We have compared the levels of inositol metabolites in three pairs of normal and transformed cells which have been matched with respect to their cell lineage, differentiation and proliferation status: (i) normal human myeloid blast cells and the human promyelocytic leukaemic cell line, HL60; (ii) human umbilical-cord T-helper cells and C8166 cells, a HTLV-1-transformed T-helper cell line; and (iii) an interleukin 3-dependent long-term culture of murine pro-B-cells (BAF3) and BAF3 cells transformed by transfection with the bcr-abl oncogene. Complex patterns of inositol metabolites were present in each of the cell populations. Although there were a number of differences in the levels of certain inositol metabolites between individual cell populations in the paired groups, we did not observe any consistent difference in the levels of inositol metabolites between the proliferating normal and transformed cells. In particular, our data do not support the reported correlation between elevated glycerophosphoinositol (GroPIns) levels and transformation of cells by membrane and cytoplasmic oncogenes which has been reported by other workers. All the cells contained high concentrations of Ins(1,3,4,5,6)P5 (between 12 and 55 microM) and InsP6 (between 37 and 105 microM). The HTLV1-transformed T-helper cells had particularly high levels of total inositol phosphates (predominantly GroPIns, an unidentified inositol bisphosphate and InsP6). The observations are discussed with reference to cell transformation and to the differentiation status of the paired populations.

1. HL60 promyeloid cells contain high intracellular concentrations of inositol polyphosphates, notably inositol 1,3,4,5,6-pentakisphosphate (InsP5) and inositol hexakisphosphate (InsP6). To determine their intracellular location(s), we studied the release of inositol (poly)phosphates, of ATP, and of cytosolic and granule-enclosed enzymes from cells permeabilized by four different methods. 2. When cells were treated with digitonin, all of the inositol phosphates were released in parallel with the cytosolic constituents. Most of the InsP5 and InsP6 was released before significant permeabilization of azurophil granules. 3. Similar results were obtained from cells preloaded with ethylene glycol and permeabilized by osmotic lysis. 4. Electroporation at approximately 500 V/cm caused rapid release of free inositol. Higher field strengths provoked release of most of the ATP, InsP5 and InsP6, but only slight release of the intracellular enzymes. Multiple discharges released approximately 80-90% of total InsP5 and InsP6. In the absence of bivalent-cation chelators, InsP5 and InsP6 were released less readily than ATP. 5. Treatment of cells with Staphylococcus aureus alpha-toxin caused quantitative release of inositol and ATP, without release of intracellular enzymes. However, inositol phosphates were released much less readily than inositol or ATP. Even after prolonged incubation with a high concentration of alpha-toxin, only approximately 50-70% of InsP2, InsP3 and InsP4 and< or = 20% of InsP5 and InsP6 were released, indicating that the high charge or large hydrated radius of InsP5 and InsP6 might limit their release through small toxin-induced pores. 6. These results indicate that most intracellular inositol metabolites are either in, or in rapid exchange with, the cytosolic compartment of HL60 cells. However, they leave open the possibility that a small proportion of cellular InsP5 and InsP6 (< or = 10-20%) might be in some intracellular bound form.

HL60 cells are human promyeloid cells that can be induced to differentiate by physiological stimuli (e.g. all-trans retinoic acid (ATRA), 1 alpha,25-dihydroxyvitamin D3 (D3), granulocyte colony-stimulating factor (G-CSF)) and by non-physiological agents such as dimethysulphoxide (DMSO) and protein kinase C-activating phorbol esters. The sensitivity of HL60 cells to physiological differentiating agents, but not to DMSO, is enhanced when cells are exposed to 'anti-inflammatory agents' (e.g. indomethacin) or are 'primed' (pretreated) with a small amount of ATRA: alone, neither treatment induces differentiation. We earlier suggested that indomethacin might act by inhibiting the endogenous formation of a differentiation-suppressing prostanoid (Bunce, C.M., et al. (1994) Leukemia 8, 595-604). Studies of the formation of prostanoids by HL60 cells and of the effects of prostanoids on these cells failed to identify any prostanoid that could be implicated in sensitization by indomethacin. 3 alpha-Hydroxysteroid dehydrogenase (3 alpha-HSD) is another target of such 'anti-inflammatory agents'. Steroid inhibitors of 3 alpha-HSD sensitized HL60 cells to inducers of differentiation in a manner similar to indomethacin. 3 alpha-HSD is a member of the aldoketoreductase enzyme family, which comprises many enzymes of similar size and primary sequence. A protein that was recognised by an antiserum to 3 alpha-HSD was found in HL60 cells, but the cells showed no detectable 3 alpha-HSD activity. The 3 alpha-HSD-like protein was strikingly down-regulated by 'priming' doses of ATRA. When treatment with a differentiation-sensitizing 'anti-inflammatory agent' or steroid was combined with ATRA "priming', the effects of the different treatments were not additive: the resulting increase in sensitivity equalled that achievable by either treatment alone. We conclude that interference with a single intracellular regulatory mechanism underlies the increases in sensitivity of cells to differentiating agents that are caused by anti-inflammatory agents, by certain steroids and by 'priming' with ATRA. Decreased activity of a yet-to-be-identified member of the aldoketoreductase family of dehydrogenases is likely to be a central feature of a previously unrecognised mechanism that controls the responsiveness of cells to environmental stimuli such as retinoids and D3.

1. An inositol trisphosphate (InsP3) distinct from Ins(1,4,5)P3 and Ins(1,3,4)P3, which we previously observed in myeloid and lymphoid cells [French, Bunce, Stephens, Lord, McConnell, Brown, Creba and Michell (1991) Proc R. Soc. London B 245, 193-201; Bunce, French, Allen, Mountford, Moore, Greaves, Michell and Brown (1993) Biochem. J. 289, 667-673], is present in WRK1 rat mammary tumour cells and pancreatic endocrine beta-cells. 2. It has been identified as Ins(1,2,3)P3 by a combination of oxidation to ribitol, a structurally diagnostic polyol, and ammoniacal hydrolysis to identified inositol monophosphates. 3. Ins(1,2,3)P3 concentration in HL60 cells changed little during stimulation by ATP or fMetLeuPhe or during neutrophilic or monocytic differentiation, and Ins(1,2,3)P3 was unresponsive to vasopressin in WRK1 cells. 4. Ins(1,2,3)P3 was usually more abundant than Ins(1,4,5)P3, often being present at concentrations between approximately 1 microM and approximately 10 microM. 5. HL60, WRK-1 and lymphoid cells also contain Ins(1,2)P2 or Ins(2,3)P2, or a mixture of these two enantiomers, as a major InsP2 species. 6. Ins(1,2,3)P3 and Ins(1,2)P2/Ins(2,3)P2 are readily detected in cells labelled for long periods, but not in acutely labelled cells. This behaviour resembles that of InsP6, the most abundant cellular inositol polyphosphate that includes the 1,2,3-trisphosphate motif, which also achieves isotopic equilibrium with inositol only slowly. 7. Ins(1,2,3)P3 is the major InsP3 that accumulates during metabolism of InsP6 by WRK-1 cell homogenates. 8. Possible metabolic relationships between Ins(1,2,3)P3, Ins(1,2)P2/Ins(2,3)P2 and other inositol polyphosphates in cells, and a possible role for Ins(1,2,3)P3 in cellular iron handling, are considered.

No abstract