At-line monitoring of a submerged filamentous bacterial cultivation using near-infrared spectroscopy

Abstract

The use of near infra-red spectroscopy (NIRS) to monitor a submerged filamentous bacterial bioprocess was investigated. An industrial strain of the filamentous bacterium Streptomyces fradiae was cultured in a 12 litre stirred tank reactor (STR) using a complex medium. This mycelial 4 phase (oil, water, gas and solid) system produced highly complex and variable matrices, therefore monitoring such a complex fluid with NIRS represented a considerable challenge. Nevertheless, successful models for four key analytes (methyl oleate, glucose, glutamate and ammonium) were built at-line (rapid off-line) using NIRS. In the present study, the methods used to formulate, select and validate the models for the key analytes are discussed, with particular emphasis on how the model performance can be critically evaluated. Since previous reports on NIRS in monitoring bioprocesses have either involved simpler matrices, or, in filamentous systems, have not discussed how NIRS models can be critically assessed, the emphasis in the present study on providing an insight into the modelling process in such a complex matrix, may be particularly important to the applicability of NIRS to such industrial bioprocesses.

Introduction

Near-infrared spectroscopy (NIRS) offers a number of potential advantages over current analytical methods used in bioprocess monitoring, which tend to be time consuming, labour intensive and costly. It is very fast, in that data is delivered within 1–2 minutes, often sample pre-treatment is not required, it is non-destructive and may allow the simultaneous analysis of several components [1]. This last point is particularly advantageous for bioprocess monitoring where many different analytes may need to be measured to allow optimisation of the bioprocess. From an industrial point of view, these potential advantages are attractive and could contribute to greatly enhanced process control options [2].

In previously reported studies various microbial processes have been monitored using NIRS, however, most of these have been limited to the “cleaner” more homogeneous cultivations of unicellular organisms, e.g. E. coli. Macaloney et al. [3] demonstrated the potential of NIRS for bioprocess monitoring by using it in the control and fault analysis of a high density E. coli unicellular process. The same group then went on to determine the concentrations of acetate, ammonium, biomass and glycerol in an E. coli fed batch fermentation [4]. Other types of microbial bioprocesses have also been investigated using NIRS, such as, lactic acid production whereby the concentration of 3 key analytes; lactic acid, glucose and biomass was monitored [5]. Ge et al. [6] monitored ethanol in a yeast fermentation non-invasively using NIRS. However, NIRS is not restricted to use in monitoring microbial cultures: it has also been applied to both mammalian [7] and insect cell cultures [8].

A common theme in many of these previous reports in the application of NIRS to submerged bioprocesses is the use of unicellular microorganisms, in chemically defined or soluble media, and, in the case of the cell culture processes, very low cell densities. From the spectroscopic viewpoint, these systems are ideal, and may permit simple transmission measurements to be made.

However, much of the worlds antibiotics, organic acids and enzymes are produced by the culture of filamentous microorganisms [9], thus, application of NIRS to such systems could have considerable benefits. These mycelial systems produce highly complex and variable matrices. They are often highly viscous, exhibit pronounced non-Newtonian characteristics, may contain solid(s), aqueous, oil and gas phases [10], and, in addition, the concentrations of reactants, products and by products can change considerably with time [2]. All of these aspects make the use of NIRS in monitoring such complex fluids highly challenging.

The first step in applying NIRS to monitoring a bioprocess is a calibration and validation which involves correlating NIR spectral data of samples of the bioprocess fluid with concentrations of the analytes of interest measured by conventional primary assay methods. This is commonly referred to as a model building process.

Although there have been reports on the application of NIRS to submerged mycelial bioprocesses [11], [12], these reports have often lacked adequate details of the models developed, and, in particular, on how the models were developed. This latter point is particularly significant as most reports on NIRS simply present a best model, but give no insights into the reasoning behind the selection and construction of the model.

If NIRS is to gain more widespread acceptance in monitoring bioprocesses, it is precisely these areas which must be more clearly elucidated, for the benefit of process operators.

In the present study, we cultivated an industrial strain of the filamentous bacterium Streptomyces fradiae for production of the antibiotic tylosin [13]. This process involves the use of two carbon sources (glucose and oil) and glutamate and ammonium as N sources. Spectroscopically, this process fluid, containing a mycelial organism, an oil phase (methyl oleate in this case), water and problematic air bubbles is very challenging [14]. Here we report on the formulation, selection and validation of models for these analytes in this complex four phase fluid with emphasis on how the models are developed, and how their performance is critically evaluated.