Post by Admin/ Traveler on Apr 14, 2019 17:04:31 GMT
Carrion’s disease: more than a neglected disease
"Meritxell Garcia-Quintanilla, Alexander A. Dichter, Humberto Guerra and Volkhard A. J. Kempf
Parasites & Vectors 2019 12:141
Received: 19 December 2018Accepted: 7 March 2019Published: 26 March 2019
Abstract:
Infections with Bartonella bacilliformis result in Carrion’s disease in humans. In the first phase of infection, the pathogen causes a hemolytic fever (“Oroya fever”) with case-fatality rates as high as ~90% in untreated patients, followed by a chronical phase resulting in angiogenic skin lesions (“verruga peruana”). Bartonella bacilliformis is endemic to South American Andean valleys and is transmitted via sand flies (Lutzomyia spp.). Humans are the only known reservoir for this old disease and therefore no animal infection model is available. In the present review, we provide the current knowledge on B. bacilliformis and its pathogenicity factors, vectors, possible unknown reservoirs, established and potential infection models and immunological aspects of the disease.
Background
Carrion’s disease is a vector-borne biphasic illness restricted to the South American Andes including Peru, Ecuador and Colombia and is endemic in Andean valleys at an altitude of 600–3200 m above sea level; it has also been described in the coastal areas of Guayas and Manabi in Ecuador [1, 2]. The causative agent of this neglected disease is Bartonella bacilliformis, which is a motile, aerobic, facultative intracellular alpha-2-proteobacterium. It infects human erythrocytes first causing a serious acute hemolytic anemia called “Oroya fever” followed by a chronic infection of endothelial cells resulting in vasculo-endothelial proliferations called “verruga peruana” as the result of the continuous angiogenic stimulus by B. bacilliformis. These two syndromes typically occur sequentially but sometimes independently. An infection with B. bacilliformis can result in a variety of different clinical manifestations such as severe illness, mild or asymptomatic illness or chronic asymptomatic bacteremia [3]. The exact factors which define the clinical course of Carrion’s disease are still unknown but it is assumed that the interplay of virulence factors of the strain, the inoculum and the fitness and individual predisposition of the host determine the severity of the clinical manifestation [4]. The existence of less virulent bacterial strains that cause mild atypical bartonellosis has been suggested, meaning Carrion’s disease is under-reported [1]. Bartonella bacilliformis is transmitted to humans by female phlebotomine sand flies (Lutzomyia spp.) which are present in high-altitude regions. Climatic changes favor the expansion of B. bacilliformis infections through sand fly proliferation [5, 6].
Oroya fever (characterized by an intraerythrocytic anemia) (Fig. 1) is more common in children than in adults and it is characterized by a plethora of symptoms including fever, hemolytic anemia, pallor, myalgia, headache, anorexia, tachycardia and hepatomegaly [5] with an immune-compromised state that facilitates secondary infections such as Toxoplasma gondii myocarditis or bacteremia with Staphylococcus aureus or Salmonella enterica [4]. In this early phase of infection, B. bacilliformis spreads into the circulatory system invading erythrocytes and leading a hemolytic anemia due to the splenic depletion of infected erythrocytes. Case-fatality rates as high as 88% have been described in the Oroya fever phase in untreated patients, meanwhile around 10% case-fatality rates have been reported for patients receiving timely antibiotic treatment [7].
The life-cycles of Bartonella spp. in their respective vectors are better known for many of the species other than B. bacilliformis. Those studies propose that Bartonella is present in the midgut of arthropod vectors and is released onto the mammalian skin in feces in order to pass to the dermal niche after erosion of the skin. The lymphatic system seems to be responsible for spreading the pathogen into the circulatory system and an intracellular presence of the bacteria (here in erythrocytes) avoids clearance by the host immune system [8, 9]. In the case of B. bacilliformis, it remains unknown if there is a dermal inoculation prior the blood spreading since the only known vectors to date are sand flies (Lutzomyia spp.) which might transmit the bacteria directly into the bloodstream. Moreover, as there are currently no animal infection models, the exact mechanisms underlying the pathobiology of this early infection state cannot be analyzed in detail in an experimental setting.
If Oroya fever is survived, the chronic verruga peruana phase can occur impressing as blood-filled nodular hemangioma-like lesions in the skin (Fig. 2). Under all human pathogenic bacteria, only the family of Bartonella has the ability to trigger angiogenic disease entities (B. bacilliformis: verruga peruana; B. henselae, B. quintana: bacillary angiomatosis, peliosis hepatis [10]). It is suggested that the abnormal endothelial cell proliferation is induced by a chronic Bartonella-infection in which the bacteria are included into vacuoles inside the capillary endothelium. Peruvian warts are mostly found on the head and extremities persisting from weeks to months. These lesions were described in the 16th century by Spanish conquerors.
In general, Carrion’s disease has been only poorly investigated; a PubMed query in December 2018 with the terminus “Bartonella bacilliformis” revealed only 258 publications, many of them from Peru where the pathogen is endemic [in contrast: Staphylococcus aureus, 112,157 publications; Trypanosoma cruzi (endemic in South America), 14,936 publications). The field suffers from a significant lack of data about many aspects of Carrion’s disease, a limited knowledge about confirmed vectors or reservoirs of B. bacilliformis and the absence of feasible animal infection models. The assumed general strategy underlying a Bartonella infection is (i) the avoidance of the host immune response and the infection of a primary niche (if this exists); (ii) the invasion of erythrocytes; and (iii) an intraerythrocytic replication [11] resulting in erythrocyte rupture [12]. Exact mechanisms involved in all these steps are not studied in detail. It is known that flagella of B. bacilliformis are not recognized by Toll-like receptor 5 (TLR5) avoiding a broad activation of the innate immune system [13] and it is assumed that adhesins might mediate autoaggregation [14] to prevent phagocytosis [11]. On the other hand, adhesins, flagellin, hemolysin, deformin or the invasion associate locus proteins A and B are some factors that have been associated with erythrocyte infections. In this review we summarize the current knowledge for B. bacilliformis with regard to vectors, pathogenicity factors and infection models.
Vectors and reservoirs for B. bacilliformis
Sand flies belonging to the genus Lutzomyia (Fig. 4) are considered the only vector for B. bacilliformis. The first evidence for the transmission of B. bacilliformis was found in 1913 when Charles Townsend captured sand flies in the train station where workers suffered from Carrion’s disease [15]. In 1929, the pioneer in analyzing Oroya fever, Hideyo Noguchi, determined which insects are responsible of the transmission of the disease by exposing Macacus rhesus monkeys to bat flies, bedbugs, buffalo gnats, fleas, horse flies, lice, mites, midges, mosquitoes, sheep ticks, ticks, and three species of sand flies (L. verrucarum, L. peruensis and L. noguchii). He injected crushed arthropods intradermally and blood cultures were analyzed for the presence of B. bacilliformis. The only vectors whose injections resulted in an infection were L. verrucarum and L. noguchii [16]. From literature, the following Lutzomyia species are suggested vectors for B. bacilliformis: L. ayacuchensis [2], L. columbiana [17], L. gomezi [17], L. maranonensis [18], L. noguchii [16], L. panamensis [17], L. peruensis [19, 20], L. pescei [5], L. robusta [21], L. serrana [2] and L. verrucarum [22]. However, the presence of B. bacilliformis DNA in these insects has only been demonstrated for L. verrucarum [22], L. peruensis [20], L. robusta [23] and L. maranonensis [18].
(Trav here, I cut out a bunch of technical writing. If you wish to see it, please follow the link at the start of the article)
Host immune response upon B. bacilliformis infections
Only little information exists about immunity in Carrion’s disease and immune response to B. bacilliformis infections. Reasons for this are the low availability of samples from the endemic areas, a hardly existing scientific attention to the disease and the lack of suitable animal infection models. There is moderate evidence that humoral and cellular immune responses are involved during Carrion’s disease. It is known that an infection with B. bacilliformis results in a lifelong humoral immunity which confers partial immunological protection [88] and this is in-line with earlier results showing that rhesus monkeys and chimpanzee which had recovered from an infection with B. bacilliformis showed complete immunity when repetitively infected [81].
Groundbreaking findings from 1929 are still valid today [89]: to study the effects of immune sera on the course of B. bacilliformis infections, rabbit immune sera and convalescent sera from infected rhesus monkeys were tested in infections of rhesus macaques. In most cases, convalescent sera delayed the formation of verruga peruana and inhibited a proliferative blood-stream infection with B. bacilliformis when simultaneously applied with the pathogen. The injection of convalescent sera after B. bacilliformis infections resulted in negative blood cultures but showed no effect on the formation of skin lesions.
In endemic regions, seropositivity (IgM, IgG) of humans can reach ~30–35%. Recent studies reported that the number of asymptomatic B. bacilliformis carriers is ~37% in post-outbreak areas and ~52% in endemic areas [51]. These asymptomatic individuals seem to represent the main reservoir of the pathogen. In an attempt to identify serum biomarkers to detect B. bacilliformis infections it was suggested to consider IgM as a marker of a recent infection and IgG as a marker of past exposure and immunity [88]. It was also shown that IgM levels correlate with low levels of eotaxin, IL-6 and VEGF and high levels of interleukin 10 (IL-10), reflecting an immunosuppression in the acute phase of Oroya fever [88]. IL-10 is a potent anti-inflammatory cytokine that plays a crucial role in limiting the host immune response to pathogens in order to prevent host damage. It was reported that some pathogens are able to utilize the immunosuppressive properties of IL-10 to limit the host immune response [90]. A decrease of the cellular mediated immune response and increased levels of IL-10 were also observed in two pregnant patients that suffered from a severe bartonellosis [91]. It is believed that B. bacilliformis induces a long lasting immunosuppression continuing after the acute phase (Oroya fever) and during the chronic phase of Carrion’s disease [88]. Due to this, levels of TH1-related and pro-inflammatory cytokines are reduced leading to persistent infections characterized by a low level-bacteremia [88]. Furthermore, the proangiogenic cytokines VEGF and eotaxin showed a positive correlation with IgG levels and a negative correlation with IgM levels in seropositive patients [88]. It has been demonstrated that B. henselae induces VEGF production in vitro and in vivo [92, 93]. It is hypothesized that with an enhanced IgG response, B. bacilliformis evades the immune system in endothelial cells to hide and replicate in this immunoprivileged niche [88].
Conclusions
Carrion’s disease is an ancient disease. There is a worrisome lack of knowledge about vectors and possible reservoir hosts of B. bacilliformis. Insights into the dynamics of pathogen transmission by Lutzomyia species might help to gain prevention strategies. Clearly, a rigorous screening of the wildlife (animals and plants) would discard or confirm the existence of other B. bacilliformis reservoir hosts apart from human beings. Molecular mechanisms underlying host infections are also widely unknown. The use of appropriate in vitro and in vivo infection models in combination with molecular strategies using bacterial mutants (e.g. generated by random and targeted mutagenesis) and recombinant protein expression strategies (e.g. via heterologous expression libraries) could help to gain deeper insights into the infection biology of this difficult to handle pathogen and might represent a basis for the development of a potential vaccine.
Notes
Meritxell Garcia-Quintanilla and Alexander A. Dichter contributed equally to this work
Abbreviations
DNA:
deoxyribonucleic acid
GFP:
green-fluorescent protein
HUVEC:
human umbilical vein endothelial cells
IgG:
immunoglobulin G
IgM:
immunoglobulin M
IL-10:
interleukin 10
MLST:
multi-locus sequence typing
mRNA:
messenger ribonucleic acid
NF-κB:
nuclear factor κB
OMP:
outer membrane protein
PCR:
polymerase chain reaction
TAA:
trimeric autotransporter adhesion
TH1:
T helper 1
TLR5:
Toll-like receptor 5
t-PA:
tissue plasminogen activator
T4SS:
type IV secretion system
VEGF:
vascular endothelial growth factor
Declarations
Acknowledgements
The authors thank Birgit Fehrenbacher and Professor Martin Schaller (Department for Dermatology, University Hospital Tübingen, Germany), Jürgen Berger and Dr Katharina Hipp (Max Planck Institute for Developmental Biology, Tübingen, Germany), Professor Martin Aepfelbacher (Medical Microbiology and Hygiene, University Hospital Hamburg Eppendorf, Germany), and Professor Mike Minnick (College of Humanities and Science, Missoula, Montana, USA) for ongoing scientific collaborations. The authors thank especially Professor Eduardo Gotuzzo, Dr Ciro Maguiña, Ms Palmira Ventosilla and Mr Enrique Pérez from the Universidad Peruana Cayetano Heredia and the Instituto de Medicina Tropical Alexander von Humboldt, Lima, Peru for supporting with knowledge and photographs. Publication of this paper has been sponsored by Bayer Animal Health in the framework of the 14th CVBD Word Forum Symposium.
"Meritxell Garcia-Quintanilla, Alexander A. Dichter, Humberto Guerra and Volkhard A. J. Kempf
Parasites & Vectors 2019 12:141
Received: 19 December 2018Accepted: 7 March 2019Published: 26 March 2019
Abstract:
Infections with Bartonella bacilliformis result in Carrion’s disease in humans. In the first phase of infection, the pathogen causes a hemolytic fever (“Oroya fever”) with case-fatality rates as high as ~90% in untreated patients, followed by a chronical phase resulting in angiogenic skin lesions (“verruga peruana”). Bartonella bacilliformis is endemic to South American Andean valleys and is transmitted via sand flies (Lutzomyia spp.). Humans are the only known reservoir for this old disease and therefore no animal infection model is available. In the present review, we provide the current knowledge on B. bacilliformis and its pathogenicity factors, vectors, possible unknown reservoirs, established and potential infection models and immunological aspects of the disease.
Background
Carrion’s disease is a vector-borne biphasic illness restricted to the South American Andes including Peru, Ecuador and Colombia and is endemic in Andean valleys at an altitude of 600–3200 m above sea level; it has also been described in the coastal areas of Guayas and Manabi in Ecuador [1, 2]. The causative agent of this neglected disease is Bartonella bacilliformis, which is a motile, aerobic, facultative intracellular alpha-2-proteobacterium. It infects human erythrocytes first causing a serious acute hemolytic anemia called “Oroya fever” followed by a chronic infection of endothelial cells resulting in vasculo-endothelial proliferations called “verruga peruana” as the result of the continuous angiogenic stimulus by B. bacilliformis. These two syndromes typically occur sequentially but sometimes independently. An infection with B. bacilliformis can result in a variety of different clinical manifestations such as severe illness, mild or asymptomatic illness or chronic asymptomatic bacteremia [3]. The exact factors which define the clinical course of Carrion’s disease are still unknown but it is assumed that the interplay of virulence factors of the strain, the inoculum and the fitness and individual predisposition of the host determine the severity of the clinical manifestation [4]. The existence of less virulent bacterial strains that cause mild atypical bartonellosis has been suggested, meaning Carrion’s disease is under-reported [1]. Bartonella bacilliformis is transmitted to humans by female phlebotomine sand flies (Lutzomyia spp.) which are present in high-altitude regions. Climatic changes favor the expansion of B. bacilliformis infections through sand fly proliferation [5, 6].
Oroya fever (characterized by an intraerythrocytic anemia) (Fig. 1) is more common in children than in adults and it is characterized by a plethora of symptoms including fever, hemolytic anemia, pallor, myalgia, headache, anorexia, tachycardia and hepatomegaly [5] with an immune-compromised state that facilitates secondary infections such as Toxoplasma gondii myocarditis or bacteremia with Staphylococcus aureus or Salmonella enterica [4]. In this early phase of infection, B. bacilliformis spreads into the circulatory system invading erythrocytes and leading a hemolytic anemia due to the splenic depletion of infected erythrocytes. Case-fatality rates as high as 88% have been described in the Oroya fever phase in untreated patients, meanwhile around 10% case-fatality rates have been reported for patients receiving timely antibiotic treatment [7].
The life-cycles of Bartonella spp. in their respective vectors are better known for many of the species other than B. bacilliformis. Those studies propose that Bartonella is present in the midgut of arthropod vectors and is released onto the mammalian skin in feces in order to pass to the dermal niche after erosion of the skin. The lymphatic system seems to be responsible for spreading the pathogen into the circulatory system and an intracellular presence of the bacteria (here in erythrocytes) avoids clearance by the host immune system [8, 9]. In the case of B. bacilliformis, it remains unknown if there is a dermal inoculation prior the blood spreading since the only known vectors to date are sand flies (Lutzomyia spp.) which might transmit the bacteria directly into the bloodstream. Moreover, as there are currently no animal infection models, the exact mechanisms underlying the pathobiology of this early infection state cannot be analyzed in detail in an experimental setting.
If Oroya fever is survived, the chronic verruga peruana phase can occur impressing as blood-filled nodular hemangioma-like lesions in the skin (Fig. 2). Under all human pathogenic bacteria, only the family of Bartonella has the ability to trigger angiogenic disease entities (B. bacilliformis: verruga peruana; B. henselae, B. quintana: bacillary angiomatosis, peliosis hepatis [10]). It is suggested that the abnormal endothelial cell proliferation is induced by a chronic Bartonella-infection in which the bacteria are included into vacuoles inside the capillary endothelium. Peruvian warts are mostly found on the head and extremities persisting from weeks to months. These lesions were described in the 16th century by Spanish conquerors.
In general, Carrion’s disease has been only poorly investigated; a PubMed query in December 2018 with the terminus “Bartonella bacilliformis” revealed only 258 publications, many of them from Peru where the pathogen is endemic [in contrast: Staphylococcus aureus, 112,157 publications; Trypanosoma cruzi (endemic in South America), 14,936 publications). The field suffers from a significant lack of data about many aspects of Carrion’s disease, a limited knowledge about confirmed vectors or reservoirs of B. bacilliformis and the absence of feasible animal infection models. The assumed general strategy underlying a Bartonella infection is (i) the avoidance of the host immune response and the infection of a primary niche (if this exists); (ii) the invasion of erythrocytes; and (iii) an intraerythrocytic replication [11] resulting in erythrocyte rupture [12]. Exact mechanisms involved in all these steps are not studied in detail. It is known that flagella of B. bacilliformis are not recognized by Toll-like receptor 5 (TLR5) avoiding a broad activation of the innate immune system [13] and it is assumed that adhesins might mediate autoaggregation [14] to prevent phagocytosis [11]. On the other hand, adhesins, flagellin, hemolysin, deformin or the invasion associate locus proteins A and B are some factors that have been associated with erythrocyte infections. In this review we summarize the current knowledge for B. bacilliformis with regard to vectors, pathogenicity factors and infection models.
Vectors and reservoirs for B. bacilliformis
Sand flies belonging to the genus Lutzomyia (Fig. 4) are considered the only vector for B. bacilliformis. The first evidence for the transmission of B. bacilliformis was found in 1913 when Charles Townsend captured sand flies in the train station where workers suffered from Carrion’s disease [15]. In 1929, the pioneer in analyzing Oroya fever, Hideyo Noguchi, determined which insects are responsible of the transmission of the disease by exposing Macacus rhesus monkeys to bat flies, bedbugs, buffalo gnats, fleas, horse flies, lice, mites, midges, mosquitoes, sheep ticks, ticks, and three species of sand flies (L. verrucarum, L. peruensis and L. noguchii). He injected crushed arthropods intradermally and blood cultures were analyzed for the presence of B. bacilliformis. The only vectors whose injections resulted in an infection were L. verrucarum and L. noguchii [16]. From literature, the following Lutzomyia species are suggested vectors for B. bacilliformis: L. ayacuchensis [2], L. columbiana [17], L. gomezi [17], L. maranonensis [18], L. noguchii [16], L. panamensis [17], L. peruensis [19, 20], L. pescei [5], L. robusta [21], L. serrana [2] and L. verrucarum [22]. However, the presence of B. bacilliformis DNA in these insects has only been demonstrated for L. verrucarum [22], L. peruensis [20], L. robusta [23] and L. maranonensis [18].
(Trav here, I cut out a bunch of technical writing. If you wish to see it, please follow the link at the start of the article)
Host immune response upon B. bacilliformis infections
Only little information exists about immunity in Carrion’s disease and immune response to B. bacilliformis infections. Reasons for this are the low availability of samples from the endemic areas, a hardly existing scientific attention to the disease and the lack of suitable animal infection models. There is moderate evidence that humoral and cellular immune responses are involved during Carrion’s disease. It is known that an infection with B. bacilliformis results in a lifelong humoral immunity which confers partial immunological protection [88] and this is in-line with earlier results showing that rhesus monkeys and chimpanzee which had recovered from an infection with B. bacilliformis showed complete immunity when repetitively infected [81].
Groundbreaking findings from 1929 are still valid today [89]: to study the effects of immune sera on the course of B. bacilliformis infections, rabbit immune sera and convalescent sera from infected rhesus monkeys were tested in infections of rhesus macaques. In most cases, convalescent sera delayed the formation of verruga peruana and inhibited a proliferative blood-stream infection with B. bacilliformis when simultaneously applied with the pathogen. The injection of convalescent sera after B. bacilliformis infections resulted in negative blood cultures but showed no effect on the formation of skin lesions.
In endemic regions, seropositivity (IgM, IgG) of humans can reach ~30–35%. Recent studies reported that the number of asymptomatic B. bacilliformis carriers is ~37% in post-outbreak areas and ~52% in endemic areas [51]. These asymptomatic individuals seem to represent the main reservoir of the pathogen. In an attempt to identify serum biomarkers to detect B. bacilliformis infections it was suggested to consider IgM as a marker of a recent infection and IgG as a marker of past exposure and immunity [88]. It was also shown that IgM levels correlate with low levels of eotaxin, IL-6 and VEGF and high levels of interleukin 10 (IL-10), reflecting an immunosuppression in the acute phase of Oroya fever [88]. IL-10 is a potent anti-inflammatory cytokine that plays a crucial role in limiting the host immune response to pathogens in order to prevent host damage. It was reported that some pathogens are able to utilize the immunosuppressive properties of IL-10 to limit the host immune response [90]. A decrease of the cellular mediated immune response and increased levels of IL-10 were also observed in two pregnant patients that suffered from a severe bartonellosis [91]. It is believed that B. bacilliformis induces a long lasting immunosuppression continuing after the acute phase (Oroya fever) and during the chronic phase of Carrion’s disease [88]. Due to this, levels of TH1-related and pro-inflammatory cytokines are reduced leading to persistent infections characterized by a low level-bacteremia [88]. Furthermore, the proangiogenic cytokines VEGF and eotaxin showed a positive correlation with IgG levels and a negative correlation with IgM levels in seropositive patients [88]. It has been demonstrated that B. henselae induces VEGF production in vitro and in vivo [92, 93]. It is hypothesized that with an enhanced IgG response, B. bacilliformis evades the immune system in endothelial cells to hide and replicate in this immunoprivileged niche [88].
Conclusions
Carrion’s disease is an ancient disease. There is a worrisome lack of knowledge about vectors and possible reservoir hosts of B. bacilliformis. Insights into the dynamics of pathogen transmission by Lutzomyia species might help to gain prevention strategies. Clearly, a rigorous screening of the wildlife (animals and plants) would discard or confirm the existence of other B. bacilliformis reservoir hosts apart from human beings. Molecular mechanisms underlying host infections are also widely unknown. The use of appropriate in vitro and in vivo infection models in combination with molecular strategies using bacterial mutants (e.g. generated by random and targeted mutagenesis) and recombinant protein expression strategies (e.g. via heterologous expression libraries) could help to gain deeper insights into the infection biology of this difficult to handle pathogen and might represent a basis for the development of a potential vaccine.
Notes
Meritxell Garcia-Quintanilla and Alexander A. Dichter contributed equally to this work
Abbreviations
DNA:
deoxyribonucleic acid
GFP:
green-fluorescent protein
HUVEC:
human umbilical vein endothelial cells
IgG:
immunoglobulin G
IgM:
immunoglobulin M
IL-10:
interleukin 10
MLST:
multi-locus sequence typing
mRNA:
messenger ribonucleic acid
NF-κB:
nuclear factor κB
OMP:
outer membrane protein
PCR:
polymerase chain reaction
TAA:
trimeric autotransporter adhesion
TH1:
T helper 1
TLR5:
Toll-like receptor 5
t-PA:
tissue plasminogen activator
T4SS:
type IV secretion system
VEGF:
vascular endothelial growth factor
Declarations
Acknowledgements
The authors thank Birgit Fehrenbacher and Professor Martin Schaller (Department for Dermatology, University Hospital Tübingen, Germany), Jürgen Berger and Dr Katharina Hipp (Max Planck Institute for Developmental Biology, Tübingen, Germany), Professor Martin Aepfelbacher (Medical Microbiology and Hygiene, University Hospital Hamburg Eppendorf, Germany), and Professor Mike Minnick (College of Humanities and Science, Missoula, Montana, USA) for ongoing scientific collaborations. The authors thank especially Professor Eduardo Gotuzzo, Dr Ciro Maguiña, Ms Palmira Ventosilla and Mr Enrique Pérez from the Universidad Peruana Cayetano Heredia and the Instituto de Medicina Tropical Alexander von Humboldt, Lima, Peru for supporting with knowledge and photographs. Publication of this paper has been sponsored by Bayer Animal Health in the framework of the 14th CVBD Word Forum Symposium.