Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Aug 9.
Published in final edited form as: Med. 2024 May 21;5(8):1016–1029.e4. doi: 10.1016/j.medj.2024.05.003

Cellular dynamics in pig-to-human kidney xenotransplantation

Wanqing Pan 1,2,*, Weimin Zhang 3,4,*, Binghan Zheng 1,*, Brendan Camellato 3,4, Jeffrey Stern 5,6, Ziyan Lin 7, Alireza Khodadadi-Jamayran 7, Jacqueline Kim 5,6, Philip Sommer 8, Karen Khalil 5,6, Elaina Weldon 5,6, Jiangshan Bai 1, Yinan Zhu 3,4, Peter Meyn 9, Adriana Heguy 9,10, Massimo Mangiola 5,6, Adam Griesemer 5,6, Brendan J Keating 3,5,6,11,†, Robert A Montgomery 5,6,†, Bo Xia 1,3,12,13,†, Jef D Boeke 3,4,†
PMCID: PMC11317223  NIHMSID: NIHMS1997640  PMID: 38776915

SUMMARY

Background:

Xenotransplantation of genetically engineered porcine organs has the potential to address the challenge of organ donor shortage. Two cases of porcine-to-human kidney xenotransplantation were performed, yet the physiological effects on the xenografts and the recipients’ immune responses remain largely uncharacterized.

Methods:

We performed single-cell and longitudinal RNA-seq analyses of the porcine kidneys to dissect xenotransplantation-associated cellular dynamics and xenograft-recipient interactions. We additionally performed longitudinal scRNA-seq of the peripheral blood mononuclear cells (PBMCs) to detect recipient immune responses across time.

Findings:

Although no hyperacute rejection signals were detected, scRNA-seq analyses of the xenografts found evidence of endothelial cell and immune response activation, indicating early signs of antibody-mediated rejection. Tracing cells’ species origin, we found human immune cell infiltration in both xenografts. Human transcripts in the longitudinal bulk-RNA-seq revealed that human immune cell infiltration and the activation of interferon gamma-induced chemokines expression occurred by 12 hours and 48 hours post-xenotransplantation, respectively. Concordantly, longitudinal scRNA-seq of PBMCs also revealed two phases of the recipients’ immune responses at 12 and 48–53 hours. Last, we observed global expression signatures of xenotransplantation-associated kidney tissue damage in the xenografts. Surprisingly, we detected a rapid increase of proliferative cells in both xenografts, indicating the activation of porcine tissue repair program.

Conclusions:

Longitudinal and single-cell transcriptomic analyses of porcine kidneys and recipient’s PBMCs revealed time-resolved cellular dynamics of xenograft-recipient interactions during xenotransplantation. These cues can be leveraged for designing gene edits and immunosuppression regimens to optimize xenotransplantation outcomes.

Funding:

This work was supported by NIH RM1HG009491 and DP5OD033430.

eTOC Blurb

Longitudinal and single-cell RNA-seq analyses of two pig-to-human kidney xenotransplantation cases revealed early signs of antibody-mediated rejection and rapid activation of porcine kidney cell proliferation indicating tissue repair. These insights can guide future strategies to optimize pig-to-human kidney xenotransplantation.

Graphical Abstract:

graphic file with name nihms-1997640-f0001.jpg

INTRODUCTION

Organ transplantation is a life-saving procedure for end-stage organ failure patients. However, the shortage of suitable donor organs has been a worldwide challenge1,2. Exploring alternative sources of organs for transplantation to meet the increasing demand is considered to be a health care imperative. Xenotransplantation of genetically engineered organs or tissues from different species has emerged as a promising solution to address this challenge3. The domesticated pig (Sus scrofa domesticus) has been identified as a suitable species due to its size similarities to human organs, quick maturation to near adult size in 6 months, and positive public perception4,5. By knocking out alpha-1,3-galactosyltransferase (αGAL-KO), genetically modified porcine organs can substantially reduce hyperacute rejection risk in non-human primates6–10. These positive results led the way to the world’s first pig-to-human kidney xenotransplantation into brain-dead decedents in 2021 to evaluate the safety and feasibility of the genetically engineered porcine kidney in humans11,12.

The porcine kidney transplants performed at our institution showed promising physiological functioning during the ~3 day study period, producing urine and not showing evidence of hyperacute rejection11. Initial histological analyses of the kidney biopsy samples during and after xenotransplantation did not detect obvious evidence of antibody-mediated rejection (AbMR), nor strong indications of porcine kidney injury11. However, due to the short duration of these human xenotransplantation trials, it is possible that the histological markers were not yet strongly expressed to a detectable level, rendering standard methods ineffective to capture signs of pathophysiological changes of the xenograft and the recipient’s immune response post xenotransplantation13.

Single-cell transcriptomic technology enables a comprehensive analysis of cellular physiology of organ transplantation and the recipient’s immune response14–17. In this study, we conducted comprehensive single-cell RNA sequencing (scRNA-seq) analyses on the first porcine-to-human kidney xenografts, as well as on longitudinal peripheral blood mononuclear cell (PBMC) samples, to characterize the intricate cellular and molecular dynamics of xenograft-recipient interactions. Such comprehensive analyses enabled the detection of major changes in both human and porcine cells upon xenotransplantation. Human transcript analyses in longitudinal bulk-RNA-seq revealed that infiltration of human immune cells into the porcine kidney occurred by 12 hours post-xenotransplantation, followed by the activation of chemokine expression between 24 – 48 hours. Further, our longitudinal scRNA-seq of recipient PBMCs unveiled a biphasic immune response, occurring at 12 and 48–53 hours post-xenotransplantation. Intriguingly, we detected a dramatic increase of proliferative cells in both porcine kidneys upon xenotransplantation. These proliferative cells express proximal tubule cell marker genes, indicating a rapid activation of a porcine kidney tissue repair program. Together with longitudinal bulk RNA-seq analyses of the xenograft, our results chart dynamic xenograft tissue physiology and time-resolved immune responses between xenografts and the recipients, thus providing key insights to guide future genetic optimization of pigs for xenotransplantation.

RESULTS

Single-cell transcriptomic landscape of pig-to-human kidney xenotransplantation

To characterize pig-to-human xenotransplantation-associated cellular dynamics in the xenografts, we performed single-cell RNA sequencing (scRNA-seq) on the xenografts and their contralateral (untransplanted) kidneys from two cases of xenotransplantation using αGAL-KO kidneys11. We collected single cell samples from the xenograft and its contralateral untransplanted kidney in parallel at the end point of each experiment (54h, Figure 1A). After quality filtering, we obtained 23,868 cells in total, including 7,080 cells from the first transplanted xenograft (and 2,819 cells from the matched control kidney), and 5,999 cells from the second transplanted xenograft (and 7,090 cells from the matched control kidney) (Figure S1). Integrative analyses of scRNA-seq data of the four samples identified 16 unsupervised cell clusters based on their transcriptomic profiles (Figure 1B), with a balanced representation of cells in the samples of both xenotransplantation cases (Figure 1C). These cell clusters represent the major cell types previously identified in mammalian kidneys, including proximal tubule cells (PTCs), thick ascending limb of the Loop of Henle (TAL), distal tubule cells (DTCs), intercalated cells (ICs), endothelial cells (ECs), and immune cells (Figures 1D and S2). Podocytes were not detected as a separate cluster of cells due to their sparsity among kidney cells, though we did identify a small population of putative podocytes in our data (Figure S2B–C)16,17.

Figure 1. Single-cell RNA-seq analyses of pig-to-human kidney xenotransplantation.

Figure 1.

Open in a new tab

(A) Schematic overview of the transcriptomic analyses of pig-to-human kidney xenotransplantation.

(B) Unsupervised clustering of the merged single cell transcriptomes across all samples (left) and visualization by cell types (right). UMAP, Uniform Manifold Approximation and Projection18.

(C) UMAP visualization of single cell distribution from each kidney sample.

(D) Dot plot of marker genes indicating cell-type identity across all cell populations. The color intensity and dot size represent average expression level and the percentage of expressed cells, respectively. Endo, endothelial cells; IC_NS, nonspecific intercalated cells; IC_TypeB, Type B intercalated cells, IC_TypeA, Type A intercalated cells; DTC, distal tubule cells; TAL_1, thick ascending limb population 1; TAL_2, thick ascending limb population 2; PT, proximal tubule cells; PT_VIM+, Vimentin-positive proximal tubule cells; PT_Prolif, proliferating proximal tubule cells.

Human immune cells infiltrate porcine xenografts

To test whether human immune cells had infiltrated into the xenograft, we mapped scRNA-seq raw sequencing reads to both human and porcine reference genomes to compare their cell-specific mapping efficiency. In principle, human cells should detect more mappable reads when mapped to the human reference genome due to sequence specificity, and vice versa. Using this approach, we identified 211 human cells in the two xenografts (146 cells in xenograft 1 and 65 cells in xenograft 2, Figure 2A). As expected, no human cells were detected in the two untransplanted control kidneys (Figure 2A). The vast majority (199/211) of the detected human cells correspond to macrophage (LYZ+ and CD163+) and NK cells (NKG7+ and GNLY+) (Figures 2B and S3A–C), while no evidence of human neutrophils was detected. Coclustering with the human infiltrated immune cells are several porcine immune cell types, including macrophages, natural killer T (NKT) cells, and T cells (Figure 2B). No other abundant human blood cell types, such as T cells or erythrocytes, were detected among these human cells, indicating that they are unlikely to come from contamination of circulating human blood cells after reperfusion.

Figure 2. Human NK cells and macrophages infiltrate into porcine kidney xenograft.

Figure 2.

Open in a new tab

(A) UMAP visualization of human (bronze) and porcine (cyan) cell distribution. The immune cell cluster is enlarged for detail.

(B) UMAP visualization of sub-clustered human and porcine immune cell types.

(C) Heatmap of human-to-porcine raw reads counting ratios (H/P ratio) of macrophage and NK cell marker genes.

(D) Violin plots of interferon-gamma signaling gene expressions in human and porcine immune cell populations in single-cell transcriptome data.

(E-F) Xenotransplantation time-resolved gene expression levels (E) and human/porcine transcript (H/P) ratio (F) of interferon-gamma signaling genes in the longitudinal bulk RNA-seq of xenograft biopsies. Statistical significance is indicated by asterisks above to the violin plots: * for p < 0.05, ** for p < 0.01, *** for p < 0.001, and **** for p < 0.0001.

To determine the time of human immune cell infiltration into the xenografts, we performed a longitudinal bulk RNA-seq of core biopsy samples from the second xenograft at times 0, 12, 24, and 48 hours post-xenotransplantation (pXTx) (Data S1). We sought to detect human transcripts from human infiltrated macrophages and NK cells by comparing raw reads of bulk RNA sequencing data mapped to the human reference genome versus those that mapped to the porcine reference genome. We found that human transcripts of both NK cell marker genes (e.g. NKG7, GNLY, KLRD1) and macrophage marker genes (e.g. LYZ, CD163, TYROBP) substantially increase at 12 hours pXTx (Figure 2C). These results suggest that human immune cell infiltration into the xenografts occurred by 12 hours pXTx.

Studying scRNA-seq data collected at 54 hours pXTx, we found significantly increased expression of interferon-gamma signaling in the xenografts compared to their untransplanted control kidneys (Figure 2D). Specifically, infiltrated human NK cells express interferon-gamma (IFNG, Figure 2D), while the infiltrated human macrophages express interferon-gamma-stimulated chemokines such as CXCL9, CXCL10 and CXCL11 (Figure 2D). These results indicate a potential early signal of impending rejection as seen in human kidney allotransplantation13,19,20.

To identify the temporal- and species-specific activation of interferon-gamma signaling, we analyzed gene expression dynamics in the longitudinal bulk RNA-seq. We found strongly increased expression of interferon-gamma signaling in the xenografts since 12 hours pXTx, which was maintained throughout the period of the study (Figure 2E). Interestingly, the persistent high chemokine expression levels over time were predominantly consisted of porcine chemokine transcripts 12 hours pXTx and 24 hours pXTx (human/porcine transcript ratio < 1), but transition to be predominantly consisted of human chemokine transcripts at 48 hours pXTx (human/porcine transcript ratio > 1, Figure 2F). Together, these results suggest early activation of host innate immune response, possibly indicating an early rejection signal in both xenografts, and revealed the temporal dynamics of xenograft-recipient immune response.

Transcriptomic signatures consistent with an antibody-mediated response

Having detected expression of marker genes corresponding to early rejection signals during xenotransplantation, we sought to distinguish different types of rejection between antibody-mediated rejection (AbMR) and acute cellular rejection (or T-cell-mediated rejection, TCMR). Though clinical diagnosis of transplantation rejection types requires a more comprehensive histological analysis21, analyzing specific marker gene transcripts can provide early evidence of rejection types in the xenograft13. To do so, we performed transcriptomic analyses of a curated list of rejection marker genes across AbMR-selective genes and TCMR-selective genes13 (Figure 3).

Figure 3. Identification of rejection signals in porcine endothelial cells and in immune cells.

Figure 3.

Open in a new tab

(A-B) Relative gene expression levels of AbMR (A) and TCMR (B) marker genes over the period of the xenotransplantation. Genes with dramatic expression changes (>2 folds) at the last time-point were highlighted in colors.

(C-D) Violin plots of AbMR marker gene expression in porcine endothelial cells (C) and TCMR marker gene expression (D) in immune cells. Gene expression levels were detected in single-cell RNA-seq data. Statistical significance is indicated by asterisks above to the violin plots: * for p < 0.05, ** for p < 0.01, *** for p < 0.001, and **** for p < 0.0001.

Endothelial cell activation due to antibody recognition is a key feature in antibody-mediated rejection22,23. Thus, we compared the AbMR-selective marker gene expressions in the endothelial cells collected from the xenografts to those of the control kidneys. We found significantly increased expression of endothelial-specific AbMR marker genes, including CDH5, CDH13, PECAM1, RAMP3, and TM4SF18 (Figure 3A). These results suggest porcine endothelial cell activation in the xenografts, indicating AbMR. Analyzing the longitudinal bulk RNA-seq transcriptome of the second xenograft, we found that many of the AbMR-selective marker genes also increased during the period of the study (Figure 3C).

Studying TCMR marker genes, we found that most of these genes fluctuate without significant changes (>2 fold) during the period of the study (Figure 3B). However, expression levels of two macrophage-expressed TCMR-selective marker genes (ANKRD22, SLAMF8)13,24,25 significantly increased across the time of the study (Figures 3B and 3D).

Global xenograft tissue damage

To fully characterize the cellular changes of porcine kidneys upon exposure to the human immune environment, we performed differentially expressed gene (DEG) analysis on the proximal tubule cell populations between xenografts and control kidneys (Figure 4A). Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) gene set enrichment analyses of the top 1000 xenograft-upregulated genes indicate increased catabolic processes and disease profiles (Figure S3D–E). The top differentially expressed genes in both xenografts are biomarkers of kidney injury and inflammatory response, including profibrotic osteopontin (SPP1)26,27, phagocytic osteoactivin (GPNMB)28,29, and S100A6 30,31(Figure 4A). Kidney injury-related genes were globally increased in the nephron cell types in both xenografts including PTCs, DTCs, TAL cells, and ICs (Figure 4B), indicating signs of holistic tissue damage when xenografts were exposed to the human immune environment.

Figure 4. Xenotransplantation-associated porcine kidney damage and the activation of a cell proliferation program.

Figure 4.

Open in a new tab

(A) Volcano plots of differentially expressed genes between xenograft and control for the first (left) and second (right) cases of xenotransplantation. Representative genes (biomarkers of kidney-injury and proliferation) were labeled. A few data dots in the plot reached a ceiling value on the y-axis, indicating system’s default lowest p-value.

(B) Violin plots of kidney-injury biomarker expression levels across major cell types in the nephron.

(C) UMAP highlight of the proliferating cell population.

(D-E) UMAP visualization of the sub-clustered proliferating cell cluster colored by organismal origin (D) and cell cycle phase assignment (E).

(F) Ridge plots of top marker genes in cells across G1, S, and G2 phases.

(G-I) Gene expression levels of proliferation marker (STMN1, G), proximal tubule cell marker (SLC34A1, H), and T-cell marker (CD3E, I) in the proliferating cell population. The smaller cluster of cells expressing T cell markers but not PTC markers (CD3E+;SLC34A1−).

(J) Time-resolved gene expression levels of kidney tissue injury marker genes (reds) and cell cycle genes (greens) in longitudinal RNA-seq.

Xenograft-specific proliferating cells indicating porcine kidney tissue repair

Concurrent with detecting global tissue damage in the nephron cells, we also identified a unique proliferating cell population among all cell types (Figure 4C). These cells were almost exclusively of porcine origin, and they highly expressed cell cycle-related genes, including Stathmin 1 (STMN1), PCNA Clamp Associated Factor (PCLAF), High Mobility Group Nucleosomal Binding Domain 2 (HMGN2), and High Mobility Group Box 2 (HMGB2) (Figure 4D–F). Sub-clustering these proliferating cells, we identified two distinct cell groups, with the majority of the cells expressing PTC marker genes, such as SLC34A1 (Figure 4G–I). Stratifying the proliferating proximal tubule cells by their sample of origin revealed that the majority of these populations was found within the two porcine xenografts (Figure 4G, and Table S1). In the first case, we detected 2 and 122 proliferating cells from control kidney and xenograft, representing 0.1% and 2.4% of total PTCs (Table S1), respectively. Similarly in the second case, we detected 12 and 322 proliferating cells from the control kidney and xenograft, representing 0.2% and 6.3% of total PTCs (Table S1), respectively.

We speculated that these proliferating cells were stimulated by xenotransplantation-associated porcine kidney damage. Indeed, compared to xenografts, fewer proliferating SLC34A1-positive cells were observed in the control kidneys (Figure 4H), which also experienced ischemic conditions during prolonged cold storage (Methods). Moreover, we expected to see a temporal order of kidney damage marker gene expression preceding the cell proliferation signals. We test this hypothesis through our longitudinal bulk RNA-seq data, which indeed revealed that the expression of genes indicating kidney injury (SPP1, GPNMB, and S100A6) peaks at 12 hours post-xenotransplantation, while the expression of genes indicating cell proliferating start to increase at 24 hours pXTx and continue to rise through 48 hours pXTx (Figure 4J).

Longitudinal scRNA-seq of recipient’s PBMCs revealed two waves of immune activation

To understand how the recipient responded to the grafting of porcine kidney across the period of xenotransplantation, we collected longitudinal PBMC samples at pre-transplantation (0 h), and 6, 12, 24, 48, and 53 hours pXTx for scRNA-seq analyses (Figure 1A). scRNA-seq of xenograft recipient’s PBMCs captured all major cell types in human PBMCs (Figures 5A and S4). The proportions of cell types within each PBMC sample collected across time points were also visualized (Figure 5B).

Figure 5. Recipient’s PBMCs show two waves of immune response expression signatures.

Figure 5.

Open in a new tab

(A) Uniform Manifold Approximation and Projection (UMAP) visualization of the second kidney xenograft recipient’s PBMC clusters. mono-CD14, Monocytes CD14; mono-CD16, CD16-positive monocytes; M1: macrophage 1; M2: Macrophage 2; NK: natural killer cells; NKT1: natural killer T cells 1; NKT2: natural killer T cells 2; T-CD8: CD8 T cells; T-CD4: CD4 T cells; T-Reg: T Regulatory cells, N.B: naïve B cells; M.B: memory B cells; P.B: plasma B cells; MGK-P: Megakaryocyte Progenitor cells; MGK: Megakaryocytes; RBC: Red blood cells (Erythrocytes)

(B) Cell type proportions of PBMC populations across the six timepoints.

(C) Temporally variable MHC class II genes expression pattern across the period of xenotransplantation in representative antigen presenting cell types.

(D-E) Relative gene expression levels of gene sets enriched at 12 hours pXTx (D) and 48–53 hours pXTx (E) across the period of xenotransplantation in representative cell types.

Antigen presentation is a pivotal process to trigger an adaptive immune response. Analyzing expression levels of major histocompatibility complex (MHC) class II genes in antigen presenting cells (monocytes, macrophages, and megakaryocyte progenitor cells), we found two waves of increased MHC gene expression: a first wave at 12 hours pXTx, and a second at 48–53 hours pXTx (Figure 5C). These two waves of increased MHC gene expression are correlated with two phases of immune gene activation at 12 hours pXTx and 48–53 hours pXTx (Figure 5D, E). PBMCs at 12 hours pXTx have upregulated gene sets in most cell types including monocytes, macrophages, T cells, B cells, and NK cells (Figure 5D, Data S2). Macrophages, NK cells, T cells, and megakaryocytes have upregulated gene sets at 48–53 hours pXTx (Figure 5E, Data S3). These genes are enriched in functions related to the activation of predominantly innate immune responses at 12 hours pXTx (Figure S5A) and the early activation of adaptive immune responses at 48 hours pXTx (Figure S5B). Interestingly, these two waves of gene expression are correlated with the trajectory of interferon-gamma expression in NK and NKT cells in the recipient’s PBMCs (Figure S5C). Together, these results from time-resolved scRNA-seq of PBMCs revealed the dynamic immune response of the recipient’s immune system to the porcine xenograft kidney.

DISCUSSION

Xenotransplantation of genetically engineered porcine organs has the potential to transform human health3,5. Although multiple trials have been performed in model organisms including non-human primates, how xenografts respond to the human physiological environment and the reciprocal activation of human immune responses remains largely unknown. Additionally, since it remains unclear how translatable the results of studies in non-human models are to human xenotransplantation, this study represents a unique opportunity to apply powerful scRNA-seq technologies to study xenotransplantation in the human decedent model. In this study, we performed scRNA-seq on two cases of pig-to-human kidney xenotransplantation to fully characterize the cellular dynamics of xenografts and the recipient’s immune response to them. Together with longitudinal bulk RNA-seq on the xenograft and scRNA-seq on the recipient’s PBMCs, our data revealed the time-resolved map of cellular and immune dynamics of xenograft-recipient interactions.

Although no hyperacute rejection signal was observed through standard histological analyses11, our results provide strong evidence that both kidney xenografts presented early signs of an AbMR phenotype at 54 hour post-xenotransplantation. The evidence obtained in this study includes 1) the infiltration and the activation of human macrophage and NK cells in the xenografts (Figure 2–3); 2) the detection of AbMR-associated marker gene expression through time-resolved RNA-seq of xenograft biopsies (Figure 3); 3) activation of endothelial cells in xenografts (Figure 3); and 4) the holistic xenograft tissue damage signals (Figure 4). Notably, these expression changes were not observed in the contralateral untransplanted kidneys that also experienced ischemia, indicating that the observed evidence reflect xenotransplantation-induced responses but not ischemic injury-induced effects. Reperfusion is associated with its own set of expressed genes, tissue injury and innate immune responses. Because our control tissues did not recapitulate this process, we cannot subtract out the full ischemia-reperfusion-induced signal in these experiments.

Our results indicating an early sign of AbMR in the xenografts are consistent with the initial observation that the recipients had low to moderate levels of preformed xenoreactive antibodies in crossmatch tests before xenotransplantation11, and with a concurrent study using immunohistology and spatial transcriptomics approaches performed on the same xenografts32. In contrast to the immunohistology study which suggested infiltration of human neutrophils into the xenograft, our single-cell transcriptomics did not detect such a cell population, potentially due to the number of detectable neutrophils remaining in the xenograft being very small.

Immune cell infiltration into transplanted organs may indicate activation of the recipient immune response and predict rejection of the graft33–36. Recipient’s macrophage and NK cell infiltration into the transplanted organ is seen in innate immune response but the extent of the response in the kidney xenograft and degree of endothelial injury seems to represent a unique phenotype. This phenomenon has been observed in studies using ex vivo perfusion of pig organ with human blood, pig-to-non-human-primate xenotransplantation, and to a lesser extent, in human allotransplantation35,37–39. Our time-resolved transcriptomic monitoring of the human transcripts in the second xenograft revealed that human immune cell infiltration happened as early as 12 hours pXTx. Notably, the infiltration of human immune cells was not likely to be induced by environmental infection as the donor pigs were exhaustively surveilled for known zoonotic pathogens and the recipients were also tested for human infections that replicate in immunocompromised hosts.

Our longitudinal RNA-seq monitoring of the xenograft revealed a two-phase activation of interferon-gamma signaling and the expression of multiple pro-inflammatory chemokines including CXCL9 and CXCL10. We found that, at 12-to-24 hours pXTx, these pro-inflammatory chemokines were mainly expressed by porcine immune cells. In contrast, at 48 hours pXTx, these pro-inflammatory chemokines were mainly expressed by the infiltrated human macrophages, indicating the activation of the recipient’s immune response. These results may explain the two waves of immune-activation gene expression observed in the recipient’s PBMCs. Our results demonstrate that porcine immune cells rapidly activated interferon-gamma signaling during xenotransplantation, and thus might have contributed to the first wave of immune response between xenograft and the recipient. In the future consideration of depleting such porcine tissue-resident immune cells pre-transplantation might reduce activation of the recipient immune response1,40,41.

Single-cell RNA-seq of the porcine kidneys identified a SLC34A1-positive proliferating cell population activated in the porcine xenografts in human studies for the first time. Proliferating or cycling cells have been observed in damaged kidney in both humans and rodents42–45, and in some rejection cases of human kidney allotransplantations15. Further studies revealed that the activation of proliferative cells occurring in damaged kidneys may represent a cellular hallmark of kidney tissue repair to potentially restore nephron structure and function43,46,47. Studies in rodents found that acute ischemic damage can activate the proliferation program of proximal tubule cells and may induce maladaptive repair of damaged kidney48,49. Our results demonstrated that the proliferative cells were dramatically more enriched in xenografts than in the untransplanted control kidneys. Moreover, time-resolved RNA-seq analysis of the xenograft biopsies revealed the sequential onset of expression of kidney damage genes and cell cycle genes. Together, these results indicate that the rapid activation of SLC34A1-positive proliferating cells was induced upon xenotransplantation. It requires a longer-duration study to reveal whether the rapid activation of cell proliferation program would help restore tissue homeostasis or trigger maladaptive repair, which may eventually damage porcine kidney function49–51.

In conclusion, our results provide comprehensive insights into the time-resolved and single-cell transcriptomic dynamics of xenograft-recipient interactions in the early stages of xenotransplantation. Beyond that, our results highlighted the power of single-cell and time-resolved transcriptomic approaches for discovering new biological insights such as the rare proliferating cell populations and the temporal regulation between xenograft and the recipient immune system. Although longer periods of follow up studies will be needed to further characterize how these transcriptomic signals manifest at the cellular and tissue level, the presented discoveries will inform future porcine genetic engineering and xenotransplantation development.

Limitations of Study

There are a few limitations in the current study. First, the main limitation of this study is the small sample size. We analyzed data from only two cases of pig-to-human kidney xenotransplantation and longitudinal monitoring of the xenograft and host peripheral blood mononuclear cell only in the second case. Second, another major limitation is the short duration of the xenotransplantation study, which lasted only 53 hours. Whether the observed features between the xenografts and the recipients would persist requires longer term follow up studies. Third, the proprietary pig organ used in this study was an alpha-1,3-galactosyltransferase knockout model, which although poses low risk of hyperacute rejection but does not include other genetic modifications to further reduce immunogenicity. These limitations of the study are inherent to the pioneering nature of this xenotransplantation protocol performed in recently deceased individuals. Consequently, our findings primarily reflect molecular correlates rather than enabling clinical diagnosis of any signs of rejection. Larger number of cases using more advanced genetically engineered pig models across longer monitoring periods are needed to comprehensively evaluate the safety and efficacy of kidney xenotransplantation, as well as dissect the time-resolved physiological changes of the xenografts and recipient immune responses.

STAR METHODS

Resource Availability

Further requests and questions should be directed to the lead contact Bo Xia at xiabo@broadinstitute.org.

Materials Availability

This study did not generate new unique reagents.

Data and Code Availability

All de-identified single-cell RNA-seq and bulk RNA-seq sample data from this study have been deposited to the GEO portal. Accession numbers are listed in the Key Resources Table. De-identified data of sequenced samples and all original code used for analysis in this study has been deposited in the Github repository: github.com/boxialaboratory/Pig-to-Human-Kidney-Xenotransplantation.

KEY RESOURCES TABLE.

REAGENT or RESOURCE SOURCE IDENTIFIER
Biological Samples
Porcine kidney tissue (untransplanted and xenograft) NYU Langone Transplant Institute NA
Human PBMC NYU Langone Transplant Institute NA
Chemicals, Peptides, and Recombinant Proteins
DPBS buffer, no calcium, no magnesium Gibco Cat#14190250
Trypsin-EDTA (0.25%) Gibco Cat#25200056
TURBO DNase Invitrogen Cat#AM2238
ACK lysing buffer Gibco Cat#A1049201
Critical Commercial Assays
Chromium Single Cell 3’ Kits (v3.1 Chemistry) 10x Genomics Cat#1000268
Dead cell removal kit Miltenyi Biotec Cat#130-090-101
RNeasy Plus kit QIAGEN Cat#74134
RNA 6000 Nano kit Agilent Cat#5067-1511
Ribo-Zero Plus rRNA Depletion Kit Illumina Cat#20037135
Countess II Automated Cell Counting Chamber Slides Invitrogen Cat#10228
Deposited Data
Raw single-cell RNA-seq data This paper GEO: GSE257542
Raw bulk RNA-seq data This paper GEO: GSE257541
HGD Duan et al.53 doi:10.1016/j.xinn.2021.100141
Gene Ontology Ashburner et al.55 doi: 10.1038/75556
Kyoto Encyclopedia of Genes and Genomes Kanehisa et al.56 doi: 10.1093/nar/28.1.27
Experimental Models: Organisms/Strains
Pig: αGTase-KO: GGTA1−/− Revivicor NA
Software and Algorithms
CellRanger 10X Genomics https://support.10xgenomics.com/single-cell-gene-expression/software/downloads/latest
Bioanalyzer 2100 Agilent https://www.agilent.com/en/product/automated-electrophoresis/bioanalyzer-systems/bioanalyzer-software
Seurat v4.3 Satija et al.52 https://doi.org/10.1038/nbt.3192
clusterProfiler v4.8.2 Wu et al.57 doi:10.1016/j.xinn.2021.100141.
R 4.3 R Core Team https://www.r-project.com
R code for analysis This paper https://github.com/boxialaboratory/Pig-to-Human-Kidney-Xenotransplantation/tree/main/scripts
doi: 10.5281/zenodo.11068840
Open in a new tab

Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

Experiment Models and Study Participant Details

Pigs genetically modified for the alpha-1,3-galactosyltransferase gene knock-out genotype were supplied by Revivicor, a subsidiary of United Therapeutics. The ethics foundations of the experimental xenograft transplantation in brain-dead human recipients were reported in the clinical report11. In summary, each of the recipients was declared brain-dead by the clinical care team according to standard criteria at site of hospitalization. Written informed consent was obtained from the family of the recipient, and the decedent was transferred to the intensive care unit (ICU) at the NYU Langone Health. The New York State Department of Health reviewed the plan for zoonotic disease surveillance, handling of hazardous materials, and embalming of the body before transport and burial. The NYU Research on Decedents Oversight Committee reviewed the protocol and provided oversight (registration approval number: #002). The institutional review board of the NYU Grossman School of Medicine approved this study (registration number S19–00192).

Method Details

Sample collection

Ethics statements, regulatory details, organ selection and procurement, and xenotransplantation procedures were previously described for the xenografts analyzed in this study11. Briefly, the transplanted and contralateral control kidneys went through the same procedure for storage and transportation. When one kidney was transplanted to the recipient, the contralateral control kidney was kept in cold preservation solution for storage and acted as a backup. Kidney tissues from xenograft and the untransplanted kidney in both cases were collected in parallel at the end point of the study (54 hours). Longitudinal core biopsy samples were collected in xenotransplantation case 2 at 0h from untransplanted control kidney (with two technical replicate samples), and then at 12h, 24h, and 48h from the xenograft (with one biopsy sample). PBMC samples were collected from the recipient in case 2 at the following timepoints: pre-transplant (0 h) and then at 6h, 12h, 24h, 48h, and 53h (right before termination) post-xenotransplantation.

Sample preparation for bulk and single-cell RNA-seq

Freshly collected kidney tissues were minced rapidly in a cell culture dish followed by washing with DPBS buffer. Minced tissue was resuspended in 5mL of 37°C pre-warmed 0.25% Trypsin plus TURBO DNase (10U/mL final concentration) for 20 min with mechanical dissociation using serological pipette every 5 min. Following digestion, dissociation mix was filtered through a 100μm strainer. The remaining large tissues on the strainer were further ground and washed to collect additional cells. Cell debris and dead cells were removed using Miltenyi Biotech Dead Cell Removal Kit. Cells were counted using Countess II Automated Cell Counter before loading for single-cell RNA-seq using 10X Genomics Chromium Single Cell 3’ Kit (v3.1 Chemistry). For bulk RNA-seq, kidney core biopsy samples were homogenized using a pestle homogenizer and total RNA was extracted from lysate using a QIAGEN RNeasy Plus kit. Total RNA quality and quantity were examined using the RNA 6000 Nano Kit and Agilent 2100 Bioanalyzer. Bulk RNA-seq library was constructed following the Illumina Ribo-Zero Plus rRNA Depletion Kit protocol (protocol number: 20037135).

Longitudinal single-cell RNA-seq of the recipient’s PBMCs were prepared at the same time after collecting all samples across timepoints. Fresh PBMCs of each timepoint were first collected through gradient centrifugation and any remaining RBCs were removed using the ACK lysing buffer followed by washing by the DPBS buffer. The prepared PBMCs for each time point were cryopreserved according to the 10X Genomics protocol (protocol number: CG00039). Upon collecting all PBMC samples, the frozen cells were thawed at the same time, then loaded for single-cell RNA-seq using 10X Genomics Chromium Single Cell 3’ Kit (v3.1 Chemistry).

Quantification and Statistical Analysis

Data Pre-processing and Quality Control of scRNA-seq

Raw single-cell sequencing data were processed with the CellRanger pipeline of 10X Genomics to generate single-cell raw gene expression matrices. Kidney single-cell sequencing data were separately mapped to both human genome (GRCh38) and porcine genome (Sscrofa11) to distinguish species origin of cells, whereas recipient PBMC single cell data were mapped to human genome (GRCh38) only. Single-cell raw expression matrices were then processed and analyzed using Seurat v4.3.052. Cells with less than 500 expressed genes, 1000 UMI, or more than 8% of transcripts from mitochondrial genes were removed from the kidney single cells, resulting in a total of 23,868 cells. A library of homologous genes between human and pig, obtained from the HGD Database53, was used as a reference to subset the filtered gene expression data for downstream analyses. Within the scope of gene expression of human-porcine one-to-one homologous genes, single cells with greater proportion of transcripts mapped to the porcine genome than the human genome are identified to be porcine cells, and vice versa to be human cells. PBMC single-cell RNA-seq datasets were filtered by removing extra-low-quality cells with less than 200 expressed genes or higher than 20% of mitochondrial genes, resulting in a total of 31,321 cells across 6 time points.

Data Normalization, Integration, and Cell Type Annotation

Each filtered dataset obtained from the kidneys and PBMC samples were normalized using SCTransform v2 from Seurat followed by integration using most commonly deemed variable features within each dataset 54. Dimensionality reduction analyses were performed on integrated kidney and PBMC datasets separately using Principal Component Analysis (PCA: npcs = 30) and Uniform Manifold Approximation and Projection (UMAP: dims = 1:30). Cell clusters were generated using the original Louvain algorithm with a resolution of 0.7 for integrated kidney dataset and 0.5 for the integrated PBMC dataset. Individual clusters were annotated based on the expression of cell type-specific markers.

Differential Expression Analysis

Differentially expressed genes between the control and xenograft samples were assessed using the Seurat FindMarkers function, whose transcripts were detected in at least 20% of cells, with a log-fold-change threshold of 0.25, and each identified DEG were tested for statistical significance using the Wilcoxon rank sum test. We filtered differentially expressed genes with log2 fold change > 0.5, and p-value < 0.01 as genes of interest for further analysis.

Cell Cycle Analyses

We applied cell cycle scoring to assess the cycling status of single cells. CellCycleScoring function of Seurat computes a score for each cell cycle phase (G1, G2/M and S) and clustering cells into the three phases of cell cycles by summing the expression levels of the phase-specific marker genes. The cell cycle scores and phase assignments were annotated for each cell, and followed by secondary validation of cell cycle marker gene expressions.

Temporal Gene Set Coexpression Clustering Analysis

To minimize the effect of noise on temporal gene expression values in PBMC data, we used the corrected UMI counts matrix adjusted for sequencing depth from the SCT assay, normalized by gene using centered log ratio transformation by applying the Seurat NormalizeData function, then scaled using Seurat ScaleData function for each cell type independently. For each cell type, we identified a set of temporally variable genes by using Seurat FindVariableFeatures and filtered out mitochondrial genes (MT), ribosomal genes (RP), or genes whose normalized expression was detected in < 20% of all cells within the given cell type. The 6 average Pearson residual values of each filtered temporally variable genes at the 6 time points generated by the AverageExpression function was used to calculate euclidean distance matrices between filtered temporally variable genes for each cell type.

Temporally Enriched Gene Set Identification

In each cell type, we identified co-expressed gene sets with strong evidence of temporal enrichment. To do so, we filtered for gene sets whose average scaled expression level at a given time point is larger than the average scaled expression level across all previous time points by 0.5. We further removed gene sets composed of less than 20 genes or are supported by less than 20 cells in the corresponding cell type at the time point of gene set enrichment. These parameters identified 13 cell types with at least one gene set showing temporal enrichment at 12 hours post-xenotransplantation (pXTx), and 8 cell types at 48–53 hours pXTx. Enriched gene sets identified from erythrocytes at both time points were excluded from visualization.

Gene Set Enrichment Analysis (GSEA) for Functional Annotations

All Gene Ontology (GO) term55 and Kyoto Encyclopedia of Genes and Genomes KEGG term56 enrichment analyses were performed using clusterProfiler (v4.8.2)57. GO terms generated from the least number of overlapping terms were visualized for GO term analyses of PBMC data. All p-values in the GSEA were adjusted for multiple tests using the Benjamin-Hochberg method58.

Analyzing Longitudinal Bulk RNA-Seq of Kidney Biopsy Samples

The bulk RNA-seq data across 0, 12, 24, and 48 hours pXTx was analyzed to understand the temporal changes in gene expression between the control and xenograft samples. Two technical replicates were conducted at 0 hours post-transplantation from the control kidneys. Gene expression at time 0 is calculated by the average value between the two replicates. The log2 fold change in gene expression was calculated for each gene across the different time points relative to the expression value at time 0. To calculate the human to pig raw count reads ratio (H/P Ratio), the absolute raw reads mapped to the human genome for every gene at a specific time point is divided by the absolute raw reads mapped to the porcine genome.

Supplementary Material

1

Data S1: Bulk RNA-seq of porcine kidney biopsies during the course of xenotransplantation, related to Figure 2.

2

Data S2: Gene sets enriched at 12 hours pXTx in PBMC single-cell data, related to Figure 5.

3

Data S3: Gene sets enriched at 48–53 hours pXTx in PBMC single-cell data, related to Figure 5.

5

Context and Significance.

Xenotransplantation of genetically engineered porcine organs holds the potential to address the challenge of organ donor shortage. Early trials of porcine kidney xenotransplantation have been performed on brain-dead human recipients, but the cellular dynamics between xenografts and the recipients’ immune responses remain largely uncharacterized. By employing single-cell and longitudinal RNA sequencing, Pan et al uncovered the infiltration of human immune cells into the xenografts and the activation of recipients’ immune response, suggesting an early sign of antibody-mediated rejection. Remarkably, the porcine kidney rapidly stimulated cell proliferation, indicating the activation of a tissue repair program post-transplantation. These insights will contribute to the design of new genetically engineered porcine models for optimized xenotransplantation outcomes.

Highlights.

  • Human immune cells infiltrated into porcine kidney xenografts as early as 12-hour pXTx

  • Longitudinal scRNA-seq of human PBMCs revealed two phases of immune response

  • Xenografts undergo global inflammatory tissue damage-associated gene expression

  • Porcine kidneys rapidly activated cell proliferation post-xenotransplantation

ACKNOWLEDGEMENTS

The authors sincerely thank the families of the decedents for their generous donation to science. The authors also thank the United Therapeutics Corporation, PBC, for supporting the kidney xenotransplantation experiments, the NYU Transplant Research Team and the NYU Langone Health Center for Biospecimen Research and Development (CBRD) team for supporting sample collection. We thank LiveOnNY for providing end-of-life family support and whole-body donation education and the NYU Langone nursing staff for their expert and compassionate care of the decedent. This work was supported in part by NHGRI RM1HG009491 to J.D.B., and partially supported by NIH DP5OD033430 to B.X. We thank Bingxu Liu, Huiyuan Zhang, and Liz Gaskell for suggestions and discussion. Graphical Abstract and parts of Figures 1A were created with BioRender.com.

Footnotes

DECLARATION OF INTERESTS

J.D.B. is a Founder and Director of CDI Labs, Inc., a Founder of and consultant to Opentrons LabWorks/Neochromosome, Inc, and serves or served on the Scientific Advisory Board of the following: CZ Biohub New York, LLC, Logomix, Inc., Modern Meadow, Inc., Rome Therapeutics, Inc., Sangamo, Inc., Tessera Therapeutics, Inc. and the Wyss Institute. R.A.M. is on scientific advisory boards for eGenesis, Sanofi, Regeneron, CareDx and Hansa Biopharma, is a consultant to Recombinetics, reports consulting fees from Hansa Medical, Regeneron, ThermoFisher Scientific, Genentech, CareDx, One Lambda, ITB Med, Sanofi and PPD Development, and reports grant support from Hansa Biopharma, all unrelated to the present work. R.A.M. also reports grant support from United Therapeutics Corporation, PBC. All other authors have no competing interests.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

REFERENCES

  • 1.Sykes M, and Sachs DH (2022). Progress in xenotransplantation: overcoming immune barriers. Nat. Rev. Nephrol. 18, 745–761. 10.1038/s41581-022-00624-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Matas AJ, Montgomery RA, and Schold JD (2023). The Organ Shortage Continues to Be a Crisis for Patients With End-stage Kidney Disease. JAMA Surg. 10.1001/jamasurg.2023.0526. [DOI] [PubMed] [Google Scholar]
  • 3.Wolbrom DH, Kim JI, and Griesemer A (2023). The road to xenotransplantation. Curr. Opin. Organ Transplant. 28, 65–70. 10.1097/MOT.0000000000001055. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Cooper DKC, Gollackner B, and Sachs DH (2002). Will the pig solve the transplantation backlog? Annu. Rev. Med. 53, 133–147. 10.1146/annurev.med.53.082901.103900. [DOI] [PubMed] [Google Scholar]
  • 5.Elisseeff J, Badylak SF, and Boeke JD (2021). Immune and genome engineering as the future of transplantable tissue. N. Engl. J. Med. 385, 2451–2462. 10.1056/NEJMra1913421. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Phelps CJ, Koike C, Vaught TD, Boone J, Wells KD, Chen S-H, Ball S, Specht SM, Polejaeva IA, Monahan JA, et al. (2003). Production of alpha 1,3-galactosyltransferase-deficient pigs. Science 299, 411–414. 10.1126/science.1078942. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Yamada K, Yazawa K, Shimizu A, Iwanaga T, Hisashi Y, Nuhn M, O’Malley P, Nobori S, Vagefi PA, Patience C, et al. (2005). Marked prolongation of porcine renal xenograft survival in baboons through the use of alpha1,3-galactosyltransferase gene-knockout donors and the cotransplantation of vascularized thymic tissue. Nat. Med. 11, 32–34. 10.1038/nm1172. [DOI] [PubMed] [Google Scholar]
  • 8.Cooper DKC, Dorling A, Pierson RN, Rees M, Seebach J, Yazer M, Ohdan H, Awwad M, and Ayares D (2007). Alpha1,3-galactosyltransferase gene-knockout pigs for xenotransplantation: where do we go from here? Transplantation 84, 1–7. 10.1097/01.tp.0000260427.75804.f2. [DOI] [PubMed] [Google Scholar]
  • 9.Pintore L, Paltrinieri S, Vadori M, Besenzon F, Cavicchioli L, De Benedictis GM, Calabrese F, Cozzi E, Nottle MB, Robson SC, et al. (2013). Clinicopathological findings in non-human primate recipients of porcine renal xenografts: quantitative and qualitative evaluation of proteinuria. Xenotransplantation 20, 449–457. 10.1111/xen.12063. [DOI] [PubMed] [Google Scholar]
  • 10.Ekser B, Rigotti P, Gridelli B, and Cooper DKC (2009). Xenotransplantation of solid organs in the pig-to-primate model. Transpl. Immunol. 21, 87–92. 10.1016/j.trim.2008.10.005. [DOI] [PubMed] [Google Scholar]
  • 11.Montgomery RA, Stern JM, Lonze BE, Tatapudi VS, Mangiola M, Wu M, Weldon E, Lawson N, Deterville C, Dieter RA, et al. (2022). Results of Two Cases of Pig-to-Human Kidney Xenotransplantation. N. Engl. J. Med. 386, 1889–1898. 10.1056/NEJMoa2120238. [DOI] [PubMed] [Google Scholar]
  • 12.Porrett PM, Orandi BJ, Kumar V, Houp J, Anderson D, Cozette Killian A, Hauptfeld-Dolejsek V, Martin DE, Macedon S, Budd N, et al. (2022). First clinical-grade porcine kidney xenotransplant using a human decedent model. Am. J. Transplant. 22, 1037–1053. 10.1111/ajt.16930. [DOI] [PubMed] [Google Scholar]
  • 13.Halloran PF, Venner JM, Madill-Thomsen KS, Einecke G, Parkes MD, Hidalgo LG, and Famulski KS (2018). Review: The transcripts associated with organ allograft rejection. Am. J. Transplant. 18, 785–795. 10.1111/ajt.14600. [DOI] [PubMed] [Google Scholar]
  • 14.Varma E, Luo X, and Muthukumar T (2021). Dissecting the human kidney allograft transcriptome: single-cell RNA sequencing. Curr. Opin. Organ Transplant. 26, 43–51. 10.1097/MOT.0000000000000840. [DOI] [PubMed] [Google Scholar]
  • 15.Wu H, Malone AF, Donnelly EL, Kirita Y, Uchimura K, Ramakrishnan SM, Gaut JP, and Humphreys BD (2018). Single-Cell Transcriptomics of a Human Kidney Allograft Biopsy Specimen Defines a Diverse Inflammatory Response. J. Am. Soc. Nephrol. 29, 2069–2080. 10.1681/ASN.2018020125. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Malone AF, Wu H, Fronick C, Fulton R, Gaut JP, and Humphreys BD (2020). Harnessing expressed single nucleotide variation and single cell RNA sequencing to define immune cell chimerism in the rejecting kidney transplant. J. Am. Soc. Nephrol. 31, 1977–1986. 10.1681/ASN.2020030326. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Andrijevic D, Vrselja Z, Lysyy T, Zhang S, Skarica M, Spajic A, Dellal D, Thorn SL, Duckrow RB, Ma S, et al. (2022). Cellular recovery after prolonged warm ischaemia of the whole body. Nature 608, 405–412. 10.1038/s41586-022-05016-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.McInnes L, Healy J, Saul N, and Großberger L (2018). UMAP: uniform manifold approximation and projection. JOSS 3, 861. 10.21105/joss.00861. [DOI] [Google Scholar]
  • 19.Magnone M, Holley JL, Shapiro R, Scantlebury V, McCauley J, Jordan M, Vivas C, Starzl T, and Johnson JP (1995). Interferon-alpha-induced acute renal allograft rejection. Transplantation 59, 1068–1070. 10.1097/00007890-199504150-00030. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Hidalgo LG, and Halloran PF (2002). Role of IFN-gamma in allograft rejection. Crit. Rev. Immunol. 22, 317–349. [PubMed] [Google Scholar]
  • 21.Roufosse C, Simmonds N, Clahsen-van Groningen M, Haas M, Henriksen KJ, Horsfield C, Loupy A, Mengel M, Perkowska-Ptasińska A, Rabant M, et al. (2018). A 2018 reference guide to the banff classification of renal allograft pathology. Transplantation 102, 1795–1814. 10.1097/TP.0000000000002366. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Nankivell BJ, and Alexander SI (2010). Rejection of the kidney allograft. N. Engl. J. Med. 363, 1451–1462. 10.1056/NEJMra0902927. [DOI] [PubMed] [Google Scholar]
  • 23.Kenta I, and Takaaki K (2020). Molecular Mechanisms of Antibody-Mediated Rejection and Accommodation in Organ Transplantation. Nephron 144 Suppl 1, 2–6. 10.1159/000510747. [DOI] [PubMed] [Google Scholar]
  • 24.Venner JM, Famulski KS, Badr D, Hidalgo LG, Chang J, and Halloran PF (2014). Molecular landscape of T cell-mediated rejection in human kidney transplants: prominence of CTLA4 and PD ligands. Am. J. Transplant. 14, 2565–2576. 10.1111/ajt.12946. [DOI] [PubMed] [Google Scholar]
  • 25.Teng L, Shen L, Zhao W, Wang C, Feng S, Wang Y, Bi Y, Rong S, Shushakova N, Haller H, et al. (2022). SLAMF8 Participates in Acute Renal Transplant Rejection via TLR4 Pathway on Pro-Inflammatory Macrophages. Front. Immunol. 13, 846695. 10.3389/fimmu.2022.846695. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Khamissi FZ, Ning L, Kefaloyianni E, Dun H, Arthanarisami A, Keller A, Atkinson JJ, Li W, Wong B, Dietmann S, et al. (2022). Identification of kidney injury released circulating osteopontin as causal agent of respiratory failure. Sci. Adv. 8, eabm5900. 10.1126/sciadv.abm5900. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Sinha SK, Mellody M, Carpio MB, Damoiseaux R, and Nicholas SB (2023). Osteopontin as a biomarker in chronic kidney disease. Biomedicines 11. 10.3390/biomedicines11051356. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Patel-Chamberlin M, Wang Y, Satirapoj B, Phillips LM, Nast CC, Dai T, Watkins RA, Wu X, Natarajan R, Leng A, et al. (2011). Hematopoietic growth factor inducible neurokinin-1 (Gpnmb/Osteoactivin) is a biomarker of progressive renal injury across species. Kidney Int. 79, 1138–1148. 10.1038/ki.2011.28. [DOI] [PubMed] [Google Scholar]
  • 29.Adler S (2010). Novel kidney injury biomarkers. J. Ren. Nutr. 20, S15–8. 10.1053/j.jrn.2010.05.005. [DOI] [PubMed] [Google Scholar]
  • 30.Cheng C-W, Rifai A, Ka S-M, Shui H-A, Lin Y-F, Lee W-H, and Chen A (2005). Calcium-binding proteins annexin A2 and S100A6 are sensors of tubular injury and recovery in acute renal failure. Kidney Int. 68, 2694–2703. 10.1111/j.1523-1755.2005.00740.x. [DOI] [PubMed] [Google Scholar]
  • 31.Chen Z, Li Y, Yuan Y, Lai K, Ye K, Lin Y, Lan R, Chen H, and Xu Y (2023). Single-cell sequencing reveals homogeneity and heterogeneity of the cytopathological mechanisms in different etiology-induced AKI. Cell Death Dis. 14, 318. 10.1038/s41419-023-05830-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Loupy A, Goutaudier V, Giarraputo A, Mezine F, Morgand E, Robin B, Khalil K, Mehta S, Keating B, Dandro A, et al. (2023). Immune response after pig-to-human kidney xenotransplantation: a multimodal phenotyping study. Lancet 402, 1158–1169. 10.1016/S0140-6736(23)01349-1. [DOI] [PubMed] [Google Scholar]
  • 33.Spahn JH, Li W, and Kreisel D (2014). Innate immune cells in transplantation. Curr. Opin. Organ Transplant. 19, 14–19. 10.1097/MOT.0000000000000041. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Puga Yung G, Schneider MKJ, and Seebach JD (2017). The Role of NK Cells in Pig-to-Human Xenotransplantation. J. Immunol. Res. 2017, 4627384. 10.1155/2017/4627384. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Itescu S, Kwiatkowski P, Artrip JH, Wang SF, Ankersmit J, Minanov OP, and Michler RE (1998). Role of natural killer cells, macrophages, and accessory molecule interactions in the rejection of pig-to-primate xenografts beyond the hyperacute period. Hum. Immunol. 59, 275–286. 10.1016/s0198-8859(98)00026-3. [DOI] [PubMed] [Google Scholar]
  • 36.Parkes MD, Halloran PF, and Hidalgo LG (2017). Evidence for CD16a-Mediated NK Cell Stimulation in Antibody-Mediated Kidney Transplant Rejection. Transplantation 101, e102–e111. 10.1097/TP.0000000000001586. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Lin Y, Vandeputte M, and Waer M (1997). Natural killer cell- and macrophage-mediated rejection of concordant xenografts in the absence of T and B cell responses. J. Immunol. 158, 5658–5667. [PubMed] [Google Scholar]
  • 38.Khalfoun B, Barrat D, Watier H, Machet MC, Arbeille-Brassart B, Riess JG, Salmon H, Gruel Y, Bardos P, and Lebranchu Y (2000). Development of an ex vivo model of pig kidney perfused with human lymphocytes. Analysis of xenogeneic cellular reactions. Surgery 128, 447–457. 10.1067/msy.2000.107063. [DOI] [PubMed] [Google Scholar]
  • 39.Ezzelarab M, Garcia B, Azimzadeh A, Sun H, Lin CC, Hara H, Kelishadi S, Zhang T, Lin YJ, Tai H-C, et al. (2009). The innate immune response and activation of coagulation in alpha1,3-galactosyltransferase gene-knockout xenograft recipients. Transplantation 87, 805–812. 10.1097/TP.0b013e318199c34f. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Kim SC, Mathews DV, Breeden CP, Higginbotham LB, Ladowski J, Martens G, Stephenson A, Farris AB, Strobert EA, Jenkins J, et al. (2019). Long-term survival of pig-to-rhesus macaque renal xenografts is dependent on CD4 T cell depletion. Am. J. Transplant. 19, 2174–2185. 10.1111/ajt.15329. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Collins BJ, Blum MG, Parker RE, Chang AC, Blair KS, Zorn GL, Christman BW, and Pierson RN (2001). Thromboxane mediates pulmonary hypertension and lung inflammation during hyperacute lung rejection. J. Appl. Physiol. 90, 2257–2268. 10.1152/jappl.2001.90.6.2257. [DOI] [PubMed] [Google Scholar]
  • 42.Toback FG (1992). Regeneration after acute tubular necrosis. Kidney Int. 41, 226–246. 10.1038/ki.1992.32. [DOI] [PubMed] [Google Scholar]
  • 43.Humphreys BD, Valerius MT, Kobayashi A, Mugford JW, Soeung S, Duffield JS, McMahon AP, and Bonventre JV (2008). Intrinsic epithelial cells repair the kidney after injury. Cell Stem Cell 2, 284–291. 10.1016/j.stem.2008.01.014. [DOI] [PubMed] [Google Scholar]
  • 44.Cochrane AL, Kett MM, Samuel CS, Campanale NV, Anderson WP, Hume DA, Little MH, Bertram JF, and Ricardo SD (2005). Renal structural and functional repair in a mouse model of reversal of ureteral obstruction. J. Am. Soc. Nephrol. 16, 3623–3630. 10.1681/ASN.2004090771. [DOI] [PubMed] [Google Scholar]
  • 45.Wen Y, Su E, Xu L, Menez S, Moledina DG, Obeid W, Palevsky PM, Mansour SG, Devarajan P, Cantley LG, et al. (2023). Analysis of the human kidney transcriptome and plasma proteome identifies markers of proximal tubule maladaptation to injury. Sci. Transl. Med. 15, eade7287. 10.1126/scitranslmed.ade7287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Benigni A, Morigi M, and Remuzzi G (2010). Kidney regeneration. Lancet 375, 1310–1317. 10.1016/S0140-6736(10)60237-1. [DOI] [PubMed] [Google Scholar]
  • 47.Witzgall R, Brown D, Schwarz C, and Bonventre JV (1994). Localization of proliferating cell nuclear antigen, vimentin, c-Fos, and clusterin in the postischemic kidney. Evidence for a heterogenous genetic response among nephron segments, and a large pool of mitotically active and dedifferentiated cells. J. Clin. Invest. 93, 2175–2188. 10.1172/JCI117214. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Gerhardt LMS, Liu J, Koppitch K, Cippà PE, and McMahon AP (2021). Single-nuclear transcriptomics reveals diversity of proximal tubule cell states in a dynamic response to acute kidney injury. Proc Natl Acad Sci USA 118. 10.1073/pnas.2026684118. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Ferenbach DA, and Bonventre JV (2015). Mechanisms of maladaptive repair after AKI leading to accelerated kidney ageing and CKD. Nat. Rev. Nephrol. 11, 264–276. 10.1038/nrneph.2015.3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Franquesa M, Flaquer M, Cruzado JM, and Grinyó JM (2013). Kidney regeneration and repair after transplantation. Curr. Opin. Organ Transplant. 18, 191–196. 10.1097/MOT.0b013e32835f0771. [DOI] [PubMed] [Google Scholar]
  • 51.Nieuwenhuijs-Moeke GJ, Pischke SE, Berger SP, Sanders JSF, Pol RA, Struys MMRF, Ploeg RJ, and Leuvenink HGD (2020). Ischemia and reperfusion injury in kidney transplantation: relevant mechanisms in injury and repair. J. Clin. Med. 9. 10.3390/jcm9010253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Butler A, Hoffman P, Smibert P, Papalexi E, and Satija R (2018). Integrating single-cell transcriptomic data across different conditions, technologies, and species. Nat. Biotechnol. 36, 411–420. 10.1038/nbt.4096. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.Duan G, Wu G, Chen X, Tian D, Li Z, Sun Y, Du Z, Hao L, Song S, Gao Y, et al. (2023). HGD: an integrated homologous gene database across multiple species. Nucleic Acids Res. 51, D994–D1002. 10.1093/nar/gkac970. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54.Stuart T, Butler A, Hoffman P, Hafemeister C, Papalexi E, Mauck WM, Hao Y, Stoeckius M, Smibert P, and Satija R (2019). Comprehensive Integration of Single-Cell Data. Cell 177, 1888–1902.e21. 10.1016/j.cell.2019.05.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55.Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP, Dolinski K, Dwight SS, Eppig JT, et al. (2000). Gene Ontology: tool for the unification of biology. Nat. Genet. 25, 25–29. 10.1038/75556. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.Kanehisa M, and Goto S (2000). KEGG: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 28, 27–30. 10.1093/nar/28.1.27. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 57.Wu T, Hu E, Xu S, Chen M, Guo P, Dai Z, Feng T, Zhou L, Tang W, Zhan L, et al. (2021). clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation (Camb) 2, 100141. 10.1016/j.xinn.2021.100141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 58.Benjamini Y, and Hochberg Y (1995). Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society: Series B (Methodological) 57, 289–300. 10.1111/j.2517-6161.1995.tb02031.x. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1

Data S1: Bulk RNA-seq of porcine kidney biopsies during the course of xenotransplantation, related to Figure 2.

NIHMS1997640-supplement-1.xlsx (3.1MB, xlsx)
2

Data S2: Gene sets enriched at 12 hours pXTx in PBMC single-cell data, related to Figure 5.

NIHMS1997640-supplement-2.xlsx (134.1KB, xlsx)
3

Data S3: Gene sets enriched at 48–53 hours pXTx in PBMC single-cell data, related to Figure 5.

NIHMS1997640-supplement-3.xlsx (58.4KB, xlsx)
5
NIHMS1997640-supplement-5.pdf (26.5MB, pdf)

Data Availability Statement

All de-identified single-cell RNA-seq and bulk RNA-seq sample data from this study have been deposited to the GEO portal. Accession numbers are listed in the Key Resources Table. De-identified data of sequenced samples and all original code used for analysis in this study has been deposited in the Github repository: github.com/boxialaboratory/Pig-to-Human-Kidney-Xenotransplantation.

KEY RESOURCES TABLE.

REAGENT or RESOURCE SOURCE IDENTIFIER
Biological Samples
Porcine kidney tissue (untransplanted and xenograft) NYU Langone Transplant Institute NA
Human PBMC NYU Langone Transplant Institute NA
Chemicals, Peptides, and Recombinant Proteins
DPBS buffer, no calcium, no magnesium Gibco Cat#14190250
Trypsin-EDTA (0.25%) Gibco Cat#25200056
TURBO DNase Invitrogen Cat#AM2238
ACK lysing buffer Gibco Cat#A1049201
Critical Commercial Assays
Chromium Single Cell 3’ Kits (v3.1 Chemistry) 10x Genomics Cat#1000268
Dead cell removal kit Miltenyi Biotec Cat#130-090-101
RNeasy Plus kit QIAGEN Cat#74134
RNA 6000 Nano kit Agilent Cat#5067-1511
Ribo-Zero Plus rRNA Depletion Kit Illumina Cat#20037135
Countess II Automated Cell Counting Chamber Slides Invitrogen Cat#10228
Deposited Data
Raw single-cell RNA-seq data This paper GEO: GSE257542
Raw bulk RNA-seq data This paper GEO: GSE257541
HGD Duan et al.53 doi:10.1016/j.xinn.2021.100141
Gene Ontology Ashburner et al.55 doi: 10.1038/75556
Kyoto Encyclopedia of Genes and Genomes Kanehisa et al.56 doi: 10.1093/nar/28.1.27
Experimental Models: Organisms/Strains
Pig: αGTase-KO: GGTA1−/− Revivicor NA
Software and Algorithms
CellRanger 10X Genomics https://support.10xgenomics.com/single-cell-gene-expression/software/downloads/latest
Bioanalyzer 2100 Agilent https://www.agilent.com/en/product/automated-electrophoresis/bioanalyzer-systems/bioanalyzer-software
Seurat v4.3 Satija et al.52 https://doi.org/10.1038/nbt.3192
clusterProfiler v4.8.2 Wu et al.57 doi:10.1016/j.xinn.2021.100141.
R 4.3 R Core Team https://www.r-project.com
R code for analysis This paper https://github.com/boxialaboratory/Pig-to-Human-Kidney-Xenotransplantation/tree/main/scripts
doi: 10.5281/zenodo.11068840
Open in a new tab

Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

RESOURCES