Synthesis and in Vitro Evaluation of Monodisperse Amino-Functional

Biomacromolecules , 2017, 18 (12), pp 4323–4330. DOI: 10.1021/acs.biomac.7b01364. Publication Date (Web): November 13, 2017. Copyright © 2017 Ameri...
1 downloads 12 Views 1MB Size
Subscriber access provided by READING UNIV

Article

Synthesis and in vitro evaluation of monodisperse amino-functional polyester dendrimers with rapid degradability and antibacterial properties Patrik Stenström, Erik Hjort, Yuning Zhang, Oliver C. J. Andrén, Simon Guette-Marquet, Marianne Schultzberg, and Michael Malkoch Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.7b01364 • Publication Date (Web): 13 Nov 2017 Downloaded from http://pubs.acs.org on November 13, 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.

Biomacromolecules 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 24

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

Biomacromolecules

Synthesis and in vitro evaluation of monodisperse amino-functional polyester dendrimers with rapid degradability and antibacterial properties Patrik Stenström†, Erik Hjorth‡, Yuning Zhang†, Oliver C. J. Andrén†, Simon GuetteMarquet‡, Marianne Schultzberg‡, Michael Malkoch*,† †Department of Fibre and Polymer Technology, KTH Royal Institute of Technology, Teknikringen 48, 114 28, 8792-8766Stockholm ‡

Department of Neurobiology, Care Sciences and Society, Section of Neurodegeneration,

Center for Alzheimer Research, Karolinska Institutet, Blickagången 6, SE-141 57 Huddinge, Sweden KEYWORDS

Dendrimers, Bis-MPA, Polycationic, Cytotoxicity, Neurotoxicity, Antibacterial ABSTRACT

Amine

functional

polymers,

especially

cationically

charged,

are

interesting

biomacromolecules for several reasons including easy cell membrane entrance, their ability to escape endosomes through the proton sponge effect, spontaneous complexation and delivery of drugs and siRNA and simple functionalization in aqueous solutions. Dendrimers, a subclass of precision polymers, are monodisperse and exhibit a large and exact number of peripheral end groups in relation to their size and have shown promise in drug delivery,

1 ACS Paragon Plus Environment

Biomacromolecules

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 24

biomedical imaging and as anti-viral agents. In this work, hydroxyl functional dendrimers of generation 1 to 5 based on 2,2-bis(methylol)propionic acid (bis-MPA) were modified to bear 6 to 96 peripheral amino groups through esterification reactions with beta-alanine. All dendrimers were isolated in high yields and with remarkable monodispersity. This was successfully esterification

accomplished (FPE)

utilizing with

the

present

advantages

imidazole-activated

of

fluoride-promoted

monomers.

Straightforward

postfunctionalization was conducted on a 2nd generation amino-functional dendrimer with tetraethylene glycol through NHS-amidation and CDI activation to full conversion with short reaction times. Fast biodegradation of the dendrimers through loss of peripheral beta-alanine groups was observed and generational- and dose dependent cytotoxicity was evaluated with a set of cell lines. An increase in neurotoxicity compared to hydroxyl-functional dendrimers was shown in neuronal cells, however, the dendrimers were slightly less neurotoxic than commercially available poly(amidoamine) dendrimers (PAMAMs). Additionally, their effect on bacteria was evaluated and the 2nd generation dendrimer was found unique inhibiting the growth of Escherichia coli at physiological conditions while being non-toxic towards human cells. Finally, these results cement a robust and sustainable synthetic route to aminofunctional polyester dendrimers with interesting chemical and biological properties.

INTRODUCTION

Dendrimers belong to the most recent category of polymers. While described by Flory in 19511 for the first time, it was not until the late 1970s that dendrimers were synthetically accomplished by Vögtle et al.2 and later cemented in two separate reports by the groups of Tomalia3 and Newkome4. They are monodisperse hyper-branched macromolecules, often in the size range of no more than a few nanometers5 with a high number of end-groups that depend on the dendritic generation. These unique structural features have positioned them as cutting-edge precision polymers being promising candidates not only for biomedical 2 ACS Paragon Plus Environment

Page 3 of 24

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

Biomacromolecules

applications such as drug delivery6-8, imaging9,

10

and anti-viral agents11-13 but also as

calibrants for mass spectrometry14. Polyamidoamine (PAMAM) dendrimers were the first to be commercialized15 and have as a result been the dominating target of application driven research. Traditional PAMAM dendrimers display peripheral amino groups that make them cationically charged. Cationic dendrimers and other such polymers and nanoparticles are of interest due to their ability to penetrate cell membranes16,

17

and escape endosomal/lysosomal pathways via the proton-

sponge effect18. Cationic polymers are also effective antibacterial agents that, in contrast to many antibiotics, can neutralize bacteria without crossing the cytoplasmic membrane by instead disrupting it upon contact19. The bacteria surface is generally negatively charged, and thus attracts cationic polymers through polyelectrolyte adsorption. The cationic groups also enable them to complex and deliver short interfering RNA (siRNA)20, and the nucleophilic amino-functionality allows for easy functionalization in aqueous solutions through NHSamidation21 and other coupling reactions. However, the applicability in biological systems of PAMAM dendrimers is limited by their inherent toxicity22, 23 and the synthetic design that requires long reaction times and large excess of monomers to reach full conversions even for lower generations24. In addition, the stability and quality of commercially available PAMAM dendrimers has recently been questioned25. On the other hand, polyester dendrimers based on 2,2-bis-(methylol)propionic acid (bisMPA) have repeatedly shown to be biocompatible and biodegradable26,

27

, and can be

synthesized through facile chemistries in multi-gram scales without the need for column chromatography28. These unique properties make them top candidates that overcome recognized drawbacks related to PAMAM dendrimers. Interestingly, bis-MPA dendrimers have been post-functionalized to display peripheral amino groups. These dendrimers have been used by Li et al. to functionalize polymer vesicles29, and the degradability and potential

3 ACS Paragon Plus Environment

Biomacromolecules

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 24

as detoxification agents of lower generations of such dendrimers have recently been assessed by Duran-Lara et al.30. They have also been shown to be less toxic than PAMAM and capable of forming complexes with plasmid DNA31. However, efficient synthetic strategies to higher generation amino-functional bis-MPA dendrimers and their biological evaluation are not yet accomplished. We herein present a facile postfunctionalization approach to monodisperse bis-MPA dendrimers of generation one to five (G1 to G5) bearing up to an unprecedented 96 peripheral primary amines. This is accomplished through robust chemistries utilizing two esterification reactions: activation of boc-protected β-alanine through anhydride coupling, conventionally used to build bis-MPA dendrimers32 as well as the modification of these scaffolds with βalanine29, and fluoride-promoted esterification (FPE) with activation by carbonyl diimidazole (CDI), which is a more recent protocol with enhanced sustainability and product purity as well as good scalability28. The solution and storage stability of the amino-functional bis-MPA dendrimers is evaluated as well as their capacity to act as scaffolds readily available for postfunctionalization protocols. Additionally, their neurotoxicity in comparison to PAMAM dendrimers and neutrally charged bis-MPA dendrimers is investigated with assays on neuroblastoma cells (SH-SY5Y). Their dose-dependent cytotoxicity is studied on human dermal fibroblasts (HDF), mouse monocytes (RAW 264.7) and glioblastoma cells (U87). Finally, an initial investigation of their antibacterial properties is performed on Escherichia coli (E.coli). MATERIALS AND METHODS Materials

All chemicals and materials were obtained from Sigma Aldrich, Merck, VWR, Acros Organics or Fisher Chemicals unless otherwise noted. Deuterated solvents were obtained from Cambridge Isotope Laboratories. β-alanine was obtained from Alfa Aesar. CDI was obtained from Carbosynth. Bis-MPA and trimethylolpropane (TMP) were obtained in kind 4 ACS Paragon Plus Environment

Page 5 of 24

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

Biomacromolecules

from Perstorp AB, Sweden. Hydroxyl-functional dendrimers with TMP (TMP-Gx-OH and 2hydroxyethyl disulfide (SS-G4-OH) cores were prepared according to previously published protocols28, 33. An amino-functional cystamine-cored 3rd generation PAMAM dendrimer (SSG3-PAMAM-NH2) was obtained from Andrews ChemServices, USA, and its trifluoroacetate (TFA-) salt (SS-G3-PAMAM-NH3+TFA-) was obtained by dissolving the dendrimer in trifluoroacetic acid (TFA) followed by evaporation and freeze-drying. Monobenzylated tetraethylene glycol was prepared according to previously published protocol34. Human Dermal Fibroblasts (HDF), mouse monocyte (RAW 264.7) and glioblastoma (U87) cells were purchased from ATCC (American Tissue Culture Collection) and were maintained in Dulbecco's Modified Eagle Medium (DMEM) containing 10 % fetal bovine serum (FBS) and 100 units per mL of penicillin as well as 100 g/L streptomycin under 5% CO2 at 37°C. E.coli K-12 was purchased from ATCC and maintained in lysogeny broth (LB). DMEM, phosphate-buffered saline (PBS), penicillin/streptomycin and trypsin-EDTA were purchased from Gibco®. LB (Lennox) was obtained from Alfa Aesar. Synthesis protocols

Synthesis protocols are presented in the supporting information. Characterization methods

Matrix-assisted laser desorption/ionization-time-of-flight mass spectrometry (MALDI-TOF MS) was performed on a Bruker UltraFlex MALDI-TOF with a SCOUT-MTP Ion Source equipped with a nitrogen laser (337 nm), a gridless ion source and a reflector. The instrument was calibrated using SpheriCal™ calibrants. Samples were prepared by mixing sample, sodium trifluoroacetate (NaTFA) and matrix in 20 µL of THF at a 1:1:40 mass ratio and applying 1 µL to a stainless steel sample plate. Matrices used were DCTB for non-polar samples and 2,5-DHB for polar samples. For SS-G4-NH3+TFA- a mixture of HABA, DCTB and 2,5-DHB at a 6:1:1 mass ratio was used as matrix to get good resolution without reducing

5 ACS Paragon Plus Environment

Biomacromolecules

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 24

the disulfide bond in the MALDI. The obtained spectra were analyzed with FlexAnalysis from Bruker Daltonics, Bremen, version 2.2. All relevant MALDI spectra can be found in the supporting information. 1

H-NMR (400 MHz) and

13

C-NMR (101 MHz) were acquired using a Bruker Avance

instrument. 1H-NMR spectra were acquired using a spectral window of 20 ppm, a relaxation delay of 1 s and 16 scans.

13

C-NMR spectra were acquired using a spectral window of 240

ppm, a relaxation delay of 2 s and 512 scans. The spectra were processed and analyzed with MestreNova version 9.0.0-12821. All relevant NMR spectra can be found in the supporting information. Degradation testing

A 0.5 mM solution of G3 amino-functional bis-MPA dendrimer was prepared in phosphate-citrate buffers of pH 4.4, 5.4, 6.4 and 7.4 with an ionic strength of 0.1 M and kept at 37°C. The pH of the solutions was re-confirmed after the addition of dendrimer. Aliquots were analyzed with MALDI at set times. Dendrimer titration

10 mL of a solution of amino-functional dendrimer (G2: 8.2 mM, G3: 7.4 mM, G4: 6.5 mM) in deionized water was kept at room temperature. The pH was monitored with a Jenway 3510 pH meter calibrated with buffers of pH 4, 7 and 10. After each addition of 0.12 M NaHCO3 the solution was agitated for 2 min with a magnetic stirrer and kept without stirring until the pH had stabilized and could be recorded. MTT Assay for HDF, RAW 264.7 and U87

Cells were washed with PBS and harvested either with trypsin for HDF and U87 or by scraping for Raw 264.7. The cells were suspended in DMEM and counted with a hemocytometer.

6 ACS Paragon Plus Environment

Page 7 of 24

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

Biomacromolecules

100 µL DMEM containing 1×105 cells were transferred into each well of a 96-well plate. Cells were incubated for 24 h at 37°C with 5% CO2 followed by exchange to media with dendrimers at concentrations of 0.1 - 100 µM. The cells were incubated with dendrimers for another 24 h followed by addition of 10 µL of 5 g/L MTT agents. 4 h later, 10% sodium dodecyl sulfate (SDS) was added to each well and the plates were incubated overnight. The absorbance at 570 nm was read by a plate reader (Infinite 200 PRO). Cells with pure medium were used as negative control. Six replicates were performed for each concentration. Culturing of human SH-SY5Y neuroblastoma cells

Cells of the SH-SY5Y neuroblastoma cell line (LGC Standards GmbH, Wiesel, Germany) were cultured in growth medium (DMEM/F12 medium from Life Technologies, Stockholm, Sweden), supplemented with 10% heat-inactivated fetal calf serum (FCS) (Life Technologies, Stockholm, Sweden), at 37°C in humidified air containing 5% CO2. The cells were routinely subcultured at 80-100% confluence. Differentiated SH-SY5Y cells have been reported previously as a model of human neuron-like cells, which have morphological and biochemical similarity to primary neurons35. To induce a neuronal differentiation, the SHSY5Y cells were cultured in growth medium containing 10 µM retinoic acid (RA) (Sigma, Stockholm, Sweden) in a Falcon T175 bottle for 5 days, with change of medium at day 3. Following this, the cells were harvested with enzyme-free cell dissociation buffer (Life Technologies, Stockholm, Sweden). After centrifugation at 500 x g for 5 min, the cell pellet was resuspended in PBS and seeded in RA-supplemented growth medium at a density of 20,000 cells/well in Falcon 48-well plates. To improve attachment of differentiated cells the wells in the plate were coated with Matrigel™ reduced of growth factors (Becton Dickinson, USA) at a dilution of 1:10 prior to seeding. After allowing the cells to attach for one day, the medium was replaced with serum-free DMEM/F12 medium containing brain-derived neurotrophic factor (BDNF) (Life Technologies, Stockholm, Sweden) at a final concentration

7 ACS Paragon Plus Environment

Biomacromolecules

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 24

of 25 ng/ml, and the experiments were started after 5 days of incubation in this differentiation medium. Treatment of neuronal cells with dendrimers

The potential neurotoxicity of four types of dendrimers was tested: SS-G4-NH3+TFA- (bisMPA based), SS-G3-PAMAM-NH3+TFA-, TMP-G4-OH, and TMP-G4-NH3+TFA-. The dendrimers were dissolved in methanol (Sigma-Aldrich, Stockholm, Sweden) to a stock concentration of 2 mM, and stored at -20°C until use. Before treatment of the cells, the dendrimers were diluted to 0.01, 0.05, 0.1, 1 and 10 µM in DMEM/F12 medium. All treatments contained the same amount of methanol, and a vehicle control without any dendrimers was included in all experiments (0.5% methanol). The differentiation medium was aspirated and replaced with treatment medium, with care taken not to disturb or detach the neuronal cells. The cells were incubated with all treatments in quadruplicates or triplicates for 24 h. Treatment of neuronal cells with β-alanine

To evaluate the potential toxic effect of β-alanine on neuronal cells, differentiated SHSY5Y cells were exposed to β-alanine in the concentration range of 0 – 1 mM in DMEM/F12 for 24 h and 48 h. The effect of β-alanine alone, as well as the HCl and TFA salts was investigated. All treatments were performed in triplicate. Measurement of neurotoxicity

To assess the potential toxicity of the dendrimers, their effect on cell viability and cell death was analyzed. As indirect measurements of these phenomena, mitochondrial activity and membrane integrity were studied using the resazurin (Sigma-Aldrich, Stockholm, Sweden) and lactate dehydrogenase (LDH) (Roche, Stockholm, Sweden) assays, respectively. A portion of the conditioned medium was removed from each well and incubated with an equal volume of the LDH reagent, and a volume of resazurin salt dissolved

8 ACS Paragon Plus Environment

Page 9 of 24

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

Biomacromolecules

in DMEM/F12 was added to the well to a final concentration of 0.01% weight/volume. As a positive control for the LDH assay, complete lysis of all cells was induced by adding 2% Triton X-100 (Sigma-Aldrich, Stockholm, Sweden) to one well in the vehicle control group. The LDH assay was incubated for 30 min at 37°C after which the absorbance at 492 nm was assessed as a measure of loss of membrane integrity. The resazurin assay was incubated for 2 h at 37°C after which the fluorescence intensity was assessed as a measure of mitochondrial activity (ex/em 570/590 nm). A Tecan Safire 2 plate reader (Tecan Nordic, Stockholm, Sweden) was used to analyze absorbance and fluorescence. In all experiments, the average of the quadruplicates or triplicates of each treatment was calculated. Data preparation and analysis of neurotoxicity results

The data from the quadruplicates or triplicates from each treatment were averaged and normalized to positive control (LDH assay), or vehicle control (resazurin assay). The Kruskal-Wallis non-parametric ANOVA in Statistica v13 (Statsoft, Uppsala, Sweden) was used to test for a significant effect of treatment, with the built-in post hoc test for multiple comparisons used for analysis of significant differences between the treatments. E. coli. viability assay

Dry dendrimer samples were sterilized with UV-irradiation for 30 min. 100 µL of either LB or PBS with 1, 10 or 100 µM of dendrimer was transferred into a 96-well plate. 104 bacteria in 10 µL broth were inoculated to each well followed by 10 µL of resazurin agent. Plates were incubated at 37°C and the fluorescence was read by a plate reader (ex/em 560/590 nm) at the following time points: 2, 3, 4, 6 and 8 h. LB/PBS with E.coli. was used as negative control and samples with pure LB/PBS were used as blank, 6 replicates were prepared for each concentration.

9 ACS Paragon Plus Environment

Biomacromolecules

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 24

RESULTS AND DISCUSSION Synthesis of amino-functional dendrimers

Boc-protected amino-functional bis-MPA dendrimers of generation one to five with TMP cores (TMP-Gx-NHBoc) were successfully synthesized from hydroxyl dendrimers through functionalization with boc-protected β-alanine. For G1-G3, in situ activation of 1.5 eq. of beta-alanine per amino group by CDI and subsequent coupling via FPE chemistry was performed. In the case of higher generation bis-MPA dendrimers (G4 and G5), anhydride chemistry was employed and catalyzed by 4-DMAP. A disulfide cored dendrimer (SS-G4NHBoc) with 32 primary amines was also synthesized by anhydride coupling. This was sought out for a direct comparison to a G3 cystamine-cored PAMAM dendrimer with an equal number of end-groups. Both the FPE chemistry as well as the anhydride route were found effective and resulted in boc-protected dendrimers of high purities and with yields that averaged around 90%. All bisMPA dendrimers synthesized through the FPE strategy required extractions as purification step while the anhydride chemistry required additional purifications by column chromatography to remove troublesome urea byproducts. The FPE chemistry using cesium fluoride as a soft inorganic catalyst has been previously shown to be an efficient coupling protocol in producing bis-MPA dendrimers of high purity without the need for extensive purification28. The multiple boc groups were removed within 1.5 h by dissolving the dendrimers in a mixture of chloroform and TFA followed by precipitation in ether and freeze-drying. This procedure resulted in pure amino-functional dendrimers as trifluoroacetate salts (TMP-GxNH3+TFA- and SS-G4-NH3+TFA-), with yields averaging 75%. MALDI was used to monitor the reactions and confirm the structural perfection of the final dendrimers. Additionally, the dendrimers were characterized by 1H-NMR and

13

C-NMR to confirm their purity. All

characterizations are presented in the supporting information. The structural perfection of the 10 ACS Paragon Plus Environment

Page 11 of 24

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

Biomacromolecules

amino-functional bis-MPA dendrimers can be seen in the MALDI-spectra found in Fig. 1. Since the MALDI was set to positive mode, the dendrimers were detected as freebases without the negative trifluoroacetate groups. To our knowledge, this is the first reported facile synthesis of higher generation amino-functional bis-MPA dendrimers with high purity and monodispersity.

11 ACS Paragon Plus Environment

Biomacromolecules

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 24

Figure 1. Examples of the structures and MALDI spectra of the synthesized aminofunctional dendrimers (top) as well as the postfunctionalized dendrimers (bottom). MALDI spectra of SS-G4-NH3+TFA- and boc-protected dendrimers are shown in the supporting information. Functionalization of the amino-functional dendrimers

To showcase the scaffolding ability of amine functional bis-MPA dendrimers the G2 was functionalized via carbamate coupling with monobenzylated tetraethylene glycol (BnTEG) and via amidation with the same species modified with a carboxylic acid linker (BnTEGCOOH) using CDI and NHS as coupling agents. The carbamate coupling was performed in acetone through activation of BnTEG with CDI. With 2 eq. per amino group of BnTEG this reaction reached almost full conversion overnight. After addition of another 2 eq. of BnTEG along with 1 eq. of trimethylamine (TEA) per amino group the reaction reached full conversion within 24 h giving monodisperse PEGylated dendrimers. The somewhat slow reaction rate can be attributed to the relatively good stability of the imidazolide-activated hydroxyls on BnTEG, but interestingly the dendrimers are stable enough in the organic solvent to allow completion of the reaction before degradation starts to occur. However, the same G2 dendrimer reacted with only 1.5 eq. per amino group of CDI-activated BnTEGCOOH to full conversion within 1 hour of reaction time at room temperature without any added catalyst. The activated carboxylic acid is more reactive and is a perfect fit for postfunctionalization of the nucleophilic amino groups of the dendrimers. BnTEGCOOH was also modified with NHS and deprotected through catalytic hydrogenation to give hydroxyl functional NHS-activated tetraethylene glycol (TEGCOONHS). The dendrimers were titrated with NaHCO3 (Fig. S4) to calculate the amount of NaHCO3 needed to reach a pH of around 8 where the NHS reaction is effective. In an aqueous solution at this pH, the dendrimer reacted exceptionally fast to full conversion in 10 minutes using only 1.5 eq. of TEGCOONHS per

12 ACS Paragon Plus Environment

Page 13 of 24

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

Biomacromolecules

amino group to give monodisperse dendrimers with a molecular weight of 5348 Da and 12 hydroxyl functional short pegs. MALDI-spectra of the three postfunctionalized dendrimers are shown in Fig. 1. Degradation studies

The influence of pH on the degradation of bis-MPA dendrimers has previously been demonstrated on hydroxyl27- and lysine30(amino)-functional dendrimers. pH-triggered degradation is an important feature since it allows for efficient elimination of substances that are used as drug carriers and it can be used to trigger drug release since the pH varies e.g. between healthy tissue and tumors36, 37. The stability of the G3 amino-functional dendrimer was monitored in physiologically relevant pH of 4.4, 5.4, 6.4 and 7.4 at 37°C with MALDI (Fig. 2).

13 ACS Paragon Plus Environment

Biomacromolecules

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 24

Figure 2. MALDI of TMP-G3-NH3+TFA- at 37°C at various pH (top) and diagram showing its amount of remaining amines in the same conditions at various time points (bottom). The dendrimer showed increasing degradation rates at higher pH, which can be explained by the increased nucleophilicity of the dendrimer’s amino groups and the surrounding water at high pH. This increase facilitates the lysis of the ester bonds through back-biting or intermolecular reactions, and the hydrolysis by water. Similar behavior has been demonstrated previously on bis-MPA dendrimers functionalized with lysine30. At physiological conditions (pH 7.4 and 37°C), the bis-MPA dendrimer degrades exceptionally fast through loss of its peripheral β-alanines. After only 3 h the dendrimer has lost its monodisperse integrity and the average dendrimer displays 17 amino groups out of its original 24, as estimated from the MALDI peak intensities. After one day, the average dendrimer has less than 6 alanines left. The two and three days assessments revealed an initiated loss of the internal bis-MPA units while the number of alanines had stagnated at a presence of around 2-3 amino groups. At pH 4.4 the average dendrimer still display above 20 amines after two days in solution at 37°C, which confirms the strong effect of pH on the degradation. Despite the rapid degradation in aqueous solutions, the dendrimers could be stored dry overnight without signs of degradation at room temperature (Fig. S5) and for at least 168 days at -20°C. After 681 days of dry storage at -20°C degradation had occurred. However, the dendrimer still had above 96% of its original structural integrity (Fig. S6). The containers were flushed briefly with nitrogen gas prior to dry storage to avoid condensation. Cytotoxicity assays

A concern regarding cationic polymers is toxicity. Cationic amino-functional PAMAM dendrimers have interesting properties for biomedical applications, but they are limited by their strong inherent toxicity, especially at higher generations. To understand the toxicity profile of the amino-functional bis-MPA dendrimers, screening was conducted against skin 14 ACS Paragon Plus Environment

Page 15 of 24

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

Biomacromolecules

cells (HDF), monocytes (RAW 264.7) and glioblastoma cells (U87) after 24 h of incubation with concentrations of 0.1-50 µM (Fig. 3I-K) and an extreme concentration of 100 µM (Fig. S3). Additionally, the TFA salt of β-alanine was evaluated at concentrations ranging from 12 µM to 2.4 mM, as well as its HCl salt for comparison. After 24 h of incubation, dose- and generation-dependent cytotoxicity was observed; however, the smaller G1 and G2 dendrimers were non-toxic at most concentrations in all three cell lines, which makes them promising candidates in biomedical applications. Slight toxicity was detected at 100 µM, mainly towards HDF and RAW 264.7 (Fig. S2). The G3 dendrimer was tolerated only at low concentrations (up to 1 µM in RAW 264.7 and U87 and up to 5 µM in HDF). The G4 and G5 were both toxic above 0.1 µM. The TFA and HCl salts of β-alanine were both non-toxic towards HDF, RAW 264.7 and U87 at up to 2.4 mM (Fig. S1). Interestingly, the results suggest that the TFA salt of the dendrimer was slightly more benign than the HCl salt.

15 ACS Paragon Plus Environment

Biomacromolecules

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 24

Figure 3. A-H shows effects of 0.1 to 10 µM of dendrimers on cell viability and cell death on differentiated SH-SY5Y neuronal cells, and vehicle control (0.5% methanol). After 24h of incubation, cell viability was measured with the resazurin assay (A-D), while cell death was measured with the LDH assay (E-H). * indicate p < 0.05 as compared to vehicle control, and 16 ACS Paragon Plus Environment

Page 17 of 24

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

Biomacromolecules

# indicates p < 0.05 compared to 10 µM dendrimer. PC = Positive control. I-K show cell viability of HDF, RAW 264.7 and U87 cells after 24 h of treatment with dendrimers at concentrations of 0.1-50 µM as measured by the MTT assay. L shows the viability of HDF, RAW 264.7 and E.coli at 37°C at different time intervals of treatment with dendrimers at a concentration of 100 µM. In vitro neurotoxicity assays

The neurotoxicity of bis-MPA dendrimers with amino groups and hydroxyl groups was assessed against human SH-SY5Y neuroblastoma cells by comparing the effects of TMP-G4NH3+TFA- and TMP-G4-OH. Moreover, a PAMAM dendrimer (SS-G3-PAMAM-NH3+TFA) and a bis-MPA dendrimer with an equal number of amino end-groups (SS-G4-NH3+TFA-) were studied in the same system. To increase the accuracy, two assays were used to individually measure cell death (LDH) and cell viability (resazurin). In general, the effects on cell death as measured by the LDH assay (Fig. 3E-H) were more pronounced than the effects on cell viability as measured by the resazurin assay (Fig. 3A-D). As expected, dendrimers with amino groups were more toxic than those with hydroxyls, which is exemplified by the difference between TMP-G4-NH3+TFA- and TMP-G4-OH. A significant decrease in cell viability was observed in the LDH assay for SS-G4-NH3+TFAand SS-G3-PAMAM-NH3+TFA- (p < 0.05), as well as for TMP-G4-NH3+TFA- (p < 0.005). However, in the multiple comparisons between treatments no significant differences could be detected when compared to the vehicle control (only aqueous buffer with 0.5% methanol). Significant cell death was observed after treatment with SS-G3-PAMAM-NH3+TFA- (p < 0.05) and TMP-G4-NH3+TFA- (p < 0.05). Although the results were suggestive of a toxic effect of 10 µM of SS-G4-NH3+TFA-, this was just void of significance (p = 0.0535), which indicates that the bis-MPA dendrimer is less toxic than its PAMAM counterpart. Furthermore, the TMP-G4-OH dendrimer, SS-G4-NH3+TFA- at 1 µM, and TMP-G4-

17 ACS Paragon Plus Environment

Biomacromolecules

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 24

NH3+TFA- at 0.01 - 0.1 µM appeared to have a slight stimulatory effect on cellular respiration; however, these observations lacked statistical significance. Nevertheless, the possibility that some types of dendrimers cause toxicity by influencing mitochondrial function is interesting and plausible. No signs of neurotoxicity of β-alanine were detected – with or without counter ions (data not shown). Antibacterial assessment

Finally, the short-term effect of the amino-functional dendrimers on E.coli bacteria was evaluated at different concentrations for G1-G4 for 2-8 h at 37°C (Fig. S3). The viability of cells at the same concentrations and time intervals was also studied. The G1 and G2 dendrimers were non-toxic towards HDF and RAW 264.7 at all concentrations and time periods, but the G1 did not show any significant antibacterial effect. Importantly, the G2 exhibited more or less complete inhibition of bacterial growth at 100 µM up to 8 h, and was non-toxic to cells at the same conditions (Fig. 3L). Even though 100 µM can be considered a high concentration for injections, it corresponds to a mass concentration of only 0.034 wt% in a water solution, which cannot be considered high in topical applications such as gels, creams or films. The study was also carried out at 37°C in broth, which is an excellent environment for bacteria. In a more challenging environment for the bacteria with lower temperature and less nutrition such as the human skin or material surfaces, an even lower concentration might be sufficiently antibacterial. In the tested conditions, bacterial growth is partly inhibited by the G2 up to 6 h also at 10 µM. When bacteria were cultured without broth, 10 µM of G2 gave similar antibacterial activity as 100 µM with broth (data not shown). The antibacterial activity of polycations is generally initiated by polyelectrolyte adsorption onto the bacteria19, and the increased bacterial inhibition of the G2 in comparison to the G1 can possibly be explained by the fact that the charge density in the latter is insufficient. In contrast to the G2 dendrimers, the G3 and G4 generation dendrimers were only significantly antibacterial when

18 ACS Paragon Plus Environment

Page 19 of 24

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

Biomacromolecules

they were also toxic against cells. It is apparent that the G2 has a good combination of charge density and mobility to enable efficient interaction with the bacteria. These results show that amino-functional bis-MPA dendrimers – TMP-G2-NH3+TFA- in particular – exhibit interesting antibacterial properties. The effect in a variety of environments and on different bacterial strains, including gram-positive bacteria, would be a suggestion for further research on these materials. CONCLUSIONS

Amino-functional dendrimers of generation 1 to 5 based on bis-MPA have been synthesized through ease-of-use protocols where multiple boc-protected beta-alanine groups are efficiently attached to hydroxyl-functional dendrimers. Dendrimers of high purity and structural integrity were obtained in high yields. The amino-functional dendrimers were obtained through simple deprotection using TFA as an acidic reagent. Successful postfunctionalizations with tetraethylene glycol were also conducted to form peptide linkages with NHS, and to form peptide and carbamate linkages with CDI. The set of aminofunctional dendrimers showed generation- and concentration-dependent cytotoxicity towards fibroblasts, monocytes and glioblastoma cells. The toxicity of amino-functional dendrimers was as expected considerably higher than that of hydroxyl-functional dendrimers, but a slightly lower neurotoxicity towards SH-SY5Y neuroblastoma cells was observed in comparison to amino-functional PAMAM dendrimers. In a 100 µM solution the 2nd generation dendrimer was able to completely inhibit growth of E.coli in broth at 37°C for up to 8 h while being non-toxic to cells at the same time conditions. This shows that cationic bisMPA dendrimers are a candidate worthy of further evaluation in the fight against bacteria when an ever increasing resistance is forcing us to find alternatives to conventional penicillin. ASSOCIATED CONTENT The supporting information is available free of charge.

19 ACS Paragon Plus Environment

Biomacromolecules

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 20 of 24

Additional cytotoxicity and anti-bacterial data, titration curves, synthesis procedures with NMR and MALDI-spectra (PDF) AUTHOR INFORMATION Corresponding Author

*Michael Malkoch, Department of Fibre and Polymer Technology, KTH Royal Institute of Technology, Teknikringen 48, 114 28, Stockholm. ACKNOWLEDGMENTS

This research was supported by Wilhelm Beckers Jubileumsfond, Wallenberg Academy Fellow (2012-0196) and The Swedish Research Council (2014-3876). REFERENCES

1. Flory, P. J., J Am Chem Soc 1952, 74, (11), 2718-2723. 2. Buhleier, E.; Wehner, W.; Vögtle, F., Synthesis 1978, 1978, (02), 155-158. 3. Tomalia, D. A.; Baker, H.; Dewald, J.; Hall, M.; Kallos, G.; Martin, S.; Roeck, J.; Ryder, J.; Smith, P., Polym J 1985, 17, (1), 117-132. 4. Newkome, G. R.; Yao, Z. Q.; Baker, G. R.; Gupta, V. K., J Org Chem 1985, 50, (11), 2003-2004. 5. Lee, C. C.; MacKay, J. A.; Frechet, J. M.; Szoka, F. C., Nat Biotechnol 2005, 23, (12), 1517-1526. 6. Kojima, C.; Kono, K.; Maruyama, K.; Takagishi, T., Bioconjug Chem 2000, 11, (6), 910917. 7. De Jesus, O. L. P.; Ihre, H. R.; Gagne, L.; Frechet, J. M. J.; Szoka, F. C., Bioconjug Chem 2002, 13, (3), 453-461.

20 ACS Paragon Plus Environment

Page 21 of 24

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

Biomacromolecules

8. Morgan, M. T.; Carnahan, M. A.; Immoos, C. E.; Ribeiro, A. A.; Finkelstein, S.; Lee, S. J.; Grinstaff, M. W., J Am Chem Soc 2003, 125, (50), 15485-15489. 9. Wiener, E. C.; Brechbiel, M. W.; Brothers, H.; Magin, R. L.; Gansow, O. A.; Tomalia, D. A.; Lauterbur, P. C., Magn Reson Med 1994, 31, (1), 1-8. 10. Ziemer, L. S.; Lee, W. M.; Vinogradov, S. A.; Sehgal, C.; Wilson, D. F., J Appl Physiol 2005, 98, (4), 1503-1510. 11. Bourne, N.; Stanberry, L. R.; Kern, E. R.; Holan, G.; Matthews, B.; Bernstein, D. I., Antimicrob Agents Chemother 2000, 44, (9), 2471-2474. 12. Bernstein, D. I.; Bourne, N.; Ayisi, N. K.; Ireland, J.; Matthews, B.; McCarthy, T.; Sacks, S., Antiviral Res 2003, 57, (3), A66-A66. 13. Jiang, Y. H.; Emau, P.; Cairns, J. S.; Flanary, L.; Morton, W. R.; McCarthy, T. D.; Tsai, C. C., AIDS Res Hum Retroviruses 2005, 21, (3), 207-213. 14. Grayson, S. M.; Myers, B. K.; Bengtsson, J.; Malkoch, M., J Am Soc Mass Spectrom 2014, 25, (3), 303-309. 15. Esfand, R.; Tomalia, D. A., Drug Discov Today 2001, 6, (8), 427-436. 16. Arvizo, R. R.; Miranda, O. R.; Thompson, M. A.; Pabelick, C. M.; Bhattacharya, R.; Robertson, J. D.; Rotello, V. M.; Prakash, Y. S.; Mukherjee, P., Nano Lett 2010, 10, (7), 2543-2548. 17. Lin, J. Q.; Alexander-Katz, A., Acs Nano 2013, 7, (12), 10799-10808. 18. Behr, J. P., Chimia 1997, 51, (1-2), 34-36.

21 ACS Paragon Plus Environment

Biomacromolecules

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 22 of 24

19. Ikeda, T.; Yamaguchi, H.; Tazuke, S., Antimicrob Agents Chemother 1984, 26, (2), 139-144. 20. Zhou, J. H.; Wu, J. Y.; Hafdi, N.; Behr, J. P.; Erbacher, P.; Peng, L., Chem Commun 2006, (22), 2362-2364. 21. Anderson, G. W.; Zimmerman, J. E.; Callahan, F. M., J Am Chem Soc 1964, 86, (9), 1839-1842. 22. Malik, N.; Wiwattanapatapee, R.; Klopsch, R.; Lorenz, K.; Frey, H.; Weener, J. W.; Meijer, E. W.; Paulus, W.; Duncan, R., J Control Release 2000, 68, (2), 299-302. 23. Leroueil, P. R.; Berry, S. A.; Duthie, K.; Han, G.; Rotello, V. M.; McNerny, D. Q.; Baker, J. R.; Orr, B. G.; Holl, M. M. B., Nano Lett 2008, 8, (2), 420-424. 24. Esfand, R.; Tomalia, D. A., Laboratory Synthesis of Poly(amidoamine)(PAMAM) Dendrimers. In Dendrimers and Other Dendritic Polymers; Frechét, J. M. J, Tomalia, D. A., Eds.; John Wiley & Sons, Ltd, Chichester, 2002, 587-604. 25. Lloyd, J. R.; Jayasekara, P. S.; Jacobson, K. A., Anal. Methods 2016, 8, (2), 263-269. 26. Gillies, E. R.; Dy, E.; Frechet, J. M.; Szoka, F. C., Mol Pharm 2005, 2, (2), 129-138. 27. Feliu, N.; Walter, M. V.; Montanez, M. I.; Kunzmann, A.; Hult, A.; Nystrom, A.; Malkoch, M.; Fadeel, B., Biomaterials 2012, 33, (7), 1970-1981. 28. Garcia-Gallego, S.; Hult, D.; Olsson, J. V.; Malkoch, M., Angew Chem Int Ed Engl 2015, 54, (8), 2416-2419. 29. Li, B.; Martin, A. L.; Gillies, E. R., Chem Commun (Camb) 2007, (48), 5217-5219.

22 ACS Paragon Plus Environment

Page 23 of 24

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

Biomacromolecules

30. Duran-Lara, E. F.; Marple, J. L.; Giesen, J. A.; Fang, Y. L.; Jordan, J. H.; Godbey, W. T.; Marican, A.; Santos, L. S.; Grayson, S. M., Biomacromolecules 2015, 16, (11), 34343444. 31. Movellan, J.; Gonzalez-Pastor, R.; Martin-Duque, P.; Sierra, T.; de la Fuente, J. M.; Serrano, J. L., Macromol Biosci 2015, 15, (5), 657-667. 32. Malkoch, M.; Malmstrom, E.; Hult, A., Macromolecules 2002, 35, (22), 8307-8314. 33. Östmark, E.; Macakova, L.; Auletta, T.; Malkoch, M.; Malmström, E.; Blomberg, E., Langmuir 2005, 21, (10), 4512-4519. 34. Stenström, P.; Andren, O. C.; Malkoch, M., Molecules 2016, 21, (3). doi:10.3390/molecules21030366 35. Encinas, M.; Iglesias, M.; Liu, Y. H.; Wang, H. Y.; Muhaisen, A.; Cena, V.; Gallego, C.; Comella, J. X., J Neurochem 2000, 75, (3), 991-1003. 36. Helmlinger, G.; Yuan, F.; Dellian, M.; Jain, R. K., Nat Med 1997, 3, (2), 177-182. 37. Zhou, Z.; Ma, X.; Murphy, C. J.; Jin, E.; Sun, Q.; Shen, Y.; Van Kirk, E. A.; Murdoch, W. J., Angew Chem Int Ed Engl 2014, 53, (41), 10949-10955.

23 ACS Paragon Plus Environment

Biomacromolecules

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

Table of Content 89x36mm (300 x 300 DPI)

ACS Paragon Plus Environment

Page 24 of 24