Dual RNA Sequencing of Mycobacterium tuberculosis-Infected Human Splenic Macrophages Reveals a Strain-Dependent Host-Pathogen Response to Infection

Abstract

Tuberculosis (TB) is caused by Mycobacterium tuberculosis (Mtb), leading to pulmonary and extrapulmonary TB, whereby Mtb is disseminated to many other organs and tissues. Dissemination occurs early during the disease, and bacteria can be found first in the lymph nodes adjacent to the lungs and then later in the extrapulmonary organs, including the spleen. The early global gene expression response of human tissue macrophages and intracellular clinical isolates of Mtb has been poorly studied. Using dual RNA-seq, we have explored the mRNA profiles of two closely related clinical strains of the Latin American and Mediterranean (LAM) family of Mtb in infected human splenic macrophages (hSMs). This work shows that these pathogens mediate a distinct host response despite their genetic similarity. Using a genome-scale host-pathogen metabolic reconstruction to analyze the data further, we highlight that the infecting Mtb strain also determines the metabolic response of both the host and pathogen. Thus, macrophage ontogeny and the genetic-derived program of Mtb direct the host-pathogen interaction.

Keywords: Mycobacterium tuberculosis; clinical strains; dual RNA-seq; human; metabolic reconstruction; splenic macrophages.

Figure 1

UT127 and UT205 clinical isolates of Mtb trigger a distinct transcriptomic response in human splenic macrophages upon infection. (A) General outline of the dual RNA-seq strategy. 5–20 × 10⁶ human splenic macrophages (hSMs) were infected for 6 h at an MOI of 10:1 with the two closely related clinical strains of the Latin American and Mediterranean (LAM) family of Mtb, UT127, and UT205. Upon total RNA-recovery, samples were RNA-seq sequenced and reads mapped to the human and Mtb genomes (see Section 4). (B) Principal component analysis (PCA) of Mtb transcriptome cluster gene expression of UT127 from UT205 inside infected hSMs. Differentially expressed genes for UT127 and UT205 were obtained by comparing the same strains cultured in a carbon-poor medium (Sauton’s). (C–E) Deduced functional categories using the GO-seq algorithm on the immune genes obtained from the ImmPORT database. (see Section 4). Infection of hSMs with UT127 was associated with 645 over-represented biological processes, while infection with UT205 was associated with 585 over-represented processes. (C) Illustrates selected common biological processes differentiating the hSM response to UT127 and UT205. (D,E) Selected unique immune biological processes over-represented in response to UT127 (D) or UT205 infection (E). Blue and red bars represent biological processes (C–E) associated with the hSMs transcriptional response to Mtb UT127 or UT205.

Figure 2

Gene expression response of UT127 and UT205 Mtb strains to hSMs adapt to hSMs by expressing distinct transcriptomic profiles. (A) Bar graph displaying the percentage of genome coverage of the reads obtained from intracellular Mtb. Reads from intracellular Mtb UT127 and UT205 of each triplicate sample were mapped to the Mtb H37Rv genome (see Section 4). Reads mapped to 80–90% of the Mtb genome, depending on the sample. (B) Bar chart indicating the amount of differentially expressed genes (DEGs) for each condition. (C) Venn diagram displaying the sharing of differentially expressed up-regulated and downregulated genes of Mtb UT127 and UT205 in infected hSMs. (D,E) Volcano plots of the genes expressed by Mtb strain UT127 (D) and UT205 (E) within 6 h of infection of human splenic macrophages. Macrophages were infected for 6 h with an MOI of 10:1. Differentially expressed genes were selected with 1.5 Log₂FC, p < 0.05, FDR < 0.05. Grey lines discriminate differentially expressed genes (DEGs). The most highly up-regulated and downregulated genes are shown.

Figure 3

Protein-Protein interaction network of commonly up-regulated genes expressed by Mtb within 6 h of hSMs infection. (A) Protein-Protein interaction network of commonly up-regulated differentially expressed genes (n = 165) as deduced from the String database. The 2 top GO categories, DIM/DIP cell wall layer assembly (FDR 0.00031) and response to hypoxia (FDR 0.00031), are encircled. (B) Significant biological processes (FDR < 0.05) associated with the 165 commonly up-regulated differentially expressed genes by hSMs infected for 6 h with M. tuberculosis.

Figure 4

UT127 and UT205 Mtb strains adapt to hSMs by expressing distinct transcriptomic profiles. (A,D) Protein-Protein interaction networks associated with the up-regulated DEGs of UT127 (n = 182) (A) and UT205 (n = 126) (D) as determined by the String database. Encircled clusters identify genes associated with selected biological processes. (A,D). Top significant biological processes (FDR < 0.05) based on Gene Ontology (GO) of up-regulated genes of UT127 ((B), blue bars) and UT205 ((E), red bars) within 6 h of infection of hSMs. (C,F) Heat maps representing the expression (Log₂ Fold) and significance (FDR, False Discovery Rate, right) of the genes associated with the GO categories, siderophore biosynthetic process, response to metal ion, transition metal homeostasis, and cellular response to iron starvation of the UT127 strain, compared to the UT205 strain (C), and the GO categories, alcohol metabolic process, cholesterol metabolic process, steroid metabolic process, and cholesterol metabolic process of the UT205 strain, compared with the UT127 strain (F).

Figure 5

A transcriptomic-based reconstruction of the host–pathogen early interaction indicates a distinct interaction. (A,B) Highly expressed and lowly expressed reactions in the Host–Pathogen Metabolic Network with Mtb UT127 and UT205 as a pathogen. (A) Number of highly expressed (RH) and lowly expressed (RL) reactions in UT127- and UT205- infected human spleen macrophages metabolic pathways. (B) Number of highly expressed (RH) and lowly expressed (RL) reactions in Mtb UT127 and UT205 metabolic pathways. Numbers inside the colored squares indicate the number of reactions. (C,D) Cumulative fluxes derived from the integration of transcriptomics data into the Host–Pathogen Metabolic Network. (A) Cumulative flux by Mtb-infected hSMs metabolic pathways; (B). Cumulative flux by Mtb UT127 and UT205 metabolic pathways.

Figure 6

Splenic macrophages infected with M. tuberculosis UT127 and UT205 display M1- and M2-like characteristics. Within 6 h of infection with Mtb, hSMs express a mixture of M1 and M2 polarization markers. Red arrows represent genes or cellular processes associated with the M1 macrophage polarization. Red arrows pointing down represent genes or cellular processes associated with the M2 macrophage polarization. There is no clear association with M1 or M2 polarization in hSMs infected with Mtb UT127 or UT205. The direction of the upward and downward arrows indicate upregulated or downregulated genes, respectively.