An Fc–Small Molecule Conjugate for Targeted Inhibition of the Adenosine 2A Receptor

Article abstract of DOI:10.1002/cbic.201600337

The adenosine A_2A receptor (A_2AR) is expressed in immune cells, as well as brain and heart tissue, and has been intensively studied as a therapeutic target for multiple disease indications. Inhibitors of the A_2AR have the

Full text of DOI:10.1002/cbic.201600337

Accepted Article
Title: An Fc-Small Molecule Conjugate for Targeted Inhibition of the Adenosine
2A Receptor
Authors: Po-Yuan Hsiao; Jay Kalin; Im-Hong Sun; Mohammed Amin; Ying-Chun
Lo; Meng-Jung Chiang; John Giddens; Polina Sysa-Shah; Kathleen Gabriel-
son; Lai-Xi Wang; Jonathan Powell; Philip A. Cole
This manuscript has been accepted after peer review and the authors have elected to
post their Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by using the Di-
gital Object Identifier (DOI) given below. The VoR will be published online in Early
View as soon as possible and may be different to this Accepted Article as a result
of editing. Readers should obtain the VoR from the journal website shown below
when it is published to ensure accuracy of information. The authors are responsible
for the content of this Accepted Article.
To be cited as: ChemBioChem 10.1002/cbic.201600337
Link to VoR: http://dx.doi.org/10.1002/cbic.201600337
A Journal of
www.chembiochem.org

ChemBioChem
10.1002/cbic.201600337
An Fc-Small Molecule Conjugate for Targeted Inhibition of the Adenosine
2A Receptor
Po-Yuan Hsiao[a], Jay H. Kalin[a], Im-Hong Sun[b], Mohammed N. Amin[c], [d], Ying-Chun
Lo[b], [e], Meng-Jung Chiang[a], [f], John Giddens[c], Polina Sysa-Shah[g], Kathleen Gabrielson[h],
Lai-Xi Wang[c], Jonathan D. Powell[b], Philip A. Cole[a], *
[a] P.Y. Hsiao, Dr. J. H. Kalin, Dr. M. J. Chiang, Prof. Dr. P. A. Cole
Department of Pharmacology and Molecular Sciences,
Johns Hopkins University School of Medicine, Baltimore, MD 21205, U.S.A.
E-mail: pcole1@jhmi.edu
[b] I. H. Sun, Dr. Y. C. Lo, Prof. Dr. J. D. Powell
Sidney-Kimmel Comprehensive Cancer Research Center,
Department of Oncology,
Johns Hopkins School of Medicine, Baltimore, MD 21205, U.S.A.
[c] Dr. M. N. Amin, J. Giddens, Prof. Dr. L. X. Wang
Department of Chemistry and Biochemistry,
University of Maryland, College Park, MD 20742, U.S.A.
[d] Dr. M. N. Amin
Center for Vaccine Development,
University of Maryland School of Medicine, Baltimore, MD 21201, U.S.A.
[e] Dr. Y. C. Lo
Department of Pathology,
Yale School of Medicine, New Haven, CT 06510, U.S.A.
[f] Dr. M. J. Chiang
Division of Viral Products, Office of Vaccines Research and Review,
Center for Biologics Evaluation and Research,
U.S. Food and Drug Administration, Silver Spring, MD 20993, U.S.A.
[g] Dr. P. Sysa-Shah
Russell H. Morgan Department of Radiology and Radiological Science,
Johns Hopkins University, Baltimore, MD 21205, U.S.A.
[h] Prof. Dr. K. Gabrielson
Department of Molecular and Comparative Pathobiology,
Johns Hopkins School of Medicine, Baltimore, MD 21205, U.S.A.
Abstract
The adenosine A2A receptor (A2AR) is expressed in immune cells as well as heart and
lung tissue and has been intensively studied as a therapeutic target for multiple disease
indications. Inhibitors of the A2AR have the potential for stimulating immune response
which could be valuable for cancer immune surveillance and mounting a response against
pathogens. One well-established potent and selective small-molecule A2AR antagonist,
ZM-241385 (ZM), has a short pharmacokinetic half-life and has the potential for systemic
toxicity due to A2AR effects in the brain and the heart. In this study, we designed an analog of
ZM for tethering to the Fc domain of the immunoglobulin IgG3 using expressed protein
ligation. The resulting protein-small molecule conjugate, Fc-ZM, retained high affinity for
both the FcγRI and the neonatal Fc receptor (FcRn). In addition, Fc-ZM was a potent A2AR
antagonist as measured in a cell-based cAMP assay. Cell-based assays also revealed that

ChemBioChem
10.1002/cbic.201600337
Fc-ZM could stimulate interferon γ production in splenocytes in a fashion that was dependent
on the presence of A2AR. We found that Fc-ZM compared with the small molecule ZM was a
superior A2AR antagonist in mice, consistent with the possibility that Fc attachment can
improve pharmacokinetic and/or pharmacodynamic properties of the small molecule.
Keywords
Antibodies. Drug design, Immunology, Inflammation, Inhibitors. Protein design, Protein
modification.
Introduction
A2AR is one of four adenosine receptor subtypes and is a G-protein coupled receptor
(GPCR) that is expressed in immune cells as well as brain and heart[1]. A2AR agonists such
as CGS21680 (CGS) have been explored as immunosuppressive agents but also have been
used in human clinical studies to assess cardiac function[2]. A2AR antagonists like
ZM-241385 (ZM) have been investigated as immunostimulatory agents and have also been
studied as treatments for neurologic disease[3]. There have been two principle challenges with
developing A2AR modulators as therapeutics. The first is that many A2AR ligands have very
short pharmacokinetic half-lives, which makes delivery of a consistent dose difficult[2, 4]. The
second is that A2AR ligands would be expected to affect physiology in multiple organs
because of the diverse distribution of the A2A receptors[1].
Previously, our group attempted to address both of these challenges by tethering the
A2AR agonist CGS to the Fc domain of an immunoglobulin[5]. The antibody Fc domain is
the key component in many biologics that confers a very long half-life, several days to weeks,
allowing for infrequent dosing[6]. The accepted basis for such a long half-life is the Fc's
interactions with the neonatal receptor (FcRn) which facilitates recycling of the Fc-containing
molecules into the circulation rather than lysosomal removal[7]. Another isoform of the Fc
receptor, FcγRI, which is commonly expressed on antigen presenting cells, can allow for
localization of Fc-containing molecules in sites of high immunoactivation through multiple
receptor engagements on the same cell, such as myeloid cells, or neighboring immune cells
like T cells and antigen presenting cells[8]. Fc fusion has been used in a variety of protein
fusion drugs that are clinically approved including etanercept, which targets the TNFα
receptor and is used for rheumatoid arthritis and other immune disorders[9]. Previously we
adapted a semisynthetic method, expressed protein ligation[10], to attach the Fc domain to the
small molecule CGS. We showed that Fc-CGS indeed had a multi-day half-life in mice and
was much more potent at treating autoimmune pneumonitis than the small molecule CGS[5].
Here we explore the pharmacologic potential of linking an A2AR antagonist ZM to the

ChemBioChem
10.1002/cbic.201600337
IgG3 Fc domain. Originally developed by the Zeneca group for neuroprotection and
treatment of cognitive disorders, ZM selectively and potently inhibits the human A2AR with a
Ki of 0.8 nM[2]. However, ZM was also reported to have a short pharmacokinetic half-life
(less than 20 min in cats), whether administered orally or intravenously[11], with no detectable
levels present in plasma 4 h after oral administration. Fc attachment could enhance its
stability by the FcRn recycling mechanism as well as by preventing hepatic metabolism by
prohibiting uptake in hepatocytes. Below we discuss our synthetic efforts toward Fc-ZM and
its initial pharmacologic characterization.
Results and Discussions
Synthesis of Fc-ZM.
In designing Fc-ZM, we considered the key structural features required for antagonistic
activity and ensured that the derivatization of the ZM small molecule would preserve its
potency for binding and inhibition. We thus pursued the chemical synthesis of A-ZM
(Scheme 1), a form of ZM in which the phenol would be derivatized with a carboxylic acid
that could be convenient for connecting a synthetic linker. Prior X-ray crystallographic
studies of ZM in complex with the A2AR revealed that the phenol was relatively solvent
exposed[12]. We thus predicted that A-ZM would retain the pharmacodynamic properties of
the parent compound ZM. The synthesis of A-ZM began with commercially available
4-hydroxybenzonitrile 1 which was alkylated with t-butyl bromoacetate to generate
intermediate 2. At the same time, the sulfone building block, intermediate 6, was prepared
according to previously published procedures[13]
.
Briefly, the commercially available
2-furonitrile starting material 3 was cyclized with aminoguanidine to generate aminotriazole
4. Condensation of 4 with N-cyanodithioiminocarbonate provided heterocyclic thioether 5
which was subsequently oxidized with mCPBA to yield sulfone intermediate 6. With the two
building blocks 2 and 6 in hand, A-ZM was generated by first hydrogenating intermediate 2
to afford phenethylamine 7. Nucleophilic displacement of the sulfone with the resulting
primary amine and subsequent base-catalyzed removal of the t-butyl group provided the
desired compound in seven overall steps. Separately, we prepared an ethylene-oxy linker
terminating in a Cys residue by solid phase synthesis for use in expressed protein ligation.
The size of this linker was projected to be of sufficient length, about 100 angstroms in
extended conformation, to allow for dual engagement of the ZM component with the A2AR
and the Fc domain with Fc receptors present on different cell types[5]. Consistent with our
previously reported Fc-CGS conjugate, we prepared C-ZM to be used in the subsequent
ligation reaction (Scheme 2).
With C-ZM in hand, it was ligated to the Fc protein via expressed protein ligation. In
order to produce Fc in its glycosylated, disulfide-linked form, we expressed an

ChemBioChem
10.1002/cbic.201600337
Fc-intein-chitin binding domain (CBD) construct containing the mouse IgG3 Fc domain (Fig.
1) in Sf9 insect cells via the baculovirus expression system as previously described[5, 14]. The
secreted Fc-intein-CBD was isolated, purified with chitin resin, and treated with sodium
2-mercaptoethanesulfonate (MESNA) to form the free Fc thioester. The Fc thioester was then
reacted with C-ZM or cysteine to make Fc-ZM and Fc, respectively. We chose expressed
protein ligation to ensure chemoselectivity at the C-terminus of the Fc protein due to its
technical simplicity and high yield in comparison with other methods. SDS-PAGE revealed
Fc and Fc-ZM with more than 90% purity based on Coomassie staining (Fig. 2A). Note that
the broad banding can be understood to be related to heterogenous Fc glycosylation as
analyzed previously[15]
.
High-resolution LC-MS confirmed the expected mass after
deglycosylation by PNGase F (Fig. 2B).
Binding to the A2AR.
To explore the potential of Fc-ZM to bind and antagonize the A2AR, we treated wild-type
mouse splenocytes with Fc-ZM and control compounds for 1 hour and measured the
subsequent production of cyclic AMP (cAMP), a known second messenger generated in
response to A2AR activation[15]. The treatments were either done alone (Fig. 3A) or in
combination with the A2AR agonist CGS (Fig. 3B and 3C), to assess receptor antagonism. We
chose a high dose (1 µM) and a more moderate dose (250 nM) of the ZM compounds to get
an approximate understanding of their potency. Importantly, the mouse splenocytes were
pre-activated with 2 µg/ml of anti-CD3 and anti-CD28 to enhance A2AR expression on the
cell surface. As expected with A2AR activation, CGS treatment led to a sharp increase in
intracellular cAMP production. Treatments with ZM, A-ZM, C-ZM, or Fc demonstrated no
changes in cAMP levels. On the other hand, when treated in combination with CGS, Fc-ZM
and its ZM-containing precursors inhibited the production of cAMP in a potent manner.
Notably, Fc-alone does not affect cAMP production induced by CGS, indicating that the
antagonism by Fc-ZM is due to a functional ZM moiety that is able to bind and inhibit A2AR
activation.
While the antagonist effects of Fc-ZM appear to have been slightly smaller than that of
A-ZM and perhaps some of the other ZM analogs at these concentrations, there could be
several reasons for this. It is possible that Fc-ZM has reduced affinity for A2AR relative to
that of A-ZM and other ZM analogs. However, the exposure to Fc-ZM is relatively short
term (1 hour) and the binding on-rate of the Fc-ZM macromolecule to A2AR may be slower
than that of the smaller ZM analogs. In addition, the concentration of Fc-ZM, which was
deduced by coomassie staining of the Fc component, may overestimate the molar level of
Fc-ZM relative to A-ZM, ZM, and C-ZM, which were determined by more conventional
methods.

ChemBioChem
10.1002/cbic.201600337
Binding to Fc Receptors.
Next we examined the ability of Fc-ZM to bind the Fc receptors. Mouse IgG3 is known
to bind to two Fc receptor isotypes: FcγRI and FcRn. FcγRI is typically found on dendritic
cells where receptor binding leads to cell activation via the immunoreceptor tyrosine-based
activation motif (ITAM)[8]. On the other hand, FcRn expression is found on most myeloid and
lymphoid cell types except T cells, NK cells, and eosinophils[8]. As mentioned, FcRn is
known to be responsible for uptake, transport, and recycling of antibodies. Binding to FcRn
occurs at the CH2-CH3 region, independent of Fc glycosylation[16]. Unlike other Fc receptors,
however, FcRn binds most Fc domains at acidic pH[8], as present in endocytic vesicles. We
therefore investigated the potential of Fc-ZM to bind both the FcγRI and the FcRn at pH 7.4
and pH 6.4 via surface plasmon resonance (SPR). In order to measure the affinities at low pH,
we immobilized FcγRI and FcRn through amine coupling to the chips and injected Fc and
Fc-ZM as analytes. As shown in Figure 4, both Fc and Fc-ZM were able to bind FcγRI and
FcRn at pH 6.4, with the affinity to FcRn being approximately one order of magnitude higher
than FcγRI. When the pH is increased from 6.4 to 7.4, affinity to FcγRI increased by about
2-fold, whereas the affinity to FcRn decreased by 6- to 10-fold (Fig. S3). These data confirm
that Fc-ZM retains the classical properties of antibody Fc domains to interact with these two
key receptor subtypes.
Ex vivo T-cell Response.
Initial characterization of Fc-ZM's immune properties was carried out in a cellular
system. We treated wild-type mouse splenocytes with Fc-ZM and controls, in the presence of
0.5 µg/ml of anti-CD3 with or without CGS inhibition for 48 hours, and measured the level
of interferon γ (IFNγ) accumulated in the supernatant. As expected, A2AR activation by CGS
decreased the amount of IFNγ produced by the splenocytes whereas ZM had no effect on
their IFNγ production (Fig. 5A). Interestingly, treatment with Fc alone resulted in increased
levels of IFNγ in the supernatant, presumably through the activation of immune cells via
FcγR engagement. The separate addition of ZM with Fc did not further stimulate the cells, in
contrast to Fc-ZM which showed enhanced activation. In fact, cells that were treated with
Fc-ZM were robust producers of IFNγ, leading to extracellular levels that were 6-fold higher
than that of the untreated cells. This increase caused by Fc-ZM suggests potential
strengthening of cell-cell interactions between lymphocytes and antigen-presenting cells. In
addition, the increase was observable even in the presence of a high concentration of CGS
(9-fold molar excess, Fig. 5B), further confirming that the ZM moiety retained its ability to
bind and inhibit the A2AR.
Moreover, testing the same treatments in mouse splenocytes lacking the A2AR (Fig. 5C
and 5D) or the FcγRs (Fig. 5E and 5F) diminished the immunostimulatory effect of Fc-ZM in

ChemBioChem
10.1002/cbic.201600337
a fashion consistent with the proposed mechanism, suggesting that the observed phenomenon
involves the A2AR and the FcγRs. Overall, the observation that Fc-ZM can increase IFNγ
production from wild-type splenocytes significantly more than either component alone serves
to illustrate the benefits of using protein-small molecule conjugates as immunomodulators.
In vivo Response to Vaccinia infection.
To examine whether the reversal of CGS effects by Fc-ZM observed ex vivo can be
extended to an in vivo setting, we looked at its effects on the acute response to Vaccinia virus
infection. In this model[17], we infected wild type C57BL/6 mice with Vaccinia virus
expressing ovalbumin and treated them with various drug combinations over the five days
post-infection. Spleens were harvested on the sixth day post-infection and the number of total
splenocytes, as an indication of systemic immune response, was quantified by counting (Fig.
6). In comparison to vehicle treatment, CGS treatment decreased the number of total
splenocytes by approximately 3-fold. This inhibition was reversed by ZM when given twice
daily to match the dosing schedule of CGS. When ZM was given once daily, the reversal was
incomplete, suggesting that ZM is largely metabolized and/or cleared from the system by the
time the second dose of CGS is administered (at least 4 hours after the first dose). By contrast,
Fc-ZM, but not Fc, was able to reverse the inhibitory effects of CGS when given only two
intraperitoneal injections total (day 1 and day 3 post-infection). As predicted, the ability of
Fc-ZM to match ZM in reversing the immuno-inhibitory effects of CGS, even at a
significantly reduced dosing frequency, is consistent with Fc-ZM's prolonged half-life
resulting in extended periods of antagonism at the A2AR. It is also important to note that the
molar dose of Fc-ZM was 50-fold less compared to that of CGS. These results suggest that
Fc-ZM, as a bivalent molecule, may have enhanced apparent potency to the A2AR, potentially
by engaging with multiple A2ARs or with A2ARs and FcRs on neighboring lymphocytes and
antigen-presenting cells simultaneously.
In vivo Cardiotoxicity of Fc-small molecule conjugates.
Notably, in the Vaccinia model described above, mice that received CGS without an
antagonist appeared to be cachectic (reduced activity and impaired breathing) shortly after the
first dose. Since the heart tissue is known to have high levels of A2AR expression, we
wondered whether the improved therapeutic profile of Fc protein-small molecule conjugates
would lead to reduced cardiovascular toxicity when compared to the small molecule drug
alone. To test this, we administered a previously synthesized Fc-small molecule conjugate,
Fc-CGS (50 nmol/kg)[5], along with PBS and CGS (5 μmol/kg) into anesthetized wild-type
mice and monitored their heart rate changes via an electrocardiogram (EKG) for 20 minutes
(Fig. 7A). Approximately 5 minutes following the intraperitoneal injection, CGS induced
tachycardia (Fig. 7B). In contrast, neither PBS nor Fc-CGS at its therapeutically appropriate

ChemBioChem
10.1002/cbic.201600337
dose[5, 15] induced tachycardia in mice. It is important to note that the bolus concentration of
CGS, at its therapeutic dose[15], is predicted to be much higher than Fc-CGS, at least for the
first 30 minutes after administration. This observation is consistent with the hypothesis that
conjugation with Fc can enhance the pharmacologic properties of small-molecule drugs such
as CGS and ZM, resulting in more efficacious and safer dosing.
Conclusion
In this study we have designed, synthesized, and tested a novel Fc protein-small
molecule conjugate, Fc-ZM, for the purpose of immune enhancement. The design strategy
allowed for a controlled, site-specific conjugation between the Fc domain protein and the
modified A2AR antagonist, A-ZM. Binding studies in vitro demonstrated the ability of Fc-ZM
to bind FcRn and FcγRI, and functionally inhibit the activation of A2AR in a manner similar
to its small-molecule precursor, ZM. In both ex vivo and in vivo immune response models,
Fc-ZM exhibited superior pharmacology in antagonizing the A2AR when compared to ZM.
These results complement prior studies that Fc conjugation can enhance the small molecule
A2AR agonist CGS[5] and highlight the versatility of this semisynthetic approach. More
broadly, we propose that Fc-small molecule conjugates may be particularly effective for
targeting class A GPCRs. In class A GPCRs, the ligand binding pocket is in the
intramembrane region of the receptor[18], and conventional antibodies may have difficulty
achieving clean agonist or antagonist activity at this location. The power and precision of
small molecules to access such targets is well-documented and Fc linkage can further
augment the pharmacologic properties of such compounds. In future studies, it will be
interesting to explore the application of Fc-ZM to augment checkpoint inhibitors in immune
surveillance of cancer.
Experimental Section:
Synthesis of A-ZM.
tert-Butyl (E)-3-(4-(cyanomethyl)phenyl)acrylate (2). A mixture of 4-hydroxybenzyl
cyanide (1, 5.00 g, 37.6 mmol) and powdered K2CO3 (12.7 g, 76.6 mmol) were dissolved in
N,N-dimethylformamide (40 mL). Then, tert-butyl bromoacetate (7.4 mL, 50 mmol) was
dissolved in N,N-dimethylformamide (10 mL) and added dropwise to the reaction at room
temperature. The mixture was vigorously stirred at 40 °C for 16 h resulting in the formation
of a white precipitate. The reaction was allowed to cool and then poured into water (150 mL).
The organic products were extracted with dichloromethane (3 x 25 mL) and the combined
organic extracts were washed with water (3 x 100 mL), brine (50 mL), dried with anhydrous
sodium sulfate, filtered, and concentrated in vacuo. The crude yellow oil was purified by
column chromatography (SiO2, 10-25% EtOAc/hexanes) to yield the desired product as a
slightly yellow, viscous oil (7.6 g, 81%). 1H NMR (500 MHz, CDCl3): δ 1.41 (s, 9H), 3.58 (s,

ChemBioChem
10.1002/cbic.201600337
13
2H), 4.44 (s, 2H), 6.80 (d, J = 8.8 Hz, 2H), 7.15 (d, J = 8.8 Hz, 2H). C NMR (125 MHz,
CDCl3): δ 22.22, 27.59, 65.19, 81.94, 114.71, 117.85, 122.53, 128.75 157.19, 167.34.
ESI-LRMS: [M+H]+ = m/z 248.2.
3-(Furan-2-yl)-1H-1,2,4-triazol-5-amine (4): 2-Furonitrile (3, 5.0 g, 0.05 mmol) was first
dissolved in 3 mL of absolute ethanol and cooled to 0 °C in an ice bath. Dry hydrogen
chloride gas was then bubbled into the mixture for 5 min after which diethyl ether was added
to precipitate a solid intermediate. The isolated solid was then dissolved in pyridine (30 mL)
and to it was added aminoguanidine nitrate (7.4 g, 0.05 mmol) at 0 °C. The mixture was
heated to reflux for 4 h after which it was cooled to RT, filtered, and the filtrate concentrated
in vacuo. The resulting crude oil was placed on ice and treated with an aqueous solution of
8M HNO3 (40 mL). The newly formed precipitate was isolated by filtration and washed with
cold H2O (10 mL) and ethanol (5 mL). The nitrate salt was suspended in near boiling H2O
(30 mL) with stirring and to it was added sodium carbonate in small portions (~1.30 g).
Heating was continued until solids were completely dissolved, then the solution was allowed
to cool to RT and subsequently placed on ice. The resulting precipitate was isolated by
filtration, washed with H2O (3 x 5 mL), and dried to provide the title compound as a colorless
prism (4, 2.37 g, 29%). 1H NMR (500 MHz, DMSO-d6): δ 6.06 (s, 2H), 6.54 (s, 1H), 6.68 (s,
13
1H), 7.68 (s, 1H), 12.08 (s, 1H). C NMR (125 MHz, DMSO-d6): δ 107.53, 111.30, 142.68,
147.68, 152.15, 157.05. ESI-LRMS: [M+H]+ = m/z 151.2.
2-(Furan-2-yl)-5-(methylthio)-[1,2,4]triazolo[1,5-a][1,3,5]triazin-7-amine (5): Compound
4 (2.18 g, 14.5 mmol) and dimethyl N-cyanodithioiminocarbonate (2.33 g, 16.0 mmol) were
placed in a round-bottomed flask fitted with a condenser under argon. The neat mixture was
heated to 170 °C for 1 h with stirring being initiated after compounds began to melt. The
reaction was then cooled to RT and the resulting solid was dry loaded and chromatographed
on silica (0-50% EtOAc/DCM) to yield the desired compound as a colorless solid (5, 1.71 g,
48%). 1H NMR (500 MHz, DMSO-d6): δ 2.51 (s, 3H), 6.71 (dd, J = 3.4 Hz, 1.8 Hz, 1H), 7.16
(dd, J = 3.4 Hz, 0.7 Hz, 1H), 7.92 (dd, J = 1.7 Hz, 0.8 Hz, 1H), 8.86 (br, 2H). 13C NMR (125
MHz, DMSO-d6): δ 13.60, 112.12, 112.57, 145.25, 145.52, 149.59, 156.19, 157.20, 173.35.
ESI-LRMS: [M+H]+ = m/z 249.2.
2-(Furan-2-yl)-5-(methylsulfonyl)-[1,2,4]triazolo[1,5-a][1,3,5]triazin-7-amine (6):
A
solution of 3-chloroperoxybenzoic acid (77% by wt., 6.58 g, 39.7 mmol) in dichloromethane
(60 mL) was added to a stirred, ice-cooled suspension of compound 5 (1.5 g, 6.1 mmol) in
dichloromethane (60 mL). The resulting suspension was allowed to warm to ambient
temperature and stirring was continued for 16 h. The solvent was evaporated and the resulting
solid triturated in ethanol (30 mL). The suspension was then cooled on ice and the precipitate

ChemBioChem
10.1002/cbic.201600337
was isolated by filtration to provide the desired material as a white, crystalline solid (1.4 g,
83%). 1H NMR (500 MHz, DMSO-d6): δ 3.37 (s, 3H), 6.76 (dd, J = 3.3 Hz, 1.7 Hz, 1H), 7.27
13
(d, J = 3.5 Hz, 1H), 7.99 (m, 1H), 9.65 (br, 2H). C NMR (125 MHz, DMSO-d6): δ 38.89,
112.34, 113.42, 144.97, 145.79, 152.19, 156.84, 157.32, 165.30. ESI-LRMS: [M+H]+ = m/z
281.0.
2-(4-(2-((7-Amino-2-(furan-2-yl)-[1,2,4]triazolo[1,5-a][1,3,5]triazin-5-yl)amino)ethyl)phe
nyl)acetic acid (A-ZM). Compound 2 (3.80 g, 15.4 mmol) was dissolved in 100 mL of
2-propanol in a Parr shaker together with 25 mL of 1N HCl and 1 g of the 10% Pd/C catalyst.
The reaction mixture was hydrogenated at 50 psi and shaken for 6 h at room temperature. The
reaction was then filtered through a bed of Celite and the filter cake was washed with
isopropanol (100 mL). The combined filtrate and wash were concentrated in vacuo and then
partitioned between saturated aqueous sodium bicarbonate (50 mL) and diethyl ether (50 mL).
The organic layer was isolated and the aqueous layer was further extracted with diethyl ether
(30 mL). The combined organic extracts were dried with anhydrous sodium sulfate, filtered,
and concentrated in vacuo to yield compound 7 as a colorless oil that was used without
further purification. Compound 7 was then diluted in acetonitrile (5 mL) and slowly added to
a solution of 6, also dissolved in acetonitrile (5 mL). The reaction mixture was stirred for 16
h at room temperature after which the solvent was removed in vacuo and the reaction
chromatographed on silica (5% MeOH/DCM) to provide a viscous yellow oil. The oil was
taken up in ethyl acetate and recrystallized to yield the penultimate ester, tert-butyl
2-(4-(2-((7amino-2-(furan-2-yl)-[1,2,4]triazolo[1,5-a][1,3,5]triazin-5-yl)amino)ethyl)phenyl)
acetate, as a white solid (yield: 0.6 g, 76%). 1H NMR (500 MHz, DMSO-d6): δ 1.41 (s, 9H),
2.78 (m, 2H), 3.45 (m, 2H), 4.59 (s, 2H), 6.67 (s, 1H), 6.82 (d, J = 8.3 Hz, 2H), 7.05 (d, J =
13
3.1 Hz, 1H), 7.16 (d, J = 8.2 Hz, 2H), 7.46 (m, 1H), 7.86 (s, 1H), 8.28 (m, 2H). C NMR
(125 MHz, DMSO-d6): δ 27.69, 33.87, 42.36, 65.00, 81.29, 111.61, 111.90, 114.32, 129.58,
132.11, 144.61, 146.19, 150.01, 155.80, 156.06, 159.19, 161.08, 167.96. ESI-LRMS: [M+H]+
= m/z 452.2.
The tert-butyl ester (0.40 g, 0.89 mmol) was dissolved in a 3:1:1 mixture of
tetrahydrofuran/methanol/water (20 mL) and to it was added lithium hydroxide (0.17 g, 7.1
mmol). The reaction was stirred at room temperature for 1 h after which it was complete as
evidenced by TLC. Then, the mixture was acidified to pH 2 with 1N HCl and the resulting
precipitate was isolated by filtration and washed with water to provide the title compound as
a white solid (0.28 g, 79%). 1H NMR (500 MHz, DMSO-d6): δ 2.78 (t, J = 7.5 Hz, 2H), 3.44
(m, 2H), 4.62 (s, 2H), 6.67 (dd, J = 3.3 Hz, 1.7 Hz, 1H), 6.84 (d, J = 8.6, 2H), 7.05 (dd, J =
13
3.5 Hz, 0.8 Hz, 1H), 7.17 (m, 2H), 7.50 (m, 1H), 7.86 (s, 1H), 8.30 (m, 2H) (Fig. S1a). C
NMR (125 MHz, DMSO-d6): δ 33.82, 42.44, 64.48, 112.01, 112.38, 114.30, 129.59, 131.94,

ChemBioChem
10.1002/cbic.201600337
144.91, 145.14, 145.53, 149.94, 156.16, 158.53, 160.99, 170.26 (Fig. S1b). ESI-HRMS:
calcd. for C18H18N7O4: [M+H]+ = m/z 396.1420, found: [M+H]+ = m/z 396.1415 (Fig. S1c).
Synthesis of C-ZM. tert-Butyl (14-amino-5-oxo-1,1,1-triphenyl-9,12-dioxa-2-thia-6-azatetra
decan-4-yl) carbamate (0.20 g, 0.33 mmol), synthesized as previously described[5], was
dissolved in anhydrous tetrahydrofuran (3 mL) and added to PAL resin (MidWest Bio-Tech)
(0.10 g, 0.09 mmol) suspended in anhydrous THF (2 mL). Glacial acetic acid (0.10 mL) was
added to this mixture and the reaction was allowed to stir at room temperature for 1 h. Then,
NaBH(OAc)3 (0.16 g, 0.77 mmol) was added and stirring was continued overnight. The resin
was then washed with methanol (5 x 5 mL), N,N-dimethylformamide (5 x 5 mL) and
dichloromethane (5 x 5 mL) in a Bio-Rad Poly-Prep Chromatography Column. The
subsequent coupling reactions with Fmoc-8-amino-3,6-dioxaoctanoic acid (139 mg, 0.35
mmol each reaction) and A-ZM (17.8 mg, 0.045 mmol prepared as previously described)
were based on standard solid phase peptide synthesis methodology. Coupling was carried out
at room temperature for 3 h with 5% diisopropylethylamine (1.8 mL) and
O-(Benzotriazol-1-yl)-N,N,N’,N’-tetramethyluronium hexafluorophosphate (HBTU, 133 mg,
0.35 mmol). Fluorenylmethoxycarbonyl deprotection was carried out at room temperature for
1 h with 20% piperidine (3 mL). The solvent used for both reactions was
N-methyl-2-pyrrolidone (NMP). The crude product was cleaved from the PAL resin with 4
mL of a 95% TFA (trifluoroacetic acid), 2.5% ddH2O and 2.5% triisopropylsilane solution at
room temperature for 1 hour, dried under vacuum, resuspended in 7 mL of ddH2O containing
0.05% TFA and then filtered through a 0.2 μm filter to remove all insoluble particles. The
filtrate was further purified by reverse-phase high performance liquid chromatography
(HPLC), Varian Dynamax Microsorb 100-5 C18 column (250 x 21.4 mm), gradient: 5% to
60% acetonitrile/ddH2O, 50 min; 60%-100% acetonitrile/H2O, 5 min; 100% acetonitrile, 10
min; 100%-5% acetonitrile/H2O, 10 minutes; the flow rate was 10 mL per minute. The final
purified product, C-ZM, was lyophilized and isolated as a fine brownish powder (16 mg,
13%). Compound characterization was done via analytical HPLC and electrospray ionization
(ESI) mass spectrometry (Figure S2a and S2b).
Synthesis of Fc-ZM. The Fc domain of the mouse IgG3 gene (aa 104-330) was expressed
and isolated as previously described[5]. Briefly, Sf9 insect cells were grown in serum-free
media (Sf-900 III, Invitrogen) in suspension culture at 27 °C. Fc-intein-CBD recombinant
protein was expressed and secreted by Sf9 insect cells to the supernatant, which was collected
by centrifugation at 1,500 rpm for 10 min and filtered through a 0.2 μm filter to remove the
cell pellet and debris. The Fc-intein-CBD fusion protein was then purified by passing the
supernatant over a bed of chitin beads (NEB, 2 mL bed volume per liter culture) by gravity
flow. The beads were washed with 50 mL of phosphate buffered saline (PBS) 5 times before

ChemBioChem
10.1002/cbic.201600337
being subjected to ligation with C-ZM. To initiate the ligation with C-ZM, 1-2 column
volumes of 400 mM sodium 2-mercaptoethanesulfonate in PBS (pH = 7.4) containing more
than 100 molar equivalents of C-ZM was added to the column. The column was purged with
argon and the reaction was carried out at room temperature over 72 h. The ligated product
was eluted from the column with 5 x 1 column volume of PBS, and the combined elutions
were dialyzed against 4 L of PBS using a 10 kDa molecular weight cutoff (MWCO) dialysis
cassette (Slide-A-Lyzer™, Thermo Fisher Scientific). Each buffer exchange lasted at least 6 h,
with a total of six buffer exchanges. The final Fc-ZM was further concentrated to 0.5-1
mg/mL and stored at -80 °C. Purified yields were in the range of 0.5 to 1 mg Fc-ZM per liter
of Sf9 cell culture.
Liquid Chromatography Mass Spectrometry (LC-MS) analysis of Fc-ZM. LC-MS was
performed on a LXQ system (Thermo Scientific) with a Poroshell 300SB-C8 column (5 µm,
75 x 1.0 mm). Fc-ZM was first digested with PNGase F (New England Biolabs) according to
the protocol provided by the manufacturer. The digested sample was then treated with 50 mM
DTT and heated at 55 °C for 20 min before subjecting it to LC-MS analysis. The LC was
performed at 60 °C eluting with a linear gradient of 20-40% acetonitrile/water containing
0.1% formic acid over 10 min at a flow rate of 0.25 mL/min.
Intracellar cAMP (cyclic AMP) measurements via enzyme-linked immunosorbent assay.
Spleens from wild type (C57BL/6) mice were harvested and crushed on a cell strainer (BD
Bioscience). The red blood cells were then removed using ACK lysis buffer. The isolated
splenocytes were seeded in 96-well plates with AE7 medium (45% RPMI 1640, 45% EHAA
Clicks, 10% fetal bovine serum, 1% antibiotics, 1% L-Glutamine, 0.1% Gentamycin, 0.05%
2-mercaptoethanol) and activated by 2 µg/mL of soluble anti-CD3 and anti-CD28 for 48 h,
then rested in fresh medium without anti-CD3 or anti-CD28 for 24 h before drug treatments.
To determine the effects of CGS, ZM, A-ZM, C-ZM, Fc, and Fc-ZM on A2AR function, the
amount of total cAMP produced in wild type (C57BL/6) splenocytes was assayed with the
cAMP Biotrak EIA system (GE Healthcare Life Science) according to the manufacturer’s
instructions. Stock concentrations of A-ZM, C-ZM, and Fc-ZM were determined by dry
weight, UV absorbance (at 260 nm), and Coomassie staining, respectively.
Surface Plasmon Resonance (SPR) Measurements. SPR measurements were performed on
a Biacore T200 instrument (GE Healthcare). FcRn and FcγRI were immobilized in different
channels on a CM5 sensor chip (GE Healthcare) using an amine coupling kit following
methods provided by the manufacturer. The sensor chip surface was activated with NHS and
EDC followed by flowing of receptors until 2000 resonance units (RU) were obtained. A
blank reference cell was used as a reference. Fc and drug conjugated Fc (Fc-ZM) were flown

ChemBioChem
10.1002/cbic.201600337
over the cell as analytes. HBS-P+ buffer (10 mM HEPES, 150 mM NaCl, 0.05% surfactant
P20, pH 7.4 or 6.4) were used as sample buffer and running buffer. Association of analytes Fc
and Fc-ZM were measured for three minutes with a flow rate of 20 L/min and allowed to
dissociate for another three minutes. The surface regeneration was achieved with 3 M MgCl2
at a flow rate of 20 L/min for 60 s. The analytes were injected in a series of two-fold
dilutions from the highest concentration used (2 M, 1 M, 500 nM, 250 nM, 125 nM, 62.5
nM, and 31.25 nM etc and blanks). Multiple individual experiments were carried out and a
typical curve is presented in the figures. All sensorgrams were reference channel and blank
subtracted for analyses. Kinetics data were obtained by global fitting of the binding data to a
1:1 steady state affinity of Langmuir binding model using BIAcore T200 evaluation software
(GE healthcare).
Mice. Mice between six- to twelve-weeks of age, including both males and females, were
used for all the experiments in this study. All mouse procedures were approved by the Johns
Hopkins University Institutional Animal Care and Use Committee. Anesthesia was achieved
using isoflurane in a closed chamber. Mice were euthanized via cervical dislocation.
C57BL/6 mice were obtained from the Jackson Laboratory. A2AR knock-out mice were bred
and housed in pathogen-free conditions on the Johns Hopkins School of Medicine East
Baltimore campus. Mice deficient of the γ-chain subunit of FcγR1, FcγRIII, and FcεRI were
purchased from Taconic Biosciences (Model 583).
Ex vivo T cell response via Interferon γ ELISA. Splenocytes from wild type (C57BL/6)
mice, A2AR knock-out mice, and mice deficient of the γ-chain subunit of FcγR1, FcγRIII, and
FcεRI (Model 583, Taconic) were isolated and cultured in round-bottom 96-well plates with
AE7 medium (45% RPMI 1640, 45% EHAA Clicks, 10% fetal bovine serum, 1% antibiotics,
1% L-glutamine, 0.1% gentamycin, 0.05% 2-mercaptoethanol) over 48 h, in the presence of
0.5 µg/ml of anti-CD3 and various drug treatments. Supernatants were harvested and the
level of IFNγ was determined via enzyme-linked immunosorbent assay (Mouse IFNγ ELISA
Ready-SET-Go!®, Affymetrix eBioscience).
In vivo response to Vaccinia virus infection. Wild type (C57BL/6) mice were infected with
Vaccinia virus expressing ovalbumin on day 1 via retro-orbital injections at 1 million
plaque-forming unit (PFU) per mouse. Throughout days 2 to 6, various treatments were given
via intraperitoneal injections, including vehicle (20% DMSO in PBS), CGS (5.0 µmol/kg,
b.i.d.), CGS (5.0 µmol/kg, b.i.d.) and ZM (8.9 µmol/kg, q.d.), CGS (5.0 µmol/kg, b.i.d.) and
ZM (8.9 µmol/kg, b.i.d.), CGS (5.0 µmol/kg, b.i.d.) and Fc (100 nmol/kg, on days 2 and 4),
CGS (5.0 µmol/kg, b.i.d.) and Fc-ZM (100 nmol/kg, on days 2 and 4). Importantly, Fc and
Fc-ZM was given one hour before CGS injections. On day 7, splenocytes were harvested and

ChemBioChem
10.1002/cbic.201600337
counted using a hemocytometer.
Statistical analysis. All graphs were created using GraphPad Prism software, and statistical
analysis was performed with GraphPad Prism. Comparisons between three or more
independent groups were assessed by one-way ANOVA with a Tukey’s multiple-comparisons
test.
In vivo cardiovascular response to CGS and Fc-CGS. Wild type (C57BL/6) mice were
anesthetized with 3% isoflurane and maintained under 2% isoflurane anesthesia throughout
EKG acquisition. Anesthetized mice were placed in a supine position on a temperature
controlled heating pad. Body temperature was monitored with rectal probe and maintained at
37–38°C. EKG probes were inserted subcutaneously and EKG signal (Standard lead II) was
obtained using a PowerLab data acquisition system (ML866) and Animal Bio Amp (ML136;
AD Instruments, Colorado Springs, CO, USA). Vehicle (PBS), 5 μmol/kg of CGS, or 50
nmol/kg of Fc-CGS was administered via intraperitoneal injections. 5 min of EKG signal
were recorded prior to each injection, followed by 20 min of recording after the injection.
LabChart Pro 7.2 software (AD Instruments, Colorado Springs, CO, USA) was used for
automated EKG tracing analysis.
Acknowledgements:
This work was supported by the National Institutes of Health and the FAMRI Foundation. We
thank Dr. Yana Li (The Eukaryotic Tissue Culture Facility at Johns Hopkins) for her help and
support with insect cell culture. We thank Safiat Ayinde (Department of Pharmacology and
Molecular Sciences, Johns Hopkins School of Medicine) for her help with peptide synthesis
and protein expression. We also thank Dr. Robert Leone, Chirag Patel, Min-Hee Oh and Jiayu
Wen (Department of Oncology, Johns Hopkins School of Medicine) for their guidance and
training in immunology research and animal care.
References:
[1]
[2]
[3]
B. B. Fredholm, I. J. AP, K. A. Jacobson, K. N. Klotz, J. Linden, Pharmacol Rev 2001, 53,
527-552.
M. de Lera Ruiz, Y. H. Lim, J. Zheng, Journal of medicinal chemistry 2014, 57,
3623-3650.
aA. Ohta, E. Gorelik, S. J. Prasad, F. Ronchese, D. Lukashev, M. K. Wong, X. Huang, S.
Caldwell, K. Liu, P. Smith, J. F. Chen, E. K. Jackson, S. Apasov, S. Abrams, M. Sitkovsky,
Proc Natl Acad Sci U S A 2006, 103, 13132-13137; bO. P. Dall'Igna, L. O. Porciuncula, D.
O. Souza, R. A. Cunha, D. R. Lara, British journal of pharmacology 2003, 138,
1207-1209; cK. Golembiowska, A. Dziubina, Neuropharmacology 2004, 47, 414-426.

ChemBioChem
10.1002/cbic.201600337
[4]
[5]
J. P. Chovan, P. A. Zane, G. E. Greenberg, Journal of chromatography 1992, 578,
77-83.
M. J. Chiang, M. A. Holbert, J. H. Kalin, Y. H. Ahn, J. Giddens, M. N. Amin, M. S. Taylor,
S. L. Collins, Y. Chan-Li, A. Waickman, P. Y. Hsiao, D. Bolduc, D. J. Leahy, M. R. Horton, L.
X. Wang, J. D. Powell, P. A. Cole, Journal of the American Chemical Society 2014, 136,
3370-3373.
[6]
[7]
D. Levin, B. Golding, S. E. Strome, Z. E. Sauna, Trends in biotechnology 2015, 33,
27-34.
aD. M. Czajkowsky, J. Hu, Z. Shao, R. J. Pleass, EMBO molecular medicine 2012, 4,
1015-1028; bS. Mitragotri, P. A. Burke, R. Langer, Nature reviews. Drug discovery
2014, 13, 655-672.
[8]
[9]
P. Bruhns, Blood 2012, 119, 5640-5649.
K. Peppel, D. Crawford, B. Beutler, The Journal of experimental medicine 1991, 174,
1483-1489.
[10] T. W. Muir, D. Sondhi, P. A. Cole, Proc Natl Acad Sci U S A 1998, 95, 6705-6710.
[11] aS. M. Poucher, J. R. Keddie, P. Singh, S. M. Stoggall, P. W. Caulkett, G. Jones, M. G.
Coll, British journal of pharmacology 1995, 115, 1096-1102; bS. M. Poucher, J. R.
Keddie, R. Brooks, G. R. Shaw, D. McKillop, The Journal of pharmacy and
pharmacology 1996, 48, 601-606.
[12] V. P. Jaakola, M. T. Griffith, M. A. Hanson, V. Cherezov, E. Y. Chien, J. R. Lane, A. P.
Ijzerman, R. C. Stevens, Science 2008, 322, 1211-1217.
[13] aP. W. R. Caulkett, G. Jones, M. McPartlin, N. D. Renshaw, S. K. Stewart, B. Wright,
Journal of the Chemical Society, Perkin Transactions 1 1995, 801-808; bA. J. Hutchison,
M. Williams, R. de Jesus, R. Yokoyama, H. H. Oei, G. R. Ghai, R. L. Webb, H. C. Zoganas,
G. A. Stone, M. F. Jarvis, Journal of medicinal chemistry 1990, 33, 1919-1924; cP. W. R.
Caulkett, G. Jones, M. G. Collis, S. M. Poucher, Google Patents, 1991.
[14] D. Bolduc, M. Rahdar, B. Tu-Sekine, S. C. Sivakumaren, D. Raben, L. M. Amzel, P.
Devreotes, S. B. Gabelli, P. Cole, eLife 2013, 2, e00691.
[15] P. E. Zarek, C. T. Huang, E. R. Lutz, J. Kowalski, M. R. Horton, J. Linden, C. G. Drake, J. D.
Powell, Blood 2008, 111, 251-259.
[16] aW. L. Martin, P. J. Bjorkman, Biochemistry 1999, 38, 12639-12647; bV. Irani, A. J. Guy,
D. Andrew, J. G. Beeson, P. A. Ramsland, J. S. Richards, Molecular immunology 2015,
67, 171-182.
[17] K. N. Pollizzi, C. H. Patel, I. H. Sun, M. H. Oh, A. T. Waickman, J. Wen, G. M. Delgoffe, J.
D. Powell, J Clin Invest 2015, 125, 2090-2108.
[18] D. Zhang, Q. Zhao, B. Wu, Molecules and cells 2015, 38, 836-842.
Figure Legends

ChemBioChem
10.1002/cbic.201600337
Scheme 1. Synthesis of A-ZM. A-ZM was made from two separate building blocks,
intermediates 2 and 6. (A) The synthetic approach to intermediate 2. (B) The synthetic route
for making intermediate 6. (C) Converting 2 and 6 into A-ZM.
Scheme 2. Synthesis of C-ZM. A-ZM was attached to an ethylene-oxy linker terminating in a
Cys residue, which was generated via solid-phase peptide synthesis (SPPS) techniques.
SPPS conditions include: (a) coupling: Fmoc-NH-(PEG)-COOH, HBTU, 5% DIPEA, NMP,
RT, 3 h, and (b) deprotection: 20% piperidine, NMP, RT, 1 h.
Figure 1. Semisynthesis of Fc-ZM by expressed protein ligation. HBM: honeybee melittin
secretion signal. CBD: chitin binding domain.
Figure 2. Characterization of purified Fc and Fc-ZM. (A) Fc and Fc-ZM stained with
Coomassie blue after being resolved by 10% SDS-PAGE. (B) Liquid chromatography mass
spectrometry (LC-MS) spectrum confirming the expected mass of Fc-ZM deglycosylated
using PNGase F. Fc-ZM calculated m/z 31163. The experimental error on this
measurement is estimated to be ±20 mass units.
Figure 3. Assessment of A2AR binding via intracellular cAMP assay. Wild-type mouse
splenocytes were pre-activated with 2 µg/mL of soluble anti-CD3 and anti-CD28 for 48 h,
followed by a 24 h resting period in fresh media without anti-CD3 and anti-CD28 before
subjecting to drug treatments. (A) Intracellular cAMP produced after 1 h incubation in media
containing various drugs at 1 μM. (B) Intracellular cAMP produced by splenocytes treated
with various drugs at 1 μM in the presence of 1 μM CGS. (C) Intracellular cAMP produced
by splenocytes treated with various drugs at 250 nM in the presence of 1 μM CGS. Results
were analyzed using one-way ANOVA test (****, P < 0.0001) followed by a Tukey's multiple
comparisons test (****, P < 0.0001).
Figure 4. Assessment of FcRn and FcγRI Binding via surface plasmon resonance at pH 6.4.
(A) and (B) shows the binding of Fc and Fc-ZM, respectively, to the immobilized mouse
neonatal Fc receptor (FcRn). (C) and (D) shows the binding of Fc and Fc-ZM, respectively, to
the immobilized mouse FcγR1.
Figure 5. Ex vivo T-cell response via interferon γ ELISA. (A), (C), and (E) shows the level of
IFNγ secreted by splenocytes harvested from wild-type mice, A2AR knock-out mice, and
FcγR1 knock-out mice (mice lacking the γ-chain subunit of FcγR1, FcγRIII, and FcεRI),
respectively, following a 48 h incubation with various drug treatments at 111 nM, in the
presence of 0.5 µg/ml of anti-CD3. (B), (D), and (F) shows the level of IFNγ secreted by

ChemBioChem
10.1002/cbic.201600337
splenocytes harvested from wild-type mice, A2AR knock-out mice, and FcγR1 knock-out
mice (mice lacking the γ-chain subunit of FcγR1, FcγRIII, and FcεRI), respectively,
following a 48 h incubation with various drug treatments at 111 nM, in the presence of 0.5
µg/ml of anti-CD3 and 1 μM CGS. Results were analyzed by one-way ANOVA followed by a
Tukey's multiple comparisons test (*, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P <
0.0001).
Figure 6. In vivo response to Vaccinia virus infection. Number of total splenocytes was
quantified in spleens collected from mice treated with vehicle (n = 19), CGS (n = 14), CGS +
ZM q.d. (n = 8), CGS + ZM b.i.d. (n = 12), CGS + Fc (n = 10), or CGS + Fc-ZM (n = 8). The
results were normalized to the vehicle treatment and statistical significance determined by
one-way ANOVA (P < 0.0001) followed by a Tukey's multiple comparisons test (*, P < 0.05;
**, P < 0.01; ***, P < 0.001; ****, P < 0.0001).
Figure 7. In vivo cardiovascular response to CGS and Fc-CGS. (A) Average heart rate of
mice treated with PBS (n = 3), CGS (n = 4), or Fc-CGS (n = 3), monitored over 20 minutes
post-administration of the treatments. Dashed line indicates the timing of the drug injections.
(B) Average heart rate at 5.5 minutes following the injection. Results were analyzed using
one-way ANOVA (***, P = 0.0002) followed by a Tukey's multiple comparisons test (***, P
< 0.001). BPM, beats per minute.
Text Suggestion for the Table of Contents
Using a semisynthetic approach, an Fc-small molecule conjugate, Fc-ZM, was made for the
targeted inhibition of the adenosine 2A receptor. Fc-ZM displayed superior pharmacologic
properties compared to the small-molecule precursor ZM and was able to functionally
interact with both A2AR and FcRs, present on lymphocytes and antigen-presenting cells,
respectively.

ChemBioChem
10.1002/cbic.201600337
Figure 1

ChemBioChem
10.1002/cbic.201600337
Figure 2

ChemBioChem
10.1002/cbic.201600337
Figure 3

ChemBioChem
10.1002/cbic.201600337
Figure 4

ChemBioChem
10.1002/cbic.201600337
Figure 5

ChemBioChem
10.1002/cbic.201600337
Figure 6

ChemBioChem
10.1002/cbic.201600337
Figure 7

ChemBioChem
10.1002/cbic.201600337

ChemBioChem
10.1002/cbic.201600337

ChemBioChem
10.1002/cbic.201600337
Scheme 1

ChemBioChem
10.1002/cbic.201600337

ChemBioChem
10.1002/cbic.201600337
Scheme2