Sources of error in the estimation of Leydig cell numbers in control and atrophied mammalian testes.

The effects of assuming (i) that testicular tissue shrinks equally regardless of species or treatment at fixing and processing, (ii) that all Leydig cells in a given testis have spherical nuclei of identical size, and (iii) that testicular volume (i.e. the reference volume) is constant regardless of species or treatment, on the estimation of Leydig cell numbers in mammalian testes were investigated. This was accomplished by comparing the results of stereological analyses of Leydig cell numerical density and Leydig cell number in control testes of hamster, guinea-pig, and rat and in atrophied testes of hamster, and rat, obtained via the disector method which is unbiased with respect to the particle shape under study, and the Floderus equation which assumes that the particles under study are identical spheres. In control hamster, and also in guinea-pig, the effects of the three assumptions on the estimates of Leydig cell number per testis were negligible, because in these two treatment groups, the total shrinkage of testis tissue at fixing and processing (ST%) was low, Leydig cell nuclear profiles were circular in section, and the average volume of a testis was close to unity (i.e. 1 cm3). By contrast, in hamsters, and rats with atrophied testes, these assumptions produced incorrect estimates in Leydig cell number per testis, because the ST% was high, the majority of Leydig cell nuclear profiles were pleomorphic, and the average volume of a testis was lower than control. In summary, this study documents that the assumptions of equal shrinkage in testis tissue at fixing and processing, a constant testicular reference volume, and spheroidal shape of Leydig cell nuclei may contribute significant errors in estimates of Leydig cell number in mammalian testes. The magnitude of the errors introduced by these assumptions depends upon the species and the experimental treatment.