Fungal pathogens of Miconia calvescens (Melastomataceae) from ...

3 downloads 205 Views 1MB Size Report
In Queensland. M. calvescens was declared a noxious weed in .... RIO DE JANEIRO: Sa˜o Roma˜o, road Lumiar–. Casimiro de Abreu, RJ, 24 Feb 1998, VIC ...
Mycologia, 99(1), 2007, pp. 99–111. # 2007 by The Mycological Society of America, Lawrence, KS 66044-8897

Fungal pathogens of Miconia calvescens (Melastomataceae) from Brazil, with reference to classical biological control Claudine D.S. Seixas Robert W. Barreto1

long), which are green adaxially and purple abaxially. We refer to this form as the highland biotype (Meyer 1996). The lowland biotype found predominantly in eastern South America and occasionally in Central America has smaller leaves with green abaxial surfaces. Because of the attractive foliage of the highland biotype it was introduced as an ornamental in many regions of the world. In some regions this proved disastrous (Gagne´ et al 1992, Meyer and Florence 1996, Meyer 1996, Baruch et al 2000). This biotype was introduced into Tahiti in 1937 where it became naturalized and slowly invaded the native forests. It now covers two-thirds of the island and forms monotypic stands in many areas (Meyer and Florence 1996). It is already present in Moorea, Raiatea and Thaa (Meyer and Florence 1996, Meyer and Malet 1997). Miconia is now regarded as a worst case example of the effect of an invasive weed on oceanic island biodiversity (Meyer 1996). An estimated 70–100 native species, including 40–50 endemics, are directly threatened by M. calvescens in French Polynesia (Meyer and Florence 1996). In Queensland M. calvescens was declared a noxious weed in May 1997, but its cultivation and commercialization still are allowed in other Australian states, a dangerous situation because the plant has all the attributes for becoming a serious weed in that country (Csurhes 1997, Csurhes and Edwards 1998). In Hawaii M. calvescens was introduced in the 1960s and since 1992 has been included in the list of noxious invasive weeds (Medeiros et al 1997). Until now the invasions in Hawaii fortunately have not become as severe as those of Tahiti (Gagne´ et al 1992), in part because of an aggressive eradication and containment program. The plant nevertheless is present on four of the main Hawaiian Islands, Hawaii, Oahu, Maui and Kauai (Loope 1997). A search for fungal pathogens that could be used as classical biocontrol agents of M. calvescens began in Jun 1995 in Brazil and was extended later to Costa Rica, the Dominican Republic and Ecuador. A description of those pathogens collected on this host and its relatives in Brazil is given below together with a discussion of their biocontrol potential based on field observations.

Departamento de Fitopatologia, Universidade Federal de Vic¸osa, MG 36571-000, Brazil

Eloise Killgore Hawaii Department of Agriculture, Division of Plant Industry, Biological Control Section, Honolulu, Hawaii 96814

Abstract: A survey of fungal pathogens of Miconia calvescens was carried out in Brazil aimed at finding potential classical biocontrol agents for management of this invasive alien weed in Hawaii. Coccodiella miconiae, Glomerella cingulata (5Colletotrichum gloeosporioides f. sp. miconiae) and the new species Guignardia miconiae and Korunomyces prostratus were found associated with foliar diseases and are described herein. Two previously undescribed spore stages of Coccodiella miconiae also were obtained allowing a complete description of this species. Pseudocercospora tamonae associated with leaf spots of other species of Miconia also was collected and also was proven to be pathogenic to M. calvescens. Key words: biological control, classical biocontrol, Hawaii, Melastomataceae, Tahiti INTRODUCTION

Miconia calvescens DC. (miconia or ‘‘cancer vert’’) belongs to the largest genus of the Melastomataceae with about 1000 species. It is a neotropical genus concentrated in the Andes (Judd and Skean 1991, Renner 1993). Few species in this genus are of practical importance but M. calvescens has become a notable exception. Once a botanical curiosity miconia is now the most devastating plant invader in Tahiti (Meyer 1996) and Hawaii (Gagne´ et al 1992). Although widely distributed from Mexico to Brazil and sometimes common locally, often in disturbed forest habitats, it never forms dense populations in its native range. In some regions in South and Central America, especially on the eastern side of the mountain ranges dividing the continents and extending from southern Mexico to Ecuador, a form of M. calvescens has large leaves (up to 1 m

MATERIALS AND METHODS

An initial survey of some Brazilian herbaria was made to locate known sites of M. calvescens in Brazil for the collecting trips. Localities in the states of Minas Gerais,

Accepted for publication 11 September 2006. 1 Corresponding author. E-mail: [email protected]

99

100

MYCOLOGIA

Espı´rito Santo, Bahia, Rio de Janeiro and Sa˜o Paulo were selected and surveyed. Ad hoc collections also were made in the states of Amazonas and Mato Grosso. In addition other species of Miconia with disease symptoms were collected when present. All specimens examined in this work were collected by R.W. Barreto. Dried specimens of diseased plants were prepared with a plant press and isolates of nonbiotrophs were obtained by direct or indirect isolation on V8 juice agar plates or VBA (vegetable broth-agar according to Pereira et al 2003), transferred to PCA (potato-carrot agar) slants and maintained at 5 C. The cultures were shipped to the Hawaii Department of Agriculture (HDOA) Plant Pathology Quarantine Facility in Honolulu for further testing (USDA permit No. 954140). Samples of biotrophic fungi were preserved on bare-rooted, living miconia plants and were dispatched to Hawaii. Identifications were made with standard keys for the genera and species. Only those that were associated with damaging disease symptoms in the field and therefore appeared to have potential as biological control agents are considered here. Semipermanent slides were prepared by scraping fungal material from diseased tissues, hand sectioning selected fragments containing fungal structures or with the help of a freezing microtome (Leitz, Kriomat) and mounting in water or unstained lactophenol. Observations, measurements, photographs and drawings were made with an Olympus BX50 light microscope, fitted with a camera and drawing tube. Leaf segments from fresh diseased leaves of M. calvescens were prepared for scanning electron microscope (SEM) observation by immersion in 2% glutaraldehyde and 2% paraformaldehyde prepared in 0.05 M C2H6AsO2Na buffer (pH 7.1) and left 5 d in a refrigerator at 5 C. Samples were washed twice with cacodylate buffer for 15 min and dehydrated by successive transfer in a graded alcohol solutions series (30, 50, 60, 70, 80, 95, 100%) and left 15 min in each solution. Samples were dried in a critical point dryer (BalzersH model CPD 020) with CO2 as transition fluid. After drying the samples were coated with gold (20 nm thick) in a sputter coater (BalzersH model FDU 010) and examined with a LEO-VP 1430 SEM. RESULTS

Coccodiella miconiae (Duby) Hino & Katuamoto, in Katuamoto, K.J Jap Bot 43:282, 1968. (FIGS. 1, 2, 3) ; Sphaeria miconiae Duby in Mem. soc. phys. et hist. nat. Gene`ve 7:405, 1835. ; Physalospora miconiae (Duby) Sacc. Syll Fung 1:447, 1882. ; Botryosphaeria miconiae (Duby) Ho¨hnel, Sitz-ber. Akad Wien 118:836, 1909. ; Phyllachora miconiae (Duby) Sacc. Ann Myc 11:547, 1913. ; Bagnisiopsis miconiae (Duby) Petrak. Hedwigia 68:275, 1928. ; Coccostroma miconiae (Duby) v. Arx & Mu¨ller, Die Gattungen der amerosporen Pyrenomyceten. Beitr. Kryptog fl Schw 11(1):263, 1954.

Disease (yellow pustule).—Lesions on living leaves:

adaxially initially punctiform, chlorotic becoming pale brown centrally, often raised and convex (pustule-like), sometimes concave; older lesions with a narrow, well defined chlorotic halo surrounded by diffuse chlorotic area becoming dark brown to black centrally, circular, up to 5 mm diam, coalescing in some areas of the leaves; abaxially, stromata initially minute and pale brown, becoming black and shiny, set inside concavities on the leaf laminae, sometimes surrounded by narrow chlorotic halo easily seen with naked eye, up to 3 mm diam; severe infection causes foliar deformation and widespread, chlorosis. Morphology.—Internal mycelium intra-intercellular, branched, septate, hyaline, 2–5 mm diam. External mycelium absent. Stromata formed abaxially, erumpent, subspherical to pulvinate, 189–1500 mm wide, to 190–473 mm high, constricted at base, 167–584 mm wide at attachment point, isolated or aggregated, initially pale brown and having one to several spermogonia on surface, becoming black and having several perithecial locules, walls composed of dark textura angularis and internal tissue hyaline to pale brown. Ascomata perithecial, embedded in stromata, spherical to subspherical, sometimes having a distorted shape, 123–218 mm diam, walls of hyaline textura angularis, approximately 12–19 mm thick (but not well differentiated from stromatal tissue). Dehiscence ostiolate, one ostiole per perithecium, 28.5–71 mm diam. Hamathecium of well developed and abundant septate, unbranched, hyaline paraphyses (,75 mm long 3 1 mm wide) and periphyses 20– 44 3 1 mm. Asci unitunicate, attached to lower part of perithecia, cylindrical, 71–100 3 7–10 mm, thickwalled, apex round to subtruncate, stalked, apical ring indistinct, 8-spored. Ascospores uniseriate, ellipsoid to subspherical, 7–12 3 6–8 mm, aseptate, eguttulate, hyaline becoming brown with age, smooth and relatively thick-walled. Spermogonia formed early on the surface of stromata, cupulate, 42.5–92.5 mm wide 3 38–83 mm high, containing receptive hyphae and abundant mucilaginous masses of tear-shaped spermatia, 2–4 3 1–1.5 mm, hyaline, smooth-walled; darkening and becoming sterile with age appearing as black, horn-like projections on stromata. Hemidothis (anamorph).— Conidiomata produced abaxially on leaves, similarly to ascomata, stromatic, multilocular, erumpent, single or forming small groups, black, shiny, having many digitate projections, each supporting a drop of milky mucilaginous conidial mass, up to 1 mm diam; walls of dark-brown textura angularis, 2–5 cells, 8–26 mm thick; locules spherical, ellipsoidal or irregular, 35–94 mm diam, arising at different levels within stromata; with long, fine, septate, hyaline paraphyses that emerge through

SEIXAS ET AL: MYCOBIOTA OF MICONIA CALVESCENS

101

FIG. 1. Coccodiella miconiae (Vic¸osa, MG); a. symptoms of yellow pustule disease on lower (left) and upper leaf surfaces of Miconia calvescens; b. close-up of abaxial side of severely infected leaf, showing the black depressions; c. close-up of adaxial side of the same leaf, showing the typical symptoms of raised spots surrounded by chlorotic haloes reminiscent of a rust fungus.

ostiole. Dehiscence ostiolate, one per locule, digitate, 19–84 mm diam. Conidiophores arising from internal walls of locules, cylindrical, tapering toward apices, straight or flexuose, 10–41 3 1–2.5 mm, 1–2-septate, branched, hyaline, smooth. Conidiogenous cells terminal, enteroblastic, cylindrical, tapering toward apices, 6.5–24.5 3 1–2 mm, hyaline, smooth. Conidiogenous loci minute, 0.5–1 mm. Conidia mucilaginous, enteroblastic, straight or curved, fusiform to falcate, 3.5–8 3 1–2 mm, apex and base rounded, aseptate (occasionally septate), guttulate, hyaline, smooth. Specimens examined. BRAZIL. MINAS GERAIS: Vic¸osa, 16 Mar 1998, VIC 19303; road Lajinha–Mutum, MG, 16 Dec 1996, VIC 19298; road Dionı´sio–Timo´teo, MG, 30 Aug 1996, VIC 19291. RIO DE JANEIRO: Sa˜o Roma˜o, road Lumiar– Casimiro de Abreu, RJ, 24 Feb 1998, VIC 19305; Sana, 24 Feb 1998, VIC 19306; road Glice´rio–Vila do Grama, 24 Feb 1998 VIC 19307; road Frade–Glice´rio, RJ, 24 Feb 1998, VIC 19308; Estrada da Grota Funda, Recreio dos Bandeirantes, Rio de Janeiro, 27 Dec 1995, VIC 19286; Boca do Mato,

Cachoeiras do Macacu´, 5 Feb 1996, VIC 19288; road Rio– Petro´polis, Xere´m, 24 Mar 1996, VIC 19290; road Rio–Sa˜o Paulo, between Barra Mansa and Arrozal, 20 Sep 1996, VIC 19292; road Lı´dice–Angra dos Reis, 20 Sep 1996; VIC 19293; Bosque da Barra, Barra da Tijuca, Rio de Janeiro, RJ, 30 Sep 1996, VIC 19294; Alto da Boa Vista, Rio de Janeiro, 30 Sep 1996, VIC 19295. BAHIA: Floresta Azul, 21 Nov 1996, VIC 19296; Reserva Biolo´gica de Una, Una, 22 Nov 1996, VIC 19297; road BR 101, km 483, between Itabuna and Ubaitaba, 19 Jan 1997, VIC 19299; road Ubaitaba–Marau´, BA, 19 Jan 1997, VIC 19300.

Glomerella cingulata (Stonem.) Spaulding and Schrenk. Sci Ser 2(17):751, 1903. Teleomorph: present in old lesions on leaves and formed in some aging cultures. Anamorph: Colletotrichum gloeosporioides (Penz.) Sacc. Atti R. Ist. Ven Sci Lett Art 6, 2, 670, 1884. Disease (anthracnose): Lesions on living leaf laminae

102

MYCOLOGIA

FIG. 2. Coccodiella miconiae (Vic¸osa, MG): a. SEM of a young spermogonial stroma erupting through abaxial leaf surface; b. section through a young spermogonial stroma; c. Close-up of a spermogonium showing a cloud of spermatia; d. close-up of a spermogonium showing receptive hyphae; e. SEM of a mature perithecial stroma (note spermogonial beaks, often confused with setae); f. group of stromata on abaxial leaf surface; g. mature stroma with perithecia and old sterile spermogonia; h.

SEIXAS ET AL: MYCOBIOTA OF MICONIA CALVESCENS and along leaf margins associated with blight-like symptoms, initiating as minute, necrotic, circular punctations becoming larger and roughly circular in the central part of the leaves, up to 3 cm diam and sometimes elliptical when growing on leaf veins, pale brown in center, periphery dark brown abaxially and gray-brown to totally gray adaxially, sometimes having a diffuse chlorotic halo; lesions sometimes coalescing leading to extensive leaf necrosis; necrotic areas easily torn and tending to fall, sometimes causing loss of parts of lamina leaving only a leaf vein skeleton. In culture: Colonies relatively fast growing (54– 86 mm diam/11 d on PDA and VBA), mycelial growth mostly within medium, having a central area of white to grayish, sparse, woolly aerial mycelium, where sporulation is concentrated in orange mucilaginous masses of conidia; reverse grayish-white with no pigmentation of medium or evidence of diurnal zonation. Ascocarp perithecia, solitary or aggregated, spherical to subspherical, 150–309 mm diam, walls composed of brown textura angularis, mostly 4– 5 cells thick, 8.5–25 mm, smooth. Dehiscence ostiolate, single, central, circular, 8.5–25.5 mm diam, lined with periphyses. Asci unitunicate, cylindrical to clavate, 44–66 3 9–11.5 mm, rounded or slightly flattened at apex, paraphysate, 8-spored. Ascospores straight to fusiform to slightly curved, allantoid, 12.5– 17.5 3 4.5–6 mm, aseptate, guttulate, hyaline, smooth. Conidiomata acervular, formed only on host leaves abaxially on laminae and veins, setose. Setae straight, cylindrical, tapering toward pointed apex, 62.5– 120 mm long, 2-septate, brown, smooth. Conidia straight, cylindrical, apices obtuse, 9–22.5 3 3– 4.5 mm, aseptate becoming 1-septate at germination, guttulate, hyaline, smooth. Appresoria variable in shape, 8–25 3 5–17.5 mm, aseptate or sometimes septate, thick walled, dark-brown, smooth. Specimens examined: BRAZIL. RIO DE JANEIRO: Belvedere, Itaguaı´, 27 Dec 1995, VIC 19287; Sana, 24 Feb 1998, VIC 19306; road Glice´rio–Vila do Grama, 24 Feb 1998, VIC 19307; Boca do Mato, Cachoeiras do Macacu´, 5 Feb, 1996, VIC 19288; Estrada da Grota Funda, Rio de Janeiro, 27 Dec 1995, VIC 19286; Mazomba, Itaguaı´, 22 Dec 1995, VIC 19285; road Rio–Petro´polis, Xere´m, 24 Mar 1996, VIC 19290; road Lı´dice–Angra dos Reis, 20 Sept 1996, VIC 19293; Sa˜o Joaquim, Piraı´, 5 Mar 1997, VIC 19302; Bosque da Barra, Barra da Tijuca, 30 Sep 1996, VIC 19294; Alto da Boa Vista, Rio de Janeiro, RJ, 30 Sep 1996, VIC 19295; road Rio–Sa˜o Paulo, between Barra Mansa and Arrozal, 20 Sep 1996 VIC 19292; road Frade–Glice´rio, RJ, 24 Feb 1998, VIC 19308. MINAS GERAIS: road Leopoldina–Cataguases, 2

103

May 1995, VIC 19284; Cristais, Vic¸osa, 13 Mar 1996, VIC 19289; road Dionı´sio–Timo´teo, 30 Aug 1996, VIC 19291. BAHIA: Floresta Azul, 21 Nov 1996, VIC 19296; road Ubaitaba–Marau´, 19 Jan 1997, VIC 19300. ESPIRITO SANTO: margins of Rio Jucu´, ES, 17 Jan 1996, VIC 19301. ˜ O PAULO: Ipeu´na, road to Serra do Cantagalo, 26 Nov SA 1998, VIC 22207.

Guignardia miconiae C.D.S. Seixas & R.W. Barreto sp.nov. ( FIG. 4) Ab Guignardia citricarpa Kiely; ascomata epiphylla, 90–285 mm diam; asci 26.5–102.0 3 24.0–31.5 mm; ascosporae ovoideus, 14.0–20.0 3 10.0–11.5 mm; appendice indistinctus; status anamorphicus Leptodothiorella differens. Etymology: on leaves of Miconia calvescens Disease (punctiform black spot): Lesions on living leaves, limited to black punctations in subcircular aggregates within green tissue when young, more visible abaxially, colonized tissue becomes discolored in older lesions with a central to subcentral gray necrotic area and a dark brown to black periphery abaxially and evenly grayish-brown to gray adaxially, 3–12 mm diam; irregularly distributed on lamina and occasionally coalescing and leading to formation of cracks in the tissue. Morphology: Internal mycelium interintracellular, branched, septate, pale brown. External mycelium absent. Ascomata stromatic, epiphyllous, semi-immersed, scattered, globose to subglobose, 90– 285 mm diam, walls composed of dark brown textura angularis, 14–74 mm thick, much thicker and melanized on upper part, smooth. Dehiscence ostiolate, central, one per ascoma, 33.5–61 mm. Interthecial filaments absent in mature ascomata. Asci bitunicate, clavate, 90–102 3 24–31.5 mm, 8-spored, pedicellate. Ascospores inordinate, ovoid, 14–20 3 10–11.5 mm, aseptate, guttulate, hyaline, smooth, mucilaginous sheath tenuous rarely perceptible. Anamorphic/spermatial stage: Leptodothiorella sp. Conidiomata pycnidial, epiphyllous, semi-immerse, scattered, sphaerical to subsphaerical, 64–145 mm diam., wall 10.5–89 mm thick, brown, smooth. Dehiscence as described for ascomata. Conidiophores covering all internal wall of the pycnidia, narrowly lageniform, 7.5–25.5 3 1.5–5.0 mm, 0–2-septate, unbranched, hyaline, smooth. Conidiogenous cells integrated, enteroblastic, cylindrical to lageniform, 6– 13 3 1–4 mm. Conidiogenous loci one per cell,

r perithecium i. asci and paraphyses; j. SEM of a conidiomatal stroma; k. section through a conidiomatal stroma. Bars: a, e 5 100 mm; b, k, j 5 200 mm; 2c 5 20 mm; d, b, i 5 50 mm; f 5 0.5 mm; g 5 115 mm.

104

MYCOLOGIA

FIG. 3. Proposed life cycle of Coccodiella miconiae (VIC 19291): a. digitate conidioma with mucilaginous drops containing conidia; b. conidiophores and conidiogenous cells showing conidiogenesis; c. conidia; d. young stroma erupting through leaf epidermis showing two spermogonia; e. spermogonium with receptive hyphae and tear-shaped spermatia. f. mature stroma with perithecia and two spermogonia (small arrow, note that the spermogonium on the left is old and inactive whereas the

SEIXAS ET AL: MYCOBIOTA OF MICONIA CALVESCENS

105

unthickened with a minute collarette. Conidia mucilaginous, enteroblastic, straight, dumbbell shaped, 4.5–5.5 3 1.0–2.0 mm, aseptate, biguttulate, hyaline, smooth. ˜ O PAULO: Bosque MuHOLOTYPE: BRAZIL. SA ´ guas da Prata, 27 Nov 1998, VIC 22211. nicipal, A ˜ O PAULO: A ´ guas da PARATYPES: BRAZIL. SA Prata, 8 Jun 2001, VIC 22212. RIO DE JANEIRO: road Vila Abraa˜o–Dois Rios, Ilha Grande, Angra dos Reis, 13 Jan 2002, VIC 22237.

strongly radial superficial growth on CVA, white, cream to pale orange, reverse as for surface; on PDA many germinated propagules observed at or near the surface of medium, rarely along or at apices of aerial hyphae; sterile. HOLOTYPE: BRASIL. RIO DE JANEIRO: Angra dos Reis, Ilha Grande, 04 Jan 2000, VIC 22213. PARATYPE: BRASIL. RIO DE JANEIRO: Angra dos Reis, Ilha Grande, road from Vila Abhraa˜o to Dois Rios, 13 Jan 2002, VIC 22218.

Korunomyces prostratus C.D.S. Seixas & R.W. Barreto sp. nov. (FIG. 5) Ab Korunomyces terminaliae Hodges & Ferreira; propaguliphora prostrata, haud distinctus ab hyphae, 3–4 usque 5–8 m m (ad propagula base) diam, propagula elementis 4.0–10.0 mm diam, propagula elementis terminalibus 7–13 mm longum, differens. Etymology. reference to the predominant formation of prostrate propagulophores and propagules. Disease (leaf blight). Lesions necrotic, initially circular, grayish-brown in center with a brown periphery, becoming irregular with age with concentric dark-brown peripheral rings often resulting in a scale like pattern, with a yellowish halo, coalescing and leading to an extensive leaf-blight; older parts of lesions tend to crack leaving irregular holes in the leaves. Morphology. External mycelium amphigenous, branched, septate, initially hyaline becoming yellow or orange. Internal mycelium indistinct. Propagulophores often difficult to distinguish from ordinary hyphae, cylindrical, simple, length indeterminate, individual cells 11–27 3 3–4 mm diam, below propagules 5–8 mm, hyaline, smooth, point of rupture indistinct or absent. Propagules subglobose to irregular when mature, formed on apex of usually prostrate hyphae or occasionally on erect propagulophores, multicellular, formed of primary branches with an initial dichotomous branching pattern, becoming dendritic later, 69–273 3 64–272 mm, branch elements 4–10 mm diam, terminal elements 4–5 3 7–13 mm, initially hyaline becoming orange when mature, smooth. In culture. relatively fast growing (3.2–6.5 cm diam after 13 d on PCA and VBA); colonies of cottonywoolly aerial mycelium showing marked diurnal zonation on PDA, dark orange with a dark orange reverse; flattened aerial mycelium surrounded centrally by an area of sparse aerial mycelium and

Pseudocercospora tamoneae (Chupp) U. Braun & Castan ˜ eda, Crypto Bot 2/3:294 (1991) (FIG. 6) ; Cercospora tamoneae Chupp, A monograph of the fungus genus Cercospora, p 383, Ithaca: published by the author, 1954.

Disease (leaf-spot). Lesions on living leaves, irregular, vein delimited, brown surrounded by a diffuse, chlorotic halo, up to 13 3 7 mm. Morphology. Internal mycelium intra- and intercellular, 2 mm diam., branched, septate, pale brown, smooth. External mycelium hypophyllous, poorly developed, septate, pale brown, smooth. Stromata erumpent, initially subglobose, becoming cylindrical, 24–43.5 3 25–7 mm, composed of grayish-brown cells. Conidiophores amphigenous, either densely fasciculate (often more than 20 per fasciculum) or produced singly on external mycelium, cylindrical, straight to slightly sinuose, 15–57 3 2–4 mm, 0–4-septate, unbranched, pale brown, smooth. Conidiogenous cells terminal and intercalary, integrated, holoblastic, proliferating sympodially, cylindrical, straight to slightly sinuose, 1–2-geniculate, 5–19.5 3 3–4 mm, pale brown, smooth. Conidiogenous loci flattened, 2– 3 mm wide, unthickened, not darkened. Conidia dry, single, holoblastic, subcylindrical tapering toward the apices, often slightly curved but sometimes straight, 53–90 3 3–5 mm, apex rounded, base subtruncate, 2– 3 mm, 6–11-septate, scar unthickened and not darkened, guttulate, pale brown, smooth. Specimens examined. BRAZIL. RIO DE JANEIRO: on Miconia jucunda (DC.) Triana, Rio de Janeiro, Barra da Tijuca, Bosque da Barra, 30 Sep 1996, VIC 19294. MATO GROSSO: on Miconia sp., Chapada dos Guimara˜es, 11 Oct 1998, VIC 22214. DISCUSSION

The genus Coccodiella includes about 25 species of biotrophic fungi that are foliar pathogens of plants belonging to 10 families. Eleven species of Coccodiella

r spermogonium on the right is still active); g. asci and ascospores, (note central, immature ascus with hyaline ascospores). Bars b, c, e and g 5 20 mm: a, d and f 5 100 mm.

106

MYCOLOGIA

FIG. 4. Guignardia miconiae (VIC 22211): a. punctiform black spot symptoms on adaxial side of infected M. calvescens leaf; b. close-up of lesions on a leaf; c. external view of upper part of pseudothecia; d. pycnidium; e. conidiophores and conidia; f. pseudothecium; g. asci and ascospores. Bars: d and 4e 5 50 mm; g 5 10 mm.

SEIXAS ET AL: MYCOBIOTA OF MICONIA CALVESCENS

107

FIG. 5. Korunomyces prostratus (VIC 22213): a. causing foliage blight on M. calvescens; b. propagule primordium; c. young propagule; d. mature propagule. Bars 5 50 mm.

are parasitic on the Melastomataceae. Katumoto (1968) recognized that the name Coccodiella, proposed for the genus by Hara in 1911, had priority over other names such as Coccostroma and Bagnisiopsis used by other authors in later works. He then proposed a series of new combinations but studied only material of the type species Coccodiella arundinaria Hara. More recently this species was examined and redescribed by Cannon (1996). Miller and Burton (1943) studied several species of Bagnisiopsis (5Coccodiella) on the Melastomataceae. These authors observed the presence of spermogonia on the stromata in this genus for the first time but failed to observe the later development of these structures in aging stromata. They misinterpreted old spermogonia (and possibly also the rostrate conidiomata) as being ornamented stromata (treated as ‘‘setae-like processes’’ by these authors) and adopted their presence as important key characters. In fact examination of fresh material of the fungus collected on M. calvescens having stromata in several stages of development, showed clearly that such ‘‘setae-like processes’’ are either old and dried spermogonia or rostrate conidiomata (FIG. 2e, g, k). The presence of these ‘‘ornaments’’ therefore is likely to be dependent on the age or lifecycle stage of the material under examination and hence is inadequate for species separation. Also, although stating that ‘‘the dimensions of ascospores are more variable than in

most Ascomycetes,’’ these authors proposed the use of spore size as critical characters for species separation in the key given in their article. These aspects, together with the high proportion of species in the genus described on Miconia (nine out of the 26 species of Coccodiella having been described from Miconia), suggest that some of these taxa in fact might be conspecific. Revision of the Coccodiella on Melastomataceae is clearly needed but is outside the scope of the present work. The fungus collected during this fieldwork fits well into several overlapping species descriptions given for Coccodiella spp. on Miconia. We decided nevertheless that for the present it would be more appropriate to use the name C. miconiae because this was proposed for a similar Coccodiella found growing on leaves of M. calvescens in Brazil (Petrak 1928). Although C. miconiae is not new to science, it is a poorly known fungus (as are the majority of the species in this genus) and the description given above (and illustrated in FIG. 3) is the first complete description of the lifecycle of this fungus, including two spore stages (spermogonial and anamorphic) that previously were not described or illustrated for this species. It is likely that the anamorphic stage with its slimy conidial mass functions in short distance splash dispersal while the ascospores provide the propagules for long distance wind dispersal. Yellow pustule disease caused by C. miconiae is

108

MYCOLOGIA

FIG. 6. Pseudocercospora tamonae (VIC 19294): a. showing leaf spot symptoms on foliage of Miconia sp. b. conidia; c. conidiophores on external mycelium; d. Conidiophore fascicle. Bars 5 10 mm.

ubiquitous on miconia and occurs throughout the year. On certain occasions the disease was almost imperceptible and few scattered stromata per leaf were present. On other occasions it was damaging, deforming and causing a general chlorosis of severely infected shoots. Six different mycoparasites of the stromata of C. miconiae commonly were observed on this fungus and in some locations they clearly play a major role in limiting the potential damage caused by yellow pustule disease. It is difficult nevertheless at this stage to explain the differences in severity observed in the field. Based on the many observations that were made of this disease C. miconiae is regarded here as one of the most promising biocontrol candidates found in the mycobiota of M. calvescens. Glomerella cingulata is a common pathogen of M. calvescens in Brazil and on several occasions it was observed in the anamorphic state causing a serious foliar anthracnose. On only one occasion, widespread dieback, starting at the flower buds and descending along the branches, also was associated with this fungus (VIC 22207, Ipeu´na, SP). The morphology is typical of G. cingulata and the data for the fungus on miconia readily fits into the description of this taxon and its C. gloeosporioides anamorphic stage given in the literature (von Arx 1957, Mordue 1971, Sutton 1980). Although isolates initially were obtained from the lowland biotype leaf form of M. calvescens that is present in Brazil, tests carried out at HDOA (Quarantine Lab) based on an isolate from VIC 19284 have shown that the fungus is host specific to M. calvescens and a new taxon at the forma specialis level has been proposed (Killgore et al 1999). A permit for the introduction of the fungus was granted by Hawaiian authorities and the fungus was introduced in selected sites in the Island of Hawaii in 1997. Later damaging outbreaks of miconia anthracnose were observed at sites unexpectedly distant from the release sites (Barreto et al 2000). This represents the first example of a program involving the classical approach to biological control that has led to the introduction of a fungal pathogen native to Brazil targeted at a weed in an alien situation (Barreto et al 2001). This is also one of the few examples where a nonbiotrophic fungus has been released as a classical biocontrol agent. Colletotrichum gloeosporioides f. sp. clidemiae was introduced to Hawaii from Panama against Clidemia hirta (L.) D. Don and regeneration of native species in infested areas was considered as striking in many areas where it was introduced (Trujillo 2005). The genus Korunomyces was, until now, monotypic. K. terminaliae was proposed by Hodges and Ferreira (1981) for a fungus causing a leaf and stem blight on Terminalia ivorensis A. Chev. These authors discussed

SEIXAS ET AL: MYCOBIOTA OF MICONIA CALVESCENS the similarity of this fungus to members of Cristulariella, Papulaspora and Aegerita and concluded that the fungus on T. ivorensis deserved the status of a new genus. Nothing has been published on this genus since then. The fungus found on M. calvescens is similar to K. terminaliae but morphological differences are regarded here as sufficient to place it as separate species. Perhaps the most significant difference between the two species is the erect position and clear achene-like form and probable dispersal function of the combination of propagule-propagulophore in K. terminliae and the predominantly prostrate condition of propagules of K. prostratus. In the new species these structures probably are not functional as dispersal units and appear to work instead as infection pads. Dispersal in this species probably is dependent on propagule elements or some spore stage that was not observed during the present work. An attempt was made at inducing teleomorph formation of Korunomyces sp. with an adaptation of the method used by Silveira and Alfenas (2001) for Thanatephorus cucumeris (A.B. Frank) Donk. A mycelial suspension produced on a semisynthetic medium (Alfenas et al 1991) was brushed on leaves of fresh, healthy cuttings of M. calvescens in Erlenmeyer flasks containing tap water. The inoculated branches were left in a mist chamber at 26 C, with a 12 h light regime for 20 d (nine daylight lamps, 40W, suspended 1 m above cuttings). Plant parts were observed every 2 d for the appearance of symptoms. Unfortunately this did not result in the production of other stages in the lifecycle of K. prostratus. Nuclei staining (HCl-Giemsa) was performed (Herr 1979) and microscopic mounts revealed that K. prostratus is multinucleate. Korunomyces prostratus was found in several locations in Brazil (states of Rio de Janeiro, Sa˜o Paulo, Minas Gerais) and also during additional surveys in Costa Rica and Ecuador associated with the highland biotype of miconia. Leaf blight often was damaging but the number of diseased leaves per plant was rarely high. Pathogenicity of Korunomyces sp. was evaluated by inoculating healthy detached leaves of Eucalyptus grandis A.W. Hill ex Maiden, M. calvescens, Terminalia catappa L., Terminalia ivorensis A. Chev. and with culture plugs. The fungus was cultivated in VBA. After 7 d mycelial plugs obtained from the margins of actively growing cultures were transferred to the abaxial and adaxial sides of the detached leaves (four plugs per leaf, four leaves per plant species). Leaves were placed in humid chambers (sealed inflated plastic bags containing trays with wet cotton pads). These were left at room temperature and examined three times a day to follow symptom development. Results of such inoculations showed that K. prostratus

109

(under such conditions) is capable of causing necrosis on leaves of M. calvescens, T. ivorensis and E. grandis but not of T. catappa. There appears to be a partial overlap of the host range of the two species of Korunomyces. Korunomyces terminaliae was capable of infecting three species of Terminalia but not Eucalyptus grandis (Hodges and Ferreira 1981). This result also indicates that considerable care needs to be taken regarding the elucidation of the host range of K. prostratus before its introduction as a classical biocontrol agent is considered. Nevertheless, if safety requirements are met and an adequate procedure of manipulating this fungus is developed, it might prove useful as a biocontrol agent. Only two species of Guignardia are described on Miconia, G. atropurpurea Chardon and G. punctiformis Chardon (Chardon et al 1940). Both were described from specimens collected on Miconia sp. from Vic¸osa, MG, Brazil. This work was undertaken at the type locality for both fungi but, during 6 y of collecting diseases of M. calvescens in the region, these fungi were never found. It is likely that both fungi have different species of Miconia as a host. Guignardia miconiae differs from G. atropurpurea by having wider asci (13–18 mm in G. atropurpurea) and ascospores that are of a different shape and size (longelliptical and 17–21 3 7–8 mm in G. atropurpurea). Guignardia punctiformis produces smaller lesions on the host (4–6 mm diam), has wider asci (30–34 mm diam) and ascospores of a different shape and size (ellipsoidal to subpyriform, 20–23 3 10–14 mm). No anamorphic state was described for either G. atropurpurea or G. punctiformis. Guignardia miconiae was rare, being found only at two distant sites (Ilha Grande in the state of Rio de ´ guas da Prata in the state of Sa˜o Paulo). Janeiro and A It has a typical hemibiotrophic habit, starting to produce its fruiting structures in green tissue that gradually yellows and become necrotic. In a recent visit to Ilha Grande damage to individual leaves was significant. Several attempts at isolating this fungus were unsuccessful. This was unexpected because fungi in this genus are known to grow well in culture. An attempt to inoculate M. calvescens with a suspension obtained by squashing fresh ascomata of material from Ilha Grande also failed to yield symptoms. For the moment it is difficult to evaluate the biocontrol potential of this fungus. Nineteen cercosporoid fungi have been described on the Melastomataceae. Many of those species have been re-examined recently but some still await a reevaluation under the new concepts for the taxa in this group (Braun 1995, 1998). In addition to these taxa there is a record of a ‘‘Cercospora’’ on the phylloplane of Metrosideros polymorpha Gaud. from Hawaii (Baker

110

MYCOLOGIA

et al 1979). Metrosideros polymorpha (local name ohia) is the dominant forest tree in the Hawaiian archipelago and a key species in any host-range test for candidate agents for use in classical biological control in Hawaii. Confirmation of this record and elucidation of the identity of the fungus, despite the relevance for this work, is not possible because no materials appear to have been deposited by the authors in a herbarium. Six species of Cercospora-like fungi have been recorded on hosts belonging to the genus Miconia (Chupp 1953, Vie´gas 1961, Farr et al 1989), Pseudocercospora erythrogena, Cercospora melastomatis, C. miconiae, C. miconicola, P. mirandensis and P. tamonae. Types of some relevant species were obtained for comparison with the material from Brazil and original descriptions were studied. The cercosporoid specimens obtained from miconia in Brazil fit well within the descriptions and are similar to the type specimen of P. tamonae and therefore were recognized as belonging to this taxon. The degree of damage caused by P. tamonae was variable. In one instance it caused minor disease (angular leaf spots) on a single plant of M. jucunda. In another situation it caused a severe disease of foliage of Miconia sp. at Chapada dos Guimara˜es. Another isolate of this fungus from Parintins was sent to HDOA Quarantine Lab for further study. Tests undertaken in Hawaii indicated that P. tamonae is capable of causing severe disease on the Hawaiian biotype of M. calvescens. This fungus unfortunately appears to have a wide host range attacking Psidium cattleianum Sabine and other plant species. It also appears to be unable to complete its cycle on the Hawaiian biotype of M. calvescens. Under these experimental conditions it did not sporulate on the inoculated plants. In addition to the above fungi specimens of Corticium sp., Melanconium sp., Myrothecium sp., Pestalotiopsis sp., Phyllachora sp. and Pythium sp., were collected on M. calvescens or other species of Miconia. None of the additional fungi collected in Brazil in association with M. calvescens were regarded as being of interest for biocontrol either because of limited host damage or because of their plurivorous habit. ACKNOWLEDGMENTS

This work forms part of a research project submitted as a doctoral dissertation to the Departamento de Fitopatologia of the Universidade Federal de Vic¸osa by C.D.S. Seixas. This study was supported by USGS BRD Pacific Islands Ecosystem Research Center, National Park Service and the Research Corporation University of Hawaii. Useful suggestions on this manuscript were made by C. Smith. SEM studies were performed at the Nu´cleo de Microscopia e

Microana´lise of the Universidade Federal de Vic¸osa. R.W. Barreto and C.D.S. Seixas also acknowledges financial support from Conselho Nacional de Desenvolvimento Cientı´fico e Tecnolo´gico CNPq. LITERATURE CITED

Alfenas AC, Peters I, Brune W, Passador GC. 1991. Eletroforese de proteı´nas e isoenzimas de fungos e esseˆncias florestais. Vic¸osa: Universidade Federal de Vic¸osa. Baker G, Dunn PH, Sakai WS. 1979. Fungus communities associated with leaf surfaces on endemic vascular plants in Hawaii. Mycologia 71:272–292. Barreto RW, Killgore E, Sugiyama LS, Santos CD. 2000. Biological control of Miconia calvescens, the ‘cancer vert’ with fungal pathogens: present and prospects. In: Third International Weed Science Congress 2000. Foz do Iguac¸u´. [Abstract]. 170 p. ———, Seixas CDS, Killgore E. 2001. Colletotrichum gloeosporioides f.sp. miconiae: o primeiro fungo fitopatogeˆnico brasileiro a ser introduzido no exterior para o controle biolo´gico cla´ssico de uma planta invasora (Miconia calvescens). In: 7 Simpo´sio De Controle Biolo´gico 2001. Poc¸os de Caldas. [Abstract] 109 p. Baruch Z, Pattinson RR, Goldstein G. 2000. Responses to light and water availability of four invasive Melastomataceae in the Hawaiian Islands. Int J Plant Sci 161:107– 118. Braun U. 1998. A monograph of Cercosporella, Ramularia and allied genera (Phytopathogenic Hyphomycetes). II. Eching: IHM-Verlag. ———. 1995. Miscellaneous notes on phytopathogenic hyphomycetes. II. Mycotaxon 55:223–241. Cannon P. 1996. Coccodiella arundinariae. Mycopathologia 135:39–40. Chardon CE, Miller JH, Muller AS. 1940. Ascomycetes from the state of Minas Gerais (Brazil). Mycologia 32:172– 204. Chupp C. 1954. A monograph of the fungus genus Cercospora. Ithaca: Published by the author. 667 p. Csurhes SM. 1997. Miconia calvescens DC. (Melastomataceae) a potentiality invasive plant in Australia’s tropical and sub-tropical rainforests. In: Meyer JY, Smith CW, eds. Proceedings First Regional Conference on Miconia Control. Papeete, Taiti. Aug 1999. Tahiti: ORSTOM. p 72–77. ———, Edwards R. 1998. Potential environmental weeds in Australia. Canberra: Environmental Australia Diversity Group. Farr D, Bills GF, Chamuris GP, Rossman AY. 1989. Fungi on plants and plant products in the United States. St Paul: APS Press. Gagne´ BH, Loope LL, Medeiros AC, Anderson SJ. 1992. Miconia calvescens: a threat to native forests of the Hawaiian Islands. Pac Sci 46:390–391. Herr L. 1979. Practical nuclear staining procedures for Rhizoctonia-like fungi. Phytopathology 69:958–961. Hodges CS, Ferreira FA. 1981. Korunomyces, a new genus of Fungi Imperfecti from Brazil. Mycologia 23:334–342.

SEIXAS ET AL: MYCOBIOTA OF MICONIA CALVESCENS Judd WS, Skean Jr JD. 1991. Taxonomic studies in the Miconiae (Melastomataceae). VI. Generic realignments among terminal-flowered taxa. Bull Fl Mus Nat His Biol Sci 36:25–84. Katumoto K. 1968. On the genus Coccodiella Hara. J Jap Bot 43:277–284. Killgore EM, Sugyiama LS, Barreto RW, Gardner DE. 1999. Evaluation of Colletotrichum gloeosporioides for biological control of Miconia calvescens in Hawaii. Pl Dis 83: 964–964. Loope LL. 1997. HNIS report on Miconia calvescens. http://www.hear.org/MiconiaInHawaii/ Medeiros AC, Loope LL, Conant P, McElvaney S. 1997. Status, ecology, and management of the invasive plant Miconia calvescens DC (Melastomataceae) in the Hawaiian Islands. Records, Hawaii Biological Survey for 1996, Bishop Museum Occasional Papers 48:23–36. Meyer JY. 1996. Status of Miconia calvescens (Melastomataceae) in the Society Islands (French Polynesia). Pac Sci 50:66–78. ———, Florence J. 1996. Tahiti’s native endangered by the invasion of Miconia calvescens DC. (Melastomataceae). J Biogeo 23:775–781. ———, Malet JP. 1997. Study and management of the alien invasive tree Miconia calvescens DC. (Melastomataceae) in the islands of Raiatea and Tahaa (Society Islands, French Polynesia): 1992–1996. Technical Report 111.

111

Honolulu: Cooperative National Park Resources Studies Unit University of Hawaii at Manoa. 56 p. Miller JH, Burton MG. 1943. Study of Bagnisiopsis species on the Melastomataceae. Mycologia 35:312–334. Mordue JEM. 1971. Glomerella cingulata. CMI Descriptions of Pathogenic Fungi and Bacteria 315:1–2. Pereira JM, Barreto RW, Ellison CA, Maffia LA. 2003. Corynespora cassiicola f.sp. lantanae: a potential biocontrol agent for Lantana camara from Brazil. Biol Cont 26:21–31. ¨ ber Bagnisiopsis und verwandt Gattung. Petrak F. 1927. U Hedwigia 68:251–290. Renner SS. 1993. Phylogeny and classification of the Melastomataceae and Memecylaceae. Nord J Bot 13: 519–540. Silveira SF, Alfenas AC. 2001. Esporulac¸a˜o e inoculac¸a˜o de Thanatephorus cucumeris (Rhizoctonia solani Ku¨hn AG1) em brotac¸o˜es de eucalipto. Sum Phytopathol 27: 217–223. Sutton BC. 1980. The Coelomycetes. Kew: CMI. Trujillo EE. 2005. History and success of plant pathogens for biological control of introduced weeds in Hawaii. Biol Cont 33:113–122. Vie´gas AP. 1961. ´Indice de Fungos da Ame´rica do Sul. Campinas: Instituto Agronoˆmico de Campinas. von Arx JA. 1957. Die Arten der Gattung Colletotrichum Cda. Phytopathol Z 29:413–468.