Research on botanicals used by indigenous populations has generally been confined to in vitro screenings of individual plants or their constituents for their antibacterial, antiviral, or antiinflammatory activities (24–27). The fact that a botanical was traditionally used for wound healing, fever, infection, edema, or rheumatic disease is taken as an indicator that the plant should be tested for its antiinflammatory properties (26). Although several in vitro assays can be used to test for antiinflammatory activities (28), most screening procedures include inhibition of cyclooxygenase and 5-lipoxygenase. These 2 enzymes are central to the pathways producing thromboxanes, prostaglandins, and leukotrienes. A list of several whole-plant extracts and isolated chemical components that have an inhibitory effect on one or both of these enzymes is provided in Table 3
. Because different assay systems were used in the various studies, a direct comparison of the results is not appropriate. Of the plants listed in Table 3, purple coneflower (Echinacea species) and stinging nettle (U. dioica), are discussed below. For the remaining plants, what little additional research exists on them is presented in the available detail.
Sanguinarine (13-methyl[1,3]benzodioxolo[5,6-
c]-1,3-dioxolo[4,5-
i]phenanthridinium) is found in the root of
Sanguinaria canadensis (bloodroot), a plant from the family of Papaveraceae that was used extensively by numerous Native American societies for blood tonification and purification, pain relief, wound healing, fevers, and numerous other purposes (
9). The use of a medicinal botanical as a tonic usually indicates that the botanical has the ability to enhance certain immune responses. It is uncertain what is meant by "blood purification" in terms of modern medicine. In vitro, sanguinarine suppressed human peripheral blood neutrophil function, including chemotaxis, adhesion, oxidative burst, degranulation, and phagocytosis, and was nontoxic at all concentrations tested (
35). Sanguinarine also strongly inhibited the activation of nuclear transcription factor
B (NF-
B), which is involved in the induction of numerous proinflammatory mediators (
36). Sanguinarine suppressed NF-
B activation by preventing phosphorylation and degradation of inhibitory
B-
, which prevents entry of NF-
B into the nucleus.
Various preparations of Hamamelis virginiana (witch hazel; Hamamelidaceae family) were taken by Native Americans for pain relief, colds, and fevers (9). A crude alcohol and water extract of H. virginiana, as well as fractions in which the hamamelitannin content was reduced by ultrafiltration, were assessed for their antiinflammatory activities in several in vitro and in vivo experimental systems (37). In vitro, elastase, a proteolytic enzyme participating in the inflammatory response, was inhibited most potently by the fraction containing the highest concentration of hamamelitannin, and this fraction exhibited the strongest antioxidant activity. Contrasting with this, and also with the finding that hamamelitannin inhibited 5-lipoxygenase activity at very low concentrations (31) (Table 3
), partial removal of hamamelitannin by ultrafiltration resulted in a stronger inhibition of croton oil–induced ear edema (37). Oral pretreatment with an ethanolic extract of the leaves of H. virginiana, before a carrageenan injection, did not prevent paw edema in rats but reduced the arthritic paw swelling induced with Freund's adjuvant, although to a lesser extent than did Polygonum bistorta, another Native American medicinal botanical (38). In the only clinical trials of the anti-inflammatory effects of H. virginiana, preparations of this botanical were applied topically rather than administered orally (39). In 2 such trials, H. virginiana was found to have a mild suppressive effect on ultraviolet light–induced erythema and itching (39, 40), whereas a similar preparation was no more effective than was the drug-free vehicle in relieving atopic eczema (41). Whether oral ingestion of H. virginiana has greater effectiveness than does topical application remains to be addressed in animal models of inflammation and possibly in humans.
S. nigra (black elder; Caprifoliaceae family) is one of many Sambucus species used by Native Americans, mostly for rheumatism and fever (9). A methanol, and particularly a butanol and a chloroform, extract of S. nigra leaves, but not flowers, significantly inhibited lipopolysaccharide (LPS)-induced synthesis of tumor necrosis factor (TNF-) by human peripheral blood mononuclear cells but had minimal effect on interleukin (IL)-1 and IL-1ß production (42). Oral administration of an aqueous extract of the aerial parts of S. nigra before a carrageenan injection significantly inhibited hind-paw edema in mice (27). A standardized extract of the berries of the black elder, which also contains raspberry extract and citric acid, inhibited viral replication of several strains of influenza viruses in vitro and was subsequently tested in vivo (43). In this double-blind, placebo-controlled study completed by 27 patients, the same extract with added glucose and honey, taken orally for 3 d after the onset of influenza virus B infection, was associated with significantly better relief of symptoms and faster recovery in the treated than in the placebo group (43). Viral antibody titers tended to be higher in the extract-treated group than in the placebo group, suggesting that this botanical extract, possibly in addition to exerting direct antiviral activity, stimulated host immune responses. To our knowledge, this was the only double-blind, placebo-controlled clinical trial to date that used elderberry extract. Although the study was limited by its small sample size, its promising results seem to warrant further in vitro and in vivo studies of elderberry preparations.
Echinacea species (purple coneflower; Compositae family)
E. angustifolia, the narrow-leafed purple coneflower, has long been used by Native Americans for pain relief and wound treatment, as an antidote against various poisons, and for symptoms associated with the common cold (Table 1) (3, 9). It was introduced by Native Americans to European settlers, who subsequently took what they thought was E. angustifolia back to Europe. It turned out, however, that the species introduced in Europe was E. purpurea, another native American plant used for medicinal purposes by the Choctaw and Delaware (9, 19). E. purpurea has since become one of the most popular medicinal botanicals in Europe and the United States (19). For medicinal purposes, besides E. purpurea and E. angustifolia, a third species, E. pallida, is commonly used.
What the American consumer calls Echinacea can be any one of the 3 above-mentioned species or a combination of 2 or even of all 3 of them, which should of course be indicated on the label. Furthermore, many Echinacea preparations, including one of the best-known European brands that has been used in numerous studies (Echinacin; Biomed, Düsendorf, Switzerland), are extracts of both root and above-ground parts, whereas in other instances the root alone or the above-ground parts alone are used. There are substantial differences in the chemical compositions and the biological activities not only between different Echinacea species, but also between their roots and aerial parts (44–46). It is noteworthy that E. purpurea does not contain echinacoside, the substance used frequently for standardizing E. angustifolia and E. pallida extracts (47). Further differences arise from the extraction procedure. For example, the polysaccharides to which many of the stimulatory effects of Echinacea species on the nonspecific immune system have been attributed are likely to be present in aqueous, but not in alcoholic, extracts (47). In addition, in the United States, Echinacea is often sold in combination with goldenseal (Hydrastis canadensis); combinations with other medicinal botanicals are common here and in Europe.
Echinacea is one of 12 commonly used herbs that physicians need to be aware of and knowledgeable about (48). At least 3 different species of Echinacea are sold under that name, yet the literature is often reviewed without regard to what particular species was used. Moreover, differences arising from different extraction procedures, solvents, and plant parts used are ignored, and little distinction is made between data obtained with purified polysaccharides and those obtained with crude extracts.
In vitro, the phagocytosis of yeast particles by human granulocytes was stimulated by commercial extracts of both E. angustifolia and E. purpurea (not further characterized; 49). In the same assay system, ethanolic root extracts of E. purpurea stimulated phagocytosis to a greater extent than did E. angustifolia or E. pallida (44). In contrast, a widely available commercial extract of both roots and aerial parts of E. purpurea significantly decreased the chemiluminescence (phagocytosis) of human granulocytes at a high concentration (1:1 dilution), and the increase seen with a 1:100 dilution was not significant (50). These results might be attributable to experimental or procedural problems, as suggested by the fact that incubation of granulocytes with phorbol myristate acetate, a known inducer of phagocytosis, did not result in an increase in phagocytotic activity. In addition, cell viability was not assessed and might have been reduced after incubation with the high concentration of this Echinacea extract. However, others also reported that a high concentration of an ethanolic extract of E. purpurea suppressed, rather than enhanced, phagocytosis (51). Human macrophages incubated in vitro with a lyophilized and reconstituted fresh-pressed juice or a reconstituted dried juice of the above-ground parts of E. purpurea, harvested at the peak of flowering, produced the cytokines TNF-, IL-1, and IL-10 at concentrations comparable with, or higher than, those seen with LPS stimulation; IL-6 production was higher than in controls, but lower than that obtained with LPS (52).
A purified polysaccharide from E. purpurea augmented the phagocytosis of yeast particles or opsonized zymosan by human granulocytes by 23% and 34%, respectively (53). Incubation of human macrophages with a purified polysaccharide from E. purpurea cell culture induced the production of TNF-, IL-1, and IL-6. In addition, this polysaccharide increased the motility of human polymorphonuclear cells (PMNs) and their cytotoxic activity against staphylococci and stimulated the proliferation of human lymphocytes (54).
Both natural killer activity and antibody-dependent cell cytotoxicity were higher after treatment with an E. purpurea/RPMI homogenate than in untreated controls in peripheral blood mononuclear cells isolated from healthy subjects or patients with chronic fatigue syndrome or AIDS (55). It was not specified whether the "fresh herbs" used for the homogenate included roots or flowers.
In vivo, ethanolic root extracts of E. purpurea, E. pallida, and E. angustifolia all significantly increased the phagocytic activity of liver and spleen macrophages in mice treated 3 times daily for 2 days by gavage (47). E. purpurea root extract was more effective than were root extracts from either of the other species. However, an extract from the aerial parts of E. purpurea stimulated phagocytosis to a much lesser extent than did E. pallida or E. angustifolia (44).
Intravenous treatment of mice with polysaccharides isolated from E. purpurea cell culture significantly increased the survival rate of healthy and immunosuppressed mice injected with lethal doses of Candida albicans or Listeria monocytogenes (56, 57). Because protection against C. albicans and L. monocytogenes is thought to be mediated mostly by PMNs and macrophages, respectively, these findings provide further indications that polysaccharides are at least partially responsible for the stimulation of the nonadaptive immune responses observed with various Echinacea extracts.
In humans, an alcohol extract of E. purpurea roots administered intravenously or orally resulted in a significant increase in PMN phagocytic activity (51, 58). However, this was not a consistent finding of clinical studies with various Echinacea preparations (59). Oral ingestion of an extract containing E. angustifolia, Eupatorium perfoliatum, and Thuja occidentalis by 23 cancer patients for 4 wk reportedly had no effect on the concentrations of the 6 cytokines measured in whole-blood cell cultures (60). Methodologic problems might largely account for this lack of effect. No attempt was made to equalize the number of cells used per assay per patient, and the cytokines were measured only in phytohemagglutinin- or pokeweed mitogen–stimulated, but not in unstimulated, cells and only after 48 and 96 h of culture. However, in contrast with the results obtained in vitro, intravenous injection of a polysaccharide from E. purpurea also did not significantly increase the concentrations of TNF- or IL-1ß or increase IFN activity in human serum or plasma, although monocytes and PMNs were induced to migrate into the peripheral blood (54).
Among the specific uses of Echinacea species reported, the Kiowa chewed ground roots of E. angustifolia for coughs and sore throats, the Cheyenne chewed roots of E. pallida for colds or took infusions of leaves and roots for sore mouth and throat, and the Choctaw chewed roots of E. purpurea or made a tincture of it as a cough remedy (9). It is this particular indication of Echinacea preparations as a cold remedy that was tested in numerous clinical trials. In a placebo-controlled, double-blind study with 180 patients with upper respiratory infections, patients given the higher (900 mg/d) dosage of an ethanolic extract from the root of E. purpurea experienced significantly fewer and milder symptoms of shorter duration than did patients treated with placebo or the lower dosage (450 mg/d) (61). Another placebo-controlled, double-blind clinical study involved 303 volunteers taking a liquid extract containing mainly E. angustifolia (as well as smaller amounts of Eupatorium perfoliatum and Baptisia tinctoria) and 306 volunteers receiving placebo (62). Treatment with this combination of botanicals was effective in diminishing the frequency of upper respiratory infections when administered prophylactically. A recent placebo-controlled, double-blind study with an ethanolic extract of E. angustifolia or E. purpurea roots showed only a trend toward a reduced risk of upper respiratory infections (63), although it is questionable whether time until onset was an appropriate outcome measure (63). In addition to the studies mentioned above, several clinical trials were performed with various Echinacea preparations, but many were not rigorously controlled or were confounded by the fact that preparations containing botanicals besides Echinacea species were used (59).
E. purpurea extracts or isolated polysaccharides were neither toxic nor mutagenic when tested in vitro and in vivo, in mice as well as in humans (64, 65). In humans, paleness, slight and transient tachychardia, and equally transient influenza-like symptoms were the most serious adverse effects reported after intravenous injection of an E. purpurea extract (51), whereas oral ingestion of an E. purpurea extract produced only slightly more adverse effects than did placebo. In at least one study, E. angustifolia tended to be associated with more adverse effects than were either E. purpurea or placebo, and the adverse effect profile was the same as with E. purpurea (63), including headache, nausea, and fatigue (62).
However, plant extracts can cause allergic reactions. Recently, a case of anaphylaxis associated with the consumption of a supplement containing a whole-plant extract of E. angustifolia and root extract of E. purpurea was reported (66). However, from the tests performed, a clear causal relation between the plant extracts and subsequent anaphylaxis could not be established. Rather, as has been suggested (67), it is possible that another ingredient of the Echinacea-containing preparation or of one of the numerous other supplements consumed by the patient were responsible. Included with the case report were the findings that a considerable percentage (19%) of patients with asthma and allergic rhinitis exhibited cross-reactivity to Echinacea extract (66). However, note that reactivity was not defined. An inappropriately chosen threshold of reactivity might account for the fact that, instead of the thousands of cases of Echinacea-induced anaphylaxis one would have expected to see over the past decade (67) on the basis of Mullins' findings (66), this was one of the first reports of such an occurrence in the literature. Thus, although the possibility of allergic reactions to Echinacea cannot be ruled out, the information summarized in the preceding paragraph indicates that, overall, Echinacea is a safe and well-tolerated medicinal botanical.
Native Americans used Echinacea species extensively for the treatment of colds and for alleviation of the symptoms associated with them, such as sore throat, cough, and fever. Evidence has been accumulating from in vitro and in vivo studies that Echinacea species indeed contain bioactive substances capable of stimulating nonadaptive immunity and of helping to prevent and allow patients to more quickly overcome upper respiratory infections. Note, however, that at this point it is not clear to what extent the results of in vitro studies, particularly those conducted with isolated polysaccharides, translate to in vivo situations. Furthermore, the lack of dose-response data from most clinical studies constitutes a serious shortcoming of the existing database on Echinacea species. In addition, the use of many different Echinacea preparations of undefined or poorly defined chemical compositions in the various studies makes comparisons of the results impossible and severely limits reproducibility. A team of researchers who have worked extensively with various Echinacea species, therefore, has long urged the use of HPLC fingerprints to document the exact chemical composition of an extract to enable correlation of specific chemical constituents with observed immunomodulatory activities and to obtain reproducible results (44, 47, 68).
Urtica dioica (stinging nettle; Urticaceae family)
Although U. dioica is a native of Europe and Asia (2), its extensive use by Native American societies throughout North America (9, 12; Table 2) suggests that the plant and its medicinal use spread rapidly after its introduction. Various parts of the stinging nettle (U. dioica) were administered externally and internally by numerous Native American societies for a variety of purposes, including as a general tonic (ie, what we would now consider an immunostimulant) and as a treatment for fevers and rheumatism (Table 2).
In vitro, by using whole blood from healthy volunteers, a commercial ethanolic extract of U. dioica, designated as IDS23, inhibited LPS-induced release of TNF- and IL-1ß but had no effect on the secretion of these cytokines when added to the culture medium in the absence of LPS (69). Inhibition was minimal at concentrations concentration could be achieved in vivo. When known constituents of U. dioica extract, such as caffeic acid, caffeic malic acid, chlorogenic acid, rutin, and quercetin, were tested individually, none inhibited the synthesis of TNF- or IL-1ß. In the same study, U. dioica extract stimulated the production of IL-6 to the same extent as did LPS, and this effect was not further augmented by the combination of these 2 agents.
Preincubation with IDS23 or IDS23/1, the water-soluble fraction of IDS23, inhibited activation of NF-B induced by incubating HeLa or L929 cells with TNF-, Jurkat cells with phorbol myristate acetate, or MonoMac6 cells with LPS (70). The binding of NF-B to DNA was inhibited considerably more strongly after preincubation with IDS23/1 than with IDS23. IDS23/1 did not interfere directly with the DNA binding process itself but rather prevented the degradation of the inhibitory molecule IB-, which binds to NF-B in the cytosol and prevents its entry into the nucleus. Another transcription factor, AP-1, was inhibited by IDS23/1, but only partially. Rheumatoid arthritis is an inflammatory disease of joints; cytokines, particularly TNF- and IL-1, and other mediators of inflammation are strongly implicated in its pathogenesis (71, 72). Both NF-B and AP-1, in turn, play a central role in the induction of these proinflammatory cytokines and other mediators. The activity of NF-B and AP-1 is upregulated in rheumatic synovia, in which transcription factors are thought to participate in various inflammatory and joint destructive processes (73–75). Thus, the results of these in vitro studies provide some indication that the Quileute and Tainarna tribes might have derived some benefit when they used infusions of U. dioica as an antirheumatic (9).
When healthy volunteers ingested capsules containing 335 mg IDS23/1 twice daily for 3 wk, the concentrations of TNF-, IL-1ß, IL-4, IL-6, and IL-10 produced in cultures of whole blood all remained below the detection limit (
7 pg/L) (76). However, LPS-stimulated concentrations of TNF- and IL-1ß were slightly but significantly reduced after 7 d and further reduced after 21 d of IDS23/1 treatment compared with baseline. Addition of IDS23/1 to these whole-blood cultures further inhibited the LPS-induced synthesis of these 2 cytokines, leading the authors to propose that the optimal concentration of biologically active constituents had not been reached. It remains to be established which component or components of U. dioica exert the cytokine inhibitory effect observed in vitro and whether such components remain intact during digestion and reach the bloodstream in a fully biologically active state. The suppression of cytokine synthesis in response to inflammatory stimuli constitutes one mechanism by which U. dioica might ameliorate rheumatic disorders. If confirmed in vivo, the ability of U. dioica to inhibit other inflammatory processes, such as cyclooxygenase activity (Table 3) and NF-B and possibly AP-1 activation, would also be expected to have beneficial effects in rheumatic diseases. Thus, science is beginning to validate the Native American concept that U. dioica is a valuable antirheumatic agent. However, the data available to date are rather limited and, though promising, await confirmation in experimental models and in clinical trials, if appropriate.
Urtica dioica agglutinin
U. dioica agglutinin (UDA) is a lectin isolated from the rhizome of the stinging nettle. In vitro, it is a murine T cell–specific mitogen with characteristics that are kinetically and functionally distinct from those of concanavalin A (Con A), another mitogen specific for T cells (77, 78). The lectin moiety of UDA recognizes specific carbohydrate structures present on major histocompatibility complex class II molecules, and this binding event results in clonal expansion of T cells expressing Vß8.3 within 3 d (79), particularly of the Jb1.1, Jb1.6, and Jb2.7 subpopulations (80).
Such selective expansion was also observed among peripheral T cells in vivo after intravenous injection of UDA into Balb/c mice and was followed by anergy and deletion via apoptosis of mature Vß8.3+ T lymphocytes (81). The percentage of Vß8.3+ T cells was not fully restored to pre–UDA-treatment concentrations, even after 2 mo. In contrast with peripheral T cells, Vß8.3+ thymocytes did not undergo clonal expansion after intravenous injection of UDA (82). Furthermore, the total number of thymocytes decreased only slightly (25%) between 2 and 8 d after the administration of UDA, and the proliferative response of thymocytes from UDA-treated animals to in vitro restimulation with UDA was 
4-fold lower than at baseline but was restored to almost normal values after 8 d.
A role for T cells bearing the Vß8 receptor has been suggested in the pathogenesis of autoimmune diseases in humans (83) and in animal models (84–86). MRL-lpr/lpr is a mutant mouse strain lacking the Fas protein, which plays a central role in apoptosis. Such mice spontaneously develop many of the symptoms found in the human autoimmune disease, systemic lupus erythematosus (SLE). A CD4+ T cell clone expressing Vß8.3-Db1.1 and Jb1.1 elements was expanded in MRL-lpr/lpr, but not MRL+/+, mice, suggesting that it contributed to the pathogenesis of SLE-like disease in the mutant strain (87). Treatment of MRL-lpr/lpr mice with intravenous injections of 100 µg UDA every other week from 6 wk to 6 mo of age resulted in the specific and sustained deletion of Vß8.3+ T cells accompanied by an absence of clinical signs of SLE, a reduction of glomerulonephritis, and a significant delay in the enlargement of lymph nodes compared with phosphate buffered saline–treated controls in both male and female animals (88). Interestingly, autoantibody production remained low in UDA-treated female MRL-lpr/lpr mice but was not affected in male mice. Despite the promising nature of the data reviewed above, more studies are needed before clinical trials can even be considered.
Answers to all your Biology Questions
