Noncyclam tetraamines inhibit CXC chemokine receptor type 4 and target glioma-initiating cells

Article abstract of DOI:10.1021/jm300862u

The three stereoisomers of the noncyclam compound 1 (1(R,R), 1(S,S), and the meso form 1(S,R)) and their corresponding tetrahydrochlorides 11 were prepared from (S)- and (R)-2-methylpiperidine. We have evaluated their inhibitory activity on the CXC chemokine receptor type 4 (CXCR4), toxicity properties, and assessment of their effect on glioma initiating cells (GICs) in comparison with the prototype compound AMD3100. The IC₅₀ values determined on human recombinant (CHO) cells showed very similar inhibitory activities albeit a lower K_B for AMD3100, with the 1(R,R) isomer being second in potency. All the compounds showed low cardiac toxicity but, contrary to AMD3100, gave maximum nonlethal doses of around 2.0 mg/kg. The CXCR4 inhibitors had an effect on the state of differentiation of GICs, decreasing the percentage of CD44+ cells in glioblastoma multiform neurospheres in vitro. Moreover, these CXCR4 inhibitors blocked the capacity of cells to initiate orthotopic tumors in immunocompromised mice.

Full text of DOI:10.1021/jm300862u

Article
pubs.acs.org/jmc
Noncyclam Tetraamines Inhibit CXC Chemokine Receptor Type 4 and
Target Glioma-Initiating Cells
Laia Ros-Blanco,† Judit Anido,‡ Ramon Bosser,§ Jose Este,∥ Bonaventura Clotet,∥ Ana Kosoy,§
́
́
§
†
,‡
,†
́
Luis Ruíz-Avila, Jordi Teixido, Joan Seoane,* and Jose I. Borrell*
́
́
†Grup d’Enginyeria Molecular, Institut Químic de Sarria,
‡Translational Research Program, Vall d'Hebron Institute of Oncology (VHIO), Vall d’Hebron University Hospital, E-08035
Barcelona, Spain and Universitat Autonoma de Barcelona, E-08193 Cerdanyola del Valles, Spain
§Janus Development SL, Parc Científic Barcelona, Baldiri Reixac 10-12, E-08028 Barcelona, Spain
∥IrsiCaixa, Hospital Universitari Germans Trias i Pujol, Universitat Auton
oma de Barcelona, E-08916 Badalona, Spain
̀
Universitat Ramon Llull, Via Augusta 390, E-08017 Barcelona, Spain
̀
̀
̀
S
* Supporting Information
ABSTRACT: The three stereoisomers of the noncyclam
compound 1 (1(R,R), 1(S,S), and the meso form 1(S,R))
and their corresponding tetrahydrochlorides 11 were prepared
from (S)- and (R)-2-methylpiperidine. We have evaluated their
inhibitory activity on the CXC chemokine receptor type 4
(CXCR4), toxicity properties, and assessment of their effect on
glioma initiating cells (GICs) in comparison with the
prototype compound AMD3100. The IC50 values determined
on human recombinant (CHO) cells showed very similar
inhibitory activities albeit a lower KB for AMD3100, with the
1(R,R) isomer being second in potency. All the compounds
showed low cardiac toxicity but, contrary to AMD3100, gave
maximum nonlethal doses of around 2.0 mg/kg. The CXCR4 inhibitors had an effect on the state of differentiation of GICs,
decreasing the percentage of CD44+ cells in glioblastoma multiform neurospheres in vitro. Moreover, these CXCR4 inhibitors
blocked the capacity of cells to initiate orthotopic tumors in immunocompromised mice.
their ability to generate detached spherical cellular structures
(neurospheres) when cultured in serum-free medium. Several
markers, most of them previously described for neuroprogenitor
cells, have been reported to identify GICs. Specifically, we and
others have shown that a glioma subpopulation of cells
expressing the cell surface protein, CD44, are enriched for
cancer stem cells.4 GICs are considered to be responsible for the
initiation, propagation, and recurrence of tumors. Moreover,
GICs are resistant to standard therapy. This indicates that GICs
are critical therapeutic targets and that more effective therapies
will result from approaches aimed at targeting the stem-cell-like
component of gliomas.5
INTRODUCTION
■
Gliomas have morphologic and gene-expression characteristics
similar to glia, the support cells of the brain, and they include
astrocytomas, oligodendrogliomas, mixed oligoastrocytomas,
and ependimomas according to their histological characteristics.
Gliomas can be divided into four clinical grades on the basis of
their histology and prognosis. Grade IV gliomas (glioblastoma,
GBM) either arise de novo or progress from lower grade to
higher grade over time; the tumor of patients originally
diagnosed with a lower grade will often progress to a GBM.
Histologically, GBMs exhibit microvascular hyperproliferation
and/or necrosis, which normally is surrounded by densely
packed tumor cell nuclei referred to as being pseudopalisades.1
GBMs have a propensity to infiltrate throughout the brain, and
this invasive nature results in the inability of surgery to
completely cure patients. GBMs are highly malignant, are usually
recalcitrant to radio- and chemotherapy, and have a median
survival of 1−2 years.2,3 Novel molecular-targeted therapies
against these devastating tumors are required.
Our work in the field of anti-HIV entry/fusion inhibitors6,7 led
to the identification of novel potent HIV-1 entry CXCR4
coreceptor inhibitors targeting CXCR4 coreceptor without
cytotoxicity at the tested concentrations.8 Among them,
compound 1 (Chart 1) was selected as a lead compound
because it presents nearly the same level of activity as the
reference compounds, 1 (EC50 = 0.019 μM) and the bicyclam
AMD3100 (2, Plerixafor, EC50 = 0.002 μM), and shows no cell
toxicity at the tested concentrations of up to 25 μg mL−1.
Recently, a subpopulation of tumor cells with stem-cell-like
properties has been identified in gliomas. These cells, called
glioma stem cells or glioma-initiating cells (GICs), are
characterized by their self-renewal capacity, their multilineage
differentiation properties, their high oncogenic potential, and
Received: April 19, 2012
Published: August 21, 2012
© 2012 American Chemical Society
7560
dx.doi.org/10.1021/jm300862u | J. Med. Chem. 2012, 55, 7560−7570

Journal of Medicinal Chemistry
Article
Chart 1. Noncyclam CXCR4 Inhibitor 1, AMD3100 (2), and Stereoisomers of Compound 1
Scheme 1. Synthesis of the Three Stereoisomers of Compound 1 (1(R,R), 1(S,S), and the meso Form 1(S,R)) and the
Corresponding Tetrahydrochlorides 11(R,R), 11(S,S), and 11(S,R)
Compound 1 was obtained and tested as a mixture of three
stereoisomers (1(R,R), 1(S,S), and the meso form 1(S,R) in an
approximate 25:25:50 ratio) due to the presence of two chiral
centers introduced from the racemic precursor N-(3-amino-
propyl)-2-pipecoline (3).
CXCR4 is a transmembrane receptor that regulates the
trafficking and homing of various cell types including stem cells
and cancer stem cells. CXCL12 is the ligand of CXCR4 and upon
binding to CXCR4 engages several signal transduction pathways
including the MAPK and the PI3K-Akt pathways. CXCR4
expression in glioma confers poor prognosis after surgery,9 and
CXCR4 blockade in mice delays tumor recurrence after
irradiation by inhibiting the recruitment of CD11b+ mono-
cytes/macrophages that participate in revascularizing the
tumor.10 These results suggest a critical role of CXCR4 in
glioma. In addition, there is much evidence that demonstrates the
involvement of the CXCL12/CXCR4 pathway in different cell
functions critical for glioma development. Among others, the
CXCL12/CXCR4 pathway11 promotes tumor vasculogenesis
and angiogenesis in tumor stem-like glioma cells.12 Similarly, the
pathway increases cell motility and proliferation in the U-118
human glioma cell line.13 In addition, the blockage of CXCR4
7561
dx.doi.org/10.1021/jm300862u | J. Med. Chem. 2012, 55, 7560−7570

Journal of Medicinal Chemistry
Article
induced a significant increase in apoptosis. Taken all together,
these results point to CXCR4 as a promising therapeutic target in
glioma. Because of the described function of CXCR4 in cancer
and stem cells, CXCR4 antagonists could be effective against
cancer-initiating cells and, hence, tumor initiation and relapse.
We systematically isolate cancer-initiating cells from patient-
derived glioma samples, glioma initiating cells (GICs) that we
culture and study. GICs grow as neurospheres in culture and can
be expanded by serial passaging. In addition, patient-derived
GICs are inoculated in the brain of NOD-SCID mice. Tumors
arising from GICs maintain the same characteristics as the
original human tumor.14
Here we report the synthesis of the three stereoisomers of
compound 1 (1(R,R), 1(S,S), and the meso form 1(S,R)), the
evaluation of their compared inhibitory activity on the CXCR4
coreceptor, the evaluation of their toxicity properties, and the
assessment of their effect on GICs and glioma initiation in
comparison with AMD3100.
NaBH4, and stirring for 24 h at room temperature, afforded
1(S,S) and 1(R,R) as yellow oils in 97% and 83% yields,
respectively (Scheme 1).
In the case of the meso form 1(S,R), being an asymmetrically
substituted compound, it was necessary to use the monop-
rotected terephthaldehyde 8 as previously described by our
group.8 Although the reductive amination could be initiated
either by using 3(S) or 3(R) at first, we preferred to start with
3(S) since it was more readily available because it proceeds from
the more commercially accessible (S)-2-methylpiperidine
(4(S)). Thus, the reductive amination of 8 with 1 equiv of
3(S) in the same reaction conditions described above for the
synthesis of the symmetrical compounds afforded 9(S) in 84%
yield. 9(S) was subsequently hydrolyzed to the aldehyde 10(S)
with 2 M HCl in 80% yield, which was treated with 1 equiv of
3(R) in the same reductive amination conditions to afford the
meso form 1(S,R) in 83% yield (56% yield from 8) (Scheme 1).
The resulting stereoisomers 1(R,R), 1(S,S), and the meso form
1(S,R) were converted into the corresponding tetrahydro-
chlorides 11(R,R), 11(S,S), and 11(S,R) upon treatment with
HCl in MeOH in almost quantitative yields. Finally, we obtained
1 and its tetrahydrochloride 11 as the mixture of the three
stereoisomers (1(R,R), 1(S,S), and the meso form 1(S,R)) for
comparison purposes by following the procedure previously
described by our group.8
RESULTS AND DISCUSSION
■
Synthesis. To obtain the three stereoisomers of compound 1
(1(R,R), 1(S,S), and the meso form 1(S,R)), it was first necessary
to prepare the homochiral amines 3(S) and 3(R), because the
diamine 3 is only commercially available as the racemic mixture.
The Gabriel protocol was selected for the synthesis of 3(S) and
3(R) starting from the corresponding enantiomeric 2-methyl-
piperidines 4(S) and 4(R) (Scheme 1). Although the resolution
of 2-methylpiperidine (4) has been described,15 finally we used
commercially available (S)-2-methylpiperidine (4(S)) and (R)-
2-methylpiperidine (4(R)) with purities of 97%, the latter one as
the HCl salt.
Antiviral Activity. As a first way to compare the biological
activity in front of the CXCR4 receptor of compounds 11(R,R),
11(S,S), and 11(S,R) and establish possible differences that can
favor one candidate over the others, we decided to evaluate the
anti-HIV activity (EC50) and cytotoxicity (CC50) in CXCR4+
MT-4 cells.
The results shown in Table 1 demonstrate that there are no
significant differences between the three isomers, at least in this
As for the nucleophilic substitution of the bromine atom
present in the commercially available N-(3-bromopropyl)-
phthalimide (5), we initially considered the same reaction
conditions we have used for the synthesis of similar diamines: an
excess of the corresponding cyclic amine in acetone at reflux for
12 h.16 However, in the case of 4(S) and 4(R), we considered a
better approach to use stoichiometric amounts of the chiral
amines to minimize their amount and avoid intermediate
purifications. A bibliographic search revealed a modification of
this protocol described by Jo et al. in which the use of solid
K2CO3 to capture HBr formed allows one to use a 1.2:1 ratio
between the chiral amine 4 and the bromo-substituted
phthalimide 5 (or even use stoichiometric amounts).17
Furthermore, the use of acetonitrile as solvent allows reducing
by half the reaction time (although we preferred to maintain the
reflux for 8 h to ensure the total conversion of the reagents). In
such modification the hydrolysis of the alkylated intermediates
6(S) and 6(R) was performed under more mild conditions using
hydrazine in EtOH instead of 6 M aqueous HCl.
Using such methodology, the 2-methylpiperidines 4(S) and
4(R) were converted to the alkylated intermediates 6(S) and
6(R) in 92% and 97% yields, respectively, which were hydrolyzed
to the desired homochiral N-(3-aminopropyl)-2-pipecolines
3(S) and 3(R) in 79% and 73% yields, respectively (total yields
greater than 60%).
The synthesis of the stereoisomers 1(S,S) and 1(R,R) was
carried out following the methodology previously described by
our group for 1.8 Thus, the treatment of terephthaldehyde (7)
with 2 equiv of the corresponding amine 3(S) and 3(R) in
anhydrous MeOH at reflux for 24 h in the presence of anhydrous
Na2SO4 as the dehydrating agent, followed by filtration and
cooling of the resultant filtrate to 0 °C, addition of 2 equiv of
Table 1. Anti-HIV Activity (EC50) and Cytotoxicity (CC50)
Measurements in MT-4 Cells of Tetrachlorohydrates 11,
11(R,R), 11(S,S), and 11(S,R)
compd
EC50
CC50
11
0.130 0.04 μM
0.125 0.04 μM
0.137 0.06 μM
0.139 0.03 μM
>44 μM
11(R,R)
11(S,S)
11(S,R)
>44 μM
>44 μM
26 10 μM
test. Consequently, we decided to work in parallel with 11(R,R),
11(S,S), and 11(S,R) in the following experiments.
ADME-TOX Studies. Since AMD3100 reported cardio
toxicity, we were interested in comparing our products with
AMD3100 in terms of cardio toxicity and affinity for the CXCR4
receptor. Therefore, both an in vitro CXCR4 antagonist effect
assay and a conventional K+ Channel patch-clamp assay were
performed for 11(R,R), 11(S,S), and 11(S,R) and AMD3100.
The antagonist effects of AMD3100, 11(R,R), 11(S,S), and
11(S,R) on the CXCR4 receptor were determined using human
recombinant (CHO) cells following the protocol described by
Zhou et al.18 The results obtained are shown in Figure 1. The
IC50 values obtained for AMD3100, 11(R,R), 11(S,S), and
11(S,R) indicate that these compounds present very similar
inhibitory activities although the KB (Dissociation constant) is
lower for AMD3100, meaning this compound binding with the
CXCR4 receptor is stronger than the others tested. The R,R
isomer of 1 was second in terms of potency. For the cardiac
toxicity determination, a conventional electrophysiological assay
7562
dx.doi.org/10.1021/jm300862u | J. Med. Chem. 2012, 55, 7560−7570

Journal of Medicinal Chemistry
Article
Figure 1. Antagonist effect and its related inhibition determination curves for 11(R,R).
using whole cell patch-clamp technique proposed by Mathes19 to
study effects of the compounds on the hERG channel, was used.
Briefly, CHO-K1 cells stably expressing hERG channel were used
and compounds were valuated at 5 concentrations to determine
the amplitude of hERG potassium channel tail currents. The
percentage inhibition of the tail current by each compound is
summarized in Table 2, showing that all of the compounds had
low cardiac toxicity.
administered with 11(R,R) at 2.5 mg/kg died immediately or
within 1 min of administration.
On the basis of the results obtained, the maximum nonlethal
dose and the minimum lethal dose of the test items when
administered intravenously to Hsd:ICR (CD-1) mice for both
sexes as well as for males and females separately were as follows:
AMD3100 (as no mortality was recorded, it was considered that
the minimum lethal dose is >4.0 mg/kg); 11(S,S) (both sexes:
maximum nonlethal dose, 1.5 mg/kg; minimum lethal dose, 2.0
mg/kg); 1(R,R) (both sexes: maximum nonlethal dose, 2.0 mg/
kg; minimum lethal dose, 2.5 mg/kg); 11(S,R) (both sexes:
maximum nonlethal dose, 2.5 mg/kg; minimum lethal dose, 3.0
mg/kg). The results obtained once more indicate that there are
not significant differences between the tetrahydrochlorides
11(R,R), 11(S,S), and 11(S,R) but pointed to a higher toxicity
in vivo of these compounds with respect to AMD3100.
Effect of CXCR4 Inhibitors on GICs and Glioma
Initiation. As a first step, to address the effect of CXCR4
inhibitors on patient-derived neurospheres, we decided to assess
how the compounds affected the signaling pathways that are
downstream of the CXCR4 receptor (the Akt and MAPK
pathways). First of all, we decided to determine whether the
ligand of the CXCR4 receptor, CXCL12, could induce the Akt
and MAPK pathways in our cells. Treatment of neurospheres
from three different patients (GBM1, GBM2, and GBM3) with
CXCL12 slightly increased phospho-Akt (p-Akt) in GBM1 and
GBM2 but not in GBM3. There was no major effect of CXCL12
treatment on phospho-MAPK (p-MAPK) (Figure 2). Two
bands were detected in the p-MAPK blot corresponding to p44
and p42 MAPK. Both kinases share many functional roles;
however, some specific functions have been assigned to each of
them.20
Table 2. Effect in the in Vitro Human CXCR4 Receptor
Functional of Compounds AMD3100, 11(R,R), 11(S,S), and
11(S,R)
compd
IC50
KB
IC50 (hERG)
AMD3100
11(R,R)
11(S,S)
0.26 μM
0.35 μM
1.1 μM
0.077 μM
0.11 μM
0.32 μM
0.24 μM
>100 μM
>100 μM
a
N.C.
11(S,R)
0.79 μM
>100 μM
a
N.C. = IC50 value not calculable. Concentration−response curve
shows less than 25% effect at the highest tested concentration.
To complement such data, we decided to study the acute
intravenous toxicity in mice by determining the maximum
nonlethal dose and the minimum lethal dose. The acute toxicities
of the test items AMD3100, 11(R,R), 11(S,S), and 11(S,R) were
evaluated after their intravenous administration to Hsd:ICR
(CD-1) mice, followed by an observation period of 14 days. In
the study, 4 groups consisting of 4 animals (2 males and 2
females) each were administered in physiological saline solution
at the doses between 1.0 and 4.0 mg/kg.
Viability/mortality and clinical signs were recorded daily
during the acclimatization and during the first 30 min and at
approximately 1, 2, 3, and 5 h after administration on test day 1.
Viability/mortality was observed twice daily, and clinical signs
were observed once daily during days 2−15. Body weights were
recorded once during the acclimatization, on days 1 and 14, and
before sacrifice (day 15). At the end of the observation period, all
surviving animals and those that were found dead were
necropsied, and a macroscopic examination was performed on
the principal organ systems and tissues to record any
macroscopic alterations related to the treatment.
We next decided to compare the effect of the CXCR4
inhibitors on the Akt and MAPK pathways using two different
concentrations of compound (1 μM and 100 nM). To perform
this experiment we decided to focus our studies on GBM2, the
patient-derived neurospheres where CXCL12 was having some
effect on p-Akt (see Figure 2). 11(S,R) (ASR226) (at 1 μM) was
the only compound that had a slight effect on p-Akt and,
interestingly, in total Akt. No major effect was observed on p-
MAPK in any of the conditions (Figure 3).
We repeated the same experiment treating the cells with the
inhibitory compounds (at 1 μM) in the presence or absence of
CXCL12. In this experiment, the effect of CXCL12 was not
evident. However, and again, a minor effect of 11(S,R) on p-Akt
No mortality was recorded in animals administered with
AMD3100 at any dose level (1, 2, 3, 3.5, and 4 mg/kg). All
animals administered with 11(S,R) at 3.0 mg/kg, one male
administered with 11(S,S) at 2.0 mg/kg, and one female
7563
dx.doi.org/10.1021/jm300862u | J. Med. Chem. 2012, 55, 7560−7570

Journal of Medicinal Chemistry
Article
Figure 2. MAPK and Akt pathway activation by CXCL12. Neuro-
spheres were dissociated and counted, and an equal number of cells were
seeded. After 24 h of EGF, FGF, and B27 starvation, cells were treated
with 100 mg/mL of CXCL12 during 1 or 2 h. Levels of p-MAPK, p-Akt,
and Tubulin were determined by immunoblotting.
Figure 4. MAPK and Akt pathway inhibition by CXCR4 inhibitors in
the presence of CXCL12. GBM2 neurospheres were dissociated and
counted. After 24 h of EGF, FGF and B27 starvation cells were treated or
left untreated with 1 μM CXCR4 inhibitors AMD3100, 11, 11(R,R),
11(S,S), and 11(S,R) during 8 h and/or CXCL12 (100 ng/mL) during 1
h. Levels of p-MAPK, MAPK, p-Akt, Akt, and Tubulin were determined
by immunoblotting.
differentiation, we analyzed the levels of CD44, a well-described
marker of cancer-initiating cells. We treated GBM neurospheres
obtained from two GBM patients with the CXCR4 inhibitors
(AMD3100, 11, 11(R,R), 11(S,S), and 11(S,R)) and analyzed
the levels of CD44 by flow cytometry. The percentage of cells
that did not express CD44 (CD44 null) increased in GBM
neurospheres from two different patients treated with the
CXCR4 inhibitors (Figure 5 and Supporting Information). As a
consequence, the ratio CD44 low/CD44 null decreased in the
presence of the CXCR4 inhibitors (Figure 5, bar graphs),
indicating that the CXCR4 inhibitors decreased the amount of
glioma-initiating cells, most likely inducing cell differentiation.
To confirm the effect of CXCR4 inhibition on glioma-
initiating cells, we decided to perform an in vivo experiment. For
this purpose, we used a glioma model that reproduces the tumor
from the patient in mice described in our previous publication by
Figure 3. MAPK and Akt pathway inhibition by CXCR4 inhibitors
AMD3100, 11(R,R), 11(S,S), and 11(S,R). GBM2 neurospheres were
dissociated, counted and an equal number of cells were seeded. Cells
were treated or left untreated with CXCR4 inhibitors (1 μM) during 8 h.
Levels of MAPK, p-Akt, Akt and Tubulin were determined by
immunoblotting.
and total Akt was observed. Moreover, a slight decrease of p-
MAPK was observed in cells treated with 11(S,S) (Figure 4).
In summary, our data indicated that 1 μM 11(S,R) had a minor
effect on the Akt pathway whereas the rest of the compounds
including the commercially available AMD3100 did not have a
major effect on the Akt and MAPK pathways. At this point, we
decided to assess the effect of the compounds on cell viability and
proliferation. To this aim, we treated neurospheres from three
patients with the compounds in the presence or absence of
CXCL12. After 7 days of treatment, cells were counted. The
number of cells after treatment is the result of a balance between
the rate of proliferation and the rate of cell death. No major effect
was observed in the conditions assayed; all drugs gave >80%
viability with the exception of AMD3100, which gave ∼65% with
GBM1, indicating that the compounds do not have a major effect
on cell proliferation nor viability.
Penuelas et al. (Figure 6).14
̃
We treated GBM neurospheres in culture with AMD3100 and
11, the tetrahydrochloride of the mixture of the three
stereoisomers (1(R,R), 1(S,S), and the meso form 1(S,R)), for
7 days. Then we inoculated in the brain of NOD-SCID mice
using a stereotactic apparatus the same number of live cells, as
analyzed by trypan blue staining, which were pretreated with the
indicated compounds. Forty days after inoculation, we
monitored tumor growth through magnetic resonance imaging
(MRI). Images from the entire mouse brain were obtained, and
tumor volume was quantified (Figure 7). Tumors generated from
the cells pretreated with the CXCR4 inhibitors generated smaller
and fewer tumors than control cells, confirming that the
inhibition of CXCR4 prevents tumor initiation through the
decrease of the number of glioma-initiating cells.
We next decided to assess the effect of CXCR4 inhibitors on
the state of differentiation of GICs. To determine the state of
7564
dx.doi.org/10.1021/jm300862u | J. Med. Chem. 2012, 55, 7560−7570

Journal of Medicinal Chemistry
Article
Figure 5. Effect of CXCR4 inhibitors on the CD44+ cell population in patient-derived GBM neurospheres. GBM neurosphere cultures were dissociated
and counted, and an equal number of cells were seeded. Cells were treated with 1 μM of the indicated inhibitors (AMD3100, 11, 11(R,R), 11(S,S), and
11(S,R)) for 7 days. The percentage of cells expressing CD44 was determined by flow cytometry. Bar graphs show the ratio of CD44 low/CD44 null
cells quantified by flow cytometry in the different conditions.
Figure 6. Scheme of mouse model for human glioblastoma.
Our results show that CXCR4 inhibition did not have a major
effect on the PI3K-Akt and the MAPK pathways in contrast to
those by Rubin et al.21 However, both results may not be
comparable because Rubin et al. used U87 and Dacoy, two
regular cell lines, whereas we used fresh patient-derived cells. The
models are different because, first, established cell lines have been
cultured in artificial conditions for a long time diverging from the
characteristics of a real tumor and, second, our cells are
undifferentiated and enriched in GICs.
CONCLUSION
■
We obtained the three stereoisomers of compound 1, 1(R,R),
1(S,S), and the meso form 1(S,R), starting from the commercially
available (S)-2-methylpiperidine (4(S)) and (R)-2-methylpiper-
idine (4(R)) in good overall yields by following our general
methodology for the preparation of such tetraamines.8 The anti-
HIV activity (EC50) and cytotoxicity (CC50) in MT-4 cells of the
corresponding tetrahydrochlorides, 11(R,R), 11(S,S), and
11(S,R), were evaluated in comparison to compounds 11 (the
7565
dx.doi.org/10.1021/jm300862u | J. Med. Chem. 2012, 55, 7560−7570

Journal of Medicinal Chemistry
Article
Figure 7. Effect of CXCR4 inhibitors AMD3100 and 11 on tumor growth. GBM1 neurospheres were dissociated and counted, and the same number of
living cells was inoculated in the brain of NOD-SCID. Forty days after inoculation, the tumor size was determined through MRI and the tumor volume
was quantified. The white box indicates an example of tumor detected by MRI.
tetrahydrochloride of the mixture of the three stereoisomers)
and AMD3100. There were no significant differences between
the three isomers. Furthermore, the IC50 values of AMD3100,
11(R,R), 11(S,S), and 11(S,R) were determined on human
recombinant (CHO) cells showing that these compounds
presented very similar inhibitory activities, although the KB was
lower for AMD3100, with the 11(R,R) isomer being second in
potency.
Toxicity studies for AMD3100, 11(R,R), 11(S,S), and 11(S,R)
were carried out to assess the safety of such compounds for
possible use in humans. All compounds tested induced low
cardiac toxicity. The evaluation of the acute intravenous toxicity
in mice showed that no mortality was recorded in animals
administered with AMD3100 at any dose level, but compounds
11(R,R), 11(S,S), and 11(S,R) gave a maximum nonlethal dose
of ∼2.0 mg/kg.
We found that these CXCR4 inhibitors affect the state of
differentiation of GICs, decreasing the percentage of CD44+
cells in GBM neurospheres in vitro. Moreover, and expected
based on the effect on the CD44+ cells, CXCR4 inhibitors
prevent the capacity of cells to initiate orthotopic tumors in
immunocompromised mice. Our results indicate that the
CXCR4 inhibitors tested could be effective against GICs,
preventing tumor initiation and recurrence and most likely
metastasis in other tumor types.
EXPERIMENTAL SECTION
■
Chemistry. Commercial reagents (Fluka, Aldrich) were used
without purification. The (S)-2-methylpiperidine (4(S)) and the (R)-
2-methylpiperidine (4(R)), this later as hydrochloride, were purchased
from Aldrich (Cat. no. 522902) and Astatech USA (Cat. no. 64253),
respectively. IR spectra were recorded in a Nicolet Magna 560 Fourier
transform infrared (FTIR) spectrophotometer. Wavenumbers (ν) are
expressed in cm−1. 1H and 13C NMR spectra were recorded in a Varian
Gemini 300HC instrument (operating at a field strength of 300 and 75.5
MHz, respectively) or a Varian Gemini 400-MR instrument (1H NMR
400 MHz and 13C NMR 100.6 MHz). Chemical shifts are reported in
parts per million (δ-scale), and coupling constants (J) are reported in Hz
by using, in the case of 1H NMR spectroscopy, tetramethylsilane (TMS)
or sodium 2,2,3,3-tetradeutero-3-(trimethylsilyl)propionate (TSPNa)
as an internal standard and, in the case of 13C NMR spectra, solvent
residual peak was taken as reference: CDCl3 at 77.0 ppm, d6-DMSO
(dimethyl sulfoxide) at 39.5 ppm, and CD3OD in D2O at 49.5 ppm.
Standard and peak multiplicities are designated as follows: s (singlet), d
(doublet), dd (doublet of doublets), t (triplet), q (quartet), qn
(quintet), m (multiplet), br (broad signal). MS data (m/z (%), EI, 70
eV) were obtained by using an Agilent Technologies 5975 spectrometer
and a Hewlett-Packard HP5988A quadrupole mass spectrometer
operating in electronic ionization (EI) mode at 70 eV and at 4 kV
accelerating potential, or a Bruker Biotoff II spectrometer operating in
electrospray ionization (ESI) mode with a time of flight (TOF) detector.
High-resolution mass spectrometry (HRMS) data were obtained by
using a VG AutoSpec (Micromass Instruments) Trisector EBE high-
resolution spectrometer (EI mode), a Bruker Biotof II mass
spectrometer (ESI-TOF mode), or a Bruker Autoflex spectrometer
(MALDI-TOF mode, HCCA matrix). Elemental microanalyses were
7566
dx.doi.org/10.1021/jm300862u | J. Med. Chem. 2012, 55, 7560−7570

Journal of Medicinal Chemistry
Article
obtained on a Carlo−Erba CHNS-O/EA 1108 analyzer. Specific optical
rotations ([α]D) were measured with a Perkin-Elmer 241 polarimeter.
Sodium D line (589 nm) and a path length of 1 dm were used for
measuring at room temperature. The purity of each tested compound
(>95%) was determined on an Agilent 1200 LC instrument using a
XTerra RP18 column (125 Å, 5 μm, 3 mm × 250 mm, with a flow rate of
0.75 mL/min and detection at 220 nm) using a 95:5 mixture of
(NaH2PO4, pH = 7 + 0.1% acetonitrile)/(acetonitrile/MeOH 75:25) as
eluent.
mmol) of hydrazine hydrate to give 0.30 g (1.9 mmol, 73%) of 3(R) as a
yellow oil. IR (film) ν (cm−1): 3273 (N−H), 2930, 2855, 2789, 1576,
1472, 1374, 1329. 1H NMR (400 MHz, CDCl3) δ (ppm): 2.86 (dt, 2J =
11.5 Hz, 3J = 4.0 Hz, 1H, H−C3), 2.73 (m, 3H, H−C1, H−C1a), 2.35
(m, 1H, H−C3), 2.26 (m, 1H, H−C5a), 2.11 (dt, 2J = 11.0 Hz, 3J = 3.0
Hz, 1H, H−C1a), 1.67−1.51 (m, 8H, H−C2, H−C2a, H−C3a, H−C4a,
3
NH2), 1.27 (m, 2H, H−C3a, H−C4a), 1.06 (d, J = 6.0 Hz, 3H, H−
C6a). 13C NMR (100.6 MHz, CDCl3) δ (ppm): 55.9 (C5a), 52.1 (C3),
51.8 (C1a), 41.0 (C1), 34.7 (C4a), 29.5 (C2), 26.2 (C2a), 24.0 (C3a),
19.1 (C6a). MS (EI) m/z (%): 156.1 (2) [M]+, 112.2 (100), 98.1 (68).
HRMS (EI) calculated for C9H20N2 [M]+: 156.1626; found: 156.1624.
(S)-N,N′-(1,4-Phenylenebis(methylene))bis(3-((S)-2-methyl-
piperidin-1-yl)propan-1-amine) Tetrahydrochloride (11(S,S)).
Terephthalaldehyde (7) (0.10 g (0.71 mmol)), 0.23 g (1.45 mmol) of
(S)-N-(3-aminopropyl)-2-pipecoline (3(S)) and Na2SO4 were mixed in
6 mL of anhydrous MeOH, and the mixture was heated at reflux under
N2 atmosphere for 24 h. The solid was filtered, and the intermediate
imine in MeOH was cooled to 0 °C and treated with 0.06 g (1.45 mmol)
of solid NaBH4. The reaction mixture was stirred at room temperature
overnight. Then water was added, and the product was extracted with
CH2Cl2. The organic layers were combined, washed with brine, dried
over anhydrous MgSO4, filtered, and concentrated to give 0.29 g (0.67
mmol, 97%) of 1(S,S) as a yellow oil. IR (film) ν (cm−1): 3282 (N−H),
2929, 2854, 2793, 1449, 1372. 1H NMR (400 MHz, CDCl3): δ (ppm)
7.30 (m, 4H, H−C2b), 3.79 (d, 4H, H−Cα), 2.92 (m, 2H, H−C1a),
2.79 (m, 2H, H−C3), 2.68 (m, 4H, H−C1), 2.44−2.33 (m, 8H, H−C3,
H−C5a, H−C1a), 2.16 (m, 2H, NH), 1.80−1.48 (m, 12H, H−C2, H−
C2a, H−C4a, H−C3a), 1.29 (m, 4H, H−C4a, H−C3a), 1.07 (d, 6H, 3J
= 8.0 Hz, H−C6a). 13C NMR (100.6 MHz, CDCl3) δ (ppm): 138.7
(C1b), 127.8 (C2b), 55.8 (C5a), 53.6 (Cα), 52.2 (C3), 52.1 (C1a), 48.2
(C1), 34.7 (C4a), 26.2 (C2a), 25.7 (C2), 23.9 (C3a), 19.1 (C6a). MS
(CI) m/z (%): 415.3 (100), 258.2 (42), 112.0 (51). HRMS (CI)
calculated for C26H47N4 [M+1]+: 415.3801; found: 415.3811. 1(S,S)
was converted to the corresponding tetrahydrochloride 11(S,S) in
almost quantitative yield upon treatment with methanolic HCl in excess
followed by concentration of the resulting solution. [α]D = +15.3 (c =
(S)-N-(3-(2-Pipecolin-1-yl)propyl)phthalimide (6(S)). A solu-
tion of 0.41 mL (0.34 g, 3.3 mmol) of (S)-(+)-2-methylpiperidine
(4(S)) in 10 mL of acetonitrile was treated with 0.82 g (3.0 mmol) of N-
(3-bromopropyl)phthalimide (5) and 0.70 g (5.0 mmol) anhydrous
K2CO3 for 8 h at reflux. After concentration, the residue was diluted with
30 mL CH2Cl2 and washed with water. The organic layers were
combined, dried over anhydrous MgSO4, filtered, and concentrated to
give 6(S) as a yellow oil (0.79 g; 92%). IR (film) ν (cm−1): 2931, 2854,
1
2790, 1772, 1712 (CO), 1396, 1032 (C−N), 720. H NMR (400
MHz, CDCl3) δ (ppm): 7.84 (dd, 3J = 5.5 Hz, 4J = 3.0 Hz, 2H, H−C6),
7.71 (dd, 3J = 5.5 Hz, 4J = 3.0 Hz, 2H, H−C7), 3.70 (m, 2H, H−C1),
2.78 (m, 2H, H−C3, H−C1a), 2.38 (m, 1H, H−C3), 2.24 (m, 1H, H−
C5a), 2.09 (m, 1H, H−C1a), 1.85 (qn, 3J = 7.5, 2H, H−C2), 1.62−1.42
(m, 4H, H−C2a, H−C3a, H−C4a), 1.24 (m, 2H, H−C3a, H−C4a),
1.03 (d, J = 6.5 Hz, 3H, H−C6a). 13C NMR (100.6 MHz, CDCl3) δ
(ppm): 168.3 (C4), 133.8 (C7), 132.2 (C5), 123.1 (C6), 55.8 (C5a),
51.8 (C3), 51.4 (C1a), 36.7 (C1), 34.5 (C4a), 26.1 (C2), 24.6 (C2a),
23.8 (C3a), 18.7 (C6a). MS (ESI-TOF) m/z (%): 287.2 (100) [M++H],
214.1 (2), 158.0 (5), 141.0 (4), 105.0 (6). HRMS (ESI-TOF) calculated
for C17H23N2O2 [M+1]+: 287.1754; found: 287.1750. Anal. calculated
for C17H22N2O2: C 71.30, H 7.74, N 9.78; found: C 70.95, H 8.03, N
9.78.
(R)-N-(3-(2-Pipecolin-1-yl)propyl)phthalimide (6(R)). A solu-
tion of (R)-(+)-2-methylpiperidine hydrochloride (4(R)·HCl) in 10%
NaOH was extracted with CH2Cl2 to obtain the free amine 4(R). A
solution of 0.83 g (8.3 mmol) of (R)-(+)-2-methylpiperidine (4(R)) in
20 mL acetonitrile was treated with 1.65 g (6.0 mmol) N-(3-
bromopropyl)phthalimide (5) and 1.50 g (11 mmol) of anhydrous
K2CO3 for 8 h at reflux. After concentration, the residue was diluted with
30 mL CH2Cl2 and washed with water. The organic layers were
combined, dried over anhydrous MgSO4, filtered, and concentrated to
give 6(R) as a yellow oil (1.67 g; 97%). IR (film) ν (cm−1): 2930, 2856,
1
1.03, MeOH). IR (KBr) ν (cm−1): 3412, 2942, 2745, 1449, 553. H
NMR (400 MHz, D2O): δ (ppm) 7.58 (s, 4H, H−C2b), 4.32 (s, 4H, H−
Cα), 3.73−3.51 (m, 2H, H−C1a), 3.41−2.98 (m, 12H, H−C1, H−C3,
H−C5a, H−C1a), 2.26−1.52 (m, 16H, H−C2, H−C2a, H−C3a, H−
C4a), 1.34 (d, 6H, 3J = 4.0 Hz, H−C6a). 13C NMR (100.6 MHz, D2O) δ
(ppm): 132.6 (C1b), 131.3 (C2b), 60.7 (C5a), 53.0 (Cα), 51.4 (C3),
50.0 (C1a), 44.8 (C1), 32.1 (C4a), 23.6 (C2a), 22.0 (C2), 20.7 (C3a),
17.8 (C6a). Anal. calculated for C26H50N4Cl4·0.5H2O: C 54.83, H 9.03,
N 9.84; found: C 54.50, H 9.37, N 9.51.
1
2791, 1772, 1713 (CO), 1378, 1031 (C−N), 720. H NMR (400
MHz, CDCl3) δ (ppm): 7.84 (m, 2H, H−C6), 7.71 (m, 2H, H−C7),
3.70 (m, 2H, H−C1), 2.77 (m, 2H, H−C1a, H−C3), 2.38 (m, 1H, H−
C3), 2.26 (m, 1H, H−C5a), 2.10 (dt, 2J = 10.0 Hz, 3J = 4.0 Hz, 1H, H−
C1a), 1.85 (qn, 3J = 7.5, 2H, HC2), 1.64−1.43 (m, 4H, H−C2a, H−C3a,
HC4a), 1.24 (m, 2H, H−C3a, H−C4a), 1.03 (d, J = 6.5 Hz, 3H, H−
C6a). 13C NMR (100.6 MHz, CDCl3) δ (ppm): 168.3 (C4), 133.8
(C7), 132.2 (C5), 123.1 (C6), 55.8 (C5a), 51.8 (C3), 51.4 (C1a), 36.7
(C1), 34.5 (C4a), 26.1 (C2), 24.6 (C2a), 23.8 (C3a), 18.7 (C6a).
(S)-N-(3-Aminopropyl)-2-pipecoline (3(S)). A solution of 0.65 g
(2.3 mmol) of (S)-N-(3-(2-pipecolin-1-yl)propyl)phthalimide (6(S))
in 10 mL of ethanol was treated with 0.43 mL (8.8 mmol) of hydrazine
hydrate under reflux for 1 h. The white precipitate of phthalhydrazide
was filtered, and the filtrate was concentrated. After diluting the residue
with 5 mL of AcOEt, a new precipitate of phthalhydrazide appeared,
which was filtered off. The filtrate was concentrated to dryness to give
0.28 g (1.81 mmol, 79%) of 3(S) as a pale yellow oil. IR (film) ν (cm−1):
3293 (N−H), 2930, 2855, 2788, 1575, 1470, 1374, 1329. 1H NMR (400
MHz, CDCl3) δ (ppm): 2.87 (m, 1H, H−C3), 2.72 (m, 3H, H−C1, H−
C1a), 2.35 (m, 1H, H−C3), 2.26 (m, 1H, H−C5a), 2.12 (m, 1H, H−
C1a), 1.80−1.50 (m, 8H, H−C2, H−C2a, H−C3a, H−C4a, NH2), 1.28
(m, 2H, H−C3a, H−C4a), 1.06 (d, 3J = 6.0 Hz, 3H, H−C6a). 13C NMR
(100.6 MHz, CDCl3) δ (ppm): 55.9 (C5a), 52.0 (C3), 51.7 (C1a), 41.0
(C1), 34.6 (C4a), 29.4 (C2), 26.1 (C2a), 23.9 (C3a), 19.0 (C6a). MS
(EI) m/z (%): 156.2 (7) [M]+, 112.2 (100), 98.2 (100). HRMS (EI)
calculated for C9H20N2 [M]+: 156.1626; found: 156.1621.
(R)-N,N′-(1,4-Phenylenebis(methylene))bis(3-((R)-2-methyl-
piperidin-1-yl)propan-1-amine) Tetrahydrochloride (11(R,R)).
As above for 1(S,S) but using 0.07 g (0.48 mmol) of terephthalaldehyde
(7), 0.15 g (0.96 mmol) of (R)-N-(3-aminopropyl)-2-pipecoline
(3(R)), and 0.04 g (0.96 mmol) of solid NaBH4 to give 0.17 g (0.4
mmol, 83%) of 1(R,R) as a yellow oil. 1H NMR (400 MHz, CDCl3): δ
(ppm) 7.31 (s, 4H, H−C2b), 3.79 (d, 4H, H−Cα), 2.91 (m, 2H, H−
C1a), 2.78 (m, 2H, H−C3), 2.68 (m, 4H, H−C1), 2.43−2.32 (m, 8H,
H−C3, H−C5a, H−C1a), 2.16 (m, 2H, NH), 1.79−1.49 (m, 12H, H−
C2, H−C2a, H−C3a, H−C4a), 1.28 (m, 4H, H−C4a, H−C3a), 1.07 (d,
6H, 3J = 8.0 Hz, H−C6a). MS (CI) m/z (%): 415.3 (100), 258.1 (31),
112.0 (53). HRMS (CI): calculated for C26H47N4 [M+1]+: 415.3801;
found: 415.3800. 1(R,R) was converted to the corresponding
tetrahydrochloride 11(R,R) in almost quantitative yield upon treatment
with methanolic HCl in excess followed by concentration of the
resulting solution. [α]D = −16.3 (c = 1.01, MeOH). IR (KBr) ν (cm−1):
3423, 2943, 2743, 1451, 553. 1H NMR (400 MHz, D2O): δ (ppm) 7.59
(s, 4H, H−C2b), 4.33 (s, 4H, H−Cα), 3.73−3.52 (m, 2H, H−C1a),
3.42−3.00 (m, 12H, H−C1, H−C3, H−C5a, H−C1a), 2.26−1.52 (m,
16H, H−C2, H−C2a, H−C3a, H−C4a), 1.37 (d, 6H, 3J = 8.0 Hz, H−
C6a). 13C NMR (100.6 MHz, D2O) δ (ppm): 132.6 (C1b), 131.3
(C2b), 60.7 (C5a), 53.0 (Cα), 51.4 (C3), 50.0 (C1a), 44.8 (C1), 32.1
(C4a), 23.6 (C2a), 22.0 (C2), 20.7 (C3a), 17.8 (C6a). Anal. calculated
for C26H50N4Cl4·0.5H2O: C 54.83, H 9.03, N 9.84; found: C 54.55, H
9.28, N 9.51.
(R)-N-(3-Aminopropyl)-2-pipecoline (3(R)). As described above
for 3(S) but using 0.74 g (2.6 mmol) of (R)-N-(3-(2-pipecolin-1-
yl)propyl)phthalimide (6(R)) in 10 mL of ethanol and 0.50 mL (10
7567
dx.doi.org/10.1021/jm300862u | J. Med. Chem. 2012, 55, 7560−7570

Journal of Medicinal Chemistry
Article
(S)-N-(4-(Diethoxymethyl)benzyl)-3-(2-methylpiperidin-1-
yl)propan-1-amine (9(S)). 4-(Diethoxymethyl)benzaldehyde (8)
(0.35 g (1.6 mmol)), 0.25 g (1.6 mmol) of (S)-N-(3-aminopropyl)-2-
pipecoline (3(S)), and Na2SO4 were mixed in 5 mL of anhydrous
MeOH, and the mixture was held at reflux under N2 atmosphere for 24
h. The solid was filtered and the intermediate imine in MeOH was
cooled to 0 °C and treated with 0.056 g (1.45 mmol) of solid NaBH4.
The reaction mixture was stirred at room temperature overnight. Then
water was added, and the product was extracted with CH2Cl2. The
organic layers were combined, washed with brine, dried over anhydrous
MgSO4, filtered, and concentrated to give 0.47 g (1.3 mmol, 84%) of the
9(S) as a brown oil. IR (film) ν (cm−1): 3283 (N−H), 2973, 2930, 2798,
1663 (CC), 1444, 1371, 1114, 1055. 1H NMR (400 MHz, CDCl3): δ
(ppm) 7.44 (m, 2H, H−C3b), 7.34 (m, 2H, H−C2b), 5.49 (s, 1H, H−
C5b), 3.82 (d, 3J = 9.0 Hz, 2H, H−Cα), 3.72−3.44 (m, 4H, H−C6b),
2.94 (m, 1H, H−C1a), 2.83 (m, 1H, H−C3), 2.71 (m, 2H, H−C1),
2.48−2.36 (m, 3H, H−C3, H−C5a, C1a), 2.21 (m, 1H, NH), 1.84−1.55
(m, 8H, H−C2, H−C2a, H−C3a, H−C4a), 1.24 (m, 6H, H−C7b), 1.09
(d, 3J = 6.5 Hz, 3H, H−C6a). 13C NMR (100.6 MHz, CDCl3) δ (ppm):
140.1 (C1b), 137.8 (C4b), 127.9 (C2b), 126.7 (C3b), 101.4 (C5b), 64.7
(C5a), 61.0 (C6b), 56.0 (C1a), 53.6 (C3), 52.2 (C1), 48.2 (Cα), 34.4
(C4a), 25.9 (C2), 25.5 (C2a), 23.8 (C3a), 18.9 (C6a), 15.1 (C7b). MS
(EI) m/z (%): 348 (3), 303 (6), 165 (14), 155 (11), 126 (15), 112
(100), 98 (34), 91 (12). HRMS (CI) calculated for C21H36N2O2 [M]+:
348.2777; found: 348.2778.
Cα), 3.74−3.52 (m, 2H, H−C1a), 3.42−2.98 (m, 12H, H−C1, H−C3,
H−C5a, H−C1a), 2.27−1.51 (m, 16H, H−C2, H−C2a, H−C3a, H−
C4a), 1.37 (d, 6H, 3J = 8.0 Hz, H−C6a). 13C NMR (100.6 MHz, D2O) δ
(ppm): 132.6 (C1b), 131.3 (C2b), 60.7 (C5a), 53.0 (Cα), 51.4 (C3),
50.0 (C1a), 44.8 (C1), 32.1 (C4a), 23.6 (C2a), 22.0 (C2), 20.7 (C3a),
17.8 (C6a). Anal. calculated for C26H50N4Cl4·0.5H2O: C 54.83, H 9.03,
N 9.84; found: C 55.06, H 9.35, N 9.52.
Biology. (A). Antiviral Activity. HIV-1 strains were titered in MT-4
cells after acute infection, and infectivity was measured by evaluating the
cytopathic effect that was induced after 5-day cultures as described.22
Anti-HIV activity (EC50) and cytotoxicity (CC50) measurements in MT-
4 cells were based on the viability of cells that had been infected or not
infected with HIV-1; all were exposed to various concentrations of the
test compound. After the MT-4 cells were allowed to proliferate for 5
days, the number of viable cells was quantified by a tetrazolium-based
colorimetric method (MTT method) as described.
(B). ADME-TOX Studies. The multigram quantities of 11, 11(R,R),
11(S,S), and 11(S,R) necessary for the ADME-Tox studies were
prepared by ENANTIA S.L. (http://www.enantia.com).
In Vitro Human CXCR4 Receptor Functional Assay. This
functional assay was determined at CEREP (http://www.cerep.fr). The
CXCR4 (antagonist effect), test (ref. 2466), was determined using
human recombinant (CHO) cells following the protocol described by
Zhou et al.18 Stimulant, SDF-1a; stimulant concentration, 1 nM;
measured component, impedance; incubation, 28 °C; detection
method, cellular dielectric spectroscopy; reference compound, MIP-II
(IC50 = 55 nM).
(S)-4-((3-(2-Methylpiperidin-1-yl)propylamino)methyl)-
benzaldehyde (10(S)) . The intermediate aminoacetal 9(S) (0.50 g
(1.42 mmol)) was treated with 5 mL of 2 M HCl at room temperature
for 2 h. The resulting mixture was basified with NaOH and extracted
with CH2Cl2. The organic layers were combined, washed with brine,
dried over anhydrous MgSO4, filtered, and concentrated to give 0.32 g
(1.14 mmol, 80%) of the desired product 10(S) as a brown oil. IR (film)
ν (cm−1): 3281 (N−H), 2929, 2852, 2797, 2729, 1701 (CO), 1606
Cardiac Toxicity Assays (hERG). Determined at CEREP (http://
www.cerep.fr), hERG (automated patch-clamp) test (ref. 2245), using
hERG CHO-K1 cell line by following the automated patch-clamp
methodology proposed by Mathes.19 Test concentration, 0.1, 1, 10 μM;
incubation, 5 min/room temperature, cumulatively; detection method,
automated whole-cell patch-clamp; reference, E-4031 (IC50 = 25.6 nM).
Acute Intravenous Toxicity Study in Mice: Determination of
Maximum Nonlethal Dose and Minimum Lethal Dose.
Determined at Harlan Laboratories S.A. (http://www.harlan.com/).
The acute toxicity of the test items AMD3100, 11(R,R), 11(S,S), and
11(S,R) was evaluated after their intravenous administration to
Hsd:ICR (CD-1) mice (vehicle: physiological saline solution, sodium
chloride 0.9%, B. Braun Medical, S.A.) followed by an observation
period of 14 days. The animals received a single dose of the test item by
intravenous injection in the lateral vein of the tail using a graduated
syringe and a 25G (0.5 × 16 mm) needle. The test item was
administered to groups consisting of two males and two females each. In
the study, 4 groups consisting of 4 animals (2 males and 2 females) each
were administered in physiological saline solution at doses of 1, 2, 3, 3.5,
and 4 mg/kg. Viability/mortality: daily during the acclimatization
period, during the first 30 min and at approximately 1, 2, 3, and 5 h after
administration on test day 1 (in common with the clinical signs) and
twice daily during days 2−15. Clinical signs: daily during the
acclimatization period, during the first 30 min and at approximately 1,
2, 3, and 5 h after administration on test day 1, and once daily during
days 2−15. All abnormalities were recorded. On the basis of the results
obtained, the maximum nonlethal dose and the minimum lethal dose of
the test items when administered intravenously to Hsd:ICR (CD-1)
mice for both sexes as well as for males and females separately were
determined.
(C). Effect of CXCR4 Inhibitors on GICs and Glioma Initiation.
Reagents. CXCL12 was purchased from R&D Systems. Specific
antibodies against p-MAPK, MAPK, p-AKT, AKT (all from Cell
Signaling), and α-Tubulin (Sigma) were used for immunoblotting.
GBM Neurospheres. GBM neurospheres were generated as
described previously.14 Human GBM specimens were obtained from
the Vall d’Hebron Hospital. The clinical protocol was approved by the
Vall d’Hebron Institutional Review Board (CEIC), with informed
consent obtained from all subjects.
Proliferation Assay. Neurosphere cultures were dissociated with
acutase during 3 min, and then cells were counted. An equal number of
cells was seeded in triplicate for each condition and treated or left
untreated with CXCR4 inhibitors (1 μM) and/or CXCL12 (10 ng/
1
(CC), 1444, 1371, 1209. H NMR (400 MHz, CDCl3): δ (ppm)
10.00 (s, 1H, H−C5b), 7.84 (d, 3J = 8.0 Hz, 2H, H−C3b), 7.50 (d, 3J =
8.0 Hz, 2H, H−C2b), 3.87 (s, 2H, H−Cα), 2.90 (m, 1H, H−C1a), 2.79
(m, 1H, H−C3), 2.66 (t, 3J = 7.0 Hz, 2H, H−C1), 2.42−2.36 (m, 2H,
H−C3, H−C5a), 2.20−1.96 (m, 2H, H−C1a, NH), 1.75−1.55 (m, 6H,
H−C2, H−C2a, H−C4a, H−C3a), 1.34−1.24 (m, 2H, H−C4a, H−
3
C3a), 1.09 (d, J = 6.5 Hz, 3H, H−C6a). 13C NMR (100.6 MHz,
CDCl3) δ (ppm): 192.0 (C5b), 147.8 (C1b), 135.3 (C4b), 129.9 (C3b),
128.5 (C2b), 56.1 (C5a), 53.7 (Cα), 52.3 (C3), 52.0 (C1a), 48.5 (C1),
34.5 (C4a), 26.0 (C2a), 25.8 (C2), 23.8 (C3a), 18.9 (C6a). MS (EI) m/
z (%): 274 (6), 183 (3), 155 (7), 126 (9), 119 (10), 112 (100), 98 (39),
91 (14). HRMS (EI) calculated for C17H26N2O [M]+: 274.2045; found:
274.2037.
3-((R)-2-Methylpiperidin-1-yl)-N-(4-((3-((S)-2-methylpiperi-
din-1-yl)propylamino)methyl)benzyl)propan-1-amine Tetrahy-
drochloride (11(S,R)). (S)-4-((3-(2-Methylpiperidin-1-yl)-
propylamino)methyl)benzaldehyde (10(S)) (0.32 g (1.14 mmol)),
0.18 g (1.44 mmol) of (R)-N-(3-aminopropyl)-2-pipecoline (3(R)),
and Na2SO4 were mixed in 8 mL of anhydrous MeOH, and the mixture
was heated at reflux under N2 atmosphere for 24 h. The solid was
filtered, and the intermediate imine in MeOH was cooled to 0 °C and
treated with 0.05 g (1.14 mmol) of solid NaBH4. The reaction mixture
was stirred at room temperature overnight. Then water was added, and
the product was extracted with CH2Cl2. The organic layers were
combined, washed with brine, dried over anhydrous MgSO4, filtered,
and concentrated to give 0.40 g (0.95 mmol, 83%) of 1(S,R) as a yellow
oil. 1H NMR (400 MHz, CDCl3): δ (ppm) 7.30 (s, 4H, H−C2b), 3.79
(d, 4H, H−Cα), 2.89 (m, 2H, H−C1a), 2.77 (m, 2H, H−C3), 2.67 (m,
4H, H−C1), 2.41−2.27 (m, 8H, H−C3, H−C5a, H−C1a), 2.16 (m, 2H,
NH), 1.75−1.48 (m, 12H, H−C2, H−C2a, H−C4a, H−C3a), 1.29 (m,
4H, H−C4a, H−C3a), 1.06 (d, 6H, 3J = 8.0 Hz, H−C6a). HRMS (CI)
calculated for C26H47N4 [M+1]+: 415.3801; found: 415.3803. 1(S,R)
was converted to the corresponding tetrahydrochloride 11(S,R) in
almost quantitative yield upon treatment with methanolic HCl in excess
followed by concentration of the resulting solution. [α]D = 0.00 (c =
1
1.03, MeOH). IR (KBr) ν (cm−1): 3446, 2944, 2681, 1456, 526. H
NMR (400 MHz, D2O): δ (ppm) 7.59 (s, 4H, H−C2b), 4.33 (s, 4H, H−
7568
dx.doi.org/10.1021/jm300862u | J. Med. Chem. 2012, 55, 7560−7570

Journal of Medicinal Chemistry
Article
mL). Media and treatments were refreshed every 2−3 days. After 7 days
cells were collected, dissociated with acutase, stained with tripan blue,
and counted with a Countess Automated Cell Counter (Invitrogen).
Analysis of the CD44+ Population by Flow Cytometry. GBM
neurospheres were dissociated and individual cells were incubated for 15
min in blocking solution containing 10 μg/mL human IgG, followed by
anti-CD44 antibody or the control IgG2b isotype, both FITC-
conjugated (BD Pharmingen). Cells were incubated for 20 min on ice
protected from light, washed in phosphate-buffered saline (PBS), and
stained with propidium iodide (Sigma) to discriminate dying cells. Cells
were then analyzed by flow cytometry (FACSCalibur; Beckton
Dickinson) or sorted (MoFlo; DAKO) after staining with CD44-FITC.
Intracranial Tumor Assay. All mouse experiments were approved
by and performed according to the guidelines of the Institutional Animal
Care Committee of the Vall d’Hebron Research Institute in agreement
with the European Union and national directives. Cells were collected,
dissociated with acutase, stained with trypan blue to assess viability, and
counted with a Countess Automated Cell Counter (Invitrogen). Live
cells (100.000) were stereotactically inoculated into the corpus striatum
of the right-brain hemisphere (1 mm anterior and 1.8 mm lateral to the
bregma; 2.5 mm intraparenchymal) of 9-week-old NOD-SCID mice
(Charles River Laboratories). Mice were euthanized when they
presented neurological symptoms or a significant loss of weight.
Magnetic resonance imaging (MRI) analysis was performed in mice
injected intraperitoneally with gadolinium diethylenetriamine penta-
acetic acid at a dose of 0.25 mmol of gadolinium/kg of body weight.
T1W magnetic resonance images were acquired in a 9.4 T vertical bore
magnet interfaced to an AVANCE 400 system (Bruker) using a spin−
echo sequence as described previously.14
GBM, grade IV gliomas or glioblastoma; GICs, glioma initiating
cells; hERG, human Ether-a-go-go-Related Gene; HIV-1, human
̀
immunodeficiency virus type 1; MAPK, mitogen-activated
protein kinase; MLD, minimum lethal dose; MND, maximum
nonlethal dose; MRI, magnetic resonance imaging; p-Akt,
phospho-Akt; p-MAPK, phospho-MAPK; PI3K, phosphatidyli-
nositol 3-kinase
REFERENCES
■
(1) Brat, D. J.; Castellano-Sanchez, A. A.; Hunter, S. B.; Pecot, M.;
Cohen, C.; Hammond, E. H.; Devi, S. N.; Kaur, B.; Van Meir, E. G.
Pseudopalisades in Glioblastoma Are Hypoxic, Express Extracellular
Matrix Proteases, and Are Formed by an Actively Migrating Cell
Population. Cancer Res. 2004, 64, 920−927.
(2) Furnari, F. B.; Fenton, T.; Bachoo, R. M.; Mukasa, A.; Stommel, J.
M.; Stegh, A.; Hahn, W. C.; Ligon, K. L.; Louis, D. N.; Brennan, C.;
Chin, L.; DePinho, R. A.; Cavenee, W. K. Malignant astrocytic glioma:
Genetics, biology, and paths to treatment. Genes Dev. 2007, 21, 2683−
2710.
(3) Zhu, Y.; Parada, L. F. The molecular and genetic basis of
neurological tumours. Nat. Rev. Cancer 2002, 2, 616−626.
(4) Anido, J.; Sae
G.; Carmona, M. A.; Prieto-San
Prudkin, L.; Cuartas, I.; Raventos
García-Dorado, D.; Lahn, M. M.; Yingling, J. M.; Rodon
́
z-Borderías, A.; Gonzal
chez, R. M.; Barba, I.; Martínez-Sae
, C.; Martínez-Ricarte, F.; Poca, M. A.;
, J.; Sahuquillo,
̀
ez-Junca,
̀
A.; Rodon
́
, L.; Folch,
́
́
z, E.;
́
́
J.; Baselga, J.; Seoane, J. TGF-beta Receptor Inhibitors Target the
CD44(high)/Id1(high) Glioma-Initiating Cell Population in Human
Glioblastoma. Cancer Cell 2010, 18, 655−668.
(5) Visvader, J. E.; Lindeman, G. J. Cancer stem cells in solid tumours:
Accumulating evidence and unresolved questions. Nat. Rev. Cancer
2008, 8, 755−768.
ASSOCIATED CONTENT
* Supporting Information
■
S
Spectral data for compounds 6(S), 6(R), 3(S), 3(R), 1(S,S),
11(S,S), 1(R,R), 11(R,R), 9(S), 10(S), and 11(S,R). High-
performance liquid chromatography (HPLC) analytical profile
for 11(S,S), 11(R,R), and 11(S,R). CD44 low/CD44 null
analysis. This material is available free of charge via the Internet at
http://pubs.acs.org.
(6) Este, J. A.; Telenti, A. HIV entry inhibitors. Lancet 2007, 370, 81−
88.
(7) Tilton, J. C.; Doms, R. W. Entry inhibitors in the treatment of HIV-
1 infection. Antiviral Res. 2010, 85, 91−100.
́
(8) Pettersson, S.; Perez-Nueno, V. I.; Ros-Blanco, L.; Puig de La
Bellacasa, R.; Rabal, M. O.; Batllori, X.; Clotet, B.; Clotet-Codina, I.;
Armand-Ugon, M.; Este, J. A.; Borrell, J. I.; Teixido, J. Discovery of novel
́
́
́
non-cyclam polynitrogenated CXCR4 coreceptor inhibitors. Chem-
AUTHOR INFORMATION
Corresponding Author
■
MedChem 2008, 3, 1549−1557.
(9) Bian, X. W.; Yang, S. X.; Chen, J. H.; Ping, Y. F.; Zhou, X. D.; Wang,
Q. L.; Jiang, X. F.; Gong, W.; Xiao, H. L.; Du, L. L.; Chen, Z. Q.; Zhao,
W.; Shi, J. Q.; Wang, J. M. Preferential expression of chemokine receptor
CXCR4 by highly malignant human gliomas and its association with
poor patient survival. Neurosurgery 2007, 61, 570−578.
*Dr. Jose
́
I. Borrell: phone, +34 932 672 059; fax, +34 932 056
266; e-mail, j.i.borrell@iqs.url.edu. Dr. Joan Seoane: phone, +34
934 894 167; fax, +34 934 893 884; e-mail jseoane@vhio.net.
Notes
(10) Tseng, D.; Vasquez-Medrano, D. A.; Brown, J. M. Targeting SDF-
1/CXCR4 to inhibit tumour vasculature for treatment of glioblastomas.
Br. J. Cancer 2011, 104, 1805−1809.
(11) Teicher, B. A.; Fricker, S. P. CXCL12 (SDF-1)/CXCR4 pathway
in cancer. Clin. Cancer Res. 2010, 16, 2927−2931.
(12) Terasaki, M.; Sugita, Y.; Arakawa, F.; Okada, Y.; Ohshima, K.;
Shigemori, M. CXCL12/CXCR4 signaling in malignant brain tumors: A
potential pharmacological therapeutic target. Brain Tumor Pathol. 2011,
28, 89−97.
The authors declare the following competing financial
́ ́
interest(s): L. Ros-Blanco, J. Este, J. Teixido, and J. I. Borrell
are co-inventors of patents WO 2008/049950 A1 and PCT/
ES2011/070407 covering the compounds and their use, licensed
to Janus Development SL.
ACKNOWLEDGMENTS
■
The authors appreciate the financial support from Janus
Development SL, which also contributed to the ADME-TOX
and GICs/glioma initiation studies design. We also appreciate
(13) do Carmo, A.; Patricio, I.; Cruz, M. T.; Carvalheiro, H.; Oliveira,
C. R.; Lopes, M. C. CXCL12/CXCR4 promotes motility and
proliferation of glioma cells. Cancer Biol. Ther. 2010, 9, 56−65.
́
the financial support from the Spanish Ministerio de Economıa y
́
(14) Penuelas, S.; Anido, J.; Prieto-Sanchez, R. M.; Folch, G.; Barba, I.;
̃
Competitividad project SAF2010-C21617-C02 (J.I.B. and B.C.)
and BFU2009-06958 (J.A.E.). L.R.-B. thanks IQS and General-
itat de Catalunya for a scholarship and a FI−B 2008−2010
scholarship, respectively.
Cuartas, I.; García-Dorado, D.; Poca, M. A.; Sahuquillo, J.; Baselga, J.;
Seoane, J. TGF-beta increases glioma-initiating cell self-renewal through
the induction of LIF in human glioblastoma. Cancer Cell. 2009, 15 (4),
315−327.
(15) Doller, D.; Davies, R.; Chackalamannil, S. A practical preparation
of (R)- and (S)-N-Boc-2-methylpiperidines. Tetrahedron: Asymmetry
1997, 8, 1275−1278.
(16) Apelt, J.; Ligneau, X.; Pertz, H. H.; Arrang, J.; Ganellin, C. R.;
Schwartz, J.; Schunack, W.; Stark, H. Development of a new class of
nonimidazole histamine H3 receptor ligands with combined inhibitory
ABBREVIATIONS USED
■
Akt, protein kinase B; CC50, half cytotoxic concentration;
CXCL12, chemokine (C−X−C motif) ligand 12; CXCR4, CXC
chemokine receptor type 4; EC50, 50% effective concentration;
7569
dx.doi.org/10.1021/jm300862u | J. Med. Chem. 2012, 55, 7560−7570

Journal of Medicinal Chemistry
Article
histamine N-methyltransferase activity. J. Med. Chem. 2002, 45, 1128−
1141.
(17) Jo, M. N.; Seo, H. J.; Kim, Y.; Seo, S. H.; Rhim, H.; Cho, Y. S.; Cha,
J. H.; Koh, H. Y.; Choo, H.; Pae, A. N. Novel T-type calcium channel
blockers: Dioxoquinazoline carboxamide derivatives. Bioorg. Med. Chem.
2007, 15, 365−373.
(18) Zhou, Y.; Larsen, P. H.; Hao, Ch.; Yong, V. W. CXCR4 is a major
chemokine receptor on glioma cells and mediates their survival. J. Biol.
Chem. 2002, 277, 49481−49487.
(19) Mathes, C. QPatch: The past, present and future of automated
patch clamp. Expert Opin. Ther. Targets 2006, 10, 319−327.
(20) Sturgill, T. W. MAP kinase: It’s been longer than fifteen minutes.
Biochem. Biophys. Res. Commun. 2008, 371, 1−4.
(21) Rubin, J. B.; Kung, A. L.; Klein, R. S.; Chan, J. A.; Sun, Y.; Schmidt,
K.; Kieran, M. W.; Luster, A. D.; Segal, R. A. A small-molecule antagonist
of CXCR4 inhibits intracranial growth of primary brain tumors. Proc.
Natl. Acad. Sci. U.S.A. 2003, 100 (23), 13513−13518.
(22) Gonzalez-Ortega, E.; Ballana, E.; Badia, R.; Clotet, B.; Este, J. A.
Compensatory mutations rescue the virus replicative capacity of VIRIP-
resistant HIV-1. Antiviral Res. 2011, 92, 479−483.
7570
dx.doi.org/10.1021/jm300862u | J. Med. Chem. 2012, 55, 7560−7570