Structure-Based Design and Synthesis of Lipid A Derivatives to Modulate Cytokine Responses.

Agonists of Toll like receptors (TLRs) have attracted interest as adjuvants and immune modulators. A crystal structure of TLR4/MD2 with E. coli LPS indicates that the fatty acid at C-2 of the lipid A component of LPS induces dimerization of two TLR4-MD2 complexes, which in turn initiates cell signaling leading to the production of (pro)inflammatory cytokines. To probe the importance of the (R)-3-hydroxymyristate at C-2 of lipid A, a range of bis- and mono-phosphoryl lipid A derivatives with different modifications at C-2 were prepared by a strategy in which 2-methylnaphthyl ethers were employed as permanent protecting group that could be readily removed by catalytic hydrogenation. The C-2 amine was protected as 9-fluorenylmethyloxycarbamate, which at a later stage could be removed to give a free amine that was modified by different fatty acids. LPS and the synthetic lipid As induced the same cytokines, however, large differences in activity were observed. A compound having a hexanoyl moiety at C-2 still showed agonistic properties, but further shortening to a butanoyl abolished activity. The modifications had a larger influence on monophosphoryl lipid As. The lipid As having a butanoyl moiety at C-2 could selectively antagonize TRIF associated cytokines induced by LPS or lipid A.

SARS-CoV-2 Spike N-Terminal Domain Engages 9-O-Acetylated α2-8-Linked Sialic Acids.

SARS-CoV-2 viruses engage ACE2 as a functional receptor with their spike protein. The S1 domain of the spike protein contains a C-terminal receptor binding domain (RBD) and an N-terminal domain (NTD). The NTD of other coronaviruses includes a glycan binding cleft. However, for the SARS-CoV-2 NTD, protein-glycan binding was only observed weakly for sialic acids with highly sensitive methods. Amino acid changes in the NTD of variants of concern (VoC) show antigenic pressure, which can be an indication of NTD-mediated receptor binding. Trimeric NTD proteins of SARS-CoV-2, alpha, beta, delta, and omicron did not reveal a receptor binding capability. Unexpectedly, the SARS-CoV-2 beta subvariant strain (501Y.V2-1) NTD binding to Vero E6 cells was sensitive to sialidase pretreatment. Glycan microarray analyses identified a putative 9-O-acetylated sialic acid as a ligand, which was confirmed by catch-and-release ESI-MS, STD-NMR analyses, and a graphene-based electrochemical sensor. The beta (501Y.V2-1) variant attained an enhanced glycan binding modality in the NTD with specificity toward 9-O-acetylated structures, suggesting a dual-receptor functionality of the SARS-CoV-2 S1 domain, which was quickly selected against. These results indicate that SARS-CoV-2 can probe additional evolutionary space, allowing binding to glycan receptors on the surface of target cells.

The SARS-CoV-2 spike N-terminal domain engages 9- O -acetylated α2-8-linked sialic acids.

SARS-CoV-2 viruses engage ACE2 as a functional receptor with their spike protein. The S1 domain of the spike protein contains a C-terminal receptor-binding domain (RBD) and an N-terminal domain (NTD). The NTD of other coronaviruses includes a glycan-binding cleft. However, for the SARS-CoV-2 NTD protein-glycan binding was only observed weakly for sialic acids with highly sensitive methods. Amino acid changes in the NTD of Variants of Concern (VoC) shows antigenic pressure, which can be an indication of NTD-mediated receptor binding. Trimeric NTD proteins of SARS-CoV-2, Alpha, Beta, Delta, and Omicron did not reveal a receptor binding capability. Unexpectedly, the SARS-CoV-2 Beta subvariant strain (501Y.V2-1) NTD binding to Vero E6 cells was sensitive to sialidase pretreatment. Glycan microarray analyses identified a putative 9- O -acetylated sialic acid as a ligand, which was confirmed by catch-and-release ESI-MS, STD-NMR analyses, and a graphene-based electrochemical sensor. The Beta (501Y.V2-1) variant attained an enhanced glycan binding modality in the NTD with specificity towards 9- O -acetylated structures, suggesting a dual-receptor functionality of the SARS-CoV-2 S1 domain, which was quickly selected against. These results indicate that SARS-CoV-2 can probe additional evolutionary space, allowing binding to glycan receptors on the surface of target cells. Synopsis: Coronaviruses utilize their N-terminal domain (NTD) for initial reversible low-affinity interaction to (sialylated) glycans. This initial low-affinity/high-avidity engagement enables viral surfing on the target membrane, potentially followed by a stronger secondary receptor interaction. Several coronaviruses, such as HKU1 and OC43, possess a hemagglutinin-esterase for viral release after sialic acid interaction, thus allowing viral dissemination. Other coronaviruses, such as MERS-CoV, do not possess a hemagglutinin-esterase, but interact reversibly to sialic acids allowing for viral surfing and dissemination. The early 501Y.V2-1 subvariant of the Beta SARS-CoV-2 Variant of Concern has attained a receptor-binding functionality towards 9- O -acetylated sialic acid using its NTD. This binding functionality was selected against rapidly, most likely due to poor dissemination. Ablation of sialic acid binding in more recent SARS-CoV-2 Variants of Concern suggests a fine balance of sialic acid interaction of SARS-CoV-2 is required for infection and/or transmission.

Synthesis of monophosphoryl lipid A using 2-naphtylmethyl ethers as permanent protecting groups.

Lipid A, which is a conserved component of lipopolysaccharides of gram-negative bacteria, has attracted considerable interest for the development of immuno-adjuvants. Most approaches for lipid A synthesis rely on the use of benzyl ethers as permanent protecting groups. Due to the amphiphilic character of lipid A, these compounds aggregate during the hydrogenation step to remove benzyl ethers, resulting in a sluggish reaction and by-product formation. To address this problem, we have developed a synthetic approach based on the use of 2-naphtylmethyl ether (Nap) ethers as permanent protecting group for hydroxyls. At the end of a synthetic sequence, multiple of these protecting groups can readily be removed by oxidation with 2,3-dichloro-5,6-dicyano-p-benzoquinone (DDQ). Di-allyl N,N-diisopropylphosphoramidite was employed to install the phosphate ester and the resulting allyl esters were cleaved using palladium tetrakistriphenylphosphine. The synthetic strategy allows late stage introduction of different fatty acids at the amines of the target compound, which is facilitated by Troc and Fmoc as orthogonal amino-protecting groups.