Biotechnol. Prog. 1990, 6,370-375
370
Fluidized Bed Adsorption for Whole Broth Extraction+ F. P. Gailliot,* C. Gleason, J. J. Wilson, and J. Zwarick Merck & Co., Inc., P.O. Box 2000, Rahway, New Jersey 07065
Fluidized bed adsorption using a high-density synthetic resin has proven to be an invaluable technique for separating novel compounds from unfiltered fermentation broths during the very early stages of fermentation development, where product concentrations are typically in the parts per million range. Previous initial downstream processing strategies consisted of cell separation from whole broth or direct extraction with waterimmiscible solvents, both of which resulted in lengthy time cycles, conflicts with existing operations requiring the use of high-cost centrifugal separators, and environmental/ solvent recovery concerns. Laboratory and subsequent pilot plant process development work along with concomitant improvements in yield, quality, and time cycles are presented for one of several fluidized bed processes piloted in Merck’s Natural Product Isolation facility.
Introduction Merck Sharp & Dohme Research Laboratories, a division of Merck & Co., Inc., has a modern Natural Product Isolation Pilot Plant (Gailliot et al., 1987) located at its Chemical Engineering Research and Development Facility in Rahway, NJ. Responsibilities of the Natural Product Isolation (NPI) Pilot Plant include developing isolation processes for novel compounds produced via fermentation, investigating new technologies related to downstream processing of fermentation broths, and preparing large quantities of investigational compounds for safety and toxicity testing, drug efficacy, and clinical trials prior to full-scale production. Material demands for new drug candidates can range from a few grams for structural determination to hundreds of kilograms for evaluation in clinical programs. Isolating quantities of a new compound is a particularly challenging task when fermentation product concentrations are in the low parts per million range before high-yielding mutants are available. The first task in developing an isolation process for a new extracellular product centers on the selection of an appropriate procedure for handling the solids present in the broth. In the 1960s, centrifugation or filtration was utilized to remove solids prior to adsorption or extraction. Colloidal solids and viscous liquors frequently made these methods both costly and inefficient. Increased cost for purchase and disposal of large quantities of filter aid coupled with product loss resulting from the ineffectiveness of washing product from the solids eventually led to the implementation of whole broth solvent extraction technology in the late 1970s. Whole broth solvent extraction (WBE) eliminated the often difficult solidliquid separation step by direct extraction with an immiscible solvent, followed by subsequent phase separation via newly developed high-speed centrifugal separators. Merck’s NPI Pilot Plant is equipped with a complete range of modern high-speed centrifugal equipment for liquid-liquid and liquid-solid separation and has successfully developed whole broth solvent extraction processes for several commercial products. Although whole broth extraction solved the difficult solid-liquid separation step of the 1960s,there are several disadvantagesassociated + Paper presented at AIChE 1989 Annual Meeting, San Francisco, CA, November 5-10, 1989.
with whole broth solvent extraction, namely, expensive centrifugal equipment, large volume tankage requirements, voluminous solvent usage, solvent recovery, and spent stream stripping costs. In 1988 the NPI Pilot Plant began investigating process alternatives to whole broth solvent extraction. The need to develop these alternatives was prompted by the facility’s requirement to handle all of the potential new product candidates in the research pipeline that had low product concentrations (
370
Fluidized Bed Adsorption for Whole Broth Extraction+ F. P. Gailliot,* C. Gleason, J. J. Wilson, and J. Zwarick Merck & Co., Inc., P.O. Box 2000, Rahway, New Jersey 07065
Fluidized bed adsorption using a high-density synthetic resin has proven to be an invaluable technique for separating novel compounds from unfiltered fermentation broths during the very early stages of fermentation development, where product concentrations are typically in the parts per million range. Previous initial downstream processing strategies consisted of cell separation from whole broth or direct extraction with waterimmiscible solvents, both of which resulted in lengthy time cycles, conflicts with existing operations requiring the use of high-cost centrifugal separators, and environmental/ solvent recovery concerns. Laboratory and subsequent pilot plant process development work along with concomitant improvements in yield, quality, and time cycles are presented for one of several fluidized bed processes piloted in Merck’s Natural Product Isolation facility.
Introduction Merck Sharp & Dohme Research Laboratories, a division of Merck & Co., Inc., has a modern Natural Product Isolation Pilot Plant (Gailliot et al., 1987) located at its Chemical Engineering Research and Development Facility in Rahway, NJ. Responsibilities of the Natural Product Isolation (NPI) Pilot Plant include developing isolation processes for novel compounds produced via fermentation, investigating new technologies related to downstream processing of fermentation broths, and preparing large quantities of investigational compounds for safety and toxicity testing, drug efficacy, and clinical trials prior to full-scale production. Material demands for new drug candidates can range from a few grams for structural determination to hundreds of kilograms for evaluation in clinical programs. Isolating quantities of a new compound is a particularly challenging task when fermentation product concentrations are in the low parts per million range before high-yielding mutants are available. The first task in developing an isolation process for a new extracellular product centers on the selection of an appropriate procedure for handling the solids present in the broth. In the 1960s, centrifugation or filtration was utilized to remove solids prior to adsorption or extraction. Colloidal solids and viscous liquors frequently made these methods both costly and inefficient. Increased cost for purchase and disposal of large quantities of filter aid coupled with product loss resulting from the ineffectiveness of washing product from the solids eventually led to the implementation of whole broth solvent extraction technology in the late 1970s. Whole broth solvent extraction (WBE) eliminated the often difficult solidliquid separation step by direct extraction with an immiscible solvent, followed by subsequent phase separation via newly developed high-speed centrifugal separators. Merck’s NPI Pilot Plant is equipped with a complete range of modern high-speed centrifugal equipment for liquid-liquid and liquid-solid separation and has successfully developed whole broth solvent extraction processes for several commercial products. Although whole broth extraction solved the difficult solid-liquid separation step of the 1960s,there are several disadvantagesassociated + Paper presented at AIChE 1989 Annual Meeting, San Francisco, CA, November 5-10, 1989.
with whole broth solvent extraction, namely, expensive centrifugal equipment, large volume tankage requirements, voluminous solvent usage, solvent recovery, and spent stream stripping costs. In 1988 the NPI Pilot Plant began investigating process alternatives to whole broth solvent extraction. The need to develop these alternatives was prompted by the facility’s requirement to handle all of the potential new product candidates in the research pipeline that had low product concentrations (
