oo4o-4020192 $3.oo+.Ml Pcrgamon Press Lid
Tetrahedron Vol. 48. No. 13, pp. 2545-2558.1992 l’riiti in Great Britain
Molecular Recognition of Immunophilins and Immunophilin-Ligand Complexes Stuart L. Schreiber,* Jun Liu, Mark W. Albers, Michael K. Rosen, Robert F. Standaert, Thomas J. Wandless, Patricia K. Somers Department of Chemistry, Harvard University 12 Ogord Street, Cambridge, iUassachusetts 02138
Abstract Immunophilin-ligand complexes have been used to identify a previously unknown step in Ca*+dependent signal ttansduction pathways. This Report, which we dedicate to Professor Harry I-I.Waaaerman, describes structural and mechanistic aspects of immunophilin research. In addition to their important medical applications, the immunosuppressants FK506, cyclosporin A, and rapamycin have proven to be valuable reagents for studying intracellular signal transduction mechanisms.1 Signal transduction refers to the prccess by which extracellular molecules ate able to influence intracellular processes. In
transcription Figure 1
Three cellular systems that am sensitive to immunophilii-ligand complexes.
recent years, much progress has been achieved in our understanding of the mechanisms of signaling at the membrane and within the nucleus of the cell. In contrast, very little is known about the mechanisms by which signals are transduced through the cytoplasm of the cell, which has been referred to as the “black box” of the signal transduction pathway.
Three systems that are sensitive to these immunosuppressants and thus have heen studied in some detail are depicted in Figure 1. Transcription of specific genes following stimulation of the T-cell receptor has heen studied in the T-cell2fP
whereas the exocytosis of secretory vesicles following stimulation of the IgE receptor has been
investigated in the mast cell.5 In addition, several events, including death and proliferation, have been examined in yeast.617 These investigations have utilized complexes of immunophilins (immunosuppressant binding proteins) and their immunosuppressive
ligands (Figure 2), which interfere with signal transduction
pathways in the
cytoplasm of the cell, following early membrane-associated events. Therefore, they serve as a window for the events that constitute cytoplasmic signaling. In this Report we describe the structural basis of immunophilii-ligand complexation and the identification of target molecules of these complexes.
Immunophilin-ligand complexes serve as probes of cytoplasmic signaling mechanisms.
The immunophilin ligands that have been utilized in our studies are shown in Figure 3. These include the cyclic peptide, cyclosporin A (CsA), the macrolides FK506 and rapamycin, and the nonnatural immunophilin ligand, 506BD.a
There exist two families of immunophilins:
the cyclophilinsg, which bind CsA, and the
FKBPs*oJl, which bind FK506 and rapamycin (Figure 4). Many of the immunophilins have catalytic properties. These enzymes catalyze the interconversion of the cis- and rr~~rotamers of a peptidyl-prolyl amide bond in peptide and protein substrates.
These rotamase enzymes are potently inhibited by their cognate immunosuppressive
ligand(s). Although a number of cyclophilin and FKBP family members have recently been characterized~~1u~~2, the focus here will be on the predominant cyclophilin, cyclophilin A, and the predominant FKBP, FKBPlZ
Molecular recognition of immunophilins
Structures of high-affinity immunophilin-ligands.
Important simihuities and differences exist in the actions of these immunosuppressants (Figure 5). Despite their dissimilar structures, CsA and IX506 interfere with a common set of signaling pathways.**4l5 These are Ca*+-dependent pathways that emanate from, for example, the T cell receptor in T lymophcytes and the IgE receptor in mast cells. Evidence has been gathered that suggests CsA and FK506 interfere with the same step in these signaling pathways and by a similar mechanism. Rapamycin, despite its structural similarity with FK506, interferes with a distinct set of C!a*+-independent signaling pathways. These include pathways emanating from lymphokine receptors, such as the interleuldn-2 receptor in T cells, and growth factor receptors in hepatocytes.t3 0
ro tamase w N
&GM Kt(FK506) = 0.4 nM K&WA) = 0.2 nM
cyclophilins CyP A Science 226,544 (1984). CyP B PNA.588, 1903 (1991). CyP C Cell 66.798 (1991). ninaA CeU65, 219 (1991).
FKBPs FKBPl2 FKBP13 FKBP25 FKBP60
Nurure 341,755 (1989) Ibid.. p. 758. PNAS88.6677 (1991). A. Galat et al.. submitted. J. Am. Chem. Sot. 113.758 (1991).
Fig 4 Two families of immunophilins. The Kt value for CsA corresponds to a measurement with cyclophilin A whereas the K, values for IX506 and rapamycin correspond measurements with FKBPIZ.
L. SCHREIRERet al.
IL-2R . l
FK!M and CsA inhibit Ca*+dependent independent signaling pathways.
signaling pathways whereas rapamycin interferes with Ca*+-
The rotamase activity of immunophilins led to an initial model of drug action whereby drug binding to the immunophilin pmduces a loss of function (Figure 6). Rotamases were envisioned as key components of signaling pathways, and their inhibition by drug binding would arrest the signal. In the past several years much evidence has accumulated that argues against the “rotamase” model of drug action. 3*4J.* Instead, it appears that the binding of drug to immunophilin results in a gain of function. l In the “active-complex” model of drug action, complexes of cyclophilin-CsA and FKRP-FIB06 interfete with @+-dependent
signaling pathways and the complex of FKFJP-
rapamycin interferes with Ca*+-independent signaling pathways. A provocative aspect of this model is that the same FKRP could be responsible for forming two inhibitory complexes with different biological activities. (It will Rotamase Model: Drug binding results in a loss in function of the immunophiliu
Signal Transduction Pathway
Complex now results
Signal Transduction Pathway
Model: Drug binding in a gain in function
Ca*+-Dependent Signaling Figure 6
Immunophilin-ligand complexation results in a gain of function.
Molecular recognition of immunophilins
be interesting to learn if a cyclophilin ligand exists that forms an inhibitory complex akin to the FKBP-rapamycin complex.) The active-complex model does not specify a role for unligated immunophilin--this is an active area of investigation. A rationale for the ability of PKBP to give rise to two distinct inhibitory complexes was suggested by the molecular structures of FK506 and rapamycin (Figure 7).l The common strucmral elements found within these macrolides were suggested to
Following binding to the presenting
molecule, FKBP, the ditferent effector elements (depicted in the gray circles) of FK506 and rapamycin were envisioned as being responsible for the selectivity exhibited by the immunosuppressants.
Evidence in support of
this view was provided by experiments centered on a molecule we have named 506BD. which stands for “the FKBP binding domain of FK506 ” .8 These studies were instrumental in distinguishing between the two biological models shown in Figure 6.
FK506 inhibitor of Ca2’-dependent
rapamycin inhibits FK506 actions by competitively binding to FKBP
P inhibitor of Ca +-independent signaling
FK506 inhibits rapamycin actions by competitively binding to PKBP
Figure 7 FE506 and rapamycin are comprised of common FKBP-binding domains and distinct effector elements (shown in the shaded circles). 506BD was designed to contain the structural elements in lX506 and rapamycin that are responsible for binding to FKBP, yet to lack the effector elements of either agent. This was achieved by bridging two atoms in FK506 that are at the intersection of the binding and effector domains (Figure 8). The indicated trartS-enoate was added in place of the allyl-substituted effector loop found in PK506. According to the active-complex model, 506BD should be able to block the ability of FK506 (but not CsA) to inhibit Ca*+-dependent signaling pathways and rapamycin to inhibit Ca*+-independent signaling pathways, by competing with FK506 and rapamycin for binding to FKBP. Since 506BD lacks the effector elements of either macrolide. the active-complex model predicts that 506BD would not interfere with either type of signal transmission pathway. A very different outcome would result if the mtamase model was operative--the ability of 506BD to bind to FKPP should result in the inhibition of rotamase activity and, thus, of signal transmission.
506BD was synthesized (longest linear sequence = 27 steps)
and found to be a potent inhibitor of the rotsmase activity of FKBP (inhibitory constant Ki = 5 no).* The analysis of the cell signaling inhibitory properties of 506BD and its effects on CsA, FK506, and rapamycin led to a clear
S. L. SCHK~ER~~U~.
= 5 nM
” PK506/506BD overlayFigure 8
A ttDns-enoate in the FKBP ligand 506BD was used to replace the effector element of FK506.
distinction between the two biological models. 506BD was found to exhibit each of the properties described above that are supportive of the active-complex model. Following studies of 506BD in T [email protected]
, these findings were extended in investigations of exocytosis in mast cells. 5 Further evidence in support of the active-complex model has been gathered in experiments in lower eukaryotes as well. G. Livi and co-workers6 showed that the deletion of FRRP12 from yeast, which are normally sensitive to nanomolar concentrations of rapamycin, results in a rapamycin-resistant strain (Figure 9). Thus, rapamycin is intrinsically inactive in yeast and is best viewed as a prodrug.
It is only after the FKEtP-rapamycin complex forms, which occurs in the mutant strain following
expression of yeast or human FRBP12 and treatment with rapamycin, that the actions of rapamycin are seen. Related findings have been obtained with cyclophilin-CsA by M. Tropschug and co-workersI
by M. N. Hall and co-workers.7
transfection yeast FKBPl2
human FKBPl2 w
WT: rapamycin sensitive (nM) Figure 9
Rapamycin Sensitive Strains @Ml
rapamycin resistant mutants
The “active-complex” model is operative in Saccharomyces cerevisiae, where it has been shown that FKBP 12 mediates the actions of rapamycin.
Molecular recognition of immunophilins
The diagram in Figure 10 summarixes our current thinking in this area. The actions of CsA and FK506 are mediated by distinct receptors (immunophilins), a cyclophilin and an FKBP, respectively. The roles of individual cyclophilin and FKBP family members have not yet been defined in most instances.
The differing effects of
rapamycin and 506BD on the actions of CsA and FK506 support a role for the individual immunophilin-drug complexes. Because rapamycin and 506BD can compete with FK506 for binding to FKBP, they block the actions of FK506 in T cells and mast cells. As mpamycin and 506BD do not bind to cyclophilin, they do not block GA’s actions. Aside from their different sensitivities to rapamycin and 506BD, CsA and FK506 have nearly identical biological properties. This would suggest that they eventually act upon a common target, which might be associated with the Ca2+-dependent immunophilin-drug
signaling pathways in T cells and mast cells.
complexes interact directly with the common target.
is that the two
Other properties expected of this
hypothetical molecule are that it should not interact with immunophilin or drug alone, or with the complex of FKBP-rapamycin.
This latter complex should interact with a distinct target(s) associated with Ca2+-independent
signaling pathways. TCR-mediated
/** 0 CYP
mediated Figure 10
Mechanistic studies suggest that the cyelophilin-CsA and FKBP-FK506 complexes act upon a common target molecule. One possibility is that this target binds directl to the two immune hilin-drug complexes (but m drug or imgmmophilin slone, or thel43 P-rapamycin camp Pex, which intcrac~ with a distinct target that is aaaoeiated with Caz+-independent signaling pathways . The indicated target may be iuvolved in a step in signal transduction that is common to b 2+dq&dcnt sigruding pathways that iriclude those emanating from the T cell receptor in T cells and the IgE receptor m mast cells.
To test this possibility, several reagents for use in affinity chromatography were prepamd (Figure 11). The fusion proteins glutathione-S-transferasePKBP12
A were expressed in
E. coli in our laboratoryl5, while the cyclophilin C fusion protein was provided to us by Jeff Friedman and Irv Weissman at StanfonLt6 These researchers had found that a 77 kDa protein from a stromal cell lime associates with cyclophilin C and that a 55 kDa protein associates with the cyclophilin C-CsA complex. cyclophilin-CsA
We found that the
complexes retain the complex of the Caz+, calmodulin-dependent
serine,Ahreonine protein phosphatase calcineurin (also named PP2B; this hetemdimeric enzyme is comprised of a 6lkDa chain (cakk.urin-A) and a 19 kDa chain (calcincurin-B)) and calm&din when treatedwith calf thymus and
S. L. SCHREIBER~~UI.
Figure 11 Five affinity reagents that were used in experiments aimed at identifying the common targetof cyclophilin-CsA and FKBP-FKSO6complexes.
Gly-Arg-Phe-Asp-ArpArg-ValHN CO-Val-Ala-Ala-Glu-COzH H’Y xo32p032Ci+&M, calcineurin 1 (+/- additive) H,N-Arg-Leu-Asp-Val-Pm-Ile-ProGly-Arg-Phe-Asp-Arg-Arg-ValHN CO-Val-Ala-Ala-Glu-COzH +
Figure 12 Specific immunophilindmg complexes inhibit the phosphataseactivity of the Cab, calmodulindependent protein phosphatase calcineurin (PP2B) when assayed with a phosphopeptide substrate. Conditions are &scribed in refmnce 15.
Molecular recognition of immunophilins
calf brain tissue extracts. Soluble cyclophilin-CsA and FKBP-PKXK complexes bid competitively to calcineurin, indicating their binding sites are similar or overlapping. In accord with a role for calcineurin as the cellular target of both CsA and FK506, neither unligated immunophilins or drugs nor the FKRP-rapamycin complex bind to the phosphatase.
Specificity is clearly evident in a study of the effects of various combinations
of drugs and
immunophilins on the phosphatase activity of calcineurin when assayed with a labeled phosphopeptide substrate (Figure 12).l5 The specificity for calcineurin has been verified in studies of immunophilins complexes with three additional cytoplasmic serine/threonine
and their drug
protein phosphatases, PPl, PP2A, and PP2C
(Unpublished results in collaboration with Philip Cohen and Carol Mackintosh). Although our experiments are limited to in virro protein binding and enzyme inhibition assays, the remarkable interactions between immunophilindrug complexes and calcineurin (Figure 13) suggest that it is an excellent candidate for the physiologically relevant
Suggested result of complex formation is interference of Ca*+dependent signaling pathways Figure 13 Two immunophilin-drug complexes bind with high affiiity to calcineurin-calmodulin to form pentarm& complexes. Based upon our in vitro biochemical experiments, we hypothesize that the protein phosphatase calcineurin is the physiological target of cyclophilin-CsA and FKRP-FK506 complexes. common target of CsA and FK506. Several plausible models are now under consideration that invoke a role for this calcium-dependent phosphatase in signal transmission pathways characterized by an initial rise in intracellular Ca2+ levels. To distinguish among these, it will be necessary to define the cellular substrates of calcineurin. Recent fmdings by Crabtree and co-workers suggest that the function of the cytosolic component of the transcription factor NF-AT (NF-ATc)l7 may be either directly or indirectly linked to the cellular actions of calcineurin.
mechanistic overview is provided in Figure 14. Activation of the T cell receptor (TCR) in T cells or the IgR receptor (IgER) in mast cells leads to a rise in the concentration of intracellular Ca2+. We propose this results in the activation of calcineurin, which either directly or indirectly activates transcription factors, such as NFAT and OAP, that bind to the enhancer of the interleukin-2 (IL-2) gene. Whereas cyclophilin-CsA and FKRP-PK506 modulate the phosphatase activity of calcineurin, FKRP-rapamycin appears to modulate the activity of a yet-to-be identified target molecule that is a component of Ca2+-independent signaling pathways. CsA and PK506 therefore behave as a “molecular glue”--they bring together two normally non-interacting proteins, their cognate immunophilin and the protein phosphatase calcineurin. The structural dissimilatity between the immunosuppressants
and the unrelated sequences of cyclophilins and FKBPs make this result all the more
remarkable. In order to understand the basis for molecular recognition within these pentameric complexes (Figure
S. L. SCHREIBER er al.
Figure 14 A mechanistic scheme for the actions of immunophilin-drug and @+-independent signal transduction pathways.
hydrophobic substrate biding groove
complexes on &+-dependent
phosphatase active site
based on struchm of PKA
small test substrate: complexation + upregulation
Figure 15 The ability of immunophilii-drug complexes to increase the rate of dephosphorylation of a small test reagent while potently inhibiting tide substrate suggests that inhibitor binding occurs to phosihog whc IS latent to the enzyme active site.
Molecular recognition of immunophilins
13), detailed structural analyses will be required.
A clue relating to the general nature of immunophilin-
drug/calcinemin complexation has been obtained from additional experiments with phosphotylated substrates of calcineurin. We fii
that a specific effect of the cyclophilin-CsA and FKBP-FK506 complexes on the phosphatase
activity of calcineurin is also seen with the small, nonpeptide substrate p-nitrophenyl phosphate. In this case, however, the immunophilindrug
complexes accelerate the rate of dephosphorylation.
We interpret these results
with the structural model shown in Figure 15. In analogy to the structure of the protein kinase PKA~~, calcineurin may contain a hydrophobic substrate binding groove adjacent to the enzyme active site. The x-ray structure of a PKA-peptide inhibitor complex reveals that the peptide binds to the substrate-binding groove through hydrophobic contacts with an N-terminal amphipathic helix, whereas its unstructured C-terminus occupies the active site. We suggest the immunophilin-drug
complexes bind to a related specificity-determining
site, as opposed to the active
site, on calcineurin. Since the small reagent presumably interacts only with active site residues on calcineurin, it is able to coexist with the immunophilin-drug complex. Although we have not achieved a detailed understanding of the structure of pentameric complexes, new insights have been gained into immunophilin-ligand complexation, particularly as it pertains to FKBP12. A stereo view of the solution conformation of human FKBPIZ, which we determined by protein NMR method&m,
provided in Figure 16. The protein is seen to consist of a twisted (right-handed) and strongly curled five-stranded antiparallel B-sheet that surrounds a short a-helix. The drug binding site consists of an array of highly conserved aromatic residues. FK506 binds to this site through contacts with the structural elements that are common to both FK506 and rapamycin (what was correctly conjectured to be the FKBP-binding
domain of PK506). These
structural insights were made possible by our ongoing collaboration with Professor Jon Clardy (Cornell University, Chemistry Department). The Clardy group has succeeded in determining the high resolution x-ray structures of both the human FKBP12PK506** and human FKBP12-rapamycin** complexes. The stereo view of the former structure (Figure 17) reveals the composite surface that serves as the calcineurin-binding
element of the complex.
As was anticipated, the effector element of FK506 is exposed to the aqueous environment in the x-ray structure, and therefore is likely to be intimately involved in the binding to calcineurin. unbound and bound FKBP is still ongoing, the effect of FKBP-binding
Whereas a detai!ed comparison of on the conformation of FK506 was
immediately apparent. The conformational changes in the x-ray structures of the unbound (Scheme 18)*3 and bound (Scheme lo)*1 drug are manifold.
A similar observation has recently been recorded in studies of the
conformation of CsA when bound to cyclophilin A. %25 As is readily apparent, them is little relationship between the unbound (Figure 20)*6 and bound (Figure 21)%25 conformations of CsA. In light of their competitive binding to calcineurin, it will be important to compare the structures of immunophilin-drug complexes once the structure of the cyclophilin-CsA complex has heen determined. In summary, the natural products CsA. FK506, and rapamycin are prodrugs (or proinhibitors). Binding to their cognate immunophilin
in viva induces a gain of function and results in the formation of the true drug
(inhibitory) entity. In the case of cyclophilin-CsA and FKBP-FK506, this gain is suggested to correspond to a modulation of the phosphatase activity of calcineurin (PPZB), which occurs through the formation of pentameric complexes. In the future, a number of important issues will be addressed. Some of the most pressing questions are listed:
S. L. SCHRE~ER~~U~.
Figure 16 Stereo view of the solution conformation of humau FKBP12 determinedby NOErestrainedMD.19
Figure17 StereoviewoftheX-raystructureofthe human FKBP12-FK506 complex determined at 1.7 A resolution.2(J
Figure 18 Stereo view of the conformation of unbound FK506.
Figure 19 Stereo view of the conformation of bound FK506.
Molecular recognition of immunophilins
Figure 20 Stereo view of the conformation of unbound CsA.26
Figure 21 Stemo view of the conformation of bound CSA.~~S
1. Is calcineurin the physiologically relevant target of the actions of CsA and PK506? 2. What is the cellular role of calcineurin (e.g. in the Go + Gt transformation of the cell cycle)? 3. What am the cellular substrates of calcineurin? How close are we to identifying each of the molecules associated with the black box of a signal transduction pathway? 4. What is the target of the FKRP-rapamycin complex? (Homologous protein phosphatase family members exist that act, like rapamycin, in Gt.) 5. Am the prodrugs mimics of an endogenous cellular regulator that also bridges immunophilins and Cal&em-in? 6. What is the structural basis for pentamer formation? Can this knowledge be put to use in structure-based drug design efforts? 7. Can calcineurin he used as a tool for the discovery of new immunosuppressants that modulate phosphatase activity by directly binding to calcineurin? These questions illustrate a number of exciting challenges that will keep those of us involved in this field of endeavor quite busy for some time to come. With a continuation of the current pace of immunophilin research, it may not take long to obtain answers to many of these questions.27 Acknowledgements
Immunophilin research was supported by grants from the National Institutes of General
Medical Sciences (GM-38627 and GM-40660, awarded to S. L. S.). J. L. is a Damon Runyon-Walter Winchell Cancer Fund Fellow (DRG-1115), P. K. S. is the recipient of an American Cancer Society Award, M. W. A. is a Howard Hughes Medical Institute Predoctoral Fellow, and M. K. R. and T. J. W. are National Science Foundation Predoctoral Fellows. M. K. R. is the recipient of an American Chemical Society Division of Organic Chemistry Graduate Fellowship sponsored by Merck, Sharp & Dohme.
S. L. SCHREIBER et al.
Schreiber, S. L. Science 1991,251,283.
Emmel, I?. A.; Venveij, C. L.; Durand, D. B.; Higgins, K. M.; Lacy, E.; Crabtree, G. R. Science 1989,246,
Dumont, F. J.; Staruch, M. J.; Koprak, S. L.; Melino, M. R.; Sigal, N. H. J. Immunol. 1990,144,251.
Bierer, B. E.; Manila, P. S.; Standaert, R. F.; Herzenberg. L. A.; Burakoff, S. J.; Crabtree, G. R.; Schreiber,
Hultsch, T.; Albers, M. W.; Schreiber, S. L.; Hohman. R. J. Proc. Nati. Acad. Sci. USA. 1991.88, 6229.
Koltin, Y.; Faucette. L.; Bergsma, D. J.; Levy, M A.; Cafferkey, R.; Koser, P. L.; Johnson, R. K.; Livi, G.
S. L. Proc. Natl. Acad. Sci. USA. 1990.87, 9231.
P. Mol. Cell. Biol. 1991,11,
1718. See also: Heitman, J.; Mowa, N. R.; Hall, M. N. Science 1991,253,
Heitman, J.; Mowa, N. R.; Hiestand. P. C.; Hall, M. N. Proc. Nufl. Acud. Sci. U.SA. 1991,88, 1948.
Bierer, B. E.; Somers, P. K.; Wandless, T. J.; Burakoff, S. J.; Schreiber, S. L. Science 1990,250. 556.
Handschumacher, R. E.; Harding, M. W.; Rice, J.; Drugge, R. J.; Speicher, D. W. Science 1984,226,544.
10. Harding, M. W.; Galat, A.; Uehling, D. E.; Schreiber, S. L. Nature 1989,341,758. 11. Siekierka, J. J.; Hung, S. H. Y.; Poe, M.; Lin. C. S.; Sigal, N. H. Nuzure 1989,341, 755. 12. Fretz, H.; Albers, M. W.; Galat, A.; Standaert, R. F.; Lane, W. S.; Burakoff, S. J.; Bierer. B. E.; Schreiber. S. L. J. Am. Chem. Sot. 1991,113,
13. Francavilla, A. et al. Hepatology, in press. 14. Tropschug, M.; Barthelmess, I. B.; Neupert, W. Nature, 1989,342,953. 15. Liu, J.; Farmer, J. D.; Lane, W. S.; Friedman, J.; Weissman, I.; Schreiber, S. L. Cell 1991,66, 807. 16. Friedman, J.; Weissman, I. Cell 1991,66, 799. 17. Flanagan, W. M.; Corthesy, B.; Bram, R. J.; Crabtree, G. R. Nature 1991,352,803. 18 Knighton, D. R.; Zheng. J.; Ten Eyck, L. F.; Xuong, N.-H.; Taylor, S. S.; Sowadski, J. M. Science 1991, 253,414. 19. Michnick, S. W.; Rosen, M. K.; Wandless, T. J.; Ksrplus. M.; Schreiber, S. L. Science 1991,251, 836. 20. Moore, J. A.; Peattie. D. A.; Fitzgibbon, M. J.; Thomson, J. A. Nurure 1991,351, 248. 21. Van Duyne, G. D.; Standaert, R. F.; Karplus, P. A.; Schreiber, S. L.; Clardy, J. A. Science, 1991,251, 839. 22. Van Duyne, G. D.; Standaert, R. F.; Schreiber, S. L.; Clardy. J. A. J. Am. Chem. Sot. 1991 in press. 23. Tanaka, H.; Kuroda, A.; Marusawa, H.; Hatanaka. H.; Kino, T.; Goto, T.; Hashimoto, M.; Taga, T. J. Am. Chem. Sot., 1987,109,
24. Weber. C.; Wider, G.; von Freyberg. B.; Traber. R.; Braun, W.; Widmer, H.; Wiithrich, K. Biochemistry, 1991,30,
25. Fesik, S. W.; Gampe, R. T.; Eaton, H. L.; Gemmecker, G.; Olejniczalc, E. T.; Neir, P.; Holzman, T. F.;
Egan, D. A.; Edalji, R.; Simmer, R.; Helfrich, R.; Hochlowski. J.; Jackson, M. Biochemistry 1991,30, 6574. 26. Loosli, H.-R.; Kessler, H.; Oschkinat, H.; Weber, H.-P.; Petcher, T. J.; Widmer, A. Helven’ca Chimica Acm, 1985,68,
27. Transcripts of related lectures are scheduled for publication elsewhere: (a) Transplantation Proceedings of the
First International Congress on FK506, and (b) The Robert A. Welch Foundation Conference on Chemical Research XXXV. Chemistry at the Frontiers of Medicine.