Synthesis and Characterization of Fatty Acid Conjugates of Niacin and Salicylic Acid

Article abstract of DOI:10.1021/acs.jmedchem.5b01961

This report describes the synthesis and preliminary biological characterization of novel fatty acid niacin conjugates and fatty acid salicylate conjugates. These molecular entities were created by covalently linking two bioactive molecules, either niacin

Full text of DOI:10.1021/acs.jmedchem.5b01961

Subscriber access provided by RMIT University Library
Article
Synthesis and Characterization of Fatty
Acid Conjugates of Niacin and Salicylic Acid
Chi B Vu, Jean E Bemis, Ericka Benson, Pradeep Bista, David Carney, Richard Fahrner, Diana
Lee, Feng Liu, Pallavi Lonkar, Jill C Milne, Andrew J Nichols, Dominic Picarella, Adam Shoelson,
Jesse Smith, Amal Ting, Allison Wensley, Maisy Yeager, Michael Zimmer, and Michael R Jirousek
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b01961 • Publication Date (Web): 19 Jan 2016
Downloaded from http://pubs.acs.org on January 25, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Synthesis and Characterization of Fatty Acid
Conjugates of Niacin and Salicylic Acid
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Chi B. Vu,* Jean E. Bemis, Ericka Benson, Pradeep Bista, David Carney, Richard Fahrner,
Diana Lee, Feng Liu, Pallavi Lonkar, Jill C. Milne, Andrew J. Nichols, Dominic Picarella, Adam
Shoelson, Jesse Smith, Amal Ting, Allison Wensley, Maisy Yeager, Michael Zimmer and Michael
R. Jirousek.
Catabasis Pharmaceuticals, One Kendall Square, Suite B14202, Cambridge, MA 02139.
KEYWORDS: niacin, omegaꢀ3 fatty acid, EPA, fatty acid niacin conjugate, SREBP, salicylic
acid, DHA, fatty acid salicylate conjugate, NFꢀκB.
ABSTRACT: This report describes the synthesis and preliminary biological characterization of
novel fatty acid niacin conjugates and fatty acid salicylate conjugates. These molecular entities
were created by covalently linking two bioactive molecules – either niacin or salicylic acid to an
omegaꢀ3 fatty acid. This methodology allows the simultaneous intracellular delivery of two
bioactives in order to elicit a pharmacological response that could not be replicated by
administering the bioactives individually or in combination. The fatty acid niacin conjugate 5
has been shown to be an inhibitor of the sterol regulatory element binding protein (SREBP), a
key regulator of cholesterol metabolism proteins such as PCSK9, HMGꢀCoA reductase, ATP
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
citrate lyase and NPC1L1. On the other hand, the fatty acid salicylate conjugate 11 has been
shown to have a unique antiꢀinflammatory profile based on its ability to modulate the NFꢀκB
pathway through the intracellular release of the two bioactives.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
INTRODUCTION
In recent years, the traditional drug discovery paradigm of optimizing a lead candidate toward a
single therapeutic target has been challenged by the successful implementation of a number of
cases involving polypharmacology. This approach, sometimes referred to as network or pathway
1
,2
pharmacology, involves the targeting of multiple therapeutic targets in order to achieve a
desired pharmacological outcome. Imatinib, for instance, was successfully developed as a kinase
3
inhibitor of BCRꢀABL, cꢀKIT, and PDGFRs for the treatment of chronic myeloid leukemia.
Since then, numerous other kinase inhibitors have been introduced against a vast array of
4
therapeutic targets. With these kinase inhibitors, pathway pharmacology was used to rationally
target a small subset of kinases known to be upꢀregulated in certain forms of cancer. Outside of
the oncology field, pathway pharmacology has also been used in the design of the antiꢀpsychotic
agent ziprasidone and the bronchodilator GSK 961081. With ziprasidone, two small
pharmacophores were merged into one single molecular entity in order to modulate multiple 5ꢀ
5
HT and dopamine receptors (5ꢀHT , 5ꢀHT , 5ꢀHT , 5ꢀHT , α , D and D receptors). With
2
A
2c
6
7
1
2
3
GSK 961081, a nonꢀhydrolysable linker was used to covalently join a muscarinic antagonist with
6
a β receptor agonist in order to modulate two distinct GPCR targets. In this report, we describe
2
a novel method of exploring pathway pharmacology involving the simultaneous intracellular
delivery of two bioactives. With this methodology, the two bioactives are first joined covalently
ACS Paragon Plus Environment
2

Page 3 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
by the use of a plasmaꢀstable linker. This process can provide a high level of extracellular
selectivity since the two molecules have been rendered essentially inactive by the covalent
linking process. However, once delivered inside cells, intracellular enzymes degrade the linker
and thereby release the two bioactives, allowing them to exert their effect on the different
therapeutic pathways or targets. Our conjugation methodology allows us to gain synergy by
targeting multiple pathways in the same cells at the same time. Furthermore, it enables us to
solve a difficult problem that arises with the simple coꢀadministration of any two bioactive
molecules, namely, the different PK and tissue distribution profile of the two bioactives. When
the two bioactive molecules are covalently linked, however, they are delivered inside cells at the
same time in equimolar concentrations. What we have observed is that the pharmacological
effect that could be produced by the simultaneous delivery of the bioactives is unique and cannot
not be replicated by administering the individual components or a combination of the individual
components. We will demonstrate the utility of our methodology with two examples. The first
example involves the use of fatty acid niacin conjugates in the area of lipid synthesis; and the
second example involves the use of fatty acid salicylate conjugates in the area of NFꢀκB
inhibition to counteract inflammation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
RESULTS AND DISCUSSION
Fatty acid niacin conjugates are molecular entities created by covalently linking niacin to an
omegaꢀ3 fatty acid, namely (5Z,8Z,11Z,14Z,17Z)ꢀeicosaꢀ5,8,11,14,17ꢀpentaenoic acid (EPA).
7
ꢀ9
10
Both niacin and EPA are wellꢀknown lipid lowering agents. EPA, in particular, has been
shown to inhibit sterol regulatory element binding proteins (SREBPs) activity in a PPARαꢀ
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
11,12
independent manner.
The SREBP pathway, in turn, is critical in modulating fatty acid
1
3
synthesis and maintaining cholesterol homeostasis. SREBPꢀ1 and SREBPꢀ2 are the two
different forms of SREBP proteins. SREBPꢀ1 is known to regulate production of genes involved
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
14
in fatty acid synthesis and oxidation such as FASN and SCDꢀ1. SREBPꢀ2, on the other hand,
controls expression of genes involved in cholesterol homeostasis, which include HMGꢀCoA
15
reductase, NPC1L1 and PCSK9. We hypothesized that if both niacin and EPA could be
delivered together inside cells, a number of lipid synthesis pathways could be simultaneously
affected to produce a unique biological response. In order to achieve simultaneous delivery to
cells, the two drugs need to be covalently linked to form a single molecular conjugate. The
resulting molecular conjugate also needs to be sufficiently stable in the plasma to remain intact
during systemic circulation. Once delivered inside targeted cells, however, intracellular enzymes
hydrolyze this molecular conjugate down to the individual components. As a fatty acid, when
EPA is covalently linked to other moieties via an amide or ester bond, those bonds could be
1
6
processed by intracellular enzymes such as fatty acid amide hydrolase (FAAH), Nꢀ
17
18
acylethanolamine hydrolyzing acid amidase (NAAA), or monacylglycerol lipase (MAGL).
Figure 1 lists the five possible fatty acid niacin conjugates (1ꢀ5), each with a different linker that
enables the covalent linkage of niacin to EPA. Compounds 1ꢀ4 were prepared by the
straightforward steps outlined in the Supporting Information Section. The diamide derivative 5
was prepared according to the general Scheme 1. Nicotinic acid was reacted with tertꢀbutyl (2ꢀ
aminoethyl)carbamate in the presence of EDC/HOBT and Et N to afford tertꢀbutyl (2ꢀ
3
(
nicotinamido)ethyl)carbamate. Upon treatment with HCl in dioxane, the HCl salt of Nꢀ(2ꢀ
aminoethyl)nicotinamide was obtained. This was then coupled with the fatty acid EPA using
EDC/HOBT and Et N to afford compound 5. This general Scheme 1 was also used to prepare a
3
ACS Paragon Plus Environment
4

Page 5 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
number of analogs that are pertinent to this report, namely the fatty acid niacin conjugates 7ꢀ10
and the fatty acid salicylate conjugates 11 and 12.
Compounds 1ꢀ5 were first assessed for plasma stability. As shown in Figure 2, the diamide
derivative 5 was the only analog among the 5 tested that showed complete plasma stability
across all 4 species. Even at the two hour time point, essentially all of the parent compound was
still present. We then investigated the cellular uptake of these fatty acid niacin conjugates using
confocal microscopy. In order to facilitate the imaging process, we replaced the omegaꢀ3 fatty
acid EPA in 5 with the fluorescent fatty acid cisꢀparinaric acid in order to form the corresponding
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
19
fatty acid niacin conjugate 10. HuHꢀ7 cells were treated with either cisꢀparinaric acid (25 ꢁM)
or the fluorescent fatty acid niacin conjugate 10 (25 ꢁM) for 30 min, fixed with 4%
formaldehyde and mounted with antiꢀfade DAPI. Figure 3A shows that the free fatty acid (i.e.
cisꢀparinaric acid) does get into cells; but it is highly concentrated in the lysosomes, and is not
distributed to the endoplasmic reticulum (ER). In sharp contrast, the fatty acid niacin conjugate
10 exhibits an ERꢀlike (periꢀnuclear) pattern, with less distribution to the lysosomes (Figure 3B).
Endocytosis is believed to be the cellular uptake mechanism for these fatty acid niacin
conjugates. A preꢀincubation experiment with Dynasore, a known endocytosis inhibitor, was
carried out. As shown in Figure 3C, treatment with 80 ꢁM of Dynasore abolished the ER
cellular localization of compound 10, indicating that these fatty acid niacin conjugates were
2
0
taken up by cells by active endocytosis, likely via clathrinꢀcoated vesicles. This uptake
mechanism for the fatty acid niacin conjugates is distinct from what has been reported with free
fatty acids. When unsaturated omegaꢀ3 fatty acids, such as EPA, are applied to cells, they are
usually incorporated first into the phospholipids and then released intracellularly by the rateꢀ
21
limiting action of phospholipase A1 (PLA1) and phospholipase A2 (PLA2). Since the fatty
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
acid niacin conjugates are taken up inside cells by active endocytosis, their intracellular
distribution to the ER is different; and so is the biological response that they could elicit.
Distribution of the fatty acid niacin conjugate to the ER also allows it to be localized in the same
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
22
subcellular compartment as the enzyme FAAH. This would allow FAAH to hydrolyze the
fatty acid niacin conjugate and release the individual components. The fatty acid niacin
conjugate 5 was then evaluated in a number of assays to see if it could be hydrolyzed in targeted
tissues down to the individual components. The rat liver lysate experiment was carried out
initially to examine the hydrolysis of compound 5. This hydrolysis experiment was carried out
°
by incubating compound 5 in rat liver lysate at 37 C and monitored for the disappearance of the
starting material. As shown in Figure 4, there was a timeꢀdependent hydrolysis of compound 5
in this assay. Furthermore, this hydrolysis is FAAHꢀdependent since addition of the FAAH
23
inhibitor PFꢀ3845 almost completely abrogated the hydrolysis of compound 5. In humans, two
forms of FAAH, FAAH1 and FAAH2, share 20% sequence homology and have different tissue
24
distribution profiles. We expressed and purified recombinant FAAH1 enzyme to determine its
role in hydrolysis of compound 5. As shown in Figure 5, recombinant FAAH1 hydrolyzed the
amide bond between the fatty acid and the linker and released EPA in a timeꢀdependent fashion.
Again, the addition of the FAAH inhibitor PFꢀ3845 also inhibited this particular hydrolysis.
The hydrolysis of compound 5 and the formation of intracellular EPA were further confirmed in
a cellꢀbased assay using HepG2 cells. As shown in Figure 6, incubation of compound 5 in
HepG2 cells resulted in a timeꢀdependent hydrolysis. In this cell line, formation of intracellular
EPA was monitored by using a deuteriumꢀlabeled EPAꢀd , in order to distinguish the fatty acid
5
that was being released from compound 5 over the endogenous EPA. Compound 9, the EPAꢀd5
version of compound 5, was prepared according to the same procedure outlined previously in
ACS Paragon Plus Environment
6

Page 7 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 1 using the appropriate deuteriumꢀlabeled fatty acid. The requisite EPAꢀd was prepared
5
2
5
by total synthesis using the reaction sequence outlined by Sandri and Viala. As shown in Figure
, there was a timeꢀdependent rate of formation of EPAꢀd during the hydrolysis of compound 9
7
5
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
in human HepG2 cell lines.
The assays described above represent some initial steps aimed at understanding the hydrolysis of
compound 5. The actual hydrolysis of compound 5 in an in vivo setting is most likely much
more complex due to the large arrays of amidases and hydrolases that are available across
various tissues and species. Thus, it is more appropriate to evaluate the oral bioavailability,
tissue distribution and hydrolysis of compound 5 in an actual animal model. When niacin is
administered to rodents, dogs, monkeys or humans, one of the major plasma metabolites is
2
6
nicotinuric acid. In a typical PK experiment where niacin was administered as a single oral
dose of 30 mg/kg to Sprague Dawley rats (plasma collection at t = 0.25, 0.5, 1, 2, 4 and 6 hr) the
Cmax and AUClast of niacin were 7467 ± 2246 ng/mL and 6980 ± 3173 hr*ng/mL respectively. In
the same PK experiment, the Cmax and AUClast of nicotinuric acid were 369 ± 80 ng/mL and 466
±
359 hr*ng/mL respectively. Both niacin and its nicotinuric acid metabolite were cleared fairly
rapidly and neither one was detected in the plasma after 2 hours. When the fatty acid niacin
conjugate 5 was dosed at 100 mg/kg po (i.e. the same molar equivalent as 30 mg/kg of niacin) a
low level of the parent compound was detected in the plasma (for compound 5, the Cmax and
AUClast were 212 ± 74 ng/mL and 172 ± 67 hr*ng/mL respectively). Interestingly, even though
niacin was not detected in the plasma, the molar equivalent dose of compound 5 produced much
higher plasma levels of nicotinuric acid (Cmax and AUC were 1131 ± 306 ng/mL and 3484 ±
last
1013 hr*ng/mL respectively). Furthermore, this type of fatty acid niacin conjugate allowed the
delivery of nicotinuric acid over a much longer period; at the 6 hour time point, a significant
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
level of nicotinuric acid could still be detected in the plasma (>200 ng/mL). Compound 5 was
then evaluated in a PK/tissue distribution study in order to further verify that the fatty acid niacin
conjugate could be hydrolyzed down to the individual components in targeted tissues. Plasma
and tissue samples were processed and analyzed by LC/MS/MS for the presence of 1) the parent
compound, 2) the niacin linker metabolite 13, 3) the linkerꢀEPA metabolite 17, 4) niacin and 5)
nicotinuric acid 16 (Figure 8). As shown in Table 1, low concentration of the parent compound
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
was detected in the plasma and liver. A significant quantity of the parent compound was,
however, present in the intestine (6,747 ng/g). The desired hydrolysis of compound 5 did indeed
take place and the key nicotinuric acid metabolite was detected not only in the plasma, but also
in both tissues (nicotinuric acid liver concentration = 1356 ng/g; nicotinuric acid intestine
concentration = 1387 ng/g). The hydrolysis of the amide bond between EPA and the linker
portion also occurred in targeted tissues and the niacin linker metabolite was detected not only in
plasma but also in both tissues (liver concentration of niacin linker metabolite = 10,910 ng/g;
intestine concentration of niacin linker metabolite = 23,899 ng/g). Lower concentration of the
linker EPA metabolite was detected in the plasma and the two isolated tissues. Even though
niacin itself was not detected in the plasma and liver tissue, it was detected in intestinal tissues
(concentration = 279 ng/g). This tissue distribution study showed that compound 5 was
extensively hydrolyzed in the intestine and a lower level of the intact parent compound was
delivered to the liver. Changing the nature of the linker group could also change the rate of
hydrolysis of the corresponding fatty acid niacin conjugate and its tissue distribution
considerably. In compounds 7 and 8, the ethylenediamine linker has been replaced with (S)ꢀ
1
pyrrolidinꢀ3ꢀamine and N ꢀmethylethaneꢀ1,2ꢀdiamine respectively. This substitution should
render the amide bond between the diamine linker and the fatty acid more resistant to hydrolysis
ACS Paragon Plus Environment
8

Page 9 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
by FAAH. Figure 9 shows the hydrolysis of compounds 5, 7 and 8 using the recombinant
FAAHꢀ1. Compound 7 was more resistant to FAAHꢀ1 hydrolysis, when compared to compound
5
. For compound 8, there was essentially no hydrolysis of the fatty acid amide bond under the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
assay conditions. Since the fatty acid conjugates 7 and 8 are more resistant to hydrolysis, more
of the intact parent compound should be delivered to the liver. PK studies using doubleꢀ
cannulated rats were carried out to compare the amount of parent compound that could
potentially be delivered to the liver. For these PK studies, Sprague Dawley rats that have been
surgically implanted with indwelling jugular vein cannula (JVC) and portal vein cannula (PVC)
were used; and serial blood collection was carried out at both the portal and jugular vein (t = 10,
20, 40 min and 1, 2, 4 and 6 hrs post dose). Table 2 summarizes selected PK parameters for
2
7
compounds 5, 7 and 8 in this double cannulated rat PK study. Consistent with the tissue
distribution study, we saw that compound 5 was extensively hydrolyzed in the intestine, as
indicated by the higher portal concentration of the niacinꢀlinker metabolite 13 (Cmax = 2,061 ±
433 ng/mL) and a lower portal concentration of the parent compound (Cmax = 418 ± 122 ng/mL).
On the other hand, with the more hydrolysisꢀresistant compound 7 and 8, we observed less
metabolism of the parent compound in the intestine; and as a result, a higher portal concentration
of the parent compound was observed along with a lower portal concentration of the niacin
linker metabolite. Compared to compound 5, a higher level of the parent compound was also
observed in the peripheral circulation for the more hydrolysisꢀresistant compounds 7 and 8
(peripheral Cmax = 564 ± 226 and 400 ± 100 ng/mL respectively). Compound 8 was then
evaluated in a rat tissue distribution study where animals (n = 6) were dosed with 100 mg/kg q.d.
for 3 days. Four hours after the last dose, plasma and tissues were collected and analyzed for the
28
parent compound 8 along with the metabolites 15, 16, and 19. Even though compound 8 was
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
dosed at a lower dose of 100 mg/kg, there was a significantly greater concentration of the
corresponding parent compound (> 60 fold increase) in the liver, when compared with compound
5
(Figure 10). This result demonstrated the dramatic shift in the tissue distribution of the intact
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
parent compound upon decreasing the rate of hydrolysis of the amide bond between the fatty
acid EPA and the linker.
Since the components are only released intracellularly with the fatty acid niacin conjugate 5, our
approach could potentially allow one to avoid the unpleasant flushing side effect of niacin. It is
widely known that niacin is capable of binding to and activating GPR109A, a Giꢀcoupled GPCR.
This process, in turn, promotes the release of prostaglandin D (PGD ) from skin cells. This
2
2
eicosanoid is a vasodilator that is responsible for the cutaneous flushing. Our analyses
demonstrated that the fatty acid niacin conjugate 5 did not interact with the GPR109A receptor.
Interaction of conjugate 5 with the GPR109A receptor, was tested using an adenylate cyclase
2
9
inhibition assay in human epidermal A431 cell line that expressed GPR109A. One of the
activities of niacin binding to GPR109A is the inhibition of adenylate cyclase. Adenylate cyclase
converts ATP to cAMP and this activity is stimulated by forskolin. As shown in Figure 11,
niacin inhibited forskolinꢀmediated cAMP production in a doseꢀdependent manner with an EC50
of 0.73 ꢁM. In this assay, niacin was able to inhibit approximately 50% of the cAMP signal
produced by forskolin. In sharp contrast, compound 5 did not functionally activate GPR109A.
No significant inhibition of cAMP production was observed when these epidermal cells were
incubated with compound 5 over the same dose range. In essence, the molecules that are being
covalently linked have been rendered inactive in the extracellular space. Covalently linking
niacin to a fatty acid moiety using the plasma stable linker described above thus represents a
potential new way of delivering the niacin component without inducing the flushing side effects.
ACS Paragon Plus Environment
10

Page 11 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
After having confirmed that the fatty acid niacin conjugate 5 could undergo the appropriate
hydrolysis to afford the desired components in targeted tissues, we then focused our attention on
the biological effect that it could induce. The SREBP proteins have been known to be essential
for cholesterol biosynthesis as well as for the uptake and biosynthesis of various fatty acids.
HMGꢀCoA reductase and PCSK9 are two of the target genes that are known to be regulated by
the mature form of the SREBPꢀ2. Both EPA and niacin, at very high concentration, could affect
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
1,12
SREBPꢀ2 activity.
At the lower concentration of 50 μM, however, both EPA and niacin,
showed little activity on SREBPꢀ2 maturation when they were incubated in human HepG2 cells.
In sharp contrast, when the fatty acid niacin conjugate 5 was incubated in human HepG2 cells,
there was a decrease in the production of mature SREBPꢀ2 protein. The effect was dose
dependent and the decrease in SREBPꢀ2 protein could be observed at the 25 and 50 μM
concentration (Figure 12). Statins have previously been shown to increase plasma level of
3
0ꢀ32
PCSK9 through its effect on LDL levels and SREBPꢀ2.
In this cell assay, the effect of
administering atorvastatin on SREBPꢀ2 was also quite pronounced. As shown in Figure 12,
when human HepG2 cells were incubated with 10 μM of atorvastatin, there was a significant
increase in the production of mature SREBPꢀ2 protein. However, treatment of human HepG2
cells with a combination of 10 μM of atorvastatin with 25 μM of the fatty acid niacin conjugate 5
resulted in a significant decrease in the production of mature SREBPꢀ2 protein. Since PCSK9 is
an immediate downstream gene of SREBPꢀ2, a decrease in the production of the SREBPꢀ2
protein would naturally lead to a decrease in the expression of PCSK9. A doseꢀdependent
decrease in the expression of PCSK9 mRNA levels in HepG2 cells was observed when cells
were treated with compound 5 is (Figure 13). This effect on the reduction of SREBPꢀ2
maturation and reduced expression of PCSK9 mRNA should translate to a corresponding effect
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
on the secretion of PCSK9. When human HepG2 cells are treated with compound 5, the
resulting cell culture supernatants can be analyzed for secreted PCSK9 levels using a humanꢀ
specific PCSK9 ELISA kit. The results from this type of assay are shown in Figure 14A and
they clearly demonstrated the unprecedented synergistic effect of compound 5 on secreted
PCSK9 levels in human HepG2 cells. In this assay, EPA did not produce any effect on secreted
PCSK9 levels at the highest tested concentration of 50 μM (when cells were incubated with EPA
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
at doses that were greater than 50 ꢀM, cellular toxicity was observed). Niacin, likewise, did not
produce any noticeable effect at doses as high as 10 mM. The combination of EPA and niacin
also did not produce any effect on secreted PCSK9 levels at the highest tested concentration of
50 ꢀM. In sharp contrast with the individual components or the combination of the individual
components, the fatty acid niacin conjugate 5 showed a synergistic inhibition on secreted PCSK9
and had an IC value of 17 ꢀM. In addition to its effects on PCSK9, high concentrations of
5
0
niacin have been reported to inhibit the secretion of Apolipoprotein B (ApoB). ApoB is the
major apolipoprotein found in VLDL and low density lipoprotein (LDL) particles secreted from
the liver into circulation. An assay was established to study the effects of Compound 5 on ApoB
secretion from the human hepatocyte line, HepG2. HepG2 cells were incubated with Compound
5, EPA, niacin or EPA plus niacin. After 18 hours, the amount of ApoB present in the cell
culture supernatant was measured using a commercially available ELISA kit. As shown in
Figure 14B, compound 5 significantly inhibited ApoB secretion in a dose dependent manner at a
significantly lower concentrations than its component parts. (Figure 14B, IC of 27 ꢁM). These
50
data demonstrate a synergistic activity of compound 5 as compared to the combination of EPA
and niacin in this assay. The measure of cellular toxicity by Alamar Blue was not significantly
different across all treatment groups
ACS Paragon Plus Environment
12

Page 13 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
33
Compound 5 was evaluated orally in the ApoE*3ꢀLeiden mice for its ability to lower PCSK9
and other lipid parameters. These transgenic mice were generated using a genomic construct
isolated from human ApoE*3 Leiden, a genetic mutation that is characterized by familial
dysbetalipoproteinemia. ApoE*3 Leiden mice, when fed on a high cholesterol Westernꢀtype
diet, have been shown to have a dramatic increase in total plasma cholesterol and triglyceride
levels. Compound 5 was dosed over a 16ꢀweek period in order to assess plasma PCSK9, total
cholesterol, and triglyceride levels, liver LDLꢀreceptor protein level, lipoprotein profile as well
as other atherosclerotic endpoints. At the conclusion of the study, ApoE*3ꢀLeiden mice that
have been treated with compound 5 showed a significant reduction in PCSK9 levels, LDL
particles including VLDL and LDL cholesterol, as well as plasma triglycerides. The entire data
set for this ApoE*3ꢀLeiden study is presented in a separate communication, along with a more
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
34
detailed discussion on the mechanism of action of compound 5.
In addition to niacin, we have also explored the possibility of coꢀdelivering salicylic acid and an
omegaꢀ3 fatty acid to cells using an analogous linker system. The fatty acid salicylate conjugate
1
1 was prepared by covalently linking salicylic acid to a different omegaꢀ3 fatty acid, namely
(4Z,7Z,10Z,13Z,16Z,19Z)ꢀdocosaꢀ4,7,10,13,16,19ꢀhexaenoic acid (DHA). The same
ethylenediamine linker used in the preparation of the fatty acid niacin conjugate 5 was employed
in this particular example. By itself, salicylic acid, has been reported to be an inhibitor of NFꢀ
3
5ꢀ37
κBꢀmediated gene expression by suppressing IKK kinase activity.
On the other hand, the
omegaꢀ3 fatty acid DHA has been shown to inhibit NFꢀκB through multiple mechanisms. DHA
has been shown to inhibit IκBα phosphorylation by binding to GPR120 and repartitioning TAB1
3
8
away from the TLR4 receptor that mediates the LPS signal. DHA has also been shown to
induce cytoplasmic retention of proꢀinflammatory NFꢀκB proteins through p105 overexpression,
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
39
thereby favoring the nuclear translocation of the p50ꢀp50 transcriptional repressor homodimer.
Additionally, DHA can be converted to the cyclopentanone ring containing A and J ꢀ
4
4
4
0,41
neuroprostanes that are analogous in structure to PGJ2.
PGJ2 has been shown to directly
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
bind cysteine residues in the DNA binding domains of both the p50 and p65 subunits of NFkB,
4
2,43
preventing them from binding to DNA.
The simultaneous intracellular delivery of salicylic
acid and DHA would therefore allow the modulation of the NFꢀκB axis via multiple pathways.
We have evaluated the fatty acid salicylate conjugate 11 in a number of cellular assays and have
observed a pattern of cellular uptake and hydrolysis that is similar to that observed previously
with the fatty acid niacin conjugate 5. In order to assess the cellular uptake of fatty acid salicylate
conjugates by confocal microscopy, we have prepared the fatty acid conjugate 12, wherein the
DHA portion has been replaced with the fluorescent cisꢀparinaric acid. Figure 3D shows that the
fatty acid salicylate conjugates are taken up into cells and are coꢀlocalized in the ER
compartment, similar to what was previously observed with the fatty acid niacin conjugate 5.
Distribution of the fatty acid salicylate conjugate 11 to the ER also facilitated its hydrolysis by
FAAH. We have studied the hydrolysis of 11 in a rat liver lysate assay and have observed that
its hydrolysis was FAAHꢀdependent. As shown in Figure 15, the addition of the FAAH inhibitor
PFꢀ3845 almost completely abrogated the hydrolysis of 11.
The hydrolysis of the fatty acid salicylate 11 has also been studied in vivo in the appropriate PK
and tissue distribution studies. When salicylic acid is dosed orally to rodents, dogs, monkeys and
44
human, salicyluric acid 21 is detected as a major metabolite (Figure 16). This observation is
consistent with the fatty acid salicylate conjugate 11 being hydrolyzed intracellularly in targeted
tissues down to the individual components (i.e. salicylic acid), resulting in salicyluric acid as one
of the major metabolites. In a PK experiment, compound 11 was dosed orally to Sprague
ACS Paragon Plus Environment
14

Page 15 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Dawley rats at 300 mg/kg and plasma was collected at t = 0, 0.5, 1, 2, 4, 8 and 24 hr for analysis
of the parent compound the potential metabolites shown in Figure 16. Compound 11 was orally
bioavailable and Cmax for the parent compound was 477 ± 193 ng/mL, along with an AUClast of
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
945 ± 1027 Hr*ng/mL. Even though the linkerꢀDHA metabolite 22 could not be detected in the
plasma, the salicylateꢀlinker metabolite 20 was detected in the plasma (Cmax = 484 ± 80 ng/mL;
AUClast = 2611 ± 501 Hr*ng/mL). In this oral rat PK study, we also observed the presence of
salicylic acid (Cmax = 75 ± 125 ng/mL; AUClast = 229 ± 132 Hr*ng/mL) which indicated that the
amide bond between salicylate and the ethylenediamine linker is degradable in cells and targeted
tissues. Concurrent with the release of salicylic acid, we also observed the presence of the
salicyluric acid metabolite 21 (Cmax = 563 ± 76.8 ng/mL; AUClast = 2937 ± 493 Hr*ng/mL).
Compound 11 was then evaluated in a rat tissue accumulation study over a 21 day period. In this
study, male Sprague Dawley rats, in groups of 4, were dosed orally with 300 mg/kg of
compound 11 once a day for 14 days. Animals were sacrificed on days 1, 3, 7 and 14; plasma
and tissues were collected 4 hr after dosing in order to determine the levels of the parent
compound 11, salicylic acid, and salicyluric acid 21. The tissues examined included brain,
epididymal fat, liver and muscle. Another group of animals was dosed orally with 300 mg/kg of
compound 11 for 14 days and then subjected to a 7ꢀday washout cycle; after which plasma and
tissues were collected on day 21. As shown in Figures 17A, steady state levels of the parent
compound were achieved in the various tissues and plasma after 7 days of dosing with the fatty
acid salicylate conjugate 11. Higher concentration of the parent compound was detected in the
tissues than in the plasma. Among the tissues that were examined, epididymal fat contained the
highest tissue concentration of the parent compound 11 (concentration = 771.96 ± 176.92 ng/g).
This was followed by the muscle and liver, wherein the tissue concentration of the parent
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
compound 11 was 425 ± 88.87 ng/g and 330.74 ± 118.82 ng/g respectively. There was no
significant level of the parent compound 11 measured in the brain tissue at any of the time points
tested. The corresponding plasma concentration of the parent compound 11 at this 7ꢀday time
point was 115.51 ± 28.45 ng/mL. After 14 days of dosing, animals were subjected to a 7ꢀday
washout cycle. Our data indicate that the parent compound 11 was completely cleared out of
plasma and tissues within 7 days (Figure 17A).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Complete hydrolysis of the fatty acid salicylate conjugate 11 occurred and salicylic acid was
detectable both in the plasma and tissues. Figure 17B shows the plasma and tissue concentration
of salicylic acid over the course of the 21 day study. Steady state levels of salicylic acid were
achieved after 7 days of dosing with 11 and a plasma concentration of 154.20 ± 65.43 ng/mL
was observed. During this period, steady state levels were also achieved in the tissues, and a
higher concentration of salicylic acid was observed in the liver (235.91 ± 95.14 ng/g) and muscle
(
265.73 ± 73.0 ng/g). Low levels of salicylic acid were detected both in the epididymal fat
65.03 ± 65.02 ng/g) and in the brain (53.32 ± 26.95 ng/g).
(
Salicyluric acid is a phase 2 metabolite formed through the intracellular conjugation of salicylic
acid and glycine. The levels of the metabolite salicyluric acid achieved a relatively constant state
after 7 days of dosing across the tissues sampled (Figure 17C). As with salicylic acid, the
highest concentrations were observed in plasma (764.41 ± 21.98 ng/mL) and liver tissues
(506.48 ± 82.44 ng/g), with lower salicyluric acid measured in fat (203.71 ± 80.19 ng/g), muscle
(165.21 ± 14.0 ng/g) and brain tissues (114.43 ± 8.87 ng/g). As observed previously with the
parent compound 11, both salicylic acid and salicyluric acid were completely cleared from the
plasma and tissues after a 7ꢀday washout cycle.
ACS Paragon Plus Environment
16

Page 17 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Having confirmed that the individual components could be released intracellularly, we then
turned our attention to the pharmacological response that could be produced by the fatty acid
salicylate conjugate 11. In an NFꢀκB reporter assay, compound 11 showed significantly more
potent and synergistic inhibition of NFꢀκB than salicylic acid or DHA, its two major
components. As shown in Figure 18, salicylic acid only showed inhibition of NFꢀκB at high
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
5
concentration of 3.2 mM, which is consistent with the previously reported values. Likewise,
DHA could only modestly inhibit NFꢀκB at a concentration of 78.7 ꢁM. In sharp contrast,
compound 11 was able to inhibit NFꢀκB with an IC value of 25.4 ꢁM. This synergistic
50
inhibition of NFꢀκB could not be replicated with the individual components or a combination of
the individual components, even at much higher concentration. Inhibition of NFꢀκB could have
many potential therapeutic applications, particularly in the area involving the enhancement of
46,47
muscle regeneration in injured and diseased skeletal muscle.
Serum levels of many NFꢀkB
4
8
49
dependent proꢀinflammatory chemokines including MMPꢀ9, and cytokines including TNFꢀα,
have been reported to be elevated relative to controls in boys with Duchenne muscular
dystrophy, underscoring the potential utility of NFꢀκB modulation in this disease. We have since
evaluated compound 11 in a number of preclinical models of Duchenne muscular dystrophy.
These included the mdx mouse model and the golden retriever model. Detailed discussion of the
5
0
results from those studies can be found in an upcoming communication.
In summary, we have described a novel method of exploring pathway pharmacology through the
use of fatty acid niacin and fatty acid salicylate conjugates. In our methodology, conjugates
comprised of two covalentlyꢀlinked bioactives are activated by specific intracellular enzymes in
order to release the two bioactives inside cells and thereby allow them to simultaneously
modulate multiple therapeutic pathways. Compared to the simple coꢀadministration of the two
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
bioactives, our conjugation methodology offers a number of advantages such as extracellular
selectivity and mechanistic synergy by the simultaneous delivery of the two bioactives to the
various intracellular targets. Both fatty acid conjugates of niacin and salicylate were able to
produce a pharmacological effect that could not be replicated by administering the individual
components or a combination of the individual components. For the fatty acid niacin conjugate
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
, a distinctive effect on SREBP inhibition was observed, along with a downꢀregulation of
cholesterol and other lipid metabolism genes, including PCSK9. For the fatty acid salicylate
conjugate 11, the intracellular release of a salicylic acid and a DHA molecule produced an
unprecedented and synergistic inhibition of NFꢀκB. The fatty acid niacin conjugate 5 has since
been advanced into human clinical trials in order to evaluate its effect on the various lipid
parameters. The fatty acid salicylate conjugate 11 has also been advanced into the clinic and is
currently being evaluated in patients with Duchenne muscular dystrophy. The results from those
clinical trials will be disclosed in due course.
EXPERIMENTAL SECTION
General information: All chemical reagents and solvents were commercially available and used
as received. Fatty acids and acid chlorides were purchased from NuꢀChek Prep. Reactions were
1
13
generally run under argon or nitrogen. H NMR (400 MHz) and C NMR (100 MHz) spectra
were recorded on a Varian 400MHz Unity Inova system in CDCl , unless otherwise noted.
3
Chemical shifts are expressed in parts per million (ppm, δ); coupling constants are in hertz (Hz).
Splitting patterns describe multiplicities s (singlet), d (double), t (triplet), q (quartet), m
(multiplet), br (broad). High resolution mass spectroscopy data was analyzed by direct flow
ACS Paragon Plus Environment
18

Page 19 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
injection, utilizing electrospray ionization (ESI) on a Waters Qtof API US instrument in the
positive mode. Analytical and other mass spectra were collected on an Agilent Technologies
1200 series system with an Agilent Technologies 6120 Quadruple LCꢀMS detector in positive
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
mode. A SiliCycle C18 XDB,3 x 100 mm column was used with a gradient of H O and
2
acetonitrile each with 0.1% formic acid; UV detection at 254 nm and 210 nm. Normal phase
flash chromatography was accomplished on Teledyne Isco systems using preꢀpacked silica gel
columns. Sample purity was determined by LCꢀMS; all compounds were of >95% purity, as
determined by at least two different HPLC methods. For the hydrolysis experiments and
PK/tissue distribution study, the samples were analyzed on the Agilent 6410 Triple Quad LC/MS
mass spectrometer. Separation was achieved using a Gemini 3 ꢁM C6 phenyl column with a
gradient using water containing 0.1% formic acid (solvent A) and methanol containing 0.1%
formic acid (solvent B).
General procedure used to prepare Nꢀ(2ꢀ((5Z,8Z,11Z,14Z,17Z)ꢀicosaꢀ5,8,11,14,17ꢀ
pentaenamido)ethyl)nicotinamide (5) and other fatty acid conjugates (6ꢀ12): In a typical run,
a mixture containing tertꢀbutyl (2ꢀaminoethyl)carbamate (1 g, 6.24 mmol), niacin (730 mg, 5.93
mmol), EDC (1.45 g, 7.50 mmol), HOBT (1.25 g, 7.50 mmol) and Et N (1.8 mL, 12.50 mmol) in
3
2
0 mL CH Cl was stirred at rt for 18 h. The resulting reaction mixture was washed with halfꢀ
2 2
saturated NH Cl, brine, dried (Na SO ) and concentrated under reduced pressure. Purification
4
2
4
by silica gel chromatography (gradient elution, CH Cl to 10% MeOH in CH Cl ) afforded tertꢀ
2
2
2
2
butyl (2ꢀ(nicotinamido)ethyl)carbamate (1.4 g, 85%). This material was taken up in 5 mL of
THF along with 8 mL of 4 N HCl in dioxane (32 mmol). The resulting reaction mixture was
stirred at rt for 8 h and then concentrated under reduced pressure to afford the HCl salt of Nꢀ(2ꢀ
aminoethyl)nicotinamide in essentially quantitative yield. This material was then taken up in 20
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
mL of CH Cl along with (5Z,8Z,11Z,14Z,17Z)ꢀeicosaꢀ5,8,11,14,17ꢀpentaenoic acid (>95%
2
2
purity, 1.5 g, 5.04 mmol), EDC (1.22 g, 6.36 mmol), HOBT (1.1 g, 6.36 mmol) and Et N (4.4
3
mL, 31.8 mmol). The resulting reaction mixture was stirred at rt for 16 h. It was then washed
with halfꢀsaturated NH Cl, brine, dried (Na SO ) and concentrated under reduced pressure. The
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
4
2
4
resulting residue was purified by silica gel chromatography (gradient elution, CH Cl to 6%
2
2
MeOH in CH Cl ) to afford 1.68 g of Nꢀ(2ꢀ((5Z,8Z,11Z,14Z,17Z)ꢀicosaꢀ5,8,11,14,17ꢀ
2
2
1
pentaenamido)ethyl)nicotinamide (compound 5, 60% overall yield, >95% purity by LC/MS). H
NMR (400 MHz, CDCl ) δ 9.06 (d, J = 2.3 Hz, 1H), 8.72 (dd, J = 4.9, 1.7 Hz, 1H), 8.11 (dt, J =
3
8
.0, 2.0 Hz, 1H), 7.64 (s, 1H), 7.38 (ddd, J = 7.9, 4.9, 0.9 Hz, 1H), 6.10 (d, J = 6.3 Hz, 1H), 5.46
5.25 (m, 9H), 3.65 – 3.49 (m, 4H), 2.90 – 2.73 (m, 8H), 2.22 (dd, J = 8.4, 6.9 Hz, 2H), 2.14 –
–
1
3
1.98 (m, 4H), 1.70 (p, J = 7.4 Hz, 2H), 0.97 (t, J = 7.5 Hz, 3H); C NMR (100 MHz, DMSOꢀd)
δ 20.0, 25.1, 26.2, 34.9, 38.1, 38.8, 39.1, 39.3, 39.5, 39.7, 39.9, 123.3, 128.9, 127.6, 127.7, 127.8,
+
127.9, 127.9, 128.0, 128.1, 129.4, 129.9, 131.5, 134.8, 148.4, 151.7, 164.9, 172.3; HRMS (ES )
+
Calcd for C H N O (M + H ) m/z 450.3121, found 450.3121.
2
8
40
3
2
The same general experimental procedure was employed for the synthesis of fatty acid salicylate
1
conjugate 11, substituting niacin for salicylic acid. H NMR (400 MHz, CDCl ) δ 12.44 (br s, 1
3
H), 8.01 (m, 1H), 7.52 (dd, J = 1.3, 7.6 Hz, 1H), 7.36 (m, 1H), 6.95 (m, 1H), 6.84 (m, 1H), 6.36
(m, 1H), 5.29ꢀ5.37 (m, 12H), 3.49ꢀ3.53 (m, 4H), 2.72ꢀ2.85 (m, 4H), 2.77ꢀ2.85 (m, 10H), 2.06ꢀ
1
3
2.42 (m, 3H), 0.97 (t, J = 7.5 Hz, 3H). C NMR (100 MHz, DMSOꢀd) δ 14.3, 20.6, 23.4, 25.5,
25.6, 25.7, 36.3, 39.4, 41.6, 76.8, 77.1, 77.4, 114.2, 118.3, 118.9, 126.2, 127.0, 127.7, 127.9,
+
1
28.1, 128.6, 129.7, 132.1, 134.2, 161.5, 170.8, 175.0; High resolution MS (ES ) Calcd for
+
C H N O (M + H ) m/z 491.3274, found 491.3278.
31
42
2
3
ACS Paragon Plus Environment
20

Page 21 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Fluorescent imaging using HuHꢀ7 cells: Human hepatocellular carcinoma cells (HuHꢀ7) were
obtained from Japan Cell Research Bank (JCRB). Two million cells were plated overnight in
10% FBS DMEM on a 4ꢀwell slide chamber (Nunc LabꢀTek II Chamber Slide System, Thermoꢀ
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
scientific). Cells were infected with CellLight ERꢀRFP, BacMam 2.0 (Molecular Probes,
C10591) or CellLight Late EndosomesꢀRFP, BacMam 2.0 (Molecular Probes, C10589) for 48
hours. Cell medium was replaced with DMEM plus compound 10 (prepared in 10% FBS), or
compound 12 for 30 minutes and fixed with 4% paraformaldehyde for 2 minutes at 37 °C and
washed two times with phosphate buffered saline for 5 minutes per wash. Slides were mounted
with ProLong Gold antifade reagent with DAPI (Molecular Probes). Images were taken with a
Zeiss 510 LSM confocal microscope with a 63X lens. Three lasers at 408 nm, 488 nm and 561
nm were used for DAPI, fatty acid conjugates 10 or 12, and ER/lysosome/endosomesꢀtracker
respectively. For the endocytosis imaging, HuHꢀ7 cells were infected with CellLight ERꢀRFP,
BacMam 2.0 (Molecular Probes) for 48 hours. Cells were pretreated with 80 µM dynasore and
then treated with compound 10 for 30 minutes, fixed with 4% formaldehyde for 2 minutes at 37
o
C and mounted with antiꢀfade DAPI (Molecular Probes). Images were taken on a Zeiss 510
LSM confocal microscope with 63X lens.
Plasma stability studies: The in vitro stability of compounds 1ꢀ5 was studied in human, mouse,
beagle and rat plasma. The plasma from the different species was diluted to 50% with 0.05 M
PBS (pH 7.4) at 37 °C. Test compounds were dissolved in DMSO to a final concentration of
1
0mM and then diluted to 1mM in MeOH. The reactions were initiated by the addition of the
test compounds to 500 µL of preheated plasma in a 96ꢀwell plate to yield a final concentration of
µM. The assays were performed at 37 °C and conducted in triplicate. Samples (50 µL) were
5
taken at 0, 30, 60 and 120 min and 200 µL acetonitrile containing an appropriate internal
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
standard (200 ng/mL) was added. The samples were subjected to vortex mixing for 1 min and
then centrifugation at 4°C for 30 min at 4,000 rpm. The clear supernatants were then analyzed by
LCꢀMSꢀMS under electrospray ionization in the positive multiple reactions monitoring (MRM)
mode.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Hydrolysis experiment using rat liver lysate: The rat liver lysate hydrolysis experiment was
performed in Eppendorf tubes, by adding the necessary amount of water for the assay first. A
10x buffer solution with final concentrations of HEPES and EDTA of 50 mM and 1 mM
respectively was prepared from 500 mM HEPES and 10 mM EDTA in water. Rat liver lysate
o
was removed from ꢀ80 C storage, thawed and centrifuged at 4000g for 5 min. A volume of the
supernatant was transferred to the Eppendorf tube yielding a final concentration in the assay of 3
mg/mL rat liver lysate. The tube was gently inverted to mix the components. The reactions are
o
o
then place in a 37 C incubator for 20 min allowing the reaction mixture to reach 37 C before
the reaction was initiated. To initiate the hydrolysis reaction, the samples were removed from
the incubator and the compounds of the invention were added, as a DMSO solution, prepared in
a manner to allow for <1% final DMSO concentration and 5 ꢀM final compound concentration.
The contents were mixed and then at time point = 0, immediately transferred into an acetonitrile
crash buffer for analysis on the LC/MS/MS in the linear range of the analyte of interest. This
process was then repeated for additional time points as needed. This protocol was repeated in the
presence of 5 ꢀM of the FAAH inhibitor PFꢀ3845.
Hydrolysis experiment using FAAH1: FAAH1ꢀ32ꢀ579 (the Nꢀterm transmembrane region
truncated) was cloned into pFastBacI (Invitrogen) with Nꢀterminal FLAG and Cꢀterminal His
tags and then expressed in SF9 insect cells at 27°C for 48 hrs. The cells were harvested and
ACS Paragon Plus Environment
22

Page 23 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
resuspended in lysis buffer (300 mM NaCl, 10% glycerol, 0.1% NPꢀ40, 50 mM TrisꢀHCl, pH 7.8
with protease inhibitor cocktail, Roche), incubated for 30 min at 4 °C and lysed by sonication.
The supernatant was incubated with AntiꢀFLAG M2 Affinity Gel (Sigma) that preꢀequilibrated
with binding buffer (300 mM NaCl, 10% glycerol, 0.1% NPꢀ40, 50 mM TrisꢀHCl, pH 7.8) for
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
.5 hrs at 4 °C. The AntiꢀFLAG M2 Affinity Gel was then put into a column and washed with 20
CVs of binding buffer and then washed with 10 CV of washing buffer (150 mM NaCl, 5%
glycerol, 5 mM ATP, 20 mM TrisꢀHCl, pH 7.8). The FAAH1 protein was eluted with 100 µg/ml
of FLAG peptide in 150 mM NaCl, 5% glycerol, 10 mM TrisꢀHCl, pH 7.4 and dialyzed against
3
00 mM NaCl, 20% glycerol, 2 mM DTT, 25 mM TrisꢀHCl, pH 7.8. The target protein was
concentrated to ~0.5 mg/mL by ultraꢀfiltration tube with the molecular cutoff of 30 kDa and
frozen with liquid nitrogen for storage at ꢀ80 °C. Compound 5 was tested in a hydrolysis
experiment using this recombinant FAAH1 at an enzyme concentration of 1 nM and substrate
concentration of 25 ꢀM. For the hydrolysis portion, compound 5 and enzyme were incubated for
2
hours at 37 °C. For the quantification portion, the samples were analyzed on the Agilent 6410
Triple Quad LC/MS mass spectrometer using the appropriate standard curves. Separation on the
LC/MS was achieved using the general protocols detailed above.
Hydrolysis experiments in HepG2 cells: HepG2 cells (from ATCC, Catalog no. HBꢀ8065)
5
were suspended in PBS (1 × 10 /mL) and preꢀincubated at 37 °C for 30 min. Compound 9 was
added to the reaction tubes at a concentration of 50 ꢀM in DMSO (1% final concentration of
DMSO). Following the addition of compound 9, an aliquot was immediately transferred from
the reaction tube and quenched into a 1.5 mL micro centrifuge tube containing acetonitrile and
an internal standard. All reaction tubes were then returned to the 37 °C incubator. Additional
aliquots were withdrawn at the following time points and processed in the same manner: 30 min,
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
h, 3 h and 4 h. The micro centrifuge tubes containing the quenched reaction were vortexed for
0 sec and then centrifuged for 5 min at 14,800 × g. The supernatant was analyzed by
3
LC/MS/MS for EPAꢀd . The standard curve for EPAꢀd was generated at a concentration range
5
5
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
from 1 to 500 nM in HepG2 cell suspension (1 × 10 /mL). Analytically pure EPAꢀd is
5
commercially available from Caymen Chemicals.
Assay used to assess activation of GPR109A: The human epidermoid A431 cell line expresses
GPR109A and was used in these studies. On the day of the assay, fresh stimulation buffer was
made with HBSS containing 5 mM HEPES, 0.5 mM IBMX and 0.1% BSA. A 3 ꢀM forskolin
solution prepared in stimulation buffer was used in the assay. The cells were trypsinized and
washed off the flask in HBSS and spun down. The cells were then reꢀsuspended in stimulation
buffer and brought up to a concentration of 100 cells/ꢀL. Ten ꢀL of this cell suspension was
added into each well of a 96ꢀwell half area plate for a total of 1000 cells/well in the assay. Five
ꢀ
L of 4X compound or vehicle only was added to each well. Following this, 5 ꢀL of the 4X
forskolin solution was added to each well for a total final volume of 20 ꢀL. The plate was
incubated at room temperature for 30 minutes. The cellular cAMP concentration was measured
with the LANCE Ultra cAMP kit: the EuꢀcAMP tracer and ULightꢀantiꢀcAMP antibody were
diluted in detection buffer at 1/50 and 1/150 dilutions, respectively. Ten ꢀL each of the Euꢀ
cAMP and ULightꢀantiꢀcAMP were added to each well. The plate was then incubated at room
temperature for 60 minutes. The plate was read on a Victor X5 multilabel plate reader
(
PerkinElmer) at an excitation of 340 nm and emission of 615 nm and 665 nm. A ratio of
65/615 nm was calculated from this reading. The data were represented as the percent
inhibition of cAMP produced by forskolin.
6
ACS Paragon Plus Environment
24

Page 25 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Immunoblotting for SREBPꢀ2: HepG2 cells were plated in DMEM 10% FBS overnight.
Compound 5, atorvastatin and EPA stock solutions were prepared at 50 mM in 100% ethanol
and dilutions (10 X of final concentration) were made in 100% FBS. HepG2 cells were replaced
with fresh DMEM plus drugs for overnight. Cells were lysed in Radioimmunoprecipitation
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
assay (RIPA) buffer (20 mM) TrisꢀHCl (pH 7.5), 150 mM NaCl,1 mM Na EDTA, 1 mM EGTA,
2
1% NPꢀ40, 1% sodium deoxycholate, 2.5 mM sodium pyrophosphate, 1 mM betaꢀ
glycerophosphate, 1 mM Na VO , 1 µg/ml leupeptin) and cell lysates were separated on 4–20%
3
4
precast polyacrylamide gels (Bioꢀrad) and transferred to Immobilon PVDF membranes
Millipore). The membrane were blocked with Odyssey blocking buffer and incubated with antiꢀ
(
SREBP2 antibody (Abcam ab72856 at 1 to 500 dilution) and antiꢀActin (Abcam ab3280 at
10,000 dilution) overnight and were incubated with LIꢀCOR secondary antibodies. Images were
scanned using LIꢀCOR Odyssey imaging system and quantified by LIꢀCOR software. Data were
analyzed by GraphPad Prism.
Determination of IC for inhibition of PCSK9 secretion in HepG2 cells: HepG2 cells (from
50
ATCC, Catalog no. HBꢀ8065) were maintained in DMEM (Invitrogen) supplemented with 10%
fetal bovine serum (Invitrogen). The day prior to the PCSK9 assay, cells were seeded in 96ꢀwell
collagen coated plates at 25,000 cells/well. The test compounds were stored at ꢀ20 °C until used.
The test article compound was dissolved in 100% ethanol to a 50 mM stock solution. This was
then diluted in FBS to a final concentration of 1 mM. This solution was placed in a sonicating
water bath for 30 minutes. Subsequent dilutions were then made in FBS supplemented with an
equivalent volume of ethanol and mixed by vortexing. Once HepG2 cells were seeded onto a
collagen coated 96ꢀwell plate (Becton Dickinson, Catalog no. 35ꢀ4407) the assay was performed
as follows: the cell medium was removed, washed once with 100 ꢁL serum free DMEM to
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
remove any residual PCSK9, and replaced with 90ꢁL of serum free DMEM. Ten ꢁL of each
compound concentration prepared in FBS was then added. Each concentration of compound was
tested in triplicate. The compound was incubated with the cells overnight for 16 hours.
Following this incubation, 10 ꢁL of AlamarBlue was added to each well and cells incubated for
another 2 hours. The plates were then removed and AlamarBlue fluorescence was measured
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(excitation, 550nm and excitation, 590nm) to assess cell viability. Cell culture supernatant was
then diluted 1:5 in 1x RD5P Calibrator Diluent and PCSK9 ELISA was then performed with
50ꢁL of this diluted sample, as per the manufacturer’s instructions. The ELISA was measured
on a Victor X5 multilabel plate reader (PerkinElmer) at an absorbance of 450nm with
background correction measured at 550nm (The PCSK9 Elisa kits can be purchased from R&D
System, Catalog no. DPC900). Six test concentrations were used to generate the IC value.
5
0
Assay for inhibition of ApoB in HepG2 cells: HepG2 cells (ATCC) were seeded at 10,000
cells per well in 96 well plates. After adhering overnight, cells were then treated with the test
compounds at 6 concentrations (2 fold dilutions starting at 50 µM for Compound 5, EPA or the
combination of EPA plus niacin, or at 10mM for niacin alone). The fatty acid conjugate 5 was
dissolved along with FBS according to the procedure outlined above. Media supernatants were
removed and ApoB concentrations measured by ELISA following manufacturer's protocol
(Mabtech AB). Percent inhibition of ApoB secretion was determined by normalizing data to
vehicle treated wells. For a given compound, an IC (concentration at which 50% of ApoB
50
secretion is inhibited) is determined by using a 4 parameterꢀfit inhibition curve model (Graph
Pad Prism). In each experiment, cell viability was determined using Alamar Blue (Invitrogen
#DAL1100).
ACS Paragon Plus Environment
26

Page 27 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Assay for inhibition of NFꢀκB: RAW 264.7 macrophages stably transfected with 3xNFꢀκBꢀ
5
1
luciferase reporter were used for the assay. Cells were plated into 96 well plates and incubated
overnight. The next day, cells were treated with Vehicle (0.1% ethanol) or test compounds for 2
hours. 10 ꢁM dexamethasone was used as a positive control on NFꢀκB inhibition. Cell viability
reagent Alamar Blue was then added to every well and cells challenged with 200 ng/mL LPS for
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
hours. Cell viability was assessed using a Victor 5 plate reader measuring fluorescence with an
excitation filter of 550 nm and an emission filter of 590 nm. Data is expressed as percent of
Vehicle control after correcting for background fluorescence. Wells were then aspirated and 100
uL Brite Lite Plus luciferase reagent (Perkin Elmer #6016761) was added to each well.
Luminescence was measured using a Victor 5 plate reader. Data is presented as percent of LPS
stimulated control calculated as 100 x (CPS sample – average CPS unstimulated control) /
(average CPS LPS stimulated control – average CPS unstimulated control). Error bars represent
standard error of the mean (SEM) for experiment performed in triplicate.
PK study in rats: Male Sprague Dawley rats (n = 4/treatment group) were administered with a
single oral dose of either niacin (30 mg/kg) or compound 5 (100 mg/kg). Compounds were
formulated as an aqueous emulsion using Tween, Peceol, PEG400 and Tocopherol. Compound
5
(170 mg) was formulated in 2.38 g of the oil mixture containing 40%Tween, 50% Peceol, 10%
PEG400 and 0.2% Tocopherol. The concentration of 5 in oil was 66 mg/g of oil. For oral
dosing, this oil formulation was then diluted with 14.25 mL of water to form a 10 mg/mL drug
emulsion and dosed as an oral gavage. Collection time points were 5, 15 and 30 min; 1, 2, 4, 6
and 8 h postꢀdose. Blood samples were collected via saphenous vein and placed in tubes
containing paraoxon (10 µL of a 200 µM solution in DMSO) and stored on an ice block until
centrifuged at 8000 rpm for six minutes at 2ꢀ8°C. The resulting plasma was kept frozen at ꢀ80
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 54
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
°
C until analysis. Compound 5, nicotinic acid, and nicotinuric acid standard curves were
generated to allow for quantitative analysis of plasma samples. Standard curves were generated
by adding 10 ꢁL of a given compound in methanol (50 ꢁg/mL) to 50 ꢁL of blank plasma in a 96
well microplate resulting in the following final concentration range: 8,333 ng/mL, 2083 ng/mL,
521 ng/mL, 130.2 ng/mL, 32.5 ng/mL, 8.1 ng/mL, 2.0 ng/mL, and 0.5 ng/mL. 300 ꢁL of
acetonitrile containing 100 ng/mL of an internal standard was added to each sample. The 96ꢀ
well microplate was vortexed for 30 seconds and was then centrifuged for 45 minutes at 2,300 x
g. 100 ꢁL of supernatant was transferred from each sample to an empty 96ꢀwell microplate for
analysis by LC/MS/MS. The instrument used was an Agilent (Agilent Technologies Inc. USA)
liquid chromatograph equipped with a binary pump (1200 series), an autosampler (1200 series)
and a degasser (1200 series). Mass spectrometric analysis was performed using an Agilent 6410
Triple Quadrupole with an ESI interface. The data acquisition and quantitation was done with
Agilent Masshunter Software. The column used was a Phenomenex Gemini C6 Phenyl 110A 50
x4.6 mm, 1.8 ꢁm. The following mobile phase gradient was used (water = solvent A; methanol
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
=
solvent B): 0.01 min = 45%A/55%B; 0.5 min = 45%A/55%B; 1 min = 5%A/95%B; 3.5 min =
5%A/95%B; 3.51 min = 45%A/55%B; 4 min = 45%A/55%B.
PK/Tissue distribution study in Sprague Dawley rats: Male Sprague Dawley rats (n = 4) with
a median age of 6ꢀ8 weeks were used for the study. Animals were dosed with 300 mg/kg of
compound 5 q.d. for 7 days (formulated as an aqueous emulsion using Tween, Peceol, PEG400
and Tocopherol in the same manner as detailed above, with the ratio of API in oil adjusted to 500
mg/g of oil). Body weight was measured and recorded daily for each rat. Two hours after the
last dose, animals were euthanized and tissues were collected as follows: The left and right lobes
ACS Paragon Plus Environment
28

Page 29 of 54
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
of the liver, as well as the remaining liver (three portions per rat) were rinsed in saline, weighed
and snap frozen in separate 50 mL conical centrifuge tubes. The small intestine was dissected
from the stomach to the cecum. The ileum was defined as the 10 cm proximal to the cecum.
The ileum was dissected and frozen in a separate tube. The remaining small intestine composed
of the jejunum and duodenum was weighed and snap frozen in separate tubes. A maximum
obtainable blood volume was also collected at termination 2 h post last dose using sodium
heparin as an antiꢀcoagulant. In order to quantify the amount of compound 5 (or any of the
metabolites shown in Figure 8) in the intestine, all three dissected sections (ileum, jejunum and
duodenum) were pooled together and homogenized in PBS (2 x). A portion of the homogenized
mixture was quenched in acetonitrile containing the appropriate internal standard and analyzed
by LC/MS/MS. Standard curves in the appropriate tissues were generated in the same manner as
described above.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
ACS Paragon Plus Environment
29