1 Introduction

Cadmium (Cd) is a relatively rare metal, which is mainly found in the earth's crust at concentrations between 0.1–0.5 μg/g, mostly in combination with zinc. It was discovered in 1817 by Stromeyer as an impurity in zinc carbonate (latin = “cadmia”). In the environment, Cd exists preferentially in the oxidation state +2 and usually does not undergo oxidation-reduction reactions. Cd occurs naturally in the environment in its inorganic form as a result of volcanic emissions, forest fires, and weathering of rocks. In addition, anthropogenic sources have increased the background levels of Cd in the atmosphere, soil, water, and living organisms (see also Chapters 2 and 3 of this book). But Cd as a chemical element cannot be degraded, therefore its concentration in the environment increases steadily and globally. Since the first systematic studies on chronic Cd poisoning were performed in the 50s (reviewed in [1]), strict safety regulations and controls have been implemented by companies and state authorities, which efficiently prevent recurrence of health hazards in exposed populations. After the health effects of Cd were first described in 1858, where respiratory and gastrointestinal symptoms were observed in persons using Cd-containing polishing agent [1], more than 10,000 studies and reviews have been published on the topic and knowledge on Cd associated health hazards and preventive measures are well established.

Why is Cd2+ the species of the metal that is noxious to organisms, toxic? Cellular concentrations of essential metal ions are tightly regulated and both deficiency and overload have adverse effects. Non-essential metal ions, such as Cd2+, compete with essential metal ions for entry into cells. Their interference with the function of essential metal ions disrupts cellular functions and leads to disease. Moreover, iron and zinc deficiencies, which are both most common in humans, enhance Cd2+ uptake.

Why should we care about Cd as a toxic compound? Not because of the rare cases of acute Cd intoxication, nor because of its health threat in an occupational context (Cd is ranked eighth in the Top 20 Hazardous Substances Priority List! [2]). Recently a shift of paradigm has occurred in our understanding of the health problems due to Cd exposure. The real challenge in the 21st century in a global setting seems to be chronic (i.e., over decades or even throughout life) low (i.e., in concentrations barely exceeding the “natural” environmental Cd concentrations) Cd exposure (CLCE), which already affects or will affect large proportions of the world populations. Nowadays CLCE results from dietary sources and cigarette smoking. Modern agriculture globally uses Cd-containing phosphate fertilizers to increase its efficiency of harvests. Plants, including tobacco, accumulate Cd, which is passed on to animals and man in the food chain. Smoking is now established as a major cause of chronic health issues, i.e., cardiovascular diseases and cancer. Still, large parts of the population do not abstain from smoking and though other components of tobacco smoke have been identified as being responsible for the development of smoking-associated diseases, evidence accumulates that Cd in tobacco smoke is one of the factors in its own right that causes smoking-associated chronic diseases.

CLCE causes chronic diseases and increases overall mortality because Cd has a very long biological half-life that may exceed 20–30 years and accumulates in organs, which can result in fibrosis, failure or, as a Class 1 human carcinogen, cancer. Hence, Cd exposure in the childhood may affect health in old age. Only recently has CLCE gained public attention with possible impact on EU legislation. Maximal exposure values for the general population to limit chronic Cd toxicity need to be re-defined as a consequence of the globally much debated CONTAM report from the European Food Safety Authority [3,4], which concluded that current Cd threshold values in food can cause kidney damage and should be reduced.

Ubiquity of Cd make it a serious environmental health problem that needs to be assessed because, besides being a well-known occupational hazard, CLCE is a health problem for ~10% of the general population that increases organ toxicity, especially nephrotoxicity, without a known threshold, implying that there is currently no safe limit for CLCE.

Hence, we found it timely and necessary to summarize and integrate past and current knowledge by reviewing pioneering studies on the mechanisms of Cd toxicity in organisms, tissues, cells, and molecules in addition to recent state-of-the art publications that take novel developments on causes and consequences of CLCE into consideration.

2 Sources and Exposures

2.1 Occupational Sources and Exposures

Areas in the vicinity of zinc mines and smelters often show pronounced Cd contamination [2,5]. Until the 1920s the industrial use of Cd was limited, but the world Cd production has increased steadily to ~20,000 tons/year in the 1980s, which has remained constant at this level until present. In 2009, China, The Republic of Korea, and Japan were responsible for more than 50% of the world production of Cd (World Bureau of Metal Statistics, Hertfordshire, United Kingdom; http://www.world-bureau.com/). About 80% of Cd produced is still used in batteries, especially nickel-Cd batteries, which have been banned by the European Union since 2004. About 10% of the yearly Cd production is used for color pigments in paints and plastics, and because of its non corrosive properties 5% is consumed for electroplating or galvanizing alloys, particularly in the aircraft industry. Further applications of Cd are solders, its use as a barrier to control nuclear fission, or as a plastic stabilizer (http://www.cadmium.org/; The International Cadmium Association (ICdA)).

Cd waste enters the air, water, and soil through industrial and household wastewater and waste incineration, burning of fossil fuels, and biomass. The primary form of Cd in occupational exposures is Cd oxide (CdO), which is predominantly inhaled from industrial dust and fumes. Improved technologies and regulations mean little Cd now enters air and water in Western Europe or Northern America. But it is still a major source of contamination in developing and newly industrializing countries.

Up to the l960s, very elevated Cd exposure levels in air were measured in some workplaces (exceeding 1 mg/m3). To decrease this level, the Agency for Toxic Substance and Disease Registry (ATSDR) has set minimal risk levels (MRL) for acute (<14 days) or chronic (>1 year) Cd inhalation exposure of 0.03 respective 0.01 μg/m3 [2] (http://www.atsdr.cdc.gov/toxprofiles/tp5.pdf), which is well below the value set by the Occupational Safety & Health Administration from the US Department of Labor (OSHA) (legal limit of 2.5–5 μg/m3 Cd averaged over an 8-hour work day). Ambient air Cd concentrations have been estimated to <10 ng/m3 (0.1–0.5 ng/m3 in rural and 2–15 ng/m3 in urban areas) but concentrations several fold higher, up to 1.2 μg/m3, can be found near lead or zinc smelters [5,6]. Road traffic and the use of wood-burning fires or stoves increase indoor Cd pollution [7].

2.2 Non-Occupational Sources and Exposures

Besides smoking, the major route of Cd intake for non-occupationally exposed humans is through food or water. Cd in drinking water contributes only to less than a few percent of the total Cd intake [8] with the exception of heavily contaminated areas. Food is the major source of Cd for the general population, which is particularly caused by the use of phosphate rock (as opposed to processed phosphoric acid and phosphate fertilizers) for agricultural purposes because most natural and agricultural soils are phosphate-deficient and plants require phosphate for growth.

Compared to those of igneous origin, phosphate rock deposits of sedimentary origin contain higher levels of Cd and are the raw material used in the manufacture of most commercial phosphate fertilizers. In 2009, the world’s largest producers of rock phosphate were China (~40%), Morocco (~17%), and the USA (~16%). Its Cd content varies between <20 to >200 mg Cd/kg rock phosphate (http://ec.europa.eu/environment/enveco/taxation/pdf/cadium.pdf). Whereas China and USA consume their own production, Morocco exports most of its phosphate production, e.g., to other African countries, the USA, and India. During the last 10 years the phosphate rock consumption of Western and Central Europe and Northern America has steadily decreased by 30–70% (http://www.fertilizer.org/). In these areas, decreased Cd contamination is due to replacement of sedimentary phosphate rock by igneous rock phosphate (from South Africa or Russia), by applying lower amounts of fertilizers, or by “decadmiation” technologies (http://ec.europa.eu/environment/enveco/taxation/pdf/cadium.pdf).

Increased Cd uptake by plants depends on plant species, pH, and other soil characteristics [9]; (http://www.environment-agency.gov.uk/static/documents/Research/SCHO0709BQRM-e-e.pdf). Soluble and insoluble complexes with organic matter are also important. Generally, Cd binds strongly to organic matter, which immobilizes it. But soil pH is the most important factor controlling Cd uptake by plants, with its mobility decreasing with increasing alkalinity [9–11]; (http://www.environment-agency.gov.uk/static/documents/Research/SCHO0709BQRM-e-e.pdf). Liming soil to increase its pH has been shown to reduce the mobility and plant availability of Cd in different soils [2,9]. Processes that acidify soil (e.g., acid rain) on the other hand may increase Cd concentrations in food.

Plant food generally contains higher concentrations of Cd than meat, egg, milk, dairy products, and fish [12]. Cereals, such as rice and wheat grains, green leafy vegetables and root vegetables, such as carrot and celeriac, accumulate Cd [9]. High levels of Cd are also found in oil seeds, cocoa beans, peanuts, and in wild mushrooms as well as in shellfish and offal products (liver and kidney), especially from older animals, such as pigs and sheep. Animal liver and kidney can have levels higher than 50 μg/kg; cereal grains concentrate Cd up to 10–150 μg/kg and shellfish accumulate 1–2 mg/kg.

The average Cd intake from food generally varies between 8 and 25 μg per day, depending on the Cd concentration of the food [8,13–16]. Up to 10% of food Cd is absorbed by the body [17]. Based on estimation of Cd intake, more than 80% of the food-Cd comes from cereals and vegetables [8]. Population with specific dietary habits, for example, vegetarians and high consumers of shellfish, can have a higher intake of Cd. Moreover, specific populations may be more susceptible to Cd due to genetic variation. Because of ionic competition, deficiencies of other metal ions, such as Zn, Ca or Fe may increase the accumulation of Cd. Populations in developing countries, where such deficiencies are prevalent, are at risk, and more particularly women of child-bearing age who are Fe-depleted and anemic.

The CONTAM report from the European Food Safety Authority recently concluded that across Europe the average dietary intake of Cd in adults is between 1.9 and 3.0 μg/kg b.w. per week, and can even reach 2.5–3.9 μg/kg b.w. per week in highly exposed adults [3] (http://www.efsa.europa.eu/en/efsajournal/doc/980.pdf). This intake is close to or slightly exceeding the threshold value of 2.5 μg/kg b.w. per week at which signs of tubular damage are noticeable in 95% of the population by age 50. This report has been criticized by the Joint Food and Agriculture Organisation/World Health Organization (FAO/WHO), which came up with a threshold value of 5.8 μg/kg b.w. per week (http://www.who.int/foodsafety/chem/jecfa/summaries/summary73.pdf). Both assessments used the same epidemiological dataset and had two primary components: a concentration-effect model that relates the concentration of Cd in urine to that of β2-microglobulin, a biomarker of renal tubular effects, and a toxicokinetic model that relates urinary Cd concentration to dietary Cd intake. The CONTAM Panel reevaluated the two assessments and maintained its previous conclusion [4]; http://www.efsa.europa.eu/de/efsajournal/doc/1975.pdf. Hence it can be concluded that the average dietary intake of Cd in adults in Europe (“CLCE”) may affect human health in the long run.

Cigarette smoking is a major non-occupational source of Cd exposure, which is inhaled mainly as CdO. Accumulation of Cd in tobacco plants can vary widely: average concentrations are 1–2 μg/g of dry weight or 0.5–1 μg per cigarette [18]. Roughly 10% of the CdO produced during cigarette smoking is inhaled with an approximate 50% absorption in the lung (see Section 3.1.1). The concentration of Cd in smokers is 4–5 times higher in blood, and 2–3 times higher in the kidneys, when compared with non-smokers [17]. Smoking is thought to roughly double the life time body burden of Cd [19]. There is also some evidence of Cd exposure from “second-hand” or “passive” smoking in children [20] and adults [21].

3 Entry Pathways, Transport, and Trafficking

3.1 Entry Pathways

Cd enters the body mainly through the lungs and the gastrointestinal (GI) tract. The absorption of Cd from the lungs is much more effective than that from the gut. However, Cd absorption from the GI tract is the main route of Cd exposure in humans.

3.1.1 Lungs

After inhalation exposure, the absorption of Cd compounds varies greatly depending on the physico-chemical properties of the Cd compounds involved, site of deposition in the lungs and particle size [22]. In the lungs, deposition, mucociliary clearance, and alveolar clearance determine the absorption of inhaled particles. Large particles, dusts (>10 μm in diameter) tend to be deposited in the upper airways, while small particles, fumes, cigarette smoke (approximately 0.1 μm in diameter) penetrate into the alveoli, which are the major site of absorption. Between 50–100% of Cd in the alveoli are transferred to the blood. In the average human population, the amount of Cd absorbed by the lungs is not greater than ~0.2 μg per day based on the assumption that up to 50% of the retained Cd is absorbed [23]. On the basis of data on organ burdens of Cd and smoking history, Elinder et al. [24] calculated that about 50% of the Cd inhaled via cigarette smoke are absorbed by the lungs.

Inhaled Cd partly dissolves in the respiratory tract lining fluid and may be found as Cd particle, as a free ion or complexed to secreted proteins or glutathione (L-γ-glutamyl-L-cysteinyl-glycine) [25], an important cellular antioxidative metabolite involved in defense against Cd-induced oxidative stress and a Cd chelator. In principle Cd may be absorbed transcellularly (via ion channels, transporters, and endocytosis) or paracellularly. Experimental studies on rat lungs have demonstrated that more Cd is accumulated by alveolar macrophages than by type II alveolar epithelial cells [26] and has been confirmed in human studies [27]. It is likely that Cd complexes are endocytosed by alveolar macrophages, resulting in either detoxification of the metal [27], toxicity [28], or inflammatory processes [29] rather than in systemic absorption of Cd. Net transport of Cd into the body is more likely to occur via alveolar epithelia. Both, para- and transcellular free Cd transport has been demonstrated through epithelial monolayers of rat and human alveolar cell lines [30]. Paracellular Cd fluxes may, however, be the consequence of direct toxic effects of Cd, which disrupts adherens junctions [31].

Recently, several solute carriers have been cloned, which transport divalent metals (including Fe and Cd) with high affinity (K m 0.1–1 μM), such as SLC39A8 or Zrt-, Irt-like protein 8 (ZIP8) [32] and SLC11A2 or divalent metal-ion transporter-1 (DMT1, also abbreviated to DCT1 for divalent cation transporter-1; or NRAMP2 for natural resistance-associated macrophage protein-2) [33,34] (see below). ZIP8 is strongly [35] and DMT1 weakly expressed [34] in rodent lung tissues. Both transporters are also expressed in human lung cell lines and may contribute to apical Cd uptake into lung cells [36,37]. In vivo studies in homozygous Belgrade rats, which are functionally deficient in DMT1 have shown that DMT1 is involved in toxic metal clearance from lungs [38]. However, overall evidence for DMT1-mediated Cd uptake by lung cells is weak. Basolateral efflux pathways for free Cd in lung epithelia are unknown.

3.1.2 Gastrointestinal Tract

The other major route of entry of Cd into the body is via the GI tract after oral exposure. Factors affecting the absorption of ingested Cd include mammalian species, type of Cd compound, dose, frequency of administration, etc.. The absorption of Cd after a single exposure ranges from 0.5–8% [39]. After chronic exposure (12 months) to Cd in drinking water rats retain less than 1% of the total amount ingested [40]. Observations in humans given radioactive Cd indicate that the average GI absorption during chronic background exposure is about 2–8% [5,41]. For a given individual, the absorption following oral exposure to Cd depends on the individual (age, body stores of Fe, Ca, and Zn; pregnancy history; lactation; etc.) and also on the levels of ions (Zn, Ca) and other dietary components ingested with Cd. Animal experiments have shown that diets with low levels of Ca and protein may increase the intestinal absorption of Cd up to 3 times [39,42] and neonatal animals absorb Cd to a much greater extent than adult animals [43]. However, data from human studies showing a relationship between GI absorption of Cd and age are missing.

In people with low body stores of Fe the absorption of Cd is higher than in subjects with normal Fe stores [44], which has been also observed in experimental animals [45]. Interestingly, dietary Cd absorption tends to be higher in females than in males [46] because of increased incidence of low Fe stores or overt Fe deficiency in women at fertile age [47,48]. Women with low body Fe stores, as reflected by low serum ferritin levels, have on average twice (about 10% but up to 20%) the normal rate of oral Cd absorption [49]. This may be explained by the close correlation between Cd absorption and the expression of DMT1, whose expression is induced by Fe deficiency and transports Fe and Cd into the mucosa cell equally well [34,50]. This situation is exacerbated during pregnancy when enterocytes have an increased DMT-1 density at the apical surface to optimize micronutrients absorption [46,48].

Other candidates for free Cd uptake in the duodenum are ZIP14, which is encoded by the SLC39A14 gene and transports Cd with high affinity [35], and the Ca2+-selective channel TRPV6, which belongs to the vanilloid family of the transient receptor potential channel (TRP) superfamily (also known as CaT1) that transports Cd2+ at concentrations ≥10 μM [51]. Cd efflux from intestinal cells by the Fe export transporter ferroportin 1 (FPN1, IREG1, MTP1), a Fe-regulated transporter implicated in the basolateral transfer of Fe from the duodenum to the circulation [52,53], has not been demonstrated so far although increased Cd uptake in Fe-deficient mice was correlated with increased expression of DMT1 and FPN1 in duodenum [54]. Interestingly, Cd absorption appears to occur independently of the enterocyte expression level of metallothioneins (MTs), small metal-binding proteins which form high-affinity complexes with divalent metals, such as Zn, Cu or Cd, and store and detoxify these metals in organs and cells (see Section 6) (reviewed in [55]).

Other nutritional forms of Cd apart from Cd2+ are Cd complexes with cysteine-rich peptides and proteins, such as Cd-metallothioneins (CdMT) (see also Chapter 11), Cd-phytochelatins (CdPC), and Cd-glutathione (CdGSH). In plants, Cd accumulates as a complex with MT or PC. PCs are glutathione (GSH)-derived peptides. A significant part of CdMT and CdPC from animal and plant food reaches the intestinal lumen in intact form [56,57] and is mainly broken down in the colon by GI bacterial fermentation [58]. Experimental evidence in rodents suggests that CdMT and CdPC may be absorbed intact by enterocytes, possibly via transcytosis [56,59]. Recently, a novel high-affinity multiligand receptor, the lipocalin-2 (24p3/Neutrophil Gelatinase-associated Lipocalin (NGAL)) receptor, has been identified, which mediates endocytosis of protein-metal complexes [60]. Current studies in the laboratory indicate that the lipocalin-2 receptor is expressed in rodent colon as well as in colon-like Caco-2 cells where it mediates transcytosis of CdMT and CdPC (F. Thévenod and C. Langelueddecke, unpublished).

3.1.3 Skin

A dermal route of entry through contamination of the skin has been described in vitro but is extremely low [61]. Percutaneous absorption of Cd chloride from water and soil into and through human skin was performed using samples of cadaver skin and did not exceed 0.6%. This route of entry may therefore be of concern only in situations where concentrated solutions would be in contact with the skin for several hours or longer.

3.1.4 Placenta

The Cd concentration of the human placenta is usually about 5–20 μg/kg wet weight [62]. The placentas of women who smoke during pregnancy have higher Cd levels than those of non-smokers [63]. But placental transfer of Cd is limited. Cd concentration in newborn blood (umbilical cord) is on average 40–50% lower than in maternal blood [64]. Transplacental transport of Cd is minimized in the normal healthy placenta presumably by the binding of Cd to MT. Placental Cd accumulation in humans [65] and experimental animals [66] may be mediated by Cd transport via placental DMT1 [67] and TRPV6 [68], which could indirectly affect the fetus.

3.2 Transport and Trafficking

Human data on the transport of Cd in the circulation from the site of absorption to the various organs are scarce. Following absorption in the lungs and/or intestine, Cd in the blood initially primarily binds to albumin and other thiol-containing high- (HMW) and low-molecular-weight (LMW) proteins in the plasma, including MT, as well as to blood cells [23]. Studies in experimental animals show that immediately after parenteral administration, most of the Cd is present in the plasma [69]. Plasma concentrations decrease rapidly during the first hours after injection of 1 mg/kg body weight, reaching a level that is less than 1% of the initial value at 24 h, and this level then decreases much more slowly. The proportion of plasma Cd bound to MT and larger proteins, respectively, varies with the length and type of exposure. During the early, fast-elimination phase, Cd in mouse plasma is mainly bound to plasma proteins with a molecular weight of 40–60 kDa (mainly albumin) [70] with low affinity (apparent stability constant ~10–5 M–1; [71]) whereas in the slower phase (more than 24 h after injection), it is partly bound to LMW proteins, mainly MT [72] and glutathione [73] with high affinity (apparent stability constants for MT and GSH ~10–25–10–14 M–1 and ~10–9 M–1, respectively [74,75]. Similar observations were made in chronically (up to 14 weeks) Cd-treated rats [76,77]. In healthy humans, the concentrations of plasma MT (to which Cd is mostly bound during CLCE) significantly differ in non-smokers and smokers (3.42 ± 2.30 ng/mL versus 4.40 ± 2.76 ng/mL) [78], whereas in occupationally exposed Cd workers plasma MT levels vary between 2–11 ng/g [79].

It is important to note that Cd also tends to concentrate in blood cells (mainly erythrocytes, but also leukocytes); only a small percentage (<10%) remains in the plasma [80]. For this reason, the monitoring of blood samples for levels of Cd exposure typically involves the analysis of whole blood. The total Cd in the leukocyte portion of the blood is negligible compared to that in the red blood cells [81]. In rodents acutely or chronically exposed to Cd, Cd in erythrocytes is partly bound to hemoglobin [82], but mostly to MT (reviewed in [5]). Red blood cells [83], lymphocytes [84], and platelets [85] contain MT, which is inducible by Cd, and contributes to the trafficking of Cd as CdMT via the systemic circulation. MT-I/II mRNA in blood and peripheral lymphocytes has recently been used as a sensitive biomarker for Cd exposure in humans [86,87]. Since MT-bound Cd is quickly cleared from the plasma by the kidneys [88,89], this protein fraction is of great importance for the transport of Cd from liver to kidney during long-term exposure [77,90–92] (see below).

The blood level of Cd largely reflects recent Cd exposure with a half life of 75–128 days (fast component); however, it also has a slow component with a half life of 7–16 years, which correlates well with total body Cd load [93] (however, as a caveat, see [94]). Non-smoking adults living in non-polluted areas have blood Cd concentrations that vary between 0.1 and 1.0 μg Cd/L in whole blood (1–10 nM) [95], and values just under 5 μg/L in smokers; values above this concentration are an indication for medical surveillance in industrial workers who do not yet have signs of renal damage.

3.3 Excretion

Once absorbed, Cd is very poorly excreted, mainly in urine and feces. In humans the amount excreted daily in urine represents only about 0.005–0.015% of the total body burden [39,96]. The mean concentration of urinary Cd in people not exposed to high Cd levels is ~0.5–2.0 μg/L [23] and increases with age [97–99] and body burden [2,100,101]. Smokers have higher urinary excretion than non-smokers [97,99]. Increased urinary Cd excretion occurs when tubular proteinuria develops [100,102]. Most of the Cd in urine is transported bound to MT. Tohyama et al. [103] found good correlation between urinary MT and Cd in the general population as well as in smokers [104]. Roels et al. [105] confirmed this correlation in Cd workers. Over a range of doses, an increase in urinary excretion of Cd is associated with an increase of Cd in the renal cortex [106–108]. Chronic studies on several mammalian species have shown that urinary excretion of Cd increases slowly for a considerable time but, as kidney dysfunction develops, a sharp increase in excretion occurs [5,106,107,109]. This leads to a decrease in renal and liver Cd concentrations [5,106].

Fecal Cd concentrations range between 0.1–0.4 μg/g wet weight [96], whilst Japanese populations display 2.5–3.5-fold higher Cd concentrations than other populations. In contrast to urinary Cd excretion there is no age dependence of fecal excretion. The average daily fecal Cd varies between 18 μg in Sweden and 40 μg in Japan. A large proportion of the GI Cd excretion is directly related to the daily intake. Hence the average daily fecal Cd is slightly higher among smokers than among non-smokers (by 1.8–3.8 μg Cd/day) [96].

It is difficult to study experimentally GI excretion after oral exposure, since it is not possible to distinguish net GI excretion from unabsorbed Cd in feces. Animal studies of GI excretion following injections of Cd (summarized in [39]) show that a few percent of the dose is excreted in the feces within the first few days after injection. After chronic exposure of rats, fecal excretion amounts to about 0.03% of the body burden, which is considerably more than the urinary excretion [107,110]. The mechanism of fecal excretion may involve a transfer of Cd via the intestinal mucosa, but biliary excretion may also be involved. The biliary excretion in the first 24 h after intravenous injection of Cd is dependent on the dose [111,112]. Biliary Cd has been partially characterized as a GSH complex [113]. Indeed, mutant rats (EHB and GY rats) that lack mrp2 activity, which exports GSH [114], exhibit impaired ability to transport Cd into bile [115–117]. Multidrug resistance-associated protein 2 (MRP2) also called canalicular multispecific organic anion transporter 1 (CMOAT) or ATP-binding cassette sub-family C member 2 (ABCC2) is a protein that in humans is encoded by the ABCC2 gene and represents a conjugate export pump expressed in the apical membrane of hepatocytes [118].

Cd is also eliminated through hair [119,120] and breast milk [121,122], but these routes are of limited importance for total excretion and do not significantly alter the biological half-time.

4 Health Effects

4.1 Acute Toxicity

Acute high-dose Cd toxicity in humans is now a rarity in Western countries. Its symptoms depend on the route of ingestion [123]. Toxicity also depends on the solubility of Cd compounds [2].

4.1.1 Inhalation

Acute intoxication mostly occurs by inhalation of fumes in an industrial setting [5]. There is usually a latent period of 4–12 hours between the exposure and the onset of symptoms [5]. In severe intoxication, patients develop a respiratory distress syndrome due to acute pneumonitis and pulmonary edema with respiratory failure, which can progress to death in 3–7 days [5,124,125]. Symptoms develop when CdO fumes reach concentrations of 200–500 μg Cd/m3 [23]. Concentrations above 1 mg/m3 in air for 8 hours lead to acute chemical pneumonitis [5] and death occurs at about 5 mg Cd/m3 for an 8-hour exposure [39]. Histology has shown congestion with intra-alveolar hemorrhage, metaplasia of the alveoli lining cells and fibrinous intra-alveolar exudates [5]. Inhalation exposure to Cd at concentrations of 5–20 mg/m3 for 50–120 min gives rise to pulmonary edema in rats and rabbits (reviewed in [5,126]).

4.1.2 Ingestion

Cd ingestion, on the other hand is mostly accidental or even intentional [127], but can also occur from heavily contaminated dust exposure, food or beverages. Liver is the major target organ of toxicity following acute Cd ingestion, and Cd hepatotoxicity is the major cause of acute Cd lethality. Symptoms begin almost immediately and include salivation, nausea, vomiting, diarrhea, and abdominal pain [5,23,128]. In cases of fatal intoxication the initial symptoms have been followed by either shock due to fluid loss and death within 24 hours, or by acute renal failure with cardiopulmonary depression, liver damage and death in 7–14 days [5,23,125,128]. Necropsy has shown pulmonary edema, pleural effusions, and ascites, and hemorrhagic necrosis of organs in the GI tract [5,128]. Damage to the pancreas also resulted in glucose intolerance [5,128]. Oral administration of Cd compounds induces epithelial desquamation and necrosis of the gastric and intestinal mucosa, together with dystrophic changes of the liver, heart, and kidneys [5]. In non-fatal cases, recovery from acute poisoning is rapid and complete. The amount of Cd absorbed is probably very limited due to vomiting and the consequential short presence of Cd in the GI tract.

4.1.3 Animal Studies

In experimental animals, the readily soluble compounds have lower LD50 values than the insoluble ones after acute oral ingestion or inhalation of Cd compounds [5,129]. In experimental animals, LD50 inhalation values are in the range of 500 to 15 000 mg/m3min for different species [5,130]. The LD50 after the injection of soluble Cd compounds is in the range of 2.5–25 mg/kg b.w. [5,131]. For most Cd compounds, the LD50 after oral administration is about 10-20 times higher than after parenteral administration. Liver damage is probably the cause of death after high parenteral Cd application [132,133].

Numerous studies on the effects of Cd on the testes and other reproductive organs have been reviewed by Gunn and Gould [134] and Barlow and Sullivan [135]. Testicular necrosis occurs in experimental animals given a single subcutaneous injection of 2–4 mg Cd/kg b.w. [136,137]. Shortly after dose levels similar to the LD50 are injected, severe endothelial damage is seen in the small vessels of the testis [136]. This damage gives rise to increased capillary permeability, which results in edema, decreased capillary blood flow, ischemia, and testicular cell necrosis [138]. A single injection of Cd salts at a dose that induces testicular hemorrhagic necrosis has been shown to induce hemorrhages and necroses in the ovaries of prepubertal rats [139], and in the ovaries of adult rats in persistent oestrus [140]. Cd-induced testicular necrosis generally results in permanent infertility. Recent findings have shed light on the underlying mechanisms that cause acute Cd-induced damage to the testis (reviewed in [141]). They involve disruption of the blood-testis-barrier by damaging effects of Cd on E-cadherins [142] as well as p38 MAPK pathway-mediated disruption of adherens junctions [143]. ZIP8, which is expressed in Sertoli cells and in vascular endothelial cells, was found in the interstitium of the testis in mouse strains sensitive to Cd toxicity in the testis [32]. Cell damage and death may then occur via formation of reactive oxygen species (ROS) [144] (see Section 5.4).

At the molecular level, oxidative stress may play a major role to account for the cellular damage observed in organs affected by acute Cd intoxication, such as lung, liver or testes. Using electron spin resonance (ESR) for ROS detection following acute Cd overload in rats and mice, Liu et al. [145,146] have provided direct evidence for Cd-generated radical formation. In Cd-treated rats, radical adducts were formed in the liver, excreted into the bile, and were probably derived from endogenous lipids [146]. Whereas the presence or absence of MT did not affect ROS formation induced by Cd [145], depletion of hepatic glutathione significantly increased ROS production and inactivation of Kupffer cells or chelation of Fe inhibited it [146]. This suggests that disruption of the cellular GSH system, inflammatory processes in the liver, and the Fenton reaction driven by Fe, mainly contribute to ROS formation in acute hepatotoxicity by Cd. No data are available for lungs.

4.2 Chronic Toxicity

Autopsy studies in humans with CLCE have demonstrated that the Cd burden differs between organs [24,96,147]. About 50% of the retained Cd was found in the kidneys and liver with one third in the kidneys alone. More recent estimates showed that the kidneys and liver together contain ~85% of the Cd body burden, and more than 60% was found in the kidneys in the age range of 30-60 years [148]. Based on calculations of organ Cd burden according to age of 160 deceased Tokyo inhabitants, Tsuchiya et al. [98,149] estimated a biological half-life of Cd in the body of 13.4 years, in the liver of 6.2 years, and in the kidney of 17.6 years, but with large individual variations. Using a similar approach with 292 persons autopsied in Stockholm a biological half-life of kidney cortex of ~30 years was estimated in non-smokers [24]. The differential distribution of Cd in different organs and the very long biological half-life of Cd in kidney cortex may explain the increased sensitivity of the kidney cortex to Cd [150] and may also account for an increased incidence of chronic kidney disease in populations with CLCE (e.g., due to smoking [151]) or of end-stage renal failure [152].

Apart from the kidneys, other organs are also affected by Cd, which may lead to damage and dysfunction as well (see Section 4.3). In fact, Cd exposure increases the risk of mortality in affected populations. This was recently demonstrated in two prospective population-based cohort studies [153,154], which showed higher risk for death in association with Cd exposure. Of note, in both studies, there was no indication of renal disease, indicating that Cd-induced damage to other organs contributed to increased mortality.

4.2.1 Teratogenicity

There have been few studies on the fetal toxicity of Cd transported across the placenta, which acts as a barrier to Cd. Nevertheless, a small amount may reach the fetus. Cd is also transferred to neonates through lactation. Maternal hypertension and decrease in birth weight have been associated with elevated levels of Cd in the neonate [155]. A depletion of zinc with increasing number of births and a progressive increase in Cd in smokers negatively affect infant birth weight [156]. Moreover, Marlowe et al. [157] found an association between mental retardation and raised concentrations of Cd in hair in school age children. But overall, there is no substantial evidence that Cd has caused teratogenic effects in humans.

Embryotoxicity and teratogenicity have been demonstrated in experimental animals treated with Cd compounds. Large doses of Cd salts induce severe placental damage and fetal deaths when given at a late stage of pregnancy, and teratogenic effects, such as exencephaly, hydrocephaly, cleft lips and palate, microphthalmia, micrognathia, clubfoot, and dysplastic tail, when given at early stages of gestation [158–161] (reviewed in [162]). Recent studies have attempted to identify the genes [163,164] and signaling pathways [165,166] that are affected by Cd and contribute to embryotoxicity and teratogenicity in experimental animals. However, in all these studies the observed effects were induced by very high concentrations of Cd and may therefore not address relevant issues of CLCE.

4.2.2 Cancer

A causal relationship has been observed between Cd exposure and the occurrence of cancer in various organs in humans and the administration of Cd have resulted in tumors of multiple organs/tissues in experimental animals (reviewed in [167], see also Chapter 15). This has led the International Agency for Research on Cancer (IARC) to classify Cd as a human carcinogen based on increased incidences of lung tumors [168]. Moreover, human Cd exposure has been associated with increased incidences of renal cancers [169,170]. Nevertheless, results of studies involving Cd carcinogenesis have been contested because of confounding factors such as co-exposure to other toxic chemicals and life style factors, for example, cigarette smoking.

Recent epidemiological studies have accounted for all these factors and endorse the role of exposure to Cd in the development of cancer of various organs in humans, such as female breast and endometrial [171,172], lung [173], pancreatic [174,175], and bladder cancer [176]. Though of 11 cohort studies, only 3 implied Cd as a cause of prostate cancer [177], a recent case-control study [178] showed an excess cancer risk in subjects with increased Cd exposure. Multiple rodent studies have confirmed that Cd exposure leads to lung cancer (reviewed in [179]). Cd cancers in the prostate of rats occur at doses below the threshold for significant testicular toxicity because of a requirement for androgen production; in contrast, the rat testis will only develop tumors if Cd is given parenterally at high doses [180]. Cd exposure can also induce tumors of the pancreas, adrenals, liver, kidney, pituitary, and hematopoietic system in mice, rats, or hamsters (reviewed in [181]).

This raises the question about the underlying mechanisms for carcinogenesis, especially those relevant at comparatively low exposure conditions. Current evidence suggests that no direct genotoxicity but rather multiple indirect mechanisms are operative to induce cancer (reviewed in [182–184]).The most important among them are alterations in gene expression patterns, interactions with the cellular DNA damage response systems, and induction of oxidative stress. For further details, the reader is referred to Chaper 15 of this book by Andrea Hartwig.

The available evidence indicates that, perhaps, oxidative stress plays a central role in Cd carcinogenesis because of its involvement in Cd-induced aberrant gene expression, inhibition of DNA damage repair, and apoptosis (reviewed in [182,185,186]). In contrast, Liu et al. [187] emphasize that CLCE-induced carcinogenesis may develop independently of formation of ROS. Exposure of liver cells to a low dose (1.0 μM) of Cd that does not produce ROS induces malignant transformation after 28 weeks of continuous exposure, as evidenced by forming highly aggressive tumors upon inoculation of cells into nude mice [188]. To examine the role of ROS in Cd carcinogenesis, cellular ROS levels in control and Cd-transformed cells were determined using fluorescence probes. In Cd-transformed cells, the intracellular levels of ROS were significantly lower at basal conditions as compared to passage-matched control cells, and these transformed cells were also highly tolerant to high dose of Cd (50 μM) induced ROS production. These data indicated the minimal role of ROS in Cd-induced malignant transformation in rat liver cells, a target for Cd carcinogenesis [187,188]. These Cd transformed cells also showed acquired apoptotic tolerance, with marked reduction of redox-sensitive JNK signal transduction pathways [189]. Hence the role of ROS in malignant transformation induced by Cd needs to be clarified further.

4.3 Target Organs

4.3.1 Kidney

As a result of Cd’s tendency to accumulate in epithelial cells of the proximal tubule (PT), the kidney is usually the primary critical target organ of chronic Cd toxicity in the body [17]. The kidney is, in effect, a “sentinel of Cd exposure” [190]. Cd is not only toxic to PT cells, but also to glomeruli and distal tubules [191]. For several decades the following scenario has prevailed to account for chronic nephrotoxicity mediated by Cd [88]: After uptake into the systemic circulation, Cd is initially found in the blood plasma either loosely associated with molecules, such as albumin, amino acids or the sulfhydryl compounds, glutathione or cysteine, or tightly bound to specific metal binding proteins such as the low-molecular-weight protein MT (MW ~7 kDa) [192]. Cd which binds with low affinity to plasma components can dissociate and bind to other target molecules on cell surfaces and also enter the cells [193]. CdMT is not available for uptake by most tissues, but can be taken up by the epithelium of the PT [89]. Cd in plasma is initially transported to the liver, taken up, and in the cell Cd induces the synthesis of MT which efficiently binds and sequesters Cd, thereby buffering its toxic effects in the cell. A small proportion of liver MT is slowly released into blood plasma as the hepatocytes in which Cd is sequestered die off, either through normal turnover or as a result of Cd injury [90,194,195]. Gunn and Gould [196] and later Nordberg and Nishiyama [197] showed that in long-term exposure and even at long time intervals after a single exposure, the level of Cd initially is highest in the liver and then gradually increases in the kidneys. The strongest evidence for the concept that the major source of renal Cd in chronic Cd exposure is derived from hepatic Cd which is transported in the form of CdMT in blood plasma was provided by Chan et al. [198], who transplanted livers of Cd-exposed rats to normal rats. The levels of Cd and MT in the liver of recipient rats were decreased (106 and 1503 micrograms/g, respectively) with time after surgery. On the other hand, renal Cd and MT levels were markedly increased (195 and 1468 micrograms/g, respectively) and most of the Cd in the kidney was bound to MT [198].

Circulating Cd is filtered by the glomerulus because of the small molecular mass of the various circulating Cd forms, and mainly internalized by the S1-segment of the PT because this segment of the PT cells possess apical transporters, metabolizing brush-border enzymes, and receptors for free Cd2+ as well as for the complexed forms of Cd2+ (for a review see [193,199]). Like other low-molecular-weight proteins [200], MT is reabsorbed from primary urine into PT cells of the kidneys by megalin/cubilin receptor-mediated endocytosis (RME) [201,202]. There is also evidence that Cd and Cd-thiol conjugates can be taken up at the basolateral surface of PT cells [203–206] although the in vivo relevance of the contributing transporters is questionable considering their low affinity to Cd. With low, or even moderate, levels of exposure, little or no Cd is excreted in the urine. Only when the body burden of Cd is large and/or kidney injury begins to appear, urinary excretion of Cd increases significantly.

Even though the CdMT complex is non-toxic, its accumulation in the PT cells may eventually cause kidney damage. A continuous catabolism of endocytosed CdMT takes place in cultured PT cells where CdMT is trafficked to lysosomes [207,208]. Cd is split from MT in lysosomes and released into the cytosol via lysosomal DMT1-mediated transport [209,210] and bound to newly formed MT in the tubular cells. It is supposed that kidney damage is prevented until a stage is reached at which the kidneys can no longer produce enough MT (see below) though evidence for this process is lacking.

The scenario that CdMT filtered by the glomerulus is responsible for chronic Cd toxicity was recently questioned [55], mainly based on the following experimental data: (i) MT-null mice, which are unable to form the CdMT complex, are hypersensitive to chronic Cd nephropathy [211], even though the accumulation of Cd in the kidneys was only 7% of that in wild-type mice; (ii) In cultured renal cells, cytotoxicity of CdMT is much less than CdCl2, corresponding to less Cd uptake and accumulation from CdMT than from CdCl2 [212,213]. Thus, it has been proposed that Cd nephrotoxicity is due to accumulation of inorganic Cd, rather than CdMT [55]. However, the studies with MT-null mice only prove that MT synthesis in renal tubular tissue protects the kidney against Cd toxicity and that in MT-null mice where no CdMT complex is formed Cd can also mediate nephrotoxicity. The studies using MT-knockout mice [211] also showed that MT-null mice accumulate much lower Cd levels in their kidneys (10 μg/g wet weight) than control mice (140 μg/g), thus showing the importance of MT for transport of Cd to the kidney and accumulation in this tissue. The in vitro studies by Prozialeck et al. [212] and Liu et al. [213] tested Cd for short periods of time or used high CdMT concentrations on LLCPK1 cells. Cells did not survive long enough after Cd or CdMT exposure. In contrast, using 10 μM free or 10 μM CdMT (1.4 μM MT and 10 μM Cd) Erfurt et al. [214] have demonstrated that in PT cells Cd and CdMT can reach similar levels of toxicity though at different time points (12 h for Cd and 72 h for CdMT), most likely because intracellular trafficking of CdMT to lysosomes delays the onset of toxicity (see above). Experiments with chronic subcutaneous injections of Cd in mice deficient in DMT1-mediated transport [215] or renal megalin/cubilin [216] should be able to elucidate the exact role of RME of CdMT in toxicity of the renal PT.

There appears to be a critical concentration of Cd in the renal cortex that, once exceeded, is associated with tubular dysfunction. This concentration depends on the individual, and chronic Cd nephropathy is seen in about 10% of the population at renal concentrations of ~200 μg/g and in about 50% of the population at about 300 μg/g [17,217]. In fact, more than 7% of general populations may have significant Cd-induced kidney alterations due to chronic exposure with kidney Cd levels as low as 50 μg/g renal cortex [17]. The current dogma is that as this threshold concentration is approached, the Cd chelating capacities (e.g., MT, GSH) of cells and adaptive antioxidant defense mechanisms to offset Cd-induced ROS and oxidative stress (GSH, ROS metabolizing enzymes) are overwhelmed, lipid peroxidation and oxidative damage ensue that lead to injury and cell death (reviewed in [187]). Reabsorptive dysfunction is then associated with a general transport defect of the PT that mimics the de Toni-Debré-Fanconi-Syndrome and is characterized by polyuria, phosphaturia, aminoaciduria, glucosuria, and low-molecular-weight proteinuria [218,219]. This transport defect may be a possible result of ROS-induced damage of the Na/K-ATPase [220,221]. Tubular apoptosis due to generation of ROS has also been described in vivo and in vitro that could also contribute to the Fanconi-like syndrome [211,220].

Pathologically, the PT lesions consist of initial tubular cell apoptosis or necrosis, depending on the degree of cellular damage and degeneration. With chronic exposure, damage progresses to interstitial inflammation, renal fibrosis, and failure [152]. The earliest manifestations of tubular toxicity are increased excretion of low-molecular-weight proteins, such as β2-microglobulin, retinol-binding protein, MT, etc. [222]. The urinary excretion of markers of cytolysis such as the lysosomal enzyme N-acetyl glucosaminidase (NAG) also increases. Tubular proteinuria may progress to glomerular damage with a decreased glomerular filtration rate (GFR), as demonstrated in studies of occupationally exposed workers [223]. Recent studies in environmentally exposed populations suggest that decreased GFR and creatinine clearance as well as increased serum creatinine may occur at similar Cd dose levels as the tubular damage [224,225]. The presence of larger proteins in urine, like albumin or transferrin, has been attributed to glomerular damage. But these proteins are also filtered by the intact glomerulus and reabsorbed via RME [200,226]. Overall, the pathogenesis of the glomerular lesion in Cd nephropathy is not well understood. In vitro studies in cultured rat glomerular mesangial cells indicate that Cd induces c-fos protooncogene via ERK- and stress-activated protein kinase-dependent MAPK pathways and inhibits both extrinsic and intrinsic apoptosis signaling, which may contribute to cell survival and mesangial cell hyperproliferation, as observed in certain forms of glomerulonephritis [227,228].

Sporadic evidence for chronic toxicity of the distal portions of the nephron (reduced urinary kallikrein excretion) induced by Cd exposure has also been obtained, both in experimental animals following CLCE [229] and in Cd-exposed workers [230], but the mechanisms of distal tubular damage are unclear.

Kidney damage may further progress to chronic kidney diseases with albuminuria in populations with low levels of exposure, e.g., in smokers [151] or even to end-stage renal disease (ESRD). An increased risk of ESRD has been observed at environmental exposure levels based on records of all persons residing in Cd-polluted areas near battery plants [152]. Cd may also potentiate diabetes-induced effects on the kidney [224,231,232].

4.3.2 Liver

The liver is a major metabolic organ for the in vivo handling of Cd (see the previous paragraph), but little is known about the metal transporters and molecular mechanisms involved in hepatic uptake of Cd. Work with isolated hepatocytes and human liver cell lines have demonstrated two possible routes of uptake: One is for the free (ionic) form of Cd (Cd2+) and the other is for the complexed form of Cd2+ [233]. Cd2+ is most likely taken up by the same metal transporters that liver cells use to acquire physiologically essential metals, notably Fe2+, Zn2+, Mn2+, and Cu2+. Among those transporters is DMT-1, which is weakly expressed in liver [34,234]. Metal transporters of the Zrt-/Irt-like protein family, such as ZIP8 and ZIP14 [35,234] are also expressed in liver. However, the localization and cellular orientation of these transporters in liver cells, has not been established. Many reports have postulated Cd2+ uptake by voltage-dependent calcium channels in hepatic cells, however, whereas whole liver expresses transcripts for L-type Ca2+ channels, hepatocytes do not [235].

In vitro studies have shown that CdMT enters liver cells by receptor-mediated endocytosis [236]. However, Sabolic et al. recently demonstrated that CdMT is not taken up by rat hepatocytes in vivo [237]. Rather, Kupffer cells endocytose CdMT, which could lead to the release of various pro-inflammatory cytokines, including interleukin 6 (IL-6) and tumor necrosis factor-α (TNF-α). However, no specific molecular mechanism for this process has been identified so far. Hence, not much is known about Cd uptake by the liver. Nevertheless, liver tissues store Cd. Of note, liver Cd in humans varies between 1.5–8 μg/g dry weight depending on age, sex, environmental exposure, etc., which is 10–20 times lower than in the kidney [24,238,239]. This difference is striking considering that the liver is the first organ targeted following oral Cd uptake.

In rodents, but not in humans, Cd is hepatotoxic after chronic exposure. However, hepatotoxicity is much less prominent than renal damage [131,240–242]. The mild morphological and functional alterations associated with chronic hepatotoxicity by Cd may be due to a pronounced induction of MT and GSH in the liver, which chelate (and thereby inactivate) Cd respective prevent oxidative stress [243–245]. Moreover, both Cd-bound MT and GSH may be released into the circulation and bile, respectively, via ABCC2-mediated efflux of Cd-GSH [115–117] and exocytotic or necrotic release of CdMT into the circulation [90,194]. Based on recent studies [246,247], Liu et al. [187] have questioned a role of ROS formation in chronic hepatotoxicity (and carcinogenesis) during long-term exposure to Cd at low doses. Adaptive mechanisms including induction of MT, GSH, and cellular antioxidants could diminish Cd-induced oxidative stress. This view is, however, in contrast to other reports from the literature [243,248].

In summary, the liver has a large reserve capacity and ability to induce anti-oxidant systems and to excrete Cd complexes luminally into the bile and basolaterally into the circulation. This may account for the lower Cd content of the liver compared to the kidney and explain why Cd toxicity has not been reported in humans. However, it cannot be excluded that subtle alterations of liver function also occur during chronic Cd exposure, which cannot be detected by liver-related biomarkers. Lee and coworkers [249] examined statistical relationships between Cd in urine and circulating antioxidants in blood of humans from the NHANES III database. Circulating levels of a number of these antioxidant markers are primarily controlled by the liver. They found inverse relationships between Cd levels in urine and the levels of circulating antioxidant markers in blood. Thus, these effects may be indirectly linked to the oxidative effects of Cd in the liver and some of these parameters could be useful as Cd biomarkers for liver toxicity in the future.

4.3.3 Respiratory System

Cd concentrations in the lungs have been determined in humans and vary between 0.4–3.0 μg/g dry weight depending on age, sex, environmental exposure, smoking status, etc. [238,250,251]. Chronic obstructive lung disease frequently occurs as a result of long-term occupational exposure to Cd dusts and fumes (reviewed in [2,5]). The severity of the disease depends on exposure time and Cd dosage. Its manifestation may be slow and become apparent only after several years of exposure. In the upper respiratory system, chronic inflammation of the nose, pharynx, and larynx have been reported [5]. In the lower respiratory tract, the changes normally involve chronic bronchitis which may lead to a mild form of obstructive lung disease with functional impairment, while following more heavy exposure this condition may progress to diffuse interstitial fibrosis of the lower airways with alveolar damage and, in severe cases, emphysema [5,100,131,252,253]. The effects on the lung increase the mortality of Cd workers with high exposure [96,252,254]. However, if working conditions are improved the pulmonary changes are mild and occur less frequently and at a later stage than renal damage [255,256]. It is not clear whether lung impairment results from long-term exposure above a critical airborne Cd concentration or from several episodes of exposure leading to permanent changes. To prevent deleterious effects on the respiratory system, exposure to CdO fumes or dust should not exceed a Cd concentration of 20 μg/m3 during a 40 hour working week for working life [5].

In animal studies, Snider et al. [257] observed signs of emphysema in rats 10 days after 5–15 daily 1 h periods of exposure to CdCl2 aerosol (10 mg/m3). CdO fumes (10–90 μg Cd/m3), Cd salts (30–100 μg/m3) or dust (10–270 μg/m3) had similar effects [258–260]. Systemic chronic Cd appears to induce lung inflammation and cell proliferation in vivo in mice with up-regulation of proinflammatory cytokines (Cox-2, IL-6) and cell cycle regulatory molecules (STAT3, Akt, cyclin D1) and subsequent cell proliferation [261]. Pulmonary inflammation and emphysema induced by Cd inhalation appears to be associated with pulmonary oxidative stress. Rats undergoing chronic inhalation with CdCl2 aerosols show significant increase of bronchoalveolar lavage fluid macrophages, neutrophils and oxidant markers (8-iso-PGF(2α), uric acid) [262]. Moreover, reduced ascorbic acid and GSH were significantly lower in Cd-exposed rats, indicating antioxidant depletion, and pulmonary emphysema developed.

In vitro studies with cultured human airway epithelial cells and in vivo experiments in mice demonstrated that Cd (2–50 μM for 24 h or instillation of 10 nM Cd and tissue processing after 72 h) activates the proinflammatory cytokine IL-8 in an NF-κB-independent but ERK1/2-dependent manner to elicit lung inflammation [263]. Short-duration exposure to lower doses of Cd increased the growth of human lung adenocarcinoma epithelial A549 cells, whereas higher doses were toxic and caused cell death [264]. Cd induced elevated expression of epidermal growth factor receptor (EGFR) along with different proinflammatory cytokines like interleukin-1 beta (IL-1β), TNF-α, and IL-6 [264]. ZIP8 expression has been recently shown to be under the control of TNF-α in a NF-κB-dependent manner in primary human lung epithelia [37,265], and is associated with increased Cd uptake and toxicity. Consistent with these findings, a significant increase in ZIP8 mRNA and protein expression was observed in the lung of chronic smokers compared to non-smokers [37]. Moreover, when small-cell lung cancer-derived cell lines, SR3A, were stably transfected with glutamate cysteine ligase catalytic subunit to produce higher levels of GSH, cells became Cd-resistant, possibly because of down-regulation of ZIP8 through GSH-dependent suppression of the ZIP8 transcription factor Sp1 [266]. Interestingly, Cd exposure of human airway epithelial cells is also accompanied by decreased expression of the epithelial ion channel cystic fibrosis transmembrane conductance regulator (CFTR) [267], which plays a central role in maintaining fluid homeostasis and lung function, and also affects transcellular GSH levels and the cellular redox state in lung epithelia [268,269]. This may additionally contribute to oxidative stress and inflammation of Cd-affected pulmonary epithelia.

Chronic exposure to Cd may also be associated with lung cancer. In a population-based prospective cohort study with a median follow-up of 17.2 years in an area close to three zinc smelters, the association between incident lung cancer and urinary Cd was assessed [173]. The risk of lung cancer was 3.58 higher in a high exposure area compared to an area with low exposure. The risk for lung cancer was increased by 70% for a doubling of 24 h urinary Cd excretion, and confounding by co-exposure to arsenic was excluded. In a recent study, prospective data from a NHANES III cohort were examined to investigate the relationship between Cd exposure and cancer mortality, and the specific cancers associated with Cd exposure, in the general population [270]. In men and women urinary Cd was associated with mortality from lung cancer though this study was limited by the possibility of uncontrolled confounding by cigarette smoking or other factors, and the limited number of deaths due to some cancers. A population-based lung cancer case-control study in Central/Eastern Europe and UK was conducted in 1998–2003, including 2853 cases and 3104 controls with adjustment for confounding factors and concluded that occupational exposure to Cd (and other metals) is an important risk factor for lung cancer [271].

In experimental studies, lung cancers were induced in rats by exposure to CdCl2 aerosols for 13 months as a function of the Cd concentration (12.5–50 μg/m3) [272]. Whereas there seems to be sufficient evidence for pulmonary carcinogenicity of inhaled Cd compounds in rats no such evidence was found in mice and hamsters, indicating significant species differences in the pulmonary response to inhaled Cd. It has been demonstrated that expression of MT in the lung after inhalation of Cd differs between species due to genetic variability in pulmonary MT, which may influence the susceptibility of rats or mice to lung carcinogenesis induced by inhalation of Cd compounds [273,274].

4.3.4 Bone

The relationship between Cd toxicity and bone disease was first described in Japan in the 1950s, when people living in the Jinzu river basin area began developing multiple bone fractures, severe bone pain, and malformations of their long bones. Fractures occurred mainly in elderly multiparous women, with a form of osteomalacia, osteoporosis, renal dysfunction and renal stone formation, a syndrome that became known as Itai-Itai disease [275,276] (reviewed in [1]). Intoxication due to relatively high Cd concentrations was eventually found to be due to industrial waste discharged into the river from an upstream zinc mine. At the concentrations encountered in Itai-Itai disease Cd disrupts renal calcium and phosphate transport as well as the actions of parathyroid hormone and vitamin D metabolism, at least partially through renal dysfunction with cellular damage and death (mainly of proximal tubules), but also inhibition and/or reduced expression of epithelial calcium channels (TRPV5) and Na-phosphate co-transporters (SLC34A1) (reviewed in [277]), and excess excretion of calcium often occurs in the urine. Hence the skeletal changes are probably related to renal Fanconi syndrome (see Section 4.3.1).

But issues with loss of bone density, height loss, and increased bone fractures have now been reported in populations exposed to far lower levels of environmental Cd than Itai-Itai victims [278]. Recent epidemiological evidence accumulates suggesting that besides kidney the bone is also a primary target organ of Cd toxicity in humans [279–283]. Sughis et al. [284] found that even in young children, low-level environmental exposure to Cd is associated with evidence of bone resorption, suggesting a direct osteotoxic effect with increased calciuria. These findings might have clinical relevance when these populations reach an older age.

Experimental studies have also addressed the question whether very low concentrations of dietary Cd that do not damage the kidney can negatively impact the skeleton by directly affecting bone tissues. Ogoshi et al. [285] reported a decrease in mechanical strength of bones in weanling rats after only 4 weeks of exposure to 5 or 10 μg Cd/mL in the drinking water – exposure levels that give urine Cd concentrations in the range of smokers. Other studies in rats designed to model human CLCE [286,287] showed that Cd affects the mineral status leading to weakening in its mechanical properties and that these processes occur prior to skeletal maturity [285,288]. Acute or chronic Cd administration resulting in low blood Cd levels increased calcium excretion in rodents well before the onset of kidney damage leading to aminoaciduria and proteinuria [289–291].

In vitro experiments with cultured cells support these in vivo studies. In a mouse osteoblast-like cell line, MC3T3-E1, 0.1 to 1 μM Cd decreased cellular alkaline phosphatase activity, a marker of osteoblast differentiation [292]. Moreover, >1 μM Cd increased prostaglandin E2 secretion by MC3T3-E1 cells, which could stimulate formation and activation of osteoclasts and lead to osteoclast-mediated bone resorption [293,294]. These effects, taken together, could clearly cause an uncoupling of the normal balance between bone formation and bone resorption. Moreover, Cd at 10 to 100 nM has been also shown to stimulate the formation and bone degrading activity of osteoclast-like multinuclear cells from progenitors in bone marrow cultures [295,296]. Some of the damaging effects of Cd on bone tissues may be caused by increased oxidative stress [297]. Further, gene expression microarray and gene knock-out mouse models have shown that Cd induces MT1 and MT2 to protect the cells but also stimulates bone demineralization via a fos-independent, but src- and p38 MAPK-dependent pathway involving osteoclast activation and resulting in the breakdown of bone matrix (reviewed in [298]).

Uptake of low micromolar Cd concentrations into human osteoblast-like MG-63 cells [299] and mouse MC3T3-E1 cells [300] has been recently demonstrated. The studies suggested that uptake of Cd and cytotoxicity are mediated by transient receptor potential melastatin-related 7 (TRPM7) channels though the permeability of TRPM7 to Cd is weak [301]. The authors also proposed that deficiency of Ca2+ or Mg2+, which are preferentially transported by TRPM7 [301], may enhance Cd uptake into osteoblast cells.

4.3.5 Cardiovascular System

Most early studies on chronic occupational or environmental Cd exposure did not find any evidence for cardiovascular disease or increased mortality due to cardiovascular disease (reviewed in [5]). Cross-sectional and prospective studies by Staessen et al. [302] showed that blood pressure, or the risk of hypertension or cardiovascular diseases risk in environmentally exposed populations were not associated with 24-h urinary Cd. Subjects with Itai-Itai disease also failed to develop hypertension [303]. Similar conclusions were obtained from other epidemiological studies [304], and from studies in Belgian populations, which did not show any correlations between measures of arterial function and blood Cd and failed to confirm that increased Cd body burden was associated with decreased arterial function [305].

In contrast to these negative findings, Thun et al. [306] found that mean systolic and diastolic blood pressures were higher in 45 Cd workers (134 and 80 mmHg, respectively) than in 32 male controls (120 and 73 mmHg respectively). Systolic but not diastolic blood pressure was significantly associated with Cd dose in multivariate analyses. Recently, Navas-Acien et al. [307,308] reported that peripheral arterial disease might be associated with increased blood and urinary Cd. Satarug et al. [309] showed a positive association between blood pressure and urinary Cd in a population sample of 200 subjects that also showed tubular dysfunction. The Atherosclerosis Risk Factors in Female Youngsters (ARFY) study showed that increased blood Cd level are associated with early atherosclerotic vessel wall thickening [310]. The population-based U.S. NHANES study found that Cd exposure is associated with an increased risk of cardiovascular disease [311–314]. Cd concentrations in cardiovascular tissues (heart, aorta) have been determined in human populations and vary between 0.05 and 1 μg/g dry wet and depend on age, environmental exposure, and other aspects [238].

Animal studies support effects of Cd on blood pressure and the cardiovascular system and suggest that Cd may cause atherosclerosis and/or be toxic to the myocardium (reviewed in [315], see also [316]). Maternal exposure during pregnancy may even reprogram cardiovascular development without detectable Cd in fetal and adult tissues of the offspring [317]. The cellular and molecular foundations and pathophysiology of vascular diseases associated with chronic Cd toxicity are beginning to emerge [318,319]. Kaji et al. [320] have demonstrated that Cd inhibits proliferation and destroys the monolayer of vascular endothelial cells in a culture system. Abu-Hayyeh et al. [321] found a positive correlation between the smoking status of humans and Cd concentrations in the aortic wall, with selective accumulation in the medial layer. Their finding that the aortic wall accumulates ~7 μmol/L Cd, in contrast to nanomolar blood concentrations, indicates that the arterial wall takes up Cd and may be a target of Cd toxicity. Uptake mechanisms are unclear. SLC39A8 (ZIP9) is expressed in endothelial cells of the testis vasculature [32] and may account for testicular hemorrhage associated with acute Cd toxicity (see acute toxicity). Other endothelial localizations of SLC39A8, however, have not been described. DMT1 is also expressed in vasculature [322] and human umbilical vascular endothelial cells [323]. In the latter study, DMT1 was increased by the inflammatory mediator TNF-α. This was associated with increased iron uptake and oxidative stress [323], and could also occur in Cd-induced vascular disease. Whether Cd protein complexes are taken up by endothelia via endocytosis is unknown. Megalin is expressed in cerebral vascular endothelia where it contributes to protein transfer [324]. However, Kaji et al. [325] showed that CdMT and CdGSH are less likely to be taken up by bovine aortic endothelial cells than Cd and also cause less cytotoxicity. Free Cd may also disrupt VE cadherin-dependent cell adherens junctions of endothelial monolayers, as evidenced by in vitro and in vivo studies (reviewed in [318]), which could account for an increased permeability of the endothelial lining and development of atherosclerosis. Cd may also inhibit endothelial proliferation [325] and induce cell death by apoptosis [326] or necrosis [320,325]. ROS formation may be the mechanism underlying vascular dysfunction in vivo [327,328] and endothelial damage and death in vitro [329,330]. An additional effect of Cd contributing to atherosclerosis may be induction of smooth muscle cell proliferation, which could induce arterial wall thickening [331] and interference with fibrinolysis [332]. Cd may also reduce vasodilatatory NO [333] by inhibiting endothelial NO synthase [334,335]. As a consequence of the release of pro-inflammatory mediators, such as TNF-α [323,336] and increased leukocyte adhesion [337,338] Cd may induce vascular inflammation and thereby promote atherosclerosis [339].

Cd has also been shown to induce cardiac damage in experimental settings by two possible mechanisms: (i) disruption of tissue structure and integrity; (ii) effects on cardiac conduction (reviewed in [5]). These effects were thought to be related to (i) decreased high-energy phosphate storage in the myocardium, (ii) reduced myocardial contractility, (iii) diminished excitability of the cardiac conduction system, and (iv) a reduction in coronary blood flow by Cd in isolated heart studies due to direct actions on the coronary vasculature [340].

Morphological alterations, including necrosis, cellular degeneration, and damage to intercalated discs, have been described in rats chronically exposed to CdO fumes or Cd gavage [341,342]. Damage of cardiac cells by Cd may involve oxidative stress to the heart by an alteration of antioxidant defense and an increased generation of ROS [343], inhibition of cardiomyocyte electron transfer chain [344], as well as a direct interactions of Cd with troponin C [345] or myoglobin [346]. Cardiac cell Cd uptake could be mediated by DMT1 [34,347], megalin [348], or the lipocalin-2 receptor [349]. Nevertheless, their exact cardiac localization remains to be determined.

Effects on cardiac excitability may be based on blocking effects of Cd on ion channels, e.g. voltage-dependent calcium channels (see [199] for a review) or Na channels of Purkinje fibres [350]. Cd may also increase outward potassium currents of ventricular myocytes and shift their voltage dependency of activation to more positive voltages [351], or interfere with Ca-dependent processes in the heart [352].

4.3.6 Nervous System

Initial reports described nervous system symptoms, including headache, vertigo, and sleep disturbance [353]. Physical examination revealed increases in knee-joint reflexes, tremor, dermographia, and sweating. In human studies, anosmia has been described among Cd workers exposed to CdO dust for long periods of time in some studies [131], but not in others [354]. At the cellular level, Cd may damage neuronal function by disrupting synaptic transmission [355], by affecting intracellular sulfhydryl homeostasis as a consequence of oxidative stress [356,357], and resulting in apoptosis or necrosis [358,359].

The blood-brain barrier (BBB) severely limits Cd access to the central nervous system. A direct toxic effect of Cd on the brain occurs only prior to BBB formation and is hence age-dependent in experimental animals [360]. The brain of newborn animals is permeable to Cd which decreases with age, probably due to increased MT expression (MT-3 in brain [361]; see below) and blood brain barrier maturation [362,363]. BBB dysfunction under certain pathological conditions increases the permeability to Cd, as described in a case report following acute Cd intoxication [364]. Additionally, the choroids plexus may accumulate high levels of Cd (reviewed in [365]). Nishimura et al. [366] observed strong MT immunostaining in ependymal cells and choroid plexus epithelium in younger rats (1–3 weeks old) poisoned with Cd. Thus, the sequestration of Cd by MT may partly contribute to the high accumulation of Cd in the choroid plexus.

Because of the functional resemblance between renal proximal tubular and inner ear epithelial cells it has been hypothesized that Cd might also impair the function of inner ear cells. In rats exposed to drinking water containing 15 ppm CdCl2 for 30 days, high Cd accumulation in ear ossicles and labyrinth was associated with signs of defective hearing and nephrotoxicity, but 5 ppm CdCl2 caused hearing loss without affecting kidney function [367]. These results suggested that hair cells are more sensitive to Cd than kidney tubule cells.

4.3.7 Reproductive System

Animal studies on chronic Cd toxicity of the reproductive system are mostly limited to male reproductive organs. They are also inconsistent due to the variability of the Cd dosage and period of exposure. To summarize the data, chronic exposure leads to smaller testis weight and reduced endocrine function, reduced spermatogenesis, reduced spermatozoa motility, decreased fertility, sterility, DNA damage in primary spermatocytes [368] and damage to testicular cells and structures, such as Sertoli cells and seminiferous tubules (early data reviewed in [5]). Recently, chronic exposition to low doses of Cd (0.015 g/L of CdCl2 in drinking water) for 3, 6, 12, and 18 months) was investigated in male mice [369]. Vascular lesions were evident from 6 months of Cd exposure and the severity of Leydig cells morphological changes increased with exposure time. Cellular degeneration was frequent after 12 months of Cd exposure. Also two Leydig cell tumors were observed after ≥12 months Cd exposure. In contrast, in a study by Zenick et al. [370] there were no significant effects on any of the parameters of reproductive toxicity or mating in male rats exposed to ≤70 ppm Cd for 70 days. Recent studies on Sertoli cells in vitro [371,372] have demonstrated a number of early molecular responses to Cd exposure. Yu et al. [371] reported dose- and time-dependent morphological changes associated with apoptosis signaling in rat Sertoli cells. They also observed disruption of the ubiquitin-proteasome system, up-regulation of cellular stress response proteins and the tumor suppressor protein p53.

In a recent review, Thompson and Bannigan [373] noted that Cd accumulates in the ovary with age and is associated with decrements in oocyte development. Whereas several animal studies noticed decreased ovarial steroid hormone synthesis and release in rats following CLCE [374,375], in vitro studies have demonstrated increased progesterone production (reviewed in [376]). The reasons for these differences are unclear.

Human studies are also variable in their outcome. No difference was found in fertility between men occupationally exposed to Cd and an appropriately matched control group (n = 118) by assessing birth experiences of their wives [377]. In contrast, a hospital sample of the general population with infertility problems showed that blood and seminal Cd were significantly higher among infertility patients than controls [378]. The percentage of motile sperm and sperm concentration inversely correlated with seminal plasma Cd among the infertility patients. In a recent study [379], human fetal male and female gonads were recovered during the first trimester and cultured with or without Cd. Cd, at concentrations as low as 1 μM, significantly decreased germ cell density and increased apoptosis, but did not affect proliferation in human fetal ovaries. Similarly, in the human fetal testis, Cd (1 μM) reduced germ cell number without affecting testosterone secretion [379]. Further experimental as well epidemiological studies are necessary to clarify the role of chronic Cd exposure on gonadotoxicity and fertility. Nevertheless, there is an increasing awareness of the possible impact of CLCE on fertility (reviewed in [380,381]).

4.3.8 Endocrine Glands

Experimental data in vitro and in animals suggest that Cd may disrupt the regulatory mechanisms of the hypothalamic-pituitary axis and change the secretory pattern of pituitary hormones like prolactin, ACTH, GH or TSH, resulting in disorders of the endocrine and/or immune system [382] (reviewed in [383]), which could account for the dysfunction of adrenal and thyroid glands observed in experimental animals during chronic exposure to Cd [384–387]. But direct effects of Cd on metabolism and secretion of peripheral glands have also been suggested [388–391]. No data are available from human studies.

Cd exposure has recently been considered to be a risk factor linked to diabetes mellitus type-2. Edwards and Prozialeck [392] and Chen et al. [393] reviewed the literature supporting a relationship between Cd exposure, elevated blood glucose levels, and the development of type-2 diabetes. In the Third National Health and Nutrition Examination Survey, a significant and dose-dependent association was found between elevations in urinary Cd levels and both impaired fasting blood glucose levels (110–126 mg/dL) and type-2 diabetes in persons >40y without renal damage [394]. A recent evaluation of Cd concentrations in biological samples of diabetes mellitus patients type-2 (age range 31–60y) has shown that the mean concentrations of blood Cd of male non-smoker and smoker diabetic patients were significantly higher than in their respective controls [395]. A prospective study, in which Cd levels are determined before the development of disease, is necessary to confirm a causal link between Cd exposure and development of diabetes. Edwards and Prozialeck [392] also provided experimental evidence indicating that Cd elevates fasting blood glucose levels in an animal model of subchronic Cd exposure before overt signs of renal dysfunction become evident. This study also showed that Cd reduces insulin levels and has direct cytotoxic effects on the pancreas. The pathogenesis of Cd-associated islet dysfunction remains to be explored. These findings also raise the possibility that Cd and diabetes-related hyperglycemia may act synergistically to damage the kidney.

4.3.9 Hematopoiesis and Hemostasis

After Cd exposure, decreased hemoglobin concentration and hematocrit are among the early signs of Cd toxicity (reviewed in [5]). Fe administration is beneficial suggesting decreased gastrointestinal absorption of Fe due to Cd [5]. Subsequent experimental studies in rats [396–399], cultured erythropoietin (EPO) producing cells [400] and populations with Itai-Itai disease [401,402] indicated that three main factors are involved in the development of Cd-induced anemia: hemolysis, Fe deficiency, and renal damage. Hemolysis can occur at the early stage of Cd exposure. Fe deficiency occurs because of competition of Cd with Fe for absorption in the intestine. However, an increase in body Fe content along with anemia is often observed in cases of parenteral Cd exposure or Itai-Itai disease, which may be caused by Fe from hemolysis, insufficient erythropoiesis, and hepatic ferritin overproduction induced by Cd-dependent interleukin-6 production [398]. Renal anemia is observed during the last phase of chronic, severe Cd intoxication, such as Itai-Itai disease. Cd suppresses renal EPO production through a direct effect, accumulation of toxic concentrations of Fe in kidney tissues, and destruction of EPO-producing cells [398,400].

Only a few studies have investigated the impact of chronic Cd exposure on immunity and hemostasis. Chronic exposure of mice to Cd by ingestion or inhalation inhibited cell-mediated immunity [403] and induced lymphopenia [404], but not in humans [405] or monkeys [406]. In contrast, in vitro studies with human peripheral blood mononuclear cells demonstrated Cd-induced interleukin-8 production with a concomitant generation of superoxide radicals [407]. Exposure of rats to Cd also increased platelet aggregation [408]. Moreover, chronic oral Cd exposure shortened prothrombin time and activated partial thromboplastin time [409]. Protein C and antithrombin also decreased in rat plasma after Cd exposure, but thrombocyte counts were not affected [409]. Hence, animal studies suggest that chronic Cd toxicity induces a hypercoagulable state and thereby increases the risk of thrombosis.

4.4 Early Biomarkers

One of the major challenges of public health and Cd toxicology is monitoring of affected populations for early signs of exposure and toxicity. Due to the kinetics of Cd in the body, which redistributes to organs such as liver and kidney and is stored there for years, blood Cd levels provide only information on acute exposure to Cd, but not on total body burden of Cd or organ damage. Nevertheless, the blood levels of Cd in non-exposed populations are normally less than 0.5 μg/L and blood levels higher than 1.0 μg/L are indicative of Cd exposure; levels higher than 5 μg/L are considered hazardous. Excellent reviews on the topic have been recently published by experts in the field [12,190,410,411]. This section summarizes the most important features of these reviews and also focuses on novel and promising early biomarkers of damage.

Many classical biomarkers detect Cd toxicity at a stage when functional damage is already advanced and perhaps irreversible. Moreover, results of recent studies have shown that even low (“normal”) levels of Cd exposure over a long period of time (CLCE) have subtle but insidious adverse effects on the kidney (see [3,4]), where traditional biomarkers may have their limitations [94]. Therefore, there is an increasing need for sensitive biomarkers of early exposure to low Cd concentrations where no overt pathology is apparent.

A Cd biomarker is “any substance or molecule that can serve as an indicator of the functional state or level of toxic injury in an organism, organ, tissue or cell” [190]. Practically, sampling of Cd biomarkers should be easy and without expensive, invasive or damaging procedures because it should be applied to large populations. Measurements should be limited to urine (and/or blood) samples because the kidney is the “sentinel of Cd toxicity” [190] (see Section 4.3.1). For the above reasons a recent suggestion to measure renal autophagy as an early biomarker of subtoxic Cd exposure is not practical [412].

Functionally, urinary markers of Cd nephrotoxicity can be classified into 3 groups: (1) Cd and Cd-binding proteins such as MT, (2) low-molecular-weight proteins, and (3) proteins and enzymes derived from the brush border, intracellular organelles or the cytosol of proximal tubule epithelial cells.

Ad (1) Urinary excretion of Cd and MT have been used both as markers of Cd exposure and of Cd-induced proximal tubule injury (reviewed in [222,413]). At early stages of Cd exposure, urinary Cd or MT most likely results from normal turnover and shedding of epithelial cells and is a reflection of the level of Cd exposure and the body burden of Cd (but see as a caveat the animal study by Thijssen et al. [94] when the body Cd burden is low). When the concentration of Cd in the epithelial cells reaches a threshold value of ~150–200 μg/g wet weight [17,217,219] Cd disrupts tubular reabsorptive processes and the excretion of Cd and MT begin to increase in a linear manner, which is associated with the onset of polyuria and proteinuria (see section on kidney). The early, linear phases of Cd and MT excretion mirror the level of Cd exposure whereas the later increases in excretion reflect Cd-induced tubular injury. Urinary levels of Cd in non-exposed populations are usually below 0.5 μg/g creatinine. The critical urinary Cd concentration that is associated with the onset of renal injury is about 2–10 μg/g creatinine, which corresponds to the critical renal cortical Cd concentration of ~150–200 μg/g tissue [17,217,219]. But there is significant evidence that even lower urinary levels of Cd may be associated with adverse renal effects [150,414,415]. The critical urinary level of MT that is associated with the onset of overt kidney injury is ~300 μg/g creatinine, which is based on a value of 2–3 μg urinary Cd/g creatinine [416,417] or even less [418].

Ad (2) Low molecular weight proteins such as β2-microglobulin, Clara-cell protein (CC-16), α1-microglobulin, and retinol binding protein are plasma proteins, which are small enough to be filtered at the glomerulus. Intact PT cells almost completely reabsorb these proteins via megalin and cubilin [419]; they are therefore not detected in the urine [420] (reviewed in [222]). As Cd accumulates and damages the PT, reabsorption of these proteins becomes impaired and they appear in the urine. β2-microglobin has been most widely employed as a standard marker for monitoring early stages of Cd exposure and toxicity. Urinary levels of β2-microglobulin greater than (300–1,000 μg/g creatinine) are indicative of PT dysfunction [421] (http://www.osha.gov/pls/oshaweb/owadisp.show_document?p_table=standards&p_id=10035) but lower critical levels have also been suggested [415].

Ad (3) Enzymes, such as N-acetyl-β-D-glucosamidase (NAG), lactate dehydrogenase (LDH), alkaline phosphatase, and α-glutathione-S-transferase (α-GST) are released from dead PT cells that are washed out into the urine. The lysosomal enzyme NAG has been very useful in the monitoring of human populations [414,422]. Recent studies suggest that α-GST may be a more sensitive early marker of Cd-induced kidney injury than NAG in human populations and animal experiments [423–425]. It is up-regulated by both Cd and oxidative stress, which could explain why it appears in urine before other cytosolic enzymes.

Another promising novel early biomarker is kidney injury molecule-1 (Kim-1). Kim-1 is a transmembrane protein that is not detectable in normal kidney but is expressed at high levels in the proximal tubule after ischemic or toxic injury (reviewed in [426]). Kim-1 regulates cell adhesion and endocytosis in regenerating cells of the injured tubule as they reform a functional epithelial barrier [427]. This process is associated with the proteolytic cleavage of the ectodomain of Kim-1 into the urine [427]. It is now used as a sensitive marker for the monitoring and diagnosis of kidney disease in humans ([428]; http://www.fda.gov/bbs/topics/NEWS/2008/NEW01850.html). Studies of chronic Cd toxicity in rats showed that Kim-1 is detected in the urine 4–5 weeks before the onset of proteinuria, i.e., when there was little PT cell death, and 2–5 weeks before the appearance of MT and CC-16 [428,429]. Recently, Pennemans et al. [430] have demonstrated that urinary Kim-1 levels positively correlated with urinary Cd concentration in an elderly population living in an area with long-term low-dose exposure to Cd, while other classical markers did not show any association. Therefore, urinary Kim-1 might be considered as a biomarker for early-stage metal-induced kidney injury by Cd.

Another very promising early biomarker of kidney damage may be neutrophil gelatinase-associated lipocalin or lipocalin-2, an acute phase 25 kDa glycoprotein which is part of the lipocalin family. NGAL is highly induced in inflammatory conditions and ischemia, and is a critical component of innate immunity to bacterial infection because it sequesters bacterial iron chelating siderophores and consequently blocks bacterial growth [431]. NGAL is also involved in growth and differentiation, especially in response to renal tissue damage and during nephrogenesis [432]. Studies in humans and experimental animals have shown that serum and urine NGAL are diagnostic and prognostic markers of acute kidney injury (reviewed in [433]) and a relevant biomarker of chronic renal failure [434]. A recent experimental study in mice demonstrated that increases in urinary NGAL (uNGAL) excretion were much more rapid (after 3–6 h) and sensitive than urinary Kim-1 (which significantly increased after 24 h) in assessing early acute kidney injury [435]. Hence, with earlier detection via Kim-1 or NGAL, it may be possible to reverse, or at least more effectively treat, Cd-induced kidney injury.

4.5 Prevention and Therapy of Toxicity

4.5.1 Prevention

In the household, practical preventive measures seek to remove Cd contaminated dusts (reviewed in [167]). In a work environment with occupational exposure to Cd fumes or dust, engineering control and ventilation should ensure that concentration levels are kept low. If constant safe air levels are not feasible, respiratory protection must be used. General rules for handling of toxic compounds should apply as well, i.e., smoking, eating, and drinking in work areas should be prohibited. Medical examination of Cd-exposed workers should be carried out at least once every year with a focus on the respiratory system and the kidneys. Cd and other early biomarker levels in blood and in urine should be checked regularly. WHO and OSHA (USA) recommend that the Cd concentration should not exceed 5–10 μg Cd/g urinary creatinine. This value is in contrast to the Cadmibel study in Belgium, which was the first to link urinary Cd with tubular proteinuria in a population study and showed that there was a 10% chance of developing tubular dysfunction at urinary Cd levels of 2–3 μg/g creatinine [436,437].

4.5.2 Therapy of Acute Poisoning

Apart from symptomatic treatment, there is currently no effective clinical treatment for Cd intoxication. One approach consists of trying to eliminate Cd using metal chelators such as ethylenediamine-N,N,N',N'-tetraacetate (EDTA) given by intravenous infusion to increase its urinary excretion [438], or if toxicity is so severe that renal function is impaired, EDTA is added to the dialysate during hemodialysis [439]. However, the role of EDTA remains unclear because of the limited number of reports in the literature and because chelation therapy for Cd may increase the uptake of Cd by the kidneys and increase the risk of nephrotoxicity. Friberg and Elinder [437] therefore recommended that EDTA is contra-indicated because of its nephrotoxicity when administered in combination with Cd. In experimental systems some chelators can reduce acute Cd-induced mortality [440]. Oral dimercaptosuccinic acid (DMSA) [441,442] or calcium disodium diethylenetriaminepentaacetate (DTPA) given parenterally [443] are the most effective antidotes, provided that treatment is started very soon after Cd ingestion.

4.5.3 Therapy of Chronic Poisoning

There is no recommended chelation treatment for chronic Cd exposure in humans [444]. However, in their review Jones and Cherian [444] highlighted a reduction of whole body, renal and hepatic levels of Cd in mice treated with compounds, such as sodium N-(4-methoxylbenzyl)-D-glucamine dithiocarbamate, or in rats using DTPA with or without 2,3-dimercaptopropanol [445]. One recent report described a combined intravenous administration of EDTA and GSH in a patient with chronic Cd intoxication [446]. Renal damage did not develop over the observation period of 6 months.

5 Cellular and Molecular Mechanisms of Toxicity

5.1 Entry Pathways

To enter the intracellular space, Cd2+ in extracellular fluids that is present as a free ion or complexed to proteins or peptides must permeate lipophilic cellular membranes. This occurs through intrinsic proteinous pathways. Free Cd2+ may be transported via ion channels and solute carriers and Cd2+ complexes may be taken up through receptor-mediated endocytosis. Cd2+ has similar chemical properties as essential metals (“ionic mimicry”) and Cd2+ complexes are analogous to endogenous biological molecules (“molecular mimicry”) [447,448]. Hence, transport (and toxicity) of Cd2+ can only occur if cells possess transport pathways for essential metals or biological molecules. After entering the intracellular compartment, higher affinity complex formation of free Cd2+ to other proteins may take place by ligand exchange, thereby also preventing back diffusion and forming a sink for Cd2+.

A variety of pathways have been suggested to allow Cd2+ entry in excitable and non-excitable cells [193]. The most likely candidates, whose molecular structures have been identified, are listed in Table 1. Proof of their Cd2+ transporting abilities has been obtained using stringent experimental approaches, such as radiotracer, fluorescent dye and/or electrophysiological transport assays combined with molecular biology techniques. Further information is available in a recent detailed review [199].

Table 1 Putative uptake pathways for Cd2+ and Cd2+-protein complexes in mammalian cells. For further details, refer to the text. M2+ = divalent metal.
Full size table

5.2 Interference with Transport and Homeostasis of Essential Metals and Biological Molecules

Several exhaustive reviews have summarized the literature on the effect of Cd on biomolecules involved in signal transduction [460], ion transport [277] and metalloproteins [461]. Cd binds to and/or interferes with a large number of transporters (e.g., pumps, channels, or carriers), signaling molecules (receptors, enzymes, transcription factors, etc.) and metalloproteins (e.g., Zn-, Ca-, Fe-, Mn-, Cu-binding proteins). Apart from metalloproteins and transporters for essential metals, most binding sites are “non-specific” low-affinity domains of biological molecules, but interaction with Cd may induce conformational changes and thereby affect their function. Binding occurs only when high concentrations of Cd are used, which will never be achieved in vivo. In fact, blood Cd concentrations in the general population are in the range of 1–10 nM [95] and may not exceed 100–300 nM in occupationally exposed workers [462]. The free Cd concentrations in the extracellular fluid that cause tissue damage are unknown but are likely to be in the submicromolar range: Acute poisoning with oral intake of a high dose of Cd resulted in a Cd concentration in the blood of ~0.2 μM [127]. Moreover, experimental studies in rats have shown that keeping artificially the free plasma Cd concentration to ~5 μM for only 30 min causes PT damage with a Fanconi-like syndrome [463] suggesting that (sub-) micromolar concentrations of Cd directly affect transport functions.

Hence, most of the studies describing direct effects of Cd on the function of transporters (reviewed in [277]) may have only in vitro or mechanistic relevance and are not likely to significantly contribute to in vivo toxicity of Cd in tissues. Indeed, few mammalian transporters are directly affected by submicromolar Cd concentrations: Ca-ATPase of rat intestinal and renal basolateral plasma membrane is blocked with an IC50 of 1.6 nM [464] and apical Na-dependent glucose and amino transporters of isolated rabbit PT (S2-segment) by nanomolar concentrations of CdMT [221]. However, it cannot be excluded that the effects observed were due to CdMT-induced endocytosis of apical membranes containing Na-dependent transporters [465]. As shown in Table 1, ZIP8 and ZIP14 (SLC39A8 and SLC39A14) as well as SLC11A2 (DMT1, DCT1, NRAMP2) transport Cd at submicromolar Cd concentrations, which indicates that Cd may disrupt Fe, Zn, and Mn homeostasis by competing for transport with these essential metals. This in turn will affect metalloproteins whose function depends on these metals (e.g., MAP kinases, PKC, NF-κB, Nrf2, Sp1, CREB, MnSOD, Cu-ZnSOD, MT, GSH, etc.; reviewed in [461]) with their resulting impact on cellular redox status, signaling, and gene transcription.

Effects of nanomolar Cd concentrations on various signaling pathways have also been described (summarized in [460]). Several observations suggest that submicromolar Cd affects Ca signaling: (i) Cd inhibits plasma membrane Ca-ATPase [464]; (ii) Cd activates protein kinase C directly [466]; (iii) long-term (14 days) Cd exposure studies in rat liver epithelial cells at doses of 0.03–2.5 μM CdCl2 showed disruption of gap junctional intercellular communication [467]; (iv) a Ca-sensing G-protein coupled receptor expressed in fibroblasts and renal MDCK cells similar to the Ca-receptor expressed in parathyroid, kidney, and gut cells is activated by submicromolar Cd with subsequent Ca mobilization [468,469]. Cd (0.01–1 μM) also interferes with NO signaling directly by blocking phosphorylation of NO synthase 3 [334] or suppressing NO synthase 2 activity indirectly by displacing MT-bound Zn [470,471]. Not being a Fenton metal, Cd cannot generate redox reactions in biological systems by itself. But because Cd interferes with SH groups, it may deplete endogenous intracellular radical scavengers, such as GSH or protein sulfhydryls (e.g., thioredoxins), inhibit the function of redox metabolizing enzymes (reviewed in [182]) but also the redox status of the cell and hence the cellular levels of redox active species. Cd has been shown to produce hydroxyl radicals in the presence of MT containing Fenton metals [472]. But Cd may also induce ROS formation by inhibiting the mitochondrial electron transfer chain [344]. Immediate early genes (c-myc and c-jun) are also up-regulated by low Cd concentrations (reviewed in [473]). However, up-regulation could be indirect and mediated by increases in cytosolic calcium or ROS signaling.

A large diversity of proteins require essential metals for their function by binding to functional groups with sulfur, nitrogen or oxygen atoms with high affinity. These metalloproteins have important cellular functions, such as metabolic, signaling or transcriptional regulation (e.g., Zn-finger and EF-hand proteins). They also play a role in the control of the cellular redox status, such as ROS metabolizing enzymes (e.g., SODs, catalases), mitochondrial electron-transport chain proteins, oxido-reductases, to name a few [474]. Cd has been shown to bind to metalloproteins with high affinity (apparent stability constants for MT and GSH ~10–25–10–14 M–1 and ~10–9 M–1, respectively [74,75]) and displace essential metals from these proteins by ligand exchange [475–477], which may result in inhibition of their function. However, Moulis [461] emphasizes that replacement of essential metals by Cd may not necessarily impact folding suggesting that inhibition of function may also be unrelated to metal replacement. This may be the case for Cd-induced inactivation of the Zn-finger tumor-suppressor protein p53 [478] where phosphorylation of p53 could be the mechanism responsible for inactivation [479]. Cd binding may shield reactive cysteine thiolates and thereby prevent formation of disulfide bridges in nascent proteins and disrupt folding, resulting in inhibition of function and/or increased degradation by cellular protein quality control systems [220,480].

Hence Cd, at free ion concentrations, which have been measured in mammalian organisms during acute or chronic exposition in vivo may not disrupt transport and homeostasis of essential metals and biological molecules directly, but rather indirectly by interfering with signal transduction pathways, by inducing ROS formation, by protein damage, and regulation of gene transcription. Nevertheless, experiments with higher Cd concentrations and shorter time periods in vitro may also be useful in their own place as a model of cumulative effects of CLCE in vivo.

5.3 Disruption of Physiological Signaling Cascades

Signaling cascades are present in all mammalian cells and are vital to physiological functions to maintain cellular and systemic homeostasis. Second messenger systems are required for the communication between different tissues and organs that are remitted via specific receptors for first messenger hormones and neurotransmitters, culminating in a cellular response, which normally entails changes in protein function and/or gene expression.

5.3.1 Cadmium and Calcium Signaling

To prevent calcium overload in the cell, which would result in cell death, Ca2+ pumps in the endoplasmic reticulum (ER) membrane (named SERCA pumps) as well as in the plasma membrane ensure that cytosolic Ca2+ is maintained at a low level (10–100 nM). Hormones, neurotransmitters and Ca2+ sensors act through G-protein coupled receptors to activate phospholipase C, which cleaves PIP2 into diacylglycerol and inositol trisphosphate (IP3) to elicit a short lived increase in the cytosolic Ca2+ concentration. Many of Ca2+’s effects are mediated through binding of Ca2+ to calmodulin and activating Ca2+-calmodulin dependent kinases (CaMKII) that are important for downstream signaling cascades. The Ca2+ ion is also important for other cellular processes, such as exocytosis, muscle contraction, and maintaining depolarization, depending on the cell type and stimulus.

The influence of Cd2+ on physiological Ca2+ signaling has been reviewed elsewhere [460] and the role of Ca2+ in Cd2+-induced cell death is detailed in Section 5.5.4.2. Many groups have reported the release of Ca2+ following Cd2+ treatment but these data have to be taken with care as Cd2+-binding fluorescent dyes, or inhibitors, such as Fura-2 or BAPTA-AM, were employed. Cd2+ can increase cytosolic Ca2+ by inactivating the SERCA pump and calmodulin-dependent Ca2+ ATPases. In addition, Cd2+ is well known to interact with calmodulin to activate CaMKII [481] as well as affect calreticulin-mediated signaling [482]. A direct interaction of Cd2+ with calmodulin is doubtful as measurements using electrospray ionization mass spectrometry demonstrated that Cd2+ had lower affinity than Ca2+, Mg2+ and Sr2+ for apo-calmodulin. However, in the presence of Ca2+ the affinities change to Ca2+ > Cd2+ > Sr2+ > Mg2+ [483]. More likely, CaMKII are activated by Cd2+-induced Ca2+ release as was demonstrated by Chen et al. in neuronal cell apoptosis [481].

5.3.2 Cadmium and cAMP Signaling

Amplification of cAMP begins with ligand binding to a G-protein coupled receptor that activates adenylyl cyclase, which in turn produces cAMP from ATP. cAMP stimulates protein kinase A (PKA), which goes on to phosphorylate target proteins, such as ion channels, or activates gene transcription. Alternatively, cAMP-dependent protein kinases are activated directly by cAMP and downstream effects depend on cell type. A review of the literature on perturbed cAMP signaling by Cd2+ was recently published [460]. Cd2+ can increase or decrease cAMP levels; the variance of effects seems to depend on the cell type. On a more unified note, Cd2+ inhibits PKA activity in both in vitro and in vivo studies.

5.3.3 Cadmium and Nitric Oxide Signaling

Nitric oxide (or nitrogen oxide) (NO) is synthesized by nitric oxide synthases (NOS) from L-arginine, oxygen and NADPH and works by activating guanylyl cyclase to amplify cGMP, which in turn activates cGMP-dependent kinases, resulting in protein phosphorylation and alteration in cellular functions. In all studies that were recently reviewed regarding Cd2+ and NO [460], NO production and NOS expression were decreased by Cd2+. More recent data support these observations [484] but also observe increases in NO or NOS expression or activity [485].

5.4 Oxidative Stress and Recruitment of Stress Signaling Pathways

The multitude of cellular signaling pathways involved in stress responses is entangled with other signaling pathways, such as those in cell death and cell survival, and cannot be separated in their entirety. The stress response protein, metallothionein, which is integral in Cd2+ detoxification has been covered in Sections 2–4 and 6 as well as in Chapters 1 and 11 of this book.

5.4.1 Oxidative Stress

Because reactive oxygen species are continuously produced from mitochondria and are highly reactive and damaging to a multitude of cellular structures, i.e., proteins and lipids, they must be quickly metabolized to non-damaging forms. Defense mechanisms include the superoxide dismutases, catalase (CAT), thioredoxins, peroxidases as well as enzymes responsible for the synthesis and reduction of the peptide glutathione [474,486]. Augmented ROS levels can lead to DNA damage (see Chapter 15) and cell death but they can also be employed as signaling molecules, which depends on the amount of ROS generated: high levels of ROS initiate cell death whereas low levels of ROS can induce cellular adaptation responses (Section 5.8).

5.4.1.1 Cadmium and Oxidative Stress

The increased formation of ROS by Cd2+ is essential to its multifaceted cellular effects and has been summarized in a number of excellent recent reviews [187,460, 487–489]. Furthermore, urine excretion of an oxidative DNA damage marker in breast-fed infants was found to positively correlate with Cd2+ concentrations in breast milk [490]. The commonly used antioxidans N-acetylcysteine binds to Cd2+, therefore, studies employing N-acetylcysteine should be taken with caution.

There are a number of mechanisms that can be affected by Cd2+ to increase ROS levels: displacement of Fenton metals [472], inhibition of the mitochondrial electron transport chain [344], decrease of antioxidant enzyme activities [182], reduction of GSH levels [491], and activation of NADPH oxidases [492]. The formation of ROS is integral to downstream signaling pathways, which are implicated in all types of stress and cell death, and could very well be one of the first responses in the cell following Cd2+ entry.

5.4.2 Endoplasmic Reticulum Stress Signaling

Endoplasmic reticulum stress develops when ER function is perturbed by accumulation of misfolded proteins, depletion of Ca2+ stores or oxidative stress in the ER lumen [493] and is sensed by three upstream signaling proteins: PERK, IRE1 and ATF6, collectively known as the unfolded protein response (UPR). They are found in the luminal ER membrane where, under ER stress, GRP78 chaperone binds to unfolded proteins allowing the ER sensors to homodimerize and activate their downstream signaling cascades. Initially, the cells try to prevent accumulation of unfolded proteins by halting transcription through the phosphorylation of PERK and eIF2α and to reduce the amount of unfolded proteins by initiating ER-associated degradation (ERAD) or autophagy to promote cell survival. But chronic or excessive ER stress can lead to activation of apoptosis mediated by the induction of CHOP, phosphorylation of JNK or activation of caspase-12 [494].

The importance of ER stress in mediating cellular life and death decisions is demonstrated by its involvement in a number of major modern diseases, including cancer, neurodegenerative and heart diseases. Some cancers have increased expression of GRP78 and GRP94, which could contribute to the ability of a cancer cell to resist proapoptotic challenges [495].

5.4.2.1 Cadmium and Endoplasmic Reticulum Stress

ER stress response genes are highly upregulated by Cd2+ treatment [496] and the pattern of genes affected appears to be distinct from other divalent heavy metals [497]. Proapoptotic CHOP is strongly upregulated by Cd2+ and this has been shown to be mediated by both the PERK and IRE1 arms of the UPR because of increases in their targets, eIF2α phosphorylation and XBP1 mRNA splicing, respectively [498–500]. Our laboratory detected phosphorylated PERK [501]. Furthermore, Cd2+ increases GRP78 indicating augmented unfolded proteins, but GRP94 has not been as reproducible. Cd2+-induced ER stress can cause phosphorylation of JNK [498], caspase activation [499] and crosstalk with mitochondria to release cytochrome c [502] to induce apoptosis.

How does Cd2+ induce ER stress? Several laboratories could attenuate ER stress and apoptosis by applying a pharmacological antioxidant or by overexpressing ROS metabolizing enzymes [487,500]. Yokouchi et al. went one step further and defined superoxide anion (O\( _2^{{ .} - } \)) as the responsible ROS for inducing ER stress. Interestingly, peroxynitrite anion (ONOO–) and hydrogen peroxide did not increase ER stress though H2O2 induced apoptosis [500]. In our recently published work, we could extend our observations to upregulation of a novel ER stress-counteracting factor, namely Bestrophin-3. Bestrophin-3 is upregulated by ER stress and appears to act by preventing CHOP upregulation as PERK and eIF2α were unaffected [501].

In summary, the UPR induced by Cd2+ can be perceived as an initial mechanism to alleviate ER stress but more often than not, CHOP is upregulated and causes apoptosis even though counteracting survival mechanisms are intact.

5.4.3 Cadmium and Mitogen Activated Protein Kinases

Mitogen activated protein kinases (MAPK) are serine/threonine-specific kinases and are activated by various stimuli to regulate a number of cellular activities. The kinases can be broadly divided into three groups: ERK, p38 MAPK and JNK. Initially it was thought that ERK mediates cell survival and p38 MAPK and JNK cause cell death, but recent data indicate otherwise.

Oxidative stress is a strong inducer of MAPK and they are therefore modulated by Cd2+, which has been summarized in the review by Thévenod [460], and JNK can be inhibited by glutathione [503]. Generally, Cd2+-activated JNK leads to apoptotic cell death, although JNK has been shown to have an antiapoptotic role too [504]. In agreement, p38 MAPK is also phosphorylated by Cd2+ in the stress response. The role of ERK in Cd2+ effects is still somewhat unclear. The literature summarized in the review by Thévenod either showed no effect or a negative effect on ERK signaling. Only two studies reported an increase in ERK [505,506]. Some recent data support a role for ERK in protection against Cd2+ toxicity and increased cell proliferation [492,507], but ERK activation by Cd2+ can also lead to cell death [506].

5.4.4 Cadmium and Nuclear Factor Kappa B

The redox sensitive transcription factor NF-κB is responsible for mediating a number of stress responses, most notably inflammation and cell survival [508]. Extracellular and intracellular stimuli transmit their signals to the NF-κB/IκB complex via IκB kinases (IKKs), which phosphorylate IκB to mark it for degradation. NF-κB can then translocate to the nucleus and activate or repress genes involved in cell stress, survival, death, and differentiation.

The effect of Cd2+ on NF-κB activity was recently reviewed [460]. In general, Cd2+ increases NF-κB transcriptional activity to upregulate multidrug resistance P-glycoprotein ABCB1 as well as genes involved in cell survival (BcL-xL), inflammation (interleukin-8), oxidative stress (metallothionein, heme oxygenase 1), and injury (ICAM-1, heat shock factor 1). With respect to cell death, opposing observations have been made in NF-κB activity. Whilst Hart et al. [509] reported increased NF-κB activity in response to oxidative stress by Cd2+, Xie and Shaikh [510] saw decreased NF-κB DNA binding and IKK activity. Furthermore, NF-κB activity is inhibited during necrotic cell death [511,512]. It is plausible that necrotic cell death was also induced in the study by Xie and Shaikh and therefore NF-κB activity was reduced. Unfortunately, necrosis was not measured. Taken together, Cd2+ increases NF-κB transcriptional activity to upregulate stress, survival and injury response genes.

5.5 Activation of Cell Death Pathways

Programmed cell death is a tightly regulated physiological process which is integral for the removal of damaged, unwanted, aged, and superfluous cells in the body and comes in various forms: apoptosis, necrosis, and autophagic cell death [513]. During apoptosis, the cell shrinks in size and volume, whilst maintaining their plasma membrane integrity whereas necrotic cell death involves swelling and rupture of the plasma membrane. Autophagy describes the breakdown of a cell’s own components, such as mitochondria or unused proteins, and has been associated with cell death. However, decisive evidence is currently lacking on whether cell death is a direct consequence of autophagy or whether these two events occur in parallel [514]. The role of autophagy in Cd2+ toxicity will be discussed in further detail in Section 5.8.

Perturbance of apoptosis execution results in a wide array of diseases such as cancer, neurodegenerative diseases, and abnormal development. The end point of the intracellular biochemical cell death process in apoptotic cells is the condensation and fragmentation of chromatin within the nucleus. The receptor-mediated pathway involves the binding of a death ligand, such as Fas, to its receptor, which activates caspase-8 and caspase-3/6 leading to cleavage of intracellular substrates and DNA condensation and/or fragmentation. The mitochondrial pathway is normally activated by other injurious compounds, such as ROS and Ca2+. The mitochondria release proapoptotic factors, which normally reside in the intracellular space, and once in the cytosol, they can either activate caspase-9 and caspase-3 to induce apoptosis, such as cytochrome c, or they can cause apoptosis in a caspase-independent manner, such as apoptosis inducing factor (AIF) and endonuclease G [515]. Cross-talk has been shown to exist between the extrinsic and intrinsic pathways [516].

5.5.1 Cadmium and Apoptosis

In early studies, apoptosis was identified as an important feature of Cd2+ toxicity in a number of organs, including the kidneys, liver, testis, and lung, using in vitro and in vivo methods [517,518]. Once taken up into the cell, Cd2+ elicits a general cellular stress response (as described in Section 5.4) that culminates in activation of apoptosis pathways beginning with mitochondria (Section 5.5.2) [460]. There is evidence for the involvement of the extrinsic pathway [519]. It is uncertain how Cd2+ might activate the receptor-mediated apoptosis pathway. Eichler and colleagues found accumulation of Fas ligand in Cd2+ exposed podocytes [519], but it must be secreted to activate the death receptor. A more plausible reason is that caspase-8 can be activated by genotoxic stress [520] which could be a result of Cd2+-induced ROS, or even by activation of the Ripoptosome complex [521] (see also Section 5.5.6).

As well as activating the classical intrinsic pathway involving activation of caspases-9 and -3, Cd2+ can also utilize caspase-independent pathways, in a context dependent manner. Our laboratory demonstrated calpain activation at early time points (3–6 hours) whereas mitochondrial damage, cytochrome c and AIF release and subsequent caspase activation was seen only after 24 hours with 10 μM Cd2+ [522]. Other studies have reported similar observations: multiple apoptosis signaling pathways can be affected by Cd2+ to induce apoptotic cell death [523,524].

5.5.2 Cadmium and Mitochondria

Generally, Cd2+ damages mitochondria. Mitochondria are the ‘powerhouses’ of a cell; they represent the metabolic and bioenergetic centers involved in a variety of functions to maintain cell sustenance including respiration, ATP production and Ca2+ homeostasis. Under normal physiological conditions, mitochondria maintain an electrochemical gradient across the inner membrane, known as the mitochondrial membrane potential (Δψm). The outer mitochondrial membrane (OMM) is generally non-selective with the voltage-dependent anion channel (VDAC) being the most abundant protein present. In contrast, permeability to ions dramatically decreases at the inner mitochondrial membrane (IMM) and is freely permeable to just a few molecules, such as O2, CO2, and NH3. It is generally accepted that mitochondrial function is altered by Cd2+; loss of Δψm, release of proapoptotic factors, swelling and inhibition of the electron transport chain have been observed in various cell lines and isolated mitochondria [344,458,525,526].

5.5.2.1 Mitochondrial Permeability Transition

The mitochondrial permeability transition describes a dramatic increase in the permeability of the IMM, which can be mediated by two mechanisms. The first involves opening of a pore, the so-called permeability transition pore (PTP), which is cyclosporin A (CsA) sensitive, to molecules <1.5 kDa, such as cytosolic solutes, which causes progressive osmotic swelling of the matrix, dissipation of Δψm, and ultimately leading to the disruption of the OMM, spilling the intermembrane contents into the cytosol [527]. The second mechanism is the formation of cytochrome c conducting “megapores” through the insertion of proapoptotic Bax into the OMM in the absence of mitochondrial swelling, which has been demonstrated in Cd2+ exposed lung fibroblasts [528].

5.5.2.2 Cadmium and the Permeability Transition Pore

The effect of Cd2+ on the PTP has been investigated in only a handful of studies. Using isolated mitochondria from liver or kidney, Cd2+ induces swelling of mitochondria, as determined by light scattering [458,525,529–531]. The role of PTP in mitochondrial swelling could be demonstrated through inhibition by CsA. Cd2+ could open the PTP by entering the mitochondrial matrix via the mitochondrial Ca2+ uniporter where it can directly bind to ANT [532] or interfere with the electron transport chain to induce ROS formation [344,530].

However, in two studies, CsA had no significant effect in blocking mitochondrial changes associated with Cd2+ [458,529], indicating that PTP opening is not always essential for MMP breakdown and apoptosis induction. So how does Cd2+ cause mitochondrial swelling in the absence of PTP opening? Though Li and colleagues still insisted that the PTP was functioning even when CsA was ineffective [529], our laboratory found that aquaporin-8 present in the IMM were activated by Cd2+ and were responsible for mediating water influx into the matrix and increasing swelling [458,533].

5.5.3 Cadmium and Bcl-2 Proteins

Whether a cell undergoes apoptosis appears to be determined by the balance of Bcl-2 family proapoptotic (e.g., Bax, Bak) and anti-apoptotic (e.g., Bcl-2, Bcl-xL) proteins [534]. Bax itself induces cytochrome c release from isolated mitochondria [535] and proapoptotic members bind to VDAC to form a large cytochrome c conducting channel [536]. In Cd2+-exposed cell lines, generally antiapoptotic Bcl-2 is decreased and/or proapoptotic Bax translocates to the mitochondria where it can release propapoptotic proteins [528,537,538] but a lack of effect of Cd2+ on Bcl-2 family proteins has also been reported [523,539]. Addition of Bcl-2 can protect cells from undergoing Cd2+-induced apoptosis [539]. Furthermore, Cd2+ exposure can increase Bcl-2 as part of a survival mechanism.

5.5.4 Other Apoptotic Signaling Pathways Involved in Cadmium Toxicity

5.5.4.1 Ceramides and Calpains

Ceramides have long been known to be intimately involved in apoptotic signaling pathways [540]. Ceramides induce cytochrome c release via ceramide channel formation [541] or activate Ca2+-dependent proteases, calpains [542]. Our laboratory were the first to show that Cd2+ increases ceramide formation, possibly via de novo synthesis, which activates calpain activity by increasing cytosolic Ca2+. Apoptosis was then executed in a caspase-independent manner [542]. Interestingly, ceramide and its metabolites are also implicated in multidrug resistance and evasion of apoptosis in tumor cells (see Section 5.8). Active calpains cleave a wide array of substrates that can lead to apoptosis as well as survival and autophagy [543]. Apoptotic substrates include caspases, Bcl-2, Bax, and AIF. In Cd2+-treated cells, calpains are activated at early time points (3–6 hours) and provide a point for crosstalk with caspases, which are activated later (24 hours) [522] and can also cleave Bax to release cytochrome c [528].

5.5.4.2 Calcium

Ca2+ release from the ER triggers the UPR (Section 5.4.2), and ultimately leads to cell death. Conclusive evidence for Ca2+ release induced by Cd2+ has been hampered by the lack of adequate tools since Cd2+ can also bind to EGTA-based fluorescent indicators and chelators, such as Fura-2 and BAPTA, due to the very similar ionic radii and often elicits stronger fluorescent intensity than Ca2+ itself [544]. Recently, aequorin, a photoprotein from luminescent jellyfish, has been tested as an appropriate Ca2+ indicator with little interference from Cd2+. Similar to an initial cell-free study by Izutsu and colleagues [545], Biagioli et al. found that aequorin did not respond to 5 μM Cd2+ but increased slightly with 15 μM Cd2+ in digitonin-permeabilized cells [546]. However, whilst Ca2+ is normally present in the cytosol in the nanomolar range, aequorin can only detect increases in Ca2+ at 0.5–10 μM. To circumvent this problem, the effect of Cd2+ on bradykinin-induced Ca2+-release was investigated. Cd2+ treatment resulted in reduced Ca2+ release by bradykinin indicating that Ca2+ homeostasis had been disturbed, which was attributed to inhibition of the SERCA in the ER by Cd2+, and lead to activation of apoptosis [499].

More recently, no interference of up to 2 mM Cd2+ with a protein-based Ca2+ sensor yellow chameleon, YC3.60 was observed. The authors concluded from their study that Ca2+ is not necessary for Cd2+ (1–30 μM)-induced transcription and that as cells succumb to metal toxicity, Ca2+ is released [547]. Further studies would be required to confirm these observations and to finally resolve the role of Ca2+ in Cd2+-induced stress and cell death signaling pathways.

5.5.5 Cadmium and Necrosis

Recent studies indicate that necrosis is indeed a form of programmed cell death, rather than a passive process of cell swelling and rupture, termed “necroptosis”, which uses many of the tools available for apoptosis induction [548].

Cd2+ is a well known necrosis inducer, especially at higher concentrations and/or in sensitive cell lines. In vivo necrosis by Cd2+ has been documented in the kidney, heart, liver, and testis. A wider variety of cell types have been employed for in vitro testing and have similarly shown both apoptosis and necrosis simultaneously or necrosis alone [359]. Using pharmacological inhibitors, Yang et al. could demonstrate a role for Ca2+, calpains, mitochondrial membrane potential, ROS formation and NF-κB in Cd2+ necrosis of CHO cells. The authors suggested that cytosolic Ca2+ overload might be important in the execution phase of necrotic cell death while sustained depletion of Ca2+ stores in ER leads to apoptosis [511].

In a follow up study, necrostatin-1 is shown to be active in blocking Cd2+-induced necrosis by inhibiting Δψm loss. Interestingly, necrostatin-1 + Cd2+ increased Ca2+ overload further over Cd2+ alone cells without a concurrent increase in calpain activity. In fact, necrostatin-1 could attenuate Cd2+ induced calpain activity [512]. In line with their previous observations, necrostatin-1 restored Cd2+-induced decrease of NF-κB activity, indicating that this transcription factor is integral to the necrotic response.

5.5.6 The Apoptosis-Necrosis Switch

Inhibition of cell death signaling pathways can initiate a cell to “re-wire” its cell death program from apoptosis to necrosis [549] through a molecular switch. Earlier studies put forward ATP [550] or caspases [551] as the mediator. Depletion of ATP would favor necrosis, which can be explained by energy demands during apoptosis execution whereas general caspase inhibitors may affect mitochondrial functions that constitute a deciding event in both cell death types. The type of ROS and redox status has been suggested as an apoptosis-necrosis mediator. H2O2 induces necrosis due to reduction of GSH levels but other ROS forms, such as superoxide anion, induce apoptosis [552]. However, this has been contested by another study where inhibiton of superoxide anion producing NADPH oxidases suppresses the switch from apoptosis to necrosis [553]. The latest data report the Ripoptosome as a signaling platform that governs the form of cell death that will be executed [521].

Necrosis occurrence is dependent on the concentration of Cd2+ applied and the levels of ATP appear to be very important. The scenario for cell death switch in Cd2+-treated cells is somewhat more complicated but ATP, GSH status and peroxide accumulation are all involved [552]. In addition, metallothionein-3 (MT-3) seems to have an important role in controlling the form of cell death. In kidney cells with low MT-3 expression, Cd2+ causes apoptosis but MT-3 overexpressing cells show necrosis by Cd2+ [554]. The mechanism by which MT-3 predisposes cells to necrotic cell death was not investigated but previous studies report a non-canonical neuronal cell growth inhibitory activity of MT-3, which may be related to its necrosis inducing abilities.

5.6 Reprogramming of Developmental Signaling Pathways

Aberrant signaling during the gestation period can lead to a multitude of developmental defects, such as stunted or absence of limbs and loss of organ function. A number of signaling pathways are crucial for development, namely, Wnt/β-catenin, Hedgehog, Notch, Hippo, and the transforming growth factor beta family (partly reviewed in [555,556]). The activity of these signaling pathways goes beyond development, remaining active in adults, where they have been implicated in a number of diseases, in particular, cancer. At the time of writing, a literature search on Notch or Hippo signaling and Cd2+ retrieved no articles.

5.6.1 Cadmium and Wnt/β-Catenin

Wnt is the mammalian form of the Drosophila gene wingless, which is mutated in flies without wings, and is secreted. An intracellular response is triggered through binding of Wnt to its receptor, Frizzled, and can be signaled through canonical and non-canonical Wnt pathways [557]. The canonical pathway involves the multifunctional protein β-catenin and is responsible for cell transformation. Cytosolic levels of β-catenin are kept low by targeting it to the proteasome for degradation through a destruction complex. Upon activation, the destruction complex is disassembled and β-catenin translocates to the nucleus where it serves as a transcriptional co-factor for the TCF/LEF transcription factor family. Target genes of TCF transcription include cyclin D1, c-myc and ABCB1, which are genes responsible for cell proliferation and survival. In another role distinct from Wnt signaling, β-catenin is found in cell borders where it forms part of the adherens junctions along with E-cadherin and α-catenin. The disruption of the adherens junctions has been associated with epithelial-to-mesenchymal transition (EMT), a phenomenon linked to malignant cell transformation and metastasis, as cells detach themselves from the extracellular matrix through loss of cell adhesion and increase their motility.

In development, Cd2+ exposure and interference with the Wnt signaling pathway can lead to birth defects (reviewed in [460]). The group of Thompson has demonstrated that the non-canonical Wnt pathway is engaged by Cd2+ leading to ventral body wall defects in a chick embryonic model [482]. In further studies, the adherens junctions was disrupted either through redistribution of β-catenin to cytosol and nucleus and actin disorganisation [558] or decreased calreticulin, E-cadherin and β-catenin mRNAs [482] in the same model. Elsewhere aberrant thymus development in embryos of Cd2+ treated pregnant mice was a result of decreased signaling activity of Sonic Hedgehog and β-catenin [559].

In addition to developmental defects, Cd2+ has been shown to target the Wnt signaling pathway as a means of exerting its carcinogenic effects. From our laboratory, we have reported translocation of β-catenin from the periphery to the nucleus and upregulation of c-myc, cyclin D1, and ABCB1 via TCF4 activation in subconfluent cells [560]. Moreover, chronically Cd2+-exposed mice exhibited significant increases in mRNA in a number of Wnts, Frizzled receptors and Lrp co-receptors [561]. Increased EMT markers (collagen I, Twist, and fibronectin, but not α-smooth muscle actin, mRNAs) were also observed [561]. Both Twist and fibronectin are under the control of β-catenin mediated transcription, thus providing the crucial link that Cd2+ supports EMT and therefore tumor progression [562].

5.6.2 Cadmium and Hedgehog

The Hedgehog family of proteins function as intercellular messengers and are secreted to regulate morphogenesis of organs and tissues and control stem cell proliferation in adults [563]. The effect of Cd2+ on developmental defects as a result of aberrant Hedgehog signaling has been reviewed elsewhere [460]. Collectively, downregulation of the Sonic Hedgehog (Shh) isoform was observed in Cd2+-treated pregnant mice that correlated with disrupted embryonic development in mice and in zebrafish. A recent study also reported decreased Shh in the thymus of Cd2+-treated pregnant mice that could be responsible for the change in thymocyte phenocyte [559]. Interestingly, the offspring demonstrated increased expression of Hedgehog signaling components indicating a mechanism of adaptation in response to depleted Shh in the embryonic environment. This may have implications in developmental malfunctions as well as tumor progression potential.

5.7 Genetic and Epigenetic Effects

A mutagen is defined as an agent that alters genetic material, usually DNA, increasing the frequency and likelihood of mutations. As a weak mutagen, Cd2+ does not directly affect the DNA structure, which is in contrast to other toxic and carcinogenic metals, such as chromium which forms DNA adducts. However, Cd2+ exposure results in changes in gene expression that ultimately influence cell characteristics and behavior.

5.7.1 Cadmium, DNA Damage, and Inhibition of DNA Repair

DNA damage by Cd2+ is linked to DNA damage by ROS and inhibition of DNA repair mechanisms. ROS, in particular hydroxyl radicals, react with the DNA bases giving rise to altered function or DNA damage [564]. Cd2+ can block repair mechanisms for DNA, such as mismatch and base excision repair [565,566], which could be attributed to Zn2+ displacement in zinc finger DNA repair proteins [567].

It has been shown that ROS induced DNA damage occurs at low levels under physiological conditions; probably as a result of mitochondrial ATP production where ROS are produced. In combination with the inhibition of DNA repair, ROS-induced DNA damage changes the genetic makeup and could contribute to promotion of carcinogenic cells, which acquire a number of mutations to achieve the hallmarks of cancer. For further details, please refer to Chapter 15.

5.7.2 Cadmium and Epigenetics

Epigenetics is the study of inheritable changes in gene expression without any modifications in the DNA sequence and is strongly elicited by environmental factors [568]. Failure to maintain epigenetic information leads to incorrect gene expression, which can be passed onto daughter cells or offspring, and apoptosis. The most studied epigenetic changes are a result of either aberrant DNA methylation or histone modifications. Cytosine nucleotides can be methylated by DNA methyltransferases (DNMT) to 5-methylcytosine. Excessive methylation, e.g. from methyl groups found in the diet, can switch off a gene through blockade of transcription factor access. Following synthesis, the DNA is wound around histone proteins to aid compaction and can influence the accessibility and thus expression of a gene. A gene is only active if it is accessible. Modification of the histone proteins may affect how the DNA is wound and therefore the accessibility of a gene.

Heavy metals, including Cd2+, may elicit their toxic and carcinogenic effects by epigenetic mechanisms. It seems that DNA hypermethylation and concomitant increase in DNMT expression is associated with malignant transformation of cells exposed to Cd2+ [569,570] or cells that develop Cd2+ resistance [571]. In contrast, cells with increased proliferation exhibited both hypermethylation [570] and hypomethylation [572], which may be dependent on the concentration and exposure time of Cd2+. Histone modification by Cd2+ has only been demonstrated in one report by Somji et al., where it affects MT-3 expression [573].

5.7.3 Cadmium and MicroRNA

MicroRNAs are short RNA sequences (21–25 nucleotides) that control gene expression at the post-transcriptional level. MicroRNAs bind to complementary messenger RNA sequences to prevent their translation and are part of a RNA-induced silencing complex. Binding to a complementary sequence causes degradation of the mRNA or inhibition of translation, resulting in gene silencing [574]. Cd2+ has been associated with modification of microRNA expression [575] but an in-depth study in a mammalian system is currently not available.

5.8 Mechanisms of Adaptation, Survival, and Carcinogenesis

5.8.1 Adaptation and Survival Mechanisms Induced by Cadmium

Cellular stress signals induced by Cd2+ forces the cell to engage counteracting adaptation and survival mechanisms in an attempt to prevent itself from dying. These mechanisms can tackle the cellular stress response in a variety of ways: removal of damaged or death-inducing proteins through degradation by autophagy or the proteasome, upregulation of pro-survival genes, activation of survival signaling pathways, and inhibition of stress and cell death signaling cascades.

5.8.1.1 Cadmium and Autophagy

Autophagy is a process of self-eating. By degrading intracellular components, damaged proteins and organelles, the cell can increase its chances of survival during stress or nutrient starvation conditions [576]. Membrane-bound vesicles called autophagosomes form to “engulf” cytosolic contents, which subsequently fuse with lysosomes to form autophagolysosomes that degrade the engulfed contents. Autophagy can be detected by (i) the conversion of LC3 from LC3-I to LC3-II or the redistribution of LC3 from cytosol to puncta; (ii) uptake of monodansylcadaverine (MDC) or acridine orange into autophagolysosomes: (iii) changes in autophagy-specific proteins, e.g., increase in Beclin-1. Persistence of autophagy can lead to cell death although cell death is not always a direct consequence of autophagy. In fact, the existence of autophagic cell death has been contested [514].

Cd2+-induced autophagy is mediated via ROS formation [577,578], GSK3β [578], AMPK [577], p38 [579] or ERK [580] activation, decreases in mTOR, PARP, and ATP [577] and Ca2+ signaling [580], though Cd2+-binding BAPTA was employed (see Section 5.5.4.2). Autophagy by Cd2+ appears to suppress apoptosis execution because changes in autophagy markers were observed in the absence of apoptosis in low Cd2+ (< 10 μM) treated rat kidneys in vivo after 3–5 days [412], endothelial cells [581], hematopoietic stem cells after 48 hours [582] and kidney proximal tubule cells after 1 h (Lee and Thévenod, unpublished data). This has led to the proposal of autophagy as an early biomarker of Cd2+ toxicity [412] (see Section 4.4). Though Cd2+ exposure eventually leads to apoptotic cell death, there is not enough evidence to conclude that it is a direct consequence of autophagy induction.

5.8.1.2 Cadmium and Nrf2-Keap1 Signaling

Nrf2-Keap1 is brought into play by oxidative stress leading to long term adaptation of the cell and enhanced survival. Upon oxidative or electrophilic stress, Nrf2 escapes its suppressor Keap1, which is inactivated by direct modification of its cysteine thiol residues, and translocates to the nucleus to bind to the antioxidant response element. Target genes of Nrf2 include antioxidant defense proteins and detoxifying enzymes [583].

Environmental insults are potent inducers of the Nrf2-Keap1 signaling pathway. In agreement, Cd2+ activates Nrf2 nuclear translocation to prevent apoptosis [584]. Though Cd2+ may activate Nrf2 through the induction of ROS, it has been demonstrated that Cd2+ can increase the stability of Nrf2 [585], possibly through direct binding of Cd2+ to the cysteine residues on Keap1.

5.8.1.3 Cadmium and Akt-PI3K Signaling

The Akt-PI3K signaling pathway is downstream of oncogenic Ras. Phosphoinositide-3-kinase (PI3K) phosphorylates PIP2 to PIP3, which binds Akt. Once correctly positioned at the membrane via binding of PIP3, Akt is activated by kinases. Active Akt then promotes cell survival primarily by counteracting proapoptotic processes, for example, antagonizing proapoptotic Bad.

Akt can also stimulate cell growth and proliferation by activating mTOR to increase protein synthesis and inhibiting GSK3β to increase β-catenin-induced transcription of proliferation genes. Cd2+ generally activates Akt signaling in various cell types, concentrations and exposure times [460,586,587], not only as a stress response, but as an adaptive reaction to prevent cell death.

5.8.2 Cadmium and Carcinogenesis

As a class I carcinogen, Cd2+ causes cancer of the lung, kidney, and the prostate in humans [168]. Chronic exposure to Cd2+ gives rise to apoptosis-resistant cells that are potentially carcinogenic [214,588] and low Cd2+ exposures induce transcription of proto-oncogenes [589]. This is partly due to oxidative DNA damage [185]. Through the induction of adaptation and survival mechanisms, cells acquire tumor cell characteristics that are known collectively as the hallmarks of cancer; they include sustaining proliferative signaling, evading growth suppressors, resisting cell death, enabling replicative immortality, inducing angiogenesis, and activating invasion and metastasis [590] that are all part of the multistep process in tumor development.

Important signaling pathways in cancer include, but are not limited to p53, cell cycle regulation, Wnt/β-catenin (see Section 5.6.1), EMT, proto-oncogenic immediate early genes, Akt-PI3K as well as the stress response pathways discussed in Section 5.4. For an in-depth review of cellular and molecular mechanisms leading to carcinogenesis by Cd2+, the reader is referred to Chapter 15 of this book as well as to recent reviews available [184,591].

5.8.2.1 Cadmium, p53, and Cell Cycle Regulation

The tumor suppressor gene, p53, is the master regulator of cell death and acts as a cell cycle guardian [592]. p53 is normally marked for degradation by the proteasome through the ubiquitylating action of Mdm2. Phosphorylation of p53 disrupts the binding of Mdm2 and p53 can elicit its effects. The tumor suppressing functions of p53 are: to initiate apoptosis of irreparable cells, arrest damaged cells at cell cycle check points, activate DNA repair proteins, and inhibit angiogenesis. If the damage cannot be repaired, the cells are removed by apoptotic cell death. Furthermore, p53 has a DNA binding domain and acts as a transcription factor to increase the expressions of proteins that are integral to its functions, such as growth arresting p21 and apoptosis-inducing Bax. In circa 50% of cancers, p53 is mutated allowing cancer cells to grow limitlessly harboring their mutations.

The double edged sword effect of Cd2+, namely apoptosis and cancer progression, is reflected in its opposing effects on p53. Cd2+ inhibits DNA base excision repair in p53-dependent and -independent manners [593], increases p53 phosphorylation as a prelude to apoptosis [594,595], and arrests cells in G1 and G2/M checkpoints via p53-dependent [510] and -independent [596] mechanisms. Cd2+ can affect p53 in two ways: p53 is activated by phosphorylation via phosphatidylinositol-3-kinase related kinases [479] or Cd2+ replaces zinc to maintain its mutated non-DNA binding form preventing apoptosis induction [597].

5.8.2.2 Cadmium and Multidrug Resistance P-Glycoprotein (MDR1/ABCB1)

ABCB1 confers cellular resistance to apoptosis inducers and is upregulated in many cancers [598]. Cd2+ increases ABCB1 expression and long-term Cd2+ exposure results in lowered apoptosis occurrence [588,599]. Initial studies supported the hypothesis that Cd2+ is transported out of the cell by ABCB1 to reduce apoptosis [600]. In contrast, our laboratory found that Cd2+ augmented ceramides, which were in turn effluxed by ABCB1 to prevent cell death [601]. This discrepancy can only be explained by different experimental conditions.

5.9 Metallohormone Effects

A metallohormone is defined as an inorganic metal ion which can mimic hormone-like activities and effects. Generally, the receptors are localized in the cytosol and upon hormone binding, they translocate to the nucleus where they exert their effects as transcription factors and modulate gene expression through DNA binding to the promoter region of target genes. The DNA binding is accomplished by zinc fingers forming a Zn2+-[Cys]4 metal-protein complex with cysteine residues.

The primary and most important androgen is testosterone. In the adult male, its active metabolite dihydrotestosterone supports spermatogenesis for reproduction as well as maintaining muscle mass and fat distribution. Estrogens are associated with development of female secondary sex characteristics and comprise of estrone, estradiol, and estriol. Steroid hormones are not exclusively gender specific and have important roles in both males and females.

5.9.1 Cadmium as a Metalloestrogen

Endocrine status is acknowledged as a prominent risk factor for breast cancer, more so than family history, and led to the focus on environmental estrogens as potential sources that contribute to a higher risk of breast cancer development. Cd2+ has been widely studied as a metalloestrogen, summarized in recent excellent reviews [602,603].

Several studies have proven that Cd2+ increases MCF-7 proliferation [604], a human breast cancer cell line that is dependent on estrogens for growth, potentially through direct binding to the ligand binding domain of the estrogen receptor [605], and can be abolished by the anti-estrogen ICI-164,384. Metalloestrogenic effects are not restricted to Cd2+ and increased cell proliferation has been observed with a number of other inorganic metal ions [606], but they do not seem to work in the same way. Only Zn2+, Cd2+, Hg2+, and Co2+ could preserve DNA binding of the estrogen receptor-α [607]. In other studies, Cd2+ did not bind to or activate the estrogen receptor or did not induce cell proliferation when measured by a recombinant yeast estrogen screen or protein labelling with sulforhodamine-B [608].

The groundbreaking report of Johnson et al. was the first study of Cd2+-induced estrogenic effects in whole animals [609]. The authors reported classical changes associated with increased estrogenic activity: increased uterine wet weight, increased height of the uterine epithelium, early onset of puberty, increased cell proliferation in the endometrium, increased mammary gland density and augmented expression of estrogen receptor driven genes in ovariectomized rats. Subsequent studies showed a partial estrogenic response [610,611]. The authors concluded that other intracellular signaling pathways are involved, namely ERK1/2.

In summary, the case of Cd2+ as a metalloestrogen is a yet unclear one. Whilst the in vitro data in mammalian cell lines are comprehensive and convincing, the in vivo data are somewhat inconsistent. Fechner and colleagues hypothesize that different cellular pools of estrogen receptor-α exist and may respond to Cd2+ differently based on the state of the cysteine residues in the ligand binding domain [612].

5.9.2 Cadmium and Androgenic Effects

Endocrine status is also a risk factor for the progression of prostate cancer and several epidemiological studies, though not all, indicate a connection to Cd2+ [613]. Alongside estrogenic effects, Cd2+ also activates the androgen receptor and increases cell proliferation, gene expression [614] as well as testosterone production. Employing a reporter gene assay, Cd2+ activated the androgen response element in human prostate epithelial and liver cells [614] but, contrary to other studies, there was no increase in proliferation. Furthermore, malignant transformation of normal prostate cells through repeated exposure to Cd2+ has been documented [615], increasing the prominence of Cd2+ as a risk factor in prostate cancer.

6 Endogenous Detoxification

The body has several means to detoxify Cd and also alleviate its toxic effects. For instance, Cd can be complexed by MTs and thereby detoxified. Several excellent reviews on MTs have been recently published [55,237,616]. MTs are low-molecular weight (MW ranging from 3.5-14 kDa), cysteine-rich metal-binding proteins. MTs have the capacity to bind both, physiological Zn2+ and toxic Cd2+, through the thiol group of its cysteine residues, which represents nearly 30% of its amino acidic residues. All cysteines are known to participate in the coordination of 7 mol of Cd or Zn per mol of MT [617]. Of the four common MTs, MT-1 and MT-2 are expressed in most tissues, MT-3 is predominantly present in brain, whereas MT-4 is restricted to certain epithelia (see also Chapter 11). MT genes are readily induced by various physiologic and toxicologic stimuli because the promoter region of MTs contains several response elements, including glucocorticoid, antioxidant, and – most importantly – metal responsive elements. The candidate metal responsive element-binding protein (MTF-1 for metal transcription factor-I) is a Zn-finger (Cys2His2) transcription factor [618].

Because the cysteines in MTs are absolutely conserved across species, it is assumed that these cysteines are necessary for function and MTs are essential for life. Their major physiological function appears to be homeostasis of essential metals, mainly Zn and Cu, and redox metabolism, which are coupled processes [616,619]. The experimentally determined apparent stability constants for MT binding to Zn are in the range of ~10–25–10–14 M–1 [74]. Their thiolate coordination environments make MTs redox-active Zn proteins that exist in different molecular states depending on the availability of cellular zinc and the redox status [616,619]: Oxidative conditions make more Zn available, while reductive conditions make less Zn available. MTs move from the cytosol to cellular compartments, are secreted or internalized by cells. MTs have been localized largely in the cell cytoplasm, but also in lysosomes, mitochondria and nuclei (reviewed in [237]).

MTs protect against Cd toxicity because the affinity of Cd to MT is about 104 = 10,000 times higher than that of Zn [74]. But, as the toxicity of Cd is also partly due to binding to charged sites of target proteins or the displacement of Zn bound to Zn metalloproteins, ZnMTs may restore structures and functions of such proteins through removal of bound Cd or through a reciprocal Cd/Zn transfer reaction [461,620]. This would confer on MT an active role in the protective response to Cd toxicity, rather than a passive one that is solely dependent on the high affinity for binding free metal ions. Mice which lack MT genes were more susceptible to renal [621], bone [622], and liver injury [245] mediated by CLCE than mice expressing MT. Therefore, MT appears to be crucial to detoxify Cd in the body. Moreover, induction of MT respective MT-transgenic mice show increased resistance against chronic Cd toxicity and lethality [623] (reviewed in [624]). Hence, MT-null mice are hypersensitive and MT-transgenic mice resistant to Cd toxicity. The other side of the coin is that MT mainly contributes to the long biological half-life of Cd in the body [625] and thereby increases the likelihood of long-term organ damage associated with chronic Cd accumulation. In humans, there are large individual variations in MT expression, possibly due to polymorphisms [626]. Indeed, polymorphisms in the human MT-2A gene can limit MT expression [627] and increase the susceptibility to organ Cd toxicity [628,629]. These differences in MT expression may be responsible for inter-individual differential predispositions to Cd toxicity [630].

Cd may also be detoxified by chelation with intracellular GSH (apparent stability constant for GSH ~10–9 M–1, [75]) and excretion through the bile, urine or pulmonary fluids. Hepatic secretion of Cd-GSH complexes has been shown to be mediated by ABCC2/MRP2 (see Sections 3.3 and 4.3.2). Moreover, L’Hoste et al. [491] have provided evidence in cultured mouse PT cells that Cd-GSH complexes are secreted via ABCC7/CFTR. A similar mechanism could be operative in pulmonary epithelia [269], where CFTR, but not MRP2 or BCRP/ABCG2, has been shown to play a crucial role for maintaining basal epithelial lining fluid GSH and increasing epithelial lining fluid GSH in response to cigarette smoke in vitro and in vivo [631].

Finally, metabolism and inactivation of Cd-induced ROS by cells and tissues can also be perceived as indirect processes that contribute to Cd detoxification. The cellular antioxidative defense mechanisms responsible for ROS scavenging have been thoroughly reviewed [474,489]. They include antioxidative enzymes (e.g., superoxide dismutases, catalases, glutathione peroxidases, antioxidative metabolites (GSH, ascorbic acid, vitamin E) and the enzymes involved in their regeneration (thioredioxins, glutaredoxins, glutathione reductases).

7 Concluding Remarks and Future Directions

Apart from summarizing current knowledge of the field, this review aimed to deliver an appraisal of missing long-term epidemiological studies that will allow narrowing down and identifying health hazards caused by CLCE in order to design landmark studies that will clarify the molecular mechanisms by which organs and cells accumulate low Cd concentrations and how CLCE causes disease. Due to space limitations, many relevant studies on the subject could not be included in this review.

Current understanding of the biological effects of Cd and the existing knowledge of Cd-induced diseases are mainly based on results obtained by exposure to high doses of the toxic metal. But there is accumulating evidence for adverse health effects even under low exposure conditions. CLCE appears to be associated with an increasing spectrum of health hazards. In addition, susceptibilities of specific populations (e.g., children, elderly), genetic variation, and cumulative risks that result from deficiencies of nutrients, such as Fe and Zn, have not been considered.

Hence, the aims for the future will be to characterize further the impact of CLCE on major health issues by prospective long-term population studies to evaluate and strengthen the evidence for the links of CLCE with widespread diseases and leading causes of death in modern societies, e.g., osteoporosis, diabetes, cardiovascular diseases, and cancer. Implementation of preventive measures will also call for further follow-up studies to demonstrate their efficacy. Cellular and molecular research will need to establish experimental models and valid hypotheses to test causal relationships between CLCE and the aforementioned disease entities. Another challenge will be to identify likely candidates for Cd uptake into cells at the low extracellular free Cd concentrations measured in a setting of CLCE in order to design strategies for prevention or therapy of Cd toxicity. Finally, the initial processes underlying Cd-induced cell death, survival and cancer signaling need to be delineated in more detail. Identifying the signals at the top of the chain of command will help to eradicate the noxious effects of Cd at their roots and ultimately contribute to the development of preventive and novel therapeutic strategies for acute and chronic Cd toxicity.