1479

Benzoic acid, salicylic acid, and the role of black galls on aspen in protection against decay M. Gabrielle Pausler, William A. Ayer, and Yasuyuki Hiratsuka

(Populus tremuloides Michx.) bearing certain types of black galls Phellinus tremulae (Bond.) Bond. & Boriss. heartwood rot than do

Abstract: Trembling aspen have a lower incidence of

non­

gall-bearing trees. Extraction of finely ground black gall tissue with ethyl acetate and separation of the acidic components of the extract led to the isolation of benzoic acid, trans-cinnamic acid, p-hydroxybenzoic acid, p-hydroxycinnamic acid, naringenin, 7'-methyl-3-hydroxynaringen, aromadendrin, and taxifolin. Bioassays revealed that among these compounds, only benzoic acid showed significant activity against P.

tremulae.

An analytical procedure was developed to measure

the concentration of benzoic acid in various types of aspen tissue. Tissue from the black galls showed a high concentration of benzoic acid, and tissue from gall-bearing trees contained significantly more benzoic acid than healthy nongalled trees. However, the amount of benzoic acid present in the gall-bearing trees may not be sufficient to prevent

Phellinus decay. It

is

suggested that perhaps the benzoic acid serves as a precursor of salicylic acid, a signal molecule in systemic acquired resistance of plants.

Resume: Les peupliers faux-tremble

(Populus tremuloides Michx.)

qui ont un certain type de

galles noires sont moins souvent affectes par la carie de coeur causee par

Phellinus tremulae

(Bond.) Bond. & Boriss. que les arbres exempts de galles. L'extraction a l'acetate d'ethyle du tissu des galles noires finement moulu et la separation des composes acides contenus dans I'extrait ont mene a l'isolation de l'acide benzoique, de l'acide

tran s-cinnamique,

de l'acide

p-hydroxybenzoYque, de l'acide p-hydroxycinnamique, de la naringenine, de la 7'-methyle-3hydroxynaringenine, de l' aromadendrine et de la taxifoline. Des bioessais ont montre que parmi ces composes, seul l'acide benzoYque avait une activite importante contre

P. tremulae.

Une

methode analytique a ete developpee pour mesurer la concentration d'acide benzoYque dans differents types de tissus du peuplier faux-tremble. Le tissu des galles noires contenait une forte concentration d'acide benzoique. Le tissu des arbres avec des galles contenait significativement plus d' acide benzoi"que que celui des arbres sains exempts de galles. Cependant, la quantite d'acide benzoYque presente dans les arbres avec des galles pourrait ne pas �tre suffisante pour prevenir la carie causee par

Phellinus.

II est possible que l'acide benzolque serve de precurseur de

l'acide salicylique, une molecule associee a la resistance systemique acquise chez les plantes. [Traduit par la Redaction]

Introduction Trembling aspen

(Populus tremuloides Michx.) is the most

widely distributed tree species in North America (Peterson and Peterson

1992). In Canada, aspen represents more than

one half of the net merchantable hardwood timber and

11% of the entire timber resource (Hunt et al. 1978). The

commercial utilization of aspen has increased dramatically in the last few years. The increasing scarcity of economi­ cally accessible softwood and the introduction of new tech­ nologies such as the chlorine-free chemothermomechanical Received November 30, 1994. Accepted April 19, 1995.

! M.G. Pansler and W.A. Ayer. Department of Chemistry, University of Alberta, Edmonton, AB T6G 2G2, Canada. Y. Hiratsuka. Forestry Canada, Northern Forestry Centre, 5320-112 Street, Edmonton, AB T6G 3S5, Canada. I

Author to whom all correspondence should be addressed.

pulping (CTMP) process have been important factors in the increased use of this hitherto neglected resource (Ondro

1989; Cheyne 1990; Peterson and Peterson 1992).

A serious limitation on the utilization of aspen is that it

is very susceptible to decay and stain caused by fungi

(Hiratsuka et al. 1990). Heartwood decay, caused by Phellinus tremulae (Bond.) Bond & Boriss., is often present in a high percentage of trees in a mature stand (Hiratsuka et al.

1990). Stored logs of aspen are very susceptible to 1987) and recently we have

blue stain fungi (Hiratsuka

reported on potential biological control methods for pro­

1994; 1994; Ayer and Jimenez 1994).

tection against t h e s e fungi (Hiratsuka et al. Chakravarty and Hiratsuka

Aspen with certain types of black galls o ccurring as stern deformities appear to have a lower incidence of

tremulae than do adjacent non-gaIl-bearing trees (Crane 1993; Crane et al. 1994). The cause of these galls is unknown

P

and they are different from those caused by poplar budgall

Can. 1. For. Res. 25: 1479-1483 (l995). Printed in Canada Ilmprime au Canada

1480 mites (Aceria parapopuli Keifer) (Crane and Hiratsuka 1994). It is possible that the black galls produce com­ pounds that protect the tree from infection, or that the tree produces phytoalexins in response to the galls and these enhance decay resistance. To investigate these possibili­ ties, we have extracted, with various solvents, the tissue of these black galls and of aspen trees bearing the galls. The extracts were then subjected to bioassay (against P. trem­ ulae) guided separation. Initial experiments were carried out on sawdust from the black galls. Separate portions of the sawdust were extracted with ethyl acetate, dichloromethane (DCM), methanol, and water. The ethyl acetate and dichloromethane extracts were separated into acidic and non acidic materials (NaHC03 extraction). Preliminary bioassays showed that these acidic extracts showed fungi­ cidal activity against P. tremulae. The components present in the acidic fractions were separated, identified, and sub­ jected to bioassay. The first objective was to identify the antifungal agent, which proved to be benzoic acid (BA), the second to monitor the level of this substance in the various wood substrates.

Materials and methods General methods

Gas chromatography was performed on a Varian 3700 gas chromatograph using a 50-m PONA(HP l ) column, 0.21 mm X 0.5 f l m film, flow rate at 55 psi of 0.6 mLimin. Mass spectra were determined on an AEI-50 mass spec­ trometer.3 Nuclear magnetic resonance (NMR) spectra eH and l C) were obtained on Bruker AM-400 or Varian Unity 500 multinuclear spectrometers. Chemical shifts are referenced to residual hydrogen (7 .26 ppm) or carbon (77.0 ppm) absorption of CDCI3• Flash chromatography was performed on silica gel 230-400 mesh, General Intermediates of Canada. Analytical thin layer chromo­ tography (TLC) was carried out on E. Merck precoated aluminum sheets of Si gel 60 F 254. FT-IR spectra were recorded on a Nicolet 7199 FTIR interferometer. All sol­ vents were distilled prior to use. Aspen and black galls

Aspen trees, some with black galls, some with heartwood rot caused by P. tremulae, and some healthy trees, were collected from the same clone near Blue Ridge, Alta., in July 1993. Black galls were excised with a chainsaw. Clean wood from a gall-bearing tree was excised 25 cm above the gall. Clean wood from a P. tremulae infected tree was removed from above the decayed area. The wood was chipped the following day and chips that were not used immediately were frozen. The chips which were used for extraction were ground to sawdust and air-dried to con­ stant weight at room temperature. Other collections were made in September 1992 and April 1993. Extraction, separation, and identification of acidic components of black galls

Air-dried sawdust from black gall tissue (0.5 kg) was sus­ pended in ethyl acetate (2 L) and vigorously stirred for 2 h, then filtered. The filtered sawdust was recovered and the extraction repeated. The combined extracts were

Can. J. For. Res. Vol.

25, 1995

concentrated to about one-third volume, then extracted three times with saturated aqueous NaHC03 (500 mL). The combined bicarbonate extracts were acidified with 2 M HCl, then back extracted with ethyl acetate (two times, 1 L). Evaporation of the ethyl acetate provided the acidic metabolites (5.0 g). Evaporation of the original ethyl acetate solution (bicarbonate washed) provided nonacidic material (21.0 g). The acidic material was separated by column chro­ matography over silica gel (150 g), eluting with hexane, then mixtures of hexane - ethyl acetate, then ethyl acetate . The fractions obtained were subjected to TLC and further purified by flash chromatography and preparative TLC where necessary. Benzoic acid (0.8 g) was eluted first, followed by trans-cinnamic acid (0.05 g), naringenin (Fig. 1, 1; 30 mg), 7'-methyl-3-hydroxynaringenin (Fig. 1, 2; 15 mg), p-hydroxybenzoic acid (30 mg), p-hydroxycinnamic acid (0.3 g), aromadendrin (Fig. 1,3; 45 mg), and taxifolin (Fig. 1, 4; 12 mg). Benzoic acid, trans-cinnamic acid, p-hydroxybenzoic acid, and p-hydroxycinnamic acid were identified by comparison (mass spectra, infrared (lR) spectra, lH NMR spectra) with authentic samples. Naringenin (Fig. 1, 1), aromadendrin (Fig. 1, 3), and taxifolin (Fig. 1, 4) were identified by comparison with published IR, NMR, and mass spectral data (see Harborne and Mabry 1982 for leading refer­ ences). The 7'-methyl-3-hydroxynaringenin (2) was iden­ tified by comparison with published IR and NMR data (Herz et al. 1972). Bioassay methods

Bioassays of crude extracts and chromatographic fractions against P. tremulae were carried out on carrot agar (pH 5.2) diffusion plates as previously described (Chakravarty and Hiratsuka 1992). In this case, 10 mg of sample was sus­ pended in 10 mL of distilled water and 1 mL of the sam­ ple was added to diffusion wells on 7 -day-old cultures of P. tremulae. After 4 weeks incubation at 22"C in the dark, the zone of inhibition around the wells was measured. The bioassays on pure BA were carried out by incorporation of BA (100 ppm, 300 ppm, 500 ppm, 700 ppm, 900 ppm, 1000 ppm) into the agar at the time of preparation. The 100 and 300 ppm BA level showed no inhibition of growth of P. tremulae relative to controls, while the 900 and 1000 ppm level showed complete inhibition. The 500 and 700 ppm levels showed 25-30% inhibition. Extraction with dichloromethane

The air-dried sawdust (50 g) was suspended in dichloromethane (400 mL) and stirred for 2 h, then fil­ tered. The filtered sawd�ust was recovered and the extrac­ tion repeated twice more. The combined DCM filtrates were dried (MgS04) and concentrated to about one-third volume, then extracted three times with saturated aque­ ous NaHC03 (100 mL). The combined bicarbonate extracts were acidified (slowly) with 2 M HCl, then back extracted with DCM (three times, equal volumes). Evaporation of the acid extract provided the crude acids (0.43 g from black gall tissue, 0.2 g from clean wood of gall tree, 0.075 g from clean wood of nongalled healthy tree). Isolation of the BA by silica gel chromatography as described above

Pausler et al. gave 77 mg from the gall tissue,2 mg from the clean wood of the gall-bearing tree, and