Work for Food – A Solution to Restricting Food Intake in Group Housed Rats?

  • Niina Kemppinen Laboratory Animal Centre, University of Helsinki
  • Anna Meller Laboratory Animal Centre, University of Helsinki
  • Kari Mauranen Department of Mathematics, University of Kuopio
  • Tarja Kohila Laboratory Animal Centre, University of Helsinki
  • Timo Nevalainen Department of Basic Veterinary Sciences, University of Helsinki and National Laboratory Animal Center, University of Kuopio


Rodents spend a great proportion of their time searching for food. The foraging drive in rats is so strong  that the animals readily work for food even when food is freely available. Commonly used ad libitum feeding  is associated with a reduced life span, increased incidence of tumours and risk of liver and kidney diseases.  It is also considered to be the most poorly controlled variable in rodent bioassays. The aim of this  study was to assess whether rats will gnaw wood in order to obtain food hidden in wooden walls, whether  this activity has a beneficial effect on controlling weight gain, and whether a typical diurnal activity rhythm  is maintained. A total of 18 BN/RijHsd and 18 F344/NHsd male rats were housed in either open or individually  ventilated cages (IVC), three rats in each cage. 10 of 36 were fitted with a telemetric transponder.  Four groups were used: two groups (diet board and plain board) with a maze made of two crossed aspen  boards, the third having a rectangular aspen tube. One maze was of plainboard, but the other included  drilled holes snugly loaded with food pellets, the “diet board”, such that the rats had to gnaw wood to reach  the food. The other two groups – and the controls – were fed ad libitum. The study used a crossover design  and the added item was changed every two weeks. Rats, added items, and amount of food left at the end of  the two week period were weighed. The statistical assessment showed that in terms of weight gain there  was a significant interaction both in IVC- (p = 0.005) and in open cages (p < 0.001) between the strains  and the group. In the F344 rats the diet board was more effective in controlling weight, but when combining  the strains, all comparisons with diet board were significant (p < 0.05). Use of strain and added item  as main effects, and age as covariate, showed that in the IVC-system there was a significant (p < 0.001)  interaction between the strain and the group, this effect being rather clear in the F344 rats in terms of  amount of food disappearing. In the open cage system, both strain and group were significant (p < 0.001)  factors; all three comparisons with diet board were significant (p < 0.001) in the amount of food disappearing.  In conclusion, the work-for-food approach appears to be a promising way of avoiding obesity  without causing untoward effects on diurnal activity in rats. Hence, the approach may have considerable  refinement and reduction potential.