Design, synthesis, and biological activity of TLR7-based compounds for chemotherapy-induced alopecia

Article abstract of DOI:10.1007/s10637-019-00793-5

Hair loss is a common dermatosis symptom and side-effect in cancer chemotherapeutics. Imiquimod application at mid and late telogen activated the hair follicle stem cells leading to premature hair cycle entry. Based on quinoline structure, a newly synthesized compound 6b displayed proliferation activity in vitro and in vivo through branch chain replacement and triazole ring cyclization. Toll-like receptors (TLRs) are also critical mediators of the immune system, and their activation is linked to various diseases. The present study aimed to expand new agonists within co-crystallization of TLR7 (PDB code: 5GMH); however, biological assays of NF-κB activity and NO-inhibition indicated that five selected compounds were TLR7 antagonists. Molecular docking indicated the binding mode differences: antagonists binding TLR7 in a different direction and interacting with adjacent TLR7 with difficulty in forming dimers.

Full text of DOI:10.1007/s10637-019-00793-5

Investigational New Drugs
https://doi.org/10.1007/s10637-019-00793-5
PRECLINICAL STUDIES
Design, synthesis, and biological activity of TLR7-based compounds
for chemotherapy-induced alopecia
Jincheng Yang1 & Kun Chen2 & Bin Wang3 & Liudi Wang3 & Shuya Qi1 & Weihua Wang1
Received: 29 March 2019 /Accepted: 13 May 2019
#
Springer Science+Business Media, LLC, part of Springer Nature 2019
Summary
Hair loss is a common dermatosis symptom and side-effect in cancer chemotherapeutics. Imiquimod application at mid and late
telogen activated the hair follicle stem cells leading to premature hair cycle entry. Based on quinoline structure, a newly
synthesized compound 6b displayed proliferation activity in vitro and in vivo through branch chain replacement and triazole
ring cyclization. Toll-like receptors (TLRs) are also critical mediators of the immune system, and their activation is linked to
various diseases. The present study aimed to expand new agonists within co-crystallization of TLR7 (PDB code: 5GMH);
however, biological assays of NF-κB activity and NO-inhibition indicated that five selected compounds were TLR7 antagonists.
Molecular docking indicated the binding mode differences: antagonists binding TLR7 in a different direction and interacting with
adjacent TLR7 with difficulty in forming dimers.
.
.
.
.
Keywords TLR7 Hair follicle stem cell Chemotherapy-induced alopecia Antagonist Synthesis
Introduction
Although CIA is usually reversible, new hair shows
graying and/or changes in the structure and texture, and the
hair growth rate might also be significantly reduced [6]. In
several patients, the regrown hair shows altered color and/or
structure and texture. In some cases, the hair density may be
reduced [7]. Permanent CIA could result from irreversible hair
follicle stem cell destruction in the hair bulge [1] or disruption
of signaling to the secondary hair germination [8].
Hair follicles undergo repeated cycles of anagen (growth),
catagen (regression), and telogen (quiescence) throughout the
life in mammals following morphogenesis [9]. Additionally,
the degree of damage to the hair follicle stem cell (HFSC)
determines whether the CIA is reversible; if chemotherapy
only has effects on the bulge of the stem cells (at the middle
of hair shaft), alopecia is reversible, while if the bulb is affect-
ed (at the bottom of hair shaft), alopecia is irreversible.
Currently, HFSC proliferation is used for the prevention or
treatment of CIA. Predominantly, CIA affects the highly pro-
liferative matrix keratinocytes of anagen hair follicles that are
located in the hair bulb. However, damaged HFSCs could also
lead to permanent hair loss [10].
Hair loss is both a common dermatosis condition and a
side-effect in patients undergoing chemotherapeutic treat-
ment for cancer. Hair loss experience causes feelings of
distorted self-image and low self-esteem, and may display
symptoms of anxiety and depression in sufferers [1].
Alopecia is a common skin disease in patients with neo-
plastic medical behaviors [2, 3] affecting all genders and
age groups [4]. Chemotherapy-induced alopecia (CIA) is
frequently described as one of the most distressing aspects
of cancer treatments [5].
* Jincheng Yang
Yangjc@cicams.ac.cn
1
Department of Pharmacy, National Cancer Center/National Clinical
Research Center for Cancer/Cancer Hospital, Chinese Academy of
Medical Sciences and Peking Union Medical College, Panjiayuan
Nanli 17, Chaoyang Qu, Beijing 100021, China
2
State Key Laboratory of Molecular Oncology/Department of
Imiquimod (Toll-like receptor-7 (TLR7) agonist) is a quin-
oline derivative and applied to mid and late telogen-activated
HFSCs, which leads to premature hair cycle entry (anagen)
[11]. In 2007, Letada et al. presented a case of a 15-year-old
girl with alopecia universalis since the age of 8 years, who
Immunology, National Cancer Center/National Clinical Research
Center for Cancer/Cancer Hospital, Chinese Academy of Medical
Sciences and Peking Union Medical College, Beijing 100021, China
3
Clinical Stem Cell Center, The Affiliated Drum Tower Hospital of
Nanjing University Medical School, Nanjing 210008, China

Invest New Drugs
experienced transient hair growth after topical application of
imiquimod [12].
various amine groups onto the main quinoline structure.
Then, the nitro-group was reduced into the amino-group,
and the final cyclization was complicated with NaNO2 in an
ice-water bath (Scheme 1).
Paus and Cotsarelis reported the development of drugs that
shield the rapidly proliferating bulb cells from chemotherapeu-
tic damage and protect the stem cells from irreversible damage
by the infiltration of autoimmune inflammatory cells or ioniz-
ing radiation [13].ThesestudiesonCIA-treatingdrugsreduced
the social and psychological effects accompanying tumor ther-
apy, andpreventionofhairlossduetochemotherapybenefitted
a large number of patients undergoing cancer treatment.
Herein, we reported a new series of synthesized quinoline de-
rivativesandpreparedanewcompoundforfurnishingvaluable
information on HFSC-proliferation drug design models.
Lipophilicity values
Lipophilicity is a significant physicochemical property determining
the distribution, bioavailability, metabolic activity, and elimination,
and hence, essential for comparison among the synthetic compounds.
The lipophilicity of the synthesized derivatives was determined ex-
perimentally from RP-TLC (Reversed Phase Thin Layer
Chromatography) as RM values and compared with the correspond-
ing theoretically calculated Clog P-values in n-octanol-buffer [14].
Compound 6b presented minimal values (Table 1), indicating that
lower the lipophilicity value, higher the biological activity.
Results
Compound synthesis
HFSC proliferation in vitro
In this study, quinoline derivatives were synthesized from 4-
hydroxyquinoline. The key step was the introduction of
The in vivo pilot test demonstrated that only compound 6b
among the four target compounds (6a–6d) promoted hair
Scheme 1 Synthesis of quinoline derivatives and structure of imiquimod.
Reagents and conditions: (a) conc HNO3, propanoic acid, ref. 1.5 h; (b)
Oxalyl chloride, DCM, 40 °C. 0.5 h; (c) Amines, TEA, CH3OH, rt. 0.5–
2 h; (d) Na2S2O4, isopropanol, H2O, ref. 2 h; (e) NaNO2, HCl, H2O, EtOH,
ice-water bath, overnight. ① Compound 6a: mp:114–116°C. 1H-
NMR(CDCl3, 400MHz): σ(ppm) 9.57(s, 1H), 8.35(t, 2H, J=7Hz), 7.85(td,
1H, J1=7.8Hz, J2=1.2Hz), 7.76(td, 1H, J1=7.8Hz, J2=1.2Hz), 5.50(quint, 1H,
J=6.8Hz), 1.92(d, 6H, J=6.8Hz). IR (KBr): 3065, 2993, 2942, 1620, 1578,
1521, 1451, 1383, 1260, 1091, 768 cm−1. MS (EI+) m/z: 235 (M+Na+).
HRMS m/s calculated for C12H12N4+H+ 235.0954, found 235.0964. ②
Compound 6b: mp:170–172°C. 1H-NMR(CDCl3, 400MHz): σ(ppm)
9.44(s, 1H), 8.42(dd, 1H, J1=8.4Hz, J2=0.8Hz), 8.27(d, 1H, J=8.4Hz),
7.83(td, 1H, J1=7.6Hz, J2=1.2Hz), 7.50(td, 1H, J1=7.8Hz, J2=1.2Hz),
5.17(t, 2H, J=5.2Hz), 4.45(t, 2H, J=5Hz). IR (KBr): 3292, 3070, 3007,
2941, 2851, 1621, 1586, 1524, 1454, 1388, 1323, 1166, 1069, 761 cm−1.
MS (EI+) m/z: 237 (M+Na+). HRMS m/s calculated for C11H10N4O+Na+
237.0747, found 237.0745. ③ Compound 6c: mp:66–68°C. 1H-
NMR(CDCl3, 400MHz): σ(ppm) 9.55(s, 1H), 8.35(d, 1H, J=8Hz), 8.23(d,
1H, J=8Hz), 7.84(t, 1H, J=7.4Hz), 7.76(t, 1H, J=7.2Hz), 4.85(d, 2H,
J=7.2Hz), 2.48(quint, 1H, J=6.4Hz), 1.08(d, 6H, J=6.4Hz). IR (KBr):
3064, 3016, 2964, 2935, 2906, 2875, 1624, 1523, 1459, 1389, 1370, 1322,
1178, 761cm−1. MS (ES+) m/z: 227 (M+H+). HRMS m/s calculated for
C13H14N4+Na+ 249.1111, found 249.1115. ④ Compound 6d: mp:168–
170°C .1H-NMR(CDCl3, 400MHz): σ(ppm) 9.68(s, 1H), 8.95(s, 1H),
8.42(d, 1H, J=7.2Hz), 8.09–8.03(m, 2H), 5.15(t, 2H, J=7.2Hz), 2.17(t, 2H,
J=7.2Hz), 1.27(s, 10H), 0.87(t, 3H, J=6.6Hz). IR (KBr): 3058, 2925, 1646,
1611, 1590, 1511, 1462, 1442, 1317, 776, 724 cm−1. MS (ES+) m/z: 283
(M+H+). HRMS m/s calculated for C17H22N4+H+ 283.1917, found 283.1924

Invest New Drugs
Table 1 Lipophilicity values: RM, Clog P, and log P
treatment. C57BL/6 and BALB/c mice were selected as two
experimental models. The first stage of chemotherapy assays
was injecting the anti-neoplastic drug into normal mice simul-
taneously when both sides of the hair were depilated, spread-
ing compound cream on the right back side and control-cream
on the left back side. The compound-treated groups showed
higher hair follicle density than the control groups (Figs. 5, 6,
7, 8). The second stage was injecting the agents into female
mice established as breast cancer model with 4 T1 cells.
Similar to the first stage, the two sides of the skin were col-
lected, and the hair follicle examined. In addition to the dif-
ference in the hair follicle density, the data also indicated weak
hair of mice in the CYP + DOX group (Fig. 9).
The breast cancer patients underwent various chemothera-
peutic regimens consisting mainly of CTX (cyclophospha-
mide), DOX (doxorubicin), or 5-FU (5-fluorouracil), as adju-
vant therapy after the surgery. 4 T1 is a commonly used mouse
triple-negative breast cancer (TNBC) cell line [18]. When
injected into BALB/c mice, 4 T1 (ATCC; CRL-2539™) spon-
taneously produces highly metastatic tumors that can metas-
tasize to the lung, liver, lymph nodes, and brain while the
primary tumor is growing in situ. Thus, we selected 4 T1 to
induce breast cancer model in mice. Subsequently, the mice
were divided into CYP-only and CYP + DOX groups. DOX
was injected after CYP injection, following which tumor gran-
ules emerged. The tumor vanished in the CYP + DOX group,
but was harvested from mice in the CYP-only group (Fig. 10).
The right back side hair were shaved on 6 female BALB/c
mice and treated with compound cream 40 μL (2 mg/mL)
once every two days. Two mice were sacrificed every week,
and two sides of the back skin were collected for H-E staining.
The right back side hair on 2 female BALB/c mice was
shaved and treated with compound cream 2000 mg/kg (cream
concentration 40%) one time. The mice were observed daily
and sacrificed on day 21. The right-side skin was collected for
H-E staining.
Imiquimod
6a
6b
6c
6d
aRM=[(1/Rf)-1]
bClog P
clog P=Co/Cw
0.879
3.236
2.077
0.176
2.435
1.647
−0.368
1.004
0.534
0.301
3.054
2.052
0.954
5.300
2.714
a RM values are the average of 10 measurements. b Clog P: Theoretical
values. c log P: experimental values
follicle proliferation in normal mouse skin. Thus, the follow-
ing assays were designed on compound 6b.
Mouse hair follicle stem cells were prepared according to
the authoritative isolation method [15] (Fig. 1). The potential
markers for hair follicle stem cells include β1-integrin, keratin
19, α6-integrin, CD71, p63, and CD34 [16, 17]. In the present
study, PE-anti-mouse CD34 antibody and FITC-anti-mouse
β1 integrin were selected to identify HFSCs. The results of
testing mixed antibodies or single antibody of β1 and CD34
did not reveal any overlap in flow cytometry, indicating the
presence of stem cells (Fig. 2).
Hair follicle proliferation in vivo
The effects of compound cream treatment on the skin
consisted of three components: normal mice, acute toxicity,
and chemotherapy treatment. The results of treated and un-
treated sides of the skins of one mouse were compared.
Normal mice group: H-E (hematoxylin-eosin) staining
showed that the number of hair follicles in the compound-
treated group was higher than that in the control group
(Fig. 3). Acute toxicity is a crucial assay for studying the drug
effect with respect to high drug density and dosage on an
organism. The current results did not show any abnormality
on the skin, mouse body, as well as the daily life of the animal,
while the number of hair follicles increased markedly (Fig. 4).
Next, the chemotherapy treatment adopted CYP
(cyclophosphamide) and DOX (Doxorubicin) as both agents
exerted common anti-neoplastic effects in breast cancer
Two back sides of 5 female BALB/c and C57 mice were
depilated with paraffin wax. The right side was treated with
compound cream (40 μL, 2 mg/mL) and the left side with
Fig. 1 Hair follicle stem cells and
proliferation from C57BL/6
Mice. Cell proliferation assay: An
equivalent of 2 × 103 cells were
seeded in 100 μL volume in 96-
well plate. Compound 6b (1, 5,
10, 20, 40 μM) was diluted with
DMEM and assessed by Cell
Counting Kit-8 (CCK-8) assay on
days 1, 3, 5, and 7 at OD 450 nm

Invest New Drugs
Fig. 2 Identification of hair follicle stem cell. Setting001: Blank;
Sample001: FITC/PE [(FITC-anti-mouse β1 integrin)/(PE-anti-mouse
CD34)]; Setting002: FITC(FITC-anti-mouse β1 integrin); Setting003:
PE(PE-anti-mouse CD34). Method: An equivalent of 1 × 105 cells were
seeded in 100 μL staining buffer. PE-anti-mouse CD34 antibody and
control cream (40 μL) while the mice were injected with CYP
(20 μg/g) one time. The cream treatment was given once per
two days. The animals were sacrificed on day 21, and two
sides of the back skin were collected for H-E staining. M1 =
Mouse 1, M2 = Mouse 2.
and still no abnormality occurred in 5 days. The left side hair
was shaved, and the compound cream was applied. On day 7
after 4 T1 injection, granules emerged on the right skin. On
day 11, one group (5 mice) was injected doxorubicin (15 μg/
g). All the mice were sacrificed on day 21; the CYP group
mice normally weighed with obvious tumor while the CYP +
DOX group mice emaciated in dry hair without tumor on the
right side. The two sides of the back skin and tumor pieces
were collected for H-E staining.
Two back sides of 5 female BALB/c and C57 mice were
depilated with paraffin wax. The right side was treated with
compound cream (40 μL, 2 mg/mL) and the left side with
control cream (40 μL) while the mice were injected with
DOX (15 μg/g). The cream was treated once per two days;
Doxorubicin (5 μg/g) was injected s.i.d three days after dep-
ilation. Mice were sacrificed on day 21, and two sides of the
back skin were collected for H-E staining.
TLR7-targeted compound designation and selection
4 T1 cells (1 × 106/100 μL) were injected into the right side
of 10 female BALB/c mice subcutaneously, but no abnormal-
ity occurred after 2 days. CYP (150 mg/kg) was injected i.p.
To the best of our knowledge, the various effects of TLR7 in
infection, inflammation, immune diseases, and tumor were
assessed to identify and enrich new TLR7-agonistic struc-
tures. The co-crystallization of TLR7 and compound R848
was used as a model to design the targeted compounds ac-
cording to the protein data bank (PDB code: 5GMH). While
virtually screening 432 structures, those containing glycosyl
were deleted, and abnormal structures (from SciFinder ) with
high scores were selected for the subsequent assays. Finally,
27 compounds with high scores were collected for the follow-
ing in vitro activity screening.
Control (day 7)
Control (day 14)
Control (day 21)
Compd(day 7)
Compd(day 14)
Compd(day 21)
Fig. 3 Compound cream treatment on normal mice
Fig. 4 Acute toxicity

Invest New Drugs
Fig. 5 Compound cream
treatment on CYP-injected C57
mice
Control (M1)
Control (M 2)
Control (M3)
Control (M4)
Control (M5)
Compd(M1)
Compd(M2)
Compd(M3)
Compd(M4)
Compd(M5)
In vitro assays of virtual screening compounds
these compounds was tested according to the protocol for TLR9
antagonists [19]. Five compounds (C2, C3, C4, C6, and C7)
(Fig. 11) showed steady TLR7 inhibition (Fig. 12).
TLR7 inhibition activity in HEK 293 cells
HEK-Blue™-hTLR7 cells plated in 96-well plate at a den-
sity of 4 × 104 cells/well were stimulated with agonist CL097
(5 μM) in the presence or absence of these compounds
(40 μM) overnight. The supernatant (20 μL) was tested with
Quanti-blue (OD 630 nm), and the residue was assessed for
cell proliferation by CCK-8 method.
The TLR7 agonist activity of all the 27 compounds was evalu-
ated by detecting the HEK-Blue activity in HEK 293 cells ex-
pressing TLR7 based on the Quanti-Blue substrate and expressed
as a change in the NF-κB activity. However, no agonists were
found in these compounds. Based on the results of inflammation-
nitric oxide (NO) inhibition assay, the TLR7 inhibitory activity of
Index of Quanti−Blue value ¼ ðA−BÞ=ðC−BÞ
where A is the average optical density of the compound-
treated cells; B is the average optical density of the control
wells (culture medium with only cells); C is the average opti-
cal density of the positive control (CL097-treated cells).
Anti-inflammatory activity in RAW264.7 cells
TLR7 is expressed on airway nerves and mediates the relaxation
of human and animal airways through NO production [20]. Also,
we evaluated the effects of these 27 compounds on the inhibitory
effect of NO on RAW264.7 cells, and the cell viability was
determined by MTT assay. Five compounds (C2, C3, C4, C6,
and C7) showed anti-inflammatory activity (Fig. 13).
Control (M1)
Control (M2)
Control (M3)
Control (M4)
RAW264.7 cells plated in 96-well plate at a density
of 4 × 104 cells/well were treated with compounds
(40 μM) and lipopolysaccharide (LPS) (1 μg/mL,
10 μL/well) within 1.5 h. The NO concentration was
tested using Griess reagent (540 nm) after 24 h in an
Compd(M1)
Compd(M2)
Compd(M3)
Compd(M4)
Fig. 6 Compound cream treatment on CYP-injected BALB/c mice

Invest New Drugs
Fig. 7 Compound cream
treatment on DOX-injected C57
mice
Control (M1)
Control (M2)
Control (M3)
Control (M4)
Control (M5)
Compd(M1)
Compd(M2)
Compd(M3)
Compd(M4)
Compd(M5)
incubator. The cell proliferation was assessed using
Ligand-receptor interaction by molecular docking
MTT (200 μL, 0.5 mg/mL, 450 nm).
Herein, we aimed to design the agonist; however, the results
indicated that neither did the compounds (C2, C3, C4, C6, and
C7) act on TLR7 agonist signaling (NF-κB pathway) nor
present the inflammatory activity like other TLR7 agonists.
The antagonistic action of TLR7 was found in the agonist
design. To gain a better understanding of the efficiency of
the ligand analog molecules, we studied the interaction of
the 5 compounds docked into the site of TLR7 receptors.
LPS−Compound
LPS−Control
NO inhibition rateð%Þ ¼
 100%
Cell ProliferationðMTTÞð%Þ
Coumpound − Negative ControlðnocellÞ
¼
 100%
Positive ControlðCellÞ − Negative ControlðnocellÞ
Fig. 8 Compound cream
treatment on DOX-injected
BALB/c mice
Control (M1)
Control (M2)
Control (M3)
Control (M4)
Control (M5)
Compd(M1)
Compd(M2)
Compd(M3)
Compd(M4)
Compd(M5)

Invest New Drugs
CYP-only group
Fig. 9 Compound cream
treatment on breast cancer
BALB/c mice
Control (M1)
Control (M2)
Control (M3)
Control (M4)
Control (M5)
Compd(M1)
Compd(M2)
Compd(M3)
Compd(M4)
Compd(M5)
CYP+DOX group
Control (M1)
Control (M2)
Control (M3)
Control (M4)
Control (M5)
Compd(M1)
Compd(M2)
Compd(M3)
Compd(M4)
Compd(M5)
Before molecular dockings, the analysis of the first site of
TLR7 indicated that the residues Phe408, Asp555, Leu557,
Ile585, and Thr586 were directly recognized as first site ligands.
The agonistic chemical ligand R848 can form hydrogen bonds
with Asp555 and Thr586 and a water bridge with Gly584. The
analysis of the binding modes of the five antagonists with TLR7
revealed the first identified site. The first antagonist can form a
strong hydrogen bond with Asp555. In addition, it also forms
three arene-H interactions with Ile585 and Thr586. The second
antagonist can form two strong hydrogen bonds with Asp555
and Gly584 and two arene-H interactions with Ile585 and
Thr586. The third antagonist can form a hydrogen bond with
Asp555 and an arene-H interaction with Tyr579. The fourth an-
tagonist can form three hydrogen bonds with the side chains of
Fig. 10 Breast tumor of BALB/c
mice from 4 T1 cells
2×
5×
10×
20×
40×

Invest New Drugs
Fig. 11 Antagonistic and agonistic compounds
Asp555 and Thr586. The fifth antagonist can form three strong
hydrogen bonds with Asp555, Gln531, and Thr532. In addition,
the antagonist can also form an arene-H interaction with the side
chain of Ile585. All the five antagonists bound to the TLR7
protein in the same site and formed hydrogen bonds with
Asp555, which was critical for the binding of compounds
(Fig. 14).
erythema, scaling, keratinocyte proliferation with acanthosis,
altered keratinocyte differentiation (parakeratosis), and a der-
mal infiltrate [25].
The patents of imiquimod are on the rings A, B, C,
and the possible branches. The designed compound 6b
replaced the atom C with N on ring A and cut the
amino-group on ring B. The key process of synthesis
was adding different aliphatic amines onto the quino-
lone ring (step c, Scheme 1). This new structure (6b)
avoided all the imiquimod patents successfully and
displayed proliferation activity in vitro and in vivo. Of
the four synthesized compounds, 6b displayed the low-
est values (Table 1) indicating highest hydrophilicity
and adequate activity in the in vivo pilot assay.
Strikingly, in the process of expansion of TLR7 agonist
structures, 5/27 selectively designed compounds (Fig. 11)
displayed TLR7 inhibition in vitro. Conversely, we docked
the compounds back to co-crystallization of TLR7 (PDB
code: 5GMH). Agonist (R848) combined into TLR7 on the
Y-axis direction, tightened the neighboring loop, and formed a
triangle to activate the agonist-TLR7 binding protein.
Moreover, it could form a dimer system with another TLR7
protein. However, the five antagonists combined on the X-
Discussion
Chemotherapy-induced alopecia is a common side effect in can-
cer treatment [1]. The current treatments of alopecia are not ef-
fective for CIA treatment or prevention, and the side effects are
not neglected; for example, androgen receptor inhibitors cause
feminization and hyposexuality in men. Interestingly, scalp
cooling is effective in the clinic [21], and other methods are only
partially proved at the preclinical stage [22, 23].
TLR7 is an intracellular toll-like receptor expressed by var-
ious cell types. The involvement of TLR7 in the development
of autoimmunity has been demonstrated by multiple groups
using various approaches [24]. Imiquimod is a toll-like recep-
tor (TLR7/8) agonist that can be applied to mouse skin to elicit
200.00%
180.00%
160.00%
140.00%
120.00%
100.00%
80.00%
60.00%
40.00%
20.00%
0.00%
A1
A2
A3
A4
A5
A6
A7
A8
B1
B2
B3
B4
B5
B6
B7
B8
C1
C2
C3
C4
C5
C6
C7
C8
D1
D2
D3 CL097
First assay average value
Second assay average value
Fig. 12 In vitro TLR7 activity of 27 compounds

Invest New Drugs
140%
120%
100%
80%
60%
40%
20%
0%
-20%
-40%
NO inhibiꢀon
Cell proliferaꢀon (MTT)
Fig. 13 Effect of compounds on the viability of RAW264.7 cells
axis direction, interacted with neighboring 2–3 TLR7 loops
like cap combination mode, and locked onto the first site po-
sition. As a result, such conservative combination is difficult
for dimer system formation with two TLR7s (Fig. 5). The
strong interactions between the five antagonists with TLR7
might induce the conformational changes and affect the bind-
ing of the agonistic chemical ligands.
Herein, we designed a series of quinoline derivatives (all
unpublished). The derivative synthesis started with aniline,
and the key step was replacing the hydroxyl group by various
amine groups. One new compound (6b) displayed prolifera-
tive activity in vitro and in vivo through branch chain replace-
ment and triazole ring cyclization. The analysis by H-E stain-
ing revealed that the size and density of hair follicles on the
1a. TLR7-antagonist1(Compd C2)
1b. TLR7-antagonist1(Compd C2)_2D
2a. TLR7-antagonist2(Compd C3)
2b. TLR7-antagonist2(Compd C3)_2D
Fig. 14 Ligand-receptor interaction schematic diagrams by molecular docking

Invest New Drugs
3a. TLR7-antagonist3(Compd C4)
3b. TLR7-antagonist3(Compd C4)_2D
4a. TLR7-antagonist4(Compd C6)
4b. TLR7-antagonist4(Compd C6) _2D
5a. TLR7-antagonist5(Compd C7)
Fig. 14 (continued)
5b. TLR7-antagonist5(Compd C7)_2D
compound-treated groups was larger and thicker than the con-
trol groups. Lipophilicity might provide some valuable infor-
mation on drug design. However, the enzymatic mechanism
indicated that 6b was not a TLR7 agonist, and the putative
mechanism is yet to be investigated. A novel series of TLR7
antagonists were designed and discovered. The structural ac-
tivity correlation studies indicated that combination direction,
interaction with neighboring TLR7, and difficulty in the

Invest New Drugs
formation of the dimer system exerted a remarkable effect on
the antagonistic activity. Among the five compounds, C6
showed excellent antagonistic activity and NO inhibition ac-
tivity in vitro.
Detection methods: 1) One pouch of QUANTI-Blue™
(InvivoGen) was solubilized into 100 mL sterile water. 2)
The QUANTI-Blue™ medium was warmed to 37 °C for
30 min. 3) A volume of 180 μL/well QUANTI-Blue™ medi-
um was dispensed into a flat-bottom 96-well plate. 4) Then,
20 μL of the sample (supernatant of HEK-Blue™-hTLR7
cells with compounds) or negative control was added. 5)
The reaction was incubated at 37 °C for 15 min–6 h. 6) The
optical density (OD) was measured at 630 nm using a micro-
plate reader.
Materials and methods
Compounds synthesis and ointment prescription
Synthesis reagents were obtained from Shanghai Molbase Data
Technology Co., Ltd., and Shanghai TopScience Co., Ltd.
Compound cream was prepared from glyceryl monostearate
(2%), stearic acid (2%), albolene (5%), paraffin liquid (20%),
glycerin (25%), water (40%), SDS (0.9%), borneol (0.5%),
DMSO (4.6%), and target compound 6b (100 mg). All the ma-
terials were bought from Molbase™. Then, the mixture of com-
ponents was heated to 80 °C for 0.5 h in a water-bath.
RAW264.7 cell culture
RAW264.7 cells (Zhejiang Ruyao Biotech, China) were
cultured in DMEM medium (supplemented with 10%
fetal bovine serum and 1% penicillin/streptomycin) at
37 °C in a humidified atmosphere at 5% CO2 and seeded
into a 96-well plate at a density of 5 × 104 cells/well and
incubated for an additional 12 h. Next, different com-
pounds (40 μM) were added, and LPS (1 μg/mL)
(Sigma-Aldrich, USA) was added after 1.5 h; the plate
was incubated for an additional 24 h.
Isolation of mouse hair stem cells
The isolation of mouse hair follicle bulge stem cells was
carried out as follows: 10-day-old mice were sacrificed
by cervical dislocation. Then, 70% ethanol was added to
cover and soak the mice for 5–10 min. The two sides of
the tentacle skin were clipped (1.5 × 1.5 cm2). After rins-
ing with PBS (phosphate-buffered saline), the hair folli-
cles were extracted from the fatty end, and the intact hair
follicles were transferred into collagenase I/dispase
(1 mg/mL) at 37 °C for 30 min. The connective tissues
were removed with a syringe needle, and the hair folli-
cles were cut off. The middle part of the follicle was
collected, and skin pieces were transferred into PBS with
penicillin and streptomycin and 0.05% trypsin at 37 °C
for 10 min. The cells were suspended in DMEM with
10% fetal bovine serum (FBS), and the stem cells were
resuspended in complete medium and cultured into 24-
well plates.
A volume of 50 μL supernatant was collected for measur-
ing the NO production, using Griess reagent (Beyotime
Biotech, China); the OD values were measured at an excita-
tion wavelength of 540 nm.
Subsequently, 200 μL MTT (0.5 mg/mL) was added into
the residue and incubated at 37 °C for 4 h. Then, the dark
supernatants were removed, the left “black slag” was solubi-
lized in 150 μL DMSO, and OD was measured at 490 nm.
Each group of samples was subjected to four parallel
experiments.
Hair follicle proliferation activity in mice
Animal study protocols were approved by the Institutional
Animal Care and Use Committee at the National Cancer
Center (Chinese Academy of Medical Sciences). C57BL/6
and Balb/c mice were purchased from Vital River
Laboratory Animal Technology (Beijing, China).
General protocol: After anesthesia by tribromoethanol,
the back skin was daubed with liquified wax. The skin
hair was removed with solidified wax. On day 9 post-
depilation, the chemotherapeutic agent was administered
by i.p., and compound cream was daubed on the right
back skin and the control cream on the left back skin.
On day 21, mice were killed, and the back skin was
collected for histological analysis by H-E staining.
Normal mice were treated with compound cream
(4 mg/kg) q2d.
Cell proliferation was measured using a CCK-8 assay
(Dojindo, Kumamoto, Japan) following the manufac-
turer’s instructions, indicating that the compound (6b)
promoted cell proliferation at low dosage (1 and
5 μM) in early days (days 3 and 5).
HEK293 cell culture and activity detection
HEK 293 cells (HEK-Blue™-hTLR7 cells, InvivoGen) were
maintained in DMEM, 4.5 g/L glucose, 10% (v/v) FBS, 50 U/
mL penicillin, 50 μg/mL streptomycin, 100 mg/mL
Normocin™, and 2 mM L-glutamine. Cells were cultured at
37 °C in a humidified atmosphere at 5% CO2; the growth
medium was renewed two times a week.
Mice exhibiting acute toxicity were treated only one time
with high dose compound cream (800 mg/kg).

Invest New Drugs
Chemotherapeutic mice (CYP) were injected 150 mg CYP/
kg once, while chemotherapeutic mice (DOX) were injected
5 mg DOX/kg q3d.
Breast cancer mice were injected with 4 T1 cells (1 × 106/
mouse), and chemical doses were administered similarly as
chemotherapeutic mice above.
Groups
Day 9
Day 5
Day 3
Day 0
Day 6
Day 7
Kill 2
Day 14
Kill 2
Day 21
1
2
3
4
5
6
Normal mice
Depilation+cream
High dose cream
CYP + cream
Kill 2
Kill
Kill
Kill
Kill
Kill
Acute toxicity mice
CYP mice
Depilation
Depilation
DOX mice
DOX + cream
Breast Cancer mice A
Breast Cancer mice B
Inject 4 T1 cells
Inject 4 T1 cells
CYP
CYP
Depilation+cream
Depilation+cream
DOX
Ethical approval This article does not contain any studies with human
participants performed by any of the authors. Animal experiments were
approved by the Institutional Animal Care and Use Committee at
National Cancer Center, Chinese Academy of Medical Sciences.
(NCC2016A050, 9 March 2016) All applicable international, national,
and institutional guidelines for the care and use of animals were followed.
Molecular docking
The agonist-bound TLR7 structure (PDB code: 5GMH,
Resolution: 2.2 Å) was obtained from Protein Data Bank. The
co-crystallized structure was prepared using Molecular
Operating Environment (MOE) 2018.0101 for correcting the
structural issues (such as break bond and miss loop); also, hydro-
gen was added, and the partial charge was estimated. The 2D
structures of the five antagonists were converted to 3D in MOE
through energy minimization. The MOE dock was used for
docking simulation of the antagonists and predicting their bind-
ing affinity to the TLR7 protein structure. The original ligand-
binding pocket served as the active docking site. A classical
triangle match was selected as the placement method, and the
number of placement poses was set to 300. The output docking
poses were evaluated by the London dG score. Then, the rigid
receptor method was employed in the refinement step. The num-
ber of the final output docking poses was set to 50 that was
minimized using Amber12:EHT force field in MOE. The
GBVI/WSA dG score was used to evaluate the binding of the
antagonists with the TLR7. Finally, the binding mode was ana-
lyzed in MOE after minimizing the structure.
References
1. Nalluri R, Harries M (2016) Alopecia in general medicine. Clin
Med 16(1):74–78
2. Chen C, Chang Y, Liu H, Chen Y (2018) Cancer risk in patients
with alopecia areata: a nationwide population-based matched co-
hort study. Cancer Med 7:2153–2159
3. Seo H, Han S, Kim J (2018) Cancer risks among patients with
alopecia areata: a population-based case-control study in Korea. J
Am Acad Dermatol 78:209–211
4. Roe H (2011) Chemotherapy-induced alopecia: advice and support
for hair loss. Br J Nurs 20:S4–S11
5. Jayde V, Boughton M, Blomfield P (2013) The experience of
chemotherapy-induced alopecia for Australian women with ovarian
cancer. Eur J Cancer Care (Engl) 22:503–512
6. Wang J, Lu Z, Au LS (2006) Protection against chemotherapy-
induced alopecia. Pharm Res (Dordrecht) 23(11):2505–2514
7. Trüeb RM (2009) Chemotherapy-induced alopecia. Semin Cutan
Med Surg 28(1):11–14
Acknowledgements The key structures have applied Chinese patent.
(CN201711279295, CN108003153A).
We thank Zhejiang Ruyao Biotech Co.,LTD for technologic support.
8. Panteleyev A (2018) Functional anatomy of the hair follicle: the
secondary hair germ. Exp Dermatol 27:701–720
9. Chen R, Miao Y, Hu Z (2019) Dynamic nestin expression
during hair follicle maturation and the normal hair cycle.
Mol Med Rep 19(1):549–554
Funding The work was supported by National Natural Science
Foundation of China (No. 81603218).
10. Ralf P, Haslam Iain S, Sharov Andrey A et al (2013) Pathobiology
of chemotherapy-induced hair loss. Lancet Oncol 14:e50–e59
11. Amberg N, Holcmann M, Stulnig G, Sibilia M (2016) Effects of
Imiquimod on hair follicle stem cells and hair cycle progression. J
Invest Dermatol 136(11):2140–2149
12. Letada PR, Sparling JD, Norwood C (2007) Imiquimod in the treat-
ment of alopecia universalis. Cutis 79(2):138–140
13. Paus R, Cotsarelis G (1999) The biology of hair follicles. N Engl J
Med 341(7):491–497
Compliance with ethical standards
Conflict of interest Author Jincheng Yang declares that he has no con-
flict of interest. Author Kun Chen declares that she has no conflict of
interest. Author Bin Wang declares that he has no conflict of interest.
Author Liudi Wang declares that she has no conflict of interest. Author
Shuya Qi declares that she has no conflict of interest. Author Weihua
Wang declares that she has no conflict of interest.

Invest New Drugs
14. Kontogiorgis CA, Hadjipavloulitina DJ (2017) Synthesis and
antiinflammatory activity of coumarin derivatives. J Med Chem
48(20):6400–6408
15. Zheng Y, Hsieh J-C, Escandon J, Cotsarelis G (2016) Isolation of
mouse hair follicle bulge stem cells and their functional analysis in a
reconstitution assay. Methods Mol Biol Chapter 8. 1453:57-69
16. Ma DR, Yang EN, Lee ST (2004) A review: the location, molecular
characterisation and multipotency of hair follicle epidermal stem
cells. Ann. Acad. Med. Singap 33:784–788
21. Young A, Arif A (2016) The use of scalp cooling for chemotherapy-
induced hair loss. Br J Nurs 25:S22–S27
22. Katikaneni R, Ponnapakkam T, Matsushita O, Sakon J, Gensure R
(2014) Treatment and prevention of chemotherapy-induced alopecia
with PTH-CBD, a collagen-targeted parathyroid hormone analog, in
a non-depilated mouse model. Anti-Cancer Drugs 25(1):30–38
23. Belen R-G, Margit J, Jamie F et al (2018) Pathogenesis and treat-
ment options for chemotherapy-induced alopecia: a systematic re-
view. Int J Dermatol 57:1417–1424
17. Meihua G, Pan Z, Chunyang L et al (2018) Protective mechanism
of adipose-derived stem cells in Remodelling of the skin stem cell
niche during Photoaging. Cell Physiol Biochem 51:2456–2471
18. Sasaki S, Baba T, Muranaka H, Tanabe Y, Takahashi C,
Matsugo S, Mukaida N (2018) Involvement of prokineticin
2-expressing neutrophil infiltration in 5-fluorouracil-induced
aggravation of breast cancer metastasis to lung. Mol Cancer
Ther 17:1515–1525
24. Rubtsova K, Marrack P, Rubtsov AV (2015) TLR7, IFNγ, and T-
bet: their roles in the development of ABCs in female-biased auto-
immunity. Cell Immunol 294(2):80–83
25. Swindell William R, Michaels Kellie A, Sutter Andrew J et al
(2017) Imiquimod has strain-dependent effects in mice and does
not uniquely model human psoriasis. Genome Med 9:24
19. Wang D, Bhagat L, Yu D, Zhu FG, Tang JX, Kandimalla ER,
Agrawal S (2009) Oligodeoxyribonucleotide-based antagonists
for toll-like receptors 7 and 9. J Med Chem 52(2):551–558
20. Drake M, Scott G, Proskocil B et al (2013) Toll-like receptor 7
rapidly relaxes human airways. Am J Respir Crit Care Med
188(6):664–672
Publisher’s note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.