N-Acetyl-β-D-Glucosaminidase

NAG is present in lysosomes of proximal tubular cells. Mean normal rates of NAG excretion rise slightly with age [204]. The molecular weight of NAG, 150 kD, precludes glomerular filtration and, therefore, ensures that excreted NAG has emanated from tubules. The enzyme exists in two major isoforms, A and B. NAG-A is released by exocytosis into the filtrate at a stable rate that is unrelated to the cellular Cd content. NAG-B remains in lysosomes and enters the filtrate after cellular injury [204,205]. Its excretion is increased by Cd toxicity [205–207]. Reported measurements of NAG excretion have contributed significantly to our current synthesis of Cd nephropathy (see Section 4.3.2, Section 4.3.3, and Section 4.3.6).

#### Kidney Injury Molecule 1

KIM1 is a transmembrane glycoprotein that participates in the restoration of adhesion between regenerated proximal tubular cells. During this process, the ectodomain of KIM1 is released into the filtrate and excreted in urine. The protein is not detectable in the absence of injury [208].

KIM1 has been studied extensively in animals as a biomarker of Cd toxicity [209–211]. In rats treated with exogenous Cd, genetic expression and urinary excretion of KIM1 (EKIM1) increased at 6 weeks; ECd and ENAG rose 3 and 6 weeks later [209,210]. Apoptosis was sparsely evident at 6 weeks and more prevalent at 12 weeks; increased KIM1 excretion accompanied apoptosis but was also observed in the absence of apoptosis [211]. If rodent studies can be extrapolated to humans, EKIM1 identifies toxicity that has not ye<sup>t</sup> led to cell death, and it is detectable before ECd rises. It appears to be the earliest appearing indicator of Cd tubulopathy.

At least four studies have examined the utility of EKIM1 in humans exposed to Cd. Pennemans and colleagues investigated a Belgian sample with chronic low-dose environmental exposure [212]. The geometric mean of Cd excretion in this group was 0.76 μg/g creatinine, which is approximately twice that reported in healthy populations and 15% of the conventional Cd threshold for tubular injury (5.24 μg/g creatinine) [41]. Pennemans and colleagues found that EKIM1 correlated with ECd even though the excretion of reabsorptive markers did not. A Thai group confirmed this correlation in a sample with higher ECd [213], but a Chinese report did not [214]. A second study from China showed that EKIM1 rose in a stepwise fashion with low-, middle-, and high-dose environmental exposure to Cd [58].

#### 4.3.2. Excretion of Cadmium

Excreted Cd is either filtered and not reabsorbed or released from tubular cells into the filtrate [215]. Several lines of evidence support the second alternative. After inducing Cd nephropathy in rabbits, Nomiyama and Foulkes infused labelled CdMT at increasing rates to create a series of steady-state plasma concentrations of the complex. Although Cd poisoning had reduced the tubular maximum for CdMT, the excretion rates of total Cd greatly exceeded those accounted for by failure to reabsorb the label [216]. Most of the excreted Cd, therefore, emanated from the tubular cells.

In kidneys from rats intoxicated with Cd, Tanimoto and colleagues showed that ECd and numbers of apoptotic, sloughed cells in tubules rose in tandem [217]. In workers with occupational exposure to Cd, ECd correlated with the renal cortical content of the metal, as measured by neutron-capture gamma-ray analysis [218]. In human accident victims, the Cd content of renal tissue at autopsy correlated with the urine Cd concentration ([136]; Figure 2; see Section 4.1.1). In transplanted kidneys, the tissue Cd content correlated with the preoperative overnight ECd of living donors [140]. Three groups of investigators found that ECd varied directly with GFR, as though the number of intact nephrons had determined the rate of appearance in urine [219–221].

Reported correlations of ECd with ENAG or EKIM1 provide additional evidence that urinary Cd emanates from tubular cells [205,212,214,222–226]. Importantly, these studies did not sugges<sup>t</sup> that markers of cell injury began to rise at a threshold level of ECd; instead, the correlations extended in a linear fashion from normal to increased levels of ECd [205,222,225,226]. EKIM1, the most sensitive indicator of Cd-induced toxicity, rose before ECd in animal studies [209,210].

Taken together, the foregoing observations sugges<sup>t</sup> that Cd excretion does not result from a failure to reabsorb filtered Cd. It is more likely that Cd, NAG, and KIM1 emanate from the same source for the same reason. We infer that ECd is itself a marker of tubular cell injury.

#### 4.3.3. Cadmium Toxicity and the Glomerular Filtration Rate (GFR)

A paradox emerges from the literature relating ECd to GFR. Three groups have reported that ECd rose with GFR in exposed populations [219–221], and three have stated that Ccr or eGFR declined steadily as ECd increased from modest levels [51,52,227–229]. To reconcile these observations, we speculate that in the progression of Cd nephropathy, cellular injury is evident before nephrons are lost; during that phase, ECd varies directly with the nephron number, which, in turn, varies directly with GFR. Cell death ensues as the burden of Cd rises; during this phase, Cd is released to filtrate at an increased rate even though nephrons are disappearing simultaneously. Estimated GFR falls as ECd continues to rise.

In Thai population samples, our group found inverse relationships between eGFR and ECd at all levels of environmental Cd exposure [52,54]. Moreover, investigators from multiple geographic regions documented a progressive decline in GFR despite mitigation or termination of occupational exposure [229–232]. This decline may have resulted from the continued transfer of CdMT from the liver to the kidneys [233], or it may have reflected continuous nephron destruction by a stable renal burden of the metal [234,235].

#### 4.3.4. Impaired Reabsorption of Small Filterable Proteins

Small proteins are readily filtered by normal glomeruli and reabsorbed and degraded by proximal tubular cells. As reabsorption of such proteins is virtually complete, excretion rates above a cuto ff value may be viewed as evidence of impaired reabsorptive capacity. The proteins most commonly studied for this purpose are β2-microglobulin (β2MG) and retinol-binding protein 4 (RBP4).
