Aromatic quinolinecarboxamides as selective, orally active antibody production inhibitors for prevention of acute xenograft rejection

Article abstract of DOI:10.1021/jm010822m

The prevention of xenograft rejection is substantially dependent on inhibiting antibodies (Ab) produced by B-cells independently of T-cell signals (TI-1). Due to their ubiquitous biochemical mechanisms of action, the immunosuppressants currently employed

Full text of DOI:10.1021/jm010822m

1986
J . Med. Chem. 2001, 44, 1986-1992
Ar om a tic Qu in olin eca r boxa m id es a s Selective, Or a lly Active An tibod y
P r od u ction In h ibitor s for P r even tion of Acu te Xen ogr a ft Rejection
Christos Papageorgiou,* Anette von Matt, J oanne J oergensen, Elsebeth Andersen, Katrin Wagner,
Christian Beerli, Thai Than, Xaver Borer, Andrea Florineth, Gretty Rihs, Max H. Schreier,†
Gisbert Weckbecker, and Christoph Heusser
Transplantation Research, NOVARTIS Pharma AG, WSJ -350.314, CH-4002 Basel, Switzerland
Received J anuary 19, 2001
The prevention of xenograft rejection is substantially dependent on inhibiting antibodies (Ab)
produced by B-cells independently of T-cell signals (TI-1). Due to their ubiquitous biochemical
mechanisms of action, the immunosuppressants currently employed not only fail to discriminate
between B- and T-cells but also have a narrow therapeutic window and, thus, their prolonged
use in complex immunosuppressive regimens is problematic. By capitalizing on the target
enzyme-bound (DHODH) structure 1b of one of these compounds, leflunomide, and modulating
part of its multiple mechanisms of action to gain selectivity, the quinoline-8-carboxamide 3
was designed as a potentially weak enzyme inhibitor but effective immunosuppressant.
Compound 3 fulfilled the mechanistic criteria set and had 10-fold B-cell over T-cell selectivity.
Its pyridyl analogue 4 was found to be a highly potent and selective B-cell immunosuppressant
with a 75-fold selectivity for B- over T-cells (as judged by the MLR data) and no general
cytotoxicity at concentrations up to 160-fold higher than those required to inhibit B-cells. In
the mouse, 4 effectively blocked TI-1 Ab production and suppressed Ab-mediated xenograft
rejection in a xenotransplantation model under a once-daily dosing regimen, with efficacy down
to 0.3 mg/kg/day po. These are the first data demonstrating the feasibility of the development
of drugs specific for impeding Ab production.
In tr od u ction
cascade inhibitor found in primates.4 Attention is cur-
rently turned toward an understanding of the im-
munological barrier mediating the next phase of xe-
nograft rejection termed acute vascular rejection (AVR).
This delayed rejection process is an Ab-mediated event
that typical T-cell immunosuppressants such as cy-
closporin A (CsA) cannot prevent, and it is likely that
the control of such Ab synthesis by B-cells will play a
pivotal role in overcoming it. With HAR and AVR
suppressed, xenografts are expected to behave similarly
to allografts and, consequently, be under the control of
the currently existing T-cell immunosuppressants.5
Regimens aimed at the control of AVR have been
extensively studied in rodent models of xenotransplan-
tation as well as in the pig-to-primate model. They focus
primarily on the inhibition of newly formed anti-GalR1,-
3Gal Ab by host B-cells in response to the xenograft6
and are based on the use of nonspecific agents such as
the de novo nucleotide inhibitors leflunomide, brequi-
nar, mycophenolic acid derivatives, and antiproliferative
macrolides of the rapamycin class.7 Despite their docu-
mented efficacy both in rodent models and in nonhuman
primate recipients of pig organs transgenic for hDAF,
the narrow therapeutic index of these compounds limits
their use in regimens for AVR prevention.8 In search of
selective inhibitors of CsA-resistant Ab formation, we
recently reported the discovery of pyrazole derivatives
that had a novel biochemical mechanism of action as
well as the desired in vitro profile but suffered from
nonoptimal pharmacokinetic properties.9 Pursuing our
efforts in the field of the inhibition of CsA-resistant
B-cell responses, we now report the design and biological
evaluation of a novel class of B-cell inhibitors that
The very high success rate of allotransplantation and
the resulting quality of life have led to the establishment
of this technique as a standard therapeutic option for
the treatment of end-stage organ failure in man. Due
to the current shortage of donor organs, xenotransplan-
tation, the transplantation of organs between two dif-
ferent species, has recently become a field of active
research because it represents the means to solve the
organ availability issue. Common to all innovative
approaches, xenotransplantation faces many challenges,
not least the immune response of the recipient against
the transplant.1 The pig-to-primate model is being
extensively utilized as the most relevant preclinical
model for the evaluation of immunosuppressive proto-
cols that will enable the long-term acceptance of trans-
genic pig organs by man.2
The transplantation of a vascularized porcine organ
into a nonhuman primate leads to an immediate and
dramatic rejection process, known as hyperacute rejec-
tion (HAR). It is the result of the species disparities and
is mediated by the binding of the naturally occurring,
pre-existing antibody (Ab) of the host to the GalR1,3Gal
epitopes found on the glycoproteins and glycolipids
expressed on pig endothelium.3 As a consequence, the
host complement is activated and rapid destruction of
the xenograft occurs. HAR can be reliably controlled by
employing pig organs transgenic for human decay
accelerating factor (hDAF), an endogenous complement
* Corresponding author (telephone 0041-61-3246188; fax 0041-61-
3243036; e-mail christos.papageorgiou@pharma.novartis.com).
† Deceased J anuary 14, 2000.
10.1021/jm010822m CCC: $20.00 © 2001 American Chemical Society
Published on Web 05/12/2001

Quinolinecarboxamides as Xenograft Rejection Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 12 1987
mide resulting from 1 during our previous investiga-
tions, we speculated that 1a could reflect the bioactive
conformation and, consequently, designed and obtained
pyrazole 2 as a novel DHODH inhibitor having a
biological activity similar to that of 1.9 In view of the
disclosed X-ray structure 1b of bioactive 1,15 the success
of the design of 2 is puzzling. Retrospectively, it can be
attributed either to binding to an allosteric position or
to the remaining presence of a 1a conformer population
that is also inhibitory but failed to crystallize in complex
with DHODH and was therefore not detected. We
subsequently initiated a novel approach based on the
structure of 1b and designed quinoline 3 as a potential
mimetic. Its probable conformation was deduced from
substructure searches in the Cambridge Crystallo-
graphic Database, which indicated that an intramolecu-
lar hydrogen bond between the amide NH and the N(1)
of quinoline should be formed, thus ensuring the
planarity of the molecule. This speculative solid state
conformation did not provide any information on the
conformation of 3 in solution but only suggested that a
similar folded structure could be energetically acces-
sible. The CN and Me groups of 1b, not being involved
in interactions with the enzyme, were not specifically
included in the designed molecule, but their electronic
and steric properties were imitated by the phenyl ring
of the quinoline. The only important functionality
missing in 3 that exists in 1b is a hydrogen bond donor
allowing the O-H f Thr356 interaction. Given that no
potent DHODH inhibition was sought, but only effects
mediated by the additional mechanism of 1b, and
considering the straightforward synthetic access to 3
and derivatives, the study of aromatic quinoline-8-
carboxamides was initiated.
F igu r e 1. Structures of antibody inhibitors.
efficiently block Ab responses in vivo and prevent
xenograft rejection upon daily oral treatment.
Design Con cep t
Two of the three current inhibitors of de novo nucleo-
tide synthesis mediate their effects by a highly specific
biochemical mechanism. Indeed, mycophenolic acid
(MPA) and brequinar (BQR) block the enzymatic activi-
ties of inosine monophosphate dehydrogenase (IMPDH)
and dihydrooroate dehydrogenase (DHODH), respec-
tively.10,11 The third inhibitor, leflunomide (1), is a
prodrug having an isoxazole ring that is quickly and
quantitatively opened in vivo as well as in vitro cellular
systems. The resulting hydroxypropenamide inhibits
DHODH at low concentrations but can operate by
additional protein tyrosine kinase-related mechanisms
at its therapeutically relevant concentrations.11,12 Of the
two hydroxypropenamide conformers 1a and 1b de-
tected so far, the latter was demonstrated to inhibit
DHODH (see below). However, because the contribution
of each conformer to the various biological mechanisms
of leflunomide is currently unknown, 1b was chosen as
a starting point for medicinal chemistry on the basis of
the hypothesis that it could lead to lymphocyte selective
substances via the modulation of only part of the
operating mechanisms. The inhibition of de novo nucle-
otide biosynthesis not being a prerequisite for efficient
B-cell inhibition,13 it was expected to identify novel and
potentially selective B-cell agents based on modifications
of 1b that would afford compounds devoid of the
ubiquitous DHODH blockade but retaining the inter-
actions of 1b with its additional targets.
To translate these mechanistic considerations into
synthetic targets, the corresponding free and enzyme-
bound conformations 1a and 1b were taken into con-
sideration, respectively (Figure 1). Both structures are
planar and have a strong intramolecular hydrogen bond
involving the enol and the amide functions. However,
in 1a the hydroxyl is the donor and the carbonyl the
acceptor, whereas the corresponding interacting pair in
1b are the amide NH and the hydroxyl oxygen.14,15 The
1a to 1b conformational change is presumably the result
of a keto-enol equilibrium leading to a torsional angle
modification. With no information available on the
DHODH-bound conformation of the hydroxypropena-
Ch em istr y a n d Str u ctu r e
The amides 3-6 were prepared, in moderate yields,
via the coupling of the acid chloride of the commercially
available quinoline-8-carboxylic acid with the corre-
sponding anilines (Scheme 1). For the synthesis of 7,
the required 7-methyl-8-carboxylquinoline 11 was ob-
tained on the basis of the Skraup reaction. The tetrahy-
droquinoline derivative 10 was quantitatively obtained
upon catalytic hydrogenation of 3. A different approach
was followed for the synthesis of the 2-trifluoromethyl
compound 9 (Scheme 2). The key step was the metala-
tion (n-BuLi/TMEDA) of the 8-bromoquinoline 15 fol-
lowed by quenching with (p-CF3)phenyl isocyanate. In
addition to 9 (30%), the reaction afforded a substantial
amount (45%) of the debrominated quinoline. Compound
1516 was obtained in a very low overall yield (2.5%) from
the highly moisture sensitive N-ethylidene-tert-butyl-
amine 12.17
To verify structurally our working hypothesis and to
gain insight for structure-activity result interpretation
(see In Vitro Biological Results), the solid state confor-
mation of two quinolines was determined crystallo-
graphically (Figure 2).18 Molecules 4 and 9 adopt planar
conformations that are stabilized by intramolecular
hydrogen bonds between the amide NH and the
quinoline nitrogen atom N(1) (Figure 3). Indeed, in
both compounds the N‚‚‚N distance is 2.7 Å and the
N-H‚‚‚N angle 143°. In agreement with 3D data on
pyridyl amides, the nitrogen atom of the pyridine in 4

1988 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 12
Papageorgiou et al.
Sch em e 1a
F igu r e 2. Structures of 8-quinolinecarboxamides.
F igu r e 3. Solid state conformation of quinolines 4 and 9
(Ortep plot with 30% displacement ellipsoids).
Ta ble 1. In Vitro Profile of Reference and Novel Compounds
compound
CsA
leflunomide (1)
3
4
5
6
7
TNP-LPSa,b
MLRa,c,e
J urkata,d,e
>10 ( 0
0.006 ( 0.004
11.4 ( 1.4
6.7 ( 1.5
>22.4 ( 5.9
23.4 ( 3.5
6.6 ( 0.9
20.2 ( 1.9
NT
>10 ( 0
34.6 ( 1.7
>44.7 ( 3.7
>50 ( 0
>50 ( 0
5.2 ( 1.2
>50 ( 0
NT
(a) (COCl)2, DMF (cat.), CH2Cl2, ArNH2; (b) acrolein, dichloro-
benzene, 140 °C, 3 h; (c) H2, Pd/C, DMF/MeOH (1.5/98.5).
5.1 ( 0.6
0.7 ( 0.1
0.3 ( 0.1
12.7 ( 1.9
4.4 ( 1.0
1.6 ( 0.4
>50 ( 0
Sch em e 2a
8
9
10
42.5 ( 3.8
19.5 ( 4.3
NT
NT
NT
NT
a
IC50 ( SE values in µM. Mean of at least four independent
b
determinations. Inhibition of murine Ab responses to the T-cell
independent B-cell antigen TNP-LPS (TI-1). c Inhibition of the
d
mouse mixed lymphocyte reaction. Inhibition of J urkat cell
proliferation. e NT, not tested.
the trinitrophenyl-lipopolysaccharide (TNP-LPS).19,20
Our goal being the identification of novel agents that
inhibit such TI-1 biological phenomena that are of
relevance in xenotransplantation, a cellular assay based
on the modulation of antibody production by TNP-LPS-
stimulated murine B-cells was employed for the com-
parative screening of the compounds. In addition, the
T-cell immunosuppressive activity of the compounds as
well as their cytotoxicity was assessed in the mouse
mixed lymphocyte reaction (MLR) and J urkat prolifera-
tion assays, respectively, as an estimate of their lym-
phocyte selectivity and therapeutic window.9 The bio-
chemical mechanism of action of the compounds was
compared to that of 1b using the human DHODH
enzymatic assay.21
(a) CH3CHO, 0 °C, pentane; (b) LDA, CF3COOEt, -75 °C, 30
min; (c) Br-aniline, AcOH, CF3COOH, 75 °C, 6 h; (d) POCl3,
heptane, 100 °C, 6 h; (e) n-BuLi, TMEDA, Et2O, -78 °C then (p-
CF3)PhNCO.
is on the opposite side of the CdO group. This preferred
conformation can be attributed to the weak intramo-
lecular H-bond between the carbonyl and the ortho
hydrogen of the pyridinyl moiety [D (C‚‚‚O) ) 2.86 Å, θ
(C-H‚‚‚O) ) 119°]. Collectively, the data indicate that
also quinolines 3-7 should be expected to have, if not
identical, very closely related overall conformations.
In Vitr o Biologica l Resu lts
The data are summarized in Table 1. As already
alluded to, CsA failed to inhibit Ab formation from TNP-
LPS-stimulated B-cells but potently and selectively
In the mouse, CsA-resistant responses are typically
elicited by type I T-cell independent antigens (TI-1) like

Quinolinecarboxamides as Xenograft Rejection Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 12 1989
Ta ble 2. In Vivo Inhibition of Immunoglobulin (Ig) Productiona
inhibited T-lymphocyte proliferation with an IC50 ) 6
nM in the MLR. Leflunomide (1) inhibited both TNP-
LPS responses and the MLR with IC50 values differing
by 2-fold. Although less potently, it also inhibited J urkat
proliferation, and the resulting window relative to the
TNP-LPS inhibition was ∼7-fold. The overall profile of
1 is compatible with that of a mild antiproliferative
agent and is the result of its ubiquitous mechanism of
action. The prototype quinoline 3 was a potent B-cell
inhibitor with 10- and >60-fold selectivities toward
T-cells and J urkat cells, respectively. The corresponding
pyridyl derivative 4 inhibited the TNP-LPS responses
with an IC50 ) 300 nM, had a 73-fold selectivity toward
T-cells, and was devoid of any effects on the J urkat cell
line. In line with their cellular profile and in contrast
to 1 that inhibits DHODH with an IC50 ) 200 nM,
neither 3 nor 4 had any effects on the enzymatic activity
at up to 50 µM concentrations. Moreover, they did not
inhibit inosine monophosphate dehydrogenase, the tar-
get of the de novo nucleotide biosynthesis inhibitor
MPA.10 Because of the reported inhibition of EGFR- and
PDGFR-Tyr kinases by leflunomide at high micromolar
concentrations, the effects of 3 and 4 on these kinases
as well as on cdk2 and the Src kinase family were
assessed and found to be negligible up to a concentration
of 30 µM.12 Therefore, compounds 3 and 4 represent
mechanistically novel, potent, and selective inhibitors
of CsA-resistant Ab production by B-cells. The currently
very limited information on the complex mechanisms
of B-cell activation, proliferation, and Ab production
does not permit a plausible explanation of the B- versus
T-lymphocyte selectivity of these quinolines.22
compound
doseb/routec
IgMd
IgGd
leflunomide (1)
50/sc
30/sc
3/sc
71 ( 10
37 ( 15
e0
85 ( 5.0
39 ( 21
e0
10/po
5 ( 21
e0
4
50/sc
30/sc
3/sc
82 ( 5
86 ( 4
73 ( 10
63 ( 12
95 ( 5
95 ( 6
84 ( 4
89 ( 9
10/po
a
b
Groups consisted of at least five mice per application. mg/
kg/day. c Vehicles for administration: (i) sc polyethoxylated castor
oil 650 mg/mL and EtOH up to 1 mL; (ii) po Labrafil 330 mg/mL,
d
EtOH 110 mg/mL, corn oil up to 1 mL. Inhibition percent.
hypothesis requiring the intamolecular NH‚‚‚N(1) H-
bond, the nonplanar tetrahydroquinoline derivative 10
was only marginally inhibiting TNP-LPS responses.
In Vivo Eva lu a tion a n d P h a r m a cok in etics
The potency and selectivity of 4 in vitro prompted its
comparative evaluation with leflunomide (1) in two
relevant in vivo models of Ab-mediated immunity.
The effects of 1 and 4 on the in vivo TI-1 Ab
production were assessed after immunization of OF1
mice with TNP-LPS and repetitive treatment with
appropriate doses of compounds followed by the deter-
mination of antibody titers in the serum.23 Upon sub-
cutaneous (sc) administration, 4 was more effective than
1 already at the 30 mg/kg/day dose, which appears to
be the dose leading to the maximum pharmacological
efficacy in this model. Furthermore, 4 was highly active
even at 3 mg/kg/day, as shown by the 73 and 84%
inhibitions of TNP-specific immunoglobulin M (IgM) and
G (IgG) titers, respectively. At this dose, leflunomide
did not significantly inhibit either IgM or IgG (Table
2). Thus, the ED50 for the quinoline was estimated to
be <3 mg/kg/day sc. Moreover, after oral application (10
mg/kg/day), 4 also blocked >60% of specific Ab forma-
tion, indicating therefore effect-bioavailability.
To correlate the in vitro activity of 4 with its phar-
macodynamic (PD) effect, the time-concentration pro-
file of the compound in the blood of OF1 mice was
quantified by LC-MS after single-dose oral (30 mg/kg)
and parenteral (10 mg/kg) applications (Figure 4). The
maximum blood concentration was achieved 0.25 h after
oral application, and 10-fold higher exposure levels (2.6
µM) than the IC50 value of 0.3 µM persisted for up to 2
h postdose. The dose-normalized AUC ratio (po/iv)
pointed to an absolute bioavailability of 13% for 4.
Having demonstrated that 4 is a potent inhibitor of
TI-1 Ab responses in vivo, its efficacy in a xenotrans-
plantation setting was investigated (Table 3). To assess
the effect of compounds on Ab-mediated rejection of
xenotransplants, the model of Syrian hamster hearts
heterotopically transplanted into nude (athymic) mice
was used.24 The rejection process in these animals is
restricted to the B-cell response due to the genetic
deficiency of mature T cells. Hence, a xenograft can be
rejected only by a T-cell-independent (TI) Ab response
because a T-cell-dependent anti-xenograft Ab antibody
formation cannot be mounted. The survival of the
animals for 4 weeks with graft histology showing no Ab-
mediated damage was considered to be a successful end
point.
Having demonstrated that 3 and 4 have the in vitro
profile we were aiming at, a limited structure-activity
investigation was undertaken to identify the key ele-
ments responsible for B-cell inhibition and selectivity.
First, the presence and influence of the position of the
trifluoromethyl substituent were assessed. In compari-
son to 3, the corresponding phenyl derivative 5 was a
weak B-cell inhibitor, whereas the ortho CF3-substituted
derivative 6, although equipotent to 1, had a cytostatic
character because it was devoid of selectivity toward
T-cells and J urkat cells. In contrast, the meta isomer 7
was a highly selective compound with only 2 times
weaker B-cell inhibitory activity as compared to that of
3. Both derivatives having an additional substituent on
the quinoline moiety were inactive in the TNP-LPS
assay. In compound 8, the 7-Me group most probably
destroyed the planarity of the molecule by positioning
the amide out of the quinoline plane. Indeed, molecular
modeling inspection using both the CHARM and the
Tripos force fields indicated that the distance between
the carbonyl O and the 7-Me moiety is quite short (D )
2.5 Å) for a nonbonded interaction. The 2-CF3 substi-
tuted compound 9 was synthesized to probe the influ-
ence of the hydrogen bond between the N(1) and the
amide NH on the biological activity. Indeed, the pres-
ence of the electron-withdrawing group was expected
to decrease the electron density of N(1) and, thus,
weaken its acceptor potential. Given the experimentally
obtained structural data showing that 4 and 9 have the
same conformation in the solid state, the lack of activity
of 9 indicates that electron-rich groups are not tolerated
at position 2. In agreement with our pharmacophore

1990 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 12
Papageorgiou et al.
tion. As a general tolerability parameter, body weight
was recorded at the time of transplant and at autopsy.
No adverse effect on normal body weight gain was seen
in any treatment group.
Con clu sion s
On the basis of the nondependency of the inhibition
of Ab production from ubiquitous mechanisms such as
the well-established de novo nucleotide biosynthesis
blockade and capitalizing both on the probable multiple
mechanism of action of 1b and on its enzyme-bound
conformation, quinoline 3 was designed as a potentially
weak inhibitor of DHODH but effective immunosup-
pressant. The compound not only fulfilled the mecha-
nistic criteria set by showing no inhibition of DHODH
but also provided additional evidence on the feasibility
of achieving selective B-cell over T-cell inhibition and,
hence, developing specific drugs impeding Ab produc-
tion. In fact, the in vitro profile of the pyridyl derivative
4 sets a novel paradigm in the B-cell immunosuppres-
sion field because it has a 75-fold selectivity for B- over
T-cells (as judged by the MLR data) and no general
cytotoxicity up to >160-fold higher concentrations than
those required for B-cell inhibition. In agreement with
the in vitro potency, both subcutaneous and oral ad-
ministration of quinoline 4 very effectively prevented
Ab production in vivo as judged by the low IgM and IgG
titers resulting from the immunization of mice with
TNP-LPS antigen. IgG titers were less sensitive to the
effects of the compound than IgM titers. Leflunomide,
the reference compound, was clearly less potent in this
setting.
F igu r e 4. Concentration of 4 in mouse blood after oral and
intravenous administration. Results are mean ( SE of four
mice per time point.
Ta ble 3. Prolongation of Xenoheart Survival and Histological
Findingsa
compound
placebo
doseb
survivalc
histologyd
5, 5, 6, 6, 6, 6
AbR (n ) 6)
leflunomide (1)
50
30
>28 (n ) 6)
no AbR (n ) 6)
AbR (n ) 2),
ChR (n ) 2)
AbR (n ) 3)
7, 7, >28, >28
Given the doses applied in the in vivo Ab pharmaco-
logical studies, the extrapolation of the pharmacokinetic
(PK) results indicates that concentrations of 4 above its
IC50 value in the corresponding cellular assay have been
obtained. The persistence of such therapeutically rel-
evant concentrations for at least 2 h postdose seems to
be sufficient to prevent the onset of Ab against TNP-
LPS for 24 h. Consequently, the data resulting from this
mechanistic model indicate a PK/PD relationship with
quinoline 4.
3
7 (n ) 3)
4
30
10
>28 (n ) 6)
11, >28 (n ) 4)
no AbR (n ) 6)
CR (n ) 1),
no AbR (n ) 4)
no AbR (n ) 4)
no AbR (n ) 4)
AbR (n ) 3)
1
0.3
0.1
>28 (n ) 4)
>28 (n ) 4)
6, 6, 7
a
b
Hamster to athymic mouse xenotransplantation model. mg/
kg/day po in Labrafil 330 mg/mL, EtOH 110 mg/mL, corn oil up
to 1 mL. c In days. > indicates mouse sacrificed with a beating
d
graft. AbR, antibody-mediated rejection; CR, cellular rejection;
ChR, chronic rejection.
Very importantly for our purpose, 4 also prevented
graft rejection in the relevant mouse xenotransplanta-
tion model. Indeed, 4 is the most potent compound
tested in suppressing Ab-mediated xenograft rejection
under a once-daily dosing regimen with efficacy down
to 0.3 mg/kg/day po. On the basis of the generated PK
data, the exposure of the animals to the quinoline can
be expected to be below the IC50 values at this dosing.
However, such extrapolations are hazardous because
not only were different mouse strains used for the
xenotransplantation (athymic C57Bl/6) and PK (OF1)
experiments but also discrepancies in the PK behavior
of a compound between transplanted and healthy
animals are not rare. In addition, no predictions can be
made concerning the sensitivity to quinoline inhibition
of anti-xenograft Ab versus TNP-LPS Ab.
Daily oral administration of leflunomide for 4 weeks
effectively controlled xenograft rejection only at the dose
of 50 mg/kg/day. At 30 mg/kg/day only two of four
animals had a beating graft at 28 days post-transplant,
histology showing an acute Ab-mediated rejection in the
two grafts rejected at day 7. The remaining two showed
signs of chronic rejection at autopsy. The 3 mg/kg/day
dose was clearly ineffective, leading to the rejection of
all grafts within 7 days by Ab attack. The quinoline 4
showed efficacy from a daily oral dose of 30 mg/kg down
to 0.3 mg/kg and, thus, was ∼100-fold superior to
leflunomide (1). At the highest dose tested, the six grafts
survived to 28 days with no histological evidence of
rejection. Likewise, at 10 and 1 mg/kg all mice were
terminated after 4 weeks with beating grafts and
normal heart histology except that of one animal of the
former group that underwent cell-mediated rejection at
day 11. The minimum effective dose of 4 in this model
was established at ∼0.3 mg/kg/day po with xenografts
showing normal myocardium with no evidence of rejec-
Overall, the in vitro and in vivo results indicate that
the quinolinecarboxamide 4 is a highly potent and
selective agent for the blockade of Ab secretion in the
xenotransplantation setting. Moreover, the lack of long-
term adverse effects observed with 10 times higher

Quinolinecarboxamides as Xenograft Rejection Inhibitors
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 12 1991
lization (hexane/EtOAc) afforded 9 (0.42 g, 30%): mp 155-157
°C; 1H NMR (DMSO-d6) δ 12.40 (1H, s), 8.98 (1H, d, J ) 8.6),
8.62 (1H, d, J ) 7.4), 8.47 (1H, d, J ) 6.6), 8.21 (1H, d, J )
8.6), 8.03 (3H, m), 7.82 (2H, d, J ) 8.6). Anal. (C18H10F6N2O):
C, H, N.
(viii) N-[4-(Tr iflu or om eth yl)p h en yl]-1,2,3,4-tetr a h yd -
r oqu in olin e-8-ca r boxa m id e, 10. A solution of 3 (0.05 g) in
MeOH (30 mL) containing DMF (0.5 mL) was hydrogenated
over Pd/C overnight. After removal of the catalyst, the crude
was concentrated to give 10 (0.46 g, 95%): mp 252-254 °C;
1H NMR (DMSO-d6) δ 10.49 (1H, s), 7.98 (2H, d, J ) 8.2), 7.80
(2H, d, J ) 8.2), 7.55 (1H, d, J ) 6.6), 7.15 (1H, m), 7.09 (1H,
d, J ) 6.6), 6.61 (1H, t, J ) 6.6), 3.35 (2H, t, J ) 4.5), 2.78
(2H, t, J ) 5.2), 1.82 (2H, m). Anal. (C17H15F3N2O): C, H, N.
dosing emphasizes the substantial therapeutic potential
of this compound in Ab-mediated diseases.
Exp er im en ta l Section
(1) Ch em istr y. Compounds were characterized by 400 MHz
proton NMR at room temperature using a Bruker MX-400 or
DPX-400 spectrometer. Chemical shifts are expressed as parts
per million downfield from tetramethylsilane, and J values
are reported in hertz. Fast atom bombardment mass spectros-
copy (Xe, 8 keV) on a VG70-SE mass spectrometer was used
for the characterization of all reported compounds. C, H, and
N analyses were carried out with all substances biologically
evaluated, and (0.4% was acceptable. For chromatographic
purifications, the flash chromatography technique was applied
using 230-400 mesh silica gel.
(a ) Qu in olin e-8-ca r boxa m id es. Gen er a l P r oced u r e. To
a solution of 8-carboxyquinoline (1.0 g, 5.78 mmol) in CH2Cl2
containing 2-3 drops of DMF was added dropwise oxalyl
chloride (0.75 mL, 8.7 mmol). The resulting suspension was
stirred for 1 h at room temperature and then cooled to 0 °C.
At this temperature, a THF solution of the aniline (3.6 g, 17.32
mmol) was slowly added and the reaction stirred for 2 h at
room temperature. The reaction mixture was subsequently
poured onto 1 N aqueous NaHSO4 solution and extracted with
ethyl acetate. The crude amide was purified by silica gel
chromatography (eluent of EtOAc/hexane ) 4:6) to afford the
title compound.
(ix) (7-Meth yl)qu in olin e-8-ca r boxylic Acid , 11. To a
suspension of 2-amino-6-methylbenzoic acid (10.0 g, 6.6 mmol)
in 1,2-dichlorobenzene (150 mL) at 140 °C was slowly added
acrolein (1.1 mL, 3.3 mmol) with stirring, and the reaction was
further stirred at this tempereature for 3 h. After the mixture
had cooled to room temperature., the solvent was evaporated,
leaving a solid crude that mainly contained the starting
benzoic acid. Chromatographic separaration (eluent of EtOAc)
1
afforded (0.14 g, 12%): mp 142-143 °C; H NMR (DMSO-d6)
14.2 (1H, broad s), 8.94 (1H, m), 8.41 (1H, d, J ) 7.2), 7.98
(1H, d, J ) 7.2), 7.57 (2H, m), 2.47 (3H, s). (C11H9NO2) C, H,
N.
(2) P h a r m a cology. (a ) In h ibition of Mu r in e TNP -LP S
Resp on ses in Vivo. OF1 mice were immunized iv with
approriate concentrations of TNP-LPS (eliciting a TI-1 anti-
body response). To determine the effect of compounds, mice
received compounds on the day of immunization (day 0) and
the three following days either by sc or by po application (total
of four compound applications). On day 6, mice were bled and
DNP-specific antibodies of the IgM and IgG class in the serum
of individual mice were determined by ELISA using DNP-
BSA-coated plates and appropriate developing reagents. Titers
resulting in a defined OD at 405 nm were recorded.
(b) Heter otop ic Ha m ster Hea r t Xen ogr a ftin g in Ath y-
m ic Mice. The heart of a Syrian hamster was heterotopically
transplanted in the abdomen of a congenitally athymic (nu/
nu) C57Bl/6 mouse, with anastomoses between the donor and
recipient’s aorta and between the donor’s right pulmonary
artery and the recipient’s inferior vena cava. Compounds were
dissolved in Labrafil 330 mg/mL, absolute ethanol 110 mg/
mL, and 1 mL of corn oil and diluted with pure corn oil (2:3
v/v). The graft was monitored daily by palpation of the
abdomen. Rejection was concluded with cessation of heartbeat
and confirmed histologically. Animals were terminated after
4 weeks, and the transplant was subjected to conventional
histology.
(3) P h a r m a cok in etics. Male mice (OF1, ∼25 g) were
administered a solution (Labrafil plus ethanol plus corn oil)
of 4 either orally or by injection in the femoral vein with a
solution (polyethoxylated castor oil plus ethanol plus aqueous
NaCl). Blood was collected from the vena cava (four mice/time
point) at 0.12, 0.25, 0.50, 1, 2, 4, 8, and 24 h after the beginning
of the experiments. Electrospray mass spectroscopy was used
for the detection of the parent compound, and quantification
was based on integration of the peaks of the HPLC chromato-
gram corresponding to the ion m/z 317.
(i) N-[4-(Tr iflu or om eth yl)p h en yl]qu in olin e-8-ca r boxa -
1
m id e, 3: yield 47%; mp 111-112 °C; H NMR (DMSO-d6) δ
13.46 (1H, s), 9.12 (1H, d, J ) 4.3), 8.58 (2H, m), 8.23 (1H, dd,
J 1 ) 8.2, J 2 ) 1.5), 8.05 (2H, d, J ) 8.5), 7.77 (2H, t, J ) 7.9),
7.72 (2H, d, J ) 7.9). Anal. (C17H11F3N2O): C, H, N.
(ii) N-[5-(Tr iflu or om eth yl)p yr id in -2-yl]qu in ole-8-ca r -
1
boxa m id e, 4: yield 45%; mp 170-172 °C; H NMR (DMSO-
d6) δ 14.35 (1H, s), 9.16 (1H, dd, J 1 ) 4.3, J 2 ) 1.6), 8.80 (2H,
m), 8.76 (1H, d, J ) 7.4), 8.66 (1H, d, J ) 8.3), 8.56 (1H, d, J
) 8.7), 8.33 (1H, d, J ) 8.1), 7.85 (1H, t, J ) 7.8), 7.77 (1H,
dd, J 1 ) 8.3, J 2 ) 4.3). Anal. (C16H10F3N3O): C, H, N.
(iii) N-(P h en yl)qu in olin e-8-ca r boxa m id e, 5: yield 52%;
mp 132-134 °C; 1H NMR (DMSO-d6) δ 13.85 (1H, s), 9.11 (1H,
d, J ) 4.3), 8.68 (1H, d, J ) 8.4), 8.52 (1H, d, J ) 8.4), 8.22
(1H, d, J ) 8.8), 7.89 (2H, d, J ) 7.2), 7.72 (1H, t, J ) 7.4),
7.67 (1H, d, J ) 8.4), 7.45 (2H, t, J ) 7.6), 7.15 (1H, t, J )
7.2). Anal. (C16H12N2O): C, H, N.
(iv) N-[2-(Tr iflu or om eth yl)p h en yl]qu in olin e-8-ca r box-
1
a m id e, 6: yield 42%; mp 128-130 °C; H NMR (DMSO-d6) δ
14.05 (1H, s), 9.08 (1H, m), 8.81 (1H, d, J ) 7.4), 8.68 (1H, d,
J ) 7.4), 8.46 (1H, d, J ) 7.4), 8.34 (1H, d, J ) 7.5), 8.82 (4H,
m), 7.37 (1H, t, J ) 7.6). Anal. (C17H11F3N2O): C, H, N.
(v) N-[3-(Tr iflu or om eth yl)p h en yl]qu in olin e-8-ca r box-
a m id e, 7: yield 22%; mp 115-116 °C; 1H NMR (CDCl3) δ 9.08
(1H, m), 8.97 (1H, d, J ) 4.3), 8.36 (1H, d, J ) 7.4), 8.16 (1H,
s), 8.09 (1H, d, J ) 7.9), 8.04 (1H, d, J ) 7.4), 7.77 (1H, t, J )
8.4), 7.58 (1H, dd, J 1 ) 8.4, J 2 ) 4.3), 7.51 (1H, t, J ) 7.9),
7.39 (1H, d, J ) 7.9). Anal. (C17H11F3N2O): C, H, N.
(vi) 7-Meth yl-N-[4-(tr iflu or om eth yl)p h en yl]qu in olin e-
8-ca r boxa m id e, 8: yield 22%; mp 190-192 °C; 1H NMR
(DMSO-d6) δ 13.40 (1H, s), 9.14 (1H, dd, J 1 ) 4.3, J 2 ) 1.6),
8.61 (1H, dd, J 1 ) 8.3, J 2 ) 1.6), 8.33 (1H, d, J ) 8.2), 8.17
(2H, d, J ) 8.1), 7.81 (2H, t, J ) 7.9), 7.70 (2H, m), 2.53 (3H,
s). Anal. (C18H13F3N2O): C, H, N.
(vii) N-[4-(Tr iflu or om eth yl)p h en yl]-2-(tr iflu or om eth -
yl)qu in olin e-8-ca r boxa m id e, 9. To a cooled (-78 °C) solu-
tion of 8-bromo-2-(trifluoromethyl)quinoline 1515 (1.0 g, 3.6
mmol) in diethyl ether (50 mL) containing N,N,N′,N′-tetram-
ethylethylenediamine (2.0 mL, 13.5 mmol) was slowly added
n-BuLi (3.12 mL of a 1.6 M solution in hexane, 4.6 mmol). After
30 min of stirring at this temperature, 4-(trifluoromethyl)-
phenylisocyanate (0.71 mL, 4.9 mmol) was added dropwise,
the cooling bath was removed, and the reaction was allowed
to reach room temperature. After addition of water and
extraction of the product with EtOAc, the organic phase was
dried over MgSO4 and concentrated. Chromatographic puri-
fication (eluent of hexane/EtOAc ) 3:1) followed by recrystal-
Ack n ow led gm en t. We express our warmest thanks
to Dr. Valerie Hungerford for running the DHODH
assay and to Dr. Claus Ehrhardt for the modeling
investigations concerning the conformation of compound
8.
Refer en ces
(1) Diamond, L. E. Progress in xenotransplantation. Expert Opin.
Invest. Drugs 1999, 8, 609-623.
(2) Lambrigts, D.; Sachs, D. H.; Cooper, D. K. C. Discordant organ
xenotransplantation in primates: world experience and current
status. Transplantation 1998, 66, 547-561.

1992 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 12
Papageorgiou et al.
(3) J oziasse, D. H.; Oriol, R. Xenotransplantation: the importance
of the GalR1,3Gal epitope in hyperacute vascular rejection.
Biochim. Biophys. Acta 1999, 1455, 403-418.
(14) Papageorgiou, C.; Zurini, M.; Weber, H.-P.; Borer, X. Lefluno-
mide’s bioactive metabolite has the minimal structural require-
ments for the efficient inhibition of human dihydroorotate
dehydrogenase. Bioorg. Chem. 1997, 25, 233-238.
(15) Liu, S.; Neidhardt, E. A.; Grossman, T. H.; Ocain, T.; Clardy, J .
Structures of human dihydroorotate dehydrogenase in complex
with anti-proliferative agents. Structure 2000, 8, 25-33.
(16) Keller, H.; Schlosser, M. 2-(Trifluoromethyl)quinolines from
anilines: a novel mode of isomerization and cyclisation. Tetra-
hedron 1996, 52, 4637-4644.
(17) Schlosser, M.; Keller, H. The Diels-Alder route to trifluorom-
ethyl-substituted carbo- and heterocycles. Liebigs Ann. Chem.
1995, 1587-1589.
(18) Rihs, G.; Papageorgiou, C. 8-Quinolinecarboxamides. Acta Crys-
tallogr. 2001, in press.
(19) J acobs, D. M.; Morrison, D. C. Stimulation of a T-independent
primary anti-hapten response in vitro by TNP-lipopolysaccha-
ride. J . Immunol. 1975, 114, 360-364.
(20) Skelly, R. R.; Muckenbeck, P.; Morrison, D. C. Stimulation of
T-independent antibody responses by hapten-lipopolysaccharides
without repeating polymeric structure. Infect. Immunol. 1979,
23, 287-293.
(21) Davies, J . P.; Cain, G. A.; Pitts, W. J .; Magolda, R. L.; Copeland,
R. A. The immunosuppressive metabolite of leflunomide is a
potent inhibitor of human dihydroorotate dehydrogenase. Bio-
chemistry 1996, 35, 1270-1273.
(22) DeFranco, A. L. The complexity of signaling pathways activated
by the BCR. Curr. Opin. Immunol. 1997, 9, 296-308.
(23) Ryffel, B.; Di Padova, F.; Schreier, M. H.; Le Hir, M.; Eugster,
H.-P.; Quesniaux, V. F. J . Lack of type 2 T cell-independent
responses and defect in isotype switching in TNF-lymphotoxin
a-deficient mice. J . Immunol. 1997, 158, 2126-2133.
(24) Kitamura, D.; Roes, J .; Ku¨hn, R.; Rajewsky, K. A B cell-deficient
mouse by targeted disruption of the membrane exon of the
immunoglobulin µ chain gene. Nature 1991, 350, 423-431.
(4) Zaidi, A.; Schmoeckel, M.; Bhatti, F.; Waterworth, P.; Tolan, M.;
Cozzi, E.; Chavez, G.; Langford, G.; Thiru, S.; Wallwork, J .;
White, D.; Friend, P. Life-supporting pig-to-primate renal xe-
notransplantation using genetically modified donors. Trans-
plantation 1998, 65, 1584-1590.
(5) Brouard, S.; Gagne, K.; Blancho, G.; Soulillou, J .-P. T cell
response in xenorecognition and xenografts: A review. Hum.
Immunol. 1999, 60, 455-468.
(6) Sato, K.; Takigami, K.; Miyatake, T.; Czismadia, E.; Latinne,
D.; Bazin, H.; Bach, F. H.; Soares, M. P. Suppression of delayed
xenograft rejection by specific depletion of elicited antibodies of
the IgM isotype. Transplantation 1999, 68, 844-854.
(7) Lambrights, D.; Van Calster, P.; Xu, Y.; Awwad, M.; Neethling,
F. A.; Kozlowski, T.; Foley, A.; Watts, A.; Chae, S. J .; Fishman,
J .; Thall, A. J .; White-Scharf, M. E.; Sachs, D. A.; Cooper, D. K.
C. Pharmacologic immunosuppressive therapy and extracorpo-
real immunoadsorption in the suppression of anti-RGal antibody
in the baboon. Xenotransplantation 1998, 5, 274-283.
(8) Paul, L. C. Adverse effects of immunosuppressive drugs on long-
term graft outcome. Transplant. Proc. 1998, 30, 4001-4004.
(9) Papageorgiou, C.; Albert, R.; Floersheim, P.; Lemaire, M.; Bitch,
F.; Weber, H.-P.; Andersen, E.; Hungerford, V.; Schreier, M. H.
Pyrazole bioisosteres of leflunomide as B-cell immunosuppres-
sants for xenotransplantation and chronic rejection: scope and
limitations. J . Med. Chem. 1998, 41, 3530-3538.
(10) Pankiewicz, C. Inhibitors of inosine monophosphate dehydro-
genase as potential chemotherapeutic agents. Expert Opin. Ther.
Patents 1999, 9, 55-65.
(11) Batt, D. G. Inhibitors of dihydroorotate dehydrogenase. Expert
Opin. Ther. Patents 1999, 9, 41-54.
(12) Fox, R. I. Mechanism of action of leflunomide in rheumatoid
arthritis. J . Rheumatol. 1998, 53 (Suppl.), 20-26.
(13) Papageorgiou, C.; Albert, R.; Andersen, E.; Zurini, M.; Hunger-
ford, V.; Floersheim, P.; Schreier, M. H. Inhibition of cyclosporin-
resistant B-cell antigen responses by pyrazoles. A tool for the
investigation of novel molecular mechanisms of B-cell activation.
Transplant. Proc. 2001, 33, 2263-2265.
J M010822M