Content uploaded by Lee Berger
Author content
All content in this area was uploaded by Lee Berger on Nov 07, 2017
Content may be subject to copyright.
DISEASES OF AQUATIC ORGANISMS
Dis Aquat Org
Vol. 68: 51–63, 2005 Published December 30
INTRODUCTION
Batrachochytrium dendrobatidis causes a potentially
fatal epidermal infection of amphibians and has
caused mass mortality, population declines and extinc-
tions (Berger et al. 1999, Speare et al. 2001, McDonald
et al. 2005). Chytridiomycosis has been recorded from
Australia, New Zealand, Europe, Africa, and South,
Central and North America, from a broad range of
habitats (Berger et al. 1999, Lips 1999, Mutschmann et
al. 2000, Bosch et al. 2001, Fellers et al. 2001, Speare et
al. 2001, Bradley et al. 2002, Weldon et al. 2004). Mor-
tality rates of up to 100% occurred during natural out-
breaks in captivity and in transmission experiments
in captive amphibians of susceptible species (Berger et
al. 1998, Longcore et al. 1999, Berger 2001, Nichols et
al. 2001), although other species can survive infection
(Ardipradja 2001, Davidson et al. 2003). Death in sus-
ceptible experimental animals usually occurs from
between 18 and 70 d post exposure and incubation
time varies with dose, fungal strain, temperature and
amphibian species (Ardipradja 2001, Berger 2001,
Nichols et al. 2001, Woodhams et al. 2003, Berger et al.
2004). Amphibians with no clinical signs frequently
carry light infections in the wild (Hopkins & Channing
2003, Hanselmann et al. 2004, Retallick et al. 2004,
McDonald et al. 2005).
The amphibian chytrid was placed in a new genus,
Batrachochytrium (Phylum Chytridiomycota, Class
Chytridiomycetes, Order Chytridiales) (Longcore et al.
1999). Photographs of an isolate from a captive blue
poison dart frog Dendrobates azureus that died at the
National Zoological Park in Washington, DC, USA,
were designated as the nomenclatural type of the spe-
cies, which was named B. dendrobatidis (Longcore et
al. 1999). The ultrastructural morphology of its zoospore,
© Inter-Research 2005 · www.int-res.com
*Email: lee.berger@jcu.edu.au
Life cycle stages of the amphibian chytrid
Batrachochytrium dendrobatidis
Lee Berger1, 2,*, Alex D. Hyatt2, Rick Speare1, Joyce E. Longcore3
1Amphibian Disease Ecology Group, School of Public Health, Tropical Medicine and Rehabilitation Science,
James Cook University, Townsville, Queensland 4811, Australia
2Australian Animal Health Laboratory, CSIRO Livestock Industries, Private Bag 24, Geelong, Victoria 3220, Australia
3Department of Biological Sciences, University of Maine, Orono, Maine 04469-5722, USA
ABSTRACT: An overview of the morphology and life cycle of Batrachochytrium dendrobatidis, the
cause of chytridiomycosis of amphibians, is presented. We used a range of methods to examine stages
of the life cycle in culture and in frog skin, and to assess ultrastructural pathology in the skin of
2 frogs. Methods included light microscopy, transmission electron microscopy with conventional
methods as well as high pressure freezing and freeze substitution, and scanning electron microscopy
with critical point drying as well as examination of bulk-frozen and freeze-fractured material.
Although chytridiomycosis is an emerging disease, B. dendrobatidis has adaptations that suggest it
has long been evolved to live within cells in the dynamic tissue of the stratified epidermis. Sporangia
developed at a rate that coincided with the maturation of the cell, and fungal discharge tubes usually
opened onto the distal surface of epidermal cells of the stratum corneum. A zone of condensed,
fibrillar, host cytoplasm surrounded some sporangia. Hyperkeratosis may be due to (1) a hyperplastic
response that leads to an increased turnover of epidermal cells, and (2) premature keratinization and
death of infected cells.
KEY WORDS: Batrachochytrium dendrobatidis · Chytridiomycosis · Fungus · Morphology ·
Ultrastructure · Transmission and scanning electron microscopy · Pathology · Amphibian
Resale or republication not permitted without written consent of the publisher
Dis Aquat Org 68: 51– 63, 2005
its occurrence on an amphibian host and small subunit
ribosomal DNA (ssu-rDNA) sequence of B. dendro-
batidis indicate that this fungus is distinctly different
from other known chytrids (Berger et al. 1998, Long-
core et al. 1999, James et al. 2000). It is the only mem-
ber of this phylum to cause disease in a vertebrate.
The life cycle of Batrachochytrium dendrobatidis is
a simple progression from zoospore to the growing
organism, called a thallus, which produces a single
zoosporangium (= container for zoospores). The con-
tents of the zoosporangium (also known as a spo-
rangium) cleave into new zoospores which exit the
sporangium through one or more papillae. Sexual
reproduction has not been observed. Colonial develop-
ment resulting from the formation of more than 1 spo-
rangium from 1 zoospore is the only known variation of
the cycle (Longcore et al. 1999). The life cycle seems to
be the same in culture and in skin. The duration of the
life cycle in vitro is 4 to 5 d at 22°C and is assumed to
be the same in amphibian skin, although this has not
been tested.
Batrachochytrium dendrobatidis discharges zoospores
through an inoperculate opening and exhibits mono-
centric or colonial growth (Longcore et al. 1999). Long-
core et al. (1999) comprehensively described the taxo-
nomic features of B. dendrobatidis and gave detailed
morphology based on light microscopic observations
of cultures and ultrastructure of serially sectioned
zoospores, but did not describe the appearance of life
cycle stages using scanning electron microscopy (SEM).
Significant ultrastructural differences were not
observed among zoospores of isolates from Australia,
the USA or Central America (Berger et al. 1998, Long-
core et al. 1999). Multilocus sequence typing (MLST)
has been used to examine genetic diversity among
fungal strains from North America, Panama, Australia
and from frogs imported from Africa, and only 5 vari-
able nucleotide positions were detected among 10 loci
(5918 bp)(Morehouse et al. 2003). These results sug-
gest that Batrachochytrium dendrobatidis is a recently
emerged clone and support the epidemiological data
showing that chytridiomycosis has been introduced
into many countries from a common source. There is
evidence that Africa is the origin (Weldon et al. 2004).
In amphibians sporangia infect cells in the stratum
granulosum and stratum corneum in the superficial
epidermis. Immature sporangia occur within the
deeper, more viable cells while mature zoosporangia
and empty sporangia are more prevalent in the outer
keratinized layers. Discharge tubes generally project
towards the skin surface and zoospores may be
released to the environment. The distribution of spo-
rangia in developing tadpoles follows the changes
in the distribution of keratinized epidermis (Marantelli
et al. 2004).
Resistant resting spores have not been found in his-
tologically prepared slides of infected skin or during
examination of fresh skin in the process of isolating the
fungus from more than 80 amphibians (J. E. Longcore
unpubl. data). MLST sequencing studies indicate that
Batrachochytrium dendrobatidis reproduces clonally,
which supports the lack, or uncommon occurrence,
of a sexually produced resting stage (Morehouse et al.
2003).
The histopathology of chytridiomycosis based on
examination of haematoxylin and eosin stained sec-
tions has been described (Berger et al. 1998, Pessier et
al. 1999, Berger 2001). Infection almost always causes
hyperkeratosis in the region of the thalli. Other
changes include irregular multifocal hyperplasia, dis-
ordered epidermal cell layers, spongiosis, erosions and
occasional ulcerations of the skin. Epidermal width is
highly variable with diffuse or focal thickening in some
areas, as well as large areas of thinning. Individual
epidermal cell pyknosis and vacuolation may occur in
scattered cells in the stratum basale or more superficial
layers near infection foci. Occasionally vacuolated
degenerate cells appear to coalesce into vesicles that
result in lifting of the epidermis and erosion. There
may be a mild inflammatory response with a slight
increase in mononuclear cells in the dermis, but in-
flammatory response in the epidermis is uncommon.
Two hypotheses have been proposed to explain how
a fungus that is restricted to the superficial epidermis
has the capacity to kill frogs (Berger et al. 1998, Pessier
et al. 1999). (1) The chytrid might release proteolytic
enzymes or other active compounds that are absorbed
through the permeable skin of the frog or, possibly,
(2) damage to skin function results in disturbance of
oxygen, water or electrolyte balance which results in
death. By examining the ultrastructural pathology in
epidermal cells we aim to improve our understanding
of the effect of Batrachochytrium dendrobatidis on
epidermal cells.
Cultures of Batrachochytrium dendrobatidis are
highly susceptible to a range of antifungal drugs and
antifungal peptides produced by frogs (Berger 2001,
Rollins-Smith et al. 2002, Johnson et al. 2003). How-
ever, since topical antifungal treatment of frogs has
variable efficacy, it appears that B. dendrobatidis is
less susceptible when present in amphibian skin
(Berger 2001). The reason for this is not known, but
one hypothesis is that thalli are protected by their
intracellular location in epidermal cells. One aim of our
transmission electron microscopy (TEM) studies is to
determine if the epidermal cell undergoes modifica-
tions that may explain this protection.
This study was undertaken to improve our under-
standing of the morphology of Batrachochytrium den-
drobatidis by using TEM and SEM of all fungal stages
52
Berger et al.: Life cycle of Batrachochytrium dendrobatidis
in culture and in epidermis. To understand the patho-
genesis of chytridiomycosis and why chytridiomycosis
appears resistant to antifungal drugs we used ultra-
structural studies of infected epidermis to examine the
effect of the fungus on amphibian cells.
MATERIALS AND METHODS
Samples. Light microscopic studies of Batrachochy-
trium dendrobatidis in culture were of isolate Mel-
bourne-Ldumerilii-98-LB-1 from a captive, ill Limno-
dynastes dumerilii. For ultrastructural studies we
examined the isolate Tully-Ndayi-98-LB-1 from a
free-living, ill Nyctimystes dayi collected from Tully,
Queensland. Cultures were isolated and maintained
on tryptone/gelatin hydrolysate/lactose (TGhL) agar or
broth using routine methods (Longcore et al. 1999).
Infected skin samples were from a wild Litoria lesueuri
from Goomburra, Queensland (AAHL 97 574/1) that
died and was fixed in 2.5% glutaraldehyde, a captive
Bufo marinus that died and was fixed in 10% formalin,
and 2 captive L. gracilenta were used for high pressure
freezing. The latter samples were a toe-clip taken from
a healthy infected frog with a light infection, and skin
from a frog that had recently died with severe chy-
tridiomycosis. These samples were kept on ice for
about 2 h before processing.
TEM. Conventional processing: Samples of skin
were fixed in 2.5% (v/v) glutaraldehyde in 0.1 M
cacodylate buffer (pH 6.8, 300 mOsm kg–1) for 1 h,
washed in buffer (3 ×20 min), post-fixed in 1% (w/v)
osmium tetroxide in 0.1 M cacodylate buffer for 1 h
followed by washing (3 ×5 min) in reverse osmosis
water. Samples were then dehydrated in graded
alcohols (70 to 100%), infiltrated with Spurrs resin by
placing them in 50/50 Spurrs/100% ethanol and then
in 2 changes of 100% Spurrs followed by embedding
in Spurrs epoxy resin at 65°C (overnight). Fixation of
cultures of Batrachochytrium dendrobatidis differed in
that glutaraldehyde was added to the culture medium
(TGhL broth) to a final concentration of 2.5%. Pre-
liminary experiments revealed that zoospores were
sensitive to the osmolality of the buffer, but TGhL
broth was found to cause no discernible artifacts in the
cytoplasm.
High pressure freezing and freeze substitution:
Samples of Batrachochytrium dendrobatidis culture
and of infected frog skin were rapidly frozen under
high pressure (2000 bars) within a Leica high-pressure
freezer. Brass planchets were soaked in a lecithin-
chloroform solution and allowed to dry before packing
with agar cultures or pieces of toe skin. A range
of embedding media was used: TGhL broth, 2.3 M
sucrose, hexadecene and hexene. The frozen plan-
chets were stored in liquid nitrogen until processing.
Substitution was initiated by placing planchets and
samples within perforated beam capsules in 1% (w/v)
osmium tetroxide/2.5% (v/v) glutaraldehyde in 100%
acetone at –90°C in the presence of a molecular sieve
within a ‘Leica CS auto’ for 8 d. The temperature was
raised to –60°C at 2.5°C h–1 and kept at – 60°C for 2 d.
The temperature was then raised to –18°C at 2.5°C h–1
then up to 0°C at 10°C h–1 and the media replaced with
100% acetone. Samples were then warmed to room
temperature (24°C) in 1 h. The substitution medium
was replaced with 1:1 propylene oxide (PO):acetone,
100% PO, 1:1 PO:epon, 100 % epon ×2 and embedded
at 60°C for 24 h. All media used in planchets resulted
in excellent preservation of samples.
Cutting and examination of samples: Thick sections
were examined with methylene blue stain by light
microscopy to identify and select the sections contain-
ing sporangia. Ultra-thin sections (70 nm) were cut on
a Leica-Reichert-Jung Ultracut E microtome, floated
and adhered onto a grid, double stained in uranyl
acetate and lead citrate and examined with a Hitachi
H7000 or Philips CM 120 transmission electron micro-
scope at 75 or 100 kV.
SEM. Critical point drying: Skin was fixed in phos-
phate buffered 2.5% glutaraldehyde for 1 h, washed in
phosphate buffer (pH 6.8, 300 mOsm kg–1) for 15 min,
post-fixed in buffered 1% (w/v) osmium tetroxide for
3 h, and rinsed in the same buffer (×10). Samples were
then placed in saturated aqueous filtered thiocarbo-
hydrazide (1% w/v) for 10 min, rinsed in distilled water
(×10), then placed in 1% (w/v) aqueous osmium
tetroxide at 0°C (30 min), rinsed in reverse osmosis
water and sequentially dehydrated in graded alcohol
(70 to 100 %). Following dehydration, tissues were
critically point dried from liquid carbon dioxide,
mounted on a stub with carbon dag, and sputter coated
with gold. Samples were viewed with a JEOL JSM 840
scanning electron microscope at 5 to 15 kV with a
working distance of between 16 and 22 mm.
Bulk-frozen hydrated samples (agar): Cultures were
examined with a JEOL 6340F field emission scanning
electron microscope fitted with an Oxford 1500 cryo
system. Blocks of 5 d old agar cultures were adhered to
stubs with OCT compound cryo adhesive (Tissue Tek)
and then plunged into melting nitrogen and trans-
ferred under vacuum to the microscope at liquid
nitrogen temperature. Samples were etched at – 95°C
for 60 s, then coated with gold/palladium in the trans-
fer chamber. Samples were examined at 2 kV and a
working distance of 20 mm.
Bulk-frozen hydrated samples (coverslip): To obtain
a cleaner view of the base of sporangia, cultures were
grown on sterile, round, plastic, 13 mm coverslips
(Thermanox), by placing coverslips in petri dishes of
53
Dis Aquat Org 68: 51– 63, 2005
active TGhL broth culture. Between 6 and 10 d, broth
was removed by rinsing the coverslips through 3
changes of distilled water. Excess water was removed
by touching the edges with filter paper, and leaving
them to dry for about 1 min. To reduce charging, cover-
slips were cut into quarters before they were adhered
to the stub with carbon dag. Stubs were plunged into
melting nitrogen then transferred under vacuum at
liquid nitrogen temperature to the Hexland chamber
and into the JEOL JSM-840 at –170°C. Extended dry-
ing at –80°C was necessary to sublimate ice, and good
results were obtained on a sample that was desiccated
overnight and examined as a freeze-dried sample;
however, an effect of freeze drying is shrinkage. Some
sporangia did not withstand this form of dehydration
and shrank, presenting a crumpled appearance. Sam-
ples were gold-coated within the antechamber of the
microscope. Samples were examined at 3 to 5 kV and a
working distance of 20 mm.
Freeze fracture: Freeze fracture preparations were
produced and examined using a Philips XL30 FEG or
JEOL JSM-840 SEM. Six to 14 d old agar cultures were
packed into rivets that were joined upright. After
plunging in melting nitrogen and placing in the
transfer chamber of the microscope, the top rivet was
knocked off, exposing the fractured surface of the
culture. The sample was etched at –96°C for 2 min to
remove about 1200 nm water, and coated with plat-
inum or gold. The sample was etched again at –100°C
for 2 min. Samples were examined at 6 kV with a
working distance of 20 mm.
RESULTS AND REVIEW OF MORPHOLOGY
Zoospore
Zoospores are the unwalled, waterborne, motile,
flagellated stage (Figs. 1 to 4). Zoospores of Batra-
chochytrium dendrobatidis are mostly spherical but
can be elongate and amoeboid when first released
from the zoosporangium (Longcore et al. 1999). They
54
Figs. 1 to 4. Batrachochytrium dendro-
batidis. Fig. 1. Light micrograph of live
cultured zoospore. Dark droplets are
probably lipid globules. Scale bar =
6 µm. Figs. 2 to 4. B. dendrobatidis.
Transmission electron micrographs of
zoospores. Fig. 2. Formalin-fixed zoo-
spores within a zoosporangium in the
skin of Bufo marinus. Zoospores are
being released and contain numerous
lipid globules that are partially sur-
rounded by the microbody and occur
at the edge of the ribosomal mass.
Scale bar = 2 µm. Fig. 3. Glutar-
aldehyde-fixed cultured zoospore.
The nonflagellate centriole (NFC) is
parallel to the kinetosome. Micro-
tubule root runs parallel to the kineto-
some and is embedded in a cone of
ribosomes. Scale bar = 0.6 µm. Fig. 4.
Glutaraldehyde-fixed cultured zoo-
spore. Nucleus is not associated with
the kinetosome and is nested in the
ribosomal mass which is surrounded
by endoplasmic reticulum. Mitochon-
dria are adjacent to the ribosomal
mass. Scale bar = 1 µm. F = flagellum;
N = nucleus; R = ribosomes; Mb =
microbody; L = lipid droplet; K = kine-
tosome; M = mitochondria; Tp = termi-
nal plate; V = vacuole; ER = endo-
plasmic reticulum; MT = microtubules
Berger et al.: Life cycle of Batrachochytrium dendrobatidis
are about 3–5 µm in diameter with a posteriorly
directed flagellum (19–20 µm in length) (Longcore et
al. 1999). After a period of motility and dispersal, the
zoospore encysts. The flagellum is resorbed and a cell
wall forms (Fig. 5).
Zoospore ultrastructure is used to differentiate
orders and genera and many important taxonomic fea-
tures are in the flagellar apparatus. The ultrastructure
of the flagellar region is best explained by diagrams
(see Fuller 1996). The features of the zoospore of Batra-
chochytrium dendrobatidis that are common to the
order Chytridiales are that the nucleus and kinetosome
are not associated, ribosomes are aggregated into a
core surrounded by endoplasmic reticulum, the micro-
body partially surrounds the lipid globules, and the
nonflagellated centriole (NFC) is parallel and con-
nected to the kinetosome (Longcore 1993, Longcore
et al. 1999). A key feature of B. dendrobatidis is the
numerous small lipid droplets with the microbodies
that are associated with the edge of the ribosomal mass
(Longcore et al. 1999) (Fig. 2). The kinetosomal root is
comprised of a group of microtubules that run parallel
to the kinetosome as they extend into the ribosomal
mass (Fig. 3). Additional key taxonomic structures not
shown here are that the microtubule root arises near
triplets 9-1 of the kinetosome, and that overlapping
fibers connect the NFC with the kinetosome. The
nucleus is partially nested in the ribosomal mass, mito-
chondria are adjacent to the ribosomal mass and
zoospores contain a single Golgi apparatus (Fig. 4).
Most members of the Chytridiales have a rumposome
abutting the edge of a single lipid globule and many
have a transition zone plug, but these are not present
in B. dendrobatidis (Berger et al. 1998, Longcore et
al. 1999).
Germling
After the zoospore has encysted, fine branching
rhizoids grow from one or more areas and the young
sporangium is known as a germling (Fig. 6). In
pure cultures, occasional germlings grew on adjacent
sporangia (Fig. 7).
55
Figs. 5 to 8. Batrachochytrium dendrobatidis. Fig. 5. Glutaraldehyde-fixed encysted zoospore. The resorbed flagellum is visible
and a cell wall has formed. Ribosomes are distributed throughout the cytoplasm. Scale bar = 2 µm. Fig. 6. Scanning electron
micrograph (SEM) of a germling showing fine rhizoids spreading out along the substrate. The culture was grown on a plastic
coverslip and prepared by freeze-drying. The crumpled surface is an artifact of freeze-drying. Scale bar = 10 µm. Fig. 7. Trans-
mission electron micrograph (TEM) of a young sporangium growing on an adjacent sporangium. Prepared by high pressure
freezing and freeze substitution (HPF/FS). Scale bar = 10 µm. Fig. 8. Live immature sporangium with rhizoids spreading out.
Scale bar = 10 µm. F = flagellum, N = nucleus, M = mitochondria
Dis Aquat Org 68: 51– 63, 2005
Developing zoosporangia
As the thalli grow, the cytoplasm becomes more
complex (Figs. 8 to 18) and becomes multinucleate
by mitotic divisions. The entire contents then cleave
and mature into rounded, flagellated zoospores. The
swollen part of the thallus is now known as a zoospo-
rangium. During sporangial development, one or more
discharge papillae form. Some young thalli become
divided by thin septa and each compartment grows
into a separate sporangium with its own discharge
tube; this mode of development is referred to as ‘colo-
nial growth’. A thallus that forms a single sporangium
has no divisions and the type of development is termed
‘monocentric’ (Longcore et al. 1999). Mature zoo-
sporangia contain fully formed flagellated zoospores
(Fig. 2). Actively motile zoospores were observed
within sporangia before they exited. After the plug
blocking the discharge tube dissolves, zoospores are
released. The length of discharge tubes is variable
even within an isolate, from virtually nothing up to
10 µm, and depends on type of media and density of
culture. Longer tubes were seen in skin and agar cul-
tures than in broth. Mature zoosporangia of many
chytrids do not discharge if the environment is too dry
for zoospore release; in amphibian tissue zoosporangia
of Batrachochytrium dendrobatidis may also retain
their zoospores until sufficient moisture is present to
induce zoospore discharge, but in culture on 1 % agar
medium, sufficient moisture is present that zoospores
discharge without additional liquid.
Empty sporangia
After zoospores have been released, sporangia are
clear, indicating that they are empty (Fig. 19). The
chitinous walls of the sporangia remain and may col-
lapse. Occasional zoospores do not escape and grow
within sporangia (Fig. 20). In frog skin bacteria may
enter through the open discharge tubes and replicate
inside (Figs. 20, 29 & 30).
56
Figs. 9 to 12. Batrachochytrium
dendrobatidis. Fig. 9. TEM (HPF/
FS) of an immature colonial spo-
rangium in skin of a Litoria graci-
lenta. A septum (S) divides the thal-
lus into 2 compartments. Scale bar
= 5 µm. Fig. 10. SEM of bulk-frozen
hydrated culture that has been
freeze-fractured. The image shows
a colonial thallus divided by a
septum. Scale bar = 5 µm. Fig. 11.
TEM (HPF/FS) of an immature
sporangium with a discharge pa-
pilla. The cell is multinucleate after
mitotic divisions, but the cytoplasm
has not yet divided. Plug blocking
the discharge papilla is clearly seen
(arrowhead). The wall over the tip
of the plug has dissolved, demon-
strating that B. dendrobatidis is
inoperculate. Early stages often
have large vacuoles (V). Transverse
sections of rhizoids occur in spaces
between sporangia. Scale bar =
5 µm. Fig. 12. TEM (HPF/FS) of a
multinucleate sporangium that is
beginning to cleave into zoospores.
Arrow indicates a cleavage line.
Scale bar = 4 µm. G = golgi, M =
mitochondria, A = agar, N = nucleus,
F = flagellum
Berger et al.: Life cycle of Batrachochytrium dendrobatidis
Colonies in culture
On agar, cultures of Batrachochytrium dendrobatidis
grow as granular, cream-colored clusters (Fig. 21). The
zoospores swarm in a film of surface liquid surrounding
each colony. They encyst at the base of the colony and
push the older sporangia upwards. Thalli grow better in
clusters and isolated zoospores placed on agar usually
die. This ‘group effect’ (Longcore et al. 1999) is unusual
in chytrids, and may indicate that the nutrient medium is
not optimal. Rhizoids spread over the surface of adjacent
sporangia and tightly intermingle with rhizoids from
other sporangia, joining them together (Figs. 22 to 25).
Zoospores appear to be attracted to colonies.
Batrachochytrium dendrobatidis in skin
The same stages of the life cycle occur within epi-
dermal cells of amphibian skin as in culture, although
it has not been determined whether thalli in skin differ
in their rate of development (Figs. 26 to 28). Up to 3
sporangia were seen within a single epidermal cell.
Immature stages occur in the deeper viable cells.
Mature zoosporangia and empty stages occur in the
sloughing stratum corneum. By the time most sporan-
gia have completed their development, they have been
carried to the skin surface with the differentiating epi-
dermal cells. Discharge tubes usually point towards
the skin surface, whereas in culture they may grow
laterally as well as upwards. Discharge
tubes usually protrude to the surface
through a hole in the epidermal cell
membrane. The cell membrane of the
keratinized skin cell adheres closely to the
surface of the discharge tube. The edge of
the cell membrane was not easily discerned
by SEM but was obvious by TEM in cross-
section (Figs. 28 & 29). Some zoosporangia
mature while still covered by cornified cell
layers that have not been sloughed, and
appear to discharge zoospores into the
intercellular spaces.
Colonial development can occur in skin
and sporangia with internal septa can be
seen in histological sections, although most
thalli in skin are not colonial. Bacteria on
the skin multiply on the layers of shedding
keratin and commonly grow in empty spo-
rangia (Figs. 29 & 30). Sporangia in the skin
are smaller than in culture, where they are
up to 40 µm in diameter (Longcore et al.
1999), suggesting that being intracellular
restricts their growth. Rhizoids were rarely
seen in skin sections examined by electron
microscopy. They could not be seen in
haematoxylin and eosin stained histology
sections, but were discernible adjacent to
sporangia when stained with the immuno-
peroxidase stain (Berger et al. 2002). The
fungi are often found in clusters on skin,
57
Figs. 13 to 16. Batrachochytrium dendrobatidis. Fig. 13. TEM (HPF/FS) of a sporangium in
skin of a Litoria gracilenta with a cytoplasm that has divided into incompletely formed
flagellated zoospores. Scale bar = 5 µm. Fig. 14. SEM of a bulk-frozen hydrated spo-
rangium that has been freeze-fractured. Image is a 3-dimensional representation of the
similar staged sporangium in Fig. 13. Scale bar = 5 µm. Fig. 15. Live sporangia with dis-
charge papillae. Internal structures of the sporangia are at various stages of zoospore
development. Scale bar = 20 µm. Fig. 16. SEM of a large zoosporangium on agar with 5
papillae visible. Zoospores are congregating and encysting around the base. Prepared
by bulk-freezing hydrated culture. Scale bar = 10 µm. N = nucleus, M = mitochondria,
F = flagellum, V = vacuole, A = agar
Dis Aquat Org 68: 51– 63, 200558
Figs. 17 to 20. Batrachochytrium den-
drobatidis. Fig. 17. TEM (HPF/ FS) of
a mature zoosporangium with dis-
charge papilla and plug. It is packed
with flagellated zoospores; reprinted
from Berger et al. (1999). Scale bar =
10 µm. Fig. 18. SEM of a zoospo-
rangium on agar with an encysted
zoospore at the end of a long dis-
charge tube. Prepared by bulk-
freezing hydrated culture. Reprinted
from Boyle et al. (2003). Scale bar =
10 µm. Fig. 19. Sporangia that have
released most of their zoospores.
Scale bar = 10 µm. Fig. 20. TEM
(HPF/FS) of an old sporangium in the
keratinized skin of a Litoria graci-
lenta. A zoospore (Z) that was not
released has encysted inside; note
thickened wall and resorbed flagel-
lum (F). A degenerate zoospore and
bacteria (B) are also inside. Scale bar
= 2 µm. M = mitochondria, R = ribo-
somal mass, N = nucleus
Figs. 21 to 25. Batrachochytrium dendrobatidis. Fig. 21. Culture on TGhL agar plate. Colonies appear as granular, cream-
coloured mounds. Fig. 22. SEM of a cluster of sporangia grown on a plastic coverslip and freeze-dried. Some sporangia have 2 or
more open discharge tubes. Threadlike rhizoids hold sporangia together. Scale bar = 10 µm. Fig. 23. SEM of thalli with 2 dis-
charge tubes demonstrating the aptness of the name ‘chytrid’ (i.e. earthen pot in Greek). Rhizoids from adjacent sporangia are
growing over the surface. Culture was grown on a plastic coverslip and freeze-dried. Scale bar = 10 µm. Fig. 24. SEM of 2 spo-
rangia showing the attraction between their rhizoids. Culture was grown on a plastic coverslip and freeze-dried. Scale bar =
10 µm. Fig. 25. SEM of freeze-fractured preparation of a bulk-frozen hydrated culture in agar. Most sporangia are immature. One
sporangium contains mature zoospores (arrow). Scale bar = 10 µm
Berger et al.: Life cycle of Batrachochytrium dendrobatidis 59
Fig. 26. Batrachochytrium dendrobatidis infecting Litoria
caerulea. Histological section of skin from a L. caerulea. Dark
immature fungal stages occur in the deeper cells (arrowhead).
Mature zoosporangia with distinct dark zoospores (Z) and old
empty stages with open discharge tubes (D) occur in the slough-
ing stratum corneum. Note the colonial thallus with an internal
septum (S). Haematoxylin and eosin stain. Scale bar = 30 µm
Fig. 27. Batrachochytrium dendrobatidis infecting Litoria gra-
cilenta. TEM (HPF/FS) of skin from a L. gracilenta with imma-
ture, solid stages deeper in the epidermis (arrowhead), and old
empty stages in the flattened, dark, keratinized cells. Infected
cells contain between 1 and 3 sporangia. Scale bar = 10 µm
Figs. 28 to 30. Batrachochytrium dendrobatidis. Fig. 28. SEM
of infected toe skin of a Litoria lesueuri. Almost all epidermal
cells in this field are infected and are bulging. Closed dis-
charge tubes protrude through the skin surface (arrow). Pre-
pared by critical point drying. Scale bar = 10 µm. Fig. 29. TEM
(HPF/FS) of skin of a Litoria gracilenta showing an old spo-
rangium containing bacteria (B) and a zoospore (Z). Dis-
charge tube is opening to the skin surface through a hole in
the condensed keratinised epidermal cell. Edge of the skin
tapers around the tube. Scale bar = 5 µm. Fig. 30. TEM
(HPF/FS) of old sporangia in skin of a L. gracilenta. Bacteria
(B) have entered the sporangium through an open discharge
tube and replicated into an ordered colony. S = septum in
an empty colonial sporangium. Scale bar = 10 µm
Dis Aquat Org 68: 51– 63, 2005
except in heavy infections where all ventral skin may
be diffusely infected.
Ultrastructural pathology in epidermis
TEM of skin from the lightly infected Litoria graci-
lenta showed that a build up in the layers of infected
keratinized cells caused the stratum corneum to
become focally thickened in areas of infected epider-
mis. Up to 4 additional layers of infected, keratinized
cells comprised the stratum corneum. These kera-
tinized cells were darker than the keratinized cells in
non-infected areas (Fig. 31).
A zone of apparently condensed host cytoplasm, up
to 2.5 µm thick, surrounded some sporangia. This zone
appeared to be mainly fibrils with no organelles
(Fig. 32). The more superficial epidermal cells con-
tained larger sporangia and the cell nuclei and
organelles such as mitochondria were located on one
side of the host cell. Near the skin surface the epider-
mal cell cytoplasm condensed to a thin layer around
the fungal thalli and host organelles were lost, in a pro-
cess similar to normal epidermal cell maturation. Cell
nuclei became dark and condensed but were not as
flattened as in normal stratum corneum. Keratinization
appeared to occur prematurely in infected cells below
the skin surface, compared with uninfected cells in the
same epidermal layer (Fig. 31). The cell junctions of
infected cells usually appeared normal (Fig. 32).
Infected cells and uninfected cells near foci of infec-
tion were acutely swollen, although mitochondria and
other organelles in these cells were intact. Nuclei of
some infected cells in the stratum granulosum were
shrunken and chromatolytic. Pathology in the deeper
epidermal cells, as distant as the basal layer, included
focal shrinkage, increased intercellular spaces, vacuo-
lation and dissolution of the cytoplasm (Fig. 31). Large,
distinctly bordered vesicles containing sparse cyto-
plasmic debris occurred within some cells.
The Litoria gracilenta that died from chytridiomyco-
sis had severe epidermal lesions (Fig. 33). Although
some intracellular changes could be attributed to
60
Fig. 31. Batrachochytrium dendrobatidis infecting Litoria
gracilenta. TEM (HPF/FS) of infected epidermis in a L. graci-
lenta adult without clinical signs. There are multiple layers of
dark infected keratinized cells, whereas away from the clus-
ter of sporangia the stratum corneum is 1 cell thick. Some
nuclei of infected cells are degenerate and chromatolytic
(arrowhead), and necrosis and dissolution of cells is shown in
the deeper epidermis (arrow). Scale bar = 18 µm
Fig. 32. Batrachochytrium dendrobatidis. Higher magnifica-
tion of an infected cell from Fig. 31. A zone that contains no
organelles surrounds the sporangium. Mitochondria and
apparently normal cell junctions are present. Scale bar = 4 µm
Fig. 33. Batrachochytrium dendrobatidis infecting Litoria
gracilenta. TEM (HPF/FS) of superficial epidermal cells from
an adult of Litoria gracilenta that died from chytridiomycosis.
A clear fibrillar zone of cytoplasm surrounds the sporangium
(*) and the chromatolytic nucleus (N) and necrotic organelles
have been displaced. Scale bar = 6 µm
Berger et al.: Life cycle of Batrachochytrium dendrobatidis
autolysis, major structural changes were evident. Mat-
uration of epidermal cells was disrupted, and shedding
cells were often pale and incompletely keratinized
with overly plump nuclei. In eroded areas cells
sloughed to expose a non-keratinized surface. Many
epidermal cells were severely necrotic with swollen,
degenerate nuclei and mitochondria. Some necrotic
infected cells appeared to lack intercellular junctions,
and bacteria were common between cells and within
discharged sporangia.
SEM revealed that surface of skin from a healthy con-
trol frog was smooth and intact with a regular pattern
demarcating adjacent cells (Fig. 34), whereas skin from
the infected Litoria lesueuri exhibited roughening due
to separation of adjacent cells, irregular rounding of
their flat surface layers, and desquamation (Fig. 35).
DISCUSSION
The life cycle of Batrachochytrium dendrobatidis
has 2 main stages: the motile, waterborne, short-lived
zoospore for dispersal, and the stationary, monocentric
thallus, which develops into a zoosporangium for asex-
ual amplification (Fig. 36). B. dendrobatidis is well
adapted to living in the dynamic tissue of the stratified
epidermis. Sporangia live inside epidermal cells, ini-
tially parasitizing cells a few layers deep, and have a
rate of development that coincides with the maturing
of the cell as it moves outwards and keratinizes. They
grow initially in living cells but complete their devel-
opment in dead superficial keratinized cells that lack
organelles. Discharge tubes have the ability to merge
with and dissolve the epidermal cell membrane and
open on to the surface of the cell, usually the surface
distal from the body. These specialized adaptations
suggest that B. dendrobatidis has long evolved to live
in skin.
The distribution of sporangia in adults and tadpoles
shows that a stratified, keratinizing epidermis is a
requirement of Batrachochytrium dendrobatidis when
occurring as a parasite (Berger et al. 1998, Marantelli
et al. 2004). However, immature sporangia grow
within the deeper cells that contain prekeratin, not
61
Fig. 34. Litoria caerulea. SEM of smooth intact healthy skin.
Toe pad is in the left part of the image. Scale bar = 100 µm
Fig. 35. Batrachochytrium dendrobatidis infecting Litoria
lesueuri. SEM of the surface of skin from the foot of an infected
adult of L. lesueuri showing the extensive degeneration and
peeling of the superficial epidermis; reprinted from Berger
et al. (1999). Scale bar = 100 µm
Fig. 36. Batrachochytrium dendrobatidis. Diagram of the life
cycle in culture. After a period of motility (usually <24 h),
zoospores encyst, resorb their flagella and form germlings.
Rhizoids appear from one or more areas. The thalli grow
larger and become mature sporangia over 4 to 5 d. Contents
of the enlarged thallus become multinucleate by mitotic divi-
sions and the entire contents cleave into zoospores while the
discharge tubes form. The discharge tube is closed by a plug
that absorbs water and deliquesces when zoospores are ready
to release. Some thalli develop colonially with thin septa
dividing the contents into multiple sporangia each with their
own discharge tube. A = zoospore, B = germling, C = devel-
oping zoosporangium, D = monocentric zoosporangium, E =
colonial thallus
Dis Aquat Org 68: 51– 63, 2005
the dense keratin of the outer, cornified layer where
mature zoosporangia are found. Keratin has not been
studied in detail in amphibians, but in mammals it is
formed when the inner surface of the plasma mem-
brane thickens and material including microfilaments,
keratohyaline and lysed material are deposited into
an amorphous-filamentous complex (Fox 1994). As
cells mature, different keratins are expressed. It is not
yet known what nutrients are being utilized by spo-
rangia in frog skin, or if it is the structure and rate of
development of stratified epidermis that make it
suitable for B. dendrobatidis compared with other
epidermal types.
Because higher temperatures (i.e. >25°C) increase
the rate of epidermal turnover and reduce growth of
the amphibian chytrid (Piotrowski et al. 2004), the loss
of infection in frogs in warm conditions (Berger et al.
2004, McDonald et al. 2005) may occur because the
fungus does not have time to complete its life cyle
before being shed with the epidermal layer.
We did not see a zoospore in the act of infecting skin,
so the method of penetration remains unknown. Long-
core et al. (1999) suggest the zoospore could encyst on
the surface then inject the nucleus and contents
through a germ tube. Other chytrids have the ability to
change from endobiotic to epibiotic growth depending
on nutrients and the substrate (Longcore 1995). The
details of the ultrastructural changes that occur within
sporangia during development also remain to be stud-
ied. Experimental infection with zoospores and a time
series of fixations are needed to trace the infection
process.
Rhizoids were seen less commonly in skin sections
than in culture, suggesting they do not grow as pro-
fusely in skin. Rhizoids may not be needed in skin as
they are not required for attachment and enzymati-
cally digested nutrients may be absorbed through the
sporangial wall. Serial sectioning could be used to
follow the path of rhizoids in skin.
The clustering of sporangia of Batrachochytrium
dendrobatidis in the skin may occur because zoospores
are attracted to foci of infection, or because zoospores
that are released from a sporangium immediately
infect adjacent skin with only a limited period of
motility and dispersal. Some zoospores appear to be
released into intercellular spaces and may not be
able to escape from the site of infection.
The optimum methods for preparing germlings,
developing sporangia and zoosporangia for TEM were
high pressure freezing and freeze substitution. For
SEM the best results were obtained by imaging bulk
frozen hydrated samples via a field emission scanning
electron microscope. These protocols minimize prepa-
ration artifacts such as shrinkage, which was apparent
in the critical point dried samples.
Batrachochytrium dendrobatidis causes complex
changes at the cellular level with 2 process being seen;
(1) the ultrastructure of infected cells appears to
undergo an active process of reorganization in the
early stages of infection and (2) dissolution of cells
occur. Examination of ultrastructural pathology re-
vealed that hyperkeratosis appeared to be partly due
to an increased turnover of epidermal cells, and the
swelling of epidermal cells near foci of infection
suggested a hyperplastic response. Stimulation of the
stratum basale leading to hyperplasia is a common
response to epidermal injury and occurs with other
epidermal infections such Sarcoptes scabiei (Skerratt
et al. 1999). The layers of skin in infected frogs were
uneven and maturation of epidermal cells appeared
disrupted. The cycle of cleavage and sloughing of
keratinized cells was also disrupted with multiple lay-
ers of keratinized cells building up before being shed.
Effects on infected cells included altered fibrils in the
cytoplasm and displacement of the nucleus by sporan-
gia. These changes in the ultrastructure may lead to
reduced penetration of antifungal drugs into the cell.
Sporangia appeared to initiate premature death and
keratinization of host cells. Thinning of the epidermis
may occur when the germination of epidermal cells
does not match the increased rate of sloughing from
the surface. Epidermal width is the result of the bal-
ance between these 2 effects and is highly variable.
We did not determine whether death is caused by
toxin release or by inhibition of skin functions, but the
dissolution of cellular cytoplasm that is visible (by both
histological and TEM methods) in epidermal cells
distant to foci of infection suggests toxicity. Over-
growth of bacteria may contribute to the pathogenesis
in terminal stages. The mechanism by which Batracho-
chytrium dendrobatidis causes death remains one of
the most important aspects yet to be understood about
this pathogen. Studies on enzymes produced by B.
dendrobatidis, effects of unpurified fungal secretions
on the health of frogs, and more detailed studies on
pathology and clinical pathology of infected frogs may
provide answers.
Acknowledgements. We are grateful to N. Cheville for assis-
tance with interpretation of pathology. We thank A. Marshall,
Latrobe University, the Centre of Electron Microscopy and
Microanalysis of South Australia at the University of Ade-
laide, and the Electron Microscopy Facility at the University
of Melbourne for assistance in using their electron microscopy
facilities. We thank G. Marantelli, H. Hines, K. McDonald and
H. Parkes for providing frog specimens, and S. Hengstberger,
T. Wise, and F. Filippi for help with photography. The work
was supported by Environment Australia, the Australian
Research Council, the Department of Environment and Heri-
tage and NSF Integrated Research Challenges in Environ-
mental Biology grant IBN 9977063.
62
Berger et al.: Life cycle of Batrachochytrium dendrobatidis
LITERATURE CITED
Ardipradja K (2001) A study of resistance towards the
emerging pathogen Batrachochytrium dendrobatidis in 4
species of Australian frogs. BSc, University of Melbourne,
Melbourne
Berger L (2001) Diseases in Australian frogs. PhD thesis,
James Cook University, Townsville
Berger L, Speare R, Daszak P, Green DE and 10 others (1998)
Chytridiomycosis causes amphibian mortality associated
with population declines in the rain forests of Australia
and Central America. Proc Natl Acad Sci USA 95:
9031–9036
Berger L, Speare R, Hyatt A (1999) Chytrid fungi and amphib-
ian declines: overview, implications and future directions.
In: Campbell A (ed) Declines and disappearances of Aus-
tralian frogs. Environment Australia, Canberra, p 23 –33
Berger L, Hyatt AD, Olsen V, Hengstberger SG, Boyle D,
Marantelli G, Humphreys K, Longcore JE (2002) Production
of polyclonal antibodies to Batrachochytrium dendrobatidis
and their use in an immunoperoxidase test for chytridio-
mycosis in amphibians. Dis Aquat Org 48:213–220
Berger L, Speare R, Hines H, Marantelli G and 10 others
(2004) Effect of season and temperature on mortality in
amphibians due to chytridiomycosis. Aust Vet J 82:31–6
Bosch J, Martínez-Solano I, García-París M (2001) Evidence
of a chytrid fungus infection involved in the decline of the
common midwife toad (Alytes obstetricans) in protected
areas of central Spain. Biol Conserv 97:331–337
Boyle DG, Hyatt AD, Daszak P, Berger L, Longcore JE, Porter
D, Hengstberger SG, Olsen V (2003) Cryo-archiving
of Batrachochytrium dendrobatidis and other chytridio-
mycetes. Dis Aquat Org 56:59–64
Bradley GA, Rosen PC, Sredl MJ, Jones TR, Longcore JE
(2002) Chytridiomycosis in native Arizona frogs. J Wildl
Dis 38:206–212
Davidson EW, Parris M, Collins JP, Longcore JE, Pessier
AP, Brunner J (2003) Pathogenicity and transmission of
chytridiomycosis in tiger salamanders (Ambystoma
tigrinum). Copeia 2003:601–607
Fellers GM, Green DE, Longcore JE (2001) Oral chytridiomy-
cosis in the mountain yellow-legged frog (Rana muscosa).
Copeia 2001:945– 953
Fox H (1994) The structure of the integument. In: Heatwole HG,
Barthalmus T (eds) Amphibian biology, Vol 1. The integu-
ment. Surrey Beatty and Sons, Chipping Norton, p 1–32
Fuller MS (1996) The flagellated fungal spore. In: Sutton B
(ed) A century of mycology. Cambridge University Press,
Cambridge, p 161–192
Hanselmann R, Rodríguez A, Lampo M, Fajardo-Ramos L and
4 others (2004) Presence of an emerging pathogen of
amphibians in introduced bullfrogs Rana catesbeiana in
Venezuela. Biol Conserv 120:115–119
Hopkins S, Channing A (2003) Chytrid fungus in northern
and western cape frog populations, South Africa. Herp
Rev 34:334– 336
James TY, Porter D, Leander CA, Vilaglys R, Longcore JE
(2000) Molecular phylogenetics of the Chytridiomycota
supports the utility of ultrastructural data in chytrid
systematics. Can J Bot 78:336– 350
Johnson M, Berger L, Philips L, Speare R (2003) Fungicidal
effects of chemical disinfectants, UV light, desiccation and
heat on the amphibian chytrid, Batrachochytrium dendro-
batidis. Dis Aquat Org 57:255–260
Lips KR (1999) Mass mortality and population declines of
anurans at an upland site in western Panama. Conserv
Biol 13:117–125
Longcore JE (1993) Morphology and zoospore ultrastructure
of Lacustromyces hiemalis gen. et sp. nov. (Chytridiales).
Can J Bot 71:414– 425
Longcore JE (1995) Morphology and zoospore ultrastructure
of Entophlyctis luteolus sp. nov. (Chytridiales): implica-
tions for chytrid taxonomy. Mycologia 87:25–33
Longcore JE, Pessier AP, Nichols DK (1999) Batrachochytrium
dendrobatidis gen. et sp. nov., a chytrid pathogenic to
amphibians. Mycologia 91:219–227
Marantelli G, Berger L, Speare R, Keegan L (2004) Distribu-
tion of Batrachochytrium dendrobatidis and keratin dur-
ing tadpole development. Pac Conserv Biol 10:173–179
McDonald KR, Méndez D, Müller R, Freeman AB, Speare R
(2005) Decline in the prevalence of chytridiomycosis in
upland frog populations in North Queensland, Australia.
Pac Conserv Biol 11:114–120
Morehouse EA, James TY, Ganley ARD, Vilgaly R, Berger L,
Murphy PJ, Longcore JE (2003) Multilocus sequence
typing suggests the chytrid pathogen of amphibians is a
recently emerged clone. Mol Ecol 12:395–403
Mutschmann F, Berger L, Zwart P, Gaedicke C (2000)
Chytridiomycosis on amphibians — first report from
Europe. Berl Münch Tierärzt Wochenschr 113:380– 383
Nichols DK, Lamirande EW, Pessier AP, Longcore JE (2001)
Experimental transmission of cutaneous chytridiomycosis
in dendrobatid frogs. J Wildl Dis 37:1– 11
Pessier AP, Nichols DK, Longcore JE, Fuller MS (1999) Cuta-
neous chytridiomycosis in poison dart frogs (Dendrobates
spp.) and White’s tree frogs (Litoria caerulea). J Vet Diagn
Invest 11:194–199
Piotrowski JS, Annis SL, Longcore JE (2004) Physiology of
Batrachochytrium dendrobatidis, a chytrid pathogen of
amphibians. Mycologia 96:9–15
Retallick RW, McCallum H, Speare R (2004) Endemic in-
fection of the amphibian chytrid fungus in a frog com-
munity post-decline. PLoS Biol 2:e351(doi: 10.1371/journal.
pbio.0020351)
Rollins-Smith LA, Carey C, Longcore J, Doersam JK, Boutte
A, Bruzgal JE, Conlon J M (2002) Activity of antimicrobial
skin peptides from ranid frogs against Batrachochytrium
dendrobatidis, the chytrid fungus associated with global
amphibian declines. Dev Comp Immunol 26:471–479
Skerratt LF, Middleton D, Beveridge I (1999) Distribution of
life cycle stages of Sarcoptes scabiei var wombati and
effects of severe mange on common wombats in Victoria.
J Wildl Dis 35:633– 646
Speare R, Alford R, Aplin R, Berger L and 11 others (2001)
Nomination for listing of amphibian chytridiomycosis as a
key threatening process under the Environment Protec-
tion and Biodiversity Conservation Act 1999. In: Speare R,
Alford R, Aplin R, Berger L and 10 others (eds) Developing
management strategies to control amphibian diseases: de-
creasing the risks due to communicable diseases. School
of Public Health and Tropical Medicine, James Cook
University, Townsville, p 186–208
Weldon C, du Preez LH, Hyatt AD, Muller R, Speare R (2004)
Origin of the amphibian chytrid fungus. Emerg Infect Dis
10:2100–2105
Woodhams DC, Alford RA, Marantelli G (2003) Emerging dis-
ease cured by elevated body temperature. Dis Aquat Org
55:65– 67
63
Editorial responsibility: Peernel Zwart,
Utrecht, The Netherlands
Submitted: March 27, 2005; Accepted: June 22, 2005
Proofs received from author(s): November 25, 2005
