Bioconjugate Chem. 1995, 6, 302-31 2
302
Construction of Coordinatively Saturated Rhodium Complexes Containing Appended Peptides Niranjan Y. Sardesai, Susanne C. Lin, Kaspar Zimmermann, a n d Jacqueline K. Barton* Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, California 91125. Received January 31, 1995@
Phenanthrenequinone diimine (phi) complexes of rhodium( 111)bearing appended peptides have been prepared using two complementary solid phase synthetic strategies. The first method involves the direct coupling of the coordinatively saturated rhodium complex containing a pendant carboxylate to the N-terminus of a resin-bound peptide, in a manner analogous to the chain-elongation step in solid phase peptide synthesis. The second involves coupling a bidentate chelator containing the pendant carboxylate to the resin-bound peptide, followed by coordination of [Rh(phi)2I3+to the bidentate chelator attached to the peptide. Peptides of length 5-30 residues have been covalently attached to rhodium complexes in 5-18% yield using both methods. Despite the low overall yields, the regioselective modification of the peptide chain afforded by these strategies is a distinct advantage over solution phase methods. With coordination complexes which are stable to peptide deprotection and cleavage conditions from the resin, the solid phase synthetic strategies are convenient to apply. Amino acid analysis, electronic spectroscopy, and circular dichroism confirm the presence of the two components in the metal-peptide chimeras; the metal-peptide complexes exhibit the combined spectral properties of the parent metal complex and the appended peptide. Significantly, plasma desorption mass spectrometry reveals a novel pattern of peptide fragmentation for the metal-peptide chimeras that is not observed in the absence of the tethered metal complex; this fragmentation facilitates the sequence analysis of the appended peptide. Thus, metal-peptide chimeras may be conveniently prepared using solid phase methodologies, and features of coordination chemistry may be exploited for new peptide design and analysis.
INTRODUCTION
There has been increased attention focused on the assembly of peptides containing coordinated transition metal complexes (I-24). The coordination geometry about a metal ion provides a rigid, well-defined center from which to append peptides for a specific function. As with larger metalloproteins, the metal ion may serve a structural role in bringing together discrete elements of peptide secondary structure- (1-28) or it may serve a functional role in catalysis (19-24). Furthermore, as is also found with larger protein systems, the presence of the transition metal center provides a convenient spectroscopic handle to assay function. Examples of the incorporation of coordination chemistry in peptide design include the application of metal ion coordination to stabilize peptide a-helices (1-8) and p-turns (91, the de novo design of three and four helix bundle proteins through crosslinking peptide helices by metal ions (1-6, 10-12), and the construction of donor-acceptor assemblies in studies of photoinduced electron transfer across peptides (14-17). Our laboratory has focused on the design of transition metal complexes to explore site-specific recognition of nucleic acids (25-28). As a part of this effort to construct smaller functional mimics of DNA-binding proteins, we became interested in incorporating appended peptides in our design. We recently reported the site-selective recognition of double helical DNA by metal-peptide constructs, in which the recognition characteristics of the complex are governed by the appended peptide (29).In these studies, peptides (13 residues) were appended to a sequence-neutral (28) metallointercalator, [Rh(phih(phen)13+(phi = 9,lO-phenanthrenequinonediimine; phen
* To whom correspondence should be addressed. @
Abstract published in Advance ACS Abstracts, April 1,1995.
= 1,lO-phenanthroline) so as to produce a sequencespecific DNA-binding molecule (29). Using this strategy, an array of metal-peptide complexes with different recognition characteristics may be constructed. Here we describe the syntheses and characterization of metal complexes containing different tethered peptides, and we explore the advantages and limitations of the methodology. Figure 1illustrates a representative metalpeptide chimera. Our strategy involves the assembly of the metal-peptide complexes on a solid support. Peptides of length 5-30 amino acids have been synthesized by standard solid phase synthesis (30-32). Thereafter, metal complexes are appended onto the resin through two distinct coupling methodologies: (i)by direct coupling of [Rh(phi)2Ll3+(L = a bidentate chelator containing a pendant carboxylate) to the peptide on the resin or (ii) by first coupling L to the resin-bound peptide, followed by coordination of [ R h ( ~ h i ) ~to ] ~ L. + Meyer and coworkers have used a similar strategy of direct coupling to synthesize a tripeptide containing [Ru(bp~)3]~+ (14). These syntheses complement solution phase coordination methodologies (1-24). The solution method relies, however, on the selective coordination of the metal center to the desired side chain functionalities on the peptide. Regioselective ligation in solution may be limited if the desired peptide sequence contains, for example, several cysteine or histidine residues. Instead, coordination directly on the resin offers control through the selective deprotection of side-chain functionalities for ligation. We also describe the characterization of these metalpeptide constructs using mass spectrometric analysis in addition to the more conventional spectrophotometric methods. Here too, the metal ion yields distinct advantages. In the presence of the appended coordination complex, a novel pattern of peptide fragmentation is
1043-1802/95/2906-0302$09.00/00 1995 American Chemical Society
Bioconjugafe Chem., Vol. 6,No. 3, 1995 303
Construction of Rh-Peptide Complexes
si
; LQQAIEQLQNAAAA
COOH
1 3 +
= 23 600 M-I cm-'. 252Cf plasma desorption mass spectrometry (PDMS) was recorded on a time-of-flight spectrometer (Bio-IodApplied Biosystems 20 K, Uppsala, Sweden). The mass scale was calibrated on the hydrogen and nitrate ions, and the experimental error is OH
[Rh(phi)2(DMF)21( O m 3
HjN
I
phen
N / @+-@ .
P
DCC/HOBt
1 P
P
P = protecting groups
? F :
o>;QJ)-@ HN
phdn
I
P
P
0
Cleavage/deprotection Purification
Scheme 3. Direct coupling method for the synthesis of metal-peptide chimeras. :
U
O
Solid Phase Peptide Synthesis (tBoc/FMOC)
H
[Rh(phi)2phe&213
J
HOBt
H 2 N G + e m P
P
P = protecting groups
HP
[Rh(phi)2phen13+
metal-peptide complexes is found to be longer than that of the corresponding free peptide. Overall Yields. Theoretical yields of metal-peptide complexes may be determined based upon the initial substitution of the resin. Actual recovered yields of pure chimera (after two rounds of HPLC purification) are found to be in the range of 5-18%. Not surprisingly, the major determinant of the yield for the reaction is the
P
P
Cleavage/deprotection Purification
initial synthesis of the peptide. Yields of metal-peptide chimeras correlate closely with the recovered yields for the individual peptides before metal-complex attachment. For small peptides (
