Identification of a novel series of potent and selective CCR6 inhibitors as biological probes

Article abstract of DOI:10.1016/j.bmcl.2018.07.042

CCR6 has been implicated in both autoimmune diseases and non-autoimmune diseases. Thus, inhibition of CCR6-dependent cell migration is an attractive strategy for their treatment. An orally available small molecule inhibitor of CCR6 could therefore be a us

Full text of DOI:10.1016/j.bmcl.2018.07.042

Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Identification of a novel series of potent and selective CCR6 inhibitors as
biological probes
Taisuke Tawaraishia,⁎, Nobuki Sakauchia,b, Kousuke Hidakaa, Kyoko Yoshikawaa,
Toshitake Okuia, Haruhiko Kunoa, Ikumi Chisakia, Kazuyoshi Asoa,b
a
Pharmaceutical Research Division, Takeda Pharmaceutical Company Limited, 26-1, Muraoka-Higashi 2-chome, Fujisawa, Kanagawa 251-8555, Japan
Axcelead Drug Discovery Partners, Inc., 26-1, Muraoka-Higashi 2-chome, Fujisawa, Kanagawa 251-0012, Japan
b
A R T I C L E I N F O
A B S T R A C T
Keywords:
CCR6
Biological probe
Autoimmune diseases
Non-autoimmune diseases
Cell migration
CCR6 has been implicated in both autoimmune diseases and non-autoimmune diseases. Thus, inhibition of
CCR6-dependent cell migration is an attractive strategy for their treatment. An orally available small molecule
inhibitor of CCR6 could therefore be a useful biological probe for the pathophysiological studies. Initial SAR
study of a hit compound provided potent N-benzenesulfonylpiperidine derivatives that suppressed CCL20-in-
duced Gi signals. By subsequent scaffold morphing of the central ring and further optimization, we identified a
novel series of 1,4-trans-1-benzenesulfonyl-4-aminocyclohexanes as potent and selective CCR6 inhibitors with
good pharmacokinetic properties. Our compounds showed good correlation between Gi signal inhibitory activity
and cell migration inhibitory activity in human CCR6-transfected CHO cells. In addition, representative com-
pound 35 potently inhibited CCR6-dependent cell migration and the increase in ERK phosphorylation in human
primary cells. Therefore, the compound could be used effectively as a biological probe against human CCR6.
Gi signal
ERK phosphorylation
1,4-trans-1-benzenesulfonyl-4-
aminocyclohexane
Chemokines have essential roles in the homeostasis of immune
systems. Among their receptors, CC chemokine receptor 6 (CCR6) is
preferentially expressed on B cells, Th17 cells, and dendritic cell sub-
sets.1,2 CC chemokine ligand 20 (CCL20), also known as macrophage
inflammatory protein-3α (MIP-3α), binds only to CCR6. Binding of
CCL20 to CCR6 induces cell migration, an event reported to depend on
the activation of both the extracellular signal-regulated kinase (ERK)
and Gi signal pathways.3 CCR6 regulates the migration of inflammatory
cells,4,5 and CCR6 antibody has been shown to inhibit CCL20-stimu-
lated cell migration.5 CCL20 is produced in tissues such as the skin,
lung, and intestinal mucosa, and its expression is increased by various
inflammatory stimuli. The association of CCR6 with autoimmune dis-
eases such as rheumatoid arthritis5,6 and psoriasis7 has been demon-
strated by genome-wide association study or in vivo study in disease
models. CCR6 is also involved in non-autoimmune diseases such as
cancer8,9 and atherosclerosis.10 Thus, inhibition of CCR6-dependent cell
migration would be an attractive strategy for the treatment of various
diseases, and an orally available small molecule inhibitor of CCR6 could
be a useful biological probe for both in vitro and in vivo pathophy-
siological studies.
migration of immune cells. First, high-throughput screening based on
CCL20-induced Gi signal inhibitory activity assessed by measuring
produced cyclic adenosine monophosphate in human CCR6-transfected
CHO cells was performed. Consequently, N-benzenesulfonylpiperidine
derivative 1 was identified as an inhibitor of human CCR6 (Fig. 1).
Interestingly, this compound had high selectivity over human CCR1 and
CCR7. Both CCR1 and CCR7 are also associated with cell migration,11
and among CCRs, CCR7 has the highest homology to CCR6 (44%
homology).12 Therefore, in the functional analysis of CCR6, it is es-
sential to use a biological probe that selects CCR6 over CCR1 and CCR7.
Based on these results, we first performed a structure-activity re-
lationship (SAR) study of hit compound 1 to discover more potent in-
hibitors. Then, we modified the central piperidine ring, which led to the
identification of a novel series of 1,4-trans-1-benzenesulfonyl-4-ami-
nocyclohexane derivatives and their in vitro POC study. Our com-
pounds demonstrated good correlation between Gi signal inhibitory
activity and cell migration inhibitory activity in human CCR6-trans-
fected CHO cells. Furthermore, representative compound 35 potently
inhibited CCR6-dependent cell migration and the increase in ERK
phosphorylation in human primary cells. Herein, we report the SAR
studies of both the hit compound and the novel series of CCR6 in-
hibitors derived from it, and the biological activities of representative
We began by searching for CCR6 inhibitors and verifying in vitro
proof of concept (POC) that these inhibitors block CCR6-dependent
⁎
Corresponding author.
E-mail address: taisuke.tawaraishi@takeda.com (T. Tawaraishi).
https://doi.org/10.1016/j.bmcl.2018.07.042
Received 12 June 2018; Received in revised form 27 July 2018; Accepted 29 July 2018
0960-894X/©2018ElsevierLtd.Allrightsreserved.
Pleasecitethisarticleas:Tawaraishi,T.,Bioorganic&MedicinalChemistryLetters(2018),https://doi.org/10.1016/j.bmcl.2018.07.042

T. Tawaraishi et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Fig. 1. Hit compound 1, N-benzenesulfonylpiper-
idine derivative effective as a human CCR6 inhibitor,
and design of a novel 1,4-trans-1-benzenesulfonyl-4-
aminocyclohexane series.
Replacement of piperidine ring of compound 2 with pyrrolidine (3) or
homopiperidine (4) led to a > 5-fold drop in potency. Then, as for the
arylamino moiety at the 4-position of the piperidine ring, transposition
of a cyano group at the 5-position on the pyridine ring to the 6-position
(5) and 4-position (6) resulted in a loss of activity. Potency relative to
Table 1
Human CCR6 inhibitory activity of N-benzenesulfonyl cyclic amine derivatives.
that of compound
2 was maintained by introduction of a tri-
fluoromethyl group (7) into the 5-position and was decreased by in-
troduction of a methyl group (8), showing that electron-withdrawing
groups at the 5-position enhanced potency. It could be considered that
cyano and trifluoromethyl groups might affect the electron density of
the aromatic ring or directly interact with the protein surface via hy-
drogen bonding to boost potency. A comparison of phenyl derivative 9
with compound 2 showed that the nitrogen atom of the pyridine ring
was non-essential for activity but had critical effects to increase meta-
bolic stability and reduce cytotoxicity [human metabolic stability (μL/
min/mg): 19 (compound 2), 69 (compound 9); cytotoxicity (% of ATP
content at 30 μM): 72.7 (compound 2), 44.9 (compound 9)]. 4-Meth-
oxyphenyl (10) and unsubstituted phenyl derivative (11) were less
potent than compound 9. Next, effect of the substituents on the ben-
zenesulfonyl moiety was investigated. Introduction of a lipophilic group
such as a chloro, trifluoromethyl, tert-butyl, or phenyl group (12–16) at
the 4-position afforded equipotent or enhanced activity relative to
compound 1. Especially, bulky groups such as the tert-butyl (14) and
phenyl group (15 and 16) showed relatively high potency among them.
On the other hand, the cyano (17) and acetamide group (18) had a
detrimental effect on activity. Meta-substitution of a methyl (19) or
phenyl group (20) also led to a decrease in activity. These results
suggest that the tolerable polarity and position of substituents in the
moiety are limited. As a result of initial SAR study, we identified several
compounds (14, 15, and 16) that were 10-fold more potent than hit
compound 1.
a
Cmpd
Ar
n
1
R
IC50 (nM)
1
4-Me
3100 (2800–3400)
2
3
4
1
0
2
4-MeO
4-MeO
4-MeO
2500 (2100–3000)
> 30000b
17,000 (15000–19000)b
5
6
1
1
4-MeO
4-MeO
7400 (6500–8300)
> 30000
7
1
1
1
1
4-MeO
4-MeO
4-MeO
4-MeO
940 (730–1200)
21,000 (18000–25000)
1900 (1300–2800)
> 30000
8
9
10
11
12
1
1
4-MeO
4-Cl
12,000 (8700–17000)
1500 (1300–1800)
We next conducted scaffold morphing of the hit series' central pi-
peridine ring, which we replaced with a cyclohexane ring (Fig. 2). We
envisioned that cyclohexane derivatives afford two isomers, 1,4-cis and
trans, and if either isomer gets the better of the other in potency, fa-
vorable conformation is clarified. It was also expected that introduction
of a cyclohexane ring might rigidify the conformation and thereby
enhance potency. Consequently, the 1,4-trans isomer 21 exhibited as
potent activity as the parent compound 14, whereas the 1,4-cis isomer
22 was less potent. In order to verify the result, a docking study using
Maestro was performed (Fig. 3).13 The key pyridine ring with a cyano
group at the 5-position overlaps considerably between potent com-
pounds 14 and 21 (shown in a cyan circle). On the other hand, the 5-
cyanopyridyl moiety of compound 22 cannot be placed in the proper
direction due to its axial conformation (shown in a white circle).
Compound 21 showed not only equipotent activity but also higher
solubility compared with the piperidine series [solubility in water and
buffer solutions adjusted to pH 6.8 described in the Japanese
Pharmacopoeia (μg/mL): < 0.10 (compound 14), 3.8 (compound 21)],
and therefore we focused on the modification of this novel cyclohexane
series to identify in vitro biological probes (Table 2). 4-Biphenyl sul-
fones 23 and 24 were equipotent to compound 21. Especially, tri-
fluoromethyl derivative 24 showed activity with an IC50 value of <
100 nM. Moreover, to exploit the surrounding environment of the
biaryl moiety, the incorporation of a nitrogen atom into the terminal
aryl group was examined. As a result, derivatives with 4-pyridyl (25)
13
14
15
16
17
18
19
20
1
1
1
1
1
1
1
1
4-CF3
4-tert-Bu
4-Ph
800 (390–1700)
220 (130–360)
290 (220–370)
170 (110–250)
12,000 (5900–24000)
> 30000
4-Ph
4-CN
4-AcNH
3-Me
> 30000
3-Ph
> 30000
a
Values shown in parentheses are 95% confidence intervals.
b
Data of racemate.
compounds to show their potential as biological probes.
Our SAR exploration of the hit series focused on three moieties (the
central cyclic amine, arylamine, and benzenesulfonyl moiety) to en-
hance inhibitory activity (Table 1). At first, we showed that 4-meth-
oxyphenyl derivative 2 and compound 1 had similar inhibitory activity.
2

T. Tawaraishi et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Fig. 2. Modification of the central ring from piper-
idine to cyclohexane. Values shown in parentheses
are 95% confidence intervals.
Table 2
Human CCR6 inhibitory activity of 1,4-trans-cyclohexane derivatives.
a
Cmpd
23
R
Ar
IC50 (nM)
CN
CF3
CF3
CF3
CF3
190 (140–260)
76 (56–100)
24 (17–34)
24
25
26
48 (34–67)
Fig. 3. Superposition of 14 (green), 21 (magenta), and 22 (yellow).
27
460 (400–530)
and 3-pyridyl group (26) exhibited potent activity, whereas that with 2-
pyridyl group (27) had reduced activity, indicating the possible inter-
action of the moiety with the protein surface through hydrogen
bonding. Encouraged by the observation, introduction of a carboxamide
or a cyano group allowing the compound to come closer to the protein
surface was next pursued. As for the carboxamide derivatives, both
para- (28) and meta-substitution (29) demonstrated improved activity
relative to the unsubstituted parent compound 24. The result was
contrary to that for ortho-substitution (30), but also consistent with the
SAR for the pyridine derivatives mentioned above. On the other hand,
there was little difference in the potency of the cyano derivatives;
however, para-cyano derivative 31 showed relatively high activity. By
putting the SAR information together, picolinamide derivatives (34 and
35) and picolinonitrile derivative (36) were prepared. They maintained
excellent inhibitory activity in accordance with expectation; especially
compound 35 had the most potent activity (IC50 = 6.0 nM) among this
series of derivatives. Thus, we successfully discovered several tool
candidates for in vitro evaluation with IC50 values of around 10 nM as
represented by compounds 34 and 35.
Versatile synthetic approaches to prepare N-benzenesulfonyl cyclic
amine derivatives A (2–20) were developed as shown in Scheme 1. In
the first route (Route 1), 1-Boc-protected cyclic amines (37a–c) were
coupled with substituted 2-chloropyridines by either direct SN2 sub-
stitution or palladium-catalyzed Buchwald reaction to afford ar-
ylamines (38a–f). After removal of the Boc group from the compound
38 series under acidic conditions, resultant amines (39a–f) were con-
verted to target compounds A by sulfonylation with various substituted
benzenesulfonyl chlorides. As the second route (Route 2), 4-(N-Boc-
amino)piperidine (40) was firstly sulfonylated to provide compound
41, which was next deprotected and neutralized to yield free amine 42.
Buchwald reaction of amine 42 using BrettPhos/BrettPhos – Pallada-
cycle under microwave irradiation gave compounds A. Due to low re-
activity of the corresponding aryl halide, compound 10 was synthesized
via the alternative route (Route 3). That is, initial sulfonylation of 4-
piperidinone hydrochloride (43) and subsequent reductive amination
with arylamine and 2-picoline borane complex successfully provided
the desired product 10.
28
29
30
CF3
CF3
CF3
13 (10–17)
12 (10–13)
900 (750–1100)
31
32
33
CF3
CF3
CF3
17 (13–22)
31 (26–37)
48 (42–55)
34
CF3
CF3
CF3
17 (15–19)
6.0 (5.0–7.1)
27 (21–36)
35
36
a
Values shown in parentheses are 95% confidence intervals.
and trans-4-N-Boc-1-cyclohexanols (45a and 45b), subsequent SN2
substitution reaction with thiophenols successfully proceeded with
stereo-chemical inversion to afford corresponding trans- and cis-4-N-
Boc-1-phenylthiocyclohexanes (47a–c), respectively. Then, oxidation of
the sulfide group of compound 47 series using m-CPBA followed by
deprotection of Boc group yielded cyclohexylamines (49a–c). Finally,
coupling reaction of amines 49 with 2-halopyridines provided the de-
sired compounds B.
Among cyclohexane derivatives, biaryl sulfone compounds were
prepared as depicted in Scheme 3. 4-Bromophenyl sulfones (50a and
50b) were subjected to Suzuki-Miyaura coupling reaction with various
boronic acids or boronic acid pinacol esters to afford target compounds
C. As an alternative route, compound 50b was firstly converted to
phenylboronic acid pinacol ester 53 using bis(pinacolato)diboron and a
palladium catalyst. Then, subsequent coupling reaction with aryl
General synthetic route for the cyclohexane derivatives B (21, 22,
50a and 50b) is illustrated in Scheme 2. Starting from tosylation of cis-
3

T. Tawaraishi et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Scheme 1. Reagents and Conditions: (i) Ar-Cl (Ar = 2-pyridyl), DIEA, DMF, 60–120 °C; (ii) Ar-Cl (Ar = 2-pyridyl), Pd2(dba)3, BINAP, Cs2CO3, toluene or DME, 80 °C;
(iii) 4 M HCl (EtOAc or CPME solution), EtOAc or EtOH, rt or 50 °C; (iv) R-C6H4SO2Cl, Et3N or DIEA, THF, 0 °C or rt; (v) NaHCO3, H2O, EtOAc, THF, rt; (vi) Ar-Br
(Ar = 2-pyridyl, phenyl), BrettPhos, BrettPhos–Palladacycle, Cs2CO3 or sodium tert-butoxide, DMF, 130 °C (under microwave irradiation); (vii) Ar-NH2, 2-picoline
borane complex, AcOH, MeOH, rt.
Scheme 2. Reagents and Conditions: (i) p-TsCl, DMAP, pyridine, rt; (ii) 4-R1-C6H4SH, K2CO3, DMF, 100 °C or Cs2CO3, acetone, 60 °C; (iii) m-CPBA, EtOAc, rt; (iv) 4 M
HCl EtOAc solution, EtOAc, rt; (v) 2-chloro-5-cyanopyridine or 2-fluoro-5-(trifluoromethyl)pyridine, DIEA, NMP or DMF, 100–120 °C.
Scheme 3. Reagents and Conditions: (i)
Ar-B(OR)2 (R = H or CMe2), Pd(PPh3)4,
K2CO3, 1,4-dioxane, EtOH, H2O,
90–100 °C (in some cases, under mi-
crowave irradiation); (ii) bis(pinaco-
lato)diboron, Pd(dppf)Cl2·CH2Cl2 or Pd
(PPh3)4, AcOK, DME or 1,4-dioxane,
90 °C; (iii) Ar-Br, Pd(PPh3)4, K2CO3,
1,4-dioxane, EtOH, H2O, 80–90 °C, or
Na2CO3, DME, H2O, 100 °C, (iv) NH4Cl,
HATU, Et3N, DMF, rt.
bromides gave compounds C as well. Carboxamides 30 and 34 were
prepared by amidation of carboxylic acids 51 and 52 with NH4Cl and
HATU.
Representative compounds were evaluated in the migration assay
using human CCR6-transfected CHO cells, and also their activity against
monkey CCR6 and subclass selectivity were assessed (Table 3). In re-
gard to human CCR6, good correlation between Gi signal inhibitory
activity and migration inhibitory activity was observed. Among them,
compound 35 exhibited the most potent activity in the migration assay
as well as the Gi signal assay. All the compounds maintained high
human CCR6 selectivity over human CCR1 and CCR7. In terms of dis-
ease model animals, we chose monkeys because the amino acid se-
quence of human CCR6 has higher homology with monkey CCR6
(> 94%) than with rodent CCR6 [75% (mouse) and 76% (rat), re-
spectively].14 Thus, we established a Gi signal assay using monkey
CCR6-transfected CHO cells. As expected, the inhibitory activity against
4

T. Tawaraishi et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Table 3
Inhibitory activity against human CCR6-dependent migration and monkey CCR6, and subclass selectivity of representative compounds.
a
Cmpd
IC50 (nM)
Migration assay
human CCR6
Gi signal assay
human CCR6
monkey CCR6
human CCR1
> 30000
> 30000
> 30000
> 30000
> 30000
> 30000
human CCR7
> 30000
> 30000
> 30000
> 30000
1
1400
(630–3200)
240
(170–340)
41
(29–59)
36
(27–49)
42
(25–69)
< 30
3100
(2800–3400)
290
(220–370)
13
(10–17)
17
(13–22)
17
(15–19)
6.0
1300
(1000–1600)
280
(250–320)
16
(11–22)
21
(14–31)
9.5
(6.5–14)
0.45
15
28
31
34
35
18,000
(13000–26000)
9400
(5.0–7.1)
(0.30–0.69)
(6300–14000)
a
Values shown in parentheses are 95% confidence intervals.
monkey CCR6 was well correlated with that against human CCR6, and
compound 35 showed the greatest activity against monkey CCR6 as
well.
In order to validate its potency as a biological probe, representative
compound 35 was subjected to in vitro evaluation in human primary
cell assays. Firstly, compound 35 significantly inhibited CCL20-induced
human B cell migration at 0.1 and 1 μM (Fig. 4). Efficacy of the com-
pound was also confirmed in a human whole blood assay, in which
CCL20-induced increase in ERK phosphorylation was detected in B cells
(Fig. 5). Compound 35 potently inhibited ERK phosphorylation in a
concentration-dependent manner. Thus, we confirmed that our com-
pounds could work effectively in vitro as biological probes against
human CCR6.
Next, we conducted pharmacokinetic study of compound 35 in cy-
nomolgus monkeys (0.1 mg/kg, iv and 1 mg/kg, po, male, N = 3) for
further application in vivo. As a result, compound 35 displayed area
under the curve values [AUC0-24h (ng * h/mL): 1552 (iv), 10,050 (po)],
volume of distribution (Vdss: 704 mL/kg), low clearance (CLt: 66 mL/h/
kg), plasma unbound fraction (fup: 0.01), and good bioavailability (BA:
65%). We also assessed off target profile of compound 35. Compound
35 did not exhibit remarkable inhibitory activities against CYPs (1A2,
2C8, 2D6, 3A4: < 50% at 10 μM; 2C9: 51% at 10 μM) and hERG
(< 50% at 10 μM with/without BSA). Those results suggested that
compound 35 would be usable as an in vivo probe as well.
Fig. 4. Effect of compound 35 in CCL20-induced human B cell migration assay.
The vertical axis is indicated by ATP content measured in relative luminescence
units (RLUs). Statistical analysis was performed using the Williams test. *:
P < 0.025 versus control. Error bars indicate SEM (N = 3).
In conclusion, we have identified a novel series of potent and se-
lective human CCR6 inhibitors, 1,4-trans-1-benzenesulfonyl-4-amino-
cyclohexanes. Among them, compound 35 potently inhibited both
CCL20-induced CCR6-dependent migration and ERK phosphorylation
in human primary cell assays. They can be utilized as in vivo biological
probes, when, for instance, monkeys are utilized as disease model an-
imals. We think that CCR6 inhibitor has an opportunity as a novel
therapeutic option for diseases including various autoimmune diseases
and cancer where CCR6-dependent cell migration contributes to the
pathophysiology, and our compounds help further understanding of the
pathophysiology of these diseases.
Acknowledgments
We wish to express our sincere thanks to Dr. Nobuyuki Negoro and
Dr. Tomoya Yukawa for their helpful discussion and support during the
preparation of the manuscript. We also thank Ms. Mika Inoue, Ms.
Tomoko Izukawa, Mr. Shinobu Sasaki, Mr. Shin-ichi Niwa, Dr.
Yoshihisa Nakada, Mr. Masaki Setoh, Dr. Yoshiyuki Fukase and Dr.
Shoji Fukumoto for their helpful support.
Fig. 5. Effect of compound 35 in a human whole blood assay. CCL20-induced
increase in ERK phosphorylation was evaluated in B cells. A and B are different
donors (N = 2).
5

T. Tawaraishi et al.
Bioorganic&MedicinalChemistryLettersxxx(xxxx)xxx–xxx
Appendix A. Supplementary data
1999;66:674–682.
4. Yamazaki T, Yang XO, Chung Y, et al. J Immunol. 2008;181:8391–8401.
5. Hirota K, Yoshitomi H, Hashimoto M, et al. J Exp Med. 2007;204:2803–2812.
6. Kochi Y, Okada Y, Suzuki A, et al. Nat Genet. 2010;42:515–519.
7. Hedrick MN, Lonsdorf AS, Shirakawa A, et al. J Clin Invest. 2009;119:2317–2329.
8. Ghadjar P, Rubie C, Aebersold DM, Keiholz U, Int J. Cancer. 2009;125:741–745.
9. Kapur N, Mir H, Clark CEIII, et al. Br J Cancer. 2016;114:1343–1351.
10. Wan W, Lim JK, Lionakis MS, et al. Circ Res. 2011;109:374–381.
11. Jung ID, Lee JS, Kim YJ, et al. Immunology. 2007;121:533–544.
12. https://blast.ncbi.nlm.nih.gov/Blast.cgi.
Supplementary data associated with this article can be found, in the
online version, at https://doi.org/10.1016/j.bmcl.2018.07.042.
References
1. Schutyser E, Struyf S, Van Damme J. Cytokine Growth Factor Rev. 2003;14:409–426.
2. Ito T, Carson WFIV, Cavassani KA, Connett JM, Kunkel SL. Exp Cell Res.
2011;317:613–619.
13. Maestro version 10.4 (Schrödinger, Inc.).
14. https://www.ncbi.nlm.nih.gov/homologene?cmd=Retrieve&dopt=
AlignmentScores&list_uids=3214.
3. Sullivan SK, McGrath DA, Liao F, Boehme SA, Farber JM, Bacon KB. J Leukoc Biol.
6