13C-NMR of ribosyl A-A-A, A-A-G, and A-U-G. Synthesis and assignment

The three RNA trinucleotides; ApApA, ApApG, and ApUpG, have been synthesized in sufficient quantity to obtain natural abundance 13C(1H)-NMR spectra at strand concentrations between 4 and 100 mM. Comparisons between 70 degrees C spectra of the three trimers and their consistuent dimers ApA, ApG, ApU, and UpG allow secure assignments to be made for most of the resonances. This paper describes the syntheses and 13C assignments of the oligomers.     

Multinuclear NMR studies of DNA hairpins. 1. Structure and dynamics of d(CGCGTTGTTCGCG)

The solution structure of the hairpin formed by d(CGCGTTGTTCGCG) has been examined in detail by a wide variety of NMR techniques. The hairpin was characterized by proton NMR to obtain interproton distances and torsion angle information. An energy-minimized model was constructed that is consistent with these data. The hairpin consists of a B-DNA stem of four C-G base pairs and a loop region consisting of five unpaired bases. Three bases in the 5' of the loop are stacked over the 3' end of the stem, and the other two bases in the 3' of the loop are stacked over the 5' end of the stem. The phosphorus NMR spectrum revealed a phosphate in the stem region with an unusual conformation, and two phosphates, P9 and P10, were found to undergo intermediate exchange between conformations. The hairpin was also synthesized with a carbon-13 label in each of the thymidine C6 carbons, and relaxation measurements were performed to determine the extent of internal motions in the loop region. The loop bases are more flexible than the stem bases and exhibit subnanosecond motions with an amplitude corresponding to diffusion in a cone of approximately 30 degrees.     

Crystal and molecular structure of methyl 4-O-methyl-beta-D-glucopyranosyl-(1-->4)-beta-D-glucopyranoside

The cellulose model compound methyl 4-O-methyl-beta-D-glucopyranosyl-(1-->4)-beta-D-glucopyranoside (6) was synthesised in high overall yield from methyl beta-D-cellobioside. The compound was crystallised from methanol to give colourless prisms, and the crystal structure was determined. The monoclinic space group is P2(1) with Z=2 and unit cell parameters a=6.6060 (13), b=14.074 (3), c=9.3180 (19) A, beta=108.95(3) degrees. The structure was solved by direct methods and refined to R=0.0286 for 2528 reflections. Both glucopyranoses occur in the 4C(1) chair conformation with endocyclic bond angles in the range of standard values. The relative orientation of both units described by the interglycosidic torsional angles [phi (O-5' [bond] C-1' [bond] O-4 [bond] C-4) -89.1 degrees, Phi (C-1' [bond] O-4 [bond] C-4 [bond] C-5) -152.0 degrees] is responsible for the very flat shape of the molecule and is similar to those found in other cellodextrins. Different rotamers at the exocyclic hydroxymethyl group for both units are present. The hydroxymethyl group of the terminal glucose moiety displays a gauche-trans orientation, whereas the side chain of the reducing unit occurs in a gauche-gauche conformation. The solid state (13)C NMR spectrum of compound 6 exhibits all 14 carbon resonances. By using different cross polarisation times, the resonances of the two methyl groups and C-6 carbons can easily be distinguished. Distinct differences of the C-1 and C-4 chemical shifts in the solid and liquid states are found.     

A conformational study of adenylyl-(3'',5'')-adenosine and adenylyl-(2'',5'')-adenosine in aqueous solution by carbon-13 magnetic resonance spectroscopy.

The solution conformation of adenylyl-(3',5')-adenosine and adenylyl-(2',5')-adenosine in both the stacked and unstacked states was studied by carbon-13 magnetic resonance spectroscopy. Large chemical shift differences between the base carbons in the dimers and those in the corresponding monomers are attributed in part to the influence of base-base interaction. Carbon-phosphorus couplings across three bonds revealed the preferred populations for certain backbone rotamers, demonstrating that significant changes in conformation about the "c(3')-O and C(5')-O bonds do not occur in the temperature or salt-induced unstacking of adenylyl-(3',5')-adenosine. However, rotations about the C(2')-O and C(5')-O bonds occur in the temperature-mediated unstacking of adenylyl-(2',5')-adenosine.     

1H NMR assignments and conformational analysis of the oligoribonucleotides CA, CAU, CAUG, ACAUG, and UCAUG: observation of pyrimidine H5-H1' long-range scalar couplings

Conformational analysis and 1H NMR spectral assignments have been carried out using COSY and RELAY methods for a series of related oligoribonucleotides including two pentamers with 5'-dangling bases. Intraresidue long-range five bond scalar coupling was observed between pyrimidine H5 and H1' protons in the COSY-45 spectra and this feature was useful for both assignment purposes and conformational analysis. The ribose ring conformations were predominantly C3'-endo with the C2'-endo population increasing at the 3'-terminus. The 5'-dangling bases were not stacked efficiently, exhibiting lower % C3'-endo values than their 3'-nearest neighbors. Backbone torsion angle population. beta t, gamma +, epsilon t, were determined using 1H-1H, 1H-31P, and 13C-31P coupling constants. From beta t and gamma + populations the U3-G4 step in CAUG was found to be less efficiently stacked than the C1-A2 and A2-U3 steps. This observation in solution is consistent with the fiber diffraction A-RNA model (S. Arnott, D.W.L. Hukins, S.D. Dover, W. Fuller and A.R. Hodgson, J. Mol. Biol. 81, 107-122, 1973) which also predicts poor stacking in a U-G dinucleotide. The epsilon t populations were greater than 65% for all C3'-O3' bonds and consistent with a right-handed A-RNA helix.     

Selective 13C labeling of nucleotides for large RNA NMR spectroscopy using an E. coli strain disabled in the TCA cycle

Escherichia coli (E. coli) is an ideal organism to tailor-make labeled nucleotides for biophysical studies of RNA. Recently, we showed that adding labeled formate enhanced the isotopic enrichment at protonated carbon sites in nucleotides. In this paper, we show that growth of a mutant E. coli strain DL323 (lacking succinate and malate dehydrogenases) on (13)C-2-glycerol and (13)C-1,3-glycerol enables selective labeling at many useful sites for RNA NMR spectroscopy. For DL323 E. coli grown in (13)C-2-glycerol without labeled formate, all the ribose carbon atoms are labeled except the C3' and C5' carbon positions. Consequently the C1', C2' and C4' positions remain singlet. In addition, only the pyrimidine base C6 atoms are substantially labeled to ~96% whereas the C2 and C8 atoms of purine are labeled to ~5%. Supplementing the growth media with (13)C-formate increases the labeling at C8 to ~88%, but not C2. Not unexpectedly, addition of exogenous formate is unnecessary for attaining the high enrichment levels of ~88% for the C2 and C8 purine positions in a (13)C-1,3-glycerol based growth. Furthermore, the ribose ring is labeled in all but the C4' carbon position, such that the C2' and C3' positions suffer from multiplet splitting but the C5' position remains singlet and the C1' position shows a small amount of residual C1'-C2' coupling. As expected, all the protonated base atoms, except C6, are labeled to ~90%. In addition, labeling with (13)C-1,3-glycerol affords an isolated methylene ribose with high enrichment at the C5' position (~90%) that makes it particularly attractive for NMR applications involving CH(2)-TROSY modules without the need for decoupling the C4' carbon. To simulate the tumbling of large RNA molecules, perdeuterated glycerol was added to a mixture of the four nucleotides, and the methylene TROSY experiment recorded at various temperatures. Even under conditions of slow tumbling, all the expected carbon correlations were observed, which indicates this approach of using nucleotides obtained from DL323 E. coli will be applicable to high molecular weight RNA systems.     

A proton NMR study of a DNA dumb-bell structure with hairpin loops of only two nucleotides: d(CACGTGTGTGCGTGCA).

One- and two-dimensional nuclear magnetic resonance (NMR) experiments were used to study the conformation of the DNA hexadecanucleotide d(CACGTGTGTGCGTGCA) in aqueous solution. NMR spectra were recorded for the compound in D2O and in H2O/D2O (90/10) over the temperature range 1 degree C-60 degrees C. Assignments of imino proton resonances and of non-exchangeable proton resonances (except for some H4', H5' and H5" resonances) are given. The 1H-NMR spectra indicate that below about 20 degrees C, the compound exists as a single monomolecular species. Between 20 degrees C and 55 degrees C the oligonucleotide occurs as a mixture of structures in fast exchange on the NMR time scale, except for the temperature region 30 degrees - 34 degrees C, where substantial line broadening indicates intermediate exchange; above 60 degrees C the single strand predominates. The imino proton spectra, chemical shift values, and scalar coupling and NOE data reveal that the monomeric form, which is exclusively present below 20 degrees C, consists of a structure with a B-DNA double helix region of six base pairs, both ends of which are closed by hairpin loops of only two nucleotides, giving the molecule a dumbbell-like structure: [sequence: see text].     

Biosynthesis of abscisic acid by the non-mevalonate pathway in plants, and by the mevalonate pathway in fungi

The biosynthetic pathways to abscisic acid (ABA) were investigated by feeding [1-(13)C]-D-glucose to cuttings from young tulip tree shoots and to two ABA-producing phytopathogenic fungi. 13C-NMR spectra of the ABA samples isolated showed that the carbons at 1, 5, 6, 4', 7' and 9' of ABA from the tulip tree were labeled with 13C, while the carbons at 2, 4, 6, 1', 3', 5', 7', 8' and 9' of ABA from the fungi were labeled with 13C. The former corresponds to C-1 and -5 of isopentenyl pyrophosphate, and the latter to C-2, -4 and -5 of isopentenyl pyrophosphate. This finding reveals that ABA was biosynthesized by the non-mevalonate pathway in the plant, and by the mevalonate pathway in the fungi. 13C-Labeled beta-carotene from the tulip tree showed that the positions of the labeled carbons were the same as those of ABA, being consistent with the biosynthesis of ABA via carotenoids. Lipiferolide of the tulip tree was also biosynthesized by the non-mevalonate pathway.     

Carbon-13 nuclear-magnetic-resonance spectra of adenine cyclonucleosides and their phosphates. Effects of neighboring groups for elucidation of fine structure of nucleosides and nucleotides.

Carbon-13 nuclear magnetic resonance spectra of adenine cyclonucleosides, which have a fixed glycosidic conformation in an anti range, and their isopropylidene and phosphate esters are reported; those of 9-beta-D-arabinofuranosyladenine and its 5'-phosphate are also presented. The chemical shifts of the base carbons are affected not only by the bridging atom but also by the position of the bridged sugar carbon which determine the planarity of the third ring formed by cyclization between the base and the sugar. The effects of glycosidic conformation on the sugar-carbon chemical shifts are discussed by comparison of the data for 8:5'-cycloadenosines with the data for adenosine and its 8-substituted derivatives. The effects of a 2'-oxygen on sugar-carbon chemical shifts have been examined by comparing the data for 2'-deoxyadenosine, arabinosyladenine and 8:2'-anhydro-8-oxy-9-beta-D-arabinofuranosyladenine. The effects of phosphomonoester groups on base and sugar carbon resonances have been examined and it is noted that these groups cause downfield shifts for C-8 of all cyclonucleotides. Data for the 3':5'-cyclic monophosphate derivative of 8:2'-anhydro-8-thio-9-beta-D-arabinofuranosyladenine suggest that the previous assignments of C-4' and C-3' for nucleoside 3':5'-cyclic monophosphates must be reversed. According to the reversed assignments, it seems that C-3' and C-5' show moderate downfield shifts and C-4' shows a marked upfield shift.     

A study of the thermal unfolding of Escherichia coli phenylalanine transfer RNA by chemical modification at elevated temperatures.

Escherichia coli tRNAPhe was modified by 3 M sodium bisulphite pH 6.0 for 24 h in the temperature range 25 degrees C (x 5 degrees C) to 55 degrees C and in the absence of added magnesium ions. The sites and extents of conversion of cytidines to uridine occurring at each temperature were determined by fingerprinting. The new sites of cytidine modification found at higher reaction temperatures were assumed to arise from breakage of secondary and tertiary structure hydrogen bonds involving cytidine residues. From these data, we conclude that hydrogen bonds within the 'complex core' of the tRNA (including the base pairs G-10 . C-25, C-11 . G-24 and C-13 . G-21 within the dihydrouridine stem and the tertiary structure base pair G-15 . C-48 melt at a lower temperature than the tertiary structure hydrogen bonds between G-19 in the dihydrouridine loop and C-56 in the TpsiC loop.     

On the base-stacking in the 5'-terminal cap structure of mRNA: a fluorescence study.

The fluorescence at 370 nm of the 7-methylguanosine residue (m7G) is found to be quenched when the base residue is involved in a stacking interaction with the adenosine residue in the cap structure m7G5' pppA of an eukaryotic mRNA. On the basis of the observed degree of quenching, the amounts of the stacked and unstacked forms in the cap structure have been determined at various temperatures and pH's. It has been found that at pH 6.2 effective enthalpy and entropy in the unstacked leads to stacked change are delta H degrees = 4.4 +/- 0.1 kcal/mole and delta S degrees = - 14.3 +/- 0.2 e.u., respectively. The pka value for the m7G residue is found to be 7.7 at 10 degrees C and 7.3 at 30 degrees C. The stacked structure seems to be less favourable in the deprotonated form that occurs in the higher pH solution. A similar analysis of some other cap structures indicates that the stacked form in m7G5' pppN structure is favourable if N is a purine nucleoside or a 2'-O-methylpyrimidine nucleoside but not for an unmethylated pyrimidine nucleoside.     

NMR study of stacking interactions and conformational adjustments in the dinucleotide-carcinogen adduct 2'-deoxycytidylyl-(3----5)-2'-deoxy-8-(N-fluoren-2-ylac etamido)guanosi ne

The conformation and dynamics of the dinucleotide d-CpG modified at the C(8) position of the guanine ring by the carcinogen 2-(acetylamino)fluorene has been investigated by high-field 1H NMR spectroscopy. A two-state analysis of chemical shift data has enabled estimation of the extent of intramolecular stacking in aqueous solution as a function of temperature. The stacking, which is mostly fluorene-cytosine, is virtually complete in the low-temperature range. The 500-MHz 1H NMR spectrum consists of two subspectra near ambient temperatures due to a 14.3 +/- 0.3 kcal/mol barrier to internal rotation about the amide bond in the stacked form. A large barrier to internal rotation about the guanyl-nitrogen bond at C(8) has also been ascertained, but separate NMR subspectra were not detected due to the predominance of one of the torsional diastereomers (alpha' = 90 degrees) in the fully stacked state. Problems of self-association and chemical exchange were identified and overcome to enable analysis of the sugar-phosphate backbone conformation utilizing coupling constants. For the exocyclic C(4')-C(5') bond of the deoxyguanosine moiety, there is a high gauche+ (gamma = 60 degrees) conformer population, which is uncommon for a purine nucleotide with a syn orientation about the glycosyl bond. The gauche- conformation (gamma = 300 degrees), which is normally present in syn purine nucleotides in solution, was not detected. The exocyclic C(5')-O(5') torsion of the deoxy-guanosine moiety remains near the classical energy minimum (beta = 180 degrees) in the major stacked conformations. The sugar ring of the deoxycytidine moiety is predominantly in the C2'-endo conformation, while the deoxyguanosine ring is a mixture of conformations, one of which appears to be unusually puckered. The results support intercalation models of modified DNA and suggest a looped-out structure, with the modified guanine being the first base in the loop. Such structures could explain the relatively rapid rate of repair and the frame-shift mutations of this type of adduct.     

Conformation of the common purine (beta) ribosides in solution: further evidence for a correlation between N-S state of the ribose moiety and syn-anti equilibrium.

The solution conformations of adenosine, guanosine and inosine in liquid ND3 have been determined by NMR. Comparison of the Karplus analysis of the proton HR spectra of the ribose moiety obtained in this solvent with the data from aqueous solutions of A and I proves that the conformations of the nucleosides are very similar in both liquids. From the analysis of the vicinal coupling constants of the ring protons it has been deduced that the S state C(2')-endo is slightly preferred. The mole fraction in S approximates 0.6 for all three nucleosides. C-13 relaxation measurements have been applied in the determination of the correlation times for rotational diffusion. Only at temperatures below - 40 degrees C is the pseudo-rotation of the furanoside ring slowed down sufficiently for it not to contribute to the measured relaxation rates. From NOE studies and T1 measurements on the individual protons it is derived that the N, C(3')-endo, form of the ribose is correlated with an anti conformation of the base (Y approximately 210 degrees to 220 degrees) and the S, C(2')-endo, form of the ribose with a syn conformation of the base (Y approximately 30 degrees to 50 degrees). The glycosyl torsion angles derived for the two conformations of A, G, and I are equal within the limits of accuracy.     

Bisulfite-induced C changed to U transitions in yeast valine tRNA.

The reaction of yeast tRNAVallab with NaHSO3 at 25 degrees and pH 5.8 has been studied. Six reactive residues have been located. C-17 in loop I is the most reactive (51% conversion) and C-73 in the first base pair of the acceptor stem the least reactive (8%). Three of the remaining reactive residues (C-39 in loop II, C-75 and C-76 near the acceptor stem) react to the same extent (36 to 38%) under the conditions of the experiment. C-37 in the anticodon reacted to a lesser extent (28%) than C-39 (36%), located just 2 residues away in the anticodon loop. No other changes were detected, but kinetic data suggest one or more additional residues may react very slowly. The C changed to U change in the anticodon (iac changed to iau) is a missense change (Val changed to Ile). Both mechanistic considerations and experimental data from the literature show that HSO3--induced deamination of cytosine residues occurs only at unstacked residues. We interpret the quantitative changes in tRNAVal to indicate that C-17 spends a large portion of its lifetime in an unstacked conformation. The stacking lifetimes of C-37, C-39, C-75, and C-76 seem to be similar but not identical. All other cytidine residues are much more tightly stacked. These results are consistent with the folded cloverleaf models that have been proposed from x-ray diffraction studies of yeast tRNAPhe. Residues C-46, C-49, C-57, and C-61, which are present in the single-stranded regions of the unfolded cloverleaf structure, do not react, suggesting that they are tightly stacked in solution under the conditions of this experiment. The data also suggest that anticodon-loop conformations other than the extremes with five bases stacked on either the 3' or 5' portion of the anticodon stem exist in solution and that the anticodon loop is flexible.     

NMR evidence of the stabilisation by the carcinogen N-2-acetylaminofluorene of a frameshift mutagenesis intermediate.

Two heteroduplexes d(C1A2C3T4C5G6C7A8C9A10C11)-d (G12T13G14T15G16G17A18G19T20G21) containing a bulged guanine either unmodified or modified with the carcinogen N-2-acetylaminofluorene (AAF) have been studied by nuclear magnetic resonance (NMR) as models of slipped mutagenic intermediates (SMI). Conformational equilibria are observed in both the unmodified and the AAF-modified heteroduplexes. The major conformation of the unmodified duplex is one where the extra guanine is stacked in the helix and the major conformation of the AAF-modified heteroduplex is one where the AAF is external to the helix. Unusual sugar proton chemical shifts of C5- and G6-AAF indicate that the AAF ring is pointing out in the 5' direction. A strong increase in the modified heteroduplex melting temperature (+15 degrees C) is observed. Moreover, in contrast to the unmodified heteroduplex, which shows extensive melting in the vicinity of the bulged guanine, the base pairs around the bulge in the AAF-modified heteroduplex remain paired at temperatures up to 30 degrees C. This exceptional stability of the site around the bulged modified guanine is suggested to be responsible for the high rate of -1 mutation induced by AAF at repetitive sequences.     

Conformational transitions in thymidine bulge-containing deoxytridecanucleotide duplexes. Role of flanking sequence and temperature in modulating the equilibrium between looped out and stacked thymidine bulge states

Structural features at extra thymidine bulge sites in DNA duplexes have been elucidated from a two-dimensional NMR analysis of through-bond and through-space connectivities in the otherwise self-complementary d(C-C-G-T-G-A-A-T-T-C-C-G-G) (GTG 13-mer) and d(C-C-G-G-A-A-T-T-C-T-C-G-G) (CTC 13-mer) duplexes in aqueous solution. These studies establish that the extra thymidine flanked by guanosines in the GTG 13-mer duplex is in a conformational equilibrium between looped out and stacked states. The looped-out state is favored at low temperature (0 degrees C), whereas the equilibrium shifts in favor of the stacked state at elevated temperatures (35 degrees C) prior to the onset of the duplex-strand transition. By contrast, the extra thymidine flanked by cytidines in the CTC 13-mer duplex is looped out independent of temperature in the duplex state. Our results demonstrate that temperature and flanking sequence modulate the equilibrium between looped-out and stacked conformations of single base thymidine bulges in DNA oligomer duplexes.     

Interactions of D-cellobiose with p-toluenesulfonic acid in aqueous solution: a (13)C NMR study

The effects of adding D(2)SO(4), and p-toluenesulfonic acid-d to D-cellobiose dissolved in D(2)O were investigated at 23°C by plotting (13)C NMR chemical shift changes (Δδ) against the acid to D-cellobiose molar ratio. (13)C Chemical shifts of all 18 carbon signals from α and β anomers of D-cellobiose showed gradual decreases due to increasing acidity in aqueous D(2)SO(4) medium. The C-1 of the α anomer showed a slightly higher response to increasing D(+) concentration in the surrounding. In the aqueous p-toluenesulfonic acid-d medium, C-6' and C-4' carbons of both α, and β anomeric forms of D-cellobiose are significantly affected by increasing the sulfonic acid concentrations, and this may be due to a 1:1 interaction of p-toluenesulfonic acid-d with the C-6', C-4' region of the cellobiose molecule.     

13C NMR of the bases of three DNA oligonucleotide duplexes: assignment methods and structural features

Natural abundance 13C NMR spectra of three DNA oligomers have been obtained. Most of the base resonances are well resolved from one another. A combination of two independent methods was used in making assignments: a one-dimensional spectral comparison method and a two-dimensional proton-detected 1H-13C correlated experiment for the protonated carbons. There are large shielding changes (between 1.62 and -1.40 ppm) upon thermal dissociation of the duplex. The shapes of the chemical shift vs temperature curves are largely independent of sequence. The base carbon resonance frequencies are sensitive to hydrogen bonding, base stacking, sugar conformation, and changes in the glycosyl torsion angle.     

Carbon-13 as a tool for the study of carbohydrate structures,conformations and interactions

The application of ¹³C-NMR spectroscopy to problems involving the structures and interactions of carbohydrates is described. Both ¹³C-enriched and natural abundance compounds were used and some advantages of the use of the stable isotope are described. Carbon-carbon and carbon-proton coupling constants obtained from 1-¹³ C enriched carbohydrates were employed in the assignment of their chemical shifts and to establish solution conformation. In all cases studied thus far, C-3 couples to C-1 only in the β-anomers while C-5 couples to C-1 only in the α-anomers. C-6 and C-2 always couple to C-1 in both anomeric species. The alkaline degradation of glucose [1-¹³ C] to saccharinic acids was followed by ¹³C-NMR. The conversion of glucose [1-¹³ C] to fructose-1,6-bisphosphate [1,6-¹³ C] by enzymes of the glycolytic pathway was shown as an example of the use of ¹³C-enriched carbohydrates to elucidate biochemical pathways. In a large number of glycosyl phosphates the ³¹P to H-1 and ³¹P to C-2 coupling constants demonstrate that in the preferred conformation the phosphate group lies between the O-5 and the H-1 of the pyranose ring. The influence of paramagnetic Mn²⁺ ions on the proton decoupled ¹³C-NMR spectra of uridine diphosphate N-acetylglucosamine indicates that the Mn²⁺ interacts strongly with the pyrophosphate moiety and with the carbonyl groups of the uracil and N-acetyl groups.     

The conformation of abscisic acid by n.m.r. and a revision of the proposed mechanism for cyclization during its biosynthesis.

The n.m.r. spectrum of abscisic acid (ABA) formed from [1,2-13C2]acetate by the fungus Cercospora rosicola shows 13C-13C coupling between C-6' (41.7 p.p.m.; 36 Hz) and the downfield 6'-methyl group (6'-Me) (24.3 p.p.m, 36 Hz). This 6'-Me, therefore, is derived from C-3' of mevalonate [Bennett, Norman & Maier (1981) Phytochemistry 20, 2343-2344]. An i.n.e.p.t. (insensitive nuclei enhanced by polarization transfer) pulse sequence demonstrated that the downfield 13C signal is produced by the 6'-Me that gives rise to the upfield 1H 6'-Me signal (23.1 d). The absolute configuration of this, the equatorial 6'-Me group, was determined as 6'-pro-R by decoupling and n.O.e. (nuclear-Overhauser-enhancement) experiments at 300 MHz using ABA, ABA in which the axial 6'-pro-S 5'-hydrogen atom had been exchanged with 2H in NaO2H and the 1',4'-cis- and 1',4'-trans-diols formed from these samples. The configuration at C-1' and at C-6' are now compatible with a chair-folded intermediate during cyclization, as proposed for beta- and epsilon-rings of carotenoids. ABA in solution exists, as in the crystalline form, with the ring in a pseudo-chair conformation. The side chain is axial and the C-3 Me and the C-5 hydrogen atoms are predominantly cis(Z).