Manganese-Based Contrast Agents for Magnetic Resonance Imaging of Liver Tumors: Structure-Activity Relationships and Lead Candidate Evaluation

Article abstract of DOI:10.1021/acs.jmedchem.8b00964

Gd-based MRI contrast agents (GBCAs) have come under intense regulatory scrutiny due to concerns of Gd retention and delayed toxicity. Three GBCAs comprising acyclic Gd chelates, the class of GBCA most prone to Gd release, are no longer marketed in Europe

Full text of DOI:10.1021/acs.jmedchem.8b00964

Subscriber access provided by University of Sunderland
Article
Manganese-Based Contrast Agents for Magnetic Resonance Imaging of Liver
Tumors: Structure Activity Relationships and Lead Candidate Evaluation
Junfeng Wang, Huan Wang, Ian Ramsay, Derek J. Erstad,
Bryan Fuchs, Kenneth Tanabe, Peter Caravan, and Eric M Gale
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.8b00964 • Publication Date (Web): 11 Sep 2018
Downloaded from http://pubs.acs.org on September 12, 2018
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 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 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.
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 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Manganese-Based Contrast Agents for Magnetic
Resonance Imaging of Liver Tumors: Structure-
Activity Relationships and Lead Candidate
Evaluation
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
Junfeng Wang,†‡ Huan Wang,†‡ Ian A. Ramsay,†‡ Derek J. Erstad,Ñ Bryan C. Fuchs,Ñ Kenneth K.
Tanabe,Ñ Peter Caravan,†§ and Eric M. Gale*†
†Athinoula A. Martinos Center for Biomedical Imaging, Department of Radiology, and §Institute
for Innovation in Imaging, Massachusetts General Hospital/ Harvard Medical School, 149
Thirteenth Street, Suite 2301, Charlestown, MA 02129.
ÑDepartment of Surgical Oncology, Massachusetts General Hospital/ Harvard Medical School,
WRN401, 55 Fruit Street, Boston, MA, 02114.
ACS Paragon Plus Environment
1

Journal of Medicinal Chemistry
Page 2 of 49
1
2
3
4
5
6
7
8
9
ABSTRACT: Gd-based MRI contrast agents (GBCAs) have come under intense regulatory
scrutiny due to concerns of Gd retention and delayed toxicity. Three GBCAs comprising acyclic
Gd chelates, the class of GBCA most prone to Gd release, are no longer marketed in Europe. Of
particular concern are the acyclic chelates that remain available for liver scans, where there is an
unmet diagnostic need and no replacement technology. To address this concern, we evaluated our
previously reported Mn-based MRI contrast agent, Mn-PyC3A, and nine newly synthesized
derivatives as liver specific MRI contrast agents. Within this focused library the transient liver
uptake and rate of blood clearance are directly correlated with logP. The complex Mn-PyC3A-3-
OBn emerged as the lead candidate due to a combination of high relaxivity, rapid blood clearance,
and avid hepatocellular uptake. Mn-PyC3A-3-OBn rendered liver tumors conspicuously hypo-
intense in a murine model and is wholly eliminated within 24 h of injection.
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
INTRODUCTION
Gd-based MRI contrast agents (GBCAs) are heavily relied upon for the radiologic diagnosis of
1-9
tissue and vascular abnormalities. However, regulatory agencies are currently re-assessing the
risk vs. benefit of GBCAs in response to well-documented safety concerns related to long-term
Gd retention.10-15 In 2017, the FDA announced a new class warning for all GBCAs and the
European Medicines Agency suspended marketing authorizations for three of the five Gd-based
agents comprised of acyclic chelators, which are more prone to Gd release than the class of agents
comprised of macrocyclic chelators (Figure S1).15-18 However, long-term Gd retention is also
documented in patients that have exclusively received macrocyclic GBCAs.19-21 EMA has allowed
two acyclic agents, Gd-DTPA-EOB and Gd-BOPTA, to remain available but according to the
EMA 2017 statement these agents are now restricted “to be used for liver scans because they are
ACS Paragon Plus Environment
2

Page 3 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
taken up in the liver and meet an important diagnostic need.”15 Liver-specific agents enable high-
sensitivity detection of liver metastases, differential diagnosis of malignant tissue from benign
abnormalities, and staging of hepatocellular lesions.22, 23 There are no liver specific macrocyclic
GBCAs or iodinated computed tomography (CT) contrast agents. Differential diagnosis cannot be
made using ultrasound imaging, CT, or contrast-free MRI.
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
The rising concerns over GBCA safety have underscored a pressing urgency to develop Gd-free
24-33
MRI contrast agents. Most research to date has focused on developing extracellular agents.
Considerably less effort has been placed on developing Gd-free liver-specific agents despite the
fact that liver cancer and metastatic liver disease are among the leading causes of cancer related
death.34, 35 In this regard, we sought to develop a Gd-free liver-specific contrast agent. We posit
that a Mn complex is the most likely GBCA alternative to succeed as a clinically viable agent.
Mn2+ is one of the few ions that can generate MRI contrast that is comparable to Gd3+.36 Unlike
Gd, Mn is a trace nutritional element that the human body can incorporate or excrete as required,
avoiding concerns of long-term accumulation.37
A new liver-specific agent should be a direct replacement for the GBCA standard of care. The
liver specificity of Gd-DTPA-EOB and Gd-BOPTA is imparted by peripheral lipophilic functional
groups that promote recognition by organic anion transporting peptides (OATPs) that are
abundantly expressed on the cell membranes of hepatocytes lining biliary sinusoids.38 Both liver
tumors and benign abnormalities such as focal nodular hyperplasia can be avidly enhanced by
contrast agents during the arterial, venous, or extracellular phases of distribution, but OATP-
mediated hepatocellular uptake enables differential diagnosis during what is referred to as the
delayed phase (15-60 min post injection) when the blood signal is diminished but liver
enhancement is strong.22, 23 Malignancies, which are comprised of non-OATP expressing cells (ie.
ACS Paragon Plus Environment
3

Journal of Medicinal Chemistry
Page 4 of 49
1
2
3
4
5
6
metastases) or OATP under-expressing cells (primary liver cancers), are rendered conspicuously
hypo-intense against normal liver during the delayed phase.
7
8
9
A number of design criteria must be met in developing a suitable Mn complex for liver-specific
imaging. We require a Mn complex that possesses relaxivity (r1 = (D1/T1) normalized to mM
contrast agent) that is comparable or better than GBCAs. The complex must be inert to Mn release
so that it can be safely delivered as a bolus injection in order to see dynamic phase tumor
enhancement. The complex must also exhibit a combination of rapid blood clearance and
preferential hepatocyte uptake in order to provide differential contrast enhancement between
benign and malignant tissue within minutes of injection. The complex must also be rapidly
eliminated as the probability of exposure to dissociated metal ions increases with dwell time.39
We hypothesized that a lipophilic, anionic Mn complex based on the previously reported Mn-
PyC3A chelator could meet the criteria required for a successful liver-targeting agent. It has been
demonstrated that Mn-PyC3A possesses relaxivity that is comparable to GBCAs and is also highly
resistant to Mn release in vivo.40, 41 However, it is difficult to predict the molecular attributes
needed to impart the pharmacokinetics and biodistribution requisite for a liver-targeting agent
(transient hepatocellular uptake, rapid blood clearance). Liver-targeting via increased lipophilicity
can promote rapid elimination from the bloodstream,42 but on the other hand increased lipophilicity
can also promote greater affinity for plasma proteins which has been shown in many cases to
prolong the circulatory lifetime of MRI contrast agents.43, 44 Slow blood clearance will blunt the
capability of a liver targeting agent to aid in differential diagnoses, or prolong the time to delayed
phase imaging. In developing a liver-specific agent, it is critical to understand the structure-activity
relationships (SAR) that influence the protein binding, relaxivity, pharmacokinetics,
biodistribution, and elimination. In this regard, we examined SAR for a focused library of anionic
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
4

Page 5 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
Mn complexes comprising Mn-PyC3A (referred to as 1 from here on) and nine newly synthesized
derivatives of varying lipophilicity (2-10, Chart 1) in order to identify a lead candidate Mn-based
contrast agent for liver imaging.
9
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
Synthesis. Complexes 2-10 were derived from Mn-PyC3A using the pyridyl-N donor group as the
modular handle (Chart 1). We targeted modifications that were unlikely to negatively impact
relaxivity or stability, were amenable to further modulation, and can be easily scaled up for large
batch preparations. Substitutions were thus introduced at the 3, 4, and 5 positions of the N-pyridyl
donor group, remote from the rapidly exchanging water co-ligand that is key for high-relaxivity.40
Pyridyl functionalization does not introduce any new chirality and the –OR or -COOR functional
groups present in 2-10 are amenable to further modulation.
Chart 1. A focused library of ten structurally related Mn-PyC3A-derived complexes were
evaluated as liver-specific MRI contrast agents.a
a(A) Structures of Mn-PyC3A and Mn-PyC3A-derived complexes, (B) derivatives of varying
lipophilicity were prepared via modifications to the 2-methylpyridine arm of the Mn chelates.
ACS Paragon Plus Environment
5

Journal of Medicinal Chemistry
Page 6 of 49
1
2
3
4
The 2-methylpyridine-derived donor arms of 2 and 3 were introduced to the ligand as the
5
6
commercially
available
2-(chloromethane)-3,5-dimethyl-5-hydroxypyridine.
The
2-
7
8
9
methane(pyridine)-derived arms of 4-10 were introduced as the corresponding 2-
(bromomethane)pyridine synthons (14d-i) which were synthesized as described in Scheme 1 and
Table 1. Briefly, the 2-methylpyridine derived starting material was converted to the
corresponding N-oxide (12d-i) by mCPBA oxidation before heating in neat acetic anhydride to
generate the corresponding pyridine-2-ylmethyl acetate derivative (13d-i). In most instances the
starting 2-methylpyridine derivatives were purchased from commercial sources, but 11g-i were
prepared by O-benzylation of the methyl nicotinate or the corresponding 5- and 3-hydroxy-2-
methylpyridines, respectively. The acetyl-O group of 13d-i was then removed to yield pyridine-2-
ylmethanol derived 14d-i either by saponification with KOH or by treatment with acetyl chloride.
Compounds 14d-i were converted to the 15d-i by treatment with PBr3. 2-(chloromethane)-3,5-
dimethyl-5-hydroxypyridine and 15d-i were then coupled with the common synthon, N,N’,N’-
trans-1,2-cyclohexylenediaminetri-tert-butylacetate, to yield O-protected ligand 16a,d-i.
Subsequent O-deprotection yielded ligands 17a-i. Complexes 2-10 were prepared by addition of 1
mol equiv. MnCl2 to an aqueous solution of ligand and then adjusted the pH to 6.5, as previously
described.40
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
Scheme 1. Synthesis of complexes 2-10a-e
ACS Paragon Plus Environment
6

Page 7 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
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
aReagents and conditions: (i) mCBPA, CH2Cl2, RT, 16 h, (ii) Ac2O, reflux, 5-7 h, (iii) 5:1 THF:
2M NaOH, RT, 2-7 h, (iv) Acetyl chloride, MeOH, RT, 16h, (v) PBr3, CH2Cl2, RT, 5 h. (vi).
N,N,N’-trans-1,2-diaminocyclohexanetri-tert-butylacetate, KI, K2CO3, MeCN, reflux, 20 h. (vii)
TFA. RT, 5 h, (viii) BBr3, CH2Cl2, -78 °C to RT, 16 h, (ix) MnCl2, H2O, adjust to pH 6.5. b11g-i
were prepared by benzylation of methyl nicotinate, 5- or 3-hydroxy-2-methylpyridines,
c
respectively. 12d-f were prepared from commercially available 2-methylpyridine derivatives.
e
d16a was prepared from 2-(chloromethane)-3,5-dimethyl-5-hydroxypyridine. Both 17a and 17b
prepared from 16a, both 17c and 17e prepared from 16e.
Table 1. Guide to intermediates synthesized en route to complexes 2-10.
Complex Corresponding R1
intermediates
denoted by:
R2
R3
1
2
3
4
5
6
7
8
-H
-H
-H
a
b
c
d
e
f
-CH3
-CH3
-OH
-H
-OCH3
-OH
-H
-CH3
-CH3
-H
-OCH3
-H
-H
-OCH3
-H
-C(O)OEt -H
-H -H
-H
g
-C(O)OBn
ACS Paragon Plus Environment
7

Journal of Medicinal Chemistry
Page 8 of 49
1
2
3
4
5
6
9
h
i
-H
-OBn
-H
-H
-OBn
-H
10
7
8
9
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
Physical characterization. The relative polarities of complexes 1-10 are described by the
estimated 1-octanol/ water partition co-efficient (est. log P) values determined from reverse phase
HPLC measurements. HPLC retention times were cast as est. log P using a calibration curve
generated from 6 reference compounds of known log P45 (Figure S2). The est. log P of 1-10 range
from 0.455 to 1.15, respectively (Table 2). Estimations of plasma protein binding by ultrafiltration
assay indicate that 1-10 have moderate plasma protein binding affinity ranging from Ka = 0.13 to
0.94 mM-1 (Table 2). Protein binding is strongly and positively correlated with est. log P (Figure
1A). A parallel ultrafiltration assay performed in 4.5% wt/vol human serum albumin (HSA)
indicates that plasma protein binding is largely attributable to HSA, which is present in plasma at
600-700 µM and is a known binding target for lipophilic anions (Figure S12).
The r1 values of complexes 1-10 range between 2.1 to 3.0 mM-1s-1 in pH 7.4 Tris buffer (1.4T,
37 °C) and are consistent with ternary complexes possessing a rapidly exchanging water co-ligand
(Table 2). However, the r1 values in human blood plasma are significantly greater and range from
3.5 to 12.2 mM-1s-1. The plasma r1 values represent between a 70% and 360% increase over the
values recorded in buffer. Blood plasma r1 correlates strongly with both est. log P (Figure 1B) and
plasma protein binding (Figure 1C). The r1 increase observed upon protein binding arises from
coupling of the rotational motions of the Mn complexes to more slowly tumbling protein
macromolecules.46, 47
The r2 values range between 5.1 and 6.4 mM-1s-1 in pH 7.4 Tris buffer but can increase by 60%
to 440% in blood plasma due to protein binding interactions. Consistent with the r1 data, plasma r2
ACS Paragon Plus Environment
8

Page 9 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
is also strongly and positively correlated with est. log P (r = 0.7243, P = 0.0178) and with plasma
Ka (r = 0.8016, P = 0.0053) (Figures 1D,E).
7
8
9
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
Table 2. Retention time (tR, method 2), est. logP, r1 and r2 in pH 7.4 Tris buffer and human blood
plasma, and protein association constants (Ka) in human blood plasma. tR corresponds to the major
diastereomer of 1-10, the tR of the minor diastereomer is typically within <0.2 min of the major
diastereomer (Figures S3-11). Relaxivity measurements were performed at 1.4T and 37 °C. Protein
binding was estimated via ultra-filtration assay. Data reported as mean±SD.
tR
(min)
r1 pH 7.4 Tris r2 pH 7.4 Tris r1 plasma
r2 plasma Ka plasma
(mM-1s-1)
est. logP
(mM-1s-1)
2.1±0.1
2.2±0.5
2.5±0.2
2.3±0.2
2.4±0.1
2.4±0.1
2.6±0.2
3.0±0.2
2.6±0.1
2.6±0.2
(mM-1s-1)
5.1±0.4
5.1±0.2
5.2±0.2
5.4±0.3
5.5±0.2
5.4±0.1
6.0±0.8
6.4±0.5
5.6±0.5
6.4±0.5
(mM-1s-1)
3.5±0.5
10.4±0.9
5.1±0.5
5.2±1.5
6.6±0.5
7.3±1.1
6.4±0.7
12.2±1.2
9.8±1.0
9.0±1.4
(mM-1)
1
2
3
4
5
6
7
8
9
10
6.96
8.05
7.03
6.61
7.40
7.48
8.01
8.63
8.57
8.86
0.575
0.922
0.601
0.455
0.748
0.750
0.909
1.09
0.13±0.095
0.63±0.062
0.22±0.077
0.22±0.12
0.60±0.040
0.41±0.18
0.58±0.080
0.60±0.12
0.94±0.067
0.87±0.23
8.2±2.1
34.5±3.8
12.4±2.1
14.2±1.7
23.8±1.8
16.3±1.4
11.6±4.2
25.3±2.1
28.5±2.3
29.1±3.7
1.07
1.15
ACS Paragon Plus Environment
9

Journal of Medicinal Chemistry
Page 10 of 49
1
2
3
4
5
6
7
8
9
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
Figure 1. Plasma protein affinity and plasma relaxivity (1.4T, 37 °C) of 1-10 are tightly correlated
with estimated log P. Data reported as mean±SD, solid lines represent fits to the data. (A) Plasma
protein Ka determined by ultrafiltration assay plotted as a function of est. log P (r = 0.9083, P =
0.0003), (B) Plasma r1 plotted as a function of est. log P (r = 0.8446, P = 0.0021), (C) Plasma r1
plotted as a function of Ka (r = 0.7527, P = 0.0120), (D) Plasma r2 plotted as a function of est. log
P (r = 0.7243, P = 0.0178), and (E) Plasma r2 plotted as a function of Ka (r = 0.8016, P = 0.0053).
The plasma relaxivity data plotted as a function of Ka confirms that the r1 and r2 dependence on
est. log P is due to an increased affinity for plasma proteins.
Dynamic MRI. In order to study the influence of changing polarity on pharmacokinetics and
biodistribution of the Mn complexes we examined the in vivo behavior of three representative
complexes in mice by dynamic MRI using a 4.7T small bore imager. We chose complexes 4 (R1
ACS Paragon Plus Environment
10

Page 11 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
= -OH, R2 = R3 = -H), 6 (R1 = -OCH3, R2 = R3 = -H), and 10 (R1 = -OBn, R2 = R3 = -H) for
evaluation, which represent the complexes of lowest, intermediate, and greatest lipophilicity. Mice
receiving 0.1 mmol/kg of Mn complex intravenously were scanned dynamically starting pre-
injection and up to 22 min post-injection using a 3D T1-weighted fast low-angle low-shot (FLASH)
sequence. Representative 3D maximum intensity projections of T1-weighted images recorded at
baseline, 1 min, 5 min, 12 min, and 22 min post-injection are shown in Figure S13. To compare
liver uptake between the three agents we included in the scanning protocol a 2D T1-weighted
FLASH acquisition capturing the liver and adjacent psoas muscle at 7 min post injection (delayed
phase), at which point blood signal is diminished but liver signal is near its peak (Figure 2).
Blood signal recorded in the vena cava normalized to pre-contrast values (nSI) plotted as a
function of time reveal bi-phasic blood elimination consistent with extracellular distribution and
rapid contrast agent washout (Figures 2A-C, Table 3). Mean blood nSI during the 0 to 6 min time
points is greater for the more lipophilic agents. We attribute this effect to the logP dependence on
plasma protein affinity, which will result in higher intravascular concentrations of the more
lipophilic contrast agents during the initial distribution equilibrium time points.
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
The elimination half-lives (t1/2) estimated from the blood signal in the dynamic scans indicates
that blood half-life correlates inversely with est. log P. Complexes 4, 6, and 10 exhibit t1/2 of
5.7±1.0 min, 3.6±1.4 min, and 2.4±0.4 min, respectively, Table 3. The elimination t1/2 of 10 is
significantly faster than that recorded for 4 and 6. The increasingly rapid blood clearance of the
more lipophilic complexes reflects increased hepatobiliary contribution to excretion. Liver nSI of
complex 10 is 105% and 55% greater than that generated by 6 and 4 at 5 min post injection (Figures
2D-F). Clearance into the bile is evidenced by an increase in gall bladder signal observed during
liver washout, shown in Figures S13. This observation is consistent with liver accumulation via a
ACS Paragon Plus Environment
11

Journal of Medicinal Chemistry
Page 12 of 49
1
2
3
4
5
6
7
8
9
hepatocellular mechanism and with hepatobiliary excretion. Quantification of liver and kidney nSI
area under the curve (AUC) recorded up to 22 min post-injection is also consistent with increased
fractional excretion via the hepatobiliary path for the more lipophilic agents (Table 3).
The increase in delayed phase liver vs. muscle contrast-to-noise ratio (DCNR) generated by 10
(61.6±15.4) is greater than that generated by 6 (48.2±15.2) and significantly greater than that
generated by 4 (22.4±16.7) (Figures 3A-G). Ex vivo analysis of liver and blood Mn content 10
min post injection are also consistent with delayed phase contrast resulting from transient
hepatocyte uptake and confirm that liver uptake increases with relative lipophilicity (Figure 3H,
Table S1, Figure S14). Complex 10 provides the strongest liver enhancement of the three
complexes tested but the real advantage for liver tumor imaging is evidenced by comparing liver-
to-blood Mn content ratios 10 min after injection (Figure 3I). The delayed phase liver-to-blood
ratio measured for 10 is 360% and 520% greater than for 6 and 4, respectively. The rapid blood
clearance and strong delayed liver enhancement effect of 10 make this complex the lead candidate
for liver imaging indications.
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
12

Page 13 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
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
Figure 2. MR signal as a function of time in the blood (A-C) and liver (D-F) for three
representative Mn-based contrast agents. Signal intensity normalized to pre-injection image (nSI)
in the blood and liver are plotted as a function of time following injection of complex 4 (A,D),
complex 6 (B,E), and complex 10 (C,F). nSI reported as mean±SD, solid lines represent best fits
to the data. (N=4 for each complex).
Table 3. The pharmacokinetics of complexes 4, 6, and 10 were estimated from the dynamic MRI
data, N =4 each. Data reported as mean±SD. *P = 0.0419 vs. complex 4, **P = 0.0030 vs. complex
4.
ACS Paragon Plus Environment
13

Journal of Medicinal Chemistry
Page 14 of 49
1
2
3
4
5
6
7
8
9
logP elimination t1/2
AUC0-22 liver
(nSI*min-1)
AUC0-22 kidney
(nSI*min-1)
(min)
5.7±1.0
3.6±1.4*
2.4±0.4**
4
0.46
0.75
1.15
10.2±6.52
21.1±12.2
23.5±4.71
21.3±8.30
20.4±4.50
14.2±4.70
6
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
Figure 3. (A-F) Axial T1-weighted images recorded at 4.7T capturing the liver and psoas muscle
collected prior to (A-C) and 7 min after (D-F) injection of 4 (A,D), 6 (B,E), and 10 (C,F). (G)
Liver vs. muscle DCNR recorded 7 min post-injection indicates that liver enhancement is greatest
for the most lipophilic complex 10. (DCNR for 10 vs. 4, P = 0.0164, N=4 for 4 and 6, N=7 for 10).
(H) Injected Mn recovered from the liver 10 min after injection is greatest for 10 (N=4 each). Note
that significant washout of 10 has already occurred by 10 min (see Fig 3F) (I) Liver:blood Mn
ratio 10 min after injection is significantly is higher for 10 than for 4 or 6 (P < 0.0001 for both).
ACS Paragon Plus Environment
14

Page 15 of 49
Journal of Medicinal Chemistry
1
2
3
4
The rapid blood clearance and strong transient liver uptake of 10 are ideal properties for an agent
5
6
designed to detect liver malignancies. Data reported as mean±SD. *P < 0.05, **** P < 0.0001.
7
8
9
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
Ex vivo quantitation of Mn content. The three complexes tested in mice are efficiently excreted.
The rates of whole body elimination reflect the blood clearance half-lives. Ex vivo analysis of
tissue Mn content recorded in 9 tissues (blood, bone, brain, heart, kidney, liver, lung, muscle,
spleen) 90 min post injection yielded 9.0±4.5% and 2.1±0.81% recovery of the injected dose
following injection of 4 and 6, respectively (Figure 4A, Tables S2-3). No significant Mn recovery
was observed following injection of the liver targeting agent 10. Mn levels had returned to baseline
within 24h of injection for all three complexes (Figure 4B, Table S4).
Figure 4. Mn levels in mice recorded 90 min (A) or 24 h (B) after injection 0.1 mmol/kg of
complex 4 (red bars), 6 (green bars), or 10 (blue bars). Endogenous Mn levels in naïve mice are
depicted by black bars. Data reported as mean±SD.
Evaluation of lead candidate in a murine liver tumor model. The efficacy of complex 10 to
detect liver tumors was tested in a mouse model. Figures 5A,B show axial T1-weighted images of
ACS Paragon Plus Environment
15

Journal of Medicinal Chemistry
Page 16 of 49
1
2
3
4
5
6
7
8
9
the liver at the level of the tumor prior to and during the delayed phase following injection of 10.
The small (<1.3 mm), hepatocyte-devoid lesion is barely perceptible in the pre-contrast image but
is rendered conspicuously hypo-intense after contrast enhancement (DCNR = 69.1±8.5 for pre- vs.
post-injection CNR, Figure 5C). The contrast enhanced imaging results were independently
confirmed by a T2-weighted pathology scan recorded at baseline and by ex vivo histopathologic
examination of the liver (Figure S15).
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
Figure 5. Liver imaging of an orthotopic mouse model of colorectal liver metastasis with complex
10 at 4.7T. (A and B) T1-weighted axial images of the liver at the level of the tumor (arrow) prior
to and during the hepatocellular phase after injection of 10, respectively. (C) Liver parenchyma
vs. tumor CNR prior to and following treatment with 10 are 14.8±3.1 and 83.8±11.5, respectively,
(N=3, P = 0.005, paired t-test). Data reported as mean±SD.
DISCUSSION
We evaluated a focused library of ten mono-anionic Mn complexes comprising Mn-PyC3A and
nine amphiphilic derivatives of varying polarity as liver specific MRI contrast agents. Across this
series of molecules the plasma protein binding affinity, rate of elimination, hepatocellular
accumulation, and delayed phase liver enhancement increase with est. logP. The blood plasma
relaxivity at 1.4T also correlates with est. logP due to increased a protein binding, but relaxation
ACS Paragon Plus Environment
16

Page 17 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
enhancement due to protein binding is a low field phenomenon and the effect will be largely
attenuated at field strength >3.0T. Each of the three complexes evaluated provided strong contrast
enhancement of the blood vessels and strong delayed phase liver specific contrast in vivo. The
most lipophilic contrast agent in the series, complex 10, demonstrated the optimal combination of
properties desired for a liver specific agent – high-relaxivity, rapid blood clearance, and avid
hepatocellular accumulation.
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
The SAR generated during this study suggests that the liver specific properties of Mn-PyC3A-
derived chelates can be reliably tuned by modulating est. logP. However, we have not yet
determined whether there is a threshold at which point increasing est. logP yields diminishing
returns. We also note that our SAR evaluation focuses exclusively on mono-anionic complexes
and that the est. logP dependence on protein binding, pharmacokinetics, and hepatocellular uptake
may differ for complexes of different charge. Regardless, the lead candidate agent identified in
this study represents a viable candidate for development as a direct replacement for the liver
specific GBCAs, The SAR outlined by this study provides an accompanying playbook for further
optimization if necessary.
Other non-GBCA liver-specific MRI contrast agents have been developed but are not directly
compatible with well-established and heavily relied upon clinical liver imaging protocols. The Mn-
based liver targeting agent Mn-DPDP was approved in the US and Europe but is not currently
marketed, largely due to its drawbacks in comparison to GBCAs. Mn-DPDP undergoes rapid and
significant dissociation upon intravenous injection and thus must be administered as a slow
infusion in order to avoid toxicity associated with exposure to large amounts of free Mn2+.48-50
A
protein-based agent named ProCA32 has also been developed as a liver specific agent.51 The safety
advantage of ProCA32 over commercial GBCAs is that the protein is engineered to bind Gd3+ with
ACS Paragon Plus Environment
17

Journal of Medicinal Chemistry
Page 18 of 49
1
2
3
4
5
6
7
8
9
1011-fold higher specificity than for endogenously competing metal ions like Zn2+. ProCA32
exhibits slow blood clearance and the time to delayed phase imaging is long relative to GBCAs
but T2/T1 ratiometric processing of the imaging data enables clear visualization of liver tumors
within minutes of injection. The agent Mn-EDTA-BTA was recently reported to provide strong
delayed phase liver enhancement,52 but >50% of the injected Mn is retained in mice 24h after
injection whereas no detectable Mn is recovered 24h after injection of the Mn complexes evaluated
in this work.
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
Conclusion. Mn-PyC3A-3-OBn (10) is a liver-specific MRI contrast agent that offers a potentially
direct replacement for the GBCA standard of care for liver scans. In addition to developing Mn-
PyC3A-3-OBn, we identified SAR that govern relaxivity, protein binding, pharmacokinetics, and
transient hepatocellular uptake of structurally related Mn complexes. This SAR can be applied to
guide further optimization of Mn-based liver-specific contrast agents, if required.
EXPERIMENTAL
General. All chemicals and solvents were purchased commercially and used without further
purification. All biologically evaluated compounds were isolated in >95% purity. Purity was
assessed by HPLC using Method A2 described under Analytical HPLC Methods.
NMR. NMR spectra were recorded on a 500 MHz Varian spectrometer at 25 °C. Chemical shifts
1
13
are reported in δ (ppm). For H and C NMR spectra, the residual solvent peaks were used as
13
internal reference except for C NMR recorded in D2O where dioxane was used as the internal
references, respectively.53 Relaxivity measurements were performed on a Bruker mq60 Minispec,
1.41 T and 37 °C. Longitudinal (T1) relaxation was acquired via an inversion recovery experiment
ACS Paragon Plus Environment
18

Page 19 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
using 10 inversion times of duration ranging between 0.05 ´ T1 and 5 ´ T1. Relaxivity (r1) was
determined from the slope of a plot of 1/T1 vs. [Mn] for at least 4 concentrations.
7
8
9
Analytical HPLC methods. Liquid chromatography-mass spectrometry (LC-MS) was performed
using an Agilent 1100 Series apparatus with an LC/MSD trap and Daly conversion dynode detector
with UV detection at 220, 254, and 280 nm. The reverse phase high pressure liquid
chromatography (RP-HPLC) methods used on this system are as follows: (A1) Luna C18 column
(100 ´ 2 mm 100 Å); eluent A: H2O/0.1 % formic acid, eluent B: MeCN/0.1 % formic acid;
gradient: 5 % B to 95 % B from 0-3 min, 95 % B from 3-4.5 min, 95 to 5% B from 4.5-5 min, 5%
B from 5-7 min; flow rate at 0.7 mL/min (used for characterization of organic molecules). (A2)
Phenomenex C4 column (150 ´ 4.6 mm 100 Å); eluent C: 10 mM ammonium acetate in water,
eluent D: 90 % MeCN and 10 % 10 mM ammonium acetate in water; gradient: 5 % D from 0-1
min, 5 to 50 % D from 1-10 min, 50 to 95 % D from 10-11 min, 95% D from 11-12 min, 95 to 5%
D from 12-13 min, 5% D from 13-15 min; flow rate at 0.7 mL/min (used for characterization of
manganese complexes).
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
Reverse-phase semi-preparative purification was performed on an Agilent 1260 Infinity
preparative HPLC system with UV detection at 220, 254, and 280 nm. The reverse phase high
pressure liquid chromatography (RP-HPLC) methods used on this system are as follows: (P1)
Phenomenex C18 column (250 x 21.8 cm); eluent A: H2O/0.1 % TFA, eluent B: MeCN/0.1 %
TFA; gradient: 5% B from 0-4 min, 5% to 95 % B from 4-20 min, 95% B from 20-24 min, 95%
to 5% B from 24-25, 5% B from 25-29 min, flow rate at 20 mL/min (used for separation of organic
molecules). (P2) Phenomenex C5 column (250 x 21.8 cm); eluent C: 10 mM ammonium acetate
in water, eluent D: 90 % MeCN and 10 % 10 mM ammonium acetate in water; gradient: 5% D
ACS Paragon Plus Environment
19

Journal of Medicinal Chemistry
Page 20 of 49
1
2
3
4
5
6
7
8
9
from 0-3 min, 5% to 95 % D from 3-23 min, 95% D from 23-26 min, 95% to 5% D from 26-28
min, 5% D from 27-31 min, flow rate at 15 mL/min (used for separation of manganese complexes).
ICP-MS. Metal concentrations were determined using an Agilent 8800-QQQ ICP-MS system. All
samples were diluted with 0.1 % Triton X-100 in 5 % nitric acid. A linear calibration curve for
each metal ranging from 0.1 ppb to 200 ppb was generated daily for the quantification.
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
Estimates of 1-octanol:water partition coefficient (est. log P). The est. log P values were
estimated using a gradient elution RP-HPLC assay that is similar to previously described
55
methods.54,
A linear calibration curve correlating logP of 6 standards (DMSO, DMF,
nitrobenzene, nitromethane, phenol, pyridine)45 to HPLC retention time (tR) under method A2
conditions was generated (Figure S2). The est log P values of complexes 1-10 were estimated from
their corresponding tR using this calibration curve.
Estimates of plasma protein binding. Measurements were performed on 50 µM solutions of
complexes 1-10 in human blood plasma (MGH blood bank) or 4.5% wt/vol human serum albumin.
150 µL of each solution was placed within a Millipore Ultra Free MC 3 kDa cutoff filtration vessel
and ~10 µL of the solution was forced through the filter by centrifugation. Mn content in each
unfiltered solution and filtrate were quantified by ICP-MS. The percentage of Mn complex bound
to protein was estimated from the difference in Mn concentrations between unfiltered solution and
filtrate. For consistency, all measurements used the same source or plasma or albumin and were
run side-by-side simultaneously. Our prior experience with this assay teaches that protein binding
can vary significantly between plasma samples acquired from different donors. In order to account
for any loss of free complex due to adherence to plastics or the filter membrane, control
experiments were performed in PBS buffer and protein binding was corrected for retention
measured under control conditions. All measurements were performed in triplicate.
ACS Paragon Plus Environment
20

Page 21 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Orthotopic mouse model of colorectal liver metastasis. All experiments were performed in
accordance with the National Institutes of Health's Guide for the Care of Use of Laboratory
Animals and approved by the Massachusetts General Hospital Institutional Animal Care and Use
Committee. Eight-to-ten-week old immunocompetent male BALB/c mice were obtained from
Charles River Labs (Wilmington, MA, USA).
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
The mc26 cell line was generously provided by the laboratory of Dr. Khalid Shah at Harvard
Medical School. Cells were plated at low density in T75 uncoated plastic flasks and cultured in
Dulbecco's Modified Eagle's Medium (DMEM) (high glucose 4.5g/L) supplemented with 10%
fetal bovine serum (FBS) and 1% Pen/Strep (100 U/ml Penicillin, 100 ug/ml Streptomycin). Cells
were grown at 370 C in a humidified, 5% CO2 incubator until they reached approximately 70%
confluence. All experiments were performed between passage 1 and 8.
Mice were anesthetized with an intraperitoneal injection of ketamine (100 mg/kg) and
xylazine (10 mg/kg). Fur was shaved in the region of the incision using an electric clipper. The
surgical field was sterilized with betadyne, and was allowed to dry for 30 seconds. Orthotopic liver
tumors were created by injecting 104 cells in 10 µL Matrigel into the left lobe of the liver. Mice
were monitored daily for disease progression by abdominal palpation, as well as for morbidity,
evidenced by decreased grooming, hunched posture, and reduced movement.
MR imaging of mice. Mice (balb/C, Charles River Laboratories, Cambridge, MA) were imaged
at 4.7T using a small bore animal scanner (Bruker Biospec) with a custom-built volume coil. The
imaging protocol at baseline consisted of multislice 2D T1-weighted fast low-angle shot (FLASH)
sequence to delineate anatomy and a 3D T1-weighted FLASH sequence. For the liver tumor mice,
a baseline 2D T2-weighted rapid acquisition with refocused echo (RARE) sequence was also
applied. After tail vein administration of 0.1 mmol/kg of Mn complex (formulated as 50 mM
ACS Paragon Plus Environment
21

Journal of Medicinal Chemistry
Page 22 of 49
1
2
3
4
5
6
7
8
9
aqueous solution), the 3D T1-weighted FLASH sequence used at baseline was repeated every 1
minute out to 6 min following contrast delivery, the 2D T1-weighted FLASH imaging was applied
at 6 min, and then the T1-weighted FLASH imaging continued in order to further monitor clearance
of the agent. 3D T1-weighted FLASH imaging parameters: coronal orientation, TR/TE/flip
angle=15 ms/2.46 ms/25°, in-plane FOV = 48x28 mm, matrix=85×70, 50 slices, slice thickness =
0.4 mm, voxel size = 0.4 x 0.4 x 0.4 mm, 1 average, and acquisition time = 0 min, 52 s. 2D T1-
weighted FLASH imaging parameters: axial orientation, TR/TE/flip angle=125 ms/2.93 ms/60°,
in-plane FOV = 33x33 mm, matrix=140×140, 9 slices, slice thickness = 1.0 mm, voxel size = 0.24
x 0.24 x 1.0 mm, 16 averages, and acquisition time = 4 min, 40s. 2D T2-weighted RARE imaging
parameters: axial orientation, TR/TE/flip angle=1286 ms/27 ms/90°, in-plane FOV = 33x33 mm,
matrix=192×192, 9 slices, slice thickness = 1.0 mm, voxel size = 0.17 x 0.17 x 1.0 mm, 4 averages,
and acquisition time = 2 min, 3 s.
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
Images were analyzed in Horos by drawing regions of interest (ROI) and measuring signal
intensity (SI) of the kidneys, liver, vena cava, and muscle. Noise was estimated from the standard
deviation (SD) of an ROI taken in the air outside the animal. Normalized SI (nSI) was calculated
by dividing SI at each time point post probe injection by the SI prior to injection. Liver-to-muscle
contrast-to-noise ratio (CNR) was calculated by dividing the absolute difference in SI between
liver and muscle by noise.
Pharmacokinetics. Blood pharmacokinetics (distribution and elimination rates, kdist and kel,
respectively) were estimated by plotting blood nSI recorded from the dynamic 3D T1-FLASH
acquisitions as a function of time and fitting the decay to a bi-exponential function. The distribution
and elimination half-lives (t1/2) were estimated by dividing ln(2) by kdist and kel, respectively
ACS Paragon Plus Environment
22

Page 23 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Ex vivo Mn quantification. The mice were euthanized 90 min and 24h after injection of the Mn
complex and tissues harvested for analysis. Tissues were dissolved in a defined weight of
concentrated nitric acid, and then an aliquot of digest of defined weight was diluted into a known
weight of ICP dilutent for ICP-MS analysis.
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
Histopathology. A cardiac terminal blood withdrawal was performed at the time of sacrifice. Liver
and tumors were extracted from mice, and tissues were formalin fixed for 48 hours prior to paraffin
embedding. Sections were cut at 5 um thickness. Representative sections were stained with
Hematoxylin & Eosin (H&E). Slides were digitally scanned for analysis using a Hamamastu
NanoZoomer-SQ Digital slide scanner.
Statistical Analysis. All statistical analysis was performed using GraphPad Prism 7. Correlations
between corresponding physical properties were determined by calculating the Pearson product-
moment correlation coefficient followed by a two-tailed t-test. Comparisons between dynamic
MRI data and 2D liver imaging data recorded following injection of complexes 4, 6, and 10 were
made by one-way multiple comparisons ANOVA followed by a Tukey’s post-hoc test.
Comparisons between basal Mn levels and Mn levels recorded after injection of complexes 4, 6,
and 10 were made by one-way multiple comparisons ANOVA followed by a Tukey’s post-hoc
test. Comparisons between liver vs. tumor CNR following prior to and after injection of complex
10 were made by a paired t-test. For all correlations and comparisons P < 0.05 was considered
significant.
Synthesis.
General method for compounds 11g-i. To a batch of the 2-methylpyridine derivative (6.0 mmol)
in THF (27 mL) was added KOH (7.0 mmol) and tetra-N-butylammonium bromide (TBAB 2.0
mmol). The mixture was stirred at room temperature for 10 min followed by addition of benzyl
ACS Paragon Plus Environment
23

Journal of Medicinal Chemistry
Page 24 of 49
1
2
3
4
5
6
7
8
9
bromide (7.2 mmol). The reaction mixture was stirred at room temperature for 48 hours. The
solvents were removed under reduced pressure, and the resulting residue was suspended in water
and extracted with DCM three times. The combined organics were washed with brine, dried over
MgSO4, filtered, and dried under reduced pressure. The crude material was adsorbed onto a plug
of silica gel and purified by chromatography, eluting with eluting with 20% EtOAc in hexanes to
give the benzyl protected pyridine derivatives 11g-i in 92-97% yield as colorless to light yellow
oils.
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
Benzyl 6-methylnicotinate (11g) was isolated in 92% yield as a light yellow oil. 1H NMR (500
MHz, CDCl3) δ 9.15 (d, J = 2.1 Hz, 1H), 8.20 (dd, J = 8.1, 2.2 Hz, 1H), 7.45 (dd, J = 4.4, 3.4 Hz,
2H), 7.41 (dd, J = 11.6, 4.6 Hz, 2H), 7.38 – 7.33 (m, 1H), 7.23 (d, J = 8.1 Hz, 1H), 5.39 (s, 2H),
2.63 (s, 3H). 13C NMR (126 MHz, CDCl3) δ 165.3, 163.3, 150.6, 137.4, 135.7, 128.6, 128.4, 128.2,
123.3, 122.9, 66.9, 24.8. LC-MS (method A1): tR = 3.57 min, m/z = 228.1[M + H]+.
5-(benzyloxy)-2-methylpyridine (11h) was isolated in 97% yield as a colorless oil. 1H NMR (500
MHz, CDCl3) δ 8.27 (s, 1H), 7.54 – 7.35 (m, 4H), 7.33 (m, 1H), 7.16 (d, J = 8.5 Hz, 1H), 7.05 (d,
13
J = 8.5 Hz, 1H), 5.07 (s, 2H), 2.49 (s, 3H). C NMR (126 MHz, CDCl3) δ 150.3, 148.1, 134.5,
133.8, 126.1, 125.6, 124.9, 120.8, 119.9, 67.9, 20.8. LC-MS (method A1): tR = 2.86 min, m/z =
200.1 [M + H]+.
3-(Benzyloxy)-2-methylpyridine (11i) was isolated in 95% yield as a colorless oil.1H NMR (500
MHz, CDCl3) δ 8.09 (dd, J = 4.7, 1.2 Hz, 1H), 7.47 – 7.37 (m, 4H), 7.34 (t, J = 7.0 Hz, 1H), 7.12
(d, J = 8.2 Hz, 1H), 7.07 (dd, J = 8.2, 4.8 Hz, 1H), 5.09 (s, 2H), 2.54 (s, 3H). 13C NMR (126 MHz,
CDCl3) δ 152.9, 149.2, 140.5, 136.5, 128.7, 128.1, 127.1, 121.7, 117.9, 69.9, 19.5. LC-MS
(method A1): tR = 2.81 min, m/z = 200.1 [M + H]+.
ACS Paragon Plus Environment
24

Page 25 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
General method for compounds 12d-i. To a batch of the corresponding 2-methylpyridine
derivative 11 (5.0 mmol) in DCM (20 mL) was added m-CPBA (6.0 mmol) at room temperature.
The resulting mixture was stirred overnight. The crude material was adsorbed onto a plug of silica
gel and purified by chromatography, eluting with a gradient of 0−20% MeOH in EtOAc, to provide
pyridine N-oxide intermediates 12d-i in 95-99% yield.
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-Methoxy-2-methylpyridine 1-oxide (12d) was isolated in 97% yield as a white solid. 1H NMR
(500 MHz, CDCl3) δ 7.95 (d, J = 7.2 Hz, 1H), 6.59 (d, J = 3.4 Hz, 1H), 6.51 (dd, J = 7.2, 3.4 Hz,
1H), 3.63 (s, 3H), 2.29 (s, 3H). 13C NMR (126 MHz, CDCl3) δ 157.4, 149.6, 139.9, 111.5, 109.8,
55.8, 18.2. ESI-MS: tR = 0.50 min, m/z = 140.1[M + H]+.
3-Methoxy-2-methylpyridine 1-oxide (12e) was isolated in 99% yield as a white solid. 1H NMR
(500 MHz, CDCl3) δ 7.82 (d, J = 6.5 Hz, 1H), 6.93 (m, 1H), 6.69 (d, J = 8.6 Hz, 1H), 3.75 (s, 3H),
2.32 (s, 3H). 13C NMR (126 MHz, CDCl3) δ 155.8, 140.5, 132.3, 121.6, 107.6, 56.2, 10.4. ESI-
MS: m/z = 140.0 [M + H]+.
3-(Ethoxycarbonyl)-2-methylpyridine 1-oxide (12f) was isolated in 95% yield as a white solid.
1H NMR (500 MHz, CDCl3) δ 8.35 (d, J = 6.4 Hz, 1H), 7.66 (d, J = 8.0 Hz, 1H), 7.16 (dd, J = 7.5,
6.9 Hz, 1H), 4.35 (q, J = 7.1 Hz, 2H), 2.73 (s, 3H), 1.36 (t, J = 7.1 Hz, 3H).13C NMR (126 MHz,
CDCl3) δ 164.8, 151.1, 141.5, 130.1, 126.8, 122.2, 62.1, 14.9, 14.1. LC-MS (method A1): tR =
2.74 min, m/z = 182.0 [M + H]+.
5-((Benzyloxy)carbonyl)-2-methylpyridine 1-oxide (12g) was isolated in 95% yield as a white
solid. 1H NMR (500 MHz, CDCl3) δ 8.81 (dd, J = 1.1, 0.5 Hz, 1H), 7.73 (dd, J = 8.1, 1.6 Hz, 1H),
7.39 – 7.28 (m, 6H), 5.32 (s, 2H), 2.51 (s, 3H). 13C NMR (126 MHz, CDCl3) δ 162.9, 153.3, 140.3,
134.9, 128. 7, 128.7, 128.4, 127.5, 126.2, 125.9, 67.7), 18.1. LC-MS (method A1): tR = 3.47 min,
m/z = 244.1 [M + H]+.
ACS Paragon Plus Environment
25

Journal of Medicinal Chemistry
Page 26 of 49
1
2
3
4
5
6
7
8
9
1
5-(Benzyloxy)-2-methylpyridine 1-oxide (12h) was isolated in 98% yield as a white solid. H
NMR (500 MHz, CDCl3) δ 8.09 (d, J = 2.4 Hz, 1H), 7.43 – 7.38 (m, 4H), 7.38 – 7.30 (m, 1H),
7.11 (d, J = 8.8 Hz, 1H), 6.88 (dd, J = 8.7, 2.4 Hz, 1H), 5.05 (s, 2H), 2.45 (s, 3H). 13C NMR (126
MHz, CDCl3) δ 155.2, 141.9, 135.2, 128.8, 128.5, 128.1, 127.5, 125.8, 114.6, 70.9, 16.9. LC-MS
(method A1): tR = 3.52 min, m/z = 216.1 [M + H]+.
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-(Benzyloxy)-2-methylpyridine 1-oxide (12i) was isolated in 97% yield as a white solid. 1H
NMR (500 MHz, CDCl3) δ 7.93 (d, J = 6.5 Hz, 1H), 7.49 – 7.32 (m, 4H), 7.32 (td, J = 8.6, 4.2 Hz,
1H), 7.02 – 6.93 (m, 1H), 6.82 (d, J = 8.6 Hz, 1H), 5.08 (s, 2H), 2.48 (s, 3H). 13C NMR (126 MHz,
CDCl3) δ 154.9, 141.1, 135.5, 132.7, 128.7, 128.4, 127.2, 121.7, 109.3, 71.1, 10.8. LC-MS
(method A1): tR = 3.11 min, m/z = 216.1 [M + H]+.
General method for the synthesis of compounds 13d-i. A batch of 2-methylpyridine N-oxide
intermediate (4.0 mmol) was dissolved in acetic anhydride (7.0 mL) and the resulting mixture was
refluxed for 5-7 hours. The reaction was cooled to room temperature and 10 mL of MeOH was
then added slowly, and the mixture was stirred at room temperature overnight. The solvent was
removed under reduced pressure, and the resulting residue was suspended in water and extracted
with DCM three times. The combined organics were washed with brine, dried over MgSO4,
filtered, and dried under reduced pressure. The crude material was adsorbed onto a plug of silica
gel and purified by chromatography, eluting with 30% EtOAc in hexanes to give the pyridine-2-
ylmethyl acetate derivatives 13d-i in 82-95% yields.
(4-methoxypyridin-2-yl)methyl acetate (13d) was isolated in 94% yield as a light yellow oil. 1H
NMR (500 MHz, CDCl3) δ 8.41 (s, 1H), 6.87 (s, 1H), 6.75 (s, 1H), 5.16 (s, 2H), 3.86 (s, 4H), 2.16
(s, 3H). 13C NMR (126 MHz, CDCl3) δ 170.6, 166.4, 157.3, 150.8, 108.8, 108.1, 66.7, 55.2, 20.9.
ESI-MS: m/z = 182.1 [M + H]+.
ACS Paragon Plus Environment
26

Page 27 of 49
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(3-methoxypyridin-2-yl)methyl acetate (13e) was isolated in 95% yield as a light yellow oil. 1H
NMR (500 MHz, CDCl3) δ 8.18 (s, 1H), 7.26 – 7.20 (m, 1H), 7.17 (d, J = 8.2 Hz, 1H), 5.26 (s,
2H), 3.84 (s, 3H), 2.12 (s, 3H). 13C NMR (126 MHz, CDCl3) δ 170.7, 153.9, 144.7, 140.8, 123.9,
117.6, 62.6, 55.4, 20.9. LC-MS (method A1): tR = 2.27 min, m/z = 182.1 [M + H]+.
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
Ethyl 2-(acetoxymethyl)nicotinate (13f) was isolated in 89% yield as a light yellow oil. 1H NMR
(500 MHz, CDCl3) δ 8.59 (d, J = 5.0 Hz, 1H), 8.14 (dt, J = 7.9, 1.4 Hz, 1H), 7.22 (dd, J = 8.25,
5.0 Hz, 1H), 5.47 (s, 2H), 4.27 (qd, J = 7.1, 1.3 Hz, 2H), 2.04 (s, 3H), 1.28 (td, J = 7.1, 1.3 Hz,
13
3H). C NMR (126 MHz, CDCl3) δ 170.5, 165.5, 156.1, 151.8, 138.5, 125.2, 122.5, 65.6, 61.5,
20.7, 14.1. LC-MS (method A1): tR = 3.47 min, m/z = 224.0 [M + H]+.
Benzyl 6-(acetoxymethyl)nicotinate (13g) was isolated in 82% yield as a white solid. 1H NMR
(500 MHz, CDCl3) δ 9.27 – 9.20 (m, 1H), 8.34 (dd, J = 8.2, 2.1 Hz, 1H), 7.42 (ddd, J = 20.7, 11.9,
7.3 Hz, 6H), 5.41 (s, 2H), 5.29 (s, 2H), 2.20 (s, 3H). 13C NMR (126 MHz, CDCcl3) δ 170.4, 164.9,
160.3, 150.8, 138.0, 135.5, 128.7, 128.5, 128.3, 125.2, 120.9, 77.27 (s), 77.02 (s), 76.76 (s), 67.1,
66.3, 20.8. LC-MS (method A1): tR = 4.32 min, m/z = 286.0 [M + H]+.
(5-(benzyloxy)pyridin-2-yl)methyl acetate (13h) was isolated in 95% yield as a light yellow oil.
1H NMR (500 MHz, CDCl3) δ 8.22 (s, 1H), 7.32 – 7.21 (m, 4H), 7.21 – 7.15 (m, 1H), 7.11 (dd, J
= 18.9, 8.5 Hz, 2H), 5.00 (s, 2H), 4.95 (s, 2H), 1.97 (s, 3H). 13C NMR (126 MHz, CDCl3) δ 170.7,
154.4, 147.8, 138.1, 135.9, 128.7, 128.3, 127.5, 123.0, 121.9, 70.4, 66.7, 20.9. LC-MS (method
A1): tR = 3.78 min, m/z = 258.0 [M + H]+.
(3-(benzyloxy)pyridin-2-yl)methyl acetate (13i) was isolated in 92% yield as a light yellow oil.
1H NMR (500 MHz, CDCl3) δ 8.18 (dd, J = 4.6, 1.3 Hz, 1H), 7.42 – 7.32 (m, 4H), 7.33 – 7.27 (m,
1H), 7.20 (dd, J = 8.3, 1.2 Hz, 1H), 7.16 (dd, J = 8.3, 4.6 Hz, 1H), 5.32 (s, 2H), 5.08 (s, 2H), 2.09
ACS Paragon Plus Environment
27

Journal of Medicinal Chemistry
Page 28 of 49
1
2
3
4
5
6
7
8
13
(s, 3H). C NMR (126 MHz, CDCl3) δ 170.9, 153.0, 145.1, 141.2, 136.0, 128.7, 128.2, 127.1,
123.9, 119.1, 70.1, 62.8, 20.9. LC-MS (method A1): tR = 3.90 min, m/z = 258.1 [M + H]+.
General method for compounds 14d-i .
9
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
Saponification procedure. A batch of pyridine-2-ylmethyl acetate derivative 13d,e,h-i (3.0
mmol) was dissolved in THF (10 mL) followed by addition of 2 mL of KOH solution (2.0 M). The
reaction mixture was stirred at room temperature for between 2-7 hours. The solvents were
removed under reduced pressure, and the resulting residue was suspended in water and extracted
with EtOAc three times. The combined organics were washed with brine, dried over MgSO4,
filtered, and dried under reduced pressure. The crude material was adsorbed onto a plug of silica
gel and purified by chromatography, eluting with eluting with 50% EtOAc in hexanes to give the
pyridin-2-ylmethanol derivatives 14d,e,h-i in 97-99% yield.
Acetyl chloride deprotection. A batch of Pyridine-2-ylmethyl acetate derivative 13f,g (3.0 mmol)
was dissolved in MeOH (10 mL) followed by addition of 1.2 equivalent of acetyl chloride. The
the mixture was stirred at room temperature overnight. The solvent was removed under reduced
pressure, and the resulting residue was suspended in water and extracted with EtOAc three times.
The combined organics were washed with brine, dried over MgSO4, filtered, and dried under
reduced pressure. The crude material was adsorbed onto a plug of silica gel and purified by
chromatography, eluting with eluting with 50% EtOAc in hexanes to give the pyridin-2-
ylmethanol derivatives 14f,g in 90-92% yield.
(4-methoxypyridin-2-yl)methanol (14d) was isolated in 97% yield as a white solid using the
saponification procedure. 1H NMR (500 MHz, CDCl3) δ 8.32 (d, J = 3.3 Hz, 1H), 8.26 – 8.17 (m,
1H), 6.85 (s, 1H), 6.73 (d, J = 2.4 Hz, 1H), 4.71 (s, 2H), 3.85 (d, J = 0.9 Hz, 3H). 13C NMR (126
MHz, CD3OD) δ 166.6, 161.4, 149.3, 109.2, 106.2, 64.2, 55.3. ESI-MS: m/z = 140.1 [M + H]+.
ACS Paragon Plus Environment
28

Page 29 of 49
Journal of Medicinal Chemistry
1
2
3
4
(3-methoxypyridin-2-yl)methanol (14e) was isolated in 99% yield as a white solid using the
5
6
1
saponification procedure. H NMR (500 MHz, CDCl3) δ 8.10 (d, J = 4.7 Hz, 1H), 7.16 (dd, J =
7
8
9
8.2, 4.8 Hz, 1H), 7.09 (d, J = 8.2 Hz, 1H), 4.70 (s, 2H), 3.81 (s, 3H). 13C NMR (126 MHz, CDCl3)
δ 152.3, 148.4, 139.3, 122.6, 116.5, 59.9, 55.2. ESI-MS: m/z = 140.1 [M + H]+.
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
Ethyl 2-(hydroxymethyl)nicotinate (14f) was isolated in 82% yield as a light yellow oil
1
following treatment with acetyl chloride. H NMR (500 MHz, CDCl3) δ 8.70 (dd, J = 4.8, 1.0 Hz,
1H), 8.33 (dd, J = 7.8, 1.4 Hz, 1H), 7.33 (dd, J = 7.5, 4.9 Hz, 1H), 5.05 (s, 2H), 4.78 (s, 1H), 4.37
(q, J = 7.1, 2H), 1.39 (t, J = 7.1, 3H). 13C NMR (126 MHz, CDCl3) δ 165.2, 160.1, 150.9, 139.1,
123.5, 121.9, 63.4, 61.5, 14.2. LC-MS (method A1): tR = 0.90 min, m/z = 182.1 [M + H]+.
Benzyl 6-(hydroxymethyl)nicotinate (14g) was isolated in 92% yield as a white solid following
1
treatment with acetyl chloride. H NMR (500 MHz, CDCl3) δ 9.20 (s, 1H), 8.32 (d, J = 8.2 Hz,
1H), 7.46 (d, J = 7.7 Hz, 2H), 7.43 – 7.31 (m, 4H), 5.40 (s, 2H), 4.84 (s, 2H). 13C NMR (126 MHz,
CDCl3) δ 164.9, 163.7, 149.9, 137.9, 135.5, 128.7, 128.5, 128.3, 124.9, 120.1, 67.1, 64.2. LC-MS
(method A1): RT = 3.44 min, m/z = 244.0 [M + H]+.
(5-(benzyloxy)pyridin-2-yl)methanol (14h) was isolated in 97% yield as a white solid following
the saponification procedure. 1H NMR (500 MHz, CDCl3) δ 8.33 (d, J = 2.7 Hz, 1H), 7.49-7.39
(m, 4H), 7.36 (t, J = 6.9 Hz, 1H), 7.32 – 7.26 (m, 1H), 7.22 (d, J = 8.6 Hz, 1H), 5.13 (s, 2H), 4.72
13
(s, 2H). C NMR (126 MHz, CDCl3) δ 154.1, 151.6, 136.4, 136.0, 128.7, 128.4, 127.5, 123.0,
121.2, 70.6, 63.9. LC-MS (method A1): tR = 3.26 min, m/z = 216.1 [M + H]+.
(3-(benzyloxy)pyridin-2-yl)methanol (14i) was isolated in 98% yield as a white solid following
the saponification procedure. 1H NMR (500 MHz, CDCl3) δ 8.70 (dd, J = 4.8, 1.6 Hz, 1H), 8.36
(dd, J = 7.9, 1.6 Hz, 1H), 7.47 – 7.41 (m, 2H), 7.41 – 7.37 (m, 2H), 7.37 – 7.34 (m, 1H), 7.32 (dd,
J = 7.8, 4.9 Hz, 1H), 5.36 (s, 2H), 5.08 (s, 2H). 13C NMR (126 MHz, CDCl3) δ 164.9, 160.3, 151.1,
ACS Paragon Plus Environment
29