Nanoparticles for Directed Immunomodulation: Mannose-Functionalized Glycodendrimers Induce Interleukin-8 in Myeloid Cell Lines

New therapeutic strategies for personalized medicine need to involve innovative pharmaceutical tools, for example, modular nanoparticles designed for direct immunomodulatory properties. We synthesized mannose-functionalized poly(propyleneimine) glycodendrimers with a novel architecture, where freely accessible mannose moieties are presented on poly(ethylene glycol)-based linkers embedded within an open-shell maltose coating. This design enhanced glycodendrimer bioactivity and led to complex functional effects in myeloid cells, with specific induction of interleukin-8 expression by mannose glycodendrimers detected in HL-60 and THP-1 cells. We concentrated on explaining the molecular mechanism of this phenomenon, which turned out to be different in both investigated cell lines: in HL-60 cells, transcriptional activation via AP-1 binding to the promoter predominated, while in THP-1 cells (which initially expressed less IL-8), induction was mediated mainly by mRNA stabilization. The success of directed immunomodulation, with synthetic design guided by assumptions about mannose-modified dendrimers as exogenous regulators of pro-inflammatory chemokine levels, opens new possibilities for designing bioactive nanoparticles.


Synthesis of Man-N 3
Man-N 3 was synthesized followed a synthetic protocol published 2 . Man-N 3 was characterized by 1

Synthesis of PPI-PEG
Anhydrous DMSO (40 mL) was dried and degassed for around 1 h under high vacuum. In a first reaction flask PPI dendrimer G4 was dissolved in 2/3 (of 40 mL) of anhydrous DMSO and then triethylamine (Et 3 N) was added to the reaction mixture under argon protection atmosphere. In a second reaction flask alpha-tert-butyloxycarbonylamino-omega-diglycolic acid octa(ethylene glycol) and BOP were dissolved in 1/3 (of 40 ml) of anhydrous DMSO. Then, the resulting reaction mixture was stirred for 1 h under argon atmosphere. PPI dendrimer solution was slowly added to the activated PEG 8 solution under protection atmosphere. This reaction mixture was then stirred for 3 days at room temperature under argon atmosphere and was then dialyzed for 2 days in water (membrane tube with MWCO 2000 g/mol). After the freeze drying a viscous liquid was obtained. Used quantities for PPI-PEG are presented in Table S1. Table S1. Used educts for the synthesis of PPI-PEG: PPI-PEG-I used for PPI-PEG-Man and PPI-PEG-II used for PPI-PEG-COD. PPI

Synthesis of maltosylated PPI-PEG-I (PPI-Mal-PEG-I) and PPI-PEG-II (PPI-Mal-PEG-II)
PPI-PEG (theoretical M w = 10567 g/mol) was stirred for one hour at 50 °C to dissolve it in sodium borate buffer (0.1 M; Na borate). After that maltose was added to the reaction mixture which was stirred for one hour at 50 °C to completely dissolve the solid maltose in the reaction mixture.
Borane*pyridine complex (BH 3 *Pyr) was lastly added to the reaction mixture, and this mixture was stirred for 7 days at 50 °C under reflux. Then the crude product was dialyzed for 4 days in water (membrane tube with MWCO of 2000 g/mol) and after freeze drying process a white solid was obtained. Used quantities for PPI-Mal-PEG-I and PPI-Mal-PEG-II are presented in Table S2.  Table S3.  After freeze drying process a white solid was obtained. Used quantities for converted PPI-Mal-PEG-NH 2 with COD are presented in Table S4.

Synthesis of PPI-PEG-Man
PPI-Mal-PEG-I-COD was dissolved in PBS buffer (0.01M), and azido-modified mannose (2) was dissolved in methanol. Both solutions were degassed with 3 freeze-pump-cycles followed by a 1h stirring at room temperature under argon atmosphere. Then adding the PPI dendrimer solution to 2containing solution and let stirred the resulting reaction for 3 days at 35 °C under argon and light protection. The crude product was then dialyzed in deionized water with steady water exchange for 2 days (membrane tube with MWCO of 2000 g/mol) and after a freeze drying a light yellow solid was obtained. Used quantities for the synthesis of PPI-PEG-Man are presented in Table S5.  (Figure S7).

Synthesis of PPI-Mal
G4 was stirred for 1h at 50 °C prior taken up in sodium borate buffer (0.1 M; Na borate). After the addition of maltose the reaction mixture was stirred for 1h at 50 °C again. Borane*pyridine complex (BH 3 *Pyr) was finally added to the reaction which was subsequently stirred for 7 days at 50 °C under reflux. The desired crude product was dialyzed for 4 days in deionized water with steady water exchange (membrane tube with MWCO of 2000 g/mol) and after a freeze drying step a white solid was obtained. Used quantities for the synthesis of PPI-Mal are presented in Table S6.
Molecular weight and composition of poly(propyleneimine) glycodendrimers are presented in Table   S7.  (Figure S8).  This precursor was used to synthesize PPI glycodendrimers OS-PEG and OS-PEG-Man.

Determination of coupled PEG-Spacer on OS-PEG and OS-PEG-Man
The  J. 2008, 14, 7030). Signal with * can be not identified, probably very less amount of NH-BOC of PPI-PEG. Signal with ** cannot be identified. Ammonium groups (NH 3 + , 8´´) can be elsewhere, but also protonated NH 2 groups of PPI-Mal-PEG-NH 2 can be elsewhere available.  J. 2008, 14, 7030). Explanation of signal with * and ** will be given below.  J. 2008, 14, 7030). Explanation of signal with * and ** will be given below.  J. 2008, 14, 7030). Explanation of signal with * and ** will be given below.

Analysis of 1 H NMR spectra of OS-PEG and OS-PEG-Man
1 H NMR spectrum of PPI-Mal-PEG-NH 2 ( Figure S4) outlines the required deprotection of tert. butyl group of NH-Boc group of PPI-Mal-PEG ( Figure S3). Typical broad singulet of tert. butyl group (8´´; slightly below 1.5 ppm) in PPI-Mal-PEG is almost disappeared in 1 H NMR spectrum of PPI-Mal-PEG-NH 2 . There could be a low content of residual signal of 8´´ for tert. butyl group of NH-Boc. This weak signal (slightly below 1.5 ppm) could be another signal which completely disappeared when deprotonated PPI-Mal-PEG-NH 2 is converted with COD derivative (Figure 1) to realize OS-PEG ( Figure S5) and OS-PEG ( Figure S6) for synthesizing OS-PEG-Man. (Figure S5 and S6) we follow the signal assignment of attached COD derivative published in following reference: Biomacromolecules 2020, 21, 199-213. Signals, 7´´

In both 1 H NMR spectra of OS-PEG
and 8´´-12´´, belong to the COD derivative in OS-PEG. Signals, 8´´, 11´´ and 12´´, are overlapped by signal of maltose units for 8´´ and PPI scaffold for 11´´ and 12´´. Proton signals, 9´´ and 10´´, belong to the cyclopropane ring in COD derivative and are visible between 1.0 -1.2 ( Figure S5 + S6), once coupled on a singular PEG spacer (Biomacromolecules 2020, 21, 199-213). Due to its hydrophobicity of cyclopropane ring in the COD derivative, we assume only broadening of the proton signals and low intensity of its protons here. It is true, one can see a broad 1 H NMR signal (e.g. Figure S5) which cannot belong to proton signals of cyclopropane ring in D 2 O, but presenting other structure units in OS-PEG (*). This broad signal (1-1.2 ppm) completely disappears when OS-PEG is converted with azido-mannose derivative to OS-PEG-Man ( Figure S7). Proton 11´´ of COD derivatives is smoothly available on singular PEG chain (1.25-1.5 ppm), but not in case of OS-PEG, while another broad 1 H NMR signal appears in this region of OS-PEG (1.25-1.5 ppm) which is not present in the precursor, PPI-Mal-PEG-NH 2 ( Figure S4). Triple bond is also not visible in FT-IR and Raman spectra, as known from terminal yne units. Beside this, reproducible 1 H NMR signal patterns are still visible in both OS-PEG, e.g. in red frame (Figure S5 and S6).    J. 2008, 14, 7030), which is presented in Figure S9.  Figure S3, while assignment of PPI scaffold is presented in Figure S9.  Table S7. M w values in Table S7 are Table S7. M w values in Table S7 are round up or down. Data for M + represents the top of mass peak. * represents degraded PPI-PEG-II. Figure S15. MALDI-TOF spectra of PPI-Mal. Corresponding value of M w is also presented in Table   S7. M w value in Table S7 is round up. Data for M + represents the top of mass peak.