The Journal of Nutrition Vol. 128 No. 7 July 1998, pp. 1179-1185

Tissue Stores of -Carotene Are Not Conserved for Later Use as a Source of Vitamin A during Compromised Vitamin A Status in Mongolian Gerbils (Meriones unguiculatus)^1,2,3

Angela J. Thatcher^*, Christine M. Lee^*, and John W. Erdman Jr.^{, 4}

^* Department of Food Science and Human Nutrition,  Division of Nutritional Sciences, University of Illinois, Urbana, IL 61801

	ABSTRACT

Abstract Introduction Methods Results Discussion References

Vitamin A (VA) deficiency remains a serious problem in the world today. Current approaches to preventing or treating VA deficiency, including dietary intervention with provitamin A compounds, rely on the body converting ingested -carotene (C) to VA. However, it is not known whether C that is already in the tissues can be used as a source of VA to prevent deficiency. The objectives of these studies were to determine whether tissue C stores are converted to VA when the Mongolian gerbils have low VA status and whether previously fed C is retained in the tissues for later conversion to VA. In the first study, gerbils were prefed diets with C (20.3 ± 6.2 nmol/g diet) (+C) or without C (C), and with VA [2.4 ± 1.5 nmol/g diet (+C diet) or 12.0 ± 4.2 nmol/g diet (C diet)] for 7 d, and then depleted of both C and VA for up to 84 d. On d 0 after the prefeeding period, hepatic C stores were 13.3 ± 9.1 nmol. These stores were significantly lower after 28d of consuming the VA/C diet (2.16 ± 1.7 nmol), even though the hepatic VA concentrations did not change. In the second study, the gerbils were prefed a VA/+C diet (74.3 ± 19.7 nmol C/g diet) for 7 d, and then fed a C-free diet either with (7.1 ± 1.4 nmol/g) or without VA for up to 34 d. Hepatic C stores after the 7-d prefeeding period were 38.1 ± 20.6 nmol, and were significantly higher than after 7d of consuming either a +VA/C (12.4 ± 10.8 mmol) or VA/C diet (11.4 ± 8.0 nmol). The results from both studies suggest that a substantial amount of hepatic C is rapidly lost when C is eliminated from the diet and therefore is not conserved to meet later VA needs. The presence of VA in the diet (Study 2) did not affect the rate of C loss from the serum and tissues. Moreover, no evidence was found that the stored C was utilized for VA. The data suggest that there may be two pools of hepatic C, one that is lost rapidly and another that is lost more slowly over time, but losses are not affected by VA status.

	INTRODUCTION

Abstract Introduction Methods Results Discussion References

It has been estimated that over 124 million preschool children worldwide are vitamin A (VA)⁵ deficient (Humphrey et al. 1992). Improving VA status is expected to prevent 1.3-2.5 million of the 8 million deaths in children <5 y old who live in countries with a high risk for developing VA deficiency (Humphrey et al. 1992). Providing a diet high in fruits and vegetables to people with low VA status increases the concentration of VA in the serum (Wadhwa et al. 1994). Pro-VA compounds such as -carotene (C) may account for up to 100% of the dietary VA value in some developing countries (Rodriguez and Irwin 1972). Because of the current emphasis on increased fruit and vegetable intake and the importance of pro-VA carotenoids in many populations, it is necessary to understand more about the potential ability of tissue C to meet VA needs. Little is known about the storage or accumulation of C in the body, and it is not known if C that is stored or accumulated in the liver and other tissues can be cleaved to provide VA during periods of low VA status. If tissue C cannot improve VA status, other approaches such as VA supplementation or fortification may be more beneficial for improving and maintaining sufficient VA status, especially during periods of low food availability.

The Mongolian gerbil (Meriones unguiculatus) has been used recently to study C and VA absorption and metabolism. The Mongolian gerbil absorbs C intact (Pollack et al. 1994); other gerbil studies in our laboratory have shown that the gerbil converts C to VA at a ratio similar to that of humans and may be an appropriate animal model for C utilization studies (Lee et al. 1998).

The goal of this research was to evaluate the efficacy of tissue C in serving as a source of VA when VA is absent from the diet. The objectives of the two studies were to determine whether previously fed C is retained in the tissues to be utilized for future VA needs and whether previously fed C found in the tissues can be converted to VA for use during mild VA deficiency.

	MATERIALS AND METHODS

Abstract Introduction Methods Results Discussion References

Animals.  Male Mongolian gerbils (n = 105, 19-21 d old and n = 128, 26-32 d old for Studies 1 and 2, respectively) were obtained from Tumblebrook Farms (Brant Lake, NY) for Study 1 and Harlan Sprague Dawley (Indianapolis, IN) for Study 2. All animals had been fed a commercial diet postweaning by the breeder. Upon arrival, gerbils were individually housed in plastic shoe box cages with beta chip bedding. They had free access to food and water. Room temperature was constant and lighting was provided on a diurnal cycle of 12 h light:12 h dark. All animal handling procedures were approved by University of Illinois Laboratory Animal Care Advisory Committee.

Diets.  Gerbils were fed a semipurified diet (Table 1) prepared in our laboratory from ingredients purchased from Harlan Teklad (Madison, WI). This diet has been successfully used for feeding studies with gerbils (Pollack et al. 1994) and provides adequate nutrition for proper growth. The diets were supplemented with C (10% in water-soluble beadlets) and/or VA provided as retinyl palmitate (Palmabeads), both gifts from Hoffmann LaRoche (Nutley, NJ). Diets were stored at 4°C until use.

Experimental design.  Study 1. Upon arrival, all gerbils (n = 105) were fed a powdered VA/C diet for 7 d to allow acclimation to the new environment (Fig. 1A). Gerbils (n = 105) were then fed the powdered diet (VA/C) for 7 more days, and 15 were killed, as described below, to determine baseline VA and C levels. The remaining gerbils were randomly assigned to one of two groups fed either +VA/+C (2.4 ± 1.5 nmol VA/g diet, 20.3 ± 6.2 nmol C/g diet) or +VA/C (12.0 ± 4.2 nmol VA/g diet) for 7 d. The goal was for the groups to achieve similar hepatic VA stores at the beginning of the study; thus the VA content of the +VA/+C prefeeding diet was reduced relative to the +VA/C diet to account for the conversion of the dietary C to VA. After the 7-d prefeeding period (d 0), 15 gerbils were killed to determine VA and C concentrations in serum and tissues. The remaining gerbils from both groups were fed a VA/C diet for 28 or 84 d. Fifteen gerbils from each group were killed at each time and VA and C concentrations in the serum and tissues were measured.

View larger version (17K): [in this window] [in a new window]   Fig 1. Study design for Studies 1 (panel A) and 2 (panel B). For Study 1, the gerbils (n = 105) were fed a carotene-free, semipurified powdered diet for 7 d, after which 15 gerbils were killed. The remaining gerbils were split into two groups, one receiving a +vitamin A (VA)/-carotene (C) diet (12.0 ± 4.2 nmol VA/g diet) and the other receiving a +VA/+C diet (2.4 ± 1.5 nmol VA/g diet, 20.3 ± 6.2 nmol C/g diet) for a 7-d prefeeding period. After the 7-d prefeeding period, 15 gerbils from each diet group were killed, and the rest were fed the VA/C diet for up to 84 d (groups of gerbils, n = 15 from each prefeeding diet group, were killed on d 28 and 84). For Study 2, the gerbils were fed the same carotenoid-free diet for 7 d, after which 12 gerbils were killed for baseline values. The remaining gerbils were fed a VA/+C diet (74.3 ± 19.7 nmol C/g diet) for 7 d, after which 12 more gerbils were killed (d 0). After this period, the rest were fed either a +VA/C diet (7.1 ± 1.4 nmol VA/g diet) or a VA /C diet for up to 34 d. Groups of 12 gerbils from each diet group were killed on d 7, 14, 24 and 34 for C and VA analysis.

Study 2. All gerbils (n = 128) were fed a powdered VA/C acclimation diet for 12 d, and for the first 5 d of this acclimation period a nonpurified diet (Purina Mills, St. Louis, MO) was added to the powdered diet for easier adaptation to the new diet (Fig. 1B). After the acclimation period, 12 gerbils were killed for baseline values, and the remaining gerbils were fed a VA/+C diet (74.3 ± 19.7 nmol C/g diet) for 7 d to increase C concentrations in the serum and tissues. At the end the 7-d prefeeding period (d 0) 12 gerbils were killed, and the remaining gerbils were divided randomly into two groups. One group received a VA/C diet, and the other group received a +VA/C diet (7.1 ± 1.4 nmol VA / g diet) for the remainder of the study. At d 7, 14, 24 and 34 after the 7-d C prefeeding period, 12 gerbils from each group were killed to follow the changes over time in concentrations of C and VA in the serum and tissues.

Gerbils were anesthetized with an intramuscular injection of premixed (10:1 v/v) ketamine HCl/xylazine (100 g / L) (Study 1) or with an inhalation of methoxyfurane (Metofane, Mallinckrodt Veterinary, Mundilein, IL) (Study 2), as recommended by the campus veterinarian. Blood samples were collected by cardiac puncture followed by severing the brachial vessels between the pectoralis major and the latissimus dorsi (Study 1) or by cervical dislocation (Study 2). Tissues (liver, kidney, lung and adrenal) were collected for subsequent analysis. Tissues were rinsed, dried, weighed and stored at 20°C until analysis, which was completed within 1 y.

Analytical methods

Tissue, serum and diet preparation.  Duplicate samples of liver, kidney or lung (0.1-0.2 g) were minced, placed into 50-mL tubes and prepared as described previously by our laboratory (Lee et al. 1998). Diet samples were analyzed six to eight times and did not need to be homogenized. Whole adrenals were pooled (n = 6-10), minced and homogenized with 5 mL ethanol/BHT. They were saponified and extracted as previously described (Lee et al. 1998). Serum samples were denatured by the addition of twice the volume of ethanol/BHT (1g/L). Retinyl laurate (synthesized in our laboratory) and echinenone (a gift from Hoffmann LaRoche) were added as internal standards for VA and C, respectively. Samples were extracted three times with a volume of hexane equal to the current volume in the tube. The combined hexane layers were evaporated completely under vacuum. Extracts were stored at 20°C until analysis within 48 h. HPLC was performed as described previously by our laboratory (Lee et al. 1998).

Statistical analysis.  Study 1. The VA and C levels of liver, adrenal, kidney and serum were compared using one-way ANOVA and Fisher's protected least significant difference (PLSD) analysis (Statview 4.5, Abacus Concepts, Berkley, CA). The percentage of retention of VA (d 28 or 84 relative to d 0) in individual liver, adrenal, kidney and serum samples was also compared between groups. The percentage of retention of VA from each gerbil was calculated by using the following formula: As expected, the +VA/C and +VA/+C group means at d 0 were not equal; thus the percentage of retention was calculated to allow comparison between the two diet groups at d 28 and 84. The percentage of retention of VA was compared using one-way ANOVA and Fisher's PLSD analysis. A comparison of percentage of retention of C was not necessary because the group that was not fed C in the diet had negligible C in serum or tissues.

Study 2. The VA and C levels of liver, adrenal, kidney, lung and serum among different groups were compared using one-way ANOVA and Fisher's PLSD analysis (Statview 4.5). Comparison of percentage of retention of VA or C was not necessary because VA and C levels did not differ among groups at d 0. Values in the text are means ± SD.

	RESULTS

Abstract Introduction Methods Results Discussion References

Study 1

Tissue and serum VA.  During the study period, there was no effect of diet on food intake, weight gain or health status (data not shown). No significant differences were found in liver VA content between the groups at d 28 or 84, regardless of the amount of C present in the tissues (Fig. 2 and Table 2). After the 7-d prefeeding period, total hepatic VA was significantly higher in both diet groups compared with the baseline value (Fig. 3). The total hepatic VA stores of each group pair (prefed either C/+VA or +C/+VA) were lower, until the final groups were approaching VA deficiency (d 84, P < 0.05). At d 84, the gerbils had liver VA concentration of 41 and 48 nmol/g for the C/+VA and +C/+VA prefeeding diets, respectively. Humans are considered VA deficient when liver concentrations of VA are <70 nmol/g (Olson 1994).

View larger version (16K): [in this window] [in a new window]   Fig 2. Liver -carotene (C) concentration in gerbils prefed diets with or without C for 7 d and then fed diets without vitamin A (VA) or C for up 28 or 84 d (Study 1). The diets on the horizontal axis refer to the prefeeding diets, containing VA and/or C. From d 0 through d 84, all groups were fed a VA/C diet. Bars represent groups means ± SD, n = 15. Bars with different letters are significantly different (P < 0.05).

View this table: [in this window] [in a new window]   Table 2. Percentage of retention of vitamin A (VA) at d 28 and 84, compared with d 0 in gerbils prefed diets with or without -carotene (C) and then fed diets without VA or C (Study 1)1,2,3

View larger version (18K): [in this window] [in a new window]   Fig 3. Liver vitamin A (VA) concentration in gerbils prefed diets with or without -carotene (C) for 7 d and then fed diets without VA or C for up 28 or 84 d (Study 1). The diets on the horizontal axis refer to the prefeeding diets, containing VA and/or C. From d 0 through d 84, all groups were fed a VA/C diet. Bars represent group means ± SD, n = 15. Bars with different letters are significantly different (P < 0.05).

Analysis of pooled adrenal tissue showed the same trend as the livers, with lower VA concentration with successive kill periods. Concentrations in both groups were significantly lower than baseline at d 84 (Table 3). The percentage of retention of VA did not differ between group pairs at d 28 or 84 (Table 2).

View this table: [in this window] [in a new window]   Table 3. Tissue vitamin A (VA) and serum retinol concentrations in gerbils prefed diets with or without -carotene (C) and then fed diets without VA or C (Study 1)1,2,3

In the kidneys, there were significant differences in VA concentrations among groups (Table 3), which appeared to be random and not related to diet or time. There were no significant differences in the percentage of retention of VA between group pairs at d 28 and at 84 (Table 2).

The groups at d 84 had significantly lower serum VA concentrations than groups at d 0 that were fed the same prefeeding diet, but VA concentrations were not different from the baseline group (Table 3). The VA concentration at d 0 was higher than at baseline in both prefeeding groups due to the 7 d of consuming VA. There were no significant differences in the percentage of retention of VA in serum between the groups prefed +VA/C and +VA/+C at d 28 or 84 (Table 2).

Tissue and serum C analysis.  Tissue concentrations of C were highest in the gerbils killed immediately after the C prefeeding period (d 0). The groups that were not prefed any C had negligible C in the serum and tissues. In the liver of gerbils prefed C, the highest level of C was at d 0. Hepatic C contents were significantly lower at d 28 and 84 (Fig. 2).

In the adrenal gland, the highest concentration of C was 4.7 ± 2.5 nmol/g at d 0 in gerbils prefed +VA/+C (Table 4). By d 28, adrenal C concentrations were not different from the baseline level. The kidneys had negligible C concentrations in all groups (Table 4).

View this table: [in this window] [in a new window]   Table 4. Tissue and serum -carotene (C) concentrations in gerbils prefed diets with or without -carotene (C) and then fed diets without vitamin A (VA) or C (Study 1)1,2

The serum had minute quantities of C and showed a pattern of change similar to those of liver and adrenals (Table 4). -Carotene concentrations at d 0 were significantly higher than baseline in the +VA/+C prefeeding group as a result of the previous 7 d of consuming C. By d 28 and 84, the serum C was significantly lower than d 0 and did not differ from the baseline concentration.

Study 2

Tissue and serum VA.  There was no effect of diet on food intake, weight gain or health of the gerbils (data not shown). Hepatic VA after the 7-d prefeeding with C (d 0) was significantly higher than baseline due to the conversion of ingested C to VA (Fig. 4). For the groups that were fed VA, hepatic VA stores were significantly higher at each successive kill time. There were no significant differences among groups not fed VA from d 0 to 34. 

View larger version (17K): [in this window] [in a new window]   Fig 4. Liver vitamin A (VA) concentration in gerbils prefed -carotene (C) and then fed a C-free diet either with or without VA for 7, 14, 24 or 34 d (Study 2). All gerbils were prefed C for 7 d and then fed the diet indicated on the horizontal axis from d 0. Bars represent group means ± SD, n = 12. Bars with different letters are significantly different (P < 0.05).

The pattern of change of VA concentrations in adrenals for this second study was similar to that of the first study (Table 5). VA levels were highest at baseline and d 0. No significant differences were found among any of the remaining groups at d 0-34. Surprisingly, adrenal VA concentration was not higher in groups that were fed VA for 34 d compared with baseline, suggesting that the adrenals glands were saturated with VA.

View this table: [in this window] [in a new window]   Table 5. Tissue vitamin A (VA) and serum retinol concentrations in gerbils prefed -carotene (C) and then fed a C-free diet either with or without VA (Study 2)1,2

The lungs of these gerbils had much higher concentrations of VA than the other extrahepatic tissues studied (Table 5). The pattern of change of VA concentrations in the lungs was similar to the pattern seen in the liver. The groups that were fed VA had significantly higher VA concentration over the course of the study, and the groups that were not fed VA were not significantly different from gerbils studied on d 0. However, the VA concentration was significantly higher at d 0 compared with baseline.

There were few significant differences in kidney and serum VA that were not related to the diets or time of feeding (Table 5).

Tissue and serum C analysis.  Liver C concentrations were significantly higher in the groups killed at d 0 than in those killed at baseline (Fig. 5). Hepatic C at d 7 was significantly lower in both the +VA and VA groups compared with d 0. No significant differences were found among the +VA and VA groups at any kill time (P > 0.5).

View larger version (17K): [in this window] [in a new window]   Fig 5. Liver -carotene (C) concentration in gerbils prefed C and then fed a C-free diet either with or without vitamin A (VA) for 7, 14, 24 or 34 d (Study 2). All gerbils were prefed C for 7 d and then fed the diet indicated on the horizontal axis, either with or without VA, from d 0. Bars represent group means ± SD, n = 12. Bars with different letters are significantly different (P <0.05).

The adrenal C had a pattern of change similar to that of liver C (Table 6). The highest concentration of C was found at d 0, immediately after the gerbils were fed C for 7 d. By d 7, C was significantly lower in both the +VA and VA diet groups compared with d 0, although these groups were not different from one another at this or subsequent times. Any C that may have been present in this lung tissue was undetectable by the HPLC system used.

View this table: [in this window] [in a new window]   Table 6. Tissue and serum -carotene (C) concentrations in gerbils prefed C and then fed a C-free diet either with or without vitamin A (VA) (Study 2)1,2

The kidney C was significantly higher at d 0 compared with baseline due to the C prefeeding period (Table 6). Interestingly, kidney C was not lower at d 7 as in the other tissues, but was lower at d 14 in both the +VA and VA groups compared with d 0.

Serum C was higher in gerbils studied on d 0 compared with all other groups.

	DISCUSSION

Abstract Introduction Methods Results Discussion References

VA deficiency is a serious problem in the world today. Suggested approaches to prevent or treat this deficiency include fortification of staple foods with VA, periodic high dose VA supplementation of the population and dietary intervention with foods rich in VA and/or pro-VA compounds such as C. In many countries, pro-VA compounds provide up to 100% of the VA value of the diet (Rodriguez and Irwin 1972). However, it has been suggested that the bioavailability of C from certain vegetables is poor (de Pee et al. 1995), and the absorption of C is affected by fat level, protein level, malnutrition, illness and many other factors (Erdman et al. 1993).

To examine the potential of C-rich foods or body C stores to prevent or treat VA deficiency, the metabolism of C in the body must be more fully understood. It is possible to achieve high tissue levels of C by feeding C; supplementation with C increases tissue C levels in human subjects (Canfield et al. 1991). It has not been determined whether C that accumulates in the tissues is specifically mobilized and cleaved to form VA.

Study 1 examined the question whether previously fed C, stored in the tissues, can be utilized for VA if VA is removed from the diet. There were no differences in the percentage of retention of serum, liver, kidney or adrenal VA between the groups with tissue C stores and those with no tissue C stores. We had expected that the groups that had tissue C stores would convert that C to VA as the VA status declined. However, liver C was significantly lower than the level at d 0, before the gerbils were even marginally VA deficient, as measured by liver VA concentrations. Thus there is no evidence that tissue C was either contributing to VA value or retained for later use. It is possible that the C was redistributed to other tissues in the body for accumulation or storage, but data from the extrahepatic tissues analyzed in this study do not support this hypothesis.

We felt that two limitations existed with this first study design. First, it is possible that the conversion of tissue C to VA began immediately once the gerbils were fed the VA diet, resulting in the rapid decrease in C levels. However, the ratio of the total hepatic C to VA content at d 0 (13 nmol C vs. 629 or 562 nmol VA) might have been too low to detect the contribution of C to VA stores. In addition, the 28-d period was perhaps too long for detection of any effect of tissue C on VA status.

The second study was similarly designed to determine whether previously fed C was retained in the tissues to be utilized for future VA needs. In this study, the gerbils were fed approximately three times more C during the prefeeding period than in Study 1 in an attempt to increase tissue C levels. This approach resulted in higher serum and tissue C levels than in the previous study. However, no differences in serum or tissue C loss were seen between the groups that were fed VA and those that were not. Liver C was significantly lower in gerbils after only 7 d of C depletion compared with gerbils before depletion. -Carotene in the adrenals, serum and kidneys was also lower in the gerbils after 7 d of depletion. During this time, there was no change in VA stores in serum or any of the tissues studied. Serum C concentration returned to baseline levels by d 7 and was not different for the rest of the study. This suggests that the level of C was maintained in the serum, probably due to the loss of C from the liver and other extrahepatic tissues. However, the level of C found in the serum represents only a small portion of the C lost from the tissues. At d 0, the ratio of hepatic C to VA was 38 nmol to 298 nmol (~1:8), and at d 7 this ratio was 12 to 713 nmol (~1:59) (for the group fed VA) and 11 to 230 nmol (~1:21) (for the group not fed VA). Thus the liver C was significantly decreased without any increase in VA stores in the liver, kidney, adrenal, lung or serum.

The high concentration of VA present in the lungs was also very interesting. The pattern of changes in VA concentration was similar to the pattern found in the liver. After feeding only C for 7 d, the VA concentration in the lungs was significantly higher than baseline, but this was not seen in any of the other extrahepatic tissues studied. This suggests that lung tissue can convert C to VA, especially when given in higher doses, such as this study.

The observations of Study 2 support the results from Study 1. The C that was stored or accumulated in the liver was lost rapidly (within 7 d) and did not cause a detectable change in liver, kidney, adrenal, lung or serum VA stores. Daily utilization rate (loss of hepatic stores) of VA by gerbils is not known, but the average daily loss was recently estimated in our laboratory to be 3.1 µg/100 g of body weight for 4- to 5-wk-old gerbils that were fed depletion diets for 8-10 wk (Lee et al. 1998). In Study 2, gerbils weighed an average of 57 g at the end of the 34-d study; therefore it can be estimated that they would lose about 6.3 nmol/d (1.8 µg/d) of VA. During the first 7 d of VA depletion, the gerbils lost 10 nmol (2.9 µg) of hepatic VA per day. During these same 7 d, the VA/C group also lost 3.1 nmol (0.89 µg) of hepatic C per day, strongly suggesting that little or no C was converted to VA for tissue needs or storage. Vitamin A utilization rate over the entire 34 d of Study 2 for this VA/C group was 3.3 nmol/d (0.94 µg/d).

Numerous studies have reported that C is converted to VA in the intestine in rats (Goodman and Huang 1965, Wang et al. 1991), and other species (Wang et al. 1991). Enzymatic conversion of C to VA has also been demonstrated in vitro in the liver and other extrahepatic tissues in monkeys, ferrets and rats (Wang et al. 1991). Lakshman et al. (1989) found that when a radiolabeled dose was administered intravenously, C is converted to VA in the liver and lung in rats. They found no difference in conversion between rats fed C for 2 wk and those that were not, but they did not investigate the VA status of the rats. In in vivo studies reported here, we examined the conversion of tissue C to VA over a period of weeks, relative to VA status. We have found that in gerbils, high levels of tissue C do not improve VA status. The amount of VA in the diet does not affect the loss of the tissue C, suggesting that gerbils do not retain tissue C for use during a period of low VA status.

Because not all of the tissues from the gerbils were collected, it is not possible to definitively state the fate of the hepatic C. However, there were no increases in serum, adrenal or kidney C. Feces were not collected and analyzed to determine the C content. It is logical to assume that many of the metabolites of the tissue C were excreted in the feces via the bile. -Carotene is thought to have other functions in the body, especially as an antioxidant (Olson 1994). It is possible that the tissue C was utilized for some of these functions other than provision of VA.

VA is stored mainly in the stellate cells in the liver, but VA is also found in the hepatocytes and other liver cells (Olson 1994). From studies with rats, it appears that there are two pools of VA in the liver (Green and Green 1994). The VA stored in the stellate cells (the largest pool) is lost slowly, whereas VA found in other cells in the liver (the smaller pool) is lost more rapidly. The data from these gerbil studies suggest that there are also two pools of C in the liver, one that is rapidly lost and the other that is lost more slowly. In contrast to VA, the largest pool of C is the rapidly lost pool. The fact that there appear to be two pools of C in the liver does not necessarily indicate that there is a specific storage mechanism for C in the liver, as with VA. The C in the liver from gerbils in Study 2 was significantly lower at d 7 compared with d 0 (Fig. 5); this C may be from the rapid turnover pool. -Carotene concentration was lower at each successive kill period, which suggests the presence of a slow turnover pool. Evaluation of the liver C from Study 1 (Fig. 2) also shows that after the initial rapid lowering of C in the first 28 d, there was a slow loss of C from d 28 to 84. Unless C is continually consumed, gerbils do not retain high levels of C in tissues. Another possibility, assuming that there is a mechanism for storage of C in the liver, is that the gerbils in these studies were fed a level of C that exceeded their capacity for storing it. The rapid loss of C would then be due to the loss of excess C that the liver could not store, whereas the subsequent slow loss of the C would be due to the loss of stored C.

-Carotene is very important as a pro-VA compound; thus it is important to understand the conversion of C to VA and the methods of storing or accumulating C in the body. This work with the gerbil model suggests that previously fed C, present in the tissues, does not improve VA status. In fact, the C is lost rapidly from the liver. There may be two pools of liver C, one that is lost rapidly and the other more slowly over time. Without continual feeding of C in the diet, most of the tissue C is eventually depleted.

The extent to which these results are applicable to humans is not known. In some countries, tissue C could serve as an important source of VA during times of food deprivation. High dose C supplementation has been considered as an alternative to high dose VA supplementation to avoid the toxic side effects seen with VA (Sommer and West 1996). If stored or accumulated C is not converted to VA when needed, this public health approach to avoiding VA deficiency would have to be revisited.

	FOOTNOTES

Manuscript received 10 November 1997. Initial reviews completed 7 January 1998. Revision accepted 16 March 1998.

	ACKNOWLEDGMENT

The authors thank Amy Moore for the synthesis of retinyl laurate.

	LITERATURE CITED

Abstract Introduction Methods Results Discussion References

• Canfield L. M., Bulux J., Serrano J. Q., Rivera C., Lima A. F., Lopez C. Y., Perez R., Khan L. K., Harrison G. G., Solomons N. W. Plasma response to oral -carotene in Guatemalan schoolchildren. Am. J. Clin. Nutr. 1991; 54:539-547[Abstract/Free Full Text]
• de Pee S., West C. E., Muhilal, Karyadi D., Hautvast J.G.A.J. Lack of improvement in vitamin A status with increased consumption of dark-green leafy vegetables. Lancet 1995; 346:75-81[Medline]
• Erdman J. W. Jr., Bierer T. L., Gugger E. T. Absorption and transport of carotenoids. Ann. N.Y. Acad. Sci. 1993; 691:76-85[Medline]
• Goodman, D. S. & Huang, H. S. (1965) Biosynthesis of vitamin A with rat intestinal enzymes. Science (Washington, DC) 149: 879-880.
• Green M. H., Green J. B. Vitamin A intake and status influence retinol balance, utilization and dynamics in rats. J. Nutr. 1994; 124:2477-2485
• Humphrey J. H., West K. P. Jr., Sommer A. Vitamin A deficiency and attributable mortality among under-5-year-olds. Bull. WHO 1992; 70:225-232[Medline]
• Lakshman M. R., Asher K. A., Attlesey M. G., Satchithanandam S., Mychkovsky I., Coutlakis P. S. Absorption, storage, and distribution of beta-carotene in normal and beta-carotene-fed rats: roles of parenchymal and stellate cells. J. Lipid Res. 1989; 30:1545-1550[Abstract]
• Lee C. M., Lederman J. D., Hofmann N. E., Erdman J. W. Jr. The Mongolian gerbil (Meriones unguiculatus) is an appropriate model for evaluation of the conversion of -carotene to vitamin A: focus on effects of vitamin A status. J. Nutr. 1998; 128:280-286[Abstract/Free Full Text]
• Olson, J. A. (1994) Vitamin A, retinoids, and carotenoids. In: Modern Nutrition in Health and Disease (Shils, M. E., Olson, J. A. & Shike, M.) Lea & Febiger, Philadelphia, PA.
• Pollack J., Campbell J. M., Potter S. M., Erdman J. W. Jr. Mongolian gerbils (Meriones unguiculatus) absorb -carotene intact from a test meal. J. Nutr. 1994; 124:869-873
• Rodriguez M. S., Irwin M. I. A conspectus of research on vitamin A requirements of man. J. Nutr. 1972; 102:909-968
• Sommer, A. & West, K. P., Jr. (1996) Vitamin A Deficiency: Health, Survival, and Vision. Oxford University Press, New York, NY.
• Wadhwa A., Singh A., Mittal A., Sharma S. Dietary intervention to control vitamin A deficiency in seven- to twelve-year-old children. Food Nutr. Bull. 1994; 15:53-56
• Wang X.-D., Tang G.-W., Fox J. G., Krinsky N. I., Russell R. M. Enzymatic conversion of -carotene into -apocarotenals and retinoids by human, monkey, ferret, and rat tissues. Arch. Biochem. Biophys. 1991; 285:8-16[Medline]

0022-3166/98 $3.00 ©1998 American Society for Nutritional Sciences