Synthesis of a macrocyclic rhodamine 110 enzyme substrate as an intracellular probe for caspase 3 activity

Article abstract of DOI:10.1016/S0040-4039(00)00708-5

The synthesis of the macrocyclic rhodamine 110 caspase 3 substrate 8 is described. The key step is a high dilution intramolecular cyclization reaction of an in situ generated primary amine with a 4-nitrophenyl ester. Substrate 8 reacts with recombinant caspase 3 to yield a fluorescent signal but virtually no signal is detected in the absence of caspase 3 or in the presence of the caspase 3 inhibitor Ac-DEVD-CHO. Notably, 8 selectively stains live cells that have been induced to undergo apoptosis with etoposide. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00708-5

Tetrahedron Letters 41 (2000) 4733±4735
Synthesis of a macrocyclic rhodamine 110 enzyme substrate
as an intracellular probe for caspase 3 activity
Anthony P. Guzikowski,* John J. Naleway, Christina T. Shipp and Rod C. Schutte
Marker Gene Technologies Inc., University of Oregon Riverfront Research Park, 1850 Millrace Dr.,
Eugene, OR 97403-1927, USA
Received 16 November 1999; revised 24 April 2000; accepted 27 April 2000
Abstract
The synthesis of the macrocyclic rhodamine 110 caspase 3 substrate 8 is described. The key step is a high
dilution intramolecular cyclization reaction of an in situ generated primary amine with a 4-nitrophenyl
ester. Substrate 8 reacts with recombinant caspase 3 to yield a ¯uorescent signal but virtually no signal is
detected in the absence of caspase 3 or in the presence of the caspase 3 inhibitor Ac-DEVD-CHO. Notably,
8 selectively stains live cells that have been induced to undergo apoptosis with etoposide. # 2000 Elsevier
Science Ltd. All rights reserved.
Keywords: cyclization; enzymes and enzyme reactions; ¯uorescence; macrocycle.
Fluorogenic enzyme substrates o€er a convenient and sensitive means to measure enzyme
activity. Substrates incorporating the peptide sequence DEVD and a ¯uorophore such as 7-amino-
4-tri¯uoromethylcoumarin1 or rhodamine 1102 (1, R110) have been prepared to measure caspase
3 activity in studies of apoptosis. However, these substrates cannot measure caspase 3 activity in
intact cells since they are impermeable to the cell membrane. Masking the charge of the free car-
boxy groups of the glutamate and aspartate residues may improve substrate permeability of the
cell membrane. We reasoned that the synthesis of a macrocyclic DEVD substrate may mask these
charges since the carboxy groups should be more prone toward intramolecular hydrogen bonding
compared to a non-cyclic analog. Therefore, we initiated studies into the synthesis of the macro-
cyclic DEVD-R110 substrate 8.
Initially, the mono-protected R110 analog 2 was prepared by the reaction of 1 with 1 equiv. of
FMOC chloride and N,N-diisopropylethylamine (DIEA) in DMF (35±40%, Scheme 1).3a Com-
pound 2 was coupled to a suitably protected DEVD peptide. The a,a-dimethyl-3,5-dimethoxy-
benzyloxycarbonyl (Ddz) group was chosen for protection of the terminal amine4 while t-butyl
esters were employed for the protection of the side chain carboxy groups.5 The coupling of 2 and
* Corresponding author. Tel: 541-342-3760; fax: 541-687-7963; e-mail: apguzik@oregon.uoregon.edu
0040-4039/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00708-5

4734
N-Ddz-DEVD tri-t-butyl ester (3) was accomplished via the mixed anhydride of 3 (isobutyl
chloroformate (IBCF), N-methylmorpholine (NMM), THF) to give the disubstituted R110 4
(55±61%).
Scheme 1. Reagents and conditions: (a) FMOC chloride, DIEA, DMF, rt; (b) IBCF, NMM, THF, ^10ꢀC; (c) 2, THF,
^10ꢀC to rt; (d) 2.5% TFA in CH2Cl2 followed by NMM, rt; (e) mixed anhydride of TrNH(CH2)5COOH in THF,6 rt;
(f) 2% piperidine in DMF, rt; (g) adipoyl chloride, collidine, THF, rt; (h) 4-nitrophenol, THF, rt; (i) 5% TFA in
CH2Cl2 followed by CH2Cl2 and DIEA, rt; (j) 25% TFA in CH2Cl2, rt; (k) Et3N, MeOH, rt
A spacer was inserted in two pieces to link the terminal amine of the peptide with the second
amino group of the R110 molecule. The ®rst piece (7 atoms) consisted of a 6-aminohexanoic acid
residue that was N-protected with the triphenylmethyl (Tr) group.3b The Ddz protecting group of
4 was removed (2.5% TFA in CH2Cl2) and the amine salt was neutralized (NMM). The free
amine was treated with the mixed anhydride of N-Tr-6-aminohexanoic acid.6 Removal of the
FMOC group (2% piperidine in DMF) provided the mono-substituted R110 5. A second spacer
piece (6 atoms) was then added to complete the macrocyclic framework. Treatment of the aromatic
amine of 5 with excess adipoyl chloride and collidine in THF a€orded the non-¯uorescent bis-
amide (not shown). Reaction of the second acid chloride residue with 4-nitrophenol gave the
active 4-nitrophenyl (4-NP) ester 6 (30%).7
The key cyclization step was now performed. Removal of the Tr group (5% TFA in CH2Cl2)
gave the primary amine salt. The solution was diluted to a concentration of 1 mM and neutralized
with DIEA. The reaction was stirred for four days at 25ꢀC to give the cyclic triester 7 (37%).

4735
MALDI mass spectral analysis of 7 gives the expected molecular ion [(M+H)+=1181].8 A second
charged species [(M+23)+] of near equal intensity is also present suggesting that polyamide 7 is
able to coordinate with sodium ions thus lending further validity to the macrocyclic structural
assignment. Removal of the t-butyl esters was accomplished with 25% TFA in CH2Cl2 to give 8,
which was isolated as the tris-Et3NH+ salt.9
Macrocycle 8 is virtually non-¯uorescent in solution since diacyl R110 molecules are constrained
to the lactone form. Treatment of 8 with recombinant caspase 3 induces formation of the ¯uorescent
carboxylate form, presumably by cleavage of the DEVD link.10 Essentially no ¯uorescence is
observed under these conditions when the caspase 3 inhibitor Ac-DEVD-CHO is present.11
Staining of live NIH3T3 and mouse melanoma D5 cell lines, induced to undergo apoptosis with
etoposide, exhibited speci®c and localized intracellular substrate turnover.12 Kinetics and speci®city
studies of such cyclic substrates are underway.
Acknowledgements
We wish to thank Professor John F. W. Keana (University of Oregon, Eugene, OR) for his
helpful synthetic insights.
References
1. Gurtu, V.; Kain, S. R.; Zhang, G. Anal. Biochem. 1997, 251, 98±102.
2. Molecular Probes, Inc., Eugene, OR, USA.
3. (a) Greene, T. W. Protective Groups in Organic Synthesis; John Wiley & Sons: New York, 1981; pp. 225±226. (b)
ibid pp. 273±274.
4. Schmitt, W.; Zanotti, G.; Wieland, T.; Kessler, H. J. Am. Chem. Soc. 1996, 118, 4380±4387.
5. Peptide 3 was purchased from BioMol Research Laboratories, Plymouth Meeting, PA, USA.
6. N-Tr-6-Aminohexanoic acid was prepared by the reaction of methyl 6-aminohexanoate hydrochloride with Tr-
chloride (DMF, DIEA) followed by saponi®cation. The corresponding mixed anhydride was prepared by treating
the protected amino acid with IBCF in the presence of NMM in THF at ^10ꢀC.
7. The preparation of a cyclic peptide employing a 4-nitrophenyl ester has been reported. Studer, R. O.; Lergier, W.
Helv. Chim. Acta 1965, 460±470.
8. The mass spectral analysis was performed by Mass Consortium Corp., San Diego, CA, USA.
9. Compound 8: 1H NMR (D2O) ꢀ 0.97±1.80 (m, 6H), 1.21 (t, J=7 Hz, 27H), 1.36±2.78 (m, 28H), 2.97 (q, J=7 Hz,
18H), 4.06 (t, J=3 Hz, 1H), 4.38 (bs, 1H), 4.68±4.80 (m, 1H), 6.64 (d, J=9 Hz, 1H), 6.65 (d, J=9 Hz, 1H), 7.11 (d,
J=9 Hz, 1H), 7.23 (d, J=8 Hz, 1H), 7.50±7.83 (m, 4H), 7.88±7.93 (m, 1H), 8.02 (d, J=8 Hz, 1H).
10. Assay conditions: [8]=10 mM; Recombinant caspase 3 was prepared in bacterial cell cultures and crude
preparations were diluted to give a measurable ¯uorescent signal. An 1:50 000 dilution gave about 1300 counts/h
over 3 h. An 1:1000 dilution gave about 5000 counts/h over 12 h. Virtually no ¯uorescence was detected in the
absence of recombinant caspase 3.
11. Garcia-Calvo, M.; Peterson, E. P.; Leiting, B.; Ruel, R.; Nicholson, D. W.; Thornberry, N. A. J. Biol. Chem. 1998,
273, 32608±32613. Experiments performed as described in Ref 10 but with the addition of Ac-DEVD-CHO (10
mM).
12. Assay conditions: NIH3T3 (mouse endothelial) and D5 (mouse melanoma) cell lines (ATCC, Rockville, MD),
induced for apoptosis by treatment with etoposide (50 mg/mL in RPMI1640 media+10% CS for 18 h), with [8]=50
mM (for 1 h) showed signi®cant and cell-speci®c labeling. These same cell lines were also stained with the
¯uorescent caspase 3 cell permeable inhibitor FITC-VAD-FMK (Promega Corp., Madison, WI) to verify
apoptosis induction. Non-induced cell lines showed little or no detectable labeling under identical conditions.