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 Dawson, H. D. Articles by Ross, A. C. Search for Related Content PubMed PubMed Citation Articles by Dawson, H. D. Articles by Ross, A. C.

---

Biochemical and Molecular Action of Nutrients

Chronic Marginal Vitamin A Status Reduces Natural Killer Cell Number and Function in Aging Lewis Rats¹

Department of Nutrition and ^* Graduate Program in Nutrition, The Pennsylvania State University, University Park, PA 16802

³To whom correspondence should be addressed.

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Natural killer (NK) cells function in the regulation of immune responses and in the surveillance of malignant or other abnormal cells. Little is known of the effects of chronic marginal vitamin A (VA) status or VA supplementation, or their interaction with age, on NK cell number and cytolytic activity. We have conducted a two-factor (diet, age) study in which male Lewis rats were fed AIN-93M diet, modified to contain either 0.3 (designated marginal), 4.0 (control) or 50 (supplemented) mg retinol equivalents (RE)/kg diet, from the time of weaning until the ages of 2.5 mo (young), 8–10 mo (middle-aged) or 18–20 mo (old). Natural killer cells were identified and quantified in peripheral blood mononuclear cells (PBMC) and spleen with the use of flow cytometry, and NK cell cytotoxicity was assayed. The number and percentage of PBMC NK cells increased with age (P < 0.0001 by two-way ANOVA). For all age groups, values were lowest in rats with marginal VA status (P < 0.0001 vs. controls). NK cell lytic activity also declined with age (P = 0.0003). As a result, NK cell lytic efficiency (lytic activity per NK cell) decreased markedly with age (P < 0.0001). Regardless of the donor's age or VA status, PBMC NK cell cytotoxicity doubled (100 ± 25% increase) after exposure to interferon- (5 x 10⁵ U/L for 1 h before assay), indicating that IFN-stimulated lytic activity was related directly to basal NK cell activity. If the relationships observed in this animal model can be applied to humans, these data suggest that elderly people consuming diets chronically low in VA may be at increased risk for infectious or neoplastic diseases.

KEY WORDS: • aging • immunophenotyping • vitamin A supplementation • liver function • rats

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
The percentage of the population 65 y of age will continue to increase dramatically in the next 20 y. Apart from the well-studied effects of dietary restriction on the aging process and of diet on cardiovascular disease, relatively little is known regarding the effects of nutritional status on the aging process or on age-related pathologies.

Nutritional surveys conducted in the United States, Canada and Europe have indicated that most healthy elderly humans meet or exceed the U.S. recommended dietary allowance (RDA)⁴ for vitamin A (VA) (Garry et al. 1982 , Hallfrisch et al. 1994 ). In some studies, up to 30% of healthy elderly people consume less than the RDA for VA (Hallfrisch et al. 1994 , Panemangalore and Lee 1992 ), whereas among institutionalized elderly people, the estimated proportion of vitamin A below the RDA is considerably higher (Azaïs-Braesco et al. 1995 ). In contrast, up to 15% of the healthy elderly population regularly consume supplements containing 1000–8000 µg retinol equivalents (RE) as preformed VA (Garry et al. 1982 , Hartz et al. 1988 ). Although this range of VA consumption in elderly humans rarely produces either overt signs of VA deficiency or toxicity (Azaïs-Braesco et al. 1995 , Hallfrisch et al. 1994 , Hartz et al. 1988 , Krasinski et al. 1989 , Stauber et al. 1991 ), it is not known whether there are more subtle effects of low or high levels of dietary VA on other physiologic systems such as the immune system. Such changes could lead to increased susceptibility to infectious disease, cancer or autoimmune diseases, which are known to increase with age (Miller 1995 ).

Natural killer (NK) cells, part of the innate immune system, play an immunoregulatory role in antibody production and cell-mediated immunity through their production of various cytokines; they function in immune surveillance through their ability to recognize and kill tumor cells or other cells with an aberrant phenotype (Kos 1998 ). Basic, clinical, epidemiologic, interventional and experimental studies have established that VA deficiency is associated with reduced resistance to a variety of infectious diseases (reviewed in Ross and Hämmerling 1994 ). However, these studies have been limited mainly to children or young animals. Previous research has shown that VA deficiency has a detrimental effect on NK cell lytic activity in young rodents (see Ross and Hämmerling 1994 ). In subsequent work, the reduction in lytic activity was attributed in part to a decrease in the number of NK cells (Zhao et al. 1994 ). In addition to a decreased bulk lytic activity due to reduced numbers of NK cells, the lytic efficiency (lytic activity per NK cell) was also decreased in VA-deficient rats. Conversely, most in vitro and in vivo studies using physiologic or high concentrations of retinoids have shown an enhancement of NK cell activity (Fraker et al. 1986 , Micksche et al. 1985 , Santoni et al. 1986 ). However, the mechanism for this stimulation is unknown, and the number of NK cells was not measured in most of these studies. Some research (Santoni et al. 1986 ) has suggested a U-shaped relationship in which both low and high doses of VA may have deleterious effects on NK cell hematopoiesis, differentiation or function. In a study of elderly nursing home residents, those taking a supplement containing 700 µg RE for 3 mo had a lower number of circulating NK cells, but this difference was not significant and NK cell activity was not assessed in that study (Fortes et al. 1998 ).

This study was designed to better understand the potential effects on NK cells of a wide range of dietary VA intakes, fed across each animal's lifetime. We wanted to produce a state of chronic marginal VA deficiency, as well as a state of chronic VA supplementation, while avoiding induction of either frank VA deficiency or toxicity. To our knowledge, this use of dietary models of VA nutrition in a long-term study is unique. Thus, considerable attention was paid to assessing the effects of these diets on growth, general health and various physical and biochemical variables associated with VA deficiency or toxicity, as well as to determining NK cell numbers, phenotype and functions. The Lewis rat was selected as the animal model for these studies because we have previously studied NK cell function and other immune responses in young VA-deficient or retinoid-supplemented rats of the same strain (Zhao et al. 1994 ). A description of T cells, including natural T cells, in these same rats is presented separately in Dawson and Ross (1999) .

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Animals, diets and experimental design.

The animal protocol used in these studies was approved by the Animal Use and Care Committee of The Pennsylvania State University. Male Lewis rats were obtained as 3- to 5-d-old pups with lactating dams from Charles River Breeding Laboratories (Kingston, NY) and were housed under conventional conditions in our animal facility. Dams were fed a purified vitamin A–free diet during lactation to reduce the accumulation of vitamin A in the offspring before weaning (Zhao et al. 1994 ). The effect of VA status and age on NK cell number and activity was assessed using a 3 x 3 two-factorial design. Rats were fed AIN-93M diet (Reeves et al. 1993 ) modified to contain 0.35, 4.0 or 50.0 mg RE (added as retinyl palmitate) per kilogram of diet. All diets were prepared by Dyets (Bethlehem, PA). These diets are designated VA-marginal, control and VA-supplemented, respectively, and contain ~0.2, 3.3 and 42 times the RE in the standard AIN-93M diet (Reeves et al. 1993 ). Rats were fed these diets from weaning until the ages of 2–3, 8–10 and 20–22 mo; rats in these age groups are designated young, middle-aged and old, respectively.

Rats had free access to food throughout the study. Food intakes were monitored for 1-wk periods at several times during the study, and all rats were weighed monthly. Starting at 5 mo of age and at intervals of ~3 mo thereafter, blood was obtained from the tail vein to monitor plasma retinol concentrations. Although not part of the data reported here, all rats were immunized in the last month of the study to determine antigen-specific antibody responses. Rats received intraperitoneal immunizations with 100 µg of tetanus toxoid 30 d and again 10 d before killing, and with 15 µg of the capsular polysaccharide antigen of Streptococcus pneumoniae, type III, 6 d before killing. Due to the large number of rats in the study and the need to conduct immune function assays on fresh tissues, rats were killed and their tissues collected in a blocked design, with one rat from each diet/age group killed per assay day (n = 9 rats/d), over the course of ~7 wk.

Tissue collection.

All rats were killed by CO₂ asphyxiation. Postmortem gross pathology was determined by a veterinary pathologist (Dr. Daniel Weinstock). Rats showing gross abnormalities, which were few, were excluded from analysis. Spleen and blood were harvested aseptically (see below). Other organs including brain and liver were removed, weighed and processed for storage or fixation for other parts of this study.

Blood collected from the vena cava into a heparinized syringe was centrifuged at 800 x g for 20 min at 4°C, plasma was removed and the blood cells were diluted 1:2 with RPMI 1640 medium (Flow Laboratories, McLean, VA). Peripheral blood mononuclear cells (PBMC) were obtained from this preparation after centrifugation on Histopaque (d = 1.083; Sigma Chemical, St. Louis, MO) overlayering 20 mL of single cell suspension on 15 mL of Histopaque. Cells were centrifuged at 1000 x g for 20 min at 20°C. The supernatant and intermediate buffy layer were collected, combined with RPMI 1640 medium and centrifuged at 1000 x g for 10 min at 4°C to remove platelets. The pelleted cells were washed twice again with RPMI 1640 medium, resuspended in RPMI 1640 medium, counted in a hemocytometer and assayed for viability by Trypan blue exclusion. Aliquots of these cells were suspended in RPMI 1640 with 10% fetal bovine serum (FBS; GIBCO BRL/Life Technologies, Gaithersburg, MD) at a concentration of 2.5 x 10⁹ cells/L. These cell preparations included lymphocytes and monocytes. Cytospin slides made from these preparations also showed some residual neutrophil contamination, particularly in the aging rats. For all experiments, the postpurification cell viability exceeded 95%.

Spleens were aseptically removed and weighed. A single-cell preparation (Zhao et al. 1994 ) was passed through 105-µm sterile nylon mesh before centrifugation. Mononuclear cells were obtained from this mixture after centrifugation on Histopaque as described for PBMC.

Biochemical assays.

Saponifiable (total) plasma and liver retinol concentrations were determined as previously described (Bowman et al. 1990 ). Plasma protein was determined by the procedure of Markwell et al. (1978) using bovine serum albumin as the standard. Plasma aspartate aminotransferase, total bilirubin, cholesterol and triglyceride concentrations were determined colorimetrically with diagnostic kits (catalogue numbers 505, 605-C, 352, and 393, respectively) as per manufacturer's instructions (Sigma Diagnostics, St. Louis, MO). Plasma albumin was measured by radial immunodiffusion using a polyclonal rabbit anti-rat albumin antibody (ICN Pharmaceuticals, Costa Mesa, CA) and rat albumin (Calbiochem, La Jolla, CA) as the standard.

Determination of peripheral blood white cell count and differential counts.

Whole blood (50 µL) was added to 450 µL of Türk's solution (1% crystal violet in diluted acetic acid). Cells were quantified manually, in duplicate, using a hemocytometer. Differential counts were determined on peripheral blood smears after Wright/Giemsa staining (Wright Stain, modified; Sigma Diagnostics). Two hundred cells were counted per slide and categorized as lymphocytes, neutrophils, eosinophils or monocytes.

NK cell percentage, number and surface antigen determination by flow cytometry.

The following antibodies were purchased from Pharmingen (San Diego, CA): anti-CD3 (G4.18, mIgG3, biotin labeled), anti-TNP isotype controls [107.3, mIgG1 fluorescein isothiocyanate (FITC) or phycoerythrin (PE), G155–178, mIgG2a, FITC labeled, J606, mIgG3, biotin labeled]. The following antibodies were from Serotec (Raleigh, NC): anti-CD68 (ED-1, mIgG1, biotin labeled), anti-Igk (OX-12, mIgG2a, FITC labeled). Anti-NKR-P1A (3.2.3, murine ascites) was kindly provided by Dr. William Chambers of the Pittsburgh Cancer Institute. A purified IgG1 fraction from this ascites was prepared by affinity purification (MAbTrap kit, Pharmacia Biotech, Piscataway NJ). FITC-labeled 3.2.3 was prepared by direct conjugation using a commercially available kit (Sigma Immunochemicals). PE-labeled streptavidin was obtained from Biosource International (Camarillo, CA).

The optimal dilutions for staining of all purified concentrated monoclonal antibodies were determined beforehand by dose-response titration and defined as the dilution of antibody that provided maximum separation of stained and unstained peaks. All antibody dilutions were made in PBS buffer (without calcium and magnesium, with 0.1% sodium azide and 1% FBS). Purified PBMC and splenocytes (0.25 x 10⁶) in 50 µL of PBS buffer were incubated with 5 µL of the appropriate dilution of each antibody in round-bottomed 96-well plates at 4°C. After incubation with antibody for 30–40 min, cells were pelleted and washed twice with 100 µL of PBS/FBS buffer. For cells labeled with biotinylated antibodies, 5 µL of a 1:100 diluted solution of PE-labeled streptavidin was added to cells, which were suspended in a volume of 50 µL. These cells were incubated for 20 min at 4°C, washed with 100 µL of PBS/FBS buffer twice, fixed in 100 µL of 1% paraformaldehyde solution and stored at 4°C for 1–3 d until analysis.

Samples were analyzed on a flow cytometer (Coulter EPICS XLII System, Coulter, Hileah, FL) at the Penn State Center for Qualitative Cell Analysis. Gating was set on either total PBMC or splenocytes, or on cells of a specific phenotype. Unstained cells were included as controls in all experiments. Additionally, irrelevant isotypic controls (for each monoclonal antibody class and label) were used to assess nonspecific binding. Flow cytometry data are expressed as a percentage of total cells, a percentage of lymphocytes (sum of T cells, B cells, and NK cells) (Kidd and Nicholson 1997 ) or as the absolute number of cells/mm³ of blood [white blood cell(WBC) count x % lymphocytes from the differential count x % cell type in the lymphocyte population from flow cytometry data].

Relative cell size, granularity and surface antigen number were also determined for NKR-P1A^bright cells. To account for any daily fluctuations in machine operation, the size and fluorescence of dual-labeled (FITC and PE) control beads (Rainbow Brite beads, Spherotec, Libertyville, IL) were also determined. The daily CV for multiple fluorescence channels of both labels and the size of the beads was always <1% and the linearity (R²) of the measurement of FITC and PE labels >0.998. Cell size was determined from the measurement of the forward angle light scattering. Granularity was determined from the measurement of the amount of light reflected from the cells at a 90° angle, or side scatter. Antigen expression/cell was determined as the mean channel number of the fluorescent marker on stained cells.

NK cell cytotoxicity assay.

NK cell cytotoxicity was assessed using the NK-sensitive YAC-1 target cell in a standard 4-h ⁵¹Cr release assay similar to that previously described (Zhao et al. 1994 ). Three- or fourfold serial dilutions of effector cells were added to 10^{4 51}Cr-labeled YAC-1 target cells to establish different effector-to-target ratios (50:1, 25:1, 12.5:1 and 6.25:1) in a 96-well plate. YAC-1 cells were also incubated alone or with 1% sodium lauryl sulfate to determine spontaneous and maximum release of ⁵¹Cr, respectively. Radioactive supernatants were mechanically harvested (Skatron, Lier, Norway) and counted (Packard Instrument, Meridian, CT). Cytotoxicity was expressed as a percentage of specific ⁵¹Cr release according to the equation:

A lytic unit is defined as the number of effector cells required to obtain 20% cytotoxicity of 10⁴ YAC-1 cells. This number was obtained by transformation of the data to describe the line that provided the best fit to the van Kroegh equation (Pross et al. 1981 ). Lytic activity was defined as the number of lytic units present in 107 effector cells (PBMC or splenocytes). Lytic efficiency was defined as the number of lytic units present in 10⁶ NK cells and was calculated from the following relationship: lytic efficiency = lytic activity/(% NKR-P1⁺ cells).

Statistical analysis.

Data from each of the cellular and biochemical analyses were analyzed for equality of variance using an F-test (Statview 4.5 for Macintosh, Abacus Concepts, Berkeley, CA). If the variance was heterogeneous, an appropriate transformation of the data was performed. Data are expressed as the mean ± SEM. A two-factor ANOVA was used to analyze the effect of diet, age or any interaction between diet and age. If there was a significant effect, a one-way ANOVA was performed (SuperANOVA 1.1, Abacus Concepts) on groups segregated according to both VA status and age. This was followed by a Tukey-Kramer post-hoc analysis to determine if there were differences due to VA status within rats of the same age group, or if there were differences due to age within rats in the same VA treatment group. If no significant interaction between diet and age was present in the two-way ANOVA, a Tukey-Kramer post-hoc analysis was performed to determine significant differences between each factor level of VA treatment or age. Because of the large number of analyses performed in this study, conservative P-values of < 0.015 for ANOVA and < 0.05 for Tukey-Kramer and least squares means test post-hoc analysis were considered significant.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Overall survival and health.

Although the study was not designed to detect differences in survival rate, it is relevant to note that all of the young and middle-aged rats survived until the end of the study and, in the cohort of old rats, two deaths occurred in each of the old, VA-marginal and the old control groups, and four deaths in the old VA-supplemented group. Seven surviving rats showed significant pathologies at autopsy and were excluded from immune function and biochemical analysis; three of these had blood cell analyses indicative of preleukemic states, three had large skin tumors and one had a tumor of the lung/mediastinum. The incidence of these pathologic conditions did not differ by diet group. In addition to these pathologies, all of the cohort of old rats eventually developed lesions on their hind feet, a condition known as pododermatitis, which is a common finding in obese aging rats (Anver et al. 1982 ). These lesions were first noticed in rats fed the marginal VA diet. Because they occurred unexpectedly, they were not quantified systematically or analyzed statistically.

On the basis of exclusions due to morbidity or mortality, the total number of evaluable rats was 16 for each of the young groups, 14–16 for the middle-aged groups, and 10–13 for the old groups. Some of the assays with smaller CV were performed on an unbiased subset of rats per treatment group.

Physical and biochemical data.

There was a significant increase in body weight with age (Table 1 ). Even though all rats continued to gain weight throughout the study, VA status became a significant factor for body weight at ~8 mo of age, with VA-marginal rats weighing less than control or VA-supplemented rats. This difference in body weight was not due to differences in food intake (data not shown). Less visceral fat was observed at autopsy in old VA-marginal rats compared with control or VA-supplemented rats.

With the exception of significantly elevated triglycerides, rats fed the VA-supplemented diet did not exhibit any of the signs characteristic of VA excess related to hepatocellular damage (elevated plasma alanine:aspartate aminotransferase or bilirubin levels, hypoalbuminemia or hepatomegaly) (Table 2 ). Similarly, no evidence of VA-induced hepatic damage was seen by gross or electron microscopic examination of the livers from VA-supplemented rats (liver data to be reported separately).

VA status had no significant effect on total WBC counts. With respect to age, there was a significant increase in WBC between middle-aged and old rats (Table 3 ). Among all ages, the percentage of lymphocytes decreased, with corresponding increases in neutrophils and monocytes. The actual number of lymphocytes decreased significantly between the young and middle-aged cohorts but, due to an increase in total WBC in the old groups, there was no decrease in the absolute number of lymphocytes in old rats. The percentage of lymphocytes was lower and the percentage of neutrophils was higher in VA-marginal rats compared with control or VA-supplemented rats. The number of neutrophils was significantly higher in VA-supplemented rats. VA status was not a significant factor on the percentage of monocytes, even though the mean values were lower in the middle-aged and old rats fed the VA-marginal diet compared with rats of the same ages fed the control diet. There were no significant differences in VA status in the percentage of monocytes, monocyte numbers or eosinophils (not shown).

There was a significant age-dependent increase in spleen weight among all age groups (P < 0.001, data not shown). VA status had no effect on spleen weight, nor was there an effect of either age or VA status on the number of cells per gram of spleen.

NK cell number and phenotype by flow cytometry.

NK cells in peripheral blood lymphocytes, determined as NKR-P1A⁺ (bright) cells by flow cytometry, increased significantly in number and percentage with age (Fig. 1 ). NKR-P1A⁺(bright) cells were exclusively CD3^– and ED-1^–, indicating a lack of T-cell or monocyte markers, respectively. Similar results were obtained when NK cells were expressed as the percentage of PBMC. For all ages, marginal VA status was associated with a lower percentage and number of NK cells in peripheral blood lymphocytes compared with control rats, whereas VA supplementation was associated with a significantly greater percentage and number of NK cells in old rats. The increase in the number of NK cells in peripheral blood during aging was highly significant (P < 0.0001), even between young and middle-aged rats. In spleen, the percentage of NK cells increased between middle-aged and old age (not shown). VA status had no effect on the percentage of NK cells in spleen.

View larger version (41K): [in this window] [in a new window]   Figure 1. Effect of vitamin A (VA) status and aging on the percentage (A) and number (B) of natural killer (NK) cells in peripheral blood lymphocytes of rats. The bars represent the mean of the values ± SEM for the number of animals (n) shown. In Figure 1 A, VA status and age were both significant main factors (two-way ANOVA, P < 0.0001) and there was a significant interaction (P < 0.001). In follow up by Tukey-Kramer test, VA status was a significant factor (marginal < control or supplemented, P < 0.01) and age was a significant factor (young < middle age < old, P < 0.01). Groups with least-squares means that differed at P < 0.05 are indicated by different letters. In Figure 1 B, VA status and age were both significant main factors (two-way ANOVA, P < 0.0001) and there was a significant interaction (P < 0.005). In follow up by Tukey-Kramer test, VA status was a significant factor (marginal < control or supplemented, P < 0.01) and age was a significant factor (young or middle age < old, P < 0.01). Groups with least squares means that differed at P < 0.05 are indicated.

NK cell cytotoxicity.

NK cell lytic activity was significantly higher in young vs. middle-aged or old rats (Table 4 ). VA status significantly affected basal NK cell lytic activity (VA-marginal < control and VA-supplemented rats, P = 0.014) in all ages. Indeed, NK cell lytic activity was nearly undetectable in several of the old rats in the VA-marginal group. In contrast to blood, there was a significant increase in NK cell lytic activity of splenocytes with age between young or middle-aged vs. old rats (not shown). Unlike in blood, there was no effect of VA status on NK cell lytic activity in spleen.

There was no effect of age or VA status on the percentage increase in NK cell lytic activity induced by in vitro incubation with interferon- (Table 4) . NK cell lytic activity increased 100 ± 25% (mean ± SD, n = 9 treatment groups) in PMBC incubated with interferon- compared with cells from the same rat incubated without interferon-.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
To our knowledge, this study represents the first controlled study of marginal VA status or VA supplementation maintained across the life span, or of the effects of a wide range of VA intake on the numbers and properties of NK cells. The level of VA in the VA-marginal diet was previously shown to permit the growth of young rats without resulting in significant liver storage of VA (Green et al. 1987 ). In this study, rats fed the VA-marginal diet for ~2–21 mo had plasma retinol concentrations that remained, throughout the study, at about one half to one third that of age-matched control rats. The level of VA in the VA-supplemented diet was chosen to create an animal model for chronic VA supplementation in humans who consume high levels of preformed VA in supplements or foods. The results of an 8-mo pilot study (Zolfaghari and Ross 1995 ) suggested that 100 mg RE/kg, which produced a very high level of hepatic storage, might cause hepatic dysfunction in a longer study; therefore, we reduced the level to 50 mg RE/kg diet for the 22-mo study reported here. It is important to note that, because diets were fed continuously from the time of weaning, age may in fact be a proxy for dietary exposure.

There was little, if any, physical or biochemical evidence of VA deficiency or toxicity in any of the rats in this study (Table 2) . However, all of the old rats in this study developed hind limb lesions characterized as ulcerative pododermatitis; thus, infection (further discussed below) may have been a contributing factor to hypoalbuminemia. An age-related increase in plasma triglycerides and cholesterol was also observed in this study; this effect has been reported previously in aging rats (Wolford et al. 1986 ) and humans (Slack et al. 1977 ). VA supplementation also caused significant hypertriglyceridemia. This effect has been noted previously in VA-supplemented and retinoid-treated humans (Pastorino et al. 1991 , Stauber et al. 1991 ).

The biochemical data from our VA-supplemented rats resemble data from a 5-y longitudinal study of healthy elderly men and women who consumed daily supplements containing up to 10 times the RDA of preformed VA (Stauber et al. 1991 ). In these elderly people, supplement use was associated with increased plasma retinyl ester concentration, which was positively correlated with plasma triglycerides, cholesterol and HDL. However, plasma retinyl ester levels were not correlated with circulating liver enzyme levels or total bilirubin. Taken together, these results suggest that circulating retinyl esters are a sign of VA excess, but may not be an indicator of significant health risk. Overall, physical signs and functional aspects of immunity were generally very good in our VA-supplemented rats.

The most obvious limitation to the interpretation of this study was the unexpected development of hind leg lesions in the cohort of old rats. However, three lines of indirect evidence suggest that these lesions did not severely compromise the study. First, the majority of the changes in immune function that were observed between middle-aged and old rats were also evident between young and middle-aged rats, before the appearance of lesions. Second, the immunologic data from this study compare favorably with previous data from studies of presumably healthy rats obtained under similar experimental conditions or data from healthy aging humans. Third, we also have examined immune phenotype in a group of healthy aging F344/BN F1 hybrid rats of similar ages, having no signs of pododermatitis. The effects of age were remarkably similar to the results obtained in this study (Dawson, unpublished results). Nonetheless, in the future, additional measures, such as frequent changes of soft bedding, should be taken to reduce the possibility of rats developing limb irritations and infection. Another measure would be to use F344/BN F1 hybrid rats, which appear less prone to these lesions. These rats were noticeably leaner and more vigorous at an advance aged than were Lewis rats; they also had fewer age-associated pathologies. Another limitation of the study is the use of a cross-sectional design, necessitated by the need to collect tissues for analysis.

In this study, WBC counts were significantly greater in old vs. middle-aged or young rats, particularly in those fed the VA-marginal diet (Table 3) . Old rats exhibited a preferential elevation in neutrophil counts over all other cell types. It is known that even subclinical infections with Staphylococcus aureus, typically found in the ulcerative pododermatitis lesions discussed above, can produce such changes in WBC and differential counts in rats (Bradfield et al. 1996 ). Second, an elevation in WBC counts with a dramatic increase in the number of neutrophils has been observed in some studies of VA deficiency in rats (Zhao et al. 1994 ). The greater elevation of WBC in VA-maginal rats may indicate that they are at increased risk of some infections. However, despite differences in total WBC and neutrophils in old rats, lymphocyte counts were not affected by age.

A main objective was to quantify and test the function of NK cells in the different diet by age groups. There was a significant expansion of NK cells (NKR-P1 bright) in the blood during aging (Fig. 1) and in spleen as reflected in the percentage of NK cells (for technical reasons, total NK cells in spleen were not determined). These data confirm earlier work in aging rats fed stock diets (Fukui et al. 1987 , Gilman-Sachs et al. 1991 , Page et al. 1995 ). The blood of VA-marginal rats contained significantly fewer NK cells and a lower percentage of NK cells at all ages. These results suggest that a chronic marginal VA deficiency, similar to acute severe VA deficiency, also reduces circulating NK cells (Zhao et al. 1994 ). Conversely, the percentage and number of NK cells in VA-supplemented old rats was significantly increased. The reasons for these differences in NK cell number may be related to differences in hematopoiesis in bone marrow or to maturational differences after pre-NK cells have emigrated from bone marrow to the periphery. It is interesting to note that a portion of dietary VA transported in chylomicrons has been shown to be taken up by bone marrow (Hussain et al. 1989 ), with considerable quantitative differences among species. Because the efficiency of VA absorption is relatively independent of dose, chylomicron VA would be expected to be present nearly in proportion to VA in the diet. Thus, the bone marrow of rats fed a diet that is chronically low in VA would likely be exposed to much less VA than the marrow cells of rats fed either the control or VA-supplemented diets.

One caveat that should be kept in mind when interpreting the flow cytometry and cytotoxicity data is that the assays were performed on mixed populations of cells containing monocyte/macrophages. The isolation of lymphocytes may result in a selective loss or enrichment of certain cellular subsets (Pelegrí et al. 1995 , Thompson et al. 1986 ). Even by light microscopy, the PBMC of old rats were noticeably different from those of younger rats. To minimize the loss of lymphocytes, the density of Histopaque was chosen to be 1.083 g/L rather than 1.077 g/L (as used in human, mouse and some rat research). The higher density of Histopaque used in our studies is closer to the optimum density (1.090 g/L) of Ficoll used to isolate rat lymphocytes and has been used previously by researchers in studies of aging rats (Goonewardene and Murasko 1993 ). As with all isolation methods based on cell density, preferential enrichment of monocyte/macrophages occurs because of the loss of granulocytes and because of the slightly lower bouyant density of monocytes relative to lymphocytes. Similarly prepared human PBMC typically contain 70–90% mononuclear cells of which 20–25% are monocytes (Stites 1994 ). This percentage is close to that for monocytes found in our middle-aged and old control rats.

Despite the increase in the percentage of blood NK cells with age, NK cell lytic activity and lytic efficiency decreased with age (Table 4) . An age-related reduction in lytic activity was observed previously (Fukui et al. 1987 , Ghoneum et al. 1987 ), but NK cell number and lytic efficiency were not studied simultaneously. Splenic NK cell lytic activity increased somewhat, in proportion to the percentage of NK cells, such that NK cell lytic efficiency was unchanged. The results of this study suggest that additional, more detailed studies are warranted to elucidate NK cell hematopoiesis, differentiation, phenotype(s) and functional activities during aging.

Neither VA status nor age was a factor in the ability of NK cells to be activated in vitro by interferon- (Table 1 and Zhao et al. 1994 ). These results suggest that the expression and/or signaling from receptors for type I interferons and subsequent signal transduction leading to increased NK cell cytotoxicity are essentially intact. Because the percentage increase in lytic activity after interferon- stimulation was not different among diet by age groups, the differences observed in basal NK cell lytic activities due to diet and age appear to be predictive of interferon-stimulated activity. These data imply that VA deficiency, whether overt (Zhao et al. 1994 ) or marginal (this study), reduces the number of NK cells and thereby the potential for heightened NK cell cytotoxic responses after stimulation by interferons. Such differences in NK cell function might affect the surveillance of abnormal self cells (e.g., cells expressing aberrant major histocompatibility proteins) or malignant cells.

In conclusion, a chronic state of marginal VA deficiency, characterized by low plasma retinol levels and essentially no liver storage of VA, was maintained across the life span of the rats in this study. Despite the lack of evidence of overt VA deficiency, there were significant differences in the number and lytic activity of NK cells in peripheral blood. Additionally, numerous age-related differences were observed, including increases in WBC counts, neutrophils, NK cells, NKR-P1 molecules per cell and NK cell granularity. The sensitivity of NK cell production and function to VA status suggests that further studies are warranted to elucidate the role(s) of dietary VA in NK cell hematopoiesis, differentiation, immune function and activity against tumor cells in vitro and in vivo.

	ACKNOWLEDGMENTS

 
We thank William Chambers for supplying the monoclonal antibody against rat NK cells, Rebecca Corwin for sharing tissues from her F344/BN rats, Meg Potter for expert care of the rats in this study and Elaine Kunze for assistance with flow cytometry.

	FOOTNOTES

 
¹ Supported by a U.S. Department of Agriculture National Needs Fellowship to H.D.D., NIH grants AG-09739 and DK-41479, and funds from the Howard Heinz Endowment.

² Current address: Laboratory of Immunology, National Institute on Aging, Baltimore, MD 21224–6825.

⁴ Abbreviations used: FBS, fetal bovine serum; FITC, fluorescein isothiocyanate; NK, natural killer; PE, phycoerythrin; PBMC, peripheral blood mononuclear cells; RDA, recommended dietary allowance; RE, retinol equivalents; VA, vitamin A; WBC, white blood cells.

Manuscript received February 11, 1999. Initial review completed March 30, 1999. Revision accepted May 12, 1999.

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1. Anver M. R., Cohen B. J., Lattuada C. P., Foster S. J. Age-associated lesions in barrier-reared male Sprague-Dawley rats: a comparison between Hap: (SD) and Crl:COBS[R]CD[R](SD) stocks. Exp. Aging Res. 1982;8:3-24[Medline]

2. Azaïs-Braesco V., Morinière C., Guesne B., Partier A., Bellenand P., Baguelin D., Grolier P., Alix E. Vitamin A status in the institutionalized elderly. Critical analysis of four evaluation criteria: dietary vitamin A intake, serum retinol, relative dose-response test (RDR) and impression cytology with transfer (ICT). Int. J. Vitam. Nutr. Res. 1995;65:151-161[Medline]

3. Bowman T. A., Goonewardene M., Pasatiempo A.M.G, Ross A. C., Taylor C. E. Vitamin A deficiency decreases natural killer cell activity and interferon production in rats. J. Nutr. 1990;120:1264-1273

4. Bradfield J. F., Schachtman T. R., McLaughlin R. M., Steffen E. K. Behavioral and physiological effects of inapparent wound infection in rats. Lab. Anim. Sci. 1996;42:572-578

5. Dawson, H. D. & Ross, A. C. (1999) Chronic marginal vitamin A status affects the distribution and function of T cells and natural killer T cells in aging Lewis rats. (in press)

6. Fortes C., Forastiere F., Agabiti N., Fano V., Pacifici R., Virgili F., Piras G., Guidi L., Bartoloni C., Tricerri A., Zuccaro P., Ebrahim S., Perucci C. A. The effect of zinc and vitamin A supplementation on immune response in the older population. J. Am. Geriatr. Soc. 1998;46:19-26[Medline]

7. Fraker L. D., Halter S. A., Forbes J. T. Effects of orally administered retinol on natural killer cell activity in wild type BALB/c and congenitally athymic BALB/c mice. Cancer Immunol. Immunother. 1986;21:114-118[Medline]

8. Fukui H., Overton W. R., Herberman R. B., Reynolds C. W. Natural killer cell activity in the rat. VI. Characterization of rat large granular lymphocytes as effector cells in natural killer and antibody-dependent cellular cytotoxic activities. J. Leukoc. Biol. 1987;41:130-142[Abstract]

9. Garry P. J., Goodwin J. S., Hunt W. C., Hooper E. M., Leonard A. G. Nutritional status in a healthy elderly population: dietary and supplement intakes. Am. J. Clin. Nutr. 1982;36:319-331[Abstract/Free Full Text]

10. Ghoneum M., Gill G., Assanah P., Stevens W. Susceptibility of natural killer cell activity of old rats to stress. Immunology 1987;60:461-465[Medline]

11. Gilman-Sachs A., Kim Y. B., Pollard M., Snyder D. L. Influence of aging, environmental antigens, and dietary restriction of expression of lymphocyte subsets in germ-free and conventional Lobund-Wistar rats. J. Gerontol. 1991;46:B101-B106[Medline]

12. Goonewardene I. M., Murasko D. M. Age-associated changes in mitogen-induced lymphoproliferation and lymphokine production in the long-lived brown-Norway rat. Mech. Ageing Dev. 1993;71:199-212[Medline]

13. Green M. H., Balmer-Green J., Lewis K. Variation in retinol utilization rate with vitamin A status in the rat. J. Nutr. 1987;117:694-703

14. Hallfrisch J., Muller D. C., Singh V. N. Vitamin A and E intakes and plasma concentrations of retinol, ß-carotene, and -tocopherol in men and women of the Baltimore Longitudinal Study of Aging. Am. J. Clin. Nutr. 1994;60:176-182[Abstract/Free Full Text]

15. Hartz S. C., Otradovec C. L., McGandy R. B., Russell R. M., Jacob R. A., Sahyoun N., Peters H., Abrams D., Scura L. A., Whinston-Perry R. A. Nutrient supplement use by healthy elderly. J. Am. Coll. Nutr. 1988;7:119-128[Abstract]

16. Hussain M. M., Mahley R. W., Boyles J. K., Lindquist P. A., Brecht W. J., Innerarity T. L. Chylomicron metabolism. Chylomicron uptake by bone marrow in different animal species. J. Biol. Chem. 1989;264:17931-17938[Abstract/Free Full Text]

17. Kidd P. O., Nicholson J.K.A. Immunotyping by flow cytometry. Rose N. R. De Macario E. C. Folds J. D. Lane H. H. Nakamura R. M. eds. Manual of Clinical Immunology 1997:229-244 ASM Press Washington, DC.

18. Kos F. J. Regulation of adaptive immunity by natural killer cells. Immunol. Res. 1998;17:303-312[Medline]

19. Krasinski S. D., Russell R. M., Otradovec C. L., Sadowski J. A., Hartz S. C., Jacob R. A., McGandy R. B. Relationship of vitamin A and vitamin E intake to fasting plasma retinol, retinol-binding protein, retinyl esters, carotene, alpha-tocopherol, and cholesterol among elderly people and young adults: increased plasma retinyl esters among vitamin A-supplement users. Am. J. Clin. Nutr. 1989;49:112-120[Abstract/Free Full Text]

20. Markwell M. A., Haas S. M., Bieber L. L., Tolbert N. E. A modification of the Lowry procedure to simplify protein determination in membrane in lipoprotein samples. Anal. Biochem. 1978;87:206-210[Medline]

21. Micksche M., Colot M., Ucheida A., Kokoschka E. M., Luger T. A., Dittrich C., Moser K., Rainer H., Lenzhofer R., Kolb R., Jakesz R., Schemper M., Kokron O., Zwick H., Scheiner A., Flicker H., Sagaster P. Immunomodulation in cancer patients by synthetic biological response modifiers. Cancer Treat. Symp. 1985;1:27-35

22. Miller R. A. Immune system. Masoro E. J. eds. Handbook of Physiology. Section 11: Aging 1995:555-590 Oxford University Press New York, NY.

23. Page G. G., Ben-Eliyahu S., Taylor A. N. The development of sexual dimorphism in natural killer cell activity and resistance to tumor metastasis in the Fischer 344 rat. J. Neuroimmunol. 1995;63:69-77[Medline]

24. Panemangalore M., Lee J. C. Evaluation of the indices of retinol and -tocopherol status in free-living elderly. J. Gerontol. 1992;47:B98-B104[Medline]

25. Pastorino U., Chiesa G., Infante M., Soresi E., Clerici M., Valente M., Belloni P. A., Ravasi G. Safety of high dose vitamin A. Randomized trial on lung cancer chemoprevention. Oncology 1991;48:131-137[Medline]

26. Pelegrí C., Rodríguez-Palmero M., Morante M. P., Comas J., Castell M., Franch A. Comparison of four lymphocyte isolation methods applied to rodent T cell subpopulations and B cells. J. Immunol. Methods 1995;187:265-271[Medline]

27. Pross H. F., Baines M. G., Rubin P., Shragge P., Patterson M. S. Spontaneous human lymphocyte-mediated cytotoxicity against tumor target cells. IX. The quantitation of natural killer cell activity. J. Clin. Immunol. 1981;1:51-59[Medline]

28. Reeves P. G., Nielsen F. H., Fahey G. C., Jr AIN-93 purified diets for laboratory rodents: final report of the American Institute of Nutrition ad hoc writing committee on the reformulation of the AIN-76A rodent diet. J. Nutr. 1993;123:1939-1951

29. Ross A. C., Hämmerling U. G. Retinoids and the immune system. Sporn M. B. Roberts A. B. Goodman D.S. eds. The Retinoids: Biology, Chemistry, and Medicine 1994:521-543 Raven Press New York, NY.

30. Santoni A., Cerruti Sola S., Giovarelli M., Martinetto P., Vietti D., Forni G. Modulation of natural killer activity in mice by prolonged administration of various doses of dietary retinoids. Nat. Immun. Cell Growth Regul. 1986;5:259-266[Medline]

31. Slack J., Noble N., Meade T. W., North W.R.S. Lipids and lipoproteins in 1,604 men and women in working populations in Northwest London. Br. Med. J. 1977;2:353-357

32. Stauber P. M., Sherry B., VanderJagt D. J., Bhagavan H. N., Garry P. J. A longitudinal study of the relationship between vitamin A supplementation and plasma retinol, retinyl esters, and liver enzyme activities in a healthy elderly population. Am. J. Clin. Nutr. 1991;54:878-883[Abstract/Free Full Text]

33. Stites D. P. Clinical laboratory methods for detection of cellular immunity. Stites D. P. Terr A. I. Parslow T. G. eds. Basic and Clinical Immunology 1994:195-215 Appleton & Lange Norwalk, CT.

34. Thompson S. C., Bowen K. M., Burton R. C. Sequential monitoring of peripheral blood lymphocyte subsets in rats. Cytometry 1986;7:184-193[Medline]

35. Wolford S. T., Schroer R. A., Gohs F. X., Gallo P. P., Brodeck M., Falk H., Ruhren R. Reference range data base for serum chemistry and hematology values in laboratory animals. J. Toxicol. Environ. Health 1986;18:161-188[Medline]

36. Zhao Z., Murasko D. M., Ross A. C. The role of vitamin A in natural killer cell cytotoxicity, number and activation in the rat. Nat. Immun. 1994;13:29-41[Medline]

37. Zolfaghari R., Ross A. C. Chronic vitamin A intake affects the expression of mRNA for apolipoprotein A-I, but not nuclear retinoid receptors, in liver of young and aging Lewis rats. Arch. Biochem. Biophys. 1995;323:258-264[Medline]

This article has been cited by other articles:

				D. E. Lackey, S. L. Ashley, A. L. Davis, and K. A. Hoag Retinoic Acid Decreases Adherence of Murine Myeloid Dendritic Cells and Increases Production of Matrix Metalloproteinase-9 J. Nutr., August 1, 2008; 138(8): 1512 - 1519. [Abstract] [Full Text] [PDF]

				H. Yamada, S. Mizuno, A. C. Ross, and I. Sugawara Retinoic Acid Therapy Attenuates the Severity of Tuberculosis While Altering Lymphocyte and Macrophage Numbers and Cytokine Expression in Rats Infected with Mycobacterium tuberculosis J. Nutr., December 1, 2007; 137(12): 2696 - 2700. [Abstract] [Full Text] [PDF]

				R. Zolfaghari, C. J. Cifelli, S. O. Lieu, Q. Chen, N.-q. Li, and A. C. Ross Lipopolysaccharide opposes the induction of CYP26A1 and CYP26B1 gene expression by retinoic acid in the rat liver in vivo Am J Physiol Gastrointest Liver Physiol, April 1, 2007; 292(4): G1029 - G1036. [Abstract] [Full Text] [PDF]

				A. C. Ross and R. Zolfaghari Regulation of Hepatic Retinol Metabolism: Perspectives from Studies on Vitamin A Status J. Nutr., January 1, 2004; 134(1): 269S - 275. [Abstract] [Full Text] [PDF]

				A. C. Ross Retinoid Production and Catabolism: Role of Diet in Regulating Retinol Esterification and Retinoic Acid Oxidation J. Nutr., January 1, 2003; 133(1): 291S - 296. [Abstract] [Full Text] [PDF]

				R. Zolfaghari and A. C. Ross Lecithin:retinol acyltransferase from mouse and rat liver: cDNA cloning and liver-specific regulation by dietary vitamin A and retinoic acid J. Lipid Res., December 1, 2000; 41(12): 2024 - 2034. [Abstract] [Full Text]

				Y. YAMAMOTO, R. ZOLFAGHARI, and A. C. ROSS Regulation of CYP26 (cytochrome P450RAI) mRNA expression and retinoic acid metabolism by retinoids and dietary vitamin A in liver of mice and rats FASEB J, October 1, 2000; 14(13): 2119 - 2127. [Abstract] [Full Text]

				N. Ahluwalia, M. A. Gordon, G. Handte, M. Mahlon, N.-Q. Li, J. L. Beard, D. Weinstock, and A. C. Ross Iron Status and Stores Decline with Age in Lewis Rats J. Nutr., September 1, 2000; 130(9): 2378 - 2383. [Abstract] [Full Text]

				H. D. Dawson, Y. Yamamoto, R. Zolfaghari, F. J. Rosales, J. Dietz, T. Shimada, N.-q. Li, and A. C. Ross Regulation of Hepatic Vitamin A Storage in a Rat Model of Controlled Vitamin A Status during Aging J. Nutr., May 1, 2000; 130(5): 1280 - 1286. [Abstract] [Full Text]

				H. D. Dawson and A. C. Ross Chronic Marginal Vitamin A Status Affects the Distribution and Function of T Cells and Natural T Cells in Aging Lewis Rats J. Nutr., October 1, 1999; 129(10): 1782 - 1790. [Abstract] [Full Text]

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 Dawson, H. D. Articles by Ross, A. C. Search for Related Content PubMed PubMed Citation Articles by Dawson, H. D. Articles by Ross, A. C.