31P NMR spectra of an oligodeoxyribonucleotide duplex lac operator-repressor headpiece complex

The interaction of a symmetric lac operator duplex, d(TGTGAGCGCTCACA)2, with the N-terminal 56-residue headpiece fragment of the lac repressor protein was monitored by 31P NMR spectroscopy. The changes in the 31P chemical shifts upon addition of the headpiece demonstrated an end point of two headpiece fragments per symmetric 14-mer duplex with each headpiece binding to the T1pG2pT3pG4pA5 ends of the duplex. The specific phosphate 31P perturbations observed are consistent with those residues implicated in protein binding by previous NMR, molecular biological, and biochemical techniques. Upon complexation, the 31P signals of phosphates G2-A5 showed upfield or downfield shifts (less than 0.2 ppm) while most other residues were unperturbed. The interactions were dependent on ionic strength. The 31P NMR data provide direct evidence for predominant recognition of the 5' strand of the 5'-TGTGA/3'-ACACT binding site.     

Effect of distortions in the deoxyribose phosphate backbone conformation of duplex oligodeoxyribonucleotide dodecamers containing GT, GG, GA, AC, and GU base-pair mismatches on 31P NMR spectra

We have previously suggested that variations in the 31P chemical shifts of individual phosphates in duplex oligonucleotides are attributable to torsional angle changes in the deoxyribose phosphate backbone. This hypothesis is not directly supported by analysis of the 1H/31P two-dimensional J-resolved spectra of a number of mismatch dodecamer oligonucleotide duplexes including the following sequences: d-(CGTGAATTCGCG), d(CGUGAATTCGCG), d(CGGGAATTCGCG), d(CGAGAATTCGCG), and d(CGCGAATTCACG). The 31P NMR signals of the dodecamer mismatch duplexes were assigned by 2D 1H/31P pure absorption phase constant time (PAC) heteronuclear correlation spectra. From the assigned H3' and H4' signals, the 31P signals of the base-pair mismatch dodecamers were identified. JH3'-P coupling constants for each of the phosphates of the dodecamers were obtained from 1H/31P J-resolved selective proton flip 2D spectra. By use of a modified Karplus relationship, the C4'-C3'-O3'-P torsional angles (epsilon) were obtained. JH3'-P coupling constants were measured for many of the oligonucleotides as a function of temperature. There exists a good linear correlation between 31P chemical shifts and the epsilon torsional angle. This correlation can be further extended to the C3'-O3'-P-O5' torsional angle (zeta) by using a linear relationship between epsilon and zeta obtained from crystal structure studies. The 31P chemical shifts follow the general observation that the more internally the phosphate is located within the oligonucleotide sequence, the more upfield the 31P resonance occurs. In addition, 31P chemical shifts show sequence- and site-specific variations. Analysis of the backbone torsional angle variations from the coupling constant analysis has provided additional information regarding the origin of these variations in 31P chemical shifts.     

Two-dimensional 1H and 31P NMR spectra and restrained molecular dynamics structure of a mismatched GA decamer oligodeoxyribonucleotide duplex

Assignment of the 1H and 31P NMR spectra of a tandem G.A mismatched base pair decamer oligodeoxyribonucleotide duplex, d(CCAAGATTGG)2, has been made by two-dimensional 1H-1H and heteronuclear 31P-1H correlated spectroscopy. Unusual downfield 31P resonances have been assigned by a pure absorption phase constant-time heteronuclear 31P-1H correlated spectrum to be associated with the phosphates on the 5'- and 3'-sides of the mismatched guanosine residue. JH3'-P coupling constants for each of the phosphates of the decamer were obtained from the 1H-31P J-resolved selective proton-flip 2D spectrum. The two most downfield-shifted 31P resonances each appear to consist of two overlapping signals that can be resolved into two distinct doublets with different coupling constants in the J-resolved spectrum. This as well as the temperature dependence of the 31P spectra demonstrates that two distinct conformations exist at lower temperatures. By use of a modified Karplus relationship, the C4'-C3'-O3'-P torsional angles (epsilon) were obtained. A linear correlation between 31P chemical shifts and the measured coupling constants is quite good (only when the larger set of coupling constants of the two most downfield 31P signals is included). The 31P chemical shifts as well as the measured coupling constants tend to follow the positional variation seen in other duplexes of interior phosphates resonating more upfield than terminal residues and of interior phosphates exhibiting smaller coupling constants; however, this pattern is disrupted at the site of the mismatch. Modeling and initial NOESY distance restrained molecular mechanics energy minimization and restrained molecular dynamics support previous observations that the mismatched guanine and adenine bases are both in anti conformations. Most significantly, the epsilon backbone torsional angle variaions calculated from the NOESY distance restrained structures are in agreement with both the crystal structure values and the measured JH3'-P coupling constants.     

Two-dimensional 1H and 31P NMR spectra and restrained molecular dynamics structure of an extrahelical adenosine tridecamer oligodeoxyribonucleotide duplex

Assignment of the 1H and 31P NMR spectra of an extrahelical adenosine tridecamer oligodeoxyribonucleotide duplex, d(CGCAGAATTCGCG)2, has been made by two-dimensional 1H-1H and heteronuclear 31P-1H correlated spectroscopy. The downfield 31P resonance previously noted by Patel et al. (1982) has been assigned by both 17O labeling of the phosphate as well as a pure absorption phase constant-time heteronuclear 31P-1H correlated spectrum and has been associated with the phosphate on the 3' side of the extrahelical adenosine. JH3'-P coupling constants for each of the phosphates of the tridecamer were obtained from the 1H-31P J-resolved selective proton-flip 2D spectrum. By use of a modified Karplus relationship the C4-C3'-O3-P torsional angles (epsilon) were obtained. There exists a good linear correlation between 31P chemical shifts and the epsilon torsional angle. The 31P chemical shifts and epsilon torsional angles follow the general observation that the more internal the phosphate is located within the oligonucleotide sequence, the more upfield the 31P resonance occurs. Because the extrahelical adenosine significantly distorts the deoxyribose phosphate backbone conformation even several bases distant from the extrahelical adenosine, 31P chemical shifts show complex site- and sequence-specific variations. Modeling and NOESY distance-restrained energy minimization and restrained molecular dynamics suggest that the extrahelical adenosine stacks into the duplex. However, a minor conformation is also observed in the 1H NMR, which could be associated with a structure in which the extrahelical adenosine loops out into solution.     

31P NMR of phosphate and phosphonate complexes of metalloalkaline phosphatases.

31P NMR spectra of phosphate and phosphonate complexes of Escherichia coli alkaline phosphatase have been obtained by Fourier transform NMR methods. One equivalent of P1i, bound to Zn(II) alkaline phosphatase, pH 8, gives rise to a single 31P resonance 2 ppm downfield from that for Pi, and assignable to the noncovalent complex, E-P. Inorganic phosphate in excess of 1 eq per enzyme dimer gives rise to a resonance at the position expected for free Pi. At pH 5.1, a second resonance appears 8.5 ppm downfield from that for free Pi, and is assignable to the covalent complex, E-P. The large downfield shift suggests that the enzyme phosphoryl group is highly strained with an O-P-O bond angle of under 100 degrees.     

Graph-theoretical assignment of secondary structure in multidimensional protein NMR spectra: Application to the lac repressor headpiece

Summary A novel procedure is presented for the automatic identification of secondary structures in proteins from their corresponding NOE data. The method uses a branch of mathematics known as graph theory to identify prescribed NOE connectivity patterns characteristic of the regular secondary structures. Resonance assignment is achieved by connecting these patterns of secondary structure together, thereby matching the connected spin systems to specific segments of the protein sequence. The method known as SERENDIPITY refers to a set of routines developed in a modular fashion, where each program has one or several well-defined tasks. NOE templates for several secondary structure motifs have been developed and the method has been successfully applied to data obtained from NOESY-type spectra. The present report describes the application of the SERENDIPITY protocol to a 3D NOESY-HMQC spectrum of the ¹⁵N-labelled lac repressor headpiece protein. The application demonstrates that, under favourable conditions, fully automated identification of secondary structures and semi-automated assignment are feasible.Abbreviations 2D, 3D two-, three-dimensional - NOESY nuclear Overhauser enhancement spectroscopy - HMQC heteronuclear multiple quantum coherence - SSE secondary structure element - SERENDIPITY SEcondary structuRE ideNtification in multiDImensional ProteIn specTra analYsis Supplementary Material available from the authors: Two tables containing the total number of mappings resulting from the graph search procedure for simulated and experimental NOE data.     

Two-dimensional 1H and 31P NMR spectra of a decamer oligodeoxyribonucleotide duplex and a quinoxaline ((MeCys3, MeCys7]TANDEM) drug duplex complex

Assignment of the 1H and 31P NMR spectra of a decamer oligodeoxyribonucleotide duplex, d(CCCGATCGGG), and its quinoxaline ((MeCys3, MeCys7]TANDEM) drug duplex complex has been made by two-dimensional 1H-1H and heteronuclear 31P-1H correlated spectroscopy. The 31P chemical shifts of this 10 base pair oligonucleotide follow the general observation that the more internal the phosphate is located within the oligonucleotide sequence, the more upfield the 31P resonance occurs. While the 31P chemical shifts show sequence-specific variations, they also do not generally follow the Calladine "rules" previously demonstrated. 31P NMR also provides a convenient monitor of the phosphate ester backbone conformational changes upon binding of the drug to the duplex. Although the quinoxaline drug, [MeCys3, MeCys7]TANDEM, is generally expected to bind to duplex DNA by bis-intercalation, only small 31P chemical shift changes are observed upon binding the drug to duplex d(CCCGATCGGG). Additionally, only small perturbations in the 1H NMR and UV spectra are observed upon binding the drug to the decamer, although association of the drug stabilizes the duplex form relative to the other states. These results are consistent with a non-intercalative mode of association of the drug. Modeling and molecular mechanics energy minimization demonstrate that a novel structure in which the two quinoxaline rings of the drug binds in the minor groove of the duplex is possible.     

Assignments of 31P NMR resonances in oligodeoxyribonucleotides: origin of sequence-specific variations in the deoxyribose phosphate backbone conformation and the 31P chemical shifts of double-helical nucleic acids

It is now possible to unambiguously assign all 31P resonances in the 31P NMR spectra of oligonucleotides by either two-dimensional NMR techniques or site-specific 17O labeling of the phosphoryl groups. Assignment of 31P signals in tetradecamer duplexes, (dTGTGAGCGCTCACA)2, (dTAT-GAGCGCTCATA)2, (dTCTGAGCGCTCAGA)2, and (dTGTGTGCGCACACA)2, and the dodecamer duplex d(CGTGAATTCGCG)2 containing one base-pair mismatch, combined with additional assignments in the literature, has allowed an analysis of the origin of the sequence-specific variation in 31P chemical shifts of DNA. The 31P chemical shifts of duplex B-DNA phosphates correlate reasonably well with some aspects of the Dickerson/Calladine sum function for variation in the helical twist of the oligonucleotides. Correlations between experimentally measured P-O and C-O torsional angles and results from molecular mechanics energy minimization calculations show that these results are consistent with the hypothesis that sequence-specific variations in 31P chemical shifts are attributable to sequence-specific changes in the deoxyribose phosphate backbone. The major structural variation responsible for these 31P shift perturbations appears to be P-O and C-O backbone torsional angles which respond to changes in the local helical structure. Furthermore, 31P chemical shifts and JH3'-P coupling constants both indicate that these backbone torsional angle variations are more permissive at the ends of the double helix than in the middle. Thus 31P NMR spectroscopy and molecular mechanics energy minimization calculations appear to be able to support sequence-specific structural variations along the backbone of the DNA in solution.     

Sequence-dependent variations in the 31P NMR spectra and backbone torsional angles of wild-type and mutant Lac operator fragments

Assignment of the 31P resonances of a series of six sequenced-related tetradecamer DNA duplexes, d(TGTGAGCGCTCACA)2, d(TATGAGCGCTCATA)2, d(TCTGAGCGCTCAGA)2, d(TGTGTGCGCACACA)2, d(TGTGACGCGTCACA)2 and d(CACAGTATACTGTG)2, related to the lac operator DNA sequence was determined either by site-specific 17O labeling of the phosphoryl groups or by two-dimensional 1H-31P pure absorption phase constant time (PAC) heteronuclear correlation spectroscopy. J(H3'-P) coupling constants for each of the phosphates of the tetradecamers were obtained from 1H-31P J-resolved selective proton flip 2D spectra. By use of a modified Karplus relationship the C4'-C3'-O3'-P torsional angles (epsilon) were obtained. Comparison of the 31P chemical shifts and J(H3'-P) coupling constants of these sequences has allowed greater insight into those various factors responsible for 31P chemical shift variations in oligonucleotides and provided an important probe of the sequence-dependent structural variation of the deoxyribose phosphate backbone of DNA in solution. These sequence-specific variations in the conformation of the DNA sugar phosphate backbone of various lac operator DNA sequences can possibly explain the sequence-specific recognition of DNA by DNA binding proteins, as mediated through direct contacts between the phosphates and the protein.     

31P NMR relaxation measurements of the phosphate backbone of a double stranded hexadeoxynucleotide in solution: determination of the chemical shift anisotropy

The five phosphates of the deoxynucleotide d(CpGpTpApCpG)₂ have been assigned by two-dimensional heteronuclear NMR spectroscopy. The chemical shift anisotropy and correlation time of each phosphate group has been determined from measurements of the spin-lattice, spin-spin relaxation rate constants and the ³¹P-{¹H} nuclear Overhauser enhancement (NOE) at three magnetic field strengths (4.7 T, 9.4 T, and 11.75 T) and two temperatures (288 K and 298 K). As expected, the relaxation data require two mechanisms to account for the observed rate constants, i.e. dipole-dipole and chemical shift anisotropy. At 9.4 T and 11.75 T, the latter mechanism dominates the relaxation, leading to insignificant NOE intensities. The correlation time, chemical shift anisotropy and effective P-H distance were obtained from least-squares fitting to the data. Comparison of the fitted value for the correlation time with that obtained from ¹H measurements shows that the molecule behaves essentially as rigid rotor on the nanosecond timescale. Large amplitude motions observed in long segments of DNA are due to bending motions that do not contribute significantly to relaxation in short oligonucleotides.Abbreviations CSA chemical shift anisotropy - NOE nuclear Overhauser enhancement Offprint requests to: A. N. Lane     

Tendencies of 31P chemical shifts changes in NMR spectra of nucleotide derivatives.

31P NMR chemical shifts of the selected mono- and oligonucleotide derivatives, including the compounds with P-N, P-C, P-S bonds and phosphite nucleotide analogues have been presented. The influence of substituents upon 31P chemical shifts has been discussed. The concrete examples of 31P chemical shifts data application in the field of nucleotide chemistry have been considered.     

Estrogen-induced changes in high-energy phosphate metabolism in rat uterus: 31P NMR studies

Changes in the concentrations of high-energy phosphate metabolites were measured by 31P NMR spectroscopy of surviving rat uteri from 0-48 h following estrogen administration. Concentrations (millimoles per kilogram wet weight) of these metabolites in the untreated immature uterus, measured at 4 degrees C, were found to be the following: creatine phosphate (CP), 2.1 +/- 0.2; nucleoside triphosphates, mainly adenosine 5'-triphosphate (ATP), 4.6 +/- 0.4; phospho monoesters, primarily sugar phosphates (SP), 5.4 +/- 0.7; and inorganic phosphate (Pi), 0.8 +/- 0.4. Adenosine 5'-diphosphate (ADP) concentration was estimated to be approximately 40 mumol/kg wet weight from the assumed equilibrium of the creatine kinase reaction. The concentration of CP, and to lesser extent ATP and SP, declined within the first 1.5-3 h after injection of 17 beta-estradiol, returned to control values between 6 and 12 h, and then increased, reaching maximal concentrations at 24 h. From the fractions of the total soluble ATP in free and Mg2+-bound forms, [free Mg2+] in the untreated uterus was estimated to be 0.2-0.4 mmol/kg wet weight. An increase in [free Mg2+] in the uterus was detected 1.5 h after estrogen injection. A subsequent parallel increase in the ratio of ATP to CP concentrations suggests that estrogen can also affect the apparent creatine kinase equilibrium by modulating [free Mg2+].     

31P NMR spectra of ethidium, quinacrine, and daunomycin complexes with poly(adenylic acid).poly(uridylic acid) RNA duplex and calf thymus DNA

31P NMR provides a convenient monitor of the phosphate ester backbone conformational changes upon binding of the intercalating drugs ethidium, quinacrine, and daunomycin to sonicated poly(A).poly(U) and calf thymus DNA. 31P chemical shifts can also be used to assess differences in the duplex unwinding angles in the presence of the drug. Thus a new 31P signal, 1.8-2.2 ppm downfield from the double-stranded helix signals, is observed in the ethidium ion-poly(A).poly(U) complex. This signal arises from phosphates which are in perturbed environments due to intercalation of the drug. This is in keeping with the hypothesis that the P-O ester torsional angle in phosphates linking the intercalated base pairs is more trans-like. Similar though smaller deshielding of the 31P signals is observed in sonicated poly(A).poly(U)-quinacrine complexes as well as in the daunomycin complexes. The effect of added ethidium ion, quinacrine, and daunomycin on the 31P spectra of sonicated calf thymus DNA is consistent with Wilson and Jones' (1982) earlier study. In these drug-DNA complexes the drug produces a gradual downfield shift in the DNA 31P signal without the appearance of a separate downfield peak. These differences are attributed to differences in the rate of chemical exchange of the drug between free and bound duplex states. The previous correlation of 31P chemical shift with drug duplex unwinding angle (Wilson & Jones, 1982) is confirmed for both the RNA and DNA duplexes.     

Characterization of the broad resonance in 31P NMR spectra of excised rat brain

31P NMR spectra of excised rat brain showed a broad resonance between-12 and -13 ppm. Subcellular fractions of brain, rich in membranes, exhibited the broad resonance and it was also present in isolated myelin, the major membrane component of brain. However, it was absent in brain cytosol (161,100 X g supernatant). Raising the temperature of the brain above 50 degrees C caused a gradual downfield chemical shift of the broad resonance, to about -1 ppm at 90 degrees C. An even larger downfield shift was produced by halothane or deoxycholate with concomitant narrowing of the line width of this resonance. Vesicles prepared from the phospholipids of excised brain or isolated myelin showed the broad resonance, and halothane produced the same downfield shift and peak sharpening in brain phospholipid vesicles as that in the intact brain. The chemical shift anisotropy was estimated to be 45 ppm for both myelin and the brain, as characteristic for biological membranes. The T1 and T2 relaxation times of the perpendicular 31P chemical shift tensor component of the broad resonance were 0.66 sec and 1.6 msec, respectively, in the same range as those for other biological membranes. Halothane-treatment of the brain increased both the T1 and T2 times considerably, as expected from the disruption of the phospholipid bilayer in a membrane. These data indicate that the broad resonance in the 31P NMR spectrum of excised rat brain originates exclusively from the phosphate head group of membrane bound phospholipids. Similar broad resonances were found in autopsied human brain and porcine spinal cord and to a lesser extent in excised rat liver and kidney.     

Unknown phosphate compounds in tail muscle of intact conscious newts by 31P NMR

Unknown phosphate resonances at 0 and -21.6 ppm have been identified in 31P NMR spectra of tail muscle of unanesthetized newts which do not correspond to known phosphate-bearing compounds in skeletal muscle cells. The concentrations of both unknowns decrease markedly during muscular activity and severe hypoxia (conditions associated with decreased intracellular pH and increased cellular levels of inorganic phosphate). The unknown at 0 ppm increases in concentration with imposition of moderate hypoxia. Our data suggest that these unknowns may be liable storage compounds for a high energy phosphate bond, and are involved in newt skeletal muscle phosphogen metabolism.     

Effect of pH and inorganic phosphate on creatine kinase inactivation: an in vitro 31P NMR saturation-transfer study.

The pseudo-first-order rate constant of rabbit muscle creatine kinase (CK), in the direction of ATP synthesis (kf), was determined by saturation-transfer 31P NMR. When pH was varied between 6.0 and 7.4, kf increased linearly at both 20 degrees C and 37 degrees c. The corresponding flux is very small between pH 6.0 and 6.5, in contrast to previous studies. Up to 50 h exposure of the CK enzyme to high concentrations of inorganic phosphate (Pi), a known inhibitor in certain situations, had negligible effect on enzymatic flux in the physiological pH range. Thus under in vivo conditions, such as in stroke, where pH falls as low as 6.2 and Pi rises to high levels, the rate of the CK reaction may be severely reduced due to pH but not due to high Pi concentrations.     

Enzyme-substrate and enzyme-inhibitor complexes of triose phosphate isomerase studied by 31P nuclear magnetic resonance.

The complex formed between the enzyme triose phosphate isomerase (EC 5.3.1.1.), from rabbit and chicken muscle, and its substrate dihydroxyacetone phosphate was studied by 31P n.m.r. Two other enzyme-ligant complexes examined were those formed by glycerol 3-phosphate (a substrate analogue) and by 2-phosphoglycollate (potential transition-state analogue). Separate resonances were observed in the 31P n.m.r. spectrum for free and bound 2-phosphoglycollate, and this sets an upper limit to the rate constant for dissociation of the enzyme-inhibitor complex; the linewidth of the resonance assigned to the bound inhibitor provided further kinetic information. The position of this resonance did not vary with pH but remained close to that of the fully ionized form of the free 2-phosphoglycollate. It is the fully ionized form of this ligand that binds to the enzyme. The proton uptake that accompanies binding shows protonation of a group on the enzyme. On the basis of chemical and crystallographic information [Hartman (1971) Biochemistry 10, 146--154; Miller & Waley (1971) Biochem. J. 123, 163--170; De la Mare, Coulson, Knowles, Priddle & Offord )1972) Biochem. J. 129, 321--331; Phillips, Rivers, Sternberg, Thornton & Wilson (1977) Biochem. Soc. Trans. 5, 642--647] this group is believed to be glutamate-165. On the other hand, the position of the resonance of D-glycerol 3 phosphate (sn-glycerol 1-phosphate) in the enzyme-ligand complex changes with pH, and both monoanion and dianon of the ligand bind, although dianion binds better. The substrate, dihydroxyacetone phosphate, behaves essentially like glycerol 3-phosphate. The experiments with dihydroxy-acetone phosphate and triose phosphate isomerase have to be carried out at 1 degree C because at 37 degrees C there is conversion into methyl glyoxal and orthophosphate. The mechanismof the enzymic reaction and the reasons for rate-enhancement are considered, and aspects of the pH-dependence are discussed in an Appendix.     

Triosephosphate isomerase catalysis is diffusion controlled. Appendix: Analysis of triose phosphate equilibria in aqueous solution by 31P NMR

The rates of the forward and reverse reactions of triosephosphate isomerase catalyzed by the wild-type and by a sluggish mutant enzyme have been studied in the absence and the presence of several viscosogenic agents. For the mutant enzyme, the kcat for which is some 10(3) times less than that for the wild-type enzyme, the value of kcat/Km with glyceraldehyde phosphate as substrate is almost unaffected by the presence of sucrose or glycerol, even though the concentration of the aldehyde form of the substrate is smaller because of hemiacetal formation. [The nature and relative amounts of the various forms of triose phosphate present in solution (free carbonyl forms, hydrates, dimers, hemiacetal adducts) have been evaluated by 31P NMR and are presented in the Appendix.] The viscosogenic agents cause the substrate to bind more tightly to the enzyme, roughly compensating for the lower substrate concentration. With dihydroxyacetone phosphate as substrate, the values of kcat/Km for the mutant enzyme increase with the addition of viscosogenic agent, consistent with the tighter binding of substrate without (in this case) any concomitant loss due to hemiketal formation. These results for the mutant enzyme (known to be limited in rate by an enolization step in the catalytic mechanism) can be used to interpret the behavior of the wild-type enzyme. Plots of the relative values of kcat/Km for catalysis by the wild-type enzyme (normalized with the corresponding data for the mutant enzyme) against the relative viscosity have slopes close to unity, as predicted by the Stokes-Einstein equation for a cleanly diffusive process. In the presence of polymeric viscosogenic additives such as poly(ethylene glycol), polyacrylamide, or ficoll, no effect on kcat/Km is seen for the wild-type enzyme, consistent with the expectation that molecular diffusion rates are unaffected by the macroviscosity and are only slowed by the presence of smaller agents that raise the microviscosity. These results show that the reaction catalyzed by the wild-type triosephosphate isomerase is limited by the rate at which glyceraldehyde phosphate encounters, or departs from, the active site.     

Scalar coupling between 31P and spin-1/2 metal nuclides in 31P NMR of complexes with a ligand containing phosphonate donor groups.

Scalar couplings between ³¹P and $\text{spin-}\text{1}\text{2}$ isotopes of cadmium, mercury, lead, and tin are reported for the respective metal complexes with the chelating agent (ethylenedinitrilo) - tetramethylenephosphonic acid.