Incorporation of a Biguanide Scaffold Enhances Drug Uptake by Organic Cation Transporters 1 and 2

Article abstract of DOI:10.1021/acs.molpharmaceut.7b00285

Membrane transporters play a significant role in the transport of many endogenous and exogenous compounds. The knowledge of transporter substrate requirements has allowed further development of drugs that utilize them to ensure tissue permeation. In this

Full text of DOI:10.1021/acs.molpharmaceut.7b00285

Subscriber access provided by AUSTRALIAN NATIONAL UNIV
Article
Incorporation of a Biguanide Scaffold Enhances
Drug Uptake by Organic Cation Transporters 1 and 2
Obinna N. Obianom, Ana L. Coutinho, Wei Yang, Hong Yang, Fengtian Xue, and Yan Shu
Mol. Pharmaceutics, Just Accepted Manuscript • DOI: 10.1021/acs.molpharmaceut.7b00285 • Publication Date (Web): 12 Jul 2017
Downloaded from http://pubs.acs.org on July 13, 2017
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.
Molecular Pharmaceutics 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
Molecular Pharmaceutics
1
2
3
4
5
6
Incorporation of a Biguanide Scaffold Enhances Drug Uptake by Organic Cation
Transporters 1 and 2
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
OBINNA N. OBIANOM#, ANA L. COUTINHO#, WEI YANG, HONG YANG, FENGTIAN
XUE*, YAN SHU*
Department of Pharmaceutical Sciences, School of Pharmacy, University of Maryland at
Baltimore, Maryland, USA
#These authors contribute equally to the work.
*Correspondence to:
Fengtian Xue, Ph.D. (fxue@rx.umaryland.edu)
OR Yan Shu, M.D., Ph.D. (yshu@rx.umaryland.edu)
Department of Pharmaceutical Sciences
School of Pharmacy
University of Maryland at Baltimore
20 N Pine Street, PH North Room 519
Baltimore, MD 21201, USA
Phone: +01 410ꢀ706ꢀ7358, Fax: +01 410ꢀ706ꢀ0886
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 2 of 49
1
2
3
4
ABSTRACT
5
6
7
8
9
Membrane transporters play a significant role in the transport of many endogenous and
exogenous compounds. The knowledge of transporter substrate requirements has allowed further
development of drugs that utilize them to ensure tissue permeation. In this study, we demonstrate
that inclusion of a biguanide functionality can potentiate uptake by the organic cation
transporters organic cation transporters 1 and 2 (OCT1 and OCT2). We synthesized 18 pairs of
structurally diverse compounds, each pair consisting of a parent amino compound and its
biguanide analog; and then assessed their cellular uptake in HEK293 cells overexpressing human
OCT1 or OCT2. Our results show that addition of the biguanide significantly improved OCT1ꢀ
and OCT2ꢀmediated transport for the majority of compounds. The biguanides also inhibited the
uptake of prototypical substrates of both transporters, 1ꢀmethylꢀ4ꢀphenylpyridinium (MPP+) and
metformin. We found that molecular weight, molecular volume, Log D (pH 7.4) and accessible
surface area were important determinants of OCT2 substrates, but none of these parameters was
a significant factor for OCT1. More so, the inhibition of MPP+ uptake correlated linearly with
that of metformin uptake for the tested biguanides in both cell lines. Taken together, we conclude
that the inclusion of the biguanide scaffold in nonꢀsubstrates of OCT1 and OCT2 increase their
propensity to become substrates and inhibitors for these transporters.
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
Keywords: Organic Cation Transporter, Metformin, Biguanide, Substrate, Liver
ACS Paragon Plus Environment

Page 3 of 49
Molecular Pharmaceutics
1
2
3
4
INTRODUCTION
5
6
7
8
9
Membrane transporters, such as those of the solute carrier superfamily including organic
cation transporters (OCT, SLC22), organic cation/carnitine transporters (OCTN, SLC22),
multidrug and toxin extrusion (MATE, SLC47), and peptide transporters (PEPT, SLC15), play a
crucial role in the absorption, distribution, elimination and toxicity of many drugs1ꢀ4. These drug
transporters are widely distributed in the body and are particularly expressed in organs of drug
disposition, such as intestine, liver, kidney and brain, where they facilitate the entry or exit of
xenobiotics and endogenous substrates5ꢀ7. Drug development has been directed to take advantage
of these transporters to improve drug bioavailability especially through the intestine8. For
instance, incorporation of the Lꢀcarnitine moiety into drug compounds valproyl, ketoprofen and
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
9
naproxen has been shown to dramatically improve their uptake by OCTN2 . Valacyclovir, a
prodrug of acyclovir, is another successful application of membrane transporter knowledge to
improve bioavailability10. Acyclovir has an oral bioavailability of 20% while valacyclovir has
incorporated in it an ester that aids its uptake via PEPT1 and PEPT2, enhancing the
bioavailability to 50%. In the same way, the tissue expression pattern of a transporter may be
utilized to achieve targeted therapy in order to improve drug exposure while decreasing offꢀ
target toxicity11ꢀ13
.
The OCTs have been implicated in the transport of endogenous and exogenous cationic
compounds, including biguanides, tetraethylammonium, catecholamines, cisplatin, and histamine
14,
15. This class of transporter proteins are present in humans in three isoforms: OCT1
(SLC22A1), OCT2 (SLC22A2) and OCT3 (SLC22A3). OCT1 is predominantly expressed on
the sinusoidal basolateral membrane of liver epithelial cells16 and OCT2 in the basolateral
membrane of renal proximal tubule epithelium cells15. OCT3 is expressed in various tissues
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 4 of 49
1
2
including brain, testis and with relatively low levels in the liver and kidney14, 17. Known to be
polyspecific in nature, this class of membrane transporters have many overlapping substrates
(reviewed in detail elsewhere 17ꢀ19). Amongst the substrates, of particular interest is the biguanide
class, which includes three previously and currently used antiꢀdiabetics – metformin, buformin,
and phenformin. This class of compounds are also transported by the MATE transporters17,
which have high expression in the liver and kidney cells to aid the extrusion of endogenous
substrates and xenobiotics20. In many cases, these MATE transporters work in conjunction with
OCTs to mediate the uptake and elimination of substrates17, 21. For instance, phenformin and
metformin exhibit enhanced bioavailability through their uptake by enterocytes and
accumulation in the liver, while being excreted by the kidney through OCTs/MATEs
transporters22, 23. Thus, the biguanide moiety may allow for efficient uptake of compounds into
the liver and kidney, which is mediated by OCTs, and subsequent MATEꢀmediated biliary and
renal excretion. We propose that inclusion of the biguanide moiety to other classes of
compounds would allow proper understanding of the structural scope of biguanides tolerated by
OCTs.
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
Although the function and clinical implication emanating from use of OCTs as
propagators of uptake into tissues have been well appreciated, there is not much information
associated with leveraging this knowledge in further modifying available compounds to attain
targeted delivery to certain tissues. This is especially important in cases where the site of action
of the compound in question is either liver or kidney24, 25. As an excellent OCT substrate, the
biguanideꢀbased drug metformin tend to undergo distribution to organs predominantly
expressing OCTs 26. For many liver diseases such as hepatitis, liver cysts, cirrhosis,
steatohepatitis and hepatocellular carcinoma, targeted delivery of drugs to liver is especially
ACS Paragon Plus Environment

Page 5 of 49
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
important to reduce drug toxicity. On the other hand, efficient elimination from the liver and
kidney is also critical to drug detoxification. If our concept could be proved, new therapeutic
agents may be further designed and developed to either target the liver via highly expressed
OCT1 or have improved elimination because of the efficient organic cation transport system in
the liver and kidney.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In this study, we sought to determine whether incorporation of the biguanide group in
structurally diverse compounds would improve their uptake by OCT1 and/or OCT2. Metformin,
eighteen parent amines (including 2ꢀphenylethylamine that is the parent compound of
phenformin) and their biguanide analogues were tested in HEK293 cells overexpressing the two
uptake transporters. The chosen pairs of compounds were within a molecular volume of 300 Å3,
having polar surface area (PSA) ranging from 81ꢀ173 Å2 and consisting largely of compounds
with lipophilic side chains due to increased number of drugs which are very lipophilic. We also
investigated if the addition of the biguanide scaffold would alter the inhibitory effects of these
compounds on OCT1 and OCT2.
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 6 of 49
1
2
3
4
EXPERIMENTAL SECTION
5
6
7
Chemical Reagents and Cell Culture
8
9
Trypsin, penicillinꢀstreptomycin solution, Dulbecco’s modified eagle medium (DMEM),
fetal bovine serum (FBS), HEPES, sodium chloride, potassium chloride, Dꢀglucose, potassium
phosphate, calcium chloride, magnesium sulfate and all chemicals used for synthesis (purity
>98%) were purchased from SigmaꢀAldrich (St. Louis, MO). PolyꢀDꢀlysine coated plates and Tꢀ
175 cell culture flasks were purchased from Corning (Lowell, MA). Acetonitrile and methanol
(HPLC grade) was obtained from Sigma Aldrich.
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
HEK293 cells obtained from ATCC were stably transfected with OCT2 or empty vector
and selected with hygromycin B as was previously described27. For OCT1 experiments, HEK293
cells were transiently transfected with a functional EGFPꢀhOCT1 plasmid as previously
described28 about 48 hours prior to uptake studies. OCT3ꢀ, MATE1ꢀ, and MATE2Kꢀ
overexpressing HEK293 cells have been previously described27, 29. The cells were grown in
DMEM supplemented with 10% (v/v) fetal bovine serum and 1% (v/v) penicillinꢀstreptomycin
solution at 37 °C in a humidified 5% CO2. Cells at 70−80% confluence were washed with
phosphate buffered saline (pH 6.8), and trypsin/EDTA (0.05%) was added for 3 minutes. The
cell suspension was collected in 5 mL of growth medium, and seeded into wells for uptake
studies.
Cells seeded in polyꢀDꢀlysine coated 24ꢀwell plates at a density of 2 × 105 cells/well were
used for uptake studies. Cell monolayers were allowed to reach total confluency prior to
experiments, about 24 hours after seeding. Before each experiment, the Krebs RingerꢀHepes
(KRH) uptake buffer (containing 25 mM HEPES, 125 mM Sodium chloride, 4.8 mM potassium
ACS Paragon Plus Environment

Page 7 of 49
Molecular Pharmaceutics
1
2
3
4
5
6
chloride, 5.6 mM Dꢀglucose, 1.2 mM potassium phosphate, 1.2 mM calcium chloride, 1.2 mM
magnesium sulfate) was prepared.
7
8
9
Immunoblot 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
Lysates of HEK293 cells were mixed 1:1 with 2x Laemmli Sample Buffer (BioꢀRad) and
resolved on a 4–15% gradient MiniꢀPROTEAN TGX™ Gel (Bioꢀrad). Western blots were then
performed on a nitrocellulose membrane, blocked with 5% milk in TBST (0.1% Tween 20 in 1X
Trisꢀbuffered saline). The membranes were incubated with antiꢀSLC22A1 or antiꢀSLC22A2
antibodies (MA5ꢀ15730 and PA5ꢀ37290, Thermo Fisher Scientific) at 1:1000 or antiꢀβꢀactin at
1:5000 (AB8227, Abcam Plc). Chemiluminescence was used to visualize the proteins using
Odyssey Fc Imaging System (LIꢀCOR Biosciences).
Real-time PCR
Total RNA was isolated from HEK293 cells stably or transiently expressing either empty
vector or vectors containing OCT transporters using the trizol reagent (Thermo Fisher Scientific)
according to the manufacturer’s instructions. The RNA was reverse transcribed to
complementary DNA, and realꢀtime polymerase chain reaction was performed using primers
designed with Oligo 5.0 and synthesized by Integrated DNA Technologies. The delta ct method
was used to analyze the gene expression and normalized to that of GAPDH 30.
Uptake Studies
The OCT1, OCT2, OCT3 and empty vector cells were washed twice with preꢀwarmed
KRH uptake buffer to remove any floating cells. Compounds diluted in uptake buffer were
replaced in the wells and incubated for the respective study conditions. For time dependent
uptake of metformin and MPP+, we incubated the cells with uptake buffer containing 50 µM
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 8 of 49
1
2
14
metformin or 5 µM MPP (containing 2.5% Cꢀmetformin or 3HꢀMPP+, respectively) for 1, 2, 5,
3
4
5
6
7
8
10 and 30 minutes. At each time point, cells were washed with cold uptake buffer three times
and lysed in 1% Triton Xꢀ100 in phosphateꢀbuffered saline for 1 hour on a shaker. The lysates
were mixed 1:10 with scintillation cocktail (Ultima Gold, PerkinElmer) and the radioactivity was
analyzed with a Packard TriꢀCarb 2500 TR liquid scintillation counter (PerkinElmer). For
screening of compounds at one concentration and oneꢀtime point, the cells were incubated with
100 µM of test compounds to allow detection across the panel of compounds studied. In each
experiment, metformin was used as reference positive control. The incubation was stopped by
removing the solution and adding 500 µL of iceꢀcold uptake buffer (without compounds). The
uptake studies with MATE1 and MATE2K overexpressing cell lines were performed as
previously described 31. Potassiumꢀbased buffer (KBB) (containing 10 mM HEPES, 140 mM
KCl, 0.4 mM KH2PO4, 1.0 mM CaCl2, 25 mM glucose, 0.8 mM MgSO4, and 30 mM NH4Cl) at
pH 7.4 was incubated with the cells for 15 minutes at 37 °C. Subsequently, we washed the cells
once with KBB uptake buffer without NH4Cl and then incubated with 100 µM of 2f or 2j in
KBB buffer without NH4Cl for 10 minutes. The cells were subsequently washed thrice and after
washing, 200 µL of lysing solution (1:1 methanol in water) containing internal standard
(compound k as described below or metformin) was added. The cell plates were shaken with the
lysing solution for at least 10 min at room temperature. The cell lysates were then centrifuged at
12,000 rpm for 15 min. Aliquots of the supernatant were transferred to new tubes and drug
concentration was quantified by LCꢀMS/MS as described below.
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
Inhibition Studies
Similar to the uptake studies, cells were seeded 24 hrs before the experiments. For
inhibition studies, the compounds were incubated at 50 µM with or without 5 µM MPP+ or 50
ACS Paragon Plus Environment

Page 9 of 49
Molecular Pharmaceutics
1
2
µM metformin containing 2.5% 3HꢀMPP+ or 14Cꢀmetformin, respectively (Moravek
Biochemicals, Brea, CA), in KRH uptake buffer. Following the incubation period, the cells were
washed with uptake buffer three times and lysed with 1% Triton Xꢀ100. Cell lysates were mixed
with scintillation cocktail at 1:10 ratio and the radioactivity was analyzed with a Packard Triꢀ
Carb 2500 TR liquid scintillation counter.
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
Liquid Chromatography–Mass Spectrometry (LC/MS/MS) Analysis
The analysis of compound in the supernatant was carried out on Agilent 1200 series
UPLC equipment (Agilent Technologies, California) equipped with a hybrid triple quadrupole
4000 QTRAP system (AB SCIEX, Massachusetts). The biguanide samples were quantified using
a Cortecs HILIC 2.7 ꢀm 2.1 x 100 mm column and that of the parent compounds were quantified
using a Cortecs C18+ 2.7ꢀm 2.1 x 75mm column. A 5 min gradient elution with 0.4 mL/min
flow rate was used for all samples. The gradient is as follows: for biguanides – 99% to 1%
mobile phase B over 1.5 min, 99% to 1% A over 2 min, and 99% B over 1.5 min; for the parent
compounds – 99% to 1% A over 1.5min, 99% to 1% B over 2 min, and 99% A over 1.5 min.
Mobile phase A was made up of 0.1% formic acid and 10 mM ammonium acetate in water while
B was made up of 0.1% formic acid in acetonitrile. The ion transitions for each of the compound
were monitored in the positive ion electrospray ionization mode and annotated (Table S1). The
peak areas obtained for each of the compound was adjusted by the peak area of the internal
standard and protein concentration, estimated using the bicinchoninic acid assay.
Computational Analysis and Compound Physicochemical Properties
All calculations were done using molecular operating environment (MOE) (Chemical
Computing Group Inc., Montreal). First, all 2D structures were converted into 3D structures in
mol2 format files and then all possible conformations were generated using the “Conformation
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 10 of 49
1
2
3
4
5
6
7
8
9
Import” application in MOE with default calculation settings. Potential energies were evaluated
using the MMFF94x force field31. Three shape descriptors were considered here to describe the
shapes of all the biguanide derivatives. This includes Accessible Surface Area (ASA), Van der
Waals Surface Area and Van der Waals Molecular Volume using Boltzmann weighted average.
LogD and molecular weight were calculated using ACD/PhysChem Suite (Advanced Chemistry
Development, Inc., Toronto, ON, Canada).
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
Data and Statistical Analysis
Statistical analysis for each sample set were carried out using the Student’s paired tꢀtest
and a pꢀvalue less than 0.05 was considered statistically significant. Data presented are an
average of at least three determinations and error bars presented in each figure denotes the
standard deviation of triplicate samples except otherwise noted.
Total uptake by OCT1ꢀ, OCT2ꢀ, or empty vectorꢀtransfected cells was calculated by
dividing the amount of compound quantified in the cell lysate by the amount of total protein
present in the well. The standard deviations were calculated using error propagation. Then the
Uptake Fold Increase was calculated as noted below:
ꢁꢂꢃꢄꢅꢆꢇꢈꢉꢊꢋꢌꢍ
Uptake Fold Increase =
ꢁꢂꢃꢄꢅꢆꢈꢎꢏꢐꢑ ꢒꢓꢔꢐꢕꢖ
ꢠꢡꢢꢣꢃꢚꢄꢃꢆ ꢡꢂꢃꢄꢅꢆ ꢇꢈꢉꢊꢋꢌꢍ ꢤꢜꢃℎ ꢜꢝℎꢜꢢꢜꢃꢙꢚ
ꢗꢆꢃꢘꢙꢚꢛꢜꢝ/ꢗꢞꢞꢟ Uptake Fold =
ꢠꢡꢢꢣꢃꢚꢄꢃꢆ ꢡꢂꢃꢄꢅꢆ ꢇꢈꢉꢊꢋꢌꢍ ꢤꢜꢃℎꢙꢡꢃ ꢜꢝℎꢜꢢꢜꢃꢙꢚ
ACS Paragon Plus Environment

Page 11 of 49
Molecular Pharmaceutics
1
2
3
4
5
6
Where ꢁꢂꢃꢄꢅꢆꢇꢈꢉꢊꢋꢌꢍ is the uptake in cells overexpressing the transporter,
ꢁꢂꢃꢄꢅꢆꢈꢎꢏꢐꢑꢊꢒꢓꢔꢐꢕꢖ is the uptake in cells transfected with the empty vector,
7
ꢠꢡꢢꢣꢃꢚꢄꢃꢆ ꢡꢂꢃꢄꢅꢆ ꢇꢈꢉꢊꢋꢌꢍ ꢤꢜꢃℎꢙꢡꢃ ꢜꢝℎꢜꢢꢜꢃꢙꢚ is the uptake of metformin or MPP+ alone and
ꢠꢡꢢꢣꢃꢚꢄꢃꢆ ꢡꢂꢃꢄꢅꢆ ꢇꢈꢉꢊꢋꢌꢍ ꢤꢜꢃℎ ꢜꢝℎꢜꢢꢜꢃꢙꢚ is the uptake of metformin or MPP+ with each
compound acting as an inhibitor. Concentration dependent uptake data were fitted to the
MichaelisꢀMenten equation and a nonꢀsaturable uptake component described by the equation
below:
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
ꢥ
ꢎꢦꢧ ∗ ꢠ
ꢥ =
+ ꢞ ∗ ꢠ
ꢨꢎ + ꢠ
In the equation above, V is the uptake rate, Vmax represents the maximum uptake rate, Km
is the MichaelisꢀMenten constant, S is the substrate concentration used, and P is the nonꢀ
saturable clearance parameter. For the correlation analysis, Pearson comparison was performed
between uptake fold increase values and physicochemical parameters (molecular weight,
molecular volume, accessible surface area, and Log D at pH 7.4) to determine their strength of
association with the fold increase. Person correlation coefficient (r) values were assessed on the
basis of low (± 0.3 to 0.5), moderate (± 0.5 to 0.7), or strong (± 0.7 to 1).
General Experimental Methods for Compound Synthesis.
Starting materials were purchased from SigmaꢀAldrich (St. Louis, MO) and used without
1
further purification. All H NMR and 13C NMR spectra were recorded on spectrometers
operating at 400 MHz and 100 MHz, respectively. Chemical shifts (δ) are expressed in ppm, and
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 12 of 49
1
2
3
4
5
6
7
8
9
coupling constants J are given in hertz. Proton chemical shifts are reported relative to residual
solvent peak (CDCl3 at 7.26 ppm, CD3OD at 3.31 ppm, and DMSOꢀd6 at 2.50 ppm). The
following abbreviations are used for signal multiplicity: s = singlet, d = doublet, t = triplet, q =
quartet, m = multiplet, br = broad, quin = quintet, sept = septet, dt = doublet of triplets. Carbon
chemical shifts are reported relative to solvent peaks (CDCl3 at 77.0 ppm and DMSOꢀd6 at 39.51
ppm). Mass spectra were recorded using electrospray as the ionization technique.
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
Synthesis and Characterization of the final products
To a solution of 1a-q (1.5 mmol) and 3 (84 mg, 1.0 mmol) in CH3CN (1.0 mL) was
o
added TMSCl (1.5 mmol, 190.4 ꢁL). The mixture was allowed to stir at 120 C overnight. After
cooling to room temperature, the resulting precipitate was filtered and washed with acetone, or
acetone and DCM (2e and 2b) or cold CH3CN (2c) (Fig. 1).
The final products consist of pairs of compounds, an amine and its biguanide analogue.
The R groups used were made of lipophilic (aꢀh, jꢀl, oꢀq, phenformin) and hydrophilic (i, m, n)
chains. As for their physicochemical properties, molecular volume were within 300 Å3; the PSA
for the lipophilic chains ranged from 81ꢀ96 Å2, and that of the hydrophilic chains ranged from
121ꢀ173 Å2. Chemical characterizations of the obtained final products are as followed.
1-n-Pentylbiguanide HCl (2a). White solid. Yield: 21%, 43.5 mg. 1H NMR (400 MHz, DMSOꢀ
d6) δ 7.56 (br, 1H), 6.97 (br, 6H), 3.10ꢀ3.04 (m, 2H), 1.44ꢀ1.41 (m, 2H), 1.26ꢀ1.25 (m, 4H), 0.87ꢀ
0.84 (m, 3H). 13C NMR (100 MHz, DMSOꢀd6) δ 159.1, 41.2, 28.9, 22.3, 14.3, 2.3. HRMS (ESIꢀ
TOF) m/z: [M + H]+ Calcd for C7H17N5: 172.1557, found: 172.1559.
1
1-(4-Phenyl-butyl)-biguanide HCl (2b). Yellow solid. Yield: 21%, 57.4 mg. H NMR (400
MHz, DMSOꢀd6) δ 7.59 (br, 1H), 7.29ꢀ7.25 (m, 2H), 7.20ꢀ7.14 (m, 3H), 6.95 (br, 5H), 3.11 (br,
ACS Paragon Plus Environment

Page 13 of 49
Molecular Pharmaceutics
1
2
2H), 2.59ꢀ2.55 (m, 2H), 1.60ꢀ1.56 (m, 2H), 1.47ꢀ1.45 (m, 2H). 13C NMR (100 MHz, DMSOꢀd6)
δ 159.1, 157.4, 142.4, 128.7, 126.1, 41.0, 35.2, 28.9, 28.6. HRMS (ESIꢀTOF) m/z: [M + H]+
Calcd for C12H19N5: 233.3190, found: 234.1722.
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
1
1-(3-Phenylpropyl)-biguanide HCl (2c). White solid. Yield: 22%, 56.7 mg. H NMR (400
MHz, DMSOꢀd6) δ 7.68 (br, 1H), 7.28ꢀ7.25 (t, J = 6.8 Hz, 2H), 7.22ꢀ7.20 (d, J = 6.8 Hz, 2H),
7.18ꢀ7.15 (t, J = 7 Hz, 1H), 6.95 (br, 6H), 3.10ꢀ3.07 (m, 2H), 2.62ꢀ2.58 (t, J = 7.4 Hz, 2H), 1.75ꢀ
13
1.71 (m, 2H). C NMR (100 MHz, DMSOꢀd6) δ 159.2, 142.0, 128.8, 128.7, 126.2, 32.8. 31.2,
2.3. HRMS (ESIꢀTOF) m/z: [M + H]+ Calcd for C11H17N5: 220.1557, found: 220.1557.
1-(p-Chlorophenethyl)-biguanide HCl (2d). White solid. Yield: 31%, 74.1 mg.1H NMR (400
MHz, DMSOꢀd6) δ 7.46 (br, 1H), 7.35ꢀ7.33 (d, J = 8 Hz, 2H), 7.28ꢀ7.26 (d, J = 8.4 Hz, 2H),
13
6.98ꢀ6.94 (br, 4H), 6.72 (s, 2H), 3.33ꢀ3.28 (m, 2H), 2.77ꢀ2.73 (t, J = 7 Hz, 2H). C NMR (100
MHz, DMSOꢀd6) δ 163.4, 158.9, 138.5, 131.3, 131.1, 128.7, 118.9, 42.6, 34.6. HRMS (ESIꢀ
TOF) m/z: [M + H]+ Calcd for C10H14ClN5: 240.1010, found 240.1023.
1
2-(2,4-Dichlorophenyl)ethyl-biguanide (2e). White solid. Yield: 1%, 4.0 mg. H NMR (400
MHz, DMSOꢀd6) δ 7.59 (s, 1H), 7.43ꢀ7.37 (m, 2H), 6.91 (br, 6H), 3.31ꢀ3.29 (m, 2H), 2.89ꢀ2.86
13
(t, J = 7 Hz , 2H). C NMR (100 MHz, DMSOꢀd6) δ 158.9, 136.0, 134.5, 132.9, 132.2, 129.0,
127.8, 40.7, 32.7. HRMS (ESIꢀTOF) m/z: [M + H]+ Calcd for C10H13Cl2N5: 274.0621, found:
274.0624.
1
2,2-Diphenylethyl-biguanide (2f). White solid. Yield: 64%, 180.3 mg. H NMR (400 MHz,
DMSOꢀd6) δ 7.30ꢀ7.29 (d, J = 3.6 Hz, 8H), 7.21ꢀ7.17 (m, 2H), 7.09ꢀ6.71 (br, 6H), 4.25ꢀ4.22 (t, J
13
= 7.6 Hz, 1H), 3.79ꢀ3.75 (t, J = 6.4 Hz, 2H). C NMR (100 MHz, DMSOꢀd6) δ 142.9, 129.0,
128.3, 126.9, 50.4, 45.6. HRMS (ESIꢀTOF) m/z: [M + H]+ Calcd for C16H19N5: 282.1713, found:
282.1718.
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 14 of 49
1
2
1
3
2-(4-Biphenyl)ethyl-biguanide (2g). White solid. Yield: 15%, 40.9 mg. H NMR (400 MHz,
DMSOꢀd6) δ 7.64ꢀ7.62 (d, J = 8 Hz, 2H), 7.60ꢀ7.58 (d, J = 8 Hz, 2H), 7.46ꢀ7.43 (t, J = 7.6 Hz,
2H), 7.35ꢀ7.33 (m, 3H), 6.95ꢀ6.91 (br, 6H), 3.38ꢀ3.36 (m, 2H), 2.83ꢀ2.79 (t, J = 7.2 Hz, 2H). 13C
NMR (100 MHz, DMSOꢀd6) δ 159.0, 140.4, 138.8, 138.6, 129.8, 129.4, 127.7, 127.1, 127.0,
42.8, 35.0. HRMS (ESIꢀTOF) m/z: [M + H]+ Calcd for C16H19N5: 282.1713, 282.1722.
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
1
1-(2-Phenoxyethyl)-biguanide (2h). White solid. Yield: 72%, 160.1 mg. H NMR (400 MHz,
DMSOꢀd6) δ 7.92 (s, 1H), 7.8ꢀ7.6.85 (br, 6H), 7.31ꢀ7.27 (t, J = 7.8 Hz, 3H), 6.95ꢀ6.92 (d, J = 8.8
13
Hz, 2 H), 4.06ꢀ4.03 (t, J = 5.2 Hz, 2H), 3.56ꢀ3.52 (m, 2H). C NMR (100 MHz, DMSOꢀd6) δ
158.5, 157.8, 130.0, 121.4, 115.0, 66.3, 40.8. HRMS (ESIꢀTOF) m/z: [M + H]+ Calcd for
C10H15N5O: 222.1349, found: 180.1250.
1
1-(p-methoxybenzyl)-biguanide HCl (2i). White solid. Yield: 16%, 40.3 mg. H NMR (400
MHz, DMSOꢀd6) δ 7.73 (br, 1H), 7.24ꢀ7.22 (d, J = 8.8 Hz, 2H), 6.97 (br, 6H), 6.90ꢀ6.88 (d, J = 8
13
Hz, 2H), 4.26ꢀ4.25 (d, J = 5.2 Hz, 2H), 3.73 (s, 3H). C NMR (100 MHz, DMSOꢀd6) δ 160.5,
158.8, 131.0, 129.1, 114.2, 55.5, 44.0. HRMS (ESIꢀTOF) m/z: [M + H]+ Calcd for C10H15N5O:
222.1349, found: 222.1358.
2,3-Dihydro-1H-inden-2-yl-biguanide HCl (2j). White solid. Yield: 9%, 23.2 mg.1H NMR
(400 MHz, DMSOꢀd6) δ 7.64 (br, 1H), 7.22ꢀ7.21 (m, 2H), 7.16ꢀ7.13 (m, 2H), 7.0 (br, 6H), 4.38
(br, 1H), 3.22ꢀ3.16, (dd, J1 = 7.2 Hz, J2 = 15.6 Hz, 2H), 2.84ꢀ2.79 (dd, J1 = 4.6 Hz, J2 = 15.8 Hz,
2H). 13C NMR (100 MHz, DMSOꢀd6) δ 141.3, 127.0, 124.9, 52.45. HRMS (ESIꢀTOF) m/z: [M +
H]+ Calcd for C11H15N5: 218.1400, found: 218.1404.
1
1-(p-Methyl)-biguanide HCl (2k). White solid. Yield: 5%, 12.3 mg. H NMR (400 MHz,
DMSOꢀd6) δ 9.60 (s, 1H), 7.22ꢀ7.21 (m, 6H), 7.10ꢀ7.08 (d, J = 7.6 Hz, 2H), 7.02 (br, 2H), 2.24
ACS Paragon Plus Environment

Page 15 of 49
Molecular Pharmaceutics
1
2
(s, 3H). 13C NMR (100 MHz, DMSOꢀd6) δ 166.1, 160.7, 141.1, 137.7, 134.3, 126.4, 25.6. HRMS
(ESIꢀTOF) m/z: [M + H]+ Calcd for C9H13N5: 192.1244, found: 192.1252.
3
4
5
6
7
8
9
1
1-(p-Chlorophenyl)-biguanide HCl (2l). White solid. Yield: 30%, 74.2 mg. H NMR (400
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
13
MHz, DMSOꢀd6) δ 9.93 (s, 1H), 7.42ꢀ7.39 (m, 6H), 7.35ꢀ7.33 (m, 2H), 7.09 (br, 2H). C NMR
(100 MHz, DMSOꢀd6) δ 166.6, 160.0, 143.0, 133.7, 132.0, 127.3. HRMS (ESIꢀTOF) m/z: [M +
H]+ Calcd for C8H10ClN5: 212.0697, found: 212.0707.
1
p-(3-Aminoguanidino)-benzoic acid (2m). Brown solid. Yield: 47%, 104.4 mg. H NMR (400
13
MHz, DMSOꢀd6) δ 10.22 (br, 1H), 7.86ꢀ7.84 (d, J = 8 Hz, 2H), 7.51 (br, 6H), 7.20 (br, 2H). C
NMR (100 MHz, DMSOꢀd6) δ 167.8, 161.9, 1540.8, 130.6, 119.6. HRMS (ESIꢀTOF) m/z: [M +
H]+ Calcd for C9H11N5O2: 221.0913, found: 222.0991.
1
4H-1-Benzopyran-4-one-biguanide HCl (2n). Brown solid. Yield: 38%, 111.7 mg. H NMR
(400 MHz, DMSOꢀd6) δ 7.89ꢀ7.87 (d, J = 8.8 Hz, 1H), 7.84ꢀ7.83 (d, J = 1.6 Hz, 1H), 7.62 (br,
13
4H), 7.31ꢀ7.29 (m, 1H), 7.21 (br, 2H), 2.35 (s, 3H). C NMR (100 MHz, DMSOꢀd6) δ 176.5,
166.8, 162.2, 157.0, 154.3, 144.6, 125.8, 118.1, 177.4, 110.2, 106.8, 20.4. HRMS (ESIꢀTOF)
m/z: [M + H]+ Calcd for C7H17N5: 160.1142, found: 260.1149.
1
1-(m-Phenoxyphenyl)-biguanide HCl (2o). White solid. Yield 27%, 80.9 mg. H NMR (400
13
MHz, DMSOꢀd6) δ 9.89 (s, 1H), 7.39ꢀ7.34 (m, 8H), 7.12ꢀ7.08 (m, 3H), 6.98ꢀ6.95 (m, 4H). C
NMR (100 MHz, DMSOꢀd6) δ 161.6, 157.7, 155.9, 152.5, 134.8, 130.4, 123.4, 123.3, 119.8,
118.3. HRMS (ESIꢀTOF) m/z: [M + H]+ Calcd for C14H15N5O: 270.1349, found: 270.1351.
1
1-[p-(p-Phenoxy)phenyl]-biguanide HCl (2p). White solid. Yield: 56%, 189.8 mg. H NMR
(400 MHz, DMSOꢀd6) δ 9.85 (s, 1H), 7.42ꢀ7.38 (m, 4H), 7.33 (br, 4H), 7.09 (br, 2H), 7.02ꢀ6.99
13
(d, J = 8.8 Hz, 2H), 6.98ꢀ6.96 (d, J = 8.4 Hz, 2H). C NMR (100 MHz, DMSOꢀd6) δ 166.4,
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 16 of 49
1
2
161.5, 160.6, 156.8, 140.0, 135.0, 131.8, 128.1, 124.8, 124.6. HRMS (ESIꢀTOF) m/z: [M + H]+
3
4
5
6
Calcd for C14H14ClN5O: 304.0960, found: 304.0956.
7
8
9
1
3-Heptylbenzyl-biguanide HCl (2q). White solid. Yield: 71%, 221.8 mg. H NMR (400 MHz,
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
DMSOꢀd6) δ 9.39 (s, 1H), 7.30ꢀ7.28 (d, J = 7.6 Hz, 2H), 7.24 (s, 4H), 7.20ꢀ7.18 (d, J = 8 Hz,
2H), 7.03 (s, 2H), 1.61ꢀ1.58 (m, 2H), 1.34ꢀ1.32 (m, 8H), 0.94ꢀ0.91 (t, J = 7.2 Hz, 3H); 13C NMR
(100 MHz, DMSOꢀd6) δ 161.4, 155.9, 137.9, 136.6, 125.8, 121.6, 34.9, 31.7, 31.5, 29.0, 29.0,
22.5, 14.4. HRMS (ESIꢀTOF) m/z: [M + H]+ Calcd for C15H25N5: 276.2183, found: 276.2182..
ACS Paragon Plus Environment

Page 17 of 49
Molecular Pharmaceutics
1
2
3
4
RESULTS
5
6
7
Validation of the overexpression of OCT1 and OCT2 in HEK293 cells
8
9
Immunoblotting and realꢀtime PCR were used to validate the overexpression of human
isoforms of OCT1 (HEKꢀOCT1) and OCT2 (HEKꢀOCT2) transgenes in HEK293 (Fig. S1). The
uptake of prototypical substrates, MPP+ and metformin, was used to assess the function of these
transporters as described previously 28. The uptake was higher in the HEKꢀOCT1 and HEKꢀ
OCT2 cells than the empty vector controls (HEKꢀP5) and was linear up to 10 minutes prior to
plateau (Fig. 2). Subsequent uptake experiments were then carried out for 10 minutes, all in
room temperature.
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
Incorporation of the biguanide moiety leads to improved uptake by OCT1 and OCT2
To ascertain if addition of the biguanide moiety led to improvement in transport by OCT1
and OCT2, the uptake of amine parent and their corresponding biguanides was assessed in the
transporter overexpressing cells and compared to that of uptake in empty vector overexpressing
cells to attain an uptake fold increase ratio (as indicated in the methods). In this study, a fold
uptake increase of 1.5 and greater over HEKꢀP5 (control) was used as the cutoff for defining
OCT substrates for both parents and biguanides (Fig. 3). Most of the biguanides, sixteen
(phenformin, 2aꢀ2p) and fourteen (phenformin, 2a-l, 2o-p) out of eighteen biguanides tested
were substrates for OCT1 and OCT2, respectively. For the parent compounds, only 1q, and three
others (1m, 1n, and 1q) surpassed the substrate threshold in OCT1 and OCT2 cells, respectively.
2m was the only compound to have a decrease in OCT2ꢀmediated transport when compared to
its amino parent. In comparison to the parent compound, addition of the biguanides moiety lead
to increase in all, but one compound (q) in OCT1 (Fig. 3a). For OCT2 cells, only four
compounds (m,n,p,q) had no improvement upon addition of the biguanides moiety (Fig. 3b).
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 18 of 49
1
2
3
4
5
6
7
8
9
While there is little or no correlation between OCT1 and OCT2 uptake observed for the
compounds assayed in this uptake experiment (Fig. S2), the results suggest that the addition of a
biguanide moiety improved transporterꢀmediated uptake of the majority of the compounds over
the parent compounds.
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
Inhibitory effects of parent and biguanide analogs on the activities of OCT1 and OCT2
Another aspect of our experiment was to determine whether the addition of biguanide
moiety would enhance the inhibitory effect of the compounds on prototypical OCT substrates.
Parent amines were assayed against MPP+ uptake while the biguanides were tested against both
MPP+ and metformin uptake. Uptake assays were firstly performed with 5 µM MPP+ in the
presence or absence of 50 µM of the biguanides or their parent compounds (Fig. 4a and b). Half
of the compounds (phenformin, 2e-f, 2h, 2k-l, and 2n-q), upon addition of the biguanide moiety,
had increased inhibition of OCT1ꢀmediated uptake of MPP+. However, the inhibition of OCT2ꢀ
mediated uptake of MPP+ was not enhanced for most of the compounds upon addition of the
biguanide moiety. In many cases (phenformin, 2bꢀe, 2g, 2iꢀj), this addition instead resulted in
less inhibition of OCT2 activity at the concentration tested. Taken together, there seem to be
increased inhibition of MPP+ by swapping amines for biguanide moieties in some cases,
however, there were more than a few instances where biguanides had lower or similar inhibition
than the amine parents, opposing a conclusion that biguanide addition cause enhanced inhibition
of either OCT1 or OCT2.
We further performed similar inhibition studies with metformin, another prototypical
substrate but also a biguanide (Fig. 4c and d). All the biguanides, except 2m, inhibited both
OCT1 and OCT2 mediated metformin uptake. 2m only inhibited OCT1 mediated metformin
uptake. In HEKꢀOCT2 cells, inhibition of metformin uptake was more potent for most of the
ACS Paragon Plus Environment

Page 19 of 49
Molecular Pharmaceutics
1
2
compounds than for their inhibition of MPP+ uptake. In addition, we found that six biguanides
(2c, 2g-h, 2j, and 2p-q) inhibited metformin transport by 50% or more in HEKꢀOCT1 cells and
thirteen biguanides (2a-f, 2h, 2j, 2n-q and phenformin) in HEKꢀOCT2 cells at the concentration
of 50 µM of metformin. Consistent with the uptake results (Fig. 3), these inhibition data overall
confirmed that addition of the biguanide scaffold led to an interaction between the compound
and OCT transporters.
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
Uptake and inhibition kinetics of 2f and 2j
To determine if incorporation of the biguanide scaffold enhanced uptake by the
transporters, we performed a concentrationꢀdependent studies with two of the biguanides with
relatively good uptake fold increase ratios, 2f and 2j (Fig. 5a). The transport profile of 0.1ꢀ1000
ꢁM of each of the compounds showed a concentrationꢀdependent increase in uptake, which was
fitted to the MichaelisꢀMenten equation with a nonꢀsaturable component (described in methods).
The uptake in OCTꢀoverexpressed cells was significantly higher than that of the mock cells. For
OCT2, 2f exhibited a moderately higher transporter affinity (Km = 12.4 ꢁM; Vmax = 1.8
pmoles/min/mg protein; P = 0.6 fl/min/mg protein) than 2j (Km = 31.7 ꢁM; Vmax = 1.1
pmoles/min/mg protein; P = 0.1 fl/min/mg protein) (Table 1), albeit there seemed to be a
significant nonꢀsaturable uptake process occurring for the former compound (Fig. 5a). The same
trend was seen for OCT1ꢀmediated uptake kinetics with the two compounds. For OCT1,
however, the two compounds had a similar transporter affinity (2f: Km = 14.2 ꢁM, Vmax = 2.0
pmoles/min/mg protein, P = 1.7 fl/min/mg protein; 2j: Km = 16.9 ꢁM, Vmax = 0.3 pmoles/min/mg
protein, P = 0.6 fl/min/mg protein). The trend in the aboveꢀdescribed uptake ratios of 2f
compared with 2j (Fig. 3) was somewhat dissimilar to their calculated kinetic parameters here,
which suggests that the uptake ratios for the various compounds at a single concentration may
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 20 of 49
1
2
3
4
5
6
not necessarily be indicative of their affinity to the transporters due to possible influence of
passive transport and other unknown mechanisms that may play a role in overall uptake.
7
8
9
Inhibition kinetics was further carried out for both 2f and 2j using similar range of
concentrations as in the above uptake kinetic studies in the presence of metformin (50 ꢁM), for
both OCT1 and OCT2 (Fig. 5b). The inhibition data were fitted to the Log [inhibitor] vs
metformin uptake to determine IC50. The obtained values noted in Table 1 show that 2j (IC50
=53.4 ꢁM) presented a relatively similar inhibitory potency towards OCT2ꢀmediated metformin
uptake as 2f (IC50 =52.7 ꢁM). However, 2j (IC50 =22.1 ꢁM) was a more potent inhibitor than 2f
(IC50 = 85.5 ꢁM) in OCT1 cells. In consideration of the uptake kinetics above (Fig. 5a), the data
indicate that IC50 may not be an accurate predictor of affinity to the transporter for a substrate
compound, attesting to the complexity of interaction that may exist between compounds and the
membrane transporters either as their substrates or as inhibitors.
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
Effect of incorporation of the biguanide moiety on uptake of 2f and 2j in OCT3, MATE2K and
MATE1 overexpressing cells
Because the OCT and MATE transporters have overlap in substrate selectivity 17, we
made another attempt to assess the uptake of the selected compounds from our study, 2f and 2j,
in HEK293 cells stably overexpressing OCT3, MATE2K, and MATE1 transporters. We have
previously validated the functions of these cells in our recent articles27, 29. Similar to the design
of the uptake studies for OCT1 and OCT2, we performed the experiment with a single
concentration of 100 ꢁM for each of the compounds. There was significant increase in uptake
fold ratio for 2j, but not 2f, in OCT3 cells (Fig. S3a). For both compounds, we found that
addition of the biguanide moiety significantly improved their uptake more than the parent
compounds for both MATE2K and MATE1 (Fig. S3b and S3c). The data suggest that addition of
ACS Paragon Plus Environment

Page 21 of 49
Molecular Pharmaceutics
1
2
3
4
5
6
the biguanide moiety may also improve uptake by OCT3, MATE1 and MATE2k; albeit, further
comprehensive experiments need to be done with other biguanides.
7
8
9
Physicochemical properties drive the uptake kinetics for biguanides
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
Physicochemical parameters are important descriptors of certain processes in drug
disposition. A number of these parameters have been correlated with the uptake properties and
inhibitory capabilities of many OCT substrates and inhibitors21, 32. Here, we assessed the
correlation between uptake fold increase and molecular weight, molecular volume, polar surface
area, accessible surface area, and calculated LogD of the tested compounds. We found no
significant correlation between the uptakes by OCT1 for the biguanides with their
physicochemical parameters (Fig.6). The lack of a rich dataset may explain the absence of
significant correlation, contrary to previous reports generally for OCT1 substrates33, 34. The
uptake mediated by OCT2 was moderately correlated with molecular volume, which has been
34
characterized as a key descriptor of uptake by OCT2 , as increase in this parameter appears to
have an inverse effect on OCT2ꢀmediated uptake. This trend was similarly observed for
molecular weight, LogD and accessible surface area in our study.
In the inhibition studies, molecular volume, molecular weight and surface accessible area
of the compounds examined moderately correlated with the biguanides inhibition of OCT1ꢀ
mediated uptake of metformin and MPP+ (Figure 7 and S4). This was not the case for their
inhibition of uptake mediated by OCT2, as the correlation coefficient was very low and
insignificant (p > 0.05). The data suggest that multiple mechanisms may be involved in the
inhibition of OCT2 by biguanides. In addition, Hacker and colleagues 35 have recently reported a
moderate correlation between inhibition of MPP+ uptake and that of metformin mediated by
OCT2. We find consistency in this correlation with our OCT2 inhibition data for both substrates
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 22 of 49
1
2
3
4
5
6
7
8
9
(r = 0.70; p = 0.00040), where increase in the inhibition of metformin uptake was correlated well
with the increase in the inhibition of MPP+ uptake by the biguanides (Figure S5). We also found
a moderate but significant correlation (r = 0.58; p = 0.0055) for OCT1. Therefore, both reference
substrates may be employed interchangeably in inhibition studies to screen compounds, as had
been previously concluded. Together, these results suggest that certain physicochemical
properties including molecular volume, molecular weight, LogD and accessible surface area are
moderate descriptors for the uptake of biguanides mediated by OCT2 (but not OCT1). Rather,
these intrinsic properties of the biguanides were moderate descriptors for their inhibition against
OCT1 (but not OCT2) activity.
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

Page 23 of 49
Molecular Pharmaceutics
1
2
3
4
DISCUSSION
5
6
7
8
Biguanides have been majorly studied for their beneficial effects as antiꢀdiabetics 36. The
model biguanide, metformin, which is still the first line treatment for diabetes, has been shown to
elicit its effect through its functions in the liver24. It has been shown that in order to permeate the
liver and accumulate at the site of action, OCT1 is expected to shuttle this cationic compound
from the extracellular space into hepatocytes. Absence of OCT1 in mice may abrogate the
glucoseꢀlowering effect of the compounds in the biguanides class – metformin, phenformin and
buformin37. Thus, it is likely that the accumulation of biguanides in the liver depends on OCT1
function. We reasoned that since the already known biguanides have been shown in vivo and in
vitro to be efficiently transported by OCTs, attaching the biguanide scaffold to some compounds
may cause them to become OCT substrates. Thus, by introducing the biguanide moiety in
compounds that are not OCT1 or OCT2 substrate, we demonstrate that this handle can be utilized
to improve cellular uptake mediated by OCT1 and OCT2.
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
We focused on whether incorporating the biguanide moiety into amines would improve
their uptake by organic cation transporters. These parent amine compounds were chosen based
on size and polarity so the impact of these two variable could be studied. It has been described
that uptake by OCT1 and OCT2 drastically decreases for compounds over 500 Å3 34. Therefore,
small side chains (R groups) were chosen so the molecular volume of the amine and biguanide
forms would not exceed 300 Å3. The R groups were composed of lipophilic (aꢀh, jꢀl, oꢀq,
phenformin) and hydrophilic (i, m, n) chains. The PSA of biguanides with lipophilic chains
ranged from 81ꢀ96 Å2, and the hydrophilic chains ranged from 121ꢀ173 Å2. A greater number of
lipophilic chains than hydrophilic chains were chosen due to our interest in investigating the
addition of the biguanide moiety to lipophilic drugs in order to increase their uptake by OCT.
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 24 of 49
1
2
3
4
5
6
7
8
9
With these structurally diverse amines and their biguanide analogues, we also investigated the
correlation between the biguanide physicochemical properties and their interaction with the two
OCTs.
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
Consistent with already existing biguanides reported in the literature37, we found that
incorporation of the biguanide scaffold in the parent compounds in this study led to an overall
increase in uptake by OCT1 and OCT2. In contrast, the uptake between the empty vector
transfected and the transporter overexpressed cells was no different for most of the parent
compounds in our study. In a particular case, the addition of the biguanide group did not
improved uptake for 1m/2m in OCT2 cells. For this pair of compounds, the PSA was the highest
amongst the studied compounds, which may have played a role in the lower uptake observed.
Besides, because the transport assays were carried out for all compounds using the incubation
conditions optimized according to metformin uptake, it is possible that some of the biguanides
might actually exhibit different and/or more complex kinetics. We observed that 1 pair (1/2q)
and 4 pairs (1/2m, 1/2n, 1/2p, and 1/2q) had only a meager, but significant increase in transport
by OCT1 and OCT2, respectively. While the exact reason remains unclear, this may be due to
their structural constituents, relatively higher passive uptake, or the site of addition of the
biguanide moiety that causes them to have a lower uptake ratio against the control cells. Overall,
this study strongly suggests that having a biguanide included in impermeable compounds will
lead to improved OCTꢀmediated uptake (Figure 8).
The uptake ratios for OCT2 were relatively higher than those for OCT1 but more
compounds had an increase in uptake by OCT1 than OCT2 when the biguanide was added to the
structure. It is noteworthy to mention that we used different systems to assess the uptakes by the
two transporters – transient transfection for OCT1 and stable expression for OCT2. Besides, we
ACS Paragon Plus Environment

Page 25 of 49
Molecular Pharmaceutics
1
2
3
4
5
6
7
8
9
did not quantitatively measure the relative expression of OCT2 compared to that of OCT1 in the
cells. Nonetheless, it has been reported that the biguanide antidiabetic drugs are more efficiently
transported by OCT2 as compared to OCT1 38.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In the case where transporterꢀmediated uptake is the major route of targeted tissue
delivery of compounds, deletion of the specific transporters may lead to drastic decrease in the
accumulation of their substrates in the tissues14, 39. For compounds whose target site is in the
specific tissue expressing OCTs transporters, ablation of this entry point may also result in
inaction of key substrate drugs24. This has become especially important in many disease states
that alter the expression of these drug carriers, including cancer and diabetes 40, 41. As such, the
concept established in this study allows for direct delivery of compounds to specific tissues
expressing OCT1/2 and lessening side effect in other tissues, with knowledge of the transporter
expressions in the tissues either in healthy individuals or in a disease state.
Tissue targeting is amongst the benefits that incorporating a biguanide moiety confers for
certain compounds. However, it is foreseeable that not all addition of the biguanide moiety may
be beneficial. Depending on the chemical site of addition, this may alter the effect of a drug
positively or negatively. The cytotoxicity of the compounds were not tested in the current study,
but this may vary from compound to compound. In the past, biguanides such as phenformin and
buformin were discontinued due to their effect of lactic acidosis42. Nevertheless, metformin has
been shown to be safe and well tolerated even at very high doses. As a future drug development,
compounds may be designed such that are cleaved to their parent compounds and metformin
upon entry into the specific tissues. This allows for maintenance of drug potency, allowing for
lower dosage administration and without much concern of toxicity resulting from metformin.
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 26 of 49
1
2
3
4
5
6
When given at lower doses, metformin has little or no effect so uncoupling the compound should
only maintains the effect of the parent compound.
7
8
9
Despite sequence similarities between OCT1 and OCT2 and an extensive list of
overlapping substrates, some specific drivers for being relatively specific substrates/inhibitors for
OCT1 or OCT2 have been identified 34, 43. The substrates of OCT1 usually are less than 500 Å3
and an increase in molecular weight and volume have a negative effect on transport. On the other
hand, OCT2ꢀmediated transport is affected positively by an increase in polar surface area (PSA),
lipophilicity, and number of hydrogen bonds21. OCT2 has a higher preference for hydrophilic
compounds than OCT1. In our study, we further uncover some key drivers for the uptake of
biguanides as well as their inhibitory capability for OCT transporters. Molecular weight,
molecular volume, accessible surface area and Log D (pH 7.4) presented significant correlation
with OCT2ꢀmediated uptake. For OCT1, however, we found no strong correlation with any of
these physicochemical parameters. We found the optimal size of biguanides for OCT2ꢀmediated
uptake to be around 200 and 250 Å3. Log P and Log D values were significant correlated with
OCT2ꢀmeidated transport; and hydrophilic biguanides (negative Log P and Log D) were greatly
transported compared to more hydrophobic compounds. An accessible surface area (ASA) of
around 450 Å2 was also correlated to a higher OCT2ꢀmeidated transport. ASA is the area
available for solvent binding on a molecule surface. We concluded that increased ASA would
decrease the likelihood of interaction between the molecule functional groups and the amino acid
residues in the transporter. This, in turn, would decrease the likelihood of molecule being
transported as observed in Figure 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
OCT1 inhibitors tend to have high molecular weight, positive net charge and high
lipophilicity. For example, nꢀalkylammonium compounds with long chains and hydrophobicity
ACS Paragon Plus Environment

Page 27 of 49
Molecular Pharmaceutics
1
2
tend to be OCT1 inhibitors32. Inhibitors of OCT2 usually have high molecular volume. OCT1
and OCT2 substrates have a positive charged functional group, but some neutral compounds can
also be transported. Zwitterion compounds do not tend to be OCT1 or OCT2 substrates34. These
structureꢀtransporterꢀactivity relationships for OCT1 and OCT2 have been reported using
inhibition data as well as substrate uptake data34, 44, 45. Molecular weight, accessible surface area
and molecular volume were moderate descriptors for OCT1 inhibition in our study, but not
OCT2. The reason for no correlation for OCT2 is unknown and may be due to difference in
interaction of the compounds with the two transporters.
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
Consistent with other reports43, we found that some of the parent compounds and most of
the biguanides assayed in this study were inhibitors of the uptake of MPP+ mediated by OCT1
and OCT2 at the concentration tested,. However, our hypothesis that the biguanides might
exhibit as more potent inhibitors, as compared to those parent compounds with an inhibition
potency, was not consistently observed either in OCT1 or OCT2 cells. We further analyzed the
uptake kinetics and that of inhibition against a typical substrate for two (2j and 2f) of the novel
biguanide substrates of OCT1 and OCT2 synthesized in this study. The results were somewhat
contradictory with our uptake fold increase ratios estimated at the concentration of 100 µM for
the two compounds. 2j had a higher uptake fold increase in OCT2 cells compared to 2f, but the
reverse was the case for their OCT1ꢀmediated uptake as well as their Km values in the
concentrationꢀdependent studies. These findings suggest that the uptake rate obtained from single
concentration incubation may not be the best indicator for how good a substrate the compounds
are in this study. In addition, the estimation of their affinity (IC50) by inhibition kinetics for both
2f and 2j did not accurately reflect the affinity (Km) observed in the uptake kinetic studies.
Overall, while addition of the biguanide scaffold certainly conferred some inhibitory effects
ACS Paragon Plus Environment

Molecular Pharmaceutics
Page 28 of 49
1
2
3
4
5
6
along with being transported by the transporter proteins, further studies need to be carried out to
fully understand specific interactions of OCT transporters with the biguanide class.
7
8
9
Current biguanide drugs have been characterized as substrates for additional transporters
such as OCT3, MATE1, and MATE2K25, 46. Interestingly, for the two compounds 2f and 2j that
were further examined in this regard, they were either none or very weak substrates for OCT3,
suggesting a possible transporter selectivity and thus tissue targeting for new biguanide
compounds. MATE1 and MATE2k, which are expressed primarily in the liver and kidney, work
in concert with OCTs to eliminate xenobiotics. Both 2f and 2j were substrates of MATE1 and
MATE2K in this study. Therefore, while further studies are warranted, we speculate that for
other compounds and those tested here, addition of biguanide functionality may be able to
provide an efficient excretion route in vivo via the MATE expressed in the kidney and liver, as
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
has been shown for other OCT substrates and biguanides47ꢀ49
.
To conclude, we have demonstrated that addition of a biguanide functional group to
structurally diverse transporterꢀimpermeable compounds enhances their OCTꢀmediated transport
(Figure 8). Physicochemical properties especially accessible surface area, molecular weight and
volume correlate with the uptake of biguanides by OCT2, but not OCT1. Targeting may be
attained in delivery of such compounds to the tissues with significant expression of these OCT
transporters including liver and kidney.
ACS Paragon Plus Environment

Page 29 of 49
Molecular Pharmaceutics
1
2
3
4
ACKNOWLEDGEMENT
5
6
7
8
The present study was supported by the National Institute of General Medical Sciences of
the National Institutes of Health (NIH) under Award R01GM099742 and by the US Food and
Drug Administration (FDA) under Award U01FD004320. The content is solely the responsibility
of the authors and does not necessarily represent the official views of the NIH and FDA.
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
ACS Paragon Plus Environment