Abstract 391 Poster Session III, Monday, 5/3 (poster 162)

We have previously demonstrated that prenatal cocaine administration in our rodent model produces reductions in dopamine (DA) and glial cell line-derived neurotrophic factor (GDNF) in central (striatum) and peripheral (carotid body) neural structures, and that the magnitude of these reductions is dependent upon the location of the fetus within the uterus. However, our findings indicated that those fetuses exposed to the lowest cocaine levels in the uterus showed the greatest reductions in DA and GDNF in both striatum and carotid body. This led us to hypothesize that perhaps it was cocaine's vasoconstrictive actions (and resulting ischemia) at the uteroplacental junction that were responsible for both limiting the passage of cocaine to the fetus as well as globally reducing DA and GDNF. Therefore, using the radiolabelled microsphere technique for blood flow quantitation, we sought to determine whether cocaine produced a gradient of vasoconstriction to different segments of the uterus (each containing a fetus) that matched the previously observed gradient of DA and GDNF reductions. Pregnant Sprague-Dawley rats were exposed to cocaine (30 mg/kg, s.c., b.i.d.) or saline (1 ml/kg, s.c., b.i.d.) from gestational day (gd) 7 to gd14. On gd15, 30 min prior to tissue sampling, anesthetized rats were challenged with a single s.c. injection of cocaine (30 mg/kg) or saline followed 29 min later by microsphere infusion (Ru103). Uterine segments containing fetuses were excised, weighed and subjected to gamma counting followed by blood flow (ml/min/g tissue) calculation. There was a clear decrease in uterine bloodflow (F(1,6)=6.82, p<0.05) as a function of cocaine exposure, as well as a clear effect of uterine position (F(1,6)=4.56, p<0.05). There was no drug vs. position interaction which suggests that a uterine gradient (decreasing bloodflow from the cervical to the ovarian ends of the uterus) of similar slope is evident in both cocaine and control groups, but that cocaine further accentuated this bloodflow gradient. These data support our hypothesis that the pattern of cocaine distribution as well as the pattern of biochemical alterations from cocaine are likely due to its vasoconstrictive properties rather than any direct neurotoxic effects cocaine may have. These data also highlight the utility of the multiparous rodent model in determining cocaine's mechanisms of actions with respect to its teratologic and neurotoxic impact.

Supported by the Justin Carl Suth SIDS Research Fund.