New Polyaminoisoprenyl Antibiotics Enhancers against Two Multidrug-Resistant Gram-Negative Bacteria from Enterobacter and Salmonella Species

Article abstract of DOI:10.1021/acs.jmedchem.0c01335

A series consisting of new polyaminoisoprenyl derivatives were prepared in moderate to good chemical yields varying from 32 to 64% according to two synthetic pathways: (1) using a titanium-reductive amination reaction affording a 50/50 mixture of cis and trans isomers and (2) a direct nucleophilic substitution leading to a stereoselective synthesis of the compounds of interest. These compounds were then successfully evaluated for their in vitro antibiotic enhancer properties against resistant Gram-negative bacteria of four antibiotics belonging to four different families. The mechanism of action against Enterobacter aerogenes of one of the most efficient of these chemosensitizing agents was precisely evaluated by using fluorescent dyes to measure outer-membrane permeability and to determine membrane depolarization. The weak cytotoxicity encountered led us to perform an in vivo experiment dealing with the treatment of mice infected with Salmonella typhimurium and affording preliminary promising results in terms of tolerance and efficiency of the polyaminoisoprenyl derivative 5r/doxycycline combination.

Full text of DOI:10.1021/acs.jmedchem.0c01335

Subscriber access provided by CORNELL UNIVERSITY LIBRARY
Article
New polyaminoisoprenyl antibiotics enhancers against two multidrug
resistant Gram-negative bacteria from Enterobacter and Salmonella species
aurelie lieutaud, cyril pieri, Jean-Michel Bolla, and Jean Michel Brunel
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.0c01335 • Publication Date (Web): 25 Aug 2020
Downloaded from pubs.acs.org on August 28, 2020
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 39
Journal of Medicinal Chemistry
1
2
3
4
New polyaminoisoprenyl antibiotics enhancers against two multidrug resistant Gram-
5
6
negative bacteria from Enterobacter and Salmonella species
7
8
9
Aurélie Lieutaud1, Cyril Pieri1, Jean Michel Bolla1*and Jean Michel Brunel1*
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 Aix Marseille Univ, INSERM, SSA, MCT, 13385 Marseille, France.
Keywords : Gram-negative bacteria, Multi-drug resistance, Efflux pump, Antibiotic enhancers,
Polyaminoisoprenyl derivatives.
Abstract
A series consisting of new polyaminoisoprenyl derivatives were prepared in moderate to good
chemical yields varying from 32 to 64% according two synthetic pathways: 1) using a titanium
reductive amination reaction affording a 50/50 mixture of cis and trans isomers 2) a direct
nucleophilic substitution leading to a stereoselective synthesis of the compounds of interest.
These compounds were then successfully evaluated for their in vitro antibiotic enhancer
properties against resistant Gram-negative bacteria of four antibiotics belonging to four
different families. The mechanism of action against Enterobacter aerogenes of one of the most
efficient of these chemosensitizing agents was precisely evaluated by using fluorescent dyes to
measure outer-membrane permeability and to determine membrane depolarization. The weak
cytotoxicity encountered led us to perform an in vivo experiment dealing with the treatment of
mice infected with Salmonella Typhimurium and affording preliminary promising results in
terms of tolerance and efficiency of the polyaminoisoprenyl derivative 5r / doxycycline
combination.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 39
1
2
3
4
Introduction
5
6
7
8
Over the last decades, antibiotic resistance has become one of the predominant general health
concerns principally due to the overuse of antibiotics, the ageing of the population mostly
subjected to infections, the increase of efficiency of intensive care units taking in charge
patients in decreased physiological state.1-3 In this context and closely associated to the failure
in the discovery of new antibiotics, there is a need to improve all the stages of infectious
diseases treatments including the management of different steps of antibiotic usage and the
improvement of the efficiency of well-known antibiotics.4-5 Amongst the human pathogenic
bacteria, the ESKAPE group (Enterococcus faecium, Staphylococcus aureus, Klebsiella
pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, Enterobacter spp) first
described by Rice et al., is the most problematic for the clinicians because of their ability to
evade and escape common therapies.6 This ESKAPE group was further considered as of major
importance by the WHO.7 On the other hand, bacteria may be intrinsically resistant or may
acquire resistance by de novo mutation or via the acquisition of resistance genes from other
organisms. Furthermore, acquired resistance mechanisms permit to a bacterium to produce
enzymes able to destroy the drug, to produce target protection proteins that prevent the drug
binding, the expression of efflux systems that avoid the drug to reach its intracellular target, the
target site modification, the production of an alternative metabolic pathway to circumvent the
action of the drug. Thus, efflux mechanisms constitute important determinants of resistance to
antibiotics in human pathogens.8-11 Additionally, susceptible Gram-negative bacteria can
become resistant particularly by changing the permeability of their membranes or by reducing
the number of channels available for drugs to diffuse through.12-13 Given the clinical
significance of antimicrobial exporters9, it is clear that efflux must be considered in formulating
strategies for treatment of drug-resistant infections, by the development of new agents less
impacted by efflux or by targeting efflux directly with efflux inhibitors restoring the activity of
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

Page 3 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
several drugs on acquired MultiDrugResistant (MDR) bugs and also allowing the susceptibility
of drugs on naturally resistant bacteria.14
7
8
9
In this context, we have previously reported the ability of a natural compound, namely geraniol,
to decrease the antibiotic resistance of a Gram-negative MDR Enterobacteria strain.15 This
compound has been determined to be a substrate of the AcrAB-TolC pump and conjointly
inhibiting the transport of other molecules including antibiotics by this pump.16 Furthermore,
we recently reported the synthesis of polyaminogeranic acid derivatives and their efficient use
as new chemosensitizers inducing a significant decrease of antibiotic resistance in Gram-
negative bacterial MDR strains.17 On the other hand, we were also able to demonstrate that
water-soluble geraniol parent amino derivative, geranylamine as well as NV716 was effective
as an efflux pump inhibitor (Figure 1).18-19
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. Structure of Geraniol, Geranylamine, PAN and NV716
In this context, we will report herein the synthesis, biological evaluation, and structure activity
relationships analysis of a series of new polyaminoisoprenyl antibiotic enhancers against
multidrug resistant Gram-negative bacteria Enterobacter aerogenes and Salmonella enterica
serovar Typhimurium.
Results and discussion
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 39
1
2
3
4
Chemistry
5
6
7
8
Using an efficient titanium reductive amination developed in our laboratory20-21, we have
envisioned a one-step synthesis procedure for the preparation of new polyaminoisoprenyl
derivatives in a Z/E mixture from both citral and farnesal according to the following synthetic
pathway (Table 1).
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 1. Titanium (IV) reductive amination reaction of citral with spermine under various
experimental conditions.
H2N
NH
NH
O
3
1) Ti(Oi-Pr)4 , MeOH
N
N
20 °C, 12 h
H
H
H
+
4
2) NaBH4
,
-78 °C, 2 h
H2N
5p
NH
H2N
1
2
Titanium
source
Entry
Solvent
Yield (%)c
1a
2b
3a
4a
5a
6a
7a
8a
64
51
30
19
33
39
30
21
Ti(Oi-Pr)4
MeOH
MeOH
CH2Cl2
Toluene
THF
Ti(Oi-Pr)4
Ti(Oi-Pr)4
Ti(Oi-Pr)4
Ti(Oi-Pr)4
Ti(OEt)4
MeOH
MeOH
MeOH
Ti(OBu)4
Ti(Ot-Bu)4
a
Reaction performed at -78ºC for 12 h in MeOH on a 0.39
mmol scale of citral, Ti(Oi-Pr)4 (2.02 mmol) and the amine (6
mmol). b Reaction performed at 0 °C. c Isolated overall yield.
First, it clearly appears that isolated yields of compound 5p are highly solvent-dependent. Thus,
the expected amino derivative 5p was obtained in 64% yield performing the reaction in MeOH
(Table 1, entry 1), whereas only moderate yields varying from 19-33% were encountered
performing the reaction in CH2Cl2, toluene and THF, respectively (Table 1, entries 3-5).
Influence of the nature of the titanium source involved was also investigated and chemical yield
variations from 21 to 64% were obtained (Table 1, entries 1, 6-8), best result being observed
using Ti(Oi-Pr)4 as titanium source. Furthermore, under these best experimental conditions,
ACS Paragon Plus Environment

Page 5 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
increasing of the reaction temperature from -78 to 0°C led to a slight decrease of the chemical
yield encountered from 64 to 51%, respectively (Table 1, entries 1 and 2).
Under the best experimental conditions (Table 1, entry 1), we have envisioned the use of
numerous different natural (such as cadaverine, putrescine, spermine...) or synthetic
polyamines in the titanium reductive amination of citral and farnesal and the results are
summarized in Table 2.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2. Polyaminoisoprenyl derivatives obtained from citral (5a-5r) or farnesal (6a-6r).
O
O
H
H
NHR
or
NHR
6a-6r
1) RNH2, Ti(Oi-Pr)4, MeOH
or
20 °C, 12 h
1
4
5a-5r
2) NaBH ,
-78 °C, 2 h
4
IC50 (µM) CHO
RNH2
Isolated Yield (%)
5a-5r
6a-6r
H2N
NH2
5a 49
6a 63
36.24
32.30
NH2
NH2
5b 33
5c (NP)
5d (NP)
6b 49
6c 58
6d 38
>150
32.06
18.06
38.38
H2N
H2N
H2N
-
-
NH2
N
H2N
5e 61
5f 51
5g 52
6e71
6f (NP)
6g 56
>150
>150
>150
85.33
-
O
N
H2N
H2N
N
57.64
O
N
H2N
5h (NP)
5i (NP)
5j 42
6h 62
6i 52
6j 58
-
-
33.75
30.19
10.47
NH2
N
NH2
N
H
H2N
H2N
NH2
N
>150
N
N
H2N
H2N
H
5k 49
5l 64
6k (NP)
6l (NP)
6m 32
30.92
>150
-
-
-
H
N
NH2
N
H
H
H
N
N
5m (NP)
43.50
NH2
N
H
H2N
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 39
1
2
3
4
5
6
7
8
9
H
N
H
N
NH2
5n 58
5o (NP)
5p 64
6n 47
6o 48
6p 72
6q 43
>150
41.71
25.32
142.79
>150
N
H
N
H
H2N
H2N
NH2
-
>150
-
H
N
NH2
NH2
N
H
H2N
H2N
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
O
O
5q (NP)
NH2
NH2
N
H2N
5r 49
6r 63
126.82
>150
NP : not prepared
Whatever the nature of the amines, the expected polyaminoisoprenyl products were obtained
mixture of the Z/E isomers in a 50/50 ratio in moderate to good chemical yields varying from
32 to 72%. It is noteworthy that these reactions have not been yet optimized, the moderate
results being explained by the difficulty to purify the expected polar products.
Antimicrobial activities
In the context of our studies, all the synthesized compounds were firstly screened for their
antimicrobial activity against Gram-negative bacterial strains (Table S-1). Two pairs of
enterobacterial strains were challenged, one E. aerogenes MDR strain Ea289 over-expressing
the efflux pump AcrAB-TolC and its ∆acrB mutant, and one S. Typhimurium MDR strain
BN10055 over-expressing the same efflux pump and its ∆acrB mutant. The data presented in
Table S-1 demonstrated that the MICs of the polyaminoisoprenyl derivatives vary from a
maximum of over 500 µM for 5l and 5n against the four strains to a minimum of 7.8 µM for
compounds 6n and 6o against the Enterobacter strain Ea289∆acrB, and 3.9 µM for 6o against
Salmonella strain BN10055∆acrB. Additionally, except for 6o no significant differences are
noticed against both Salmonella strains indicating that none of the compounds is a substrate of
the efflux pump AcrAB-TolC.
ACS Paragon Plus Environment

Page 7 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In this study, our goal was to identify molecules capable of decreasing antibiotic resistance of
Gram-negative MDR strains. Thus, we chose drugs for which these strains had high MIC values
and our compounds were tested for their ability to decrease resistance of the MDR strains to
erythromycin, doxycycline, nalidixic acid and chloramphenicol in a synergistic assay. The
compounds were used at a sub-inhibitory concentration corresponding to a quarter of their
respective MICs to avoid an intrinsic action and to ascertain that the effect observed resulted
from the combination of the molecules used. For the two MDR strains EA289 and BN10055
we determined the ratio between the MIC of each antibiotic when applied alone and the MIC
of the antibiotic in the presence of a desired compound. The cumulative MIC ratios for each
compound are presented in Figure 2A and 2B, respectively.
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

Journal of Medicinal Chemistry
Page 8 of 39
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. Cumulative MIC ratios for each compound for their ability to decrease resistance of
MDR strains EA289 (Figure 2A) and BN10055 (Figure 2B) to erythromycin, doxycycline,
nalidixic acid and chloramphenicol in a synergistic assay.
ACS Paragon Plus Environment

Page 9 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Although three compounds (5e, 5f, and 5g) are ineffective, the twenty-three other ones
exhibited an efficiency at least in one of the combinations with the antibiotics. In the case of
Enterobacter, 4 compounds (6b, 6c, 6h, 6j) showed a cumulative score over 150 (Figure 2A)
while 5 others (6b, 6j, 6e, 5r, 6r) had the same behavior in the case of Salmonella (Figure 2B).
In this context, the data of Figure 2A and 2B highlighted that 6b and 6j are the most efficient
and promising 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
We also observed, that several compounds decreased the MIC of the antibiotic under the
threshold of susceptibility according to the Comité de l’Antibiogramme de la Société Française
de Microbiologie (CA-SFM)22, (≤ 2 mg/L for the four antibiotics tested) (Table 3).
Table 3: Decrease of chloramphenicol, nalidixic acid, doxycycline and erythromycin
resistance in the presence of the polyaminoisoprenyl enhancers 5a-6r.
a
Gain factor
Ea289
BN10055
Adjuvant
Chl
Nal Doxy Ery
Chl
Nal Doxy Ery
b
1024 4096
32
32
16
2
256 1024 4096
64
16
16
2
256
1
No
5a
5b
5e
5f
4
8
4
8
4
32
1
8
8
16
32
4
1
2
2
1
1
1
1
2
2
1
1
1
5g
5j
1
1
2
1
2
1
2
1
16
16
8
8
16
32
32
32
16
16
4
32
64
16
16
64
128
64
128
64
16
128
32
64
64
16
32
Nd
8
5k
5l
8
c
c
c
c
2
Nd
2
Nd
1
Nd
4
Nd
c
5n
5p
5r
8
4
8
Nd
8
16
8
16
32
8
4
16
16
8
16
16
32
32
16
16
32
16
32
32
2
4
8
16
128
256
256
32
64
16
6a
6b
6c
6d
6e
6g
6h
6i
4
8
32
64
4
16
16
16
16
8
16
16
8
64
64
64
32
32
32
8
8
2
16
8
16
8
16
4
8
32
8
128 128
32 64
8
4
8
8
8
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 39
1
2
3
4
5
6
7
8
9
6j
6m
6n
8
8
32
16
8
64
32
16
2
128
8
16
8
128
16
32
8
64
4
8
1
4
4
8
16
8
2
6o
2
1
1
4
2
6p
16
4
32
2
16
8
8
32
4
64
16
128
64
8
Ndc
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
6q
2
8
4
6r
32
16
16
8
16
16
4
16
8
8
1
PAßN
8
16
8
a
Gain factor determined as the ratio of the MIC of the antibiotic
alone for each strain to the MIC of the same antibiotic in the presence
of the considered compound at a concentration corresponding to its
MIC/4. A gain factor greater than 1 indicates an improvement of
activity. MIC determined for the antibiotics alone. Nd, Not
determined.
b
c
This is particularly the case against Ea289 for most compounds in combination with
doxycycline. Concerning erythromycin five compounds were observed to decrease the MIC of
Ea289 under the threshold. To a lesser extent, the same observations can be drawn for the
combination of seven compounds and doxycycline against Salmonella. Noticeably, derivatives
5k, 6b, 6e, 6h, 6i and 6j decrease the MIC of doxycycline towards these two strains.
Surprisingly, whereas these compounds in the presence of erythromycin led to an efficient
combination against Enterobacter, none of these significantly improve the MIC of erythromycin
against Salmonella.
Taken together these results suggested that the two compounds 6b and 6j might be the most
active. However, this analysis did not consider the concentration used to get the synergy with
the different antibiotics. Consequently, in order to generate comparable data we decided to
correlate the efficiency according to the concentration used by introducing an “Efficiency
Parameter” (EP) that corresponds to the sum of gain for each compound obtained for the four
antibiotics tested and divided by the concentration of the molecule used in the assay (Table S-
2). It is noteworthy that the gain factor was defined as corresponding to the ratio of the MIC of
the antibiotic tested alone to the MIC of the antibiotic obtained in the presence of the considered
adjuvant.
ACS Paragon Plus Environment

Page 11 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
This EP varies from 0.01 to 7.17 in arbitrary units allowing us to classify the compounds into
three groups: one group corresponding to non-efficient molecules, a second group with
molecules with an average score of EP varying from 0.01 to 0.57 and a third one with an EP
higher than 1 and reaching 7.17 in the best case (Table S-2).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
It is noticeable that PAßN, the reference inhibitor of Gram-negative efflux has through this
classification an efficiency of 0.47 making it part of the second group. The third group
comprised 13 compounds including the three compounds previously identified. This third group
appeared as the most efficient. To better characterize our compounds and considering this kind
of molecules as putative drug adjuvants, the cytotoxicity of each molecule has been
investigated.
Cytotoxicity was assayed with Chinese hamster ovary (CHO) cells for all the compounds which
present IC50 ranging from 1 to 150 µM. A major part of our compounds exhibited an IC50
comprised between 18.06 and 142.79 µM while 10 compounds had IC50 over 150 µM
suggesting that they were minimally toxic. A toxicity index varying from 0.06 to 19.2 was
subsequently defined as the ratio of IC50 to the concentration used leading to a decrease of MIC
of at least 8 times. Thus, we were able to directly compare the EP of each compound versus its
toxicity index as plotted in Figure 3.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 39
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 3. Efficiency/toxicity index correlation for all the polyaminoisoprenyl derivatives
It is noteworthy that two compounds 5r and 6r exhibited the best cytotoxicity index and 6j and
6b the best efficiency, respectively, and that they were used at similar concentration against the
considered bacteria in order to compare their efficiency in the same range of concentration (see
Table S-2). In this context, 5r and 6r showed efficiency index about 2 times lower than 6j but
presented a better toxicity index. By comparison, PAßN (red dot on Figure 3) presented a
moderate to weak efficiency index of 0.47 and a toxicity index of 3.95. Thus, six compounds
namely 6p, 6r, 5p, 5r, 6n, and 6m were identified to possess better parameters than PAßN.
Additionally, 5r was identified as the most potent compound by considering all these criteria.
We suppose, as already postulated by others23 that the death process could involve electrostatic
interactions and is related to the high density of charges exposed at the surface of bacterial
membranes. Thus, the electrostatic compensation of the negative charges of the bacterial
envelope could be provided by the cationic charges of the substrate leading for the bacteria to
a loss of their natural counterions. Based on the number of positive charges involved (number
of nitrogen atoms present in the structure) the different compounds were classified and three
groups of molecules emerged (Table 4).
ACS Paragon Plus Environment

Page 13 of 39
Journal of Medicinal Chemistry
1
2
3
4
Table 4. Stucture-activity relationships analysis
Name
5j
Structure
Efficiencya Indexb
5
6
7
8
0.01
0.06
NH2
N
N
H
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
1.66
3.58
7.17
1.98
3.86
1.34
N
5k
6i
N
N
H
H
NH2
N
N
H
H
6j
NH2
N
N
H
H
N
NH2
0.00
0.06
5l
N
H
N
H
H
N
NH2
NH2
0.57
1.61
4.80
9.15
5p
6p
N
N
H
H
H
N
N
H
N
H
NH2
N
H
N
3.54
3.65
19.20
16.23
6r
5r
NH2
NH2
N
H
N
NH2
H
N
N
H
H
N
NH2
0.00
1.49
0.06
5.37
5n
N
N
H
H
H
H
N
N
6m
NH2
N
N
H
H
H
H
N
NH2
N
2.88
10.69
6n
N
N
N
H
H
H
a Efficiency parameter that corresponds to the sum of gain factor for each compound reported
for the four antibiotics tested and divided by the concentration of the molecule used in the
assay.b Index parameter is determined by the ratio of the IC50 to the concentration giving a
decrease of MIC of at least 8 times for almost 3 antibiotics on both species.
The most efficient compound 6j was encountered in the first group with an efficiency score of
7.17. This compound is a farnesyl derivative, with a triterpene structure linked to a linear
triamine (three positive charges). Otherwise, the parent geranyl compound 5j, was neither
efficient nor showed a good index of toxicity indicating that the length of the terpene moiety is
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 39
1
2
3
4
5
6
7
8
of major importance. While the activity of 6i is near half of the activity of 6j, its index of toxicity
is about 3 times higher suggesting that the nature of the amino-methyl group plays an important
role for toxicity.
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
On the other hand, some compounds including four amino groups such as 6r and 5r in their
close present interesting results in terms of activity and toxicity. By comparing 5l and 5p it
clearly appears that the number of carbon atoms between the amino groups remains crucial on
the outcome of the observed antimicrobial activities whereas the number of positive charges is
conserved. Furthermore, comparison of the structure for derivatives 6p and 5p lead to the
conclusion that the length of the terpene chain involved is of major importance for the activity
encountered whereas the polyamine chain is conserved. Moreover, the presence of a ramified
amine in compounds 6r and 5r improve their efficiency and toxicity index with respect to their
parent derivatives 5p and 6p involving a linear spermine group. In the third group, the
importance of the length of the terpene moiety with respect to the encountered efficiency was
confirmed by studying typically derivatives 5n and 6m. It is noteworthy that an attempt to
correlate the efficiency or cytotoxicity of the different compounds with their intrinsic
lipophilicity by considering the LogP parameter, which reflects the true behavior and
bioavailability of an ionizable compound in a solution at a given pH, failed (Table S-2).
All these derivatives have been prepared as a 50/50 mixture of cis and trans isomers. At this
stage, it appears of interest to determine the impact of each isomer on the observed biological
activities. In a first approach and due to its interesting biological data, the synthesis of pure Z-
5r and E-5r derivatives has been envisioned with the use of a titanium amination reaction
involving pure isomers of neral or geranial as starting materials.
Surprisingly, a subsequent titanium amination reaction in methanol in the presence of tris(3-
aminopropyl)amine 7 led to a mixture of the Z/E isomers of 5r in a 50/50 ratio in a non-
optimized 35% isolated yield (Scheme 1).
ACS Paragon Plus Environment

Page 15 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
NH2
5r
N
Z-
NH2
N
H
1) Ti(Oi-Pr)4, MeOH
NH2
NH2
CHO
+
20 °C, 5-6h
N
H2N
+
50/50 Z/ E mixture
NH2
2) NaBH4, 0 ºC, 2 h
35% overall yield
7
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 equiv.
1
geranial Z-
N
N
5r
E-
(90% pure form)
H
1 equiv.
NH2
Scheme 1. Titanium reductive amination of geranial Z-1 with tris(3-aminopropyl) amine 7.
Thus, whatever the experimental conditions applied (data not shown) it was impossible to
obtain the expected polyamino derivatives in a pure stereoselective form. We have been able to
demonstrate that this racemization occurs by mixing the pure aldehyde and the amine in the
absence of titanium isopropoxide. Indeed, after two days the corresponding imines were
obtained as a mixture of the two isomers suggesting an isomerization process during the imine
formation.24-30
These disappointing results led us to envision another strategy involving a direct nucleophilic
substitution of the considered polyamine on geranyl chloride 8 and neryl chloride 9 derivatives
(Scheme 2).
NH2
CH2Cl
N
5r
Z-
N
H
8
53%
NH2
NH2
NH2
NEt3 (2.2 equiv.)
THF, 20 °C, 24 h
or
+
N
H2N
or
7
NH2
5r
1.25 equiv.
N
N
CH2Cl
E-
H
9
48%
NH2
Scheme 2. Synthesis of pure Z-5r and E-5r by nucleophilic substitution of tris(3-
aminopropyl)amine 7 on geranyl chloride 8 and neryl chloride 9 derivatives.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 39
1
2
3
4
5
6
7
8
9
Under these experimental conditions, the expected products were obtained as pure isomers Z-
5r and E-5r in 53 and 48% isolated yield, respectively. Their biological efficiency and
cytotoxicity have been evaluated involving different salt formulations as summarized in Table
S-3).
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
It clearly appears that no significant differences were obtained in terms of biological activities
and cytotoxicity for the different isomers and formulations used.
Animal study
Because of these outcomes, we have used a mixture of Z-5r/E-5r to investigate the potent
mechanism of action of these derivatives against Gram-negative bacteria and evaluate their
efficiency in a very preliminary in vivo animal model. Thus, to ascertain the in vivo activity of
compound 5r, mice were used as animal model to determine their tolerance to this compound
as well as the in vivo activity of 5r/doxycycline combination against S. Typhimurium
infection.31 In a first approach, the results indicate a good tolerance of the mice towards 5r to a
concentration of 50 µM and they did not show signs of acute toxicity (Figure 4A). On the other,
the treatment by a doxycycline/5r combination (64 mg/Kg for doxycycline and 0.5 mg/Kg for
5r, respectively) of S. Typhimurium infected mice lead to a 43% survival rate whereas all the
mice died by using a treatment involving saline solution or doxycycline alone as illustrated in
Figure 4B.
Figure 4. A) Mice survival 24h after administration with saline or 5r (4.06, 8.12, 16.25, 32.5
mg/Kg). B) Survival of mice infected with S. Typhimurium following administration with
ACS Paragon Plus Environment

Page 17 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
saline, doxycycline alone or a doxycycline/5r combination solution (64 mg/Kg for doxycycline
and 0.5 mg/Kg for 5r, respectively). Survival of mice number corresponds to the number of
mice euthanized.
7
8
9
These results, even if there are preliminary, tend to demonstrate the potent in vivo enhancing
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
activity of polyaminoisoprenyl compounds when associated with doxycycline against S.
Typhimurium. Nevertheless, since the mechanism of action of these compounds remained
unclear, we undertook to perform four different experiments to determine how derivative 5r (as
a Z/E mixture) targeted our Gram-negative E. aerogenes EA289 bacteria model, a kanamycine-
susceptible derivative of the MDR clinical isolate Ea27.32
Mechanism of action
We have previously determined that compound 5r strongly decrease the MIC of four antibiotics
against MDR Enterobacter and Salmonella strains, and that this activity was not or weakly
observed against their isogenic ∆AcrAB derivatives. This observation suggests that the efflux
pump AcrAB-TolC could be the target of the involved compound. Nevertheless, such an
inhibition of the efflux pump may result from various actions including either direct interaction
with the pump by blocking the antibiotic transfer or competing with the antibiotic during its
transfer, by energy disruption or by inhibition of the pump assembly. MIC determination
involves the incubation of bacteria in the presence of the selected compound and of the
antibiotic for about 18 hours. Thus, the observed synergy may result from the combination of
numerous other effects. To eliminate such unwanted observations, we set up a series of real
time assays allowing us to follow the interaction of the compound more accurately with the
bacteria (Figure 5)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 39
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
Figure 5. Evaluation of the possible modes of action of 5r against EA289: A) Outer membrane permeabilization assay by evaluating the rate of
nitrocefin hydrolysis. B) H33342 accumulation assay (cell wall penetration) C) Efflux performance of the bacteria by evaluating glucose-triggered
1,2’-diNA efflux. D) Transmembrane potential disruption assay (diSC3(5) assay).
ACS Paragon Plus Environment

Page 19 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
Each experiment was performed using standard molecules as positive control, at a concentration
of 100 µM.33 In a first attempt we have envisioned to determine if a permeabilization and/or
disruption of the outer membrane could occur. Thus, outer-membrane permeabilization was
followed by monitoring the nitrocefin hydrolysis in the periplasm by the constitutively
expressed ß-lactamase since an important color change from yellow to red would be observed
in the positive.34-35 It clearly appears that during the first 200 seconds of incubation, no outer
membrane permeabilization was observed by using 5r whereas in the same time benzalkonium
chloride induced a complete outer membrane disruption (Figure 5A). Thus, even after 15
minutes, a slight effect was observed for 5r without reaching the maximum observed for
benzalkonium chloride.
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
Furthermore, membrane perturbation can also be observed by monitoring H33342 dye entrance
into the bacteria (Figure 5B). Fluorescence of this compound correlates with its binding to DNA
in the cytoplasm of the bacteria. Thus, even if the H33342 dye entrance occurred in non-treated
bacteria, this kinetic is widely increased by adding compound 5r. Taken together, these data
suggested that derivative 5r was unable to strongly disrupt outer membrane integrity but
allowed the diffusion of molecules such as H33342 into the bacteria. Furthermore, these
observations for nitrocefin and H33342 are in full agreement with the fact that their respective
LogD are -0.95 and 3.76 suggesting a facilitated diffusion of H33342 relative to nitrocefin
through the membranes of the bacteria. Moreover, since the bacterial strain considered is devoid
of porins36 which are used by ß-lactam compounds such as nitrocefin to cross the outer
membrane of Gram-negative bacteria, our results confirm the property of 5r to disrupt the
integrity of the outer membrane and improve its permeability.
On the other hand, we investigated the ability of derivative 5r to act as an inhibitor of the
AcrAB-TolC efflux pump. In this context, the bacteria were loaded with 1,2’dNA dye, a
19
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 39
1
2
3
4
5
6
7
8
9
substrate of the AcrAB-TolC efflux pump.37 Bacteria were then incubated with the compound
(100 µM) before addition of glucose as energy source. The active transport of 87% of the dye
is observed for non-treated bacteria (Figure 5C, black curve). By using PAßN, a well-known
reference for efflux inhibition, a 25% retention of the dye within the bacterial membranes was
observed (Figure 5C, blue curve), whereas addition of derivative 5r led to a stronger inhibition,
resulting in 68% retention of the dye (Figure 5C, orange curve).
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
RND efflux pumps, such as the AcrAB-TolC considered here, use the proton gradient across
the inner membrane as energy source. Thus, we asked if the efflux inhibition observed in Figure
5C by 5r, may result in the disruption of the transmembrane potential. For that, we used the
membrane potential sensitive probe 3,3′-Dipropylthiadicarbocyanine iodide (DiSC3(5)) which
concentrated across the inner membrane and self-quenched its fluorescence. If a compound
impairs the membrane potential, DiSC3(5) is released into the medium leading to a fluorescence
increase. Figure 5D summarizes the results obtained by using compound 5r and polymyxin B,
known to strongly disrupt the membrane. A decrease of 14% of the membrane potential was
observed with polymyxin B, while no change was encountered by using compound 5r
suggesting that this later did not inhibit the efflux transport observed in Figure 5C by disrupting
the transmembrane potential.
All these data suggest that the observed synergy, between compound 5r and the tested intra-
cytoplasmic targeted antibiotics, results from a dual effect i.e. a specific OM permeabilization
combined with an AcrAB-TolC pump impairment. Thus, although we cannot exclude the
impairment of the pump, this latter could be the result of a side effect of the outer membrane
permeabilization. Nevertheless, a specific inhibition of AcrAB-TolC remains possible and
researches in this regard are under current investigation.
3. Conclusion
20
ACS Paragon Plus Environment

Page 21 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
An original chemical strategy has been developed affording new polyaminoisoprenyl
compounds in moderate to good yields. Amongst all the synthetized derivatives, some of them
exhibited a strong effect against resistant Gram-negative bacteria of four antibiotics belonging
to four different families. Thus, this activity was correlated to the ability of the
polyaminoisoprenyl derivatives to alter bacterial outer membrane integrity. Studies are now
underway to determine if this restoration of antibiotic susceptibility occurs also by a direct
interaction of the molecule with the efflux pump or by another mechanism.
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. Experimental section
4.1 Materials
All the solvents were purified according to reported procedures, and the reagents used were
commercially available. Methanol, ethyl acetate and dichloromethane were purchased from
Sigma-Aldrich and used without further purification. Column chromatography was performed
on Merck silica gel (70-230 mesh). 1H NMR and 13C NMR spectra were recorded in MeOD on
a Bruker AC 300 spectrometer working at 300 MHz and 75 MHz, respectively (the usual
abbreviations are used: s: singlet, d: doublet, t: triplet, q: quadruplet, m: multiplet). All chemical
shifts are given in ppm. Mass spectroscopy analysis has been performed by the Spectropole
(Analytical Laboratory) of Aix-Marseille University (Marseille). The purity of the compounds
was checked by analytical HPLC (C18 column, eluent CH3CN-water-TFA (90:10:0.025,v/v/v),
0.5-1 mL/Min) with PDA detector spanning from 210 nm to 310 nm. All compounds possessed
purity above 95%, as determined by analytical HPLC-PDA at 210 nm.
All bacterial experiments were performed on the BAC-Screen platform (UMR-MD1), with a
Freedom EVO 150 liquid handling system (Tecan Lyon-France) and were independently
repeated at least three times. Antibiotics (chloramphenicol, doxycycline and nalidixic acid)
21
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 39
1
2
3
4
5
6
7
8
9
were purchased from Sigma (St Quentin Fallavier-France); erythromycin-lactobionate was
purchased from Amdipharm ltd (Dublin-Ireland). They were dissolved in ethanol, water or
dimethyl sulfoxide (DMSO), as further precised. The chemicals benzalkonium chloride,
phenylalanine-arginine beta-naphthylamide (PAßN), polymyxin-B and carbonyl cyanide m-
chlorophenylhydrazone (CCCP) were purchased from Sigma-Aldrich (St Quentin
Fallavier, France). Nitrocefin was purchased from Oxoid (Basingstoke, England), 1,2'-
Dinaphthylamine (1,2'-diNA) was purchased from T.C.I (Zwijndrecht, Belgium), 3,3' -
Dipropylthiadicarbocyanine iodide (diSC3(5)) was purchased from Anaspec (Freemont, USA),
Hoechst 33342 (H33342) was purchased from Molecular probes (Eugene, USA).
Benzalkonium chloride, PAßN and polymyxin-B were dissolved in double distilled water and
stored at -20°C until use. Nitrocefin was dissolved in DMSO 5%. 1,2'-diNA, diSC3(5), H33342
were dissolved in DMSO 100%.
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
4.2 General procedure for the synthesis of compounds 5a-6r
The general synthetic pathway is illustrated for the preparation of compound 5p.
A mixture of citral (345 mg, 2.27 mmol), titanium(IV)isopropoxide (645 mg, 2.27 mmol) and
spermine (2.27 mmol) in absolute methanol (5 mL) was stirred at room temperature for 12
hours. Sodium borohydride (172 mg, 4.5 mmol) was then added at 0ºC and the resulting mixture
was stirred for an additional 2 hours. The reaction was then quenched by adding water (1 mL).
Stirring was maintained at room temperature for 20 minutes. After filtration over a pad of Celite
washing with methanol and ethylacetate, the solvents were removed under vaccuum and the
crude amine was purified by flash chromatography on silicagel. using CH2Cl2/MeOH/ NH4OH
(7/3/1) as eluent affording the expected coupling product 5p in 64% yield.
Compound 5a, 49% yield; Yellow solid; 1H NMR (MeOD):  = 5.08-4.90 (m, 2H), 3.01-2.95
(m, 2H), 2.68-2.64 (m, 5H), 2.09-2.03 (m, 3H), 1.62-1.40 (m, 12H). 13C (MeOD):  = 142.24,
22
ACS Paragon Plus Environment

Page 23 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
142.16, 132.80, 132.78, 123.05, 116.84, 116.78, 51.76, 51.72, 44.19, 41.20, 31.80, 26.20, 26.14,
25.70, 17.80, 17.14. C12H24N2 m/z 197.1945 (100%, (M+H+)).
7
8
9
Compound 5b, 33% yield; Yellow solid; 1H NMR (MeOD):  = 5.05-4.93 (m, 2H), 2.93-2.78
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
(m, 4H), 2.18-1.92 (m, 5H), 1.63-1.18 (m, 15H). C (MeOD):  = 136.20, 133.14, 124.83,
120.12, 47.18, 44.12, 40.60, 39.17, 33.14, 26.82, 25.38, 18.01, 17.84. C13H26N2 m/z 211.2075
(100%, (M+H+)).
Compound 5e, 61% yield; Yellow solid; Mixture of Z/E isomers; 1H NMR (MeOD):  = 5.08-
4.82 (m, 2H), 2.95-2.70 (m, 4H), 2.59-2.40 (m, 6H), 2.02-2.00 (m, 4H), 1.83-1.66 (m, 4H),
13
1.61-1.51 (m, 12H). C (MeOD):  = 142.21, 131.83, 131.81, 122.95, 117.35, 53.70, 51.29,
48.22, 44.01, 43.67, 31.78, 27.61, 26.15, 25.82, 23.47, 17.51, 15.92. C17H32N2 m/z 265.2578
(100%, (M+H+)).
Compound 5f, 51% yield; Yellow solid; Mixture of Z/E isomers; 1H NMR (MeOD):  = 5.08-
4.96 (m, 2H), 3.33-2.95 (m, 8H), 2.76-2.71 (m, 2H), 2.23-2.03 (m, 6H), 1.82-1.59 (m, 12H).
13C (MeOD):  = 177.21, 177.15, 143.24, 141.24, 141,21, 132.80, 123.05, 117.32, 115.34,
48.12, 48.09, 46.94, 46.89, 44.82, 44.01, 31.80, 31.78, 31.09, 29.78, 28.07, 27.52, 26.82, 19.43,
18.03, 17.95, 16.45. C17H30N2O m/z 279.2364 (100%, (M+H+)).
Compound 5g, 52% yield; Yellow solid; 1H NMR (MeOD):  = 5.05-4.81 (m, 2H), 3.51-3.33
(m, 4H), 2.93-2.71 (m, 6H), 2.27-2.01 (m, 6H), 1.62-1.53 (m, 14H). 13C (MeOD):  = 141.84,
131.75, 131,21, 123.05, 114.32, 114.12, 66.85, 66.79, 52.81, 51.82, 48.19, 48.09, 43.13, 31.79,
31.09, 27.49, 26.18, 25,23, 17.46, 16.83. C17H32NO m/z 281.2516 (100%, (M+H+)).
Compound 5j, 42% yield; Yellow solid; Mixture of Z/E isomers; 1H NMR (MeOD):  = 5.04-
13
4.97 (m, 2H), 2.91-2.75 (m, 5H), 2.18-1.97 (m, 7H), 1.67-1.08 (m, 21H). C (MeOD):  =
136.77, 132.43, 131.45, 125.41, 118.12, 55.84, 55.54, 44.12, 39.68, 39.45, 39.37, 38.12, 30.60,
27.07, 26.62, 25.70, 17.18, 16.95, 14.15. C17H35N3 m/z 282.2846 (100%, (M+H+)).
23
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 39
1
2
3
4
5
6
7
8
9
Compound 5k, 49% yield; Mixture of Z/E isomers, Yellow solid; 1H NMR (MeOD):  = 5.15-
4.95 (m, 2H), 2.97-2.68 (m, 6H), 2.50-2.48 (m, 4H), 2.09-1.98 (m, 4H), 1.68-1.48 (m, 19H),
0.95 (t, J = 6Hz, 6H). 13C (MeOD):  = 136.10, 136.04, 132.80, 124.05, 123.92, 117.32, 49.51,
47.93, 47.67, 47.30, 44.00, 40.78, 39.68, 26.62, 25.92, 25.70, 24.63, 17.80, 17.03, 16.90, 16.13.
C20H41N3 m/z 324.3334 (100%, (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
Compound 5l, 64% yield; Yellow solid; Mixture of Z/E isomers; 1H NMR (MeOD):  = 5.06-
13
4.94 (m, 2H), 3.04-2.56 (m, 16H), 2.03-1.99 (m, 2H), 1.64-1.58 (m, 14H). C (MeOD):  =
142.32, 142.29, 131.95, 123.05, 116.82, 51.32, 51.24, 49.32, 48.19, 48.03, 48.01, 44.12, 41.12,
31.79, 31.67, 26.15, 26.10, , 17.75, 17.69, 16.83. C16H34N4 m/z 283.2817 (100%, (M+H+)).
Compound 5n, 58% yield; Mixture of Z/E isomers, Yellow solid; 1H NMR (MeOD):  = 5.07-
13
4.92 (m, 2H), 3.02-2.52 (m, 27H), 2.09-1.58 (m, 15H). C (MeOD):  = 141.34, 141.29,
131.77, 131.54, 124.56, 122.95, 118.89, 117.02, 117.01, 52.46, 51.30, 51.14, 49.64, 48.19,
48.12, 44.15, 42.45, 41.03, 41.01, 31.77, 26.18, 17.96,17.79, 16.92. C20H44N6 m/z 369.3661
(100%, (M+H+)).
Compound 5p, 64% yield; Mixture of Z/E isomers, white solid; 1H NMR (MeOD):  = 5.27-
5.10 (m, 3H), 3.36-3.19 (m, 4H), 2.74-2.63 (m, 10H), 2.12-2.08 (m, 6H), 1.76-1.31 (m, 22H).
13C (MeOD):  = 140.12, 133.18, 132.82, 125.60, 124.18, 123.30, 50.83, 48.38, 47.98, 47.94,
41.39, 41.22, 33.49, 30.40, 30.30, 28.56, 27.98, 27.95, 26.42, 26.38, 24.14, 18.24, 16.83.
C20H42N4 m/z 339.3482 (100%, (M+H+)).
Compound 5r, 49% yield; Mixture of Z/E isomers, Yellow solid; 1H NMR (MeOD):  = 5.27-
5.10 (m, 3H), 3.34-3.19 (m, 4H), 2.76-2.53 (m, 9H), 2.12-2.09 (m, 6H), 1.77-1.63 (m, 20H).
13C (MeOD):  = 141.10, 140.53, 133.66, 133.29, 125.99, 125.90, 124.60, 123.71, 54.95, 53.97,
53.50, 49.15, 48.42, 48.37, 41.80, 41.60, 33.88, 30.97, 28.37, 28.34, 26.87-26.48, 19.33, 18.72,
17.32. C19H40N4 m/z 325.3326 (100%, (M+H+)).
24
ACS Paragon Plus Environment

Page 25 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Compound 6a, 63% yield; Mixture of Z/E isomers, Yellow solid; 1H NMR (MeOD):  = 5.30-
4.99 (m, 3H), 3.01-2.92 (m, 2H), 2.70-2.67 (m, 5H), 2.09-1.93 (m, 7H), 1.62-1.49 (m, 15H).
13C (MeOD):  = 140.02, 139.84, 137.15, 137.09, 136.12, 125.17, 124.23, 116.80, 116.78,
51.78, 51.54, 44.19, 41.20, 39.90, 35.48, 26.82, 26.76, 26.32, 25.66, 23.30, 17.62, 16.82.
C17H32N2 m/z 265.2578 (100%, (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
Compound 6b, 49% yield; Mixture of Z/E isomers, Yellow solid; 1H NMR (MeOD):  = 5.18-
5.01 (m, 4H), 3.22-3.08 (m, 3H), 2.84-2.64 (m, 4H), 2.09-1.90 (m, 8H), 1.66-1.02 (m, 18H).
13C (MeOD):  = 141.15, 141.09, 136.77, 136.64, 136.44, 136.33, 132.43, 132.17, 125.41,
125.37, 54.34, 47.25, 47.17, 41.33, 40.90, 40.80, 40.14, 32.95, 30.53, 28.86, 27.73, 27.77,
27.65, 27.42, 26.02, 25.97, 23.73, 17.82, 16.52, 16.25. C18H34N2 m/z 279.2745 (100%, (M+H+)).
Compound 6c, 58% yield; Mixture of Z/E isomers, Yellow solid; 1H NMR (MeOD):  = 5.19-
4.99 (m, 3H), 2.73-2.56 (m, 4H), 3.32-3.06 (m, 3H), 2.73-2.56 (m, 4H), 2.03-1.80 (m, 9H),
1.67-1.09 (m, 20H). 13C (MeOD):  = 141.52, 141.41, 136.54, 136.42, 132.45, 132.24, 132.18,
125.73, 125.41, 125.37, 125.05, 121.15, 120.11, 120.02, 54.446, 46.88, 41.15, 40.90, 40.80,
33.38, 33.07, 32.96, 28.79, 28.73, 27.83, 27.41, 27.34, 26.05, 26.00, 24.39, 23.82, 23.76, 22.17,
17.86, 16.60, 16.19. C19H36N2 m/z 293.2884 (100%, (M+H+)).
Compound 6d, 38% yield; Mixture of Z/E isomers, Yellow solid; 1H NMR (MeOD):  = 5.28-
5.11 (m, 4H), 3.43-3.19 (m, 3H), 2.75-2.61 (m, 4H), 2.14-2.06 (m, 9H), 1.77-1.40 (m, 20H).
13C (MeOD):  = 140.67, 140.58, 136.61, 136.28, 132.21, 132.15, 125.83, 125.40, 125.37,
125.13, 125.00, 121.92, 121.28, 47.39, 41.87, 40.89, 40.17, 33.03, 27.83, 27.76, 27.38, 25.94,
25.55, 23.73, 17.79, 16.45, 16.12. C20H38N2 m/z 307.3052 (100%, (M+H+)).
Compound 6e, 71% yield; Mixture of Z/E isomers, Yellow solid; 1H NMR (MeOD):  = 5.14-
4.92 (m, 3H), 2.93-2.68 (m, 6H), 2.56-2.48 (m, 4H), 2.09-1.98 (m, 6H), 1.68-1.48 (m, 21H).
13C (MeOD):  = 140.12, 140.00, 136.10, 136.01, , 124.03, 123.92, 117.42, 54.70, 51.38, 48.19,
25
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 39
1
2
3
4
5
6
43.27, 39.72, 31.40, 27.50, 26.34, 25.42, 25.34, 25.20, 23.56, 17.62, 16.83, 15.18. C22H40N2
m/z 333.3194 (100%, (M+H+)).
7
8
9
Compound 6g, 56% yield; Mixture of Z/E isomers, Yellow solid; 1H NMR (MeOD):  = 5.09-
4.92 (m, 3H), 2.91-2.61 (m, 6H), 2.56-2.48 (m, 4H), 2.08-1.98 (m, 6H), 1.67-1.46 (m, 21H).
13C (MeOD):  = 140.32, 140.14, 139.86, 137.61, 136.68, 132.48, 132.25, 125.83, 125.43,
125.32, 66.70, 66.32, 53.73, 51.92, 48.19, 44.00, 39.37, 31.20, 27.49, 26.82, 26.30, 25.66,
17.62, 16.90, 15.84. C22H40N2O m/z 349.3120 (100%, (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
Compound 6h, 62% yield; Mixture of Z/E isomers, Yellow solid; 1H NMR (MeOD):  = 5.09-
4.92 (m, 3H), 2.91-2.61 (m, 6H), 2.56-2.48 (m, 4H), 2.08-1.98 (m, 6H), 1.67-1.46 (m, 21H).
13C (MeOD):  = 140.11, 139.95, 137.89, 136.58, 132.58, 132.15, 125.99, 125.43, 125.17,
52.36, 52.03, 51.68, 51.56, 48.19, 43.82, 40.29, 39.77, 31.02, 27.34, 26.82, 26.78, 26.28, 25.89,
25.66, 24.32, 17.54, 16.90, 15.84. C25H48N4 m/z 405.3875 (100%, (M+H+)).
Compound 6i, 52% yield; Mixture of Z/E isomers, Yellow solid; 1H NMR (MeOD):  = 5.17-
4.97 (m, 3H), 3.20-3.18 (m, 4H), 2.66-2.54 (m, 7H), 2.02-1.90 (m, 10H), 1.63-1.24 (m, 24H).
13C (MeOD):  = 140.12, 140.00, 137.62, 136.68, 132.58, 132.26, 125.83, 125.43, 125.37,
125.07, 47.43, 47.28, 41.14, 40.90, 40.78, 31.96, 28.84, 27.78, 27.44, 26.03, 25.99, 22.74,
17.84, 16.57, 16.21. C21H41N3 m/z 335.3342 (100%, (M+H+)).
Compound 6j, 58% yield; Mixture of Z/E isomers, Yellow solid; 1H NMR (MeOD):  = 5.29-
5.14 (m, 4H), 3.34-3.31 (m, 3H), 2.79-2.63 (m, 4H), 2.48-2.45 (m, 4H), 2.26 (s, 3H), 2.14-1.93
(m, 8H), 1.77-1.63 (m, 18H). 13C (MeOD):  = 140.97, 137.51, 137.3, 133.23, 132.96, 126.74,
126.29, 126.24, 126.02, 125.82, 57.79, 57.49, 50.90, 49.16, 48.35, 43.18, 41.75, 41.64, 29.53,
28.70, 28.28, 26.88, 26.83, 24.61, 18.70, 17.04. C22H43N3 m/z 349.3463 (100%, (M+H+)).
Compound 6m, 32% yield; Mixture of Z/E isomers, Yellow solid; 1H NMR (MeOD):  = 5.12-
13
4.89 (m, 3H), 3.05-2.48 (m, 21H), 2.10-1.61 (m, 23H). C (MeOD):  = 136.22, 136.11,
135.60, 135.45, 131.09, 125.07, 124.66, 124.18, 116.78, 116.67, 51.32, 49.69, 49.62, 48.22,
26
ACS Paragon Plus Environment

Page 27 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
48.06, 44.17, 39.53, 39.03, 26.85, 26.54, 25.89, 23.42, 17.61, 16.82. C23H47N5 m/z 394.3786
(100%, (M+H+)).
7
8
9
Compound 6n, 47% yield; Mixture of Z/E isomers, Yellow solid; 1H NMR (MeOD):  = 5.12-
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
4.89 (m, 3H), 3.07-2.53 (m, 25H), 2.11-1.58 (m, 24H). C (MeOD):  = 136.57, 136.10,
135.32, 135.15, 131.09, 125.17, 124.86, 124.78, 116.74, 116.69, 51.39, 49.75, 49.67, 48.19,
48.16, 44.20, 41.09, 39.78, 39.57, 26.95, 26.32, 25.47, 23.25, 17.52, 16.92. C25H52N6 m/z
437.4235 (100%, (M+H+)).
Compound 6o, 48% yield; Mixture of Z/E isomers, Yellow solid; 1H NMR (MeOD):  = 5.03-
4.94 (m, 3H), 2.97-2.95 (m, 2H), 2.67-2.2.61 (m, 4H), 2.01-1.90 (m, 6H), 1.61-1.32 (m, 37H).
13C (MeOD):  = 136.10, 135.00, 134.89, 131.09, 125.17, 125.09, 124.36, 117.24, 117.12,
48.36, 44.27, 42.60, 41.12, 34.55, 29.67, 29.37, 28.87, 28.45, 28.02, 27.52, 27.49, 27.42, 23.12,
17.60, 16.43. C27H52N2 m/z 405.4110 (100%, (M+H+)).
Compound 6p, 72% yield; Yellow solid; 1H NMR (MeOD):  = 5.05-4.93 (m, 3H), 2.93-2.57
(m, 14H), 2.19-1.92 (m, 10H), 1.63-0.97 (m, 23H). 13C (MeOD):  = 142.24, 134.83, 131.09,
124.86, 124.17, 117.32, 47.90, 47.69, 47.64, 46.61, 44.01, 40.60, 39.83, 33.97, 31.20, 26.97,
26.30, 25.66, 25.11, 24.90, 17.62, 16.90, 15.93. C25H50N4 m/z 407.4033 (100%, (M+H+)).
1
Compound 6q, 43% yield; Yellow solid; Mixture of Z/E isomers, H NMR (MeOD):  = 5.03-
13
4.95 (m, 3H), 3.40-2.88 (m, 14H), 2.17-1.45 (m, 31H). C (MeOD):  = 136.20, 136.14,
134.98, 134.83, 131.02, 125.18, 125.07, 124.84, 117.12, 70.64, 69.07, 69.00, 68.87,
46.73,44.25, 39.72, 39.57, 38.86, 33.61, 27.14, 27.09, 26.32, 25.62, 23.33, 17.62, 16.27.
C25H48N2O2 m/z 409.3715 (100%, (M+H+)).
Compound 6r, 63% yield; Yellow solid; 1H NMR (MeOD):  = 5.03-4.95 (m, 3H), 2.93-2.64
13
(m, 8H), 2.36-1.92 (m, 14H), 1.65-1.39 (m, 23H). C (MeOD):  = 142.31, 134.93, 131.09,
124.76, 124.17, 117.39, 52.88, 51.68, 48.19, 44.00, 40.30, 39.33, 31.20, 30.98, 27.67, 26.93,
26.30, 25.66, 17.62, 16.90, 15.32. C24H48N4 m/z 393.3869 (100%, (M+H+)).
27
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 39
1
2
3
4
5
6
7
8
Preparation of geranyl chloride (trans (1-Chloro-3,7-dimethyl-octa-2,6-diene)) 7
To a solution of geraniol (40.1 g, 0.26 mol) and anhydrous LiCl (23 g, 0.54 mol) in CH2Cl2
(250 mL) are added dropwise triethylamine (50 mL, 0.37 mol) and mesyl chloride (30 mL, 0.39
mol). The reaction is stirred at room temperature for 16 h. and then washed with 10 % HCl (2
x 100 mL), sat. Na2CO3 (2 x 100 mL) and brine (2 x 100 mL). The organic layer is dried
(Na2SO4) and the solvents removed in vacuo. The crude dark oil is distillated under reduced
pressure to afford the pure expected trans 1-Chloro-3,7-dimethyl-octa-2,6-diene as a colorless
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
oil. 27.5 g (61 %). H NMR (CDCl3, 400 MHz) : δ = 5.48-5.43 (m, 1H), 5.10-5.07 (m, 1H),
4.10-4.12 (d, J = 6.0 Hz, 2H), 2.14-2.04 (m, 4H), 1.74 (s, 3H), 1.1.69 (s, 3H), 1.61 (s, 3H).
Preparation of neryl chloride (cis (1-Chloro-3,7-dimethyl-octa-2,6-diene)) 8
To a solution of nerol (40.1 g, 0.26 mol) and anhydrous LiCl (23 g, 0.54 mol) in CH2Cl2 (250
mL) are added dropwise triethylamine (50 mL, 0.37 mol) and mesyl chloride (30 mL, 0.39
mol). The reaction is stirred at room temperature for 16 h. and then washed with 10 % HCl (2
x 100 mL), sat. Na2CO3 (2 x 100 mL) and brine (2 x 100 mL). The organic layer is dried
(Na2SO4) and the solvents removed in vacuo. The crude dark oil is distillated under reduced
pressure to afford the pure expected cis 1-Chloro-3,7-dimethyl-octa-2,6-diene 8 as a colorless
oil. 26 g (58%). 1H NMR (CDCl3, 400 MHz) : δ = 5.40-5.35 (m, 1H), 5.01-4.99 (m, 1H), 4.04-
4.02 (d, J = 8.0 Hz, 2H), 2.04-1.97 (m, 4H), 1.66 (s, 3H), 1.62 (s, 3H), 1.53 (s, 3H). 13C NMR
(CDCl3, 75 MHz) : δ = 142.69, 131.90, 123.55, 120.27, 41.08, 39.41, 26.19, 26.52, 17.64,
16.04.
Scale-up synthesis of cis-5r tartrate salt
28
ACS Paragon Plus Environment

Page 29 of 39
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
To a solution of tris-(3-aminopropyl)amine (50 mL, 0.25 mol) and triethylamine (60 mL, 0.44
mol) in distillated THF (100 mL) is added dropwise neryl chloride 8 (35 g, 0.20 mol) in
distillated THF (200 mL). The reaction mixture is stirred at room temperature for 24 h. and
evaporated to dryness. The crude residue is purified by column chromatography (eluant
CH2Cl2/MeOH/conc.NH4OH, 7:3:1) to afford the pure desired cis {3-[Bis-(3-amino-propyl)-
amino]-propyl}-(3,7-dimethyl-octa-2,6-dienyl)-amine 5r as a pale yellow oil. 31.1 g (48%).
Then, a solution of cis {3-[Bis-(3-amino-propyl)-amino]-propyl}-(3,7-dimethyl-octa-2,6-
dienyl)-amine 5r (23.1 g, 0.07 mol) and tartaric acid (42.8 g, 0.28 mol) in anhydrous methanol
(200 ml) is stirred at room temperature for 16 h. Methanol is evaporated until residual volume
of about 50 mL is reached. Diethyl ether is added (about 500 mL) and the resulting precipitate
is filtered off to give the pure desired salt as a white solid. 48.4 g (74%). 1H NMR (D2O, 400
MHz) : δ = 5.24-5.20 (m, 1H), 5.10 (m, 1H), 4.50 (s, 6H), 3.66-3.61 (m, 2H), 3.30-3.24 (m,
6H), 3.06-3.02 (m, 6H), 2.10 (m, 9H), 1.75 (s, 3H), 1.63 (s, 3H), 1.56 (s, 3H). 13C NMR (D2O,
75 MHz) : δ = 146.96, 134.12, 123.34, 113.42, 72.75, 49.86, 45.00, 43.04, 36.37, 31.31, 25.58,
24.82, 22.62, 21.49, 16.69. MS (ESI) C19H40N4 m/z 325.4 (100 %, (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
Scale-up synthesis of trans-5r
To a solution of tris-(3-aminopropyl)amine (50 ml, 0.25 mol) and triethylamine (60 mL, 0.44
mol) in distillated THF (100 mL) is added dropwise geranyl chloride 7 (35 g, 0.20 mol) in
distillated THF (200 mL). The reaction mixture is stirred at room temperature for 24 h. and
evaporated to dryness. The crude residue is purified by column chromatography (eluant
CH2Cl2/MeOH/conc.NH4OH, 7:3:1) to afford the pure desired trans {3-[Bis-(3-amino-propyl)-
amino]-propyl}-(3,7-dimethyl-octa-2,6-dienyl)-amine 5r as a pale yellow oil. 35.2 g (54%). 1H
NMR (MeOD, 400 MHz) : δ = 5.30 (t, J = 7.0 Hz, 1H), 5.14 (t, J = 6.9 Hz, 1H), 3.39 (d, J =
7.0 Hz, 2H), 2.85 (t, J = 7.2 Hz, 4H), 2.75 (t, J = 7.4 Hz, 2H), 2.58 (t, J = 6.9 Hz, 4H), 2.56 (t,
29
ACS Paragon Plus Environment