Glycosyltransferases for biosynthesis of oligosaccharides and genes encoding them

This work leading to the present invention was partially supported by a fund under approval number AI-10615 from the Public Health Service. Therefore, this administrative body may own some of the rights of the present invention.<u style="single">Field of invention</u> The present invention relates to glycosyltransferases useful for biosynthesis of oligosaccharides, genes encoding such glycosyltransferases and recombinant methods for producing the enzymes, and oligosaccharides produced by the methods.

<u style="single">Neisseria and lipo-oligosaccharides (LOS)</u> Neisseria species generally live in many mammalian hosts, but humans are the only species to suffer from invasive disease by members of this species. Neisseria meningitidis is an etiologic factor for sepsis and meningitis that can occur as a mumps disease. Neisseria gonorrhoeae is a causative agent of gonorrhea or its various complications. These organisms, especially gonococci, have been shown to significantly skillfully alter the antigenic array of molecules exposed on their surface, especially their adhesive pili and opacity-related (opa) proteins. Genetic mechanism for this pili modification (Meyer et al., Cell, 1982, 30:45; Haas and Meyer, Cell 1986,44: 107; Koomey et al., Genetics, 1987, 117: 391; Swanson and Koomey, American Society for Microbiology, Washington, 743-761) and opa Protein expression (Stern et al., Cell, 1986, 47:61; Meyer et al., Ann. Rev. Microbiol., 1990, 44: 451; Bhat et al., Molec. Microbiol., 1991, 5: 1889) is well understood. Like other Gram-negative bacteria, Neisseria ssp. Has LPS in the fronds of its exodermis [Johnston and Gotschlich, J. Bacteriol., 1974, 119: 250).

In contrast to the high molecular weight LPS with repetitive O-chains found in many intestinal bacteria, the LPS of Neisseria ssp. Has a medium size and is therefore often referred to as lipooligosaccharide or LOS. There is. The molecular size of LOS is similar to that found in crude LPS variants of Salmonella ssp., But this material has considerable antigenic diversity. In the case of N. meningitidis, a serological typing method has been developed that divides the strain into 12 immune types [Zollinger and Mandrell, Infect. Immun., 1977, 18: 424; Zollinger and Mandrell, Infect. Immun., 1980, 28: 451). A fairly complete understanding of the structure of N. meningitidis LPS (recent overview: Verheul et al., Microbiol. Rev., 1993, 57:34) was the work of Jennings and his collaborators (Jennings). (Jennings) et al., Carbohyd. Res., 1983, 121: 233; Michon et al., J. Biol. Chem., 1990, 265: 7243; Gamian et al., J. Biol. Chem., 1992, 267: 922; Pavliak et al., J. Biol. Chem., 1993, 268: 14146). In the case of Neisseria gonoroae, the antigenic alteration is very prominent and elusive by serological classification. In part, this is due to the heterogeneity of LOS synthesized by a particular strain, and LOS preparations often contain several closely spaced SDS-PAGE bands (Mandrell et al., Etc., Infect. Immun., 1986, 54: 63). In addition, studies using monoclonal antibodies have shown that N. gonorrhoeae can alter the serological characteristics of the LOS it expresses, and that this antigenic change occurs with a frequency of 10-2 to 10-3. This means that the presence of certain genetic mechanisms is required to achieve these high frequency mutations [Schneider et al., Infect. Immun., 1988, 56: 942; Apicella et al., Infect. Immun., 1987, 55: 1755).

Due to the molecular heterogeneity and antigenic alterations of LOS produced by Neisseria gonorrhoeae, structural chemical determination of this antigen has proven difficult, and definitive information based on extremely elaborate analysis is available. Only recently available [Yamasaki et al., Biochemistry, 1991, 30: 10566; Kerwood et al., Biochemistry, 1992, 31: 12760; John et al.,. Boil.Chem., 1991, 266: 19303; Gibson et al., J. Bacteriol., 1993, 175: 2702). These are summarized in Fig. 1. Of particular interest is the presence of the tetrasaccharide Galβ1 4 GlcNAcβ1 3Galβ1 4Glcβ1 4, which is a complete mimic of the sphingolipid paragloboside lacto-N-neotetraose [Mandrel et al., J. .Ex.Med.,1988,168: 107; Tsai and Civin, Infect. Immun., 1991, 59: 3604). In LOS, this tetrasaccharide often has an associated N-acetylgalactose amine residue (GalNAcβ1 3Galβ1 4 GlcNAcβ1 3Galβ1 4Glcβ1 4) and is therefore similar to a ganglioside. In some strains of N. gonorrhoeae, alternating side chains are found, which have the structure Galα1 4Galβ1 4Glcβ1 4hep R [John et al., J. Biol. Chem., 1991, 266: 19303. ). It is an analog of the sugar moiety of globo-glycolipid [Mandrell et al., Infect. Immun., 1992, 60: 3017] and is characteristically found in immune form L1 of Neisseria meningitidis. It is a structure.

The LOS molecule has several biological activities. It is a powerful endotoxin molecule that is thought to be a toxin involved in adrenal cortex necrosis. They act as target antigens for most of the bactericidal activity present in normal or convalescent human serum [Rice et al., J. Immunol., 1980, 124: 2105]. Neisseria gonorrhoeae has a very familiar thin sialyltransferase activity, which allows the use of externally supplied CMP-NANA and adds N-acetylneuraminic acid to the LOS on the surface of the organism. [Nairn et al., J.Gen. Microbiol., 1988, 134: 3295; Parsons et al., Microb. Pathog., 1989, 7:63; Mandrell et al., J. Ex. Med., 1990, 171: 1649]. Group B and C meningococci have the ability to synthesize CMP-NANA and often sialylate their LOS without the need for exogenous CMP-NANA [Mandrell et al., J. Bacteriol ., 1991, 173: 2823]. In the Neisseria meningitidis strain 6275 immune type L3, its sialic acid unit is bound to the terminal Gal of lacto-N-neotetraose (α2 3) [Yamasaki et al., J. Bacteriol., 1993. , 175: 4565].

Concentrations of CMP-NANA found in various host environments are sufficient to sustain this reaction [Apicella et al., J. Infect.Dis., 1990, 162: 506]. This sialylation of LOS makes gonococci resistant to the antibody-complement-dependent bactericidal action of serum [Parsons et al., Microb. Pathog., 1989, 7:63]. This resistance acts not only on the above-mentioned bactericidal action mediated by antibodies against LOS, but also on other surface antigens [Wetzler et al., Infect. Immun., 1992, 60:39]. Van Putten demonstrated that exposure of Neisseria gonorrhoeae to CMP-NANA significantly reduced the ability of the gonococcus to invade epithelial cells in tissue culture [van Putten, EMBO J., 1993. , 12:40 43]. These findings strongly suggest that the ability of Neisseria gonorrhoeae to alter the chemistry of the LOS gives them the ability to cope with different host environments [Mandrell and Apicella), Immunobiology., 1993, 187: 382].

Perhaps most effectively, LOS changes have been found to have been screened in vivo during human infection. A well-characterized experimental strain of N. gonorrhoeae, MS11mk variant A, was used to inoculate applicants [Swanson et al., J. Ex. Med., 1988, 168: 2121]. In two subjects infected within a period of 4-6 days, the gonococcal group recovered in their urine gradually transformed into two variants expressing antigenically different LOS [Schneider]. Etc., J.Ex.Med., 1991, 174: 1601]. Structural analysis showed that the inoculated mutant A produced a truncated LOS containing only the β-lactosyl group bound to Hepl, while one of the newly generated mutants (mutant C) was a complete LOS. Was produced [Kerwood et al., Biochemistry, 1992, 31: 12760]. This fact suggests that the addition of the accompanying sugar GalNAcβ1 3Galβ1 4 GlcNAcβ1 3 is thought to be under the control of the framemutation mechanism.

Little information is available on LOS synthetic genes in Neisseria. The main developments are the creation [Dudas and Apicella, Infect. Immun., 1988, 56: 499] and biochemical characteristics of five piocin mutants (called 1291a-e) of Neisseria gonorrhoeae strain 1291. It was attached [John et al., J. Biol. Chem., 1991, 266: 19303]. Immunological and biochemical data show that 1291a, 1291c, 1291d and 1291e produced LOS with a shortened lacto-N-neotetraose chain sequence, with mutant 1291e also having a glucose substitution on the heptose. Showed that there is no such thing. Mutant 1291b has another LOS structure: GALα1 4Galβ1 4Glc (see Fig. 1) is synthesized. Only the genetic basis of mutant 1291e is currently elucidated. This is a mutation in phosphoglucomutase (pgm), the latter interfering with the synthesis of UDP-glucose, resulting in the addition of the first residue of the lacto-N-neotetraose unit [Zhou et al. , J.Biol.Chem., 1994,269: 11162; Sandlin and Stein, J.Bacteriol., 1994,176: 2930]. It has also been shown that galE mutants of Neisseria gonorrhoeae or Neisseria gonorrhoeae produce incised LOS while retaining the non-synthetic capacity of UDP-galactose [Robertson et al., Molec. Microbiol. ., 1993, 8:891; Jennings et al., Molec. Microbiol., 1993, 10: 361].

Biosynthesis of oligosaccharides Oligosaccharides are polymers with varying numbers of residues, bonds and subunits. The basic subunits are carbohydrate monosaccharides or sugars such as mannose, glucose, galactose, N-acetylglucoseamine, N-acetylgalactoseamine and the like. The number of different possible steric oligosaccharide chains is enormous. Oligosaccharides and polysaccharides play important roles in protein function and activity by acting as half-life modifiers and, in some cases, giving structures. As pointed out above, oligosaccharides are essential for the antigenic alterations of Neisseria, especially Neisseria gonorrhoeae, and the resulting immune disqualification.

Although many classical techniques have been developed for the synthesis of carbohydrates, it is difficult to require selective protection and deprotection for these techniques. Organic synthesis of oligosaccharides is hampered by more glycosidic bond instability, difficulty in achieving regioselective sugar binding, and generally low synthesis yields. Simply put, unlike experience with peptide synthesis, traditional synthetic organic chemistry cannot synthesize even fairly simple oligosaccharides quantitatively and with high reliability. Recent advances in the synthesis of oligosaccharides have been made to the isolation of glycosyltransferases. These enzymes can be used in vitro to synthesize oligosaccharides and polysaccharides (see, eg, Roth, US Pat. No. 5,180,674, dated January 19, 1993). The advantage of biosynthesis with glycosyltransferases is that the glycosidic bonds formed by these enzymes are significantly steric and position-specific. However, each enzyme catalyzes the binding of a particular sugar residue to another particular acceptor molecule, such as an oligosaccharide or lipid. Thus, the synthesis of certain oligosaccharides may be limited by the availability of glycosyltransferases (see Ross's International Patent Application WO 93/13198, dated 8 July 1993).

Another drawback of biosynthesis is that the glycosyltransferase itself is usually present only in very low content in cells. It is difficult to have a sufficient amount of the enzyme to allow industrial practice. Therefore, there is a great demand for glycosyltransferases in the art. In addition, there is a demand for genes encoding such glycosyltransferases to provide an unrestricted source of glycosyltransferases by recombinant techniques. Citation of any reference herein should not be understood as a license to make such a reference available as a prior art of the invention.

<u style="single">Outline of the invention</u> An object of the present invention is to provide a nucleic acid encoding a glycosyltransferase, a protein encoded by the nucleic acid, and a method for synthesizing an oligosaccharide using the glycosyltransferase of the present invention. Thus, in one aspect, the invention corresponds to or under moderately strict conditions a nucleic acid corresponding to the LOS locus of Neisseria, eg, the nucleotide sequence shown in FIG. 2 (SEQ ID NO: 1). Provided is a purified nucleic acid that can be hybridized with a nucleic acid having a nucleotide sequence complementary to the above. Preferably, the nucleic acid of the invention comprises a coding sequence portion of the gene at the LOS locus, i.e. a portion of the nucleotide sequence shown in FIG. 2 encoding a functionally active glycosyltransferase (SEQ ID NO: 1). Be hybridized.

In certain embodiments, the present invention has a nucleotide sequence that corresponds to or is complementary to a portion of the nucleotide sequence shown in FIG. 2 that encodes a functionally active glycosyltransferase. Regarding nucleic acids. In a further aspect, the nucleic acid encodes a functionally active glycosyltransferase. In one particular aspect, the invention is intended to provide a nucleic acid having a nucleotide sequence that corresponds to or is complementary to the nucleotide sequence shown in FIG. 2 (SEQ ID NO: 1). The functionally active glycosyltransferases of the present invention are characterized by catalyzing a reaction selected from the following groups: Addition of Gal β1 4 to GlcNAc or Glc; Addition of GalNAc or GlcNAc β1 3 to Gal; and addition of Gal α1 4 to Gal. Most preferably, this nucleic acid encodes a functionally active glycosyltransferase. However, the nucleic acids of the invention contain oligonucleotides that are useful as primers for the polymerase chain reaction (PCR) or as probes for the level of transcription of the glycosyltransferase gene and its presence.

In certain embodiments embodied herein, the nucleic acid is a glycosyl having an amino acid sequence of SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6 or SEQ ID NO: 8. Encode a transferase. The present invention is also related to an expression vector containing a nucleic acid encoding the glycosyltransferase of the present invention, which is further functionally linked to an expression control sequence. Thus, the invention extends to recombinant host cells transformed with such expression vectors. In another aspect, the invention also relates to a method for producing a glycosyltransferase, which comprises culturing the recombinant host cell under conditions that allow the expression of the glycosyltransferase and the expression. Includes a step of recovering the glycosyltransferase. In the first aspect, the present invention is a glycosyltransferase having an amino acid sequence of SEQ ID NO: 3, SEQ ID NO: 4, SEQ ID NO: 5, SEQ ID NO: 6 or SEQ ID NO: 8, functionally thereof. Intended to provide an active fragment. The present invention also contemplates compositions containing a glycosyltransferase bound to a solid phase carrier, wherein the glycosyltransferase is a glycosyltransferase having the amino acid sequence of SEQ ID NO: 3, or a functionally active fragment thereof. , A glycosyltransferase having an amino acid sequence of SEQ ID NO: 8, or a functionally active fragment thereof, a glycosyltransferase having an amino acid sequence of SEQ ID NO: 4, or a functionally active fragment thereof, and SEQ ID NO: It is selected from the group consisting of a glycosyltransferase having an amino acid sequence of 5 or a functionally active fragment thereof, and a glycosyltransferase having an amino acid sequence of SEQ ID NO: 6 or a functionally active fragment thereof.

Given the novel glycosyltransferases and the genes encoding them, the present invention also provides methods for preparing oligosaccharides, such as two or more sugars. In certain embodiments, the present invention contacts a reaction mixture containing activated GalNAc or GlcNAc with an acceptor moiety containing a Gal residue in the presence of a glycosyltransferase having the amino acid sequence of SEQ ID NO: 3. Method of adding GalNAc or GlcNAcβ1 3 to Gal, including the step of allowing the reaction mixture containing activated Gal to be added to GlcNAc or Glc residue in the presence of a glycosyltransferase having the amino acid sequence of SEQ ID NO: 8. A method of adding Galβ1 4 to GlcNAc or Glc, including contacting with a group-containing acceptor moiety; a reaction mixture containing activated Gal of a glycosyltransferase having the amino acid sequence of SEQ ID NO: 4. Gal α1 4 Gal, including the step of contacting with an acceptor moiety containing a Gal residue in the presence Method of adding to; a reaction mixture containing activated GalNAc or GlcNAc is contacted with an acceptor moiety containing a Gal residue in the presence of a glycosyltransferase having an amino acid sequence of SEQ ID NO: 5. , GalNAc or GlcNAc β1 3 is added to Gal; and the reaction mixture containing activated Gal contains a GlcNAc or Glc residue in the presence of a glycosyltransferase having the amino acid sequence of SEQ ID NO: 6. It is also related to the method of adding Galβ1 4 to GlcNAc or Glc, which includes the step of contacting the acceptor portion.

In a preferred embodiment, the oligosaccharide is prepared on a carrier that is harmless to mammals, especially humans, such as lipid isoprenoids or polyisoprenoid alcohols. One specific example of such a carrier is a recall phosphate. In a specific embodiment, the oligosaccharide of the present invention is attached to the carrier via an unstable bond, thus allowing the oligosaccharide to be chemically separated from the lipid carrier. Also, oligosaccharide transferases can be used, for example, to transfer the oligosaccharides from lipid carriers to proteins. In yet another embodiment, the glycosyltransferases of the invention can be expressed in eukaryotic expression systems to glycosylate proteins expressed in such systems. One of the important advantages of the present invention is to achieve the synthesis of Neisseria oligosaccharide antigens independently of the significantly harmful lipid A. Although theoretically desirable for vaccine preparation, the use of natural LOS from Neisseria has not been successful. LOS lipid A protein is a potent endotoxin and is extremely harmful. The LOS Chemical treatment, for example by hydrolysis, destroys the antigenicity of the oligosaccharide and releases useless products. Thus, for the preparation of vaccines, it is highly desirable to have a source of Neisseria oligosaccharides attached to non-toxic lipids.

Therefore, the present invention provides glycosyltransferases and a large number of oligosaccharides such as Galα1 4Galβ1 4Glc, Galβ1 4GlcNAcβ1 3Galβ1 4Glc or GalNAcβ1 3Galβ1 4 GlcNAcβ1 3Galβ1 4Glc (but not limited to these). Also provides a method for preparing. Therefore, it is a main object of the present invention to provide a glycosyltransferase useful for synthesizing oligosaccharides. A further object of the present invention is to provide a method for synthesizing oligosaccharides characteristic of Neisseria meningitidis and N. gonorrhoeae. An object of the present invention is to provide a method for synthesizing oligosaccharides characteristic of mammalian oligosaccharides, further including blood group core oligosaccharides. Yet another object of the present invention is to provide a vaccine that contains the oligosaccharide units of LOS but does not contain lipid A. Another object of the present invention is to synthesize therapeutically useful oligosaccharides. These and other objects of the invention will become apparent with reference to the drawings and detailed description below.

<u style="single">Detailed description of the invention</u> As mentioned above, the present invention provides five novel glycosyltransferases, genes encoding these glycosyltransferases, and methods for biosynthesizing oligosaccharides using such glycosyltransferases. The glycosyltransferases of the present invention can be used for in vitro biosynthesis of various oligosaccharides, such as lacto-N-neotetraose, which are the core oligosaccharides of human blood group antigens. Cloning and expression of the glycosyltransferases of the present invention can be achieved using the standard methods described herein. Such glycosyltransferases are useful for in vitro oligosaccharide biosynthesis, or instead genes encoding such glycosyltransferases transfect cells, such as yeast cells or eukaryotic cells. And can provide another glycosylation of proteins and lipids.

The present invention is based, in part, on the discovery and cloning of loci associated with LOS biosynthesis of Neisseria gonorrhoeae derived from N. gonorrhoeae strain F62. This locus has 5 open reading frames. The first and second reading frames are homologous, respectively, but the fourth and fifth reading frames are not the same. Arranged in between is an additional reading frame that is less homologous to the E. coli rfaI and rfaJ genes and both glycosyltransferases are associated with LPS center (core) biosynthesis. The second and fifth reading frames show strong homology with the Haemophilus influenza lex-1 or lic2A gene, but this gene does not contain the CAAT repeat sequence. Deletions of these five genes, gene combinations, and the entire locus are constructed and transformed into the parent Neisseria gonorrhoeae strain F62. This LOS phenotype is then analyzed by SDS-PAGE and reactivity with monoclonal antibodies. Analysis of this gonococcal mutant revealed that these four genes were found on the inner central region substrate Glc β1-4Hep-R, GalNAcβ1-3Galβ1-4GlcNAc. It is shown to be a glycosyltransferase added with β1-3Galβ1-4. A gene homologous to E. coli rfal / rfaJ is associated with the addition of α-linked galactose residues in the biosynthesis of another LOS structure Gal α1-4Galβ1-4Gacβ1-4Hep-R. These genes encode LOS glycosyltransferases and are therefore named lgtA, lgtB, lgtC, lgtD and lgtE. Analysis of the DNA sequence revealed that lgtA, lgtC, and lgtD had a poly-G region, and in the case of strain F62, they were 17, 10 and 11 bp, respectively. Therefore, these three LOS biosynthetic enzymes are prone to immature termination and potential due to changes in the reading frame. Perhaps these structural properties are responsible for the high frequency genetic variation of Neisseria gonorrhoeae LOS.

The abbreviations used throughout the specification are listed below: lipopolysaccharide, LPS; lipooligosaccharide, LOS; N-acetyl-citidine monophosphate, CMP-NANA; wild type, wt; Gal, galactose; Glc, Glucose; NAc, N-acetyl (eg GalNAc or GlcNAc). In the present invention, molecular biology, microbiology, and recombinant DNA techniques common in the art can be used. Such techniques are well documented in the literature. For example, Sambrook, Fritzch and Maniatis's Molecular Cloning: Experimental Manual (Second Edition (1989), Cold Spring Harbor Laboratory Publishing, Cold Spring Harbor, NY) (hereinafter referred to as Sambrook et al. (1989)); DNA Cloning: Practical Methods "(Volumes I and II) (DNGlover, 1985);" Oligonucleotide Synthesis "(MJGait, 1984);" Nucleic Acid Hybridization "[BD Hames and SJ Higgins Ed. (1985)]; "Transcribing and translation" [BD Hames and SJ Higgins ed. (1984)]; "Animal cell culture" [RI Freshney ed. (1986)]; "Immobilized cells and enzymes" [IRL press (1986)]; See B. Perbel, Practical Guide to Molecular Cloning (1984).

Therefore, the following terms in the specification of the present application have the definitions described below. Cells are "transformed" by such DNA when introduced into the exogenous or heterologous DNA; the cell expresses one or more genes encoded by such DNA. can do. This transformed DNA may or may not be incorporated into the chromosomal DNA constituting the cell genome (covalent bond), or may be contained in an autonomously proliferating replicon. For example, in prokaryotic, yeast and mammalian cells, transformed DNA can be retained in episomal elements such as plasmids. A "clone" is a cell population derived from a single cell or a common ancestor by mitosis.

A "nucleic acid molecule" is either a single-stranded or double-stranded helix, ribonucleoside (adenosine, guanosine, uridine or cytidine; "RNA molecule") or deoxyribonucleoside (deoxyadenosin, deoxyguanosine, deoxythymidine or deoxy). Thymidine; means DNA molecule). Double-stranded DNA-DNA, DNA-RNA and RNA-RNA helices are possible. The term nucleic acid molecule, particularly DNA or RNA molecule, means only the primary or secondary structure of the molecule and does not limit any particular tertiary structure. Thus, the term specifically includes double-stranded DNA found in linear or circular DNA molecules (eg, restriction fragments), viruses, plasmids, and chromosomes. For the structure of a particular double-stranded DNA molecule, the sequences are described herein according to the usual practice of producing only sequences in the 5'to 3'direction along the non-transcriptional strand of DNA (ie). , This strand has a sequence homologous to mRNA). A recombinant DNA molecule is a DNA molecule that has undergone molecular biological manipulation.

When a single strand derived from a nucleic acid molecule can be annealed under conditions of suitable temperature and ionic strength of the solution, the nucleic acid molecule can be associated with other nucleic acid molecules such as cDNA, genomic DNA or RNA. It is hybridizable (see Sambrook et al. (1989) above). Temperature and ionic strength conditions determine the "stringency" of hybridization. Hybridization requires that the two nucleic acids have complementary sequences, although hybridization stringency can result in base-to-base mismatches. The reasonable stringency of the hybridized nucleic acid depends on the length of the nucleic acid, the degree of complementarity, and variables well known in the art. As the degree of similarity or homology between the two nucleic acid sequences increases, the Tm value of the hybrid of nucleic acids having these sequences increases. The relative stability of nucleic acid hybridization (corresponding to a relatively high Tm) decreases in the following order: RNA: RNA, DNA: RNA, DNA: DNA. Nucleotide length 100 For the above hybrids, the formula for Tm has been obtained (Sambrook et al., 9.50-9.51 above). For hybridization with shorter nucleic acids, ie oligonucleotides, the location of the mismatch becomes even more important, and the length of this oligonucleotide determines its specificity (Sambrook et al., 11.7-11.8, supra). The shortest length of hybridizable nucleic acid is preferably at least about 10 nucleotides; more preferably at least about 15 nucleotides; most preferably at least about 20 nucleotides.

A "coding sequence" of DNA is a double-stranded DNA sequence transcribed and translated into a polypeptide in vivo, under the control of a suitable regulatory sequence. The boundary region of this coding sequence is determined by the starting codon at the 5'(amino) terminal and the translation stop codon at the 3'(carboxyl) terminal. The coding sequence includes, but is not limited to, prokaryotic sequences, eukaryotic mRNA-derived cDNAs, eukaryotic (eg, mammal) DNA-derived genomic DNA sequences, and even synthetic DNA sequences. Absent. If intended for eukaryotic expression of this coding sequence, the polyadenylation signal and transcription termination sequence would normally be located at 3'of the coding sequence.

Transcriptional and translational regulatory sequences are DNA regulatory sequences such as promoters, enhancers, terminators, etc. that are used for the expression of coding sequences in host cells. Either lgtA, lgtB or lgtC, even though the individual genes encoding the glycosyltransferases of the invention are found in a single locus with very short non-coding sequences between them. The phase mutation that results in the deletion of the gene does not prevent the restart of transcription of the gene downstream of it. Thus, the loci provided herein include transcription of the transcription initiation sequence of Neisseria. On the other hand, the coding sequences of the present invention can be manipulated for expression under the control of heterologous regulatory sequences. The "promoter sequence" is a DNA that can bind RNA polymerase in the cell and initiate transcription of the downstream (3'direction) coding sequence. It is an adjustment area. For the purposes of clarifying the present invention, this promoter sequence binds to the transcription initiation site at the 3'end thereof and contains at least the bases or elements required to initiate transcription at a level detectable in the background described above. It extends upstream (5'direction) to include the number. Within this promoter sequence will be a transcription initiation site (conveniently limited by mapping with nuclease S1, for example), as well as a protein binding domain (consensus sequence) that contributes to RNA polymerase binding. .. Eukaryotic promoters often include, but are not always, "TATA" and "CAT" boxes.

The coding sequence is "under control" of the cell transcription and translation control sequence if RNA polymerase transcribes this coding sequence into mRNA and then translates it into the protein encoded by this coding sequence. .. The "signal sequence" can be included before the code sequence described above. This sequence encodes the N-terminal signal peptide of the polypeptide, which causes the host cell to transfer the polypeptide to the cell surface or intracellular organella, or to secrete the polypeptide into its medium. Instructed and the signal peptide is usually selectively cleaved by a protein transport machinery. Signal sequences can be found in relation to various proteins endemic to prokaryotic and eukaryotic cells. Integration of the signal sequence is desired for high levels of expression of the glycosyltransferases of the invention by bacteria, yeast, insect cells (baculovirus), or eukaryotic cells and avoids the effects of exogenous glycosylation in host cells. Can be done.

Molecules are "antigenic" when they are capable of specifically interacting with antigen-recognizing molecules of the immune system, such as immunoglobulins (antibodies) or T cell antigen receptors. As mentioned earlier, the carbohydrate (oligosaccharide) portion of Neisseria's LOS is an important antigenic determinant, which determines the serotype of N. meningitidis (Zollinger and Mandrell, Infect. Immun., 18). : 424 (1977); Zollinger and Mandrell, Infect. Immun., 28: 451 (1980)). The antigen portion of a molecule can be the immunodominant portion to the antibody, or is the moiety used to generate an antibody against the molecule by combining the antigen moiety to the immunosensitizing carrier molecule. Can be done. Molecules that are antigenic need not be immunogenic in their own right, i.e., they can elicit an immune response without a carrier.

A composition containing "A" (where "A" is a single protein, DNA molecule, vector, etc.) is one of the species to which the protein, DNA, vector (A and B belongs) in the composition. Substantially "B" (where "B" is one or more contaminating proteins, DNA molecules, vectors, etc.) when at least about 75% by weight of (depending on the category) is "A". ) Is not included. A is preferably contained in at least about 90% by weight of the A + B species of the composition, most preferably at least about 99% by weight. Furthermore, the composition, which is substantially free of contaminants, preferably contains only a single molecular weight species having the activity or properties of the species of interest.

The phrase "medically acceptable" means that the molecular entity and composition are physiologically acceptable and are allergic or similar, such as stomach upset, dizziness, etc., typically when administered to humans. It means that it does not cause adverse reactions. Preferably, the term "pharmaceutically acceptable" as used herein is either approved by a US federal regulatory authority or state government, or is generally used by the United States Pharmacopeia or other animals, especially humans. It means that it was listed in the approved drug collection. The term "carrier" means a diluent, adjuvant, excipient, or vehicle administered with the aforementioned compounds. Such pharmaceutical carriers are sterile liquids such as water and oils, which are oils of petroleum, animal, plant or synthetic origin, such as peanut oil, soybean oil, mineral oil, sesame oil. be able to. Water or saline, and aqueous dextrose and glycerol solutions are preferably used as carriers, especially as injectable solutions. The pharmaceutically acceptable compositions of the present invention do not contain Lipid A, which has a reactive effect on mammalian subjects, especially human subjects.

The term "adjudicant" means a compound or mixture that enhances the immune response to an antigen. Some adjuvants serve as tissue deposits that slowly release antigens and as lymphoid activators that nonspecifically enhance the immune response (Hood et al., Immunology, 2nd Edition, 1981, pp. 384, Benjamin). / Cummings, Menlo Park, CA). Primary challenges with antigen alone without an adjuvant often fail to elicit a humoral or cell-mediated immune response. The adjuvants are complete Freund's adjuvant, incomplete Freund's adjuvant, saponin, inorganic gels such as aluminum hydroxide, surfactants such as lysolecithin, pluronic polyols, polyanions, peptides, oil or hydrocarbon emulsifiers, Keyhole Limpet Hemocyanin, Dinitrophenol, and BCG (Basil Carmet-Gelan) and Corynebacterium Contains, but is not limited to, human adjuvants such as parvum) that may be useful. This adjuvant is preferably pharmaceutically acceptable.

<u style="single">Isolation of glycosyltransferase gene</u> The present invention provides a full-length coding sequence for the LOS locus of Neisseria, and thus any one of the genes encoding the glycosyltransferase properties of that locus, referred to herein as lgt. Or get all 5 species. Any Neisseria bacterial cell can be potentially utilized as a nucleic acid source for molecular cloning of the lgt gene. In the following specific embodiments, these genes are isolated from Neisseria gonoloae. This DNA is a genome obtained from a desired cell, from cloned DNA (eg, a DNA library), by standard methods known in the art, by chemical synthesis, by cloning of cDNA, or by purification. It can be obtained by cloning DNA or fragments thereof (Sambrook et al., Supra, (1989); Glover, DM (eds.), 1985, DNA Cloning: Practical Methods, Volumes I, II, MRL Press, Inc. , Oxford, UK). For example, N. Goronoea's genomic DNA is Sau3A to create a phage genomic library. It is digested with a restriction endonuclease or endonuclease such as BamHI / EcoRI and inserted into a phage vector digested with a restriction endonuclease or endonuclease such as BamHI / EcoRI. Whichever is the source, the gene must be cloned as a molecule into a vector suitable for transmission of the gene.

In molecular cloning of genes derived from genomic DNA, DNA fragments are generated, some of which are thought to encode the desired gene. This DNA is cleaved at specific sites using various restriction enzymes. Alternatively, DNAse can be used in the presence of manganese to make DNA fragments, or the DNA can be physically sheared, as in sonication. The linear DNA fragment can then be separated according to size by standard methods, including but not limited to gel electrophoresis of agarose and polyacrylamide, and column chromatography.

Once a DNA fragment is generated, identification of a particular DNA fragment containing the desired lgt gene can be achieved in many ways. For example, the generated DNA fragments can be screened by hybridization of nucleic acids to labeled probes synthesized with the sequences described herein (Benton and Davis, Science, Science, 196: 180). 1977); Grunstein and Hogness, Proc. Natl. Acad. Sci. USA, 72: 3961 (1975)). These DNA fragments, which are substantially homologous to the probe, will hybridize. The present invention provides specific examples of DNA fragments that can be used as hybridization probes for glycosyltransferases, such as SEQ ID NO: 1.

As mentioned above, the presence of this gene can be detected by an assay based on the physical, chemical or immunological properties of the expressed product. For example, electrophoretic mobility, isoelectric focusing behavior, proteolytic digestion map, proteolytic activity, or functional properties, especially glycosyltransferase activity, similar or similar ability of the Lgt protein to mediate transglycosylation to receptor molecules. It is a DNA clone that produces the same protein. Instead, the putative lgt gene can be mutated and its role as a glycosyltransferase is established by detecting various structures of LOS oligosaccharides.

Other methods of isolating lgt genomic DNA include, but are not limited to, the chemical synthesis of the known sequence encoding Lgt, such as the SEQ ID No: 1 gene sequence itself. In another embodiment, the DNA of the lgt gene can be isolated by PCR using oligonucleotide primers determined from the nucleotide sequences described herein. Other methods are also possible and are within the scope of the present invention.

The next identified and isolated gene is then inserted into a suitable cloning vector. A number of vector-host systems known in the art can be used. Possible vectors include, but are not limited to, plasmids or modified viruses, the vector system must be compatible with the host cell used. In certain aspects of the invention, the lgt coding sequence is inserted into an E. coli cloning vector. Other examples of vectors include, but are not limited to, bacteriophages such as λ derivatives, pBR322 derivatives or plasmids such as pUC plasmid derivatives, such as pGEX vectors, pmal-c, pFLAG and the like. is there. Insertion into a cloning vector is achieved, for example, by ligation of the DNA fragment into a cloning vector with complementary attachment ends. However, if this complementary restriction site used for DNA fragmentation is not present in the clonal vector, the ends of the DNA molecule can be enzymatically modified. Instead, place any desired site in the DNA. It can be produced by ligating a nucleotide sequence (linker) at the end; these ligated linkers contain a specific chemically synthesized oligonucleotide that encodes a restriction endonuclease recognition sequence. be able to. In a specific embodiment, PCR primers having such a linker site can be used to amplify DNA for cloning. Recombinant molecules can be introduced into host cells by transformation, transfection, infection, electroporation, etc., resulting in the production of many copies of this gene sequence.

Transformation of a host cell with an isolated lgt gene or a recombinant DNA molecule incorporating a synthesized DNA sequence allows the production of multiple copies of the gene. Therefore, this gene can be obtained by growing the transformant, isolating the recombinant DNA molecule from the transformant, and if necessary, by recovering the inserted gene from the isolated recombinant DNA. , Can be obtained in large quantities. The present invention further relates to vectors containing the enzyme (fragment) and a gene encoding a truncated form of a derivative of Lgt's having the same functional activity as Lgt. The production and use of fragments and derivatives associated with Lgt is within the scope of the present invention. In a specific embodiment, the fragment or derivative is functionally active, i.e. can mediate glycosyl transfer to the accepting molecule.

Cleaved fragments of the glycosyltransferase can be prepared by removing the N-terminal, C-terminal, or internal regions of proteins that are not needed for functional activity. Usually, such removed portions contain only a few, eg, 1-5 amino acid residues, but larger fragments can be removed. Chimeric molecules that contain all or functionally active moieties of the glycosyltransferases of the invention that are bound to other proteins, such as fusion proteins, are also considered. The glycosyltransferase fusion protein comprises at least the functionally active portion of the non-glycosyltransferase protein bound via a peptide bound to the functionally active portion of the glycosyltransferase polypeptide. This non-glycosyltransferase sequence can be the amino- or carboxyl-terminal of the glycosyltransferase sequence. Expression of the fusion protein can result in an enzymatically inert glycosyltransferase fusion protein. Recombinant DNA encoding such a fusion protein The molecule comprises a sequence encoding at least the functionally active portion of a non-glycosyltransferase protein bound in-frame of the glycosyltransferase coding sequence, and preferably a particular protease, such as thrombin or factor Xa. It encodes a cleavage site, preferably a glycosyltransferase-non-glycosyltransferase junction. In a specific embodiment, the fusion protein is capable of expressing Escherichia coli.

Specifically, Lgt derivatives can be made by modifying the encoding nucleic acid sequence by substitutions, additions or deletions that result in functionally equivalent molecules. Due to the degeneracy of the nucleotide coding sequence, another DNA sequence encoding substantially the same amino acid sequence as the lgt gene can be used in the practice of the present invention. These include nucleotide sequences containing all or part of the lgt gene, which encode the same amino acid residue in the sequence, thus causing a silent change, modified by the substitution of another codon. , Not limited to this. Similarly, the Lgt derivatives of the present invention contain, as the primary amino acid sequence, a modified sequence in which a functionally equivalent amino acid residue is replaced with a residue in a sequence that results in a conservative amino acid substitution. Includes, but is not limited to, those having all or part of the amino acid sequence of.

For example, one or more amino acid residues in the sequence can be replaced with another amino acid of similar polarity, which acts functionally equivalently, resulting in a silent change. Substitutes for the amino acid in the sequence can be selected from other members of the class to which this amino acid belongs. For example, non-polar (hydrophobic) amino acids are alanine, leucine, isoleucine, valine, proline, phenylalanine, tryptophan, and methionine. Polar neutral amino acids are glycine, serine, threonine, cysteine, tyrosine, asparagine, and glutamine. The positively charged (basic) amino acids are arginine, lysine and histidine. The negatively charged (acidic) amino acids are aspartic acid and glutamic acid.

The genes encoding the Lgt derivatives and analogs of the present invention can be generated by various methods known in the art (eg, Sambrook et al. (1989), supra). This sequence is cleaved at the appropriate site with restriction endonucleases and then further enzymatically modified, isolated and ligated in vitro, if desired. In the generation of a gene encoding an Lgt derivative or analog, the same translation reading frame as the lgt gene, where the modified gene is not disrupted by a translational arrest signal in the gene region where the desired activity is encoded. Care must be taken to ensure that it remains in the.

In addition, lgt nucleic acid sequences can be mutated in vitro or in vivo to form and / or disrupt translation, initiation and / or termination sequences, or to form coding regions and / or novel restriction endos. The existing sequence is disrupted to form a nuclease site or to facilitate further in vitro modification. Any method for mutations known in the art can be used, which is a site-specific mutation method in vitro (Hutchinson, C. et al., J. Biol. Chem., 253: 6551 (1978); Zoller and Smith's paper, DNA, 3: 479-488 (1984); Oliphant et al., Gene, 44: 177 (1986); Hutchinson et al., Proc. Natl. Acad. Sci. USA , 83: 710 (1986)), including, but not limited to, the use of TAB® linkers (Pharmacia). PCR is preferred for site-specific mutations (Higuchi's paper, 1989, DNA with PCR Manipulation , PCR Techniques: Principles and Applications of DNA Proliferation, H. Erlich (eds.) Stockton Press, Chapter 6, pp. 61-70). The lgtA, lgtB and lgtC genes are particularly susceptible to phase mutation mutations, It should be noted that it contains long poly-G elongations.

<u style="single">Expression of glycosyltransferase</u> The gene encoding Lgt, or a functionally active fragment or other derivative thereof, may be inserted into a suitable expression vector, i.e. a vector containing the elements required for transcription and translation of the coding sequence of the inserted protein. it can. The expression vector also contains a replication origin. The required transcription and translation signals can also be supplied by the native lgt gene and / or its flanking region. Various host-vector systems are capable of expressing the coding sequences of proteins. However, it is preferable to use a bacterial expression system to provide high levels of expression of the protein, which is more likely to be a natural conformation. Possible host-vector systems are mammalian cell lines infected with a virus (eg, vaccinia virus, adenovirus, etc.); insect cell line, infected with a virus (eg, baculovirus); microorganisms such as yeast, including yeast vectors, Alternatively, it includes, but is not limited to, bacteria transformed with bacteriophage, DNA, plasmid DNA, or cosmid DNA. The expression elements of the vector differ in their intensity and specificity. Host used- Depending on the vector system, any one of many suitable transcription and translation elements can be used.

It is preferred that the peripheral type of Lgt (including the signal sequence) results in transport of the protein to the peripherals of Escherichia coli or to a Bacillus subtilus-based expression system. Any of the aforementioned methods for inserting a DNA fragment into a vector can be used to construct an expression vector having a chimeric gene consisting of a suitable transcription / translation control signal and the protein coding sequence. These methods can include recombinant DNA and synthetic methods in vitro, as well as recombinants (genetic recombination) in vivo.

Expression of the nucleic acid sequence encoding the glycosyl transferase or peptide fragment can be regulated by a second nucleic acid sequence so that these glycosyl transferases or peptides are transformed in the recombinant DNA molecule in the host. , Expressed. For example, expression of glycosyltransferases can be regulated by any of the promoter / enhancer elements known in the art, but these regulatory elements must function in the host selected for expression. .. For expression in bacteria, a bacterial promoter is required. A eukaryotic virus, or a eukaryotic promoter containing a tissue-specific promoter, is used when a vector containing the lgt gene is injected directly into the subject for transient expression, as described in detail below. Preferably, it provides heterologous protection against bacterial infections. Promoters that can be used to control the expression of the lgt gene include, but are not limited to: the SV40 early promoter region (Benoist). And Charnbon, Nature, 290: 304-310 (1981), Promoters Included in 3'Long End Repeats of Raus Sarcoma Virus (Yamamoto et al., Cell, 22: 787-797 (1980)), Herpes Timidin kinase promoter (Wagner et al., Proc. Natl.Acad.Sci.USA, 78: 1441-1445 (1981)), regulatory sequence of metallotionine gene (Brinster et al., Nature, 296: 39-42 (1982)) ); Eukaryotic expression vectors such as the β-lactamase promoter (Villa-Kamaroff et al., Proc. Natl. Acad. Sci. USA, 75: 3727-3731 (1978)), or the tac promoter (De Boer et al., Paper, Proc.Natl.Acad.Sci.USA, 80: 21-25 (1983)); and Useful Proteins Derived from Recombinant Bacteria (Scientific American, 242: 74-94 (1980)), etc. reference.

Expression vectors containing the lgt gene insertion fragment are identified by four common methods: (a) PCR amplification of the desired plasmid DNA or specific mRNA, (b) nucleic acid hybridization, (c) marker. The presence or absence of gene function and (d) expression of the inserted sequence. In the first method, the nucleic acid can be amplified by PCR along with integration of radioactive nucleotides or staining with ethidium bromide for detection of amplified products. In the second method, the presence of a foreign gene in the expression vector can be detected by nucleic acid hybridization using a probe containing a sequence homologous to the inserted lgt gene. In a third method, the recombinant vector / host system is subjected to specific "marker" gene functions (eg, β-galactosidase activity, PhoA activity, thymidine kinase activity, antibiotic resistance, etc., resulting from the insertion of foreign genes in the vector. It can be identified and selected based on the presence or absence of a transforming phenotype (formation of inclusions in baculovirus). If the lgt gene is inserted into the marker gene sequence of the vector, lgt A recombinant having an inserted fragment can be confirmed by not exerting its marker gene function. In the fourth method, the recombinant expression vector can be identified by assaying for the activity of the lgt gene product expressed in the recombinant. Such assays can be based, for example, on the physiological or functional properties of the lgt gene product, such as glycosyltransferase activity, in an in vitro assay system. Once suitable host systems and growth conditions have been established, recombinant expression vectors are grown and prepared in large quantities.

<u style="single">Oligosaccharide biosynthesis</u> The glycosyltransferase of the present invention can be used for the biosynthesis of oligosaccharides. The glycosyltransferase of the present invention is capable of stereospecifically complexing a specifically activated saccharide unit with a specific accepting molecule. Such activated saccharides generally consist of uridine, guanosine, and cytidine diphosphate derivatives of the saccharides, which provide nucleoside diphosphate as a free radical. Therefore, these activated saccharides are saccharide-UDP, saccharide-GDP or saccharide-CDP. In a specific embodiment, the activated saccharides are UDP-GlcNAC, UDP-GalNAc or UDP-Gal.

As used herein, the term "receptive molecule" means a molecule by which a glycosyltransferase transfers an activated sugar. As is well known in the art, carbohydrate synthesis proceeds by the continuous coupling of sugar residues to lipids such as, for example, dolicole phosphate. In eukaryotic cells that glycosylate proteins, oligosaccharides or polysaccharides are transferred from the activated lipid carrier to the luminal polypeptide of the endoplasmic reticulum. In prokaryotic cells, carbohydrates are synthesized directly on the lipid A molecule. Presumably, the glycosyltransferases of the present invention react with the central portion of growing carbohydrate and lipid molecules. Thus, in a preferred embodiment, the accepting molecule or carrier comprises a lipid, preferably a polyisoprenoid alcohol lipid such as dolicole phosphate. Maximum synthetic efficiency is provided by using Lipid A as a carrier. Lipid A is a carrier for direct administration of the oligosaccharides produced in the subject, for example in vaccine preparations. Although not useful, this is suitable when using labile linkage for continuous cleavage (under mild conditions) and separation of oligosaccharides from lipid carriers. .. It should be further noted that the glycosyltransferase only effectively acts on the addition of specifically activated saccharides to saccharide residues on the accepting molecule corresponding to the naturally occurring accepting molecule. Is. For example, LgtE catalyzes the transition from Gal to Glc β1 4Hep. Thus, when a glycosyltransferase mediates the attachment of GalNAc to Glc, the nature of the Glc residue (eg, whether directly or indirectly attached to the carrier) is a reaction. It will affect efficiency. Effective synthesis is unlikely to occur in the absence of carriers or in the absence of non-lipid carriers. However, even if inefficient synthesis is desired, the practice of the present invention is not limited to the use of receptor molecules, including lipids, but extends to saccharides, polysaccharides, polypeptides, glycoproteins and the like.

For the synthesis of oligosaccharides, glycosyltransferases are contacted with the appropriate activated saccharides and the appropriate acceptor molecules under conditions effective for transfer and covalent attachment of the saccharide to the acceptor. Suitable and best time, temperature and pH conditions for the transfer of a particular saccharide unit can be determined through routine tests; in general, physiological conditions will be accepted. Certain co-reagents may be desired; for example, they more efficiently contact glycosyltransferases with activated saccharides and accepting molecules in the presence of divalent cations. I can let you.

In the present invention, the glycosyltransferase enzyme is covalently or non-covalently immobilized on a solid support, such as sephadex, cepharose, or a poly (acrylamide-co-N-acrylic oxysucciimide) (PAN) resin. can do. A particular reaction can proceed in an isolated reaction solution with easy separation of the solid phase enzyme from the reaction product. Enzyme immobilization involves a specific glycosyltransferase attached to a solid support, with supports randomly placed in the column in a certain order or in adjacent regions, and a passage of reaction solution through the column. Also consider continuous biosynthetic flows, such as with elution of the desired oligosaccharide at the ends. Efficient methods of attachment of glycosyltransferases to solid supports and the use of such immobilized glycosyltransferases are all cited herein by reference, Roth, 19 January 1993. It is described in US Pat. No. 5,180,674.

Oligosaccharides prepared using the glycosyltransferases of the present invention, such as disaccharides, can be used as further synthetic accepting molecules regardless of whether the glycosyltransferases of the present invention or glycosyltransferases known in the art are used. Available (eg, Roth's US Pat. No. 5,180,674 and Roth's International Patent Publication No. WO 93/13198, published July 8, 1993, both cited herein as references. ing.). The oligosaccharide compositions of the present invention are widely useful in therapeutic and diagnostic applications. For example, this saccharide composition can be useful as a protective agent for cell surface receptors in the treatment of numerous diseases associated with cell adhesion. To mention a little more, saccharide compositions useful as dietary supplements, antibacterial agents, anti-metastatic agents, anti-inflammatory agents (eg, binding to inflammation-related lecithin or cell surface receptors) are expected by the present invention. As mentioned above, the glycosyltransferases of the present invention can be used with other glycosyltransferases known in the art, or are recognized as synthetic complex oligosaccharides or polysaccharides.

Furthermore, the glycosyltransferases of the present invention can be used for the synthesis of oligosaccharide representatives of oligosaccharides found in various strains of Neisseria. For example, the structure of another oligosaccharide can be prepared by deletion of the oven reading frame from the locus, or by the selection of only a few glycosyltransferases of the invention for synthesis. They can be used in vaccine preparations that are effective against Neisseria variants, especially in subunit vaccines against Neisseria gonorrhoeae or Neisseria gonorrhoeae. Furthermore, the glycosyltransferases of the present invention can be used in the preparation of oligosaccharides corresponding to oligosaccharides associated with human glycolipids. Therefore, in a specific embodiment, the present invention describes the sphingolipid paragloboside lacto-N-. It provides the synthesis of an oligosaccharide corresponding to neotetraose; an oligosaccharide that mimics a ganglioside; and a mimicry of the saccharide moiety of a globoglycolipid whose structure is characteristic of Neisseria meningitidis immunotype L1. The oligosaccharides of the present invention correspond to the oligosaccharide centers of blood group antigens, and therefore the preparation of such blood group antigens is very useful for diagnostic or therapeutic purposes.

Therefore, the preparation of structural GalNAc β1 3Galβ1 4GlcNAc β1 3Galβ1 4Glc (ie, ganglioside) involves the following sequence of steps: a. , A glycosyltransferase having the amino acid sequence SEQ ID NO: 6, or a step of contacting in the presence of a functionally active fragment thereof; b. A reaction mixture containing activated GlcNAc, a receiving moiety containing a Gal β1 4 Glc residue. In the presence of a glycosyltransferase having the amino acid sequence SEQ ID NO: 3, or a functionally active fragment thereof, a step of contacting; c. A reaction mixture containing activated Gal contains GlcNAcβ1 3Galβ1 4Glc residues. Contacting the accepting moiety in the presence of a glycosyltransferase with the amino acid sequence SEQ ID NO: 8 or a functionally active fragment thereof; d. A reaction mixture containing activated GalNAc, Gal β1 4GlcNAc β1 3Gal β1 Amino acid sequence SEQ ID NO: 5 at the receiving moiety containing 4Glc residue The step of contacting in the presence of a glycosyltransferase having the above, or a functionally active fragment thereof.

Similarly, the preparation of structural Gal β1 4GlcNAc β1 3Gal β1 4Glc (ie, lacto-N-neoteotranose) involves the following sequence of steps: a. Glc residue of the reaction mixture containing activated Gal. A step of contacting a group-containing accepting moiety in the presence of a glycosyltransferase having the amino acid sequence SEQ ID NO: 6 or a functionally active fragment thereof; b. A reaction mixture containing activated GlcNAc, Gal β1 4Glc The step of contacting the accepting moiety containing the residue in the presence of a glycosyltransferase having the amino acid sequence SEQ ID NO: 3 or a functionally active fragment thereof; c. GlcNAcβ1 3Galβ1 of the reaction mixture containing activated Gal. This is a step of contacting the receiving moiety containing the 4Glc residue in the presence of a glycosyltransferase having the amino acid sequence SEQ ID NO: 8 or a functionally active fragment thereof.

In another embodiment, the preparation of structural Gal α1 4Gal β1 4Glc (ie, globoglycolipid) comprises the following sequence of steps: a. Receiving a reaction mixture containing activated Gal with a Glc residue. The step of contacting in the presence of a glycosyltransferase having the amino acid sequence SEQ ID NO: 6 or a functionally active fragment thereof; b. A reaction mixture containing activated Gal containing a Gal β1 4 Glc residue. A step of contacting the receiving moiety in the presence of a glycosyltransferase having the amino acid sequence SEQ ID NO: 4, or a functionally active fragment thereof. Such oligosaccharides can be prepared using Lipid A as a carrier. When the obtained glycolipid is used as a vaccine, it is preferable to use a non-toxic lipid as a carrier, for example, dolicole phosphate.

<u style="single">Vaccination</u> Active immunity to the Niseria strain can be induced by immunosensitization (vaccination) with an oligosaccharide of an immunological amount prepared according to the present invention mixed with an adjuvant, in which case the oligosaccharide is the vaccine. It becomes an antigen component of. This oligosaccharide is preferably complexed with a carrier protein. Alternatively, if the antigen is a glycolipid, it can be incorporated into liposomes. Although this oligosaccharide on Lipid A is toxic and the vaccination-induced active immunity of the present invention can result in an immediate immune response, this oligosaccharide alone causes bacterial infection. It is not possible.

The choice of adjuvant depends on the subject to be vaccinated. Preferably, a pharmaceutically acceptable adjuvant is used. For example, for human vaccines, avoid oil or hydrocarbon emulsifying adjuvants, including Freund's complete and incomplete adjuvants. One example of an adjuvant suitable for human use is alum (alumina gel). However, vaccines for animals can include adjuvants that are not suitable for human use. The vaccine of the present invention, that is, a vaccine containing an oligosaccharide corresponding to the antigenic determinant of a Neisseria strain, is administered by a parenteral route including, but not limited to, intramuscular, abdominal cavity, and venous administration. be able to. Administration of a sufficient amount of Neisseria oligosaccharide to inhibit the adhesion of Neisseria to its target cells is also effective in treating meningococcal or Neisseria gonorrhoeae infections. This requirement can be determined by one of the common techniques using standard methods.

<u style="single">Expression of glycosyltransferases for intracellular glycosylation</u> The present invention further contemplates transformation of a host cell carrying the glycosyltransferase of the present invention. Where possible, expression of glycosyltransferases in cells lacking one or more endogenous glycosyltransferases results in the novel glycosylation of lipids and proteins in such eukaryotic cells and the novel glycosylation of lipids in prokaryotic cells. May occur. For example, transformation of bacteria with non-toxic lipid molecules causes Neisseria oligosaccharides to be expressed on such bacteria, which can then be used directly in whole cell vaccines. Separately, expression of glycosyltransferases in yeast, insect or mammalian cell lines can result in novel glycosylation of lipids and proteins expressed by these cells.

<u style="single">Antibodies to Neisseria oligosaccharides and their diagnosis and treatment</u> Just as this oligosaccharide can be used in vaccines, this oligosaccharide can be used to produce antibodies against it. This antibody can then be used for testing specific strains of bacteria and for passive immunity. Antibodies include, but are not limited to, polyclonal antibodies, monoclonal antibodies, chimeric antibodies, single chain antibodies, Fab fragments, and Fab expression libraries. Various methods known in the art can be used to produce polyclonal antibodies against oligosaccharides. Antibodies can be produced by immunizing various host animals by injection with oligosaccharides. Animals include, but are not limited to, rabbits, mice, rats, sheep, goats, and the like.

In one embodiment, oligosaccharides may be bound to immunogenic carriers such as bovine serum albumin (BSA), keyhole limpet hemocyanin (KLH) and the like. Various adjuvants can be used to increase the immune response, depending on the host animal species. Various methods of providing antibody molecules from continuous cell lines in culture can be used to prepare monoclonal antibodies against oligosaccharides, fragments thereof, analogs, or derivatives. These methods include the hybridoma technology first developed by Koehler and Milstein (1975, Nature 256: 495-497), the trioma technology, and the human B cell hybridoma technology (Kozbor et al., 1983, Immunology Today 4:72). , And EBV hybridoma technology for producing human monoclonal antibodies (Cole et al., Monoclonal Antibodies and Cancer Therapy, Alan R. Liss., Pp. 77-96), but is not limited thereto.

In an additional aspect of the invention, modern techniques can be utilized to produce monoclonal antibodies in germ-free animals (PCT / US90 / 02545). According to the present invention, a human antibody can be used, which human antibody can be used using a human hybridoma (Cote et al., 1983, Proc. Natl. Acad. Sci. USA 80: 2026-2030) or in vitro. It can be obtained by transforming human B cells with the EBV virus (Cole et al., 1985, Monoclonal Antibodies and Cancer Therapy, Alan R. Liss., Pp. 77-96). In fact, according to the present invention, a technique developed for the production of "chimeric antibodies" (genes from mouse antibody molecules specific for oligosaccharides, along with genes from human antibody molecules with appropriate biological activity). By splicing (Morrison et al., 1984, J. Bacterol. 159-870; Neuberger et al., 1984, Nature 312: 604-608; Takeda et al., 1985, Nature 314: 452-454)) can be used.

This antibody is within the scope of the present invention. Such human or humanized chimeric antibodies are used in the treatment of human diseases because these antibodies do not induce a much more immune response than exogenous antibodies, in particular an allergic response by themselves. Is preferable. According to the present invention, oligosaccharide-specific single-chain antibodies can be produced by applying the techniques described for the production of single-chain antibodies (US Pat. No. 4,946,778). In yet another aspect of the invention, a monoclonal Fab having the desired specificity for oligosaccharides using the techniques described for the construction of Fab expression libraries (Huse et al., 1989, Science 246: 1275-1281). Fragments, derivatives thereof, or analogs can be identified quickly and easily.

Antibody fragments containing the idiotypes of antibody molecules can be produced by known techniques. For example, such fragments include F (ab') 2 fragments that can be formed by pepsin digestion of antibody molecules, and Fab'fragments that can be formed by reducing the disulfide bridge of F (ab') 2. And Fab fragments that can be formed by treating antibody molecules with papain and reducing agents, but are not limited to these. Known techniques for antibody production include, for example, radioimmunoassay (ELISA) (enzyme-linked immunosorbant assay), "sandwich" immunoassay, immunoradiometric assay, gel diffusion precipitin reaction, immunodiffusion assay. , In situ immunoassay immunoassay) (for example, using colloidal gold, enzyme or radioisotope labeling), Western blot, prescription reaction, aggregation analysis (eg, gel aggregation analysis, hemoaglutination analysis), complementary fixation analysis, immunoassay The desired antibody can be screened by fluorescence analysis, protein A analysis, immunoelectrophoresis analysis and the like.

In one embodiment, antibody binding is detected by detecting the label on the primary antibody. In another aspect, the primary antibody is detected by detecting the binding of the secondary antibody or reagent to the primary antibody. In yet another embodiment, the secondary antibody is labeled. Many techniques are known in the art for the detection of binding in immunoassay and are within the scope of the present invention. For example, in order to select an antibody that recognizes a particular oligosaccharide, hybridomas formed against a substance that binds to an oligosaccharide containing such an epitope can be analyzed. To select an antibody specific for an oligosaccharide from Neisseria of a particular species or strain, it can be selected based on positive binding to the oligosaccharide expressed and isolated by that species or strain. .. The above antibodies are known in the art for the localization and activity of oligosaccharides, such as Western blotting, in situ imaging of oligosaccharides, their concentration in suitable physiological samples. Can be used for measurement.

Diagnosis of Gram-positive bacterial infections can use any desired immunoassay format known in the art. These antibodies can be labeled with labeling substances such as enzymes, fluorescent substances, color formers, radioisotopes, dyes, colloidal gold, latex particles, and chemiluminescent substances for detection in vitro. These antibodies can also be labeled with, for example, radioisotopes (preferably technetium or iodine), magnetic resonance shift reagents (eg, gadolinium and manganese), or radiopaque reagents for detection in vivo.

The nucleic acids of the invention and their sequences can also be used to diagnose Neisseria infections, in particular to identify specific strains or to determine which glycosyltransferase gene is mutated. For example, the lgt gene or a hybridizable fragment thereof can be used for in situ hybridization with a sample from a patient at risk of Neisseria infection. In another embodiment, a PCR amplification method using a probe based on the lgt gene of the present invention can be utilized to identify a specific gene segment of Neisseria. In one aspect of the invention, hybridization with a probe or PCR primer is performed under stringent conditions, or with a sequence specific for a particular strain or a limited number of bacterial strains, or both. This allows the diagnosis of infection by a particular strain (or multiple strains). Hybridization can also be performed under less stringent conditions, or the sequences may match in any or all strains of the bacterium, which allows the diagnosis of such infections. .. The present invention will be better understood from the following exemplary description detailing the configuration and method.

Example This example describes the locus of Neisseria gonorrhoeae strain F62, which has five genes. Four of these genes sequentially add GalNAcβ1 3Galβ1 4GlcNAcβ1 3Galβ1 4 to the substrate Glcβ1 4Hep R in the inner core region (Yamasaki et al., 1991, Biochemistry 30: 10566). The fifth gene is involved in the addition of α-linked galactose residues in the biosynthesis of another LOS structure Galα1 4Galβ1 4Glcβ1 4Hep R (John et al., 1991, J.Biol.Chem.266). : 19303). DNA sequence analysis revealed that the 1st, 3rd and 4th reading frames contained the poly-G region. These are 17, 10 and 11 bp, respectively, in strain F62. Therefore, three of the LOS biosynthetic enzymes are at risk of premature arrest due to changes in the reading frame, as reported for the Neisseria gonorrhoeae pilC gene (Johnson et al., 1991, EMBO J. 10: 477; Rudel et al., 1992, Molec. Microbiol. 6: 3439). These structural features appear to be responsible for the frequent genetic alterations of Neisseria gonorrhoeae LOS (Schneider et al., 1988, Infect. Immun. 56: 942).

Materials and Methods Reagents and Chemicals: Most chemicals were obtained from Sigma Chemicals (St. Louis, MO). Restriction enzymes were purchased from the New England Biolabs (Beverly, MA). Medium and Growth Conditions: E. coli strains were grown on solid or liquid LB medium (Sunbrook et al., 1989, Cold Spring Harbor Laboratory Press, Cold Spring Harbor). Antibiotics were added as appropriate. Carbenicillin was used at 50 μg / ml and erythromycin was used at 200 μg / ml. Neisseria gonoloea strain F62 was grown on GC agar medium (Swanson, 1978, Infect. Immun. 19: 320) or GC agar medium containing 2 μg / ml erythromycin. For isolation of LOS or genomic DNA, gonorrhea 1.5% Proteose peptone broth (Diff collaborations, Detroit MI), 30 mM phosphate, 8.5 mM NaCl, 1% Isovitalex (Becton Dickinson Microbiology Systems, Cockiesville) , MD).

Recombinant DNA method: The plasmid was purified using a Qiagen column or QIA prep spin column obtained from Qiagen (Chatsworth, CA). Digestion with restriction enzymes, gel electrophoresis, ligation with T4 DNA polymerase and transformation of E. coli were performed according to Sambrook et al., 1989, Cold Spring Harbor Laboratory Press, Cold Spring Harbor. Southern hybridization was performed on Highbond N + Membrane Amersham with labeled DNA using the Amersham (Arlington Heights, IL) ECL kit. Genomic DNA was isolated as described by Moxon et al. 1984, J. Clin. Invest. 73: 298.

A gene bank of Neisseria gonorrhoeae strain F62 genomic DNA binds approximately 20 kb fragments obtained by incomplete digestion with Sau3A to BamHI / EcoRI digested λ2001 (Karn et al., Gene 32: 217 (1984)). Constructed by letting. Phage libraries were screened by hybridization with random-primer-labeled plasmid pR10PI, and 5 clones were isolated by plaque purification. Phage from these clones were purified by sedimentation in CsCl and subsequent levitation (Cold Spring Harbor Laboratory, Cold Spring Harbor, NY (1980) by Davis et al.) To isolate DNA. From one of these clones, two ClaI fragments of 4.9 and 3.4 kb were isolated by gel electrophoresis and recovery by Geneclean II (BIO 101 Inc., La Jolla. CA.). These are Stratagene (La ClaI cleavage from Jolla.CA.) was combined with pBluescript II SK and named p4900 and p3400, respectively. The p4900 was subdivided into two clones containing the PstI site in the insert and containing the 2.1 and 2.8 kb inserts. The clone containing the 2.8 kb insert was named pPstCla. Inserts of p3400 and pPstCla are chain termination methods using Sequence II (United States Biochemical Co., Cleveland.OH) (Sanger et al., Proc. Natl. Acad. Sci. USA 74: 5463 (1977). ) To determine the sequence. All sequences shown in FIG. 2 were completed in both directions.

The insertions and deletions shown in FIG. 6 were performed as follows. For I1, I3, Δ1 and Δ2, plasmid pPstCla cleaved with BsaBI, AscI, and StyI and double-cut with StyI and BsaBI was used, respectively. In 12 and Δ3, plasmid p3400 cleaved with AgeI or StyI was used. The complete locus was assembled by cloning the ClaI-ApaI fragment from p3400 into ClaI and pPstCla cleaved with ApaI, and named the plasmid pLOS5. Deletions Δ4 and Δ5 were constructed using pLOS5 by digestion with StyI and BbsI or digestion with StyI alone. In all cases (except digestion with BsaBI), the cleaved plasmid was treated with the Klenow fragment of E. coli DNA polymerase to smooth the ends and insert ermC'(erythromycin resistance marker). The ermC'gene was introduced from the plasmid pIM13 (J. Bacteriol. 169: 5131 (1987) by Projan et al.), ClaI-Hind. It was isolated as a III fragment and cloned to the same site in plasmid pHSS6 (Proc. Natl. Acad. Sci. USA 83: 735 (1986) by Seifert et al.). From this plasmid, it was excised as a Not I fragment, its ends blunted by treatment with the Klenow fragment of DNA polymerase, and purified by gel electrophoresis and recovery with Geneclean II.

Piliated Neisseria Transformation of gonorrhoeae strain F62 was performed on GC agar medium (Swanson's Infect. Immun. 19) containing a plasmid isolated from E. coli (Klugman et al. Infect. Immun. 57: 2066 (1989)) and erythromycin 2 μg / ml. It was performed using the transformant selected in: 320 (1978)). The goodness of fit of the genome exchange of each gonococcal transformant was confirmed by sequencing the upstream and downstream junctions of the ermC'gene in their genomic DNA using PCR techniques. Two biotinylated primers, GCCGAGAAAACTATTGGTGGA (SEQ ID NO: 9) and AAAACATGCAGGAATTGACGAT (SEQ ID NO: 10), were synthesized: each of these was used as the basis for the ermC'sequence near its upstream and downstream ends. Primers were designed so that their 3'ends pointed outward from the ermC'gene. Each of these primers is putative Used with appropriate primers that match the sequence of the LOS locus near insertion). PCR was performed in 25 cycles according to the instructions for GeneAmp PCR Reagent Kit from Perkin Elmer (Branchburg, NJ). In all cases, a product of the expected size was obtained. The DNA sequences of these products are obtained by purifying PCR products with magnetic streptavidin beads from Dynal (Lake Success, NY) and the method developed by Hultman et al. (Hultman et al. Nucleic). Based on Acids Res. 17: 4937 (1989)), it was determined by sequencing using the Sequence II kit according to the Synal protocol. The sequences were analyzed by a computer program in the GCG package of Genetics Computer Group (Madison, WI).

Immunological Methods Monoclonal antibodies 17-1-L1 (L1), 9-2-L378 (L3), 2-1-L8 (L8) were obtained as filtered ascites fluid. Antibodies 1-1-M were obtained as ascites fluid and 3F11 and 4C4 were obtained as the upper layer of tissue culture. LOS was extracted from each N. gonorrhoeae variant by the thermal phenol-water method (Westphal and Jann Academic Press, New York 83-91 (1965)) and Johnston et al. J.Exp.Med.143:741 (1976). )) Purified as described. LOS was diluted to 200 μg / ml in Western blot buffer described by Towbin et al. (Proc. Natl. Acad. Sci. USA 76: 4350 (1979) by Towbin et al.) And blotting 1.5 μl aliquots. Millipore on 3MM Whatman filter paper soaked in buffer Spotted on Immobilon-P membrane from Corp (Bedford, MA). The spots were allowed to absorb into the membrane for 2 minutes and the strips were placed in blocking buffer for at least 60 minutes. The blocking buffer consisted of 150 mM NaCl, 10 mM Tris-HCl, 10 mM, pH 7.5, 5 mM MgCl2, and 3% gelatin dissolved in 0.02% NaN2. The strips were washed 3 times in the same buffer containing 1% gelatin. Strips were treated with monoclonal antibody diluted in blocking buffer for 2 hours. The antibody available as ascites fluid was diluted 1/1000 and the antibody available as the upper layer of tissue culture was diluted 1/10. The strips were washed and incubated with phosphatase-conjugated anti-IgG, IgA, IgM (1/1000 dilution) from Cappel (Organon Teknika Co., West Chester, PA) for 60 minutes as described above (Blake). Et al. Analyt. Blochem. 136: 175 (1984)), washed and stained.

Gel Electrophoresis Gel electrophoresis of LOS samples was performed as described in Lesse et al. (Lesse et al. J. Immunol. Meth. 126: 109 (1990)) and silver-stained gels (Hitchcock and Brown J. Bacteriol. 154-269 (1983)).<u style="single">result</u> Cloning of the LOS locus: In an attempt to isolate the Neisseria gonorrhoeae porin gene, colony blots showed a pBR322 clone containing a 4.9 kb ClaI fragment that reacted with rabbit antiserum against purified porin. It was repeatedly isolated. An immunoreactive subclone, pR10PI consisting of a 1305 bp RsaI-ClaI fragment, was induced and its DNA sequence was determined. This sequence is known to be involved in the synthesis of LPS of its kind, Haemophilus influenzae, so-called lex-1 (Cope et al., 1991, Molec. Microbiol. 5: 1113), or It had homology to the gene isolated from lic2A (High et al., 1993, Molec. Microbiol. 9: 1275). Neisseria gonorrhoeae genomic DNA digested with ClaI using subclone pR10PI as a probe Southern blots showed hybridization with two fragments, 4.9 and 3.4 kb. However, digestion with other restriction enzymes gave only a single band. Notably, digestion with BfaI gave only a single band of 4.1 kb, suggesting that the two copies were closely bound (data not shown).

The λ2001 bank of Neisseria gonorrhoeae F62 DNA was screened by hybridization with pR10PI and 5 clones were isolated. One of those clones was digested with ClaI or BfaI and gave the same pattern as what appears to be genomic DNA when tested by Southern hybridization using pR10PI as a probe. A suitable ClaI fragment of this λ2001 clone was isolated and cloned into the ClaI site of pBluescript II SK-. The entire sequence of the 3400 ClaI fragment was determined. Mapping of the clone containing the 4900 bp ClaI fragment showed that a single PstI site was present in the clone of about 2.8 kb from one end and could be divided into two subclones. The terminal partial sequence of the 2.1 kb subunit contains a coding frame homologous to the E. coli COOH-terminal of the α subunit of glycoyl-tRNA synthesizer (glyS) and most of the β subunit of this gene. Was shown (Webster et al., 1983, J. Biol. Chem. 258: 10637). There was a predicted length of DNA that needed to match the E. coli sequence. This clone was not further tested.

DNA sequence at the LOS locus: Figure 2 shows a summary of the features found by sequence analysis of the two clones. Following the glyS gene, five close-located open reading frames were found. The last frame has a sequence typical of the rho independent terminal signal 46 bp downstream of the stop codon. Subsequently, there is a region of approximately 100 bp with significant homology to the IS1106 Neisseria insert (Knight et al., 1992, Molec. Microbiol. 6: 1565). A further explanation of the nature of this locus, presented below, is that five open reading frames encode LOS glycosyltransferases, hence they were named lgtA-lgtE. Indicates that.

The investigation of internal homology within this locus shows that the DNA encoding the first two genes (lgtA, lgtB) is repeated as the 4th and 5th genes (lgtD, lgtE), and the ones that are inserted are additional. It was shown to be an open reading frame, lgtC. This means that lgtB and the pR10PI probe containing a small portion of the lgtC gene hybridize with two ClaI fragments, but only with the only BfaI fragment (see the BfaI portion of the LOS locus in Figure 2). This is in agreement with the data obtained by the Southern hybridization shown above. More specifically, 16 bp following the stop codon of the tRNA synthesizer (glyS) is the starting point of the stem loop structure immediately followed by the consensus ribosome binding site (rbs), and within 6 bp, the initiation of lgtA. There is a TTG that is considered to be a codon. 2871 bp downstream from the start of the trunk loop (immediately followed by the stop codon of lgtC) is the TTG of the trunk loop structure, rbs and lgtD, with downstream sequences that are highly homologous to about 500 bp. There is an almost complete repeat of the start codon. This sequence then varies to some extent. However, at the starting point of lgtB and lgtE, the homology is again almost complete for about 200 bases and then fluctuates towards the later part of the orf. The similarities of this homologous protein are shown in Figures 3 and 4. These comparisons show that the primary structure is almost completely conserved at the N-terminal portion of the molecule, with increasing variability towards the COOH terminal of the protein.

The lgtC sequence inserted between the repeating parts of the locus is not repeated within the locus or in the gonococcal genome (data not shown). This appears to be homologous to the closely related genes E. coli rfaI or rfaJ genes that act as glucosyltransferases in core LPS biosynthesis (Pradel et al., 1992, J. Bacteriol. 174: 4736). The similarity between rfaI and lgtC is shown in Figure 5. Three of these genes, contained within their coding frame, were found to run of guanosine, which encodes glycine elongation (see Figure 2). These poly-G regions were found in lgtA (17bp), lgtC (10bp) and lgtD (11bp), in which case the number of G residues was the number that maintained a complete reading frame (in each case). See Figures 3 and 5). Changes in 1 or 2 G bases in each of the three genes cause premature termination of transcription.

LOS phenotype of gonococcus F62 due to deletion of LOS locus: To determine the function of the lgt gene, insertion or deletion of the LOS locus was constructed with a plasmid grown in Escherichia coli. Insertions or deletions in each case were marked with the ermC'gene, an excellent selective marker in N. gonorrhoeae (Klugman et al., 1989, Infect. Immun. 57: 2066). This construction is summarized in Figures 6, I1, I2 and I3, showing the insertion of the ermC'gene into each of the BsaBI, AgeI and AscI sites. Similarly, the deletion was constructed by excising a portion of the plasmid and replacing it with an erythromycin marker. White arrows indicate the genes to be discussed. Each of these plasmids is N. gonorrhoeae strain F62 The transformant was selected on an erythromycin-containing plate. The fidelity of the genomic variation of each prototype of the gonococcal transformant was verified by sequence analysis of the upstream and downstream binding points of the ermC'gene. To simplify the nomenclature in this report, the gonococcal variant was given the same name used to identify the plasmid construct in Figure 6.

The mutant LOS was examined by SDS-PAGE and compared with the LOS of strain 1291e. This strain was first isolated by Dudas and Apicella (Dudas and Apicella, Infect.Immun., Vol. 56, p. 499 (1988)) as the first wild-type piosin-resistant mutant of strain 1291. , Chemically and genetically extensively characterized. Chemical analysis has shown that this mutant completely lacks the lacto-N-neotetraose substitution on heptose 1 (John et al., J. Biol. Chem., Vol. 266, p. 19303). (1991)). The genetic basis for this mutant has been identified (Zhou et al., J. Biol. Chem., Vol. 269, p. 11162 (1994), Sandlin and Stein, J. Bacteriol. , Vol. 176, p. 2930 (1994)), which encodes for phosphoglucomutase pgm It is a mutation of a gene. This mutation inhibits the synthesis of UDP-glucose, thereby inhibiting the addition of glucose to heptose. As shown in Figure 7, the parental wild-type F62 strain resulted in two major LOS bands, the appearance of which is the SDS-PAGE pattern (Schneider) previously published by other researchers. Et al., Amer.Soc.Microbiology, Washington, pp. 400-405 (1985)). Mutants were arranged on the gel according to the size of the major band contained in the mutant. This size decreases in four distinct steps, from the highest band of F62 wt LOS to the size of Δ4 or 12 LOS. The lgtE product is initially the first because the 12 mutants (with insertion into the last gene in the locus, lgtE) have the same phenotype as Δ4 (completely deleted at the locus). It is suggested that the biosynthesis step of. Thus, the enzyme encoded by lgtA-D is complete but has no substrate to act on. Mutant Δ5 (locus excluding lgtE is deleted) is one step larger LOS This supports the idea that this gene contributes to the first biosynthetic process. Note that the LOS of both the 12 and Δ4 mutants is clearly higher than that of strain 1291e, which is known to be unable to add to glucose, the first residue in the lacto-N-neotetraose chain. Should. These data suggest that lgtE encodes a galactosyltransferase that attaches to the first galactose of lacto-N-neotetraose.

In addition, dot blot techniques were used to examine LOS samples for their reactivity with monoclonal antibodies. The monoclonal antibody used and its reported specificity are shown in FIG. The reactions observed for LOS obtained from the parent strain and mutant are summarized in Figure 8. Reactivity of parent F62 with 1-1-M, 3F11 and L8 has traditionally been described by Mandrell et al. (Mandrel et al., Amer.Soc.Microbiology, Washington, pp. 379-384 (1985)) and Yamasaki. ) Et al. (Yamasaki et al., Mol. Immunol., Vol. 28, p. 1233 (1991)).

Mutants Δ4 and I2 did not react with either antibody. However, Δ5 reacts strongly with antibodies C4C and L8, indicating the presence of the first galactose residue. This is consistent with SDS-PAGE results (see Figure 6) and supports the role of lgtE as a galactosyltransferase. This also indicates that the upstream deletion of lgtE does not significantly inactivate its function due to the polar effect. The LOS of the F62 wt parent had strong reactivity with L3 and weak reactivity with 3F11. The reactivity of 3F11 is known to be blocked by the addition of GalNAc residues (Schneider et al., J.Exp.Med., Vol. 174, p. 1601), but this is not the case with L3 antibodies. Was there. wt LOS reacted with 1-1-M, an antibody that is reactive in the presence of terminal GalNAc residues. Reactivity with 1-1-M was lost in Δ3, which has a deletion only in lgtD. This suggests that this gene encodes a GalNAc transferase.

Reactivity with antibody L1 (specific for alternative LOS structures capped with α1 4Gal) was not observed with wt LOS and was absent in all deletions affecting 11 and lgtC. The reactivity was strongest in Δ1 with only a deletion of lgtA. It should be noted that this mutant also has no reactivity with 3F11 and L3. These two findings suggest that lgtA encodes for GlcNAc transferase and, in the absence of this residue, the substrate for the action of lgtC to produce an alternative LOS structure is this incomplete strand. .. It should be noted that the size of the LOS product seen in Figure 7 is consistent with the immunological data. This result suggests that lgtC encodes an α-Gal transferase. This is further supported by the weak reactivity of mutant Δ3 with antibody L1. Mutant Δ3 has a deletion of lgtD and cannot be added to the terminal GalNAc, and the α-Gal transferase is lacto-N-. It is possible to modify the neotetraose group to produce Pi-like globosides (Mandrel, Infect. Immun., Vol. 60, p. 3017 (1992)). Mutant 13 (with inactive lgtB) lost reactivity with 1-1-M, 3F11 and L1, leaving only weak reactivity with L3. These observations, along with product size, suggest that lgtB encodes a galactosyltransferase that adds Gal β1 4 to GlcNAc residues. Himalectin RCA-I is specific for terminal galactose in the β-chain (Nicolson and Blaustein, Biochim.Biophys.Acta, Vol. 266, p. 543 (1972), Lin) and Li, Eur.J.Biochem., Vol. 105, p. 453 (1980)), which was used to confirm the presence of this structure on LOS samples. Wild-type, Δ3, Δ2 and Δ5 LOS using ELISA tests, terminal βGal (See FIG. 7), whereas Δ4, I2, Δ1 and I3 were non-reactive (data not shown).

<u style="single">Consideration</u> A locus containing 5 open reading frames was cloned. From the effects of the eight identified mutants within this locus on the size and serological reactivity of LOS produced by gonococcal transformants, these genes are lacto-N-neotetra. It is suggested that it is a glycosyltransferase that controls the biosynthesis of most of the aus chain. From the obtained data, it is possible to identify the function of each of these genes. It should be noted that the structurally very closely related lgtB and lgtE also act on the apparently very similar biosynthesis of adding Galβ1 4 to GlcNAc or Glc, respectively. Similarly, the closely related lgtA and lgtD add GlcNAc or GlcNAcβ1 3, respectively, to the Gal residue. LgtC, which is unrelated to other genes in the locus, controls the addition of Gal α1 4.

The DNA sequence showed that three of the genes (lgtA, lgtC and lgtD) contained a region of guanosine encoding a glycine residue in the protein. These provide a potential mechanism for frequent mutations in the expression of the gene. It has been well reported that such slippage in the poly-G region regulates the expression of the Neisseria gonorrhoeae pilC gene and, as a result, affects the adhesion of packs to human epithelial cells ( Rudel et al., Molec.Microbiol., Vol. 6, p. 3439 (1992)). In strain F62, the number of bases in each of the three poly-G regions is such that the protein is in the frame, which is the ability of the F62 wild-type to produce complete LOS, including the addition of terminal GalNAc. Matches with.

The three aspects of LOS biosynthesis appear to be dependent on high frequency mutations. The first is the addition of terminal GalNAc (lgtD). This causes a change in reactivity with the monoclonal antibody 1-1-M, and this phase variation has been reported by Van Putten (1993, EMBO J. 12: 4043). Similarly, changes in lgtA cause the failure of GlcNAc addition to the growth chain, shortening LOS to β-lactosyl levels. This is a 3.6 kilodalton molecule, a very normal form of LOS in Neisseria gonorrhoeae, which provides resistance to the bactericidal effects of normal human serum (Schneider et al., 1985, Infect. Immun. 50: 672). .. In vitro mutations between variants A and C of MS11mk from β-lactosyl chains to complete LOS (which had selective advantages in vivo in healthy individuals) restored functional expression of GlcNAc transferase lgtA. It can be inferred that it can be explained by. Finally, β-lactosyl (pk-like globotriose) or lacto-N- Addition of α1 4Gal mutations to any of the neotetraose groups (Pi-like globosides) (Mandrell, 1992, Infect. Innun. 60: 3017) is under the control of lgtC expression. There will be. The activity of lgtC transferase is inferior to competing with other transferases due to its precursor, and its activity is apparent only when either lgtA or lgtD is inactive. The GlcNAc transferase lgtA should be inactive due to the synthesized Galα1 4Galβ1 4Glc trisaccharide, and the GalNAc transferase lgtD is dormant due to the expression of Pi-like globoside Galα1 4Galβ1 4GlcNAc β1 3Galβ1 4Glc. Must be.

Comparable high-frequency antigenic variation in Haemophilus influenzae LOS has also been noted, with two isolated loci, licl (Weiser et al., 1989, Cell 59: 657) and lic2 (high (high). High) et al., 1993, Molec. Microbiol. 9: 1275) attributed to changes in translation frames caused by shifts in the number of CAAT iterations. The shift that allows the expression of the lic2 gene correlates with the expression of an epitope having the structure Galα1 4Galβ1 . The lic2 gene is homologous to lgtB and lgtE, and galactosyltransferase binds to Glc or GlcNAc at Galβ1 4, respectively, which appears to be its function in Haemophilus influenzae LOS synthesis. It should be noted that both of these mucosal pathogenic bacteria develop a frameshift mechanism to cause antigenic variation in LOS, while the lic2 gonococcal homologues (lgtB and lgtE) do not contain poly-G tracts. is there.

While the frame-shift mechanism described above is suitable for on / off regulation of gene expression, the structure of its locus also contributes to more precise regulation of the level of gene expression. It has been demonstrated that the growth rate affects the molecular weight distribution and the antigenic properties produced by the LOS species (Morse et al., 1983, Infect. Immun. 41:74). Although the size of the RNA transcript has not been measured, it appears that lgtA, lgtB and lgtC (eg, where the poly-G tract maintains its coding frame) are transcribed together. The terminal codon of lgtA and the start codon of lgtB actually overlap, and the distance between TAA of lgtB and ATG of lgtC is only 11 base pairs. Similarly, the stop codon of lgtD and the start codon of lgtE are only 18 base pairs apart. If this mechanism is off-positioned in any of the three genes that are mutated, transcription can be effectively restarted from the beginning of the next gene. This ability to restart transcription is evident in the mutants constructed in this experiment.

The correlation between LOS structure and function is still in its infancy. A major advance in this area is the development of an understanding of its molecular structure and its ability, which is often apparently related to its reactivity with many well-characterized monoclonal antibodies. In addition to this, in the in vivo environment that provides CMP-NANA, the reality is that organisms sialylate or do not sialylate LOS, depending on whether the synthesized LOS has a competent receptor structure. It is well known that sialylation induces a serum-resistant state in many strains. However, the effect of sialization on local infections has not been well studied. Vampatin showed that sialylation of LOS had a significant inhibitory effect on epithelial cell invasion without significantly altering adhesion (Vampatin, 1993, EMBO). J.12: 4043). His work suggests that in mucosal infections, LOS structures that cannot be sialylated may be important for efficient cell invasion. According to the context of this report, such a structure can be achieved either by efficient addition of terminal GalNAc or by shortening the LOS chain by resting GlcNAc transferase. The correlation between LOS chemistry and biological responses is complicated by the leakage of real LOS mutants isolated by piosin selection (Dudas and Apicella, 1988, Infect. Immun. 56: 499; Sandlin et al., 1993, Infect. Immun. 61: 3360). This is actually the mutant 1291e Illustrated in, in addition to the large low molecular weight bands, additional higher bands are shown (see Figure 7). Its new insights into the biosynthetic genetic features of Neisseria gonorrhoeae LOS will allow the construction of complete mutants. For example, Δ4 and Δ5 are stable mutants because they no longer contain genes with poly-G tracts. Expression of genes containing poly-G tracts will be stabilized by designing the region to encode glycine with other codons.

The scope of the present invention is not limited by the specific embodiments described herein. This is because such embodiments are intended as a brief description of one aspect of the invention, and any functionally equivalent embodiment is within the scope of the invention. In fact, various variations of the invention in addition to those shown and described herein will be apparent to those skilled in the art from the earlier description and accompanying drawings. Such modifications are intended to fall within the scope of the appended claims. It is also understood that the size of all base pairs given to a nucleotide is an approximation and is used for descriptive purposes. Various references are cited herein and their disclosures are incorporated herein by reference in their entirety.

Embodiments of the present invention are shown below. 1. Purified nucleic acid that can hybridize to a nucleic acid having a nucleotide sequence corresponding to or complementary to the nucleotide sequence shown in FIG. 2 (SEQ ID NO: 1 in the Sequence Listing) under moderate stringent conditions. .. 2. A nucleotide sequence that corresponds to or is complementary to a portion of the nucleotide sequence shown in Figure 2 (SEQ ID NO: 1 in the Sequence Listing) that encodes a functionally active glycosyltransferase under moderate stringent conditions. The nucleic acid according to 1 above, which can hybridize to a nucleic acid having. 3. The nucleic acids described in 2 above, which encode a functionally active glycosyltransferase. 4. The nucleic acid according to 1 above, which has a nucleotide sequence corresponding to or complementary to a part of the nucleotide sequence shown in FIG. 2 (SEQ ID NO: 1 in the Sequence Listing) encoding a functionally active glycosyltransferase.

5. The nucleic acids described in 4 above, which encode a functionally active glycosyltransferase. 6. The nucleic acid according to 1 above, which has a nucleotide sequence corresponding to or complementary to the nucleotide sequence shown in FIG. 2 (SEQ ID NO: 1 of SEQ ID NO: in the Sequence Listing). 7. The nucleic acid according to 3 above, wherein the functionally active glycosyltransferase catalyzes a reaction selected from the group consisting of a) to c) below: a) Add Galβ1 4 to GlcNAc or Glc; b) Add GalNAc or GlcNAcβ1 3 to Gal; and c) Add Galα1 4 to Gal. 8. The nucleic acid according to 3 above, which encodes a glycosyltransferase having the amino acid sequence of SEQ ID NO: 3 in the sequence listing. 9. The nucleic acid according to 3 above, which encodes a glycosyltransferase having the amino acid sequence of SEQ ID NO: 8 in the sequence listing. 10. The nucleic acid according to 3 above, which encodes a glycosyltransferase having the amino acid sequence of SEQ ID NO: 4 in the sequence listing. 11. The nucleic acid according to 3 above, which encodes a glycosyltransferase having the amino acid sequence of SEQ ID NO: 5 in the sequence listing. 12. The nucleic acid according to 3 above, which encodes a glycosyltransferase having the amino acid sequence of SEQ ID NO: 6 in the sequence listing. 13. An expression vector containing the nucleic acids described in 3 above, which are functionally bound to the expression control sequence. 14. Recombinant host cells transformed with the expression vector described in 13 above . 15. Method for producing glycosyltransferase including the following steps: a) Culturing the recombinant host cells according to 14 above under conditions that allow expression of the glycosyltransferase; and b) Recovering the expressed glycosyltransferase.

16. A glycosyltransferase having the amino acid sequence of SEQ ID NO: 3 in the Sequence Listing or a functionally active fragment thereof. 17. A glycosyltransferase having the amino acid sequence of SEQ ID NO: 8 in the Sequence Listing or a functionally active fragment thereof. 18. A glycosyltransferase having the amino acid sequence of SEQ ID NO: 4 in the Sequence Listing or a functionally active fragment thereof. 19. A glycosyltransferase having the amino acid sequence of SEQ ID NO: 5 in the Sequence Listing or a functionally active fragment thereof. 20. A glycosyltransferase having the amino acid sequence of SEQ ID NO: 6 in the Sequence Listing or a functionally active fragment thereof. 21. A composition containing a glycosyltransferase bound to a solid phase support, wherein the glycosyltransferase is selected from the group consisting of the following a) to e): a) SEQ ID NO: 3 in the sequence listing. Glycosyltransferase having the amino acid sequence of, or a functionally active fragment thereof; b) Glycosyltransferase having the amino acid sequence of SEQ ID NO: 8 in the Sequence Listing or its functionally active fragment; c) A glycosyltransferase having the amino acid sequence of SEQ ID NO: 4 in the Sequence Listing or a functionally active fragment thereof; d) A glycosyltransferase having the amino acid sequence of SEQ ID NO: 5 in the Sequence Listing or a functionally active fragment thereof; and e. ) A glycosyl transferase having the amino acid sequence of SEQ ID NO: 6 in the Sequence Listing or a functionally active fragment thereof.

22. A method of adding GalNAc or GlcNAcβ1 3 to Gal, in which a reaction mixture containing activated GalNAc or GlcNAc is contacted with a receptor component containing a Gal residue in the presence of the glycosyltransferase described in 16 above. Methods involving letting. 23. A method of adding Gal β1 4 to GlcNAc or Glc, in which a reaction mixture containing activated Gal is added to the receptor component containing GlcNAc or Glc residue in the presence of the glycosyltransferase described in 17 above. Methods involving contact. 24. A method of adding Gal α1 4 to Gal, in which a reaction mixture containing activated Gal is contacted with a receptor component containing a Gal residue in the presence of the glycosyltransferase described in 18 above. How to include. 25. A method of adding GalNAc or GlcNAcβ1 3 to Gal, in the presence of the glycosyltransferase described in 19 above, a reaction mixture containing activated GalNAc or GlcNAc in a receptor component containing a Gal residue. Methods involving contact. 26. Gal β1 4 to GlcNAc or Glc A method comprising contacting a reaction mixture containing activated Gal with a receptor component containing GlcNAc or a Glc residue in the presence of the glycosyltransferase described in 20 above.

27. A method for producing an oligosaccharide having the structure Gal α1 4Gal β1 4Glc, which comprises continuously carrying out the following steps: a) The amino acid sequence of SEQ ID NO: 6 in the sequence listing, or its functionally active The step of contacting the receptor component containing the Glc residue with the reaction mixture containing the activated Gal in the presence of a glycosyltransferase having a fragment; and b) the amino acid sequence of SEQ ID NO: 4 in the Sequence Listing, or its function. A step of contacting a reaction mixture containing activated Gal with a receptor component containing a Gal β1 4Glc residue in the presence of a glycosyltransferase having an active fragment. 28. Structure Gal β1 4 A method for producing an oligosaccharide having Glc, which is a reaction mixture containing activated Gal in a receptor component containing a Glc residue in the presence of the glycosyltransferase described in 20 above. Methods involving contacting. 29. Structure GlcNAcβ1 3Galβ1 A method for producing an oligosaccharide having 4Glc, in the presence of the glycosyltransferase described in 16 above, Gal β1 A method involving contacting a reaction mixture containing activated GalNAc with a receptor component containing a 4Glc residue. 30. Structure Gal β1 4GlcNAc β1 3Gal β1 A method for producing oligosaccharides having 4Glc, which is active in the receptor component containing GlcNAc β1 3Gal β1 4Glc residue in the presence of the glycosyltransferase described in 17 above. A method comprising contacting a reaction mixture containing a glycosylated Gal. 31. Structure GalNAc β1 3Gal β1 4GlcNAc β1 3Gal β1 A method for producing oligosaccharides having 4Glc, which contains Gal β1 4GlcNAc β1 3Gal β1 4Glc residues in the presence of the glycosyltransferase described in 19 above. A method comprising contacting a body component with a reaction mixture containing activated GalNAc.

32. Method for producing oligosaccharide having structure GalNAcβ1 3Galβ1 4GlcNAc β1 3Galβ1 4Glc, which comprises continuously carrying out the following steps: a) Amino acid sequence of SEQ ID NO: 6 in the sequence listing, or its function The step of contacting a reaction mixture containing activated Gal with a receptor component containing a Glc residue in the presence of a glycosyltransferase having an active fragment in; b) The amino acid sequence of SEQ ID NO: 3 in the Sequence Listing, or The step of contacting the reaction mixture containing activated GlcNAc with the receptor component containing Gal β1 4Glc residue in the presence of a glycosyltransferase having the functionally active fragment; c) SEQ ID NO: in the sequence listing. Step of contacting a reaction mixture containing activated Gal with a receptor component containing GlcNAcβ1 3Galβ1 4Glc residues in the presence of a glycosyltransferase having an amino acid sequence of 8; and d) SEQ ID NO: 5 in the sequence listing. Gal β1 4GlcNAc β1 3Gal β1 in the presence of the amino acid sequence of A step of contacting a reaction mixture containing activated GalNAc with a receptor component containing a 4Glc residue.

33. Method for producing oligosaccharide having structure Gal β1 4GlcNAc β1 3Gal β1 4Glc, which comprises continuously carrying out the following steps: a) Amino acid sequence of SEQ ID NO: 6 in the sequence listing, or its function The step of contacting a reaction mixture containing activated Gal with a receptor component containing a Glc residue in the presence of a glycosyltransferase having an active fragment; b) The amino acid sequence of SEQ ID NO: 3 in the Sequence Listing, or the amino acid sequence thereof. The step of contacting the reaction mixture containing activated GlcNAc with the receptor moiety containing Gal β1 4Glc residue in the presence of a glycosyltransferase having a functionally active fragment; and c) SEQ ID NO: in the Sequence Listing. A step of contacting a reaction mixture containing activated Gal with a receptor component containing GlcNAcβ1 3Galβ1 4Glc residues in the presence of a glycosyltransferase having an amino acid sequence of 8.

<figref num="1">Alternating structure found in Neisseria gonorrhoeae LOS. R1 refers to the inner core region of LOS consisting of two keto-deoxy-octulosonic acid (KDO) residues. These are in turn attached to the lipid A structure. R2 in Neisseria gonorrhoeae is typically GlcNAcβ1 2Hepα1 3. The structure in the upper figure contains tetrasaccharides equivalent to lacto-N-neotetraose found in paragloboside glycolipids. In many strains, this tetrasaccharide has a terminal GalNAcβ1 3. The figure below shows an alternating trisaccharide structure with bound terminal Galα1 4. This trisaccharide is found in the L1 serotype Meningococcus and some N. gonorrhoeae strains. These two structural parts identified by the monoclonal antibody used in this study are shown, these are (4C4) (Dudas and Apicella, Infect. Immun., 1988, 56: 499), 3F11. (Mandrell et al., J.Ex.Med., 1988, 168: 107; Yamasaki et al., Mol.Immunol., 1991, 28: 1233), 1-1-M (Yamasaki et al., Mol.Immunol., 1991, 28: 1233), 2-1-L8 (Kerwood) Etc., Biochemistry, 1992, 31:12760; Schneider et al., J.Ex.Med., 1991, 174: 1601; Schneider et al., Infect. Immun., 1985, 50: 672), 9-2 -L378 and 17-1-L1.</figref><figref num="2A">2A is a gene map of the LOS locus based on the DNA sequence. Sequence information bp 1-2725 is from plasmid pPstCla and bp 2725-5859 is from plasmid p3400 (see Materials and Methods section). IS means a region of the sequence that has homology to the previously reported Neisseria insertion sequence IS1106 (Knight et al., Molec. Microbiol., 1992, 6: 1565). The position of the reading frame of lgtA-E is shown. Three spreads of poly-G are found in lgtA (17bp), lgtC (10bp) and lgtD (11bp), which are indicated by vertical black bars.</figref><figref num="2B">2B is the amino acid sequence of LgtA (SEQ ID NO: 3).</figref><figref num="2C">2C is the amino acid sequence of LgtB (SEQ ID NO: 8).</figref><figref num="2D">2D is the amino acid sequence of LgtC (SEQ ID NO: 4).</figref><figref num="2E">2E is the amino acid sequence of LgtD (SEQ ID NO: 5).</figref><figref num="2F">2F is the amino acid sequence of LgtE (SEQ ID NO: 6).</figref><figref num="2G">2G is the nucleic acid sequence of the lgt locus (SEQ ID NO: 1).</figref><figref num="2H">2H is the nucleic acid sequence of the lgt locus (SEQ ID NO: 1).</figref><figref num="2I">2I is the nucleic acid sequence of the lgt locus (SEQ ID NO: 1).</figref><figref num="2J">2J is the nucleic acid sequence of the lgt locus (SEQ ID NO: 1).</figref><figref num="2K">2K is the nucleic acid sequence of the lgt locus (SEQ ID NO: 1).</figref><figref num="2L">2L is the nucleic acid sequence of the lgt locus (SEQ ID NO: 1).</figref><figref num="2M">2M is the nucleic acid sequence of the lgt locus (SEQ ID NO: 1).</figref><figref num="3A">3A is the homology of the lgtA and lgtD protein products. The primary structures of the two proteins are very similar, especially in the first half of the sequence. The glycine residue starting at position 86 reflects the coding of the poly-G region in each gene. The Bestfit program of the GCG package is used and the symbols (-), (), () represent the degree of similarity based on the Dayhoff PAM-250 matrix.</figref><figref num="3B">3B is the homology of the lgtA and lgtD protein products. The primary structures of the two proteins are very similar, especially in the first half of the sequence. The glycine residue starting at position 86 reflects the coding of the poly-G region in each gene. The Bestfit program of the GCG package is used and the symbols (-), (), () represent the degree of similarity based on the Dayhoff PAM-250 matrix.</figref><figref num="4A">4A is the homology of lgtB and lgtE protein products. The primary structures of the two proteins are very similar, especially in the first half of the sequence. These sequences also include Haemophilus influenzae, lex-1 (Cope et al., Molec. Microbiol., 1991, 5: 1113) or lic2A (High et al., Molec. Microbiol., 1993). , 9: 1275) also has considerable homology to the gene. See Figure 3 for the meaning of the symbols.</figref><figref num="4B">4B is the homology of lgtB and lgtE protein products. The primary structures of the two proteins are very similar, especially in the first half of the sequence. These sequences also include Haemophilus influenzae, lex-1 (Cope et al., Molec. Microbiol., 1991, 5: 1113) or lic2A (High et al., Molec. Microbiol., 1993). , 9: 1275) also has considerable homology to the gene. See Figure 3 for the meaning of the symbols.</figref><figref num="5A">5A is the homology of the protein products of rfaI and lgtC. E. Kori's rfaI and rfaI genes are very closely related. They function as glycosyltransferases for the two glucose residues in the LPS core region (Pradel et al., J. Bacteriol., 1992, 174: 4736). Glycine at positions 54-56 of lgtC is encoded by the spread of the poly-G. See Figure 3 for the meaning of the symbols.</figref><figref num="5B">5B is the homology of the protein products of rfaI and lgtC. E. Kori's rfaI and rfaI genes are very closely related. They function as glycosyltransferases for the two glucose residues in the LPS core region (Pradel et al., J. Bacteriol., 1992, 174: 4736). Glycine at positions 54-56 of lgtC is encoded by the spread of the poly-G. See Figure 3 for the meaning of the symbols.</figref><figref num="6">Deletion at the LOS locus. The three insertions and five deletions of the LOS locus were constructed as detailed in the section on methods below. The restricted sites used are shown. Insertions are indicated by triangles and the degree of deletion is indicated by stippling boxes. White arrows indicate the reading frame collapsed by this construction. In each of the constructs, the erythromycin marker ermC'was inserted at the insertion or deletion site.</figref><figref num="7">Silver-stained SDS-PAGE of LOS preparations. Gel electrophoresis of 375 ng of purified LOS sample was performed and stained as described in the Materials and Methods section. Above the gel shows the structure of the LOS with a major band presumed to be present in each of the preparations. These structures are based on the reactivity with the monoclonal antibody shown in FIG. 8, which is shown in the figure to simplify the description of the observed patterns. R represents the inner core region and lipid A. 1291e is a piocin-resistant mutant (Dudas and Apicella, Infect. Immun., 1988, 56: 499).</figref><figref num="8">Reactivity of LOS derived from strain F62 wt and mutant with monoclonal antibody. Monoclonal antibodies have been abbreviated as follows: 17-1-L1 (L1), 9-2-L378 (L3), 2-1-2L8 (L8). Purified LOS was applied to Immobilon-P membranes, reacted with antibodies and developed as described in the Materials and Methods section. The specificity of the monoclonal antibody is summarized in FIG.</figref>