Compstatin

Expression of compstatin in Escherichia coli: Incorporation of unnatural amino acids enhances its activity
Madan Katragadda, John D. Lambris ¤
Protein Chemistry Laboratory, Department of Pathology and Laboratory Medicine, School of Medicine, University of Pennsylvania, Philadelphia, PA 19104, USA
Received 4 September 2005, and in revised form 11 November 2005
Available online 15 December 2005

Abstract

Compstatin, a 13-residue cyclic peptide, is a complement inhibitor that shows therapeutic potential. Several previous approaches have improved the activity of this peptide several-fold. In the present study, we have expressed and purified compstatin from Escherichia coli in an eVort to increase its potency and to generate it in high yield in a more economical fashion. An intein-based expression system was used to express compstatin in fusion with chitin-binding domain and Ssp DnaB intein, which were later cleaved from the expressed molecule at room temperature and pH 7.0 to yield pure compstatin in one step. The expressed compstatin showed activity similar to the synthetic compstatin in an ELISA-based assay. The same expression system and purification strategy were used to incorporate three tryptophan analogs, 6-fluoro-tryptophan, 5-hydroxy-tryptophan, and 7-aza-tryptophan, into compstatin. Interestingly, incorporation of 6-fluoro- tryptophan increased the activity three-fold relative to wild-type compstatin; in contrast, incorporation of 5-hydroxy- or 7-aza-trypto- phan rendered compstatin less active than the wild-type form.
 2005 Elsevier Inc. All rights reserved.

Keywords: Compstatin; Peptide; Intein; Expression; Tryptophan analog

Compstatin is a 13-residue cyclic peptide inhibitor that binds to the third component of complement and abrogates the activation of complement cascade [1–7]. Compstatin, as shown using NMR, folds into a loop-like structure with a type I beta turn in solution and remains cyclic through a disulfide bond bridging residues Cys-2 and Cys-12 [8]. This peptide was originally discovered by screening a phage- displayed peptide library; all the subsequent design and development were carried out using the peptides that had been chemically synthesized employing solid-phase peptide chemistry [1,9]. The drawbacks of this solid-phase peptide synthesis approach include its high cost and the possible presence of residual chemicals that can interfere with the biological activity.

* Corresponding author. Fax: +1 215 573 8738.
E-mail address: [email protected] (J.D. Lambris).
URL: www.complement.us (J.D. Lambris).

The goal of the present study was to express compstatin in bacteria using an intein-based vector and subsequently incorporate tryptophan analogs into the peptide, since substituting tryptophan for valine at position 4 has been shown to increase the activity of compstatin several-fold [7]. Since most of the tryptophan analogs are available cur- rently as a mixture of D and L forms, solid-phase peptide synthesis becomes cumbersome, as the peptides containing D and L forms of tryptophan (analogs) have to be separated. Previous studies carried out with compstatin analogs containing D-amino acids have shown no activity for these peptides. Hence, only peptides containing L-amino acids are desired, and separation of peptides with L-tryptophan ana- logs from the mixture results in lower yields. Therefore, we have attempted to express compstatin in Escherichia coli to increase the yield in a more economical way, since E. coli is capable of incorporating only L-amino acids, and thus pro- vides a useful system for efficiently incorporating amino acids even when present as a DL mixture [10–13].

1046-5928/$ – see front matter  2005 Elsevier Inc. All rights reserved. doi:10.1016/j.pep.2005.11.016

In the present study, we have used an intein-based expres- sion system to express compstatin. Inteins have a unique capacity to cleave themselves from exteins in the process of generating a functional protein. This property has been exploited by Xu and co-workers [14] to design expression sys- tems in which the protein of interest is fused with an intein and later cleaved under defined conditions that support intein cleavage. This expression system is cost-eVective, as the pro- tein of interest can be purified in a single step, and no enzyme cleavage is required to obtain the final product. As a result, this system has been widely employed to address several applications involving proteins and peptides [14–23]. Using this expression system, we succeeded to express compstatin and its analogs, some of which showed increased activity.

Experimental procedures

Reagents

The pTWIN 1 vector was obtained from New England Biolabs (Beverly, MA, USA). SOB medium, sodium phos- phate, sodium chloride (NaCl),1 ethylenediamine tetraacetic acid (EDTA), ammonium acetate, isopropyl- beta-D-thiogalactopyranoside (IPTG), M9 minimal medium, 6-fluoro-DL-tryptophan, 7-aza-DL-tryptophan, and 5-hydroxy-L-tryptophan were obtained from Sigma– Aldrich (St. Louis, MO, USA).

Construction of the compstatin gene

A compstatin analog with the sequence NH2-GIC- VWQDWGAHRCTN-OH, designated [G( 1)/V4W/H9A/ N14], was expressed in fusion with the chitin-binding domain (CBD) and the DnaB intein (New England Bio- labs). Back-translation of the peptide sequence, taking into consideration the codon usage for E. coli, provided the fol- lowing genetic code, which was used to generate a synthetic gene for this peptide with the sequence:
5′-ATTTGCGTTTGGCAGGATTGGGGTGCGCAC CGTTGCACCAATTAA-3′.
To clone the synthetic gene into the pGEM-T vector, a 5′ flanking region containing a SapI site and 3′ flanking region containing a PstI site were designed. To construct
the synthetic gene, the four overlapping oligonucleotides shown below were designed using DnaWorks software [24] and synthesized at Invitrogen (Carlsbad, CA, USA):
5′-GGTGGTGCTCTTCCAACGGTATTTGCGTTT GGCAGGA-3′
5′-TTGGGGTGCGCACCGTTGCACCAATTAACT GCAGG-3′

1 Abbreviations used: Ac, acetyl group; OH, hydroxy; 6fW, 6-fluoro- tryptophan; NaCl, sodium chloride; EDTA, ethylenediaminetetraacetic acid; IPTG, isopropyl-β-D-thiogalactopyranoside; CBD, chitin-binding domain; MALDI, matrix-assisted laser desorption ionization; TOF, time of flight; HPLC, high-performance liquid chromatography; PHMB, para- hydroxyl mercuryl benzoate.

3′-CAACGTGGTTAATTGACGTCCGC-5′
3′-CATAAACGCAAACCGTCCTAACCCCACGCG TGG-5′.
The overlapping DNA fragments were assembled by PCR as described by Heyneker and co-workers [25].

Cloning of the compstatin gene

The resulting gene was amplified using the primers 5′-C GCCTGCAGTTAATTGGT-3′ and 5′-GGTGGTGCTC TTCCAACG-3′.
The PCR-amplified fragments of compstatin were then cloned into the pGEM-T vector, and the resulting clone was digested with PstI and SapI. The digested compstatin gene was further subcloned into the expression vector pTWIN1, which had been predigested with PstI and SapI. The sequence of the clone was verified at the DNA Sequencing Facility of the University of Pennsylvania. Fig. 1 shows the vector construct depicting diVerent genetic elements.

Expression and purification of compstatin

To express compstatin, ER2566 E. coli cells transformed with the compstatin clone were grown in SOB medium (20 g/L tryptone, 5 g/L yeast extract, 0.5 g/L NaCl, 2.5 mM
KCl, and 10 mM MgCl2) at 37 °C. When an OD600 of 0.7 was reached, expression was induced by the addition of IPTG to a final concentration of 0.3 mM, followed by an additional incubation at 37 °C for 4 h. Cells were collected
by centrifugation and lysed by sonication in 20 mM phos- phate buVer, pH 8.5, with 500 mM NaCl and 1 mM EDTA (buVer A) supplemented with 0.2% Tween 20. The cell extract was centrifuged, and the soluble fraction was applied to a chitin-binding column pre-equilibrated with buVer A. The insoluble pellet was solubilized using buVer A containing 8 M urea, and after diluting 10 times using buVer A was applied to the column. The column was washed with 100 mL of buVer B1, followed by a quick wash with three column volumes of 50 mM ammonium acetate, pH 7.0 (buVer B). The column was incubated at room tem- perature for 20 h, and the peptide was eluted with buVer B, lyophilized, and further purified on a reversed-phase C18 high-performance liquid chromatography (HPLC) column using a 0–90% acetonitrile gradient. The purified peptide was identified and characterized using matrix-assisted laser desorption ionization (MALDI)-time of flight (TOF) mass spectrometry.

Expression of tryptophan analogs of compstatin in Escherichia coli

To express compstatin analogs containing tryptophan derivatives, the pTWIN1-compstatin clone was transformed into the ER2566 Trp 82 auxotroph (obtained from Bjorkl- und) [23]. Expression was carried out in M9 minimal medium supplemented with 1 mM L-tryptophan as described above.

Fig. 1. Construction of the compstatin gene. A schematic showing the location of compstatin insert in the pTWIN1 vector. Also shown are the locations of the other genetic elements that assist in the replication and expression of the plasmid. This picture is drawn using the ApE-plasmid editor software (http:// www.biology.utah.edu/jorgensen/wayned/ape).

Cells were grown to an OD600 of 0.8–1.0, then collected by centrifugation, and resuspended in fresh minimal medium containing 2 mM of the desired tryptophan analog(s):
6-fluoro-tryptophan, 7-aza-tryptophan, or 5-hydroxy-trypto- phan. The expressed compstatin analogs were further purified as described above.

Activity measurements for expressed compstatin

The complement inhibitory activity of compstatin analogs was assayed as described elsewhere [7]. In brief, an antigen– antibody complex was coated onto an ELISA plate, and the deposition of C3b in the presence of the compstatin analog was measured. The absorbance data obtained at 405 nm were translated into % inhibition based on the absorbance corre- sponding to 100% complement activation. The % inhibition was plotted against the peptide concentration, and the result- ing dataset was fit to the logistic dose–response function using Origin 7.0 software. IC50 values were obtained from the fitted parameters that achieved the lowest z2 value.

Results and discussion

Construction and cloning of the gene

The gene encoding compstatin, together with the flank- ing restriction enzyme sites, is 74 nucleotides long. The gene was assembled using short overlapping fragments encom-

passing the entire 74 nucleotides. Upon cloning into the pTWIN1 vector, the SapI site was lost, while the PstI site was regenerated. Flanking the multiple cloning site in this vector is a region that encodes an intein, Ssp DnaB fused with the CBD at its N-terminus. While the CBD facilitates the isolation of the fusion protein from the cell extract, the intein activates self-cleavage and releases compstatin free of any added (extra) amino acids. Fig. 1 shows the final con- struct depicting the sequence of the insert and other genetic elements that contribute to the expression. The sequence of the clone was verified using the Ssp DnaB intein forward primer.

Expression and purification of compstatin

The fusion protein CBD-DnaB-compstatin was purified from the soluble fraction of cell lystate on a chitin column followed by a reverse-phase C18 column; the final yield was 2 mg/L of the culture. Fig. 2 (lanes 2 and 3) shows a 13% SDS–PAGE gel of compstatin in fusion with Ssp DnaB– CBD. The peptide was further characterized using MALDI-TOF spectrometry and HPLC; single peaks (Fig. 4) in each of these profiles indicate that the peptide was obtained in pure form.
To incorporate tryptophan analogs into compstatin, the clone bearing the compstatin gene was transformed into a tryptophan-auxotrophic ER2566 strain. Previous studies have indicated that E. coli is capable of incorporating these

Fig. 2. SDS–PAGE analysis of expressed compstatin-fusion proteins. A 13% SDS–PAGE gel showing the induction of the compstatin-fusion gene expression by 0.3 mM IPTG at 30 °C in the presence of tryptophan (lanes
1 and 5), 6-fluoro-tryptophan (lanes 2 and 6), 5-hydroxy-tryptophan (lanes 3 and 7), and 7-aza-tryptophan (lanes 4 and 8). Lanes 1–4 represent soluble fraction, while lanes 5–8 represent insoluble cell pellet solubilized in 8 M urea. Molecular weights of the standards are expressed in kilodalton.

analogs without modifying its translation machinery [11,26– 33]. Following induction with IPTG, the fusion protein con- taining tryptophan resulted in the soluble fraction (lane 1). In contrast, the fusion protein with 6-fluoro-tryptophan incorporated instead of tryptophan expressed in lower levels and resulted in inclusion bodies as shown in Fig. 2 (lane 2, soluble; lane 6, insoluble). This distinction could have occurred as a result of 6-fluoro-tryptophan rendering the fusion protein more hydrophobic compared to tryptophan. In the case of 5-hydroxy-tryptophan and 7-aza-tryptophan incorporation, due to the low expression levels, we were unable to assess if the fusion protein is localized in soluble or insoluble fraction (Fig. 2). Following single-step purifica- tion, pure compstatin was obtained with the yields of 1.2 mg/ L for 6-fluoro-tryptophan, 0.2 mg/L for 5-hydroxy-trypto- phan, and 0.05 mg/L for 7-aza-tryptophan. The low yield for these peptides can be attributed to poor growth of cells in the presence of tryptophan analogs. As shown in Fig. 3, in the presence of tryptophan analogs, 6-fluoro-tryptophan

and 5-hydroxy-tryptophan, E. coli exhibits slower growth while in the presence of 7-aza-tryptophan, the growth was minimal. The incorporation of tryptophan analogs was con- firmed using MALDI as indicated by the spectra shown in the inset of Figs. 4 and 5 and the molecular masses are listed in Table 1. Both native compstatin and its analogs in the final pure forms were shown to be cyclized through a disul- fide bond, as evidenced by their inability to react with para- hydroxyl mercuryl benzoate (PHMB) [34] (data not shown). The cyclization could have occurred naturally during the process of purification.
All the peptides were further purified on a reversed-phase C18 HPLC column. For each of the peptides containing 6- fluoro-tryptophan and 5-hydroxy-tryptophan, three peaks were observed in the chromatogram (Fig. 5). The sequence of expressed compstatin contains two tryptophans at positions 4 and 7 and ideally single cyclized species of compstatin con- taining tryptophan analog at both the positions is expected. However, if the medium is contaminated with tryptophan through a carry over from the media exchange during the induction, peptides with tryptophan incorporated simulta- neously at both the positions or at either of the positions are also expressed. The mass identification of the peptides (Table 1; Fig. 5) eluted in each of the three peaks observed in the chromatograms using MALDI indicated that diVerential incorporation of the tryptophan analogs occurred. None of these peptides represent the linear non-cyclized compstatin as evidenced from the masses obtained from the MALDI spec- tra. In addition, the mass spectra of the peptides representing each of these peaks indicated that the peptides were eluted in increasing order of the degree of incorporation of the fluoro analog. In contrast, the elution order was reversed in the case of peptides containing 5-hydroxy-tryptophan. These diVeren- tial changes in the elution order reflect the relative hydropho- bicity of the peptides altered as a result of the incorporation of the tryptophan analogs. For example, 5-hydroxy-tryptophan is hydrophilic, while 6-fluoro-tryptophan is hydrophobic.

1.6

1.4

1.2

1.0

0.8

0.6

0.4

0.2

0.0

-0.2

Tryptophan
5-fluoro-tryptophan 5-hydroxy-tryptophan

Activity of expressed compstatin and its analogs

The expressed compstatin [G(¡1)/V4W/H9A/N14] exhibited an IC50 of 1.2 µM, which is similar to that observed for the synthetic Ac-V4W/H9A analog [7]. It was previously suggested that acetylation neutralizes the charge on the N-terminus of compstatin, thus preserving the hydrophobic cluster required for the activity [35]. Our findings in the present study indicate that glycine contrib- utes in preserving the hydrophobic cluster, thus eliminating the need for acetylation. All the expressed compstatin ana- logs except the 7-aza-tryptophan analog were found to be active at the concentrations tested. However, the peptide

-200 0 200 400 600 800 1000 1200 1400
Time(min)

Fig. 3. Growth curves of the Trp-82 auxotrophic ER2566 strain. The plot shows the optical density of the cells grown in the presence of tryptophan (squares), 6-fluoro-tryptophan (circles), and 5-hydroxy-tryptophan (triangles) as a function of time.

showed diVerent levels of activity relative to the synthetic analog, Ac-V4W/H9A (Fig. 6; Table 2): compstatin con- taining 6-fluoro-tryptophan exhibited a three-fold higher activity than did the Ac-V4W/H9A analog, suggesting that increasing the hydrophobicity of the indole ring by adding fluorine atom increases the activity of the peptide (Fig. 6).

1600

1400

1200

1000

800

600

400

200

0
0 10 20 30 40 50 60
Time (minutes)
Fig. 4. Purification of expressed compstatin. Compstatin was expressed using an intein-mediated expression system and purified using reversed-phase HPLC employing a linear gradient of 5–90% acetonitrile. An HPLC chromatogram representing purified compstatin is shown. A MALDI spectrum of the purified compstatin is shown in the inset. The m/z ratio represents the masses of the protonated species.

A 800

600

400

200

0

B 300

200

100

0

C 200

100

0
0 10 20 30 40 50 60
Time (minutes)
Fig. 5. Characterization of compstatin analogs. Shown in the picture are HPLC chromatograms of the expressed compstatin analogs containing (A) 6-fluoro-tryptophan, (B) 5-hydroxy-tryptophan, and (C) 7-aza-tryptophan. The corresponding mass spectra for the fully incorporated compstatin analogs are shown in the inset. The m/z ratio represents the masses of the protonated species.

We also investigated the incorporation of the less hydro- phobic tryptophan analogs 5-hydroxy-tryptophan and 7-aza-tryptophan as a means of assessing the significance of

the hydrophobic nature of the indole ring in the interaction of compstatin with C3. In contrast to the results obtained with the 6-fluoro analog, the compstatin analog containing

Table 1
Molecular masses of compstatin and its analogs
Expressed peptide Expected molecular weight

Observed molecular weight

with 7-aza-tryptophan shows the relative importance of this carbon atom. One possible explanation is that the nitrogen at position 7 of the indole ring reduces the hydro- phobic character of the tryptophan. These data suggest that

G( 1)/V4W/H9A/N14 1743.7406 1743.74
G( 1)/V4(6fW)/W7(6fW)/ 1779.7217 1779.91
H9A/N14
a

the degree of hydrophobicity of the indole ring influences the activity of compstatin.
In conclusion, fully active compstatin was obtained by

G( 1)/V4(5-OH -W)/W7 (5-OH-W)/H9A/N14
G( 1)/V4(7-aza-W)/W7 (7-aza-W)/H9A/N14
a Represents hydroxy.

1775.7148 1775.94

1745.7311 1745.85

expression in E. coli, with an yield of 2 mg/L of culture. Incorporation of unnatural amino acids with varying degrees of hydrophobicity at positions 4 and 7 led to a 126- fold increase in activity in the case of the 6-fluoro analog and suggest that increasing the hydrophobicity at these two positions can increase the activity of compstatin.

Acknowledgments

We sincerely thank Mikael Björklund, Department of Biological and Environmental Sciences, University of Hel- sinki, Finland for providing the tryptophan-auxotrophic strain of ER2566, and critically reading the manuscript. We thank Charissa Wiley and Lynn Spruce for their help with HPLC purification. We also thank Deborah McClellan for editorial assistance. This work was supported by NIH Grants GM 62134 and GM 069736.

References

Fig. 6. Activity of expressed compstatin and its analogs. Plots of percent complement inhibition versus peptide concentration for Ac-V4W/H9A (squares) and expressed compstatin with tryptophan (circles), 6-fluoro- tryptophan (stars), 5-hydroxy-tryptophan (hexagons), and 7-aza-trypto- phan (diamonds).

Table 2
Complement inhibitory activity of the expressed peptides
G(¡1)/V4(6fW)/W7(6fW)/H9A/N14 0.43 126
G(¡1)/V4(5-OHc-W)/W7(5-OH-W)/H9A/N14 33 1.6
G(¡1)/V4(7-aza-W)/W7(7-aza-W)/H9A/N14 122 0.44

a Relative to the activity of the peptide I(CVVQDWGHHRC)T-NH2.
b Synthetic peptide.
c Represents hydroxy.

5- hydroxy-tryptophan showed a 28-fold loss in activity when compared to the Ac-V4W/H9A analog, and the pep- tide containing 7-aza-tryptophan showed no activity at all at any of the concentrations tested. These results indicate that lowering the hydrophobicity of the indole ring results in a loss of activity (Fig. 6). The 7-aza-tryptophan resem- bles tryptophan in molecular structure, except that it has nitrogen instead of a carbon atom at position 7 of the indole ring. The loss in activity observed after substitution

[1] A. Sahu, B.K. Kay, J.D. Lambris, Inhibition of human complement by a C3-binding peptide isolated from a phage-displayed random pep- tide library, J. Immunol. 157 (1996) 884–891.
[2] A.M. Soulika, M.M. Khan, T. Hattori, F.W. Bowen, B.A. Richardson,
C.E. Hack, A. Sahu, L.H. Edmunds Jr., J.D. Lambris, Inhibition of heparin/protamine complex-induced complement activation by compstatin in baboons, Clin. Immunol. 96 (2000) 212–221.
[3] D. Morikis, J.D. Lambris, Structural aspects and design of low-molec- ular-mass complement inhibitors, Biochem. Soc. Trans. 30 (2002) 1026–1036.
[4] B.P. Morgan, C.L. Harris, Complement therapeutics; history and cur- rent progress, Mol. Immunol. 40 (2003) 159–170.
[5] A. Sahu, D. Morikis, J.D. Lambris, Compstatin, a peptide inhibitor of complement, exhibits species-specific binding to complement compo- nent C3, Mol. Immunol. 39 (2003) 557–566.
[6] M.C. Holland, D. Morikis, J.D. Lambris, Synthetic small-mole- cule complement inhibitors, Curr. Opin. Investig. Drugs 5 (2004) 1164–1173.
[7] B. Mallik, M. Katragadda, L.A. Spruce, C. Carafides, C.G. Tsokos, D. Morikis, J.D. Lambris, Design and NMR characterization of active analogues of compstatin containing non-natural amino acids, J. Med. Chem. 48 (2005) 274–286.
[8] D. Morikis, N. Assa-Munt, A. Sahu, J.D. Lambris, Solution structure of compstatin, a potent complement inhibitor, Protein Sci. 7 (1998) 619–627.
[9] A. Sahu, A.M. Soulika, D. Morikis, L. Spruce, W.T. Moore, J.D. Lam- bris, Binding kinetics, structure–activity relationship, and biotransfor- mation of the complement inhibitor compstatin, J. Immunol. 165 (2000) 2491–2499.
[10] E.A. Pratt, C. Ho, Incorporation of fluorotryptophans into proteins of Escherichia coli, Biochemistry 14 (1975) 3035–3040.
[11] C. Evrard, J. Fastrez, J.P. Declercq, The incorporation of a non-nat- ural amino acid (aza-tryptophan) may help to crystallize a protein and to solve its crystal structure. Application to bacteriophage lambda lysozyme, Acta Crystallogr. D Biol. Crystallogr. 55 (Pt. 2) (1999) 430–435.

[12] T.A. Alekseev, N.I. Dergousova, E.D. Shibanova, E.A. Azeeva, E.V. Kriukova, T.A. Balashova, P.V. Dubovskii, A.S. Aesen’ev, V.I. Tset- lin, [5-Fluoro-tryptophan-containing N-terminal domain of the alpha-subunit of the Torpedo californica acetylcholine receptor: preparation in E. coli and 19F NMR study], Bioorg. Khim. 29 (2003) 384–390.
[13] J. Broos, E. Gabellieri, E. Biemans-Oldehinkel, G.B. Strambini, Efficient biosynthetic incorporation of tryptophan and indole analogs in an integral membrane protein, Protein Sci. 12 (2003) 1991–2000.
[14] S. Chong, Y. Shao, H. Paulus, J. Benner, F.B. Perler, M.-Q. Xu, Pro- tein splicing involving the Saccharomyces cerevisiae VMA intein. The steps in the splicing pathway, side reactions leading to protein cleav- age, and establishment of an in vitro splicing system, J. Biol. Chem. 271 (1996) 22159–22168.
[15] S. Chong, G.E. Montello, A. Zhang, E.J. Cantor, W. Liao, M.Q. Xu, J. Benner, Utilizing the C-terminal cleavage activity of a protein splicing element to purify recombinant proteins in a single chromatographic step, Nucleic Acids Res. 26 (1998) 5109–5115.
[16] T.C. Evans Jr., M.Q. Xu, Intein-mediated protein ligation: harnessing nature’s escape artists, Biopolymers 51 (1999) 333–342.
[17] T. Otomo, K. Teruya, K. Uegaki, T. Yamazaki, Y. Kyogoku, Improved segmental isotope labeling of proteins and application to a larger protein, J. Biomol. NMR 14 (1999) 105–114.
[18] M.W. Southworth, K. Amaya, T.C. Evans, M.Q. Xu, F.B. Perler, Puri- fication of proteins fused to either the amino or carboxy terminus of the Mycobacterium xenopi gyrase A intein, Biotechniques 27 (1999) 110–114, 116, 118–120.
[19] M.L. Lesaicherre, R.Y. Lue, G.Y. Chen, Q. Zhu, S.Q. Yao, Intein- mediated biotinylation of proteins and its application in a protein microarray, J. Am. Chem. Soc. 124 (2002) 8768–8769.
[20] S.F. Singleton, R.A. Simonette, N.C. Sharma, A.I. Roca, Intein-medi- ated affinity-fusion purification of the Escherichia coli RecA protein, Protein Expr. Purif. 26 (2002) 476–488.
[21] W. Wu, D.W. Wood, G. Belfort, V. Derbyshire, M. Belfort, Intein- mediated purification of cytotoxic endonuclease I-TevI by insertional inactivation and pH-controllable splicing, Nucleic Acids Res. 30 (2002) 4864–4871.
[22] M.Q. Xu, T.C. Evans Jr., Purification of recombinant proteins from
E. coli by engineered inteins, Methods Mol. Biol. 205 (2003) 43–68.
[23] M. Bjorklund, H. Valtanen, H. Savilahti, E. Koivunen, Use of intein- directed peptide biosynthesis to improve serum stability and bioactiv- ity of a gelatinase inhibitory peptide, Comb. Chem. High Throughput Screen 6 (2003) 29–35.

[24] D.M. Hoover, J. Lubkowski, DNAWorks: an automated method for designing oligonucleotides for PCR-based gene synthesis, Nucleic Acids Res. 30 (2002) e43.
[25] W.P. Stemmer, A. Crameri, K.D. Ha, T.M. Brennan, H.L. Heyneker, Single-step assembly of a gene and entire plasmid from large numbers of oligodeoxyribonucleotides, Gene 164 (1995) 49–53.
[26] D.C. Kombo, G. Nemethy, K.D. Gibson, J.B. Ross, S. Rackovsky,
H.A. Scheraga, EVects on protein structure and function of replacing tryptophan with 5-hydroxytryptophan: single-tryptophan mutants of the N-terminal domain of the bacteriophage lambda repressor, J. Pro- tein Chem. 15 (1996) 77–86.
[27] H. Adachi, H.M. Rajh, G.I. Tesser, J.J. de Pont, R.T. Jensen, J.D. Gardner, Interaction of tryptophan-modified analogues of cholecys- tokinin-octapeptide with cholecystokinin receptors on pancreatic acini, Biochim. Biophys. Acta 678 (1981) 358–363.
[28] N.R. Glover, A.S. Tracey, Nuclear magnetic resonance and restrained molecular dynamics studies of the interaction of an epidermal growth factor-derived peptide with protein tyrosine phosphatase 1B, Bio- chemistry 38 (1999) 5256–5271.
[29] B. Koksch, C. Dahl, G. Radics, A. Vocks, K. Arnold, J. Arnhold, J. Sieler, K. Burger, Fluoro-modified chemotactic peptides: fMLF ana- logues, J. Pept. Sci. 10 (2004) 67–81.
[30] M. Maeda, K. Kawasaki, T. Taguchi, Y. Kobayashi, Y. Yamamoto, K. Shimokawa, M. Takahashi, K. Nakao, H. Kaneto, Amino acids and pep- tides. XXIII. Leu-enkephalin analogs containing a fluorinated amino acid at position 2, 4 or 5, Chem. Pharm. Bull. (Tokyo) 42 (1994) 1658–1662.
[31] J.F. Parsons, G. Xiao, G.L. Gilliland, R.N. Armstrong, Enzymes har- boring unnatural amino acids: mechanistic and structural analysis of the enhanced catalytic activity of a glutathione transferase containing 5-fluorotryptophan, Biochemistry 37 (1998) 6286–6294.
[32] F. Yang, L. Yu, D.Y. He, C.A. Yu, Y. Shichida, Protein–ubiquinone interaction in bovine heart mitochondrial succinate-cytochrome c reductase. Synthesis and biological properties of fluorine substituted ubiquinone derivatives, J. Biol. Chem. 266 (1991) 20863–20869.
[33] N.C. Yoder, K. Kumar, Fluorinated amino acids in protein design and engineering, Chem. Soc. Rev. 31 (2002) 335–341.
[34] S.P. George, M.B. Rao, Conformation and polarity of the active site of xylanase I from Thermomonospora sp. as deduced by fluorescent chemoaffinity labeling: site and significance of a histidine residue, Eur. J. Biochem. 268 (2001) 2881–2888.
[35] D. Morikis, M. Roy, A. Sahu, A. Troganis, P.A. Jennings, G.C. Tso- kos, J.D. Lambris, The structural basis of compstatin activity exam- ined by structure–function-based design of peptide analogs and NMR, J. Biol. Chem. 277 (2002) 14942–14953.