Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Jan 31.
Published in final edited form as: Bioorg Med Chem Lett. 2014 Dec 17;25(3):547–553. doi: 10.1016/j.bmcl.2014.12.024

Characterization of TRIF Selectivity in the AGP Class of Lipid A Mimetics: Role of Secondary Lipid Chains

Juhienah K Khalaf a,1, William S Bowen b, Hélène G Bazin a, Kendal T Ryter a, Mark T Livesay a, Jon R Ward a, Jay T Evans a, David A Johnson a,*
PMCID: PMC4330088  NIHMSID: NIHMS650673  PMID: 25553892

Abstract

TLR4 agonists that favor TRIF-dependent signaling and the induction of type 1 interferons may have potential as vaccine adjuvants with reduced toxicity. CRX-547 (4), a member of the aminoalkyl glucosaminide 4-phosphate (AGP) class of lipid A mimetics possessing three (R)-3-decanoyloxytetradecanoyl groups and D-relative configuration in the aglycon, selectively reduces MyD88-dependent signaling resulting in TRIF-selective signaling, whereas the corresponding secondary ether lipid 6a containing (R)-3-decyloxytetradecanoyl groups does not. In order to determine which secondary acyl groups are important for the reduction in MyD88-dependent signaling activity of 4, the six possible ester/ether hybrid derivatives of 4 and 6a were synthesized and evaluated for their ability to induce NF-κB in a HEK293 cell reporter assay. An (R)-3-decanoyloxytetradecanoyl group on the 3-position of the D-glucosamine unit was found to be indispensable for maintaining low NF-κB activity irrespective of the substitutions (decyl or decanoyl) on the other two secondary positions. These results suggest that the carbonyl group of the 3-secondary lipid chain may impede homodimerization and/or conformational changes in the TLR4–MD2 complex necessary for MyD88 binding and pro-inflammatory cytokine induction.

Keywords: Lipid A mimetic, TLR4 agonist, TRIF selectivity, NF-κB activity, AGP, glycolipids, vaccine adjuvant

Toll-like receptors (TLRs) are a family of pattern recognition receptors on innate immune cells that recognize pathogen-specific components of microbial invaders. The recognition of microbial ligands by TLR receptors triggers intracellular signaling via the cytoplasmic Toll-interleukin receptor (TIR) domain common to all TLRs, resulting in the release of pro-inflammatory cytokines, chemokines, and anti-microbial defensins, and in the expression of co-stimulatory molecules. Expression of these factors drives the innate immune response to infection as well as the recruitment and activation of antigen-presenting cells and effector B and T cells involved in adaptive immunity.1, 2

Lipopolysaccharide (LPS, endotoxin), the main cell surface component of Gram-negative bacteria, is the natural glycolipid ligand that binds TLR4 and its accessory molecule, MD-2, to form a stable TLR4–MD-2 receptor complex, triggering an initial innate immune response.3 Although cellular activation through the TLR4–MD-2 receptor is architecturally complex4 and involves many signaling elements, TLR4–MD-2 receptor signaling proceeds mainly through two intracellular pathways: the MyD88-dependent pathway, and the TRIF-dependent pathway, also known as the MyD88-independent pathway.5 Signaling through the MyD88-dependent pathway involves binding of two adaptor proteins, MyD88 and the TIR-adaptor TIRAP (Mal), to the cytoplasmic domain of TLR4 and induces early NF-κB activation and the release of pro-inflammatory cytokines such as TNF-α and IL-1β.6 The TRIF-dependent pathway, on the other hand, relies on cytoplasmic adaptor proteins TRIF (TIR domain-containing adapter inducing IFN-β) and the TRIF-related molecule TRAM, and induces later and lower levels of NF-κB activation, resulting in lower expression of mediators of inflammation and toxicity. The TRIF-dependent pathway also activates the nuclear translocation of the transcription factor IRF-3, resulting in expression of type I interferons (IFN-α/β) and IFN-inducible genes.7 IFN-dependent signaling downstream of TRIF, in turn, is involved in the up-regulation of major histocompatibility complex (MHC) and co-stimulatory molecules on dendritic cells, mediators of antigen stimulation and T-cell activation and proliferation that are crucial for an effective adaptive immune response to infectious agents and heterologous vaccine antigens.8

Although LPS is a potent stimulator of host defense systems via its interaction with the TLR4–MD-2 receptor complex, the pathophysiology of LPS and its active principle, lipid A, preclude their use as adjuvants in human vaccines. The toxicity of the LPS from Salmonella minnesota R595, however, has been ameliorated by the selective hydrolysis of certain groups, including the anomeric phosphate, to produce a TLR4-active product approved for human use known as monophosphoryl lipid A (MPL). MPL, which comprises several less highly acylated compounds in addition to the major, hexa-acyl component 1 (Figure 1), is an effective adjuvant in prophylactic and therapeutic vaccines and shows an excellent safety profile in humans.9 The reduced toxicity of MPL has been attributed to selective induction of the TRIF signaling pathway and MyD88-independent factors such as IP-10 and MCP-1 coupled with threshold levels of MyD88-dependent cytokines.10 By the same token, the lower virulence of some bacterial strains as well as the decreased toxicity of certain LPS molecules, including Salmonella LPS, has also been attributed to selective TRIF signaling.5, 11 However, the structural variability within individual lipid A or LPS preparations and the potential presence of other bioactive substances often make it difficult to draw definite conclusions about which structural features are responsible for a particular immune response. Thus, considerable effort has been directed towards the synthesis of not only individual natural lipid A components12 but also subunit analogs of lipid A in which the disaccharide backbone of lipid A has been replaced with a structural motif more amenable to systematic structure-activity relationship (SAR) and mechanism of action investigations.13, 14

Figure 1.

Figure 1

Open in a new tab

Synthetic and naturally derived lipid A mimetics

In the course of our own SAR studies on lipid A, we identified a new class of TLR4-active glycolipids known as aminoalkyl glucosaminide 4-phosphates (AGPs).15 The immunostimulatory activity of the AGP class of synthetic lipid A mimetics, which have the general structure 1 (Figure 1), was found to depend greatly on the length of the secondary lipid chain length (R1-R3) as well as the structure of the aglycon moiety.15, 16 Maximum TLR4 agonist activity in human in vitro models was observed with seryl-based AGPs (R4=CO2H, n=1) containing 10-carbon secondary acyl or alkyl groups (R1, R2, R3=decanoyl or decyl), whereas the corresponding seryl derivatives possessing 6-carbon secondary lipid chains were potent TLR4 antagonists in human systems.17 CD14, a protein involved in the shuttling of LPS to MD-2, was not required for MyD88-dependent agonist activity in the AGP series in vitro but did enhance responses, particularly for lower potency agonists.16 Site-directed mutagenesis studies18 and structural studies with secondary acyl hybrid AGP molecules16 pointed to the particular importance of secondary lipid chain R1 in determining TLR4 activity. These observations are consistent with a TLR4–MD-2–AGP complex in which the terminal methylene units of secondary lipid chain R1 of the AGP molecule interact with the dimerization interface created by hydrophobic patches on MD-2 and the TLR4 ectodomain to form a symmetrical ‘m’ shaped TLR4–MD-2–AGP homodimer.18 Dimerization of ligand-complexed TLR4–MD-2 is thought to be if not prerequisite to TLR4 activation19 at least promotive of more rapid signaling.20

Such an orientation of the AGP molecule in the MD-2 hydrophobic pocket with the R1 terminus interacting with the TLR4 ectodomain corresponds to that determined crystallographically for TLR4 antagonists eritoran (E5564) bound to a hybrid human TLR4–MD-2 heterodimer21 and lipid IVa bound to human MD-2,22 but is opposite (i.e., rotated 180 degrees) to that shown crystallographically for hexa-acyl E. coli LPS bound to the hybrid TLR4–MD-2 heterodimer, wherein the lipid chain amide-linked to the reducing sugar interacts with hydrophobic residues of both MD-2 and the TLR4 ectodomain at the dimerization interface.4 While the latter ‘agonist’ or ‘LPS-like’ orientation may be favored by crystallization conditions and/or the presence of a divalent counter ion, as well as by structural changes made to the TLR4 molecule to permit solubilization/co-crystallization,21 the above data suggest that the AGP class of lipid A mimetics bind in an ‘antagonist’ or ‘eritoran-like’ orientation to the human TLR4–MD-2 heterodimer to induce dimerization and signaling. Nonetheless, given the C2 symmetry of certain TLR4 agonists14 and antagonists,23 the pseudosymmetry of the receptor itself, as well as the striking effect of different metal counter ions on LPS activity,24 it is likely that some TLR4–MD-2 ligands modulate immune responses by inserting into MD-2 in both orientations.

Because MPL’s beneficial adjuvant effects have been associated with a bias toward the TRIF-dependent pathway, members of the AGP class of lipid A mimetics were also screened for differential induction of MyD88- and TRIF-dependent signaling pathways in the hope of identifying a TRIF-selective AGP for potential use as a vaccine adjuvant with improved safety and efficacy profiles. A comparison of seryl-based AGPs CRX-527 (3) and CRX-547 (4) possessing 10-carbon secondary acyl chains and differing only in the configuration of the seryl stereocenter (Figure 2) showed that the D-seryl-based AGP 4 (‘D-isomer’) induced significantly lower levels of MyD88-dependent cytokines relative to the L-isomer 3 in human primary PBMC-derived monocytes and dendritic cells but similar levels of TRIF-dependent chemokines.25 The relative responses of CRX-527 and CRX-547 in these cell-based assays correlated strongly with their MyD88-dependent NF-κB activity in a human embryonic kidney (HEK) cell based reporter assay, using either MD2/TLR4 or MD2/TLR4/CD14 receptor transfectants (data not shown; see also reference 25). The inverted configuration of the seryl carboxyl group, a bioisostere of the anomeric phosphate of lipid A, in CRX-547 likely disrupts electrostatic binding to positively charged amino acids of MD-2 or TLR4 and results in altered receptor dimerization and/or conformational changes in TLR4, which affect adaptor protein (MyD88 and/or Mal/TIRAP) binding and subsequent intracellular signaling. In fact, it was recently shown that the TRIF-selectivity of congeneric MPL is due to impaired CD14-dependent homodimerization of the TLR4-MD-2-MPL complex at the cell surface and concomitant reduction in MyD88-dependent signaling.26

Figure 2.

Figure 2

Open in a new tab

Structures of secondary ether and ester lipid AGPs

Previously, we17 and others27 showed that in some cases the chemical and metabolic instability of secondary fatty acyl residues present in lipid A mimetics such as CRX-527 can be overcome without a loss in bioactivity by replacing the ester linkages with more stable ether moieties (e.g. CRX-601 (5), Figure 2). Accordingly, we synthesized the ether lipid analog of D-isomer 4 containing three secondary decyl groups fully expecting that the cognate ether lipid 6a would retain the TRIF selectivity exhibited by 4.28 Quite surprisingly, the three ester-to-ether conversions restored or ‘rescued’ substantial NF-κB activity in ether lipid D-isomer 6a relative to the L-isomer 5, which itself was virtually indistinguishable from the corresponding ester lipid 3 in terms of NF-κB induction in a human cell line (specific experiment not shown, cf. Figure 3). The structural basis for the TRIF selectivity of CRX-547 (4), and subject of this note, was evaluated by synthesizing a chemical factorial of the six possible ether/ester hybrid derivatives of CRX-547 (4) and 6a in order to determine which secondary acyl residue (or residues) in 4 is likely responsible for the observed reduction in MyD88-dependent signaling. Three of the hybrids, 6b–d, possess a single (R)-3-decyloxytetradecanoyl (E–ether lipid) group either O- or N- linked to the sugar moiety or N-linked to the seryl unit in combination with two (R)-3-decanoyloxytetradecanoyl (A–acyl) groups at the other two sites (abbreviated EAA (6b), AEA (6c), AAE (6d); wherein the ether lipid E is in the ‘left-hand’, ‘middle’, or ‘right-hand’ position on the AGP backbone, respectively). The other three hybrids, 6e–g, possess an (R)-3-decanoyloxytetradecanoyl (A) group at any one of the three backbone positions in combination with ether lipids (E) at the other two sites (abbreviated AEE (6e), EAE (6f), EEA (6g)) (Scheme 1).

Figure 3.

Figure 3

Open in a new tab

NF-κB Induction by AGP derivatives in HEK293 human cell line stably transfected with plasmids expressing TLR4, MD-2, CD14 and NF-κB SEAP.

Scheme 1.

Scheme 1

Open in a new tab

(a) (i) 8, 13 or 14, BF3-OEt2 (2 equiv), CH2Cl2; (ii) 10% Pd/C, H2, THF, 50–60% with 13 and 14; 80% with 8 (2 steps); (b) (i) NH4OH or Mg(OMe)2, MeOH; (ii) BnBr, n-Bu4NBr, CH2Cl2, aq NaHCO3; (iii) TBDMS-Cl, pyr, 38–56% (3 steps); (c) (R)-3-decyl or decanoyloxytetradecanoic acid (1 equiv), EDC-MeI (1 equiv), 4-pyrrolidinopyridine (0.05 equiv), CH2Cl2, 0 ºC→rt, 75–88%; (d) (i) Zn, AcOH, 20% aq THF, rt; (ii) (R)-3-decyl or decanoyloxytetradecanoic acid, (2 equiv), EEDQ, CH2Cl2, rt, 43–58% (2 steps); (e) TFA (10–20 equiv), CH2Cl2, rt, quant; (f) (i) Zn, AcOH, 60 °C; (ii) (R)-3-decyl or decanoyloxytetradecanoic acid, (1 equiv), EDDQ (1.4 equiv), CH2Cl2, rt, 62–66% (3 steps); (g) TBDMS-Cl (3.9 equiv) 90–93%; (h) (i) Zn, AcOH, 20% aq THF, rt; (ii) (R)-3-decyl or decanoyloxytetradecanoic acid (1 equiv), EDC-MeI (1 equiv), CH2Cl2, rt, 54–56% (2 steps); (i) (i) (BnO)2PNi-Pr2 (1.4 equiv), 4,5-dicyanoimidazole (1.4 equiv), CH2Cl2, rt; (ii) m-chloroperbenzoic acid (3 equiv), 0 ºC, 59–78% (2 steps); (j) 10% Pd/C, H2, THF, 90–98%.

The ether lipid AGP 6a and the ether/ester hybrids 6b-g were synthesized by the general route developed earlier for the preparation of CRX-527, CRX-601 and other seryl AGPs using the N-2,2,2-trichlorethoxycarbonyl (Troc) method for stereoselective β-glycosylation in the first step but differing with respect to the order in which the fatty acyl groups (A or E) are introduced along the synthetic pathway (Scheme 1).13a, 17, 29

The syntheses of 6a (EEE) and the ‘left-hand’ hybrids 6b (EAA) and 6e (AEE), in which the N-acyl groups of the seryl and glucosamine units are the same, commenced with readily available 1,3,4,6-tetra-O-acetyl-2-deoxy-2-(2,2,2-trichloroethoxycarbonylamino)-β-D-glucopyranoside (7)30 and N-(2,2,2-trichloroethoxycarbonyl)-D-serine benzyl ester (8)31 (Scheme 1, route A). Stereoselective β-glycosylation of 8 with 7 in the presence of BF3-etherate followed by a series of deprotection (benzyl ester and O-acetyl cleavage) and re-protection (esterification and 6-O-silylation) steps afforded common advanced intermediate 9 in 45% overall yield from 7. Selective carbodiimide-mediated 3-O-acylation of 9 with (R)-decanoyloxytetradecanoic acid29 or (R)-3-decyloxytetradecanoic acid17 followed by reductive removal of the N-Troc groups with activated Zn/AcOH in 20% aq. tetrahydrofuran (THF) at room temperature and N,N-di-acylation with the requisite fatty acid in the presence of 2-ethoxy-1-ethoxycarbonyl-1,2-dihydroquinoline (EEDQ) furnished the desired tri-acyl intermediates 12a, 12b, and 12e.

The syntheses of the ‘right-hand’ hybrids 6g (EEA) and 6d (AAE) and ‘middle’ hybrids 6f (EAE) and 6c (AEA), which possess different acyl groups on the two amino groups, began with the β-glycosylation of N-acylated D-serine benzyl esters 1315 and 1432 with glycosyl donor 7 followed by a sequence of protecting group manipulations analogous to that used in the preparation of 9 to give the monoacylated AGP scaffolds 15 and 16 (Scheme 1, route B). However, the introduction of the 3-O-acyl group onto these scaffolds made it difficult to subsequently remove the remaining N-Troc group flanked by two bulky acyl residues, requiring more rigorous deprotection conditions and leading to the partial loss of the t-butyldimethylsilyl (TBDMS) group and by-product formation. This problem was overcome in the case of tri-acyl intermediates 12d,g by first removing the TBDMS group and then performing N-Troc deprotection with Zn/AcOH at 60 °C; subsequent N-acylation with the appropriate fatty acid and re-installation of the TBDMS group gave 12d and 12g (Scheme 1, route C).

Alternatively, this problem could be overcome by reversing the order in which O- and N-linked fatty acyl groups were introduced onto the AGP backbone, which also obviated the need to temporarily remove the 6-O-silyl group. Accordingly, in the case of pseudosymmetric (‘middle’) hybrids 6c (AEA) and 6f (EAE), the N-Troc group of 15 and 16 was removed under mild conditions (Zn, AcOH, 20% aq. THF) at room temperature, and the resulting amino diol intermediate sequentially N- and O-acylated with the appropriate fatty acids under carbodiimide conditions to provide 12c and 12f (Scheme 1, route D). Dibenzyl phosphorylation of tri-acyl derivatives 12a–g via the phosphoramidite method and sequential deprotection of the TBDMS and benzyl groups produced the target compounds 12a–g in >95% purity as their triethylammonium salts after reverse-phase chromatography and lyophilization of the free acids from 2% aqueous triethylamine.33

The 6 hybrid compounds 6b–g along with CRX-527 (3), CRX-547 (4), CRX-601 (5) and 6a were formulated at 2 mg/mL in 2% aq. glycerol as previously described25 and evaluated for NF-κB induction in a HEK293 cell reporter assay. The utility of this assay for measuring relative activity for TLR4 agonists and the correlation of NF-κB nuclear translocation and MyD88-dependent inflammatory cytokine induction from primary human cells was previously demonstrated for CRX-527, CRX-547, and other AGPs.16,25 To measure the relative induction of NF-κB after stimulation with AGPs, HEK293 cells stably transfected with plasmids for expression of human TLR4, MD-2, CD14 and an NF-κB-inducible promoter driving transcription of secreted embryonic alkaline phosphatase (SEAP) were treated with a dose response of each compound for 24 hours and cell culture supernatants were assayed for relative SEAP activity as previously described.25

As noted above and shown in Figure 3, the D-seryl AGP CRX-547 (AAA) exhibited significantly reduced NF-κB activity in comparison to the L-isomer CRX-527 and corresponding D- and L-secondary ether lipids 6a (EEE) and CRX-601, indicating that one or more of the secondary acyl residues of CRX-547 in combination with R absolute configuration in the aglycon unit (i.e., D-seryl unit) is responsible for the reduced NF-κB activity in the seryl AGP series. While it is tempting to speculate that the secondary lipid chain R3 proximal to the R-stereocenter in the aglycon would likely account for TRIF selectivity of CRX-547 (and MyD88 activity of 6a), quite the opposite was found to be the case: replacing the left-hand secondary acyl chain (R1) of CRX-547 (AAA) with an ether lipid to form compound EAA (3b) led to an intermediate NF-κB response and partial rescue of MyD88 signaling; replacing a second and third secondary acyl chain in the middle (R2) and right-hand positions (R3) further enhanced MyD88 activity (compounds EEA, EAE, EEE). Conversely, analogs of CRX-547 possessing an ester-linked lipid in the left-hand position (R1) distal to the aglycon unit were the least active of the analogs tested irrespective of the substitutions in the other two secondary positions (i.e., AEE, AAE, AEA) and similar in MyD88 activity to CRX-547 (AAA).

The critical importance of the secondary acyl residue R1—both its H-bond accepting carbonyl oxygen and 10-carbon chain length34—for reduced NF-κB activity and concomitant TRIF selectivity is consistent with the importance of a 10-carbon R1 acyl (or alkyl) group for AGP agonist activity and may also be related to the species selectivity of the secondary ether lipid derivative of CRX-526 (compound 2, R4=CO2H (S), n=1, R1=R2=R3= hexyl), which shows potent TLR4 antagonist activity similar to CRX-526 in human cell assays, but weak agonism in murine models.17 Taken together, these observations suggest that the R1 carbonyl group (C=O) likely plays a pivotal role not only in hydrogen bonding to an H-bond donor in murine MD-2 at the entrance to the hydrophobic pocket and interfering with murine TLR4 activation in the case of CRX-526, but also in H-bonding to human MD-2 and markedly interfering with MyD88 signaling for certain D-seryl AGPs (viz. AAA, AAE, AEA, AEE). The results are also consistent with an eritoran-like orientation of AXX hybrid AGP molecules (X = A or E) in the MD-2 hydrophobic pocket, wherein the R1 carbonyl group might impede homodimerization and/or conformational changes in TLR4 necessary for MyD88-dependent signaling.

In summary, the syntheses of the secondary ether lipid D-seryl AGP 6a and six secondary ester/ether hybrid derivatives of 6a and CRX-547 (6b-g) were carried out in order to further define the structural requirements for MyD88-dependent signaling activity in the AGP class of lipid A mimetics; the new AGPs were prepared using the Troc method for stereoselective β-glycosylation and adjusting the order in which the fatty acyl groups (A or E) are attached to the AGP backbone to achieve the desired substitution patterns. A comparison of D-seryl hybrid molecules 6b–g (EAA, AEA, AAE, AEE, EAE, EEA) to parent molecules CRX-547 (AAA) and 6a (EEE) and corresponding L-seryl derivatives CRX-527 (3) and CRX-601 (5) for their ability to induce NF-κB in the HEK293 cell reporter assay showed that a secondary acyl group in the left-hand position (R1) was indispensable for reduced activation through NF-κB and, by extension, reduced MyD88-dependent signaling activity (TRIF selectivity) in the D-seryl class of AGPs (viz. AAA, AAE, AEA, AEE). On the other hand, the effect on NF-κB activity of a secondary ether group in the left-hand position and either an ether or ester group at the other two positions was not as definitive: NF-κB activity was variable and correlated generally with the overall number of ether lipids present (EEE > EAE~EEA > EAA), suggesting that the EXX hybrid molecules (X = A or E) may enter the MD-2 pocket in both orientations predicted by crystallographic studies, but with the eritoran-like orientation placing the R1 ether lipid at the dimerization interface predominating.

The possibility of an AGP entering the MD-2 pocket in both orientations is further supported by molecular modeling studies using the Molecular Operating Environment (MOE) program.35 The hybrid molecule AEE (6e) was docked in the TLR4–MD-2 complex (Protein Data Bank ID code 3FXI) in both LPS-like and eritoran-like orientations as shown in Figure 4(a) and 4(b), respectively. Molecular dynamics simulations with the MOE program using the AMBER12:EHT force field predicted comparable potential energies for the two orientations of AEE (6e) in the TLR4–MD-2 complex at the dimerization interface. Similar results were obtained for other hybrids as well as the parent molecules AAA (4) and EEE (6a) in the two orientations (data not shown).

Figure 4.

Figure 4

Open in a new tab

Molecular models of AEE (6e) docked in MD-2 in LPS-like (a) and eritoran-like (b) orientations and interacting at the dimerization interface with the TLR4 ectodomain (primary TLR4 of the TLR4–MD-2 complex not shown). The TLR4 ectodomain and MD-2 are shown in blue and gray, respectively. The AEE molecule bound to the hydrophobic cavity of MD-2 is shown as ball and stick with C, N, O and P atoms shown as cyan, blue, red and purple, respectively; side chains of certain amino acid residues participating in interactions between MD-2, TLR4 and AEE are shown in orange.

These study results provide valuable insight to agonist interactions with the human TLR4–MD-2 receptor complex and should be useful in the design and development of novel vaccine adjuvants or immunomodulators with enhanced immunostimulatory response and reduced toxicity. Studies are currently underway to confirm whether the essential interactions of the R1 carbonyl group of TRIF-selective AGPs with human TLR4–MD-2 take place predominantly proximal (eritoran-like orientation) as proposed here or distal (LPS-like orientation) to the dimerization interface. As with subtle structural changes in the TLR4–MD-2 receptor due to species differences and TLR4 polymorphism,36 slight structural variations in the TLR4 ligand itself also appear to affect ligand orientation in the receptor and the resulting immune response.

Acknowledgments

The authors wish to thank Dr. Nupur Dutta for compound formulation and analytical characterization of formulations. This work was sponsored by GlaxoSmithKline Biologicals SA and supported in part by the National Institutes of Allergy and Infectious Diseases (NIAID) contract nos. HHSN266200400008C and HHSN27200900008C (to Corixa Corporation d/b/a GlaxoSmithKline Vaccines). Any opinions, findings and conclusions or recommendations expressed herein are those of the authors and do not necessarily reflect the views of the NIAID. All authors participated in the design, implementation, or analysis as well as in the interpretation of the study results and drafting of the manuscript or revising it critically for intellectual content. All authors had full access to the data and approved the manuscript before it was submitted by the corresponding author.

Footnotes

Competing Interests

J.K.K., H.G.B., K.T.R., M.T.L., J.R.W., J.T.E., and D.A.J. are, or were at the time of the study, employees of GlaxoSmithKline (GSK) Vaccines. H.G.B., K.T.R., M.T.L., J.R.W., J.T.E., and D.A.J. own shares or options to purchase shares in GSK. W.S.B., J.T.E., and D.A.J. are designated inventors on U.S. patent No. 8,629,116 related to this work.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References and notes

  • 1.Ulevitch RJ. Nat Rev Immunol. 2004;4:512. doi: 10.1038/nri1396. [DOI] [PubMed] [Google Scholar]
  • 2.Duthie MS, Windish HP, Fox CB, Reed SG. Immunol Rev. 2010;239:178. doi: 10.1111/j.1600-065X.2010.00978.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Visintin A, Latz E, Monks BG, Espevik T, Golenbock DT. J Biol Chem. 2003;278:48313. doi: 10.1074/jbc.M306802200. [DOI] [PubMed] [Google Scholar]
  • 4.Park BS, Song DH, Kim HM, Choi BS, Lee H, Lee JO. Nature. 2009;458:1191. doi: 10.1038/nature07830. [DOI] [PubMed] [Google Scholar]
  • 5.Zughaier SM, Zimmer SM, Datta A, Carlson RW, Stephens DS. Infect Immun. 2005;73:2940. doi: 10.1128/IAI.73.5.2940-2950.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Bjoerkbacka H, Fitzgerald KA, Huet F, Li X, Gregory JA, Lee MA, Ordija CM, Dowley NE, Golenbock DT, Freeman MW. Physiol Genomics. 2004;19:319. doi: 10.1152/physiolgenomics.00128.2004. [DOI] [PubMed] [Google Scholar]
  • 7.a) Yamamoto M, Sato S, Mori K, Hoshino K, Takeuchi O, Takeda K, Akira S. J Immunol. 2002;169:6668. doi: 10.4049/jimmunol.169.12.6668. [DOI] [PubMed] [Google Scholar]; b) Yamamoto M, Sato S, Hemmi H, Hoshino K, Kaisho T, Sanjo H, Takeuchi O, Sugiyama M, Okabe M, Takeda K, Akira S. Science. 2003;301:640. doi: 10.1126/science.1087262. [DOI] [PubMed] [Google Scholar]
  • 8.a) Hoebe K, Janssen EM, Kim SO, Alexopoulou L, Flavell RA, Han J, Beutler B. Nat Immunol. 2003;4:1223. doi: 10.1038/ni1010. [DOI] [PubMed] [Google Scholar]; b) McAleer JP, Rossi RJ, Vella AT. J Immunol. 2009;182:5322. doi: 10.4049/jimmunol.0803616. [DOI] [PMC free article] [PubMed] [Google Scholar]; c) Kolumam GA, Thomas S, Thompson LJ, Sprent J, Murali-Krishna K. J Exp Med. 2005;202:637. doi: 10.1084/jem.20050821. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.a) Schwarz TF, Huang L, Medina Doris MR, Valencia A, Lin T, Behre U, Catteau G, Thomas F, Descamps D. J Adolesc Health. 2012;50:187. doi: 10.1016/j.jadohealth.2011.11.004. [DOI] [PubMed] [Google Scholar]; b) Garcon N, Segal L, Tavares F, Van Mechelen M. Vaccine. 2011;29:4453. doi: 10.1016/j.vaccine.2011.04.046. [DOI] [PubMed] [Google Scholar]
  • 10.Mata-Haro V, Cekic C, Martin M, Chilton PM, Casella CR, Mitchell TC. Science. 2007;316:1628. doi: 10.1126/science.1138963. [DOI] [PubMed] [Google Scholar]
  • 11.Mogensen TH, Paludan SR, Kilian M, Oestergaard L. Scand J Immunol. 2006;64:646. doi: 10.1111/j.1365-3083.2006.01856.x. [DOI] [PubMed] [Google Scholar]
  • 12.a) Kotani S, Takada H, Tsujimoto M, Ogawa T, Takahashi I, Ikeda T, Otsuka K, Shimauchi H, Kasai N. Infect Immun. 1985;49:225. doi: 10.1128/iai.49.1.225-237.1985. [DOI] [PMC free article] [PubMed] [Google Scholar]; b) Kusumoto S, Fukase K. Chemical Record. 2006;6:333. doi: 10.1002/tcr.20098. [DOI] [PubMed] [Google Scholar]; c) Gaekwad J, Zhang Y, Zhang W, Reeves J, Wolfert MA, Boons GJ. J Biol Chem. 2010;285:29375. doi: 10.1074/jbc.M110.115204. [DOI] [PMC free article] [PubMed] [Google Scholar]; d) Fujimoto Y, Shimoyama A, Suda Y, Fukase K. Carbohydr Res. 2012;356:37. doi: 10.1016/j.carres.2012.03.013. [DOI] [PubMed] [Google Scholar]
  • 13.a) Lam C, Schuetze E, Hildebrandt J, Aschauer H, Liehl E, Macher I, Stuetz P. Antimicrob Agents Chemother. 1991;35:500. doi: 10.1128/aac.35.3.500. [DOI] [PMC free article] [PubMed] [Google Scholar]; b) Matsuura M, Kiso M, Hasegawa A. Infect Immun. 1999;67:6286. doi: 10.1128/iai.67.12.6286-6292.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]; c) Johnson DA. Curr Top Med Chem. 2008;8:64. doi: 10.2174/156802608783378882. [DOI] [PubMed] [Google Scholar]
  • 14.Ishizaka ST, Hawkins LD. Expert Rev Vaccines. 2007;6:773. doi: 10.1586/14760584.6.5.773. [DOI] [PubMed] [Google Scholar]
  • 15.Johnson DA, Sowell CG, Johnson CL, Livesay MT, Keegan DS, Rhodes MJ, Ulrich JT, Ward JR, Cantrell JL, Brookshire VG. Bioorg Med Chem Lett. 1999;9:2273. doi: 10.1016/s0960-894x(99)00374-1. [DOI] [PubMed] [Google Scholar]
  • 16.Stover AG, Da Silva CJ, Evans JT, Cluff CW, Elliott MW, Jeffery EW, Johnson DA, Lacy MJ, Baldridge JR, Probst P, Ulevitch RJ, Persing DH, Hershberg RM. J Biol Chem. 2004;279:4440. doi: 10.1074/jbc.M310760200. [DOI] [PubMed] [Google Scholar]
  • 17.Bazin HG, Murray TJ, Bowen WS, Mozaffarian A, Fling SP, Bess LS, Livesay MT, Arnold JS, Johnson CL, Ryter KT, Cluff CW, Evans JT, Johnson DA. Bioorg Med Chem Lett. 2008;18:5350. doi: 10.1016/j.bmcl.2008.09.060. [DOI] [PubMed] [Google Scholar]
  • 18.Resman N, Vasl J, Oblak A, Pristovsek P, Gioannini TL, Weiss JP, Jerala R. J Biol Chem. 2009;284:15052. doi: 10.1074/jbc.M901429200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Saitoh Si, Akashi S, Yamada T, Tanimura N, Matsumoto F, Fukase K, Kusumoto S, Kosugi A, Miyake K. J Endotoxin Res. 2004;10:257. doi: 10.1179/096805104225005904. [DOI] [PubMed] [Google Scholar]
  • 20.Tsukamoto H, Fukudome K, Takao S, Tsuneyoshi N, Kimoto M. Int Immunol. 2010;22:271. doi: 10.1093/intimm/dxq005. [DOI] [PubMed] [Google Scholar]
  • 21.Kim HM, Park BS, Kim JI, Kim SE, Lee J, Oh SC, Enkhbayar P, Matsushima N, Lee H, Yoo OJ, Lee JO. Cell. 2007;130:906. doi: 10.1016/j.cell.2007.08.002. [DOI] [PubMed] [Google Scholar]
  • 22.Ohto U, Fukase K, Miyake K, Satow Y. Science. 2007;316:1632. doi: 10.1126/science.1139111. [DOI] [PubMed] [Google Scholar]
  • 23.a) Hawkins LD, Christ WJ, Rossignol DP. Curr Top Med Chem. 2004;4:1147. doi: 10.2174/1568026043388123. [DOI] [PubMed] [Google Scholar]; b) Gradisar H, Keber MM, Pristovsek P, Jerala R. J Leukocyte Biol. 2007;82:968. doi: 10.1189/jlb.1206727. [DOI] [PubMed] [Google Scholar]
  • 24.Garidel P, Rappolt M, Schromm AB, Howe J, Lohner K, Andrae J, Koch MHJ, Brandenburg K. Biochim Biophys Acta, Biomembr. 2005;1715:122. doi: 10.1016/j.bbamem.2005.07.013. [DOI] [PubMed] [Google Scholar]
  • 25.Bowen WS, Minns LA, Johnson DA, Mitchell TC, Hutton MM, Evans JT. Sci Signaling. 2012;5:ra13. doi: 10.1126/scisignal.2001963. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Tanimura N, Saitoh SH, Ohto U, Akashi-Takamura S, Fujimoto Y, Fukase K, Shimizu T, Miyake K. Int Immunol. 2013;26:307. doi: 10.1093/intimm/dxt071. [DOI] [PubMed] [Google Scholar]
  • 27.a) Vasan M, Wolfert MA, Boons GJ. Org Biomol Chem. 2007;5:2087. doi: 10.1039/b704427e. [DOI] [PMC free article] [PubMed] [Google Scholar]; b) Christ WJ, Asano O, Robidoux AL, Perez M, Wang Y, Dubuc GR, Gavin WE, Hawkins LD, McGuinness PD, Mullarkey MA. Science. 1995;268:80. doi: 10.1126/science.7701344. [DOI] [PubMed] [Google Scholar]; c) Erwin AL, Munford RS. J Biol Chem. 1990;265:16444. [PubMed] [Google Scholar]
  • 28.Johnson DA. Expert Rev Vaccines. 2013;12:711. doi: 10.1586/14760584.2013.811189. [DOI] [PubMed] [Google Scholar]
  • 29.Bazin HG, Bess LS, Livesay MT, Ryter KT, Johnson CL, Arnold JS, Johnson DA. Tetrahedron Lett. 2006;47:2087. [Google Scholar]
  • 30.Boullanger P, Jouineau M, Bouammali B, Lafont D, Descotes G. Carbohydr Res. 1990;202:151. doi: 10.1016/0008-6215(90)84077-8. [DOI] [PubMed] [Google Scholar]
  • 31.Compound 8 was prepared in 94% yield from D-serine benzyl ester and 2,2,2-trichloroethyl chloroformate (Troc-Cl) under Schotten-Baumann conditions: colorless oil; 1H NMR (CDCl3, 400 MHz) δ 7.38–7.26 (m, 5H), 5.93 (d, J = 7.2 Hz, 1H), 5.24 (d, J = 2.0 Hz, 2H), 4.74 (d, J = 1.6 Hz, 2H), 4.52–4.49 (m, 1H), 4.13–4.05 (m, 1H), 3.99–3.94 (m, 1H), 2.00 (t, J = 6.0 Hz, 1H); HRMS (ESI-TOF) m/z: Calcd for C13H14Cl3NO5 [M+H]+ 370.0016, found 370.0018.
  • 32.Compound 14 was prepared in 57% yield by N-acylation of D-serine benzyl ester with (R)-3-decyloxytetradecanoic acid 17 in the presence of 1-(3-dimethylaminopropyl)-3-ethylcarbodiimide methiodide (EDC-MeI): white solid; mp 67–68°C; 1H NMR (CDCl3, 400 MHz) δ 7.35 (br s, 5H), 7.30 (d, J = 7.2 Hz, 1H), 5.22 (s, 2H), 4.70–4.72 (m, 1H), 3.96 (t, J = 5.2 Hz, 2H), 3.59–3.62 (m, 1H), 3.42–3.46 (m, 2 H), 2.54 (t, J = 6.0 Hz, 1H), 2.38–2.49 (m, 2H), 1.47–1.56 (m, 4H), 1.45 (br s, 32H), 0.88 (t, J = 6.6 Hz, 6H); HRMS (ESI-TOF) m/z: Calcd for C34H59NO5 [M+H]+ 562.4471, found 562.4481.
  • 33.Spectral data for compounds 6a–g. Compound 6a (EEE): 1H NMR (CDCl3 + few drops CD3OD, 400 MHz) δ 5.12 (t, J = 9.8 Hz, 1H), 4.65 (t, J = 3.4 Hz, 1H), 4.45 (d, J = 8.4 Hz, 1H), 4.26 (q, J = 9.8 Hz, 1H), 3.99–3.88 (m, 4H), 3.75–3.33 (m, 11H, peaks partially masked by CD3OH peak), 3.07 (q, J = 7.2 Hz, 4H, CH2 of Et3N (~2/3 equiv), 2.64–2.58 (m, 1H), 2.51–2.31 (m, 4H), 2.14 (dd, J = 5.2, 15.2 Hz, 1H), 1.56–1.26 (m, 114H), 0.88 (t, J = 6.2 Hz, 18H); 13C NMR (CDCl3 + few drops CD3OD, 100 MHz) δ 172.4, 172.1, 171.6, 171.5, 101.7, 76.8, 76.5, 76.3, 76.1, 75.7, 73.4, 70.7 (2), 69.7, 69.6, 69.2, 60.3, 53.7, 52.4, 45.8, 41.4, 41.0, 39.2, 34.6, 34.0, 31.9, 30.0 (2), 29.8 (2), 29.7 (3), 29.6, 29.5 (2), 29.3, 26.2, 26.1, 26.0, 25.3, 25.2, 25.0, 22.6, 14.0, 8.4; HRMS (ESI-TOF) m/z: Calcd for C81H157N2O16P [M]− 1445.1271, found 1445.1236. Compound 6b (EAA): 1H NMR (CDCl3 + few drops CD3OD, 400 MHz) δ 5.30–5.24 (m, 1H), 5.16 (t, J = 9.8 Hz, 1H), 5.11 (br s, 1H), 4.65 (br s, 1H), 4.47 (d, J = 8.0 Hz, 1H), 4.26–4.21 (m, 1H, peaks partially masked by CD3OH peak), 3.99–3.94 (m, 3H), 3.84 (t, J = 9.4 Hz, 1H), 3.78–3.75 (m, 2H), 3.60–3.54 (m, 1H), 3.47–3.37 (m, 2H), 3.11 (q, J = 7.2 Hz, 4H, CH2 of Et3N (~2/3 equiv), 2.66–2.28 (m, 10H), 1.66–1.50 (m, 12H), 1.35–1.27 (m, 98H), 0.89 (t, J = 6.4 Hz, 18H); 13C NMR (CDCl3 + few drops CD3OD, 100 MHz) δ 174.3, 174.0, 171.6, 171.4, 171.2, 170.9, 101.7, 76.1, 75.6, 73.0, 71.3, 71.2, 69.8, 69.6, 60.3, 53.9, 52.3, 46.0, 40.9, 40.8, 39.1, 34.5 (2), 34.3, 34.1, 31.9 (2), 29.8, 29.7 (3), 29.6 (4), 29.5 (2), 29.4 (3), 29.3 (2), 29.2 (2), 26.0, 25.4, 25.2, 25.1, 25.0, 22.7, 14.0, 8.4; HRMS (ESI-TOF) m/z: Calcd for C81H153N2O18P [M]− 1473.0856, found 1473.0803. Compound 6c (AEA): 1H NMR (CDCl3 + few drops CD3OD, 400 MHz) δ 7.70 (d, J = 8.0 Hz, 1H), 5.28–5.22 (m, 1H), 5.18–5.11 (m, 1H), 5.07 (t, J = 9.8 Hz, 1H), 4.62 (br s, 1H), 4.41 (d, J = 8.0 Hz, 1H), 4.23 (q, J = 9.6 Hz, 1H), 3.99–3.87 (m, 4H), 3.54–3.50 (m, 2H), 3.42–3.33 (m, 3H, peaks partially masked by CD3OH peak), 3.08–3.04 (br q, J = 6.4 Hz, 2H, CH2 of Et3N (~1/3 equiv), 2.69 (dd, J = 6.8, 16.8 Hz, 1H), 2.58–2.39 (m, 4H), 2.30–2.20 (m, 5H), 1.59–1.45 (m, 12H), 1.26 (br d, J = 3.6 Hz, 95H), 0.88 (t, J = 6.4 Hz, 18H); 13C NMR (CDCl3 + few drops CD3OD, 100 MHz) δ 174.1, 173.8, 172.6, 171.8, 170.5, 101.7, 76.1, 75.7, 73.5, 71.0, 70.4, 70.1, 69.9, 69.1, 60.1, 57.8, 53.4, 52.4, 45.6, 40.9, 40.8, 38.8, 34.4, 34.3, 34.1, 31.9, 31.8, 29.9, 29.8, 29.7 (3), 29.6 (2), 29.5 (3), 29.4 (2), 29.3 (3), 29.2, 29.1 (2), 26.2, 25.2, 25.3, 25.1, 25.0 (2), 22.6, 14.0, 10.8, 8.3, 7.7; HRMS (ESI-TOF) m/z: 1472.0778, found 1472.0795. Compound 6d (AAE): 1H Calcd for C81H153N2O18P [M–H]− NMR (CDCl3 + few drops CD3OD, 400 MHz) δ 5.18–5.12 (m, 2H), 5.06 (t, J = 9.8 Hz, 1H), 4.43 (d, J = 8.4 Hz, 1H), 4.37 (br s, 1H), 4.20–4.13 (m, 1H), 3.97–3.82 (m, 4H), 3.74–3.68 (m, 2H), 3.50–3.46 (m, 1H), 3.44–3.39 (m, 2H), 3.06 (q, J = 7.2 Hz, 4H, CH2 of Et3N (~2/3 equiv), peaks partially masked by CD3OH peak), 2.68 (dd, J = 7.2, 16 Hz, 1H), 2.56–2.27 (m, 9H), 1.54 (br s, 12H), 1.26 (br s, 98H), 0.88 (t, J = 6.6 Hz, 18H); 13C NMR (CDCl3 + few drops CD3OD, 100 MHz) δ 174.2, 173.8, 171.8, 170.7, 101.2, 76.5, 75.4, 73.3, 71.2, 70.7, 70.2, 69.5, 69.2, 60.3, 53.7, 52.3, 45.8, 41.4, 40.7, 39.1, 34.6, 34.5, 34.4, 34.3, 34.2, 31.8 (2), 29.9, 29.7, 29.6 (3), 29.5, 29.4, 29.3 (2), 29.2, 29.1, 26.1, 25.3 (2), 25.2, 25.0, 24.9, 22.6, 14.0, 8.4; HRMS (ESI-TOF) m/z: Calcd for C81H153N2O18P [M–H]− 1472.0778, found 1472.0789. Compound 6e (AEE): 1H NMR (CDCl3 + few drops CD3OD, 400 MHz) δ 5.14–5.08 (m, 2H), 4.67 (br s, 1H), 4.45 (d, J = 8.4 Hz, 1H), 4.28–4.20 (m, 1H), 3.99–3.93 (m, 3H), 3.86 (t, J = 9.0 Hz, 1H), 3.76–3.69 (m, 3H), 3.55–3.38 (m, 5H, peaks partially masked by CD3OH peak), 3.07 (br d, J = 7.2 Hz, 8H, CH2 of Et3N (~1 1/3 equiv), 2.67 (dd, J = 6.8, 17.2 Hz, 1H), 2.54–2.22 (m, 7H), 1.55–1.42 (m, 12H), 1.35–1.26 (m, 106H), 0.88 (t, J = 6.4 Hz, 18H); 13C NMR (CDCl3 + few drops CD3OD, 100 MHz) δ 174.2, 172.7, 172.3, 171.8, 170.7, 101.7, 77.6, 76.8, 76.4, 75.9, 73.6, 70.3, 70.0, 69.8, 69.4, 60.5, 53.9, 52.5, 46.0, 41.6, 41.1, 39.1, 34.8 (2), 34.7, 34.4, 34.3, 32.2, 32.1, 30.2, 30.1, 30.0, (3), 29.9, 29.8, (2), 29.7, 29.6 (2), 29.4, 26.4, 26.3, 25.5, 25.4, 25.2, 22.9, 14.3, 8.6; HRMS (ESI-TOF) m/z: Calcd for C81H155N2O17P [M–H]− 1458.0985, found 1458.0993. Compound 6f (EAE): 1H NMR (CDCl3 + few drops CD3OD, 400 MHz) δ 5.13–5.09 (m, 2H), 4.59 (br s, 1H), 4.46 (d, J = 8.4 Hz, 1H), 4.26–4.18 (m, 1H), 3.99–3.96 (m, 3H), 3.86 (t, J = 9.2 Hz, 1H), 3.76–3.69 (m, 3H), 3.54–3.40 (m, 5H, peaks partially masked by CD3OH peak), 3.08 (br q, J = 6.8 Hz, 6H, CH2 of Et3N (~1 equiv), 2.61 (dd, J = 7.2, 16 Hz, 1H), 2.50–2.28 (m, 7H), 1.58–1.51 (m, 12H), 1.26 (br s, 103H), 0.88 (t, J = 6.6 Hz, 18H); 13C NMR (CDCl3 + few drops CD3OD, 100 MHz) δ 174.0, 171.6 (2), 171.5 (2), 101.8, 76.5, 76.0, 75.5, 73.1, 71.0, 70.7, 70.3, 69.6, 69.5, 59.9, 54.1, 53.6, 45.7, 41.5, 41.0, 39.0, 34.4, 34.1, 31.8 (3), 29.8, 29.7, 29.6 (3), 29.5 (2), 29.4 (3), 29.3 (2), 29.2, 29.1, 26.0, 25.9, 25.2, 25.1, 25.0, 22.6, 13.9, 8.3; HRMS (ESI-TOF) m/z: Calcd for C81H155N2O17P [M–H]− 1458.0985, found 1458.0991. Compound 6g (EEA): 1H NMR (CDCl3 + few drops CD3OD, 400 MHz) δ 5.26–5.20 (m, 1H), 5.09 (t, J = 10 Hz, 1H), 4.45 (d, J = 8.4 Hz, 1H), 4.31 (t, J = 3.6 Hz, 1H), 4.26–4.18 (m, 1H, peaks partially masked by CD3OH peak), 3.99–3.40 (m, 12H), 3.13 (q, J = 7.6 Hz, 8H, CH2 of Et3N (~1 1/3 equiv), 2.65–2.22 (m, 8H), 1.60–1.54 (m, 12H), 1.38–1.27 (m, 106H), 0.89 (t, J = 4.8 Hz, 18H); 13C NMR (CDCl3 + few drops CD3OD, 100 MHz) δ 173.5, 172.2, 171.6, 171.4, 170.3, 101.6, 76.1, 76.0, 75.4, 73.1, 70.9, 70.8, 69.5, 69.0, 60.2, 53.4, 52.2, 45.8, 40.9, 40.7, 39.0, 34.3, 34.2, 34.0, 33.9, 31.8, 31.7, 29.8, 29.7 (2), 29.6, 29.5 (3), 29.4 (2), 29.3 (2), 29.2 (2), 29.1, 29.0, 26.1, 25.9, 25.1, 24.9 (2), 22.5, 13.9, 8.3; HRMS (ESI-TOF) m/z: Calcd for C81H155N2O17P [M–H]− 1458.0985, found 1458.0982.
  • 34.The 14-carbon homolog of CRX-547 (compound 2, R4=CO2H (R), n=1, R1=R2=R3= tetradecanoyl) was found to be virtually indistinguishable from its L-seryl isomer15 in terms of NF-κB activity in the HEK293 system; however, both of these compounds were substantially less potent than CRX-527 (3) in this assay (data not shown). These results indicate that secondary acyl chain length—and R1 chain length in particular25—is a critical determinant of TRIF selectivity as well as TLR4 agonist activity in general.
  • 35.Molecular Operating Environment (MOE) Chemical Computing Group Inc; 1010 Sherbooke St. West, Suite #910, Montreal, QC, Canada, H3A 2R7: 2012. [Google Scholar]
  • 36.a) Ohto U, Fukase K, Miyake K, Shimizu T. Proc Natl Acad Sci USA. 2012;109:7421. doi: 10.1073/pnas.1201193109. [DOI] [PMC free article] [PubMed] [Google Scholar]; b) Ferwerda B, McCall MBB, Verheijen K, Kullberg BJ, van der Ven AJAM, van der Meer JWM, Netea MG. Mol Med. 2008;14:346. doi: 10.2119/2007-00135.Ferwerda. [DOI] [PMC free article] [PubMed] [Google Scholar]; c) Yamakawa N, Ohto U, Akashi-Takamura S, Takahashi K, Saitoh SI, Tanimura N, Suganami T, Ogawa Y, Shibata T, Shimizu T, Miyake K. Int Immunol. 2013;25:45. doi: 10.1093/intimm/dxs084. [DOI] [PubMed] [Google Scholar]

RESOURCES