Surveillance and genotype characterization of zoonotic trypanosomatidae in Didelphis marsupialis in two endemic sites of rural Panama.

Didelphis marsupialis has been reported as a competent reservoir for trypanosomatid parasites infections. The aim of this study was to measure Trypanosoma cruzi, T. rangeli, and Leishmania spp. infection rates and to characterize discrete typing units (DTUs) of T. cruzi in D. marsupialis from two Chagas disease endemic sites in Panama. Blood from 57 wild-caught D. marsupialis were examined from two rural communities, Las Pavas (N = 18) and Trinidad de las Minas (N = 39). Twenty-two (38.60%) opossums were positive for flagellates by general hemoculture. T. cruzi infection was confirmed by positive hemoculture and/or kDNA based PCR performed in 31/57 (54.39%) blood samples from opossums. T. rangeli infection was confirmed by hemoculture and/or TrF/R2-Primer PCR assay applied on 12/57 (21.05%) blood samples. Nine (15.79%) D. marsupialis harbored T. cruzi/T. rangeli coinfections. All opossums tested negative for Leishmania spp. by PCR assays based on kDNA and HSP70 gene amplification. There was a significant association between T. cruzi infection and site (Fisher exact test, p = 0.02), with a higher proportion of T. cruzi infected opossums in Las Pavas (77.78%, n = 14/18) compared to Trinidad de las Minas (43.59%, n = 17/39). A significant association was found between habitat type and T. cruzi infection in opossums across both communities, (X2 = 6.91, p = 0.01, df = 1), with a higher proportion of T. cruzi infection in opossums captured in forest remnants (76%, 19/25) compared to peridomestic areas (37.5%, 12/32). T. rangeli detection, but not T. cruzi detection, may be improved by culture followed by PCR. TcI was the only DTU detected in 22 T. cruzi samples using conventional and real-time PCR. Eight T. rangeli positive samples were characterized as KP1(-)/lineage C. Trypanosome infection data from this common synanthropic mammal provides important information for improved surveillance and management of Chagas disease in endemic regions of Panama.

Introduction

Chagas disease and American cutaneous leishmaniasis, caused by vector-borne parasites in the family Trypanosomatidae, pose a significant health burden to many people throughout the Americas (de Lima et al., 2006). The protozoan parasite Trypanosoma cruzi, cause of Chagas disease, infects approximately 7 million people worldwide and cycles between many wild and domestic mammalian reservoir host species and triatomine vectors (Saldaña et al., 2005; WHO, 2021a). Based on genetic and biological characteristics, T. cruzi isolates have been divided into six discrete typing units (DTUs): TcI, TcII, TcIII, TcIV, TcV and TcVI (Zingales et al., 2012) and an additional bat-associated genotype TcBat (Marcili et al., 2009). All DTUs are infectious to humans with mounting evidence that different DTUs have distinct clinical presentations (Zingales, 2018). DTUs display genetic and biogeographic diversity, nevertheless, the association of these subpopulations with biological and epidemiological characteristics is not clearly defined (Izeta-Alberdi et al., 2016; Jansen et al., 2020). Identifying T. cruzi in wild reservoirs and their DTU/reservoir host relationships contribute to a better understanding of T. cruzi ecology and transmission patterns within a given geographic area (Jansen et al., 2017).

Originally, T. cruzi was limited to wild mammal reservoirs, but human settlements led to establishments and overlap of sylvatic and domestic mammal transmission cycle (Herrera and Urdaneta-Morales, 1992; Jansen et al., 2017). Throughout the Americas, natural T. cruzi infection has been identified in more than 180 species of mammals (Sousa, 1972; Noireau et al., 2009; Rodriguez and Loaiza, 2017). Across its range, the common opossum Didelphis marsupialis, an opportunistic mammal who adapts well to anthropogenic landscapes, is a reservoir host in the maintenance of T. cruzi circulation, serving as a link between wild and domestic cycle (Jansen et al., 2017; Pinho et al., 2000). Opossums acquire T. cruzi infection by feeding on triatomines or infected small mammals, mucous membrane triatomine bite wound contamination with infected triatomine feces, or contact with secretions of the anal glands of other infected opossums (Jansen et al., 2018). The main vector of Chagas disease in Panama, Rhodnius pallescens, frequently feeds from opossum blood (Pineda et al., 2008; Saldaña et al., 2012). Trypanosoma rangeli, which is also transmitted by R. pallescens, is not pathogenic in mammals, including humans, commonly co-circulates with T. cruzi in rural Panama, and may cause cross-reactivity with T. cruzi in some immunodiagnostic tests (Saldaña et al., 2005). In Panama, human infection with T. rangeli is up to 10 times more frequent than with T. cruzi (Sousa, 1972), thus justifying the search and characterization of this generalist hemoflagellate. In addition to its role as a T. cruzi reservoir, D. marsupialis can also play an important role in Leishmania spp. transmission (Travi et al., 1994, 1998b). Leishmania spp. are transmitted from wild mammal reservoirs to humans by phlebotomine sandfly vector bites (WHO, 2021b). Leishmania infections have been detected in D. marsupialis in Brazil (Cabrera et al., 2003; Schallig et al., 2007), and Colombia (Corredor et al., 1989; Travi et al., 1994, 1998a). L. mexicana, L. infantum, and L. chagasi infections, including co-infections with T. cruzi, have been detected in D. marsupialis in Venezuela (Viettri et al., 2018, 2019). Arguably, the role of D. marsupialis in Leishmania transmission is less well understood compared to its role in T. cruzi transmission throughout its range.

Although observations suggest D. marsupialis plays an important role in Chagas disease epizootiology in Panama, the last report being from the 1970s (Sousa, 1972), contemporary information regarding zoonotic trypanosomatidae infection prevalence and genotype in D. marsupialis across endemic rural landscapes in Panama is lacking. Accordingly, this study aims to evaluate the frequency of Trypanosomatid infection and the detection of T. cruzi DTUs in common opossums from sites near two rural communities in central Panama.

Materials and methods

Study area

We carried out a descriptive, cross-sectional study in peridomestic areas and forest remnants near communities of Las Pavas (LP), District of La Chorrera (9°6′15″N, 79°53′9″W, 50–156 m above sea level, near the west bank of the Panama Canal), and further west in Trinidad de Las Minas (TM), District of Capira (8° 46′32″N, 79° 59′45″W), 230 m above sea level. The potential vegetation at both sites is tropical rainforest (Holdridge, 1967), characterized by a mosaic of forest patches, riparian forest remnants, cattle pasture, regenerating forest, and human settlements. There is a marked dry season mid-December to March and a rainy season throughout the rest of the year.

Capture and sampling

For 5 consecutive nights (640 trap nights) during wet (November) and dry seasons (March) 2013-2015, D. marsupialis (N = 57) were captured using 32 Tomahawk traps distributed in four 150 m long transects separated by 50 m within each site (2 forest remnants located 1 and 3 km from each community, and a peridomestic site 300 m around houses), Fig. 1. Captured animals were anesthetized with Ketamine (5–7.5 mg/kg) and 500 μl-3 ml blood was drawn from the caudal vein of each opossum and placed in microtubes with EDTA.

Fig. 1

Map showing the communities of Las Pavas (LP) (top set of images) and Trinidad de Las Minas (TM) (bottom set of images) with the number of opossums captured and infected with T. cruzi in the 3 collection sites in each community. A. Map with the geographic location of the LP and TM communities in the country of Panama. Satellite view of the P: Peridomicile (B), R1: remnant 1 (C) and R2: remnant 2 (D) collection site each with its 4 transects in the LP community. Satellite view of the P: Peridomicile (E), R1: remnant 1 (F) and R2: remnant 2 (G) collection site each with its 4 transects in the TM community.

Trypanosomatid diagnostics

Blood was cultured for trypanosomatids (Vásquez et al., 1997). DNA of blood samples collected at time of capture and blood samples cultured for six weeks after collection were extracted using a commercial kit (QIAamp® DNA Blood Mini Kit (Quiagen). PCR were performed using S35/S36 primers that amplify a 330 bp segment of the T. cruzi variable region minicircle and the primer pairs TrF/R2 that amplify a 620 bp fragment of T. rangeli snoRNA-c11 gene (Vallejo et al., 1999; Pavia et al., 2007). T. cruzi and T. rangeli positive samples were typed by a miniexon gene-based approach (Fernandes et al., 2001) and a PCR-RFLP approach targeting the COII gene (de Sá et al., 2013), respectively. T. cruzi DTUs were detected by a real time PCR strategy based on the amplification of a set of the specific markers SL-IR (TcI-TcIII), COII (TcII-TcIV), ND1 (TcV) and 18S rDNA (TcVI) that in combination discriminate all T. cruzi DTUs (Muñoz-San Martín et al., 2017). DNA samples from opossum blood were also tested for Leishmania spp. using a B1/B2 primers-PCR based assay that amplify a 750 bp kinetoplastic segment specific for Leishmania Viannia (Vergel et al., 2005) along with a PCR approach based on the amplification of a 1230 bp segment of the Leishmania sp. HSP70 gene (Montalvo et al., 2012).

Statistical analysis

RStudio (RStudio Team version 1.31, 2020). RStudio: Integrated Development for R. RStudio, PBC, Boston, MA URL http://www.rstudio.com/) was used for statistical analyses. Chi square or Fisher's exact test were used to evaluate any associations between T. cruzi infection, T. rangeli infection, T. cruzi/T. rangeli coinfection and site (TM vs LP), habitat type, season, year-season, and opossum sex. Basic descriptive and bivariate analyses were performed in the MASS package in R (and tables calculating the proportion of opossums infected with 95% confidence intervals using confint function (Venables and Ripley, 2004). We used the package ‘fsmb’ version 0.7.1 (Minakawa, 2007) for diagnostic agreement analysis (Cohen's kappa test).

Ethics

The Institutional Animal Care and Use Committee (014/CIUCAL-ICGES/13) of the Instituto Conmemorativo Gorgas de Estudios de la Salud (ICGES) approved this study, which was done in accordance with Law No. 23 of 15 January 1997 (Animal Welfare Assurance) of the Republic of Panama.

Results

The presence of T. cruzi and T. rangeli were evaluated in blood samples from 57 captured D. marsupialis (31 females, 25 males, and one unidentified sex) by PCR of DNA directly isolated from blood and by PCR performed on DNA samples obtained after 6 week of blood culture. Trypanosomes were detected in 35.09% (20/57, 95% CI; 22.91%, 48.87%) opossums by hemoculture.

Overall T. cruzi positivity in opossums (including individuals as positive who tested positive from non-cultured and cultured blood) was 54.39% (31/57, 95% CI; 41.59%,66.63%). Overall T. rangeli positivity in opossums was 21.05% (12/57, 95% CI; 12.47%,33.29%). T. cruzi/T. rangeli coinfections were detected in 15.79% (9/57, 95% CI; 15.79%,27.36%) of opossums. Table 1 shows PCR results for T. cruzi and T. rangeli infection in opossums from each community.

Table 1

Trypanosoma cruzi and T. rangeli infections in Didelphis marsupialis from two Chagas disease endemic sites in Panama. Data show the following: infected/total number sampled, percent infected (%), and upper and lower bounds of the 95% confidence interval for infection.

Site	T. cruzi			T. rangeli			Coinfection
	PD	Forest Remnant	Total	PD	Forest Remnant	Total	PD	Forest Remnant	Total
Las Pavas (LP)	3/5	11/13	14/18	3/5	2/13	5/18	2/5	2/13	4/18
	60.00	84.61	77.78	60.00	15.38	27.78	40.00	15.38	22.22
	23.07-	57.76-	54.78-	23.07-	4.33-	12.50-	11.76-	4.33-	9.00-
	88.20	95.67	91.0	88.24	42.23	50.87	76.93	42.23	45.21
Trinidad de las Minas (TM)	9/27	8/12	17/39	4/27	3/12	7/39	2/27	3/12	5/39
	33.33	66.67	43.59	14.81	25.00	0.1795	74.07	25.00	12.82
	18.64-	39.06-	29.30-	5.92-	8.89-	8.98-	2.05-	8.89-	5.60-
	52.17	86.19	59.02	32.48	53.23	32.67	23.37	53.23	26.71
Total	12/32	19/25	31/57	7/32	5/25	12/57	4/32	5/25	9/57
	37.50	76.00	54.39	21.87	20.00	21.05	12.50	20.00	15.79
	22.93-	56.57-	41.59-	11.02-	8.86-	12.47-	4.97-	8.86-	8.54-
	54.75	88.50	66.63	38.75	39.13	33.29	28.07	39.13	27.36

There was a significant association between T. cruzi infection and community (X-squared = 4.51, df = 1, p-value = 0.03), with a higher proportion of T. cruzi infected opossums in LP (77.78%) than TM (43.59%). Across both communities, there was a significant association between habitat type and T. cruzi infection (Chi squared = 6.91, df = 1, p = 0.009) with a higher proportion of T. cruzi-infected opossums in forest remnants (19/25, 76%) compared to peridomiciliary sites (12/32, 37.50%). There was no significant association between community and T. rangeli infection (Fisher's exact test, p = 0.49) or T. cruzi-T. rangeli coinfection (Fisher's exact test, p = 0.44). Although there was a greater number of opossums captured at both sites in the wet season (N = 46) compared to the dry season (N = 11), there was no significant association between season of capture (wet vs dry) and T. cruzi infection (Fisher's exact test p-value = 0.52) or T. rangeli infection (Fisher's exact test p-value = 0.52), nor T. cruzi-T. rangeli coinfection (Fisher's exact test, p = 0.67). There was no significant association between opossum sex and T. cruzi infection (X-squared = 1.6, df = 1, p-value = 0.2059), T. rangeli infection (X-squared = 7.2547e-32, df = 1, p-value = 1), and T. cruzi/T. rangeli coinfection (Fisher's exact test, p = 1).

For T. cruzi, there was 70.9% crude agreement between PCR of cultured and non-cultured blood and a Cohen's Kappa of 0.415 (95% CI; 0.173, 0.656, Z = 3.06, p = 0.001), indicating moderate agreement. For T. rangeli detection, there was 81.18% crude agreement between cultured and non-cultured PCR diagnostic tests, and a Cohen's kappa of 0.068, 95% CI; −0.455, 0.590, Z = 0.247, p = 0.402, indicating slight agreement between the two methods. For coinfection, although there was 87% crude agreement between non cultured- and cultured blood PCR tests, there was slight agreement between the two methods (Cohen's kappa = 0.112, 95% CI; −0.421,0.689, Z = 0.446, p = 0.33).

Fifteen out of 31 T. cruzi positive samples by PCR were characterized as DTU 1 (8 from LP and 7 from TM). Nineteen out of 31 T. cruzi positive samples were successfully characterized as DTU 1 by real time PCR (8 from LP and 11 from TM). All T. rangeli positive samples (N = 8) were successfully characterized as a lineage corresponded to and KP1(-)/lineage C. All 57 samples tested negative for Leishmania spp. by both the PCR targeting kinetoplastic DNA and the PCR assay based on the amplification of HSP70 gene.

Discussion

Didelphis opossums in our study are frequently infected with T. cruzi and T. rangeli in rural habitats in central Panama, with a higher percentage of T. cruzi-infected opossums (54.38%) than T. rangeli-infected opossums (21.05%). In studies from central Panama in 1932, T. cruzi was detected by parasitological methods in 24.6% of D. marsupialis (N = 81) (Clark and Dunn, 1932). In later studies in central Panama (early 1970's), trypanosome infection was detected by hemoculture in over 50% of D. marsupialis, with T. cruzi identified in 20% and T. rangeli in 28% of animals (Sousa, 1972). In our study, the higher T. cruzi infection rate in opossums compared to previous reports may be a true increase in prevalence, or due to improved molecular detection methods. Molecular assays are highly effective and sensitive for the detection of T. rangeli infection (Vallejo et al., 1999; Pavia et al., 2007; de Sá et al., 2013). In addition, these methods allow the evaluation of the genetic diversity of parasite populations, thus enabling a better understanding of the infection behavior among a given reservoir population.

Diagnostic agreement results between different methods (PCR direct from DNA extracted from blood and PCR from DNA extracted after hemoculture) suggest that T. rangeli detection, but not T. cruzi detection, may be improved by culture followed by PCR.

Throughout Latin America, T. cruzi infection rates in D. marsupialis range between 3% to over 80% (Herrera and Urdaneta-Morales, 1992; Grisard et al., 2000; Pinto et al., 2006; Rodriguez-Mongui et al., 2019; Magalhães et al., 2021). T. cruzi infection rates in our study are similar to opossums from rural landscapes of Costa Rica (Zeledón et al., 1970), the Brazilian Amazon (Roque et al., 2013; Magalhães et al., 2021), Colombia (Cantillo-Barraza et al., 2015, 2020; Rodriguez-Mongui et al., 2019) and Venezuela (Herrera and Urdaneta-Morales, 1992; de Lima et al., 2006).

Although no opossums were positive for Leishmania spp. infection by PCR, this does not mean that they do not play a role in Leishmania spp. transmission in the study area. Ardila et al. (2019) sampled 65 tissues (ear and tail biopsies) and blood from 19 D. marsupialis in a Leishmania endemic area of Colombia and did not detect Leishmania spp. (Ardila et al., 2019). Leishmania may not have been detected due to chance, and/or because low prevalence in opossums may require higher sample size/sample effort. Sample type may also play a role, as Leishmania DNA may be present in other tissues, but not in leukocytes in the peripheral blood sample. Also, Leishmania spp. infection may be rare or absent due to sandfly feeding preferences or low opossum-sandfly contact in the study area. Improved detection in future studies could include xenodiagnostics, serotesting for Leishmania exposure, multiannual longitudinal surveys, or more invasive sampling of additional tissues such as skin biopsies. There were significant differences in T. cruzi infection across community and habitat type. D. marsupialis abundance is favored by a greater number of forested patches, conserved areas, agricultural crops, which could explain the higher opossum trap success in the more forested TM compared to LP, where cattle pastures are extensive and forest remnants are smaller and irregularly distributed. Although it is more deforested, LP has a higher relative abundance of Attalea palms where R. pallescens vectors live. Opossums may occupy Attalea palms at a higher frequency in LP compared to TM, resulting in higher opossum-vector contact in LP, because in this pasture-dominated landscape, there are less alternative nesting and hiding places for opossums. The higher T. cruzi infection rates in opossums in forest remnants (Table 1) correlate with studies that found higher infection rates of T. cruzi in R. pallescens vectors in forest remnants compared to peridomiciliary sites (Gottdenker et al., 2012), suggesting more T. cruzi-triatomine contact occurs with opossums and possibly other competent reservoirs in these remnants (Travi et al., 1994; Ruiz-Pina and Cruz-Reyes, 2002; Roque et al., 2013; Orozco et al., 2013; Jansen et al., 2017; Rodriguez-Mongui et al., 2019).

Lower T. rangeli compared to T. cruzi infection in opossums correlates with previous studies in the vector R. pallescens describing a higher proportion of T. cruzi-infected vectors compared to T. rangeli, and similar T. cruzi-T. rangeli co-infection rates across habitat types (Gottdenker et al., 2016). Interestingly, studies from the same geographic region of LP show T. rangeli infection is relatively common in children (Saldaña et al., 2005). In this regard, little is known about the immune response induced by a T. rangeli infection, however it has been reported that immunization of dogs with T. rangeli antigens induces protection against Chagas infection (Basso et al., 2016). The search and genetic characterization of T. rangeli in other wild reservoirs is a requirement for understanding the eco-biology of this parasite.

Only TcI was detected in D. marsupialis in central Panama, corresponding to studies in other countries where TcI predominates in D. marsupialis (Legey et al., 2003; Jansen et al., 2017). Although T. cruzi circulates widely in Chagas disease endemic areas in Panama (Sousa et al., 2006; Samudio et al., 2007; Brandao et al., 2008; Saldaña et al., 2012), TcI is the predominant DTU reported to date in acute chagasic patients, and in triatomine species R. pallescens and Triatoma dimidiata. These results, the previous demonstration of TcI in both patients and triatomines (Sousa et al., 2006; Samudio et al., 2007; Prescilla-Ledezma et al., 2021), and the fact that principal vectors of T. cruzi in Panama frequently feed on opossum blood (Pineda et al., 2008), show that D. marsupialis is an important reservoir in local T. cruzi transmission cycles. However, in Panama as in other regions, genetic heterogeneity present in TcI parasites of opossums as well as the presence of mixed trypanosome infections and DTU characterization in other wild mammalian reservoirs are critical to our understanding of T. cruzi infection ecology (Cura et al., 2010; Jansen et al., 2018; Prescilla-Ledezma et al., 2021). In our study, the KP1(-)/lineage C was the only genetic group found in eight T. rangeli positive samples, agreeing with the genotype reported so far in Panama (Salazar-Antón et al., 2009; Saldaña et al., 2018).

To summarize, this study contributes to understanding of parasite, host, and environmental relationships of Chagas and Leishmania infection across central Panama, allowing for comparison of wildlife infection rates across wider geographic areas and landscapes through the Americas.

Ethical statement

This study was approved by the institutional animal care and use committee: 014/CIUCAL-ICGES/13.

Funding

This research was funded by , grant number Project COL11-043.

Declaration of competing interest

The authors declare that there are no conflicts of interest.