**4. Serum and Urine Cytokines and Chemokines**

In the recent years, there has been a rise in the number of studies published in the field of urinary biomarkers, seizing the non-invasive nature of the sampling. The introduction and spread of novel multiplex assays that allow multiple simultaneous cytokine measurements in the same procedure has become a valuable tool in this kind of research. The presence in urine of the different chemokines and cytokines evaluated isillustrative of the pathophysiologic processes occurring in this disease.

Monocyte chemoattractant protein-1 (MCP-1) has a key role in the recruitment of monocytes, neutrophils, and lymphocytes in tissue inflammation processes. In the kidney, it is produced by TECs, endothelial cells of the peritubular capillaries, and macrophages themselves. MCP-1 has been identified as a chemokine involved in autoimmune diseases [18]. Wu et al. found higher MCP-1 levels in a cohort of 40 patients with drug-induced ATIN compared to controls. Urinary MCP-1 correlated and predicted the severity of the acute lesions in kidney biopsies from these patients [19]. In the same line, Yun et al. described higher MCP-1 serum and urine concentration among 113 ATIN patients from different causes in their bead-based multiplex assay [20]. Interestingly, Dantas et al. found that urinary MCP-1 concentration finely correlated with the amount of tubulointerstitial infiltrate but not with the glomerular infiltrate in a cohort of patients with glomerular autoimmune diseases [21]. Other authors also reported higher urinary levels of MCP-1 in patients with systemic lupus erythematosus (SLE) that correlated with the extension of the tubulointerstitial infiltrate, thus confirming the utility of MCP-1 in urine as a biomarker of acute infiltration of kidneys with dense affectation of this compartment [22].

Tumor necrosis factor alpha (TNF-α) is a proinflammatory cytokine produced by macrophages and monocytes. It participates in signaling cascades that lead to cell apoptosis or necrosis. Several authors reported higher levels of TNF-α in serum and urine samples from ATIN patients. Moledina et al. reported a higher urinary TNF-α in patients with ATIN that helped in the differential diagnosis of patients with ATN [23]. The same group showed that urinary TNF-α was also higher in a large series of ATIN cases compared to other causes of inflammatory kidney disease, and that urinary TNF-α correlated well with the number of TNF-α-positive cells in the renal biopsy [24]. Aoyagi et al. reported the value of serum TNF-α levels in the follow-up of a case of TINU, which dramatically decreased during the first week after treatment initiation [25]. Moledina et al. found higher urinary IL-9 along with higher TNF-α levels in patients with ATIN. IL-9 is involved in allergic responses and induces mast cell accumulation, which isin turn a source of TNF-α [23,24]. TNF-α inducessecretion of IL-6, among other cytokines. This is a cytokine with a local proinflammatory effect in the early stages of inflammation that also induces acute-phase production of proteins such as CRP [26]. Numerous

authors reported elevated IL-6 concentration in urine and plasma from patients with ATIN compared to healthy controls [20,27,28].

Other unspecific tubular AKI biomarkers have also been evaluated in ATIN. *N*-acetyl-β-D-glucosaminidase (NAG) is a lysosomal enzyme present in TECs which is a marker of proximal renal tubular damage [29]. Neutrophil gelatinase-associated lipocalin (NGAL) is a protein contained in the neutrophil granules but also in other human tissues such as kidney TECs. Some authors have demonstrated NGAL release during inflammatory processes [30]. α1-microglobulin is a low-molecular-weight protein that acts as radical scavenger and reductase [31]. NAG, α1-microglobulin, and NGAL are filtered in the glomeruli and completely reabsorbed by proximal TECs. Thus, urinary presence of these three molecules indicates proximal TEC damage and dysfunction. Many authors have found high levels of these biomarkers in urine and have used combination strategies to increase sensitivity and specificity of ATIN diagnosis [19,32–34]. Figure 1 illustrates the role of the suggested cytokines and chemokines in ATIN.

**Figure 1.** Cytokines and chemokines suggested as biomarkers in acute tubulointerstitial nephritis. Tubular endothelial cells (TECs) secrete MCP-1, which is a powerful chemoattractant for inflammatory cells towards the interstitium. Incoming macrophages release TNF-α, that primes and activates lymphocytes to secrete various proinflammatory interleukins (IL). In parallel, TNF-α can damage TECs inducing acute tubular necrosis (ATN). As a consequence of ATN, some markers of the injury of TECs (NAG, NGAL, and α1MG ) are found in urine.MCP-1, monocyte chemoattractant protein-1; TNFα, tumor necrosis factor α; NAG, N-acetyl glucosamidinadase; NGAL, neutrophil gelatinase-associated lipocalin; α1MG, α1 microglobulin.

Renal regional complement activation has been examined as an ATIN biomarker in a study by Zhao et al. Soluble urinary C5b9 was found to correlate with the degree of interstitial inflammatory infiltrates and also with tubular dysfunction in a cohort of 44 patients with ATIN of different etiologies [35].

In addition to the diagnostic value in the acute stage of the disease, urinary biomarkers have been evaluated in other settings. Shi et al. evaluated a panel of biomarkers in the follow-up of 54 patients with ATIN along a median time of 38 months. They found that NAG, matrix metalloproteinase 2 (MMP)2, and MMP9 clearly correlated with the rate of glomerular filtration rate (GFR) decline. The authors performed a multiple linear regression analysis to exclude other factors, and confirmed that higher concentration of these biomarkers at diagnostic was independently associated with faster progression of chronic kidney disease [34]. Table 1 summarizes the published evidence of ATIN serum and urinary cytokines and chemokines as biomarkers.


**Table 1.** Summary of the main publications related to serum and urinary biomarkers of acute tubulointerstitial nephritis (ATIN).

MCP-1 macrophage chemoattractant protein; NAG N-acetyl-neuraminidase; α1-MG α1-microglobulin; NGAL neutrophil gelatinase-associated lipocalin; IL-4 interleukin-4, IL-10 interleukin-10, transforming growth factor-β; GFR glomerular filtration rate; TNFα Tumor Necrosis Factor α; IL-8 Interleukin-8; IFNγ interferon γ; IL-6 interleukin-6; KIM-1 kidney injury molecule; IL-1β interleukin 1 β, IFNα2 interferon α2; IL-17A interleukin-17A, IL-18 interleukin-18; IL-23 interleukin-23; IL-12p70 interleukin-12p70; IL-9 interleukin-9; IL-5 interleukin-5.

## **5. Cellular Biomarkers**

In the context of drug-induced ATIN, the most common cause of this group of diseases, in vivo cellular assays are key to demonstrate the existence of drug hypersensitivity and identify the offending drug. Throughout their lives, but also during a previous ATIN episode, people are frequently exposed to multiple drugs. Proper identification of which exact medication is the culprit allows for safe discontinuation. Binding of T cells with a drug is a complex process that generates a cascade of events that can be measured in its different steps. These cellular assays are based on the demonstration of the lymphocyte proliferation response and the cytokine secretion response when exposed to the suspected drug.

On the one hand, the activation phenotype of lymphocytes after drug exposure can be assessed using fluorescence cytometry techniques. Detection of activation markers onthe surface of lymphocytes after incubation with the offending drug—such as CD25, CD69, or Human Leukocyte Antigen (HLA)-DR—may indicate hypersensitivity [36,37].

Also, activation of T cells after drug exposure triggers a specific pattern of cytokine secretion. These cytokines can be detected in the supernatant of the stimulated cell cultures using enzymelinked assays. Many authors have reported the release of high amounts of IL-5 in the setting of hypersensitivity. Mauri-Hellweg et al. reported high IL-5 concentration in supernatant of lymphocyte cultures from patients with known hypersensitivity to diverse antiepileptic drugs after culturing with these drugs37. Zanni et al. found a predominant Th2 response with prominent IL-5 release among a cohort of 13 patients affected by lidocaine hypersensitivity [38]. Sachs et al. found higher IL-5 concentration and, to a lesser extent IL-10 and IFN-γ, was even more sensitive for the detection of drug hypersensitivity than lymphocyte proliferation tests [39].

Enzyme-linked immunospot (ELISpot) assay is a sensitive test that allows the ex vivo measurement of the production of cytokines by lymphocytes in response to a certain stimulus. It provides quantitative information of the amount of responder cells. There are some reports of the usefulness of this test in evaluating drug hypersensitivity reactions. Tanvarasethee et al. studied 25 patients with cephalosporin-induced maculopapular exanthema. They found that the combined quantification of INF-γ and IL-5 spots by ELISpot assay was more sensitive than skin tests to diagnose cephalosporin hypersensitivity [40]. Rozieres et al. proved the specificity of the IFN-γELISpot assay for the demonstration of Type IV delayed hypersensitivity reaction (DHR) detecting amoxicillin-specific T cells. Twenty of twenty-two patients with known amoxicillin DHR had detectable amoxicillin-specific T cells. Interestingly, none of the control patients with IgE-mediated amoxicillin hypersensitivity nor the healthy controls presented amoxicillin-specific T cells [41]. At the moment, there is little evidence on the usefulness of ELISpot assays specifically targeting patients with ATIN. Punrin et al., in a cohort of patients with drug-induced ATIN, reported a positive IFN-γELISpot assay in 50% of them [42].

The lymphocyte transformation test (LTT) measures the proliferation of lymphocytes in response to the pure form of a suspicious drug. In vitro, prior to the exposure, lymphocytes are incubated with 3H-thymidine or carboxyfluoresceinsuccinimidyl ester (CFSE). The attenuation of radioactivity incorporation or dilution of CFSE with each cell division after incubation with the offending drug, measured in a β-counter or flow cytometry, is an indicator of cell proliferation. Positive LTT has been reported in the setting of ATIN induced by β-lactams and NSAIDs [43]. Koda et al. described a case of ATIN in the setting of the treatment with nivolumab and lansoprazole [44]. The LTT demonstrated reactivity against lansoprazole and not against nivolumab, indicating which was the culprit drug for ATIN.
