Synthesis and Evaluation of Doxorubicin-Loaded Gold Nanoparticles for Tumor-Targeted Drug Delivery

Article abstract of DOI:10.1021/acs.bioconjchem.7b00756

Doxorubicin is an effective and widely used cancer chemotherapeutic agent, but its application is greatly compromised by its cumulative dose-dependent side effect of cardiotoxicity. A gold nanoparticle-based drug delivery system has been designed to overc

Full text of DOI:10.1021/acs.bioconjchem.7b00756

Subscriber access provided by RMIT University Library
Article
Synthesis and Evaluation of Doxorubicin-Loaded
Gold Nanoparticles for Tumor-Targeted Drug Delivery
Yongle Du, Long Xia, Ami Jo, Richey M. Davis, Philippe Bissel, Marion Ehrich, and David George Ian Kingston
Bioconjugate Chem., Just Accepted Manuscript • DOI: 10.1021/acs.bioconjchem.7b00756 • Publication Date (Web): 20 Dec 2017
Downloaded from http://pubs.acs.org on December 21, 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.
Bioconjugate Chemistry 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 31
Bioconjugate Chemistry
1
2
3
4
5
Synthesis and Evaluation of Doxorubicin-Loaded Gold
Nanoparticles for Tumor-Targeted Drug Delivery
Yongle Du,† Long Xia,† Ami Jo,‡ Richey M. Davis,‡ Philippe Bissel,§ Marion F. Ehrich,§ and
6
7
8
9
David G. I. Kingston*,†
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
†Department of Chemistry and the Virginia Tech Center for Drug Discovery, Virginia Tech, Blacksburg,
Virginia 24061, United States
‡Department of Chemical Engineering and the Virginia Tech Center for Drug Discovery, Virginia Tech,
Blacksburg, Virginia 24061, United States
§Virginia-Maryland College of Veterinary Medicine and the Virginia Tech Center for Drug Discovery,
Virginia Tech, Blacksburg, Virginia 24061, United States
Supporting Information
ABSTRACT:
Doxorubicin is an effective and widely used cancer chemotherapeutic agent,
but its application is greatly compromised by its cumulative dose-dependent side effect of
cardiotoxicity. A gold nanoparticle-based drug delivery system has been designed to overcome
this limitation. Five novel thiolated doxorubicin analogs were synthesized and their biological
activities evaluated. Two of these analogs and PEG stabilizing ligands were then conjugated to
gold nanoparticles, and the resulting Au-Dox constructs were evaluated. The results show that
release of native drug can be achieved by the action of reducing agents such as glutathione or
under acidic conditions, but that reductive drug release gave the cleanest drug release. Gold
1
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 2 of 31
1
2
3
4
5
6
7
8
9
nanoparticles (Au-Dox) were prepared with different loadings of PEG and doxorubicin, and one
formulation was evaluated for mammalian stability and toxicity. Plasma levels of doxorubicin in
mice treated with Au-Dox were significantly lower than in mice treated with the same amount of
doxorubicin, indicating that the construct is stable under physiological conditions. Treatment of
mice with Au-Dox gave no histopathologically observable differences from mice treated with
saline, while mice treated with an equivalent dose of doxorubicin showed significant
histopathologically observable lesions.
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
 INTRODUCTION
The anthracycline antibiotic doxorubicin (1, Figure 1) was discovered in the late 1960’s
as a minor component of the fermentation broth used to produce its analog daunorubicin (2).1
Daunorubicin itself was discovered in the 1950’s from a strain of Streptomyces peucetius
isolated from a soil sample collected near the Adriatic Sea;2 it had good bioactivity against
mouse tumors, and was eventually marketed for treatment of leukemia.3 Doxorubicin was
identified as a hydroxylated analog of daunorubicin and was initially named adriamycin after the
source of the original soil sample, but was later renamed doxorubicin by the World Health
Organization.
O
OH
O
R
OH
CH3O
O
OH
O
O
NH3
Cl
HO
=
1
2
R
R
OH
H
=
Figure 1. Doxorubicin (1) and daunorubicin (2).
2
ACS Paragon Plus Environment

Page 3 of 31
Bioconjugate Chemistry
1
2
3
4
5
6
7
8
Doxorubicin has a wide spectrum of activity and is used today, either alone or in
combination with other drugs, for the treatment of a variety of cancers, including acute
leukemias and ovarian and breast cancers. It is one of the most effective cancer
9
chemotherapeutic agents available,4-7 and is on the World Health Organization list of essential
medicines for both adults and children.8 Regrettably its use is severely limited by various side
effects, including myelosuppression, vomiting, diarrhea, and especially life-threatening
cardiotoxicity.9, 10 Two liposomal formulations of doxorubicin have been approved for clinical
use. Doxil®, available in the U.S., is a pegylated liposomal formulation of doxorubicin that is
approved for treatment of Kaposi’s sarcoma, ovarian cancer, and multiple myeloma; it is much
less cardiotoxic than native doxorubicin, but it does have some new side-effects.11 Myocet®
(liposome encapsulated doxorubicin citrate), available in the U.K., also is less cardiotoxic than
doxorubicin and can be better tolerated by patients, while having an equivalent therapeutic
efficacy.12 While these formulations of doxorubicin are an improvement over the native drug,
there is still room for further improvement, and it is a measure of doxorubicin’s clinical
effectiveness that it is so widely used in spite of its very significant side effects. Clearly however
a selective targeted drug delivery method would offer the opportunity to make doxorubicin an
even better option for cancer treatment.
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
The basic idea of targeting cancer chemotherapeutic agents is well known, and several
approaches to targeting doxorubicin have been published. These include targeted delivery to
hepatocarcinoma cells by laponite nanodiscs,13 the use of magnetoliposomes,14 and the use of
doxorubicin-loaded poly(lactic-co-glycolic acid) hollow microcapsules.15 The nanoparticle
approach to drug delivery is attractive, since the enhanced permeability and retention effect
assures a measure of selectivity,16 although the magnitude of this effect has been questioned by a
3
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 4 of 31
1
2
3
4
5
6
7
8
9
meta-analysis which showed that only 0.7% of the administered dose of nanoparticle drugs is
delivered to solid tumors.17 Even this relatively low number is however higher than that of naked
drug molecules,18 and in the particular case of doxorubicin even this low delivery efficiency
could be significant if the remaining drug never reached the heart.
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
The use of gold nanoparticles (AuNP) for drug delivery is particularly attractive since
they can be generated in a range of sizes and can readily be functionalized with a variety of
ligands,19 and numerous studies have been made of AuNPs conjugated with doxorubicin. In
some cases protonated doxorubicin was absorbed directly onto the surface of negatively charged
AuNP under acidic conditions,20 and a one-pot synthesis of doxorubicin-loaded AuNP was
achieved in the presence of oxygen to generate radicals.21 Several studies used various polymers
to link doxorubicin to AuNP. Thus doxorubicin was loaded onto anionic polyaspartic acid AuNP
by ionic complexation,22 or it was bound to DNA which was itself bound to the AuNP.23 Another
study used DNA-caged AuNP that allowed T7 exonuclease and pH responsive controlled release
of doxorubicin.24 A study of doxorubicin-loaded PEG-AuNPs and their further antibody
targeting showed that toxicity was reduced and therapeutic efficacy increased for the antibody
targeted system.25 On the other hand, a study of PEGylated gold nanoparticles with doxorubicin
attached by covalent peptide or pH-active hydrazone linkers revealed that gold nanostructures
modified with doxorubicin were more toxic to A549 and HeLa cells than free doxorubicin.26 The
use of folate-modified pegylated AuNP to conjugate doxorubicin gave a drug system that
showed a much higher cytotoxicity than that of free doxorubicin, as well as an increased water-
solubility.27 Another study of folic acid-coated AuNP gave a construct with enhanced drug
accumulation and retention in multidrug resistant hepG2-R cancer cells compared with free
doxorubicin.28 Grafting a thiolated doxorubicin onto AuNPs via a cleavable disulfide linkage
4
ACS Paragon Plus Environment

Page 5 of 31
Bioconjugate Chemistry
1
2
3
4
5
6
7
8
9
(Au-PEG-SS-DOX) improved drug uptake as compared with native doxorubicin, although the
drug released was a doxorubicin derivative.29 A study of a formulation with a pH-sensitive
hydrazone bond as linker between doxorubicin and pegylated AuNP showed greatly enhanced
solubility and an excellent pH response profile of drug release for the drug construct.30 A
variation of this approach grafted doxorubicin by a cleavable hydrazone linker onto the
hydrophobic inner shell of a folate-conjugated poly(L-aspartate-doxorubicin)-b-poly(ethylene
glycol) copolymer to give a drug construct that released doxorubicin readily at pH 5.3.31
Ultrasmall gold-doxorubicin conjugates were found to target apoptosis-resistant cancer cells.32
The use of AuNPs for targeted drug delivery19, has been reviewed, as has the use of iron and gold
nanoparticles for doxorubicin delivery,19, 33 and the cytotoxicity of AuNPs has recently been
evaluated, with the finding that PEG-coated nanospheres, nanorods, and nanostars show low
cytotoxicity to U87 cells and fibroblasts.34
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
The use of AuNPs conjugated with the cytokine tumor necrosis factor (TNF) is a
particularly powerful approach to drug targeting, since TNF not only targets the nanoparticle to
tumors but also reduces the interstitial fluid pressure, allowing for enhanced uptake of the
chemotherapeutic payload.35 In the present study we investigated the conjugation of doxorubicin
to AuNPs in preparation for future studies to add TNF to the drug construct.
 RESULTS
Design and Synthesis of New Thiolated Doxorubicin Analogs. A key consideration in
any approach to drug delivery by AuNPs is whether the drug payload is released from the
nanoparticle, and if so how this will occur. Release of the native drug from the nanoparticle is
preferable, since it allows the drug to interact with its target unhindered by any linker, and so we
5
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 6 of 31
1
2
3
4
prepared the five doxorubicin analogs 3 – 7 (Figure 2) that were designed to release free
5
6
doxorubicin in the tumor cell.
7
8
9
O
O
OH
O
O
O
OH
O
O
O
OH
O
OH
OH
OH
OH
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
OH
OH
CH3O
OH
O
CH3O
OH
O
CH3O
OH
O
O
S
O
O
S
S
O
NH
NH
NH
O
O
HO
O
O
HO
O
O
O
HO
N
4H
O
O
O
O
S
S
O
3
4
5
S
O
O
O
NH
O
O
OH
OH
O
N
O
O
OH
OH
S
O
O
O
OH
S
OH
OH
O
O
CH3O
CH3O
O
O
O
O
NH3
NH3
Cl
Cl
HO
HO
6
7
Figure 2. Doxorubicin analogs 3 – 7.
The 7-O-thiobenzyl derivative 3 was designed to undergo reductive cleavage from the
AuNP by glutathione in the reducing tumor environment and to release native doxorubicin, as
previously described for a similar paclitaxel derivative.35 The proposed mechanism of
doxorubicin release is shown in Figure 3, with glutathione displacing analog 3 from the gold
surface to give deacetyl 3, which undergoes self-immolation (arrows) leading to doxorubicin, a
thioquinone methide, and carbon dioxide. The thioquinone methide then undergoes hydration to
(4-mercaptophenyl)methanol.
6
ACS Paragon Plus Environment

Page 7 of 31
Bioconjugate Chemistry
1
2
3
4
5
6
7
8
9
O
O
OH
OH
O
O
OH
OH
O
OH
OH
OH
OH
OH
O
OH
O
CH3O
O
O
CH3O
GSH
O
H+
O
O
HS
S
GS
NH
NH
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
O
O
self-immolation
3
Gold-bound
O
OH
OH
O
OH
OH
S
HS
H
2O
+
OH
CH3O
O
O
O
CO2
1
OH
NH2
Figure 3. Proposed mechanism of doxorubicin release from gold-bound analog 3 by glutathione (GSH).
The synthesis of analog 3 is shown in Scheme 1. Commercially available 4-
mercaptobenzoic acid (8) underwent acetylation, reduction and acylation with 4-nitrophenyl
chloroformate to obtain compound 11, which was then reacted directly with doxorubicin
hydrochloride (1) to give final product 3 in 63% overall yield.
Scheme 1. Synthesis of doxorubicin derivative 3a
O
O
3
OH
OH
O
O
OH
CH2OH
O
O
COOH
COOH
d
OH
c
a
b
S
CH3O
O
O
O
S
S
NO2
SH
S
NH
O
O
HO
O
O
8
9
10
11
O
a(a) Ac2O, pyridine, CH2Cl2, 6 h 0 ºC to rt, 98%; (b) Me2S.BH3, THF, -10 ºC to rt, 4 h, 85%: (c) 4-nitrophenyl
chloroformate, pyridine, CH2Cl2, 4 h 0 ºC to rt, 88%; (d) Et3N, DMF, 1, 0 ºC to rt, 2 h, 86%;
The syntheses of analogs 4 – 7 are described in the Supporting Information.
7
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 8 of 31
1
2
3
4
5
6
7
8
9
Antiproliferative Activities of the Analogs. The antiproliferative activities of the five
analogs in the A2780 ovarian cancer cell line were determined (Table 1). All the analogs had
similar IC50 values to that of doxorubicin, and three of the five (compounds 4, 6, and 7) even had
lower values than that of doxorubicin. Although this result is somewhat surprising, it does not
affect the selection of analog for linking to gold, since all the analogs were designed to release
intact doxorubicin.
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 1. Antiproliferative Activities of Compounds 3 – 7
Against the A2780 Human Ovarian Cancer Cell Line
IC50 values (nM)
Doxorubicin (1)
26 ± 3
3
4
5
6
7
31 ± 5 10 ± 4 28 ± 3 11 ± 5 11 ± 1
Stability Studies of the Analogs. The HPLC chromatograms of doxorubicin and analogs 3
and 4 (Figures S1 and S2) were used to establish a calibration curve (Figure S3) for analysis of
doxorubicin so as to determine the concentration of doxorubicin released from these analogs
under various conditions (Scheme 2). Only very slow loss of analog was observed for both 3 and
4 at pH 7.4, with about 90% of the original analog remaining after 3 days at 37 ºC (Figure S4).
At pH 4.6, however, analog 3 retained its stability, but analog 4 underwent conversion to
doxorubicin, with complete loss of 4 after 3 days and formation of doxorubicin (Fig. S5). The
half-lives of analog 3 were calculated to be 471 hours at pH 7.4 and 421 hours at pH 4.6; as
expected 3 is stable at both neutral and acidic pH values. For analog 4 the corresponding values
were 376 and 9.6 hours, consistent with the designed lability of 4 under acidic conditions. Only
about 55% of the doxorubicin present in 4 was detected as free doxorubicin, presumably due to
slow degradation under the conditions of the experiment. Although doxorubicin only
decomposes slowly at pH 4, with less than 10% degradation over 40 hours,36 this stability is
8
ACS Paragon Plus Environment

Page 9 of 31
Bioconjugate Chemistry
1
2
3
4
5
6
7
8
presumed to be due to protonation of the amine group. Since this group is replaced by an amide
in 4, it is possible that acidic decomposition of the anthraquinone moiety proceeds more readily
on the unprotonated 4 than on doxorubicin itself.
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
Analog 3 was selected for detailed examination based on the decomposition observed for
4 under acidic conditions and on the fact that analogs 5 – 7 also require acidic conditions for
drug release.
Scheme 2. Doxorubicin release from analogs 3 and 4 under two different conditions.
O
O
3
OH
O
OH
OH
CH3O
OH
O
O
S
at
at
7.4, 37 o
4.6, 37 o
NH
conversion
conversion
Slow
Slow
C
C
pH
pH
O
O
HO
O
O
O
3
OH
OH
O
OH
OH
CH3O
O
O
OH
O
OH
O
OH
NH2
conversion
at
7.4, 37 o
C
4.6, 37 o
C
Slow
pH
HO
conversion
at
Rapid
pH
CH3O
O
OH O
O
S
S
NH
O
O
HO
O
O
4
Preparation of AuNPs. Gold nanoparticles (AuNPs) were prepared by two variations of the
Wang method; method A with added AgNO3, and method B without added AgNO3.37
37
Method A. AuNPs prepared by the Wang method with added AgNO3 used a trace amount of
silver nitrate to produce quasi-spherical AuNPs (Fig. S6). The citrate concentration was adjusted
to produce nanoparticles with an average diameter of 41 nm, polydispersity index (PDI) of 0.212
and a Zeta potential of -26 ± 3 mV (Fig. S7a). PEGylated gold nanoparticles were prepared by
treatment of a solution of gold nanoparticles with PEG methyl ether thiol; these had average
9
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 10 of 31
1
2
3
4
5
6
7
8
9
diameter of 49 nm, PDI of 0.317, and a Zeta potential of -16 ± 2 mV (Fig. S7b). Doxorubicin-
functionalized nanoparticles were prepared by treating a solution of nanoparticles prepared as
described above with doxorubicin analog 3 and mPEG thiol in CH3OH, followed by repeated
centrifugation and resuspension to remove excess reagents. This gave soluble doxorubicin-
loaded nanoparticles Au-3. Incubation of Au-3 at pH 4.6 in the presence of 10 mM glutathione at
37 ºC for 72 h and analysis of the resulting solutions by HPLC showed a single peak which
matched that of doxorubicin. The area of the peak indicated that the drug loading of 3 was 0.78
mg analog per 1 mg of AuNP.
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
37
Method B. AuNPs were prepared by the Wang method without added AgNO3. The citrate
concentration was adjusted to produce approximately spherical nanoparticles with an average
diameter of 34.6 ± 0.1 nm as determined by TEM (Figures 4 and 5), and supported by dynamic
light scattering (Figure S8). The average diameters of the doxorubicin-functionalized
nanoparticles differed from the nonfunctionalized nanoparticles by less than 5% (Figures S9-
S20). The DLS of these and other nanoparticle preparations also showed signals for other size
particles, presumably due to weak aggregation of particles during storage. However, the
measured particle hydrodynamic diameter of the major population aligned with the size ranges
seen through TEM. Long-term sonication of these samples caused further destabilization and
aggregation of samples.
10
ACS Paragon Plus Environment

Page 11 of 31
Bioconjugate Chemistry
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. TEM images of naked AuNPs.
Figure 5. Image analysis of >400 AuNP, 5 nm bins fit to a LogNormal distribution (see equation in Supporting
Information)
Drug loading and release of nanoparticles as a function of the concentration of
PEG. Doxorubicin-functionalized nanoparticles were prepared by treating 50 mL of a solution
of AuNP containing 2.5 mg AuNP with 6.67 mg of 3 in 100 mL CH3OH and either 13.3 mg or
26.6 mg of PEG methyl ether thiol (mPEG thiol) in 1.0 mL water with stirring for 48 h, followed
by repeated centrifugation and resuspension to remove excess reagents. The residue was
dissolved in 1 mL water to give soluble doxorubicin-loaded nanoparticles Au-3-1 (Figures S9 –
S11) and Au-3-2 (Figures S12 – S14) respectively, with gold contents of 1.05 and 1.70 mg as
11
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 12 of 31
1
2
3
4
5
6
7
8
9
determined by ICP-MS; the low recovery of AuNP reflects handling losses during the repeated
centrifugation and resuspension purification procedure. Au-3-1 and Au-3-2 were then incubated
with glutathione as described above for 72 h to ensure complete drug release, and the resulting
solutions were analyzed by HPLC. Both Au-3-1 and Au-3-2 gave a major peak which matched
that of doxorubicin. The areas of the peaks combined with ICP-MS data indicated that Au-3-1
had a loading of 0.018 mg per mg AuNP, while Au-3-2 had a loading of 0.012 mg per mg AuNP.
This result indicates that the concentration of mPEG thiol affects drug loading, with the higher
concentration of mPEG thiol reducing drug loading.
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
Drug loading and release of nanoparticles as a function of doxorubicin
concentration. Doxorubicin-functionalized nanoparticles were prepared by treating a solution
of AuNP prepared as described above with a fixed concentration of mPEG thiol of 88.67 µg/mL
(17.7 µM) and concentrations of 3 of 6.67, 13.34 and 20 µg/mL (8.87, 17.7, and 26.6 µM) in
CH3OH, followed by centrifugation and resuspension to remove excess reagents. This gave
soluble doxorubicin-loaded nanoparticles Au-3-1, Au-3-3 (Figures S15 – S17) and Au-3-4
(Figures S18 – S20) with gold contents of 1.05, 0.39 and 1.79 mg as determined by ICP-MS.
Samples of each of the three doxorubicin-loaded nanoparticles were incubated in the presence of
5 mM glutathione and analyzed as described above, and the resulting chromatograms of each
exhibited a major peak which matched that of doxorubicin. The area of the peak and the ICP-MS
data indicated that Au-3-1 had a loading of 0.018 mg per mg AuNP, while Au-3-3 had a loading
of 0.034 mg per mg AuNP and Au-3-4 had a loading of 0.050 mg per mg AuNP. These data
indicated unsurprisingly that the doxorubicin loading increased as the initial concentration of
doxorubicin analog 3 increased.
12
ACS Paragon Plus Environment

Page 13 of 31
Bioconjugate Chemistry
1
2
3
4
5
6
7
8
9
Pharmacokinetics of doxorubicin and Au-3 in mouse plasma. Treatment of seventy-
two CD-1 mice with either doxorubicin hydrochloride at 6 mg/Kg or Au-3 containing 6 mg/Kg
doxorubicin, followed by euthanasia in groups of 4 at ½, 1, 2, 4, 6, 12, 16, 24 and 48 hours post-
dose, was followed by analysis of blood plasma samples for doxorubicin by LC-MS/MS.
Doxorubicin concentration in plasma from the doxorubicin group started at 35 ng/mL and fell to
less than 5 ng/mL over 16 hours (Figure 6 and Tables S4 and S5). On the other hand the
doxorubicin concentration in plasma from the Au-3 group started at 0.14 ng/mL and fell to
almost undetectable levels over 16 hours (Figure 7 and Tables S6 and S7). This experiment
indicated that almost all of the doxorubicin administered as the nanoparticle formulation Au-3
remained bound to the nanoparticle in mouse blood, and was not released into the plasma.
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
70
60
50
40
30
20
10
0
0
10
20
30
40
50
Time (hr)
Figure 6. Median doxorubicin concentration in plasma from mice treated with 6 mg/kg doxorubicin.
13
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 14 of 31
1
2
3
4
5
6
7
8
9
0.45
0.40
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
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
0
10
20
30
40
50
Time (hr)
Figure 7. Median doxorubicin concentration in plasma from mice treated with 6 mg/kg doxorubicin as Au-3. Note the different
concentration scale from Figure 4.
Safety of doxorubicin and Au-3 to mice. Forty-eight male, CD-1 mice were given a single
IV dose of saline vehicle (dose group 1), 6 mg/kg doxorubicin hydrochloride (dose group 2) or 6
mg/kg doxorubicin as Au-3 (dose group 3). The dose of doxorubicin was based on equivalent
human doses38 and previous experience of toxic effects that followed IV administration of
doxorubicin to rodents.39 The dose chosen was below that capable of causing the overt
neurotoxicity previously reported yet within the possibility of dosages used for therapeutic
purposes in people.
Neurobehavioral Assessment analysis. As part of the safety study, mice were evaluated
for neurobehavioral endpoints as described in the Experimental Section using the endpoints of
Table S8. Only the tail condition showed some variation between the treatment groups, but the
proportions of abnormal tails were not significantly different between the groups at any of the
time points (P-values > 0.05) (Table S9).
Effect of treatment on body weight. At baseline the groups of mice were balanced for
weight, and each group showed moderate weight gain during the study. At 2 and 4 days of post-
14
ACS Paragon Plus Environment

Page 15 of 31
Bioconjugate Chemistry
1
2
3
4
5
6
7
8
9
dosing, the groups were not significantly different, but at 14 days, the least squares mean for
group 3 (35.62 SE 0.40) was significantly greater than the least squares mean for group 1 (34.26
SE 0.39) (0.0488). Similarly, the least squares mean for group 3 (35.62 SE 0.40) was
significantly greater than the least squares mean for group 2 (33.41 SE 0.40) (0.0008). Group 1
was not significantly different from group 2 (Figure 8). Least squares means (standard error) are
presented in Table S10.
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 8. Effect of treatment on body weight in grams. Group 1 saline; group 2 doxorubicin.HCl; group 3 Au-3.
Clinical Pathology. Review of the chemistry panel and hematology panel showed that 4/4
mice given native doxorubicin (group 2) had significantly decreased reticulocyte counts 2 and 4
days after dosing (range of 2-33 compared to range of 201-340 in group 1 control mice). At 14
days post dosing 3 of 4 mice had counts above the range of the controls (414, 466, 745), with this
15
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 16 of 31
1
2
3
4
5
6
also providing significant difference from controls (p<0.05). No such effects were noted with
the mice given Au-3 (group 3, Figure S7).
7
8
9
Histopathology. Non-blinded histologic examination was performed on the following H&E-
stained sections of paraffin-embedded tissues: heart, lung, stomach, small intestine, large
intestine, liver, pancreas, kidney, adrenal, testes, spleen, thymus, skeletal muscle, vertebral bone
marrow, brain, spinal cord, and some dorsal root ganglions. Samples were obtained from
animals sacrificed on post-dosing days 2, 4 and 14. Histopathological lesions, viewed after
staining with hematoxylin and eosin, were restricted to the group of mice receiving conventional
doxorubicin by IV injection. On post-dosing day 2 there was marked erythroid hypoplasia and
reduced overall cellularity of the hematopoietic elements of the marrow, seen in 4/4 Group 2
mice. Myeloid cells were prominent, but whether this was due to hyperplasia or was relative due
to the erythroid hypoplasia could not be determined. On day 4 erythroid hypoplasia was still
evident in 3/4 mice, along with myeloid hyperplasia. By day 14 the bone marrow of the Group 2
animals had returned to normal. No changes in megakaryocytes was detected in the 2 or 4 day
Group 2 mice. The appearance of the bone marrow in Group 3 (6 mg/kg doxorubicin-AuNP)
was similar to that of Group 1 (vehicle control) at all stages. Occasional incidental background
changes in various tissues, of no significance, were seen in all Groups.
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
 DISCUSSION
The above data show that doxorubicin analog 3 is stable both in pH 7.4 and 4.6 buffer solutions
and that it binds well to AuNPs in the presence of mPEG thiol. The loading of 3 on AuNP can be
varied by adjusting the ratio of 3 and mPEG thiol, and the nanoparticle-bound forms of 3 all
release doxorubicin cleanly from the nanoparticles under reducing conditions. The highly
16
ACS Paragon Plus Environment

Page 17 of 31
Bioconjugate Chemistry
1
2
3
4
5
6
7
8
9
doxorubicin-loaded nanoparticle form Au-3 was evaluated pharmacologically, and is stable in
mouse blood. Significantly lower plasma levels of doxorubicin were noted in plasma samples
from mice given Au-3 when compared to those given conventional doxorubicin, suggesting that
its nanoparticle-bound form contributed to protection from pathological defects. Treatment of
mice with Au-3 gave no histopathologically observable differences from mice treated with
saline, while mice treated with doxorubicin showed significant histopathologically observable
lesions. These results suggest that Au-3 is has significantly reduced toxicity to normal cells as
compared with doxorubicin. Whether the nanoparticle formulation of doxorubicin, with its
longer half-life, would be effective in a clinical situation would require efficacy studies in
diseased subjects. However, since TNF is known to target AuNPs to tumors,35 these results
suggest that linking of TNF to Au-3 would provide a potent delivery vehicle for doxorubicin that
lacks some of the major side effects of this potent chemotherapeutic agent.
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
 EXPERIMENTAL SECTION
The following standard conditions apply unless other stated. All chemicals were purchased from
Sigma-Aldrich and used as received unless otherwise stated. Doxorubicin hydrochloride was
purchased from Wonda Science Inc, USA, amino-dPEG™4 acid was obtained from
QuanatBioDesign, Ltd., Powell, OH, and methoxypolyethylene glycol thiol (5 kDa) was
obtained from Creative PEGworks. All glassware was oven-dried and all reactions were
conducted under nitrogen. Dichloromethane (CH2Cl2) was distilled from CaCl2, tetrahydrofuran
(THF) was distilled from CaH2, anhydrous methanol (MeOH) was purchased directly from
Sigma-Aldrich. Thin layer chromatography (TLC) was used to monitor the reaction progress on
silica gel UV254. Purification was achieved either by column chromatography over silica gel 60
17
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 18 of 31
1
2
3
4
5
6
(220–240 mesh), or preparative thin layer chromatography (PTLC) on glass-backed plates coated
with silica UV254. HPLC was performed on Shimadzu SCL-10AVP system using a column
7
8
1
13
purchased from Phenomenex (Luna 5μ C18 (2) 25 x 4.6 mm). H and C NMR spectra were
9
1
obtained on an Agilent U4-DD2 spectrometer at 400 MHz for H and 100 MHz for 13C or a
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
1
13
Bruker Avance II spectrometer at 500 MHz for H and 125 MHz for C. All chemical shifts
1
13
were referenced to the solvent peaks of CDCl3 (7.26 ppm for H and 77.16 ppm for C) unless
otherwise stated. An Agilent 6220 accurate-mass TOF LC/MS instrument was used to obtain
high-resolution mass spectra of pure compounds, an Agilent 6470 Triple Quad LC-MS/MS was
used for analysis of samples from biological media at the Virginia Tech Toxicology Diagnostic
Laboratory, and an ICP-MS (Thermo Electron X-Series) following APHA Standard Method
3125-B (American Public Health Association et al., 1998) was used for analysis of
concentrations of gold content in the Civil and Environmental Engineering Department at
Virginia Tech. Specific protocols for animal experiments were reviewed and approved by
Virginia Tech’s Institutional Animal Care and Use Committee (IACUC).
4-(Acetylthio)benzoic acid (9). A 25 mL round bottom flask was oven-dried, cooled to room
temperature and charged with a magnetic stirring bar, 4-mercaptobenzoic acid 8 (154 mg, 1.0
mmol), and a rubber septum. Dry CH2Cl2 (5 mL) was added via syringe to dissolve the solid and
followed by adding C5H5N (0.323 mL, 4.0 mmol). An ice bath was used to cool the solution to 0
ºC, followed by vigorous stirring for 5 min. After the solid was dissolved completely, Ac2O (0.1
mL, 1.06 mmol) was added over a period of 10 min via syringe. The resulting mixture was
warmed to room temperature and stirred for 6 h under nitrogen (balloon). The reaction was
diluted with EtOAc (100 mL), washed with 0.6 M HCl solution (2 x 10 mL), brine (2 x 20 mL),
and dried with anhydrous Na2SO4. After evaporating the solvent by rotary evaporation (25 ºC), a
18
ACS Paragon Plus Environment

Page 19 of 31
Bioconjugate Chemistry
1
2
3
4
white crude product was obtained and purified by PTLC (eluted with CHCl3:MeOH, 13:1) to
5
6
yield 9 (192 mg, 0.98 mmol) as a white powder in a yield of 98%. 1H NMR (CD3OD, 500 MHz,
7
8
9
13
δ): δ 8.05 (dd, J = 8.5, 2 Hz, 2H), 7.54 (dd, J = 8.5, 2 Hz, 2H), 2.44 (s, 3H); C NMR (126
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
MHz, δ): δ 194.2, 169.1, 135.2, 135.0, 132.8, 131.2, 30.3; HRESIMS: (m/z) [M – H]– calcd for
C9H7O3S, 195.0121; found 195.0124.
S-(4-(hydroxymethyl)phenyl) ethanethioate (10). A 10 mL round bottom flask was oven-
dried, cooled to room temperature and charged with a magnetic stirring bar, compound 9 (192
mg, 0.98 mmol), and a rubber septum. The flask was evacuated for 5 min and flushed with
nitrogen. Dry THF (3 mL) was added to dissolve the solid. An ice-salt bath was used to cool the
solution to –10 ºC, followed by vigorous stirring for 5 min. A solution of (CH3)2S•BH3 in THF
(1.0 mL, 2.0 mmol) was added over a period of 10 min via syringe. The resulting mixture was
warmed to room temperature and stirred for 4 h under nitrogen (balloon). H2O (2 mL) was added
to remove unreacted (CH3)2S•BH3, the solution was diluted with EtOAc (100 mL), washed with
brine (2 x 20 mL), and dried with anhydrous Na2SO4. After evaporating the solvent by rotary
evaporation (30 ºC), a liquid crude product was obtained, which was purified by PTLC (eluted
with CHCl3:MeOH, 15:1) to yield 10 (151 mg, 0.83 mmol) as a colorless liquid in a yield of
85%. 1H NMR (CD3OD, 500 MHz, δ): δ 7.36 (m, 4H), 4.60 (s, 2H), 2.34 (s, 3H); 13C NMR (126
MHz, δ): 194.5, 143.2, 134.2, 127.2, 126.6, 63.2, 28.6; HRESIMS: (m/z) [M + H]+ calcd for
C9H11O2S+, 183.0474; found 183.0466.
S-4-((((4-nitrophenoxy)carbonyl)oxy)methyl)-phenyl ethanethioate (11). A 50 mL
round bottom flask was oven-dried, cooled to room temperature and charged with a magnetic
stirring bar, compound 10 (151 mg, 0.83 mmol), and a rubber septum. Dry CH2Cl2 (10 mL) was
added via syringe to dissolve the liquid followed by adding C5H5N (0.202 mL, 2.5 mmol). An ice
19
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 20 of 31
1
2
3
4
5
6
7
8
9
bath was used to cool the solution to 0 ºC, followed by vigorous stirring for 5 min. A solution of
4-nitrophenyl chloroformate (168 mg, 0.83 mmol) in 5 mL CH2Cl2 was added over a period of 10
min via syringe. The resulting mixture was warmed to room temperature and stirred for 4 h
under nitrogen (balloon). The reaction was diluted with EtOAc (100 mL), washed with saturated
NaHCO3 solution (2 x 10 mL), H2O (2 x 10 mL) and brine (2 x 10 mL), and dried with
anhydrous Na2SO4. The solid crude obtained after evaporation was purified by column
chromatography (silica gel, 100 g, eluted with hexane:CH2Cl2, 5:1) to yield 11 (253 mg, 0.73
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
1
mmol) as a white powder in a yield of 88%. H NMR (CDCl3, 500 MHz, δ): 8.27 (dd, J = 9, 2
13
Hz, 2H), 7.47 (m, 4H), 7.38 (dd, J = 9, 2 Hz, 2H), 5.31 (s, 2H), 2.44 (s, 3H); C NMR (126
MHz, δ): 193.7, 155.5, 152.5, 145.5, 135.6, 134.8, 129.3, 129.0, 125.4, 121.9, 70.2, 30.4;
HRESIMS (m/z) [M + NH4]+ calcd for C16H17N2O6S+, 365.0807; found 365.0795.
Doxorubicin Analog 3: S-(4-(((((2S,3S,4S,6R)-3-hydroxy-2-methyl-6-(((1S,3S)-
3,5,12-trihydroxy-3-(2-hydroxyacetyl)-10-methoxy-6,11-dioxo-1,2,3,4,6,11-
hexahydrotetracen-1-yl)oxy)tetrahydro-2H-pyran-4yl)carbamoyl)oxy)
methyl)phenyl) ethanethioate. A 10 mL round bottom flask was oven-dried, cooled to room
temperature and charged with a magnetic stirring bar, doxorubicin hydrochloride (20 mg, 0.0345
mmol), compound 11 (20 mg, 0.0576 mmol), and a rubber septum. Anhydrous DMF (5 mL) was
added by syringe to dissolve the solids. An ice bath was used to cool the solution to 0 ºC,
followed by vigorous stirring for 5 min. Then Et3N (0.013 mL, 0.0988 mmol) was added to the
solution over a period of 5 min via syringe. The resulting mixture was warmed to room
temperature and stirred for 2 h under nitrogen (balloon). The reaction was diluted with EtOAc
(30 mL), washed with 0.6 M HCl solution (2 x 5 mL), saturated sodium NaHCO3 (2 x 5 mL),
brine (2 x 10 mL), and dried with anhydrous Na2SO4. The crude product was purified by PTLC
20
ACS Paragon Plus Environment

Page 21 of 31
Bioconjugate Chemistry
1
2
3
4
5
6
7
8
9
(eluted with CHCl3:MeOH, 13:1) to yield 3 (22.3 mg, 0.0297 mmol) as a red solid in a yield of
86%. 1H NMR (CDCl3, 500 MHz, δ): 13.95 (s, 1H, -OH), 13.21 (s, 1H, -OH), 8.02 (dd, J = 8.0,
1.1 Hz, 1H), 7.78 (t, J = 8.0 Hz, 1H), 7.39 (dd, J = 8.0, 1.1 Hz, 1H), 7.34 (m, 4H), 5.50 (d, J =
4.0 Hz, 1H), 5.27 (d, J = 3.5 Hz, 1H), 5.19 (d, J = 8.6 Hz, 1H, -OH), 5.04 (s, 2H), 4.75 (dd, J =
5.0, 2.0 Hz, 2H), 4.53 (s, 1H, -OH), 4.13 (q, J = 6.5 Hz, 1H), 4.07 (s, 3H), 3.86 (m, 1H), 3.66 (d,
J = 5.6 Hz, 1H), 3.25 (dd, J = 18.8, 2.0 Hz, 1H), 3.02 (t, J = 5.0 Hz, 1H, -OH), 3.00 (d, J = 18.8
Hz, 1H), 2.39 (s, 3H), 2.32 (dt, J = 14.7, 2.0 Hz, 1H), 2.16 (dd, J = 14.7, 3.5 Hz, 1H), 2.05 (d, J =
7.5 Hz, 1H, -OH), 1.87 (dd, J = 13.2, 5.1 Hz, 1H), 1.77 (td, J = 13.2, 4.0 Hz, 1H), 1.25 (d, J = 6.5
Hz, 3H); 13C NMR (126 MHz, δ): 214.0, 194.0, 187.2, 186.8, 161.2, 156.3, 155.8, 155.5, 138.0,
135.9, 135.6, 134.7, 133.7, 133.6, 128.8, 127.8, 121.0, 120.0, 118.6, 111.7, 111.6, 100.9, 69.8,
69.7, 67.4, 66.2, 65.7, 56.8, 47.2, 35.8, 34.1, 30.4, 30.3 17.0; HRESIMS: (m/z) [M + Na]+ calcd
for C37H37NO14SNa+, 774.1832: found 774.1840.
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
Stability studies of analog 3. Calibration curves for free doxorubicin and analog 3 were
constructed as described in the SI (Fig. S1). Analog 3 was then dissolved in a 1:3 mixture of
methanol and PBS buffer (pH 7.4) or sodium acetate buffer (pH 4.6); methanol was used to
increase the solubility of the analog. Both solutions were stirring vigorously at 37 ºC. Aliquots
were taken at various time points over a 72 h range, and their concentrations were determined by
quantitative HPLC analysis on a Phenomenex Luna 5μ C18 column, 25 x 4.6 mm, with gradient
elution from 30% CH3CN:70% H2O at 0 min to 100% CH3CN:0% H2O at 30 min and detection
by UV lamp at 253 nm. The concentration of analog at each time point was calculated from the
calibration curves (Fig. S2).
Preparation of monodisperse spherical AuNPs. A 25 mL round bottom flask was
charged with 1.5 mL sodium citrate aqueous solution (1 wt %) and a magnetic stirring bar.
21
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 22 of 31
1
2
3
4
5
6
7
8
9
Aqueous HAuCl4 (5.0 mL containing 12.5 mg Au as) was added, and H2O (6.0 mL) was added
to bring the volume of the solution to 12.5 mL. This mixture was incubated for 5 min before
rapid addition to 237.5 mL of boiling H2O in a 500 mL two-necked round bottom flask with
vigorous stirring. The color of the reaction solution changed quickly from colorless, grayish blue
to purple and then to ruby red within 1 min. The reaction solution was refluxed for 30 min under
stirring to promote formation of uniform quasi-spherical AuNPs.
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
Excess sodium citrate and other small molecules were removed by dialysis using dialysis
tubing with a MW cutoff of 3,500 Da and immersing it in 4 L water with vigorous stirring. The
water was changed every 2 hours over an 8 hour period and the resulting solution (250 mL) was
used for the following steps.
Transmission Electron Microscopy. Aliquots of 0.01 mL of the AuNP suspensions were
deposited on carbon-coated 200-mesh copper grids. The aqueous droplet was left to dry in
ambient conditions, typically for ~1 hour. Grids were imaged using a Philips EM420 electron
microscope with an accelerating voltage of 120 kV to enhance the contrast of the electron-dense
gold. Enough frames were collected to capture at least 400 particles per replicate. Nanoparticle
diameter (d) was estimated using the formula d = A/π, where A is the cross-sectional area of a
particle measured using ImageJ software.
Dynamic Light Scattering. Aliquots of 0.1 mL of each suspension were diluted to 1 mL with
DI H2O. The hydrodynamic diameter (HD) and PDI were measured in polystyrene cuvettes
under 25 oC after an equilibration time of 120 seconds using a Malvern Zetasizer NanoZS.
Synthesis of Doxorubicin Loaded AuNPs. A 50 mL aqueous solution of AuNPs
containing 2.5 mg AuNP as described above, a 100 mL solution of 1 mg doxorubicin analog 3 in
CH3OH, and a 1 mL aqueous solution of either 13.3 mg or 26.6 mg mPEG thiol (5 kDa) were
22
ACS Paragon Plus Environment

Page 23 of 31
Bioconjugate Chemistry
1
2
3
4
5
6
7
8
9
mixed and stirred for 48 h at room temperature. Unbound drugs and mPEG thiol were removed
by centrifugation on an Eppendorf Centrifuge 5810 R with a rotor diameter of 10 cm at 14,000
rpm (10,956 x g) for 20 min at room temperature, pipetting of the supernatant and re-dispersal of
the nanoparticle pellet in water. This process was repeated three times to remove the unbound
material completely. The residue was dissolved in 1.0 mL water to make Au-3-1 and Au-3-2
respectively. Repetition of this process with 2 mg of 3 and 13.3 mg mPEG thiol gave Au-3-3,
and with 3 mg of 3 and 13.3 mg mPEG thiol gave Au-3-4. Four batches of each sample were
prepared.
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
ICP-MS of Doxorubicin Loaded AuNPs. A 10 µL aliquot of a 1.0 mL solution of Au-3-1
and 25 µL aliquots of solutions of Au-3-2, Au-3-3 and Au-3-4 in 10 mL water were used to
determine the concentration of Au by ICP-MS. The results were 1054, 4256, 982, and 4473 ppb
respectively, corresponding to 0.53, 0.85, 0.196 and 0.895 mg Au in 500 µL.
Drug Release Studies of Analogs 3-1 – 3-4 from AuNP. A 500 µL portion of each of the
four AuNP solutions prepared above was mixed with 500 µL sodium acetate buffer containing
10 mM GSH, and the pH was adjusted to 4.6. The solution was stirred vigorously at 37 ºC for 72
h. The solution was then centrifuged at a speed of 14,000 rpm for 20 min. The four different
supernatants were evaporated and dissolved in 450 µL MeCN and 50 µL 0.6 M HCl, and
aliquots (100 μL) of the solution were injected onto an HPLC column and analyzed by
quantitative HPLC. Au-3-1, Au-3-2, Au-3-3 and Au-3-4 each gave a single peak which matched
that of doxorubicin. The areas of the peaks were compared with those of the standard curve, and
corresponding to masses of 0.00192, 0.00208, 0.00132 and 0.00899 mg per injection,
corresponding to 0.0096, 0.0104, 0.0066 and 0.045 mg per sample. After adjustment for the
differing masses of nanoparticles in each sample, these loadings correspond to drug loadings of
23
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 24 of 31
1
2
3
4
5
6
0.018, 0.012, 0.034, and 0.050 mg doxorubicin per 1 mg of AuNP for Au-3-1 to Au-3-4,
respectively.
7
8
9
Synthesis of Doxorubicin Loaded AuNPs Au-3 for Biological Studies. A 10 mL
aqueous solution of AuNPs containing 10 mg AuNPs, a 20 mL solution of 10 mg doxorubicin
analog 3 in CH3OH, and a 3 mL aqueous solution of 30 mg mPEG thiol (5 kDa) were mixed and
stirred for 48 h at room temperature. Unbound drugs and mPEG-thiol were removed by
centrifugation on an Eppendorf Centrifuge 5810 R with rotor diameter of 10 cm at 14,000 rpm
(10,956 x g) for 20 min at room temperature followed by removal of the supernatant by pipette
and re-dispersal of the nanoparticle pellet in water. This process was repeated three times to
remove the unbound material completely. The residue was dissolved in 1 mL water to give a
solution designated Au-3. Incubation of Au-3 at pH 4.6 in the presence of 10 mM glutathione at
37 ºC for 72 h and analysis of the resulting solutions by HPLC showed a single peak which
matched that of doxorubicin. The area of the peak indicated that the drug loading of 3 was 0.78
mg analog per 1 mg of AuNP.
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
Pharmacokinetics of Doxorubicin and Au-3 in Mouse Plasma. Seventy-two male,
CD-1 mice (Envigo) were used for the pharmacokinetic study. All animals were group housed
on Diamond Dry Cellulose Bedding (Envigo), given a standard pelleted rat chow and water ad
libitum. Mice were randomly assigned to dose groups. Dose groups consisted of either 6 mg/kg
doxorubicin hydrochloride (dose group 2) or Au-3 containing 6 mg/kg doxorubicin (dose group
3). Animals were intravenously dosed at time 0, and 4 animals per group were sacrificed at ½, 1,
2, 4, 6, 12, 16, 24 and 48 hours post-dosing with euthanasia by intraperitoneal injection of 200
mg/kg of sodium pentobarbital. Up to 1 mL of whole blood was collected in heparinized
syringes from the inferior vena cava to obtain plasma for LC-MS/MS analysis. The samples were
24
ACS Paragon Plus Environment

Page 25 of 31
Bioconjugate Chemistry
1
2
3
4
5
6
7
8
9
transferred to Vacutainer® blood collection tubes with added lithium heparin and centrifuged to
separate the plasma. The plasma supernatant was removed and stored in Eppendorf tubes at -70
ºC until analyses could be done. For analysis the samples were extracted with acetonitrile and the
top organic layer containing doxorubicin was evaporated to dryness and analyzed by LC-MS/MS
with a H2O:CH3CN solvent gradient containing 1% formic acid (Table S1). Doxorubicin
extracted from blood was compared with that extracted from buffer, using free doxorubicin as
the internal standard (Figure S6). Daunorubicin was used as internal standard.
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
Determination of Pharmacokinetic Properties. Pharmacokinetic calculations were based
on 3 plasma samples per time point. Inspection of the time course plot showed that data
followed an approximate one-compartment model for the doxorubicin group. Median plasma
concentrations were entered into PKSolver40 for determination of elimination rate constant (kel),
apparent volume of distribution (Vd), clearance (Cl) and half-life (t1/2). The maximal
concentration (Cmax) and the time to Cmax (Tmax) were determined from observing the graph of
median plasma concentrations versus time. Concentrations for the Au-3 group were all below 1
ng/mL and appeared to show more random variation over time.
Safety of Doxorubicin and Au-3 to Mice. Forty-eight male, CD-1 mice (Envigo) were
used for the safety study and maintained as described above. Dose groups consisted of either a
saline vehicle (dose group 1), 6 mg/kg doxorubicin hydrochloride (dose group 2) or Au-3 6
mg/kg AuNPs (dose group 3). All solutions were given once by IV injection of 0.05 mL volume
into the lateral tail vein.
A baseline neurobehavioral assessment was performed prior to dosing (study day -3) and
then again on study days 2, 4 (n = 10/group/timepoints) and 14 (n = 8/group). Each assessment
consisted of clinical observations scored as normal/abnormal or yes/no (observational category is
25
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 26 of 31
1
2
3
4
5
6
7
8
9
present/absent) and a recording of body weight. Physiological endpoints (weight, body
temperature) and observational endpoints (general appearance, mobility, nervous system
functioning) were compared between vehicle-dosed and test-agent dosed mice. Mice (4/group)
were sacrificed by pentobarbital overdose at 2, 4 and 14 days and underwent examination for
detrimental effects as determined by clinical pathology and histopathology. The following tissues
were collected and immersion-fixed in 10% neutral buffered formalin: brain, heart, lung,
thymus, stomach, duodenum/pancreas, small intestine, large intestine, liver, kidney, adrenal
gland, spleen, testes, gastrocnemius muscle and lumbar spinal column. Sections of tail were also
collected from animals sacrificed on day 14. Lumbar spinal column and tail samples were
decalcified overnight in EDF® prior to all tissues being trimmed and analyzed by the
Histopathology Laboratory of the Veterinary Medical Teaching Hospital. There the samples
were dehydrated and cleared and embedded in Paraplast®. The samples were then sectioned at 5
µm thickness, stained with hematoxylin and eosin (H&E) for pathological examination by light
microscopy (Figure S7).
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
For clinical pathology, up to 1 mL of whole blood was collected using heparinized
syringes. Blood samples were transferred to two Microtainer® tubes, one with added lithium
heparin and one with K2EDTA (up to 500 µL each). Samples were then analyzed by the Clinical
Pathology Laboratory of the Veterinary Medical Teaching Hospital for blood chemistry (Table
S9).
Statistical Analysis of Behavior Data. Categorical endpoints of the Neurobehavioral
Assessment investigated are listed in Table S6. Subsequently the proportion of abnormal tails
were compared between the groups at each time point using the Mantel-Haenszel Chi-Square.
For the continuous endpoint of body weight, the treatment groups were compared using mixed-
26
ACS Paragon Plus Environment

Page 27 of 31
Bioconjugate Chemistry
1
2
3
4
5
6
7
8
model analysis of variance (at baseline) and mixed-model analysis of covariance (with baseline
measurements as covariates) at each time point during follow up. All analyses controlled for
block as specified in the study design.
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
 ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the ACS Publications website at
DOI:10.1021/acs.bioconjchem.7b00164y
Synthetic procedures for analogs 4 – 7, 1H and 13C NMR spectra of all new compounds,
experimental description of bioanalytical methods, Tables S1 – S3, and Figures S1 – S25.
 AUTHOR INFORMATION
Corresponding Author
*D.G.I.K.: e-mail, dkingston@vt.edu; phone, +1 540-231-6570
Yongle Du and Long Xia contributed equally to this work.
Notes
The authors declare the following competing financial interests: D.G.I.K. holds stock options in
CytImmune Sciences, Inc., a company that is developing AuNP-based drug delivery systems.
 ACKNOWLEDGMENTS
This work was supported in part by a grant from the Virginia Tech Institute for Critical
Technology and Applied Science “A Nanotechnology Approach to Overcome the Barriers of
Cancer Drug Delivery.” The authors thank Peggy Brodie, Qingxi Su and Ming Wang for
determining the IC50 values of the analogs against the A2780 ovarian cancer cell line.
27
ACS Paragon Plus Environment

Bioconjugate Chemistry
Page 28 of 31
1
2
3
4
5
6
7
8
9
Drs. Bernard S. Jortner, Katie Boes, and Stephen Werre of the College of Veterinary Medicine at
Virginia Tech provided results for histopathology, clinical pathology, and statistics (including
pharmacokinetics), respectively. Jonathan Hinckley dosed the mice, did behavioral observations,
collected blood and tissue samples, and prepared the formalin-fixed tissue for histopathological
processing and staining.
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
 REFERENCES
(1)
Arcamone, F., Cassindelli, G., Fantini, G., Orezzi, P., Pol, C., and Spalla, C. (1969)
Adriamycin, 14-hydroxydaunomycin, a new antitumor antibiotic from S. peucetius var.
caesius. Biotechnol. Bioeng. 11, 1101-1110.
(2)
(3)
(4)
Arcamone, F.-M., Gaetani, M., Scotti, T., and Dimarco, A. (1961) Isolamento ed attivita
antitumorale di un antibiotico da Streptomyces sp. Giorn. Microbiol. 9, 83-90.
Arcamone, F.-M. (2009) Fifty Years of Chemical Research at Farmitalia. Chem. Eur. J.
16, 7774.
Arcamone, F.-M. (2012) Anthracyclines, in Anticancer Agents from Natural Products
(Cragg, G. M., Kingston, D. G. I., and Newman, D. J., Eds.) pp 383-406, CRC Press,
Boca Raton, FL.
(5)
(6)
Lown, J. W. (1993) Discovery and development of anthracycline antitumour antibiotics.
Chem. Soc. Rev. 22, 165-176.
Minotti, G., Menna, P., Salvatorelli, E., Cairo, G., and Gianni, L. (2004) Anthracyclines:
molecular advances and pharmacologic developments in antitumor activity and
cardiotoxicity. Pharmacol. Rev. 56, 185-229.
(7)
Weiss, R. B. (1992) The anthracyclines: will we ever find a better doxorubicin? Sem.
Oncol. 19, 670-686.
(8)
(9)
Hill, S. (2015), World Health Organization.
Carvalho, F. S., Burgeiro, A., Garcia, R., Moreno, A. J., Carvalho, R. A., and Oliveira, P.
J. (2014) Doxorubicin-induced cardiotoxicity: From bioenergetic failure and cell death to
cardiomyopathy. Med. Res. Rev. 34, 106-135.
(10) Menna, P., Salvatorelli, E., Gianni, L., and Minotti, G. (2008) Anthracycline
cardiotoxicity. Top. Curr. Chem. 283, 21-44.
(11) Barenholz, Y. (2012) Doxil® — The first FDA-approved nano-drug: Lessons learned. J.
Controlled Release 160, 117-134.
(12) Batist, G., Barton, J., Chaikin, P., Swenson, C., and Welles, L. (2002) Myocet (liposome-
encapsulated doxorubicin citrate): a new approach in breast cancer therapy. Expert Opin.
Pharmacother. 3, 1739-1751.
(13) Chen, G., Li, D., Li, J., Cao, X., Wang, J., Shi, X., and Guo, R. (2015) Targeted
doxorubicin delivery to hepatocarcinoma cells by lactobionic acid-modified laponite
nanodisks. New J. Chem. 39, 2847-2855.
(14) Joniec, A., Sek, S., and Krysinski, P. (2016) Magnetoliposomes as potential carriers of
doxorubicin to tumours. Chemistry 22, 17715-17724.
28
ACS Paragon Plus Environment

Page 29 of 31
Bioconjugate Chemistry
1
2
3
4
5
6
7
8
(15) Liu, W., Wen, S., Shen, M., and Shi, X. (2014) Doxorubicin-loaded Poly(lactic-co-
glycolic acid) Hollow Microcapsules for Targeted Drug Delivery to Cancer Cells. New J.
Chem. 38, 3917-3924.
(16) Maeda, H. (2012) Macromolecular therapeutics in cancer treatment: The EPR effect and
beyond. J. Controlled Release 164, 138-144.
9
(17) Wilhelm, S., Tavares, A. J., Dai, Q., Ohta, S., Audet, J., Dvorak, H. F., and Chan, W. C.
W. (2016) Analysis of nanoparticle delivery to tumours. Nat. Rev. Mater. 1, 16014.
(18) Torrice, M. (2016) Does nanomedicine have a delivery problem? Chem. Eng. News 94
(25), 16-19.
(19) Kumar, A., Zhang, X., and Liang, X.-J. (2013) Gold nanoparticles: emerging paradigm
for targeted drug delivery system. Biotechnol. Adv. 31, 593-606.
(20) Mirza, A. Z., and Shamshad, H. (2011) Preparation and characterization of doxorubicin
functionalized gold nanoparticles. Eur. J. Med. Chem 346, 1857-1860.
(21) Chaudhary, A., Dwivedi, C., Gupta, A., and Nandi, C. K. (2015) One pot synthesis of
doxorubicin loaded gold nanoparticles for sustained drug release. RSC Adv. 5, 97330-
97334.
(22) Khandekar, S. V., Kulkarni, M. G., and Devarajan, P. V. (2014) Polyaspartic acid
functionalized gold nanoparticles for tumor targeted doxorubicin delivery. J. Biomed.
Nanotechnol. 10, 143-153.
(23) Alexander, C. M., Hamner, K. L., Maye, M. M., and Dabrowiak, J. C. (2014)
Multifunctional DNA-gold nanoparticles for targeted doxorubicin delivery. Bioconjugate
Chem. 25, 1261-1271.
(24) Zhang, Z.-M., Gao, P.-C., Wang, Z.-F., Sun, B.-W., and Jiang, Y. (2015) DNA-caged
gold nanoparticles for controlled release of doxorubicin triggered by a DNA enzyme and
pH. Chem. Commun. 51, 12996-12299.
(25) Spadavecchia, J., Movia, D., Moore, C., Maguire, C. M., Moustaoui, H., Casale, S.,
Volkov, Y., and Prina-Mello, A. (2016) Targeted polyethylene glycol gold nanoparticles
for the treatment of pancreatic cancer: from synthesis to proof-of-concept in vitro studies.
Int. J. Nanomed. 11, 791-822.
(26) Swiech, O. A., Opuchlik, L. J., Wojciuk, G., Stepkowski, T. M., Kruszewski, M., and
Bilewicz, R. (2016) Doxorubicin carriers based on Au nanoparticles - effect of shape and
gold-drug linker on the carrier toxicity and therapeutic performance. RSC Adv. 6, 31960-
31967.
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
(27) Asadishad, B., Vossoughi, M., and Alemzadeh, I. (2010) Folate-receptor-targeted
delivery of doxorubicin using polyethylene glycol-functionalized gold nanoparticles. Ind.
Eng. Chem. Res. 49, 1958-1963.
(28) Cheng, J., Gu, Y.-J., Cheng, S. H., and Wong, W.-T. (2013) Surface functionalized gold
nanoparticles for drug delivery. J. Biomed. Nanotech. 9, 1362-1369.
(29) Gu, Y.-J., Cheng, J., Man, C. W.-Y., Wong, W.-T., and Cheng, S. H. (2012) Gold-
doxorubicin nanoconjugates for overcoming multidrug resistance. Nanomedicine:
Nanotechnology, Biology and Medicine 8, 204-211.
(30) Aryal, S., Grailer, J. J., Pilla, S., Steeber, D. A., and Gong, S. (2009) Doxorubicin
conjugated gold nanoparticles as water-soluble and pH-responsive anticancer drug
nanocarriers. J. Mater. Chem. 19, 7879-7884.
29
ACS Paragon Plus Environment