Evidence for effects of chronic lead exposure on blood pressure in experimental animals: an overview.

Information obtained in a number of experimental studies conducted over the last 40 years on the effects of lead on blood pressure is reviewed. Differences in animal species, age at beginning of exposure, level of lead exposure, indices of lead burden, and blood pressure effects of each study are reported. In several of the high-dose experiments, hypertension was observed, but nephrotoxicity of lead may have contributed to its development. Moreover, in other high-dose experiments, no hypertension was observed, and in at least one experiment, the evidence suggested that lead could reduce an elevated blood pressure. In contrast, the lower dose experiments consistently demonstrated a hypertensive effect. Overall, the data suggest a biphasic dose response. Establishment of an appropriate animal model to study blood pressure effects of lead will require careful assessment of dietary interactions with lead, unstressed blood pressure monitoring with standardized techniques, and documentation of biologically effective lead burden. Future research should examine lead exposure at more environmentally appropriate levels in order to determine the validity of associating this pollutant with blood pressure effects in the human population.


Introduction
The literature reporting effects of lead exposure on blood pressure in experimental animals is scattered among international journals and proceedings over at least a 40-year period; in many cases, the authors' knowledge of similar or divergent experimental findings was not noted. It is not surprising that the results sometimes appear inconsistent. The purpose of this paper is to review these studies, examining the lead dose, route of exposure, age at initiation of exposure, when the effects were first observed, how long the animals were followed, and any indication of the internal lead burden. These findings will orient the reader to the papers in this issue, which discuss several possible targets of lead's cardiovascular toxicity.
An underlying question for most researchers is whether the lead dose administered should be of sufficient magnitude to induce nephrotoxicity prior to the onset of hypertension or if hypertension may develop prior to renal damage. Over the years, the experimental lead dose range for rats has varied between 70 mg per day (1,2)  lead in drinking water (3). Clearly, in the former case it is likely that renal pathology can be, observed, while in the latter, only subtle renal changes are observed.
The route of administration varies somewhat among studies, but it is usually oral (by gavage, in drinking water, or mixed in the food). The amount absorbed varies widely and depends on other nutritional factors and the age of the animal. The length of time for which exposure continues has been between 2 months and 2 years. The length of exposure before initiating monitoring and for continuation of monitoring varies greatly; in fact, some investigators have not followed a large enough group of control and experimental animals to detect statistical significance between the groups. Blood pressure monitoring is usually performed indirectly (by tail cuff plethysmography), either on the anesthetized or unanesthetized animal, or directly by insertion of an arterial catheter and detection by a pressure transducer. For repeated measurements in the same animal, the plethysmography is preferable but has the disadvantage that the animals are prewarmed, which is enough of a stress that some lead-exposed rats did not survive this procedure (4). If the animals are unanesthetized, they must be conditioned to the restrainer so that unstressed values are obtained. Anesthetics also affect the blood pressure, so that this stress must also be evaluated. At the termination of the exposure period, it is quite helpful to have a direct blood pressure measurement to correlate with the indirect measurements previously obtained.
Species studied include rats, dogs, and pigeons. Sex of the animal may have a role in determining susceptibility to cardiovascular toxicity of lead, although this has not been clearly established. Animal housing and handling conditions are especially important if the animals are to be monitored in the unanesthetized, unstressed state. Indicators of the body burden of lead (such as blood or kidney lead concentrations) have frequently not been performed; if so, the extent of a doseresponse relationship between the amount of lead administered, blood lead, and blood pressure changes could be quantified. bRats exposed to 10 mg lead served as controls.
Experimental Findings Table 1 presents a brief summary of the literature citations in which rats were administered relatively high lead doses. The experiments of Griffiths and Lirdauer (1) and Diaz-Rivera and Horn (2) used similar protocols, administering 70 mg of lead per day by gavage to rats; this dose produced hypertension in animals that survived the dosing period. Nephrotoxic effects were clearly observed in the animals that were examined. Pardoe (5) and Padilla et al. (6) also used similar protocols (30 to 40 mg for the initial dose, with even higher doses in the subsequent exposure period). The lead exposure did not elevate blood pressure with respect to controls; there were marked changes in renal histology, although there were no decrements in several measured parameters of renal function. Pardoe reported that the tubular transport maximum for paminohippuric acid (PAH) was elevated by 20% at 3 months, indicating an increase in tubular excretory function that paralleled the increase in kidney weight. This effect appeared to be reversible in the later period of the experiment. Of the earlier experiments, Padilla et al. (6) are the only authors to report blood lead concentrations; however, the validity of the blood lead analysis appears questionable considering both the high control values and the relatively low values in the exposed group. The paper with the most relevance for comparison to more recent studies is that of Cottier et al. (7). Lead phosphate in a dose of 20 mg per week ad-ministered SC elevated systolic pressure above 130 mm Hg. Lead phosphate contains 76% lead, which would be equivalent to 15 mg lead per week. Twice this dose (40 mg/week) was "too toxic." In the report (to be described later) by Victery et al. (8), a rat consuming 100 ppm lead in drinking water might average 30 mL/day x 7 days/week, or 21 mg/week. Extrapolation between these two papers should be performed with caution to the different routes of administration (SC vs. oral). There are no reported blood lead values in the Cottier paper.
The last paper in this group (9) used a very different treatment protocol: 1% lead was administered for 6 weeks and the blood pressure and renal function tests were performed 19 weeks later. Mean blood lead concentration exceeded 150 ,ug/dL at the end of exposure and fell to 24 tg/dL at the time of the procedure. Blood pressure was elevated by 13 mm Hg and there were decreases in whole kidney glomerular filtration rate (GFR) and single nephron GFR, as well as decreases in renal plasma flow. Table 2 presents the more recent experimental reports of blood pressure effects when lower levels of lead exposure was used beginning as early as in utero. A similar presentation of this information is incorporated in the comprehensive review paper of Sharp et al. (10). Perry and Erlanger (3) report initial results from a large series of studies. The most striking findings from among the published reports from this group are that significant increases in systolic blood pressure are found in rats exposed to as little as 0.1 ppm lead in the treatment drinking water, with a dose-dependent relationship holding for the higher exposure regimens (up to 5 ppm). All animals were fed a rye-based diet with low levels of trace minerals; mineral supplements were present in the drinking water. However, no blood lead values were obtained. In the early 1980s, a number of similar studies were performed by a group of Italian scientists and by Vander and colleagues at the University of Michigan. Rats were exposed to doses of lead between 5 and 500 ppm in the drinking water. The studies reported by lannaccone et al. (11) and Carmignani et al. (12) used 50 ppm beginning at weaning to 6 months of age. Under pentobarbital anesthesia, these animals had an approximate 50-mm Hg increase in blood pressure compared to controls and were studied for a number of other cardiovascular parameters, some of which are discussed in the papers published in this volume (13)(14)(15)(16). Blood pressure was not followed over time, so it is not known when the pressure change occurred or whether there was evidence of a dose-response relationship. Blood lead values for the 50-ppm lead-exposed group were approximately 40 ,ug/dL, compared to 14 ,ug/dL for controls, as reported by lannaccone (10).
In the Michigan studies reported by Victery et al. (8,1 7), 0, 5, and 25 ppm lead in drinking water was administered to rats beginning in utero and continuing to 5 to 6 months of age in one study, and 100 and 500 ppm doses were used in the other. Blood pressure was monitored approximately every 2 weeks in conscious, restrained rats by tail cuff plethysmography. Although no changes in blood pressure were detected in the lower dose experiments, there were significant changes in the renin-angiotensin system of 25-ppm animals (15). Blood lead concentrations averaged 5.6 and 18.2 pg/dL for the 5-and 25-ppm animals, respectively. Exposure to 100 ppm (but not 500 ppm) produced a significant 17-mm Hg elevation in blood pressure beginning at 3 1/2 months of age and continuing until 6 months of age. Renal function was not evaluated. Kidney weights were significantly elevated at 100 ppm, and some morphologic changes were observed at 500 ppm. Blood lead levels averaged 40 ,ug/dL for 100 ppm and 70 gg/dL for 500 ppm. Table 3 presents the results of two recent studies evaluating the effects of lead on the cardiovascular system of rats of the spontaneously hypertensive strain (SHR). The paper by Wiecek et al. (18) reports a lead dosedependent decrease in blood pressure in the SHR rats and control (WKR) rats, and the authors conclude that their findings are contradictory to the findings of Victery et al. (8,17). However, the findings are not contradictory because it appears that these authors have made a mistake in calculation of the lead dose used by Victery et al. and actually treated their animals with 2.5 to 50 times higher doses than our hypertensive dose. This is verified by their statement that blood lead values in their rats were "much higher" than those reported by Victery et al. (8,17).
In another study, Evis et al. (4) administered lead (at 250 and 1000 ppm) to SHR and control rats. The authors evaluated a number of cardiovascular parameters including conscious and anesthetized blood pressure. The higher dose of lead significantly increased blood pressure in both groups of rats, which had reported blood lead concentrations of 40 ,ug/dL. Interestingly, lead accelerated the development of hypertension in the SHR animals and also increased their sensitivity to the warming required for dilation of the tail artery used for measuring conscious blood pressure (seven of eight animals died).
Results obtained in other species (pigeons and dogs) are summarized in Table 4. In pigeons (19), a low dose of lead (in combination with a number of other dietary parameters) produced a significant elevation in  blood pressure, which could be alleviated by increasing the calcium concentration in the drinking water. The exposure of two dogs to very high levels of lead did not invoke an increase in blood pressure but did produce acute episodes of lead poisoning (20). In a paper by Fine et al. (21), three young dogs received a low dose of lead (1 mg/kg daily, blood lead concentration of 36 ig/dL) and were found to develop a persistent (from 10 days to 20 weeks) 12-mm Hg increase in mean arterial pressure when compared to their paired control dogs. Additional determinations of the renin profile show that renin is initially elevated (at 28 days exposure) but then returns to normal levels while blood pressure remains elevated.

Conclusions and Directions for Future Research
This short overview illustrates the complexity of interpreting data from experimental models that have been used to evaluate the relationship of lead exposure at different levels to changes in blood pressure. Based on the collective results of the studies reviewed, it appears that lead administered at relatively nontoxic levels produces an elevation in blood pressure that is sustained over a considerable portion of the animal's life span. In contrast, high doses are not consistently associated with hypertension and may actually lower blood pressure. Could this be related to certain protective or adaptive mechanisms reducing lead's effects or to an effect on the general health of the animals? In any case, these high doses may not be relevant to the human situation. For the experimental literature to be specifically relevant to human environmental exposures, findings of alterations in cardiovascular parameters in the blood lead range of 5 to 40 ,ug/dL would be most suitable.
A number of blood pressure regulatory control systems have been shown to be altered by various lead exposure protocols. These include several hormone and neurobehavioral regulatory systems, changes in vascular smooth muscle reactivity, cardiac muscle contractility, changes in cell membrane cation transport systems, and possible effects on vascular endothelial cells. Changes in at least one of these systems have been reported at all of the lower lead dose ranges, with some exceptions at higher doses. Which systems are most important contributors to a significant, persistent elevation in blood pressure have not been established definitively. There may be a number of sequential changes that occur and are no longer observable depending on the time of examination. However, it appears that there are sufficient experimental data available to confirm that chronic lead exposure at low levels can affect blood pressure levels in a consistent fashion.
Inevitably, the dose range and shape of the doseresponse curve become issues for debate because they are relevant to the question of whether or not the lead exposure range is comparable to that of the general human population or only to those individuals occupationally exposed to lead. It appears from some experimental studies that the blood pressures measured at higher doses are lower than would be expected, suggesting a biphasic dose-response curve.
The exact dose-response range for this finding has not been well established, primarily because of different methodology used by each group of investigators. A standardized protocol for preparing and administering the lead dose, dietary components, rat strain, and age-at-initiation of exposure should be developed. Methodology for measuring blood pressure on a continuing basis in the conscious animal has improved so that tail cuff measurements can be performed without warming, and recordings can be obtained on several animals at once. These techniques should be applied in a carefully controlled setting along with validation of the biological markers of lead exposure.
Once these conditions have been well established, other important questions should be addressed. These include: Can the effect on adult blood pressure be established by an early exposure period at environmentally appropriate levels? Can the effect on blood pressure be reversed by removal of the lead source and reducing the lead burden? Answers to these questions would be valuable because they would help to evaluate the need to continue reductions in environmental lead exposure to the human population.