|
The baffled microtiter plate: Increased oxygen transfer and improved online monitoring in small scale fermentations M. Funke, S. Diederichs, F. Kensy, C. Müller, J. Büchs
Biotechnology and Bioengineering (2009), Vol. 103, No. 6, pp 1118-1128
Shaken microtiter plates (MTP) are user friendly and inexpensive and therefore the favourite small-scale vessel for microbial screening and high throughput experiments. Despite its broad range of applications, little is known about culture conditions inside the microtiter plate.
Thus, oxygen limited cultivation conditions may disguise the effects of the variables being studied, resulting in the selection of suboptimal strains, media or culture conditions. Compensating for this in later process steps may be either expensive or unfeasible.
Altering the usual round well design is one possible option to enhance the oxygen transfer capacity (OTR) of a shaken vessel. This study focuses on the effects baffles and well geometry modifications have on the OTRmax in microtiter plates.
Potential problems when introducing baffles include splashing, “out-of-phase” behaviour of the liquid and wetting of the well seal. It is also preferable to maximise the liquid height in the well to allow optical analysis of the culture broth by online monitoring systems such as the BioLector. This instrument performs fibre-optic measurements through the well bottom and constant liquid height at all shaking frequencies is the key parameter.
In this study thirty cross-sections of different well geometries were systematically examined for maximum oxygen transfer capacity, liquid height at the well center and filling volume during orbital shaking (shaking diameter 3 mm).
OTRmax and specific mass transfer coefficient (kLa) were obtained from the sulfite system at increasing shaking frequencies (500 – 1000 rpm) and different filling volumes (200 – 600 µL) for a set of interrelated well geometries.
The well geometry clearly affects OTRmax. Introducing baffles results in higher oxygen transfer at shaking frequencies less than or equal to 800 rpm with any filling volume. The use of MTPs with baffles rather than round 48-well plates, could double the maximum oxygen transfer capacity to more than 100 mmol/L/h (kLa > 600 1/h). In contrast, pronounced baffles lead to a decrease in the OTRmax values at higher shaking frequencies (> 800 rpm), most likely caused by the out-of-phase behaviour of the liquid.
E. coli K12, requiring high amounts of oxygen (< 110 mmol/L/h) for unlimited growth, was used as the biological test system for all 30 well geometries. The bacterial growth rate increased with more distinct baffles, validating the strong correlation between well geometry and OTRmax.
However, when considering the filling height at the well centre, pronounced baffles disturb the liquid rotation at higher shaking frequencies and prevent it from climbing up the wall of the well. As mentioned previously, sufficient cover of the well bottom is essential for stable online monitoring.
A maximum filling volume of 1300 µL, without touching the well seal, was only possible (at 1000 rpm) in almost round-shaped wells. Pronounced baffles reduced the usable filling volume to 800 µL (33 % of well volume), while flower and star-geometries were the exception to the rule, with liquid volumes up to 1200 µL.
To determine the best MTP design for aerobic cultivation and monitoring microorganisms, a compromise has to be found between (i) strong baffles for a stable liquid height during orbital shaking, (ii) moderate baffles for sufficient oxygen supply and (iii) low baffles needed for adequate filling volumes without spilling the culture broth. The six-petal flower with a 5mm edge diameter proved to be the optimal well geometry.
Direct comparisons were made between microbial cultivations of different volumes in a flowerplate with baffles and a round-well MTP. The flowerplate culture with 1200 µL showed similar growth kinetics to the 800 µL culture in the round wells (1000 rpm shaking freuquency). Hence, the culture volume in the flowerplate can be approx. 400 µL higher before reaching an oxygen limitation level matching that of the standard well.
The outcome of this publication is an optimized MTP suitable for screening and process development with optimized oxygen supply and enabling meaningful experiments and scale-up to stirred tank reactors.
|
 |
