Effect of feed restriction on Hershberger and pubertal male assay endpoints

BACKGROUND: The Hershberger and male pubertal onset assays have been identified as possible Tier I screening tests to detect endocrine‐active compounds (EACs). Both tests rely on changes in reproductive and/or accessory sex gland (ASG) weights in young animals. Because chemical treatment may affect growth rate, the relationship between body weight and reproductive/ASG weights was examined using feed restriction (FR) to produce a targeted 10% decrease in body weight. METHODS: In the male pubertal onset assay, 23‐day‐old rats (12/group) were given ad lib feed or FR until euthanized at 45, 49, 52, 56, or 59 days of age. Despite a 10% body weight differential, pubertal onset was not significantly delayed and testes weights were conserved. Absolute prostate, ventral prostate, seminal vesicle, epididymides, and liver weights were decreased by FR. Relative weights for the prostate, ventral prostate, and seminal vesicles were similar to controls, but relative epididymides and liver weights still exhibited FR‐mediated changes. In the Hershberger assay, male rats (12/group) castrated at 36 days of age were given ad lib feed or FR in the presence or absence of testosterone propionate (T) from 46–55, 50–59, or 56–65 days of age. At 56, 60, and 66 days of age, rats were euthanized. In untreated animals, FR did not alter absolute ventral prostate, seminal vesicles, or Cowper's gland weights; however, absolute and relative weights of the levator ani‐bulbocavernosus muscles (LABC) were affected. In T‐treated animals, absolute organ weights (the ventral prostate, seminal vesicles, LABC, and glans penis) were relatively insensitive to FR. The weight of the Cowper's gland was affected only at 66 days of age. RESULTS: These data show that reproductive and ASG organ weight endpoints in the Hershberger and male pubertal onset assays can be influenced by FR levels that produce a 10% change in terminal body weight. CONCLUSIONS: The establishment of objective criteria for a positive or negative result is problematic due to the confounding effects of body weight on some endpoints. Furthermore, a 10% decrease in body weight seems to be excessive as a requirement for high‐dose toxicity in these assays due to possible indictment of agents that are not EACs, as well as potential masking of EAC effects coincident with body weight changes. Minimally, caution must be used in interpreting assay results in the presence of a 10% body weight change, recognizing the possible confounding effects of this degree of growth suppression. Birth Defects Res B 68:363–374, 2003. © 2003 Wiley‐Liss, Inc.