Post on 20-Apr-2023
Elsevier Editorial System(tm) for Infection,
Genetics and Evolution
Manuscript Draft
Manuscript Number: MEEGID-D-14-00013R1
Title: High levels of Trypanosoma cruzi DNA detemined by qPCR and
infectiousness to Triatoma infestans support dogs and cats are major
sources of parasites for domestic transmission
Article Type: Research paper
Keywords: Trypanosoma cruzi, Real-time PCR, Infectiousness, Reservoir
Corresponding Author: Dr. Marta Victoria Cardinal, PhD
Corresponding Author's Institution:
First Author: Gustavo F Enriquez
Order of Authors: Gustavo F Enriquez; Jacqueline Búa; Marcela M Orozco;
Sonia Wirth; Alejandro G Schijman; Ricardo E Gürtler; Marta Victoria
Cardinal, PhD
Abstract: The competence of reservoir hosts of vector-borne pathogens is
directly linked to its capacity to infect the vector. Domestic dogs and
cats are major domestic reservoir hosts of Trypanosoma cruzi, and exhibit
a much higher infectiousness to triatomines than seropositive humans. We
quantified the concentration of T. cruzi DNA in the peripheral blood of
naturally-infected dogs and cats (a surrogate of intensity of
parasitemia), and evaluated its association with infectiousness to the
vector in a high-risk area of the Argentinean Chaco. To measure
infectiousness, 44 infected dogs and 15 infected cats were each exposed
to xenodiagnosis with 10-20 uninfected, laboratory-reared Triatoma
infestans that blood-fed to repletion and were later individually
examined for infection by optical microscopy. Parasite DNA concentration
(expressed as equivalent amounts of parasite DNA per mL, Pe/mL) was
estimated by real-time PCR amplification of the nuclear satellite DNA.
Infectiousness increased steeply with parasite DNA concentration both in
dogs and cats. Neither the median parasite load nor the mean
infectiousness differed significantly between dogs (8.1 Pe/mL and 48%)
and cats (9.7 Pe/mL and 44%), respectively. The infectiousness of dogs
was positively and significantly associated with parasite load and an
index of the host's body condition, but not with dog's age, parasite DTU
and exposure to infected bugs in a random-effects multiple logistic
regression model. Real-time PCR was more sensitive and less time-
consuming than xenodiagnosis, and in conjunction with the body condition
index, may be used to identify highly infectious hosts and implement
novel control strategies.
246 words
Dr Nancy Sturm
Editor
Infection, Genetics and Evolution
Buenos Aires, March 31, 2014
Dear Dr Sturm,
We are pleased to resubmit the manuscript entitled “Use of real-time PCR for
quantification of Trypanosoma cruzi DNA in domestic dogs and cats and its association with
infectiousness to Triatoma infestans” co-authored by Gustavo F Enriquez, Jacqueline Bua,
María M. Orozco, Sonia Wirth, Alejandro G. Schijman, Ricardo E. Gürtler, and Marta V.
Cardinal for publication in Infection, Genetics and Evolution.
In the revised version of this manuscript, we have addressed all the concerns the referees raised
on the original manuscript. Below you will find a list detailing modifications done and
answers to referees’ comments. Also, following Reviewer 2’s suggestion we have changed
the title of the manuscript to make it more attractive and now reads: “High levels of
Trypanosoma cruzi DNA determined by qPCR and infectiousness to Triatoma infestans
support dogs and cats are major sources of parasites for domestic transmission”.
The materials in this manuscript have resulted from our original work and have not
been submitted elsewhere. I hereby certify that all authors have read the manuscript and
concur with this submission. None of the material has been published or is under
consideration for publication elsewhere, including the Internet. The authors have no financial
or other relationship that might lead to a conflict of interest regarding the contents of this
manuscript. We look forward to hearing from you at your earliest convenience.
Sincerely,
Marta Victoria Cardinal
CONICET Scientific Investigator
UNIVERSIDAD DE BUENOS AIRES
FACULTAD DE CIENCIAS EXACTAS Y NATURALES
Departamento de Ecología, Genética y Evolución
Laboratorio de Eco-Epidemiología
Ciudad Universitaria. 1428 Buenos Aires, Argentina
Tel-Fax: (+54-11) 4576-3318
Email: mvcardinal@ege.fcen.uba.ar
Cover Letter
MEEGID-D-14-00013
Referees' comments and our responses (in italics)
Reviewer #1: This is manuscript represents a lot of molecular data. Release of molecular data sets is a condition
of MEEGID publications thus this manuscript should be accepted subject to three additional appendice
comprising:
1_All raw qPCR data (Figure 1)
2_All infectiousness data linked with the corresponding qPCR data and the corresponding dog/cat sample (Fig 2
to 3)
3_The GIS locations of the dogs and cats in the study
In response to your request, we provide the information in an appendix and a table: Appendix contains
individual data on dog and cat infectiousness to the vector, qPCR and DTU identification, plus kDNA-PCR
results. GIS locations are provided in a table after this set of responses only for confidential use for reviewers
and editors. We are not sure in which way detailed GIS locations are relevant to our manuscript. If the editor
decides they are relevant, we will not object to add this information to the manuscript.
Appendix. Individual dog and cat infectiousness to the vector, DTU identification, kDNA-PCR and qPCR
results.
Identification
number Host
Infectiousness
(%)
kDNA-
PCR
Bloodstream
parasitic load
(Pe/mL) DTU
15 Dog 61 Positive 18.90 TcVI
16 Dog 68 Positive 50.19 TcVI
17 Dog 5 Positive 0.73 TcVI
18 Dog 61 Positive 10.38 TcVI
19 Dog 72 Positive 1.44 TcVI
21 Dog 60 Positive 8.09 TcVI
21 Dog 0 Positive 8.09 TcVI
23 Dog 0 Positive 20.49 TcI
37 Dog 0 Positive 17.00 TcII / V / VI
38 Dog 32 Positive 4.91 TcVI
39 Dog 5 Positive 1.44 TcVI
63 Dog 89 Positive 28.15 TcVI
71 Dog 50 Positive 12.59 TcVI
77 Dog 75 Positive 50.87 TcVI
81 Dog 75 Positive 2.67 TcV
91 Dog 95 Positive 21.85 TcV
96 Dog 16 Positive 1.38 TcV
98 Dog 0 Negative 2.54 TcII / V / VI
101 Dog 0 Negative 1.06 TcVI
*Reply to Reviewers
102 Dog 85 Positive 3.18 TcVI
103 Dog 5 Negative 1.74 TcVI
104 Dog 16 Positive 3.03 TcV
105 Dog 5 Positive 1.29 TcII / V / VI
106 Dog 65 Positive 20.75 TcVI
108 Dog 95 Positive 214.53 TcVI
112 Dog 15 Positive 1.21 TcVI
114 Dog 80 Positive 34.39 TcVI
117 Dog 85 Positive 39.32 TcVI
118 Dog 35 Positive 8.11 TcVI
119 Dog 0 Negative 1.31 TcII / V / VI
122 Dog 90 Positive 97.36 TcVI
128 Dog 20 Positive 1.13 TcVI
130 Dog 35 Positive 5.85 TcVI
139 Dog 45 Positive 28.83 TcVI
152 Dog 0 Positive 1.94 TcII / V / VI
153 Dog 63 Positive 5.16 TcVI
154 Dog 30 Positive 44.43 TcVI
157 Dog 95 Positive 26.63 TcVI
159 Dog 85 Positive 0.76 TcVI
166 Dog 30 Positive 81.42 TcVI
272 Dog 84 Positive 7.26 TcV
274 Dog 84 Positive 60.79 TcII / V / VI
276 Dog 45 Positive 1.46 TcV
284 Dog 68 Positive 15.21 TcVI
22 Cat 0 Positive 7.42 TcI
25 Cat 67 Positive --a ND
b
26 Cat 95 Positive 1.60 TcVI
62 Cat 95 Positive 96.91 TcVI
78 Cat 79 Positive 202.08 TcVI
79 Cat 0 Negative 0.00 NDb
82 Cat 0 Positive --a ND
b
99 Cat 0 Positive 9.73 TcI
100 Cat 95 Positive 1101.78 TcVI
129 Cat 25 Positive 3.07 TcVI
132 Cat 35 Positive 2.85 TcV
270 Cat 78 Positive 114.41 TcV
278 Cat 50 Positive 2.17 TcVI
781 Cat 75 Positive 20.03 TcVI
797 Cat 63 Positive 20.06 TcVI
a. DNA quantifications could not be normalized. b. ND: Not determined. DTU identifications were
unsuccessful.
This is an interesting and relevant study, which has considerable application to Chagas disease epidemiology. It
continues Prof Gurtler's in examining the important epidemiological role of dogs in T. cruzi transmission. The
TcI association in both cats and dogs is interesting but I agree with the authors there isn't a big enough sample
size (total of 3) to make any conclusions and as the authors point out could be a consequence of the vector.
However, the association between cats and DTU specific infectiousness looks to be an error of designating
sample groups and must be removed throughout the manuscript: Figure 3 (mistakenly labelled 'Figure 2' in the
manuscript) shows 2 TcV's with approx. 80% infectiousness, whilst 8 TcVI infected cats shows 25 - 100%
infectiousness. There are not enough TcV infected cats to conclude, TcVI is associated within higher
infectiousness against TcV. The authors grouped TcI with TcV and compared this against TcVI, genetically the
authors should have grouped TcV with TcVI and compared it to TcI, where up on they will have found there
isn't a large enough sample size to draw a conclusion. Thus the authors are left saying there is a strong DTU
association with cats for TcVI but not TcV and not dogs. This is either remarkable or else a result of group
designation.
316 The infectiousness of infected cats was significantly associated with ...
317 ........ DTU (OR = 15.04, 95% CI = 1.58-142.52,
318 P < 0.01) (Wald <chi>2 = 10.44, P = 0.02) (Table 1).
Thank you for these insightful comments on our data. Following your suggestions, in the revised version we
have deleted the comparison of infectiousness and identified DTUs for cats due to the small sample size of
examined cats.
Regarding dogs, we have originally designated DTU groups bearing in mind that TcVI is frequently associated
with dogs and cats; and TCV with human infections in the domestic transmission cycle of T. cruzi in the
Argentine Chaco (Diosque et al., 2003; Cardinal et al., 2008). Therefore, we examined if the infectiousness to
the vector differed between these DTUs. Given the small number of TcV-infected animals, in the previous
version we decided to group them with TcI- or TcIII-infected animals. Following your suggestions, in the
revised version we have added a new category level within DTUs and repeated the analysis for dogs.
Revised text in M&M (lines 277-285):
“The independent variables were the bloodstream parasite load (a continuous variable, in Pe/mL), body
condition (two levels: 0 = good, 1 = regular or poor); age of the host (a continuous variable, in months);
exposure to infected bugs, as measured by the abundance of T. infestans infected with T. cruzi captured per 15
min-person per site in domiciles, kitchens and storerooms (i.e., typical resting sites of dogs and cats),
categorized in low, medium and high-risk levels, 0 = 0; 1 = 1-5, and 2 ≥ 5 bugs, respectively; and identified
DTU (0 = TcV, 1 = TcVI; 2 = TcI or TcIII). Identified DTU was excluded as a predictor in cats due to the small
sample size.”
>The authors should discuss whether dog quality is a confounding variable and clarify what a good and average
conditioned dog is please.
A more detailed description of body condition has been included in the revised version.
Revised text in M&M (lines136-140):
“The host’s body condition was used as an index of the nutritional status of dogs as described by Petersen et al.
(2001). Each dog was classified as having a good, regular or poor body condition based on the degree of
muscle development; external evidence of bone structure, state of fur coat, existence of fatty deposits, and facial
expression (Petersen et al., 2001).”
In my opinion it is likely to be a confounding effect of quality of dwelling. In other words, a homeowner
unwilling to keep their pets in good condition is unlikely to keep their home or peridomestic environments in
good condition which could result in higher vector burdens. Thus this could be a confounding variable.
Comparisons against Leishmania are not valid because T. cruzi is asymptomatic, unlike Leishmania infected
dogs, and what the authors are claiming is the dogs overall condition is linked to its immunological status which
is linked to its ability to control T. cruzi parasitaemia. The parsimonious explanation is a confounding variable
of either peridomestic environment or else a superinfection, such as worms. The authors need a better
discussion on the matter and it will help decide hypotheses for future research. There are obvious means of
spatial analysis to assess whether dog parasitaemia is linked with dog genetics and again the authors must
present this raw data.
These interesting points deserve further explanations. First, in most rural areas of the Argentine Chaco, dogs
are not considered pets or companion animals by their owners because they play important roles in the family
subsistence economy. The main functions usually recognized are goat/cow herding, hunting or guardian. Only
3% of 111 dogs included in this study were considered pets by their owners.
Second, a poor quality of dwelling (e.g., walls in poorer conditions) and poor dog body condition are the
common denominator of rural poverty. The former is associated with higher domestic abundance of T. infestans
(Gurevitz et al., 2011) and greater risk of dog infection (Cardinal et al. in press, Am J Trop Med Hyg). If
infectious dogs were present in these households, the abundance of infected bugs would increase and dogs or
other hosts would face higher chances of (re)infection. Some experimental data support that exposure to
reinfection may increase parasitaemia. Therefore, houses with higher infected-bug abundance would entail a
higher risk of (re)infection and perhaps, enhanced infectiousness. To account for household clustering of living
conditions we have included the household as a new random effect in our regression analysis for dogs, while
keeping the individual host as another random effect nested within the household. However, given that these
results were qualitatively equivalent to the previous ones, with very minor changes in parameter estimates, we
decided not to include this new random effect in order to keep the model as parsimonious as possible.
Previous evidence showed that dog body condition (indexed by ECA) was closely associated with its nutritional
status (especially the hematocrit) and infectivity to the vector (Petersen et al., 2001). Therefore, we
hypothesized that body condition impacted the dogs’ immune system and the host’s ability to control
parasitaemia, in which case the index of body condition is not simply a confounder. A confounder effect would
imply a spurious relation between ECA and infectiousness to the vector. Nutritional deficiencies have also been
associated with a decreased immune response and increased parasitaemia in experimental conditions
(Carlomagno et al., 1987; Keusch, 2003; Machado-Coelho et al., 2005). A competent immune system is needed
to control T. cruzi parasitaemia, and whenever the immune response is suppressed reactivation of Chagas
disease and high parasitaemia are recorded (Burgos et al., 2005; Martin et al., 2004; Sartori et al., 2007).
Superinfection with helminths and other parasites are most likely in poor rural settings as in our study area,
and could affect the nutritional status of the dogs, its body condition and immune response.
Typos have been designated on the document.
Corrections were made where needed.
> Reviewer #2: Dogs and cats are known to constitute important sources of parasites for domestic transmission
of T. cruzi. This study confirmed that infectiousness of dogs is associated with the bloodstream parasitic load
using a qPCR method. This work simply described the application of a previously standardized qPCR method
for quantification of T. cruzi in dogs and cats. The study was well-conducted, results are sound and clearly
presented. However, the Introduction and Discussion failed to clearly show the relevance and current
knowledge of this issue, and have several questionable points. The manuscript, an extension of successive
articles recently published by the authors about T. cruzi infection in dogs and cats, can be much improved
before accepted for publication by the IGE.
>Please, find bellow some comments that, maybe, can be useful to clarify and improve the manuscript,
especially for readers non-familiarized with this specific topic.
Thank you for your helpful comments.
> 1. The goal of the study is to identify "superspreader" hosts. The concept of "superspreader" was not clearly
defined, apparently, for the authors only animals showing high parasite loads are "superspreader"(amplification
hosts?), despite they are domestic animals of restricted "spread". In the literature, 'superspreading' hosts play a
key role in disease epidemics.
In the revised version we have now included a definition of “superspreader”. (Lines 90-92)
The study dogs do not usually spend the night in another house, but they were reported to visit frequently
nearby forest patches either alone or in groups; as they are not restrained they may move freely. Additionally,
dogs imported from infested villages without vector control were at a higher relative odds of infection than
other dogs native to villages under sustained vector surveillance and control (Cardinal et al., 2007), showing
that infected dogs could “spread” T. cruzi infection outside their places of birth.
Revised text in the Introduction (Lines 87-94):
“There is strong evidence on the clustering and heterogeneity in the transmission of pathogens, where 20% of
individuals in a population would be responsible for 80% of the transmission (Wilson et al., 2001; Woolhouse
et al., 1997; Courtenay et al., 2014). The small fraction of individuals that infect disproportionately more
susceptible contacts than most infected individuals became known as “superspreaders” of disease (Stein 2011).
Trypanosoma cruzi infection, although mainly a vector-borne disease, shows that a small fraction of the
infected dogs and cats were highly infectious to xenodiagnosis bugs and could be considered “superspreaders”
(Gürtler et al., 2007).”
In lines 103-107:
“In this study we used for the first time a qPCR protocol to quantify the load of T. cruzi DNA in the peripheral
blood of naturally-infected dogs and cats from a well-defined rural area in the Argentinean Chaco. We also
evaluated the association among infectiousness to the vector, bloodstream parasite load and other potential
predictors of infectiousness, and identified highly infectious hosts.”
> 2. Although domestic dogs and cats are major domestic reservoir hosts of T. cruzi in many places, studies on
these animals have been concentrated in Argentina; many articles are from the authors of this manuscript. The
concentration of studies in this country (and also in the literature cited in this manuscript) does not implicate
that data from other countries and authors are irrelevant. In fact, data of dogs infected with T. cruzi and
implication to human infection from other countries add relevance and broader interest to this manuscript.
Right. We have now added three references to further illustrate the importance of dogs in the transmission of T.
cruzi throughout the Americas (Lines 51-53).
“Domestic dogs and cats are major reservoir hosts of T. cruzi in the domestic environment throughout the
Americas (Cardinal et al., 2008, in press; Crisante et al., 2006; Gürtler et al., 2007; Noireau et al., 2009;
Ramírez et al., 2013).”
> 3. Introduction: "Therefore, dogs and cats frequently constitute the largest sources of parasites for domestic
transmission of T. cruzi. Everywhere? References?
The revised text (Lines 56-59):
“Therefore, dogs and cats frequently constitute the largest source of T. cruzi for domestic transmission where
Triatoma infestans is the main vector species as in the Gran Chaco region (Gürtler et al., 2007).”
> 4. Introduction: The minimum number of parasites required to infect a bug via xenodiagnosis and the
percentage of infected insects depend on T. cruzi strains (DTUs?) and triatomine species, and not simply to the
number of trypomastigotes ingested or inoculated.
This sentence has been re-written (Lines 65-69):
“However, the percentage of infected insects could be highly variable and depends on T. cruzi DTU and strain,
triatomine species and not simply on the number of trypomastigotes ingested or inoculated (Alvarenga and
Leite, 1982; Garcia et al., 2007; de Lana et al., 1998; Neal and Miles, 1977; da Silveira Pinto et al., 2000).”
> 5. Results. "No clear pattern was observed between DTU and host infectiousness or bloodstream parasitic
load. However, the two cats and the only dog infected with TcI had nil infectiousness despite having 7.4, 9.7
and 20.5 Pe/mL, respectively (Figure 3). Sequence analysis of SL-IR amplicons from identified TcI showed that
they were more closely related to TcI than to other DTUs (i.e., 98-100% sequence identity)."
We have rewritten the last sentence (Lines 325-330):
“No clear association was observed between DTU and host infectiousness or bloodstream parasite load
(Figure 3). However, the two cats and the only dog infected with TcI had nil infectiousness despite having 7.4,
9.7 and 20.5 Pe/mL, respectively (Figure 3) (Appendix). In these three individuals TcI was confirmed by
sequence analysis of Spliced Leader region (SL-IR I) (i.e., 98-100% sequence identity).”
"Indeed, bug infection in xenodiagnosis-negative, TcI-infected dogs and cats was rare and only revealed by
means of a highly sensitive kDNA-PCR applied to the bugs rectal-ampoule lysates. The substantial variations in
metacyclogenesis and trypomastigote numbers among triatomine species (Carvalho-Moreira et al., 2003; Loza-
Murghía et al., 2010; Perlowagora-Szumlewicz et al., 1994) may explain the nil infectiousness (as measured by
optical microscopy??) recorded in TcI-infected dogs and cats".
This sentence has been re-written: (Lines 452-458):
“Indeed, TcI-infected dogs and cats were rare and could only be identified by means of a highly sensitive
kDNA-PCR applied to the rectal-ampoule lysates of xenodiagnostic bugs or GEB samples. The substantial
variations in metacyclogenesis and trypomastigote density among triatomine species (Carvalho-Moreira et al.,
2003; Loza-Murghía et al., 2010; Perlowagora-Szumlewicz et al., 1994) may explain the nil infectiousness (as
measured by optical microscopy) recorded in TcI-infected dogs and cats. However, samples sizes were small
and this relationship should be further investigated.”
> The authors assessed the DTU of T. cruzi infecting all infected dogs and cats. However, results are not clear
and deserve to be better discussed. What means "SL-IR amplicons from identified TcI showed that they were
more closely related to TcI?
Given that two unspecific bands were persistently obtained in the SL-IR-PCR from GEB samples, the expected
amplicon for T. cruzi DNA of 475 pb was purified and sequenced to confirm TcI identification.
Revised text in (Lines 325-330)
“No clear association was observed between DTU and host infectiousness or bloodstream parasite load
(Figure 3). However, the two cats and the only dog infected with TcI had nil infectiousness despite having 7.4,
9.7 and 20.5 Pe/mL, respectively (Figure 3) (Appendix). In these three individuals TcI was confirmed by
sequence analysis of Spliced Leader region (SL-IR I) (i.e., 98-100% sequence identity).”
In a previous article, the authors (Enriquez et al., 2013a: Discrete typing units of Trypanosoma cruzi identified
in rural dogs and cats in the humid Argentinean Chaco. Parasitol 12, 1-6) were unable to detect TcI in dogs
from the areas investigated in this work (the same dogs and cats?).
These dogs and cats belong to the same communities and samples were taken at the same time as those included
in Enriquez et al. (2013a).
In Enriquez et al. (2013a) all DTU identifications were made from DNA samples from cultured parasites.
However, 7 dogs and 5 cats examined were not included because the infecting parasites could not be isolated
(i.e., all were negative by xenodiagnosis or in vitro cultures were unsuccessful). In the current work, due to the
need to identify the DTUs or at least achieve a partial identification for normalizing the quantification of
parasites, we attempted to identify the DTUs from other sources as guanidine blood samples or from the rectal
ampoule.
This sentence has been re-written (Lines 229-236):
“Identification of DTUs infecting dogs and cats based on isolated parasites was reported elsewhere (Enriquez
et al., 2013a). In the current study, additional DTU identifications from seropositive, xenodiagnosis-negative
dogs (n = 6) and cats (n = 5) and from a seropositive, xenodiagnosis-positive dog were attempted directly from
GEB or rectal ampoule samples of OM-negative, kDNA-PCR-positive bugs used in xenodiagnostic tests of these
animals. These additional dogs and cats belonged to the same villages, and samples were taken at the same
time as those included in Enriquez et al. (2013a).”
How to explain the lack of detection of TcI infection in dogs (by hemoculture and genotyping) in
countries/regions where predominates TcI? In general, TcI exhibit high metacyclogenesis in culture and
triatomines bugs; hemocultures are much more easer to be obtained for TcI than for the other DTUs. It was
recently demonstrated that wild T. infestans is naturally infected with TcI in Bolivia. Substantial differences in
parasitic load between dogs/cats and humans could be associated to distinct and preferential DTUs infecting
these hosts?
TcVI and TcV are more frequently found in the domestic cycle of transmission than TcI in Argentina, unlike the
situation in the Amazon basin, Colombia, Venezuela and Mexico, where TcI has also been identified in dogs
(Ramirez et al., 2013). Also, the lack of detection of TcI in dogs by xenodiagnosis could be due to different
susceptibility of T. infestans or to the specific genotypes of TcI involved (Cura et al., 2010; Herrera et al.,
2009). In our study area, TcI was not found in T. infestans and was only detected in few specimens of
peridomestic Triatoma sordida collected along many years (Maffey et al., 2012). However, we found a high
prevalence of TcI infection (by xenodiagnosis and kDNA-PCR) in Didelphis albiventris opossums, probably
indicating this DTU is predominantly associated with the local sylvatic transmission cycle.
> 6. Discussion: "Quantitative PCR was more sensitive and less time-consuming than xenodiagnosis, and alone
or in conjunction with the body condition index, may be helpful to identify "superspreader" hosts, to replace
xenodiagnosis (or hemoculture) ...."
>In a recent article from the same authors (Enriquez et al., 2013b: Detection of Trypanosoma cruzi infection in
naturally infected dogs and cats using serological, parasitological and molecular methods, ActaTrop126:211-7)
they concluded: "The high sensitivity of kDNA-PCR to detect T. cruzi infections in naturally infected dogs and
cats supports its application as a diagnostic tool complementary to serology and may replace the use of
xenodiagnosis or hemoculture". It will be interesting to see a comparative analysis between kDNA-PCR (very
easy to be done dispensing equipment still sophisticate for most CH endemic countries) and qPCR to detect T.
cruzi infections regardless the parasite load, i.e., in both "superspreader" and "non-superspreader" hosts.
Following your suggestion, in the revised version we have now included a comparative analysis between kDNA-
PCR and qPCR to detect T. cruzi infection. However, in agreement with your suggestions about the role that
superspreader hosts play in disease, we believe that categorizing dogs and cats into “high or low
infectiousness” would be more appropriate than strictly qualifying them as superspreaders.
We re-wrote the following section and incorporated a paragraph in the discussion
Revised text in M&M (Lines 159-164):
“Trypanosoma cruzi-infected dogs and cats were also classified as having high or low infectiousness to T.
infestans; the “high infectiousness” category included dogs and cats that were responsible for ≥80% of all
infected triatomines used in the xenodiagnostic tests, and the “low infectiousness” included the remaining
seropositive dogs and cats and that were also examined by xenodiagnoses.”
Results (Lines 359-370):
“3.4. Profile of infected dogs and cats with high or low infectiousness”
“Nearly half of the dogs (50%) and cats (47%) showed high infectiousness and were responsible of at least
80% of the infected xenodiagnosis bugs. The median parasite load was nearly 10 times greater for dogs with
high infectiousness than for dogs with low infectiousness and this difference was statistically significant
(Kruskal-Wallis test, P < 0.01). For cats this relationship was 32 times greater (Kruskal-Wallis test, P = 0.02)
(Table 3). Quantitative PCR could not distinguish between highly infectious and non-highly infectious hosts,
given that it detected T. cruzi DNA in 100% of the seropositive dogs and in 93% of the seropositive cats,
whereas kDNA-PCR detected 89% and 87% of them, respectively (Table 3). Four seropositive dogs and one
seropositive cat that were both qPCR-positive and kDNA-PCR- negative showed very low infectiousness to the
vector (range, 0-5%).”
Table 3. Relationship among qualitative kDNA-PCR, qPCR and parasite load to detect T. cruzi-infected dogs
and cats with different levels of infectiousness
Host Infectiousnessa
No. positive
xenodiagnosis (No.
examined)
Mean
infectiousness
(%)
Median parasite
load (Q1-Q3)
No.
Positive
kDNA (%)
No.
Positive
qPCR (%)
Dogs High 22 (22) 79 21.3 (7.3-39.3) 22 (100) 22 (100)
Low 16 (22) 19 2.4 (1.3-12.6) 17 (77) 22 (100)
Total 38 (44) 48 8.1 (1.5-26.6) 39 (89) 44 (100)
Cats High 7 (7) 79 96.1(18.1-202.1) 7 (100) 7 (100)
Low 3 (8) 14 3.0 (2.2-7.4) 6 (75) 7 (88)
Total 10 (15) 44 9.7 (2.9-96.9) 13 (87) 14 (93)
a. The “high infectiousness” category included dogs and cats that were responsible for ≥80% of all infected
triatomines used in the xenodiagnostic tests. The “low infectiousness” category included all remaining
seropositive dogs and cats examined by xenodiagnosis.
Discussion (Lines 459-468)
“In our study, approximately 50% of the infected dogs and cats were responsible for at least 80% of the
infected T. infestans used in xenodiagnosis. All of them were positive by qPCR and kDNA-PCR. However, dogs
and cats with very low infectiousness to the vector (<20%) were also frequently positive by these techniques.
These results suggest that the quantitative estimates of parasite load and not the qualitative outcomes of qPCR
or kDNA-PCR would be a more precise tool to identify highly infectious dogs and cats. A wide variability in
infectiousness (range, 0-63% for dogs and 35-67% for cats) was observed between 1-5 Pe/mL. Courtenay et al.
(2014) recently showed that quantitative PCR was needed to identify highly infectious L. infantum-infected
dogs.”
> 7. Discussion: "There was a small fraction of highly infectious hosts (i.e., "superspreaders") that may
contribute disproportionally to (peri)domestic transmission and pathogen persistence" Is this comment about
hosts of T. cruzi?? The wild mammals are never "superspreaders"? This statement must be explained and
reference must be cited.
In the revised version we have included additional explanations (Lines 415-423):
“Dogs and cats had similar infectiousness to the vector and this distribution was aggregated, as in studies
conducted elsewhere (Gürtler et al., 2007). There was a fraction of highly infectious dogs and cats that may
contribute disproportionally to (peri)domestic transmission and persistence of T. cruzi. This heterogeneous
distribution of parasite burden in host populations is a generalized pattern in macroparasites and
microparasites (Wilson et al., 2001; Woolhouse et al., 1997) and may be used for targeted control strategies.
Trypanosoma cruzi-infected Dasypus novemcinctus armadillos and Didelphis albiventris opossums also have
high infectiousness (Orozco et al., 2013).”
In lines 469-474:
“Quantitative PCR was more sensitive and less time-consuming than xenodiagnosis, and alone or in
conjunction with the body condition index, may be helpful to identify highly infectious hosts; replace
xenodiagnosis (or hemoculture) to assess treatment efficacy or infection status, and contribute to novel control
strategies (e.g., targeted treatment or vaccination) that reduce the risks of domestic transmission of T. cruzi.”
>Finally, the title can be more atractive. Suggestion High levels of Trypanosoma cruzi DNA detemined by
qPCR in dogs and cats and infectiousness to Triatoma infestans support these hosts as major sources of
parasites for domestic transmission of Chagas disease
Thank you for the suggestion! Just to make it shorter we suggest:
High levels of Trypanosoma cruzi DNA determined by qPCR and infectiousness to Triatoma infestans support
dogs and cats are major sources of parasites for domestic transmission
We added a new result to assess the OM false negative rate in dogs.
Revised text in M&M (Lines 193-195):
“The rectal-ampoule samples of 94 OM-negative bugs used in xenodiagnosis of the seropositive dogs was
examined by kDNA-PCR to assess the false-negative rate of OM procedures.”
In Results, lines 319-320
“Thirteen percent of 94 OM-negative bugs were positive by kDNA-PCR; thus, the corrected infectiousness to
the vector would rise to 54%.”
Table. Individual dog and cat infectiousness to the vector, DTU identification, house location, kDNA-PCR and
qPCR results.
Identification
number Host
Infectiousness
(%)
kDNA-
PCR
Bloodstream
parasite load
(Pe/mL) DTU
GIS
locations
(X)
GIS
locations
(Y)
15 Dog 61 Positive 18.9 TcVI 192273 7141605
16 Dog 68 Positive 50.19 TcVI 192273 7141605
17 Dog 5 Positive 0.73 TcVI 192273 7141605
18 Dog 61 Positive 10.38 TcVI 192273 7141605
19 Dog 72 Positive 1.44 TcVI 192273 7141605
21 Dog 60 Positive 8.09 TcVI 192509 7142642
21 Dog 0 Positive 8.09 TcVI 192509 7142642
23 Dog 0 Positive 20.49 TcI 192509 7142642
37 Dog 0 Positive 17.00 TcII / V / VI 192483 7142945
38 Dog 32 Positive 4.91 TcVI 192592 7142622
39 Dog 5 Positive 1.44 TcVI 192519 7141807
63 Dog 89 Positive 28.15 TcVI 192519 7141807
71 Dog 50 Positive 12.59 TcVI 192519 7141807
77 Dog 75 Positive 50.87 TcVI 192519 7141807
81 Dog 75 Positive 2.67 TcV 193639 7140089
91 Dog 95 Positive 21.85 TcV 193684 7140046
96 Dog 16 Positive 1.38 TcV 191530 7141117
98 Dog 0 Negative 2.54 TcII / V / VI 191530 7141117
101 Dog 0 Negative 1.06 TcVI 191530 7141117
102 Dog 85 Positive 3.18 TcVI 191530 7141117
103 Dog 5 Negative 1.74 TcVI 191525 7140815
104 Dog 16 Positive 3.03 TcV 191525 7140815
105 Dog 5 Positive 1.29 TcII / V / VI 192124 7139869
106 Dog 65 Positive 20.75 TcVI 192124 7139869
108 Dog 95 Positive 214.53 TcVI 192124 7139869
112 Dog 15 Positive 1.21 TcVI 192124 7139869
114 Dog 80 Positive 34.39 TcVI 191406 7140272
117 Dog 85 Positive 39.32 TcVI 191406 7140272
118 Dog 35 Positive 8.11 TcVI 191406 7140272
119 Dog 0 Negative 1.31 TcII / V / VI 191406 7140272
122 Dog 90 Positive 97.36 TcVI 192964 7138634
128 Dog 20 Positive 1.13 TcVI 192964 7138634
130 Dog 35 Positive 5.85 TcVI 192964 7138634
139 Dog 45 Positive 28.83 TcVI 192964 7138634
152 Dog 0 Positive 1.94 TcII / V / VI 190995 7140873
153 Dog 63 Positive 5.16 TcVI 190995 7140873
154 Dog 30 Positive 44.43 TcVI 191703 7140101
157 Dog 95 Positive 26.63 TcVI 191703 7140101
159 Dog 85 Positive 0.76 TcVI 191703 7140101
166 Dog 30 Positive 81.42 TcVI 191703 7140101
272 Dog 84 Positive 7.26 TcV 190995 7140873
274 Dog 84 Positive 60.79 TcII / V / VI 190250 7140985
276 Dog 45 Positive 1.46 TcV 190250 7140985
284 Dog 68 Positive 15.21 TcVI 190250 7140985
22 Cat 0 Positive 7.42 TcI 190407 7141226
25 Cat 67 Positive NDa ND
b 191149 7139883
26 Cat 95 Positive 1.6 TcVI 191149 7139883
62 Cat 95 Positive 96.91 TcVI 191149 7139883
78 Cat 79 Positive 202.08 TcVI 191202 7139841
79 Cat 0 Negative 0 NDb 190457 7139884
82 Cat 0 Positive NDa ND
b 190498 7139473
99 Cat 0 Positive 9.73 TcI 191776 7133939
100 Cat 95 Positive 1101.78 TcVI 191776 7133939
129 Cat 25 Positive 3.07 TcVI 191776 7133939
132 Cat 35 Positive 2.85 TcV 191259 7134302
270 Cat 78 Positive 114.41 TcV 191259 7134302
278 Cat 50 Positive 2.17 TcVI 191705 7134296
781 Cat 75 Positive 20.03 TcVI 213943 7091603
797 Cat 63 Positive 20.06 TcVI 193608 7086903
a. DNA quantifications could not be normalized. b. ND: No determined. DTU identifications were
unsuccessful.
Highlights
1. We measured the parasitic load of dogs and cats infected with T. cruzi by qPCR.
2. In both hosts, parasitic load was similar and higher than in chronic patients.
3. Infectiousness to the vector was associated with parasitic load in both hosts.
4. qPCR may be used to identify “superspreader” hosts and to assess treatment
efficacy or infection status.
*Highlights (for review)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
High levels of Trypanosoma cruzi DNA detemined by qPCR and infectiousness to 1
Triatoma infestans support dogs and cats are major sources of parasites for domestic 2
transmission 3
Enriquez G.F.,1,2
Bua J.,3
Orozco M.M.,1,2
Wirth S.,4
Schijman A.G.,5
Gürtler R.E.,1,2
4
Cardinal M.V. 1,2
5
6
1Laboratory of Eco-Epidemiology, Faculty of Exact and Natural Sciences, University of 7
Buenos Aires, Argentina. 8
2Institute of Ecology, Genetics and Evolution of Buenos Aires (UBA-CONICET). 9
3National Institute of Parasitology Dr M. Fatala Chaben, National Administration of 10
Laboratories and Institutes of Health Dr. C.G. Malbrán, Buenos Aires, Argentina. 11
4Laboratory of Agro-biotechnology, Faculty of Exact and Natural Sciences, University 12
of Buenos Aires, Argentina. 13
5Laboratory of Molecular Biology of Chagas Disease, Institute for Research on Genetic 14
Engineering and Molecular Biology (INGEBI-CONICET).
15
16
Running title: Trypanosoma cruzi quantification and infectiousness in dogs and cats 17
18
Corresponding author: M. Victoria Cardinal, Laboratory of Eco-Epidemiology, Faculty 19
of Exact and Natural Sciences, University of Buenos Aires, Argentina, 1428 Buenos 20
Aires, Argentina. Tel/Fax: +54-11-4576-3318. mvcardinal@ege.fcen.uba.ar 21
22
*ManuscriptClick here to view linked References
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Abstract
23
The competence of reservoir hosts of vector-borne pathogens is directly linked to its 24
capacity to infect the vector. Domestic dogs and cats are major domestic reservoir hosts 25
of Trypanosoma cruzi, and exhibit a much higher infectiousness to triatomines than 26
seropositive humans. We quantified the concentration of T. cruzi DNA in the peripheral 27
blood of naturally-infected dogs and cats (a surrogate of intensity of parasitemia), and 28
evaluated its association with infectiousness to the vector in a high-risk area of the 29
Argentinean Chaco. To measure infectiousness, 44 infected dogs and 15 infected cats 30
were each exposed to xenodiagnosis with 10-20 uninfected, laboratory-reared Triatoma 31
infestans that blood-fed to repletion and were later individually examined for infection 32
by optical microscopy. Parasite DNA concentration (expressed as equivalent amounts of 33
parasite DNA per mL, Pe/mL) was estimated by real-time PCR amplification of the 34
nuclear satellite DNA. Infectiousness increased steeply with parasite DNA 35
concentration both in dogs and cats. Neither the median parasite load nor the mean 36
infectiousness differed significantly between dogs (8.1 Pe/mL and 48%) and cats (9.7 37
Pe/mL and 44%), respectively. The infectiousness of dogs was positively and 38
significantly associated with parasite load and an index of the host‟s body condition, but 39
not with dog‟s age, parasite DTU and exposure to infected bugs in a random-effects 40
multiple logistic regression model. Real-time PCR was more sensitive and less time-41
consuming than xenodiagnosis, and in conjunction with the body condition index, may 42
be used to identify highly infectious hosts and implement novel control strategies. 43
246 words 44
Keywords: Trypanosoma cruzi, Real-time PCR, Infectiousness, Reservoir 45
46
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
1. Introduction 47
Trypanosoma cruzi, the etiologic agent of Chagas disease, has been found 48
infecting approximately 180 species of mammals (WHO, 2002). The competence of 49
reservoir hosts of vector-borne pathogens is directly linked to its capacity to infect the 50
vector. Domestic dogs and cats are major reservoir hosts of T. cruzi in the domestic 51
environment throughout the Americas (Cardinal et al., 2008, in press; Crisante et al., 52
2006; Gürtler et al., 2007; Noireau et al., 2009; Ramírez et al., 2013). They are usually 53
highly infectious to triatomine bugs, and display much higher infectiousness (defined as 54
the proportion of uninfected insects that become infected after blood-feeding once on an 55
infected host) than T. cruzi-seropositive humans (Gürtler et al., 2007; 1996). Therefore, 56
dogs and cats frequently constitute the largest source of T. cruzi for domestic 57
transmission where Triatoma infestans is the main vector species as in the Gran Chaco 58
region (Gürtler et al., 2007). 59
The chances of a triatomine bug becoming infected with T. cruzi after blood-60
feeding on an infected host mainly depend on the probability of ingesting the parasite 61
(determined by the intensity of parasitemia and bloodmeal size), and its successful 62
establishment and reproduction in the insect gut. The minimum number of T. cruzi 63
parasites required to infect a bug through artificial xenodiagnosis was estimated to be 64
one (Moll-Merks et al., 1987). However, the percentage of infected insects could be 65
highly variable and depends on T. cruzi DTU and strain, triatomine species and not 66
simply on the number of trypomastigotes ingested or inoculated (Alvarenga and Leite, 67
1982; Garcia et al., 2007; de Lana et al., 1998; Neal and Miles, 1977; da Silveira Pinto 68
et al., 2000). 69
Previous studies that sought to estimate parasitemia in naturally-infected hosts 70
have traditionally relied on direct counting of T. cruzi parasites under optical 71
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
microscopy. However, new technologies have become available for this purpose (i.e., 72
flow citometry, DNA amplification). Real-time or quantitative PCR (qPCR) is useful 73
for the quantification of parasite DNA in different host tissues and its applications have 74
substantially increased in recent years. In Chagas disease, qPCR has been mainly 75
applied to human patients and has been successfully used in the follow-up of treatment 76
outcomes, diagnosis of infection, and to investigate parasite vertical transmission (Bua 77
et al., 2012; Duffy et al., 2009; Moreira et al., 2013; Schijman et al., 2011). Recent 78
studies using qPCR to measure the concentration of T. cruzi DNA (i.e., a surrogate of 79
intensity of parasitemia) found that the median bloodstream parasite load of chronic 80
patients was ≤ 2 parasite equivalents/mL and differed significantly among countries 81
(Freitas et al., 2011; Moreira et al., 2013). Higher parasite loads were recorded in 82
chronic patients from Argentina and Colombia (Duffy et al., 2013; Moreira et al., 2013), 83
in patients coinfected with HIV (Freitas et al., 2011), in seropositive women giving 84
birth to T. cruzi-infected newborns (Bua et al., 2012) and in newborns diagnosed by 85
optical microscopy after delivery (Bua et al., 2013). 86
There is strong evidence on the clustering and heterogeneity in the transmission of 87
pathogens, where 20% of individuals in a population would be responsible for 80% of 88
the transmission (Wilson et al., 2001; Woolhouse et al., 1997; Courtenay et al., 2014). 89
The small fraction of individuals that infect disproportionately more susceptible 90
contacts than most infected individuals became known as “superspreaders” of disease 91
(Stein 2011). Trypanosoma cruzi infection, although mainly a vector-borne disease, 92
shows that a small fraction of the infected dogs and cats were highly infectious to 93
xenodiagnosis bugs and could be considered “superspreaders” (Gürtler et al., 2007).The 94
infectiousness to the vector T. infestans of seropositive dogs was shown to be 95
aggregated (Gürtler et al., 2007) and associated with an index of the host‟s body 96
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
condition (i.e., external clinical aspect, ECA) (Petersen et al., 2001), and not associated 97
with the Discrete Typing Unit (DTU, i.e., genotypes) of T. cruzi (Cardinal et al., 2008). 98
Mixed outcomes were registered for age of the dog and exposure to infected bugs 99
(Gürtler et al., 2007; 1996; 1992). The functional relationship between parasitemia and 100
infectiousness to the vector in dogs and cats naturally infected with T. cruzi (or any 101
other mammalian nonhuman host) has not been investigated. 102
In this study we used for the first time a qPCR protocol to quantify the load of T. 103
cruzi DNA in the peripheral blood of naturally-infected dogs and cats from a well-104
defined rural area in the Argentinean Chaco. We also evaluated the association among 105
infectiousness to the vector, bloodstream parasite load and other potential predictors of 106
infectiousness, and identified highly infectious hosts. 107
108
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
2. Materials and Methods 109
2.1. Study area 110
Field work was conducted in the municipality of Pampa del Indio (26° 2′ 0″ S, 111
59° 55′ 0″ O), Chaco Province, Argentina. The study area was described elsewhere 112
(Gurevitz et al., 2011). Prior to community-wide residual spraying with pyrethroid 113
insecticides, a cross-sectional survey of all villages revealed that 45.9% of the inhabited 114
houses were infested with T. infestans (Gurevitz et al., 2011), and the overall prevalence 115
of T. cruzi infection was 27% in bugs, 26% in dogs and 29% in cats in August–116
December 2008 (Cardinal et al., in press). 117
118
2.2. Study design 119
Nested within this larger survey, for current purposes we selected two 120
neighboring villages (10 de Mayo and Las Chuñas, with 60 inhabited houses) showing 121
the highest bug prevalence of infection with T. cruzi (61%) at baseline. Within these 122
villages, all the houses with at least one T. cruzi-infected T. infestans at baseline (38 of 123
60 houses) were selected to conduct a cross-sectional xenodiagnostic survey of all dogs 124
and cats. A total of 99 (56.3%) dogs and 29 (74.4%) cats were examined by 125
xenodiagnosis to increase the chances of isolating parasites from a large number of 126
individuals. Each head of household was informed on the objectives of the study; an 127
informed oral consent was requested and obtained in all cases. 128
Blood samples (up to 7 mL) were drawn by venipuncture and allowed to clot at 129
ambient temperature. Each serum was separated after centrifugation at 3,000 rpm during 130
15 min., allocated in triplicate vials and preserved at -20 ºC at the field laboratory. 131
Blood samples were also used (2 mL in dogs and 1 mL in cats) for DNA extraction and 132
PCR; these were immediately mixed with an equal volume of guanidine hydrochloride 133
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
6 M, EDTA 0.2 M pH 8.00 buffer (GEB), and stored at 4ºC. Serum samples were 134
transported in dry ice to the main laboratory at the end of each survey, whereas GEB 135
samples were kept in a cooler. The host‟s body condition was used as an index of the 136
nutritional status of dogs as described by Petersen et al. (2001). Each dog was classified 137
as having a good, regular or poor body condition based on the degree of muscle 138
development; external evidence of bone structure, state of fur coat, existence of fatty 139
deposits, and facial expression (Petersen et al., 2001). Only dogs aged 1 year or more 140
were categorized to avoid potential confounders due to growth effects and acute 141
infections (Petersen et al., 2001). Handling and examination of dogs and cats was 142
conducted according to the protocol approved by the „Dr. Carlos Barclay‟ Independent 143
Ethical Committee for Clinical Research from Buenos Aires, Argentina (IRB No. 144
00001678, NIH registered and Protocol N ° TW-01-004). 145
146
2.3. Xenodiagnosis 147
Dogs and cats were examined by xenodiagnosis with 10 or 20 uninfected, 148
laboratory-reared fourth-instar nymphs of T. infestans exposed to the animal‟s belly 149
during 20 min, followed by a 10-minute re-exposure period if initially most bugs had 150
not blood-fed to repletion (Enriquez et al., 2013a). Each bug was individually examined 151
by optical microscopy (OM) (400×) for T. cruzi infection 30 and 60 days after exposure. 152
The infectiousness to the vector of dogs and cats that were seropositive for T. cruzi was 153
calculated as the total number of insects infected with T. cruzi divided by the total 154
number of insects exposed to the animal and examined for infection at least once 155
(Gürtler et al., 1992). Bug mortality rate at 30 days post-exposure was 3% in both dogs 156
and cats; and bugs exposed to dogs molted after a single blood meal (12%) more 157
frequently than bugs fed on cats (8%), though these differences were not statistically 158
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
significant (Enriquez et al., 2013b). Trypanosoma cruzi-infected dogs and cats were 159
also classified as having high or low infectiousness to T. infestans; the “high 160
infectiousness” category included dogs and cats that were responsible for ≥80% of all 161
infected triatomines used in the xenodiagnostic tests, and the “low infectiousness” 162
included the remaining seropositive dogs and cats and that were also examined by 163
xenodiagnoses. 164
165
2.4. Serodiagnosis 166
Dogs and cats aged 4 months or more were diagnosed serologically whereas 167
younger animals were examined only by xenodiagnosis because maternally-derived 168
antibodies to T. cruzi could induce a false-positive result. Seventy (70%) dogs and 27 169
(93%) cats examined by xenodiagnosis were also examined by serodiagnosis. Sera were 170
tested for antibodies to T. cruzi by indirect hemagglutination assay (IHA) following the 171
manufacturer‟s instructions (Wiener Laboratories S.A.I.C., Buenos Aires, Argentina), 172
an in-house enzyme-linked immunosorbent assay (ELISA), and an indirect 173
immunofluorescence test (IFAT) (Ififluor Parasitest Chagas, Laboratorio IFI, Buenos 174
Aires, Argentina). The serodiagnosis methodologies and results of dog and cat 175
populations have been reported elsewhere (Enriquez et al., 2013b). An animal was 176
considered “seropositive” if reactive by at least two assays. 177
178
2.5. Molecular analysis 179
2.5.1. DNA extraction 180
Guanidine-EDTA blood samples (GEB) of 2 and 4 mL were heated in boiling 181
water for 10 and 15 min, respectively (Britto et al., 1993). Prior to DNA extraction, 200 182
pg of an internal amplification control DNA (IAC) was added to 400 µL GEB aliquot 183
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
(Duffy et al., 2009). Total DNA was purified using a commercial kit (DNeasy Blood & 184
Tissue Kit, QIAGEN Sciences, Maryland, USA) and following manufacturer´s 185
instructions avoiding the use of proteinase K and the addition of buffer AL, as reported 186
(Duffy et al., 2009). Purified DNA was eluted in 200 µL of distilled water and used as 187
template for PCR and qPCR amplification. 188
189
2.5.2. Polymerase chain reaction 190
DNA from seropositive dogs and cats that were also examined by xenodiagnosis 191
were tested by means of qualitative and qPCR assays. The qualitative PCR assay was 192
targeted to the minicircles of the kinetoplast (kDNA-PCR). The rectal-ampoule samples 193
of 94 OM-negative bugs used in xenodiagnosis of the seropositive dogs was examined 194
by kDNA-PCR to assess the false-negative rate of OM procedures. In addition, we 195
included samples from 16 dogs and 7 cats that resided in the study area and were not 196
infected by T. cruzi as determined by ELISA, IHA and xenodiagnosis (i.e., control 197
hosts). The comparison of kDNA-PCR results and xenodiagnosis has been reported 198
elsewhere (Enriquez et al., 2013b). 199
200
2.5.3. Real-time PCR assay 201
The bloodstream parasite load in dogs and cats was quantified by amplifying a 202
T. cruzi satellite DNA flanked by the Sat Fw (5'-GCAGTCGGCKGATCGTTTTCG-3') 203
and Sat Rv (5'-TTCAGRGTTGTTTGGTGTCCAGTG -3') oligonucleotides (Duffy et 204
al., 2009). The standard curve for DNA quantification (generated with T. cruzi CL 205
Brener clone), reagents, control samples and cycling profile for T. cruzi DNA 206
amplification were performed as previously described for human samples (Bua et al., 207
2012). Samples were run in duplicates using a commercial kit (SYBR GreenER qPCR 208
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
SuperMix Universal, Invitrogen, Life Technologies, Grand Island, NY, USA) and 8 μL 209
of template DNA in a 20 μL final volume. 210
The standard calibration curve for IAC was carried out using blood samples of 211
dogs not infected with T.cruzi (i.e., dogs residing in non-endemic areas). Samples were 212
spiked with 400 pg, 40 pg and 4 pg of IAC. The IAC was quantified using 0.6 μM of 213
primers, IS Fw (5'-AACCGTCATGGAACAGCACGTAC-3') and IS Rv (5'-214
CTAGAACATTGGCTCCCGCAACA-3') (Duffy et al., 2009) using a commercial kit 215
(Quantitec SYBR Green PCR kit, QIAGEN Sciences, Maryland, USA). After 15 min of 216
pre-incubation at 95 ºC, PCR amplification was carried out for 40 cycles (94 ºC for 15 s, 217
59 ºC for 30 s and 72 ºC for 30 s). 218
Given that the number of satellite DNA repeats differs among different T. cruzi 219
DTUs, DNA quantification was normalized according to the identified DTU (Duffy et 220
al., 2009). Parasite DNA concentration was expressed as equivalent amounts of parasite 221
DNA per ml (Pe/mL). An ABI 7500 thermocycler (Applied Biosystems, Carlsbad, CA, 222
USA) was used in all qPCR amplifications. 223
224
2.5.4. DTU identification 225
Parasite DTUs were identified by PCR targeted to three genomic markers: spliced-226
leader (SL) DNA, 24Sα ribosomal RNA genes and A10 marker as described by Burgos 227
et al. (2007). PCR products were analyzed in 2.5% or 3% agarose gels (Invitrogen, 228
USA) and UV visualized after staining with Gel Red (GenBiotech). Identification of 229
DTUs infecting dogs and cats based on isolated parasites was reported elsewhere 230
(Enriquez et al., 2013a). In the current study, additional DTU identifications from 231
seropositive, xenodiagnosis-negative dogs (n = 6) and cats (n = 5) and from a 232
seropositive, xenodiagnosis-positive dog were attempted directly from GEB or rectal 233
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
ampoule samples of OM-negative, kDNA-PCR-positive bugs used in xenodiagnostic 234
tests of these animals. These additional dogs and cats belonged to the same villages, and 235
samples were taken at the same time as those included in Enriquez et al. (2013a). DNA 236
was extracted from the bugs‟ rectal ampoule using DNAzol® (Invitrogen, USA) as 237
described previously (Maffey et al., 2012). DTU identifications were performed 238
introducing the following modifications to avoid unspecific amplifications within the 239
Spliced Leader region. SL-IR I, using primers TC2 and UTCC, was used to identify TcI 240
(475 bp) as follows: one cycle at 94 ºC for 3 min, 6 cycles at 62 ºC for 40s, 6 cycles at 241
60 ºC for 40 s, 40 cycles at 58 ºC for 40s and final extension step at 72 ºC for 10 min; 242
and SL-IR II, using primers TC1 and UTCC, was used to identify TcII, TcV, and TcVI 243
(425 bp) as follows: one cycle at 94 ºC for 3 min, 5 cycles at 68 ºC for 40s, 5 cycles at 244
66 ºC for 40 s, 5 cycles at 64 ºC for 40 s, 3 cycles at 62 ºC for 40 s, 30 cycles at 60 ºC 245
for 40s and final extension step at 72 ºC for 10 min. To enhance T. cruzi detection, PCR 246
reactions were carried out at a final volume of 50 μL of reaction mix, with 8 μL of DNA 247
sample as template and using Taq Platinum DNA polymerase (Invitrogen, USA). Given 248
that two unspecific bands in the SL-IR I PCR were persistently obtained in GEB 249
samples, the expected amplicon of 475 pb was purified and sequenced to confirm TcI 250
identification. 251
Incomplete identifications of DTUs were achieved in some cases in which only a 252
425bp-band was obtained in the SL-IR II PCR and no bands in the other PCR assays 253
due to low DNA content. These cases were described as TCII / V / VI. 254
255
2.6. Sequence analysis of SL-IR I 256
The PCR-amplified DNA was purified with Qiaquick Gel Extraction kit 257
(Qiagen) and cloned into pGEMT-easy vector following the manufacturer´s instructions 258
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
(Promega, Madison, WI). Recombinant plasmids were screened by digestion with 259
EcoRI (New England Biolabs) and sequenced using T7 and SP6 universal primers by 260
Macrogen (Inc., Seoul, Korea) in an ABI PRISM 3730 DNA Analyzer automated 261
sequencer (Applied Biosystems) and aligned manually or using MEGA 5.1 software 262
(Tamura et al., 2007). Newly reported sequences are available at GenBank under 263
accession numbers KF964667, KF964668 and KF964669. A consensus of forward and 264
reverse sequences was created and a BLAST search was performed in GenBank 265
database to compare sequence identity. 266
2.7. Data analysis 267
Agresti-Coull binomial 95% confidence intervals (CI) were used for proportions. 268
The relationship between infectiousness to the vector of dogs and cats infected with T. 269
cruzi and potential predictors was analysed using maximum likelihood multiple logistic 270
regression implemented in Stata (Stata 10.1, Stata Corp, College Station, Texas). Only 271
dogs and cats aged 1 year or more were included in the regression to avoid potential 272
confounders due to growth effects. Because the bugs used in xenodiagnosis were 273
clustered on individual dogs or cats, observations are not independent. Therefore, the 274
xtmelogit command was used to include random effects that measure the residual 275
effects due to each subject on the probability of bug infection. The dependent variable 276
was the infection status of each bug used in xenodiagnosis. The independent variables 277
were the bloodstream parasite load (a continuous variable, in Pe/mL), body condition 278
(two levels: 0 = good, 1 = regular or poor); age of the host (a continuous variable, in 279
months); exposure to infected bugs, as measured by the abundance of T. infestans 280
infected with T. cruzi captured per 15 min-person per site in domiciles, kitchens and 281
storerooms (i.e., typical resting sites of dogs and cats), categorized in low, medium and 282
high-risk levels, 0 = 0; 1 = 1-5, and 2 ≥ 5 bugs, respectively; and identified DTU (0 = 283
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
TcV, 1 = TcVI; 2 = TcI or TcIII). Identified DTU was excluded as a predictor in cats 284
due to the small sample size. The body condition status was divided into two levels 285
because of the low number of dogs with a poor body condition. Interaction terms were 286
added stepwise and dropped from the final model if not significant at a nominal 287
significance level of 5%. The Wald test examined the hypothesis that all regression 288
coefficients are 0. 289
To reflect the potential ability of a dog in a given body condition category to 290
infect a bug after feeding to repletion, we calculated an Infectious Potential Index (IPI) 291
for each category (Petersen et al., 2001) as the product of the proportion of seropositive 292
dogs, the proportion of seropositive dogs with a positive xenodiagnosis, and the median 293
infectiousness of seropositive dogs with a positive xenodiagnosis in each category. 294
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
3. Results 295
3.1. Parasite concentration 296
A total of 44 seropositive dogs and 14 seropositive cats that had been examined 297
by xenodiagnosis were assayed by qPCR, and additionally, a xenodiagnosis-positive, 298
seronegative cat was also tested. All of these samples were qPCR-positive except one 299
seropositive cat that was negative by xenodiagnosis and kDNA-PCR (Table 3). All 300
control dogs (n = 16) and cats (n =7) that were both seronegative and xenodiagnosis-301
negative were also negative by kDNA-PCR and qPCR. In two xenodiagnosis-negative, 302
qPCR-positive cats, DNA quantifications could not be normalized because DTU 303
identifications were unsuccessful, therefore they were excluded from further analysis. 304
The frequency distribution of parasite DNA concentration estimated by qPCR 305
showed a negative binomial distribution in both host populations (Figure 1). The 306
median bloodstream parasite load was slightly higher in cats (median = 9.7 Pe/mL, first-307
third quartiles [Q1-Q3] = 2.9-96.9, range 0.0-1101.8) than in dogs (median = 8.1 308
Pe/mL, Q1-Q3 = 1.5-26.6, range 0.7-214.5), although this difference was not 309
statistically significant (Kruskal-Wallis test, P = 0.5). If the two seropositive, 310
xenodiagnosis-negative cats in which DNA quantifications could not be normalized had 311
been infected either with TcI or TcVI (i.e., the extreme cases for normalization), their 312
parasitemia would range between <1 and 10 Pe/mL. 313
314
3.2. Infectiousness to the vector and parasite load 315
Overall, 86% of dogs (n = 44) and 64% of cats (n = 14) seropositive for T. cruzi 316
were positive by xenodiagnosis. The mean infectiousness to T. infestans did not differ 317
significantly between dogs (48%, 95% confidence interval [CI] = 33-62%) and cats 318
(44%, 95% CI = 10-63%). Thirteen percent of 94 OM-negative bugs were positive by 319
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
kDNA-PCR; thus, the corrected infectiousness to the vector would rise to 54%. The 320
frequency distribution of infectiousness showed a bimodal pattern in both host 321
populations (Figure 2). 322
The infectiousness of infected dogs and cats increased steeply with parasite load, 323
reaching maximum infectiousness (95%) at 96.9 Pe/mL for cats and 13.0 Pe/mL for 324
dogs (Figure 3). No clear association was observed between DTU and host 325
infectiousness or bloodstream parasite load (Figure 3). However, the two cats and the 326
only dog infected with TcI had nil infectiousness despite having 7.4, 9.7 and 20.5 327
Pe/mL, respectively (Figure 3) (Appendix). In these three individuals TcI was 328
confirmed by sequence analysis of Spliced Leader region (SL-IR I) (i.e., 98-100% 329
sequence identity). 330
Random-effects multiple logistic regression model showed that infectiousness to 331
the vector of seropositive dogs was significantly associated with the bloodstream 332
parasite load (OR = 1.02, 95% CI = 1.00-1.04, P < 0.05) and body condition (OR = 333
4.45, 95% CI = 1.24-16.01, P = 0.02) (Wald χ2
= 11.76, P < 0.01) (Table 1). No 334
association was found between infectiousness and age of the dog, exposure to infected 335
bugs at the dog‟s house and DTU. The median parasite load in dogs with poor or regular 336
body condition (9.93 Pe/mL, Q1-Q3 = 3.03-28.15) was approximately 4 times higher 337
than in dogs with good condition (2.67 Pe/mL, Q1-Q3 = 1.38-18.90), but this difference 338
was not statistically significant (Kruskal-Wallis, P = 0.17). 339
The infectiousness of infected cats was significantly associated with parasite load 340
(OR = 1.02, 95% CI = 1.00-1.04, P = 0.02) (Wald χ2
= 4.83, P = 0.09) (Table 1). No 341
significant association between infectiousness and age of the cat and exposure to 342
infected bugs was found by random-effects multiple logistic regression analysis. 343
344
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
3.3. Body condition and infectiousness 345
The classification of 111 dogs aged ≥ 1 year showed that 52%, 43% and 5% of 346
them were in a good, regular and poor body condition, respectively (Table 2). The mean 347
(±SD) age (3.0±2.6, 4.0±2.6, and 5.5±6.9 years) and the proportion of female dogs 348
(36%, 39% and 40%, χ2 = 0.2, df: 2, P = 0.9) were similar among categories ranging 349
from good to poor conditions, respectively. The prevalence of T. cruzi infection varied 350
between 50% and 80% and was not significantly different among body condition 351
categories (χ2
= 1.7, df: 2, P = 0.2). The median infectiousness increased from 30% in 352
dogs with good condition to 63% and 80% in animals with a regular and poor condition, 353
respectively. Dogs with a poor condition had an infectious potential index (IPI) 2.2 354
times greater than dogs with a regular condition, and 4.9 times greater than those in a 355
good condition (Table 2). However, “poor” dogs had a very low sample size of 5 versus 356
the total sample size of 44. 357
358
3.4. Profile of infected dogs and cats with high or low infectiousness 359
Nearly half of the dogs (50%) and cats (47%) showed high infectiousness and 360
were responsible of at least 80% of the infected xenodiagnosis bugs. The median 361
parasite load was nearly 10 times greater for dogs with high infectiousness than for dogs 362
with low infectiousness and this difference was statistically significant (Kruskal-Wallis 363
test, P < 0.01). For cats this relationship was 32 times greater (Kruskal-Wallis test, P = 364
0.02) (Table 3). Quantitative PCR could not distinguish between highly infectious and 365
non-highly infectious hosts, given that it detected T. cruzi DNA in 100% of the 366
seropositive dogs and in 93% of the seropositive cats, whereas kDNA-PCR detected 367
89% and 87% of them, respectively (Table 3). Four seropositive dogs and one 368
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
seropositive cat that were both qPCR-positive and kDNA-PCR- negative showed very 369
low infectiousness to the vector (range, 0-5%). 370
371
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
4. Discussion 372
By means of a real-time quantitative PCR protocol, this study shows that the 373
parasite load detected in bloodstream of dogs and cats naturally infected with T. cruzi 374
was closely associated with their infectiousness to the vector, and on average, was 4-5 375
times higher than that of chronic Chagas disease patients examined elsewhere (Freitas et 376
al., 2011; Moreira et al., 2013). 377
The median bloodstream parasite load was 8.1 Pe/mL in dogs and 9.7 Pe/mL in 378
cats, whereas in chronic patients was ≤ 2 Pe/mL (Freitas et al., 2011; Moreira et al., 379
2013). Furthermore, the sensitivity of qPCR in seropositive, chronic patients ranged 380
from 41% to 76% (Duffy et al., 2013; Moreira et al., 2013; Piron et al., 2007), whereas 381
in our study 93-100% of seropositive cats and dogs were qPCR-positive. These 382
substantial differences in parasite load most likely explain the much lower proportion of 383
T. cruzi-seropositive humans with a positive xenodiagnosis and their much lower 384
infectiousness to bugs relative to seropositive dogs and cats (Coura et al., 1991; Freitas 385
et al., 2011; Gürtler et al., 1996). 386
In our study, the infectiousness of dogs and cats increased non-linearly from 0% 387
to 85% with small increases in bloodstream parasite load. Two direct implications 388
follow from these observations: first, a positive xenodiagnosis can be obtained with 389
very low parasite load (i.e., 0.7 and 1.6 Pe/mL for dogs and cats, respectively), although 390
the ultimate fraction of bugs that become infected may be highly variable. Second, 391
given that 96% of the dogs and 90% of cats had more than 0.7-1.6Pe/mL, a large 392
proportion of them is expected to be highly infectious to the vector, as our study attests. 393
The wide variability in the infectiousness of dogs and cats with very low 394
bloodstream parasite load may be related, at least in part, to unrecorded variations in the 395
effective bloodmeal size achieved by the bugs used in xenodiagnosis. The actual 396
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
infectiousness is expected to be greater than 44-48% because of the limited sensitivity 397
of optical microscopy when the intensity of rectal infections is low (false-negative rate 398
was 13%). 399
There were a few cases in which the high infectiousness observed could not be 400
explained by the estimated parasite load of the individual host on which the bugs had 401
fed, as in a TcVI-infected dog with a regular body condition and 0.8 Pe/mL which 402
infected 17 of 20 insects. Perhaps the differential origin of the blood samples for qPCR 403
(venous blood) and xenodiagnosis (capillary blood ingested by the bugs) may explain 404
this apparent inconsistency (Ferreira et al., 2007). Parasite DNA concentrations differed 405
between tissues in dogs naturally infected with Leishmania infantum (Ferreira et al., 406
2013). Technical errors in the quantification of bloodstream parasite load may be ruled 407
out because all blood samples from dogs with very low parasite load and large 408
infectiousness were repeated and the initial outcome corroborated (results not shown). 409
Although rather unlikely, we cannot rule out the presence of inhibitors that affected the 410
performance of qPCR despite IAC amplifications were consistently satisfactory. New 411
protocols based on multiplex qPCR amplificating both DNA targets (i.e., DNA of T. 412
cruzi and IAC) in the same reaction tube could enhance the reproducibility of qPCR 413
assays (Duffy et al., 2013). 414
Dogs and cats had similar infectiousness to the vector and this distribution was 415
aggregated, as in studies conducted elsewhere (Gürtler et al., 2007). There was a 416
fraction of highly infectious dogs and cats that may contribute disproportionally to 417
(peri)domestic transmission and persistence of T. cruzi. This heterogeneous distribution 418
of parasite burden in host populations is a generalized pattern in macroparasites and 419
microparasites (Wilson et al., 2001; Woolhouse et al., 1997) and may be used for 420
targeted control strategies. Trypanosoma cruzi-infected Dasypus novemcinctus 421
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
armadillos and Didelphis albiventris opossums also have high infectiousness (Orozco et 422
al., 2013). 423
Multiple logistic regression analysis showed that the infectiousness of dogs was 424
associated with the bloodstream parasite load and the body condition of the individual 425
host. The latter is likely reflecting the combined effects of the immune system and the 426
nutritional status of the host (Keusch, 2003; Machado-Coelho et al., 2005; Petersen et 427
al., 2001). Dogs with a regular or poor body condition showed a nearly four-fold higher 428
median parasite load than dogs in good conditions. The former were probably less 429
effective in controlling parasitemia for reasons that remain to be identified. 430
Exposure to repeated reinfections with T. cruzi via inoculations was positively 431
correlated with a slight and transient increase in parasitemia in experimentally infected 432
dogs and mice (Machado et al., 2001; Bustamante et al., 2007). In our study, however, 433
exposure to infected bugs was not associated with increased infectiousness in dogs; 434
previous surveys yielded mixed outcomes on this point (Gürtler et al., 2007; 1992). A 435
possible explanation for these discrepancies is that our metric of exposure in the 436
regression model may not represent in a precise manner the actual likelihood of 437
reinfection the dogs faced throughout its home range in the past. In addition, we found 438
no association between infectiousness and age of dogs (as in one of three previous 439
studies: Gürtler et al., 2007; 1996; 1992) and parasite DTU (Cardinal et al., 2008). 440
Other factors not evaluated here, such as concomitant parasitic infections, stress levels, 441
inoculum size, and genetic background of host and parasite, may eventually mask the 442
effects of age, DTU and reinfections on infectiousness. Moreover, the existence of a 443
polymorphism in the genes involved in the immune response could also affect 444
infectiousness, as in dogs infected with L. infantum (Quinnell et al., 2003). In cats, 445
infectiousness was only associated with parasite load. TcVI-infected cats were more 446
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
infectious that the few ones infected with TcV or TcI, but samples sizes were too small 447
to make any sound conclusion. 448
The concurrent finding of TcI infections with bloodstream parasite load ranging 449
from 7.4 to 20.5 Pe/mL and nil infectiousness (as determined by optical microscopy) 450
suggest a differential behavior of TcI or a different susceptibility of the vector species 451
involved. Indeed, TcI-infected dogs and cats were rare and could only be identified by 452
means of a highly sensitive kDNA-PCR applied to the rectal-ampoule lysates of 453
xenodiagnostic bugs or GEB samples. The substantial variations in metacyclogenesis 454
and trypomastigote density among triatomine species (Carvalho-Moreira et al., 2003; 455
Loza-Murghía et al., 2010; Perlowagora-Szumlewicz et al., 1994) may explain the nil 456
infectiousness (as measured by optical microscopy) recorded in TcI-infected dogs and 457
cats. However, samples sizes were small and this relationship should be further 458
investigated. 459
In our study, approximately 50% of the infected dogs and cats were responsible 460
for at least 80% of the infected T. infestans used in xenodiagnosis. All of them were 461
positive by qPCR and kDNA-PCR. However, dogs and cats with very low 462
infectiousness to the vector (<20%) were also frequently positive by these techniques. 463
These results suggest that the quantitative estimates of parasite load and not the 464
qualitative outcomes of qPCR or kDNA-PCR would be a more precise tool to identify 465
highly infectious dogs and cats. A wide variability in infectiousness (range, 0-63% for 466
dogs and 35-67% for cats) was observed between 1-5 Pe/mL. Courtenay et al. (2014) 467
recently showed that quantitative PCR was needed to identify highly infectious L. 468
infantum-infected dogs. 469
Quantitative PCR was more sensitive and less time-consuming than 470
xenodiagnosis, and alone or in conjunction with the body condition index, may be 471
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
helpful to identify highly infectious hosts; replace xenodiagnosis (or hemoculture) to 472
assess treatment efficacy or infection status, and contribute to novel control strategies 473
(e.g., targeted treatment or vaccination) that reduce the risks of domestic transmission of 474
T. cruzi. 475
476
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
ACKNOWLEDGMENTS 477
We are grateful to Carolina Cura and Margarita Bisio for laboratory assistance on 478
molecular biology. Romina Piccinali, Marina Leporace, Francisco Petrocco, Laura 479
Tomassone, Leonardo Ceballos, Julián Alvarado-Otegui, Sol Gaspe and Nala for field 480
or laboratory assistance, and to Lucía Maffey, Pilar Fernández, Fernando Garelli, Juan 481
Gurevitz, Yael Provecho, Jimena Gronzo and Carla Cecere for valuable comments. 482
Special thanks to the villagers of Pampa del Indio for kindly welcoming us into their 483
homes and cooperating with the investigation. 484
485
FINANCIAL SUPPORT 486
This study received financial support from International Development Research 487
Center (Eco-Health Program); the UNICEF/UNDP/World Bank/WHO Special 488
Programme for Research and Training in Tropical Diseases (TDR) to REG; Agencia 489
Nacional de Promoción Científica y Tecnológica to REG; Agencia Nacional de 490
Promoción Científica y Tecnológica and GlaxoSmithKline to REG, and University of 491
Buenos Aires to REG. JB, MMO, SW, AGS, REG and MVC are members of 492
CONICET Researcher‟s Career. 493
494
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
References 495
Alvarenga, N.J., Leite, M.A., 1982. Injection of Trypanosoma cruzi into the gut of 496
triatomine bugs: number required to infect the vector. Trans. R. Soc. Trop. Med. 497
Hyg. 76, 708-709. 498
Britto, C., Cardoso, M.A., Wincker, P., Morel, C.M., 1993. A simple protocol for the 499
physical cleavage of Trypanosoma cruzi kinetoplast DNA present in blood 500
samples and its use in polymerase chain reaction (PCR)-based diagnosis of 501
chronic Chagas disease. Mem. Inst. Oswaldo Cruz 88, 171-172. 502
Bua, J., Volta, B.J., Velazquez, E.B., Ruiz, A.M., Rissio, A.M., Cardoni, R.L., 2012. 503
Vertical transmission of Trypanosoma cruzi infection: quantification of parasite 504
burden in mothers and their children by parasite DNA amplification. Trans. R. 505
Soc. Trop. Med. Hyg. 106, 623-628. 506
Bua, J., Volta, B.J., Perrone, A.E., Scollo, K., Velázquez, E.B., Ruiz, A.M., De Rissio, 507
A.M., Cardoni, R.L., 2013. How to improve the early diagnosis of Trypanosoma 508
cruzi infection: relationship between validated conventional diagnosis and 509
quantitative DNA amplification in congenitally infected children. PLoS. Negl. 510
Trop. Dis. 7, e2476. 511
Burgos, J.M., Altcheh, J., Bisio, M., Duffy, T., Valdares, H.M.S., Seidenstein, M.E., 512
Piccinali, R., Freitas, J.M., Levin, M.J., Machi, L., Macedo, A.M., Freilij, H., 513
Schijman, A.G., 2007. Direct molecular profiling of minicircle signatures and 514
lineages of Trypanosoma cruzi bloodstream populations causing congenital 515
Chagas disease. Int. J. Parasitol. 37, 1319-1327. 516
Bustamante, J.M., Novarese, M., Héctor, W., Rivarola, H.W., Lo Presti, M.S., 517
Fernández, A.R., Enders, J.E., Fretes, R., Paglini-Oliva, P.A., 2007. Reinfections 518
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
and Trypanosoma cruzi strains can determine the prognosis of the chronic 519
chagasic cardiopathy in mice. Parasitol. Res. 100, 1407-1410. 520
Cardinal, M.V., Lauricella, M.A., Ceballos, A.L., Lanati, L., Marcet, P.L., Levin, M.J., 521
Kitron, U., Gürtler, R.E., Schijman, A.G., 2008. Molecular epidemiology of 522
domestic and sylvatic Trypanosoma cruzi infection in rural northwestern 523
Argentina. Int. J. Parasitol. 38, 1533-1543. 524
Carvalho-Moreira, C.J., Spata, M.C., Coura, J.R., Garcia, E.S., Azambuja, P., Gonzalez, 525
M.S., Mello, C.B., 2003. In vivo and in vitro metacyclogenesis of two strains of 526
Trypanosoma cruzi in the vectors Triatoma pseudomaculata and Rhodnius 527
neglectus: short/long-term and comparative study. Exp. Parasitol. 103, 102-111. 528
Crisante, G., Rojas, A., Teixeira, M.M.G., Añez, N., 2006. Infected dogs as a risk factor 529
in the transmission of human Trypanosoma cruzi infection in western 530
Venezuela. Acta Trop. 98, 247-254. 531
Coura, J.R., de Abreu, L.L., Willcox, H.P., Petana, W., 1991. Evaluation of the 532
xenodiagnosis of chronic Chagas patients infected ten years or over in an area 533
where transmission has been interrupted--Iguatama and Pains, west Minas 534
Gerais State, Brazil. Mem. Inst. Oswaldo Cruz 86, 395-398. 535
Courtenay, O., Carson, C., Calvo-Bado, L., Garcez, L.M., Quinnell, R.J., 2014. 536
Heterogeneities in Leishmania infantum infection: using skin parasite burdens to 537
identify highly infectious dogs. PLoS. Negl. Trop. Dis. 8, e2583. 538
da Silveira Pinto, A., de Lana, M., Britto, C., Bastrenta, B., Tibayrenc, M., 2000. 539
Experimental Trypanosoma cruzi biclonal infection in Triatoma infestans: 540
detection of distinct clonal genotypes using kinetoplast DNA probes. Int. J. 541
Parasitol. 30, 843-848. 542
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
de Lana, M., da Silveira Pinto, A., Barnabé, C., Quesney, V., Noël, S., Tibayrenc, M., 543
1998. Trypanosoma cruzi: compared vectorial transmissibility of three major 544
clonal genotypes by Triatoma infestans. Exp Parasitol. 90, 20-25. 545
Duffy, T., Bisio, M., Altcheh, J., Burgos, J.M., Diez, M., Levin, M.J., Favaloro, R.R., 546
Freilij, H., Schijman, A.G., 2009. Accurate real-time PCR strategy for 547
monitoring bloodstream parasitic loads in Chagas disease patients. PLoS. Negl. 548
Trop. Dis. 3, e419. 549
Duffy, T., Cura, C.I., Ramirez, J.C., Abate, T., Cayo, N.M., Parrado, R., Bello, Z.D., 550
Velazquez, E., Muñoz-Calderon, A., Juiz, N.A., Basile, J., Garcia, L., Riarte, A., 551
Nasser, J.R., Ocampo, S.B., Yadon, Z.E., Torrico, F., de Noya, B.A., Ribeiro, I., 552
Schijman, A.G., 2013. Analytical performance of a multiplex Real-Time PCR 553
assay using TaqMan probes for quantification of Trypanosoma cruzi satellite 554
DNA in blood samples. PLoS. Negl. Trop. Dis. 7, e2000. 555
Enriquez, G.F., Cardinal, M.V., Orozco, M.M., Lanatti, L., Schijman, A.G., Gürtler, 556
R.E., 2013a. Discrete typing units of Trypanosoma cruzi identified in rural dogs 557
and cats in the humid Argentinean Chaco. Parasitology 12, 1-6. 558
Enriquez, G.F., Cardinal, M.V., Orozco, M.M., Schijman, A.G., Gürtler, R.E., 2013b. 559
Detection of Trypanosoma cruzi infection in naturally infected dogs and cats 560
using serological, parasitological and molecular methods. Acta Trop. 126, 211-561
217. 562
de Freitas, V.L., da Silva, S.C., Sartori, A.M., Bezerra, R.C., Westphalen, E.V., Molina, 563
T.D., Teixeira, A.R., Ibrahim, K.Y., Shikanai-Yasuda, M.A., 2011. Real-time 564
PCR in HIV/Trypanosoma cruzi coinfection with and without Chagas disease 565
reactivation: association with HIV viral load and CD4 level. PLoS. Negl. Trop. 566
Dis. 5, e1277. 567
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Ferreira, R.A., Lazzari, C.R., Lorenzo, M.G., Pereira, M.H., 2007. Do haematophagous 568
bugs assess skin surface to detect blood vessel? PLoS. One 2, e932. 569
Ferreira, S. de A., Almeida, G.G., Silva, S. de O., Vogas, G.P., Fujiwara, R.T., de 570
Andrade, A.S., Melo, M.N., 2013. Nasal, oral and ear swabs for canine visceral 571
leishmaniasis diagnosis: new practical approaches for detection of Leishmania 572
infantum DNA. PLoS. Negl. Trop. 7, e2150. 573
Garcia, E.S., Ratcliffe, N.A., Whitten, M.M., Gonzalez, M.S., Azambuja, P., 2007. 574
Exploring the role of insect host factors in the dynamics of Trypanosoma cruzi-575
Rhodnius prolixus interactions. J. Insect. Physiol. 53, 11-21. 576
Gurevitz, J.M., Gaspe, M.S., Enriquez, G.F., Vassena, C., Alvarado-Otegui, J.A., 577
Provecho, Y.M., Mougabure-Cueto, G., Picollo, M.I., Kitron, U., Gürtler, R.E., 578
2011. Unexpected pyrethroid resistance and control failures of Chagas disease 579
vector in Argentina. J. Med. Entomol. 49, 1379-1386. 580
Gürtler, R.E., Petersen, R.M., Rubel, D.N., Lauricella, M.A., Wisnivesky-Colli, C., 581
1992. Infectivity to the vector Triatoma infestans of dogs infected with 582
Trypanosoma cruzi in north-west Argentina. Ann. Trop. Med. Parasitol. 86, 111-583
119. 584
Gürtler, R.E., Cecere, M.C., Castañera, M.B., Canale, D., Lauricella, M.A., Chuit, R., 585
Cohen, J.E., Segura, E.L., 1996. Probability of infection with Trypanosoma cruzi 586
of the vector Triatoma infestans fed on infected humans and dogs in northwest 587
Argentina. Am. J. Trop. Med. Hyg. 55, 24-31. 588
Gürtler, R.E., Cecere, M.C., Lauricella, M.A., Cardinal, M.V., Kitron, U., Cohen, J.E., 589
2007. Domestic dogs and cats as sources of Trypanosoma cruzi infection in rural 590
northwestern Argentina. Parasitology 134, 69-82. 591
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Keusch, G.T., 2003. The history of nutrition: malnutrition, infection and immunity. J. 592
Nutr. 133, S336–S340. 593
Loza-Murguía, M., Noireau, F., 2010. Vectorial capacity of Triatoma guasayana 594
(Wygodzinsky&Abalos) (Hemiptera: Reduviidae) compared with two other 595
species of epidemic importance. Netrop. Entomol. 39, 799-809. 596
Machado-Coelho, G., Caiaffa, W., Genaro, O., Magalhães, P., Mayrink, W., 2005. Risk 597
factors for mucosal manifestation of American cutaneous leishmaniasis. Trans. 598
R. Soc. Trop. Med. Hyg. 99, 55–61. 599
Machado, E.M.M., Fernandes, A.J., Murta, S.M.F., Vitor, R.W.A., Camilo, D.J.Jr., 600
Pinheiro, S.W., Lopes, E.R., Adad, S.J., Romanha, A.J., Pinto Dias, J.C., 2001. 601
A study of experimental reinfection by Trypanosoma cruzi in dogs. Am. J. Trop. 602
Med. Hyg. 65, 958-965. 603
Maffey, L., Cardinal, M.V., Ordóñez-Krasnowski, P.C., Lanati, L.A., Lauricella, M.A., 604
Schijman, A.G., Gürtler, R.E., 2012. Direct molecular identification of 605
Trypanosoma cruzi Discrete Typing Units in domestic and peridomestic 606
Triatoma infestans and Triatoma sordida from the Argentine Chaco. 607
Parasitology 139, 1570-1579. 608
Moreira, O.C., Ramírez, J.D., Velázquez, E., Melo, M.F., Lima-Ferreira, C., Guhl, F., 609
Sosa-Estani, S., Marin-Neto, J.A., Morillo, C.A., Britto, C., 2013. Towards the 610
establishment of a consensus real-time qPCR to monitor Trypanosoma cruzi 611
parasitemia in patients with chronic Chagas disease cardiomyopathy: a substudy 612
from the BENEFIT trial. Acta Trop. 125, 23-31. 613
Moll-Merks, C., Werner, H., Dönges, J., 1987. Suitability of in-vitro xenodiagnosis: 614
development of Trypanosoma cruzi in Triatoma infestans depending on larval 615
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
stage of bugs and number of trypomastigotes taken during in-vitro blood meal. 616
Zentralbl. Bakteriol. Mikrobiol. Hyg. 268, 74-82. 617
Neal, R.A., Miles, R.A., 1977. The number of trypomastigotes of Trypanosoma cruzi 618
required to infect Rhodnius prolixus. Rev. Inst. Med. Trop. Sao Paulo 19, 177-619
181. 620
Noireau, F., Diosque, P., Jansen, A.M., 2009. Trypanosoma cruzi: adaptation to its 621
vectors and its hosts. Vet. Res. 40(2):26. 622
Perlowagora-Szumlewicz, A., Moreira, C.J., 1994. In vivo differentiation of 623
Trypanosoma cruzi--1. Experimental evidence of the influence of vector species 624
on metacyclogenesis. Mem. Inst. Oswaldo Cruz 89, 603-618. 625
Petersen, R.M., Gürtler, R.E., Cecere, M.C., Rubel, D.N, Hansen, D., Lauricella, M.A., 626
Carlomagno, M., 2001. Association between nutritional indicators of dogs 627
seroractive for Trypanosoma cruzi in a rural area of northwestern Argentina. 628
Parasitol. Res. 87, 208-214. 629
Piron, M., Fisa, R., Casamitjana, N., López-Chejade, P., Puig, L., Vergés, M., Gascón, 630
J., Gómez i Prat, J., Portús, M., Sauleda, S., 2007. Development of a real-time 631
PCR assay for Trypanosoma cruzi detection in blood samples. Acta Trop. 103, 632
195-200. 633
Orozco, M.M., Enriquez, G.F., Alvarado-Otegui, J.A., Cardinal, M.V., Schijman, 634
A.G., Kitron, U., Gürtler, R.E., 2013 New sylvatic hosts of Trypanosoma cruzi 635
and their reservoir competence in the humid Chaco of Argentina: a longitudinal 636
study. Am. J. Trop. Med. Hyg. 88, 872-882. 637
Quinnell, R.J., Kennedy, L.J., Barnes, A., Courtenay, O., Dye, C., Garcez, L.M., Shaw, 638
M.A., Carter, S.D., Thomson, W., Ollier W.E.R., 2003. Susceptibility to visceral 639
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
leishmaniasis in the domestic dog is associated with MHC class II 640
polymorphism. Immunogenetics 55, 23-28. 641
Ramírez, J.D., Turriago, B., Tapia-Calle, G., Guhl, F., 2013. Understanding the role of 642
dogs (Canis lupus familiaris) in the transmission dynamics of Trypanosoma 643
cruzi genotypes in Colombia. Vet. Parasitol. 196, 216-219. 644
Schijman, A.G., Bisio, M., Orellana, L., Sued, M., Duffy, T., Mejia Jaramillo, A.M., 645
Cura., Auter, F., Veron, V., Qvarnstrom, Y., Deborggraeve, S., Hijar, G., Zulantay, 646
I., Lucero, R.H., Velazquez, E., Tellez, T., Sanchez Leon, Z., Galvão, L., Nolder, 647
D., Monje Rumi, M., Levi, J.E., Ramirez, J.D., Zorrilla, P., Flores, M., Jercic, M.I., 648
Crisante, G., Añez, N., De Castro, A.M., Gonzalez, C.I., Acosta Viana, K., 649
Yachelini, P., Torrico, F., Robello, C., Diosque, P., Triana Chavez, O., Aznar, C., 650
Russomando, G., Büscher, P., Assal, A., Guhl, F., Sosa Estani, S., Da Silva, A., 651
Britto, C., Luquetti, A., Ladzins, J., 2011. International study to evaluate PCR 652
methods for detection of Trypanosoma cruzi DNA in blood samples from Chagas 653
disease patients. PLoS. Negl. Trop. Dis. 5, e931. 654
Stein, R.A., 2011. Super-spreaders in infectious diseases. Int. J. Infect. Dis. 15, 510-655
513. 656
Tamura, K., Peterson, D., Peterson, N., Stecher, G., Nei, M., Kumar, S., 2007. MEGA 657
5: Molecular Evolutionary Genetics Analysis using Maximum Likelihood, 658
Evolutionary Distance, and Maximum Parsimony Methods. Mol. Biol. Evol. 28, 659
2731-2739. 660
World Health Organization (WHO), 2002. Control of Chagas Disease: Second Report 661
of the WHO Expert Committee, World Health Organization, Geneva, 57-58. 662
Wilson, K., Bjørnstad, O.N., Dobson, A.P., Merler, S., Poglayen, G., Randolph, S.E., 663
Read, A.F., Skoping, A., 2001. Heterogeneities in macroparasite infections: 664
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
patterns and processes. In: Hudson, PJ., Rizzoli, A., Grenfell, BT., Heesterbeek, 665
H., Dobson, AP., editors. The Ecology of Wildlife Diseases. Oxford University 666
Press; Oxford, UK: p. 6-44. 667
Woolhouse, M.E, Dye, C., Etard, J.F., Smith, T., Charlwood, J.D., Garnett, G.P., 668
Hagan, P., Hii, J.L., Ndhlovu, P.D., Quinnell, R.J., Watts, C.H., Chandiwana, 669
S.K., Anderson, R.M., 1997. Heterogeneities in the transmission of infectious 670
agents: implications for the design of control programs. Proc. Natl. Acad. Sci. 671
94, 338-342. 672
673
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
Figure 1. Frequency distribution of parasite load of T. cruzi-infected dogs (black bars) 674
and cats (white bars); Pampa del Indio, Chaco, 2008. 675
676
Figure 2. Frequency distribution of infectiousness to T. infestans of T. cruzi-infected 677
dogs (black bars) and cats (white bars); Pampa del Indio, Chaco, 2008. 678
679
Figure 3. Distribution of infectiousness to T. infestans of T. cruzi-infected dogs (A) and 680
cats (B) according to parasite DNA quantification by qPCR and identified DTU; Pampa 681
del Indio, Chaco, 2008. The figure excludes two seropositive cats from which DNA 682
quantifications could not be normalized. 683
684
Table 1. Infectiousness to the vector T. infestans of dogs and cats infected for T. cruzi
according to potential risk factors; Pampa del Indio, Chaco, 2008.
Factor
Dogsa
Catsa
Median
infectiousness
% (first-third
quartiles) OR (95% CI)
Median
infectiousness
% (first-third
quartiles) OR (95% CI)
Host body condition
– –
Good 16 (5-68) 1
– –
Regular/Poor 64 (32-84) 4.45 (1.24-16.01)
– –
Parasitemia
(Pe/mL) – 1.02 (1.00-1.04)
– 1.02 (1.00-1.04)
Exposure to infected bugs (abundance of infected-bug captured per 15 min-person per
site)
0 55 (35-75) 1
67 (25-75) 1
1-5 68 (35-84) 0.41 (0.03-6.58) 79 (78-79) 0.21 (0.01-5.17)
>5 20 (5-65) 0.40 (0.02-9.43) 0 (0-50) 0.20 (0.01-6.66)
Age of host – 1.00 (0.99-1.02) – 1.18 (0.93-1.48)
DTU
TcV 30 (16-45) 1 – –
TcI or TcIII 38 (0-75) 0.58 (0.02-22.42) – –
TcVI 62 (30-80) 2.22 (0.45-10.94) – –
a. Three dogs and four cats aged <1 year were not included in the regression.
Table(1)
Table 2. Relationship between host body condition and infectiousness to T. infestans of
dogs infected with T. cruzi aged ≥1 year; Pampa del Indio, Chaco, 2008.
Host body
condition
Percentage of
seropositive
dogs (No.
examined)
Percentage of
seropositive dogs with
a positive
xenodiagnosis (No.
examined)
Median
infectiousness of
xenodiagnosis-
positive dogs
Infectious
Potential
Index
Good 55 (56) 78 (23) 30 0.13
Regular 50 (50) 93 (25) 63 0.29
Poor 80 (5) 100 (3) 80 0.64
Total 54 (111) 85 (41) 61 0.28
Table(2)
Table 3. Relationship among qualitative kDNA-PCR, qPCR and parasite load to detect
T. cruzi-infected dogs and cats with different levels of infectiousness
Host Infectiousnessa
No. positive
xenodiagnosis
(No.
examined)
Mean
infectiousness
(%)
Median parasite
load (Q1-Q3)
No.
Positive
kDNA
(%)
No.
Positive
qPCR
(%)
Dogs High 22 (22) 79 21.3 (7.3-39.3) 22 (100) 22 (100)
Low 16 (22) 19 2.4 (1.3-12.6) 17 (77) 22 (100)
Total 38 (44) 48 8.1 (1.5-26.6) 39 (89) 44 (100)
Cats High 7 (7) 79 96.1(18.1-202.1) 7 (100) 7 (100)
Low 3 (8) 14 3.0 (2.2-7.4) 6 (75) 7 (88)
Total 10 (15) 44 9.7 (2.9-96.9) 13 (87) 14 (93)
a. The “high infectiousness” category included dogs and cats that were responsible for
≥80% of all infected triatomines used in the xenodiagnostic tests. The “low
infectiousness” category included all remaining seropositive dogs and cats examined by
xenodiagnosis.
Table(3)
Parasite equivalents per mL
0 1-30 >30-60 >60-90 >90-120>120-150 >150
No
. o
f h
ost
s
0
5
10
15
20
25
30
35
40
Dogs
Cats
Figure(1)
Infectiousness to bugs (%)
0-20 21-40 41-60 61-80 81-100
Per
centa
ge
of
host
s
0
10
20
30
40
50
60
Dogs
Cats
Figure(2)
Parasite equivalents per mL
0 50 100 150 200 250
Infe
ctio
usn
ess
to b
ugs
(%)
0
20
40
60
80
100
TcV
TcVI
TcI
TcIII
TcII/V/VI
A
Parasite equivalents per mL
0 40 80 120 160 200 1100
Infe
ctio
usn
ess
to b
ugs
(%)
0
20
40
60
80
100
TcV
TcVI
TcI
B
Figure(3)
Supplementary MaterialClick here to download Supplementary Material: Appendix.docx