Mobilization of Vitamin A Stores in Rats after Administration of 2,3,7,8-Tetrachlorodibenzo-p-dioxin: A Kinetic Analysis

Sean K. Kelley*,{dagger},1, Charlotte B. Nilsson*, Michael H. Green{dagger}, Joanne Balmer Green{dagger} and Helen Håkansson*,2

* The Institute of Environmental Medicine, Karolinska Institutet, P.O. Box 210, S-177 77 Stockholm, Sweden; and {dagger} Graduate Program in Physiology and Nutrition Department, Pennsylvania State University, University Park, Pennsylvania 16802 Part of this work was presented in the doctoral theses of Sean Keith Kelley ("The Effects of 2,3,7,8-Tetrachlorodibenzo-p-dioxin on Vitamin A Kinetics in Rats: A Compartmental Model," Pennsylvania State University, University Park, PA, 1997) and Charlotte B. Nilsson ("Studies on the Effects of 2,3,7,8-Tetrachlorodibenzo-p-dioxin on Vitamin A Homeostasis," Karolinska Institutet, Stockholm, Sweden, 1999).

Received November 9, 1999; accepted February 14, 2000


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
2,3,7,8-Tetrachlorodibenzo-p-dioxin (TCDD) is a highly toxic environmental contaminant that prevents the normal accumulation of vitamin A in liver and causes increased excretion of vitamin A. To determine what alterations in vitamin A metabolism occur first in response to TCDD treatment, we administered TCDD (7.0 µg/kg b.w.) orally to rats that had received a nonperturbing (tracer) iv dose of [3H]vitamin A-labeled plasma (n = 3) or lymph (n = 3) 21 days earlier. Within a few days after TCDD administration, fraction of the injected radiolabel in plasma, which had been in a terminal slope when plotted on a semilog scale, increased and remained elevated until the experiment was terminated (day 42). At that time, liver vitamin A levels were 65% lower in TCDD-perturbed rats than in controls. Using model-based compartmental analysis and compartmental models developed previously for control rats (S. K. Kelley et al., 1998, Toxicol. Sci, 44:1–13), we determined the minimal changes needed to account for the perturbation in plasma [3H] tracer responses after TCDD administration. We determined that the effects of TCDD could be explained by adjusting the value of one fractional transfer coefficient corresponding to the mobilization of vitamin A from large, slowly turning-over pools. We speculate that this change corresponds to an increased fractional rate of retinyl ester hydrolysis, and that it precedes the TCDD-associated increased irreversible utilization and excretion of vitamin A.

Key Words: 2,3,7,8-tetrachlorodibenzo-p-dioxin; TCDD; rat; vitamin A; model-based compartmental analysis; kinetics.


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Exposure to the persistent environmental contaminant 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) affects many physiologic functions, including vitamin A homeostasis (see reviews by Pohjanvirta and Tuomisto [1994] and Zile [1992]). Three compounds that show the biological activity of vitamin A are discussed in this paper (retinol, retinyl esters, and retinoic acid). In rats, the observable impacts of TCDD on vitamin A metabolism include a dose-dependent inhibition of vitamin A accumulation in liver, an increase in kidney vitamin A, an increase or no change in plasma retinol concentration, and increases in excretion of retinol or its metabolites (Bank et al., 1989Go; Håkansson and Ahlborg, 1985Go; Håkansson et al., 1991Go; van Birgelen et al., 1995Go). It is not yet known what underlying mechanisms are responsible for these changes, although increased vitamin A catabolism may be involved (Bank et al., 1989Go; Fiorella et al., 1995Go). In addition, the changes in retinol esterification activity seen in kidneys and liver perisinusoidal stellate cells from TCDD-treated rats may account for observed changes in renal and hepatic retinyl ester levels in these animals (Nilsson et al., 1996Go).

To investigate the sites and quantitative impacts of TCDD on vitamin A dynamics, we recently applied model-based compartmental analysis (Foster and Boston, 1983Go; Green and Green, 1990aGo) to data on the kinetics of [3H]vitamin A in rats treated with repeated oral doses of TCDD (Kelley et al., 1998Go). Our models predict that repeated exposure to TCDD causes a more rapid movement of vitamin A into and out of storage pools in the liver and possibly other tissues (presumably due to increases in esterification and hydrolysis of retinol), as well as an increase in the vitamin A disposal rate. We wanted to discern whether increased retinyl ester hydrolysis was driving the observed increase in vitamin A degradation, whether increased degradation led to a subsequent increase in retinyl ester hydrolysis, or whether both responses occurred simultaneously. To distinguish between these possibilities, we administered TCDD to perturb the kinetics of previously injected [3H]vitamin A in rats and observed the resultant changes in plasma tracer ([3H]) response. We hypothesized that if the initial effect of TCDD was to mobilize vitamin A from liver stores, this would be reflected by an increase in plasma tracer concentrations. If, on the other hand, increased degradation was the earliest effect, we would observe an initial fall in plasma tracer concentrations after TCDD treatment. By applying model-based compartmental analysis to data collected after the TCDD perturbation, we were able to identify and quantitate the kinetic processes that were initially affected by TCDD administration.


    MATERIALS AND METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Animals and diets.
Studies were carried out at the Karolinska Institute, Stockholm, Sweden; animal experiments were approved by the local committee on ethical animal experimentation. All procedures were done under light filtered through transparent films of titanium 35 (TESAB Solna AB, Solna, Sweden) to prevent the photo-oxidation of vitamin A.

This study was conducted at the same time as two related experiments that have been recently described in detail (Kelley et al., 1998Go), using extra rats that were not needed for that project. Animals described here were treated identically to control rats in the larger experiments (except for TCDD treatment) and analytical procedures were the same.

Weanling male Sprague-Dawley rats (n = 6; B&K Universal, Sollentuna, Sweden) were housed in shoebox cages and had free access to tap water and a commercial diet (catalogue # R34; Lactamin, Stockholm, Sweden) containing 4.2 µmol retinyl palmitate/kg for 6 to 7 weeks before initiation of kinetic studies. Assuming average daily food intakes of 15–20 g/day, rats consumed 63 to 84 nmol vitamin A/day, an amount that we correctly presumed would put rats in a slight positive vitamin A balance (Kelley et al., 1998Go).

Kinetic studies.
The experimental protocol is outlined in Figure 1Go. To overview, changes in plasma tracer concentration were monitored for 21 days after administration of tritiated vitamin A to six rats. At 21 days, when changes in plasma tracer concentrations had entered a terminal slope, TCDD was administered to perturb vitamin A kinetics, and plasma tracer concentrations were monitored for an additional 21 days. Compartmental analysis was then used to identify primary effects of TCDD on vitamin A metabolism.



View larger version (9K):
[in this window]
[in a new window]
 
FIG. 1. Design of kinetic studies. On day 0, [3H]vitamin A was administered as [3H]retinol-labeled plasma to three rats and as [3H]vitamin A-labeled lymph chylomicrons to three rats. On day 21, rats received an oral dose of 7 µg TCDD/kg of body weight; on days 28 and 35, the dose was 0.7 µg/kg. See Materials and Methods section for more details.

 
For kinetic studies, [3H]vitamin A was incorporated in vivo into its two physiologic plasma transporters (the retinol-binding protein [RBP]/transthyretin [TTR] complex and lymph chylomicrons) as described by Kelley et al. (1998) and Green and Green (1990b). The two carriers were used because in the 1998 study our aim was to model the kinetics of chylomicron (dietary) vitamin A before that vitamin A entered the holoRBP pool. For the current kinetic analysis, TCDD was administered so long after the tritiated vitamin A that the difference in carriers is of no consequence to our hypothesis. [3H]retinol-labeled plasma (~0.4 g containing ~180 kBq; group 1, n = 3) or [3H]vitamin A-labeled lymph (~0.25 g containing ~120 kBq; group 2, n = 3) was administered intrajugularly to nonfasting anesthetized rats and anesthesia (methoxyflurane; Pitman-Moore Inc., Mundelein, IL) was removed. For group 1, serial blood samples (n = 24; 0.1–0.25 ml) were collected from a caudal vein into heparinized tubes from 10 min to 21 days after administration of [3H]retinol-labeled plasma; for group 2, 26 samples were collected beginning 5 min after dosing until day 21. Aliquots of plasma were frozen under a nitrogen atmosphere for later analysis of tritium and, in some cases, retinol concentration (see below). Urine and feces were collected as described by Kelley et al. (1998) for 10 days after administration of [3H]retinol-labeled plasma in group 1 rats, and from the time of administration of [3H]vitamin A-labeled lymph chylomicrons until day 4 in group 2 rats, then from 17 to 20 days and 21 to 24 days. Rats were housed in shoebox cages except during these times. Excreta were frozen for later analysis of tritium (see below).

Twenty-one days after administration of [3H]vitamin A, an oral dose of 7.0 µg TCDD (provided by Dow Chemicals, Stockholm, Sweden) in 2 ml corn oil/kg body weight was administered to perturb vitamin A metabolism. In order to maintain a TCDD quasi-steady state as discussed by Flodström et al. (1991), rats also received two additional maintenance doses of TCDD (0.7 µg/kg body weight) on days 28 and 35. Blood was sampled as described above at 4, 8, 12, and 24 h, and 2 days (group 1 only) after TCDD administration, and then at 4-day intervals until day 42 of the study.

Forty-two days after administration of labeled plasma or lymph, rats were euthanized with CO2. Livers were excised, weighed, and frozen for later analysis (see below). The thymus glands were excised, weighed, and then returned to the carcass. Carcasses were weighed and frozen at –16°C until they were analyzed for radioactivity (see below).

Plasma and tissue analyses.
Plasma samples (40–200 µl) and aliquots of the vitamin A-labeled doses were analyzed for tritium (model 1409; Wallac Sverige AB, Upplands Väsby, Sweden) using Ecoscint A (National Diagnostics, Hintze, Stockholm, Sweden) as scintillation solution. Samples were counted twice to a final 2{sigma} error of 1%. After background correction, net counts/minute (cpm) were converted to disintegrations/min (dpm) using an external standard channels ratio method.

After addition of retinyl acetate (internal standard) to selected plasma samples, vitamin A was extracted (Kelley et al., 1998Go; Thompson et al., 1971Go) and analyzed by reverse phase HPLC using a Nucleosil 5µ C18 Resolve column (150 x 4.5 mm; Phenomenex, Torrance, CA) and methanol:water (90:10 v/v, 1 ml/min) as mobile phase. Retinol and retinyl acetate peaks were detected by UV absorbance at 328 nm (model 486; Waters Assoc., Milford, MA). Using an internal standard method, the retinol and retinyl acetate peak areas obtained from integration (MiniChrome v1.66; Fisons Instr., VG Data Systems, Cheshire, UK) were used, in conjunction with mass:area ratios for retinol and retinyl acetate standards, to determine the amount of retinol in samples.

Radioactivity in retinol and retinyl esters was determined in aliquots of the lymph chylomicron dose and in plasma samples collected during the first 120 min after administration of [3H]vitamin A to group 2 rats. Samples were extracted (Kelley et al., 1998Go; Thompson et al., 1971Go), and retinol and retinyl esters were separated on columns of deactivated aluminum oxide (Ross, 1982Go). Solvent-free fractions were analyzed for tritium using Ecoscint O (National Diagnostics).

Aliquots of freeze-dried liver were saponified and lipids were extracted into hexane (Green et al., 1985Go). Extracts were analyzed for tritium as described for plasma, using Ecoscint O as scintillation solution, and for vitamin A content by HPLC (see above). Tritium in carcasses was determined using the method of Adams et al. (1995). Aliquots of ground carcass were extracted using hexane:isopropanol:sodium sulfate. Solvent was allowed to evaporate, and solvent-free extracts were analyzed for tritium after addition of Ecoscint O.

The tritium content of urine was determined by liquid scintillation spectrometry after addition of Ecoscint A. Freeze-dried feces were ground and aliquots were extracted using methanol (Håkansson and Ahlborg, 1985Go, Kelley et al., 1998Go). Solvent-free extracts were analyzed for radioactivity after addition of Ecoscint O.

Kinetic analysis.
For each animal in the current study, we calculated fraction of the administered dose of tritiated vitamin A in plasma versus time, in liver and carcass at 42 days, and in pools of urine and feces as described by Kelley et al. (1998). Data on fraction of the dose in plasma from the time of administration of [3H]vitamin A until day 21 were fit to previously obtained models (Kelley et al., 1998; see Results) using the Simulation, Analysis and Modeling computer program (SAAM31; Berman and Weiss, 1978) and its conversational version (CONSAM; Berman et al., 1983). The model parameters or fractional transfer coefficients (L[I,J]s, the fraction of compartment J's tracer transferred to compartment I per day) for each rat were determined by weighted nonlinear regression analysis in CONSAM using a fractional standard deviation (FSD) of 5% as weighting factor. These L(I,J)s were then fixed for each rat. To accommodate the perturbation of the existing vitamin A system due to administration of TCDD on day 21, we added a time-interrupt function into each input file just prior to the time at which a change in plasma radioactivity was observed. Then the appropriate model L(I,J)s were made adjustable one at a time in order to identify the minimal change(s) needed to fit data obtained after the TCDD perturbation. Finally, mean L(I,J)s ± SD were calculated for each group.

Statistical analysis.
Descriptive data and kinetic parameters are presented as arithmetic group mean ± 1 standard deviation. Using Minitab (Ryan et al., 1985Go), a standardized paired t-test with an alpha level of 0.05 was used to test for significant effects of TCDD treatment.


    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Descriptive and Kinetic Data
Body weights for group 1 rats (n = 3) averaged 393 ± 31 g at the beginning (14 weeks of age) and 484 ± 62 g at the end of the kinetic study (17 weeks of age); for group 2, corresponding values were 400 ± 34 and 530 ± 12 g. As shown in Table 1Go, body weight gain from days 21 to 42 and relative liver weights were significantly affected by TCDD treatment in group 1 rats compared to controls studied concurrently by Kelley et al. (1998); these parameters were not significantly affected in group 2 rats. Relative thymus weights and plasma retinol concentrations were not significantly affected by the TCDD perturbation. In both groups, liver vitamin A content and hepatic recovery of radioactivity were significantly lower in TCDD-treated rats than in controls on day 42: liver vitamin A levels were 65% lower and recovery of label was 54% lower in TCDD-treated rats in both groups. Fraction of dose recovered in carcass was not affected by TCDD in either group. For group 2 rats, excretion of radioactivity during the 4 days after TCDD administration amounted to 1.35 ± 0.19% of the dose in urine and 2.29 ± 0.15% in feces, significantly higher than the amounts excreted during the 4 days before TCDD administration (0.48 ± 0.02% in urine and 0.83 ± 0.07% in feces).


View this table:
[in this window]
[in a new window]
 
TABLE 1 Body and Organ Weights, Plasma and Liver Vitamin A Levels, and Recovery of Radioactivity in Liver and Carcass in Control and TCDD-Perturbed Rat
 
Plasma tracer response curves for two representative rats are shown in Figure 2Go. The curves for tracer response versus time had reached a terminal slope by ~ day 15 after administration of [3H]vitamin A, and are similar during the first 21 days to those for control rats studied concurrently (Kelley et al., 1998Go). When TCDD was administered on day 21, there were no obvious perturbations in plasma tracer concentrations during the first 12 h when we had expected to see changes. However, by 6 days after TCDD treatment, plasma tracer concentrations increased in all rats and remained elevated until the end of the experiment (day 42). These data indicate that TCDD administration on day 21 caused an increased mobilization of tracer into plasma, after a lag time of several days. In general, effects of TCDD on plasma tracer response profiles were less dramatic and occurred later in group 2 than group 1. In both groups, the maintenance doses of TCDD on days 28 and 35 did not cause observable changes in plasma tracer concentrations. We note that the control rats described by Kelley et al. (1998) received a weekly intubation of corn oil (1 ml/kg body weight) when TCDD-treated rats in that study received doses of TCDD in corn oil. As we observed no effect of corn oil administration on plasma [3H] tracer responses in that study, we were able to eliminate the possibility that the perturbations in plasma tracer levels in the current experiment were due to the corn oil vehicle.



View larger version (14K):
[in this window]
[in a new window]
 
FIG. 2. Working compartmental models developed to fit tracer data from plasma, carcass, irreversible loss (1-[fraction of dose in plasma + liver + carcass]), and excreta (urine and feces) of control rats injected with [3H]retinol-labeled plasma (left panel) or [3H]vitamin A-labeled lymph (right panel) by Kelley et al. (1998). Circles represent compartments; the rectangle is a delay component. Interconnectivities are the adjustable model parameters (fractional transfer coefficients [L(I,J)s] or the fraction of compartment J's tracer transferred to compartment I each day). Values for the L(I,J)s are for control rats in Kelley et al. (1998). The asterisk indicates the site of tracer introduction and U(I) represents dietary input into the system. In the left panel, compartment 1 represents plasma retinol and compartments 2–4 are extravascular pools of vitamin A; compartments 31 (urine) and 32 (feces) are sites of tracer output. In the right panel, compartments 5, 7, 9, and 11, and delay component 6 were used to fit data for plasma tracer disappearance and hepatic processing of chylomicron vitamin A. Compartment 1 represents plasma retinol and compartments 2 and 3 are extravascular vitamin A pools. Plasma tracer data collected until day 21 for rats in group 1 in the current study were fit to the model in the left panel; data for group 2 rats were fit to the model in the right panel.

 
Model-Based Compartmental Analysis
Model-based compartmental analysis (Foster and Boston, 1983Go; Green and Green, 1990aGo) was used to identify kinetic parameters describing vitamin A metabolism, by fitting the data obtained in the current study to models developed by Kelley et al. (1998) for data collected concurrently from two groups of control rats given the same preparations of [3H]vitamin A. In the model developed by Kelley et al. (1998) for control rats (n = 7) injected with [3H]retinol/RBP/TTR (Fig. 2Go, left), plasma retinol (compartment 1, the site of tracer introduction) exchanges with vitamin A in both a fast and slowly turning-over pool (compartments 2 and 3). In addition, vitamin A in compartment 3 exchanges with a very slowly turning-over pool (compartment 4). System input (dietary vitamin A) was modeled into compartment 1 and output to urine and feces was via compartment 3. We postulate that vitamin A in compartment 2 includes rapidly turning-over intracellular retinol and retinol in interstitial fluid, as well as vitamin A that has been filtered by the kidney and is in the process of being reabsorbed. We postulate that compartments 3 and 4 include vitamin A in retinyl ester-containing storage pools, primarily in the liver. The model developed by Kelley et al. (1998) for rats that received [3H]vitamin A-labeled lymph (n = 8; Fig. 2Go, right) shows more initial complexity related to the plasma clearance (compartments 5 and 11) and hepatic processing of newly absorbed (chylomicron) vitamin A (compartments 7 and 9 and delay element 6). This model predicts that after hepatic processing, retinol is secreted into plasma compartment 1 bound to RBP/TTR, where it exchanges with vitamin A in both rapidly and slowly turning-over pools (compartments 2 and 3, respectively). Addition of a second slowly turning-over pool (compartment 4) was not statistically justified, but otherwise the interconnections among compartments 1, 2, and 3, and their postulated physiologic significance, are the same as in the first model.

As is evident in Figure 3Go, these models provide a good fit to observed plasma tracer data for rats in the current study during the first 21 days following administration of [3H]vitamin A. Model-derived fractional transfer coefficients [L(I,J)s] for the six rats in the current study (Table 2Go) were very similar (Fig. 2Go) to those determined for the larger groups of control rats studied for 42 days (Kelley et al. 1998Go).



View larger version (8K):
[in this window]
[in a new window]
 
FIG. 3. Plasma tracer response profiles for one representative rat from each group. Data are fraction of injected [3H] remaining in plasma versus time after injection of [3H]retinol-labeled plasma (group 1; left panel) or [3H]vitamin A-labeled lymph chylomicrons (group 2; right panel). A perturbing dose of TCDD (7 µg/kg) was administered on day 21. Symbols correspond to observed data; lines are the responses predicted by the compartmental models shown in Figure 2Go using the fractional transfer coefficients shown in Table 2.Go

 

View this table:
[in this window]
[in a new window]
 
TABLE 2 Model-Predicted Fractional Transfer Coefficient [L(I,J)s] for Proposed Model of Vitamin A Metabolism in TCDD-Perturbed Rats
 
We then fixed these L(I,J)s for each rat and introduced a time-interrupt function in the SAAM input files following the administration of TCDD on day 21. The time and minimal adjustments needed in each rat's model to explain the plasma tracer data collected after day 21 were determined. For both groups, adjusting the value of a single L(I,J) was necessary and sufficient to describe the postperturbation response. Specifically, in group 1, an increase in L(3,4) from 0.0482 to 0.2548/day (Table 2Go) was sufficient to explain the effects on tracer kinetics after day 21. This change corresponds to more than a 4-fold increase in the mobilization of tracer from very slowly turning-over compartment 4 to storage compartment 3. When, during data fitting, L(4,3) was allowed to decrease while L(3,4) increased, a slightly better fit (based on visual examination) to plasma tracer data was obtained. However, addition of the extra parameter was not statistically justified. The result of an abrupt increase in L(3,4) was not immediately observed in the simulated plasma tracer response curves, as compartment 3 acted as an intermediate (buffer) pool, leading to a gradual rise in plasma tracer concentrations. For group 2, L(1,3) was the parameter that was most sensitive to TCDD exposure. An increase in L(1,3) from 0.103 to 0.241/day (Table 2Go) was sufficient to explain the changes in plasma tracer response after TCDD. This adjustment corresponds to nearly a 1.3-fold increase in the mobilization of tracer from the slowly turning-over vitamin A pool (compartment 3) into plasma. In contrast to the case for group 1 rats, the increase in L(1,3) caused an abrupt increase in the simulated plasma tracer responses, as there was no intermediate pool to buffer the response.


    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
As has been shown in other studies (Kelley et al., 1998, and references therein; see review by Zile, 1992), TCDD treatment caused a substantial depletion in hepatic vitamin A stores (Table 1Go), and an increased excretion of radioactivity in urine and feces. In the current study, in which rats received three weekly doses of TCDD (7.0, 0.7, and 0.7 µg/kg body weight) beginning 21 days after administration of labeled vitamin A, liver vitamin A levels were 65% lower, and hepatic recovery of radioactivity was 54% lower, than in control rats studied concurrently (Kelley et al., 1998Go). In that study, liver vitamin A levels were 98% lower, and hepatic recovery of radioactivity was 95–98% lower in TCDD-treated rats given a loading dose of TCDD (3.5 µg/kg) 9 days before tracer administration, and then seven weekly maintenance doses (0.7 µg/kg) during the 42-day kinetic study, compared to control rats. Those more severe effects confirm results of other studies, which showed that many smaller doses of TCDD have a greater effect on liver vitamin A levels than a smaller number of larger ones, even if the total dose is lower (Håkansson et al., 1991Go; van Birgelen et al., 1995Go).

The fact that plasma tracer levels increased rather than decreased following TCDD administration (Fig. 3Go) indicates that the initial effect of TCDD was to mobilize vitamin A from body sites that had appreciable amounts of radioactivity, rather than to increase degradation of vitamin A. The initial effect seemed to be followed by an increase in vitamin A degradation, as evidenced by the increased excretion of radioactivity during the 4 days following TCDD administration in group 2 rats.

Because the increased levels of radioactivity in plasma did not occur until several days after TCDD administration (Fig. 3Go), we hypothesize that the increased mobilization of vitamin A from body sites with appreciable amounts of radioactivity occurs secondary to an alteration in protein synthesis and/or degradation. In support of this idea, it is known that TCDD induces numerous liver enzymes (see review by Pohjanvirta and Tuomisto, 1994). For example, greatly increased expression of cytochrome P450 1A1 and 1A2 is already evident 24 h after TCDD administration (Santostefano et al., 1998Go), indicating rapid enzyme induction by TCDD. Several lines of evidence indicate that members of the cytochrome P450 family are involved in the catabolism of retinoic acid (Spear et al., 1988Go; Sonneveld et al., 1998Go). Thus, induction of cytochrome P450 enzymes by TCDD may result in altered vitamin A catabolism.

Using compartmental analysis, we determined which model parameter(s) (L[I,J]s) were most sensitive to TCDD treatment. For rats that received [3H]retinol-labeled plasma (group 1), L(3,4) (Fig. 2Go, left), the fractional transfer of label to compartment 3 from compartment 4, was the parameter most sensitive to TCDD treatment. By adjusting this one parameter, we obtained a good fit to plasma tracer data obtained after TCDD administration (Fig. 3Go, left). The model predicts a 428% increase in L(3,4) after TCDD treatment (Table 2Go). We speculate that compartment 4 represents large, slowly turning-over pools, predominantly retinyl esters in hepatic perisinusoidal stellate cells. Corresponding to the increase in L(3,4), it seems likely that the increased radioactivity in plasma after TCDD administration probably originated from labeled retinyl esters in these slowly turning-over vitamin A pools, as the model predicts that these sites contain enough tracer to cause such a sustained perturbation. Physiologically, mobilization of vitamin A from compartment 4 would represent an increased fractional rate of retinyl ester hydrolysis. A similar conclusion can be drawn from the model developed for rats that received labeled vitamin A in chylomicrons (group 2), in which a 133% increase in L(1,3) (Table 2Go and Fig. 2Go, right) was sufficient to explain the effect of TCDD on plasma tracer response. Here, compartment 3 is presumably dominated by large vitamin A stores in the hepatic perisinusoidal stellate cells.

As it has previously been reported that TCDD and dioxin-related substances either inhibit or have no effect on various hepatic retinyl ester hydrolase activities (Chen et al., 1992; Jensen et al., 1987; Powers et al., 1987; Mercier et al., 1990; Nilsson et al., submitted for publication), it may be that nonspecific carboxyl ester hydrolase activities are responsible for the increased mobilization of vitamin A stores after TCDD treatment. Further work is needed to identify which enzyme is responsible for the increased mobilization and to determine if this is the same enzyme involved in normal retinyl ester hydrolysis.

When both L(3,4) and L(4,3) were allowed to vary when fitting the post-TCDD plasma tracer data for group 1 [with L(3,4) increasing and L(4,3) decreasing], a visually better fit was obtained. Although including the second parameter was not statistically justified, this result may indicate that there is an additional, secondary factor contributing to the response observed after TCDD exposure. Biologically, L(4,3) is thought to represent the esterification of retinol by lecithin:retinol acyl transferase (LRAT) for storage in the slow turning-over stellate cell pool. Our observation supports the results of Nilsson et al. (1996), which showed that TCDD reduces hepatic perisinusoidal stellate cell LRAT activity. In contrast, in rats undergoing a rapid depletion in liver vitamin A levels due to TCDD treatment (Kelley et al., 1998Go), while L(4,3) increased compared with controls, the estimated rate of transfer [L(4,3)* mean liver vitamin A level] was less than 50% of that in control rats (253 versus 580 nmol/day in TCDD-treated versus controls).

It is worth noting that the fractional transfer coefficient describing irreversible loss in group 2 (L[0,3]) did not need to change following TCDD treatment, indicating that the initial effect of TCDD was not to increase the fraction of compartment 3 being excreted per day. Rather, the observed increased tracer excretion in group 2 was caused by a mass action effect (i.e., an increase in the pool in compartment 3, from which system output was modeled). A similar effect was predicted for group 1; no change was needed in the fractional transfer coefficients involved in irreversible loss, indicating that the predicted increase in excretion of tracer was due to a mass action consequence of the increased mobilization of vitamin A stores.

In conclusion, this work illustrates the usefulness of model-based compartmental analysis in discriminating between possible mechanisms by which a system perturbation affects isotope kinetics. Our results indicate that the major impact of TCDD on vitamin A dynamics is to increase mobilization of vitamin A from storage sites, most likely liver perisinusoidal stellate cells; subsequently, vitamin A degradation increases. Because the effect of TCDD on plasma tracer concentrations was not evident for several days after TCDD administration, it seems likely that the increased mobilization is secondary to changes in protein levels or other effects of TCDD on liver parenchymal- and/or stellate cells.


    ACKNOWLEDGMENTS
 
This project was supported by funds from the Swedish Environmental Protection Agency, Visiting Scientist grants from the Karolinska Institute, and a Fulbright Fellowship to S.K.K. We thank Ellu Manzoor and Christina Trossvik for their generous technical contributions to this project, and Dr. Annika Hanberg and Christina Trossvik for performing lymph duct cannulations.


    NOTES
 
1 Current address: Genetech, Inc., 1 DNA Way, South San Francisco, CA 94080. Back

2 To whom correspondence should be addressed. Fax: 46 8 33 44 67. E-mail: helen.hakansson{at}imm.ki.se. Back


    REFERENCES
 TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Adams, W. R., Smith, J. E., and Green, M. H. (1995). Effects of N-(4-hydroxyphenyl)retinamide on vitamin A metabolism in rats. Proc. Soc. Exp. Biol. Med. 208, 178–185.[Abstract]

Bank, P. A., Salyers, K. L., and Zile, M. H. (1989). Effect of tetrachlorodibenzo-p-dioxin (TCDD) on the glucuronidation of retinoic acid in the rat. Biochim. Biophys. Acta 993, 1–6.[ISI][Medline]

Berman, M., Beltz, W. F., Greif, P.C., Chabay, R., and Boston, R. C. (1983). CONSAM User's Guide. U.S. Govt. Printing Office, Washington, DC. PHS Publ. 1983–421–132: 3279.

Berman, M., and Weiss, M. F. (1978). SAAM Manual. U.S. Govt. Printing Office, Washington, DC. DHEW Publ. 78–180.

Chen, L.-C., Berberian, I., Koch, B., Mercier, M., Azais-Braesco, V., Glauert, H. P., Chow, C. K., and Robertson, L. W. (1992). Polychlorinated and polybrominated biphenyl congeners and retinoid levels in rat tissues: structure-activity relationships. Toxicol. Appl. Pharmacol. 114, 47–55.[ISI][Medline]

Fiorella, P. D., Olson, J. R., and Napoli, J. L. (1995). 2,3,7,8-Tetrachlorodibenzo-p-dioxin induces diverse retinoic acid metabolites in multiple tissues of the Sprague-Dawley rat. Toxicol. Appl. Pharmacol. 134, 222–228.[ISI][Medline]

Flodström, S., Busk, L., Kronevi, T., and Ahlborg, U. G. (1991). Modulation of 2,3,7,8-tetrachlorodibenzo-p-dioxin and phenobarbital-induced promotion of hepatocarcinogenesis in rats by the type of diet and vitamin A deficiency. Fundam. Appl. Toxicol. 16, 375–391.[ISI][Medline]

Foster, D. M., and Boston, R. C. (1983). The use of computers in compartmental analysis: The SAAM and CONSAM programs. In Compartmental Distribution of Radiotracers (J. S. Robertson, Ed.), pp. 73–142. CRC Press, Boca Raton, FL.

Green, M. H., and Green, J. B. (1990a). The application of compartmental analysis to research in nutrition. Annu. Rev. Nutr. 10, 41–61.[ISI][Medline]

Green, M. H., and Green, J. B. (1990b). Experimental and kinetic methods for studying vitamin A dynamics in vivo. Methods Enzymol. 190, 304–317.[Medline]

Green, M. H., Uhl, L., and Green, J. B. (1985). A multicompartmental model of vitamin A kinetics in rats with marginal liver vitamin A stores. J. Lipid Res. 26, 806–818.[Abstract]

Håkansson, H., and Ahlborg, U. G. (1985). The effect of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) on the uptake, distribution and excretion of a single oral dose of [11, 12–3H]retinyl acetate and on the vitamin A status in the rat. J. Nutr. 115, 759–771.[ISI][Medline]

Håkansson, H., Manzoor, E., and Ahlborg, U. G. (1991). Interaction between dietary vitamin A and single oral doses of 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) on the TCDD-induced toxicity and on the vitamin A status in the rat. J. Nutr. Sci. Vitaminol. 37, 239–255.[ISI][Medline]

Jensen, R. K., Cullum, M. E., Deyo, J., and Zile, M. H. (1987). Vitamin A metabolism in rats chronically treated with 3,3`,4,4`,5,5`-hexabromobiphenyl. Biochim. Biophys. Acta 926, 310–320.[ISI][Medline]

Kelley, S. K., Nilsson, C. B., Green, M. H., Green, J. B., and Håkansson, H. (1998). Use of model-based compartmental analysis to study effects of 2,3,7,8-tetrachlorodibenzo-p-dioxin on vitamin A kinetics in rats. Toxicol. Sci. 44, 1–13.[Abstract]

Mercier, M., Pascal, G., and Azais-Braesco, V. (1990). Retinyl ester hydrolase and vitamin A status in rats treated with 3,3`,4,4`-tetrachlorobiphenyl. Biochim. Biophys. Acta 1047, 70–76.[ISI][Medline]

Nilsson, C. B., Hanberg, A., Trossvik, C., and Håkansson, H. (1996). 2,3,7,8-Tetrachlorodibenzo-p-dioxin affects retinol esterification in rat hepatic stellate cells and kidney. Environ. Toxicol. Pharmacol. 2, 17–23.[ISI]

Pohjanvirta, R., and Tuomisto, J. (1994). Short-term toxicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin in laboratory animals: effects, mechanisms, and animal models. Pharmacol. Rev. 46, 483–549.[ISI][Medline]

Powers, R. H., Gilbert, L. C., and Aust, S. D. (1987). The effect of 3,4,3`,4`-tetrachlorobiphenyl on plasma retinol and hepatic retinyl palmitate hydrolase activity in female Sprague-Dawley rats. Toxicol. Appl. Pharmacol. 89, 370–377.[ISI][Medline]

Ross, A. C. (1982). Retinol esterification by rat liver microsomes. Evidence for a fatty acyl coenzyme A: retinol acyltransferase. J. Biol. Chem. 257, 2453–2459.[Abstract/Free Full Text]

Ryan, B. F., Joiner, B. L., and Ryan, T. A., Jr. (1985). Minitab Handbook, 2nd Ed. PWS, Boston, MA.

Santostefano, M. J., Wang, X., Richardson, V. M., Ross, D. G., De Vito, M. J., and Birnbaum, L. S. (1998). A pharmacodynamic analysis of TCDD-induced cytochrome P450 gene expression in multiple tissues: dose- and time-dependent effects. Toxicol. Appl. Pharmacol. 151, 294–310.[ISI][Medline]

Sonneveld, E., van den Brink, C. E., van der Leede, B. M., Schulkes, R-K., Petkovich, M., van der Burg, B. and van der Saag, P. T. (1998). Human retinoic acid (RA) 4-hydroxylase (CYP26) is highly specific for all-trans-RA and can be induced through RA receptors in human breast and colon carcinoma cells. Cell Growth Differ. 9, 629–637.[Abstract]

Spear, P. A., Garcin, H. and Narbonne, J-F. (1988). Increased retinoic acid metabolism following 3,3`,4,4`,5,5`-hexabromobiphenyl injection. Can. J. Physiol. Pharmacol. 66, 1181–1186.[ISI][Medline]

Thompson, J. N., Erdody, P., Brien, R., and Murray, T. K. (1971). Fluorometric determination of vitamin A in human blood and liver. Biochem. Med. 5, 67–89.[ISI][Medline]

van Birgelen, A. P., van der Kolk, J., Fase, K. M., Bol, I., Poiger, H., Brouwer, A., and van den Berg, M. (1995). Subchronic dose-response study of 2,3,7,8-tetrachlorodibenzo-p-dioxin in female Sprague-Dawley rats. Toxicol. Appl. Pharmacol. 132, 1–13.[ISI][Medline]

Zile, M. H. (1992). Vitamin A homeostasis endangered by environmental pollutants. Proc. Soc. Exp. Biol. Med. 201, 141–153.[Abstract]