Bordetella vaccines comprising LPS with reduced reactogenicity

Abstract

The current invention lies in the field of medicine and more specifically in the field of vaccinology. The current invention concerns a novel Bordetella LPS and a modified bacterium of the genus Bordetella comprising such modified LPS. The LPS of the invention has a reduced endotoxicity in comparison to an unmodified Bordetella LPS. The modified LPS of the invention is therefore particularly suitable for use in inducing or stimulating an immune response in a subject, wherein the immune response is induced or stimulated against a Bordetella infection. The modified Bordetella LPS of the invention is obtainable by introducing in a Bordetella cell the expression of a heterologous acyl transferase. In particular, the modified Bordetella cell of the invention has an increased expression of an heterologous LpxA, LpxL or LpxD acyl transferase.

Claims

1. A genetically modified Bordetella pertussis, Bordetella parapertussis or Bordetella bronchiseptica bacterium, wherein the bacterium is modified compared to the wild-type Bordetella pertussis, Bordetella parapertussis or Bordetella bronchiseptica bacterium in that it has a genetic modification that introduces a heterologous acyl transferase activity, wherein the genetic modification that introduces heterologous acyl transferase activity confers to the cell at least one of a heterologous LpxA, LpxL and LpxD acyl transferase activity and wherein the genetic modification introduces the expression of at least one of a heterologous LpxA, a LpxL, and a LpxD acyl transferase, wherein i) the LpxA acyl transferase has SEQ ID NO: 1, or a variant thereof having at least 95% amino acid sequence identity with SEQ ID NO. 1; ii) the LpxL acyl transferase has SEQ ID NO: 2, or a variant thereof having at least 95% amino acid sequence identity with SEQ ID NO. 2; and/or iii) the LpxD acyl transferase has SEQ ID NO: 4, or a variant thereof having at least 95% amino acid sequence identity with SEQ ID NO. 4, wherein expression of the heterologous acyl transferase results in a B. pertussis, B. parapertussis or B. bronchiseptica LPS having a lipid A moiety that is modified as compared to the lipid A moiety of a wild-type B. pertussis, B. parapertussis or B. bronchiseptica LPS in that the length of at least one acyl chain is shorter and wherein the length of the acyl chain at the 3 position of the modified lipid A moiety does not have a greater length than the acyl chain of the wild-type B. pertussis, B. parapertussis or B. bronchiseptica lipid A moiety at the same 3 position.

2. The genetically modified bacterium according to claim 1, wherein the bacterium is modified compared to the wild-type B. pertussis, B. parapertussis or B. bronchiseptica bacterium in that it has a genetic modification that introduces a heterologous UDP-2,3-diacylglucosamine pyrophosphatase activity.

3. The genetically modified bacterium according to claim 2, wherein the genetic modification introduces the expression of a heterologous lpxH UDP-2,3-diacylglucosamine pyrophosphatase, and wherein the LpxH UDP-2,3-diacylglucosamine pyrophosphatase has at least 95% sequence identity with SEQ ID NO: 5.

4. The genetically modified bacterium according to claim 3, wherein the lpxH UDP-2,3-diacylglucosamine pyrophosphatase has SEQ ID NO: 5.

5. Bordetella LPS, wherein the LPS is obtainable from a genetically modified B. pertussis, B. parapertussis or B. bronchiseptica bacterium according to claim 1.

6. The Bordetella LPS according to claim 5, wherein at least 70% of the LPS has a modified lipid A moiety, wherein the lipid A moiety is modified as compared to the lipid A moiety of a wild-type Bordetella pertussis, Bordetella parapertussis or Bordetella bronchiseptica LPS in that the length of at least one acyl chain is shorter, and wherein the shorter acyl chain selected from the group consisting of: i) the acyl chain at the 3′ position of the lipid A moiety is C10; ii) the primary acyl chain at the 2′ position of the lipid A moiety is C12; iii) the secondary acyl chain at the 2′ position of the lipid A moiety is C12; and iv) the acyl chain at the 2 position of the Lipid A moiety is C12, and wherein the length of the acyl chain at the 3 position of the modified lipid A moiety does not have a greater length than the acyl chain of the wild-type Bordetella lipid A moiety at the same 3 position.

7. The Bordetella LPS according to claim 6, wherein the length of the acyl chain at the 3 position of the modified lipid A moiety is not greater than C10.

8. The Bordetella LPS according to claim 7, wherein the length of the acyl chain at the 3 position of the modified lipid A moiety has the same length as the acyl chain of the wild-type Bordetella lipid A moiety at the same 3 position.

9. The Bordetella LPS according to claim 8, wherein the length of the acyl chain at the 3 position is C10.

10. An OMV obtainable from the genetically modified bacterium as defined in claim 1.

11. A composition comprising at least one of: Bordetella LPS obtainable from a genetically modified bacterium according to claim 1; a genetically modified bacterium of claim 1; and an OMV obtainable from the genetically modified bacterium of claim 1, wherein optionally the composition is a pharmaceutical composition further comprising a pharmaceutically accepted excipient.

12. The composition according to claim 11, wherein the genetically modified bacterium is inactivated.

13. The composition according to claim 11, wherein the composition further comprises at least one non-Bordetella antigen.

14. The genetically modified bacterium according to claim 1, wherein the bacterium is a genetically modified Bordetella pertussis.

15. The genetically modified bacterium according to claim 14, wherein the bacterium is a genetically modified Bordetella pertussis B213 strain.

16. The genetically modified bacterium according to claim 1, wherein the bacterium has a genetic modification that increases the expression of a lipid A 3-O-deacylase, and wherein the lipid A 3-O-deacylase has at least 95% sequence identity with SEQ ID NO: 25.

17. The genetically modified bacterium according to claim 1, wherein the modified bacterium further comprises a genetic mutation that reduces or eliminates the activity and/or expression of an endogenous LpxA acyl transferase and/or an endogenous LpxD acyl transferase, wherein the endogenous LpxA acyl transferase has at least 95% sequence identity with SEQ ID NO: 28 and wherein the endogenous LpxD acyl transferase has at least 95% sequence identity with SEQ ID NO: 29.

18. The genetically modified bacterium according to claim 1, wherein i) the variant of the LpxA acyl transferase has at least 98% amino acid sequence identity with SEQ ID NO: 1; ii) the variant of the LpxL acyl transferase has at least 98% amino acid sequence identity with SEQ ID NO: 2; and/or iii) the variant of the LpxD acyl transferase has at least amino acid sequence identity with SEQ ID NO: 4.

19. The genetically modified bacterium according to claim 1, wherein i) the LpxA acyl transferase is SEQ ID NO: 1; ii) the LpxL acyl transferase is SEQ ID NO: 2; and/or iii) the LpxD acyl transferase is SEQ ID NO: 4.

20. The genetically modified bacterium according to claim 1, wherein i) the variant of the LpxA acyl transferase is a heterologous LpxA acyl transferase obtained from a Gram negative bacterium; ii) the variant of the LpxL acyl transferase is a heterologous LpxL acyl transferase obtained from a Gram negative bacterium; and iii) the variant of the LpxD acyl transferase is a heterologous LpxD acyl transferase obtained from a Gram negative bacterium.

21. A method for inducing or stimulating an immune response in a subject in need thereof, comprising administering to the subject the composition of claim 11, wherein the immune response is stimulated or induced against a Bordetella infection.

22. The method of claim 21, wherein the subject in need thereof suffers from whooping cough.

23. The method of claim 21, wherein the Bordetella infection is a Bordetella pertussis infection.

Description

DESCRIPTION OF THE FIGURES

(1) FIG. 1. Lipid A structure of wild-type and genetically modified B. pertussis. (A) Lipid A structure of wild-type B. pertussis. B) predicted lipid A structure of B. pertussis expressing LpxA.sub.(Nm) ΔlpxA. C) Lipid A structure of B. pertussis expressing LpxA.sub.(Pa)ΔlpxA. D) Lipid A structure of B. pertussis expressing LpxL.sub.(Nm)ΔlpxL and E) Lipid A structure of B. pertussis expressing LpxL.sub.(Pg)ΔlpxL and F) Lipid A structure of B. pertussis expressing LpxD.sub.(Pa)ΔlpxD and ΔlpxA, ΔlpxL and ΔlpxD indicate inactivation of the chromosomal lpxA, lpxL and ΔlpxD genes, respectively.

(2) FIG. 2. Implication of the expression of heterologous enzymes on growth. A) The OD.sub.590 of cultures of B213 and derivatives expressing LpxL.sub.(Nm), LpxA.sub.(Pa), or LpxL.sub.(Pg) from pMMB67EH plasmids, after 18 h of growth in Verweij medium in the presence of 1 mM IPTG is shown. The starting OD.sub.590 was 0.05. Data are from one representative experiment performed in duplicate of which average and standard variation are given. The growth defect of the strain expressing LpxL.sub.(Nm) was reproduced in two additional experiments. B) LpxD.sub.(Pa) The OD at 590 nm (OD.sub.590) of cultures of B213 and B213-pLpxD.sub.Pa clone 4 (cl4) and clone 5 (cl5) after 12 and 24 h of growth in liquid Verweij medium in the presence of 1 mM of IPTG is shown. The starting OD.sub.590 was 0.05.

(3) FIG. 3. Structural analysis by ESI-MS of lipid A. Negative-ion lipid A mass spectra were obtained by in-source collision-induced dissociation nano-ESI-FT-MS of intact LPS isolated from cells of B213, B213 expressing LpxA.sub.(Pa) (B213-pLpxA.sub.(Pa)), ΔlpxA mutant of B213 expressing LpxA.sub.(Pa) (B213ΔlpxA-pLpxA.sub.(Pa)) backgrounds, B213 expressing LpxL.sub.(Nm) (B213-pLpxL.sub.(Nm)), B213 expressing LpxL.sub.(Pg) (B213-pLpxL.sub.(Pg)) and B213 expressing LpxD.sub.(Pa) (B213-pLpxD.sub.(Pa)) (clones 4 and 5). Bacteria were grown for 12 h in liquid Verweij medium in the presence of 1 mM of IPTG. A major singly-deprotonated ion at m/z 1557.97 was interpreted as the typical B. pertussis lipid A structure: a diglucosamine (2 GlcN) penta-acylated (three 3OH—C14, one 3OH—C10 and one C14) with two phosphates residues (2 P) as illustrated in FIG. 1. Additional singly-deprotonated lipid A ions were detected in different derivatives and their interpretations are also indicated. Only the m/z range covering lipid A ions is shown.

(4) FIG. 4. Stimulation of HEK293 cells expressing hTLR4 (A, C) or mTLR4 (B, D) with purified LPS (A, B) or whole-cell preparations of B213 and LPS mutant derivatives (C, D). LPS preparations and bacterial suspensions, adjusted to an OD.sub.590 of 0.1, were serially diluted. After incubation for 2 h with HEK293 cells expressing mTLR4 or 4 h with HEK293 cells expressing hTLR4, alkaline phosphatase activity was determined by adding substrate and measuring the OD at 630 nm. One representative experiment is shown.

(5) FIG. 5. Stimulation of HEK293 cells expressing hTLR4 with LPS purified from B213, B213ΔlpxA-pLpxA.sub.(Pa), and B213-pLpxD.sub.(Pa) cl4 and cl5. Purified LPS at a concentration of 2 μg/ml was serially diluted, added to the cultured cells and incubated for 4 h. The OD at 630 nm resulting of SEAP activity is provided.

(6) FIG. 6. In vivo pyrogenicity. Pyrogenicity in rabbits induced by mutant Bordetella pertussis LPS purified from an lpxA.sub.(Pa) mutant and from an lpxD.sub.(Pa) mutant, and with OMVs extracted from the lpxD.sub.(Pa) mutant, all in comparison to B. pertussis wildtype LPS and OMVs. Pyrogenicity is expressed as area under curve for 0-48 h and 0-8 h intervals.

EXAMPLES

Example 1

(7) Material and Methods

(8) Plasmids, Strains and Growth Conditions

(9) Table 1 lists all plasmids and strains used in this study. B. pertussis strains were cultured on Bordet-Gengou agar (Difco) supplemented with 15% defibrinated sheep blood (Biotrading) for 48 h at 35° C. To grow the bacteria in liquid cultures, bacteria were collected from solid medium and diluted in Verweij medium [16] to an OD.sub.590 of 0.05 and incubated in 125-ml square media bottles with constant shaking at 175 rpm. In some assays, the bacteria were inactivated by incubation for 1 h at 60° C., resuspended in PBS and adjusted to an OD.sub.590 of 0.5. E. coli strains were grown in lysogeny broth (LB) or LB agar at 37° C.

(10) For all strains, media were supplemented with kanamycin (100 μg ml.sup.−1), gentamicin (10 μg ml.sup.−1), ampicillin (100 μg ml.sup.−1), nalidixic acid (50 μg ml.sup.−1), or streptomycin (300 μg ml.sup.−1) when required, and with 0.1 or 1 mM isopropyl-β-D-1-thiogalactopyranoside (IPTG) for E. coli or B. pertussis, respectively, to induce protein expression.

(11) TABLE-US-00001 TABLE 1 Used plasmids and strains Plasmids/Strains Characteristics Plasmids pMMB67EH Broad host-range vector, Ptac, lacl.sup.q, Amp.sup.R pKAS32 Allelic exchange suicide vector, Amp.sup.R pMMB67EH-PagL.sub.(Pa) pMMB67EH harboring pagL from P. aeruginosa PAO1 pMMB67EH-LpxA.sub.(Nm) pMMB67EH harboring lpxA from N. meningitidis H44/76 pMMB67EH-LpxA.sub.(Pa) pMMB67EH harboring lpxA from P. aeruginosa PAO1 pMMB67EH-LpxL.sub.(Nm) pMMB67EH harboring lpxL from N. meningitidis H44/76 pMMB67EH-LpxL.sub.(Pg) pMMB67EH harboring lpxL from Po. gingivalis ATCC33277 pMMB67EH-LpxD.sub.(Pa) pMMB67EH harboring lpxD from P. aeruginosa PAO1 pKA32-ABGH LpxL::gm pKAS32 derivative harboring lpxL.sub.1-lpxL.sub.2 knockout construct, Amp.sup.R, Gm.sup.R pRTP113368K2a lpxL2 knockout construct, Amp.sup.R, Kan.sup.R (kan in similar orientation as the operon) pRTP113368 k1a lpxL2 knockout construct, Amp.sup.R, Kan.sup.R (kan in opposite orientation as the operon) pRT669 lpxA knockout construct, Amp.sup.R, Kan.sup.R (kan in opposite orientation as the lpxA gene) Strains Escherichia coli DH5α F.sup.−, Δ(lacZYA-argF)U169 thi-1 hsdR17 gyrA96 recA 1 endA 1 supE44 relA1 phoA ϕ80 dlacZΔM15 SM10λpir thi thr leu fhuA lacY supE recA::RP4-2-Tc::Mu λpir R6K Kan.sup.R BL21(DE3) Contains gene for T7 DNA polymerase BL21-pLpxA.sub.(Nm) BL21(DE3) carrying pMMB67EH-LpxA.sub.(Nm) BL21-pLpxA.sub.(Pa) BL21(DE3) carrying pMMB67EH-LpxA.sub.(Pa) BL21-pLpxL.sub.(Nm) BL21(DE3) carrying pMMB67EH-LpxL.sub.(Nm) BL21-pLpxL.sub.(Pg) BL21(DE3) carrying pMMB67EH-LpxL.sub.(Pg) BL21-pLpxD.sub.(Pa) BL21(DE3) carrying pMMB67EH-LpxD.sub.(Pa) Bordetella pertussis B213 Nal.sup.R Str.sup.R derivative of strain Tohama I B213-pLpxA.sub.(Pa) B213 carrying pMMB67EH-LpxA.sub.(Pa) B213 ΔlpxA-pLpxA.sub.(Pa) B213 carrying pMMB67EH-LpxA.sub.(Pa) with an inactivated lpxA gene B213-pLpxL.sub.(Nm) B213 carrying pMMB67EH-LpxL.sub.(Nm) B213-pLpxL.sub.(Pg) B213 carrying pMMB67EH-LpxL.sub.(Pg) B213-pLpxD.sub.(Pa) B213 carrying pMMB67EH-LpxD.sub.(Pa) AmpR, ampicilin resistance; GmR, gentamicin resistance, KanR, kanamycin resistance; StrR, streptomycin resistance
Genetic Manipulations

(12) PCRs were performed using High Fidelity Polymerase (Roche Diagnostics GmbH, Germany). PCR mixes consisted of 1 μl of template DNA, 200 μM dNTPs (Fermentas), 0.25 μM of different primer combinations (SEQ ID NO: 7-24, see Table 2), 0.5 U DNA polymerase, and PCR buffer. The mixtures were incubated for 10 min at 95° C. for DNA denaturation, followed by 30 cycles of 1 min at 95° C., 0.5 min at 58° C. and elongation at 72° C. for 1 min per kbp of expected amplicon size. The reaction was terminated with an extended elongation step for 10 min at 72° C. The resulting products were separated on 1% agarose gels by electrophoresis and visualized using ethidium bromide.

(13) Genes encoding LPS biosynthesis enzymes of different bacteria were amplified by PCR from bacterial stocks and cloned into broad host-range expression vector pMMB67EH. To this end, PCR products and plasmid pMMB67EH-PagL.sub.(Pa) were purified using the Clean-Up System and Plasmid Extraction kit, respectively, both provided by Promega. Purified plasmid and PCR products were digested with the restriction enzymes (Fermentas, The Netherlands) for which sites were included in the primers (SEQ ID NOs: 7-24, 26 and 27 see Table 2) and subsequently ligated together. To knock out the chromosomal lpxA and lpxL genes, the plasmids were used.

(14) E. coli DH5a was transformed with ligation products or plasmids following standard protocols. Correct clones were elected by PCR, and plasmids were purified and sequenced at the Macrogen sequencing service (Amsterdam). Then, plasmids were transferred to E. coli strain SM10λpir by transformation and subsequently to B. pertussis strain B213 by conjugation using ampicillin and nalidixic acid for selection and counter selection, respectively. To generate chromosomal mutations, the knockout plasmids, which contained a rpsL gene conferring streptomycin sensititvity (Skorupsky and Taylor, 1996) were integrated into the chromosome by single crossover by selecting for kanamycin- or gentamicin-resistant transconjugants; the resulting bacteria had lost streptomycin resistance. Subsequently, to select for plasmid loss by a second crossover, bacteria were cultured in liquid medium and mutants were selected on plates with streptomycin and kanamycin or gentamicin. The presence of the plasmids in B. pertussis transconjugants and the proper generation of knockout mutants were verified by PCR.

(15) To express the target enzyme LpxD.sub.Pa in B. pertussis, vector pMMB67EH was used lpxD was amplified by PCR from P. aeruginosa strain PAO1 using a proof—reading enzyme (High Fidelity Polymerase, Roche Diagnostics GmbH) with primers LpxD-Fw Pa Ndel (having SEQ ID NO:26) and LpxD-rev-His Pa HindIII (having SEQ ID NO:27). The primers both contain sequences for restriction enzymes to facilitate cloning and the reverse primer also contains a sequence encoding a Hiss-tag to facilitate the detection of the recombinant protein via western blotting. After cloning the PCR product behind the tac promoter on pMMB67EH, the correct sequence of the insert was confirmed.

(16) RNA Extraction and RT-PCR

(17) To obtain RNA, cells from exponentially growing cultures were collected by centrifugation for 10 min at 5000 rpm in an Eppendorf Centrifuge 5424, adjusted to an OD.sub.550 of 4, and resuspended in trizol (Invitrogen, U.K.). Then, 200 μl of chloroform were added per ml of trizol, followed by centrifugation at 5000 rpm for 30 min. The resulting upper layer was mixed with an equal amount of ice-cold 75% ethanol. Next, RNA was isolated using the Nucleospin RNA II kit (Macherey-Nagel, U.S.A.) according to the manufacturer's instructions. The resulting solution was treated with Turbo DNA free (Ambion, Germany) for 1 h at 37° C. to remove genomic DNA followed by DNase inactivation according to recommendations of the manufacturer to generate pure RNA. This was used immediately to generate cDNA using the Transcriptor High Fidelity cDNA Synthesis Kit (Roche, The Netherlands). RNA, cDNA and chromosomal DNA were used as templates in PCRs with primers (see Table 2, SEQ ID NOs: 7-24, 26 and 27) to determine the generation of specific transcripts.

(18) Electrophoretic Techniques

(19) Whole cell lysates were adjusted to an OD.sub.600 of 5.0 solubilized in 1:1 in double-strength sample buffer and heated for 10 min at 100° C. For LPS visualization, after boiling, whole cell lysates were treated with proteinase K during 1 h at 37° C. Proteins and LPS were separated on 14% and 16% acrylamide gels, respectively, after which they were stained with Coomassie brilliant blue G250 or silver stain, respectively.

(20) LPS Purification and Analysis

(21) LPS were extracted from bacteria with hot phenol-water (Westphal, 1965) and purified further by solid phase extraction (SPE) on C8 reversed-phase cartridges. Briefly, bacteria were collected from culture suspensions by centrifugation, suspended with water at 70° C. and mixed with 0.8 volumes of phenol at the same temperature. After separating the aqueous and phenolic phase by centrifugation, the aqueous phase was prepared for SPE by adding one volume of 0.356 M triethylammonium acetate (TEAA) pH 7 (solvent A) and ⅓ volume of 2-propanol:water:triethylamine:acetic acid (70:30:0.03:0.01, v/v) pH 8.9 (solvent B). LPS extracts were purified simultaneously by SPE on reversed-phase Sep-Pak C8 cartridges (3 ml syringe-barrel-type Vac cartridge, 200 mg of C8 resin, Waters) using a 20-position vacuum manifold (Waters). Cartridges were conditioned for SPE by applying consecutively 3×1 ml of solvent B, 2-propanol:water:triethylamine:acetic acid (10:90:0.03:0.01, v/v) pH 8.9 (solvent C), 0.07 mM TEAA pH 7 (solvent D) and solvent A under vacuum. Then, samples were loaded into the cartridges and each cartridge was washed with 3×1 ml of solvents A, D and C, in this order. LPS were eluted from the columns by applying 2×0.3 ml of solvent B. Eluates were dried in a centrifugal vacuum concentrator and suspended in water. The purity and integrity of purified samples were judged by Tricine-SDS-PAGE combined with LPS silver and Coomassie staining. For analysis of lipid A structure, negative-ion nano-electrospray ionization—Fourier transform-mass spectrometry (nano-ESI-FT-MS) of purified LPS was performed on an LTQ Orbitrap XL instrument (Thermo Scientific). LPS samples were dissolved in a mixture of 2-propanol, water and triethylamine (50:50:0.001, vol/vol/vol) pH 8.5 and infused into the mass spectrometer by nano-ESI using gold-coated pulled glass capillaries, as described previously (Pupo et al, 2014) (Kondakov, A., and Lindner, B. (2005) Structural characterization of complex bacterial glycolipids by Fourier transform mass spectrometry. Eur J Mass Spectrom (Chichester, Eng) 11, 535-546). The spray voltage was set to −1.2 kV and the temperature of the heated capillary to 250° C. Under these ionization conditions no appreciable fragmentation of LPS was produced. To record lipid A mass spectra, nano-ESI-FT-MS of LPS was performed with in-source collision-induced dissociation (CID) at a potential difference of 100 V. In-source CID under this setting produced intense fragment ions corresponding to intact lipid A domains, which originate from the rupture of the labile linkage between the non-reducing lipid A glucosamine and Kdo, with minimal lipid A fragmentation, as shown in the mass spectra of the lipid A of the wild-type B213 strain (FIG. 3a).

(22) Eukaryotic Cell Lines Culture and Stimulation

(23) Human NF-κB/SEAP reporter HEK293 cells transfected either with human or mouse TLR4 in combination with MD-2 and CD14 were purchased from InvivoGen. Both cell lines contain an NF-κB-inducible secreted embryonic alkaline phosphatase (SEAP) reporter gene, which is expressed after TLR signaling. The cells were grown in HEK-Blue culture medium as described before [19]. SEAP was detected in culture supernatants after adding the substrate Quanti-Blue (InvivoGen). The human monocytic cell line MonoMac6 (MM6; DSMZ) was grown in Iscove's modified Dulbecco's medium (IMDM; Gibco) supplemented with 10% heat-inactivated FCS, 50 U/ml penicillin, and 50 μg/ml streptomycin. All cell lines were cultured at 37° C. in a 5% saturated CO.sub.2 atmosphere.

(24) For TLR4 signaling, HEK-Blue cell lines (2.5×10.sup.4) were incubated with serial dilutions of purified LPS or heat-inactivated whole cell preparations in a 96-well plate. After 2, 4 or 6 h of incubation at 37° C., supernatants were collected and incubated for 2 h with Quanti-Blue substrate and the OD.sub.630 was measured using an enzyme-linked immunosorbent assay (ELISA) reader.

(25) Results

(26) Expression of Heterologous LpxLs and LpxAs in B. pertussis

(27) To modify the length of the primary acyl chains at the 3 and 3′ positions and of the only secondary acyl chain in B. pertussis lipid A, we made use of LpxA and LpxL acyl transferases from other bacteria. B. pertussis lipid A contains 3OH—C10 and 3OH—C14 chains at the 3 and 3′ positions, respectively (FIG. 1a). We investigated whether the replacement of LpxA by the corresponding enzymes from Neisseria meningitidis (LpxA.sub.(Nm)) or Pseudomonas aeruginosa (LpxA.sub.(Pa)) would result in modifications of the acyl chains. Similarly, we investigated whether substitution of LpxL of B. pertussis by the corresponding enzymes from Porphyromonas gingivalis (LpxL.sub.(Pg)) or N. meningitidis (LpxL.sub.(Nm)) would result in a modification of the acyl chains.

(28) The genes for the heterologous enzymes were cloned into the broad host-range expression vector pMMB67EH under the control of the tac promoter. The expression of the recombinant enzymes LpxA and LpxL was first evaluated in the cloning host E. coli BL21(DE3) by RT-PCR (data not shown). These assays confirmed the presence of transcripts of the genes of interest when the bacteria were grown with IPTG whilst these transcripts were much less abundant or undetectable when the bacteria were grown in the absence of IPTG. Protein expression was also detected by SDS-PAGE; LpxA proteins were expressed in higher abundance than LpxL proteins (data not shown).

(29) The plasmids were then transferred to B. pertussis strain B213, a derivative of strain Tohama I (Table 1). Surprisingly, although pMMB67EH-LpxA.sub.(Nm) was successfully introduced in B. pertussis as evidenced by PCR, the transconjugants failed to grow after being replated on plates containing ampicillin for plasmid maintenance and no IPTG. Thus, apparently, even uninduced expression levels of LpxA from N. meningitidis are lethal in B. pertussis, although this is not the case in E. coli. The LPS predicted to be produced by the introduction of the Lpx.sub.(Nm) enzyme is depicted in FIG. 1B. We noticed also that expression of LpxL.sub.(Nm) impaired growth (Fig S1C). All other recombinant strains grew as the wild-type (FIG. 2 (S1C)).

(30) Expression of the recombinant protein LpxD.sub.(Pa) was first tested in Escherichia coli strain BL21(DE3), to which the plasmid was transferred. The resulting strain is called BL21-pLpxD.sub.Pa. Addition of IPTG to the culture did not affect growth (data not shown). Expression of the protein was not detected on regular Coomassie blue-stained gels on which whole cell lysates of BL21(DE3) and BL21-pLpxD.sub.Pa were analyzed. However, Western blotting assays showed a reaction of an anti-Hiss-tag antibody with a band of the expected size of LpxD.sub.Pa (36.4 kDa) in BL21-pLpxD.sub.Pa (data not shown), which was not present in the sample of BL21(DE3). Next, the plasmid was transferred to B. pertussis B213 by conjugation using E. coli strain SM10λpir as donor, and two transconjugants were saved. In the presence of 1 mM IPTG, both clones showed a growth defect as compared with the parent (FIG. 2B), which was more pronounced for clone 5. Western blotting assays confirmed the expression of the enzyme in both clones (data not shown), while no differences in expression levels were observed.

(31) Analysis of Recombinant Lipid a Structures

(32) The lipid A structures were then analyzed by nano-ESI-MS using purified LPS extracted from whole cells grown in the presence of 1 mM IPTG for 12 h (logarithmic growth). For the wild-type strain, a major peak was observed at m/z 1557.97 that corresponds with the expected bis-phosphorylated penta-acetylated lipid A (FIG. 3A). In the strain expressing LpxA.sub.(Pa), the spectrum revealed, besides the ion at m/z 1557.97, two additional ions at m/z 1501.91 and 1529.94 (FIG. 3B). The ion at m/z 1501.91 corresponds with a substitution of the primary 3OH—C14 acyl chain at the 3′ position by 3OH—C10 (FIG. 1C), whilst the m/z 1529.94 ion indicates the presence of a hydroxylated fatty acid with an intermediary C12 chain length. The relative abundance of the two new species was only 48 and 75% relative to the wild-type structure at m/z 1557.97, which could be due to the expression of the endogenic lpxA on the chromosome. Therefore, we decided to knock out the chromosomal lpxA copy. MS analysis of the resulting strain evidenced the complete loss of the m/z 1557.97 ion and a drastic decrease of the abundance of the m/z 1529.94 ion leaving a major peak of m/z 1501.91 corresponding to the substitution (FIG. 1C and FIG. 3C).

(33) MS analysis of B213-LpxL.sub.(Nm) detected the wild-type m/z 1557.97 ion as a minor species, whilst a major peak of m/z 1529.94 corresponded with a substitution of the secondary C14 acyl chain by C12 (FIG. 1D and FIG. 3D). Attempts to delete the chromosomal lpxL failed. B. pertussis contains two adjacent lpxL homologues on the chromosome, but only one, called lpxL2, is active under laboratory growth conditions [20]. Different constructs were used to delete the lpxL2 gene partially or completely; however, in spite of considerable efforts, we could not obtain the desired knockout.

(34) Apparently, the lpxL2 gene sequence rather than the enzyme is essential for B. pertussis considering that expression of LpxL.sub.(Nm) in the wild-type strain already altered about 65% of the lipid A structure. This could be due to a polar effect of lpxL2 disruption on expression of the downstream gene dapF, which encodes for L,L-DAP epimerase that catalyzes L,L-diaminopimelate (DAP) into meso-DAP Meso-DAP is vital for cell wall synthesis and lysine biosynthesis. However, also attempts to inactivate the lpxL.sub.2 gene in the presence of meso-DAP failed. MS analysis of B213-LpxL.sub.(Pg) evidenced a drastic reduction in the abundance of the m/z 1557.98 ion and the appearance of a new peak of m/z 1586.01 that corresponds with a substitution of the C14 by a C16 chain (FIG. 1E and FIG. 2e). In summary, heterologous expression of LpxA.sub.(Pa), LpxL.sub.(Nm), and LpxL.sub.(Pg) in B. pertussis resulted in LPS alterations as depicted in FIG. 1.

(35) Analysis of B213-pLpxD.sub.(Pa) clones 4 and 5 revealed that in both mutants, ions at m/z 1557.97 were found corresponding with the standard penta-acylated lipid A as found in the wild-type (FIGS. 3F and 3G). In addition, abundant ions at m/z 1529.94 and 1501.91 were found corresponding with the reduction of the length of one or two acyl chains from 3OH—C14 to 3OH—C12, respectively (FIGS. 3F and 3G). The abundance of these additional major ions varied between both clones; the relative abundance of the peak at m/z 1529.94 was lower in clone 5 than in clone 4, and vice versa for the peak at m/z 1501.91. Thus, clone 5 had a higher amount of lipid A molecules with an entire modification of the length of both acyl chains in the lipid A than clone 4, but considerable amounts of unaltered lipid A remained in both cases. These results show that the expression of LpxD.sub.Pa modified the structure of lipid A as shown in FIG. 1.

(36) Differential Activation of TLR4 by the LPS Variants

(37) We next investigated whether the altered structure affects the toxicity of the LPS. To this end, purified LPS preparations were added to cultures of HEK293 cells expressing the human or mouse TLR4 complex (hTLR4 and mTLR4, respectively). After exposure, the activation of the receptor was evaluated by the expression of a reporter gene (FIG. 4). Interestingly, LPS from B213-pLpxA.sub.(Pa) stimulated hTLR4 much less than LPS from the wild-type strain (FIG. 4A). The residual activation still detected was due to the expression of the chromosomal lpxA gene, since it was totally eliminated after inactivation of this gene (FIG. 4A). Thus, the length of the primary acyl chain at the 3′ position is relevant for the activation of hTLR4 by pertussis LPS. LPS from B213-pLpxL.sub.(Nm) and B213-pLpxL.sub.(Pg) reduced and increased hTLR4 activation, respectively (FIG. 4A). Hence, stimulation of hTLR4 correlates with the length of the secondary acyl chain in the order C16>C14>C12.

(38) To evaluate the toxicity of the altered LPS of the strains expressing LpxD.sub.(Pa), preparations of purified LPS from both mutants and the wild-type were added to cultures of HEK-Blue cells expressing the human TLR4 receptor (hTLR4) (InvivoGen). As a negative control, we used purified LPS from strain B213ΔlpxA-pLpxA.sub.(Pa), which contains a 3OH—C10 acyl chain instead of 3OH—C14 at the 3′ position and did not activate hTLR4. The activation of the receptor was evaluated by the expression of a SEAP reporter gene after 4 h of exposure, and the results are presented in FIG. 5. LPS from B213-pLpxD.sub.(Pa) clone 4 and clone 5 showed considerably lower activity than wild-type LPS. The SEAP activity of cells stimulated with LPS from clone 5 was as low as that of non-stimulated cells, while cells stimulated with LPS from clone 4 showed low residual activity, perhaps in agreement with a somewhat less efficient modification of the acyl chains in this clone (FIG. 3). Cells stimulated with LPS from B213ΔlpxA-pLpxA.sub.(Pa) showed even lower SEAP activity than non-stimulated cells.

(39) Stimulation of HEK293 cells expressing mTLR4 with LPS preparations from wild-type strain B213 resulted in a stronger response than observed in the cells expressing hTLR4 (compare FIGS. 4A and B). LPS preparations from B213 cells expressing the heterologous enzymes were slightly less effective in stimulating these cells (FIG. 4B). However when the chromosomal lpxA gene was inactivated in B213 expressing LpxA.sub.(Pa), the resulting LPS did not activate mTLR4 at all (FIG. 4B). Thus, the human and mouse TLR4 are activated differently by modified pertussis LPS, but the length of the primary acyl chain at the 3′ position is critical in both cases. It was reported previously that the decreased toxicity of B. pertussis LPS that had lost the primary acyl chain at the 3 position was nullified in whole-cell preparations by its increased release from the membranes [2]. Taking this into account, we wished to determine the biological activity of whole-cell preparations. Expression of heterologous LPS enzymes in strain B213 affected the stimulation of HEK293 cells expressing hTLR4 similarly in whole-cell and purified LPS preparations (compare FIGS. 4A and C). Stimulation of HEK293 cells expressing mTLR4 by whole-cell preparations was barely affected by the expression of the heterologous enzymes in B213 (FIG. 4D). However, whole-cell preparations of the ΔlpxA mutant of B213 expressing LpxA.sub.(Pa) failed to activate these cells (FIG. 4D).

DISCUSSION

(40) Their reactogenicity has led to the replacement of whole-cell pertussis vaccines by subunit vaccines, which, however, do not provide satisfactory protection. The development of new, less reactogenic whole-cell vaccines could offer a solution. LPS is, to a considerable extent, responsible for the toxicity of the cellular pertussis vaccines [2]. In the present study, we investigated if modification of acyl chain length in B. pertussis lipid A could result in reduced toxicity. We altered the length of primary acyl chain at the 3′ position and of the secondary acyl chain at 2′ position by expression of heterologous LpxA and LpxL acyltransferases. We found that reduction in the length of both acyl chains resulted in a drastic decrease in LPS toxicity (FIG. 4).

(41) More precisely, substitution of a 3OH—C10 acyl chain for the 3OH—C14 chain present at the 3′ position of lipid A abolished endotoxicity. In addition, substitution of the secondary C14 chain attached to the primary acyl chain at the 2′ position by a C12 reduced endotoxicity. Consistently, the endotoxicity increased when this chain was substituted by a C16.

(42) It was further observed that the reduction of the length of the acyl chains at the 2 and 2′ positions of B. pertussis lipid A also has a large impact on the activation of hTLR4. Considering that the effect on toxicity is obtained independent of the position of the shorter acyl chain, the total volume of the hydrophobic moiety of the lipid A molecule is apparently important for the proper binding to and activation of the hTLR4 complex. The novel LPS species generated seem to function as hTLR4 antagonist as they could deplete any hTRL4 response (FIG. 5) even in the presence of considerable amounts of wild-type LPS (FIG. 3). This may not to be the case for LPS with a shorter secondary acyl chain as LPS from B213-pLpxL.sub.(Nm) showed residual activity in activating hTLR4 even though it contained lower amounts of the wild-type LPS than the purified LPS preparations from the strains expressing LpxA.sub.(Pa) or LpxD.sub.(Pa). It should be noted that expression of LpxD.sub.(Pa) in B. pertussis caused growth defects. Similarly, our previous results showed a growth defect of B. pertussis expressing LpxL.sub.(Nm).

(43) In summary, our results demonstrate the importance of the acyl-chain length for activation of the immune system and for endotoxicity of B. pertussis LPS.

(44) Our results also revealed a different effect of the LPS modifications on activation of mouse and human TLR4, where, in most cases, mTLR4 was less sensitive to the modifications. Previous studies reported species-dependent differences regarding TLR4 activation [24; 25]. These differences are explicable by interspecies variation in MD-2 and TLR4. These differences limit extrapolation of data from experimental animals to humans in vaccine trials [25]. Importantly, however, the LPS of the lpxA knockout mutant of strain B213 expressing LpxA.sub.(Pa) failed to activate both mTLR4 and hTLR4 in vitro allowing for extrapolation of results of planned experiments in mice to humans.

(45) It is remarkable that the acyl chains at the 3 and 3′ positions of B. pertussis lipid A differ in length [4]. LpxA catalyzes the first reaction in the lipid A biosynthetic pathway by transferring an acyl chain of a specific length onto the 3 position of GlcNAc in UDP-GlcNAc. The exact length of this acyl chain is defined by a hydrocarbon ruler in LpxA [26]. Later in the pathway, LpxH removes UMP in a proportion of the population of UDP-diacylglucosamine (UDP-DAG) precursors generating lipid X, after which LpxB links a UDP-DAG and a lipid X molecule generating a mono-phosphorylated, tetra-acylated glucosamine disaccharide in which the acyl chains at positions 3 and 3′ are both derived from the original acylation by LpxA and, therefore, usually identical. Only rarely, LPS species with different acyl chain length at the 3 and 3′ positions are found in nature. Consistent with the different acyl-chain length, expression studies in E. coli showed that B. pertussis LpxA has reduced chain-length specificity, but acyl chains of various lengths were incorporated at both the 3 and 3′ positions [27]. Thus, the impeccable asymmetry in B. pertussis lipid A must be explained by chain-length specificity of an enzyme downstream in the pathway, which, we hypothesize, is LpxH. In our work, the expression of LpxA.sub.(Pa) resulted in two 3OH—C10 chains at positions 3 and 3′, which was tolerated. However, the expression of LpxA.sub.(Nm), which would result in two primary 3OH—C12 chains at these positions (FIG. 1), appeared to be lethal. This can be explained if LpxH of B. pertussis can remove UMP only from UDP-DAG molecules containing a short 3OH—C10 chain at the 3 position. Indeed when we expressed LpxH.sub.(Nm) in B. pertussis the asymmetry disappeared, confirming our LpxH hypothesis (data not shown). Similarly, the heterologous expression of both LpxA.sub.(Nm) and LpxH.sub.(Nm) resulted in viable cells (data not shown). Hence to obtain a modified Bordetella lipid A moiety having an acyl-chain at the 3-position that is longer than 3OH—C10, may (in addition to a modified acyl transferase) require the presence of a modified LpxH, such as LpxH.sub.(Nm).

(46) In conclusion, our approaches to reduce the toxicity of whole-cell B. pertussis vaccines by lipid A engineering as disclosed herein were effective. Our results show that the endotoxic activity of B. pertussis LPS is largely determined by the length of its fatty acyl chains. For the first time, we succeeded to engineer a strain that is totally devoid of endotoxic activity in in vitro assays. Importantly, this LPS did also not activate mTLR4 in vitro allowing for extrapolation of data obtained in planned animal studies to humans. Hence, our findings will allow for the generation of new cellular vaccines for B. pertussis and other pathogens.

(47) TABLE-US-00002 TABLE 2A SEQ ID NOs and corresponding protein and organism SEQ ID NO Protein Organism* 1 LpxA Pa (PA01) 2 LpxL Nm (H44/76) 3 LpxL Pg (ATCC33277) 4 LpxD Pa (PA01) 5 LpxH Nm (H44/76) 6 LpxA Nm (H44/76) 25 PagL Bb and Bp (GenBank WP-003813842.1) 28 LpxA Bpe (GenBank: CAE41721.1) 29 LpxD Bpe (GenBank: CAE41719.1) 30 LpxH Bpe (Genbank: CAE42187.1) 31 LpxL Bpe (Genbank: CAE43342.1) 32 LpxL Pa (Genbank: AAG06812.1) *Pa = Pseudomonas aeruginosa, Nm = Neisseria meningitidis, Pg = Porphyromonas gingivalis, Bb = B. bronchiseptica, Bp = Bordetella parapertussis, Bpe = Bordetella pertussis

(48) TABLE-US-00003 TABLE 2B SEQ ID NOs and primer names SEQ ID NO Primer name Obtained product 7 LpxA.sub.(Nm) Fw pMMB67EH-LpxA.sub.(Nm) 8 LpxA.sub.(Nm) Rev 9 LpxA.sub.(Pa) Fw pMMB67EH-LpxA.sub.(Pa) 10 LpxA.sub.(Pa) Rev 11 LpxL.sub.(Nm) Fw pMMB67EH-LpxL.sub.(Nm) 12 LpxL.sub.(Nm) Rev 13 LpxL.sub.(Pg) Fw pMMB67EH-LpxL.sub.(Pg) 14 LpxL.sub.(Pg) Rev 15 LpxA.sub.(Nm) Fw RT lpxA.sub.(Nm) 16 LpxA.sub.(Nm) Rev RT 17 LpxA.sub.(Pa) Fw RT lpxL.sub.(Pa) 18 LpxA.sub.(Pa) Rev RT 19 LpxL.sub.(Nm) Fw RT lpxL.sub.(Nm) 20 LpxL.sub.(Nm) Rev RT 21 LpxL.sub.(Pg) Fw RT lpxL.sub.(Pg) 22 LpxL.sub.(Pg) Rev RT 23 Amp Fw RT amp 24 Amp Rev RT 26 LpxD.sub.(Pa) FW pMMB67EH-LpxD.sub.(Pa) 27 LpxD.sub.(Pa) Rev-His

Example 2

(49) lpxD.sub.(Pa) and lpxA.sub.(Pa) LPS Mutants Show Reduced Pyrogenicity in Rabbits

(50) Bordetella pertussis mutants were constructed with an altered lipid A moiety in their LPS through heterologous expression of lpxA and lpxD genes from Pseudomonas aeruginosa. Specifically, B. pertussis B1917 strains were constructed wherein either the chromosomal lpxA gene or the lpxD gene was replaced with the corresponding P. aeruginosa versions. In both cases, this resulted in the synthesis of LPS with the expected shortened acyl chains, as shown by mass spectrometry.

(51) In order to test the effect of these alterations in vivo, a rabbit pyrogenicity test was conducted with LPS purified from the lpxA mutant and from the lpxD mutant, and with OMVs extracted from the lpxD mutant, all in comparison to the wildtype. OMV (nOMV) were extracted by detergent-free extraction of the bacterial biomass with EDTA as chelating agent, essentially as described by van de Waterbeemd et al. (2010, Vaccine, 28(30):4810-6).

(52) New Zealand White rabbits were injected intramuscularly with 0.5 ml of solution containing nOMVs (50 μg of protein) or purified LPS (10 μg), and acellular pertussis vaccine and saline as controls. The following groups were used (5 animals per group): 1. Vehicle Control (saline) 2. Reference vaccine (acP) 3. Vaccine 1: B1917 nOMV ompA prn 4. Vaccine 2: B1917 nOMV ompA prn lpxD 5. Vaccine 3: B1917 LPS lpxD 6. Vaccine 4: B1917 LPS lpxA 7. Vaccine 5: B1917 LPS wildtype
Body temperature was measured using an external scanner from subcutaneously implanted transponders, at 0.5, 1, 2, 4, 6, 24 and 48 hrs after injection. The results are shown in Table 3 and in FIG. 6.
Results

(53) A statistically significant rise in body temperature is seen with vaccine 1 (1, 2 and 4 h after injection) and vaccine 5 (4 h after injection). With purified LPS, there is a clear fever peak induced by the wildtype, but not by the lpxD and lpxA mutants. With OMVs, there is a more prolonged period of fever, both for wildtype and lpxD mutant, but lower for the latter. This is to be expected, as OMVs contain other pyrogenic components in addition to LPS.

CONCLUSIONS

(54) The data demonstrate that mutant Bordetella LPS having a lipid A moiety wherein the length of at least one acyl chain is shorter as compared to the lipid A moiety of a wild-type Bordetella show a clearly reduced pyrogenicity in rabbits. The above observed in vitro data with HEK cells expressing TLR4 are therefore corroborated by these in vivo data.

(55) TABLE-US-00004 TABLE 3 Data of pyrogenicity study in rabbits. TempFirstInj Temp (sc) Temp (sc) Temp (sc) Temp (sc) Temp (sc) Temp (sc) Temp (sc) pretreat first inj first inj first inj first inj first inj first inj first inj (° C.) (° C.) (° C.) (° C.) (° C.) (° C.) (° C.) (° C.) [G] [C] [C1] [C] [C1] [C] [C] [C] Sex: Male 0 (PreDos) 0 (05hPtD) 0 (1hPstD) 0 (2hPstD) 0 (4hPstD) 0 (6hPstD) 1 (24hPtD) 2 (48hPtD) Vehicle Mean 38.52 38.40 39.00 38.44 38.18 38.40 38.50 38.48 SD 0.60 0.42 0.32 0.59 0.61 0.20 0.60 0.28 N 5 5 5 5 5 5 5 5 Reference Mean 38.92 38.72 37.98* 38.66 38.84 38.72 38.38 38.28 vaccine SD 0.26 0.36 0.89 0.17 0.29 0.46 0.56 0.61 N 5 5 5 5 5 5 5 5 Vaccine 1 Mean 37.90 38.34 39.78* 39.64* 40.18** 39.00 38.98 38.28 SD 0.69 0.38 0.38 0.40 0.70 0.95 0.48 0.44 N 5 5 5 5 5 4 5 5 Vaccine 2 Mean 38.20 38.52 39.32 39.52 39.54 38.76 38.76 38.08 SD 1.14 0.53 0.48 0.77 1.27 0.86 0.78 0.77 N 5 5 5 5 5 5 5 5 Vaccine 3 Mean 38.06 37.84 39.20 38.76 39.38 38.22 37.96 38.36 SD 0.48 0.69 0.32 0.62 1.18 0.46 0.68 0.36 N 5 5 5 5 5 5 5 5 Vaccine 4 Mean 38.06 38.46 38.68 38.90 38.54 38.48 38.12 38.38 SD 0.86 0.31 0.24 0.59 0.51 0.54 0.45 0.41 N 5 5 5 5 5 5 5 5 Vaccine 5 Mean 38.50 38.16 38.64 39.40 40.46** 38.76 38.74 38.34 Pos. cntrl SD 0.70 1.04 0.81 1.11 0.76 0.63 0.35 0.59 N 5 5 5 5 5 5 5 5 [G] - Ancova/Anova & Dunnett [C] - Ancova/Anova & Dunnett {Covariate: Temp (Sc) FirstInj (pretreat)}: *= p < 0.05 [C1] - Ancova/Anova & Dunnett(Rank) {Covariate: Temp (Sc) FirstInj (pretreat)}: *= p < 0.05; **= p < 0.01

REFERENCES

(56) [1] Clark T A. Changing pertussis epidemiology: everything old is new again. J Infect Dis 2014 Apr. 1; 209(7):978-81. [2] Geurtsen J, Steeghs L, Hamstra H J, et al. Expression of the lipopolysaccharide-modifying enzymes PagP and PagL modulates the endotoxic activity of Bordetella pertussis. Infect Immun 2006 October; 74(10):5574-85. [3] Peppler M S. Two physically and serologically distinct lipopolysaccharide profiles in strains of Bordetella pertussis and their phenotype variants. Infect Immun 1984 January; 43(1):224-32. [4] Raetz C R, Whitfield C. Lipopolysaccharide endotoxins. Annu Rev Biochem 2002; 71:635-700. [5] Rietschel E T, Schade U, Jensen M, Wollenweber H W, Luderitz O, Greisman S G. Bacterial endotoxins: chemical structure, biological activity and role in septicaemia. Scand J Infect Dis Suppl 1982; 31:8-21. [6] Palsson-McDermott E M, O'Neill L A. Signal transduction by the lipopolysaccharide receptor, Toll-like receptor-4. Immunology 2004 October; 113(2):153-62. [7] Alexander C, Rietschel E T. Bacterial lipopolysaccharides and innate immunity. J Endotoxin Res 2001; 7(3):167-202. [8] Conti P, Dempsey R A, Reale M, et al. Activation of human natural killer cells by lipopolysaccharide and generation of interleukin-1 alpha, beta, tumour necrosis factor and interleukin-6. Effect of IL-1 receptor antagonist. Immunology 1991 August; 73(4):450-6. [9] Cinel I, Dellinger R P. Advances in pathogenesis and management of sepsis. Curr Opin Infect Dis 2007 August; 20(4):345-52. [10] Crowell D N, Anderson M S, Raetz C R. Molecular cloning of the genes for lipid A disaccharide synthase and UDP-N-acetylglucosamine acyltransferase in Escherichia coli. J Bacteriol 1986 October; 168(1):152-9. [11] Coleman J, Raetz C R. First committed step of lipid A biosynthesis in Escherichia coli: sequence of the lpxA gene. J Bacteriol 1988 March; 170(3):1268-74. [12] Loppnow H, Brade H, Durrbaum I, et al. IL-1 induction-capacity of defined lipopolysaccharide partial structures. Expression of foreign LpxA acyltransferases in Neisseria meningitidis results in modified lipid A with reduced toxicity and retained adjuvant activity. Cell Microbiol 2002 September; 4(9):599-611. [14] Wyckoff T J, Lin S, Cotter R J, Dotson G D, Raetz C R. Hydrocarbon rulers in UDP-N-acetylglucosamine acyltransferases. J Biol Chem 1998 Dec. 4; 273(49):32369-72. [15] Raetz C R, Reynolds C M, Trent M S, Bishop R E. Lipid A modification systems in gram-negative bacteria. Annu Rev Biochem 2007; 76:295-329. [16] Verwey W F, Thiele E H, Sage D N, Schuchardt L F. A SIMPLIFIED LIQUID CULTURE MEDIUM FOR THE GROWTH OF HEMOPHILUS PERTUSSIS. J Bacteriol 1949 August; 58(2):127-34. [17] Westphal O, K. Jann. Bacterial lipopolysaccharides extraction with phenol-water and further applications of the procedure. Methods carbohydrates chemistry 1965; 5:83-91. [18] Pupo E, Hamstra H J, Meiring H, van der Ley P. Lipopolysaccharide engineering in Neisseria meningitidis: structural analysis of different pentaacyl lipid A mutants and comparison of their modified agonist properties. J Biol Chem 2014 Mar. 21; 289(12):8668-80. [19] Brummelman J, Veerman R E, Hamstra H J, et al. Bordetella pertussis naturally occurring isolates with altered lipooligosaccharide structure fail to fully mature human dendritic cells. Infect Immun 2015 January; 83(1):227-38. [20] Geurtsen J, Angevaare E, Janssen M, et al. A novel secondary acyl chain in the lipopolysaccharide of Bordetella pertussis required for efficient infection of human macrophages. J Biol Chem 2007 Dec. 28; 282(52):37875-84. [21] Park B S, Song D H, Kim H M, Choi B S, Lee H, Lee J O. The structural basis of lipopolysaccharide recognition by the TLR4-MD-2 complex. Nature 2009 Apr. 30; 458(7242):1191-5. [22] DeMarco M L, Woods R J. From agonist to antagonist: structure and dynamics of innate immune glycoprotein MD-2 upon recognition of variably acylated bacterial endotoxins. Mol Immunol 2011 October; 49(1-2):124-33. [23] Cunningham M D, Seachord C, Ratcliffe K, Bainbridge B, Aruffo A, Darveau R P. Helicobacter pylori and Porphyromonas gingivalis lipopolysaccharides are poorly transferred to recombinant soluble CD14. Infect Immun 1996 September; 64(9):3601-8. [24] Akashi S, Nagai Y, Ogata H, et al. Human MD-2 confers on mouse Toll-like receptor 4 species-specific lipopolysaccharide recognition. Int Immunol 2001 December; 13(12):1595-9. [25] Steeghs L, Keestra A M, van Mourik A., et al. Differential activation of human and mouse Toll-like receptor 4 by the adjuvant candidate LpxL1 of Neisseria meningitidis. Infect Immun 2008 August; 76(8):3801-7. [26] Smith E W, Zhang X, Behzadi C, Andrews L D, Cohen F, Chen Y. Structures of Pseudomonas aeruginosa LpxA Reveal the Basis for Its Substrate Selectivity. Biochemistry 2015 Sep. 29; 54(38):5937-48. [27] Sweet C R, Preston A, Toland E, et al. Relaxed acyl chain specificity of Bordetella UDP-N-acetylglucosamine acyltransferases. J Biol Chem 2002 May 24; 277(21):18281-90.