Report of a conference held 29 February – 3 March 2000 at the Hotel Divani Palace, Athens, Greece under the auspices of the EU Biomed 2 programme
This conference was organised as a result of award of a grant to develop a Concerted Action to investigate the development and application of hyphenated chromatography techniques with NMR spectroscopy and mass spectrometry. The participants in this Concerted Action are Professor Steen Honoré Hansen of the Royal Danish School of Pharmaceutical sciences (Coordinator of the programme), Professor Jeremy K. Nicholson and Professor John C. Lindon of Imperial College, University of London, Professor Ian D. Wilson of AstraZeneca Pharmaceuticals, Macclesfield, UK, Dr. Manfred Spraul of Bruker Analytik, Rheinstetten, Germany and Professor Jostein Krane of University of Trondheim, Norway. As part of this programme it was agree to organise a conference/workshop ion the basic techniques and recent advances in the subject. The organisation of the conference was coordinated by Professor Hansen. In total, 86 delegates attended including many from the Concerted Action participating groups but also many from other organisations. The scientific sessions were split into four separate half day symposia and in addition there was a session to view 17 scientific posters submitted by the participants. The full programme is attached at the end of this document.
The first session on methodology was chaired by Professor Ian Wilson and included two overview lectures, the first on analytical NMR spectroscopy and hyphenated NMR techniques by Jeremy Nicholson of Imperial College, London, UK. This provided an overview of the techniques as applied to biomedical applications, in particular to HPLC-NMR studies of a tripeptide library relevant to combinatorial chemistry synthesis and to the characterisation of a mixture of 10 isomers of a chiral drug and in this case HPLC-CD (circular dichroism) proved useful for determining absolute configurations of the isomers and NMR was complementary in that it gave unambiguous information on cis and trans isomers. In addition the application of HPLC-NMR-MS to the identification of metabolites of substituted anilines was highlighted. The need for ultimate NMR sensitivity was stressed and the provision of NMR probes operating at cryogenic temperatures was urged.
The second lecture was given by John Lindon, also of Imperial College, London, UK, and comprised an overview of the technologies used for HPLC-NMR and HPLC-NMR-MS and capillary techniques such as CE-NMR and CEC-NMR. In HPLC-NMR-MS the two detectors can be operated in parallel or in series and examples of both modes were given from the field of xenobiotic metabolism in both plants (maize) and animals (drug metabolites in urine). In addition, the recent coupling of ICPMS with HPLC was mentioned and preliminary results using bromine atom-specific detection for identifying drug metabolites was shown.
The next talk was by Lee Griffiths of AstraZeneca, Macclesfield, UK, who described the use of software for automatic identification of compounds arising from multiple parallel syntheses using a robotic approach. He pointed out the usefulness of various methodologies, including flow through NMR probes, peak trapping of analytes from HPLC runs, a generic LC-MS system and new software for NMR spectral interpretation. Thus it is possible to autoanalyse the NMR spectra and allow the computer to make a decision on whether to use HPLC-NMR if necessary to trap and identify impurity peaks.
Finally, Udo Brinkman gave a fascinating lecture on "Hyphenation—hype or fascination" in which he described the use of LC-LC and LC-GC and GC-GC, i.e column switching studies, to analyse for pollutants in water, focussing on the needs for speed, sensitivity and selectivity. Examples included the use of LC-ESI-MS/MS for analysis of pesticides and their breakdown products and also cytotoxins such as zeatin. He showed how by selection of techniques it was possible to increase sensitivity by going to smaller samples and carrying out less solvent extraction. He showed a detection limit of 1 mg kg–1, using only using 1 mL of milk to detect banned compounds. He also mentioned SPE-GC to assay for 87 compounds in raw water. Finally, he showed GC-TOFMS studies yielding very fast MS results as a consequence of the TOF. Also he showed interesting two-dimensional GC-GC results, using two capillaries, one of 50 m and the second of 1 m such that all alcohols came out on one trajectory in the 2D plot and aromatics and ethers came out along different 2D trajectories. To identify the thousands of detected compounds, TOFMS would be needed.
The second session was chaired by Jeremy Nicholson and was concerned with pharmaceutical applications of HPLC-NMR-MS. The first talk was by Ulla Sidelmann from Novo-Nordisk, Denmark, and she covered applications of HPLC-NMR to drug metabolism. The speed and efficiency of the approach was the main reason it was used in drug metabolism studies. She showed results on the metabolites of tolfenamic acid using 800 MHz LC-NMR, and results on metabolites of naproxen using LC-NMR-MS. The next application covered the problem of transacylating ester glucuronide metabolites of carboxylate drugs. Thus many drugs form such glucuronides which can undergo an internal rearrangement to successively transfer the aglycone moiety to the 2, 3 and 4 positions of the glucuronic acid species which can then also mutarotate to their a anomers. These isomers are resistant to b-glucuronidase and have been implicated in drug toxicity problems. Thus, it is of importance to understand the kinetics of these reactions in relation to the half-life of the drug in the body as well as the protein binding propensities of these isomers. It was shown how HPLC-NMR could be use to identify the individual isomers and also to measure the reaction kinetics using model fluorobenzoic acid glucuronides and also prebenecid and isoxepac glucuronides. Finally, examples of drug metabolite problems from microsomal incubations within NovoNordisk were highlighted.
Next Olivia Corcoran from Imperial College, London, UK, gave an explanation of why transacylation reactions are important in drug development because of the possibility of macromolecular binding and showed examples of drugs which produce glucuronides such as naproxen and diclofenac and showed how HPLC-NMR can be used to identify these. Following this, it was indicated how HPLC-NMR is used to study the reaction kinetics and the identification of a previously unidentified component was elucidated. This turned out to be the 1-O-a-isomer which is formed from the 2-O-a-isomer and this reaction has always been discounted in the past. This observation has now been shown to be general.
Next Rasmus Mortenesen, Royal Danish School of Pharmacy, Copenhagen, Denmark, gave a talk showing how that whilst HPLC-NMR is necessary to identify the individual transacylated isomeric glucuronides, HPLC with UV detection is much faster for providing the kinetic information. He pointed out that most literature studies only focussed on the 1-O-b isomer degradation and that this was insufficient and irrelevant to the toxicity. He used naproxen glucuronide as an example. A number of careful studies were explained including the effect on the kinetics of adding acetonitrile which is necessary to effect an HPLC separation. He also showed that using stop-flow HPLC-NMR to obtain kinetic information on individual isomers and using HPLC-UV gave essentially the same results and that nothing new could be determined by the time-consuming HPLC-NMR process on single isomers. Transacylation in D2O is slower than in H2O and reasons for this were proposed. Also kinetics are different in plasma and whilst transacylation is generally slower, other enzymes are present and hydrolysis can be much faster.
The final talk in this session was from John Shockcor of Dupont Pharmaceuticals, Newark, Delaware, USA. He described his fully-integrated HPLC-NMR-MS system with the splitter just after the column and before the UV diode array detector. He discussed the physical arrangement of the equipment in his laboratory including a QTOF-MS. He also described the latest commercial software for controlling the chromatography, the NMR and the MS. The method has been applied to identifying glutathione and glucuronide metabolites of DPC963, an anti-HIV compound. He gave an example of time-slicing over an HPLC peak to study metabolites in a rat bile fraction after SPEC in which he carried out a stop-flow NMR acquisition every 30 s of the chromatogram giving an overnight data acquisition period. He advocated the use of acetonitrile-d3 as opposed to protio-acetonitrile as the benefit-to-cost ratio was high. Identification of the metabolites was based on one-dimensional NMR spectra mainly, but TOCSY two-dimensional spectra were used as appropriate.
The third session was chaired by John Lindon and was devoted to advances in methodology. This was opened by Ernst Bayer, University of Tubingen, Germany, who spoke about on-line coupling of electro-driven separation techniques with NMR spectroscopy and MS. He outlined the benefits of capillary separation methods and pointed out the better separation efficiency of techniques such as capillary electrophoresis (CE) and capillary electrochromatography (CEC). He showed the development of a method of packing capillary columns for CEC without the use of frits and gave application results using CEC-MS in the enkephalin area. He went on to describe RF coil design for CE-NMR spectroscopy and compared results from CE-NMR with those from HPLC-NMR on alkylbenzoates. He pointed out that stop-flow operation is possible as well as two-dimensional NMR methods for molecular identification using caffeine and aspartame in soft drinks as an example. The very small amount of eluent required means that deuterated solvents can be used cheaply. Next, he discussed coordination ion mass spectrometry for cases where electrospray ionisation gives poor results. Thus pre- or post-column separation, a charged metal complex is produced by injection of a reagent and this has now been applied to a variety of applications including unsaturated fatty acids in lipids using HPLC, CE and CEC with the last approach being the best. If silver complexes are possible then one can use the silver isotope pattern to gain extra information. In MS/MS studies it was shown that the product ions can also still contain the metal ion, and sometimes MS can be used to derive stereochemical information, for example in carbohydrates where a separation of trisaccharides from St John's Wort was achieved, with MS/MS used to identify the compounds.
This talk was followed by Andrew Webb from the Beckman Institute for Advanced Sciences and Technology, Urbana, Illinois USA, who gave a lecture on hyphenation of micro-separation techniques with high sensitivity detectors. He initially described the design of micro coil detectors and discussed the benefits and disadvantages of such systems. One of the main problems is broadening of NMR peaks with very small coils because of magnetic susceptibility mismatching and this was ovecome using a perfluorocarbon to surround the RF coils and improve the susceptibility matching with the sample. Using this approach it is now possible to get reasonable signal-noise ratios on 1 nmole in a few mL volume in 1 min. The detector has been used in conjunction with capillary HPLC to analyses amino acid mixtures and terpenoids. Again deuterated solvents can be used at low cost. Both on-flow and stop-flow experiments are possible. Because the detector cell is so small there is a problem in stop-flow experiments of the analyte diffusing outside the detection region. However, for 1H NMR, it was postulated that a detection limit of 0.002 nmoles was possible after several hours data accumulation. Next he went on to describe the possibility of multiple samples being measured simultaneously with a 4 or 16 chamber detector cell. For example, this would increase throughput for small molecule studies or for protein structure determination where a series of experiments each taking 1 week is performed—this could all be done simultaneously. The main problem is cross talk between the detector coils for the different sample chambers but this has now been solved using a new generation of RF switches in conjunction with a single detector and careful consideration of coil geometry and positioning. The design shown had four cells each of 350 mm diameter by 1 mm long and of 25 nL volume, with a cross talk typically of 0.2%. This system has been tested for 1H NMR spectroscopy and also heteronuclear 1H-13C HSQC. If the switch is used to also switch the voltages for CE or CEC then it is possible to stop a peak in one cell and allow the flow to continue in the other cells, and hence do stop-flow analysis without stopping the flow. Finally, the design of RF coils was re-addressed in that it is desirable to avoid horizontal solenoidal coils and this was done by turning them vertically but with a turning pitch otherwise no signal would result. This gives about the same signal-noise as a horizontal solenoid but without the band broadening. The problem of sample diffusing outside the detector region was solved by designed a bubble cell and optimising the shape empirically.
The next speaker was Ulrich Braumann of Bruker Analytik, Rheinstetten, Germany speaking on behalf of his colleague Manfred Spraul. His talk was concerned with the full automation of HPLC-NMR-MS based on the Bruker approach. He gave a background to HPLC-NMR-MS specifically in terms of how to operate in the most efficient manner. If MS was to be used as an intelligent detector, then automation is a necessity. He discussed the relative merits of serial and parallel NMR and MS operation and made some points about instrument siting. To avoid compromised chromatography in long runs with many stops for stopped flow NMR operation, it is possible to park chromatographic peaks of interest in loops for later analysis. This type of operation was elaborated in detail. He also gave an overview of the latest software for controlling the HPLC-NMR-MS complex and showed an example of the separation of glycosides from apple peel.
The final talk of the session was by Gerhard Bringmann from the University of Wurzburg, Germany who described the use of HPLC-NMR, HPLC-MS and HPLC-CD as a triad of methods for separating and identifying natural products. He concentrated on identifying the structures of alkaloids from plants and marine sources. Many of the compounds such as isoquinoline alkaloids have pharmaceutical potential and studies are being directed towards hyphenated methods because of the small amounts of sample often available. He gave examples using stopped-flow NMR including the use of two-dimensional methods especially the ROESY technique with WET solvent suppression for obtaining inter-proton distance information and hence molecular relative stereochemistry . The absolute configuration of chiral compounds was ascertained using HPLC-CD and in cases where good model compounds of known stereochemistry are not available, he maintained that it is now possible to calculate CD spectra sufficiently precisely using semi-empirical MO methods such that absolute stereochemistry can be obtained by comparison of experimental and calculated spectra. For conformationally flexible molecules, it is necessary to average over all low energy conformers weighted by their Boltzmann populations.
After the end of the oral presentations, a period was set aside for the scientific poster presenters to discuss their work with the conference attendees and this was followed by the Conference dinner in a local restaurant.
The final session of the meeting was held on Friday 3 March and was largely devoted to studies of natural products using principally HPLC-NMR and HPLC-MS methods. The first talk was due to be presented by Professor K. Hostettmann of the University of Lausanne, Switzerland, but he was unable to attend and instedd it was given by his colleague Jean-Luc Wolfender and was concerned with the role of HPLC-NMR and HPLC-MS in the discovery of pharmacologically interesting compounds from plants. He gave the strategy that their group have evolved for this purpose involving solvent extraction and bioassay methods followed by further separation techniques. He showed some examples of how anti-fungal compounds can be detected by spraying TLC plates with fungal spores and observing regions where they fail to grow. One major activity is elimination of known compounds and an example of xanthones with monoamine-oxidase activity was shown where HPLC-MS gave characteristic molecular weights which could be used to search a database of natural products, and also fragmentation patterns for glycosides which help poinpont positions of conjugation. Another interesting example concerned three compounds found in the root bark of an African tree, Bobgunnia madagascariensis, where HPLC-NMR and HPLC-MS could not give full structural information. There was sufficient material for an x-ray crystal structure determination of the main compound, but it was clear from the NMR spectra that the other two compounds were in a dynamic equilibrium. Here stopped-flow HPLC-NMR was useful in showing that the two compounds comprised hemi-acetal and ring opened forms of the same compound. The main compound is a very potent anti-fungal agent against Candida albicans and is now in clinical trial in the USA. Further examples were given where stopped-flow two-dimensional NMR spectra such as TOCSY, NOESY and HSQC were useful in compound identification.
Next Klaus Albert from the University of Tubingen, Germany, talked about the identification of carotenoid comounds from spinach and also in bovine and chicken retinal tissue, showing how HPLC-NMR can be used for detailed characterisation of the various Z and E isomers. The compounds are both air and light sensitive and great care must be taken in the experimental approach. Concentration of the analytes was achieved using highly selective C30 solid-phase extraction phases. In addition, HPLC-MS using APCI gave an estimated 0.6 ng detection limit. He then went on to give an overview of NMR flow probe design criteria including the needs for miniaturisation and showed examples of CEC-NMR studies of beer extracts using deuterated solvents.
Bernt Schneider of the Max Planck Institute for Chemical Ecology, Jena, Germany, spoke on the use of HPLC-NMR coupled with 13C and 2H labellling studies to identify biosynthetic pathways in plants, particularly in the context of dihydrophenylpropanoids. He used root cultures grown in sterile conditions for the labelling studies and HPLC-NMR was used to identify the 13C and 2H labelled molecules and to determine the position of the labels within the molecules. With this information, metabolic routes coud be ascertained.
Next Jean-Luc Wolfender of the Univeristy of Lausanne, Switzerland, gave a second talk on the use of HPLC-NMR and HPLC-MS in a dereplication strategy for identifying known and unknown natural products from plants. He highlighted the dificulties of variable ionisation in MS and the interferences caused by solvent supression methods in NMR. He pointed out that they usually use electrospray ionisation for the MS stage as it is informative with regard to sugar fragnmentation. For example, MS/MS is used to identify positions of methoxylation and glycosidation based on fragmentation pattermns. Often, metabolites appear in related structural families and these can often be monomeric and dimeric, both symmetrically and unsymmetrically. He also pointed out the future potential use of HPLC-MS-MS libraries.
The final talk was from Ian Wilson of AstraZeneca Pharmaceuticals, Macclesfield, UK, on multiply hyphenated techniques and he subtitled his talk as "Is there more than HPLC-NMR?". He pointed out that as well as connecting the instruments together it was vital to ensure that the spectroscopists also communicated well. He showed an example of the structural identification of plant ecdysteroids using HPLC-NMR-MS and then went on to talk about additional hyphenation such as HPLC-UV-NMR-IR-MS. He showed how IR could be placed after the NMR stage and the eluent from the NMR probe could be sprayed on to a rotating germanium disk such that reflectance FT-IR spectra of individual compounds could be obtained from a model mixture of phthalates. This approach can also be used without the HPLC, simply doing flow injection UV(DAD)-IR-NMR-TOFMS. However, for an example of the separation of a mixture of nonsteroidal anti-inflammatory compounds, the HPLC stage was also required.
In his closing remarks at the end of the conference, Steen Honoré Hansen thanked all the speakers and the participants for their highly interactive approach during the meeting. He also acknowledged the staff responsible for the local organisation including both in the hotel and for the travel arrangements. Following representations from a number of participants, he expressed the hope that such a meeting could be held again in the future in about 18 months to two years time.
John Lindon
Imperial College of Science, Technology and Medicine, London
18.00-19.00 Registration
19.00-21.00 Reception / welcome party
8.30- 9.00 Registration
9.00- 9.10 Welcome address Steen Honoré Hansen
SESSION I Methodology. Chair: Ian D. Wilson
9.15-10.00 Jeremy K. Nicholson: “Analytical NMR Spectroscopy and Hyphenated NMR Techniques”
10.00- 10.45John C. Lindon: “Overview of HPLC-NMR-MS technologies”.
10.45-11.15 Coffee break
11.15-12.00 Lee Griffiths: “LC-NMR in multiple paralellel synthesis. Strategy and development”.
12.00-12.45 Udo A. Th. Brinkman: “Hyphenation - type and fascination”.
12.45-13.45 Lunch
14.00 Excursion (Including Dinner)
SESSION II Pharmaceutical Applications of LC-NMR-MS Chair: Jeremy K. Nicholson
9.00-9.45 Ulla G. Sidelmann: “Application of directly coupled HPLC-NMR within the field of drug metabolism”.
9.45-10.15 Olivia Corcoran: ”HPLC- 1H NMR studies of the chemical reactivity of carboxylate drug glucuronides”.
10.15-10.45 Rasmus W. Mortensen: ”Studies of S-naproxen glucuronide degradation kinetics by LC-NMR: Comparison with offline HPLC methods.
10.45-11.15 Coffee Break
11.15-12.00 John P. Shockcor: “The characterization of unique glutathione conjugates of the non-nucleoside HIV-1 reverse transcriptase inhibitor DPC 963 by LC-NMR-MS".
12.00-13.30 Lunch
SESSION III Advances in Techniques Chair: John C. Lindon
13.30-14.15 Ernst Bayer: “On-line coupling of electro-driven separation methods (CEC, CE) with MS and NMR”.
14.15-15.00 Andrew Webb: “Hypenation of microseparation techniques with high sensitive NMR detection”.
15.00-15.30 Coffee break
15.30-16.15 Manfred Spraul: “LC-NMR/MS under full automation with application to natural products”.
16.15-17.00 Gerhard Bringmann: “Absolute stereostructures of natural products directly from extracts: By the analytical ”Triade”, HPLC coupled to NMR, MS-MS and CD”.
17.00-18.30 POSTER SESSION
20.00 Conference dinner
SESSION IV Natural Product Applications of LC-NMR-MS Chair: Steen Honoré Hansen
9.00- 9.45 K. Hostettmann: “The role of LC-MS and LC-NMR in the discovery of lead compounds from plants”.
9.45-10.30 Klaus Albert: “Investigation of carotenoid stereoisomers in functional foods and in biological tissues by combining matrix solid phase dispersion with HPLC-MS and HPLC-NMR coupling”.
10.30-11.00 Coffee break
11.00-11.30 Bernd Schneider: “Biosynthesis of plant secondary metabolism observed by HPLC-NMR hyphenation”.
11.30-12.00 Jean-Luc Wolfender: “Potential of LC-MSn and LC-NMR for detailed on-line structural investigation of plant metabolites”.
12.00-12.45 Ian D. Wilson: “Mixture characterisation with multiply hyphenated spectroscopies: HPLC-UV(DAD)-FT-IR-NMR-MS”.
12.45 Closing remarks Steen Honoré Hansen.
13.00 Lunch
Departure (Check out of Hotel is no later than 12.00).