Kidney Tissue Proteome Profiles in Short Versus Long Duration of Delayed Graft Function - A Pilot Study in Donation After Circulatory Death Donors

Introduction Delayed graft function (DGF) is often defined as the need for dialysis treatment in the first week after a kidney transplantation. This definition, though readily applicable, is generic and unable to distinguish between “types” of DGF or time needed to recover function that may also significantly affect longer-term outcomes. We aimed to profile biological pathways in donation after circulatory death (DCD) kidney donors that correlate with DGF and different DGF durations. Methods A total of N = 30 DCD kidney biopsies were selected from the UK Quality in Organ Donation (QUOD) biobank and stratified according to DGF duration (immediate function, IF n = 10; “short-DGF” (1–6 days), SDGF n = 10; “long-DGF” (7–22 days), LDGF n = 10). Samples were matched for donor and recipient demographics and analyzed by label-free quantitative (LFQ) proteomics, yielding identification of N = 3378 proteins. Results Ingenuity pathway analysis (IPA) on differentially abundant proteins showed that SDGF kidneys presented upregulation of stress response pathways, whereas LDGF presented impaired response to stress, compared to IF. LDGF showed extensive metabolic deficits compared to IF and SDGF. Conclusion DCD kidneys requiring dialysis only in the first week posttransplant present acute cellular injury at donation, alongside repair pathways upregulation. In contrast, DCD kidneys requiring prolonged dialysis beyond 7 days present minimal metabolic and antioxidant responses, suggesting that current DGF definitions might not be adequate in distinguishing different patterns of injury in donor kidneys contributing to DGF.

10,000xg for 10min.20μg of protein for each sample were added to the filter and centrifuged at 13,000xg for 20min.Samples were reduced by addition of 200mM dithiothreitol (DTT, Sigma Aldrich, Gillingham, UK) and centrifuged at 13,000xg for 20min.Cysteine residues were alkylated by incubation with 200mM iodoacetamide (in 8M urea buffer, Sigma Aldrich) for 30min.After incubation, filters were centrifuged at 10,000xg for 10min and washed with 250μL of 8M urea (in Tris/HCl, pH 8.5) by centrifuging at 13,000xg for 10min.Afterwards, samples were washed twice with 250μL of 0.05M NH5CO3 and centrifuged at 13,000xg for 10min.Filter units were transferred to fresh collection tubes and trypsin was added at a 1:50 (trypsin:protein) ratio.Trypsin digestion was performed overnight at 37°C with gentle agitation.Upon completion of digestion, samples were briefly vortexed, and filters centrifuged upside down at 5,000xg for 1min to collect samples.Subsequently, 20uL of 1% Formic Acid (FA) were added to stop the trypsin reaction.One hundred (100) μL of 0.5M NaCl were added to the filters and centrifuged upside down at 5,000xg for 1min.Finally, to ensure complete peptide recovery, 100μL of MilliQ-H2O were added and centrifuged at 10,000xg for 10min.

Peptide purification
Peptide digests were purified and desalted on a C18 reverse phase column (Sep-Pak light C18 cartridges, Waters, Dublin, Ireland) according to manufacturer's instructions.Briefly, the columns were equilibrated with 5mL buffer B (65% Acetonitrile (ACN), 35% MilliQ-H2O, 0.1% formic acid (FA)), followed by 10mL buffer A (98% MilliQ-H2O, 2% ACN, 0.1% FA).Samples were then loaded into the columns, washed with 10mL buffer A and subsequently eluted using 600μL buffer B twice.Peptide fractions were dried by centrifugation under vacuum overnight.Pellets were re-suspended in 40μL of buffer A with final concentration of 0.5 µg/µL, and 1μL was analyzed by LC-MS/MS.

LC-MS/MS for protein identification
Equal amounts (500ng) of peptide material were analysed by liquid chromatography-tandem mass spectrometry (LC-MS/MS), using nano-UHPLC coupled to a hybrid quadrupole-orbitrap mass spectrometer (Q Exactive, Thermo Scientific) as described in 16 .Briefly, peptides were separated by a C18 Easy spray column (0.75µm x 50cm, Thermo Fisher) at a flow rate of 250 μL/min, using a 60minute linear gradient from 97% buffer A (H2O with 0.1% FA) to 40% buffer B (90% ACN with 0.1% FA).After separation, the peptides were ionized by electrospray ionization and injected into the mass spectrometer.Higher energy collisional dissociation (HCD) was induced on the ten most abundant ions per full MS scan to produce MS/MS spectra.

Western blot validation of mass spectrometry results
An independent set of n=13 samples was selected from the QUOD biobank.Proteins were extracted from the biopsies by homogenization with RIPA buffer and sonication.Protein concentration was determined by BCA assay (Pierce Thermo Scientific, Life technologies ltd, Paisley, UK) and 7ug protein/lane were loaded on a Bis-Tris 4-12% Criterion gel (Bio-Rad), following sample reduction.The following antibodies at the following dilutions were used after transfer of the proteins from the gels to PVDF membranes: NGAL 1:1,000 (ab125075, Abcam, Cambridge, UK), Ferritin light chain 1:20,000 (FtL, ab109373, Abcam), RNA-binding protein FUS 1:500 (ab124923, Abcam) and β-actin 1:16,000 (ab6276, Abcam).Membranes were imaged using a Li-COR Clx Odyssey system and IRDye secondary antibodies (Li-COR, Lincoln, NE, USA).Blots were analyzed with the Image Studio lite software.Statistical analysis was performed with GraphPad Prism (version 9.4.1).The differences highlighted in the procurement proteomes of grafts with immediate function, compared to DGF of either short or long duration, suggest that donor molecular factors contribute to initiation and extension of DGF and also support the hypothesis that short and longer duration DGF reflect distinct entities.The fact that the duration of DGF might be a clinical factor of importance was also recently highlighted in a published registry data analysis that indicated that prolonged DGF in DCD kidneys is associated with poorer death-censored graft failure.14

SDGF
The focus of this study was to perform an integrated analysis of the tissue proteome of wellmatched DCD kidneys with opposing DGF outcomes posttransplant, in order to identify molecular changes underpinning DGF.Relative protein expressions were integrated through gene ontology analysis in order to map the molecular pathways differentially expressed in grafts with future DGF.The selected 7day cut-off between short and long duration reflected the median DGF duration for grafts with DGF in the QUOD biobank and is also reflective of current clinical definitions of DGF.However, we also analyzed whether protein expression was significantly correlated with DGF duration as a continuum.
From the pathway analysis of short vs long DGF, a seemingly paradoxical picture emerges with comprehensive activation of stress responses in grafts with short DGF, but not in grafts with longer DGF.Absence of a molecular stress response in long DGF might be ascribed to metabolic deficiencies, also observed in these grafts, which render them unable to sustain energy-demanding processes, including protein transcription and translation.At the proteomic level we were also able to detect metabolic deficit proteolysis in DGF grafts compared to control kidneys and this was present already at time of procurement.While short DGF kidneys present downregulation of aerobic respiration and mitochondrial ATP synthesis compared to immediate function, they also show some resilience and parallel upregulation of peroxisome lipid metabolism, as an alternative energy source and as previously shown in a rodent IRI model.22 This was not the case for long DGF kidneys, which in turn present further downregulation of metabolic pathways (TCA cycle and carboxylic acid metabolism).Hence this metabolic deficit might be behind the apparent lack of stress response in the kidneys with longer DGF and, interestingly, is already present at time of procurement.It appears that these molecular and metabolic features render the kidneys less able to cope with the IR-related stress of the preservation and transplant procedure, causing the grafts to take longer time to recover posttransplant.These features were also confirmed when analyzing the associations between protein expression and DGF duration as a continuum.Proteins involved in gene transcription, protein transport and quality control were all found to be negatively correlated with DGF duration (lower levels of these markers associated with longer DGF).

Limitations
19 Discuss limitations of the study, taking into account sources of potential bias or imprecision.Discuss both direction and magnitude of any potential bias

Figure S1 .
Figure S1.Panther statistical enrichment tests of n=3,378 proteins (and relative fold change) for SDGF vs IF (A) and LDGF vs IF (B).The statistical enrichment test (Mann-Whitney Rank-Sum, U test) analyses if the expression of any ontology class (e.g.GO Biological Processes) significantly deviates from the reference distribution of all proteins in the dataset.A) Top 10 GO biological processes significantly regulated in SDGF vs IF.B) Top 10 GO biological processes significantly regulated in LDGF vs IF.GO terms are presented in the y axis, the x axis reports -log10 (adjusted p values) (p<0.05 and FDR<0.05 after multiple testing correction).

Figure S2 .
Figure S2.Panther statistical enrichment tests of n=3,378 proteins (and relative fold change) for LDGF vs SDGF.The statistical enrichment test (Mann-Whitney Rank-Sum, U test) analyses if the expression of any ontology class (e.g.GO Biological Processes) significantly deviates from the reference distribution of all proteins in the dataset.The figure presents the top 10 GO biological processes significantly regulated in LDGF vs SDGF.GO terms are presented in the y axis, the x axis reports -log10 (adjusted p values) (p<0.05 and FDR<0.05 after multiple testing correction).

Figure S3 .
Figure S3.Representative Western blot analysis validation of NGAL, FUS and Ferritin light chain (FtL).An independent cohort of n=13 DCD donor kidney samples were selected from the QUOD biobank and stratified according to outcome (n=5 SDGF, n=4 LDGF and n=4 IF).Proteins were extracted and loaded on a gel for western blot analysis (A) of NGAL, FUS and FtL.Intensity of the bands was calculated and normalized to the average β-actin signal for all samples.No significant differences were found between the samples (panels B, C and D), despite trends correlating with the mass spectrometry data.

Table S1 . Additional recipient demographics
. For each outcome group (SDGF, LDGF and IF) the table reports the recipient diabetes status (number of recipients with or without diabetes), whether the recipients were on dialysis at time of transplant (yes vs no, n) and the immunosuppression status at time of transplant.Statistical differences between the groups were calculated by Chi-square test for trend.Recipients received combination of drugs for immunosuppression, therefore we tested whether there was a significant difference in the distribution of recipients across drug type.MMF= Mycophenolate Mofetil *Data not available for n=1 recipient.

Table S2 . Top 20 proteins significantly up-or down-regulated in the short DGF (SDGF, n=10) group compared to immediate function (IF, n=10).
Negative log2 fold change values indicate downregulation in SDGF compared to IF.

Table S3 . Top 20 proteins significantly up-or down-regulated in the long DGF (LDGF, n=10) group compared to immediate function (IF, n=10).
Negative log2 fold change values indicate downregulation in LDGF compared to IF.

Table S5 . Representative differentially expressed proteins in SDGF or LDGF with known reported roles in kidney injury. Protein
name, localization and reported role are listed alongside Log2 Fold change.Negative log2 fold change values indicate downregulation.Ns= not significant.
STROBE Statement-checklist of items that should be included in reports of observational studies Participants 6 (a) Cohort study-Give the eligibility criteria, and the sources and methods of selection of participants.Describe methods of follow-up Case-control study-Give the eligibility criteria, and the sources and methods of case ascertainment and control selection.Give the rationale for the choice of cases and controls Cross-sectional study-Give the eligibility criteria, and the sources and methods of selection (d) Cohort study-If applicable, explain how loss to follow-up was addressed Case-control study-If applicable, explain how matching of cases and controls was addressed Funding 22 Give the source of funding and the role of the funders for the present study and, if applicable, for the original study on which the present article is based Give information separately for cases and controls in case-control studies and, if applicable, for exposed and unexposed groups in cohort and cross-sectional studies.Note: An Explanation and Elaboration article discusses each checklist item and gives methodological background and published examples of transparent reporting.The STROBE checklist is best used in conjunction with this article (freely available on the Web sites of PLoS Medicine at http://www.plosmedicine.org/,Annals of Internal Medicine at http://www.annals.org/,and Epidemiology at http://www.epidem.com/).Information on the STROBE Initiative is available at www.strobe-statement.org. *