QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Philcox, J. C. Articles by Rofe, A. M. Search for Related Content PubMed PubMed Citation Articles by Philcox, J. C. Articles by Rofe, A. M.

---

Article

Metallothionein in Mice Reduces Intestinal Zinc Loss during Acute Endotoxin Inflammation, but Not during Starvation or Dietary Zinc Restriction

Division of Clinical Biochemistry, Institute of Medical and Veterinary Science, Adelaide, SA 5000 Australia

¹To whom correspondence should be addressed.

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Normal metallothionein [(MT)+/+] and MT-null (MT-/-) mice were used to examine the influence of MT on Zn retention and the metabolic consequences of 2 d food deprivation, with and without inflammation induced by intraperitoneal injection of bacterial endotoxin lipopolysaccharide (LPS). LPS reduced fecal Zn concentration in MT+/+ mice from 5.9 ± 0.2 µmol/g on d 1 to 2.2 ± 0.2 µmol/g on d 2, but not in MT-/- mice, 5.9 ± 0.2 and 5.7 ± 0.5 µmol/g, respectively. MT+/+ mice fed an 8 mg Zn/kg diet and injected with LPS excreted 40% less Zn over 2 d than their MT-/- counterparts. Starvation for 2 d did not lower fecal Zn concentration in either genotype, although in MT+/+ mice, urinary Zn excretion was reduced from 12.7 ± 1.3 nmol on d 1 to 5.9 ± 1.8 nmol on d 2 and plasma Zn concentration was lowered to 9.8 ± 0.4 µmol/L. Zn was not reduced in urine or plasma of MT-/- mice, with respective values of 10.8 ± 2.0 nmol on d 1, 9.3 ± 2.9 nmol on d 2 and 13.0 ± 1.0 µmol/L. LPS injection resulted in much higher total liver Zn (677 ± 27 nmol) and MT (106 ± 2 nmol Cd bound/g) than starvation (Zn = 405 ± 21, MT = 9 ± 3) in MT+/+ mice after 2 d, but did not further reduce urinary Zn. LPS-injected MT-/- mice had no rise in liver Zn or fall in plasma and urine Zn. MT-/- mice fed a Zn-deficient (0.8 mg Zn/kg) diet lost 10% of body weight over 25 d compared with no loss in MT+/+ mice. Despite this, MT-/- mice excreted no more Zn via the gut than did MT+/+ mice. In summary, MT inhibits intestinal Zn loss when highly expressed. When uninduced, typically during Zn deficiency, MT appears to conserve Zn and body mass by reducing only urinary and other nonintestinal Zn losses.

KEY WORDS: • mice • metallothionein • zinc • endotoxin

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Metallothionein (MT)² is a cysteine-rich, intracellular metal-binding protein, incorporating 4–7 g-atom Zn/mol. MT plays a part in normal Zn homeostasis, but has a more prominent role in inflammation-mediated tissue Zn redistribution [De et al. 1990 ; for reviews, see Hamer (1986) and Kagi (1987) ]. Cytokines and counterregulatory hormones induce MT synthesis in the liver, the chief Zn recipient (Abe et al.1987 ), in which increases in MT protein can be demonstrated as early as 2–4 h after an inflammatory insult, preceding the appearance of acute phase proteins in the plasma. The dependence of liver Zn accumulation on MT synthesis has been confirmed using MT-null (MT-/-) mice administered either a bacterial endotoxin lipopolysaccharide (LPS) by intraperitoneal injection (Philcox et al. 1995 , Rofe et al. 1996 ) or Zn (Coyle et al. 1995 ). Although it was shown over 20 years ago that endotoxemia in rats results in a cytokine-driven increase in Zn absorption and retention (Pekarek and Evans 1975 and 1976 ), the involvement of MT remains to be proven. In humans, LPS injection limits Zn loss in the urine, presumably by cytokine-directed Zn redistribution (Gaetke et al. 1997 ).

Food deprivation causes a much smaller increase in MT than does LPS. This increase is approximately threefold baseline, largely in response to a rise in plasma glucagon (Tran et al. 1998 ). There is evidence that MT at this low concentration may benefit Zn retention. For example, after 20 h of food deprivation, MT-/- mice had greater reductions than normal (MT+/+) mice in plasma Zn (5.3 vs. 1.7 µmol/L) and liver Zn (27 vs. 8%) (Philcox et al. 1995 ). Furthermore, MT-/- mice excrete a greater proportion of subcutaneously administered ⁶⁵Zn from the pancreas and probably from intestinal mucosa than do MT+/+ mice (Rofe et al. 1999b ). We recently reported that MT+/+ mice are better able than MT-/- mice to absorb ⁶⁵Zn from a 50-mg Zn/kg egg white–based test meal (Coyle et al. 1999 ), indicating that MT at low concentration may also enhance absorption of Zn from solid food.

This study evaluates the contribution of MT to the tissue distribution and balance of Zn by comparing responses of MT-/- and MT+/+ mice to the following three challenges in which Zn retention is paramount, but MT induction differs widely: 1) food deprivation for 2 d, 2) acute inflammation superimposed on food deprivation for 2 d and 3) severe dietary Zn deficiency for 5 wk.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Experiment 1: food deprivation and inflammation for 2 d

Mice lacking both MT-1 and MT-2 in a mixed genetic background of OLA129 and C57BL6J strains, produced by Michalska and Choo (1993) , were used. Control (MT+/+) mice were C57BL6J supplied by the Animal Resources Center, Canning Vale, Western Australia. The mice best suited for metabolic cage studies were full-grown males because they have an excellent yield of blood for the many analyses performed, have reasonable urine outputs and well-defined epididymal fat depots. There is also no potential for mismatched estrus cycles confusing result interpretation. Hybrid (MT+/-) mice were used for the experiment involving LPS inflammation, in which we found large differences in measured responses between MT+/+ and MT-/- mice. Differences in the starved and Zn-deficient MT+/+ and MT-/- mice were too small to warrant inclusion of a starved or Zn-deficient MT+/- group. The MT+/- mice were first generation crosses between MT+/+ and MT-/- mice.

The mice were housed individually in metabolic cages; n = 6 for each of the following treatments: 1) MT+/+ and MT-/- mice starved for 2 d and 2) MT+/+, MT-/- and MT+/- mice given a single intraperitoneal injection of LPS (from Escherichia coli 0111.B4: Sigma Chemical, St. Louis, MO), then starved for 2 d. The LPS dose of 5 mg/kg was administered in 0.2 mL of sterile physiological saline, which is well tolerated and produces a maximal MT response (Rofe et al. 1996 ).

All mice, 10–12 wk old, were matched for initial body weight, MT+/+, 26.82 ± 0.34 g (mean ± SEM) and MT-/-, 26.83 ± 0.59 g in the starvation experiment and MT+/+, 27.20 ± 0.17 g; MT-/-, 27.04 ± 0.23 g; and MT+/-, 27.07 ± 0.21 g in the LPS experiment. All mice had had free access to a commercial nonpurified diet (Milling Industries, South Australia) containing 30 mg Zn/kg, before being placed in the metabolic cages. This diet was the same as that used by Coyle et al. (1993) , but with no supplementary Zn. Water consumption and body weight were measured on d 0, 1 and 2. Urine and feces excreted on the previous day were removed and weighed on d 1 and 2. Two consecutive 1-d fecal collections were necessary, d 1 to clear exogenous Zn from the gut and d 2 to gauge the Zn excretory response to the experimental treatment. This approach was validated by gavaging mice with 37 Bq ⁶⁵Zn in 0.5 mL of 20 g/L albumin and measuring ⁶⁵Zn in blood, urine, heart, liver, kidneys, muscle and skin as well as stomach, small intestine, cecum, colon and feces at timed intervals. In general, the highest concentrations of ⁶⁵Zn in muscle, skin and the organs were reached at 4 h, and remained unchanged for the next 12 h. All ⁶⁵Zn had cleared the gut lumen, including the cecum and colon, within 16 h and was shown to be present in the feces. Complete resorption of retroperitoneal fat in some MT+/+ mice indicated a practical limit of 2 d for the duration of starvation.

    Short-term dietary adjustment to 8 mg Zn/kg. MT+/+ and MT-/- mice were housed together (n = 3/group) in a wire-floored cage to establish a common feeding pattern, while preventing coprophagy. They were fed a diet containing 8 mg Zn/kg for 5 d to lower the amount of exogenous Zn in feces on d 1. The procedure was otherwise identical to the LPS experiment above. Mouse body weight before LPS injection was MT+/+, 25.95 ± 0.25 g and MT-/-, 25.90 ± 0.23 g. These mice were third generation derivatives from backcrossing MT-/- to control C57BL6J mice.

    Metabolic cages. The metabolic cages were constructed in-house. They had a minimal surface area for urine collection (collection funnel diameter 75 mm) and the feces were retained on a 55-mm diameter stainless steel grid (1.8-mm mesh), anchored beyond reach of the mice to prevent coprophagy. At the end of each collection period, after the removal of feces, dried urine spots were carefully rinsed from the funnel and mesh with the minimum amount of distilled water (3–7 mL) required to completely remove all traces. The rinsings were diluted to 10 mL, acidified with HCl and assayed for Zn by atomic absorption spectrophotometry (AAS). Contamination of urine and rinsings with Zn from the feces was negligible. Urine Zn concentrations matched those typical of bladder aspirations.

Approximately one third of urinary Zn excretion was contained in the rinsings. Zn concentration was that of the urine in the collection tube, and the total urinary Zn excretion was the sum of Zn present in the collection tube and rinsings. To gain a closer approximation of the urinary volume as excreted, the volume in the collection tube was multiplied by the sum of total Zn in collection tube and rinsings, and divided by the total Zn in the collection tube alone. That is, the volume was increased by the estimated one third that did not flow into the collection tube. This volume was used to calculate the difference between fluid intake and urinary excretion. Other components of fluid balance such as respiratory water vapor loss were not measured. The actual fluid deficits hence exceed our reported values.

After 2 d in metabolic cages, the mice were anesthetized with halothane and 1 mL of blood withdrawn for assay of plasma Zn and metabolites (glucose, lactate, ß-hydroxybutyrate); for that assay, 0.1 mL of blood was immediately added to 0.4 mL of ice-cold 0.8 mol/L perchloric acid. Plasma fibrinogen was measured in the LPS experiment. The mice were killed by cervical luxation and the livers excised without delay, 200 mg homogenized in 0.8 mL of ice-cold 0.8 mol/L perchloric acid for metabolite and glycogen analysis, and the remainder frozen for MT and Zn assay. Other organs and tissues removed and weighed included kidneys, heart, spleen, epididymal and retroperitoneal fat pads, skin (including hair) and carcass. The carcass weight included head, paws and tail. Liver and feces were dried (70°C, 2 d) before nitric acid digestion and Zn assay by flame AAS (PE3030, Perkin-Elmer, Norwalk, CT). MT was analyzed by a Cd-heme method using a 1:5 dilution of homogenate in 0.01 mol/L Tris-HCl, pH 8.2 (Eaton and Toal 1982 ). The perchloric acid supernatants from blood and liver were neutralized with saturated potassium bicarbonate and the metabolites were measured by previously described enzymatic techniques (Rofe and Williamson, 1983 ) in a semiautomated procedure (ABA 100; Abbott Diagnostics, Pasadena, CA). Fibrinogen was measured using Fibriquick reagent (Organon Teknika, Boxtel, Netherlands) and a Roche Cobas-Fibro analyzer (Roche Diagnostic Systems, French’s Forest, NSW, Australia)

Experiment 2: chronic severe dietary Zn deficiency

This experiment was to determine the influence of MT at uninduced concentrations on the excretion of Zn from mice fed a severely Zn-restricted diet. A diet containing 0.8 mg Zn/kg was used (Coyle et al. 1999 ) for the Zn deficiency study. Egg white provided the nitrogen and cellulose the fiber, both chosen for low intrinsic Zn content. MT+/+ and MT-/- mice were used, n = 16/genotype, with 8 mice per wire-floored plastic cage. To confirm normal growth and food intake from this cohort, four mice from each genotype were housed similarly and fed a 100 mg/kg Zn-replete diet. All mice used in Experiment 2 were females, 6–8 wk old, weighing 21.0 ± 0.4 g, to allow for some further growth. They were third-generation derivatives from backcrossing MT-/- to C57BL6J mice. The study was terminated at 37 d, when the trends in weight loss were firmly established. As the period of Zn deficiency extended, the total dietary intake fell more in MT-/- than in MT+/+ mice, adding a progressive protein-energy shortfall to the metabolic picture of the MT-/- mice. It was not possible to correct for variation in food intake by pair feeding because the differences were too small to be measured accurately and compensated for. The calculations of Zn intake, final body mass and Zn composition correct for this variation in food consumption.

Individual mice were weighed; water and food intakes per cage were measured daily for the first 11 d and thereafter three times per week. Feces excreted for the previous 24 h were collected at the same times, dried at 70°C, then weighed; a 200-mg aliquot was digested in nitric acid, diluted in 0.01mol/L HCl and assayed by AAS for Zn. At the end of the experiment, the mice were treated as in the earlier experiments, with additional assay of Zn concentrations in skin, kidneys, abdominal muscle (central region of rectus abdominis containing a high proportion of red, oxidative fibers) and hind limb muscle (predominantly white quadriceps, from the rectus femoris).

Statistical analysis

Repeated-measures ANOVA using the General Linear Model on Minitab (Minitab, State College, PA) was used. Experiment 1: differences between food-deprived MT+/+ and MT-/- mice with or without LPS were compared by two-way ANOVA in a 2 x 2 factorial design. Variability was expressed as the root mean square error (RMSE), an estimate of the standard deviation across treatments. When interactions were significant, Tukey’s post-hoc test was used to distinguish differences between specific means (Tukey 1949 ). One-way ANOVA was used for comparison among LPS-treated MT+/+, MT+/- and MT-/- mice. Experiment 2: one-way ANOVA was used for calculating significant differences in weight loss among genotypes. Where appropriate, in both experiments, data have been expressed as means ± SEM and significant differences determined by two-tailed Student’s t test for independent samples.

The Institute of Medical and Veterinary Science Animal Ethics Committee approved all experimentation.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Experiment 1

    Weight loss. Starved MT+/+ and MT-/- mice injected with LPS lost 27 ± 5 and 40 ± 4% less weight on d 1 than their starved, uninjected counterparts. MT-/- mice also lost 28 ± 5% less weight on d 1 after LPS than did MT+/+ and MT+/- mice (-3.32 ± 0.12 g) (Table 1 ). Weight loss was 59 ± 2 (MT+/+) and 53 ± 4% (MT-/-) less on d 2 than d 1 in starved mice. LPS-injected mice had a more uniform loss on d 2 (i.e., MT+/+, 2.39 ± 0.07 g; MT-/-, 2.25 ± 0.20 g; and MT+/-, 2.33 ± 0.11 g).

    Body composition. Carcass (mainly skeletal muscle) weight is an index of lean body mass, representing tissue protein, and at d 2 did not differ among groups. LPS-injected MT-/- mice had a significantly higher mean skin weight than all other groups, caused largely by visibly greater subcutaneous fat accumulation. Livers and spleens were significantly larger in all mice injected with LPS. In those tissues in which there were differences between MT+/+ and MT-/- mice, namely, spleen, epididymal fat and retroperitoneal fat, the LPS MT+/- mice had intermediate values of 0.10 ± 0.01, 0.25 ± 0.05 and 0.05 ± 0.01 g, respectively. The two last-mentioned values differed significantly from those of MT-/- (P < 0.001, one-way ANOVA), but not from MT+/+ mice.

    Liver Zn and MT and plasma Zn. To best represent the liver Zn pool size, irrespective of treatment-induced changes in hepatic size, hydration or fat content, we used total liver Zn content as an index of hepatic Zn status (Table1). Total liver Zn (nmol) in male mice with free access to food (not shown in Table1) was MT+/+, 579 ± 20 (n = 16, body weight 26.04 g); MT-/-, 576 ± 33 (n = 18, body weight 26.11 g) and MT+/- 569 ± 19 (n = 5, body weight 25.08 g.) The total liver Zn of these mice was adjusted to a body weight of 27 g, to match the Experiment 1 groups before food deprivation.

MT+/+ mice had hepatic Zn concentrations of 459 ± 4 nmol/g wet liver after 2 d starvation and 704 ± 22 nmol/g after LPS injection and starvation. The corresponding values in MT-/- mice were 417 ± 15 and 409 ± 13, respectively. LPS-injected MT+/- mice had hepatic Zn concentrations of 702 ± 18 nmol/g. LPS-stimulated MT concentrations in MT+/+ and MT+/- mice were 106 ± 12 and 70 ± 6 nmol Cd bound/g wet liver, respectively, and not detectable above assay background in the MT-/- mice. MT was 8.7 ± 2.2 nmol Cd bound/g in the MT+/+ mice starved for 2 d. The mean plasma Zn concentration of the LPS MT+/- mice was 11.8 ± 0.4 µmol/L.

It is noteworthy that LPS-injected MT+/- mice synthesized sufficient MT within 2 d of LPS injection to have a liver Zn content not different from that of similarly treated MT+/+ mice, despite having a 34% lower hepatic MT concentration (as well as 31% less fibrinogen). The MT+/- mice had no symptoms of LPS toxicity on d 2, unlike their MT+/+ counterparts, appearing to recover more rapidly from the effects of LPS. Their higher total liver Zn:MT ratio may thus indicate that MT changed in parallel with the inflammatory process, which peaked earlier and resolved more rapidly. As liver MT levels fall, Zn previously sequestered in the liver is released into the plasma compartment to normalize plasma Zn, as was observed.

    Metabolites. LPS administration to MT+/+ mice significantly decreased liver glucose (-37%) blood glucose (-36%) and blood lactate (-47%), compared with starvation (Table 1) . A similar trend, significant only for blood lactate, was seen in MT-/- mice, with lower values in general, typified by a liver glucose after food deprivation 34% lower than in MT+/+ mice. The absence of MT was associated with hepatic glycogen reserves that were one fourth of normal. Blood glucose was also significantly lower (by 36%) in food-deprived MT-/- mice than their MT+/+ counterparts, but did not decrease further with LPS treatment. Liver and blood lactate of MT+/+ and MT-/- mice did not differ. This was also true of liver ß-hydroxybutyrate, which showed little variation among groups. The blood ß-hydroxybutyrate after LPS injection, on the other hand, was 69% higher in MT-/- mice than MT+/+ mice. LPS MT+/- mice had liver concentrations of glycogen, 53.6 + 7.7 µmol/g; glucose, 22.4 ± 1.7 µmol/g; and lactate, 5.4 ± 0.5 µmol/g as well as blood glucose, 7.7 ± 0.2 mmol/L, that were between corresponding values from starved and LPS MT+/+ mice. Blood ß-hydroxybutyrate (1.7 ± 0.1 mmol/L) of MT+/- mice was intermediate between LPS MT-/- and LPS MT+/+ mice and significantly different from each (P < 0.05). These results demonstrated no difference in carbohydrate metabolism between MT+/- and MT+/+ mice, although MT+/- mice appeared less affected by LPS on d 2.

    Acute phase response to LPS injection. Plasma fibrinogen concentrations, 2 d after LPS injection were used as an objective index of the acute phase response. They were 5.80 ± 0.23, 4.46 ± 0.45 and 3.98 ± 0.20 g/L for MT+/+, MT-/- and MT+/- mice, respectively, 2–3 times unstimulated concentrations. MT+/+ mice had significantly higher fibrinogen concentrations than other mice; MT-/-, P < 0.05; MT+/-, P < 0.001, Student’s t test. There was evidence of a more prolonged inflammatory reaction in MT+/+ mice, despite the use of identical doses of LPS. Although all mice had very similar signs and symptoms of LPS toxicity (piloerection, hunched posture and torpor) on d 1, MT+/+ mice deteriorated over d 2, whereas MT-/- mice remained unchanged and MT+/- mice improved.

    Zinc excretion in feces and urine. In Experiment 1, mice lost almost all Zn via the gut, with renal excretion contributing only 2–3% of the total. Figure 1 shows total fecal Zn loss over the two 1-d collection periods. The main determinant of total 2-d Zn loss was fecal Zn excretion on d 1; at the highest extreme, this loss was 1380 ± 123 nmol in starved MT+/+ mice and, at the lowest, 407 ± 72 nmol in the LPS-treated MT-/- mice. This variation obscured any differences in total 2-d Zn excretion attributable to MT. The low Zn output in MT-/- mice must be interpreted in the context of their much lower (dry) fecal mass on d 1, i.e., 69 ± 12 vs. 130 ± 11 mg in MT+/+ mice. More importantly, d 2 fecal mass did not differ between the two groups of LPS-treated mice (67 ± 8 and 62 ± 8 mg), and no fecal matter remained in the gut of any mouse at killing. A component of the lower d 1 fecal output in MT-/- mice may have been an inflammation-related protraction of intestinal transit because LPS-treated MT-/- mice had a much higher d 2 fecal output than their uninjected, starved counterparts (12 ± 2 mg); however, uninjected MT+/+ mice also had very low d 2 fecal excretions (8 ± 3 mg). The cause of the high d 1 fecal Zn output in MT+/+ mice was thus more likely to be greater food consumption immediately before the experiment. Starved MT+/+ mice had a 45% higher d 1 mean fecal output than the mean from all other groups (161 ± 3 vs. 111 ± 9 mg dried feces). Therefore, in mice previously fed a commercial nonpurified diet, the d 2 change in Zn concentration (Fig. 2 ) rather than the change in total fecal Zn excretion was a better reflection of the effect of MT on gut Zn loss. The fall in total fecal Zn excretion from starved mice over 2 d was caused by a gross reduction in volume that overwhelmed a minor increase in concentration. The three LPS-injected mouse genotypes had identical fecal Zn concentrations on d 1; these fell by 61 ± 5 and 52 ± 6% in MT+/+ and MT+/- mice, respectively (P < 0.001, t test) over 2 d, but were unchanged in MT-/- mice (4 ± 9%).

View larger version (22K): [in this window] [in a new window]   Figure 1. Total fecal Zn excretion in metallothionein (MT)+/+ and MT-/- mice after starvation and in MT+/+, MT+/- and MT-/- mice after injection of the endotoxin lipopolysaccharide (LPS; Experiment 1). The results represent means ± SEM, n = 5–6. aSignificantly different from other LPS d 1 excretions, P < 0.05. bSignificantly different from other LPS d 2 excretions, P < 0.05.

View larger version (13K): [in this window] [in a new window]   Figure 2. Consecutive daily (d 1 and 2) fecal Zn concentration in metallothionein (MT)+/+ and MT-/- mice during starvation, and in food-deprived MT+/+, MT+/- and MT-/- mice after injection of the endotoxin lipopolysaccharide (LPS; Experiment 1). The results represent means ± SEM, n = 5–6. aSignificantly different from d 1 concentration, P < 0.001. bSignificantly different from d 2 concentrations in other LPS-injected mouse genotypes, P < 0.001.

View larger version (43K): [in this window] [in a new window]   Figure 3. Consecutive daily total urinary Zn excretion in metallothionein (MT)+/+ and MT-/- mice during starvation, and in food deprived MT+/+, MT+/- and MT-/- mice after injection of the endotoxin lipopolysaccharide (LPS; Experiment 1). The results represent means ± SEM, n = 5–6. aSignificantly different from d 1 excretions, P < 0.05.

The MT-mediated reduction in fecal Zn was seen earlier in LPS-treated mice fed the low Zn diet, in which a trend (P = 0.25) toward a lower excretion by MT+/+ than MT-/- mice on d 1 became significant over 2 d. The 48-h fecal Zn excretion was 251 ± 18 (MT+/+) and 416 ± 37 nmol (MT-/-) (P < 0.03, Student’s t test). Compared with their counterparts fed the nonpurified diet, LPS-treated mice fed the 8 mg Zn/kg diet had lower d 1 fecal Zn concentrations of 2.41 ± 0.25 (MT+/+), and 3.23 ± 0.27 (MT-/-). Day 2 values were 2.50 ± 0.98 and 4.39 ± 1.24 µmol Zn/g, respectively. MT+/+ mice fed the 8 mg Zn/kg diet had d 1 and 2 fecal Zn concentrations that were not different from the d 2 concentration of MT+/+ mice fed the nonpurified diet (2.22 ± 0.23 µmol/g). Total d 1 and 2 urinary Zn losses, in nmol Zn/d, were 7.8 ± 2.5 and 2.7 ± 0.9 (MT+/+) and 5.2 ± 2.0 and 9.1 ± 4.4 (MT-/-), consistent with earlier results.

Experiment 2: response to chronic severe dietary Zn deficiency

We found that food utilization, growth rate and liver Zn content of most cohorts of our MT-/- mice closely matched those of MT+/+ mice up to age 12 wk. However, it was deemed advisable to confirm that this similarity was maintained in the cohort of smaller female mice used in Experiment 2. Baseline Zn concentrations in muscle were also required.

Mice fed the 100 mg Zn/kg diet gained a total of 4.7 ± 0.6 g (MT+/+) and 4.6 ± 0.5 g (MT-/-) over 40 d. Growth was linear, giving a regression slope of 0.1093, MT+/+ (r² = 0.95) and 0.1088, MT-/- (r² = 0.98), representing a mean weight gain of 109 mg/d for each over the experimental period. Food consumption was 3.4 ± 0.075 g/d for both genotypes. Conversion of dietary protein into body mass was thus indistinguishable between genotypes at this stage of growth, supporting earlier findings in our MT-/- mouse populations.

Feeding a 0.8 mg Zn/kg diet caused MT-/- mice to lose more weight than MT+/+ mice (Fig. 4 ). MT+/+ mice were better able to maintain initial body weight, losing only 1 g by 30 d, in contrast to MT-/- mice (a 3-g loss). Daily food intake over the first 11 d was 3.32 g/mouse (MT+/+) and 3.10 g/mouse (MT-/-). The difference in daily food intake between MT+/+ and MT-/- mice increased over the experimental period, from 0.2 g initially to 0.55 g at 35 d. Fecal Zn excretion (Fig. 5 ), with large day-to-day variation, was 24 nmol Zn/d less in MT-/- mice than in MT+/+ mice over the experimental period. Despite lower fecal Zn losses, MT-/- mice lost weight at a greater rate than MT+/+ mice, amounting to a 2.0 g difference at 37 d, equivalent to ~10 nmol Zn/d. Their daily dietary Zn shortfall was equivalent to 6 nmol/mouse. The tissue Zn levels at 37 d (corrected for differences in weight loss) indicated that the MT-/- mice had lost, on average, ~8 nmol Zn/d more than MT+/+ mice from their remaining tissues. In summary, the daily Zn deficit between oral intake and fecal excretion averaged 18 nmol/mouse less in MT-/- than in MT+/+ mice, despite evidence that the MT-/- mice lost 18 nmol/d more Zn than MT+/+ mice from catabolized tissues.

View larger version (23K): [in this window] [in a new window]   Figure 4. Weight loss in metallothionein (MT)+/+ and MT-/- mice fed a Zn-deficient diet (0.8 mg Zn/kg) for 37 d. The results represent means ± SEM n =16 mice/genotype. aWeight loss in MT-/- mice was significantly different from MT+/+ mice from d 17, P < 0.05. Over the entire period, weight loss was significantly different between genotypes, P < 0.001.

View larger version (13K): [in this window] [in a new window]   Figure 5. Fecal Zn excretion in metallothionein (MT)+/+ and MT-/- mice fed a Zn-deficient diet (0.8 mg Zn/kg) for 37 d. The results are shown as the pooled mean daily excretion (2 cages per genotype, 8 mice per cage). The regression for MT+/+ mice is y = 161.49e-0.0428x (r = -0.729) and for MT-/- mice is y = 111.51e-0.0343x (r = -0.601).

Plasma Zn concentrations at 37 d were 10.9 ± 0.7 (MT+/+) and 12.2 ± 0.9 µmol/L (MT-/-) in Zn-replete control mice, and 6.5 ± 0./6 and 6.1 ± 0.3 µmol/L, respectively, in mice fed the 0.8 mg Zn/kg diet.

Liver Zn concentrations did not differ between genotypes, 378 ± 13 nmol/g (MT+/+) and 360 ± 10 nmol/g (MT-/-), in mice fed the 0.8 mg Zn/kg diet for 37 d. At this time, hepatic MT was 4.8 ± 0.5 nmol Cd bound/g wet liver in the Zn-deficient MT+/+ mice, not different from control mice (MT 4.8 ± 0.5 nmol Cd/g). Although hepatic Zn concentrations fell minimally, the livers shrank by 30% as a proportion of body weight in both genotypes, reducing total hepatic Zn by 43% in MT+/+ and 50% in MT-/- mice compared with their Zn-replete counterparts. The lower liver Zn content in the MT-/- mice resulted from their greater overall weight loss.

Muscle was sampled from two sites, i.e., the abdominal region containing mainly red (oxidative) fibers, and a portion of the quadriceps, mainly white (glycolytic) fibers. Zn loss from the two regions was different (Fig. 6 ). Both genotypes had significantly lower Zn concentrations in predominantly red abdominal muscle when mice consumed the 0.8 mg Zn/kg diet, (MT+/+; P < 0.001, MT-/- P < 0.02, Student’s t test). On the other hand, MT+/+ mice maintained predominantly white quadriceps Zn concentrations, whereas MT-/- mice had a reduction of 29% (P < 0.01, t test).

View larger version (46K): [in this window] [in a new window]   Figure 6. Muscle Zn concentrations from metallothionein (MT)+/+ and MT-/- mice fed a Zn-deficient diet (0.8 mg Zn/kg) for 37 d. The results represent means ± SEM (n = 16/genotype). Shaded bars represent control mice (mean ± SEM, n = 4/genotype) from the same cohort maintained on a Zn-replete diet. aSignificantly different between diets, P < 0.02. bSignificantly different between genotypes, P < 0.05.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
MT has long been implicated in the regulation of absorption and excretion of Zn by the intestine (Richards and Cousins 1975 and 1976 ), and the action of MT in restricting Zn absorption at high dietary Zn concentrations has gained acceptance [for review, see Cousins (1985) ]. However, the influence of MT in enhancing Zn absorption/retention when dietary Zn is low remains unclear. In this study, we demonstrated in mice that MT, when highly induced, greatly restricts loss of Zn from the gastrointestinal tract, but does not do so at the lower concentrations found during starvation and severe dietary Zn deficiency. A MT-dependent reduction in renal, and presumably other extraintestinal Zn losses was found, but appeared to be secondary to less efficient carbohydrate metabolism caused by the absence of MT. These findings are discussed, with emphasis on the metabolic consequences of MT deletion.

LPS inflammation: high MT induction.

Conservation of body Zn during acute inflammation appears to rely mainly on increased ZnMT accumulation in the liver. The potential also exists for MT in other tissues, including the small intestine (Coyle et al. 1999 ) and pancreas (De Lisle et al. 1996 , Rofe et al. 1999b ) to reduce Zn loss from its major route of excretion, the gut, although findings regarding inflammation in this context are scanty at best. More information is available for pregnancy, during which elevations of MT inducers common to inflammation, including steroid hormones (Quaife et al. 1986 ) and cytokines, especially interleukin-6, occur (De et al. 1992 ). Progressive hepatic MT accumulation in pregnancy is associated with an increased uptake of Zn from the diet (Rofe et al. 1999a ). Fetuses from mice overexpressing MT have also been found to be more resistant to the effects of maternal Zn deficiency (Dalton et al. 1996 ). Furthermore, MT ameliorates the effects of Zn deficiency during the immediate postnatal period, when MT-/- mouse pups fed reduced Zn intakes have been shown to have renal malformations linked to low hepatic Zn reserves (Kelly et al. 1996 ). MT has also been shown to protect the fetus during pregnancy when Zn intake of the dam is limited (Andrews and Geiser 1999 , Rofe et al. 1999a ). The studies on pregnancy show a MT-dependent increase in extraction of exogenous Zn from the diet, but do not give an indication of the contribution of endogenous Zn retention.

The present study, in which LPS inflammation elicited a 61% reduction in fecal Zn concentration over 2 d in starved MT+/+, but not in MT-/- mice, indicates that MT restricts the loss of endogenous Zn. Furthermore, MT+/+ mice fed a low Zn diet before LPS administration excreted 165 nmol (40%) less Zn in the feces and 5 nmol less Zn in the urine over 2 d than did their MT-/- counterparts. In this experiment, assuming that MT+/+ and MT-/- mice started with the same liver Zn content, the difference in liver Zn of 275 nmol between genotypes at d 2 could be attributed to 170 nmol more Zn retained and 105 nmol redistributed from other tissues in MT+/+ mice. Because the combined muscle, skin and bone mass of a 27-g mouse is ~12–13 g, the LPS-stimulated transfer would require a reduction of <10 nmolZn/g within donating tissues, not demonstrable by conventional Zn measurements. What is clear is that the MT-related increase in liver Zn exceeds the total Zn conserved, demonstrating the capacity of this organ to be solely responsible for retention of endogenous Zn, at least in the short term, with no dietary input.

Starvation: mild MT induction.

Starvation induced a threefold elevation in hepatic MT, associated with a 50% decrease in urinary Zn loss on d 2 in MT+/+ mice. Much higher, LPS-stimulated MT concentrations did not cause a further reduction. Urinary Zn did not fall in response to either challenge in MT-/- mice. Although fecal volume fell markedly during starvation, fecal Zn concentration did not decrease in either genotype. The contribution of MT to the retention of body Zn during starvation thus resulted entirely from a reduction in urinary Zn excretion, amounting to a mere 7 nmol/d, which is trivial in terms of murine Zn reserves. Wastney and co-workers (1999) showed that human preterm infants, gestational age 32 wk, enterally fed a diet containing 15 ± 0.2 mg Zn/kg, excreted ~3% of total Zn loss in the urine, not unlike the proportion seen here in mice. However, a 50% reduction in renal Zn excretion by adult humans, who may excrete a greater proportion of Zn in the urine and who are able to withstand prolonged periods of inadequate food intake, may be more important.

Retention of an additional 70 nmol Zn in the liver possibly contributed to starved MT+/+ mice having a plasma Zn concentration 25% lower than that of their MT-/- counterparts. However, the added stimulus of LPS injection, which caused a 290 nmol (17%) elevation in liver Zn in MT+/+ above MT-/- mice at 2 d, was not associated with a further fall in plasma Zn concentration (9.0, LPS vs. 9.8 µmol/L, starved). The well-described LPS-induced inverse relationship between liver and plasma Zn was not seen, because at d 2, equilibrium between hepatic uptake and Zn release from other tissues had occurred. Nadir plasma concentrations of 2–5 µmol Zn/L were seen much earlier, at 4–8 h.

Metabolic implications from 2-d experiments.

Starved LPS-injected MT-/- mice lost significantly less fluid and weight over 1 d than did their MT+/+ counterparts, probably secondary to a reduction in glycogen consumption/metabolism of carbohydrate fuels, and hence a lower production of metabolic water. Previous measurements in MT-/- mice revealed lower reserves of hepatic glycogen, as well as depressed liver glucose concentrations (Rofe et al. 1996 and 1998 ). This was apparent in the present study, in which MT-/- mice had much lower hepatic glycogen levels, despite no evidence of increased utilization. The absence of MT may mimic a functional Zn deficiency in certain situations; it may be pertinent that rats fed a 1 mg Zn/kg diet for 5 d before self-selecting food from a three-choice macronutrient diet significantly reduced carbohydrate and increased fat intake (Kennedy et al. 1998 ). Fuel substitution, albeit in endogenous substrates, may also occur in MT-/- mice, which appear to have impaired ability to synthesize and store carbohydrate. They develop larger fat reserves and may preferentially metabolize lipids at a higher ratio to other fuels earlier during starvation. By burning a greater proportion of more energy dense fuel at what may also be a lower rate of metabolism, MT-/- mice would conserve body weight, particularly lean body mass, in the shorter term (Robinson and Williamson 1980 ). Blood ß-hydroxybutyrate was higher in MT-/- mice, as were urinary ketones, supporting this explanation. The livers of MT-/- mice contained grossly more fat 2 d after LPS injection, suggesting a mismatch between fat mobilization from peripheral tissues and the metabolic capacity to oxidize it in the liver. Calorimetry and respiratory quotient determinations are required to determine fuel usage by MT-/- mice, both qualitatively and quantitatively.

MT+/- mice were used in the LPS experiment to gain a better indication of which metabolic anomalies in MT-/- mice were likely to be caused directly by MT deficiency, rather than result from differences in mouse strain. MT+/- mice had liver and blood glucose and lactate concentrations, as well as fluid and weight loss, similar to those of MT+/+ mice, suggesting that the abnormal carbohydrate metabolism of MT-/- mice is likely to be symptomatic of MT deficiency. On the other hand, the lipid-related findings of MT+/- mice were intermediate between MT-/- and MT+/+ mice, giving no clear indication that the adiposity of MT-/- mice was caused by the absence of MT. Another possible cause is a strain-dependent proclivity to obesity, unmasked in the MT-/- mice. Obesity in C57BL6J mice fed high fat diets has been reported (Black et al. 1998 , Opara et al. 1996 , Surwit et al. 1995 ).

Beattie and co-workers (1998 and 1999) described spontaneous obesity in MT-/- mice derived from the same stock as those used here. Their mice, in which >20% of males weighed 46–59 g at 22–39 wk, had a 6-g weight increase over C57BL6J mice by 8 wk of age. In the cohorts of MT-/- mice used in our study, excessive weight gain began at a later stage, after the mice had exceeded 28 g (male) or 22 g (female). Beattie et al. (1999) also found that more severe obesity corresponded to greater ob gene mRNA and plasma leptin concentration, with severe leptin resistance manifested by increased food consumption and higher plasma insulin concentration. Although in some ways MT-/- mice are similar to leptin receptor deficient (db/db) mice, they differ in other respects (e.g., blood glucose remains low and onset of obesity is delayed). A strong association between obesity and defective carbohydrate metabolism, and a requirement for MT in the regulation of key Zn metalloenzymes in carbohydrate metabolism (Maret et al. 1999 ) may seem compelling evidence for an influence of MT on adiposity. Wide variations in weight gain between MT-/- cohorts and equivocal lipid metabolic indices in MT+/- mice suggest otherwise.

Chronic gross dietary Zn deficiency.

Intestinal inhibition of Zn excretion in response to dietary Zn deficiency had no demonstrable MT-related component in mice, in common with the finding after 2 d starvation and in agreement with an earlier report (Flanagan et al. 1983 ). An acute decrease in fecal Zn from both genotypes was a simple reflection of the much lower Zn concentration within the ingesta. In the light of recent findings (Coyle et al. 1999 ), the more negative gastrointestinal Zn balance in MT+/+ than MT-/- mice over 37 d of dietary Zn deficiency was unexpected. Nonetheless, the onset of weight loss was delayed by 3 wk in MT+/+ mice, and the calculated tissue Zn loss was higher in MT-/- than in MT+/+ mice. This can be accounted for only by excessive urinary Zn excretion in MT-/- mice. The implication is that high renal Zn loss may be part of a cycle in which the lack of MT-facilitated Zn incorporation into certain enzymes of carbohydrate metabolism (Brand and Klieneke 1996 ) condemns chronically Zn-depleted MT-/- mice to a more catabolic course. Zn from catabolized muscle and other tissues is released into the plasma; this supports the Zn requirement for energy production, but it also enables a larger proportion to be lost in the urine, exacerbating the dietary Zn deficiency. A greater proportion of this Zn would be expected to be bound to the amino acids cysteine and histidine, facilitating urinary excretion (Yunice et al. 1978 ). A conserving action of renal MT in MT+/+ mice also cannot be excluded.

Further evidence of the value of MT in glycolysis may be found as differences between genotypes in cumulative loss of Zn (at 37 d) from red and white muscle. Both Zn-deficient MT+/+ and MT-/- mice had similar Zn concentrations in predominantly red muscle, one fourth lower than counterparts in Zn-replete mice. This is in agreement with an earlier finding in rats of a 19% lower Zn concentration in dry soleus (red) muscle from Zn-deficient rats compared with that of weight-restricted controls (O’Leary et al. 1979 ). Zn concentration from oxidative (red) muscle in Zn-deficient rodents therefore appears to be independent of MT, perhaps because most Zn is incorporated into carbonic anhydrase III (Jeffery et al. 1982 ), which has a greater binding affinity for Zn than MT. O’Leary and co-workers (1979) were unable to demonstrate a reduction in Zn concentration in white muscle from Zn-deficient rats, consistent with our finding in Zn-deficient MT+/+ mice. Zn deficient MT-/- mice, on the other hand, had significantly lower Zn concentrations in the predominantly white, more glycolytic quadriceps muscles than both Zn-replete MT-/- mice and Zn-deficient MT+/+ mice. All mice may also have lost a greater proportion of glycolytic muscle fibers in response to Zn depletion as suggested by the finding that accumulation of DNA and protein is slowed more in glycolytic than oxidative muscles in rats fed a Zn-deficient diet (Giugliano and Millward 1987 ). Nevertheless, our ratios of red abdominal muscle Zn:white quadriceps muscle Zn were within a narrow range from 2.5:1 to 2.9:1 in both Zn-replete and Zn-deficient MT-/- mice as well as in Zn-replete MT+/+ mice. This ratio was lowered (to 1.8:1) only in Zn-deficient MT+/+ mice, suggesting a MT-dependent transfer of Zn to, or preservation of Zn within glycolytic muscle. MT may thus assist in maintaining Zn in a form and concentration appropriate for the adequate metabolic function of glycolytic muscle.

Findings from this dietary Zn deficiency experiment may seem inconsistent with our recent report of a MT-dependent lowering of endogenous Zn secretion from pancreas and intestine (Rofe et al. 1999b ). However, that study was shorter term and used Zn-replete mice in a sensitive ⁶⁵Zn tracking procedure. Conditions were different in the Zn deficiency experiment, in which all mice became hypozincemic and had gross reductions in fecal Zn excretion. Increased catabolism in the MT-/- mice may also have resulted in their renal Zn loss taking precedence over gut loss. The conclusion is that the net effect of all MT-dependent influences on gut Zn loss is negligible during prolonged severe dietary Zn deficiency.

In summary, threefold or lower inductions of MT have no net conserving effect on the intestinal processing of Zn in mice, and reduce only the minor (<3%), nongut components of Zn excretion. The cumulative effect of increased urinary Zn excretion in MT-/- mice fed a Zn-deficient diet becomes important over time. The Zn-conserving action of MT at low concentration may derive from more efficient marshalling of tissue Zn resources for essential enzyme activity.

Conditions involving high (>30-fold in rodents) MT induction, including inflammation, infection and pregnancy, on the other hand, result in significant retention of Zn by the liver, and a gross reduction of intestinal Zn excretion, improving Zn uptake and markedly reducing Zn loss from the body. Cui and co-workers (1998) demonstrated up-regulation of MT-1 gene expression in response to interleukin-1 in Zn-deficient rats. Because chronic Zn undernutrition causes increased susceptibility to infection, and is often associated with a greater exposure to environmental pathogens, a resulting induction of MT is highly adaptive in terms of Zn balance and survival.

	FOOTNOTES

 
² AAS, atomic absorption spectrophotometry; LPS, lipopolysaccharide; MT, metallothionein

Manuscript received December 6, 1999. Initial review completed January 20, 2000. Revision accepted March 30, 2000.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Abe S., Matsumi M., Tsukioki M., Mizukawa S., Takahashi T., Iijima Y., Itano Y., Kosaka F. Metallothionein and zinc metabolism in endotoxin shock rats. Experientia 1987;52:587-593

2. Andrews G. K., Geiser J. Expression on the mouse metallothionein-I and -II genes provides a reproductive advantage during maternal dietary zinc deficiency. J. Nutr. 1999;129:1643-1648[Abstract/Free Full Text]

3. Beattie J. H., Wood A. M., Newman A. M., Bremner I., Choo K.H.A., Michalska A. E., Duncan J. S., Trayhurn P. Obesity and hyperleptinemia in metallothionein (-I and -II) null mice. Proc. Natl. Acad. Sci. U.S.A. 1998;95:358-363[Abstract/Free Full Text]

4. Beattie J. H., Wood A. M., Newman A. M., Bremner I., Choo K.H.A., Michalska A. E., Duncan J. S., Trayhurn P. Obesity and hyperleptinemia in a colony of metallothionein (-I and -II) null mice. Klaassen C. D. eds. Metallothionein IV 1999:505-510 Birkhauser Verlag Berlin, Germany.

5. Black B. L., Croom J., Eisen E. J., Petro A. E., Edwards C. L., Surwit R. S. Differential effects of fat and sucrose on body composition in A/J and C57BL/6 mice. Metabolism 1998;47:1354-1359[Medline]

6. Brand I. A., Kleineke J. Intracellular zinc movement and its effect on the carbohydrate metabolism of isolated rat hepatocytes. J. Biol. Chem. 1996;271:1941-1949[Abstract/Free Full Text]

7. Cousins R. J. Absorption, transport, and hepatic metabolism of copper and zinc: special reference to metallothionein and ceruloplasmin. Physiol. Rev. 1985;65:238-309[Free Full Text]

8. Coyle P., Philcox J. C., Rofe A. M. Corticosterone enhances the zinc and interleukin-6–mediated induction of metallothionein in cultured hepatocytes. J. Nutr. 1993;123:1464-1470

9. Coyle P., Philcox J. C., Rofe A. M. Hepatic zinc in metallothionein-null mice following zinc challenge: in vivo and in vitro studies. Biochem. J. 1995;309:25-31

10. Coyle P., Philcox J. C., Rofe A. M. Metallothionein-null mice absorb less Zn from an egg-white diet, but a similar amount from solutions, although with altered intertissue Zn distribution. J. Nutr. 1999;129:372-379[Abstract/Free Full Text]

11. Cui L., Takagi Y., Wasa M., Iiboshi Y., Inoue M., Khan J., Sando K., Nezu R., Okada A. Zinc deficiency enhances interleukin-1- induced metallothionein-1 expression in rats. J. Nutr. 1998;128:1092-1098[Abstract/Free Full Text]

12. Dalton T., Fu K., Palmiter R. D., Andrews G. K. Transgenic mice that overexpress metallothionein-1 resist dietary zinc deficiency. J. Nutr. 1996;126:825-833

13. De S. K., McMasters M. T., Andrews G. K. Endotoxin induction of murine metallothionein gene expression. J. Biol. Chem. 1990;265:15267-15274[Abstract/Free Full Text]

14. De S. K., Sandford T. H., Wood G. W. Detection of interleukin-6, and tumour necrosis factor in the uterus during the second half of pregnancy in the mouse. Endocrinology 1992;131:14-20[Abstract/Free Full Text]

15. De Lisle R. C., Sarras M. P., Jr, Hidalgo J., Andrews G. K. Metallothionein is a component of exocrine pancreas secretion: implications for zinc homeostasis. Am. J. Physiol. 1996;271:C1103-C1110[Abstract/Free Full Text]

16. Eaton D. L., Toal B. F. Evaluation of the Cd/hemoglobin assay for the rapid determination of metallothionein in biological tissues. Toxicol. Appl. Pharmacol. 1982;66:134-142[Medline]

17. Flanagan P. R., Haist J., Valberg L. S. Zinc absorption, intraluminal zinc and intestinal metallothionein in zinc-deficient and zinc-replete rodents. J. Nutr. 1983;113:962-972

18. Gaetke L. M., McClain C. J., Talwalkar R. T., Shedlofsky S. I. Effects of endotoxin on zinc metabolism in human volunteers. Am. J. Physiol. 1997;272:E952-E956[Abstract/Free Full Text]

19. Giugliano R., Millward D. J. The effects of severe zinc deficiency on protein turnover in muscle and thymus. Br. J. Nutr. 1987;57:139-155[Medline]

20. Hamer D. H. Metallothionein. Annu. Rev. Biochem. 1986;55:913-951[Medline]

21. Jeffery D., Edwards Y. H., Jackson M. J., Jeffery S., Carter N. D. Zinc and carbonic anhydrase III distribution in mammalian muscle. Comp. Biochem. Physiol. 1982;73:971-975

22. Kägi J.H.R., Kojima Y. Chemistry and biochemistry of metallothionein. Kägi J.H.R. Kojima Y. eds. Metallothionein II 1987:25-62 Birkhäuser. Verlag Basel, Switzerland.

23. Kelly E. J., Quaife C. J., Froelick G. J., Palmiter R. D. Metallothionein I and II protect against zinc deficiency and zinc toxicity in mice. J. Nutr. 1996;126:1782-1790

24. Kennedy K. J., Rains T. M., Shay N. F. Zinc deficiency changes preferred macronutrient intake in subpopulations of Sprague-Dawley outbred rats and reduces hepatic pyruvate kinase gene expression. J. Nutr. 1998;128:43-49[Abstract/Free Full Text]

25. Maret W., Jacob C., Vallee B. L., Fischer E. H. Inhibitory sites in enzymes: zinc removal and reactivation by thionein. Proc. Natl. Acad. Sci. U.S.A. 1999;96:1936-1940[Abstract/Free Full Text]

26. Michalska A., Choo K.H.A. Targeting and germ-line transmission of a null mutation at the metallothionein I and II loci in mouse. Proc. Natl. Acad. Sci. U.S.A. 1993;90:8088-8092[Abstract/Free Full Text]

27. O’Leary M. J., McClain C. J., Hegarty P.V. J. Effect of zinc deficiency on the weight, cellularity and zinc concentration of different skeletal muscles of the post-weanling rat. Br. J. Nutr. 1979;42:487-495[Medline]

28. Opara E. C., Petro A., Tevrizian A., Feinglos M. N., Surwit R. S. L-Glutamine supplementation of a high fat diet reduces body weight and attenuates hyperglycemia and hyperinsulinemia in C57BL/6J mice. J. Nutr. 1996;126:273-279

29. Pekarek R. S., Evans G. W. Effect of acute infection and endotoxemia on zinc absorption in the rat. Proc. Soc. Exp. Biol. Med. 1975;150:755-758[Medline]

30. Pekarek R. S., Evans G. W. Effect of leukocytic endogenous mediator (LEM) on zinc absorption in the rat. Proc. Soc. Exp. Biol. Med. 1976;152:573-574[Medline]

31. Philcox J. C., Coyle P., Michalska A., Choo C.H.A., Rofe A. M. Endotoxin-induced inflammation does not cause hepatic zinc accumulation in mice lacking metallothionein gene expression. Biochem. J. 1995;308:543-546

32. Quaife C., Hammer R. E., Mottet N. K., Palmiter R. D. Glucocorticoid regulation of metallothionein during murine development. Dev. Biol. 1986;118:549-555[Medline]

33. Richards M. P., Cousins R. J. Mammalian zinc homeostasis: requirements for RNA and metallothionein synthesis. Biochem. Biophys. Res. Commun. 1975;64:1215-1223[Medline]

34. Richards M. P., Cousins R. J. Metallothionein and its relationship to the metabolism of dietary zinc in rats. J. Nutr. 1976;106:1591-1599

35. Robinson A. M., Williamson D. H. Physiological role of ketone bodies as substrates and signals in mammalian tissues. Physiol. Rev. 1980;60:143-187[Free Full Text]

36. Rofe A. M., Barry E. F., Shelton T. L., Philcox J. C., Coyle P. Paracetamol hepatotoxicity in metallothionein-null mice. Toxicology 1998;125:131-140[Medline]

37. Rofe A. M., Philcox J. C., Coyle P. Trace metal, acute phase and metabolic response to endotoxin in metallothionein-null mice. Biochem. J. 1996;314:793-797

38. Rofe A. M., Philcox J. C., Sturkenboom M., Coyle P. Zinc homeostasis during pregnancy in metallothionein-null mice on a low zinc diet. Klaassen C. D. eds. Metallothionein IV 1999a:309-314 Birkhauser Verlag Berlin, Germany.

39. Rofe A. M., Williamson D. H. Metabolic effects of vasopressin infusion in the starved rat: reversal of ketonaemia. Biochem. J. 1983;212:231-239[Medline]

40. Rofe A. M., Winters N., Hinskens B., Philcox J. C., Coyle P. The role of the pancreas in intestinal secretion in metallothionein-null mice. Pancreas 1999b;19:69-75[Medline]

41. Surwit R. S., Feinglos M. N., Rodin J., Sutherland A., Petro A. E., Opara E. C., Kuhn C. M., Rebuffe-Scrive M. Differential effects of fat and sucrose on the development of obesity and diabetes in C57BL/6J and A/J mice. Metabolism 1995;44:645-651[Medline]

42. Tran C. D., Butler R. N., Philcox J. C., Rofe A. M., Howarth G. S., Coyle P. Regional distribution of metallothionein and zinc in the mouse gut: comparison with metallothionein-null mice. Biol. Trace Elem. Res. 1998;63:239-251[Medline]

43. Tukey J. W. Comparing individual means in the analysis of variance. Biometrics 1949;5:99-114

44. Wastney M. E., Angelus P. A., Barnes R. M., Subramanian K. N. Zinc absorption, distribution, excretion, and retention by healthy preterm infants. Pediatr. Res. 1999;45:191-196[Medline]

45. Yunice A. A., King R. W., Kraikitpanitch S., Haywood C. C., Lindeman R. D. Urinary zinc excretion following infusions of zinc sulfate, cysteine, histidine, or glycine. Am. J. Physiol. 1978;235:F40-F45[Medline]

This article has been cited by other articles:

				Abstract Section: 11th International Symposium on Trace Elements in Man and Animals Abstracts J. Nutr., May 1, 2003; 133(5): 203E - 282. [Full Text] [PDF]

				W. Waelput, D. Broekaert, J. Vandekerckhove, P. Brouckaert, J. Tavernier, and C. Libert A Mediator Role For Metallothionein in Tumor Necrosis Factor-induced Lethal Shock J. Exp. Med., December 3, 2001; 194(11): 1617 - 1624. [Abstract] [Full Text] [PDF]

				B. Ye, W. Maret, and B. L. Vallee Zinc metallothionein imported into liver mitochondria modulates respiration PNAS, February 8, 2001; (2001) 41619198. [Abstract] [Full Text]

				B. Ye, W. Maret, and B. L. Vallee Zinc metallothionein imported into liver mitochondria modulates respiration PNAS, February 27, 2001; 98(5): 2317 - 2322. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Philcox, J. C. Articles by Rofe, A. M. Search for Related Content PubMed PubMed Citation Articles by Philcox, J. C. Articles by Rofe, A. M.