Anatomical MRI and [18F]FDG PET/CT imaging of Schistosoma mansoni in a NMRI mouse model

Schistosomiasis represents one of the most devastating worm parasitosis in the world. Current diagnostic methods are insufficient to determine the infection grade and the disease related organ damage. We herein investigated whether discrimination of infection grade and its correlation to liver damage could be accurately performed by multimodal imaging in a mouse model of Schistosoma mansoni infection. Therefore, groups of uninfected and infected mice underwent MRI and [18F]FDG PET/CT imaging. Anatomical MRI images were used for liver volumetry and for quantification of hepatic granulomas. For PET/CT images a volume of interest based analyses were employed to calculate the [18F]FDG uptake in liver, portal vein, spleen and abdomen. Herein, we demonstrate that the combined use of [18F]FDG-PET/CT and MRI represents an appropriate diagnostic tool for Schistosoma mansoni infection, but fails to discriminate the infection grade and the linked organ damage. Only the splenic [18F]FDG uptake in the 25 cercariae group (5.68 ± 0.90%ID/cc) and 50 cercariae group (4.98 ± 1.43%ID/cc) was significantly higher compared to the control group (2.13 ± 0.69%ID/cc). Nevertheless, future multimodal imaging studies with new radiopharmaceuticals could build a highly sensitive and specific basis for the diagnosis and evaluation of organ damage of schistosomiasis.

Schistosomiasis is caused by digenetic blood trematodes of the genus Schistosoma spp. and represents one of the most devastating worm parasitosis in the world, accounting for more than 260 million infected people, mainly in tropic and sub-tropic regions 1 . In developing countries the disease has a significant impact on economic and public health 2 . The main human pathogenic species of schistosomes are Schistosoma mansoni (S. mansoni), Schistosoma japonicum (S. japonicum) and Schistosoma haematobium (S. haematobium) 3 . People become infected by skin contact to waterborne cercariae shed by infected fresh water snails 3,4 . In the body the cercariae undergo a variety of morphological changes and transform into juvenile larvae known as schistosomula 5 . Schistosomula invade blood vessels and migrate through the heart and the lung into the vascular system of the liver, where they mature into adult male and female worms. After maturation, paired worms migrate against the blood flow to the veins of the small intestine (S. japonicum), colon (S. mansoni) or bladder (S. haematobium) 6 . Here, the female parasites produce several hundred eggs per day 7 . Parasite eggs provoke local inflammatory lesions within the vessel wall which pave the way to reach the gut or the bladder wall to be finally shed into the environment through feces or urine of the host 8 . Parasite eggs contain miracidia that hatch by fresh water contact to infect fresh water snails for completion of developmental cycle of schistosomes 2,8 . However, most of the parasitic eggs (S. mansoni and S. japonicum) are flushed back to the liver via bloodstream to become entrapped within the small liver sinusoids. Here they provoke T-cell mediated inflammation resulting in the generation of granulomatous lesions followed by vigorous fibrous tissue formation 1,9 . After long-term infection (5-20 years) chronic progressive liver fibrosis leads to portal hypertension and its clinical sequelae, such as ascites and gastrointestinal/ esophageal varices bleeding associated with high mortality rate [8][9][10][11][12][13] .

Scientific Reports
| (2020) 10:17343 | https://doi.org/10.1038/s41598-020-74226-2 www.nature.com/scientificreports/ Currently, the gold standard therapy is based on the anthelmintic drug Praziquantel. It eliminates adult worms, but does not cure chronic liver or intestinal damage and provides no protection against future or recurrent infections 14 . Moreover, mass drugging with Praziquantel paves the way for emergence of Praziquantel resistant schistosome strains 15 . Steinemann et al. have estimated more than 779 million people worldwide holding a risk for infection with schistosomes in future 16 .
Besides the urgent need for new therapeutic options, it is necessary to have thorough assessment of the disease progression and evaluation of drug effectiveness of future therapy applications against schistosomiasis. In clinical practice, parasitological techniques such as filtration of urine and examination of stool (microscopy-based Kato-Katz or Telemann) are still the methods of choice for diagnosis and monitoring of drug efficacy. But these methods have low sensitivity, especially in low-prevalence areas, have a significant time-delay as they are dependent on eggs and therefore do not detect early infections and cannot accurately display the disease burden 6,[17][18][19] . Immunodiagnostic techniques are in fact more sensitive and easier to use, but antibodies can be detected even years after infection. This makes it difficult to distinguish between active and previous exposure 17, [20][21][22] . DNA (circulating cell-free parasite DNA) and RNA (circulating miRNAs) based diagnostic methods have been proven to be highly sensitive and specific 23 . The requirement for specialized laboratories makes the broad field application of these methods a problematic issue. In addition, the detection of the intensity of infection as an indicator of morbidity 24 and parameter for drug effectiveness is barely covered by any of the above methods.
Modern imaging techniques like positron emission tomography / computed tomography (PET/CT) and magnetic resonance imaging (MRI) could improve diagnostic of schistosomiasis, in particular with a focus on assessing the degree of infection, and could help to evaluate the efficiency of novel immunoprotective applications during the course of the disease. Due to the high metabolism of glucose by the parasite worms 25 20 . In addition, they also showed a strong correlation between [ 18 F]FDG uptake in the portal vein area and worm burden when more than 50 worms are present 20 . Moreover, the characterization of a mouse model of schistosomiasis using MRI demonstrated that this in vivo imaging technique is capable of monitoring liver damage during the course of the disease 26 . The pathological findings obtained in this MRT study were consistent with known characteristics of infected people, confirming the convenience of MRI imaging as an appropriate instrument to detect schistosomiasis and determine the efficacy of new developed therapies 26 . In light of these results, we assume that a combined use of PET/CT and MRI could be a beneficial technique for clinicians and researchers to visualize and characterize the worm load in infected individuals and to evaluate new treatments for schistosomiasis in vivo.
Therefore we herein performed for the first time a multimodal [ 18 F]FDG-PET/CT and anatomical MRI study in a mouse model of S. mansoni to investigate the usability of this combination for determining the degree of infection and its correlation to possible tissue or organ damage. In addition, the results of multimodal imaging were validated in relation to the histological findings.

MRI.
In all infected groups (25,  Quantification of multifocal hyperintensities (lesions) in the liver (Fig. 1B) has shown a correlation between the number of lesions and the degree of infection. The highest amount of lesions was found in the group infected with 100 cercariae (14.15 ± 4.18 lesions per 10 mm 2 ), compared to the group infected with 25 cercariae (6.01 ± 3.19 per 10 mm 2 , p = 0.016-25 cercariae vs. 100 cercariae) and 50 cercariae group (11.77 ± 3.34 per 10 mm 2 , p = 0.716-50 cercariae vs. 100 cercariae). The groups 50 and 100 displayed a significant difference in number of lesions compared to the uninfected mice. (p < 0.001 for control vs. 100 cercariae and p = 0.003 for control vs. 50 cercariae).
PET/CT. With regard to the PET/CT images, high radioactive uptakes were observed in the heart, bladder and kidneys of control and infected mice (Fig. 4). However, compared to the controls, the infected mice showed further regions of intensive radioactive accumulation including the abdomen (arrowheads), portal vein (asterisks) and spleen (arrows) (Fig. 4B-D). Qualitative analysis of the PET/CT images showed slightly more radioactive intensity in the liver of infected mice compared to control mice.
For analysis of correlation between the infection burden and the [ 18 F]FDG uptake, the % injected dose per cubic centimeter (%ID/cc) was calculated in the liver, portal vein, spleen and abdomen of each animal. In general, the total uptake of [ 18 F]FDG in all investigated regions was elevated in infected mice compared to uninfected animals, whereas the radioactivity levels in the particular regions were quite similar between the groups of mice infected with different numbers of cercariae (Fig. 3).

Discussion
In addition to the diagnostics of schistosomiasis, the assessment and quantification of worm and egg burden as well as the evaluation of infection-related tissue damage is a challenge 28 . Direct detection methods like Kato-Katz or Telemann stool microscopy or indirect detection of anti-bodies against schistosome antigens in the blood serum 29 are not suitable for drawing conclusions about parasite associated organ damage. In human infection there is no gold standard for the evaluation of portal fibrosis 30 . Wedge or percutaneous liver biopsies are invasive methods that are unsuitable for long-term or follow-up assessment. Ultrasound scans or elastography methods have proven to be more suitable for weighing periportal fibrosis in a human schistosome infection 31,32 . For decades, the murine model of S. mansoni infection is used for experimental research and development of anti-fibrotic drugs or vaccines against the parasite. None of the above-mentioned detection methods used in humans are suitable or available for determination of drug effectiveness in long-term follow-up studies. In this study, we represent a multimodal imaging examination including PET/CT and MRI in a S. mansoni mouse model to weigh the potential of this combination for quantification of the infection burden and to monitor schistosomiasis-induced pathology. To our knowledge, this is the first time that such a combined in vivo imaging approach of S. mansoni has been applied. Furthermore, anatomical MRI and PET/CT data were cross-correlated with histological analyses. Overall, with our multimodal imaging study at the described murine model of S. mansoni infection we found diseases related pathological differences between uninfected and infected mice. In line with a previous MRI study 27 that investigated the liver in a S. mansoni mouse model, we could also identify multifocal hyperintensities in anatomical T2-weighted images which are associated with granulomas within the liver of S. mansoni infected mice. These detectable hepatic granulomas are caused by the deposited parasite eggs which are produced by mate worm pairs living mainly in the veins of the colon and then flushed back to the liver via the blood stream 6,33 .
Mice infected with 100 cercariae displayed the highest number of hepatic granulomas. However, a significant difference was quantifiable only between light (25 cercariae) and high infection (100 cercariae) by the presented imaging analysis. This finding suggests that the T2-weighted MRI imaging is not sensitive enough to differentiate all grades of infection via quantification of egg granulomas in the liver. The experimental design chosen here is fairly artificial. We aimed here to cause three different degrees of liver damage in terms of the number of granulomatous lesions within the livers. We paid special attention to the different, widely spread infection intensities in order to investigate whether these differences can be quantified by MRI or PET/CT. We have used the system of 25, 50 and 100 cercariae in the past to analyze mild, moderate and severe clinical picture 34 . However, histological analyses of the granuloma sizes and collagen fibers within the livers using HE and SR staining revealed comparable results. These findings underline the assumption that beside the worm burden other factors like host genetic background, conditions of infection and parasite strains must play a role in the development of disease www.nature.com/scientificreports/ severity during S. mansoni infection 13 . Moreover, we observed a high inter-individuality regarding the disease progression and its representation in our MRI data analysis. While an increase in liver weight was detected ex vivo, our volumetric analyses of the liver derived from the anatomical T2-weighted MRI images showed no significant increases of liver sizes as signs of hepatomegaly, typical for schistosome infection. According to the data of the liver volume measurement the infection intensity was not distinguishable. These discoveries are contrary to the findings of a previous MRI study with S. mansoni infected mice. Masi and colleagues have shown an increase in liver volume of 19% six weeks after infection with 30 cercaria of a Venezuelan S. mansoni strain 26 . This discrepancy between increased liver weight and steady liver volume found in our study could be explained by the disease related liver tissue alterations (e.g. formation of fibrosis or deposition of eggs) which are accompanied by density changes 35 .
Our results of the MRI analysis during S. mansoni infection suggest that the determination of the hepatic granulomas could serve as a marker to distinguish between uninfected and infected patients in clinical practice. However, in agreement with our histological analyses a classification of the infection grade was not clearly possible based on the MRI derived data of the liver. From our point of view, future studies on a larger cohort of animals and additional examinations of S. mansoni infection affected organs for example spleen and intestines 26 are needed to establish MRI as appropriate diagnostic tool for quantitative assessment of disease burden. Furthermore, mri techniques which probe functional and quantitative parameters beyond anatomical imaging and T2/T2* relaxometry like perfusion and diffusion imaging should be investigated regarding their potential to determination the infection grade during S. mansoni infection.
PET imaging revealed an increased [ 18 F]FDG accumulation following infection. Livers, portal veins, spleens and abdomen represent the primarily affected regions by schistosome infection. An elevated [ 18 F]FDG uptake was detectable in all infected mice compared to the healthy controls. This observation goes in line with a previous PET study which also showed a disease related higher uptake of [ 18 F]FDG in the abdomen, livers and portal veins of S. mansoni infected mice 20 . However, additional to this work we could identify a higher splenic   20 reported in an PET imaging study of S. mansoni infected mice a positive linear correlation between the total [ 18 F]FDG uptakes in the portal veins as well as livers and worm burden. However, the ability to quantify the worm burden based on this relationship was limited in mice with low worm burden and only powerful in mice with more than 50 worms 20 . Since only approx. 60% of the penetrated cercariae mature to adult worms in mice 39 , our mice infected with 25, 50 and 100 cercariae displayed a burden of around 15, 30 and 60 worms respectively. Based on this and in conjunction with Salem et al. we presume that PET imaging with [ 18 F]FDG is not optimal to quantify small numbers of present parasites. This drawback could be solved in the future by using a radiotracer that is directly incorporated by the worms via i.e. oral ingestion. For example, radioactively labeled albumin could be used for the accurate localization of worms and quantification of worm burden, because it was already shown that fluorescence marked albumin is ingested and digested by all development stages of schistosomes 40 . In principle, however, both studies are consistent in regard to the MRI and PET findings, but only a weak correlation was found between the liver lesions obtained by MRI analysis and the total [ 18 F]FDG uptake in livers, portal veins, spleens and abdomen. This underlines the importance of the combined usage of MRI and PET/CT for accurate diagnosis and monitoring of S. mansoni infection. We are aware that the high inter-animal variance and the relatively small number of investigated animals per group present a limitation of our study. Therefore, a multimodal imaging study in the same experimental way but with a larger cohort of animals and a higher infection intensity (> 100 cercariae) in a longitudinal approach until beyond 10 weeks after infection would be advantageous or, in order to achieve a more human-like periportal fibrosis low number of cercariae and extended infection time (Andrade ZA, 1997). However, regarding the evaluation of periportal fibrosis or more severe clinical forms, a PET/CT tracer that specifically binds to fibrotic tissue would be required. This would be a great achievement for the future.
In conclusion, this study describes an in vivo scan protocol to characterize S. mansoni infection in a mouse model using anatomical MRI imaging and [ 18 F]FDG-PET/CT. Morphological MRI imaging is capable to visualize

Schistosoma mansoni infection model and study design. Schistosoma mansoni (S. mansoni; Belo
Horizonte strain) was held in a life cycle using Biomphalaria glabrata (B. glabrata) fresh water snails (Brazilian strain) as intermediate hosts and female NMRI mice as definitive hosts. Six-to eight-weeks-old NMRI mice (Charles River Laboratories, Germany) were housed in an animal facility with a 12:12 h light/dark cycle with ad libitum water and free access to standard chow (SSNIFF, Soest, Netherland). S. mansoni cercariae were obtained by mass shedding of 5-10 infected B. glabrata after day light exposure. With our experimental design we aim to investigate three widely differing infection intensities and degrees of liver damage 34 . Each mouse was exposed to parasites by sitting in  PET/CT imaging. All PET/CT scans were performed on a small animal PET/CT scanner (Inveon MM-PET/ CT, Siemens Medical Solutions, Knoxville, TN, USA) according a standard protocol. Mice were anaesthetized using isoflurane (4% for induction and 1-2.5% maintenance during preparation and scanning) and injected intravenously with a mean dose of 16.72 ± 1.60 MBq [ 18 F]FDG via a custom-made micro catheter placed in a tail vein. After an uptake period of 60 min, mice were imaged in prone position for 15 min. During the imaging session, respiration of the mice was controlled and core body temperature was constantly kept at 38 °C via a heating pad. For attenuation correction and anatomical reference whole body CT scans were acquired. Each PET data set was corrected for random coincidences, dead time, scatter and attenuation. CT images were reconstructed with Feldkamp algorithm. PET data were first Fourier rebinned into 41 a 2D dataset from which real-space images were reconstructed with an ordered subset expectation maximization (OSEM) algorithm with 16 subsets and 4 iterations 41,42 . Data were decay-corrected to the time of injection.   . For image analysis the whole liver was segmented and the volume calculated to detect hepatomegaly due to parasite egg deposition. In a second step, multifocal hyper intense spots around the hepatic lesions, referable to worm egg deposition, were manually counted in three consecutive slices (Fig. 1) and the number of lesions per ten cubic millimeters were calculated.
Image analysis PET/CT. PET/CT datasets were transferred to the PMOD software. For the analysis of the PET images the CT images were first co-registered by rigid fusion in order to correct the minor motion differences of the animal due to the deep anesthesia breathing or bed translation in the gantry from PET to the CT. Next, on coronal PET images volumes of interest (VOIs) around abdomen, spleen, kidneys, bladder, portal vein and heart were manually drawn with the aid of visual inspection of co-registered CT images (Fig. 4). In the same way, a further VOI (0.2 × 0.2 × 0.2 cm) was placed in the liver of each animal. The VOIs of heart, bladder and kidneys were used to accurately separate the uptakes of [ 18 F]-FDG in the VOIs of abdomen (VOI abdomen ), spleen (VOI spleen ), portal vein (VOI portal vein ) and liver (VOI liver ) from each animal. Quantification of PET data was expressed as mean percent injected dose per cubic centimeter (%ID/cc) for each animal.
Assessment of infection-related organ alterations. One part of the right lobe of each liver (lobe three) was fixed in 4% neutral buffered formaldehyde solution and embedded in paraffin. Four µm thin sections were stained with hematoxylin/eosin (HE) to differentiate hepatic granulomas from unaffected liver parenchyma. To stain hepatic collagen I/III, liver slices were stained with 0.1% Sirius Red (SR) solution dissolved in aqueous saturated picric acid for 1 h and washed in 0.5% acetic acidified water. Morphometric analysis of granuloma size and of Sirius red positive areas were performed using ImageJ software (v1.47v; National Institute of Health). www.nature.com/scientificreports/