Graphiolales: Basidiomycetes parasitic on palms

Pinnt SV.st.e.mntics nnd Eunlutinn

P1. Syst. Evol. 140, 251--277 (1982)

O by Springer-Verlag 1982

Graphiolales: Basidiomycete8 P a r a s i t i c

on Palms1

By

F. Oberwinkler, Tfibingen, R. J. Bandoni, Vancouver, P. Blanz, G. Deml, and L. Kisimova-Horovitz, Tfibingen

(Received July 14, 1981) Key Words: Basidiomycetes, Heterobasidiomycetes, Graphiolales, ord. nov., Graphiola; Arecaceae, Phoenix: basidiomycetous yeasts. - - Dimorphism, ultrastructure, septal pore apparatus, dikaryon, meiosis, synaptonemal complex, basidia, basidiospores, spore-germination, diazonium blue B-test, urease-activity, ferrichrome. Abstract: Graphiola phoenicis was restudied by light microscopy and investigated in detail with the scanning and the transmission electron microscopes. Hyphae of the fruitbody are mainly dikaryotic. Karyogamy occurs in cells which are interpreted as meiosporangia (basidia), and which develop in chains. Shortly after karyogamy, meiosis takes place in these basidia. Primary, sessile meiospores are then formed which later divide and produce thick-walled diaspores. The latter germinate either by hyphae or by yeast-like budding. The nutritional requirements of pure cultures of the yeast stage were also investigated. Life cycle, karyological criteria, ultrastructural details, and chemical tests clearly show that Graphiola belongs in the Basidiomycetes. The taxonomic .position within the Heterobasidiomycetes is discussed and the order Graphiolales Is validated.

Graphiola phoenicis (Mouc.) POITEAU parasitizes leaves of Phoenix (Arec,aceae, Palmae). I t occurs not only in the natural range of the host, but also on species which are cultivated as o r n a m e n t a l trees, e.g. Phoenix canariensis in the tropics, subtropical regions and in greenhouses. This parasite was classified as a P y r e n o m y c e t e by FarES (1823), KuNz~, (1826), DUBV (1830) and MONTAGNE(1859), while POITEAU (1824), who introduced the generic n a m e Graphiola, aligned it with the Myxomyc, etes, as did LI~V~:ILT,1;:(1848). Affiliation with the rusts was 1 Part 15 in a series "Studies in Heterobasidiomycetes" from the Lehrstuhl Spezielle Botanik der Universit~t Tfibingen, and the Department of Botany, University of British Columbia.

0378-2697/82/0140/0251/$ 05.40

252

F. ()m:j~wlxl~j,l~r et al. :

proposed by CHEVALIER (1826), Cot~J)i (1842), BONOI~I)EN (1851) and BAIl, (1858) and it seems t h a t TU~,ASNE also (1854) perferred such an interpretation. I t was EDUARD FmCHER (1883) who first critically investigated Graphiola phoenicia' a hundred years ago. H e came to the conclusion t h a t this organism and related ones could be best compared with the smuts. Almost 40 years later (1921), he proposed the family Graphiolaceae and included SYDow's genus Stylina as a second one in this taxon. H e discussed the taxonomic p o s i t i o n at some length, then concluded: ,,Vielmehr ist jede Diskussion fiber die Stellung dieser kleinen, gut charakteristierten Pilzgruppe verfriiht, solange wir fiber ihren Entwicklungsgang und ihre Zytologie niehts wissen." Three years later, K-ILLIAN(1924) published a karyologieal s t u d y of Graphiola phoenicis which he carried out in RENE MAIItE'S l a b o r a t o r y in Algiers; he concluded t h a t the species belonged with the Ustilaginales. The occurrence of Graphiola in J a p a n was reported by KO~AYASH1 (1952), who also discussed the possibility of a new order, Graphiolale8 (article in J a p a n e s e ; cited in TUBAKI & YOKOYAMA1971). TUBAK1 & YOKOYAMA (1971) studied the cultural aspects of Graphiola and mentioned the possible phylogenetic relationship with the red yeasts. Meanwhile, HUGHES (1953) was of the opinion t h a t the spore-producing h y p h a e of Graphiola are conidiophores; he included the genus in the Deuteromycetes, an alternative already proposed by YON HOHNEL (1909) and discussed by FISCHER (1921). I n his review of the Heterobasidiomycetes, DONK (1973) assessed all the published d a t a , trusted KILLIAN'S findings and called the group Graphiolales. KENDRIOK & CARMICHAEL (1973) and CAr~MmnAEL &al. (1980) listed the genus Graphiola in their survey of the Fungi Imperfeeti, and KEND,t~OK & WATLING (1979) did not m a k e a decision on the proper taxonomic position of the genus. Methods and Materials

The descriptions and illustrations in this paper were derived from study of living material of Graphiola phoenicis collected in the following localities: Canary Islands, Tenerife, Puerto de la Cruz, on Phoenix canariensis, Feb_ 16, 1978, leg. L. K ISIMOvA-Hoaovvrz & F. O BERWINK[,En.FO 25205 ;--West Germany, Tfibingen, Botanical G~rden, on Phoenix canariensis, April 10, 1978, leg. J. FrANTZ & L. KIS~M0vA-HoRoVITZ,FO 25633, 26634; May 24, 1978, leg. P. B LANZ PB 4349; March 24, 1981, leg. L. KtSlmOvA-HoROVITZFO 31575, 31576 ;--Greece, Crete, Pinikodasos near Vai, on Phoenix theophrasti, May 31, 1979, leg. F. O BEIt,WINKLER& L. K ISIMOvA-HoRoVJTZFO 28695. Low iron media, cultural conditions, assay for sideramine production, isolation, and identification of sideramines are described in DEmL & OS~[r WINKLEr(1981 b).

(/raphiolale.~': Ba,~'idiomycetc,~Parasitic on Palms

253

For scanning electron microscopy, the fruitbodies were fixed in glutaraldehyde and osmium tetroxide, washed with distilled water, dehydrated in a alcohol series, followed by critical-point drying according to SAUTTEH'Sprocedure (1977). Shortly before examination in a Cambridge steroscan S 4-10, the material was fixed on a specimen holder, broken, and coated with goldpaladium. For transmission electron microscopy, material was fixed in glutaraldehyde aad osmium tetroxide, washed with distilled water, stained in aqueous uranyl acetate, dehydrated in an ethanol series, and embedded in epoxid resin according to Splml~ (1969). Ultrathin sections were mounted on unsupported mesh copper grids, and examined in a Zeiss EM 9 S-2 transmission electron microscope. Results

Young developmental stages of Graphiola phoenicis can be found in the host tissue as pustulate aggregations which cause swellings of the leaves and which finally break through the epidermis. Mature fruiting bodies are often opposed to each other on opposite sides of a leaf (Fig. ll), grow to ca. 1/e mm high, and reach approximately 11/2 mm in diameter. An outer peridial layer (Figs. ! 1; 12 op) is composed of thick walled hyphae, 3-6~m in diameter (Fig. 1), which are irregularly branched apically (Fig. 13) and which bear scattered simple septa. In thick sections, these hyphae appear blackish; in squashed mounts, however, they have a greenish tint and are also characterized by substances which are partly soluble in K O H and Lactophenol. The peridial hyphae originate from thin-walled basal hyphae, 2-3 ~m in diam., with regular terminal branching (Fig. 2) and irregular growth between the host cells (Fig. 3). Short side-branches of these hyphae often attach to palm cells, form a tiny haustorial neck (Fig. 3), penetrate the host cell wall and expand inside the cell into irregularly ramified haustoria (Figs. 3, 16 ha). The mode of penetration is comparable to t h a t found in other basidiomycetous parasites, specifically the rusts and smuts. From the basal hyphal layer (Figs. 3, 12 bh), not only the outer peridium but also the inner peridium (Fig. 11 ip) develops~ I t is a fragile layer which can be easily overlooked and commonly is attached to the outer peridium in mature fruitbodies. A third sterile hyphal system, the hyphal strands, is present and also starts from the basal hyphal layer (Figs. 4, 11 hs, 31 hs, 32 hs). Cells of the strands are thin-walled below (Figs. 31, 32 hs), become increasingly thickened upward, and finally show only very thin lumina (Figs. 4, 25, 26). The gradually thickening of the cell walls may result from a deposition of cell wall material at the inner side and consequently leads to a reduction of the cytoplasm. Cross sections (Fig. 25)

254

F. ()J~EI(WINI~IA.]R e t a]. :

show a multilamellar construction of these hyphal walls. The original hyphal cell chain is visible in the thickened p a r t s (Figs. 4, 26) by constrictions which lack the fibrillar ultrastructural differentiation. Finally, the generative hyphae also develop from the basal hyphal layer (Figs. 5, 7, 11, 12) with c o m p a r a t i v e l y short, ramified and thin-walled cells of a diameter of 2-3 ~m. T h e y elongate considerably (Figs. 5, 14) and are divided into short cylindrical cells which later swell (Figs. 6, 7, 14, 18). Towards the apex of this chain, outgrowths of the cells can be observed which are inconspicuous at first (Figs. 5, 6, 7, 14) and expand to ca. 1/3 of the diameter of the m o t h e r cell (Figs. 6, 7, 14, 15). These sessile, p r i m a r y spores can be separated from the chain together with the mother cell which finally collapses. The p r i m a r y spore itself is further divided (Figs. 7, 19) into secondary spores, 3-6 ~m in diameter and provided with p e r m a n e n t l y thickened wails (Fig. 28), the periphery of which becomes w a r t y (Figs. 7, 17, 20, 27, 28). The exospore sculpturing gradually develops and finally is visible in ultrathin sections as electron dense spots inserted in the outer layer of the considerably thickened walls of m a t u r e spores (Fig. 20). Spore production is accompanied by continuous enlongation of the hyphal strands ; these protrude from the fruitbodies and move hygroscopically, thus assisting in spore release. Irl water or on artificial media, the spores germinate readily and rapidly. The spore wall ruptures (Fig. 9) and a short projection develops on which elongate buds are produced. The budding continues to yield a yeast colony (Fig. 10). However, the spores also are able to germinate by hyphae, 1.5-2 ~m in diameter (Figs. 37-40), which branch and later become septate. The cell walls of the y e a s t and of the germ tubes are Figs. 1-10. Graphiola phoenicis. - - Fig. 1. Squashed hyphae from the outer peridium. - - Fig. 2. Basal hyphal layer (bh) of the peridium. - - Fig. 3. Hyphae between the host cells and penetrating into them with haustoria (ha); upper part : formation of the basal hyphal layer (bh) from which hyphal strands and generative hyphae originate. - - Fig. 4. Terminal part of the hyphal strands, showing closely packed, thick-walled hyphae. - - Fig. 5. Fascicle of generative hyphae (gh) with basal hyphal branching and, apically, the beginning of primary spore budding. - - Fig. 6. Two generative hyphae demonstrating the cell sequence and the centrifugal growth of primary spores. - - Fig. 7. A single generative hypha, showing the nuclear behavior: basal and middle part with dikaryotic cells (dk), then karyogamy (K !) and the beginning of primary spore (ps) formation by lateral outgrowths; meiosis (R!) in the basidium (ba) and further growth of the primary spores. - - Fig. 8. Separation of the uppermost basidium wit h primary spore still attached; septation of primary spore to form secondary spores (ss) which finally separate from each other. - - Fig. 9. Germination of secondary spores by rupturing of the spore wall and yeast-like budding. - - Fig. 10. Different stages of budding of yeast cells

~~

o '

~[f

~

CJ

~

C~

~

c

~.

256

F. ()I~EICW[NKIA.]Ir e t ~1. :

thin and hyaline. We used ultrathin sections also to s t u d y the nuclear behavior of Graphiola phoenicis with the transmission electron microscope. Dikaryotic cells were easily found in the generative hyphae (Figs. 7, 31), and also in the basal hyphal layer connected with the peridium, in the haustoria, and in the hyphal strands (Fig. 32). The dikaryotic stage in the generative h y p h a e is present in the l)asal cells, the elongate ones of the middle p a r t and the very short apical cells (Fig. 7). When apical cells begin to swell, only monokaryotic stages are visible and the nuclei are a p p a r e n t l y bigger than of tbe /)inucleate stage. P r e d o m i n a n t l y in cells which are beginning to produce p r i m a r y spores, structures are visible which strongly resemble s y n a p t o n e m a l complexes (Figs. 33, 34, 35 sy). When the primary spores reach a p p r o x i m a t e l y 1/2 the size of the mother cell, more than one nucleus can often be found there" (Fig. 30). Occasionally, stages also are visible which m a y represent the migration of one nucleus into the daughter cell (Fig. 36). Many secondary spores seem to be mononu(tleate (Fig. 20); however, there are also others which are binucleate. Through spore germination, mononucleate yeasts (Figs. 9, 37, 38) or hyphal cells (Figs. 39, 40) are produced. Rarely two germ tubes develop from one spore (Fig. 37). The mechanism of spore germination obviously resembles strongly the outgrowth of the p r i m a r y spores (Figs. 29, 30, 33, 36). The original mother cell wall is ruptured, and the innermost 1)art extends to produce the bu(l(ling initial. During further development of the p r i m a r y spore, its cell wall thickens slightly, but is already often ornamented with knots which only partly protrude to the outside. The sel)tal pore apparatus, another i m p o r t a n t ultrastructural character, was also studied. All median sections (Figs. 21 24) show simple pores, which, however, v a r y considerably during a certain stage of development. Several times (Figs. 21, 24) electron dense arches covering the septal opening could be detected ; other sections (Figs. 22, 23) do not show these structures. Pure cultures of the yeast stage developed pinkish colored colonies. These were unable to ferment glucose, but strongly assimilated glucose

Figs. 11 14. SEM-mierographs of Graphiola phoenicis. Fig. 11. l,~mgitudinal section through two opposed fruitbodies on the host leaf, sh()wing outer (op) and inner peridium (ip), hyphal strands (h,~), generative hyphae (gh) and leaf tissue (It) of" the host. Bar equals 200 tzm. --- Fig. 12. Detail of' a longitudinally sectioned fruiting budy with basal hyphae (bh), outer peridium (op) and generative hyphae (gh). Note tile hyphal arrangements in different parts. Bar equals 100 ~m. - - Fig. 13. Hyphal tills of the outer peridium. Bar equals 5 ~m. - - Fig. 14. Palisade of generative hyphae with different stages of pri mary spore development and successive separation of basidia. Bar equals 10 ~m

(,'raphiolal~.s: l?,asidiom!tcetes

Parasitic on lJalms

Figs, 11-14

257

258

F. ()]~J~Jm'[.~K[,EJ~ eta]. :

and sucrose, also galactose fairly well, and maltose, and melibiose to some minor extent. However, lactose was not utilized as a carbon source. After three days growth, weak splitting of arbutin could be observed, while after 10 days this reaction was strong. Urease was detectable within 2 to 3 d a y s and the diazonium blue B test was positive with 3 week old yeast cultures. The sideramine, ferrichrome, could also be found.

Discussion EDUARD FISCHER's studies (1883) on Graphiola phoenicis unraveled the most i m p o r t a n t morphological and developmental characteristics of this species. His findings, which were verified b y our own light microscopic investigation, were explained and illustrated in some detail. W i t h the aid of SEM and T E M techniques, we were able to confirm his results and to obtain some additional information on the species. The gross morphology of the f r u i t b o d y is absolutely unique, a fact which explains why the Graphiolaceae were Placed in such different m a j o r fungal groups as Myxomycetes (PoITEAU 1824, L]~VEILLI~1848), Pyrenomycetes (FRIES 1823, KUNZE 1826, DUBY 1830), Deuteromycetes (VON HOHNEL 1909, HUGHES 1953), rusts (CHEVALIER 1826, CORDA 1842, BONORDEN 1857, BAIL 1858), and s m u t s (FISCHER 1883, KILLIAN 1924, TUBAKI & YOKOYAMA1971). The most critical analysis of FISCHER (1883, 1921) finally led to the establishment of a separate order within the Basidiomycetes (KOBAYASHI 1952, DONK 1973). FISCHER (1. C.) first elucidated the main morphological features, i.e. two peridial layers, hyphal strands, and generative hyphae, and he properly interpreted those according to their functions. H e also was the first to suggest a possible relationship with the s m u t s (1883). Almost four decades later (1921) he returned to his research of earlier years. H e was still convinced of his earlier findings, but was also open minded as to other interpretations, e.g. including the Graphiolaceae in the Fungi I m perfecti, provided t h a t i m p o r t a n t d a t a could be found to support this conclusion.

Figs. 15-18. SEM-micrographs of Graphiola phoenicis. -- Fig. 15. Upper region of generative hyphae with basidia (ba) and different stages of primary spore (ps) development. Bar equals 2 Izm. - - Fig. 16. Tissue of palm leaf with intercellular hyphae (ih) and haustoria (ha) inside the host cell. Bar equals 3 ~m. - - Fig. 17. Secondary spores attached to one another, showing the blunt warty outer wall. Bar equals l ~zm. - - Fig. 18. Generative hyphae from the middle region with short cylindrical cells. Bar equals 1 ~m

(Iraphioltde.s': Ila,s'idiom..qcele.s

18

171. ,~.vst. Evol.. Vol. 140. No. 4

F i g s . 15 18

P a r a s i t i c on l ' a l m s

259

260

F. ()BJ']R%VINI(I,I,]I{ct al. :

F r o m our own knowledge, we can conclude t h a t there still are no other possible g r o u p s to which the Graphiolaceae can be linked. T h e rust aecia are only superficially similar a n d no true fruitbodies with a c o m p a r a b l e c o n s t r u c t i o n p a t t e r n are k n o w n in the smuts. W e are aware of the similarities of Farysia olivacea, as was F ]SellER (1921), especially concerning the h y p h a l strands. B u t peridial layers and generative h y p h a e differ so u n e q u i v o c a l l y t h a t a close relationship seems to be improbable. I n the course of a c o m p a r a t i v e u l t r a s t r u c t u r a l s t u d y of a v a r i e t y of yeasts KREGER-VAN RIJ & VEENHUIS (1971) f o u n d a r e m a r k able diversity in cell wall architecture. T h e wall t u r n e d out to be multilamellar in Basidiomycetes, with several a l t e r n a t i n g electrondense and - t r a n s p a r e n t layers, while Ascomycetes walls a r e c o m p o s e d of only two lamellae, a s d a r k outer one and a conspicuously thicker a n d t r a n s p a r e n t inner layer. DONK (1973 a), s t r o n g l y emphasizing this character, compiled the available and usable T E M - p i c t u r e s f r o m different publications to d e m o n s t r a t e its t a x o n o m i c i m p o r t a n c e ; he correctly refered to " t h e strict application of s t a n d a r d m e t h o d s " as well as the c o m p a r i s o n of w h a t he called " p r i m a r y walls". Such walls are n o t easily f o u n d in Graphiola phoenicis, b u t the generative h y p h a e m a y be suitable, t h o u g h even meiosporangial walls which are clearly multilayered (Figs. 34, 35) a p p e a r to be a l r e a d y thickened. Walls of y e a s t cells seem to be multilamellate, too (Fig. 43). All septa in Graphiola phoenicis are efibulate and it can be a s s u m e d t h a t this c h a r a c t e r is uniform also in the other species of the genus and of the family Graphiolaceae. We t h e n have a situation which is c o m p a r a b l e to t h a t of the Uredinales and Septobasidiales. I t seems t h a t also the Cryptobasidiales are simple s e p t a t e - - i n so far as we can judge from our own investigations. I t was n o t surprising for us to find simple septal pores in Graphiola phoenicis (Figs. 21 24; we are u n a b l e to interpret MOORE'S T E M - p i c t u r e of a Graphiola phoenicis s e p t u m , t a k e n f r o m a CBS-strain, 1972). Simple septal pores are well k n o w n in the Uredinales (EBRLtCH & al. 1968, LITTLEFIF,LD & BRACKER 1971, COFFZY & al. 1972, MIMS & al. 1976,

Figs. 19 26. TEM-micrographs of Graphiolaphoenicis. All bars equal l am. - Fig. 19. Primary spore divided into two secondary spores ; spore below with two nuclei (nu). -- Fig. 20. Secondary spore; note the thick cell wall and exospore warts at the periphery. - - Fig. 21. Generative hypha with median section of the septal pore. - - Figs. 22 24. Median sections of thick walled hyphae of the outer peridium, showing septal pores. - - Figs. 25, 26. Sections of the upper part of' hyphal strands; note the concentric wall layers in the cross section of Fig. 25 and the electron transparent regions of septal constrictions in the longitudinal section of Fig. 26

(;r~tphiolale,~': Ba,~'~[dio~llyce, te.~. l)~rasiti(,

la*

Figs. 19 26

on P a l m s

261

262

F. ()*~l~l'wl•

et ai. :

SEBALD 1977), Septobasidiales (DYKSTRA 1974, SEBALD 1977), and are also reported from the Atractiellales (OBERWINKLER & BANI)ONI, in press), some Cystobasidiaceae of the Auriculariales s.l. (SF,I~AIm 1977, KHAN & KIMBI>~OUGH1980; compare also O Bm~WlNKLER& ]3 ANI)ON1 1981, in press), Ustilaginales (I~AMBEI~G& MfZLAU(;HIAN1980), Exobasidiales (BLANZ 1977, 1978) and Cryptobasidiales (unpubl. data). The originally emphasized, but certainly oversimplified concept t h a t Basidiomycetes possess dolipores had to be modified when rust septa became known. There are several observations reported on septal pore types which nowadays are difficult to interpret taxonomically. The Auriculariales contain t a x a with dolipores in the A uricularia-Hirneola-group (S m~ALD 1977, MOORE 1978, PATTON & MAI~CHANT1978, TU & K Imm~ou(~n 1978, M('.LAU('HLJN 1980) while Herpobasidium struthiopteridis (SEBALD 1977), Eocronartium muscicola (SEBALD 1977, KHAN & KIMBROU(IH 1980) and species of severa} more genera (OBERWINKLER& BANDONI, in press) are simple pored. The other example are the Ustilaginales s.1. with simple pored species in Ustilago (RAMBERC & McLAUGHLTN 1980), Rhodosporidium and Leucosporidium (Moo[r 1972), and in the Tilletiales with a v a r i e t y of different pore types, e.g. dolipores in Tilletia (DEML 1977), and Entorrhiza (DEML & OBERWINKLER1981), while in Entyloma several modified types (DEML 1977) including simple pored (F. O. unpublished data) m a y be present. Therefore, the ultrastructural features of the septa] pore types also favor interpreting a possible relationship of the Graphiolales with the rust-smut complex. However, this would not be justified without additional supporting evidence of i m p o r t a n t characteristics. Certainly we have to preferentially consider the basidia. Until now, the basidium in Graphiola has not been recognized. KILL, AN (1924) examined the nuclear behavior extensively, and was able to demonstrate the dikaryon in the spore-producing hyphae. He also observed the diploid nuclei and supposed t h a t meiosis occurred in the spores shortly before germination. In contrast to this interpretation, we ibund t h a t the diploid nuclei undergo meiosis in those cells which produce the p r i m a r y spores (Fig. 7 P~ !), as indicated by chromosomal a r r a n g e m e n t s which we interpret as s y n a p t o n e m a l complexes (Figs. 33-35 sy). Figs. 27 32. TEM-micrographs of Graphiola phoenicis. All bars equal 5 ~m. - Fig. 27. Basidia budding off primary spores and detached primary spores. - Fig. 28. Secondary spores, mostly in detached pairs; note the asymmetrical exopore sculpturing. - - Fig. 29. Budding basidia; in the upper left corner, cells of a hyphal strand. Fig. 30. Basidium with budded off primary spore; note the two nuclei in tile basidial cell. --- Fig. 31. Basal hyphal layer with parts of the generative hyphae. - - Fig. 32. Upper portions of generative hyphae with basidial chains, budding basidia and separated basidia; hs = hyphal strand

(~raphiolale.~: Ba,~idioJllycete,~. P a r a s i t i c

Figs. 27 32

()Jl P a l m s

263

264

F. ()I}EI,'WIXI,~LEI~ e t

al. :

F o r t u n a t e l y , one of several different lots of fixed material from various fresh collections could be used for electron microscopy studies of nuclear behavior. Though we were unable to obtain outstanding EM pictures, we feel justified in interpreting the present ones with some caution. Relatively often, distal cells of generative hyphae were observed which were beginning to " s p r o u t " and within which a single large nucleus was present. In the nucleoplasm of these cells, triplets of parallel dense strands of constant structure and m e a s u r e m e n t s were observed. These consisted of an inconspicuous central element, on both sides of which there were electron t r a n s p a r e n t regions and, outside these, prominent electron dense elements. In s y n a p t o n e m a l complexes, according to MOSES (1968), " t h e center-to-center distance separating the lateral elements i~reasonably constant, varying a b o u t 0.1 to 0.15 in the cases reported, although extremes have been observed". The structures in our TEM-pictures agree fairly well with these proportions. Similar proportions have also been found in Basidiomycetes, first in Coprinus cinereus (= C. lagopus) by L u (1966, 1967), and in other t a x a in the Agaricales as listed by WELLS (1977), Aphyllophorales (Poria latemarginata by SETHFF & al. 1974), and also in the Heterobasidiomyceres, e.g., Myxarium subhyalinum ( - - S e b a c i n a subhyalina, WELLS 1971), and Auricularia fuscosuccinea (McLAUGHLIN 1979). GULL & NEWSAM (1975) studied species of Agaricus, Amanita, Coprinus, Hypholoma, Paneolus, and Russula, and found t h a t the lateral components of the s y n a p t o n e m a l complexes in Basidiomycetes do not show the banded substructure characteristic for Ascomycetes. Unfortunately, such details as banded substructures cannot be clearly observed in our pictures. S y n a p t o n e m a l complexes are accepted as characteristic of meiotic prophase in m a n y eukaryotes, documenting effective synapsis and thus distinguishing this from somatic pairing. From these observations, we deduce t h a t the upper cells of Graphiola generative hyphae which are mononucleate are meiosl)orangia and, consequently, the cells t h a t are budded off are sessile basidiospores. Next we considered the ontogeny of the basidium and its corresponding morphology. The generative h y p h a e produce apical cells successively and in a centrifugal manner. These short-cylindrical cells switch from a dikaryotic to a m o n o k a r y o t i c stage, the latter re-

Figs. 33, 34. TEM-micrographs of Graphiola phoenieis. Bars equal 0.5~ml -Fig. 33. Basidium with nucleus showing synaptonemal complexes (sy) of chromosomes and budding off of a primary spore. - - Fig. 34. Nucleus in a basidium with synaptonemaI complexes (sy)

(Irctphiolales: l]asidiom.qreles

P a r a s i t i c on P a l m s

Figs. 33, 34

265

266

F. Om':,~'WJNK,,EJ~et al. :

presenting the diplophase. Several diploid cells, present in a chain, form a sequence of probasidia which are reminiscent of teliospore columns in rusts like Cronartium, Chrysomyxa (excluding Uhrysomyxa abietis), Ceroteli~,,m, Crossopsora, etc., or even multi-celled teliospores like those of the l)hraffmidi~,m group. However, the next developmental step clearly reveals major differences. The Oraphiola probasidium and m a t u r e basidium are exomort)hologically identical, i.e. the meiosporangium. However, in the rusts mentioned above the prol)asidium ( - t e l i o s l ) o r e ) germinates to produce the final I)asidium (metabasidium, commonly the meiosporangium) of the auricularioid type. Therefore, we also see no homologous structures in the smuts. F,s(~HI,:I{ (1883) tried to compare Oraphiola with Tt~,b~,rcin,ia trie~#alis a n d ,gorosporiam ,~'apo,zar~iaeaccording to the information avai]al)le to him, including the excellent illustrations of W(),~)XIN (1881), t)ut he correctly concluded t h a t a connection between the two groups seemed iml)rohable. We consider it i m p o r t a n t t h a t Tuburciniadevelops a Tilletialike basidium and t h a t the smutspores of Sorosporit~m germinate to produce a basidium of the Ustilaqo tyt)e, I)oth obviously unlike the Graphiola meiosporangium. This is true also for Doassansia alismati, and Sphaeelotheca (= Ustila.qo) hydropiperi.% two more t a x a which FJscHI~I{ (I.e.) compared with Graphiola, stressing t)'uitl)ody-likc structures in these smuts. When ]{ACIBORSKI (!909) described Faryaia javanica, a s m u t parasitizing female flowers of a Oavanian species of Carex, he noticed hyphal strands in the sori and therefore concluded, ,,Sie scheint daher ein Analogon zu der G a t t u n g Graphiola zu bilden". FISCHER (1921) was so strongly impressed by this discovery and interpretation t h a t he restudied the species from a SCHIFFNEIr collection made in J a v a ; this material was sent to him by VON HOHNEL (1909) who also examined the fungus and emphasized the connection with Graphiola. F r o m the basal hyphal layer, an evanescent peridial layer, a conspicuous mass of generative hyphae, and hyphal strands are derived. Unfortunately, nothing is known ~bout the karyological behavior of the generative hyphae or the germination of the spores, facts which are needed to disentangle possible correlations. All spore-producing mechanisms in the Ustilaginales and Tilletiales have in c o m m o n the germination of the smutspore to produce the promycelium, i.e., the basidium.

Figs. 35, 36. TEM-micrographs of Graphiola phoenicia'. Bars equal 0.5 ~m. - Fig. 35. Basidium with nucleus and synaptonemal complex (sy). -- Fig. 36. Basidium with attached primary spore; note the two nuclei one of which is dividing into the spore; spore wall ornamented with knots

(Iraphiolale~.: Ba.sidiomycele.s

P a r a s i t i c on P a l m s

Figs. 35, 36

267

268

F. O BERW]NKLERet al. :

Though the meiosporangium in Graphiola is finally released from the generative hypha, it does not have the structure and function of the smutspore (Figs. 7, 33, 36). I t remains more or less thin-walled and it is the locus of the subsequent meiosis which permits delivery of haploid nuclei to the simultaneously budded primary spores. The Graphiola basidium is a distal cell of the generative hypha, not the germ tube or " p r o m y c e l i u m " of a teliospore. Further, the basidia are developed in a linear sequence in each generative hypha. They mature basipetally, separate apically as sporulating meiosporangia, and thus make room for those further down the chain. We know of no homologous structural development in other Basidiomycetes. Chains of meiosporangia are known from the Homobasidiomycetes in Repetobasidium (E R~IKSSON1958, OBERWINKLER ! 9 6 5 ) a n d in some genera of the Heterobasidiomycetes, e.g., Sirobasidium (LAGERHEIM& PATOUILLARD !892), Fibulobasidium (BANDONI !979), occasionally in Tulasnella and Gloeotulasnella (MARTIN 1945), and rust genera such as Chrysopsora and Trichopsora. However, not all of these represent homologous structures, the closest parallel being t h a t of Sirobasidium. In this genus, basidia in chains mature basipetally, the "epibasidia" or primary basidiospores are released from the basidium, then sporulate (BANDONI 1957}. I t seems necessary now to explain some of the terms which we used to describe the reproductive process in Graphiola. We call those cells in which k a r y o g a m y and later meiosis occur "holobasidia", since they are not divided by septa during their ontogeny. More specifically, they can be called caducous holobasidiate heterobasidia; they are structures set free from generative hyphae and are a part of a heterobasidiomycetous life cycle. Further, we use the terms " p r i m a r y - " and "secondary spores" in a simple descriptive manner. The bud-like cells developing on the basidium might be compared to sterigmata, but their walls show ultrastructural spore ornamentation; they later divide to form at least two propagules. Our term, " p r i m a r y spores" is therefore equivalent to FmCHE~'s {.1883) "Sporeninitialen". The name merely indicates t h a t these propagules are not the primary ones produced b y the basidia. From this presentation, we finally have to prove the heterobasidiomycetous relationship. Germinating secondary spores (Figs. 9, 37, 39) give important hints in t h a t yeast cells bud off and are capable of further growth to form colonies in pure culture. Thus, Graphiola shows the dimorphism which we again consider as characteristically heterobasidiomycetous (OBERWlNKLER1978). I t was already stated t h a t the simple septal pore type is not only in agreement with this interpretation, but leaves no other alternative. TUBAKI & YOKOYAMh (1971) reported the pinkish color of Graphiola phoenicis yeast colonies; they also refered to the similarly pigmented

(~ .,raphwlale:,~: Ba,~idiomycet(~,~Parasitic

on l)alms

/ J38Y

269

/

lO IJm

I

l

Figs. 37 40. Graphiola phoenicis. Secondary spores germinated on artificial media. - - Fig. 37. Sequence of budding ; one spore producing two germ tubes. - Fig. 38. Successive figures of yeast budding; note the asymmetric outgrowth of the bud. - - Fig. 39. Different sts of germination with hyphae; note mononucleate germ tube, simple septate hypha, and developing branches; one hyphal fragment budding. - - Fig. 40. Germination by hyphae and formation of adventitious septa with protoplast migration

270

F. Om,:~w~xl