Ischemia occurs when the blood supply to tissue becomes restricted, causing a shortage of oxygen that is needed for the production of adenosine triphosphate (ATP). As a result of ischemia, cells switch from highly efficient aerobic respiration (i.e., 29–30 ATP molecules per glucose molecule) to far less efficient anaerobic respiration (i.e., 2 ATP molecules per glucose molecule) (26, 27). This causes a decline in intracellular ATP levels and increased levels of lactate, which contribute to intracellular acidosis and cell membrane injury (Figure 3) (28). ATP depletion also affects intracellular calcium (Ca²⁺) levels because Ca²⁺ pumps, which are ATP-dependent, cannot pump Ca²⁺ out of cells (29). Increased levels of intracellular Ca²⁺ subsequently cause activation of various calcium-dependent enzymes, such as calpains, endonucleases, and phospholipases. In turn, these enzymes contribute to the generation of reactive oxygen species (ROS) and the breakdown of the cell membrane, among other processes (30). It is therefore unsurprising that epithelial cells, in particular proximal tubular cells, are highly susceptible to ischemia, given the fact that these cells require a lot of ATP for mediating ion transport (31).

Upon reperfusion, the blood supply is restored, and the ischemic environment induces an inflammatory response. The re-introduction of oxygen, rewarming of the kidney graft, and pH normalization is detrimental to the previously ischemic cells. Sudden increases in oxygen levels lead to extensive ROS production. The production of ROS is normally counteracted by antioxidants, but the high levels after reperfusion overwhelm this protection mechanism (32). pH normalization leads to a further increase in intracellular Ca²⁺ levels that, together with the generated ROS, further activate calcium-dependent enzymes (e.g., calpains, etc.). Activated calpains promote inflammation and damage cells by selectively degrading a large number of intracellular proteins, including signaling proteins, cytoskeletal proteins, and transcription factors (33, 34). The electron transport chain may also become unstable. This causes the formation of mitochondrial permeability transition pores which, in turn, promote the release of cytochrome C into the cytosol to activate caspases—the key effector proteins in apoptosis (30).

Fibrogenesis represents a well-coordinated, multicellular regenerative response that is triggered upon injury (Figure 4). When renal tissue gets injured, an inflammatory response is elicited, involving various immune cells, such as T-cells, dendritic cells, macrophages, and neutrophils (40). Triggered by inflammatory cytokines, including but not limited to tumor necrosis factor alpha (TNF-α), interleukin 1 beta (IL-1β), interleukin 6 (IL-6), and transforming growth factor beta 1 (TGF-β1), these immune cells are recruited to affected sites and become activated to remove tissue debris as well as dead and damaged cells (41). TGF-β1 in particular also promotes the accumulation of myofibroblasts, which are key effector cells in wound-healing. Myofibroblasts have been shown to arise from various sources, namely: proliferation and differentiation of resident interstitial fibroblasts (50%), infiltration and differentiation of fibrocytes (35%), endothelial-to-mesenchymal transition (EndoMT) (20%), and epithelial-to-mesenchymal transition (EMT) (5%) (42–45). The extent of EndoMT and EMT, however, remains controversial and it is currently unclear whether these processes plays a substantial role in humans. Once myofibroblasts become activated, characterized by de novo expression of alpha smooth muscle actin (α-SMA), these cells aim to restore tissue integrity by producing and secreting ECM proteins, especially collagens and fibronectins (43).

The provisional, collagen-rich matrix provides a scaffold for cell growth (e.g., to facilitate tubular regeneration). In normal conditions, the renewal rate of tubular epithelial cells is low. This rate, however, drastically increases after (ischemic) injury, with the aim of restoring the tubular epithelium via dedifferentiation (46). Indeed, fully differentiated tubular epithelial cells have been shown to repair proximal tubules without the contribution of intralobular stem cells. This process comprises four stages. Stage one is characterized by apoptosis or necrosis of damaged tubular epithelial cells as well as inflammation. In the second stage, tubular cells flatten and lose their brush border as well as polarity. EMT has also been shown to occur in this stage. Stage three is marked by increased secretion of growth factors, which stimulate tubular epithelial cells to enter the cell-division cycle. In the last stage, the tubules become repopulated with daughter-cells, and morphological features of tubular structures recover (46). After these processes, kidney function is restored.

In some cases, the wound-healing process becomes severely dysregulated, resulting in the uncontrolled deposition of ECM proteins into the intertubular, extraglomerular, extravascular space of the kidney (Figure 5) (47, 48). This phenomenon is called “interstitial fibrosis.” Fibrosis is a disorder in which the excessive deposition of ECM proteins afflict the kidney architecture and function. The root-cause of the shift from fibrogenesis to fibrosis remains unclear. Some aspects, however, have been shown to play a role, such as the persistence and duration of injury. Chronic injury, for example, contributes to the unbridled recruitment of immune cells as well as activation of myofibroblasts. Given enough time, the ECM changes with respect to its composition (i.e., collagen and fibronectin-rich), organization (i.e., densely crosslinked fibers), and stiffness. In fibrosis, these changes to the ECM are thought to have passed a point-of-no-return, wherein the ECM itself starts to perpetuate myofibroblast activation. One of the key mediators during this pathophysiological process is TGF-β, which binds to its cell membrane type I and II receptor, causing receptor phosphorylation and so activation of its downstream signaling mediators SMAD2 and SMAD3 (Figure 5) (49).

Interstitial fibrosis is often observed together with tubular atrophy, as both pathologies can be the result of IRI (50). TA is characterized by the disappearance and/or flattening of tubular epithelial cells, contraction of tubular lumens, and thickening as well as wrinkling of tubular basement membranes (Figure 5) (15). Unsurprisingly, these processes reduce the tubular reabsorption of water and small molecules. Like fibrosis, the pathogenesis of TA involves various mechanisms, of which the extent of involvement has not been fully elucidated. Apoptosis, for example, could become persistent, effectively reducing the number of cells as a result of (chronic) injury (50). Schelling have also observed increased levels of senescence in remaining epithelial cells (50). Senescent cells have lost the ability to divide and therefore further impede tubular dedifferentiation. Sometimes, chronic histological damage is already visible as early as 3 months post-transplantation (51).

SCS is the cheapest and simplest technique for preserving donor kidneys. When using this approach, kidney grafts are first flushed with ice-cold preservation solution and then stored on ice until further use (52, 54). So far, great efforts have been put into the development of preservation solutions, each with the purpose of maintaining the viability of grafts. The benefits and drawbacks of various solutions are discussed in an extensive review written by Hosgood et al. (55). Nonetheless, SCS suffers from several limitations, such as the fact that hypothermic and ischemic conditions as well as nutritional deficiencies do not fully stop cellular respiration or prevent the formation of metabolic waste products (e.g., lactate). As a result, cells switch to anaerobic respiration, depleting energy reserves and leading to acidification of the cellular environment (52, 54, 56). The cold ischemia time (CIT) is therefore directly associated with DGF (21, 23). Each additional hour of CIT, for example, significantly increases the risk of graft failure and mortality, especially for grafts from extended criteria and DCD donors (24, 57).

Hypothermic machine perfusion (HMP) has been introduced as an alternative to SCS because it offers superior organ preservation, especially for grafts of suboptimal quality. HMP refers to the controlled flow of perfusion solution through grafts at a temperature of 4°C (58). University of Wisconsin machine perfusion (UW-MP) solution is used most often. UW-MP contains nutrients and metabolites to support the reduced cellular metabolism during HMP (59). It has been shown that HMP protects endothelial cells and reduces pro-inflammatory cytokine expression (e.g., TNF-α and IL-1β) (60–62). In a seminal study, Moers et al. (63) described that, in comparison to SCS, HMP led to reduced serum creatinine levels post-transplantation and a reduced risk of graft failure. On top of that, the use of HMP significantly increased the 1-year graft survival rate compared to SCS (94 vs. 90%, p = 0.04).

Nowadays, HMP is frequently performed using oxygenated perfusion solution, as it has been shown to promote ATP synthesis, preserve mitochondrial homeostasis and reduce interstitial fibrosis in porcine kidneys (64–66). Jochmans et al. (67) demonstrated that oxygenation (100% oxygen at 100 mL/min) lessened post-transplant complications and significantly increased the 1-year graft survival rate compared to HMP without oxygen (95 vs. 89%, p = 0.028). Therefore, oxygenated HMP may be a promising therapy to reduce the odds of developing IF/TA, but this will require long-term follow-up studies.

Last but not least, HMP allows for sampling of the perfusate to assess the viability, as elegantly shown by Moers et al. (68) who analyzed glutathione-S-transferase, N-acetyl-β-D-glucosaminidase, and heart-type fatty acid binding protein to predict DGF. Emerging techniques, such as proteomics, also appear promising, as reported by van Leeuwen et al. (69), who found out that the HMP perfusate composition associates with the 1-year transplant outcome.

Normothermic machine perfusion (NMP) represents an advanced form of preservation. During NMP, organs are perfused with a blood-based solution at 37°C, mimicking the physiological conditions to support metabolic activity and organ function (52). The perfusate composition between NMP protocols varies greatly, as the exact metabolic needs of an isolated kidney are still not fully characterized (70, 71). Having said that, studies have revealed the importance of using an oxygen carrier (e.g., red blood cells or artificial oxygen carriers), even when supraphysiological concentrations of oxygen are used (72, 73). Short-term outcomes for porcine kidneys that were subjected to NMP have shown to be superior compared to those subjected to SCS (74–76). Studies comparing HMP and NMP have not revealed differences in transplant outcomes yet (77).

Like HMP, NMP also allows for sampling and subsequent analysis of perfusate. NMP, however, offers another benefit as it can be used to assess kidney function (i.e., urine production, creatinine clearance, and fractional sodium excretion) as metabolism is restored. Carefully monitoring graft function during NMP and measuring biomarkers in the perfusate can facilitate decision-making during transplantation (52, 78, 79). In spite of that, standardized viability criteria are lacking because no one has managed to predict post-transplant outcomes yet. Although NMP itself seems to offer superior graft preservation in comparison to SCS, supplementing perfusate with pharmacologically-active compounds may further improve the preservation of metabolically active donor kidneys.

Current approaches for suppressing fibrosis are mostly focused on minimizing damage to kidney grafts (e.g., donor age, ischemia time, and machine perfusion). With the shortage of donor kidneys, however, these strategies are not sufficient. Supplementing NMP solutions with anti-fibrotic drugs could be a promising strategy to suppress the onset of fibrosis and, in turn, reduce graft failure. Unfortunately, research into this area is limited, as conventional models such as cell cultures and animal studies have a limited predictive value when it comes to assessing the efficacy of anti-fibrotic drugs. Cell cultures are useful for investigating cell-specific responses but do not recapitulate the intricate cell-cell and cell-matrix interactions as well as mechanical cues that are observed in fibrosis. Of course, animal models using rodents provide considerably more useful data, owing to the presence of a functional immune system and interorgan crosstalk, yet interspecies differences remain (79–81). We therefore carefully considered the models used whilst selecting relevant literature on anti-fibrotic compounds. It was also important to take into account the potency of compounds as well as the rate at which anti-fibrotic effects were produced. These factors are crucial, as the maximum NMP time is currently limited to 24 h (80).

To identify promising anti-fibrotic compounds, we reviewed literature describing the use of precision-cut kidney slices (PCKS). PCKS are viable explants, with consistent dimensions that can be used to study fibrogenesis and fibrosis as well as the efficacy of anti-fibrotic compounds (81). The advantage of PCKS is that they contain a large number and diversity of cells, accurately reflecting the renal architecture (82). On top of that, PCKS can be prepared from (fibrotic) human tissue, creating an unprecedented opportunity to validate the molecular basis of anti-fibrotic drug candidates in humans. PCKS can also be quickly prepared in a reproducible manner by means of a specialized workflow (Figure 7). In short, cylindrical cores are prepared from kidney tissue using a biopsy puncher, after which they are placed into a Krumdieck tissue slicer and sliced under hypothermic and oxygenated conditions. PCKS are subsequently transferred to culture plates containing fresh and pre-warmed culture medium and are then incubated at 37°C, 80% O₂, and 5% CO₂, whilst being gently shaken (82). Taking into account the advantages and flexibility of PCKS (e.g., using different timepoints, compound concentrations etc.), it becomes clear that their use represents a valuable intermediate step before assessing anti-fibrotic compounds in an NMP setup.

Several drug candidates have been shown to have anti-fibrotic effects on PCKS (Table 1). Each of these small-molecule drugs produced effects already within 48 h of incubation. We hypothesize that these compounds could be promising for use during NMP, as each inhibited fibrosis via the TGF-β/SMAD signaling pathway (Figure 8). The respective studies will be discussed in this section.

Apart from small-molecule drugs, no other candidates have been tested ex vivo with the purpose of attenuating fibrogenesis and/or fibrosis in kidney tissue. Therefore, it would be particularly useful to investigate other therapeutic modalities as well, including therapeutic proteins and monoclonal antibodies, or perhaps even more advanced approaches, such as cell-based therapies or (lipid-based) nanoparticles encapsulating nucleic acids for transiently silencing or enhancing the expression of specific genes. Combinatorial therapies, affecting various stages of renal injury and fibrogenesis, could also be developed. Some of the previously described modalities have already been tested in kidneys to attenuate IRI, but not with the focus on fibrosis (105, 106). Of course, there are some points to consider when developing such therapies. It is important to investigate the distribution kinetics of respective modalities in donor kidneys when administered during NMP. Cells, for instance, are unlikely to uniformly distribute through the tissue. The same applies to nanoparticles. Lastly, when supplementing perfusion solutions with therapeutic proteins and/or monoclonal antibodies, it is imperative to check whether shear stress affects their stability (e.g., denaturation, aggregation, etc.) (107).