Growth Dynamics of Artemisia annua Hairy Roots in Three Culture Systems Yoo Jeong Kim, 1 Pamela J. Weathers, 2 Barbara E. Wyslouzil 1 1 Department of Chemical Engineering, Worcester Polytechnic Institute, 100 Institute Road, Worcester, Massachusetts 01609; telephone: 508-831-5493; fax: 508-831-5853; e-mail: barbaraw@wpi.edu 2 Department of Biology and Biotechnology, Worcester Polytechnic Institute, 100 Institute Road, Worcester, Massachusetts 01609 Received 26 June 2002; accepted 3 February 2003 DOI: 10.1002/bit.10685 Abstract: The transient growth of Artemisia annua hairy roots was compared for cultures grown in shake flasks and in bubble column and mist reactors. Instantaneous growth rates were obtained by numerically differentiat- ing the transient biomass measurements. Specific sugar consumption rates showed good agreement with litera- ture values. From the growth rate and sugar consump- tion rate, the specific yield and maintenance coefficient for sugar were determined for all three culture systems. These values were statistically indistinguishable for roots grown in shake flasks and bubble columns. In con- trast, the values for roots grown in bubble columns and mist reactors were statistically different, suggesting that sugar utilization by roots grown in these two systems may be different. By measuring respiration rates in the bubble column reactor we also determined the actual biomass yield and maintenance coefficient for O 2 and CO 2 . Together with an elemental analysis of the roots, this allowed us to obtain a reasonable carbon balance. © 2003 Wiley Periodicals, Inc. Biotechnol Bioeng 83: 428–443, 2003. INTRODUCTION Bioreactor design remains a key issue that must be ad- dressed before secondary metabolite production from hairy roots can be commercially successful. Developing a deeper understanding of how transformed roots respond to different reactor environments is particularly important, and the goal of our research is to quantitatively assess the relative effec- tiveness of liquid- and gas-phase reactors for culturing these plant tissues. In an earlier study (Kim et al., 2001), we found that the level of artemisinin contained in Artemisia annua hairy roots was significantly higher in the mist reactor than in the bubble column, and that the roots grown in shake flasks contained a negligible amount of artemisinin. On the other hand, a side-by-side comparative study of the overall biomass productivity showed that both the effective growth rate and a 2-week average growth rate were almost always higher in the bubble column than in the mist reactor (Kim et al., 2002). In this paper, we compare the transient growth of A. annua hairy roots in different reactor environments and de- termine the nutrient utilization rates. Nutrient utilization rates for hairy root cultures are usually reported as the ap- parent biomass yield—the ratio of the amount of nutrient consumed to the increase in biomass (Bhadra and Shanks, 1997; Kwok and Doran, 1995). The apparent yield can be further subdivided into the actual yield, a conversion effi- ciency of the nutrient consumed for biomass gain, and the maintenance coefficient, the nutrient required to maintain viability of existing biomass (Schnapp et al., 1991). These values can provide deeper insight into the bioconversion process and may clarify the advantages that one reactor design may have over another. Obtaining actual yield and maintenance coefficients requires following transient bio- mass closely, a task that is more difficult for hairy root cultures than cell suspension cultures. Thus, there is little data of this type available in the literature. As in our previous work, we chose a bubble column and a nutrient mist reactor as representative liquid- and gas- phase reactors, respectively. We cultured the YUT16 hairy root clone of A. annua in these two reactors as well as in shake flasks. We first compared the biomass productivity and the changes in the major nutrients observed in the re- actors directly to those observed in shake flask experiments as a function of time. We then estimated transient biomass in the reactors during culture, using correlations based on shake flask experiments, and determined the instantaneous specific growth rates. The specific growth rates were related to the specific sugar consumption rates to determine the actual biomass yield and the maintenance coefficients for sugar as a function of the culture environment. For the bubble column reactors, we also measured the specific res- piration rate, related this to the specific growth rate, and calculated the actual biomass yield and the maintenance coefficient for oxygen and carbon dioxide. Finally, using both actual yields and maintenance coefficients for sugar and oxygen from the bubble column reactors, we performed a carbon balance to determine the reliability of our results. Correspondence to: Barbara E. Wyslouzil, Department of Chemical En- gineering, 100 Institute Road, Worcester, MA 01609. Contract grant sponsors: NSF; NIH Contract grant numbers: BES-9414858; R21AI39170-01 © 2003 Wiley Periodicals, Inc.