Sequence-Defined Heteromultivalent Precision Glycomacromolecules Bearing Sulfonated/Sulfated Nonglycosidic Moieties Preferentially Bind Galectin-3 and Delay Wound Healing of a Galectin-3 Positive Tumor Cell Line in an In Vitro Wound Scratch Assay

Within this work, a new class of sequence-defined heteromultivalent glycomacromolecules bearing lactose residues and nonglycosidic motifs for probing glycoconjugate recognition in carbohydrate recognition domain (CRD) of galectin-3 is presented. Galectins, a family of β-galactoside-binding proteins, are known to play crucial roles in different signaling pathways involved in tumor biology. Thus, research has focused on the design and synthesis of galectin-targeting ligands for use as diagnostic markers or potential therapeutics. Heteromultivalent precision glycomacromolecules have the potential to serve as ligands for galectins. In this work, multivalency and the introduction of nonglycosidic motifs bearing either neutral, amine, or sulfonated/sulfated groups are used to better understand binding in the galectin-3 CRD. Enzyme-linked immunosorbent assays and surface plasmon resonance studies are performed, revealing a positive impact of the sulfonated/sulfated nonglycosidic motifs on galectin-3 binding but not on galectin-1 binding. Selected compounds are then tested with galectin-3 positive MCF 7 breast cancer cells using an in vitro would scratch assay. Preliminary results demonstrate a differential biological effect on MCF 7 cells with high galectin-3 expression in comparison to an HEK 293 control with low galectin-3 expression, indicating the potential for sulfonated/sulfated heteromultivalent glycomacromolecules to serve as preferential ligands for galectin-3 targeting.


Introduction
Many processes in tumorigenesis are the result of dysregulated protein expression and the presentation of abnormal glycan motifs on the cell surface. [1][2][3] One family of proteins known to be involved in tumor biology is the galectins. Galectins consist of a conserved carbohydrate recognition domain that is known to bind β-galactoside terminating glycans such as those terminating in lactose (Lac) or poly N-acetyl-lactosamine (LacNAc).

Solid Phase Synthesis
General protocols for the solid phase synthesis were described for batch sizes of 0.1 mmol resin. All reactions were performed at room temperature in a polypropylene syringe reactor with a frit on a shaker.

Resin Preparation and Fmoc
Cleavage-0.1 mmol resin (800 mg, resin loading 0.25 mmol g −1 ,) was transferred into a 10 mL reactor and 5 mL DCM was added to swell the resin for 1 h. After washing the resin ten times with 5 mL DMF, Fmoc was cleaved by adding 5 mL of 25% piperidine in DMF three times for 10 min. In between the deprotection steps, the resin was washed three times with 5 mL DMF, and after the last deprotection, the resin was washed ten times with 5 mL DMF.

Building Block and Amino
Acid Coupling-0.5 mmol building block (5 eq) and 260 mg PyBOP (0.5 mmol, 5 eq) were dissolved in 3 mL DMF, and 0.2 mL DIPEA (1 mmol, 10 eq) was added. After flushing the solution with nitrogen for 1 min, the solution was added to the resin and the reaction was shaken for 1-1.5 h. After that, the liquid content was discarded, and the resin was washed ten times with 5 mL DMF.

Terminal-NH 2 Capping-
The resin was treated with 3 mL acetic anhydride two times for 15 min. In between, the resin was washed with DMF. After the last capping step, the resin was washed five times with 5 mL MeOH and five times with 5 mL DMF. [71] -For conditioning, the resin was washed ten times with 5 mL of tetrahydrofuran (THF)/H 2 O (1/1). For deprotection, the resin was treated two times for 1 h with 5 mL 0.2 M LiOH in THF/H 2 O (1/1). In between, the resin was washed three times with 5 mL THF/H 2 O (1/1). After deprotection, the resin was washed alternately five times with each 5 mL of H 2 O, DMF, and DCM.

Carbohydrate
Conjugation-CuAAC-Azido carbohydrate derivatives (3 eq/ alkyne group) were dissolved in 2 mL DMF. Separately, CuSO 4 (50 mol%/alkyne) and sodium ascorbate (50 mol%/alkyne) were each dissolved in 0.2 mL MilliQ water. The carbohydrate solution was first added to the resin, followed by sodium ascorbate and CuSO 4 . After shaking the reaction mixture overnight, the resin was washed sequentially with 5 mL of DMF, a solution of 0.2 M sodium diethyldithiocarbamate in DMF and water (1/1), and water, DMF, and DCM until no more color changes were observed after the treatment with the diethyldithiocarbamate solution.
washing three times with 10 mL of the 1.6 M acetic acid solution, followed by three times with 10 mL of the 0.16 M acetic acid solution. The sample was dissolved in 10 mL water and the solution was loaded to the resin into the syringe. The syringe was shaken for 1 h. The supernatant was recovered, and the resin was washed three times with 2 mL water. The combined water phases were loaded onto new, freshly activated resin and shaken for 1 h. The supernatant was collected, and the resin was washed three times with 2 mL water. The combined liquid phases were lyophilized to obtain the crude product as a white solid.

2.2.11
. Additional Note 2-For the amine and fluorescein isothiocyanate (FITC) derivatives, the Fmoc group of the last EDS building block remained until the end of the solid phase assembly. Since the basic deprotection conditions for the MDS sidechain could result in loss of Fmoc groups, Boc-protected β-alanine was used as final building block for structures 6a,c and 12a,c. FITC conjugation was performed in solution on purified glycomacromolecules as reported previously, [74] and the corresponding FITC conjugates were repurified by preparative chromatography.

2.2.12.
Additional Note 3-Glycomacromolecules were used as isolated after precipitation, TFA removal, and preparative purification. All samples have high purities (see RP-HPLC, Supporting Information), however, they contain small amounts of deletion sequences that are individually assigned and quantified according to HPLC spectra (see the Supporting Information). The ESI-MS-spectrum of the main peak is given, but the analysis of each individual peak is not further shown. 1 1 1 1 1 1 1 1 1

SPR-Inhibition Studies
The SPR-inhibition studies were performed on a lactose-functionalized CM5 senor chip on a Biacore X100 from GE Healthcare Life Science. The lactose-functionalized CM5 chip was prepared using the "surface preparation wizard" for the sensor chip CM5. The evaluation was performed using the evaluation software provided by GE Healthcare. The response unit of the Gal-3 binding event without and after incubation with the ligands was taken 155 s after start of the sample injection. The response unit (RU) for only Gal-3 represents the 100% binding and 0% inhibition event. The value of the inhibition with the glycomacromolecules was referred to the response unit of only Gal-3. All measurements were performed in triplicates.

Cell Lines and Tissue Culturing
Tissue culture was performed in a SterilGARD III Advance SG 603 laminar flow hood from Baker Company. Cell cultures were observed using a Zeiss Axiovert 25 inverted microscope. All cell lines and culture media were purchased from ATCC. Cell line HEK-293 (#CRL-1573) was grown in Eagle's minimum essential medium (EMEM) (# 30-2003) and supplemented with 10% fetal bovine serum (FBS) and 1% Pen/Strep. Cell line MCF7 (ATCC HTB-22) was cultured in EMEM with 0.1 mg mL −1 insulin, 10% FBS, and 1% Pen/Strep. Cells were cultured in 75 cm 2 tissue culture flask from TPP at 37 °C and 5% CO 2 in a CO 2 water jacketed incubator 3110 from Scientific Inc. Once weekly, the medium was changed and the trypsinization of confluent cells was performed with trypsin-ethylenediaminetet-raacetic acid solution from ATCC (#30-2101) for subculturing as recommended by the supplier. Cell counting was performed with a 2%-trypan-blue solution in PBS from VWR and dispos-able hemacytometers from Incyto C-chip.

Flow Cytometry
2.6.1. Galectin-Antibody Analysis with Flow Cytometry-Human galectin-1 biotinylated antibody, human galectin-3 biotinylated antibody, and goat IgG biotinylated antibody as isotype control were purchased from R&D systems (#BAF1152, BAF1154, and BAF108). For surface staining, cells were suspended at 5 × 10 6 cells mL −1 in Dulbecco's phosphate-buffered saline buffer containing 0.1% w/w bovine serum albumin (BSA) and 0.1% w/w sodium azide. 100 μL of the cell suspension (500 000 cells) were incubated with 3 μL of human BD Fc block (#564220) from BD biosciences for 10 min at room temperature. Without washing in between, either 3 μL of the biotinylated human galectin antibodies or the isotype control (each 0.5 μg μL −1 stock solution) was added and the cells were incubated for an additional 1 h on ice. After that, the cells were washed three times with cooled PBS+ buffer (containing BSA and sodium azide) by centrifugation at 780 rpm and 4 °C for 5 min, followed by resuspension of the pellet in PBS. After the last centrifugation step, the cell pellet was resuspended in 100 μL of PBS+ and the cells were stained with 10 μL of a 0.002 μg mL −1 solution of Streptavidin-PE conjugate for 20 min at room temperature. The cells were washed three time with on ice cooled PBS+, followed by the resuspension of the cell pellet in 300 μL of PBS+ buffer for the flow cytometry measurements using the Accuri C6 flow cytometer. For intracellular staining, the cells were fixed before staining using a Fixation/Permeabilization Solution Kit from BD Bioscience. Therefore, a cell pellet of 500 000 cells was suspended in 1 mL Fix/Perm. Solution provided by the kit for 20 min on ice. Cells were washed three times with 1 mL BD Perm/Wash Buffer. After that, the staining was performed as described for the surface staining using permeabilization buffer for the washing steps in between to ensure permeability. A total of 100 000 cells were counted with a medium flow rate. Evaluation of the FACS result was performed with the FCS Express 4 program.

Studies of FITC-Conjugated Glycooligomers using Flow Cytometry-
The entire procedure was performed while avoiding light exposure to the samples. 500 000 cells in 90 μL were seeded into 96-well plates. 10 μL of FITC-conjugated derivatives of the glycoconjugates, prepared as 2000 × 10 −6 and 1000 × 10 −6 M stock solutions in water, were added to the cells resulting in a total volume of 100 μL containing 200 × 10 −6 and 100 × 10 −6 M of the ligands, respectively. After incubation of the cells for 3 h at 37 °C and 5% CO 2 , the content of each well was transferred to a centrifuge tube and the well was washed one time with 200 μL PBS, which was afterward transferred to the same corresponding centrifuge tube. After washing the cells three times with 1 mL PBS buffer, the cells were fixed with 100 μL of a Cytofix solution for 20 min on ice. Fixed cells were washed three times with 1 mL PBS+ before measuring with an Accuri 6 flow cytometer. 20 000 cells were collected with a slow flow rate of 14 μL min −1 . In between samples, two washing steps were performed at a high flow rate of 66 μL min −1 for 1 min, the first of which involved a bleach solution containing 4% of hypochlolrite, followed by MilliQ water. A slow flow rate and the washing steps were needed to avoid clogging the system.

Fluorescence Microscopy
Cells were grown on 24 mm cover slips (# 229174) in 6-well plates (#229106) purchased from Celltreat. Cover slips were first coated with a 0.01% poly-L-lysine solution from Sigma Aldrich for 5 min, followed by three washing steps with sterile MilliQ water. Cover slips were transferred into 6-well plates and allowed to dry at room temperature for 2 h. The cell lines were trypsinized, counted, centrifuged, and resuspended to a final concentration of 70 000 cells mL −1 in the corresponding total growth medium. Next, 3 mL of the cell suspension was transferred to each well. The cells were grown on the cover slips for 2 days at 37 °C at 5% CO 2 . The staining procedure was performed in petri dishes (  x stock solution in 1 mL 70% ethanol right before usage. The cover slips were coated with the dilution for 30 s. After that, the cover slip was washed carefully three times with PBS buffer, freed from an excess of liquid by dapping on dust-free tissue and flipped onto glass slides for microscopy, using one drop of ProLong Gold as antifade mountant (#P10144). Fluorescence microscopy was performed with an Olympus DP80 coupled with Prior Scientific Launches L200S fluorescence illumination system. for 48 h at 37 °C and 5% CO 2 , the plates were centrifuged at 1000 rpm, 4 °C for 5 min. 60 μL of the supernatant was carefully removed, followed by the addition of 50 μL PBS and 50 μL MTT reagent. After incubation for 3 h at 37 °C and 5% CO 2 , the plates were centrifuged and 75 μL of the supernatant was removed. The purple precipita-tion was then dissolved by adding 200 μL of dimethyl sulfoxide to each well and shaking the plate for 3 h at room temperature covered with aluminum foil. The plates were read out with a Synergy H1 microplate reader from Biotek at 590 nm without a lid.

Scratch Wound In Vitro Test
Cells were seeded at 400 000 cells per well in 12-well plates with a final volume of 1 mL cell culture medium. Cells were grown for 48 h until a dense monolayer was reached. A wound field was created with a 10 μL pipet tip using a line on the back of the plate as guide.
After creation of the wound field, the medium was removed and replaced by 360 μL total growth medium and

Results and Discussion
In this study, we endeavored to better understand the role of nonglycosidic motifs in binding galectin-3. Therefore, we synthesized a series of glycomacromolecules of similar size and molecular weight, bearing either lactose alone (homomultivalent) or lactose alternating with several nonglycosidic motifs (heteromultivalent). Nonglycosidic motifs were designed to present aromatic residues with either neutral, amine, or sulfonated/sulfated groups to explore the role of these groups on galectin-3 binding. [75] SPPoS was used to synthesize lactose-based glycomacro-molecules 1-8 as potential ligands for galectin-3, and glucose derivatives 9-13 as nonbinding controls (Scheme 1). SPPoS uses tailor-made building blocks for the stepwise assembly of monodisperse, sequence-defined oligo(amidoamines) on solid support by applying standard Fmoc-peptide coupling protocols. The building blocks used for this study include: TDS [69] for introducing an alkyne moiety in the side chain that can be used for site-selective conjugation of azido-functionalized carbohydrates via copper-catalyzed azide-alkyne cycloaddition (CuAAC), MDS [71] for introducing a carboxylic group in the side chain for conjugation via amide coupling, and EDS [70] for introducing an ethylene glycol motif in the main chain. Homomultivalent glycomacromolecules 1-3, and 9 were synthesized according to previously established methods using TDS and EDS (see the Supporting Information), and vary in the number of glyco-sidic residues. [ 69,70,76] With the exception of 1 and 3, which were designed to represent mono-and higher valent analogs, respectively, all heteromultivalent glycomacromolecules carry three glycosidic residues and two nonglycosidic moieties.
Heteromultivalent glycomacromolecules were synthesized by replacing the EDS building blocks of compound 2 with MDS. [71] The carboxylic side chain of the MDS building block and the alkyne moiety of the TDS building block enabled orthogonal post-modification of the scaffolds via amine coupling and CuAAC, respectively. As nonglycosidic moieties, aryl residues bearing different amine or sulfonic acid and sulfonate function-alities were used. [77] For Lac(1,3,5)-Ph(1-SO 3 H,4-OH)(2,4)-6,7, attachment of the nonglycosidic sidechain after carbohydrate conjugation as described above was unsuccessful, potentially due to steric hindrance of the lactose residues. Therefore, the synthetic route was altered to reverse the amide and CuAAC coupling steps. This strategy resulted in the desired product after deprotection, cleavage, and purification (see Figures S57-S60, Supporting Information). Furthermore, shorter sulfated heterostructures 8 and 13 were synthesized using Fmoc-L-Tyr(4-SO 3 H)-OH instead of the MDS building block, by applying standard protocols to give the desired products after deprotection, cleavage, and purification.
The final glycomacromolecules were deprotected, cleaved from the resin, purified by ionexchange chromatography, and preparative RP-HPLC, and isolated with high purities (as determined by RP-HPLC analysis) (see Figures S2-S110, Supporting Information). Notably, glycomacromolecules were synthesized with different end-functionalities: free (subgroup a) and capped amine (subgroup b) or FITC-conjugated derivatives (subgroup c) for different studies (Scheme 1).
To investigate the binding avidities of the heteromultivalent glycomacromolecules as ligands for galectin-3, competitive-inhibition binding studies were performed. These studies test the ability of the glycomacromolecules to competitively inhibit the binding of galectin-3 to the glycoprotein asialofetuin in an ELISA-type assay. We chose to compare the results from our galectin-3 studies to studies with galectin-1 to determine preferential binding between these two lectins, which have a conserved carbohydrate recognition domain (CRD), but are members of different galectin classes. The ELISA-type assay performed gives the half-maximum inhibitory concentration (IC 50 ) of each ligand as a measure of avidity toward galectin-3 or galectin-1, where lower IC 50 values correspond to higher avidity ( Table 1).
As expected, for studies with galectin-3, homomultivalent structures show increased binding with increasing valency, while studies with galectin-1 revealed only small changes in binding with increasing valency. This is expected since galectin-3 forms pentameric oligomers and galectin-1 is known to form dimers. [78] However, when comparing trivalent homomultivalent glycomacromolecule 2b lacking aromatic residues with trivalent heteromultivalent analogs 4b-8b bearing aromatic residues, binding to galectin-3 increased, while binding to galectin-1 decreased.
For better comparison, IC 50 values were normalized to the IC 50 value of free lactose giving the relative inhibitory potential (Table 1: RIP) with respect to free lactose. Heteromultivalent glycomacromolecules containing aromatic residues and three lactose units (4b-8a/b) revealed a trend of increased binding to galectin-3 by a factor of 1.5-to 3-fold, and decreased binding with galectin-1 by a factor of 1.5-to 2-fold in comparison to 2b. Notably, more significant increases in binding were observed for glycomacromolecules bearing sulfonated (6b and 7b) and sulfated (8b) aromatic motifs in binding studies with galectin-3 in comparison to galectin-1. In addition, no statistically significant differences for different end groups were noted demonstrating that both capped and free amines resulted in similar outcomes for the compounds tested. One explanation for the increased binding results observed with galectin-3 is that hydrophobic interactions may be forming between the Galectin-3 CRD and the aromatic motifs of the glycomacromolecules. Preliminary support for this hypothesis is provided by previous studies focusing on the role of aromatic residues in galectin-3 binding [79][80][81] However, future studies involving 15N-1 H HSQC (heteronuclear single quantum coherence) experiments would need to be performed to confirm specific binding interactions within the binding pocket.
Similar trends were observed by applying the aforementioned glycomacromolecules in SPR bind assays. [82] For SPR experiments, a fixed concentration of galectin-3 and ligand was used to determine the ability of each ligand to reduce galectin-binding to a lactosefunctionalized CM5 chip. A statistically significant reduction in binding was observed between sul-fonated and sulfated derivatives (6b-8b) in comparison to 2b indicating that these glycomolecules had a greater inhibitory effect than compounds 4b and 5b ( Figure 1). This finding is in accordance with studies investigating the influence of negatively charged glycans on galectin-3 binding which showed that higher affinities could be obtained in comparison to uncharged glycans. [29,31 , 32] No statistical significance was observed between the binding of 5b (the best nonsulfonated/sufated binder) and 6b/8b. As expected, when replacing the binding carbohydrate ligands (lactose) with a nonbinding carbohydrate ligand (glucose, 9b-13b), we see no inhibition in both ELISA and SPR studies (see Figures S111-S117, Supporting Information) confirming that the lactose is required for binding to galectin-3 and that only the specific combination of lactose and sulfonated or sulfated nonglycosidic motifs leads to higher avidity ligands.
To further investigate the effect of glycomacromolecules as ligands of galectin-3, we performed in vitro cell studies with a human breast cancer cell line MCF 7 which is known to overexpress galectin-3. [83] Immunostaining of untreated live cells confirmed that MCF 7 cells exhibit surface expression of galectin-3 and no expression of galectin-1, while internal staining of fixed, permeabilized MCF 7 cells revealed increased staining, demonstrating that a substantial intracellular reserve of galectin-3 is present in these cells ( Figure S118, Supporting Information).
As a prerequisite for further testing, cell toxicity of glycomacromolecules selected for cell experiments (1-3, 6, 8,9,12-14) was determined via MTT cell viability assay. Results demonstrated no detectable differences in viability between the vehicle control and cells treated with the glycomacromolecules after 48 h incubation ( Figure S119/120, Supporting Information). To test the general ability of glycomacromolecules to associate with the cells, flow cytometry studies were performed using FITC-conjugated derivatives 1c-3c, 6c, 8c, 9c, 12c, 13c at two different concentrations. After fixation of the cells, analysis of stained cells via flow cytometry showed a dose-dependent staining for both cell lines for all compounds ( Figure S121/122, Supporting Information). The lack of a significant difference in the mean-FITC values suggests a nonse-lective association (Figure 2A).
Preliminary fluorescence microscopy studies were then performed to analyze localization of the glycomacromolecules. Exemplary comparison of compounds 3c (3b as best homomultivalent binder) and 8c (8b as best heteromultivalent binder) shows a general staining ( Figure  2). However, for compound 8c, there appears to be an enrichment of fluorescence around the nucleus in comparison to the cytosol. A similar pattern was observed for cells stained with 6c, 12c, and 13c (Figures S123-S127, Supporting Information). This seems to be related to the presence of the aryl sulfonated/sulfated motifs. One possible explanation for this finding is that the negative charge of these motifs might lead to an attractive interaction with positively charged nucleoporins. [84] However, at this point, it is not possible to determine the exact nature of the glycomacromolecule-cellular association.
Since the tested glycomacromolecules showed no cytotoxic behavior and positive association with the cells, we decided to study their influence on wound closure, which is known to be mediated by galectin-3. [85][86][87] This was accomplished by performing an in vitro scratch wound assay as described by Dion and co-workers [88,89] using compounds 1a-3a, 6a, 8a-9a, and 12-13a. These compounds represent the direct precursor of the FITC derivatives. Cells were cultivated as a monolayer and a "wound field" was created ( Figure  3A). Cells were then incubated with the glycomacromolecules, and the width of the wound field was observed under an inverted microscope. The distance analysis at different time points was used to create a wound closure curve ( Figure 3B). In this model, ligand binding to galectin-3 is expected to result in a reduction in wound closure over time.
In this study, different effects were observed for wound closure for the different glycomacromolecules (Figures S128-S137 and Tables S1-S3, Supporting Information). Generally, glucose-functionalized glycomacromolecules resulted in a slightly faster wound closure [90] In comparison, the lactose derivatives led to a delayed wound closure, especially sulfonic acid derivative 6a and sulfated derivative 8a. For example, after 48 h compound 8a differed with 57 ± 5% an almost 20% difference from the corresponding glucose derivative 13a showing 76 ± 5% closure (Table S2, Supporting Information). These results are in agreement with similar studies on other ligand systems targeting lectins involved in cell migration. [77,[86][87][88][89] For example, Dion and coworkers observed a delay in wound healing of around 20% for the treatment of keratinocytes with lactosamine-based (2-naphthyl)methyl compounds inspired by TD139, which is currently in clinical trials. [57,88,89] Raman and co-workers achieved a delay of 20-30% through the incubation of MCF 7 cells with a nucleoside analog addressing RNA helicase. [91] To examine the effects of sustained exposure of the glyco-macromolecules on the MCF 7 cells, we performed a "dosing" study by introducing additional aliquots of glycomacromolecules 6a, 8a, 12a, and 13a after 12, 24, 36, and 48 h giving in total an additional 100 mol% (Figure 3b). In this experiment, the differences in wound closure were even more significant, yielding delays of 53 ± 7% for 8a compared to 80 ± 4% for 13a after 48 h (Figures S131, S135, and Table S3, Supporting Information) indicating a dose-specific response, further demonstrating that glycomacromolecule ligands with higher inhibition potentials in the ELISA and SPR studies delayed wound closing in a galectin-3 cell line. Negative controls replacing lactose by glucose side chains showed no effects in similar studies, confirming that it is the combination of carbohydrate and nonglycosidic motifs that enables high avidity and selective binding.
While our preliminary results demonstrate the potential for the aforementioned glycomacromolecules to serve as preferential ligands for galectin-3, additional studies including the use of additional cell lines as well MCF 7 cells transfected with galectin-3 siRNA are necessary to further confirm the involvement of galectin-3 in delaying wound closure as observed in this study.

Conclusion
In conclusion, we introduced the synthesis of heteromulti-valent lactose-functionalized glycomacromolecules bearing nonglycosidic motifs to investigate the impact of these motifs on galectin-3 binding. The aforementioned structures were successfully tested for binding to galectin-3 using ELISA and SPR. Results revealed that the incorporation of nonglycosidic motifs, especially sulfonated or sulfated motifs, could be used to achieve RIPs that were similar to more highly glycosylated structures demonstrating that binding could be modulated by nonglycosidic components. An in vitro wound scratch assay using a galectin-3 positive MCF 7 breast-cancer cell line further revealed that structures containing sulfonated and sulfated nonglycosidic motifs delayed wound closing by almost 20%. This work demonstrates the potential for heteromultivalent glycomacromolecules bearing nonglycosidic sulfonated or sulfated motifs to serve as preferential ligands for galectin-3 in comparison to analogs without these motifs. Future studies will involve the investigating heteromultivalent glycomacromolecules-lectin binding with 15N-1 H HSQC to better understand the nature of the binding events investigated here as well as with other members of the galectin family. In addition, we plan to investigate the biological effects of the glycomacromolecules-bearing sulfonated/sulfated motifs in in vitro scratch wound assays with additional cell lines expressing galectin-3.

Supplementary Material
Refer to Web version on PubMed Central for supplementary material. Overview of synthesized structures. The terminal number for each glycomacromolecule represents the overall number of combined building blocks (EDS, TDS, and MDS). Results of the inhibition-competition binding studies of glycomacromolecules 1b-8b to Gal-1 and Gal-3 in the ELISA-type assay.