TOXIC EXPOSURE OF THE URINARY TRACT
Many environmental and industrial pollutants that appear to be well tolerated at low doses can produce renal damage at high doses; and some nephrotoxic agents that are poorly tolerated at any dose are used injudiciously or otherwise find their way into the environment. In addition, a growing list of therapeutic and diagnostic agents are capable of causing renal injury, and the problem of renal disease due to recreational drug use is growing.
The extent to which those and other agents result in clinical renal insufficiency and the nature of the population at risk are incompletely defined. However, many forms of acute and chronic renal failure occur for unknown reasons, and the incidence of end-stage renal disease (ESRD) has marked racial and regional differences, so nephrotoxicants might well present a serious health hazard. It has been estimated that one in five patients hospitalized with acute renal failure has been exposed to one or more nephrotoxic agents (Rasmussen and Ibels, 1982).
The identification of agents with nephrotoxic potential is hampered by several obstacles. Foremost is a constantly changing environment of chemical hazards. In addition, variations in diagnostic criteria make it difficult to identify conditions due to prolonged exposure to nephrotoxicants.
The lack of availability of simple and reliable tests for early renal injury and the long period between exposure to environmental nephrotoxicants and the onset of definable disease seriously limit the ability to define a cause-effect relationship. Standard indexes of renal function are rather insensitive markers of injury. Each human kidney contains some 800,000–1,000,000 nephrons, and a nephron can have an average filtration rate of 50 μ1/min. The total glomerular filtration rate of both kidneys (1,600,000–2,000,000 nephrons) exceeds 100 ml/min. After acute or chronic exposure to a nephrotoxicant, hypertrophy of less severely damaged nephrons tends to counterbalance the atrophy of the most severely damaged nephrons. It is conceivable that one-third of the nephrons of the two kidneys could be lost without a noticeable reduction in the whole glomerular filtration rate if
the remaining 1,330,000 nephrons (assuming a normal total of 2,000,000) hypertrophied so that their filtration rate increased by 50%. Such fine adjustment rarely occurs; the presence of a substantial amount of structural and functional damage might be impossible to determine precisely with standard tests of renal function.
Finally, there are almost no epidemiologic data on the number of people who develop ESRD as a result of acute or chronic exposure to environmental nephrotoxicants. All the estimates depend on inferences drawn from inconclusive sources, such as surveys of patients entering dialysis and transplantation programs. The results of several of the surveys indicate that substantial gaps exist in the ability to identify the primary abnormality leading to ESRD (Burton and Hirschman, 1979; Easterling, 1977; Evans et al., 1981; NIH, 1990; Rostand et al., 1982). For example, data compiled by the U.S. Renal Data System (USRDS) for the years 1987–1990 indicate that disease of unknown cause made up 6.6% of the cases of ESRD (NIH, 1993). Patients with interstitial nephritis not due to analgesics abuse were 3.4% of the total. Because those groups of patients include some who have been exposed to xenobiotics, such as heavy metals, it is possible that for 10.0% of the patients with ESRD, environmental and occupational nephrotoxicants might be of primary importance in the etiology of the disease. Even in other persons with ESRD, exposure to environmental pollutants might have been a factor in the onset or progression of the disease. Information on occupational history or other factors that would implicate a patient's environment in his or her potentially catastrophic illness is rarely available.
On the basis of the available social and demographic data, several notable groupings might be relevant to the incidence of renal diseases. ESRD is found in a disproportionately high percentage of minority, ethnic, and racial groups in the United States. Native Americans, blacks, and Hispanics, especially Mexican Americans, have overall ESRD rates about 3–4 times greater than the rate in whites (USRDS, 1991). Although the reasons for the increased susceptibility of minority groups to developing ESRD are unknown (Rostand, 1992), several possibilities have been suggested (Feldman et al., 1992). They can be separated into two broad categories: differences in the access to preventive health care and renal-replacement therapy, and physiologic heterogeneity among racial groups that might increase renal sensitivity to toxic exposures.
A study of 9,390 black and white New York state residents who began treatment between 1982 and 1988 sought to determine whether the incidence of ESRD due to the three most frequent causes (diabetic glomerulosclerosis, hypertensive nephrosclerosis, and glomerulonephritis) was related to socioeconomic status (Byrne et al., 1994). A clear effect of socioeconomic status on the incidence of ESRD due to diabetes or hypertension was demonstrated in whites, but, perhaps because of overriding factors, no such effect was seen in
blacks. It was suggested that vigorous pursuit of other epidemiologic factors in the development of the progression of renal disease in blacks and of the possible relevance of the different structural, physiologic, and vascular renal responses between blacks and whites is indicated.
The hypothesis that treated ESRD is associated with socioeconomic status—independently of the known associations with race, age, and sex—and the hypothesis that the higher incidence of treated ESRD among blacks could be explained by differences in socioeconomic status have been examined in a study that linked the information from the USRDS and the Bureau of Health Professions Area Resource File (Young et al., 1994). An inverse association between the incidence of treated ESRD and socioeconomic status, as estimated by average income of county of residence, was found after adjustment for race, sex, and age. Differences in socioeconomic status appeared to explain some of the difference between blacks and whites in the incidence of treated ESRD. That many patients who develop ESRD do so for unknown reasons, are proportionately more likely to be members of minority groups, and come from economically marginal backgrounds is consistent with the possibility that environmental factors influence the development of such disease.
In the United States in 1990, 165,000 people with irreversible renal failure received renal therapy for ESRD with the aid of chronic dialysis, and 9,800 renal transplantations were performed (NIH, 1993). In that year, total medical payments to provide maintenance dialysis, kidney transplantation, and all related health services to ESRD patients were in excess of $6.39 billion. In view of the limited rehabilitation achieved by dialysis, the complications associated with transplantation, and the tremendous costs involved in each, substantial efforts are required to identify the specific causes of renal disease and the factors that determine progressive and irreversible decline in renal function.
THE URINARY TRACT
To understand the inherent limitations in the detection of early renal injury, we consider here several elements of normal renal function. In the sections that follow, we treat the mechanisms that lead to renal toxicity once exposure to a nephrotoxic substance occurs, the host factors that modify the response and the nature and extent of the physiologic adjustments to injury; it is presumably the responses to injury that give rise to the various markers described in later chapters.
Renal Blood Flow
The kidneys are highly vascular organs with a blood flow of about 1,000 and 1,200 ml/min in women and men, respectively, of average height and weight. That flow is about one-fifth of the resting cardiac output. Within the
kidney, 85–90% of blood flows through the cortex and only 10–15% through the medulla.
The initial step in the formation of urine is the production of an ultrafiltrate of plasma (filtration under pressure results in the retention of colloids but permits the passage of crystalloids). The unique interposition of the glomerular capillaries between the afferent and efferent arterioles is fundamental to the formation of an ultrafiltrate. Each minute, the kidneys produce 100–140 ml of glomerular filtrate with an osmolality of 280–290 mOsmol/L. In 24 hours, this amounts to 150–200 L of filtrate and over 40,000 mOsmol of solute. The amount and characteristics of the filtrate are influenced by the area available for filtration and the electric charges on the glomerular capillaries.
Tubular Reabsorption and Secretion
Once the glomerular filtrate is formed, it passes through a complex series of tubular structures where it is modified in such a fashion that waste products are excreted in the urine, critical body constituents are conserved, and the body's fluid volume is regulated. More than 99% of the filtered solute and water is reabsorbed. The principal oxygen-consuming work performed by the kidney is electrolyte reabsorption. Urine volume depends on the dietary intake of water, endogenous water production, insensible water losses, and the ability to concentrate or dilute the urine. The final osmolality of the urine can be as high as 1,400 mOsmol/L or as low as 40 mOsmol/L. The mechanisms by which the kidneys adjust the final composition of the urine are varied. Some substances, which are protein-bound, escape filtration only to be added to the urine by the process of tubular secretion. Other substances that are freely filtered—such as amino acids and glucose—are in normal circumstances completely reabsorbed by the tubules. These processes generally require the expenditure of energy and are particularly vulnerable to the effects of toxicants.
Renal function is modified by several extrarenal and intrarenal hormones. The major extrarenal hormones—aldosterone, vasopressin, and parathyroid hormone—modulate the excretion of sodium, water, and phosphorus, respectively. Intrarenal hormones—such as renin, prostaglandins, and kallikreins—affect renal blood flow, glomerular filtration, and tubular function. The kidneys also produce erythropoietin, which stimulates red-cell production; synthesize vitamin D from its precursor; and participate in the metabolism of several hormones, such as insulin. A complex group of peptide mediators, cytokines, influence cell growth and function;
these are produced locally or systemically and have the potential to influence the response to injury.
MECHANISMS OF RENAL TOXICITY
Susceptibility to Injury
The susceptibility of the kidney to toxic damage is related to various aspects of renal function. First, because blood flow to the kidney per gram of tissue is greater than that to most other organs, the total amount of toxicant delivered can be disproportionately high. Second, the processes of glomerular filtration, tubular reabsorption, and secretion tend to concentrate a toxicant that reaches the kidney. Third, the high metabolic rate of tubular epithelial cells leaves the kidney vulnerable to the actions of metabolic inhibitors. Fourth, the kidney can metabolically alter various endogenous and exogenous chemicals; this generally produces compounds with reduced biologic activity, but occasionally compounds with increased biologic activity are formed. Fifth, the mechanism of countercurrent exchange, which allows the kidney to form a concentrated urine, can prolong the residence time of a toxicant in the kidney.
Direct Toxic Effects
The nephrotoxicity of environmental pollutants is determined by their particular chemical properties, the duration and extent of exposure, and the nature of the host response. Manifestations of toxicity are related to the site of action in the kidney, the degree of damage produced, and the ability of the kidney to compensate for the loss of function or to repair injury.
Renal Vascular Injury
Involvement of the renal vasculature leads to changes in renal vascular resistance with a redistribution of blood flow in the kidney, a decrease in total blood flow, or both. The kidney can also lose its ability to autoregulate its blood flow. To the extent that renal plasma flow determines the rate of glomerular filtration, a decrease in the clearance of a number of substances can be expected to accompany changes in renal vascular resistance. Those effects might be mediated by anatomic changes in the renal vasculature, by changes in the sensitivity to systemic or local vasoactive substances, or by changes in muscular reflexes within the vascular walls themselves.
Glomerular Capillary Injury
The primary effect of a toxicant might be to change the ultrafiltration coefficient of the glomerular capillary membranes. That coefficient is a product of the glomerular capillary surface area and hydraulic conductivity. A decrease in either results in a proportional decrease in the filtration rate. Changes
in pore size or configuration and neutralization of the fixed negative charges can have generalized or selective effects on the ability of various substances to pass the glomerular barrier.
Renal Tubular-Cell Injury
The most firmly established effects of nephrotoxicants are on renal tubular epithelial cells, in particular those of the proximal tubule. Indeed, toxicity might be limited to a specific cell type or to a particular organelle in the cell. For example, some compounds have their major effect on tubular epithelial cell membranes, and others selectively alter the function of lysosomes, mitochondria, nuclei, or the endoplasmic reticulum (Fowler, 1982). Disruption of some organelles—particularly those which provide energy for cellular respiration—can lead to cell death.
The medullary countercurrent multiplication system provides an efficient mechanism for eliminating waste products and minimizing body-water losses. The system is such that drugs and their metabolites can accumulate in the medullary interstitium. Their chemical properties determine whether the accumulated substances initiate an inflammatory response.
The proximal tubule's organic acid and base transport systems provide an important route of elimination of molecular species that, as a result of their charge or size, do not undergo glomerular filtration but still require urinary elimination. Chemicals that interact with the organic ion-transport system can accumulate in cells or achieve high concentrations in the urine.
Other tubular mechanisms that can be impaired include those involved in electrolyte excretion and water metabolism. In addition, the mechanism of pinocytosis—whereby high molecular-weight molecules, if filtered, are recaptured from the proximal tubule fluid—can be disrupted. Finally, mechanisms of tubular epithelial cell regeneration can be compromised by nephro-toxicants.
Indirect Toxic Effects
Evidence has accumulated that the toxicity of environmental agents can in part be mediated by immunologic mechanisms that result in glomerular or tubulointerstitial disease (Wilson, 1982). There are four major categories of immunologic mechanisms. In Type I, or immediate hypersensitivity, damage results from the binding of antigen to IgE antibodies fixed to mast cells and basophils. In Type II, or cytotoxic reaction, damage results from the reaction of antibodies with cell-bound antigens and leads to activation of the complement cascade and cell death. Type III, or immune-complex reaction, stems from the formation of immune complexes in situ or in the circulation and leads to tissue damage. Type IV, or delayed hypersensitivity, is mediated primarily by T lymphocytes.
Data are insufficient to implicate a Type I response in mediating the effects of nephrotoxicants. But, various agents, including xenobiotics, might alter glomerular or tubular basement-membrane structures so that autoantigens are produced, autoantibodies or sensitized lymphocytes are formed and come to rest in the glomerulus through the process of filtration arrest, and a Type II response occurs. The Type III response is likely to be more important and might involve either the deposition of immune complexes formed in the intravascular compartment producing a serum-sickness-like reaction, or an antibody-antigen reaction in the extravascular compartment that results in inflammation produced by antigen-reactive cells, rather than antibodies—the so-called Arthus reaction. In either Type III case, the toxicant can act either as a full antigen or as a hapten. The haptens are small antigenic determinants that are covalently coupled to larger carrier molecules. Alternatively, various antibody-antigen complexes can be formed in situ. Once bound to tissue, the complexes fix complement, activating the complement cascade and triggering an in situ inflammatory response. In this situation, material previously trapped or planted in renal structures serves as the antigen; this material can be cationic proteins that are sequestered in the glomerulus or in vascular cells, where they become planted antigens. Later, a circulating antibody can attach to such antigens and result in in situ immune-complex formation. In addition, when structural damage is produced as a result of direct toxic effects, local auto-antigens can be produced. Environmental agents can also have a primary effect on other antibody-antigen interactions, favoring the formation of complexes of such a size or composition that nephritogenic immune reactions occur.
Apart from those considerations is the possibility that the Type IV mechanism, once thought to be of little importance in the development of renal injury, plays a role in some forms of interstitial disease. In the Type IV, macrophages, either leukocytes or monocytes, invade glomeruli and initiate a local inflammatory response mediated by cytokines, thromboxanes, and leukotriene prostaglandins.
Nephrotoxicants might also produce novel antigens that are capable of stimulating an autoimmune response in keeping with any or all of the four mechanisms.
Under most circumstances, foreign substances (xenobiotics), once absorbed, are distributed to various tissues where they undergo biotransformation with the production of innocuous metabolites, which are then eliminated. The enzymes responsible for biotransformation include various mixed-function oxidases in microsomes. Nonmicrosomal biotransformation can also occur. At times, these reactions result in the augmentation of toxicity. Although activity of these enzymes in the kidney as a whole is at a low level, certain specific
cell types show considerable activity in this regard. The site of transport of many xenobiotics is the proximal tubule. During transport, they must cross two cell membranes: the contraluminal and the luminal. Most xenobiotics are transported by the transport system for hydrophobic anions, some by the system for hydrophobic cations, and some by both (Ullrich and Rumrich, 1993). Several environmental pollutants—most notably the polybrominated biphenyls (PBBs), the polychlorinated biphenyls (PCBs), some halogenated dibenzodioxins, and hexachlorobenzene—have the capacity to alter activity of mixed-function oxidases (Kluwe and Hook, 1980). A result of this activity is that renal injury occurs after exposure to metabolites that by themselves would be otherwise well tolerated.
Necrosis and Apoptosis
There are two discrete types of cell death. Most commonly described is necrosis, an inflammatory process in response to cell injury from a variety of causes, such as exposure to nephrotoxicants or ischemia. The other type of cell death, apoptosis or programed cell death, is a finely controlled, active process that affects scattered individual cells, rather than tracts of contiguous cells, as occurs with necrosis. Various external and internal stimuli regulate apoptosis. Low levels of stimuli, such as ionizing radiation and toxins, can initiate apoptosis (Duvall and Wyllie, 1986). As described in the final section of this chapter, oncogenes are genetic loci ordinarily carried on tumor viruses that are responsible for neoplastic transformation. Under some circumstances, the proto-oncogene c-myc can drive cells into apoptosis (Evan et al., 1992). The bcl-2-proto-oncogene protects against apoptosis (Hockenbery et al., 1992). The tumor-suppressor gene p53 promotes differentiation, maturation, and apoptosis (Clarke et al., 1993).
Although the precise control mechanisms are unknown, cell injury can induce synthesis of genes that stimulate the apoptosis process. Apoptosis has been demonstrated in oxidant injury to renal tubular epithelial cells, renal arterial stenosis, ischemia-reperfusion injury, hydronephrosis, polycystic kidney disease, glomerulonephritis, lead nitrate injury, radiation nephropathy, and analgesic nephropathy. Further research on the mechanism of apoptosis might identify means of reversing this process and preventing atrophy.
HOST FACTORS IN RENAL TOXICITY
Various host factors can alter renal toxicity by influencing the metabolism of xenobiotics, by minimizing the concentrations of toxic substances in the kidney, or by decreasing susceptibility to cell injury by other mechanisms. The various host factors are important in determining populations at risk, as discussed in detail in Chapter 3. Physiologic variation associated with age alters the response to various toxicants, as do
some forms of underlying renal insufficiency. Other environmental concerns, such as nutritional status, can also influence toxicity.
Changes with Renal Growth and Aging
The susceptibility of the kidneys to toxic damage increases with age. This increased susceptibility can be related to a number of anatomic and functional changes that occur over time (Darmady et at., 1973; Epstein, 1979; Friedman et al., 1972; Spitzer, 1982). Those changes can be separated into phases of growth, maturation, and aging. At birth, the human kidney contains a full complement of nephrons, those of the outer cortex being relatively small and incompletely differentiated. During the first year, the glomeruli mature and grow as renal blood flow and cardiac output increase and renal vascular resistance decreases. The length and volume of the tubular structures rapidly increase, especially the convolutions of the proximal tubules. The increase in tubular mass is reflected in a pronounced increase in kidney weight. After the first year, growth slows. By the age of 4 years, the nephrons in the outer cortex have longer proximal convoluted tubules than those of the inner cortex and the ability to concentrate urine, to excrete excess acid, to maintain sodium balance, and to transport organic ions increases substantially.
By age 18–20 years, maturity is reached, and a plateau in structural and functional development is obtained that persists until around the age of 40. Thereafter, regressive changes take place. Prominent among those are changes within the renal vasculature and glomeruli; in particular, glomeruli appear to shrink with proportional decreases in the length and volume of the proximal convoluted tubules. Renal blood flow decreases at about 10% per decade and faster after the age of 60 years. The glomerular filtration rate also decreases with age, as does the ability to conserve sodium and the ability to produce maximally concentrated urine. With regard to the decrease in the glomerular filtration rate, it should be noted that muscle mass and creatinine production also decrease, so the serum creatinine concentration remains constant; this is why the serum creatinine concentration cannot be used reliably as a measure of glomerular function in the elderly.
The kidney's susceptibility to injury increases with age. This is most apparent with renal ischemic injury and has been demonstrated in both clinical settings (Balslov and Jorgenson, 1963; Groeneveld et al., 1991; Kiley et al., 1960; McMurray et al., 1978; Swan and Merrill, 1953) and laboratory settings (Kunes et al., 1978). Similarly, kidneys of young animals are relatively resistant to ischemic insults (Kunes et al., 1978). Changes in subcellular structures, such as mitochondria and lysosomes, and variation in xenobiotic biotransformation might be important determinants in age-related responses to environmental nephrotoxicants. Age-related alter-
ations in the immune system might also play a role.
Changes with Diet
Malnutrition by itself does not seem to lead to parenchymal renal disease, but dietary inadequacy might result in developmental abnormalities in the very young and physiologic defects in adults. For instance, if caloric deficiency in the mother occurs early in the growth phase in the fetus, when cell multiplication is rapid, the kidneys might not achieve their proper weight or number of nephrons. Under normal circumstances humans have a full complement of nephrons at birth, and malnutrition after birth would not necessarily be expected to have an adverse effect on kidney size. However, the kidneys of some infants that died from protein-calorie malnutrition reportedly showed signs of chronic contraction and scarring.
In adults, physiologic defects pre-dominate; these can be acute or chronic and are generally reversible. For example, fasting is associated with a natriuresis that can be abolished by ingestion of carbohydrate. The natriuresis has been related to alterations in glucagon concentration (Spark, 1975) and the need to excrete anions produced as a consequence of continuing metabolism (Sigler, 1975). With prolonged malnutrition, although renal blood flow and glomerular filtration rate are thought to be normal, other physiologic measures might be adversely affected. In particular, abnormal responses to salt and water loads are reflected in a propensity for edema.
When normal kidneys are stimulated to undergo compensatory hypertrophy, dietary protein restriction retards the response. In chronic renal disease in both experimental animals (Alfrey and Tomford, 1982) and humans (Walzer, 1982), reducing phosphorus intake slows the progressive decline in function, presumably by minimizing the adverse effects of hypertrophy of residual nephrons. Nutritional status clearly can be an important determinant of the ultimate effects of exposure to environmental nephrotoxicants.
Influence of Pre-existing Renal Function
Nephron Number at Birth
Injury to a population of nephrons can be underestimated because of the so-called reserve capacity of the kidney (i.e., the compensatory increase in function of normal or less severely injured nephrons). It might be expected that an inherited reduction in the number of nephrons in a kidney could explain in part the highly variable rates of expression and progression of human renal disease. Persons with a greater number of nephrons at birth might be able to sustain renal function after initial injury better than those with fewer nephrons at birth. Indeed, it has been postulated that those born with nephron numbers at the low end of the distribution curve can demonstrate accelerated declines in
renal function after initial renal injury (Brenner and Anderson, 1989). Females have smaller kidneys and 10% fewer glomeruli than males (McLachlan et al., 1977), and age-related loss of renal function is faster in North American blacks than in whites (Boyle, 1970).
Acute Renal Disease
A combination of insults, which by themselves are mild and individually well tolerated, can result in unexpectedly severe acute renal failure. For example, mild tubular injury produced by gentamicin (Zager and Sharma, 1983) or amino acid infusion (Zager and Venkatachalam, 1983) sometimes potentiates the effect of ischemia. Even relatively mild renal ischemic injury increases the sensitivity of the kidney to damage by a number of nephrotoxic agents, including aminoglycosides (Zager, 1988) and radiocontrast agents (Humes et al., 1987). Similarly, studies in rats subjected to renal ischemia (Ding et al. 1991) and studies in human renal-transplant patients have shown that the later administration of cyclosporine has a deleterious effect on renal function.
Chronic Renal Disease
Patients with diabetes, severe atherosclerosis, or any type of pre-existing renal disease and patients who are hypovolemic but otherwise healthy are all at risk for the development of renal injury from nephrotoxicants. Perhaps the most compelling data on the influence of exposure to potential nephrotoxicants and the rate of progression of renal disease come from studies of the relationship between hydrocarbon exposure, glomerulonephritis, and other forms of nonneoplastic renal disease.
Physiologic Responses to Renal Injury
Physiologic Changes with Compensatory Hypertrophy
To some extent, the changes in individual nephron structure and function observed during the phases of growth and aging (Tucker and Blantz, 1977) can occur in a kidney undergoing compensatory hypertrophy (Deen et al., 1974; Finn, 1982; Hayslett, 1979), recovering from an acute insult, or adapting to chronic disease. The consequences of exposure to environmental nephrotoxicants can be expected to differ from the consequences in normal kidneys. For example, in chronic disease, some of the reserve capacity of the kidney has been used; that is, adaptive changes have taken place in individual nephrons in response to abnormalities in others.
It is helpful to consider the response that occurs in the nephrons of a normal kidney after the surgical or traumatic loss of its mate. The anatomic response is marked by a combination of hypertrophy and hyperplasia without the formation of additional nephrons. There is a homogeneous response of individual
nephrons that is most marked by growth in the proximal convoluted tubule and an increase in glomerular size. The corresponding functional responses are marked by an increase in renal blood flow with parallel decreases in pre-glomerular and postglomerular vascular resistances. The glomerular filtration rate increases but tends to lag behind the increase in renal blood flow. Consequently, the filtration fraction falls. The increase in renal plasma flow is a major factor in the increase of the glomerular filtration rate. Increases in the glomerular capillary hydrostatic pressure and in the surface area or hydraulic permeability of the glomerular capillary membranes have also been found. Before any of the changes in renal blood flow or glomerular filtration, tubular function changes so that salt and water excretion quickly increase to equal that from both kidneys while two were still functioning—an appropriate response aimed at maintaining homeostasis. As noted earlier, many of these changes resemble those observed during the normal growth phase and are more prominent in young and mature kidneys than in aged kidneys.
In contrast with the homogeneous response of individual nephrons in a normal kidney undergoing compensatory hypertrophy, the individual nephrons in a diseased kidney demonstrate a considerable degree of heterogeneity. Indeed, within such a kidney, regressive features (marked by necrosis and atrophy) coexist with progressive features (marked by hypertrophy and hyperplasia). The anatomic changes are accompanied by marked functional changes, the most prominent of which is the great variation in the filtration rates of individual nephrons (Allison et al., 1973). In many ways, hypertrophy of some nephrons tends to counterbalance atrophy in others (Finn, 1983). Nephrotoxic damage to hypertrophied nephrons of a diseased kidney causes a greater decline in the whole kidney glomerular filtration rate than does damage to an equal number of nephrons in a normal kidney.
Factors Associated with Progressive Low of Renal Function
A concern over the early detection of renal injury is appropriate and is heightened by the propensity of many forms of renal disease, once established, to progress relentlessly. It has been suggested that some proportion of nephrons are irreversibly injured after an initial injury. Adaptive changes in the remaining nephrons allow the whole kidney filtration rate to remain at near normal, so an acute injury might go undetected.
Those adaptive changes can lead to premature demise of nephrons that have undergone the greatest degree of hypertrophy. Because a prominent finding in markedly hypertrophic kidneys is an increase in the number of sclerotic glomeruli, it has been proposed that the hyperperfusion or increase in the glo-
merular blood flow that is associated with compensatory hypertrophy is deleterious to the glomerular capillaries (Brenner et al., 1982; Schimamuria and Morris, 1975). It is also possible that an increase in the glomerular capillary hydrostatic pressure or a primary alteration in the glomerular capillary membranes themselves can contribute to progressive glomerular damage and tubular atrophy. Whatever the cause, the loss of those nephrons contributes to the progressive decrease in whole kidney filtration rate and stimulates other, less-involved nephrons to undergo a similar process, thus repeating a destructive cycle. Eventually, the compensatory processes in some nephrons are unable to keep up with the progressive loss of other nephrons, and major alterations in renal function occur.
CLINICAL EFFECTS OF CHEMICAL EXPOSURE ON THE KIDNEY
Acute Renal Failure
Acute renal failure is marked by a progressive rise in the serum concentration of creatinine and other nitrogenous compounds. It has several causes, many of which are due to ischemic or nephrotoxic agents. Acute renal failure is often but not always accompanied by a reduction in urinary output to less than 500 cm3 /day. Even if urinary flow is unaltered, inevitable compositional changes in the urine reflect parenchymal injury and tend to separate postischemic and nephrotoxic renal failure from reduction in renal function that occurs merely as a result of alterations in systemic hemodynamics.
When acute renal failure occurs in association with drugs and toxicants, the overall mortality rate is about 37%. In the majority of those who survive, life-sustaining renal function can be expected to return, but recovery is usually not complete. In a small percentage, recovery of renal function does not occur. In the remainder, various degrees of structural and functional impairment persist indefinitely.
Acute tubulointerstitial nephritis is marked by interstitial edema and infiltration of the interstitium with inflammatory cells, some of which appear later in the urine. Both structural evidence and functional evidence of tubular epithelial cell injury is present. Chronic tubulointerstitial nephritis is distinguished by the presence of interstitial fibrosis and tubular atrophy. The manifestations of tubulointerstitial nephritis depend on the extent of injury, the tubular segments most severely involved, and the degree of compensation achieved by the less severely involved nephrons.
The proximal tubule is responsible for the reabsorption of some 60–70% of filtered sodium and water and nearly all
the filtered glucose, amino acids, and low-molecular-weight proteins. The predominant site of phosphate reabsorption and bicarbonate reabsorption and regeneration is also the proximal tubule. Individual cell types in the proximal tubule are identified with one or more of these processes. Damage to these cells can be expected to result in the appearance in urine of the substances ordinarily reabsorbed or metabolized. Damage to more distal structures—including the loop of Henle, the distal convoluted tubule, and the collecting duct—is accompanied by abnormalities in the ability to concentrate and dilute urine. The former effect is more pronounced and can result in polyuria. Acidification of urine also occurs at distal sites, where damage can lead to metabolic acidosis.
There is continuing interest in the urinary presence of various cellular enzymes and low-molecular-weight proteins as markers of recent renal damage. One such enzyme is N-acetyl-beta-glucosaminidase (NAG), which is released from injured renal tubular cells; the activity of this enzyme is often the only clinical-chemistry value that is increased in the urine in workers exposed to inorganic mercury, a classical renal poison. The appearance in the urine of low-molecular-weight proteins, such as beta2-microglobulin, is also a marker of renal injury. Under normal conditions, low molecular weight proteins are filtered through the glomerulus and undergo complete reabsorption by the action of the renal tubules; in tubular disorders, as can be found in people exposed to cadmium, reabsorption can be incomplete and high urinary concentrations can result.
Exposure to environmental cadmium can cause chronic interstitial nephritis. Proteinuria and other signs of renal dysfunction have been found in people who live in cadmium-polluted areas or near smelters. Qualitative analysis of the urinary protein most commonly reveals a small albumin fraction and large alpha2, beta, and gamma protein fractions. Cadmium-produced injury to proximal tubular epithelial cells also results in the presence of aminoaciduria, enzymuria, and glycosuria. A decrease in the tubular reabsorption of phosphorus, an increase in the fractional excretion of uric acid, and an increase in cadmium excretion can be present. Those abnormalities were highlighted by the results of recent CADMIBEL studies in Belgium (Buchet et al., 1990). The studies compared the relative sensitivities of various urinary biologic markers, which included retinol-binding protein, NAG, beta2-microglobulin, amino acids, and calcium as biologic markers of cadmium-induced nephropathy among environmentally exposed human populations. All were found to be significantly increased in association with increased 24-hour urinary cadmium excretion. There were, however, differences in the relative sensitivities. For example, increased excretion of retinol-binding protein, beta2-microglobulin, and amino acids occurred at lower urinary cadmium excretion levels than the other markers.
Two types of lead nephropathy can
be found in association with lead poisoning. The first is an acute form marked by generalized defects of proximal tubular function with aminoaciduria, glycosuria, and phosphaturia. These abnormalities most often occur in children after several months of heavy lead ingestion. The defects are generally rapidly reversible. Some patients will develop the chronic form. This condition is an indolent disease that is difficult to separate from other forms of chronic, slowly progressive renal insufficiency. Its incidence is difficult to determine. Evidence of prior excessive lead absorption might be found by administering EDTA and then measuring urinary lead excretion.
Chronic glomerulonephritis disease is an insidious process generally accompanied by albuminuria and microscopic hematuria. Its onset is often impossible to date, and its diagnosis is usually delayed until a secondary complication occurs, such as hypertension, anemia, or metabolic bone disease. Lacking such a complication, the diagnosis might not be suspected until an abnormal urinary sediment is seen during routine examination. In patients with chronic renal failure, the progressive decline in the glomerular filtration rate is too slow and the deviation from the steady-state condition too small to result in day-to-day changes in the serum creatinine or blood urea nitrogen concentration. A decrease in kidney size confirms the chronic and irreversible nature of this condition. Examination of tissue obtained by percutaneous renal biopsy can confirm that the primary process involved glomerular structures. Such a distinction is often clouded by nonspecific changes that occur in all forms of chronic renal disease.
Rapidly Progressive Glomerulonephritis
Occasionally, a more aggressive form of glomerular disease occurs in which renal function is lost over a period of weeks or months. This so-called rapidly progressive glomerulonephritis can be identified by the presence of glomerular epithelial crescents in renal-biopsy specimens. The kidney is sometimes the only organ affected; at other times the renal abnormalities are part of a systemic disease that can result from severe vasculitis.
The combination of prolonged, excessive exposure to hydrocarbons with unidentified host factors can predispose to glomerular injury or aggravate injury due to other causes (Yaqoob and Bell, 1994). An apparent association of Goodpasture's syndrome—a form of rapidly progressive glomerulonephritis—with exposure to petroleum products has been reported (Beirne and Brennan, 1972; Bombassei and Kaplan, 1992). It has also been claimed that previous exposure to hydrocarbon solvents is a common feature in some groups of patients with crescentric glomerulonephritis or proliferative glomer-
ulonephritis (Zimmerman et al., 1975). Additional presumptive evidence that hydrocarbons can produce glomerular damage has come from the observation that remissions and exacerbations of the nephrotic syndrome follow removal from and re-exposure to solvents (Cagnoli et al., 1980). Other studies have noted a historical relationship between exposure to organic solvents and a wide spectrum of renal disorders, including tubular necrosis, interstitial disease, glomerulonephritis, and neoplasia (Nelson et al., 1990).
In a case-control study, hydrocarbon exposure was significantly higher in those with primary glomerulonephritis than in a group of normal subjects and an internal control group (Yaqoob et al., 1992a). Those with glomerulonephritis had a significantly greater exposure to petroleum products, greasing and degreasing agents, and paints and glue with resulting estimated risks of developing glomerulonephritis 15.5, 5.3, and 2.0 times greater than normal, respectively. In another study of patients with diabetes mellitus, hydrocarbon exposure was found to be significantly greater in those with incipient (microalbuminuria) and overt (macroalbuminuria) diabetic nephropathy than in those with no clinical evidence of nephropathy, with odds ratios of 4.0 and 5.8, respectively (Yaqoob et al., 1992b).
The role of hydrocarbon exposure on the progression of renal failure in patients with primary glomerulonephritis has also been studied (Bell et al., 1985; Ravnskov, 1986). Patients with primary glomerulonephritis and progressive renal failure have heavier hydrocarbon exposure and worse renal impairment at presentation than those with stable or improving function (Yaqoob et al., 1993). Moreover, patients with declining renal function were more likely to have continued occupational hydrocarbon exposure after the diagnosis of glomerulonephritis.
Several cross-sectional studies comparing measures of renal dysfunction in hydrocarbon-exposed and-nonexposed workers have suggested an association between hydrocarbon exposure and renal injury (Askergren et al., 1981; Askergren, 1984; Franchini et al., 1983; Hotz et al., 1991; Lauwerys et al., 1985; Mutti et al., 1992; Solet and Robins, 1991; Viau et al., 1987). For example, exposed subjects have been found to have a slight but significant increase in the abnormalities found at urinalysis (Askergren, 1984).
Increased proteinuria and tubular enzymuria (lysozyme and beta-glucuronidase) in the absence of albuminuria, indicative of tubular rather than glomerular dysfunction, has been seen in a large group of subjects exposed to aliphatic and acyclic hydrocarbons (Franchini et al., 1983). A separate study of 20,000 workers showed that the prevalence of proteinuria was higher in those with hydrocarbon exposure than in nonexposed subjects. Others have found a higher mean albuminuria and urinary excretion of renal antigen and a higher prevalence of antilaminin antibodies in a group of 53 male refinery workers (Viau et al., 1987).
The nephrotic syndrome is marked by the presence of heavy proteinuria (generally in excess of 3 g per 24 hours), hypoalbuminemia, and edema. In addition, abnormalities in lipid metabolism with hypercholesterolemia, hypertri-glyceridemia, and lipiduria are common. At the onset, the glomerular filtration rate might not be reduced and is occasionally increased. The nephrotic syndrome can occur in conjunction with a variety of systemic diseases. Or it can be a manifestation of a primary glomerular injury without a definable etiology and thus be termed idiopathic; in this situation, the condition is classified according to the appearance of glomeruli on light and electron microscopic sections and on the basis of various immunofluorescent patterns. Chronic occupational exposure to gold, bismuth, and mercury salts produces pathologic lesions similar to those found in some forms of the idiopathic nephrotic syndrome. One of the most common types, membranous glomerulonephropathy, is associated with an increased frequency of the histocompatibility leukocyte antigen (HLA) DR3 (Klouda et al., 1979). It has been suggested that exposure to environmental or occupational agents, such as formaldehyde, can act as a ''triggering'' agent in genetically susceptible persons (Breysse et al., 1994).
The association between kidney function and hypertension is based on both clinical and experimental observations reported over the last 150 years. Renal disease in the form of diabetic nephropathy, glomerulonephritis, interstitial nephritis, obstructive uropathy, polycystic kidney disease, pyclonephritis, and vasculitis is the most common cause of secondary hypertension in humans. Conversely, primary hypertension ranks second only to diabetes as an etiology for patients entering treatment for ESRD in the United States (NIH, 1993). It was recently proposed that ischemic nephropathy accounts for the rising incidence of hypertension-induced ESRD. Thus, the age-old question of cause versus effect appropriately characterizes the interaction between the kidney and hypertension.
It is germane to this report that some drugs or toxicants, often through their action on the kidney, are recognized causes of hypertension. Examples include amphetamines, estrogens and oral contraceptives, steroids, sympathomimetic drugs, tricyclic antidepressants, cisplatin, cyclosporine, licorice, lead, and ethanol.
Clinically, hypertension is usually detected as an incidental finding during a health evaluation, inasmuch as mild to moderate blood-pressure increases are usually asymptomatic. However, in a small percentage of patients, hypertension has a sudden severe onset accompanied by headache, nausea, vomiting, and mental confusion, which demand immediate emergency treatment; in these cases, underlying renal disease is often present.
It is estimated that one-fourth of
adults in the United States have hypertension and suggested that treatment decisions should be based on the associated evidence of target-organ damage that is traceable to increased blood pressure. Today, a multitude of antihypertensive drugs are available, so virtually any hypertensive patient can be successfully treated without intolerable side effects of administered drugs. Long term studies are being conducted to evaluate the effectiveness of antihypertensive treatment in preserving renal function and preventing renal damage. For patients whose hypertension is the result of chemical or drug nephrotoxicity, recognition and withdrawal of the offending agent are the appropriate clinical strategies. In addition, coexisting hypertension often emerges as a statistically significant risk factor in studies of nephrotoxicity.
CANCER OF THE BLADDER, KIDNEY, AND PROSTATE
Tumor-Suppressor Genes, Oncogenes, and Growth Factors
According to Barrett and Huff (1991), the multistep process of carcinogenesis can be operationally divided into initiation, promotion, and progression. Initiation consists of the first heritable alterations that predispose a cell to neoplastic transformation; promotion is the clonal expansion of initiated cells; and progression is the acquisition of other changes that are required for a cell to become fully malignant. It is now believed that more than two changes are required for neoplastic conversion of a cell; additional clonal evolutions most likely occur in the later stages and make the distinction between promotion and progression difficult. However, it is important not to confuse the two processes. Promotion involves the multiplication of the initiated cell, whereas progression involves the acquisition of additional, heritable changes in the initiated cell. Promotion can lead to progression, although with a low frequency. Progression can occur as a direct result of a chemical on an initiated cell. Increasing the target size of the population of initiated cells by promotional mechanisms will increase the probability of secondary spontaneous or chemically induced changes and therefore progression. The rate-limiting step in malignant development is the acquisition of additional genetic changes in an initiated cell. Therefore, a weak mutagenic effect can be at least as important as a potent tumor-promoting effect for cancer development.
A better way to define the multistep process of carcinogenesis is to identify the genetic alterations in tumor cells and attempt to determine how chemical carcinogens affect the neoplastic process that leads to these changes. There is now convincing evidence of the importance of two classes of genes in the carcinogenic process: proto-oncogenes and tumor-suppressor genes. Proto-oncogenes are a family of cellular genes with at least 40 members, which appear to be involved in normal cellular growth and development; activation or inappropri-
ate expression of these genes results in proliferative signals involved in neoplastic growth. Tumor-suppressor genes are less well defined, but they might also function in the control of normal cellular division and possibly differentiation. For a tumor cell to emerge, suppressor genes must be inactivated or lost. The number of tumor-suppressor genes is unknown.
The control of cellular growth resides in a complex, interacting system of positive (oncogenes) and negative (tumor-suppressor genes) controls. Each cell in a tissue responds to unique rules in the form of specific genes that control its growth. The subversion of these systems is responsible for the development of cancer. Indeed, specific tumor-suppressor genes and oncogenes peculiar to colon cancer have been identified.
Growth factors are peptides that regulate cellular growth and usually function in paracrine or endocrine modes rather than in an autocrine mode. The complex communication among various cells becomes subverted during tumorigenesis, because cells often move from paracrine to autocrine control (Nathan and Sporn, 1991). In some tissues, such as the prostate, growth is also under strong hormonal influences. For example, prostatic differentiation is under endocrine control during embryologic development. That is followed by a second sequence of changes associated with puberty. During adulthood, both hormonally dependent and hormonally independent cells can be found. Finally, the prostate undergoes hypertrophy as a result of dysregulated growth control among its various constituent cells. Each factor that controls growth can function as an important biologic marker. Also to be considered are cell surface receptors for the growth factors and the many proteins that are triggered by the growth peptides.
Cancer of the Bladder
In 1989, there were 47,000 new cases of bladder cancer and 10,000 deaths due to bladder cancer (Smart, 1990). In that year, bladder cancer accounted for 5% of all new cancer cases and 2.2% of all cancer deaths. The incidence of bladder cancer was 29.1 per 100,000 males and 7.7 per 100,000 females. Male bladder-cancer mortality was 5.9 per 100,000.
Bladder cancer presents an interesting paradigm of the mechanism by which environmental or occupational toxicants can be initiators or promoters of cancer development or factors in its progression. Indeed, bladder cancer is highly correlated with occupational exposure to xenobiotics. For example, Silverman (Silverman et al., 1989a,b, 1992) reported that 20–25% of cases of bladder cancer in white males, 27% in nonwhite males, and 11% in white females were associated with occupational exposure to toxicants. The relationship is not new: In 1895, Rehn first reported that bladder cancers could be caused by specific carcinogens in a group of workers exposed to aniline dyes. Various occupations have since been associated with bladder cancer. Table 2-1 contains na-
tional estimates, according to occupation, of the number of workers exposed to agents known to cause bladder cancer in animals. Over 200 chemicals have been suggested as associated with bladder cancer, and a few are documented human carcinogens (Anonymous, 1990). A list of compounds classified by IARC as human bladder carcinogens is contained in Table 2-2.
The bladder is a specialized neuromuscular organ whose muscular layer, or muscularis, is protected by a unique impermeable epithelium, the urothelium. This layer is three to seven cells thick and is capable of distention. The luminal layer consists of terminally differentiated cells, the so-called umbrella cells,
TABLE 2-1 Exposure to Animal Bladder Tumorigens in Selected Occupations During 1980s
No. of Workers
Miscellaneous machine operators
Machine operators, NECa
a Not elsewhere classified.
Source: Adapted from Ruder et al., 1990.
TABLE 2-2 Compounds Classified as Human Bladder Carcinogensa
which have a lifetime measured in months. This cell layer seems to be responsible for the bladder's impermeability to urinary solutes, perhaps because of prominent desmosomes, tight junctions between cells, specialized ion pumps, and a thick layer of highly charged glycosaminoglycan (Parsons et al., 1990). The bladder also contains components of the cytochrome P-450 system (Vanderslice et al., 1985), so it might have some role in xenobiotic detoxification or bioactivation, in addition to being subjected to the effects of carcinogens excreted in the urine (Kadlubar et al., 1992).
The urothelium is normally quiescent, but it is capable of re-entering the cell
cycle in response to injury. Epidermal growth factor (EGF), which is present at high concentrations in the urine, has been suggested as a major factor in urothelial growth and differentiation. Only the basal cell layer normally expresses EGF receptor (Messing et al., 1987). It is possible that injury to these cells and exposure to urine trigger rapid growth.
Bladder cancer, like many other cancers, occurs as a result of the interaction of genetic predisposition, occupational exposure, and a variety of cofactors. One of the prime nonoccupational causes of bladder cancer is cigarette-smoking, with an increasing incidence of disease in women that correlates with increased tobacco use (Cole and Hoover, 1971). Whether low-level arsenic exposure is synergistic with known carcinogens, such as cigarette-smoking, is unknown. These examples illustrate the complex interaction between genetic factors and multiple occupational or environmental toxicants.
Cancer of the Kidney
About 18,000–20,000 new cases of renal cancer are diagnosed each year in the United States. This malignancy, which makes up 2–3% of all cancers, ranks eleventh in cancer incidence and results in 8,000 deaths a year in the United States. White males have the highest mortality, 4.8 per 100,000; black females have the lowest mortality, 2.0 per 100,000.
About 85% of the renal cancers diagnosed are renal-cell cancers, whose incidence is about twice as high in men as in women. In 1986, the incidence of renal-cell carcinoma per 100,000 was 11.3 in white males, 11.9 in black males, and 5.6 in white females and black females. Kidney cancer is increasing in the United States; Huff reported in 1991 that percentage age-adjusted increases (per 100,000) in mortality (and in incidence) were 9.1% (21.7%) in whites, 38.2% (19.8%) in blacks, 7.8% (21.3%) in white males, 44.0% (26.9%) in black males, 13.5% (23.0%) in white females, and 34.7% (11.9%) in black females (Huff and Haseman, 1991).
Environmental agents have been implicated in the development of neoplasms of the renal parenchyma. Renal adenomas and adenocarcinomas account for about 85% of renal neoplasms. Commonly referred to as hypernephromas, these tumors arise from cells of the proximal convoluted tubule. They account for 2.1% and 1.60% of all cancer deaths in males and females, respectively. Squamous-cell carcinomas are much less common and account for 5–6% of renal neoplasms. Neither nephroblastomas nor renal sarcomas have been associated with renal carcinogens. Transitional-cell carcinomas of the renal pelvis and ureter can be induced by the same carcinogens that produce bladder tumors. Workers in the aniline-dye, rubber, textile, and plastic industries have a higher incidence of these tumors, which overall account for 7–8% of renal neoplasms. A list of occupations found to have excess risks for kidney cancer is contained in Table 2-3.
Renal cancers in humans have been associated with exposures to tobacco smoke, some environmental and occupational factors (e.g., coke-oven emissions
TABLE 2-3 Occupations Found to Have Excess Risks for Kidney Cancera
and possibly rubber-industry byproducts), and therapeutic agents, particularly analgesic mixtures containing phenacetin (Amico et al., 1991). A large epidemiologic study confirmed the association of lead and renal cancer in humans (Steenland et al., 1992). It is also suspected that arsenic is related to cancers of the kidney (Scandinavian Committee on Enzymes, 1985). However, most causes of kidney cancers are unknown, and these malignancies remain an important human health problem. The true incidence and mortality in the human population may be considerably higher than those reported. As evidence of that, more than one-third of the reports of the relatively few routine autopsies in the United States disclose undiagnosed cancers (Azzopardi and Evans, 1971). According to Holm-Nielsen and Olsen (Bauer, 1988), renal adenomas (minute cortical foci of proliferating tubular or papillary epithelium) often are present in 15–22% of all adult kidneys. Whether these small tumors (typically 2-3 mm, up to 6 mm) should be regarded as carcinomas or as benign precursors of renal-cell carcinomas remains controversial (Bauer, 1988).
Data from a number of animal and some human studies (IARC, 1987) generally support a relationship between lead-induced chronic renal disease and renal adenocarcinoma (Steenland et al., 1992), but an increased incidence of renal cancer has also been reported among lead-exposed workers without statistically increased rates of chronic renal disease (Steenland et al., 1992). The role of renal lead-binding proteins that both animals and humans mediate individual susceptibility to renal cancer from lead has been hypothesized (Fowler et al., 1994) as a mechanism for explaining observed variability among lead-exposed persons.
Recent evidence suggests that the original Knudson hypothesis—that the chromosomal regions often lost or mutated in tumors harbor tumor-suppressor genes—is correct. Several forms of renal cancer should be considered, including those with Wilms's tumor, the Von Hippel-Lindau disease, and the Li-Fraumeni syndrome, all of which represent the inheritance of a deleted or otherwise inactive suppressor gene (Latif et al., 1993). Studies of Wilms's tumor were the first to suggest that chromosomal defects in cancer cells harbored tumor-suppressor genes (Knudson and Strong, 1972). The Wilms's tumor-locus gene is a tumor-suppressor gene on chromosome 11p13. Germline mutations in WT-1 are associated with both the heritable and sporadic forms of Wilms's tumor and urogenital abnormalities. People heterozygous for mutations of the WT-1 gene are predisposed to Wilms's tumor (Haber and Housman, 1992). The tumor cells have lost heterozygosity and contain mutants at both alleles; this is consistent with the Knudson model of genetic predisposition to cancer. The WT-1 gene encodes a nuclear protein that possesses a so-called zinc finger domain governing DNA-binding specificity.
Von Hippel-Lindau disease is a relatively rare, dominantly inherited tumor disorder characterized by retinal angiomatosis and cerebellar hemangioblastoma (Latif et al., 1993). Renal-cell carcinoma is a frequent cause of death in
this disease. It is inherited as an autosomal dominant trait. The gene has been mapped to the short arm of chromosome 3.
The Li-Fraumeni syndrome is a familial tumor syndrome associated with malignant tumors in various organs. Inherited mutations of the tumor-suppressor p53 gene have been described. Like several other previously described tumor-suppressor gene products, p53 is thought to regulate transcription of genes critical to the control of cell growth and differentiation and in this way to be involved in the regulation of the cell cycle. The p53 gene is a promising candidate marker for cancer susceptibility (Harris, 1993). It is on chromosome 17p and is the most frequently mutated gene of human tumors in the United States. Evidence that the p53 gene is an anti-oncogene came from the observations that the wild-type form of the protein inhibits oncogene-mediated transformation of cells and that the growth of human cancers with endogenous p53 mutations was inhibited. In a detailed analysis of a group of families with multiple cancers first described by Li and Fraumeni, it was shown that the p53 gene was inherited in a mutated form and that cancer resulted only when additional mutations accumulated; this is consistent with the clinical appearance of cancer at ages 10–40 (Harris and Hollstein, 1993). That different carcinogens cause different characteristic mutations in the p53 gene suggests that the location and characteristics of these mutations can reveal clues about etiology and molecular pathogenesis of cancer. Thus, screening for germline mutations of the p53 gene could be used to identify a population at risk for cancer after exposure to occupational or environmental toxicants.
The retinoblastoma gene product regulates the cell cycle by maintaining cells in a non-proliferating state. It does so by binding to transcription factors presumably related to the cell cycle, thereby inactivating them (Marx, 1991). Binding is regulated by phosphorylation and dephosphorylation events—events also regulated during the cell cycle. The retinoblastoma gene binds to the adeno-viral E1A oncogenic protein, thus inactivating it and allowing the cells to enter the cell cycle continuously. Both alleles at the retinoblastoma-gene locus on chromosome 13 are defective in retinoblastoma and other cancers, including prostatic and bladder cancer (DeCaprio et al., 1989).
The importance of gasoline exposure and various aliphatic hydrocarbons in the induction of renal-cell carcinoma is controversial (Kadamani et al., 1989). The linkage to gasoline and hydrocarbon exposure stems from earlier studies in male rats that had an increased incidence of renal-cell carcinoma; however, the extension of the information obtained in the rats to humans has not proved informative (see Chapter 5). For example, the deposition of alpha2u-globulin, which has been observed in rats, has not been observed in humans (see Chapter 6). A review of the pathologic changes that were present in cases of renal-cell carcinoma in which hydrocarbon exposure was thought to play a role (Pitha et al., 1987) did not detect important changes in the normal kidney cells adjacent to
the carcinomatous cells. Those results do not substantiate the importance of hydrocarbon exposure in either the development of subclinical nephrotoxicity or the pathogenesis of renal-cell carcinoma.
Unsubstantiated evidence that occupational exposure is a factor in renal carcinoma indicates that development of biologic markers for individual risk assessment might assist in identification. The increased incidence in males might be attributed to the hormonal milieu or to smoking, which has been more frequent in men in the past. Recent studies indicate only a weak association with exposure to dry-cleaning agents, but there is a further need for individual risk assessment. The potential causative agents for renal-cell carcinoma have been extensively reviewed by Schulte and Kaye (1988).
Keys to the understanding of renal cancer might come from such seemingly diverse disciplines as toxicology, molecular biology, and cancer biology, which share an interest in understanding the regulation of the cell cycle. For example, the interaction of inherited defects in the p53 tumor-suppressor gene and environmental toxicants has proved to be an important lesson in the induction of cancer. Similarly, inheritance of the disabled WT-1 tumor suppressor gene is a well-known cause of a form of renal-cell cancer.
Cancer of the Prostate
Prostatic cancer is the most frequent cancer in men in the United States, with an annual incidence of 87.8 per 100,000. Its mortality is much lower, 23.8 per 100,000, but it is still the second most common cause of cancer death in men and the most common cause of cancer death in older men. The annual cost of diagnosis and care is more than $1 billion. In 1992, there were about 132,000 diagnosed cases of prostatic cancer and 36,000 related deaths. On the basis of autopsy studies, it is believed that more than 40% of men over 50 have undiagnosed prostatic cancer. It is projected that 10–12% of men alive today will have clinically manifested prostatic cancer and that 2% will die of it.
Blacks in the United States have a higher incidence of prostatic cancer than do whites in the United States or blacks in some other countries. For example, several studies show an odds ratio of 1.8-2.0 for prostatic cancer for blacks compared with whites in the United States. The incidence of prostatic cancer is markedly higher among blacks in the United States (100 per 100,000) than among blacks in Nigeria (10 per 100,000). Not only is the incidence of the disease higher, but prostatic cancer mortality is higher among blacks than among whites. That has been attributed to later diagnosis and increased risk among blacks, which might be related to genetic factors that influence the response to occupational or environmental exposures. The increased incidence of prostatic cancer associated with the migration of ethnic groups and the modulation of disease by diet support a xeno-biotic etiology for prostatic cancer. The importance of environmental factors has been brought into focus by differences in
the incidence of disease in various countries, and several hypotheses have been offered to explain the etiology. For instance, the low incidence of prostatic cancer and high incidence of breast cancer in males in Egypt have been attributed to increased estrogen levels in patients with bilharziasis infection of the liver (Bouffloux, 1980; El-Aaser et al., 1985; Sherif et al., 1980). A further reflection of environmental or occupational exposure comes from a 28-year Japanese study of latent prostatic cancer found in surgical specimens. During the study, an increase in overall incidence of prostatic carcinoma was observed; the number of high-grade (Gleason 3–5) cases increased, and the incidence of low-grade (Gleason 1–2) cases remained the same. These observations indicate that the changes are related to two diseases that develop along separate tracks.
In Hawaii, with its many ethnic groups, Kolonel (Kolonel et al., 1983) found a high correlation between dietary fat (both total fat and animal saturated and unsaturated fats) and incidence of prostatic cancer. The importance of saturated fats was substantiated by Hankin (Hankin et al., 1992), and of both saturated and unsaturated fats by Hursting (Hursting et al., 1990). Although cigarette-smoking as an etiologic factor for prostatic cancer has been considered, recent evidence suggests that the risk associated with cigarette-smoking, if any, is small (Hsing et al., 1990).
Increasingly, evidence indicates that prostatic cancer is a multistep process in which a series of events are required for a normal malignant cell to give rise to a fully malignant cancer cell. A large number of clinically undetectable cancer cells are present in a prostatic cancer. Although defined histologically as cancer, they might not have undergone all the steps to malignancy (Carter at al., 1990). That is an important consideration because widespread screening for prostate-specific antigen (PSA)—see Chapter 6—has led to an increase in detection of the occult form of prostatic cancer. Consequently, many men are being offered the option of surgical intervention or radiation therapy without proof from clinical trials that screening with PSA will enhance patient survival. Screening is detecting many cases of prostatic cancer, some of which might be biologically inactive.
Alterations in biologic markers that are associated with prostatic cancer reflect the subversion of growth control. The complex organization of the prostate is an example of the multiplicity of cellular interactions critical to homeostasis. The gland contains neuroendocrine cells, basement-membrane cells, associated stromal cells, and, of great importance for the continued reproductive success of the species, functional epithelial cells. Any of the several cellular types can abort normal regulatory control and participate in a new program of tumorigenesis. In the past, the influences of various hormones on the control of growth have been the primary focus of research and therapy. With the recognition that the tumor cells that progress are hormone-independent, increased attention has been given to other mechanisms of tumorigenesis.
The scientific challenge is to determine which of the cells are primary targets of xenobiotics; to understand the genetic factors that allow the transition from controlled paracrine growth to independent autocrine growth; and to define the early histopathologic malignant alterations and correlate them with early biochemical markers. Each of the histopathologic manifestations of disordered cell growth—proliferation, invasion, and metastasis—may be accompanied by subtle, but detectable and quantitative alterations. A key to the recognition of early biochemical profiles might come from study of cells surrounding those which have undergone malignant transformation, given the assumption that a gradient of changes can be detected.
In the adult male, homeostasis in the prostate is under the influence of androgens and is mediated by a series of diffusible peptides, or growth factors (EGF, transforming growth factor-(TGF) beta, insulin-like growth factors (IGFs), and fibroblast growth factor). The interaction of these and other growth factors is complex. Two examples are illustrative. First, TGF inhibits prostatic epithelial cell growth in the presence of EGF, but this inhibitory effect is abrogated in the presence of fibroblast growth factor, FGF, (McKeehan and Adams, 1988). Second, low concentrations of TGF are involved in a negative-feedback mechanism for the control of growth of prostatic stroma and epithelial cells. Obliteration of androgens—either surgically by orchiectomy or medically by use of a dihydroxysterone antagonist—results in prostatic-cell involution; this is followed by an increase in the prostatic growth factors, receptors, and cofactors, such as EGF and IGFs (Fiorelli et al., 1991a,b). In animals, administration of exogenous androgen decreases the production of TGF messenger RNA and its receptor to normal levels. The epithelial cells in turn regulate stromal cells through secretion of fibroblast growth factor, which influences angiogenesis, chemokinesis, and extracellular matrix-protein production (Gospodarowicz, 1991; Rifkin and Moscatelli, 1989). Alterations in one or more of those factors probably influence the metastatic potential of prostatic cancer or facilitate or inhibit the growth of transformed cells. Furthermore, it is not surprising that the growth of this highly regulated cellular system can be influenced by aging, by alterations in hormonal control, or by exogenous factors, such as xenobiotics or diet. One of the features of prostatic cancer is that the cells eventually become resistant to androgen suppression, proliferate rapidly and metastasize.
The prostate constitutes a unique environment that imposes special constraints on the development of biologic markers. Two of the most important constraints are that most men will eventually develop some degree of benign prostatic hyperplasia (BPH) and that in later life there is a high incidence of microfocal cancer. Implications of those facts are that there is almost always some confounding pathology and the problem is not to detect cancer, but rather to detect cancer that will become aggressive.
As with other organs, clues to the selection of markers lie in the molecules involved in the regulation of cell growth and differentiation, because it is the subversion of these systems that leads to cancer. The prostate is subject to complex growth and differentiation control. There are interacting stromal and epithelial elements and both androgen-dependent and androgen-independent mechanisms. The androgen-independent mechanisms seem to be operative during embryonic development; at puberty, there is a switch to androgen-dependent mechanisms. Clinically dangerous cancers seem to involve a reawakening of the androgen-independent mechanisms.
The biologic controls seem to involve several important principles: There are both positive (or stimulatory) and negative or (inhibitory) signals; an example of the former is platelet-derived growth factor, and an example of the latter is TGF although, to complicate the picture, TGF can inhibit epithelial cells and stimulate fibroblasts. Receptor processing at the membrane is complex, involving a receptor, a transmembrane domain, and an intracellular domain; the latter produces changes in cellular second messengers and modulates proteins, such as the ras p2l protein and other G proteins. Receptors are likely to be hormone-dependent. Extracellular-matrix proteins transmit growth or differentiation control signals to cells through linkages of integrins and other surface molecules. Autoregulation can liberate cells from exogenous controls. Growth peptides can be modified postranslationally; for example, TGF-alpha exists in the prepro form, which can be activated by appropriate proteases. The growth peptides can have actions on cells other than their main target cell.
It is difficult to know which factors are important to study clinically in these circumstances. For example, it can be difficult to determine whether the decrease in EGF or the increase in FGF is the important factor. Given the increasing number of growth factors identified, it is not feasible to examine all peptide growth factors in clinical studies. It is important to examine alterations in prostatic histopathology and cytology, as well as to correlate them with biochemical events, such as changes in the androgen receptor in the stroma, in androgen concentrations in plasma, in the amount being expressed by the testes or adrenal cortex, or even in the mechanisms that control the processing. One of the features of BPH is the increase in the stroma, compared with the endothelial or epithelial parts of the tissue, which results in such clinical manifestations as difficulty urinating or, in severe cases, hydronephrosis. Biochemical and structural analysis might provide clues as to the mechanisms in the increased amount of stromal formation in BPH.
The bladder has several potential targets for toxic xenobiotics. First, the protective glycosaminoglycan layer is subject to inactivation by amine compounds, which are among the most potent bladder carcinogens. Inactivation of
this layer by any one of several mechanisms increases the effect of later carcinogen exposure, whereas the administration of other glycosaminoglycans or heparin mitigates the carcinogenic effect of xenobiotics (Bodenstab et al., 1983). Second, primary and secondary alterations in the neuromuscular function of the bladder can produce changes in bladder physiology; any abnormality that causes obstruction of urine flow from the bladder alters bladder physiology and can lead to hypertrophy and altered neuromuscular function (Levin et al., 1990). Third, because there is a link between nerve function and inflammation, abnormalities in the nerves that innervate the bladder—similar to that described after exposure to organophosphate pesticides (Hohenfellner et al., 1992)—can produce an inflammatory response. Such a mechanism might also be involved in interstitial cystitis (described later). That is, considerable evidence suggests that at least some cases of interstitial cystitis result from urinary toxicants (Messing et al., 1992) that inactivate the glycosaminoglycan layer and expose the muscularis to various urinary solutes. It is not known whether this inactivation is a result of exposure to endogenous or exogenous agents.
Interstitial cystitis is a syndrome characterized by pain, urgency and frequency of urination, and cystoscopic abnormalities, all of which occur without known causes. Studies of the epidemiology of the disorder have not pointed to any one causal agent (Held et al., 1990; Koziol et al., 1993). One estimate, which might be low, placed the number of diagnosed cases in the United States at 43,500 in 1985. The reported incidence is much higher in women, but the reported incidence in men could reflect underdiagnosis. A subpopulation at special risk has not been identified, but Jewish women are overrepresented and black women are underrepresented. Although the etiology of interstitial cystitis is not known and there is no firm evidence that it results from xenobiotic exposure, the economic and human import of this condition merits attention and consideration of all possible causes.
ANIMAL MODELS OF INTERSTITIAL CYSTITIS
Several animal models of various aspects of interstitial cystitis have been reported recently. Ratliff and co-workers reported that some strains of mice develop an autoimmune cystitis that mimics the ulcerative form of interstitial cystitis, even to the point of showing bladder-permeability changes (Bullock et al., 1992). Stein and Parsons (1991) reported that the chronic instillation of protamine into rabbit bladders produced inflammation and a breakdown of the permeability defenses of the bladder that could persist after removal of the challenge and therefore provide an example of the ''toxic urine'' and "epithelial dysfunction" models. Buffington and co-workers reported that a spontaneous syndrome of cats manifested by frequent, apparently painful urination with sterile urine is apparently similar to interstitial
cystitis in showing decreased urinary glycosaminoglycan excretion.
Advances in understanding and using biologic markers should assist in identifying xenobiotics that are toxic to the urinary tract. The functional role of the urinary tract, including clearance of toxic substances from the blood, predisposes it to xenobiotic exposure and toxicity. Historically, identification of the type and amount of xenobiotic exposure has been difficult, often because of the interval between exposure and the onset of disease. Blacks and other minority groups, for reasons that are not entirely apparent, are at higher risk.
In diseases such as bladder cancer, xenobiotics associated with particular occupations are strongly implicated, and their mutagenic effects may be important. However, in kidney cancer and other renal diseases, a number of host factors, the typically low levels of exposure to multiple xenobiotics, and such confounding variables as smoking and genetic susceptibility often mask the epidemiologic significance of individual xenobiotics. A powerful approach toward unraveling the complexities of xenobiotic exposure is to integrate biologic markers of susceptibility with biologic markers of effect.
Some xenobiotics are known to cause acute renal failure. Heavy metals and organic solvents stand out in this regard. There are several well-established associations between xenobiotic exposure and the development of chronic renal failure, as exemplified by exposures to lead and cadmium. The association of bladder cancer and occupational exposure to aniline dyes serves as a paradigm for the potential adverse health effects of xenobiotics.
Environmental agents have also been implicated in the development of neoplasms in the kidney. Some of these can be facilitated by acquired or inherited genetic defects. The association of xenobiotic exposure and such conditions as prostate cancer and interstitial cystitis is less certain but merits attention.