Temperature and salmonid reproduction: implications for aquaculture.

Fish reproduction is likely to be affected by increasing water temperatures arising from climate change. Normal changes in environmental temperature have the capacity to affect endocrine function and either advance or retard gametogenesis and maturation, but above-normal temperatures have deleterious effects on reproductive processes. In Atlantic salmon Salmo salar, exposure to elevated temperature during gametogenesis impairs both gonadal steroid synthesis and hepatic vitellogenin production, alters hepatic oestrogen receptor dynamics and ultimately results in reduced maternal investment and gamete viability. Exposure to high temperature during the maturational phase impairs gonadal steroidogenesis, delaying or inhibiting the preovulatory shift from androgen to maturation-inducing steroid production. There are also deleterious effects on reproductive development of female broodstock of rainbow trout Oncorhynchus mykiss and Arctic charr Salvelinus alpinus when they are exposed to elevated temperature. Less is known about temperature effects on male fishes but inhibition of spermiation has been observed in S. salar and O. mykiss. Among wild stocks, the response to elevated temperature will involve behavioural thermoregulation with consequent change in geographical ranges and the possibility of local extinctions in some regions. For domesticated stocks, containment in the culture environment precludes behavioural thermoregulation and aquaculturists will be required to develop adaptive strategies in order to maintain productivity. The most direct strategy is to manage the thermal environment using one or more of a range of developing aquaculture technologies. Alternatively, there is potential to mitigate the effects of elevated temperature on reproductive processes through endocrine therapies designed to augment or restore natural endocrine function. Studies largely on S. salar have demonstrated the capacity for synthetic luteinizing hormone-releasing hormone to offset the inhibitory effects of elevated temperature on maturational events in both sexes, but the potential for hormone therapy to provide protection during gametogenesis is still largely unexplored.

Habitat-related variation in reproductive endocrine condition in the coral reef damselfish Acanthochromis polyacanthus.

Spiny damselfish Acanthochromis polyacanthus are brood protectors with no larval dispersal stage, with the result that characteristics of local populations are likely to reflect local habitat conditions. In order to assess the possible effect of habitat on reproductive characteristics, spiny damselfish were captured by divers in 1999 and 2001 from reefs around Lizard Island in the northern section of Australia's Great Barrier Reef, chosen to represent a range of coral cover characteristics. Fish were bled underwater immediately after capture, then blood and fish were placed on ice at the end of the dive for transport to the laboratory where plasma was separated for subsequent measurement of testosterone (T) and 11-ketotestosterone (11KT) in males, and T and 17beta-estradiol (E2) in females. Ovaries from fish captured in 2001 were dispersed to isolate vitellogenic follicles, fecundity and follicle size were determined, then follicles were incubated in Leibowitz L15 medium alone or with human chorionic gonadotropin (hCG), to assess steroidogenic capacity. In 1999 there were significant site to site variations in plasma T and E2 levels in females, and in 2001, in E2 in females, and in T and 11KT in males. Highest hormone levels were recorded from sites of both low and high coral cover (a measure of presumptive habitat quality), but there was consistently low steroid production in fish from a site of high coral cover and fish density. An initial expectation that poor reproductive condition might be associated with degraded coral sites was not met. Vitellogenic follicles from fish captured in 2001 showed increased in vitro production of E2 and to a lesser extent, T, with increasing follicle size, and this was further augmented by treatment with hCG. Comparison of regression slopes of log E2 production versus follicle size showed that fish from sites where there were generally low levels of plasma steroids also had impaired in vitro steroidogenic capacity, and that this effect partially disappeared when follicles were stimulated with hCG. Reduced steroidogenic capacity was strongly associated with low fecundity, indicating that low in vitro and in vivo E2 production were reflected in reduced reproductive capacity. As the effect was most consistent at a site where fish density (and subsequent competition for planktonic food) was high, it is suggested that nutritional status associated with habitat characteristics may regulate reproductive endocrine condition in spiny damselfish. It is clear that local factors other than coral cover can generate site variation in reproductive performance.

Differential effects of temperature and maturity stage on hepatic estrogen receptor characteristics of Atlantic salmon.

In order to determine if elevated temperature during vitellogenesis had a detrimental effect on hepatic estrogen receptors of Atlantic salmon (Salmo salar), 3H-estradiol saturation binding analysis, using one- and two-site binding models, was carried out on extracts of hepatic cytosols from fish held at 14, 18 or 22 degrees C over the austral period of peak vitellogenesis (February to April). With one-site binding analysis, no temperature related difference in either receptor affinity (Kd) or number (Bmax) was found at each sampling point, but there was an apparent decrease in both affinity and number at each temperature over the period of the study. However, some analyses, notably at 22 degrees C during February, were best described using a two-site binding model. At this temperature and time, there was a clear separation of binding affinity into high and low components (Kd = 0.67+/-S.E. 0.05 and 20+/-S.E. 5.6 nM, respectively) (n = 4), which suggests that February was a critical time of temperature related hepatic sensitivity to estrogen. These results support those of other studies where we found that February was also a sensitive time with respect to temperature impairment of in vitro follicular estrogen synthesis, and the greatest period of in vivo temperature sensitivity.