Gluconamide Cross-Linking in

Nov 30, 2017 - Department of Bioengineering, Graduate School of Engineering, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan. â...
0 downloads 7 Views 3MB Size
Subscriber access provided by READING UNIV

Article

Polyplex micelles with phenylboronate/gluconamide crosslinking in the core exerting promoted gene transfection through spatiotemporal responsivity to intracellular pH and ATP concentration Naoto Yoshinaga, Takehiko Ishii, Mitsuru Naito, Taisuke Endo, Satoshi Uchida, Horacio Cabral, Kensuke Osada, and Kazunori Kataoka J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.7b08816 • Publication Date (Web): 30 Nov 2017 Downloaded from http://pubs.acs.org on November 30, 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.

Journal of the American Chemical Society 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 28 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

Journal of the American Chemical Society

Polyplex micelles with phenylboronate/gluconamide crosslinking in the core exerting promoted gene transfection through spatio-temporal responsivity to intracellular pH and ATP concentration.

Naoto Yoshinaga,† Takehiko Ishii,† Mitsuru Naito,‡ Taisuke Endo,§ Satoshi Uchida,† Horacio Cabral,† Kensuke Osada,*,† and Kazunori Kataoka*,∥,⊥ †

Department of Bioengineering, Graduate School of Engineering, The University of Tokyo, 7-3-1

Hongo, Bunkyo-ku, Tokyo 113-8656, Japan. ‡

Center for Disease Biology and Integrative Medicine, Graduate School of Medicine, The University

of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan. §

Department of Material Engineering, Graduate School of Engineering, The University of Tokyo,

7-3-1 Hongo, Bunkyo-ku, Tokyo 113-8656, Japan. ∥Innovation

Center of NanoMedicine (iCONM), Kawasaki Institute of Industrial Promotion, 3-25-14

Tonomachi, Kawasaki-ku, Kawasaki, 210-0821, Japan ⊥Policy

Alternatives Research Institute, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo

113-0033, Japan. *Corresponding authors: Professor Kensuke Osada and Professor Kazunori Kataoka E-mail: [email protected] and [email protected]

KEYWORDS:

Polyplex

micelle,

Gene

delivery,

Phenylboronic

ATP-responsivity

ACS Paragon Plus Environment

acid,

pH-responsivity,

Journal of the American Chemical Society 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 28

Abstract Polyplexes as gene delivery carriers require integrated functionalities to modulate intracellular trafficking for efficient gene transfection. Herein, we developed plasmid DNA (pDNA)-loaded polyplex micelles (PMs) from poly(ethylene glycol) (PEG)-based block catiomers derivatized with 4-carboxy-3-fluorophenylboronic acid (FPBA) and

D-gluconamide

(GlcAm) to

form pH- and ATP-responsive crosslinking in the core. These PMs exhibited robustness in the extracellular milieu, smooth endosomal escape after cellular uptake, and facilitated pDNA decondensation triggered by increased ATP concentration inside of the cell. Laser confocal microscopic observation revealed that FPBA installation enhanced the endosomal escapability of the PMs; presumably, this effect resulted from the facilitated endo/lysosomal membrane disruption triggered by the released block catiomers with hydrophobic FPBA moieties in the side chain from the PM at lower pH condition of endo/lysosomes. Furthermore, the profile of intracellular pDNA decondensation from the PMs was monitored using Förster resonance energy transfer (FRET) measurement by flow cytometry; these observations confirmed that PMs optimized for ATP-responsivity exerted effective intracellular decondensation of loaded pDNA to attain promoted gene transfection.

1. Introduction Polyplex micelles (PMs) have attracted attention as non-viral vectors applicable to gene therapy; these structures electrostatically self-assemble from negatively-charged plasmid DNA (pDNA) and block catiomers composed of polycation segment and non-ionic hydrophilic segment, such as poly(ethylene glycol) (PEG).1-6 They have a unique core-shell architecture with hydrophilic segments forming a shell to prevent the polyplex core composed of pDNA and polycation from non-specific interaction with biological components in harsh in vivo conditions. Moreover, PMs can be integrated with a variety of functions by manipulating block catiomer structures; such

ACS Paragon Plus Environment

Page 3 of 28 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

Journal of the American Chemical Society

modifications can facilitate gene transfection by (1) protecting the loaded pDNA under physiological conditions en route to the cellular target; (2) attaining effective cellular uptake and subsequent endosomal escape; and (3) achieving effective intracellular decondensation of loaded pDNA for efficient gene transcription. Current efforts have been devoted to optimize these functions, including stabilization of PM structures by introducing hydrophobic moieties,7-9 incorporation of chemical motifs to facilitate endosomal escape,10-12 and introduction of chemical crosslinking that disintegrates selectively in the intracellular environment.13-14 Nevertheless, these functions often conflict with one another, and present difficulties to the practical application of PMs as non-viral gene vectors. Here, we focused on a phenylboronic acid (PBA) group to overcome multiple issues such as stability in the extracellular milieu, endosomal escapability, and controlled intracellular decondensation of pDNA payload. In aqueous milieu, PBA and its derivatives are known to be in equilibrium between an uncharged trivalent form and a negatively-charged tetravalent form.15-19 We note that tetravalent boronate selectively forms stable ester linkage with diol compounds in aqueous solutions (Scheme 1), which is facilitated by an increase in both pH and the concentration of diol compounds.20-21 This dual responsivity of PBA endows PBA-modified block catiomers with unique functions relevant to gene delivery. Indeed, by fine tuning the acid dissociation constant (pKa) of PBA moieties in the polymer chain, we developed PBA-modulated delivery systems that selectively released their encapsulated cargos in response to an increase in intracellular concentration of adenosine triphosphate (ATP), 22-23 as ATP has a ribose structure with strong binding affinity for PBA moieties. Recently, this concept of ATP-responsive nanocarriers was further applied to pDNA-loaded polyplexes,24 implying a wide utility for PBA-modulated carrier systems in ATP-responsive intracellular delivery of nucleic acid compounds. Other unique properties of PBA are pH-responsive change in hydrophobicity and diol-binding ability. These properties may enable a PBA-modulated delivery system with facilitated endosomal escape, which is believed to be critical to increasing transfection efficiency of polyplex systems.

ACS Paragon Plus Environment

Journal of the American Chemical Society 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 28

Accordingly, we developed a simple yet potent strategy for constructing novel PBA-modulated PMs, optimizing to exert the multiple functionalities of pH-responsive endosomal escape and ATP-mediated pDNA decondensation. Our strategy entailed mixing pDNA with a set of block catiomers: one with PBA moieties and the other with polyol (D-gluconamide (GlcAm)) moieties. This strategy allows the selective formation of phenylboronate ester intermolecular crosslinks between PBA- and GlcAm-modulated block catiomers, stabilizing the condensed polyplex core after self-assembly with pDNA. Importantly, the ribose structure in ATP has a higher affinity for PBA than the open-chained polyol structure in GlcAm.18, 25 Thus, PBA/GlcAm crosslinking may be cleaved inside of the cell due to the exchange reaction with ATP in elevated concentration. This cleavage would eventually induce release of the pDNA cargo by facilitating disintegration of the PM; this disintegration would result from the replaced negative charges of phosphate groups in ATP. Moreover,

to

allow

the

PBA/GlcAm

crosslinks

to

be

cleaved

in

the

endosome,

4-carboxy-3-fluorophenylboronic acid (FPBA), whose pKa is estimated to be 7.2 after introduction to the cationic side chain of the catiomer via amide coupling,26 was used in this study to drastically shift the equilibrium of FPBA from tetravalent to trivalent corresponding to a change in pH from 7.4 (physiological pH) to 5.5 (late endosomal pH). Furthermore, the increase of the hydrophobic FPBA fraction with uncharged trivalent form in the endosomes due to the pH decrease may promote the association of the block catiomers with endosomal membranes and assist the catiomer-mediated membrane destabilization, facilitating the translocation of PMs into cytoplasm (Scheme 2). In this regard, poly{N’-[N-(2-aminoethyl)-2-aminoethyl]aspartamide} (PAsp(DET)),12,

27

which shows

lower cytotoxicity to cultured cells comparing with a commercial transfection agent of linear polyethyleneimine,28 was chosen as a platform catiomer of block copolymers for introducing FPBA or GlcAm groups into the side chain with taking an advantage of the ability of PAsp(DET) to disrupt endosomal membrane selectively at acidic pH condition of late endosomes. The PM structure was firstly optimized from the standpoints of stability and dual pH- and ATP-responsivity in buffer

ACS Paragon Plus Environment

Page 5 of 28 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

Journal of the American Chemical Society

solution, followed by cell-based experiments for validating the enhanced cellular uptake, endosomal escape, and ATP-responsive pDNA decondensation inside of the cell. Moreover, the potency of the optimized FPBA/GlcAm-modulated PMs for enhancing transfection was evaluated by measuring condensation status of pDNA inside of the cells, using intracellular Förster resonance energy transfer (FRET) method.

2. Results and Discussion 2.1 Synthesis of PEG-b-PAsp(DET/FPBA) and PEG-b-PAsp(DET/GlcAm) FPBA and GlcAm groups were introduced separately to the side chain of the PAsp(DET) segment of PEG-b-PAsp(DET) (Scheme 3) to avoid the formation of intrachain crosslinking, which may occur if FPBA and GlcAm groups are co-conjugated into the same strand of the block catiomer. Introduction ratios of FPBA and GlcAm were systematically varied to find the suitable composition for crosslinking while managing the stability and responsivity of PMs to pH and ATP. Eventually, the introduction ratios of FPBA in the PAsp(DET) side chain were controlled to be 14, 26, 59, and 89%, while those of GlcAm were 9, 14, 27, and 54%, as determined by 1H-NMR spectra (Figure S1). These block catiomers with varying introduction ratios of primary amino group to FPBA or GlcAm are denoted as PEG-b-PAsp(DET/FPBAX%) and PEG-b-PAsp(DET/GlcAmY%), respectively. Note that all of the PEG-b-PAsp(DET/FPBA) were soluble in aqueous solution at pH 7.4, showing comparable light scattering intensity with that of the PEG-b-PAsp(DET), even though 39% of the introduced FPBA groups are estimated to adopt hydrophobic trivalent structure at pH 7.4, based on the pKa value.

2.2

Characterization

of

PMs

prepared

from

PEG-b-PAsp(DET/FPBA)

and

PEG-b-PAsp(DET/GlcAm) PEG-b-PAsp(DET/FPBAX%) and PEG-b-PAsp(DET/GlcAmY%) solutions were mixed with

ACS Paragon Plus Environment

Journal of the American Chemical Society 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 28

varying composition to prepare a library of block catiomer solutions. These blended solutions exhibited comparable light scattering intensities with that of unmodified PEG-b-PAsp(DET) solution, suggesting

that

PEG-b-PAsp(DET/FPBA)

and

PEG-b-PAsp(DET/GlcAm)

did

not

form

intermolecular crosslinking in the mixture by forming phenylboronate ester linkages (data not shown). Presumably, the electrostatic repulsion between these block catiomers prevented their close association for the formation of intermolecular FPBA/GlcAm ester linkages. Next, the blended solution of block catiomers was mixed with pDNA solution at pH 7.4 in a residual charge ratio (r) of 1.5 to form PM (details are described in the section entitled “Preparation of polyplex micelles” in Supporting Information). The DLS measurement of the samples prepared from all of the combinations of block catiomers indicated the formation of particles with cumulant diameter of 75-96 nm and PDI of 0.15-0.25 (Table 1). These values are consistent with those previously reported for PMs prepared from similar block catiomers,27 indicating the formation of PMs from the mixture of PEG-b-PAsp(DET/FPBA) and PEG-b-PAsp(DET/GlcAm). Hereafter, PMs prepared from PEG-b-PAsp(DET/FPBAX%) and PEG-b-PAsp(DET/GlcAmY%) are denoted as BX/GY PM. The efficiency of core crosslinking was next evaluated by examining the tolerability of PM against sodium dextran sulfate (DS), which causes the release of loaded pDNA from PM by polyion exchange reaction. First, to find the best combination of PEG-b-PAsp(DET/FPBAX%) and PEG-b-PAsp(DET/GlcAmY%) for PM stabilization, the tolerability of all the PMs against DS addition was evaluated by gel electrophoresis. The electrophoregrams (Figure 1a-d) revealed the release behavior of pDNA from PMs upon the addition of DS with residual sulfate groups of 8 times equivalents to the phosphate groups of pDNA (A/P ratio of 8). Obviously, there was an optimum range in the introduction ratio of FPBA and GlcAm residues into the PAsp(DET) segment to suppress pDNA release from PMs, and B59/G27 and B59/G54 PMs (Figure 1c) were stable enough to show no release of the loaded pDNA at this condition. With further increase of the A/P ratio to 9, only the B59/G27 PM kept pDNA in the PM core (Figure 1e), scoring as the most stable PM against

ACS Paragon Plus Environment

Page 7 of 28 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

Journal of the American Chemical Society

DS exchange. The optimal tolerability against DS exchange of this formulation may be explained by the balance between the crosslinking density in the core and the efficacy of electrostatic interaction between pDNA and block catiomers. Thus, an increase in the number of FPBA and GlcAm residues in each of the PAsp(DET) strands is apparently favorable for increasing the crosslinking density, though they reduce the charge density in PAsp(DET/FPBA) strand due to the presence of the negatively charged tetravalent (-B--(OH)3) form, impairing the formation of continuous ion-pair repeats along the chain. Furthermore, as shown in Figure 1f, the stabilization occurred only when both PEG-b-PAsp(DET/FPBA59%) and PEG-b-PAsp(DET/GlcAm27%) were involved in the PM structure, indicating the significant role of the intermolecular crosslinking formation through phenylboronate ester linkage between FPBA and GlcAm residues to increase the tolerability against DS exchange reaction.

2.3 Responsivity of FPBA/GlcAm crosslinking against ATP and pH ATP-responsivity of PMs was evaluated by gel electrophoresis under various ATP concentrations at pH 7.4. Note that DS was not added in this experimental condition. Thus, the PMs formed through cumulative electrostatic interaction between the charged residues in pDNA and block catiomer can tolerate the electric field applied for electrophoresis. Regardless of ATP concentration, there was no observable migration of pDNA in the electrophoregram for the PM prepared from PEG-b-PAsp(DET) (Figure 2a). Instead, pDNA release from B59/G27 PM clearly occurred depending on ATP concentration. At ATP concentration under 0.5 mM, B59/G27 PM revealed no release of the loaded pDNA, while evident release of pDNA was observed in the range over 1 mM ATP (Figure 2c). ATP-responsivity was further investigated for B26/G14 PM with less introduction degree of FPBA and GlcAm units, expecting a reduced sensitivity to ATP concentration. Indeed, the B26/G14 PM released the loaded pDNA at ATP concentrations above 3 mM (Figure 2b), which was higher than that for B59/G27 PM (1 mM). A consistent trend was observed in the DLS measurement,

ACS Paragon Plus Environment

Journal of the American Chemical Society 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

as both B59/G27 and B26/G14 PMs increased their size and PDI values with ATP concentration, with the former being more sensitive to ATP concentration than the latter. Thus, B59/G27 PM increased the size and PDI values in the 1 mM ATP range, while B26/G14 PM increased those values in ATP concentrations over 3 mM (Table S1). These results of increased size and PDI of the PMs with ATP addition were consistent to the decondensation of pDNA in the PM core due to the cleavage of boronate crosslinking induced by ATP. Importantly, the ATP responsivity of the B59/G27 PM is relevant for their application to selective intracellular release of pDNA, since there is a substantial change in the ATP concentration from ~0.4 mM in the extracellular environment to ~3 mM at the intracellular condition.29-31 Thus, these results indicate the potential of the crosslinked PMs for selective responsivity against the elevation of ATP concentration within cells. As aforementioned, introducing FPBA residues into the side chain of PAsp(DET) segment concomitantly reduces the number of positive charges per PAsp(DET) strand, decreasing the local charge density in the core and impairing the formation of ion pair repeats with phosphates in pDNA strand. Accordingly, it is safe to assume that the ATP-responsive FPBA/GlcAm crosslinking may play a major role in maintaining the integrity of PMs, although initial PM formation is driven through electrostatic attraction between the block catiomer and pDNA. Next, the pH-sensitivity of the PMs was compared at pH 7.4 and 5.5, mimicking the extracellular and late endosomal pH, respectively. Previous studies have demonstrated that some portion of PEG-b-PAsp(DET) was released from PM at pH 5.5 to compensate electrostatic repulsion in the PM core caused by the promoted protonation of 1,2-diaminoethane units in PAsp(DET) segment (53% at pH 7.4 and 85% at pH 5.5).9, 12 Thus, it is reasonable to expect that the release of block catiomers from B59/G27 PM occurs if the part of FPBA/GlcAm crosslinks in the PM core may cleave under acidic environment due to the shift in the equilibrium of FPBA (pKa = 7.2) from charged tetravalent (-B--(OH)3) form to uncharged trivalent (-B-(OH)2) form. Indeed, by ultracentrifugation, it was demonstrated that certain portions of both PEG-b-PAsp(DET/FPBA59%)

ACS Paragon Plus Environment

Page 8 of 28

Page 9 of 28 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

Journal of the American Chemical Society

and PEG-b-PAsp(DET/GlcAm27%) were released from B59/G27 PM by decreasing pH from 7.4 to 5.5 (Table 2). It should be noted that the PMs appeared to keep the structure even after the release of a portion of block catiomers because the size and PDI values for B59/G27 PM at both pH 7.4 and pH 5.5 were comparable (Table S2). The released block catiomers from PMs in the endosomes are assumed to facilitate translocation of the PMs into cytoplasm via their integrity-perturbing interaction with endosomal membrane. To address the membrane disrupting capability of PEG-b-PAsp(DET/FPBA59%) and PEG-b-PAsp(DET/GlcAm27%), the permeability of YO-PRO 1, i.e. a fluorescent dye impermeable to the intact cellular membrane, into HuH-7 cells was examined in the presence of these block catiomers in the culture media.9 We observed a significant increase in YO-PRO 1 fluorescence in HuH-7 cells when decreasing the pH from 7.4 to 5.5 for all of the examined block catiomers (PEG-b-PAsp(DET), PEG-b-PAsp(DET/FPBA59%), and PEG-b-PAsp(DET/GlcAm27%)) (Figure 3), indicating that these block catiomers facilitated YO-PRO 1 permeability through their membrane disruptive ability, particularly at late-endosomal pH. PEG-b-PAsp(DET/GlcAm27%) showed lower enhancement in YO-PRO 1 permeability than the original PEG-b-PAsp(DET) at pH 5.5, probably due to the decreased charge density. Interestingly, PEG-b-PAsp(DET/FPBA59%) had comparable enhancement in permeability of YO-PRO 1 to PEG-b-PAsp(DET), despite the lowered positive charge density compared to PEG-b-PAsp(DET) and PEG-b-PAsp(DET/GlcAm27%). This apparent contradictory observation may be accounted by the hydrophobicity of uncharged trivalent FPBA groups at pH 5.5 to facilitate the PEG-b-PAsp(DET/FPBA59%) interaction with cellular membranes. Note that 98% of FPBA residues are estimated to take uncharged hydrophobic form at pH 5.5 from the pKa value. In fact, we have previously reported that PEG-b-PAsp(DET) installed with cholesteryl groups (PEG-b-PAsp(DET)-Cholesteryl) induced higher membrane permeability of YO-PRO 1 compared to the original PEG-b-PAsp(DET) due to the facilitated membrane association through the hydrophobic cholesteryl group.9

ACS Paragon Plus Environment

Journal of the American Chemical Society 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 28

2.4 Effect of FPBA/GlcAm crosslinking in the core on cellular uptake and gene transfection efficiency of PMs The cellular uptake efficiency of PMs was evaluated in HuH-7 cells. B59/G27 PM showed approximately

10-fold

higher

cellular

uptake

than

other

PMs

prepared

solely

from

PEG-b-PAsp(DET), PEG-b-PAsp(DET/FPBA59%), and PEG-b-PAsp(DET/GlcAm27%) (Figure 4a; open bar). Previous studies suggested that anionic glycosaminoglycans (GAGs), such as heparan sulfate (HS) and chondroitin sulfate, existing on the outer surface of cellular membrane, may destabilize the PM structure through their exchange with loaded pDNA as revealed in the model experiments using DS (Figure 1).32 Thus, the cellular uptake of PMs to HuH-7 cells was evaluated after the treatment of Heparanase II, which reduces the amount of HS on the cell membrane. Actually, the pretreatment of HuH-7 cells with Heparanase II somewhat increased the cellular uptake of the PMs prepared from solely PEG-b-PAsp(DET), PEG-b-PAsp(DET/FPBA59%), and PEG-bPAsp(DET/GlcAm27%), while the uptake of B59/G27 PM was unchanged to keep significantly higher efficiency (Figure 4a; closed bar). Thus, it is natural to conclude that the FPBA/GlcAm crosslinking in B59/G27 PM substantially contributes to increase its resistance against the attack from GAGs on the cellular membrane, thereby, achieving significantly higher efficiency of cellular uptake than the PMs without crosslinking. The B59/G27 PM also showed approximately 20-fold higher gene expression than the other PMs (Figure 4b). To get further insights in the mechanisms involved in the enhanced transfection of B59/G27 PM, the transfection efficiency was compared with that of the PEG-b-PAsp(DET) PM after adjusting the uptaken pDNA amount to the same level between the B59/G27 PM and the PEG-b-PAsp(DET) PM, which was attained by reducing the dose of the B59/G27 PM to cells (Figure S2). Interestingly, the B59/G27 PM still achieved significantly higher transfection efficiency than the PEG-b-PAsp(DET) PM (Figure 4c), indicating that factors other than the increased cellular uptake

ACS Paragon Plus Environment

Page 11 of 28 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

Journal of the American Chemical Society

played a role in the promoted gene transfection by B59/G27 PM. Then, to clarify these factors, we focused on the intracellular trafficking of these PMs, as described in the following section.

2.5 Effect of FPBA/GlcAm crosslinking on intracellular trafficking of PMs The endosomal escape of Cy3-labeled pDNA loaded in the PM was investigated in HuH-7 cells under confocal laser scanning microscopy (CLSM). The late endosomes/lysosomes were labeled with LysoTracker Green (Figure 5a-d; green pixels). Alternatively, the red pixels correspond to Cy3-labeled pDNA escaped into cytoplasm from endosomal/lysosomal compartments, and the yellow pixels represent the co-localization of Cy3-labeled pDNA and late endosomes/lysosomes, indicating the Cy3-labeled pDNA still entrapped in late endosomal/lysosomal compartments. Apparently, all the images obtained for PEG-b-PAsp(DET) PM, B59/G27 PM, B26/G14 PM, and PEG-b-PAsp(DET/FPBA59%) PM were observed with clear red spots (Figure 5a-d, respectively), demonstrating the considerable escape of pDNA from late endosomes/lysosomes. The quantification of the co-localization ratio showed that the B59/G27 PM had even higher endosomal escapability of pDNA than PEG-b-PAsp(DET) PM (Figure 5e). As demonstrated in the model experiment of block catiomer release from B59/G27 PM in acidic milieu (Table 2), the lowered pH in the endosomes may facilitate the cleavage of phenylboronate ester linkages as well as the protonation of amino groups in entrapped PM, promoting the release of PEG-b-PAsp(DET/FPBA) with hydrophobic trivalent FPBA moieties from the PM to induce the effective disruption of endosomal membranes. Noteworthy, the B26/G14 PM and the PEG-b-PAsp(DET/FPBA59%) PM also elicited higher endosomal escapability than the PEG-b-PAsp(DET) PM (Figure 5e), which agreed with the contribution of the hydrophobic trivalent FPBA moieties to the effective disruption of endosomal membrane. The results of Figure 5 revealed that some portion of pDNA was indeed translocated into cytoplasm from the endosomal/lysosomal compartment. Thus, the next issue it to prove that the loaded pDNA in the B59/G27 PM is eventually decondensed in response to an elevated

ACS Paragon Plus Environment

Journal of the American Chemical Society 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

concentration of cytoplasmic ATP. To tackle this issue, Cy3/Cy5 double-labeled pDNA was used in the following experiments to measure FRET efficiency. As shown in Figure 6, the loading into PMs induced significant increase in the FRET signal from Cy3/Cy5 double-labeled pDNA compared to the free form at the condition without ATP due to the pDNA condensation in the PM core.33-34 The FRET efficiency significantly decreased with increasing ATP concentration for the B59/G27 PM (Figure 6; closed bar), while such a remarkable decrease was not observed for PEG-b-PAsp(DET) PM (Figure 6; open bar). This result indicates that the pDNA in B59/G27 PM undergoes decondensation in response to ATP, which is consistent with ATP-triggered pDNA release from the PM as indicated in the gel electrophoresis results (Figure 2c). Notably, we observed a significant decrease in the FRET efficiency of the B59/G27 PM in the presence of 3 mM ATP, i.e. intracellular ATP concentration, to a value comparable to that of naked pDNA, suggesting that the unpackaging of pDNA occurred at this ATP concentration. Of note, the B26/G14 PM revealed to have lower ATP-responsivity than B59/G27 PM (Figure 6; hatched bar), which is consistent with the trend observed in the gel electrophoresis (Figure 2b). The intracellular FRET efficiency of B59/G27 and B26/G14 PMs was then monitored in HuH-7 cells by flow cytometry (Figure 7a and b). Note that, as shown in Figure S3 and 5e, both PMs revealed similar cellular uptake efficiency and endosomal escapability into HuH-7 cells, and thus, it may be safe to directly compare the FRET efficiency as the marker for pDNA condensation loaded in these two PMs. After 6 h of incubation, the B59/G27 PM has already revealed lower FRET efficiency than the B26/G14 PM (Figure 7c). After 12 h of incubation, further decrease in FRET efficiency was observed for the B59/G27 PM, while the B26/G14 PM still maintained the high efficiency comparable to that observed at 6 h of incubation. Drop in FRET efficiency became more obvious for the B59/G27 PM at 24 h of incubation (Figure 7a, b). These results demonstrated the facilitated decondensation of pDNA occurred inside of the cells for the B59/G27 PM compared to the B26/G14 PM. The major mechanism of pDNA decondensation and the following release from

ACS Paragon Plus Environment

Page 12 of 28

Page 13 of 28 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

Journal of the American Chemical Society

polyplexes inside of the transfected cell is believed to be an exchange with intracellular polyanionic compounds. The B59/G27 PM has originally higher tolerability against polyion exchange than B26/G14 PM, as clearly seen from the results of the exchange reaction with DS for PMs that have experienced the intracellular pH modulation process, i.e. PMs first incubated at the endo/lysosomal pH of 5.5 and then at the cytoplasmic pH of 7.4 followed by the incubation with DS (Figure S4). The result of the intracellular FRET measurement is not in line with this trend of tolerability measured in the ATP-free buffer, but it is supportive to the mechanism that elevated intracellular ATP concentration facilitates the cleavage of FPBA/GlcAm crosslinking in B59/G27 PM to make it more susceptible against polyanion exchange inside of the cell, as demonstrated in the model experiments shown in Figures 2c and 6.

3. Conclusion We developed FPBA/GlcAm crosslinked PMs with the aim of improving the major problems in gene carriers, i.e. ensuring secure delivery and cellular uptake of pDNA by enhancing the structural stability of PMs in extracellular conditions, improving the efficiency of pDNA translocation from endosomal compartments to cytoplasm by integrating chemical moieties to disrupt the integrity of endosomal membrane upon pH drop, and facilitating the selective decondensation of pDNA inside of the cell for smooth gene transfection. The optimization of the block catiomer composition enables these distinct functionalities of FPBA/GlcAm crosslinked PMs to work synergistically, and perform efficient gene transfection. Furthermore, the modulated stability of the PMs inside of the cell triggered by intracellular ATP was dynamically monitored using FRET measurement. These observations provided insights into the molecular mechanisms of smart PM systems sensing intracellular chemical signals, i.e. change in pH and ATP concentration, in a spatio-temporal manner to ultimately exert enhanced gene transfection compared to non-sensitive PM systems.

ACS Paragon Plus Environment

Journal of the American Chemical Society 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

Supporting Information Materials, Methods, Additional data referenced in the text; DLS measurement in the presence of ATP and at pH 5.5, 1H-NMR spectra, cellular uptake efficiencies of PMs at different dose, cellular uptake efficiencies of B59/G27 PM and B26/G14 PM, tolerability of PMs against DS after the intracellular pH modulation.

Acknowledgements This research was financially supported by the Japan Science and Technology Agency (JST) through the Center of Innovation (COI) Program (Center of Open Innovation Network for Smart Health) and the Precursory Research for Embryonic Science and Technology (PRESTO) (Molecular Technology and Creation of New Functions), and the Japan Society for the Promotion of Science (JSPS) through the Specially Promoted Research Program (grant number 25000006) and the Core-to-Core Program (A. Advanced Research Networks).

References 1. Katayose, S.; Kataoka, K., In Advanced Biomaterials in Biomedical Engineering and Drug Delivery Systems; Ogata, N.; Kim, S. W.; Feijen, J.; Okano, T., Eds. Springer Japan, Tokyo, 1996. 2. Kabanov. A. V.; Kabanov, V. A, Adv. Drug Delivery Rev. 1998, 30, 49-60. 3. Lee, M.; Kim, S. W., Pharm. Res. 2005, 22, 1-10. 4. Park, T. G.; Jeong, J. H.; Kim, S. W., Adv. Drug. Deliver. Rev. 2006, 58, 467-486. 5. Mintzer, M. A.; Simanek, E. E., Chem. Rev. 2009, 109, 259-302. 6. Lachelt, U.; Wagner, E., Chem. Rev. 2015, 115, 11043-11078. 7. Wang, D. A.; Narang, A. S.; Kotb, M.; Gaber, A. O.; Kim, S. W.; Mahato, R. I., Biomacromolecules 2002, 3, 1197-1207.

ACS Paragon Plus Environment

Page 14 of 28

Page 15 of 28 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

Journal of the American Chemical Society

8. Anwer, K.; Barnes, M. N.; Fewell, J.; Lewis, D. H.; Alvarez, R. D., Gene Ther. 2010, 17, 360-369. 9. Chen, Q.; Osada, K.; Ge, Z. S.; Uchida, S.; Tockary, T. A.; Dirisala, A.; Matsui, A.; Toh, K.; Takeda, K. M.; Liu, X. Y.; Nomoto, T.; Ishii, T.; Oba, M.; Matsumoto, Y.; Kataoka, K., Biomaterials 2017, 113, 253-265. 10. Yessine, M. A.; Leroux, J. C., Adv. Drug Deliver. Rev. 2004, 56, 999-1021. 11. Walker, G. F.; Fella, C.; Pelisek, J.; Fahrmeir, J.; Boeckle, S.; Ogris, M.; Wagner, E., Mol. Ther. 2005, 11, 418-425. 12. Miyata, K.; Oba, M.; Nakanishi, M.; Fukushima, S.; Yamasaki, Y.; Koyama, H.; Nishiyama, N.; Kataoka, K., J. Am. Chem. Soc. 2008, 130, 16287-16294. 13. Park, Y.; Kwok, K. Y.; Boukarim, C.; Rice, K. G, Bioconjugate Chem. 2002, 13, 232-239. 14. Miyata, K.; Kakizawa, Y.; Nishiyama, N.; Harada, A.; Yamasaki, Y.; Koyama, H.; Kataoka, K., J. Am. Chem. Soc. 2004, 126, 2355-2361. 15. Lorand, J. P.; Edwards, J. O., J. Org. Chem. 1959, 24, 769-774. 16. Kitano, S.; Kataoka, K.; Koyama, Y.; Okano, T.; Sakurai, Y., Makromol. Chem. Rapid Commun. 1991, 12, 227-233. 17. Singh, N.; Willson, R. C., J. Chromato. A 1999, 840, 205-213. 18. Springsteen, G.; Wang, B. H., Tetrahedron 2002, 58, 5291-5300. 19. Yan, J.; Springsteen, G.; Deeter, S.; Wang, B., Tetrahedron 2004, 60, 11205-11209. 20. Hartley, J. H.; Phillips, M. D.; James, T. D., New J. Chem. 2002, 26, 1228-1237. 21. Brooks, W. L. A.; Sumerlin, B. S., Chem. Rev. 2016, 116, 1375-1397. 22. Naito, M.; Ishii, T.; Matsumoto, A.; Miyata, K.; Miyahara, Y.; Kataoka, K., Angew. Chem. Int. Ed. 2012, 51, 10751-10755. 23. Biswas, S.; Kinbara, K.; Niwa, T.; Taguchi, H.; Ishii, N.; Watanabe, S.; Miyata, K.; Kataoka, K.; Aida, T., Nat. Chem. 2013, 5, 613-620. 24. Kim, J.; Lee, Y. M.; Kim, H.; Park, D.; Kim, W. J., Biomaterials 2016, 75, 102-111.

ACS Paragon Plus Environment

Journal of the American Chemical Society 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

25. Shiino, D.; Murata, Y.; Kataoka, K.; Koyama, Y.; Yokoyama, M.; Okano, T.; Sakurai, Y., Biomaterials 1994, 15, 121-128. 26. Matsumoto, A.; Ishii, T.; Nishida, J.; Matsumoto, H.; Kataoka, K.; Miyahara, Y., Angew. Chem. Int. Ed. 2012, 51, 2124-2128. 27. Kanayama, N.; Fukushima, S.; Nishiyama, N.; Itaka, K.; Jang, W. D.; Miyata, K.; Yamasaki, Y.; Chung, U. I.; Kataoka, K., ChemMedChem 2006, 1, 439-444. 28. Masago, K.; Itaka, K.; Nishiyama, N.; Chung, U. I.; Kataoka, K., Biomaterials 2007, 28, 5169-5175. 29. Traut, T. W., Mol. Cell. Biochem. 1994, 140, 1-22. 30. Leist, M.; Single, B.; Castoldi, A. F.; Kuhnle, S.; Nicotera, P., J. Exp. Med. 1997, 185, 1481-1486. 31. Gorman, M. W.; Feigl, E. O.; Buffington, C. W., Clinical Chem. 2007, 53, 318-325. 32. Ruponen, M.; Ya-Herttuala, S.; Urtti, A. Biochim. Biophys. Acta 1999, 1415, 331−341. 33. Matsumoto, Y.; Itaka, K.; Yamasoba, T.; Kataoka, K., J. Gene Med. 2009, 11, 615-623. 34. Schneider, S.; Lenz, D.; Holzer, M.; Palme, K.; Suss, R., J. Control. Release 2010, 145, 289-296.

ACS Paragon Plus Environment

Page 16 of 28

Page 17 of 28 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

Journal of the American Chemical Society

Schemes

Scheme 1. Equilibria of PBA in aqueous solution in the presence of diol compounds.

ACS Paragon Plus Environment

Journal of the American Chemical Society 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

Scheme 2. Schematic illustration of intracellular trafficking of FPBA/GlcAm-crosslinked PM directing to smooth gene expression via cumulative processes of cellular entry, endosomal escape, and ATP-responsive pDNA release.

ACS Paragon Plus Environment

Page 18 of 28

Page 19 of 28 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

Journal of the American Chemical Society

Scheme 3. Synthetic schemes of PEG-b-PAsp(DET/FPBA) and PEG-b-PAsp(DET/GlcAm) from PEG-b-PAsp(DET).

ACS Paragon Plus Environment

Journal of the American Chemical Society 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

Tables Table 1. Cumulant diameter and polydispersity index (PDI) of PMs determined by dynamic light scattering at 25 °C.

Table 2. Number of binding block catiomers to pDNA at pH 7.4 and 5.5.

ACS Paragon Plus Environment

Page 20 of 28

Page 21 of 28 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

Journal of the American Chemical Society

Figures

Figure 1. Electrophoregrams of PMs in the presence of dextran sulfate (DS). (a-d) Electrophoregrams of PMs prepared from PEG-b-PAsp(DET/FPBAx%) (x = 14, 26, 59, and 89) and PEG-b-PAsp(DET/GlcAmy%) (y = 9, 14, 27, and 54) at A/P = 8. (e) Electrophoregrams of PMs prepared from the combination of PEG-b-PAsp(DET/FPBA59%) with PEG-b-PAsp(DET/GlcAmy%) (y

=

27

and

54) at

A/P =

9.

(f) Electrophoregrams

of

PEG-b-PAsp(DET)

PM,

PEG-b-PAsp(DET/FPBA59%) PM, PEG-b-PAsp(DET/GlcAm27%) PM, and B59/G27 PM at A/P = 9.

ACS Paragon Plus Environment

Journal of the American Chemical Society 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

Figure 2. Electrophoregrams of (a) PEG-b-PAsp(DET) PM, (b) B26/G14 PM, and (c) B59/G27 PM after incubated with different concentration of ATP.

ACS Paragon Plus Environment

Page 22 of 28

Page 23 of 28 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

Journal of the American Chemical Society

Figure 3. Membrane disruptive ability of PEG-b-PAsp(DET), PEG-b-PAsp(DET/FPBA59%) and PEG-b-PAsp(DET/GlcAm27%) at pH 7.4 (open bar) and pH 5.5 (closed bar). The residual positive charge concentrations ([N+]PEG-b-PAsp(DET), [N+]PEG-b-PAsp(DET/GlcAm), and [N+-B-]PEG-b-PAsp(DET/FPBA)) were adjusted to be same for all the samples at pH 7.4. (mean ± SEM, n = 6, **: p < 0.01).

ACS Paragon Plus Environment

Journal of the American Chemical Society 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

Figure 4. Effect of FPBA/GlcAm crosslinking in the polyplex core on the uptake and the gene transfection efficiency against HuH-7 cells. (a) Uptake efficiency of PMs into untreated- (open bar) or Heparanase II-treated (closed bar) HuH-7 cells. (means ± SEM, n = 4, *: p < 0.05, **: p < 0.01). (b) Transfection efficiency of PMs loaded with pDNA encoding luciferase against HuH-7 cells after 48 h of incubation. (c) Transfection efficiency of PEG-b-PAsp(DET) PM and B59/G27 PM loaded with pDNA encoding luciferase against HuH-7 cells in the condition with adjusting uptaken pDNA amount to the same level after 48 h of incubation. (means ± SEM, n = 4, *: p < 0.05, **: p < 0.01).

ACS Paragon Plus Environment

Page 24 of 28

Page 25 of 28 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

Journal of the American Chemical Society

Figure 5. CLSM observation of PMs internalized into HuH-7 cells. Intracellular distribution of Cy3-labeled pDNA (red) loaded in (a) PEG-b-PAsp(DET) PM, (b) B59/G27 PM, (c) B26/G14 PM, and (d) PEG-b-PAsp(DET/FPBA59%) PM after 6 h of transfection. Late endosomes/lysosomes were stained with LysoTracker Green (green). Nucleus was stained with Hoechst 33342 (blue). (e) Co-localization ratio of Cy3-labeled pDNA with late endosomes/lysosomes. (mean ± SEM, n = 50, *: p < 0.05)

ACS Paragon Plus Environment

Journal of the American Chemical Society 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

Figure 6. FRET efficiency from Cy3/Cy5 double-labeled pDNA loaded in PEG-b-PAsp(DET) PM (open bar), B26/G14 PM (hatched bar), and B59/G27 PM (closed bar) after incubated with different concentration of ATP. Mean FRET efficiency of Cy3/Cy5 double-labeled pDNA in free form is indicated as a solid line. (mean ± SEM, n = 3).

ACS Paragon Plus Environment

Page 26 of 28

Page 27 of 28 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

Journal of the American Chemical Society

Figure 7. Estimation of the PM dissociation by measuring intracellular FRET efficiency from Cy3/Cy5 double-labeled pDNA loaded in PMs. Histograms of intracellular FRET signals of Cy3/Cy5 double-labeled pDNA loaded in (a) B59/G27 PM and (b) B26/G14 PM after 6, 12, and 24 h of transfection. (c) Intracellular FRET efficiency of Cy3/Cy5 double-labeled pDNA loaded in B59/G27 PM (closed circle) and B26/G14 PM (open circle) evaluating by flow cytometer analysis after 6, 12, and 24 h of transfection. (mean ± SEM, n = 3, *: p < 0.05, **: p < 0.01).

ACS Paragon Plus Environment

Journal of the American Chemical Society 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

ACS Paragon Plus Environment

Page 28 of 28