A Novel Fusarium Species Causes a Canker Disease of the Critically Endangered Conifer, Torreya taxifolia

Abstract

A canker disease of Florida torreya (Torreya taxifolia) has been implicated in the decline of this critically endangered species in its native ranee of northern Florida and southeastern Georgia. In surveys of eight Florida torreya sites, cankers were present on all dead trees and 71 to 100% of living trees, suggesting that a fungal pathogen might be the causal agent. To identify the causal agent, nuclear ribosomal internal transcribed spacer region (ITS rDNA) sequences were determined for 115 fungi isolated from cankers on 46 symptomatic trees sampled at three sites in northern Florida. BLASTn searches of the GenBank nucleotide database, using the ITS rDNA sequences as the query, indicated that a novel Fusarium species designated Fsp-1 might be the etiological agent. Molecular phylogenetic analyses of partial translation elongation factor I-alpha (EF-1) and RNA polymerase second largest subunit (RPB2) gene sequences indicate that Fsp-1 represents a novel species representing one of the earliest divergences within the Gibberella clade of Fusarium. Results of pathogenicity experiments established that the four isolates of Fsp-1 tested could induce canker symptoms on cultivated Florida torreya in a growth chamber. Koch's postulates were completed by the recovery and identification of Fsp-1 from cankers of the inoculated plants.

Full text

Plant Disease / June 2011 633
A Novel Fusarium Species Causes a Canker Disease of the Critically
Endangered Conifer, Torreya taxifolia
Jason A. Smith, School of Forest Resources and Conservation, University of Florida, Gainesville; Kerry O’Donnell, Bacterial Food-
borne Pathogens and Mycology Research Unit, Agricultural Research Service, U.S. Department of Agriculture, Peoria, IL; Lacey L.
Mount, Department of Plant Pathology, University of Florida, Gainesville; Keumchul Shin, Kelly Peacock, and Aaron Trulock,
School of Forest Resources and Conservation, University of Florida, Gainesville; Tova Spector, Florida Park Service, Department of
Environmental Protection, Panama City; and Jenny Cruse-Sanders and Ron Determann, Atlanta Botanical Garden, Atlanta, GA
Abstract
Smith, J. A., O’Donnell, K., Mount, L. L., Shin, K., Peacock, K., Trulock, A., Spector, T., Cruse-Sanders, J., and Determann, R. 2011. A novel
Fusarium species causes a canker disease of the critically endangered conifer, Torreya taxifolia. Plant Dis. 95:633-639.
A canker disease of Florida torreya (Torreya taxifolia) has been impli-
cated in the decline of this critically endangered species in its native
range of northern Florida and southeastern Georgia. In surveys of eight
Florida torreya sites, cankers were present on all dead trees and 71 to
100% of living trees, suggesting that a fungal pathogen might be the
causal agent. To identify the causal agent, nuclear ribosomal internal
transcribed spacer region (ITS rDNA) sequences were determined for
115 fungi isolated from cankers on 46 symptomatic trees sampled at
three sites in northern Florida. BLASTn searches of the GenBank nu-
cleotide database, using the ITS rDNA sequences as the query, indi-
cated that a novel Fusarium species designated Fsp-1 might be the
etiological agent. Molecular phylogenetic analyses of partial transla-
tion elongation factor 1-alpha (EF-1) and RNA polymerase second
largest subunit (RPB2) gene sequences indicate that Fsp-1 represents a
novel species representing one of the earliest divergences within the
Gibberella clade of Fusarium. Results of pathogenicity experiments
established that the four isolates of Fsp-1 tested could induce canker
symptoms on cultivated Florida torreya in a growth chamber. Koch’s
postulates were completed by the recovery and identification of Fsp-1
from cankers of the inoculated plants.
Canker disease of Florida torreya, known as gopherwood or
stinking cedar (Torreya taxifolia Arn.), hereafter referred to as
CDFT, appears to have contributed to this plant being listed as
critically endangered by the U.S. Fish and Wildlife Service. Cur-
rently, this endemic taxaceous conifer is restricted to bluffs and
ravines along the Apalachicola River in Gadsden and Liberty coun-
ties in Florida and Decatur County in Georgia (16). T. taxifolia is
considered the rarest conifer in North America and one of the most
endangered species in the world (4).
Although the decline of Florida torreya was first observed in the
late 1930s (1), the tree was still common in its habitat in northern
Florida and southeastern Georgia through the 1950s. The rapid
decline of the species in the early 1900s was initially attributed to
an unknown fungal disease based on the abundance of leaf spots
and stem cankers (7). Due to decline, and to the lack of seed-bear-
ing trees (1), Florida torreya was considered to be destined for
extinction (7). In addition to the decline, this species has been
negatively impacted by changes in hydrology, forest structure,
heavy deer browse, and a loss of reproductive capability (17). Flor-
ida torreya stems killed by disease often re-sprout in a manner
reminiscent of American chestnut following Chestnut Blight. Flor-
ida torreya has declined by more than 99% over the past century
from an estimated population of 357,500 individuals in 1914 to
approximately 1,350 in the 1990s (19,20), to current estimates of
400 to 600 individuals (T. Spector, personal communication). Trees
in their native range have not reproduced from seed for several
decades (18). Despite several attempts to conclusively determine
the causal agent of Florida torreya decline, disease etiology has not
been elucidated (1–3,8,22). In the first pathology studies conducted
on T. taxifolia (1), it was noted that leaf spots, needle necrosis,
defoliation, and stem lesions were common on native and culti-
vated T. taxifolia. Several pathogens were commonly isolated from
symptomatic needles (Macrophoma sp., Rhizoctonia solani,
Sphaeropsis sp., and Sclerotium rolfsii); however, no pathogens
were isolated from cankered stems and Koch’s postulates were not
completed. Subsequently, El-Gholl (3) reported Fusarium laterit-
ium as a causal agent by demonstrating this species’ capacity to
cause leaf spots. Alfieri et al. (2) completed Koch’s postulates with
F. lateritium as a leaf spot pathogen, but did not address whether
this species could induce the canker disease. While Schwartz et al.
(22) implicated Pestalotiopsis microspora as the causal agent of the
canker disease, no information was given on canker development,
morphology, or ability to cause mortality. Artificial inoculations
using P. m ic ro sp or a resulted in stem canker development (8), but
stem mortality was not observed. These reports are considered to
be inconclusive given that Pestalotiopsis spp. are considered to be
weak opportunistic pathogens (23). Subsequent studies implicated
a Scytalidium sp. due to frequent isolation from cultivated and
naturally occurring Florida torreya. Artificial inoculations resulted
in small lesions on needles, but cankers were not observed.
In addition to biotic causes of decline, several studies have re-
ported on changes in soils, drought, global warming, sunlight
exposure, and fire regime as possible causes of decline (21). Some
of these environmental changes are thought to have occurred be-
cause of the building of the Woodruff Dam along the Apalachicola
River in 1957, and changing land uses in the surrounding areas.
However, none of these environmental hypotheses has been
demonstrated as a cause of the decline.
Current efforts to manage this endangered species have been
hindered by a lack of understanding of the current and historic
causes of disease of Florida torreya. As a result, various agencies
have taken different approaches to manage Florida torreya depend-
ing on which cause the decline is attributed to. For these reasons,
more information is needed about the etiology of CDFT in order to
develop sound management practices. Accordingly, the present
study was conducted to: (i) assess canker incidence among natural
populations of T. taxifolia, and (ii) identify the causal agent of
CDFT.
Corresponding author: Jason A. Smith, E-mail: jasons@ufl.edu
Accepted for publication 26 January 2011.
doi:10.1094 / PDIS-10-10-0703
© 2011 The American Phytopathological Society

634 Plant Disease / Vol. 95 No. 6
Materials and Methods
Field surveys. During 2008 to 2010, surveys were conducted for
all living T. taxifolia on public and private lands in Gadsden and
Liberty counties in Florida where this species had been reported
previously. Surveys were conducted using historical maps/data,
personal knowledge of the sites, and visual searching. Once lo-
cated, individual trees were measured for height and diameter at
ground line (DGL). Due to stem heights commonly <1 m, diameter
at breast height could not be measured for most specimens. The
condition of each tree and occurrence of stem cankers was re-
corded. Position on slope, soil conditions, associated flora, occur-
rence of leaf spots and canopy cover/light intensity were also re-
corded. For canker incidence, site means were analyzed using one-
way analysis of variance (ANOVA) and Duncan’s multiple range
test (P < 0.05) (SAS ver. 9.1).
Isolation and tentative identification of fungi. Stem tissue
from the margins of 150 cankers were collected in the field from
46 diseased trees collected from three locations and returned to the
laboratory. Samples were cut to approximately 10 mm2 and surface
sterilized in 5% sodium hypochlorite for 30 s, followed by a 10-s
rinse in sterile H2O. The samples were placed on 2% potato dex-
trose agar (PDA; DIFCO, Detroit, MI) and incubated at room tem-
perature for 5 to 7 days. Fungal colonies were then subcultured on
PDA for subsequent identification and inoculation experiments.
A total of 129 isolates were used for identification based on
ribosomal internal transcribed spacer region (ITS-rDNA) se-
quences. DNA was extracted using the Qiagen DNeasy Plant Mini
kit (Qiagen Inc., Valencia, CA) following the manufacturer’s in-
structions except that samples were ground using liquid nitrogen
and a mortar and pestle prior to extraction, and the 65°C incubation
step was increased to 1 h. PCRs using the universal primers ITS1
and ITS4 (26) were performed with the following reaction mixture:
1 µl of diluted (1:100) template DNA, 1 µl of each primer (10 µM),
9.5 µl of ddH2O, and 12.5 µl of Amplitaq Gold Master Mix (Ap-
plied Biosystems, Emeryville, CA). PCRs were performed in a MJ
Mini thermocycler (BioRad Inc., Hercules, CA) with the following
thermocycling profile: 94°C for 5 min followed by 35 cycles of
94°C for 1 min, 55°C for 1 min, 72°C for 1 min, followed by a
final extension step at 72°C for 5 min. PCR amplicons were visual-
ized on a 1.5% agarose gel and were purified prior to sequencing
using the EXOSAPit kit (USB Corp., Cleveland, OH). Amplicons
were sequenced at the University of Florida Interdisciplinary Cen-
ter for Biotechnology Research using an ABI3730 DNA Analyzer
(Applied Biosystems Inc., Foster City, CA). Forward and reverse
sequences were edited and contigs were aligned using ChromasPro
ver. 1.5 software (Technelysium Inc., Tewantin QLD, Australia).
Edited sequences were used for BLASTn searches of the GenBank
nucleotide database (http://www.ncbi.nlm.nih.gov/). Fungi were
identified based on top BLAST results (lowest e-value, highest
score, and greatest similarity).
Inoculation experiments. Seven of the most frequently isolated
fungi (Botryosphaeria obtusa [Bo-Tt1], Fusarium cf. lateritium
[Fl-Tt1], F. solani species complex [Fs-Tt1], Fusarium sp. 1 [Fsp-
1], Fusarium sp. 2 [Fsp-2], Lasiodiplodia theobromae, and Pes-
talotiopsis sp.) were used in inoculation experiments with potted,
seed-grown T. taxifolia provided by the Atlanta Botanical Garden.
For all three inoculation experiments (Table 1), inoculation points
were made on each plant by making a wound using a sterile single-
edge blade to make a vertical slit under the bark approximately 10
× 5 mm in size. The inoculum, a mycelium plug approximately 5
mm2, was inserted into the wound, which was wrapped in Parafilm.
For mock-inoculated plants, wounds were made in the same man-
ner, but sterile PDA plugs were used instead of mycelium plugs.
The plants were maintained in a growth chamber with 16-h day
length and temperatures of 25°C (light) and 18°C (dark) for 6
months following inoculation. For inoculation experiment 1 (IE1),
one isolate of all species were used to inoculate the stems of one
plant each of 2-year-old seedlings (average height approximately
25 cm). One mock-inoculation was done on a separate plant as a
control (Table 1). Each inoculation point was made on each plant
at approximately 5 cm above the ground line. The Parafilm was
removed after 2 weeks, and plants were monitored for symptom
development at weekly intervals. Cankers were measured 3 months
following inoculation, and percent stem circumference cankered
was determined and mortality was measured after 6 months.
In inoculation experiment 2 (IE2), three different isolates of Fsp-1
were used (NRRL 54152 = 542, NRRL 54154 = 587, and NRRL
54155 = 596) to verify results obtained in IE1. Despite causing a
canker in experiment IE1, FSSC isolate Fs-Tt1 was not used in
subsequent experiments because it was infrequently isolated from
cankers in the field (<1%) and was not considered a likely candi-
date causal agent. All of the methods were the same, except a total
of six plants were inoculated with two replicates per isolate. One
mock-inoculation on a separate plant was included as a control
(Table 1).
Inoculation experiment 3 (IE3) was performed using plants of
two different sizes: 2-year-old seedlings as used before (average
height = 27 cm; average stem diameter = 0.6 cm), and larger sap-
lings, approximately 5 years old (average height of 89 cm and
average stem diameter of 1.2 cm). The larger saplings approximate
the size of the trees re-sprouting and displaying canker symptoms
in the field. Seven different isolates of Fsp-1 were used to inoculate
plants (Torreya isolates listed in Table 2). Each isolate was inocu-
lated onto one plant of the smaller size and two plants each of the
larger size with three inoculation points on each plant for a total of
nine inoculations per isolate. Two plants, one of each size, were
mock inoculated (three inoculations per plant as described above)
and served as controls (Table 1). Inoculation points on the smaller
plants were distanced approximately 3 cm apart on the stems start-
ing at 2.5 cm above ground line (and spiraling up the stem) and 5
cm apart on the larger plants starting at 10 cm above ground line.
In addition to percent stem circumference girdled for each canker,
mortality was assessed after 6 months.
Molecular phylogenetics. Isolates of Fsp-1 were cultured in
yeast-malt broth for 2 to 3 days at 24°C, after which total genomic
DNA was isolated from freeze-dried mycelium using a hexadecyl-
trimethyl-ammonium bromide (Sigma-Aldrich, St. Louis, MO)
protocol published previously (10). Seven isolates of the novel
Fusarium sp.-1 (Fsp-1) of T. taxifolia (NRRL 54149-54155; Table
2) were characterized genetically by analyzing DNA sequences of
the nuclear ribosomal internal transcribed spacer region (ITS
rDNA), and partial sequences of translation elongation factor (EF-
1
α
) and RNA polymerase second largest subunit (RPB2). The ITS
rDNA region was PCR amplified and sequenced with primers
ITS1> TCCGTAGGTGAACCTGCGG and ITS4> TCCTCC
GCTTATTGATATGC; EF-1
α
was PCR amplified with primers
EF-1> ATGGGTAAGGARGACAAGAC and EF-2> GGARGT
ACCAGTSATCATG and sequenced with EF-3> GTAAGG
AGGASAAGACTCACC and EF-22T> AGGAACCCTTAC
CGAGCTC; and RPB2 was PCR amplified and sequenced as two
contiguous fragments using primers 5f2> GGGGWGAYCAGA
Tab le 1 . Numbers of seedlings and treatments in inoculation experiments
Experiment no.
Total plants No. inoculated
with Fsp-1 No. Fsp-1
isolates used No. inoculated
with other fungi No. mock-
inoculated Inoculations
per plant
IE1 8 1 1 6 1 1
IE2 7 6 3 0 1 2
IE3 16 14 7 0 2 3

Plant Disease / June 2011 635
AGAAGGC × 7cr> CCCATRGCTTGYTTRCCCAT and 7cf>
ATGGGYAARCAAGCYATGGG × 11ar> GCRTGGATCTTR
TCRTCSACC as previously described (10,11,14,26). Platinum Taq
DNA polymerase (Invitrogen Life Technologies, Carlsbad, CA)
was used in all PCR reactions. Following PCR amplification, am-
plicons were sized by gel electrophoresis in 1.5% agarose gels
(Invitrogen) run in 1× TAE buffer (15), after which they were
stained with ethidium bromide and visualized over a UV trans-
illuminator. Prior to sequencing with Applied Biosystems BigDye
version 3.1 Terminator reaction mix (ABI, Emeryville, CA), ampli-
cons were purified using Montage96 filter plates (Millipore Corp.,
Billerica, MA). Sequencing reaction mixes were conducted in a
10-µl volume and contained 2 µl of ABI BigDye version 3.1-termi-
nator reaction mix, 2 to 4 pmol of a sequencing primer, and ap-
proximately 50 ng of amplicon (10). Sequencing reaction mixes
were purified using ABI XTerminator and then run on an ABI 3730
automated sequencer. Sequence chromatograms were edited for
accuracy and aligned with Sequencher version 4.9 (Gene Codes,
Ann Arbor, MI) prior to being exported as NEXUS files for subse-
quent analyses.
To assess the genetic diversity of the Florida torreya pathogen,
we conducted phylogenetic analyses of the aligned partial EF-1
α
and RPB2 gene sequences employing maximum parsimony in
PAUP (ver. 4.0b10; Sinauer Associates, Inc., Sunderland, MA; 24)
and maximum likelihood in GARLI (ver. 0.951; 27). Maximum
parsimony (MP) analyses were conducted in PAUP using the
branch-and-bound option for an exact solution. The general-time-
reversible model with a proportion of invariant sites and gamma
distributed rate heterogeneity was employed as the model of nu-
cleotide substitution when using maximum likelihood (ML) as the
optimality criterion. To investigate evolutionary relationships,
RPB2 sequences were obtained for five Fsp-1 isolates and 19 addi-
tional phylogenetically diverse fusaria. Over half of the latter were
represented by fusaria reported to be tree pathogens (6,9; Table 2).
The best ML tree, based on 10 independent analyses of the RPB2
dataset, yielded a log-likelihood of –9930.87 (Fig. 1). Clade sup-
port was assessed by 1,000 MP and ML bootstrap pseudoreplicates
of the data (24,27). DNA sequences generated in this study have
been deposited in GenBank under accession numbers HM068337–
HM068363.
Results
Field surveys. A total of 225 trees were located on eight sites
with an average of 28 trees and a range of 15 to 47 trees per site.
Average height and diameter at ground line among all eight sites
were 118.5 cm and 1.9 cm, respectively. Average height ranged
from 77.7 to 262.6 cm, and average DGL ranged from 0.87 to 4.4
cm. Canker incidence ranged from 71 to 100% depending on site
(Figs. 1 and 2) with an average incidence of 93.4%. However,
means for canker incidence were only statistically significant be-
tween the TNC Sweetwater Tract and all other sites (at the P <
0.05 level; see Fig. 2).
Isolation and tentative identification of fungi. Samples were
collected from cankers from a total of 46 (20% of the currently
known and surveyed population) trees from three sites (Torreya
State Park, TNC-Sweetwater Tract, and TNC-Aspalaga Tract)
representing a total of 150 samples. From these samples, nuclear
ribosomal ITS rDNA sequences were obtained for 115 isolates.
These sequences were used to query the GenBank database to
identify the isolates. The most frequently isolated species appeared
to represent a novel unidentified Fusarium species (28.7% of all
isolates and 44.0% of all sampled trees), designated Fsp-1, based
on the results of the BLASTn searches (e-values > 0.0). Unidenti-
fied fungi constituted the next most common group (21.7% of all
isolates and 40.3% of all sampled trees), followed by saprophytes
(22.5% of all isolates and 50.0% of all sampled trees) and Pestalo-
tiopsis spp. (10.8% of all isolates and 30.4% of all sampled trees).
The remaining isolates were identified as known or potential
pathogens (B. obtusa, F. cf. lateritium, members of the FSSC [12],
unidentified Fusarium sp-2 [Fsp-2], L. theobromae) but constituted
a low percentage (3.9% of all isolates and only 10.9% of all sam-
pled trees).
For the Torreya State Park samples, cankers from 31 trees were
sampled, resulting in 91 isolates, with 78 being successfully se-
Tab le 2 . Isolates used in the phylogenetic analysis
NRRL no.a Equivalent no.b Fusarium speciesc Host/substrate Geographic origin
13622 FRC L-55 Fusarium cf. lateritium Ulmus americana LA-USA
20956 CBS 123670 Fusarium verticillioides Zea mays CA-USA
22161 ATCC 18692 FSSC 13-a (Fusarium solani f. sp. robiniae) Robinia pseudoacacia Japan
22163 ATCC 18690 FSSC 22-a (Fusarium solani f. sp. xanthoxyli) Xanthoxylum piperitum Japan
22230 ATCC 44934 FSSC 17-b (Fusarium solani f. sp. mori) Morus alba Japan
22316 ATCC 66906 Fusarium staphyleae Staphylea trifolia NJ-USA
22944 CBS 217.76 Fusarium proliferatum Cymbidium sp. Germany
25226 BBA 69662 Fusarium mangiferae Mangifera indica India
25331 CBS 405.97 Fusarium circinatum Pinus radiata CA-USA
25486 CBS 258.52 Fusarium xylarioides Coffea sp. Ivory Coast
28387 PD 90/1377 Fusarium commune Unknown The Netherlands
31011 BBA 69079 FIESC 12-a (Fusarium sp.) Thuja sp. Germany
31041 LI #95 Fusarium virguliforme Glycine max IL-USA
31084 PH-1 = CBS 123657 Fusarium graminearum Zea mays MI-USA
34936 CBS 123668 Fusarium oxysporum f. sp. lycopersici Solanum lycopersicum The Netherlands
36148 CBS 109638 Fusarium buxicola Buxus sp. Belgium
36575 CBS 976.97 FIESC 20-b (Fusarium sp.) Juniperus chinensis HI-USA
37021 FRC L-0110 Fusarium cf. lateritium Coffea sp. New Guinea
45880 VanEtten 77-13-4 FSSC 11-c (Fusarium solani f. sp. pisi) Pisum sativum Unknown
54149 JAS #481 (5005-08 canker) Fusarium sp. (Fsp-1) Torreya taxifolia FL-USA (Aspalaga)
54150 JAS #499 (4005-08 recently dead) Fusarium sp. (Fsp-1) Torreya taxifolia FL-USA (Gregory House)
54151 JAS #510 (5003-08 shoot dieback) Fusarium sp. (Fsp-1) Torreya taxifolia FL-USA (Aspalaga)
54152 JAS #542 (5006-09 canker) Fusarium sp. (Fsp-1) Torreya taxifolia FL-USA (Aspalaga)
54153 JAS #545 (5003-08 canker) Fusarium sp. (Fsp-1) Torreya taxifolia FL-USA (Aspalaga)
54154 JAS #587 (4008-08 canker) Fusarium sp. (Fsp-1) Torreya taxifolia FL-USA (Gregory House)
54155 JAS #596 (4025-09 canker) Fusarium sp. (Fsp-1) Torreya taxifolia FL-USA (Gregory House)
a NRRL, Agricultural Research Service Culture Collection, Peoria, IL.
b ATCC, American Type Culture Collection, Manassas, VA; BBA, Biologische Bundesanstalt fűr Land-und Forstwirtschaft, Institute fűr Mikrobiologie,
Berlin, Germany; CBS, CBS-KNAW Fungal Biodiversity Center, Utrecht, The Netherlands; FRC, Fusarium Research Center, Department of Plan
t
Pathology, The Pennsylvania State University, University Park, PA; JAS, Jason A. Smith, University of Florida, Gainesville, FL.
c Strains of formae speciales (f. sp.) within the Fusarium solani species complex (FSSC) represent four phylogenetically distinct species. NRRL 13622 and
37021 are listed as Fusarium cf. lateritium because it is unclear which of these two species, if any, is authentic for this species.

636 Plant Disease / Vol. 95 No. 6
quenced (85.7%). The most common species isolated was Fsp-1,
isolated from 28.5% of all samples and 51.6% of all trees. This was
followed by Pestalotiopsis sp. at 8.8% of all samples and 22.6% of
all trees sampled. The remainder of the isolates could not be identi-
fied based on rDNA-ITS sequences (23.2% of all samples and
70.9% of all trees) or were considered to be nonpathogenic fungi
(24.2% of all samples and 54.4% of all trees). A second unidenti-
fied Fusarium species (Fsp-2) based on BLASTn searches com-
prised 1.0% of all samples and 3.2% of all trees sampled.
Cankers from five trees were sampled at the TNC-Aspalaga
Tract, resulting in 14 isolates from which ITS rDNA sequence data
were obtained. The most common species isolated were Fsp-1
(21.4% of all samples and 14.3% of all trees sampled) and Pestalo-
tiopsis sp. (21.4% of all samples and 14.3% of all trees sampled).
The remainder of the isolates could not be identified using the
rDNA-ITS sequence data (42.9% of all samples and 100% of all
trees sampled) or appeared to represent saprophytic species (14.3%
of all samples and trees sampled).
Cankers from 10 trees were sampled at the TNC-Sweetwater
tract, from which ITS rDNA sequence data were obtained from the
24 isolates recovered. The most commonly isolated species was
Fsp-1, isolated from 33.4% of all samples and 40.0% of all trees
sampled. This was followed by Pestalotiopsis sp. at 20.9% of all
samples and 30.0% of all trees sampled. The remainder of the iso-
lates could not be identified to species using the ITS rDNA se-
quence data (16.6% of all samples and 40.0% of all trees sampled),
or they appeared to represent saprophytes (12.5% of all samples
and 30.0% of all trees sampled). The rest of the samples were
identified as known plant pathogens (B. obtusa, F. cf. lateritium,
members of the FSSC [12], and L. theobromae) (16.6% of all iso-
lates and 30.0% of all sampled trees), but not found at other loca-
tions.
Inoculation experiments. No cankers formed following inoc-
ulation with B. obtusa, F. cf. lateritium, Fusarium sp-2 (Fsp-2), L.
theobromae, and Pestalotiopsis sp. in IE1. The FSSC isolate and
Fusarium sp-1 (Fsp-1) caused cankers (% stem girdle; % SG =
100%) that caused mortality above the inoculation point after 6
months (Fig. 3C and D). Since Fs-Tt1 was infrequently recovered
in this study, additional experiments focused on Fsp-1. In IE2, all
three isolates of Fsp-1 induced cankers on all plants. Cankers were
not observed on the negative controls. The average % SG for iso-
late NRRL 54152 = 542 (65.0%), for NRRL 54154 = 587 (65.0%),
and for NRRL 54155 = 596 (80.0%). One of each of the two repli-
cates of isolates NRRL 54154 and 54155 caused mortality after 6
months. Sporodochia of Fsp-1 were observed on the cankered tis-
sue at approximately 4 months postinoculation. In IE3, all three
isolates of Fsp-1 tested caused cankers on all plants inoculated;
cankers were not induced on the plants included as negative con-
trols. The average % SG for all isolates inoculated onto the larger
plants was 16.9% (control = 0%) and on the smaller plants was
84.1% (control = 0.0%). Isolates NRRL 54149 = 481 and NRRL
54152 = 542 both caused sapling mortality after 6 months. After 4
months, cankers that formed from the three individual inoculation
points began to coalesce to form large single cankers (Fig. 3A and
B). For all inoculation experiments, Fsp-1was recovered from the
advancing canker margin on each plant at the end of the experi-
ment. The identity of the fungi recovered was determined to be
Fsp-1 based on analyses of their ITS rDNA sequence.
Molecular phylogenetics. DNA sequence data were obtained
from portions of three nuclear genes to identify the Florida torreya
pathogen (Fsp-1) and to characterize its genetic diversity (Table 2).
ITS rDNA sequences of the six isolates sequenced (NRRL 54149–
54151, 54153–54155) were identical and yielded an alignment of
453 nucleotide positions. A search of GenBank (http://www.ncbi.
nlm.nih.gov/), using one of the ITS rDNA sequences as the query,
identified Fusarium lateritium BBA 65675 (AY188920, 96% iden-
tity), F. buharicum (FBU34581, 97% identity), and Fusarium sp.
IP-87 (DQ780424, 97% identity) as the best matches. Partial EF-
1
α
sequences of the seven isolates (NRRL 54149-54155) yielded
an alignment of 689 bp. A branch-and-bound search of the EF-1
α
dataset identified four equally most parsimonious trees four steps
in length. Five unique haplotypes were identified among the seven
sequences (Fig. 4). BLAST searches of GenBank and the FUSA-
RIUM-ID database (http://isolate.fusariumdb.org/index.php; 5),
using a partial EF-1
α
sequence as the query, failed to identify any
sequences producing significant alignments. This query identified
an EF-1
α
sequence of F. redolens NRRL 52619 (GU250581) with
the best maximum identity, but at only 86%. Similar BLAST
searches of GenBank, using a partial RPB2 sequence as the query,
also identified sequences with a maximum identity of only 86% as
the top hits (ex., F. concolor EF470115 and F. brachygibbosum
GO505482). The results of these three BLAST searches, coupled
with comparisons of the partial EF-1
α
and RPB2 gene sequences
with more inclusive fusarial databases of these two genes (K.
O’Donnell, unpublished), strongly indicated that the Florida tor-
reya pathogen represented a novel species of Fusarium.
To assess phylogenetic relationships of the Florida torreya
pathogen within Fusarium, MP and ML analyses were conducted
on partial RPB2 sequences (1,772 bp alignment) of five Fsp-1 iso-
lates together with a phylogenetically diverse set of 19 Fusarium
spp.; over half of the latter have been reported to be tree pathogens
(Table 2; Fig. 5). The results of these analyses were highly concor-
dant in placing the Florida torreya pathogen as a novel phylo-
genetically distinct species representing one of the earliest diverg-
ing lineages within the Gibberella clade of Fusarium.
Discussion
Field studies indicate that the existing Florida torreya population
is severely affected by canker disease. However, linking any cur-
rent threat, including CDFT, with the historical decline of this criti-
cally endangered species is difficult to accomplish. Although sev-
eral authors hypothesized the role of a pathogen in the initial
decline of Florida torreya (2,3,22), no pathogen was conclusively
demonstrated to be the cause of decline. The rapid die off and
subsequent windthrow of dead trees (21) suggests the involvement
of a root pathogen (e.g., Phytophthora cinnamomi); however, no
root pathogen has been shown to cause significant damage to
Fig. 1. Typical stem cankers (yellow arrows) observed on main stem of tree and
basal sprouts of Torreya taxifolia at natural sites.
Fig. 2. Canker incidence in eight Florida torreya sites surveyed during 2008 to
2010. Letters above bars represent homogenous subsets (Duncan’s test, P < 0.05).

Plant Disease / June 2011 637
Fig. 3. Symptoms on cultivated plants inoculated with Fusarium sp. (Fsp-1). A, Cankers coalescing; B, bark scraped away to reveal lesion; C, mortality of inoculated seedling;
D, seedling with stem mortality above inoculation point and basal sprouting. Symptoms observed at 4 weeks (A and B) and 6 weeks (C and D) postinoculation, respectively.

638 Plant Disease / Vol. 95 No. 6
Florida torreya. Moreover, the fact that basal sprouts frequently de-
velop from older root systems suggests that the roots are not cur-
rently affected by disease.
Inoculation experiments verify that CDFT is caused by an un-
described Fusarium sp. Although Fusarium spp. have been
implicated as causal agents of disease of Florida torreya in the past,
none has been demonstrated to either induce cankers and stem
mortality or cause symptoms that are similar to those observed in
the field. A pathogen reported as F. lateritium caused leaf spots on
Florida torreya, but apparently not cankers (3). The novel
Fusarium species has not been detected in previous pathology
studies of Florida torreya. However, because published studies on
CDFT did not include molecular phylogenetic data, it is possible
that Fsp-1 was misidentified as a described species using morpho-
logical species recognition.
Analyses of multilocus DNA sequence data, which included
BLAST searches of GenBank and the FUSARIUM-ID databases
(5), phylogenetic analyses of partial RPB2 gene sequences, and
DNA sequence comparisons with more inclusive EF-1
α
and RPB2
gene sequence databases (K. O’Donnell, unpublished data),
strongly suggest the Florida torreya pathogen represents a novel,
phylogenetically distinct species within Fusarium (25). Phyloge-
netic placement of this novel pathogen as one of the earliest diverg-
ing lineages within the Gibberella clade was greatly facilitated by
the use of partial RPB2 sequence data, which has a significant
advantage over ITS rDNA and partial EF-1
α
data, in that it can be
easily aligned across the phylogenetic breadth of Fusarium (11).
ITS rDNA sequence data have low utility within Fusarium because
highly divergent paralogs or xenologs have been reported within
the Gibberella fujikuroi and F. oxysporum species complexes and
because it contains relatively little phylogenetic signal (10). By
way of contrast, RPB2 sequence data are highly informative phy-
logenetically, as evidenced by the large number of parsimony
informative characters in the dataset analyzed in the present study
(Fig. 1) and in previously published studies of other fusaria
(11,12,13).
Florida torreya faces numerous challenges to its future survival
in its natural habitat. In addition to the canker disease, deer rou-
tinely cause damage to stems from antler rubbing. It is unclear
whether they are attracted by the tree’s aroma or seek out Florida
torreya for some other unknown reason. Whether the wounds
caused by deer serve as infection courts for pathogens, including
the new Fusarium sp. (Fsp-1), is unclear and warrants further
study. Additionally, since lesions on the larger plants in IE3 re-
sulted in less stem girdling and no mortality, the host response to
infection, particularly under different stress conditions, needs to be
investigated. In addition to more research on the biology and man-
agement of CDFT, more work is needed to assess the various fac-
tors involved in decline of Florida torreya and how the species can
be protected from extinction.
Acknowledgments
We thank the Florida Parks Department staff and numerous volunteers who
were instrumental in carrying out the surveys for Florida torreya. We also thank
Stacy Sink for generating the DNA sequence data reported in this study and
Nathane Orwig for collecting the DNA sequence data in the NCAUR DNA core
facility. Mention of firm names or trade products does not imply that they are
endorsed or recommended by the U.S. Department of Agriculture over other
firms or similar products not mentioned.
Literature Cited
1. Alfieri, S. A., Jr., Martinez, A. P., and Wehlburg, C. 1967. Stem and needle
blight of Florida torreya (Torreya taxifolia Arn.). Proc. Fla. State Hortic.
Soc. 80:428-431.
Fig. 4. One of four most-parsimonious phylograms inferred for partial EF-1
α
gene
sequences for seven isolates of the Tor r e y a ta xi fo li a (Florida torreya) pathogen. Five
unique, but closely related, haplotypes are represented in the midpoint-rooted tree.
Fig. 5. One of 24 equally most-parsimonious trees inferred from partial RPB2 gene
sequences for 24 fusaria, including five isolates of the Florida torreya pathogen
(highlighted in dark gray). The whole genome sequence of the four fusaria high-
lighted in light gray is publically available. Note that the Florida torreya isolates form
a phylogenetically distinct lineage, representing one of the earliest divergences
within the Gibberella clade of Fusarium.

Plant Disease / June 2011 639
2. Alfieri, S. A., Jr., Schoulties, C. L., Langdon, K. R., and El-Gholl, N. E.
1987. Leaf and stem disease of Torreya taxifolia in Florida. Plant Pathology
Circ. No. 291. Florida Department of Agriculture and Consumer Service,
Division of Plant Industry.
3. El-Gholl, N. E. 1985. Fusarium lateritium causing needle spots on To rre ya
taxifolia in Florida. Plant Dis. 69:905.
4. Farjon, A. 2010. A Handbook of the World’s Conifers. Brill Academic Pub-
lishers, Leiden, Netherlands.
5. Geiser, D. M., del Mar Jiménez-Gasco, M., Kang, S., Makalowska, I.,
Veeraraghavan, N., Ward, T. J., Zhang, N., Kuldau, G. A., and O’Donnell,
K. 2004. FUSARIUM-ID v.1.0: A DNA sequence database for identifying
Fusarium. Eur. J. Plant Pathol. 110:473-479.
6. Gerlach, W., and Nirenberg, H. 1982. The genus Fusarium – A pictorial
atlas. Mitt. Biol. Bundesanst. Land-Forstwirtsch, Berlin-Dahlem 209:1-406.
7. Godfrey, R. K., and Kurz, H. 1962. The Florida torreya destined for extinc-
tion. Science 136:900-902.
8. Lee, J. C., Yang, X., Schwartz, M., Strobel, G., and Clardy, J. 1995. The
relationship between an endangered North American tree and an endophytic
fungus. Chem. Biol. 2:721-727.
9. Leslie, J. F., and Summerell, B. A. 2006. The Fusarium Laboratory Manual.
Blackwell Publ., Ames, IA.
10. O’Donnell, K., Cigelnik, E., and Nirenberg, H. 1998. Molecular systematics
and phylogeography of the Gibberella fujikuroi species complex. Mycolo-
gia 90:465-493.
11. O’Donnell, K., Sarver, B. A. J., Brandt, M., Chang, D. C., Noble-Wang, J.,
Park, B. J., Sutton, D. A., Benjamin, L., Lindsley, M., Padhye, A., Geiser,
D. M., and Ward, T. J. 2007. Phylogenetic diversity and microsphere array-
based genotyping of human pathogenic fusaria, including isolates from the
multistate contact lens-associated U.S. keratitis outbreaks of 2005 and
2006. J. Clin. Microbiol. 45:2235-2248.
12. O’Donnell, K., Sutton, D. A., Fothergill, A., McCarthy, D., Rinaldi, M. G.,
Brandt, M. E., Zhang, N., and Geiser, D. M. 2008. Molecular phylogenetic
diversity, multilocus haplotype nomenclature, and in vitro antifungal resis-
tance within the Fusarium solani species complex. J. Clin. Microbiol.
46:2477-2490.
13. O’Donnell, K., Sutton, D. A., Rinaldi, M. G., Gueidan, C., Crous, P. W., and
Geiser, D. M. 2009. Novel multilocus sequence typing scheme reveals high
genetic diversity of human pathogenic members of the Fusarium incar-
natum-F. equiseti and F. chlamydosporum species complexes within the
United States. J. Clin. Microbiol. 47:3851-3861.
14. O’Donnell, K., Ward, T. J., Geiser, D. M., Kistler, H. C., and Aoki, T. 2004.
Genealogical concordance between the mating type locus and seven other
nuclear genes supports formal recognition of nine phylogenetically distinct
species within the Fusarium graminearum species complex. Fungal Genet.
Biol. 41:600-623.
15. Sambrook, J., Fritsch, E. F., and Maniatis, T. 1989. Molecular Cloning: A
Laboratory Manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY.
16. Schwartz, M. W. 1993. Allozyme variation of the endangered Florida tor-
reya (Torreya taxifolia). Can. J. For. Res. 23:2598-2602.
17. Schwartz, M. W., and Hermann, S. 1993. The continuing population decline
of Torreya taxifolia Arn. Bull. Torrey Bot. Club 120:275-286.
18. Schwartz, M. W., and Hermann, S. M. 1999. Is slow growth of the endan-
gered Torreya taxifolia (Arn.) normal? J. Torrey Bot. Soc. 126:307-312.
19. Schwartz, M. W., Hermann, S. M., and van Mantgem, P. J. 2000. Population
persistence in Florida torreya: Comparing modeled projections of a declin-
ing coniferous tree. Conserv. Biol. 14:1023-1033.
20. Schwartz, M. W, Hermann, S. M., and van Mantgem, P. J. 2000. Estimating
the magnitude of decline of the Florida torreya (Torreya taxifolia Arn.).
Biol. Conserv. 95:77-84.
21. Schwartz, M. W., Hermann, S. M., and Vogel, C. 1995. The catastrophic
loss of Torreya taxifolia: assessing environmental induction of disease hy-
potheses. Ecol. Appl. 5:501-516.
22. Schwartz, M. W., Porter, D. J., Hermann, S. M., and Strobel, G. 1996.
Occurrence of Pestalotiopsis microspora on Torreya taxifolia in Florida.
Plant Dis. 80:600.
23. Sinclair, W., and Lyon, H. H. 2005. Diseases of Trees and Shrubs. Cornell
University, Ithaca, NY.
24. Swofford, D. L. 2002. PAUP*. Phylogenetic analysis using parsimony
(*and other methods), version 4.0b10. Sinauer Associates, Sunderland, MA.
25. Taylor, J. W., Jacobson, D. J., Kroken, S., Kasuga, T., Geiser, D. M., Hib-
bett, D. S., and Fisher, M. C. 2000. Phylogenetic species recognition and
species concepts in fungi. Fungal Genet. Biol. 31:21-32.
26. White, T. J., Bruns, T., Lee, S., and Taylor, J. 1990. Amplification and direct
sequencing of fungal ribosomal RNA genes for phylogenetics. Pages 315-
322 in: PCR Protocols: A Guide to Methods and Applications. M. A. Innes,
D. H. Gelfand, J. J. Sninsky, and T. J. White, eds. Academic Press, San
Diego, CA.
27. Zwickl, D. J. 2006. Genetic algorithm approaches for the phylogenetic
analysis of large biological sequence data sets under the maximum likeli-
hood criterion. Ph.D. diss. University of Texas, Austin.