In vivo Targeting of Alveolar Macrophages via RAFT-Based Glycopolymers

Abstract

Targeting cell populations via endogenous carbohydrate receptors is an appealing approach for drug delivery. However, to be effective, this strategy requires the production of high affinity carbohydrate ligands capable of engaging with specific cell-surface lectins. To develop materials that exhibit high affinity towards these receptors, we synthesized glycopolymers displaying pendant carbohydrate moieties from carbohydrate-functionalized monomer precursors via reversible addition-fragmentation chain transfer (RAFT) polymerization. These glycopolymers were fluorescently labeled and used to determine macrophage-specific targeting both in vitro and in vivo. Mannose- and N-acetylglucosamine-containing glycopolymers were shown to specifically target mouse bone marrow-derived macrophages (BMDMs) in vitro in a dose-dependent manner as compared to a galactose-containing glycopolymer (30- and 19-fold higher uptake, respectively). In addition, upon macrophage differentiation, the mannose glycopolymer exhibited enhanced uptake in M2-polarized macrophages, an anti-inflammatory macrophage phenotype prevalent in injured tissue. This carbohydrate-specific uptake was retained in vivo, as alveolar macrophages demonstrated 6-fold higher internalization of mannose glycopolymer, as compared to galactose, following intratracheal administration in mice. We have shown the successful synthesis of a class of functional RAFT glycopolymers capable of macrophage-type specific uptake both in vitro and in vivo, with significant implications for the design of future targeted drug delivery systems.

1. Introduction

The targeted delivery of small molecule drugs and biologics continues to be a major objective towards improving therapeutic efficiency through the mitigation of off-target effects and reduction in required dose [1]. However, few delivery carriers are capable of recognizing cell-specific ligands while avoiding nonspecific cellular uptake. To overcome this limitation, carbohydrate-based materials have been investigated due to their biocompatibility, target specificity, and ability to facilitate receptor-mediated uptake through cell-surface lectins (carbohydrate-binding proteins) [2,3].

Macrophages are an attractive therapeutic target because they play an important role in the inflammatory response and wound healing [4]. A number of macrophage subsets have been described that are associated with distinct phenotypes. Classically activated macrophages (M1) are critical for host defense whereas alternatively activated macrophages (M2) are important in injury resolution and wound healing. In response to acute injury, the predominant macrophage phenotype is pro-inflammatory (M1). However, an overexuberant M1 macrophage response can result in collateral tissue damage and impaired wound healing. Likewise, while M2 macrophages are important for appropriate wound healing, excess M2 response can result in tissue fibrosis. Indeed, dysregulated macrophage function is associated with a wide range of conditions including chronic ulcers, allergic asthma, atherosclerosis, autoimmune disorders, and fibrotic diseases [4]. Therefore, developing a drug delivery platform capable of directly targeting and modulating polarized macrophages has important therapeutic implications, for example, targeting of alveolar macrophage represents a promising strategy for addressing pulmonary inflammatory conditions [5,6].

Carbohydrates are known to play a significant role in the inflammatory response through mediating cell-cell recognition of immune cells, including macrophages [7]. Macrophages are a promising target for carbohydrate-based therapeutics as they express carbohydrate binding receptors which internalize bound material via receptor-mediated endocytosis [8]. One such carbohydrate binding receptor is the macrophage mannose receptor, an endocytic protein that is highly expressed on macrophages, including the alveolar macrophage [9]. The mannose receptor mediates the uptake and internalization of extracellular ligands including potentially harmful extracellular glycoproteins with terminal mannose, fucose, or N-acetylglucosamine, and pathogens with high densities of mannose on their surface [10,11]. The murine mannose receptor, MRC-1 (CD206), contains eight extracellular C-type lectin-like domains (CTLDs) [12]. Simple monosaccharides exhibit low affinities towards single MRC-1 CTLDs, with dissociation constants in the low millimolar range [13]. Many natural glycans enhance these weak binding events by clustering multiple glycosides together, thereby allowing multivalent interactions to be made with a multidomain lectin receptor (e.g. the mannose receptor) leading to a significant increase in overall avidity [14]. Ligands capable of exhibiting this multivalent behavior through simultaneous engagement of multiple CTLDs on the mannose receptor are potent facilitators of macrophage-specific uptake [15]. While mannose displays the highest affinity toward mannose receptor CTLDs, other sugars (e.g. fucose, N-acetylglucosamine and glucose) are also recognized by these lectin-like domains [16].

Through the presentation of multiple pendant carbohydrates, synthetic glycopolymers provide a promising platform to facilitate mannose receptor-mediated binding and subsequent endocytosis [2]. While such compounds have been previously synthesized to probe lectin-carbohydrate binding behavior [17], only recently have structurally well-defined, homogeneous glycopolymers capable of multivalent interactions been successfully prepared [18]. For example, multivalent glycopolymers can be synthesized via the functionalization of an individual monosaccharide with a vinyl-containing compound resulting in a glycomonomer [19]. These glycomonomers can be polymerized through free radical polymerization to yield a glycopolymer with pendant carbohydrates [20,21]. Use of reversible addition-fragmentation chain transfer (RAFT) polymerization for this glycopolymer synthesis offers precise control over the reaction, resulting in predictable molecular weights, narrow molecular weight distributions, and the ability to develop complex polymer architectures [22]. Lowe et al. first demonstrated the successful RAFT polymerization of a glycomonomer by using glucosefunctionalized methacrylate monomer in aqueous conditions; the resultant material exhibited a low polydispersity and displayed “living” properties characteristic of the RAFT technique [19]. Additionally, through the modification of the chain transfer agent (CTA) and the incorporation of comonomers, facile telechelic and pendant polymer functionalization is achievable [23-25]. By combining the versatility of the RAFT process with carbohydrate synthetic techniques, structurally complex glycosylated materials capable of mimicking the multivalent binding activity of biological carbohydrate compounds can be realized., e.g. glycopolymer micelles [25,26], stars [18], nanoparticles [27,28], “clickable” constructs [23] [29-31], and glycosylated block copolymers [26] [32-34]. By displaying multiple functional carbohydrates, these materials can be used to target specific cell populations via carbohydrate-dependent uptake mechanisms, allowing for the design of diagnostic and therapeutic glycosylated constructs.

Glycosylation of drug delivery vehicles has been explored as a means to access alveolar macrophages in vivo, a cell implicated in the pathogenesis of pulmonary conditions [6]. Chono et al. demonstrated that mannosylating liposomes enhanced their uptake by rat alveolar macrophages in vivo following intratracheal administration, work which has been extended by Wijagkanalan et al. [35-37]. In light of this previous work, there has yet to be a study systemically evaluating the uptake of well-defined, multivalent carbohydrate materials by macrophages both in vitro and in pulmonary tissue. Herein, we describe the synthesis of glycomonomers and employed RAFT polymerization to develop a family of fluorescently-labeled glycopolymers capable of macrophage-specific targeting.

2. Materials and methods

2.1. Materials

Materials were purchased from Sigma-Aldrich unless otherwise specified. 4,4′-Azobis(4-cyanovaleric acid) (V501) was obtained from Wako Chemicals USA, Inc. 4-Cyano-4-(ethylsulfanylthiocarbonyl) sulfanylvpentanoic acid (ECT) and Pyridyl disulfide ethyl methacrylamide (PDSEMA) was synthesized as described previously [38].

2.2. General synthesis of glycomonomers

TMSOTf (cat.) was added to a mixture of trichloroacetimidate sugar donor [39] (1g, 2.0 mmol) and 2-hydroxyethyl methacrylate (1.2 eq) in dichloromethane at room temperature. The reaction mixture was stirred at room temperature for 20 min and then quenched by the addition of triethylamine. The protected sugar-hydroxyethyl methacrylate was obtained after removal of solvent under reduced pressure, followed by purification via flash silica column chromatography (82-89% yield). The procedure for preparing protected GlcNAc-hydroxyethyl methacrylate is described in Supporting Information. For deprotection, sugar-hydroxyethyl methacrylate was added to 1% sodium methoxide in methanol and the mixture was stirred at 25 °C for 15 min. The reaction mixture was neutralized with glacial acetic acid. The fully deprotected glycomonomer was obtained after evaporation of solvent under reduced pressure, followed by purification via flash silica column chromatography (70-82% yield).

2.3. Synthesis of glycopolymers

The RAFT polymerization of glycopolymers was conducted in a heterogeneous solvent system of dH₂O/ethanol (3:1 vol:vol) at 70°C with 15 wt% monomer under a nitrogen atmosphere for 4h using ECT and V501 as the chain transfer agent (CTA) and radical initiator, respectively. The initial CTA to monomer molar ratio ([CTA]₀:[M]₀) was 50:1, the initial CTA to initiator molar ratio ([CTA]₀:[I]₀) was 20:1, and the initial molar feed ratios of glycomonomer to PDSEMA was 9:1. The resultant glycopolymer was isolated by dilution into dH₂O followed by lyophilization. The glycopolymer was further purified by redissolution into dH₂O, chromatographic separation with a PD-10 desalting column (GE Healthcare), and further lyophilization to obtain the final polymer.

2.4. Glycopolymer characterization

Monomer conversion and incorporation were determined by ¹H-NMR (500 MHz, D₂O). Conversion was determined to be greater than 99% due to the absence of resonances associated with vinyl protons on the glycomonomer (δ 6.07) and PDSEMA (δ 5.77) following polymerization. Copolymer composition was calculated from an aromatic PDSEMA proton (δ 8.40) and a methylene proton vicinal to the ester group (δ 4.15 Man) or methine proton on the pyranose ring (δ 4.36 Gal and δ 4.51 GlcNAc). Molecular weights (M_n) and polydispersity indices (PDI) were determined by gel permeation chromatography (GPC) using a Viscotek GPCmax VE2001 and refractometer VE3580 (Viscotek) with Tosoh TSK-GEL α-3000 (2X) and α-4000 columns connected in series (Tosoh Bioscience). HPLC-grade N,N-dimethylacetamide (DMAc; 0.03% w/v LiBr; 0.05% w/v BHT) was used as the eluent at a flow rate of 0.85 mL/min while column temperature was maintained at 50°C. Ab solute number average molecular weights were calculated from dn/dc values that were determined for each glycopolymer (pManEMA: 0.101, pGalEMA: 0.100, pGlcNAcEMA: 0.130). PDSEMA incorporation was further validated by reduction of the glycopolymers in the presence of Bond-Breaker TCEP solution (~150 molar excess per polymer at 50 mM; Thermo Scientific) followed by spectroscopic measurement of liberated pyridine-2-thione (ε₃₄₃ = 8080 M cm⁻¹).

2.5. Fluorophore labeling of glycopolymers

Glycopolymers (~10 mg mL⁻¹, in 0.1 M sodium phosphate pH 7.4 with 0.15 M NaCl buffer) were incubated in immobilized TCEP disulfide reducing gel (~10 molar excess per polymer; Thermo Scientific) for 2h followed by elution with PBS. Alexa Flour 488 (AF488) C5 maleimide (10 mg mL⁻¹ in DMSO) was added to the reduced polymer in solution resulting in an approximately equimolar amount of fluorophore to reduced PDS groups and an overall polymer concentration of ~2 mg mL⁻¹. The reaction proceeded overnight at room temperature followed by removal of excess fluorophore by a PD-10 desalting column and lyophilization. Fluorophore labeling efficiency was determined by spectroscopic measurement of the trithiocarbonate species on the glycopolymer end group (ε₃₁₀ = 20,000 M cm⁻¹) and the Alexa Fluor 488 (ε₄₉₅ = 73,000 M cm⁻¹).

2.6. Lectin agglutination assay

The ability of the glycopolymers to bind a mannose-specific lectin, Concanavalin A (ConA), was assessed by an agglutination assay. 1 μM ConA was mixed with 10 μM glycopolymer (based on number of carbohydrate repeats) and the solution turbidity was measured by UVVis spectroscopy at 350 nm at one minute intervals for 30 minutes.

2.7. In vitro glycopolymer uptake assay

Animal protocol was approved by University of Washington Institutional Animal Care and Use Committee. Primary mouse lung fibroblasts (MLF) were isolated and cultured as previously described (Choi 2009). The murine lung epithelial cell line MLE12 was obtained from American Type Culture Collection (ATCC). MLF and MLE12 uptake assays were performed in colorless DMEM supplemented with Glutamax (Life Technologies) and Nutridoma-SP (Roche). Bone marrow derived macrophages (BMDM) were isolated from femurs and tibias of 8-12 week-old C57BL6 mice as previously described [40] and cultured in RPMI-1640 containing 10% Fetal Bovine Serum and 30% L929-conditioned medium for 7-10 days. To obtain differentiated macrophages, BMDMs (M0) were treated with 50 ng mL⁻¹ E. coli 0111:B4 LPS (Sigma) for 24 hours (M1) or 20 ng mL⁻¹ each of IL-4 and IL-13 (Life Technologies) for 48 hours (M2). Cells were seeded in 12-well plates and allowed to adhere overnight. Cells were rinsed with DPBS, and then incubated with indicated concentration of Alexa488-labeled glycopolymers in colorless RPMI supplemented with Glutamax and Nutridoma-SP. For competition experiments, unlabeled glycopolymer was added 15 min prior to addition of labeled glycopolymer. At the end of incubation, the cells were washed and lifted with cold DPBS aided by cell scrapers. Internalized polymer was measured as fluorescence intensity of the cells using Guava EasyCyte Plus System (Millipore) and data analyzed using CellQuest 2.0 (BD Biosciences). All experiments were performed in triplicate and repeated at least 3 times.

2.8. In vivo glycopolymer uptake assay

8-12 week-old C57BL6 mice (Jackson Laboratories) underwent intratracheal instillation with 10 μM Alexa-488 glycopolymers or control (unlabeled) polymer in 50 μL DPBS. After 15-30 min, the lungs were lavaged with 1 mL of DPBS containing 0.6 mM EDTA. Bronchoalveolar lavage cells were centrifuged and rinsed to remove unincorporated polymers and resuspended in PBS for flow analysis by Guava as above [41]. In some cases, BAL cells were spun onto microslides, and nuclei were counterstained with DAPI. Images were obtained using a Nikon Eclipse 80i microscope with a DS Camera Head DS-5M for fluorescent microscopy.

2.9. Statistical analysis

Means of more than two groups of data were compared using one-way analysis of variance (ANOVA) for analysis of one independent variable or two way ANOVA, for analysis of two independent variables, followed by Tukey’s honestly significant difference (HSD) post hoc test. Student T-test was used for comparison of paired parametric data. For non-parametric data, Mann-Whitney’s U test was performed. All tests were two-tailed and p values ≤ 0.05 were considered significant.

3. Results and discussion

3.1. Glycopolymer synthesis and characterization

Carbohydrates are attractive ligands for imparting biological targeting functionality to polymeric systems. Carbohydrate-ligands can be easily synthesized in large scale [42], are stable to indefinite storage at ambient temperatures [43], and can leverage low-affinity binding interactions through multivalency [44]. Well-defined glycopolymers were prepared via the RAFT polymerization of synthesized glycomonomers (Scheme 1). We selected mannose (Man) and N-acetylglucosamine (GlcNAc) because of their known mannose receptor binding and galactose (Gal) due to its lack of mannose receptor interactions [16,45]. Each monosaccharide was functionalized with ethyl methacrylate (EMA), yielding the glycomonomers: ManEMA, GlcNAcEMA, and GalEMA (Supplementary Fig. S1). These glycomonomers were polymerized via the RAFT technique. A pyridyl disulfide comonomer, pyridyl disulfide ethyl methacrylamide (PDSEMA), was incorporated into the polymerization (at a 10% molar feed ratio) to provide a conjugatable handle for a maleimide-containing fluorophore. Glycopolymers (subsequently referred to as pManEMA, pGlcNAcEMA, and pGalEMA) were successfully synthesized under controlled conditions with consistent size and composition as determined by gel permeation chromatography (GPC) and ¹H-NMR spectroscopy (Table 1, Supplementary Fig. S2 and S3). The glycopolymers exhibited narrow molecular weight distributions with polydispersity indices (PDI) of 1.2 and resultant block lengths of 11400 – 13400 g mol⁻¹. The monomer incorporations of pyridyl disulfide groups per polymer were also similar between the glycopolymers (3.3 – 5.2). These findings demonstrate the first direct RAFT polymerization of the ManEMA and GlcNAcEMA glycomonomers used here while the GalEMA glycomonomer has been previously polymerized via this technique [28,46].

Scheme 1.

RAFT-mediated glycopolymer synthesis and subsequent fluorophore conjugation and proposed mechanism of MRC-1-mediated macrophage uptake of the glycopolymers.

Table 1.

Molecular weights, compositions, conversions, and labeling efficiency of glycopolymers.

Glycopolymer	Mna (g/mol)	PDIa	%Conversionb	PDS/ polymerc	Alexa488/ polymerc
pManEMA	11400	1.2	>99	5.2	0.93
pGalEMA	12200	1.2	>99	4.5	0.95
pGlcNAcEMA	13100	1.2	>99	3.3	1.2

^aAbsolute number average molecular weights and polydispersity index (PDI) as determined by gel permeation chromatography (GPC).

^bDetermined by 1H-NMR

^cDetermined by UV-Vis spectroscopy; ratio represents number of pyridyl disulfide (PDS) groups per polymer chain

A maleimide-functionalized fluorophore (Alexa Fluor 488, AF488) was conjugated to the glycopolymers through the pyridyl disulfide (PDS) groups following reduction with tris(2-carboxyethyl)phosphine (TCEP). Labeling efficiency was similar among the three glycopolymers as determined by UV-Vis spectroscopy: 0.93 – 1.2 fluorophores/polymer, and there was no measurable increase in solution turbidity suggestive of a lack of glycopolymer aggregation (Supplementary Fig. S4). Moreover, dynamic light scattering (DLS) measurements of the glycopolymers following labeling demonstrated there is no significant aggregation of the materials into larger, macromolecular assemblies as mean particle diameters of less than 8 nm were observed (approximately 4 – 7 nm; Supplementary Fig. S5). To initially determine lectin-binding activity, each glycopolymer was incubated with Concanavalin A (ConA) [47], a mannose-specific-binding lectin known to bind mannose and glucose, but not galactose or N-aceltylglucosamine [48]. pManEMA was found to agglutinate ConA, as measured by an increase in solution turbidity, showing that the material is capable of multivalent CRD-binding, a prerequisite for mannose receptor engagement (Supplementary Fig. S6). pGlcNAcEMA and pGalEMA did not induce ConA agglutination, demonstrating that the glycopolymer binding activity is carbohydrate-specific.

3.2. In vitro macrophage uptake of glycopolymers

We first examined whether the glycopolymers were differentially internalized by murine bone marrow-derived macrophages (BMDMs), a cell type known to express the mouse mannose receptor, MRC-1. BMDMs were incubated with increasing doses of AF488-labeled glycopolymers for varying lengths of time and cell uptake was assessed by flow cytometry and fluorescent microscopy (Fig. 1 and 2). We found that pManEMA and pGlcNAcEMA, but not pGalEMA, were internalized efficiently by BMDMs in a time dependent fashion. The uptake of pManEMA and pGlcNAcEMA were 30- and 19-fold higher than pGalEMA, respectively. Significant uptake occurred by 15 min and increased up to 6 hours (Figure 2 and data not shown). Mannose-binding protein, a soluble multi-domain C-type lectin in the same family as MRC-1, is known to exhibit carbohydrate-binding specificities defined by interactions with the vicinal, equatorial hydroxyl groups, C-3 and C-4, that are shared between several sugars [49]. Therefore, the ability of polymerized Man and GlcNAc to target macrophages is not surprising. BMDMs demonstrated a cytoplasmic punctate distribution of internalized pManEMA and pGlcNAcEMA, consistent with localization in the endosome/lysosome vesicles. No cell fluorescence above background was observed with pGalEMA. We also observed a dose-dependent response on uptake for both pManEMA and pGlcNAcEMA with saturation of pManEMA uptake occurring at 0.5 μM (Fig. 2); no significant dose-dependent effects were observed for pGalEMA uptake.

Fig. 1. In vitro BMDM glycopolymer uptake.

Representative histograms (left) and fluorescence microscopy images (right) of BMDMs incubated in vitro with indicated AF488-glycopolymer (1.5 μM) for 4 hr. Black line in histograms represents no polymer control. Inset shows a higher magnification image of the AF488-pManEMA treated cells.

Fig. 2. Dose dependent internalization.

BMDMs were incubated with AF488-glycopolymers for 15 min at 37°C. Data are reported as mean fluorescence intensity ± standard deviation from three independent experiments.

The effect of glycopolymer competition on uptake was assessed by incubating BMDMs with labeled glycopolymer in the presence of excess unlabeled glycopolymer. Competition with unlabeled pManEMA was more efficient than pGlcNAcEMA at attenuating uptake of labeled pManEMA and pGlcNAcEMA, suggesting that the binding affinity of MRC-1 towards pManEMA is higher than pGlcNAcEMA (Fig. 3). This finding is consistent with the mannose receptor’s stronger affinity for Man residues over GlcNAc [8]. We presume the minimal uptake of pGalEMA by BMDMs was due to nonspecific macropinocytosis, supported by the lack of competition from any of the glycopolymers.

Fig. 3. Competitive glycopolymer uptake.

BMDMs were preincubated with 15 μM unlabeled glycopolymer for 15 min followed by 1 μM AF488-glycopolymer for 30 min at 37°C. NC represents no competition. Data are rep orted as mean fluorescence intensity ± standard deviation from three independent experiments.

3.3. In vitro glycopolymer uptake by polarized macrophages

We tested whether the polarization state of macrophages altered the uptake of pManEMA. Macrophages can be polarized into a classic “pro-inflammatory” (M1) or alternative “pro-resolution” (M2) state, depending on the local environment of cytokines and other immune-stimulating compounds. Macrophages can be differentiated in vitro by LPS to the M1, “classically” activated state which results in the secretion of pro-inflammatory cytokines. Incubation with IL-4 and IL-13 leads to differentiation into M2, or “alternatively” activated macrophages, which are considered a pro-resolution and anti-inflammatory phenotype [50]. M2 macrophages secrete pro-fibrotic cytokines, such as tumor growth factor β (TGFβ) and platelet-derived growth factor (PDGF), that act on nearby fibroblasts to promote a fibroproliferative response. M2 macrophages also secrete matrix metalloproteases (MMPs) and tissue inhibitor of metalloproteinases (TIMPs) that regulate matrix remodeling. Additionally, they produce chemokines that attract other inflammatory cells (e.g. monocytes and dendritic cells) that subsequently clear apoptotic cells and debris, dampening the inflammatory response [51,52]. Due to the orthogonal roles played by the M1 and M2 macrophage phenotype, differential targeting of activated macrophages is attractive for therapeutic drug delivery applications. Human alveolar macrophages adopting a M2 polarization are believed to play an important role in the pathogenesis of pulmonary fibrosis, highlighting this phenotype as a potential clinical target [5].

BMDMs were cultured in either LPS or IL-4/IL-13 to obtain either the M1 or M2 polarization, respectively. Macrophages activated by the alternative pathway (M2) showed increased internalization of pManEMA whereas classically activated macrophages (M1) had similar internalization of pManEMA as naïve macrophages (M0) (Fig. 4). These differences were retained over a two-hour time course. At this later time point, M2 macrophages had internalized pManEMA at an approximately 3- and 5-fold higher level than M0 and M1 macrophages, respectively. M2 macrophages have higher expression of mannose receptor [53], providing further support that this receptor is facilitating glycopolymer uptake.

Fig. 4. Time course of mannose uptake in stimulated macrophages.

AF488-pManEMA (1 μM) uptake by unstimulated (M0), LPS-stimulated (M1), or IL-4/IL-13 stimulated (M2) BMDM’s (n=3).

3.4. In vitro cell-specific glycopolymer uptake

To determine cell-type specificity, we compared uptake and internalization of labeled glycopolymers by BMDMs, primary mouse lung fibroblasts (MLF) and the murine lung epithelial cell line, MLE-12. The latter two cell-types were selected as they are representative of the cell phenotypes encountered in the lung, the target site of our in vivo study. Neither MLF nor MLE-12 had significant internalization of any of the glycopolymers at any timepoint examined (Fig. 5), whereas BMDMs had significant uptake of pManEMA and pGlcNAcEMA. These results are consistent with mannose receptor-mediated uptake of the glycopolymers by macrophages.

Fig. 5. Cell specificity of glycopolymer internalization.

Internalization of AF488-glycopolymers (1.5 μM) by BMDMs, mouse lung epithelial cells (MLE-12), or primary mouse lung fibroblasts (MLF) after 2h (A) or 24h (B) incubation.

3.5. In vivo macrophage uptake of glycopolymers

To determine whether macrophage internalization of glycopolymers retained the same carbohydrate specificity in vivo, we administered Alexa 488-labeled glycopolymers intratracheally in normal mice and measured uptake by alveolar macrophages. Bronchoalveolar lavage (BAL) was performed at different timepoints following administration, and BAL cells were analyzed for uptake of the glycopolymers. We verified that greater than 90% of the BAL cells were alveolar macrophages by staining with a macrophage cell surface marker (F4/80, data not shown). We found that alveolar macrophages had similar in vivo uptake of glycopolymers as BMDM uptake in vitro: pManEMA and pGlcNAcEMA were readily internalized by alveolar macrophages as early as 30 minutes after instillation whereas pGalEMA had minimal internalization (Fig. 6). Flow cytometric analysis showed that uptake of pManEMA was up to 6-folds higher than the pGalEMA at 15 min (data not shown). Ex vivo examination of alveolar macrophage demonstrated a similar punctate distribution pattern as observed in BMDMs (Fig. 6 inset). These findings are consistent with previous work examining the uptake of mannosylated liposomes by alveolar macrophages following intratracheal administration in rats. Chono et al. demonstrated that adding mannose to liposomes resulted in a 2.2 fold increase in uptake over a 24 hour period as compared to bare liposomes [35]. Wijagkanalan et al. supported these findings by showing a significant increase in the internalization of mannosylated liposomes versus unmodified liposomes after two hours [36]. The results presented here demonstrate that these functional glycopolymers are active within the complex biological milieu found within the lung. Local delivery via intratracheal instillation also has the advantage of less systemic toxicity and off target effects, lower doses, and better drug stability. For example local delivery of therapeutics to the lung is a feasible option in intubated patients suffering from acute respiratory distress syndrome.

Fig. 6. Internalization of glycopolymers by alveolar macrophages in vivo.

Mice were given 10 μM AF488-glycopolymers intratracheally. Bronchoalveolar lavage (BAL) was performed after 30 min and BAL cells were analyzed by flow cytometry. Data are reported as mean fluorescence intensity ± standard deviation from three independent experiments. *p<0.01 compared to no polymer.

4. Conclusions

Synthetic glycopolymers provide a promising platform to leverage mannose receptor-mediated endocytosis for the intracellular delivery of therapeutic cargo into macrophage populations in the lung. In this study, structurally well-defined glycopolymers were synthesized via RAFT polymerization and shown to specifically engage macrophages both in vitro and in vivo in a carbohydrate-dependent manner. We demonstrate for the first time that macrophages activated by the alternative pathway (M2) showed increased internalization of mannose glycopolymers and, to a lesser degree, N-acetylglucosamine glycopolymers, whereas classically activated macrophages (M1) exhibit similar internalization of glycopolymers as naïve macrophages (M0). Demonstration of specific uptake of pManEMA by alveolar macrophages in vivo is promising for future therapeutic applications. While not explored in this work, the pyridyl disulfide handle on these glycopolymers could be conjugated to other maleimide- or thiol-functionalized compounds, e.g. small molecule drugs and biological macromolecules. Coupled with the versatility of RAFT-based polymer synthesis, glycopolymers are a promising strategy for the design of targeted polymeric drug delivery systems.