NBD-labeled derivatives of the immunomodulatory drug FTY720 as tools for metabolism and mode of action studies

Article abstract of DOI:10.1016/j.bmcl.2005.09.038

Fluorescently labeled chiral analogs of the immunomodulatory drug FTY720 and its corresponding phosphates with variable aliphatic spacers between the aromatic ring and the NBD label have been synthesized. Determining the influence of the spacer on the in

Full text of DOI:10.1016/j.bmcl.2005.09.038

Bioorganic & Medicinal Chemistry Letters 16 (2006) 84–87
NBD-labeled derivatives of the immunomodulatory drug FTY720
as tools for metabolism and mode of action studies
Peter Ettmayer,a,* Thomas Baumruker,a Danilo Guerini,b Diana Mechtcheriakova,a
Peter Nussbaumer,a Markus B. Streiff b and Andreas Billicha
aNovartis Institutes for BioMedical Research, Brunnerstrasse 59, A-1230 Vienna, Austria
bNovartis Institutes for BioMedical Research, Basel, Switzerland
Received 29 July 2005; revised 16 September 2005; accepted 19 September 2005
Available online 19 October 2005
Abstract—Fluorescently labeled chiral analogs of the immunomodulatory drug FTY720 and its corresponding phosphates with var-
iable aliphatic spacers between the aromatic ring and the NBD label have been synthesized. Determining the influence of the spacer
on the in vitro phosphorylation rate by SPHK1 and 2 resulted in the identification of NBD-(R)-AAL 1c,d which are phosphorylated
with an efficiency comparable to that of the unlabeled FTY720 analog (R)-AAL. Furthermore, the NBD-(R)-AAL phosphates 10c,d
were proven to be a functional S1P receptor agonist.
Ó 2005 Elsevier Ltd. All rights reserved.
FTY720 is a potent immunomodulatory agent with po-
tential usefulness in the control of organ transplant
rejection and for treatment of autoimmune diseases
such as autoimmune diabetes, multiple sclerosis, and
systemic lupus erythematosus.1,2 FTY720 functions as
a prodrug that needs to be phosphorylated by sphingo-
sine kinases in vivo with the resulting FTY720 phos-
phate being the active principle.3,4 FTY720 phosphate
functions as an agonist at four of the five sphingosine-
1-phosphate (S1P) receptors (S1P1, S1P3, S1P4, and
S1P5).5,6 Recent reports propose that binding of
FTY720 phosphate to the S1P1 receptor on T-cells
leads to receptor internalization and concomitant down
regulation, thereby blocking direct migration of T-cells
toward high levels of S1P in the bloodstream resulting
in inhibiting the egress of T-cells from lymphoid or-
gans.7 In addition, alternative mechanisms are under
investigation.8
We now report on the synthesis and characterization of
nitrobenzo-2-oxa-1,3-diazole- (NBD) labeled chiral
FTY720 derivatives based on the functional FTY720
analog (R)-AAL5 with variable lengths of the aliphatic
chain between the NBD label and the aromatic ring.
The (R) enantiomer was chosen because it had been
shown previously that (S)-AAL is not a substrate of
SPHK and that the corresponding (S)-phosphate does
not signal through S1P receptors5 (Fig. 1).
Using recombinant human SPHK1 and 2, we describe
the substrate properties of these labeled (R)-AAL deriv-
atives by quantifying the conversion rate to the corre-
sponding phosphates. In addition, the S1P receptor
OH
FTY720
OH
For further understanding of the mode of action, subcel-
lular localization, and metabolism of both FTY720 and
its phosphate, fluorescently labeled, bioactive analogs
would be ideal tools. Recently, we and others reported
on the synthesis and successful application of pyrene
ester- and NBD-labeled sphingosine derivatives.9,10
NH2
O
(R)-AAL
OH
NH2
NBD
(CH2)
O
n
Fluorescent derivatives 1a-d
OH
Keywords: Immunomodulation; Fluorescence; Sphingolipids; Kinase;
Receptors.
n = 4,6,8,11
NH2
*
Corresponding author. Tel.: +43 1 80105/2530; fax: +43 1 80105/
9530; e-mail: peter.ettmayer@vie.boehringer-ingelheim.com
Figure 1. Design of fluorescently labeled (R)-AAL derivatives.
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.09.038

P. Ettmayer et al. / Bioorg. Med. Chem. Lett. 16 (2006) 84–87
85
Table 1. Rates of phosphorylation of NBD-labeled (R)-AAL deriva-
tives by SPHK1 and 2
binding profile and functional activity of the corre-
sponding phosphates were determined.
O
N
N
H
N
The key intermediate 2 was prepared by the Schoellkopf
protocol as recently described.11 Alkylation of 2 with the
commercially available. x-Bromo-alkylphthalimides
3a,b provided the orthogonally protected amino alco-
hols 5a,b in 28–32% yield (Scheme 1). The relatively
low yield is due to partial hydrolysis of the phthalimide
group. Utilizing the dibenzyl-protected mesylates 4c,d
under identical alkylation conditions provided 6c,d in
71–76% yield. Deprotection of 5a,b using hydrazine in
ethanol provided the primary amines 7a,b in quantita-
tive yield. Selective removal of the dibenzyl-protecting
groups by refluxing 6c,d and 10% Pd/C in a methanolic
ammonium formate solution provided 7c,d in quantita-
tive yield. 7a–d were reacted with NBD–Cl in THF/TEA
to provide 8a–d (70–75% yield). Acidic cleavage of the
Boc group provided the NBD-labeled (R)-AAL deriva-
tives 1a–d12 in quantitative yield.
O
n
NO2
HO
NH2
1a-d
Compound
n
SPHK1 (%)
SPHK2 (%)
Sphingosine
FTY720
(R)-AAL
—
—
—
4
100.0
0.6
4.3
100.0
14.1
22.9
<0.1
9.0
1a
1b
1c
1d
<0.1
0.7
6
8
11
1.9
5.6
6.5
119.0
Values are given relative to the rate for D-sphingosine.9
We selected amino alcohols 1c,d for further chemical
conversion to the corresponding NBD-labeled phos-
phates (R)-AAL-P (10c,d). Thus, 10c,d13 were obtained
in good yield using standard phosphoamidate chemistry
(Scheme 2).
The NBD-labeled (R)-AAL derivatives 1a–d were used
as substrates for human recombinant SPHK1 and 2.9
The rates of phosphorylation were determined by assess-
ing the incorporation of radiolabeled phosphate upon
incubation with [c-32P]ATP and the enzymes. Results
are reported in Table 1 relative to the natural substrate
sphingosine. All FTY720 derivatives were more effi-
ciently converted by SPHK2 compared to SPHK1. As
observed already with the fluorescently labeled sphingo-
sine derivatives, conversion rates improved with the
length of the aliphatic chain between the head group
and the NBD label. While 1a is not a substrate for
SPHK1 and SPHK2, phosphorylation rates for 1b,c
are almost comparable to those of FTY720 and (R)-
AAL. The phosphorylation of 1d by SPHK2 in vitro
was even superior to the natural substrate sphingosine.
To check whether the NBD label would affect the S1P
receptor binding profile, we also prepared the NBD-
labeled S1P derivative 13 using the trimethyl phosphite
procedure described by Hakogi et al.10 (Scheme 2). After
experiencing some difficulties in isolating 12 in reason-
able yield, we found that the phosphorylation of the
Boc-protected NBD-sphingosine 119 worked best by
adding 2.2 equiv of CBr4 and 2.6 equiv of trim-
ethylphosphite to a solution of 11 in a minimum amount
O
N
N
H
O
N
O
P
i
n
1c,d
O
O
n
61-68%
NO2
O
O
n
or
NH
O
S
9c,d
Br
OH
N
O
O
N(Bn)2
N
O
H
O
N
O
O
N
n
O
O
ii
3a,b n = 4, 6
4c,d n = 8, 11
HO
HO
P
HO
HO
NO2
O
64-76%
NH
i
10c,d
NH2
2
O
O
O
O
O
N+ O
N+ O
N+ O
O
NRR´
R
R´
#
n
phthalimide 5a,b
Bn Bn
ii
6c,d
7a-d
8c-d
N
O
N
iii
iv
N
O
H
H
H
NBD
NH
O
N
HN
HN
HN
N
N
iii
50%
HO
iv
O
O
(CH2)10
O
(CH2)10
(CH2)10
56%
v
O
O
N
N
HO
HO
H
N
n
O
n
4
6
8
O
O
N
H
NH2
N
H
1a
1b
1c
OPO3(CH3)2
OPO3H2
OH
NO2
HO
1d 11
NH2
12
11
13
Scheme 1. Variation of the label. Reagents and conditions: (i) K2CO3,
EtOH/DMF 3/1, 50 °C; (ii) N2H4ÆH2O, EtOH, reflux; (iii) 10% Pd/C,
ammonium formiate, MeOH, reflux; (iv) NBD–Cl, Et3N, THF;
(v) 10% aq TFA, rt.
Scheme 2. Preparation of the labeled phosphates. Reagents and
conditions: (i) (a) 1H-tetrazole, di-tert-butyl-diethylphosphoramidite,
THF, rt, (b) H2O2; (ii) 10% aq TFA, rt; (iii) CBr4, pyridine, P(OCH3)3,
À10 °C to rt; (iv) TMS-Br, acetonitrile, rt.

86
P. Ettmayer et al. / Bioorg. Med. Chem. Lett. 16 (2006) 84–87
2.4
of dry pyridine at À10 °C and subsequently allowing the
A
B
S1P
10d
10c
S1P
10d
10c
reaction mixture to warm to room temperature.
2.0
1.6
1.2
0.8
0.4
0.0
3
Motivated by the good in vitro phosphorylation of 1c,d
by SPHK2, we investigated the S1P-receptor binding
profile of the corresponding NBD-labeled (R)-AAL
phosphates 10c and d in comparison to that of
(R)-AAL phosphate, S1P and NBD-labeled S1P 13
(Table 2). The [c-35S]GTPcS binding assays were
performed as described previously.5 NBD-S1P (13)
was found to be an agonist for all S1P receptors tested.
While its potency (pEC50) is 1–2 log steps lower
compared to that of unlabeled S1P, it is still in the nano-
mole-range. The efficacy (Emax) is reduced, especially for
the S1P1 receptor. The NBD-labeled (R)-AAL-phos-
phates 10c,d also functioned as agonists at the S1P
receptors with slightly reduced potency and efficacy
compared to that of the unlabeled derivative. As previ-
ously shown for (R)-AAL-phosphate,5 10c,d do not
signal via the S1P2 receptor (data not shown).
2
1
0
1E-12 1E-10 1E-8
1E-6
1E-12 1E-10 1E-8
1E-6
Figure 2. Ca2+ mobilization by 10c and d compared to that of the
endogenous agonist sphingosine-1-phosphate (S1P). CHO cells
expressing S1P1 (A) and S1P3 (B) were treated with different
concentrations of the agonists and signals were recorded by
a
fluorescent image plate reader. Plotted is molar concentration against
fold change in fluorescence.14
A
B
controls
In line with the binding data, 10c,d induced Ca2+ mobi-
lization in S1P1 and S1P3 overexpressing CHO cells.14
For the S1P1 receptor the potency and efficacy was com-
parable to that of S1P and was significantly weaker for
the S1P3 receptor (Fig. 2). Taking both the S1P1 bind-
ing data and the Ca2+ mobilization assay into account,
10c and d were found to be equally potent and effective
functional S1P receptor agonists.
1d
10d
Figure 3. (A) Rapid incorporation of 1d into endothelial cells
(HUVEC). Cells were incubated for 15 min with 1 lM 1d under
normal growth conditions; ER, endoplasmic reticulum; Mito, mito-
chondria. (B) Metabolic conversion of 1d assayed by thin-layer
chromatography. Cells were incubated for 0.5, 1, and 3 h with 1 lM
1d followed by lipid extraction.
We used the labeled amino alcohol 1d to examine the up-
take, subcellular distribution, and metabolism (conver-
sion to the phosphate) in human endothelial cells
(HUVEC). Fluorescently labeled (R)-AAL 1dwas rapidly
incorporated into the cells within 5 min after addition and
showed predominant distribution to the endoplasmic
reticulum. The plasma membrane and, in some cells,
mitochondria were visibly labeled as well (Fig. 3A). 1d
was converted intracellularly to the phosphate 10d as
shown by thin-layer chromatography9 (Fig. 3B). The con-
version efficiency was comparable to that of FTY720
(data not shown). These data indicate that the NBD label
does not hinder enzymatic phosphorylation within cells.
Because of the superior aqueous solubility of 10c15 com-
pared to 10d in cell culture medium, we used the NBD-
labeled (R)-AAL phosphate 10c and NBD-S1P 13 to
examine the MAP kinase/ERK activation in HUVEC.16
At 1 lM, both NBD-labeled agonists were able to trig-
ger ERK phosphorylation (Fig. 4) with an efficiency
comparable to that of unlabeled FTY720 phosphate.
Table 2. Potency (pEC50) and efficacy (Emax) values of phosphorylated compounds at human S1P receptors
Receptor
Parameter
Compound
S1P
NBD-S1P 13
(R)-AAL-P
10c
10d
a
S1P1
S1P3
S1P4
S1P5
pEC50
Emax
9.0 (8.1–9.7)b
1.00
7.2 (7.0–7.3)
0.67 (0.58–0.76)
9.1 (8.9–9.2)
0.91 (0.80–1.00)
7.7 (7.4–7.9)
0.76 (0.74–0.78)
9.0 (8.7–9.3)
0.82 (0.76–0.87)
pEC50
Emax
9.75 (9.1–10.5)
1.00
8.3 (8.1–8.4)
0.99 (0.78–1.20)
8.2 (8.0–8.5)
0.43 (0.32–0.49)
7.7 (7.5–7.9)
0.87 (0.68–1.05)
7.5 (7.2–7.8)
0.69 (0.56–0.82)
d
d
pEC50
Emax
8.9 (8.1–9.5)
1.00
7.2 (7.0–7.4)
1.06 (1.02–1.10)
8.9 (8.1–9.5)
1.53
7.7 (7.4–7.9)
0.27 (0.16–0.37)
c
c
pEC50
Emax
8.7 (8.4–9.1)
1.00
7.3 (7.1–7.6)
0.80 (0.71–0.88)
8.7 (8.4–9.1)
0.75 (0.53–1.00)
8.7 (8.3–9.1)
0.27 (0.23–0.31)
a pEC50, Àlog molar concentration of compounds resulting in 50% of maximal GTPcS binding; Emax, maximal [c-35S] binding as a fraction of the
S1P signal (set at 100% = 1.0).
b Average value of three to six experiments (range of values).
c The shape of binding curve did not allow the calculation of an EC50
d Not an agonist up to 10 lM.
.

P. Ettmayer et al. / Bioorg. Med. Chem. Lett. 16 (2006) 84–87
87
References and notes
1
2
3
4
5
p-ERK1/2
actin
1. Brinkmann, V.; Lynch, K. R. Curr. Opin. Immunol. 2002,
14, 569.
2. Brinkmann, V.; Cyster, J. G.; Hla, T. Am. J. Transplant.
2004, 4, 1019.
3. Paugh, S. W.; Payne, S. G.; Barbour, S. E.; Milstien, S.;
Spiegel, S. FEBS Lett. 2003, 554, 189.
4. Billich, A.; Bornancin, F.; Devay, P.; Mechtcheriakova,
D.; Urtz, N.; Baumruker, T. J. Biol. Chem. 2003, 278,
47408.
5. Brinkmann, V.; Davis, M. D.; Heise, C. E.; Albert, R.;
Cottens, S.; Hof, R.; Bruns, C.; Prieschl, E.; Baumruker,
T.; Hiestand, P.; Foster, C. A.; Zollinger, M.; Lynch, K.
R. J. Biol. Chem. 2002, 277, 21453.
Figure 4. Activation of ERK1/2 by NBD-S1P 13 (2), NBD-(R)-AAL
phosphate 10c (3), S1P (4) and FTY720 phosphate (5) in endothelial
cells. Short-starved cells were incubated for 10 min with 1 lM of
compound and collected for Western blot analysis. Membranes were
incubated with antibodies directed to phosphorylated ERK1/2 and re-
probed with anti-b-actin antibodies to ensure equal loading of the
samples; 1, unstimulated cells; 2–5, compounds as indicated in
brackets.
6. Mandala, S.; Hajdu, R.; Bergstrom, J.; Quackenbush, E.;
Xie, J.; Milligan, J.; Thornton, R.; Shei, G. J.; Card, D.;
Keohane, C.; Rosenbach, M.; Hale, J.; Lynch, C. L.;
Rupprecht, K.; Parsons, W.; Rosen, H. Science 2002, 296,
346.
7. Graler, M. H.; Goetzl, E. J. FASEB J. 2004, 18, 551.
8. Hla, T. Semin. Cell Dev. Biol. 2004, 15, 513.
9. Ettmayer, P.; Billich, A.; Baumruker, T.; Mecht-cheriak-
ova, D.; Schmid, H.; Nussbaumer, P. Bioorg. Med. Chem.
Lett. 2004, 14, 1555.
10. Hakogi, T.; Shigenari, T.; Katsumura, S.; Sano, T.;
Kohno, T.; Igarashi, Y. Bioorg. Med. Chem. Lett. 2003,
13, 661.
11. Cottens, S.; Albert, R.; Zecri, F.; Buehlmayer, P.; Spanka,
C.; Brinkmann, V.; Nussbaumer, P.; Ettmayer, P.; Hoe-
genauer, K.; Gray, N.; Pan, S.; Hinterding, K. Synthesis
2003, 11, 1667.
50
40
30
20
10
0
control
S1P (R)-AAL-P 13
10c
Figure 5. Activity of test compounds at 1 lM in an in vitro
angiogenesis model. The capillary-like network formation on Matrigel
upon activation of HUVEC for 8 h is shown. Effects were quantitated
12. Characteristic data for 1d: 1H NMR (DMSO-d6,
400 MHz) d: 9.53 (br s; 1H), 8.48 (d; J = 8.9 Hz; 1H),
7.75 (br s; 3H), 7.05 (d; J = 6.6 Hz; 2H), 6.81 (d;
J = 6.6 Hz; 2H), 6.38 (d; J = 8.9 Hz; 1H), 5.50 (br s; 1H),
3.86 (t; J = 6.5 Hz; 2H), 3.43 (m; 2H), 2.50 (m; 2H), 1.80–
1.60 (m; 6H), 1.42–1.15 (m; 14H), 1.18 (s; 3H). Electro-
spray MS: 528.3 (M+H)+, C28H41N5O5, calcd 527.3.
by a direct counting of the number of branching points per
microscopic field.
To further characterize the biological activities of NBD-
labeled phosphates in a cellular system, we investigated
the effect of NBD-(R)-AAL phosphate 10c and NBD-
S1P 13 on morphogenic differentiation of endothelial
cells using an in vitro angiogenesis assay on Matri-
gel,16,17 wherein the number of branching points reflects
the potency of the stimulus. As shown in Figure 5, both
unlabeled (R)-AAL phosphate and S1P and their NBD-
labeled analogs 10c and 13 strongly promoted the capil-
lary-like network formation.
1
13. Characteristic data for 10d: H NMR (CD3OD + 1 drop
DCL, 500 MHz): d = 8.51 (d; 1H), 7.13 (d; 2H), 6.81 (d;
2H), 6.34 (d; 1H), 4.09/4.01 (ABX system, 2H), 3.91 (t;
2H), 3.54 (br s; 2H), 2.63 (m; 2H), 2.04–1.87 (m; 2H),
1.81–1.69 (m; 2H), 1.50–1.29 (m; 14H), 1.41 (s; 3H).
Electrospray MS: 608.4 (M+H)+, C28H42N5O8P, calcd
607.28.
14. CHO overexpressing S1P receptors were plated in black
Costar plate. Cells were loaded with 2 lM Fluo4AM
(Molecular Probes), 5 mM probenecid for 1 at 37 °C,
rinsed, and transferred to the fluorescent image plate
reader (FLIPR). The cells were pre-treated for 15–25 min
with 10 lM ATP and, after measuring the baseline
fluorescence (Fb) for 40 s, the agonist was added to
determine the peak of the fluorescence Fm (3–5 min).
Pretreatment with ATP allows efficient coupling of S1P
receptors to Ca2+ mobilization as described for other
GPCRs by Werry et al. Biochem. J. 2003, 374, 281.
15. A 10 mM stock solution of 10c in DMSO/1 N HCl (15/2 v/
v) is diluted with assay medium.
16. Sanchez, T.; Estrada-Hernandez, T.; Paik, J. H.; Wu,
M.-T.; Venkataraman, K.; Brinkmann, V.; Claffey, K.;
Hla, T. J. Biol. Chem. 2003, 278, 47281.
17. Lucerna, M.; Mechtcheriakova, D.; Kadl, A.; Schab-
bauer, G.; Schaefer, R.; Gruber, F.; Koshelnick, Y.;
Mueller, H.-D.; Issbruecker, K.; Clauss, M.; Binder, B.
R.; Hofer, E. J. Biol. Chem. 2003, 278, 11433.
In summary, we prepared fluorescence-labeled chiral
FTY720 analogs and their corresponding phosphates,
which can serve as valuable tools for many biological
investigations. Amino alcohol derivatives with an octa-
nyl (1c) or undecanyl (1d) spacer between the NBD-label
and the phenoxy ring were proven to be efficiently phos-
phorylated by SPHK2 in vitro and in vivo. 1d was rap-
idly taken up by cells and was distributed preferentially
to the endoplasmic reticulum. The corresponding phos-
phates 10c,d were proven to be potent and efficacious
agonists for the S1P receptors 10c,d induced Ca2+ mobi-
lization in S1P1 and S1P3 overexpressing CHO cells
comparable to that of the endogenous ligand S1P. Fur-
thermore, the NBD label did not interfere with ERK
activation and the pro-angiogenesis effect of S1P and
FTY720.