phosphatase enzyme of pancreatic islets, Waddell &. Burchell [1] ... latter value is similar to that reported by Waddell & .... Ian D. WADDELL and Ann BURCHELL.
Biochem. J. (1990) 266, 619-624 (Printed in Great Britain)
619
TTERS BlCEHCLE n LL Ij J&U-i I
Rat liver glucose-6-phosphatase: a corrigendum In their article on the microsomal glucose-6phosphatase enzyme of pancreatic islets, Waddell & Burchell [1] report maximal velocities of 0.18 and 0.52,amol/min per mg of liver microsomal proteins in intact and disrupted hepatic microsomes, respectively. They compare these values with 0.00275,tmol/min per mg of protein, as allegedly reported by Giroix et al. [2] for rat liver glucose-6-phosphatase activity, and conclude that the procedure used by the latter authors resulted in the inactivation of liver microsomal glucose-6-phosphatase. In doing so, the former authors overlook the two following facts. First, the measurement reported by Giroix et al. was obtained in the presence of 2.0 mM-Dglucose 6-phosphate; hence, allowance should be made for the Km of 2.4 mm reported by the same authors in order to calculate the maximal velocity. Second, the results reported by Giroix et al. were expressed on a liver wet weight basis (see Table 2 in [2]) and not per mg of microsomal protein. If it is assumed that microsomal proteins only represent approx. 15 ,ug per mg of liver wet weight [3], and if allowance is indeed made for the Km, it is easy to calculate that the value reported by Giroix et al. would correspond to a maximal velocity close to 0.4 ,tmol/min per mg of liver microsomal protein. The latter value is similar to that reported by Waddell & Burchell.
Willy J. MALAISSE Laboratory of Experimental Medicine, Brussels Free University, Boulevard de Waterloo 115, B-1000 Brussels, Belgium 1. Waddell, I. D. & Burchell, A. (1988) Biochem. J. 255, 471-476 2. Giroix, M.-H., Sener, A. & Malaisse, W. J. (1987) Mol. Cell. Endocrinol. 49, 219-225 3. Ramirez, R., Zahner, D., Marynissen, G., Sener, A. & Malaisse, W. J. (1989) Biochem. J. 261, 509-513 Received 12 September 1989
Glucose-6-phosphatase activity in fed rat liver In Waddell & Burchell (1988) we reported that there high levels of microsomal glucose-6-phosphatase activity in rat pancreatic islets (1.5 + 0.5,tmol/min per mg of microsomal protein). In complete contrast, while our paper was in preparation Giroix et al. (1987) reported finding only extremely low levels of glucose-6phosphatase activity in pancreatic islets and they conare very
Vol. 266
cluded "that rat islet cells are virtually devoid of true glucose-6-phosphatase activity". In the discussion of our paper we suggested that the most likely reason Giroix et al. (1987) had found no glucose-6-phosphatase activity in islets was that it had been inactivated. There were three reasons why we made this suggestion: 1. we had found that the glucose-6-phosphatase activity in pancreatic islets was very unstable; 2. several steps used by Giroix et al. (1987) cause inactivation of glucose-6-phosphatase; 3. the control liver glucose-6-phosphatase activities that they reported were low and variable. In the preceding letter, Malaisse (1990) states that the values of fed rat liver glucose-6-phosphatase reported in Giroix et al. (1987) are not low but are very similar to those reported in Waddell & Burchell (1988); we cannot agree with this statement. Giroix et al. (1987) did not subtract the non-specific phosphatase activity from the values they report or measure microsomal intactness, making the data somewhat difficult to interpret. In addition, in Giroix et al. (1987), activities of the glucose6-phosphatase enzyme are expressed in a variety of different units, i.e. pmol/min per islet and pmol/min per jug of protein (introduction), pmol/min per islet and pmol/min per jug wet wt. (Fig. 1), pmol/min per ,tg of protein (Table 1), pmol/min per 103 cells, fmol/min per islet, fmol/min per ,ug wet weight and pmol/min per ,tg of microsomal protein (results), and this has led to some confusion. We then added to the confusion because in the discussion of Waddell & Burchell (1988) we converted one of the values given by Giroix et al. (1987) into ,umol/min per mg of protein, which are the only units that we have ever used in our papers. In his letter, Malaisse assumes that we had converted to ,amol/min per mg of microsomal protein. This we could not do, as no data were given in Giroix et al. (1987) to indicate their yields of microsomal protein, etc. We were simply trying to indicate that the rat liver homogenate values in Giroix et al. (1987) were low compared to our values. However with hindsight, we did not explain the units clearly. Nor did we demonstrate how low and different the results in Giroix et al. (1987) are, not only compared to our data, but also to glucose-6-phosphatase values in the literature. We apologise for these omissions and we therefore welcome this opportunity to clear up the misunder-
standing. In his letter, Malaisse (1990) uses the activity of fed rat liver homogenate glucose-6-phosphatase given in pmol/ min per ,tg wet wt. of liver in Table 2 of Giroix et al. (1987) as the basis of his calculations. For simplicity in Table 1, we have not discussed how the calculations should or should not have been done but we have directly compared the value in Table 2 of Giroix et al. (1987) with the activity found routinely in this laboratory
BJ Letters
620 Table 1. Glucose-6-phosphatase enzyme activity in fed rat liver homogenates
Glucose-6-phosphatase activity (amol/min per g wet wt. of liver)
Source of data
Giroix et al. (1987) Our data (six separate preparations) Table V of Alvares & Nordlie 18.3 (1977) Table IV of Nordlie et al. 13.4+0.8 (1982) Table IX of Arion et al. 12.6+0.8 (1983) Table 1 of Nordlie & 19.8+ 1.3* Jorgenson (1981) Table 3 of Johnson et al. 14.3 + 1.6 (1984) value This is per g wet wt. of hepatocytes rather than
2.75+0.04 17.6 + 0.9
liver.
and with all the literature values that we can find expressed in terms of per g of liver (most values in the literature are expressed as per mg of protein). There is no doubt that Table 1 clearly shows that the value reported by Giroix et al. (1987) is low. In another table (Table 1), Giroix et al. (1987) reported glucose-6-phosphatase activity expressed as pmol/min per ,ug of protein expressed as percentage of the activity of rat liver homogenates with 2 mM-substrate. They unfortunately do not give the actual activity in rat liver homogenates but list rat liver homogenate values as 100+5.46 and rabbit liver enzyme (purchased from Sigma) as 442.27+17.01. The activity of the rabbit liver enzyme (purchased from Sigma) in the presence of 1O mM-substrate was given as 57.7 + 3.0 pmol/min per ,ug of microsomal protein on p. 224 (column 2). Assuming a Km of 2.4 and multiplying by 100/442.27 gives a value of approx. 13 pmol/min per,g for fed rat liver glucose-6phosphatase with 10 mM-substrate. This value needs to be corrected for the non-specific phosphatase levels in rat liver homogenates given on p. 223 of Giroix et al. (1987) by multiplying by 0.833 to give 10.8 pmol/min per ,tg or approx. 0.01 ,umol/min per mg of homogenate protein. This value, although only approximate, is clearly quite different to the value of 0.4 ,mol/min per mg of microsomal protein calculated by Malaisse (1990) using data from Table 2 of Giroix et al. (1987) and from literature values, e.g. 0.098 ,mol/min per mg of homogenate protein (Nordlie et al., 1968). A. B. is
a
Lister Institute Research Fellow.
Ian D. WADDELL and Ann BURCHELL Departments of Child Health and Obstetrics & Gynaecology, University of Dundee, Ninewells Hospital and Medical School, Dundee DD1 9SY, Scotland, U.K.
Arion, W. J., Schulz, L. O., Lange, A. J., Telford, J. N. & Wells, H. E. (1983) J. Biol. Chem. 258, 12661-12669 Alvares, F. L. & Nordlie, R. C. (1977) J. Biol. Chem. 252, 8404-8414
Giroix, M.-H., Sener, A. & Malaisse, W. J. (1987) Mol. Cell Endocrinol. 49, 219-225 Johnson, W. T., Nordlie, R. C. & Klevey, L. M. (1984) Biol. Trace Element Res. 6, 369-378 Malaisse, W. J. (1990) Biochem. J. 266, 619 Nordlie, R. C. & Jorgenson, R. A. (1981) J. Biol. Chem. 256, 4768-4771 Nordlie, R. C., Arion, W. J., Hanson, T. L., Gildsdorf, J. R. & Horne, R. N. (1968) J. Biol. Chem. 243, 1140-1446
Nordlie, R. C., Alvares, F. L. & Sukalaski, K. A. (1982) Biochim. Biophys. Acta 719, 244-250 Waddell, I. D. & Burchell, A. (1988) Biochem. J. 255, 471-476
Received 31 October 1989
The pathophysiological significance of increased tight-junctional permeability during oestrogen cholestasis A recent paper in this journal by Kan et al. [1] showed that the cholestatic oestrogen metabolite oestradiol 17,/glucuronide increases tight junctional permeability in isolated perfused rat liver. The authors base the importance of their findings on the hypothesis that oestrogen cholestasis in man and experimental animals is due to D-ring glucuronide conjugates of oestrogens. Although the authors acknowledge that their experiments do not prove a causal relationship between the increased tight junctional permeability and the reduction of the bile flow, this paper nevertheless will rekindle the discussion about this highly controversial topic. This discussion has been ongoing since 1969, when Forker [2] first established the hypothesis that oestrogen cholestasis may be due to the regurgitation of osmotically active bile constituents through the paracellular pathway, thereby reducing the bile flow. Although this is still a valid concept to explain some cases of intrahepatic cholestasis, considerable evidence against the validity of this hypothesis for the oestrogen cholestasis accumulated over the years. Before research efforts are again invested to test this hypothesis, these new data need to be interpreted using the present-day knowledge on the pathogenesis of oestrogen cholestasis. Studies aimed at supporting the hypothesis that Dring glucuronide conjugates are the toxic metabolites causing oestrogen cholestasis rely on indirect evidence. These oestrogen metabolites are detected in urine and bile of humans and experimental animals. The intravenous infusion of these oestrogen metabolites in rats reduced bile flow together with other characteristics of oestrogen cholestasis such as inhibition of bile acid transport (see [3] for review). Besides this indirect supportive evidence, however, there is no definitive proof that the D-ring glucuronide conjugates of oestrogens are the only cause of, or even contribute to, the oestrogen cholestasis in vivo. Until this hypothesis is substantiated, it is necessary to differentiate clearly between experimental findings obtained with isolated oestrogen metabolites and the parent compounds. A second important aspect in this discussion is the dose of oestrogen used in these experiments. Vore and coworkers demonstrated that a bolus dose of 10-133,mol/kg is required for the most potent cholestatic metabolites to inhibit bile flow by 50 % [4,5]. Kan et al. [1] infused 13.6 ,cumol/kg in the isolated perfused liver within 1 min.
1990
