Characterisation of the pitch canker fungus, Fusarium circinatum, from Mexico

, Department of Microbiology and Plant Pathology, Forestry and Agricultural Biotechnology Institute (FABI), University of Pretoria , Pretoria 0002, South Africa t Present address: Onderstepoort Veterinary Institute, Private Bag X5, Onderstepoort 0/10, South Africa ' Department of Plant Pathology, University of California, Davis, California 95616, USA * Corresponding author, e-mail: teresa.coutinho@fabi.up.ac.za


Introduction Pitch canker. caused by Fusarium circinatum Nirenberg and
O'Donnell (=F subglutinans (Wollenw. and Re inking) Nelson et al. f. sp . pini Correll et al.), was first reported in the southeastern United States (Hepting and Roth 1946) . F eireinatum is now found throughout this region where it has caused significant losses on a wide variety of pine species. This led to the suggestion that pitch canker is endemic to the area . More re cently, pitch ca nker was identified and reported in California (McCain et al. 1987), predominately on Pinus radiata planted in landscape settings (Correll ef al. 1991). Since 1992, it has been recognised as a threat to native P radiata stands in Catifornia (Storer et al. 1994(Storer et al. , 1998. F circinatum is at so found in Japan (Muramoto et al. 1988, Kobayashi andKawabe 1992) and South Africa (Vitioen et al. 1994(Vitioen et al. , 1997 In South Africa, th e fungus was reported from forestry nurseries where it has resulted in serious losses of P patufa (Vitioen et al. 1994), P ellioNii, P greggii and P radiata seedlings. Stem cankers on targer trees such as those found in the United States (Hepting andRoth 1946, Dwinell et al. 1977) have not been seen (Wingfietd ef al. 1999). Pitch canker has been reported in Mexico on a variely of native pin e speci es (Santos andTovar 1991 , Guerra-Santos 1999) and this is thought to be the origin of the pathogen , for areas such as South Africa where Mexican pines are commonly propagated.
from cones from apparently healthy trees. Morphological characteristics of the pitch canker fungus and isolates of F. subglutinans from other hosts are very similar. Therefore, pathogenicity tests, sexual compatibility studies am1 histone H3-RFLPs were used to characterise isolates. Isolates collected from Pinus spp. from Mexico were identified as F. circinatum. In this study we have thus confirmed that F. circinatum occurs on pines in Mexico and that the affected trees can be asymptomatic.
Fusarium circinatum has been isolated from stems and branches of trees (Hepting andRoth 1946, Dwinell et al. 1977), root collars of pine seedlings (Barnard andBlakeslee 1980 , Vi tioen e/ al. 1994), fema te strobiti, mature cones and seeds (Dwinell et al. 1977, Miller and Bramlett 1978, Barrows-Broaddus 1986, 1987, Storer ef al. 1998, Dwinell 1999. Recently. Storer et al. (1998) isotated F circinatum from pine seedlings originating from seeds collected from cones on diseased as well as asym ptomatic branches. Those authors hypothesised that F. circinatum ca n be carried inside seeds and may remain dormant until germina tion, which increases the possi bi lity of seedling infections. The implication of seed transmission is serious, since current treatments may be ineffective in elim in ating the pathogen . This woutd increase the possibilily of introducing th e pathogen into uninfested areas (Barrows-Broaddus and , Storer et al. 1998. Fusarium subglutinans sensu lata includes species occurring on a wide variety of hosts, including pineapple, maize, mango, pine and sugarcane (Booth 1971). Correll et al. (199 1) distinguished pine and non-pine F subglutinans isolates based on pathogenicity to pines. Those authors proposed that F subglutinans from pine shou td be designated as F subglutinans f. sp. pini based on its exctusive pathogenicity to pine trees (Correll et al. 1991). Restriction frag-ment pattern s of the mtONA and random amplified polymorphic DNA (RAPO) also indicated that pine isolates differed from non-pine isolates (Correll el al. 1992, Viljoen et al. 1997 .Sexual compatibility among isolates ca using pitch canker on pines confirmed that this group corresponded to a distinct biological species (Viljoen el a/. 1997). O'Donnell et a/. (1998) showed pine isolates to be phylogenetically distinct and Nirenberg and O'Donnell (1998) (Steenkamp el al. 1999). Sexual compatibility as well as histone H3-RFLPs can, th erefore , be used to separate the eight different mating populations (biological species), designated by the letters A to H, in this complex. Heterothallic F. circinatum isolates reside in mating population H of the G, fujikuroi complex and tester strains representing opposite mating types have been selected and designated (Coutinho et al. 1995, Britz el al. 1998. Sexual compatibility of field isolates with tester strains of mating population H (M RC 6213 and MRC 7488) provide a firm basis for the identification of field isolates as F circinatum.
In this stUdy, isolations from pine trees in Mexico showing typical canker symptoms we re made. The possibl e association of F. circinatum with asymptomati c cones was also investiga ted. The identity of these isolates was verified using morphology (Nelson el a/. 1983, Nirenberg and as well as pathogenicity, sexual compatibility and histone H3-RFLP comparisons.

Isolates, cultural and morphological characteristics
Fusarium circinatum strains (MRC 7568-7587) were collected in Laguna Atezca and Hidalgo, Mexico, from ca nkers occurring on branches of native stands of P. palula and P. greggii (Table   1). Strains MRC 7568-7569 were isolated from apparently healthy cones collected from asymptomatic P. For isolations, pine cones were immersed in 70% ethanol for 2min. Small pieces (approxima tely 5mm2) of the asymptomatic cone scale tissue and wood pieces (2mm2) from the infected tree branches were removed and plated on 2% malt extract agar (MEA). Fungi were allowed to grow for seven days at room temperature. Small agar pieces (approximately 5mm' ) from the edges of th e colonies were transferred to 90mm plastic Petri dishes containing carnation leaf agar (CLA) (Fisher el al. 1982). Cultures were incubated at 23"C under near-ultraviolet and cool-white light with a 12h photoperio d to stimulate culture and conidium development. Isolates were identified and morphological features we re compared with F. circinalum tester stra ins (MRC 6213 and MRC 7488) after 10 to 14 days, using a light microscope. Diagnostic characteristics used were those specified by Nelson el al. (1983) and Nirenberg and O'Donnell (1998) (1999). PCR reactions were performed using pri mers (H3-1a and H3-1 b) for amplification of the histone H3 gene PCR product as described by Glass and Donaldson (1995). The same cond itions as those described by  were Digests were performed using this restriction enzyme in a total reaction volume of 20~1 containing 5U of the enzyme. Sodium ch loride was added to the reactions with th e enzyme Ode 1 to a final concentration of 1 OOmM . All digestion reactions were incuba ted at 3rC for 7h.
Restriction fragments were separated using agarose gels (Promega Corporation, Madison, Wisconsin, USA) in the presence of ethidium bromide (0.1 ~g/ml). RFLP fragmen ts were electrophoresed on 3% (w/v) agarose gels a nd visualised using a n ul trav iolet transil luminator (Ultra-Violet Product) . The visualised RFLP fragme nts we re ph otographed using a gel documentation system (Microsoft Corporation) and evaluated using the methods described by Steenkamp et a/. (1999).

Sexual compatibility
Crosses to determine sexual compatibility were made on carrot agar as described by Klittich and Leslie (1988), except that 300g fres h ca rrots were used rath er tha n the recommended 400g. All crosses were also done on V8-agar (325 ml canned V8-juice per liter and 2% agar, pH 5.8-6.2 crossed with ea ch other in all possible combinations . The isolates in this study were also crossed against themselves as a negative control, i.e. where no mature pe rithecia should be produced. Reciprocal crosses, where the isolates corresponded to the male and female parents we re reve rsed , were done for all crosses. All the cross es recorded as positive were repeated at least once. Crosses were exam ined week ly and scored as positive when ascospores were observed , either by their exudation from perithecia or after cru shi ng these structu res. The viability of the ascospores was determined by streaking a portion of the ascospore cirrhus onto the surface of 2% water agar plates and estimating the percentage germination after 24h.

Pathogenicity tests
Due to quarantine constrai nts in South Africa, pathogeniCity tests of F. circinatum isolates from Mexico were conducted in greenhouse facilities of the Department of Plant Pathology,

Universi ty of California. Tests to confirm pathogenicity were
performed at approximately 25' C during the day and 18' C at night with a 12 h day/night cycle. Tests were performed on P. radiata seedlings, 3-4 years of age using 19 F. circinatum isolates collected in Mexico (Table 1), as well as, two isolates of F. circinatum (FSP 24 and FSP 34), known to be pathogenic to pines in Californ ia. All F circinatum isolates were grown on potato dextrose agar (PDA) at 25' C for 7-10 days.

Inoculations were performed by making a small wound in the
seedling stems and plaCing a spore suspension (approximately 500 spores in distilled water) into each wou nd. Each isolate was inocula ted into two P. radiata seedlings. The lesion leng ths under the bark of the inoculated P radiata plants were measured 4"1 days after inoculation.

Isolates, cultural and morphological characteristics
Isolates obtained from diseased pine branches, cankers and asymptoma tic co nes were identified as F subglutinans based on morphological characteristics de scribed by Nelson et al. (1983). These isolates could also be identified as F. circ;naturn based on the characteristics proposed by Nirenberg and O'Donnell (1998). Bra nched an d proliferating conidiophares were observed and the polyphialides had 2-5 conidiogenous openings (Figure 1 c) . Sterile coiled hyphae ( Figure  1d) and lunate macroconidia (Figure 1e), reported by Nirenberg and O'Donnell (1998) to distinguish F. circinaturn from similar Fusarium spp. in Liseo/a and related sec tion, were observed.

RFLPs of histone H3 gene
The PCR products obtain ed using th e primers H3-1 a and H3-1 b we re approximately 500 base pairs (bp) in size. None of the PCR products from isolates belonging to the E mating population (MRC 6483 and MRC 6512) of the G. fujikuro complex and the F. subg/utinans isolate from maize (MRC 1077) were cut by Ode 1, whereas the F. subglutinans iso-

Sexual compatibility
Using sexual compatibility tests, we were able to verify that isolates from pine in Mexico belong to mating population H and, therefore, F. circinatum (Figure 1). Perithecia with exuding ascospores (Figure la 7568,7569,7572,7591,7592,7593,7595,7597,7599 and 7600 produced perithecia with viable ascospores when crossed with the mating population H tester strain, MRC 6213. These isolates only produced fertile crosses when MRC 6213 was used as the female parent and are thus female-sterile. No mature perithecia with viable ascospores resulted from crosses amongst F. circinatum isolates from Mexico.
All the fertile crosses recorded in this study were repeated and identical results were obtained in at least two different tests. The same results were also obtained on both carrot this study varied between 85-98%. None of the isolates in this study produced perithecia when crossed wilh themselves as negative controls.

Palhogenicity tesls
The 21 selected F. circinatum isolates gave lesions aher 41 days that varied in length between 10 to 90mm. These isolates were, thus, pathogenic to P. radiala seedlings in the glasshouse. The pathogenicity of the F. circinatum isolates from California (FSP 24 and FSP 34) was consistent with that reported previously (Correll et al. 1991, Gordon et af. 1996).

Discussion
In this study, we were able to identify F. circinatum from branches, cankers and asymptomatic cones trom Mexico, based on a wide range of criteria. The association of the pitch canker fungus with asymptomatic cones demonstrates the possibility of spread on apparently healthy seeds such as reported by Storer et al. (1998). F. circinatum isolates considered in this study were collected from P. patula, P. greggii, P. teocote and P. leiophyl/a in Mexico. P. patufa, P. elliottii, and P. radiala are the most important, commercially planted species in South Africa (Hinze 1993). More recently, P. greggii has also become important to the forestry industry in South Africa (Malan 1994). The isolation of F. circinalum from P. patula and P. greggii in native stands in Mexico, indicates that the pitch canker fungus could have been introduced into South Africa from Mexico. This would most likely have occurred through seed importation. The finding emphasise the importance of screening seed for F. circinatum infection before it is exported. Seed treatment with fungicides would also reduce the chances of new introductions occurring, although it might not effectively eliminate intemal infections (Storer el a/. 1998(Storer el a/. , 1999. Small seed lots, where the risk can be reduced through propagation under controlled conditions, may be an acceptable practice. The importation of large collections of seed for commercial planting can not be managed effectively, and shou ld be avoided. In this study, only a small number (less than 28%) of F. cirdnatum isolates from Mexico were able to cross with the single F. circinatum tes ter (MRC 6213), from South Africa. It is possible that the isolates from Mexico that did not cross with MRC 6213 all belonged to the same mating type as this South African tester. Alternatively, low fertility or sterility (Perkins 1994) cou ld explain why none of these isolates cross with the tester strain of the opposite mating type. Female-sterility (Leslie and Klein 1996) cou ld also have contributed to the lack of sexual compatibility seen between Mexican F circinatum isolates. The low level of sexual compatibility might also suggest that the population of F. circinatum in Mexico is evolving towards an asexual life history (Leslie and Klein 1996). This is ind icated by the hig h percentage of female-sterile isolates found in the sexual compatibility study. Despite the low level of sexual compatibility, sexual crosses confi rm ed that some of th e Mexican isolates belonged to mating population H of the G. fujikuroi complex.  showed that F. subglutinans isolates from vario us plant hosts can be distinguished from one another with histone H3-RFLPs. Those authors concl uded that this technique could be used for routine identification of F. circinatum. In this study, the histone H3-RFLP tec hnique was crit ical fo r positive identification of F d rdnatum. This was particularly due to the sometimes inconclusive and time-con suming pathogenicity tests and the low fertility among the F. cirdnatum isolates from Mexico.