Design, synthesis, and biological activity of novel Magmas inhibitors

Article abstract of DOI:10.1016/j.bmcl.2011.03.050

Magmas (mitochondria associated, granulocyte-macrophage colony stimulating factor signaling molecule), is a highly conserved and essential gene, expressed in all cell types. We designed and synthesized several small molecule Magmas inhibitors (SMMI) and a

Full text of DOI:10.1016/j.bmcl.2011.03.050

Bioorganic & Medicinal Chemistry Letters 21 (2011) 3479–3482
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Design, synthesis, and biological activity of novel Magmas inhibitors
Paul T. Jubinsky a,b, , Mary K. Short b, Mohmoud Ghanem c, Bhaskar C. Das b,d,
⇑
⇑
a Pediatric Hematology/Oncology, Yale University Medical School, New Haven, CT 06510, United States
b Department of Developmental & Molecular Biology, Albert Einstein College of Medicine, Bronx, NY 10461, United States
c Department of Biochemistry, Albert Einstein College of Medicine, Bronx, NY 10461, United States
d Department of Nuclear Medicine, Albert Einstein College of Medicine, Bronx, NY 10461, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Magmas (mitochondria associated, granulocyte–macrophage colony stimulating factor signaling mole-
cule), is a highly conserved and essential gene, expressed in all cell types. We designed and synthesized
several small molecule Magmas inhibitors (SMMI) and assayed their effects on proliferation in yeast. We
Received 29 January 2011
Revised 10 March 2011
Accepted 14 March 2011
Available online 21 March 2011
found that the most active compound 9 inhibited growth at the 4
lM scale. This compound was shown by
fluorometric titration to bind to Magmas with a Kd = 33 M. Target specificity of the lead compound was
l
established by demonstrating direct binding of the compound to Magmas and by genetic studies. Molec-
ular modeling suggested that the inhibitor bound at the predicted site in Magmas.
Ó 2011 Elsevier Ltd. All rights reserved.
Keywords:
Magmas inhibitors
Guanidine pharmacophores
Chemical genetics
LRD-limited
Rational design
Molecular modeling
Tim16
Pam16
Mitochondrial protein import
Our laboratory identified Magmas (mitochondria associated
granulocyte–macrophage colony stimulating factor signaling mol-
ecule), as a GM-CSF inducible gene in myeloid cells.1 Magmas pro-
tein localizes to the inner membrane of mitochondria, and protein
expression is both developmentally regulated and tissue specific.2
The protein is highly conserved, and is required for mitochondria
function and cell survival.3 Immuno-precipitation and immuno-
affinity chromatography studies show that Magmas mostly associ-
ates with other mitochondrial proteins. Reduction in Magmas
expression results in decreased mitochondrial function (Soumit
Roy and Jubinsky, unpublished data).
The yeast ortholog of Magmas, known as Tim16 or Pam16 has
been shown to functionally interact with another protein, Tim14,
to regulate the importation of peptides into the mitochondria
through an import complex.4 Studies have established that the
functional Tim14/Tim16 complex is a heterodimer.4b,5
Elevated levels of Magmas protein occurred in a subset of
advanced-grade, invasive adenocarcinoma. However, many sam-
ples of similar stage and grade had staining patterns identical to
normal prostate epithelium. Increased staining resulted from
higher amounts of Magmas in the abnormal cells and not from in-
creased numbers of mitochondria. It indicates that the Magmas
expression may affect response to therapy, possibly by altering
mitochondrial function. Data from other laboratories also supports
the importance of Magmas expression in human cancer.7
Based on the above findings, we believed that small molecule
Magmas inhibitors (SMMI) could be beneficial for studying mito-
chondrial function and for diagnosing and treating human disease.
The design of potential compounds was aided by analyzing
information known about the structure of Magmas derived from
crystallography data.5a
Small molecules likely to interact with highly conserved func-
tional regions of Magmas were conceived and a corresponding
small chemical library containing oxazine, chromene and guani-
dine pharamacophore groups were synthesized. Instead of the
large libraries typically used in high-throughput screening our
library consists of 10 lead compounds likely to contain a fairly
complete range of structural motifs that interact with Magmas.
We have termed this method the ‘limited design approach’
The biological activity of the compounds was tested in
Saccharomyces cerevisiae proliferation assays. Human Magmas
Magmas expression in prostate cancer samples suggested a role
for this protein in human cancer.6 Magmas staining is minimal to
absent in normal prostate tissue by immuno-histochemistry.
⇑
Corresponding authors. Tel.: +203 785 2754; fax: +203 737 2228 (P.T.J.); tel.: +1
17184302422; fax: +1 17184308853 (B.C.D.)
E-mail addresses: paul.jubinsky@yale.edu (P.T. Jubinsky), bdas@aecom.yu.edu
(B.C. Das).
0960-894X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2011.03.050

3480
P. T. Jubinsky et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3479–3482
and Tim16 share 42% amino acid identity, and human Magmas is
able to functionally replace endogenous Tim16.8
Our small molecule library (Scheme 1) was synthesized by the
chemical modification of existing reactions.9 The chemical synthe-
sis of lead compound 9 is shown in (Scheme 2).
The fluorescence emission spectra (kexc = 295 nm, 2.5 nm slit
width, and 10 mm path length) of free or ligand bound Magmas
(yeast or mouse) were acquired using a FluroMax-3 spectrofluo-
rometer thermostated at 25 °C. The dissociation constants for com-
pound 9 were obtained from the fluorometric titrations of the
In brief, acid 16 was reacted with guanidine hydrochlo-
ride,1,10carbonyldiimidazole in DMF–dioxane mixture. The crude
reaction mixture was evaporated and the residue was purified by
fractional crystallization from methanol, to give compound 9 white
crystals mp 165–167 °C.10
A library of randomized mutations in yeast Magmas (TIM16)
were made by error-prone PCR. The genes encoding the mutations
were spliced into plasmid pRS315 and introduced into a haploid
yeast strain containing a wild-type copy of TIM16 on plasmid
pRS316 and a null genomic replaced by a HIS3 cassette. The plas-
mid containing the wild type (wt) TIM16 gene was lost by selection
in media containing FOA11 and two mutants were identified for
having altered sensitivity to SMMI (small molecule Magmas inhib-
itor) 9 in proliferation assays. The mutations in TIM16 were deter-
mined from DNA sequencing of the plasmid from each mutant
strain and were found to result in amino acid changes N84T and
A104S in Tim16.
S. cerevisiae strain J47D in log phase growth was plated in 96
well dishes at 1.5 ꢀ 104 cells/well in YEPD media. Magmas inhibi-
tors were diluted in DMSO and added at various concentrations
(range 0–1 mM) to each well in triplicate at a final DMSO concen-
tration of 1%. Samples were cultured at 30 °C in a Bioscreen C MBR
incubator (Growth Curves USA, Piscataway, NJ) and optical density
(OD) readings at 600 nM were measured every 30 min for the indi-
cated time in the plots. Cell proliferation data was analyzed and
plotted using Microsoft Excel. Standard deviations were smaller
than 2% of the mean OD value (n = 3) and each experiment was
performed at least two independent times.
yeast or mouse Magmas (3 lM) with compound 9 in 50 mM Bis–
Tris at pH 7.1. During titrations the dilution does not exceed 10%,
and all the fluorescence emissions spectra were corrected for dilu-
tion. Compound 9 stock solutions were prepared fresh prior to each
titration. The dissociation constants for the binding of compound 9
to yeast Magmas was determined by fitting the fluorometric titra-
tions data to Eq. 1, where F0, FSMMI and FSMMI1 are the intrinsic
fluorescence of free Magmas, compound 9 bound Magmas and
Magmas at saturating SMMI 9 concentration, respectively, [SMMI]
is the total ligand concentration, Kd is the dissociation constant of
Magmas–SSMI complex.
ꢀ
ꢁ
FFLB1½SMMIꢁ
Kd þ ½SMMIꢁ
FSMMI
¼
þ F0
ð1Þ
Structural modeling was performed using ICM-Molsoft LLC soft-
ware. The binding mode for compound 9 was predicted by system-
atic docking calculations with the ICM software accounting for
both ligand and protein flexibility.12 Docking of compound 9 to
individual Tim14 and Tim16 monomers the Tim14–Tim16 hetero-
dimer was modeled using data from PDB entry 2GUZ.5a
Reduced Magmas function caused by siRNA knockdown or by
mutations in Magmas causes growth inhibition and cell cycle ar-
rest. To assay whether the compounds were active we measured
their ability to inhibit yeast proliferation. Yeast is a good organism
to test SMMI activity because Magmas is fully functional in TIM16
null yeast and because the target region is identical across all
eukaryotic species.
The concentration dependent effects of the compounds on yeast
proliferation are shown in Table 1 (for detail, please see Supple-
mentary data) Compound 5 was inactive at all concentrations
while all the other compounds completely inhibited yeast prolifer-
ation at 100 lM. At 10 lM, 6 was partially inhibitory and 9 fully
inhibited growth. The lead compound 9, was further tested to
determine the lower range of activity (Fig. 1). Compound 9 at
3
l
M had a proliferation profile similar to that observed without
inhibitor. In contrast, significant and reproducible inhibition was
observed at 4 M, and resulted a lengthening of the time to half
l
maximal growth by 1.8 times.
To genetically demonstrate that effects of 9 were mediated
through Magmas, yeast strains containing several different Mag-
mas mutations expressed on a plasmid were screened for altered
compound 9 concentration dependent growth. All of these stains
express similar levels of Magmas protein, which is approximately
four-fold higher than the haploid control (wt genomic) (Short
and Jubinsky, unpublished data). Comparing the dose response of
wt genomic (TIM16) to wt plasmid (human Magmas) showed that
Scheme 1. Small focused molecular library (1–10).
Scheme 2. Synthesis of compound 9. Compound 9 was synthesized from acid 16.

P. T. Jubinsky et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3479–3482
3481
over-expression of Magmas reduces sensitivity of yeast to SMMI 9
(Supplementary Fig. 1). Yeast strains containing Magmas N84T
proliferate identically to wt plasmid without SMMI present, but
had an increased resistance to compound 9. In contrast, yeast
strains containing the Magmas A104S mutation are more sensitive
to the effects of SMMI 9.
We have also cultured cell lines with 9 and observed that the
morphological and biochemical features were similar to those in
cells with siRNA-mediated reduced Magmas expression, providing
additional evidence of target specificity (Roy and Jubinsky, unpub-
lished data).
cence properties of Magmas with 1 tryptophan and 5 tyrosine res-
idues, upon excitation at 295 nm showed emission maximum
centered at 365 nm and a shoulder at ꢂ315 nm, consistent with
the fluorescence emission maxima for tryptophan (typically at
340 nm) and tyrosine (typically at 310 nm), respectively. This
significant red-shift (15 and 5 nm for tryptophan and tyrosine
fluorescence, respectively) suggests solvent exposed Trp and Tyr
residues in Magmas.
The fluorometric titration of Magmas with compound 9 showed
linear decrease in both tryptophan and tyrosine intrinsic fluores-
cence intensity with a single isosbestic point at ꢂ420 nm, consis-
To test whether SMMI directly binds to Magmas compound 9,
the SMMI with the highest biological activity in vivo, was used in
a fluorometric titration of Magmas (Fig. 2). The intrinsic fluores-
tent with a single step binding process of SMMI to yeast
Magmas. The ꢂ4-fold quenching of the tryptophan and tyrosine
intrinsic fluorescence at saturating SMMI concentration suggest
polarity (hydrophobic) changes in the environment of Trp and
Tyr upon SMMI 9 binding to Magmas.11 A dissociation constant
(Kd) of 33.5 3.5 lM was determined for the binding of compound
9 to yeast magmas by fitting the titration data to Eq. 1 (Fig. 3, in-
Table 1
Inhibition of yeast proliferation by five different synthesized compounds
Compound
1 mM
100
lM
10
lM
1
lM
sert). The dissociation constant obtained from the binding of com-
pound 9 to murine Magmas was similar (36 lM).
5
4
6
8
9
ꢃ
i
+
i
ꢃ
+
+
+
+
ꢃ
ꢃ
ꢃ
ꢃ
ꢃ
ꢃ
ꢃ
Molecular modeling of SMMI docking to Tim14 and Tim16
monomers and the Tim14-Tim16 heterodimer showed that the
lowest energy binding resulted with the Tim14-Tim16 complex.14
Lowest energy binding was found with the Tim14–TIim16 com-
plex. Compound 9 was predicted to bind at the dimer interface
establishing multiple hydrophobic contacts with both Tim14
(L100, K111, L114, Q115, and T120) and Tim16 (F95) residues
(Fig. 3) and hydrogen bonding Tim14 (E121 and K168).
ꢃ
+
+
(+) Complete inhibition (no proliferation).
(ꢃ) No inhibition (normal proliferation).
( ) Partial inhibition.
(i) Compound insoluble.
Figure 3. 2-Dimentional-interaction diagram of predicted binding contacts
between compound and the Tim14–Tim16 heterodimer. (hydrophobic contact:
red; hydrogen bond: blue). Compound 9 establishes multiple hydrophobic contacts
with Tim16 (F95) and Tim14 (L100, K111, L114, Q115, and T120). L100 is located at
the Tim14 N-terminal arm. The guanidinium moiety forms hydrogen bonds with
E121 and K168. Residue numbers are those used in PDB entry 2GUZ.5a
Figure 1. Dose response of compound 9 on yeast proliferation. Yeast cells were
cultured in the indicated concentration of SMMI 9 and the optical density was
measured.
Figure 2. Compound
(kx = 295 nm) of yeast Magmas during the fluorometric titration of SMMI in a
concentration range between 0 (curve 1) and 79.36 M (curve 11). Inset, fluores-
cence emission intensity values at 365 as a function of SMMI 9. The curve is a fit of
the data to Eq. 1. Fluorescence emission spectra were recorded at protein
9 binding to Magmas. Fluorescence emission spectra
Figure 4. Model of SMMI 9 binding in the Tim14–Tim16 interface pocket. The
predicted Tim14–Tim16 binding pocket surface is color coded according to protein
binding properties (white: neutral surface; green: hydrophobic surface; red:
hydrogen bonding acceptor potential; blue: hydrogen bond donor potential;
brown; SMMI 9).
l
concentration of ꢂ3 lM in Bis–Tris (50 mM, pH 7.1, 25 °C).

3482
P. T. Jubinsky et al. / Bioorg. Med. Chem. Lett. 21 (2011) 3479–3482
5. (a) Mokranjac, D.; Bourenkov, G.; Hell, K.; Neupert, W.; Groll, M. EMBO J. 2006,
Steric and physicochemical complementarity between com-
25, 4675; (b) Kozany, C.; Mokranjac, D.; Sichting, M.; Neupert, W.; Hell, K. Nat.
Struct. Mol. Biol. 2004, 11, 234.
6. Jubinsky, P. T.; Short, M. K.; Mutema, G.; Morris, R. E.; Ciraolo, G. M.; Li, M. J.
Mol. Histol. 2005, 36, 69.
pound 9 and the Tim14–Tim16 heterodimer interface was ob-
served in the model (Fig. 4). Because the hydrophobic substituted
cyclohexyl appears to interact with residues (F95 and L100) that
stabilize the heterodimer through multiple van der Walls interac-
tions, it is predicted that compound 9 binding results in complex
dissociation. In addition, compound 9 likely interferes with the
N-terminal arm of Tim14 that interacts with Tim16 helix III, known
to be necessary for complex formation and function.5a
The binding mode prediction supports the importance of a
hydrophilic guanidinium moiety to correctly align the small mole-
cule at the dimer interface, along with hydrophobic groups at the
opposite end of the molecule to induce an alternative conformation
for the N-terminal arm of Tim14. Confirmation of this model by
NMR, Magmas mutagenesis and crystal structures are planned.
Several SMMIs that are biologically active in yeast were synthe-
sized. Over-expression of Magmas and the isolation of Magmas
mutants with both increased and decreased sensitivity to 9
showed that Magmas is a target in vivo, while the fluorometric
titration experiments demonstrated a direct physical interaction.
We also showed that compound 9 is predicted to functionally
interact with Magmas using computational modeling. SMMIs
should prove useful for studying mitochondrial function and may
have clinical relevance for a variety of human diseases. The effects
of SMMI on CNS malignancies are under evaluation, as are efforts
to develop agents that can monitor Magmas function in vivo.
7. (a) van de Vijver, M. J.; He, Y. D.; Van’t Veer, L. J.; Dai, H.; Hart, A. A.; Voskuil, D.
W.; Schreiber, G. J.; Peterse, J. L.; Roberts, C.; Marton, M. J.; Parrish, M.; Atsma,
D.; Witteveen, A.; Glas, A.; Delahaye, L.; van der Velde, T.; Bartelink, H.;
Rodenhuis, S.; Rutgers, E. T.; Friend, S. H.; Bernards, R. N. Engl. J. Med. 2002,
347, 1999; (b) Desmedt, C.; Piette, F.; Loi, S.; Wang, Y.; Lallemand, F.; Haibe-
Kains, B.; Viale, G.; Delorenzi, M.; Zhang, Y.; d’Assignies, M. S.; Bergh, J.;
Lidereau, R.; Ellis, P.; Harris, A. L.; Klijn, J. G.; Foekens, J. A.; Cardoso, F.; Piccart,
M. J.; Buyse, M.; Sotiriou, C. Clin. Cancer Res. 2007, 13, 3207; (c) Tagliati, F.;
Gentilin, E.; Buratto, M.; Mole, D.; degli Uberti, E. C.; Zatelli, M. C. Endocrinology
2010, 151, 4635.
8. (a) Sinha, D.; Joshi, N.; Chittoor, B.; Samji, P.; D’Silva, P. Human Mol. Genet. 2010,
19, 1248; (b) Elsner, S.; Simian, D.; Iosefson, O.; Marom, M.; Azem, A. Int. J. Mol.
Sci. 2009, 10, 2041.
9. (a) Shridhar, D. R.; Reddy Sastry, C. V.; Bansal, O. P.; Pulla Rao, P. P. Synthesis
1981, 19, 2; (b) Liu, F.; Evans, T.; Das, B. C. Tetrahedron Lett. 2008, 49, 1578.
10. Detail experimental procedures for compound (9): The syntheis of acid 16
involves the reaction with methyl magnesium bromide with b-cyclocitral in
THF to give the alcohol as a yellow oil.13 The alcohol gave satisfactory spectral
data and was directly converted to the triphenyl phosphine salt by treatment
with triphenylphosphine hydrobromide in methanol. Recrystalization of the
salt from methanol/ether (1:6) gave a yellow crystalline solid.14 Formation of
the Witting reagent from the salt in ether was accomplished with n-
butyllithium in hexane at room temperature (dark-red color). The Witting
reagent was treated with methyl 4-formybenoate in ether at ꢃ78 °C for 10–
15 min and then stirred at room temperature under a nitrogen atmosphere for
30 h. The crude ester was purified by flash column chromatography (hexane/
ethyl acetate: 98:2) to give a brown oil in 85% yield.15 The ester was saponified
to generate a white solid which was filtered, washed with water, and dried. The
product was recrystallized from hot ethanol and washed with dry hexane to
give acid as white crystals (87%) yield. The structure was confirmed by 1H, 13
NMR and NOE experiment, HMBC, and HRMS. mixture of acid 16
C
A
Acknowledgements
(300 mg;1.05 mmol) in dry DMF (4 mL) and CDI (171 mg;1.05 mmol) was
stirred at room temperature under nitrogen atmosphere for 1 h. Guanidine
base was prepared by consecutive additive of sodium-tert-butoxide (201.8 mg,
2.1 mmol) and guanidine hydrochloride (200 mg, 2.1 mmol) to a dry mixture of
dimethylformamide/dioxane (1:1, 8 mL) under nitrogen, this mixture was
heated to 50–55 °C for 20 min, and then the NaCl was filtered. The solution of
CDI and acid was added to the guanidine base solution. The mixture was stirred
at room temperature for 6 h. The progress of the reaction was monitored by
TLC. After the reaction was completed the mixture was evaporated and the
DMF was removed under vaccum. The residue suspended in cold water (8 mL).
The crystaline solid was filtered, washed with water, and dried, then purified
by fractional crystalization from methanol to give a white solid 340 mg (87%).
Mp: 165–167 °C. 1H NMR (300 MHz, acetone-d6): 7.90–7.82(d, J = 8.0 Hz, 2H),
7.40 (d, J = 8 Hz, 2H), 6.47(s, 1H);2.55 (m, 2H), 2.19 (q, 3H), 1.86 (s, 3H);1.44
(m, 2H), 1.05 (s, 6H) and 1.03 (t, J = 8 Hz, 3H) 13C NMR CDCl3: 177.2;163.8,
147.6, 141.5, 140.9, 137.2, 129.06, 128.7, 127.5, 121.6, 39.1, 35.9, 28.6,
This study was supported by the James B. Ax Family Founda-
tion. We thank E. Richard Stanley and Donna Alex for their contin-
ued encouragement and Ruben Abagyan and Macro Neves for their
assistance with the modeling of SMMI and Magmas.
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2011.03.050.
References and notes
27.5, 24.4, 22.9, 14.9 and 14.6 ESI-MS: Calcd for
326.2227; found: 326.2223.
C
20H27 N3O ([M+H]+)
1. Jubinsky, P. T.; Messer, A.; Bender, J.; Morris, R. E.; Ciraolo, G. M.; Witte, D. P.;
Hawley, R. G.; Short, M. K. Exp. Hematol. 2001, 29, 1392.
2. Jubinsky, P. T.; Short, M. K.; Mutema, G.; Witte, D. P. J. Histochem. Cytochem.
2003, 51, 585.
3. Peng, J.; Huang, C. H.; Short, M. K.; Jubinsky, P. T. In Silico Biol. 2005, 5, 251.
4. (a) Li, Y.; Dudek, J.; Guiard, B.; Pfanner, N.; Rehling, P.; Voos, W. J. Biol. Chem.
2004, 279, 38047; (b) D’Silva, P. R.; Schilke, B.; Walter, W.; Craig, E. A. Proc. Natl.
Acad. Sci. U.S.A. 2005, 102, 12419; (c) Frazier, A. E.; Dudek, J.; Guiard, B.; Voos,
W.; Li, Y.; Lind, M.; Meisinger, C.; Geissler, A.; Sickmann, A.; Meyer, H. E.;
Bilanchone, V.; Cumsky, M. G.; Truscott, K. N.; Pfanner, N.; Rehling, P. Nat.
Struct. Mol. Biol. 2004, 11, 226.
11. Boeke, J. D.; Trueheart, J.; Natsoulis, G.; Fink, G. R. Methods Enzymol. 1987, 154,
164.
12. Totrov, M.; Abagyan, R. Proteins 1997, Suppl 1, 215.
13. Fernandez-Mateos, A.; Mateos Buron, L.; Martin de la Nava, E. M.; Rubio
Gonzalez, R. J. Org. Chem. 2003, 68, 3585.
14. Rosenberger, M.; Neukom, C. J. Org. Chem. 1982, 47, 1782.
15. Waugh, K. M.; Berlin, K. D.; Ford, W. T.; Holt, E. M.; Carrol, J. P.; Schomber, P. R.;
Thompson, M. D.; Schiff, L. J. J. Med. Chem. 1985, 28, 116.