Immune System Evaluation of Echinacea Purpurea Root Extract in Female Harlan Sprague Dawley (Hsd:Sprague Dawley SD) Rats

Abstract

Echinacea is commonly used as a dietary supplement owing to its putative anti-inflammatory, antiviral, and antibacterial activities. Unfortunately, determining risk posed to human health is challenging. A novel combination of chemical and biological characterization was employed for selection of a representative extract sample used for in vivo investigation of the immunomodulatory properties of Echinacea (E.) purpurea. In vitro cytokine profiling showed the extract could stimulate pro-inflammatory cytokine production in the absence of detectable endotoxin. An in-depth in vivo evaluation of the immune response following a 28-day oral exposure to the selected E. purpurea extract was conducted in adult Harlan Sprague Dawley (HSD) female rats at doses up to 2000 mg/kg/day. There was no evidence that E. purpurea caused overt toxicity or effects on humoral immunity. However, cell-mediated immunity was increased but only at low doses and with no evidence of a dose-response. In addition, E. purpurea treatment enhanced natural killer cell activity and the relative percentage of macrophages/monocytes in the spleen of rats from the highest treatment group. These results suggest that E. purpurea may have immunomodulatory properties that support antiviral and antibacterial activity but may also foster an immunostimulatory environment that could exacerbate some immune-mediated conditions.

1. Introduction

There are 11 species within the genus Echinacea, herbaceous flowering plants in the daisy family (Asteraceae; Compositae), of which 3, E. purpurea, E. pallida and E. angustifolia, are currently employed medicinally in the US and Europe (Barnes et al., 2010). Interest in Echinacea originates from reports that it possesses notable immunomodulatory, anti-inflammatory, antioxidant, antiviral and antibacterial properties (Catanzaro et al., 2018; Nagoor Meeran et al., 2021). Arguably, the most convincing evidence regarding the benefits of the medicinal use of Echinacea originates from observational epidemiological/clinical studies suggesting that it can be effective in reducing the duration and severity of viral infections (Grimm & Müller, 1999; Karsch-Völk et al., 2015; Percival, 2000; Rondanelli et al., 2018; Schoop et al., 2006; Woelkart et al., 2008).

E. purpurea is the most studied and utilized species on the market. The reported effects on the immune system suggest that E. purpurea modulates innate immunity, characterized by macrophage activation, leading to changes in both proinflammatory and anti-inflammatory cytokines (Aucoin et al., 2021; Fonseca et al., 2014; Fu et al., 2017; LaLone et al., 2007; Luettig et al., 1989; Roesler et al., 1991; Steinmuller et al., 1993; Stevenson et al., 2005) as well as increases in Natural Killer (NK) cell function (See et al., 1997; Zhai et al., 2007). Although changes in adaptive immunity have also been described, the data are more limited. For example, several studies demonstrated slight enhancement in humoral immunity (Bodinet C et al., 2002; Freier et al., 2003; Rehman et al., 1999; Zhai et al., 2007) as well as increased cytokine production following in vitro treatment of T cells (Fonseca et al., 2014; Morazzoni et al., 2005).

To help clarify the immune effects of this dietary supplement, an in-depth examination of the impact of oral exposure to E. purpurea on the immune response in adult Harlan Sprague Dawley (HSD) female rats was investigated. As part of these efforts, we identified an extract with characteristics similar to E. purpurea root XRM (reference material) (Ryan et al., 2019; Waidyanatha et al., 2020), while at the same time possessing immunostimulatory properties in vitro. These studies are unique as they employed a highly defined phytochemically characterized E. purpurea root extract and battery of validated rodent immune function tests (Luster et al., 1988).

2. Materials and Methods

2.1. Echinacea Purpurea Extract Procurement

To support test material selection, 24 extracts of E. purpurea samples and 6 reference materials were procured from 10 suppliers and designated as EC1-30 (Table 1). Root extract reference materials (XRM) [e.g., E. purpurea root extract (EC15) and E. angustifolia root extract (EC16)] and verified botanical reference materials (VBRM) [e.g., E. pallida root extract (EC14) were procured to help guide lot selection. Materials were stored at either room temperature or −20°C.

Table 1:

Procured Echinacea Samples

Sample ID	Sample Name	Lot #	Supplier
EC1	Echinacea Purpurea Extract BL	9809078	Acta Pharmacal (Sunnyvale, CA)
EC2	Echinacea Purpurea Extract BL	9805112	Acta Pharmacal (Sunnyvale, CA)
EC3	Echinacea Purpurea Extract BL	0201-7402	Quality Botanical Ingredients (South Plainfield, NJ)
EC4	Echinacea Purpurea Extract BL	0601-6233	Quality Botanical Ingredients (South Plainfield, NJ)
EC5	Echinacea Purpurea Extract BL	87471	Stryka Botanicals Co., Inc. (Belle Mead, NJ)
EC6	Echinacea Purpurea Extract BL	87693	Stryka Botanicals Co., Inc. (Belle Mead, NJ)
EC7	Echinacea Purpurea Extract BL	SF-20120508	Scandinavian Formulas (Sellersville, PA)
EC8	Echinacea Purpurea Herb BL	1212260361	Pharmachem Labs (Kearny, NJ)
EC9	Echinacea Purpurea Herb BL	60531	Gojira Fine Chemicals (Bedford Hights, OH)
EC10	Echinacea Purpurea Herb BL	RCX042K14	Amax NutraSource (City of Industry, CA)
EC11	Echinacea Purpurea Root BL	SF-EPE-20130701-13-15	Scandinavian Formulas (Sellersville, PA)
EC12	Echinacea Purpurea Root BL	C20120602	Maypro (Purchase, NY)
EC13	Echinacea Purpurea Root BL	D189/013/A13	Maypro (Purchase, NY)
EC14	Echinacea Pallida Root VBRM	00030683-010	Chromadex (Irvine, CA)
EC15	Echinacea Purpurea Root Extract XRM	00030217-012	Chromadex (Irvine, CA)
EC16	Echinacea Angustifolia Root XRM	00030212-020	Chromadex (Irvine, CA)
EC17	Echinacea Purpurea Flowers VBRM	00030802-125	Chromadex (Irvine, CA)
EC18	Echinacea Purpurea Leaf and Stem VBRM	00030784-058	Chromadex (Irvine, CA)
EC19	Echinacea Purpurea Root VBRM	00030785-059	Chromadex (Irvine, CA)
EC20	Echinacea Purpurea Root Extract BL	SF-20170220	Scandinavian Formulas (Sellersville, PA)
EC21	Echinacea Purpurea Root Extract BL	SF-20170409	Scandinavian Formulas (Sellersville, PA)
EC22	Echinacea Purpurea Root Extract BL	0905703	Gojira Fine Chemicals (Bedford Heights, OH)
EC23	Echinacea Purpurea Root Extract BL	0905704	Gojira Fine Chemicals (Bedford Heights, OH)
EC24	Echinacea Purpurea Root Extract BL	0629702	Gojira Fine Chemicals (Bedford Heights, OH)
EC25	Echinacea Purpurea Root Extract BL	1005703	Gojira Fine Chemicals (Bedford Heights, OH)
EC26	Echinacea Purpurea Root Extract BL	0629701	Gojira Fine Chemicals (Bedford Heights, OH)
EC27	Echinacea Purpurea Root Extract BL	170702	Acetar via PlusPharma (Vista, CA)
EC28	Echinacea Purpurea Root Extract BL	0911701	Gojira Fine Chemicals (Bedford Heights, OH)
EC29	Echinacea Purpurea Root Extract BL	1009702	Gojira Fine Chemicals (Bedford Heights, OH)
EC30	Echinacea Purpurea Root Extract BL	1009703	Gojira Fine Chemicals (Bedford Heights, OH)

EC = Echinacea; BL = Bulk Lot; VBRM = Verified Botanical Reference Material; XRM = Extract Reference Material. Note: EC13 reflects well-characterized test lot for additional studies (Waidyanatha et al 2020).

2.2. In Vitro Assessment of Immunomodulatory Properties of Echinacea Purpurea Extracts

As a first step to select a representative material for in vivo testing, an in vitro human whole blood culture system was used to examine cytokine production following treatment with E. purpurea extracts and reference materials (EC1-30). Cytokine production from human whole blood has been used previously to determine the immunomodulatory properties of novel pharmaceuticals (Wolf et al., 2012) as well as environmental chemicals (Johnson, Luster, et al., 2025). Details of the in vitro test system have been previously published (Johnson, Luster, et al., 2025). Briefly, heparinized human whole blood was obtained from StemExpress now CGT Global, Fulsom, CA and no personal identifying information or demographics were provided. Blood was used within 24 hours of draw. E. purpurea materials were prepared in 20:80 ethanol:water to the highest solubility and then diluted to 75x working solutions using sterile saline at 5 concentrations as a 2-fold dilution series. The final ethanol concentration in the culture tubes of the highest tested concentration (the original extract prepared in 20:80 ethanol:water) was 0.27%. Test samples (40 μL) were added to triplicate TruCulture^® tubes (Rules Based Medicine, Austin, TX) containing 2 mL of TruCulture^® growth media and anti-human CD3 (0.4 μg/mL) and anti-human CD28 (0.33 μg/mL) antibodies as stimulants. Blood (1 mL) from three donors was added to separate culture tubes and incubated static at 37°C for 24 ± 1 hours. Following incubation, the soluble and cellular fractions were separated, and the supernatant was frozen at -70°C until used for assessment of cytokine production. Each concentration of test sample was tested as a single TruCulture^® tube using blood from three independent donors for a biological replication of n=3.

Culture supernatants were analyzed for selected cytokines using a 96-well plate multiplex V-Plex immunoassay (Meso Scale Discovery, MSD, Rockville, MD) composed of individual spots for INFγ, IL-10, IL-12p40, IL-13, IL-1β, IL-2, IL-4, IL-6, IL-8, and TNFα. The electrochemiluminescent multiplex immunoassay was performed as validated by the manufacturer. Electrochemiluminescence signals were acquired using a Meso QuickPlex SQ 120 reader (MSD) running Methodical Mind v1.0.38. Signals were analyzed using MSD Discovery Workbench v4.0.13 and cytokine concentrations for samples were extrapolated from the standard curve. Cytokine concentrations for extract treated samples were compared to vehicle treated samples within donors to calculate a fold change in production. The area under the curve (AUC) for fold change for all concentrations was calculated and the sum of the AUC for three donors is reported as a measure of effect size for the impact of E. purpurea on in vitro cytokine production. Endotoxin contamination was measured using the Kinetic Chromogenic Limulus Amebocyte Lysate Assay with a lower limit of reporting of 5 EU/mL (Charles River Laboratories, Inc., Charleston, SC).

2.3. Chemical Characterization of Select Echinacea Purpurea Extracts

To further support selection of a representative sample of E. purpurea extract for assessment of innate, cell-mediated, and humoral immune function in vivo, 4 lots were selected for chemical characterization using methods published previously (Waidyanatha et al., 2020): E. purpurea root extract was selected as a comparator to the extract utilized in previous toxicity studies (EC13), and 3 lots (EC22, EC23, and EC24) were identified based on the cytokine profiles similar to EC13 (above). E. purpurea root extract reference material (EC15) was also included in the analysis for comparison. Briefly, this characterization included 1) authentication by high performance-thin layer chromatography (HPTLC), 2) estimation of Echinacea marker constituents utilizing high performance liquid chromatography (HPLC) coupled with charged aerosol detection (CAD), and 3) an assessment for contaminants (microbial counts, pesticides, mycotoxins, and heavy metals). Brief methods and results from these analyses are provided in the CEBS Supplemental Figures 1 and 2 and Supplemental Table 1.

EC23 lot 0905704 was received in 8 containers. Based on the cytokine release, lowest levels of microbial counts, endotoxin, heavy metals and pesticides, and comparability to the E. purpurea profiles in EC13 and the standard reference lot EC15, EC23 (lot 0905704; Gojira Fine Chemicals, LLC, Bedford Heights, OH) was selected for the in vivo studies. Lot 0905704 was received in 8 containers. The material in multiple containers was homogenized by blending, repackaged, and a lot number 02252019 was assigned. Additional characterization of lot 02252019 was conducted and results are provided in CEBS Supplemental Tables 2 and 3.

2.4. Echinacea Purpurea Dose Formulations for In Vivo Studies:

Dose formulations were prepared in 0.5% aqueous methyl cellulose (Spectrum Chemical MFG. Group, New Brunswick, NJ) by Battelle (Columbus, OH) at target concentrations of 0, 25, 50, 100, 200, and 400 mg/mL and analyzed using a validated analytical method using marker constituents and utilizing HPLC with ultra violet detection at 254 nm (r² = 1.00; relative standard deviation (RSD) ≤ 1%; relative error (RE) ≤± 3.3 with respect to all 3 marker constituents). Three marker constituents including caftaric acid (CASRN 67879-58-7), chicoric acid (CASRN 6537-80-0), and dodeca-2E,4E,8Z,10(E/Z)-tetraenoic acid isobutylamide (CASRN 866602-52-0) were selected to measure dose formulation concentrations. All formulations were analyzed prior to dose administration and were within 10% of the target and had a RSD ≤5%. The 0 mg/mL dose formulations were below the limit of quantitation for all marker compounds. Formulation homogeneity and stability when stored at refrigerated conditions up to 42 days were confirmed prior to study start using the same constituents. All dose formulations were used within 42 days.

2.5. In Vivo Experimental Animals and Study Design:

The in vivo studies were conducted in compliance with the U.S. Food and Drug Administration Good Laboratory Practices for Nonclinical Laboratory Studies (Title 21 of the Code of Federal Regulations, Part 58). The studies were performed at BRT, which is an AAALAC accredited research facility with the Office of Laboratory Animal Welfare Assurance. Animal procedures were approved by BRT’s Institutional Animal Care and Use Committee prior to conducting these studies (protocol number I99067) in compliance with the Guide for the Care and Use of Laboratory Animals. In addition, the conduct and reporting of the in vivo studies are compliant with the ARRIVE guidelines 2.0. Sentinel animals (5) were purchased and arrived with the study animals and were used for health assessment by a detailed gross necropsy performed by a laboratory animal veterinarian followed by submitting blood spots and feces to IDEXX BioAnalytics (Columbia, MO) for pathogen serology and parasite testing, respectively. Female Harlan Sprague Dawley rats (Hsd:Sprague Dawley SD) were housed in Tecniplast individually ventilated caging with 3 rats per cage with Sani-Chip^® bedding (PJ Murphy Forest Products Corp., Montville, NJ). They received NTP-2000 (Zeigler Feeds, Gardners, PA) feed and facility tap water ad libitum. Animals were provided with certified contaminant screened Rat Retreats^™ polycarbonate shelters (Bio-Serv, Flemington, NJ) as enrichment. Animals were treated daily for 28 consecutive days by oral gavage in a volume of 5 mL/kg to provide doses of 0, 125, 250, 500, 1000, or 2000 mg/kg bodyweight/day (n=12 per treatment/cohort). The top dose for this study in female rats was set at the potential highest achievable concentration in 0.5% methylcellulose and the remaining lower doses set to half increments of the top dose. The calculated human equivalent dose (HED) ranged from 20-324 mg/kg (Reagan-Shaw et al., 2008). For reference purposes, previous studies have shown the absence of acute toxicity at 15,000 mg/kg after a single oral dose or 8000 mg/kg repeated daily dosing for 4 weeks in rats (Mengs et al., 1991). The vehicle control group received 0.5 % aqueous methylcellulose alone. A positive control group (n=8 per cohort) was included in which rats were dosed with cyclophosphamide (CPS; Sigma Aldrich, St. Louis, MO [CASRN 6055-19-2, Lot MKBS0021V]) at a concentration of 15 mg/kg/day in saline by intraperitoneal injection for up to 14 days prior to study termination (see Table 2 for details). Historical data for the multiple endpoints was used to determine the number of animals per group identified above. Animals were evaluated for general health condition and only healthy animals were released to study, although all animals from the order were healthy. Animals released to study were randomized using Provantis v9.3.2.1 (Instem, Philadelphia, PA) to balance group bodyweights within ± 20% prior to the start of dose administration.

Table 2:

Endpoints examined for each cohort

Cohort	Study Endpoints
Cohort 1: Immunopath	Clinical observations Body and organ weights Hematology Pathology (enhanced immunopathology for immune tissues, traditional histology for non-immune tissues) CPS administered to positive control group for last 4 days
Cohort 2: SRBC	Clinical observations Body and organ weights Antibody response to SRBC Serum anti-SRBC IgM Anti-SRBC IgM producing B-cell enumeration by ELISpot CPS administered to positive control group for last 4 days
Cohort 3: KLH	Clinical observations Body and organ weights Antibody response to KLH – In-life 5 days following immunization (IgM) and at terminal necropsy (IgG) CPS administered to positive control group for last 14 days
Cohort 4: NK/Pheno	Clinical observations Body and organ weights T-cell proliferation Immunophenotyping of the spleen NK cell activity in spleen CPS administered to positive control group for last 4 days
Cohort 5: CTL	Clinical observations Body and organ weights CTL response to influenza infection CPS administered to positive control group for last 8 days

Dose formulations were prepared by RTI (Durham, NC) and analyzed using a validated analytical method utilizing HPLC with evaporative light scattering detection (r = 0.9974; relative standard deviation (RSD) ≤ 6.7%; relative error (RE) ≤± 10%). All CPS dose formulations were within 10% of target concentration prior to administration and following dosing. RTI confirmed the identity of CPS by infrared and nuclear magnetic resonance spectra (proton, ¹³C and ³¹P) and purity (> 99%) by HPLC with evaporative light scattering detection. As it was not possible to conduct all immune function tests using the same animals, the animals were divided into the minimum number of cohorts required by grouping compatible immune function tests resulting in 5 separate cohorts as shown in Table 2. Dosing was performed by dose group without randomization and was short in duration (~1 hour).

At scheduled necropsy, animals were euthanized by CO₂ inhalation using 100% CO₂ introduced at 3.65 LPM into a 7.3 L chamber to displace 50% of the atmosphere per minute until breathing ceased and no pedal reflex was observed. Severing the diaphragm was used for confirmation of death. For the immunopathology cohort, animals were randomized for necropsy to minimize the potential impact of the long necropsy (4-5 hours). All other cohorts were euthanized and samples collected in animal order without randomization as the necropsies were short duration (~1 hour).

2.6. General Toxicology and Histopathology (Cohort 1):

Clinical observations including evaluation of skin and fur, eyes and mucous membranes, changes in activity level, gait and posture were recorded at least twice daily. Body weights were recorded before administration and then weekly thereafter until scheduled study termination. Body weights for the positive control group were recorded only between days 14-28 because CPS was administered during the final 4-14 days of the study (Table 2); therefore, body weight gains were not reported for this group. Clinical observations and bodyweights were used to identify signs of overt toxicity throughout the study including greater than 10% bodyweight loss so that study management and the veterinarian could determine if humane euthanasia was required. At necropsy, the liver, spleen, lungs, thymus, kidneys, and adrenal glands were weighed and examined and then fixed in 10% neutral buffered formalin, along with bone marrow (femur), gastrointestinal tract with Peyer’s patches, and mesenteric and popliteal lymph nodes (LN). Tissues were sectioned at 4–6 μm and stained with hematoxylin and eosin for histopathological evaluation. The lymphoid organs were evaluated using enhanced histopathology guidelines (Elmore, 2012) while non-lymphoid organs were evaluated by traditional histopathology. All evaluations were conducted by a veterinary pathologist in accordance with the NTP Immunotoxicity Study Pathology Specifications (NTP, 2016). Necropsy technicians and veterinary pathologists were blinded to the identity of the dose group until evaluations were completed.

2.7. Hematology (Cohort 1):

At study termination, the animals were anesthetized with CO₂ and blood (~250 μl) collected from the retroorbital site and placed into tubes containing K₂EDTA. Immediately following blood collection and before recovery of consciousness, animals were returned to the CO₂ chamber for euthanasia. Complete blood cell counts with white blood cell differential and reticulocyte counts were determined the day of collection on an Advia 120 hematology analyzer using associated V.6.3.2-MS software (Siemens Medical Solutions USA, Inc., Malvern, PA).

2.8. Humoral Immunity (Cohorts 2 and 3):

Humoral mediated immunity was evaluated by measuring the T-cell-dependent antibody responses (TDAR) to sheep red blood cells (SRBC; Cohort 2) and keyhole limpet hemocyanin (KLH; Cohort 3). Following immunization with SRBC (2x10⁸ cells/mL in Alsever’s; Colorado Serum Company, Denver, CO), the number of antibody producing cells in the spleen were measured using an antibody forming cell assay (AFC), the number of IgM antibody producing B cells were measure using ELISpot, and the serum titers of circulating anti-SRBC IgM were measured using ELISA (Life Diagnostics, West Chester, PA). A separate group of rats was immunized with KLH (Stellar Biotechnologies, Inc., Los Angeles, CA), and serum anti-KLH IgM and IgG titers determined using ELISA (Life Diagnostics). We have described these methods in detail previously (Watson et al., 2021).

2.9. Natural Killer (NK) Cell Activity (Cohort 4):

NK activity was determined using a target cell killing assay as previously described (Johnson et al., 2023; Watson et al., 2021). Spleen effector cells were separated from red blood cells and adherent cells, and the single cell suspensions were adjusted to achieve the desired effector-to-target ratios. Effector cells (100 μl) were added to wells of round-bottom microtiter plates containing 100 μl of YAC-1 target cells (1x10⁵ cells/mL) labeled with ⁵¹Cr at 100 μCi per 1x10⁶ target cells for 90 minutes. Spontaneous-release and total ⁵¹Cr release controls were prepared separately by adding 100 μL of YAC-1 target cells to appropriate wells containing 100 μL of RPMI 1640 or Triton X-100 for spontaneous and total release, respectively. The plates were centrifuged at 250xg for 5 minutes, then incubated at 37°C/5%CO₂ for 4 hours. Culture supernatants were harvested and release of ⁵¹Cr was determined using a Packard Cobra II AutoGamma counter. Percent specific ⁵¹Cr release (NK lytic activity) was calculated using the formula [(E-S)/(T-S)] x 100; where E is the ⁵¹Cr release from target cells in the presence of effector cells, S is the spontaneous release of ⁵¹Cr from target cells alone, and T is the maximum release of ⁵¹Cr from target cells in the presence of Triton X-100.

2.10. Immunophenotyping (Cohort 4):

Immune cell populations in the spleen were enumerated using flow cytometry as described previously (Johnson et al., 2023; Watson et al., 2021). Immunophenotypes are presented as absolute cell numbers and percentages of CD45⁺ lymphocytes for lymphoid cells or percentages of total myeloid cells, as appropriate. Antibody cocktails for discrimination of immune cell populations were the following.

1. CD45, CD3, CD45RA, and CD161a

2. CD45, CD3, CD4, CD8a

3. CD11b/c, CD103, RP-1, and CD161a.

Details of the antibodies used for these studies have been published previously (Johnson, Walker, et al., 2025). Cell populations examined in the spleen included total lymphocytes, total T cells, CD4+ T cells, CD8+ T cells, B cells, NK cells, monocytes/macrophages, eosinophils, and neutrophils. In addition, T:B cell and CD4+:CD8+ T cell ratios were determined. Following staining, samples were stored at 2-8 °C, protected from light, until analyzed on an Accuri C6 flow cytometer using CFlow Plus v 1.0.264.21 (BD Biosciences). In all cases, a minimum of 20,000 events/sample was acquired.

Lymphocyte gating was performed on CD45+ populations. The following lymphocyte subsets were identified; T-cells (CD3+CD45RA−), B-cells (CD3−CD45RA+), NK cells (CD3−CD161a+), T-helper cells (CD3+CD4+), and T-cytotoxic cells (CD3+CD8+). Myeloid cells were gated based on being positive for CD11b with low-to-mid intensity staining for CD103. Myeloid populations were differentiated from NK cells based on lack of CD161a. Further differentiation was based on expression of RP-1 with positive cells being neutrophils and negative cells differentiated using SSC into monocytes/macrophages with low granularity and eosinophils with high granularity.

2.11. Cell Mediated Immunity (Cohort 5):

To stimulate an in vivo cell-mediated immune response, rats from Cohort 5 were infected with influenza virus [~ 2 x 10⁵ plaque forming units (pfu)/rat] via intranasal instillation (Burleson et al., 2018) eight days prior to scheduled termination. The ex vivo CTL assay was performed using lung effector cells isolated from the influenza-exposed animals eight days following infection as previously described (Johnson et al., 2023; Watson et al., 2021). Briefly, on the day of the assay UMR-106 target cells (ATCC, Manassas, VA) were infected with influenza virus (multiplicity of 10) in vitro and then labeled with Chromium-51 (⁵¹Cr at 100 μCi per 1x10⁶ target cells for 90 minutes). Lung effector cells isolated from lung tissue digests were separated from red blood cells and adherent cells and combined with labeled target cells in U-bottomed microtiter plates at the appropriate effector-to-target ratios (50:1, 25:1, or 12.5:1). Plates were briefly centrifuged to facilitate effector-to-target cell contact and incubated at 37°C/5%CO2 for 6 hours. Culture supernatants were harvested and release of ⁵¹Cr was determined using a Packard Cobra II AutoGamma counter. Percent specific ⁵¹Cr release (CTL lytic activity) was calculated using the formula [(E-S)/(T-S)] x 100; where E is the ⁵¹Cr release from target cells in the presence of effector cells, S is the spontaneous release of ⁵¹Cr from target cells alone, and T is the maximum release of ⁵¹Cr from target cells in the presence of Triton X-100.

2.12. Data collection and statistical analysis:

Data were collected into Provantis v9.3.2.1 (Instem, Philadelphia, PA) and calculation of endpoints was performed within this validated electronic data collection and management system. Results are presented as mean ± SEM (N). Both trend tests and pairwise comparisons were conducted. Trend tests evaluate whether there is evidence of an overall monotonic response while pairwise comparisons identify the specific dose levels which differ from the vehicle control. Jonckheere’s test was used to test dose-related trends (Jonckheere, 1954). If a significant trend was detected at p ≤ 0.01, a trend-sensitive test (Williams’ or Shirley’s test, depending on the historical distribution of the endpoint) was used for pairwise comparisons to preserve statistical power under this framework. Body and organ weight data, which typically exhibit a normal distribution, were analyzed using parametric multiple comparison procedures. Williams’ test was used if the trend was significant at p ≤ 0.01 (Williams, 1986), or Dunnett’s test was used if the trend was not significant at p ≤ 0.01 (Dunnett, 1955). Data for all other endpoints were analyzed using non-parametric multiple comparison procedures (Shirley or Dunn’s tests). If a significant trend was observed at p ≤ 0.01, Shirley’s test was used (Shirley, 1977); if the trend was not significant at p ≤ 0.01, Dunn’s test was used (Dunn, 1964).(Jonckheere, 1954) (Williams, 1986)(Dunnett, 1955)(Shirley, 1977)(Dunn, 1964)Body and organ weight data for the positive control were compared to the vehicle control group using the t-test. Positive control group data for all other endpoints were compared to the vehicle control group using the Wilcoxon rank sum test (Mann & Whitney, 1947). Treated and positive control data that were different from the vehicle control or showed a trend with increasing dose at p≤0.05 were considered statistically significant. Extreme values were identified using the Dixon & Massey outlier test (Dixon & Massey, 1957) prior to analysis. All flagged outliers were examined by NIEHS personnel and implausible values were eliminated from the final analysis. If outliers were removed, the exact group size for each endpoint is provided in the respective table or figure.

3. Results

Summary findings pertinent for evaluating immunomodulation are presented below. All study tables, including results that are discussed in this section and individual animal data, can be found in the Chemical Effects in Biological Systems (CEBS) database https://doi.org/10.22427/DATA-500-103-002-000-1.

3.1. Echinacea purpurea extract in vitro immune response, chemical characterization, and test material selection

Prior to selecting a representative extract to assess the immunomodulatory potential of E. purpurea in vivo, extracts and reference materials were screened for immunomodulation using an in vitro human whole blood culture system. In addition, samples were evaluated for the presence of endotoxin to help select a representative extract. Individual cytokine levels, as well as total cytokine level, expressed as area under the curve (AUC) of the fold change through the concentration range, are shown in Table 3. The in vitro analyses showed stimulation of cytokine production following treatment with 29 of 30 extracts (EC5 had no effect), although qualitative and quantitative differences were observed. While the predominant cytokines produced were proinflammatory, specifically IL-1β, IL-6, IL-8, and TNFα, the total levels produced often spanned several orders of magnitude difference between extracts. Other cytokines included IFNγ, IL-10, and IL-4 were stimulated by some extracts, although the levels produced were much lower. To address whether the proinflammatory cytokine production was due to endotoxin, a common contaminant of botanical extracts, endotoxin levels were quantified. Endotoxin levels showed marked variability between extracts (Table 3). Ten samples had undetectable levels of endotoxin (<5 EU/mL) while the remaining 20 samples had endotoxin levels ranging from 14.3 to 9522.8 EU/mL. Upon comparing cytokine responses to EC13, the well-characterized E. purpurea extract used for previous toxicity evaluations (Ryan et al., 2020; Waidyanatha et al., 2020), EC22, EC23, and EC24 were selected as candidates for the in vivo immunotoxicity assessment.

Table 3:

Cytokine production from human whole blood following in vitro treatment with E. purpurea extracts

Sample ID	Cytokines										Sum of AUC	Overall Response	Endotoxin (EU/mL)
	INFγ	IL- 10	IL- 12p70	IL- 13	IL-1β	IL-2	IL-4	IL-6	IL-8	TNFα
EC1	BLQ	3.94	BLQ	BLQ	6.80	BLQ	2.12	21.19	28.63	7.51	70.19	↑	< 5
EC2	BLQ	BLQ	BLQ	BLQ	7.67	BLQ	BLQ	54.46	29.50	4.74	96.37	↑	< 5
EC3	BLQ	BLQ	BLQ	BLQ	5.32	BLQ	BLQ	60.15	28.60	7.98	102.05	↑↑	< 5
EC4	BLQ	BLQ	BLQ	BLQ	1.38	BLQ	BLQ	16.42	24.58	5.92	48.30	↑	< 5
EC5	BLQ	BLQ	BLQ	BLQ	BLQ	BLQ	BLQ	1.03	2.55	0.73	4.32	NC	< 5
EC6	BLQ	BLQ	BLQ	BLQ	BLQ	BLQ	BLQ	11.78	12.81	5.30	29.89	↑	< 5
EC7	2.19	42.40	1.97	BLQ	330.50	BLQ	5.73	1471.00	210.10	365.00	2428.88	↑↑↑	< 500
EC8	1.50	12.41	1.37	BLQ	79.64	BLQ	4.72	1481.00	228.90	152.50	1962.03	↑↑↑	14.3
EC9	25.39	48.93	5.82	BLQ	445.70	2.84	18.31	5164.00	267.60	767.10	6745.69	↑↑↑	1407.7
EC10	1.58	2.32	BLQ	BLQ	7.10	BLQ	1.72	757.90	135.80	21.62	928.04	↑↑	< 5
EC11	1.07	4.17	BLQ	BLQ	16.38	BLQ	1.29	599.10	81.86	34.09	737.97	↑↑	117.9
EC12	2.89	16.67	2.94	BLQ	114.00	BLQ	6.47	2867.00	657.80	209.90	3877.67	↑↑↑	63.4
EC13	BLQ	3.29	0.63	BLQ	23.19	BLQ	2.70	342.50	88.90	34.35	495.56	↑↑	41.4
EC14	8.09	23.35	5.81	BLQ	798.20	BLQ	13.28	5741.00	791.70	573.40	7954.83	↑↑↑	1184.5
EC15	BLQ	BLQ	BLQ	BLQ	1.96	BLQ	BLQ	108.30	43.69	6.67	160.62	↑↑	< 5
EC16	1.88	2.61	BLQ	BLQ	32.20	BLQ	2.31	629.60	235.50	23.57	927.67	↑↑	73.5
EC17	23.54	38.09	5.30	BLQ	663.90	BLQ	18.87	5339.00	755.50	700.40	7544.60	↑↑↑	4028.8
EC18	11.47	32.62	3.56	BLQ	483.80	BLQ	14.05	3603.00	723.70	471.60	5343.80	↑↑↑	1380.2
EC19	6.89	42.72	3.77	BLQ	744.50	BLQ	9.52	3687.00	519.00	460.80	5474.20	↑↑↑	9522.8
EC20	2.96	17.49	1.44	BLQ	133.30	BLQ	4.79	2460.00	433.40	375.00	3428.38	↑↑↑	76.4
EC21	4.84	23.24	2.81	BLQ	346.70	BLQ	11.51	2087.00	329.10	290.70	3095.91	↑↑↑	74.9
EC22	3.24	24.19	3.48	BLQ	181.40	1.09	16.54	4954.00	811.70	555.90	6551.54	↑↑↑	14.7
EC23	BLQ	1.14	BLQ	BLQ	6.96	BLQ	BLQ	180.60	145.70	18.28	352.68	↑↑	< 5
EC24	7.26	53.03	8.00	BLQ	342.30	BLQ	24.25	3246.00	977.90	648.00	5306.74	↑↑↑	1653.6
EC25	BLQ	3.13	0.41	BLQ	27.78	BLQ	3.36	524.90	303.40	70.89	933.86	↑↑	6.3
EC26	BLQ	BLQ	BLQ	BLQ	12.20	BLQ	1.56	127.30	71.46	23.00	235.52	↑↑	< 0.5
EC27	2.76	7.48	1.30	BLQ	82.20	BLQ	5.14	827.80	171.10	79.25	1177.03	↑↑↑	24.9
EC28	0.78	3.46	BLQ	BLQ	75.78	4.07	2.03	1930.00	251.90	132.30	2400.32	↑↑↑	730.7
EC29	BLQ	2.83	BLQ	BLQ	67.75	4.07	4.15	1667.00	223.30	123.60	2092.70	↑↑↑	598.2
EC30	BLQ	BLQ	BLQ	BLQ	3.47	4.07	BLQ	169.50	87.74	3.69	268.47	↑↑	< 0.5

BLQ – Below limit of quantification for the assay. Cytokine levels (sum of AUC values based on the fold change from donor-specific vehicle treatment for the concentration range tested for three donors) are shown for each sample (EC1-30) to allow for comparison of previously NTP evaluated material (EC13) vs. Echinacea reference materials (e.g. EC15 -19) vs additional lots for selection of a new test article for NTP studies (EC1-3, 5-12, 20-30). Overall Cytokine Response was determined by summing individual AUC values for each cytokine/EC sample (Sum of AUC). Arrows graphically represent the “Overall Response.” Values <10 were deemed as No Change (NC) in cytokine response. Endotoxin levels are shown for each stock sample extract.

Chemical characterization of EC22, EC23, and EC24 was the final step to select a candidate extract for assessing the immunomodulatory potential of E. purpurea in vivo. Briefly, authentication by HP-TLC 2 demonstrated EC22, EC23, and EC24 have markers (high levels of chicoric and caftaric acids) of E. purpurea and low levels of markers (cynarin and echinacoside) for alternate species E. angustifolia and E. pallida (CEBS Supplemental Figure 1). The levels of constituents were found to be similar among the selected lots and aligned with the lot used in previous toxicity studies (EC13) and the E. purpurea root XRM EC15 (Ryan et al., 2020; Waidyanatha, et al., 2020) (CEBS Supplemental Figure 2). While no mycotoxins were detected, the samples contained varying levels of microbes including E. Coli and Salmonella, as well as varying levels of pesticides and heavy metals (CEBS Supplemental Table 1).

EC23 was selected for use in the in vivo studies reported here for the following reasons: 1) in vitro immunostimulatory response (i.e., cytokine release), 2) lowest contaminant levels, and 3) comparable levels of E. purpurea marker constituents to extracts used in previous studies and the E. purpurea root XRM EC15. While other lots, such as EC22 and EC24, showed significantly more immunostimulatory activity relative to EC23, they contained significant levels of endotoxin which by itself will produce inflammatory responses.

3.2. In vivo assessment of the impact of Echinacea Purpurea on the immune system:

There were no indications of any adverse clinical changes following E. purpurea or CPS treatment in this study (CEBS Summary Table I05). Treatment with E. purpurea or CPS did not affect body weight or body weight gains throughout the study (CEBS Summary Tables I04 and I04G). Organ weights were examined in the immunopathology cohort which only received E. purpurea treatment to avoid confounding or interaction with immunostimulants or infectious agents used in other cohorts. There were no statistically significant changes in organ weights, including the thymus, spleen, liver, lung or kidney, resulting from E. purpurea treatment, although a slight, but statistically significant decrease in relative adrenal weight was observed at the lowest exposure level (CEBS Summary Tables PA06). As expected, rats treated with CPS had a pronounced and statistically significant decrease in spleen and thymus weights. Histologic examination revealed that there was a minor increase in germinal centers in the spleen and mesenteric lymph nodes of rats treated with E. purpurea relative to the vehicle treated rats, although this effect was inconsistent and may not be related to treatment (CEBS Summary Table PA02, PA03, and PA10X).

Hematological evaluation, which included white blood cell differentials, demonstrated several statistically significant changes associated with E. purpurea exposure (CEBS Summary Tables M03 and M04). A small but statistically significant decrease in the absolute and relative number of reticulocytes occurred in all treatment groups administered the test chemical but there were no associated changes to the erythrocyte count or hemoglobin concentration. The reticulocyte count of the control animals was historically high (i.e., biological variability; internal data) and may explain the significant pairwise comparisons; reticulocyte counts of the treated animals were within historically normal range for rats. In addition, there was a small but statistically significant increase in the number of large unstained cells (LUC) at the 2000 mg/kg exposure level although the LUC numbers were within the normal range for rats. When observed in low numbers, LUCs usually represent intermediate to large lymphocytes, which is a normal finding in rat peripheral blood. The percentage of eosinophils was significantly decreased in animals exposed to 500-2000 mg/kg E. purpurea. Other statistically significant changes in hematological values were mild and inconsistent and likely due to biological variability and not considered related to E. purpurea treatment. CPS treatment in rats resulted in marked leukocytopenia as expected.

Spleen cell populations were enumerated using flow cytometry and are presented as absolute cell numbers and as relative percentages of CD45⁺ lymphocytes for lymphoid cells or percentages of total myeloid cells, as appropriate (CEBS Summary Tables M06). There was no effect on total lymphocytes, T cells, CD4⁺ T cells, CD8⁺ T cells, B cells, NK cells, monocytes/macrophages, eosinophils, or neutrophils in the spleen that could be attributed to E. purpurea treatment. In addition, B:T cell and CD4⁺:CD8⁺ T cell ratios were unaffected by E. purpurea treatment. Treatment with CPS produced a decrease in the number of almost all immunophenotypes, consistent with leukocytopenia.

Humoral mediated immunity was assessed using the TDAR following immunization with SRBCs by measuring the number of AFCs in the spleen, circulating anti-SRBC IgM levels and the number of individual anti-SRBC IgM producing B cells. In addition, the TDAR to KLH was assessed by quantifying titers of anti-KLH IgM and IgG antibodies in the serum following KLH immunization. There were no statistically significant E. purpurea treatment-related effects on TDAR to SRBC (Figure 1A and 1B) or KLH (Figure 1C and 1D). These and other measures of humoral immunity are provided in CEBS Summary Tables M08, M09, and M19. CPS treatment suppressed the TDAR to SRBC and KLH.

Figure 1.

TDAR specific for SRBC and KLH in female rats treated with E. purpurea. (A) AFC/10⁶ splenocytes to SRBC antigen; (B) anti-SRBC IgM producing B-cells/10⁶ splenocytes; (C) anti-KLH IgM antibody titer in the serum; (D) anti-KLH IgG antibody titer in the serum. Values represent mean ± SEM. SEM – standard error of the mean; CPS – cyclophosphamide. Significantly different pairwise comparisons of CPS groups from vehicle controls are indicated by p≤0.05 (*) or **p≤0.01 (**). n=11-12 per treatment group except for CPS where n=8.

Cell mediated immune function was assessed by determining the ability of cytotoxic T lymphocytes (CTL) isolated from the lungs of influenza infected rats to destroy influenza infected target cells. E. purpurea treatment increased the CTL response in the 250 mg/kg treatment group at the 25:1 E:T cell ratios, and in the 2000 mg/kg treatment group at the 50:1 E:T ratio (Figure 2; CEBS Summary Table M12). Overall, there appeared to be a biphasic trend with increased CTL responses at low doses that returned to vehicle levels at intermediate doses and then increased at the highest dose. CPS treatment significantly reduced the CTL response. Treatment of rats with E. purpurea did not affect proliferation of T-cells (CEBS Summary Table M11), another measure related to cell-mediated immunity.

Figure 2.

CTL activity in female rats treated with E. purpurea. Values represent mean ± SEM. SEM – standard error of the mean; E:T - effector:target cell ratio; CPS - cyclophosphamide. Significantly different pairwise comparisons of treated doses or CPS groups from vehicle controls are indicated by p≤0.05 (*) or p≤0.01 (**). n= 11-12 per treatment group except for CPS where n=8.

The impact of E. purpurea on innate immune function was determined using a NK cell killing assay. NK cell activity was significantly enhanced in a positive dose-response trend with the highest dose of E. purpurea showing significantly increased activity relative to the vehicle control group at all E:T cell ratios (Figure 3; CEB Summary Table M15). NK cell activity was also significantly increased in the CPS treatment group. This is not unexpected as NK cells are non-proliferating cells, and the toxicity of CPS requires DNA intercalation into proliferating cells.

Figure 3.

NK cell activity in female rats treated with E. purpurea. Values represent mean ± SEM. SEM – standard error of the mean; E:T - effector:target cell ratio; CPS - cyclophosphamide. Statistically significant trends with increasing dose of E. purpurea are indicated by upward arrows over the control groups (↑↑, p≤0.01). Statistically significant pairwise comparisons of treated doses or CPS groups from vehicle controls are indicated by asterisks over the appropriate treated dose or CPS group (**, p≤0.01). n= 11-12 per treatment group except for CPS where n=8.

4. Discussion

The present studies demonstrate that E. purpurea has selected effects on the immune system in the absence of any apparent toxicity. In addition to stimulating the production of specific cytokines in human blood cell cultures, E. purpurea was shown to enhance cell mediated immune responses (i.e., CTL) as well as NK cell function but not humoral immunity in vivo in HSD rats. Female rats were used for these studies as it has been demonstrated that they tend to be more sensitive to immunotoxicity and show greater dynamic range in immune responses than males (Klein and Flanagan, 2016). Consistent with this concept, South and Exon (2001) reported that female rats fed echinacea extracts showed reduced T-dependent antibody production whereas male rats did not, supporting that females may be more sensitive than males to the immunomodulatory effects of echinacea (South & Exon, 2001). By using females only, our study design greatly reduced the number of animals that would have been required to assess the immune effects of echinacea in both sexes.

To compare the rat toxicology doses with reported human exposures, the HED calculated from the highest rat dose (2000 mg/kg/day) corresponds to approximately 324 mg/kg/day in humans (i.e., ~25 g/day for a 70-kg adult). In the Dietary Supplement Label Database label dataset, adult recommendations vary widely, ranging from approximately 125 mg/day to 4,050 mg/day for standard capsule/tablet products, with most products clustering around 400–2,400 mg/day (https://dsld.od.nih.gov/). While the doses used in these studies exceed commonly recommended human doses of echinacea, they are not far removed from the upper range of doses reported in clinical trials. Published clinical trials demonstrate substantial variability in echinacea dosing, with reported regimens ranging from single doses of 5–100 mg, to repeated doses of 500 mg/day to 1.5 g/day, to as high as 2.4–8 g/day depending on formulation, indication, and duration of use. For example, common cold prevention studies have used 2,400 mg/day for several months and up to 4,000 mg/day during acute illness, while randomized trials in respiratory tract infections have administered as much as 16.8 g/day during the first several days of treatment (Aucoin et al., 2021b; Jawad et al., 2012; Sumer et al., 2023). These studies reinforce that there is no single standardized or regulated echinacea dose in products available to the public or in clinical use. Importantly, the dose range tested in our study is within or near the broad range already explored in humans, supporting the translational relevance of our dosing strategy while also strengthening the toxicological relevance of the results.

Immunophenotyping studies indicated that the observed effects were not likely due to increases in T cell numbers but rather functional changes. Immunophenotyping studies, however, did reveal a dose-responsive increase in the relative percentage of macrophages/monocytes that was significantly different from the vehicle group in rats from the high dose treatment group. The increase in CTL activity was observed at lower dose levels, however, this effect decreased in the intermediate doses and increased again at the highest dose. Similar dose relationships have been observed for other immunomodulators including lipopolysaccharide which shows differential anti-inflammatory and proinflammatory effects depending on dose level (Charoensaensuk et al., 2024; Ding et al., 2020; Hirohashi & Morrison, 1996; Mizobuchi et al., 2020). The use of multiple E:T cell ratios is based upon the belief that antigen concentration plays a role in determining which predominant T-cell types proliferate in a particular immunological situation (Gajewskj et al., 1989). Cytotoxic T cells are antigen-specific effector cells important in host defense as they can destroy foreign cells, cancer cells, and virus infected cells. It is well established that proinflammatory cytokines, such as IL-12 and IFNγ can promote the expansion, survival, and development of CTLs (Cox et al., 2011; Hervas-Stubbs et al., 2010; Kim & Harty, 2014). Activation of CTL responses at lower doses of E. purpurea were likely due to modulation of cytokines known to drive differentiation, maturation, and activation of CTLs (Sun et al., 2023), which then showed a hormetic-like decrease until the highest dose of 2000 mg/kg. Similar effects on cytokine production with higher induction at low and high concentrations of echinacea was observed following in vitro treatment of human mononuclear leukocytes (Kapai et al., 2010). In the present study, low dose stimulation of CTL may be related to direct proinflammatory cytokine induction by multiple immune cells following E. purpurea treatment, whereas the elevated CTL response in the high dose may be related to the activation of NK cells which are known to produce cytokines and facilitate activation of CTL, especially during infections (Cook et al., 2014).

NK cell activity was increased in rats treated with the highest dose of E. purpurea tested, 2000 mg/kg. Enhancement of NK cell activity has been observed previously in young adult (Park et al., 2021) and aged (Brousseau & Miller, 2005) mice. Additionally, other phytochemicals, including genistein, curcumin, ginseng, and resveratrol, have been reported to increase NK cell activity in laboratory rodent models (rats and mice) as well as ex vivo human NK cells (Grudzien & Rapak, 2018). In the current study, NK cell numbers were not altered in the spleen indicating that the change in NK cell activity caused by E. purpurea was due to NK cell activation rather than increased numbers of NK cells. NK cells represent a major component of the innate lymphoid cell (ILC) population, composing 8-15% of circulating cells in human blood and 2-6% in rodent blood (Caligiuri, 2008). They exhibit cytotoxic effects due to direct or indirect target cell recognition, and their clinical significance has been established in individuals with primary NK cell deficiency who are highly susceptible to influenza and hepatitis C virus infections as well as latent viral infections such as herpes viral infection (Jost & Altfeld, 2013; Mace & Orange, 2019). At the same time, NK cells can contribute to poor outcomes for certain infections potentially through exerting negative regulatory pressures on viral T-cell responses, as has been shown in patients with chronic hepatitis B (Fisicaro et al., 2019). Besides their common feature of being highly efficient producers of cytokines such as TNFα and IFNγ, NK cell function is tightly regulated by cytokines (Wu et al., 2017).

A strength of the present study was that we challenged human peripheral blood cells with multiple extracts of E. purpurea and examined cytokine responses in vitro. Although this approach was designed for selection of a test article for in vivo testing of immunotoxicity, the data can also bridge the immunomodulatory findings across species and provide mechanistic insights. In addition, leveraging data from non-animal in vitro stimulation of human immune cells contributes to a reduction in animals use as ex vivo cytokine production for the present study would have required an additional cohort of animals. In vitro treatment of human whole blood with the E. purpurea extract used for the present in vivo assessments resulted in increased production of IL-1β, TNFα, IL-6, and IL-8. These proinflammatory cytokines have been shown to modulate responses of CTL and NK cells. NK cells are highly responsive to cytokines in their environment, particularly to IL-2 and IL-12 and elevated levels of TNFα have been shown to augment activation-induced NK cell production of IFNγ (Almishri et al., 2016). Following activation, NK cells have been shown to produce high levels of IL-8 (Poznanski et al., 2017), a cytokine that was also elevated following in vitro treatment in the present study. In this respect, activated NK cells can act as a bridge between innate and adaptive immunity, specifically through production of cytokines like IFNγ known to drive differentiation and activate CTL cells (Ge et al., 2012) and production of IL-8 which can serve as a potent chemoattractant for CTL cells (Taub et al., 1996). Direct effects of cytokines on CTL cells including CD8 T-cell polarization towards CTL phenotype by IL-1β (Van Den Eeckhout et al., 2020), TNFα-mediated stimulation of CTL activity (Ye et al., 2018), and IL-6-driven T-cell expansion (Li et al., 2018) also occur. In this context, the potential for E. purpurea to cause subtle alterations in cytokine signaling may facilitate activation and crosstalk between innate and adaptive immune responses which can have important benefits in disease states such as viral infection (Cook et al., 2014). Our in vitro and in vivo findings are supported by similar elevations in cytokines and NK activity in Korean adults taking an echinacea extract for 8 weeks as part of a clinical trial (Lee et al., 2024).

The preceding discussion highlights how our study results compare with the variable immunological effects of echinacea reported in experimental studies and clinical trials (Bałan et al., 2016; Bodinet C et al., 2002; Burger et al., 1997; Catanzaro et al., 2018; Currier & Miller, 2001; Freier et al., 2003; Melchart et al., 1994; Rehman et al., 1999; See et al., 1997; Zhai et al., 2007). A systematic review of more than 200 publications similarly found inconsistent effects on pro-inflammatory cytokines and highlighted major differences in study design, extract type, dose, and regimen as likely sources of this variability (Aucoin et al., 2021). An additional major confounder is the complexity and variability of the echinacea products studied, including differences in species, plant parts used, extraction methods, and limited information on endotoxin contamination (Todd et al., 2015). In our in vitro and in vivo assessments, we confirmed if endotoxin or other common contaminants were present. The lack of activity in our baseline measurements for cytokines suggest that contamination from endotoxin or other immunoactive substances was minimal. However, we cannot completely rule out the possibility that trace levels of contaminants contributed to the observed effects.

Taken together, the results of the present studies are consistent with the hypothesis that E. purpurea can directly induce immunomodulatory cytokine responses from immune cells as evidenced from in vitro treatment of human whole blood, with the NK cell and macrophage/monocyte populations likely being the most affected. These changes in cytokine secretion correlate with the stimulatory effects observed on CTL and NK cell functions in our in vivo studies in HSD rats. The representative lot of E. purpurea contained the expected phenolic constituent profile of chicoric acid, caftaric acid and chlorogenic acid, as well the alkamides dodecatetraenoic acid and isobutylamides (Supplemental File Figure 2); many of these constituents and their metabolites are thought to support immunomodulatory, anti-inflammatory, and antioxidant activities (Ahmadi, 2024; Ahmadi et al., 2024; Barrett, 2003). Although this investigation did not specifically evaluate the impact of individual constituents on immunomodulation, we hypothesize they are contributing to the response observed in this in vivo assessment. It is important to note, however, that while E. purpurea contributes to the efficacy of natural immunity it may equally foster an inflammatory environment that could exacerbate immune mediated conditions such as autoimmune disease. In any case, the selection of an appropriate and high-quality product for testing is critical to generate robust safety data that can be compared widely across studies. We screened multiple materials, along with available reference materials, to select an authentic (i.e., confirmed to be E. purpurea with comprehensive chemical characterization and DNA barcoding) lot for this study. However, during our evaluation we observed substantial variability among commercially available materials, and some products marketed as E. purpurea were not representative of authentic E. purpurea. Thus, the findings of this study may not be representative of all products currently marketed and sold as E. purpurea.

Data Availability Statement

All study tables, including results that are discussed in this section and individual animal data, can be found in the Chemical Effects in Biological Systems (CEBS) database https://doi.org/10.22427/DATA-500-103-002-000-1.