Folia Microbiol. 54 (3), 217–229 (2009) http://www.biomed.cas.

cz/mbu/folia/

Genetic Variation of Phoma sorghina Isolates

from Southern Africa and Texas
S. PAŽOUTOVÁ
Institute of Microbiology, Academy of Sciences of the Czech Republic, v.v.i., 142 20 Prague, Czech Republic
e-mail pazouto@biomed.cas.cz
fax +420 296 442 332
Received 18 December 2008
Revised version 30 March 2009

ABSTRACT. Genetic variability of Phoma sorghina, a ubiquitous facultative phytopathogen, was investi-
gated on 41 isolates cultivated from surface-sterilized sorghum grains originating from South Africa and
Texas; pearl millet isolates from Namibia were also included. Most of the isolates from Texas produced intense
red pigments, especially on Czapek–Dox agar plates. Many African isolates formed conspicuous dark radial
substrate hyphae with intercalated chlamydospores on oatmeal plates. Conidial dimensions and shape were
very variable (mean lengths 4.5–5.7 μm). Haplotypes were defined based on 53 markers from banding patterns
obtained with rep-PCR (primers: M13core, ERIC IR). The shared geographic origin was partially reflected in
the clades of the haplotype phylogram. The values of GST were intermediate; 16–37 % of the variation was
found between the populations. Nm values of gene flow were 0.84–1.15. Average gene diversity HE was mo-
derate (0.256). Sequences of ITS-rDNA were obtained from 21 isolates. Allele 1 was found in 9 isolates
scattered throughout the clades, allele 2 occurred in 6 isolates (5 of them from the same clade), alleles 3 and 4
were shared by two isolates each and two isolates were unique. Alleles 1 and 2 were also found among highly
related sequences from GenBank. All shared an 8-bp deletion near the 5´ end of ITS2 that was not found in any
other Phoma/Didymella species and which may be a typical marker for P. sorghina. Among related species,
members of legume-associated Ascochyta/Didymella complex, Epicoccum spp., D. applanata and P. glome-
rata were found.

Abbreviations

AFLP amplified fragment length polymorphism NNI nearest neighbor interchange (method)
ERIC repetitive intergenic consensus (sequence) OA oatmeal agar
GTR general time reversible (substitution model) PCR polymerase chain reaction
ITS internal transcribed spacer RAPD random amplification of polymorphic DNA
MEA 2 % malt extract, 2 % agar (medium) UPGMA unweighted pair group cluster method with arithmetic averages

Phoma sorghina (SACC.) BOEREMA, DORENB. & KESTEREN is a cosmopolitan fungus belonging to
the Peyronellae section of the genus Phoma (Pleosporales) (Aveskamp et al. 2008; Boerema et al. 2004). It
occurs in a wide variety of plants, in soil, and other substrates. It is also encountered as a facultative plant
pathogenic fungus with preference for Gramineae in tropical regions, often reported as a cause of leaf spot
(Amaral et al. 2004; Venkatasubbaiah et al. 1992). It may also occur as an endophyte (de Souza Borges and
Tallarico Pupo 2006; Feldman et al. 2008). Together with Alternaria tenuissima, Curvularia lunata and Fusa-
rium spp., P. sorghina causes grain mould of sorghum and millets (Forbes et al. 1992). During the seed
colonization it may produce toxins harmful to humans (Rabie et al. 1975).
The teleomorph of P. sorghina is supposed to be either Leptosphaeria sacchari BREDA DE HAAN,
(according to Index Fungorum) or Didymella holci (TEHON) ARX (VON ARX 1987) (according to MycoBank),
both leaf pathogens of grasses belonging to Pleosporales. None of these teleomorph–anamorph relationships
was confirmed by DNA sequence comparison.
Morphological studies have shown considerable variability (White and Morgan-Jones 1983) mani-
festing itself in culture macromorphological (e.g., intensity of soluble red pigment production) and micro-
morphological features (conidial size and shape). This fact together with the wide range of substrates colo-
nized suggests that a high degree of genetic variability may exist within the species. In their study of glume
blight, de Souza et al. (1988) detected variable levels of aggressivity in P. sorghina isolates when inoculated
on the same rice cultivar.
Differences in ribosomal DNA ITS regions (ITS-rDNA) have been used for detecting variation inside
and between fungal species. In the GenBank, there is only one sequence attributed to P. sorghina (AF046022)
(Reddy et al. 1998). In species related to Phoma/Didymella complex, little variation has been repeatedly
218 S. PAŽOUTOVÁ Vol. 54

observed in the ITS-rDNA regions. The studies of the Ascochyta legume pathogens have shown that species
clearly delimited by their host specificity may have identical sequences (Faris Mokaiesh et al. 1996; Peever et
al. 2007). Abeln et al. (2002) found almost identical sequences in Phoma exigua host-specific varieties, although
their AFLP patterns considerably differed. Similarly, Somai et al. (2002) detected two distinct groups inside
Didymella bryoniae using RAPD but their rDNA were 99 % identical. Castell-Miller et al. (2008) found the
same situation in Phoma medicaginis isolates.
The objectives of this study were to assess the morphological and genetic variability of P. sorghina
isolates from sorghum and pearl millet and to establish the phylogenetical position of P. sorghina among the
Phoma/Didymella species based on ITS-rDNA. The samples were obtained from southern Africa and Texas.
Because of known problems with reproducibility of RAPD fingerprinting due to low annealing tem-
peratures, another PCR method was selected to distinguish among the P. sorghina genotypes. The method,
called rep-PCR, is based on ERIC primers, originally derived from bacterial repetitive sequences (Versalovic
et al. 1991) and M13core primer (Lieckfeldt et al. 1993). With its application of stringent annealing tempe-
ratures of >50 °C, the method was used for the development of an automated typing system (Healy et al. 2004).
The rep-PCR method has been also applied to identify genetic variation of fungi (Arora et al. 1996; Healy et
al. 2005; Tymon and Pell 2005).

MATERIALS AND METHODS

Isolation and cultivation. Isolates were obtained from harvested sorghum and pearl millet grain col-
lected from two African regions (South Africa, six locations, sorghum; Namibia, one location, pearl millet)
(Fig. 1) and two sorghum growing locations in Texas (Table I). Each grain sample corresponded to a single
field. From each sample, a total of 100–200
seeds was disinfected by 10–15 min in 1.5 %
sodium hypochlorite, rinsed in distilled water
thrice and plated on MEA (2 % malt extract, 2 %
agar). Grains with black dots (identified under
the microscope as pycnids of Phoma spp.) were
preferably chosen for plating. All isolates re-
sembling Phoma (producing pycnidia, chla-
mydospore-like cells or gray-green nonsporu-
lating mycelium) were subcultured on MEA
slants. Each isolate in Table I originated from
a single seed.
Morphology. Colonies were grown on
MEA, OA (both, Difco) and Czapek–Dox me-
dium at 24 ºC. Conidia from the specimens and
cultures were mounted in 1 % cotton blue–lac-
Fig. 1. African locations: 1 – Namibia, 2 – Potchefstroom, 3 – Del- tic acid and photographed and measured using
mas, 4 – Gladdedrif, 5 – Platrand, 6 – Bethlehem, 7 – Cedara. Olympus BX51 microscope equipped with
a digital camera CAMEDIA and image-pro-
cessing software QuickPHOTO Camera 2.2.
Spore measurements were made on 50–100 pycnidial conidia from cultures on MEA. The statistical treatment
of spore size data was done using Kyplot 2.0 beta 15 (Yoshioka 2002) available at http://www.
pricelesswarehome.org/WoundedMoon/win32/kyplot.html.
DNA analyses. DNA was purified from a 4-d-old mycelium grown on MEA plates overlaid with
cellophane using an UltraClean Microbial DNA Isolation Kit (Mo-Bio Laboratories, USA) according to the
manufacturer’s manual.
Repetitive elements-based PCR (rep-PCR) analyses have been done using primers M13core (5´-GAG
GGT GGC GGT TCT-3´) and ERIC IR (5´-ATG TAA GCT CCT GGG GAT TCA-3´). The reaction mixture
(20 μL) contained 0.2 mmol/L deoxynucleotides, 20 pmol of the primer, DynaZyme reaction buffer, 1 U
DynaZyme polymerase (both FinnZymes, Finnland) and MgCl2 in a total concentration of 1.75 mmol/L.
Cycling conditions in a Mastercycler Gradient thermocycler (Eppendorf, Germany) were: 3 min at 94 °C and
then 30 cycles of 30 s 94 °C, 90 s 52 °C, and 3 min 68 °C, followed by a 8-min incubation at 68 °C. Amplicons
were separated on 2 % agarose gels in 0.5× Tris–borate–EDTA buffer; DNA banding patterns were visualized
by staining with ethidium bromide and read manually. The presence or absence of DNA bands were converted
in 1–0 matrix that was analyzed by the FreeTree software (Hampl et al. 2001) where distance estimation was
2009 GENETIC VARIATION OF P. sorghina ISOLATES 219

performed using the Dice coefficient. An unweighted pair group cluster method with arithmetic averages
(UPGMA) was used to infer tree topology. Bootstrap analyses (1000×) were included (Felsenstein 1985).

Table I. Phoma sorghina isolates: their origin, morphotype and conidia size

Conidia dimensionsc
Isolatea Location Host Yearb Morphotype
length width

TX03_01 Texas, Beeville sorghum-1 2003 3 5.0 ± 0.6 2.2 ± 0.2

TX03_02 sorghum-2 2003 3 4.9 ± 0.6 2.2 ± 0.2
TX03_03 ditto 2003 3 – –
TX03_04 sorghum-3 2003 3 4.7 ± 0.4 1.9 ± 0.2
TX03_05 ditto 2003 3 6.4 ± 0.9 2.1 ± 0.2
TX03_06 sorghum-4 2003 3 5.2 ± 0.6 2.2 ± 0.2
TX03_07 ditto 2003 3 4.9 ± 0.5 2.2 ± 0.3
TX03_08 ditto 2003 3 5.0 ± 0.6 2.2 ± 0.2
TX04_16 Texas, Corpus Christi sorghum 2004 3 5.5 ± 0.5 2.2 ± 0.1
TX04_17 ditto 2004 1 – –
TX04_18 ditto 2004 3 5.5 ± 0.5 2.2 ± 0.1
SA04_02 SA, Potchefstroom ditto 2004 1 6.4 ± 0.6 2.9 ± 0.2
SA04_06 ditto 2004 2 – –
SA05_03 sorghum white 2005 3 5.4 ± 0.5 2.1 ± 0.3
SA05_04 ditto 2005 2 5.1 ± 0.4 2.0 ± 0.2
SA04_08 SA, Bethlehem sorghum 2004 1 6.6 ± 0.8 3.3 ±0.2
SA05_06 ditto 2005 2 5.6 ± 0.6 2.3 ± 0.2
SA05_13 SA, Platrand sorghum PAN 8706 W 2005 2 – –
SA05_14 ditto 2005 2 5.1 ± 0.5 2.4 ± 0.2
SA05_40 ditto 2005 3 6.8 ± 1.0 2.6 ± 0.3
SA05_41 ditto 2005 3 – –
SA05_44 sorghum NK283 2005 2 – –
SA05_15 SA, Delmas sorghum PAN 8706 W 2005 2 5.3 ± 0.6 2.1 ± 0.2
SA05_16 ditto 2005 2 5.2 ± 0.6 2.1 ± 0.2
SA05_17 ditto 2005 3 – –
SA05_18 ditto 2005 2 5.2 ± 0.7 2.1 ± 0.2
SA05_20 ditto 2005 2 4.9 ± 0.5 2.0 ± 0.2
SA05_21 ditto 2005 2 5.0 ± 0.6 1.8 ± 0.2
SA05_38 SA, Gladdedrif sorghum PAN 8446 2005 2 – –
SA05_01 SA, Cedara sorghum white 2005 2 – –
NM05_22 Namibia Pearl millet Kangara 2005 3 – –
NM05_23 ditto 2005 2 – –
NM05_24 ditto 2005 2
NM05_26 ditto 2005 1 6.1 ± 0.7 2.8 ± 0.3
NM05_28 Pearl millet Kantana 2005 3 – –
NM05_31 ditto 2005 2 4.9 ± 0.6 2.6 ± 0.2
NM05_32 ditto 2005 2 – –
NM05_33 ditto 2005 3 – –
NM05_34 Pearl millet Okashana 2005 2 5.2 ± 0.5 2.8 ± 0.3
NM05_35 ditto 2005 3 – –
NM05_42 ditto 2005 1 5.8 ± 0.6 2.6 ± 0.3

aIsolates from Africa (SA) were collected by N.W. McLaren (University of the Free State, Bloemfontein, South Africa), those from
Texas were collected by G.N. Odvody (Texas AgriLife Research and Extension Center, Corpus Christi, TX, USA).
bOf isolation. cIn µm; means ± SD; (–) – no sporulation observed.

Nuclear rDNA region containing internal transcribed spacers (ITS1, ITS2) and 5.8S region was ampli-
fied using primers ITS5 (White et al. 1990) and ITS4s (Kretzer et al. 1996). The reaction conditions were as
follows: 1 cycle of 3 min at 95 °C, 30 s at 55 °C and 1 min at 72 °C; 30 cycles of 30 s at 95 °C, 30 s at 55 °C and
1 min at 72 °C; 1 cycle 30 s at 95 °C, 30 s at 55 °C and 10 min at 72 °C. The reaction mixture consisted of
DynaZyme buffer, 1 U DynaZyme (both Finnzymes), 0.2 mmol/L deoxynucleotides, 30 pmol of each primer,
and 5–50 ng of DNA in a total of 30 μL. The amplicons were custom-sequenced at Macrogen (Korea) and
the representative sequences deposited at GenBank under accession numbers EU626191– EU626197 (Table II).
220 S. PAŽOUTOVÁ Vol. 54

Table II. Alleles of ITS-rDNA sequences obtained from Phoma sorghina isolates and from GenBank

Allele Organism Origin Location Accession no. Reference

1 SA05_01 EU626192 this study

SA05_06 ditto
SA05_22 ditto
SA05_26 ditto
SA05_38 ditto
SA05_42 ditto
TX03_01 ditto
TX03_06 ditto
TX03_07 ditto
Ampelomyces sp. Coffea arabica EF672292 Vega et al.,
Vega607 (as endophyte) unpublished
USA, Hawaii
dothideomycete, 113 Paspalum notatum USA, Florida EU680502 Feldman et al. 2008
dothideomycete, 7685 Nomophila nearctica USA, North EU680546 Feldman et al. 2008
Carolina
2 SA04_06 EU626191 this study
SA05_04 ditto
SA05_14 ditto
SA05_18 ditto
SA05_41 ditto
SA05_44 ditto
“Phoma herbarum” a Quercus aliena China, Shanxi AB369456 Tian C.M.,
unpublished
Ampelomyces sp. Po59 Podocarpus falcatus Ethiopia AY207323 Gure 2004
Ampelomyces sp. Po61 Podocarpus falcatus Ethiopia AY207317 Gure 2004
ascomycete. shz-102 Hippophae China EU682958 Sun et al.,
rhamnoides unpublished
dothideomycete, 001 shirasu soil China, Fujian EU828349 Wang et al.,
unpublished
Ampelomyces sp. A10 Aquilaria sinensis China, EU781667 Wang et al.,
(as endophyte) Guangdong unpublished

3 “Phoma herbarum” a Capsicum annuum Mexico EU082106 Rivera-Jimenez

strain M16 et al., unpublished
SA04_02 EU626195 this study
SA04_08 ditto
4 TX04_16 EU626193 ditto
TX04_17 ditto
Unique SA05_03 EU626197 ditto
TX03_02 EU626194 ditto
ascomycete sp. LM243 sand USA, Hawaii EF060576 Mahdi 2006
Phoma sorghina rice grains Denmark AF046022 Reddy et al. 1998
Ampelomyces sp. Po58 Podocarpus falcatus Ethiopia AY207322 Gure 2004
ascomycete sp. CE-2007 EU204641 Yanxiang et al.,
unpublished
fungus, uncultured soil USA, DQ420945 Waldrop et al. 2006
Minnesota
fungus SK3RW1M Avicennia marina China, EU791918 Pan et al.,
(as endophyte) GuangXi unpublished
Ampelomyces sp. Po62 Ethiopia AY513942 Gure 2004
ascomycete LM107 marine invertebrate USA, Hawaii EF060476 Mahdi 2006
ascomycete sp. RM6-1 Suberites zeteki DQ993634 Wang et al. 2008
dothideomycete, F6 Paspalum laeve USA, North EU680489 Feldman et al. 2008
Carolina
dothideomycete, 7687 Paspalum laeve EU680531 Feldman et al. 2008
(as endophyte)

aObvious misidentification of Phoma isolates.

Sequences similar to P. sorghina were obtained from GenBank (Table II and III). Two types of se-
quences were sought: those identical with sequences from any of the isolates studied or sharing the typical
deletion at the beginning of the ITS region, and those of related Phoma/Didymella species that originated from
2009 GENETIC VARIATION OF P. sorghina ISOLATES 221

collection isolates and/or were described in other phylogenetical studies of the taxon. Alignment and calcu-
lations of the identity matrices were performed in BioEdit (Hall 1999). The phylogenetic tree was recon-
structed using the maximum likelihood method implemented in PhyML (Guindon and Gascuel 2003). The
GTR substitution model with 6 categories was selected. Proportion of invariant sites (0.755) and -shape
parameter (0.525) were directly estimated from the data. The base frequencies (A, C, G, T) were 0.23119,
0.24425, 0.22397, and 0.30058, respectively. The values of the rate matrix (rAC, rAG, rAT, rCG, rCT, rGT)
were 19.77911, 28.67836, 8.60620, 60.70783, and 1.0, respectively. For tree search, the NNI method was
applied. Reliability for internal branches was assessed using the bootstrapping method (300 bootstrap repli-
cates) and aLRT non-parametric branch support based on a Shimodaira–Hasegawa-like procedure. All algo-
rithms used were implemented in PhyML Online (v3.0 aLRT) (Guindon et al. 2005) at the Phylogeny.fr
website.

Table III. ITS-rDNA sequence phylogenetically close to Phoma sorghina

Species Accession no. Reference

Ascochyta lentis AY131201 Ellwood and Oliver, unpublished

Ascochyta pisi EU338442 Davidson et al., unpublished
Ascochyta viciae-villosae EU167560 Simon et al., unpublished
Cerebella andropogonis AJ306620 Pažoutová and Kolínská 2003
Didymella applanata AJ428534 Lindqvist-Kreuze et al. 2003
Didymella fabae EU167566 Simon et al., unpublished
Didymella ligulicola AY157886 Pethybridge et al. 2004
Didymella phacae EU167570 Simon et al., unpublished
Didymella pinodes AY152551 Verkley et al. 2004
Didymella rabiei DQ383949 Peever et al. 2007
Didymella bryoniae AF297228 Goodwin et al. 2001
Epicoccum nigrum AJ853755 Meyer and Huynh, unpublished
Leptosphaerulina americana AY278318 Abler 2003
Leptosphaerulina trifolii AY131203 Ellwood and Oliver, unpublished
Phoma epicoccina AF149931 Arenal et al. 2000
Phoma exigua AF268190 Abeln et al. 2002
Phoma glomerata AF126819 Sullivan and White 2000
Phoma herbarum AY293791 Sullivan et al., unpublished
Phoma macrostoma DQ093700 Menkis et al. 2006
Phoma medicaginis DQ096584 Castell-Miller et al. 2008
Phoma pinodella EU167565 Simon et al., unpublished
Phoma sojicola EU167568 Simon et al., unpublished

The amount of genetic variation was calculated following three parameters: the percentage of poly-
morphic loci, gene diversity (HE) and Shannon Information Index. HE is equivalent to the proportion of loci
heterozygous per individual under Hardy–Weinberg expectations (= expected heterozygosity) and was cal-
culated by the unbiased method of Nei (1973) adjusting for the differences in sample size. Shannon’s Infor-
mation index (Lewontin 1972) was defined as I = –pi ln pi (i = 1,2) where pi was the frequency of the
presence or absence of a given rep-PCR band. In addition, GST (proportion of total variation that is distributed
among populations; for haploids it should be equal to FST) (Nei 1987) and Nm (estimate of gene flow derived
from GST) were calculated. All these calculations were performed using PopGene 1.32 (Yeh et al. 2000). Ana-
lysis of molecular variance (AMOVA) on allele frequencies was performed with the software Arlequin 3.11
(Excoffier et al. 2005) to test the partitioning of genetic variability. Population differentiation based on FST
values and its significance assessment were done by random permutation tests, also in Arlequin.

RESULTS

Among the fungi that grew out of surface-sterilized grains, Alternaria sp., Colletotrichum graminis,
C. gloeosporoides, Curvularia lunata and some Fusaria were most often encountered. Phoma-like fungi were
rather rare. These isolates were identified as Phoma sorghina and P. glomerata, one isolate of P. epicoccina
was also found.
222 S. PAŽOUTOVÁ Vol. 54

Morphology. The appearance of colonies of P. sorghina was very variable; therefore three media
were used to detect markers that might be shared by any of the regional or host-specific groups. The growth of
isolates was rapid on all media, cultures attaining 60–80 mm in diameter after 7 d.
The appearance of 14-d-old colonies on MEA was defined by the proportion of three features:
(a) dense gray-green felty aerial mycelium, (b) fluffy to funiculose white aerial mycelium overgrowing the
gray-green mycelial “layer”, (c) the formation of soluble pink to salmony pigment, often in clear droplets, on
the white aerial hyphae. A combination of these features led to dark gray-green felty colonies with missing
white overgrowth, gray-green colonies with white overgrowth and pinkish tufts, to colonies with sparse
whitish mycelium partially colored salmon, orange or reddish due to the soluble pigment production; the
gray-green felt was missing. In the latter cultures, a reddish soluble pigment was also excreted to agar.
No clear-cut difference between populations was observed; however, isolates from Namibia tended
to dark gray-green felty appearance with some white overgrowth and none to medium salmon pigment pro-
duction.
On the OA medium, aerial mycelium was cream to beige, later gradually darkening to grey green
from the center, with pinkish to orange exudate. Conidiation began after 10 d and was more intensive than on
the MEA medium. Pycnids occurred either single or in aggregates, immersed in agar as well as on the colony
surface. Unpigmented dictyochlamydospores as well as botryoid dark chlamydospores were formed on sub-
strate mycelium and also later on aerial hyphae. Most of the African isolates produced on OA conspicuous
dark radial substrate hyphae with numerous intercalated chlamydospores (Fig. 2) that were not observed in the
cultures of Texas isolates. Dark radial substrate hyphae resembled pigmented reverse at the first sight but no
soluble dark pigments were produced into agar.

Fig. 2. Colony morphology of P. sorghina morphotypes (1–3) differing by red pigment production on Czapek–Dox agar plates after a 7-d
incubation; upper row – obverse, lower row – reverse.

On Czapek–Dox agar, formation of red pigments was stimulated more than on OA (Fig. 3). In most
isolates that otherwise did not produce species-typical red pigments either on MEA or OA, at least pinkish
coloration of the reverse was observed. Mycelium was felty to funiculose, whitish, later salmon to red due to
extracellular pigment production, after 10–14 d acquiring a gradually darkening gray-green color. Sectored
colonies have been also found.
The isolates were sorted in three groups according to the intensity of the reverse coloration and
reddish exudate production attained on the 7th day of cultivation on Czapek–Dox agar (Table I). Isolates of the
2009 GENETIC VARIATION OF P. sorghina ISOLATES 223

morphotype 1 did not produce any pigment into the agar; none or a small amount of pink exudate was ob-
served on aerial hyphae. Isolates of morphotype 2 produced pale pink to pale salmon agar pigmentation, with
reddish exudate or red sectoring of the aerial
hyphae. In morphotype 3, abundant pigmentation
of the whole agar plate and intense coloration of
aerial hyphae by red exudate were observed.
Abundant red pigment production in the reverse
predominated in the Texas isolates; the African
ones were mostly low to medium producers.
Conidiation was not found in all iso-
lates; some formed only nonsporulating pycni-
dia-like structures. Therefore, the measurements
of conidial dimensions were made only in 24
isolates grown on OA (Table I). Conidia were
hyaline, unicellular, elliptic to oval. The mean
size and the size range between isolates varied.
Conidial dimensions found in P. sorghina were
somewhat smaller than stated in Boerema et al.
(2004), mostly averaging 4.5–5.5 µm in length.
Variability. Amplification of DNA from
41 isolates using two primers resulted in 53 re-
producible rep-PCR bands sized 300–3000 bp.
There were two monomorphic bands produced
with the primer M13core that were shared by all
isolates, but no such common bands were obser-
ved with primer ERIC I. Common banding pat-
terns were found only among location-related
isolates.
The UPGMA analysis of the Dice ma-
trix of rep-PCR data shows 65 % similarity
within the isolates (Fig. 4). The shared geogra- Fig. 3. Radial substrate hyphae from oatmeal agar cultures, typical
phic and host origin was partially reflected in the of African isolates, with numerous intercalated chlamydospores
and dark pigmentation; bar = 50 μm.
haplotype grouping and supported by a signifi-
cant bootstrap for two clades. Namibian pearl
millet isolates grouped together, with a single Bethlehem isolate. Isolates from Delmas were significantly
grouped with those of Platrand (135 km distance) but not with the one from 70 km distant Gladdedrif. In Texas
samples, the Beeville isolates from 2003 formed an unsupported monophyletic group containing two
supported subclades. The four isolates sampled at Corpus Christi (70 km distance from Beeville) in 2004 were
closer to South African isolates than to the Beeville ones. The last supported clade consisting of four isolates
was rather heterogeneous (Bethlehem and Potchefstroom from 2004, and Cedara and Gladdedrif from 2005).
Three different estimators of genetic diversity were applied: percentage of polymorphic loci, gene
diversity (as expected heterozygosity HE) (Nei 1973), and Shannon’s information index (I). Percentage of
polymorphic loci was high: 24–77 %. Values of HE were moderate, ranging from 0.073 to 0.229 in regional
populations (Table IV) and 0.256 when calculated over all populations and loci. The number of the pairwise
differences between isolates averaged to 9.61 ± 5.28. Higher diversity was found in the sorghum isolates (both

Table IV. Standard diversity indices of Phoma sorghina populations

Polymorphic Number of pairwise

Population Haplotypes HEa Shannon’s Ia
loci, % differencesa

Namibia 11 24.53 4.25 ± 2.28 0.073 ± 0.150 0.113 ± 0.219

South Africa 17 77.36 12.82 ± 6.05 0.229 ± 0.179 0.355 ± 0.249
Texas 11 52.83 11.75 ± 5.76 0.202 ± 0.206 0.298 ± 0.297
Total 39 96.23 9.61 ± 5.28 0.256 ± 0.152 0.405 ± 0.194

aValue ± SD.
224 S. PAŽOUTOVÁ Vol. 54

from Africa and Texas) than in the Namibian population from pearl millet. Only two pairs of clonal isolates
were found in the samples from the same field at Platrand and at Delmas.

Fig. 4. UPGMA dendrogram of P. sorghina isolates (haplotypes) based on Dice coefficient similarity matrix
obtained from rep-PCR data; bootstrap supports over 50 % are given at the nodes; isolates with the name in
bold have been sequenced; A1–A4 – alleles of ITS-rDNA.

Population differentiation. As in some locations only small numbers of isolates were found, these sub-
populations were pooled into three regional populations: South Africa, Namibia and Texas.
Gene flow and genotypic differentiation between regions was estimated using a coefficient of gene
differentiation (GST) and Nm (Table V). The values of GST were intermediate; 16–37 % of the variation was
2009 GENETIC VARIATION OF P. sorghina ISOLATES 225

found between the populations. Values for Nm showed enough of gene flow even over long distances to over-
come genetic differentiation (Wright 1931). The Nm values were lowest between Namibian pearl millet isolates
and Texas sorghum isolates. Population differentiation based on pairwise FST was significant (p < 0.0001).

Table V. Population differentiation and gene flow between Phoma sorghina populations

Populationsa Genetical distanceb FSTc GST Nm

NM × SA 0.158 0.391* 0.303 1.15

SA × TX 0.097 0.211* 0.156 2.70
NM × TX 0.199 0.498* 0.373 0.84

aNM – Namibia; SA – South Africa; TX – Texas. bUnbiased (Nei 1987).

cCalculated using Arlequin; *p < 0.001.

Analysis of molecular variance (AMOVA) (Table VI) between the three regional populations has
shown that 65 % of the total variation was due to within-population differences and 35 % of the variation was
due to differences among populations (FST = 0.349, p = 0.01). When the analysis was done between popula-
tions from sorghum (pooled) and pearl millet, the results were similar: 66 % of the variation was caused by dif-
ferences within the host and 34 % was due to between-host differences (FST = 0.338, p = 0.01). In comparison
of pooled African vs. Texas population, the within-population difference was the highest, 78 %, whereas bet-
ween these populations only 22 % of variation was found.

Table VI. Analysis of molecular variance

Sum Variance Percentage

Source df p
of squares component of variation

Among regional populations 2 82.72 2.75 34.90 0.00

Within populations 38 195.47 5.14 65.10
Total 40 278.20 7.90

Between host plants 1 53.22 2.95 33.82 <0.01

Within hosts 39 224.97 5.77 66.18
Total 40 278.20 8.72

Between Africa and Texas 1 34.13 1.73 21.67 <0.01

Within populations 39 244.06 6.26 78.32
Total 40 278.20 8.00

rDNA analysis. Relationship among isolates was also examined using ITS-rDNA. Of the 21 repre-
sentative sequences, allele 1 was found in 9 isolates scattered throughout the phylogram (Fig. 5). Allele 2
occurred in 6 isolates, 5 of them from the Platrand–Delmas clade, one from a more distant Potchefstroom
isolate. Alleles 3 and 4 were shared by two isolates each and two sequences were unique. Only alleles 1 and 2
were found among sequences deposited in GenBank (3 and 7, respectively) (cf. Table II).
Sequences from the database that formed a clade with seed isolates of P. sorghina (P. sorghina com-
plex on Fig. 5) were obtained from a wide spectrum of endophytic, epiphytic, sea and soil isolates or uncul-
tured fungi (Table II). All shared an 8-bp deletion starting at 38th base from the 3´ end of 5.8S rDNA that was
not found in any other Phoma/Didymella species in the alignment and which is a typical marker for the
P. sorghina complex.
The rDNA alignment had 478 positions, 51 positions were variable and 35 positions were phylo-
genetically informative; for resolving relationships inside the P. sorghina complex, only 12 informative sites
remained. Each allele was represented only once. The sequence identity inside the P. sorghina clade (sup-
ported by 82 % bootstrap and 96 % aLRT) was 98–99.7 %, but even the outgroup species Didymella bryoniae
and D. ligulicola were still no less than 92 % identical with the other species and isolates.
Among the species closely related to P. sorghina (forming a clade with 51 % bootstrap and 81 %
aLRT support), all members of the “Ascochyta complex” of legume pathogens described by Peever et al.
(2007) were found. Also P. glomerata, Didymella applanata, Epicoccum spp., two species of legume-associ-
226 S. PAŽOUTOVÁ Vol. 54

ated Leptosphaerulina, P. macrostoma and P. herbarum belonged to this group. Inside the P. sorghina com-
plex, no significant bootstrap supports were found at any node, but aLRT values were comparable with those
found on bootstrap-supported nodes.

Fig. 5. Dendrogram constructed from ITS-rDNA sequences of P. sorghina isolates and their closest
neighbors from the GenBank databases using PhyML; bootstrap values (300 replicates) higher than 50 %
are shown over the branches in bold, the aLRT supports over 70 % are given under the branches.

DISCUSSION

Except for more intense formation of red pigments in the isolates from Texas and the dark radial
substrate hyphae with intercalated chlamydospores in the African isolates, no other group- or region-specific
morphological marker has been found. The wide range of conidia sizes was found across all regions and hosts.
From the two supposed teleomorphs of P. sorghina, Didymella holci is the more probable candidate; its coni-
2009 GENETIC VARIATION OF P. sorghina ISOLATES 227

dial dimensions are closer to those observed here for P. sorghina. The fungus also produces a pink pigment
into corn-meal agar (Anahosur and Sivanesan 1978) resembling the P. sorghina pigmentation. Moreover, all
species shown here to be phylogenetically close to P. sorghina belonged to the Phoma/Didymella group which
was in a recent phylogenetic study of de Gruyter et al. (2009) redefined as a newly erected family Didymel-
laceae in the order Pleosporales.
The relatedness of isolates was not always correlated with their geographical distance. Isolates from
locations Platrand and Delmas clustered together, but isolate from the neighboring Gladdedrif differed. Pot-
chefstroom and Bethlehem isolates were close despite the locations being distant. On the other hand, Beeville
isolates from 2003 were quite different from those found in Corpus Christi in 2005. As the fungus clearly sur-
vives the surface treatment of seeds, the similarity might reflect grain transfer more than spreading by natural
means.
Gene diversity was moderate and only two pairs of clonal isolates were found each in samples from
the same field. Certain host preferences might be suggested based on most of pearl millet isolates from Namibia
grouping together. However, a sorghum isolate from Bethlehem grouped with millet isolates.
The partitioning of genetic variation showed that most variation (63–78 %) occurred within popula-
tions defined according to regions, host or continent (due to the high number of distinct haplotypes); 22–35 %
of the variation was caused by differences among regional populations. More homogeneous, often clonal,
“epidemic” population structure (Maynard Smith et al. 2000) that results from rapid spreading of invasive
clones replacing native populations was not found in P. sorghina; this may also reflect the absence of any
selection pressure in the population. The population continuity at the same location might be interrupted by
a dry season(s), unfavorable for grain mould development. However, P. sorghina can persist as a soil fungus
which may preserve a part of the original haplotypes in the field from one growing season to another. Re-
cently, P. sorghina was isolated as a causal agent of maize leaf spot in Brazil (Amaral et al. 2004) which may
represent yet another source of inoculation from plant debris. These factors can enhance haplotype diversity at
a location.
The P. sorghina populations showed differentiation but also some degree of genetic exchange even
among continents. South Africa sorghum population exchanges genes with both Texas sorghum and Namibia
pearl millet populations. No evidence for a presence of specific lineages or cryptic speciation has been found.
Values of FST (Arlequin) and GST (PopGene) differed (although their proportions were identical) which may
be caused by different assumptions of the algorithms. The Arlequin algorithm assumes that the data are co-
dominant and haplotypic whereas in PopGene algorithm, codominant markers and Hardy–Weinberg equi-
librium is assumed. However, with the present haploid data set neither the codominance nor the equilibrium
could be tested, also due to the small amount of isolates from the same location.
Small or no intraspecific differences in ITS-rDNA sequences in combination with higher diversity
observed by DNA fingerprinting methods were found also in other Phoma species. Castell-Miller et al. (2008)
analyzed P. medicaginis isolates and found no ITS-rDNA differences, mean HE = 0.23 and 55 % similarity as
measured by the Dice coefficient. These observations are very close to those presented here for P. sorghina
(HE = 0.256 and 65 %). However, the FST between northern and southern P. medicaginis populations from
Minnesota was only 1.99 % and the population differentiation was not significant. Abeln et al. (2002) found in
P. exigua a different situation in that the ITS-rDNA sequences of the host-specific varieties were highly similar,
the AFLP patterns of these varieties differed, but inside the variety the patterns were very close. Peever et al.
(2007) found nearly identical ITS-rDNA sequences among legume-associated Ascochyta/Didymella species,
whereas other genes clearly delimited clusters of host-specific varieties.
Alleles of GenBank ITS-rDNA sequences closest to P. sorghina have been obtained mostly from endo-
phytes of various plants and trees, present in stems and seeds. Closely related sequences were found in endo-
phytes but also in soil fungi and even in isolates associated with marine organisms. With the limited amount of
knowledge it is not possible to decide now, if all these fungi belong to one ubiquitous species, or if there is
a subspecies structure. At least, the presence of the ITS-rDNA alleles 1 and 2 (first detected in the isolates
from grains) in various plant endophytes from distant locations suggests, that P. sorghina may be a univer-
salist species.

This work was supported by the Institutional Research Concept no. AV 0Z 5020 903 and by the INTSORMIL (Sorghum, Mil-
let and Other Grains CRSP, USAID). Thanks are due to Prof. N.W. McLaren and Prof. G.N. Odvody for kindly providing grain samples.

REFERENCES

ABELN E.C.A., STAX A.M., DE GRUYTER J., VAN DER AA H.A.: Genetic differentiation of Phoma exigua varieties by means of AFLP
fingerprints. Mycol.Res. 106, 419–427 (2002).
ABLER S.W.: Ecology and Taxonomy of Leptosphaerulina spp. Associated with Turfgrasses in the United States. Virginia Polytechnic
Institute and State University, Blacksburg (USA) 2003.
228 S. PAŽOUTOVÁ Vol. 54

AMARAL A.L.D., CARLI M.L.D., NETO J.F.B., SOGLIO F.K.D.: Phoma sorghina, a new pathogen associated with phaeosphaeria leaf spot
on maize in Brazil. Plant Pathol. 53, 259 (2004).
ANAHOSUR K.H., SIVANESAN A.: Mycosphaerella holci. CMI Descriptions of Fungi and Bacteria (No. 59). Sheet 584 (1978).
ARENAL F., PLATAS G., MONTE E., PELAEZ F.: ITS sequencing support for Epicoccum nigrum and Phoma epicoccina being the same
biological species. Mycol.Res. 104, 301–303 (2000).
ARORA D.K., HIRSCH P.R., KERRY B.R.: PCR-based molecular discrimination of Verticillium chlamydosporium isolates. Mycol.Res. 100,
801–809 (1996).
VON ARX J.A.: Plant pathogenic fungi. Nova Hedwigia Beihefte 87, 1–288 (1987).
AVESKAMP M.M., DE GRUYTER J., CROUS P.W.: Biology and recent developments in the systematics of Phoma, a complex genus of
major quarantine significance. Fungal Diversity 31, 1–18 (2008).
BOEREMA G.H., DE GRUYTER J., NOORDELOOS M.E., HAMERS M.E.C.: Phoma Identification Manual: Differentiation of Specific and
Infra-Specific Taxa in Culture. CABI Publishing, Wallingford (UK) 2004.
CASTELL-MILLER C.V., SZABO L.J., GALE L.R., O’NEILL N.R., SAMAC D.A.: Molecular variability of a Minnesota population of Phoma
medicaginis var. medicaginis, the causal agent of spring black stem and leaf spot of alfalfa. Can.J.Plant Pathol. 30, 85–96
(2008).
EXCOFFIER L., LAVAL G., SCHNEIDER S.: Arlequin ver. 3.0: an integrated software package for population genetics data analysis. Evol.
Bioinf.Online 1, 47–50 (2005).
FARIS MOKAIESH S., BOCCARA M., DENIS J.B., DERRIEN A., SPIRE D.: Differentiation of the “Ascochyta complex” fungi of pea by bio-
chemical and molecular markers. Curr.Genet. 29, 182–190 (1996).
FELDMAN T.S., O’BRIEN H.E., ARNOLD A.E.: Moths that vector a plant pathogen also transport endophytic fungi and mycoparasitic anta-
gonists. Microb.Ecol. 56, 742–750 (2008).
FELSENSTEIN J.: Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39, 783–791 (1985).
FORBES G.A., BANDYOPADHYAY R., GARCIA G.: A review of sorghum grain mold, in W.A.J. de Milliano, R.A. Frederiksen, G.D. Bengston,
Eds): Sorghum and Millets Diseases: a Second World Review. ICRISAT, Pantacheru (Andhra Pradesh, India) 1992.
GOODWIN S.B., DUNKLE L.D., ZISMANN V.L.: Phylogenetic analysis of Cercospora and Mycosphaerella based on the internal transcribed
spacer region of ribosomal DNA. Phytopathology 91, 648–658 (2001).
DE GRUYTER J., AVESKAMP M.M., WOUDENBERG J.H.C., VERKLEY G.J.M., GROENEWALD J.Z., CROUS P.W.: Molecular phylogeny of
Phoma and allied anamorph genera: towards a reclassification of the Phoma complex. Mycol.Res. 113, 508–519 (2009).
GUINDON S., GASCUEL O.: A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Syst.Biol. 52,
696–704 (2003).
GUINDON S., LETHIEC F., DUROUX P., GASCUEL O.: PHYML Online – a web server for fast maximum likelihood-based phylogenetic
inference. Nucl.Acids Res. 33, W557–W559 (2005).
GURE A.: Seed-Borne Fungi of the Afromontane Tree Species Podocarpus falcatus and Prunus africana in Ethiopia. PhD Thesis. Swedish
University of Agricultural Sciences, Uppsala 2004.
HALL T.A.: BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucl.Acids
Symp.Ser. 41, 95-98 (1999).
HAMPL V., PAVLÍČEK A., FLÉGR J.: Construction and bootstrap analysis of DNA fingerprinting-based phylogenetic trees with a freeware
program FreeTree: application to trichomonad parasites. Internat.J.Syst.Evol.Microbiol. 51, 731–735 (2001).
HEALY A., REECE K., WALTON D., HUONG J., SHAH K., KONTOYIANNIS D.P.: Identification to the species level and differentiation
between strains of Aspergillus clinical isolates by automated repetitive-sequence-based PCR. J.Clin.Microbiol. 42, 4016–4024
(2004).
HEALY M., REECE K., WALTON D., HUONG J., FRYE S., RAAD I.I., KONTOYIANNIS D.P.: Use of the DiversiLab System for species and
strain differentiation of Fusarium species isolates. J.Clin.Microbiol. 43, 5278–5280 (2005).
KRETZER A., LI Y., SZARO T.M., BRUNS T.D.: Internal transcribed spacer sequences from 38 recognized species of Suillus sensu lato:
phylogenetic and taxonomic implications. Mycologia 88, 776–785 (1996).
LEWONTIN R.C.: The apportionment of human diversity. Evolutionary Biol. 6, 381–398 (1972).
LIECKFELDT E., MEYER W., BÖRNER T.: Rapid identification and differentiation of yeasts by DNA and PCR fingerprinting. J.Basic
Microbiol. 33, 413–426 (1993).
LINDQVIST-KREUZE H., HELLQVIST S., KOPONEN H., VALKONEN J.P.T.: Phoma–Didymella complex on hybrid arctic bramble with
wilting symptoms. Plant Pathol. 52, 567–578 (2003).
MAHDI L.: A Survey of Hawaiian Marine Fungi and Yeast. MSc Thesis. University of Hawai‘i, Manoa 2006.
MAYNARD SMITH J., FEIL E.J., SMITH N.H.: Population structure and evolutionary dynamics of pathogenic bacteria. Bioessays 22,
1115–1122 (2000).
MENKIS A., VASILIAUSKAS R., TAYLOR A.F.S., STENSTROM E., STENLID J., FINLAY R.: Fungi in decayed roots of conifer seedlings in
forest nurseries, afforested clear-cuts and abandoned farmland. Plant Pathol. 55, 117–129 (2006).
NEI M.: Analysis of gene diversity in subdivided populations. Proc.Nat.Acad.Sci.USA 70, 3321–3323 (1973).
NEI M.: Molecular Evolutionary Genetics. Columbia University Press, New York 1987.
PAŽOUTOVÁ S., KOLÍNSKÁ R.: Cerebella relationship to Epicoccum and their closest relatives among Dothideales. Czech Mycol. 54,
155–160 (2003).
PEEVER T.L., BARVE M.P., STONE L.J., KAISER W.J.: Evolutionary relationships among Ascochyta species infecting wild and cultivated
hosts in the legume tribes Cicereae and Vicieae. Mycologia 99, 59–77 (2007).
PETHYBRIDGE S.J., SCOTT J.B., HAY F.S.: Genetic relationships among isolates of Phoma ligulicola from pyrethrum and chrysanthemum
based on ITS sequences and its detection by PCR. Australas.Plant Pathol. 33, 173–181 (2004).
RABIE C.J., VAN RENSBURG S.J., VAN DER WATT J.J., LUBBEN A.: Onyalai – the possible involvement of a mycotoxin produced by Phoma
sorghina in the aetiology. South Afr.Med.J. 57, 1647–1650 (1975).
REDDY P.V., PATEL R., WHITE J.F. Jr.: Phylogenetic and developmental evidence supporting reclassification of cruciferous pathogens
Phoma lingam and Phoma wasabiae in Plenodomus. Can.J.Bot. 76, 1916–1922 (1998).
SIMON U.K., WEISS M.: Intragenomic variation of fungal ribosomal genes is higher than previously thought. Mol.Biol.Evol. 25, 2251–
2254 (2008).
2009 GENETIC VARIATION OF P. sorghina ISOLATES 229

SOMAI B.M., DEAN R.A., FARNHAM M.W., ZITTER T.A., KEINATH A.P.: Internal transcribed spacer regions 1 and 2 and random amplified
polymorphic DNA analysis of Didymella bryoniae and related Phoma species isolated from cucurbits. Phytopathology 92,
997–1004 (2002).
DE SOUZA N., ZAMBOLIM L., THIÈBAUT J.T.L.: Variabilidade de isolados de Phoma sorghina em arroz. Pesq.Agropec.Bras. 23, 1139–
1141 (1988).
DE SOUZA BORGES W., TALLARICO PUPO M.: Novel anthraquinone derivatives produced by Phoma sorghina, an endophyte found in
association with the medicinal plant Tithonia diversifolia (Asteraceae). J.Braz.Chem.Soc. 17, 929–934 (2006).
SULLIVAN R.F., WHITE J.F.: Phoma glomerata as a mycoparasite of powdery mildew. Appl.Environ.Microbiol. 66, 425–427 (2000).
TYMON A.M., PELL J.K.: ISSR, ERIC and RAPD techniques to detect genetic diversity in the aphid pathogen Pandora neoaphidis.
Mycol.Res. 109, 285–293 (2005).
VENKATASUBBAIAH P., VAN DYKE C., CHILTON W.: Phytotoxic metabolites of Phoma sorghina, a new foliar pathogen of pokeweed.
Mycologia 84, 715–723 (1992).
VERKLEY G.J.M., STARINK-WILLEMSE M., VAN IPEREN A., ABELN E.C.A.: Phylogenetic analyses of Septoria species based on the ITS
and LSU-D2 regions of nuclear ribosomal DNA. Mycologia 96, 558–571 (2004).
VERSALOVIC J., KOEUTH T., LUPSKI R.: Distribution of repetitive DNA sequences in eubacteria and application to fingerpriting of bac-
terial genomes. Nucl.Acids Res. 19, 6823–6831 (1991).
WALDROP M.P., ZAK D.R., BLACKWOOD C., CURTIS C.D., TILMAN D.: Resource availability controls fungal diversity across a plant
diversity gradient. Ecol.Lett. 9, 1127–1135 (2006).
WANG G., LI Q., ZHU P.: Phylogenetic diversity of culturable fungi associated with the Hawaiian sponges Suberites zeteki and Gelliodes
fibrosa. Antonie van Leeuwenhoek 93, 163–174 (2008).
WHITE J.F. Jr., MORGAN-JONES G.: Studies in the genus Phoma. II. Concerning Phoma sorghina. Mycotaxon 18, 5–13 (1983).
WHITE T.J., BRUNS T., LEE S., TAYLOR J.: Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics, pp.
315–322 in M.A. Innis, D.H. Gelfand, J.J. Sninsky, T.J. White (Eds): PCR Protocols: a Guide to Methods and Applications.
Academic Press, San Diego (USA) 1990.
WRIGHT S.: Evolution in Mendelian populations. Genet.Mol.Biol. 16, 97–159 (1931).
YEH F.C., YANG R., BOYLE T.J., YE Z., XIYAN J.M.: PopGene32, Microsoft Windows-Based Freeware for Population Genetic Analysis.
1.32. Molecular Biology and Biotechnology Centre, University of Alberta, Edmonton (Alberta, Canada) 2000.
YOSHIOKA K.: KyPlot – a user-oriented tool for statistical data analysis and visualization. Comp.Stat. 17, 425–437 (2002).