4,5,6,7,8,9-Hexahydroxy-2-oxo-nonanoic acid

• Molecular Formula: C9H16O9
• Molecular Weight: 268.22 g/mol
• SMILES: C(C(C(C(C(C(CO)O)O)O)O)O)C(=O)C(=O)O
• InChI: InChI=1S/C9H16O9/c10-2-5(13)7(15)8(16)6(14)3(11)1-4(12)9(17)18/h3,5-8,10-11,13-16H,1-2H2,(H,17,18)
• InChIKey: FQHUAUMYHAJTDH-UHFFFAOYSA-N

4,5,6,7,8,9-Hexahydroxy-2-oxo-nonanoic Acid (KDN): A Structurally Distinct Deaminated Sialic Acid for Glycobiology Research and Procurement

4,5,6,7,8,9-Hexahydroxy-2-oxo-nonanoic acid (CAS 22594-61-2, molecular formula C₉H₁₆O₉, molecular weight 268.22 Da), systematically named 3-deoxy-D-glycero-D-galacto-2-nonulosonic acid and commonly abbreviated as KDN (ketodeoxynonulosonic acid), is a nine-carbon monosaccharide belonging to the sialic acid family [1]. Unlike the predominant mammalian sialic acid N-acetylneuraminic acid (Neu5Ac), KDN is deaminated at the C5 position, bearing a hydroxyl group in place of the N-acetamido moiety [2]. First discovered in rainbow trout egg polysialoglycoproteins in 1986, KDN has since been identified across evolutionary diverse organisms from bacteria to humans, where it occurs as the free sugar, as CMP-KDN, and conjugated to glycoproteins and glycolipids [3].

Structurally distinct deaminated sialic acid: C5 hydroxyl replaces the N-acetamido group of Neu5Ac

Sialidase-resistant ketosidic linkages support stable glycoconjugate probe development

Occurs across bacteria to humans; enables evolutionary and comparative glycobiology studies

• [1] ChEBI. Sialosonic acid (CHEBI:176465). European Bioinformatics Institute. CAS Registry Number: 22594-61-2. View Source
• [2] Chao L, Wennekes T. Toward the Synthesis of 3,3-Difluoro Deaminated Sialic Acid (C3DFKDN). European Journal of Organic Chemistry. First published 25 June 2025. DOI: 10.1002/ejoc.202500456. View Source
• [3] Inoue S, Kitajima K. KDN (deaminated neuraminic acid): dreamful past and exciting future of the newest member of the sialic acid family. Glycoconjugate Journal. 2006;23(5-6):277-290. DOI: 10.1007/s10719-006-6484-y. View Source

Why Neu5Ac, Neu5Gc, or KDO Cannot Substitute for 4,5,6,7,8,9-Hexahydroxy-2-oxo-nonanoic Acid in Critical Research and Industrial Applications

KDN occupies a unique biochemical niche that no other sialic acid or nonulosonic acid can fill. The defining structural feature—a C5 hydroxyl replacing the N-acetamido group of Neu5Ac—fundamentally alters enzyme recognition, glycan stability, and immunological properties [1]. C5-acetamido-bearing sialic acids (Neu5Ac, Neu5Gc) are susceptible to canonical bacterial and animal sialidases, while KDN ketosidic linkages are refractory to these enzymes, conferring resistance to exosialidase cleavage [2]. Furthermore, the KDN biosynthetic pathway proceeds via a dedicated KDN-9-phosphate synthetase that condenses mannose-6-phosphate with phosphoenolpyruvate—an enzymatic route distinct from the Neu5Ac pathway that uses N-acetylmannosamine-6-phosphate [3]. These divergent properties mean that substituting Neu5Ac, Neu5Gc, or the eight-carbon analog KDO for KDN in experimental or production workflows will yield fundamentally different enzymatic kinetics, glycan stability profiles, and immunological readouts.

C5 substitution alters enzyme recognition

The C5 hydroxyl of KDN fundamentally changes sialidase and sialyltransferase kinetics compared to N-acetylated sialic acids.

Glycan stability profile may not transfer

KDN ketosidic linkages are often refractory to common sialidases; Neu5Ac and Neu5Gc are susceptible and may yield divergent stability readouts.

Distinct biosynthetic pathway requires dedicated tools

KDN biosynthesis proceeds via a dedicated KDN-9-phosphate synthetase, not the Neu5Ac pathway; the eight-carbon analog KDO also lacks the relevant C9 structure.

• [1] Chao L, Wennekes T. Toward the Synthesis of 3,3-Difluoro Deaminated Sialic Acid (C3DFKDN). European Journal of Organic Chemistry. First published 25 June 2025. DOI: 10.1002/ejoc.202500456. View Source
• [2] Angata T, Matsuda T, Kitajima K. Synthesis of neoglycoconjugates containing deaminated neuraminic acid (KDN) using rat liver α2,6-sialyltransferase. Glycobiology. 1998;8(3):277-284. DOI: 10.1093/glycob/8.3.277. PMID: 9451037. View Source
• [3] Angata T, Nakata D, Matsuda T, Kitajima K, Troy FA 2nd. Biosynthesis of KDN (2-keto-3-deoxy-D-glycero-D-galacto-nononic acid). Identification and characterization of a KDN-9-phosphate synthetase activity from trout testis. Journal of Biological Chemistry. 1999;274(33):22949-22956. DOI: 10.1074/jbc.274.33.22949. PMID: 10438460. View Source

Head-to-Head Quantitative Evidence: 4,5,6,7,8,9-Hexahydroxy-2-oxo-nonanoic Acid (KDN) vs. N-Acetylneuraminic Acid (Neu5Ac) and Other Sialic Acids

Exosialidase Resistance of KDN α2→6Gal Linkage vs. Neu5Ac α2→6Gal Linkage in Neoglycoconjugates

In a direct head-to-head comparison using rat liver α2,6-sialyltransferase to synthesize KDN-transferrin and Neu5Ac-transferrin, the KDNα2→6Gal linkage was completely resistant to exosialidase treatment, whereas the Neu5Acα2→6Gal linkage was sensitive to enzymatic cleavage [1]. The Vmax/Km for incorporation of KDN from CMP-KDN into asialotransferrin was 0.0527 min⁻¹, approximately half that of Neu5Ac from CMP-Neu5Ac (0.119 min⁻¹). Additionally, Sambucus sieboldiana agglutinin (SSA) showed binding preference for KDN-transferrin over Neu5Ac-transferrin [1].

Exosialidase Resistance

Head-to-head

KDNα2→6Gal resistant; Neu5Acα2→6Gal sensitive

Supports sialidase-resistant glycoconjugate probe development

Vmax/Km: 0.0527 (KDN) vs 0.119 min⁻¹ (Neu5Ac); rat α2,6-sialyltransferase

Evidence Dimension	Exosialidase sensitivity of α2→6Gal-linked sialic acid on transferrin
Target Compound Data	KDNα2→6Gal linkage: resistant to exosialidase treatment; Vmax/Km = 0.0527 min⁻¹
Comparator Or Baseline	Neu5Acα2→6Gal linkage: sensitive to exosialidase treatment; Vmax/Km = 0.119 min⁻¹
Quantified Difference	KDN linkage is categorically resistant vs. Neu5Ac linkage sensitive; KDN incorporation efficiency (Vmax/Km) is 44% of Neu5Ac (0.0527/0.119)
Conditions	Rat liver Galβ1→4GlcNAc α2,6-sialyltransferase; CMP-KDN and CMP-Neu5Ac donor substrates; asialotransferrin acceptor; exosialidase from Arthrobacter ureafaciens

Why This Matters

For researchers developing stable glycoconjugate therapeutics, probes, or diagnostic reagents, KDN provides inherent exosialidase resistance that Neu5Ac cannot offer, eliminating the need for chemical stabilization against glycosidase degradation.

• [1] Angata T, Matsuda T, Kitajima K. Synthesis of neoglycoconjugates containing deaminated neuraminic acid (KDN) using rat liver α2,6-sialyltransferase. Glycobiology. 1998;8(3):277-284. DOI: 10.1093/glycob/8.3.277. PMID: 9451037. View Source

KDNase SM: Absolute Substrate Selectivity for KDN over Neu5Ac and Neu5Gc

Kitajima et al. (1994) isolated and characterized KDNase SM from Sphingobacterium multivorum, a novel sialidase that specifically hydrolyzes KDN α-ketosidic linkages but does not release Neu5Ac or Neu5Gc from any tested substrate [1]. The enzyme was assayed using 4-methylumbelliferyl-KDN (4-MU-KDN) and demonstrated no activity toward 4-MU-Neu5Ac, N-acetylneuraminyllactose, colominic acid, or other Neu5Ac- and Neu5Gc-containing glycoconjugates. KDNase SM activity was not inhibited by the classical N-acylneuraminidase inhibitor 2,3-dehydro-2-deoxy-N-acetylneuraminic acid (Neu5Ac2en) [1]. The enzyme exhibited broad linkage specificity for KDN, hydrolyzing α2→3, α2→6, and α2→8 ketosidic linkages from natural substrates including (KDN)GM3 and KDN-glycoprotein [1].

KDNase SM Selectivity

Head-to-head

Absolute specificity: cleaves KDN, zero activity on Neu5Ac/Neu5Gc

Enables selective KDN detection in complex samples

Purified KDNase SM; fluorometric assay with 4-MU substrates

Evidence Dimension	Substrate hydrolysis by purified KDNase SM
Target Compound Data	KDN: cleaved from 4-MU-KDN, (KDN)GM3, KDN-glycoprotein, and KDN-containing N-glycans (α2→3, α2→6, α2→8 linkages)
Comparator Or Baseline	Neu5Ac and Neu5Gc: no detectable release from 4-MU-Neu5Ac, N-acetylneuraminyllactose, colominic acid, or any Neu5Ac/Neu5Gc-glycoconjugate tested
Quantified Difference	Absolute selectivity: KDNase SM hydrolyzes KDN substrates; zero detectable activity on Neu5Ac/Neu5Gc substrates
Conditions	Affinity-purified KDNase SM from Sphingobacterium multivorum; fluorometric assay with 4-MU substrates; pH and temperature optima as reported

Why This Matters

KDNase SM is an essential analytical tool for selectively detecting and quantifying KDN in complex biological samples containing Neu5Ac and Neu5Gc; procurement of KDN is a prerequisite for generating the 4-MU-KDN substrate needed for this assay.

• [1] Kitajima K, Kuroyanagi H, Inoue S, Ye J, Troy FA 2nd, Inoue Y. Discovery of a new type of sialidase, "KDNase," which specifically hydrolyzes deaminoneuraminyl (3-deoxy-D-glycero-D-galacto-2-nonulosonic acid) but not N-acylneuraminyl linkages. Journal of Biological Chemistry. 1994;269(34):21415-21419. PMID: 8063773. View Source

Loach Liver Sialidase Km: KDN Exhibits 1.7-fold Higher Affinity than Neu5Ac

Li et al. (1994) identified a novel sialidase from loach (Misgurnus fossilis) liver capable of efficiently cleaving both KDN and Neu5Ac glycosides [1]. Kinetic analysis using fluorogenic 4-methylumbelliferyl α-ketoside substrates yielded a Km of 0.07 mM for KDN versus 0.12 mM for Neu5Ac, representing a 1.7-fold higher apparent affinity for the deaminated sialic acid [1]. The enzyme was also active toward natural substrates KDNα2→3Galβ1→4GlcCer and KDNα2→6N-acetylgalactosaminitol, establishing this as the first reported sialidase capable of efficiently cleaving glycosidically linked KDN [1].

Loach Sialidase Affinity

Head-to-head

Km: 0.07 mM (KDN) vs 0.12 mM (Neu5Ac)

Higher enzyme affinity for KDN reported in loach liver sialidase

4-MU substrates, pH 4.6; first sialidase cleaving glycosidically linked KDN

Evidence Dimension	Michaelis constant (Km) for 4-methylumbelliferyl α-ketoside substrates
Target Compound Data	Km (4-MU-KDN) = 0.07 mM
Comparator Or Baseline	Km (4-MU-Neu5Ac) = 0.12 mM
Quantified Difference	1.7-fold lower Km for KDN (0.07 vs. 0.12 mM), indicating higher enzyme affinity for KDN
Conditions	Loach (Misgurnus fossilis) liver sialidase; pH optimum 4.6; 4-methylumbelliferyl α-ketoside substrates

Why This Matters

The higher affinity for KDN indicates that KDN-specific or KDN-preferring sialidases exist in nature, reinforcing that KDN is not merely a Neu5Ac surrogate but a biologically distinct substrate requiring dedicated analytical and enzymatic tools.

• [1] Li YT, Yuziuk JA, Li SC, Nematalla A, Hasegawa A, Kimura M, Nakagawa H. A novel sialidase capable of cleaving 3-deoxy-D-glycero-D-galacto-2-nonulosonic acid (KDN). Archives of Biochemistry and Biophysics. 1994;310(1):243-246. DOI: 10.1006/abbi.1994.1163. PMID: 8161211. View Source

CMP-KDN Synthetase: KDN is the Preferred Substrate over Neu5Ac and Neu5Gc in Trout Testis

Terada et al. (1993) characterized CMP-KDN synthetase from rainbow trout testis and demonstrated that KDN is the kinetically preferred nonulosonic acid substrate [1]. Vmax/Km analysis revealed values of 4.4 × 10⁻³ min⁻¹ for KDN, compared to 2.3 × 10⁻³ min⁻¹ for Neu5Ac and 1.8 × 10⁻³ min⁻¹ for Neu5Gc [1]. This represents a 1.9-fold preference for KDN over Neu5Ac and a 2.4-fold preference over Neu5Gc. Notably, the calf brain CMP-sialic acid synthetase showed the opposite preference, with Neu5Ac and Neu5Gc as preferred substrates, highlighting the unique species- and tissue-specific biochemical machinery for KDN activation [1].

CMP-KDN Synthetase Preference

Head-to-head

Vmax/Km: 4.4×10⁻³ (KDN) vs 2.3×10⁻³ (Neu5Ac)

KDN is the kinetically preferred substrate in trout testis synthetase

1.9-fold preference over Neu5Ac; 2.4-fold over Neu5Gc; pH 9–10, 25 °C

Evidence Dimension	Catalytic efficiency (Vmax/Km) of CMP-sialic acid synthetase
Target Compound Data	KDN: Vmax/Km = 4.4 × 10⁻³ min⁻¹
Comparator Or Baseline	Neu5Ac: Vmax/Km = 2.3 × 10⁻³ min⁻¹; Neu5Gc: Vmax/Km = 1.8 × 10⁻³ min⁻¹
Quantified Difference	1.9-fold higher Vmax/Km for KDN vs. Neu5Ac; 2.4-fold vs. Neu5Gc
Conditions	Partially purified CTP:CMP-3-deoxynonulosonate cytidylyltransferase from rainbow trout (Oncorhynchus mykiss) testis; pH 9-10; 25°C; 25 mM Mg²⁺

Why This Matters

For chemoenzymatic synthesis of CMP-KDN—the activated sugar nucleotide donor required for all KDN-transferase reactions—the trout testis enzyme provides the only known efficient route; mammalian CMP-Sia synthetases are 15-fold less active toward KDN, making procurement of KDN and access to the appropriate synthetase critical for glycoconjugate production.

• [1] Terada T, Kitazume S, Kitajima K, Inoue S, Ito F, Troy FA 2nd, Inoue Y. Synthesis of CMP-deaminoneuraminic acid (CMP-KDN) using the CTP:CMP-3-deoxynonulosonate cytidylyltransferase from rainbow trout testis. Identification and characterization of a CMP-KDN synthetase. Journal of Biological Chemistry. 1993;268(4):2640-2648. PMID: 8428944. View Source

Species-Divergent CMP-Sialic Acid Synthetase Activity: Trout Enzyme Favors KDN, Murine Enzyme Strongly Discriminates Against It

Nakata et al. (2001) cloned and compared recombinant CMP-sialic acid synthetases from rainbow trout and mouse [1]. The recombinant trout enzyme showed high activity toward both substrates, with Vmax/Km values of 1.1 min⁻¹ for KDN and 0.68 min⁻¹ for Neu5Ac (1.6-fold preference for KDN) [1]. In stark contrast, the recombinant murine enzyme had a Vmax/Km of 0.23 min⁻¹ for KDN versus 3.5 min⁻¹ for Neu5Ac—a 15-fold discrimination against KDN [1]. This species divergence explains why mammalian systems have historically been poor sources for KDN activation and underscores the need to use the appropriate enzymatic toolkit when working with KDN.

Species-Divergent Activation

Cross-study

Trout: 1.6-fold KDN preference; Murine: 15-fold discrimination against KDN

Trout synthetase required for efficient CMP-KDN production

Recombinant enzymes; Vmax/Km trout 1.1 (KDN) vs 0.68 (Neu5Ac); murine 0.23 vs 3.5

Evidence Dimension	Recombinant CMP-Sia synthetase Vmax/Km for KDN and Neu5Ac
Target Compound Data	Trout enzyme: KDN Vmax/Km = 1.1 min⁻¹; Murine enzyme: KDN Vmax/Km = 0.23 min⁻¹
Comparator Or Baseline	Trout enzyme: Neu5Ac Vmax/Km = 0.68 min⁻¹; Murine enzyme: Neu5Ac Vmax/Km = 3.5 min⁻¹
Quantified Difference	Trout: 1.6-fold KDN preference (1.1/0.68); Murine: 15-fold discrimination against KDN (0.23 vs. 3.5)
Conditions	Recombinant enzymes expressed from cloned cDNA; trout enzyme 432 amino acids, 53.8% identity to murine; Vmax/Km determined under standardized conditions

Why This Matters

Researchers aiming to produce CMP-KDN for glycoconjugate synthesis must select the trout-derived synthetase over mammalian enzymes; the 65-fold relative difference in KDN activation efficiency between trout and murine enzymes directly determines the feasibility of preparative-scale CMP-KDN production.

• [1] Nakata D, Münster AK, Gerardy-Schahn R, Aoki N, Matsuda T, Kitajima K. Molecular cloning of a unique CMP–sialic acid synthetase that effectively utilizes both deaminoneuraminic acid (KDN) and N-acetylneuraminic acid (Neu5Ac) as substrates. Glycobiology. 2001;11(8):685-692. DOI: 10.1093/glycob/11.8.685. PMID: 11479279. View Source

Optimal Research and Industrial Applications for 4,5,6,7,8,9-Hexahydroxy-2-oxo-nonanoic Acid (KDN) Based on Quantitative Evidence

Synthesis of Exosialidase-Resistant Neoglycoconjugate Therapeutics and Diagnostic Probes

KDN is the sialic acid of choice for constructing glycoconjugates that must resist degradation by host and microbial sialidases. As demonstrated by Angata et al. (1998), KDNα2→6Gal linkages are completely refractory to Arthrobacter ureafaciens exosialidase, while the corresponding Neu5Ac linkages are quantitatively cleaved [1]. This property is critical for developing stable glycoprotein therapeutics, antibody-drug conjugates with sialylated linkers, and glycan-based diagnostic probes intended for use in sialidase-rich environments such as inflamed tissues, tumor microenvironments, or bacterial co-cultures. The slightly reduced incorporation efficiency of KDN relative to Neu5Ac (Vmax/Km 0.0527 vs. 0.119 min⁻¹) is a manageable trade-off for the conferred stability.

Selective Detection and Quantification of KDN in Cancer Biomarker Research Using KDNase SM

The absolute substrate selectivity of KDNase SM—which hydrolyzes KDN α-ketosides but shows zero activity toward Neu5Ac and Neu5Gc substrates—makes this enzyme an indispensable analytical tool [2]. Elevated free KDN levels have been documented in fetal cord red blood cells (2.4-fold higher than adult controls), ovarian cancer cells, and throat cancers (where free KDN accounts for ~94% of total KDN vs. only ~6.5% free Neu5Ac) [3][4]. Procurement of high-purity KDN is essential for synthesizing the 4-MU-KDN fluorogenic substrate and for use as an authentic standard in LC-MS/MS quantification workflows targeting KDN as a potential oncodevelopmental biomarker.

Chemoenzymatic Production of CMP-KDN Using Trout-Derived Synthetase for Glycan Remodeling

Preparative-scale synthesis of CMP-KDN—the obligate activated donor for all sialyltransferase-mediated KDN incorporation—requires the rainbow trout CMP-KDN synthetase, which exhibits a 1.9-fold Vmax/Km preference for KDN over Neu5Ac (4.4 × 10⁻³ vs. 2.3 × 10⁻³ min⁻¹) [5]. Recombinant trout enzyme maintains a favorable 1.6-fold KDN preference (Vmax/Km 1.1 vs. 0.68 min⁻¹), whereas the murine ortholog discriminates 15-fold against KDN [6]. This enzymatic route, combined with commercially sourced KDN, enables the scalable production of CMP-KDN for in vitro glycan remodeling, synthesis of KDN-containing oligosaccharides, and generation of KDN-based glycan arrays for Siglec and lectin screening.

Anti-Virulence Research: KDN as a Natural 'Trojan Horse' Against Sialic Acid-Scavenging Bacterial Pathogens

KDN offers a unique anti-pathogen strategy distinct from Neu5Ac. Saha et al. (2021) demonstrated that nontypeable Haemophilus influenzae (NTHi) incorporates environmental KDN into its lipooligosaccharide, which blunts Neu5Ac-mediated virulence attributes including resistance to complement and whole blood killing, complement C3 deposition, IgM binding, and Siglec-9 engagement [7]. Unlike Neu5Ac, which pathogens use for molecular mimicry of host glycans, Kdn-terminated glycans are structurally distinct from vertebrate sialoglycans and elicit natural human anti-Kdn antibodies [7]. Upper airway Kdn administration reduced NTHi infection in a humanized mouse model, positioning KDN as a potential therapeutic intervention against pathogens that scavenge host sialic acids [7].

Application

Selection Property

Validation Focus

Exosialidase-resistant glycoconjugate synthesis

Sialidase resistance profile

Exosialidase cleavage assay

KDN detection in glycan biomarker research

KDNase SM substrate selectivity

KDN-specific enzymatic assay

Chemoenzymatic CMP-KDN production

Trout CMP-KDN synthetase preference

CMP-KDN synthetase activity assay

Anti-virulence pathway research with sialic acid-scavenging bacteria

KDN incorporation into bacterial LPS

Host-pathogen interaction model

• [1] Angata T, Matsuda T, Kitajima K. Synthesis of neoglycoconjugates containing deaminated neuraminic acid (KDN) using rat liver α2,6-sialyltransferase. Glycobiology. 1998;8(3):277-284. DOI: 10.1093/glycob/8.3.277. PMID: 9451037. View Source
• [2] Kitajima K, Kuroyanagi H, Inoue S, Ye J, Troy FA 2nd, Inoue Y. Discovery of a new type of sialidase, "KDNase," which specifically hydrolyzes deaminoneuraminyl (3-deoxy-D-glycero-D-galacto-2-nonulosonic acid) but not N-acylneuraminyl linkages. Journal of Biological Chemistry. 1994;269(34):21415-21419. PMID: 8063773. View Source
• [3] Inoue S, Lin SL, Chang T, Wu SH, Yao CW, Chu TY, Troy FA 2nd, Inoue Y. Identification of free deaminated sialic acid (2-keto-3-deoxy-D-glycero-D-galacto-nononic acid) in human red blood cells and its elevated expression in fetal cord red blood cells and ovarian cancer cells. Journal of Biological Chemistry. 1998;273(42):27199-27204. DOI: 10.1074/jbc.273.42.27199. PMID: 9765240. View Source
• [4] Wang B, Brand-Miller J. LC-MS/MS glycomic analyses of free and conjugated forms of the sialic acids, Neu5Ac, Neu5Gc and KDN in human throat cancers. Glycobiology. 2015;25(12):1362-1374. DOI: 10.1093/glycob/cwv051. View Source
• [5] Terada T, Kitazume S, Kitajima K, Inoue S, Ito F, Troy FA 2nd, Inoue Y. Synthesis of CMP-deaminoneuraminic acid (CMP-KDN) using the CTP:CMP-3-deoxynonulosonate cytidylyltransferase from rainbow trout testis. Identification and characterization of a CMP-KDN synthetase. Journal of Biological Chemistry. 1993;268(4):2640-2648. PMID: 8428944. View Source
• [6] Nakata D, Münster AK, Gerardy-Schahn R, Aoki N, Matsuda T, Kitajima K. Molecular cloning of a unique CMP–sialic acid synthetase that effectively utilizes both deaminoneuraminic acid (KDN) and N-acetylneuraminic acid (Neu5Ac) as substrates. Glycobiology. 2001;11(8):685-692. DOI: 10.1093/glycob/11.8.685. PMID: 11479279. View Source
• [7] Saha S, Coady A, Sasmal A, Kawanishi K, Choudhury B, Gerardy-Schahn R, Inoue S, Inoue Y, Kitajima K, Ram S, Varki A, Nizet V. Exploring the Impact of Ketodeoxynonulosonic Acid in Host-Pathogen Interactions Using Uptake and Surface Display by Nontypeable Haemophilus influenzae. mBio. 2021;12(1):e03226-20. DOI: 10.1128/mBio.03226-20. View Source