Human Milk Oligosaccharide Specificities of Human Galectins

Mar 27, 2017 - The affinities of thirty-two free human milk oligosaccharides (HMOs) for four human galectin proteins, a stable mutant of hGal1 (hGal-1...
1 downloads 0 Views 1MB Size
Subscriber access provided by University of Newcastle, Australia

Article

Human Milk Oligosaccharide Specificities of Human Galectins. Comparison of Electrospray Ionization Mass Spectrometry and Glycan Microarray Screening Results Km Shams-Ud-Doha, Elena N Kitova, Pavel I. Kitov, Yves St-Pierre, and John S. Klassen Anal. Chem., Just Accepted Manuscript • Publication Date (Web): 27 Mar 2017 Downloaded from http://pubs.acs.org on March 27, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Analytical Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Human Milk Oligosaccharide Specificities of Human Galectins. Comparison of Electrospray Ionization Mass Spectrometry and Glycan Microarray Screening Results Km Shams-Ud-Doha,1,§ Elena N. Kitova,1 Pavel I. Kitov,1 Yves St-Pierre2 and John S. Klassen1* 1

Alberta Glycomics Centre and Department of Chemistry, University of Alberta, Edmonton, Alberta, Canada T6G 2G2 2

INRS-Institut Armand-Frappier, Laval, Québec, Canada H7V 1B7

*Corresponding Author’s address: Department of Chemistry University of Alberta Edmonton, AB CANADA T6G 2G2 Email: [email protected] Telephone: (780) 492-3501 Fax: (780) 492 8231

§

Current address: Sanford Burnham Prebys Medical Discovery Institute, 10901 North Torrey

Pines Road, La Jolla, California 92037 USA

1 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 29

Abstract The affinities of thirty-two free human milk oligosaccharides (HMOs) for four human galectin proteins – a stable mutant of hGal1 (hGal-1), a C-terminal fragment of hGal-3 (hGal-3C), hGal-7 and an N-terminal fragment of hGal-9 (hGal-9N) – were measured using electrospray ionization mass spectrometry (ESI-MS). The binding data show that each of the four galectins recognize the majority of the HMOs tested (hGal-1 binds thirty-two HMOs, hGal-3C binds twenty-six, hGal-7 binds thirty-one and hGal-9N binds twenty-six). Twenty-five of the HMOs tested bind all four galectins, with affinities ranging from 103 M-1 to 105 M-1. The reliability of the ESI-MS assay for quantifying the affinities of HMOs for lectins was established from the agreement found between the ESI-MS data and affinities of a small number of HMOs for hGal-1, hGal-3C and hGal-7 measured by isothermal titration calorimetry (ITC). Comparison of the relative affinities (of fourteen HMOs) measured by ESI-MS with the reported specificities of hGal-1, hGal-3, hGal-7 and hGal-9 for these same HMOs established using the shotgun human milk glycan microarray (HM-SGM-v2) showed fair-to-poor correlation, with evidence of false positives and false negatives in the microarray data. The results of this study suggest that HMO specificities of lectins established using microarrays may not accurately reflect their true HMObinding properties and that the use of “in solution” assays such as ESI-MS and ITC is to be preferred.

2 ACS Paragon Plus Environment

Page 3 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Introduction Human milk contains a variety of bioactive components, including proteins, glycoproteins, fat globules and free oligosaccharides (commonly referred to as human milk oligosaccharides or HMOs).1,2 HMOs contain lactose (Lac, β-D-Gal(1→4)-β-D-Glc) at the reducing end and are extended at the non-reducing end by N-acetyllactosamine (LacNAc) type 1 (β-D-Gal(1→3)-β-DGlcNAc) or type 2 (β-D-Gal(1→4)-β-D-GlcNAc) repeating units; fucose (Fuc) and sialic acid (Sia) residues can be attached via α-(1→2), α-(1→3) or α-(1→4), and α-(2→3) or α-(2→6) linkages, respectively.3 The total mass concentration of HMOs in breast milk is reported to be between 5 and 23 g L-1, depending on lactation period, and more than two hundred different HMOs have been identified so far.4-8 Fucosylated oligosaccharides are the most abundant HMOs (6 to 12 g L-1); non-fucosylated neutral and sialylated HMOs are present at concentrations of 1 to 6 g L-1 and 0.5 to 3 g L-1, respectively.9,10 Although HMOs are non-digestible and concentrated in the intestinal tracts of breast fed infants, they can also enter systemic circulation and have effects outside of the gastrointestinal tract on local and systemic levels.11 The reported health effects of HMOs are numerous - they serve as prebiotics and anti-adhesive antimicrobials against intestinal and urinary pathogens, can modulate the immune response and regulate gene expression in intestinal epithelial cells.12-18 Given the importance of HMOs to infant health, there is currently considerable interest in identifying their protein receptors in humans, as well as those associated with viral and bacterial pathogens. At present, shotgun and defined human milk glycan (HMG) microarrays, developed by Cummings and coworkers, represent the dominant technology for evaluating the HMO-binding specifities of lectins.8,19-22 The most recent shotgun HMG microarray (version 2, HM-SGM-v2) consists of two-hundred and forty-seven oligosaccharide fractions prepared from pooled human

3 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 29

milk, as well as twelve commercially available HMOs.8,20 The HMOs used in the array are modified at the reducing end by reductive amination (such that the terminal Glc residue is in open ring conformation) and immobilized on an N-hydroxysuccinimide-activated glass slide.23-26 Of the two-hundred and forty-seven HMG fractions present in the HM-SGM-v2, the structures present in thirty-two of the fractions were interrogated using multiple stages of mass spectrometry (i.e, MSn) and using lectins and antibodies with known specificities.8,20 The defined HMG microarray consists of eight commercially available HMOs derivatized by either reductive amination or using a chemistry that maintains the reducing end ring structure in pyranose form (giving in total sixteen modified HMO structures).21 Typically, 1-10 mg of each oligosaccharide is used to prepare conjugates for immobilization on the surface (50-100 fM of glycan conjugate per spot).19,26 The minimum requirement for unlabelled lyophilized protein is 0.5 mg.27 Screening of the microarrays is carried out by applying 50-100 µL of a solution of target protein to the array, at different concentrations (0.2-200 µg mL-1), followed by a washing step.8,21,25 Binding is detected by fluorescence, using fluorophore-labeled streptavidin or specific antibodies.8,21,22 Screening results are reported in terms of relative fluorescence units (RFU), which are believed to reflect the relative affinities of the glycans for the target protein.8,19,21,22 Notably, absolute affinities cannot be established from microarray data. Using the HMG microarrays, the HMO specificities for number of human, plant and pathogen-generated lectins have been investigated recently, including human galectins (hGal-1, hGal-2, hGal-3, hGal-4, hGal-7, hGal-8 and hGal-9),21 C-type lectins (DC-SIGN, langerin, dectin-2, MGL),22 siglecs 1, 5, 7, 9 and 10,22 plant lectins AAL, UEA-I, LTL, SNA, RCA-I, ECL, GSL-II, ConA, MAL-I,19 antibodies,19 human influenza viruses,19 parvovirus minute

4 ACS Paragon Plus Environment

Page 5 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

viruses of mice19 and rotaviruses.8 Notably, the binding data suggest that many of lectins investigated recognize HMOs. While the HMG microarrays represent a sensitive and high-thoroughput approach to HMO screening, they have several drawbacks. First, it is well known that binding data acquired from glycan microarrays may exhibit a dependence on the size/nature of the nature of the linker used to immobilize the oligosaccharide on the surface.28,29 Moreover, in the case of HMOs, the terminal lactose residue often plays an important role in molecular recognition30,31 and any modification to the reducing end could influence lectin binding. Also, it is well known that low affinity interactions are often missed in glycan array screening.32-34

To the best of our

knowledge, no comprehensive comparisons of the HMO binding specificities determined by HMG microarrays with data acquired using other binding assays have been reported. Electrospray ionization (ESI)-MS has recently emerged as rapid and sensitive method for detecting protein-carbohydrate interactions and measuring their affinities in vitro.35,36 Notably, the assay, which relies on the quantification of free and ligand-bound protein ions, is label-free and doesn’t require immobilization of any of the binding partners and, as such, is ideally suited for studying lectin-HMO interactions.37 The assay, which typically takes 1-2 min per measurement, has sample requirements similar to those of HMG array screening. For example, given that typical protein concentrations used in ESI-MS measurements are in the low µM range, 0.5 mg of protein is sufficient for 102 - 103 measurements (depending on protein MW). Similarly, 0.1 mg of a 1000 Da HMO would allow for 102 - 103 individual measurements (depending on affinity). Moreover, the ESI-MS assay is quantitative and can measure association constants (Ka) in the 102 – 107 M-1 range.36,38 As described in detail elsewhere, there are a number of potential sources of error in ESI-MS affinity measurements.36 However, these sources of error are well

5 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 29

understood and their effects on affinity measurements can be minimized using established protocols.36 As a result, protein-carbohydrate affinities measured by ESI-MS are generally found to be in good agreement with values measured using other binding assays.39,40 Here, we used ESI-MS to measure the affinities of thirty-two pure HMOs for four human galectin proteins - a stable mutant of hGal1 (hGal-1), a C-terminal fragment of hGal-3 (hGal3C), hGal-7 and an N-terminal fragment of hGal-9 (hGal-9N). Comparison of these affinities with available binding data measured using isothermal titration calorimetry (ITC) served to establish the reliability of the ESI-MS affinities. ITC is considered the gold standard for quantifying protein-carbohydrate interactions in vitro. However, the assay requires significant quantities of protein and carbohydrate (several mg of each). Moreover, ITC is not readily applied to weak (90% in all cases, are listed in Table S2

6 ACS Paragon Plus Environment

Page 7 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

(Supporting Information). The HMOs L6, L9, L16, L18, L24 and L30 were chemically modified according to the procedure described in Scheme S1 (Supporting Information). The structures of the modified HMOs (open-ring structures designated as L6-MA, L9-MA, L16-MA, L18-MA, L24-MA and L30-MA; closed-ring structures designated as L6-PA, L9-PA, L16-PA, L18-PA, L24-PA and L30-PA) are shown in Figure S1 (Supporting Information). Stock solutions of each oligosaccharide were prepared by dissolving a known mass of compound in a known volume of ultra-filtered water (Milli-Q, Millipore, Billerica, MA) to achieve a final concentration of ~1 mM. All stock solutions were stored at -20 °C until needed. Proteins S-carboxyamidomethylated oxidation resistant (C2S substituted to improve stability) recombinant hGal-1 (dimer MW 29 235 Da) was a gift from S. Sato (Laval University).41 Notably, the C2S modification is believed not to affect the carbohydrate affinity of the protein.42 The galectin proteins, hGal-3 (MW 26 152 Da), the recombinant fragment of the C-terminus (residues 107–250) carbohydrate recognition domain of hGal-3 (hGal-3C, MW 16 327 Da)43 and recombinant fragment of N-terminus (residues 1–148) carbohydrate recognition domain of human galectin-9 (hGal-9N, MW 18 408 Da)44 were gifts from Prof. C. Cairo (University of Alberta). Bovine ubiquitin (MW 8 565 Da), which served as a reference protein (Pref) for the ESI-MS binding measurements, was purchased from Sigma-Aldrich Canada (Oakville, Canada). Recombinant hGal-7 (dimer MW 29 888 Da) was produced and purified as described previously.45 All proteins were dialyzed against an aqueous solution of 200 mM ammonium acetate (pH 6.8) using 0.5 mL Amicon microconcentrator (EMD Millipore, Billerica, MA), MW cut-off of 10 kDa and stored at -20 oC until required. The concentrations of protein stock solutions were estimated by UV absorption (280 nm).

7 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 29

Isothermal Titration Calorimetry Affinity measurements were carried out on L9 binding to hGal-3C using a VP-ITC (MicroCal, Inc., Northampton, MA) at 30 oC. Each of hGal-3C and L9 were dissolved in PBS buffer (0.14 M NaCl, 2.7 mM KCl, 0.01 M Na2HPO4, 1.8 mM KH2PO4 and 14 mM β-mercaptoethanol at pH 7.4) and stored at 4°C until used. At the concentrations needed for ITC, ~0.1 mM, hGal-3C was found to be unstable in aqueous ammonium acetate solutions and precipitate. Therefore, PBS buffer was used for the ITC measurements. All solutions were filtered through 0.22 µm cutoff filters (EMD Millipore, Billerica, MA) and extensively degassed with stirring prior to use. For the hGal-3C and L9 interaction, a 0.109 mM hGal-3C solution (in the sample cell) was titrated with a 1.1 mM L9 solution. Mass spectrometry The ESI-MS binding measurements were carried out in positive ion mode using a 9.4T ApexQe FTICR mass spectrometer (Bruker Daltonic, Billerica, MA) and a Synapt G2 quadrupole-ion mobility separation-time-of-flight (Q-IMS-TOF) mass spectrometer (Waters UK Ltd., Manchester, UK). In both cases, nanoflow ESI (nanoESI) was performed using borosilicate glass tips (1.0 mm o.d., 0.68 mm i.d.) pulled to ~5 µm o.d. at one end using a P-2000 micropipette puller (Sutter Instruments, Novato, CA). A capillary voltage of ~1.0 kV was applied to a Pt wire in the nanoESI tip to carry out ESI. A brief description of the instrumental conditions and data analysis procedures used is given as Supporting Information. Results and Discussions a. HMOs binding specificities of hGal-1, hGal-3C, hGal-7 and hGal-9N The four galectin proteins, hGal-1, hGal-3C, hGal-7 and hGal-9N, are readily amenable to ESIMS binding measurements. Shown in Figures 2a-d are representative mass spectra acquired for

8 ACS Paragon Plus Environment

Page 9 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

aqueous ammonium acetate (pH 6.8, 25 °C) solutions of each of the galectins. For prototype galectins hGal-1 and hGal-7, which are known to exist as noncovalent dimers in neutral solution, ion signal corresponding to protonated homodimer (at charge states from +9 to +11) dominated the mass spectra (Figures 2a and 2c). Signal corresponding to protonated monomer ions was also detected but at low abundance (~4% of protein signal for hGal-1, ~5% for hGal-7. The ESI mass spectra acquired for hGal-3C and hGal-9N revealed signal for the monomer ions, at charge states +7 and +8, consistent with the known oligomeric state of hGal-3, a chimera type galectin, and hGal-9, a tandem-repeat galectin (Figures 2b and 2d).46 For hGal-9N, signal corresponding to multiple adducts, including Ni2+ (presumably originating from Ni-column used to purify the Histagged hGal-9N), were detected (Figure S2, Supporting Information). For the purposes of illustration, ESI mass spectra acquired for solutions of each galectin with L9 (commonly referred to as LNT) are shown in Figures 2e-h. At the concentrations investigated, both free and L9-bound hGal-3C and hGal-9N monomer ions were detected (Figures 2f and 2h, respectively). In the case of hGal-9N, it was established, based on the similarities in the abundance ratios of L9-bound to unbound hGal-9N (i.e., R), that the presence of adducts does not measurably affect L9 binding (Figure S2, Supporting Information). Analysis of the mass spectra revealed that the abundance ratios of ligand-bound to unbound protein (i.e., R) measured for the different protein adducts species are indistinguishable, within experimental error. For dimeric hGal-12 and hGal-72, ions corresponding to one bound L9 were detected, (Figures 2e and 2g). From the measured R values, the affinities of L9 for the four galectin proteins were determined. Analogous binding measurements were carried out at multiple galectin/HMO concentrations for all thirty-two HMOs. The resulting Ka values are listed in Table S1 (Supporting Information) and, for ease of comparison, plotted in Figure 3. Affinity

9 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 29

measurements were also performed on L33 binding to the four galectin proteins (Table S1, Supporting Information). Inspection of the binding data reveals that all four galectins recognize the majority of the HMOs tested, with many of the interactions having moderate-to-high affinities (i.e., >104 M-1). Remarkably, hGal-1 binds to all thirty-two HMOs; the affinities range from 0.2 x 104 M-1 to 5.3 x 104 M-1. Eight of the HMO ligands exhibit affinities ≥104 M-1, with L9 having the highest affinity for hGal-1. With the exception of L27, hGal-7 recognizes all of the HMOs, although the affinities span a narrow range, from 0.1 x 104 M-1 to 1.4 x 104 M-1. Both hGal-3C and hGal-9N bind to twenty-six HMOs, with all but two and five of these, respectively, having affinities >104 M-1; L5 has the highest affinity for hGal-3C (13.0 x 104 M-1) and L4 has the highest affinity for hGal-9N (41.5 x 104 M-1). Analysis of the binding data also reveals that twenty-five of the HMOs (L1-L19, L21-L23 and L30-L32) are common ligands to all four galectins, although there are significant differences in the affinities for each protein. This “relaxed” HMO specificity is not surprising given that the amino acid residues that make up the carbohydrate binding site are highly conserved among all human galectins.46 It is also notable that both L16 and L9, which are the most abundant HMOs found in the milk of secretors and non-secretors, respectively,9 bind to all four galectins with moderate-to-high affinity. This finding reveals that, regardless of secretor status of the mother, all human milk is expected to contain mg L-1 amounts of HMOs that have relatively high affinities for these and, presumably, other human galectins. The important role of the Lac and LacNAc motifs in the recognition of HMOs by galectins is highlighted by the binding data in Table S1 (Supporting Information). For example, that L24, L26-L29 exhibit weak or no binding to the four galectins can be explained by the α(1→3) fucosylation of Glc.30,47 The relatively high affinities measured for L17 and L21, which

10 ACS Paragon Plus Environment

Page 11 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

are also fucosylated at Glc, suggest that the LacNAc motif present in these oligosaccharides serves instead as the binding epitope. This conclusion is also supported by the measurable affinities of L33 for the four galectins. The binding data also reveal that α(2→6) sialylation of terminal Gal or GlcNAc residues, adjacent to Lac, may also lead to a significant reduction in affinity. For example, L25, which is α(2→6) sialylated Lac, doesn’t bind hGal-3C or hGal-9N, while α(2→6) sialylation of L3, L11 and L9 (to give L14, L20 and L22, respectively) leads to decrease or loss of affinity for hGal-9N. These findings, which are in agreement with previous studies,30,48 highlight the significance of the 6-OH group of Gal in the Lac motif in HMOs for the binding of human galectins. α(2→3) sialylation of terminal Gal has a less pronounced effect on binding, consistent with X-ray crystal structures that show that the 3-OH group in Gal is not involved in interactions with hGal-1, hGal-3C, hGal-7 and hGal-9N.49-52 b. Comparison of ESI-MS and ITC binding data The affinities for a small number of HMOs for hGal-1, hGal-3C and hGal-7 have been measured, previously and in the present study, using ITC.21,49,53 ITC is generally regarded as the gold standard for quantifying the affinities of protein-carbohydrate interactions in vitro. Consequently, comparison of the binding data reported here with the ITC-derived values provides an excellent test of the reliability of the HMO affinities measured by ESI-MS. Plotted in Figure 4 are reported ITC-derived affinities (corrected to 25 °C) of L30 for hGal-1 and hGal-3C with L3049,53 and L3, L6, L9, L16 and L24 for hGal-7,21 as well as the affinity L9 for hGal-3C (Figure S3, Supporting Information) and the corresponding affinities measured by ESI-MS. Notably, there is good linear correlation between the data sets, with a slope of 0.88 ± 0.04. The agreement between the affinities measured by ITC and ESI-MS, which is consistent with the findings for other protein-

11 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 29

carbohydrate interactions,39,54,55 is significant as it establishes the reliability of the ESI-MS assay for quantifying HMO affinities and establishing lectin HMO specificities. c. Comparison of ESI-MS affinities and HM-SGM-v2 microarray data The HMO binding data measured in the present study, while of interest on their own, also provide an unprecedented opportunity to evaluate the reliability of the HMG microarrays for establishing HMO specificities. Shown in Figures 5 and S5 (Supporting Information) are comparisons of the ESI-MS affinities of L2-L4, L6, L9, L11, L13, L15, L16, L20, L22, L24, L26, L30 with reported data for the binding of the conjugates of these same HMOs to hGal-1, hGal-3, hGal-7 and hGal-9, which were measured using the HM-SGM-v2 array.21 Although absolute affinities can’t be measured using microarrays, the relative HMO affinities (or at least trends in affinities) can be inferred from the reported RFU values.8 Overall, the correlation between the ESI-MS and HM-SGM-v2 data sets is modest to poor. Moreover, there is evidence of both false positives and false negatives in the microarray data. A brief summary of the results of this comparison is given below for each of the galectins. hGal-1: The best agreement between the microarray data and ESI-MS affinities is found for hGal-1 (Figure 5a and S5a, Supporting Information); an average correlation coefficient of 0.71±0.1 is found for the three protein concentrations (2 µg mL-1, 20 µg mL-1 and 200 µg mL-1) used for the array screening.21 Although, L9 was correctly identified as the strongest binder (of these fourteen HMOs) from the array data, the trends in RFU observed for the other HMO ligands (at all protein concentrations) deviate from the trend in affinities measured by ESI-MS. A striking example is the trend observed for L3, L6 and L9 – according to the array, these three ligands have similar RFU values, while the Ka values of L3 and L6 are seven-fold and four-fold lower, respectively, than that of L9.

12 ACS Paragon Plus Environment

Page 13 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

hGal-3/hGal-3C: Comparison of the ESI-MS (measured for hGal-3C) and HMG array data (measured for hGal-3) yields correlation coefficients of 0.56 ±0.03 (average of values obtained for 2 µg mL-1, 20 µg mL-1 and 200 µg mL-1) (Figure 5b and S5b, Supporting Information). Although the highest affinity ligands (L4 and L3) identified from the array data (at all three protein concentrations) are consistent with the ESI-MS results, the trends for the other ligands are in poor agreement. Most notably, L30 is preferred over all but L3 and L4 according to the array, while the ESI-MS data indicate that it is the weakest of the HMO ligands. Moreover, L24 and L26, which do not bind to hGal-3C (according to the ESI-MS data), are found to exhibit weak but measurable RFU values, comparable to those measured for the majority of the HMOs considered. These results indicate either the array screening gave false positives or that the signal-to-noise in these measurements is too low to reliably distinguish many HMO binders from non-binders. It must be noted that the HM-SGM-v2 array data reported by Noll et al. were obtained using hGal-3,21 while the ESI-MS affinity measurements were performed on the hGal-3C fragment. It is, therefore, possible that some of the apparent discrepancies between the array and ESI-MS data may be due differences in the HMOs specificities of the two proteins. To rule this out, the affinities of seven HMOs (L6, L7, L9, L16, L18, L24 and L30) for hGal-3 were measured by ESI-MS. According to the ESI mass spectrum acquired for a 200 mM aqueous ammonium acetate solution (pH 6.8, 25 °C) of (a) hGal-3 (3 µM) (Figure S4a, Supporting Information), a fraction of hGal-3 is unfolded under the solution conditions used for the ESI-MS binding measurements. Additionally, there is evidence of several isoforms of hGal-3. The MW of one of the most abundant isoforms (26 021 Da) is consistent with the known sequence (but devoid of N-terminal Met);57 the other major isoforms having either slightly larger (26 061 Da)

13 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 29

or lower MWs (≤25 235 Da). Given the resulting uncertainty in the concentrations of the different forms of hGal-3, a competitive ESI-MS assay was employed, wherein hGal-3C was added to the hGal-3 solution. As described in the Experimental section, the ratios of the abundance ratios of HMO-bound to free protein (measured for hGal-3C and the different forms of hGal-3) will reflect their affinity ratios (eq S3), independent of the protein concentration. Shown in Figure S4b (Supporting Information), is a representative ESI mass spectrum acquired in positive ion mode for a 200 mM aqueous ammonium acetate solution (pH 6.8, 25 °C) of hGal3 (5 µM), hGal-3C (4 µM), Pref (2 µM) and L16 (20 µM). Notably, the ratio of R values measured for the 26 021 Da isoform of hGal-3 (i.e., hGal-3ꞌ) and hGal-3C is 0.97 ± 0.02 (Table S3, Supporting Information), indicating similar Ka values for L16 binding to the two proteins. Analogous measurements performed on L6, L7, L9, L18, L24 and L30 produced similar results. hGal-7: Overall, there is quite poor agreement between the ESI-MS affinities and RFU values obtained from the HM-SGM-v2 array - the correlation coefficients are 0.31 (2 µg mL-1) and ~0.4 (20 µg mL-1 and 200 µg mL-1) (Figure 5c and S5c, Supporting Information). According to the array data, L3 exhibits the strongest binding of the fourteen HMOs and all but L3 and L9 exhibit weak or no binding. This finding contrasts the ESI-MS affinity data, which indicate that hGal-7 recognizes twelve of the fourteen HMOs tested, with affinities ranging from 3 x 103 to 1.1 x 104 M-1; L9 was found to exhibit the highest affinity. Also notable is the fact that no binding between hGal-7 and L24 was detected by ESI-MS, which can be explained by α(1→3) fucosylation of the Glc residue. However, the HM-SGM-v2 data obtained at 200 µg mL-1 Gal-7 indicates binding to L24 (i.e. a false positive result). hGal-9/hGal-9N: The HM-SGM-v2 array screening of the fourteen HMOs against hGal-9, which contains two non-equivalent carbohydrate binding sites, produced uniformly weak

14 ACS Paragon Plus Environment

Page 15 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

fluorescence signal21 and there was no correlation between the HM-SGM-v2 data and the ESIMS affinities (measured for hGal-9N fragment) (Figure 5d and S5d, Supporting Information). According to the ESI-MS binding data, hGal-9N binds strongly to the majority of these HMOs – eleven are ligands with affinities as high as 4x105 M-1 (L4). Moreover, according to the array data, Lac (L30) is preferred over the other thirteen HMOs by hGal-9 (at all concentrations); this disaccharide has the lowest affinity of the HMO ligands measured by ESI-MS. Open-ring vs closed-ring HMG derivatives: The aforementioned comparative analysis reveals that the HMO specificities of hGal-1, hGal-3, hGal-7 and hGal-9 established using the HMSGM-v2 array exhibit modest-to-poor correlation with the corresponding HMO affinities. The array screening also produced false positives and false negatives. As noted elsewhere,21 one possible source of error in the data generated with the HM-SGM-v2 is the modification of the reducing end Glc, which converts it to the “open-ring” form. Given the importance of the Lac moiety to binding, it is reasonable that this modification could influence HMO interactions with lectins. Using the defined HMG array, which contains immobilized HMOs with Glc in both the “open-ring” and “closed-ring” forms, Noll et al. investigated the effect of Glc modification.21 It was reported that, overall, the “closed-ring” HMOs were preferred by most galectins.21 The most dramatic effect was found for hGal-9, where only the “closed-ring” forms of L3, L6 and L9 exhibited detectable binding (no binding was observed for either form of the other HMOs).21 Importantly, however, neither the open-ring” nor “closed-ring” HMO conjugates in the defined HMG array yield trends in binding that are consistent with the ESI-MS affinity data. To gain quantitative insights into the effect of Glc modification on HMO binding to galectins, we produced “open-ring” and “closed-ring” conjugates of L6, L9, L16, L18, L24 and L30 (Figure S1, Supporting Information) and measured their affinities for hGal-1, hGal-3C,

15 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 29

hGal-7 and hGal-9N using ESI-MS. Inspection of the binding data (Table S4 and Figure S6, Supporting Information) reveals that the affinities for “open-ring” compounds (L6-MA, L9-MA, L16-MA, L18-MA, L24-MA and L30-MA) are, in all cases, lower than those of the corresponding free HMOs. The affinities of the “closed-ring” conjugates (L6-PA, L9-PA, L16PA, L18-PA, L24-PA and L30-PA) also tended to be lower than those of the free HMOs, although in a small number of cases, similar affinities were measured. Overall, the “closed-ring” conjugates were preferred to the “open-ring” conjugates by the galectins. Finally, it should be noted that the affinity trends of neither the “closed-ring” nor “open-ring” conjugates match the trends established for the corresponding free HMOs. Conclusions In summary, the affinities of thirty-two of the most abundant HMOs for hGal-1, hGal-3C, hGal-7 and hGal-9N were quantified using ESI-MS. It was found that the four galectin recognize the majority of the HMOs tested (hGal-1 binds thirty-two HMOs, hGal-3C binds twenty-six, hGal-7 binds thirty-one and hGal-9N binds twenty-six). Twenty-five of the HMOs tested bind all four galectins, with affinities ranging from 103 M-1 to 105 M-1. The reliability of the ESI-MS assay for quantifying the affinities of HMOs for lectins was established from the agreement found between the ESI-MS data and affinities of a small number of HMOs for hGal-1, hGal-3C and hGal-7 measured by ITC. Comparison of the relative affinities (of fourteen HMOs) measured by ESIMS with the reported HMO specificities of hGal-1, hGal-3, hGal-7 and hGal-9 (for these same HMOs) established from the HM-SGM-v2 microarray21 showed fair-to-poor correlation, with evidence of false positives and false negatives in the microarray data. ESI-MS binding measurements performed on galectins and “open-ring” and “closed-ring” conjugates of some of the HMOs revealed that modification of the Glc residue alters both the absolute and relative

16 ACS Paragon Plus Environment

Page 17 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

affinities. Consequently, the discrepancies in the HMO specificities established from the ESI-MS measurements and HM-SGM-v2 microarray data may be due, at least in part, to the chemical modifications of the HMOs required for immobilization. Taken together, the findings of this study suggest that HMO specificities of lectins established using microarrays may not accurately reflect their true HMO-binding properties and that the use of “in solution” assays, such as ESIMS and ITC, is to be preferred. Supporting Information Experimental details, affinities, HMOs structures and purities, mass spectra, ITC data, comparison of affinities and RFU values, synthesis of “open-ring” and “closed-ring” conjugates. This information is available free of charge via the Internet at http://pubs.acs.org/. Acknowledgements The authors are grateful for financial support provided by the National Sciences and Research Council of Canada and the Alberta Glycomics Centre. We also acknowledge C. Cairo (University of Alberta) and S. Sato (Laval University) for generously providing proteins used in this study.

17 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 29

References (1)

Lara-Villoslada, F.; Olivares, M.; Sierra, S.; Rodríguez, J.M.; Boza, J.; Xaus, J. Brit. J. Nutr. 2007, 98, S96-S100.

(2)

Hill, D.R.; Newburg, D. S. Nutr. Rev. 2015, 73, 463-476.

(3)

Kobata, A. Proc. Jpn. Acad. Sci., Ser. B 2010, 86, 731-747.

(4)

Wu, S.; Tao, N.; German, J.B.; Grimm, R.; Lebrilla, C.B. J. Proteom. Res. 2010, 9, 41384151.

(5)

Wu, S.; Grimm, R.; German, J.B.; Lebrilla, C.B. J. Proteom. Res. 2011, 10, 856-868.

(6)

Ruhaak, L.R.; Lebrilla, C.B. Adv. Nutr. 2012, 3, 406S-414S.

(7)

Mariño, K.; Lane, J.A.; Abrahams, J.L.; Struwe, W.B.; Harvey, D.J.; Marotta, M.; Hickey, R.M.; Rudd, P.M. Glycobiology, 2011, 21, 1317-1330.

(8)

Yu, Y.; Lasanajak, Y.; Song, X.; Hu, L.; Ramani, S.; Mickum, M.L.; Ashline, D.J.; Prasad, B.V.; Estes, M.K.; Reinhold, V.N.; Cummings, R.D.; Smith, D. F. Mol. Cell. Proteom. 2014, 13, 2944-2960.

(9)

Hong, Q.; Ruhaak, L.R.; Totten, S.M.; Smilowitz, J.T.; German, J.B.; Lebrilla, C.B. Anal. Chem. 2014, 86, 2640-2647.

(10)

Xu, G.; Davis, J.C.; Goonatilleke, E.; Smilowitz, J.T.; German, J.B.; Lebrilla, C.B. J. Nutr. 2017, 147, 117-124.

(11)

Goehring, K.C.; Kennedy, A.D.; Prieto, P.A.; Buck, R.H. PloS One 2014, 9, e101692.

(12)

Smilowitz, J.T.; Lebrilla, C.B.; Mills, D.A.; German, J.B.; Freeman, S.L. Annual Rev. Nutr. 2014, 34, 143-169.

(13)

Bode, L. J. Nutr. 2006, 136, 2127-2130.

(14)

Bode, L.; Jantscher-Krenn, E. Adv. Nutr. 2012, 3, 383S-391S.

(15)

Bode, L. Glycobiology 2012, 22, 1147-1162. 18 ACS Paragon Plus Environment

Page 19 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

(16)

Bhargava, P.; Li, C.; Stanya, K.J.; Jacobi, D.; Dai, L.; Liu, S.; Gangl, M.R.; Harn, D.A.; Lee, C.-H. Nat. Med. 2012, 18, 1665-1672.

(17)

Newburg, D. S.; Ruiz-Palacios, G.M.; Morrow, A.L. Annu. Rev. Nutr. 2005, 25, 37-58.

(18)

He, Y.; Lawlor, N.T.; Newburg, D. S. Adv. Nutr. 2016, 7, 102-111.

(19)

Yu, Y.; Mishra, S.; Song, X.; Lasanajak, Y.; Bradley, K.C.; Tappert, M.M.; Air, G.M.; Steinhauer, D.A.; Halder, S.; Cotmore, S.; Tattersall, P.; Agbandje-McKenna, M.; Cummings, R. D.; Smith, D. F. J. Biol. Chem. 2012, 287, 44784-44799.

(20)

Ashline, D.J.; Yu, Y.; Lasanajak, Y.; Song, X.; Hu, L.; Ramani, S.; Prasad, V.; Estes, M.K.; Cummings, R.D.; Smith, D.F. Mol. Cell. Proteom. 2014, 13, 2961-2974.

(21)

Noll, A.J.; Gourdine, J.P.; Yu, Y.; Lasanajak, Y.; Smith, D. F.; Cummings, R. D. Glycobiology 2016, 26, 655-669.

(22)

Noll, A.J.; Yu, Y.; Lasanajak, Y.; Duska-McEwen, G.; Buck, R.H.; Smith, D.F.; Cummings, R.D. Biochem. J. 2016, 473, 1343-1353.

(23)

Song, X.; Xia, B.; Stowell, S.R.; Lasanajak, Y.; Smith, D.F.; Cummings, R.D. Chem. Biol. 2009, 16, 36-47.

(24)

Song, X.; Heimburg-Molinaro, J.; Smith, D.F.; Cummings, R.D. Curr. Protoc. Chem. Biol. 2011, 3, 53-63.

(25)

Song, X.; Heimburg-Molinaro, J.; Smith, D.F.; Cummings, R.D. Methods Mol. Biol. 2012, 800, 163-171.

(26)

Heimburg‐Molinaro, J.; Song, X.; Smith, D.F.; Cummings, R.D. Curr. Protoc. Protein Sci. 2011, Chapter 12: Unit 12.10.

(27)

Welcome to the home for Functional Glycomics research. (2016, October 3). Retrieved, http://www.functionalglycomics.org

19 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(28)

Page 20 of 29

Grant, O.C.; Smith, H.M.; Firsova, D.; Fadda, E.; Woods, R.J. Glycobiology 2014, 24, 1725.

(29)

Kilcoyne, M.; Gerlach, J.Q.; Kane, M.; Joshi, L. Anal. Methods 2012, 4, 2721-2728.

(30)

Hirabayashi, J.; Hashidate, T.; Arata, Y.; Nishi, N.; Nakamura, T.; Hirashima, M.; Urashima, T.; Oka, T.; Futai, M.; Muller, W.E. BBA-Gen. Subjects 2002, 1572, 232-254.

(31) Ohtsuka, I.; Sadakane, Y.; Higuchi, M.; Hada, N.; Hada, J.; Kakiuchi, N.; Sakushima, A. Bioorg. Med. Chem. 2011, 19, 894-899. (32)

Arthur, C.M.; Rodrigues, L.C.; Baruffi, M.D.; Sullivan, H.C.; Heimburg-Molinaro, J.; Smith, D.F.; Cummings, R.D.; Stowell, S.R. Examining galectin binding specificity using glycan microarrays. In Galectins: Methods and Protocols; Stowell, S.R., Cummings, R.D., Eds.; Methods in Molecular Biology 1207; Humana Press: New York, 2015; pp 115-131.

(33)

Paulson, J.C.; Blixt, O.; Collins, B.E. Nat. Chem. Biol. 2006, 2, 238-248.

(34)

He, X.G.; Gerona-Navarro, G.; Jaffrey, S.R. J. Pharmacol. Exp. Ther. 2005, 313, 1-7.

(35)

El-Hawiet, A.; Kitova, E.N.; Klassen, J.S. Biochemistry 2012, 51, 4244-4253.

(36)

Kitova, E.N.; El-Hawiet, A.; Schnier, P.D.; Klassen, J.S. J. Am. Soc. Mass Spectrom. 2012, 23, 431-441.

(37)

Han, L.; Shams-Ud-Doha, K.; Kitova, E.N.; Klassen, J.S. Anal. Chem. 2016, 88, 82248231.

(38)

Han, L.; Kitov, P.I.; Kitova, E.N.; Tan, M.; Wang, L.; Xia, M.; Jiang, X.; Klassen, J.S. Glycobiology 2013, 23, 276-285.

(39)

Yao, Y.; Shams-Ud-Doha, K.; Daneshfar, R.; Kitova, E.N.; Klassen, J.S. J. Am. Soc. Mass Spectrom. 2015, 26, 98-106.

20 ACS Paragon Plus Environment

Page 21 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

(40)

El-Hawiet, A.; Kitova, E.N.; Klassen, J.S. Biochemistry 2012, 51, 4244-4253.

(41)

Ouellet, M.; Mercier, S.; Pelletier, I.; Bounou, S.; Roy, J.; Hirabayashi, J.; Sato, S.; Tremblay, M. J. J. Immunol. 2005, 174, 4120-4126.

(42)

Hirabayashi, J.; Kasai, K. J. Biol. Chem. 1991, 266, 23648-23653.

(43)

Diehl, C.; Genheden, S.; Modig, K.; Ryde, U.; Akke, M. J. Biomol. NMR 2009, 45, 157169.

(44)

Nagae, M.; Nishi, N.; Murata, T.; Usui, T.; Nakamura, T.; Wakatsuki, S.; Kato, R. Glycobiology 2009, 19, 112-117.

(45)

Labrie, M.; Vladoiu, M.; Leclerc, B.G.; Grosset, A.A.; Gaboury, L.; Stagg, J.; St-Pierre, Y. PLoS One. 2015,10, e0131307.

(46)

Cummings, R.D.; Liu, F.-T. Galectins. In Essentials of Glycobiology, 2nd edition; Varki, A., Cummings, R.D., Esko J.D., Freeze, H.H., Stanley, P., Bertozzi, C.R., Hart, G.W., Etzler, M.E., Eds.; Cold Spring Harbor Laboratory Press: New York, 2009; Chapter 33.

(47)

Lobsanov, Y.D.; Gitt, M.A.; Leffler, H.; Barondes, S.H.; Rini, J.M. J. Biol. Chem. 1993, 268, 27034-27038.

(48)

Sato, S.; St-Pierre, C.; Bhaumik, P.; Nieminen, J. Immunol. Rev. 2009, 230, 172-187.

(49)

López-Lucendo, M.F.; Solís, D.; André, S.; Hirabayashi, J.; Kasai, K.; Kaltner, H.; Gabius, H.J.; Romero, A. J. Mol. Biol. 2004, 343, 957-970.

(50)

Su, J.; Zhang, T.; Wang, P.; Liu, F.; Tai, G.; Zhou, Y. Acta Biochim. Biophys. Sin. 2015, 47, 192-198.

(51)

Leonidas, D.D.; Vatzaki, E.H.; Vorum, H.; Celis, J.E.; Madsen, P.; Acharya, K.R. Biochemistry 1998, 37, 13930-13940.

21 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(52)

Page 22 of 29

Nagae, M.; Nishi, N.; Nakamura-Tsuruta, S.; Hirabayashi, J.; Wakatsuki, S.; Kato, R. J. Mol. Biol. 2008, 375,119-135.

(53)

Diehl, C.; Engström, O.; Delaine, T.; Håkansson, M.; Genheden, S.; Modig, K.; Leffler, H.; Ryde, U.; Nilsson, U.J.; Akke, M. J. Am. Chem. Soc. 2010, 132, 14577-14589.

(54) Wang, W.; Kitova, E.N.; Klassen, J.S. Anal. Chem. 2003, 75, 4945-4955. (55)

Rademacher C.; Shoemaker G.K.; Kim H.S.; Zheng R.B.; Taha H.; Liu C.; Nacario R.C.; Schriemer D.C.; Klassen J.S.; Peters T.; Lowary T.L. J. Am. Chem. Soc., 2012, 129, 10489-10502.

(56)

http://www.uniprot.org/, UniProtKB entry P17931 (accessed 22/01/2017).

22 ACS Paragon Plus Environment

Page 23 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Figure captions Figure 1.

Structures of the HMOs L1 – L32 and the disaccharide L33. Monosaccharide key: glucose ( ), galactose ( ), N-acetylgalactosamine ( ), N-acetylglucosamine ( ), sialic acid ( ), fucose ( ).

Figure 2.

ESI mass spectra acquired in positive ion mode for a 200 mM aqueous ammonium acetate solution (pH 6.8, 25 °C) of (a) hGal-1 (3 µM); (b) hGal-3C (5 µM); (c) hGal-7 (14 µM); (d) hGal-9N (10 µM); (e) hGal-1 (3 µM), Pref (2 µM) and L9 (5 µM); (f) hGal-3C (3 µM), Pref (5 µM) and L9 (6 µM); (g) hGal-7 (7 µM), Pref (4 µM) and L9 (10 µM); (h) hGal-9N (5 µM), Pref (3 µM) and L9 (5 µM).

Figure 3.

Summary of the association constants (Ka) for HMO (L1–L32) binding to (a) hGal1, (b) hGal-3C, (c) hGal-7 and (d) hGal-9N measured by ESI-MS in 200 mM aqueous ammonium acetate solutions (pH 6.8, 25 °C).

Figure 4.

Comparison of the association constants (Ka) for HMO binding to hGal-7 ( ), hGal3C ( ) and hGal-1 ( ) at 25 °C and pH 6.8 measured by ESI-MS and ITC.

Figure 5.

Comparison of the normalized RFU values measured for binding of fourteen HMOs (L2-L4, L6, L9, L11, L13, L15, L16, L20, L22, L24, L26, L30) with (a) hGal-1 ( , 200 µg mL-1), (b) hGal-3 ( , 200 µg mL-1), (c) hGal-7 ( , 200 µg mL-1) and (d) hGal-9 (20 µg mL-1) using HM-SGM-v2 microarray21 and association constants (Ka) measured by ESI-MS ( ) for (a) hGal-1, (b) hGal-3C, (c) hGal-7 and (d) hGal9N in 200 mM aqueous ammonium acetate solutions (pH 6.8, 25 °C). Errors correspond to one standard deviation.

23 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 29

Figure 1

24 ACS Paragon Plus Environment

Page 25 of 29

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 2

25 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 29

Figure 3

26 ACS Paragon Plus Environment

Page 27 of 29

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4

27 ACS Paragon Plus Environment

Analytical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 29

Figure 5

28 ACS Paragon Plus Environment

Page 29 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

TOC graphic

29 ACS Paragon Plus Environment