Fatty Acid Cysteamine Conjugates as Novel and Potent Autophagy Activators That Enhance the Correction of Misfolded F508del-Cystic Fibrosis Transmembrane Conductance Regulator (CFTR)

Article abstract of DOI:10.1021/acs.jmedchem.6b01539

A depressed autophagy has previously been reported in cystic fibrosis patients with the common F508del-CFTR mutation. This report describes the synthesis and preliminary biological characterization of a novel series of autophagy activators involving fatty acid cysteamine conjugates. These molecular entities were synthesized by first covalently linking cysteamine to docosahexaenoic acid. The resulting conjugate 1 synergistically activated autophagy in primary homozygous F508del-CFTR human bronchial epithelial (hBE) cells at submicromolar concentrations. When conjugate 1 was used in combination with the corrector lumacaftor and the potentiator ivacaftor, it showed an additive effect, as measured by the increase in the chloride current in a functional assay. In order to obtain a more stable form for oral dosing, the sulfhydryl group in conjugate 1 was converted into a functionalized disulfide moiety. The resulting conjugate 5 is orally bioavailable in the mouse, rat, and dog and allows a sustained delivery of the biologically active conjugate 1.

Full text of DOI:10.1021/acs.jmedchem.6b01539

Subscriber access provided by UNIV OF REGINA
Article
Fatty Acid Cysteamine Conjugates as Novel and Potent Autophagy
Activators that Enhance the Correction of Misfolded F508del-
Cystic Fibrosis Transmembrane Conductance Regulator (CFTR)
Chi B Vu, Robert J. Bridges, Cecilia Pena-Rasgado, Antonio E. Lacerda, Curtis
Bordwell, Abby L. Sewell, Andrew J. Nichols, Sachin Chandran, Pallavi S. Lonkar,
Dominic Picarella, Amal Ting, Allison M. Wensley, Maisy Yeager, and Feng Liu
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b01539 • Publication Date (Web): 15 Dec 2016
Downloaded from http://pubs.acs.org on December 16, 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 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Fatty Acid Cysteamine Conjugates as Novel and
Potent Autophagy Activators that Enhance the
Correction of Misfolded F508delꢀCystic Fibrosis
Transmembrane Conductance Regulator (CFTR)
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
¶
¶
ǂ
ǂ
Chi B. Vu*, Robert J. Bridges , Cecilia PenaꢀRasgado , Antonio E. Lacerda , Curtis Bordwell ,
ǂ
Abby Sewell , Andrew J. Nichols, Sachin Chandran, Pallavi Lonkar, Dominic Picarella, Amal
Ting, Allison Wensley, Maisy Yeager and Feng Liu.
Catabasis Pharmaceuticals, One Kendall Square, Suite B14202, Cambridge, MA 02139.
¶
Rosalind Franklin University of Medicine and Science, Chicago Medical School, 3333 Green
ǂ
Bay Road, North Chicago, IL 60064. Charles River Laboratories, 14656 Neo Parkway,
Cleveland, OH 44128.
ABSTRACT
A depressed autophagy has previously been reported in cystic fibrosis patients with the common
F508delꢀCFTR mutation. This report describes the synthesis and preliminary biological
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 56
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
characterization of a novel series of autophagy activators involving fatty acid cysteamine
conjugates. These molecular entities were synthesized by first covalently linking cysteamine to
docosahexaenoic acid. The resulting conjugate 1 synergistically activated autophagy in primary
homozygous F508delꢀCFTR human bronchial epithelial (hBE) cells at submicromolar
concentrations. When conjugate 1 was used in combination with the corrector lumacaftor and
the potentiator ivacaftor, it showed an additive effect, as measured by the increase in the chloride
current in a functional assay. In order to obtain a more stable form for oral dosing, the sulfhydryl
group in conjugate 1 was converted into a functionalized disulfide moiety. The resulting
conjugate 5 is orally bioavailable in the mouse, rat and dog and allows a sustained delivery of the
biologically active conjugate 1.
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
Network pharmacology is a versatile and important tool for a medicinal chemist when faced with
a challenging therapeutic target. This drug design technique, sometimes referred to as pathway
1
ꢀ4
pharmacology or polypharmacology, involves the modulation of multiple therapeutic pathways
in order to achieve a desired pharmacological response. We have recently described a variation
of this technique which allowed network pharmacology to take place when two bioactives are
5
delivered intracellularly at the same time and in equimolar concentrations. In our approach, a
plasma stable linker was used to first covalently join the two bioactives. Once delivered inside
cells, intracellular enzymes could readily cleave the linker to simultaneously release the two
bioactives. Since multiple biological pathways were simultaneously impacted by this type of
intracellular delivery, the resulting pharmacology that could be produced was unique and could
not be replicated by giving the two bioactives either individually or in combination. We have
ACS Paragon Plus Environment

Page 3 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
demonstrated the utility of our methodology with two examples: the first involves the use of a
fatty acid niacin conjugate in the inhibition of sterol regulatory element binding protein (SREBP)
and the second involves the use of a fatty acid salicylate conjugate in the inhibition of NFꢀκB.
Both of these therapeutic targets have historically been known to be difficult to attenuate by
traditional medicinal chemistry approaches. In this communication, we will demonstrate how
this approach could be used again to address another challenging therapeutic target; this time
involving the activation of autophagy. Even though a depressed autophagy has been implicated
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
6
in numerous diseases, this report focuses primarily on its impact in cystic fibrosis (CF).
CF is a deadly orphan disease that affects over 70,000 patients worldwide, predominantly in the
Caucasian population. It is caused by a defective cystic fibrosis transmembrane conductance
7
regulator (CFTR), a cAMPꢀdependent ion channel that regulates the flow of chloride ions across
epithelial cell membranes. A malfunctioning CFTR ion channel will cause an imbalance in ion
and fluid transport; and over time, this can have a debilitating effect on the pancreas, liver,
kidneys, intestine and especially the lungs. Even though there are more than 1900 reported
mutations in CF, more than 80% of CF patients worldwide carry the F508delꢀCFTR mutation, in
either the homozygous or heterozygous form. With this particular mutation, the misfolded
F508delꢀCFTR cannot be transported out of the endoplasmic reticulum (ER), and instead is
degraded by the proteasome before it can rise to the cell surface. A restoration of F508delꢀCFTR
function therefore requires the use of two agents: 1) a CFTR corrector to help transport the
misfolded CFTR to the cell surface; 2) a CFTR potentiator to help keep this ion channel open at
the cell surface. Intense research efforts in this field have recently resulted in the approval of the
8
combination consisting of the corrector lumacaftor (VXꢀ809) and the potentiator ivacaftor (VXꢀ
9
7
70) in CF patients with the homozygous F508delꢀCFTR mutation (Figure 1). However, the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 56
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
clinical efficacy for this particular combination of drugs is modest, with an average improvement
1
0
of 3% in lung function. Therefore, numerous efforts are currently being taken to identify more
1
1
effective CFTR correctors or potentiators, as well as agents that can potentially improve 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
12
efficacy of the current combination of lumacaftor and ivacaftor. In this report, we will
demonstrate that the autophagy pathway is intricately linked to the trafficking of the CFTR; and
an effective autophagy activator can enhance the transport of the misfolded CFTR to the cell
surface and be useful as part of a triple drug combination to treat CF patients with the F508delꢀ
CFTR mutation.
RESULTS AND DISCUSSION
Autophagy and the trafficking of the F508delꢀCFTR
Normal cells typically use autophagy as a means for removing sources of reactive oxygen
species, damaged mitochondria, misfolded proteins, and foreign pathogens. The autophagy
process involves the initial formation and elongation of a membrane sac called the phagophore,
1
3
which eventually develops into an autophagosome. This is a doubleꢀmembrane vesicle that is
used to sequester the dysfunctional cellular components into the cytoplasmic cargo. The
autophagosomes themselves will eventually fuse with the lysosomes to enable the degradation of
the cargo. This process is used to essentially convert damaged cellular constituents into amino
acids, lipids, sugars, and nucleotides, all basic building blocks for the regeneration of new
proteins and organelles. In CF patients with the common F508delꢀCFTR mutation, a depressed
autophagy could be traced back to decreased levels of the Beclinꢀ1, a key protein that is needed
1
4
for the autophagosome formation. The defective F508delꢀCFTR is believed to induce an upꢀ
regulation of reactive oxygen species (ROS) and a corresponding increase in transglutaminase
ACS Paragon Plus Environment

Page 5 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
TG2) activity. The increased TG2 activity, in turn, drives the crosslinking and the subsequent
disabling of Beclinꢀ1. Once autophagy is depressed, there is a corresponding accumulation of
p62. Since p62 has been shown to regulate aggresome formation, its accumulation effectively
traps the misfolded CFTR and prevents its trafficking to the cell surface. The sequestered CFTR
is then targeted for rapid degradation by the proteasome (Figure 2). A successful activation of
autophagy could potentially allow this sequence of events to be reversed, and thereby allow more
of the F508delꢀCFTR to rise to the cell membrane. Previous proofꢀofꢀconcept studies using
human epithelial cell lines (CFBE41oꢀ and IB3ꢀ1 cells) that were transfected with the F508delꢀ
CFTR have shown that over expression of Beclinꢀ1 or knock down of p62 via siRNA did indeed
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
result in more of the misfolded CFTR being transported to the cell surface. Even though this
autophagyꢀbased approach has the potential to address one of the underlying causes of CF, it has
not been investigated extensively in the clinic. This is perhaps due in part to the lack of safe and
effective agents that could selectively activate this particular pathway. Among the agents that
have been reported to date (i.e., brefeldin A, rapamycin, valproic acid, trehalose, tunicamycin,
carbamazepine, penitrem A) some could only activate autophagy at high concentrations; while
others displayed many offꢀtarget side effects or nonꢀdrug like properties, which precluded their
1
6
evaluation in the clinic. Cysteamine and its disulfide form, namely cystamine, appear to be
promising leads for autophagy activation, based on their safe and long term use in the clinic for
1
7
F508del
patients with nephropathic cystinosis. Administration of cysteamine to CFTR
mice over a
5
week period showed a reduction in inflammatory markers (TNF, Cxcl2) and an increase in
body weight, CFTR protein and the autophagy marker Beclinꢀ1. Maiuri and coworkers have
shown that cystamine (at 250 ꢁM) could activate autophagy in brushed nasal epithelial cells from
CF patients (homozygous F508delꢀCFTR) to allow the trafficking of the misfolded CFTR to the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 56
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
18
cell membrane. Cysteamine has since been evaluated in a pilot clinical trial involving
1
9, 20
homozygous F508delꢀCFTR CF patients, in combination with epigallocatechin gallate.
Cysteamine administration, however, comes with many challenges because of its poor PK profile
i.e., short halfꢀlife) and the associated GI side effects (i.e., nausea, emesis). Because of its low
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
(
potency, it is not clear if cysteamine could be administered at doses high enough to allow a
restoration of autophagy without resulting in significant offꢀtarget effects. Furthermore,
cysteamine itself has an unpleasant thiol odor and taste which can present significant patient
compliance issues upon chronic dosing.
As outlined below, we have identified a more effective way of activating autophagy via the use
of fatty acid cysteamine conjugates. These covalent conjugates have the potential of filling the
current gap in the availability of more potent, yet safe and effective autophagy activators suitable
for definitive proofꢀofꢀconcept studies. These NCEs were prepared by covalently linking
cysteamine to the omegaꢀ3 fatty acid (4Z,7Z,10Z,13Z,16Z,19Z)ꢀdocosaꢀ4,7,10,13,16,19ꢀ
hexaenoic acid (DHA). This process allows the simultaneous delivery of the two bioactives
without having to address the individual PK and tissue distribution of the individual components.
Our fatty acid cysteamine conjugates synergistically activated autophagy in cultured primary
homozygous F508delꢀCFTR hBE cells at concentrations that were significantly lower than what
had previously been reported with cysteamine.
Synergistic activation of autophagy with a fatty acid cysteamine conjugate
Cysteamine has been reported by Maiuri and coꢀworkers to activate autophagy in certain human
epithelial cell lines (CFBE41oꢀ and IB3ꢀ1 cells) by restoring Beclinꢀ1 via the inhibition of
1
4
TG2. However, the effect was modest and a high concentration of cysteamine (250 ꢁM) was
ACS Paragon Plus Environment

Page 7 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
required. We therefore hypothesized that if we could simultaneously deliver cysteamine along
with a second bioactive, another pathway could perhaps work in concert with the TG2 inhibition
to amplify the autophagy activation. We selected DHA as this second bioactive. This omegaꢀ3
fatty acid, by itself, has previously been shown to increase autophagosome formation via 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
2
1, 22
AMPK pathway.
In order to achieve a simultaneous intracellular delivery, we employed the
5
same methodology that we described earlier by first covalently linking cysteamine to DHA to
form the fatty acid cysteamine conjugate 1 (Figure 3). The amide bond that is present in the
covalent conjugate 1 is susceptible to cleavage by the intracellular enzyme, fatty acid amide
hydrolase (FAAH). We have evaluated this hydrolysis process using the purified recombinant
5
enzyme FAAHꢀ1. As shown in Figure 4, recombinant FAAHꢀ1 readily hydrolyzed the amide
bond to release DHA in a timeꢀdependent fashion. Interestingly, FAAH is localized inside cells
at the ER, the same compartment where autophagy is initiated. The covalent conjugate 1 allows
cysteamine and DHA to be delivered at the same time, in equimolar concentrations, to the same
subcellular compartment in which the mutant F508delꢀCFTR is trapped by the p62 aggresomes.
This simultaneous intracellular delivery enables an unprecedented level of synergy to occur in
terms of autophagy activation. Figure 5 summarizes an experiment where cultured primary
homozygous F508delꢀCFTR hBE cells were incubated with the following treatment groups: 1)
vehicle; 2) DHA (250 ꢁM); 3) cysteamine (250 ꢁM); 4) combination of cysteamine (250 ꢁM) +
DHA (250 ꢁM); 5) fatty acid cysteamine conjugate 1 (3 ꢁM). It should be noted that the fatty
acid cysteamine conjugates have limited water solubility and need to be solubilized along with a
protein carrier, such as FBS, prior to the serial dilution with the assay media in order to obtain
the optimal effect. Compound 1, at a concentration of 3 ꢁM, increased Beclinꢀ1; and this, in turn,
activated autophagy, as noted by the increase in the ratio of LC3ꢀII to LC3ꢀI, a commonly used
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 56
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
23
marker for this catabolic process. Autophagy activation was further accompanied by a
corresponding decrease in p62, a key protein that participates in the sequestration of the F508delꢀ
CFTR. This synergistic activation in autophagy was observed only with the covalent fatty acid
cysteamine conjugate 1; and not with the individual components (i.e., cysteamine, DHA) or a
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
2
4
combination with the individual components. In cultured primary CF hBE cells obtained from
this particular donor, cysteamine, by itself, did not activate autophagy even at a concentration of
2
50 ꢁM. The omegaꢀ3 fatty acid DHA, as well as the combination consisting of cysteamine and
DHA, did not activate autophagy in primary CF hBE cells at a concentration of 250 ꢁM.
We next assessed how restoration of autophagy would affect the overall correction of the
F508delꢀCFTR upon chronic exposure to the CFTR corrector lumacaftor and the potentiator
ivacaftor. Figure 6 summarizes an experiment where cultured primary homozygous F508delꢀ
CFTR hBE cells (patient code KKCFFT006F) were incubated for 24 hr with 1 (3 ꢁM) along with
2
5
lumacaftor (3 ꢁM) and ivacaftor (0.10 ꢁM) in differentiation media. Compared to the vehicleꢀ
treated group, treatment with lumacaftor/ivacaftor resulted in the expected increase in the CFTR
band B (the immature, coreꢀglycosylated isoform) and band C (the mature, complexꢀ
glycosylated isoform). When lumacaftor and ivacaftor were used in combination with the fatty
acid cysteamine conjugate 1 (3 ꢁM), a restoration of autophagy was observed, as noted by the
increase in the level of Beclinꢀ1 and the ratio of LC3ꢀII to LC3ꢀI (Figure 6C). Consistent with the
previously proposed mechanism, an increase in autophagy was accompanied by a corresponding
decrease in the level of p62, which, in turn, allowed some of F508delꢀCFTR to rise to the cell
surface. As shown in Figures 6A and 6B, the combination consisting of 1 (3 ꢁM) and
lumacaftor/ivacaftor clearly produced an increase in both the CFTR band B and band C. With
cultured primary homozygous F508delꢀCFTR hBE cells from this particular donor
ACS Paragon Plus Environment

Page 9 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
KKCFFT006F), we have observed autophagy activation even when 1 was used at the lower
concentrations of 0.018ꢀ0.60 ꢁM (this data is shown in the Supporting Information). Maiuri and
1
4
coworkers have previously reported that once autophagy is activated, a reduction in the
turnover of plasma membraneꢀlocated F508delꢀCFTR in human CF airway epithelial cells is
observed. This could potentially translate to an increase in the chloride current across the
F508delꢀCFTR when the proper CFTR corrector and potentiator are available. The next stage of
our validation involves measurement of the chloride current across the F508delꢀCFTR hBE cells.
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
A triple combination to correct F508delꢀCFTR
A successful correction of the F508delꢀCFTR needs to translate to an actual increase in the
chloride current to be of benefit to CF patients. This important functional activity was assessed
using the TECCꢀ24 semiꢀautomated format with cultured primary homozygous F508delꢀCFTR
hBE cells. Unlike the traditional Ussing chamber assays, where the short circuit current (I ) is
SC
measured as an index of CFTR correction, the TECCꢀ24 assay measures the equivalent current
(
IEQ). The TECCꢀ24 assay format offers some advantages over the Ussing chamber assay, one of
which is the higher throughput; and the other involves the ability to measure the IEQ with the test
compounds continuously present in the recording medium. However, since the IEQ was recorded
in 5 min intervals, one disadvantage with the TECCꢀ24 in the present format is the lack of
sensitivity of the assay during the forskolin activation phase. Nevertheless, once steady state has
been achieved, the bumetanideꢀinhibited CFTR chloride current obtained from a TECCꢀ24 assay
represents a reliable measure of functional efficacy. There are two basic protocols that are used
to assess functional activity: 1) the acute ivacaftor protocol; and 2) the chronic ivacaftor preꢀ
incubation protocol. In the acute ivacaftor protocol, cultured primary homozygous F508delꢀ
CFTR hBE cells were incubated with the fatty acid cysteamine conjugate 1 along with the CFTR
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 56
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
corrector lumacaftor for 24 hr. Forskolin and the CFTR potentiator ivacaftor were then added
acutely just after benzamil was added to the apical side to inhibit sodium transport prior to the
IEQ measurements. This is the standard set of assay conditions that were used originally to assess
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
8
the functional activity of the corrector lumacaftor. In the chronic ivacaftor preꢀincubation
protocol, cultured primary homozygous F508delꢀCFTR hBE cells were incubated with the fatty
acid cysteamine conjugate 1 along with both the corrector lumacaftor and the potentiator
ivacaftor for 24 hr prior to the IEQ measurements. This protocol was introduced more recently
26
27
based on the findings from Cholon and Veit that the potentiator ivacaftor actually had a
destabilizing effect on the corrected CFTR upon chronic preꢀincubation with lumacaftor; and the
chloride current measured under this protocol was significantly less than what was previously
reported with the acute addition of ivacaftor. When a triple combination is used in a clinical
setting, this chronic preꢀincubation protocol might be more representative of what cells would be
exposed to at a given time period. We have assessed the functional activity of 1 using this
chronic ivacaftor preꢀincubation protocol and have found a significant additive effect on top of
28
lumacaftor/ivacaftor. Figure 7 summarizes an experiment using the chronic ivacaftor preꢀ
incubation protocol. Here, primary homozygous F508delꢀCFTR hBE cells (patient code
o
CFFT0018I) was incubated in differentiation media for 24 hr at 37 C with the following
treatment groups: 1) vehicle + ivacaftor (0.10 ꢁM); 2) lumacaftor (3 ꢁM) + ivacaftor (0.10 ꢁM);
3) 1 (0.075 ꢁM) + lumacaftor (3 ꢁM) + ivacaftor (0.10 ꢁM). The next day, the differentiation
media was replaced with Coon’s F12 (without serum or bicarbonate) media; with the same
2
9
concentrations of the test compounds added back to this new media. To initiate the run,
benzamil was added first to the apical side to block currents through the epithelial sodium
channel (ENaC). Forskolin was added and the IEQ was recorded over a 27 min period. The
ACS Paragon Plus Environment

Page 11 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
antagonist bumetanide was then added to the basolateral side to inhibit the NaꢀKꢀ2Cl
cotransporter and block chloride secretion from the CFTR. The resulting bumetanideꢀinhibited
2
CFTR chloride current (ꢂBumetanide, ꢁA/cm ) is a reflection of the functional activity of 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
now corrected F508delꢀCFTR. As shown in Figures 7A and 7B, the triple combination
consisting of 1 (0.075 ꢁM) along with lumacaftor/ivacaftor showed a statistically significant
increase in the bumetanideꢀinhibited CFTR chloride current (ꢂBumetanide = 3.08 ± 0.24
2
2
ꢁ
A/cm ) over the lumacaftor/ivacaftor treatment group (ꢂBumetanide = 1.91 ± 0.22 ꢁA/cm ).
This increase in the CFTR chloride current could also be expressed as a % of the
lumacaftor/ivacaftor control group, with the appropriate subtraction of the ꢂBumetanide value
corresponding to the vehicle group from both the triple combination and the lumacaftor/ivacaftor
treatment group, in order to allow for a more direct comparison across the different assay runs.
For this particular run, the triple combination allowed a 184% increase in the bumetanideꢀ
inhibited CFTR chloride current relative to the lumacaftor/ivacaftor treatment group; a nearly
twoꢀfold increase in the response (Figure 7B, with the appropriate subtraction of the
2
ꢂBumetanide value for the vehicle group of 0.52 ± 0.14 ꢁA/cm ). In this assay format using
primary cells from this particular patient donor, we did not observe a noticeable change in the
forskolinꢀstimulated ꢂI for the triple combination. However, there was a statistically
EQ
significant increase in the AUC for the triple combination over the lumacaftor/ivacaftor
2
treatment group (Figure 7C, triple combination AUC = 173.3 ± 18.5 ꢁA/(27 min * cm ) vs
2
lumacaftor/ivacaftor control group AUC = 134.2 ± 5.6 ꢁA/(27 min * cm ); vehicle group AUC =
2
2
2.3 ± 4.9 ꢁA/(27 min * cm )). With this particular assay setup, the solubility of 1 in the Coon’s
F12 media was a limiting factor. At concentrations of > 0.075 ꢁM, some of the precipitated, oily
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 56
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
fatty acid conjugates began to interfere with the current recording. The IEQ readings became
3
0
much more variable and the results were more difficult to interpret.
Triple combinations using primary CF hBE cells from different donors
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
We have also evaluated the functional activity of 1 in a similar TECCꢀ24 assay format using
primary CF hBE cells from different donors. Figure 8 summarizes a TECCꢀ24 assay that was
carried out at a different laboratory, using cultured primary homozygous F508delꢀCFTR hBE
cells from patient code KKCFFT006F. These primary hBE cells are of the same patient code as
the one used in the band B/C western blots shown in Figure 6. Using the chronic ivacaftor preꢀ
incubation protocol, cells were treated with the following groups: 1) vehicle + ivacaftor (0.10
ꢁ
M); 2) lumacaftor (3 ꢁM) + ivacaftor (0.10 ꢁM); 3) 1 (0.15 ꢁM) + lumacaftor (3 ꢁM) +
ivacaftor (0.10 ꢁM). As detailed in the experimental section, there are some minor differences
between the two TECCꢀ24 assay formats. However, the basic protocol remains essentially the
same. One major difference is in the antagonist that is used to block the chloride secretion from
the CFTR. With this set up, CFTR ꢀinh was used instead of bumetanide; and the antagonistꢀ
1
72
inhibited CFTR chloride current, ꢂI (CFTR ꢀ172), represents essentially the same functional
EQ
inh
readout as the ꢂBumetanide described in Figure 7. With this particular TECCꢀ24 set up, using
cells from this particular patient donor, a significant increase in the peak forskolin response was
observed with the triple combination (Figure 8A). As shown in Figure 8B, the peak forskolin
2
ꢂI for the triple combination was 24.47 ± 0.62 ꢁA/cm . This increase in the peak forskolin
EQ
response was statistically significant over the lumacaftor/ivacaftor treatment group (ꢂIEQ = 16.61
2
±
2.16 ꢁA/cm ). When expressed as a % of the lumacaftor/ivacaftor control group, with the
2
appropriate subtraction of the ꢂI value of 7.50 ± 0.44 ꢁA/cm for the vehicle group, the triple
EQ
combination with 1 afforded a 186% increase in the peak forskolin response; a nearly twoꢀfold
ACS Paragon Plus Environment

Page 13 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
increase in the response. In terms of the CFTR ꢀ172 inhibited chloride current, the triple
inh
2
combination (ꢂI = 7.47 ± 0.16 ꢁA/cm ) still showed a statistically significant increase over the
EQ
2
lumacaftor/ivacaftor treatment group (ꢂI = 5.45 ± 0.48 ꢁA/cm ). When expressed as a % of the
EQ
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
lumacaftor/ivacaftor control group, this represents a 149% increase in the antagonistꢀinhibited
chloride current (with the appropriate subtraction of the vehicle group ꢂIEQ = 1.37 ± 0.23
2
ꢁ
A/cm ). There was also a statistically significant increase in the AUC for the triple combination
2
(
triple combination AUC = 607.3 ꢁA/(60 min * cm ) vs lumacaftor/ivacaftor control AUC =
2
2
430.6 ± 40.5 ꢁA/(60 min * cm ); vehicle group AUC = 137.2 ± 20.3 ꢁA/(60 min * cm ). Even
though the solubility of 1 was still limited in this TECCꢀ24 assay set up, it did not appear to
interfere with the IEQ measurements as much as in the previous set up. We have evaluated 1 in
concentrations of up to 0.6 ꢁM and observed an additive effect on top of lumacaftor/ivacaftor
(the entire data set is shown in the Supporting Information, along with the corresponding band
B/band C data for this lower concentration of 0.15 ꢁM of conjugate 1). In this particular assay
setup, at concentrations of ≥ 1.2 ꢁM of 1, solubility became an issue in the Coon’s F12 media;
31
and we began to experience an inhibition in the chloride current.
A statistically significant increase in the peak forskolin response was also observed when 1 was
evaluated in the same TECCꢀ24 setup using primary cells from a different patient donor. Figure
9
summarizes an experiment where primary homozygous F508delꢀCFTR hBE cells (patient code
o
KKCFFT004I) were treated with the following groups for 24 hr at 37 C prior to IEQ
measurements: 1) vehicle + ivacaftor (0.10 ꢁM); 2) lumacaftor (3 ꢁM) + ivacaftor (0.10 ꢁM); 3)
1
(0.30 ꢁM) + lumacaftor (3 ꢁM) + ivacaftor (0.10 ꢁM). Under this chronic ivacaftor preꢀ
incubation protocol, there was a statistically significant increase in the peak forskolinꢀactivated
CFTR chloride current with the triple combination. Even though the magnitude of the increase
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 56
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
was smaller than what was observed with the previous patient code, the ꢂIEQ (peak forskolin) for
the triple combination still represent a 167% increase over the lumacaftor/ivacaftor treatment
2
group (triple combination, ꢂIEQ = 8.24 ± 1.18 ꢁA/cm vs lumacaftor/ivacaftor control group ꢂIEQ
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
2
2
=
6.250 ± 0.36 ꢁA/cm ; vehicle group, ꢂI = 3.90 ± 0.88 ꢁA/cm ). With primary cells from this
EQ
particular donor, we began to see an inhibition in the chloride current at concentrations ≥ 0.60
M of 1 (additional data is presented in the Supporting Information). Having evaluated 1 in
ꢁ
TECCꢀ24 setups from different laboratories, using primary CF cells from different patient
donors, we have noticed a wide range of responses. With some primary hBE cells, we did not
experience any chloride current inhibition until we reached a concentration of ≥ 1.2 ꢁM of 1. In
others, we began to experience significant variability and some chloride current inhibition at
much lower concentrations. There are many contributing factors to the observed variations,
including: a) different laboratories cultured and maintained primary hBE cells from different
donors on different TECCꢀ24 filter inserts; b) different protocols were used to remove the mucus
film on the apical side of the hBE monolayers prior to IEQ measurements; c) different sources of
FBS were used to solubilize 1 prior to the serial dilution. However, one observation that was
consistent across the different setups was the additive effect of 1 on top of lumacaftor/ivacaftor
when used at lower concentrations, where solubility of 1 in the recording media was less of a
factor. Thus far, because of the low solubility of 1 in the Coon’s F12 media using this TECCꢀ24
assay format, we have not been able to assess its functional activity at the higher concentration of
3
ꢁM (i.e, the concentration where we observed a significant increase in the CFTR band C,
Figure 6). In order to evaluate 1 at the higher concentrations, we have looked into different
32
functional assay formats, including the forskolinꢀinduced swelling of organoids. The results
from those studies will be disclosed in a separate communication.
ACS Paragon Plus Environment

Page 15 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Design of a plasmaꢀstable and orally bioavailable fatty acid cysteamine conjugate
Even though the covalent conjugate 1 showed promising functional activity, when used in
combination with lumacaftor/ivacaftor, it is rather unstable when exposed to air, presumably
because of the sulfhydryl moiety. A more stable form is therefore needed for the appropriate in
vivo studies. Converting the sulfhydryl group to the disulfide form, as shown with the cystamine
derivative 2, should enhance the overall stability of the covalent conjugate (Figure 3). In theory,
the bisꢀfatty acid cystamine conjugate 3 should allow the intracellular delivery of the same
covalent conjugate 1. In practice, however, this type of bisꢀfatty acid was rather difficult to
formulate for oral dosing, as well as for standard in vitro work. We, therefore, sought to replace
one of the two fatty acids with a smaller substituent. Based on our earlier work with fatty 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
5
niacin conjugates, similar to the one shown with structure 6, we selected nicotinic acid as the
capping group for our first proofꢀofꢀconcept compound. What we found was that the cystamine
linker in 4 was still unstable in the plasma. To improve the plasma stability of our fatty acid
cysteamine conjugate, a bisꢀgeminal methyl group was installed adjacent to the disulfide
33
moiety. As shown in Figure 10, compound 5 was stable in rat, mouse, dog, and human plasma.
5
Unlike the related compound 6, which had significant inhibitory activity against SREBP, the
fatty acid cysteamine conjugate 5, with this type of disulfide linker, did not show any SREBP
activity at the highest tested concentration of 50 ꢁM. Compound 5, in turn, was prepared
according to the synthetic sequence outlined in Scheme 1. Cysteamine was first reacted with 1,2ꢀ
di(pyridinꢀ2ꢀyl)disulfane (7) to form the mixed disulfide derivative 8. Treatment of 8 with DHA
afforded the fatty acid derivative 9. This was subsequently treated with 1ꢀaminoꢀ2ꢀ
methylpropaneꢀ2ꢀthiol to afford the bisꢀgeminal methyl derivative 9. Standard amide coupling
with nicotinic acid afforded the fatty acid cysteamine conjugate 5. Unlike compound 1, the fatty
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 56
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 cysteamine conjugate 5 is now sufficiently stable at room temperature and can be
conveniently stored in the freezer over an extended period (initial stability > 1 year). Compound
5
can be conveniently formulated as a selfꢀemulsifying dispersion (SED) using an aqueous
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
mixture consisting of Tween, PEG400, and Peceol (glyceryl monoleate Type 40) for oral gavage
administration. In a typical PK experiment, Sprague Dawley rats were dosed orally with 30
mg/kg of compound 5 (Figure 11). Serial blood collection was carried out at 0.125, 0.25, 0.5, 1,
2
, 4, 5, 8 and 12 hrs. In order to quantify the unstable thiol metabolite 1, the corresponding
Ellman’s adduct 15 (Figure 12) was prepared and used to prepare the necessary standard curve.
A small quantity of the parent compound 5 was detected in the plasma upon oral dosing (Cmax
4.6 ± 4.39 ng/mL; AUClast = 37.7 ± 16.2 h*ng/mL). There was a significantly higher and more
=
2
sustained level of the biologically active metabolite 1 (Cmax = 777 ± 435 ng/mL, AUClast = 2929
34
±
1282 h*ng/mL) and this was detected in the plasma even at the 8 and 12 hr time point. This
plasma level would support the feasibility of either a BID or QD dosing and is sufficient to
3
5
support the functional activity displayed by 1 in the various assays discussed earlier.
Even as the disulfide form, when 5 was administered to primary hBE cells, the disulfide bond
should undergo the appropriate intracellular reduction to afford the biologically active metabolite
1
. The functional activity of the disulfide derivative 5 was evaluated in the TECCꢀ24 assay
format, and some of the results are described herein. Figure 13 summarizes an experiment where
5 was evaluated along with the corrector lumacaftor, upon acute exposure to the potentiator
ivacaftor (additional TECCꢀ24 assay data for 5 under the chronic ivacaftor preꢀincubation
protocol, as well as other functional data, can also be found in the Supporting Information). With
this setup using primary hBE cells from this particular patient code (homozygous F508delꢀ
CFTR, CFFT028H), there was no significant change in the forskolin response. However, a twoꢀ
ACS Paragon Plus Environment

Page 17 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
fold increase in the bumetanideꢀinhibited CFTR chloride current was observed when 5 (0.0375
ꢁ
M) was used in combination with lumacaftor and ivacaftor (triple combination, ꢂBumetanide =
2
2
3
.73 ± 0.41 ꢁA/cm vs lumacaftor/ivacaftor control, ꢂBumetanide = 1.54 ± 0.04 ꢁA/cm ). Figure
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
4 summarizes an experiment where 5 (0.018 ꢁM) was evaluated in a different laboratory using
primary homozygous F508delꢀCFTR hBE cells from another patient code (KKCFFT0012I)
along with lumacaftor and ivacaftor. With this setup, although there was no change in the peak
forskolin response, there was a significant additive effect for the triple combination when we
looked at other parameters such as the CFTR 172ꢀinhibited CFTR chloride current (triple
inh
2
combination, ꢂIEQ = 3.05 ± 0.18 ꢁA/cm vs lumacaftor/ivacaftor control, ꢂI = 1.74 ± 0.29
EQ
2
2
ꢁ
A/cm ; vehicle group, ꢂI = 0.18 ± 0.11 ꢁA/cm ), and the AUC (triple combination, AUC =
EQ
2
326.36 ± 17.16 ꢁA/(60 min * cm ) vs lumacaftor/ivacaftor control, AUC = 209.24 ± 40.42
2
2
ꢁ
A/(60 min * cm ); vehicle group, AUC = 63.83 ± 5.44 ꢁA/(60 min * cm )). When expressed as
a % of the lumacaftor/ivacaftor control group, both of these responses showed a ≥ 180% increase
for the triple combination, a nearly twoꢀfold increase over the lumacaftor/ivacaftor treatment
group. In total, cultured primary homozygous F508delꢀCFTR hBE cells from five different
patients were used to demonstrate the functional activity of conjugate 1 and the disulfide
derivative 5 in the TECCꢀ24 assay format, using either the chronic ivacaftor preꢀincubation
protocol or the standard acute ivacaftor protocol.
CONCLUSION
In summary, we have demonstrated the use of fatty acid cysteamine conjugates as novel
autophagy activators that could enhance the correction of misfolded F508delꢀCFTR, when used
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 56
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
in combination with established CFTR modulators. Previous work by Maiuri and coꢀworkers
has demonstrated that autophagy could play a role in the trafficking of the mutant F508delꢀ
1
4, 15
CFTR.
However, the reagent employed in their studies, cysteamine, was a weak autophagy
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
activator; and the effect was observed only at 250 ꢁM in certain cell lines. This high
concentration of cysteamine would be difficult to achieve in the various tissues because of its
poor PK properties. By employing our covalent linking methodology, we have demonstrated that
the fatty acid cysteamine conjugate 1 could synergistically activate autophagy in primary
homozygous F508delꢀCFTR hBE cells at much lower concentrations. This level of activity could
not be replicated by using the individual components (i.e., cysteamine, DHA) or a combination
of the individual components, even at much higher concentrations. Immunoblotting and chloride
conductance assays using cultured primary homozygous F508delꢀCFTR hBE cells were used to
verify the functional activity of the fatty acid cysteamine conjugate 1. Depending on the primary
CF cells used, we have observed an improvement of nearly twoꢀfold in either the peak forskolin
response or in the bumetanideꢀinhibited CFTR chloride current when the conjugate 1 was used in
combination with lumacaftor/ivacaftor. In order to obtain a more stable form for oral dosing, as
well as to improve patient compliance, we converted the sulfhydryl group present in 1 to the
corresponding odorless disulfide derivative. The resulting conjugate 5 is orally bioavailable in
the mouse, rat and dog when formulated as a selfꢀemulsifying dispersion. This disulfide form
allowed a sustained delivery of the biologically active metabolite 1, with plasma concentrations
that were significantly higher than the concentrations at which functional activity was observed
in primary homozygous F508delꢀCFTR hBE cells (0.075ꢀ0.3 ꢁM). Since the biologically active
metabolite 1 is so effective in activating autophagy, we are anticipating that a much lower dose
of it would be needed in an in vivo setting. This would minimize the unpleasant smell, taste,
ACS Paragon Plus Environment

Page 19 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
body odor and offꢀtarget side effects that are associated with the chronic administration of high
doses of cysteamine.
This communication focuses primarily on the effect of autophagy on the correction of the
misfolded F508delꢀCFTR. However, it is not the only significant biological activity that is
associated with an autophagy activator such as 1 and its disulfide form 5. The autophagy
mechanism has previously been reported to have an important role in the innate and adaptive
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
6, 37
immunity and this could have a significant impact on the intracellular clearance of bacteria.
While an autophagy activator can be helpful in enhancing CFTR correction, we believe that this
unique antiꢀbacterial effect can potentially be just as useful in CF. It is widely known that the
lack of a proper functioning CFTR in the lungs of CF patients causes a chronic bacterial
3
8ꢀ40
infection which further deteriorates their lung function.
Pseudomonas aeruginosa is a
41
particularly difficult pathogen to treat because of its ability to internalize within cells. Airway
epithelial cells thus become a reservoir of intracellular bacteria during chronic P. aeruginosa
infection. Autophagy activation enables an alternative mechanism to clear the bacterial infection
42
out of the bronchial epithelial cells; and therefore, could potentially be useful when used in
combination with antiꢀinfective agents. We have evaluated the ability of compound 5 (CATꢀ
5571) to activate autophagy and induce an intracellular clearance of P. aeruginosa in a number
4
3
of in vitro and in vivo studies. The results from those studies will be discussed in more detail in
a separate communication.
EXPERIMENTAL SECTION
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 56
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
General Information: All chemical reagents and solvents were commercially available and
used as received. (4Z,7Z,10Z,13Z,16Z,19Z)ꢀDocosaꢀ4,7,10,13,16,19ꢀhexaenoic acid (DHA) was
purchased from NuꢀChek Prep. Lumacaftor was purchased from Astatech; and ivacaftor was
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
1
purchased from DꢀLꢀChiral Chemicals. Reactions were generally run under argon or nitrogen. H
1
3
NMR (400 MHz) and C NMR (100 MHz) spectra were recorded on a Varian 400MHz Unity
Inova system in CDCl , unless otherwise noted. Chemical shifts are expressed in parts per
3
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 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 mode. A SiliCycle C18 XDB, 3 × 100 mm column
was used with a gradient of H O and acetonitrile each with 0.1% formic acid; UV detection at
2
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 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).
(4Z,7Z,10Z,13Z,16Z,19Z)-N-(2-Mercaptoethyl)docosa-4,7,10,13,16,19-hexaenamide (1):
The fatty acid cysteamine conjugate 1 was generated fresh from the wellꢀcharacterized disulfide
form prior to each assay, and was not isolated due to its unstable nature. In a typical run, the bisꢀ
ACS Paragon Plus Environment

Page 21 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
fatty acid derivative (4Z,4'Z,7Z,7'Z,10Z,10'Z,13Z,13'Z,16Z,16'Z,19Z,19'Z)ꢀN,N'ꢀ
4
4
(
disulfanediylbis(ethaneꢀ2,1ꢀdiyl))bis(docosaꢀ4,7,10,13,16,19ꢀhexaenamide) (125 mg, 0.162
mmol) was taken up in absolute EtOH (1.75 mL). Racemic dithiothreitol (DTT, 30 mg, 0.194
mmol) was added at rt, followed by aqueous NaOH (1 N solution, 250 ꢁL). The resulting
reaction mixture was stirred at rt for 30 minutes to allow for a complete reduction of the disulfide
group. It was then diluted with DMSO (4.5 mL) to form a 50 mM DMSO stock solution of 1.
This freshly prepared stock solution was purged with inert gas and used directly for the assay.
This stock solution should not be kept at rt for more than 3 hr. If necessary, the stock solution
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
o
o
could be stored at ꢀ80 C up to 5 days or at ꢀ20 C for up to 3 days. Upon thawing to rt, it should
be used (without refreezing) within 3 hr and then discarded appropriately.
N-(2-((2-((4Z,7Z,10Z,13Z,16Z,19Z)-docosa-4,7,10,13,16,19-
hexaenamido)ethyl)disulfanyl)-2-methylpropyl)nicotinamide (5): (4Z,7Z,10Z,13Z,16Z,19Z)ꢀ
Nꢀ(2ꢀ((1ꢀAminoꢀ2ꢀmethylpropanꢀ2ꢀyl)disulfanyl)ethyl)docosaꢀ4,7,10,13,16,19ꢀhexaenamide (10,
3
.2 g, 6.5 mmol) and nicotinoyl chloride (1.8 g, 13 mmol) were taken up in CH Cl (30 mL).
2 2
o
Triethylamine (3 mL, 19.5 mmol) was added dropwise at 0 C. The resulting mixture was stirred
at rt for 18 hr. It was then diluted with water and extracted with CH Cl . The combined organic
2
2
layers were washed with water (2 ×50 mL), brine (50 mL), dried over Na SO , and concentrated
2
4
under reduced pressure. The resulting residue was purified by silica gel chromatography
(pentanes/EtOAc) to afford Nꢀ(2ꢀ((2ꢀ((4Z,7Z,10Z,13Z,16Z,19Z)ꢀdocosaꢀ4,7,10,13,16,19ꢀ
hexaenamido)ethyl)disulfanyl)ꢀ2ꢀmethylpropyl)nicotinamide (2.2 g, 57% yield) as a light brown
1
oil. H NMR (400 MHz, DMSOꢀd6): δ 9.01 (s, 1H), 8.71ꢀ8.65 (m, 2H), 8.21ꢀ8.17 (m, 1H), 8.04ꢀ
8.00 (m, 1H), 7.52ꢀ7.47 (m, 1H), 5.39ꢀ5.25 (m, 12H), 3.49ꢀ3.47 (m, 2H), 3.35ꢀ3.28 (m, 3H),
13
2.82ꢀ2.78 (m, 11H), 2.29ꢀ2.22 (m, 2H), 2.13ꢀ1.76 (m, 4H), 1.29 (s, 6H), 0.94ꢀ0.88 (m, 3H); C
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 56
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
NMR (100 MHz, DMSOꢀd6, 5 overlapping peaks corresponding to the CH group present in the
omegaꢀ3 fatty acid portion): δ 172.08, 165.92, 152.26, 148.98, 135.59, 131.99, 130.54, 129.27,
1
28.58, 128.52, 128.36, 128.31, 128.21, 128.15, 127.39, 123.81, 51.95, 48.35, 39.18, 38.69,
5.66, 25.79, 25.69, 25.65, 25.60, 23.56, 20.53, 14.59; MS calcd for C H N O S : 595.3;
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
3
3
4
49
3
2 2
+
+
+
found, 596.3 [M+H] . High resolution MS (ES ) calcd for C H N O S (M + H ) m/z,
3
4
49
3
2 2
596.3270; found, 596.3345.
2-(Pyridin-2-yldisulfaneyl)ethan-1-amine (8): A solution containing 1,2ꢀdi(pyridinꢀ2ꢀ
yl)disulfane (7, 26 g, 0.227 mmol) in MeOH (200 mL) was added dropwise at rt to a solution
containing cysteamine (50 g, 0.227 mmol) in MeOH (200 mL). The resulting reaction mixture
was stirred at rt for 2 h under an inert atmosphere of nitrogen and then concentrated under
reduced pressure.
The resulting residue was purified by silica gel chromatography
1
(
CH Cl /MeOH = 10/1) to afford 2ꢀ(pyridinꢀ2ꢀyldisulfanyl)ethanꢀ1ꢀamine (39 g, 92% yield). H
2
2
NMR (400 MHz, DMSOꢀd6): δ 8.50 (m, 1 H), 7.78ꢀ7.92 (m, 2 H), 7.28 (m, 1 H), 4.58 (br s, 2
13
H),
2.80ꢀ3.00
(m,
4
H).
C
NMR
(100
MHz,
=
DMSOꢀd6):
δ 159.37, 150.19, 138.65, 122.28, 119.96, 49.04, 38.49. Rf
0.22
+
(
CH Cl /MeOH/AcOH=85:15:5). MS calcd for C H N S : 186.03; found, 187 [M+H] .
2
2
7
10
2 2
(4Z,7Z,10Z,13Z,16Z,19Z)-N-(2-(Pyridin-2-yldisulfaneyl)ethyl)docosa-4,7,10,13,16,19-
hexaenamide (9): A mixture containing 2ꢀ(pyridinꢀ2ꢀyldisulfanyl)ethanꢀ1ꢀamine (8, 5 g, 26.8
mmol), DHA (9.2 g, 26.8 mmol), and HATU (10.2 g, 26.8 mmol) were taken up in CH Cl (100
2
2
mL) and stirred at rt. Triethylamine (18 mL, 40.3 mmol) was then added dropwise at rt. The
resulting reaction mixture was stirred at rt for 18 hr. It was then diluted with water and extracted
with CH Cl . The combined organic layers were washed with water (3 × 100 mL), brine (100
2
2
ACS Paragon Plus Environment

Page 23 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
mL), dried over anhydrous Na SO , and concentrated under reduced pressure. The resulting
2
4
residue was purified by silica gel chromatography (pentanes/EtOAc) to afford
(4Z,7Z,10Z,13Z,16Z,19Z)ꢀNꢀ(2ꢀ(pyridinꢀ2ꢀyldisulfanyl)ethyl)docosaꢀ4,7,10,13,16,19ꢀ
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
hexaenamide (10 g, 75% yield) as a light brown oil. H NMR (400 MHz, CDCl ): ): δ 8.49 (d, J
3
= 4.4 Hz, 1 H), 7.65 (m, 1 H), 7.53 (d, J = 8.0 Hz, 1 H), 7.10ꢀ7.20 (m, 2 H), 5.22ꢀ5.40 (m, 12 H),
3
.55 (m, 2 H), 2.95 (m, 2 H), 2.80ꢀ2.95 (m, 8 H), 2.42 (m, 2 H), 2.05 (m, 2 H), 0.97 (t, J = 7.2
1
3
Hz, 3 H). C NMR (100 MHz, CDCl , 3 overlapping peaks due to the CH group of the omegaꢀ3
3
fatty acid) δ 172.46, 165.75, 159.17, 149.73, 137.01, 132.04, 129.25, 128.56, 128.11, 128.08,
1
2
4
27.87, 127.01, 121.31, 121.06, 38.87, 38.62, 37.23, 36.60, 31.43, 29.38, 25.84, 25.64, 25,54,
3.48, 20.57, 14.3. R = 0.50 (EtOAc/Hexane=1:2). MS calcd for C H N OS : 496.3; found,
f
29 40
2
2
+
97.5 [M+H] .
(4Z,7Z,10Z,13Z,16Z,19Z)-N-(2-((1-Amino-2-methylpropan-2-yl)disulfaneyl)ethyl)docosa-
4
,7,10,13,16,19-hexaenamide (10): 1ꢀAminoꢀ2ꢀmethylpropaneꢀ2ꢀthiol (1.14 g, 8 mmol) was
added dropwise at rt to a solution containing (4Z,7Z,10Z,13Z,16Z,19Z)ꢀNꢀ(2ꢀ(pyridinꢀ2ꢀ
yldisulfanyl)ethyl)docosaꢀ4,7,10,13,16,19ꢀhexaenamide 9 (4 g, 8 mmol) in a 1:1 mixture of
MeOH/DMF (10 mL). The resulting reaction mixture was stirred at rt for 18 hr. It was then
diluted with water and extracted with EtOAc. The combined organic layers were washed with
water (2 ×50 mL), brine (50 mL), dried over Na SO , and concentrated under reduced pressure.
2
4
The resulting residue was purified by silica gel chromatography (pentanes/EtOAc) to afford
(4Z,7Z,10Z,13Z,16Z,19Z)ꢀNꢀ(2ꢀ((1ꢀaminoꢀ2ꢀmethylpropanꢀ2ꢀyl)disulfanyl)ethyl)docosaꢀ
1
4
,7,10,13,16,19ꢀhexaenamide (3.2 g, 81% yield) as a light brown oil. H NMR (400 MHz,
CDCl ): 6.37 (m, 1 H), 5.25ꢀ5.40 (m, 12 H), 4.30ꢀ4.50 (m, 2 H), 3.55 (m, 2 H), 2.95 (s, 2 H),
3
2.75ꢀ2.90 (m, 12 H), 2.40 (m, 2 H), 2.25 (m, 2 H), 2.05 (m, 2 H), 1.36 (s, 6 H), 0.97 (t, J = 7.2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 56
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
13
Hz, 3 H). C NMR (100 MHz, CDCl , 4 overlapping peaks due to CH group of the omegaꢀ3
3
fatty acid): 172.88, 132.05, 129.33, 128.58, 128.30, 128.29, 128.27, 128.21, 128.11, 128.07,
1
27.87, 127.01, 50.90, 49.23, 38.99, 38.49, 36.35, 29.71, 25.64, 25.54, 25.46, 23.36, 20.58,
4.32. R = 0.25 (CH Cl /MeOH=10:1). MS calcd for C H 2 OS : 490.31; found, 491.30
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
1
f
2
2
28 46 3
2
+
[M+H] .
5
-((2-((4Z,7Z,10Z,13Z,16Z,19Z)-Docosa-4,7,10,13,16,19-hexaenamido)ethyl)disulfaneyl)-
2
-nitrobenzoic acid (15): In a typical run, (4Z,7Z,10Z,13Z,16Z,19Z)ꢀNꢀ(2ꢀ(pyridinꢀ2ꢀ
yldisulfanyl)ethyl)docosaꢀ4,7,10,13,16,19ꢀhexaenamide (compound 9, 1.0 g, 2.02 mmol) was
taken up in 20 mL of MeOH/DMF (1:1) along with 5,5'ꢀdisulfanediylbis(2ꢀnitrobenzoic acid)
(
Ellman’s reagent, 0.8 g, 2.02 mmol). The resulting reaction mixture was stirred at rt for 4 h.
Then it was concentrated under reduced pressure and purified by preparativeꢀHPLC using an
aqueous mixture of CH CN, buffered with 0.1% TFA, to provide 5ꢀ((2ꢀ
3
(
(4Z,7Z,10Z,13Z,16Z,19Z)ꢀdocosaꢀ4,7,10,13,16,19ꢀhexaenamido)ethyl)disulfanyl)ꢀ2ꢀ
nitrobenzoic acid (0.337 g, 28.6% yield) as a brown oil. This Ellman’s adduct was then used to
prepare the standard curves for the quantification of the fatty acid cysteamine conjugate 1 in the
various mouse, rat and dog PK studies. This Ellman’s adduct was stable in the ꢀ80 freezer for
approximately 2 months. After this period, a fresh batch of the Ellman’s adduct would need to be
1
remade. H NMR (400 MHz, DMSOꢀd6): δ 8.02ꢀ8.05 (m, 2 H), 7.88ꢀ7.91 (m, 2 H), 5.28ꢀ5.36
(m, 12 H), 3.28ꢀ3.33 (m, 2 H), 2.75ꢀ2.90 (m, 12 H), 2.23ꢀ2.25 (m, 2 H), 2.00ꢀ2.11 (m, 4 H), 0.89ꢀ
13
0
.93 (t, 3 H). C NMR (100 MHz, DMSOꢀd6, 3 overlapping peaks corresponding to the CH
group present in the omegaꢀ3 fatty acid portion): δ 174.52, 167.63, 145.78, 145.04, 132.17,
1
29.87, 129.49, 128.71, 128.52, 128.42, 128.19, 128.07, 127.99, 127.89, 127.13, 126.55, 124.78,
+
5
3.62, 38.57, 37.85, 36.27, 25.77, 25.76, 25.75, 25.68, 23.51, 20.70, 14.42. MS (ES ) calcd for
ACS Paragon Plus Environment

Page 25 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
+
C H N O S 584.24; found 585 [M+H] .This basic procedure could be repeated to form the
31
46
2
5 2
corresponding Ellman’s adduct for the thiol metabolites shown in Figure 12.
Immunoblotting: Primary CF cells (homozygous F508delꢀCFTR hBE, source: ChanTest,
KKCFFT006F) were prepared and grown on Snapwell filter inserts as previously described. The
test compounds, solubilized in FBS according to the procedure outlined above and diluted to the
desired concentration, were then added to the individual Snapwell filter inserts in differentiation
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
o
media at 37 C. Twentyꢀfour hours after the drug addition, cells were snap frozen and later lysed
in RIPA buffer. The amounts of proteins were determined by BioꢀRad protein assay. Fifty ꢁg
of total cell lysates were analyzed by immunoblotting with antiꢀBeclinꢀ1, antiꢀp62 and antiꢀLC3
antibodies. The immunoꢀreactivity was normalized with actin as the loading control. Data were
presented as Beclinꢀ1/actin, p62/actin and LC3ꢀII/LC3ꢀI ratio compared to vehicle treated
samples. Antibodies against LC3A/B antibodies (Cell signaling, 12741), Beclinꢀ1 (Cell
Signaling Technology, 3495), p62 (Cell Signaling Technology, 5114) and βꢀactin (Cell Signaling
Technology, 4970) were used as primary antibodies.
Determination of CFTR band B/band C: Materials and reagents: 1) Lysis Buffer (Thermo
Scienfic Pierce IP Lysis Buffer, PIꢀ87788); 2) Protease Inhibitors: Complete Ultra Tablets, Mini,
EDTAꢀFree (ThermoFisher Scientific 88666); 3) Sample Loading Buffer: NuPAGE LDS Sample
Buffer, 4x (Life Technologies NP0007) NuPAGE Reducing Agent, 10x (Life Technologies
NP0004); 4) Running Buffer: NuPAGE TrisꢀAcetate SDS Running Buffer, 20x (ThermoFisher
Scientific LA0041); 5) Gel: NuPAGE 3ꢀ8% TrisꢀAcetate Gel, 1.5mm x 15 wells
(
EA03785BOX); 6) Transfer System: Xcell II Blot Module (ThermoFisher Scientific EI0002); 7)
Transfer Buffer: NuPAGE Transfer Buffer, 20x (Life Technologies NP0006ꢀ1); 8) Blocking
Buffer: Blotting Grade Blocker (BioꢀRad 170ꢀ6404, 5% in TBS containing 0.2% Tweenꢀ20); 9)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 56
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
PVDF: Amersham HyBond (GE Healthcare Life Sciences 10600023); 10) Chemiluminescent
Detection Reagent: SuperSignal West Pico OR SuperSignal West Femto (ThermoFisher
Scientific, 34080 and 34095, respectively).
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
The test compounds, solubilized in FBS according to the procedure outlined below and diluted
to the desired concentration, were added to primary homozygous F508delꢀCFTR hBE cells,
o
grown on individual Snapwell filter inserts, in differentiation media at 37 C. Twenty four hours
after the drug addition, cells were lysed on ice, using iceꢀcooled buffers, as follows: 1) The
growth medium was removed and cells were rinsed with 2 × 500ꢁL ice cold DPBS. 2) The
second wash was removed and 150 ꢁL of DPBS was added to each well. 3) Cells were gently
removed from Snapwell inserts by directly rinsing them with DPBS. 4) Cells were collected in
suspension and placed into microcentrifuge tube on ice. 5) An additional 200 ꢁL of ice cold
DPBS was added to each well and steps 3ꢀ4 were repeated. 6) An additional 700 ꢁL of ice cold
DPBS was added to each microcentrifuge tube. 7) Tubes were centrifuged at 1,000 x g for 5
minutes at 4 °C. 8) The supernatant was removed as much as possible without disturbing the cell
pellet (some mixing of the pellet with mucus may occur). 9) Each cell pellet was gently reꢀ
suspended in 30 ꢁL of ice cold lysis buffer with protease inhibitor and incubated on ice for 20
min. 10) Samples were centrifuged at 12,500 x g for 10 min at 4 °C to pellet cellular debris. The
supernatant was removed to a fresh tube.
For the SDSꢀPAGE and electrophoretic transfer: Each sample was prepared by adding: a)
15ꢁL of the lysate; b) 5ꢁL 4X LDS Sample Buffer; c) 2.5ꢁL 10x Reducing Agent; d) 2.5ꢁL
MilliꢀQ water (Note: samples are not to be heated before loading gel). 2) The gel and running
buffer were prepared according to manufacturer’s protocol. Gel was run at 150V constant for 1
hr and proteins were transferred to PVDF overnight at 10V constant in Xcell II Blot Module.
ACS Paragon Plus Environment

Page 27 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
For the Western blot, the PVDF membrane was removed from transfer sandwich and placed
directly into 5% Blocking buffer. This was allowed to incubate for 1 hr at rt with shaking. The
primary antibody was prepared by adding 5ꢁL UNCꢀ596 primary antibody to 10mL Blocking
Buffer (per membrane). Membranes were incubated with primary antibody overnight at 4 °C
with gentle shaking. Membranes were removed from 4°C and rinsed 3 × 10 minutes in TBST at
rt. The secondary antibody was prepared by adding 1 ꢁL goat antiꢀMouse IgGꢀHRP in 10mL
Blocking Buffer (per membrane). Secondary antibody was applied to the membrane and
incubated for 1 hr at rt with shaking. Membranes were rinsed 3 × 10 minutes in TBST, and 2 ×
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
10 minutes in TBS. SuperSignal West Femto was applied to membranes and then allowed to
incubate for 5 minutes in low light. Membranes were exposed to Xꢀray film at regular intervals
for up to 1 hour. Membrane was washed with 3 × 5 minutes to remove substrate. The subsequent
primary antibody was prepared by adding 0.5ꢁL Mouse antiꢀNa/KꢀATPase antibody in 10mL
blocking buffer. Membranes were incubated with primary antibody for 1 hour at rt with shaking
and then washed 3 × 10 minutes in TBST. SuperSignal West Pico was applied to membranes and
incubated for 5 min in low light. Membranes were exposed to Xꢀray film at regular intervals for
up to 1 hour.
In order to increase sensitivity at the lower concentrations (e.g. 0.15 ꢁM of conjugate 1), the
following modified protocols were used: Cells were rinsed in DPBS two times, followed by
direct lysis in IP Lysis buffer (Thermo) for 20 minutes. Cellular debris was pelleted, and
supernatants were used directly in samples without determination of protein concentration.
Unheated samples were run on a 3ꢀ8% TrisꢀAcetate gel (Life Technologies) at 150V for 2.5
hours. Separated proteins were transferred to PVDF using the Xcell Blot Module (Life
Technologies) for 1 hour at 25V constant. After overnight transfer, blots were blocked in 5%
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 56
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
Blotting Grade Blocker (BioꢀRad) for 1 hour at room temperature (RT) and followed by primary
antibody treatment.
Hydrolysis experiment using FAAH-1: FAAHꢀ1ꢀ32ꢀ579 (the Nꢀterm transmembrane region
truncated) was cloned into pFastBacI (Invitrogen) with Nꢀterminal FLAG and Cꢀterminal His
tags and then expressed in SF9 insect cells. Compound 1 was tested in a hydrolysis experiment
using this recombinant FAAHꢀ1 at an enzyme concentration of 1 nM and substrate concentration
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
of 25 ꢁM. For the hydrolysis portion, compound 1 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.
Automated equivalent current (I ) assay for compound evaluation on F508del-CFTR
EQ
chloride channel function in primary cultures of CF hBE cells:
(TECCꢀ24 assay setup # 1, Rosalind Franklin University, primary homozygous F508delꢀCFTR
hBE cells, patient code CFFT0018I or CFFT0028H). Primary CF hBE cells from patients
homozygous for F508delꢀCFTR were grown on Costar 24ꢀwell HTS filter plates at an air liquid
interface for four to six weeks. Cells were fed three times a week with differentiation media and
o
maintained in an incubator at 37 C and 90% humidity. The mucus film was removed three days
before the experiment by incubating the apical surface of the cells with 70 ꢁL of differentiation
media plus 3 mM DTT for approximately 30 min followed by aspiration of the mucus and
media. The apical surface was washed again with 70 ꢁL of PBS for 30 min 24 hrs before the
cells were treated with test compounds. For the experiment, the test and control compounds
were applied to cells as prepared dilutions of DMSO stock solutions into FBS and differentiation
media to the basolateral side only. The fatty acid cysteamine conjugates are highly lipophilic and
ACS Paragon Plus Environment

Page 29 of 56
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
need to be solubilized with FBS in order to obtain the optimal effect. This was done by diluting
the DMSO stock solution with FBS prior to the serial dilution: 100 ꢁL of a 25 mM DMSO stock
solution of the test compound was diluted in 10.0 mL of FBS in a centrifuge tube to prepare an
intermediate 250 ꢁM intermediate dilution 10× stock (1% DMSO, 99% FBS). To prepare the
0.3 ꢁM solution of the fatty acid conjugate in 10 mL of differentiation media, 12 ꢁL of the 10×
stock solution was added to 9000 ꢁL of differentiation media and 988 ꢁL of 1% DMSO, 99%
FBS solution. Other concentrations of the test compound could be prepared by adding the
appropriate volume of the 10× stock solution to the differentiation media and 1% DMSO, 99%
FBS solution. Both the vehicle and the positive control group were similarly solubilized with
FBS and diluted with differentiation media to give a final concentration of 10 % FBS. Cells were
incubated with either the control or test compounds for 24 hr. For the chronic ivacaftor preꢀ
incubation protocol, all treatment groups (vehicle, positive control and fatty acid cysteamine
conjugateꢀtreated groups) were incubated along with 0.10 ꢁM of ivacaftor for 24 hr. For the
electrophysiological measurements, the differentiation media was changed to an experimental
media that have been supplemented with either the control or test compounds (in the same
concentration used during the 24 hr incubation period). HEPESꢀbuffered Coon’s F12 media
without serum or bicarbonate was used as the experimental assay solution. The fatty acid
cysteamine conjugates were solubilized in this experimental assay solution using the same FBS
solubilization protocol described above (substituting the differentiation media with the HEPESꢀ
buffered Coon’s F12 media without serum and bicarbonate). Cells were incubated for 4 hr in a
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
o
o
CO free incubator at 37 C. Cells were then placed in preꢀwarmed heating blocks at 36 C ±
2
0.5 for an additional 15 min before initiating the electrophysiological measurements.
ACS Paragon Plus Environment