Chemical Synthesis and Biological Evaluations of Adiponectin Collagenous Domain Glycoforms

Article abstract of DOI:10.1021/jacs.1c02382

The homogeneously glycosylated 76-amino acid adiponectin collagenous domains (ACDs) with all of the possible 15 glycoforms have been chemically and individually synthesized using stereoselective glycan synthesis and chemical peptide ligation. The following biological and pharmacological studies enabled correlating glycan pattern to function in the inhibition of cancer cell growth as well as the regulation of systemic energy metabolism. In particular, hAdn-WM6877 was tested in detail with different mouse models and it exhibited promising in vivo antitumor, insulin sensitizing, and hepatoprotective activities. Our studies demonstrated the possibility of using synthetic glycopeptides as the adiponectin downsized mimetic for the development of novel therapeutics to treat diseases associated with deficient adiponectin.

Full text of DOI:10.1021/jacs.1c02382

pubs.acs.org/JACS
Article
Chemical Synthesis and Biological Evaluations of Adiponectin
Collagenous Domain Glycoforms
Hongxiang Wu,§ Yiwei Zhang,§ Yuanxin Li,§ Jianchao Xu, Yu Wang,* and Xuechen Li*
Cite This: J. Am. Chem. Soc. 2021, 143, 7808−7818
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: The homogeneously glycosylated 76-amino acid adiponectin
collagenous domains (ACDs) with all of the possible 15 glycoforms have been
chemically and individually synthesized using stereoselective glycan synthesis
and chemical peptide ligation. The following biological and pharmacological
studies enabled correlating glycan pattern to function in the inhibition of
cancer cell growth as well as the regulation of systemic energy metabolism.
In particular, hAdn-WM6877 was tested in detail with different mouse
models and it exhibited promising in vivo antitumor, insulin sensitizing, and
hepatoprotective activities. Our studies demonstrated the possibility of using
synthetic glycopeptides as the adiponectin downsized mimetic for the development
of novel therapeutics to treat diseases associated with deficient adiponectin.
trimers and is less active than the full-length adiponectin,
especially in its insulin-sensitizing and hepatoprotective func-
tions.21,22 These results indicate the importance of the glyco-
sylation and collagenous domain for adiponectin’s function.
Thus, exploration of the glycosylated adiponectin collagenous
domain is highly demanded. However, the biofunction of the
glycosylated collagenous domain and the role of glycosylation
position and degree remain elusive due to its unavailability.
The mammalian adiponectin collagenous domain contains
four 5-(2S,5R)-hydroxyl lysine residues (at 65, 68, 77, 101)
that are glycosylated with a glucosyl-galactose disaccharide
(Figure 1).12,13 The glycosylated domain cannot be accessed
using the DNA recombinant process, which hampers the detailed
study of this domain’s function. In addition, two-dimensional
gels showed multiple adiponectin isoforms,23−25 which are likely
resulting from heterogeneous glycoforms, while no information
is available on their relative distribution.
In general, heterogeneity issues associated with proteins
with post-translational modifications make it difficult to correlate
structure to function at a molecular level. Protein chemical
synthesis provides a means to generate homogeneous proteins
with site-specific and structure-defined natural or unnatural
modification(s).34 To fill in this knowledge gap, we undertook a
chemical approach to generate the homogeneously glycosylated
INTRODUCTION
■
Adiponectin is a circulating glycoprotein mainly produced
from adipocytes.1−4 It is a key regulator of glucose and lipid
metabolism in peripheral organs such as skeletal muscle and
liver, thus increasing the systemic insulin sensitivity and energy
homeostasis.5−7 A decrease in the production of adiponectin
is involved in the pathogenesis of insulin resistance, type 2
diabetes, steatohepatitis, cardiovascular diseases, and certain
types of cancers.8−10 In this context, adiponectin supplementation
could be a highly beneficial approach with therapeutic potentials
for metabolic, cancer, and cardiovascular diseases. Nevertheless,
adiponectin-based therapeutics are not possible at present,
mainly due to the difficulty in obtaining the full-length glycosylated
human adiponectin.11
Human adiponectin is a polypeptide with 244 amino acids
and comprises four structurally distinct domains from NH2- to
CO2H-terminus: a signal peptide, a variable region, a colla-
genous domain, and a globular domain (Figure 1). The colla-
genous domain is glycosylated with 2-O-α-D-glucopyranosyl-
D-galactose disaccharide via hydroxylysine.12,13 The glycan
structure is very different from the common O-linked or
N-linked glycoproteins, which represents a huge challenge to
convert the full-length human adiponectin protein into a viable
drug via recombinant approaches. Although efforts have been
made to identify a minimal structure capable of eliciting the
required pharmacological agonist activities, the structure−activity
relationships of the individual domain within adiponectin
have not been well-defined.14−19
Received: March 3, 2021
Published: May 12, 2021
Full-length adiponectin produced from expression in
bacteria has been shown to be biologically inactive, due to the
lack of glycosylation.15,20 The globular domain forms exclusively
https://doi.org/10.1021/jacs.1c02382
J. Am. Chem. Soc. 2021, 143, 7808−7818
© 2021 American Chemical Society
7808

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
a
Scheme 1. Synthesis of the (2S,5R)-Hydroxylysine Building Block
a
Reagents and conditions: (a) BnBr, KHCO3, DMF, 4 h, 75%. (b) NaIO4, THF/MeOH/H2O, 0 °C, overnight, 90%. (c) Xylene, reflux, 2 d, 67%.
(d) DCC, EtSH, DMAP, 90%. (e) 10% Pd/C, Et3SiH, 3 h, 86%. (f) 1 M vinylmagnesium bromide, THF, −30 °C, 3 h, 55%. (g) Boc-L-Ala-OH,
EDCI, DMAP, DCM. (h) 1 M LiOH, THF/H2O, 1 h, 85%. (i) DIAD, triphenylphosphine, p-nitrobenzoic acid, THF, 0 °C, 1 h. (j) Grubbs II, CuI,
Et2O, overnight, 60%. (k) 10% Pd/C, EA, 5 h; BnBr, DMF, KHCO3, 3 h, 75%.
adiponectin collagenous domains with site-specific glycan(s).
Herein, we report our efforts to develop a facile strategy to
chemically synthesize adiponectin collagenous domains with
all possible 15 glycoform combinations. This achievement has
allowed us to define the biological role of the glycosylated
adiponectin collagenous domains and further to correlate the
glycosylation pattern to function.
first task was to address the issue associated with a large-scale
synthesis of the (2S,5R)-hydroxylysine building block. Although
several synthetic routes have been previously reported, none
of them could be likely applicable for the large-scale synthesis.26−28
Our synthesis is shown in Scheme 1. The key step of our
synthetic strategy was the reaction between L-α-vinylglycine
1.3 and 1-amino-3-butene-2-ol derivative 2.3-1 via olefin
metathesis. The carboxylic acid of Fmoc-Met-OH was first
masked by a Bn group to give 1.1, which was converted to
the 1.2 by NaIO4 oxidation. This intermediate underwent
pyrolysis to afford the corresponding product 1.3 in overall
RESULTS AND DISCUSSION
■
Synthesis of the Hydroxyl Lysine Building Block.
Prior to the efforts for adiponectin chemical synthesis, the
7809
https://doi.org/10.1021/jacs.1c02382
J. Am. Chem. Soc. 2021, 143, 7808−7818

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Scheme 2. Retrosynthetic Analysis of Glycosylated Hydroxylysine
a
Scheme 3. Synthesis of Glycosylated Hydroxylysine
a
Reagents and conditions: (a) See Supporting Information. (b) AgOTf, PhSCl, TTBP, DCM/CH3CH2CN = 1:1, −78 °C, 2 h, 41%, only β. (c) DDQ,
DCM/H2O = 18:1, rt. (d) MeONa, MeOH/DCM = 1:1, 5 h, 90%. (e) AgOTf, PhSCl, DMF, DCM, AW-300 molecular sieves, −15 °C, overnight, 53%,
only α. (f) AgOTf, PhSCl, TTBP, DCM/CH3CH2CN = 1:1, −78 °C, 2 h, 43%, only β. (g) LiOH, CaCl2, THF, i-PrOH, overnight, 51%.
7810
https://doi.org/10.1021/jacs.1c02382
J. Am. Chem. Soc. 2021, 143, 7808−7818

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 2. Synthesis of glycopeptide hAdn-WM77. (A) Synthetic route for hAdn-WM-77: (a) CH3CN/H2O/diethylamine (4.5/4.5/1, v/v/v), rt,
2 h, lyophilization; (b) hAdn-WM-a, pyridine/acetic acid (1/1, mol/mol), rt, 4 h. (B) UPLC−MS trace of purified hAdn-WM77. (C) ESI-MS
spectrum of purified hAdn-WM77. ESI-MS calcd for C302H470N92O106S, [M + 4H]4+ m/z = 1780.4, found 1780.4; [M + 5H]5+ m/z = 1424.5,
found 1424.6; [M + 6H]6+ m/z = 1187.3, found 1187.3; [M + 7H]7+ m/z = 1017.8, found 1017.7; [M + 8H]8+ m/z = 890.7, found 890.8;
[M + 9H]9+ m/z = 791.9, found 792.0; [M + 10H]10+ m/z = 712.8, found 712.6; [M + 11H]11+ m/z = 648.1, found 648.4.
a
Scheme 4. Synthesis of glyACD Glycoforms
a
WM: The first two amino acids at the N-terminus were used to label compounds.
50% yield. On the other hand, Boc-Gly thioester 2.1 was
subjected to the Fukuyama reduction,29 and the resultant
aldehyde 2.2 underwent a vinylmagnesium addition to form a
racemic alcohol 2.3. Subsequently, we used diastereomeric
resolution via formation of diastereomeric esters 2.4-1 and
2.4-2, which could be readily separated and purified by silica
gel column. The ester compound 2.4-1 was treated with
LiOH to regenerate the optical pure product 2.3-1, allowing
cross olefin metathesis30 via the Grubbs’ catalyst with compound
1.3 to produce the precursor 3, which underwent hydrogenation
and reinstallation of Bn group to obtain (2S,5R)-hydroxylysine
building block 4 (HLBB) with orthogonal protecting groups.
Moreover, the undesired diastereomer 2.4-2 with inverse
chiral center could be transformed into the desired one. This
synthetic route can be efficiently performed up to 10 g.
Synthesis of the Glycosylated Hydroxylysine. The
second challenge is the scalable preparation of the glycosylated
hydroxylysine. According to the retrosynthetic analysis
(Scheme 2), the designed glycosylated hydroxylysine building
block could be synthesized via a stepwise approach (route 1)
or convergent approach (route 2). The protecting groups for
the disaccharide need to be compatible with Fmoc-solid phase
peptide synthesis (SPPS). Benzyl group (Bn) was first used,
which, however, was found to be very difficult to remove
from the final glycopeptides. In the end, a more acid-labile
group, 4-methylbenzyl (MBn), was used.
We first tested out the feasibility of the straightforward
linear synthesis (route 1) (Scheme 3). Galactosyl donor 5.1
with 2-Ac protecting group to promote β-selectivity via
neighboring group participation effect was employed to glyco-
sylate compound 4. However, the glycosylation product 6.1
was not observed due to the formation of the corresponding
ortho-ester. Several glycosyl donors and glycosylation conditions
(see Supporting Information) were attempted to overcome
7811
https://doi.org/10.1021/jacs.1c02382
J. Am. Chem. Soc. 2021, 143, 7808−7818

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
this synthetic obstacle, yet there was not any improvement to
suppress the ortho-ester formation. To overcome this obstacle,
p-methoxybenzyl (PMB) was used as in compound 5.2.
Without the neighboring group participation, the β-1,2-trans-
O-glycosylation was achieved with assistance of the solvent
effect. After screening of several glycosylation conditions, the
galactosylated hydroxylysine 6.2 was obtained in 42% yield
with only the β-glycoside formed. This promising result not
only allowed for the next step synthesis but also demon-
strated the feasibility of stereoselective glycosylation of the
ε-amine Boc protected 5-hydroxylysine. To continue the
second glycosylation, we attempted the PMB deprotection of
6.2 to afford compound 7. Unfortunately, the PMB deprotection
reaction was messy. Likely, the MBn group might be sensitive
to oxidants and incompatible with PMB groups (Scheme 3).
With this failure, we turned to the convergent approach
(route 2) (Scheme 3). The challenge of this route lies in the
final β-1,2-trans-glycosylation between the disaccharide and
the hydroxylsine building block, considering its steric hindrance
and lack of β-selectivity driving factor plagued with a glycoside
at the C-2 position. To this end, the 5.1 underwent a hydrolysis
to generate a 2-OH free product 5.1a in quantitative yield,
which was subjected as the glycosyl acceptor to react with
thioglycoside 8 for disaccharide moiety construction. The key
point was how to selectively introduce an α-1,2-cis glucosyl
bond. Fortunately, when the DMF/DCM solvent system31
and PhSCl/AgOTf initiator32 were employed, the desired
disaccharide thioglycoside with α-O-glycosyl linkage could be
obtained in good yield with exclusive α-stereoselectivity. Subse-
quently, the disaccharide thioglycoside moiety 9 was successfully
glycosylated with the HLBB 4 under AgOTf/PhSCl/TTBP
conditions, and the usage of DCM/CH3CH2CN solvent system
was the key to induce exclusive β-1,2-trans-glycosylation affording
the desired product 10. Next, a Fmoc-compatible ester hydrolysis
protocol (LiOH, CaCl2)33 was employed to afford the building
block 11. This is a convenient and scalable synthetic route,
which permitted the synthesis of MBn protected α-D-glucopyr-
anosyl-(1−2)-β-galactopyranosyl (2S,5R)-hydroxylysine building
block for the glycopeptide fragment preparation (Scheme 3).
Synthesis of Glycosylated Adiponectin Collagenous
Domain (glyACD). With all the building blocks in hand, we
continued with the total synthesis of glyACD. The attempt
with direct SPPS failed to generate any desired product.
After several trials, we finalized the synthesis via ligating two
peptide fragments via Ser/Thr ligation (Figure 2).34−37 The
47-amino acid-glycopeptide hAdn-WM77-b was prepared
using SPPS, followed by TMSOTf treatment to remove the
MBn groups present on the glycans, affording hAdn-WM77-b
after HPLC isolation (8.9% yield based on the resin loading).
It is worth noting that the replacement of Bn protecting groups
of the disaccharide to MBn protecting groups was critical, as
all MBn groups could be cleanly removed without cleaving
the glycosidic linkage. The resultant glycopeptide underwent
deFmoc conditions, which then was, without HPLC purification,
directly ligated with peptide salicylaldehyde ester hAdn-WM-a
under Ser/Thr ligation conditions in 30.8% yield over two steps
after HPLC isolation. With the same strategy, the syntheses
of glyACD with all possible glycoforms (15 in total) have
been successfully completed (Scheme 4).
Table 1. Comparison of the Emax and EC50 for the
Antiproliferative Activity of glyACD in Human Breast
Cancer MDA-MB-231 Cells
a
After 24 h of incubation, the cell number was manually counted to
b
calculate the maximum inhibition rate (Emax). Half-maximal effective
concentration. This peptide contains a tiny amount of small
c
molecular weight impurity. hAdn: human adiponectin.
for details of the assay). After 24 h of incubation, the cell
number was manually counted to calculate the maximum inhi-
bition rate (Emax) and the half-maximal effective concentra-
tion (EC50) (Table 1). When compared to those of full-length
human adiponectin, hAdn-WM exhibited a significantly reduced
Emax (by ∼3-fold) and a much higher EC50 (by ∼19-fold).
The Emax of glyACD ranged from ∼26.0% to ∼53%, which
was lower than that of full-length adiponectin but higher than
hAdn-WM. Among all glyACD, hAdn-WM656877101 with
tetraglycans exhibited the lowest EC50 (Table 1).
The glyACD with mono-, di-, and triglycans exhibited lower
Emax than that of hAdn-WM656877101, which contained
tetraglycans (Figure 3A). There were no significant differences
between the EC50 of glyACD containing mono-, di-, and
triglycans, all significantly higher than that of hAdn-
WM656877101 (Figure 3B). In the presence of 20 μg/mL
hAdn-WM, the Emax of human adiponectin was significantly
decreased (66.7
4.29% vs 54.2
2.95%%, P < 0.05). By
contrast, hAdn-WM656877101 did not inhibit the Emax of
human adiponectin (Figure 3C). In the chessboard assay,
co-incubation with 5, 10, or 20 μg/mL of hAdn-WM reduced
the antiproliferative activity of adiponectin at a concentration
from 0.74 to 2.5 μg/mL (Figure 3D), indicating antagonistic
effects (combination index of >1). By contrast, co-incubation
with 2.5, 5, 10, or 20 μg/mL of hAdn-WM656877101 enhanced
the antiproliferative activity of adiponectin at a concentration
lower than 2.5 μg/mL (Figure 3D), suggesting that the
combination generated synergistic effects (combination index
of <1). The chessboard competition assay was performed to
Antiproliferative Activity of the glyACD. The human
breast cancer MDA-MB-231 cells were used to evaluate the
antiproliferative activities of adiponectin, the 15 glyACD, and
nonglycosylated hAdn-WM peptides (see Supporting Information
7812
https://doi.org/10.1021/jacs.1c02382
J. Am. Chem. Soc. 2021, 143, 7808−7818

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 3. GlyACD and adiponectin elicit synergistic antiproliferative activity in human MDA-MB-231 cells. After serum starvation for 24 h,
MDA-MB-231 cells were treated with different concentrations of hAdn-WM, glyACD, full-length human adiponectin, or their combinations.
After incubation for another 24 h, cells were harvested for manual counting. (A) The Emax of adiponectin or peptides with no glycans
(hAdn-WM) and the average Emax of each mono-, di-, tri-, and tetraglycan variants were calculated for comparison. (B) The EC50 of adiponectin
or peptides with no glycans (hAdn-WM) and the average EC50 of each mono-, di-, tri-, and tetraglycan variants were calculated for comparison.
(C) The MDA-MB-231 cells were treated with different concentrations of adiponectin in the absence or presence of low (2.5 μg/mL) or high
(20 μg/mL) doses of hAdn-WM or hAdn-WM656877101. The percentage inhibition on cell growth was calculated for comparison. (D) The
MDA-MB-231 cells were treated with 0, 0.22, 0.33, 0.49, 0.74, 1.11, 1.67, 2.5 μg/mL adiponectin in combination with 0, 0.3125, 0.625, 1.25,
2.5, 5, 10, 20 μg/mL hAdn-WM or hAdn-WM656877101. The percentage inhibition on cell growth was calculated and indicated by different
◆
intensity of the blue color. Data are presented as the mean SEM (n = 6): ∗, P < 0.05 vs full-length human adiponectin; , P < 0.05 vs
▲
peptide with no glycans (hAdn-WM); , P < 0.05 vs peptide with tetraglycans (hAdn-WM656877101). The significant differences between
groups were analyzed by t test or two-way ANOVA (GraphPad Prism 8.0.2, GraphPad Software, Inc., San Diego, CA, USA).
mimic the in vivo conditions, in which the adiponectin levels
vary among different individuals. This result provides important
information on the optimal dose for the best synergistic effects.
Moreover, the nonglycosylated peptide hAdn-WM inhibits the
antiproliferative activity of the hAdn, suggesting the essentiality
of the glycan.
regular basis. Compared to the vehicle group, tumor development
was significantly attenuated in mice implanted with MDA-MB-231
cells pretreated with hAdn-WM6877 (Figure 4A,B). Body
weight did not show significant differences between the vehicle
and treatment groups. However, the percentage tumor weights
were significantly decreased in both nude and NOD/Scid mice
implanted with MDA-MB-231 cells pretreated with hAdn-
WM6877 (Figure 4A,B).
The transgenic MMTV PyVT mice lacking the Adipoq alleles
(PyVT-AKO) exhibit spontaneous mammary tumor development
starting from the age of 7 or 8 weeks.38 The PBS, hAdn-WM,
or hAdn-WM6877 (40 (μg/mouse)/day) was intraperitoneally
injected into PyVT-AKO mice from the age of 8 weeks. Tumor
development was monitored on a weekly basis. Treatment with
hAdn-WM6877 significantly inhibited mammary tumor develop-
ment in PyVT-AKO mice (Figure 4C). After 5 weeks of treat-
ment, tumors were collected for examination. When compared to
the vehicle group, the tumor-to-body weight ratios were
significantly decreased (by over 2-fold) in PyVT-AKO mice
treated with hAdn-WM6877 (Figure 4C). Collectively, the above
results demonstrate that the diglycan peptide hAdn-WM6877
elicited potent anti-breast-cancer activity in mouse models.
Metabolic Functions of the hAdn-WM6877 glyACD.
The adiponectin knockout (AKO) mice of the C57BL/6J
Anti-Breast-Cancer Activity of the hAdn-WM68
glyACD. Among all variants, the tetraglycan WM656877101
demonstrates the highest inhibition rate and the lowest EC50,
but the overall yield of its chemical synthesis is the lowest,
requiring a large amount of the glycosylated hydroxylysine
building block. Thus, considering the synthetic efficiency and
overall bioactive performances of Emax and EC50 of all the
variants, the diglycan peptide hAdn-WM6877 was selected
and synthesized to a larger amount (20 mg) for examining
the anti-breast cancer activity (see Supporting Information for
details). As the tumor can be generated with subcutaneous
injection of tumor cell into immunocompromised mice, Nude
and Nod/Scid mice were adopted for in vivo study. First,
the MDA-MB-231 cells treated with phosphate buffered
saline (PBS), hAdn-WM, or hAdn-WM6877 for 24 h were
implanted orthotopically into the third right mammary fat pad
of athymic nude (Figure 4A) or NOD/Scid mice (Figure 4B).
The development of mammary tumors was monitored on a
7813
https://doi.org/10.1021/jacs.1c02382
J. Am. Chem. Soc. 2021, 143, 7808−7818

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 4. Anti-breast-cancer activity of the hAdn-WM6877 glyACD in different mouse models. (A) Human breast cancer MDA-MB-231 cells
were incubated with phosphate buffer saline (PBS), 20 μg/mL hAdn-WM or hAdn-WM6877 for 24 h. A total number of 5 × 106 cells
were harvested from each treatment group and injected into the third right mammary fat pad of nude mice. Tumor size was measured by
vernier caliper every 2 or 3 days and calculated as volumes for comparison (left). Mice were sacrificed after 4 weeks. Tumors were collected
to examine the wet tissue weights. The results were presented as the tumor-to-body weight ratios for comparison (middle and right).
(B) Human breast cancer MDA-MB-231 cells were incubated with PBS, 20 μg/mL hAdn-WM or hAdn-WM6877 for 24 h. A total
number of 2 × 105 cells were harvested from each treatment group and injected into the third right mammary fat pad of NOD/Scid mice.
Tumor development was monitored as in panel A for comparison (left). Mice were sacrificed after 30 days. Tumors were collected
to calculate the tissue-to-body weight ratios for comparison (middle and right). (C) Equal volume of PBS, hAdn-WM, or hAdn-WM6877
(40 (μg/mouse)/day) was intraperitoneally injected into PyVT-AKO mice. Tumor development was monitored weekly (left). Mice
were sacrificed after 5 weeks of treatment. Tumors were collected to compare the tissue-to-body weight ratios (middle and right). Data are
presented as the mean
analyzed by t test or two-way ANOVA.
SEM (n = 6): ∗, P < 0.05 vs corresponding vehicle controls. The significant differences between groups were
background was used to study the metabolic functions of
hAdn-WM6877. AKO mice were given a high-fat diet (HFD)
for 12 weeks, starting from the age of 4 weeks. The PBS or
hAdn-WM6877 peptide (40 (μg/mouse)/day) was administered
intraperitoneally during the last 4 weeks of HFD treatment.
Compared to the vehicle control group, daily injection with
hAdn-WM6877 significantly attenuated the gain of body weight
and body fat mass (Figure 5A). After 4 weeks of injection, mice
treated with hAdn-WM6877 exhibited significantly improved
glucose and insulin tolerance (Figure 5B).
The metabolic performance was evaluated by the Compre-
hensive Laboratory Animal Monitoring System (CLAMS; see
Supporting Information for details). Compared to those of
the vehicle controls, treatment with hAdn-WM6877 significantly
increased oxygen consumption (VO2) during both dark (2345.0
102.9 vs 2689.1
(2070.7
142.2 (mL/kg)/h, P < 0.05) and light
33.6 vs 2416.9
142.7 (mL/kg)/h, P < 0.05)
cycles (Figure 5C, left) and increased carbon dioxide
release (VCO2) during dark (2077.1 68.0 vs 2320.2
74.1 (mL/kg)/h, ∗, P < 0.05) cycles (Figure 5C, middle).
7814
https://doi.org/10.1021/jacs.1c02382
J. Am. Chem. Soc. 2021, 143, 7808−7818

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 5. Metabolic functions of the hAdn-WM6877 glyACD in dietary obese mice. The adiponectin deficient (AKO) mice of the C57BL/6J
background were fed a high fat diet (HFD) starting from the age of 4 weeks. After 8 weeks of HFD, equal volume of PBS or hAdn-WM6877
(40 (μg/mouse)/day) was intraperitoneally injected into AKO mice for another 4 weeks. (A) Body weight and fat mass composition were
recorded weekly and calculated for comparison. The results are presented as fold changes against the starting points (before injection).
(B) After 4 weeks of treatment, intraperitoneal glucose (left) and insulin (right) tolerance tests were performed as described in the methods
for comparison. (C) The indirect calorimetry was performed to examine the oxygen consumption (VO2), carbon dioxide production VCO2,
and respiratory exchange ratio (RER) for comparison. (D) At the end of treatment, mice were sacrificed after fasting with food removal for
16 h. Serum was collected to measure circulating levels of glucose, insulin, triglyceride, and cholesterol for comparison. Data are presented as
the mean SEM (n = 6): ∗, P < 0.05 vs corresponding vehicle controls. Differences between mean values were determined using two-way
ANOVA with Fisher’s LSD test (panels A, B, and C). The significant differences between groups were analyzed by t test or two-way ANOVA.
As a result, the respiratory exchange ratio (RER) during
the light cycle was significantly lower in mice treated with
hAdn-WM6877 than those of the vehicle controls (0.84
(Figure 6B, right). When compared to the vehicle control
animals, the mRNA expression levels of steatohepatitis-associated
genes, including those encoding tumor necrosis factor α (TNFA),
monocyte chemoattractant protein-1 (CCL2), low-density
lipoprotein receptor (LDLR), and collagen types I (COL1)
and VI (COL6), were significantly decreased in livers of mice
treated with hAdn-WM6877 (Figure 6C). In summary, the
results suggest that hAdn-WM6877 mimicked adiponectin to
elicit insulin-sensitizing, anti-inflammatory, and hepatoprotective
functions in AKO mice challenged with HFD.
0.04 vs 0.92
0.02, P < 0.01). The fasting serum levels of
triglyceride and total cholesterol were also significantly decreased
in mice treated with hAdn-WM6877 (Figure 5D).
H&E and Oil Red O staining revealed that daily treatment
with hAdn-WM6877 significantly reduced the lipid accumulation
in livers of HFD-fed AKO mice (Figure 6A). The triglycerides
and cholesterol contents were decreased by ∼51.9% and
∼35.4% in livers of mice treated with WM6877 when compared
to the vehicle control animals (Figure 6B, left).
The circulating levels of liver injury markers, alanine trans-
aminase (ALT) and aspartate aminotransferase (AST), were
both significantly decreased by treatment with hAdn-WM6877
CONCLUSION
■
Adiponectin is a circulating hormone produced abundantly
from adipose tissue and possesses many therapeutic potentials
for metabolic, cancer, and cardiovascular diseases. However,
7815
https://doi.org/10.1021/jacs.1c02382
J. Am. Chem. Soc. 2021, 143, 7808−7818

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 6. HAdn-WM6877 glyACD alleviates HFD-induced fatty liver injuries. The adiponectin deficient (AKO) mice of the C57BL/6J
background were fed a high fat diet (HFD) starting from the age of 4 weeks. After 8 weeks of HFD, equal volume of PBS or hAdn-WM6877
(40 (μg/mouse)/day) was intraperitoneally injected into AKO mice for another 4 weeks. At the end of treatment, mice were sacrificed after
fasting with food removal for 16 h. (A) Liver tissue sections were prepared for hematoxylin and eosin [H&E] and Oil Red O (left) staining
to evaluate the accumulation and distribution of lipid droplets. (B) Triglyceride and cholesterol contents in liver samples were examined by
biochemical assays (left). ALT/AST in blood circulation was examined by biochemical assays (right). (C) Quantitative PCR (QPCR) was
performed to measure the gene expression levels of TNFA, CCL2, LDLR, COL1, and COL6 in liver tissue. Data are presented as the mean
SEM (n = 6): ∗, P < 0.05 vs corresponding vehicle controls (n = 6). Differences between mean values were determined using the unpaired
Student’s t tests.
biochemical and pharmacological studies of the human adipo-
nectin correlating the domain structure to function have been
restricted by protein heterogeneity and difficulty in obtaining
the homogeneous collagenous domain with site-specific modi-
fication(s). In this study we have developed an accessible
and scalable synthetic route of the glycosylated adiponectin
collagenous domain by using stereoselective glycan synthesis
and chemical peptide ligation, which led to the achievement
of 15 homogeneously glycosylated variants of ACD for the
first time. Subsequently, the biological activities and pharmaco-
logical properties of the glycosylated adiponectin peptides have
been evaluated and compared with the full-length human
adiponectin. The results demonstrate that the glycan plays
an important role of the collagenous domain of adiponectin
in the inhibition of cancer cell growth as well as the regulation
of systemic energy metabolism.
activity, enhanced fatty acid consumption, and improved glucose
tolerance and insulin sensitivity, which opens the door to
explore the opportunity of using the synthetic glycopeptide as
a potential adiponectin downsized mimic supplementary in
metabolic diseases treatment, and the extensive knowledge in
adiponectin glycosylation might trigger the development of
new drug and mechanism studies.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/jacs.1c02382.
■
sı
Full details of chemical synthesis, protocols of bio-
logical experiments, NMR spectra, and UPLC−MS
traces (PDF)
The key feature of this work is to demonstrate the power
of chemical synthesis of glycosylated adiponectin collagen
domain in systematically addressing the role of glycosylation
on activity and specificity. This study not only shows how
chemistry facilitates the solution to the unsolved biological
issues associated with posttranslational glycosylation at a
molecular level but also paves the way to the total synthesis
of glycosylated full-length human adiponectin for further
therapeutic study. More importantly, our in vivo studies
showed that hAdn-WM6877 exhibited impressive antitumor
AUTHOR INFORMATION
Corresponding Authors
■
Yu Wang − Department of Pharmacology and Pharmacy,
State Key Laboratory of Pharmaceutical Biotechnology, The
University of Hong Kong, Hong Kong, P. R. China;
Email: yuwanghk@hku.hk
Xuechen Li − Department of Chemistry, State Key
Laboratory of Synthetic Chemistry, The University of Hong
Kong, Hong Kong, P. R. China; orcid.org/0000-0001-
5465-7727; Email: xuechenl@hku.hk
7816
https://doi.org/10.1021/jacs.1c02382
J. Am. Chem. Soc. 2021, 143, 7808−7818

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(10) Dalamaga, M.; Diakopoulos, K. N.; Mantzoros, C. S. The role
of adiponectin in cancer: a review of current evidence. Endocr. Rev.
2012, 33, 547−594.
(11) Liu, Y.; Vu, V.; Sweeney, G. Examining the potential of
developing and implementing use of adiponectin-targeted therapeu-
tics for metabolic and cardiovascular diseases. Front. Endocrinol.
2019, 10, 842.
(12) Wang, Y.; Xu, A.; Knight, C.; Xu, L. Y.; Cooper, G. J. S.
Hydroxylation and glycosylation of the four conserved lysine
residues in the collagenous domain of adiponectin. Potential role
in the modulation of its insulin-sensitizing activity. J. Biol. Chem.
2002, 277, 19521−19529.
(13) Wang, Y.; Lu, G.; Wong, W. P. S.; Vliegenthart, J. F. G.;
Gerwig, G. J.; Lam, K. S. L.; Cooper, G. J. S.; Xu, A. Proteomic and
functional characterization of endogenous adiponectin purified from
fetal bovine serum. Proteomics 2004, 4, 3933−3942.
(14) Yamauchi, T.; Kamon, J.; Waki, H.; Terauchi, Y.; Kubota, N.;
Hara, K.; Mori, Y.; Ide, T.; Murakami, K.; Tsuboyama-Kasaoka, N.;
Ezaki, O.; Akanuma, Y.; Gavrilova, O.; Vinson, C.; Reitman, M. L.;
Kagechika, H.; Shudo, K.; Yoda, M.; Nakano, Y.; Tobe, K.; Nagai,
R.; Kimura, S.; Tomita, M.; Froguel, P.; Kadowaki, T. The fat-
derived hormone adiponectin reverses insulin resistance associated
with both lipoatrophy and obesity. Nat. Med. 2001, 7, 941−946.
(15) Berg, A. H.; Combs, T. P.; Du, X.; Brownlee, M.; Scherer, P.
E. The adipocyte-secreted protein Acrp30 enhances hepatic insulin
action. Nat. Med. 2001, 7, 947−953.
(16) Kim, Y.; Lim, J. H.; Kim, M. Y.; Kim, E. N.; Yoon, H. E.;
Shin, S. J.; Choi, B. S.; Kim, Y.; Chang, Y. S.; Park, C. W. The
adiponectin receptor agonist AdipoRon ameliorates diabetic
nephropathy in a model of type 2 diabetes. J. Am. Soc. Nephrol.
2018, 29, 1108−1127.
(17) Abou-Samra, M.; Selvais, C. M.; Dubuisson, N.; Brichard, S.
M. Adiponectin and its mimics on skeletal muscle: insulin
sensitizers, fat burners, exercise mimickers, muscling pills or
everything together. Int. J. Mol. Sci. 2020, 21, 2620−2640.
(18) Sayeed, M.; Gautam, S.; Verma, D. P.; Afshan, T.; Kumari,
T.; Srivastava, A. K.; Ghosh, J. K. A collagen domain-derived short
adiponectin peptide activates APPL1 and AMPK signaling pathways
and improves glucose and fatty acid metabolisms. J. Biol. Chem.
2018, 293, 13509−13523.
(19) Okada-Iwabu, M.; Yamauchi, T.; Iwabu, M.; Honma, T.;
Hamagami, K.; Matsuda, K.; Yamaguchi, M.; Tanabe, H.; Kimura-
Someya, T.; Shirouzu, M.; Ogata, H.; Tokuyama, K.; Ueki, K.;
Nagano, T.; Tanaka, A.; Yokoyama, S.; Kadowaki, T. A small-
molecule AdipoR agonist for type 2 diabetes and short life in
obesity. Nature 2013, 503, 493−499.
Authors
Hongxiang Wu − Department of Chemistry, State Key
Laboratory of Synthetic Chemistry, The University of Hong
Kong, Hong Kong, P. R. China
Yiwei Zhang − Department of Pharmacology and Pharmacy,
State Key Laboratory of Pharmaceutical Biotechnology, The
University of Hong Kong, Hong Kong, P. R. China
Yuanxin Li − Department of Pharmacology and Pharmacy,
State Key Laboratory of Pharmaceutical Biotechnology, The
University of Hong Kong, Hong Kong, P. R. China
Jianchao Xu − Department of Chemistry, State Key
Laboratory of Synthetic Chemistry, The University of Hong
Kong, Hong Kong, P. R. China
Complete contact information is available at:
https://pubs.acs.org/10.1021/jacs.1c02382
Author Contributions
§H.W., Y.Z., and Y.L. contributed equally.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Research Grants Council
Collaborative Research Fund of Hong Kong (Grant C7017-18G),
General Research Fund (Grants 17124420 and 17153016),
and the Area of Excellence Scheme of the University Grants
Committee of Hong Kong (Grants AoE/P-706/16 and AoE/
M-707/18).
ABBREVIATIONS
■
hAdn, human adiponectin; AKO, adiponectin knockout
REFERENCES
■
(1) Scherer, P. E.; Williams, S.; Fogliano, M.; Baldini, G.; Lodish,
H. F. A novel serum protein similar to C1q, produced exclusively in
adipocytes. J. Biol. Chem. 1995, 270, 26746−26749.
(2) Hu, E.; Liang, P.; Spiegelman, B. M. AdipoQ is a novel
adipose-specific gene dysregulated in obesity. J. Biol. Chem. 1996,
271, 10697−10703.
(3) Nakano, Y.; Tobe, T.; Choi-Miura, N. H.; Mazda, T.; Tomita,
M. Isolation and characterization of GBP28, a novel gelatin-binding
protein purified from human plasma. J. Biochem. 1996, 120, 803−
812.
(4) Maeda, K.; Okubo, K.; Shimomura, I.; Funahashi, T.;
Matsuzawa, Y.; Matsubara, K. cDNA cloning and expression of a
novel adipose specific collagen-like factor, apM1 (adipose most
Abundant gene transcript 1). Biochem. Biophys. Res. Commun. 1996,
221, 286−289.
(20) Fruebis, J.; Tsao, T.; Javorschi, S.; Ebbets-Reed, D.; Erickson,
M. R. S.; Yen, F. T.; Bihain, B. E.; Lodish, H. F. Proteolytic cleavage
product of 30-kDa adipocyte complement-related protein increases
fatty acid oxidation in muscle and causes weight loss in mice. Proc.
Natl. Acad. Sci. U. S. A. 2001, 98, 2005−2010.
(21) Shapiro, L.; Scherer, P. E. The crystal structure of a
complement-1q family protein suggests an evolutionary link to
tumor necrosis factor. Curr. Biol. 1998, 8, 335−338.
(22) Yamauchi, T.; Kamon, J.; Ito, Y.; Tsuchida, A.; Yokomizo, T.;
Kita, S.; Sugiyama, T.; Miyagishi, M.; Hara, K.; Tsunoda, M.;
Murakami, K.; Ohteki, T.; Uchida, S.; Takekawa, S.; Waki, H.;
Tsuno, N. H.; Shibata, Y.; Terauchi, Y.; Froguel, P.; Tobe, K.;
Koyasu, S.; Taira, K.; Kitamura, T.; Shimizu, T.; Nagai, R.;
Kadowaki, T. Cloning of adiponectin receptors that mediate
antidiabetic metabolic effects. Nature 2003, 423, 762−769.
(23) Wang, Y.; Lam, K.; Yau, M.; Xu, A. Post-translational
modifications of adiponectin: mechanisms and functional implica-
tions. Biochem. J. 2008, 409, 623−633.
(5) Kadowaki, T.; Yamauchi, T.; Kubota, N.; Hara, K.; Ueki, K.;
Tobe, K. Adiponectin and adiponectin receptors in insulin
resistance, diabetes, and the metabolic syndrome. J. Clin. Invest.
2006, 116, 1784−1792.
(6) Achari, A. E.; Jain, S. K. Adiponectin, a therapeutic target for
obesity, diabetes, and endothelial dysfunction. Int. J. Mol. Sci. 2017,
18, 1321−1337.
(7) Goldstein, B. J.; Scalia, R. G.; Ma, X. L. Protective vascular and
myocardial effects of adiponectin. Nat. Clin. Pract. Cardiovasc. Med.
2009, 6, 27−35.
(8) Li, S.; Shin, H. J.; Ding, E. L.; van Dam, R. M. Adiponectin
levels and risk of type 2 diabetes: a systematic review and meta-
analysis. JAMA 2009, 302, 179−188.
(24) Peake, P. W.; Hughes, J. T.; Shen, Y.; Charlesworth, J. A.
Glycosylation of human adiponectin affects its conformation and
stability. J. Mol. Endocrinol. 2007, 39, 45−52.
(25) Richards, A. A.; Stephens, T.; Charlton, H. K.; Jones, A.;
Macdonald, G. A.; Prins, J. B.; Whitehead, J. P. Adiponectin
(9) Kim, D. H.; Kim, C.; Ding, E. L.; Townsend, M. K.; Lipsitz, L.
A. Adiponectin levels and the risk of hypertension: a systematic
review and meta-analysis. Hypertension 2013, 62, 27−32.
7817
https://doi.org/10.1021/jacs.1c02382
J. Am. Chem. Soc. 2021, 143, 7808−7818

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
multimerization is dependent on conserved lysines in the
collagenous domain: evidence for regulation of multimerization by
alterations in posttranslational modifications. Mol. Endocrinol. 2006,
20, 1673−1687.
(26) Marin, J.; Didierjean, C.; Aubry, A.; Briand, J.-P.; Guichard,
G. Diastereoselective hydroxylation of 6-Substituted piperidin-2-
ones: an efficient synthesis of (2S,5R)-5-hydroxylysine and related
α-Amino Acids. J. Org. Chem. 2002, 67, 8440−8449.
(27) Adamczyk, M.; Johnson, D. D.; Reddy, R. E. Collagen cross-
links: Synthesis of pyridinoline, deoxypyridinoline and their
analogues. Tetrahedron 1999, 55, 63−88.
(28) Herbert, K. R.; Williams, G. M.; Cooper, G. J. S.; Brimble, M.
A. Synthesis of glycosylated 5-hydroxylysine, an important amino
acid present in collagen-like proteins such as adiponectin. Org.
Biomol. Chem. 2012, 10, 1137−1144.
(29) Fukuyama, T.; Lin, S.; Li, L. Facile reduction of ethyl thiol
esters to aldehydes: application to a total synthesis of (+)-neo-
thramycin a methyl ether. J. Am. Chem. Soc. 1990, 112, 7050−7051.
(30) Scholl, M.; Trnka, T. M.; Morgan, J. P.; Grubbs. Increased
ring closing metathesis activity of ruthenium-based olefin metathesis
catalysts coordinated with imidazolin-2-ylidene ligands. Tetrahedron
Lett. 1999, 40, 2247−2250.
(31) Lu, S.; Lai, Y.; Chen, J.; Liu, C.; Mong, K. T.
Dimethylformamide: an unusual glycosylation modulator. Angew.
Chem., Int. Ed. 2011, 50, 7315−7320.
(32) Martichonok, V.; Whitesides, G. M. Stereoselective α-
Sialylation with Sialyl Xanthate and Phenylsulfenyl Triflate as a
Promotor. J. Org. Chem. 1996, 61, 1702−1706.
(33) Capicciotti, C. J.; Trant, J. F.; Leclere, M.; Ben, R. N. Synthe-
sis of C-Linked Triazole-Containing AFGP Analogues and Their
Ability to Inhibit Ice Recrystallization. Bioconjugate Chem. 2011, 22,
605−616.
(34) Zhang, Y.; Xu, C.; Lam, H. Y.; Lee, C. L.; Li, X. Protein
chemical synthesis by serine and threonine ligation. Proc. Natl. Acad.
Sci. U. S. A. 2013, 110, 6657−6662.
(35) Lee, C. L.; Liu, H.; Wong, C. T. T.; Chow, H. Y.; Li, X.
Enabling N-to-C Ser/Thr ligation for convergent protein synthesis
via combining chemical ligation approaches. J. Am. Chem. Soc. 2016,
138, 10477−10484.
(36) Tan, Y.; Wu, H.; Wei, T.; Li, X. Chemical protein synthesis:
advances, challenges and outlooks. J. Am. Chem. Soc. 2020, 142,
20288.
(37) Liu, H.; Li, X. Serine/Threonine ligation: origin, mechanistic
aspects, and applications. Acc. Chem. Res. 2018, 51, 1643−1655.
(38) Lam, J. B. B.; Chow, K. H. M.; Xu, A.; Lam, K. S. L.; Liu, J.;
Wong, N.; Moon, R. T.; Shepherd, P. R.; Cooper, G. J. S.; Wang, Y.
Adiponectin Haploinsufficiency Promotes Mammary Tumor Devel-
opment in MMTV-PyVT Mice by Modulation of Phosphatase and
Tensin Homolog Activities. PLoS One 2009, 4, e4968.
7818
https://doi.org/10.1021/jacs.1c02382
J. Am. Chem. Soc. 2021, 143, 7808−7818