WHY DISCARD THE PERITONEAL MACROPHAGES OF PATIENTS ON CAPD?

It is difficult to obtain enough human macrophages for extensive in vitro study, hence our relatively poor understanding of the mechanisms whereby human, as opposed to murine, macrophages kill tumours. Yet millions of macrophages are discarded daily in the peritoneal effluents of patients on continuous ambulatory peritoneal dialysis (CAPD) (Maddox et al, 1984). If these were cryopreserved, they would be available for study. In order to evaluate this proposition we examined peritoneal dialysates from 17 patients on CAPD and found that seven contained > 3 × 106 macrophages. Cells were harvested from four patients over a short period of time and cryopreserved over liquid nitrogen. The causes of their renal failure were hypertension (n = 2; patients A, D) glomerulonephritis (B) and sarcoidosis (C); one patient had familial monocyte esterase deficiency (D), his peritoneal macrophages were similarly esterase deficient (Hall et al, 1998). On the day of examination all samples from a patient were thawed and pooled. The viability of the pooled cells varied from 83% to 90% and examination of cytospins demonstrated that the suspensions contained 57–61% macrophages, 30–41% lymphocytes, 1–6% mesothelial cells, and <1–3% polymorphs. Further purification of macrophages was unnecessary for the assays chosen for this study. Respiratory burst activity (RBA) was examined on the day of thawing in the presence or absence of vitamin D3 (2 × 10−8 m), zinc chloride (1 mg/ml) or both. Reactive oxygen species (ROS) produced from aliquots of 1 × 105 macrophages/ml were measured using the ABEL cell activation test kit with pholasin(K) (Knight Scientific Ltd) and a Bio-Orbit 1251 luminometer, after stimulation with phorbol myristate acetate (PMA) 0.8 μm final concentration. The results shown in Fig 1 demonstrate the retention of RBA. In two patients the peak emission value equated to that described for peripheral blood neutrophils (Knight & McCafferey, 1997). The mean lag time of 9.4 (±2.1) seconds from PMA addition to start of light emission was shorter than that (15–20 s) described for peripheral blood neutrophils (Knight & McCafferey, 1997). Zinc appeared to enhance ROS production. . Activation of cryopreserved CAPD-M by PMA in the presence of either vitamin D3 (D3) ZnCl or both or in the absence of stimulants (MAC) is shown for four patients. ROS was measured by the mV of light released from Pholasin(K)/second. The control(s) were assay buffer ± ZnCl or vitamin D3. Tumour necrosis factor α (TNF-α) and nitrite levels were measured in duplicate in supernatants of thawed and washed cells (4 × 105 macrophages in each) cultured for 24 h with bacterial lipopolysaccharide 10 μg/ml, interferon-γ 10 μg/ml, phytohaemagglutinin (PHA) 50 μg/ml, vitamin D3 2 × 10−8/ml, or polyribonucleotide polyinosinic-polycytidilic acid (Poly IC) 50 μg/ml in the presence or absence of zinc chloride and without stimulants, using a Quantikine HS human TNF-α immunoassay kit (R&D Systems Europe Ltd, U.K.) and Griess reagent respectively. The range of TNF-α produced was 1.9–7 μg/106 macrophages; these levels were similar to those produced by fresh CAPD macrophages (Kimmer et al, 1996) and peritoneal macrophages from normal peritoneal cavities, women with endometriosis or from ascitic fluid of patients with portal hypertension (Rana et al, 1996); PHA was the only stimulant causing a significant increase overall above unstimulated cells (P < 0.05 Mann-Whitney U test). Supernatant nitrite levels ranged from <225 nm to 4.2 μm/106 macrophages; poly IC significantly increased nitrite production both in the presence and absence of zinc (P < 0.05 Mann-Whitney U test); zinc alone did not significantly increase nitrite production. Poly IC has previously been shown to be a potent inducer of nitrite production (Snell et al, 1997). Therefore large numbers of functionally intact macrophages may be obtained for extensive investigation of tumouricidal mechanisms. Also, the best methods for up-regulation of monocyte/macrophage tumouricidal activity could be determined and applied to immunotherapy; the effectiveness of candidate drugs in the enhancement of antitumour activity against human tumour cell lines could be assessed; and the effect of monocyte/macrophage carboxylesterase on the activation of pro-drugs measured, e.g. irinotecan is converted to a form 39 times more active by carboxylesterase and the variable therapeutic dose might, it has been suggested, be due to the variable carboxylesterase activity of subjects (Kojima et al, 1998). This hypothesis could be tested using CAPD macrophages of varying carboxylesterase activity. From patients with 3 × 106 macrophages/effluent, 60–80 × 106 macrophages can be available weekly. We suggest that this rich source of human macrophages should be used.