Background

Leishmania infantum (Kinetoplastida, Trypanosomatidae) and Anaplasma platys (Rickettsiales, Anaplasmataceae) are the causative agents of canine leishmaniosis (CanL) and canine infectious cyclic thrombocytopenia (CICT), respectively [1]. While L. infantum is primarily transmitted by phlebotomine sand flies of the genera Phlebotomus and Lutzomyia (Diptera, Psychodidae) [2], alternative modes of transmission have been investigated. These include blood transfusions [3, 4], as well as sexual [5, 6] and vertical transmission [7–12]. In particular, the latter has been hypothesized to occur in naturally infected dogs, resulting in both stillborn and live puppies [7–10]. Vertical transmission of L. infantum has been suggested to represent an important mechanism for the dissemination and maintenance of this pathogen, especially in the (apparent) absence of known competent vectors in non-endemic areas [10, 12–14]. For instance, a high prevalence of infection (i.e. 20 %) has been recorded in dogs from an area of the United States where the presence of phlebotomine sand fly vectors is yet to be confirmed [9, 10]. Rhipicephalus sanguineus (sensu lato) (Acari, Ixodidae), also referred to as the ‘brown dog tick’ or ‘kennel tick’, has been indicated as a potential vector of A. platys, mainly as a consequence of the detection of bacterial DNA in this arthropod species [15–18]. However, the role of ticks in the transmission of this bacterium has yet to be fully proven. Similarly, and to the best of our knowledge, thus far, no studies have assessed the potential occurrence of vertical transmission of A. platys in dogs, or of vertical transmission of L. infantum during the first half of gestation. Here, we explored the potential occurrence of vertical transmission of L. infantum and A. platys in naturally infected dogs, during early gestation, from a hyperendemic area for both pathogens.

Methods

Ethics statement

Animals involved in this study were enrolled in a sterilization program in 2008 and 2011 and were part of a previous study for the prevention and control of canine vector borne diseases [19]. The study above was carried out according to the guidelines for Good Clinical Practice (GCP GL9 VICH, 2000) and authorized by the Italian Ministry of Health (authorization number 72/2009C, n°. 69062; 11/28/2008).

Animals

All animals enrolled in the present study were housed in a private dog shelter in Putignano, province of Bari, Apulia region, Italy, where the incidence of L. infantum (i.e. 47.6 %) and A. platys (i.e. 17.2 %) infections have previously been determined [19–21]. Twenty asymptomatic mixed-breed bitches were screened for infection by L. infantum and A. platys by whole blood and bone marrow cytological examination. In order to confirm the infection status, dogs were also tested by indirect immunofluorescent antibody test (IFAT) on whole blood (L. infantum) and pathogen DNA amplification from whole blood, bone marrow and skin samples (both L. infantum and A. platys) (Table 1). All bitches were subjected to clinical and abdominal ultrasound examinations for verification of the pregnancy status, followed by an ovariohysterectomy. Ten dogs (i.e. nine positive and one negative control) (aged from 7 months to 1 year old) were from 25 to 35 days pregnant (Table 1). Following surgery, two foetuses were randomly collected from the uteri of each pregnant bitch (Table 2) and foetal crown-rump length was measured to confirm their gestational age [22]. The remaining (age-matched) 10 dogs (i.e. nine positive and one negative control) enrolled in the study were non-pregnant bitches (Table 1).

Table 1 Distribution of Leishmania infantum and Anaplasma platys infections in different tissue samples from pregnant and non-pregnant bitches (Pos/Tot, %)
Full size table
Table 2 Distribution of Leishmania infantum and Anaplasma platys infection in foetal tissue from naturally infected pregnant bitches. Starting Quantity (SQ) mean value of parasite load per PCR reaction
Full size table

Sample collection and diagnostic procedures

At the time of ovariohysterectomy, amniotic fluid and tissues samples (n = 210) including whole blood, skin, uterus, ovaries, placenta, liver, spleen, umbilical cord and bone marrow were collected from individual bitches and foetuses, in sterile conditions, and stored at -20 °C until extraction of genomic DNA (Tables 1 and 2). Whole blood samples (4 ml) were collected from the brachial or jugular veins of bitches and aliquots (2 ml) were centrifuged at 1678 g for 10 min, the sera were separated and stored in individually labelled Eppendorf tubes at -20 °C until tested. An IFAT was performed using promastigotes of L. infantum zymodeme MON1 as antigen [19] to investigate previous exposure to Leishmania parasites. While under general anaesthesia, bone marrow samples were obtained from bitches by aspiration from the iliac crest using Rosenthal needles (16 or 18 Gauge) and stored at -20 °C in individual tubes with 1 ml of phosphate buffered saline (PBS) until subsequent testing. Aliquots of whole blood and bone marrow samples were smeared on slides and microscopically examined by staining with May-Grünwald-Giemsa Quick Stain (Bio Optica Spa, Italy). Skin tissue samples of approximately 0.5 cm2 were collected from the interscapular region of bitches and processed as described previously [19]. Uterus, ovaries and placenta biopsy samples were collected and stored in individual sterile tubes (Table 1).

Amniotic fluid (1 ml) was collected for each foetus from the intact uterus using a sterile needle, and stored in individual tubes. Umbilical cord, spleen and liver samples (1 cm2) were collected by a sterile scalpel and stored in tubes containing PBS. The bone marrow samples were collected by cutting the tip of the sternum and stored in tubes with 1 ml of PBS (Table 2).

DNA extraction and parasite load determination

Total DNA was extracted from whole blood, skin, uterus, ovaries, placenta, liver, spleen and umbilical cord samples using the Genomic DNA Purification Kit (Gentra Systems, Minnesota, USA), while genomic DNA from amniotic fluid and bone marrow samples was extracted using QIAampDNA Micro Kit (Qiagen, GmbH, Hilden, Germany). For each sample, two qPCR reactions were individually performed for the detection and quantification of L. infantum and A. platys nucleic acids, using primers and probes targeting, respectively, the kinetoplast minicircle DNA (kDNA) and 16S rRNA gene, as described previously [18, 23]. DNA extracted from lymph nodes and whole blood from L. infantum and A. platys-infected dogs were included as positive controls. Quantification of DNA of L. infantum and A. platys was performed using a 10-fold dilution series of standard DNA from promastigotes (log phase concentration, 1.7 × 106 parasites/ml) of L. infantum (zymodeme MON-1) and from A. platys-infected blood with a concentration of 5.6 × 105 infected platelets/100 μl. The detection limits of the qPCRs were assessed using serial dilutions from 1.7 × 10-2 to 1.7 × 10-7 parasites (L. infantum) and from 2.24 × 102 to 2.24 × 10-6 infected platelets (A. platys) per reaction (2 μl of DNA template), respectively.

Results

Dogs included in the study, with the exception of negative controls, were serologically positive for L. infantum (IFAT titre up to 1:160) and for A. platys by pathogen DNA amplification from whole blood, bone marrow and skin samples (both L. infantum and A. platys) (Tables 1, 3 and 4). Eighteen out of 54 (33.3 %) tissue samples collected from pregnant bitches were positive to L. infantum by qPCR; of these, only one placental tissue sample was positive for this parasite, at a detected load of 1.2 × 10-3/PCR reaction (Tables 1 and 3). No uterine and ovarian samples were positive for L. infantum (Tables 1 and 3). Of the 100 foetal samples collected, only one umbilical cord was L. infantum-positive, albeit with a low parasite load of 8.14 × 10-4/PCR reaction (Table 2). Conversely, 17/45 (37.8 %) tissue samples collected from non-pregnant bitches were qPCR positive for L. infantum (Tables 1 and 4), with the highest parasite load detected in ovarian and uterine samples (i.e. up to 3.09 × 10 and 7.51/PCR reaction, respectively) (Table 4).

Table 3 Real-time PCR detection and quantification of Leishmania infantum and Anaplasma platys DNA. in tissue samples from pregnant bitches. Starting Quantity (SQ) mean value of parasite load per PCR reaction
Full size table
Table 4 Real-time PCR detection and quantification of Leishmania infantum and Anaplasma platys DNA in tissue samples from non-pregnant bitches. Starting Quantity (SQ) mean value of parasite load per PCR reaction
Full size table

Sixteen out of 54 (29.6 %) tissues from pregnant bitches were qPCR positive for A. platys, including three uterine, two ovarian and one placental sample with parasite loads ranging from 6.2 × 10-3 to 2.73 × 10-1/PCR reaction (Tables 1 and 3). Of the foetal tissues samples, five tested positive for A. platys (n = 2 liver; n = 3 spleen) with parasite loads ranging from 2.83 × 10-2 to 1.65 × 10-1/PCR reaction (Table 2). Of the 45 tissue samples collected from non-pregnant bitches (Tables 1 and 4), seven uteri and six ovaries scored qPCR positive for A. platys, with parasite loads ranging from 6.8 × 10-2 to 2.05/PCR reaction (Tables 1 and 4).

Ten bitches (n = 6 pregnant; n = 4 non-pregnant) were shown to be co-infected with both pathogens (Tables 1, 3 and 4). Two uterine samples from non-pregnant bitches were qPCR positive for L. infantum and A. platys, with parasite loads ranging from 2.17 × 10-3 to 2.32 × 10-3/PCR reaction and from 6.8 × 10-2 to 2.05/PCR reaction, respectively (Table 4). Tissues from dogs selected as negative controls on the basis of cytological examination of whole blood and bone marrow were qPCR negative for both pathogens (Tables 3 and 4).

Discussion

This study provides new data regarding vertical transmission of L. infantum and A. platys in naturally infected dogs. Based on the samples examined, vertical transmission of L. infantum from an infected bitch to the offspring could be less frequent during the first half of gestation (i.e. 25–35 days). Indeed, a previous study investigating vertical transmission of L. infantum in eight bitches and their 53 foetuses during late pregnancy (i.e. 50–60 days) showed that 32 % of foetuses and 49 % of the corresponding placentas were PCR positive for this pathogen [24]. In contrast, the detection of L. infantum DNA in only one placenta, moreover with a low parasite load (i.e. 1.21 × 10-3), supports our hypothesis that this pathogen is unlikely to be transmitted during this phase of gestation. However, based upon the qPCR findings from non-pregnant bitches, L. infantum may be able to spread to the ovary, uterus and blood, with the former tissues displaying a higher parasite load (i.e. 3.09 × 10) compared to the latter (i.e. 2.41 × 10-1). Nevertheless, albeit promising, our preliminary data needs confirming in studies including a significantly larger number of infected bitches.

The contrasting findings observed for L. infantum between pregnant and non-pregnant bitches might be related to the hormonal changes associated with pregnancy and the different stages of the reproductive cycle. Increased oestrogen concentration in non-pregnant bitches may cause hyperaemia of the uterus and ovary tissues [25], which in turn may have been responsible for the increase in pathogen concentration in these sites. Alternatively, the fact that the uterus and ovary tissues of pregnant bitches were negative for L. infantum could be associated to the high blood flow resistance that characterises the canine placenta throughout the first half of gestation, and that progressively decreases in conjunction with the development of the foetal and placental circulation throughout the second half of gestation [26].

In previous studies, DNA from L. infantum was detected from foetal tissue samples (i.e. bone marrow, liver and spleen) over the last ten days of pregnancy (i.e. 50–60 days) and in newborn puppies [7, 10, 12, 24, 27, 28], but not from the uteri of infected pregnant bitches [27]. The detection of L. infantum amastigotes mostly in lymphoreticular organs (e.g. spleen, bone marrow, lymph nodes and liver) of late-gestation foetuses or newborn puppies [24], could mirror the distribution of parasites in tissues and organs of adult dogs. Therefore, it is tempting to speculate that the lack of L. infantum-PCR positive samples from foetuses examined in the present study may be linked to the immature state of the foetal immune system in early gestation. Lymphocytic infiltration of the foetal lymph nodes and spleen becomes evident from 45 to 52 days of gestation, which temporally corresponds to the development of the bone marrow that, at this time, contains a large number of hematopoietic stem cells [29].

This study provides first evidence supporting the hypothesis that A. platys may be vertically transmitted from the pregnant bitch to the offspring. Indeed, liver, spleen and uterine, ovarian, and placental tissues of foetuses and mothers positive for this pathogen were also positive for A. platys, with comparable pathogen loads (i.e. 2.83 × 10-2 in foetuses and 2.78 × 10-2 in mothers). The presence of the marginal haemophagous zones in the dog placenta, filled with extravasated maternal blood [30] could justify the detection of A. platys-infected platelets.

The above factors may also justify the finding of a higher A. platys load (i.e. 2.05) than that of L. infantum (i.e. 2.32 × 10-3) in the uterus of co-infected non-pregnant bitches in this study. The possibility of vertical transmission of A. platys in dogs is in agreement with knowledge of this event occurring for Anaplasma marginale and Anaplasma phagocytophilum in cattle [31, 32], whereas the minimal data relative to in utero transmission of A. phagocytophilum in dogs [33] did not support transplacental transmission. Controlled laboratory studies involving puppies from bitches infected with A. platys should be addressed to better understand this transmission pattern. To mimic a natural infection, bitches should be infected with A. platys by blood transfusion or by a competent vector.

However, confirmatory studies on naturally infected pregnant bitches (both symptomatic and asymptomatic) in early (25–35 days) and late pregnancy (50–60 days) and on their foetuses, need to be carried out in order to confirm or confute this hypothesis.

Conclusion

Based on the results of the present study, transmission of L. infantum from infected mothers to their offspring appears unlikely or less frequent during the first half of gestation, whereas vertical transmission may occur during the later stages of gestation. In contrast, vertical transmission of A. platys may occur during the early stages of gestation, and throughout its entire course. If confirmed, these findings will pose additional challenges for the development of disease control strategies in both endemic and non-endemic areas.