Printed in the United States of America MONOCYTE SECRETION OF fl-HEXOSAMINIDASE IN PATIENTS WITH OBSTRUCTIVE JAUNDICE

Monocyte hydrolases are harmful when secreted inappropriately. In this study we have investigated the levels of one of the hydrolases, β-hexosaminidase in patients with obstructive jaundice. These patients showed markedly elevated plasma levels, and their monocytes show increased spontaneous secretion and total enzyme content. The plasma enzyme levels correlate with monocyte enzyme content as well as bile salt, and bilirubin levels, the high levels may also reflect Kupffer cell damage, as these cells clear the enzyme. Compared with controls monocytes from jaundiced patients show reduced enzyme secretion after PMA stimulation, in vitro, and unchanged secretion after zymozan stimulation. There is a difference between plasma enzyme levels in benign and malignant patients but this does not provide a clear distinction between the two groups. We conclude that patients with obstructive jaundice have increased blood level of β-hexosaminidase, and that activated monocytes partly contribute to this change.


INTRODUCTION
fl-hexosaminidase (EC 3.2.1 52) is one of the lysosomal hydrolases that is critically deficient in Tay-Sachs and Sandoff syndrome'2. This enzyme has been found to be increased in several other diseases: e.g. diabetes mellitus3, sepsis4, enterocolitis5, and some malignancies6'7'8'9. Hultberg 1 reported in 1981 that this enzyme was also increased in the serum from obstructive jaundice patients and together with other 11 researchers suggested this might be due to the decreased clearance by Kupffer cells in these disease conditions.
Monocytes are an important component of the reticuloendothelial system. This cell population is known to secrete many substances including proteins, enzymes, cytokines, lipid metabolites and other immune regulators. It has been shown that monocytes/macrophages secrete lysosomal enzymes spontaneously 12 and in response to stimuli13'14. Previous work has shown that patients with obstructive jaundice have altered immune responses15'6'17A8. The monocyte function however in jaundice has not yet been fully studied. The relationship between monocyte secretion of lysosomal enzymes and plasma fl-hexosaminidase level in obstructive jaundice is poorly understood.
In this study we have investigated the monocyte fl-hexosaminidase secretion under stimulated and non-stimulated conditions and its relation with plasma flhexosaminidase and bile salts levels.

Patients and Controls
Fifty eight patients with obstructive jaundice were included in the study. Twenty nine had a cholangiocarcinoma, 7 had pancreatic carcinoma and 1 gallbladder cancer, the remaining 21 had benign disease which was due to gallstones in 19 and a benign biliary stricture in two. Forty-two controls were either non-jaundiced patients admitted for elective surgery (n 32) or normal volunteers (n 10). The patients in the control group comprised 20 with benign diseases (14 with gallstones, 5 with inguinal hernias and 1 with haemorrhoids) and 13 with malignant disease (colon cancer with or without liver metastases). Patients with sepsis were excluded. The median age for the jaundiced patients was 64.7y (IQR 57.3-74.8) and for the controls 53.8y (IQR 42. 5-70.5). Informed consent to participate in the study was obtained from all the patients. Blood was taken immediately after admission, before any procedures were carried out.

Cell Preparation
All materials were purchased from Sigma unless otherwise stated. 20mls peripheral blood were taken into a Sterilin bottle (Sterilin, England) containing 200U heparin. Blood was put immediately onto ice and processed, at 4C, as soon as possible (within 60mins). Plasma was obtained after centrifugation of blood at 1500 rpm for 20 minutes it was then stored at -80C until needed. Mononuclear cells (MNC) were obtained by Ficoll-Hypaque separation. A small portion of the resulting cell suspension was used for non-specific esterase staining to estimate the monocyte proportion in the MNC preparation, this was: 29% (IQR 25%-36%) in jaundice and 27% (25%-31%) in controls. MNC's at 2 106 were then added to a multiwell plate (NUNC, Denmark) at 100/A/well and allowed to adhere to the plastic surface at 37C in 5 % CO2 for 60 minutes. Non adherent cells were then washed off and the monocytes were re-suspended in RPMI 1640 with 2.0% bovine serum albumin (BSA, fraction V, Boehringer Mannheim GmbH, Germany). Serum was avoided in cell culture because it contains a high concentration of fl-hexosaminidase.

Cell Stimulation
Aliquots of cells in triplicate were then treated with (i) culture medium only for estimation of spontaneous secretion, with (ii) phorbol myristate acetate (PMA) at a concentration 5.0/g/ml, or with (iii) opsonised zymozan particles at 0.5 mg/ml made up as follows; 10mg zymozan was heat inactivated at 100C for 10 minutes and washed, 2.0 ml pooled human serum was added to the particles and incubated at 37C for 30 minutes after washing 10 times at 2000 rpm for 10 minutes the particles were re-suspended in culture medium and stored at -80C. Some cells were also treated by 10% Triton X100 to cause the release of the total enzyme content of the monocytes. After 4 hours culture with either medium, one of the stimulants or with Triton the plates were centrifuged at 1000 rpm for 20 minutes. Supernatants were harvested with a multichannel pipette and stored at -80C before analysis. Cell

Zymozan stimulated secretion
Stimulation with opsonised zymozan particles caused control monocytes to secrete more lysosomal enzyme and this is consistent with the work of others 13,14 confirming that opsonised zymozan particles are the most potent stimulator of

Monocyte total enzyme contents
As measured in the supernatant from Triton lysed monocytes the total enzyme content from jaundiced patients was 0.46 (0.32-0.56) U/million cells which was higher than controls 0.28 (0.18-0.36) U/million cells with p<0.01. Analysis of subgroups of patients showed a significant difference between the benign and malignant jaundiced patients, but not between other groups, this data is shown in Table 4.

fl-Hexosaminidase
The plasma enzyme levels were significantly increased in obstructive jaundice patients, 39.7 (28.4-54.5) U/I compared with controls, 13.4 (9.5-19.7) U/I. (p<0.001). The differences between benign controls (12.2(9.5-16.7) U/l) and benign jaundiced (31.8(28.2-47.6) U/l), and also between malignant controls (18.9(3.7-25.5) U/I) and malignant jaundiced (46.7(30.4-63.5) U/l) all reached statistical significance. There was also a difference between the benign and malignant jaundiced patients, but no difference between control groups was seen. Plasma biochemistry Plasma biochemistry from patients and controls is summarised ,in Table 5. The controls all had values within the normal range for this institution.
The relationships between the monocyte lysosomal secretions, plasma enzyme levels, bile salts and biochemistry in jaundice patients As shown in Table 6, the increased monocyte total enzyme contents and decreased Table 5 Plasma  Table 6 The correlation coefficients between monocyte hexosaminidase activity and liver function response to PMA stimulation have a close relationship with the degree of jaundice (correlation indes r 0.39, p 0.01 and r 0.47, p < 0.01 respectively).
Plasma fl-hexosaminidase level war also positively correlated with monocyte total enzyme contents (r=0.31), bile sale levels (r=0.44), alkaline phosphatase (r=0.35) and bilirubin (r=0.35) DISCUSSION fl-hexosaminidase is one of the monocyte lysosomal hydrolases. This enzyme is cleared from the circulation by Kupffer cells in the liver2'21, by the kidney 22 and some other tissues. In blood however, this enzyme exists as its precursor form and the mode of clearance of the precursor is not clear23. Hultberg in 1981 reported that the concentration of this enzyme increased in the blood of patients with obstructive jaundice. This enzyme has also been found to be increased in animals after hepatectomy24, or with sepsis4. These studies suggested that the increases were due to damaged Kupffer cell clearance rather than increased production. The correlations between plasma fl-hexoaminidase level and bile salt levels has also been previously reported in pati,,,; with intrahepatic cholestasis11. We confirm from our study that the blood levels of fl-hexosaminidase are elevated in both benign and malignant jaundiced patients. Our data also shows that plasma/3hexosaminidase is positively correlated with bile salt and bilirubin levels. The plasma fl-hexosaminidase level has been reported to be a marker to distinguish between patients with benign or malignant jaundice25. Our data only partly supports this conclusion as there is a considerable overlap between the two groups.
Jaundiced monocytes have an increased total enzyme content ( Table 4) and this is correlated with plasma/%hexosaminidase level suggesting that monocytes contribute in part to the increased plasma enzyme level in addition to the depressed Kupffer cell clearance of the enzyme. Therefore the increased plasma /3hexosaminidase is not only a measure for the function of the fixed cells of the reticuloendothelial system 26'27 but also partly reflects peripheral monocyte activation. The change in monocyte total enzyme content in jaundiced patients is different from that reported by Holdstock et al. 28, where the enzyme levels were found to be reduced in cirrhotic patients. The changes seen here are similar to those in the small number of jaundiced patients in Holdstock's report.
Monocytes secrete some lysosomal enzyme spontaneously (Table 1) and increased amounts in response to stimulation. Indeed we have shown that PMA ( Table 2) and opsonised zymozan particles (Table 3) can stimulate /3hexosaminidase secretion from control monocytes however PMA has an inhibitory effect on release of enzyme from jaundiced cells (that is, a lower proportion of the totaly enzyme content of the cell is released; although the content is higher in jaundice). In the case of zymozan stimulation jaundice and control cells release the same proportions of their respective totals. It is known that zymozan has its effects on monocyte lysosomal secretion via mannose-like receptors on the cells. Our data may suggest that the capacity for phagocytosis of these particles remains intact in obstructive jaundice. Indeed previous work in this laboratory has shown that monocytes from jaundiced patients phagocytose particles (bacteria and latex beads) normally (Puntis, M.C.A. data presented at the 2nd World Congress of HPB Surgery 1988). These results are different from some other reported investigations 29 which may be due to the material used in the study. Phorbol myristate acetate is a kinase C activator and its effects involve a cell surface receptor and intracellular mechanisms3. The reason for the inhibitory effect on the release of/%hexosaminidase from jaundiced monocytes in response to this stimulus is not clear at present. Monocytes from jaundiced patients also show increased spontaneous secretion of/%hexosaminidase. Two possible explanations for this are: (i) membrane fatty acids may be modified, as it has been shown that in jaundiced patients essential fatty acids are deficient31. The membrane fatty acids are an important component of several cell functions involving the cell surface and intracellular mechanisms32. We have previously reported that in vitro supplementation with n-6 fatty acids may inhibit the spontaneous secretion of this enzyme in obstructive jaundice33. Essential fatty acids and their metabolites are involved in the immune response 34 and therefore may play an important role in the altered monocyte function that we report here in jaundiced patients. (ii) Monocyte secretion of cytokines is modified in jaundiced patients. We have,already reported that tumour necrosis factor (TNF) and interleukin-1 can stimulate monocyte lysosomal enzyme secretion 35 and TNF release by monocytes is increased in jaundiced patients36.
The effects in vivo of the altered lysosomal enzyme release in jaundice are not yet clear. It is known that monocyte lysosomal enzyme may cause tissue damage in amoebiasis of the gut 37 and inflammatory bowel disease3, also fl-hexosaminidase from other sources such as mucosa can cause lesions in other tissues39. We suggest that the increased spontaneous secretion in jaundice may also be harmful to patients. In the presence of infection, for example, monocyte/macrophages may fail to respond adequately and this again may well be harmful to patients.