Leveraging Stereoelectronic Effects in Biofilm Eradication: Synthetic β-Amino Human Milk Oligosaccharides Impede Microbial Adhesion as Observed by Scanning Electron Microscopy

Abstract

Alongside Edward, Lemieux was among the earliest researchers studying negative hyperconjugation (i.e. the anomeric effect) or the preference for gauche conformations about the C1-O5 bond in carbohydrates. Lemieux also studied an esoteric, if not controversial, theory known as the reverse anomeric effect (RAE). This theory is used to rationalize scenarios where predicted anomeric stabilization does not occur. One such example is the Kochetkov amination, where reducing end amines exist solely as the β-anomer. Herein, we provide a brief account of Lemieux’s contributions to the area of stereoelectronic effects and apply this knowledge toward the synthesis of β-amino human milk oligosaccharides (βA-HMOs). These molecules were evaluated for their ability to inhibit growth and biofilm production in Group B Streptococcus (GBS) and Staphylococcus aureus. While the parent HMOs lacked antimicrobial and antibiofilm activity, their β-amino derivatives significantly inhibit biofilm formation in both species. Field Emission Gun-Scanning Single Electron Microscopy (FEG-SEM) revealed that treatment of the β-amino HMOs significantly inhibits bacterial adherence and eliminates the ability of both microbes to form biofilms.

Graphical Abstract

Introduction

Antimicrobial resistance threatens public health.¹ According to estimates published by the Centers for Disease Control (CDC), 2.8 million infections in the Unites States each year are caused by antibiotic-resistant pathogens.² A frontier approach to address bacterial infections is the development of molecules which weaken microbial behaviors that are harmful to the host.³ For example, molecules that regulate biofilms can be employed alongside or in lieu of antibiotics.^4–7 Such therapeutics, particularly ones that operate through a non-bactericidal mechanism, are desirable as they do not exert evolutionary pressure on the organism to acclimate and evolve resistance.

In an effort to discover novel scaffolds to address antibiotic resistance, we started a program aimed at assessing the antimicrobial properties of human milk oligosaccharides (HMOs).⁸ Initial studies revealed that HMOs govern bacterial growth and biofilm assembly.⁹ We discovered that HMO extracts possess bacteriostatic and antibiofilm activity against gram-positive pathogens, including Streptococcus agalactiae (GBS). In second-generation work, we observed that HMOs potentiate several intracellular-targeting antibiotics by, among other likely mechanisms, increasing cellular permeability.^10–11 Recently, we observed that while heterogeneous HMOs possess potent antibiofilm activity against GBS, single-entity HMOs are largely devoid of antibiofilm activity against this pathogen.^12–13 In a subsequent study, however, we observed that conversion of the prebiotic 2’-fucosyllactose (2’-FL, 1) to its β-amino-variant (βA-2’-FL, 2) generates a compound with impressive antibiofilm activity against GBS. We hypothesized that positively charged β-amino HMO species employ an antibiofilm effect by acting as a surfactant, preventing the microbe from adhering to surfaces. The biofilm extracellular matrix is stabilized through anionic substances including negatively charged extracellular DNA and polymeric sugars, which is known to be destabilized by cationic molecules.^14–15 While the mechanism behind this activity is unknown, we note that a related study by Aiassa and co-workers has shown that D-glucosamine reduces both adhesion and biofilm formation in a Staphylococcus epidermidis model.¹⁶ To contrast, D-glucose promotes adhesion, biofilm production, and growth.

Recently, we began to question whether the antibiofilm activity of βA-2’-FL 2 was unique, or if other βA-HMOs would serve as antibiofilm agents. Moreover, while the mechanisms that different bacteria employ to form biofilms varies, depending on environmental conditions and specific strain attributes, it is known that Streptococcal and Staphylococcal species share mechanistic pathways for producing biofilms.¹⁷ Thus, we hypothesized that the antibiofilm activity of βA-HMOs would be observed in an additional gram-positive organism, Staphylococcus aureus. Accordingly, we synthesized various βA-HMOs and assayed their abilities to inhibit biofilm production in GBS and methicillin-resistant Staphylococcus aureus (MRSA). Before we describe the production of these molecules and the results of our biological inquiry, we digress to provide a brief account of why the β-amine is the sole product produced during the Kochetkov amination. The knowledge of this phenomenon is derived primarily from Lemieux’s contribution to the area of carbohydrate-based stereoelectronic effects.

Stereoelectronic Effects in Carbohydrates

Reducing end glycosyl amines are precursors to glycoproteins as these amines accommodate selective derivatization with a variety of acylating reagents (e.g. the Lansbury reaction).^18–23 To synthesize a glycosyl amine, a carbohydrate is reacted with an excess of an ammonia source in a process known as the Kochetkov amination.²⁴ While superficially simple, the reaction generates the β-anomer exclusively. Interestingly, the reaction rarely goes to completion and typically reaches an equilibrium that depends on the structure of the starting materials and products, as well as the reaction conditions, vide infra. To appreciate why this reaction produces a single anomer, one must revisit a primary contribution from the Lemieux laboratory: characterization of the anomeric effect (Figure 1).

Figure 1.

Stereoelectronic effects in carbohydrates. A. The endo anomeric effect. B. The exo anomeric effect. C. The anomeric effect generalized in a Newman projection. D. The reverse anomeric effect.

The anomalous preference for the axial orientation of electronegative substituents at the anomeric center was first disclosed by Edward in 1955 and later characterized by Lemieux in 1958. In pyranosides, this stereoelectronic effect is operable in O5−C1−X systems (Figure 1A). It is valued at ca. 1.0 kcal/mol and can be described as an endo- or exo-anomeric effect (Figure 1A and 1B, respectively). The endo-anomeric effect refers to the preference of electronegative substituents, X, at the anomeric center to be oriented axially. This preference is dictated partly by a stabilizing n_O5→σ*_C1-X interaction. If the electronegative substituent is oriented equatorially, the system lacks bond “resonance” via back-bonding of an oxygen lone-pair to the antibonding (σ*) orbital of the C1-X bond. Moreover, dipole–dipole repulsion in the equatorial anomer favors axial orientation. The dipole repulsion argument is further supported by the reduction of the anomeric effect in polar solvents.²⁵ It should be noted, however, that computational studies suggest the anomeric effect is dominated by solely electrostatic interactions.²⁶

Contrary to the endo-anomeric effect, the exo-anomeric effect is driven by a lone pair of electrons on the anomeric substituent. It has been observed that there is a preference for this substituent to adopt a gauche conformation around the anomeric carbon. This confirmation enables an n_X→σ*_C1-O5 stabilizing effect (Figure 1B). Because equatorial glycosides lack an endo-anomeric effect, the exo-anomeric effect predominates. The exo-anomeric effect has important consequences for glycoside conformation in solution as it limits the number of possible conformations the molecule can adopt; conformational integrity is critical to biological function.²⁷ In summation, the anomeric effect is the general preference for gauche conformations about the C1-X bond in the O5-C1-X network where X is an electronegative heteroatom with a nonbonding pair of electrons (Figure 1C).

After detailing negative hyperconjugative effects, Lemieux described a more dubious aspect of hyperconjugation known as the reverse anomeric effect (Figure 1D). This effect attempts to explain the tendency of positively-charged, electronegative groups at the anomeric center to adopt an equatorial orientation. The reverse anomeric effect is often used to explain the stereochemical outcome of the Kochetkov amination. The longstanding argument was that if electronegative substituents prefer an axial orientation due to an n_O5→σ*_C1-X interaction, then a positively charged group would exist in an equatorial orientation to avoid the n_O5→σ*_C1-X interaction. However, if one properly interprets a positively charged functional group as electronegative, then the concept of a reverse anomeric effect is unreasonable. Indeed, Pinto and Perrin have independently shown that neutral and protonated amines show an equatorial preference, likely due to accentuated steric effects (although this assertation is based on data from biased systems. While the origin of selectivity for the configuration of glycosyl amines remains a contested topic,²⁸ one could argue that in the “no-bond” resonance picture of the anomeric effect, positively charged groups should be oriented in an axial fashion. Thus, the greater equatorial preference observed in the Kochetkov amination is likely attributed to accentuated steric effects, (in the case of alkyl amino groups) and generally to favorable electrostatic interactions (for any amine).

Results and Discussion

While the Kochetkov amination is the state-of-the-art method to synthesize β-amino glycans, the reaction is not without its complications. First, while the reaction does produce the β−glycosyl amine, the product exists in equilibrium with both anomers of the starting material. Second, the reaction traditionally requires upwards of 50 equivalents of an ammonium salt, which complicates purification. Lastly, the reaction generally requires long reaction times (e.g. 48 to 120 hours). While elevated temperatures increase reaction rate, they also increase formation of diglycosylamino byproducts when using monosaccharides as starting material. To address these shortcomings, microwave irradiation has been employed for the Kochetkov amination.^29–30 Counter to thermal conditions, microwave irradiation requires only 5 fold excess of an ammonium salt and generally proceeds to equilibrium within an hour at 50°C. Additionally, microwave irradiation suppresses formation for dimerized byproducts. While microwave irradiation has proven useful in Kochetkov amination reactions of monosaccharides, the reaction conditions do not accommodate carbohydrates that are disaccharides or larger. For example, the Seeberger lab demonstrated that when cellobiose is exposed to microwave mediated Kochetkov conditions, only 17% of the amino sugar is observed.²⁹

The present study commenced with determining the ideal conditions to achieve Kochetkov amination on complex oligosaccharides. Using 2’-FL, 1, as a model substrate, we compared the outcomes of microwave and thermal mediated reaction conditions. Beginning with microwave mediated reactions, we first employed (NH₄)₂CO₃ as the ammonia source. In this reaction, we observed that conversion of 2’-FL to its amino sugar occurred in 1:2 selectivity (Table 1, entry 1). Interestingly, both longer (3 h) and shorter (5 min) reaction times did not enhance production of the system. When the reaction was conducted using (NH₄)HCO₃, NH₄Cl, and a solution of NH₃ in methanol (Table 1, entries 2–4), each system failed to provide superior results. Our investigation of microwave mediated reaction conditions ended with focusing on entry 1 and exchanging CH₃OH for dimethylsulfoxide (DMSO). Interestingly, the change in reaction conditions provided an increased conversion 1:2.7 from starting material to product (Table 1, entry 5). Given that the conversion of 2’-FL to its β-amine was generally low under all conditions examined, our next point of investigation was a thermal reaction. Contrary to standard conditions that employ 50 to 100 equivalents of ammonium salt, we used 5-fold excess in the reaction. We used (NH₄)₂CO₃ as the source of ammonia as it provided the highest conversion under microwave conditions. The reactions stirred for 48 h, were warmed to 40°C and three solvents evaluated (DMSO, H₂O, and CH₃OH). The reaction was conducted at a lower temperature than the microwave variant in order to mitigate diglycosylamine formation. DMSO provided a similar result thermally, as it did under microwave irradiation (Table 1, entry 6, 1:2.2 ratio). H₂O decreased productivity of the reaction (Table 1, entry 7, 1:0.6 ratio). In an interesting development, using CH₃OH as solvent provided a 1:4.7 conversion of starting material to product (Table 1, entry 8).

Table 1.

Evaluation of the Kochetkov amination on 2’-FL.

	NH3 Source	NH3 Source Fold-Excess	Solvent	Time	Ratiod 1 : 2
1a	(NH4)2CO3	5	CH3OH	1 h	1 : 2
2a	(NH4)HCO3	5	CH3OH	1 h	1: 0.6
3a	NH4Cl	5	CH3OH	1 h	1 : 0
4b	NH3/CH3OH	3	CH3OH	1 h	1 : 0
5a	(NH4)2CO3	5	DMSO	1 h	1 : 2.7
6c	(NH4)2CO3	5	DMSO	48 h	1 : 2.2
7c	(NH4)2CO3	5	H2O	48 h	1 to 0.6
8c	(NH4)2CO3	5	CH3OH	48 h	1 : 4.7

^aReaction conditions: HMO 1 (1.0 equiv., 0.2 mmol) and ammonia source (amount of fold-excess) were placed into a microwave vial and diluted with solvent (2.5 mL). The microwave vial was sealed and irradiated for 1 h at 200 W and 50°C. After cooling to ambient temperature, the reaction mixture was lyophilized to dryness, providing the βA-HMO 2, which was used without further purification.

^bReaction follows the same general procedure but 7N NH₃ in CH₃OH is used as the solvent.

^cReaction follows the same general procedure but is heated thermally at 40 °C for 48 h.

^dRatio was calculated according to Table S1 in the SI.

We next sought to evaluate the effectiveness of the optimized reaction conditions on additional HMOs that would be used in the biological assays. Additional compounds were selected based on two criteria. First, we evaluated HMOs with structural features different to that of 2’-FL, i.e. more steric bulk near the reducing end, acidic residues, or longer chains. Second, we evaluated HMOs that had shown no antimicrobial or antibiofilm activity in previous screens.^12–13 This would enable facile evaluation of the antibiofilm activity of the βA-HMO variants. Based on these criteria, we selected the following HMOs: 3-fucosyllactose (3-FL), 6’-sialyllactose (6’-SL), and lacto-N-tetraose (LNT) (Figure 2). Using the optimized thermal conditions (Figure 2), each HMO was converted to its β−amine (2–5). The thermal yields and ratio of starting material to product are shown in blue. For comparison, we conducted the reaction on the optimal microwave conditions (yields and ratios shown in red). Complex oligosaccharides are converted to their amino sugars in superior yields under thermal conditions, when using methanol as solvent. While purification was not a necessary step in conversion to the βA-HMOs, the loss of product mass can be explained in three ways: transfer of product to conical tubes between sequential lyophilizations; permeabilization through the Kimwipe barrier during lyophilization; or hydrated starting material HMO elevating the initial mass. With βA-HMOs 2-5 in hand, we moved to evaluate their antimicrobial and antibiofilm properties against GBS (strain GB00590) and S. aureus (MRSA, strain USA300). Antimicrobial activity was assessed by monitoring bacterial growth in the presence of each βA-HMO over 24 h, while antibiofilm activity was assessed by evaluating biofilm production levels at 24 h. Importantly, biofilm levels were expressed as ratios of biofilm to biomass as a means of accounting for any accompanying antimicrobial activity. Growth and biofilm trends for each βA-HMO were compared to those of their respective parent HMOs as well as those of bacteria grown in the absence of any HMO additive. For all assays, HMO additives (βA-HMO and natural HMOs) were dosed at ca. 5 mg/mL as this approximates the low end of physiological HMO concentrations. The βA-HMOs were added as a mixture of starting material HMO and βA-HMO product in the ratio as reported in Figure 2. Moreover, previous work from our lab has shown that this concentration is non-lethal for GBS and S. aureus thus ensuring our ability to evaluate compounds for antibiofilm activity.^{9, 31}

Figure 2.

Synthesis of βA-HMOs (2–5). Anomeric proton shifts used to calculate the ratio is reported in the SI (Table S1).

As expected, based on our prior evaluation of the antimicrobial activity of βA-2’-FL 2 against GBS, none of the βA-HMOs used in the present study significantly inhibited GBS growth at any point (Figure 3A). Moreover, once again, no parent HMO exhibited significant antimicrobial activity. Similar results were observed in S. aureus as no βA-HMO or parent HMO was found to significantly inhibit S. aureus growth at any point in 24 h (Figure 3B). Gratifyingly however, while no parent HMOs exhibited antibiofilm activity, all βA-HMOs (2-5) significantly inhibited biofilm production in both GBS and S. aureus (Figure 3C and D). Impressively, βA-HMOs 2-5 decreased biofilm production in GBS and S. aureus by an average of 62 and 42%, respectively. Also notable is the observation that all βA-HMOs reduced GBS and S. aureus biofilm production to similar extents despite their structural differences. This finding supports our hypothesis that the positively charged carbohydrates function as antibiofilm agents by serving as surfactants rather than by engaging a specific cellular target.

Figure 3.

Evaluation of HMOs and (βA-HMO) dosed at ca. 5 mg/mL on S. agalactiae (GB00590) and S. aureus (USA300). Growth was quantified via OD₆₀₀ readings at 0, 2, 4, 6, 7, 8, and 24 hours. Mean OD₆₀₀ for each time point is indicated by the corresponding symbols. Biofilm was quantified via OD₅₆₀ readings at 24 h. Biofilm production is expressed as a ratio of biofilm/biomass (OD₅₆₀/OD₆₀₀). Growth of GB00590 (A) and USA300 (B) in the presence of parent HMOs and βA-HMOs. Biofilm production of GB00590 (C) and USA300 (D) in the presence of parent HMOs and βA-HMOs. Data displayed represent the relative mean growth or biofilm/biomass ratios ± SEM of three independent experiments, each with three technical replicates. Statistical analysis was performed in (C) and (D) in which **** represents p < 0.0001, *** represents p = 0.0010, and ** represents p = 0.0081 by one-way ANOVA, with post hoc Dunnett’s multiple comparison test comparing biofilm production of HMO supplemented conditions to biofilm production in either GB00590 or USA300 in HMO-free THB media.

To further interrogate the perturbations in bacterial biofilm formation exerted by the prebiotic β-amino HMO derivatives, high resolution FEG-SEM analyses were employed. Our analyses revealed that S. agalactiae GB00590 adhered to the surface of the coverslip and formed colonies of bacterial cells stacked in clumps with defined tertiary architecture, which we have previously defined as criteria for biofilm formation (Figure 4).^32–34 The addition of 2’-FL or 6’-SL compounds did not significantly alter bacterial cell adherence to the abiotic substrate, nor did it alter the ability of the bacterial cells to stack on top of each other to form biofilms. Interestingly, the addition of 3-FL did not impact bacterial adherence to the coverslip, but it was correlated with a slight decrease in biotic adherence between bacterial cells to form the tertiary architecture of the biofilm. But, interestingly, the addition of the β-amino derivatives resulted in a significant inhibition of bacterial adherence to the coverslip and abolished the ability for GB00590 to form biofilms. Similarly, our analyses revealed that S. aureus USA300 adhered to the coverslip and formed biofilms which were characterized by the presence of a fibrous and globular extracellular matrix as has been previously observed (Figure 5).^32–34

Figure 4.

High resolution field-emission gun scanning electron microscopy analyses of S. agalactiae strain GB00590 bacterial biofilm formation. FEG-SEM imaging of bacterial biofilms were performed on GBS samples grown in medium alone (Medium Alone), or medium supplemented with 2’-FL (+2’-FL), βA-2’-FL (+βA-2’-FL), 3-FL (+3-FL), βA-3-FL (+βA-3-FL), 6’-SL (+6’-SL), or βA-6’-SL (+βA-6’-SL). The addition of β-amino variants significantly inhibits GBS biofilm formation. Micrographs were collected at 20,000x magnification and magnification bars indicate 5 μm.

Figure 5.

High resolution field-emission gun scanning electron microscopy analyses of S. aureus strain USA300 bacterial biofilm formation. FEG-SEM imaging of bacterial biofilms were performed on USA300 samples grown in medium alone (Medium Alone), or medium supplemented with 2’-FL (+2’-FL), βA-2’-FL (+βA-2’-FL), 3-FL (+3-FL), βA-3-FL (+βA-3-FL), 6’-SL (+6’-SL), or βA-6’-SL (+βA-6’-SL). The addition of β-amino variants significantly inhibits USA300 biofilm formation. Micrographs were collected at 20,000x magnification and magnification bars indicate 5 μm.

The addition of 2’-FL or 3-FL compounds did not significantly alter bacterial cell adherence to the abiotic substrate, nor did it alter the ability of the bacterial cells to stack on top of each other to form biofilms. Interestingly, the addition of 6’-SL did not impact bacterial adherence to the coverslip, but it was correlated with a slight decrease in both the size and number of bacterial aggregates. But, interestingly, the addition of the β-amino derivatives resulted in a significant inhibition of bacterial adherence to the coverslip and abolished the ability for USA300 to form biofilms. The addition of the β-amino derivatives to both S. aureus and GBS resulted in significant abrogation of biofilm formation as determined by quantitative colorimetric assays and high-resolution SEM analyses. Previous work by our lab has demonstrated that other glycosides (such as synthetic ellagic acid) also have the capacity to inhibit early stage adhesion of bacterial biofilm formation.³²

Conclusion

To conclude, we have applied a thermal mediated Kochetkov amination to the synthesis of βA-HMOs that show anti-biofilm activity across two gram-positive test strains. The addition of the βA-HMOs, as determined by FEG-SEM, reveals a significant inhibition of bacterial surface adherence and the ability of the bacteria to form biofilms. An evaluation of genetic changes that occur when microbes engage βA-HMOs, and the spectrum of microbes that are susceptible to these interesting molecules will be reported in due course.

Materials and Methods

All reactions were performed in a microwave compatible vial. The reagents and solvents were purchased from commercial sources and were used as received unless mentioned otherwise. 2’-fucosyllactose, 3-fucosyllactose, and 6’-siallylactose were purchased from Carbosynth. Lacto-N-tetrose (LNT) was synthesized as previously described.³⁵ The MW experiments were conducted in a closed reaction vessel using an Anton Paar G10 Monowave 200 (8.0 mL microwave vial), capped with reusable snap caps and silicon-Teflon septa. All reactions were irradiated for 1 h at 200 W and 50 °C. ¹H NMR spectra were obtained on a Bruker 600 MHz spectrometer and are reported relative to deuterated solvent signals. Data for ¹H NMR spectra are presented as follows: chemical shift (δ ppm), multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, p = pentet, m = multiplet, br = broad, app = apparent), coupling constants (Hz) and integration. Deuterated methanol was standardized to 3.31 ppm. ¹³C NMR spectra were obtained on a Bruker 151 MHz spectrometer and are reported relative to deuterated solvent signals. Deuterated methanol was standardized to 49.0 ppm. Structural assignments were made with additional information from gHSQC experiments. High-resolution mass spectra (HRMS) were obtained from the Department of Chemistry, Vanderbilt University using a Synapt G2-S HDMS TOF (Milford, Ma, USA) mass spectrometer.

Microwave Activated Kochetkov Amination Procedure

HMO (0.1 mmol) was dissolved in reaction solvent (2 mL), ammonia source (5 × mass of HMO) was added and the reaction was irradiated for 1 h at 200 W and 50°C. The reaction mixture was diluted to 45 mL with water in a 50 mL conical centrifuge tube, frozen with liquid nitrogen, and lyophilized repeatedly until a constant mass of white solid was obtained. Ratio of conversion was determined by integration of the C-1 anomeric protons of the starting material to that of the desired product (Table S1).

Thermally Activated Kochetkov Amination Procedure

HMO (0.1 mmol) was dissolved in reaction solvent (2 mL), ammonia source (5 × mass of HMO) was added and the reaction warmed for 48 h at 40°C in an oil bath. The heating medium was silicone oil purchased from Sigma-Aldrich (St. Louis, MO, USA) and was carried out in a Pyrex® crystalizing dish. The reaction mixture was diluted to 45 mL with water in a 50 mL conical centrifuge tube, frozen with liquid nitrogen, and lyophilized repeatedly until a constant mass of white solid was obtained. Ratio of conversion was determined by integration of the C-1 anomeric protons of the starting material to that of the desired product (Table S1).

Compound characterization

(2S,3S,4R,5S,6S)-2-(((2S,3R,4S,5R,6R)-2-(((2R,3S,4R,5R)-6-amino-4,5-dihydroxy-2-(hydroxymethyl)tetrahydro-2H-pyran-3-yl)oxy)-4,5-dihydroxy-6-(hydroxymethyl)tetrahydro-2H-pyran-3-yl)oxy)-6-methyltetrahydro-2H-pyran-3,4,5-triol (2, βA-2’-FL): White solid, 37 mg, 77%, 1:4.7 ratio of HMO:βA-HMO; Spectral data for 2 was consistent with known values.¹⁵

(2S,3S,4R,5S,6S)-2-(((3R,4R,5R,6R)-2-amino-3-hydroxy-6-(hydroxymethyl)-5-(((2S,3R,4S,5R,6R)-3,4,5-trihydroxy-6-(hydroxymethyl)tetrahydro-2H-pyran-2-yl)oxy)tetrahydro-2H-pyran-4-yl)oxy)-6-methyltetrahydro-2H-pyran-3,4,5-triol (3, βA-3-FL): White solid; 42 mg, 86%, 1:12 ratio of HMO:βA-HMO; R_f 0.15 (60 : 30 : 5 : 5 CHCl₃ : CH₃OH : AcOH : H₂O); ¹H NMR (600 MHz, MeOD): δ 5.42 (d, J = 3.9 Hz, 0.92H), 4.84 – 4.78 (m, 2.33H), 4.40 (d, J = 7.3 Hz, 1.02H), 3.97 (d, J = 8.7 Hz, 1.09H), 3.94 (dd, J = 10.2, 3.4 Hz, 1.38H), 3.92 – 3.65 (m, 13.23H), 3.53 – 3.42 (m, 4.29H), 3.38 (dt, J = 9.8, 3.1, 3.1 Hz, 1.05H), 3.28 (t, J = 8.8, 8.8 Hz, 1.01H), 1.20 – 1.18 (m, 3H);¹³C{¹H} NMR (151 MHz, MeOD): δ 102.4, 98.9, 85.5, 78.9, 77.2, 76.4, 75.3, 73.5, 73.1, 72.4, 71.5, 69.8, 69.5, 69.0, 68.6, 66.0, 61.5, 39.1, 15.2; IR (ATR) ν = 3344, 3099, 2978, 1617, 1490 cm⁻¹; HR-ESI-MS (m/z): calcd for C₁₈H₃₂NO₁₄⁻ (M-H)⁻ 486.1823, found 486.1816.

(2R,4S,5R,6R)-5-acetamido-2-(((2R,3R,4S,5R,6S)-6-(((2R,3S,4R,5R)-6-amino-4,5-dihydroxy-2-(hydroxymethyl)tetrahydro-2H-pyran-3-yl)oxy)-3,4,5-trihydroxytetrahydro-2H-pyran-2-yl)methoxy)-4-hydroxy-6-((1R,2R)-1,2,3-trihydroxypropyl)tetrahydro-2H-pyran-2-carboxylic acid (4, βA-6’-SL): White solid; 50 mg, 80%, 1:10 ratio of HMO:βA-HMO; R_f 0.10 (60 : 30 : 5 : 5 CHCl₃ : CH₃OH : AcOH : H₂O); ¹H NMR (600 MHz, MeOD): δ 5.13 (d, J = 3.7 Hz, 0.51H), 4.53 (d, J = 7.9 Hz, 0.62H), 4.37 – 4.32 (m, 2.21H), 4.05 (dq, J = 10.1, 7.6, 7.4, 7.4 Hz, 2.34H), 4.01 (d, J = 8.6 Hz, 1.00H), 3.94 – 3.76 (m, 14.65H), 3.76 – 3.71 (m, 2.36H), 3.72 – 3.61 (m, 9.14H), 3.59 – 3.49 (m, 8.46H), 3.50 – 3.42 (m, 4.23H), 3.28 – 3.22 (m, 0.60H), 3.17 – 3.12 (m, 0.86H), 2.81 (dtd, J = 12.4, 5.6, 5.1, 2.8 Hz, 2.24H), 2.04 – 2.00 (m, 6.72H), 1.71 – 1.63 (m, 2.18H); ¹³C{¹H} NMR (151 MHz, MeOD): δ 173.6, 173.1, 103.8, 100.1, 96.5, 85.1, 80.4, 80.3, 76.2, 75.6, 75.1, 74.8, 74.7, 74.4, 73.2, 72.8, 71.8, 71.0, 69.9, 69.2, 68.8, 68.4, 63.3, 63.1, 60.9, 52.4, 41.1, 39.0, 21.4; IR (ATR) ν = 3489, 3002, 2978, 1725, 1627, 1490 cm⁻¹; HR-ESI-MS (m/z): calcd for C₂₃H₃₉N₂O₁₈⁻ (M-H)⁻ 632.2276, found 632.2270.

N-((2S,3R,4R,5S,6R)-2-(((2S,3R,4S,5S,6R)-2-(((2R,3S,4R,5R)-6-amino-4,5-dihydroxy-2-(hydroxymethyl)tetrahydro-2H-pyran-3-yl)oxy)-3,5-dihydroxy-6-(hydroxymethyl)tetrahydro-2H-pyran-4-yl)oxy)-5-hydroxy-6-(hydroxymethyl)-4-(((2R,3R,4S,5R,6R)-3,4,5-trihydroxy-6-(hydroxymethyl)tetrahydro-2H-pyran-2-yl)oxy)tetrahydro-2H-pyran-3-yl)acetamide (5, βA-LNT): White solid; 54 mg, 76%, 1:4.8 ratio of HMO:βA-HMO; Spectral data for 5 was consistent with known values.²⁸

Bacterial Strains and Culture Conditions

S. agalactiae strain GB00590 is a clinical isolate provided by Dr. Shannon Manning at Michigan State University. S. aureus strain USA300 is the laboratory-adapted strain USA300 JE2; USA300 JE2 is derived from the parental community-associated methicillin-resistant S. aureus isolate USA300. All strains were grown on tryptic soy agar plates supplemented with 5% sheep blood (blood agar plates) at 37 °C in ambient air overnight. Strains were subcultured from blood agar plates into 5 mL of Todd-Hewitt broth (THB) and incubated under shaking conditions at 180 rpm at 37 °C in ambient air overnight. Following overnight incubation, bacterial density was quantified through absorbance readings at 600 nm (OD₆₀₀) using a Promega GloMax-Multi Detection System plate reader. Bacterial numbers were determined using the predetermined coefficient of 1 OD₆₀₀ = 10⁹ CFU/mL.

Bacterial Growth Assays

Bacterial strains were grown overnight as described above and used to inoculate fresh THB at a multiplicity of infection (MOI) of 10⁶ colony forming units per 200 μL of growth medium in 96 well tissue culture treated, sterile polystyrene plates (Corning, Inc). HMOs and βA-HMOs were dissolved in DI water to achieve a concentration of 80 mg/mL and filtered through a 0.2 μm syringe filter. HMOs or βA-HMOs were added to achieve a final carbohydrate concentration of ca. 5 mg/mL. Bacteria grown in THB in the absence of any HMOs served as the control. Cultures were grown under static conditions at 37 °C in ambient air for 24 h. Growth was quantified through spectrophotometric reading at OD₆₀₀ with readings taken at 0, 2, 4, 6, 7, and 8 hours then a final reading at 24 hours.

Bacterial Biofilm Assays

Bacterial strains were grown overnight as described above and used to inoculate fresh THB at a multiplicity of infection (MOI) of 10⁶ colony forming units per 200 μL of growth medium in 96 well tissue culture treated, sterile polystyrene plates (Corning, Inc.). HMOs and βA-HMOs were dissolved in DI water to achieve a concentration of 80 mg/mL and filtered through a 0.2 μm syringe filter. HMOs or βA-HMOs were added to achieve a final carbohydrate concentration of ca. 5 mg/mL. Bacteria grown in THB in the absence of any HMOs served as the control. Cultures were incubated under static conditions at 37 °C in ambient air for 24 h. Bacterial growth was quantified through absorbance readings at an optical density of 600 nm (OD₆₀₀). Following growth quantification, the culture medium was removed, and wells were washed gently with phosphate buffered saline (PBS, pH 7.4) to remove nonadherent cells. The remaining biofilms were stained with a 10% crystal violet solution for 10 min. Following staining, wells were washed with PBS and allowed to dry at room temperature for at least 30 min. The remaining crystal violet stain was solubilized with 200 μL of 80% ethanol/20% acetone solution. Biofilm formation was then quantified through absorbance readings at an optical density of 560 nm (OD₅₆₀). Results are expressed as biofilm/biomass ratios (OD₅₆₀/OD₆₀₀).

High resolution field-emission gun scanning electron microscopy (FEG-SEM) analyses

Bacterial biofilms were analyzed via FEG-SEM as previously described.^32–34 Briefly, bacterial cells were cultured in biofilms adhering to glass coverslips coated with poly-L-lysine overnight in the culture conditions described above. HMOs and βA-HMOs were dissolved in DI water to achieve a concentration of 80 mg/mL and filtered through a 0.2 μm syringe filter. HMOs or βA-HMOs were added to achieve a final carbohydrate concentration of ca. 5 mg/mL. The following day, bacterial cells were fixed in a solution of 2.5% glutaraldehyde, 2.0% paraformaldehyde, and 0.05 M sodium cacodylate buffer pH 7.4. Samples were dehydrated with sequential washes of increasing concentrations of ethanol before being subjected to critical point drying, mounting on aluminum stubs, and sputter coating with 20 nm of gold-palladium. Samples were viewed using an FEI Quanta 250 field-emission gun scanning electron microscope at 5 kEV with a spot size of 2.5.

Statistical Analysis

All data shown signify three independent experiments each with three technical replicates. Data are expressed as the mean ± SEM. Statistical analyses were performed in GraphPad Prism Software v. 8.2.1. Statistical significance for growth was determined using two-way ANOVA with post hoc Dunnett’s multiple comparison test comparing growth in the presence of HMOs or βA-HMOs to growth in media alone. Statistical significance for biofilm production was determined using one-way ANOVA with post hoc Dunnett’s multiple comparison test comparing biofilm production in the presence of HMOs or βA-HMOs to biofilm production in media alone.