic agents against the HIV-1 virus (French patent application FR 2 570 278).
The cloning and the first version of the sequence of the gene encoding human CD4 has been described by Maddon et al. (Cell 42 (1985) 93-104), and a corrected version by Littmann et al. (Cell 55 (1988) 541): the CD4 molecule is a member of the super-family of immunoglobulins and specifically, it carries a V1 N-terminal domain which is substantially homologous to the immunoglobulin heavy chain variable domain (Maddon P. J. et al., Cell 42 (1985) 93-104). Experiments involving in vitro DNA recombination, using the gene coding for the CD4 molecule, have provided definite proof that the product of the CD4 gene is the principal receptor of the HIV-1 virus (Maddon P. J. et al., Cell 47 (1986) 333-348). The sequence of this gene as well as its utilization as an anti-HIV-1 therapeutic agent are discussed in International patent application WO 88 013 040 A1.
The manipulation of the CD4 gene by the techniques of DNA recombination has provided a series of first generation soluble variants capable of antiviral action in vitro (Smith D. H. et al., Science 238 (1987) 1704-1707; Traunecker A. et al., Nature 331 (1988) 84-86; Fischer R. A. et al., Nature 331 (1988) 76-78; Hussey R. E. et al., Nature 331 (1988) 78-81; Deen K. C. et al., Nature 331 (1988) 82-84), and in vivo (Watanabe M. et al., Nature 337 (1989) 267-270). In all cases, it was observed during various in vivo assays in animals (rabbit, monkey) as well as during phase I clinical trials, that the first generation soluble CD4 variant consisting of the CD4 molecule lacking the two domains in the C-terminal region has a very short half-life: approximately 15 minutes in rabbits (Capon et al., Nature 337 (1989) 525-531), while 50% of first generation soluble CD4 administered intramuscularly to Rhesus monkeys remained bioavailable for 6 hours (Watanabe et al., Nature 337 (1989) 267-270). In addition, Phase I clinical trials conducted on 60 patients presenting AIDS or ARC (“Aids Related Complex”) indicated that the half-life of the Genentech product varied between 60 minutes (intraveinous administration) and 9 hours (intramuscular administration) (AIDS/HIV Experimental Treatment Directory, AmFAR, May 1989). Clearly, a therapeutic agent with such a weak stability in vivo constitutes a major handicap. In effect, repeated injections of the product, which are costly and inconvenient for the patient, or an administration of the product by perfusion, become necessary to attain an efficient concentration in plasma. It is therefore especially important to find derivatives of the CD4 molecule characterized by a much higher in vivo half-life.
The part of the CD4 molecule which interacts with the HIV-1 virus has been localized to the N-terminal region, and in particular to the V1 domain (Berger E. A. et al., Proc. Natl. Acad. Sci. USA 85 (1987) 2357-2361). It has been observed that a significant proportion (about 10%) of HIV-1-infected subjects develop an immune response against the CD4 receptor, with antibodies directed against the C-terminal region of the extra-cellular portion of the receptor (Thiriart C. et al., ADS 2 (1988) 345-352; Chams V. et al., AIDS 2 (1988) 353-361). Therefore, according to a preferred embodiment of the present invention, only the N-terminal domains V1 or V1V2 of the CD4 molecule, which carry all the viral binding activity, will be used in fusion with the stable transporter function derived from albumin.
On the basis of the homology observed with the variable domain of immunoglobulins, several laboratories have constructed genetic fusions between the CD4 molecule and different types of immunoglobulins, generating hybrid immunoglobulins with antiviral action in vitro (Capon D. J. et al., Nature 337 (1989) 525-531; Traunecker A. et al., Nature 339 (1989) 68-70; also see International patent application WO 89 02922). However, the implication of the FcγRIII receptor (type 3 receptor for the Fc region of IgG's), which in humans is the antigen CD16 (Unkeless J. C. and Jacquillat C., J. Immunol. Meth. 100 (1987) 235-241), in the internalization of the HIV-1 virus (Homsy J. et al., Science 244 (1989) 1357-1360) suggests an important role of these receptors in viral propagation in vivo. The receptor, which has been recently cloned (Simmons D. and Seed B., Nature 333 (1988) 568-570), is mainly located in the membranes of macrophages, polynuclear cells and granulocytes, but in contrast to CD4, the CD16 receptor also exists in a soluble state in serum (Khayat D. et al., J. Immunol. 132 (1984) 2496-2501; Khayat D. et al., J. Immunol. Meth. 100 (1987) 235-241). It should be noted that the membraneous CD16 receptor is used as a second route of entry by the HIV-1 virus to infect macrophages, due to the presence of facilitating antibodies (Homsy J. et al., Science 244 (1989) 1357-1360). This process of infection which involves an “Fc receptor” at the surface of target cells (for example the CD16 receptor), and the Fc region of antibodies directed against the virion, is named ADE (“Antibody Dependent Enhancement”); it has also been described for the flavivirus (Peiris J. S. M. et al., Nature 289 (1981) 189-191) and the Visna-Maedi ovine lentivirus (Jolly P. E. et al., J. Virol. 63 (1989) 1811-1813). Other “Fc receptors” have been described for IgG's (FcγRI and FcγRII for example) as well as for other classes of immunoglobulins, and the ADE phenomenon also involves other types of “Fc receptors” such as that recognized by the monoclonal antibody 3G8 (Homsy J. et al., Science 244 (1989) 1357-1360; Takeda A. et al., Science 242 (1988) 580-583). One can thus call into question the efficiency of hybrid antiviral macromolecules which depend uniquely on fusions between immunoglobulins and all or part of a receptor normally used by a virus such as HIV-1 for its propagation in the host; in effect, the presence of a functional Fc fragment on such molecules could actually facilitate viral infection of certain cell types. It is also important to obtain CD4 derivatives that can be used at high therapeutic concentrations.
A different type of chimeric construction involving the bacterial protein MalE and the CD4 molecule has been studied (Clément J. M. et al., C.R. Acad. Sci. Paris 308 series m (1989) 401-406). Such a fusion allows one to take advantage of the properties of the MalE protein, in particular regarding the production and/or purification of the hybrid protein. In addition, the construction of a genetic fusion between the CD4 molecule and a bacterial toxin has also been described (Chaudhary V. K. et al., Nature 335 (1988) 369-372). In these cases, utilization of a genetic fusion involving a bacterial protein for therapy in humans can be questionable.
The discovery of the role of the ADE phenomenon in the propagation of certain viruses, in particular lentiviruses including HIV-1, justifies the search for alternatives to both the development of an anti-AIDS vaccine, and to the development of therapeutic agents based solely on fusions between immunoglobulins and molecules capable of binding the virus. This is why the anti-AIDS therapeutic agents described in the present invention are based on the fusion of all or part of a receptor used directly or indirectly by the HIV-1 virus for its propagation in vivo, with a stable plasma protein, devoid of enzymatic activity, and lacking the Fc fragment.
In particular, the present invention concerns the coupling, mainly by genetic engineering, of human albumin variants with a binding site for the HIV-1 virus. Such hybrid macromolecules derived from human serum albumin are characterized by the presence of one or several variants of the CD4 receptor arising from the modification, particularly by in vitro DNA recombination techniques (mutation, deletion, and/or addition), of the N-terminal domain of the CD4 receptor, which is implicated in the specific interaction of the HIV-1 virus with target cells. Such hybrid macromolecules circulating in the plasma represent stable decoys with an antiviral function, and will be designated by the generic term HSA-CD4. Another goal of this invention concerns the coupling of human albumin variants with variants of the CD16 molecule, which is implicated in the internalization of viruses including HIV-1 (to be designated by the generic term HSA-CD16), and in general the coupling of albumin variants with molecules capable of mimicking the cellular receptors responsible for the ADE phenomenon of certain viruses, and in particular the lentiviruses.
The principles of the present invention can also be applied to other receptors used directly or indirectly by a human or animal virus for its propagation in the host organism. For example:
Another goal of the present invention concerns the development of stable hybrid macromolecules with an anticancer function, obtained by the coupling of albumin variants with molecules able to bind growth factors which, in certain pathologies associated with the amplification of the corresponding membraneous proto-oncogenes, can interact with their target cells and induce a transformed phenotype. An example of such receptors is the class of receptors with tyrosine kinase activity (Yarden Y. and Ulrich A., Biochemistry 27 (1988) 3113-3119), the best known being the epidermal growth factor (EGF) and the colony stimulating factor I (CSF-I) receptors, respectively coded by the proto-oncogenes c-erbB-1 (Downward J. et al., Nature 307 (1984) 521-527) and c-fms (Sherr C. J. et al., Cell 41 (1985) 665-676). Another example of such receptors includes the human insulin receptor (HIR), the platelet-derived growth factor (PDGF) receptor, the insulin-like growth factor I (IGF-I) receptor, and most notably the proto-oncogene c-erbB-2, whose genomic amplification and/or overexpression was shown to be strictly correlated with certain human cancers, in particular breast cancer (Slamon D. J. et al., Science 235 (1987) 177-182; Kraus M. H. et al., EMBO J. 6 (1987) 605-610). Furthermore, the principles put forth in the present invention can be equally applied to other receptors, for example the interleukin 6 (IL-6) receptor, which has been shown in vitro to be an autocrine factor in renal carcinoma cells (Miki S. et al., FEBS Lett., 250 (1989) 607-610).
As indicated above, the hybrid macromolecules of interest are substantially preferably proteinic and can therefore be generated by the techniques of genetic engineering. The preferred way to obtain these macromolecules is by the culture of cells transformed, transfected, or infected by vectors expressing the macromolecule. In particular, expression vectors capable of transforming yeasts, especially of the genus Kluyveromyces, for the secretion of proteins will be used. Such a system allows for the production of high quantities of the hybrid macromolecule in a mature form, which is secreted into the culture medium, thus facilitating purification.
The preferred method for expression and secretion of the hybrid macromolecules consists therefore of the transformation of yeast of the genus Kluyveromyces by expression vectors derived from the extrachromosomal replicon pKD1, initially isolated from K. marxianus var. drosophilarum. These yeasts, and in particular K. marxianus (including the varieties lactis, drosophilarum and marxianus which are henceforth designated respectively as K. lactis, K. drosophilarum and K. fragilis), are generally capable of replicating these vectors in a stable fashion and possess the further advantage of being included in the list of G.R.A.S. (“Generally Recognized As Safe”) organisms. The yeasts of particular interest include industrial strains of Kluyveromyces capable of stable replication of said plasmid derived from plasmid pKD1 into which has been inserted a selectable marker as well as an expression cassette permitting the secretion of the given hybrid macromolecule at high levels.
Three types of cloning vectors have been described for Kluyveromyces:
The vectors derived from plasmid pKD1 described in European patent application EP 0 241 435 A2 are also very unstable since even the most performant vector (P3) is lost in approximately 70% of the cells after only six generations under nonselective growth conditions.
An object of the present invention concerns the utilization of certain plasmid constructions derived from the entire pKD1 plasmid; such constructions possess significantly higher stability characteristics than those mentioned in European patent application EP 0 241 435 A2. It will be shown in the present invention that these new vectors are stably maintained in over 80% of the cells after 50 generations under nonselective growth conditions.
The high stability of the vectors used in the present invention was obtained by exploiting fully the characteristics of plasmid pKD1. Besides an origin of replication, this extrachromosomal replicon system possesses two inverted repeats, each 346 nucleotides in length, and three open reading frames coding for genes A, B et C, whose expression is crucial for plasmid stability and high copy number. By analogy with the 2μ plasmid of S. cerevisiae, which is structurally related to plasmid pKD1 (Chen X. J. et al., Nucl. Acids Res. 14 (1986) 4471-4480), the proteins encoded by genes B et C are probably involved in plasmid partitioning during mitotic cell division, and may play a role in the negative regulation of gene A which encodes a site-specific recombinase (FLP). It has been shown that the FLP-mediated recombination between the inverted repeats of the 2μ plasmid of S. cerevisiae is the basis of a mechanism of autoregulation of the number of plasmid copies per cell: when copy number becomes too low to permit the production of sufficient quantities of the products of genes B and C, which act as repressors of gene A, the FLP recombinase is induced and the plasmid replicates according to a rolling circle type model, which amplifies copy number to about 50 copies per cell (Futcher A. B., Yeast 4 (1988) 2740).
The vectors published in European patent application EP 0 241 435 A2 do not possess the above-mentioned structural characteristics of plasmid pKD1 of K. drosophilarum: vector A15 does not carry the complete sequence of pKD1, and vectors P1 and P3 carry an interrupted A gene, thereby destroying the system of autoregulated replication of resident plasmid pKD1. In contrast, the pKD1-derived constructs used in the present invention maintain the structural integrity of the inverted repeats and the open reading frames A, B and C, resulting in a notably higher stability of the plasmid as well as an increased level of secretion of the therapeutically active hybrid macromolecules.
The expression cassette will include a transcription initiation region (promoter) which controls the expression of the gene coding for the hybrid macromolecule. The choice of promoters varies according to the particular host used. These promoters derive from genes of Saccharomyces or Kluyveromyces type yeasts, such as the genes encoding phosphoglycerate kinase (PGK), glyceraldehyde-3-phosphate dehydrogenase (GPD), the lactase of Kluyveromyces (LAC4), the enolases (ENO), the alcohol dehydrogenases (ADH), the acid phosphatase of S. cerevisiae (PHO5), etc . . . These control regions may be modified, for example by in vitro site-directed mutagenesis, by introduction of additional control elements or synthetic sequences, or by deletions or substitutions of the original control elements. For example, transcription-regulating elements, the so-called “enhancers” of higher eukaryotes and the “upstream activating sequences” (UAS) of yeasts, originating from other yeast promoters such as the GAL1 and GAL10 promoters of S. cerevisiae or the LAC4 promoter of K. lactis, or even the enhancers of genes recognized by viral transactivators such as the E2 transactivator of papillomavirus, can be used to construct hybrid promoters which enable the growth phase of a yeast culture to be separated from the phase of expression of the gene encoding the hybrid macromolecule. The expression cassette used in the present invention also includes a transcription and translation termination region which is functional in the intended host and which is positioned at the 3′ end of the sequence coding for the hybrid macromolecule.
The sequence coding for the hybrid macromolecule will be preceded by a signal sequence which serves to direct the proteins into the secretory pathway. This signal sequence can derive from the natural N-terminal region of albumin (the prepro region), or it can be obtained from yeast genes coding for secreted proteins, such as the sexual pheremones or the killer toxins, or it can derive from any sequence known to increase the secretion of the so-called proteins of pharmaceutical interest, including synthetic sequences and all combinations between a “pre” and a “pro” region.
The junction between the signal sequence and the sequence coding for the hybrid macromolecule to be secreted in mature form corresponds to a site of cleavage of a yeast endoprotease, for example a pair of basic amino acids of the type Lys−2-Arg−1 or Arg−2-Arg−1 corresponding to the recognition site of the protease coded by the KEX2 gene of S. cerevisiae or the KEX1 gene of K. lactis (Chen X. J. et al., J. Basic Microbiol. 28 (1988) 211-220; Wesolowski-Louvel M. et al., Yeast 4 (1988) 71-81). In fact, the product of the KEX2 gene of S. cerevisiae cleaves the normal “pro” sequence of albumin in vitro but does not cleave the sequence corresponding to the pro-albumin “Christchurch” in which the pair of basic amino acids is mutated to Arg−2-Glu−1 (Bathurst I. C. et al., Science 235 (1987) 348-350).
In addition to the expression cassette, the vector will include one or several markers enabling the transformed host to be selected. Such markers include the URA3 gene of yeast, or markers conferring resistance to antibiotics such as geneticin (G418), or any other toxic compound such as certain metal ions. These resistance genes will be placed under the control of the appropriate transcription and translation signals allowing for their expression in a given host.
The assembly consisting of the expression cassette and the selectable marker can be used either to directly transform yeast, or can be inserted into an extrachromosomal replicative vector. In the first case, sequences homologous to regions present on the host chromosomes will be preferably fused to the assembly. These sequences will be positioned on each side of the expression cassette and the selectable marker in order to augment the frequency of integration of the assembly into the host chromosome by in vivo recombination. In the case where the expression cassette is inserted into a replicative vector, the preferred replication system for Kluyveromyces is derived from the plasmid pKD1 initially isolated from K. drosophilarum, while the preferred replication system for Saccharomyces is derived from the 2μ plasmid. The expression vector can contain all or part of the above replication systems or can combine elements derived from plasmid pKD1 as well as the 2μ plasmid.
When expression in yeasts of the genus Kluyveromyces is desired, the preferred constructions are those which contain the entire sequence of plasmid pKD1. Specifically, preferred constructions are those where the site of insertion of foreign sequences into pKD1 is localized in a 197 bp region lying between the SacI (SstI) site and the MstII site, or alternatively at the SphI site of this plasmid, which permits high stability of the replication systems in the host cells.
The expression plasmids can also take the form of shuttle vectors between a bacterial host such as Escherichia coli and yeasts; in this case an origin of replication and a selectable marker that function in the bacterial host would be required. It is also possible to position restriction sites which are unique on the expression vector such that they flank the bacterial sequences. This allows the bacterial sequences to be eliminated by restriction cleavage, and the vector to be religated prior to transformation of yeast, and this can result in a higher plasmid copy number and enhanced plasmid stability. Certain restriction sites such as 5′-GGCCNNNNNGGCC-3′ (SfiI) or 5′-GCGGCCGC-3′ (NotI) are particularly convenient since they are very rare in yeasts and are generally absent from an expression plasmid.
The expression vectors constructed as described above are introduced into yeasts according to classical techniques described in the literature. After selection of transformed cells, those cells expressing the hybrid macromolecule of interest are inoculated into an appropriate selective medium and then tested for their capacity to secrete the given protein into the extracellular medium. The harvesting of the protein can be conducted during cell growth for continuous cultures, or at the end of the growth phase for batch cultures. The hybrid proteins which are the subject of the present invention are then purified from the culture supernatant by methods which take into account their molecular characteristics and pharmacological activities.
The present invention also concerns the therapeutic application of the hybrid macromolecules described therein, notably in the treatment and the prevention of AIDS, as well as the cells which are transformed, transfected, or infected by vectors expressing such macromolecules.
The examples which follow as well as the attached figures show some of the characteristics and advantages of the present invention.
The diagrams of the plasmids shown in the figures are not drawn to scale, and only the restriction sites important for the constructions are indicated.
General Cloning Techniques.
The classical methods of molecular biology such as preparative extractions of plasmid DNA, the centrifugation of plasmid DNA in cesium chloride gradients, agarose and polyacrylamide gel electrophoresis, the purification of DNA fragments by electroelution, the extraction of proteins by phenol or phenol/chloroform, the precipitation of DNA in the presence of salt by ethanol or isopropanol, transformation of Escherichia coli etc . . . have been abundantly described in the literature (Maniatis T. et al., “Molecular Cloning, a Laboratory Manual”, Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y., 1982; Ausubel F. M. et al. (eds), “Current Protocols in Molecular Biology”, John Wiley & Sons, New York, 1987), and will not be reiterated here.
Restriction enzymes are furnished by New England Biolabs (Biolabs), Bethesda Research Laboratories (BRL) or Amersham and are used according to the recommendations of the manufacturer.
Plasmids pBR322, pUC8, pUC19 and the phages M13 mp8 and M13mp18 are of commercial origin (Bethesda Research Laboratories).
For ligations, the DNA fragments are separated by size on agarose (generally 0.8%) or polyacrylamide (generally 10%) gels, purified by electroelution, extracted with phenol or phenol/chloroform, precipitated with ethanol and then incubated in the presence of T4 DNA ligase (Biolabs) according to the recommendations of the manufacturer.
Filling in of 5′ ends is carried out using the Klenow fragment of E. coli DNA polymerase I (Biolabs) according to manufacturer recommendations. Destruction of 3′ protruding termini is performed in the presence of T4 DNA polymerase (Biolabs) as recommended by the manufacturer. Digestion of 5′ protruding ends is accomplished by limited treatment with S1 nuclease.
In vitro site-directed mutagenesis is performed according to the method developed by Taylor et al. (Nucleic Acids Res. 13 (1985) 8749-8764) using the kit distributed by Amersham.
Enzymatic amplification of DNA fragments by the PCR technique (Polymerase-catalyzed Chain Reaction, Saiki R. K. et al., Science 230 (1985) 1350-1354; Mullis K. B. and Faloona F. A., Meth. Enzym. 155 (1987) 335-350) is carried out on a “DNA thermal cycler” (Perkin Elmer Cetus) according to manufacturer specifications.
Nucleotide sequencing is performed according to the method developed by Sanger et al. (Proc. Natl. Acad. Sci. USA, 74 (1977) 5463-5467), using the Amersham kit.
Transformation of K. lactis with foreign DNA as well as the purification of plasmid DNA from K. lactis are described in the text.
Unless indicated otherwise, the bacterial strains used are E. coli MC1060 (lacIPOZYA, X74, galU, galK, strAr), or E. coli TG1 (lac proA, B, supE, thi, hsdD5/F′traD36, proA+B+, lacIq, lacZ, M15).
All yeast strains used are members of the family of budding yeasts and in particular of the genus Kluyveromyces. Examples of these yeasts are given in the text. The K. lactis strain MW98-8C (α, uraA arg, lys, K+, pKD1°) was often used; a sample of this strain has been deposited on Sep. 16, 1988 at the Centraalbureau voor Schimmelkulturen (CBS) at Baarn (Netherlands) under the registration number CBS 579.88.
An MstII-SmaI restriction fragment corresponding to the V1V2 domains (where V1 and V2 designate the first two N-terminal domains of the CD4 molecule) was generated by the technique of enzymatic amplification (PCR) according to the following strategy: the lymphoblastic cell line CEM13, which expresses high quantities of CD4 receptor, was used as the source of messenger RNAs coding for the receptor. Total RNA was first purified from 3×108 cells of this line by extraction with guanidium thiocyanate as originally described by Cathala et al. (DNA 4 (1983) 329-335); 50 μg of RNA prepared in this manner then served as matrix for the synthesis of complementary DNA (cDNA) using the Amersham kit and the oligodeoxynucleotide Xol27 as primer (
E.2.1. Construction of Plasmid pXL869 Coding for Prepro-HSA.
The NdeI site of plasmid pXL322 (Latta M. et al., Bio/Technology 5 (1987) 1309-1314) including the ATG translation initiation codon of prepro-HSA was changed to a HindIII site by oligodeoxynucleotide-directed mutagenesis using the following strategy: the HindIII-BglII fragment of pXL322 containing the 5′ extremity of the prepro-HSA gene was cloned into vector M13mp18 and mutagenized with oligodeoxynucleotide 5′-ATCTAAGGAAATACAAGCTT-ATGAAGTGGGT-3′ (the HindIII site is underlined and the ATG codon of prepro-HSA is shown in bold type); the phage obtained after this mutagenesis step is plasmid pXL855 whose restriction map is shown in
E2.2. Construction of Expression Cassettes for Prepro-HSA Expressed Under the Control of the PGK Promoter of S. cerevisiae.
Plasmid pYG12 contains a 1.9 kb SalI-BamHI restriction fragment carrying the promoter region (1.5 kb) and terminator region (0.4 kb) of the PGK gene of S. cerevisiae (
Plasmid pYG208 is an intermediate construction generated by insertion of the synthetic adaptor BamHI/SalI/BamHI (5′-GATCCGTCGACG-3′) into the unique BamHI site of plasmid pYG12; plasmid pYG208 thereby allows the removal of the promoter and terminator of the PGK gene of S. cerevisiae in the form of a SalI restriction fragment (
The HindIII fragment coding for prepro-HSA was purified from plasmid pXL869 by electroelution and cloned in the “proper” orientation (defined as the orientation which places the N-terminal of the albumin prepro region just downstream of the PGK promoter) into the HindIII site of plasmid pYG208 to generate plasmid pYG210. As indicated in
E.2.3. Optimization of the Expression Cassette.
The nucleotide sequence located immediately upstream of the ATG translation initiation codon of highly expressed genes possesses structural characteristics compatible with such high levels of expression (Kozak M., Microbiol. Rev. 47 (1983) 1-45; Hamilton R et al., Nucl. Acid Res. 15 (1987) 3581-3593). The introduction of a HindIII site by site-directed mutagenesis at position-25 (relative to the ATG initiation codon) of the PGK promoter of S. cerevisiae is described in European patent application EP No 89 10480.
In addition, the utilization of oligodeoxynucleotides Sq451 and Sq452 which form a HindIII-BstEII adaptor is described in the same document and permits the generation of a HindIII restriction fragment composed of the 21 nucleotides preceding the ATG initiator codon of the PGK gene, followed by the gene coding for prepro-HSA. The nucleotide sequence preceding the ATG codon of such an expression cassette is as follows (the nucleotide sequence present in the PGK promoter of S. cerevisiae is underlined):
The cloning strategy used for the in-frame construction of the hybrid molecule prepro-HSA-V1V2 is illustrated in
The MstII-SmaI restriction fragment carrying the V1V2 domains of the CD4 receptor, obtained as described in Example 1, was cloned into plasmid pYG18 cut by the same enzymes to generate recombinant plasmid pYG303 whose restriction map is shown in
E4.1. Isolation and Purification of Plasmid pKD1.
Plasmid pKD1 was purified from K. drosophilarum strain UCD 51-130 (U.C.D. collection, University of California, Davis, Calif. 95616) according to the following protocol: a 1 liter culture in YPD medium (1% yeast extract, 2% Bacto-peptone, 2% glucose) was centrifuged, washed, and resuspended in a solution of 1.2 M sorbitol, and cells were transformed into spheroplasts in the presence of zymolyase (300 μg/ml), 25 mM EDTA, 50 mM phosphate and β-mercaptoethanol (1 μg/ml). After washing in a solution of 1.2 M sorbitol, spheroplasts corresponding to 250 ml of the original culture were resuspended in 2.5 ml of 1.2 M sorbitol to which was added the same volume of buffer (25 mM Tris-HCl, pH 8.0; 50 mM glucose; 10 mM EDTA). The following steps correspond to the alkaline lysis protocol already described (Birnboim H. C. and Doly J. C., Nucleic Acids Res. 7 (1979) 1513-1523). DNA was purified by isopycnic centrifugation in a cesium chloride gradient.
E4.1 Construction of Plasmid pCXJ1.
The intermediate construction pUC-URA3 (
Plasmid pCXJ1 (
E.4.3. Construction of an In-Frame Fusion Between ORF1 of the Killer Plasmid of K. lactis and the Product of the Bacterial Gene aph[3′]-I of Transposon Tn903.
Plasmid pKan707 was constructed as a vector to be used in wild type yeasts. This plasmid was generated by insertion of the aph[3′]-I gene of bacterial transposon Tn903 coding for 3′-aminoglycoside phosphotransferase (APH), expressed under control of a yeast promoter, into the SaI of plasmid PCXJ1.
In the first step, the bacterial transcription signals of the aph[3′]-I gene were replaced by the Pk1 promoter isolated from the killer plasmid k1 of K. lactis as follows: the 1.5 kb ScaI-PstI fragment of plasmid k1 was cloned into the corresponding sites of vector pBR322, to generate plasmid pk1-PS1535-6 (
Plasmid pUC-kan1 is an intermediate construction obtained by insertion of the 1.25 kb EcoRI fragment carrying the aph[3′]-I gene of Tn903 (Kanamycin Resistance Gene Block™, Pharmacia), into the EcoRI site of plasmid pUC19 (
E4.4. Construction and Stability of Plasmid pKan707 in K. lactis.
Plasmid pCXJ1 was cleaved by HindIII, treated with the Klenow fragment of E. coli DNA polymerase I, then ligated with the 1.2 kb ScaI-HincII fragment coding for the ORF1-APH fusion expressed under control of the K. lactis Pk1 promoter deriving from plasmid pUC-Kan202. The resulting plasmid (pKan707,
The SalI restriction fragment coding for the hybrid protein prepro-HSA-V1V2 expressed under control of the PGK promoter of S. cerevisiae was purified by electrolution from plasmid pYG306 cut by the corresponding enzyme, and then cloned into the SalI site of plasmid pKan707, to generate plasmids pYG308A and pYG308B which are distinguished only by the orientation of the SalI fragment in relation to the vector pKan707. A restriction map of plasmid pYG308B is shown in
Plasmid pYG221B is a control construction coding for prepro-HSA alone; this plasmid was constructed as for plasmid pYG308B (prepro-HSA-V1V2): the SalI fragment coding for prepro-HSA expressed under control of the PGK promoter was purified from plasmid pYG210 and cloned into the SalI site of plasmid pKan707 to generate plasmid pYG221B (
The nucleotide sequence of the MstII-HindIII fragment of plasmid pYG308B is included in the sequence of the MstII-SmaI fragment shown in
Transformation of yeasts of the genus Kluyveromyces and in particular K. lactis strain MW98-8C, was performed by treating whole cells with lithium acetate (Ito H. et al., J. Bacteriol. 153 (1983) 163-168), adapted as follows. Cells were grown in shaker flasks in 50 ml of YPD medium at 28° C., until reaching an optical density of 0.6-0.8, at which time they were harvested by low speed centrifugation, washed in sterile TE (10 mM Tris HCl pH 7.4; 1 mM EDTA), resuspended in 3-4 ml of lithium acetate (0.1 M in TE) to give a cell density of 2×108 cells/ml, then incubated 1 hour at 30° C. with moderate agitation. Aliquots of 0.1 ml of the resulting suspension of competent cells were incubated 1 hour at 30° C. in the presence of DNA and polyethylene glycol (PEG4000, Sigma) at a final concentration of 35%. After a 5 minute thermal shock at 42° C., cells were washed twice, resuspended in 0.2 ml sterile water, and incubated 16 hours at 28° C. in 2 ml YPD to allow for phenotypic expression of the ORF1-APH fusion protein expressed under control of promoter Pk1; 200 μl of the resulting cell suspension were spread on YPD selective plates (G418, 200 μg/ml). Plates were incubated at 28° C. and transformants appeared after 2 to 3 days growth.
After selection on rich medium supplemented with G418, recombinant clones were tested for their capacity to secrete the mature form of albumin or the hybrid protein HSA-V1V2. Certain clones corresponding to strain MW98-8C transformed by plasmids pYG221B (prepro-HSA) or pYG308B (prepro-HSA-V1V2) were incubated in selective liquid rich medium at 28° C. Culture supernatants were prepared by centrifugation when cells reached stationary phase, then concentrated by precipitation with 60% ethanol for 30 minutes at 20° C. Supernatants were tested after electrophoresis through 8.5% polyacrylamide gels, either by direct Coomassie blue staining of the gel (
After ethanol precipitation of the culture supernatants corresponding to the K. lactis strain MW98-8C transformed by plasmids pYG221B (prepro-HSA) and pYG308B (prepro-HSA-V1V2), the pellet was resolubilized in a 50 mM Tris-HCl buffer, pH 8.0. The HSA-CD4 and HSA proteins were purified by affinity chromatography on Trisacryl-Blue (IBF). An additional purification by ion exchange chromatography can be performed if necessary. After elution, protein-containing fractions were combined, dialyzed against water and lyophylized before being characterized. Sequencing (Applied Biosystem) of the hybrid protein secreted by K. lactis strain MW98-8C revealed the expected N-terminal sequence of albumin (Asp-Ala-His . . . ), demonstrating the proper maturation of the protein.
The isoelectric point was determined by isoelectrofocalization to be 5.5 for the HSA-V1V2 protein and 4.8 for HSA.
The HSA-V1V2 protein is recognized by the monoclonal mouse antibodies OKT4A and Leu3A directed against human CD4, as well as by a polyclonal anti-HSA serum (
The proteins corresponding to albumin (negative control) and to the HSA-V1V2 fusion purified from culture supernatants of K. lactis strain MW98-8C transformed respectively by plasmids pYG221B (prepro-HSA) and pYG308B (prepro-HSA-V1V2) as in examples 7 and 8, were tested in vitro for antiviral activity and compared to the entire soluble CD4 molecule purified from CHO (Chinese Hamster Ovary) cells. Protein concentrations are expressed in molarity and were determined both by methods to measure proteins in solution as well as by comparison of successive dilutions of each protein after electrophoretic migration in polyacrylamide gels followed by silver nitrate staining.
The HSA-CD4 protein is also able to inhibit viral infection of permissive cells in cell culture. This inhibition was measured either by assaying the production of viral antigens (viral p24) using the kit ELAVIA-AG1 (Diagnostics Pasteur), or the kit p24-ELISA (Dupont), or by measuring the reverse transcriptase activity by the technique of Schwartz et al. (Aids Research and Human Retroviruses 4 (1988) 441-448). The experimental protocol was as follows: the product of interest at a final concentration X was first preincubated with supernatants of CEM13 cells infected by the isolate LAV-Bru1 of virus HIV-1 (dilution 1/250, 1/2500 and 1/25000) in a total volume of 1 ml of culture medium (RPMI 1640 containing 10% fetal calf serum, 1% L-glutamine and 1% penicillin-streptomycin-neomycin). The mixture was then transfered onto a pellet of 5×105 permissive cells (e.g. MT2, CEM13, or H9) and incubated in tubes for 2 hours at 37° C. for infection to occur. The infection could also be carried out on microtitration plates with 104 cells per well in 100 μl of complete medium. A volume of 100 μl of the virus that had been preincubated with the product to be tested was then added, followed by 50 μl of the product at 5× concentration. Cells were then washed twice with 5 ml RPMI 1640 and resuspended in culture medium at a density of 2:5×105 cells/ml. 100 μl of this suspension was then aliquoted into each well of microtitration plates which already contain 100 μl of the product at 2× concentration, and the plates were incubated at 37° C. in a humid atmosphere containing 5% CO2. At different days (D3-D4-D6-D8-D10-D12-D14-D16-D19-D21 and D25), 100 μl of supernatant was removed and the p24 viral production as well as the reverse transcriptase activity were assayed. Cells were then resuspended and distributed onto microtitration plates for assays of cell viability (MTT) as described above. To the 50 μl remaining on the original plates, 200 μl of culture medium containing the product to be tested at concentration X were added, and infection was allowed to progress until the next sampling. For the cell viability test, 10 μl of MTT at 5 mg/ml filtered on 0.2 μm filters was added to each well and plates were incubated 4 hours at 37° C. in a humid atmosphere containing 5% CO2. Then to each well was added 150 μl of an isopropanol/0.04 N HCl mixture, and the Formazan crystals were resuspended. Optical density from 520 to 570 nm was measured on a Titertek plate reader; this measure reflects cell viability (Schwartz et al., Aids Research and Human Retroviruses 4 (1988) 441-448).
It has been shown that first generation soluble CD4 possesses a half-life of 20 minutes in rabbits (Capon D. J. et al.; Nature 337 (1989) 525-531). We have therefore compared the half-life in rabbits of the HSA-CD4 hybrid to soluble CD4 and to recombinant HSA produced in yeast and purified in the same manner as HSA-CD4. In these experiments, at least 2 male NZW(Hy/Cr) rabbits weighing 2.5-2.8 kg were used for each product. Rabbits were kept in a room maintained at a temperature of 18.5-20.5° C. and a humidity of 45-65%, with 13 hours light/day. Each product was administered in a single injection lasting 10 seconds in the marginal vein of the ear. The same molar quantity of each product was injected: 250 μg of CD4 per rabbit, 400 μg of HSA per rabbit, or 500 μg of HSA-CD4 per rabbit, in 1 ml physiologic serum. Three to four ml blood samples were taken, mixed with lithium heparinate and centrifuged 15 min at 3500 rpm; samples were then divided into three aliquots, rapidly frozen at −20° C., then assayed by an ELISA method. Blood samples from rabbits injected with CD4 were taken before injection (To), then 5 min, 10 min, 20 min, 30 min, 1 h, 2 h, 4 h and 8 h after injection. Blood samples from rabbits injected with HSA-CD4 or HSA were taken at To, 30 min, 1 h, 2 h, 4 h, 8 h, 24 h, 32 h, 48 h, 56 h, 72 h, 80 h, 96 h, 104 h and 168 h after injection.
Assays of the CD4 molecule were carried out on Dynatech M129B microtitration plates previously covered with the HSA-CD4 hybrid protein. Increasing concentrations of CD4 or the samples to be assayed were then added in the presence of the mouse monoclonal antibody OKT4A (Ortho-Diagnostic, dilution 1/1000); after incubation and washing of the plates, the residual binding of antibody OKT4A was revealed by addition of antibodies coupled to peroxidase (Nordic, dilution 1/1000) and directed against mouse IgG. Measurements were made at OD 405 nm in the presence of the peroxidase substrate ABTS (Fluka).
Assays of recombinant HSA were carried out on Dynatech M129B microtitration plates previously covered with anti-HSA serum (Sigma Ref. A0659, dilution 1/1000); increasing concentrations of HSA or samples to be measured were then added, followed by addition of anti-HSA serum coupled to peroxidase (Nordic, dilution 1/1000). Measurements were made at OD 405 nm as above.
Two different assays were done for the HSA-CD4 hybrid: either the assay for the HSA moiety alone, using the same methods as for recombinant HSA, or an assay for the HSA moiety coupled with an assay for the CD4 moiety. In the latter case, microtitration plates were covered first with anti-HSA serum (Sigma Ref. A0659, dilution 1/1000), then incubated with the samples to be assayed. The mouse monoclonal antibody Leu3A directed against CD4 was then added, followed by antibodies coupled to peroxidase (Nordic, dilution 1/1000) and directed against mouse antibodies. Measurements were made at 405 nm as described above.
The curves for each of these assays are given in
These results underscore the following points:
It is noteworthy that the volume of distribution of HSA and HSA-CD4 is close to that of the blood compartment, and therefore suggests a distribution of the product limited to the extracellular compartment.
E.11.1. Introduction of AhaII and BglII Sites at the End of the Prepro Region of HSA.
Introduction of the AhaII restriction site was carried out by site-directed mutagenesis using plasmid pYG232 and oligodeoxynucleotide Sq1187, to generate plasmid pYG364. Plasmid pYG232 was obtained by cloning the HindIII fragment coding for prepro-HSA into the vector M13mp9. The sequence of oligodeoxynucleotide Sq1187 is (the AhaII site is in bold type):
It should be noted that creation of the AhaII site does not modify the protein sequence of the N-terminal of mature HSA. The construction of plasmid pYG364 is shown in
Plasmid pYG233 was obtained in analogous fashion, after site-directed mutagenesis of plasmid pYG232 using oligodeoxynucleotide Sq648 (the codons specificying the amino acid pair Arg-Arg situated at the end of the prepro region of HSA are in bold type, and the BglII site is underlined):
The creation of this restriction site does not change the protein sequence of the prepro region of HSA. In contrast, the first amino acid of the mature protein is changed from an aspartate to a serine; plasmid pYG233 therefore codes for a mature HSA modified at its N-terminal (HSA*,
E11.2. Introduction of the Prepro Region of HSA Upstream of the CD4 Receptor.
The introduction of the prepro region of HSA upstream of the V1V2 domains of the CD4 receptor was accomplished by site-directed mutagenesis, to generate plasmid pYG347 as shown in
Plasmid pYG347 was obtained by site-directed mutagenesis of plasmid pYG332 with oligodeoxynucleotide Sq1092 (
Plasmid pYG347 therefore carries a HindIII fragment composed of the 21 nucleotides preceding the ATG codon of the PGK gene of S. cerevisiae, the ATG translation initiation codon, and the prepro region of HSA (LPHSA) immediately followed by the V1V2 domains of the CD4 receptor.
E.11.3. Introduction of an AhaII Site at the End of the V1 Domain of the CD4 Receptor.
The introduction of an AhaII site at the end of the V1 domain of the CD4 receptor was accomplished by site-directed mutagenesis using oligo-deoxynucleotide Sq1185 and a derivative of plasmid pYG347 (pYG368,
Plasmid pYG362 therefore carries a HindIII-AhaII fragment composed of the 21 nucleotides preceding the ATG codon of the PGK gene of S. cerevisiae followed by the coding sequence of the HSA prepro region fused to the V1 domain of the CD4 receptor, according to example E.11.2. In a fusion such as the example given here, the V1 domain of the CD4 receptor carries 106 amino acids and includes the functional binding site of the HIV-1 viral glycoprotein gp120.
E.11.4. Introduction of an AhaII Site at the End of the V2 Domain of the CD4 Receptor.
The introduction of an AhaII site at the end of the V2 domain of the CD4 receptor was accomplished by site-directed mutagenesis using oligo-deoxynucleotide Sq1186 and plasmid pYG368, to generate plasmid pYG363 (
Other variants of plasmid pYG363 were generated by site-directed mutagenesis in order to introduce an AhaII at different places in the V2 domain of the CD4 receptor. In particular, plasmid pYG511, shown in
E11.5. Generic Constructions of the Type V1-HSA.
The plasmids described in the preceding examples allow for the generation of HindIII restriction fragments coding for hybrid proteins in which the receptor of the HIV-1 virus (fused to the signal sequence of HSA) precedes HSA. For example, plasmids pYG362 and PYG364 are respectively the source of a HindIII-AhaII fragment (fusion of the HSA prepro region to the V1 domain of the CD4 receptor), and an AhaII-NcoI fragment (N-terminal region of mature HSA obtained as in example E.11.1.). The ligation of these fragments with the NcoI-KpnI fragment (C-terminal region of HSA and terminator of the PGK gene of S. cerevisiae) in an analogue of plasmid pYG18 cut by HindIII and KpnI generates plasmid pYG371 whose structure is shown in
E11.6. Generic Constructions of the Type V1V2-HSA.
Hybrid proteins of the type V1V2-HSA were generated by the following strategy: in a first step, plasmids pYG511 (
E.12.1. Introduction of a Stop Codon Downstream of the V1 Domain of the CD4 Receptor
Conventional techniques permit the introduction of a translation stop codon downstream of the domain of the CD4 receptor which is responsible for the binding of the HIV-1 viral glycoprotein gp120. For example, a TAA codon, immediately followed by a HindIII site, was introduced by site-directed mutagenesis downstream of the V1 domain of the CD4 receptor. In particular, the TAA codon was placed immediately after the amino acid in position 106 of the CD4 receptor (Thr106) using oligodeoxynucleotide Sq1034 and a plasmid analogous to plasmid M13-CD4 as matrix. The sequence of oligodeoxynucleotide Sq1034 is (the stop codon and the HindIII site are in bold type):
E12.2 Constructions of the Type CD4HSA-CD4.
The plasmids described in examples E.11.5. et E.11.6. which exemplify generic constructions of the type CD4-HSA allow for the easy generation of bivalent constructions of the type CD4-HSA-CD4. Plasmids pYG374 (V1-HSA-V1V2) or pYG375 (V1-HSA-V1) illustrate two of these generic constructions: for example, the small MstII-HindIII fragment of plasmid pYG371 which codes for the last amino acids of HSA can be replaced by the MstII-HindIII fragment coding for the last 3 amino acids of HSA fused to the V1V2 domains of the CD4 receptor (plasmid pYG374,
The construction of HindIII fragments coding for bivalent hybrid proteins of the type V1V2-HSA-V1V2 has already been described in
E.12.3. Introduction of a Dimerization Domain.
For a given hybrid protein derived from albumin and carrying one or several binding sites for the HIV-1 virus, it may be desirable to include a polypeptide conferring a dimerization function, which allows for the agglomeration of trapped virus particles. An example of such a dimerization function is the “Leucine Zipper” (LZ) domain present in certain transcription regulatory proteins (JUN, FOS . . . ). In particular, it is possible to generate a BglII-AhaII fragment coding, for example, for the LZ of JUN, by the PCR technique by using the following oligodeoxynucleotides and the plasmid pTS301 (which codes for an in-frame fusion between the bacterial protein LexA and the LZ of JUN, T. Schmidt and M. Schnar, unpublished results) as matrix (BglII and AhaII sites are underlined):
This BglII-AhaII fragment (
The introduction of carefully selected restriction sites that permit the construction of genes coding for hybrid proteins of the type LZ-CD4-HSA or LZ-CD4-HSA-CD4 is also possible, using conventional in vitro mutagenesis techniques or by PCR.
E-13.1. Strategy Using Bal31 Exonuclease.
Proteins secreted by strain MW98-8C transformed by expression plasmids for HSA-CD4 hybrid proteins in which the CD4 moiety is carried on the MstII-HindIII fragment in the natural MstII site of HSA (plasmid pYG308B for example), were analyzed.
E.13.2. Mutation of Dibasic Amino Acid Pairs.
Another way to prevent cleavage by endoproteases with specificity for dibasic amino acid pairs is to suppress these sites in the area of the hinge region between the HSA and the CD4 moieties (
Plasmid M13-ompA-CD4 is a derivative of plasmid M13-CD4 in which the signal sequence of the ompA gene of E. coli is fused in frame to the CD4 receptor using the MstII site generated by PCR upstream of the V1 domain (example 1).
E13.3. Introduction of a Synthetic Hinge Region.
In order to promote an optimal interaction between the CD4 moiety fused to HSA, and the gp120 protein of the HIV-1 virus, it may be desirable to correctly space the two protein moieties which form the building blocks of the hybrid protein HSA-CD4. For example, a synthetic hinge region can be created between the HSA and CD4 moieties by site-directed mutagenesis to introduce a fragment of troponin C between amino acids 572 and 582 of mature HSA (
Similar techniques also permit the introduction of such a synthetic hinge region between the HSA and CD4 moieties (junction peptide,
For a given protein secreted by cells at high levels, there exists a threshold above which the level of expression is incompatible with cell survival. Hence there exist certain combinations of secreted protein, promoter utilized to control its expression, and genetic background that are optimal for the most efficient and least costly production. It is therefore important to be able to express the hybrid proteins which are the object of the present invention under the control of various promoters. The composite genes coding for these proteins are generally carried on a HindIII restriction fragment that can be cloned in the proper orientation into the HindIII site of a functional expression cassette of vectors that replicate in yeasts. The expression cassette can contain promoters that allow for constitutive or regulated expression of the hybrid protein, depending on the level of expression desired. Examples of plasmids with these characteristics include plasmid pYG105 (LAC4 promoter of K. lactis,
| Number | Date | Country | Kind |
|---|---|---|---|
| 89 10480 | Aug 1989 | FR | national |
| Number | Date | Country | |
|---|---|---|---|
| Parent | 10073118 | Feb 2002 | US |
| Child | 10794774 | Mar 2004 | US |
| Parent | 09551635 | Apr 2000 | US |
| Child | 10073118 | Feb 2002 | US |
| Parent | 09004319 | Jan 1998 | US |
| Child | 09551635 | Apr 2000 | US |
| Parent | 08479146 | Jun 1995 | US |
| Child | 09004319 | Jan 1998 | US |
| Parent | 08295078 | Aug 1994 | US |
| Child | 08479146 | Jun 1995 | US |
| Parent | 08121236 | Sep 1993 | US |
| Child | 08295078 | Aug 1994 | US |
| Parent | 07955243 | Oct 1992 | US |
| Child | 08121236 | Sep 1993 | US |
| Parent | 07561879 | Aug 1990 | US |
| Child | 07955243 | Oct 1992 | US |
