Design, synthesis, and optimization of a series of 2-azaspiro[3.3]heptane derivatives as orally bioavailable fetal hemoglobin inducers

Article abstract of DOI:10.1016/j.bmcl.2020.127425

Pharmacological reactivation of the γ-globin gene for the production of fetal hemoglobin (HbF) is a promising approach for the management of β-thalassemia and sickle cell disease (SCD). We conducted a phenotypic screen in human erythroid progenitor cells

Full text of DOI:10.1016/j.bmcl.2020.127425

Journal Pre-proofs
Design, synthesis, and optimization of a series of 2-azaspiro[3.3]heptane de‐
rivatives as orally bioavailable fetal hemoglobin inducers
Katsushi Katayama, Tomoyuki Tsunemi, Kazuo Miyazaki, Kouichi Uoto,
Ryosuke Yoshioka, Hideki Terashima, Maki Terakawa, Kyoko Yamashiro,
Munetada Haruyama, Hiroaki Maeda, Tomohiro Makino
PII:
S0960-894X(20)30536-9
DOI:
Reference:
https://doi.org/10.1016/j.bmcl.2020.127425
BMCL 127425
To appear in:
Bioorganic & Medicinal Chemistry Letters
Received Date:
Revised Date:
Accepted Date:
30 May 2020
8 July 2020
16 July 2020
Please cite this article as: Katayama, K., Tsunemi, T., Miyazaki, K., Uoto, K., Yoshioka, R., Terashima, H.,
Terakawa, M., Yamashiro, K., Haruyama, M., Maeda, H., Makino, T., Design, synthesis, and optimization of a
series of 2-azaspiro[3.3]heptane derivatives as orally bioavailable fetal hemoglobin inducers, Bioorganic &
Medicinal Chemistry Letters (2020), doi: https://doi.org/10.1016/j.bmcl.2020.127425
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. Please note that, during the production process, errors
may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
©
2020 Published by Elsevier Ltd.

Bioorganic & Medicinal Chemistry Letters
Design, synthesis, and optimization of a series of 2-azaspiro[3.3]heptane derivatives
as orally bioavailable fetal hemoglobin inducers
*
,a
a
b
a
a
Katsushi Katayama , Tomoyuki Tsunemi , Kazuo Miyazaki , Kouichi Uoto , Ryosuke Yoshioka , Hideki
a
a
a
a
a
Terashima , Maki Terakawa , Kyoko Yamashiro , Munetada Haruyama , Hiroaki Maeda , Tomohiro
a
Makino
Asubio Pharma Co. Ltd., 6-4-3 Minatojima-Minamimachi, Chuo-ku, Kobe, Hyogo 650-0047, Japan
a
Present address: Shinagawa R&D Center, Daiichi-Sankyo Co. Ltd., 1-2-58 Hiromachi, Shinagawa-ku, Tokyo 140-8710, Japan
Present address: MiCAN Technologies, Inc., KKVP, Goryo-ohara, Nishikyo-ku, Kyoto 615-8245, Japan
Corresponding author. E-mail address: katayama.katsushi.ne@daiichisankyo.co.jp
b
*
ARTICLE INFO
ABSTRACT
Article history:
Received
Received in revised form
Accepted
Pharmacological reactivation of the γ-globin gene for the production of fetal hemoglobin (HbF) is a
promising approach for the management of β-thalassemia and sickle cell disease (SCD). We
conducted a phenotypic screen in human erythroid progenitor cells to identify molecules that could
induce HbF, which resulted in identification of the hit compound 1. Exploration of structure–activity
relationships and optimization of ADME properties led to 2-azaspiro[3.3]heptane derivative 18, which
is more rigid and has a unique structure. In vivo using cynomolgus monkeys, compound 18 induced a
significant dose-dependent increase in globin switching, with developable properties. Moreover,
compound 18 showed no genotoxic effects and was much safer than hydroxyurea. These findings
could facilitate the development of effective new therapies for the treatment of β-hemoglobinopathies,
including SCD.
Available online
Keywords:
γ-globin
globin switching
fetal hemoglobin
β-thalassemia
sickle cell disease (SCD)
structure–activity relationship (SAR)
2
020 Elsevier Ltd. All rights reserved.
Hemoglobinopathies are a group of inherited single-gene
diseases caused by the abnormal production or structure of
hemoglobin molecules. β-Thalassemia is a monogenic disorder
characterized by reduced or absent synthesis of the β-globin
chain, one of the main components of hemoglobin A (HbA,
drug approved by the US Food and Drug Administration for the
specific treatment of SCD. HU has the potential to induce
expression of the γ-globin (HBG) gene and reduce expression of
1
2
the β-globin (HBB) gene. Although HU is important in the
treatment of sickle cell patients, the effect is limited in many
1
–3
S
13–15
α β ). In contrast, sickle cell disease (SCD) is caused by the β -
patients.
Most notably, there is considerable variation in
2
2
globin mutation, which leads to the production of hemoglobin S,
resulting in the formation of long and rigid multistranded fibers
that transform the shape of red blood cells (RBCs) into elongated
patient responses to HU, and many adult patients still require
clinical disease management because of the limited induction of
1
6
HbF and the high toxicity. Therefore, there is an urgent need to
identify new types of agent that can induce the production of
HbF with greater efficiency and lower toxicity for the treatment
4
–6
crescents.
Hemoglobin within RBCs is known to bind to
oxygen molecules in the lungs, and deliver oxygen to tissues
throughout the body. In adults, hemoglobin normally consists of
four protein subunits: two subunits of α-globin and two subunits
of β-globin. However, it has been observed that the γ-globin
chain, rather than the β-globin chain, is the dominant α-globin
counterpart in the RBCs of newborns, resulting in the formation
of fetal hemoglobin (HbF). Soon after birth, γ-globin is replaced
1
7
of various hemoglobinopathies.
In an effort to identify novel inducers of HbF with desirable
pharmacokinetic (PK) and pharmacodynamic (PD) properties, we
first conducted a high-throughput phenotypic screen using human
1
8,19
umbilical cord blood-derived erythroid progenitor cells
and
identified the hit compound 1 as having weak potency (EC50 >10
μM, Emax 24%; Fig. 1). Compound 1 was shown to have an EC50
of 2.8 μM against human erythroid progenitor cells derived from
7
–10
by β-globin through a process referred to as globin switching.
Reactivation of HbF has been identified as an effective
+
20
therapeutic approach for β-hemoglobinopathies, including β-
bone marrow-derived (BM-)CD34 hematopoietic stem cells,
7,8,10
thalassemia and SCD.
Specifically, pharmacological
and demonstrated favorable physicochemical properties with a
molecular weight (MW) of 305 g/mol and a log D (pH 7.4) of
1.4. Compound 1 was also shown to have comparable efficacy to
reactivation of the γ-globin gene can increase the production of γ-
globin and HbF to compensate for the defective β-globin,
resulting in reduced hemoglobinopathies and effective
+
known HbF inducers, such as decitabine, in the BM-CD34 cell
1
1
erythropoiesis. Therefore, induction of HbF is considered an
appropriate therapy for β-thalassemia and SCD. The
ribonucleotide reductase inhibitor hydroxyurea (HU) was the first
assay. Despite these prospects for potential therapeutic
implementation, we identified several issues with compound 1
that indicated the potential for further improvement.

a
b
29
30
31
32
1
c
d
BM-CD34 EC50: 2.8 μM
metabolic stability (%) human/monkey: 87/10
monkey p.o. (10 mg/kg) AUClast: 4365 ng*h/mL
33
34
9
t1/2: 1.2 h
Fig. 1. Chemical structure, activity, and ADME profile of hit compound 1.
e
f
First, compound 1 showed poor metabolic stability (MS) in
monkey liver microsomes (10%), but high MS in human liver
microsomes (87%; Fig. 1). Second, compound 1 had a poor PK
profile in monkeys (low area under the plasma drug
concentration–time curve from zero to last concentration
35
36
37
38
g
h
3
9
40
41
[
AUC ] and short elimination half-life; for further details, see
last
Table 5). Herein, we describe our efforts to overcome these
defects to find novel compounds with an improved PK profile
while maintaining pharmacological efficacy.
i
j
The synthetic procedure for the target compounds 1–8 is
described in Scheme 1. Amide bond formation with
commercially available compounds 24a–24g was accomplished
by activation with HABU or EDCI to afford compounds 1–7,
respectively. Compound 8 was obtained by amidation of
carboxylic acid 24a with amine 26, followed by hydrolysis and
then condensing with dimethylamine. The synthesis of
compounds 9–11 is described in Scheme 2. Commercially
available pyrrole-2-carboxylate 29 was alkylated, followed by
42
10
11
Scheme 2. Synthesis of compounds 9–11. Reagents and conditions: a) NaH,
DMF, rt to 80 °C, 1 h, 77%; b) 3-amino-N,N-diethylpropanamide, K
DMF, rt to 80 °C, 2 h, 12%; c) DMSO, 150 °C, 4 h, 87%; d) Pd(PPh
CO , dioxane, H
h, 33%; f) DIPEA, CH
9%; h) AcOH, H O, rt to 90 °C, 2 h, 100%; i) NH
17%; j) NaH, MeI, DMF, 0 °C to rt, 2 h, 76%.
2
CO
Cl
3
,
,
3
)
2
2
K
2
3
2
O, rt to 120 °C, 1 h, μW, 33%; e) MeOH, rt to 50 °C, 16
Cl , 0 °C, 1 h, 45%; g) NaH, EtOH, 0 °C to rt, 17 h,
OAc, EtOH, rt, 17 h,
2
2
8
2
4
further N-alkyation to give amine 32.
This intermediate 32 was converted to lactam 33 by heating
in DMSO at 150 °C, followed by Suzuki coupling with borate 34
to afford compound 9. The synthesis of compounds 10 and 11
commenced with an aza-Michael addition of amine 36 to methyl
acrylate 35, affording compound 37. N-Acylation was
subsequently performed, followed by lactamization to afford
compound 40. After decarboxylation of compound 40, the
Hantzsch reaction between piperidinedione 41 and bromoketone
42 was utilized to afford compound 10. Compound 11 was
obtained by N-methylation of compound 10. Compounds 12–14
were obtained by amidation of the corresponding amine (Scheme
a
R = p-fluorophenyl;24a
R = o-fluorophenyl; 24b
R = m-fluorophenyl; 24c
R = phenyl; 24d
25
R = p-fluorophenyl; 1
R = o-fluorophenyl; 2
R = m-fluorophenyl; 3
R = phenyl; 4
R = cyclohexyl; 24e
R = cyclohexyl; 5
3
). Compounds 15 and 16 were synthesized from compounds 43a
a
and 43b in a manner analogous to that of compound 8. The
synthesis of compounds 17–23 is described in Scheme 4.
Commercially available compounds 24a and 24d were
condensed with amine 46 to afford compound 47 and 48,
respectively.
24f
25
6
a
a
R = CH CH CH N(Me) , 12
2
2
2
2
24a
R = CH CH CH CON(Me) , 13
2
2
2
2
R = CH CON(Me) , 14
2
2
24g
25
7
HCl
a
a
b
24a
26
24a
R = CH(CH )CH CO Et, 43a
R = CH(CH )CH CO Et, 44a
3 2 2
3
2
2
27
R = CH CH(CH )CO Me, 43b R = CH CH(CH )CO Me, 44b
2 3 2 2 3 2
b
c
a
R = CH(CH )CH CO H, 45a
R = CH(CH )CH CON(Me) , 15
3 2 2
3
2
2
28
8
R = CH CH(CH )CO H, 45b
R = CH CH(CH )CON(Me) , 16
2 3 2
2
3
2
Scheme 1. Synthesis of compounds 1–8. Reagents and conditions: a) HABU
Scheme 3. Synthesis of compounds 12–16. Reagents and conditions: a)
corresponding amine, HATU or HBTU, DIPEA, DMF, 0 °C to rt, 1 h, 15%–
or EDCI, DIPEA or Et
rt, 2 h, 52%.
3 2 2
N, DMF, rt, 1 h, 25%–62%; b) LiOH-H O, THF, H O,

7
0%; b) LiOH-H
2 2 2
O, THF, H O, rt, 4 h; c) Me NH-HCl, HATU, DIPEA,
DMF, 0 °C to rt, 1 h, 47%–71% over 2 steps.
1
2
2.8
>5
56
37
87/10
a, for 47
b, for 48
>100/4
3
>5
-
>100/55
2
2
4a; R1 = F
4d; R1 = H
47; R1 = F
48; R1 = H
46
4
5
1.7
>5
51
-
81/4.7
c
d
>100/34
a
Compound concentration that induces HbF by 50% if the maximum
induction of HbF by decitabine is taken as 100%. Details of in vitro assays
+
19
17; R1
18; R1
using BM-CD34 cells are described in the paper by Makino et al.
9; R1 = F
0; R1 = H
= F
= H
4
5
b
e, for 19
f, for 20, 21
g, for 22
Maximum effect of the compound as a percentage of that of decitabine.
+
Human BM-CD34 cells were used in the assay.
c
Remaining percentage of test compounds after 0.5 h incubation with
human/monkey liver microsomes (0.5 mg/mL).
R2 = SO CH , 19
Based on the results in Table 1, the substituent at the phenyl ring
was fixed as a para-fluoro group. We next investigated the acyl
isoxazole core (Table 2). Replacing the isoxazole ring with a
thiazole (6) or thiazolidine (7) ring decreased potency, as did N-
methylation of compound 1 (8), suggesting the possibility that
the intramolecular hydrogen bond between amide N-H and the
oxygen bond of isoxazole would be important for potency. Based
on these results, we next designed fused ring derivatives (9–11),
but none of these compounds exhibited any potency.
2
3
R2 = COCH CH , 20
2
3
R2 = COCH(CH ) , 21
3
2
R2 = CH CH , 22
2
3
h
i
24d
51
52
23
Scheme 4. Synthesis of compounds 17–23. Reagents and conditions: a) 24a,
COMU, Et N, DMF, 0 °C to rt, 3 h, 56%; b) (i) 24d, SOCl , Et N, DMF,
CH Cl , rt to 50 °C, 4 h; (ii) 46, Et N, CH Cl , 0 °C to rt, 1 h, 91% over 2
steps; c) TFA, CH Cl , 0 °C, 3 h, 54%–85%; d) AcCl, Py., CH Cl , 0 °C to rt,
h, 80%–88%; e) MsCl, Py, CH Cl , 0 °C to rt, 3 h, 71%; f) carbonyl
chloride, Py, CH Cl , DMF, 0 °C to rt, 3 h, 90%–96%; g) acetaldehyde,
NaBH(OAc) , THF, 0 °C to rt, 1 h, 61%; h) HATU, DIPEA, DMF, 0 °C to rt,
h, 41%; i) (i) TFA, CH Cl , 0 °C, 3 h, 61%; (ii) AcCl, Py, CH Cl , 0 °C to
rt, 1 h, 86%.
Table 2. SAR of acyl isoxazole moieties
3
2
3
2
2
3
2
2
2
2
2
2
1
2
2
2
2
3
1
2
2
2
2
+
MS
(human/mo
nkey) (%)
BM-CD34
EC50 (μM)
b
Compound
a
E
max (%)
After removal of the Boc group under acidic conditions,
acetylation of the resultant amines 49 and 50 afforded the target
compounds 17 and 18, respectively. Compound 19 was obtained
by N-sulfonylation of compound 50. Likewise, compounds 20
and 21 were obtained by carbonylation with corresponding
carbonyl chloride. Compound 22 was obtained by reductive
amination with acetaldehyde. Compound 23 was synthesized
from compounds 24d and 51 in a manner analogous to that of
compound 17.
Initial evaluation of the structure–activity relationships (SAR)
was directed towards modifications of substituents at the 3-
position of the isoxazole ring (Table 1). Shifting the fluoro
substituent from the para (1) to the ortho (2) or meta (3) position
significantly reduced potency. Deletion of the fluoro group in
compound 1 (4) resulted in comparable potency, but compound 4
did not exhibit improved MS in monkey liver microsomes.
Furthermore, the cyclohexyl derivative (5) exhibited decreased
potency.
c
1
2.8
>5
>5
56
-
87/10
53/0.1
90/32
6
d
7
-
8
9
>5
>5
-
-
98/36
94/41
Table 1. SAR of substituents at the 3-position of the isoxazole ring
1
1
0
1
>5
>5
-
-
99/39
>100/42
+
MS
BM-CD34
EC50 (μM)
E
max
(%)
Compound
R
a
b
(human/monkey)
c
(%)

a
Compound concentration that induces HbF by 50% if the maximum
d
induction of HbF by decitabine is taken as 100%. Details of in vitro assays
16
17
F
F
>5
-
61/52
84/79
+
19
using BM-CD34 cells are described in the paper by Makino et al.
b
Maximum effect of the compound as a percentage of that of decitabine.
+
Human BM-CD34 cells were used in the assay.
2.9
64
c
Remaining percentage of test compounds after 0.5 h incubation with
human/monkey liver microsomes (0.5 mg/mL).
d
Diastereomer mixture.
1
8
H
H
2.8
3.9
54
45
>100/64
76/87
Next, we explored the SAR of the dimethylamide group
Table 3). Replacement of dimethylamide with a dimethylamino
12) decreased potency, suggesting that the amide moiety was
(
(
1
2
9
0
important for potency. Lengthening (13) or shortening (14) the
alkyl chain by a single carbon atom retained its single-digit
micromolar potency. The introduction of a methyl group in the
linker between the isoxazole and dimethyl amide moieties
decreased potency (15, 16). Based on the loss of potency in
compounds 15 and 16, it was considered that the space in the
linker position was narrow. Therefore, 2-azaspiro[3.3]heptane
derivatives were designed and synthesized with the aim of
increasing potency by conformational rigidification; as expected,
compound 17 exhibited potency with EC50 values of 2.8.
Likewise, deletion of a fluoro group of compound 17 (18) led to
retention of the potency. Replacing the acetyl group of compound
H
H
>5
>5
47
27
68/69
79/55
2
2
1
2
H
H
5.0
>5
50
-
100/87
50/17
1
8
with
methylpropionyl (21), or ethyl (22) group decreased potency.
-Azaspiro[3.3]heptane, which is more rigid and has a unique
structure, is recognized as the bioisostere of 4-
a
methanesulfonyl (19), propionyl (20), 2-
2
23
2
1,22
23–
aminopiperidine,
and this moiety is known to improve MS.
2
6
a
As expected, compounds 17 and 18 exhibited improved MS in
Compound concentration that induces HbF by 50% if the maximum
monkey liver microsomes, despite having an increased number of
carbons compared with compound 1. This seems to be
attributable to increased polarity and lower cLogP (e.g., 1.70 vs.
induction of HbF by decitabine is taken as 100%. Details of in vitro assays
+
19
using BM-CD34 cells are described in the paper by Makino et al.
b
Maximum effect of the compound as a percentage of that of decitabine.
1
.35 in compounds 1 and 17, respectively) in these spiro
+
Human BM-CD34 cells were used in the assay.
derivatives. In addition, acylazetidine derivative 23 having a non-
rigid structure was synthesized for comparison, and compound
c
Remaining percentage of the test compounds after 0.5 h incubation with
2
3 showed decreases in both potency and MS in monkey liver
human/monkey liver microsomes (0.5 mg/mL).
microsomes; thus, the superiority of the spiro structure was
confirmed. Based on these results, representative compounds 17
and 18 were further profiled in additional ADME, off-target, and
PK assays (Tables 4 and 5).
d
Racemate.
e
Not tested.
Table 3. SAR of dimethylamide moieties
R2
R1
+
MS
(human/
monkey)
1
2
BM-CD34
EC50 (μM)
E
max
(%)
Compound
R
F
R
a
b
c
1
2.8
>5
4.1
6.1
>5
56
87/10
e
Fig. 2. Cell viability measured using propidium iodide viability assay. Data
are shown as mean ± SD of three biological replicates.
1
1
1
1
2
3
4
5
F
F
F
F
-
N.T.
Next, compounds 17, 18, and 23 were used to verify the
cytotoxicity of a series of 2-azaspiro[3.3]heptane derivatives and
its analog (Fig. 2). All compounds tested in the cell viability
assay using propidium iodide were less toxic. In particular,
compound 17 was non-toxic and even compound 18 was less
toxic (only about 20%) at 15 μM, which is a fivefold higher
concentration than each EC50 (2.9 and 2.8 μM, respectively).
e
55
41
-
N.T.
84/11
91/35
d

Table 4. ADMET profiles of compounds 1, 17, and 18.
a
Solubility
% Direct inhibition of
CYP at 10 μM
(1A2/2C9/2D6/3A4)
6/6/5/<0
b
% hERG
inhibition at ERM
10 μM
Toxicity
(IVtMN/Ames
assays)
Negative/negative
Negative/negative
Negative/negative
Compoun
d
% Plasma protein
binding (monkey)
MBI
(%)
(μg/mL)
(JP1/JP2)
c
1
>100/>100
>100/>100
>100/>100
58
72
59
97
98
94
7.1
N.T.
18
N.T.
c
1
1
7
8
1/<0/1/<0
3/<0/2/<0
Negative
Negative
a
JP1/JP2: Japanese Pharmacopoeia first/second test fluid (pH = 1.2/6.8).
b
c
Percentage remaining after 100 μM test compound was reacted with CYP3A4 probe substrates for 30 min in human liver microsomes.
Not tested.
Table 5. Monkey PK profiles of compounds 1, 17, and 18.
a
PO (n=2)
Compound
b
Dose (mg/kg)
C
max (ng/mL)
AUClast (ng·h/mL)
½
t (h)
1
1
1
10
1
1362
460
4365
2354
4212
1.0
5.9
3.7
7
8
1
545
a
Dosing vehicle: 0.5% HPC.
Representative compounds were analyzed to determine their
significant induction compared with the vehicle control group at
a dose of 30 mg/kg.
solubility, protein binding, cytochrome P450 (CYP) inhibition,
and mechanism-based inhibition (MBI), as well as off-target and
genotoxic potential (Table 4). Compounds 1, 17, and 18
exhibited better solubility at both acidic pH and neutral pH, and
these compounds were moderately bound to monkey plasma
protein (58%–72%). Compounds 1, 17, and 18 did not exhibit
CYP inhibitory activity at a concentration of 10 μM, which is a
clinically relevant concentration, and also did not exhibit MBI of
CYP3A4. Furthermore, these compounds showed low human
ether-a-go-go-related gene (hERG) inhibition at a concentration
of 10 μM, and were negative in the Electrophilic Reactive
Metabolite (ERM), Ames, and in vitro micronucleus (IVtMN)
assays. In a previous study, we confirmed that the lead compound
In summary, we have developed potent and orally
bioavailable inducers of HbF originating from hit compound 1. 2-
Azaspiro[3.3]heptane derivative 18, which is more rigid and has
a unique structure, showed improved potency and favorable PK
properties, including good MS in monkey liver microsomes.
Notably, in vivo using normal cynomolgus monkeys, compound
18 dose-dependently induced γ-globin gene expression. There is
a need to identify new, safe, and effective inducers of HbF that
solve issues such as HU cytotoxicity, and compound 18 is
expected to be useful in developing therapeutic agents for the
treatment of β-thalassemia and SCD. Further, synthesis of
compound 18 required only three steps, so it can easily be scaled
up and is developable in terms of production. At present, our
laboratory is conducting further structure optimization and
detailed mechanistic analysis using compound 18 to afford
clinical candidates.
1
did not inhibit any epigenetic modulatory enzymes, such as
19
histone deacetylase or DNA methyltransferases. Therefore, it is
reasonable to assume that compounds 1, 17, and 18 are not
genotoxic, and that they are likely to be much safer than HU or
1
4,27
decitabine, both of which are known epigenetic modulators.
Compounds 1, 17, and 18 were subjected to a PK study in
monkey (Table 5). All compounds were administered orally at
doses of 1–10 mg/kg. Among compounds 1, 17, and 18, the
overall profile of compound 18 in monkey was the most desirable
because it showed high maximum plasma exposure (AUClast =
4
3
212 ng·h/mL) to the dose and a long elimination half-life (t =
.7 h) at a dose of 1 mg/kg. Based on its PK properties
½
(especially given high AUC), compound 18 was selected for
further in vivo study. The ability of compound 18 to induce HbF
in vivo was investigated in normal cynomolgus monkeys (Fig. 3).
This experiment was based on the protocols described by Makino
1
9
et al. Expression of HBG and HBB mRNA was quantified
before (pre) and after (post) 6 consecutive days of oral
administration of compound 18. Post/pre ratios were calculated
for HBG and HBB to determine changes in individual gene
expression. In addition, the ratio of post/pre HBG expression to
post/pre HBB gene expression was calculated as an indicator of
Fig. 3. Induction of HbF by compound 18 in normal cynomolgus monkeys.
Cynomolgus monkeys were administered compound 18 (15 or 30 mg/kg,
twice a day) or vehicle orally from Day 0 to Day 6. Data show the mean
(±SD) fold increase in the ratio of Hb γ-/β-subunit mRNA expression (n = 3
1
9
HbF induction. We also attempted to evaluate the HbF protein
expression level directly from bone marrow cells according to
or 4; *p < 0.05, Dunnett’s test). In one monkey, transient hypoglycemia was
observed after the oral administration of 30 mg/kg compound 18, and this
monkey was excluded from the experiment. There were no other safety issues
observed throughout the experiment.
1
9
methods described by Makino et al. However, we could not
detect HbF protein at either post- and pre-administration, due to
the limited sensitivity of the assay. Based on these findings,
compound 18 dose-dependently induced HbF, with statistically
Declaration of interest

The authors declare that they have no known competing financial
interests or personal relationships that could have influenced the
work reported in this paper.
13. Bakanay SM, Dainer E, Clair B, et al. Blood. 2005;105:545.
1
1
4. McGann PT, Ware RE. Curr Opin Hematol. 2011;18:158.
5. Steinberg MH, McCarthy WF, Castro O, et al. Am J Hematol.
2
010;85:403.
1
1
6. Steinberg MH, Lu ZH, Barton FB, Terrin ML, Charache S, Dover GJ.
Blood. 1997;89:1078.
7. For a recent example, see: Benowitz AB, Eberl HC, Erickson-Miller
CL, et al. Bioorg Med Chem Lett. 2018;28:3676–3680.
8. Kurita R, Suda N, Sudo K, et al. PloS One. 2013;8:e59890.
9. Makino T, Haruyama M, Katayama K, et al. Biochem Pharmacol.
Appendix A. Supplementary data
Supplementary data to this article can be found online at XXXX.
1
1
Acknowledgments
2
020;171:113717.
2
2
0. Lee JA, Berg EL. J Biomol Screen. 2013;18:1212.
1. Radchenko DS, Pavlenko SO, Grygorenko OO, Volochnyuk DM,
Shishkina SV, Shishkin OV, Komarov IV. J Org Chem.
The authors thank Dr. Daisuke Nakai for conducting the
pharmacokinetic evaluation and Dr. Tsuyoshi Muto for helpful
discussions on medicinal chemistry and manuscript preparation.
2
010;75:5941–5952.
2
2
2. Burkhard JA, Wagner B, Fischer H, Schuler F, Müller K, Carreira EM.
Angew Chem Int Ed. 2010;49:3524–3527.
3. Boiteau JG, Ouvry G, Arlabosse JM, et al. Bioorg Med Chem.
2018;26:945–956.
References
1
2
.
.
Galanello R, Origa R. Orphanet J Rare Dis. 2010;5:11.
Musallam KM, Taher AT, Rachmilewitz EA. Cold Spring Harbor
Persp Med. 2012;2:a013482.
24. Caldwell RD, Qiu H, Askew BC, Bender AT, Brugger N, et al. J Med
Chem. 2019;62:7643–7655.
3
4
5
6
.
.
.
.
Higgs DR, Engel JD, Stamatoyannopoulos G. Lancet. 2012;379:373.
Weatherall DJ. Blood. 2010;115:4331.
Xuejin L, Ming D, George L, et al. J Biomech. 2017;50:34–41.
Lanzkron S, Strouse JJ, Wilson R, et al. Ann Intern Med.
25. Zhang X, Sheng X, Shen J, et al. ACS Med Chem Lett. 2020.
https://pubs.acs.org/doi/10.1021/acsmedchemlett.9b00395
26. Gehling VS, McGrath, JP, Duplessis M, et al. ACS Med Chem Lett.
2020;11:1213–1220
2
018;148:939.
27. Molokie R, Lavelle D, Gowhari M, et al. PLoS Med.
2017;14:e1002382.
7
8
9
1
1
.
.
.
Bauer DE, Kamran SC, Orkin SH. Blood. 2012;120:2945.
Ginder GD. Transl Res. 2015;165:115.
Vinjamur DS, Bauer DE, Orkin SH. Br J Haematol. 2018;180:630.
Graphical Abstract
0. Wilber A, Nienhuis AW, Persons DA. Blood. 2011;117:3945.
1. Fard AD, Hosseini SA, Shahjahani M, Salari F, Jaseb K. Int J Hematol
Oncol Stem Cell Res. 2013;7:47.
1
2. Charache S, Terrin ML, Moore RD, et al. N Engl J Med.
1
995;332:1317.
Declaration of interests
The authors declare that they have no known
☒
competing financial interests or personal
relationships that could have appeared to influence
the work reported in this paper.
☐
The authors declare the following financial
interests/personal relationships which may be
considered as potential competing interests:
Design, synthesis, and optimization of a series of 2-
Leave this area blank for abstract info.
azaspiro[3.3]heptane derivatives as orally bioavailable
fetal hemoglobin inducers
*,a
a
b
a
a
Katsushi Katayama , Tomoyuki Tsunemi , Kazuo Miyazaki , Kouichi Uoto , Ryosuke Yoshioka ,
a
a
a
a
a
Hideki Terashima , Maki Terakawa , Kyoko Yamashiro , Munetada Haruyama , Hiroaki Maeda ,
a
Tomohiro Makino