1 Soy isoflavones modulate immune function in healthy ... - CiteSeerX

4 downloads 44 Views 249KB Size Report
Department of Food Science and Human Nutrition, Washington State ... Supported by a grant from the Washington State Attorney General's Office; soymilk. 15.
1 1

Soy isoflavones modulate immune function in healthy postmenopausal women

2 3

Tracy A. Ryan-Borchers, Jean Soon Park, Boon P. Chew, Michelle K. McGuire, Lisa R.

4

Fournier, Kathy A. Beerman

5 6 7

From the Department of Food Science and Human Nutrition (TAR-B, JSP, BPC, MKM,

8

KAB), and Department of Psychology (LRF), Washington State University, Pullman,

9

WA 99164-6376.

10 11

Address correspondence about the manuscript and reprint requests to B. P. Chew,

12

Department of Food Science and Human Nutrition, Washington State University,

13

Pullman, WA 99164-6376.

14 15

Supported by a grant from the Washington State Attorney General’s Office; soymilk

16

provided by White Wave, Inc. (Boulder, CO); and isoflavone (Novasoy®) and placebo

17

supplements provided by the Archer Daniels Midland Co. (Decatur, IL).

18 19 20

Running head: Isoflavones modulate postmenopausal immunity

2 1

ABSTRACT

2

Background: After menopause, the immune system may be compromised due to effects

3

of aging and diminishing concentrations of estrogen, an immune-modulating hormone.

4

Isoflavones, plant-derived compounds with estrogenic and antioxidant properties, may

5

offer immunological benefits to women in this stage of life.

6

Objective: The objective of this study was to evaluate the effects of soy isoflavones,

7

both soymilk and supplement form, on markers of immunity and oxidative stress in

8

postmenopausal women.

9

Design: Postmenopausal women aged 50-65 y (n = 52) enrolled in this 16 wk double-

10

blind, placebo-controlled trial were randomized to one of three experimental groups: 1)

11

Control – 706 mL/d cow’s milk plus a placebo supplement; 2) Soymilk – 71.6 mg

12

isoflavones derived from 706 mL/d soymilk plus a placebo supplement; and 3)

13

Supplement – 70 mg isoflavones in a supplement plus 706 mL/d cow’s milk. Plasma and

14

24-h urine samples were obtained at baseline and at 16 wk. Immune variables included

15

lymphocyte subsets, cytokine production, as well as markers of inflammation and

16

oxidative damage.

17

Results: Baseline immune variables did not differ among groups. Both isoflavone

18

interventions increased (p < 0.05) B cell populations and concentrations of plasma

19

interferon-gamma (IFN-γ) and decreased (p < 0.05) concentrations of 8-hydroxy-2-

20

deoxy-guanosine (8-OHdg), an oxidative marker of DNA damage. Isoflavone treatments

21

did not influence concentrations of urinary 8-isoprostane (8-iso), plasma interleukin-2

22

(IL-2), tumor necrosis factor-alpha (TNF-α) or C-reactive protein (CRP).

3 1

Conclusions: Soymilk and supplemental isoflavones modulate B cell populations and

2

IFN-γ concentrations, and appear protective against DNA damage in postmenopausal

3

women.

4 5

KEY WORDS

6

soymilk

7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

Isoflavones, immune function, postmenopausal women, DNA damage,

4 1

INTRODUCTION

2

Isoflavones are a subclass of flavonoids, a broad group of polyphenolic compounds

3

widely distributed in foods of plant origin. Biological effects attributed to flavonoids

4

include anti-estrogenic and pro-estrogenic, as well as antioxidant and antiproliferative

5

actions (1). The three main isoflavone derivatives are daidzein, genistein and glycitein,

6

which are found predominantly in soybeans and other legumes (2). During digestive and

7

absorptive processes, isoflavones often undergo further metabolic transformations (3).

8

For example, daidzein can be converted to equol by intestinal microflora. The biological

9

activities of this mammalian derived isoflavone are thought to be stronger than its parent

10 11

compound, daidzein (4). Isoflavones have been classically defined as phytoestrogens, compounds that exert

12

estrogenic effects. The presence of the phenolic ring enables isoflavones to bind to both

13

types of estrogen receptors, Erα and Erβ (5). Unlike endogenous estrogen, however,

14

isoflavones bind with a much higher affinity to Erβ, which suggests they may be more

15

accurately defined as selective estrogen receptor modulators (SERMs) capable of both

16

pro-estrogenic and anti-estrogenic effects (6). This hormonal mechanism explains, in

17

part, how isoflavones protect against age-related diseases, including cardiovascular

18

disease (7, 8), osteoporosis (9), and cancers of the breast (10) and prostate (11, 12).

19

Antioxidant and antiproliferative properties of isoflavones offer additional, important

20

mechanisms for their protection against many prevalent chronic diseases (13, 14).

21

Cellular damage resulting from oxidative stress is believed to be a major contributor to

22

the etiology of cardiovascular disease through the oxidation of LDL, and cancer by

23

causing DNA strand breaks which may lead to mutations (15-17). In humans,

5 1

isoflavones have been found to prevent LDL oxidation in vivo (16), as well as inhibit

2

DNA damage in vitro (17).

3

The immune system encompasses an array of defenses, which help guard against the

4

development of age-related diseases. However, like other bodily systems, its functions

5

can be adversely affected by factors such as oxidative damage and hormonal changes

6

(18). Therefore, the immune system may benefit from the various biological properties

7

of isoflavones. Enhanced immune responses have been found in animals fed genistein,

8

the isoflavone most abundant in soy foods (19, 20). In humans, consumption of

9

isoflavone-containing soy foods modulated cytokine production (21), while daidzein and

10

genistein glucuronides enhanced the activity of natural killer (NK) cells in vitro (22).

11

Postmenopausal women, in particular, are susceptible to chronic diseases associated

12

with aging, as well as major shifts in hormonal status. It is possible that isoflavones may

13

improve immune function in this population. Also, it is important to determine whether

14

dietary or supplement forms of isoflavones affect immunological variables similarly.

15

Therefore, the purpose of this study was to evaluate the effects of soy isoflavones,

16

obtained from either soymilk or supplement, on immune and oxidative markers in

17

postmenopausal women.

18 19 20 21 22 23

6 1

SUBJECTS AND METHODS

2

Subjects and study design

3

This study was conducted as part of a larger research project designed to investigate the

4

effects of soy isoflavones on cognitive function. Thus, the sample sizes were determined

5

based on expected changes and variability in cognitive, rather than immunologic

6

variables. Healthy postmenopausal women between 50 and 65 y of age were recruited

7

regionally from the states of Idaho and Washington. Potential subjects (n = 300) were

8

screened by telephone to determine study eligibility. Women free of major health

9

conditions, with a body mass index (BMI) between 18 and 40 kg/m2, and who had not

10

menstruated for at least one year were eligible to participate in the study. Additional

11

exclusion criteria included legume allergies, smoking, kidney stones and antibiotic

12

therapy within the past six months.

13

Telephone screening resulted in 150 women who were both eligible and interested in

14

enrollment; a total of 117 of these women completed the enrollment process. Of these

15

women, 5 dropped out of the study, all for personal reasons. After exclusion of women

16

using hormone replacement therapy (HRT), the first 52 women completing the cognitive

17

study composed the subset for this secondary investigation of immune and antioxidant

18

variables. All study procedures were approved by the Institutional Review Board (IRB)

19

of Washington State University.

20

Prior to intervention, subjects providing informed consent adhered to a 4 wk adjustment

21

diet that minimized intake of foods containing isoflavones. Following this washout

22

period, the 16 wk intervention period began. Subjects were enrolled in groups of 20 to 28

23

women; these groups were then considered to be treatment blocks. Within each block,

7 1

subjects were randomly assigned to receive: 1) cow’s milk and placebo supplement

2

(Control group); 2) soymilk (White Wave, Inc.; Boulder, CO) and placebo supplement

3

(Soymilk group, 71.6 + 3.1 mg isoflavones/d based on 30 samples); or 3) cow’s milk and

4

isoflavone tablets (Supplement group, 70 mg isoflavones/d). Randomization was

5

performed using random number generation. Researchers and subjects remained blinded

6

to group assignment throughout the study.

7

Nutrient composition and caloric value of the two types of milk was nearly identical

8

(Table 1). The isoflavone content and composition of the supplements (30 mg daidzein,

9

33 mg genistein and 7 mg glycetin) were formulated to match those of the soymilk and

10

values are verified by Archer Daniels Midland Co. (Novasoy®; Decatur, IL). Placebos

11

were composed of maltodextrin with 10% caramel color. The isoflavone content of the

12

soymilk was determined by HPLC (P. Murphy, Iowa State University) and is presented in

13

Table 1. Milk products were delivered to each subject weekly.

14

Subjects were instructed to consume a total of 706 mL milk/d (soymilk or cow’s milk)

15

for a total of 16 wk. Milk was consumed in the morning (353 mL) and in the evening

16

(353 mL). The cow’s milk was treated with flavoring and color agents to resemble the

17

flavor and appearance of soymilk. Supplements (isoflavone or placebo tablet) were taken

18

with the milk, twice per day, throughout the duration of the intervention period. The

19

research team’s dietitian provided individual instruction, along with written guidelines, to

20

the women regarding replacement of usual dairy or other similar food items with the

21

study milk to allow for consistent macronutrient intake and prevention of weight gain

22

during the trial. Compliance was assessed via personal communication, dietary records

23

and appearance of isoflavone metabolites in plasma and urine.

8 1

Blood and urine samples were obtained at baseline (0 wk) and post-intervention

2

(16 wk). Blood was drawn into evacuated collection tubes (10 mL) coated with EDTA.

3

An aliquot of whole blood from each subject was used for flow cytometry analysis

4

immediately after the collection. The remainder was centrifuged for 30 min and aliquots

5

of the plasma were frozen at -80 °C until analyzed for interleukin-2 (IL-2), interferon-

6

gamma (IFN-γ) and tumor necrosis factor-alpha (TNF-α); the acute-phase protein, C-

7

reactive protein (CRP), the oxidative marker, 8-hydroxy-2-deoxy-guanosine (8-OHdg)

8

and isoflavone concentrations. Two, 24-h urine collections (0 wk and 16 wk) were

9

obtained from participants. Total urine volume was recorded, and aliquots frozen at

10

-20 °C until analyzed for the oxidative marker, 8-isoprostane (8-iso), as well as urinary

11

isoflavone concentrations.

12

Analytical procedures

13

Isoflavone concentrations

14

The isoflavone concentrations in plasma and urine were measured by HPLC as

15

previously described (23, 24). Briefly, β-glucuronidase-sulfatase H-2 (EC 3.2.1.31, from

16

Helix pomatia, Sigma Chemical, St. Louis, MO) was added to plasma (1 mL) and urine

17

(5 mL), and the mixtures were incubated at 37 °C for 20 h to release isoflavone

18

aglycones. Samples were then loaded onto Extrelut QE columns (EM Science,

19

Gibbstown, NJ) and extracted with 8 mL (plasma) or 12 mL (urine) of ethyl acetate. The

20

eluent was collected and dried under nitrogen gas. Extracted isoflavones were dissolved

21

in 80% methanol in water for HPLC analysis. An (Waters, Milford, MA) HPLC system

22

equipped with a photodiode array detector (PDA 996) was used to distinguish the

23

ultraviolet spectra of the specific isoflavone compounds: daidzein, equol and genistein.

9 1

Isoflavones were separated in a µ-Bondapak C18 reverse-phase column (Waters, Milford,

2

MA; 3.9 mm x 30 cm). The samples were eluted using a linear gradient of 40% to 70%

3

methanol in 30 min at a flow rate of 1 mL/min. Retention times for the isoflavones were

4

as follows: daidzein at 14.9 min, equol at 16.8 min, and genistein at 19.5 min. Daidzein,

5

equol and genistein were detected at 248, 280 and 260 nm, respectively. The identity of

6

each isoflavone was confirmed by comparing its absorption spectrum with that of a

7

corresponding standard. Plasma and urine isoflavone concentrations were calculated

8

based on standard curves constructed for daidzein (Valeant Pharmaceuticals, USA),

9

genistein (Calbiochem, USA) and equol (Sigma-Aldrich, USA).

10

Lymphocyte subsets

11

Specific populations of lymphocytes were measured using dual color flow cytometric

12

analysis (FACScalibur equipped with Cell Quest software, BD Biosciences, San Diego,

13

CA): Total T cells (CD3+CD19-), T cytotoxic cells (Tc; CD3+CD8+), T helper cells (Th;

14

CD3+CD4+), Th/Tc ratio (CD3+CD4+/CD3+CD8+), B cells (CD3-CD19+) and NK cells

15

(CD3-/CD16+56+). Whole blood aliquots obtained from subjects were treated with a lysis

16

buffer to avoid contamination by erythrocytes (25). After centrifugation, the resultant

17

cell pellet was washed, and cells were labeled with monoclonal antibodies conjugated

18

with fluorescein isothiocyanate (FITC) or phycoerythrin (PE): anti-CD3 conjugated to

19

FITC, and anti-CD8, anti-CD4 or anti-CD19 conjugated to PE (Caltag Laboratories,

20

Burlingame, CA). Cell suspensions were fixed with paraformaldehyde and analyzed

21

using a leucogate, or automatically established lymphocyte analysis gate. The leucogate

22

helps define and distinguish the lymphocytes from other blood cell types. A total of 2000

23

events were counted for each sample.

10 1

Cytokine production

2

Plasma samples were analyzed by enzyme-linked immunosorbent assays (ELISA) for

3

IFN-γ (Human IFN-γ BD OptEIA™ ELISA Set, BD Biosciences, San Diego, CA), TNF-

4

α (BD OptEIA™ Set Human TNF, BD Biosciences), and IL-2 (BD OptEIA™ Set

5

Human IL-2, BD Biosciences). In these assays, wells were coated with capture antibody,

6

diluted in coating buffer and incubated overnight at 4 °C. Samples were pipetted into

7

appropriate wells and incubated for 2 h at room temperature. Detection antibody with

8

avidin-horseradish peroxidase was added to plates for color development and incubated

9

for 30 min. A stop solution was then added and absorbance read on a plate reader (Bio-

10

Tek Instruments, Inc., Winooski, VT) at 450 nm.

11

C-reactive protein

12

Plasma samples were analyzed for CRP by ELISA (Human CRP ELISA kit, Alpha

13

Diagnostic International, San Antonio, TX). The human CRP kit is a sandwich-type

14

ELISA based on the CRP in samples binding simultaneously to two antibodies, one fixed

15

to the microtiter well plates and the other conjugated to horseradish peroxidase. After

16

color development, absorbance was read at 450 nm. Concentration of CRP in samples

17

was based on the standard curve.

18

Lipid peroxidation

19

8-Isoprostane was measured in purified urine by an enzyme immunoassay (EIA)

20

method (8-Isoprostane EIA kit, Cayman Chemical Company, Ann Arbor, MI). This

21

assay is based on the competition between 8-iso and an 8-iso-acetylcholinesterase

22

(AChE) tracer. Concentration of tracer is held constant, while the amount of 8-iso in

23

urine samples varies. The 8-iso antibody binds to the IgG mouse monoclonal antibody

11 1

that has been previously attached to the wells. After washing, color is developed with

2

Ellman’s Reagent and absorbance measured at 410 nm.

3

DNA damage

4

Plasma concentrations of 8-OHdg were measured by ELISA (Bioxytech 8-OHdg-EIA™

5

kit, OXIS Health Products, Inc., Portland, OR). Samples and standards were added to

6

microtiter plate wells that had been precoated with 8-OHdg. The 8-OHdg present in both

7

the wells and samples competes for binding with the monoclonal antibody to 8-OHdg.

8

Plates were washed, leaving only the antibody that had bound to the 8-OHdg coating the

9

wells. Secondary antibody conjugated to horseradish peroxidase was then added and

10

bound to the monoclonal antibody remaining on the plate. Chromogen was added

11

following another wash step, color was developed, and absorbance measured at 450 nm.

12

Statistical analysis

13

Data were analyzed by ANOVA using the General Linear Model procedure of SAS

14

statistical software (version 8; 26). The initial statistical model included the effects of

15

block, treatment, period, and the interaction of treatment by period. Upon preliminary

16

statistical analyses, it was determined that there were no effects of block. Thus, this

17

variable was removed from the statistical models prior to further analyses. Treatment

18

means were compared using Tukey test; group BMI means (0 wk and 16 wk) were

19

evaluated using paired t tests, and equol production among the groups was analyzed by

20

Chi-square analysis. Statistical significance was set at p < 0.05.

21 22 23

12 1

RESULTS

2

Subjects

3

Demographic characteristics are presented in Table 2. A total of 52 (95%)

4

postmenopausal women completed the study. Mean age of study participants did not

5

differ by group (Table 2). There was no significant group difference in body mass index

6

(BMI) at baseline or post-intervention. Ninety-four percent of the study participants were

7

Caucasian and the majority (88%) received education beyond the high school level.

8

Study treatments, including cow’s milk, soymilk and supplemental tablets, were tolerated

9

well overall. Compliance with study protocol was very good as indicated by verbal

10

reports and written dietary records from the women, as well as by the absence (control) or

11

presence (isoflavone interventions) of isoflavone metabolites in plasma and urine samples

12

of the participants.

13

Isoflavone concentrations

14

At baseline, plasma and urinary isoflavone concentrations were non-detectable among

15

the women in all experimental groups, confirming dietary compliance during the

16

adjustment period. Following the 16 wk intervention, plasma concentrations of genistein

17

and equol were significantly higher (p < 0.05) in both isoflavone treatment groups, but

18

remained non-detectable in the control group (Figure 1). Plasma daidzein was detectable

19

in all women and was significantly higher (p < 0.05) in the supplement group compared

20

to control. Urinary concentrations of daidzein, genistein and equol were higher (p