A tutorial for learning and teaching macromolecular crystallography - version 2010
Annette Faust, Sandra Puehringer, Nora Darowski, Santosh Panjikar, Venkataraman Parthasarathy, Andrea Schmidt, Victor S. Lamzin, Kay Diederichs,
Uwe Mueller and Manfred S. Weiss
References: A. Faust etal. (2008). J. Appl. Cryst. 41, 1161-1172.
A. Faust et al. (2010). J. Appl. Cryst. 43 (in press).
Experiment 3: Molecular Replacement on monoclinic
Lysozyme is a 129 amino acid enzyme that dissolves bacterial cell walls by catalyzing the hydrolysis of l,4-|3-linkages between N-acetylmuramic acid and N-acetyl-D-glucosamine residues in the peptidoglycan layer and between N-acetyl-D-glucosamine residues in chitodextrins. It is abundant in a number of secreted fluids, such as tears, saliva and mucus. Lysozyme is also present in cytoplasmic granules of the polymorphonuclear neutrophils (Voet et al., 2006). Large amounts of lysozyme can also be found for instance in egg-whites. The crystal structure of hen egg-white lysozyme (HEWL) based on crystals belonging to the tetragonal space group P432i2, was the first enzyme structure published (Blake et al., 1965). Over the years, HEWL has been crystallized in many different crystal forms (for an overview see Brinkmann et al, 2006) and has become a standard object for methods developments but also for teaching purposes.
10 20 30 40 50 60 70
KVFGRCELAAAMKRHGLDNYRGYSLGNWVCAAKFESNFNTQATNRNTDGSTDYGILQINSRWWCNDGRTP
80 90 100 110 120 129
GSRNLCNIPCSALLSSDITASVNCAKKIVSDGNGMNAWVAWRNRCKGTDVQAWIRGCRL
Figure 1: Amino acid sequence of hen egg-white lysozyme.
In this experiment, the structure of monoclinic HEWL is determined by Molecular Replacement (MR) using the structure of tetragonal HEWL as a search model. MR is a method to determine a structure in cases where a similar structure is already known. If the similar structure can be correctly oriented and positioned in the unit cell of the structure to be solved, it can be used as a starting point for phase calculation and refinement. Currently, about two thirds of all new structures deposited with the PDB (Berman et al, 2000) are solved using MR (Long et al, 2008).
1 Crystallisation
Chemicals:    hen egg-white lysozyme (M ~ 14600 g/mol, Fluka cat. no. 62970) CH3COONa (M = 82.03 g/mol, Sigma cat. no. S2889) CH3COOH (M = 60.0 g/mol, Sigma cat.no. 537020) NaN03 (M = 84.99 g/mol, Sigma cat. no. S5506) Milli-Q water
Paraffin oil (Fluka cat. no. 76235)
Monoclinic HEWL (Mr = 14.6 kDa, Fluka cat. no. 62970) crystals (Figure 2) were prepared according to a recipe described before (Saraswathi et al., 2002) by mixing 12 |u.l of protein solution (20 mg/ml lysozyme in 50 mM sodium acetate pH 4.5) and 12 |u.l of reservoir solution containing 50 mM sodium acetate pH 4.5 and 4% sodium nitrate and equilibrating the drop against the reservoir. Crystals belonging to the monoclinic space group P2i with unit-cell parameter a = 27.4 A, b = 62.3 A, c = 59.5 A and |3 = 90.5° grew within a few days. They were cryo-protected using paraffin oil and usually diffracted X-rays to better than 1.3 A.
			A	
200nm			300nm	
Figure 2: Monoclinic HEWL crystals.
2 Data Collection
X-ray diffraction data have been collected at beam line BL 14.2 at the BESSY-II synchrotron in Berlin Adlershof. The beam line was at the time of this experiment equipped with a MARCCD detector (165mm) from the company MARRESEARCH (Norderstedt, Germany) and a MARdtb goniostat (MARRESEARCH, Norderstedt, Germany).
The relevant data collection parameters are given below:
wavelength 0.9 A
detector distance: 100 mm
oscillation range/image: 1.0°
no. of images: 360
exposure time/image: 3.4 sec
path to images: exp3/data
image names: exp3_lyso_molrep_l_###.img
Figure 3: Diffraction image of a crystal of monoclinic HEWL displayed using different contrast levels. The resolution rings are shown at 5.4, 2.7, 1.8 and 1.4A, respectively.
3 Data Processing
The collected diffraction data were indexed, integrated and scaled using the program XDS (Kabsch, 1993, 2010a,b). XDS is simply run by the command xds. If a multi-processor machine is available, the command xds_par can be used, which calls a parallel version of XDS and consequently runs much faster. XDS needs only one input file, which must be called XDS.INP. No other name is recognized by the program. The file XDS.INP contains all relevant information about the data collection, from beam parameters to detector parameters and crystal parameters (if known) as well as the data collection geometry. In XDS.INP one can also define the steps through which the program should go. This is done by using the parameter JOBS. The following command, which is equivalent to JOBS= ALL would make XDS run through all eight steps XYCORR, INIT, COLSPOT, IDXREF, DEFPIX, XPLAN, INTEGRATE and CORRECT.
JOBS= XYCORRINIT COLSPOT IDXREF DEFPIX XPLAN INTEGRATE CORRECT
In the XYCORR step, tables of spatial correction factors are set up (if required). INIT calculates the gain of the detector and produces an initial background table. COLSPOT identifies strong reflections which are used for indexing. IDXREF performs the actual indexing of the crystal. DEFPIX identifies the regions on the detector surface which are used for measuring intensities, XPLAN helps to devise a data collection strategy, INTEGRATE integrates the reflection intensities of the whole data set and CORRECT scales and merges symmetry-related reflections and multiple measurements. It also prints out data processing statistics. After completing each individual step, a log-file with a name corresponding to the step {STEP-namely) is written.
Action 1: edit the supplied file XDS.INP and insert the relevant information about the data collection, namely the data collection wavelength, crystal-to-detector distance, the direct beam coordinates, the total number of images, and rotation increment per image and of course very importantly the path to and the names of the image files. XDS is able to recognize compressed images; therefore it is not necessary to unzip the data before using XDS. The image name given must not include the zipping-format extension (*.img instead of *.img.bz2). Further, XDS has a very limited string length (80) to describe the path to the images. Therefore it may be necessary to create a soft link to the directory containing the images by using the command In -s /path/to/images/ ./images. The path to the images in XDS.LNP will then be ./images/. If the space group and cell dimensions are known, the relevant information should be written into XDS.LNP, if they are not known just set the parameter SPACE GROUP NUMBER= 0.
Action 2: run XDS until the indexing step, with the parameter JOBS set to:
JOBS= XYCORR INIT COLSPOT IDXREF
The output file IDXREF.LP contains the results of the indexing. It needs to be checked carefully whether the indexing is correct, since all subsequent steps assume the correctness of the indexing step. The most relevant parameters to look for are the STANDARD DEVIATION OF SPOT POSITION and the STANDARD DEVIATION OF SPrNDLE POSITION. The first one should be in the order of 1 pixel, whereas the second one depends to some extent on the rotation increment per image but also on the mosaicity of the crystal. If it is 0.1° it is very good, if it is 0.5° it might still be ok, if it is larger than 1.0° the indexing has probably not worked. The table with the entries SUBTREE and POPULATION is also very interesting to look at. The first SUBTREE should have by a large margin more entries than all others. Also, the input parameters, such as the crystal-to-detector distance should after refinement not deviate too much from the input values.
The most common problem with the IDXREF step is that it often finishes with the message !!! ERROR !!! INSUFFICIENT PERCENTAGE (<70%) OF INDEXED REFLECTIONS
This means that less than 70% of the reflections that were collected in the COLSPOT step are not indexed, which may happen because of ice rings on the frames, split reflections or simply wrong input parameters. However, if all indicators of correct indexing are fine (see above) and no obvious errors can be identified then this message can be safely ignored and data processing can be continued.
If IDXREF was run with SPACE_GROUP_NUMBER= 0, an assumption of the correct Bravais lattice may be made at this stage. As a rule of thumb, choose the lattice of the highest possible symmetry, with a QUALITY OF FIT-value as low as possible (usually < 10). These numbers are printed in IDXREF.LP in the paragraph DETERMINATION OF LATTICE CHARACTER AND BRAVAIS LATTICE. Then, re-run the IDXREF step with the parameter SPACE GROUP NUMBER corresponding to the assumed Bravais lattice in XDS.LNP. Alternatively, it is possible to make no assumption of the Bravais lattice at this point, and to simply continue with data integration. In this case the program will assume that space group determination should not be based on the metric symmetry of the lattice, but should be postponed to the CORRECT step (see below).
Nevertheless, if the true lattice is known, it should yield a good, i.e. low QUALITY OF FIT-number. For monoclinic HEWL the correct space group is P2i (space group number 4) with unit cell parameters a = 27.40 A, b = 62.30 A, c = 59.50 A and p = 90.50°
Action 3: After the determination of the Bravais lattice and the cell parameters all images need to be integrated and corrections (radiation damage, absorption, detector etc.) will have to be calculated. This can be done in a further XDS run.
JOBS= DEFPIX XPLAN INTEGRATE CORRECT
The CORRECT step produces a file called CORRECT.LP, which contains the statistics for the complete data set after integration and corrections. For the statistics to be meaningful, the correct Laue symmetry has to be established first. To this end, the CORRECT step compares the statistics in all possible Laue groups. The correct Laue group is the one with the highest symmetry, which at the same time still exhibits an acceptable Rr.i.m./Rmeas- CORRECT writes a file named XDSASCII.HKL, which contains the integrated and scaled reflections.
The CORRECT step also performs a refinement of all geometric parameters and the cell dimensions based on all reflections of the data set. These parameters may be more accurate than the ones obtained from the indexing step. Therefore, one may try to use the refined parameters and to re-run the last XDS job. In order to not overwrite the original results, it is advisable to save all current files to a temporary directory. Then, the file GXP ARM.XDS should be renamed or copied to XPARM.XDS and XDS be re-run. In case the original results are better, they can be copied back to the original directory.
While XDS will usually identify the correct Laue group, it does not determine the actual space group of the crystal. The decision about the existence of screw axes is left to the user. Indications, which screw axes may be present can be obtained from the table REFLECTIONS OF TYPE H,0,0 0,K,0 0,0,L OR EXPECTED TO BE ABSENT (*) in the file CORRECT.LP. Alternatively, the program POINTLESS (Evans, 2005) offers an automatic way of assigning the space group. POINTLESS can be run with the command pointless XDSIN XDS_ASCII.HKL. In the output the possible space groups together with their probabilities are given. Some space group ambiguity still remains at this stage, since it is impossible to distinguish between enantiomorphic space groups e.g. P3i and P32, or P4i2i2 and P432i2 just based on intensities. This ambiguity has to be resolved later during structure solution. The parameter SPACEGROUPNUMBER corresponding to the determined space group as well as the cell parameters should be entered into the file XDS.INP for running the next step.
Action 4: Finally, outlier reflections are identified by CORRECT by comparing their intensity to the average intensity in their respective resolution shells. These outliers may be removed, if there is a clear indication and reason for their existence: for example, ice rings often produce very strong reflections at specific d-spacings. The outliers are flagged as 'alien' in the file CORRECT.LP and their removal can simply be achieved by writing the outliers into a file called REMOVE.HKL. By re-running XDS with the command
JOBS= CORRECT
in XDS.INP, these outliers are then disregarded. This last action can be repeated until no more additional outliers are identified. However, the outlier removal has to be handled very carefully because strong reflections may also arise from non-crystallographic symmetry and in particular from the presence of pseudo-translation. A command to identify only the most extreme outliers would be awk '/alien/ {if (strtonum($5) > 19) print $0}' CORRECT.LP » REMOVE.HKL This command will remove outliers only when their Z-score is above 19. Hints to suitable criteria for outlier rejection can be found in the XDSwiki (http://strucbio.biologie.uni-konstanz.de/xdswiki), where this question is treated specifically in the article "Optimization".
Action 5:, The CORRECT step can be followed up by running the scaling program XSCALE, which is part of the XDS program package. This serves three purposes: a) the user may specify the limits of the resolution shells for which statistics should be printed, b) several XDSASCII.HKL files may be scaled together and c) correction factors for radiation damage may be applied to the data (see also the article "XSCALE" in the XDSwiki). XSCALE is run by simply typing xscale (or xscale_par to speed up the computation on a multi-processor machine) provided that a file XSCALE. INP defining the input and output files is present. As above in the CORRECT step, outliers may be rejected. XSCALE writes out a *.ahkl file, which can be converted with XDSCONV to be used within the CCP4-suite (Collaborative Computational Project, 1994) or other programs.
Both CORRECT and XSCALE will produce all necessary output for assembling a table with all relevant data processing statistics, which is necessary for a publication.
Table 1: Data processing statistics (from XSCALE.LP).
o Resolution limits [A]	10.0 - 1.60 (1.70-1.60)
o Unit cell parameters, a, b, c, ß [A,°]	27.40, 62.30, 59.50, 90.50
Space group	P2i
Mosaicity [°]	0.5
Total number of reflections	317,620
Unique reflections	43,663
Redundancy	7.3 (5.5)
Completeness [%]	99.4 (94.5)
I/o(D	27.4 (4.7)
Rr.i.m. / Rmeas [%]	4.2 (40.2)
Wilson B-factor [A2]	18.9
Action 6: finally, the processed intensity file needs to be converted to certain file formats, which are used by other programs to perform the necessary structure determination steps. This can be achieved using the program XDSCONV, which can simply be run by using the command xdsconv provided that a file called XDSCONV.INP is present. XDSCONV.INP just needs to contain information about the name of the input file and about the name and type of the output file. If a CCP4-type file is required, XDSCONV reformats the reflection output file from XSCALE and creates an input file F2MTZ.INP for the final conversion of the reflection file to binary mtz-format, which is the standard format for all CCP4 programs (CCP4, 1994).
OUTPUT_FILE=lyso_molrep.hkl CCP4
INPUT_FILE=lysp_molrep.ahkl_
To run the CCP4 programs F2MTZ and CAD, just type the two commands
flmtz HKLOUT temp.mtz < F2MTZ.INP
and
cad HKLIN1 temp.mtz HKLOUT lyso_molrep_ccp4.mtz « eof
LABIN FILE 1 ALL
END
eof
Some CCP4 programs need the intensities of the Bijvoet pairs as input. For those the second parameter on the OUTPUT_FILE= line should be CCP4_I instead of CCP4. Alternatively, the file XDSASCII.HKL can be converted to mtz-format using the CCP4-programs COMBAT or POTNTLESS (Evans, 2005) and this mtz-file can be used as an input file for the scaling program SCALA (Evans, 2005) in CCP4. More information on this can be found in the articles "Pointless" and "Scaling with SCALA" in the XDSwiki.
4 Structure Solution
The structure can be solved using the MR-protocol (run in the advanced version) of AUTO-RICKSHAW: the EMBL-Hamburg automated crystal structure determination platform (Panjikar et al, 2005; 2009). AUTO-RICKSHAW can be accessed from outside EMBL under www.embl-hamburg.de/Auto-Rickshaw/LICENSE (a free registration may be required, please follow the instructions on the web page). In the following the automatically generated summary of AUTO-RICKSHAW is printed together with the results of the structure determination:
The structure was solved using the MR-protocol of Auto-Rickshaw with tetragonal HEWL (PDB entry 193L, Vaney et al., 1996) as a starting model. The input diffraction data (file XDSASCII.HKL) were uploaded and then prepared and converted using programs of the CCP4-suite. The molecular replacement step was done using MOLREP (Vagin and Teplyakov, 1997) with a resolution cut-off of 4 A to find the two molecules in the asymmetric unit. Despite a very high initial R-factor of 73% (correlation coefficient 43%), the solution was correct as was demonstrated by subsequent refinement. This was performed to a resolution of 3.0 A using the program CNS (Bruenger et al, 1998) in four consecutive steps: rigid body refinement, a minimization step, B-factor refinement and a second minimization step. At this point the R- and Rfree-values were 24.9 and 33.5%, respectively. Further refinement was done in REFMAC5 using all available data to R- and Rfree-values of 28.3 and 31.5%. The model was completed and further modified using COOT and refined using REFMAC5. Figure 6 shows the final electron density with some nitrate ions clearly visible. For more detailed information see the AUTO-RICKSHAW output (directory exp3/struct_sol) . Figure 3 shows the cartoon representation of the two molecules in the asymmetric unit. Clear electron density can be found where the nitrate ions are bound (see Figure 4).
Figure 3: Cartoon representation of the two HEWL molecules in the asymmetric unit of the monoclinic crystals.
Figure 4: Experimental electron density map showing the bound nitrate ions. The (2F0bS-Fcaic, ocCaic)-map (blue) is contoured at 1.2 a, the (F0bs-Fcaic, acaic)-map (green and red) at +3.0 and - 3.0 a, respectively.
5 References
Berman, H. M, Westbrook, J., Feng, Z., Gilliland, G., Bhat, T. N., Weissig, H., Shindyalov, I. N. &Bourne, P. E. (2000). Nucl. Ac. Res. 28, 235-242.
Blake, C. C. F., Koenig, D. F., Mair, G. A., North, A. C. T., Philipps, D. C. & Sarma, V. R. (1965). Nature 206, 757-761.
Brinkmann, C, Weiss, M. S. & Weckert, E. (2006). Acta Cryst. D62, 349-355.
Bruenger, A. T., Adams, P. D., Clore, G. M., DeLano, W. L., Gros, P., Gross-Kunstleve, R. W., Jiang, J. S., Kuszewski, J., Nilges, M., Pannu, N. S., Read, R. J., Rice, L. M., Simonson, T. & Warren, G. L. (1998). Acta Cryst. D54, 905-921.
Collaborative Computational Project, Number 4 (1994). Acta Cryst. D50, 760-763.
Emsley, P. & Cowtan, K. (2004). Acta Cryst. D60, 2126-2132.
Evans, P. (2005). Acta Cryst. D62, 72-82.
Kabsch, W. (1993). /. Appl. Cryst, 26, 795-800.
Kabsch, W. (2010a). Acta Cryst. D66, 125-132.
Kabsch, W. (2010b). Acta Cryst. D66, 133-144.
Long, F., Vagin, A. A., Young, P. & Murshudov, G. N. (2008). Acta Cryst. D64, 125-132.
Morris, R. J., Perrakis, A. & Lamzin, V. S. (2002). Acta Cryst. D58, 968-975.
Mueller-Dieckmann, C, Panjikar, S., Schmidt, A., Mueller, S., Kuper, J., Geerlof, A., Wilmanns, M., Singh, R. K., Tucker, P. A. &Weiss, M. S. (2007). Acta Cryst. D63, 366-380.
Murshudov, G. N., Vagin, A. A. & Dodson, E. J. (1997). Acta Cryst. D53, 240-255.
Panjikar, S., Parthasarathy, V., Lamzin, V. S., Weiss, M. S. & Tucker, P. A. (2005). Acta Cryst. D61, 449-457.
Panjikar, S., Parthasarathy, V., Lamzin, V. S., Weiss, M. S. & Tucker, P. A. (2009). Acta Cryst. D65, 1089-1097.
Perrakis, A., Morris, R. J. & Lamzin, V. S. (1999). Nature Struct. Biol. 6, 458-463. Saraswathi, N. T., Sankaranarayanan, R. & Vijayan, M. (2002). Acta Cryst. D58, 1162-1167. Vagin, A., Teplyakov, A. (1997). /. Appl. Cryst, 30, 1022-1025.
Vaney, M. C, Maignan, S., Ries-Kautt, M. & Ducruix, A. (1996). Acta Cryst. D52, 505-517.
Voet, D., Voet, J. & Pratt, C. W. (2006). Fundamentals in Biochemistry - Life at the molecular level, 2nd Edition, John Wiley & Sons, Inc., Hoboken, NJ, USA.