Three-dimensional co-culture system using kidney organoids and alloreactive PBMC for inducing allogeneic condition

To set up our in vitro kidney transplant rejection model, which we refer to as the allogeneic condition in a culture dish, we initially conducted HLA mismatch scoring for HLA-A, B, C, DRB1, and DQB1 loci in hiPSCs employed in this study. Table S1 presents the results of high-resolution WT-hiPSC and HC PBMC HLA-typing for both class-I (A, B, C) and class-II (DRB1, DQB1) antigens using NGS technology. We established a co-culture system by combining a kidney organoid derived from WT-hiPSC with HLA-mismatched PBMC from HC for 24 h as described in the experimental workflows (Fig. 1A). For the syngeneic control group, we established an HC hiPSC line from this HC’s PBMC using a reprogramming method with episomal vectors (Fig. S1). Primary humoral alloimmune activation following allogeneic conditions in the co-culture system was assessed by quantifying the induction of HLA molecules, HLA-ABC, and -DR in whole-well cells via flow cytometric analysis. The proportion of HLA-ABC and HLA-DR-positive cells significantly increased in the allogeneic group compared to the syngeneic group (HLA-ABC, 53.2 ± 0.4% vs. 41.8 ± 0.7%, #P < 0.05 vs. syngeneic group; HLA-DR, 38.8 ± 0.7% vs. 49.1 ± 0.4%, #P < 0.05 vs. syngeneic group) (Figs. 1B-E).

Fig. 1

Vitro recapitulation of allogeneic conditions using co-culture with HLA-mismatched PBMCs and kidney organoids. (A) Schematic illustration of experimental conditions for allogeneic conditions using HLA-mismatched PBMCs and kidney organoids derived from hiPSCs. Representative flow cytometric dot plots and graphs showing the HLA-ABC + cells (B and C) and HLA-DR + cells (D and E) of kidney organoids in the syngeneic group and allogeneic group. Data is presented as mean ± SE. #P < 0.05 vs. syngeneic group

Increased HLA expression in the kidney organoids in the allogeneic condition

Next, we assessed the expression of HLA molecules in the nephron structure, including podocalyxin (PODXL) in glomerular epithelial cells, lotus tetragonolobus lectin (LTL) in the proximal tubules, and e-cadherin (ECAD) in the distal tubules. Immunofluorescence staining and examination with confocal microscopy revealed that immunoreactivity for HLA-ABC and –DR was elevated in the allogeneic group compared to the syngeneic group (Fig. 2A and B). Following co-culture with alloreactive PBMC, the kidney organoids disintegrated into single cells. These cells were then stained with a cocktail of antibodies targeting HLA-ABC, -DR, PODXL, LTL, and ECAD for flow cytometric analysis. Flow cytometric dot plots and quantitative graphs demonstrated that the percentages of HLA-ABC, -DR-positive cells containing nephron structure markers were significantly higher in the allogeneic group than in the syngeneic group (PODXL + HLA-ABC +, 20.9 ± 0.4% vs. 17.3 ± 0.5%, #P < 0.05 vs. syngeneic group; LTL + HLA-ABC +, 9.2 ± 0.4% vs. 6.0 ± 0.3%, #P < 0.05 vs. syngeneic group; ECAD + HLA-ABC +, 7.7 ± 0.2% vs. 2.8 ± 0.1%, #P < 0.05 vs. syngeneic group; PODXL + HLA-DR +, 17.8 ± 0.4% vs. 13.9 ± 0.4%, #P < 0.05 vs. syngeneic group; LTL + HLA-DR +, 3.9 ± 0.2 vs. 2.8 ± 0.2%, #P < 0.05 vs. syngeneic group; ECAD + HLA-DR +, 6.0 ± 0.2% vs. 2.0 ± 0.1%, #P < 0.05 vs. syngeneic group) (Fig. 2C-F).

Fig. 2

Expression of HLA-ABC and HLA-DR in the kidney organoids after co-culture with HLA-mismatched PBMC. After co-culturing with PBMC and maturing kidney organoids for 24 h, the kidney organoids were double labeled with antibodies against HLA-ABC or HLA-DR, as well as PODXL, under both the syngeneic condition (A) and the allogeneic condition (B). Scale bar = 50 μm. Representative flow cytometric dot plots and their quantitative graphs show the HLA-ABC + cells (C and E) and HLA-DR + cells (D and F) in the nephron markers, PODXL, LTL, and ECAD in the syngeneic group and allogeneic group. PODXL, podocalyxin; LTL, lotus tetragonolobus lectin; ECAD, e-cadherin. Data are presented as mean ± SE. #P < 0.05 vs. syngeneic group

Reduced cell viability of the kidney organoids in the allogeneic condition

We investigated the effects of HLA-mismatched PBMC on the viability of kidney organoid cells in a co-culture system. The viable cells were labeled using a fluorescence fixable viable cell dye. A histogram of cell viability displayed a reduced median value in the allogeneic group at 1633 compared with 2480 in the syngeneic group (Fig. 3A). The Annexin V and PI assay indicated that early (Annexin V positive, PI negative) and late apoptotic cells (Annexin V and PI positive) were significantly more prevalent in the allogeneic group than in the syngeneic group (Q2, 16.4 ± 0.5% vs. 7.3 ± 0.03%, *P < 0.05 vs. the syngeneic group; Q3, 18.9 ± 0.4% vs. 11.9 ± 2.4%, *P < 0.05 vs. the syngeneic group). Consistent with the result of Fig. 3A, the percentage of live cells (Annexin V and PI negative) was also lower in the allogeneic group compared to the syngeneic group (65.0 ± 0.4% vs. 75.1 ± 0.1%, *P < 0.05 vs. the syngeneic group).

Fig. 3

Effect of cell viability of the kidney organoids after co-culture with HLA-mismatched PBMC and kidney organoids. (A) Representative histogram showing cell viability measured by a fixable viable cell dye in syngeneic and allogeneic groups of kidney organoids. (B) Flow cytometry analysis of apoptotic cell death induced under allogeneic conditions using annexin V-FITC/PI staining. (C-F) The gating strategy includes single PI positive (Q1), double positive (Q2), single annexin V positive (Q3), and double negative (Q4). Data are presented as mean ± SE. *P < 0.05 vs. syngeneic group

Tacrolimus reduced the HLA expression in a dose -dependent manner

Figure 4 illustrates the impact of tacrolimus (Tac) treatment on HLA-ABC and -DR expression in kidney organoids under allogeneic conditions. Immunofluorescence staining and confocal microscopy revealed a gradual decrease in the immunoreactivity of HLA-ABC and -DR as the Tac dose was increased (Fig. 4A and B). Additionally, flow cytometric dot plots and their quantitative graphs also indicate that Tac treatment significantly reduced the percentage of HLA-ABC and -DR-positive cells across all nephron structure markers - PODXL, LTL, and ECAD- in a Tac dose -dependent manner relative to the group treated with 0 ng/mL of Tac (PODXL + HLA-ABC +, 7.1 ± 0.9% in the Tac 0 ng/mL, 5.1 ± 0.2% in the Tac 1 ng/mL, 2.2 ± 0.04% in the Tac 10 ng/mL, 1.4 ± 0.04% in the Tac 100 ng/mL; LTL + HLA-ABC +, 4.3 ± 0.1% in the Tac 0 ng/mL, 4.0 ± 0.2% in the Tac 1 ng/mL, 3.0 ± 0.1% in the Tac 10 ng/mL, 1.6 ± 0.2% in the Tac 100 ng/mL; ECAD + HLA-ABC +, 17.7 ± 0.4% in the Tac 0 ng/mL, 14.7 ± 0.5% in the Tac 1 ng/mL, 11.1 ± 0.3% in the Tac 10 ng/mL, 5.8 ± 0.5% in the Tac 100 ng/mL; PODXL + HLA-DR +, 14.0 ± 1.2% in the Tac 0 ng/mL, 11.2 ± 0.3% in the Tac 1 ng/mL, 6.0 ± 0.4% in the Tac 10 ng/mL, 3.2 ± 0.2% in the Tac 100 ng/mL; LTL + HLA-DR +, 13.9 ± 0.7% in the Tac 0 ng/mL, 10.6 ± 0.3% in the Tac 1 ng/mL, 7.2 ± 0.2% in the Tac 10 ng/mL, 3.5 ± 0.4% in the Tac 100 ng/mL; ECAD + HLA-DR +, 6.5 ± 0.1% in the Tac 0 ng/mL, 4.7 ± 0.2% in the Tac 1 ng/mL, 3.4 ± 0.1% at Tac 10 ng/mL, 1.8 ± 0.05% in the Tac 100 ng/mL, #P < 0.05 vs. 0 ng/mL of Tac group; $P < 0.05 vs. 1 ng/mL of Tac group; &P < 0.05 vs. 10 ng/mL of Tac group).

Fig. 4

Effect of tacrolimus on the expression of HLA-ABC and HLA-DR in the kidney organoids after co-culture with HLA-mismatched PBMC. Tacrolimus was administered during the co-culture of PBMC with mature kidney organoids for 24 h. Subsequently, the kidney organoids were double -labeled with antibodies against HLA-ABC (A) or HLA-DR (B) and PODXL under an allogeneic condition. Scale bars represent 50 μm. Representative flow cytometric dot plots and corresponding quantitative graphs illustrate HLA-ABC + cells (C and E) and HLA-DR + cells (D and F) among the nephron markers PODXL, LTL, and ECAD in the allogeneic group. Tac, tacrolimus; PODXL, podocalyxin; LTL, lotus tetragonolobus lectin; ECAD, e-cadherin. Data are presented as mean ± SE. #P < 0.05 vs. 0 ng/mL of Tac group; $P < 0.05 vs. 1 ng/mL of Tac group; &P < 0.05 vs. 10 ng/mL of Tac group

Influence on helper and cytotoxic T cell subsets of the PBMC under in vitro allogeneic condition

In this study, we analyzed the distribution of T cells (CD3+) and their subsets, i.e., TH (CD4+) and TC (CD8+) cells. Based on CD45RO and CCR7 surface expression levels, both TH and TC cells were further classified into four subsets: naïve (CD4+CD45RO-CCR7+), central memory (CM; CD4+CD45RO+CCR7+), effector memory (EM; CD4+CD45RO+CCR7-), and effector (CD4+CD45RO-CCR7+) cells. Additional subsets of TH cells analyzed included TH1 (CD4+CXCR3+CCR6-), TH2 (CD4+CXCR3-CCR6-), and TH17 (CD4+CXCR3-CCR6+). The gating strategies are depicted in Fig. 5A.

Fig. 5

Comparisons of CD3 + T cell subsets of PBMC after co-culturing with HLA-mismatched kidney organoids between the syngeneic and allogeneic groups with or without Tac. (A) Gating strategy for T cell subsets. (B) Percentage of total T (CD3 +), TH (CD3 + CD4 +), and TC (CD3 + CD8 +) cells in each group. (C) Percentage of TH subsets, EM TH (CD3 + CD4 + CD45RO + CCR7-), CM TH (CD3 + CD4 + CD45RO + CCR7 +), Eff TH (CD3 + CD4 + CD45RO- CCR7-), and Naïve TH cells (CD3 + CD4 + CD45RO- CCR7 +) in each group. (D) Percentage of TC subsets, EM TC (CD3 + CD8 + CD45RO + CCR7-), CM TC (CD3 + CD8 + CD45RO + CCR7 +), Eff TC (CD3 + CD8 + CD45RO- CCR7-), and Naïve TC (CD3 + CD8 + CD45RO- CCR7 +) cells. (E) Percentage of TH subsets, TH1 (CD3 + CD4 + CCR6- CXCR3 +), TH2 (CD3 + CD4 + CCR6- CXCR3-), and TH17 (CD3 + CD4 + CCR6 + CXCR3-) cells. TH, Helper T cells; TC cytotoxic T cells; EM, effector memory; CM, central memory; Eff, effector; PBMC, Peripheral blood mononuclear cell; Tac, Tacrolimus Data are presented as mean ± SE. *P < 0.05 vs. syngeneic group, #P < 0.05 vs. allogeneic group without tacrolimus

To examine the immunosuppressive effect of tacrolimus, PBMCs were co-cultured with mature kidney organoids under allogeneic conditions for 24 h in the presence or absence of tacrolimus (10 or 100 ng/mL). In Fig. 5B, the frequencies of total T cells, TH cells, and TC cells were significantly altered in the allogeneic group compared to the syngeneic group, (T cells, 78 ± 0.2% vs. 81 ± 0.3%; TH cells, 56 ± 0.4% vs. 55 ± 0.2%; TC cells, 34 ± 0.4% vs. 35 ± 0.2%, *P < 0.05 vs. syngeneic group) and these changes were attenuated by tacrolimus treatment (T cells, 67 ± 3.2% and 71 ± 1.3%; TH cells, 54 ± 0.2% and 47 ± 1.2%; TC cells, 34 ± 0.3% and 34 ± 0.3%, *P < 0.05 vs. syngeneic group, #P < 0.05 vs. allogeneic group without tacrolimus, allogeneic group + Tac10 or Tac100, respectively). As shown in Fig. 5C, the percentage of effector TH cells was significantly elevated in the allogeneic group (51 ± 0.4% vs. 43 ± 0.3%, *P < 0.05 vs. syngeneic group), while EM and CM TH cells were reduced (EM TH cells, 25 ± 0.6% vs. 27 ± 0.3%; CM TH cells, 3.8 ± 0.1% vs. 5.1 ± 0.2%, *P < 0.05 vs. syngeneic group); these changes were partially reversed by tacrolimus treatment (Eff TH cells, 38 ± 0.4% and 34 ± 0.3%; EM TH cells, 28 ± 0.5% and 37 ± 0.9%; CM TH cells, 5.4 ± 0.1% and 6.4 ± 0.1%, *P < 0.05 vs. syngeneic group, #P < 0.05 vs. allogeneic group without tacrolimus, allogeneic group + Tac10 or Tac100, respectively). Similarly, effector TC cells and EM TC cells were significantly increased in the allogeneic group (EM TC cells, 6.2 ± 0.2% vs. 4.6 ± 0.2%; Eff TC cells, 17.7 ± 0.2% vs. 16.2 ± 0.2%, *P < 0.05 vs. syngeneic group), and tacrolimus treatment significantly suppressed the increase in effector TC cells (Eff TC cells, 16 ± 0.4% and 15 ± 0.3%, *P < 0.05 vs. syngeneic group, #P < 0.05 vs. allogeneic group without tacrolimus, allogeneic group + Tac10 or Tac100, respectively) (Fig. 5D). Furthermore, as shown in Fig. 5E, the proportion of TH1 cells was markedly decreased in the allogeneic group and further reduced by tacrolimus treatment. (TH1 cells, 10.5 ± 0.2% vs. 31.2 ± 0.3%; *P < 0.05 vs. syngeneic group; 6.9 ± 0.2% and 5 ± 0.3%. #P < 0.05 vs. allogeneic group without tacrolimus, allogeneic group + Tac10 or Tac100, respectively). In contrast, the proportion of TH2 and TH17 cells was elevated in the allogeneic group compared to the syngeneic group (TH2 cells, 86.5 ± 0.6% vs. 65.2 ± 0.3%; TH17 cells, 2.7 ± 0.1% vs. 2.0 ± 0.1%, *P < 0.05 vs. syngeneic group), and this increase was partially suppressed by tacrolimus at both concentrations (TH2 cells, 85 ± 1.1% and 77 ± 1.6%; TH17 cells, 2.1 ± 0.1% and 1.8 ± 0.1%, #P < 0.05 vs. allogeneic group without tacrolimus, allogeneic group + Tac10 or Tac100, respectively).

Transcriptomic analysis of the kidney organoids derived allogeneic condition versus syngeneic condition

To investigate the differences in gene expression profiles of kidney organoids between syngeneic and allogeneic conditions, we conducted a transcriptomic analysis using RNA-sequencing. The results are illustrated in the hierarchical clustering heat map (Fig. 6A), scatter plot (Fig. 6B), and volcano plot (Fig. 6C), which depict global transcriptomic variations between the groups. A total of 43,424 RNA transcripts were identified in the kidney organoids; moreover, we discovered 256 DEGs with a p-value of < 0.05. Among these, 93 genes were two-fold upregulated while 163 genes were two-fold downregulated. Additionally, a bubble or radar chart displays the Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis, predicting the potential functions of the DEGs (Fig. 6D and E). The NF-κB and TNFα signaling pathways were ranked at the top, suggest that the co-cultivation of kidney organoids with HLA-mismatched PBMCs reflects kidney allograft rejection. Figure 6 presents the qRT-PCR results for NF-κB, IκBα, TNFα, and IL-6 which play critical roles in the NF-κB and TNFα signaling pathways (NF-κB, 1.35 ± 0.32; IκBα, 0.60 ± 0.1; TNFα, 1.22 ± 0.15; IL-6, 1.83 ± 0.11, *P < 0.05 vs. syngeneic group).

Fig. 6

Whole transcriptome analysis of the kidney organoids of allogeneic group versus syngeneic group. (A) Heat map and hierarchical clustering of RNA-sequencing data from the kidney organoids of both the syngeneic and allogeneic groups. Scatter plot (B) and volcano plot (C) illustrate the significant transcriptional changes in genes associated with allogeneic conditions. (D) The size and color of each bubble indicate the number of significantly changed genes enriched in the pathway and –log10 (p-value), respectively. The Y-axis represents the pathway name, and the x-axis represents the fold enrichment factor. (E) A radar chart displays normalized enrichment scores for genes with significant transcriptional changes in the allogeneic group across hallmark pathways; pathways more overrepresented in the allogeneic group than in the syngeneic group are highlighted with red brackets. (F) qRT-PCR result for mRNA level of NF-κB, IκBα, TNFα, and IL-6 in the kidney organoids. Data are presented as mean ± SE. *P < 0.05 vs. syngeneic group