, E. A. V. Zauza
, M. J. Wingﬁeld
and C. Mohammed
Ensis Forest Biosecurity & Protection, Private Bag 12, Hobart, Tas. 7001, Australia.
Department of Plant Pathology, Federal University of Vic¸osa, Vic¸osa, MG 36571-000, Brazil.
Forestry and Agricultural Biotechnology Institute, University of Pretoria, Pretoria, South Africa.
University of Tasmania, Private Bag 12, Hobart, Tas. 7001, Australia.
Corresponding author. Email: firstname.lastname@example.org
is native to South America where it can cause severe disease in eucalypt plantations and other introduced Myrtaceae.
The pathogen has recently expanded its geographical range to Hawaii, increasing concerns about the potential for an
incursion in Australia. This paper reviews the taxonomy, biology, impact and options for control of P. psidii. It also
discusses the probable impact if an incursion were to occur in Australia and the preparations that must be made to mitigate
Several papers calling attention to Puccinia psidii as a
biosecurity threat to Australia and New Zealand have been
published in recent years (Navaratnam 1986; Ridley et al. 2000;
Mireku and Simpson 2002; Tommerup et al. 2003). Puccinia
psidii is a native of South and Central America where it was ﬁrst
described on guava (Winter 1884), hence its vernacular name of
guava rust. In recent years, research has been conducted into the
taxonomy, biology, host range, actual and potential distribution
and options available for control of P. psidii. This review attempts
to encapsulate the reasons why this organism is considered to be
a threat to Australia and to summarise the research that has been
published since the last review (Coutinho et al. 1998).
J.A. Simpson, K. Thomas & C.A. Grgurinovic have many
synonyms (Table 1). A recent review of species of Uredinales
pathogenic on species of Myrtaceae (Simpson et al. 2006)
described eight rust species, including U. psidii, U. rangelii
and U. seclusa, which are all anamorphs of P. psidii sensu
are Phakopsora rossmaniae and its anamorph Physopella
The only two species that are known from Australia are
= Uredo xanthostemonis), on Xanthostemon
which was unknown in Australia until it was detected by
New Zealand quarantine on a shipment of cut ﬂowers and was
subsequently found on Kunzea ericifolia near Perth in Western
Australia (Shivas and Walker 1994). Shivas and Walker (1994)
were of the opinion that P. cygnorum is quite distinct from
such as P. boroniae; rDNA sequences support this (M. Glen,
unpubl. data). It thus seems possible that other undiscovered
rusts exist on Myrtaceae hosts in Australia.
Jamaica (Simpson et al. 2006). Its status as a species distinct
from U. psidii is based on the presence of a tonsure on the
urediniospores and subtle differences in size, shape and wall
thickness of the urediniospore. However, molecular phylogenetic
and morphological analyses of further collections should be
pursued to support the morphological distinction. Uredo seclusa
is known only from the type collection, on an unidentiﬁed species
of Myrtaceae from S˜ao Paulo, Brazil. Phakopsora rossmaniae
and Physopella jueli are also known only from Brazil, on species
of Campomanesia. A full description of spores and a key to rusts
on Myrtaceae is provided in Simpson et al. (2006).
Symptoms on a range of hosts have been described and illustrated
(Coutinho et al. 1998; Tommerup et al. 2003; Alfenas et al.
2004) and appear on various websites (Agricultural Research
Service USDA 2006; PaDIL 2006; University of Hawaii 2006).
Lesions are produced on young, actively growing leaves and
shoots, as well as on fruits and sepals (Figs 1 and 2). Lesions are
brown to grey with masses of bright yellow or orange-yellow
urediniospores. Occasionally, lesions have sori containing dark
brown teliospores or a mixture of the two spore types. Older
lesions have purpling of their margins on leaves and shoots of
many Eucalyptus, Melaleuca and Callistemon hosts. Lesions on
ﬂeshy fruits of Eugenia, Psidium and Syzygium may not have
obvious margins due to their being covered with heavy spore
masses when young and rot caused by secondary pathogens
as the fruits ripen. Severe rust disease in young trees may kill
shoot tips, causing loss of leaders and a bushy habit. Proliﬁc
© Australasian Plant Pathology Society 2007
M. Glen et al.
(Arthur & Mains)
Puccinia actinostemonis H. S.
Jackson & Holway
erroneously reported as
P. barbacensis Rangel
Myrtaceae, genus not identiﬁed
Species of Eugenia, Marlierea,
Myrtaceae indeterminate, originally
reported as Acacia sp.
U. rangelii J. A. Simpson,
K. Thomas & C. A. Grgurinovic
branching and galling in eucalypts is a symptom of previous
rust infection. Persistent localised lesions and stem swellings on
Melaleuca quinquenervia have also been reported and illustrated
(Rayachhetry et al. 2001
). Similar symptoms may occur in other
species but have not been recorded because many host species
have been tested only at the seedling stage.
Symptoms on resistant hosts
On resistant plants, the pathogen may induce a hypersensitive
reaction (HR) expressed as ﬂecks or necrotic lesions generally
with no sporulation (Junghans et al. 2003) However, depending
on the level of resistance, punctiform pustules may be formed
over the brown, necrotic lesions. This type of reaction is typical
of a single gene controlling resistance, as previously detected
in E. grandis (Junghans et al. 2003) and in several other pure
species and hybrids (A. C. Alfenas, unpubl. data).
Puccinia psidii is considered to be an autoecious species
with an incomplete lifecycle (Fig. 3). With the exception of
spermogonia, all stages are produced on the same Myrtaceous
host. Aecia and aeciospores are morphologically identical to
uredinia and urediniospores (Figueiredo 2001; A. C. Alfenas
and E. A. V. Zauza, unpubl. data). It has recently been suggested
that P. psidii may be heteroecious with an unknown aecial
host (Simpson et al. 2006) but this seems doubtful given the
multiple observations, in independent laboratories, of infections
on uredinial hosts (E. grandis and S. jambos) inoculated with
teliospores or basidiospores (Figuiredo 2001; A. C. Alfenas and
E. A. V. Zauza, unpubl. data).
Under natural conditions, P. psidii produces abundant
comparatively rare, although teliospores are more frequent on
hosts. Frequency on all hosts is higher in warmer months
(Ferreira 1983). Aeciospores have not been observed or
recognised in nature due to their similarity to urediniospores
Production of teliospores can be stimulated by incubation
of infected hosts at temperatures outside the optimal range
for urediniospore production. Ruiz et al. (1989b) found that
the number of urediniospores and teliospores produced on
inoculated E. grandis was signiﬁcantly higher at 20
C and 25
than at 30
C. Alfenas et al. (2003) noted teliospore formation
on Eucalyptus globulus and E. viminalis at 28
C but not at 22
With a variable temperature regimen, Aparecido et al. (2003b)
between 21 and 35
Basidiospores have been produced free of urediniospores
(Figueiredo 2001). Eighteen days after inoculation, aecia
and aeciospores were produced that were morphologically
Spermogonia, however, have not been observed (Figueiredo
2001). The current understanding of the P. psidii life cycle is
illustrated in Fig. 3.
Conditions for germination and infection
Urediniospore germination and infection are affected by
temperature, leaf wetness, light intensity and photoperiod (Ruiz
et al. 1989b). Several studies have agreed that high humidity
or leaf wetness and low light for a minimum of 6 h following
inoculation are necessary for successful germination and
infection (Piza and Ribeiro 1988; Ruiz et al. 1989a, 1989c).
Several studies have determined different optimum temperatures
for urediniospore germination. Lack of consistency may have
been caused by variation in methodology among the studies.
Variable factors included the substrate on which germination
occurred, the length of incubation before germination was
assessed and even the type of water (or oil) in which the
spores were resuspended. Suspension of the spores in mineral
oil rather than water increased the germination rate (Furtado
This inconsistency among studies may also be due to variation
among the rust biotypes. In one study, a temperature range of
(1) susceptible, (2) resistant and (3) hypersensitive response.
C gave the highest germination rate for urediniospores
germination rate at 15
C (Aparecido et al. 2003a
Light exposure during the initial stages of infection
Rust on other Myrtaceae species. (a, b) Infected fruit of Eugenia uniﬂora. (c, d ) Flower buds and guava fruit. (e, f ) Urediniosori on leaves and ﬂower
buds of Syzygium jambos.
produced on E. grandis seedlings exposed to 3640 lx than on
those exposed to 1092 lx (Ruiz et al. 1989a
, 1989b). In this
study, infection and spore production were not affected by the
source (host species or location) of the inoculum.
Germination on host leaves is also more proliﬁc than
on water agar. Tessmann and Dianese (2002) investigated
whether plant compounds may have a stimulatory effect on
urediniospore germination and found that germination was
Basidiospore germination, host penetration,
haustorium development, colony and aeciosori
formation with aeciospores
Inoculation of young
Aeciospore germination, host
development, colony and
colony and teliosori
host penetration, urediniosori
Schematic life cycle of Puccinia psidii
enhanced by an extract from leaves of S. jambos. The stimulatory
compound was identiﬁed as the hydrocarbon, hentriacontane.
Low concentrations of hentriacontane (20–200 mg/L) almost
doubled the germination rate, but higher concentrations
(2000–20 000 mg/L) did not lead to a germination rate greater
than could be obtained in pure mineral oil. The authors suggest
that certain hydrocarbons may have a role in overcoming
self-inhibition and this may account for the effect of both
mineral oil and hentriacontane. In contrast, Salustiano et al.
and those of three other rusts.
Teliospores germinated in vitro and basidiospores were
produced at temperatures ranging from 12 to 24
C, basidiospore production occurred after 48 h compared
A histopathological study of the infection process on detached
leaves revealed no difference between susceptible and resistant
E. grandis genotypes during the processes of urediniospore
germination, appressorium formation and host penetration
(Xavier et al. 2001). Ninety percent of the urediniospores had
germinated within 6 h of inoculation and 90% of these formed
appressoria within 18 h whereas a low percentage entered
through stomata without appressorium formation. Infection pegs
from the appressoria penetrated between the anticlinal walls of
the epidermal cells and colonised the mesophyll, as previously
reported for S. jambos (Hunt 1968). In resistant genotypes, a
hypersensitive reaction was seen after 48 h, in contrast with
susceptible genotypes, where macroscopic disease symptoms
were observed 3–5 days after inoculation and urediniospore
formation after 12 days.
The environmental conditions important for the in vitro or
be strongly correlated with rust epidemics in the ﬁeld (Ruiz
Studies conducted on E. cloeziana coppice showed that rust
progress and severity varied from year to year according to
the environmental conditions. Periods of high relative humidity
longer than 8 h and temperatures in the range of 15–25
; Carvalho et al.
1994). In a year-long study on S. jambos in central Brazil
(Tessmann et al. 2001), disease incidence and severity were
highly correlated with periods of relative humidity over 90% or
leaf wetness periods greater than 6 h and nocturnal temperatures
between 18 and 22
C. In another study, Blum and Dianese (2001)
airborne urediniospores, the number of infected young S. jambos
shoots and the number of nights with temperatures below
was observed between midday temperature and the number of
Knowledge of the potential survival time of the different spore
types is vital to assessing the risk associated with various
pathways possible for an incursion. Aparecido et al. (2003a)
examined the germination rate of urediniospores at ﬁve ages (10,
14, 21, 28 and 34 days old) from S. jambos and Psidium guajava.
Germination was assessed after a 5-h incubation on water agar
at temperatures ranging from 12 to 24
C. They found that
those from P. guajava, and that 14-day-old urediniospores from
C, had the highest germination rate of
was 30% for 21-day-old urediniospores incubated at 15
No germination was recorded for 34-day-old urediniospores
recorded a germination rate of 3% for 31-day-old urediniospores
from S. jambos. An even longer survival time was recorded
for urediniospores from P. guajava, stored at 4
C and 40%
viability, albeit low at 3%, after 100 days. In this study,
germination was assessed after a 48 h incubation at 20
germination. A factorial experiment was also conducted by
Suzuki and Silveira (2003) to develop a model of survival time
based on temperature and relative humidity.
In current studies, urediniospores from Eucalyptus spp.
maintained viability after 90 days at 15
C and 35–55% relative
C (V. M. Lana,
of urediniospores during a sea voyage from South America
to Australia, with temperatures around 30
C and 70% RH,
for the spores, while still viable, to reach a susceptible
host in suitable environmental conditions for germination.
Theoretically, teliospores survive longer than urediniospores,
although less empirical information is available for teliospores
of P. psidii.