Tuberculosis Imaging Program

National Institute Of Allergy And Infectious Diseases

Investigators

Linked publications, trials & patents

Abstract

The major activities of this research project in the past have centered around using our optimized methods and procedures for imaging rhesus macaques, NZW rabbits, and common marmosets on PET/CT scanners while conducting a variety of chemotherapy and basic immunology experiments in Mtb models. We have also optimized the imaging of mice on the LFER scanner to fulfill the need in the TRS, but because of the resolution and speed of the unit, it is not ideal for mice. The experiments in 2022 included various Mtb infections of rhesus macaques with TBU and IIIU and the Vaccine Research Center (VRC). Two Rabbit, four marmoset, and two mouse infections with Mtb with TRS. We have also conducted two Sars CoV-2 infection experiments in rhesus macaques with the TBS. In these experiments we use our optimized HU and 18F-fluorodeoxyglucose (FDG)-PET ranges to automate lesion identification. In MTB lesions the HU distributed from -550 to 100 HU with very few voxels above 100 HU. When possible, we applied an automated method that segregates low- and high-density ranges using a whole lung technique. COVID-19 lesions are much softer and often have only slightly elevated HU (-600) associated with a larger region of elevated SUV. Our analysis of PET images has been based on changes in total activity in the lung and in individual lesions where the SUV>2.5 for TB lesion identification and SUV> 1.5 for COVID-19. Some specific projects that are published are mentioned below: We have conducted experiments where BAL fluid and blood was collected regularly from Mtb infected macaques so that IIIU researchers could determine the kinetics of the various innate effector cells in the blood and on airway surfaces during progressive infection. Previously they uncovered an unexpected association between eosinophils and Mtb infection. In humans with refractory tuberculosis, eosinophils were decreased in the blood but enriched in resected human lung lesions and autopsy granulomas. In 2022, they showed that eosinophils migrate into the lungs of macaques and mice as early as one week after Mtb exposure. In mice, this influx requires cell-intrinsic expression of the oxysterol receptor GPR183, which is highly expressed on human and macaque eosinophils as well (PMID: 35905725). The work has led to the discovery that eosinophils are among the earliest cells from circulation to sense and respond to Mtb infection of alveolar macrophages and revealed a role for GPR183 in the migration of eosinophils into infected lung tissue. We collaborated with the TBS to examine the possibility that HIV co-infection with MTB leads to CD4 T cell depletion in lung tissue before it is evident in blood. Mtb-infected macaques with a stable subclinical infection were inoculated with SIV, and two weeks later any changes in their infection perimeters were queried to determine if and what early changes occur after SIV infection (PMID: 35649361). PET/CT scans just prior SIV infection and 12 days later showed no significant change in lesion volume, or FDG uptake, but there was a moderate but significant increase the lesion bacteria load in SIV co-infected animals in that time. At that time Mtb-specific CD4 T cells were dramatically depleted from the SIV+ granulomas, before CD4 T cell loss was observable in blood, airways, and lymph nodes. Using live cell imaging, CD4 T cells were preferentially depleted from the granuloma core and cuff relative to B cell-rich regions. In addition, CD4 T cell motility was reduced within the granuloma, suggesting that depleted effector cells could not be replaced and control of Mtb replication in the granuloma might follow as is seen when Mtb latently infected macaques are depleted of CD4 cells. In collaboration with TRS and investigators at Rutgers University, we continued to use rabbits and MTB lesions from the rabbit model to elucidate the drug penetration into caseum as a mimic of human lesions and drug activity in this nutrient-limiting hypoxic condition. These experiments revealed limited but rapid penetration of streptomycin, amikacin, and kanamycin to the sites of TB disease and supports the development of aminoglycoside analogs with improved efficacy and tolerability (PMID: 34252307). In the case of clarithromycin, it had favorable tissue penetration in the rabbit, but lacked significant bactericidal activity in some lesion compartments. These experiments and modeling indicated that its clinical activity is limited by pharmacodynamic, rather than pharmacokinetic, factors (PMID: 35099272) In Sars-CoV-2 investigations with TBU, they used rhesus macaques to model protective primary immune responses in tissues during COVID-19 (PMID: 35271298). Using a high dose of a pre-Alpha strain of Sars-CoV-2, we found FDG-avid lung abnormalities of mild to moderate opacity (HU range -600 to -100) that peaked on day 3 and waned by day 9 although fibrin+ microthrombi were still detectable in the lungs on day 10 post-infection. Virus-specific effector CD8+ and CD4+ T cells became detectable in the BAL on days 7-10 and lung tissue on day 10. Interestingly, lung lesions identified by PET/CT scanning had significantly more of these virus-specific effector T cells than did other normal lung regions although there was no difference in viral genomes at that time. Other body regions (spleen, nasal turbinates, tonsils) were also evaluated by PET/CT in this small group of animals but no clear change in FDG uptake was noted.