Cell-specific calcineurin inhibition by a modified cyclosporin

Full text of DOI:10.1021/ja9636146

J. Am. Chem. Soc. 1997, 119, 1805-1806
1805
Cell-Specific Calcineurin Inhibition by a Modified
Cyclosporin
Peter J. Belshaw and Stuart L. Schreiber*
Howard Hughes Medical Institute
Department of Chemistry and Chemical Biology
HarVard UniVersity, Cambridge, Massachusetts, 02138
ReceiVed October 16, 1996
The immunosuppressive drug cyclosporin A (CsA) is widely
used to suppress graft rejection following transplantation in
humans. Its cellular effects are mediated by a complex of CsA
and its intracellular, soluble receptor cyclophilin (Cyp).1 The
Cyp-CsA complex is a specific inhibitor of the protein
phosphatase calcineurin (Cn).2 Consequently, CsA has also
been used to illuminate the cellular functions of Cn. For
example, its use revealed that, in T cells, Cn mediates the nuclear
translocation and activation of NFAT,3 a transcription factor
essential for T cell activation. As a therapeutic agent, CsA has
several undesirable side effects including hypertension, CNS
toxicity, and nephrotoxicity. These side effects result from CsA-
mediated inhibition of Cn in cells outside the immune system.4
We now report the synthesis of R-cyclopentylsarcosine11-
CsA (CsA*), a rationally modified CsA that does not bind to
Cyp and consequently does not inhibit Cn in cells. We also
describe a modified Cyp with compensatory mutations in its
CsA-binding pocket (F113G, C115M, S99T; Cyp*) that promote
high affinity complexation with CsA*. The resulting Cyp*-
CsA* complex presents a composite surface that binds Cn with
high affinity. As a result, Cn inhibition by CsA* is restored in
cells expressing Cyp*. These results provide a method to
control the inhibition of Cn spatially and temporally in animals
by targeting the expression of Cyp* to specific cells and tissues.
Modifying receptors or ligands by adding even a few atoms
can abolish their binding, often changing the function of the
molecules involved. When these loss-of-interaction substituents
are introduced into proteins by mutation5 and small molecules
by synthesis,6 information about the cellular function of interact-
ing molecules can be obtained. To achieve our goal of cell-
specific Cn inhibition, it was essential that the modified
receptor-ligand complex have a composite surface suited for
Cn binding, since altered Cyp-CsA complexes can be envi-
sioned that will no longer bind Cn (lower right schematic in
Figure 1).
Figure 1. Strategy for producing selective inhibitors of calcineurin.
A modified cyclosporin with additional atoms added to its cyclophilin
binding surface abolishes its ability to bind to both cyclophilin and
calcineurin. The cyclosporin binding surface of cyclophilin is re-
engineered to have a complementary shape for the modified cyclosporin,
allowing binding to the modified cyclophilin. The resulting complex
may retain the ability to bind and inhibit calcineurin only if its composite
surface is similar to that of the native Cyp-CsA complex (lower left
vs lower right schematic).
We designed a variant of CsA having two additional
methylenes (CH2) attached to the two methyl groups of the
MeVal11 side chain, resulting in the cyclopentyl side chain of
CsA*. The crystal structure of the CypA-CsA complex7 shows
residue 11 of CsA directly contacting Cyp, binding in a deep
hydrophobic pocket in the active site of Cyp (Figure 2A). The
additional atoms were expected to reduce the binding of CsA*
to Cyp significantly, presumably through steric interactions
Figure 2. Graphical representation of the binding interfaces between
Cyp, CsA, and variants. The solid blue surface and white mesh represent
the solvent accessible surfaces of CsA and Cyp, respectively, (nonpolar
hydrogens omitted). The contact side chains of Cyp are represented as
tubular bonds and viewed from the inside of the protein, the lowest
protrusion of solid blue corresponds to the side chain of residue 11 of
CsA. (A) Cyp-CsA crystal structure. (B) Hypothetical model of Cyp-
(S99T, F113G, C115M)-CpSar11-CsA (Cyp*-CsA*) complex.10
(1) Handschumacher, R. E.; Harding, M. W.; Rice, J.; Drugge, R. J.;
Speicher, D. W. Science 1984, 226, 544-547.
between the cyclopentyl side chain and the receptor. To select
possible receptors capable of binding CsA*, we generated
computer models of complexes between CsA* and several Cyp
mutants. On the basis of these models, we selected three
mutations in residues lining the valine11 binding pocket of Cyp,
one to remove the offending steric interaction (F113G) and two
others (S99T, C115M) to improve the fit between the new
receptor and ligand (Figure 2B). The synthesis of CsA* first
required the asymmetric synthesis of R-cyclopentylsarcosine
using a combination of the methods of Evans8 and Dorrow9
and the subsequent incorporation of this non-natural amino acid
(2) Liu, J.; Farmer, J. J.; Lane, W. S.; Friedman, J.; Weissman, I.;
Schreiber, S. L. Cell 1991, 66, 807-815.
(3) Northrop, J. P.; Ullman, K. S.; Crabtree, G. R. J. Biol. Chem. 1993,
268, 2917-2923.
(4) Lyson, T.; Ermel, L. D.; Belshaw, P. J.; Alberg, D. G.; Schreiber, S.
L.; Victor, R. G. Circ. Res. 1993, 73, 596-602.
(5) Brown, E. J.; Beal, P. A.; Keith, C. T.; Chen, J.; Shin, T. B.; Schreiber,
S. L. Nature 1995, 377, 441-446.
(6) Liu, J.; Albers, M. W.; Wandless, T. J.; Luan, S.; Alberg, D. G.;
Belshaw, P. J.; Cohen, P.; MacKintosh, C.; Klee, C. B.; Schreiber, S. L.
Biochemistry 1992, 31, 3896-3901.
(7) Ke, H.; Mayrose, D.; Belshaw, P. J.; Alberg, D. G.; Schreiber, S. L.;
Chang, Z. Y.; Etzkorn, F. A.; Ho, S.; Walsh, C. T. Structure 1994, 2, 33-
44.
S0002-7863(96)03614-1 CCC: $14.00 © 1997 American Chemical Society

1806 J. Am. Chem. Soc., Vol. 119, No. 7, 1997
Communications to the Editor
Figure 4. Cellular assay for Cn-mediated NFAT-signaling.13 CpSar11-
CsA has little effect on NFAT-signaling even at the highest concentra-
tions tested yet potently inhibits signaling in human Jurkat-TAg T cells
expressing Cyp(S99T, F113G, C115M) (Cyp*). In this assay, CsA has
an IC50 of 15 nM in cells transfected with NFAT-SEAP alone (data
not shown).
MeIle11-CsA complexes and therefore inhibition of Cn. The
bulkier CpSar11-CsA used here provides greater selectivity
for inhibiting Cn only in cells expressing Cyp*.
Cyclophilins are expressed ubiquitously and have been found
in virtually all organisms and tissues. Cn is expressed in many
tissues with the highest levels of expression occurring in the
brain.11 The roles of Cn in cells outside the immune system
are not known with certainty. The ability to inhibit Cn in a
tissue-specific fashion, for example, through the expression of
Cyp* under the control of a tissue-specific promoter in
transgenic animals, may provide a method to dissect the many
roles of Cn in different cells types. Indeed, the use of a T cell
specific promoter in a transgenic mouse was recently used to
direct the expression of a designed protein that is activated by
a synthetic ligand.12 Since cyclosporin affects many cell subsets
within the immune system, the question of whether inhibition
of Cn in T cells or other calcineurin-dependent immune cells
alone (e.g., B cells, mast cells, neutrophils) is sufficient to
prevent rejection of transplanted organs or ameliorate autoim-
mune disorders has particular relevance to the development of
new immunosuppressants. This technique may also find ap-
plications in human gene therapy, for example, in the treatment
of graft-vs-host disease following bone marrow transplantation.
In this scenario, Cyp* would be expressed in the transplanted
bone marrow by gene transfer ex vivo so that CsA* could be
used tissue specifically in vivo. Since many signaling proteins
are expressed ubiquitously yet have context-dependent roles in
signaling, this approach to cell-specific targeting may provide
a widely useful experimental window to study cells and tissue
subsets within an organism.
Figure 3. Synthesis of CpSar11-CsA. H2N-nonapeptide (H2N-Abu-
Sar-MeLeu-Val-MeLeu-Ala-DAla-MeLeu-MeLeu-Ac-N-MeValinol es-
ter) was synthesized from CsA as described previously.10 Detailed
experimental protocols and spectra are available from the authors upon
request.
into the CsA macrocycle10 (Figure 3). To determine the binding
characteristics of CsA*, we purified recombinant CypA and
Cyp(S99T, F113G, C115M) (Cyp*) expressed in Escherichia
coli and measured the binding constants for CsA* with a
fluorescence-binding assay.10 As anticipated, CsA* has little
affinity for wild-type Cyp (Kd > 15 µM) and has a high affinity
for the mutant Cyp* (Kd ) 9 nM).
To test whether the new receptor-ligand combination is able
to inhibit the cellular function of Cn, we used a cellular assay
for NFAT-signaling6 where human T cells (Jurkat) were
transiently transfected with a reporter gene responsive to the
transcription factor NFAT either alone or in combination with
Cyp or Cyp*. Stimulation of these cells with phorbol ester
(PMA) and ionomycin mimics signaling through the T cell
receptor and activates NFAT by a Cn-dependent pathway. As
anticipated, CsA* has little or no effect on reporter gene
expression at concentrations up to 400 nM in cells expressing
either endogenous cyclophilins alone or coexpressing recom-
binant, wild-type cyclophilin, and yet in cells expressing Cyp*,
CsA* potently inhibits NFAT signaling with IC50 ) 25 nM
(Figure 4). Thus, we have made cells conditionally sensitive
to a small molecule ligand, dependent on the expression of a
dominant allele of its receptor protein.
Acknowledgment. We thank Joseph Schoepfer for previous
synthetic endeavors that inspired the work presented here. We thank
the National Institute of General Medical Sciences for support of this
research and the NSERC for a predoctoral fellowship to P.J.B. S.L.S.
is an Investigator at the Howard Hughes Medical Institute.
In a previous study we engineered a similar Cyp-CsA-based
receptor-ligand pair: Cyp(S99T, F113A)-MeIle11-CsA.10 In
this NFAT-signaling assay, we found that although MeIle11-
CsA potently inhibits NFAT-signaling in cells expressing CypA-
(S99T, F113A), MeIle11-CsA still inhibited NFAT-signaling
in cells transfected with wild-type CypA (IC50 ≈ 300 nM, data
not shown). Presumably, this is due to the high concentration
CypA ([CypA] > Kd), which promotes formation of the Cyp-
JA9636146
(11) Klee, C. B.; Krinks, M. H.; Manalan, A. S.; Cohen, P.; Stewart, A.
A. Methods Enzymol. 1983, 102, 227-244.
(12) Spencer, D. M.; Belshaw, P. J.; Chen, L.; Ho, S. N.; Randazzo, F.;
Crabtree, G. R.; Schreiber, S. L. Curr. Biol. 1996, 6, 839-847.
(13) CypA and mutants were subcloned from pGEX-based expression
vectors into pBJ5, a eukaryotic expression vector containing an N-terminal
FLAG epitope tag. The signaling assay was performed essentially as
previously described.6 The data for each transfection are presented as percent
SEAP activity relative to the signal for [CpSar11-CsA] ) 0 (triplicate
readings).
(8) Evans, D. A.; Britton, T. C.; Ellman, J. A.; Dorow, R. L. J. Am.
Chem. Soc. 1990, 112, 4011-4030.
(9) Dorow, R. L.; Gingrich, D. E. J. Org. Chem. 1995, 60, 4986-4987.
(10) Belshaw, P. J.; Schoepfer, J. G.; Liu, K.-Q.; Morrison, K. L.;
Schreiber, S. L. Angew. Chem., Int. Ed. Engl. 1995, 34, 2129-2132.