microorganisms Article Experimental Mycobacterium microti Infection in Bank Voles (Myodes glareolus) Enric Vidal 1 , Judit Burgaya 1 , Lorraine Michelet 2 , Claudia Arrieta-Villegas 1 , Guillermo Cantero 1 , Krystel de Cruz 2 , Jennifer Tambosco 2 , Michelle Di Bari 3 , Nonno Romolo 3 , Maria Laura Boschiroli 2 and Bernat Pérez de Val 1, * 1 IRTA, Centre de Recerca en Sanitat Animal (CReSA, IRTA-UAB), Campus UAB, 08193 Bellaterra, Spain; enric.vidal@irta.cat (E.V.); judit.burgaya@gmail.com (J.B.); claudia.arrieta@irta.cat (C.A.-V.); guillermo.cantero@irta.cat (G.C.) 2 Animal Health Laboratory, National Reference Laboratory for Tuberculosis, Paris-Est University, Anses, 94700 Maisons-Alfort, France; Lorraine.michelet@anses.fr (L.M.); krystel.decruz@anses.fr (K.d.C.); Jennifer.tambosco@anses.fr (J.T.); maria-laura.boschiroli@anses.fr (M.L.B.) 3 Department of Food Safety, Nutrition and Veterinary Public Health, Istituto Superiore di Sanità, 00161 Rome, Italy; michele.dibari@iss.it (M.D.B.); romolo.nonno@iss.it (N.R.) * Correspondence: bernat.perez@irta.cat Abstract: Voles are maintenance hosts of Mycobacterium microti. In line with the goal to eradicate tuberculosis (TB) in livestock, the role of this mycobacteria needs to be assessed since it might interfere with current M. bovis/M. caprae surveillance strategies. To better understand the pathogenesis of TB  in voles, an experimental infection model was set up to reproduce M. microti infection in laboratory  Bank voles (Myodes glareolus). Two infection routes (intragastric and intraperitoneal) and doses Citation: Vidal, E.; Burgaya, J.; (105 and 106 CFU/0.1 mL) were assessed. Voles were culled at different post-infection time points. Michelet, L.; Arrieta-Villegas, C.; Serology, histopathology, acid-fast bacilli staining, qPCR, and mycobacterial culture from tissues Cantero, G.; de Cruz, K.; Tambosco, J.; were performed. In addition, qPCR from feces and oral swabs were conducted to assess bacterial Di Bari, M.; Romolo, N.; Boschiroli, shedding. The model allowed us to faithfully reproduce the disease phenotype described in free- M.L.; et al. Experimental Mycobacterium microti Infection in ranging voles and characterize the pathogenesis of the infection. Most animals showed multifocal and Bank Voles (Myodes glareolus). diffuse granulomatous lesions in the liver and spleen, respectively. Less frequently, granulomas were Microorganisms 2022, 10, 135. observed in lungs, lymph nodes, muscles, and salivary gland. Mycobacterial DNA was detected in https://doi.org/10.3390/ feces from a few animals but not in oral swabs. However, one contact uninfected vole seroconverted microorganisms10010135 and showed incipient TB compatible lesions, suggesting horizontal transmission between voles. Academic Editor: Antonella d’Arminio Monforte Keywords: voles; wild rodents; tuberculosis; Mycobacterium microti; experimental infection Received: 18 November 2021 Accepted: 18 December 2021 Published: 10 January 2022 1. Introduction Publisher’s Note: MDPI stays neutral Mycobacterium microti, a member of the Mycobacterium tuberculosis complex (MTBC), with regard to jurisdictional claims in is the main causative agent of tuberculosis (TB) in wild rodents, particularly field voles published maps and institutional affil- (Microtus agrestis), bank voles (Myodes glareolus), wood mice (Apodemus sylvaticus), yellow- iations. necked mice (Apodemus flavicollis), and shrews (Sorex araneus) [1–5]. Cases of TB due to M. microti have also been described in humans [6], a wide-range of domestic species, including cattle [7,8], goats [9], pigs [10], dogs [11], cats [12,13], and wildlife other than micromam- mals, such as meerkats [14], squirrel monkeys [15], badgers [16], foxes [17], deer [18], and Copyright: © 2022 by the authors. wild boar [19,20], presenting a distribution across different European countries, including Licensee MDPI, Basel, Switzerland. France, UK, Italy, Spain, Germany, Austria, and Switzerland [3,9,18–21]. This article is an open access article distributed under the terms and Although differentiation of M. microti from the other members of MTBC is possible conditions of the Creative Commons mainly by genotyping methods [21], it is rarely diagnosed because of the apparently lower Attribution (CC BY) license (https:// susceptibility in humans and livestock [6,7,9], but also due to M. microti’s slow growing creativecommons.org/licenses/by/ rate that that impedes its isolation by classical bacteriological methods [22]. However, 4.0/). the extended distribution of this mycobacteria, together with the newly sensitive hosts Microorganisms 2022, 10, 135. https://doi.org/10.3390/microorganisms10010135 https://www.mdpi.com/journal/microorganisms Microorganisms 2022, 10, 135 2 of 15 presents a concern for the identification and management of its role in the control strategies of animal TB. Indeed, it has been suggested that it may interfere with current M. bovis/M. caprae diagnosis conducted in livestock in the frame of TB eradication campaigns [8]. This poses a major concern in low bovine TB prevalence settings, such as in the Pyrenees, where M. microti has recently been detected in different hosts [20,23]. Since voles are described as the main maintenance host for this M. microti, an experi- mental infection was designed to reproduce the natural disease and better understand the pathogenesis and transmission of TB in voles. 2. Materials and Methods 2.1. Experimental Animals and Housing Bank voles (Myodes glareolus) were bred at the facilities of the Istituto Superiore di Sanità (Rome, Italy). Then, experimental voles, aged between 80 and 167 days, were transported to the BLS-3 facilities of IRTA-CReSA (Bellaterra, Spain) and were kept in a controlled environment at a room temperature of 22 ◦ C, 12 h light–darkness cycle and 60% relative humidity in HEPA filtered cages (both air inflow and extraction) in ventilated racks. The voles were fed ad libitum, observed daily, and their clinical status assessed once a week. Periodically, voles were weighed and oral swabs and individual fecal samples were obtained. Males and females were housed separately and a maximum of 4 animals per cage. All experiments took place at the Biocontainment Level 3 animal facility at IRTA-CReSA. 2.2. Study Design and Experimental Infection Two separate experiments were conducted sequentially. 2.2.1. Experiment 1 For the first experiment, a M. microti field strain used as inoculum was initially isolated in a dog of the Ariege region (France). The strain was selected on the basis of its spoligotype profile, SB0423 (Mbovis.org), which has been detected in different mammals in cross-border regions of France [23] and Spain [20] (Ariege, Haute Garonne, and Catalonia). The inoculum was prepared at Anses (Maisons-Alfort, France) as follows: The isolate was subcultured in Middlebrook 7H9 medium with mycobactin and a titer of 3 × 107 CFU/mL was determined by measurement of optic density at the beginning of the stationary phase. Once the master seed of M. microti was titered, it was diluted in sterile PBS at final suspension of 106 CFU/mL. This experiment involved two groups (n = 8, male and female balanced, thus, a total of 16 animals) each group with a different inoculation route: (1) intragastric (IG), 0.1 mL of 105 CFU/0.1 mL of M. microti using a plastic flexible intragastric mice probe and a 1 mL syringe; and (2) intraperitoneal (IP) via inoculation of 0.1 mL of 106 CFU/mL of M. microti with a 25 G needle and a 1 mL syringe. The inoculation was carried out under general gaseous anesthesia (Isofluorane 5%). Two time points were established at 28- and 114-days post inoculation (DPI) for euthanasia and postmortem examination. Animals that died spontaneously or had to be euthanized for humanitarian reason were also analyzed. 2.2.2. Experiment 2 For the second experiment, the inoculum was prepared as following: A M. microti SB0423 isolate, obtained from the spleen of a vole of the experiment 1 (IP group, euthanized at 28 dpi), was subcultured in Middlebook 7H9 medium (BD Diagnostics, Sparks, NV, USA) for 28 days at 37 ± 1 ◦ C. This second inoculum was titrated by qPCR as following: an aliquot of 1 mL of the suspension was inactivated at 75 ◦ C for 1 h. In parallel, an aliquot of the M. microti master seed (3 × 107 CFU/mL) used for the challenge inoculum of the experiment 1, was also inactivated and then ten-fold serially diluted to establish titrated standards. DNA samples were extracted and amplified using the ID Gene™ spin univer- sal extraction and ID Gene™ Mycobacterium tuberculosis complex Duplex commercial kits (ID.vet, Grabels, France), respectively, according to the manufacturer procedures. Amplifi- Microorganisms 2022, 10, 135 3 of 15 cation was performed in a 7500 fast real-time PCR system (Applied Biosystems, Walham, MA, USA). M. microti CFU genomic equivalents were calculated by the extrapolation of Ct value obtained for each DNA sample according to previously described [24]. Once titrated, the remaining subcultured suspension was diluted in sterile PBS at a final suspension of 107 CFU/mL. In this experiment, the IP route was chosen and a higher dose was used, it consisted of a group (n = 14, male and female balanced) inoculated IP with 0.1 mL of 107 CFU/mL of M. microti. Four additional unchallenged animals were housed in 4 cages, each of them sharing cage with 3 inoculated animals to assess the occurrence of horizontal transmission. An early time point was established at 28–37 dpi (a first batch of 3 animals were euthanized at 29 dpi and were grouped with other three animals that were found dead at 28 and 37 dpi, due to TB-unrelated causes) and a second late time point at 69 dpi for euthanasia and postmortem examination. Two additional voles remained uninoculated and unexposed to infected animals and were used as negative controls. 2.3. Ethics Statement All experimental animal procedures were approved by the Animal Research Ethics Commission of the Generalitat de Catalunya (procedure number FUE-2020-01337124 and ID 46KZF9ZM4), according to current European legislation for the protection of experimental animals (86/609/EEC, 91/628/EEC, 92/65/EEC, and 90/425/EEC). 2.4. Postmortem Examination and Pathological Evaluation Animals were humanely euthanized upon reaching end point criteria or at established time points by general gaseous anesthesia (Isofluorane 5%) and subsequent decapitation. After visual examination of the viscera, samples of liver, spleen, and lung tissue were obtained and frozen at −75 ◦ C, the remaining tissues were immediately immersed in formalin. Two weeks later the tissues were routinely processed for histopathology, i.e., paraffin embedded, 3–4 µm sections were obtained and further stained with Hematoxylin and Eosin (HE) for morphological assessment. Ziehl Neelsen (ZN) stain was used to detect AFB. The tissues processed for microscopic evaluation included: kidney, liver, lung, thoracic lymph node (LN), spleen, stomach, intestines, mesenteric LN, cervical LN, inguinal LN, and any tissue with visible lesions. To objectively assess the progression of the TB lesions, a pathology index was estab- lished based on a mixed scoring system that integrated different parameters evaluated separately: the absence or presence of lesions (0/1) inguinal LN, cervical LN, and peri- toneum; for the spleen a semiquantitative scoring system (0–3) was established (from 0—absence of lesion— to 3—extensive splenomegaly due to diffuse granulomatous inflam- matory infiltrate—scores 1, 2, and 3 were assigned to mild, moderate, and evident levels of the same lesional pattern, respectively); thoracic lesions were evaluated as summatory of the absence/presence (0/1) of TB lesions in the lung and thoracic lymph nodes com- bined. Finally, lesions in the hepatic tissue were quantitatively evaluated using Image J software [25] including different parameters: number of granulomas found in five 10× randomly selected liver fields, the area of the granulomas was calculated, as well as the absence/presence (0/1) of a central area on necrosis and absence/presence (0/1) of min- eralization. For the pathology index, the liver score included three parameters: necrosis (0/1), Mineralization (0/1) and a third one regarding the size of granulomas. For this third parameter a score of 0 was given when the animal’s total liver granuloma area median was below (smaller granulomas) the global liver granuloma area median (5729.985 µm2 ) and 1 when it was above (bigger granulomas). 2.5. Mycobacterial Cultures Liver, spleen, and lung tissues were thawed and sliced with sterile scissors and subsequently mechanically homogenized in 1.5 mL of sterile distilled water, and an aliquot of 1 mL of each homogenate was separated for subsequent bacterial load assessment by Microorganisms 2022, 10, 135 4 of 15 qPCR. The remaining 0.5 mL was decontaminated with 0.5 mL of oxalic acid at 5% w/v for 30 min and then neutralized with 250 µL of NaOH 1 M. Afterwards, samples were centrifuged at 2451× g for 30 min. Supernatants were discarded and pellets were suspended in 1 mL of sterile PBS. Suspensions were cultured as following: 100 µL were inoculated in BBL tubes and incubated in BACTEC MGIT 320 system (BD diagnostics, Sparks, MD, USA), 100 µL were cultured in Middlebrook 7H11 plates (BD diagnostics, Sparks, MD, USA), and a swab was immersed in the remaining suspension for culture in Löwenstein–Jensen with pyruvate and Coletsos solid medium tubes (BD Diagnostics, Sparks, MD, USA). Growth in positive cultures were confirmed as MTBC by multiplex PCR [26]. A culture was considered negative when no growth was observed in Middlebrook 7H11 plates (at 28 days), BACTEC MGIT (at 42 days) or solid medium tubes (at 90 days). 2.6. Mycobacterial Load Assessment by qPCR Aliquots of 1 mL of liver, spleen, and lung homogenates of each vole were inactivated at 75 ◦ C for 1 h. The M. microti master seed was used to generate the qPCR standard curve. DNA samples from homogenates and standards were extracted and amplified, and M. microti CFU genomic equivalents were calculated as described above. 2.7. Mycobacterial Excretion Assessment by qPCR Oral swabs and feces were collected from each at animal at days: 7, 14, 21, 28, and 56 p.i. (experiment 1), and 13, 41, and 69 p.i. (experiment 2). Oral swabs were immersed and cut in microtubes containing 1 mL of PBS, while feces was suspended and homogenized in 1 mL of PBS. All samples were inactivated at 100 ◦ C for 10 min at stored at −20 ◦ C. DNA was extracted from oral swabs using an LSI MagVetTM Universal Isolation Kit (Life Technologies, Villebon sur Yvette, France) with a KingFisherTM Flex automate (Ther- moFisher Scientific, Villebon sur Yvette, France), following the manufacturer’s instructions. DNA was extracted from feces with the FastDNA™ Spin Kit for Soil (MP Biomedicals, Eschwege, Germany). Real-time PCR were carried out in a 25 µL reaction mix contain- ing TaqMan™ Fast Advanced Master Mix (ThermoFisher Scientific, Villebon sur Yvette, France), 300 nM forward and reverse primers, 250 nM probes, sterile water, and 5 µL of DNA template. Thermocycling conditions were 50 ◦ C for 2 min (1 cycle), followed by one cycle of 20 s at 95 ◦ C and 40 cycles of 3 s at 95 ◦ C and 30 s at 60 ◦ C. PCR inhibition was tested (Diagenode, Seraing, Belgium). Positive detection of M. microti was established on the basis of a positive response for IS1081 and IS6110 (Mycobacterium tuberculosis complex) [27]. 2.8. Serology Blood was obtained upon sacrifice through decapitation. Sera were separated and a MTBC IgG ELISA was carried out and interpreted as previously described [28]. Briefly, sera were analyzed for antibodies against the MTBC-specific MPB83 antigen (Lionex, Braunschweig, Germany) by a homemade IgG indirect ELISA. A sample was classified as positive (i.e., seroconversion) when ∆OD (sample wells Optical Density—OD—at 450 nm minus the blank well OD at 450 nm) ≥0.2. 2.9. Data Analysis A randomized design was used for data analyses. Differences in the spleen pathology score, liver granuloma area, overall pathology index, and bacterial loads in the spleen (log10 CFU equivalents) between the early and late infection groups were compared using the one-tailed Mann–Whitney non-parametric test. Correlation between the spleen pathology score and bacterial loads in the spleen was performed by a non-parametric Spearman rank test. The statistical analysis and visualizations were carried out using the R software version 4.1.1, and the following packages: dplyr (1.0.7), tidyverse (1.3.1), ggplot2 (3.3.5), ggpubr (0.4.0), hrbrthemes (0.8.0), viridis (0.6.1), viridisLite (0.4.0), heatmaply (1.2.1), and plotly (4.9.4.1). Microorganisms 2022, 10, 135 5 of 15 3. Results 3.1. Comparison of Infection Routes In experiment 1, at 28 dpi no macroscopic or microscopic lesions were observed in neither IP nor the IG groups. In the IP group 2/4 voles seroconverted and 3/4 yielded low titer positive PCR and culture results from the spleen samples. No indication of TB infection was found in any of the IG animals euthanized at 28 dpi (see Table 1 for a summary of experiment 1 results and Supplementary File S1 for the complete data set). Table 1. Summary of the results for Experiment 1. The numbers indicate the number of positive results/total animals per group (or available samples). dpi: days post inoculation. Experiment 1 MTBC qPCR MTBC Culture TB Compatible Lesions Timepoint Serology Route Spleen Liver Lung Spleen Liver Lung Spleen Liver Lung Early 0/4 0/4 0/4 0/4 0/4 0/4 0/4 0/4 0/4 0/4 Intragastric (23–28 dpi) (105 CFU) Late 2/3 3/4 1/4 2/4 2/4 3/4 1/4 2/3 2/4 0/4 (58–114 dpi) Early 2/4 3/4 0/4 0/4 3/4 0/4 0/4 0/4 0/4 0/4 Intraperitoneal (17–28 dpi) (105 CFU) Late 2/2 4/4 2/4 2/4 2/4 2/4 1/4 2/3 3/3 0/3 (95–114 dpi) At late time points, the four animals of the IP group showed positive qPCR in the spleen with similar or higher titers than the inoculum (from ~5 × 104 to ~107 CFU equiva- lents, see Supplementary File S1), whereas the IG group showed 3 out of 4 positive spleens, one of them with a lower titer than the inoculum (~8 × 102 CFU equivalents). An animal of the IP group was found dead at 58 dpi (the cause of death could not be determined) and it already showed TB compatible lesions. The TB lesions in this animal were mild and were not considered to be the cause of death. The remaining animals of the IP group, sacrificed at 114 dpi when the end point of the study was established, showed evident macroscopic and microscopic TB compatible lesions. Macroscopically a moderate increase in the spleen size (splenomegaly) was observed only in the two animals of the IP group sacrificed at 114 dpi. Microscopically small granu- lomas were observed in the liver with a miliary distribution, composed of macrophages, the occasional multinucleated giant cell, and small foci of mineralization. The splenic parenchyma was invaded by a diffuse granulomatous infiltrate with scant multinucleated giant cells and multifocal foci of mineralization (Figure 1). The same infiltrate could be observed in cervical, thoracic, mesenteric, and inguinal lymph nodes. In one animal a focal area of granulomatous infiltrate could be observed in the salivary gland. Finally, and probably because of the inoculation procedure, a subcutaneous granulomatous infiltrate was observed in the left inguinal region as well as foci of granulomatous infiltrate in the peritoneum, in contact with pancreatic tissue. No lesions were observed in the lung or the kidneys. A moderate to low amount of AFB was observed in all the lesions. The results obtained with the IG inoculated group at later time points were inconsis- tent: one animal was found dead at 95 dpi with lesions similar to those described above for the IP group at 114 dpi, including splenomegaly, but it also showed a visible subcutaneous granuloma, consisting of a histiocytic proliferation infiltrating the skeletal muscle of the left anterior limb. Ziehl Neelsen staining evidenced that most of the macrophages were heavily loaded with abundant AFB (Figure 2). This animal also showed the highest bacterial load of the experiment in the spleen, liver, and lungs (~4 × 108 , ~3 × 105 and ~3 × 107 CFU equivalents calculated by qPCR, respectively). Regarding the two remaining animals, euth- anized at the end point of the experiment (114 dpi), one showed no lesions (yet a positive culture result in the liver) and the other had a lesional pattern equivalent to that observed Microorganisms 2022, 10, 135 6 of 15 Microorganisms 2022, 9, x 6 of 15 in the IP group, including granulomatous miliary hepatitis, diffuse granulomatous splenitis peritoneum, and thoracic in contact and with inguinal pancreatic tissue. granulomatous No lesions were observed in the lung or the lymphadenitis. kidneys. A moderate to low amount of AFB was observed in all the lesions. Figure 1. TB lesions in intraperitoneally inoculated bank voles: (A): Mock inoculated bank vole nec- Figure 1. TB lesions in intraperitoneally inoculated bank voles: (A): Mock inoculated bank vole ropsy. Notice the thin, almost transparent, spleen (black arrowhead) (BV2). (B): Bank vole eu- necropsy.29 thanized Notice the thin, dpi (BV15, 106almost transparent, CFU dose), notice aspleen (black moderate arrowhead) increase (BV2). in spleen (B): size Bankarrowhead) (black vole eutha- nized 29 dpi (BV15, 10 6 CFU dose), notice a moderate increase in spleen size (black arrowhead) and and the presence of a granulomatous subcutaneous lesion in the inguinal region (white arrowhead). the presence (C): Necropsyofofa granulomatous subcutaneous a bank vole euthanized lesion at 69 dpi in the (BV40, 10inguinal region 6 CFU dose). (white Notice thearrowhead). (C): severe spleno- Necropsy of a bank vole euthanized at 69 dpi (BV40, 10 6 CFU dose). Notice the severe splenomegaly megaly (black arrowhead) and also a small inguinal subcutaneous granuloma (white arrowhead). (D): (blackLiver of BV16 and arrowhead) CFUadose, (106 also small69 dpi) showing inguinal larger granuloma subcutaneous granulomawith a central (white area of(D): arrowhead). necrosis Liver (asterisk). HE, 6 bar 100 µm. (E): Salivary gland of BV8 (10 4 CFU dose, 114 dpi). Notice the granulom- of BV16 (10 CFU dose, 69 dpi) showing larger granuloma with a central area of necrosis (asterisk). atous HE, barinfiltration 100 µm. of theSalivary (E): glandular parenchyma gland of BV8 (10(asterisk) 4 CFU dose,as well 114asdpi). the surrounding adipose tissue Notice the granulomatous (on the right). HE, bar 100 µm. infiltration of the glandular parenchyma (asterisk) as well as the surrounding adipose tissue (on the right). HE, bar 100 µm. The results obtained with the IG inoculated group at later time points were incon- sistent: one animal was found dead at 95 dpi with lesions similar to those described above for the IP group at 114 dpi, including splenomegaly, but it also showed a visible subcuta- neous granuloma, consisting of a histiocytic proliferation infiltrating the skeletal muscle of the left anterior limb. Ziehl Neelsen staining evidenced that most of the macrophages were heavily loaded with abundant AFB (Figure 2). This animal also showed the highest bacterial load of the experiment in the spleen, liver, and lungs (~4 × 108, ~3 × 105 and ~3 × 107 CFU equivalents calculated by qPCR, respectively). Regarding the two remaining an- imals, euthanized at the end point of the experiment (114 dpi), one showed no lesions (yet a positive culture result in the liver) and the other had a lesional pattern equivalent to that Microorganisms 2022, 10, 135 7 of 15 observed in the IP group, including granulomatous miliary hepatitis, diffuse granuloma- tous splenitis and thoracic and inguinal granulomatous lymphadenitis. Figure 2. TB lesions in an intragastric inoculated bank vole (BV20, 95 DPI): (A): Macroscopically an Figure 2. TB lesions in an intragastric inoculated bank vole (BV20, 95 DPI): (A): Macroscopically an evident increase in size of the spleen (white arrowhead) was observed, along with a subcutaneous evident increase in size of the spleen (white arrowhead) was observed, along with a subcutaneous granuloma in the axillar region (black arrowheads). (B): Microscopically the axillar lesion corre- granuloma sponded to in the axillar a diffuse region (blackinflammatory granulomatous arrowheads).proliferation (B): Microscopically (asterisk) the thataxillar lesionmuscle infiltrated corre- sponded to a diffuse granulomatous inflammatory proliferation (asterisk) that infiltrated fibers. Remains of skeletal muscle fibers can be observed on the bottom half of the image. Hematox- muscle fibers. ylin andRemains of skeletal Eosin staining (HE),muscle bar 100fibers µm. can bemost Inset: observed on the bottom macrophages half of loaded were heavily the image. withHema- abun- toxylin dant and Eosin acid-fast staining bacilli Ziehl–bar (AFB). (HE), 100 µm. Neelsen Inset:(ZN), staining mostbarmacrophages 25 µm. (C): were heavily A single smallloaded with granuloma could be observed abundant acid-fastin the lung bacilli (asterisk). (AFB). HE, Barstaining Ziehl–Neelsen 50 µm. Inset: (ZN), abundant bar 25 µm.AFB (C):were observed A single small in this granu- loma could be observed in the lung (asterisk). HE, Bar 50 µm. Inset: abundant AFB were observed in this lesion. ZN, bar 25 µm. (D): Multifocal small granulomas were observed in the liver (black arrowheads) some of which showed punctiform mineralization foci (white arrowhead). HE, bar 100 µm. Inset: a few AFB were observed in these lesions. ZN, bar 25 µm. (E): A diffuse granulomatous infiltrate was observed in the spleen with multinucleated giant cells (black arrowhead) and focal mineralization foci (white arrowhead). HE, Bar 50 µm. Inset: abundant AFB were observed in this lesion. ZN, bar 25 µm. Microorganisms 2022, 10, 135 8 of 15 MTBC DNA was not detected from oral swabs by PCR in any animal, even in that with the granulomatous sialadenitis. Regarding the feces, in the first experiment only one animal showed double positivity to both IS6110 and IS1081 targeted qPCRs (criterion to consider a sample as positive). This result was obtained in one animal (BV-25) of the intragastric group at 7 dpi, probably corresponding to the remains of the inoculum since the feces of this animal were negative at later time points (Supplementary File S1). 3.2. Comparison of IP Dosages For the experiment 2 the IP infectious route, which had demonstrated a more consistent result, was selected as well as an inoculum with a higher titer (106 CFU/0.1 mL) to try to optimize the model and shorten the disease onset (See Table 2 for a summary of experiment 2 results and Supplementary File S1 for the complete data set). Table 2. Summary of the results for Experiment 2. The numbers indicate positive results/total animals per group (or available samples). dpi: days post inoculation. Experiment 2 MTBC qPCR M. microti Culture TB Compatible Lesions Timepoint Serology Route Spleen Liver Lung Spleen Liver Lung Spleen Liver Lung Early 3/3 6/6 5/6 6/6 4/6 3/6 2/6 4/5 6/6 2/6 Intraperitoneal (28–37 dpi) (106 CFU) Late 8/8 8/8 5/8 4/8 8/8 7/8 0/8 8/8 8/8 2/8 (69 dpi) Late Contact 1/4 0/4 0/4 0/4 0/4 0/4 0/4 0/4 0/4 0/4 (69 dpi) All the voles euthanized at the early time point (28–37 dpi) seroconverted and devel- oped TB compatible pathology comparable or slightly more severe than the one observed in 114 dpi animals of the first experiment. Additionally, 2/5 voles also presented pulmonary granulomas and 3/5 showed granulomatous infiltrate in the peritoneum (Table 2, Sup- plementary File S1). PCR revealed M. microti titers in the spleen equivalent to, or slightly higher, than those contained in the inoculum (Table 2). All the animals euthanized at a late time point in experiment 2 (69 dpi) showed a more severe and advanced pathology, also characterized by miliary granulomatous hepatitis, but in this case some of the granulomas (in 3/8 animals) were larger and had a central area of necrosis. Diffuse granulomatous splenitis was also observed fully obliterating the splenic parenchyma. Granulomatous lymphadenitis could be observed in all studied lymph nodes (thoracic, cervical, mesenteric, and inguinal) but only 2/8 voles had granulomatous lesions in the lung. As with experiment 1 all IP inoculated animals showed some degree of subcutaneous inguinal granulomatous infiltrate as well as peritoneal granulomatous infiltrates, the latter were observed not only in the vicinity of pancreatic parenchyma but also surrounding the kidneys, without infiltrating the capsule nor the renal parenchyma. 3.3. Disease Progression To study the progression of the disease several parameters were semiquantitatively and quantitatively evaluated in early and late time points in the animals from Experiment 2 (Figures 3–5). Microorganisms 2022,10, Microorganisms2022, 9, 135 x 99 of of1515 Figure3.3.Splenic Figure Spleniclesions lesionsprogression: progression:(A):(A):Spleen Spleenpathology pathologyscoring. scoring.From From0—absence 0—absenceofoflesion lesiontoto 3—extensive 3—extensivesplenomegaly. splenomegaly.Scores Scores1,1,2,2,and and33correspond correspondtotomild, mild,moderate, moderate,and andevident evidentlevels levelsofof thesame the samelesion lesionpattern. pattern.Scale Scalebars bars100µm. 100µm.(B): (B):Spleen Spleenpathology pathologyscore scoredifference differencebetween betweenthetheearly early andlate and latetime timepoints. points. Significances Significances determined by by one-tailed one-tailedMann –Whitney test. Mann–Whitney test.(C): (C):Differences Differencesin inthe thebacterial bacterialload loadbetween betweenthetheearly early and and late time points late time points as aslog CFU equivalents log1010 CFU equivalentsmeasured measuredby by qPCR. Significances determined by one-tailed Mann – Whitney test. (D): Correlation qPCR. Significances determined by one-tailed Mann–Whitney test. (D): Correlation determined by determined by betweenthe between thebacterial bacterialload load(log CFUequivalents) (log10CFU equivalents)andandthe thespleen spleenpathology pathologyscore. score.Spearman Spearmanrank rank 10 test, ρ: Spearman rho. test, ρ: Spearman rho. The pathology scoring system applied for the spleen (from ‘0’ normal spleen to ‘3’ severe granulomatous splenomegaly, Figure 3A) revealed a statistically significant increase in the late time point (p = 0.021, Figure 3B). The average score for the early time point was 1 and for the late time point was 1.875. The late time point also showed an increased bacterial load measured in this tissue by the qPCR although not statistically significant (p = 0.091, Figure 3C). Accordingly, pathology score and bacterial load showed a statistically significant direct correlation (Spearman ρ = 0.7, p = 0.008, Figure 3D). Neither the number nor the average area of liver granulomas in five 10× random fields showed statistically significant differences between time points. However, qualitative differences were observed: in the late time point, higher number of larger granulomas were found (Figure 4), and additionally 3/8 animals had granulomas with central necrosis that were not observed in any of the early time point animals. Mineralization was observed in both time points indistinctly and none of the granulomas presented capsulation. At the late time point, the animals presented a higher absolute number of granulomas compared Microorganisms 2022, 10, 135 10 of 15 Microorganisms 2022, 9, x 10 of 15 Microorganisms 2022, 9, x 10 of 15 to the early time point (62% increase). This increase is mainly explained by a larger number of small granulomas, which is compatible with a recent generation. Figure 4. Liver pathology measured as the granuloma area. The median granuloma area (µm2 ) at2the Figure 4. Liver pathology measured as the granuloma area. The median granuloma area (µm ).at early and Figure late stages 4. Liver is represented byashorizontal pathology lines. Number ofmedian granulomas: N = 94area at early stage stages ismeasured represented the granuloma area.Number The granuloma N = 94(µm 2).at the early and late by horizontal lines. of granulomas: at early andearly the N = 152, at a late and late stage. stages Significances is represented bydetermined by one-tailed horizontal lines. NumberMann–Whitney of granulomas: test. N = 94 at early stage and N = 152, at a late stage. Significances determined by one-tailed Mann–Whitney test. stage and N = 152, at a late stage. Significances determined by one-tailed Mann–Whitney test. Figure Figure 5.5. Pathology progression summary. Pathology progression summary. (A): (A): Heatmap Heatmap of of normalized normalized scores scores for for each each bank bank vole. vole. Figure The rows5. Pathology correspond correspond toprogression to the ids summary. (BV-XX), (BV-XX), andand (A): the Heatmap columns of to normalized the score scores given to for each each organ bank vole. according The rows to the the correspond disease disease to the ids progression progression (BV-XX), and/or and/or and the columns toof the presence/absence the presence/absence the of score given lesions. Spleen: Spleen:to each 0–3 organ 0–3 accordingaccording according to the to the disease spleen progression score. Liver: and/or to 0–3 according thethepresence/absence presence/absence ofoflesions. Spleen: calcification, 0–3 according central necrosis, to the and the presence/absence of calcification, spleen score.size granuloma Liver: 0–3 media (>total according= 1, to the media total media = 1, total thoracic lymphmedia = 1,cLN node.