Mathematical description of bikaverin production in a fluidized bed bioreactor Ma. del Carmen Cha´vez-Parga, Omar Gonza´lez-Ortega, Guadalupe Sa´nchez-Cornejo, Ma. de la Luz X. Negrete- Rodrı´guez, Guillermo Gonza´ lez-Alatorre and Eleazar M. Escamilla-Silva* Laboratorio de Biotecnologı´a y Bioingenierı´a, Departamento de Ingenierı´a Quı´mica, Instituto Tecnolo ´gico de Celaya, Av. Tecnolo ´gico y A. Garcı´a, Cubas s/n, 38010 Celaya, Gto. Me´xico *Author for correspondence: Tel.: +52-461-61-175-75, Fax: 52-461-61-177-44, E-mail: eleazar@iqcelaya.itc.mx Keywords: Bikaverin, fluidized bioreactor, Gibberella fujikuroi, immobilized, mathematical model Summary Growth of Gibberella fujikuroi in submerged cultures occurs as micelles or filamentous hyphae dispersed in fluid and pellets or stable, spherical agglomerations. Gibberella fujikuroi growth, substrate consumption and bikaverin production kinetics obtained from submerged batch fermentation were fitted to three different sigmoid models: two and three-parameter Gompertz models and one Logistic model. Growth fitting was used to compare between models and select the best one by means of an F test. The best model for describing growth was the two-parameter Gompertz model and was used for glucose consumption and bikaverin production fitting. Data from eight different schemes of fermentations were analysed and parameter estimation was carried out by means of minimization of residual sum of squares. Some characteristic values obtained with the two-parameter Gompertz model fit are: l ¼ 0.028 h )1 , Y x/s ¼ 0.1089 g substrate/g biomass, a ¼ 0.1384 g product/g biomass. Introduction Bikaverin is a red pigment with specific anti-protozoal activity against Leishmania brasiliensis (Balan et al. 1970; Desjeux 1992) and anti-tumour activity (Fuska et al. 1975). Additionally, bikaverin and its derivatives have a cytotoxic effect on in vitro proliferating cells of Erlich ascites carcinoma, Sarcoma 37 and leukaemia L-5178 (Fuska et al. 1975) and is a fermentation product of Gibberella fujikuroi or Fusarium sp. (Corn- forth & Ryback 1971; Kjaer et al. 1971). It has been chemically synthesized (Barton et al. 1976; Katagiri et al. 1981). Kjaer (1971), reported the chemical and spectroscopic characteristics of bikaverin, while the crystal and molecular structures were described by de Boer et al. (1971). The formation of bikaverin precedes that of gibberellin (Escamilla-Silva et al. 2000) and both secondary metabolites are produced from the primary metabolite acetyl-CoA. Bikaverin is synthe- sized via the polyketide route while gibberellin is synthesized through the isoprenoid pathway (Jones & Pharis 1987). The industrial production of these secondary metab- olites is done with cultures of mycelia in liquid (submerged) or solid substrate fermentation. Models of mould growth and metabolic production based on characteristics of mycelial physiology are important to understand, design, and control those industrial fermentation processes (Fredickson et al. 1979; Nielsen & Villadsen 1992; Bailey & Ollis 1998, Escamilla-Silva et al. 2001 a, b). In other words these models enable us to obtain information in a practical way, facilitating fermentation analysis, and can be used to solve prob- lems that may appear during the fermentation process. Furthering our research work on metabolite production (Escamilla-Silva et al. 1999, 2000, 2001a, b), we now offer an unstructured model based on the Gompertz and Logistic function is now offered to describe the kinetic studies of biomass and bikaverin production using G. fujikuroi in submerged batch fermentation in a fluidized bed bioreactor. Mathematical models for bikaverin production A great variety of mathematical models for batch fermentations are currently available (Fredickson et al. 1979; Humphrey 1979; Gutke 1980; Kossen & Oos- terhuis 1985; Aynsley & Ward 1990; Nielsen & Villadsen 1992; Larralde-Corona et al. 1994). In general, models used for describing growth may or may not be coupled with substrate consumption and have two or three adjustable parameters in the simplest cases. Equations (1)–(3) present models studied that are independent of substrate consumption. dx dt ¼ kxe l t  ax ð1Þ World Journal of Microbiology & Biotechnology (2005) 21:683–688 Ó Springer 2005 DOI 10.1007/s11274-004-3854-0