Physiological Interpretation Of Three And Fourcompartment Models

Although large, complex models such as the ones developed by Green et al. (1985) and Lewis et al. (1990) are important to our understanding of whole-body vitamin A metabolism, three- and four-compartment models that view the vitamin A system from the plasma space can be developed from more manageable experiments and with less extensive modeling. Of course, with the latter approach, the investigator is not able to postulate definitive correspondence between anatomical sites and the model's compartments as is possible in a large model developed after sampling multiple tissues as well as plasma. However, this limitation does not negate the usefulness of the more straightforward approach.

In the studies discussed here which use this approach, either a three-(Fig. 2) or four-compartment model (Fig. 10) has provided a good fit to vitamin A kinetic data. Compartment 1 is clearly the plasma pool of retinol bound to RBP and transthyretin. It is worth emphasizing that plasma retinol acts as one kinetically homogeneous pool with a mean transit time (Table I) of about 2 h. In the postabsorptive state after a vitamin A-rich meal, there would be a transient increase in plasma retinyl esters. These would be in a kinetically distinguishable pool from compartment 1. On the basis of kinetics, we hypothesize that compartment 2 represents retinol that has either entered interstitial fluid or been filtered by the kidneys as holoRBP; it also likely includes a relatively rapidly turning-over intracellular pool of retinol in some of the organs shown in Fig. 9 (e.g., carcass).

When liver vitamin A is not changing substantially during the course of a kinetic study, then plasma tracer data can often be fit to a three-compartment model (Fig. 2), with the majority of compartment 3 corresponding to liver retinyl ester stores. However, when liver vitamin A balance is negative, such as when TCDD was administered (Green and Green, 2003) or when liver vitamin A balance was positive as in the case of iron deficiency (Jang et al., 2000), or in the LPS-induced inflammation study (Gieng, 2006), then a fourth compartment is required to fit the tracer and tracee data (liver vitamin A) (Figs. 7 and 10). The kinetic behavior of vitamin A in compartment 3 in the three-compartment model versus compartment 4 in the four-compartment model is very similar; that is, all of the slowest turning-over vitamin A lumps into one compartment (compartment 3) in a steady state condition. When liver vitamin A is changing substantially over time (i.e., is not in a steady state), then compartment 4 is needed as liver vitamin A decreases or increases compartment 3; compartment 4 contains sufficient tracer to feed the plasma retinol tracer compartment as liver tracee is depleted or expands. This was especially important in the inflammation study, in which the model predicted that during inflammation, mobilization of retinol from liver retinyl ester stores was inhibited, likely due to unavailability of apoRBP. The recycling of retinol from compartment 4 and to a lesser extent from compartment 2 prevented plasma retinol concentration from dropping even further into a vitamin A-deficient-like state. Thus, compartment 4 in Fig. 10 plays an important role in plasma retinol homeostasis. It will be interesting in future kinetic studies to determine the location of compartment 4 vitamin A. Likely candidates for this extrahepatic pool of kinetically active vitamin A that plays an important role in whole-body vitamin A metabolism are adipose tissue, small intestine, and skin. Future research will be needed to identify the source(s) of apoRBP involved in vitamin A turnover from this pool (plasma versus the organs themselves).

0 0

Post a comment