Using cell cycle measurements with flow cytometry to predict the growth rate of Walleye Pollock Gadus chalcogrammus larvae

Abstract

Growth rate during early life stages can be an important factor in the recruitment process of marine fishes, and in this study we develop a methodology based on cell cycle measurements to predict the growth rate of Walleye Pollock (Gadus chalcogrammus) larvae. Results from cell-cycle analysis of muscle cell nuclei of laboratory-reared Walleye Pollock larvae measured with flow cytometry were used as covariates in a generalized additive model to predict the growth rate (growth in length per day, mm d−1) of individual larvae from the day exogenous feeding began (first feeding) to the time of capture (r2 = 0.79). Additional covariates used were temperature and standard length. A generalized additive model to classify a larva as fast- or slow-growing was also formulated using the same covariates. Validation testing with an independent set of 45 laboratory-reared larvae showed that 33% of the laboratory growth rates (i.e., growth rate based on age and size) fell within the 95% confidence interval of the predicted growth rates, and predicted growth rates were significantly less than otolith-derived growth rates. The growth classification model was more accurate than the growth rate model, correctly classifying the growth rate type (fast or slow growing) of 71% of the same set of larvae showing that flow cytometric cell cycle analysis may be better suited for classifying larvae as fast- or slow-growing rather than for predicting absolute growth rate. Predicted growth rates for larvae ≤11 mm in length collected from the Gulf of Alaska were within the range of published values for that area, and were not significantly different than corresponding otolith-derived growth rates. The growth rates of Gulf of Alaska larvae >11 mm were overestimated when compared to both otolith-derived growth rates and published values, and that may be due in part because those larvae were outside of the size range used to formulate the model. Predicted growth rate and growth classification of late-stage, field-collected Walleye Pollock larvae would be improved by adding larvae as large as 15 mm to the model to encompass the size range of larvae typically collected during spring ichthyoplankton surveys conducted in the Bering Sea or Gulf of Alaska. Flow cytometric cell cycle analysis offers promise as an alternative method for determining growth rate of fish larvae when otolith daily increments cannot be reliably counted.