!"#$%& (&)*+&,-.+/& 0+/&#1&%1$23$.+ 4(506 !7,&) 8$.9 (:;< =&%1$2>&#?

!"#$%&'$(#)"'%* ,-.("(/#$%0)" )1 !2345, !"6(7(#)$/ 89&&$'//'/ :(/'%/'

@,&%2AB*27% *%, C9*#*B.&#$>*27% 7D =+%.9&2B =E*)) !7)&B")& !*B#7B+B)& F%.*G7%$1.1 7D 0"E*% @H-IJF
(*3$, H$3$%G1.7%K =&.9" F)&L*%,&#K M")$*% <7%,K 5$E7.9+ <#$GG1K F%,#&8 ;#*)&+K =.&/9&% 0*)&K 5*%+* H*%,1E*%K N$B9*#, !*#2%&))$K O&))&+ =97#.1)&&3&1K :$BP 5&##&Q *%, :*.9*% R*)19
WWW.ENSEMBLETX.COM
99 Erie Street Cambridge, MA 02139 USA (617) 492-6977
@H-IJ?@H-IJN&B&/.7# C7E/)&L&1 @%37)3&
="S1.*%2*) T#7.&$%-T#7.&$% ="#D*B&1
=+%.9&2B*))+ FBB&11$S)& !*B#7B+B)$B C9&E$B*) !*Q&#?
;"(<9' :'/(=" >*'.'"#/ 1)$ !"6(7(0"= ?$)#'("3?$)#'(" @).&*'A'/
@H-IJF !*B#7B+B)& H&*,1 @,&%2A&,? B))- >"$(C6.'"#/ %"-
@)"D$.'- E("-("= #) % F9"C0)"%* 8(#' )1 G9.%" !2345,
C7%B)"1$7%1 *%, U".)77P
@H-IJF @%9$S$.7#1 ="//#&11 @H-IJF- $%,"B&, C+.7P$%& *%,
C9&E7P$%& T#7,"B27% $% 0"E*% NF=; *%, 05VW C&))1
C7E/7"%,1 XL9$S$. (71&-N&1/7%1& @%9$S$27% $%
* !"#$%& (50 @%Y*EE*27% !7,&)
C+.7P$%& C&)) 5+/& X%,/7$%. XCZ[ 4!6
Puman lL-17A Puman P129 Cro! 0.043
Puman lL-17A Puman 8ASl lL-6 1.8
Murlne lL-17A Mouse 313 lL-6 6
Puman lL-1# Puman 8ASls lL-6 >>23
Puman lL-2 Mouse P12 Cell prollferauon >>23
Puman lL-6 Puman PeLa S1A13 phosphorylauon >>23
Puman lL-13 Mouse P12 Cell prollferauon >>23
Murlne lL-13 Mouse P12 Cell prollferauon >23
Puman lL-22 Puman P129 CxCL-1 (Cro!) >> 30
Puman 1nl! Puman P129 or 8ASls lL-6 >>23
X-\]W\Z @H-IJF @%9$S$.7# =/&B$AB*))+ <)7BP1
@H-IJF-,&/&%,&%. @%,"B27% 7D C+.7P$%&1 *%, C9&E7P$%&1
Lnsemble has ldenued and opumlzed lnhlblLors of human and murlne lL-17A uslng lLs proprleLary
lnLegraLed macrocycle drug dlscovery plauorm.
1hese lnhlblLors blnd lL-17A wlLh nM amnlLy, compeLe wlLh lL-17A blndlng Lo lLs cellular recepLor,
lnhlblL speclcally lL-17A lnducuon of cyLoklnes and chemoklnes ln cell assays , and are selecuve for
lL-17 lnduced cellular responses vs. responses lnduced by oLher pro-lnammaLory cyLoklnes.
1he lL-17A lnhlblLors are acuve ln vlvo ln murlne models of acuLe lnammauon when dosed
lnLraperlLoneally or by oral gavage ln self-emulslfylng soluuon vehlcles.
Cne lnhlblLor descrlbed (L-36041) ls orally acuve ln a murlne ClA model. 1hls compound dlsplays
slmllar acuvlLy as an anu-lL-17A anubody. 1he compound suppresses ln paw lnammauon, pannus
formauon, and bone resorpuon.
LorLs are underway Lo furLher lmprove Lhe compounds Lo oral poLency and bloavallablllLy and Lo
examlne Lhe compounds ln oLher chronlc models of human auLolmmune/lnammaLory dlsease.
!"#$% '()"#$*+,- ./*(/,01 23%+4,$5 6,7/+75$1*1 8'.269 ,% :; 2)/"#<
=>?@ 5)4)51 #)$1"/)- 73 A>=2. BC (/1D E+1*?,%-"F0+% G,*( =>:H.D
AIJC1 -)*)/#,%)- G)/) :DKL BD@L CDMH $%- :DN O P+/ A?MNQMJL
A?MJC:KL A?MJH@B $%- ACM@CN:L /)1E)F04)53D
FS1.#*B.
<*BPG#7"%,^T"#/71&: lL-17A has been demonsLraLed Lo be a key pro-lnammaLory cyLoklne ln
human rheumaLold arLhrlus and ln several rodenL models of arLhrlus. SynLheuc macrocycles are
more amenable Lo opumlzauon for meLabollc sLablllLy and oral absorpuon Lhan bloLherapeuucs.
1he almof Lhls lnvesugauon was Lo ldenufy hlgh-amnlLy macrocycle blnders of human lL-17A, Lo
quanufy Lhelr lnhlblLory poLency agalnsL cyLoklne producuon ln human cells, and Lo deLermlne lf
acuve compounds could lnhlblL a delayed-Lype hypersensluvlLy response ln mlce.
!&.97,1: unA programmed chemlsLry (uC) llbrarles were generaLed Lo synLheslze ln vlLro
llbrarles of non-pepudlc synLheuc macrocycles of molecular welghL 600- 1000 kua. Compounds
blndlng Lo lmmoblllzed lL-17A were ldenued by C8 and unA sequenclng. 1wo compounds
were resynLheslzed and characLerlzed by 1) compeuuve LLlSA Lo deLermlne amnlLy for human
lL-17A, 2) lnhlbluon of lL-17A-drlven lL-6 producuon ln human rheumaLold arLhrlus synovlal
broblasLs (8ASl) and human P1-29 adenocarclnoma cells, 3) lnhlbluon of oLher pro-
lnammaLory human cyLoklne acuvlues, such as lL-1, lL-6, lL-22, and 1nlo, and 4) emcacy ln a
delayed-Lype hypersensluvlLy (u1P) mouse model. 1he u1P model used a 1-uoro-2,4-
dlnlLrobenzene (unl8) sensluzer, whlch was applled Lo Lhe anlmals aL day 0. Cn day 7,
compounds dlssolved ln uMSC were dosed by lnLraperlLoneal (l.p.) ln[ecuon aL a dose of 10 mg/
kg. A second appllcauon of unl8 was performed on Lhe le ear 30 mln aer compound doslng.
Aer 24 hours, le ear edema was measured by change ln ear welghL compared Lo Lhe rlghL ear,
and levels of lnl-y ln ear ussue homogenaLes were quanued by LLlSA.
N&1").1: 1wo synLheuc macrocycles ldenued ln Lhls lnvesugauon, L-34933 and L-33018, were
characLerlzed by a compeuuon LLlSA wlLh human lL-17A, and deLermlned Lo have a dlssoclauon
consLanL (kd) = 2 nM. L-34933 and L-33018 were found Lo lnhlblL lL-17A wlLh LC30 of 2.0 and 2.1
M ln 8ASl, and 43 and 20 nM ln P129 cells, respecuvely. 8oLh compounds were lnacuve (LC30 >
23 M) ln a bauery of cellular assays for Lhe human cyLoklnes lL-1, lL-6, lL22, and 1nlo. A
slngle l.p. dose of 10 mg/kg of L-34933 or L-33018 ln Lhe murlne u1P model suppressed edema
vs. vehlcle conLrol by 30 or 34 respecuvely (p < 0.03 vs. vehlcle conLrol). ln comparlson, a raL
anu-mouse lL-17A lgC1 (3 mg/kg, l.p.) resulLed ln 76 lnhlbluon of edema. lnl-y levels ln ussue
homogenaLes were also suppressed by L-34933, L-33018, or anu-lL-17A Ab vs. vehlcle conLrol by
72, 62 or 73, respecuvely (p < 0.03 for all groups vs. vehlcle conLrol group).
C7%B)"1$7%: Cur daLa provlde evldence LhaL synLheuc macrocycles can be ldenued LhaL blnd
poLenLly and speclcally Lo human lL-17A, and acL as lnhlblLors of lL-17A-sumulaLed lL-6
producuon ln 8ASl and P129 cells. 1hese compounds are also anu-lnammaLory ln an lL-17-
dlrecLed murlne u1P model. rlor Lo Lhls lnvesugauon, such speclc lnhlblLors of Lhe lL-17A-
lL17recepLor lnLeracuon were llmlLed Lo polypepudes.
T#7,"B23& !&,$B$%*) C9&E$1.#+ C*E/*$G% ?$)-9C'- G(=63%H"(#I
!245, 7("-'$/
C7%B)"1$7%1
T#7,"B23& !&,$B$%*) C9&E$1.#+ C*E/*$G%
T#7,"B&, G(=63%H"(#I G9.%" !2345, E("-'$/
U#*))+ F,E$%$1.&#&, @H-IJF @%9$S$.7# X-\_[]I
@1 F%2-$%Y*EE*.7#+ *%, ($1&*1&-E7,$D+$%G $% !"#$%& C@F !7,&)
@H-IJF @%9$S$.7#1 F,E$%$1.&#&, S+ U#*) `*3*G& ="//#&11
X,&E* *%, C+.7P$%& T#7,"B27% $% !"#$%& (50 !7,&)
<$7B9&E$B*) C9*#*B.&#$>*27% 7D @H-IJF <$%,&#1
(&E7%1.#*.&1 @).&'00)" J(#6 !2345K, %"- 8*)J LM3K%#'/
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
1E-06 1E-05 0.0001 0.001 0.01 0.1 1 10 100 1000
IL17A
-IL17R
A Inhibition
Compound Concentration (nM)
E-035018
E-034935
E-035762
E-036041
C7E/&227% XH@=F
@CZ[ a%!b
IcV
Icd
Jc_
VJc
A compeuuve LLlSA blndlng assay ln whlch selecLed anu-lL-17A
macrocycles dlsrupL Lhe lnLeracuon of human lL-17A Lo
lL-178A.
1he blndlng of anu-lL-17A macrocycles Lo human lL-17A was lnvesugaLed
uslng 8lacore S8 Lechnlques. 1he lnhlblLors dlsplayed rapld Cn-raLes
of blndlng and slow Cll-raLes of dlssoclauon fromLhe cyLoklne LargeL.
L
E-34935
E-35018
E-35762
E-36041
=%(,7,0+% +P =>:H. =%-"F)- R/+! 1)F/)0+% ,% !SBQ $-)%+F$/F,%+#$ F)551 ,%
1)/"#?P/)) #)-,"#< =>?@ 5)4)51 #)$1"/)- $* NK (/1D E+1*?,%-"F0+% G,*( =>?:H.D
AIJC1 -)*)/#,%)- G)/) CDCNJL CDC::L CDDMM $%- :DC O P+/ A?MNQMJL A?MJC:KL
A?MJH@B $%- A?M@CN:L /)1E)F04)53D
Male Balb/c mice were administered a 0.5% 1-fluoro-2,4-dinitrobenzene (DNFB) solution to the shaved abdomen on Day 0 and
Day 1. On Day 7, the mice were dosed i.p. with inhibitor, and 30 minutes later a 0.2% DNFB solution was applied to the shaved
left ear, and an inactive solution applied to the right ear. 24 hours later, the mice were euthanized and ear edema (left right ear
weight) was determined. Cytokine/chemokine levels were determined by ELISA in ear tissue homogenates.
CRO = Washington Biotechnology Inc., Baltimore, MD USA.
DTH protocol as described in prior panel, except that in these experiments the inhibitors were dosed by oral gavage.
Inhibitors were formulated in self-emulsifying solutions of Cremophor/Water (20:80) or TPGS/PEG-400/Water (20:60:20).
Anti-IL17-A antibody (affinity purified rat anti-mouse IgG1 $ isotype, BioLegend) was dosed i.p. in PBS.
Methods
IL-17A inhibitors were evaluated in a
murine CIA model. In this study (CRO =
Bolder BioPATH, Inc., Bolder, CO USA).
DBA-1 mice (10/group) were injected
intradermally with 150 L of Freunds
Complete Adjuvant containing bovine type
II collagen (2 mg/ml) at the base of the tail
on day 0 and again on day 21. On study
days 2425, onset of arthritis occurred and
mice were randomized into treatment
groups. Randomization into each group
was done after swelling was obviously
established in at least one paw (score of
1), and attempts were made to ensure
approximately equal mean arthritis scores
of 0.25 across the groups at the time of
enrollment. Treatment was initiated after
enrollment and continued as indicated
(see below) through arthritis day 10. Mice
were terminated on day 11. Clinical scores
were given for each of the paws (right
front, left front, right rear, left rear) on
arthritis days 111.
Oral delivery of anti-IL-17A inhibitors inhibits chronic inflammation in the murine CIA model. A) Summed clinical arthritis scores All
Paws (scored 0-5), B) Clinical arthritis score with AUC calculation All Paws, C) Individual histopathology parameters (All Joints), D)
Histopathology sum (All Joints), E) Photomicrographs of forepaws (left to right) nave, vehicle treated, 5 mg/kg IP anti-IL-17A antibody,
30 mg/kg E-36041, p.o.