Studies on the Mechanism of Toxicity of DDT and Polychlorinated Biphenyls: Disruption of Osmoregulation in Marine Fish.

In 1970 studies were initiated at the Mount Desert Island Biological Laboratory to explore mechanisms of toxicity underlying the high sensitivity of bony fish, teleosts, to organochlorine pollutants. The starting point was several reports that DDT [1 ,1, 1-trichloro-2, 2-bis (p-chlorophenyl) ethane] inhibited Na, K-ATPase (Na+, K+-activated adenosine triphosphatase), as well as the knowledge that this enzyme appears to play a central role in osmoregulation by marine teleosts. In the face of a desiccative environment, these fish maintain body fluid hypotonicity (Fig. 1) by drinking sea water, absorbing water coupled with salts across the intestinal epithelium, and eventually secreting the NaCl across the gill epithelium while retaining the free water. The primary driving mechanism in both intestine and gill is the Na pump with which Na, K-ATPase appears to be intimately involved. For fuller discussion, see (2) and (3). After demonstrating that DDT did, in fact, inhibit Na, K-ATPase activity in homogenates of intestinal mucosa and gill filaments from several marine teleosts (2), attention was focused on the sea water-adapted eel (Anguilla rostrata), in which osmoregulation has been well investigated. This eel can also adapt to fresh water and can be used for future DDT studies.

t To whom reprint requests should be sent: Dr. A. M. Guarino, Bldg. 10

Materials and Methods
Eels, Anguilla rostrata, and killifish, Fundulus heteroclitus, were captured in estuaries along the Maine coast. They were adapted to and maintained in sea water for about three weeks before use. Non-radioactive puriss grade p,p'-DDT was procured from Aldrich Chemical Co., Milwaukee, Wis. 14C-DDT was obtained from Amersham/ Searle, Arlington Heights, Ill. and had a specific activity of 24 mCi/mM. Aroclor 1221 was donated by Monsanto Co., St. Louis, Mo. Plasma milliosmolarity was determined by a standard freezing point depression technique, while Na+ and K+ were analyzed by means of a flame photometer. Levels of 14C-DDT were determined as described previously (4). The fractional 22Na space of isolated gills was measured according to the procedure of Kamiya (5) where eel gills are incubated for 1 hr at 150 in oxygenated sea water. Control gills were incubated with the solvent 0.5% N, N-dimethylformamide (DMF), and experimental gills were exposed to 50 ppm DDT in 0.5% DMF. Gill homogenates were assayed for Na, K-ATPase activity as described previously (2). Acute toxicity and osmoregulatory studies were conducted in killifish using a simple static system. For each experiment ten sea water-adapted fish (5 males and 5 females, each weighing about 5 g) were placed in an aluminum or enameled metal container holding 2 liters of sea water and maintained at 14-16°throughout the exposure time. In control containers 2 ml of ethanol were added to the sea water; in experimental containers Aroclor 1221 (mixture of PCBs with 21% chlorine content) or p, p'-DDT was first dissolved in the ethanol. These solutions formed a cloudy suspension when added to the sea water, and in addition some separation of the Aroclor was noted, making the actual concentrations unknown (Aroclor 1248 was not tested due to much greater separation). Blood was drawn by cardiac puncture with lightly heparinized glass capillary tubes. Whole blood from 3 to 5 fish was pooled and centrifuged. The 0.05-0.1 ml of serum was diluted, and the osmolarity and concentration of Na and K ions were measured as for the eel. * Mean ISE for 6 animals exposed to 1 ppm of 'IC-DDT for 6 hr.

Results
Ability to osmoregulate was followed in individual eels (about 100 g) placed in 150 sea water (2 liters) containing 0.1% ethanol and 1 ppm DDT, a level which is fatal by 10 hr. In Table 1 it is seen that, at 6 hr, plasma milliosmolarity of    Table 4), our eel data provide table one can see that, while much of this increase circumstantial evidence that disruption of osmowas due to plasma Na, plasma K also doubled in regulation is a primary toxic effect of DDT, i.e., these animals. In additional experiments with 1 the DDT levels in gut and gill of fatally-dosed fish ppm 14C-DDT, the fish had taken up about half were high enough to have affected osmoregulatory the radioactivity by 6 hr. The levels of radio-Na transport under in vitro conditions. Both water activity for some tissues are indicated in Table 2 absorption from gut sacs and Na content of and ranged from 7-22 ppm. Using isolated eel gills isolated gills are directly dependent on Na pump the fractional 22Na space was found to be increased activity and can be altered similarly by obtaining significantly (Table 3) by 50 ppm DDT. Data in a highly specific Na, K-ATPase inhibitor. More- Table 3 also shows that Na, K-ATPase activity over, from the published (3) dose-response curve was significantly inhibited (38%) by 50 ppm for gut homogenate, 7 ppm DDT in mucosal DDT. tissue should still inhibit Na, K-ATPase activity by 45%, provided, of course, that tissue content on a wet weight basis is equivalent to homogenate content based on added DDT. The data in this same curve also indicate that Mg-ATPase is less sensitive to DDT than Na, K-ATPase. Certainly, until more information is available, the present data would appear to provide a partial explanation as to why a whole body burden of less than 10 ppm DDT is often lethal to fish. It is also recognized that DDT may exert a primary toxic effect on the central nervous system; by 6 hr our fatally-dosed eels were hyperactive and easily stimulated to convulsions by touch. Such symptoms, however, could be secondary to alterations to osmolarity of plasma and cerebrospinal fluid.
Last summer, because of the chemical similarity between the insecticide DDT and the industrial PCBs (polychlorinated biphenyls) pollutants in the environment, acute toxicity and osmoregulatory effects were compared using killifish. The toxicity data, Table 5, show that, with respect to the acute lethal level in sea water, the Aroclor 1221 was less toxic than DDT by a factor of about 100. There was a "dose-response" in terms of the theoretical concentration in sea water and number of deaths for both Aroclor 1221 and DDT, but the actual dose which the fish received is unknown since the tissue levels were not measured.
The osmolarity and Na concentration of the serum (Table 6) consistently increased (toward sea water values) in fish exposed to lethal levels of either Aroclor 1221 or DDT. The Aroclor and DDT levels selected for these serum-composition studies were those that were potentially lethal to most of the fish, but low enough to have killed only some at the time (6 or 24 hr) samples were taken from those still surviving. With the higher levels tested, most of the increases in serum osmolarity and Na were statistically significant when compared to ethanol controls.
Also, serum osmolarity and K were elevated in all ethanol-treated controls (P < 0.05, except for 24-hr osmolarity); this elevation may have been due to the ethanol or to confinement of the fish in the experimental containers. Additional data ( Table 7) show that in the extreme cases, when blood was sampled before the fish died but after they had completely lost their righting ability and remained sideways, the serum osmolarity and

Discussion
Overall, these results suggest that lethal levels of Aroclor 1221 and DDT both decrease the ability of killifish to osmoregulate to about the same degree. Also killifish and eels seem about equally sensitive to DDT. Thus, inhibition of Na, K-ATPase by appropriate levels of Aroclor 1221 was anticipated. Using a homogenate of King O'Norway (Hemitripterus americanus) intestinal mucosa 250 ppm of Aroclor 1221 inhibited the Na,K-ATPase by 51% and 22% in two experiments. Moreover, recently-published work by Yap et al. (6) showed that several of the Aroclor mixtures inhibited the enzyme from fish brain and kidney.
Concerning the molecular basis for Na, K-ATPase inhibition by DDT and PCBs, one may speculate that these lipophilic compounds interact with the phospholipid-activating component of this lipoprotein enzyme (7,8). It is well known that many lipophilic compounds, e.g., cyclohexanone, inhibit the enzyme and evidence of proton interaction between DDT and a phospholipid, lecithin, recently has been reported (9). Moreover, other organochlorine insecticides, including endrin, also inhibit Na,K-ATPase from fish brain (10), and endrin has been observed to disrupt osmoregulation in both a marine and a fresh water teleost (11,12). Thus, it seems reasonable to predict that many organochlorine Environmental Health Perspectives pollutants will be found to disrupt osmoregulation in fish.