- Email: [email protected]
Microtiter plates as mini-bioreactors: miniaturization of fermentation methods
Review
TRENDS in Microbiology
Vol.15 No.10
Microtiter plates as mini-bioreactors: miniaturization of fermentation methods Wouter A. Duetz Enzyscreen, Biopartner Center Leiden, Wassenaarseweg 72, 2333 AL Leiden, The Netherlands
In the past decade, the use of microtiter plates for microbial growth has become widespread, particularly in industry. In parallel, research in academia has provided a thorough insight into the complex relation between well dimensions, culture volumes, orbital shaking conditions and surface tension on the one hand, and oxygen-transfer rates and degrees of mixing on the other. In this review, I will discuss these fundamental issues and describe the current applications of microtiter plates in microbiology. Microtiter plates can now be considered a mature alternative to Erlenmeyer shake flasks. Progression from Erlenmeyer flasks to microtiter plates The adoption of the shake flask in the 1880s as the vessel of choice for the growth of microorganisms in liquid cultures was pragmatic in the sense that the Erlenmeyer flask had been invented 20 years earlier, for use in chemical laboratories. More than a century later, the relatively large culture volumes in Erlenmeyer flasks gradually became obsolete for many research projects. Modern microbiologists showed a similar degree of pragmatism by adopting another existing vessel type – this time the microtiter plate – for the cultivation of heterogeneous culture collections and mutant libraries. The 96-well microtiter plate (MTP) was developed originally by the Hungarian Gyula Takatsy for serological studies in the 1950s (http://www.microplate. org). As is common with new technologies, the pioneers of growing microbial cultures in MTPs in the 1980s and 1990s were not hypercritical and were prepared to tolerate suboptimal quality levels in terms of mixing, oxygen transfer rates (OTRs) and culture-to-culture variability. In the course of the past 7 years, however, the knowledge and technology in relation to the cultivation in MTPs has evolved rapidly. The now satisfactory quality of growth in combination with the logistic advantages of using cheaply available microtiter plates and, perhaps even more importantly, readily available compatible side equipment, such as multipipettes, pipetting robots, microplate readers and autosamplers, is likely to ensure that the MTP will have an increasingly important role in microbiology in the decades to come. Corresponding author: Duetz, W.A. ([email protected]). Available online 24 October 2007. www.sciencedirect.com
Several reviews have been published recently that discuss aspects of MTP technology that are not covered by this review. Micheletti and Lye [1] review microscale bioprocessing and include work on enzymatic bioconversions, miniature bioreactors and down-stream processing on a small scale. Kumar et al. [2], Betts and Baganz [3] and Fernandes and Cabral [4] focus on mini-bioreactors in general but include some sections on MTPs. The review of Kumar et al. [2], for example, includes work on MTPs with integrated optical sensing of dissolved oxygen by immobilization of fluorophores at the bottom of polystyrene wells. In this review, I explain the fundamental physicochemical effects of miniaturization of cultivation systems, focusing specifically on various types of MTPs. Subsequently, I describe successful applications of MTP technology in various fields of microbiology. I also put forward suggestions for improvements in this technology and suggest future research needed in this area. Physicochemical effects of miniaturization The size of the bulk of suspended cultures in 300 ml Erlenmeyer flasks is a factor 104 larger than that of an average microbial cell (5 cm versus 5 mm), whereas this factor is reduced to 103 in a 96-MTP well. It is reasonable to assume that the direct effects of this dimensional change on the physiology of an individual microbial cell are minimal. However, two indirect effects have significant consequences for an individual cell. One is the intrinsically higher ratio of the gas–liquid exchange area to the volume of the bulk of the liquid, which results in relatively high specific OTRs under static conditions. The second physical effect of miniaturization is an increased importance of the surface tension, which counteracts the flow and movement of the liquid under the impact of bubbling or g-forces arising from orbital shaking. The importance of surface tension for growth in MTPs was first addressed by Hermann and coworkers [5] and was later studied in more detail by Doig [6]. The latter study showed that the air–liquid surface area during shaking is strongly diminished by the surface tension in vessels with a diameter of less than 8 mm (for 96- and 384-round-well MTPs but not for 24-roundwell MTPs; Figure 1). An alleviating, although also complicating, factor is the reducing effect that culture-broth components (e.g. cell-membrane components, proteins
0966-842X/$ – see front matter ß 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.tim.2007.09.004
470
Review
TRENDS in Microbiology Vol.15 No.10
Figure 1. High-speed photograph illustrating the ability of a g-force of 2.5, generated by orbital shaking at 300 rpm and a shaking diameter of 50 mm, to generate a rotating movement of an aqueous solution of bromocresol blue (0.1% w/w, surface tension of 72 dynes/cm) in round polyacrylate vessels with various diameters. The theoretical angle of the liquid–air surface with the horizontal plane of 688 is only reached at vessel diameters of 12 mm and higher. The practical relevance of this surface tension effect for 96 low-well MTPs (6 mm) was first demonstrated by Hermann et al. [5]. The angle at a vessel diameter of 16 mm is more than the theoretical 688 since the shaking diameter for the bulk of the liquid is effectively 60 mm instead of 50 mm. The practical absence of any movement of the liquid at 3 and 4 mm is due to the high surface tension. Quantitative approaches of this surface-tension effect can be found in Refs [5] and [6].
and extracellular glycolipids) exert on the surface tension [6–8]. As a consequence, the modeling of OTRs and fluid mechanics in shaken microwells [6] remains difficult and dependent on the characteristics of both the medium broth and the type of microbial strain. The two physical effects of miniaturization mentioned here counteract each other with respect to the maximal OTRs that can be obtained. As a consequence, a satisfactory OTR can be achieved despite a poor degree of mixing within the bulk of the liquid. This is especially relevant when working with cells that tend to accumulate in stagnant zones at the bottom of wells, such as yeast cells. Focusing solely on the results of OTR measurements might lead to erroneous conclusions with regard to the suitability of small culture vessels. Well-closure systems A crucial aspect of microbial growth in MTPs is the closure system of the individual wells, which should prevent crosscontamination during vigorous shaking, permit a limited degree of exchange of headspace air and limit evaporation. An identical physical condition in all wells is a further requirement for quantitative studies; the wells in the corners should have exactly the same characteristics as the wells in the middle of the microtiter plate. This is especially important for media optimization and mutant screening, in which productivity improvements as low as 5% should be detectable. Duetz and coworkers [9] described a closure system consisting of a soft silicone layer with a small hole above the center of each well and a layer of cotton wool. Such a sandwich cover prevented spillage of the culture fluid during orbital shaking and ensured that exchange of headspace air is sufficient and occurs solely through the center holes. Evaporation rates of 20 ml per well per day (at 50% humidity and 30 8C) were reported for 6 mm long holes with a diameter of 1.5 mm [9]. Zimmermann [10] assessed the applicability of a broad range of plastic seals in terms of water loss and oxygen-supply rate. No product was found that satisfies both high O2 permeability and high watervapor retention. At present, a suitable alternative to ‘mechanically’ limiting the gas-exchange area (as described here) is not available. www.sciencedirect.com
Gas–liquid oxygen-transfer rates A sufficiently high exchange rate of headspace air is a prerequisite but not a guarantee that the cells growing in the wells are supplied with an adequate amount of oxygen; gas–liquid transfer is usually the chief limiting factor in this respect. Virtually all publications in this area have focused on orbital shaking as a mode to increase the gas–liquid exchange area. Stirring and bubbling are generally considered to be impractical for large numbers of cultures. Methods for determining OTRs To compare various types of miniature growth vessels and incubation conditions in quantitative terms, reliable methods for OTR determinations are essential. The cobalt-catalyzed oxidation of sulphite and the dynamic gassing-out method are used most commonly. The cobalt-catalyzed oxidation of sulphite method is suitable for volumes of less than 1 ml [11]. A minor disadvantage of this method is the high salt concentration applied (usually 0.5 M), which reduces the maximal solubility of oxygen to a level well below that in low-salt culture media, which generally contains 0.1–0.2 M salts. Because this method systematically underestimates the maximal OTR, a correcting factor of an additional approximately 30% is necessary when translating the results to microbial cultures (J. Buchs, pers. commun.). A second drawback of this method is the high surface tension in comparison to microbial cultures (see earlier) leading to an underestimate of potentially achievable OTRs in small-scale cultures that have a well diameter of less than 8 mm. Deriving OTRs from the oxygen-limited growth phase of growth curves of aerobic strains of Pseudomonas putida [9] or Bacillus subtilis [12] provides a more realistic comparison with ‘real life’ conditions (also in terms of surface tension) but is more laborious. For both strains, a good correlation with non-biological methods within 15% was reported. In recent years, fast enzymatic methods based on glucose oxidase and catechol 2,3-dioxygenase (C23O) have been published [13,14]. Both methods are suitable for cultures in the 10–200 ml range. The C23O method is more elegant because the colored compound (a semi-aldehyde) is formed in a single-step reaction in a 1:1 proportion to the
Review
TRENDS in Microbiology
amount of oxygen consumed. The glucose-oxidase method shows a less clear stoichiometry and therefore requires calibration with another method (such as the cobalt-catalyzed oxidation of sulphite) but has the advantage that all reagents are available commercially. A more detailed discussion on the pros and cons of the methods available can be found in the study of Ortiz-Ochoa [14]. Optimal orbitalshaking incubation conditions for various MTPs are described later. Cultivation in polypropylene square-well MTPs Earlier literature on cultivation in MTPs [9,15] focused on polypropylene MTPs with square deep-wells (8 8 mm in the horizontal plane, 40 mm deep, a well volume of 2.4 ml and culture volumes between 0.5 and 1 ml). OTRs were derived from the oxygen disappearance due to the cobaltcatalyzed oxidation of sulfite and from the oxygen-limited growth phase of growth curves of P. putida. The OTR values of these two methods were the same within an error margin of 15% if a proper correction factor (see earlier) is applied. The highest OTR (38 mmol O2 liter 1 h 1) was measured during orbital shaking at 300 rpm, with a shaking diameter (the orbit of the shaker used) of 50 mm and a culture volume of 0.5 ml. Such OTRs enabled exponential (non-oxygen-limited) growth of a P. putida strain up to a density of approximately 2 g dry wt liter 1. Further, O2-limited growth yielded maximal cell densities of 9 g dry wt liter 1 [9]. The square shape of the wells supposedly contributes to the relatively high OTRs by causing a turbulent flow pattern [15] (Figure 2). This turbulence does not cause splashing in 96- or 24-square-well MTPs at 300 rpm, as often occurs in baffled-shake flasks, however, it did result in wave-like structures at the surface causing an increase in the air–liquid surface area and probably an increase in the steepness of the oxygen gradient at the surface. In addition, the turbulent liquid flow seems to ‘break’ the surface tension, which does not exert a visible negative effect on the air–liquid area (in contrast to 8 mm round wells; compare Figure 1 and 2). An unanticipated result of this study [9] was the strong influence of the shaking diameter. Although, intuitively, one might predict that, with decreasing culture dimensions, a smaller shaking diameter could suffice, in practice, a shaking diameter of 25 mm resulted in threefold-lower OTRs than a shaking
Vol.15 No.10
471
diameter of 50 mm. In 24-square deep-well MTPs, the influence of the shaking diameter on the OTR was much less pronounced: culture volumes of 2.5 ml gave rise to OTRs of 39 and 50 mmol O2 liter 1 h 1 at a shaking diameter of 25 and 50 mm, respectively [13]. These results can be explained by photographic analysis of the shaking patterns: the 17 17 mm wells require an angle of the surface with the horizontal plane of approximately 458 (corresponding to a g-force of 1) for an optimal increase in the liquid surface area, whereas, in 8 8 mm square wells, an angle of approx 708 is required. A problem associated with the cultivation of microorganisms in polypropylene deep-well MTPs is the leakage of toxic compounds [16] from the plastic during incubation. Extraction of these contaminants by sequential soaking of MTPs in boiling NaOH and hot HCl before firsttime use was reported to reduce this problem sufficiently [9]. The polypropylene square deep-well MTP is unsurpassable currently in terms of total culture volume per MTP and the degree of turbulence that can be achieved. Cultivation in polystyrene round-well MTPs Polystyrene round-well MTPs (especially 96- and 48-well plates) have received much attention in more recent years. The advantages of polystyrene MTPs include their transparency, which enables the direct reading of optical densities, the absence of toxic substances leaking out from polystyrene and the presence of two walls separating the wells rather than one, which further minimizes the risk of cross-contamination. Hermann and Buchs [5] yielded a thorough insight into the influence of various shaking parameters on the OTRs in 96-low-well MTPs (well diameter: 6 mm; depth: 11 mm). Also, for this type of MTP, a strong influence of the shaking diameter on the OTR was found (Figure 3). OTRs of 15 mmol O2 liter 1 h 1 (20 mmol O2 liter 1 h 1, if corrected for the lower oxygen solubility) were achieved at 300 rpm, a shaking diameter of 50 mm and a culture volume of 0.2 ml. Similar OTRs at smaller shaking diameters were only reached at much higher frequencies (Figure 3). The maximum OTR in round wells is approximately half the OTR for a 96-deep-square well if the same filling height is used (mainly owing to the higher turbulence in square deep-well plates). The viscosity appeared to
Figure 2. Shaking pattern in a square microwell of 2 ml (8 8 mm) during orbital shaking at 300 rpm, a shaking diameter of 50 mm and a working volume of 0.75 ml. The relatively high turbulence in such cultures in square deep-well plates results in wave structures at the surface and relatively high OTRs (24 mmol O2 liter 1 h 1) in comparison to round-well MTPs at the same ‘filling height’. Reprinted with permission from [15]. www.sciencedirect.com
472
Review
TRENDS in Microbiology Vol.15 No.10
Figure 3. Effect of the shaking frequency and the shaking diameter on the maximum oxygen-transfer capacity (OTRmax) and specific mass-transfer coefficient (kLa) in conventional polystyrene 96-roundwell MTPs (200 ml filling volume, 22 8C, 0.5 M Na2SO3). The shaking diameter exerts a strong influence on the OTR. Reprinted with permission from [5].
exert little effect but the surface tension has a strong effect on the ‘liquid height’ (a measure for the extent to which a centrifugal force is capable of throwing the bulk of the liquid up the wall of a well). This ‘liquid height’ was significantly lower for water than for ethanol, which have surface tensions of 72 and 22 dynes/cm, respectively. The influence of the surface tension on the OTR was not quantified in this study but was later assessed using growth curves of P. putida (Duetz, W.A, unpublished, method detailed in [9]). These studies yielded an OTR of 26 mmol O2 liter 1 h 1, rather than 20 mmol O2 liter 1 h 1, as derived from the chemical method [5]. When comparing the shaking patterns in wells filled with water and wells filled with a P. putida culture grown on a glucose-mineral
medium, it is plausible that this 30% increase in OTR is not an artifact and is caused by the change in surface tension (Figure 4). Using the same growth-curve method, culture volumes of 150 and 100 ml yielded OTRs of 32 and 39 mmol O2 liter 1 h 1, respectively, at 300 rpm, with a shaking diameter of 50 mm and at 30 8C. Kensy and coworkers [17] used polystyrene 48-roundwell MTPs with wells twice the diameter of 96-low-well MTPs (11 mm instead of 6 mm) for a similar study. A central goal of this study was to find the upper limit of the OTR that can be reached in MTPs by orbital shaking. The maximum OTR obtained (280 mmol O2 liter 1 h 1 at 1450 rpm, a shaking diameter of 3 mm and a working volume of 300 ml) is the highest reported for MTPs so
Figure 4. Comparison of the shaking pattern at the indicated volumes of an aqueous solution (a) without cells and (b) a culture of P. putida in wells from a polystyrene 96low-well MTP during orbital shaking at 300 rpm and a shaking diameter of 50 mm (g-force = 2.5). The larger ‘liquid height’ in the cultures is caused by the lowering effect that culture broth components (e.g. cell-membrane components, proteins and extracellular glycolipids) exert on the surface tension [6–8]. www.sciencedirect.com
Review
TRENDS in Microbiology
Vol.15 No.10
473
Table 1. OTRs in various types of MTPs at various filling heights during orbital shaking at 300 rpm and a shaking diameter of 50 mm Type of MTP 24-square deep well 96-square deep well 96-square deep well 48-low well, no cells 96-low well, no cells 96-low well, cells 96-low well, cells
volume (height) 2.5 ml (8.7 mm) 0.5 ml (8.5 mm) 0.75 ml (12.8 mm) 0.6 ml (6.0 mm) 0.2 ml (6.2 mm) 0.2 ml (6.2 mm) 0.15 ml (4.7 mm)
OTR 50 mmol 38 mmol 24 mmol 27 mmol 20 mmol 26 mmol 32 mmol
O2 O2 O2 O2 O2 O2 O2
liter liter liter liter liter liter liter
1
h 1 h 1 h 1 h 1 h 1 h 1 h
1a 1a 1a 1b 1b 1a 1a
Refs Duetz et al., 2004 [13] Duetz et al., 2000 [9] Duetz et al., 2000 [9] Kensy et al., 2004 [17] Hermann et al., 2003 [5] Duetz, W.A, Unpublished Duetz, W.A, Unpublished
a
Derived from the oxygen-limited growth phase of growth curves of Pseudomonas putida. Determined by the cobalt-catalyzed oxidation of sulphite but corrected for the lower oxygen solubility (original values multiplied by 1.3 to compensate for low oxygen solubility: see text). b
far. These high values, however, do not imply that 48-well MTPs give rise to intrinsically higher OTRs compared with 96-low-well MTPs under more standard shaking conditions. A culture volume of 600 ml shaken at 300 rpm at a shaking diameter of 50 mm resulted in an OTR of 21 mmol O2 liter 1 h 1 (27 mmol O2 liter 1 h 1 if corrected for the lower oxygen solubility). If this OTR is compared with the OTRs measured in 96-low-well MTPs incubated under the same orbital-shaking conditions and a similar filling height, the difference is small (Table 1). An advantage of 48-well plates is that the fourfold higher culture volume reduces the effect of random pipetting errors and variations in inoculum sizes. However, the poor compatibility of the 48-well format with a lot of side equipment (such as microplate readers and multipipettes) is a distinct disadvantage. It might be concluded that MTPs with round wells give rise to lower OTRs than square-well MTPs at the same filling height and shaking conditions but that this disadvantage can be compensated for readily by using smaller culture volumes. The logistic advantages of the use of nontoxic and disposable polystyrene round-well MTPs – in combination with ever more sensitive assays – make it likely that they will gradually replace square-well MTPs for those applications where turbulence is not a major issue. Application of MTPs Mutant screenings Open literature on the application of MTPs for mutant screenings is scarce and mainly limited to metabolic flux analysis studies in academic settings [18–24]. The group of Uwe Sauer used 96-square deep-well MTPs to study glucose metabolism in 137 knockout mutants of B. subtilis using 13C tracer technology [18]. The application of small culture volumes in this research area is particularly appealing in the light of the high costs of 13C-labeled compounds. Culture volumes of 1.2 ml in combination with the cover systems described in Ref. [9] enabled the detection of changes in fluxes as low as 5–10%. A good correlation with results in 30 ml shake-flask cultures was observed. A comparative study from the same research group [20] concluded that batch-growth characteristics of E. coli in shake flask and 2-ml deep-well microtiter plates were indistinguishable. Standard deviations of growth rates and 13C glucose-uptake rates (among triplicates of 1.2 ml cultures in 96-square deep-well MTPs) were reported to be 7–8%. A similar study on 37 knockout mutants of Saccharomyces cerevisiae showed standard deviations in www.sciencedirect.com
growth rates of duplicates below 5% using 1.2 ml cultures in 96-square deep-well MTPs [21]. Standard deviations of a similar magnitude were reported for growth rates and fluxes through the pentose-phosphate pathway for 14 hemiascomycetous yeasts, including strains from the genera Saccharomyces, Kluyveromyces, Pichia, Yarrowia and Candida [22]. Weis et al. [23] optimized the growth conditions of Pichia pastoris in 96-square deep-well plates to limit cell death and to enable the accurate screening of enzyme variants in gene-expression libraries. The application of MTPs for mutant screenings in industry has become standard practice rapidly in recent years, for example, in screening high-productivity mutants for secondary metabolites. Minas et al. [24] reported standard deviations in the actinorhodin production by two Streptomyces coelicolor strains of 2% and 6% for 12 and six duplicates, respectively, using 1 ml cultures in 96square deep-well MTPs. The scarcity of further open literature in this area leaves me no choice other than to draw a general and subjective picture based mainly on personal contacts with representatives from the industry. If standard deviations in measured parameters between 5% and 10% are acceptable, the application of polystyrene 96-lowwell MTPs with maximal culture volumes of 0.2 ml is often feasible, although restricted to strains giving rise to homogeneous cultures, such as Bacillus, E. coli or yeast, and only if used in combination with very accurate pipetting robots and a well-defined inoculation size. If standard deviations below 5% are required for experiments, such as mature production bioprocesses where improvements per mutant cycle are very low, minimum culture volumes of 1 ml are indicative, especially for inhomogeneous cultures from, for example, Streptomyces mutant banks. Depending on the strain type, the use of square deep-well MTPs or 48or 24-low-well MTPs might give satisfactory results. The final choice is a trade-off between throughput per unit of time and the time-expenditure on what later appear to be false positives. In all mutant-screening projects, the inoculum size appears to be a crucial factor and measures to synchronize growth curves of all mutants are essential for low standard deviations of independent duplicates. The number of false positives could be reduced if end-point measurements were corrected for small differences in lag times and growth rates. Such corrections would be possible with the availability of methodologies to follow the growth in the wells of a MTP during continuous shaking conditions in parallel. Recent work from the Jochen Buchs group [25] addressed this concern by generating growth curves of Vibrio natriegens and E. coli using light scattering as well as intracellular and/or protein fluorescence.
474
Review
TRENDS in Microbiology Vol.15 No.10
OTRs and degrees of turbulence in the primaryscreening phase, similar in magnitude to those in the production facilities, help to limit the number of ‘wellplate wonders’ or mutants that perform reproducibly better in the primary screening but not in multiliter fermenters. Discovery of new secondary metabolites 24-square well MTPs are often used in the discovery of new secondary metabolites (for reviews, see [26,27]). The relative ease with which high OTRs are achieved [13] ensures a high biomass and sufficient secondary metabolites formed by anabolic pathways requiring oxygenases. Furthermore, culture volumes of 2–4 ml enable a straightforward separation of cells, pellets or mycelia from the supernatant by centrifugation and adequate amounts of extracted compounds for multiple bioassays, such as those for antibiotic or anticancer activities. Pelaez [26] advocated the use of MTPs with the argument that they permit an increase in the number of growth conditions that can be assigned to each strain and so enable a more complete exploitation of the metabolic potential of selected strains. 96-square deep-well MTPs are also applied in this area of research [26], particularly in combination with direct solvent extraction of the whole culture with no prior separation of biomass and supernatant. Miscellaneous applications Casey [28] described the successful use of MTPs to generate dose–response curves of test organisms (E. coli, B. subtilis and S. cerevisiae) to bioactive compounds formed by Streptomyces hygroscopicus. Diaz et al. [29] described a rapid assay for methionine using a methionine auxotroph E. coli strain growing in MTPs. Doig et al. [30] used several microwell formats (96-round-, 96-deep square- and 24-round-well microtiter plates) for quantification of the kinetics of a Baeyer-Villiger oxygenase expressed in E. coli. In a similar study, FerreiraTorres et al. [31] used 96-square deep-well MTPs to compare libraries of recombinant cyclohexanone monooxygenases in E. coli and Acinetobacter calcoaceticus. The use of MTPs for the screening of heterogeneous culture collections for regiospecific hydroxylations was described for a range of educt–product combinations [32–35]. Conclusions In the past 7 years, MTPs have become a mature alternative to Erlenmeyer flasks for the batch cultivation of microorganisms and have helped to make the handling and screening of large numbers of strains and mutant libraries less time consuming. Orbital shaking is suitable to generate OTRs in MTPs equaling or even surpassing the OTRs achieved traditionally in Erlenmeyer flasks, if the right shaking conditions are used. The development of suitable cover systems has ruled out cross-contaminations. Overall, these developments have contributed to the speed with which the productivity of large-scale bioprocesses can be improved and new secondary metabolites can be discovered. Further research is necessary for continued progress in this area (Box 1). www.sciencedirect.com
Box 1. Further research for improving the quality of growth in microtiter plates Studies on the degree to which various shapes of wells, in combination with various orbital-shaking conditions, mimic the hydrodynamic conditions that microbial cells encounter in large cultivation systems, such as stirred-tank bioreactors. Quantification of local fluid velocities. The development of filter materials combining high diffusion coefficients for O2 and CO2 with low diffusion coefficients for water vapor. The cultivation of slow-growing strains would benefit from the integration of such materials in well-closure systems. Adaptation of this technology for anaerobic cultivations, which would benefit flourishing research fields, such as the industrial production of ethanol. The development of technologies enabling the parallel quantitation of biomass levels in time during orbital shaking. Such data would, for example, enable end-point measurements of secondary metabolite levels to be corrected for small differences in lag times and growth rates of the screened mutants.
References 1 Micheletti, M. and Lye, G.J. (2006) Microscale bioprocess optimisation. Curr. Opin. Biotechnol. 17, 611–618 2 Kumar, S. et al. (2004) Minibioreactors. Biotechnol. Lett. 26, 1–10 3 Betts, J.I. and Baganz, F. (2006) Miniature bioreactors: current practices and future opportunities. Microbial. Cell. Fact. 5, Art. No. 21 4 Fernandes, P. and Cabral, J.M.S. (2006) Microlitre/millilitre shaken bioreactors in fermentative and biotransformation processes - a review. Biocatal. Biotrans. 24, 237–252 5 Hermann, R. et al. (2003) Characterization of gas-liquid mass transfer phenomena in microtiter plates. Biotechnol. Bioeng. 81, 178–186 6 Doig, S.D. et al. (2005) Modelling surface aeration rates in shaken microtitre plates using dimensionless groups. Chem. Eng. Sc. 60, 2741– 2750 7 Marshall, M.S. (1924) Surface tension of culture mediums. J. Infect. Dis. 35, 526–536 8 Bailey, J.E. and Ollis, D.F. (1986) Biochemical engineering fundamentals. 2nd edn., New York: McGraw-Hill 9 Duetz, W.A. et al. (2000) Methods for intense aeration, growth, storage, and replication of bacterial strains in microtiter plates. Appl. Environ. Microbiol. 66, 2641–2646 10 Zimmermann, H.F. et al. (2003) Rapid evaluation of oxygen and water permeation through microplate sealing tapes. Biotechnol. Prog. 19, 1061–1063 11 Hermann, R. et al. (2001) Optical method for the determination of the oxygen-transfer capacity of small bioreactors based on sulfite oxidation. Biotechnol. Bioeng. 74, 355–363 12 Doig, S.D. et al. (2005) Characterisation of a novel miniaturised bubble column bioreactor for high throughput cell cultivation. Biochem. Eng. J. 23, 97–105 13 Duetz, W.A. and Witholt, B. (2004) Oxygen transfer by orbital shaking of square vessels and deepwell microtiter plates of various dimensions. Biochem. Eng. J. 17, 181–185 14 Ortiz-Ochoa, K. et al. (2005) A novel method for the measurement of oxygen mass transfer rates in small-scale vessels. Biochem. Eng. J. 25, 63–68 15 Duetz, W.A. and Witholt, B. (2001) Effectiveness of orbital shaking for the aeration of suspended bacterial cultures in square deepwell microtiter plates. Biochem. Eng. J. 7, 113–115 16 Becker, R.F. et al. (1996) Additives. In Polypropylene handbook (Moore, E.P., Jr, ed.), pp. 177–210, Cincinnati, Ohio, Hanser-Gardner 17 Kensy, F. et al. (2005) Oxygen transfer phenomena in 48-well microtiter plates: Determination by optical monitoring of sulfite oxidation and verification by real-time measurement during microbial growth. Biotechnol. Bioeng. 89, 698–708 18 Fischer, E. and Sauer, U. (2005) Large-scale in vivo flux analysis shows rigidity and suboptimal performance of Bacillus subtilis metabolism. Nat. Gen. 37, 636–640
Review
TRENDS in Microbiology
19 Sauer, U. (2004) High-throughput phenomics: experimental methods for mapping fluxomes. Curr. Opin. Biotechnol. 15, 58–63 20 Fischer, E. et al. (2004) High-throughput metabolic flux analysis based on gas chromatography-mass spectrometry derived C-13 constraints. Anal. Biochem. 325, 308–316 21 Blank, L.M. et al. (2005) Large-scale C-13-flux analysis reveals mechanistic principles of metabolic network robustness to null mutations in yeast. Genome Biol. 6 (6), Art. No. R49 22 Blank, L.M. et al. (2005) Metabolic-flux and network analysis in fourteen hemiascomycetous yeasts. FEMS Yeast Res. 5, 545– 558 23 Weis, R. et al. (2004) Reliable high-throughput screening with Pichia pastoris by limiting yeast cell death phenomena. FEMS Yeast Res. 5, 179–189 24 Minas, W. et al. (2000) Streptomycetes in micro-cultures: growth, production of secondary metabolites, and storage and retrieval in the 96-well format. Antonie Van Leeuwenhoek 78, 297–305 25 Samorski, M. et al. (2005) Quasi-continuous combined scattered light and fluorescence measurements: a novel measurement technique for shaken microtiter plates. Biotechnol. Bioeng. 92, 61–68 26 Pelaez, F. (2006) The historical delivery of antibiotics from microbial natural products – can history repeat? Biochem. Pharmacol. 71, 981– 990 27 Singh, S.B. and Barrett, J.F. (2006) Empirical antibacterial drug discovery – foundation in natural products. Biochem. Pharmacol. 71, 1006–1015
Vol.15 No.10
28 Casey, J.T. et al. (2004) Development of a robust microtiter plate-based assay method for assessment of bioactivity. J. Microbiol. Methods 58, 327–334 29 Diaz, I.B.Z. et al. (2002) Adaptation of a methionine auxotroph Escherichia coli growth assay to microtiter plates for quantitating methionine. J. Rapid Meth. Autom. Microbiol. 10, 217–229 30 Doig, S.D. et al. (2002) The use of microscale processing technologies for quantification of biocatalytic Baeyer-Villiger oxidation kinetics. Biotechnol. Bioeng. 80, 42–49 31 Ferreira-Torres, C. et al. (2005) Microscale process evaluation of recombinant biocatalyst libraries: application to Baeyer-Villiger monooxygenase catalysed lactone synthesis. Bioprocess Biosys. Eng. 28, 83–93 32 Duetz, W.A. et al. (2001) Biotransformation of D-limonene to (+) transcarveol by toluene-grown Rhodococcus opacus PWD4 cells. Appl. Environ. Microbiol. 67, 2829–2832 33 Spain, J.C. et al. (2003) Production of 6-phenylacetylene picolinic acid from diphenylacetylene by a toluene-degrading Acinetobacter strain. Appl. Environ. Microbiol. 69, 4037–4042 34 O’Connor, K.E. et al. (2005) Isolation and characterization of a diverse group of phenylacetic acid degrading microorganisms from pristine soil. Chemosphere 61, 965–973 35 Van Beilen, J.B. et al. (2005) Biocatalytic production of perillyl alcohol from limonene by using a novel Mycobacterium sp cytochrome P450 alkane hydroxylase expressed in Pseudomonas putida. Appl. Environ. Microbiol. 71, 1737–1744
Elsevier celebrates two anniversaries with a gift to university libraries in the developing world In 1580, the Elzevir family began their printing and bookselling business in the Netherlands, publishing works by scholars such as John Locke, Galileo Galilei and Hugo Grotius. On 4 March 1880, Jacobus George Robbers founded the modern Elsevier company intending, just like the original Elzevir family, to reproduce fine editions of literary classics for the edification of others who shared his passion, other ‘Elzevirians’. Robbers co-opted the Elzevir family printer’s mark, stamping the new Elsevier products with a classic symbol of the symbiotic relationship between publisher and scholar. Elsevier has since become a leader in the dissemination of scientific, technical and medical (STM) information, building a reputation for excellence in publishing, new product innovation and commitment to its STM communities. In celebration of the House of Elzevir’s 425th anniversary and the 125th anniversary of the modern Elsevier company, Elsevier donated books to ten university libraries in the developing world. Entitled ‘A Book in Your Name’, each of the 6700 Elsevier employees worldwide was invited to select one of the chosen libraries to receive a book donated by Elsevier. The core gift collection contains the company’s most important and widely used STM publications, including Gray’s Anatomy, Dorland’s Illustrated Medical Dictionary, Essential Medical Physiology, Cecil Essentials of Medicine, Mosby’s Medical, Nursing and Allied Health Dictionary, The Vaccine Book, Fundamentals of Neuroscience, and Myles Textbook for Midwives. The ten beneficiary libraries are located in Africa, South America and Asia. They include the Library of the Sciences of the University of Sierra Leone; the library of the Muhimbili University College of Health Sciences of the University of Dar es Salaam, Tanzania; the library of the College of Medicine of the University of Malawi; and the University of Zambia; Universite du Mali; Universidade Eduardo Mondlane, Mozambique; Makerere University, Uganda; Universidad San Francisco de Quito, Ecuador; Universidad Francisco Marroquin, Guatemala; and the National Centre for Scientific and Technological Information (NACESTI), Vietnam. Through ‘A Book in Your Name’, these libraries received books with a total retail value of approximately one million US dollars.
For more information, visit www.elsevier.com www.sciencedirect.com
475
TRENDS in Microbiology
Vol.15 No.10
Microtiter plates as mini-bioreactors: miniaturization of fermentation methods Wouter A. Duetz Enzyscreen, Biopartner Center Leiden, Wassenaarseweg 72, 2333 AL Leiden, The Netherlands
In the past decade, the use of microtiter plates for microbial growth has become widespread, particularly in industry. In parallel, research in academia has provided a thorough insight into the complex relation between well dimensions, culture volumes, orbital shaking conditions and surface tension on the one hand, and oxygen-transfer rates and degrees of mixing on the other. In this review, I will discuss these fundamental issues and describe the current applications of microtiter plates in microbiology. Microtiter plates can now be considered a mature alternative to Erlenmeyer shake flasks. Progression from Erlenmeyer flasks to microtiter plates The adoption of the shake flask in the 1880s as the vessel of choice for the growth of microorganisms in liquid cultures was pragmatic in the sense that the Erlenmeyer flask had been invented 20 years earlier, for use in chemical laboratories. More than a century later, the relatively large culture volumes in Erlenmeyer flasks gradually became obsolete for many research projects. Modern microbiologists showed a similar degree of pragmatism by adopting another existing vessel type – this time the microtiter plate – for the cultivation of heterogeneous culture collections and mutant libraries. The 96-well microtiter plate (MTP) was developed originally by the Hungarian Gyula Takatsy for serological studies in the 1950s (http://www.microplate. org). As is common with new technologies, the pioneers of growing microbial cultures in MTPs in the 1980s and 1990s were not hypercritical and were prepared to tolerate suboptimal quality levels in terms of mixing, oxygen transfer rates (OTRs) and culture-to-culture variability. In the course of the past 7 years, however, the knowledge and technology in relation to the cultivation in MTPs has evolved rapidly. The now satisfactory quality of growth in combination with the logistic advantages of using cheaply available microtiter plates and, perhaps even more importantly, readily available compatible side equipment, such as multipipettes, pipetting robots, microplate readers and autosamplers, is likely to ensure that the MTP will have an increasingly important role in microbiology in the decades to come. Corresponding author: Duetz, W.A. ([email protected]). Available online 24 October 2007. www.sciencedirect.com
Several reviews have been published recently that discuss aspects of MTP technology that are not covered by this review. Micheletti and Lye [1] review microscale bioprocessing and include work on enzymatic bioconversions, miniature bioreactors and down-stream processing on a small scale. Kumar et al. [2], Betts and Baganz [3] and Fernandes and Cabral [4] focus on mini-bioreactors in general but include some sections on MTPs. The review of Kumar et al. [2], for example, includes work on MTPs with integrated optical sensing of dissolved oxygen by immobilization of fluorophores at the bottom of polystyrene wells. In this review, I explain the fundamental physicochemical effects of miniaturization of cultivation systems, focusing specifically on various types of MTPs. Subsequently, I describe successful applications of MTP technology in various fields of microbiology. I also put forward suggestions for improvements in this technology and suggest future research needed in this area. Physicochemical effects of miniaturization The size of the bulk of suspended cultures in 300 ml Erlenmeyer flasks is a factor 104 larger than that of an average microbial cell (5 cm versus 5 mm), whereas this factor is reduced to 103 in a 96-MTP well. It is reasonable to assume that the direct effects of this dimensional change on the physiology of an individual microbial cell are minimal. However, two indirect effects have significant consequences for an individual cell. One is the intrinsically higher ratio of the gas–liquid exchange area to the volume of the bulk of the liquid, which results in relatively high specific OTRs under static conditions. The second physical effect of miniaturization is an increased importance of the surface tension, which counteracts the flow and movement of the liquid under the impact of bubbling or g-forces arising from orbital shaking. The importance of surface tension for growth in MTPs was first addressed by Hermann and coworkers [5] and was later studied in more detail by Doig [6]. The latter study showed that the air–liquid surface area during shaking is strongly diminished by the surface tension in vessels with a diameter of less than 8 mm (for 96- and 384-round-well MTPs but not for 24-roundwell MTPs; Figure 1). An alleviating, although also complicating, factor is the reducing effect that culture-broth components (e.g. cell-membrane components, proteins
0966-842X/$ – see front matter ß 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.tim.2007.09.004
470
Review
TRENDS in Microbiology Vol.15 No.10
Figure 1. High-speed photograph illustrating the ability of a g-force of 2.5, generated by orbital shaking at 300 rpm and a shaking diameter of 50 mm, to generate a rotating movement of an aqueous solution of bromocresol blue (0.1% w/w, surface tension of 72 dynes/cm) in round polyacrylate vessels with various diameters. The theoretical angle of the liquid–air surface with the horizontal plane of 688 is only reached at vessel diameters of 12 mm and higher. The practical relevance of this surface tension effect for 96 low-well MTPs (6 mm) was first demonstrated by Hermann et al. [5]. The angle at a vessel diameter of 16 mm is more than the theoretical 688 since the shaking diameter for the bulk of the liquid is effectively 60 mm instead of 50 mm. The practical absence of any movement of the liquid at 3 and 4 mm is due to the high surface tension. Quantitative approaches of this surface-tension effect can be found in Refs [5] and [6].
and extracellular glycolipids) exert on the surface tension [6–8]. As a consequence, the modeling of OTRs and fluid mechanics in shaken microwells [6] remains difficult and dependent on the characteristics of both the medium broth and the type of microbial strain. The two physical effects of miniaturization mentioned here counteract each other with respect to the maximal OTRs that can be obtained. As a consequence, a satisfactory OTR can be achieved despite a poor degree of mixing within the bulk of the liquid. This is especially relevant when working with cells that tend to accumulate in stagnant zones at the bottom of wells, such as yeast cells. Focusing solely on the results of OTR measurements might lead to erroneous conclusions with regard to the suitability of small culture vessels. Well-closure systems A crucial aspect of microbial growth in MTPs is the closure system of the individual wells, which should prevent crosscontamination during vigorous shaking, permit a limited degree of exchange of headspace air and limit evaporation. An identical physical condition in all wells is a further requirement for quantitative studies; the wells in the corners should have exactly the same characteristics as the wells in the middle of the microtiter plate. This is especially important for media optimization and mutant screening, in which productivity improvements as low as 5% should be detectable. Duetz and coworkers [9] described a closure system consisting of a soft silicone layer with a small hole above the center of each well and a layer of cotton wool. Such a sandwich cover prevented spillage of the culture fluid during orbital shaking and ensured that exchange of headspace air is sufficient and occurs solely through the center holes. Evaporation rates of 20 ml per well per day (at 50% humidity and 30 8C) were reported for 6 mm long holes with a diameter of 1.5 mm [9]. Zimmermann [10] assessed the applicability of a broad range of plastic seals in terms of water loss and oxygen-supply rate. No product was found that satisfies both high O2 permeability and high watervapor retention. At present, a suitable alternative to ‘mechanically’ limiting the gas-exchange area (as described here) is not available. www.sciencedirect.com
Gas–liquid oxygen-transfer rates A sufficiently high exchange rate of headspace air is a prerequisite but not a guarantee that the cells growing in the wells are supplied with an adequate amount of oxygen; gas–liquid transfer is usually the chief limiting factor in this respect. Virtually all publications in this area have focused on orbital shaking as a mode to increase the gas–liquid exchange area. Stirring and bubbling are generally considered to be impractical for large numbers of cultures. Methods for determining OTRs To compare various types of miniature growth vessels and incubation conditions in quantitative terms, reliable methods for OTR determinations are essential. The cobalt-catalyzed oxidation of sulphite and the dynamic gassing-out method are used most commonly. The cobalt-catalyzed oxidation of sulphite method is suitable for volumes of less than 1 ml [11]. A minor disadvantage of this method is the high salt concentration applied (usually 0.5 M), which reduces the maximal solubility of oxygen to a level well below that in low-salt culture media, which generally contains 0.1–0.2 M salts. Because this method systematically underestimates the maximal OTR, a correcting factor of an additional approximately 30% is necessary when translating the results to microbial cultures (J. Buchs, pers. commun.). A second drawback of this method is the high surface tension in comparison to microbial cultures (see earlier) leading to an underestimate of potentially achievable OTRs in small-scale cultures that have a well diameter of less than 8 mm. Deriving OTRs from the oxygen-limited growth phase of growth curves of aerobic strains of Pseudomonas putida [9] or Bacillus subtilis [12] provides a more realistic comparison with ‘real life’ conditions (also in terms of surface tension) but is more laborious. For both strains, a good correlation with non-biological methods within 15% was reported. In recent years, fast enzymatic methods based on glucose oxidase and catechol 2,3-dioxygenase (C23O) have been published [13,14]. Both methods are suitable for cultures in the 10–200 ml range. The C23O method is more elegant because the colored compound (a semi-aldehyde) is formed in a single-step reaction in a 1:1 proportion to the
Review
TRENDS in Microbiology
amount of oxygen consumed. The glucose-oxidase method shows a less clear stoichiometry and therefore requires calibration with another method (such as the cobalt-catalyzed oxidation of sulphite) but has the advantage that all reagents are available commercially. A more detailed discussion on the pros and cons of the methods available can be found in the study of Ortiz-Ochoa [14]. Optimal orbitalshaking incubation conditions for various MTPs are described later. Cultivation in polypropylene square-well MTPs Earlier literature on cultivation in MTPs [9,15] focused on polypropylene MTPs with square deep-wells (8 8 mm in the horizontal plane, 40 mm deep, a well volume of 2.4 ml and culture volumes between 0.5 and 1 ml). OTRs were derived from the oxygen disappearance due to the cobaltcatalyzed oxidation of sulfite and from the oxygen-limited growth phase of growth curves of P. putida. The OTR values of these two methods were the same within an error margin of 15% if a proper correction factor (see earlier) is applied. The highest OTR (38 mmol O2 liter 1 h 1) was measured during orbital shaking at 300 rpm, with a shaking diameter (the orbit of the shaker used) of 50 mm and a culture volume of 0.5 ml. Such OTRs enabled exponential (non-oxygen-limited) growth of a P. putida strain up to a density of approximately 2 g dry wt liter 1. Further, O2-limited growth yielded maximal cell densities of 9 g dry wt liter 1 [9]. The square shape of the wells supposedly contributes to the relatively high OTRs by causing a turbulent flow pattern [15] (Figure 2). This turbulence does not cause splashing in 96- or 24-square-well MTPs at 300 rpm, as often occurs in baffled-shake flasks, however, it did result in wave-like structures at the surface causing an increase in the air–liquid surface area and probably an increase in the steepness of the oxygen gradient at the surface. In addition, the turbulent liquid flow seems to ‘break’ the surface tension, which does not exert a visible negative effect on the air–liquid area (in contrast to 8 mm round wells; compare Figure 1 and 2). An unanticipated result of this study [9] was the strong influence of the shaking diameter. Although, intuitively, one might predict that, with decreasing culture dimensions, a smaller shaking diameter could suffice, in practice, a shaking diameter of 25 mm resulted in threefold-lower OTRs than a shaking
Vol.15 No.10
471
diameter of 50 mm. In 24-square deep-well MTPs, the influence of the shaking diameter on the OTR was much less pronounced: culture volumes of 2.5 ml gave rise to OTRs of 39 and 50 mmol O2 liter 1 h 1 at a shaking diameter of 25 and 50 mm, respectively [13]. These results can be explained by photographic analysis of the shaking patterns: the 17 17 mm wells require an angle of the surface with the horizontal plane of approximately 458 (corresponding to a g-force of 1) for an optimal increase in the liquid surface area, whereas, in 8 8 mm square wells, an angle of approx 708 is required. A problem associated with the cultivation of microorganisms in polypropylene deep-well MTPs is the leakage of toxic compounds [16] from the plastic during incubation. Extraction of these contaminants by sequential soaking of MTPs in boiling NaOH and hot HCl before firsttime use was reported to reduce this problem sufficiently [9]. The polypropylene square deep-well MTP is unsurpassable currently in terms of total culture volume per MTP and the degree of turbulence that can be achieved. Cultivation in polystyrene round-well MTPs Polystyrene round-well MTPs (especially 96- and 48-well plates) have received much attention in more recent years. The advantages of polystyrene MTPs include their transparency, which enables the direct reading of optical densities, the absence of toxic substances leaking out from polystyrene and the presence of two walls separating the wells rather than one, which further minimizes the risk of cross-contamination. Hermann and Buchs [5] yielded a thorough insight into the influence of various shaking parameters on the OTRs in 96-low-well MTPs (well diameter: 6 mm; depth: 11 mm). Also, for this type of MTP, a strong influence of the shaking diameter on the OTR was found (Figure 3). OTRs of 15 mmol O2 liter 1 h 1 (20 mmol O2 liter 1 h 1, if corrected for the lower oxygen solubility) were achieved at 300 rpm, a shaking diameter of 50 mm and a culture volume of 0.2 ml. Similar OTRs at smaller shaking diameters were only reached at much higher frequencies (Figure 3). The maximum OTR in round wells is approximately half the OTR for a 96-deep-square well if the same filling height is used (mainly owing to the higher turbulence in square deep-well plates). The viscosity appeared to
Figure 2. Shaking pattern in a square microwell of 2 ml (8 8 mm) during orbital shaking at 300 rpm, a shaking diameter of 50 mm and a working volume of 0.75 ml. The relatively high turbulence in such cultures in square deep-well plates results in wave structures at the surface and relatively high OTRs (24 mmol O2 liter 1 h 1) in comparison to round-well MTPs at the same ‘filling height’. Reprinted with permission from [15]. www.sciencedirect.com
472
Review
TRENDS in Microbiology Vol.15 No.10
Figure 3. Effect of the shaking frequency and the shaking diameter on the maximum oxygen-transfer capacity (OTRmax) and specific mass-transfer coefficient (kLa) in conventional polystyrene 96-roundwell MTPs (200 ml filling volume, 22 8C, 0.5 M Na2SO3). The shaking diameter exerts a strong influence on the OTR. Reprinted with permission from [5].
exert little effect but the surface tension has a strong effect on the ‘liquid height’ (a measure for the extent to which a centrifugal force is capable of throwing the bulk of the liquid up the wall of a well). This ‘liquid height’ was significantly lower for water than for ethanol, which have surface tensions of 72 and 22 dynes/cm, respectively. The influence of the surface tension on the OTR was not quantified in this study but was later assessed using growth curves of P. putida (Duetz, W.A, unpublished, method detailed in [9]). These studies yielded an OTR of 26 mmol O2 liter 1 h 1, rather than 20 mmol O2 liter 1 h 1, as derived from the chemical method [5]. When comparing the shaking patterns in wells filled with water and wells filled with a P. putida culture grown on a glucose-mineral
medium, it is plausible that this 30% increase in OTR is not an artifact and is caused by the change in surface tension (Figure 4). Using the same growth-curve method, culture volumes of 150 and 100 ml yielded OTRs of 32 and 39 mmol O2 liter 1 h 1, respectively, at 300 rpm, with a shaking diameter of 50 mm and at 30 8C. Kensy and coworkers [17] used polystyrene 48-roundwell MTPs with wells twice the diameter of 96-low-well MTPs (11 mm instead of 6 mm) for a similar study. A central goal of this study was to find the upper limit of the OTR that can be reached in MTPs by orbital shaking. The maximum OTR obtained (280 mmol O2 liter 1 h 1 at 1450 rpm, a shaking diameter of 3 mm and a working volume of 300 ml) is the highest reported for MTPs so
Figure 4. Comparison of the shaking pattern at the indicated volumes of an aqueous solution (a) without cells and (b) a culture of P. putida in wells from a polystyrene 96low-well MTP during orbital shaking at 300 rpm and a shaking diameter of 50 mm (g-force = 2.5). The larger ‘liquid height’ in the cultures is caused by the lowering effect that culture broth components (e.g. cell-membrane components, proteins and extracellular glycolipids) exert on the surface tension [6–8]. www.sciencedirect.com
Review
TRENDS in Microbiology
Vol.15 No.10
473
Table 1. OTRs in various types of MTPs at various filling heights during orbital shaking at 300 rpm and a shaking diameter of 50 mm Type of MTP 24-square deep well 96-square deep well 96-square deep well 48-low well, no cells 96-low well, no cells 96-low well, cells 96-low well, cells
volume (height) 2.5 ml (8.7 mm) 0.5 ml (8.5 mm) 0.75 ml (12.8 mm) 0.6 ml (6.0 mm) 0.2 ml (6.2 mm) 0.2 ml (6.2 mm) 0.15 ml (4.7 mm)
OTR 50 mmol 38 mmol 24 mmol 27 mmol 20 mmol 26 mmol 32 mmol
O2 O2 O2 O2 O2 O2 O2
liter liter liter liter liter liter liter
1
h 1 h 1 h 1 h 1 h 1 h 1 h
1a 1a 1a 1b 1b 1a 1a
Refs Duetz et al., 2004 [13] Duetz et al., 2000 [9] Duetz et al., 2000 [9] Kensy et al., 2004 [17] Hermann et al., 2003 [5] Duetz, W.A, Unpublished Duetz, W.A, Unpublished
a
Derived from the oxygen-limited growth phase of growth curves of Pseudomonas putida. Determined by the cobalt-catalyzed oxidation of sulphite but corrected for the lower oxygen solubility (original values multiplied by 1.3 to compensate for low oxygen solubility: see text). b
far. These high values, however, do not imply that 48-well MTPs give rise to intrinsically higher OTRs compared with 96-low-well MTPs under more standard shaking conditions. A culture volume of 600 ml shaken at 300 rpm at a shaking diameter of 50 mm resulted in an OTR of 21 mmol O2 liter 1 h 1 (27 mmol O2 liter 1 h 1 if corrected for the lower oxygen solubility). If this OTR is compared with the OTRs measured in 96-low-well MTPs incubated under the same orbital-shaking conditions and a similar filling height, the difference is small (Table 1). An advantage of 48-well plates is that the fourfold higher culture volume reduces the effect of random pipetting errors and variations in inoculum sizes. However, the poor compatibility of the 48-well format with a lot of side equipment (such as microplate readers and multipipettes) is a distinct disadvantage. It might be concluded that MTPs with round wells give rise to lower OTRs than square-well MTPs at the same filling height and shaking conditions but that this disadvantage can be compensated for readily by using smaller culture volumes. The logistic advantages of the use of nontoxic and disposable polystyrene round-well MTPs – in combination with ever more sensitive assays – make it likely that they will gradually replace square-well MTPs for those applications where turbulence is not a major issue. Application of MTPs Mutant screenings Open literature on the application of MTPs for mutant screenings is scarce and mainly limited to metabolic flux analysis studies in academic settings [18–24]. The group of Uwe Sauer used 96-square deep-well MTPs to study glucose metabolism in 137 knockout mutants of B. subtilis using 13C tracer technology [18]. The application of small culture volumes in this research area is particularly appealing in the light of the high costs of 13C-labeled compounds. Culture volumes of 1.2 ml in combination with the cover systems described in Ref. [9] enabled the detection of changes in fluxes as low as 5–10%. A good correlation with results in 30 ml shake-flask cultures was observed. A comparative study from the same research group [20] concluded that batch-growth characteristics of E. coli in shake flask and 2-ml deep-well microtiter plates were indistinguishable. Standard deviations of growth rates and 13C glucose-uptake rates (among triplicates of 1.2 ml cultures in 96-square deep-well MTPs) were reported to be 7–8%. A similar study on 37 knockout mutants of Saccharomyces cerevisiae showed standard deviations in www.sciencedirect.com
growth rates of duplicates below 5% using 1.2 ml cultures in 96-square deep-well MTPs [21]. Standard deviations of a similar magnitude were reported for growth rates and fluxes through the pentose-phosphate pathway for 14 hemiascomycetous yeasts, including strains from the genera Saccharomyces, Kluyveromyces, Pichia, Yarrowia and Candida [22]. Weis et al. [23] optimized the growth conditions of Pichia pastoris in 96-square deep-well plates to limit cell death and to enable the accurate screening of enzyme variants in gene-expression libraries. The application of MTPs for mutant screenings in industry has become standard practice rapidly in recent years, for example, in screening high-productivity mutants for secondary metabolites. Minas et al. [24] reported standard deviations in the actinorhodin production by two Streptomyces coelicolor strains of 2% and 6% for 12 and six duplicates, respectively, using 1 ml cultures in 96square deep-well MTPs. The scarcity of further open literature in this area leaves me no choice other than to draw a general and subjective picture based mainly on personal contacts with representatives from the industry. If standard deviations in measured parameters between 5% and 10% are acceptable, the application of polystyrene 96-lowwell MTPs with maximal culture volumes of 0.2 ml is often feasible, although restricted to strains giving rise to homogeneous cultures, such as Bacillus, E. coli or yeast, and only if used in combination with very accurate pipetting robots and a well-defined inoculation size. If standard deviations below 5% are required for experiments, such as mature production bioprocesses where improvements per mutant cycle are very low, minimum culture volumes of 1 ml are indicative, especially for inhomogeneous cultures from, for example, Streptomyces mutant banks. Depending on the strain type, the use of square deep-well MTPs or 48or 24-low-well MTPs might give satisfactory results. The final choice is a trade-off between throughput per unit of time and the time-expenditure on what later appear to be false positives. In all mutant-screening projects, the inoculum size appears to be a crucial factor and measures to synchronize growth curves of all mutants are essential for low standard deviations of independent duplicates. The number of false positives could be reduced if end-point measurements were corrected for small differences in lag times and growth rates. Such corrections would be possible with the availability of methodologies to follow the growth in the wells of a MTP during continuous shaking conditions in parallel. Recent work from the Jochen Buchs group [25] addressed this concern by generating growth curves of Vibrio natriegens and E. coli using light scattering as well as intracellular and/or protein fluorescence.
474
Review
TRENDS in Microbiology Vol.15 No.10
OTRs and degrees of turbulence in the primaryscreening phase, similar in magnitude to those in the production facilities, help to limit the number of ‘wellplate wonders’ or mutants that perform reproducibly better in the primary screening but not in multiliter fermenters. Discovery of new secondary metabolites 24-square well MTPs are often used in the discovery of new secondary metabolites (for reviews, see [26,27]). The relative ease with which high OTRs are achieved [13] ensures a high biomass and sufficient secondary metabolites formed by anabolic pathways requiring oxygenases. Furthermore, culture volumes of 2–4 ml enable a straightforward separation of cells, pellets or mycelia from the supernatant by centrifugation and adequate amounts of extracted compounds for multiple bioassays, such as those for antibiotic or anticancer activities. Pelaez [26] advocated the use of MTPs with the argument that they permit an increase in the number of growth conditions that can be assigned to each strain and so enable a more complete exploitation of the metabolic potential of selected strains. 96-square deep-well MTPs are also applied in this area of research [26], particularly in combination with direct solvent extraction of the whole culture with no prior separation of biomass and supernatant. Miscellaneous applications Casey [28] described the successful use of MTPs to generate dose–response curves of test organisms (E. coli, B. subtilis and S. cerevisiae) to bioactive compounds formed by Streptomyces hygroscopicus. Diaz et al. [29] described a rapid assay for methionine using a methionine auxotroph E. coli strain growing in MTPs. Doig et al. [30] used several microwell formats (96-round-, 96-deep square- and 24-round-well microtiter plates) for quantification of the kinetics of a Baeyer-Villiger oxygenase expressed in E. coli. In a similar study, FerreiraTorres et al. [31] used 96-square deep-well MTPs to compare libraries of recombinant cyclohexanone monooxygenases in E. coli and Acinetobacter calcoaceticus. The use of MTPs for the screening of heterogeneous culture collections for regiospecific hydroxylations was described for a range of educt–product combinations [32–35]. Conclusions In the past 7 years, MTPs have become a mature alternative to Erlenmeyer flasks for the batch cultivation of microorganisms and have helped to make the handling and screening of large numbers of strains and mutant libraries less time consuming. Orbital shaking is suitable to generate OTRs in MTPs equaling or even surpassing the OTRs achieved traditionally in Erlenmeyer flasks, if the right shaking conditions are used. The development of suitable cover systems has ruled out cross-contaminations. Overall, these developments have contributed to the speed with which the productivity of large-scale bioprocesses can be improved and new secondary metabolites can be discovered. Further research is necessary for continued progress in this area (Box 1). www.sciencedirect.com
Box 1. Further research for improving the quality of growth in microtiter plates Studies on the degree to which various shapes of wells, in combination with various orbital-shaking conditions, mimic the hydrodynamic conditions that microbial cells encounter in large cultivation systems, such as stirred-tank bioreactors. Quantification of local fluid velocities. The development of filter materials combining high diffusion coefficients for O2 and CO2 with low diffusion coefficients for water vapor. The cultivation of slow-growing strains would benefit from the integration of such materials in well-closure systems. Adaptation of this technology for anaerobic cultivations, which would benefit flourishing research fields, such as the industrial production of ethanol. The development of technologies enabling the parallel quantitation of biomass levels in time during orbital shaking. Such data would, for example, enable end-point measurements of secondary metabolite levels to be corrected for small differences in lag times and growth rates of the screened mutants.
References 1 Micheletti, M. and Lye, G.J. (2006) Microscale bioprocess optimisation. Curr. Opin. Biotechnol. 17, 611–618 2 Kumar, S. et al. (2004) Minibioreactors. Biotechnol. Lett. 26, 1–10 3 Betts, J.I. and Baganz, F. (2006) Miniature bioreactors: current practices and future opportunities. Microbial. Cell. Fact. 5, Art. No. 21 4 Fernandes, P. and Cabral, J.M.S. (2006) Microlitre/millilitre shaken bioreactors in fermentative and biotransformation processes - a review. Biocatal. Biotrans. 24, 237–252 5 Hermann, R. et al. (2003) Characterization of gas-liquid mass transfer phenomena in microtiter plates. Biotechnol. Bioeng. 81, 178–186 6 Doig, S.D. et al. (2005) Modelling surface aeration rates in shaken microtitre plates using dimensionless groups. Chem. Eng. Sc. 60, 2741– 2750 7 Marshall, M.S. (1924) Surface tension of culture mediums. J. Infect. Dis. 35, 526–536 8 Bailey, J.E. and Ollis, D.F. (1986) Biochemical engineering fundamentals. 2nd edn., New York: McGraw-Hill 9 Duetz, W.A. et al. (2000) Methods for intense aeration, growth, storage, and replication of bacterial strains in microtiter plates. Appl. Environ. Microbiol. 66, 2641–2646 10 Zimmermann, H.F. et al. (2003) Rapid evaluation of oxygen and water permeation through microplate sealing tapes. Biotechnol. Prog. 19, 1061–1063 11 Hermann, R. et al. (2001) Optical method for the determination of the oxygen-transfer capacity of small bioreactors based on sulfite oxidation. Biotechnol. Bioeng. 74, 355–363 12 Doig, S.D. et al. (2005) Characterisation of a novel miniaturised bubble column bioreactor for high throughput cell cultivation. Biochem. Eng. J. 23, 97–105 13 Duetz, W.A. and Witholt, B. (2004) Oxygen transfer by orbital shaking of square vessels and deepwell microtiter plates of various dimensions. Biochem. Eng. J. 17, 181–185 14 Ortiz-Ochoa, K. et al. (2005) A novel method for the measurement of oxygen mass transfer rates in small-scale vessels. Biochem. Eng. J. 25, 63–68 15 Duetz, W.A. and Witholt, B. (2001) Effectiveness of orbital shaking for the aeration of suspended bacterial cultures in square deepwell microtiter plates. Biochem. Eng. J. 7, 113–115 16 Becker, R.F. et al. (1996) Additives. In Polypropylene handbook (Moore, E.P., Jr, ed.), pp. 177–210, Cincinnati, Ohio, Hanser-Gardner 17 Kensy, F. et al. (2005) Oxygen transfer phenomena in 48-well microtiter plates: Determination by optical monitoring of sulfite oxidation and verification by real-time measurement during microbial growth. Biotechnol. Bioeng. 89, 698–708 18 Fischer, E. and Sauer, U. (2005) Large-scale in vivo flux analysis shows rigidity and suboptimal performance of Bacillus subtilis metabolism. Nat. Gen. 37, 636–640
Review
TRENDS in Microbiology
19 Sauer, U. (2004) High-throughput phenomics: experimental methods for mapping fluxomes. Curr. Opin. Biotechnol. 15, 58–63 20 Fischer, E. et al. (2004) High-throughput metabolic flux analysis based on gas chromatography-mass spectrometry derived C-13 constraints. Anal. Biochem. 325, 308–316 21 Blank, L.M. et al. (2005) Large-scale C-13-flux analysis reveals mechanistic principles of metabolic network robustness to null mutations in yeast. Genome Biol. 6 (6), Art. No. R49 22 Blank, L.M. et al. (2005) Metabolic-flux and network analysis in fourteen hemiascomycetous yeasts. FEMS Yeast Res. 5, 545– 558 23 Weis, R. et al. (2004) Reliable high-throughput screening with Pichia pastoris by limiting yeast cell death phenomena. FEMS Yeast Res. 5, 179–189 24 Minas, W. et al. (2000) Streptomycetes in micro-cultures: growth, production of secondary metabolites, and storage and retrieval in the 96-well format. Antonie Van Leeuwenhoek 78, 297–305 25 Samorski, M. et al. (2005) Quasi-continuous combined scattered light and fluorescence measurements: a novel measurement technique for shaken microtiter plates. Biotechnol. Bioeng. 92, 61–68 26 Pelaez, F. (2006) The historical delivery of antibiotics from microbial natural products – can history repeat? Biochem. Pharmacol. 71, 981– 990 27 Singh, S.B. and Barrett, J.F. (2006) Empirical antibacterial drug discovery – foundation in natural products. Biochem. Pharmacol. 71, 1006–1015
Vol.15 No.10
28 Casey, J.T. et al. (2004) Development of a robust microtiter plate-based assay method for assessment of bioactivity. J. Microbiol. Methods 58, 327–334 29 Diaz, I.B.Z. et al. (2002) Adaptation of a methionine auxotroph Escherichia coli growth assay to microtiter plates for quantitating methionine. J. Rapid Meth. Autom. Microbiol. 10, 217–229 30 Doig, S.D. et al. (2002) The use of microscale processing technologies for quantification of biocatalytic Baeyer-Villiger oxidation kinetics. Biotechnol. Bioeng. 80, 42–49 31 Ferreira-Torres, C. et al. (2005) Microscale process evaluation of recombinant biocatalyst libraries: application to Baeyer-Villiger monooxygenase catalysed lactone synthesis. Bioprocess Biosys. Eng. 28, 83–93 32 Duetz, W.A. et al. (2001) Biotransformation of D-limonene to (+) transcarveol by toluene-grown Rhodococcus opacus PWD4 cells. Appl. Environ. Microbiol. 67, 2829–2832 33 Spain, J.C. et al. (2003) Production of 6-phenylacetylene picolinic acid from diphenylacetylene by a toluene-degrading Acinetobacter strain. Appl. Environ. Microbiol. 69, 4037–4042 34 O’Connor, K.E. et al. (2005) Isolation and characterization of a diverse group of phenylacetic acid degrading microorganisms from pristine soil. Chemosphere 61, 965–973 35 Van Beilen, J.B. et al. (2005) Biocatalytic production of perillyl alcohol from limonene by using a novel Mycobacterium sp cytochrome P450 alkane hydroxylase expressed in Pseudomonas putida. Appl. Environ. Microbiol. 71, 1737–1744
Elsevier celebrates two anniversaries with a gift to university libraries in the developing world In 1580, the Elzevir family began their printing and bookselling business in the Netherlands, publishing works by scholars such as John Locke, Galileo Galilei and Hugo Grotius. On 4 March 1880, Jacobus George Robbers founded the modern Elsevier company intending, just like the original Elzevir family, to reproduce fine editions of literary classics for the edification of others who shared his passion, other ‘Elzevirians’. Robbers co-opted the Elzevir family printer’s mark, stamping the new Elsevier products with a classic symbol of the symbiotic relationship between publisher and scholar. Elsevier has since become a leader in the dissemination of scientific, technical and medical (STM) information, building a reputation for excellence in publishing, new product innovation and commitment to its STM communities. In celebration of the House of Elzevir’s 425th anniversary and the 125th anniversary of the modern Elsevier company, Elsevier donated books to ten university libraries in the developing world. Entitled ‘A Book in Your Name’, each of the 6700 Elsevier employees worldwide was invited to select one of the chosen libraries to receive a book donated by Elsevier. The core gift collection contains the company’s most important and widely used STM publications, including Gray’s Anatomy, Dorland’s Illustrated Medical Dictionary, Essential Medical Physiology, Cecil Essentials of Medicine, Mosby’s Medical, Nursing and Allied Health Dictionary, The Vaccine Book, Fundamentals of Neuroscience, and Myles Textbook for Midwives. The ten beneficiary libraries are located in Africa, South America and Asia. They include the Library of the Sciences of the University of Sierra Leone; the library of the Muhimbili University College of Health Sciences of the University of Dar es Salaam, Tanzania; the library of the College of Medicine of the University of Malawi; and the University of Zambia; Universite du Mali; Universidade Eduardo Mondlane, Mozambique; Makerere University, Uganda; Universidad San Francisco de Quito, Ecuador; Universidad Francisco Marroquin, Guatemala; and the National Centre for Scientific and Technological Information (NACESTI), Vietnam. Through ‘A Book in Your Name’, these libraries received books with a total retail value of approximately one million US dollars.
For more information, visit www.elsevier.com www.sciencedirect.com
475