Intrahepatic microbes govern liver immunity by programming NKT cells
Joshua C. Leinwand, Bidisha Paul, Ruonan Chen, Fangxi Xu, Maria A. Sierra, Madan M. Paluru, Sumant Nanduri, Carolina G. Alcantara, Sorin A.A. Shadaloey, Fan Yang, Salma A. Adam, Qianhao Li, Michelle Bandel, Inderdeep Gakhal, Lara Appiah, Yuqi Guo, Mridula Vardhan, Zia Flaminio, Emilie R. Grodman, Ari Mermelstein, Wei Wang, Brian Diskin, Berk Aykut, Mohammad Khan, Gregor Werba, Smruti Pushalkar, Mia McKinstry, Zachary Kluger, Jaimie J. Park, Brandon Hsieh, Kristen Dancel-Manning, Feng-Xia Liang, James S. Park, Anjana Saxena, Xin Li, Neil D. Theise, Deepak Saxena, George Miller
Joshua C. Leinwand, Bidisha Paul, Ruonan Chen, Fangxi Xu, Maria A. Sierra, Madan M. Paluru, Sumant Nanduri, Carolina G. Alcantara, Sorin A.A. Shadaloey, Fan Yang, Salma A. Adam, Qianhao Li, Michelle Bandel, Inderdeep Gakhal, Lara Appiah, Yuqi Guo, Mridula Vardhan, Zia Flaminio, Emilie R. Grodman, Ari Mermelstein, Wei Wang, Brian Diskin, Berk Aykut, Mohammad Khan, Gregor Werba, Smruti Pushalkar, Mia McKinstry, Zachary Kluger, Jaimie J. Park, Brandon Hsieh, Kristen Dancel-Manning, Feng-Xia Liang, James S. Park, Anjana Saxena, Xin Li, Neil D. Theise, Deepak Saxena, George Miller
View: Text | PDF
Research Article Hepatology Microbiology

Intrahepatic microbes govern liver immunity by programming NKT cells

Abstract

The gut microbiome shapes local and systemic immunity. The liver is presumed to be a protected sterile site. As such, a hepatic microbiome has not been examined. Here, we showed a liver microbiome in mice and humans that is distinct from that of the gut and is enriched in Proteobacteria. It undergoes dynamic alterations with age and is influenced by the environment and host physiology. Fecal microbial transfer experiments revealed that the liver microbiome is populated from the gut in a highly selective manner. Hepatic immunity is dependent on the microbiome, specifically the bacteroidetes species. Targeting bacteroidetes with oral antibiotics reduced hepatic immune cells by approximately 90%, prevented antigen-presenting cell (APC) maturation, and mitigated adaptive immunity. Mechanistically, our findings are consistent with presentation of bacteroidetes-derived glycosphingolipids to NKT cells promoting CCL5 signaling, which drives hepatic leukocyte expansion and activation, among other possible host-microbe interactions. Collectively, we reveal a microbial/glycosphingolipid/NKT/CCL5 axis that underlies hepatic immunity.

Authors

Joshua C. Leinwand, Bidisha Paul, Ruonan Chen, Fangxi Xu, Maria A. Sierra, Madan M. Paluru, Sumant Nanduri, Carolina G. Alcantara, Sorin A.A. Shadaloey, Fan Yang, Salma A. Adam, Qianhao Li, Michelle Bandel, Inderdeep Gakhal, Lara Appiah, Yuqi Guo, Mridula Vardhan, Zia Flaminio, Emilie R. Grodman, Ari Mermelstein, Wei Wang, Brian Diskin, Berk Aykut, Mohammad Khan, Gregor Werba, Smruti Pushalkar, Mia McKinstry, Zachary Kluger, Jaimie J. Park, Brandon Hsieh, Kristen Dancel-Manning, Feng-Xia Liang, James S. Park, Anjana Saxena, Xin Li, Neil D. Theise, Deepak Saxena, George Miller

×

Figure 8

The microbiome promotes the expansion and maturation of liver myeloid cells.

Options: View larger image (or click on image) Download as PowerPoint
The microbiome promotes the expansion and maturation of liver myeloid ce...
(A) Six-week-old female mice were treated with either broad-spectrum antibiotics or vehicle. CD45+ liver leukocytes were purified by FACS and analyzed by single-cell RNA-Seq. The distribution of cellular clusters was determined using the t-SNE algorithm. Each cluster is identified by a distinct color. Percentage of cellular abundance in each cluster in each respective cohort is depicted in pie charts. (B–E) The frequency of diverse APC subsets among CD45+ liver leukocytes in mice treated with broad-spectrum antibiotics or vehicle was determined by flow cytometry. Data are representative of experiments performed more than 4 times in replicates of 5. **P < 0.01; ***P < 0.001. (F–H) Expression of activation markers in liver APC subsets in mice treated with broad-spectrum antibiotics or vehicle was determined by flow cytometry and is shown in spider plots. Data are representative of experiments performed more than 4 times in replicates of 5. (I) Volcano plot showing differential gene expression in the hepatic myeloid cell cluster for mice treated with broad-spectrum antibiotics versus vehicle based on single-cell RNA-Seq used for A. (J) Violin plots comparing normalized log expression of select genes in the hepatic myeloid cell cluster for each treatment group shown in A. *P < 0.05; **P < 0.01; ****P < 0.0001. (K) Changes in upstream regulators in the hepatic myeloid cell cluster for each treatment group shown in A. *P < 0.05; **P < 0.01; ***P < 0.001; ****P < 0.0001. (L and M) Naive mice were immunized twice with OVA-pulsed hepatic macrophages harvested from antibiotic- or vehicle-treated donors. CD8+ T cell activation was determined 1 week after the last immunization by measuring the frequency of OVA-dextramer+ CTLs (L) and their surface phenotype (M). This experiment was performed twice. n = 5/group. *P < 0.05; **P < 0.01. (N) Liver macrophages harvested from antibiotic- or vehicle-treated donors were pulsed with α-GalCer and used to immunize naive mice. The frequency of CD1d-restricted NKT cells was determined 1 week after the second immunization by flow cytometry. This experiment was performed twice. n = 5/group. *P < 0.05.