Synthesis of hyaluronan-amikacin conjugate and its bactericidal activity against intracellular bacteria in vitro and in vivo

Article abstract of DOI:10.1016/j.carbpol.2017.10.061

Aminoglycosides are often subtherapeutic to intracellular infections due to their high hydrophilicity. Here we used hyaluronic acid (HA) as a carbohydrate carrier of the aminoglycoside antibiotic, amikacin (AM) to deal with intracellular bacterial infections. The hyaluronan-amikacin conjugate (HA-AM) was synthesized by ‘click’ reaction between HA-propargyl amide (HAPA) and AM-azide. This conjugate with little cytotoxicity retained antibiotic effects on planktonic bacteria and showed better intracellular bactericidal activity than the antibiotic did. In addition, this conjugate was more efficient in reducing bacteria burden in an in vivo acute infection model than amikacin did. These results suggested that hyaluronic acid conjugation could reduce the dosage of antibiotic in treatment of intracellular bacterial infection, and further helping to alleviate the emergence of drug resistance in bacteria due to a long-term, high-dosage treatment.

Full text of DOI:10.1016/j.carbpol.2017.10.061

Accepted Manuscript
Title: Synthesis of hyaluronan-amikacin conjugate and its
bactericidal activity against intracellular bacteria in vitro and
in vivo
Authors: Zhaojie Wang, Yuanhao Qiu, Chunyan Hou,
Dongdong Wang, Feifei Sun, Xiaojun Li, Feng Wang, Hong
Yi, Haibo Mu, Jinyou Duan
PII:
S0144-8617(17)31217-1
DOI:
Reference:
https://doi.org/10.1016/j.carbpol.2017.10.061
CARP 12910
To appear in:
Received date:
Revised date:
Accepted date:
11-8-2017
5-10-2017
17-10-2017
Please cite this article as: Wang, Zhaojie., Qiu, Yuanhao., Hou, Chunyan., Wang,
Dongdong., Sun, Feifei., Li, Xiaojun., Wang, Feng., Yi, Hong., Mu, Haibo., &
Duan, Jinyou., Synthesis of hyaluronan-amikacin conjugate and its bactericidal
activity against intracellular bacteria in vitro and in vivo.Carbohydrate Polymers
https://doi.org/10.1016/j.carbpol.2017.10.061
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Manuscript: CARBPOL-D-17-02734
Title
Synthesis of hyaluronan-amikacin conjugate and its bactericidal activity against
intracellular bacteria in vitro and in vivo
Zhaojie Wang, Yuanhao Qiu, Chunyan Hou, Dongdong Wang, Feifei Sun, Xiaojun Li,
Feng Wang, Hong Yi, Haibo Mu∗ and Jinyou Duan∗
Shaanxi Key Laboratory of Natural Products & Chemical Biology，College of
Chemistry & Pharmacy, Northwest A&F University, Yangling 712100, Shaanxi,
China
∗ Correspondence:
Haibo Mu, PhD
Tel.: +86-29-87092226;
E-mail: mhb1025@nwsuaf.edu.cn
Or
Jinyou Duan, PhD
Tel.: +86 29-87092226;
E-mail: jduan@nwsuaf.edu.cn
Highlights



HA-AM was obtained through reaction of HA with amikacin via “click” chemistry.
HA-AM was more effective against intracellular bacteria than amikacin did.
HA-AM had undetectable cytotoxicity.

Abstract
Aminoglycosides are often subtherapeutic to intracellular infections due to their
high hydrophilicity. Here we used hyaluronic acid (HA) as a carbohydrate carrier of
the aminoglycoside antibiotic, amikacin (AM) to deal with intracellular bacterial
infections. The hyaluronan-amikacin conjugate (HA-AM) was synthesized by ‘click’
reaction between HA-propargyl amide (HAPA) and AM-azide. This conjugate with
little cytotoxicity retained antibiotic effects on planktonic bacteria and showed better
intracellular bactericidal activity than the antibiotic did. In addition, this conjugate
was more efficient in reducing bacteria burden in an in vivo acute infection model
than amikacin did. These results suggested that hyaluronic acid conjugation could
reduce the dosage of antibiotic in treatment of intracellular bacterial infection, and
further helping to alleviate the emergence of drug resistance in bacteria due to a
long-term, high-dosage treatment.
Keywords: Hyaluronic acid; conjugate; Amikacin; Click chemistry; Intracellular
bacteria.
1. Introduction
The intracellular bacteria related infections are always hard to be cured because
these intracellular bacteria (e.g. L. monocytogenes, P. aeruginosa, S. aureus and M.
tuberculosis) can exploit a huge range of niches within their hosts, protect them from
some immune effectors such as antibodies, further gain access to and proliferate
within the host cell cytosol, making the infection latent or recurrent (Ray, Marteyn,
Sansonetti, & Tang, 2009; Joller et al., 2010; Leber et al., 2008).

For now, the treatment of infections caused by intracellular bacteria still presents
a number of challenges, because their intercellular lifestyle protect them not only
from the attack of the immune system but also from the action of the antibiotics
(Imbuluzqueta, Gamazo, Ariza, & Blanco-Prieto, 2010). There are more than two
thirds of prescribed antibiotics are ineffective against intracellular pathogens (Butler
& Cooper, 2011). One of the major reasons is the limited intracellular active
concentrations of antibiotics are often subtherapeutic, resulting in low effectiveness
against intracellular pathogens and the emergence of antibiotic resistance (Baharoglu,
Krin, & Mazel, 2013). To improve the therapeutic effect, high doses of antibiotics are
often given, generating many side effects and toxicity. Aminoglycoside antibiotics, for
instance, are poorly effective in treating intracellular pathogens in vitro, due to their
high hydrophilicity and poor penetration into the eukaryotic cell membrane (Maurin
& Raoult, 2001). But high-does or longtime aminoglycosides treatment are known to
cause ototoxicity (Francis et al., 2013) and nephrotoxicity (Mingeot-Leclercq &
Tulkens, 1999).
Hyaluronic acid (HA), is a nonsulfated glycosaminoglycan which consisted of
alternating residues of β-4 linked D-glucuronic acid and β-3 linked
N-acetyl-D-glucosamine (W. Zhang et al., 2014; Orellana et al., 2016). As a
component of extracellular matrix, HA is readily chemically modified and adapted for
medical uses (Dřímalová et al., 2005; Guo, Xie et al., 2014; Naderi-Meshkin et al.,
2014), which shows various applications in tissue engineering or drug delivery (Borke,
Winnik, Tenhu, & Hietala, 2015; Dosio, Arpicco, Stella, & Fattal, 2016; Thibeault et

al., 2011). Several chemical methods that use usually involving the activation of the
carboxylate
groups
with
reagents,
such
as
N-(3-dimethylaminopropyl)-N'-ethy-lcarbodiimide
(EDC)
and
1-hydroxypyrrolidine-2,5-dione (NHS) or 2-chloro-4,6-dimethoxy-1,3,5-triazine
(CDMT), have been devised to facilitate the amidation of the HA carboxylate groups
(Borke, Winnik, Tenhu & Hietala, 2015; Bulpitt & Aeschlimann, 1999;
Huerta-Angeles et al., 2011; Kuo, Swann & Prestwich, 1990; Nakajima & Ikada,
1995; Schanté et al., 2011). 1-[Bis (dimethylamino) methylene]-1H-1,2,3-triazolo
[4,5-b] pyridinium-3-oxid-hexafluorophosphate (HATU) is one of condensing
reagents for solid-phase peptide synthesis (SPPS) that has been used in the reaction of
carboxyl and amino peptide bonds (Saito & Wada, 2014).
Nowadays, it is possible to encapsulate, incorporate or even conjugate
biologically active molecules into liposomes in order to deliver antibiotics
intracellularly and hence to treat infections (Imbuluzqueta et al., 2010). However,
liposomal antibiotics also display some drawbacks, such as instability of the vesicles,
low drug entrapment and difficulty in sterilization procedures for liposomal
antibiotics (Drulis-Kawa & Dorotkiewicz-Jach, 2010). Previously we demonstrated
that a conjugate between HA and streptomycin through reductive amination could
resolve the conundrum above (H. Y. Qiu et al., 2017). We asked whether this strategy
was also suitable to amikacin, belonging to the third generation of aminoglycoside
antibiotic family. The advantage of polysaccharide-antibiotic conjugate is that it
possesses definite molecular structure and was ready to be prepared and sterilized. In

the present study, an HA-AM conjugates was synthesized by ‘click’ reaction between
HAPA and AM-azide. The intracellular killing capacity of this conjugate was explored
in vitro and in vivo (Fig. 1).
2. Results and discussion
2.1. Synthesis and characterization of HAPA derivatives
Several methods have been reported for the preparation of alkyne modified HA,
but none of them was a chemical reaction in an easy way with HATU-mediated
amidation. We set out first to confirm that, in our hands, the HATU-mediated
amidation proceeded well under the conditions reported previously using the same
amines ranging in the same pKa=6.5~7.0 (Borke, Winnik, Tenhu & Hietala, 2015).
HATU in the presence of DMSO could be used to activate carboxyl functional groups
in polysaccharides. In order to increase the solubility of polysaccharides in DMSO,
hyaluronic acid sodium salt was exchanged for HA by cation exchange resin and the
triethylamine was added to the mixture. Using this methodology, the amine bearing
alkynyl terminal groups were attached to HA quickly and easily (Scheme 1). After
reaction, the HAPA derivatives were purified, isolated.
The FT-IR analysis confirmed that the chemical reaction between the units and
the amine by the presence of two distinct bands at 1645 cm−1 attributed to (C=O)
stretching band and (–NH) deformation band, respectively. A small band at 2156 cm−1
revealed the presence of a terminal alkynyl group (Fig. 2A).
1
The degree of substitution (DS) values were resolved by H NMR spectroscopy in
1
mixed solution of DOH/DMSO-d5 (v/v=1:1). The H NMR spectroscopy showed

typical HA proton signals at 2.0 ppm belonging to –N–COCH3 group, skeletal signals
at 3.4~3.9 ppm and anomeric resonances at 4.4~4.6 ppm. The signal peaks of the
propargyl protons appeared at 4.0~4.1 ppm (-NHCH2C≡CH) and 2.8~3.0 ppm
(–NHCH2C≡CH), respectively (Fig. 3A). The DS of HAPA derivatives was
determined by1H NMR spectroscopy as the integral ratio of the protons of methylene
in amide, which was introduced relative to the three N-acetyl protons of HA.
2.2. Synthesis and characterization of azide-modified amikacin (AM-N3)
One group has prepared and assayed 6'-N-metheyl and 6'-N-ethyl derivatives of
amikacin. They showed that these derivatives were hardly affected by the 6'-N-acetyl
transferase which has been shown to inactivate amikacin (Umezawa, H., Iinuma, K.,
Kondo, S., & Maeda, K., 1975). The primary alcohol in the 6’’ position on these
molecules was accessible to modification and was substituted for a variety of
hydrogen bond donors and acceptors of different sizes. Analogues with guanidinium
groups replacing the 6’’-hydroxy group have been shown to display increased A-site
affinity and, in some cases, superior antibacterial activity. Beyond imparting greater
affinity for the A-site, some modifications could potentially disrupt recognition by
aminoglycoside-modifying enzymes, the most common mechanism of
aminoglycoside deactivation (Disney, M. D., & Childs‐Disney, J. L. 2007). Such
derivatives could exhibit greater antibacterial potency against resistant bacteria (Fair
et al., 2014).
In order to site-specifically immobilize the aminoglycosides onto a HAPA, the
azide-modified aminoglycoside was synthesized. The parent aminoglycoside was

converted into three key intermediates (Fair et al., 2014). Firstly, four amines of AM
were globally tert-butyloxycarbonyl (Boc)-protected using excess amount of
di-tert-butyl dicarbonate (Tsitovich et al., 2010). The single primary alcohol of (Boc)4
amikacin was then selectively converted into a sterically demanding sulfonate of
active intermediate (Fair et al., 2012; Childs-Disney, Pushechnikov, Aminova, &
Disney, 2007). Syntheses of AM derivatives were completed according to
modification of previously published procedures in which a primary hydroxy group
was activated by reaction with 2,4,6-triisopropylbenzenesulfonyl chloride in dry
pyridine (Tsitovich et al., 2010). The resulting adduct was displaced with sodium
azide (Mccoy. 2014; Tsitovich et al., 2010). The AM-N3 were synthesized in above
way. Acidic deprotection of all Boc groups using a one to one mixture of
trifluoroacetic acid (TFA) in dichloromethane, followed by column chromatography
purification (Fair et al., 2014; Tsitovich et al., 2010). Fig.2B showed the FTIR spectra
of amicacin (a), 6''-O-TIPBS-tetra-N-Boc-amikacin (b), azide-modified amikacin (c).
As shown in Fig.2B curve (c), a new adsorption peak at 2114 cm–1 was assigned to
the stretching vibration of azide groups, which indicated that AM chains were
successfully functionalized by azide groups. Spectra in Fig.2B curve (b) was
normalized with respect to the benzenesulfonyl stretching peak at 2924 cm-1. The
peaks at 1645 cm-1 and 1046 cm-1 in Fig.2B curve (a), (b), (c) corresponded to the
C-H stretching vibration of AM chains.
2.3. Synthesis and characterization of HA-AM
The azido-aminoglycosides and polysaccharide derivatives were arrayed onto

alkyne-agarose microarrays in the presence of copper (I) (Crescenzi et al., 2007). The
water-soluble polysaccharide derivatives bearing side chains endowed with either
azide or alkyne terminal functionality was mixed together in aqueous solution. When
the latter two types of derivatives are mixed together in aqueous solution they give
rise to a 1,3-dipolar cycloaddition reaction resulting in fast gelation (in the presence of
catalytic amounts of Cu(I)) at room temperature. “Click chemistry” using azides and
acetylenes is most effective when performed in water or the mixture of water and
organic co-solvents （e.g. tret-butanol, ethanol, DMSO, THF and acetonitrile）
(Gruškienė, Čiuta & Makuška, 2009; Parrish et al., 2005). AM-N3 grafted HAPA in a
mixed solution of water and DMSO, under the conditions of copper sulfate and
sodium ascorbate, via a click reaction. The presence of a weak singlet assigned to the
proton of the triazole ring at δ 7.9~8.0 ppm (Gruškienė, Čiuta & Makuška, 2009).
1
Formation of HA-AM copolymers were confirmed by H NMR spectra. The
propargyl peaks at 2.8~3.0 ppm disappeared, while the signal originating from the
newly formed triazole ring at 7.9~8.0 ppm was indicated (Fig. 3B). The overt signal
protons of glycosidic ring groups of AM at 3.1~4.2 ppm appeared in the spectra and
cross overlapped with those signals of the protons of skeletal signals of the HA
backbone. The remaining typical proton signals of AM were prevailed at 5.4 ppm
(-OH), 5.0 ppm (-CH2-), 2.1 ppm (-CH2-), 1.8 ppm (-CH2-).
The DS of HA-AM conjugate (containing 13% amikacin) was determined by 1H
NMR spectroscopy as the integral ratio of the protons of the -CH- of the triazole ring,
which was introduced relative to the two anomeric protons of HA ( Fig. 3B signing 1

and 4).
2.4. Determination of molecular weight results by HPGPC
As shown in Fig. 4, both HA and HA-AM were eluted as a symmetrical narrow
peak on HPGPC, indicating a high purity. The molecular weight was estimated to be
149.450 kDa, in reference to standard dextrans.
2.5. HA-AM conjugate eliminated intracellular bacteria effectively
Viability assay indicated that HA-AM (containing 13% amikacin) almost had no
cytotoxicity on RAW 264.7 cells at tested concentration (Fig. S2). The cell survival
rate (CSR) was above 80% even HA-AM up to the dosage of 300 μg/mL.
Macrophages function at the front line of immune defences against incoming
pathogens, and therefore, are a common target for those bacterial pathogens that
benefit from avoiding an encounter with the immune system, as well as those that are
aiming to secure systemic spread (Parsek & Singh, 2003). After RAW264.7
macrophages were infected with P. aeruginosa, an opportunistic pathogen that could
survive inside macrophages (Miao, Ernst, Dors, Mao, & Aderem, 2008), amikacina
lone had a mild effect on bacterial counts after 12 h treatment compared to blank
control (Fig. 5A). A combination of amikacin and hyaluronan did not improve
reduction of intracellular bacteria counts dramatically. However, the HA-AM
(containing 13% amikacin) conjugate (100 μg/mL) could kill more P. aeruginosa in
macrophages, although hyaluronan had no bacterial activities at all (Fig. 5A).To see
whether the HA-AM conjugate was able to clear other intracellular bacteria in
macrophages, two other intracellular organisms, S. aureus (a common cause of skin

infections and respiratory disease) (Naimi et al., 2003) and L. monocytogenes (the
causal organism of the serious foodborne illness listeriosis) (Freitag, Port, & Miner,
2009) were tested. As expected, the HA-AM conjugate (100 μg/mL) had stronger
bactericidal activities towards S. aureus (Fig. 5B) and L. monocytogenes (Fig. 5C)
intracellular than the respective mixture with equivalent amount of hyaluronan and
amikacin did. Taken together, these results clearly indicated that the HA-AM
conjugate had a potential to eliminate multiple intracellular bacteriain macrophages.
2.6. HA-AM conjugate promoted intracellular bacteria clearance in vivo
In order to evaluate bactericidal efficiency of HA-AM (containing 13% amikacin)
conjugate against bacteria in vivo, mice were infected intraperitoneally with log-phase
L. monocytogenes (105 CFU/mouse). 24 h later, mice received a daily subcutaneous
injection of HA-AM (10 mg/kg), amikacin alone (AM) or the mixture with equivalent
amount of HA and amikacin (HA+AM), or PBS for three days. Similar to in vitro
findings, there was an improvement in the reduction of bacteria burden in the spleen
and kidney for HA-AM conjugate, although amikacin alone or in combination with
HA reduced the viable bacteria counts in some extent (Fig. 6).
2.7. Discussion
In an earlier study we demonstrated that covalent-coupling chitosan to
streptomycin significantly improved intracellular bactericidal capacity of
streptomycin towards multiple organisms within phagocytic or nonphagocytic cells
(Mu, Niu, et al., 2016). However, chitosan-streptomycin conjugate showed an
undesirable cytotoxicity to RAW264.7 cells, maybe because of the poly cationic

features of chitosan. Furthermore chitosan has been reported exhibited cytotoxicity at
high concentrations (>500 μg/mL) (Huang, Khor, & Lim, 2004). Recently we
synthesized a conjugate between HA and streptomycin through reductive amination
from our group (Qiu et al., 2017). We asked whether this strategy was suitable for
other antibiotic in the aminoglycoside family.
The HA-AM conjugate was synthesized by ‘click’ reaction between
HA-propargyl amide and AM-azide. The HATU was used as coupling agent to carry
out the amidation of HA in DMSO under mild conditions with excellent yields
(59.3%). Compared to other coupling agents, such as EDC/NHS (Crescenzi, Cornelio,
Di Meo, Nardecchia, & Lamanna, 2007) or CDMT/NMM (Bergman, Elvingson,
Hilborn, Svensk, & Bowden, 2007) with yeild of 31% and 50%, this method
described here was economical and convenient. This reaction condition meets the
modification requirements for HA.
HA could be quickly transported into the cell by carriers (e.g. CD44), which
exists on the cell surface of epithelial cells or certain types of hematopoietic cells
(Mackay et al., 1994). CD44 receptors were over-expressed on certain cellular
populations such as activated inflammatory cells (Vafaei, Esmaeili, Amini, Atyabi,
Ostad & Dinarvand, 2016). Our earlier studies also discovered over-expressed CD44
receptor in infected RAW 264.7 (Yin et al., 2008; Zhang et al., 2014), thus HA-AM
can be delivered into these infected cells by binding to CD44 receptor with high
efficiency and be used for eradication of intracellular bacteria (Fig. 7). Compared to
amikacin alone, HA-AM could reduce bacteria counts in infected RAW264.7 cells

(Fig. 5) or infected mice (Fig. 6), although HA-AM didn’t show better activities
against planktonic bacteria (Table S1). On the other hand, this indicated that HA
conjugation did not reduce the antibiotic activity of amikacin.
3. Experimental
3.1. Materials
Hyaluronic acid sodium salt (Mw = 129 kDa) was obtained form Aladin.
Amikacin disulfate (AM, 99%), Propargyl amine (PA), dichloromethane (DCM),
1-[Bis(dimethylamino)me-thylene]-1H-1,2,3-triazolo[4,5-b]-pyridinium-3-oxid-hexaf
luorophosphate(HATU,
99%),
di-tert-butyl
dicarbonate (BOC, 99.5%),
2,4,6-triisopropylbenzenesulfonyl chloride (TIPBS-Cl, 99%), sodium ascorbate (BC),
trifluoroacetic acid (TFA, 99.8%), pyridine (99%), Fluorescein5(6)-isothiocyanate
(FITC) were purchased from Aladdin and used as received. Sodium azide, methanol
(MeOH), triethylamine (TEA, 99.5%), N, N-dimethylformamide (DMF) were
purchased by Xiya (Shangdong, China). 5-(4,6-Dichlorotriazinyl)aminofluorescein
(5-DTAF) was buy from AAT Bioquest (USA). Water was obtained from a Millipore
Milli-Q water purification system. Dialysis was conducted in CelluSep T2 regenerated
cellulose tubular membranes (Membrane Filtration Products, U.S.) with a molecular
1
weight cut off (MWCO) of 5,000 Da. Samples for HNMR spectroscopy were
prepared in deuterium oxide (D2O, 99.96% D) and dimethyl sulfoxide-d6 (DMSO-d6,
99.80% D) obtained from Aladdin. DOWEX 50WX8 cation resin exchange (H type)
was prepared. High performance gel permeation chromatography (HPGPC) was used
1
on a Waters PL-GPC50. H was carried out at room temperature on a BRUKER

Avance III 500 MHz (Switzerland). The Fluorescence Microscopy was used on
OLYMPUS BX53-DP72 (JAPAN). Other chemicals were reagent grade and used
without further purification unless otherwise noted.
Listeria monocytogenes (ATCC 19114), Staphylococcus aureus (ATCC 29213),
Pseudomonas aeruginosa (PAO1) were generous gifts received from Prof. Xia
(College of Food Science and Engineering, Northwest A&F University). Female
Kunming mice (weighing 28± 2 g) were purchased from School of medical
laboratory animal science, The Fourth Military Medical University, China.
3.2. Preparation of HA form
As described (Huerta-Angeles et al., 2016), 1g hyaluronic acid sodium salt was
dissolved in 100 mL distilled water. After filtrated through DOWEX 50WX8 cation
resin exchange (H type)，the resulting solution was concentrated by rotary evaporator
and lyophilized. The molecular weights of the polymer after the cationic exchange
were determined by HPGPC. The obtained HA acid form was further used for the
production of the amide derivate.
3.3. Synthesis of HAPA derivatives
The HAPA was synthesized as previous described (Huerta-Angeles et al., 2016).
Briefly, HA acid form (200mg, 0.04mmol) was dissolved in 12mL of dry DMSO at
60 ℃. After dissolution of the polymer, the solution was cooled to room temperature.
Then, 0.611 mL of TEA (3.3 mmol) was added to the reaction solution and allowed to
be stirred for 15 min. At this moment 36mg of HATU (0.09mmol) was added. The
mixture was stirred for 30 min at room temperature under nitrogen atmosphere (N2).

After that, the propargyl amine (10 uL, 0.09mmol) was added to the reaction and
stirred at room temperature for 24h under N2. The products were dialyzed (Mw cut off
= 5000Da) against 0.1 M of sodium chloride for 24 h, and then dialyzed against water
for five days. The samples were freeze-dried and named as HAPA. The DS was
determined by 1H NMR as the integral ratio of the enantiomers protons of methylene
α and β in amide which was introduced relative to the three N-acetyl protons of HA.
FT-IR spectra were recorded using 500 cm-1 to 4000 cm-1 collecting 32 scans range
with a resolution of 4 cm-1. Samples were studied as KBr pellets.
3.4. Synthesis of azide-modified amikacin (AM-N3)
As described (Childs-Disney, Pushechnikov, Aminova, & Disney, 2007; Disney
& Barrett, 2007; Fair et al., 2014; Tsitovich et al., 2010; Michael, Wang, & Tor, 1999),
all amines of aminoglycoside antibiotics were globally tert-butyloxycarbonyl
(Boc)-protected using di-tert-butyl dicarbonate. The primary alcohol in the 6'' position
was modified by sulfonate of active intermediate. Then, the 6'' position was displaced
by azide group. Finally, the protecting group BOC was removed in trifluoroacetate.
3.5. Synthesis of Hyaluronan acid-amikacin conjugate (HA-AM) by “click
chemistry”
In a typical procedure, HAPA 0.5mg, 0.017mmol (12.5μmol of propargyl groups)
and AM-N3 (10 mg, 17.7 μmol, 1.1eq) were dissolved in 0.5 mL DMSO and 1.0 mL
H2O. The solution was bubbled with argon gas for 15min to remove oxygen.
.
Solutions of copper (II) sulfate pentahydrate (CuSO4 5H2O, 0.006 mmol, 0.6 eq) and
sodium ascorbate (0.021 mmol, 1.3 eq) in 0.5 mL water were prepared separately and

also deoxygenated by bubbling with argon gas for 10 min. Subsequently, the copper
and ascorbate solution were added to the first mixture under an inert atmosphere via a
syringe. The mixture was stirred under constant N2 at room temperature for 24h. The
product was purified by dialysis against 0.1 M of sodium chloride for one day, and
then going on to be dialyzed against water for three days and then lyophilized. The
pure product (HA-AM) was obtained in 61% yield.
3.6. Determination of molecular weight results by high performance gel
permeation chromatography (HPGPC)
As previously (Mu, Niu, et al., 2016), the molecular weight was measured using
HPGPC on three columns (Waters Ultrahydrogel 250, 1,000 and 2,000; 30 cm×7.8
mm; 6 μm particles) in series. The columns were calibrated with T-series Dextrans
(5.2, 10, 48.6, 668, 2,000 kDa). Sodium acetate (3 mM) was used as eluant and the
flow rate was kept at 0.5 mL/min. A 50 μL aliquot was injected for each run. The
calibration curve of Log(Mw) vs. elution time (t) is:
Log Mw = -0.1869t + 12.061
Mw = molecular weight.
3-1
t = Retention time.
3.7. Cytotoxic activity of HA-AM
The cytotoxicity test was performed using the MTT method (Mu, Tang, et al.,
2016). The RAW264.7 cells (5×103cells/well) were cultured in DMEM (high glucose)
medium supplemented with 10% fetal bovine serum. Approximately 8000 RAW264.7
(200 μL RPMI-1640 complete medium) was placed in 96-well plates for 8 h. Then

different concentration of the HA-AM (100 μg/mL, 200 μg/mL, 300 μg/mL) were
added and continued for 24 h, the medium was removed. The complete medium
containing 0.5 mg/mL MTT was added for 4 h, and then removed the medium. Then
100 μL DMSO was added to dissolve the blue violet crystal formazan. After 15 min,
the absorbance was measured at 570 nm using a microplate reader. PBS was used as
blank control.
3.8. Anti-intracellular bacteria activities
As previously (Mu, Niu, et al., 2016), the RAW 264.7 cells were plated on
24-well plate for 6h. Then, macrophages were infected at an MOI=2 for 1 h, washed
thrice with PBS, and media with 50 μg/mL gentamicin was added for another 1h.
After washed thrice with PBS, cells were incubated in 1ꢀmL fresh medium
supplemented with compounds for HA-AM, HA+AM, HA and AM for 12 h. Then
cells were lysed with 0.25% Triton X-100 for 30ꢀmin. Serial dilutions were plated on
TSB plates and colonies were counted the next day to determine CFU.
All data were made using GraphPad Prism 5.0. Results were expressed as mean
±SD. Data were analyzed by t-test and P < 0.05 was considered statistically
significant.
3.9. In vivo bacterial infections
Female Kunming mice (weighing 28±2 g) were infected intraperitoneally with
5×105 CFU of log-phase mouse passaged. Next day, mice received subcutaneous
injection of HA-AM, HA+AM, AM or PBS (10 mg/kg of HA-AM, n=5 mice) once
every day for three times. Spleen and kidney were harvested. CFU were determined

by dilution plating as previously described (Mu, Niu, et al., 2016).
3.10. The minimum inhibitory concentration (MIC) determinations for
extracellular bacteria
MIC of HA-AM against bacteria was performed using microdilution assay.
Briefly, two-fold serial dilutions of procedures were prepared in 96-well microplates
in TSB with the final concentration of 25% after P. aeruginosa, S. aureus, L.
monocytogenes (105 CFU/mL) addition. All assays were carried out in triplicate. The
antibacterial activities were examined after incubation at 37°C for 24 h (Bastian,
Warszawik, Panduru, Arenz & Herrmann, 2013; Mu, Zhang, Zhang, Cui, Wang &
Duan, 2012).
3.11. Immunofluorescence
This was measured by fluorescent labels (FITC and 5-DTAF). The FITC bond
amino and hydroxyl groups of HA-AM. The 5-DTAF labeled hydroxyl groups of HA
and HA-AM. As described (Li, G. G. M, Li et al., 2002), Fluorescent labeling of
HA-AM was obtained. RAW264.7 cells were plated on coverslips 12 h before
incubation with fluorescent labeling of HA-AM conjugate (containing 13% amikacin),
equivalent HA alone, PBS for 4 h. Then the coverslips were fixed in 4%
paraformaldehyde for 15ꢀminutes at room temperature . The fluorescent was stained
by the Fluorescence Microscopy (OLYMPUS, Japan). For blocking experiments,
cells were incubated with Anti-CD44 (20 μg/mL, eBioscience, CA) or free hyaluronic
acid (100 μg/mL) for 1 h prior treatment with HA-AM.

4. Conclusion
Our results highlighted that HA conjugation could improve intracellular
bactericidal capacity of amikacin against different pathogens. In vivo acute infection
models indicated it elicited a better therapeutic effect on intracellular infection than
amikacin did. Moreover, HA-AM showed neglectable cytotoxicity on RAW264.7
cells. These might be contribute to reduce antibiotic consumption in intracellular
bacterial infection treatment, and then help to reduce the emergence of antibiotic
resistance due to a long-term, high-dosage treatment.
Acknowledgement
This work was supported by the National Natural Science Foundation of China
(NSFC) (Grant No. 31570799), and Fundamental Research Funds for the Central
Universities, Northwest A&F University (2452017026、Z109021713)

References
Baharoglu, Z., Krin, E., & Mazel, D. (2013). RpoS plays a central role in the SOS induction by sub-lethal
aminoglycoside concentrations in Vibrio cholerae. PLoS Genet, 9(4), e1003421.
Bergman, K., Elvingson, C., Hilborn, J., Svensk, G., & Bowden, T. (2007). Hyaluronic Acid Derivatives
Prepared in Aqueous Media by Triazine-Activated Amidation. Biomacromolecules, 8(7),
2190-2195.
Borke, T., Winnik, F. M., Tenhu, H., & Hietala, S. (2015). Optimized triazine-mediated amidation for
efficient and controlled functionalization of hyaluronic acid. Carbohydrate polymers, 116, 42-50.
Bulpitt, P., & Aeschlimann, D. (1999). New strategy for chemical modification of hyaluronic acid:
preparation of functionalized derivatives and their use in the formation of novel biocompatible
hydrogels. Journal of Biomedical Materials Research Part A, 47(2), 152-169.
Butler, M. S., & Cooper, M. A. (2011). Antibiotics in the clinical pipeline in 2011. The Journal of
antibiotics, 64(6), 413-425.
Childs-Disney, J. L., Wu, M., Pushechnikov, A., Aminova, O., & Disney, M. D. (2007). A small molecule
microarray platform to select RNA internal loop− ligand interactions. ACS chemical biology, 2(11),
745-754.
Crescenzi, V., Cornelio, L., Di Meo, C., Nardecchia, S., & Lamanna, R. (2007). Novel hydrogels via click
chemistry: synthesis and potential biomedical applications. Biomacromolecules, 8(6), 1844-1850.
Disney, M. D., & Childs‐Disney, J. L. (2007). Using Selection to Identify and Chemical Microarray to
Study the RNA Internal Loops Recognized by 6′‐N‐Acylated Kanamycin A. ChemBioChem, 8(6),
649-656.
Dosio, F., Arpicco, S., Stella, B., & Fattal, E. (2016). Hyaluronic acid for anticancer drug and nucleic acid
delivery. Advanced drug delivery reviews, 97, 204-236.
Z, D., & A, D. (2010). Liposomes as delivery systems for antibiotics. International Journal of
Pharmaceutics, 387(1–2), 187-198.
Fair, R. J., Hensler, M. E., Thienphrapa, W., Dam, Q. N., Nizet, V., & Tor, Y. (2012). Selectively
Guanidinylated Aminoglycosides as Antibiotics. ChemMedChem, 7(7), 1237–1244.
Fair, R. J., McCoy, L. S., Hensler, M. E., Aguilar, B., Nizet, V., & Tor, Y. (2014). Singly Modified Amikacin
and Tobramycin Derivatives Show Increased rRNA A‐Site Binding and Higher Potency against
Resistant Bacteria. ChemMedChem, 9(9), 2164-2171.
Francis, S. P., Katz, J., Fanning, K. D., Harris, K. A., Nicholas, B. D., Lacy, M., et al. (2013). A novel role of
cytosolic protein synthesis inhibition in aminoglycoside ototoxicity. Journal of Neuroscience, 33(7),
3079-3093.
Freitag, N. E., Port, G. C., & Miner, M. D. (2009). Listeria monocytogenes—from saprophyte to
intracellular pathogen. Nature Reviews Microbiology, 7(9), 623-628.
Gruškienė, R., Čiuta, G., & Makuška, R. (2009). Grafting of poly (ethylene glycol) to chitosan at c (6)
position of glucosamine units via “click chemistry” reactions. Chemija, 20(4), 241-249.
Guo, J., Xie, Z., Tran, R. T., Xie, D., Jin, D., Bai, X., et al. (2014). Click chemistry plays a dual role in
biodegradable polymer design. Advanced Materials, 26(12), 1906-1911.
Huerta-Angeles, G., Šmejkalová, D., Chládková, D., Ehlová, T., Buffa, R., & Velebný, V. (2011).
Synthesis of highly substituted amide hyaluronan derivatives with tailored degree of substitution
and their crosslinking via click chemistry. Carbohydrate polymers, 84(4), 1293-1300.

Huerta-Angeles, G., Brandejsová, M., Knotková, K., Hermannová, M., Moravcova, M., Šmejkalová, D., &
Velebný, V. (2016). Synthesis of photo-crosslinkable hyaluronan with tailored degree of
substitution suitable for production of water resistant nanofibers. Carbohydrate polymers, 137,
255-263.
Huang, M., Khor, E., & Lim, L.-Y. (2004). Uptake and cytotoxicity of chitosan molecules and
nanoparticles: effects of molecular weight and degree of deacetylation. Pharmaceutical research,
21(2), 344-353.
Imbuluzqueta, E., Gamazo, C., Ariza, J., & Blanco-Prieto, M. J. (2010). Drug delivery systems for
potential treatment of intracellular bacterial infections. Frontiers in Bioscience Journal & Virtual
Library, 15(2):397-417.
Joller, N., Weber, S. S., Müller, A. J., Spörri, R., Selchow, P., Sander, P., et al. (2010). Antibodies protect
against intracellular bacteria by Fc receptor-mediated lysosomal targeting. Proceedings of the
National Academy of Sciences, 107(47), 20441-20446.
Kuo, J. W., Swann, D. A., & Prestwich, G. D. (1990). Chemical modification of hyaluronic acid by
carbodiimides. Bioconjugate chemistry, 2(4), 232-241.
Li Fu-Chuan, G. G. M. e.-Y., Li Ying -Xia, Li Jing , Miao Ben-Chun, Guan Hua-Shi. (2002). Studies on
Fluorescent Labeling of Marine Sulfated Polysaccharide 911. Chemical journal of chinese
universities, 23(9), 1704-1708.
Leber, J. H., Crimmins, G. T., Raghavan, S., Meyer-Morse, N. P., Cox, J. S., & Portnoy, D. A. (2008).
Distinct TLR-and NLR-mediated transcriptional responses to an intracellular pathogen. PLoS
Pathog, 4(1), e6.
Mackay, C. R., & Günthert, U. (1994). Expression and modulation of CD44 variant isoforms in humans.
Journal of Cell Biology, 124(1-2), 71-82.
Maurin, M., & Raoult, D. (2001). Use of aminoglycosides in treatment of infections due to intracellular
bacteria. Antimicrobial agents and chemotherapy, 45(11), 2977-2986.
Miao, E. A., Ernst, R. K., Dors, M., Mao, D. P., & Aderem, A. (2008). Pseudomonas aeruginosa activates
caspase 1 through Ipaf. Proceedings of the National Academy of Sciences, 105(7), 2562-2567.
Mccoy, L. C. S. (2014). Exploring biomolecular interactions: New ligands and novel assays.
Dissertations & Theses - Gradworks.
Mu, H., Tang, J., Liu, Q., Sun, C., Wang, T., & Duan, J. (2016). Potent Antibacterial Nanoparticles
against Biofilm and Intracellular Bacteria. Scientific reports, 6, 18877-18877.
Mu, H., Niu, H., Wang, D., Sun, F., Sun, Y., & Duan, J. (2016). Chitosan conjugation enables intracellular
bacteria susceptible to aminoglycoside antibiotic. Glycobiology, 26(11), 1190-1197.
Naimi, T. S., LeDell, K. H., Como-Sabetti, K., Borchardt, S. M., Boxrud, D. J., Etienne, J., et al. (2003).
Comparison of community-and health care–associated methicillin-resistant Staphylococcus aureus
infection. Jama, 290(22), 2976-2984.
Naderi-Meshkin, H., Andreas, K., Matin, M. M., Sittinger, M., Bidkhori, H. R., Ahmadiankia, N., et al.
(2014). Chitosan-based injectable hydrogel as a promising in situ forming scaffold for cartilage
tissue engineering. Cell Biology International, 38(1), 72-84.
Nakajima, N., & Ikada, Y. (1995). Mechanism of amide formation by carbodiimide for bioconjugation
in aqueous media. Bioconjugate chemistry, 6(1), 123-130.
Orellana, S. L., Giacaman, A., Pavicic, F., Vidal, A., Moreno‐Villoslada, I., et al. (2016). Relevance of
charge balance and hyaluronic acid on alginate‐chitosan sponge microstructure and its influence
on fibroblast growth. Journal of Biomedical Materials Research Part A, 104(10), 2537-2543.

Parrish, B., Breitenkamp, R. B., & Emrick, T. (2005). PEG- and peptide-grafted aliphatic polyesters by
click chemistry. Journal of the American Chemical Society, 127(20), 7404-7410.
Parsek, M. R., & Singh, P. K. (2003). Bacterial biofilms: an emerging link to disease pathogenesis.
Annual Reviews in Microbiology, 57(1), 677-701.
Qiu, Y., Hou, Y., Sun, F., Chen, P., Wang, D., Mu, H., Zhang, X., Ding, K., & Duan, J. (2017). Hyaluronic
acid conjugation facilitates clearance of intracellular bacterial infections by streptomycin with
neglectable nephrotoxicity. Glycobiology, 861-867.
Ray, K., Marteyn, B., Sansonetti, P. J., & Tang, C. M. (2009). Life on the inside: the intracellular lifestyle
of cytosolic bacteria. Nature Reviews Microbiology, 7(5), 333-340.
Saito, K.,
&
Wada, T. (2014). 3-Nitro-l,2,4-triazol-l-yl-tris(pyrrolidin-1-yl) phosphonium
hexafluorophosphate (PyNTP) as a condensing reagent for solid-phase peptide synthesis.
Tetrahedron Letters, 55(12), 1991-1993.
Schanté, C. E., Zuber, G., Herlin, C., & Vandamme, T. F. (2011). Chemical modifications of hyaluronic
acid for the synthesis of derivatives for a broad range of biomedical applications. Carbohydrate
polymers, 85(3), 469-489.
Umezawa, H., IINUMA, K., KONDO, S., & MAEDA, K. (1975). Synthesis and antibacerial activity of
6'-N-alkyl derivatives of 1-N-[(S)-4-amino-2-hydroxybutyryl]-kanamycin. The Journal of antibiotics,
28(6), 483-485.
Thibeault, S. L., Klemuk, S. A., Chen, X., & Johnson, B. H. Q. (2011). In vivo engineering of the vocal fold
ECM with injectable HA hydrogels—late effects on tissue repair and biomechanics in a rabbit
model. Journal of Voice, 25(2), 249-253.
Tsitovich, P. B., Pushechnikov, A., French, J. M., & Disney, M. D. (2010). A Chemoenzymatic Route to
Diversify Aminolgycosides Enables a Microarray‐Based Method to Probe Acetyltransferase Activity.
ChemBioChem, 11(12), 1656-1660.
Vafaei, S. Y., Esmaeili, M., Amini, M., Atyabi, F., Ostad, S. N., & Dinarvand, R. (2016). Self-assembled
hyaluronic acid nanoparticles as a potential carrier for targeting the inflamed intestinal mucosa.
Carbohydrate polymers, 144, 371.
Yin, M. L., Groof, A. J. C. d., Bhattacharjee, M. K., Figurski, D. H., Schon, & A., E. (2008). Bacterial
conjugation in the cytoplasm of mouse cells. Infection & Immunity, 76(11), 5110-5119.
Zhang, A., Mu, H., Zhang, W., Cui, G., Zhu, J., & Duan, J. (2013). Chitosan coupling makes microbial
biofilms susceptible to antibiotics. Scientific reports, 3, 3364.
Zhang, W., Mu, H., Dong, D., Wang, D., Zhang, A., & Duan, J. (2014). Alteration in immune responses
toward N-deacetylation of hyaluronic acid. Glycobiology, 24(12), 1334-1342.

Figure legends:
Fig.1. Schematic of the procedure for Anti-intracellular bacteria.
Fig.2. (A).FT-IR spectra of HA (1) and HA-Propargyl amide Derivatives (2),
(B).FT-IR spectra of Amicacin (a) and 6''-O-TIPBS-tetra-N-Boc-amikacin (b) and
azide-modified amikacin (c).
Fig.3. (A).1H NMR spectra of HAPA and HA, (B).1H NMR spectra of amikacin (AM),
HA and HA-MA conjugate (HA-AM).
Fig.4. HPGPC chromatograms of HA, HAPA, HA-AM conjugate and AM on waters
ultrahydrogel in series.
Fig.5. The colony-forming units of residual P. aeruginosa (A), L.monocytogenes (B)
and S.aureus (C) in infected macrophages treated with different samples.
Fig.6. HA-AM conjugate removed intracellular bacteria efficiently in Kunming mice
(Kunming mice peritoneal were infected with L. monocytogenes and treated with
HA-AM, HA+AM, AM or PBS at a dose of 10 mg/kg. Mice were harvested 120 hpi
for bacterial colony counts. Data points represent each mouse, means ± SD from all
mice. n = 5 mice per group.)
Fig. 7. (A) The conjugation facilitated the entry of amikacin into host cells via a
CD44-mediated pathway. (B) The conjugation promoted amikacin entrance into
macrophages.
Scheme.1. Synthesis of HA-AM conjugate with AM-N3 derivatives and HAPA via
“click chemistry” reactions.

Fig. 1

Fig. 2

Fig. 3

Fig. 4

Fig. 5

Fig. 6

Fig. 7