Biochemical

and Molecular Roles of nutrients

Dietary Fat Influences la Antigen Expression and Immune Cell Populations in the Murine Peritoneum and Spleen1»2 Department of Animal Sciences and *Department of Molecular Microbiology and Immunology, University of Missouri, Columbia, MO 65211 interact with, and activate, T cells (1, 2). Major histocompatibility class II (la) antigen expression on cellular membranes has been shown to correlate quantitatively with the immune response (3). The expression of la antigens is inducible and influenced by certain lymphokines, prostaglandins (PG),5 alphafetoprotein, glucocorticoids and stress (4—8). Immune cell lymphokine production, PG biosynthesis, cellular membrane fluidity and its fatty acid composition are influenced by dietary fat (9-13). However, little is known about the effect of dietary fat on la antigen expression and immune cell populations. The purpose of this study was to determine whether dietary fat influences la expression and the populations of T cells, B cells and macrophages in the peritoneum and spleen of both Listeria monocytogenes (LM)-infected and noninfected mice. Four di etary fats were chosen to compare their effect on these selected aspects of the immune system. The fats were rich in either linoleic acid [18:2(n-6(], oleic acid (18:l(n-9)], (fl-3) polyunsaturated fatty acids (PUFA),

ABSTRACT Peritoneal cells (PEC) and splenocytes were obtained from Osterìamonocytogene (LM)-lnfected or noninfected mice fed a 20% fat diet rich in either (n-3) polyunsaturated fatty acids [(n-3) PÜFA diet], linoleate [(n-6) PÜFAdiet], oleate (MONO diet), or saturated fatty acids (SAT diet) for 6 wk and were assessed for T cells, B cells, macrophages and la ex pression by flow cytornetric analysis. In the peritoneum of noninfected mice, dietary fat did not affect total cell yield or the percentage of B cells, macrophages or Ia+ cells, but the (n-3) P(lFA-fed group had a greater per centage of T cells than did the other groups. Among the LM-infected mice, the (n-3) PUFA-fed group generally had the highest percentage of B cells and the lowest percentages of T cells, macrophages and Ia+ cells in the peritoneum. Listeria monocytogene infection elevated peritoneal T cell numbers in all mice except the (n-3) PÜFA-fedgroup. The density of la molecules on PEC was 40% lower in mice fed the (n-3) PÜFAdiet. In the spleen, dietary fat also influenced the immune cell popu lations and Ia+ cells. Two-color staining of spleen cells revealed that Ia+ splenocytes were predominately B cells. These data demonstrate that dietary fats influence la expression and immune cell populations and that the effects observed in one immune tissue or cell type may not be readily extrapolated to others. J. Nutr. 122: 1219-1231, 1992. INDEXING KEY WORDS:

•dietary fat •la antigen

financial support for this research was provided by the Food for the 21st Century Program and the University of Missouri Agri culture Experiment Station. Contribution from the Missouri Agriculture Experiment Sta tion, Journal Series Number 11,545. 3Current address: Oklahoma Medical Research Foundation, 825

fish oil •mice immune cell populations

N.E. 13th St., Oklahoma City, OK 73104. ^o whom correspondence and reprint requests should be ad dressed. Abbreviations used: LM, Listeria monocytogenes; MFI, mean fluorescence intensity; mAb, monoclonal antibody,- PEC, peritoneal (exúdate) cells; PG, prostaglandin; PUFA, polyunsaturated fatty acids. Diet abbreviations: MONO diet, purified diet containing (by weight) 20% high oleate sunflower oil; (n-3) PUFA diet, purified diet containing (by weight) 17% menhaden fish oil and 3% high linoleate sunflower oil; (n-6) PUFA diet, purified diet containing (by weight) 20% high linoleate sunflower oil; SAT diet, purified diet containing (by weight) 17% coconut oil and 3% high linoleate sunflower oil.

The immune response is a complex biological process and is highly regulated through the interac tions among antigen-presenting cells, T cells, B cells and their mediators. Alteration of the immune cell populations and their mediators can influence immune responsiveness (1). The expression of major histocompatibility antigens is also crucial during an immune response, because major histocompatibility antigens are required for antigen-presenting cells to 0022-3166/92

$3.00 ©1992 American

Institute

of Nutrition.

Received 30 July 1991. Accepted

1219

14 January 1992.

Downloaded from https://academic.oup.com/jn/article-abstract/122/6/1219/4754845 by University of Rhode Island user on 09 December 2018

SHÜ-CAI HUANG,3 MICHAEL L MISFELDT* AND KEVIN L PRITSCHE4

1220

HUANG

ET AL.

TABLE 1 Fatty acid composition of the diets Diet1 Fatty acid2

jn-6) PUFA

(n-3) PUFA

SAT

g/100 g fatty

n-9)18:2(n-6)18:3(n-6)18:3(n-3) and

18:4(n-3)20:5(n-3)22:5(n-3|22:6(n-3)SFA and

(8-18:0)MUFATotal

Downloaded from https://academic.oup.com/jn/article-abstract/122/6/1219/4754845 by University of Rhode Island user on 09 December 2018

acids8:010:012:014:016:016:l(n-7)18:018:l(n-7

MONO

(n-3)Total |n-6)Total PUFANDNDND1.06.8tr5.016.967.9tr0.40.70.2ND12.817.01.368.169.4NDNDNDtr3.8tr4.577.012.0tr0.31.00.3ND8.377.11.612.013.6NDNDND5.815.57.73.513.212.3 'The (n-6) PUFA diet contained 20% linoleic sunflower oil (Beatrice/Hunt-Wesson, Fullerton, CA); the MONO diet contained 20% oleic sunflower oil (SVO Enterprises, Painesville, OH); the (n-3) PUFA diet contained 17% menhaden fish oil (Zapata Haynie, Reedville, VA) plus 3% linoleic sunflower oil; the SAT diet contained 17% coconut oil (Karlshamns USA Inc., Harrison, NJ) plus 3% linoleic sunflower oil. 2Fatty acids are denoted by the number of carbons: the number of double bonds, followed by the position of the first double bond relative to the methyl-end ("n-"|. SFA - saturated fatty acids, MUFA - monounsaturated fatty acids, PUFA = polyunsaturated fatty acids,- tr = trace (90%. Determination of immune cell population and la antigen expression. To assess the T cell population, a monoclonal antibody (mAb) against Thy-1.2 was used. This mAb reacts with all T cells from mice that express the Thy-1.2 alÃ-ele(15). A mAb specific for B220 was used to analyze B cells. B220 is a pan-B marker and is expressed on all B cells and a very small portion of immature T cells in the thymus (16). A mAb directed specifically against mouse macro phages, F4/80, was used for the determination of the macrophage population. F4/80 is expressed solely on the macrophages in mice (17). For analysis of la an tigen expression, a mAb against the la molecule, I-Ek, was used. All mAb used in this study were purified from hybridoma supernatant by protein G affinity chromatography. The isolated PEC or splenocytes were separated into six equal aliquots (0.5-1 x IO6 cells per aliquot for PEC, 1 x IO6 cells per aliquot for splenocytes) and incubated with one of the following mAb or a non specific control: 1}TIB120 (a rat IgG mAb against the la molecule, I-Ek, from ATCC, Rockville, MD); 2) TIB107 (a rat IgG mAb against the Thy 1.2 marker on T cells from ATCC); 3) F4/80 (a rat mAb against a macrophage-specific marker from ATCC); 4} PBS as a nonspecific control; 5) an R-phycoerytherin con jugated mAb (R-PE B220) against the B220 marker specific for B cells (a rat IgG2a isotype, Caltag Labora tories, South San Francisco, CA); 6) an R-

1221

1222

HUANG

for phospholipid fatty acid composition. Splenocytes from noninfected mice were also subjected to lipid extraction, phospholipid separation and fatty acid analysis. Total cellular lipids were extracted as described by Sun (18). Lipids were resuspended in chloroformmethanol (2:1, v/v), applied to HPTLC silica gel 60 plates (E. Merck, Darmstadt, Germany) and developed in a solvent system containing hexane-diethyl etheracetic acid (85:15:2, v/v/v) for 30 min. After develop ment, solvent on the plates was removed by blowing with an air gun for 10 min, and the TLC plates were sprayed with 2',7'-dichlorofluorescein (2.5 mmol/L methanol). The total phospholipid spot was visualized under a UV lamp and was scraped into 30-mL screwcapped tubes (Teflon-lined caps). Twenty-five micrograms of 17:0 fatty acid methyl ester was added to each tube as an internal standard. Fatty acid trans-

methylation was conducted by adding 1 mL of 0.5 mol/L NaOH per tube for 10 min, followed by 2 mL of chloroform for another 10 min. Fatty acid methyl esters were then extracted by adding 0.75 mL of water. Samples (bottom layer) were then filtered through sodium sulfate and the fatty acid methyl esters were analyzed using a Hewlett-Packard 5890 gas Chromatograph (Hewlett-Packard, Norwalk, CT) equipped with a 30-m x 0.25-mm i.d. fused silica capillary column (Supelco, Bellefonte, PA). The fatty acid methyl esters were identified by comparing rel ative retention times of commercial standard PUFA-1 and PUFA-2 (Supelco). Data are expressed as moles per 100 moles of the total fatty acid methyl esters identified. Three PEC samples and three splenocyte samples per treatment were analyzed for fatty acid composition of total phospholipids. Determination of prostaglandin E¿production by PEC. Thioglycollate-elicited PEC (2 x IO6) were resus pended in 0.5 mL of RPMI-1640 containing 2.5 ug of the calcium ionophore, A23187 (Sigma Chemical, St. Louis, MO). The cells were cultured in a 37°C, 5% CÛ2, humidified incubator for 1 h. Supernatants were harvested by centrifugation for 1 min at 12,000 x g and stored at -80°C for later analysis for PG. Prosta glandin £2was analyzed by enzyme immunoassay (AMI Research Products, Cambridge, MA). Anti-PGE2 is reported by the manufacturer to have a cross-reac tivity of 50% with PGEi and 6% with PGAi and

TABLE 2 Cell yield and flow cytometric analysis of la antigen expression and immune cell populations in the peritoneum and Listeria monocytogene (LM)-infected mice fed various dietary fats1'2 Noninfected (n-6) PUFA diet

MONO diet(n-3)

LM-infected (n = 12)3

(n - 9) PUFA dietSAT

1CT6)Total

of noninfected

PUFA diet

dietcell diet(n-6) number

MONO diet

(n-3) PUFA diet

(x

cellyieldIa+

0.76ab± 1.12a 5.35 ±0.33b 5.92 ±0.56ab 0.41± 3.020.32 0.39± cells4T 0.19± 0.32± 0.30 2.63 ±0.20 2.43 ±0.23 1.330.09ab 0.15± 0.05b± 0.05b± 0.12a± 0.15a 0.73 ±0.08ab 0.43 ±0.07b cells4B 0.250.31 0.17b± 0.10b 1.02 ±0.06b 1.51 ±0.18a cells4Macrophages43.561.570.321.640.62±±±±±0.61 0.15± 1.430.15 0.16± 0.27atotal± 0.27a 1.21 ±0.16ab 0.83 ±0.17b 0.12percentageIa+ 0.52± 0.143.621.530.441.590.50±±±±±0.340.210.07a0.190.113.371.690.281.810.58± the44.210.315.024.2± 3.4ab± 3.8ab 48.7 44.81.3b 1.2± 1.6± 0.8b± 1.2b± 0.4ab± 1.2b 13.4 8.02.9 cellsB 2.7b± 2.7b 19.4 2.3± 48.82.1 3.2± cellsMacrophages42.78.046.316.0±±±±1.8 3.5a 22.1 2.07.883.100.921.081.77of 15.5± 3.046.511.650.715.1±±±±2.01.5a2.21.944.67.947.713.4±

cellsT

SAT

±1.5a 41.4 ±1.1a 6.9 ±2.0ab 26.1 ±2.2ab 13.5

7.19 ± 3.18 0.89 1.06 1.78

±1.6b 46.4 ±0.7C 11.3 ±3.3a 15.1 ±2.7b 25.4

± ± ± ± ± ± ± ±3.8a

'Values are means ±SEM.Values within each challenge group (i.e., noninfected or LM-infected) that do not share a common letter are significantly different from each other (P < 0.05|. 2Weanling female C3H/Hen mice were fed a diet containing either 20% high linoleate sunflower oil [(n-6) PUFA diet), 20% high oleate sunflower oil (MONO diet), 17% menhaden fish oil plus 3% high linoleate sunflower oil ((n-3) PUFA diet], or 17% coconut oil plus 3% high linoleate sunflower oil (SAT diet) for 6 wk before immune cell collection and analysis. 3Listeria monocytogene-infected mice were injected intraperitoneally with 5 x IO4 live LM, and peritoneal exúdate cells were obtained at d 4 of infection. 4Immune cells were stained for I-Ek (la antigens), Thy 1.2 (T cells), B220 (B cells) or F4/80 (macrophages) microfluorimetry

as described in Materials

and Methods.

and analyzed by flow

Downloaded from https://academic.oup.com/jn/article-abstract/122/6/1219/4754845 by University of Rhode Island user on 09 December 2018

control mice were injected intraperitoneally with 3 mL of sterile thioglycollate broth (40 g/L). Four days later, PEC were collected from the mice as described previously. Cells were washed twice with RPMI-1640 medium and counted. Cell viability was determined using the trypan blue exclusion test and was generally >95%. An aliquot (2 x IO6 cells) was removed for measuring PG production. The remaining cells were washed once with PBS and resuspended in 1 mL of 10 mmol/L EDTA and stored at -20°C for later analysis

ET AL.

FAT AND IMMUNE

CELL POPULATIONS

ro q es -H d d d d

H **

SO

-H-H-H-H-H

SI

Tf

h^

O

-H-H-H-H

O 03 m o

m so ec o\

GO CO SO Tt

Tt

—

Tt

O

—

U

J3 E

S^'S

—i

dö

o

D « O« U _^ "?

o o

E "J

u

— d d d

-H -H -H -H

s •«• J3 ^iu

-H-H-H-H

g O

¿

C

ü £

T3

_

f-

*

fO

*

C â„¢

f > 3 " W w u O T3 _i U 2 fa H ta

so O eo so O in e gï Sn S u rt — u >,

sii ° E |S g E o

§ff -"

a Ic -2 c

J3

3 "S.

Downloaded from https://academic.oup.com/jn/article-abstract/122/6/1219/4754845 by University of Rhode Island user on 09 December 2018

u

U,

1223

o\ •*

es 1^

es

AND la ANTIGEN

1224

HUANG

RESULTS Animal weight gain, spleen weights, total number of splenocytes and peritoneal cells. Dietary fat sources did not influence body weight gain of either LM-infected or noninfected mice (data not shown). The number of PEC in noninfected animals was not altered by dietary fat. However, when mice were chal lenged with Listeria, the number of peritoneal cells was significantly affected by dietary fat (P < 0.05), with the mice fed the MONO diet having lower cell yield than mice fed the (n-6) PUFA diet (Table 2). The total number of PEC in all treatment groups increased significantly upon LM infection (P < 0.0001). Spleen weight and spleen cell yield were higher (P < 0.05) in noninfected mice fed the MONO and (n-3) PUFA diets than those fed the (n-6) PUFA and SAT diets (Table 3). Listeria infection caused an enlargement of the spleens (P < 0.0001). Spleens were 12-22% heavier in the LM-infected mice fed the MONO and (n-3) PUFA diets than in those fed the SAT diets (P < 0.05). Furthermore, spleen cell yields of LM-infected mice were lowest in those fed the SAT diet (P < 0.05). Because dietary fat influenced the cell yields, we ana lyzed and present our immune cell population and la antigen data in both relative (i.e., percentage) and absolute terms (i.e., total number of positive cells). la antigen expression. In noninfected mice, dietary fat did not influence the percentage or the number of PEC that expressed la antigen (Table 2). However, in LM-infected mice, the percentage of Ia+ PEC was sig nificantly influenced by dietary fat source (P < 0.05). In addition, dietary fat had a profound effect (P < 0.005) on the relative number of la molecules ex pressed on PEC as determined by la MFI (Fig. 1). Mice fed the (n-3) PUFA diet expressed significantly fewer la molecules on their PEC compared with the groups fed the (n-6) PUFA and SAT diets before, and com pared with all groups after, LM infection. In the spleen of noninfected mice, dietary fat did not influence the percentage of Ia+ cells (Table 3). However, total number of Ia+ cells per spleen was higher in mice fed the MONO and (n-3) PUFA diets than in those fed the (n-6) PUFA and SAT diets (P < 0.05). Listeria infection increased (P < 0.05) the total

number of splenic Ia+ cells in mice fed the (n-6) PUFA, (n-3) PUFA and SAT diets compared with their respective noninfected groups. Although the per centage of splenic Ia+ cells was highest in the LMinfected mice fed the SAT diet, the total number of splenic Ia+ cells was similar for all diet treatment groups. Dietary fat also influenced (P < 0.05) the la MFI of splenocytes from the LM-infected mice but not those from the noninfected mice (Fig. 1). Feeding the (n-3) PUFA and MONO diets resulted in a small, but significant, reduction in la MFI compared with feeding the SAT diet (P < 0.05). We also examined whether dietary fat affected the splenic B cell population that expressed la antigens (B+/Ia+)in some of these mice. Data from flow cytometric analysis for B220 (B cells) and la expression of splenocytes from both LM-infected and noninfected mice are shown in Figure 2. The B+/Ia+ and B~/Ia~ populations accounted for -90% of the cells in the spleen, with each representing -45%. The remaining cell types (i.e., B+/Ia~and B~/Ia+)made up about 3-5% of the total cell population. In the spleens of nonin fected mice there was no effect of dietary fat source on the relationship between B220 and la expression. However, when mice were infected with Listeria, there were shifts in B220 and la expression that differed with the dietary fat source. For example, the percentage of splenic B+/Ia+ increased to >50% and B~/Ia~ populations decreased to MONO diet > (n3) PUFA diet (P < 0.05).

DISCUSSION In the present investigation, we have demonstrated that dietary fat influenced la antigen expression on peritoneal cells in mice, especially those infected with Listeria. This is significant because the ex pression of la molecules on antigen-presenting cells is required for antigen presentation and activation of T cells, the crucial step during an immune response (1, 2). The reduction in la antigen expression of PEC due to (n-3) PUFA (e.g., fish oil) feeding may benefit pa tients with autoimmune diseases associated with over-expression of la antigens. Consumption of fish oil by arthritic patients has been shown to reduce the incidence and course of the disease (21). A beneficial effect of feeding fish oil to autoimmune-susceptible mice was observed by Kelley et al. (22). These authors showed that the percentage of resident peritoneal macrophages that expressed la antigen increased spontaneously in the MRL-lpr mice fed a safflower oil [rich in (n-6) PUFA diet], whereas fish oil feeding blunted the age-related increase in la antigen ex pression in these mice. Recently, Mosquera et al. (23)

Downloaded from https://academic.oup.com/jn/article-abstract/122/6/1219/4754845 by University of Rhode Island user on 09 December 2018

Non-infected

ET AL.

FAT AND IMMUNE

CELL POPULATIONS

AND la ANTIGEN

1227

TABLE 4 Effect of dietary fat on fatty acid content of murine peritoneal exúdate cell total phospbolipids1 Diet2 Fatty acid3

(n-6) PUFA

MONO

(n-3) PUFA

SAT

SEM

(8-18:0)MUFATotal

Downloaded from https://academic.oup.com/jn/article-abstract/122/6/1219/4754845 by University of Rhode Island user on 09 December 2018

mol14:016:016:118:018:118:2|n-6)20:4|n-6)20:5(n-3)22:l|n-9)22:4(n-6)22:5|n-6)22:5|n-3)22:6(n-3)SFA moll '100

6b13.6a17.1d30.6b2.0a29.2b1.9a23.4b10.2b6.5b13.4a0.3e1.3b4.7a3.7a0.6e0.9 PUFATotal (n-3) PUFATotal (n-6) PUFA0.4b28.2b1.1*27.1a6.1b12.9a11.8a0.4e1.0b4.8a2.3b0.5e0.8e55.7b8.3e1.7e34.3a36.0a0.5b28.5b1.0b24.2b16.4a4.6e9.1b1.5b2.4a3.4b3.2a1.7b1.9b53.2e19.8a5.2b2 ^Values are means, with pooled SEMshown in the right-hand column. Values with different letters within the same fatty acid group are significantly different at P < 0.05; n-3 per treatment. 2The (n-6) PUFA diet contained 20% linoleic sunflower oil (Beatrice/Hunt-Wesson, Fullerton, CA); the MONO diet contained 20% oleic sunflower oil (SVO Enterprises, Painesville, OH); the (n-3) PUFA diet contained 17% menhaden fish oil (Zapata Haynie, Reedville, VA) plus 3% linoleic sunflower oil; the SAT diet contained 17% coconut oil (Karlshamns USA Inc., Harrison, NJ) plus 3% linoleic sunflower oil. 3Fatty acids are denoted by the number of carbons: the number of double bonds, followed by the position of the first double bond relative to the methyl-end ("n-"). SFA = saturated fatty acids, MUFA = monounsaturated fatty acids, PUFA = polyunsaturated fatty acids.

showed that administration of fish oil to mice or rats by esophageal gavage reduced the percentage of peritoneal macrophages that expressed la antigen in comparison to a saline-gavaged control group. Unfor tunately, these authors failed to control for the effect of the increased energy intake and reduced food intake in the oil-gavaged mice. In addition, stress from the esophageal gavage may have distorted the results because stress influences la antigen expression (8). Our use of flow cytometric analysis has several important advantages over the traditional fluorescent microscopic technique used by Mosquera et al. (23) and Kelley et al. (22). First, it allowed us to determine not only the percentage of la positive cells, but also the relative number of la molecules expressed per cell. Second, it overcame the limitation of micro scopic analysis from a relatively small number of cells (e.g., 200 cells per sample). Third, it allowed us to measure overall Ia+ cells instead of limiting the analysis to just the adherent cell population (i.e., macrophages). This is important because other cells, such as B cells, can express la antigen and function as antigen-presenting cells (1). We showed here that di etary fat affected not only the percentage of PEC that

expressed la antigen but also the relative number of la molecules expressed on the cell surface during in fection (Table 2, Fig. 1). Our data suggest that the impact of dietary fat on la expression is not uniform across all immune tis sues. In the spleen of the LM-infected mice, the effect of dietary fat on la expression was of a lower mag nitude compared with the effect in the peritoneum. Splenic Ia+ cell data are almost superimposable on the splenic B cell data (Table 3). Therefore, it seems that the effect of dietary fat on la antigen expression in the spleen can be explained, in large part, by the changes in the B cell population. By two-color staining, we further demonstrated that the influence of dietary fat source on the percentage of Ia+ cells in the spleen of LM-infected mice was limited to the B cell population that expresses la antigens (Fig. 2). It is not understood at present why consuming fish oil reduces murine PEC la expression. It has been shown that PGE2 suppresses la antigen expression (5). However, because PEC from mice fed the (n-3) PUFA diet produced less PGE compared with PEC from mice fed other fat sources, the reduction in la ex pression due to the (n-3) PUFA feeding would not be related to the suppressive effect of PGE. Similarly, Mosquera et al. (23) showed that the reduction in the

1228

HUANG

ET AL.

TABLE5 Effect of dietary fat on phospholipid fatty acid content of murine splenocytes1 Diet-1 Fatty acid3

(n-6) PUFA

MONO

(n-3) PUFA

SAT

SEM

|8-18:0)MUFATotal

Downloaded from https://academic.oup.com/jn/article-abstract/122/6/1219/4754845 by University of Rhode Island user on 09 December 2018

mol14:016:016:118:018:118:2(n-6)20:4(n-6)20:5(n-3)22:l(n-9)22:4(n-6)22:5(n-6)22:5(n-3)22:6(n-3)SFA mol/100

lb23.9a2.7b20.6b23.2b1.5a39.1a2.6a26.5b8.6b5.3b3.5b4.7a1.5b0.4b1.3C2.9a4.8a67.1a12.7b12.4a1 4b2.9b23.3ab26.2a0.20.90.11.00.50.40.40.10.10.10.10

PUFATotal (n-3) PUFATotal (n-6) PUFA0.5b35.5b0.9b27.8ab6.7C11.2a10.6a0.5b1.2b2.0a2.4b0.8C0.7b63.8a8.8e2.0b26.2a28.2a0.4b30.1e1.3b23.5e19.5a3.5e11.3a0.4b3.2a1.9a3.9a0.7C1.6b54. Values are means, with pooled SEMshown in the right-hand column. Values with different letters within the same fatty acid group are significantly different at P < 0.05; n - 3 per treatment. 2The (n-6) PUFA diet contained 20% linoleic sunflower oil (Beatrice/Hunt-Wesson, Fullerton, CA); the MONO diet contained 20% oleic sunflower oil (SVO Enterprises, Painesville, OH); the (n-3) PUFA diet contained 17% menhaden fish oil (Zapata Haynie, Reedville, VA) plus 3% linoleic sunflower oil; the SAT diet contained 17% coconut oil (Karlshamns USA Inc., Harrison, NJ) plus 3% linoleic sunflower oil. 3Fatty acids are denoted by the number of carbons: the number of double bonds, followed by the position of the first double bond relative to the methyl-end ("n-"|. SFA »saturated fatty acids, MUFA - monounsaturated fatty acids, PUFA = polyunsaturated fatty acids.

percentage of la-positive macrophages in the peritoneum of mice due to fish oil gavaging was not correlated with the PGE2 level. We postulate that several mechanisms may be involved in the modu lation of la expression by dietary fish oil. First, the reduction in la antigen expression caused by the (n-3) PUFA feeding may be related to the incorporation of (n-3) PUFA into membrane phospholipids (Tables 4 and 5). Such changes in the fatty acid composition of phospholipids can influence cell membrane fluidity and thus membrane-associated receptor and enzyme activities (10, 24). Second, dietary fish oil has been shown to increase the cAMP levels in rat heart homogenates (24). Because cAMP has been shown to play a major role in the suppression of la antigen expression (25), the reduction in la antigen expression caused by the (n-3) PUFA feeding may be related to the suppressive effect of cAMP. Third, fish oil feeding has been shown to increase susceptibility to lipid peroxidation (26). An increase in lipid peroxidation due to fish oil or (n-3) PUFA feeding may influence the expression of la antigen, because free radicals can suppress la antigen expression (27). Fourth, fish oil feeding has been shown to reduce the responsiveness of macrophages to y-interferon stimulation (28). Be cause y-interferon is required to up-regulate la antigen

expression during Listeria infection (4), the decrease in the cellular response to y-interferon due to fish oil or (n-3) PUFA feeding may reduce the expression of la antigen. Clearly, further research is warranted in order to understand the underlying mechanisms in volved in the modulation of la antigen expression by dietary fat. Our data further demonstrate that both the per centage and the total number of T cells in the peritoneum of noninfected mice fed the (n-3) PUFA diet were generally higher than those in mice fed other diets (Table 2, Fig. 3). Such an increase in T cells may be related to the enhancement by fish oil of cellular-associated interleukin-1 level in resident peritoneal macrophages (12), because interleukin-1 has been shown to influence the activation and prolif eration of T cells (1, 2). Interestingly, we observed that Listeria infection resulted in significant increases in both the percentage and the total number of T cells in all treatments except the (n-3) PUFA-fed group (Table 2, Fig. 3). It is possible that the activation of T cells (e.g., helper T cells) is impaired in part because of the reduction of la antigen expression during in fection (1, 3). In addition, proliferation of the T cells in the peritoneum may be reduced by (n-3) PUFA feeding (29, 30). Migration or infiltration of T cells

FAT AND IMMUNE CELL POPULATIONS

1229

high (n-6) PUFA diet or high saturated fatty acid diet upon challenge with sheep red blood cells. Dietary fat did not influence the percentage or total number of peritoneal macrophages in noninfected mice as assessed by the expression of the F4/80 marker (Table 2). A similar observation was made by Somers et al. (28), who counted cells with a light microscope and based their results on cell mor phology. However, when mice were infected with Listeiia, both the percentage and the total number of peritoneal macrophages were lower in mice fed the (n3) PUFA diet compared with those fed the (n-6) PUFA or SAT diets. Such an effect could be due to the lower chemotactic leukotriene production observed in fish oil-fed animals (10). However, in the spleen such a lowering effect due to fish oil feeding was not ob served. In our study both the percentage and the number of splenic macrophages in (n-3) PUFA-fed mice were not significantly different than those in mice fed the other diets. This indicates that the ef fects of dietary fat source on the macrophage popu lation may differ according to the location of macro phages or the site of infection. We observed that dietary fat influenced the PEC yield in the LM-infected mice but not in the nonin fected mice (Table 2). Similar observations have been reported (39) for fish oil-fed vs. corn oil-fed BALB/c mice both before and after a viral challenge. Such an effect seems to correlate with the PEC phospholipid fatty acid composition. Locomotion and/or migration of immune cells has been shown to be influenced by differences in membrane fatty acid composition (31). In addition, changes in chemotactic leukotriene pro duction induced by dietary fats may in part account for such an effect, because leukotrienes have a pro found effect on immune cell migration (10). Unlike results in the peritoneum, the spleen weight and spleen cell yield were higher in mice fed the MONO and (n-3) PUFA diets than in those fed the (n-6) PUFA and SAT diets (Table 3). The enlargement of the spleen and the increase in spleen cell yield in the absence of any overt immune challenge was observed previously in fish oil-fed BALB/c mice (39). In con trast, Kelley et al. (22) showed that feeding MRL-lpr mice a fish oil diet significantly reduced the total number of splenocytes compared with feeding a safflower oil [rich in (n-6) PUFA] diet. The MRL-lpr mice are autoimmune-prone mice that suffer from ex cessive lymphoproliferation, lymphoid hyperplasia and over-expression of la antigen with age. The con trast between the data of Kelley et al. (22) and our data suggests that autoimmune-prone mice may re spond differently to dietary fish oils than do normal mice. In conclusion, the source of dietary fat can in fluence murine peritoneal and splenic immune cell populations and la antigen expression. Clearly, the impact of dietary fat was of greater magnitude in the peritoneum than in the spleen. Furthermore, although

Downloaded from https://academic.oup.com/jn/article-abstract/122/6/1219/4754845 by University of Rhode Island user on 09 December 2018

into the peritoneum may also be influenced by cel lular membrane fatty acid composition and chemotactic leukotriene production (10, 31). Further more, fish oil feeding has been shown to cause a reduction in interleukin-1 production upon a bacterial endotoxin challenge (32). Such a reduction in inter leukin-1 may decrease the activation and proliferation of T cells (1, 2). In the spleen, the percentage of T cells in LM-infected mice fed the (n-3) PUFA diet was lower than in those fed the (n-6) PUFA diet (Table 3). Interestingly, when our spleen data were analyzed based on the cell number (Table 3, Fig. 3), this dif ference was no longer observed. Thus, caution should be exercised in the interpretation of data concerning the effect of dietary fats on immune cell populations. Both T helper and T cytotoxic cells play an im portant role in immune response and survival after Listeria infection (33). Rubin et al. (34) showed that there was no difference in survival between the fish oil-fed and lard-fed (NZB/NZWJF! mice subjected to an LDso dose of Listeiia infection. Unfortunately, the strain of mice they used has been shown to carry a gene resistant to Listeiia infection, whereas the strain of mice used in our study is genetically susceptible to Listeiia infection (35). Because the dose of Listeiia we used to infect the mice is 10 times less than the LDso (14), no animals died during the period of infection. It would be of interest to assess the impact of fish oil feeding on the survival of these susceptible strains of mice (e.g., C3H/Hen mice) upon a Listeiia challenge at the LDso dose. The greater number of peritoneal B cells in the (n-3) PUFA-fed mice than in the other groups during LM infection may be due to the differences in B cell proliferation modulated by the dietary fats. Leslie et al. (36) showed that spleen cells from arthritis-prone mice fed fish oil had significantly higher proliferative responses to the B cell mitogen lipopolysaccharide. The decrease in PGE production due to the (n-3) PUFA feeding may in part account for the elevation in B cells, because PGE has been shown to impair B cell responsiveness (37). Unlike the peritoneum, the spleen from LM-infected mice fed the MONO and (n3) PUFA diets had a lower percentage of B cells com pared with spleen from mice fed the SAT diet. In contrast to the percentage, the number of splenic B cells was higher in the noninfected mice fed the MONO or (n-3) PUFA diet compared with those fed the (n-6) PUFA or SAT diet. This is directly correlated with the higher total spleen cell yield in these mice. It is noteworthy that in our study the groups fed the (n6) PUFA and SAT diets had similar splenocyte phospholipid fatty acid composition (Table 5). This may explain why the number of splenic B cells did not differ between mice fed the (n-6) PUFA and SAT diets. Similarly, Erickson et al. (38) observed no differences in the percentage and the number of splenic B cells among mice fed an essential fatty acid-deficient diet,

AND la ANTIGEN

1230

HUANG

there are some similarities, cells from the spleen and peritoneum often responded differently to the various dietary fats. This suggests that observations of immunomodulation by diet in one type of cell or tissue may not be readily extrapolated to others.

We are grateful to Zapata Haynie Inc. (Reedville, VA) for supplying the menhaden oil, Karlshamns USA Inc. (Harrison, NJ) for supplying coconut oil, and SVO Enterprises (Painesville, OH) for supplying the high oleate sunflower oil used in this research project. We thank Louis Henry and David Lafrenz for their help in flow microfluorimetry analysis, Grace Y. Sun for her technical advice in analysis of phospholipid fatty acids, John N. Berg for providing Listeria monocytogenes, and Nancy A. Cassity, research specialist, for technical assistance.

LITERATURE CITED 1. Vitetta, E. S., Femandez-Botran, R., Myers, C. D. & Sanders, V. M. (1989) Cellular interactions in the humoral immune re sponse. Adv. Immunol. 45: 1-105. 2. Unanue, E. R. &. Cerottini, J-C. (1989) Antigen presentation. FASEB I. 3: 2496-2502. 3. Malis, L. A., Glimcher, L. K, Paul, W. E. & Schwartz, R. H. (1983) Magnitude of response of histocompatibility-restricted T-cell clones is a function of the product of the concentrations of antigen and la molecules. Proc. Nati. Acad. Sci. U.S.A. 80: 6019-6023. 4. Koga, T., Mitsuyama, M., Manda, T., Watanabe, Y. & Nomoto, K. (1987) Gamma interferon-mediated increase in the number of la-bearing macrophages during infection with Listeiia monocytogenes. Infect. Immun. 55: 2300-2303. 5. Snyder, D. S., Beller, D. I. & Unanue, E. R. (1982) Prostaglandins modulate macrophage la expression. Nature (Lond.) 299: 163-165. 6. Lu, C. Y., Changelian, P. S. & Unanue, E. R. (1984) aFetoprotein inhibits macrophage expression of la antigens. J. Immunol. 132: 1722-1727. 7. Snyder, D. D. & Unanue, E. R. (1982) Corticosteroids inhibit murine macrophage la expression and interleukin-1 produc tion. J. Immunol. 129: 1803-1805. 8. Zwilling, B. S., Dinkins, M., Christner, R., Paris, M., Griffin, A., Hilburger, M., McPeek, M. & Pearl, D. (1990) Restraint stress-induced suppression of major histocompatibility complex class n expression by murine peritoneal macrophages. J. Neuroimmunol. 29: 125-130. 9. Johnston, P. V. & Marshall, L. A. (1984) Dietary fat, prostaglandins and the immune response. Prog. Food & Nutr. Sci. 8: 3-35. 10. Kinsella, ]. E., Lokesh, B., Broughton, S. & Whelan, ]. (1990) Dietary polyunsaturated fatty acids and eicosanoids: potential effects on the modulation of inflammatory and immune cells: an overview. Nutrition 6: 24-44. 11. Endres, S., Ghorbani, R., Kelley, V. E., Georgilis, K., Lonnemann, G., van de Meer, J.W.M., Cannon, J. G., Rogers, T. S., Klempner, M. S., Weber, P. C., Schaefer, E. J., Wolff, S. M. & Dinarello, C. A. (1989) The effect of dietary supplementation with n-3 polyunsaturated fatty acids on the synthesis of inter leukin-1 and tumor necrosis factor by mononuclear cells. N. Engl. J. Med. 320: 265-271.

12. Lokesh, B. R., Sayers, T. ]. & Kinsella, J. E. (1990) Interleukin-1 and tumor necrosis factor synthesis by mouse peritoneal mac rophages is enhanced by dietary n-3 polyunsaturated fatty acids. Immunol. Lett. 23: 281-286. 13. Meydani, S. N., Endres, S., Woods, M. M., Goldin, B. R., Soo, C., Morrill-Labrode, A., Dinarello, C. A. & Gorbach, S. L. (1991) Oral (ji-3) fatty acid supplementation suppresses cytokine production and lymphocyte proliferation: comparison between young and older women. J. Nutr. 121: 547-555. 14. Beller, D. L., Kiely, J-M. & Unanue, E. R. (1980) Regulation of macrophage populations. I. Preferential induction of la-rich peritoneal exúdales by immunologie stimuli. J. Immunol. 124: 1426-1432. 15. Ledbetter, J. A. & Herzenberg, L. A. (1979) Xenogeneic mono clonal antibodies lo mouse lymphoid differeniiaiion aniigens. Immunol. Rev. 47: 63-90. 16. Cause, W. C., Mountz, J. D. & Steinberg, A. D. (1988) Charac terization and differentialion of CD4~, CD8~ ihymocytes sorted with the Ly-24 marker. J. Immunol. 140: 1-7. 17. Ausi vu, J. M. & Gordon, S. (1981) F4/80, a monoclonal an tibody directed specifically againsi ihe mouse macrophage. Eur. J. Immunol. 11: 805-815. 18. Sun, G. Y. (1988) Preparaiion and analysis of acyl and alkenyl groups of glycerophospholipids from brain subcellular mem branes. In: Lipids and Related Compounds (Boulton, A. A., Baker, G. B. & Horrocks, L. A., eds.), pp. 63-82. Humana Press, Clifton, NJ. 19. Cochran, W. G. & Cox, G. M. (1957) Analysis of ihe resulis of a series of experimenis. In: Experimental Designs (Cochran, W. G. & Cox, G. M., eds.), pp. 545-568. John Wiley & Sons, New York, NY. 20. SAS Institute Inc. (1988) SAS User's Guide: Statistics. SAS Institule, Cary, NC. 21. Kremer, J. M., Lawrence, D. A., Jubiz, W., Digiacomo, R., Rynes, R., Bartholomew, L. E. & Sherman, M. (1990) Dietary fish oil and olive oil supplementation in patients with rheumaioid arthriiis: clinical and immunologie effects. Arthrilis Rheum. 33: 810-820. 22. Kelley, V. E., Ferreiti, A., Izui, S. & Strom, T. B. (1985) A fish oil diet rich in eicosapentaenoic acid reduces cyclooxygenase metabolites, and suppresses lupus in MRL-lpr mice. J. Im munol. 134: 1914-1919. 23. Mosquera, J., Roddriguez-Iturbe, B. & Parra, G. (1990) Fish oil dietary supplementation reduces la expression in ral and mouse peritoneal macrophages. Clin. Immunol. Immunopath. 56: 124-129. 24. Alam, S. Q., Ren, Y-F. & Alam, B. S. (1988) (3H)Forskolin- and (3H)dihydroalprenolol-binding sites and adenylate cyclase ac

25.

26.

27.

28.

29.

30.

tivity in heart of rats fed diels coniaining differenl oils. Lipids 23: 207-213. Figueiredo, F., Uhing, R. J., Okonogi, K., Gellys, T. M., Johnson, S. P., Adams, D. O. & Prpic, V. (1990) Aciivaiion of the cAMP cascade inhibits an early event involved in murine macrophage la expression. J. Biol. Chem. 265: 12317-12323. Meydani, M., Natiello, F., Goldin, B., Free, N., Woods, M., Schaefer, E., Blumberg, J. B. & Gorbach, S. L. (1991) Effect of long-term fish oil supplementation on vitamin E staius and lipid peroxidaiion in women. J. Nuir. 121: 484-491. Grüner,S., Volk, H-D., Falck, P. & Baehr, R. V. (1986) The influence of phagocytic stimuli on the expression of HLA-DR aniigens; role of reactive oxygen intermediales. Eur. J. Im munol. 16: 212-215. Somers, S. D., Chapkin, R. S. & Erickson, K. L. (1989) Alteraiion of in vilro murine periioneal macrophage function by dietary enrichment with eicosapemaenoic and docosahexaenoic acids in menhaden fish oil. Cell. Immunol. 123: 201-211. Fritsche, K. L., Cassity, N. A. & Huang, S.-C. (1991) Effect of dietary fai source on anlibody produclion and lymphocyle proliferaiion in chickens. Poult. Sci. 70: 611-617. Kelley, D. S., Branch, L. B., Love, J. E., Taylor, P. C., Rivera, Y.

Downloaded from https://academic.oup.com/jn/article-abstract/122/6/1219/4754845 by University of Rhode Island user on 09 December 2018

ACKNOWLEDGMENTS

ET AL.

FAT AND IMMUNE CELL POPULATIONS

31. 32.

34.

1231

35. Cheers, C. & McKenzie, I.F.C. (1978) Resistance and suscepti bility of mice to bacterial infection: genetics of listeriosis. Infect. Immun. 19: 755-762. 36. Leslie, C., Meydani, S., Cathcart, E. S., Hayes, K. C., Gonnerman, W. A. & Dubey, D. (1984) Enhancement of B cell function in fish oil fed arthritis susceptible mice. Fed. Proc. 43: 1991 (abs.|. 37. Schad, V. C. & Phipps, R. P. (1988)Two signals are required for accessory cells to induce B cell unresponsiveness: tolerogenic Ig and prostaglandin. J. Immunol. 141: 79-84. 38. Erickson, K. L., Adams, D. A. & Scibienski, R. J. (1986)Dietary fatty acid modulation of murine B-cell responsiveness. J. Nutr. 116: 1830-1840. 39. Pritsche, K. L. & Johnston, P. V. (1990)Effect of dietary omega3 fatty acids on cell-mediated cytotoxic activity in BALB/c mice. Nutr. Res. 10: 577-588.

Downloaded from https://academic.oup.com/jn/article-abstract/122/6/1219/4754845 by University of Rhode Island user on 09 December 2018

33.

M. & lacono, J. M. (1991| Dietary linolenic acid and immunocompetence in humans. Am. J. Clin. Nutr. 53: 40-46. Novo, C., Fonseca, E. & Freitas, A. A. (1987)Altered fatty acid membrane composition modifies lymphocyte localization in vivo. Cell. Immunol. 106: 387-396. Billiar, T. R., Bankey, P. E., Svingen, B. A., Curran, R. D., West, M. A., Holman, R. T., Simmons, R. L. & Cerra, F. B. (1988) Fatty acid intake and Kupffer cell function: fish oil alters eicosanoid and monokine production to endotoxin stimula tion. Surgery (St. Louis) 104: 343-349. Kaufmann, S.H.E. (1988) Listeiia monocytogenes specific Tcell lines and clones. Infection 16 (suppl. 2): 128-136. Rubin, R. H., Wilkinson, R. A., Xu, L. & Robinson, D. R. (1989) Dietary marine lipid does not alter susceptibility of (NZB x NZW)FI mice to pathogenic microorganisms. Prostaglandins 38: 251-262.

AND la ANTIGEN