Novel inhibitors of leukocyte transendothelial migration

B i o o r g an i c C h e m i s tr y 9 2 (2 019 )1 03250

Contents lists available at ScienceDirect

Bioorganic Chemistry

journal homepage: www.elsevier.com/locate/bioorg

Novel inhibitors of leukocyte transendothelial migration

Tamar Getter^a,1, Raanan Margalit^b,e, Shirin Kahremany^a,1, Laura Levy^a, Eliav Blum^a,

Netaly Khazanov^a, Nimrod Y. Keshet-Levy^a,c, Tigist Y. Tamir^d, M. Ben Major^d, Ron Lahav^e,

⁎

Soﬁa Zilber^c, Hanoch Senderowitz^a, Paul Bradﬁeld^f, Beat A. Imhof^g, Evgenia Alpert^e,

,

⁎

Arie Gruzman^a,

^aDivision of Medicinal Chemistry, Department of Chemistry, Faculty of Exact Sciences, Bar-Ilan University, Ramat-Gan, Israel

^b“Science in Action”, Ness-Ziona, Israel

^cDepartment of Pathology, Shaare Zedek Medical Center, Jerusalem, Israel

^dDepartment of Pharmacology and the Lineberger Comprehensive Cancer Center, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599, USA

^e“AltA-ZuZ Therapeutics“, Ness-Ziona, Israel

^f“MesenFlow Technologies“, Geneva, Switzerland

^gDepartment of Pathology and Immunology, University of Geneva, Geneva, Switzerland

A R T I C L E I N F O

A B S T R A C T

Keywords:

Leukocyte transendothelial migration is one of the most important step in launching an inﬂammatory immune

response and chronic inﬂammation can lead to devastating diseases. Leukocyte migration inhibitors are con-

sidered as promising and potentially eﬀective therapeutic agents to treat inﬂammatory and auto-immune dis-

orders. In this study, based on previous trioxotetrahydropyrimidin based integrin inhibitors that suboptimally

blocked leukocyte adhesion, twelve molecules with a modiﬁed scaﬀold were designed, synthesized, and tested in

vitro for their capacity to block the transendothelial migration of immune cells. One of the molecules, namely,

methyl 4-((2-(tert-butyl)-6-((2,4,6-trioxotetrahydropyrimidin-5(2H)-ylidene) methyl) phenoxy) methyl)

benzoate, (compound 12), completely blocked leukocyte transendothelial migration, without any toxic eﬀects

on immune or endothelial cells (IC₅₀= 2.4 µM). In vivo, compound 12 exhibited signiﬁcant therapeutic eﬀects in

inﬂammatory bowel disease (IBD)/Crohn’s disease, multiple sclerosis, fatty liver disease, and rheumatoid ar-

thritis models. A detailed acute and chronic toxicity proﬁle of the lead compound in vivo did not reveal any toxic

eﬀects. Such a type of molecule might therefore provide a unique starting point for designing a novel class of

leukocyte transmigration blocking agents with broad therapeutic applications in inﬂammatory and auto-immune

pathologies.

Leukocyte transmigration

Trioxotetrahydropyrimidin derivatives

Multiple sclerosis

IBD/Crohn’s disease

Arthritis

Fatty liver

Abbreviations: ANOVA, analysis of variance; AUC, area under the curve; CII, type II collagen; CNS, central nervous system; CXCL, chemokine (C-X-C motif) ligand;

CIA, collagen-induced arthritis; CFA, complete Freund’s adjuvant; CD, chow diet; CRN, clinical research network; DDW, double distillated water; DAI, disease activity

index; DMEM, Dulbecco's modiﬁed Eagle medium; DMF, dimethylformamide; DMSO, dimethyl sulfoxide; DSS, dextran sulfate sodium salt; EAE, experimental

autoimmune encephalomyelitis; ECs, endothelial cells; ESI, electrospray ionization; FCS, fetal calf serum; FLIP, fatty liver inhibition of progression; F¹⁹-NMR,

Fluorine-19 nuclear magnetic resonance spectroscopy; GLP, good laboratory practice; H&E, hematoxylin and eosin; HPLC, high-performance liquid chromatography;

HUVEC, human umbilical vein endothelial cells; HRMS, high-resolution mass spectra; IBD, inﬂammatory bowel disease; Interferon γ, (IFNγ); IP, intraperitoneal; IFD,

induced ﬁt docking; IV, intravenous; JAM-C, junction adhesion molecule 1; KO, knockout; LBRC, lateral border recycling compartment; LDT, leucine–aspartic

acid–threonine; MAdCAM-1, mucosal vascular addressin cell adhesion molecule 1; MOG, myelin oligodendrocyte glycoprotein; NW, normal water; NAFLD, non-

alcoholic fatty liver dystrophy; NAS, NAFLD activity score; PECAM-1, platelet endothelial cell adhesion molecule 1; RPMI, Roswell park memorial institute; PPAR-γ,

peroxisome proliferator-activated receptor; PO, per os (oral administration); PTX, pertussis toxin; PBS, phosphate buﬀer solution; RIPA, radioimmunoprecipitation

assay buﬀer; ROC, receiver operating characteristic; SC, sub-cutaneous; SEM, standard error measurements; THF, tetrahydrofuran; TLC, thin-layer chromatography;

TMS, tetramethylsilane TNF-α, tumor necrosis factor α; VCAM-1, vascular cell adhesion protein 1; WD, western diet

^⁎Corresponding authors.

E-mail addresses: genia.a@ayalapharma.com (E. Alpert), gruzmaa@biu.ac.il (A. Gruzman).

¹Present addresses: Department of Ophthalmology, Gavin Herbert Eye Institute, UCI Health, Orange County, CA, USA.

https://doi.org/10.1016/j.bioorg.2019.103250

Received 30 May 2019; Received in revised form 2 September 2019; Accepted 3 September 2019

A v a i l a b l e o n l i n e 0 9 S ep t e m b e r 2 019

T. Getter, et al.

BioorganicChemistry92(2019)103250

1. Introduction

Inﬂammatory and autoimmune diseases represent a signiﬁcant

proportion of devastating human pathologies [1–4]. Dozens of drugs

are currently in use to treat these disorders [5–7]. However, often

complete remission is not achieved and most of the drugs have serious

side eﬀects [8,9]. Several autoimmune and chronic inﬂammatory dis-

eases, such as Crohn’s disease, multiple sclerosis, ankylosing spondy-

litis, psoriasis, rheumatoid arthritis, or lupus erythematosus and many

others are not treatable or have poor treatment response and are often

fatal [10,11].

Thus, the discovery and design of novel drugs against inﬂammatory

and autoimmune diseases is one of the major directions in the pharma-

ceutical industry and in academia. For example, there were around 400

ongoing clinical trials for treating rheumatoid arthritis in 2018 alone.

https://www.centerwatch.com/clinical-trials/listings/therapeutic-area/

19/rheumatology/.

Therefore, the molecular processes that deﬁne and regulate the re-

cruitment of leukocytes as part of inﬂammation have been identiﬁed as

potential targets in drug development since the 1990s [12–14]. His-

torically, integrins have been the focus of drug development and their

inhibition was generated by monoclonal antibodies, peptides/peptido-

mimetics, and small organic molecules [15]. Several integrin antago-

nists or modulators of their binding ligands have exhibited clinically

signiﬁcant beneﬁts, leading to continued medical interest in further

development of novel integrin inhibitors [16]. Currently, almost all

human integrins have been used as a target for developing eﬀective

antagonists [14].

Fig. 1. Ligand-based pharmacophore model developed using the LigandScout

program. (A) A pharmacophore model design based on the trioxotetrahy-

dropyrimidin derivative structures described by Harriman et al. (B) The phar-

macophore model includes two aromatic rings (blue)- AR, one hydrophobic

(yellow)- H, three hydrogen acceptor features (red)- HBA, and one hydrogen

donor (green)- HBD. (C) A pharmacophore mapped 2D feature representation of

the designed compounds. The features are color-coded, as described above.

Integrin α4 β7 plays an important role in the leukocyte transmi-

gration process [15]. One of the most important binding partners of

integrin α4 β7 is mucosal vascular addressin cell adhesion molecule 1

(MAdCAM-1) [17]. This protein is predominantly expressed in the

mesenteric lymph nodes and Peyer’s patches [18,19]. However, under

inﬂammatory stimuli, the expression level of MAdCAM-1 in vascular

endothelial cells is upregulated [20]. The protein translocated to the

plasma membrane and leads to augmented adhesion of α4 β7 integrin-

positive lymphocytes to the endothelium [21]. In particular, the inter-

action between α4 β7 and MAdCAM-1 is considered to be one of the

most vital events in regulating leukocyte traﬃcking in the gastro-

intestinal tract [22,23]. It is therefore not surprising that both proteins

have been shown to be involved in the pathophysiology of in-

ﬂammatory bowel diseases (IBD) [24], and they have been targeted by

several drugs, for example, by vedolizumab, a humanized monoclonal

antibody that recognizes the gut-speciﬁc α4β7 integrin [25,26].

In addition to biological drugs, small-molecule inhibitors of that

interaction were also developed. In 1996 Fong’s group at Genentech

showed that MAdCAM-1 binds to α4 β7 integrin via the speciﬁc amino

acid sequence leucine–aspartic acid–threonine (LDT) motif [27]. Since

then, several mimicking LDT moiety inhibitors have been developed

[28,29]. In addition, some very potent MAdCAM-1 and α4 β7 integrin

interaction blockers have been developed without targeting the LDT

sequence [30–32]. In 2008, Briskin’s group at Millenium Pharmaceu-

ticals published a paper in which they described the discovery of a

trioxotetrahydropyrimidin-based antagonist that speciﬁcally blocks the

α4 β7 integrin MAdCAM-1 interaction, without indicating that this

inhibition was mediated by the LDT sequence. All compounds reported

by the group were selected via high-throughput screening results based

on the static whole-cell adhesion assay [33]. In addition, several other

trioxotetrahydropyrimidin derivatives were reported as anti-in-

ﬂammatory agents [34].

and B lymphocyte transendothelial migration in vitro. A signiﬁcant in

vivo eﬀect of compound 12 was also observed in multiple sclerosis,

arthritis, fatty liver, and in IBD mouse models. Compound 12 did not

display any toxic eﬀects in mice in concentrations up to 200 mg/kg, at

multi-dose administration during a 28-day period. Surprisingly, com-

pound 12 did not block MAdCAM-1 as we had previously anticipated.

Based on in silico calculation, we now hypothesize that compound 12

instead may block platelet endothelial cell adhesion molecule

1

(PECAM-1) based transendothelial migration of leukocytes. Such types

of molecules might therefore provide a unique starting point for de-

signing a novel class of leukocyte transmigration blocking agents with

broad therapeutic applications.

2. Results and discussion

2.1. Chemistry

Based on the trioxotetrahydropyrimidin derivative structures, de-

scribed by Harriman et al. [33] (Supporting Information Chart 1), we

designed a ligand-based pharmacophore model using the LigandScout

platform (Fig. 1A) [35,36]. The resulting pharmacophore model con-

tained three hydrogen acceptor features, which represented the oxygen

of the carbonyl moieties, one hydrogen donor representing the amine

donor, aromatic rings, and two merged hydrophobic aromatic features,

representing the aromatic substitution rings in the structures of all

trioxotetrahydropyrimidin derivatives. In order to simplify the struc-

tures of the designed compounds and to avoid the formation of cis and

trans isomers by substituting the amide moiety of trioxotetrahy-

dropyrimidin (similarly to all compounds that were developed by

Harriman et al.), we decided to omit one of the merged pharmacophore

features that represent the substitution of the amide moiety of triox-

otetrahydropyrimidin.

Here we describe the design, synthesis, and biological evaluation of

twelve trioxotetrahydropyrimidin derivatives that signiﬁcantly diﬀered

in structure from the set of compounds previously described by the

group of M. Briskin. The lead compound, methyl 4-((2-(tert-butyl)-6-

((2,4,6-trioxotetrahydropyrimidin-5(2H)-ylidene) methyl) phenoxy)

methyl) benzoate (12), completely inhibited monocyte, T lymphocytes,

The pharmacophore (Fig. 1B), which was changed accordingly, was

used for aligning the designed compounds 1–4, 11–14, 26, and 22–24,

and for evaluating their suitability for the pharmacophore (Fig. 1C). By

2

T. Getter, et al.

BioorganicChemistry92(2019)103250

The synthesis procedure for more complex novel trioxotetrahydropyr-

imidin derivatives is presented in Scheme 2. Using another classical

organic synthesis method, Fischer-Speier esteriﬁcation, a 4-bromo-

methylbenzoic acid was reﬂuxed with methyl alcohol in the presence of

a catalytic amount of H₂SO₄to obtain the corresponding ester: methyl

4-(bromomethyl) benzoate, (5) [41]. This was followed by reaction of 5

with commercially available hydroxybenzaldehydes, using the proce-

dure described by Zidar et al. [42]. In which the bromine moiety

changes in situ with a better leaving group (iodine) and by using a weak

basic compound (potassium carbonate) to deprotonate the phenol

moiety to couple with iodobenzyl to obtain the corresponding ether.

The resulting substituted aldehydes; methyl 4-((2-formylphenoxy) me-

thyl) benzoate (6), methyl 4-(3-formylphenoxy) benzoate (7), methyl 4-

((2-(tert-butyl)-6-formylphenoxy) methyl) benzoate (8), and methyl 4-

((4-formyl-2,6-dimethylphenoxy) methyl) benzoate (9) were coupled

with barbituric acid using Knoevenagel condensation (without the ad-

dition of a basic catalyst) to obtain methyl 4-((2-((2,4,6-trioxotetrahy-

dropyrimidin-5(2H)-ylidene) methyl) phenoxy) methyl) benzoate (10),

methyl 4-((3-((2,4,6-trioxotetrahydropyrimidin-5(2H)-ylidene) methyl)

phenoxy) methyl) benzoate (11), methyl 4-((2-(tert-butyl)-6-((2,4,6-

trioxotetrahydropyrimidin-5(2H)-ylidene) methyl) phenoxy) methyl)

benzoate (12), and methyl 4-((2,6-dimethyl-4-((2,4,6-trioxotetrahy-

dropyrimidin-5(2H) ylidene) methyl) phenoxy) methyl) benzoate (13),

Scheme 1. Synthesis of four benzylidene pyrimidine derivative compounds

1–4: EtOH:H₂O 1:1, barbituric acid, under reﬂex overnight.

performing this step, we developed compounds with the potential to

block leukocyte transmigration process.

First, four trioxotetrahydropyrimidin derivatives were synthesized,

as shown in Scheme 1, using classical Knoevenagel condensation be-

tween acidic hydrogens located between two carbonyls of trioxote-

trahydropyrimidin and the corresponding aromatic aldehydes. Usually,

a weak basic amine is used to catalyze such a reaction; however, the

high acidic properties of a trioxotetrahydropyrimidin lead to deproto-

nation without the addition of a basic catalyst. First, four trioxote-

trahydropyrimidin derivatives were obtained: 8H-benzo[7,8] chromeno

[2,3-d] pyrimidine-8,10(9H)-dione, (1) 5-(3-acetylbenzylidene) pyr-

imidine-2,4,6(1H,3H,5H)-trione, (2) 5-(3,4-dihydroxybenzylidene)

pyrimidine-2,4,6(1H,3H,5H)-trione, (3) 5-(anthracen-9-ylmethylene)

pyrimidine-2,4,6(1H,3H,5H)-trione, and (4) using known synthetic

procedures [37,38].

respectively.

Compound

4-((2-((2,4,6-trioxotetrahydropyrimidin-

5(2H)-ylidene) methyl) phenoxy) methyl) benzoic acid (14) was ob-

tained using strong basic catalysis of a hydrolysis of the ester bond in

the presence of sodium hydroxide. Following an almost identical syn-

thetic pathway, but using as a starting material a 3-(chloromethyl)-

benzoic acid, an additional set of novel trioxotetrahydropyrimidin de-

rivatives were synthesized, as shown in Scheme 3. This set consists of

methyl

3-((2-methyl-6-((2,4,6-trioxotetrahydropyrimidin-5(2H)-yli-

dene) methyl) phenoxy) methyl) benzoate (19), methyl 3-((2-((2,4,6-

trioxotetrahydropyrimidin-5(2H)-ylidene) methyl) phenoxy) methyl)

benzoate (20), methyl 3-((4-((2,4,6-trioxotetrahydropyrimidin-5(2H)-

ylidene) methyl) phenoxy) methyl) benzoate (21), 3-((2-methyl-6-

((2,4,6-trioxotetrahydropyrimidin-5(2H)-ylidene) methyl) phenoxy)

methyl) benzoic acid (22), 3-((2-((2,4,6-trioxotetrahydropyrimidin-

5(2H)-ylidene) methyl) phenoxy) methyl) benzoic acid (23), 3-((4-

((2,4,6-trioxotetrahydropyrimidin −5(2H) ylidene) methyl) phenoxy)

methyl) benzoic acid (24), and methyl 3-((3-((4,6-dioxo-1,3,5-tria-

zinan-2-ylidene) methyl) phenoxy) methyl) benzoate (26).

The structures of compounds 3 and 4 were already reported in the

scientiﬁc literature [39,40]. However, compounds 1 and 2 are novel. It

is important to mention that compound 1 was obtained unexpectedly.

We planned to synthesize a molecule with a free hydroxyl group on the

ortho to the aldehyde position. However, as

a result of an in-

tramolecular reaction, the formation of a 3,6-dihydro-2H-pyran deri-

vative was observed. We presume that the driving force for the cycli-

zation was the formation of a stable six-member ring with conjugated

double bonds. As a result, only an O-cyclic compound was obtained

with high yield (around 95%) and no compound with free hydroxyl.

In Scheme 4 shown the synthetic pathway for the preparation of ¹⁹

F

labeled derivative of compound 12 which was obtained as described in

Scheme 2. Synthetic pathways of four benzylidene pyrimidine derivative compounds 11–14: (i) MeOH, H₂SO₄, under reﬂex for 5 h. (ii) K₂CO₃, KI, salicylaldehyde,

CH₃CN, 70 °C for 5 h. (iii) EtOH/H₂O 1:1, barbituric acid, under reﬂex overnight. (iv) 2 M NaOH, 1,4 dioxan:MeOH:H₂0 1:1:1, under reﬂex for 5 h.

3

T. Getter, et al.

BioorganicChemistry92(2019)103250

Scheme 3. Synthetic pathways of four benzylidene pyrimidine derivative compounds 22–24 and 26: (i) MeOH, H₂SO₄, under reﬂex for 5 h. (ii) K₂CO₃, KI, CH₃CN,

70 °C for 5 h. (iii) EtOH/H₂O 1:1, barbituric acid, reﬂex overnight. (iv) 2 M NaOH, 1,4 dioxan:MeOH:H₂0 1:1:1, under reﬂex for 5 h.

Scheme 1. Compound 30 was used for the determination of the level of

representing capillary ﬂow rates in vivo (Supporting Information Fig.

S1). Using phase contrast bio-imaging, we were able to ﬁlm (Movie 1)

the perfused leukocytes undergoing capture, their ﬁrm adhesion to

endothelial cells, followed by transendothelial migration and thus de-

termine how this process may be aﬀected by the tested compounds. A

summary of the results is presented in Fig. 2. Brieﬂy, a monocyte sus-

pension was ﬂowed over tumor necrosis factor α (TNF- α) activated

HUVECs for the ﬁrst 5 min of the experiment, before switching to the

wash buﬀer for 40 min. The adherent monocytes appeared as phase-

white, which rapidly changed to phase-gray as the cells became ﬁrmly

adherent. Monocytes that transmigrated through the HUVEC mono-

layer changed from a phase-gray to a phase-black appearance. This

change in the cells’ appearance and the phases in migration enabled us

to measure every aspect of monocyte recruitment. Under control con-

ditions, the captured monocytes became adherent within 1 min, and

transmigrated within 15 min of ﬂow co-culture. Four compounds: 4, 11,

12, and 26 displayed a signiﬁcant inhibitory eﬀect in this screening

system.

HeLa cell labeling by ¹⁹F NMR.

2.2. In vitro study of the anti-inﬂammatory eﬀects

The synthesized compounds were evaluated in vitro using a bio-

imaging platform that provides a detailed analysis of individual leu-

kocyte interactions with vascular endothelial cells (ECs) under phy-

siological ﬂow that occurs during inﬂammatory reactions [43]. This

technique allowed us to determine the possible eﬀects of compounds on

four major mechanisms that deﬁne human leukocyte traﬃcking using

primary cells: (1) the initial capture of leukocytes in free-ﬂow onto

endothelial luminal surfaces, which is rapidly followed by (2) ﬁrm

adhesion, and (3) the migration of monocytes on the endothelial surface

(pre-transmigration event). Luminal movement can then lead to mi-

gration of the leukocyte across the endothelial barrier into the ablum-

inal compartment, termed leukocyte trans-endothelial migration (be-

tween EC junctions) area, followed by (4) migration into the abluminal

extracellular matrix or eventually, reverse transmigration back into the

vascular lumen.

Another parameter that was measured in the monocyte transmi-

gration assay is the level of adhesion of monocytes to the endothelial

cells. This parameter was used to determine whether the inhibitory

eﬀects of active compounds are related to their intervention with ad-

hesion or to the transmigration pathway itself. The eﬀect of the most

active compounds 4, 12, and 26 on the adhesion of monocytes was

tested. We dropped in this stage compound 11, because it was less

active then 12 and 26 but its activity was similar to 4. The adhesion was

The underlying traﬃcking mechanisms of human monocytes were

investigated using a ﬂow assay bio-imaging system in the presence of

the tested compounds [44]. A monolayer of human umbilical vein en-

dothelial cells (HUVECs) was cultured in ﬂow chambers that allowed

human monocytes, T- or B- lymphocytes to be perfused over the en-

dothelial monolayer under physiological shear stress conditions

Scheme 4. Synthetic pathways of ﬂuorinated benzylidene pyrimidine derivative compound 30: (i) MeOH, H₂SO₄, under reﬂex for 5 h. (ii) K₂CO₃, KI, salicylaldehyde,

CH₃CN, 70 °C for 5 h. (iii) EtOH/H₂O 1:1, barbituric acid, under reﬂex overnight. Compound 28 is a known molecule.

4

T. Getter, et al.

BioorganicChemistry92(2019)103250

Fig. 2. Monocyte transmigration on the surface of activated HUVECs under ﬂow. Primary human monocytes co-cultured on TNF-α-activated HUVECs. Test com-

pounds at 30 µM or DMSO (blue line) were added to cells for 45 min. Co-cultures were kept under a ﬂow of 0.05 Pa. (A). Compounds 12 (red line) and 26 (green line).

Total blocking of transmigration. (B) Compounds 4 (red line) and 11 (green line). High levels of blocking. (C) Compounds 13 (red line), 23 (green line), and 24

(purple line). Low levels of blocking. (D) Compounds 1 (red line), 2 (green line), 3 (purple line), 14 (orange line), and 22 (black line). No eﬀect. n = 3. Mean

*P ≤ 0.05.

SE,

found to be unaﬀected for all three tested compounds (Fig. 3A). The

eﬀect on the transmigration process was then determined again for all

three molecules (Fig. 3B).

located at para positions relative to each other. Compounds 24 and 23

do not have any methylated substitutions on the benzyl ring and the

aromatic orientation of the benzyl ester relative to the etheric bond is at

the meta position, compared with para in compound 12. Thus, these

diﬀerences might be responsible for reducing the inhibitory activity of

both compounds. Compounds 1–3, 14, and 22 were inactive.

Compounds 2 and 3 contain two rings: one is trioxotetrahydropyr-

imidin and the other is a benzylidene motif; both are similar to com-

pound 12. However, compound 12 has an additional third aromatic

substituted ring and both compounds 2 and 3 lack this structural

moiety. These data highlight the importance of the third aromatic ring

for biological activity. Compound 1 is the most rigid molecule synthe-

sized with intra-cyclization, aﬀording saturated four-member ring mo-

lecules that reduce activation. Finally, compounds 14 and 22, which

lack both t-butyl substitution, and methyester, which are critical for the

activity, exhibited no biological activity.

Two compounds, 12 and 26, exhibited a very eﬀective inhibition of

monocyte transmigration. However, compound 12 completely blocked

monocyte transmigration at a 30 µM concentration, and not even a

single blocked monocyte undergoing transmigration could be recorded

(Movie 1).

In general, from the most active compound, 12, several conclusions

of the structure-activity relationship results could be drawn. For ex-

ample, compound 4 has an anthracene moiety instead of the di-aro-

matic-substituted one in compounds 12 and 26. This compound ex-

hibited an inhibitory eﬀect similar to compounds 26 and 12 but with

less potency. This means that bulky substitution for trioxotetrahy-

dropyrimidin might be critical for achieving high-eﬃcacy compounds.

Compound 26 is similar to the most potent and active compound 12.

However, the absence of t-butyl substitution and the meta position of

the benzyl ether instead of the ortho one in compound 12 decreased

dramatically the activity of compound 26. Consequently, we propose

that the t-butyl moiety might be critical for binding to a hydrophobic

pocket of target protein. Compound 11 has no t-butyl substitution and

the trioxotetrahydropyrimidin groups are located at the meta position

relative to the etheric bond, compared with the ortho position in com-

pound 12, which signiﬁcantly aﬀects the potency of the binding.

Compounds 13, 23, and 24 were low-aﬃnity inhibitors of monocyte

transmigration. Compound 13 has two signiﬁcant structural diﬀer-

ences, compared with compound 12. First, two methyl groups are lo-

cated at ortho positions relative to the etheric aromatic group. Second,

the benzyl ester moiety and the trioxotetrahydropyrimidin fragment are

Based on these results, it could be concluded that the activity of the

identiﬁed compounds was via inhibition of the transendothelial mi-

gration process and that the eﬀect was not attributed to the adhesion

level of monocytes relative to the endothelial cells. The dose response

eﬀect of compound 12 was investigated in the monocyte/HUVEC

system by an identical ﬂow assay (Movie 2). We chose this molecule as

a lead compound for further biological investigation and an in-depth

study. Serial dilutions of compound 12, starting from 100 µM to 3 nM,

were used to establish the concentration that would give a 50 percent

inhibition (IC₅₀) using the monocyte transmigration ﬂow assay (Fig. 4).

As we have already observed in the ﬁrst screening, complete blocking of

monocyte transmigration was achieved at a concentration range of 30

to 100 µM, whereas partial blocking was achieved at concentrations as

5

T. Getter, et al.

BioorganicChemistry92(2019)103250

Fig. 5. Eﬀect of compound 12 on T cell recruitment. Co-culture was treated as

described in Fig. 4 (only for a longer time: 60 min) by compound 12 (at 30 µM

concentration red line) and by DMSO (blue line), Panel A represents the cap-

tured T cells. Panel B represents the % of T cell transmigration, n = 3.

Fig. 3. Adhesion (sub-Figure A) and transmigration (sub-Figure B) patterns of

human monocytes co-cultured on TNF-α-activated HUVECs under ﬂow in the

presence of compounds 12 (purple line), 26 (green line), and 4 (red line). Co-

culture was treated as described in Fig. 2 (the time was shorter: 30 min, DMSO

Mean

SE, *P ≤ 0.05.

is denoted by the blue line), n = 3. Mean

SE, *P ≤ 0.05.

3. In our hands, compound 12 also had a potent eﬀect on B cell re-

cruitment, blocking both B cell adhesion to vascular endothelium and

transendothelial migration. Compound 12 contrasted with Tysabri in an

unexpected manner. Whereas Tysabri has been described as blocking T

cell traﬃcking, it had no eﬀect on the number of B cells recruited. It had

only a moderate eﬀect on the blockade of B cell adhesion to en-

dothelium vascular; however, this was in contrast to the eﬃciency of

adherent B cells transmigrating at a higher rate. The net eﬀect was that

the number of transmigrating B cells in the ablumen of HUVECs was

identical to that of the control. This could occur for a number of reasons

(e.g., adhesion blockade of a speciﬁc B cell subset); the proﬁle of the

adherent versus the transmigrated B cells is being actively investigated.

This observation has many implications regarding how this transla-

tional approach can impact early drug development; for example, we

found that despite the poor blockade eﬀect on B cells’ eﬀect on Tysabri,

it is still a very eﬀective drug for treating MS and other inﬂammatory

conditions.

Fig. 4. Dose response eﬀect of compound 12 on monocyte transmigration on

the surface of HUVECs. Co-culture was treated as described in Fig. 3 by in-

creased concentrations of compound 12; 3 nM (red line), 30 nM (green line),

300 nM (purple line), 3 µM (black line), 30 µM (orange line) and 100 µM (gray

Based on these observations, we propose that compound 12 inter-

acts with vascular or leukocyte adhesion receptors that support leuko-

cyte traﬃcking. Therefore, we thought that compound 12 would

compete with an antibody for binding with three well-known adhesion

proteins: MAdCAM-1, vascular cell adhesion protein 1 (VCAM-1), and

junction adhesion molecule 1 (JAM-C). Commercially available ELISA

kits were used for this determination using HeLa (cervical carcinoma)

and Jeko-1 (mantle cell lymphoma) cell lines. Indeed, both types of

cells are not ideal in vitro models for inﬂammatory/auto-immune dis-

eases. However, HeLa cells express a constitutive level of MAdCAM-1

and VCAM-1 on their plasma membrane [49,50]. We planned to detect

the possible inhibition by compound 12 of the binding between

MAdCAM or VCAM-1 and their speciﬁc ligands (provided by the kits).

The use of HeLa cells was recommended by the provider of the kits. In

an identical assay we used Jeko-1 cells to test eventual blocking of JAM-

line). DMSO (blue line), n = 3. Mean

SE, *P ≤ 0.05.

low as 3 µM. The compound lost its activity at concentrations below

300 nM. The IC₅₀value for compound 12 was 2.4 µM, with no indica-

tion of cell toxicity at higher concentrations.

To determine whether our compound aﬀected lymphocyte traf-

ﬁcking, compound 12 was tested in a T lymphocyte recruitment assay.

Compound 12 had a profound eﬀect on both adhesion and transen-

dothelial migration. Similarly to monocytes, T cell transmigration was

completely blocked by compound 12 at a concentration of 30 µM

(Fig. 5). Compound 12 was also tested in a novel B lymphocyte re-

cruitment assay. For this purpose, a well-known drug Tysabri © (Na-

talizumab), an anti- α4 β1 integrin antibody, which blocks T cell re-

cruitment, was used as a control [45–48], as shown in Fig. 6 and Movie

6

T. Getter, et al.

BioorganicChemistry92(2019)103250

Fig. 6. Eﬀect of compound 12 and Tysabri on B cell recruitment. Co-culture was treated as described in Fig. 2 (only for a longer time: 60 min) by compound 12 or by

Tysabri (red line), both at a 30 µM concentration, in both experiments DMSO was added to control samples, (blue line). (Panels A, B, C) indicate the eﬀect of

compound 12 and (Panels D, E, F) indicate the eﬀect of Tysabri. Panels A and D represent the captured B cells. Panels B and E represent the % of B cell transmigration

and Panels C and F represent the transmigrated B cells, n = 3. Mean

SE, *P ≤ 0.05.

Fig. 7. Investigation of the possible inhibitory ef-

fect of compound 12 on the binding activity of

diﬀerent adhesion proteins using commercially

available kits. (A) JAM-C, compound 12 was used

at 30 µM. (B) VCAM-1, (C) MAdCAM-1. (D) Panel of

integrins, n = 6, Mean

SE. no signiﬁcant diﬀer-

ences were observed (NS, none signiﬁcant).

7

T. Getter, et al.

BioorganicChemistry92(2019)103250

C binding by compound 12 [51]. Fig. 7A, B and C shows that compound

12, at a concentration at which it displayed total inhibition of monocyte

transmigration, did not reduce the binding activity of these three tested

proteins to their speciﬁc antibodies. In addition, Jeko-1 cells were used

for identiﬁcation of possible eﬀect of compound 12 on the binding

between set of integrins to their antibodies. A commercial available kit

for measurement of the integrines binding was used (Fig. 7D). Also in

this assay no competition between compound 12 and antibodies were

detected. Obviously, we cannot draw any conclusions based on the

negative results obtained in those experiments. To detect a decreased

binding between a protein and its antibody in the presence of com-

pound 12 might be possible only if they compete for the same epitope

(the binding site of compound 12 is unknown), thus the lack of eﬀect

can be still explained by diﬀerent epitopes of binding for the antibody

and compound 12.

buﬀered formalin for future mounting in paraﬃn blocks. Sections from

each organ of each animal were then stained with Hematoxilin/Eosin

(H/E) and histological evaluation of the tissue was performed by mi-

croscopy to identify signs of toxicity. No signiﬁcant changes in white

and red blood were observed in blood from mice treated with com-

pound 12 in both administered routes, as shown in Supporting

Information Figs. S3 and S4. In addition, all biochemical tests on blood

from mice treated with compound 12 conﬁrmed that there were no

signs of toxicity, as shown in Supporting Information Figs. S5, S6, and

S7. Macro-evaluation of mouse tissues also revealed no signiﬁcant

changes in tissue morphology in animals treated by compound 12, as

shown in Supporting Information Figs. S8 and S9.

In addition, a good laboratory practice (GLP) toxicity experiment

was also conducted. Compound 12 was administered orally, daily for

28 days to C57 black mice, in three doses. The maximal dose was supra

pharmacological (200 mg/kg/day). The detailed protocol is presented

in the Supporting Information. The main conclusion, which was derived

from the study, is that compound 12 has no toxic eﬀects. Body weight,

clinical observations, necroscopy results, and blood tests were similar in

treated and non-treated compound 12 mice.

The next question was: “Does compound 12 target white blood cells

or endothelium, or both?” We therefore developed a strategy where

monocytes and HUVEC cells were treated separately with a test com-

pound, before co-culture in the ﬂow assay chamber. Treatment of

monocytes only, blocked monocyte transmigration (100%). Pre-treat-

ment of endothelium by compound 12 only before ﬂowing the mono-

cytes, induces rapid reverse transmigration when monocytes were

subsequently ﬂown of the endothelium (Supporting Information Fig.

S2A). This eﬀect might explain the potent therapeutic eﬀect of com-

pound 12, which we identiﬁed in vivo (discussed below) and the ability

of compound 12 to revert existing inﬂammation. Targeted reverse

leukocyte transendothelial migration would be a novel mechanism in

treatment of autoimmune diseases that would raise a lot of interest.

Leukocyte retention in tissue is a key factor in chronic disease and

drives in human chronic inﬂammation. Bringing leukocytes back into

the blood stream by reverse transendothelial migration could resolve

the inﬂammation. Additional in vitro-related results indicated pheno-

typic manifestations of a distinctive movement within the HUVEC in-

tercellular junctions observed after treatment with compound 12. We

termed this manifestation as “Fluttering”, because it was characterized

by a rapid ﬂuctuating movement (similar to butterﬂy wings) at the

intercellular junctions of adjacent HUVECs, and predominantly within

the tricellular corners. This was associated with absence of leukocyte

adhesion or migration within these regions (Supporting Information

Fig. S2B and Movie 4). “Fluttering” was observed in all HUVECs co-

cultured with monocytes that were treated with compound 12. In all

cases, junctional integrity was maintained, and no physical breakdown

was observed at the endothelial intercellular contacts. This therefore

represents a phenotype that is consistent with a profound disruption in

leukocyte transmigration.

The possible therapeutic eﬀect of compound 12 was evaluated in an

acute inﬂammation model of IBD/Crohn’s disease. C57BL mice were

treated with dextran sulfate sodium salt (DSS) to induce colitis [53]. On

day 5, water containing DSS was changed to regular drinking water and

treatment with compound 12 was initiated. Compound 12 was ad-

ministered at 15 mg/kg by intravenous injection (IV), IP, or sub-

cutaneous injection (SC) injections using a q3w regimen (administered

3 times per week) and at 30 mg/kg by PO using a qd regimen (once

daily). The treatment was continued for 10 days and thereafter, mice

were euthanized. For determining the severity of the disease and the

possible therapeutic eﬀects of compound 12, a disease activity index

(DAI), based on the methods reported by Park et al. [54], was applied.

The DAI score was calculated as the sum of the weight loss score, the

diarrheal score, and the hematochezia score (Table 1).

After 10 days of treatment, and an additional 24 h after the last dose

was administered, a disease activity index in all experimental groups

was estimated. The obtained results were summarized in Fig. 8. Com-

pound 12 signiﬁcantly decreased the index in all four treated groups.

The maximal eﬀect was observed in the PO-administered mice; how-

ever, this may be because of the higher dose (30 mg/kg versus 15 mg/

kg), which the PO-administered mice received compared to other mice

that received compound 12 by IV, IP, or SC. It is important to mention

that the additional autopsy examination of mouse colon samples re-

vealed that the colon length was shorter by approximately 2 cm in all

treated groups compared with the untreated animals. In addition, the

presence of gross blood in the stool or anus and the formation of a stool

that readily becomes paste on the anus were not detected in the treated

groups compared with the control mice. In general, colons treated by

compound 12 mice looked almost identical to those of naïve mice

(Fig. 9A and B). A histological score was based on the parameters re-

ported by Park et al.[54] and the score was signiﬁcantly lower in mice

treated with compound 12, as shown in Fig. 9C.

2.3. In vivo studies of leukocyte migration and inﬂammatory pathologies

Having identiﬁed an anti-inﬂammatory eﬀect in vivo, we tested

compound 12 in a series of toxicity studies. To increase the adminis-

tered dose and to avoid using dimethyl sulfoxide (DMSO) as a solvent, a

simple methylcellulose-based formulation was developed as part of this

study. Methylcellulose is an FDA-approved component for medical

formulations and in the concentrations used for this study (0.332%),

methylcellulose has been shown to be inert [52]. Eight experimental

groups with ﬁve female 6–7-week-old C57BL/6J mice were treated with

compound 12 and one control group with non-treated (naïve) mice was

used. Mice were administered a daily application of compound 12

suspension at 5, 10, and 15 mg/kg by intraperitoneal injection (IP) or at

10, 20, and 30 mg/kg by oral gavage (PO) for 7 days. Mice from all

experimental groups were observed for signs of toxicity immediately

post dosing and following 2 h post-administration daily. After 24 h of

the last administered dose, the mice were euthanized. Whole blood was

collected and analyzed for biochemical markers, white/red blood

counts, and white blood cell diﬀerentiation. Liver, spleen, lungs, heart,

and kidneys were collected from each mouse and ﬁxed with 4%

Having established the striking therapeutic eﬀect of compound 12

Table 1

Scoring criteria for the disease (IBD) activity index.

Score

Weight loss (%)

Stool consistency^b

Hematochezia^a

Absence

0

1

2

3

4

None

0–10

11–15

16–20

˃20

Normal

Loose stool^b

Diarrhea

Presence

a

b

The presence of gross blood in the stool or anus.

The formation of a stool that readily becomes a paste on the anus of a

mouse.

8

T. Getter, et al.

BioorganicChemistry92(2019)103250

Table 2

EAE clinical score.

Score

Clinical manifestation

1

2

3

4

5

Limp tail

Partial hind leg paralysis

Complete hind leg paralysis

Complete hind and partial front leg paralysis

Moribund

soon as they showed the ﬁrst signs of EAE (paralysis score 0.5–1). The

treatment group was administered compound 12 at 15 mg/kg dose

using a q3w regimen by SC injections. The mock-treated (vehicle)

control group was administered a volume of solvent similar to the

treated group. The body weight of animals was determined prior to the

experiment and the administered volume of compound 12 formulation

was adjusted according to the body weight. The EAE score (the severity

of the disease) was calculated based on criteria shown in Table 2. Be-

sides the clinical score (which mirrors the neurological symptoms), the

body weight of each mouse was measured as an additional parameter

for estimating the possible therapeutic eﬀect of compound 12. Fig. 10A

shows that starting already from the tenth day, the onset of the disease

in mice treated with compound 12 was delayed, and that the body

weight at day 20 of treatment returned to normal. This contrasted

signiﬁcantly with non-treated and control group mice where the loss in

body weight diﬀered by as much as 4 g approximately. These diﬀer-

ences closely correlated with the measured changes in clinical scores.

Starting from the 10th day, mice treated with compound 12 showed no

increase in the disease score and by day 20, the score even decreased to

score values of 2 to 1.5. In contrast, non-treated and control mice

reached a score above 2.5 in the same timeframe (Fig. 10B).

Fig. 8. Eﬀect of compound 12 on the DAI score in IBD/the Crohn’s disease

mouse model. The experiment was conducted as described in the Results sec-

tion. After 10 days of treatment, and an additional 24 h after the last adminis-

tered dose, a disease activity index in all experimental groups was estimated.

n = 3. Mean

SE, *P ≤ 0.05 compared to treatment by vehicle control, **

compared to non-treated control. ANOVA test.

in an acute IBD model, we also tested the eﬀect of compound 12 in a

chronic inﬂammatory multiple sclerosis model in mice. Female 7–8-

week-old C57BL/6J mice were immunized with an emulsion of myelin

oligodendrocyte glycoprotein (MOG) in complete Freund’s adjuant

(CFA) according to standard protocols, followed by administering per-

tussis toxin (PTX) in phosphate buﬀer solution (PBS) [55]. After the

onset of experimental autoimmune encephalomyelitis (EAE) after

9 days, mice were assigned by enrollment to groups, one at a time, as

Fig. 9. Eﬀect of compound 12 on the length and histology of colon in the Crohn’s disease mouse model. (A) The experiment was conducted as described in the Results

section. After 10 days of treatment, and an additional 24 h after the last administered dose, mice were euthanized; their colons were isolated, ﬁxed, and photo-

graphed. (B) Colon length measurement. (C) The intestinal lumen was investigated (mucosa, gross lesions, the architecture of the epithelial structure, the content of

inﬂammatory cells (mostly lymphocytes) within the lamina propria). Lymphoid follicles were investigated in caecum submucosae. n = 3. Mean

compared to those treated by vehicle control, ** compared to the non-treated control.

SE, *P ≤ 0.05

9

T. Getter, et al.

BioorganicChemistry92(2019)103250

(however, substantially diﬀerent ones from our reported compounds

are described here) in the nonalcoholic fatty liver disease mouse model

[56,57]. The authors proposed that their compounds would activate

peroxisome proliferator-activated receptor (PPAR-γ); however, they did

not prove this hypothesis at the molecular level and did not test the

eﬀect of the lead compound on leukocyte transmigration [58].

Our ﬁnal evaluated chronic inﬂammatory in vivo model investigated

the eﬀect of compound 12 on rheumatoid arthritis, an autoimmune

disorder that can be induced in mice by injection of collagen (collagen-

induced arthritis, CIA) [59]. To this end, male DBA/1 mice (7–8 weeks

of age) were injected with collagen type II (CII) + CFA emulsion on day

0 by SC. On day 21 the animals received the same amount of the

emulsion as a booster. On day 22, the mice were randomized, and

treatment was initiated. On days 0, 22, and 43, the paws of all mice

were measured using calipers to establish baseline and experimental

end-point values for paw thickness. The CIA scoring was conducted on a

scale of 0 to 16 (0 to 4 for each paw, adding the scores for all 4 paws).

For this study, compound 12 was administered IP (15 mg/kg, four times

per week) for 21 days. Fig. 12 shows that after two weeks, compound

12 signiﬁcantly reduced the clinical score in the treated mice. This

eﬀect was observed for the duration of the experiment. In addition,

compound 12 also improved local arthritis manifestations, in both fore

and hind limbs, where the thickness of both limbs was signiﬁcantly

reduced (Fig. 13A and B). Finally, all limb pathologies were assessed by

histology using hematoxylin and eosin (H&E) stains in paraﬃn-em-

bedded sections. Joints were scored for inﬂammation, cartilage da-

mage, pannus formation, and bone resorption on a scale of 0 to 3 for

each readout (Supporting Information Fig. S11). In Fig. 13C the in-

dividual score values represent the summarized results. Compound 12

decreased the total histological score by more than 50%. The body

weight of treated animals remained relatively constant during the ad-

ministration of the compound; thus, the decreased tissue thickness was

not related to weight loss and a perceived reduction of stress on the

associated joints (Supporting Information Fig. S12).

Fig. 10. Eﬀect of compound 12 on the mouse body weight and paralysis score.

The experiment was conducted as described in the Results section. During the

21 days of treatment, the body weight, (A), and the paralysis clinical score, (B),

were measured daily in all experimental groups. Data shown represent the

Mean

SEM, n = 10, *P < 0.0001 compared to those treated by vehicle

2.4. In silico study of PECAM-1 binding

control, ** compared to the non-treated control.

Since we did not see that compound 12 inhibits antibody binding to

MAdCAM, despite that the trioxotetrahydropyrimidin derivatives de-

veloped by Harriman et al. aﬀected this protein, we hypothesized that

compound 12 may have a diﬀerent mechanism of action. We conducted

in silico experiments to investigate and identify a possible molecular

target of compound 12. One of the vascular adhesion molecules that

appeared as such a target is the PECAM-1. The role of this receptor in

inﬂammation, leukocyte transendothelial migration, and cancer has

highlighted a number of opportunities in drug development. PECAM-1

has been shown to be involved in the late-stage progression in cancer,

highlighting a potential mechanism in disease stratiﬁcation, which is

dependent on the stage of treatment [60]. PECAM-1 is also known to

regulate T cell traﬃcking into the central nervous system (CNS) during

critical pathologies, demonstrating that targeting this receptor may

block multiple cell populations in both inﬂammation and cancer [61].

PECAM-1 not only resides at the junction; it is also a constituent of a

recycling compartment of endothelial cells, termed the lateral border

recycling compartment (LBRC), which modulates the transmigration of

leukocytes through the endothelial cell-cell junction [62]. The active

and inactive compounds were docked into the predicted binding sites of

the monomeric and dimeric structures of PECAM-1 in order to de-

termine which site can best distinguish between active and inactive

compounds measured by the area under the receiver operating char-

acteristic (ROC) - curve, which ranges between 0.5 for a separation not

better than random, and 1 for a perfect separation, (Fig. 14A). ROC

values calculated for sites on the monomeric structure were all below

0.6, thereby ruling out these sites (data not shown) [56]. On the other

hand, one of the sites identiﬁed on dimeric PECAM-1 (site 4) produces

an AUC (area under the curve)-ROC value of 0.73 (Fig. 14B) with lower

We then decided to investigate the eﬀect of compound 12 in chronic

inﬂammatory pathology, which is related to insulin resistance and the

metabolic syndrome: non-alcoholic fatty liver dystrophy (NAFLD),

which in the last decade has attracted much attention owing to a

forecast of increased frequency in the USA, Europe, and other in-

dustrialized countries. The classical in vivo model for NAFLD was con-

ducted in mice placed on a high-fat diet for a prolonged time period to

simulate a western diet (14 weeks prior to treatment initiation).

Compound 12 was administered on day 85 of the fat diet (15 mg/kg, for

SC and IP, 30 mg/kg for PO), three times per week for 65 days.

Compound 12 signiﬁcantly reduced the body weight of mice in all

administered roots (Supporting Information Fig. S10). After euthanasia

of the mice and a histopathological analysis of the livers, it was found

that compound 12 dramatically reduced the level of fat in the liver

almost to the normal histology (the SC administration results are shown

in Fig. 11A). In addition, even the amount of fat in the physiological

depot (mesenteric space) was signiﬁcantly reduced because of com-

pound 12 administration (Fig. 11B). The activity score of the disease in

the mice under SC administration was also similar to that of the naïve

mouse group (zero values in both categories, Fig. 11C). IP and PO-ad-

ministered compound 12 likewise signiﬁcantly reduced the disease

score. Compound 12 also signiﬁcantly decreased the liver weight after

treatment by all routes of administration mentioned above, as shown in

Fig. 11D. Finally, the mesenteric fat mass measurements were also

signiﬁcantly reduced by compound 12 treatment given PO, SC, and IV

(Fig. 11E).

It is important to mention that already in 2011 Liang Ma et al. re-

ported the signiﬁcant eﬀect of trioxotetrahydropyrimidin derivatives

10

T. Getter, et al.

BioorganicChemistry92(2019)103250

Fig. 11. Eﬀect of compound 12 on fat accumulation in the NAFLD mouse model. Male mice (3–4 weeks of age) were fed ad libitum a high-fat diet, no high-

carbohydrate diet (modiﬁed WD) with 42% kcal from fat containing 0.1% cholesterol (Harlan TD.88137) with normal water (without a high fructose-glucose

solution) for 14 weeks prior to initiating treatment. Control mice were fed with a standard chow diet (CD, Harlan TD.7012) with normal water (NW). Compound 12

was administered by IP injections at 15 mg/kg dose using a q3w regimen for two months. Liver (A) and mesenteric fat (B) histologies were assessed using H&E stains

in paraﬃn-embedded sections. (C) In addition to IP injection, compound 12 was also administered by SC (15 mg/kg) and PO (30 mg/kg). Steatohepatitis was deﬁned

by the presence of steatosis, inﬂammation, and hepatocellular ballooning, according to the FLIP algorithm. The severity of steatosis, lobular inﬂammation, and

hepatocellular ballooning was scored using the CRN criteria. The NAS was calculated by adding the scores of steatosis (the maximum score is 3), which is deﬁned by

more than 66% of hepatocytes containing fat droplets, with lobular inﬂammation (the maximum score is 3), which is deﬁned by > 4 foci/200 ﬁeld hepatocyte

ballooning (maximal score 2), which is deﬁned by many cells/prominent ballooning. The maximum NAS score combining all three features is 8. In all mice,

hepatocyte ballooning, as well as portal and perisinusoidal ﬁbrosis was not detected. (D) Liver weight. After 60 days of the treatment, livers were isolated and

weighed. (E) Mesenteric fat weight. After 60 days of treatment, mesenteric fat was isolated and weighed. Data shown represent the Mean

SEM, n = 8, n = 6,

**P < 0.002. ANOVA test. The double asterisk shows a signiﬁcant diﬀerence in the obtained results compared to the non-treated control and also to the vehicle-

treated group.

values obtained for all other sites (0.64, 0.52, 0.65, and 0.37 for sites 1,

2, 3, and 5, respectively). Thus, we hypothesize that site 4, located at

the interface between the two monomers of PECAM-1, can potentially

serve as the binding site for compound 12 (Fig. 14C and D). If indeed

compound 12 binds to PECAM-1 but only to its activated form (dimer),

such a diﬀerence may represent a very important advantage to clinics,

because it suggests that treatment with compound 12 might selectively

block PECAM-1 in the vascular region and not on platelets. This theo-

retical observation might also explain the lack of general toxicity of

compound 12 in mice [63]. It is important to mention that PECAM-1

knockout (KO) mice have proinﬂammatory phenotype results, for ex-

ample, in atherogenic diet-induced steatohepatitis [60]. However, also

a high level of PECAM-1 leads to increased transmigration of immune

cells to organs [64,65]. In our case, compound 12 prevented the de-

velopment of NAFLD in mice as we showed above. Therefore, other

molecular targets of compound 12 might be involved in its biological

eﬀect. The exact mechanism of action of compound 12 is currently

under investigation.

2.5. ¹⁹F NMR detection of ﬂuorinated compound 12 (compound 30) in

HeLa cell versus PC12 cell

Fluorine exhibits an excellent NMR trace ability and it is totally

absent in animal cells and ﬂuids [66,67]. Thus, ¹⁹F NMR is in intensive

use in drug discovery as a pharmacokinetic and pharmacodynamics tool

[68]. To further evaluate the possibility that compound 12 binds to

PECAM1, we introduced a ﬂuorine atom instead of hydrogen in the

ortho position in the second aromatic ring of compound 12 (Scheme 4).

We supposed that if the compound binds to PECAM1, we should detect

¹⁹F NMR signal only in the cells which express the protein. PECAM1 is

in use as a speciﬁc marker of endothelial cells in diﬀerent im-

munohistological technics and it is absent in neurons. Thus, we planned

to expose both types of cells to compound 30 and conduct ¹⁹F NMR

measurements to the obtained lysates. Due to technical limitations in-

stead of endothelial cells, we used epithelial cells (HeLa cell line) versus

neuronal PC12 cells. Such exchange was possible, because of the fact

that HeLa cells express PECAM1 selectively only on the plasma mem-

brane [69,70]. Indeed, in exposed to compound 30 HeLa cells, ¹⁹F was

detected (Supporting Information Fig. S13) and in PC12 (neuron-like

11

T. Getter, et al.

BioorganicChemistry92(2019)103250

developing a leukocyte transmigration inhibitor (compound 12). The

lead compound was active in IBD/Crohn’s disease, multiple sclerosis,

fatty liver disease, and rheumatoid arthritis mouse models at pharma-

cological doses. The clinical score was improved by compound 12 by

75%, 40%, 100% and 46% respectively. Compound 12 did not exhibit

any toxic eﬀects in mice at the administered doses. Based on in silico

calculations, we postulate that compound 12 blocks PECAM-1, its mono

or heterodimeric form found in vascular endothelial junctions. Such

types of molecules might therefore provide a unique possibility for

developing novel classes of leukocyte transmigration blocking agents

with broad therapeutic applications, especially for the treatment of

acute inﬂammation and chronic autoimmune diseases. The exact

binding partner of compound 12 now is under intensive investigation.

Beside the possible use of compound 12 as a drug candidate for the

developing of novel therapeutics against auto-immune diseases or in-

ﬂammation related disorders, compound 12 might also serve as a car-

rier for the delivery of anti-inﬂammatory, anti-thrombotic and anti-

oxidant agents to the pathologically altered vasculature [71,72]. Fur-

thermore, anchoring of some drug delivery systems to compound 12

might provide trans-endothelial delivery of drugs and genetic materials

[73,74].

Fig. 12. Eﬀect of compound 12 on the CIA score in the arthritis mouse model.

Male DBA/1 mice (7–8 weeks of age) were injected with CII + CFA emulsion on

day 0 (SC) at the base of the tail with a volume of 0.05 mL/animal. On day 21

the animals received the same amount of the emulsion as a booster. On day, 22

mice were randomized and the treatment was initiated (SC injections with a

15 mg/kg dose using the q3w regimen. On days 0, 22, and 43, the paws of all

mice were measured using calipers to establish baseline and end point values

for paw thickness. The CIA scoring was on a scale of 0 to 16 (0 to 4 for each

4. Experimental section

paw, adding the scores for all 4 paws). Data shown represent the Mean

SEM,

4.1. Materials

n = 8, n = 7 (not treated). *P < 0.002. ANOVA test.

All chemical reagents, solvents, and acids were purchased from

Acros Organic, Alfa Aesar, Bio-Lab Ltd., and Merck, and were all used as

received. Anhydrous tetrahydrofuran (THF) was obtained by distilla-

tion from a boiled blue colored mix containing sodium (1% w/v) and

benzophenone (0.2% w/v). Anhydrous dimethylformamide (DMF) was

obtained by distillation under vacuum and stored over 4 A° molecular

sieves. Column chromatography was performed on silica gel 60

(230–400 mesh; Merck). Analytical and preparative high-performance

liquid chromatography (HPLC) (Young Lin Instruments, Anyang, Korea)

were performed on LUNA C18 preparative (10 µm, 100 × 30 mm) or

analytical (5 µm, 250 × 4.6 mm) columns, both from Phenomenex, Inc.

(Torrance, CA, USA). HPLC puriﬁcation was carried out with an in-

creasing linear gradient of CH₃CN in H₂O. The purity of all synthesized

compounds was conﬁrmed by elemental analysis and for compounds 29

and 30 by analytical HPLC. Analytical thin-layer chromatography (TLC)

was carried out on pre-coated silica gel 60 F254 (Merck) sheets using

UV absorption and iodine physical adsorption for visualization. Mass

spectra were recorded on a Finnigan Model 400 instrument using a

QToF microspectrometer (Micromass, Milford, MA, USA), using elec-

trospray ionization (ESI) in the positive ion mode. 6200 Series

Accurate-Mass Time-of-Flight (TOF) LC/MS from Agilent Technologies

(Santa Clara, CA, USA) was used for PK analysis. Data were processed

cells), ¹⁹F was not detected at all. Those results might indirectly support

our hypothesis that compound 12 binds to PECAM1.

2.6. PK study

Preliminary PK study was conducted in C57 black mice. Compound

12 was administrated PO by a single dose (gavage, 50 mg/kg) in the

minimal amount of DMSO (70 µL). The level of the compound was

measured by LC-MS after the extraction from the plasma by chloroform

(due to high lipophilicity of the molecule). The compound reaches the

peak concertation in the blood after 2.5 h (around 15 µM) and after 9 h

the level of the compound was undetectable (Fig. 14S). Suggesting that

compound 12 might be the main pharmacological active specie, al-

though the broad PK study with identiﬁcation of possible active me-

tabolites are still needed.

3. Conclusions

Modiﬁed trioxotetrahydropyrimidin scaﬀolds were used for

Fig. 13. Eﬀect of compound 12 in the

Collagen-Induced Arthritis (CIA) model in

DBA/1 mice (limb thickness and histolo-

gical score). (A) Fore limb thickness, (B)

Hind limb thickness, (C) Histological score.

Data shown represent the mean

SEM,

n = 8. *P ≤ 0.05 compared to those treated

by vehicle control, ^ compared to the non-

treated control.

12

T. Getter, et al.

BioorganicChemistry92(2019)103250

Fig. 14. Binding sites and the binding mode

of compound 12 within the PECAM-1 dimer

structure. (A) Predicted binding sites within

the PECAM-1 dimer structure. Site 1: light

blue, site 2: green, site 3: blue, site 4: red,

and site 5: purple. The best ROC curve

(calculated for

7 active and 13 inactive

compounds) was obtained for site 4. (B)

ROC curve (AUC of 0.73) obtained for site 4.

The black 45° line represents random

ranking. (C) Representation of compound

12 at binding site 4. D) CPK representation

of compound 12 at binding site 4.

using mass L-ynX ver. 4.1 calculation and de-convolution software

(Waters Corp., Milford, MA, USA). High-resolution mass spectra

(HRMS) were obtained using an LTQ Orbitrap XL (Thermo Scientiﬁc,

Waltham, MA, USA). Melting points were measured with a Fisher-Johns

melting point apparatus (Waltham, MA, USA). The ¹H NMR and ¹³C

NMR were recorded at room temperature on a Bruker Advance NMR

spectrometer (Vernon Hills, IL) operating at 300, 400, 600, and

700 MHz and were in accordance with the assigned structures.

Chemical shift values were reported relative to tetramethylsilane

(TMS), which was used as an internal standard. Chemical shifts were

expressed in δ (ppm) and coupling constants (J) in hertz. The splitting

pattern abbreviations are as follows: s, singlet; d, doublet; t, triplet; q,

quartet; quint, quintet; m, unresolved multiplet due to the ﬁeld strength

of the instrument; dd, doublet of doublet. Dulbecco’s modiﬁed Eagle’s

medium (DMEM), Roswell park memorial institute (RPMI) 1640

medium, fetal calf serum (FCS),

4.2. Synthetic procedures

4.2.1. General procedure for the synthesis of methylated benzoates (5, 15

and 28) [41]

To

a

solution of 4-bromomethylbenzoic acid, 3-chlor-

omethylbenzoic acid or 4-(bromomethyl)-2-ﬂuorobenzoic acid

(2.30 mmol) in MeOH (5.6 mL), concentrated sulfuric acid (0.14 mL)

was added. The resulting mixture was reﬂuxed for 5 h. Then, the mix-

ture was cooled to room temperature and evaporated in vacuo. H₂O

(20 mL) was added to the reaction mix in an ice-water bath, and the

resulting solid was ﬁltered and washed with cold water. The solid

material was partitioned between Et₂O/AcOEt 1:1 and Na₂CO₃. The

organic layer was dried over Na₂SO₄and evaporated in vacuo to aﬀord

the products.

and trypsin were purchased from ^LB^-gio^lulo^tag^micⁱaⁿl^eI^,nd^pu^enstⁱr^ciⁱe^llsⁱⁿ(^/B^se^tt^rh^e-^pH^toa^mem^yceⁱkⁿ,

Israel). Heparin, hydrocortisone, DMSO, PBS tablets, Medium 199,

radioimmunoprecipitation (RIPA) buﬀer, methylcellulose and Type II

collagen were purchased from Merck-Sigma–Aldrich Chemicals

(Rehovot, Israel). TNF-α, IFNγ, and CXCL-12 and -13 were purchased

from Peprotech, Rehovot, Israel. Endothelial cell growth supplement

was from Upstate Biotechnology (Boston, MA). T-, B-, monocyte se-

lection kits were from Miltenyi Biotec (Bergisch Gladbach, Germany).

Kits for determining the blocking activity of compound 12 of JAM-C

were purchased from Ray-Biotech (Norcross, GA, USA) of VCAM-1 from

Abcam (Cambridge, UK), Beta Integrin-Mediated Cell Adhesion Array

Kit was purchased from Chemicon (Temecula, CA, USA) and MAdCAM-

1 from ELAab (Guangguguoji, Wuhan, China). The bicinchoninic acid-

based protein concentration determination kit was purchased from

Pierce (Waltham, MA). Santa Cruz Biotechnology supplied complete

Freund's adjuvant.

4.2.2. Methyl 4-(bromomethyl)benzoate, (5)

The title compound was obtained starting from 3-chlor-

omethylbenzoic acid (white powder, yield 48%): ¹H NMR (400 MHz,

CDCl3): δ 7.96 (d, J = 8.41 Hz, 2H), 7.40 (d, J = 8.41 Hz, 2H), 4.45 (s,

2H), 3.87 (s, 3H) ppm. ¹³C NMR (100.6 MHz, CDCl3): δ 166.28, 142.59,

129.95, 128.98, 52.10, 32.27 ppm. MS (ESI, SQ), m/z (%): (229 ⁷⁹Br,

231 ⁸¹Br) [M+H]⁺

.

4.2.3. Methyl 3-(chloromethyl)benzoate (15)

The title compound was obtained starting from 3-chlor-

omethylbenzoic acid (colorless syrup, yield 75%): ¹H NMR (400 MHz,

CDCl3): δ 8.06 (s, 1H), 7.99 (d, J = 7.76 Hz, 1H), 7.58 (d, J = 7.76 Hz,

1H) 7.429 (t, J = 7.76 Hz, 1H), 4.61(s, 2H), 3.92 (s, 3H) ppm. ¹³C NMR

(100.6 MHz, CDCl3):

δ 166.54, 137.91, 132.99, 130.73, 129.67,

129.55, 128.28, 52.23, 45.52 ppm. GC-MS (EI),70 ev, m/z (%): 184

[M]⁺

.

4.2.4. General procedure for the synthesis of 6, 7, 8, 9, 16, 17, 18, and 25

[42]

A suspension of 5, 15 or 28 (6.55 mmol) and the corresponding

benzaldehydes

(6.55 mmol)

including

potassium

carbonate

13

T. Getter, et al.

BioorganicChemistry92(2019)103250

(13.10 mmol), and potassium iodide (7.86 mmol) was heated at 70 °C

for 5 h. The solvent was evaporated, the residue was dissolved in ethyl

acetate (50 mL), washed with saturated aqueous NaHCO₃solution

(2 × 20 mL), water (2 × 20 mL), and brine (2 × 10 mL), dried over

Na₂SO₄, and concentrated under reduced pressure. The crude product

was puriﬁed by ﬂash column chromatography using ethyl acetate/

petroleum ether as an eluent.

4.2.11. Methyl 3-((3-formylphenoxy)methyl)benzoate (25)

(white powder, yield 41%) ¹H NMR (300 MHz, CDCl3): δ 9.95 (s,

1H), 8.12 (s, 1H), 8.01 (d, J = 7.12 Hz, 1H), 7.63(d, J = 7.12 Hz, 1H),

7.53–7.40 (m, 4H), 7.31–7.17 (m, 1H), 5.13 (s, 2H), 3.91 (s, 3H) ppm.

¹³C NMR (75.49 MHz, CDCl3) δ 192.02, 166.76, 159.06, 137.83,

136.81, 131.86, 130.57, 130.19, 129.33, 128.77, 128.54, 123.86,

122.06, 113.25, 69.55, 52.21 ppm. MS (ESI, SQ), m/z (%): 271.2 [M

+H]⁺, 293 [M+Na]⁺

.

4.2.5. Methyl 4-((2-formylphenoxy)methyl)benzoate (6)

(white powder, yield 60%) ¹H NMR (300 MHz, CDCl3): δ 10.57 (s,

1H), 8.07 (d, J = 8.43 Hz, 2H), 7.86 (d, J = 7.50 Hz, 1H), 7.53 (d,

J = 8.43 Hz, 2H), 7.76 (t, J = 7.50 Hz, 1H), 7.06 (t, J = 7.50 Hz, 1H),

7.01 (d, J = 8.44 Hz, 1H), 5.25 (s, 2H), 3.92 (s, 3H) ppm. ¹³C NMR

4.2.12. Methyl 3-((4-formylphenoxy)methyl)benzoate (18)

The title compound was obtained starting from 28 and 30 (white

powder, yield 51%) ¹H NMR (300 MHz, CDCl3): δ 9.83 (s, 1H), 8.09 (s,

1H), 7.98 (d, J = 7.56 Hz, 1H), 7.79 (d, J = 8.32 Hz, 2H), 7.60 (d,

J = 7.56 Hz, 1H), 7.43 (t, J = 7.56 Hz, 1H), 7.04 (d, J = 8.32 Hz, 2H),

5.10 (s, 2H), 3.87 (s, 3H) ppm. ¹³C NMR (75.49 MHz, CDCl3) δ 190.58,

166.50, 163.34, 136.53, 131.89, 131.81, 130.54, 130.23, 129.32,

128.76, 128.45, 115.05, 69.46, 52.10 ppm. MS (ESI, SQ), m/z

(75.49 MHz, CDCl3):

δ 189.48, 166.68, 160.64, 141.18, 135.94,

130.03, 128.73, 126.85, 125.19, 121.31, 112.90, 69.76, 52.20 ppm. MS

(ESI, SQ), m/z (%):293.1 [M+Na]⁺

.

(%):271.2 [M+H]⁺, 293 [M+Na]⁺

.

4.2.6. Methyl 4-((3-formylphenoxy)methyl)benzoate (7)

(yellow powder, yield 54%) ¹H NMR (300 MHz, CDCl3): δ 9.96 (s,

1H), 8.06 (d, J = 8.27 Hz, 2H), 7.54–7.44 (m, 5H), 7.29–7.22 (m, 1H),

5.17 (s, 2H), 3.91 (s, 3H) ppm. ¹³C NMR (75.49 MHz, CDCl3) δ 191.95,

166.74, 159.01, 141.48, 137.87, 130.23, 129.96, 129.89, 127.02,

124.00, 122.11, 113.15, 69.49, 52.16 ppm. MS (ESI, SQ), m/z (%):

4.2.13. Methyl 4-((2-(tert-butyl)-6-formylphenoxy)methyl)-2-ﬂuorobenzoate

(29)

(white powder, yield 50%, m.p. 70 °C) ¹H NMR (400 MHz, CDCl3): δ

10.26 (s, 1H),

8 (t, J = 7.77 Hz, 1H), 7.74 (dd, J = 7.63 Hz,

J = 1.75 Hz, 1H), 7.65 (dd, J = 7.87 Hz, J = 1.75 Hz, 1H), 7.36 (d,

J = 11.33 Hz, 1H), 7.32 (d, J = 8 Hz, 1H), 7.23 (t, J = 7.6 Hz, 1H), 5.08

(s, 2H), 3.95 (s, 3H), 1.43 (s, 9H) ppm. ¹³C NMR (100.6 MHz, CDCl3): δ

189.98, 164.61 (d, J = 3.53 Hz), 162.13 (d, J = 261.3 Hz), 160.67,

144.05 (d, J = 8.61 Hz), 143.93, 133.8, 132.53, 129.91, 128.94,

124.54, 121.72 (d, J = 3.49 Hz), 118.06 (d, J = 10.17 Hz), 114.99 (d,

J = 23.88 Hz), 78.06, 52.37, 35.27, 30.85 ppm. ¹⁹F (376 MHz, CDCl₃):

δ −109.02 (dd, J = 11.33 Hz, J = 7.4 Hz) ppm. MS (APCI) found m/z

345.1497 (calcd for C₂₀H₂₁FO₄: 345.1497 [M+Na]⁺).

271.1 [M+H]⁺, 293.2 [M+Na]⁺

.

4.2.7. Methyl 4-((2- (tert-butyl)-6- formylphenoxy) methyl)benzoate (8)

(oﬀ white powder, yield 43%, m.p. 84 °C) ¹H NMR (300 MHz,

CDCl3):

δ 10.30 (s, 1H), 8.11 (d, J = 8.04 Hz, 2H), 7.75 (dd,

J = 7.80 Hz, J = 1.65 Hz, 1H), 7.64 (dd, J = 7.81 Hz, J = 1.62 Hz, 1H),

7.58 (d, J = 8.04 Hz, 2H), 7.21 (t, J = 7.81 Hz, 1H), 5.12 (s, 2H), 3.94

(s, 3H), 1.44 (s, 9H) ppm. ¹³C NMR (75.49 MHz, CDCl3): δ 190.16,

166.78, 161.26, 143.93, 141.53, 133.72, 129.99, 129.91, 128.36,

126.53, 124.37, 79.35, 52.18, 35.26, 30.86 ppm. MS (ESI) found m/z

349.14102 (calcd for C₂₀H₂₂O₄: 349.14103 [M+Na]⁺). Anal. Calcd for

4.2.14. General procedure for the coupling of trioxotetrahydropyrimidin

and the corresponding aldehydes to obtain: 1, 2, 3, 4, 10, 11, 12, 13, 19,

20, 21, 26 and 30 [57]

C

₂₀H₂₂O₄: C, 73.60; H, 6.79. Found: C, 74.32; H, 7.03.

The corresponding aldehydes (3.5 mmol), ethanol (10 mL), distilled

water (10 mL), and barbituric acid (3 mmol) were reﬂuxed overnight.

The formed solids were collected by sucking ﬁltration and washed with

boiling water (3x15 mL), ethanol (3x15 mL), and ether (3x15 mL). The

solids obtained were dried in vacuum.

4.2.8. Methyl 4-((4-formyl-2,6-dimethylphenoxy)methyl)benzoate (9)

(yellow powder, yield 94%, m.p. 82 °C) ¹H NMR (400 MHz, CDCl3):

δ 9.87 (s, 1H), 8.07 (d, J = 8.4 Hz, 2H), 7.57 (s, 2H), 7.54 (d,

J = 8.4 Hz, 2H), 4.92 (s, 2H), 3.92 (s, 3H), 2.33 (s, 6H) ppm. ¹³C NMR

(75.49 MHz, CDCl3):

δ 191.52, 166.69, 160.86, 141.99, 132.52,

4.2.15. 8H-benzo[7,8]chromeno[2,3-d]pyrimidine-8,10(9H)-dione (1)

(orange powder, yield 46%, m.p. > 300 °C) ¹H NMR (600 MHz,

DMSO‑d₆): δ 11.48 (s, 1H), 9.57 (s, 1H), 8.85 (d, J = 8.28 Hz, 1H), 8.51

(d, J = 9.45 Hz, 1H), 8.17 (d, J = 8.28 Hz, 1H), 7.86 (d, J = 9.45 Hz,

1H), 7.85 (t, J = 8.28 Hz, 1H), 7.74 (t, J = 8.28 Hz, 1H) ppm. ¹³C NMR

(150.9 MHz, DMSO‑d₆) δ 167.37, 161.20, 156.28, 154.27, 140.08,

138.20, 130.19, 129.59, 129.16, 129.04, 127.21, 122.98, 116.76,

114.51, 112.98 ppm. MS (ESI) found m/z 265.06097 (calcd for

132.07, 130.75, 129.66, 127.24, 73.25, 52.13, 16.46 ppm. MS (ESI)

found m/z 299.12793 (calcd for C₁₈H₁₈O₄: 299.12779 [M+H]⁺). Anal.

Calcd for C₁₈H₁₈O₄: C, 72.47; H, 6.08. Found: C, 71.92; H, 5.87.

4.2.9. Methyl 3-((2-formyl-6-methylphenoxy)methyl)benzoate (16)

(oﬀ white powder, yield 61%, m.p 45 °C) ¹H NMR (300 MHz,

CDCl3): δ 10.29 (s, 1H), 8.11 (s, 1H), 8.05 (d, J = 7.75 Hz, 1H), 7.70 (d,

J = 7.66 Hz, 1H), 7.65 (d, J = 7.66 Hz, 1H), 7.49 (t, J = 7.66 Hz, 1H),

7.48 (d, J = 7.66 Hz, 1H), 7.18 (t, J = 7.66 Hz, 1H), 5.01 (s, 2H) ppm.

¹³C NMR (75.49 MHz, CDCl3): δ 190.12, 166.73, 159.87, 137.71,

136.67, 132.43, 130.65, 129.70, 129.49, 128.86, 126.92, 124.72,

52.25, 15.95 ppm. MS (ESI) found m/z 307.09402 (calcd for C₁₇H₁₆O₄:

307.09408 [M+Na]⁺). Anal. Calcd for C₁₇H₁₆O₄: C, 71.82; H, 5.67.

Found: C, 71.78; H, 5.88.

C

₁₅H₈N₂O₃: 265.06077 [M+H]⁺). Anal. Calcd for C₁₅H₈N₂O₃: C,

68.18; H, 3.05; N, 10.60. Found: C, 68.18; H, 3.12; N, 11.01.

4.2.16. 5-(3-acetylbenzylidene)pyrimidine-2,4,6(1H,3H,5H)-trione (2)

(yellow powder, yield 20%, m.p 231 °C) ¹HNMR (300 MHz,

DMSO‑d₆): δ 11.45 (s, 1H), 11.29 (s, 1H), 8.63 (s, 1H), 8.35 (s, 1H),

8.21 (s, 1H), 8.06 (d, J = 7.45 Hz, 1H), 7.61 (t, J = 7.45 Hz, 1H), 2.61

(s, 3H) ppm. ¹³CNMR (75.49 MHz, DMSO‑d₆) δ 197.45, 163.05, 161.45,

153.38, 150.12, 136.74, 136.24, 133.11, 132.27, 130.80, 128.29,

120.13, 26.67 ppm. MS (ESI) found m/z 259.07166 (calcd for

4.2.10. Methyl 3-((2-formylphenoxy)methyl)benzoate (17)

(white powder, yield 44%) ¹H NMR (300 MHz, CDCl3): δ 10.52 (s,

1H), 8.10 (s, 1H), 7.98 (d, J = 7.63 Hz, 1H), 7.81(d, J = 7.63 Hz, 1H),

7.63 (d, J = 7.63 Hz, 1H), 7.55–7.38 (m, 2H), 7.08–6.95 (m, 2H), 5.16

(s, 2H), 3.89 (s, 3H) ppm. ¹³C NMR (75.49 MHz, CDCl3): δ 189.00,

166.27, 160.43, 136.37, 135.68, 131.43, 130.32, 129.08, 128.57,

128.15, 128.06, 124.81, 120.86, 112.71, 69.48, 51.87 ppm. MS (ESI,

C

₁₃H₁₀N₂O₄: 259.07133 [M+H]⁺). Anal. Calcd for C₁₃H₁₀N₂O₄: C,

60.47; H, 3.90; N, 10.85. Found: C, 60.54; H, 3.86; N, 10.83.

4.2.17. 5-(3,4-dihydroxybenzylidene)pyrimidine-2,4,6(1H,3H,5H)-trione

(3)

SQ), m/z (%): 271.2 [M+H]⁺, 293.1 [M+Na]⁺

.

(orange powder, yield 46%, m.p > 300 °C) ¹H NMR (600 MHz,

14

T. Getter, et al.

BioorganicChemistry92(2019)103250

DMSO‑d₆): δ 11.21 (s, 1H), 11.09 (s, 1H), 10.39 (bs, 1H), 9.46 (bs, 1H),

8.195 (d, J = 2.09 Hz, 1H), 8.121 (s, 1H), 7.63 (dd, J = 8.51 Hz,

J = 2.09 Hz, 1H), 6.86 (d, J = 8.51 Hz, 1H) ppm. ¹³C NMR (150.9 MHz,

DMSO‑d₆) δ 164.09, 162.14, 155.90, 152.25, 150.13, 144.74, 131.18,

124.16, 121.22, 115.26, 113.57 ppm. MS (ESI) found m/z 249.05086

(calcd for C₁₁H₈N₂O₅: 249.05060 [M+H]⁺). Anal. Calcd for

409.13941 [M+H]⁺). Anal. Calcd for C₂₂H₂₀N₂O₆: C, 64.70; H, 4.94;

N, 6.86. Found: C, 65.03; H, 4.76; N, 6.77.

4.2.23. Methyl 3-((2-methyl-6-((2,4,6-trioxotetrahydropyrimidin-5(2H)-

ylidene) methyl) phenoxy) methyl) benzoate (19)

(yellow powder, yield 46%, m.p. 198 °C) H¹-NMR (600 MHz,

DMSO‑d₆): δ 11.31 (s, 1H), 11.12 (s, 1H), 8.33 (s, 1H), 7.97 (s, 1H),

7.91 (d, J = 7.81 Hz, 1H), 7.76 (d, J = 7.81 Hz, 1H), 7.65 (d,

J = 7.51 Hz, 1H), 7.49 (t, J = 7.81 Hz, 1H), 7.40 (d, J = 7.51 Hz, 1H),

7.71 (t, J = 7.51 Hz, 1H) ppm. ¹³C NMR (150.9 MHz, DMSO‑d₆): δ

165.92, 162.98, 161.05, 156.62, 150.35, 150.21, 137.22, 134.71,

133.42, 130.73, 129.92, 129.84, 129.93, 128.90, 127.35, 123.25,

119.47, 75.37, 52.18, 15.85 ppm. MS (ESI) found m/z 395.12393 (calcd

for C₂₁H₁₈N₂O₆: 395.12376 [M+H]⁺). Anal. Calcd for C₂₁H₁₈N₂O₆: C,

63.96; H, 4.60; N, 7.10. Found: C, 63.45; H, 4.55; N, 7.01.

C

₁₁H₈N₂O₅: C, 53.23; H, 3.25; N, 11.29. Found: C, 53.45; H, 3.21; N,

11.81.

4.2.18. 5-(anthracen-9-ylmethylene)pyrimidine-2,4,6(1H,3H,5H)-trione

(4)

(red powder, yield 54%, m.p > 300 °C) ¹H NMR (400 MHz,

DMSO‑d₆): δ 11.57 (s, 1H), 11.17 (s, 1H), 9.01 (s, 1H), 8.63 (s, 1H),

8.12 (s, 2H), 7.97 (s, 2H), 7.52 (s, 4H) ppm. ¹³C NMR (75.46 MHz,

DMSO‑d₆) δ 162.28, 160.41, 151.20, 150.42, 130.50, 129.38, 128.57,

127.69, 127.55, 126.12, 125.42 ppm. MS (ESI) found m/z 317.009259

(calcd for

C

₁₉H₁₂N₂O₃: 317.09207 [M+H]⁺). Anal. Calcd for

4.2.24. Methyl 3- ((2- ((2,4,6 -trioxotetrahydropyrimidin- 5(2H)-ylidene)

methyl) phenoxy) methyl) benzoate (20)

C₁₉H₁₂N₂O₃: C, 72.15; H, 3.82; N, 8.86. Found: C, 72.12; H, 3.79; N,

8.72.

(orange powder, yield 56%, m.p. 200 °C) ¹H NMR (300 MHz,

DMSO‑d₆): δ 11.46 (s, 1H), 11.27 (s, 1H), 8.68 (s, 1H), 8.17 (s, 1H),

8.10 (d, J = 8.21 Hz, 1H), 8.04 (d, J = 7.62 Hz, 1H), 7.85 (d,

J = 7.33 Hz, 1H), 7.68 (t, J = 7.59 Hz, 1H), 7.61 (t, J = 7.87 Hz, 1H),

7.31 (d, J = 8.50 Hz, 1H), 7.116 (t, J = 7.62 Hz, 1H) ppm. ¹³C NMR

(75.49 MHz, DMSO‑d₆): δ 165.90, 163.18, 161.30, 157.66, 150.15,

149.72, 137.38, 133.70, 132.38, 132.04, 129.81, 128.96, 128.59,

127.88, 122.08, 119.87, 118.92, 112.41, 69.20, 52.12 ppm. MS (ESI)

found m/z 381.10834 (calcd for C₂₀H₁₆N₂O₆: 381.10811 [M+H]⁺).

Anal. Calcd for C₂₀H₁₆N₂O₆: C, 63.16; H, 4.24; N, 7.37. Found: C,

63.16; H, 4.32; N, 7.12.

4.2.19. Methyl 4-((2-((2,4,6-trioxotetrahydropyrimidin- 5(2H)-ylidene)

methyl) phenoxy)methyl) benzoate (10)

(yellow powder, yield 64%) ¹H NMR (300 MHz, DMSO‑d₆): δ 11.35

(s, 1H), 11.17 (s, 1H), 8.58 (s, 1H), 8.06–7.93 (m, 3H), 7.60 (d,

J = 8.28 Hz, 2H), 7.50 (t, J = 7.29 Hz, 1H), 7.16 (d, J = 8.28 Hz, 1H),

7.01 (t, J = 7.62 Hz, 1H), 5.34 (s, 2H), 3.86 (s, 3H) ppm. ¹³C NMR

(150.9 MHz, DMSO‑d₆): δ 165.87, 163.22, 161.32, 157.64, 150.14,

149.65, 142.05, 133.71, 132.40, 129.28, 129.06, 127.32, 122.07,

119.93, 118.94, 112.39, 69.25, 52.06 ppm. MS (ESI, SQ), m/z

(%):381.3 [M+H]⁺

.

4.2.25. Methyl

3-((4-((2,4,6-trioxotetrahydropyrimidin-5(2H)-ylidene)

4.2.20. Methyl 4-((3-((2,4,6- trioxotetrahydropyrimidin- 5(2H)-ylidene)

methyl) phenoxy) methyl) benzoate (11)

methyl) phenoxy) methyl) benzoate (21)

The title compound was obtained starting from 34 and 5 (yellow

powder, yield 71%, m.p. 227 °C) 1H NMR (300 MHz, DMSO‑d₆): δ

11.33 (s, 1H), 11.20 (s, 1H), 8.37 (d, J = 8.34 Hz, 2H), 8.26 (s, 1H),

7.95 (d, J = 7.47 Hz, 1H), 7.77 (d, J = 7.60 Hz, 1H), 7.58 (t,

J = 7.60 Hz, 1H), 7.16 (d, J = 8.34 Hz, 2H), 5.34 (s, 2H), 3.87 (s, 3H)

ppm. ¹³C NMR (75.45 MHz, DMSO‑d₆): δ 165.93, 163.798, 162.18,

162.06, 154.74, 150.12, 137.31, 137.14, 132.46, 129.83, 128.99,

128.72, 128.22, 125.40, 115.71, 114.60, 68.88, 52.14 ppm. MS (ESI)

found m/z 381.10818 (calcd for C20H16N2O6: 381.10811 [M+H]⁺).

Anal. Calcd for C20H16N2O6: C, 63.16; H, 4.24; N, 7.37. Found: C,

62.39; H, 4.63; N, 7.24.

(yellow powder, yield 57%, m.p. 256 °C) ¹H NMR (300 MHz,

DMSO‑d₆): δ 11.41 (s, 1H), 11.26 (s, 1H), 8.25 (s, 1H), 7.99 (d,

J = 7.84 Hz, 2H), 7.91 (s, 1H), 7.71–7.57 (m, 3H), 7.405 (t,

J = 8.10 Hz, 1H), 7.22 (d, J = 8.10 Hz, 1H), 5.25 (s, 2H), 3.86 (s, 3H)

ppm. ¹³C NMR (75.49 MHz, DMSO‑d₆): δ 165.93, 163.27, 161.49,

157.43, 154.19, 150.07, 142.26, 133.84, 129.27, 129.12, 128.95,

127.47, 126.29, 119.32, 118.93, 118.6468.62, 52.06 ppm. MS (ESI)

found m/z 403.09000 (calcd for C₂₀H₁₆N₂O₆: 403.09006 [M+Na]⁺).

Anal. Calcd for C₂₀H₁₆N₂O₆: C, 63.16; H, 4.24; N, 7.37. Found: C,

63.23; H, 4.35; N, 7.56.

4.2.21. Methyl 4-((2-(tert-butyl)-6-((2,4,6- trioxotetrahydropyrimidin-

5(2H)- ylidene) methyl) phenoxy) methyl) benzoate (12)

4.2.26. Methyl 3-((3-((2,4,6-trioxotetrahydropyrimidin- 5(2H)- ylidene)

methyl) phenoxy) methyl) benzoate (26)

(yellow powder, yield 14%, m.p. 184 °C) ¹H NMR (300 MHz,

DMSO‑d₆): δ 11.33 (s, 1H), 11.19 (s, 1H), 8.34 (s, 1H), 7.99 (d,

J = 8.08 Hz, 2H), 7.70 (d, J = 7.85 Hz, 1H), 7.61 (d, J = 8.08 Hz, 2H),

7.48 (d, J = 7.85 Hz, 1H), 7.12 (t, J = 7.85 Hz, 1H), 4.94 (s, 2H), 3.94

(s, 3H), 1.38 (s, 9H) ppm. ¹³C NMR (100.6 MHz, DMSO‑d₆): δ 165.88,

162.95, 161.14, 157.91, 150.35, 150.20, 142.04, 141.62, 130.30,

130.27, 129.21, 129.05, 127.75, 127.37, 122.78, 119.60, 76.96, 52.07,

34.64, 30.50 ppm. MS (ESI) found m/z 437.17087 (calcd for

The title compound was obtained starting from 33 and 5 (yellow

powder, yield 17%, m.p. 158 °C) ¹H NMR (300 MHz, DMSO‑d₆): δ 11.40

(s, 1H), 11.25 (s, 1H), 8.25 (s, 1H), 8.06 (s, 1H), 7.93–7.83 (m, 2H),

7.75 (d, J = 7.54 Hz, 1H), 7.66 (d, J = 7.95 Hz, 1H), 7.57 (t,

J = 7.54 Hz, 1H), 7.41 (t, J = 7.95 Hz, 1H), 7.22 (d, J = 8.37 Hz, 1H),

5.24 (s, 2H), 3.87 (s, 3H) ppm. ¹³C NMR (150.9 MHz, DMSO‑d₆): δ

165.96, 163.26, 161.48, 157.49, 154.24, 150.05, 137.57, 133.85,

132.32, 129.79, 129.11, 128.94, 128.55, 128.06, 126.19, 119.32,

118.89, 118.73, 68.67, 52.11 ppm. MS (ESI) found m/z 381.10840

C

₂₄H₂₄N₂O₆: 437.17071 [M+H]⁺). Anal. Calcd for C₂₄H₂₄N₂O₆: C,

66.05; H, 5.54; N, 6.42. Found: C, 66.12; H, 5.53; N, 6.81.

(calcd for

C

₂₀H₁₆N₂O₆: 381.10811 [M+H]⁺). Anal. Calcd for

C

₂₀H₁₆N₂O₆: C, 63.16; H, 4.24; N, 7.37. Found: C, 63.19; H, 4.29; N,

4.2.22. Methyl 4-((2,6- dimethyl-4-((2,4,6- trioxotetrahydropyrimidin

-5(2H)- ylidene) methyl) phenoxy) methyl) benzoate (13)

7.14.

(yellow powder, yield 73%, m.p. 281 °C) ¹H NMR (400 MHz,

DMSO‑d₆): δ 11.35 (s, 1H), 11.20 (s, 1H), 8.19 (s, 1H), 8.02 (d,

J = 8.26 Hz, 2H), 7.99 (s, 2H), 7.66 (d, J = 8.26 Hz, 2H), 5.00 (s, 2H),

3.87 (s, 3H), 2.97 (s, 6H) ppm. ¹³C NMR (100.6 MHz, DMSO‑d₆): δ

165.94, 163.54, 161.69, 159.16, 154.52, 150.11, 142.49, 135.17,

130.44, 129.24, 129.09, 128.26, 127.80, 117.32, 72.68, 52.07,

16.14 ppm. MS (ESI) found m/z 409.13950 (calcd for C₂₂H₂₀N₂O₆:

4.2.27. Methyl 4-((2- (tert-butyl)-6-((2,4,6- trioxotetrahydropyrimidin-

5(2H)- ylidene) methyl) phenoxy) methyl) −2-ﬂuorobenzoate (30)

(yellow powder, yield 39%, m.p. 210 °C) ¹H NMR (400 MHz,

DMSO‑d₆): δ 11.31 (s, 1H), 11.18 (s, 1H), 8.31 (s, 1H), 7.93 (t,

J = 7.61 Hz, 1H), 7.68 (d, J = 7.51 Hz, 1H), 7.47 (d, J = 6.85 Hz, 1H),

7.42 (d, J = 7.71 Hz, 1H), 7.41 (d, J = 11.98 Hz, 1H), 7.12 (t,

J = 7.81 Hz, 1H), 4.93 (s, 2H), 3.87 (s, 3H), 1.37 (s, 9H) ppm. ¹³C NMR

15

T. Getter, et al.

BioorganicChemistry92(2019)103250

(100.6 MHz, DMSO‑d₆): δ 163.64 (d, J = 3.08 Hz), 162.9, 161.09,

160.8 (d, J = 257.47 Hz), 159.51, 157.62, 150.178, 150.102, 144.66

(d, J = 8.52 Hz), 141.58, 131.89, 130.23, 130.23, 127.71, 122.87,

119.77, 117.28 (d, J = 10.43 Hz), 115.35 (d, J = 23.26 Hz), 76, 52.27,

4.3. Biological part

4.3.1. In vitro human leukocyte traﬃcking assays.

Leukocyte traﬃcking assays were conducted as described previously

[75,76]. Brieﬂy, human umbilical vein endothelial cells (HUVECs) were

isolated by collagenase treatment of umbilical veins and maintained

(passages 2–4) in Medium 199 containing 10% FCS, 15 ug/mL en-

dothelial cell growth supplement (Upstate Biotechnology), 100 ug/mL

heparin, and 50 uM hydrocortisone. For the ﬂow assays, HUVECs were

cultured after maximum of 3–5 passages in chamber slides (m-slide VI

0.1 ibiTreat, IBIDI), previously coated with 0.2% gelatin [76–78]

(Sigma-Aldrich and 1 mg/ml Collagen G (Biochrom AG) for 2–3 days

and then treated for 24 h with acute or chronic inﬂammation protocols

using tumor necrosis factor α (TNF- α) at 1000 U/mL alone or in

combination with interferon γ(IFN γ), at 500 U/ml for chronic in-

ﬂammatory protocols. The compounds to be tested and leukocytes were

added at 30 μM (Supporting Information Fig. S1A). Monocytes, T lym-

phocytes, and B lymphocytes were puriﬁed from EDTA blood collected

from healthy donors using Ficoll gradients and negative selection kits

(Miltenyi Biotec) for those types of cells. The slides containing the

cultured HUVEC monolayers were then mounted to a ﬂow system set at

37 °C and ﬂow was generated over the HUVEC monolayer by perfusing

a wash medium or a medium containing the leukocyte suspension using

a calibrated pump (Supporting Information Fig. S1B). The ﬂow rate,

expressed as a measurement of shear stress, was representative of small

venules/capillaries (0.05 Pascal (Pa)). The ﬁrst stage of the assay was a

washing procedure, where the wash buﬀer was perfused over the HU-

VECs for 10 min (Supporting Information Fig. S1C, step 1). For chronic

activation protocols, an additional ﬂow step of 15 min with wash buﬀer

containing 1 µM of CXCL12 for T cell assays and 1 µM CXCL13 for B cell

assays. Wash buﬀer was then perfused over the HUVECs for 10 min to

remove unbound chemokine, before puriﬁed leukocytes were perfused

over the HUVECs for 5 min (Supporting Information Fig. S1C, step 2),

followed by 45 min with wash buﬀer (Supporting Information Fig. S1C,

step 3). Throughout steps 2–3, images of the captured leukocytes under

ﬂow are made using phase-contrast microscopy, and a high-resolution

camera. Individual images are recorded every 30 s and compiled into

movie sequences, allowing analysis of individual leukocytes over large

areas. Leukocytes adherent to the surface of the HUVECs have a phase-

white appearance, whereas those that have transmigrated have a phase-

black appearance. For individual cell tracking, the phase appearance of

each leukocyte was marked at 1-minute intervals in deﬁned areas by

running the movies forth and back multiple times [44].

34.62, 30.52 ppm. ¹⁹F (376 MHz, DMSO‑d₆):

δ −110.47 (dd,

J = 11.98 Hz, J = 7.52 Hz) ppm. MS (ESI) found m/z 455.1621 (calcd

for C₂₄H₂₃FN₂O₆: 455.1613 [M+H]⁺).

4.2.28. General procedure for the synthesis of benzoic acid derivatives: 14,

22, 23, 24 [42]

To a stirred solution of 10, 19, 20, and 21 (5.55 mmol) in 1,4-di-

oxane/methanol/water (1:1:1) (30 mL), 2 M NaOH (5.55 mL,

11.10 mmol) was added dropwise. After 5 h the mixture was acidiﬁed to

pH 3 with 1 M HCl, the organic phase was evaporated, the residue was

dissolved in ethyl acetate (50 mL), washed with 10% citric acid (2x20

mL) and brine (2x15 mL), dried over Na₂SO₄, and concentrated under

reduced pressure.

4.2.29. 4-((2-((2,4,6-trioxotetrahydropyrimidin-5(2H) ylidene) methyl)

phenoxy) methyl) benzoic acid (14)

(yellow powder, yield 53%) ¹H NMR (600 MHz, DMSO‑d₆): δ 12.97

(bs, 1H), 11.33 (s, 1H), 11.16 (s, 1H), 8.58 (s, 1H), 8.05 (d, J = 8.79 Hz,

1H), 7.97 (d, J = 8.44 Hz, 2H), 7.57 (d, J = 8.44 Hz, 2H), 7.49 (t,

J = 8.79 Hz, 1H), 7.17 (d, J = 8.79 Hz, 1H), 7.00 (t, J = 8.79 Hz, 1H)

ppm. ¹³C NMR (150.9 MHz, DMSO‑d₆) δ 166.91, 163.23, 161.32,

157.67, 150.14, 149.65, 141.55, 133.71, 132.40, 130.40, 129.41,

127.16, 122.03, 119.88, 118.92, 112.35, 69.28 ppm. MS (ESI, SQ), m/z

(%): (389.1) [M+Na]⁺

.

4.2.30. 3-((2-methyl-6-((2,4,6-trioxotetrahydropyrimidin-5(2H)-ylidene)

methyl)phenoxy) methyl) benzoic acid (22)

(yellow powder, yield 59%, m.p 234 °C) ¹H NMR (600 MHz,

DMSO‑d₆): δ 11.32 (s, 1H), 11.14 (s, 1H), 8.39 (s, 1H), 7.98 (s, 1H),

7.91 (d, J = 7.82 Hz, 1H), 7.79 (d, J = 7.82 Hz, 1H), 7.65 (d,

J = 7.41 Hz, 1H), 7.49 (d, J = 7.82 Hz, 1H), 7.41 (d, J = 7.41 Hz, 1H),

7.10 (t, J = 7.41 Hz, 1H), 4.90 (s, 2H), 2.30 (s, 3H) ppm. ¹³C NMR

(150.9 MHz, DMSO‑d₆): δ 166.91, 162.98, 161.06, 156.66, 150.15,

136.98, 134.65, 132.78, 130.88, 130.64, 129.88, 129.16, 128.98,

128.61, 127.12, 123.17, 119.52, 75.48, 15.74 ppm. MS (ESI) found m/z

403.09021 (calcd for C₂₀H₁₆N₂O₆: 403.09006 [M+Na]⁺).

4.2.31. 3-((2-((2,4,6-trioxotetrahydropyrimidin-5(2H) ylidene) methyl)

phenoxy) methyl) benzoic acid (23)

(yellow powder, yield 49%, m.p. 242 °C) ¹H NMR (600 MHz, DMSO-

6D): δ 11.31 (s, 1H), 11.13 (s, 1H), 8.55 (s, 1H), 8.2 (s, 1H), 7.99 (d,

J = 8.05 Hz, 1H), 7.91(d, J = 7.47 Hz, 1H), 7.71(d, J = 8.05 Hz, 1H),

7.54 (t, J = 8.05 Hz, 1H), 7.50 (t, J = 7.47 Hz, 1H), 7.19 (d,

J = 8.38 Hz, 1H), 7.00 (t, J = 7.47 Hz, 1H) ppm. ¹³C NMR (150.9 MHz,

DMSO-6D): δ 167.00, 163.25, 161.36, 157.81, 150.18, 149.75, 137.18,

133.80, 132.49, 131.76, 131.03, 128.84, 128.26, 122.06, 119.87,

118.90, 112.46, 69.44 ppm. MS (ESI) found m/z 367.09259. calcd for

Adhesion events were recorded as the total number of cells per unit

ﬁeld (mm²). Transmigration events are presented as a percentage of

total leukocytes captured from ﬂow per unit ﬁeld. All experiments were

carried out using triplicate ﬁelds and are presented as a mean value

with + standard error measurements (+SEM). Statistical analyses as-

sume parametric distributions, and are conducted using Student’s t-test.

P values from signiﬁcance scores are presented in the ﬁgures as follows:

* P < 0.05, ** P < 0.01, *** P < 0.005 [44].

C

₁₉H₁₄N₂O₆: 367.09246 [M+H]⁺. Anal. Calcd for C₁₉H₁₄N₂O₆: C,

62.30; H, 3.85; N, 7.65. Found: C, 62.32; H, 3.90; N, 7.58.

4.4. Cell cultures

4.2.32. 3-((4-((2,4,6-trioxotetrahydropyrimidin −5(2H) ylidene) methyl)

phenoxy) methyl) benzoic acid (24)

Jeko-1 and PC12 cells were grown in RPMI 1640 medium (11.1 mM

glucose) supplemented with 10% FCS, 1% glutamine, and antibiotics

(100 μg/mL penicillin, 100 μg/mL streptomycin) at 37 °C in a 5% CO₂

humidiﬁed atmosphere. HeLa cells were grown in DMEM, 22.5 mM

glucose, 10% FCS, 1% glutamine, and antibiotics (100 μg/mL penicillin,

100 μg/mL streptomycin) at 37 °C in a 5% CO₂humidiﬁed atmosphere.

(yellow powder, yield 61%, m.p. 289 °C) ¹H NMR (600 MHz, DMSO-

6D): δ 11.30 (s, 1H), 11.17 (s, 1H), 8.37 (d, J = 9 Hz, 2H), 8.26 (s, 1H),

7.93 (d, J = 7.79 Hz, 1H), 7.73 (d, J = 7.79 Hz, 1H), 7.55 (t,

J = 7.79 Hz, 1H), 7.16 (d, J = 9 Hz, 2H), 5.33 (s, 2H), 2.09 (s, 3H) ppm.

¹³C NMR (150.9 MHz, DMSO‑d₆): δ 166.98, 163.77, 162.22, 162.03,

154.74, 150.08, 137.28, 136.90, 132.01, 130.99, 128.85, 128.79,

128.38, 125.38, 115.71, 114.61, 68.99 ppm. MS (ESI) found m/z

367.09265 (calcd for C₁₉H₁₄N₂O₆: 367.09246 [M+H]⁺). Anal. Calcd

for C₁₉H₁₄N₂O₆: C, 62.30; H, 3.85; N, 7.65. Found: C, 62.12; H, 3.91; N,

7.43.

4.5. Cell lysate preparation.

4.5.1. Lysis of suspension culture (Jeko-1 cell line)

Cells were grown to a density of 2 million cells/mL. Thereafter, cells

were centrifuged (300g for 5 min) in a conical tube for 5 min at room

16

T. Getter, et al.

BioorganicChemistry92(2019)103250

temperature. The obtained pellet was washed with 10 mL of ice cold

PBS and then spun for an additional time (300g for 5 min). PBS was

aspirated from the tube supernatant. RIPA Lysis buﬀer (50 µL) with

protease inhibitor mix was added per 1 × 10⁶cells. The lysates were

kept on ice for 15 min and then sonicated by Branson Digital Soniﬁer®

set at 50% amplitude) three times for two seconds, each with at least a

1 min rest on ice between each two-second pulse. Thereafter, the lysates

were incubated for an additional 15 min on ice and centrifuged at

13,000g for 5 min at 4 °C. The protein concentration was determined by

the bicinchoninic acid method. Aliquots were stored at −20 °C.

4.10. ¹⁹FNMR of compound 30 in HeLa and PC12 cells lysates

Cells were cultured as described above in Petri dishes. Compound

30 (2 mg in 100 µL of DMSO) was introduced into the medium (12 mL)

Control cells received only 100 µL of DMSO. The cells were incubated

for 30 min. The medium was taking out by vacuum aspiration. The cells

were washed with PBS (room temperature) 10 mL × 3 times. The cells

were lysed with hypertonic lysis buﬀer (0.1 M, Tris, pH = 8.8; 0.15 M,

NaCl, 0.25% SDS in water, 200 µL per plate). The obtained lysates were

centrifuged in 13.500 rpm for 20 min at 4 °C. The obtained lysates were

taken into NMR tubes. The 300 µL of D₂O were added to each tube and

¹⁹FNMR measurements were conducted.

4.5.2. Lysis to adherent cells (HeLa cell line)

HeLa cells were grown to approximately 80% conﬂuence. After the

medium was aspirated, the dishes were kept on ice for all the lysis

preparation steps. Cells were washed one time with 10 mL of ice cold

PBS. An excess of PBS was aspirated. Then 500 µL of RIPA lysis buﬀer

with protease inhibitor mix was added to each plate and the buﬀer was

swirled very well for an equal distribution of the buﬀer. Afterwards, cell

scraper cells were scraped and transferred to a 15 mL test tube. The

obtained lysates were incubated on ice for 15 min and sonicated by

Branson Digital Soniﬁer (at 50% amplitude) three times for two seconds

each with at least 1 min rest on ice between each two-second pulse.

Thereafter, the lysates were centrifuged at 13,000g for 5 min at 4 °C.

The supernatant was collected into new microtubes. The protein con-

centration was determined by the bicinchoninic acid method. Aliquots

were stored at −20 °C.

4.11. Methylcellulose-based suspension

1% methyl cellulose in double distillated water (DDW) solution was

prepared by adding 0.5 g methyl cellulose into 50 mL of DDW and

stirring the solution at the lowest possible speed (without making

bubbles) overnight at room temperature. Compound 12 stock solution

was prepared at a concentration of 20 mg/mL at room temperature

following an established protocol: First, 30 mg of the compound were

placed in a glass vial together with a small magnet. Thereafter, 1.5 mL

of 1% methylcellulose were added by very slow circular swirling

movements. The entire surface of the vial was gently covered by the

solvent and powder was not allowed to ﬂoat above the liquid. The re-

sulting suspension was stirred at the lowest possible speed up to 48 h at

room temperature. An intermediate stock solution of compound 12 at a

10 mg/mL concentration was prepared by diluting the obtained stock

solution (adding the stock solution directly to equal amounts of DDW).

The obtained stock solution was stirred at the lowest possible speed for

another hour at room temperature. Finally, a compound 12 working

solution at 3 mg/mL was prepared by diluting an intermediate stock

solution (v/v) with DDW accordingly. The working solution of com-

4.6. Determination of MAdCAM-1 binding aﬃnity in the presence of

compound 12 by the EIAab® ELISA kit

This immunoassay kit allows for the in vitro quantitative determi-

nation of human hMAdCAM-1 concentrations in serum, plasma, tissue

homogenates, and cell culture supernates. The plate provided in this kit

was pre-coated with an antibody speciﬁc to hMAdCAM-1. HeLa cell

lysates (50 µg of total protein) were placed in each well. The lysates

were incubated with compound 12 or DMSO on ice with gentle shaking

for 30 min. Thereafter, the samples were treated with kit reagents and

processed according to the kit manufacturer’s instructions. Finally, the

optical density of each well was detected using a micro plate reader at

450 nm.

pound 12 (gentle suspension) was stirred at

a

low speed

(~80–100 rpm) for 30 min − 1 h prior to use in the animal experi-

ments. The working solution of the compound (3 mg/mL) was prepared

on the day that the experiment starts. Compound 12 is stable for at least

7 days at a 10 mg/mL concentration, while constantly being stirred at

room temperature.

4.12. Animals

4.7. Determination of JAM-C binding aﬃnity in the presence of compound

Mice were handled according to the regulations formulated by the

Israel Animal Care and Use Committee (Ministry of Health) and the

University Medical Center, University of Geneva (Geneva, Switzerland).

All experiments were approved by the Animal Care Committee of

Geneva and by the Swiss National Veterinary Law. In Israel all ex-

periments were approved by the Animal Care Committee of the

Ministry of Health.

12 by the RayBio® human JAM-C ELISA kit

Jeko-1 cell lysates were used for the assay. First, 50 µg of total

protein was placed in each well. The lysates were incubated with

compound 12 or DMSO on ice with gentle shaking for 30 min. The

RayBio® Human JAM-C ELISA kit was used for determining the possible

inhibitory eﬀect of compound 12 on JAM-C binding. This assay em-

ploys an antibody speciﬁc for human JAM-C coated on a 96-well plate.

After the pre-incubation with compound 12, lysates were processed

according to the kit manufacturer’s instructions. Finally, the optical

density of each well was detected using a micro plate reader at 450 nm.

4.13. Acute toxicology

Eight experimental groups with 5 female 6–7-week-old C57BL mice

each and one control group were used for the experiment. Mice were

injected using compound 12 or a suitable control using a methylcellu-

lose formulation. Mice from all experimental groups were observed for

a possible change in their behavior: immediately post-dosing and fol-

lowing 2 h post-administration. Mice from all groups were euthanized

with isoﬂurane 24 h after the last administered dose. Sampling for WBC

and White Blood Diﬀerentiation: 200 µL of whole blood were collected

into EDTA tubes (Greiner MiniCollect K3EDTA tubes). The remaining

whole blood (300–400 µL) was collected for serum preparation (Greiner

4.8. Determination of VCAM-1 binding aﬃnity in the presence of compound

12 by the human VCAM-1 ELISA Kit

The assay was conducted as we described above (MAdCAM-1

assay).

4.9. Determination of β integrins mediated Jaco-1 cell adhesion in the

presence of compound 12 by the β Integrin-Mediated cell adhesion Kit

MiniCollect

Z Serum Tubes). All blood tests were conducted in

American Medical Labs, Herzeliah, Israel. Liver, spleen, lungs, heart,

and kidneys were collected from each mouse and placed into vials with

4% buﬀered formalin for future preparation of paraﬃn blocks. Sections

The assay was conducted as we described in manufacturer protocol

for dispersed cells.

17

T. Getter, et al.

BioorganicChemistry92(2019)103250

from each organ of each animal were stained with H&E and histological

evaluation of the tissue was performed for identiﬁcation of macrosigns

of possible toxicity.

4.18. Collagen-Induced arthritis (CIA) model in DBA/1 Mice

Male DBA/1 mice (7–8 weeks of age) were injected with CII (Type II

collagen) + CFA (complete Freund's adjuvant) emulsion on day 0 (SC)

to stimulate CIA [80]. On day 21, the animals received the same

amount of the emulsion as a booster. Twice between days 0 and 18 after

immunization (before enrollment begins), both hind paws of all mice

were measured using calipers to establish the baseline values for paw

thickness. On day 22 the mice were enrolled into groups when they ﬁrst

exhibited paw swelling and treatment was initiated. A compound 12

was administered by IP injections at 15 mg/kg dose using the q3w re-

gimen. The mock-treated (vehicle) control group was administered IP

with the same volume and regimen as the treatment group. Adminis-

tration was performed at a constant volume dosage based on individual

body weight. The severity of CIA was monitored daily. The CIA scoring

is on a scale of 0 to 16 (0 to 4 for each paw, adding the scores for all 4

paws). On day 43 (24 h after the last administered dose), all paw

thicknesses were measured and the mice were euthanized. All limbs

from each mouse were placed in individually labeled tubes, ﬁxed in 4%

buﬀered formalin, paraﬃn embedded, and sectioned. The histological

score was assessed using H&E stained sections. Joints were scored for

inﬂammation, cartilage damage, pannus formation, and bone resorp-

tion on a scale of 0 to 3 for each readout. Two-way ANOVA statistical

analysis was performed.

4.14. Chronic toxicology (GLP guided experiment)

The experiment was conducted by the “Science and Action” com-

pany according to GLP protocol. The data and detailed protocol are

presented in the SI.

4.15. IBD mouse model

Female 7–8-week-old C57BL mice were treated with 1.5% dextran

sodium sulfate in drinking water. Compound 12 was administered as

described. Determination of the individual body weights of animals was

made prior to each dosing and the administered volume was adjusted.

During the administration of the test compound, the mouse body

weights were measured daily. The DAI score was monitored daily, ac-

cording to criteria mentioned in Table 1. Mice from all groups were

euthanized with isoﬂurane 24 h after administering the last dose. The

colon was collected from each mouse from all groups, ﬁlled with 4%

buﬀered formalin, and the colon length was measured.

4.16. MS mouse model

4.19. Docking

Female 7–8-week-old C57BL mice were treated with OG35-55/CFA

according to the standard protocol [79]. After the onset of EAE, mice

were assigned to the groups (treated and untreated as described). The

body weight of animals was determined prior to starting the experiment

and the administered volume of the formulation of compound 12 was

adjusted according to the body weight during the experiment; the body

weight was measured daily. The severity of the EAE score was mon-

itored daily, according to criteria mentioned in Table 2. Mice from all

groups were euthanized with isoﬂurane 24 h after the last administered

dose.

Prior to docking, the protein (PECAM-1 IgL1-2 trans-homophilic

dimer, PDB code: 5GNI) was prepared using the prepared protein pro-

tocol, as implemented in Maestro (Schrödinger, USA) to assign the

correct protonation states to titrate-able residues. The structures of the

active (12, 26, 4, 11, 13, 23, and 24) and inactive compounds (1, 2, 3,

14, and 22) were processed by Ligprep as implemented in Maestro, to

assign correct protonation states at a physiological pH. Prior to docking,

potential druggable binding sites in PECAM-1 were identiﬁed using

SiteMap. Five possible binding sites were identiﬁed (Fig. 14A). Grid

boxes were centered on the identiﬁed sites. Docking of all compounds

to all sites was performed using the induced ﬁt docking (IFD) protocol

in which residues within 10 A˚ of any of the resulting top 20 ligand

poses (from the initial docking) were subjected to a conformational

search and minimization, whereas residues outside the region were kept

ﬁxed [81–83]. The IFD score consists of GlideScore SP + 0.05 × Pri-

meEnergy [83]. The resulting poses from the docking procedure were

sorted by the IFD score and subjected to enrichment calculation, as

implemented in Maestro to produce ROC curves. These plots were used

to rank the diﬀerent binding sites in terms of their ability to distinguish

between active and inactive compounds.

4.17. The High-Fat Diet-Induced NAFLD mouse model.

C57BL/6 male mice were fed ad libitum a high-fat diet, no high-

carbohydrate diet (modiﬁed western diet, WD) with 42% kcal from fat

containing 0.1% cholesterol (Harlan TD.88137) with normal water

(without high fructose-glucose solution) for 14 weeks prior to initiating

the treatment. Control mice were fed a standard chow diet (CD, Harlan

TD.7012) with normal water (NW). A working formulation of com-

pound 12 was administered on the 85th day of the high-fat diet (15 mg/

kg, for SC and IP using a 30 mg/kg q3w regimen for PO along with a qd

regimen) to a mock-treated (vehicle) control group by IP injections at

the same volume and regimen as a correspondingly treated group for

65 days. The body weight of animals was determined prior to initiating

the treatment and the administered volume of compound 12 was ad-

justed according to the body weight. All animals were euthanized 24 h

after the last administered dose. At necropsy, liver and mesenteric fat

wet weights were determined. Liver and mesenteric fat were ﬁxed with

4% buﬀered formalin, embedded in paraﬃn sections, and H&E stained.

Steatohepatitis was deﬁned by the presence of steatosis, inﬂammation,

and hepatocellular ballooning, according to the fatty liver inhibition of

progression (FLIP) algorithm. The severities of steatosis, lobular in-

ﬂammation, and hepatocellular ballooning were scored using the

NASH-Clinical Research Network (CRN) criteria. The NAFLD activity

score (NAS) was calculated by adding the scores of steatosis (the

maximum score was 3). It is deﬁned by more than 66% of hepatocytes

containing fat droplets; lobular inﬂammation (the maximum score of 3)

is deﬁned by > 4 foci/200 ﬁeld; hepatocyte ballooning (the maximum

score of 2) is deﬁned by many cells or prominent ballooning. Ordinary

one-way ANOVA statistical analysis was performed.

4.20. PK study

Twelve C57 black mice received compound 12, PO (gavage, 50 mg/

kg) in DMSO 70 µL per mice. Three mice received only DMSO and were

used as a control for zero time point. At time 30 min, 2.5 h, 6 h and 9 h,

three mice per time point were euthanized and whole blood was taken

from heart aspiration. The blood was centrifuged for 2000 rpm/10 min.

Plasma was treated by acetonitrile (70 µL per y 200 µL of plasma) and

extracted by chloroform (4 mL per 1 mL of plasma). Chloroform was

transferred to a clean tube and evaporated until dryness. The obtained

residue was dissolved in 100 µL of DMSO and injected to LC-MS.

Spiking and calibration curve were generated using plasma of the non-

treated mice.

5. Patent

All data described in the manuscript are protected by patent appl.

No. 62/552,491.

18

T. Getter, et al.

BioorganicChemistry92(2019)103250

Author contributions

disease and ulcerative colitis, Inﬂammopharmacology 20 (1) (2012) 1–18 .

[21] H. Hoshino, M. Kobayashi, J. Mitoma, Y. Sato, M. Fukuda, J. Nakayama, An integrin

alpha4beta7*IgG heterodimeric chimera binds to MAdCAM-1 on high endothelial

venules in gut-associated lymphoid tissue, J. Histochem. Cytochem. 59 (6) (2011)

572–583 .

The manuscript was written through the contributions of all au-

thors. T.G., R.M., P.B., E.A., R. L., B.I., H.S., and A.G., designed the

experiments. T.G., R.M., E.A., P.B., S.Z., N.K., N. K-L., S.K., E.B., N.H.

and L.L. performed the experiments. All authors have approved the

ﬁnal version of the manuscript.

[22] T. Ando, R.R. Langley, Y. Wang, P.A. Jordan, A. Minagar, J.S. Alexander,

M.H. Jennings, Inﬂammatory cytokines induce MAdCAM-1 in murine hepatic en-

dothelial cells and mediate alpha-4 beta-7 integrin dependent lymphocyte en-

dothelial adhesion in vitro, BMC Physiol. 7 (2007) 10 .

[23] E.J. Park, J.R. Mora, C.V. Carman, J. Chen, Y. Sasaki, G. Cheng, U.H. von Andrian,

M. Shimaoka, Aberrant activation of integrin alpha4beta7 suppresses lymphocyte

migration to the gut, J. Clin. Invest. 117 (9) (2007) 2526–2538 .

Acknowledgements

[24] S. Ghosh, R. Panaccione, Anti-adhesion molecule therapy for inﬂammatory bowel

disease, Therap. Adv. Gastroenterol. 3 (4) (2010) 239–258 .

This study was supported by a Bar-Ilan University new faculty grant

(to A.G). T.G was the recipient of a Lev Zion Fellowship, Israel. S.K. is

grateful for the support of her work by the Wolf Foundation, Israel. We

thank Mr. Steven Manch for the English editing.

[25] S. Vermeire, W.J. Sandborn, S. Danese, X. Hebuterne, B.A. Salzberg, M. Klopocka,

D. Tarabar, T. Vanasek, M. Gregus, P.A. Hellstern, J.S. Kim, M.P. Sparrow,

K.J. Gorelick, M. Hinz, A. Ahmad, V. Pradhan, M. Hassan-Zahraee, R. Clare,

F. Cataldi, W. Reinisch, Anti-MAdCAM antibody (PF-00547659) for ulcerative co-

litis (TURANDOT): a phase 2, randomised, double-blind, placebo-controlled trial,

Lancet 390 (10090) (2017) 135–144 .

Appendix A. Supplementary material

[26] M.L. Scribano, Vedolizumab for inﬂammatory bowel disease: From randomized

controlled trials to real-life evidence, World J. Gastroenterol. 24 (23) (2018)

2457–2467 .

Supplementary data to this article can be found online at https://

doi.org/10.1016/j.bioorg.2019.103250.

[27] J.L. Viney, S. Jones, H.H. Chiu, B. Lagrimas, M.E. Renz, L.G. Presta, D. Jackson,

K.J. Hillan, S. Lew, S. Fong, Mucosal addressin cell adhesion molecule-1: a struc-

tural and functional analysis demarcates the integrin binding motif, J. Immunol.

157 (6) (1996) 2488–2497 .

References

[28] H.N. Shroﬀ, C.F. Schwender, A.D. Baxter, F. Brookﬁeld, L.J. Payne, N.A. Cochran,

D.L. Gallant, M.J. Briskin, Novel modiﬁed tripeptide inhibitors of alpha 4 beta 7

mediated lymphoid cell adhesion to MAdCAM-1, Bioorg. Med. Chem. Lett. 8 (13)

(1998) 1601–1606 .

[1] A. Watad, S. Azrielant, N.L. Bragazzi, K. Sharif, P. David, I. Katz, G. Aljadeﬀ,

M. Quaresma, G. Tanay, M. Adawi, H. Amital, Y. Shoenfeld, Seasonality and au-

toimmune diseases: The contribution of the four seasons to the mosaic of auto-

immunity, J. Autoimmun. 82 (2017) 13–30 .

[29] J. Boer, D. Gottschling, A. Schuster, M. Semmrich, B. Holzmann, H. Kessler, Design

and synthesis of potent and selective alpha(4)beta(7) integrin antagonists, J. Med.

Chem. 44 (16) (2001) 2586–2592 .

[2] L. Wang, F.S. Wang, M.E. Gershwin, Human autoimmune diseases: a comprehensive

update, J. Intern. Med. 278 (4) (2015) 369–395 .

[30] N.J. Dubree, D.R. Artis, G. Castanedo, J. Marsters, D. Sutherlin, L. Caris, K. Clark,

S.M. Keating, M.H. Beresini, H. Chiu, S. Fong, H.B. Lowman, N.J. Skelton,

D.Y. Jackson, Selective alpha4beta7 integrin antagonists and their potential as

antiinﬂammatory agents, J. Med. Chem. 45 (16) (2002) 3451–3457 .

[31] G.M. Castanedo, F.C. Sailes, N.J. Dubree, J.B. Nicholas, L. Caris, K. Clark,

S.M. Keating, M.H. Beresini, H. Chiu, S. Fong, J.C. Marsters Jr., D.Y. Jackson,

D.P. Sutherlin, Solid-phase synthesis of dual alpha4beta1/alpha4beta7 integrin

antagonists: two scaﬀolds with overlapping pharmacophores, Bioorg. Med. Chem.

Lett. 12 (20) (2002) 2913–2917 .

[3] E. Kalaitzoglou, I. Popescu, R.C. Bunn, J.L. Fowlkes, K.M. Thrailkill, Eﬀects of type

1 diabetes on osteoblasts, osteocytes, and osteoclasts, Curr. Osteoporos. Rep. 14 (6)

(2016) 310–319 .

[4] G. Azizi, H. Abolhassani, M.H. Asgardoon, T. Alinia, R. Yazdani, J. Mohammadi,

N. Rezaei, H.D. Ochs, A. Aghamohammadi, Autoimmunity in common variable

immunodeﬁciency: epidemiology, pathophysiology and management, Expert Rev.

Clin. Immunol. 13 (2) (2017) 101–115 .

[5] M.D. Rosenblum, I.K. Gratz, J.S. Paw, A.K. Abbas, Treating human autoimmunity:

current practice and future prospects, Sci. Transl. Med. 4 (125) (2012) 125sr1 .

[6] A. Sala, E. Proschak, D. Steinhilber, G.E. Rovati, Two-pronged approach to anti-

inﬂammatory therapy through the modulation of the arachidonic acid cascade,

Biochem. Pharmacol. 158 (2018) 161–173 .

[32] A.B. Dyatkin, Y. Gong, T.A. Miskowski, E.S. Kimball, S.M. Prouty, M.C. Fisher,

R.J. Santulli, C.R. Schneider, N.H. Wallace, P.J. Hornby, C. Diamond, W.A. Kinney,

B.E. Maryanoﬀ, B.P. Damiano, W. He, Aza-bicyclic amino acid carboxamides as

alpha4beta1/alpha4beta7 integrin receptor antagonists, Bioorg. Med. Chem. 13

(24) (2005) 6693–6702 .

[7] M. Del Grossi Moura, L. Cruz Lopes, M.T. Silva, S. Barberato-Filho, R.H.L. Motta,

C.C. Bergamaschi, Use of steroid and nonsteroidal anti-inﬂammatories in the

treatment of rheumatoid arthritis: Systematic review protocol, Medicine

(Baltimore) 97 (41) (2018) e12658 .

[33] G.C. Harriman, M. Brewer, R. Bennett, C. Kuhn, M. Bazin, G. Larosa, P. Skerker,

N. Cochran, D. Gallant, D. Baxter, D. Picarella, B. Jaﬀee, J.R. Luly, M.J. Briskin,

Selective cell adhesion inhibitors: Barbituric acid based alpha4beta7–MAdCAM

inhibitors, Bioorg. Med. Chem. Lett. 18 (7) (2008) 2509–2512 .

[8] J. Saad, M.V. Pellegrini, Nonsteroidal anti-inﬂammatory drugs (NSAID) toxicity,

StatPearls, Treasure Island (FL), (2018).

[34] C. Xu, A.R. Wyman, M.A. Alaamery, S.A. Argueta, F.D. Ivey, J.A. Meyers, A. Lerner,

T.H. Burdo, T. Connolly, C.S. Hoﬀman, T.C. Chiles, Anti-inﬂammatory eﬀects of

novel barbituric acid derivatives in T lymphocytes, Int. Immunopharmacol. 38

(2016) 223–232 .

[9] C.C. Chow, S.S. Simons Jr., An approach to greater speciﬁcity for glucocorticoids,

Front. Endocrinol. (Lausanne) 9 (2018) 76 .

[10] M. Jastrzebska, M.E. Czok, P. Guzik, Autoimmune diseases, their pharmacological

treatment and the cardiovascular system, Cardiol. J. 20 (6) (2013) 569–576 .

[11] S.L. Thomas, C. Griﬃths, L. Smeeth, C. Rooney, A.J. Hall, Burden of mortality as-

sociated with autoimmune diseases among females in the United Kingdom, Am. J.

Public Health 100 (11) (2010) 2279–2287 .

[35] V. Temml, T. Kaserer, Z. Kutil, P. Landa, T. Vanek, D. Schuster, Pharmacophore

modeling for COX-1 and -2 inhibitors with LigandScout in comparison to Discovery

Studio, Future Med. Chem. 6 (17) (2014) 1869–1881 .

[36] G. Wolber, T. Langer, LigandScout: 3-D pharmacophores derived from protein-

bound ligands and their use as virtual screening ﬁlters, J. Chem. Inf. Model. 45 (1)

(2005) 160–169 .

[12] G.S. Hamilton, R.E. Mewshaw, C.M. Bryant, Y. Feng, G. Endemann, K.S. Madden,

J.E. Janczak, J. Perumattam, L.W. Stanton, X. Yang, et al., Fluorenylalkanoic and

benzoic acids as novel inhibitors of cell adhesion processes in leukocytes, J. Med.

Chem. 38 (10) (1995) 1650–1656 .

[37] J.D. Figueroa-Villar, A.A. Vieira, Nuclear magnetic resonance and molecular mod-

eling study of exocyclic carbon-carbon double bond polarization in benzylidene

barbiturates, J. Mol. Struct. 1034 (2013) 310–317 .

[13] F.A. Acker, H.P. Voss, H. Timmerman, Chemokines: structure, receptors and func-

tions. A new target for inﬂammation and asthma therapy? Mediators Inﬂamm 5 (6)

(1996) 393–416 .

[38] E. Knoevenagel, Condensation von Malonsäure mit aromatischen Aldehyden durch

Ammoniak und Amine, Ber. Dtsch. Chem. Ges. 31 (3) (1898) 2596–2619 .

[39] K.M. Khan, M. Ali, M. Khan, M. Taha, S. Perveen, NH4Cl mediated new protocol for

the synthesis of 5-arylidene barbiturates, Lett. Org. Chem. 8 (1) (2011) 28–32 .

[40] J. Lasri, G. Gajewski, M.F.C.G. da Silva, M.L. Kuznetsov, R.R. Fernandes,

A.J.L. Pombeiro, Solvent-dependent reactivities of acyclic nitrones with beta-dike-

tones: catalyst-free syntheses of endiones and enones, Tetrahedron 68 (35) (2012)

7019–7027 .

[14] K. Ley, J. Rivera-Nieves, W.J. Sandborn, S. Shattil, Integrin-based therapeutics:

biological basis, clinical use and new drugs, Nat. Rev. Drug Discov. 15 (3) (2016)

173–183 .

[15] P. Vanderslice, D.G. Woodside, Integrin antagonists as therapeutics for in-

ﬂammatory diseases, Expert Opin. Invest. Drugs 15 (10) (2006) 1235–1255 .

[16] S.L. Goodman, M. Picard, Integrins as therapeutic targets, Trends Pharmacol. Sci.

33 (7) (2012) 405–412 .

[41] S. Vassiliou, M. Xeilari, A. Yiotakis, J. Grembecka, M. Pawelczak, P. Kafarski,

A. Mucha, A synthetic method for diversiﬁcation of the P1' substituent in phos-

phinic dipeptides as a tool for exploration of the speciﬁcity of the S1' binding

pockets of leucine aminopeptidases, Bioorg. Med. Chem. 15 (9) (2007) 3187–3200 .

[42] N. Zidar, T. Tomasic, R. Sink, A. Kovac, D. Patin, D. Blanot, C. Contreras-Martel,

A. Dessen, M.M. Premru, A. Zega, S. Gobec, L.P. Masic, D. Kikelj, New 5-benzyli-

denethiazolidin-4-one inhibitors of bacterial MurD ligase: design, synthesis, crystal

structures, and biological evaluation, Eur. J. Med. Chem. 46 (11) (2011)

5512–5523 .

[17] M.J. Briskin, L.M. McEvoy, E.C. Butcher, MAdCAM-1 has homology to im-

munoglobulin and mucin-like adhesion receptors and to IgA1, Nature 363 (6428)

(1993) 461–464 .

[18] E.E. Sikorski, R. Hallmann, E.L. Berg, E.C. Butcher, The Peyer's patch high en-

dothelial receptor for lymphocytes, the mucosal vascular addressin, is induced on a

murine endothelial cell line by tumor necrosis factor-alpha and IL-1, J. Immunol.

151 (10) (1993) 5239–5250 .

[19] R.K. Pachynski, S.W. Wu, M.D. Gunn, D.J. Erle, Secondary lymphoid-tissue che-

mokine (SLC) stimulates integrin alpha 4 beta 7-mediated adhesion of lymphocytes

to mucosal addressin cell adhesion molecule-1 (MAdCAM-1) under ﬂow, J.

Immunol. 161 (2) (1998) 952–956 .

[43] E.C. Butcher, Leukocyte-endothelial cell recognition: three (or more) steps to spe-

ciﬁcity and diversity, Cell 67 (6) (1991) 1033–1036 .

[44] P.F. Bradﬁeld, C. Scheiermann, S. Nourshargh, C. Ody, F.W. Luscinskas,

G.E. Rainger, G.B. Nash, M. Miljkovic-Licina, M. Aurrand-Lions, B.A. Imhof, JAM-C

[20] S. Thomas, D.C. Baumgart, Targeting leukocyte migration and adhesion in Crohn's

19

T. Getter, et al.

BioorganicChemistry92(2019)103250

regulates unidirectional monocyte transendothelial migration in inﬂammation,

border recycling compartment, J. Exp. Med. 205 (4) (2008) 951–966 .

Blood 110 (7) (2007) 2545–2555 .

[63] C. Paddock, D. Zhou, P. Lertkiatmongkol, P.J. Newman, J. Zhu, Structural basis for

PECAM-1 homophilic binding, Blood 127 (8) (2016) 1052–1061 .

[45] H.S. Schultz, S.L. Reedtz-Runge, B.T. Backstrom, K. Lamberth, C.R. Pedersen,

A.M. Kvarnhammar, A. Consortium, Quantitative analysis of the CD4+ T cell re-

sponse to therapeutic antibodies in healthy donors using a novel T cell: PBMC assay,

PLoS One 12 (5) (2017) e0178544 .

[64] K. Hashimoto, N. Kataoka, E. Nakamura, K. Hagihara, M. Hatano, T. Okamoto,

H. Kanouchi, Y. Minatogawa, S. Mohri, K. Tsujioka, F. Kajiya, Monocyte trans-en-

dothelial migration augments subsequent transmigratory activity with increased

PECAM-1 and decreased VE-cadherin at endothelial junctions, Int. J. Cardiol. 149

(2) (2011) 232–239 .

[46] A. Mathias, S. Perriot, M. Canales, C. Blatti, C. Gaubicher, M. Schluep,

B. Engelhardt, R. Du Pasquier, Impaired T-cell migration to the CNS under ﬁngo-

limod and dimethyl fumarate, Neurol. Neuroimmunol. Neuroinﬂamm. 4 (6) (2017)

e401 .

[65] K. Hashimoto, N. Kataoka, E. Nakamura, K. Tsujioka, F. Kajiya, Oxidized LDL

speciﬁcally promotes the initiation of monocyte invasion during transendothelial

migration with upregulated PECAM-1 and downregulated VE-cadherin on en-

dothelial junctions, Atherosclerosis 194 (2) (2007) e9–e17 .

[47] M.K. Racke, Y. Yang, A.E. Lovett-Racke, Is T-bet a potential therapeutic target in

multiple sclerosis? J. Interferon Cytokine Res. 34 (8) (2014) 623–632 .

[48] C. Zanotti, M. Chiarini, F. Serana, A. Sottini, E. Garrafa, F. Torri, L. Caimi, S. Rasia,

R. Capra, L. Imberti, Peripheral accumulation of newly produced T and B lym-

phocytes in natalizumab-treated multiple sclerosis patients, Clin. Immunol. 145 (1)

(2012) 19–26 .

[66] D. Bartusik, B. Tomanek, Detection of (19)F-labeled biopharmaceuticals in cell

cultures with magnetic resonance, Adv. Drug Deliv. Rev. 65 (8) (2013) 1056–1064 .

[67] M. Srinivas, P. Boehm-Sturm, C.G. Figdor, I.J. de Vries, M. Hoehn, Labeling cells for

in vivo tracking using (19)F MRI, Biomaterials 33 (34) (2012) 8830–8840 .

[68] B. Gaye, A. Adejare, Fluorinated molecules in the diagnosis and treatment of neu-

rodegenerative diseases, Future Med. Chem. 1 (5) (2009) 821–833 .

[69] S. Xu, X. Du, G. Feng, Y. Zhang, J. Li, B. Lin, L. Yang, S. Fu, J. Wu, Eﬃcient in-

hibition of cervical cancer by dual drugs loaded in biodegradable thermosensitive

hydrogel composites, Oncotarget 9 (1) (2018) 282–292 .

[49] S.K. Alford, G.D. Longmore, W.F. Stenson, C. Kemper, CD46-induced im-

munomodulatory CD4+ T cells express the adhesion molecule and chemokine re-

ceptor pattern of intestinal T cells, J. Immunol. 181 (4) (2008) 2544–2555 .

[50] T.T. Ho, J.O. You, D.T. Auguste, siRNA Delivery Impedes the Temporal Expression

of Cytokine-Activated VCAM1 on Endothelial Cells, Ann. Biomed. Eng. 44 (4)

(2016) 895–902 .

[70] B. Klementsen, L. Jorgensen, Distribution of adhesion molecules on HeLa cells,

platelets and endothelium in an in vitro model mimicking the early phase of me-

tastasis An immunogold electron microscopic study, APMIS 105 (7) (1997)

546–558 .

[51] C. Donate, C. Ody, T. McKee, S. Ruault-Jungblut, N. Fischer, P. Ropraz, B.A. Imhof,

T. Matthes, Homing of human B cells to lymphoid organs and B-cell lymphoma

engraftment are controlled by cell adhesion molecule JAM-C, Cancer Res. 73 (2)

(2013) 640–651 .

[71] S. Muro, V.R. Muzykantov, Targeting of antioxidant and anti-thrombotic drugs to

endothelial cell adhesion molecules, Curr. Pharm. Des. 11 (18) (2005) 2383–2401 .

[72] B.S. Ding, C. Gottstein, A. Grunow, A. Kuo, K. Ganguly, S.M. Albelda, D.B. Cines,

V.R. Muzykantov, Endothelial targeting of a recombinant construct fusing a

PECAM-1 single-chain variable antibody fragment (scFv) with prourokinase facil-

itates prophylactic thrombolysis in the pulmonary vasculature, Blood 106 (13)

(2005) 4191–4198 .

[52] T.L. Rogers, D. Wallick, Reviewing the use of ethylcellulose, methylcellulose and

hypromellose in microencapsulation. Part 1: materials used to formulate micro-

capsules, Drug Dev. Ind. Pharm. 38 (2) (2012) 129–157 .

[53] Y.H. Li, H.T. Xiao, D.D. Hu, S. Fatima, C.Y. Lin, H.X. Mu, N.P. Lee, Z.X. Bian,

Berberine ameliorates chronic relapsing dextran sulfate sodium-induced colitis in

C57BL/6 mice by suppressing Th17 responses, Pharmacol. Res. 110 (2016)

227–239 .

[73] S. Muro, C. Gajewski, M. Koval, V.R. Muzykantov, ICAM-1 recycling in endothelial

cells: a novel pathway for sustained intracellular delivery and prolonged eﬀects of

drugs, Blood 105 (2) (2005) 650–658 .

[54] Y.H. Park, N. Kim, Y.K. Shim, Y.J. Choi, R.H. Nam, Y.J. Choi, M.H. Ham, J.H. Suh,

S.M. Lee, C.M. Lee, H. Yoon, H.S. Lee, D.H. Lee, Adequate dextran sodium sulfate-

induced colitis model in mice and eﬀective outcome measurement method, J.

Cancer Prev. 20 (4) (2015) 260–267 .

[74] C. Garnacho, S.M. Albelda, V.R. Muzykantov, S. Muro, Diﬀerential intra-endothelial

delivery of polymer nanocarriers targeted to distinct PECAM-1 epitopes, J. Control.

Release 130 (3) (2008) 226–233 .

[55] C.S. Moore, A.L. Hebb, M.M. Blanchard, C.E. Crocker, P. Liston, R.G. Korneluk,

G.S. Robertson, Increased X-linked inhibitor of apoptosis protein (XIAP) expression

exacerbates experimental autoimmune encephalomyelitis (EAE), J. Neuroimmunol.

203 (1) (2008) 79–93 .

[75] P.F. Bradﬁeld, A. Menon, M. Miljkovic-Licina, B.P. Lee, N. Fischer, R.J. Fish,

B. Kwak, E.A. Fisher, B.A. Imhof, Divergent JAM-C expression accelerates mono-

cyte-derived cell exit from atherosclerotic plaques, PLoS ONE 11 (7) (2016)

e0159679 .

[56] L. Ma, S. Li, H. Zheng, J. Chen, L. Lin, X. Ye, Z. Chen, Q. Xu, T. Chen, J. Yang,

N. Qiu, G. Wang, A. Peng, Y. Ding, Y. Wei, L. Chen, Synthesis and biological activity

of novel barbituric and thiobarbituric acid derivatives against non-alcoholic fatty

liver disease, Eur. J. Med. Chem. 46 (6) (2011) 2003–2010 .

[76] W. He, S. Holtkamp, S.M. Hergenhan, K. Kraus, A. de Juan, J. Weber, P. Bradﬁeld,

J.M.P. Grenier, J. Pelletier, D. Druzd, C.S. Chen, L.M. Ince, S. Bierschenk, R. Pick,

M. Sperandio, M. Aurrand-Lions, C. Scheiermann, Circadian expression of mi-

gratory factors establishes lineage-speciﬁc signatures that guide the homing of

leukocyte subsets to tissues, Immunity 49 (6) (2018) pp. 1175–1190 e7 .

[77] A. Sidibe, P. Ropraz, S. Jemelin, Y. Emre, M. Poittevin, M. Pocard, P.F. Bradﬁeld,

B.A. Imhof, Angiogenic factor-driven inﬂammation promotes extravasation of

human proangiogenic monocytes to tumours, Nat. Commun. 9 (1) (2018) 355 .

[78] P. Ropraz, B.A. Imhof, T. Matthes, B. Wehrle-Haller, A. Sidibe, Simultaneous study

of the recruitment of monocyte subpopulations under ﬂow in vitro, J. Vis. Exp.

(141) (2018).

[57] L. Ma, C. Xie, Y. Ran, X. Liang, L. Huang, H. Pei, J. Chen, J. Liu, Y. Sang, H. Lai,

A. Peng, M. Xiang, Y. Wei, L. Chen, Synthesis and biological evaluation of 5-ben-

zylidenepyrimidine-2,4,6(1H,3H,5H)-trione derivatives for the treatment of obe-

9958–9972 .

[58] H. Zheng, S. Li, L. Ma, L. Cheng, C. Deng, Z. Chen, C. Xie, M. Xiang, W. Jiang,

L. Chen, A novel agonist of PPAR-gamma based on barbituric acid alleviates the

development of non-alcoholic fatty liver disease by regulating adipocytokine ex-

pression and preventing insulin resistance, Eur. J. Pharmacol. 659 (2–3) (2011)

244–251 .

[79] O. Ozgun-Acar, G. Celik-Turgut, I. Gazioglu, U. Kolak, S. Ozbal, B.U. Ergur,

S. Arslan, A. Sen, G. Topcu, Capparis ovata treatment suppresses inﬂammatory

cytokine expression and ameliorates experimental allergic encephalomyelitis model

of multiple sclerosis in C57BL/6 mice, J. Neuroimmunol. 298 (2016) 106–116 .

[80] A. Hoehne, M.L. James, I.S. Alam, J.A. Ronald, B. Schneider, A. D'Souza,

T.H. Witney, L.E. Andrews, H.C. Cropper, D. Behera, G. Gowrishankar, Z. Ding,

T. Wyss-Coray, F.T. Chin, S. Biswal, S.S. Gambhir, [(18)F]FSPG-PET reveals in-

creased cystine/glutamate antiporter (xc-) activity in a mouse model of multiple

sclerosis, J. Neuroinﬂammation 15 (1) (2018) 55 .

[59] T. Kanazawa, K. Tamano, K. Sogabe, T. Endo, H. Ibaraki, Y. Takashima, Y. Seta,

Intra-articular retention and anti-arthritic eﬀects in collagen-induced arthritis

model mice by injectable small interfering RNA containing hydrogel, Biol. Pharm.

Bull. 40 (11) (2017) 1929–1933 .

[60] H. DeLisser, Y. Liu, P.Y. Desprez, A. Thor, P. Briasouli, C. Handumrongkul,

J. Wilfong, G. Yount, M. Nosrati, S. Fong, E. Shtivelman, M. Fehrenbach, G. Cao,

D.H. Moore, S. Nayak, D. Liggitt, M. Kashani-Sabet, R. Debs, Vascular endothelial

platelet endothelial cell adhesion molecule 1 (PECAM-1) regulates advanced me-

tastatic progression, Proc. Natl. Acad. Sci. USA 107 (43) (2010) 18616–18621 .

[61] Z. Qing, M. Sandor, Z. Radvany, D. Sewell, A. Falus, D. Potthoﬀ, W.A. Muller,

Z. Fabry, Inhibition of antigen-speciﬁc T cell traﬃcking into the central nervous

system via blocking PECAM1/CD31 molecule, J. Neuropathol. Exp. Neurol. 60 (8)

(2001) 798–807 .

[81] T.A. Halgren, Identifying and characterizing binding sites and assessing drugg-

ability, J. Chem. Inf. Model. 49 (2) (2009) 377–389 .

[82] R. Farid, T. Day, R.A. Friesner, R.A. Pearlstein, New insights about HERG blockade

obtained from protein modeling, potential energy mapping, and docking studies,

Bioorg. Med. Chem. 14 (9) (2006) 3160–3173 .

[83] W. Sherman, T. Day, M.P. Jacobson, R.A. Friesner, R. Farid, Novel procedure for

modeling ligand/receptor induced ﬁt eﬀects, J. Med. Chem. 49 (2) (2006) 534–553 .

[62] Z. Mamdouh, G.E. Kreitzer, W.A. Muller, Leukocyte transmigration requires ki-

nesin-mediated microtubule-dependent membrane traﬃcking from the lateral

20

Article Doi

Novel inhibitors of leukocyte transendothelial migration

DOI: 10.1016/j.bioorg.2019.103250

Source and publish data:

Authors:

Article abstract of DOI:10.1016/j.bioorg.2019.103250

Full text of DOI:10.1016/j.bioorg.2019.103250

Products guided by the article

R&D Labs maybe for 191167-05-2

Relevant to this article

Hot Product