Enzymatic digestion with pronase and DNAase was used to isolate Kupffer cells from mouse liver. The characteristics of these cells were found to be similar to those of peritoneal macrophages, except that in the initial suspension the percentage of Kupffer cells with Fc receptors was low, C receptors were absent and the ingestion of opsenized bacteria was very poor, because of the effect of pronase on the cell membrane. After 24 h incubation in vitro all these characteristics return. The in vitro and 1 h-pulse [(3)H]thymidine labeling of the Kupffer cells is low (0.8 and 1 percent, respectively) indicating that in essence these cells do not divide. It was also shown that the small percentage of in vitro labeled Kupffer cells was recently derived from the circulation. After an intravenous injection of zymosan the in vitro labeling index of the Kupffer cells increased 16-fold, but it was proven that these dividing cells were immature mononuclear phagocytes very recently recruited from the bone marrow. The labeling of Kupffer cells aider one or four injections of [(3)H]thymidine reached a peak of 10.4 percent at 48 h or 24.1 percent at 60 h, respectively, indicating that these cells are derived from labeled monocytes. Further evidence for this conclusion was obtained by the absence of an increase of labeled Kupffer cells during treatment with hydrocortisone, which causes a monocytopenia during which no circulating monocytes are available to migrate to the tissues. Labeling studies in animals X-irradiated with hind-limb shielding gave a Kupffer cell labeling index of 5-10 percent of the normal values, which confirms their bone marrow origin. A quantitative study on the production of labeled monocytes in the bone marrow and their transit through the circulation showed that in the normal steady state at least 56.4 percent of the monocytes leaving the circulation become Kupffer cells. Considering the Kupffer cells as kinetically homogeneous this gives a mean turnover time of the total population of Kupffer cells of 21 days.

No abstract

The effect of inorganic zinc on the absorption of inorganic iron (Fe+2) from a solution was assessed in two studies on healthy male volunteers. In the first study coadministration of 344 mumol of zinc had no effect (p less than 0.5) on the absorption of 842 mumol of radiolabeled Fe, assessed by the area under plasma Fe increment time curve during the 3 (AUC3) and 6 (AUC6) h postadministration (Fe alone AUC3 = 176.4 +/- 39.3; AUC6 = 387 +/- 101; Fe + Zn AUC3 = 180 +/- 33.1; AUC6 = 396 +/- 73.1 mumol.h-1.L-1), total plasma content of 59Fe, and whole-body retention of 59Fe. In the second study only the plasma appearance of Fe was monitored. After administration of 421 mumol of Fe alone, the AUC3 and AUC6 were 167 +/- 21.2 and 429.4 +/- 57 mumol.h-1.L-1, respectively; these were reduced to 56.4 +/- 17 and 119 +/- 34 (p less than 0.002) by 421 mumol Zn and further reduced by 1048 mumol Zn to 33 +/- 15 and 43.4 +/- 23.8 mumol.h-1.L-1 (p less than 0.001), respectively. It is concluded that Zn can impair the intestinal absorption of Fe.

The turnover of a radiolabeled (65Zn) pool of endogenous zinc was monitored by using a whole-body counter in eight patients with celiac disease (CD) and analyzed by using a two-compartment model. The biological half-life of the first compartment (1-3 wk postadministration) was similar in healthy volunteers (122 +/- 34 d, means +/- SD) and untreated patients (97 +/- 21 d). The second compartment in the patients (3-12 wk postadministration) was shorter (159 +/- 22.5 d. p less than 0.001) than were reference values (218 +/- 27 d) but increased (291 +/- 71 d) after the patients started gluten-free diets. The percentage absorption of 65Zn (9.25 kBq) from a test meal containing 31 mumol (2 mg) zinc was similar in untreated patients (30.0 +/- 13%) and healthy volunteers (32.5 +/- 12.4%). These data show that in mild untreated CD increased turnover and loss of endogenous zinc occurs whereas the absorption of zinc from a customary zinc intake may be normal. The pathophysiological basis of this loss was not investigated.

SUMMARY.Random serum transferrin saturation (TS) was measured in 1194 patients attending a diabetic clinic. Twenty-one patients had TS> 55070 and in three of these patients repeat random TS was < 55070. Seventeen patients were recalled for fasting serum TS and ferritin measurement. Ten patients had fasting TS> 55%. The diagnosis of haemochromatosis was confirmed by liver biopsy in a total of six patients, three of whom were previously unsuspected. Haemochromatosis was the possible diagnosis in a further four patients. Family studies using HLA typing confirmed haemochromatosis in four family members, three of whom were asymptomatic. We conclude that measurement of TS is a simple and effective method of finding cases of haemochromatosis in the diabetic clinic. Additional key phrases: transferrin saturation; ferritin; liver iron content; serum iron; HLAHereditary haemochromatosis (HH) is an autosomal recessive disorder of iron metabolism." The defective gene lies close to the histocompatibility locus antigen (HLA) on the short arm of chromosome 6.2 Untreated, homozygotes for the disorder absorb excessive amounts of iron which is deposited in various organs and can eventually lead to the clinical signs and symptoms of the disease such as heart failure, cirrhosis, arthritis, pigmentation and diabetes. ' The prevalence of HH may be as high as 4· 5 per 1000,4 one of the commonest autosomal recessive disorders. Not all homozygotes, however, will absorb sufficient iron to become symptomatic. Clinical haemochromatosis develops five times more frequently in men than in women, even though the male: female ratio for homozygosity is the same. It is likely that most cases of haemochromatosis are not diagnosed during life. 5Although HH fulfils the classic criteria of a medical condition suitable for screening, no screening programme exists in the UK and there may be formidable obstacles to the establishment of such a programme in the healthy population."Diabetes is a condition classically associated with HH. In addition to the effect of iron deposition in the pancreatic islet beta cells there may also be a genetic link between the two Correspondence; Dr R M Evans, Biochemistry Department.conditions which would explain the frequent occurrence of diabetes in the siblings of patients with haemochromatosis and diabetes." The diabetes in haemochromatotic patients is similar to other patients with type II diabetes, with normal or exaggerated glucagon response but impaired insulin response to glucose." Depending on the population studied, between 20% and 80% of cases of HH have diabetes rnellitus.P-'? and therefore screening the diabetic population is likely to result in finding new cases of HH. Subsequent investigation of family members of these cases is likely to be successful, because these individuals may be motivated to participate since they have been exposed to morbidity due to the condition in one of their relatives. However, there have been conflicting reports regarding the prevalence of haemochromatosis in diabetics with r...