Original Research Ingestion of a Dietary Supplement Containing Dehydroepiandrosterone (DHEA) and Androstenedione Has Minimal Effect on Immune Function in Middle-Aged Men

Marian L. Kohut, PhD, James R. Thompson, MS, Jeff Campbell, BA, Greg A. Brown, MS, Matthew D. Vukovich, PhD, Dave A. Jackson, MS, Doug S. King, PhD Department of Health and Human Performance, Iowa State University, Ames, Iowa Key words: aging, cytokines, lymphocyte, hormones, androstenedione, DHEA

Objective: This study investigated the effects of four weeks of intake of a supplement containing dehydro- epiandrosterone (DHEA), androstenedione and herbal extracts on immune function in middle-aged men. Design: Subjects consumed either an oral placebo or an oral supplement for four weeks. The supplement contained a total daily dose of 150 mg DHEA, 300 mg androstenedione, 750 mg Tribulus terrestris, 625 mg chrysin, 300 mg indole-3-carbinol and 540 mg saw palmetto. Measurements: Peripheral blood mononuclear cells were used to assess phytohemagglutinin(PHA)-induced lymphocyte proliferation and cytokine production. The cytokines measured were interleukin (IL)-2, IL-4, IL-10, IL-1␤, and interferon (IFN)-␥. Serum free testosterone, androstenedione, estradiol, dihydrotestosterone (DHT) were also measured. Results: The supplement significantly increased serum levels of androstenedione, free testosterone, estradiol and DHT during week 1 to week 4. Supplement intake did not affect LPS or ConA proliferation and had minimal effect on PHA-induced proliferation. LPS-induced production of IL-1beta, and PHA-induced IL-2, IL-4, IL-10, or IFN-gamma production was not altered by the supplement. The addition of the same supplement, DHEA or androstenedione alone to lymphocyte cultures in vitro did not alter lymphocyte proliferation, IL-2, IL-10, or IFN-␥, but did increase IL-4. In addition, serum HDL-C concentration significantly declined. Conclusion: These findings suggest that, although chronic intake of a complex dietary supplement contain- ing DHEA, androstenedione and herbal extracts increases serum androgen levels, it has minimal effect on immune function in middle-aged men.

INTRODUCTION nutritional supplements containing different combinations of these androgenic steroid hormones are sold, the potential ef- The plasma concentration of several androgenic steroid fects of these supplements on immune function have not been hormones declines with advancing age [1]; therefore, the older characterized. adult population may consume supplements containing andro- Immune function may decline with age, and the results from genic steroids in an effort to reverse age-associated changes several studies suggest that supplementation with the steroid such as the decrease in immune function. In addition, strength/ dehydroepiandrosterone (DHEA) may modify the age-related power athletes may use these substances in an attempt to changes of immune response. It is well documented that the enhance performance. Various combinations of herbal extracts production of the cytokine interleukin-2 (IL-2) declines with have been added to the androgenic steroid-containing supple- age [2]. It has been suggested that a dysregulation in cytokine ments that are sold over the counter, in an attempt to maximize balance (reduced T helper type 1 (Th1) cytokines, increased T the “androgenic effect” of these dietary supplements. Although helper type 2 (Th2) cytokines) may also accompany the aging

Address correspondence to: Marian L. Kohut, Ph.D., Department of Health and Human Performance, Iowa State University, 235 Forker Building, Ames IA 50011. E-mail: [email protected].

Journal of the American College of Nutrition, Vol. 22, No. 5, 363–371 (2003) Published by the American College of Nutrition

363 DHEA, Androstenedione and Immunity process [2–5]. IL-2 and interferon-␥ (IFN-␥) are Th1 cytokines preventing conversion to DHT [22], indole-3-carbinol and that drive cell-mediated immune responses, whereas Th2 cyto- chrysin, which prevent aromatization of androgens [23,24], and kines promote humoral-mediated immunity. The age-related Tribulus terrestris, suggested to increase serum lutenizing hor- dysregulation in Th1/Th2 cytokine balance may contribute to mone, although this claim has not been verified in a controlled an impaired ability to defend against infection [3]. Studies trial. These herbal extracts are not known to exhibit significant using animal models have demonstrated that one of the steroids immunomodulatory properties. The purpose of this study was contained in the over-the-counter “androgenic dietary supple- to investigate the immunological effects of four weeks of oral ments,” dehydroepiandrosterone (DHEA), may reverse the age- ingestion of a supplement containing DHEA, androstenedione, related decline of immune function and provide protection from saw palmetto, Tribulus terrestris, chrysin and indole-3-carbinol infection [6]. Aged mice treated with DHEA exhibit enhanced in middle-aged men. Serum levels of several steroid hormones interleukin (IL)-2 and interferon (IFN)-␥ production with no were also monitored. In addition, blood samples from middle- effect on IL-4 or IL-10, perhaps reversing the age-associated aged men were collected to assess the in vitro effects of the change in Th1/Th2 cytokine balance [4,5]. supplement and its androgenic components. In addition to its potential effect on cytokines, other studies have shown that the same component, DHEA, contained in “androgenic dietary supplements” has a protective effect METHODS against infection with numerous pathogens (M. tuberculosis, West Nile virus, Herpes simplex type, Coxsackie virus) in mice Subjects [7–9]. However, it is important to note that lifelong treatment Sixteen healthy men between the ages of 50 and 59 not with DHEA did not reverse immunosenescence or enhance taking medications or dietary supplements were recruited for resistance to infection in rodents [10]. Fewer human studies this double-blind study. Written informed consent was obtained exist, and there is limited evidence suggesting that DHEA can from each participant, and the protocol was approved by the reverse immunosenescence. In several trials, oral DHEA ad- Iowa State University Human Subjects Review Board. Each ministration did not significantly improve the immune response subject completed a medical history, to eliminate subjects with to influenza immunization [11–13]. The effect of “androgenic known chronic disease, and a dietary evaluation. Subjects were dietary supplements” on overall immune function is not well randomly assigned to consume either a rice flour oral placebo studied in humans. Based on two reports of healthy, older (n ϭ 8) or an oral dose of the dietary supplement (n ϭ 8) for adults, some aspects of immune function may be enhanced by 28 days. The supplement contained 150 mg DHEA, 300 mg DHEA treatment, whereas others show no effect or are inhib- androstenedione, 750 mg Tribulus terrestris, 625 mg chrysin, ited [14,15]. The effect of a complex “androgenic dietary 300 mg indole-3-carbinol and 540 mg saw palmetto. The sup- supplement” on immune responsiveness has not been studied in plement was consumed in three equal doses taken before 0900, humans to our knowledge. at 1500 and before bedtime. The supplement was supplied by Other weak androgenic steroids include dihydrotestosterone Experimental and Applied Sciences, Golden, CO, and content (DHT), androstenedione, androstenediol and androstenetriol. of the supplement was verified by HPLC at an independent These steroids may also enhance immunity and protect against laboratory (Integrated Biomolecule, Tuscon, AZ). Subjects infection [16–19]. DHT is the most potent androgen [20] and were asked to maintain normal diet and activity patterns binds to the androgen receptor more tightly than any other throughout the study. Subjects recorded dietary intake and androgen [21]. Some investigators have demonstrated that the activity for the two days prior to each blood draw. Diet analysis beneficial effect of these other steroid hormones, in terms of was performed with Nutritionist 4 software. Six additional immune function, may be greater than DHEA [17–19]. How- young male subjects (ages 21 to 32) were included only as an ever, the immunoenhancing effect of these steroids or combi- age comparison group. These young subjects also completed a nation of steroids has only been observed in animal models, medical history and a dietary evaluation to eliminate any sub- and, to our knowledge, these have not been tested in humans. jects with known chronic disease and/or consuming dietary The metabolic fate of ingested DHEA and androstenedione supplements. The young subjects did not consume oral placebo is not clear. DHEA may be converted to androstenedione or or dietary supplement. Instead, this group was included only to androstenediol, and either of these steroids may undergo further document the age-related changes of immune function. conversion to testosterone, a more potent androgen. Testoster- one can undergo further conversion to DHT by the enzyme Blood Collection and Peripheral Blood 5-alpha reductase or may undergo aromatization resulting in Mononuclear Cell (PBMC) Isolation estrogen formation. In order to maximize the “androgenic” effect of these steroids and reduce the possibility of steroid Fasting blood samples were collected from the older sub- conversion to DHT or estrogens, several herbal extracts have jects between 6:30 and 8:00 am prior to supplementation and been added to nutritional supplements. These extracts include once per week during supplementation. Blood was collected saw palmetto extract, which may inhibit 5-alpha reductase, between 6:30 and 8:00 am from the young subjects at week 0,

364 VOL. 22, NO. 5 DHEA, Androstenedione and Immunity

week 1 and week 4. PBMC were isolated by density gradient exception that cells were not frozen and thawed. Instead, all centrifugation (Ficoll-Paque Plus, Amershem Pharmacia Bio- blood was collected, cells were isolated and cultured on the tech, NJ). Cells were washed two times with sterile Hank’s same day. Cells were counted and adjusted to 1 ϫ 106 cells/ Balanced Salt Solution (HBSS; Sigma Chemical Co. St. Louis, mL. 100 ␮L of cells were added to 96 well plates for the MO), suspended in RPMI (Life Technologies, Rockville, MD) mitogen-induced proliferation assay and cytokine production. supplemented with 10 mM Hepes (Sigma Chemical Co.), 50 The proliferation assay was performed as described above. U/mL penicillin, 50 mg/mL streptomycin (both from Life Both the proliferation assay and the cytokine assay used mito- Technologies) and 5% fetal bovine serum (FBS; JRH Bio- gen and supplement components at the following doses. Wells sciences, Lenexa, KS), and counted using a hemocytometer. contained one of the following: PHA (5 ␮g/mL), LPS (100 All cells were then frozen in RPMI media containing 30% FBS ng/mL) or media plus DHEA (10Ϫ5–10Ϫ9 M), androstenedione and dimethylsulfoxide (DMSO) and stored in liquid nitrogen (10Ϫ5–10Ϫ9 M), complete dietary supplement with DHEA and until later analysis. All cell samples were thawed on the same androstenedione concentrations at 10Ϫ5–10Ϫ9M, or media day to perform the mitogen-stimulated proliferation and cyto- alone. The steroids were first dissolved in ethanol and then kine production assays to limit inter-assay variability. After diluted to the appropriate concentrations. The media control thawing, the cells were washed twice in supplemented RPMI wells contained the same final concentration of ethanol as the and adjusted to the appropriate concentration. Cell viability was wells containing steroid (0.00001–0.1% ethanol). Plates were Ն 85% after thawing, and cell counts were adjusted based on incubated for 72 hours at 37°Cin5%CO2, and proliferation the number of live cells. was assessed using MTT as described above. In the cytokine

analyses, cells were incubated 37°Cin5%CO2 for 24 to 48 hours, Peripheral Blood Mononuclear Cell (PBMC) and the supernatant was collected from the appropriate wells Proliferation according to the following time points. IL-2 and IFN␥ were Ϯ PBMC proliferation was measured using a colorimetric measured after 24 hours incubation with PHA supplement assay described by Mosmann (J Immunol Methods 65:55–63, components, IL-4 and IL-10 were measured after 48 hours incu- Ϯ 1983). Cells were plated at 1 ϫ 105 cells per well in 96-well, bation with PHA supplement components, and IL-1 was mea- Ϯ flat bottom tissue culture plates (Corning, Corning, NY), con- sured after 24 hours incubation with LPS supplement compo- taining a final concentration of 5 ␮g/mL phytohemagglutinin nents. A standard ELISA kit specific for each of the cytokines to (PHA; Sigma Chemical Co.) or media alone in a total volume be assayed was used (OptEIA Sets, PharMingen, San Diego, CA). ␮ of 100 L. Plates were incubated at 37°Cin5%CO2 for 72 hours. During the last four hours of incubation, 10 ␮Lof5.0 Serum Hormones and Blood Chemistry mg/mL 3-(4,5-dimethylthiozol-2yl)-2,5-diphenyltetrazolium The serum concentrations of androstenedione, estradiol, free bromide (MTT, Sigma Chemical Co.), in PBS was added to testosterone and total testosterone from the older subjects were each well. Then, 100 ␮L of 0.04N HCl in isopropanol was measured with commercial radioimmunoassay kits (Diagnostic added to dissolve the colored precipitate. Absorbance was read Products Corp. Los Angeles, CA and Diagnostic Systems Labo- at a dual wavelength of 570 and 630 nm using an automated ratory, Webster, TX). An ELISA kit was used to measure serum plate reader (BioRad Instrument Co., Richmond, CA). level of dihydrotestosterone (Immuno-Biological Laboratories, Hamburg, Germany). A commercial laboratory performed blood Cytokine Production: IL-1␤, IL-2, IL-4, IL-10, chemistry analysis (Quest Diagnostics, Wood Dale, IL). IL-12 and IFN-␥ Cytokine concentration was measured in supernatant 24 or Statistics 48 hours after the addition of LPS or PHA. Control wells Statistical analyses were performed to determine the effect received media without mitogens. IFN-␥ and IL-2 was mea- of supplement using a 2 factor mixed ANOVA (time ϫ sup- sured after 24 hours in culture with PHA (5 ␮g/mL). IL-4 and plement). A 2 factor mixed ANOVA (time ϫ age) was used to IL-10 was measured after 48 hours in culture with PHA (5 test the effect of age. A one-way ANOVA was used to test the ␮g/mL). IL-1␤ was measured after 24 hours in culture with potential effects of the supplement components added to cells LPS, whereas IL-12 was measured after 48 hours in culture in vitro. All analyses were performed with SPSS software with LPS (100 ng/mL). A standard ELISA kit specific for each (SPSS Inc., Chicago, IL). of the six cytokines to be assayed was used (OptEIA Sets, PharMingen, San Diego, CA). RESULTS In Vitro Effects of DHEA, Androstenedione and Supplement Diet/Activity Blood samples were collected from five men (40 to 58 years There was no difference between the placebo and supple- of age), and PBMC were isolated as described above with the ment groups with respect to any of the nutrients analyzed with

JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION 365 DHEA, Androstenedione and Immunity

the Nutritionist-4 software. Also, activity patterns were not different between the two groups.

PHA-Induced Proliferation With respect to PHA-induced proliferation, there were main effects of supplement treatment and time (Fig. 1). PHA-in- duced proliferation did not differ between placebo and supple- ment groups at baseline (week 0) or at week 4, but was greater in subjects consuming the supplement at week 1 (p ϭ 0.023) (Fig. 1). A significant main effect of age was observed (young Ͼ old, p Ͻ 0.001), and a significant age ϫ time interaction (p ϭ 0.036) was observed, suggesting that PHA proliferation did not change over time in young subjects.

Cytokine Production The production of IFN-␥ and IL-2 (Th1 cytokines) was not altered by consumption of the dietary supplement at any time point throughout the four-week period (Fig. 2). However, a significant effect of age was observed for IL-2 (young Ͼ old, p Ͻ 0.001) and IFN-␥ (young Ͼ old, p ϭ 0.001). Dietary supplementation also did not have an effect on the production of IL-4 and IL-10 (Th2 cytokines) at week 1 or at week 4 (Fig. 3). A significant main effect of age was observed with respect to IL-4 (young Ͻ old, p ϭ 0.025); however, we did not observe an effect of age on IL-10 production. Finally, the production of the pro-inflammatory cytokine IL-1␤ was not altered by the supplement (Fig. 4).

In Vitro Effect of DHEA, Androstenedione or Fig. 2. PHA-induced production of the Th1 cytokines, IL-2 and IFN-␥ ␮ Supplement on Lymphocyte-Induced Proliferation in supernatant from PBMC cultured with PHA at 5 g/mL. PBMC were obtained from middle-aged subjects consuming either placebo or PBMC were cultured with DHEA, androstenedione or the supplement, or from young subjects consuming no supplement at the complete dietary supplement Ϯ PHA. No significant prolifera- time points indicated. Results are expressed as mean Ϯ SEM. tion was observed in wells containing hormones or supplement without PHA (results not shown). As expected, PHA did induce proliferation, but the addition of DHEA, androstenedione or the complete supplement did not alter proliferation at any of the concentrations tested (10Ϫ5 M, 10Ϫ7 M, 10Ϫ9 M) (Fig. 5).

In Vitro Effect of DHEA, Androstenedione or Supplement on Cytokine Production PHA was added to cultures to induce cytokine production. Wells that contained hormones or supplement but not PHA did not produce detectable levels of any of the cytokines measured (results not shown). The in vitro addition of DHEA (10Ϫ7 M), androstenedione (10Ϫ7 M) or supplement (10Ϫ7 M) did not alter the production of IL-2, IFN-␥, or IL-10 (Figs. 6, 7). Fig. 1. PHA-induced proliferation in PBMC from middle-aged subjects However, IL-4 production was enhanced by the addition of the ϭ consuming either placebo or supplement and young subjects that did supplement to PBMC cultures (p 0.004) (Fig. 7). There was not consume placebo or supplement. Results are expressed as mean Ϯ a trend towards enhanced IL-4 production in wells containing SEM. Asterisk indicates significant differences (p Ͻ 0.05) between DHEA (p ϭ 0.065) and in wells containing androstenedione placebo and supplement. (p ϭ 0.094).

366 VOL. 22, NO. 5 DHEA, Androstenedione and Immunity

Fig. 4. LPS-induced production of IL-1␤ in supernatant from PBMC culture with LPS at 100 ng/mL. PBMC were obtained from middle- aged subjects consuming either placebo or supplement at the time points indicated. Results are expressed as mean Ϯ SEM.

Fig. 3. PHA-induced production of the Th2 cytokines, IL-4 and IL-10 in supernatant from PBMC cultured with PHA at 5 ␮g/mL. PBMC were obtained from middle-aged subjects consuming either placebo or supplement, or from young subjects that did not consume placebo or supplement at the time points indicated. Results are expressed as Fig. 5. PBMC obtained from middle-aged men and incubated with mean Ϯ SEM. PHA (5 ␮g/mL) plus media, the dietary supplement (containing andro- stenedione and DHEA at 10Ϫ7 M), androstenedione (10Ϫ7 M) or Ϫ7 Ϯ Serum Hormone Levels DHEA (10 M). Results are expressed as mean SEM. Although serum levels of androstenedione, free testoster- one, DHT and estradiol did not differ between groups at base- androgens and herbal extracts designed for maximal “andro- line, dietary supplementation resulted in increased serum levels genic effect” do not strongly support the use of oral androgen of androstenedione, free testosterone, DHT and estradiol at plus herbal supplementation with respect to immune benefits. It weeks 1 to 4 (Table 1). Total testosterone was not altered by has been suggested that weak androgens such as DHEA may dietary supplementation. Total cholesterol and LDL-choles- reverse the age-associated dysregulation of the Th1/Th2 cyto- terol did not change (results not shown); however, a significant kine balance by increasing the production of Th1 cytokines, decrease in HDL-cholesterol concentration was observed in IL-2 and IFN-␥, with little or no effect on the Th2 cytokines, subjects consuming the supplement (Table 1). IL-4 and IL-10 [4,5]. However, it is important to note that a DHEA-induced enhancement of Th1 cytokine production is not consistently found among rodent studies and decreases of Th1 DISCUSSION cytokines have also been reported [25]. In our study we exam- ined the effect of an “androgenic dietary supplement” contain- Although some rodent studies have noted an enhancement ing DHEA on both Th1 and Th2 cytokine production in healthy of immune function following administration of weak andro- older adults. Four weeks of oral intake of a supplement con- gens, our results in humans using a supplement containing taining DHEA (150 mg/day), androstenedione (300 mg/day),

JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION 367 DHEA, Androstenedione and Immunity

Fig. 6. IL-2 and IFN-␥ in supernatant from PBMC obtained from Fig. 7. IL-10 and IL-4 in supernatant from PBMC obtained from middle- middle-aged men and incubated with PHA (5 ␮g/mL) plus media, the aged men and incubated with PHA (5 ␮g/mL) plus media, the dietary dietary supplement (containing androstenedione and DHEA at 10Ϫ7 M), supplement (containing androstenedione and DHEA at 10Ϫ7 M), andro- androstenedione (10Ϫ7 M) or DHEA (10Ϫ7 M). Results are expressed as stenedione (10Ϫ7 M) or DHEA (10Ϫ7 M). Results are expressed as mean Ϯ SEM. mean Ϯ SEM. Asterisk indicates significant difference (p Ͻ 0.05) between cultures containing media alone and cultures containing dietary supple- plus herbal extracts, had no effect on PHA stimulated produc- ment. A trend (p Ͻ 0.1) towards a difference between cultures containing tion of IL-2, IL-4, IL-10 or IFN-␥. We did observe an effect of media alone and cultures containing DHEA or androstenedione is indi- age on Th1 and Th2 cytokine production that is consistent with cated by ϩ. other research (decreased Th1, increased Th2). The Th1 cyto- kines, IL-2 and IFN-␥, were lower among the 50 to 59 year old androstenedione in those subjects consuming the supplement subjects as compared to younger subjects whereas production compared to those subjects consuming placebo, yet this change of the Th2 cytokine, IL-4, was greater among the older sub- in serum level of androstenedione did not appear to be associ- jects. We have also consistently observed that the variability in ated with altered cytokines. cytokine production is greater in older subjects than young In addition, we added DHEA, androstenedioene or supple- subjects. Differing rates of immunosenescence among older ment in vitro to cultures containing PBMC from middle-aged subjects may explain this larger degree of variability. Based on men and found no effect on the production of IL-2, IL-10 and our findings, it appears that four weeks of oral intake of a IFN-␥. However, we did observe an increase in IL-4 production complex dietary supplement containing DHEA, androstenedi- in cultures containing the supplement and a trend towards one and herbal extracts was not sufficient to reverse the age- increased IL-4 in cultures containing DHEA or androstenedi- related changes in Th1/Th2 cytokine balance. However, given one. An increase in IL-4 (Th2 cytokine) does not support the that we did not test the effect of DHEA or androstenedione theory that these steroids can reverse the age-associated change supplementation alone, it remains possible that oral intake of in cytokine balance. Instead, IL-4 has been reported to increase these steroids alone or in combination may have altered Th1/ with age. Therefore, our results suggest that oral intake of a Th2 cytokines. We did document an increase in serum levels of supplement containing DHEA and androstenedione, or addition of

368 VOL. 22, NO. 5 DHEA, Androstenedione and Immunity

Table 1. Serum Hormone and HDL-Cholesterol Concentrations in Placebo- and Supplement-Treated Groups

Androstenedione Free Testosterone (nmol/L) Estradiol (nmol/L) (pmol/L) DHT (pmol/L) HDL-C (mg/dL) Group Week Mean Ϯ S.E. Mean Ϯ S.E. Mean Ϯ S.E. Mean Ϯ S.E. Mean Ϯ S.E. Supplement 0 7.5 Ϯ 1.0 0.20 Ϯ 0.03 54.5 Ϯ 5.4 1904 Ϯ 259 40.2 Ϯ 3.1 Placebo 0 4.8 Ϯ 0.3 0.20 Ϯ 0.01 49.4 Ϯ 4.5 1961 Ϯ 263 45.3 Ϯ 5.0 Supplement 1 35.6 Ϯ 4.9* 0.36 Ϯ 0.05* 80.3 Ϯ 10.4* 3058 Ϯ 225* 36.0 Ϯ 2.0* Placebo 1 5.5 Ϯ 0.4 0.17 Ϯ 0.01 51.9 Ϯ 5.5 1980 Ϯ 289 44.4 Ϯ 4.7 Supplement 2 33.7 Ϯ 5.0* 0.36 Ϯ 0.06* 72.0 Ϯ 8.4* 3097 Ϯ 174* 33.5 Ϯ 2.0* Placebo 2 5.5 Ϯ 0.5 0.19 Ϯ 0.02 52.7 Ϯ 4.8 2095 Ϯ 313 44.7 Ϯ 5.0 Supplement 3 38.4 Ϯ 6.5* 0.32 Ϯ 0.05* 74.2 Ϯ 10.4* 2952 Ϯ 201* 34.7 Ϯ 2.4 Placebo 3 6.3 Ϯ 0.8 0.18 Ϯ 0.02 46.7 Ϯ 4.3 1874 Ϯ 229 46.2 Ϯ 5.1 Supplement 4 48.6 Ϯ 15.5* 0.34 Ϯ 0.06* 74.3 Ϯ 9.4* 2878 Ϯ 227* 33.7 Ϯ 2.3* Placebo 4 5.5 Ϯ 0.6 0.19 Ϯ 0.02 47.0 Ϯ 4.5 1929 Ϯ 288 44.0 Ϯ 5.0

these steroids to mononuclear cell cultures in vitro, do not reverse proliferation following either DHEA supplementation or intake the age-related dysregulation of Th1/Th2 cytokine balance. of “androgenic dietary supplements.” The findings from our study differ with the results from a This is the first study to our knowledge to examine oral study by Khorram et al. [14] in that we did not observe intake of any supplement containing androstenedione on im- increased IL-2 production; however, subjects in the Khorram mune responsiveness in humans. The results from several studies study only consumed DHEA, not an “androgenic dietary sup- using rodents suggest that weak androgens such as andro- plement.” In their study, men 53 to 69 years of age consumed stenediol and/or androstenetriol have more potent immuno- 50 mg of DHEA per day for 20 weeks. However, the increase modulatory effects than DHEA [17,18]. It has been suggested in IL-2 was observed after 20 weeks of treatment, but not seen that the metabolic conversion of DHEA to other androgens is after four weeks of supplementation. Although these authors necessary for regulation of immune function. Based on our reported an increase in several immune parameters during the results, it appears that the addition of androstenedione to a 20-week trial [14], it has been suggested that some of these effects supplement containing DHEA and other herbal extracts thought may reflect a multiple comparisons artifact [10]. A separate study to prevent the conversion of androgens to DHT or estrogen involving postmenopausal women reported no change in IL-2 does not enhance immune function in older men. It is possible production after three weeks of DHEA supplementation (50 mg/ that the combination of steroids and/or extracts contained in the day) [15]. Based on these two studies and our results, it is difficult supplement inhibited any immunoenhancing properties of an- to make definite conclusions regarding the effects of DHEA and drostenedione. However, our in vitro results are not consistent “androgenic dietary supplements” of IL-2. with this possibility since androstenedione alone did not have The ability of T cells to proliferate in response to mitogen or significant immunomodulatory effects. Also, we did observe ele- antigen also declines with age [2]. We have also observed that vated plasma levels of androstenedione in the supplement group, mitogen-induced proliferation is greater in young subjects than but this alteration in serum level did not appear to be associated in older subjects. Again, the variability in the amount of pro- with changes in lymphocyte proliferation. liferation tends to be greater among older subjects. With respect Although we did not independently test the potential immu- to mitogen-induced T cell proliferation, we observed an in- nomodulatory properties of the herbal extracts by separate oral crease in PHA-stimulated proliferation only at week one of administration of these compounds, the herbal extracts con- supplementation. Although this increase was statistically sig- tained in the supplement are not known to significantly impact nificant, it was a small effect and did not restore the amount of lymphocyte function. Indole-3-carbinol has been shown in sev- proliferation to the level seen in younger subjects. Also, given eral studies to inhibit tumor cell growth, but this appears to be the responses over time, with higher levels of proliferation at via a direct effect on the tumor cell itself rather than via week 4 in both groups and a decrease only in placebo at week activation of immune response [26,27]. In one study, large 1, it is doubtful that the statistical difference at week 1 is doses of indole-3-carbinol were fed to rats, and no significant physiologically meaningful. In the study by Khorram et al. changes were seen in lymphocyte proliferation, Natural Killer [14], an increase in PHA-induced proliferation was observed in cell activity or delayed type hypersensitivity (DTH) response subjects consuming 50 mg/day of DHEA at week 12 of sup- [28]. In a separate study, chrysin was reported to enhance plementation, but not at week 4 or week 20. The other human killing of L-929 tumor cells by increasing tumor necrosis factor study involving postmenopausal women reported a decrease in (TNF-␣) cytotoxicity, but no effects on lymphocyte function mitogen-induced T cell proliferation after three weeks of were reported [29]. The extract Serenoa repens derived from DHEA supplementation [15]. Taken together these studies and saw palmetto has been shown to reduce symptoms of benign our results do not provide strong support for enhanced T cell prostatic hyperplasia, but the mechanism of action involves

JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION 369 DHEA, Androstenedione and Immunity inhibition of 5-alpha-reductase and inhibition of DHT binding ACKNOWLEDGMENT to androgen receptors in prostate cells, rather than involvement of immune cells [30]. Therefore, there is not strong evidence This study was funded in part by a grant from Experimental that any of the herbal extracts contained in the supplement have and Applied Sciences, Golden, CO. direct effects on lymphocyte function. Although enhancement of the immune response has been reported in many, but not all, studies involving rodents and DHEA, the majority of human studies do not observe large REFERENCES immune benefits [11–13,15,31]. Perhaps one explanation for the different findings is that the adrenal glands of mice do not 1. Gray A, Feldman HA, McKinlay JB, Longcope C: Age, disease, synthesize androgens, and the circulating level of DHEA in and changing sex hormone levels in middle-aged men: results of mice has been reported to be very low or undetectable [32]. The the Massachusetts Male Aging Study. J Clin Endocrinol Metab dose of androgens administered either via diet or injection in 73:1016–1025, 1991. many rodent studies likely results in pharmacological levels of 2. Miller RA: The aging immune system: primer and prospectus. Science 273:70–73, 1996. androgen. In contrast, DHEA levels in humans altered by 3. Shearer GM: Th1/Th2 changes in aging. Mech Ageing Dev 94: DHEA administration may increase twofold to tenfold and are 1–5, 1997. generally considered to be within the physiological range 4. Inserra P, Zhang Z, Aredestani SK, Araghi-Nikman M, Liang B, [33,12]. At higher doses of androgen there may be negative Jiang S, Shaw D, Moliter M, Elliott K, Watson RR: Modulation of health consequences, such as the decline in serum HDL-cho- cytokine production by dehydroepiandrosterone (DHEA) plus mel- lesterol noted in this study, found as part of a larger investiga- atonin (MLT) supplementation of old mice. Proc Soc Exp Biol tion on androgenic steroid intakes in men 30 to 59 years of age Med 218:76–82, 1998. [34]. Therefore, in rodents, supraphysiological doses of andro- 5. Daynes RA, Dudley DJ, Araneo BA: Regulation of murine lym- gen may confer some immune benefit, and it is possible that phokine production in vivo II. Dehydroepiandrostenedione is a very large doses of androgen may enhance immune function in natural enhancer of interleukin-2 synthesis by helper T cells. Eur humans. However, there may be negative health effects in J Immunol 20:793–802, 1990. 6. Daynes RA, Araneo BA: Prevention and reversal of some age humans associated with supraphysiological doses such as re- associated changes in immunologic responses by supplemental duced HDL-cholesterol and increased risk of breast or ovarian DHEAS therapy. Aging Immunol Infect Dis 3:135–154, 1992. cancer in women [34,35]. As part of a larger study on andro- 7. Hernandez-Pando R, De La Luz Streber M, Orozco H, Arriaga K, genic dietary supplement intake in men 30 to 59 years of age, Pavon L, Al-Nakhli SA, Rook GAW: The effects of androstenediol the intake of supplements containing the same ingredients, total and dehydroepiandrosterone on the course and cytokine profile of daily dose of 150 mg DHEA, 300 mg androstenedione, 750 mg tuberculosis in BALB/c mice. Immunol 95:234–241, 1998. Tribulus terrestris, 625 mg chrysin, 300 mg indole-3-carbinol 8. Ben-Nathan D, Lustig S, Kobiler D, Danenberg HD, Lupu E, and 540 mg saw palmetto [34] or a total daily dose of 300 mg Feurstein G: Dehydroepiandrosterone protects mice inoculated androstenedione [36], demonstrated the negative health conse- with West Nile Virus and exposed to cold stress. J Med Virol quences of steroid intake, in particular, the decrease in HDL- 38:159–166, 1992. cholesterol found in all age groups with all androgenic supple- 9. Loria RM, Inge TH, Cook SS, Szaki AK, Regelson W: Protection against lethal viral infection with the native steroid dehydroepi- ment combinations. In addition, four weeks of oral intake of androstenedione. J Med Virol 26:301–314, 1988. this supplement (DHEA (150 mg/day), androstenedione (300 10. Miller RA, Chrisp C: Lifelong treatment with oral DHEA sulfate mg/day) plus herbal extracts) did not reverse the age-related does not preserve immune function, prevent disease, or improve decrease in serum total testosterone. The detailed serum hor- survival in genetically heterogenous mice. J Am Geriatr Soc 47: mone concentrations and serum lipid concentrations resulting 960–966, 1999. from oral intake of this supplement are reported elsewhere [34]. 11. Danenberg HD, Ben-Yehuda A, Zakay-Rones Z, Gross DJ, Fried- Taken together, the findings from this study demonstrate man G: Dehydroepiandrosterone treatment is not beneficial to the that oral intake of a supplement containing DHEA (150 mg/ immune response to influenza in elderly subjects. J Clin Endocri- day), androstenedione (300 mg/day) plus herbal extracts for nol Metab 82:2911–2914, 1997. four weeks does not enhance immune response among middle 12. Degelau J, Guay D, Hallgren MS: The effects of DHEAS on aged men. Serum steroid hormones and HDL-cholesterol are influenza vaccination in aging adults. J Am Geriatr Soc 45:747– 751, 1997. altered in subjects consuming the supplement, but these 13. Evans TG, Judd ME, Dowell T, Poe S, Daynes RA, Araneo BA: changes may lead to negative health consequences [34]. There- The use of oral dehydroepiandrosterone sulfate as an adjuvant in fore, it is possible that higher doses or a longer time period of tetanus and influenza vaccination of the elderly. Vaccine 14:1531– consumption of the supplement given in this study or some 1537, 1996. other combination of the ingredients found in the supplement 14. Khorram O, Vu L, Yen SS: Activation of immune function by may enhance immune function, but the health risks may out- dehydroepiandrosterone in age-advanced men. J Gerontol A Biol weigh any potential health benefits. Sci Med Sci 52:M1–7, 1997.

370 VOL. 22, NO. 5 DHEA, Androstenedione and Immunity

15. Casson PR, Andersen RN, Herod HG: Oral dehydroepiandros- Indole-3-carbinol is a negative regulator of estrogen receptor-alpha terone treatment in physiologic doses modulates immune function signaling in human tumor cells. J Nutr 130:2927–2931, 2000. in postmenopausal women. Am J Obstet Gynecol 169:1536–1539, 27. Chinni SR, Li Y, Upadhyay S, Koppolu PK, Sarkar FH: Indole- 1993. 3-carbinol induced cell growth inhibition, G1 cell cycle arrest and 16. Araneo BA, Dowell T, Diegel M, Daynes RA: Dihydrotestosterone apoptosis in prostate cancer cells. Oncogene 24:2927–2936, 2001. exerts a depressive influence on the production of interleukin-4 28. Exon JH, South EH, Magnuson BA, Hendrix K: Effects of indole- (IL-4), IL-5, and ␥-interferon, but not IL-2 by activated murine T 3-carbinol on immune responses, aberrant crypt foci, and colonic cells. Blood 78:688–699, 1991. crypt cell proliferation in rats. J Toxicol Environ Health 62:561– 17. Padgett DA, Loria RM: In vitro potentiation of lymphocyte acti- 573, 2001. vation by dehydroepiandrosterone, androstenediol, and androsten- 29. Habtemariam S: Flavonoids as inhibitors or enhancers of the etriol. J Immunol 153:1544–1452, 1994. cytotoxicity of tumor necrosis factor-alpha in L929 tumor cells. J 18. Loria RM, Padgett DA, Huynh PN: Regulation of the immune Nat Prod 60:775–778, 1997. response by dehydroepiandrosterone and its metabolites. J Endo- 30. Plosker GL, Brogden RN: Serenoa repens (Permixon). A review of crinol 150:S209–S220, 1996. its pharmacology and therapeutic efficacy in benign prostatic hy- 19. Ben-Nathan D, Padgett DA, Loria RM: Androstenediol and dehy- perplasia. Drugs Aging 9:379–395, 1996. droepiandrosterone protects mice against lethal bacterial infections 31. Dyner TS, Lang W, Geaga J: An open-label dose-escalation trial of and lipopolysaccharide toxicity. J Med Microbiol 48:425–431, oral DHEA tolerance and pharmacokinetics in patients with HIV 1999. disease. J AIDS 6:459–465, 1993. 20. Mooradian AD, Morley JE, Korenman SG: Biological actions of 32. van-Weerden WM, Bierings HG, van-Steenbrugge GJ: Adrenal androgens. Endocr Rev 8:1–28, 1987. glands of mouse and rat do not synthesize androgens Life Sci 21. Mean F, Pellaton M, Magrini G: Study on the binding of dihy- 50:857–861, 1992. drotestosterone, testosterone and oestradiol with sex hormone 33. Ben-Yehuda A, Danenberg HD, Zakay-Rones Z, Gross DJ, Fried- binding globulin. Clin Chim Acta 80:171–180, 1977. man G: The influence of sequential annual vaccination and of 22. Sultan C, Terraza A, DeVillier C, Carilla E, Briley M, Lorre C, DHEA administration on the efficacy of the immune response to DeScomps B: Inhibition of androgen metabolism and binding by a influenza vaccine in the elderly. Mech Ageing Dev 102:299–306, liposterolic extract of “Serona repens B” in human foreskin fibro- 1998. blasts. J Steroid Biochem 20:515–519, 1984. 34. Brown GA, Vukovich MD, Martini ER, Kohut ML, Franke WD, 23. Kellis JT, Vivkery LE: Inhibition of human estrogen synthetase Jackson DA, King DS: Effects of androstenedione-herbal supple- (aromatase) by flavones. Science 225:1032–1034, 1984. ment on serum sex hormone concentrations in 30–59 year old men. 24. Bradlow HL, Sepkovic DW, Telang NT, Osborne MP: Indole-3- Int J Vitam Nutr Res 71:27–31, 2001. carbinol. A novel approach to breast cancer prevention. Ann NY 35. Sirrs SM, Bebb RA: DHEA:panacea or snake oil? Can Fam Phy- Acad Sci 768:180–200, 1995. sician 45:1723–1728, 1999. 25. Moynihan JA, Callahan TA, Kelley SP, Campbell LM: Adrenal 36. Brown GA, Vukovich MD, Martini ER, Kohut ML, Franke WD, hormone modulation of type 1 and type 2 cytokine production by Jackson DA, King DS: Endocrine responses to chronic androsten- spleen cells: dexamethasone and dehydroepiandrosterone sup- dione intake in 30–56 year old men. J Clin Endocrinol Metab presses interleukin-2, interleukin-4, and interferon-gamma produc- 85:4074–4080, 2000. tion in vitro. Cell Immunol 184:58–64, 1998. 26. Meng Q, Yuan F, Goldberg ID, Rosen EM, Auborn K, Fan S: Received May 24, 2002; revision accepted December 12, 2002.

JOURNAL OF THE AMERICAN COLLEGE OF NUTRITION 371