X-ray sequence and crystal structure of luffaculin 1, a novel type 1 ribosome-inactivating protein
© Hou et al; licensee BioMed Central Ltd. 2007
Received: 26 October 2006
Accepted: 30 April 2007
Published: 30 April 2007
Protein sequence can be obtained through Edman degradation, mass spectrometry, or cDNA sequencing. High resolution X-ray crystallography can also be used to derive protein sequence information, but faces the difficulty in distinguishing the Asp/Asn, Glu/Gln, and Val/Thr pairs. Luffaculin 1 is a new type 1 ribosome-inactivating protein (RIP) isolated from the seeds of Luffa acutangula. Besides rRNA N-glycosidase activity, luffaculin 1 also demonstrates activities including inhibiting tumor cells' proliferation and inducing tumor cells' differentiation.
The crystal structure of luffaculin 1 was determined at 1.4 Å resolution. Its amino-acid sequence was derived from this high resolution structure using the following criteria: 1) high resolution electron density; 2) comparison of electron density between two molecules that exist in the same crystal; 3) evaluation of the chemical environment of residues to break down the sequence assignment ambiguity in residue pairs Glu/Gln, Asp/Asn, and Val/Thr; 4) comparison with sequences of the homologous proteins. Using the criteria 1 and 2, 66% of the residues can be assigned. By incorporating with criterion 3, 86% of the residues were assigned, suggesting the effectiveness of chemical environment evaluation in breaking down residue ambiguity. In total, 94% of the luffaculin 1 sequence was assigned with high confidence using this improved X-ray sequencing strategy. Two N-acetylglucosamine moieties, linked respectively to the residues Asn77 and Asn84, can be identified in the structure. Residues Tyr70, Tyr110, Glu159 and Arg162 define the active site of luffaculin 1 as an RNA N-glycosidase.
X-ray sequencing method can be effective to derive sequence information of proteins. The evaluation of the chemical environment of residues is a useful method to break down the assignment ambiguity in Glu/Gln, Asp/Asn, and Val/Thr pairs. The sequence and the crystal structure confirm that luffaculin 1 is a new type 1 RIP.
Assignment of the sequence of luffaculin 1 by X-ray sequencing method (total number of residues: 241)
Sequence assignment method
Number of identified residues
Both molecule A and B have clear electron density and can be assigned unambiguously
Using method 1 plus electron density of the second molecule for those residues where the first molecule has weak electron density, 66% of residues (160) can be deduced
160 (9 additional residues were assigned: A123, A185, A206, A215, A219, A221, A222, A234 and B209)
Using methods 1 and 2 plus chemical environment evaluation without using sequence comparison, 86% of residues (208) can be deduced.
208 (exclude 28*, 29, 33*, 48*, 56*, 93*, 97, 109*, 111*, 122*, 132*, 142*, 156*, 157*, 169*, 177*,179, 190*, 197*, 202*, 204*, 205*, 213*, 216*, 217, 218, 220*, 223*, 225, 226*, 228, 237*, 241*)
Using method 3 plus the sequence comparison, 94% of residues (227) can be deduced.
227 (exclude 28*, 29, 97, 109**, 111**, 190*, 216*, 217, 218, 220*, 225, 228, 223*, 237*)
Neither molecule A nor B has clear density map, but the sequence is highly conserved among RIPs
4 (A&B109*, A&B179, A&B111*, A&B223*)
Neither molecule A nor B has clear density map, and the sequence is not conserved among RIPs
6 (A&B29, A&B97, A&B217, A&B218, A&B225, A&B228)
The ribosome-inactivating proteins (RIPs) are RNA N-glycosidases [5, 6] that inactivate ribosome by cleaving a single N-C glycosidic bond between adenine and ribose at A4324 in the 28S eukaryotic mammalian rRNA or at A2660 in the 23S Escherichia coli rRNA. The cleaved N-C glycosidic bond is located in a loop containing a GAGA sequence and highly conserved in rRNAs from bacteria, plants and animals. The removal of one adenine from rRNA by RIPs prevents the binding of elongation factor II (EF-2) to the 60S subunit, resulting in the termination of protein translation. In E. coli, the cleavage by RIP affects the combination of EF-Tu and EF-G. Both EF-G and EF-Tu protect bases in the universally conserved loop around position 2660 of 23S rRNA. This loop is also the site of action of cytotoxins that alter the structure of a region of rRNA that interacts with EF-Tu and EF-G and thus abolish protein synthesis. RIPs from plants can be classified into three types based on the structure of the genes and mature proteins . Type 1 RIPs, such as trichosanthin , bryodin , α, β-momorcharin [10, 11], luffin a and b  and cucurmosin , have alkaline isoelectric points and molecular weights ranging from 26 to 31 kDa. They typically contain a single polypeptide chain and have the potent ability to inhibit protein synthesis in the cell free system but are relatively non-toxic to the intact cells. Type 2 RIPs, such as ricin  and abrin , consist of two chains, chain A and chain B, linked by disulfide bridges. The A chain is homologous to type 1 RIPs and possesses the ribosome-inactivating activity; the B chain, containing a lectin domain, binds to galactosyl-terminated receptors on the target cell surface, facilitating the entry of the A chain into the cytoplasm of the cell. Thus, some, but not all, type 2 RIPs are more potent toxin than type 1 RIPs because type 1 RIPs have difficulty in entering into cells. Type 3 RIP includes JIP60  (jasmonate-induced protein) from maize, which consists of an N-terminal domain similar to type 1 RIPs and an unrelated C-terminal domain of unknown function. Most RIPs are glycoproteins, with varying amount and type of sugars.
RIPs have received wide attentions due to their potential therapeutic applications in medicine and transgenic reagents in agriculture. In medicine, they have been found to possess various pharmacological activities including abortifacient , antifungal , anti-tumor [19, 20], antivirus and HIV-1 integrase inhibitory activity [21, 22]. Plants transfected with RIP genes exhibit broad-spectrum resistance to viral and fungal infection [23, 24] in the plant defense system.
We identified a new type 1 RIP, luffaculin 1 . It is a basic protein with a pI of 8.86 by IEF analysis and has a molecular mass of about 28 kDa based on the mobility on SDS-PAGE. Luffaculin 1 not only possesses rRNA N-glycosidase activity as expected , but also inhibits proliferation of tumor cells, induces apoptosis  and differentiation on tumor cells .
Here, we report the high resolution (1.4 Å) crystal structure of luffaculin 1 and the protein sequence derived from this crystal structure. The structural comparison with other RIPs provides a structural basis to understand their possible biological activity. The amino-acid sequence of luffaculin 1 has not been determined by the traditional cDNA method. We demonstrated that the primary structure of luffaculin 1 can be derived with a high degree of confidence from the high-resolution electron density. The existence of two independent luffaculin 1 molecules in the asymmetric unit allows the cross-validation of this X-ray sequence, further increasing the reliability of the sequence assignment.
Results and discussion
Quality of the model
Data collection and model refinement statistics for luffaculin 1
a = 39.135 Å, b = 46.813 Å, c = 83.571 Å, α = 89.068°, β = 80.009°, γ = 72.143°
No. of water molecules
No. of carbohydrates
4 per asymmetric unit
No. of polyethylene glycols
R.m.s.deviations from ideal geometry
Bond lengths (Å)
Bond angles (°)
Fig. 1 shows a ribbon representation of luffaculin 1. The structure of luffaculin 1 contains two domains: a large N-terminal domain composed of eight α-helices and eight β-strands, and a smaller C-terminal domain consisting of two α-helices (α9 and α10) and two β-strands (β9 and β10). The secondary structure of luffaculin 1 is typical of type 1 RIPs: six β-strands of N-terminal domain (β1, β4, β5, β6, β7 and β8) form a mixed β-sheet. Eight helices of N-terminal have canonical geometry  and enclose the active site cleft. Helices α1 and α3 are part of the crossover connections between the parallel strands of the β-sheet. Helices α7 and α8 are contiguous in sequence and a single residue (Phe 163) assumes a non-helical dihedral conformation, introducing a bend between the two helices. Two β-strands of C-terminal domain (β9 and β10) are connected by a loop whose length varied among different RIPs.
The crystals of luffaculin 1 contain two enzyme molecules (A and B) in the asymmetric unit. A comparison of molecules A and B shows that the overall structures of these two molecules are almost identical (Fig. 2) with rmsd of 0.181 Å for 221 Cα atoms. Some deviation between two molecules occurs at the terminal of the α9-helix that is involved in the crystal packing.
X-ray sequence of luffaculin 1
Based on the electron density, majority of the residues can be assigned unambiguously. Some examples are illustrated in Fig. 3b for residues 46, 94 and 129. The two protein molecules (A and B) in the crystal should have the identical amino-acid sequence. This redundancy provides an additional level of validation on the electron-density-based sequence assignment (Fig. 3a and 3b). In the case that electron density is disordered or weak in one molecule, the electron density of the other molecule allows the identification of the sequence (9 residues are in such case, see Table 2 and Fig. 3c). Fig. 3c shows an example where residue 185 has weak electron density in molecule B but can be clearly identified as Ile in molecule A. 66% of residues (160 out of a total of 241 residues, see Table 2) of luffaculin 1 can be identified with confidence purely based on the high resolution electron density. Similar to our results, a previous study  suggested that 60% of residues can be identified reliably based on electron density.
Four residues (109, 111, 179 and 223) have weak electron density in both molecule A and molecule B (Table 2), and thus cannot be assigned based on electron density. All these four residues are located either at loop region or at the surface of the molecules. However, these four residues are highly conserved among various RIPs (Fig. 4), and are thus tentatively assigned according to the sequences of their homologous proteins. Residues (29, 97, 217, 218, 225 and 228) do not have enough electron density of side chains in both molecules and are not conserved in RIP sequences, and thus cannot be assigned in this study. They are currently tentatively assigned as Ala or Gly.
In summary, the evaluation of chemical environment greatly facilitates to break down the ambiguity (Table 2): 32 out of a total of 36 Val/Thr pairs and 16 out of a total of 38 Asp/Asn and Glu/Gln pairs were assigned by this method. By using electron density and evaluation of chemical environment, 86% of residues were assigned with confidence. Assignment based on sequence comparison, although not absolutely reliable, further increases the number of the identified residues to 227 (94% of a total of 241 residues).
X-ray sequencing method has been successfully used to determine the amino-acid sequence of PAP-Saci , a Pokeweed antiviral protein, and trichomaglin . The sequence of PAP-Saci was obtained from the exceptional quality of the electron density at 1.7 Å resolution, combined with the known sequence of the two PAP-S isoforms. The authors claimed that almost all amino-acid side chains were identified with a high degree of certainty (with the exception of Asp/Asn and Glu/Gln ambiguities). The X-ray sequence of trichomaglin was obtained by combining those derived from electron density at 2.2 Å resolution with the partial sequence information from mass spectroscopic analysis and the experimentally determined N-terminal sequence. 60% of the X-ray sequence was thus demonstrated to be highly reliable. In this paper we got the X-ray sequence of luffaculin 1 based on the high resolution (1.4 Å) electron density, cross-validated by the second molecule in the asymmetric unit of the crystals, and combined with the evaluation of the chemical environment of selected residues.
Comparison with other RIPs and the active site
Fig. 4 shows the structure-based alignment of luffaculin 1 with selected type 1 RIPs, including luffaculin 1, luffin a, luffin b, α-momorcharin, trichosanthin and bryodin. The amino-acid sequence of luffaculin 1 shows high degree of sequence identities to other RIPs: 94% for luffin a, 83% for luffin b, 73% for α-momorcharin, 64% for bryodin, and 63% for trichosanthin, respectively. The active site is the most conserved region at both sequence and structure level as observed. This indicates that the function and the enzymatic mechanisms of luffaculin 1 are probably the same as other RIPs .
We present the 1.4 Å resolution crystal structure of luffaculin 1 and its X-ray sequence. This sequence was derived based on the high resolution electron density, validated against the second molecule present in the crystals, the evaluation of the chemical environment of selected residues, and the sequence comparison with other homologues. A total of 86% (without using sequence comparison) or 94% (with sequence comparison) of luffaculin 1 residues can be assigned with confidence by this approach. The luffaculin 1 is quite similar to luffin a at both sequence and structural levels, suggesting its functions as an RIP.
Purification and crystallization
Luffaculin 1 was extracted and purified by extraction with acetate buffer, ammonium sulfate fractional precipitation and cation exchange chromatography . It was eluted as a single symmetrical peak in a cation exchange column (Mono S, Amersham Pharmacia Biotech) and gave a single band with an apparent molecular weight of about 28 kDa by reducing SDS-PAGE. The purified luffaculin 1 was thoroughly dialyzed against deionized water and lyophilized. For crystallization, lyophilized powder of luffaculin 1 was dissolved to a concentration of 15.8 mg/mL and then crystallized by hanging drop vapor diffusion method  at room temperature by mixing 2 μL protein (15.8 mg/mL) with an equal volume of reservoir solution (28% (w/v) PEG 6000, 0.1 M citrate buffer pH4.5, containing 0.02% (w/v) sodium azide) and equilibrating against 800 μL of the same reservoir solution. The crystal was briefly dipped into a cryoprotectant, 20% of glycerol (final concentration) in the reservoir solution, before data collection.
Data collection and processing
Diffraction data of the crystals were collected using synchrotron radiation (APS SER-CAT beamline 22ID) at low temperature (100K) to improve the diffraction quality and to decrease the radiation decay. A 1.4 Å resolution data set was obtained and the diffraction data were processed with the program package HKL2000 . The crystals belong to space group P1, with unit-cell parameters a = 39.135 Å, b = 46.813 Å, c = 83.571 Å, α = 89.068°, β = 80.009°, γ = 72.143°. Matthews coefficient calculations  show two molecules present in the asymmetric unit and the value of Vm is 2.49 Å Da-1 corresponding to a solvent content of 48%. The statistics for the data set are summarized in Table 1. The final merged data set of 94795 unique reflections has high quality with an Rmerge of 0.03 and an averaged signal to noise ratio of 21.8.
Structure determination and model refinement
The structure of luffaculin 1 was solved by the molecular replacement method (AMORE [41, 42]) using a homology model built based on the sequence of luffin a [43, 44]. Luffin a and luffaculin 1 were both purified from the same genus but different species in the Cucurbitaceae family, and were expected to share high sequence homology and structure similarity. Rotational solutions in the resolution range of 8-4 Å showed two clear peaks with the high correlation coefficients of 0.223 and 0.199, respectively, and the corresponding third highest value was 0.071. Then, the translation vector of the first solution was fixed because of the space group P1 of luffaculin 1 and the translation search was performed for the second solution, giving a higher correlation coefficient of 0.429. The refinement was performed with the program CNS  in a resolution range from 50-1.4 Å. A total of 5% of the data was randomly selected for Rfree calculation throughout the whole refinement. After a starting cycle of rigid body refinement, the R factor was 0.3360 and Rfree was 0.3518. Simulated annealing and restrained individual B-factor refinement were then performed. The sigma A weighted 2Fo-Fc and Fo-Fc electron density maps were used to guide the model building process. The model was examined and manually rebuilt with the graphic program O . In the final stage of the refinement, water molecules were added by CNS at locations where electron density was stronger than 3.0σ in sigma A-weighted Fo-Fc maps and had reasonable hydrogen-bond interaction with the protein, and then were inspected with program O. The carbohydrates and the polyethylene glycol (PEG) molecules [see Additional file 1] were visible at this stage in 2Fo-Fc and Fo-Fc electron density maps and were built in. Luffaculin 1 was crystallized from the solution containing PEG6000. It is inevitable that there are some low molecular weight polyethylene glycol molecules existed in the PEG6000 and these small molecules most likely penetrate into crystals. Such cases can also be found in the PDB data bank, for example, in PDB entry 2FD6 . The final model has R factor and Rfree of 0.213 and 0.232, respectively, containing 492 water molecules, one PEG1 (tetraethylene glycol), two PEG2 (diethylene glycol) and four N-acetylglucosamines (NAG) in the asymmetric unit. Data collection and model refinement statistics are listed in Table 1.
Financial support from the National Science Foundation of China (No. 39970872, 30625011), the Natural Science Foundation of Fujian Province (C97052), Special Fund of Fujian Development and Reform Commission and State Key Laboratory of Structural Chemistry, and NSF-EPSCoR of USA are gratefully acknowledged. Use of the Advanced Photon Source was supported by the US Department of Energy, Office of Science, Office of Basic Energy Sciences, under contract No.W-31-109-Eng-38. We thank Yujun Wang of University of Alabamaba in Huntsville as well as the staffs of the APS SER-CAT beamline 22ID for help with data collection. The coordinates of luffaculin 1 have been deposited in PDB (code 2OQA).
- Liang SP: Techniques of protein sequencing at the turn of the centuries. Chinese Bulletin of Life Sciences 1999, 11(1):31–35.Google Scholar
- Chambery A, de Donato A, Bolognesi A, Polito L, Stirpe F, Parente A: Sequence determination of lychnin, a type 1 ribosome-inactivating protein from Lychnis chalcedonica seeds. Biological chemistry 2006, 387(9):1261–1266.View ArticlePubMedGoogle Scholar
- Hogg T, Kuta Smatanova I, Bezouska K, Ulbrich N, Hilgenfeld R: Sugar-mediated lattice contacts in crystals of a plant glycoprotein. Acta Crystallogr D Biol Crystallogr 2002, 58(Pt 10 Pt 1):1734–1739.View ArticlePubMedGoogle Scholar
- Gan JH, Yu L, Wu J, Xu H, Choudhary JS, Blackstock WP, Liu WY, Xia ZX: The three-dimensional structure and X-ray sequence reveal that trichomaglin is a novel S-like ribonuclease. Structure 2004, 12(6):1015–1025.View ArticlePubMedGoogle Scholar
- Barbieri L, Battelli MG, Stirpe F: Ribosome-inactivating proteins from plants. Biochim Biophys Acta 1993, 1154(3–4):237–282.View ArticlePubMedGoogle Scholar
- Stirpe F, Barbieri L: Ribosome-inactivating proteins up to date. FEBS Lett 1986, 195(1–2):1–8.View ArticlePubMedGoogle Scholar
- Damme EJMV, Hao Q, Chen Y, Barre A, Vandenbussche F, Desmyter S, Rougé P, Peumans WJ: Ribosome-Inactivating Proteins: A Family of Plant Proteins That Do More Than Inactivate Ribosomes. Critical Reviews in Plant Sciences 2001, 20(5):395–465.View ArticleGoogle Scholar
- Pan KZ, Lin YJ, Zhou KJ, Fu ZJ, Chen MH, Huang DR, Huang DH: The crystal and molecular structure of trichosanthin at 2.6 A resolution. Sci China B 1993, 36(9):1069–1081.PubMedGoogle Scholar
- Stirpe F, Barbieri L, Battelli MG, Falasca AI, Abbondanza A, Lorenzoni E, Stevens WA: Bryodin, a ribosome-inactivating protein from the roots of Bryonia dioica L. (white bryony). Biochem J 1986, 240(3):659–665.PubMed CentralView ArticlePubMedGoogle Scholar
- Husain J, Tickle IJ, Wood SP: Crystal structure of momordin, a type I ribosome inactivating protein from the seeds of Momordica charantia. FEBS Lett 1994, 342(2):154–158.View ArticlePubMedGoogle Scholar
- Yuan YR, He YN, Xiong JP, Xia ZX: Three-dimensional structure of beta-momorcharin at 2.55 A resolution. Acta Crystallogr D Biol Crystallogr 1999, 55 (Pt 6): 1144–1151.View ArticleGoogle Scholar
- Masahiro K, Hiromi N, Gunki F: Isolation and Characterization of Two Luffins, Protein-biosynthesis Inhibitory Proteins from the seeds of Luffa cylindrica. Agric Biol Chem 1988, 52(5):1223–1227.View ArticleGoogle Scholar
- Chen MH, Ye XM, Cai JH, Lin YJ: Crystallization and preliminary crystallographic study of cucurmosin, a ribosome-inactivating protein from the sarcocarp of Cucurbita moschata. Acta Crystallogr D Biol Crystallogr 2000, 56 (Pt 5): 665–666.View ArticleGoogle Scholar
- Rutenber E, Katzin BJ, Ernst S, Collins EJ, Mlsna D, Ready MP, Robertus JD: Crystallographic refinement of ricin to 2.5 A. Proteins 1991, 10(3):240–250.View ArticlePubMedGoogle Scholar
- Tahirov TH, Lu TH, Liaw YC, Chen YL, Lin JY: Crystal structure of abrin-a at 2.14 A. J Mol Biol 1995, 250(3):354–367.View ArticlePubMedGoogle Scholar
- Chaudhry B, Muller-Uri F, Cameron-Mills V, Gough S, Simpson D, Skriver K, Mundy J: The barley 60 kDa jasmonate-induced protein (JIP60) is a novel ribosome-inactivating protein. Plant J 1994, 6(6):815–824.View ArticlePubMedGoogle Scholar
- Jin YC: Intra-amniotic injection of crystal trichosanthin for induction of labour in second trimester pregnancy. Shengzhi Yu Biyun 1985, 5(1):15–17, 20.PubMedGoogle Scholar
- Leah R, Tommerup H, Svendsen I, Mundy J: Biochemical and molecular characterization of three barley seed proteins with antifungal properties. J Biol Chem 1991, 266(3):1564–1573.PubMedGoogle Scholar
- Bolognesi A, Polito L: Immunotoxins and other conjugates: pre-clinical studies. Mini Rev Med Chem 2004, 4(5):563–583.View ArticlePubMedGoogle Scholar
- Pastan II, Kreitman RJ: Immunotoxins for targeted cancer therapy. Adv Drug Deliv Rev 1998, 31(1–2):53–88.View ArticlePubMedGoogle Scholar
- Au TK, Collins RA, Lam TL, Ng TB, Fong WP, Wan DC: The plant ribosome inactivating proteins luffin and saporin are potent inhibitors of HIV-1 integrase. FEBS Lett 2000, 471(2–3):169–172.View ArticlePubMedGoogle Scholar
- Wang JH, Nie HL, Tam SC, Huang H, Zheng YT: Anti-HIV-1 property of trichosanthin correlates with its ribosome inactivating activity. FEBS Lett 2002, 531(2):295–298.View ArticlePubMedGoogle Scholar
- Parikh BA, Tumer NE: Antiviral activity of ribosome inactivating proteins in medicine. Mini Rev Med Chem 2004, 4(5):523–543.View ArticlePubMedGoogle Scholar
- Wang P, Tumer NE: Virus resistance mediated by ribosome inactivating proteins. Adv Virus Res 2000, 55: 325–355.View ArticlePubMedGoogle Scholar
- Lin JK, Chen MH, Xie JM, Zhao R, Ye XM, Shi XL, Wang ZR: Purification and Characterization of Two Luffaculins, Ribosome-inactivating Proteins from Seeds of Luffa acutangula. Chinese Journal of Biochemistry and Molecular Biology 2002, 18(5):609–613.Google Scholar
- Chen MH, Shi XL, Ye XM, Xie JM, Zhao R, Rao PF, Wang ZR: Study on the secondary structure and bioactivities of Luffaculin 1. Chinese J Struct Chem 2004, 23: 232–235.Google Scholar
- Zhao R, Xie JM, Xu JH, Chen MH, Lin JK, Ye XM: Apoptosis-inducing effect of luffaculin on K562 cells. Pharmacology and Clinics of Chinese Materia Medica 2001, 17(6):24–26.Google Scholar
- XIE JM, Chen MH, Xu Y, Zhao R, Xu JH, Yu ZX: Differentiation-inducing effect of Luffaculin-1 on B16 melanoma cells. Chin J Cancer Prev Treat 2005, 12(20):1521–1524.Google Scholar
- Laskowski RA, Rullmannn JA, MacArthur MW, Kaptein R, Thornton JM: AQUA and PROCHECK-NMR: programs for checking the quality of protein structures solved by NMR. J Biomol NMR 1996, 8(4):477–486.View ArticlePubMedGoogle Scholar
- Savino C, Federici L, Ippoliti R, Lendaro E, Tsernoglou D: The crystal structure of saporin SO6 from Saponaria officinalis and its interaction with the ribosome. FEBS Lett 2000, 470(3):239–243.View ArticlePubMedGoogle Scholar
- Rudd PM, Wormald MR, Stanfield RL, Huang M, Mattsson N, Speir JA, DiGennaro JA, Fetrow JS, Dwek RA, Wilson IA: Roles for glycosylation of cell surface receptors involved in cellular immune recognition. J Mol Biol 1999, 293(2):351–366.View ArticlePubMedGoogle Scholar
- Wormald MR, Dwek RA: Glycoproteins: glycan presentation and protein-fold stability. Structure 1999, 7(7):R155–60.View ArticlePubMedGoogle Scholar
- Brunger AT, Adams PD, Clore GM, DeLano WL, Gros P, Grosse-Kunstleve RW, Jiang JS, Kuszewski J, Nilges M, Pannu NS, Read RJ, Rice LM, Simonson T, Warren GL: Crystallography & NMR system: A new software suite for macromolecular structure determination. Acta crystallographica 1998, 54(Pt 5):905–921.Google Scholar
- Jones TA, Zou JY, Cowan SW, Kjeldgaard M: Improved methods for building models in electron density maps and the location of errors in these models. Acta Crystallogr A 1991, 47: 110–119.View ArticlePubMedGoogle Scholar
- Monzingo AF, Robertus JD: X-ray analysis of substrate analogs in the ricin A-chain active site. J Mol Biol 1992, 227(4):1136–1145.View ArticlePubMedGoogle Scholar
- Fermani S, Falini G, Ripamonti A, Polito L, Stirpe F, Bolognesi A: The 1.4 anstroms structure of dianthin 30 indicates a role of surface potential at the active site of type 1 ribosome inactivating proteins. Journal of structural biology 2005, 149(2):204–212.View ArticlePubMedGoogle Scholar
- Robertus JD, Monzingo AF: The structure of ribosome inactivating proteins. Mini Rev Med Chem 2004, 4(5):477–486.View ArticlePubMedGoogle Scholar
- Mcpherson A: Preparation and Analysis of Protein Crystals. New York , John Wiley; 1982:82–159.Google Scholar
- Otwinnowski Z, Minor W: Processing of X-ray diffraction data collected in oscillation model. Methods Enzymol 1997, 276: 307–326.View ArticleGoogle Scholar
- Matthews BW: Solvent content of protein crystals. J Mol Biol 1968, 33(2):491–497.View ArticlePubMedGoogle Scholar
- Navaza J: AmoRe-an automated package for molecular replacement. Acta Crystallogr A 1994, 50: 157–163.View ArticleGoogle Scholar
- Navaza J: Implementation of molecular replacement in AMoRe. Acta Crystallogr D Biol Crystallogr 2001, 57(Pt 10):1367–1372.View ArticlePubMedGoogle Scholar
- Guex N, Peitsch MC: SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling. Electrophoresis 1997, 18(15):2714–2723.View ArticlePubMedGoogle Scholar
- Schwede T, Kopp J, Guex N, Peitsch MC: SWISS-MODEL: An automated protein homology-modeling server. Nucleic Acids Res 2003, 31(13):3381–3385.PubMed CentralView ArticlePubMedGoogle Scholar
- Huai Q, Mazar AP, Kuo A, Parry GC, Shaw DE, Callahan J, Li Y, Yuan C, Bian C, Chen L, Furie B, Furie BC, Cines DB, Huang M: Structure of human urokinase plasminogen activator in complex with its receptor. Science (New York, NY 2006, 311(5761):656–659.View ArticleGoogle Scholar
- DeLano WL: The PyMOL Molecular Graphics System.2002. [http://www.pymol.org]Google Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.