PEDS Advance Access originally published online on April 12, 2007
Protein Engineering Design and Selection 2007 20(4):171-177; doi:10.1093/protein/gzm009
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
A chimeric cysteine protease of Plasmodium berghei engineered to resemble the Plasmodium falciparum protease falcipain-2
Department of Medicine, San Francisco General Hospital, University of California, PO Box 0811, San Francisco, CA 94143, USA
3 To whom correspondence should be addressed. E-mail: philip.rosenthal{at}ucsf.edu
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
The cysteine proteases falcipain-2 and falcipain-3 are hemoglobinases and potential targets for chemotherapy directed against Plasmodium falciparum, the most important human malaria parasite. Most in vivo evaluations of candidate antimalarials are conducted in murine malaria models, and falcipain homologs from rodent malaria parasites differ importantly from falcipain-2 and falcipain-3. We expressed berghepain-2, the single homolog of falcipain-2 and falcipain-3 of the rodent parasite P. berghei, in Escherichia coli, and characterized the refolded active enzyme. Berghepain-2 was biochemically very similar to the previously characterized rodent plasmodial protease vinckepain-2, but differed from falcipain-2 and falcipain-3 in its fine substrate and inhibitor specificity. We then used homology modeling and evolutionary trace analysis to predict key amino acids that mediate functional differences between falcipain-2 and berghepain-2. Thirteen amino acids were sequentially altered to replace berghepain-2 residues with those in falcipain-2. Mutant enzymes varied in activity and sensitivity to inhibitors. A berghepain-2 mutant with eight substitutions retained good activity and demonstrated fine substrate and inhibitor sensitivity more similar to that of falcipain-2 than berghepain-2. These results suggest that, to facilitate drug discovery, we can produce mutant animal model malaria parasites with biochemical properties more like those of the key drug target, P. falciparum.
Keywords: malaria/Plasmodium berghei/Plasmodium falciparum/protease
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Malaria is one of the most important infectious disease problems in the world (Greenwood et al., 2005
). A key factor responsible for the failure to control this disease is increasing resistance of malaria parasites to available drugs (Wongsrichanalai et al., 2002). Thus, the identification and characterization of new targets for antimalarial chemotherapy is an urgent need. Among potential new targets for chemotherapy are proteases that degrade hemoglobin in erythrocytic parasites, and thereby play a key role in the parasite life cycle (Rosenthal, 2002). Among the proteases that mediate hemoglobin hydrolysis are the cysteine proteases falcipain-2 (Shenai et al., 2000) and falcipain-3 (Sijwali et al., 2001b). Disruption of the falcipain-2 gene led to markedly diminished hemoglobin hydrolysis in early trophozoites, proving a role for the protease in hemoglobin processing (Sijwali and Rosenthal, 2004). Disruption of the falcipain-3 gene has not been possible, although replacement with a tagged functional copy was successful, suggesting that this protease activity is required by erythrocytic parasites (Sijwali et al., 2006). Cysteine protease inhibitors block hemoglobin hydrolysis and parasite development in vitro (Rosenthal et al., 1988) and cure mice with malaria (Rosenthal et al., 1993), validating the falcipains as promising drug targets.
A limitation of antimalarial drug discovery is that most in vivo studies must be performed in mice infected with rodent malaria parasites. Murine models are adequate for evaluating some classes of compounds, such as chloroquine analogues, but they are seriously deficient for studies of cysteine protease inhibitors, as the homologs of falcipain-2 and falcipain-3 are quite different in rodent parasites. Known rodent parasites contain a single falcipain-2/falcipain-3 homolog. Earlier we showed that vinckepain-2, a homolog from the rodent malaria parasite Plasmodium vinckei, differed importantly from falcipain-2 in its substrate preference and inhibitor sensitivity (Singh et al., 2002). We now report characterization of berghepain-2, the homolog from the more widely studied rodent parasite P. berghei; this enzyme has also recently been evaluated after expression as a fusion protein (Chan et al., 2005). In addition, with an eye toward improving our animal model, we engineered berghepain-2 to create an enzyme more similar to falcipain-2 and falcipain-3 than is the wild-type protease.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Materials
Synthetic peptide substrates were obtained from Bachem, and fluoromethyl ketones were a gift from Dr Robert Smith, Prototek. All other chemicals were purchased from Sigma.
Cloning and expression of berghepain-2
The berghepain-2 gene encodes 469 amino acids (GenBankTM AY063763
[GenBank]
). A berghepain-2 fragment (327 amino acids) encoding the complete mature domain of the protease and 73 C-terminal amino acids of the prodomain (–73 BP-2) was amplified from pTOP-BP2 (Singh et al., 2002) using berghepain-2 specific primers (FP 5'-TCTAGGATCCAACAATTTAGAATCTG-3' RP 5'-AATAAGCTTTTATTCAATTATAGGAGCATAACC-3'). The PCR fragment was digested with BamHI and HindIII, gel purified and ligated into the Escherichia coli expression vector pRSET-A (Invitrogen), which had been digested with the same enzymes to generate pRSET-BP-2. The construct was transformed into BL21 (DE3)-pLysS strain E. coli (Invitrogen), the insert sequence was verified by sequencing DNA isolated from multiple positive clones, one of the clones was grown overnight, and the culture was then treated with 0.5 mM isopropyl-ß-D-thiogalactosidase (IPTG) for 4 h at 37°C before harvesting the bacteria. Expression of protein was analyzed by SDS–PAGE.
A multiple sequence alignment of 57 papain-family cysteine proteases was generated by CLUSTALW (Thompson et al., 1994). This alignment was used to thread the full-length berghepain-2 and falcipain-2 sequences onto the structure of the human cysteine protease cathepsin K (Sivaraman et al., 1999). A homology model was then generated for berghepain-2 from the threaded alignment using MODELLER v4 (John and Sali, 2003). SPDBV Deep View was used to visualize and manipulate models and structures (Guex and Peitsch, 1997). Evolutionary trace analysis (Lichtarge et al., 1996), an approach to highlight important structural and functional residues using sequence conservation patterns, was performed on the same multiple sequence alignment via the software package JEvTrace (Joachimiak and Cohen, 2002). A homology model for berghepain-2, threaded through the recently solved structure of falcipain-2 (Wang et al., 2006), was used specifically for Fig. 1. The final homology model was generated by MODELLER v5. Mutations in berghepain-2 were added to the structure via SPDV Deep View. Images were composed in UCSF Chimera (Pettersen et al., 2004) before being rendered using Pixie v1.7.3 (http://www.cs.utexas.edu/~okan/Pixie/pixie.htm).
|
Construction of berghepain-2 mutants
In vitro site-directed mutagenesis was performed to generate a series of berghepain-2 mutants using the Quick-Change Multi Site-Directed Mutagenesis Kit (Stratagene) according to the manufacturer's instructions. Briefly, PCR was performed with Pfu Turbo DNA polymerase using pRSET-BP-2 DNA as a template and specific primers to introduce desired mutations (Table I), and template DNA was then digested with the methylase-specific DNAse DpnI at 37°C for 1 h. Berghepain-2 mutants with multiple mutations were generated in sequential steps using the same techniques. Mutated DNAs were then transformed into XL-10 Gold Ultra Competent E. coli (Stratagene), four to five recombinant clones were picked for each mutation and sequences were verified. Clones bearing the desired mutations were then transformed into BL21 (DE3)-pLysS cells, sequences were again verified and expression was carried out as described earlier.
|
Purification and refolding of wild-type and mutant berghepain-2
In all cases, recombinant proteins were expressed as insoluble inclusion bodies. Inclusion bodies were prepared by suspension of bacteria in 2 M Urea, 2.5% (v/v) Triton X-100, sonication and centrifugation at 12 000g for 20 min. Inclusion bodies were solubilized with Gu–HCl buffer (6 M guanine–HCl, 20 mM Tris–HCl, pH 8.5, 500 mM NaCl and 10 mM imidazole) and centrifuged, and soluble fractions were subjected to nickel-nitrilotriacetic acid (Ni-NTA) chromatography based on His-tags present in all recombinant proteases. Falcipain-2 and falcipain-3 were purified and refolded as described earlier (Shenai et al., 2000; Sijwali et al., 2001b). Refolding of wild-type and mutant berghepain-2 was carried out exactly as described earlier for vinckepain-2 (Singh et al., 2002). Briefly, for each protein, 2 mg of Ni-NTA purified protein was reduced with dithiothretol (DTT; 10 mM), diluted in 100 ml refolding buffer (100 mM Tris–HCl, pH 8.5, 1 mM EDTA, 400 mM L-arginine-HCl, 1 mM GSH and 0.5 mM GSSG), and incubated at 4°C for 21 h. The refolded samples were concentrated to 10 ml using a Centricon Plus-70 filter (Millipore) with a 10 kDa cut-off membrane and 30 ml holding capacity. To allow processing of active enzymes, the pH of the refolded samples was adjusted to 5.5, and DTT was added to a final concentration of 5 mM. The material was then run through 0.22 µm filters and incubated at 37°C for 2 h. The pH was adjusted to 6.5 with 1 M Tris–HCl, pH 8.0. The refolded and processed proteins were then applied to a Q-Sepharose column (bed volume 0.5 ml, Pharmacia), pre-equilibrated with 20 mM Bis-Tris–HCl, pH 6.5 and maintained at 4°C. The column was washed with at least 10 bed volumes of the same buffer, and protein was eluted with a 0–0.4 M linear gradient of sodium chloride. Fractions were collected and analyzed by SDS–PAGE. Activity was monitored by hydrolysis of benzyloxy carbonyl-Leu-Arg-7- N-4-methylcoumarin (Z-Leu-Arg-AMC). Active fractions were pooled, concentrated to 200 µl and stored (at concentrations of 0.1–0.2 µM) in 20 mM Bis-Tris–HCl, pH 6.5, with 50% glycerol at – 20°C.
Assays of enzyme activity and inhibition
Fluorometric assays of recombinant wild-type and mutant berghepain-2 were carried out as described earlier (Rosenthal et al., 1996), except for changes in substrates and enzyme concentrations, as below. In brief, reaction mixtures (0.35 ml) contained 20 µl of each refolded enzyme (final concentration 10 nM) in 100 mM sodium acetate, pH 5.5 (or with change in pH as described), 10 mM DTT. Reactions were started by adding fluorogenic substrates (Z-Leu-Arg-AMC or Z-Phe-Arg-AMC, 50 µM), and activity was measured as the release of AMC (excitation 355 nm; emission 460 nm) over 30 min at room temperature with a Labsystems Fluoroscan II spectrofluorometer. Activities were compared as fluorescence over time. To compare enzyme stabilities, enzymes were incubated in 100 mM sodium acetate, pH 5.5, 10 mM DTT, at the indicated time points 10 µl of the incubation mixture was removed and added to 290 µl of reaction buffer (100 mM sodium acetate, pH 5.5, 10 mM DTT and 50 µM Z-Leu-Arg-AMC), and activity was measured as described earlier. Assessments of inhibition of proteases by cysteine protease inhibitors were performed and IC50 determinations calculated as previously described (Shenai et al., 2003). For each enzyme, identical concentrations were used for assessment of the effects of each inhibitor on the hydrolysis of Z-Leu-Arg-AMC.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Expression, purification and biochemical characterization of berghepain-2
Berghepain-2 is the single homolog of falcipain-2 and falcipain-3 in P. berghei. The berghepain-2 gene has an open reading frame of 1407 bp that encodes a 469 amino acid protein, including a prodomain of 215 amino acids and mature domain (responsible for catalytic activity) of 253 amino acids, with an architecture similar to that of falcipains and homologous plasmodial cysteine proteases (Rosenthal, 2004). Its predicted sequence is very similar to that of vinckepain-2, the previously characterized homolog from the related murine parasite P. vinckei (78% identity of mature domains; Singh et al., 2002) and less similar to those of falcipain-2 (53%), falcipain-3 (49%) (Fig. 2), and homologs from other primate malaria species (Rosenthal et al., 2002). A construct encoding a portion of the prodomain and all of mature berghepain-2 was amplified, cloned into a plasmid vector and transformed into E. coli. The protease was expressed as insoluble protein, solubilized, purified by affinity chromatography and refolded using protocols similar to those used previously for falcipains (Sijwali et al., 2001a). The refolded protease was processed to an enzymatically active protein of predicted size for the mature form upon exposure to an acidic buffer (Fig. 3).
|
|
General features of berghepain-2 were very similar to those of vinckepain-2 expressed in the same manner (Singh et al., 2002
), and also similar to those recently reported for berghepain-2 expressed as a fusion with maltose-binding protein (Chan et al., 2005). The enzyme required a reducing environment and had an acidic pH optimum (Fig. 4A), as is typical for falcipains and other papain-family proteases. Berghepain-2 shared with falcipain-2, falcipain-3 and vinckepain-2 a strong preference for the cleavage of peptide substrates with Leu at the P2 position (Z-Leu-Arg-AMC). As with vinckepain-2, but unlike the P. falciparum enzymes, berghepain-2 had very low activity against Z-Phe-Arg-AMC. Consistent with this observation, berghepain-2 activity was strongly inhibited by the broad-acting cysteine protease inhibitors E-64 and leupeptin, but relatively weakly inhibited by the fluoromethylketone inhibitor Z-Phe-Arg-CH2F (Fig. 4B). None of the plasmodial cysteine proteases studied was inhibited by inhibitors of serine (PMSF), aspartic (pepstatin) or metallo (EDTA) proteases.
|
Modeling of berghepain-2 and identification of key residues for site-directed mutagenesis
We were interested in engineering berghepain-2 to make it more similar to falcipain-2 and potentially establish an animal model parasite of value for in vivo studies of protease inhibitors. We therefore modeled the structure of berghepain-2 on the solved structure of cathepsin K (Fig. 1). As is the case for falcipain-2 and falcipain-3, berghepain-2 has nine cysteine residues and conservation of papain-family active site residues (Q19, C25, H158, N174 and W176; papain numbering; Singh et al., 2002). Homology modeling identified 13 positions at which amino acids expected to have functional importance differed between berghepain-2 and falcipain-2. Evolutionary trace and multiple sequence alignment analyses were used to refine predictions and identify nine residues predicted to most critically mediate the substrate and inhibitor specificity differences observed between berghepain-2 and falcipain-2: K21 N, A23G, F61Y, P69 N, T130S, A132S, G135A, A156L and Q205D (for each residue, the P. berghei sequence is provided before and the P. falciparum sequence after the amino acid number, using papain numbering). Four additional mutations (N59K, E64 N, I67L and Y70 N) were deemed somewhat less likely to mediate differences between the enzymes.
Generation and characterization of mutant forms of berghepain-2
Site-directed mutagenesis was used to generate constructs with the nine single mutations of greatest interest and constructs containing combinations of these and four additional mutations. In each case, the wild-type berghepain-2 residue was replaced by that in falcipain-2 at the same location. All the mutant forms of berghepain-2 were expressed at approximately the same level as the wild-type protein, but the proteolytic activity achieved after refolding varied considerably (Table II). We previously found that, for vinckepain-2, the single substitution of A23G led to remarkably greater activity. Similarly, for berghepain-2, single substitutions at either positions 21 (K21 N) or 23 (A23G) yielded mutants with activity much greater than that of the wild-type enzyme. Other point mutations led to enzymes with activities similar to those of the wild-type enzyme (Table II). Interestingly, although the K21 N and A23G substitutions led to mutants with much higher activities than the wild-type enzyme, the double mutant containing both substitutions was much less active than either single mutant, with only a modest increase in activity compared with the wild-type enzyme. Consideration of mutants with multiple substitutions showed that one construct, MBP 2.13, which contained eight substitutions, retained activity near that of the wild-type enzyme. Additional mutations led to markedly diminished activity. Five mutants were tested against both Z-Leu-Arg-AMC and Z-Phe-Arg-AMC. Of note, MBP 2.13 was quite unstable (Fig. 5), but it was much more active against Z-Phe-Arg-AMC than was wild-type berghepain-2, demonstrating in this regard a substrate specificity more similar to that of falcipain-2 than wild-type berghepain-2 (Fig. 6).
|
|
|
Inhibition of wild-type and mutant berghepain-2
To further assess the impact of the substitution of berghepain-2 residues with amino acids found in falcipain-2, we compared the inhibition of falcipain-2, berghepain-2 and MBP 2.13 by two cysteine protease inhibitors, E-64 and Mu-Leu-Hphe-CH2F (Table III). Although the inhibitors were active against both enzymes, the degrees of inhibition differed markedly between falcipain-2 and berghepain-2. With falcipain-2, the IC50 for inhibition was about 50-fold lower for the fluoromethyl ketone compared with E-64, and for berghepain-2, the IC50 for the fluoromethyl ketone was 300-fold greater. Remarkably, MBP 2.13 was much less well inhibited by E-64 and much better inhibited by Mu-Leu-Hphe-CH2F than the wild-type enzyme, such that the ratio of IC50s for the two inhibitors was very similar to that of falcipain-2. Thus, although the extensive mutations placed in the MBP 2.13 enzyme led to some loss of stability, these alterations created an enzyme that based on activities against substrates and susceptibility to inhibitors, was considerably more similar to falcipain-2 than the wild-type enzyme.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Cysteine protease inhibitors directed against falcipains are promising new compounds for the treatment of falciparum malaria. However, a problem with drug discovery directed against these proteases is that animal model parasites contain proteases that are quite different from those of the key human parasite. We previously studied vinckepain-2, and showed that it had important differences from falcipain-2, in particular regarding substrate and inhibitor preferences. To expand these studies, and to consider the engineering of an improved animal model parasite, we studied berghepain-2, the homologous protease from the most commonly studied rodent malaria parasite, and we considered the possibility of engineering a mutant berghepain-2 that might lead to an animal model parasite with a protease repertoire more similar to that of P. falciparum. We found that berghepain-2 shares key features with its close homolog from P. vinckei and that mutagenesis modeled on falcipain-2 led to proteases with properties more similar to those of falcipain-2, albeit at times at the cost of significant loss of activity and stability of mutant proteases.
Recent work from other groups showed that, as expected, substitution of active site residues resulted in complete loss of falcipain-2 activity and replacement of two cysteines markedly decreased activity (Goh and Sim, 2004; Kumar et al., 2004). Substitutions at two other residues predicted to impact upon proteolysis had only modest effects, however, and so beyond the obvious importance of active site amino acids and disulfide bonds, the roles of individual residues in mediating the activity of falcipains has been uncertain. In order to attempt to design a more ‘falcipain-2-like’ berghepain-2, we used homology modeling and evolutionary trace analysis to predict key determinants of the fine specificity of related falcipain proteases. We then mutated berghepain-2 and evaluated the impacts of a number of mutations of interest.
Vinckepain-2 and berghepain-2 both contain a residue near the active site cysteine that is very unusual for papain family proteases. As with vinckeipain-2 (Singh et al., 2002), replacement of this residue with that seen in falcipain-2 and many related proteases (A23 
G) led to a dramatic gain in specific activity. A second nearby substitution (K21 N) led to a similarly large increase in activity compared with that of the wild-type enzyme. Surprisingly, a double mutant containing both these mutations was much less active than the single mutants, with specific activity similar to that of wild-type berghepain-2. This result highlights the interdependency of residues in mediating enzyme function and the risks inherent in attempting to augment activity via mutagenesis at multiple residues. As A23 is predicted to lie in the S1 pocket of berghepain-2 and K21 is also very near the active site, we predict that the single substitutions facilitated substrate entry into the active site pocket and thereby enhanced catalysis. The double mutant did not maintain the enhanced activity, presumably because of the extent of alteration in the protein brought about by the substitution of two key amino acids. Conservative substitution of another amino acid (T130 S), which is predicted to reside in the S2 pocket, led to a more modest increase in the specific activity of berghepain-2. Other single substitutions led to little change in specific activity or a modest loss of activity. Some changes in enzyme activity following mutagenesis were consistent with predictions whereas others were not easily explained, reflecting our limited appreciation of structure-function relationships, especially for studies based on modeled structures.
We were most interested in introducing multiple mutations in berghepain-2, each designed to substitute the wild-type sequence with that seen in falcipain-2 at residues predicted to impact on fine enzyme specificity. Our results were complex. Mutants with multiple mutations generally had significantly lower specific activity than did the wild-type protease. In addition, fine features of activity were altered in some mutants. Notably, with a berghepain-2 mutant containing eight amino acid substitutions (MBP 2.13), improved activity against the substrate Z-Phe-Arg-AMC, which is well hydrolyzed by falcipain-2, but not berghepain-2, was seen. Considering protease inhibitors, the MBP 2.13 mutant demonstrated a pattern of inhibition by cysteine protease inhibitors much more similar to that of falcipain-2 than berghepain-2, although the mutant enzyme was about an order of magnitude less sensitive than falcipain-2 (as measured by IC50 determination) to both E-64 and a vinyl sulfone inhibitor. These results must be interpreted with caution, as the multi-substituted berghepain-2 mutants had relatively low activity and stability, such that kinetics could not be studied in detail. Indeed, the poor stability of MBP 2.13 will limit its utility as a tool for selecting drug leads in an animal model using a rodent malaria parasite. Nonetheless, the results suggest that it may be possible to generate improved animal model proteases. With improved systems for gene replacement, particularly in P. berghei, replacement of the wild-type berghepain-2 with mutated berghepain-2 may be a reasonable strategy. Alternatively, the approach of replacing the berghepain-2 gene with intact falcipain-2 or falcipain-3 genes is worthy of study.
|
Footnotes |
|---|
1 Present address: Division of Infectious Diseases and Environmental Medicine, Thomas Jefferson University, Philadelphia, PA 19107, USA
2 Present address: Novartis Vaccines and Diagnostics Inc., 4560 Horton St., M/S 4.3, Emeryville, CA 94608-2916, USA
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
We thank Jiri Gut and Belinda J. Lee for excellent technical assistance. This work was supported by grants from the National Institutes of Health (AI35800 and RR01081) and Medicines for Malaria Venture. P.J.R. is a Doris Duke Charitable Foundation Distinguished Clinical Scientist.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionAcknowledgementsReferences |
|---|
Chan C., Goh L.L., Sim T.S. FEMS Microbiol. Lett (2005) 249:315–321.[CrossRef][Web of Science][Medline]
Goh L.L., Sim T.S. Biochem. Biophys. Res. Commun (2004) 323:565–572.[CrossRef][Web of Science][Medline]
Greenwood B.M., Bojang K., Whitty C.J., Targett G.A. Lancet (2005) 365:1487–1498.[CrossRef][Web of Science][Medline]
Guex N., Peitsch M.C. Electrophoresis (1997) 18:2714–2723.[CrossRef][Web of Science][Medline]
Joachimiak M.P., Cohen F.E. Genome Biol (2002) 3. RESEARCH0077.
John B., Sali A. Nucleic Acids Res (2003) 31:3982–3992.
Kumar A., Dasaradhi P.V., Chauhan V.S., Malhotra P. Biochem. Biophys. Res. Commun (2004) 317:38–45.[CrossRef][Web of Science][Medline]
Lichtarge O., Bourne H.R., Cohen F.E. J. Mol. Biol (1996) 257:342–358.[CrossRef][Web of Science][Medline]
Petterson E.F., Goddard T.D., Huang C.C., Couch G.S., Greenblatt D.M., Meng E.C., Ferrin T.E. J. Comput. Chem. (2004) 25:1605–1612.[CrossRef][Web of Science][Medline]
Rosenthal P.J. Curr. Opin. Hematol (2002) 9:140–145.[CrossRef][Web of Science][Medline]
Rosenthal P.J. Int. J. Parasitol. (2004) 34,:1489–1499.[CrossRef][Web of Science][Medline]
Rosenthal P.J., McKerrow J.H., Aikawa M., Nagasawa H., Leech J.H. J. Clin. Invest (1988) 82:1560–1566.[Web of Science][Medline]
Rosenthal P.J., Lee G.K., Smith R.E. J. Clin. Invest (1993) 91:1052–1056.[Web of Science][Medline]
Rosenthal P.J., Olson J.E., Lee G.K., Palmer J.T., Klaus J.L., Rasnick D. Antimicrob. Agents Chemother (1996) 40:1600–1603.[Abstract]
Rosenthal P.J., Sijwali P.S., Singh A., Shenai B.R. Curr. Pharm. Des (2002) 8:1659–1672.[CrossRef][Web of Science][Medline]
Shenai B.R., Sijwali P.S., Singh A., Rosenthal P.J. J. Biol. Chem (2000) 275:29000–29010.
Shenai B.R., Lee B.J., Alvarez-Hernandez A., Chong P.Y., Emal C.D., Neitz C.D., Roush W.R., Rosenthal P.J. Antimicrob. Agents Chemother (2003) 47:154–160.
Sijwali P.S., Rosenthal P.J. Proc. Natl Acad. Sci. USA (2004) 101:4384–4389.
Sijwali P.S., Brinen L.S., Rosenthal P.J. Protein Expr. Purif (2001a) 22:128–134.[CrossRef][Web of Science][Medline]
Sijwali P.S., Koo J., Singh N., Rosenthal P.J. Mol. Biochem. Parasitol (2006) 150:96–106.[CrossRef][Web of Science][Medline]
Sijwali P.S., Shenai B.R., Gut J., Singh A., Rosenthal P.J. Biochem. J (2001b) 360:481–489.[CrossRef][Web of Science][Medline]
Singh A., Shenai B.R., Choe Y., Gut J., Sijwali P.S., Craik C.S., Rosenthal P.J. Biochem. J (2002) 368:273–281.[CrossRef][Web of Science][Medline]
Sivaraman J., Lalumiere M., Menard R., Cygler M. Protein Sci (1999) 8:283–290.[Web of Science][Medline]
Thompson J.D., Higgins D.G., Gibson T.J. Nucleic Acids Res (1994) 22:4673–4680.
Wang S.X., Pandey K.C., Somoza J.R., Sijwali P.S., Kortemme T., Brinen L.S., Fletterick R.J., Rosenthal P.J., McKerrow J.H. Proc. Natl Acad. Sci. USA (2006) 103:11503–11508.
Wongsrichanalai C., Pickard A.L., Wernsdorfer W.H., Meshnick S.R. Lancet Infect. Dis (2002) 2:209–218.[CrossRef][Web of Science][Medline]
Received June 13, 2006; revised January 18, 2007; accepted January 30, 2007.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||