POLYPHENOLS FROM BULGARIAN MEDICINAL PLANTS WITH ANTI-INFECTIOUS ACTIVITY Stephanie Ivancheva,a Nadya Manolova,b Julia Serkedjieva,a Valentin Dimov,b and Nina Ivanovska b aBotanical Institute Bulgarian Academy of Sciences 1113 Sofia, Bulgaria bInstitute of Microbiology Bulgarian Academy of Sciences 1113 Sofia, Bulgaria ABSTRACT Three Bulgarian medicinal plants-Geranium macrorrhizum L. and G. sanguineum L. (Geraniaceae), and Epilobium hirsutum L. (Onagraceae) were analyzed phytochemically. Different polyphenols like flavonoids and tannins have been found to be principal constituents of the plants. A series of water or alcohol extracts was obtained, and their antiinfectious activity was tested. A significant inhibitory effect of wateralcohol extract and of four fractions from the polyphenolic mixture of E. hirsutum on the reproduction of influenza viruses in vitro, in ovo, and in vivo was established. Four extracts from G. macrorrhizum and three extracts from G. sanguineum were studied for in vitro inhibition of the growth of some Gram-negative bacteria (Klebsiella pneumoniae, Proteus vulgaris, Escherichia coli, Pseudomonas aeruginosa), Grampositive bacterium (Staphylococcus aureus), and fungus (Candida albicans). Some geranium extracts caused a strong increase of the survival rate in an infection with K. pneumoniae in mice. Augmentation of the nonspecific host resistance in relation to the influence of the extracts on the classical complement activation pathway was also studied. INTRODUCTION The study of medicinal plants as a source of new biologically active substances and their detailed theoretical and practical examination is a promising up-to-date Plant Polyphenols, Edited by R.W. Hemingway and P.E. Laks, Plenum Press, New York, 1992
717
718
Ivan che va
approach. We have set up a program for isolation of polyphenols from Bulgarian plants and determination of their biological activities. Two species offamily Geraniaceae-Geranium macrorrhizum L. and Geranium sanguineum L., and Epilobium hirsutum L. - family Onagraceae - have been investigated for flavonoids and tannins and their biological activity studied. The compounds are derivatives of quercetin (3), kaempferol, myricetin, gallotannins, catechins, polyphenolic acids. l ,2,3,4 G. macrorrhizum is a grassy perennial plant, spread primarily along the mountains of the Balkan Peninsula and partly also in the Carpatians and the Alps. In Bulgaria, it is known under the name 'Zdravets,' which means 'healthy' in Bulgarian. It is used for production of an essential oil containing mainly sesquiterpene compounds with germacrons as a principal component. The whole plant is rich in tannins with more in the stems than in the green foliage. No alkaloids and cardiological glycosides have been found. A methanol extract from leaves or stems possesses strong hypotensive activity, and cardiotonic, capillary, and sedative action as well. l G. sanguineum is also a perennial grassy plant. In Bulgarian folk medicine, the root systems are used as astringent, and for anti-inflammatory treatments of diarrhea, gastric-enteric catarrh, and dysentery. It is applied also in cases of scrofula and tuberculosis. E. hirsutum is a perennial plant widely distributed in Bulgaria. Different extracts of the plant are effective folk methods for treatment of skin infections. 3 Many plant products including polyphenolic substances have been subjected to intensive studies for inhibition of viral replication. 5,6, 7,8 The antiviral effect of polyphenolic complexes and their fractions from G. sanguineum and E. hirsutum were tested on the reproduction of influenza viruses in diverse systems. The information about the use of geranium extracts or decoctions in traditional medicine for the cure of several infectious diseases, especially caused by Gramnegative bacteria, determined our studies on the effects of polyphenolic extracts in regard to in vitro and in vivo anti-infectious properties.
MATERIALS AND METHODS Plant Material G. macrorrhizum, G. sanguineum, and E. hirsutum were collected when flowering in Bulgaria, and voucher specimens were deposited in the herbarium at the Botanical Institute, Sofia. Extraction of Polyphenols Rhizomes and overground parts of G. macrorrhizum were extracted with diethyl ether, chloroform, ethylacetate, methanol, butanol, or water. G. sanguineum rhizomes were extracted with methanol and butanol. Paper chromatography (DPC) was on Whatman 3 with the solvent systems: butanol/acetic acid/water (4:1:2.2) and 15 percent acetic acid. Thin layer chromatography was on silica gel with the solvent system: toluene/methylethylketone (9: 1) ; polyamide: toluene /b enzene / methylethylketone / methanol (30: 90:2: 1.5).
Polyphenols with Anti-infectious Activity
719
Column chromatography was on Sephadex LH-20, followed by chromatography on polyamide Sl1 and silica gel S. Identification was based on standard chemical and physical methods-IR, UV, NMR, and MS, according to Harborne9 and Markham1o , by chemical transformation, and direct comparison with authentic samples. E. hirsutum was routinely extracted, and a water-alcohol extract (WAE) and a total polyphenolic mixture were obtained. From the latter, four fractions (E 1-E4 ) were subsequently prepared. Influenza Viruses A/PR/8/34 (HINl) and A/Hong Kong/l/68 (H3N2) were used as allantoic fluids with infectious titers 6.5-7.5 19 E1D so /O.l mL and as 10 percent suspension of virus infected mice lungs with infectious titer 3.5 19 LD 5o /O.1 mL. Pathogenic Microorganisms
Klebsiella pneumoniae, strain 52145 (Institute Pasteur, Paris); Proteus vulgaris: strain D52; Escherichia coli: strain WF+ (Central Institute of Microbiology and Immunology, Jena); Pseudomonas aeruginosa: strain 2 (Chemica-Pharmaceutical Institute, Sofia); Staphylococcus aureus: strain 209 and Candida albicans: strain 562 (Institute of Microbiology, Sofia). Antiviral Activity
In vitro tissue cultures. In vitro tissue cultures of surviving chorioallantoic membranes (CAM) from 11-day old embrionated hen's eggs, prepared according to Horvath,11 were used. The antiviral activity was estimated by the difference in the infectious titers of control and treated viruses, calculated according to Reed and Muench and expressed in 19 1Dso/mL. The hemagglutination titers were determined according to Hirst. Embrionated hen's eggs. Ten ll-day-old eggs were inoculated into allantoic cavity with 100 E1D5o/mL of the virus and treated with the substance 1 hour before or simultaneously with viral infection. The antiviral activity was estimated by the difference of infectious and hemagglutination titers of control and treated viruses and by means of the protective index (PI) according to Ilyenko. 12 Influenza virus infection. Infection was produced in white mice, strain ICR, 18 g with 10 LD50 A/PR/8 and A/Hong Kong inoculated nasally. The substances had been applied intraperitoneally for 5 days following the infection in a single dose of 35 or 70 mg/kg daily. The antiviral activity was determined by the difference of the infectious titers from the lungs of control and treated animals determined in CAM and by reduction of mortality rate for 21 days. Lung suspensions were prepared by a standard method. The difference in infectious titers ~1 19 was considered significant. The results are the mean values of three experiments. Antibacterial and Antifungal Activities
In vitro tests. A modification of the diffusion method of Bauer13 was used. All the microorganisms were cultured for 24 hours at 37°C in digest agar. Bacterial
720
Ivancheva
colonies were washed down with saline, and the suspensions were diluted to the final concentration of 107 cells/mL. Each plate with digest agar (100 mm diameter) was inoculated with 0.2 mL of the microorganism tested. Five wells (10 mm diameter) were then made and filled with 0.2 mL of various concentrations of the geranium extracts. After 24 hours of incubation at 37°C, the inhibition zones were measured. The following antibiotics in their minimal inhibitory concentrations were included as positive controls: Chloramphenicol (EGYT, Hungary) for Klebsiella pneumoniae; Amicacin (Bristol-Myers, United States) for Proteus vulgaris; Keflin (Lilly, United States) for Escheria coli and Pseudomonas aeruginosa; Benzylpenicillin (Medexport, USSR) for Staphylococcus aureus and Nystatin (Pharmachim, Bulgaria) for Candida albicans. Experimental infection. The inoculation dose for Klebsiella pneumoniae infection (25-30 cells per mouse, s.c.) was chosen after preliminary experiments to achieve a survival rate of approximately 50 percent. The course of infection was followed up to the eighth day, when the survival rate (S, in percents) and mean survival time (MST, in days) were calculated. 14 The protection induced was estimated by differences (.6.), defense indices (DI), and efficacy values as shown in table 6. Complement assay. Classical pathway (CP) complement activity was determined using a modified version of a standard microtiter assay after the methods of Klerx and others.15 The samples of geranium extracts were serially diluted in 0.1 mL veronal buffer saline (VSB 2 +) in 96-well microtiter plate. Immediately, 0.1 mL of diluted guinea pig serum was added to each well. Serum dilution was chosen as to give 50 percent hemolysis (CH50). After a preincubation at 37°C for 30 min, 0.05 mL of a 2.10 8 suspension of sensitized sheep erythrocytes were added per well. The plates were incubated at 37°C for 60 min and then centrifuged at 3000 rpm for 10 min. To quantify hemolysis, 0.05 mL of supernatants were mixed with 0.2 mL water, and the light absorbance at 405 nm was measured. RESULTS AND DISCUSSION
We present some results of investigation of both overground and rhizome parts of G. macrorrhizum, G. sanguineum, and E. hirsutum. From the overground green part of G. macrorrhizum, compounds (1) -+ (8) were isolated. From the rhizome of the same plant were isolated: acetovanillone (8), retusin (1), ermanin (4), {3sitosterol-D-glucoside( daucosterin)( 9) ,{3-sitosterol( 10) ,1-galloyl-{3-D-glucopyranose (glycogallin) (11), tannic acid (12). The use ofreagents specific for tannins allowed gallotannins and catechins to be established. 1 The geranium extracts are numbered in table 1. The polyphenolic complex (PC=GS s ) from rhizome of G. sanguineum contains gallotannins, catechins, flavonoid-derivatives of quercetin (3), kaempferol, myricetin, morin, and chinic acid. 2 The phytochemical investigation of the total polyphenolic mixture from E. hirsutum identified more than six components. It was mainly flavonols and related
721
Polyphenols with Anti-infectious Activity
(1) Retusin: R-Rs-R~CH3
Rs
(2) Rannasin: R=Rt+t Rs-R:rCH=C~ (31 Quercetin: Rs·R~ OH
0
(-4) Ermanin: R-H; Rs-CH3
(5) Kumatakenin: R-Di~ Rs-H (6)
OR s
Isakaempferid: R-Rs-H OH
m Maltal
(8) Acetavanllan
~ OH
0
0
~
OC~
OH
polyphenolic substances. The presence of kaempferol, quercetin (3), and myricetin was shown by chromatographic analysis in fractions Es and E 4 •S The polyphenolic complex (PC), isolated from G. sanguineum, inhibited the reproduction of influenza viruses type A and B in vitro, in OVO, and in vivo. 6 When influenza viruses were treated with PC, their hemagglutination, neuraminidase, and infectious activities were reduced. 16 It was assumed that the PC - specific inhibitory effect in vitro was mediated by partial inactivation of viral glycoproteins. The complex also inhibited the reproduction of herpes simplex virus in vitro. 17 The antiviral effect of PC was related to the presence of large quantities of polyphenolic substances. 2 ,18 These findings are discussed in detail in the Chapter by Serkedjieva in this book. The effect ofWAE from E. hirsutum was studied on the reproduction of influenza A viruses. The results are presented in table 2. Applied 1 h before or simultaneously with the viral infection, the WAE reduced the infectious titer of HINI strain with 2 - 2.5 Ig and the titer of H3N2 strain with 1.66 - 3.0 19 in ovo. The protective effect (PI) was 69.23 percent for HINI and 71.59 percent for H3N2 strains when the extract was applied simultaneously with the virus. In the course of experimental influenza infection in mice with HINI, the WAE in doses of 35 and 70 mg/kg applied simultaneously with the virus inhibited its
722
Ivancheva
Table 1. Polyphenol Extracts from Geranium Species Tannins b
Extracts"
Flavonoid b
G. macrorrhizum (rhizome) GM 1 (water) GM 2 (30 percent ethanol) GM3 (70 percent ethanol) GM 4 (90 percent ethanol)
26.75 43.66 49.21 71.42
0.050 0.256 0.143 0.242
G. sanguineum rhizome GSs=PC c (methanol) GS 6 (butanol)
11.02 2.10
0.Q25 0.076
G. sanguineum (leaves) GSs (methanol)
8.50
0.126
"Extracted by water or alcohols. bpercentage in dryed extracts. cThe polyphenolic complex (PC) was the designation of the GSs in our previous papers. 2 ,6,16-18
Table 2. Effect of WAE of E. hirsutum on the Reproduction of Influenza Viruses In Ovo
Strains
Schedule of application ofWAE"
Inhibition IgEIDso mL PI (percent)
A (HINl)
1 h before Simultaneously
2.50 2.00
n.m b 69.23
A (H3N2)
1 h before Simultaneously
1.66 3.00
n.m. 71.59
"In a concentration of 1.33 mg/mL. bn.m.-not made.
Polyphenols with Anti-infectious Activity
723
reproduction, reduced the lung infectious titer with 1.3 - 2.5 19, and increased the number of survivors (table 3). The extract did not affect the reproduction of virus H3N2 in vivo. The extract was tested on some other viruses in vitro, but there was no inhibitory effect on the reproduction of fowl plague virus in BHK12 cells, of vaccinia virus in CEF, and of herpes simplex virus type 1 in HeLa cells. 19 The antiviral effect of the fractions from E. hirsutum was studied in the course of an experimental influenza infection in mice with virus H3N2. The results of these experiments are presented in table 4. All fractions significantly inhibited the reproduction of virus H3N2; the infectious titer in the lungs of treated animals was reduced with 1.5 - 4.5 19. The effect of fractions El and E3 was most pronounced. The mortality rate was decreased in all experimental groups. Most effective was fraction E3 (PI = 100 percent). The number of surviving animals was also increased in the groups of mice treated with fractions ETE4 in the course of an experimental infection with virus HIN1. Chantrill and others20 considered the tannins to be the active principle of the antiviral activity of an extract from E. hirsutum. The total polyphenolic mixture from this plant used in our experiments was free of tannins. Our results on the inhibitory activity of WAE from E. hirsutum and the fractions on the reproduction of influenza virus in vitro, in ava, and in vivo can be related to the presence of other polyphenolic substances, mainly flavonols. Besides the antiviral activity, antibacterial and antifungal actions of geranium extracts were studied. The ability of all extracts to inhibit the growth of different microorganisms was tested toward Gram-negative pathogens (K. pneumoniae, P. aeruginosa, E. coli, P. vulgaris), Gram-positive pathogens (S. aureus) , and fungi (C. albicans). Table 5 presents only the positive results. No effect was observed in the cases of K. pneumoniae, P. vulgarisa, and C. albicans. Some extracts (GM1' GM 2 , and GM 4 ) exhibited slight inhibition against P. aeruginosa and E. coli, but the zones observed were significantly smaller as compared to antibiotics controls. All geranium extracts except GS 6 suppressed the growth of S. aureus, but the values were negligible. Table 3. Effect of WAE of E. hirsutum on the Reproduction of Influenza Viruses In Vivo
Inhibition Dose a
35 70
A (HlNl)
A (H3N2)
19 IDsomL
19 IDsomL
2.50 1.30
0.25 0.10
aDoses in mg/kg.
724
Ivancheva Table 4. Inhibition of Influenza Viruses by Fractions of E. hirsutum In Vivo
Fractions"
Inhibition A (H3N2) A (HlNl) 19 IDso mL PI (percent) 19IDso mL PI (percent)
4.50 1.50 4.50 1.84
00.00 33.30 66.60 50.30
El E2 E3 Ei
10.0 20.0 100.0 41.0
"In dose of 70 mg/kg. bNot made.
Due to the absence of effective antiviral drugs, research of new substances from plant origin is very important. In contrast, the availability of antibiotics restricts the attention paid to the development of plant substances as antibacterials. Nevertheless, some bactericidal compounds were recently isolated from medicinal plants. 21 In spite of use of geranium in Bulgarian traditional medicine for nonspecific treatment of gastrointestinal infections caused mainly by Gram-negative agents, our data showed that the direct bactericidal effect of the extracts was too limited. Table 5. In Vitro Inhibition of Bacterial Growth by Geranium Extracts
Strains
Ps. aeruginosa
E. coli S. aUTeus
Extracts
2000
GM l GM2 GM. GM l GM2 GM3 GM. GS s GS 8
13.3±2.1b 4.8±0.3 3.2±0.3 2.2±OA 3.0±0.0 4.3±1.S 3.3±0.3 1.0±1.0 3.8±0.3
Concentrations" SOO 1000
2S0
Control 19.7±0.6c
9.2±1.6 3.7±0.3 2.0±0.0 1.8±1.1
7.0±1.0 2.S±0.S
6.3±0.6 - d
0.S±0.7
-11.2±0.8f
3.2±0.6 2.3±1.2
1.8±0.3 2.0±0.0
12.6±0.6e
0.7±0.3
"Concentrations of geranium extracts in J.Lg mL. bInhibitory zones estimated as a half of differences between the mean diameter of each zone and the diameter of the well. Results are expressed as mean ± S.D. of three experiments performed in triplicate. The data in each line are significantly different as compared to the relevant control with antibiotics (p
717
718
Ivan che va
approach. We have set up a program for isolation of polyphenols from Bulgarian plants and determination of their biological activities. Two species offamily Geraniaceae-Geranium macrorrhizum L. and Geranium sanguineum L., and Epilobium hirsutum L. - family Onagraceae - have been investigated for flavonoids and tannins and their biological activity studied. The compounds are derivatives of quercetin (3), kaempferol, myricetin, gallotannins, catechins, polyphenolic acids. l ,2,3,4 G. macrorrhizum is a grassy perennial plant, spread primarily along the mountains of the Balkan Peninsula and partly also in the Carpatians and the Alps. In Bulgaria, it is known under the name 'Zdravets,' which means 'healthy' in Bulgarian. It is used for production of an essential oil containing mainly sesquiterpene compounds with germacrons as a principal component. The whole plant is rich in tannins with more in the stems than in the green foliage. No alkaloids and cardiological glycosides have been found. A methanol extract from leaves or stems possesses strong hypotensive activity, and cardiotonic, capillary, and sedative action as well. l G. sanguineum is also a perennial grassy plant. In Bulgarian folk medicine, the root systems are used as astringent, and for anti-inflammatory treatments of diarrhea, gastric-enteric catarrh, and dysentery. It is applied also in cases of scrofula and tuberculosis. E. hirsutum is a perennial plant widely distributed in Bulgaria. Different extracts of the plant are effective folk methods for treatment of skin infections. 3 Many plant products including polyphenolic substances have been subjected to intensive studies for inhibition of viral replication. 5,6, 7,8 The antiviral effect of polyphenolic complexes and their fractions from G. sanguineum and E. hirsutum were tested on the reproduction of influenza viruses in diverse systems. The information about the use of geranium extracts or decoctions in traditional medicine for the cure of several infectious diseases, especially caused by Gramnegative bacteria, determined our studies on the effects of polyphenolic extracts in regard to in vitro and in vivo anti-infectious properties.
MATERIALS AND METHODS Plant Material G. macrorrhizum, G. sanguineum, and E. hirsutum were collected when flowering in Bulgaria, and voucher specimens were deposited in the herbarium at the Botanical Institute, Sofia. Extraction of Polyphenols Rhizomes and overground parts of G. macrorrhizum were extracted with diethyl ether, chloroform, ethylacetate, methanol, butanol, or water. G. sanguineum rhizomes were extracted with methanol and butanol. Paper chromatography (DPC) was on Whatman 3 with the solvent systems: butanol/acetic acid/water (4:1:2.2) and 15 percent acetic acid. Thin layer chromatography was on silica gel with the solvent system: toluene/methylethylketone (9: 1) ; polyamide: toluene /b enzene / methylethylketone / methanol (30: 90:2: 1.5).
Polyphenols with Anti-infectious Activity
719
Column chromatography was on Sephadex LH-20, followed by chromatography on polyamide Sl1 and silica gel S. Identification was based on standard chemical and physical methods-IR, UV, NMR, and MS, according to Harborne9 and Markham1o , by chemical transformation, and direct comparison with authentic samples. E. hirsutum was routinely extracted, and a water-alcohol extract (WAE) and a total polyphenolic mixture were obtained. From the latter, four fractions (E 1-E4 ) were subsequently prepared. Influenza Viruses A/PR/8/34 (HINl) and A/Hong Kong/l/68 (H3N2) were used as allantoic fluids with infectious titers 6.5-7.5 19 E1D so /O.l mL and as 10 percent suspension of virus infected mice lungs with infectious titer 3.5 19 LD 5o /O.1 mL. Pathogenic Microorganisms
Klebsiella pneumoniae, strain 52145 (Institute Pasteur, Paris); Proteus vulgaris: strain D52; Escherichia coli: strain WF+ (Central Institute of Microbiology and Immunology, Jena); Pseudomonas aeruginosa: strain 2 (Chemica-Pharmaceutical Institute, Sofia); Staphylococcus aureus: strain 209 and Candida albicans: strain 562 (Institute of Microbiology, Sofia). Antiviral Activity
In vitro tissue cultures. In vitro tissue cultures of surviving chorioallantoic membranes (CAM) from 11-day old embrionated hen's eggs, prepared according to Horvath,11 were used. The antiviral activity was estimated by the difference in the infectious titers of control and treated viruses, calculated according to Reed and Muench and expressed in 19 1Dso/mL. The hemagglutination titers were determined according to Hirst. Embrionated hen's eggs. Ten ll-day-old eggs were inoculated into allantoic cavity with 100 E1D5o/mL of the virus and treated with the substance 1 hour before or simultaneously with viral infection. The antiviral activity was estimated by the difference of infectious and hemagglutination titers of control and treated viruses and by means of the protective index (PI) according to Ilyenko. 12 Influenza virus infection. Infection was produced in white mice, strain ICR, 18 g with 10 LD50 A/PR/8 and A/Hong Kong inoculated nasally. The substances had been applied intraperitoneally for 5 days following the infection in a single dose of 35 or 70 mg/kg daily. The antiviral activity was determined by the difference of the infectious titers from the lungs of control and treated animals determined in CAM and by reduction of mortality rate for 21 days. Lung suspensions were prepared by a standard method. The difference in infectious titers ~1 19 was considered significant. The results are the mean values of three experiments. Antibacterial and Antifungal Activities
In vitro tests. A modification of the diffusion method of Bauer13 was used. All the microorganisms were cultured for 24 hours at 37°C in digest agar. Bacterial
720
Ivancheva
colonies were washed down with saline, and the suspensions were diluted to the final concentration of 107 cells/mL. Each plate with digest agar (100 mm diameter) was inoculated with 0.2 mL of the microorganism tested. Five wells (10 mm diameter) were then made and filled with 0.2 mL of various concentrations of the geranium extracts. After 24 hours of incubation at 37°C, the inhibition zones were measured. The following antibiotics in their minimal inhibitory concentrations were included as positive controls: Chloramphenicol (EGYT, Hungary) for Klebsiella pneumoniae; Amicacin (Bristol-Myers, United States) for Proteus vulgaris; Keflin (Lilly, United States) for Escheria coli and Pseudomonas aeruginosa; Benzylpenicillin (Medexport, USSR) for Staphylococcus aureus and Nystatin (Pharmachim, Bulgaria) for Candida albicans. Experimental infection. The inoculation dose for Klebsiella pneumoniae infection (25-30 cells per mouse, s.c.) was chosen after preliminary experiments to achieve a survival rate of approximately 50 percent. The course of infection was followed up to the eighth day, when the survival rate (S, in percents) and mean survival time (MST, in days) were calculated. 14 The protection induced was estimated by differences (.6.), defense indices (DI), and efficacy values as shown in table 6. Complement assay. Classical pathway (CP) complement activity was determined using a modified version of a standard microtiter assay after the methods of Klerx and others.15 The samples of geranium extracts were serially diluted in 0.1 mL veronal buffer saline (VSB 2 +) in 96-well microtiter plate. Immediately, 0.1 mL of diluted guinea pig serum was added to each well. Serum dilution was chosen as to give 50 percent hemolysis (CH50). After a preincubation at 37°C for 30 min, 0.05 mL of a 2.10 8 suspension of sensitized sheep erythrocytes were added per well. The plates were incubated at 37°C for 60 min and then centrifuged at 3000 rpm for 10 min. To quantify hemolysis, 0.05 mL of supernatants were mixed with 0.2 mL water, and the light absorbance at 405 nm was measured. RESULTS AND DISCUSSION
We present some results of investigation of both overground and rhizome parts of G. macrorrhizum, G. sanguineum, and E. hirsutum. From the overground green part of G. macrorrhizum, compounds (1) -+ (8) were isolated. From the rhizome of the same plant were isolated: acetovanillone (8), retusin (1), ermanin (4), {3sitosterol-D-glucoside( daucosterin)( 9) ,{3-sitosterol( 10) ,1-galloyl-{3-D-glucopyranose (glycogallin) (11), tannic acid (12). The use ofreagents specific for tannins allowed gallotannins and catechins to be established. 1 The geranium extracts are numbered in table 1. The polyphenolic complex (PC=GS s ) from rhizome of G. sanguineum contains gallotannins, catechins, flavonoid-derivatives of quercetin (3), kaempferol, myricetin, morin, and chinic acid. 2 The phytochemical investigation of the total polyphenolic mixture from E. hirsutum identified more than six components. It was mainly flavonols and related
721
Polyphenols with Anti-infectious Activity
(1) Retusin: R-Rs-R~CH3
Rs
(2) Rannasin: R=Rt+t Rs-R:rCH=C~ (31 Quercetin: Rs·R~ OH
0
(-4) Ermanin: R-H; Rs-CH3
(5) Kumatakenin: R-Di~ Rs-H (6)
OR s
Isakaempferid: R-Rs-H OH
m Maltal
(8) Acetavanllan
~ OH
0
0
~
OC~
OH
polyphenolic substances. The presence of kaempferol, quercetin (3), and myricetin was shown by chromatographic analysis in fractions Es and E 4 •S The polyphenolic complex (PC), isolated from G. sanguineum, inhibited the reproduction of influenza viruses type A and B in vitro, in OVO, and in vivo. 6 When influenza viruses were treated with PC, their hemagglutination, neuraminidase, and infectious activities were reduced. 16 It was assumed that the PC - specific inhibitory effect in vitro was mediated by partial inactivation of viral glycoproteins. The complex also inhibited the reproduction of herpes simplex virus in vitro. 17 The antiviral effect of PC was related to the presence of large quantities of polyphenolic substances. 2 ,18 These findings are discussed in detail in the Chapter by Serkedjieva in this book. The effect ofWAE from E. hirsutum was studied on the reproduction of influenza A viruses. The results are presented in table 2. Applied 1 h before or simultaneously with the viral infection, the WAE reduced the infectious titer of HINI strain with 2 - 2.5 Ig and the titer of H3N2 strain with 1.66 - 3.0 19 in ovo. The protective effect (PI) was 69.23 percent for HINI and 71.59 percent for H3N2 strains when the extract was applied simultaneously with the virus. In the course of experimental influenza infection in mice with HINI, the WAE in doses of 35 and 70 mg/kg applied simultaneously with the virus inhibited its
722
Ivancheva
Table 1. Polyphenol Extracts from Geranium Species Tannins b
Extracts"
Flavonoid b
G. macrorrhizum (rhizome) GM 1 (water) GM 2 (30 percent ethanol) GM3 (70 percent ethanol) GM 4 (90 percent ethanol)
26.75 43.66 49.21 71.42
0.050 0.256 0.143 0.242
G. sanguineum rhizome GSs=PC c (methanol) GS 6 (butanol)
11.02 2.10
0.Q25 0.076
G. sanguineum (leaves) GSs (methanol)
8.50
0.126
"Extracted by water or alcohols. bpercentage in dryed extracts. cThe polyphenolic complex (PC) was the designation of the GSs in our previous papers. 2 ,6,16-18
Table 2. Effect of WAE of E. hirsutum on the Reproduction of Influenza Viruses In Ovo
Strains
Schedule of application ofWAE"
Inhibition IgEIDso mL PI (percent)
A (HINl)
1 h before Simultaneously
2.50 2.00
n.m b 69.23
A (H3N2)
1 h before Simultaneously
1.66 3.00
n.m. 71.59
"In a concentration of 1.33 mg/mL. bn.m.-not made.
Polyphenols with Anti-infectious Activity
723
reproduction, reduced the lung infectious titer with 1.3 - 2.5 19, and increased the number of survivors (table 3). The extract did not affect the reproduction of virus H3N2 in vivo. The extract was tested on some other viruses in vitro, but there was no inhibitory effect on the reproduction of fowl plague virus in BHK12 cells, of vaccinia virus in CEF, and of herpes simplex virus type 1 in HeLa cells. 19 The antiviral effect of the fractions from E. hirsutum was studied in the course of an experimental influenza infection in mice with virus H3N2. The results of these experiments are presented in table 4. All fractions significantly inhibited the reproduction of virus H3N2; the infectious titer in the lungs of treated animals was reduced with 1.5 - 4.5 19. The effect of fractions El and E3 was most pronounced. The mortality rate was decreased in all experimental groups. Most effective was fraction E3 (PI = 100 percent). The number of surviving animals was also increased in the groups of mice treated with fractions ETE4 in the course of an experimental infection with virus HIN1. Chantrill and others20 considered the tannins to be the active principle of the antiviral activity of an extract from E. hirsutum. The total polyphenolic mixture from this plant used in our experiments was free of tannins. Our results on the inhibitory activity of WAE from E. hirsutum and the fractions on the reproduction of influenza virus in vitro, in ava, and in vivo can be related to the presence of other polyphenolic substances, mainly flavonols. Besides the antiviral activity, antibacterial and antifungal actions of geranium extracts were studied. The ability of all extracts to inhibit the growth of different microorganisms was tested toward Gram-negative pathogens (K. pneumoniae, P. aeruginosa, E. coli, P. vulgaris), Gram-positive pathogens (S. aureus) , and fungi (C. albicans). Table 5 presents only the positive results. No effect was observed in the cases of K. pneumoniae, P. vulgarisa, and C. albicans. Some extracts (GM1' GM 2 , and GM 4 ) exhibited slight inhibition against P. aeruginosa and E. coli, but the zones observed were significantly smaller as compared to antibiotics controls. All geranium extracts except GS 6 suppressed the growth of S. aureus, but the values were negligible. Table 3. Effect of WAE of E. hirsutum on the Reproduction of Influenza Viruses In Vivo
Inhibition Dose a
35 70
A (HlNl)
A (H3N2)
19 IDsomL
19 IDsomL
2.50 1.30
0.25 0.10
aDoses in mg/kg.
724
Ivancheva Table 4. Inhibition of Influenza Viruses by Fractions of E. hirsutum In Vivo
Fractions"
Inhibition A (H3N2) A (HlNl) 19 IDso mL PI (percent) 19IDso mL PI (percent)
4.50 1.50 4.50 1.84
00.00 33.30 66.60 50.30
El E2 E3 Ei
10.0 20.0 100.0 41.0
"In dose of 70 mg/kg. bNot made.
Due to the absence of effective antiviral drugs, research of new substances from plant origin is very important. In contrast, the availability of antibiotics restricts the attention paid to the development of plant substances as antibacterials. Nevertheless, some bactericidal compounds were recently isolated from medicinal plants. 21 In spite of use of geranium in Bulgarian traditional medicine for nonspecific treatment of gastrointestinal infections caused mainly by Gram-negative agents, our data showed that the direct bactericidal effect of the extracts was too limited. Table 5. In Vitro Inhibition of Bacterial Growth by Geranium Extracts
Strains
Ps. aeruginosa
E. coli S. aUTeus
Extracts
2000
GM l GM2 GM. GM l GM2 GM3 GM. GS s GS 8
13.3±2.1b 4.8±0.3 3.2±0.3 2.2±OA 3.0±0.0 4.3±1.S 3.3±0.3 1.0±1.0 3.8±0.3
Concentrations" SOO 1000
2S0
Control 19.7±0.6c
9.2±1.6 3.7±0.3 2.0±0.0 1.8±1.1
7.0±1.0 2.S±0.S
6.3±0.6 - d
0.S±0.7
-11.2±0.8f
3.2±0.6 2.3±1.2
1.8±0.3 2.0±0.0
12.6±0.6e
0.7±0.3
"Concentrations of geranium extracts in J.Lg mL. bInhibitory zones estimated as a half of differences between the mean diameter of each zone and the diameter of the well. Results are expressed as mean ± S.D. of three experiments performed in triplicate. The data in each line are significantly different as compared to the relevant control with antibiotics (p
