Interactions of molecules with nucleic acids. I. An algorithm to generate nucleic acid structures with an application to the B-DNA structure and a counterclockwise helix.

An algorithm is developed that enables the routine determination of backbone conformations of nucleic acids. All atomic positions including hydrogen are specified in accord with experimental bond lengths and angles but with theoretically determined conformational angles. For two Watson-Crick base pairs at a separation of 3.38 Å, and perpendicular to a common helical axis, minimum energy configurations are found for all 10 combinations at helical angles of α ～ 36°–38°, corresponding to the B-DNA structure with C(2′)-endo sugar puckers. Backbone configurations exist only within the range 35.5° ? α ? 42°, which suggests the origin of the 10-fold helix. Calculated stacking energies for the B-DNA structure increases for each of the clustered groups of base pairs: G·C with G·C, G·C with A·T, and A·T with A·T, and they are in approximate agreement with experimental observations. The counter-clockwise helix is examined, and physically meaningful structures are found only when the helical axes of successive base pairs are disjointed.     

Helix bending as a factor in protein/DNA recognition

Normal vectors perpendicular to individual base pairs are a powerful tool for studying the bending behavior of B-DNA, both in the form of normal vector plots and in matrices that list angles between vectors for all possible base pair combinations. A new analysis program, FREEHELIX, has been written for this purpose, and applied to 86 examples of sequence-specific protein/DNA complexes whose coordinates are on deposit in the Nucleic Acid Data Base. Bends in this sample of 86 structures almost invariably follow from roll angles between adjacent base pairs; tilt makes no net contribution. Roll in a direction compressing the broad major groove is much more common than that which compresses the minor groove. Three distinct types of B-DNA bending are observed, each with a different molecular origin: (1) Localized kinking is produced by large roll at single steps or at two steps separated by one turn of helix. (2) Smooth, planar curvature is produced by positive and negative roll angles spaced a half-turn apart, with random side-to-side zigzag roll at intermediate points, rather than a tilt contribution that might have been expected theoretically. (3) Three-dimensional writhe results from significant roll angles at a continuous series of steps. Writhe need not change the overall direction of helix axis, if it is continued indefinitely or for an integral number of helical turns. A-DNA itself can be formally considered as possessing uniform, continuous writhe that yields no net helix bending. Smooth curvature is the most intricate deformation of the three, and is least common. Writhe is the simplest deformation and is most common; indeed, a low level of continuous writhe is the normal condition of an otherwise unbent B-DNA helix of general sequence. With one exception, every example of major kinking in this sample of 86 structures involves a pyrimidine–purine step: C–A/T–G, T–A, or C–G. Purine–purine steps, especially A–A, show the least tendency toward roll deformations. © 1998 John Wiley & Sons, Inc. Biopoly 44: 361–403, 1997     

Proton and phosphorus NMR studies of d-CpG(pCpG)n duplexes in solution. Helix-coil transition and complex formation with actinomycin-D.

The Watson–Crick imino and amino exchangeable protons, the nonexchangeable base and sugar protons, and the backbone phosphates for d-CpG(pCpG)_n, n = 1 and 2, have been monitored by high-resolution nmr spectroscopy in aqueous solution over the temperature range 0°–90°C. The temperature dependence of the chemical shifts of the tetramer and hexamer resonances is consistent with the formation of stable duplexes at low temperature in solution. Comparison of the spectral characteristics of the tetranucleotide with those of the hexanucleotide with temperature permits the differentiation and assignment of the cytosine proton resonances on base pairs located at the end of the helix from those in an interior position. There is fraying at the terminal base pairs in the tetranucleotide and hexanucleotide duplexes. The Watson–Crick ring imino protons exchange at a faster rate than the Watson–Crick side-chain amino protons, with exchange occurring by transient opening of the double helix. The structure of the d-CpG(pCpG)_n double helices has been probed by proton relaxation time measurements, sugar proton coupling constants, and the proton chemical shift changes associated with the helix–coil transition. The experimental data support a structural model in solution, which incorporates an anti conformation about the glycosyl bonds, C(3) exo sugar ring pucker, and base overlap geometries similar to the B-DNA helix. Rotational correlation times of 1.7 and 0.9 × 10^?9 sec have been computed for the hexanucleotide and tetranucleotide duplexes in 0.1 M salt, D₂O, pH 6.25 at 27°C. The well-resolved ³¹P resonances for the internucleotide phosphates of the tetramer and hexamer sequences at superconducting fields shift upfield by 0.2–0.5 ppm on helix formation. These shifts reflect a conformational change about the ω,ω′ phosphodiester bonds from gauche-gauche in the duplex structure to a distribution of gauche-trans states in the coil structure. Significant differences are observed in the transition width and midpoint of the chemical shift versus temperature profiles plotted in differentiated form for the various base and sugar proton and internucleotide phosphorous resonances monitoring the d-CpG(pCpG)_n helix–coil transition. The twofold symmetry of the d-CpGpCpG duplex is removed on complex formation with the antibiotic actinomycin-D. Two phosphorous resonances are shifted downfield by ～2.6 ppm and ～1.6 ppm on formation of the 1:2 Act-D:d-CpGpCpG complex in solution. Model studies on binding of the antibiotic to dinucleotides of varying sequence indicate that intercalation of the actinomycin-D occurs at the GpC site in the d-CpGpCpG duplex and that the magnitude of the downfield shifts reflects strain at the O-P-O backbone angles and hydrogen bonding between the phenoxazone and the phosphate oxygens. Actinomycin-D is known to bind to nucleic acids that exhibit a B-DNA conformation; this suggests that the d-CpG(pCpG)_n duplexes exhibit a B-DNA conformation in solution.     

Circular dichroism of flow-oriented nucleic acids. II. Comparison between observed and calculated spectra.

The ultraviolet circular dichroism (CD) of oriented DNA and RNA molecules is calculated by an extension of Johnson and Tinoco's theory [(1969) Biopolymers 7 , 727–749] for unoriented molecules. The calculations are carried out for molecular models of A-DNA, B-DNA, planar B-DNA, C-DNA, and RNA-11-α. The calculated curves are compared to measured spectra [(1975) J. Mol. Biol. 92 , 449–466] Chung and Holzwarth, for oriented solutions of DNA in buffer, DNA in 6 M LiCl or in ethylene glycol, and double-stranded viral RNA. The calculation, which considers only base–base interactions, predicts that the CD of B-DNA, measured with light propagating parallel to the helix axis, should be large and semiconservative, whereas the CD for light propagating perpendicular to the helix axis should be nonconservative. These predictions agree qualitatively with the experimental observations for DNA in buffer; agreement improves if one assumes the bases to be exactly perpendicular to the helix axis. For the other geometries, agreement is less satisfactory, but qualitative agreement with experiment is obtained and the signs of the specific CD spectra are in accord with observations.     

The complex of ethidium bromide with genomic DNA: structure analysis by polarized Raman spectroscopy

Structural properties of the complex formed between genomic DNA and the intercalating drug ethidium bromide (EtBr) have been determined by use of a Raman microscope equipped with near-infrared laser excitation. The polarized spectra, which were obtained from oriented fibers of the EtBr:DNA complex, are interpreted in terms of the relative orientations of the phenanthridinium ring of EtBr and bases of DNA. Quantification of structure parameters of EtBr and DNA in the complex were assessed using Raman tensors obtained from polarized Raman analyses of oriented specimens of EtBr (single crystal) and DNA (hydrated fiber). We find that the phenanthridinium plane is tilted by 35+/-5 degrees from the plane perpendicular to the fiber (DNA helix) axis. Assuming coplanarity of the phenanthridinium ring and its immediate base neighbors at the intercalation site, such bases would have a tilt angle closer to that of A-DNA (20 degrees) than to that of B-DNA (6 degrees). The average base tilt in stretches of DNA between intercalation sites remains that of B-DNA.     

Molecular structure of an A-DNA decamer d(ACCGGCCGGT)

The molecular structure of the DNA decamer d(ACCGGCCGGT) has been solved and refined by single-crystal X-ray-diffraction analysis at 0.20 nm to a final R-factor of 18.0%. The decamer crystallizes as an A-DNA double helical fragment with unit-cell dimensions of a = b = 3.923 nm and c = 7.80 nm in the space group P6(1)22. The overall conformation of this A-DNA decamer is very similar to that of the fiber model for A-DNA which has a large average base-pair tilt and hence a wide and shallow minor groove. This structure is in contrast to that of several A-DNA octamers in which the molecules all have low base-pair-tilt angles (8-12 degrees) resulting in an appearance intermediate between B-DNA and A-DNA. The average helical parameters of this decamer are typical of A-DNA with 10.9 base pairs/turn of helix, an average helical twist angle of 33.1 degrees, and a base-pair-tilt angle of 18.2 degrees. However, the CpG step in this molecule has a low local-twist angle of 24.5 degrees, similar to that seen in other A-DNA oligomers, and therefore appears to be an intrinsic stacking pattern for this step. The molecules pack in the crystal using a recurring binding motif, namely, the terminal base pair of one helix abuts the surface of the shallow minor groove of another helix. In addition, the GC base pairs have large propeller-twist angles, unlike those found most other A-DNA structures.     

The potentially Z-DNA-forming sequence d(GTGTACAC) crystallizes as A-DNA

(GT)n/(CA)n sequences have stimulated much interest because of their frequent occurrence in eukaryotic DNA and their potential for forming the left-handed Z-DNA structure. We here report the X-ray crystal structure of a self-complementary octadeoxynucleotide, d(GTGTACAC), at 2.5 A resolution. The molecule adopts a right-handed double-helical conformation belonging to the A-DNA family. In this alternating purine-pyrimidine DNA minihelix the roll and twist angles show alternations qualitatively consistent with Calladine's rules. The average tilt angle of 9.3 degrees is between the values found in A-DNA (19 degrees) and B-DNA (-6 degrees) fibers. It is envisaged that such intermediate conformations may render diversity to genomic DNA. The base-pair tilt angles and the base-pair displacements from the helix axis are found to be correlated for the known A-DNA double-helical fragments.     

Mediation of the A/B-DNA helix transition by G-tracts in the crystal structure of duplex CATGGGCCCATG

The crystal structure of the DNA dodecamer duplex CATGGGCCCATG lies on a structural continuum along the transition between A- and B-DNA. The dodecamer possesses the normal vector plot and inclination values typical of B-DNA, but has the crystal packing, helical twist, groove width, sugar pucker, slide and x-displacement values typical of A-DNA. The structure shows highly ordered water structures, such as a double spine of water molecules against each side of the major groove, stabilizing the GC base pairs in an A-like conformation. The different hydration of GC and AT base pairs provides a physical basis for solvent-dependent facilitation of the A↔B helix transition by GC base pairs. Crystal structures of CATGGGCCCATG and other A/B-DNA intermediates support a ‘slide first, roll later’ mechanism for the B→A helix transition. In the distribution of helical parameters in protein–DNA crystal structures, GpG base steps show A-like properties, reflecting their innate predisposition for the A conformation.     

Three-dimensional structure of d(GGGATCCC) in the crystalline state.

The structure of the self-complementary octamer d(GGGATCCC) has been analysed by single crystal X-ray diffraction methods at a nominal resolution of 2.5 A. With acceptable stereochemistry of the model the crystallographic R factor was 16.6% after restrained least-squares refinement. In the crystal, d(GGGATCCC) forms an A-DNA double helix with slightly varying conformation of the two strands. The average displacement of the base pairs from the helix axis is unusually large and is accompanied by pronounced sliding of the base pairs along their long axes at all dinucleotide steps except for the central AT. With 12 base pairs per complete turn the helix is considerably underwound. As observed with most oligodeoxyribonucleotides analysed by X-ray crystallography so far, the octamer displays reduced base pair tilt, increased rise per base pair and a more open major groove compared with canonical A-DNA. We propose that, based on these parameters, three A-helical sub-families may be defined; d(GGGATCCC) then is a representative of the class with intermediate tilt, rise, and major groove width.     

The extended and eccentric E-DNA structure induced by cytosine methylation or bromination

Cytosine methylation or bromination of the DNA sequence d(GGCGCC)2 is shown here to induce a novel extended and eccentric double helix, which we call E-DNA. Like B-DNA, E-DNA has a long helical rise and bases perpendicular to the helix axis. However, the 3'-endo sugar conformation gives the characteristic deep major groove and shallow minor groove of A-DNA. Also, if allowed to crystallize for a period of time longer than that yielding E-DNA, the methylated sequence forms standard A-DNA, suggesting that E-DNA is a kinetically trapped intermediate in the transition to A-DNA. Thus, the structures presented here chart a crystallographic pathway from B-DNA to A-DNA through the E-DNA intermediate in a single sequence. The E-DNA surface is highly accessible to solvent, with waters in the major groove sitting on exposed faces of the stacked nucleotides. We suggest that the geometry of the waters and the stacked base pairs would promote the spontaneous deamination of 5-methylcytosine in the transition mutation of dm5C-dG to dT-dA base pairs.     

Assessing the effects of time and spatial averaging in 15N chemical shift/15N-1H dipolar correlation solid state NMR experiments

The effect of time and spatial averaging on ¹⁵N chemical shift/¹H-¹⁵N dipolar correlation spectra, i.e., PISEMA spectra, of -helical membrane peptides and proteins is investigated. Three types of motion are considered: (a) Librational motion of the peptide planes in the -helix; (b) rotation of the helix about its long axis; and (c) wobble of the helix about a nominal tilt angle. A 2ns molecular dynamics simulation of helix D of bacteriorhodopsin is used to determine the effect of librational motion on the spectral parameters. For the time averaging, the rotation and wobble of this same helix are modelled by assuming either Gaussian motion about the respective angles or a uniform distribution of a given width. For the spatial averaging, regions of possible ¹⁵N chemical shift/¹H-¹⁵N dipolar splittings are computed for a distribution of rotations and/or tilt angles of the helix. The computed spectra show that under certain motional modes the ¹⁵N chemical shift/¹H-¹⁵N dipolar pairs for each of the residues do not form patterns which mimic helical wheel patterns. As a result, the unambiguous identification of helix tilt and helix rotation without any resonance assignments or on the basis of a single assignment may be difficult.     

Molecular mechanics calculations of dA12·dT12 and of the curved molecule d(GCTCGAAAAA)4·d(TTTTTCGAGC)4

Using the AMBER software package (Weiner and Kollman 1981) substantially modified for electrostatic contributions, the structural energies of the double-stranded oligonucleotides dA₁₂·dT₁₂ and d(GCTCGAAAAA)₄·d(TTTTTCGAGC)₄ were minimized. Using various starting structures for the molecule dA₁₂·dT₁₂, one final structure is obtained which possesses the experimentally determined properties of poly(dA)·poly(dT). This structure is an A-form-B-form-hybrid structure similar to that of Arnott et al. (1983). The dA-strand is similar to an A-form while the dT-strand is similar to normal B-form. This structure and separately optimized B-form sequence stretches were used to construct the double-stranded fragment d(GCTCGAAAAA)₄ which again was optimized. This sequence, when imbedded in a DNA fragment as contiguous repeats, shows a gel migration anomaly which has been interpreted as stable curvature of the DNA (Diekmann 1986). The calculated structure of this sequence indeed has a curved helix axis and is discussed as a model for curved DNA. A theoretical formalism is presented which allows one to calculate the structural parameters of any nucleic acid double helix in two different geometrical representations. This formalism is used to determine the parameters of the base-pair orientations of the curved structure in terms of wedge as well as cylindrical parameters. In the structural model presented here, the curvature of the helix axis results from an alternation of two different DNA structures in which the base-pairs possess different angles with the helix axis (cylinder tilt). Resulting from geometric restraints, a negative cylinder tilt angle correlates strongly with the closing of the minor groove (wedge roll). The blocks with different structure are not exactly coincident with the dA₅-blocks and the B-DNA stretches. Within the dA₅ block, base-pair tilt and wedge roll adopt large values which proceed into the 3 flanking B-DNA sequence by about one base-pair. These properties of the structure calculated here are discussed in terms of different models explaining DNA curvature.     

Spatial translational motions of base pairs in DNA molecules: Application of the extended matrix generator method

We have used the elementary generator matrices outlined in the preceding paper to examine the conformational plasticity of the nucleic acid double helix. Here we investigate kinked DNA structures made up of alternating B- and A-type helices and intrinsically curved duplexes perturbed by the intercalation of ligands. We model the B-to-A transition by the lateral translation of adjacent base pairs, and the intercalation of ligands by the vertical displacement of neighboring residues. We report a complete set of average configuration-dependent parameters, ranging from scalars (i.e., persistence lengths) to first- and second-order tensor parameters (i.e., average second moments of inertia), as well as approximations of the associated spatial distributions of the DNA and their angular correlations. The average structures of short chains (of lengths less than 100 base pairs) with local kinks or intrinsically curved sequences are essentially rigid rods. At the smallest chain lengths (10 base pairs), the kinked and curved chains exhibit similar average properties, although they are structurally perturbed compared to the standard B-DNA duplex. In contrast, at lengths of 200 base pairs, the curved and kinked chains are more compact on average and are located in a different space from the standard B- or A-DNA helix. While A-DNA is shorter and thicker than B-DNA in x-ray models, the long flexible A-DNA helix is thinner and more extended on average than its B-DNA counterpart because of more limited fluctuations in local structure. Curved polymers of 50 base pairs or longer also show significantly greater asymmetry than other DNAs (in terms of the distribution of base pairs with respect to the center of gravity of the chain). The intercalation of drugs in the curved DNA straightens and extends the smoothly deformed template. The dimensions of the average ellipsoidal boundaries defining the configurations of the intercalated polymers are roughly double those of the intrinsically curved chain. The altered proportions and orientations of these density functions reflect the changing shape and flexibility of the double helix. The calculations shed new light on the possible structural role of short A-DNA fragments in long B-type duplexes and also offer a model for understanding how GC-specific intercalative ligands can straighten naturally curved DNA. The mechanism is not immediately obvious from current models of DNA curvature, which attribute the bending of the chain to a perturbed structure in repeating tracts of A · T base pairs. © 1994 John Wiley & Sons, Inc.     

Nucleic acid-metal interactions: V. The effect of silver(I) on the structures of A- and B-DNA forms

The interaction of silver(I) with DNA has been studied with uv LD in aqueous solution and in a humid anisotropic poly(vinyl alcohol) host (B-DNA) and also in 80% ethanolic solution (A-DNA). Addition of silver ions has a pronounced effect on the dichroic spectra of DNA, indicating that the DNA structure is significantly altered. By correlation with calculated reduced LD spectra, using intensities and moments of the corresponding electronic transitions of the DNA bases, the experimental spectra of DNA at high silver content may be interpreted in terms of tilt and roll angles of the bases in the double helix. In ethanolic A-DNA solution there is a pronounced decrease in the orientation by flow of DNA, suggesting that the complexation of DNA to silver may be accompanied by the formation of compact tertiary structures.     

Base sequence, local helix structure, and macroscopic curvature of A-DNA and B-DNA

Given a specified DNA sequence and starting with an idealized conformation for the double helix (A-DNA or B-DNA), the dependence of conformational energy on variations in the local geometry of the double helix can be examined by computer modeling. By averaging over all thermally accessible states, it is possible to determine 1) how the optimum local structure differs from the initial idealized conformation and 2) the energetic costs of small structural deformations. This paper describes such a study. Tables are presented for the prediction of helix twist angles and base pair roll angles for both A-DNA and B-DNA when the sequence has been specified. Local deviations of helix parameters from their average values can accumulate to produce a net curvature of the molecule, a curvature that can be sharp enough to be experimentally detectable. As an independent check on the method, the calculations provide predictions for the longitudinal compressibility (Young's modulus) and the average torsional stiffness, both of which are in good agreement with experimental values. In examining the role of sequence-dependent variations in helix structure for the recognition of specific sequences by proteins, we have calculated the energy needed to deform the self-complementary hexanucleotide d(CAATTG) to match the local geometry of d(GAATTC), which is the sequence recognized by the EcoRI restriction endonuclease. That energy would be sufficient to reduce the binding of the incorrect sequence to the protein by over 2 orders of magnitude relative to the correct sequence.     

A note on sequence-dependence of DNA structure

A circular dichroism study was conducted on the solution structure of several different oligonucleotides, whose X-ray structures have been solved. It is suggested that in aqueous solution the oligonucleotides can form structures that maintain geometrical elements which are typical of B-DNA, A-DNA, and their intermediate forms. It is shown that 5'GGATGGGAG:5'CTCCCATCC, which forms an A-DNA helix in the crystal state (McCall et al. 1986), in aqueous solution maintains an A-DNA like structure at temperatures below 10 degrees C. At temperatures between 10 degrees C and 25 degrees C it shows a tendency to form an intermediate structure between A-DNA and B-DNA. Also, it is shown that TFE does not cause a transition from B-DNA to A-DNA helix in short DNA fragments, but instead disrupts the helix.     

Chiral phosphorothioate analogues of B-DNA. The crystal structure of Rp-d[Gp(S)CpGp(S)CpGp(S)C]

The compound Rp-d[Gp(S)CpGp(S)CpGp(S)C], an analogue of the deoxyoligomer d(G-C)3, crystallizes in space group P2(1)2(1)2(1) with a = 34.90 A, b = 39.15 A and c = 20.64 A. The structure, which is not isomorphous with any previously determined deoxyoligonucleotide, was refined to an R factor of 14.5% at a resolution of 2.17 A, with 72 solvent molecules located. The two strands of the asymmetric unit form a right-handed double helix, which is a new example of a B-DNA conformation and brings to light an important and overlooked component of flexibility of the double helix. This flexibility is manifest in the alternation of the backbone conformation between two states, defined by the adjacent torsion angles epsilon and zeta, trans . gauche-(BI) and gauche-. trans (BII). BI is characteristic of classical of B-DNA and has an average C(1') to C(1') separation of 4.5 A. The corresponding separation for BII is 5.3 A. Each state is associated with a distinct phosphate orientation where the plane of the PO2 (or POS) group is alternately near horizontal or vertical with respect to the helix axis. The BI and BII conformations are out of phase on the two strands. As a consequence, on one strand purine-pyrimidine stacking is better than pyrimidine-purine, while the converse holds for the other strand. At each base-pair step, good and bad stacking alternate across the helix axis. The pattern of alternation is regular in the context of a fundamental dinucleotide repeat. Re-examination of the B-DNA dodecamer d(C-G-C-G-A-A-T-T-C-G-C-G) shows that the C-G-C-G regions contain the BI and BII conformations, and the associated dual phosphate orientation and asymmetric base stacking. Different mechanisms are used in the two structures to avoid clashes between guanine residues on opposite strands, a combination of lateral slide, tilt and helical twist in the present structure, and base roll, tilt and longitudinal slide (Calladine rules) in the dodecamer. The flexibility of the phosphate orientations demonstrated in this structure is important, since it offers a structural basis for protein-nucleic acid recognition.     

The Crystal Structure of d(CCCCGGGG): A New A-Form Variant with an Extended Backbone Conformation

Abstract

The crystal structure of d(CCCCGGGG) has been determined at a resolution of 2.25Å. The oligomers crystallize as A-DNA duplexes occupying crystallographic two-fold axes. The backbone conformation is, in general, similar to that observed in previously reported crystal structures of A-DNA fragments, except for the central linkage, where it adopts an extended structure resulting from all trans conformation at the P-05′-C5′-C4′ bonds. This type of conformation facilitates interstrand stacking between the guanines at the C-G site. The local helix twist at this step is very small (25°) compared to an overall average of 33.5°. The unique structure of the C-G base-pair step, namely the extended backbone and the distinct stacking geometry, may be an important feature in the recognition mechanism between double- stranded DNA molecules and restriction endonucleases such as Msp I, which cuts the sequence CCGG very specifically with a rate unaffected by neighboring base pairs.     

Conformational characteristics of the trinucleoside diphosphate d(ApApA) from energy-minimization studies

Energy-minimization studies were carried out on the trinucleoside diphosphate d(ApApA). The potential energy contributions from nonbonded, electrostatic, hydrogen-bonding, and torsional interactions were minimized by treating the 13 relevant dihedral angles as simultaneous variables. For the C(3′)-endo trimer, 14 low-energy conformations are within 10 kcal/mol above the lowest energy found, compared to only 3 in the case of the C(2′)-endo trimer. This result shows the flexible character of the C(3′)-endo unit. The hairpin-type, loop-promoting conformer with (ω′,ω) = (101°, 59°) was found to be the most favored structure at the 3′-terminus of d(ApApA). The predicted U- and L-type bend conformers were found to lie within 5 kcal/mol, compared to the lowest energy B-DNA structure. The A-DNA and Watson-Crick DNA types of helical conformers also lie within very small energy barriers. The phosphate group at the 5′-end of the nucleotide residue has a definite influence on the base of the corresponding nucleotide, keeping it in the normal anti-region, and hence on the base-stacking property. The results are compared with the presently available experimental data, mainly with the tRNA^Phe crystal.