Preconditioning donors with corticosteroids improves early lung graft immunity

Abstract

Background: Preclinical studies have recently revealed the critical role of innate immunity in determining lung transplantation outcomes. Although the International Society for Heart and Lung Transplantation recommends high-dose corticosteroid administration to donors, this practice is inconsistently applied worldwide. Investigating its impact on the donor lung's innate immune response - an unexplored area - could provide valuable evidence to support adoption of donor preconditioning with corticosteroids, beyond their traditional administration to recipients.

Method: We used a cross-circulatory pig platform that consists of a donor lung placed extracorporeally and connected to the circulation of a recipient pig whose leukocytes are fluorescently labeled.

Results: Donor preconditioning - compared to recipient's treatment alone - reduced the presence of CD3^pos T-cells in the graft from both the donor and recipient, and enhanced the anti-inflammatory profile of alveolar macrophages, at least during the first 10 hours of donor-recipient interaction. The alveolar macrophages isolated from corticosteroid-preconditioned pig lungs exhibited decreased gene expression of T-cell-attracting chemokines during the 10-hour reperfusion period, correlating with the reduced T-cell infiltration. Similarly, human lung macrophages showed lower expression of these T-cell-attracting chemokines and higher anti-inflammatory profiles with corticosteroid treatment.

Conclusion: Our results show that the early immune status of lung grafts is improved by treating donors with corticosteroids through macrophage-targeted mechanisms. This finding provides an immunological rationale for expanding the implementation of donor preconditioning with corticosteroids.

Keywords: T cells; corticosteroids; innate immunity; lung transplantation; macrophages; preconditioning; translational research.

Figure 1

Experimental set up. (A) Pigs (donors (D) and recipients (R)) were split in 3 groups: UT includes donors and recipients that received no treatment, CR includes recipients that received 20 mg/kg methylprednisolone 60 min before cross-circulation, and CDR includes donors that received 20 mg/kg methylprednisolone 12 h before lung procurement and recipients that received 20 mg/kg methylprednisolone 60 min before cross-circulation. (B) The cross-circulation’s experimental set up has been previously described in (10). CFSE (25 mg) was injected 30 min before cross-circulation, and it was shown to label the blood cell compartment (10). Blood and lung tissue biopsies were collected at different time points. (C) Flow cytometry profile of isolated cells from a lung biopsy 6 h after the onset of cross-circulation; the gates show the CFSE^pos cells (recipient (R)) and CFSE^neg cells (donor (D)).

Figure 2

Preconditioning the donor with CS modifies the composition of the cell subsets recruited from the recipient in the lung graft. Cells were gated as presented in the workflow ( Additional File 3 ) and the percent among CFSE^pos cells is reported. Each pig is labeled with a unique colored symbol throughout the paper (UT no treatment in red, CR corticosteroids to recipient only in black, CDR corticosteroids to donor and recipient in blue). When the log-transformed data followed a normal distribution, a paired t-test was performed to compare the data between timing and an unpaired t-test was performed to compare the data between groups; alternatively, a Wilcoxon test was performed. The p-values were corrected for multiple testing, * < 0.05, ** < 0.01, *** < 0.001, NS stands for non-significant, p-values > 0.05 and < 0.2 are indicated. The mean and sd values are reported in Additional File 9 .

Figure 3

Preconditioning the donor with CS modifies the representation of the donor cell subsets in the lung graft. Cells were gated as presented in the workflow ( Additional File 3 ) and the percent among CFSE^neg cells is reported. Each pig is labeled with a unique colored symbol throughout the paper (UT no treatment in red, CR corticosteroids to recipient only in black, CDR corticosteroids to donor and recipient in blue). When the data followed a normal distribution, a paired t-test was performed to compare the data between timing and an unpaired t-test was performed to compare the data between groups, alternatively a Wilcoxon test was performed. The p-values were corrected for multiple testing, * < 0.05, ** < 0.01, *** < 0.001, NS stands for non-significant, p-values > 0.05 and < 0.2 are indicated. The mean and sd values are reported in Additional File 9 .

Figure 4

Reduction of donor CD3^pos T cells in lungs by the corticosteroid treatment. (A) Representative photograph of immunohistochemical CD3 staining of untreated (UT group, left) and corticosteroid-treated (right, CDR group) donor pig lungs. The lung samples were obtained upon pig donor death, 12 h post-methylprednisolone injection (CDR). (B) The mean number of CD3+ T cells per mm² in 4 pigs per group was calculated from ten randomly-selected high power field (HPF, 0.1 mm² area) per slide, 3 slides per sample. As the data distribution did not pass the normality test, a two tailed Wilcoxson test was performed. (*, p-value < 0.05).

Figure 5

Cytokine gene expression ratios in the recipient CD14^pos and CD16^pos CFSE^pos MoCs and in AMs upon cross-circulation and effects of CS treatment. Gene expression arbitrary values were calculated from RT-qPCR data normalized to a house keeping gene (RSP24) and to an internal calibrator, using the 2^−ΔΔCT method; the gene expression data were obtained from flow cytometry sorted CFSE^posCD172A^posCD14^pos cells, CFSE^posCD172A^posCD16^pos cells and AMs from the UT (3 pigs, red), CR (4 pigs, black) and CDR groups (4 pigs, blue). A ratio between the IL10 gene expression values and the different inflammatory cytokine values was calculated. The RT-qPCR 2^−ΔΔCT of TNFA, IL10, CXCL8, CCL2, IL6 results are shown in Supplementary File 12 . As the log-transformed values passed the normality test, a paired t-test was used to identify statistically significant differences between timings and an unpaired t-test was used to identify statistically significant differences between groups. The p-values were corrected for multiple testing, * < 0.05, ** < 0.01, *** < 0.001, NS stands for non-significant, p-values > 0.05 and < 0.2 are indicated.

Figure 6

Macrophages are major expressors of CXCL9,10 and 11 transcripts in the pig lung. (A) Single cell RNA-seq was conducted with the 10X genomics technology (v3 chemistry) on 20 x 10⁴ isolated cells from a pig lung (post isolation with anti-MHC class II immunobeads). The data were pre-processed for a high quality transcriptome (> 500 genes and < 50000 reads per cell), clustered using the Seurat package and projected into a UMAP reduced space. The clusters were annoted based on the top marker genes extracted from each cluster using the Seurat FindMarkers function ( Additional File 5 ). (B) The gene expressions for CD163, CCL5, CXCL9, 10 and 11 are displayed in the UMAP space with the red color representing the maximal expression level and the gray color representing absence of expression. The monocyte/macrophage cell types are indicated by a green dashed line.

Figure 7

Preconditioning the donor with CS reduces the gene expression of T-cell-attracting chemokines in AMs. In (A) the AMs from lung biopsies collected at T0H of 4 untreated pigs and of 4 corsticosteroid treated pigs were sorted with flow cytometry. In (B) the AMs from lung biopsies collected after T6H and T10H cross-circulation in the UT group (3 pigs), CR group (3 pigs) and CDR group (4 pigs) were sorted with flow cytometry. In (A, B) the CFSE^negCD163^hi CD172A^hi SSC^hi gating was used for the AM sorting, the RNA of the sorted cells was extracted and subjected to RT-qPCR for CXCL9, CXCL10, CXCL11 and CCL5 detection. Gene expression arbitrary values were calculated from RT-qPCR data normalized to a house keeping gene (RSP24) and to an internal calibrator, using the 2^−ΔΔCT method. As the log-transformed values passed the normality test, a paired t-test was used to identify statistically significant differences between timings and an unpaired t-test was used to identify statistically significant differences between groups. The p-values were corrected for multiple testing, * < 0.05, ** < 0.01, NS stands for non-significant, p-values > 0.05 and < 0.2 are indicated.

Figure 8

Effects of CS treatment on T-cell-attracting chemokine and inflammatory gene expression in human lung cells and macrophages. (A) Human lung cells from 8 patients were cultured without or with 20 µg/ml methylprednisolone (CS). The CXCL10 protein levels in the supernatants were measured by ELISA after 12 and 42 h. A specific shape symbol is attributed to each patient. (B, C) Human lung cells from 5 patients were cultured for 12 h without or with 20 µg/ml methylprednisolone, then detached with accutase and CD172A^pos macrophages were immunomagnetically sorted and then lysed for RNA extraction. Gene expression arbitrary values were calculated from RT-qPCR data normalized to a house keeping gene (RSP18) and to an internal calibrator, using the 2-DDCT method. A specific shape symbol is attributed to each patient. In B, gene expression data for T-cell attracting chemokines and CD172A are shown. In C, gene expression ratios for inflamamatory genes are shown. When the log-transformed values passed the normality test, a paired t-test was used to identify statistically significant differences (CCL5, CXCL9, CXCL10, IL10:cytokine ratios); alternatively a wilcoxson test was used (CXCL11). The p-values are reported as * < 0.05, ** < 0.01. *** < 0.005. NS, non significant.