Transcriptional Profiling Suggests T Cells Cluster around Neurons Injected with Toxoplasma gondii Proteins

Abstract

Toxoplasma gondii’s tropism for and persistence in the central nervous system (CNS) underlies the symptomatic disease that T. gondii causes in humans. Our recent work has shown that neurons are the primary CNS cell with which Toxoplasma interacts and which it infects in vivo. This predilection for neurons suggests that T. gondii’s persistence in the CNS depends specifically upon parasite manipulation of the host neurons. Yet, most work on T. gondii-host cell interactions has been done in vitro and in nonneuronal cells. We address this gap by utilizing our T. gondii-Cre system that allows permanent marking and tracking of neurons injected with parasite effector proteins in vivo. Using laser capture microdissection (LCM) and RNA sequencing using RNA-seq, we isolated and transcriptionally profiled T. gondii-injected neurons (TINs), Bystander neurons (nearby non-T. gondii-injected neurons), and neurons from uninfected mice (controls). These profiles show that TIN transcriptomes significantly differ from the transcriptomes of Bystander and control neurons and that much of this difference is driven by increased levels of transcripts from immune cells, especially CD8+ T cells and monocytes. These data suggest that when we used LCM to isolate neurons from infected mice, we also picked up fragments of CD8+ T cells and monocytes clustering in extreme proximity around TINs and, to a lesser extent, Bystander neurons. In addition, we found that T. gondii transcripts were primarily found in the TIN transcriptome, not in the Bystander transcriptome. Collectively, these data suggest that, contrary to common perception, neurons that directly interact with or harbor parasites can be recognized by CD8+ T cells. IMPORTANCE Like other persistent intracellular pathogens, Toxoplasma gondii, a protozoan parasite, has evolved to evade the immune system and establish a chronic infection in specific cells and organs, including neurons in the CNS. Understanding T. gondii’s persistence in neurons holds the potential to identify novel, curative drug targets. The work presented here offers new insights into the neuron-T. gondii interaction in vivo. By transcriptionally profiling neurons manipulated by T. gondii, we unexpectedly revealed that immune cells, and specifically CD8+ T cells, appear to cluster around these neurons, suggesting that CD8+ T cells specifically recognize parasite-manipulated neurons. Such a possibility supports evidence from other labs that questions the long-standing dogma that neurons are often persistently infected because they are not directly recognized by immune cells such as CD8+ T cells. Collectively, these data suggest we reconsider the broader role of neurons in the context of infection and neuroinflammation.

A transcriptional network required for Toxoplasma gondii tissue cyst formation is dispensable for long-term persistence

“…The inflammatory signature of mice infected with ΔROCY1 and ΔBFD1 parasites was also significantly different as determined using gene set enrichment analysis (GSEA; (Subramanian et al ., 2005)), where pathways like “HALLMARK INFLAMMATORY RESPONSE” and “HALLMARK IL2 STAT5” were similarly and significantly enriched based on a false-discovery rate of 0.05 (Figure S5D). We then used the informatics tool Cibersort (Newman et al ., 2019) which can be used to infer percentages of different cell populations in bulk RNAseq data and that has recently been applied to brain RNAseq data from T. gondii -infected mice (Merritt et al ., 2020). Given our interest in the inflammatory profile of mice infected with wild type and cyst-deficient T. gondii (ΔROCY1 and ΔBFD1), we used the precomputed “LM22” as the signature gene file and used it to estimate the relative proportions of 22 immune cell types using our brain RNAseq data.…”

“…We then used the informatics tool Cibersort (Newman et al, 2019) which can be used to infer percentages of different cell populations in bulk RNAseq data and that has recently been applied to brain RNAseq data from T. gondii-infected mice (Merritt et al, 2020). Given our interest in the inflammatory profile of mice infected with wild type and cyst-deficient T. gondii (ΔROCY1 and ΔBFD1), we used the precomputed "LM22" as the signature gene file and used it to estimate the relative proportions of 22 immune cell types using our brain RNAseq data.…”

“…It has long been assumed that tissue cyst formation is a response to immune pressure within the host and is required for T. gondii persistence in the host (Kim et al, 2007;Tomita et al, 2013). In mice the primary site of persistence is within the brain where it resides within neurons and can cause lethal encephalitis when immune surveillance is altered by experimental treatment or disease (Cabral et al, 2020;Jones et al, 1996;Suzuki et al, 1989;Takashima et al, 2008). A less than complete understanding of how this process is regulated transcriptionally and executed by downstream effectors presents a barrier to the development of new therapies targeting this drug-refractory life stage and limits our ability to understand the importance of cyst formation itself in T. gondii infection, persistence and transmission to definitive or intermediate hosts.…”

Deficiency in astrocyte CCL2 production reduces neuroimmune control of Toxoplasma gondii infection

“…found that CCL2 expression is upregulated in regions of the brain cells without significant interaction with T . gondii , but was found within 120 μm of a parasite-interacted cell [ 49 ]. In the periphery, innate immune recognition of T .…”

“…However, it is also clear that direct invasion of host cells in the brain by T. gondii is not required for CCL2 production. Merritt et al found that CCL2 expression is upregulated in regions of the brain cells without significant interaction with T. gondii, but was found within 120 μm of a parasite-interacted cell [49]. In the periphery, innate immune recognition of T. gondii profilin by TLR11/12 induces CCL2 production and monocyte recruitment [33].…”