A Specific, Glycomimetic Langerin Ligand for Human Langerhans Cell Targeting

Langerhans cells are a subset of dendritic cells residing in the epidermis of the human skin. As such, they are key mediators of immune regulation and have emerged as prime targets for novel transcutaneous cancer vaccines. Importantly, the induction of protective T cell immunity by these vaccines requires the efficient and specific delivery of both tumor-associated antigens and adjuvants. Langerhans cells uniquely express Langerin (CD207), an endocytic C-type lectin receptor. Here, we report the discovery of a specific, glycomimetic Langerin ligand employing a heparin-inspired design strategy and structural characterization by NMR spectroscopy and molecular docking. The conjugation of this glycomimetic to liposomes enabled the specific and efficient targeting of Langerhans cells in the human skin. We further demonstrate the doxorubicin-mediated killing of a Langerin+ monocyte cell line, highlighting its therapeutic and diagnostic potential in Langerhans cell histiocytosis, caused by the abnormal proliferation of Langerin+ myeloid progenitor cells. Overall, our delivery platform provides superior versatility over antibody-based approaches and novel modalities to overcome current limitations of dendritic cell-targeted immuno- and chemotherapy.


Supporting Figures
. KI determination for sulfated GlcNAc derivatives.
The KI determination for heparin-derived GlcNAc derivatives via 19 F R2-filtered NMR revealed the impact of sulfation patterns on monosaccharide affinity. Obtained KI values are given in Table S1. The purity of 2, 15 and 16 was determined to be >95% using analytical reversed-phase HPLC (gradient: 3% to 30% acetonitrile in 0.1% TFA in H2O in 2.5 min). The assignment of the two peaks for 2 to the αand the β-anomer were validated via 1 H NMR and ESI MS. S4 Figure S3. KI determination for GlcNS analogs 1 to 5.
Competitive binding experiments served to determine the affinities for the GlcNS analog library. Obtained KI values are given in Table S2. The purity of 21 was determined to be >95% using analytical reversed-phase HPLC (gradient: 3% to 30% acetonitrile in 0.1% TFA in H2O in 2.5 min).   a. In presence of DC-SIGN, the 19 F NMR resonance of the trifluoroacetamido group of reporter ligand 24 displayed substantial line broadening. This phenomenon was utilized to develop the 19 F R2-filtered NMR reporter displacement assay following the procedure previously published for Langerin 3 . The spectrum in absence of DC-SIGN is processed with an exponential apodization function at 1.8 Hz. b. The interaction between 24 and DC-SIGN was quantified via the transversal relaxation rate R2,obs using the CPMG pulse sequence [4][5] . Representative decay curves in the presence of DC-SIGN are shown. c. The Ca 2+ -dependency of the interaction was validated via addition of EDTA. The R2,obs value in absence of DC-SIGN was determined from four independent experiments. The standard error for the experiments in presence of DC-SIGN and EDTA is derived directly from the fitting procedure. d. Relaxation dispersion experiments for 24 in presence of DC-SIGN indicate at negligible exchange contribution R2,ex at a νCPMG value of 500 Hz. In contrast to Langerin, a considerable exchange contribution is observed a lower νCPMG values. e. A representative binding isotherm for the affinity determination for 24 is shown. The KD value was determined from three independent titrations. The obtained KD and additional parameters from the fitting procedure are given in Table S3.  a to c. The mapping of CSP values on the X-ray structure of Langerin in complex with GlcNAc or Man (PDB code: 4N32 or 3P5F) validated a Ca 2+ -dependent binding mode for 2 and 16 as indicated by CSPs observed for E285 and K299 [7][8] . Additionally, CSPs were observed for N297, A300 and S302, residues also affected upon recognition of Man or 21. By contrast, Y251 and I250 displayed considerably increased CSP values compared to 21, while a relative decrease was observed for K313. This decrease was accompanied by a relative increase for the proximal T314. Notably, residues that display considerably increased CSP values can predominantly associated with F315 and N307 which were not assigned. This also hold true for W252 and W306 that displayed smaller relative increases. Accordingly, the observed CSP pattern might be induced by interactions formed between 2 or 16 and F315 rather than K313. Similar to Man and 21, CSPs were also observed in remote regions of the C-type lectin-like domain fold, particularly for K257 and G259 in the short loop region. This might indicate a modulation of the previously reported allosteric network 1 . d. A comparison of titrations with 16 and 21 revealed distinct CSP trajectories for residues associated with the carbohydrate binding site such as E285 or W252 while trajectories of residues located in remote regions of the C-type lectin-like fold such as K257 were conserved. The abrogation of the STD effect after addition of 10 mM EDTA validated the Ca 2+ -dependent binding mode 16. STD NMR spectra were recorded at saturation times tsat of 2.0 s and are magnified 12-fold. S12 Figure S12. STD NMR epitope mapping for Man analog 21.

S5
a. STD NMR experiments served to investigate the interaction of 21 with Langerin. STD NMR spectra were recorded at saturation times tsat of 0.4 s and are magnified 8-fold. b. The epitope for 21 was determined from build-up curves and suggests a solvent exposed orientation for acetylated ethylamino linker ( Figure S13).  a. A pharmacophore model was defined to guide the initial placement of 16 in the carbohydrate binding site of Langerin (PDB code: 4N32) and to constrain the orientation of the Glc scaffold during the force field-based refinement of docking poses 7 . All features displayed require an oxygen atom within the indicated spheres. b. Four out of ten generated docking poses resemble the depicted conformation of 16. The selected docking pose predicted the formation of π-π interactions between the phenyl ring and F315 as well as the formation of a hydrogen bond between the sulfonamide group and N307. The acetylated ethylamino linker displays high solvent exposure. Accordingly, this docking pose is consistent with both 15 N HSQC and STD NMR experiments. c. The depicted alternative conformation of 16 is representative for three out of ten generated docking poses. The selected docking pose predicts the formation of cation-π interaction between the phenyl ring and K313 as well as the formation of a hydrogen bond between the sulfonamide and E293. The acetylated ethylamino linker displays high solvent exposure. However, this docking pose was less consistent with the 15 N HSQC NMR results, particularly the relative decrease of CSP values for K313. The molecular docking study afforded three additional unique docking poses for 16 that were excluded due unfavorable dihedral angles for the sulfonamide linker. The receptor surface is colored according to its lipophilicity (lipophilic: red, hydrophilic: blue). The liposomes are monodisperse (d = 160±60 nm) and stable (Ζ potential = -27±6 mV) for up to seven months when stored at 4°C in PBS. d. 1 H NMR experiments with liposomes 22 revealed two states for 16 as observed for resonances corresponding to H1' and H2' of the phenyl ring. Both states display a linewidth ν0.5 smaller than 30 Hz, indicating accessibly and residual flexibility due to their presentation on extended PEG chains. The alternative state potentially corresponds to targeting ligands oriented towards the lumen of the liposomes.  a. The gating strategy to evaluate the induction of cytotoxicity and activation by liposomes 22 with LCs is exemplarily shown after incubation at a total lipid concentration [L]T of 2.7 μM for 48 h. Using epidermal cell suspension, LCs were identified as HLA-DR + -CD45 + -CD1a high cells with an additional viability gate to analyze their activation. Apoptosis was monitored by staining for active caspase 3. The activation of LCs was monitored via expression of CD80 and CD83. b. Additionally, the activation of LCs was analyzed by ELISA to monitor TNF-α secretion in one independent experiment. In contrast to the addition of poly I:C and LPS, liposomes 22 did not activate LCs.   11 . Stereoselectivity at the anomeric position of the mannose and the glucosamine scaffold was analyzed by measuring 1 J C1,H1 coupling constants 12 .

Supporting Tables
NMR spectra were processed with MestReNova 13 . ESI-MS analysis was conducted using an 1100 Series LC/MS coupled to a G1946D ESI-Q spectrometer (Agilent). HR ESI-MS analysis was conducted using an Acquity H-Class UPLC/MS coupled to a Xevo G2-S ESI-Q-TOF spectrometer (Waters). Preparative HPLC was performed on a 1100 Series LC/MS (Thermo Scientific) using a preparative Nuleodur C18 column (Machery Nagel). Analytical HPLC was performed on an Aqcuity UPLC system using an analytical BEH C18 column (Waters).
The ratio of αand β-anomer was determined to be 10:1 via 1 H NMR. Here, only chemical shifts corresponding to the β-anomer are documented. As a result of the low signal intensities for the α- anomer the corresponding resonances were not assigned.

2-deoxy-2-3'-(methyl)phenylsulfonylamido-D-glucopyranose
Glucosamine hydrochloride (49 mg, 227 µmol) was dissolved in DMSO (4.6 ml) via addition of DIPEA (240 µl, 1.39 mmol). Next, 3-methylphenylsulfonyl chloride (84 µl, 580 µmol) was slowly added and the reaction mixture was stirred for 2 h at room temperature. The reaction was quenched with MeOH and solvents were removed in vacuo. Residual DMSO was removed via lyophilization. The ratio of αand β-anomer was determined to be 10:1 via 1 H NMR. Here, only chemical shifts corresponding to the β-anomer are documented. As a result of the low signal intensities for the α- anomer the corresponding resonances were not assigned. The ratio of αand β-anomer was determined to be 4:1 via 1 H NMR. Here, only chemical shifts corresponding to the β-anomer are documented. As a result of the low signal intensities for the α-

2-Acetamidoethyl-α-D-mannopyrannoside
Intermediate 20 was prepared as previously published 10    S40  S42   For each spectrum 2048 scans were recorded. The relaxation delay d 1 was set to 2 s and the saturation time t sat was set to 2.00 s.

2-Acetamidoethyl-2-deoxy-2-N-tosyl-β-D-glucospyrannoside (16)
Epitope mapping. The binding epitope for 16 was determined at a concentration of 500 µM. For each spectrum 512 scans were recorded. The relaxation delay d1 was set to 6 s and spectra were recorded at 5 different saturation time tsat varying from 0.25 to 6.00 s. Equation 4 served to derive the STD effect STD for each analyzed resonance from the corresponding on-and off-resonance spectra 28 .
I0 represents the integral of a resonance in the off-resonance spectrum and Isat represents the integral of a resonance in the on-resonance spectrum. The apparent saturation rate ksat and the maximal STD effect STDmax were derived from Equation 5 in a two parameter fit 29 . Standard errors were derived directly from the fitting procedures. These parameters were used to calculate the initial slope of the STD build-up curves STD'0 via Equation 6.
STD'0 values were normalized and mapped on the corresponding ligand structure. Only resonances for which at least part of a multiplet was isolated were considered for the epitope mapping.

Liposomal Formulation
PEGylated liposomes were prepared via thin film hydration and subsequent pore extrusion as previously published 34  Doxorubicin was encapsulated into liposomes as previously described 34 . Briefly, the thin film was

Lipid-ELLA
The Lipid-ELLA was conducted as previously published 35

Cytokine ELISA
To assess the immunogenicity of targeted and non-targeted liposomes, a human TNF-α ELISA MAX TM (BioLegend) was performed according to the manufacturer´s instructions. Briefly, 20,000 enriched CD1a + epidermal cells were co-cultured with liposomes at a total lipid concentration [Lipid] T of 16 µM lipid concentration in a final volume of 100 μl culture medium for 15 h at 37°C and 5% S55 CO 2 . The positive control was co-cultured with 20 µg/ml poly I:C (Sigma Aldrich) and 100 ng/ml LPS (Invivogen), whereas the negative control remained untreated. After centrifugation at 400 g for 5 min the supernatant was analyzed in the TNF-α ELISA.

Colorimetric Cytotoxicity Assay
To assess the cytotoxicity of doxorubicin-containing liposomes with THP-1 cells, 50,000 cells per well were added to 96 well microtiter plates (Corning) and incubated in presence of liposomes at a given total lipid concentration [Lipid] T in a final volume of 100 μl for 48 h at 37°C and 5% CO 2 . Cell viability was analyzed using CellTiter 96 (Promega) by measuring formazan absorbance at 490 nm.
Complete killing was defined as the signal measured after cell lysis using 0.5 Triton X-100 and Triton-X-100 and 100 mM glycine for 10 min at room temperature. After washing two times and subsequent blocking with PBS with 3% BSA for 30 min at room temperature, cells were incubated with primary antibodies for EEA1 (rabbit, clone C45B10) and Rab5 (rabbit, clone C8B1) (Cell Signaling Technology) in PBS with 3% BSA for 1h at room temperature. After washing three times for 5 min with PBS, cells were incubated with fluorophore-labeled secondary antibody (Invitrogen) for 30 min at room temperature followed by three additional washing steps. DAPI (1:1000 dilution, Thermo Fisher Scientific) in PBS was incubated for 5 min at room temperature, following by washing with PBS for 5 min at room temperature. Coverslips were mounted using Roti-Mount (Roth). For Rab9 visualization, cells were transfected 1 day after seeding with 500 ng YFP-Rab9 per well using Lipofectamine 2000 (Invitrogen) according to the manufacturer's protocol. After 1 day, cells were incubated with liposomes and fixed as described above. After fixation, cells were washed two times