The filamentous fungus Penicillium chrysogenum analysed via flow cytometry—a fast and statistically sound insight into morphology and viability
Filamentous fungi serve as production host for a number of highly relevant biotechnological products, like penicillin. In submerged culture, morphology can be exceptionally diverse and is influenced by several process parameters, like aeration, agitation, medium composition or growth rate. Fungal growth leads to several morphological classes encompassing homogeneously dispersed hyphae and various forms of hyphal agglomerates and/or clump structures. Eventually, the so-called pellet structure can be formed, which represents a hyphal agglomerate with a dense core. Pellet structures can hinder oxygen and substrate transport, resulting in different states of viability, which in turn affects productivity and process control. Over the years, several publications have dealt with methods to either gain morphological insight into pellet structure or determine biomass viability. Within this contribution, we present a way to combine both in a flow cytometry–based method employing fluorescent staining. Thereby, we can assess filamentous biomass in a statistically sound way according to (i) morphology and (ii) viability of each detected morphological form. We are confident that this method can shed light on the complex relationship between fungal morphology, viability and productivity—in both process development and routine manufacturing processes.
KeywordsFilamentous fungi Penicillium chrysogenum Flow cytometry Viability Morphology Pellets
Successful cultivation strategies involving filamentous fungi need to consider the organism’s morphology. For example, Penicillium chrysogenum comprises several morphological forms when growing in submerged culture, ranging from homogenously dispersed hyphae to compact, hyphal agglomerates known as pellets (Veiter et al. 2018). Each morphological class affects viability, productivity and performance in different ways. From a process control standpoint, pellets are favoured as rheology, gas–liquid mass transfer and mixing are facilitated. However, pellet morphology also leads to active and non-active zones within the pellet due to limitations in transport of substrates and products, especially oxygen (Dynesen and Nielsen (2003)). These zones also affect productivity, as production of penicillin is happening in the non-growing cytoplasm found in the pellet’s outer layer (Baumgartl et al. 1984). In turn, the pellet’s core exhibits hyphal degradation, a decline in viability and no productivity (Ehgartner et al. 2017a, b). Naturally, these variations across all morphological forms complicate viability estimation and by extension determination of growth rate, substrate uptake rates and yields.
Hence, a quantitative approach to assess viable biomass is of utmost importance. Determination of viability can be performed employing at-line chemical methods such as fluorescent staining or physical techniques using various sensors. Dielectric spectroscopy, infrared spectroscopy and fluorescence have been comprehensively studied in the scope of filamentous fungi (Ronnest et al. 2011; Ehgartner et al. 2017a, b). While these methods enable real-time measurement, they cannot take into account morphological aspects directly. In this respect, flow cytometry is a potent alternative. Biomass morphology can be classified according to size and form through analysis of light scatter signals (Dubelaar et al. 1999; Ehgartner et al. 2017a, b; Pekarsky et al. 2018). To assess viability, fluorescent staining is regularly used in flow cytometry coupled with fluorescence detectors (Langemann et al. 2016; Attfield et al. 2000; Pekarsky et al. 2018). For filamentous fungi such studies are scarce, mainly due to the large particle sizes of fungal biomass (Dubelaar et al. 1999). Recent studies encompass Aspergillus niger microcolonies and Trichoderma (de Bekker et al. 2011; Delgado-Ramos et al. 2014), but are lacking detailed morphological analysis. Specific applications of flow cytometry for morphological classification of Penicillium chrysogenum were recently published (Ehgartner et al. 2017a, b); however, they did not include viability assessment yet.
In this publication, we quantitatively employ flow cytometry to combine detailed morphological insights with viability assessment. The developed method is at-line and potentially online applicable, statistically sound due to the high number of measured particles, and can estimate viable layers in specific morphological classes, such as pellets and large hyphal agglomerates. Furthermore, we have verified our results with established state-of-the-art methods such as a plate reader method for viability assessment as well as confocal laser microscopy for determination of a viable pellet layer. In the following, these points will be discussed: (i) differentiation of viable biomass against complex media background, (ii) morphological analysis and assessment of viability, (iii) comparison of flow cytometry viability assessment with the state-of-the-art plate reader method, (iv) analysis of large element morphology and viable layer and (v) comparison of results from flow cytometry with confocal laser microscopy.
Materials and methods
Spore suspensions of the P-14 P. chrysogenum candidate strain for penicillin production descending from the P-2 P. chrysogenum candidate strain (American Type Culture Collection with the access number ATCC 48271) (Lein 1986) were provided by Sandoz GmbH (Kundl, Austria) and used for all experiments.
Three cultivations (FB1 and FB2) were performed in a Techfors-S bioreactor (Infors HT, Bottmingen, Switzerland) with a 10-l working volume. The batch was cultivated with an initial volume of 6.5 l in the first mentioned bioreactor and inoculated with 2 × 108 spores/l. During batch phase pH was not controlled. The end of the batch was defined per default as an increase in pH of 0.5 by convention. After the batch, the broth was diluted with fed-batch medium (15% broth, 85% medium) and fed-batches were started with an initial volume of 6.5 l. Batch and fed-batch media were similar as described elsewhere (Posch and Herwig 2014).
During the fed-batch phase, pH was kept constant at 6.5 ± 0.1 by addition of 20% (w/v) KOH or 15% (v/v) H2SO4, respectively. pH was measured using a pH probe (Hamilton, Bonaduz, Switzerland). After additional 12-h nitrogen and phenoxyacetate feeds were started at constant rates (6.5 ml/h for nitrogen and 2 ml/h for phenoxyacetate). In the first 24 h of the fed-batch, 500 g/l glucose solution was fed at a constant rate of 1.01 ml/(l∙h). Afterwards, a three-times increase in feeding rate was carried out leading to a constant rate of 3 ml/(l/h).
The stirrer was equipped with three six-bladed Rushton turbine impellers, of which two were submersed and one was installed above the maximum liquid level for foam destruction. Fermentation temperature was kept at 25 °C via a cooling/heating jacket. Aeration was controlled at 1 vvm in batch and initial fed-batch with mass flow controllers (Vögtlin, Aesch, Switzerland). Dissolved oxygen concentration was measured using a dissolved oxygen probe (Hamilton, Bonaduz, Switzerland) and controlled between 40 and 90% during batch and between 40 and 60% during fed-batch, via adjustment of stirrer speed. The initial agitation conditions were 325 rpm stirring speed in batch and 500 rpm in fed-batch. CO2 and O2 concentrations in the off gas were analysed with an off-gas analyser (M. Müller AG, Switzerland).
Both cultivations were similarly conducted in a standard manner to generate biomass for method development. Only in FB2 was this strategy slightly altered: in order to measure a sudden viability decline, aeration was switched from air to N2 for FB2 at a process time of 160 h, which caused an immediate drop in dissolved oxygen concentration and CO2 concentration in the off gas.
Samples from fed-batch cultivations were diluted 1:10 into phosphate-buffered saline (50 g/l of 2.65 g/l CaCl2 solution, 0.2 g/l KCl, 0.2 g/l KH2PO4, 0.1 g/l MgCl∙6 H2O, 8 g/l NaCl and 0.764 g/l Na2HPO4 + 2 H2O) and stained with propidium iodide (Sigma-Aldrich, St. Louis, Missouri/USA; 20 mM stock dissolved in DMSO ≥ 99.9%, diluted with phosphate-buffered saline to a final concentration of 20 μM). In order to study different viability stages, some samples were subjected to microwave treatment for 30 s at 940 W in a microwave oven. After incubating for 5 min, the sample was further stained with fluorescein diacetate (Sigma-Aldrich, St. Louis, Missouri, USA; stock solution of 5 g/l dissolved in acetone ≥ 99.9% to a final concentration of 5 mg/l). After incubation of 5 min, the sample was further diluted (1:100 in the same buffer) for flow cytometric analysis. Metabolic activity is shown by fluorescein diacetate (FDA) treatment resulting in green fluorescence through esterase activity. PI fluorescence is a result from DNA intercalation in cells with compromised membranes (Pekarsky et al. (2018)).
A CytoSense flow cytometer (CytoBuoy, Woerden, Netherlands) with two forward scatter (FSC), one sideward scatter (SSC) and two fluorescence channels (green, red) was used for particle analysis. The implemented laser had a wavelength of 488 nm. The configuration of the filter set was 515–562 ± 5 nm for the green fluorescence channel (FL-green, used for fluorescein diacetate) and 605–720 ± 5 nm for the red fluorescence channel (FL-red, used for propidium iodide). The device was equipped with a PixeLINK PL-B741 1.3MP monochrome camera for in flow image acquisition. For data treatment, the software CytoClus3 (CytoBuoy, Woerden, Netherlands) and a custom-programmed Matlab 2016b script (MathWorks, Natick, Massachusetts, USA) were used.
The CytoSense flow cytometer provides multiple data points per channel per particle. This signal shape is achieved for both scatter channels as well as green and red fluorescence channels (Dubelaar et al. 1999). These pulse shapes are the basis for multiple curve parameters (Ehgartner et al. 2017a, b). Except for length parameters in micrometres, all parameters are in arbitrary units, as the user can set sensitivity levels SSC and fluorescence detectors. Setting of sensitivity levels was aligned with plate reader viability assessment. The most relevant parameters for the here presented study are the following parameters: maximum (maximum of signal curve), total (area under curve), length (length of the signal), sample length (length of signal above trigger level) and fill factor (similarity of the curve to a block; 0–1; higher if block-shaped).
At-line viability measurement via a plate reader
Confocal laser fluorescence microscopy
Confocal fluorescence microscopy was used as a method to distinguish between viable and dead parts of the pellets. FDA was used to stain metabolically active, viable hyphae while PI was used to stain dead cells. Microscopic images of the pellets were taken using a confocal fluorescent microscope (TE2000-E, Nikon, Japan).
One hundred microlitres of the bioreactor sample was diluted with 900 μl of PBS buffer and then centrifuged for 2 min at 500 rpm at room temperature. Eight hundred microlitres of the supernatant was discarded and replaced with 800 μl of PBS buffer. Afterwards, 10 μl of 200 μM PI reagent (prepared from 20 mM stock solution by 1:100 dilution) was added and the sample was incubated for 10 min in the dark. Twenty microlitres of the sample was then applied on a cover slide, and the slide was then placed on the microscope table. After focusing, 2 μl of freshly prepared 50 mg/l FDA reagent (Sigma-Aldrich, St. Louis, Missouri, USA; prepared with PBS buffer from a stock solution of 5 g/l dissolved in acetone) was added and a cover slide was placed on the sample. Lasers and the respective detector systems (PI: ex. 543 nm, em. 580 nm; FDA: ex. 488 nm, em. 507 nm) were activated separately. The gain for the 507-nm channel was adjusted according to FDA-related fluorescence intensity increase. The pellet was focused with the maximum intension of the PI stained area as criterion. Pictures were taken for at least 10 pellets per sample.
Differentiation of viable biomass against complex media background
Based on initial measurements of the fed-batch medium with and without cells, a distinction of fungal cells from the media background was possible. Particles exceeding a green fluorescence signal of 500 were classified as viable cells. Within previously set gates, viable cells and dead cells are easily differentiated from the media background as displayed in Fig. S1.
In principle, this differentiation is also possible without the use of fluorescent staining. However, results are negatively influenced by the media particle content (as demonstrated in Fig. S2) making a sharp distinction impossible. Naturally, fluorescence intensity of unstained biomass is 10–50 times lower as well, which further complicates differentiation.
Morphological analysis and assessment of viability
Measurement errors of common parameters determined by previously described methods
Plate reader staining method
Confocal laser microscopy
Viable layer large elements
Viable layer pellets
Comparing flow cytometry viability assessment with the state-of-the-art plate reader method
Analysis of large element morphology and viable layer
Comparing results from flow cytometry with confocal laser microscopy
Advantages, disadvantages and comparability to other methods
Within this contribution, we present a novel combination of morphological analysis and viability assessment based on flow cytometry. This signifies a faster alternative to image analysis via microscopy and more statistical reliability due to the large number of particles being measured in sort time spans (as previously established by Ehgartner et al. 2017a, b). In addition, enhanced insight into viability is generated simultaneously through fluorescent staining: Overall viability, viability of morphological classes and the viable layer of large elements can be determined. This viability data is enhanced by morphological parameters like pellet and large element compactness.
To verify this technique, results were compared to data from respective state-of-the-art methods, namely at-line viability measurement via plate-reader for overall viability and confocal laser microscopy for determination of the viable pellet layer. To generate sufficient amounts of biomass with diverse morphology and viability states, bioreactor cultivations in fed-batch mode were conducted and extensively sampled. Each sample was subjected to flow cytometry and plate-reader viability measurement. For determination of overall viability, the flow cytometry method was superior as the effects of a sudden drop of dissolved oxygen were registered more reliably. A selection of samples from FB1 was also analysed using confocal laser microscopy to determine viable layers across pellets. Results of the flow cytometry method were in accordance with reference measurements. Furthermore, the method was applicable in complex media with high particle background.
The main distinguishing feature of the flow cytometry method is that viability in different morphological classes can be determined, even down to individual particles. Other methods generally only provide an overview on viability. This is especially useful in later process stages: small hyphal elements tend to be viable, while degradation in larger agglomerates and pellets is observable over time. Such large elements can be analysed in detail; thereby, viable and non-viable biomass sections are identified and quantified over each particle. However, a diverse morphology is a challenging thing and needs to be addressed: to guarantee comparable information content across all process phases, a compromise in fluorescence detector sensitivity settings must be found for individual strain/media combinations: in early process phases, detectors must be sensitive enough to detect viable biomass; in later stages, signal saturation needs to be avoided when possible. Furthermore, it should be noted that fluorescence spectral overlap might result in misleading signals. This is especially true for large elements harbouring considerable green fluorescence from FDA, which can also be registered by the red fluorescence detector as a misleading artefact (Bagwell and Adams 1993). Consequently, the ratio between red and green fluorescence needs to be checked regularly. Deviations in this ratio occur due to saturation effects from green fluorescence signals and spectral overlap or might indicate viability decline.
Disadvantages also include size-exclusion effects: due to the large size and compact nature of fungal pellets, they might be excluded at the opening of the sampling tube. As a result, small elements are generally over-represented while more information can be obtained from the evaluation of large elements. If the measurement goal is characterization of large elements, a simple solution to the size-exclusion issue would be to increase measurement times or set trigger factors in the software according to particle size. However, a representative overview on morphology respecting all size classes of morphology is more challenging. Depending on the fungal species and/or strain to be analysed, certain adjustments of the sampling tube could be considered, like a wider tubing or a cone at the end of the sampling tube.
Applicability of the method
We envision this method to be a further milestone in the at-line characterization of complex fungal biomass (with a clear potential for online application through automated sampling systems) in process development and routine manufacturing processes. Based upon previous method development (as published by Ehgartner et al. 2017a, b), we enhanced morphological classification to analyse viability across all morphological forms with a special emphasis on the pellet’s viable layer. As a result, we are now able to combine morphological analysis with viability assessment in an at-line environment with potential online applicability through the use of automated sampling and sample processing. For this purpose, sampling, dilution and addition of fluorescent dyes needs to be performed in a modular process analytical (PAT) system with a flow cytometer connected.
We are confident that this method can shed a light on the complex and extensively researched relationship between fungal morphology, viability and productivity (Veiter et al. 2018; Wucherpfennig et al. 2011; Krull et al. 2013). While this method was developed for P. chrysogenum, we see the possibility to broaden applicability towards other filamentous fungi and by extent further agglomerate forming organisms such as yeast (Pekarsky et al. 2018).
Strains for the experiments were gratefully provided by Sandoz GmbH (Kundl, Austria).
Open access funding provided by TU Wien (TUW). This study was funded by the Austrian Federal Ministry of Science, Research and Economy in course of the Christian Doppler Laboratory for Mechanistic and Physiological Methods for Improved Bioprocesses (grant number 171).
Compliance with ethical standards
Conflict of interest
The authors declare that they have no conflict of interest.
This article does not contain any studies with human participants or animals performed by any of the authors.
- Lein, J (1986) “The Panlabs penicillin strain improvement program.” Vanek Z, Hostalek Z (eds) Overproduction of microbial metabolites. Butterworths, Boston, pp 105–139Google Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.