Causes and consequences of Aspergillus niger pelleting behavior in industrial cultivation

The greater goal of this thesis is to understand the growth patterns of the filamentous fungus Aspergillus niger in submerged cultivations. The resulting morphology is reported to have impact on productivity of the diverse products that are manufactured by A. niger.
Existing mechanistic growth models and knowledge had to be consolidated and expanded with confirming trials and testing of various parameters to influence the morphology to create the ideal morphology for the specific production process. The topics included in this thesis cover morphology altering by pH, power input by agitation, presence and concentration of particles, initial conidia concentration and ion concentration.
The outstanding feature of this thesis is the tracking of morphology with particle size analysis by laser diffraction. Assuming spherical particles, first conidia and in the later course of the fermentation pellets, the Fraunhofer approximation was deemed to suffice in precision detail for displaying a volume based density function. To follow development of conidia aggregation, germination and pellet formation, the bimodal distribution could be split into modes of conidia and pellets to differentiate their respective particle size development.
Using the lab strain Aspergillus niger AB1.13, several growth conditions were evaluated in terms of influence on morphology engineering towards either freely dispersed mycelia or pelleted growth with defined diameter to prevent particle internal gradients. A conception is presented with the core control handle to determine when pellet formation takes place being pH. Initial cultivation at a pH of 3 prevented the first conidia aggregation step and the following aggregation could be controlled by shifting pH upwards to 5.5. The particle size could be tuned by adjusting parameters like power input and initial conidia concentration which possess inferior influence on particle size and biomass development compared to pH.
The resulting model was applied onto an industrial glucoamylase producing strain in courtesy of Novozymes A/S. The media was as close to the production environment as the settings of the model allowed. The result was that this strain behaved very differently from the suggested model, and all settings that were evaluated experimentally resulted in mixed morphology with both freely dispersed mycelia and pellets being present. In most cases, the pellets were also fragmenting before cultivation end thus making particle size engineering meaningless.
The model was then applied to the Aspergillus niger wildtype BO-1 strain to validate its general usefulness. Being the third strain to be tested, a third growth pattern could be detected. This indicates that morphology engineering probably must be evaluated for each strain separately.
The question arose if morphology matters to produce enzymes, in this case production of glucoamylase with BO-1. With introduction of different concentration of cellulose particles, the respective morphologies could be tailored. Employing the achieved morphologies as inoculum for cultivations, it was shown that freely dispersed BO-1 mycelia morphology is superior in producing glucoamylase despite oxygen limitation.
In conclusion, production processes should be checked for their morphology and the respective strain preference of producing enzymes.