Abstract
Polysaccharides and glycoproteins are in part responsible for the immunostimulatory effect of Echinacea. spp. (Asteraceae). A major problem, however, is finding a suitable analytical method to quantify them in the botanical raw material or in Echinacea. extracts. An HPLC method has been developed using an evaporative light scattering detector (ELSD) detector. The fraction containing polysaccharides and glycoproteins (PS/GP) was obtained by precipitation of Echinacea. spp. extracts with EtOH. The PS/GP fraction was passed through a 5-kDa molecular weight filter to remove smaller molecules and then analyzed by HPLC-ELSD. The chromatogram shows two peaks, one corresponding to PS/GP with a molecular weight (MW) above 48.6 kDa and one to PS/GP with a MW below 48.6 kDa. The PS/GP fraction is expressed as an arabinogalactan equivalent. This method has been validated and is suitable for quality control of raw material and extracts. The results showed that the composition of polysaccharides differs among species of Echinacea., among organs of the same species, among provenance, and among extracts based on the solvent used.
Introduction
Most of the dietary supplements on the market in the United States are standardized to one or several marker compounds. These markers are often not the only molecules related to the activity of the plant and sometimes may have no activity at all. In the case of Echinacea purpurea. (L.) Moench (Asteraceae), cichoric acid is one of the commonly used markers, whereas echinacoside is the marker most frequently used for Echinacea angustifolia. DC (Asteraceae).
Many papers have pointed out the importance of polysaccharides and glycoproteins (PS/GP) as compounds partly responsible for the immunostimulatory effect of Echinacea. from plant material (Stimpel et al., Citation1984; Wagner et al., Citation1985; Beuscher et al., Citation1990; Bodinet et al., Citation1993; See et al., Citation1997; Alban et al. Citation2002) or cell cultures (Wagner et al., Citation1988; Luettig et al., Citation1989; Roesler et al., Citation1991aCitation1991b; Steinmüller et al., Citation1993; Emmendorffer et al., Citation1999).
Different polysaccharides and glycoproteins have been found in Echinacea. species
An inulin-type fructan (6 kDa), heterogeneous polysaccharides (10–50 kDa), an acidic arabinogalactan (70 kDa), and an arabinogalactan-protein (1200 kDa) have been isolated from E. purpurea. herb–pressed juice (Stuppmen, 1985; Blaschek et al., Citation1998; Classen et al., Citation2000). A 4-O.-methylglucuronoarabinoxylan (35 kDa, PSI), an acid rhamnoarabinogalactan (45 kDa, PSII), a xyloglucan (79.5 kDa), and EPS (Echinacea Purified Polysaccharides) (450 kDa) were also isolated from E. purpurea. (Proksch and Wagner, Citation1987; Stimpel et al., Citation1984; Stuppner, Citation1985; Blaschek et al., Citation1998; Wichtl, Citation2002). Glycoproteins (17, 21, 30, and 40 kDa) have been reported in E. angustifolia. and E. purpurea. roots (Beuscher et al., Citation1987, 1990).
A major problem has been finding a suitable method to quantify these large molecules in raw material and extracts. In our review of the existing methods for polysaccharide quantification from Echinacea. species, we saw two non-peer-reviewed methods emerging from a government report (Stuart et al., Citation2004) and a patent publication (Giori et al., Citation2005). Stuart et al. (Citation2004) described an HPLC method using a refractive index (RI) detector. They stated that the RI was 100-times less sensitive than the evaporative light scattering detection (ELSD) for the detection of polysaccharides. Furthermore, the preparation of their samples was time consuming as they were using dialysis. We used a molecular filter instead of dialysis in order to save time in the sample preparation. More recently, a gravimetric method was described (Brovelli et al., Citation2005); this method is simple but will quantify the mono- and oligosaccharides as well as the low-molecular-weight polysaccharides.
Initially, we used a modified phenol/sulfuric acid assay (Dubois et al., Citation1956), which included a step with a molecular weight filter, but the high variability of the results from sample to sample prompted us to develop another method (Gafner et al., Citation2001). ELSD was the detection method of choice in terms of sensitivity to nonvolatile and weakly chromophoric or nonchromophoric compounds such as carbohydrates (Young & Dolan, Citation2004). Moreover, as no polysaccharide standards from Echinacea. are commercially available, the ELSD provides a more uniform response to structurally similar analytes than do light-absorbing detectors. For many analyte classes, users can create an universal calibration set from a single analyte to quantify all analytes of the same class (Young & Dolan, Citation2004). For this reason, an arabinogalactan isolated from Larix. sp. (Pinaceae) was choosen to quantify the polysaccharides from Echinacea. spp. as it is commercially available.
We report here an analytical method for the quantification of the glycoproteins and polysaccharides using HPLC-ELSD for quality control of raw material or extracts. The method quantifies the polysaccharide/glycoprotein fraction of Echinacea. spp. as an arabinogalactan equivalent.
Materials and Methods
General
Glycerin USP (Univar, Salem, MA, USA) and Crystal spring water (Auburn, ME, USA) were used for the Echinacea. spp. extraction. For analytical purposes, demineralized water was obtained from a Direct Q Ultrapure water system (Millipore, Bedford, MA, USA). Other solvents used were HPLC-grade MeCN (Fisher Chemicals, Fairlawn, NJ, USA) and HPLC-grade alcohol (Fisher Chemicals).
Standards
Arabinogalactan UF Powder (lot no. 2UF0103801) was purchased from Larex Inc. (White Bear Lake, MN, USA). Dextran standards were purchased from Phenomenex (Torrance, CA, USA).
Plant material and extraction
Echinacea purpurea. (L.) Moench. (Asteraceae) certified organic herb was purchased from Trout Lake Farm (Trout Lake, WA, USA) and Blessed Herbs (Oakham, MA, USA). E. purpurea. roots were obtained from San Francisco Herbs & Natural Food Co. (Fremont, CA, USA), Trout Lake Farm, and Blessed Herbs. E. angustifolia. DC (Asteraceae) roots were purchased from Trout Lake Farm and Blessed Herbs.
Tea
Five grams of plant material (E. purpurea. herb, E. purpurea. root or E. angustifolia. root) were extracted in 100 mL of hot water for 30 min.
Glycerite and ethanol extract
Five grams of plant material (Echinacea purpurea. herb, E. purpurea. root or E. angustifolia. root) were extracted in 100 mL of 65% aqueous glycerin or 65% aqueous ethanol, for 24 h.
Sample preparation
Each of the extracts was prepared in triplicate. A volume of 500 µL of the sample (tea, glycerite, or ethanol extract) was mixed with 2.5 mL of HPLC-grade ethanol and then centrifuged for 10 min. The supernatant was removed using a Pasteur pipette, and the pellet was dissolved in 1 mL of hot water (80°C). This solution was put on a 5-kDa molecular filter (Amicon Ultra-4 centrifugal filter devices, Millipore), 1 mL of hot water was added, and the solution was then centrifuged for 1 h at 3500 rpm. The residue left on the filter was recovered with 4 × 0.5 mL of hot water and the volume adjusted to 2 mL in a volumetric flask.
HPLC analysis
The analyses were performed on an Agilent HPLC system (Series 1100) with a quaternary pump and automatic sample injector (Agilent Technologies, Burlington, MA, USA) equipped with a Sedex 75 ELSD detector (Sedere, Lawrenceville, NJ, USA). The parameters on the ELSD were gain, 9; temperature, 50°C. Separation was achieved on a BioSep-SEC-S 2000 column (300 × 4.6 mm i.d.; Phenomenex). Eluent (A) H2O and (B) MeCN: 0 min, 0% B; 10 min, 0% B; 11 min, 100% B; 15 min, 100% B; 16 min, 0% B; 25 min, 0% B. The flow rate was 0.7 mL/min. Each sample was injected once.
Method validation
A glycerin extract prepared from 5 g of a tea bag cut of E. purpurea. herb extracted 24 h in 100 mL aqueous glycerin (65%) was used for the method validation.
Instrument precision
One sample prepared from E. purpurea. aqueous glycerin (65%) extract was injected 10 times.
Intermediary precision
Six samples of the E. purpurea. extract were prepared by analyst no. 1 and injected once. On the following day, six samples of the same extract, prepared by analyst no. 2, were injected.
Accuracy
Samples (0.5 mL) ofE. purpurea. glycerite 65% were spiked with four different amounts (4.00, 6.83, 8.94, 10.25 mg) of the arabinogalactan standard, and the percentage of recovery was determined.
Linearity
A quadratic function gave the best fit for the calibration curve of thearabinogalactan standard, detected by the ELSD with r2 value of 0.99999.
Results and Discussion
Method development
A fraction containing polysaccharides and glycoproteins (PS/GP) was obtained by precipitation of E. purpurea. extract with EtOH. The precipitate was taken up in water and passed through a 5-kDa molecular weight centrifugal filter to remove smaller molecules. The PS/GP fraction (MW > 5 kDa) was dissolved in water and analyzed by HPLC with an ELSD detector.
In order to evaluate the time necessary to precipitate the polysaccharides in an ethanol solution, two samples were placed in a refrigerator at 0°C and two at room temperature for 24 h and then extracted, and two samples were extracted after 10 min. The results showed no significant difference. In order to save time, we chose the 10-min method for the precipitation of the PS/GP fraction.
In order to evaluate the benefit or lack thereof of a step to remove the free proteins, the samples were boiled for 5 min and then centrifuged in order to eliminate the free proteins (Classen et al., Citation2000) and the polysaccharides were then determined. There was no difference in the results between boiled samples and the control. We concluded that boiling was unnecessary. We also used trichloroacetic acid (TCA) (results not shown) on the polysaccharide fraction to precipitate proteins, but the TCA trace was detectable by the ELSD and interfered with the polysaccharides in the chromatogram.
We compared three different drift tube temperatures on the ELSD, which were 40°C, 50°C, and 60°C. There was no significant difference between the results.
The temperature of the drift tube was set at 50°C. This temperature is sufficient to evaporate the mobile phase. This low temperature (compared with the boiling temperature of water) is explained by the fact that the ELSD produces a mist of microdroplets, and evaporating the microdroplet mist is far easier (less heat is needed) than evaporating a liquid.
Method validation
As standards for Echinacea. polysaccharides and glycoproteins are not commercially available, we chose to use the arabinogalactan from Larix. sp. and quantified the PS/GP fraction as an arabinogalactan equivalent. The molecular weight of this arabinogalactan varies between 16 and 24 kDa. The cutoff of the molecular filter was chosen at 5 kDa as the molecular weight reported for PS/GP isolated from Echinacea. spp. varies between 6 and 1200 kDa (Stuppner, Citation1985; Blaschek et al., Citation1998; Classen et al., Citation2000; Wichtl, Citation2002).
The HPLC-ELSD method () allows a good separation, linearity (r2 = 0.99999), and repeatability (2.36% of relative standard deviation). The recovery rate (95.47±1.21%) was in the acceptable range (80–100%). The time (approximately 2 h) consumed for the preparation of the sample is principally due to the centrifugation steps.
Table 1. Validation of the method of analysis of the glycoprotein/polysaccharide fraction using Echinacea purpurea. glycerite
Instrument precision (repeatability)
Ten injections of a sample of aqueous glycerin (65%) extract of E. purpurea. herb were measured in order to establish the precision. The average concentration of the fraction PS/GP was 2.65%, expressed in terms of dry weight, with a relative standard deviation of 2.36%.
Intermediate precision
Six samples were prepared by analyst no. 1 from a 65% glycerin extract of E. purpurea. and injected. On the following day, six samples prepared by analyst no. 2 from the same extract were injected. Analyst no. 1 obtained a concentration of 7.06% and analyst no. 2 obtained 7.21% (the mean of the two analysts' results was 7.14% and the relative standard deviation 0.57%). The samples used to measure the instrument precision and intermediary precision were from a different source, which explains the difference in the results obtained.
Accuracy
A sample (0.5 mL) of E. purpurea. glycerite 65% was spiked with four different amounts (4.00, 6.83, 8.94, 10.25 mg) of arabinogalactan; 95.47% of the standard was recovered with a RSD of 1.21%
Linearity
The relation between peak area and amount of standard was found to be quadratic [y. = 41.23925 × 2 + 249.61107 × − 146.45137; where y. is area and x. is amount (µg)] for the arabinogalactan standard, detected by the ELSD with r2 value of 0.99999.
The limit of quantification was 0.95 µg with a signal-to-noise ratio of 5.2, and the limit of detection for a signal-to-noise ratio of 2.2 was 0.88 µg.
Analysis of E. angustifolia. and E. purpurea.
Plant species
A comparison with dextran standards () confirmed that the dextran standard with a median molecular weight of 48.6 and 5.2 kDa appears between 2.5–4 min and 4–5.5 min, respectively.
Figure 1 HPLC-ELSD chromatogram of polysaccharides/glycoproteins in E. purpurea. herb, E. purpurea. roots, and E. angustifolia. tea compared with dextran standard.
E. angustifolia. roots contain principally the 5–48.6 kDa fraction (). E. purpurea. roots contain both a 5–48.6 kDa fraction and above 48.6-kDa compounds with a predominance of the fraction under 48.6 kDa. E. purpurea. aerial parts contain mainly the fraction above 48.6 kDa and a smaller quantity of 5–48.6 kDa.
The glycerite of E. purpurea. roots contains an equivalent quantity () of the fractions 5–48.6 kDa and above 48.6 kDa. The ethanol extract of E. purpurea. roots contains mainly the 5–50 kDa polysaccharide fraction. By comparison with a glycoprotein standard, we saw that the glycoproteins had a chromophore that can be detected at 330 nm; therefore, we conclude that glycerin allowed a better extraction of the glycoproteins than ethanol, possibly because of the pH of the glycerin.
Figure 2 HPLC-ELSD-UV chromatogram of the glycerin and EtOH extracts of E. purpurea. roots.
Solvent
In this experiment, the PS/GP content varied between 0.84% and 5.98% of dry plant weight, and it was higher with a hot water extraction. In an experiment with different solvents (), using the same plant lot, it was found that the hot water was better for extracting the PS/GP fraction than 65% glycerin and 65% ethanol in both E. purpurea. root and E. angustifolia. root. For the glycerin and the ethanol extract, the water was added before the organic solvent, which explains the higher level of polysaccharides than previously stated by Gafner and collaborators (2001).
Table 2. Concentration of polysaccharides/glycoproteins expressed as a percentage of dry Echinacea. extract
Cut
The content of PS/GP may vary according to the type of cut and the plant part (). The tea-bag cut gave a higher yield than the cut-and-sifted. The content of PS/GP is generally higher in the root than in the aerial part, but it varies depending on the origin of the plant material and the lot. According to Stuart and colleagues (2004), the average concentration of each plant section over the course of the plant growth revealed that the stem and root contain the highest polysaccharide content with the leaf and flower being significantly lower. The results vary also in the different organs as a function of the growth stage. The level of polysaccharides was higher in the root, flower, and leaf at senescence. The different stages of growth at the time of harvest may be one of the reasons why our results vary depending on origin and plant part.
Table 3. Concentration of polysaccharides/glycoproteins expressed as a percentage of dry extract weight in Echinacea. from different origins and with different cuts
In our experiment, the level of PS/GP varies depending on the origin of the plant material and cut from 0.84–7.21% in the aerial parts of E. purpurea., to 1.91–5.67% in E. purpurea. roots, and to 0.12–2.98% in E. angustifolia. roots. This method quantifies both the polysaccharides and the glycoproteins. Other peer-reviewed and non-peer-reviewed methods differ from ours principally in the use of a step to precipitate the proteins. The level of polysaccharides obtained by Stuart et al. (Citation2004) varies from 0.34% to 1.81% in different plant parts at different stages of development. The Australian group used an alkaline water solution for the extraction and further precipitated the proteins with trichloroacetic acid, then centrifuged and precipitated the polysaccharides in the supernatant with an 80% ethanol solution overnight, finally centrifuging and resuspending the pellet in 2% sodium stearate to remove insoluble material. The polysaccharides were precipitated again in a 75% ethanolic solution overnight, centrifuged, and a water solution was dialysed with a 5-kDa molecular weight cutoff against water for 24 h. The preparation of the sample was time-consuming as it required at least 3 days and 2 nights to precipitate the polysaccharides and 24 h of dialysis. To quantify the polysaccharide content, they used HPLC coupled with an RI detector. The level of polysaccharides obtained by this group is lower than what we obtained principally because they were measuring only polysaccharides and not glycoproteins. According to their experience, the limit of detection for the E. purpurea. was 100-times lower with an ELSD than with an RI.
In a patent by the Indena group, the roots of E. angustifolia. were extracted with water at a temperature ranging from 20°C to 100°C and then concentrated to a small volume and the residue dissolved in a water-ethanol solvent mixture (50–70% v/v) to obtain a precipitate. The suspension was filtered and then the precipitate washed. The determination of the polysaccharide content was done by HPLC using an ELSD detector as well.
Dalby-Brown et al. (Citation2005) used a series of solvent mixtures to extract 1-year-old E. purpurea. root (EtOH-H2O; 80:20, 50:50 followed by a water extract). The extract was subsequently dialysed at a 3.5-kDa cutoff to remove the low-molecular-weight compounds to obtain a yield of 4.4%. This method gives results comparable with our results as they also quantify the glycoproteins, but in an industrial context, the dialysis is difficult to realize for a quality control purpose.
Brovelli et al. (Citation2005) used a gravimetric method that consisted of precipitating the polysaccharides three-times with 60–80% aqueous ethanol to obtain 1.91% and 4.20% of crude polysaccharides in dried Echinacea. aerial parts. This method is easy but requires a long time for drying (15 h). It includes also the simple sugars in the final results.
Conclusions
This method shows good repeatability and good recoveries and confirmed that HPLC using ELSD as detector is a suitable method for the analysis of the PS/GP fraction of Echinacea. spp. The time for the preparation of the sample is reasonable (≈ 2 h) if we compare with the other methods used for the polysaccharide quantification. The results showed that the composition of polysaccharides differs among different species of Echinacea., among different organs of the same species, and also among different origins. The E. angustifolia. root extract contains the highest proportion of low-molecular-weight molecules. E. purpurea. roots contain polysaccharides from both the 5–48.6 kDa and above 48.6 kDa groups, whereas E. purpurea. aerial parts contain mainly the PS/GP fraction with a molecular weight above 48.6 kDa. Hot water and 65% glycerin were the best solvents for isolating the PS/GP fraction with a molecular weight above 48.6 kDa.
Acknowledgment
We thank Elizabeth R. Dumas for her help with editing the manuscript.
References
- Alban S, Classen B, Brunner G, Blaschek W (2002): Differentiation between the complement modulating effects of an arabinogalactan-protein from Echinacea purpurea. and heparin. Planta Med 68: 1118–1124.
- Beuscher N, Kopanski L, Ernwein C (1987): Modulation of immune response by polymeric substances from Baptisia tinctoria. and Echinacea angustifolia.. Immunomodulators and non-specific host defence mechanisms against microbial infections. Adv Biosci 68: 329–336.
- Beuscher N, Scheit KH, Bodinet C, Egert D (1990): Modulation of host resistance by polymeric substances from Baptisia tinctoria. and Echinacea purpurea.. Immunotherapeutic. In: Masihi KN, Lange W, eds., Prospects of Infectious Diseases 1st ed., Berlin, Springer-Verlag, pp. 59–63.
- Blaschek W, Döll M, Franz G (1998): Echinacea-Polysaccharide. Z Phytother 19: 255–262.
- Bodinet C, Willigmann I, Beuscher N (1993): Host-resistance increasing activity of root extracts from Echinacea. species. Planta Med 59: A672–A673.
- Brovelli EA, Rua D, Roh-Schmidt H, Chandra A, Lamont E, Noratto GD (2005): Human gene expression as a tool to determine horticultural maturity in a bioactive plant (Echinacea purpurea. L. Moench). J Agric Food Chem 53: 8156–8161.
- Classen B, Witthohn K, Blaschek W (2000): Characterization of an arabinogalactan-protein isolated from pressed juice of Echinacea purpurea. by Yariv reagent. Carbohydr Res 327: 497–504.
- Dalby-Brown L, Barsett H, Landbo AK, Meyer AS, Molgaard P (2005): Synergistic antioxidative effects of alkamides, caffeic acid derivatives, and polysaccharide fractions from Echinacea purpurea. on in vitro. oxidation of human low-density lipoproteins. J Agric Food Chem 53: 9413–9423.
- Dubois M, Gilles KA, Hamilton JK, Rebers PA, Smith F (1956): Colorimetric method for determination of sugars and related substances. Anal Chem 28: 350–356.
- Emmendorffer AC, Wagner H, Lohmann-Matthes ML (1999): Immunologically active polysaccharides from Echinacea purpurea. plant and cell cultures. In: Wagner H, ed., Immunomodulatory Agents from Plants. Basel, Birkhäuser, pp. 89–103.
- Gafner S, Bergeron C, Keil J, Batcha LL, Angerhofer CK (2001): Comparison of extraction efficiency of 50% ethanol versus 65% glycerin: HPLC analysis of Echinacea purpurea. (L.) Moench, Ginkgo biloba. L., Hypericum perforatum. L., Panax quinquefolius., L. and Valeriana officinalis. L. American Society of Pharmacognosy Annual Meeting, New Brunswick, NJ.
- Giori A, Anelli A, Morazzoni P, Di Pierro F (2005): Echinacea angustifolia. extracts. U.S. Patent no. 6,881,426.
- Luettig B, Steinmuller C, Gifford GE, Wagner H, Lohmann-Matthes ML (1989): Macrophage activation by the polysaccharide arabinogalactan isolated from plant cell cultures of Echinacea purpurea.. J Natl Cancer Inst 81: 669–675.
- Proksch A, Wagner H (1987): Structural analysis of a 4-O.-methyl-glucuronoarabinoxylan with immuno-stimulating activity from Echinacea purpurea.. Phytochemistry 26: 1989–1993.
- Roesler J, Steinmuller C, Kiderlen A, Emmendorffer A, Wagner H, Lohmann-Matthes ML (1991a): Application of purified polysaccharides from cell cultures of the plant Echinacea purpurea. to mice mediates protection against systemic infections with Listeria monocytogenes. and Candida albicans.. Int J Immunopharmacol 13: 27–37.
- Roesler J, Emmendorffer A, Steinmuller C, Luettig B, Wagner H, Lohmann-Matthes ML (1991b): Application of purified polysaccharides from cell cultures of the plant Echinacea purpurea. to test subjects mediates activation of the phagocyte system. Int J Immunopharmacol 13: 931–941.
- See DM, Broumand N, Sahl L, Tilles JG (1997): In vitro. effects of Echinacea. and ginseng on natural killer and antibody-dependent cell cytotoxicity in healthy subjects and chronic fatigue syndrome or acquired immunodeficiency syndrome patients. Immunopharmacology 35: 229–235.
- Steinmüller C, Roesler J, Grottrup E, Franke G, Wagner H, Lohmann-Matthes ML (1993): Polysaccharides isolated from plant cell cultures of Echinacea purpurea. enhance resistance of immunosuppressed mice against systematic infections with Candida albicans. and Listeria monocytogenes.. Int J Immunopharmacol 15: 605–614.
- Stimpel M, Proksch A, Wagner H, Lohmann-Matthes ML (1984): Macrophage activation and induction of macrophage cytotoxicity by purified polysaccharide fractions from the plant Echinacea purpurea.. Infect Immun 4: 845–849.
- Stuart DL, Wills RBH, Dickeson TM (2004): Optimisation of Polysaccharides in Processed Echinacea purpurea. RIRDC publication. No. 04/118. Barton, Australia.
- Stuppner H (1985): Chemische und Immunologische Untersuchungen von Polysacchariden us der Gewebekultur von Echinacea purpurea. (L.) Moench. PhD thesis, Ludwig-Maximilians-Universität, Munich, Germany.
- Wagner H, Proksch A, Riess-Maurer I, Vollmar A, Odenthal S, Stuppner H, Jurcic K, Le Turdu M, Fang JN (1985): Immunostimulating action of polysaccharides (heteroglycans) from higher plants. Arzneimittelforschung 35: 1069–1075.
- Wagner H, Stuppner H, Schafer W, Zenk M (1988): Immunologically active polysaccharides of Echinacea purpurea. cell cultures. Phytochemistry 27: 119–126.
- Wichtl M (2002): Echinacea angustifolia radix. Herbal Drugs and Phytopharmaceuticals 3rd ed, Boca Raton, CRC Press, pp. 179–182.
- Young CS, Dolan JW (2004): Success with evaporative light-scattering detection, part II: Tips and techniques. LC. GC Europe 17: 192–199.