Hatching Fish—When Should Animal Tracking Begin?

Zebrafish inclusion in IACUC protocols (new April 2015)

PHS policy requires that all live vertebrate animals be included in the EU IACUC approved protocol. The NIH OLAW considers fish species to be “live vertebrate animals” at “hatching.” It is recognized that this is an imprecise stage in zebrafish and hatching may occur naturally or as part of a procedure such as dechorionation.

These new requirements were sent out to all researchers and staff listed on zebrafish protocols. The response was immediate and palpable. Most investigators had heard of the hatching recommendation but had not expected that it would apply to them as there was already an established written standard in place, describing the time of inclusion as 7 dpf based on the requirement for feeding^2,3 and normal entry into the facility system, where counting of animals naturally occurs. Investigators discussed with the IACUC director that they had not been consulted regarding these changes and that to their knowledge the zebrafish community also had not been consulted regarding the OLAW recommendation. One concern was that all animal storage would have to be inspected and daily environmental, population, health, and mortality documentation would have to be maintained with the same criteria normally applied for other animals covered under the IACUC program. All spaces where live animals are held would have to be reviewed based on criteria normally applied to surgical or live animal procedure space. Because there are now several zebrafish locations on the EU campus, these changes would require considerable effort to apply and implement. They challenged that the hatching rule was not based on the science regarding the comparative development of zebrafish and that the traditional standards of practice had developed over years of the successful use of zebrafish as a model system.

Point by point they argued that fish are not comparable to newly hatched chickens as a standard for independence. Unlike chicken/avian species, on which OLAW guidelines were based,^4,5 zebrafish are essentially a free-swimming larvae before 7 dpf because of the internal yolk supply that eliminates the demand for exogenous feeding before that time. Chicken also lay fewer eggs, and brood to protect these and provide warmth for development, during which time the yolk is used up, which requires that chicklets feed after hatching. Zebrafish lay large number of unprotected eggs, which hatch early in development relative to the chicklets, and are capable of escaping predation, not requiring brooding or protection from the parent. However, because they hatch with a yolk sack and have not fully developed all organs that would allow for exogenous feeding, they cannot feed at the hatching stage and attempting to feed them would likely result in considerably reduced water quality and environment. Thus applying the definition of independence for fish based on bird biology makes no sense and is even detrimental for animal welfare.

In addition, embryos/fry in early stages of development may be dechorionated for experimental purposes and develop normally in the petri dish environment. ⁶ The zebrafish investigators also reminded the IACUC that pain categories would also not apply before 8 dpf and so do not demand the necessity of protection. ⁷

They further appealed that the current expectations represented a significant increase in cost, effort to count, document, monitor, and report with no obvious benefit for animal care or research. The possibility of research being out of compliance for any of the new requirements did not appear to be justified by any benefit from this policy change, they argued.

As for documenting numbers, investigators argued that it is simply not practical because of the large numbers involved on a regular basis. Survival at this early stage is highly variable even within a wild-type population, and health issues related to mortality data are not practically of general concern for reporting purposes. The zebrafish mating has wide variation from day to day, and excess numbers of the early stages are collected to assure work progresses and are terminated by the 7 dpf stage. It is at the nursery point of entry that numbers are naturally and consistently recorded to assure population density is appropriate for the feeding and housing system used. In fact, the effort to track and count all animals or to feed them unnecessarily before 7 dpf took away time from animal caretakers that they would have normally used to provide a clean, biosafe environment.

Applying animal housing standards to the laboratory environment would also likely result in sequestering all live zebrafish work to a centralized, dedicated facility environment, causing most current arrangements to become out of compliance. The result would be a significant increase in the cost of zebrafish research, thus eliminating the capacity for small-scale discovery-based or exploratory start-up projects that have been the common starting point for the adoption of zebrafish as an alternative to mammalian studies. Such centralization would also limit the access of researchers to embryo or larval stages.

The EU IACUC director anticipated some resistance to change, but the responses to the OLAW hatching recommendation and concerns regarding the downstream implications were reasonable and the new regulation seemed to result in additional confusion and conflict where no problems had existed before. It seemed that in this case, the traditional system fits the natural biology and the normal research mode. The director could not defend the proposed policy change based on the reasoning of principles shared by the research community and was concerned that regulation without reason breeds contempt and distrust. At the same time, animal welfare requirements resulting from inclusion of larvae before 7 dpf suggested that their program might not be in compliance with the GUIDE (http://www.nap.edu/catalog/12910/guide-for-the-care-and-use-of-laboratory-animals-eighth) with regard to regulating, monitoring, and documenting zebrafish use and care at the early embryonic stages.

The OLAW interpretation of PHS policy regarding the timing of animal welfare coverage has potential implications that seem to run counter to traditional organization and work in most zebrafish research laboratories. This policy interpretation will result in significant shift of effort from research activities to count and document activities with no apparent benefit to animal welfare at a time in zebrafish development when feeding, management of pain, distress, or suffering are not required for biological reasons. The IACUC staff charged with implementation experience a similar burden with no apparent benefit to animal welfare.

How should the EU IACUC proceed?

Is this OLAW interpretation of PHS policy justified as applied to zebrafish? In the case of zebrafish, hatching seems a biologically arbitrary and imprecise time point for welfare documentation. Should zebrafish researchers be required to have an IACUC protocol that covers all life stages? When should animal counting and tracking begin? When should feeding be required? How can the EU IACUC, EU researchers, the zebrafish research community, and OLAW work together to reconsider the hatching policy in a way that is optimal for research, account properly for the biology of this model organism and the animal welfare standards that are established to improve both research and animal care as established by the GUIDE?