Download citation
Download citation
link to html
A new `rule' for the association of hydrogen-bond donors and acceptors in crystal structures is presented. It implies that ranks are assigned to each donor and each acceptor (1 is best, 2 is next best etc.), and that hydrogen bonds should be formed between donors and acceptors in rank order. L-Ser-L-Ala, C6H12N2O4, is used together with its retroanalogue, L-Ala-L-Ser, and three other pairs of dipeptide retroanalogues to illustrate this rule and the reasons why it may not always be followed.

Supporting information

cif

Crystallographic Information File (CIF) https://doi.org/10.1107/S0108270100001463/de1127sup1.cif
Contains datablocks I, global

hkl

Structure factor file (CIF format) https://doi.org/10.1107/S0108270100001463/de1127Isup2.hkl
Contains datablock I

Comment top

One of the empirical `Hydrogen-Bond Rules' (Etter, 1990) states that `the best proton donors and acceptors remaining after intramolecular hydrogen bond formation form intermolecular hydrogen bonds to one another'. This rule may tentatively be expanded to include all acceptors and donors in a new hydrogen-bond Network Rule (NR) as follows: `after intramolecular hydrogen-bond formation, proton donors and acceptors associate in rank order'. In principle, any structure can be analysed with reference to this rule. Two previous papers (Görbitz & Backe, 1996; Görbitz, 1999a) have demonstrated that studies of dipeptide retroanalogues are quite useful. The present work concludes this investigation. \scheme

The structure of L-Ser-L-Ala is shown in Fig. 1. The structure of its retroanalogue, L-Ala-L-Ser, was presented by Jones et al. (1978a). For both compounds, the three best hydrogen-bond donors, with ranks 1, 2 and 3, are the three amino N—H atoms, while the carboxylate group has acceptor ranks from 1 to 4. As far as the two additional donors (>N—H and –CH2—OH) are concerned, it is not obvious which should be assigned rank 4 and which rank 5. This is also true for assigning ranks to the two acceptors >CO and –CH2—OH, although judging by statistical values for the donor···O distances with carbonyl and water acceptors (Görbitz, 1989), one can tentatively assign rank 5 to the hydroxyl group and rank 6 to >CO. With this set of ranks, we find that hydrogen bonding in the L-Ala-L-Ser structure (Fig. 2 b; Jones et al., 1978a) strictly follows the extended hydrogen-bond rule, as does L-Ser-Gly (Jones et al., 1978b), in which precisely the same types of interactions occur. L-Ser-L-Ala (Fig. 2a) represents only a modest deviation from this pattern, in that one of the amino N—H protons is accepted by the L-Ser side-chain hydroxyl group rather than by the main chain carboxylate group. As discussed previously, however, hydrogen bonding in Gly-L-Ser (Görbitz, 1999a) is quite different. It is interesting to see how the introduction of a small hydrophobic group [H (Gly) methyl (Ala)] gives more similar hydrogen-bond interactions within the retroanalogue pair. It can be seen from Fig. 2 how these methyl groups generate small hydrophobic columns along the shortest axis in each structure. This is a common motif for the aggregation of hydrophobic groups in the crystal structures of peptides (Görbitz & Etter, 1992).

Dipeptide structures have previously (Görbitz, 1999a) been retrieved from the Cambridge Structural Database (Allen & Kennard, 1993) and divided into three categories (counting Gly as a hydrophilic residue): A) hydrophilic structures with abundant hydrogen bonding. Dx is typically 1.40–1.60 Mg m-3; B) dipeptides with one hydrophobic residue. Dx in the range 1.25–1.40 Mg m-3 (1.44 for L-Ser-L-Ala represents an extreme value); C) strictly hydrophobic dipeptides. Dx in the range 1.05–1.20 Mg m-3.

The L-Ala-L-Ser/L-Ser-L-Ala pair belongs to group B, as does the L-Val-L-Glu (Eggleston, 1984)/L-Glu-L-Val (Görbitz & Backe, 1996) pair. The Gly-L-Ser/L-Ser-Gly pair discussed above, on the other hand, belongs to group A. Furthermore, we have data for a fourth pair, L-Val-L-Ala (Görbitz & Gundersen, 1996) and L-Ala-L-Val (Görbitz, 2000), belonging to group C. It should be added that dipeptides (as well as other peptides) often include cocrystallized water or organic solvent molecules. Therefore, the structures of retroanalogues may not contain the same hydrogen bond donors and acceptors [e.g. Gly-L-Asp·2H2O (Eggleston & Hodgson, 1982)/L-Asp-Gly·H2O (Eggleston et al., 1984) and L-Ala-L-Leu·0.5H2O (Görbitz, 1999b)/L-Leu-L-Ala·4H2O (Görbitz, 1997)], and a direct comparison of hydrogen-bonding patterns is rendered less straightforward.

Some relevant data for the four known pairs of solvent-free retroanalogue pairs are given in Table 3. The most obvious observation is that an increasing number of hydrophobic groups in a molecule (meaning an unchanged or reduced number of hydrophilic groups) makes hydrogen bonding within a pair more similar, but it does not imply that the NR is followed more rigorously. This apparent contradiction may be explained by considering the problems associated with arranging three main chain carboxylate groups (as required by the NR) around each amino group when usually rather bulky hydrophobic side chains are present. In fact, only about one out of eight dipeptide structures display three amino N—H+···-O—C carboxylate hydrogen bonds (Görbitz, 1999a), usually when Gly (or less frequently Ala) is either an N-terminal or a C-terminal residue. We believe that deviations from the NR can, at least in part, be explained by such inherent steric constraints, combined with the need always to segregate hydrophobic groups into distinct regions of the crystal.

The experimental material discussed in this paper is limited, but the results should give at least an indication of the general trends for short linear peptides. It is clear that the NR is not very robust, as it is followed (completely or almost) by only three out of the eight structures discussed here. Nevertheless, the rule could be a useful tool in the analysis of two- and three-dimensional hydrogen-bond networks in crystal structures of a variety of organic compounds. Further results from such investigations would be most interesting.

Experimental top

The title compound was obtained from Sigma. The specimen used for data collection was the only obvious single-crystal resulting from a series of slow evaporation experiments with aqueous solutions of the dipeptide at room temperature.

Refinement top

The hydroxylic H-atom was refined isotropically. Other peptide H atoms were placed geometrically and refined with constraints to keep all C—H/N—H distances and all C—C—H/C—N—H angles on one C or N atom the same. Uiso values were 1.2Ueq of the carrier atom, or 1.5Ueq for hydroxyl, methyl and amino groups. Free rotation was permitted for amino and methyl groups.

Computing details top

Data collection: SMART (Siemens, 1995); cell refinement: SAINT (Bruker, 1997); data reduction: SAINT; program(s) used to solve structure: SHELXTL (Sheldrick, 1997); program(s) used to refine structure: SHELXTL; molecular graphics: SHELXTL; software used to prepare material for publication: SHELXTL.

Figures top
[Figure 1] Fig. 1. The asymmetric unit of L-Ser-L-Ala with the atomic numbering. Displacement ellipsoids are shown at the 50% probability level and H atoms are shown as spheres of arbitrary size.
[Figure 2] Fig. 2. The unit cell and hydrogen bond pattern for a) L-Ser-L-Ala and b) L-Ala-L-Ser (Jones et al., 1978a). Views are along the 4.849 Å a axis and the 4.859 Å c axis, respectively.
L-Seryl-L-Alanine top
Crystal data top
C6H12N2O4F(000) = 188
Mr = 176.18Dx = 1.443 Mg m3
Monoclinic, P21Mo Kα radiation, λ = 0.71073 Å
a = 4.8488 (1) ÅCell parameters from 4637 reflections
b = 14.8294 (4) Åθ = 3–35°
c = 6.0228 (2) ŵ = 0.12 mm1
β = 110.534 (1)°T = 153 K
V = 405.55 (2) Å3Block, colourless
Z = 20.42 × 0.35 × 0.17 mm
Data collection top
Siemens SMART CCD
diffractometer
1837 independent reflections
Radiation source: fine-focus sealed tube1786 reflections with I > 2σ(I)
Graphite monochromatorRint = 0.033
Detector resolution: 8.3 pixels mm-1θmax = 35.0°, θmin = 2.8°
Sets of exposures each taken over 0.6° ω rotation scansh = 77
Absorption correction: empirical (using intensity measurements)
SADABS (Sheldrick, 1996)
k = 2323
Tmin = 0.950, Tmax = 0.980l = 99
6572 measured reflections
Refinement top
Refinement on F2Primary atom site location: structure-invariant direct methods
Least-squares matrix: fullSecondary atom site location: difference Fourier map
R[F2 > 2σ(F2)] = 0.057Hydrogen site location: inferred from neighbouring sites
wR(F2) = 0.127H atoms treated by a mixture of independent and constrained refinement
S = 1.34 w = 1/[σ2(Fo2) + (0.0432P)2 + 0.176P]
where P = (Fo2 + 2Fc2)/3
1837 reflections(Δ/σ)max = 0.002
120 parametersΔρmax = 0.42 e Å3
1 restraintΔρmin = 0.26 e Å3
Crystal data top
C6H12N2O4V = 405.55 (2) Å3
Mr = 176.18Z = 2
Monoclinic, P21Mo Kα radiation
a = 4.8488 (1) ŵ = 0.12 mm1
b = 14.8294 (4) ÅT = 153 K
c = 6.0228 (2) Å0.42 × 0.35 × 0.17 mm
β = 110.534 (1)°
Data collection top
Siemens SMART CCD
diffractometer
1837 independent reflections
Absorption correction: empirical (using intensity measurements)
SADABS (Sheldrick, 1996)
1786 reflections with I > 2σ(I)
Tmin = 0.950, Tmax = 0.980Rint = 0.033
6572 measured reflections
Refinement top
R[F2 > 2σ(F2)] = 0.0571 restraint
wR(F2) = 0.127H atoms treated by a mixture of independent and constrained refinement
S = 1.34Δρmax = 0.42 e Å3
1837 reflectionsΔρmin = 0.26 e Å3
120 parameters
Special details top

Experimental. The data collection nominally covered a hemisphere of reciprocal space by a combination of five sets of exposures at a 5.0 cm crystal-to-detector distance.

Refinement. Refinement of F2 against ALL reflections.

Fractional atomic coordinates and isotropic or equivalent isotropic displacement parameters (Å2) top
xyzUiso*/Ueq
O10.3135 (4)0.47457 (13)0.8180 (3)0.0191 (3)
O20.6018 (4)0.67387 (12)0.6509 (3)0.0192 (3)
O30.5258 (5)0.75305 (12)0.9401 (3)0.0239 (4)
O40.8348 (4)0.36651 (13)0.3557 (3)0.0176 (3)
H50.736 (9)0.331 (3)0.279 (7)0.026*
N10.4237 (4)0.32040 (13)0.6268 (3)0.0132 (3)
H10.249 (4)0.3435 (6)0.550 (4)0.020*
H20.425 (3)0.2967 (12)0.761 (3)0.020*
H30.463 (3)0.2784 (12)0.539 (4)0.020*
N20.7847 (4)0.52517 (12)0.9235 (3)0.0126 (3)
H40.942 (7)0.5142 (5)0.9261 (3)0.015*
C10.6498 (4)0.39249 (13)0.6773 (4)0.0113 (3)
H110.828 (7)0.3690 (9)0.770 (3)0.014*
C20.6754 (5)0.42783 (14)0.4456 (4)0.0145 (4)
H210.469 (5)0.4370 (3)0.322 (3)0.017*
H220.780 (3)0.4891 (15)0.4759 (8)0.017*
C30.5663 (4)0.46816 (14)0.8143 (4)0.0118 (3)
C40.7375 (4)0.60723 (13)1.0380 (3)0.0114 (3)
H410.600 (5)0.5944 (5)1.113 (3)0.014*
C51.0251 (5)0.63652 (16)1.2284 (4)0.0183 (4)
H511.100 (3)0.5892 (12)1.334 (4)0.027*
H520.9901 (14)0.6862 (16)1.310 (3)0.027*
H531.161 (4)0.6521 (16)1.1574 (17)0.027*
C60.6114 (5)0.68322 (14)0.8585 (4)0.0132 (3)
Atomic displacement parameters (Å2) top
U11U22U33U12U13U23
O10.0129 (6)0.0200 (8)0.0263 (8)0.0018 (6)0.0093 (6)0.0080 (6)
O20.0300 (9)0.0156 (7)0.0129 (6)0.0026 (7)0.0086 (6)0.0010 (6)
O30.0413 (11)0.0157 (7)0.0147 (7)0.0119 (7)0.0098 (7)0.0017 (6)
O40.0170 (7)0.0212 (8)0.0170 (7)0.0030 (6)0.0090 (6)0.0077 (6)
N10.0151 (7)0.0117 (7)0.0137 (7)0.0026 (6)0.0064 (6)0.0012 (6)
N20.0103 (7)0.0105 (7)0.0173 (7)0.0002 (5)0.0054 (6)0.0020 (6)
C10.0127 (8)0.0098 (7)0.0119 (8)0.0013 (6)0.0051 (6)0.0010 (6)
C20.0186 (9)0.0129 (8)0.0138 (8)0.0009 (7)0.0080 (7)0.0010 (7)
C30.0125 (7)0.0099 (7)0.0126 (7)0.0003 (6)0.0038 (6)0.0000 (6)
C40.0134 (8)0.0101 (7)0.0112 (7)0.0009 (6)0.0048 (6)0.0012 (6)
C50.0169 (9)0.0165 (9)0.0176 (9)0.0018 (8)0.0012 (8)0.0046 (8)
C60.0160 (8)0.0112 (8)0.0124 (7)0.0015 (7)0.0051 (7)0.0017 (6)
Geometric parameters (Å, º) top
O1—C31.238 (3)C1—C31.529 (3)
O2—C61.242 (3)C1—C21.536 (3)
O3—C61.276 (3)C1—H110.9174
O4—C21.418 (3)C2—H211.0255
O4—H50.75 (4)C2—H221.0255
N1—C11.484 (3)C4—C51.525 (3)
N1—H10.8790C4—C61.533 (3)
N1—H20.8790C4—H410.9445
N1—H30.8790C5—H510.9322
N2—C31.334 (3)C5—H520.9322
N2—C41.456 (3)C5—H530.9322
N2—H40.7740
C2—O4—H5111 (3)C1—C2—H22109.5
C1—N1—H1109.5H21—C2—H22108.1
C1—N1—H2109.5O1—C3—N2124.20 (19)
H1—N1—H2109.5O1—C3—C1121.18 (19)
C1—N1—H3109.5N2—C3—C1114.62 (18)
H1—N1—H3109.5N2—C4—C5109.89 (17)
H2—N1—H3109.5N2—C4—C6111.75 (16)
C3—N2—C4122.27 (17)C5—C4—C6110.74 (17)
C3—N2—H4118.9N2—C4—H41108.1
C4—N2—H4118.9C5—C4—H41108.1
N1—C1—C3108.85 (16)C6—C4—H41108.1
N1—C1—C2110.39 (17)C4—C5—H51109.5
C3—C1—C2111.15 (17)C4—C5—H52109.5
N1—C1—H11108.8H51—C5—H52109.5
C3—C1—H11108.8C4—C5—H53109.5
C2—C1—H11108.8H51—C5—H53109.5
O4—C2—C1110.89 (17)H52—C5—H53109.5
O4—C2—H21109.5O2—C6—O3125.1 (2)
C1—C2—H21109.5O2—C6—C4119.51 (19)
O4—C2—H22109.5O3—C6—C4115.35 (18)
N1—C1—C3—N2163.35 (17)N1—C1—C3—O117.3 (3)
C1—C3—N2—C4172.41 (17)C2—C1—C3—O1104.5 (2)
C3—N2—C4—C680.4 (2)C2—C1—C3—N274.9 (2)
N2—C4—C6—O210.1 (3)C3—N2—C4—C5156.3 (2)
N1—C1—C2—O476.6 (2)C5—C4—C6—O2112.8 (2)
C3—C1—C2—O4162.47 (17)N2—C4—C6—O3170.7 (2)
C4—N2—C3—O16.9 (3)C5—C4—C6—O366.5 (2)
Hydrogen-bond geometry (Å, º) top
D—H···AD—HH···AD···AD—H···A
N1—H1···O4i0.881.972.831 (3)167
N1—H2···O3ii0.881.852.723 (3)173
N1—H3···O2iii0.881.892.718 (3)157
N2—H4···O1iv0.772.202.947 (2)162
O4—H5···O3iii0.75 (4)1.88 (4)2.623 (3)173 (4)
Symmetry codes: (i) x1, y, z; (ii) x+1, y1/2, z+2; (iii) x+1, y1/2, z+1; (iv) x+1, y, z.

Experimental details

Crystal data
Chemical formulaC6H12N2O4
Mr176.18
Crystal system, space groupMonoclinic, P21
Temperature (K)153
a, b, c (Å)4.8488 (1), 14.8294 (4), 6.0228 (2)
β (°) 110.534 (1)
V3)405.55 (2)
Z2
Radiation typeMo Kα
µ (mm1)0.12
Crystal size (mm)0.42 × 0.35 × 0.17
Data collection
DiffractometerSiemens SMART CCD
diffractometer
Absorption correctionEmpirical (using intensity measurements)
SADABS (Sheldrick, 1996)
Tmin, Tmax0.950, 0.980
No. of measured, independent and
observed [I > 2σ(I)] reflections
6572, 1837, 1786
Rint0.033
(sin θ/λ)max1)0.806
Refinement
R[F2 > 2σ(F2)], wR(F2), S 0.057, 0.127, 1.34
No. of reflections1837
No. of parameters120
No. of restraints1
H-atom treatmentH atoms treated by a mixture of independent and constrained refinement
Δρmax, Δρmin (e Å3)0.42, 0.26

Computer programs: SMART (Siemens, 1995), SAINT (Bruker, 1997), SAINT, SHELXTL (Sheldrick, 1997), SHELXTL.

Selected geometric parameters (Å, º) top
O1—C31.238 (3)O3—C61.276 (3)
O2—C61.242 (3)N1—C11.484 (3)
N1—C1—C3—N2163.35 (17)N2—C4—C6—O210.1 (3)
C1—C3—N2—C4172.41 (17)N1—C1—C2—O476.6 (2)
C3—N2—C4—C680.4 (2)
Hydrogen-bond geometry (Å, º) top
D—H···AD—HH···AD···AD—H···A
N1—H1···O4i0.881.972.831 (3)167
N1—H2···O3ii0.881.852.723 (3)173
N1—H3···O2iii0.881.892.718 (3)157
N2—H4···O1iv0.772.202.947 (2)162
O4—H5···O3iii0.75 (4)1.88 (4)2.623 (3)173 (4)
Symmetry codes: (i) x1, y, z; (ii) x+1, y1/2, z+2; (iii) x+1, y1/2, z+1; (iv) x+1, y, z.
Comparison of hydrogen-bond types in the structures of dipeptide retroanalogues top
pairDx (Mg m-3)H-bond typesfollows NRReference
Gly-L-Ser/L-Ser-Gly1.55/1.60rather differentno/yes(a)/(b)
L-Ala-L-Ser/L-Ser-L-Ala1.42/1.44very similaryes/almost(c)/(d)
L-Val-L-Glu/L-Glu-L-Val1.31/1.38similarpartly/partly(e)/(f)
L-Ala-L-Val/L-Val-L-Ala1.07/1.04similar*partly/partly(g)/(h)
Notes: (a) Görbitz (1999a), (b) Jones et al. (1978a), (c) Jones et al. (1978b), (d) present work, (e) Eggleston (1984), (f) Görbitz & Backe (1996), (g) Görbitz, (2000), (h) Görbitz & Gundersen (1996); * isomorphous.
 

Follow Acta Cryst. C
Sign up for e-alerts
Follow Acta Cryst. on Twitter
Follow us on facebook
Sign up for RSS feeds