Clements: “3357_c007” — 2007/11/9 — 12:42 — page 95 — #1 7 Bioaccumulation Bioavailability is a widely accepted concept based on the implicit knowledge that before an organism may accumulate or show a biological response to a chemical, that element or compound must be available to the organism. While the concept of bioavailability is widely accepted, the processes that control it are poorly understood. (Benson et al. 1994) 7.1 OVERVIEW With the exception of radionuclides, a toxicant must first come into contact with the initial site of action before an effect can manifest. This might, in some cases, involve a straightforward contact with a biological surface where a localized effect occurs. Even in this case, the toxicant will penetrate to some extent into cells. In many other cases, the toxicant moves from the surface of first contact into the organism where it interacts elsewhere with the site(s) of action. Once within the organism, many processes modify, redistribute, or remove the toxicant. Bioaccumulation is the net result of these uptake, transformation, translocation, and elimination mechanisms. Toxicant qualities influencing bioaccumulation-associated processes will be described here. 7.2 UPTAKE 7.2.1 C ELLULAR MECHANISMS Atoxicant present in an external medium or one already inside the organism can come in contact with and then be moved acrossacellmembrane. External media might be gaseous (e.g., inhaledair), liquid (e.g., inhaled or imbibed water), or solid (e.g., ingested food). A toxicant already in the organism, perhaps moving within the circulatory system, can be present in its original form, complexed, or transformed to some metabolite(s) or conjugate(s). Regardless, the same general mechanisms are involved in the toxicant transport into and out of cells. Before discussing cellular transport mechanisms, it is important to mention that some substances might not pass into cells, but instead enter by moving through the tight junctions between cells. Solvent drag of a substance into a fish through gill cell junctions (see Evans et al. 1999) is one example of such movement by this paracellular pathway. Gill cell junction permeability can increase substantially under conditions that interfere with calcium’s normal role of maintaining tight seals between adjacent cells (e.g., low water pH) (Booth et al. 1988, Cuthbert and Maetz 1972, Newman and Jagoe1994). Anotherexample of paracellulartransport occurs inthe lugworm, Arenicola marina, whose survival depends on coping with poisonous sulfide in its environment. Sulfide emanating from anoxic sediments enters the worm but is quickly oxidized to the less toxic thiosulfate. Thiosulfate then diffuses out through the body wall of the lugworm primarily via cell junctions (Hauschild et al. 1999). Correctly so, most treatments of toxicant movement into or out of cells focus on passage across cell membranes. The cell membrane is a phospholipid bilayer with proteins interspersed between and extending into the phospholipid layers. According to Singer and Nicholson’s classic fluid mosaic membrane model (Singer and Nicholson 1972), lipids float within one layer of the membrane to 95 © 2008 by Taylor & Francis Group, LLC Clements: “3357_c007” — 2007/11/9 — 12:42 — page 96 — #2 96 Ecotoxicology: A Comprehensive Treatment associate in different ways but do not often move from one phospholipid layer to the other (Simkiss 1996). Not only do consequent differences exist in lipid characteristics between the outer and inner lipid layers, patches or macrodomains of cell surface membrane lipids and proteins form within a layer (Gheber and Edidin 1999). The lateral transport of membrane components is influenced by several factors that impart a dis- tinctively dynamic and heterogeneous nature to membranes (Jacobson et al. 1995). A membrane component can have its movement restricted by another cluster of membrane components, or it can move about in a random or directed manner. Connection of a membrane component to the cell cyto- skeleton often directsmovement. As describedby the raft hypothesis, resulting lateral heterogeneities in cell surface components produce functional heterogeneity on the membrane surface (Edidin 2001, Mayor and Rao 2004). The plasma membrane presents an intriguing mix of dynamic activities in which components may randomly diffuse, be confined transiently to small domains, or experience highly directed movements. (Jacobson et al. 1995) Given the complex and dynamic nature of cell membranes, it should be no surprise that chemicals pass to and fro across cell membranes in many ways (Figure 7.1). Some nonionized, lipid-soluble chemicals diffuse passively through the lipid bilayer. This diffusion that Simkiss (1996) calls the “lipid route” forms the basis for the pH-Partition Theory discussed in Section 7.2.3. Other chemicals move by passive diffusion (filtration) through ion channels or pores. Many hydrophilic molecules smaller than 100 Da enter cells this way, although exceptions include ions with large hydration spheres that restrict movement through channels (Timbrell 2000). 1 Gated ion channels also allow passive diffusion, but how they function depends on chemical and electrical conditions. Diffusion facilitated by a carrier molecule is faster than predicted for simple diffusion, although it also does not require the expenditure of energy. Diffusion can involve two ions synchronously exchanged between Diffusion Ion channel Gated ion channel Across lipid bilayer Facilitated diffusion Carrier molecule Active transport Endocytosis (phagocytosis and pinocytosis) Ion pump Paracellular transport FIGURE 7.1 Diagram of routes of chemical uptake into cells and the paracellular route. 1 For simple metal ions, hydration sphere size generally increases with increasing ion charge or decreasing ion size. © 2008 by Taylor & Francis Group, LLC Clements: “3357_c007” — 2007/11/9 — 12:42 — page 97 — #3 Bioaccumulation 97 the outside and inside of the membrane. As with the ATPase-mediated exchange discussed in the last chapter, the energy-requiring active transport of a chemical can occur up an electrochemical gradient. Finally, endocytosis can be an important avenue for moving chemicals into and out of the cell. As an example, iron can be assimilated by binding to a membrane-associated transferrin protein, with subsequent movement of the iron–transferrin complex to a membrane “coated pit” region, and incorporation into a vesicle that then passes into the cell (Simkiss 1996). At this point, some specific examples of cellular transport mechanisms might foster an appreci- ation for the diversity of avenues by which toxicants are taken up, transported within, and eliminated from cells. A few important ones are provided here with brief mention of detoxification mechan- isms that will be described again later in this chapter. In reading these examples, it is important to understand that mechanisms often work in concert to facilitate uptake, transformation, and elimina- tion. This point of the body’s simultaneous use of several mechanisms to regulate internal chemical concentrations can be illustrated with a straightforward example peculiar to elasmobranchs that, unlike other fishes, retain urea for osmoregulatory purposes. Urea accumulates in the elasmobranch, Squalus acanthias, as a result of two important mechanisms (Fines et al. 2001). First, elasmobranch gill cells have a protein transport mechanism that moves urea in a direction (inward) opposite to that seen in gills of most other fishes. Second, the gill phospholipid bilayer is modified so that it is less soluble to urea than those of other fishes. A toxicological example of mechanisms working in concert is the movement and resulting toxicity of arsenic species. Low phosphate concentrations result in accelerated As(V) uptake involving a shared energy-requiring uptake mechanism. As(V) is reduced to As(III) by arsenate reductase inside the cell and then removed from the cell by an ATPase-dependent pump (Huang and Lee 1996). Removal would be very much slower if As(V) was not first converted to As(III). It should be clear from these examples that understanding of cellular movement of chemicals requires consideration of interactions among mechanisms and processes. Active transport is important for many different toxicants or their metabolites.Arelevant example is gut epithelial cell metabolism and membrane movement of benzo[a]pyrene. After entering a gut epithelial cell, benzo[a]pyrene is metabolized by Phase I reactions (CYP1A1 and CYP1B1) and then conjugated. The resulting benzo[a]pyrene-3-sulfate and benzo[a]pyrene-1-sulfate are actively transported back toward the gut lumen by ATP-binding cassette (ABC) transport proteins (Buesen et al. 2002). 2 Another example of active transport is the previously mentioned transport of metals by the gill ATPases. The basolateral membrane of gill cells, which have high Na + /K + -ATPase activity (Evans et al. 1999), is the site ofATPase-mediated transport of many metals such as silver in rainbow trout (Oncorhynchus mykiss) gills (Bury et al. 1999). Many toxicants and natural metabolic products are organic anions. Important examples include chlorinated haloalkenes such as the solvent trichloroethylene and chlorinated phenoxyacetic acid herbicides such as 2,4-D(2,4-dichlorophenoxyacetic acid) (Sweet 2005). Some mercury complexes and many conjugated toxicants or their metabolites are also organic anions. Therefore, it should be no surprise that a family of organic anion transporters (OATs) exists to move organic anions into and out of cells. As one important example, OATs are involved in active transport across the renal proximal tubules (Sweet 2005). Ionic mercury conjugated with cysteine, N-acetylcysteine, and glutathione is transported by OATs of rabbit renal proximal tubule cells (Zalups and Barfuss 2002). Reduced glutathione that is involved in detoxification of some poisons and in combating oxidative stress can also be moved across cell membranes via OATs. Toxicants present as organic anions are subject to renal elimination via this mechanism, but kidney damage can occur if a toxicant was accumulated by OATs to very high concentrations in the associated cells. 2 The ATP-binding ABC transporters are members of a very large family of membrane transporters that move a diversity of chemicals including phospholipids, peptides, steroids, polysaccharides, amino acids, nucleotides, organic anions, drugs, toxicants, xenobiotics, and their conjugates (Hoffmann and Kroemer 2004). Most relevant to this discussion, they pump toxicants from cells. © 2008 by Taylor & Francis Group, LLC Clements: “3357_c007” — 2007/11/9 — 12:42 — page 98 — #4 98 Ecotoxicology: A Comprehensive Treatment Box 7.1 Cadmium and Cells Epithelial cell transport of cadmium is a good example of a toxicant movement requiring a range of mechanisms (Zalups and Ahmad 2003). Cadmium can compete for transport sites associated with movement of essential elements (calcium, iron, or zinc). For example, epithelial cells of the intestine have a zinc transporter system through which cadmium also enters cells. Cadmium also enters cultured intestinal epithelial cells via a calcium-binding protein (Pigman et al. 1997) andthecrustacean gillby diffusion facilitated bya calcium-bindingprotein (Rainbow and Black 2005). This general type of entry route is categorized as ionic mimicry or ionic homology. Characteristic of the second general route of entry, cadmium can conjugate with thiol-containing compounds such as glutathione or cysteine, and the resulting conjugates pass through membranes by facilitated diffusion mechanisms designed for organic anion transport (Pigman et al. 1997, Zalups and Almad 2003). Aduayom et al. (2003) and Pigman et al. (1997) found evidence of such movement in cultured intestinal epithelial cells. Zalups and Ahmad (2003) refer to this second general route as molecular mimicry or molecular homology. As a third mechanism, cadmium associated with proteins such as metallothionein or albumin can enter the cell by endocytosis. Finally, Pigman et al. (1997) suggest that cadmium might also move into intestinal epithelial cells by passive diffusion. These mechanisms of cellular transport of cadmium manifest to differing degrees throughout the body’s tissues, resulting in differential uptake, distribution among the organs, localized effects, and elimination. Figure 7.2 illustrates broadly how these processes result in the complex cadmium transformations and dynamics in the mammalian body. FIGURE 7.2 A general illustration of the consequences of cadmium uptake, conver- sion, and elimination from cells of the vari- ous organs of the mammalian body. Shown in the figure are cadmium bound to pro- tein (CdP), metallothionein (CdM), cysteine (CdC), homocysteine (CdH), gluatathione (CdG), N-acetylcysteine (CdN), and other thiol-containing compounds (CdO). (Derived from Figure 1 in Zalups and Ahmad (2003) to which the reader is referred for a more comprehensive description.) Lungs Inhaled Ingested Cd Inhaled Cd Liver CdM CdC 2 CdG 2 CdC 2 Bile Intestine Inhaled Cd Absorbed Cd CdM CdP CdO CdM CdP CdO Feces CdC 2 CdM CdP CdG 2 CdO CdC 2 CdG 2 CdM CdP CdP CdG 2 CdC 2 CdP CdC 2 CdP CdO CdN CdM CdG 2 CdN CdH Urine Blood CdM CdO CdG 2 CdO CdM CdG 2 CdC 2 CdP Cd-oligopeptides Kidney CdG 2 CdO CdM CdP Endocytosis contributes to uptake and elimination from cells. Kupffer cells present in the mammalian liver sinusoids phagocytize some toxicants present in the blood (Timbrell 2000). Small particles of iron oxide or iron saccharate are phagocytized by cells associated with oyster gills (Galtsoff 1964). It is also well established that the cells of the molluscan digestive gland © 2008 by Taylor & Francis Group, LLC Clements: “3357_c007” — 2007/11/9 — 12:42 — page 99 — #5 Bioaccumulation 99 Inhalation of vapors during preening Direct dermal exposure Ingestion of food Ingestion during preening FIGURE 7.3 Three routes of exposure illustrated with a common loon coming into contact with spilt oil. The three routes include ingestion, direct dermal exposure, and inhalation. Ingestion involves consumption of tainted food and incidental ingestion during preening of oiled feathers. Dermal exposure can involve contact with dissolved oil components such as polycyclic aromatic hydrocarbons and physical contact with the floating oil. Inhalation of volatile components can occur especially during preening of oiled feathers. (hepatopancreas) phagocytize particles (Purchon 1968) including those containing contaminants. Cells of this gland also eliminate material by expulsion from vacuoles into ducts leading to the gut. The toxicant-containing materials are then removed from the individual after incorporation into feces. Molluscs and other invertebrates sequester toxic metals in intracellular granules and can eliminate the sequestered metals by cellular release of metal-rich granules into the ducts of the digestive gland (Lowe and Moore 1979, Mason and Nott 1981, Simkiss 1981, Wallace et al. 2003). 7.2.2 ROUTES OF ENTRY INTO ORGANISMS Considering contaminant movement at a higher level of organization, contaminants can enter an organism through the gut, respiratory surfaces, and dermis (Figure 7.3). These are the classic entry routes discussedthoroughly inmammalian toxicologyand quantifiedcarefully duringrisk assessment activities. Some aspects of toxicant movements in the associated organs have already been described briefly in Sections 5.2 (dermal), 5.3 (respiratory surfaces), and 5.5 (gut). Oral exposure involves direct ingestion in food or imbibed water, or ingestion during grooming, preening, or pica. Some chemicals that enter initially by inhalation can also be swept back up from within the lungs, swallowed, and gain entry into the digestive tract. The ingestion route becomes more complicated for nonhuman species. Some invertebrates pos- sess elaborate feeding structures involved in respiration (e.g., lugworms) or locomotion (e.g., cope- pods). Contaminant entry to such individuals is influenced by the demands of respiration and movement in addition to feeding. Uptake after ingestion can change if a contaminant or co-contaminant damages gastrointestinal tissues. This phenomenon of malabsorption is well studied in pharmacology. For instance, Gibaldi (1991) describes how the cancer treatment drug 5-fluorouracil damages the intestinal epithelium, allowing movement of large polar molecules that otherwise could not pass through the gut wall. Similarly, ethanol damage increases movement of toxic chemicals as large as 5000 Da across the guts of alcoholics (Bjarnason et al. 1984, Gibaldi 1991). Keshavarzian et al. (1999) recently proposed that © 2008 by Taylor & Francis Group, LLC Clements: “3357_c007” — 2007/11/9 — 12:42 — page 100 — #6 100 Ecotoxicology: A Comprehensive Treatment ethanol-induced malabsorption of endotoxins is responsible for liver cirrhosis of many alcoholics. Relative to nonhuman species, gastric erosion and hemorrhaging were apparent in oiled sea otters (Enhydra lutris) after the Exxon Valdez spill (Lipscomb et al. 1996). This created a condition that would foster malabsorption. Uptake can also change with induction of cellular mechanisms or physiological processes already described. As an example, both digesta retention time in the gut and lipid absorption decrease after ingestion of crude oil by river otters (Lontra canadensis) (Ormseth and Ben-David 2000). Such changes were speculated to contribute to the drop in body weight and general condition observed in sea otters living near the 1989 Exxon Valdez spill. Respiratory uptake from air or water is a major entry route for animals. Inhaled toxicants can be gaseous, associated with liquid aerosols, or incorporated into solids for air breathing species or life stages. As examples, the sea otters (E. lutris) studied by Lipscomb et al. (1996) showed pulmonary interstitial emphysema, suggesting lung contact via inhalation of oil spill-related volatile compounds. Peterson et al. (2003) also indicate that harbor seals (Phoca vitulina) living in the area of the Exxon Valdez spill were exposed via inhalation of oil fumes enough to produce brain lesions and other damage. Lead associated with roadside dust can penetrate deeply into terminal bronchioles and alveoli with subsequent dissolution (Biggins andHarrison 1980). Black kite (Milvus migrans) nesting near an incinerator accumulate lead via respiration (Blanco et al. 2003). Water-breathing species can also be exposed to toxicants in gaseous, liquid (dissolved or micelles), and solid phases. The importance of the respiratory route varies for aquatic species that can have different respiratory strategies as was clearly illustrated by Buchwalter et al. (2003) with diverse aquatic insect species exposed to the organophosphate insecticide, chlorpyrifos. Most atten- tion is focused on uptake from water and gas phases; however, as we saw earlier, relative to metal uptake from particles on oyster gills (Section 7.2.1), particulate-associated toxicants on gills can be taken up under certain conditions. For example, lead adsorbed to gibbsite can gain entry into goldfish (Carassius auratus) after gibbsite particles adhere to gill surfaces and the associated lead desorbs (Tao et al. 1999). Obviously, damage to respiratory surfaces can modify uptake in ways similar to that described for the ingestion route. Considering all the species living within ecosystems, it would be a mistake to focus only on these routes of entry that emerge out of classic mammalian toxicology—even for the animal kingdom. The paradoxically high arsenic concentration in the giant clam, Tridacna maxima, tissues is a good illustration of this point. The Great Barrier Reef giant clams and their symbiotic zooxanthellae live in phosphorus-deficient waters. The zooxanthellae within the clam tissues actively take up and metabolize arsenate in their attempt to extract as much of the meager amount of phosphate as possible from surrounding waters. Once taken up, the arsenic is converted to various organic forms and accumulates to extremely high concentrations in various tissues of the giant clam and other invertebrates containing symbionts (Benson and Summons 1981). This symbiont exposure route does not fit neatly into the context of the three classic routes of exposure. Nor does an endoparasite’s exposure to a toxicant present in a host’s tissues and body fluids. Focusing for a moment on plants, stomatal entry of gaseous air pollutants such as sulfur dioxide (Kimmerer and Kozlowski 1981), or uptake via aerial or terrestrial roots, e.g., arsenic uptake from soils (Wauchope 1983), might be important to consider. In urban areas, particulate-associated con- taminants contact plants by simply settling onto their exposed surfaces. Treatment of these routes of entry would not necessarily involve minor changes to the methods applied for respiratory (sto- mata?), ingestion (roots?), or dermal (plant surfaces?) routes of entry for animals. Microscopic organisms can also require a different vantage for assessing toxicant entry. As an important example, treatment of toxicant entry into unicellular algae might be most effective if the context developed above for cellular movement of toxicants was adopted instead (e.g., Crist et al. 1992, Klimmek et al. 2001, Morris et al. 1984). As another, but more extreme, example involving arsenic, some microbes use arsenic oxyanions to generate energy (Oremland and Stolz 2003) and, for this reason, would require a very different vantage for discussions of exposure routes and associated uptake calculations. © 2008 by Taylor & Francis Group, LLC Clements: “3357_c007” — 2007/11/9 — 12:42 — page 101 — #7 Bioaccumulation 101 7.2.3 FACTORS MODIFYING UPTAKE What general rules exist regarding the uptake of contaminants by these routes? Some rules of thumb emerge despitethe diversity of relevant species, toxicants, and media. Most are based on the tendency of particular toxicants to engage in specific reactions, including transport, and to preferentially associate with a certain phase. Several key themes are sketched out here. Tendencies to partition between aqueous and lipid phases are often used to predict uptake and bioaccumulation of nonpolar organic chemicals. This point can be illustrated for fish gill uptake of nonpolar compounds differing in lipophilicity(as measured by the logarithm of the octanol:water par- tition coefficient, log K ow ) (Figure 7.4). Connell and Hawker (1988) speculate that uptake increases with log K ow because membrane permeation by the chemical increases, but only to a point. At a cer- tain point, the large molecular size of the increasingly lipophilic chemicals begins to impede their diffusion in aqueous phases of the fish: these molecules have such low diffusion coefficients that the curve plateaus above approximately log K ow of 6 for many nonionic chemicals (Connell 1990). This decreasing of uptake rates results from steric hindrance as contaminant molecules attempt to pass through the cavities between membrane molecules (Connell 1990). The influence of lipophilicity depends on the route of entry. Lipophilicity is less important for chemical uptake in lungs. Boethling and MacKay (2000) generalize, “substances with solubility in water equal to or greater than their solubility in lipids are likely to be absorbed from the lung. Polar substances are generally absorbed better from the lung than nonpolar substances due to greater Log K 1 (1/day) K x (L /kg/h) K 1 (fugacity-based uptake rate) Log K ow Gobas and MacKay (1987) Erickson and McKim (1990) Connell and Walker (1988) 012345678910 012345678910 012345678910 FIGURE 7.4 Gill uptake (K 1 , K x ) versus log K ow . (Panels derived from Figure 3 of Connell and Hawker (1988), Figure 1 of Erickson and McKim (1990), and Figure 1 of Gobas and MacKay (1987).) © 2008 by Taylor & Francis Group, LLC Clements: “3357_c007” — 2007/11/9 — 12:42 — page 102 — #8 102 Ecotoxicology: A Comprehensive Treatment water solubility.” However, increasing lipophilicity of ingested chemicals does tend to increase uptake from food (Gibaldi 1991) but, similar to the limits discussed for neutral chemical uptake by gills, most chemicals with log K ow values greater than roughly 5 display diminished uptake after ingestion because they are sparingly soluble in gastric juices (Boethling and MacKay 2000). For dermal exposure of mammals, compounds with higher lipid solubility tend to have greater rates of uptake than less lipid-soluble compounds. The pH Partition Theory (Hogben et al. 1959, Shore et al. 1957) is a central one for dealing with ionizable toxicants. This theory, often invoked when dealing with absorption of ingested chemicals, is based on the simplifying (and sometimes insufficient) assumption that the gut can be envisioned as a simple lipid barrier. Nonionized forms of acidic or basic toxicants pass through this lipid barrier much more readily than ionized forms. This being the case, one can insert a compound’s pK a and the pH of the gut region where uptake is to occur into the appropriate Henderson–Hasselbalch equation (also see Equation 3.5 and associated discussion) to predict how readily the chemical might cross the lipid barrier. Henderson–Hasselbalch equations for weak (monobasic) acids (Equation 7.1) or (monoacidic) bases (Equation 7.2) are the following: f u = 1 1 + 10 pH−pK a (7.1) f u = 1 1 + 10 pK a −pH (7.2) where f u is the proportion of the toxicant remaining unionized. The calculated degree of ionization of a chemical and this theory are often adequate for approximating weak acid or base uptake by passive diffusion but, as evidenced by our previous discussions, many uptake mechanisms exist beyond simple diffusion of an uncharged chemical through the lipid route. Many of the mechanisms discussed above can facilitate uptake of an ionized compound. Regardless, based on pH Partition Theory, one can generally predict for compounds ingested by humans that weak organic bases tend to be taken up in the intestine and weak organic acids tend to be taken up in the stomach as well as the intestine (due to its high surface area and blood flow) (Abou-Donia et al. 2002). Gibaldi (1991) discusses a very instructive elaboration of the pH Partition Theory made by Hogerle and Winne (1978) that incorporates the depth of the unstirred layer immediately adjacent to the mucosal cells, pH of the media immediately adjacent to the mucosal cell membrane, amount of surface area available for uptake, and uptake of both the ionic and nonionic forms of the chemical: Absorption rate = CA (T/D) +{1/P u [f u +f i (P i /P u )]} (7.3) where C = toxicant concentration, A = area over which absorption is occurring, T = unstirred layer thickness, D = the chemical’s diffusion coefficient, f u = fraction of chemical unionized, f i = fraction of chemical ionized, P u = permeability of the unionized chemical, and P i = permeability of the ionized chemical. The pH immediately adjacent to the mucosal cell membrane is used instead of the general gut lumen pH in this model. Both the unionized and ionized forms are assumed to be taken up, albeit, at distinct rates. Another set of general theories govern our current predictions of metal uptake. A basic premise relative to metal toxicity is that the metal must first be in solution before being capable of interacting with a site of action. This is a sound premise if applied insightfully, not dogmatically. Where exactly the metal must be in solution is crucial to consider. Aparticulate-associated metal that is taken into the cell by phagocytosis can dissolve within the cell and cause an adverse effect. A lead halide particle inhaled deeply can release lead upon contact with moist respiratory surfaces. Finally, dissolved aluminum that encounters elevated pH conditions at the gill surface microlayer and precipitates will become associated with and cause harm due to its presence on the gill as a solid (colloid). © 2008 by Taylor & Francis Group, LLC Clements: “3357_c007” — 2007/11/9 — 12:42 — page 103 — #9 Bioaccumulation 103 Beyond this point that the dissolved metal is the most bioavailable form, Mathews (1904) proposed the ionic hypothesis more than a century ago. The ionic form of any dissolved metal is the most active form relative to biological uptake or effect. Beginning in the late 1930s, this context was expanded to correlate the relative toxicities of mono-, di-, and trivalent metal ions to their respective abilities to form complexes with biomolecules (e.g., Biesinger et al. 1972, Binet 1940, Fisher 1986, Jones 1939, 1940, Jones and Vaughn 1978, Kaiser 1980, Loeb 1940, McGuigan 1954, Newman and McCloskey 1996, Newman et al. 1998, Williams and Turner 1981). A series of ancillary theories have recently emerged around these well-established theories. The ionic hypothesis was augmented by what is now called the free ion activity model (FIAM): the free metal ion is the most important dissolved species relative to determining dissolved metal uptake and effect (Campbell and Tessier 1996). If applied with insight, the FIAM is an excellent rule of thumb; however, there are cases in which it should not be expected to apply. For example, Simkiss (1983, 1996) indicates that the neutral chloro complex of mercury, HgCl 0 2 , is lipophilic and potentially available for uptake via the lipid route. Charged uranium complexes as well as the free uranium ion have significant bioactivity (Markich et al. 2000). Another ancillary model goes under the name of the biotic ligand model (BLM): the bioactivity of a metal manifests if and when the amount of metal–biotic ligand complexes reaches a critical concentration (Di Toro et al. 2001, Santore et al. 2001). The BLM focuses on the activities of dissolved metal–ligand complexes and the metal–biotic ligand complexes formed at crucial sites on organism surfaces such as gill surfaces. The competition among other dissolved cations and dissolved ligands are also considered. The FIAM and BLM are often applied together to imply or predict a relationship between dissolved metal concentration (in bulk water or sediment inter- stitial water) and some adverse consequence. As with the FIAM, insightful use of the BLM can generate valuable explanations or predictions, but unthoughtful application can produce contradic- tions. As examples, the uptake rate of zinc by Chlorella kessleri was not directly related to the concentration of the free (aquated) ion, Zn 2+ , leading Hassler and Wilkinson (2003) to challenge the FIAM–BLM model. The discrepancy was attributed to the synthesis of new membrane-associated zinc transporters that were involved in active transport, i.e., transport against an electrochem- ical gradient. Using this same algal species, Hassler et al. (2004) found that the FIAM–BLM did correctly predict lead uptake in the absence of competitors but it failed to do so in the pres- ence of the competing ion, Ca 2+ . Work to this point suggests that prediction from FIAM–BLM is extremely useful but must be applied with a clear understanding of important underlying processes and modifiers. Newman et al. (1998) recently developed quantitative ion character–activity relationships (QICARs) that quantitatively predict metal activity based on Hard Soft Acid Base (HSAB) theory and FIAM–BLM. These QICARs are quantitative extensions of work by many others (e.g., Biesinger et al. 1972, Binet 1940, Fisher 1986, Jones 1939, 1940, Jones and Vaughn 1978, Kaiser 1980, Loeb 1940, McGuigan 1954, Williams and Turner 1981) showing relationships between metal bioactivity and metal–ligand binding tendencies. In some cases, such QICARs also allow bioactivity prediction for binary metal mixtures (Ownby and Newman 2003) (see Box 9.4). Last, Di Toro et al. (1990) combined aspects of various theories (HSAB, Ionic Theory especially FIAM) to make general predictions about sediment metal availability/bioactivity. In their approach, they assumed the following: (1) the dissolved ion is the most bioactive form of a metal, 3 (2) sulfides of many metals of concern are much less soluble than iron (and manganese) sulfides found in anoxic sediments, (3) the relatively large amount of iron (and manganese) sulfides often found in anoxic sediment provides a solid-phase sink for any metal in the sediments, (4) metals associated with solid- phase sulfides are essentially unavailable relative to that dissolved in interstitial waters, and (5) metal 3 Or, minimally, is a good indicator of bioavailable metal. © 2008 by Taylor & Francis Group, LLC Clements: “3357_c007” — 2007/11/9 — 12:42 — page 104 — #10 104 Ecotoxicology: A Comprehensive Treatment bioactivity in anoxic sediments can be predicted if one knows the amount of metal in the sediment and the amount of sulfides in the sediment available to react with and remove the metal from the interstitial waters. To this end, sediments are extracted with cold 1 N HCl, and the amounts of sulfide (acid volatile sulfides or AVS) and simultaneously extracted metal (SEM) are determined. A metal might be available to interact adversely with biota if the amount of metal in the sediment exceeds the capacity of the AVS to remove it from the interstitial waters, that is, if SEM–AVS > 0. Again, this SEM–AVS approach is very useful if used insightfully: there are cases in which considerable understanding is required to correctly interpret or predict from AVS–SEM information. For example, Lee et al. (2000) demonstrated clearly that many benthic organisms take up significant amounts of metals from solid sediment phases, contradicting a major assumption of the SEM–AVS approach. 4 Other researchers (Chen and Mayer 1999, Fan and Wang 2003) have applied biomimetic methods to provide support for the SEM–AVS approach for some metals. Chen and Mayer (1999) showed that the SEM–AVS approach gave similar predictions as their biomimetic approach except the presence of a threshold SEM–AVS suggested that other phases also contributed to availability. Fan and Wang (2003) found poor agreement between biomimetic and SEM–AVS studies of several metals. All of these recent permutations are ancillary to the established ionic and HSAB theories. If applied thoughtfully, the FIAM, BLM, QICAR, and SEM–AVS models can provide invaluable insights. Unfortunately, their dogmatic application or rejection is the source of some confusion in the field at this time. (See Chapter 36 for further details about dogmatic rejection of emerging paradigms.) 7.3 BIOTRANSFORMATION As described in Chapter 3, an organic toxicant can be subject to transformations that influence its retention. As a recent example, wide variation in the ability of invertebrates to metabolize PAH was found to lead to significant differences in PAH retention (Rust et al. 2004). Some might be eliminated immediately upon entry to a cell as in the case where the P-glycoprotein acts as a barrier to xenobiotic absorption (Box 3.1). If an organic toxicant gains entry into the individual, it can be eliminated in its original form by a variety of mechanisms already discussed or it can be converted to a form more amenable for elimination. It can undergo Phase I transformations and be eliminated in its new form(s). Alternatively, the Phase I metabolite(s) can be conjugated with one of a variety of endogenous compounds andthen eliminated. Inthe exampleofbenzo[a]pyrene metaboliteconjugates given above, the metabolites are transported via the circulatory system to the gut where they are moved back into the lumen via the ABC active transport proteins. As we have already discussed, final removal of the organic chemical or its metabolites can occur via a variety of other mechanisms, for example, organic anion removal via OATs. Inorganic toxicants can also undergo changes after entering the organism. With the simple case of sulfide in lugworms, we saw that elimination occurred with oxidation to thiosulfate and simple thiosulfate diffusion out of the worm via the paracellular route. Toxic metal transformations tend to be more involved than this. Metals can become complexed with a ligand and transported to the site of elimination in that form. Lyon et al. (1984) found that metal elimination from crayfish (Austropotamobius pallipes) hemolymph was linked to metal–ligand binding tendencies as predicted by HSAB theory. Metals can be taken into cell vesicles and sequestered in granules (e.g., Coombs and George 1977) (also Section 4.5.1) or cysteine-rich proteins. 5 Metal-rich granules can be emptied 4 The complicated nature of metal uptake from sediments has stimulated the application of empirical methods for determ- ining bioavailability. The most recent adaptation from human pharmacology is the biomimetic approach. With this approach, an aliquot of sediment, soil, or food is placed into a solution that mimics digestive juices (e.g., Leslie et al. 2002, Rodriguez and Basta 1999) or into digestive juices themselves (e.g., Mayer et al. 1996, Weston and Maruya 2002, Yan and Wang 2002). The amount of toxicant released into solution is used to predict uptake after ingestion. 5 There is currently a regulatory movement afoot to relate toxicant effects to a critical body residue, instead of an environ- mental concentration. The complexity of relating a concentration in, for example, a sediment to a realized effect has prompted © 2008 by Taylor & Francis Group, LLC [...]... as lead can be excreted via the mammalian liver by an active transport mechanism (Rozman and Klaassen 1996) Abou-Donia et al (2002) also make the generalization that a molecular weight greater than 325 Da, structure containing two or more aromatic rings, or the presence of a polar group all tend to favor biliary excretion for weak organic acids Gibaldi (1991) provides a cut-off point of molecular weight... then reabsorbed (Abou-Donia et al 2000, Rozman and Klaassen 1996) Toxicant binding with circulating plasma proteins can strongly influence their ability to participate in glomerular filtration Tubular secretion, active transport of a compound from the blood capillary into the renal tubule, is also important (Gibaldi 1991) Anionic and cationic organic compounds are actively transported into the proximal tubules... W.-X., Metal exposure and bioavailability to a marine deposit-feeding Sipuncula, Sipunculus nudus, Environ Sci Technol., 36, 40– 47, 2002 Zalups, R.K and Ahmad, S., Molecular handling of Cadmium in transporting epithelia, Toxicol Appl Pharmocol., 186, 163–188, 2003 Zalups, R.K and Barfuss, D.W., Renal organic anion transport system: A mechanism for the basolateral uptake of mercury–thiol conjugates along... Environmental and Industrial Applications, 2nd ed., Williams, P.L., James, R.C., and Roberts S.M (eds.), John Wiley & Sons, Inc., New York, 2000, pp 129–143 Morris, R.J., McArtney, M.J., Howard, A. G., Arbab-Zavar, M.H., and Davis, J.S., The ability of a field population of diatoms to discriminate between phosphate and arsenate, Mar Chem., 14, 259–265, 1984 Newman, M.C and Jagoe, C.H., Ligands and the bioavailability... (Middendorf and Williams 2000, Rozmann and Klaassen 1996) Also occurring at this point in the process is tubular reabsorption that involves active reabsorption of chemicals such as water, salts, amino acids, and glucose Many drugs, especially lipid-soluble compounds, are subject to passive reabsorption because of the concentration gradient created during water reabsorption (Gibaldi 1991) Passive reabsorption... lipid-soluble compounds that are deposited in fatty tissues (Rozmann and Klaassen 1996) Toxicants associated with inhaled particles or liquid can be eliminated by the mucociliary escalator process (Dallas 2000) Mucus-entrapped toxicants are swept from the respiratory tract by the coordinated motion of a complex of cilia The toxicant can then be ejected from the respiratory system, or swallowed and gain... polychlorinated biphenyl and DDT in milk (Aguilar and Borrell 1994) Egg layers such as birds and fish transfer lipophilic contaminants to eggs Insects can eliminate metals by incorporation into the exoskeleton before molting (Lindqvist and Block 1994) Some contaminants such as metals and metalloids are lost via hair (e.g., Akagi et al 1995) or feathers (e.g., Becker et al 1994) Plants can eliminate contaminants... of body load of organochlorine pollutants with age in fin whales (Balaenoptera physalus), Arch Environ Contam Toxicol., 27, 546–554, 1994 Akagi, H., Malm, O., Branches, F.J.P., Kinjo, Y., Kashima, Y., Guimaraes, J.R.D., Oliveira, R.B., et al., Human exposure to mercury due to goldmining in the Tapajos River basin, Amazon, Brazil: Speciation of mercury in human hair, blood and urine, Water, Air, Soil... broods, Arch Environ Contam Toxicol., 27, 162–1 67, 1994 Benson, A. A and Summons, R.E., Arsenic accumulation in Great Barrier Reef invertebrates, Science, 211, 482–483, 1981 Benson, W.H., Alberts, J.J., Allen, H.E., Hunt, C.D., and Newman, M.C., Bioavailability of inorganic contaminants, In Bioavailability Physical, Chemical, and Biological Interactions, Hamelink, J.L., Landrum, P.F., Bergman, H.L., and... 2081–2088, 1 978 Kaiser, K.L.E., Correlation and prediction of metal toxicity to aquatic biota, Can J Fish Aquat., 37, 211–218, 1980 Keshavarzian, A. , Holmes, E.W., Patel, M., Iber, F., Fields, J.Z., and Pethkar, S., Leaky gut in alcoholic cirrhosis: A possible mechanism for alcoholic-induced liver damage, Am J Gastroenterol., 94, 200–2 07, 1999 Kimmerer, T.W and Kozlowski, T.T., Stomatal conductance and sulfur . (Gibaldi 1991). For example, metals such as lead can be excreted via the mammalian liver by an active transport mechanism (Rozman and Klaassen 1996). Abou-Donia et al. (2002) also make the generalization. Glomerular filtration can remove molecules as large as 60,000 70 ,000 Da but some are then reabsorbed (Abou-Donia et al. 2000, Roz- man and Klaassen 1996). Toxicant binding with circulating plasma proteins. phosphate and arsenate, Mar. Chem., 14, 259–265, 1984. Newman, M.C. and Jagoe, C.H., Ligands and the bioavailability of metals in aquatic environments, In Bioavail- ability. Physical,Chemical and