Extracted RNA from macrophages was analysed. with a standard false discovery rate (FDR) of 0.05; most of the genes showed either significant up- or down-regulated expression in lungs exposed to 90% O2. Following the hyperoxia, the highest level of gene up-regulation was a (+3.64) Log2 fold change, found for Cyp4a12b. In contrast, the greatest degree of gene down-regulation was the (-5.21) Log2 fold change found for Spink1.
The administration of hMSCs had an extensive impact on the changes in gene expression level. A total of 421 genes were found to be up-regulated and 651 genes were down-regulated following hMSCs therapy. The changes were expressed as Log2 fold-changes. In addition, the highest level of gene up-regulation was a (+2.72) Log2 fold change, found for Lpar3, and the lowest level of gene up-regulation was the (+0.30) Log2 fold change, found for Btk. In contrast, a total of 651 genes were down-regulated. The greatest degree of gene down-regulation was the (-4.28) Log2 fold change found for Hspa1b, and the smallest was the (-0.26) Log2 fold change, found for Mapre1.
A total of 139 genes were shown to be up-regulated by hyperoxia and then significantly down-regulated following the administration of hMSCs. Of these genes, the highest level of up-regulation was a (+2.74) Log2 fold change, found for Rhov, and the lowest level of gene up-regulation was a (+0.15) Log2 fold change, found for Tcf25, as shown in Figure 2. A list of the 30 most up-regulated genes (among genes up-regulated by hyperoxia) following hMSCs administration is shown in Figure 3.
Macrophage total-RNA gene pathways
The use of integrity pathway analysis (IPA) showed the ability of hMSCs to ameliorate the effects of hyperoxia following the analysis of specific gene expression combinations, that likely contribute to a number of physiological and pathological consequences, as shown in Figure 4.
Gene pathway analysis showed that hyperoxia resulted in an up-regulation of expression that may indicate an increased risk of a number of pathological conditions, but the risk of these was significantly reduced by hMSCs administration. These pathological conditions include lung inflammation (ACE, ARIH2, C5AR1, CAV1, CCL22, CCR1, CD44, CD86, CFLAR, CSF2RB, CYP51A1, HLAA, HRH1, ICAM1, IL12B, IL1RN, IMPDH1, LILRB4, MAP2K3, MAP3K14, NFKB1, NFKBIZ, NLRC4, PLAT, PTGS2, RELB, SLC7A11, SQSTM1, TICAM1, TLR2, TRAF4, TREM1, and VDR), and lung structural damage (ACE, C5AR1, CAV1, CCR1, CD44, CYP51A1, ICAM1, IL12B, MAP2K3, NFKB1, PTGS2, SOCS3, TICAM1, TLR2, and TREM1).
Analysing macrophage-related gene expression and macrophage specific gene pathways revealed the specific contribution of macrophages in hyperoxic lung injury, and the effectiveness of hMSCs in altering macrophage function and phenotypes. Using NGS, macrophage total-RNA gene expression was analysed to assess specific genes characteristic of M1-proinflammatory macrophages, including Cd86, Stat1, Socs3, Slamf1, Tnf, Fcgr1, Il12b, Il6, Il1b, and Il27ra. These genes were up-regulated in lung macrophages isolated from mice exposed to hyperoxia compared to mice receiving hMSCs following hyperoxia, as shown in Figures 5 and 6. Genes related to M2 anti-inflammatory macrophages showed an overall down-regulation in macrophages isolated from lungs of hyperoxic mice receiving hMSCs. These genes included Arg1, Stat6, Mrc1, Il27ra, Chil3, and Il12b. Hyperoxia alone led to a significant up-regulation in M1-associated gene expression together with the down-regulation of M2-associated gene expression. The administration of hMSCs was found to inhibit the expression of M1 macrophage genes and up-regulate M2 associated gene expression. The results are summarised in Figure 7.
Macrophage phenotypes and associated gene pathways
Following the analysis of gene expression, the gene pathways of macrophages were assessed using IPA. Our total RNA sequencing data showed changes in the expression of macrophage-associated genes. Investigating the contribution of these gene expression changes in macrophages that have associated effects on lung morphology and the development of fibrosis could clarify the mechanism of the macrophage’s response to hyperoxia with and without hMSCs therapy.
The IPA showed the ability of hMSCs to alter macrophage gene expression. These altered genes can be summarised as:
- Genes affecting macrophage infiltration (ICAM1, IL12B, IL1A, IL1RN, KRAS, MALT1, MAP3K14, MAP3K8, NDRG1, NFKB1, NFKB2, NFKBIZ, NINJ1, PDK4, PLAT, PLEC, PRDM1, PTGS2, SOCS3, SPP1, and STIM1).
- Genes affecting macrophage function (ACE, CCL22, CD40, CD44, HLA-A, ICAM1, IL1RN, INHBA, MAP2K3, PDK4, PLAT, PTGS2, TGM2, TICAM1, and TLR2).
- Genes affecting macrophage recruitment (BRAF, C5AR1, CCL22, CCR1, CD40, CD44, ETS1, ICAM1, IL1RN, KRAS, MAP3K8, SPP1, TLR2, and VDR).
- Genes affecting macrophage development (CAV1, DCSTAMP, IL1A, IL1RN, MAP3K8, NFKB1, NFKB2, PRDM1, and RELB).
- Genes affecting macrophage maturation (BRAF, C5AR1, CAV1, CCL22, CCR1, CD44, CRKL, ICAM1, IL12B, IL1A, IL1RN, LITAF, MAP3K8, NDRG1, NINJ1, NLRC4, PLEC, PRDM1, PTGS2, SPP1, TICAM1, and TLR2); and
- Altered are genes affecting the immune response of leukocytes (AFF1, BHLHE40, BRAF, C5AR1, CD80, CD83, CD86, CFLAR, DUSP10, ETS1, ICAM1, ICOSLG/LOC102723996, IL10RA, IL12B, IRF1, JAG1, LEPR, LFNG, LILRB4, MALT1, MMP19, PRDM1, PTGS2, RELB, SOCS3, ZBTB1, and Zfp35) and infiltration (CCL22, CCNT1, CCR1, CCR7, CD40, CD44, CD80, CD86, HLA-A, KRAS, MALT1, NFKB1, NFKB2, PTGS2, TNIP1, and ZC3H12A), activation of M2 macrophages and the involvement of the macrophages in the deposition of lung collagens Alpha1, I,II, and IV (IL1A, IL1RN, MAP3K8, and MMP19).
Small-RNA sequencing
We used small-RNA sequencing to provide further information on the effects of hyperoxia and the administration of hMSCs on the small RNAs transcriptomes that were evident in the CD45+CD11b+CD11c+ macrophages. The 0.05 FDR was also applied for the small-RNA analysis. A total of 1,098 transcriptomes were expressed as either significantly up- or down-regulated following the administration of hMSCs in mice exposed to 90% O2. Importantly, these results indicated different types of small RNA transcriptomes expressed that included small nucleolar RNAs (snoRNA and snRNA), ribosomal-RNA (r-RNA), miscellaneous RNA (misc-RNA), microRNA (mi-RNA), large intergenic non-coding RNAs (linc-RNAs) and mitochondrial- RNA (mt-tRNAs).
Up- and down-regulated small-RNA expressiontranscriptomes following hyperoxia
A total of 378 small-RNA transcriptomes were up-regulated. The highest-level of transcriptome up-regulation was the (+5.04) Log2 fold change found for Gm24144-snRNA and the lowest level of transcriptome up-regulation was the (+0.24) Log2 fold change found for Snord91a-snoRNA. Furthermore, a total of 676 small-RNA transcriptomes were down-regulated. The greatest level of transcriptome down-regulation was a (-7.98) Log2 fold change found for Snord88c-snoRNA and the lowest level down transcriptomes down-regulation was (-0.15) Log2 fold change found for Gm24357-snoRNA.
Alterations in transcriptome expression induced by hMSCs in small-RNA transcriptomes that were up and down-regulated by hyperoxia
A total of 42 small-RNA transcriptomes were shown to be up-regulated by hyperoxia and then significantly inhibited by hMSCs. In these transcriptomes, the highest level of up-regulation was a (+4.71) Log2 fold change, found for A530013C23Rik, and the lowest level of transcriptomes up-regulation was a (+0.56) Log2 fold change, found for Gm22886. In addition, a total of 53 transcriptomes were shown to be down-regulated by hyperoxia and then subsequently significantly re-expressed in lungs from mice exposed to hyperoxia and administered hMSCs. The greatest degree of down-regulation in these genes was a (-3.40) Log2 fold change found for Gm24920 and the smallest degree of down-regulation was a (-0.42) Log2 fold change found for Gm13571.