Kidney Injury Molecule-1: A Translational Journey

Abstract

Kidney injury molecule-1 (KIM-1, also named TIM-1 and HAVCR-1) was identified as the most highly upregulated protein in the proximal tubule of the kidney after injury. This protein is present with injury in multiple species including man, and also after a large number of acute and chronic insults to the kidney. It is a type-1 membrane protein whose ectodomain is released into the lumen of the tubule. The ectodomain is heavily glycosylated and stable and appears in the urine after injury. It has been qualified by the United States Food and Drug Administration and the European Medicines Agency for preclinical assessment of nephrotoxicity and on a case-by-case basis for clinical evaluation. As a biomarker in humans, its utility has been demonstrated in acute and chronic injury and in renal cell carcinoma, a condition similar to injury, where there is dedifferentiation of the epithelial cell. KIM-1 is a phosphatidylserine receptor which recognizes apoptotic cells directing them to lysosomes. It also serves as a receptor for oxidized lipoproteins and hence is important for uptake of components of the tubular lumen which may be immunomodulatory and/or toxic to the cell. KIM-1 is unique in being the first molecule, not also present on myeloid cells, that transforms kidney proximal epithelial cells into semi-professional phagocytes. Data suggest that KIM-1 expression is protective during early injury, whereas in chronic disease states, prolonged KIM-1 expression may be maladaptive and may represent a target for therapy of chronic kidney disease.

INTRODUCTION

We identified kidney injury molecule-1 (KIM-1, also named TIM-1 and HAVCR-1) as an important molecule for kidney biology by using techniques of representational difference analysis, a polymerase chain reaction − based technique, in a study designed to identify genes whose expression increases 24 to 48 hours after ischemia in the rat (1). The goal was to identify potential new therapeutic targets for treatment of acute kidney injury (AKI). KIM-1 mRNA was increased at 48 hours post-ischemia in the rat, more than any other mRNA in this analysis. This marked elevation in expression was subsequently confirmed by other groups (2). We raised antibodies to the gene product and established that protein expression was markedly upregulated specifically in the proximal tubule. The protein is localized primarily to the apical membrane of the cell, although it is also present in the basolateral membrane when there is significant loss of cell-cell structures that separate apical from basolateral compartments of the membrane, as seen in advanced injury. Shortly after its initial discovery, we found that KIM-1 was produced by the human kidney and that the ectodomain was present in human (3), rat and mouse urine (4,5) when there was injury to the kidney.

The gene encoding KIM-1 is conserved across species including zebrafish, rodents, dogs, primates, and humans. KIM-1 is a type I cell membrane glycoprotein which contains an extracellular immunoglobulin- and mucin-like domain, with N- and O-glycosylation sites. It has a transmembrane domain and short intracellular domain with intracellular tyrosine phosphorylation sites. The heavily glycosylated ectodomain is cleaved by metalloproteases.

KIM-1 AS A BIOMARKER OF KIDNEY INJURY

Much of the early focus of research related to KIM-1 pertained to its use as a urinary biomarker of kidney injury. There were a number of initial publications showing studies on rats and mice attesting to its sensitivity and specificity for kidney injury (4–6). A major advance in the demonstration of the utility of KIM-1 in preclinical nephrotoxicity was derived from studies performed by the Predictive Safety Testing Consortium (PSTC). This consortium was made up primarily of pharmaceutical companies and the United States Food and Drug Administration (FDA), all focused on a pre-competitive identification of biomarkers that could identify nephrotoxicity in a timely, sensitive, and specific way. The goal was in line with the focus of the Critical Path Initiative of the FDA to identify new biomarkers to predict drug toxicity in pre-clinical studies which can act as surrogate endpoints and/or aid in making efficacious and cost-saving decisions or terminating drug development more quickly (7). The PSTC was also very interested in advancing the use of the best performing biomarkers to clinical utility. Extensive rat toxicology studies were performed to compare the diagnostic performance of a number of urinary biomarkers, including KIM-1, to blood urea nitrogen (BUN) and serum creatinine (SCr) as predictors of kidney tubular damage as adjudicated by histopathology. Eight nephrotoxicants were used in these studies; dose-response and time course studies were performed. Ischemia/ reperfusion studies were incorporated also. Specificity was determined by using two known hepatotoxicants without known nephrotoxicity. KIM-1 was found to outperform SCr, BUN, and urinary N-acetyl-β-D-glucosaminidase (NAG) in multiple rat toxin models of kidney injury. On the basis of these studies, the FDA and the European Medicines Agency (EMA) qualified KIM-1 in the context of non-clinical drug development for the detection of acute drug-induced kidney toxicity. Qualification is the result of evaluation of evidence that leads to a conclusion that a biomarker can be linked to a biological process or clinical endpoint in a fit-for-purpose context (8). In this PSTC study, KIM-1 performed better than any other biomarker tested with area under the curve (AUC) in receiver operator curve (ROC) analysis of between 0.91 to 0.99 (9) (Fig. 1).

Fig. 1.

Receiver operator curve (ROC) analysis from four nephrotoxicant studies showing the area under the receiver operator curve for KIM-1, BUN, N-acetyl-β-D-glucosaminidase (NAG), and serum creatinine (SCr) as a function of histopathology injury grade which was used as the gold standard. The higher the score on a 0−3 scale, the more severe the injury. Urinary KIM-1 and NAG levels were normalized to urinary creatinine. Reprinted with permission from Vaidya et al. (9).

The utility of tissue expression and urinary KIM-1 has also been shown in a number of clinical studies in community studies of early diabetes (10), patients exposed to potential environmental toxins (11), hospitalized patients with AKI (12), and in transplant recipients who develop graft dysfunction (13,14). KIM-1 has also been associated with the risk of incident congestive heart failure in the Framingham Heart Study participants (15). KIM-1 levels in subjects with chronic heart failure can predict worsening renal function and can be used to identify patients at risk for poor clinical outcomes (16). Levels of urinary KIM-1 can predict regression of proteinuria in patients with diabetes mellitus and can predict outcome in a critically ill population (17). Urinary KIM-1 levels correlate well with histological and functional injury in patients with a variety of kidney diseases. Urinary KIM-1 levels have been found to be associated with insulin resistance in an elderly community population even before the development of diabetes or recognized kidney disease (10). This has important implications for patient selection for drug studies as well as therapeutic intervention. One of the major problems with clinical studies evaluating new biomarkers is that the gold standard used, SCr, is insensitive and non-specific, making it difficult to truly evaluate a novel biomarker such as KIM-1 (18). This is a prominent problem that has held up the entire field of biomarker development in humans. If we were satisfied with BUN and creatinine, we would not need additional biomarkers. Yet most studies continue to measure the performance of the novel biomarker against these “legacy” markers. We have shown that even if a biomarker is 100% specific and sensitive, its performance would look inadequate when compared to SCr (18).

We have reported that urinary KIM-1 is also a very good biomarker for renal cell carcinoma (19,20). This is consistent with our initial reports showing that KIM-1 expression occurs with dedifferentiation of the proximal tubule epithelial cell, which is also a property of renal cell carcinoma cells that are derived from the proximal tubule.

THE FUNCTION OF KIM-1

KIM-1 is a phosphatidylserine receptor that recognizes apoptotic cells and directs them to lysosomes (21). It also serves as a receptor for oxidized lipoproteins. KIM-1 is unique in being the first molecule, not also present on myeloid cells, that transforms kidney proximal epithelial cells into semi-professional phagocytes. As a consequence of its role in enhancing the clearance of dead cells by the surviving tubular cells, KIM-1 may play an important role in the modulation of the innate immune response in AKI. Our data (unpublished) indicates that KIM-1 protects the kidney early after injury via processes facilitated by apoptotic cell uptake.

With respect to chronic kidney disease, however, the situation is different. When KIM-1 is expressed in renal epithelial cells of mice, they developed spontaneous and progressive interstitial kidney inflammation with fibrosis leading to renal failure with anemia, proteinuria, hyperphosphatemia, hypertension, cardiac hypertrophy, and death (22) analogous to progressive kidney disease in humans. KIM-1 is expressed in chronic kidney disease in humans (23,24). Thus, sustained KIM-1 expression promotes kidney fibrosis and provides a link between acute and recurrent injury with progressive chronic kidney disease. This highly suggests that KIM-1 may represent a therapeutic target for the treatment of chronic kidney disease and renal cell carcinoma.

CONCLUSIONS

In conclusion, KIM-1 is the most upregulated protein in the proximal tubule with any form of kidney injury that affects this nephron segment. Urinary KIM-1 is an excellent biomarker for kidney injury as shown in pre-clinical studies and increasingly in clinical studies. KIM-1 is a phosphatidylserine receptor which mediates uptake of apoptotic cells, and oxidized lipids into the proximal tubule. Acute KIM-1 expression is adaptive and it reduces the innate immune response to injury. Chronic KIM-1 expression is, by contrast, maladaptive leading to chronic kidney disease and KIM-1 may serve as an important therapeutic target for a disease process, chronic progressive kidney disease, for which we have little therapeutically to offer patients.

DISCUSSION

Williams, Boston: Thank you for a great talk. Do you know from experimental models whether, at the range in which you see elevation with normal creatinine with nephrotoxic drugs, you can use this to appropriately adjust dosage of a drug in a sense of trying to reduce renal toxicity?

Bonventre, Boston: As part of the Predictive Safety Testing Consortium, there were dose responses done for all of those toxins, and there is a clear dose response relationship in KIM-1 expression with increasing dose of the toxin. We do know, therefore, that if you reduce the dose of the toxin you will see a corresponding reduction in the level of urinary KIM-1.

Boyer, New Haven: Joe, thank you for that very nice presentation. As you know, it is often very difficult to distinguish hepatorenal syndrome from patients with intrinsic renal disease. I was wondering if you've had a chance to look at KIM-1 in this situation.

Bonventre, Boston: We have seen that KIM-1 is upregulated in many patients who have liver disease. The study hasn't been completed yet. But it is clearly up, and we are interpreting it as a reflection of the early signs of kidney disease, even before the creatinine goes up.

Boyer, New Haven: So then it would not be helpful in making a distinction… .

Bonventre, Boston: You mean hepatorenal versus something else.

Boyer, New Haven: Yes

Bonventre, Boston: It's a marker of any kind of kidney injury. Kim-1 levels in hepatorenal might be a little bit less than in ATN let's say. That's what we do see — but KIM-1 is not going to distinguish one form of kidney injury from another unless the other form of kidney injury doesn't affect the proximal tubule.

Boyer, New Haven: Normally, we usually think of hepatorenal as a completely normal… .

Bonventre, Boston: Right. So in that context, if there is no kidney injury — if creatinine is up and there is no KIM-1 in the urine — then that would be confirmatory that there is no tubular injury.

Molitoris, Indianapolis: Joe, you've made tremendous progress, and I am particularly interested in your chronic kidney disease ideas now. One thing struck me funny, and that is, how are you proposing — and maybe you know — KIM-1 enters the bloodstream since it's cleaved into the tubular lumen?

Bonventre, Boston: So, the issue is how it gets into the bloodstream. When there is injury to the tubule, the epithelial cell loses its polarity. So, once polarity is lost, then you get a release into the bloodstream as well as into the lumen.