No detection of chytrid in first systematic screening of Bombina variegata pachypus (Anura: Bombinatoridae) in Liguria, northern Italy

Abstract

The Apennine Yellow-bellied toad Bombina variegata pachypus, a small anuran endemic to peninsular Italy, has been declining throughout its range over the last 30 years. Although mortality by chytridiomycosis, caused by the fungus Batrachochytrium dendrobatidis, was first reported for the species in 2004, its role in the decline has not yet been assessed. Between 2011 and 2012 we sampled eight populations of B. v. pachypus in Liguria, northern Italy, swabbing 86 and 143 individuals respectively, corresponding to between 24 and 80% of the estimated individuals within each population. We did not detect chytrid in any the samples collected. For the three largest populations in the region, we can rule out infections of prevalence greater than 10% with at least 98% confidence. Research at a larger scale is urgently needed to clarify the role of B. dendrobatidis in the decline of this and other amphibians in Italy.

Full text

ISSN 1827-9635 (print) © Firenze University Press
ISSN 1827-9643 (online) www.fupress.com/ah
Acta Herpetologica 8(1): 59-63, 2013
No detection of chytrid in rst systematic screening of Bombina variegata
pachypus (Anura: Bombinatoridae) in Liguria, northern Italy
S C,*, A M, F P
1 ARC Centre of Excellence for Environmental Decisions, School of Botany, University of Melbourne, 3010 VIC, Australia. *Correspond-
ing author. E-mail: canessas@unimelb.edu.au
2 Faculty of Veterinary Medicine, Ghent University, Merelbeke 9820, Belgium
Submitted on: 2012, 7th Decembre; revised on: 2013, 10th April; accepted on: 2013, 19th April.
Abstract. e Apennine Yellow-bellied toad Bombina variegata pachypus, a small anuran endemic to peninsular Italy,
has been declining throughout its range over the last 30 years. Although mortality by chytridiomycosis, caused by
the fungus Batrachochytrium dendrobatidis, was rst reported for the species in 2004, its role in the decline has not
yet been assessed. Between 2011 and 2012 we sampled eight populations of B. v. pachypus in Liguria, northern Ita-
ly, swabbing 86 and 143 individuals respectively, corresponding to between 24 and 80% of the estimated individuals
within each population. We did not detect chytrid in any the samples collected. For the three largest populations in
the region, we can rule out infections of prevalence greater than 10% with at least 98% condence. Research at a larg-
er scale is urgently needed to clarify the role of B. dendrobatidis in the decline of this and other amphibians in Italy.
Keywords. Bombina, chytrid, condence, Liguria, monitoring, population size, prevalence.
e amphibian chytrid fungus Batrachochytrium dend-
robatidis is the main pathogen driver of the global amphib-
ian decline (Fisher et al., 2009). Die-os and extinctions
have occurred in all continents (Berger et al., 1998; Con-
radie et al., 2011; Swei et al., 2011) including several cas-
es in Europe (Garner et al., 2005). In Italy, infection by
chytridiomycosis has been documented for several species
in the last decade (reviewed in Tessa et al., 2013): further
information on the distribution of B. dendrobatidis in Italy
is needed to understand the causes of local and regional
declines and inform conservation eorts.
e Apennine yellow-bellied toad Bombina variegata
pachypus (Anura: Bombinatoridae) was the rst amphib-
ian species in Italy with conrmed mortality by chytridi-
omycosis, reported by Stagni et al. (2004) on captive
individuals caught from three populations in the north-
eastern Apennines. Recent evidence suggests chytrid is
present in populations of B. v. pachypus along the Ital-
ian peninsula, with infections dating back as early as the
late 1970s (Canestrelli et al., 2013). B. v. pachypus has
declined steeply over the last 30 years, with widespread
local disappearances: although habitat loss is believed to
represent the main driver of the decline (Mirabile et al.,
2009; Canessa et al., 2013), chytridiomycosis is accepted
as a potential threat to the species (see for example its
Red List assessment in IUCN, 2011). However, to our
knowledge, no systematic assessment of the spread of
chytrid among populations of B. v. pachypus in Italy has
been undertaken to date.
In order to clarify this issue at least in part, we con-
ducted surveys throughout the northern Italian region of
Liguria in 2011 and 2012. A recent assessment of the sta-
tus of B. v. pachypus in this area showed that over 50%
of the known populations disappeared between 2001 and
2010: habitat loss relating to the abandonment of tradi-
tional farming practices has been suggested as the main
cause of decline (Canessa et al., 2013). A small-scale sur-
vey (ve individuals tested) carried out in 2005 at a single
site in the region failed to detect any presence of B. dend-
robatidis (Adams et al., 2008).

60 Stefano Canessa et al.
In 2011, between 31 July and 4 August we visited all
seven known breeding sites of B. v. pachypus in Liguria,
located within the catchments of the Lavagna and Vara
rivers in the eastern part of the region, encompassing an
area of approximately 800 km2. In 2012, we repeated our
visits at ve of these seven populations: the three sites
believed to host the largest populations were sampled
three times during the 2012 season in order to maxi-
mize the overall proportion of the population that could
be tested (22 June, 24 July and 26 July). We also tested
all the animals present at a small captive breeding centre
within the study area: these individuals had been caught
in the wild and kept in a semi-natural enclosure with no
contact with other amphibians during the study period.
We captured all detected individuals by hand and
photographed them to record their unique ventral pat-
tern: this allowed us to calculate the number of unique
individuals tested in addition to the total number of
swabs collected at sites that were visited multiple times.
To minimize the chance of spreading pathogens, we fol-
lowed recommendations by Speare et al. (2004) when
handling individuals, and sites within dierent catch-
ments were never sampled during the same day. We col-
lected samples as described by Hyatt et al. (2007) using
cotton-tip swabs and then released individuals at the
point of capture. Samples were stored at –4 °C until fur-
ther analysis. We only tested post-metamorphic indi-
viduals: although pathogen prevalence can be higher in
tadpoles, facilitating detection, we chose to avoid any
impact to the few remaining populations of the species in
the region that could result from applying lethal or non-
lethal sampling to larvae. Swabbing is less sensitive than
other methods, and can still damage tadpole mouthparts
and facilitate infections in addition to stress resulting
from handling (Retallick et al., 2006).
B. dendrobatidis DNA in the samples, real-time
Taqman PCR assays were conducted on a CFX96 Real
Time System (BioRad). Amplication conditions, primer
and probe concentrations were according to Boyle et al.
(2004). Due to economic constraints, samples were run in
duplicate. Consistent results between the two runs were
classied as correct: we considered a test result as posi-
tive where both qPCR runs gave results above 0.1 GE,
with standard deviation smaller than the average of both
repeats and suitable amplication curves.
After obtaining the test results, we estimated the
mean prevalence (with 95% credible intervals) of the
pathogen in the population using a Bayesian equal-tailed
Jerey prior (Brown et al., 2001). To assess the sensitivity
of our sampling program, we then calculated the proba-
bility of detecting 5, 10 and 30% infections with the num-
ber of individuals sampled at each site, by solving
C = 1 – (1 – p)n
where C is the probability of detecting at least one infect-
ed individual from n tested in a population with p preva-
lence of the disease (DiGiacomo and Koepsell, 1986).
Additionally, for the three largest populations in the
study we were able to use an existing multi-year mark-
recapture database to estimate population size at the
time of sampling (Canessa, unpubl. data). To account
for open-population dynamics such as temporary emi-
gration, oen observed for B. variegata (Hartel, 2008),
we used an open-population Jolly-Seber model (Wagner
et al., 2011). We t the model in a Bayesian framework
using date- and population-dependent recapture rates
(code in Kéry and Schaub, 2011). We used JAGS (Plum-
mer, 2005) to obtain samples from three Markov chains
over 100,000 iterations, aer discarding the rst 10,000 as
a burn-in.
In 2011, we collected 81 swabs from 81 individuals in
seven wild populations, corresponding to between 24 and
42% of the estimated sizes of respective populations (as
estimated from 95% condence intervals from the Jolly-
Seber model): at two sites we captured only one and four
individuals respectively (Table 1). In 2012, we captured
and swabbed a total of 143 individuals at four of the
wild locations on four separate occasions, correspond-
ing to between 66 and 80% of the estimated sizes of the
tested populations. Several individuals were re-captured
and swabbed in two or three occasions in 2012, yielding
a total of 224 swabs for that season. All individuals at the
captive breeding centre were also caught and tested (four
in 2011 and three in 2012): we did not calculate con-
dence intervals or sampling power for this site.
All samples tested negative for B. dendrobatidis. We
captured more than 10 individuals only at four sites: in
2011, this resulted in upper 95% condence intervals (CI)
of prevalence between 0.09 and 0.13 (Table 1). For three
of these populations, in 2012 we were able to reduce the
upper CIs to 0.05 or less; at the fourth site, only seven
individuals were captured in 2012, resulting in an upper
CI of 0.23. For the three most intensively sampled pop-
ulations, we can be at least 97% condent that chytrid
would have been detected at these populations for a
prevalence of at least 10%, and more than 83% condent
for a prevalence of at least 5% (Table 1). For the smaller
populations sampled (less than 10 individuals), we could
make no reliable inference regarding the prevalence of
infection, although in sites with more than one capture,
the probability of having missed large infections (> 30%
prevalence) was still lower than 50%.
Although reliable estimates of abundance could not
be obtained for the smaller populations in the study, we

61
No detection of chytrid in Ligurian B. variegata
observed relatively high recapture rates (> 40%) for the
larger populations, suggesting the small sample sizes at
some sites might reect actual abundances. Nevertheless,
the estimates we present for populations with less than 10
sampled individuals can only be considered as prelimi-
nary and do not allow reliable inference. For the largest
populations in the region the large sample sizes, particu-
larly for 2012, allows us to rule out with sucient con-
dence infections of prevalence greater than 10%.
Additional information about the expected preva-
lence of chytrid infection in B. v. pachypus is needed to
make more precise inference: for example. Canestrelli et
al. (2013) recently found that prevalence of B. dendroba-
tidis in present and historical specimens of B. v. pachy-
pus across the Italian peninsula we never below 12%.
Amongst populations of B. variegata in Hungary, the
infection rate has been estimated between 12% and 29%
(Vörös, pers. comm.). In the Netherlands and Flanders
(Belgium), Spitzen-van der Sluijs et al. (2010) reported a
prevalence of infection of 5.9% for B. variegata out of 255
individuals sampled. Holding this relatively high base rate
as a guideline, the failure to detect chytrid in the Ligurian
populations may indicate lower susceptibility due to local
conditions, bias related to the date of sampling (patterns
of chytrid infection are highly seasonal: see Kriger and
Hero, 2007), or possibly actual absence of the pathogen
from the region at present. In addition, breeding sites of
B. v. pachypus in Liguria have been regularly monitored
throughout the breeding season since 2008 and no obvi-
ous sign of chytrid infection (e.g. individual mortality or
abnormal behaviour) has ever been reported (Arillo et al.,
2009; Arillo et al., 2011).
It is dicult to infer whether chytridiomycosis might
have had a role in past local extinctions in the region.
Most disappearances in the region in recent years have
occurred at sites in agricultural areas (Canessa et al.,
2013). However, the existing literature suggests such sites
might be less likely to host permanent foci of chytrid
infection, due to their low elevation and open vegeta-
tion types, resulting in higher water temperatures (> 30
°C in August) and regular desiccation, unsuitable for the
pathogen (Johnson et al., 2003). For example, Hyne et al.
Table 1. Results for all screened populations of B. v. pachypus in 2011 and 2012.


Population
Fav Lo1 Lo2 LoP Pav Per Pin Te v
2011
Total swabs14 17 21 8 17 1 13 1
Individuals swabbed24 17 21 8 17 1 13 1
Population size3-68 ± 2 67 ± 2 -35 ± 4 -47 ± 3 -
% tested4- 25.0 31.3 - 48.6 - 27.7 -
Prevalence (95% CI)50.36 0.11 0.09 0.21 0.11 0.77 0.13 0.77
Detection p = 0.0560.19 0.58 0.66 0.34 0.58 0.05 0.49 0.05
p = 0.170.34 0.83 0.89 0.57 0.83 0.10 0.75 0.10
p = 0.380.76 1.00 1.00 0.94 1.00 0.30 0.99 0.30
2012
Total swabs1- 106 60 - 7 - 51 -
Individuals swabbed2- 55 43 - 7 - 35 -
Population size3-69 ± 2 69 ± 2 -35 ± 4 -48 ± 2 -
% tested4- 79.7 62.3 - 20.0 - 72.9 -
Prevalence (95% CI)5- 0.04 0.04 - 0.23 - 0.05 -
Detection p = 0.056- 0.94 0.89 - 0.30 - 0.83 -
p = 0.17- 1.00 0.99 - 0.52 - 0.97 -
p = 0.38- 1.00 1.00 - 0.92 - 1.00 -
1 Total number of swabs collected
2 Total numbers of unique individuals swabbed, where repeated visits were carried out (Lo1, Lo2 and Pin in 2012)
3 Population size at the time of sampling (± standard deviation), estimated using the Jolly-Seber open population model – a dash indicates
no estimate could be made for that population/year
4 Proportion of the estimated population size (mean) that was tested
5 Upper 95% equal-tailed Jerey prior condence interval (lower interval is 0 in all cases).
6, 7, 8 Probability of detecting at least one infected individual if the pathogen were present in the population with a prevalence of 0.05, 0.1
and 0.3 respectively.

62 Stefano Canessa et al.
(2009) found lowland species persisting without notice-
able population declines in the presence of chytrid in an
agricultural landscape with relatively warm climate. Con-
versely, particularly high impacts of chytrid worldwide
have occurred in forest areas with permanent hydrop-
eriods and cooler mesoclimates (Berger et al., 1998): all
remaining large populations of B. v. pachypus in Liguria
occur in such areas and therefore require special atten-
tion. At the moment, monitoring must continue across
all known sites to ensure prompt detection of possible
outbreaks. Quantifying the eciency of monitoring (e.g.
capture rates of individuals and test sensitivities) can help
inform screening programs to achieve the best results for
the eort invested (Lachish et al., 2012).
Although B. v. pachypus is recognized as an endan-
gered species in Italy, this is, to our knowledge, the rst
systematic screening of any of its populations for B. den-
drobatidis. Conservation plans for the species include
habitat restoration and ex-situ initiatives (Di Cerbo and
Ferri, 2002; Gentilli et al., 2002; Arillo et al., 2011). Addi-
tional information about the role of chytridiomycosis
in the decline of B. v. pachypus is urgently needed, both
to inform actions (for example, habitat restoration may
not be sucient where disease is the main threat) and
to raise awareness of the need to implement safety pro-
tocols (Phillott et al., 2010). In general, further research
about the distribution and prevalence of chytrid amongst
amphibian species in Italy should be seen as a prior-
ity, particularly in areas of known and ongoing declines.
Surveys at the local and regional scale, whilst relatively
inexpensive, can provide valuable information about the
spread and impact of the pathogen and help optimize
conservation strategies.
ACKNOWLEDGEMENTS
We thank A. Morán Ordoñez, F. Oneto, D. Ottonello, P.
Piana and S. Salvidio for help with surveys. Comments from
two anonymous referees helped improve an earlier version of
the manuscript. Capture and handling permits granted by the
Ministero dell’Ambiente e della Tutela del Territorio (DPN-
2010-370, 12-01-297 2010). Field work was supported by fund-
ing from the Societas Europaea Herpetologica and the Society
for the Study of Amphibians and Reptiles. Manuscript prepa-
ration was supported by the University of Melbourne and the
ARC Centre of Excellence for Environmental Decisions.
REFERENCES
Adams, M.J., Galvan, S., Scalera, R., Grieco, C., Sindaco,
R. (2008): Batrachochytrium dendrobatidis in amphib-
ian populations in Italy. Herpetol. Rev. 39: 324-326.
Arillo, A., Braida, L., Canessa, S., Ferravante, C., Oneto,
F., Ottonello, D., Salvidio, S. (2011): Azioni di sal-
vaguardia di Bombina pachypus (Bonaparte, 1838) in
Liguria. Pianura 27: 171-173.
Arillo, A., Braida, L., Canessa, S., Oneto, F., Ottonello, D.,
Raineri, A. (2009): Problematiche di conservazione delle
popolazioni di Bombina pachypus (Bonaparte, 1838) in
Liguria. Boll. Mus. Ist. Biol. Univ. Genova 71: 89.
Berger, L., Speare, R., Daszak, P., Green, D.E., Cunning-
ham, A.A., Goggin, C.L., Slocombe, R., Ragan, M.A.,
Hyatt, A.D., McDonald, K.R., Hines, H.B., Lips, K.R.,
Marantelli, G., Parkes, H. (1998): Chytridiomycosis
causes amphibian mortality associated with population
declines in the rain forests of Australia and Central
America. Proc. Natl. Acad. Sci. U. S. A. 95: 9031-9036.
Boyle, D.G., Boyle, D.B., Olsen, V., Morgan, J.A.T., Hyatt,
A.D. (2004): Rapid quantitative detection of chytridio-
mycosis (Batrachochytrium dendrobatidis) in amphib-
ian samples using real-time Taqman PCR assay. Dis.
Aquat. Org. 60: 141-148.
Brown, L.D., Cai, T.T., DasGupta, A. (2001): Interval
estimation for a binomial proportion. Stat. Sc. 16:
101-117.
Canessa, S., Oneto, F., Ottonello, D., Arillo, A., Salvidio,
S. (2013): Land abandonment may reduce distur-
bance and aect the breeding sites of an endangered
amphibian in northern Italy. Oryx 47: 280-287.
Canestrelli, D., Zampiglia, M., Nascetti, G. (2013): Wide-
spread occurrence of Batrachochytrium dendrobatidis
in contemporary and historical samples of the Endan-
gered Bombina pachypus along the Italian peninsula.
PloS One: e63349.
Conradie, W., Weldon, C., Smith, K.G., Du Preez, L.H.
(2011): Seasonal pattern of chytridiomycosis in com-
mon river frog (Amietia angolensis) tadpoles in the
South African Grassland Biome. Afr. Zool. 46: 95-102.
Di Cerbo, A.R., Ferri, V. (2002): Guidelines for the
improvement of aquatic habitats of Bombina variegata
and Bombina pachypus. Riv. Idrobiol. 40: 73-84.
DiGiacomo, R.F., Koepsell, T.D. (1986): Sampling for
detection of infection or disease in animal popula-
tions. J. Am. Vet. Med. Assoc. 189: 22-23.
Fisher, M.C., Garner, T.W.J., Walker, S.F. (2009): Global
emergence of Batrachochytrium dendrobatidis and
Amphibian chytridiomycosis in space, time, and host.
Annu. Rev. Microbiol. 63: 291-310.
Garner, T.W.J., Walker, S., Bosch, J., Hyatt, A.D., Cun-
ningham, A.A., Fisher, M.C. (2005): Chytrid fungus
in Europe. Emerg. Infect. Dis. 11: 1639-1641.
Gentilli, A., Scali, S., Barbieri, F., Bernini, F. (2002): A
three-year project for the management and the con-
servation of amphibians in Northern Italy. Biota 3:
27-33.

63
No detection of chytrid in Ligurian B. variegata
Hartel, T. (2008): Movement activity in a Bombina varie-
gata population from a deciduous forested landscape.
North-Western J. Zool. 4: 79-90.
Hyatt, A.D., Boyle, D.G., Olsen, V., Boyle, D.B., Berger,
L., Obendorf, D., Dalton, A., Kriger, K.M., Hero, J.M.,
Hines, H.B. (2007): Diagnostic assays and sampling
protocols for the detection of Batrachochytrium dend-
robatidis. Dis. Aquat. Org. 73: 175-192.
Hyne, R.V., Spolyarich, N., Wilson, S.P., Patra, R.W.,
Byrne, M., Gordon, G., SánchezBayo, F., Palmer,
C.G. (2009): Distribution of frogs in rice bays within
an irrigated agricultural area: links to pesticide usage
and farm practices. Environ. Toxicol. Chem. 28:
1255-1265.
IUCN. (2011): Bombina pachypus. In: IUCN Red List of
reatened Species, version 2011.2. 2011.2. Retrieved
27 January 2012.
Johnson, M.L., Berger, L., Phillips, L., Speare, R. (2003):
Fungicidal effects of chemical disinfectants, UV
light, desiccation and heat on the amphibian chytrid,
Batrachochytrium dendrobatidis. Dis. Aquat. Org. 57:
255-260.
Kéry, M., Schaub, M. (2011): Bayesian population analy-
sis using WinBUGS: a hierarchical perspective. Aca-
demic Press, Burlington.
Kriger, K.M., Hero, J.M. (2007): Large-scale seasonal vari-
ation in the prevalence and severity of chytridiomyco-
sis. J. Zool. 271: 352-359.
Lachish, S., Gopalaswamy, A.M., Knowles, S.C.L., Shel-
don, B.C. (2012): Site-occupancy modelling as a novel
framework for assessing test sensitivity and estimating
wildlife disease prevalence from imperfect diagnostic
tests. Methods Ecol. Evol. 3: 339-348.
Mirabile, M., Melletti, M., Venchi, A., Bologna, M.A.
(2009): e reproduction of the Apennine yellow-
bellied toad (Bombina pachypus) in central Italy.
Amphibia-Reptilia 30: 303-312.
Phillott, A.D., Speare, R., Hines, H.B., Skerratt, L.F., Mey-
er, E., McDonald, K.R., Cashins, S.D., Mendez, D.,
Berger, L. (2010): Minimising exposure of amphibians
to pathogens during eld studies. Dis. Aquat. Org. 92:
175-185.
Plummer, M. (2005): JAGS: just another Gibbs sampler.
Proceedings of the 3rd International Workshop on
Distributed Statistical Computing (DSC 2003), Vien-
na, Austria.
Retallick, R.W., Miera, V., Richards, K.L., Field, K.J., Col-
lins, J.P. (2006): A non-lethal technique for detecting
the chytrid fungus Batrachochytrium dendrobatidis on
tadpoles. Dis. Aquat. Org. 72: 77.
Speare, R., Berger, L., Skerratt, L.F., Alford, R., Mendez,
D., Cashins, S., Kenyon, N., Hauselberger, K., Rowley,
J. (2004): Hygiene protocol for handling amphibians
in eld studies. Unpublished report by the Amphibian
Diseases Group, James Cook University, Townsville.
Spitzen-van der Sluijs, A., Zollinger, R., Bosman, W., Pas-
mans, F., Martel, A., Van Rooij, P., Clare, F. (2010):
Short report: Batrachochytrium dendrobatidis in
amphibians in the Netherlands and Flanders (Bel-
gium). Stichting RAVON. Nijmegen, the Netherlands.
Stagni, G., Dall’olio, R., Fusini, U., Mazzotti, S., Scoccian-
ti, C., Serra, A. (2004): Declining populations of apen-
nine yellow-bellied toad Bombina pachypus in the
northern Apennines (Italy): is Batrachochytrium dend-
robatidis the main cause? Ital. J. Zool. 71: 151-154.
Swei, A., Rowley, J.J.L., Rodder, D., Diesmos, M.L.L.,
Diesmos, A.C., Briggs, C.J., Brown, R., Trung Tien,
C., Cheng, T.L., Chong, R.A., Han, B., Hero, J.M., Huy
Duc, H., Kusrini, M.D., Duong i uy, L., McGuire,
J.A., Meegaskumbura, M., Min, M., Mulcahy, D.G.,
Neang, T., Phimmachak, S., Rao, D., Reeder, N.M.,
Schoville, S.D., Sivongxay, N., Srei, N., Min, M.S., Rao,
D.Q. (2011): Is chytridiomycosis an emerging infec-
tious disease in Asia? PLoS ONE: e23179.
Tessa, G., Angelini, C., Bielby, J., Bovero, S., Giacoma, C.,
Sotgiu, G., Garner, T.W.J. (2013): e pandemic path-
ogen of amphibians, Batrachochytrium dendrobatidis
(Phylum Chytridiomicota), in Italy. Ital. J. Zool. 80:
1-11.
Wagner, N., Pellet, J., Lotters, S., Schmidt, B.R., Schmitt,
T. (2011): The superpopulation approach for esti-
mating the population size of’prolonged’breeding
amphibians: Examples from Europe. Amphibia-Rep-
tilia 32: 323-332.