Precise measurement of pHe in vivo may be of clinical value for both diagnosis and selection of therapy. pHe measurements made by the 31P probe 3‐aminopropylphosphonate (3‐APP) were compared with those made by the 19F probe, 3‐[N‐(4‐fluor‐2‐trifluoromethylphenyl)‐sulphamoyl]‐propionic acid (ZK‐150471) in three solid tumour types, human HT29 xenografts, murine RIF‐1 fibrosarcomas and Lettre tumours grown subcutaneously in mice. No significant differences were observed when probe measurements of pHe were compared at 20–60 min post‐administration, although very low pHe values (ca. 6.0) were recorded in two out of eight Lettre tumours by ZK‐150471. The more rapid pHe measurements possible using ZK‐150471 showed that during the first 20 min post‐administration significant increases occurred in pHe which were greatest in the more necrotic tumours. Since isolated cell experiments showed that ZK‐150471 was non‐toxic and did not enter the cells, this early increase in pHe may reflect gradual penetration by ZK‐150471 of the reportedly alkaline necrotic space in the tumours. The wide chemical shift range, improved signal‐to‐noise and absence of signal overlap allowed a more rapid and precise measurement of pHe by ZK‐150471 compared to 3‐APP. These characteristics suggest that ZK‐150471 is currently the preferred pHe probe for non‐invasive MRS. Copyright © 1999 John Wiley & Sons, Ltd.

Summary Quantification of metabolite or drug concentrations in living tissues requires determination of intra-and extracellular volumes. This study demonstrates how this can be achieved non-invasively by 31p magnetic resonance spectroscopy (MRS) employing dimethyl methylphosphonate (DMMP) as a marker of total water space, 3-aminopropylphosphonate (3-APP) as a marker of extracellular space and P, and 3-APP as markers of intracellular pH (pH,) and extracellular pH (pH) respectively. The MRS measurements of the tumour volumes were validated by classic radiolabelling methods using 3H20 and ['4C]inulin as markers of total and extracellular space respectively. The extracellular volume fraction measured by radiolabelling of RIF-1 tumours was 23 ± 0.83% (mean ± s.e.m. n = 9), not significantly different (P > 0.1) from that found by MRS (27 ± 2.9%, n = 9, London, and 35 + 6.7, n = 14, Baltimore). In untreated RIF-1 tumours, pHl was about 0.2 units higher than PHe (P < 0.01). 5-Fluorouracil (5FU) treatment (165 mg kg') caused no significant changes in either PHe or per cent extracellular volume. However significant increases in pH1 48 h after treatment (P < 0.01) correlated with decreased tumour size and improved bioenergetic status [NTPfinorganic phosphate (P.) ratio]. This study shows the feasibility of an MR method (verified by a gold standard') for studying the effects of drug treatment on intra-and extracellular spaces and pH in solid tumours in vivo. In a 'H-MRS study of RIF-1 tumours responding to 5-fluorouracil (5FU) therapy. a decrease in trimethylamine and lactate signals was detected (Shungu et al. 1992b). However. Whether these decreases resulted from a decrease in the intracellular quantities of these metabolites or from a chance in the intracellular volume fraction is not known. and information on volume fraction would allow us to distinguish between these two possibilities. In addition. distinction between intra-and extracellular compartments is particularly important in pH measurements of normal and tumour tissue. The hydrogen ion (H+) concentration of the intraand extracellular milieu can influence drug uptake or repair of cellular damage (Hult and Larson. 1976: Hofer and Mivechi. 1980: Nissen and Tanneberger. 1981. As 3-APP also serves as an indicator of extracellular pH (pH) (Gillies et al. 1994) and inorganic phosphate (P) is predominantly in the intracellular compartment and is. hence. an endogenous indicator of intracellular pH (pH) under most conditions (Stubbs et al. 1992). we have also monitored changes in pH and pHe of RIF-1 tumours following treatment with 5FIT. Several agents such as 5FUT or radiation are known to induce changes in pH following treatment (Tozer et al. 1989: Li et al. 1991. However. the contribution of the intra-and extracellular compartments to these chances has not been adequately defined. USA. 606Intra-and extracellular volume and pH of tumours after 5FU 607 (1980). Tumours were between 300 and 1000 mm3 when used. Mice were anaesthetized with a combination of ketamine (50...

Summary Apoptosis was induced by treating L1210 leukaemia cells with mechlorethamine, and SW620 colorectal cells with doxorubicin. The onset and progression of apoptosis were monitored by assessing caspase activation, mitochondrial transmembrane potential, phosphatidylserine externalization, DNA fragmentation and cell morphology. In parallel, 31 P magnetic resonance (MR) spectra of cell extracts were recorded. In L1210 cells, caspase activation was detected at 4 h. By 3 h, the MR spectra showed a steady decrease in NTP and NAD, and a significant build-up of fructose 1,6-bisphosphate (F-1,6-P) dihydroxyacetonephosphate and glycerol-3-phosphate, indicating modulation of glycolysis. Treatment with iodoacetate also induced a build-up of F-1,6-P, while preincubation with two poly(ADP-ribose) polymerase inhibitors, 3-aminobenzamide and nicotinamide, prevented the drop in NAD and the build-up of glycolytic intermediates. This suggested that our results were due to inhibition of glyceraldehyde-3-phosphate dehydrogenase, possibly as a consequence of NAD depletion following poly(ADP-ribose) polymerase activation. Doxorubicin treatment of the adherent SW620 cells caused cells committed to apoptosis to detach. F-1,6-P was observed in detached cells, but not in treated cells that remained attached. This indicated that our observations were not cell line-or treatment-specific, but were correlated with the appearance of apoptotic cells following drug treatment. The 31 P MR spectrum of tumours responding to chemotherapy could be modulated by similar effects.