THE RELATIONSHIPS AMONG REPRODUCTIVE END-POINTS IN SWISS MICE, USING THE REPRODUCTIVE ASSESSMENT BY CONTINUOUS BREEDING DATABASE

The database of Continuous Breeding mouse studies was evaluated to det ermine the relationships between the functional indicators of reproduc tion (pup measures) and the various necropsy endpoints collected for m ales and females. Of 72 chemicals studied, both males and females were affected in 33 studies, while females and/or conceptuses were affecte d in 7, Two compounds affected only males, 17 studies were negative, a nd in 13 studies with effects it was not possible to clearly determine the affected gender(s), Greater F-0 dam weight was correlated with in creased pup mass per litter; this relationship was strongest for the f irst litter, and weakest for the fifth litter. For both generations of treated females (F-0 and F-1), longer estrous cycles correlated with reduced numbers of pups; the relationship was stronger in F-0 than in F( )females and was not seen in controls. Sperm parameters had differe nt distributions in treated mice than in control mice. Fertility (tota l live pups/number of pairs cohabited) was reduced if there were > sim ilar to 15% sperm abnormalities or if sperm motility (moving/ not movi ng) was < approximate to 37%. Both of these relationships appeared to have thresholds. Epididymal sperm count in treated animals, however, w as linearly related to fertility, even within the control range, sugge sting strongly that other factors are important. Using both treated an d control data together, combining sperm count with motility could exp lain much (r = 0.77) of the variation in fertility; adding morphology did not significantly improve the correlation, The model was almost as strong using count and morphology, in which case adding motility did not strengthen the model, This analysis of these studies shows that wh ile some endpoints (e.g., random-estrous-cycle-point ovary weight) cor relate poorly with fertility, other necropsy endpoints (epididymal spe rm count and motility, estrous cycle length, and testis and epididymal weights) can be useful (though not complete) surrogates of overall re productive function, Indeed, over many studies, epididymal sperm count in treated animals correlates with fertility so well that even small reductions (approximate to 20%) in count result in reduced fertility, suggesting that mice may be better models of human fertility than was previously believed. (C) 1997 Society of Toxicology.